ENZYMES IN FARM ANIMAL NUTRITION...ENZYMES IN FARM ANIMAL NUTRITION 1 Z:\Customer\CABI\A3882 - Bedford + Partridge - Enzymes\A3957 - Bedford - Enzymes #P.vp 17 November 2000 14:57:14

$Page 1: ENZYMES IN FARM ANIMAL NUTRITION...ENZYMES IN FARM ANIMAL NUTRITION 1 Z:\Customer\CABI\A3882 - Bedford + Partridge - Enzymes\A3957 - Bedford - Enzymes #P.vp 17 November 2000 14:57:14$
A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-

ENZYMES IN FARM ANIMAL NUTRITION

1Z:\Customer\CABI\A3882 - Bedford + Partridge - Enzymes\A3957 - Bedford - Enzymes #P.vp17 November 2000 14:57:14

Color profile: DisabledComposite Default screen




Edited by

MICHAEL R. BEDFORD

and

GARY G. PARTRIDGE

FinnfeedsMarlborough

WiltshireUK

CABI Publishing


ENZYMES IN FARM ANIMALNUTRITION



CABI Publishing is a division of CAB International

CABI PublishingCAB InternationalWallingfordOxon OX10 8DEUK

Tel: +44 (0)1491 832111Fax: +44 (0)1491 833508Email: [email protected] site: http://www.cabi.org

CABI Publishing10 E 40th Street

Suite 3203New York, NY 10016

USA

Tel: +1 212 481 7018Fax: +1 212 686 7993

Email: [email protected]

© CAB International 2001. All rights reserved. No part of this publication may bereproduced in any form or by any means, electronically, mechanically, by photocopying,recording or otherwise, without the prior permission of the copyright owners.

A catalogue record for this book is available from the British Library, London, UK.

Library of Congress Cataloging-in-Publication DataEnzymes in farm animal nutrition / edited by M.R. Bedford and G.G. Partridge.

p. cm.Includes bibliographical references.ISBN 0-85199-393-1 (alk. paper)

1. Enzymes in animal nutrition. 2. Feeds--Enzyme content. 3. Animal feeding. I.Bedford, M. R. (Michael Richard), 1960- II. Partridge, G. G. (Gary G.), 1953-

SF98.E58 E59 2000636.08′52--dc21 00-044426

ISBN 0 85199 393 1

Typeset in Garamond by AMA DataSet LtdPrinted and bound in the UK by Biddles Ltd, Guildford and King’s Lynn




ContentsContentsContents

Contributors vii

Preface ix

1. The Current Feed Enzyme Market and Likely Trends 1C. Sheppy

2. Enzymology and Other Characteristics of Cellulases and Xylanases 11M.K. Bhat and G.P. Hazlewood

3. Enzymatic Characteristics of Phytases as they Relate to Their Use inAnimal Feeds 61D.D. Maenz

4. Analysis of Feed Enzymes 85B.V. McCleary

5. Maize: Factors Affecting its Digestibility and Variability in itsFeeding Value 109J.D. Summers

6. Vegetable Protein Meals and the Effects of Enzymes 125J. Thorpe and J.D. Beal

7. Enzyme Supplementation of Poultry Diets Based on Viscous Cereals 145M. Choct

8. The Role and Efficacy of Carbohydrase Enzymes in Pig Nutrition 161G.G. Partridge

v

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-Contents

Contents



9. Interaction between Cereal Identity and Fat Quality and Content inResponse to Feed Enzymes in Broilers 199S. Dänicke

10. Digestion of Phosphorus and Other Nutrients: the Role of Phytasesand Factors Influencing Their Activity 237E.T. Kornegay

11. Enzymes in Ruminant Diets 273T.A. McAllister, A.N. Hristov, K.A. Beauchemin, L.M. Rode andK.-J. Cheng

12. Microbial Interactions in the Response to Exogenous EnzymeUtilization 299M.R. Bedford and J. Apajalahti

13. Enzymes: Screening, Expression, Design and Production 315K. Clarkson, B. Jones, R. Bott, B. Bower, G. Chotani and T. Becker

14. Liquid Application Systems for Feed Enzymes 353P. Steen

15. Process Stability and Methods of Detection of Feed Enzymes inComplete Diets 377M.R. Bedford, F.G. Silversides and W.D. Cowan

16. Future Horizons 389R.R. Marquardt and M.R. Bedford

Index 399

vi Contents

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-Contents



ContributorsContributorsContributors

J. Apajalahti, Danisco-Cultor Innovation, Kantvik, FIN-02460, FinlandJ.D. Beal, University of Plymouth, Seale-Hayne Faculty, Newton Abbot, Devon

TQ12 6NQ, UKK.A. Beauchemin, Agriculture and Agri-Food Canada, Lethbridge Research Centre,

PO Box 3000, Lethbridge, Alberta, Canada T1J 4B1T. Becker, Genencor International Inc., 925 Page Mill Road, Palo Alto, CA

94304-1013, USAM.R. Bedford, Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UKM.K. Bhat, Food Materials Science Division, Institute of Food Research, Norwich

Research Park, Colney, Norwich NR4 7UA, UKR. Bott, Genencor International Inc., 925 Page Mill Road, Palo Alto, CA

94304-1013, USAB. Bower, Genencor International Inc., 925 Page Mill Road, Palo Alto, CA

94304-1013, USAK.-J. Cheng, Faculty of Agricultural Sciences, University of British Columbia,

Vancouver, British Columbia, Canada V6T 1Z4: Present address; Institute ofBioAgricultural Sciences, Academia Sinica, Taipei, Taiwan

M. Choct, School of Rural Science and Natural Resources, Department of AnimalScience, The University of New England, Armidale, NSW 2351, Australia

G. Chotani, Genencor International Inc., 925 Page Mill Road, Palo Alto, CA94304-1013, USA

K. Clarkson, Genencor International Inc., 925 Page Mill Road, Palo Alto, CA94304-1013, USA

W.D. Cowan, Novo Nordisk, Krogshoevej 36, Bagsvaerd, Denmark 2880S. Dänicke, Institute of Animal Nutrition, Federal Agricultural Research Centre,

Braunschweig (FAL), Bundesallee 50, D-38116 Braunschweig, GermanyG.P. Hazlewood, Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UKA.N. Hristov, Agriculture and Agri-Food Canada, Lethbridge Research Centre,

PO Box 3000, Lethbridge, Alberta, Canada T1J 4B1B. Jones, Genencor International Inc., 925 Page Mill Road, Palo Alto, CA

94304-1013, USA

vii

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-Contributors

Contributors



E.T. Kornegay (deceased), Department of Animal and Poultry Sciences, VirginiaPolytechnic Institute and State University, 3060 Litton-Reaves Hall,Blacksburg, VA 24061, USA

D.D. Maenz, Prairie Feed Resource Centre (Canada) Inc., Department of Animaland Poultry Science, University of Saskatchewan, Saskatoon, Saskatchewan,Canada S7N 0W0

R.R. Marquardt, Department of Animal Science, University of Manitoba,Winnipeg, Manitoba, Canada R3T 2N2

T.A. McAllister, Agriculture and Agri-Food Canada, Lethbridge Research Centre,PO Box 3000, Lethbridge, Alberta, Canada T1J 4B1

B.V. McCleary, Megazyme International Ireland Limited, Bray Business Park, Bray,Country Wicklow, Ireland

G.G. Partridge, Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UKL.M. Rode, Agriculture and Agri-Food Canada, Lethbridge Research Centre,

PO Box 3000, Lethbridge, Alberta, Canada T1J 4B1: Present address; RosebudTechnology Development Ltd, Lethbridge, Alberta, Canada T1K 3J5

C. Sheppy, Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UKF.G. Silversides, Denman Island, British Columbia, CanadaP. Steen, Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UKJ.D. Summers, Poultry Industry Centre, RR#2, Guelph, Ontario, Canada

N1G 6H8J. Thorpe, University of Bristol, Department of Clinical Veterinary Science,

Langford House, Langford, Bristol BS40 5DU, UK

viii Contributors

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-Contributors



PrefacePrefacePreface

The rearing and feeding of domestic animals and the use of enzymes in processessuch as brewing have been distinct parts of human life for many thousands ofyears, but it is only recently that these two disciplines have crossed paths. The firstcommercial use of feed enzymes dates back to 1984 in Finland, where opportunitiesexisted to improve significantly the nutritional quality of barley-based rations byinclusion of enzymes derived from the brewing industry (A. Haarisilta, SuomenRehu). The years since then have seen an exponential increase in the usage of manyenzyme types in rations for poultry and, to a lesser extent, swine. Significant recentinterest has also been shown by the ruminant sector. Scientific studies describing theuse of exogenous enzymes in animal nutrition dates back to the mid 1920s andthey now number in excess of 1300 papers for broilers alone (Rosen, 2000, personalcommunication). This rapidly expanding field is becoming increasingly multi-disciplinary as more is understood of the mode of action of feed enzymes. Due tothe complexity of this field it is timely to produce a single source from where theuninitiated and well versed alike can ground themselves with the basics required tograsp the subject. Furthermore, it is the intention of this book to provide sufficientdetail to enable an understanding of the complexities of response to such products inall classes of farm animal livestock.

It is of interest that not only the usage of enzymes has increased but also thescope of their use. This is likely to continue as the price of enzyme application falls,with improvements in both enzyme efficacy and production costs making previouslyuneconomic solutions more attractive. Particular attention is drawn to the ruminantsector, where usage is in its infancy but research is demonstrating that significantgains can be made. The challenge is to find methods of predicting enzyme responseso that enzyme application in all classes of livestock becomes increasingly a sciencerather than an art.

ix

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-Preface

Preface



In memoriam

It is with great sadness that this book is produced after the death of Dr E.T.Kornegay, a significant contributor to this publication and to the field of phytaseresearch in poultry and pigs. Dr Kornegay was particularly active and enthusiastic inhis enzyme phytase research and will be sadly missed.

x Preface

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-Preface



C. SheppyThe Current Market and Likely Trends1

Global Animal Feed Production

As with any publication dealing with the application of a particular technology to acertain industry, it is important to assess the commercial environment in which thetechnology is being applied.

Despite the general euphoria and enthusiasm surrounding the dawn of a newmillennium, the challenges faced by the global animal feed industry have a familiarfeel. Human health, environmental safety and animal welfare are all issues thatcontinue to be to the fore of consumers’ minds and are at the top of the agenda ofthe media and politicians alike. Although tending to emerge in Europe and NorthAmerica these issues are now of worldwide importance, a development that mirrorsthe increasing globalization of the food chain and feed industry. Recent controversiessurrounding genetically modified crops, antibiotic growth promoters and dioxin-contaminated feedstuffs provide some very real examples of this. A direct resultof these ‘health scares’ has been the adoption, and in some cases the imposition, ofincreased regulatory, testing and quality control procedures aimed at maintainingand improving the trust between animal producers and the ultimate consumers oftheir products.

Faced with these challenges, how is the industry currently faring? Althoughglobal compound feed production actually fell by 5% in 1998 to 575 Mt (the firstsuch fall in over 50 years), 1999 experienced something of a recovery, albeit small, to586 Mt (Table 1.1).

The Asian economic crisis of 1997–1998 saw the collapse of the feed industry inmany previously booming markets and with the benefit of hindsight it is probablyfair to say that the previous rates of growth were unsustainable. As demand for meat,milk and eggs tumbled, so did feed production: Indonesian feed output plummetedby 40%; Korean production fell by 30%; Thai output was down nearly 25%. Eventhe Chinese market appeared to stumble, with an approximate 2% decline inindustrially manufactured feed to below 55 Mt. These Asian falls were followed by ageneral stagnation of the global feed market in 1998, with EU feed production fallingby 6% and the non-EU European countries witnessing a 12% decline. Only North

©CAB International 2001. Enzymes in Farm Animal Nutrition(eds M.R. Bedford and G.G. Partridge) 1

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-1

The Current Feed EnzymeMarket and Likely Trends

C. SHEPPYFinnfeeds, PO Box 777, Marlborough, Wiltshire, SN8 1XN, UK

1



America (+1%), Latin America (+1%) and the Middle East and African countries(+10%) experienced any growth in their feed industries in 1998. As clearly seen inTable 1.1, the recovery in 1999 was global, with all major regions showing somegrowth as the Asian economies improved and demand picked up in other markets(Gill, 2000).

Despite the fall, or correction, in feed production seen in 1998, one trend withinthe feed industry continues unabated – the rationalization of capacity and theconcentration and integration of production into the major food-producingcompanies. Driven by the need for cost reduction, increased regulatory requirementsand hygiene and safety standards, fewer than 3800 feed mills now manufacture morethan 80% of the world’s compound feed. However, there is still plenty of room forfurther consolidation and rationalization as the ten largest feed manufacturers stillonly account for approximately 9–10% of global output.

Although pioneered in the poultry industry, food company integration nowextends to most of the major farmed species – poultry, swine, beef, dairy cattle andaquatic. Swine feed is still the largest proportion of feed produced but broiler feedproduction continues to grow at the expense of beef cattle feeds (Table 1.2), spurredon by increased consumer demand for cheap, safe and healthy meat products (Gill,2000).

So what does the future hold for the feed industry? Most informedcommentators believe that the drop in feed production seen in 1998 was a ‘one-off’or correction caused by the dramatic collapse experienced by the Asian economies.Muller (1999) comments that:

in the last 25 years compound feed production has always grown parallel with theincrease in world population. The decline in per capita ‘consumption’ of compoundfeed from 104 kg to 97 kg in 1998 does not represent a reverse trend in global com-pound feed production, but instead must be categorized correctly in conjunction withthe Asian crisis.

With the world population set to grow from the current 6 billion to approximately7.8 billion over the coming two decades, the demand for food, including animal

2 C. Sheppy


1998 production(Mt)

1999 production(Mt)

Percentage difference(1999/1998)

North AmericaAsia-PacificEuropean UnionLatin AmericaNon-EU EuropeMiddle East and AfricaOtherTotal

159.8128.1114.9

63.347.421.639.9

575.0

160.7132.2116.4

65.448.624.038.7

586.0

+0.6+3.2+1.3+3.3+2.5

+11.1−3.1+2.0

Table 1.1. Global animal feed production, 1998 and 1999. Source: Gill (1999, 2000).



protein, is set to increase at an even greater rate as the growing populations in thedeveloping economies alter their dietary habits. Set against this scenario, the futurefor compound feed production looks positive. However, those companies who wishto continue in the industry and expand their businesses will need to become moreadept at operating in a market environment where many of the key influencingfactors are beyond their control.

Background to Enzyme Technology

Enzymes exist practically everywhere. They are naturally occurring and are producedby all living organisms and, as nature’s catalysts, they speed up the chemical reactionsthat enable all living things, from simple single-celled organisms, through plants andinsects to humans, to function. Without them food could not be digested.

So far over 3000 different enzymes have been discovered. Enzymes, as with allproteins, are made from chains of amino acids. They speed up or catalyse reactionsby binding to their substrate and ‘stabilize’ the entire reaction process through toproduct formation, so that far less activation energy is required to move the reactionforwards. As a result, the rate of reaction progression is greatly increased for any givenenergy status. It is the three-dimensional shape and position of the reactive aminoacids within the molecule that confer the catalytic properties of enzymes. Conditionsthat significantly alter the structure of the active enzyme often result in loss ofactivity. As a result, enzymes are very sensitive to the environment in which theyfunction and many work best in mild temperatures and mid-range pH.

Humans have made use of enzymes, often unknowingly, throughout history.Cheese making and the use of malted barley in brewing are examples of theharnessing of the power of enzymes. Modern enzyme technology really started in1874 following the first documented production of a refined enzyme that wasprepared from the contents of calves’ stomachs. That enzyme, rennet, is still usedtoday in cheese manufacture.

Since then the technology to identify, extract and produce enzymes on acommercial scale has progressed dramatically and they are now used in manyindustrial processes. Currently enzymes are used in detergents, paper production

The Current Market and Likely Trends 3


Sector 1993 (%) 1998 (%)

SwineBroilerBeefDairyLayerAquaticOther

31241117

836

3127

917

853

Table 1.2. Global animal feed production by species. Source: Hoffman (1999).



and in leather and textile processing. In addition, they are widely used in the foodand drink industry in products ranging from bread and cheese to fruit juice, wine andcoffee. The global market value for enzymes in 1995 was estimated to be worth US$1billion (Godfrey and West, 1996) and is forecast to rise to between US$1.7 billionand US$2 billion by 2005.

Why use enzymes in animal feeds?

The principal rationale for the use of enzyme technology is to improve the nutritivevalue of feedstuffs.

All animals use enzymes in the digestion of food, those produced either by theanimal itself or by the microbes present in the digestive tract. However, the digestiveprocess is nowhere near 100% efficient; for example, swine are unable to digest15–25% of the food they eat. Therefore, the supplementation of animal feeds withenzymes to increase the efficiency of digestion can be seen as an extension of theanimal’s own digestive process.

In many animal production systems feed is the biggest single cost andprofitability can depend on the relative cost and nutritive value of the feeds available.Often, the limiting factor when formulating rations is the animal’s ability to digestdifferent constituent parts of the feed raw materials, particularly fibre. Despite recentadvances, the potential nutritional value of feedstuffs is not achieved at the animallevel. This inefficiency in the utilization of nutrients can result in a cost to the farmer,the food company and the environment. Put simply, there are four main reasons forusing enzymes in animal feed:

1. To break down anti-nutritional factors that are present in many feed ingredients.These substances, many of which are not susceptible to digestion by the animal’sendogenous enzymes, can interfere with normal digestion, causing poor performanceand digestive upsets.2. To increase the availability of starches, proteins and minerals that are eitherenclosed within fibre-rich cell walls and, therefore, not as accessible to the animal’sown digestive enzymes, or bound up in a chemical form that the animal is unable todigest (e.g. phosphorus as phytic acid).3. To break down specific chemical bonds in raw materials that are not usuallybroken down by the animal’s own enzymes, thus releasing more nutrients.4. To supplement the enzymes produced by young animals where, because of theimmaturity of their own digestive system, endogenous enzyme production may beinadequate.

In addition to improving diet utilization, enzyme addition can reduce the variabilityin nutritive value between feedstuffs, improving the accuracy of feed formulations.Trials have shown that ensuring feed consistency in this way can increase the unifor-mity of groups of animals, thus aiding management and improving profitability. Thegeneral health status of animals can also be indirectly influenced, resulting in fewer of

4 C. Sheppy




the non-specific digestive upsets that are frequently provoked by fibre components inthe feed.

Of increasing importance and relevance to the feed industry are the environ-mental benefits of harnessing enzyme technology. Since the animal better utilizes thefeed, less is excreted. This results in manure volume being reduced by up to 20% andnitrogen excretion by up to 15% in pigs and 20% in poultry. As significant is theopportunity for enzymes to reduce phosphorus pollution.

How do they work and how are they used in animal feeds?

Broadly speaking, four types of enzymes currently dominate the animal feed market:enzymes to break down fibre, protein, starch and phytic acid.

Fibre-degrading enzymesOne of the main limitations to digestion is the fact that monogastrics (pigs and poul-try) do not produce the enzymes to digest fibre. In diets containing ingredients suchas wheat, barley, rye or triticale (the main ‘viscous’ cereals), a large proportion of thisfibre is soluble and insoluble arabinoxylan and β-glucan (White et al., 1983; Bedfordand Classen, 1992). The soluble fibre can increase the viscosity of the contents of thesmall intestine, impeding the digestion of nutrients and thereby reducing the growthof the animal. It has also been linked with the incidence of digestive disorders such asnon-specific colitis in swine, and sticky litter and hock burns in poultry.

The fibre content of wheat and barley can vary considerably according tovariety, growing location, climatic conditions, etc. This in turn means that there canbe considerable variability in the nutritional value of these ingredients and hencediets containing them. In breaking down the fibre, enzymes (e.g. xylanase targetingarabinoxylans, β-glucanase targeting β-glucans) can reduce this variability innutritional value, giving rise to improvements in the performance of the feed and theconsistency of the response. An added benefit is the reduced incidence of certaindigestive disorders.

Protein-degrading enzymesVarious raw materials contribute to the protein content in the diet and ultimately theamino acids that fuel lean meat deposition. There is considerable variability in thequality and availability of protein from the different raw materials typically found inmonogastric diets. Within the primary vegetable protein sources such as soybeanmeal, certain anti-nutritional factors (ANFs), such as lectins and trypsin inhibitors,can lead to damage to the absorptive surface of the gut, impairing nutrient digestion.In addition, the underdeveloped digestive system of young animals may not be ableto make optimal use of the large storage proteins found in the soybean meal (glycininand β-conglycinin).

The addition of a protease can help to neutralize the negative effects of theproteinaceous ANFs in addition to breaking down the large storage proteinmolecules into smaller, absorbable fractions.





Starch-degrading enzymesTo many nutritionists maize is viewed as the ‘gold standard’ of raw materials. Mostnutritionists do not consider maize digestion as being poor: in fact most would arguethat it is better than 95% digested. However, recent evidence presented by Noy andSklan (1994) suggests that, at the ileal level, starch digestibility rarely exceeds 85% inbroilers between 4 and 21 days of age. The addition of an amylase to animal feed canhelp to expose the starch more rapidly to digestion in the small intestine, and indoing so lead to improved growth rates from enhanced nutrient uptake.

At weaning, piglets often suffer a growth check because of changes in theirnutrition, environment and immune status. The addition of an amylase, usually inconjunction with other enzymes, to augment the animal’s endogenous enzymeproduction has been shown to improve nutrient digestibility and absorption and,hence, growth rate for a range of diets (Close, 1995).

Phytic acid-degrading enzymesPhosphorus is required for bone mineralization, immunity, fertility and growth andis an essential mineral for all animals. Swine and poultry digest only about 30–40%of the phosphorus found in feedstuffs of vegetable origin, with the remainder beingtied up in a form inaccessible to the animal – phytic acid. In many instancesadditional phosphorus must be added to the diet to meet the animal’s requirement.More than half of the phosphorus consumed from such feedstuffs is excreted in thefaeces, which can result in major environmental pollution. By adding a phytase tothe diet, the phytic acid is broken down, liberating more of the phosphorus for use bythe animal.

The two main benefits of phytase supplementation are, firstly, the reduction infeed costs from the reduced additional supplementation of phosphorus to the dietand, secondly, environmental from reduced excretion of waste products and thethreat of pollution.

The Current Feed Enzyme Market

It should come as no surprise that the adoption of enzyme technology by the animalfeed industry has, to a large degree, been driven by the mode of action of these fourparticular types of enzymes. Taken in conjunction with the different structure of thevarious segments of the animal feed industry, the background to the adoption ofthe technology becomes even clearer.

The global poultry industry, particularly the broiler industry, is highlyintegrated and dominated by a relatively small group of large companies. Forexample, Tyson’s alone in the USA is estimated to produce some 2 billion broilersper year, considerably larger than the entire annual production of the UK poultryindustry. The company controls every step of the production chain from the growingof the raw materials and the breeding of the birds through to the processing of themeat and ultimately the retailing of the finished product. This integration andconcentration of the majority of industry capacity in a relatively small number of

6 C. Sheppy




hands makes the adoption of new technology a rapid occurrence, particularly whenthe value of the opportunity is significant. The value of enzyme addition to broilerfeeds will typically deliver a return on investment (ROI) well in excess of 2 : 1.

By contrast the swine industry is considerably more fragmented, with manymore producers and steps in the chain. In addition, the efficacy response to enzymeaddition has been more variable and harder to measure commercially.

Although first considered in the 1950s, it was not until the 1980s that theanimal feed industry really began to understand how to harness the power of enzymetechnology properly.

As already discussed, feed grains such as wheat and barley contain relativelyhigh levels of fibre that monogastrics are unable to digest. If the problem element ofthe fibre can be broken down, the animal has greater access to the available nutrients,so overcoming the negative impacts of incorporating the ingredient in the diet.Spurred on by a plentiful supply of cheap barley, European poultry nutritionistsand enzymologists investigated the opportunity to reduce the negative impact ofincluding barley in broiler diets by adding β-glucanase to the diet. This proved to bea success with the ‘rule of thumb’ being adopted that ‘barley + β-glucanase = wheat’.

Encouraged by the outstanding success of this approach, wheat became thenext target for enzymatic enhancement via the use of xylanase. Work this timeconcentrated on the hypothesis that ‘wheat + xylanase = maize’.

Again this approach was successful and the mid-1990s saw the rapid acceptanceof the application of enzyme technology to the animal feed industry. It is probably noexaggeration to say that by 1996 in excess of 80% of all European broilers diets thatcontained a viscous cereal (wheat, barley, etc.) would have also contained a fibre-degrading enzyme – certainly an impressive adoption rate for a new technology in thefeed industry.

Taking a global perspective across all species and diet segments, currentestimates of market penetration suggest that approximately 65% of all poultry feedscontaining viscous cereals also contain a fibre-degrading enzyme. Given the verydifferent structure of the industry and the difficulties in accurately measuring anefficacy response, it is no surprise that current penetration into the swine industry isconsiderably lower, approaching 10%.

In terms of geographical spread, the utilization of fibre-degrading enzymes isconcentrated on those regions where viscous cereals are the predominant energysource, namely Europe, Canada and Australia/New Zealand. (This is not to discountcompletely the USA, South America and the rest of Asia-Pacific, where there are reg-ular ‘window’ opportunities depending on the post-harvest price ratio between maizeand, for example, wheat.) It is for this reason that fibre-degrading enzymes are largelyseen as a ‘European’ niche product. To gain global acceptance it will be necessary forthe enzyme manufacturers to break significantly into the North American andAsia-Pacific maize–soybean markets. Maize–soybean diets are historically regarded asthe ‘gold standard’ in terms of consistency of animal response, though most nutri-tionists recognize that these raw materials are more variable than is often presumed.

There is increasing evidence and acceptance that this ‘gold standard’ can beimproved upon and that there is a role for enzymes in these diets that do not





superficially present a problem related to fibre or viscosity. The focus of considerableresearch-and-development (R&D) dollars over the last 10 years has seen the firstgeneration of maize–soybean enzymes for poultry diets, launched in 1996. Despiteinitial mixed results the industry is now beginning to understand more fully how bestto apply this technology to reap an economic reward. Current estimates suggest thatthis market segment is worth approximately US$20 million and that 5% of broilerdiets based upon maize–soybean now contain an enzyme. The combined marketfor these viscous and non-viscous (carbohydrase) enzymes is currently (1999/2000)estimated to be in excess of US$100 million.

One enzyme concept that already has both global acceptance and globalapplication is phytase. The type of carbohydrate source present does not drive itsmode of action, as phytic acid is present at varying levels in all vegetable feed rawmaterials. Estimates suggest that the phytase market is currently worth up to US$50million, with approximately 8% of global swine and poultry feeds containingphytase. This again is an exceptional adoption rate for a new technology, given theshort period since the concept was launched.

The driving force behind this success has been increasing concern about theimpact of phosphorus on environmental pollution. In a number of countrieslegislation has either been imposed (e.g. The Netherlands) or is being imposed (e.g.the Delmarva Peninsula in the USA) on pig and poultry farmers. The historicalapplication of phytase is to replace a proportion of the supplementary calciumand phosphorus added to the diet, but now phytase suppliers are increasinglyconcentrating on the apparent potential of the product to improve the availability ofcertain other nutrients (e.g. amino acids and trace minerals) in the diet. The aim is toimprove the ROI and so the uptake of this new technology.

The largest applications for phytase are currently into laying hens and swine,given the scope to remove larger quantities of supplementary phosphorus and theease of application (note that many layer and swine diets are not heat treated). Thereremains significant scope to increase phytase application and application uptakein the broiler segment as the value equation improves and the enzyme industryovercomes the issues surrounding liquid application and the thermostability ofenzymes.

Future Needs and Opportunities

The adoption of enzyme technology has forced the animal feed industry to challengetraditional assumptions about diet formulation, ingredient selection, nutrientrequirements and the desired productive response. Ten years ago many Europeannutritionists viewed wheat in the same light as North Americans view maize –basically as an ingredient that could not be improved. As enzyme technologybecomes more accepted and widespread throughout the global animal feed industry,so the enzyme producers must continue to develop new products and applicationsto overcome the existing obstacles to growth, as well as to offer new solutions andopportunities.

8 C. Sheppy




The thermal stability of enzymes and their ability to survive the heat processingsteps in the manufacture of pelleted animal feed is of major concern to the industry.This, along with concerns over enzyme assays and the need for full traceabilitythroughout the food chain, will force the enzyme manufacturers to increase theirefforts to discover new enzyme products that can survive the increasingly highprocessing temperatures, deliver in vivo efficacy and be recovered in standard qualitycontrol procedures. Alternatively, improved post heat treatment liquid applicationsystems will offer the industry short- to medium-term solutions.

The adoption of existing enzyme technology in the animal feed sector hasincreased as knowledge of the interaction between the enzyme, its substrate and theenvironment of the animal’s gut has evolved and improved. It is very likely that theincreased uptake of enzymes in the maize–soybean type diets will be driven by anenhanced understanding of which ingredients will respond to enzyme treatment andthe amount of enzyme required to deliver the most economic and cost-effectiveresponse. The ultimate scenario would see a computerized and automated ‘on-line’process that tests the raw material as it enters the feed mill and then doses theeconomically optimal amount of enzyme into the mixer as the feed is produced.

The decision by the EU authorities to ban the use of certain antibiotic growthpromoters, and the knock-on effect in other countries around the world, has forcedfeed manufacturers and meat producers to explore alternatives. Enzymes are regularlyput forward as a technology that will allow producers to continue to produce safe,nutritious and cheap meat, milk and eggs whilst meeting the ever increasingconsumer demand for these products. It is important to note here that there are manyproducts being promoted opportunistically on the back of the antibiotic growthpromoter ban. When searching out realistic and sustainable solutions to fill this gap,it is important that the feed industry is rigorous and consistent in its demand forscientifically credible technologies and products.

Despite the impressive uptake of the technology, only approximately 10% ofmonogastric feeds today contain a feed enzyme, giving a total market value in excessof US$150 million. As such, perhaps the enzyme supply industry needs to ask somesearching questions of itself as to why uptake has not been faster, particularly in thoseapplications where there appears to be a sound commercial case. Standardized andmore transparent quality control procedures, improved heat stability, more accurateliquid application systems, clearer presentation of technical information andproducts that deliver an even more consistent productive response are just some ofthe factors raised by the feed industry to explain lack of use. It is clear that thereremains huge untapped potential for enzyme technology in the animal feed industry.

References

Bedford, M.R. and Classen, H.L. (1992) Reduction of intestinal viscosity through manipula-tion of dietary rye and pentosanase concentration is effected through changes in the car-bohydrate composition of the intestinal aqueous phase and results in improved growthrate and food conversion efficiency of broiler chicks. Journal of Nutrition 122, 560–569.





Close, W. (1995) Enzymes in action, feed mix enzymes special issue Enzymes - Nature’sCatalyst. Misset pp. 17–20.

Gill, C. (1999) Feed International, January 1999, Watt Publishing, pp. 4–10.Gill, C. (2000) Feed International, January 2000, Watt Publishing, pp. 4–6.Godfrey, T. and West, S.I. (1996) Introduction to industrial enzymology. In: Godfrey, T. and

West, S.I. (eds) Industrial Enzymology, 2nd edn. Macmillan, UK, pp. 1–8.Hoffman, P. (1999) Prospects for feed demand recovery. International Feed Markets ’99,

October 1999, Agra Europe.Muller, A. (1999) Kraftfutter, September 1999, pp. 303–311.Noy, Y. and Sklan, D. (1994) Digestion and absorption in the young chick. Poultry Science,

366–373.White, W.B., Bird, H.R., Sunde, M.L., Marlett, J.A., Prentice, N.A. and Burger, W.C. (1983)

Viscosity of β-D-glucan as a factor in the enzymatic improvement of barley for chicks.Poultry Science 62, 853–862.

10 C. Sheppy




M.K. Bhat and G.P. HazlewoodCharacteristics of Cellulases and Xylanases2

Introduction

Cellulose and hemicellulose are the major plant structural polysaccharides andaccount for approximately 70% of plant biomass (Ladisch et al., 1983). It has beenestimated that the amount of carbon fixed by plants during photosynthesis is over100 billion tons per annum (Ryu and Mandels, 1980). In addition to their pivotalrole in maintaining the structural integrity of plants, cellulose and hemicellulose serveas a major source of nutrients for herbivores and as renewable substrates for theproduction of food, animal feed, paper and pulp as well as textiles (Ryu and Mandels,1980; Gilbert and Hazlewood, 1993; Beguin and Aubert, 1994).

It has been unambiguously established that cellulose and hemicellulose can beconverted to soluble sugars by enzymes of mainly microbial origin collectively termedcellulases and hemicellulases (Mandels, 1985; Viikari et al., 1993). Microorganismsincluding fungi, bacteria and actinomycetes produce mainly three types of cellulasecomponents, namely endoglucanase, exoglucanase and β-glucosidase, either asseparate entities or in the form of an aggregated complex, for cellulose hydrolysis(Wood, 1985; Lamed and Bayer, 1987, 1988; Bhat and Bhat, 1997). The hemi-cellulose fraction of the cell walls of most species of land plants contains mainly xylanand mannan and requires a more extensive repertoire of enzymes to effect completehydrolysis to soluble sugars (Biely et al., 1992; Hazlewood and Gilbert, 1998a). Thetwo main enzymes involved are endoxylanases (xylanases) and endomannanases(mannanases), which attack the backbone structure (Viikari et al., 1993). Otherhemicellulases, including β-xylosidase, β-mannosidase, α-L-arabinofuranosidase, α-D-glucuronidase, α-galactosidase, acetyl and phenyl esterases, remove side-chains andsubstituents (Biely et al., 1992; Coughlan, 1992; Coughlan and Hazlewood, 1993a).

The interesting biochemical and catalytic properties of cellulases and hemi-cellulases, coupled with their key role in the natural world and their tremendous



Enzymology and OtherCharacteristics of Cellulases

and XylanasesM.K. BHAT1 AND G.P. HAZLEWOOD2

1Food Materials Science Division, Institute of FoodResearch, Norwich Research Park, Colney,

Norwich NR4 7UA, UK; 2Finnfeeds, PO Box 777,Marlborough, Wiltshire SN8 1XN, UK

2



potential for biotechnological applications, have stimulated research aimed atunderstanding the detailed biochemistry, molecular biology and structure–functionrelationships of these enzymes. This chapter summarizes current knowledge ofthe enzymology and other characteristics of cellulases and xylanases from bothfundamental and applied viewpoints. Further background information can be foundin Ryu and Mandels (1980), Wood (1985, 1992a,b), Lamed and Bayer (1988),Coughlan (1992), Coughlan and Hazlewood (1993a,b), Gilbert and Hazlewood(1993), Bayer et al. (1994), Beguin and Aubert (1994), Bhat and Bhat (1997, 1998)and Hazlewood and Gilbert (1998a).

Structure of Cellulose and Hemicellulose

Plant cell walls consist mainly of cellulose (40–45%), hemicellulose (30–35%) andlignin (20–23%) (Ladisch et al., 1983). Cellulose is a linear polymer of glucose linkedby β-1,4-glycosidic bonds, having a simple primary and complex tertiary structures.The repeating unit of cellulose is cellobiose (Fig. 2.1a).

The degree of polymerization (DP; number of glucose residues) per cellulosechain varies from 500 to 14,000 (Marx-Figini and Schultz, 1966). In plant cell walls,the cellulose chains are oriented in parallel with varying degrees of order. In someregions the cellulose chains are highly ordered and strongly hydrogen bonded to formcrystallites, whereas loosely arranged cellulose molecules form the amorphous regions(Fig. 2.1b). Native crystalline cellulose has a structure specified as type I, which canbe converted to type II by alkali treatment (Beguin and Aubert, 1994). These twotypes of cellulose differ in their intrachain hydrogen bonding. In addition, nativecellulose may be composed of two slightly different forms of type I cellulose, calledIα and Iβ, which differ in their intermolecular hydrogen bonding (Atalla and VanderHart, 1984).

The degree of crystallinity of cellulose varies with its origin and its treatmentafter isolation (Hoshino et al., 1992). It ranges from 0% for amorphous andacid-swollen cellulose to nearly 100% for cellulose from Valonia macrophysa (Beguinand Aubert, 1994). Cotton cellulose is approximately 70% crystalline, while thedegree of crystallinity of commercial celluloses varies from 30 to 70% (Fan et al.,1980; Wood, 1988). The crystalline regions of cellulose are rigid and not easilyaccessible to endo-acting cellulases, while the amorphous regions are easily attackedby either dilute acid, endoglucanases or exoglucanases (Sinitsyn et al., 1990). Thus,for the complete hydrolysis of cellulose, either concentrated acid or a completecellulase system capable of attacking both amorphous and crystalline regions isnecessary.

Hemicellulose, the second most abundant plant structural polysaccharide, ispresent in association with cellulose in the walls of most plant species, and can beextracted by alkali. Based on the main sugar residues present in the polymer back-bone, hemicelluloses can be termed xylans, glucomannans, galactans or arabinans.The two main types of hemicelluloses are generally considered to be xylans and

12 M.K. Bhat and G.P. Hazlewood




glucomannans (Timell, 1967; Whistler and Richards, 1970; Stephen, 1983; Puls andSchuseil, 1993; Viikari et al., 1993).

Xylans from annual plants are more heterogeneous than xylans from perennialplants and are designated as arabinoxylans. The two main types of arabinoxylansare: (i) highly branched and without uronic acid substitution – found in cerealendosperms; and (ii) much less branched and substituted with uronic acid and/orwith 4-O-methyl ether and galactose – present in lignified tissues. Arabinoxylans ofgraminaceous plants contain acetic and phenolic acids (ferulic, p-coumaric) whichare esterified to the backbone xylose units and the arabinose side groups, respectively(Fig. 2.1c) (Hartley and Ford, 1989). In addition to enzymes (endoxylanase and

Characteristics of Cellulases and Xylanases 13


Fig. 2.1. Structure of cellulose and arabinoxylan. (a) Glucose and cellobiosereprinted from FEMS Microbiology Reviews, Vol. 13, Beguin, P. and Aubert, J.-P.,The biological degradation of cellulose, pp. 25–58, 1994, with permission fromElsevier Science. (b) Cellulose microfibrils, showing crystalline and paracrystalline(amorphous) regions, reprinted from Biotechnology Bioengineering Symposium,Vol. 5, Cellulose as a Chemical and Energy Resource, Wilke, C.R. (ed.), Physicaland chemical constraints in the hydrolysis of cellulose and lignocellulosicmaterials, pp. 163–181, 1975, with permission from John Wiley & Sons, Inc.,New York. (c) A typical cereal arabino-4-O-methylglucuronoxylan, reprintedfrom Biotechnology and Applied Biochemistry, Vol. 17, Coughlan, M.P. andHazlewood, G.P., β-1,4-D-Xylan-degrading enzyme systems: biochemistry,molecular biology and applications, pp. 259–289, 1993, with permission fromPortland Press, London.



β-xylosidase), which cleave the backbone structure, accessory enzymes likeα-L-arabinofuranosidase, acetyl and phenyl esterases and α-D-glucuronidase arenecessary for the removal of side-chains and the complete hydrolysis of arabinoxylans(Biely et al., 1992).

Enzymology of Cellulases and Xylanases

Source

Cellulases and xylanases are produced by a wide range of bacteria and fungi,including aerobes, anaerobes, mesophiles, thermophiles and extremophiles. Aerobicfungi and bacteria generally produce extracellular cellulases and hemicellulases.Interestingly, anaerobic bacteria (Clostridium thermocellum, C. cellulovorans,Ruminococcus albus, R. flavefaciens, Fibrobacter succinogenes, Acetivibrio cellulolyticus)and anaerobic fungi (Neocallimastix frontalis, N. patriciarum, Piromyces equi) producecellulases in the form of a multienzyme aggregated complex (Groleau and Forsberg,1981; Lamed et al., 1987; Wood, 1992a; Gilbert and Hazlewood, 1993; Beguin andLemaire, 1996; Bhat and Bhat, 1997).

Most of the early studies were carried out on the biochemistry and enzymologyof cellulases from aerobic mesophilic fungi, Trichoderma viride, T. reesei, Penicilliumpinophilum, Sporotrichum pulverulentum, Fusarium solani, Talaromyces emersonii andTrichoderma koningii (Coughlan and Ljungdahl, 1988). In the past two decades,it has been recognized that other microorganisms such as thermophilic fungi(Sporotrichum thermophile, Thermoascus aurantiacus, Chaetomium thermophile,Humicola insolens), mesophilic anaerobic fungi (N. frontalis, N. patriciarum,P. communis, Sphaeromonas communis, P. equi, Orpinomyces sp.), mesophilic andthermophilic aerobic bacteria (e.g. Cellulomonas fimi, Pseudomonas fluorescenssubsp. cellulosa, Cellvibrio sp., Microbispora bispora, Clostridium cellulolyticum andC. cellulovorans), mesophilic and thermophilic anaerobic bacteria (A. cellulolyticus,Bacteroides cellulosolvens, F. succinogens, R. albus, R. flavefaciens, C. thermocellum andC. stercorarium), as well as actinomycetes (Thermomonospora fusca), produce highlyactive cellulase and hemicellulase systems (Bhat and Maheswari, 1987; Aubertet al., 1988; Beguin and Lemaire, 1996; Claeyssens et al., 1998). In addition,hyperthermophilic microorganisms – namely, Thermotoga sp., Pyrococcus furiosus andThermofilum sp., which grow between 85 and 110°C – produce extremely stablecellulolytic and hemicellulolytic enzymes (Simpson et al., 1991; Antranikian, 1994;Winterhalter and Liebl, 1995).

In addition to many of those listed above, other microorganisms such asAspergillus and Cryptococcus produce xylanases and xylan debranching enzymes(Coughlan and Hazlewood, 1993a,b; Viikari et al., 1993). The evidence currentlyavailable suggests that the xylan- and mannan-degrading enzymes of anaerobicmicroorganisms may be produced as integral components of the aggregatedmultienzyme cellulase complexes that are characteristic of these species (Beguin andLemaire, 1996; Hazlewood and Gilbert, 1998b).





Enzyme assays

The measurement of cellulase and xylanase activities is hampered by the nature ofsubstrates used and the complexity of the enzyme systems produced by differentmicroorganisms. In order to overcome these problems, numerous assays have beendeveloped (Wood and Bhat, 1988; Biely et al., 1992). The substrates and themethods used to measure cellulase and xylanase activities are summarized in Table2.1. For more detailed information on assays for accessory enzymes involved in thedegradation of substituted xylans, the reader is referred to Coughlan and Hazlewood(1993a).

Quantitative assays for cellulases and xylanasesThe cellulase system contains mainly endoglucanase (EC 3.2.1.4, 1,4-β-D-glucanglucanohydrolase), exoglucanase or cellobiohydrolase (EC 3.2.1.91, 1,4-β-D-glucancellobiohydrolase) and β-glucosidase or cellobiase (EC 3.2.1.21, β-D-glucosideglucohydrolase). Endoglucanase activity is generally determined by measuring thereducing sugars released from either carboxymethyl (CM-) or hydroxyethyl (HE-)cellulose (Wood and Bhat, 1988). This activity can also be measured by determiningeither the decrease in viscosity of CM-cellulose, the swelling of cotton fibre in alkalior the decrease in turbidity of amorphous cellulose. In addition, substituted,unsubstituted, radio- and reduced end-labelled cello-oligosaccharides have been usedto characterize endoglucanases (Bhat et al., 1990). Interestingly, some endo-glucanases catalyse transferase reactions and act synergistically with cellobiohydrolaseduring the solubilization of crystalline cellulose (Wood et al., 1988; Claeyssens et al.,1990a).

Cellobiohydrolase (CBH; exoglucanase) activity is determined by measuring thereducing sugars released from either Avicel or H3PO4-swollen cellulose (Wood andBhat, 1988). Besides, CBH activity can be measured by determining either therelease of dyed cellobiose from dyed Avicel or the decrease in turbidity of amorphouscellulose. Substituted and unsubstituted cello-oligosaccharides have been used tocharacterize CBHs (Claeyssens et al., 1989).

β-Glucosidase activity is generally determined by measuring the release ofglucose and o-/p-nitrophenol from cellobiose and o-/p-nitrophenyl β-D-glucoside,respectively (Wood and Bhat, 1988). Also, the increase in reducing power of cello-oligosaccharides can be used as a measure of β-glucosidase activity.

Total cellulase activity, comprising endoglucanase, exoglucanase andβ-glucosidase, is measured by determining the solubilization of either cotton fibre,filter paper or Avicel. The release of dyed soluble fragments from dyed Avicel can alsobe used to measure total cellulase activity. It is generally accepted that either cottonfibre or filter paper is the best substrate, but some consider Avicel is ideal formeasuring total cellulase activity. All three substrates contain a high proportion ofcrystalline cellulose and are therefore useful for determining total cellulase activity(Wood and Bhat, 1988).

Xylanase (EC 3.2.1.8) activity is generally determined by measuring the reduc-ing sugars released from xylan by either the Somogyi–Nelson or the dinitrosalicylic







Enzy

mes

Subs

trate

Assa

y

1.Q

uant

itativ

eas

says

Endo

gluc

anas

e(C

M-c

ellu

lase

,en

doce

llula

se)

CM

-cel

lulo

se;h

ydro

xyet

hylc

ellu

lose

Cot

ton;

amor

phou

sce

llulo

se(W

ood,

1988

)

Subs

titut

edan

dun

subs

titut

edce

llo-o

ligos

acch

arid

es

Rel

ease

ofre

duci

ngsu

gars

(Som

ogyi

,195

2);d

ecre

ase

invi

scos

ity(W

ood

and

Bhat

,198

8)Sw

ellin

gin

alka

li(M

arsh

etal

.,19

53);

rele

ase

ofre

duci

ngsu

gars

(Som

ogyi

,195

2);

decr

ease

intu

rbid

ity(N

umm

i eta

l.,19

81)

Incr

ease

inre

duci

ngsu

gars

(Som

ogyi

,195

2);a

naly

sis

ofpr

oduc

tsby

HPL

C(B

hat e

tal.,

1990

)

Cel

lobi

ohyd

rola

se(e

xogl

ucan

ase,

exoc

ellu

lase

,Av

icel

ase)

Avic

el;h

ydro

cellu

lose

and

dyed

Avic

elAm

orph

ous

cellu

lose

Subs

titut

edan

dun

subs

titut

edce

llo-o

ligos

acch

arid

es

Rel

ease

ofre

duci

ngsu

gars

(Som

ogyi

,195

2);d

yed

cello

bios

e(W

ood

and

Bhat

,198

8)R

elea

seof

redu

cing

suga

rs(S

omog

yi,1

952)

;dec

reas

ein

turb

idity

(Num

mi e

tal.,

1981

)In

crea

sein

redu

cing

suga

rs(S

omog

yi,1

952)

;rel

ease

ofp-

nitro

phen

olfro

mp-

nitro

phen

yl-β

-D-c

ello

bios

ide

(Des

hpan

deet

al.,

1984

);an

alys

isof

prod

ucts

byH

PLC

(Bha

t eta

l.,19

90)

β-G

luco

sida

se(c

ello

bias

e)C

ello

bios

eC

ello

-olig

osac

char

ides

o-or

p-N

itrop

heny

l-β-D

-glu

cosi

des

Rel

ease

ofgl

ucos

e(W

ood

and

Bhat

,198

8)In

crea

sein

redu

cing

suga

rs(S

omog

yi,1

952)

Rel

ease

ofo-

orp-

nitro

phen

ol(W

ood

and

Bhat

,198

8)

Tota

lcel

lula

seC

otto

n

Filte

rpap

er;h

ydro

-cel

lulo

se;A

vice

l;So

lka

Floc

Dye

dAv

icel

Cel

lulo

sere

sidu

e(W

ood,

1969

);re

leas

eof

redu

cing

suga

rs,w

eigh

tlos

san

dlo

ssin

tens

ilest

reng

th(W

ood

and

McC

rae,

1972

)R

elea

seof

redu

cing

suga

rs(S

omog

yi,1

952;

Mille

r,19

59)

Rel

ease

ofso

lubl

edy

edfra

gmen

ts(W

ood

and

Bhat

,198

8)

Xyla

nase

(end

oxyl

anas

e)Bi

rchw

ood

xyla

n;oa

tspe

ltxy

lan

Inso

lubl

exy

lan

Arab

inox

ylan

;CM

-xyl

anR

BB-x

ylan

;Ost

azin

eBr

illian

tRed

(OBR

)-xyl

an[1

-3 H]-l

abel

led,

unsu

bstit

uted

and

subs

titut

edxy

lo-o

ligos

acch

arid

es

Rel

ease

ofre

duci

ngsu

gars

(Som

ogyi

,195

2;M

iller,

1959

;Bai

ley

etal

.,19

92)

Dec

reas

ein

turb

idity

(Num

mi e

tal.,

1985

)D

ecre

ase

invi

scos

ity(S

engu

pta

etal

.,19

87)

Rel

ease

ofR

BB-o

rOBR

-dye

dfra

gmen

ts(B

iely

etal

.,19

85)

Anal

ysis

ofpr

oduc

tsby

HPL

Can

dsu

bseq

uent

radi

oact

ive

coun

ting,

ifne

cess

ary

(Bha

tet

al.,

1990

;Bie

lyet

al.,

1992

)

Tabl

e2.

1.Su

bstra

tes

and

assa

ysus

edfo

rdet

erm

inin

gth

ece

llula

sean

dxy

lana

seac

tiviti

es.





2.Q

ualit

ativ

eas

says

Endo

gluc

anas

eC

M-c

ellu

lose

;RBB

-CM

-cel

lulo

se;

OBR

-hyd

roxy

ethy

lcel

lulo

seM

ethy

lum

bellif

eryl

- β-D

-cel

lobi

osid

e(M

eUm

bGlc

2);m

ethy

lum

bellif

eryl

- β-D

-cel

lotri

osid

e(M

eUm

bGlc

3)

Det

ectio

nby

usin

gC

ongo

Red

and

zone

clea

ring

(Cou

ghla

n,19

88)

Rel

ease

ofm

ethy

lum

bellif

eron

e(M

eUm

b)an

dde

tect

ion

byU

V(v

anTi

lbeu

rgh

etal

.,19

88)

Cel

lobi

ohyd

rola

se(e

xogl

ucan

ase)

RBB

-Avi

cel

p-N

itrop

heny

l-β-D

-cel

lobi

osid

eZo

necl

earin

g(C

ough

lan,

1988

)R

elea

seof

p-ni

troph

enol

(van

Tilb

eurg

het

al.,

1988

)

β-G

luco

sida

sep-

Nitr

ophe

nyl-β

-D-g

luco

side

;met

hylu

mbe

llifer

yl- β

-D-g

luco

side

(MeU

mbG

lc)

Det

ectio

nof

p-ni

troph

enol

(Bha

teta

l.,19

93)o

rMeU

mb

byU

V(v

anTi

lbeu

rgh

etal

.,19

88)

Xyla

nase

Solu

ble

xyla

n;R

BB-x

ylan

;OBR

-xyl

anM

ethy

lum

bellif

eryl

-β-D

-xyl

obio

side

(MeU

mbX

yl2);

met

hylu

mbe

llifer

yl- β

-D-x

ylot

riosi

de(M

eUm

bXyl

3)

Det

ectio

nby

usin

gC

ongo

Red

and

zone

clea

ring

(Bie

lyet

al.,

1992

)D

etec

tion

ofM

eUm

bby

UV

(Bie

lyet

al.,

1992

;Kal

ogia

nnis

etal

.,19

96)



acid (DNS) method (Somogyi, 1952; Miller, 1959; Bailey et al., 1992). The DNSmethod tends to give a higher value than the Somogyi–Nelson method. Nevertheless,xylanase activity determined by the DNS method, standardized with respect to aparticular xylan, is very reproducible (Biely et al., 1992). Based on a round-robin testthat involved 20 laboratories, Bailey et al. (1992) proposed a suitable procedure forpreparing xylan and measuring xylanase activity. In addition, a viscometric methodusing a soluble xylan (arabinoxylan or carboxymethyl xylan) has been used forthe determination of xylanase activity in samples containing high background sugarvalues (Sengupta et al., 1987). Although the above methods detect the ability of anenzyme sample to hydrolyse xylan, they do not reveal the type of products released,which may be important when evaluating the ability of an enzyme to degrade xylan.Hence, the hydrolysis products should be analysed to obtain detailed information onthe mode of action of the xylanase in question.

Another specific assay, which can be used to determine xylanase activity in thepresence of either large amounts of reducing sugars or viable cells utilizing xylanfragments (Biely, 1985), involves the use of soluble and covalently dyed xylan(Remazol Brilliant Blue) (RBB-xylan). Although this assay is highly sensitive totemperature and ionic strength, interlaboratory tests revealed that the release oflow molecular weight dyed fragments is a useful alternative for measuring xylanaseactivity (Biely et al., 1992). A nephelometric (turbidometric) assay using insolublexylan has also been found to be suitable for determining xylanase activity (Nummiet al., 1985). Furthermore, [1-3H]-labelled and substituted xylo-oligosaccharideshave been used to characterize xylanases (Biely et al., 1992; Bennett et al., 1998).

Qualitative assays for cellulases and xylanasesQualitative assays have been developed either to select microbial strains producinghigh levels of cellulases and xylanases, or to identify and/or characterize theseenzymes in a given sample. Insoluble xylan and RBB-xylan are ideal substratesto select microorganisms producing xylanase on solid agar medium (Sprey and Lam-bert, 1983; Farkas et al., 1985). Similarly, methods using either RBB-CM-celluloseor CM-cellulose stained with Congo Red can be used to select microorganismsproducing endoglucanase activity (Kluepfel, 1988).

The capacity of Congo Red to complex with either polymeric xylan or CM-cellulose, but not with small oligosaccharide products, is conveniently used to detectxylanase and endoglucanase activities in zymograms after the fractionation of proteinmixtures by SDS/IEF-PAGE (Coughlan, 1988). RBB-xylan and RBB-CM-cellulosecan be used for similar purposes (Biely et al., 1985). The advantages of the lattermethod are: (i) that the dyed fragments released from RBB-xylan/CM-cellulosediffuse from the detection gel into the separating gel and further help to identifythe position of a xylanase or endoglucanase, which can subsequently be eluted; and(ii) the hydrolysis of the dyed substrate can be visually followed and the reactionterminated when necessary. The use of these substrates facilitates the identification ofmultiple forms of xylanase or endoglucanase produced by different microorganisms.Likewise, chromogenic and fluorogenic cello- and xylo-oligosaccharides have beensuccessfully used to identify multiple forms of cellulases and xylanases after the





separation of crude enzyme mixtures by SDS/IEF-PAGE (Biely et al., 1992;Kalogiannis et al., 1996).

Substrate specificity

EndoglucanasesEndoglucanases specifically cleave the internal β-1,4-glycosidic bonds of amorphous,swollen and substituted celluloses as well as cello-oligosaccharides. These enzymes aregenerally inactive towards crystalline cellulose and cellobiose. Some endoglucanasesattack barley glucan with mixed β-1,3 and β-1,4 linkages (Petre et al., 1986). Usingsubstituted, unsubstituted, [1-3H]-labelled and reduced cello-oligosaccharides, Bhatet al. (1990) demonstrated a marked difference in the substrate specificities ofendoglucanases purified from P. pinophilum. Thus, endoglucanases III and IV wereactive on cellotriose and higher cello-oligosaccharides, whereas endoglucanases II, Vand I required at least four, five and six glucose residues, respectively. Although suchvariation in substrate specificity of endoglucanases was unexpected, it was speculatedthat microorganisms secrete multiple endoglucanases with a wide range of substratespecificities to effect efficient hydrolysis of complex cellulosic substrates.

CBHs (cellobiohydrolases; exoglucanases)Like endoglucanases, the CBHs are highly active on amorphous and swollencelluloses, but degrade crystalline cellulose and cello-oligosaccharides rather poorly(Wood and Bhat, 1988). These enzymes are specific for β-1,4 linkages of thecellulose chain, but are inactive on cellobiose, CM- and hydroxyethyl-celluloses. Ingeneral, CBHs attack cellulose chains from the non-reducing end and releasecellobiose (Wood et al., 1988). Recent kinetic studies and high-resolution structuraldata confirmed that there are two classes of CBHs (Barr et al., 1996). The class oneenzymes (e.g. CBH I from T. reesei and two exoglucanases E4 and E6 from T. fusca)hydrolyse the cellulose chain preferentially from the reducing end, while class twoCBHs (e.g. CBH II from T. reesei and E3 from T. fusca) release cellobiose specificallyfrom the non-reducing end (Barr et al., 1996; Teeri, 1997).

b-Glucosidasesβ-Glucosidases can be classified as either aryl β-D-glucosidases (hydrolysingexclusively aryl-β-D-glycosides), cellobiases (hydrolysing diglucosides and cello-oligosaccharides) or β-glucosidases with broad substrate specificities. Mostβ-glucosidases characterized so far show broad substrate specificities and hydrolysearyl and alkyl β-D-glycosides and β-1,1-, β-1,2-, β-1,3-, β-1,4- and β-1,6-linkeddiglucosides, as well as substituted and unsubstituted cello-oligosaccharides(Bhat et al., 1993; Christakopoulos et al., 1994a). Interestingly, an intracellularβ-glucosidase from S. thermophile was found to be an aryl-β-glucosidase, whereas twoextracellular β-glucosidases from the same organism hydrolysed only cellobiose(Bhat et al., 1993). Some β-glucosidases showed activity towards H3PO4-swollenand CM-celluloses (Sadana et al., 1988), but most β-glucosidases are inactive





towards these and other polymeric substrates such as Avicel, filter paper and cotton(Woodward and Wiseman, 1982).

XylanasesIn general, xylanases are specific for the internal β-1,4 linkages of polymericxylan and are designated as endoxylanases. Based on their action on differentpolysaccharides, endoxylanases have been classified as either specific or non-specific(Coughlan, 1992; Coughlan et al., 1993). The specific endoxylanases are active onxylans with only β-1,4 linkages, whereas non-specific endoxylanases hydrolyseβ-1,4-linked xylans, β-1,4 linkages of mixed xylans and other β-1,4-linked polymerssuch as CM-cellulose. In fact, the determination of the kinetic constants ratio kcat/Km

for xylan and CM-cellulose, for a particular non-specific xylanase, should confirmwhether the enzyme in question is a xylanase or a cellulase. Generally, endoxylanaseactivity and affinity for xylo-oligosaccharides decrease with decreasing DP (Coughlanet al., 1993). Most endoxylanases are specific for unsubstituted xylosidic linkagesof xylans and release both substituted and unsubstituted xylo-oligosaccharides. Incontrast, some endoxylanases are specific for xylosidic linkages adjacent tosubstituted groups in the main chain xylan. For example, two endoxylanases fromAspergillus niger (pI 8.0 and 9.6) showed little or no action on xylo-oligomers orxylans from which the arabinose substituents were removed (Frederick et al., 1985).Furthermore, the endoxylanase with pI 9.6 cleaved the xylan chain only in thevicinity of 4-O-methylglucuronic acid, and this substitution was reported to be anabsolute requirement for the action of the enzyme (Nishitani and Nevins, 1991).Nevertheless, both enzymes failed to release any substituent as a free product, whichindicated that these substitutions are necessary for the proper orientation of thesubstrate in the active site.

Most endoxylanases, active on mixed xylan (rhodymenan with β-1,3 and β-1,4linkages), are specific for the β-1,4 linkages (Coughlan, 1992). Also, endoxylanaseshave been grouped either as debranching or non-debranching based on their abilityto release arabinose in addition to hydrolysing the main chain (Coughlan et al.,1993). Although some endoxylanases purified to homogeneity hydrolysed bothmain-chain xylan and arabinose side-chains (Matte and Forsberg, 1992), there isalways a question as to whether the release of arabinose is due to an intrinsic propertyof the enzyme or is due to the presence of a trace contaminant.

Kinetics and subsite mapping

Kinetic constants for the hydrolysis of CM-cellulose and xylan by purifiedendoglucanases and xylanases have been determined (Bhat et al., 1989; Coughlanet al., 1993). It is often difficult to determine the kinetic constants of these enzymesusing polymeric substrates. Therefore, substituted (chromogenic or fluorogenic)and [1-3H]-labelled cello- and xylo-oligosaccharides have been prepared andconveniently used for this purpose (Bhat et al., 1990; Claeyssens et al., 1990a,b; Bielyet al., 1992). When an enzyme attacks more than one glycosidic bond in a particular





oligosaccharide, the kinetic constants (Km and kcat) for the cleavage of all glycosidicbonds must be determined in order to understand the mode of action of the enzyme,as well as to identify the preferred site of attack. Using this approach, the numberof glycosyl-binding sites present in different endoglucanases, CBHs and xylanaseshave been determined biochemically (Claeyssens et al., 1989; Coughlan et al., 1993;Bhat et al., 1994). Detailed information on glycosyl-binding sites and active sitearchitecture of the above enzymes has also been derived by determining theirthree-dimensional structures with bound substrates or substrate analogues in theactive site (Davies et al., 1995; Sakon et al., 1997). In contrast, the kinetic constantsof β-glucosidases have been determined using either cellobiose or pNPC (Bhat et al.,1993; Christakopoulos et al., 1994a), and have been shown to depend on enzymesource and preference for alkyl or aryl-glycosides (Woodward and Wiseman, 1982;Bhat et al., 1993).

Inhibition

Cellulases and xylanases are often inhibited by the presence of high concentrations oftheir hydrolysis products. For example, most CBHs are inhibited by cellobiose (Woodand McCrae, 1986), even though cellobiose (> 10 mM) stimulates the activity ofCBH II from P. pinophilum towards H3PO4-swollen cellulose. Similarly, endoglu-canases are inhibited by cellobiose at or above 100 mM (Bhat et al., 1989). However,glucose (up to 100 mM) showed little effect on many CBHs and endoglucanases(Wood et al., 1988; Bhat et al., 1989). In contrast, β-glucosidases are inhibited byglucose and other mono- and disaccharides, such as xylose, fucose, galactose, maltose,lactose and meliobiose (Bhat et al., 1993). Besides, nojirimycin and gluconolactoneare known to be potent inhibitors of β-glucosidases (Bhat et al., 1993).

Like cellulases, endoxylanases are believed to be inhibited by high concentra-tions of xylobiose but not by xylose. In general, the activity of cellulases and xylanasesis neither activated nor inhibited by metal ions and reducing agents, though theaggregated cellulase systems of anaerobic bacteria may require divalent cations andreducing agents for their activity towards crystalline cellulose, but not towardsswollen and substituted celluloses (Lamed and Bayer, 1988).

Transferase activity

Many cellulases and endoxylanases catalyse both hydrolysis and transferase (ortransglycosylase) reactions, especially in the presence of high concentrations ofoligomeric substrates or alcohol (Bhat et al., 1990, 1993; Christakopoulos et al.,1994a–c). The essential difference between hydrolysis and transferase reactions isthat, in the former, water acts as an acceptor of the glycosyl moiety, whereas in thelatter the acceptor is an alcohol or a sugar molecule.

Transferase reactions have considerable significance from both fundamental andapplied viewpoints. It is generally accepted that the transferase products of cellobiose





and xylobiose could be the true inducers of microbial cellulases and xylanases (Biely,1993; Bhat and Bhat, 1997). In addition, enzymatic transglycosylation can be usedto synthesize high value alkylglycosides and oligosaccharides with desired linkages,which have uses both for research and for commercial applications (Christakopouloset al., 1994a–c).

Mode of action

Fungal cellulaseAll endoglucanases attack swollen and substituted celluloses and the amorphousregions of cellulose randomly and release glucose, cellobiose and cello-oligosaccharides (Table 2.2) (Wood, 1985; Coughlan and Ljungdahl, 1988). The



Enzyme Mode of action

Endoglucanase or endocellulase(1,4-β-D-glucan glucanohydrolase;EC 3.2.1.4)

–G–G–G–G–G–G–G–G–G–G–G–↑ ↑

Cleaves 1,4-β linkages at random (Wood and Bhat, 1988)

Cellobiohydrolase (CBH) orexocellulase (1,4-β-D-glucancellobiohydrolase; EC 3.2.1.91)

G–G–G–G–G–G–G–G–G–G–G–↑ (type I) ↑ (type II)

Releases cellobiose from both reducing (type II) andnon-reducing ends (type I) (Wood et al., 1988; Vrsanskaand Biely, 1992; Barr et al., 1996; Gilkes et al., 1997)

Exoglucanase or glucohydrolase(1,4-β-D-glucan glucohydrolase;EC 3.2.1.74)

G–G–G–G–G–G–G–G–G–G–↑

Releases glucose from the non-reducing end (McHale andCoughlan, 1980; Wood and McCrae, 1982)

β-Glucosidase or cellobiase(β-D-glucoside glucohydrolase;EC 3.2.1.21)

G–G; G–G–G–G; G–G–G–G–G↑ ↑ ↑ ↑ ↑

Releases glucose from cellobiose and hydrolyses short-chaincello-oligosaccharides from both reducing and non-reducingends by releasing one glucose unit at a time (Wood andBhat, 1988; Christakopoulos et al., 1994a)

Xylanase or endoxylanase(1,4-β-D-xylan xylanohydrolase;EC 3.2.1.8)

S S| |

X–X–X–X–X–X–X–X–X–X–X–X–X–↑ | ↑

SCleaves 1,4-β linkages of xylan at random with preference to

unsubstituted regions (Biely et al., 1992; Coughlan et al.,1993; Coughlan and Hazlewood, 1993a)

Table 2.2. Mode of action of cellulases and xylanase.



CBHs hydrolyse H3PO4-swollen cellulose and Avicel sequentially by removingcellobiose units from either reducing or non-reducing ends of the cellulose chain(Wood et al., 1988; Vrsanska and Biely, 1992; Barr et al., 1996). Endoglucanaseand CBH act synergistically to effect extensive hydrolysis of crystalline cellulose.Subsequently β-glucosidase completes the hydrolysis by converting the resultantcello-oligosaccharides and cellobiose to glucose (Wood, 1985).

The modes of action of endoglucanase, CBH and β-glucosidase from severalsources have been extensively studied using substituted, unsubstituted and labelledcello-oligosaccharides (van Tilbeurgh et al., 1988; Bhat et al., 1990; Claeyssens andHenrissat, 1992). Based on these studies, it was concluded that most endoglucanasesattack internal glycosidic bonds of cello-oligosaccharides and release mainlycellobiose and cellotriose, while CBHs hydrolyse the second glycosidic bond fromeither the reducing or non-reducing end of the cello-oligosaccharides (Table 2.2).However, β-glucosidase sequentially removes one glucose unit from either thereducing end, the non-reducing end or both ends (Table 2.2).

The viscosity of a cellulose solution is directly related to the DP of individualcellulose chains. An endoglucanase that attacks the cellulose chain randomlyproduces the largest decrease in the viscosity of a cellulose solution per unit increasein reducing power. Based on the relationship between viscosity reduction andthe increase in reducing end groups, the multiple endoglucanases from T. koningiiand P. pinophilum were classified as more or less randomly acting enzymes(Wood and McCrae, 1978; Bhat et al., 1989). Using a similar approach, the mode ofaction of endoglucanases and CBHs (exoglucanases) from C. fimi, P. pinophilum andC. thermocellum has been clarified (Gilkes et al., 1984; Bhat et al., 1989, 1994; Woodet al., 1989).

Some fungi, such as P. pinophilum and T. emersonii, produce an exoglucanasethat catalyses the removal of glucose from the non-reducing end of cellodextrins,but does not interact cooperatively with endoglucanases during the hydrolysis ofcrystalline cellulose (McHale and Coughlan, 1980; Wood and McCrae, 1982).

Anaerobic fungi such as N. frontalis, N. patriciarum and P. equi use a differentmechanism to degrade cellulose when compared with aerobic fungi (Wood, 1992a).It has been reported that N. frontalis produces a multicomponent enzyme complexcalled crystalline cellulose-solubilizing factor (CCSF), which has a molecular massof 700 kDa and consists of a number of subunits with molecular mass rangingfrom 68 to 135 kDa (Wood, 1992a). Functionally, CCSF resembles the cellulasesfrom the aerobic fungi in solubilizing crystalline cellulose, since it acts synergisticallywith an endoglucanase and a β-glucosidase of N. frontalis which are not part ofCCSF (Wood, 1992a). Furthermore, it was demonstrated that CCSF of N. frontalisconsists of endoglucanase, β-glucosidase and perhaps also CBH. Although themode of action of CCSF on cellulose is not clearly understood, it is believed thatall subunits of CCSF act cooperatively to degrade cellulose (Wood et al., 1994).More recent molecular biology studies of individual cellulases and hemicellulasesfrom anaerobic fungi support the view that these organisms produce an aggregatedcellulase complex analogous to the cellulosome of C. thermocellum (Hazlewood andGilbert, 1998b).





Bacterial cellulaseBacterial cellulases are known to adopt different mechanisms for the hydrolysis ofcellulose. For example, the aerobes Cellulomonas, Pseudomonas, Thermoactinomycetes,T. fusca and Microbispora and the anaerobe C. stercorarium produce a cellulase systemsimilar to that of aerobic fungi, and degrade cellulose by the cooperative interactionof the different cellulase components (Beguin et al., 1992; Wood, 1992b; Gilbertand Hazlewood, 1993). In contrast, the anaerobic thermophilic bacteriumC. thermocellum degrades crystalline cellulose very effectively by means of a highmolecular mass multienzyme complex called the cellulosome (Lamed and Bayer,1988; Beguin and Lemaire, 1996). The cellulosome of C. thermocellum has beenextensively studied and used as a model system for understanding how anaerobicbacteria, including rumen bacteria, degrade cellulose (Wood, 1992a). Theproduction of cellulosomes is not restricted to C. thermocellum alone. Various otheranaerobic bacteria, such as R. albus, R. flavefaciens, F. succinogenes, A. cellulolyticus andC. cellulovorans also produce cellulosomes on their cell surface which possess similarproperties to the cellulosome from C. thermocellum (Groleau and Forsberg, 1981;Lamed et al., 1987; Beguin and Lemaire, 1996).

Cellulosomes of C. thermocellum degrade crystalline cellulose extensively inthe presence of Ca2+ and DTT (Lamed and Bayer, 1988). According to currentunderstanding, the cellulosome contains a large molecular mass, non-catalyticsubunit termed scaffoldin (CipA, S1 or SL) which possesses a cellulose-bindingdomain (CBD) and numerous duplicated attachment sites called cohesins (Bayeret al., 1994; Beguin and Lemaire, 1996; Beguin and Alzari, 1998; Bhat and Bhat,1998). In addition to catalytic domains, the enzymatic subunits of the cellulosomeeach contain a highly conserved duplicated docking domain termed a dockerin,which interacts with the cohesins of the scaffoldin polypeptide (Bayer et al., 1994).Some of the enzymatic subunits also contain a CBD. The exact mechanism by whichthe cellulosome achieves cellulolysis is unclear, but it is evident that the efficiency ofthe multienzyme complex is a function of its quaternary structure, and is dependenton both endoglucanase/CBH synergism and clustering of the complex on to thesubstrate surface (Beguin and Lemaire, 1996).

Fungal and bacterial endoxylanasesThe mode of action of endoxylanases has been extensively studied (Coughlan, 1992;Coughlan and Hazlewood, 1993a,b; Biely et al., 1997). In general, these enzymescleave the internal β-1,4 linkages of the xylan backbone and release mainly xylobiose,xylotriose and substituted oligomers having two to four residues (Table 2.2). It hasbeen reported that most endoxylanases cleave the xylan backbone leaving thesubstituent at the non-reducing end of the xylosyl chain or the oligosaccharide(Dekker, 1985). Moreover, the presence of glucopyranosyl uronic acid stericallyhinders the hydrolysis of the second and third xylosidic linkages to the right of thebranch points, by the xylanases from S. dimorphosporum and Trametes hirsuta(Kubackova et al., 1978; Comtat and Joseleau, 1981). Nevertheless, the xylanasesfrom A. niger and C. sacchari were found to cleave the substituent on the reducingend and in the middle of the oligosaccharide chain, respectively (Dekker, 1985).





By analysing the structures of the substituted products released during theextensive hydrolysis of different xylans by purified endoxylanases, it was possible todraw the following conclusions about their modes of action.

1. An endoxylanase from Streptomyces sp. yielded arabinoxylotrioside (IV, Fig. 2.2) asthe smallest product from wheat arabinoxylan. This revealed that the minimal spatialrequirement for the attack of β-1,4 linkage is two adjacent unsubstituted xylose units.2. An endoxylanase from M. verrucaria liberated arabinoxylobiose (I, Fig. 2.2) fromwheat straw xylan, indicating that this enzyme attacks the glycosidic linkage adjacentto the substitution.3. Hydrolysis of spear grass hemicellulose by an endoxylanase from Ceratocystisparadoxa yielded products I and III (Fig. 2.2), while a xylanase from A. niger liberatedproducts II and V (Fig. 2.2).4. An extensive hydrolysis of larchwood glucuronoxylan by an endoxylanase fromT. aurantiacus yielded xylose, xylobiose and aldotetrauronic acid (VII, Fig. 2.2). Thisenzyme did not cleave main-chain β-1,4 linkages between the residues carrying thesubstituent and the adjacent xylose on either side.5. Action of endoxylanase from S. dimorphosporum on redwood arabinoglucuro-noxylan released products I, III, VI and VIII (Fig. 2.2). This endoxylanase appeared torequire at least a xylotriose group (IX) with an unsubstituted (at O-2 position) xylosylresidue at the non-reducing end and an unsubstituted (at O-2 and O-3 positions)central xylosyl residue, while the reducing end could be substituted or unsubstituted(Fig. 2.2).

Based on amino acid sequence similarities and hydrophobic cluster analysis,Henrissat and Bairoch (1993, 1996) grouped endoxylanases in glycosyl hydrolasefamilies 10 and 11. Family 10 includes mainly acidic and high molecular massendoxylanases, while basic and low molecular mass endoxylanases belong to family11. Studies on the mode of action of endoxylanases from these two families revealedthat: (i) endoxylanases of family 10 hydrolyse heteroxylans (substituted xylan) andhomoxylans (rhodymenan with β-1,4 and β-1,3 linkages) to a higher degree thanthose of family 11; (ii) only endoxylanases of family 10 are capable of cleavingthe glycosidic linkages of xylan either closer or adjacent to substituents, such asmethylglucuronic acid (MeGlcA) and acetic acid; and (iii) family 10 endoxylanasesrelease smaller oligosaccharides from glucuronoxylan, acetylxylan and rhodymenanthan family 11 (Biely et al., 1997). In addition, it was reported that the endoxylanasesof family 10 require two unsubstituted xylose residues between the branchpoints, whereas family 11 endoxylanases require at least three unsubstitutedxylose residues in a sequence (Fig. 2.3a–c) (Biely et al. 1997). The predicted modeof action of endoxylanases from families 10 and 11 on rhodymenan is shown inFig. 2.3b.

Studies of the mode of action and biochemical characteristics of endoxylanases Iand III from Aspergillus awamori indicated that they belong to families 10 and 11,respectively (Kormelink et al., 1993). Thus, endoxylanase I (family 10) cleaves β-1,4linkages adjacent to arabinose substitution as well as internal β-1,4 linkages, whereasendoxylanase III (family 11) cleaves only the internal β-1,4 linkages of cereal





arabinoxylan (Fig. 2.3c) (Kormelink et al., 1993). The isolation and identificationof methylglucuronic acid and arabinose-linked oligosaccharides released frommethylglucuronoxylan and arabinoxylan by endoxylanases of family 10 revealed thatthe substitution at position 3 could cause greater steric hindrance of enzyme attack



Fig. 2.2. Chemical structures of substituted products released by differentendoxylanases. Reprinted from Xylans and Xylanases, Visser, J., Beldman, G.,Kusters-van Someren, M.A. and Voragen, A.G.J. (eds), Coughlan, M.P., Towardsan understanding of the mechanism of action of main chain-hydrolysing xylanases,pp. 111–139, 1992, with permission from Elsevier Science.



than at position 2. Furthermore, only endoxylanases of family 10 releasedmonoacetylated xylobiose from acetylxylan, with the exception of a family 11endoxylanase from T. reesei (Biely et al., 1997).

Endoxylanases from families 10 and 11 can be distinguished further, based ontheir mode of action on substituted and unsubstituted oligosaccharides. In general,the family 10 endoxylanases hydrolyse xylotriose and xylotetraose more rapidly thanthose from family 11. Also, endoxylanases of family 10 hydrolyse pNPC, pNPXyl2,pNPXylGlc, MeUmbXyl2 and MeUmbXylGlc, while endoxylanases of family 11hydrolyse only pNPXyl2 and MeUmbXyl2. Thus, the ability to cleave β-1,4 linkagesbetween xylose and glucose residues is one of the major differences betweenendoxylanases from families 10 and 11. Based on this information, Biely et al. (1997)concluded that if a model β-1,4-glucoxylan is available, it is most likely that theendoxylanases of family 10 will cleave certain β-1,4-glucopyranosyl linkages, butthose from family 11 will not.



Fig. 2.3. Mode of action of endoxylanases from families 10 and 11 on (a) 4-O-methyl-D-glucurono-D-xylan, (b) rhodymenan, and (c) cereal L-arabino-D-xylan.The arrows indicate the sites of attack on different substrates. Reprinted fromJournal of Biotechnology, Vol. 57, Biely, P., Vrsanska, M., Tenkanen, M. andKluepfel, D., Endo-β-1,4-xylanase families: differences in catalytic properties,pp. 151–166, 1997, with permission from Elsevier Science.



Synergism

Between cellulasesSynergism is an enhanced effect of two or more enzymes when acting cooperatively,compared with their additive effect. Giligan and Reese (1954) first demonstratedsynergism between cellulase components during the hydrolysis of cellulose.Subsequently several research groups have demonstrated synergism between endo-and exoglucanases during the solubilization of crystalline cellulose (Wood et al.,1988, 1989; Klyosov, 1990; Bhat et al., 1994). Five different types of synergism havebeen reported between fungal cellulase components: (i) a non-hydrolytic proteincalled C1 and endoglucanase (Reese et al., 1950); (ii) β-glucosidase and eitherendoglucanase or CBH (Eriksson and Wood, 1985); (iii) two immunologicallyrelated and distinct CBHs (Wood and McCrae, 1986); (iv) endoglucanase and CBHfrom either the same or different microorganisms (Wood et al., 1989); and (v)between two endoglucanases (Klyosov, 1990). In addition, synergism betweenbacterial and fungal cellulases, as well as between the subunits of the C. thermocellumcellulosome, has been reported (Bhat et al., 1994; Wood et al., 1994).

Synergism between fungal cellulase components has been studied mostextensively (Coughlan and Ljungdahl, 1988; Wood et al., 1988, 1989; Klyosov,1990). The most interesting types of synergism are between: (i) endoglucanase/exoglucanase (CBH); (ii) exoglucanase (CBH)/exoglucanase (CBH); and (iii)endoglucanase/endoglucanase. An early model postulated by Wood and McCrae(1972) suggested that cellulose chains cleaved by endoglucanase become the substratefor exoglucanase, and that these two enzymes cooperatively degrade the cellulose.However, this model did not explain the synergism seen between two differentCBHs, or the inability of CBH to synergize with endoglucanases from differentmicroorganisms. Subsequently, using highly purified endoglucanases and CBHsfrom P. pinophilum, it was demonstrated that only two endoglucanases (EGIII andEGV), which were strongly adsorbed on to cellulose, best synergized with CBHs Iand II (Fig. 2.4; Wood et al., 1989). The authors explained the observed synergismbetween CBHs I and II and endoglucanases in terms of the different stereo-specificities of these two enzymes.

The attack of a cellulose chain by a stereospecific endoglucanase will generateonly one of two possible types of non-reducing ends, which will be hydrolysed by astereospecific CBH. The successive removal of cellobiose by CBH will exposeanother chain-end of a different configuration, which will be attacked by the otherstereospecific CBH. Hydrolysis of the two chain-ends by two CBHs actingrandomly, together with attack of the cellulose chains by another stereospecificendoglucanase to generate a reducing end of a different configuration, wouldfacilitate the synergism observed between these enzymes. Nevertheless, it was arguedthat the above synergism could be due to the strong adsorption of CBHs andendoglucanases on to cellulose (Klyosov, 1990). Furthermore, Klyosov (1990)reported that two endoglucanases which adsorbed strongly on to cellulose degradedthe cellulose synergistically. Although the adsorption appeared to be important, it isdifficult to explain the contribution of adsorption towards synergistic interaction





between endoglucanases and CBHs. Recent studies (Barr et al., 1996; Teeri, 1997)revealed that there are two classes of CBHs (exoglucanases) which attack cellulosechains from both reducing and non-reducing ends. Based on these results, it wasspeculated that the synergism observed between CBH/CBH (exo/exo) was due totheir ability to expose new hydrolysis sites to each other, as well as their ability to actfrom reducing and non-reducing ends (Barr et al., 1996). Thus, two CBHs actingfrom reducing and non-reducing ends and an endoglucanase appeared to be essentialfor the effective hydrolysis of crystalline cellulose. The presence of anotherendoglucanase with different substrate specificity would increase the synergisticefficiency further, as was observed in the case of the P. pinophilum cellulase system(Wood et al., 1989).

Recent studies of the C. thermocellum cellulase system revealed that twocellulosomal exoglucanases (S5 and S8 subunits) and an endoglucanase (S11 subunit),together with the S1 (scaffoldin) subunit, are essential for maximum synergismduring the hydrolysis of crystalline cellulose (Bhat et al., 1994). Furthermore, the S1

subunit mediated the formation of an enzyme complex with the S5, S8 and S11



Fig. 2.4. Synergism between P. pinophilum CBHs (I and II) and endoglucanases(EI to EV) in solubilizing cotton fibre. Reproduced from Biochemical Journal,Vol. 260, Wood, T.M., McCrae, S.I. and Bhat, K.M., The mechanism of fungalcellulase action. Synergism between enzyme components of P. pinophilumcellulase in solubilizing hydrogen bond-ordered cellulose, pp. 37–43, 1989,with permission from Portland Press, London.



subunits (Bhat et al., 1994). Using recombinant scaffoldin (CipA) polypeptides andthe endoglucanase CelD, it was demonstrated that a truncated scaffoldin polypeptidewith a CBD and a single cohesin domain was adequate for the maximum activity ofCelD towards Avicel (Kataeva et al., 1997; Beguin et al., 1998). It was also reportedthat linkage of the CelD–CipA complex with the CBD was critical for maximumsynergism during Avicel solubilization, rather than the clustering of catalyticdomains (Kataeva et al., 1997; Beguin et al., 1998). These results strongly indicatethat, in the case of the aggregated C. thermocellum cellulase system, assembly of anenzyme complex is crucial for the maximum synergistic interaction of subunitsduring the solubilization of crystalline cellulose. Other studies using recombinantcellulosomal components have confirmed the important role of the CipA CBD inefficient cellulolysis (Ciruela et al., 1998).

Between xylan-degrading enzymesEfficient and complete hydrolysis of xylan requires the synergistic action of main-and side-chain-cleaving enzymes with different specificities (Coughlan et al., 1993;Coughlan and Hazlewood, 1993a). For xylan-degrading enzymes, it is difficult todemonstrate the synergism by measuring only the reducing sugars produced, becauseof the heterogeneous nature of the substrate. Hence, the hydrolysis products mustbe separated, identified and quantified in order to obtain a detailed picture of thesynergistic action of xylan-degrading enzymes. In fact, three types of synergy betweensuch enzymes have been described: (i) homeosynergy; (ii) heterosynergy; and (iii)antisynergy (Coughlan et al., 1993). Homeo- and heterosynergies could be eitheruni- or biproduct. Homeosynergy is the synergistic interaction between two or moredifferent types of side-chain-cleaving enzymes or between two or more types ofmain-chain-cleaving enzymes (Coughlan et al., 1993). Homeosynergy is best demon-strated for Neurospora crassa (Deshpande et al., 1986), Talaromyces byssochlamydoides(Yoshioka et al., 1981) and Trichoderma harzianum (Wong et al., 1986), where endo-xylanases of different specificities, or mixtures of endoxylanases and β-xylosidases,cooperate in the hydrolysis of xylan. Another example of homeosynergy is reportedbetween ferulic acid esterase from A. oryzae and α-L-arabinofuranosidase from P.capsulatum, where the former enzyme facilitates the release of arabinose fromferuloylated arabinoxylan by the latter (Coughlan et al., 1993).

Heterosynergy is the synergisitc interaction between side-chain- andmain-chain-cleaving enzymes (Coughlan et al., 1993). Heterosynergy is uniproductif the action of the main-chain-cleaving enzyme facilitates the release of a substituentby the side-chain-cleaving enzyme, or vice versa. Heterosynergy is biproduct if theextent of liberation of substituent, and the hydrolysis of the main chain, by the actionof combined enzymes exceeds the sum of the actions of the individual enzymes.Heterosynergy has been reported between ferulic acid esterases and endoxylanases(Faulds and Williamson, 1991; Tenkanen et al., 1991), α-L-arabinofuranosidasesand endoxylanases (Tuohy et al., 1992), acetyl xylan esterases and endoxylanases(Biely et al., 1986; Lee et al., 1987), as well as between α-glucuronidases andendoxylanases (Puls et al., 1987). A good example of heterosynergy is shown inFig. 2.5.





Antisynergy occurs when the action of one type of enzyme prevents the action ofa second enzyme (Coughlan et al., 1993). This can be observed in the case of eitherarabinoxylan xylanohydrolase (Frederick et al., 1985) or glucuronoxylanglucuronohydrolase (Nishitani and Nevins, 1991), an enzyme that cleaves the mainchain linkages only in the vicinity of a particular substituent. For example, in the caseof glucuronoxylan glucuronohydrolase, the presence of substituents is essential for itsaction on the main-chain and the removal of the substituent by the relevantside-chain-cleaving enzyme would preclude its action on the main chain. Althoughantisynergy is unlikely to occur in vivo, it may be observed in in vitro experimentswhere only one or two enzymes are used in a specific application.

Catalytic mechanism

Cellulases and xylanases catalyse stereoselective hydrolysis of the glycosidic bond witheither retention or inversion of configuration around the anomeric centre (Sinnott,1990; Davies and Henrissat, 1995). Inversion occurs by way of a single displacementreaction, while retention is via double displacement (Sinnott, 1990). Both reactionmechanisms are assisted by general acid/base catalysis (Fig. 2.6).

During inversion, the protonation of the glycosidic oxygen and aglyconedeparture are accompanied by a concomitant attack of a water molecule that isactivated by the catalytic base. This single nucleophilic substitution yields a productwith opposite stereochemistry to the substrate. In contrast, the retention mechanisminvolves two steps: glycosylation and deglycosylation. In the glycosylation step, thecatalytic residue, acting as a general acid, protonates the glycosidic oxygen with



Fig. 2.5. Heterosynergistic interactions in the hydrolysis of (a) feruloylxylan and(b) arabinoxylan by fungal enzymes. Abbreviations: Ara, α-L-arabinofuranosidase;Fae, ferulic acid esterase; Xyl, xylanase; exp, expected; obs, observed. Reprintedfrom Hemicellulose and Hemicellulase, Coughlan, M.P. and Hazlewood,G.P. (eds), Coughlan, M.P., Tuohy, M.G., Filho, E.X.F., Puls, J., Claeyssens, M.,Vrsanska, M. and Hughes, M.M., Enzymological aspects of microbial hemi-cellulases with emphasis on fungal systems, pp. 53–84, 1993, with permissionfrom Portland Press, London.



concomitant C-O breaking of the scissile glycosidic bond. A deprotonatedcarboxylate, functioning as a nucleophile, attacks the anomeric centre to give acovalent glycosyl–enzyme intermediate. The deglycosylation step involves the attackof a water molecule, assisted by the conjugate base of the general acid, to release thefree sugar with overall retention of configuration. Both steps proceed via transitionstates with considerable oxocarbonium ion character. In both retention and inversionmechanisms, the position of the proton donor is within hydrogen-bonding distanceof the glycosidic oxygen. In retaining enzymes, the nucleophilic catalytic base is inclose vicinity to the sugar anomeric carbon, whereas in inverting enzymes this base ismore distant, since it must accommodate a water molecule between the base and thesugar. This difference results in an average distance between the two catalytic residuesof 5.5 Å in retaining enzymes, as opposed to 10 Å in inverting enzymes (Davies andHenrissat, 1995).



Fig. 2.6. (a) Retention and (b) inversion mechanisms of glycosyl hydrolases.Reprinted from Structure, Vol. 3, Davies, G. and Henrissat, B, Structures andmechanisms of glycosyl hydrolases, pp. 853–859, 1995, with permission fromCurrent Biology Ltd, London.



The stereochemical course of hydrolysis has been determined for glycosylhydrolases belonging to 42 of the 64 known families, of which families 1, 3, 5–9, 12,26, 43–45, 48, 60 and 61 represent cellulases, and 10 and 11 represent xylanases(Henrissat and Davies, 1997; Davies, 1998; Henrissat, 1998). Enzymes of families 1,3, 5, 7, 10–12 and 26 proceed with retention of configuration, while enzymes fromfamilies 6, 8, 9, 43, 45 and 48 proceed with inversion of configuration (Table 2.3).Interestingly, all representatives of a particular family follow the same catalyticmechanism (Gebler et al., 1992). Thus, it is possible to predict the catalyticmechanism of a given cellulase or xylanase accurately once its family is known.

Other Characteristics of Cellulases and Xylanases

Induction and regulation

Induction and regulation of microbial cellulases and xylanases have been the subjectof detailed investigation (Ryu and Mandels, 1980; Kubicek, 1992; Biely, 1993;Kubicek et al., 1993; Saloheimo et al., 1998). Both cellulases and xylanases areinducible enzymes. It has been reported that cellulose or cellulose-rich material andxylan or xylan-rich material are the best carbon sources for the production of highlevels of cellulases and xylanases by many microorganisms (Ryu and Mandels, 1980;Biely, 1993). Since these polysaccharides cannot enter the microbial cell, it isgenerally accepted that the low levels of cellulases and xylanases constitutivelyproduced by the microorganisms hydrolyse the corresponding polysaccharides tosoluble sugars, which enter the microbial cell and either act as, or get converted into,true inducers and stimulate the transcription of cellulase and xylanase genes (Biely,1993; Kubicek et al., 1993). In fact, it has been suggested that the conidial-boundCBH of T. reesei hydrolyses cellulose to cellobiose and cellobiono-δ-1,5-lactone(CBL), which promotes cellulase synthesis (Kubicek et al., 1988). Although cello-biose and xylobiose are assumed to be the natural inducers of cellulases and xylanases,the rapid utilization of these disaccharides suggested that the transglycosylatedproducts of these disaccharides or their positional isomers could in fact be the trueinducers (Ryu and Mandels, 1980; Biely, 1993). Sophorose (β-1,2-glucobiose), apositional isomer of cellobiose, is reported to be an effective inducer of cellulases inT. reesei and other species of Trichoderma, even at low concentrations (Sternberg andMandels, 1979). However, the inability of sophorose to induce cellulase synthesisin other fungi (Moloney et al., 1983), and the ability of cellobiose to induce theproduction of cellulase by T. reesei, in the presence of either a β-glucosidase inhibitoror in a β-glucosidase defective mutant (Fritscher et al., 1990), strongly indicates thatcellobiose could be the true inducer of the cellulase system. Also, the positionalisomers of xylobiose (Xylβ,1-2Xyl; Xylβ,1-3Xyl) were found to be poor inducers ofxylanase in C. albidus compared with xylobiose (Xylβ,1-4Xyl), which suggests thatthe latter is probably the true inducer of xylanase (Biely, 1993).

Initial studies using protein synthesis inhibitors suggested that the productionof cellulase is regulated at the translational level (Nisizawa et al., 1972). However, the







Fam

ilyEn

zym

eO

rgan

ism

ECnu

mbe

rM

echa

nism

Ref

eren

ce

1β-

Glu

cosi

dase

Trifo

lium

repe

ns3.

2.1.

21R

eten

tion

Barre

ttet

al.,

1995

3β-

Glu

cosi

dase

Fung

iand

bact

eria

3.2.

1.21

Ret

entio

nW

arre

n,19

98

5En

doA

Cel

BC

elC

Cel

HEG

IIIEn

do5

EGZ

C.c

ellu

loly

ticum

C.t

herm

ocel

lum

C.t

herm

ocel

lum

C.t

herm

ocel

lum

T.re

esei

T.fu

sca

E.ch

rysa

nthe

mi

3.2.

1.4

3.2.

1.4

3.2.

1.4

3.2.

1.4

3.2.

1.4

3.2.

1.4

3.2.

1.4

Ret

entio

nR

eten

tion

Ret

entio

nR

eten

tion

Ret

entio

nR

eten

tion

Ret

entio

n

Geb

ler e

tal.,

1992

Geb

ler e

tal.,

1992

Geb

ler e

tal.,

1992

Geb

ler e

tal.,

1992

Geb

ler e

tal.,

1992

Geb

ler e

tal.,

1992

Barra

set

al.,

1992

6C

BHII

Exog

luca

nase

–E3

Endo

gluc

anas

eC

bhA

Cen

A

T.re

esei

T.fu

sca

T.fu

sca

C.f

imi

C.f

imi

3.2.

1.91

3.2.

1.91

3.2.

1.4

3.2.

1.91

3.2.

1.4

Inve

rsio

nIn

vers

ion

Inve

rsio

nIn

vers

ion

Inve

rtion

Rou

vine

net

al.,

1990

Barr

etal

.,19

96Sp

ezio

etal

.,19

93M

eink

eet

al.,

1994

With

ers

etal

.,19

88

7C

BHI

Endo

gluc

anas

eI

Endo

gluc

anas

eI

Endo

gluc

anas

eI

T.re

esei

H.i

nsol

ens

T.re

esei

F.ox

yspo

rum

3.2.

1.91

3.2.

1.4

3.2.

1.4

3.2.

1.4

Ret

entio

nR

eten

tion

Ret

entio

nR

eten

tion

Div

neet

al.,

1994

Dav

ies

and

Hen

rissa

t,19

95C

laey

ssen

set

al.,

1990

bSu

lzen

bach

eret

al.,

1996

;Mac

kenz

ieet

al.,

1997

8C

elA

Cel

CC

.the

rmoc

ellu

mC

.cel

lulo

lytic

um3.

2.1.

43.

2.1.

4In

vers

ion

Inve

rsio

nAl

zari

etal

.,19

96Fi

erob

eet

al.,

1993

Tabl

e2.

3.Li

stof

cellu

lase

san

dxy

lana

ses

belo

ngin

gto

diffe

rent

fam

ilies

ofgl

ycos

ylhy

drol

ases

and

thei

rcat

alyt

icm

echa

nism

.





9En

doD

Cen

BEn

doI

Endo

4

C.t

herm

ocel

lum

C.f

imi

T.fu

sca

T.fu

sca

3.2.

1.4

3.2.

1.4

3.2.

1.4

3.2.

1.4

Inve

rsio

nIn

vers

ion

Inve

rsio

nIn

vers

ion

Juy

etal

.,19

92G

eble

r eta

l.,19

92G

eble

r eta

l.,19

92Sa

kon

etal

.,19

97

10Xy

lana

seA

CeX

Xyn

ZXy

lana

seA

S.liv

idan

sC

.fim

iC

.the

rmoc

ellu

mP.

fluor

esce

ns

3.2.

1.8

3.2.

1.91

3.2.

1.8

3.2.

1.8

Ret

entio

nR

eten

tion

Ret

entio

nR

eten

tion

Der

ewen

daet

al.,

1994

With

ers

etal

.,19

88G

eble

r eta

l.,19

92H

arris

etal

.,19

94

11Xy

lana

seXy

lana

seXy

lana

seXy

lana

se

B.ci

rcul

ans

B.su

btilis

S.co

mm

une

T.re

esei

3.2.

1.8

3.2.

1.8

3.2.

1.8

3.2.

1.8

Ret

entio

nR

eten

tion

Ret

entio

nR

eten

tion

Wak

arch

uket

al.,

1994

Geb

ler e

tal.,

1992

Geb

ler e

tal.,

1992

Hav

ukai

nen

etal

.,19

96

12En

dogl

ucan

ase

(Cel

B)S.

livid

ans

3.2.

1.4

Ret

entio

nSu

lzen

bach

eret

al.,

1997

26En

dogl

ucan

ase

Bact

eria

3.2.

1.4

Ret

entio

nW

arre

n,19

98

43Xy

lana

ses

Fung

iand

bact

eria

3.2.

1.8

Inve

rsio

nW

arre

n,19

98

45En

doV

H.i

nsol

ens

3.2.

1.4

Inve

rsio

nD

avie

set

al.,

1995

48C

BHb

Endo

gluc

anas

e–

E6C

.fim

iT.

fusc

a3.

2.1.

913.

2.1.

4In

vers

ion

Inve

rsio

nSh

enet

al.,

1995

Barr

etal

.,19

96



presence of high levels of cellulase mRNA in T. reesei grown on solka floc and thecomplete absence of cellulase mRNA in T. reesei grown on glucose revealed thatthe cellulase expression is regulated at transcriptional level (El-Gogary et al., 1989;Saloheimo et al., 1998). The addition of low levels (1–2 mM) of sophorose resultedin the accumulation of very high levels of mRNA in T. reesei (Ilmen et al., 1997).Nevertheless, no cellulase mRNA was detected when sophorose was added toglucose-grown culture or glucose was added to cellulose-grown culture of T. reesei(Ilmen et al., 1997). These results clearly indicated that both induction and glucoserepression are operating at the transcriptional level during the regulation of cellulasesynthesis. Furthermore, all cellulase inducers promoted the synthesis of a constantratio of three major cellulases such as CBH I, CBH II and endoglucanase I in T.reesei. This indicates that the expression of cellulase genes is coordinatedly regulated,at least in T. reesei (Kubicek, 1992; Saloheimo et al., 1998).

In contrast, the induction pattern of hemicellulase genes appeared to be variableeven in T. reesei. For example, the expression pattern of xyn I and II genes wassomewhat different (Saloheimo et al., 1998). Also, genes coding for debranchingenzymes were efficiently induced by xylan having the appropriate substitutions.Thus, regulation of hemicellulase genes appears to be more complex than cellulasegenes, even though both sets of genes are regulated through induction and cataboliterepression. Interestingly, two activator genes corresponding to proteins called ACEI and II, responsible for the activation of cellulase expression, were isolated fromT. reesei (Saloheimo et al., 1998).

Recent studies on induction of cellulases and xylanases in Schizophilumcommune revealed that cellulose or cellulose-rich substrates greatly stimulated theformation of both cellulases and xylanases by this fungus. Also, other carbon sourcessuch as cellobiose, lactose, sophorose and L-sorbose induced production of bothcellulase and xylanase (Haltrich et al., 1995). Although the production of xylanasesalong with cellulases on media containing cellulose has been reported in thecase of Aspergillus sp., S. rolfsii, T. aurantiacus, P. pinophilum, etc., there are distinctdifferences between these strains and S. commune (Brown et al., 1987; Yu et al.,1987; Lachke and Deshpande, 1988; Bailey and Poutanen, 1989). Thus, thesefungi produced both cellulase and xylanase when xylan was present as a contaminantwith cellulose, and its removal from cellulose reduced their ability to producexylanase. In contrast, the removal of xylan from Avicel induced higher levels ofboth cellulase and xylanase production by S. commune. In addition, cellulosefrom Acetobacter pasteurianus, free of xylan, strongly induced the production ofboth cellulase and xylanase in S. commune. Furthermore, cellobiose, the repeatingunit of cellulose, induced the production of both cellulase and xylanase byS. commune, which strongly suggests that this fungus appears to possess acommon regulatory mechanism for the synthesis of cellulases and hemicellulases(Haltrich et al., 1995). A similar regulatory system appears to operate in anotherwood-rotting fungus, Polyporus adustus (Eriksson and Goodell, 1974). Thus, themechanism governing induction and regulation of cellulases and xylanases appearsto be quite complicated, and markedly different from one microorganism toanother.





Physicochemical properties

Most fungal endoglucanases and CBHs (exoglucanases) have molecular masses rang-ing from 20 to 100 kDa, whereas the molecular mass of β-glucosidase ranges from 50to 300 kDa (Coughlan and Ljungdahl, 1988; Bhat et al., 1993). In general, fungalcellulases are optimally active between pH 4.0–6.0, whereas bacterial enzymes show apH optimum of 6.0–7.0 (Wood, 1985). Many fungal endoglucanases and CBHs arehighly glycosylated while bacterial enzymes are less glycosylated (Wood et al., 1988;Bhat et al., 1989). Endoglucanases, CBHs and β-glucosidases from mesophilic fungiand bacteria are optimally active between 40 and 55°C (Wood et al., 1988; Bhatet al., 1989; Christakopoulos et al., 1994a). Cellulases from thermophilic andhyperthermophilic microorganisms are optimally active between 60–80 and90–110°C, respectively (Khandke et al., 1989; Bhat et al., 1993; Antranikian, 1994).

Fungal and bacterial xylanases are almost exclusively single subunit proteins withmolecular masses ranging from 8.5 to 85 kDa, and with pI values of 4.0–10.3(Coughlan et al., 1993). Based on their physicochemical properties, most xylanasesmay be assigned to two classes: proteins with molecular mass < 30 kDa are usuallybasic, whereas those with molecular mass > 30 kDa are usually acidic (Wong et al.,1988). Also, most xylanases have acidic pH optima (Wong et al., 1988), with theexception of xylanases from an alkalophilic thermophilic Bacillus sp., which areoptimally active at alkaline pH and are particularly suitable for paper and pulpapplications. In addition, xylanases from thermophiles are optimally active at60–80°C, while those from mesophiles show optimal activity from 40–55°C(Coughlan et al., 1993; Bennett et al., 1998).

Multiplicity of cellulases and xylanases

Most microorganisms capable of degrading cellulose and xylan produce multipleforms of cellulases and xylanases with either very similar substrate specificities orvarying substrate specificities (Wong et al., 1988; Bhat et al., 1989). Based on bio-chemical studies, it has been reported that T. reesei and P. pinophilum produce twoforms of CBHs and four to seven forms of endoglucanases (Gong et al., 1979; Woodet al., 1988; Bhat and Wood, 1989), whereas A. niger, T. viride and T. emersoniiproduce 15, 13 and 11–13 forms of xylanases, respectively (Biely, 1985; Coughlanet al., 1993). Nevertheless, it is not certain whether the above multiple formsof cellulases and xylanases are the products of multiple genes or the result ofpost-translational modification of a single gene product by differential glycosylationor partial proteolysis. Also, it is often difficult to distinguish the multiple forms ofcellulases and xylanases based on substrate specificity, because of the lack of definedand structurally characterized substrates, and the preponderance of enzymes withbroad substrate specificity. It has been speculated that the apparent broad substratespecificity of some cellulases and xylanases could be due to traces of contaminatingprotein(s) present in purified samples. However, it has been clearly shown thatP. pinophilum produces seven to eight forms of endoglucanase, which differ with





respect to their pI values, during the early logarithmic growth phase, and furtherincubation only increases their concentration with minor variations in relativeproportions (Bhat and Wood, 1989).

The above results suggest that multiple endoglucanases must be the result ofexpression of different genes, rather than post-translational modification. This facthas been confirmed by the isolation and characterization of many cellulase andxylanase genes from a number of bacteria and fungi (Gilbert and Hazlewood, 1993).It has been shown conclusively that the isomeric forms of cellulases and xylanases,observed in several microorganisms, are the consequence of large multigene familiesand are not the result of processing of a single gene product. For example, at least 21endoglucanase, three exoglucanase, six xylanase, and two β-glucosidase genes, andone xylan esterase gene, have been isolated from C. thermocellum (Hazlewood et al.,1988, Sakka et al., 1989; Hayashi et al., 1997). Likewise, multiple cellulase andxylanase genes have been isolated and characterized from strains of Ruminococcus,Pseudomonas, Cellulomonas, Bacillus, Fibrobacter and Butyrivibrio, etc. (Gilkes et al.,1991a; Flint and Zhang, 1992; Gilbert and Hazlewood, 1993; Hazlewood andGilbert, 1998b–d). Furthermore, multiple cellulase and xylanase gene productshave been characterized and their structural and functional differences have beendemonstrated. Even though it is clear that multiple forms of cellulases and xylanasesare the products of different genes, a significant role for post-translationalmodification in the production of some cellulase and xylanase isoforms cannot becompletely ruled out.

Adsorption

Adsorption of cellulases on to cellulose not only facilitates physical contact betweenenzyme and substrate, but also plays an important role in the efficiency of theenzymatic hydrolysis of crystalline cellulose (Klyosov, 1990). Cellulases thatadsorbed to a greater extent showed an increase in the rate and extent of crystallinecellulose hydrolysis when compared with cellulases that adsorbed less strongly(Klyosov, 1990). Moreover, the ability of endoglucanases to adsorb on to crystallinecellulose varies significantly. For example, an endoglucanase from Aspergillus foetidusshowed weak adsorption on to cellulose, whereas that from C. thermocellum adsorbedvery strongly on to cellulose (Klyosov, 1990). The discovery of CBDs in somecellulolytic enzymes explains in part the differences in adsorption ability of cellulases(Gilkes et al., 1991b). However, a correlation between the hydrophobicity ofendoglucanases and their ability to adsorb on to cellulose has also been reported(Klyosov, 1990). A study of the adsorption of 12 cellulase preparations and theirability to hydrolyse amorphous and crystalline cellulose revealed that a 20-foldincrease in the adsorption capability of endoglucanases does not change the rate ofhydrolysis of CM-cellulose, but increases the rate of hydrolysis of amorphous andcrystalline celluloses by twofold and 50-fold, respectively (Klyosov, 1990). Moreconvincingly, Wood and co-workers (1989) demonstrated that a mixture of CBHs Iand II from P. pinophilum significantly synergized and effected extensive hydrolysis





of crystalline cellulose when mixed with strongly adsorbed endoglucanases III and Vfrom the same source, but not in the presence of weakly adsorbed endoglucanases I,II and IV.

Architecture and classification

The presence of separate catalytic and cellulose-binding domains in cellulases wasfirst demonstrated by proteolytic cleavage of cellulases from C. fimi and T. reesei(Tomme et al., 1988; Gilkes et al., 1991b). Typically the two domains are linked by ahighly glycosylated amino acid sequence (six to 59 residues), rich in proline andhydroxyamino acids. Some linkers are relatively rich in either Asp or Glu, or both.The occurrence of this type of cellulase architecture has been extensively reviewed byTomme et al. (1995b). It was subsequently shown that many xylan-degradingenzymes are also modular in structure. For example, in Pseudomonas fluorescens subsp.cellulosa, four xylanases, an arabinofuranosidase and an acetylxylan esterase all havemodular architecture and, like the cellulases from this organism, all contain bothcatalytic and non-catalytic CBDs, joined together by serine-rich linkers or hinges(Hazlewood and Gilbert, 1998a,c,d). Moreover, a similar pattern of architecturehas been reported for xylanolytic enzymes from R. flavefaciens, C. fimi and Cellvibriomixtus (Flint et al., 1993; Millward-Sadler et al., 1994, 1995). Although the majorityof polysaccharide-binding domains mediate specific binding to cellulose, bindingdomains with specificity for xylan have been described in xylanases from C. fimi(Black et al., 1995) and T. fusca (Irwin et al., 1994). Other interesting non-catalyticdomains described in modular xylanases include thermostabilizing domains andthe so-called Nod B domains. Thermostabilizing domains have been found inxylanases from thermophilic (Fontes et al., 1995) and mesophilic (Clarke et al.,1996) bacteria and appear to confer stability in a broad sense on the enzymes thatcontain them. Nod B domains occur in xylanases from P. fluorescens, C. mixtus andC. fimi and have the capacity to release acetyl groups from acetylated xylan(Hazlewood and Gilbert, 1998a).

Cellulose-binding domainsThe CBDs of cellulases and xylanases from a wide range of microorganisms havebeen extensively characterized (Tomme et al., 1995b; Linder and Teeri, 1997). Basedon their amino acid sequences, CBDs have been classified into more than 10 families(Tomme et al., 1995b). Most of the CBDs characterized so far belong to families I, IIand III, while the remaining families have a few or only one member in each family.Family I contains fungal CBDs of 30–40 amino acid residues; the best characterizedare from CBH I, CBH II, and endoglucanases II and III from T. reesei. All otherfamilies contain bacterial CBDs, which are much larger in size than family I CBDs.Family II CBDs are found almost exclusively at either the N or C terminus ofthe enzyme and consist of typically 95–108 amino acids. The well-characterizedCBDs of this family are from Cex, CenA and CenB from C. fimi, CelA, CelB, CelC,XylA, XylB, XylC and XylD from P. fluorescens subsp. cellulosa (Tomme et al.,





1995c). Family III CBDs are found in cellulosome-producing bacteria, namely C.thermocellum and C. cellulovorans, and contain 132–172 amino acids (Tomme et al.,1995b). Based on amino acid sequence differences and other properties of familyIII CBDs, Bayer et al. (1998) proposed three subfamilies called IIIa, b and c. FamilyIIIa includes scaffoldin CBDs from C. thermocellum, C. cellulovorans and C.cellulolyticum; family IIIb includes the CBDs of free cellulases listed in family III,while family IIIc includes CBDs that lack cellulose-binding and anchoring residues.Family IV CBDs are found in enzymes from C. fimi, C. cellulolyticum and otherbacteria, and contain 125–168 amino acids. CBDs of families V (E. chrysanthemi,EgZ, 63 amino acids), VII (C. thermocellum, CelE, 240 amino acids) and VIII(D. discoidum, CelA, 152 amino acids) are the sole members of their families, whilefamilies VI, IX and X contain three to seven members with 85–90, 170–189 and51–55 amino acids, respectively (Tomme et al., 1995b).

The three-dimensional structures of CBDs of families I (CBD of T. reesei CBH;Kraulis et al., 1989), II (CBD of C. fimi Cex; Xu et al., 1995) and IV (CBD of C. fimiβ-1,4-glucanase; Johnson et al., 1996) have been determined by nuclear magneticresonance (NMR), while those of families IIIa (CBD of C. thermocellum scaffoldin;Tormo et al., 1996) and IIIc (CBD of T. fusca cellulase E4; Sakon et al., 1997) weredetermined by X-ray crystallography.

Although there is an overall functional similarity between CBDs from differentfamilies, it is clear that they differ with respect to their affinity for different types ofcellulose (Linder et al., 1995, 1996; Tomme et al., 1995c, 1996). Also, the presenceof two successive CBDs in an enzyme is believed to result in synergistic bindingand an increase in affinity for the ligand (Linder et al., 1996; Tomme et al., 1996).A number of approaches has been used to investigate the role of CBDs in celluloseand xylan hydrolysis, including truncation of enzymes, domain swappingand site-directed mutagenesis (Tomme et al., 1988, 1995b; Reinikainen et al.,1992; Black et al., 1997; Srisodsuk et al., 1997), but a unifying role for CBDs hasyet to be established and their biological function continues to be a matter forconjecture. Although the removal of CBDs from endoglucanases and CBHs doesnot affect the activity of the respective enzymes against soluble substrates, there isevidence that CBDs play a major role in potentiating the activity of cellulases againstthe more resistant forms of insoluble cellulose (Tomme et al., 1988; Coutinho et al.,1993; Hall et al., 1995). Indeed, an isolated family II CBD from C. fimi CenA wasshown to open the structure of crystalline cellulose, making it more accessibleto enzyme attack (Din et al., 1991). The fact that CBDs occur with high frequencyin xylan-degrading enzymes which have no activity against cellulose suggeststhat these domains potentiate catalytic activity of hemicellulases by promotingclose contact between the enzymes and their native substrates in plant cell walls.Support for this view has come from a number of recent studies (Hazlewood andGilbert, 1998b,c,d), including one which showed that the activities of a xylanaseand an arabinofuranosidase from P. fluorescens subsp. cellulosa, against plantcell walls, were significantly enhanced by the presence of their CBDs (Black et al.,1997).





Catalytic domainsThe catalytic domains of cellulases and xylanases are relatively large and representmore than 70% of the total protein. Sequence analysis of these domains from variouscellulases and xylanases revealed extensive conservation of sequence and, usinghydrophobic cluster analysis, the catalytic domains of cellulases and xylanaseswere initially grouped into 11 structurally related families (Henrissat et al., 1989).According to this classification, the members of a single family possess the sameprotein fold and display the same stereoselectivity (Gebler et al., 1992). Thissuggested that the enzymes of a given family would follow either retaining orinverting mechanisms (Table 2.3). Subsequently, the catalytic domains of some 1000glycosyl hydrolases have been classified into 64 families (Henrissat and Bairoch,1993, 1996; Henrissat, 1998). The cellulases and xylanases are contained in families1, 3, 5–12, 26, 43–45, 60 and 61 (Henrissat and Davies, 1997; Davies, 1998;Henrissat, 1998). This sequence-based classification has been enormously beneficialin reflecting common structural features and evolutionary relationships, and as a toolfor deducing mechanistic information.

Structure/function relationships

Crystal structures have been determined for more than 20 glycosyl hydrolases,including β-glucosidases, endoglucanases, exoglucanases and xylanases from families1, 5–12 and 45. The main structural and functional properties of cellulases andxylanases studied so far are summarized in Table 2.4. Interestingly, members offamilies 1, 5 and 10 show a typical eightfold α/β barrel structure, commonly knownas a TIM barrel, but these families include β-glucosidase, endoglucanase andxylanase, respectively (Table 2.4). Families 5, 8, 9, 12, 26 and 45 contain mainlyendoglucanases, families 10 and 11 include only xylanases, while families 6 and 7contain both endoglucanases and CBHs.

The catalytic domains of T. reesei CBH II and endoglucanase (E2) from T. fuscabelong to family 6 and show a distorted α/β barrel structure (Rouvinen et al., 1990;Spezio et al., 1993). Although both these enzymes belong to the same family andpossess a similar three-dimensional structure, there is a clear difference in their activesite architecture, with an enclosed tunnel consisting of four glucose-binding sitesin the case of CBH II, and an open active site in the case of the endoglucanase. Thecatalytic domains of CBH I from T. reesei and endoglucanase I from H. insolensbelong to family 7 and show a similar β-jelly roll structure (Divne et al., 1994;Sulzenbacher et al., 1996, 1997). However, CBH I has an extended active site tunnelwith seven glycosidic binding sites, while endoglucanase I from H. insolens shows amore open structure suitable for endo-action.

CelA from C. thermocellum belongs to family 8 and folds into an α6/α6 barrelconsisting of six internal and six external helices (Alzari et al., 1996). A strictlyconserved residue, Glu-95, at the centre of the substrate-binding cleft, serves as aproton donor. Three glucose-binding sites are identified on either side of the scissileglycosidic linkage. Furthermore, the subsites do not lie in a straight line, which forces







Enzy

me

Fam

ilyO

rgan

ism

Stru

ctur

eFu

nctio

nR

efer

ence

β-G

luco

sida

se1

T.re

pens

Eigh

tfold

α/β-

barre

lwith

pock

etty

peac

tive

site

Cle

aves

disa

ccha

rides

from

the

non-

redu

cing

end

Barre

ttet

al.,

1995

Endo

gluc

anas

eA

Cel

C

5 5

C.c

ellu

loly

ticum

C.t

herm

ocel

lum

Eigh

tfold

α/β-

barre

lwith

groo

vety

peac

tive

site

Asab

ove

with

anad

ditio

nals

egm

ent

Cle

aves

cellu

lose

chai

nra

ndom

ly

Asab

ove

Duc

ros

etal

.,19

95

Dom

ingu

ezet

al.,

1995

CBH

II

Endo

gluc

anas

e-E2

6 6

T.re

esei

T.fu

sca

Dis

torte

d(α

/β)b

arre

lwith

tunn

elsh

ape

activ

esi

te

Asab

ove

with

anop

enac

tive

site

Rel

ease

sce

llobi

ose

from

the

non-

redu

cing

end

and

faci

litat

esth

read

ing

ofce

llulo

sech

ain

thro

ugh

itR

ando

mly

actin

gen

zym

e

Rou

vine

net

al.,

1990

Spez

ioet

al.,

1993

CBH

I

Endo

gluc

anas

eI

7 7

T.re

esei

H.i

nsol

ens

β-Je

llyro

llw

itha

long

activ

esi

tetu

nnel

β-Je

llyro

llw

itha

long

subs

trate

bind

ing

groo

ve

Rel

ease

sce

llobi

ose

from

the

redu

cing

end

ofce

llulo

sech

ain

Ran

dom

lyac

ting

enzy

me

Div

neet

al.,

1994

Sulz

enba

cher

etal

.,19

96,

1997

Cel

A8

C.t

herm

ocel

lum

α/α

Barre

l–si

xin

tern

alan

dsi

xex

tern

alhe

lices

with

groo

vety

peac

tive

site

Ran

dom

lyac

ting

enzy

me

with

six

glyc

osyl

bind

ing

site

sAl

zari

etal

.,19

96

Cel

D9 9

C.t

herm

ocel

lum

T.fu

sca

α/α

Barre

lwith

two

dist

inct

dom

ains

,an

dex

tend

edac

tive

site

Poss

ess

both

fam

ily9

cata

lytic

dom

ain

with

an( α

/α) 6

barre

lfol

d,an

da

fam

ilyIII

CBD

,hav

ing

anan

ti-pa

ralle

l β-s

andw

ich

fold

Ran

dom

lyac

ting

enzy

me

with

six

glyc

osyl

bind

ing

site

sSh

ows

both

endo

-and

exog

luca

nase

activ

ities

with

six

glyc

osyl

bind

ing

site

s

Juy

etal

.,19

92

Sako

net

al.,

1997

Tabl

e2.

4.St

ruct

ure

and

func

tion

rela

tions

hips

ofce

llula

ses

and

xyla

nase

s.





Exog

luca

nase

–C

eX

Xyla

nase

A

Xyla

nase

A

10 10 10

C.f

imi

S.liv

idan

s

P.flu

ores

cens

Eigh

t-par

alle

lstra

nded

α/β

prot

ein

with

anop

encl

efta

ctiv

esi

te,a

ndbe

long

sto

4/7

supe

rfam

ily

( α/β

) 8Ba

rrelw

ithpo

cket

type

activ

esi

tean

dbe

long

sto

4/7

supe

rfam

ily

(α/β

) 8Ba

rrel,

belo

ngs

to4/

7su

perfa

mily

and

with

anun

expe

cted

Ca2+

bind

ing

site

whi

chst

abiliz

esth

eca

taly

ticco

re

Rel

ease

sce

llobi

ose

from

the

non-

redu

cing

end

ofce

llulo

sech

ain

and

hydr

olys

esxy

losi

dic

and

gluc

osid

iclin

kage

sH

ydro

lyse

sxy

lan

and

pNPC

pref

eren

tially

,and

poss

esse

ssi

xgl

ycos

ylbi

ndin

gsi

tes

Ran

dom

lycl

eave

sxy

lan

chai

nan

dpo

sses

ses

five

glyc

osyl

bind

ing

site

s

Whi

teet

al.,

1994

Der

ewen

daet

al.,

1994

Har

riset

al.,

1994

Xyla

nase

Xyla

nase

Xyla

nase

IXy

lana

seII

Xyla

nase

11 11 11 11 11

B.ci

rcul

ans

T.ha

rzia

num

T.re

esei

T.re

esei

T.la

nugi

nosu

s

β-Je

llyro

llw

ithan

over

alls

truct

ure

ofpa

rtial

lycl

osed

right

hand

Asab

ove

Asab

ove

Asab

ove

Asab

ove

Ran

dom

lyac

ting

enzy

me

with

five

glyc

osyl

bind

ing

site

sAs

abov

eAs

abov

eAs

abov

ebu

twith

four

glyc

osyl

bind

ing

site

sAs

abov

ebu

twith

seve

ngl

ycos

ylbi

ndin

gsi

tes

Wak

arch

uck

etal

.,19

94

Cam

pbel

leta

l.,19

93To

rrone

nan

dR

ouvi

nen,

1997

Torro

nen

etal

.,19

94;

Hav

ukai

nen

etal

.,19

96G

rube

r eta

l.,19

98

Endo

gluc

anas

e(C

elB)

12S.

livid

ans

β-Je

llyro

llw

ithan

ti-pa

ralle

lβ-s

heet

sar

rang

edin

asa

ndw

ich

fash

ion,

and

open

activ

esi

te

Ran

dom

lyac

ting

enzy

me

with

five

glyc

osyl

bind

ing

site

sSu

lzen

bach

eret

al.,

1997

Endo

gluc

anas

e45

H.i

nsol

ens

Six-

stra

nded

β-ba

rrelw

ithan

open

clef

tact

ive

site

Rel

ease

sce

llobi

ose

from

the

redu

cing

end

and

poss

esse

ssi

xgl

ycos

ylbi

ndin

gsi

tes

Dav

ies

etal

.,19

93



the substrate to bend at the active site and thus requires lower energy for cleavage ofthe glycosidic bond. CelD of C. thermocellum belongs to family 9 and has a globularand slightly elongated shape with a small N-terminal β-barrel closely packed against alarge, mostly α-helical domain (Juy et al., 1992). An extended active site withsix glycosyl-binding sites has been suggested since it was possible to fit cellohexaoseinto the active site. Recently, the three-dimensional structure of E4 from T. fusca,belonging to the same family, has been determined (Sakon et al., 1997). This enzymecontains both a family 9 catalytic domain with an α/α6 barrel fold, and a family IIICBD, having an antiparallel β-sandwich fold. In addition, this enzyme functions asboth endo- and exoglucanase and possesses an extended glycosyl-binding site, similarto that of C. thermocellum CelD.

Xylanases from families 10 and 11 follow the retaining mechanism and possessessential glutamates in their active sites. Family 10 xylanases show a typical eightfoldα/β barrel structure (Harris et al., 1994), while family 11 xylanases possess a β-jellyroll architecture with two parallel β-sheets and an α-helix (Torronen and Rouvinen,1997; Pickersgill et al., 1998). Family 12 endoglucanase from Streptomyces lividansshows a β-jelly roll topology with antiparallel β-sheets arranged in a sandwichmanner, whereas the family 45 endoglucanase from H. insolens has a six-strandedβ-barrel structure (Davies et al., 1993; Sulzenbacher et al., 1997). The family 12enzyme possesses five glycosyl-binding sites and cleaves the glucan chain randomly,while the family 45 enzyme has six glycosyl-binding sites and attacks between the −1and +1 subsites at the reducing end of the cellulose chain.

Applications

In addition to their established use as animal feed additives, cellulases and xylanaseshave a wide range of applications in food, paper and pulp, textile, fuel and chemicalindustries (Table 2.5) (Mandels, 1985; Gilbert and Hazlewood, 1993; Beguin andAubert, 1994; Bhat and Bhat, 1997). In a wider context, these enzymes havepotential uses in waste management, pollution treatment, protoplast production,genetic and protein engineering as well as other research and developmentprogrammes (Table 2.5) (Beguin and Aubert, 1994; Bayer et al., 1995; Bhat andBhat, 1997). Thus, cellulases and xylanases have a wide range of potentialapplications, but further research is necessary to exploit the commercial potentialof these enzymes to the fullest extent.

Conclusions

Basic and applied research carried out during the past decade has shown thatcellulases and xylanases are of great intrinsic interest, and have a range of noveland valuable uses in biotechnology. Early biochemical studies of purified enzymes,coupled with more recent molecular studies utilizing recombinant DNA tech-nologies, have provided detailed information on the biochemistry, architecture,







Indu

stry

and

rese

arch

Enzy

me

Appl

icat

ions

1.Fo

od,b

rew

ery

and

win

eC

ellu

lase

s

Xyla

nase

s

Extra

ctio

nan

dcl

arifi

catio

nof

fruit

juic

esan

doi

lfro

mse

eds;

incr

easi

ngth

eso

akin

gef

ficie

ncy

and

hom

ogen

eous

wat

erad

sorp

tion

ofce

real

s;im

prov

ing

the

dige

stio

nof

ball-

mille

dlig

noce

llulo

sean

dth

enu

tritiv

equ

ality

offe

rmen

ted

food

s;pr

oduc

tion

ofce

llo-o

ligos

acch

arid

esan

dsu

gard

eriv

ativ

es;r

elea

sing

flavo

urs,

enzy

mes

and

prot

eins

;im

prov

ing

the

filtra

tion

ofbe

eran

din

crea

sing

the

arom

aof

win

es(C

ough

lan,

1985

;Man

dels

,198

5;Be

guin

and

Aube

rt,19

94;C

aldi

niet

al.,

1994

)C

larif

icat

ion

offru

itju

ices

and

win

es;p

rodu

ctio

nof

food

thic

kene

rs;i

mpr

ovin

gth

ete

xtur

ean

dqu

ality

ofba

kery

prod

ucts

,pro

duct

ion

ofxy

lo-o

ligos

acch

arid

es,x

ylito

l(ar

tific

ials

wee

tene

r)an

dsu

gard

eriv

ativ

es(M

aat e

tal.,

1992

;Viik

ari e

tal.,

1993

)

2.An

imal

feed

Cel

lula

ses

and

xyla

nase

sAs

afe

edsu

pple

men

tfor

poul

try,p

igs

and

rum

inan

ts;p

retre

atm

ento

ffor

ages

;deh

ullin

gof

cere

algr

ains

;tre

atm

ent

ofsi

lage

(Bed

ford

and

Cla

ssen

,199

2;Se

lmer

-Ols

enet

al.,

1993

;Viik

arie

tal.,

1993

;Lew

iset

al.,

1996

)

3.Pa

pera

ndpu

lpXy

lana

ses

Pre-

blea

chin

gof

kraf

tpul

psan

dde

bark

ing;

refin

ing

pulp

fibre

san

dpr

epar

atio

nof

diss

olvi

ngpu

lps

(Won

gan

dSa

ddle

r,19

92,1

993;

Viik

ari e

tal.,

1993

;Rya

net

al.,

1998

)

4.Te

xtile

Cel

lula

ses

Bios

toni

ng(re

mov

ing

exce

ssdy

efro

mth

ede

nim

fabr

icin

pre-

fade

dje

ans)

,use

inw

ashi

ngpo

wde

rsfo

rret

aini

ngth

eor

igin

ality

ofco

tton

fabr

ics

afte

rsev

eral

was

hing

cycl

esby

rem

ovin

gth

ece

llulo

sem

icro

fibril

sas

wel

las

rest

orin

gth

eso

ftnes

san

dco

lour

brig

htne

ssof

cotto

nfa

bric

s(M

ande

ls,1

985;

Begu

inan

dAu

bert,

1994

)

5.W

aste

man

agem

ent

Cel

lula

ses

and

xyla

nase

sTr

eatm

ento

fagr

icul

tura

land

mun

icip

alw

aste

s;co

mpo

stin

gan

den

zym

atic

proc

essi

ngof

indu

stria

lpol

luta

nts

(Riv

ard

etal

.,19

94;P

hilip

pidi

san

dSm

ith,1

995;

Scot

teta

l.,19

95;B

hata

ndBh

at,1

997)

6.R

esea

rch

and

deve

lopm

ent

Cel

lula

ses

and

xyla

nase

sPr

oduc

tion

ofpr

otop

last

san

dhy

brid

stra

in;g

enet

icen

gine

erin

g;af

finity

purif

icat

ion;

unde

rsta

ndin

gth

epl

antc

ell

wal

lstru

ctur

ean

dch

emis

try(B

egui

nan

dAu

bert,

1994

;Bay

eret

al.,

1995

;Bha

tand

Bhat

,199

7)

Tabl

e2.

5.Ap

plic

atio

nsof

cellu

lase

san

dxy

lana

ses.



catalytic mechanisms and structure–function relationships of these importantenzymes. Two clear models have emerged for native cellulase–hemicellulase systems.In the first model, exemplified by aerobic fungi and bacteria, individual cellulases andhemicellulases are secreted freely by the host, without aggregating, and interactsynergistically to degrade their target substrates in plant cell walls. Anaerobic bacteria,and apparently also anaerobic fungi, exhibit an alternative mechanism for cellulase/hemicellulase hydrolysis in which the individual enzymes involved form an orderedmultienzyme complex or cellulosome. In both models, individual enzymes aretypically modular in structure and, in addition to one or more catalytic domains,contain non-catalytic functional domains, the most common of which is acellulose-binding domain that enables the enzyme to cluster on the surface ofcellulosic substrates. Structural studies of single enzymes have resulted in highresolution crystal structures for representatives of more than 20 glycosyl hydrolasesfamilies, and the identification of key catalytic residues.

Notwithstanding the progress of recent research, there are still manyunanswered questions regarding: (i) the biological significance and prevalence ofmultiple domains in cellulases and xylanases; (ii) the relative efficiencies of themultienzyme and disaggregated cellulase/hemicellulase systems; and (iii) the roles ofindividual enzymes during the solubilization of complex lignocellulosic substrates.Having the answers to such questions will not only enhance understanding of howcellulases and xylanases function, but should also enable these enzymes to be usedmore effectively as animal feed additives.

Acknowledgements

Financial support of the BBSRC is gratefully acknowledged.

References

Alzari, P.M., Souchon, H. and Dominguez, R. (1996) The crystal structure of endoglucanaseCelA, a family 8 glycosyl hydrolase from Clostridium thermocellum. Structure 4, 265–275.

Antranikian, G. (1994) Extreme thermophilic and hyperthermophilic microorganisms andtheir enzymes. Advanced Workshops in Biotechnology on ‘Extremophilic Microorganisms’,NTUA, Athens, Greece, pp. 1–30.

Atalla, R.H. and Vander Hart, D.L. (1984) Native cellulose: a composite of two distinctcrystalline forms. Science 223, 283–284.

Aubert, J.-P., Beguin, P. and Millet, J. (1988) Biochemistry and genetics of cellulose degrada-tion. Proceedings of FEMS Symposium No. 43, Academic Press, London, 428 pp.

Bailey, M.J. and Poutanen, K. (1989) Production of xylanolytic enzymes by strains ofAspergillus. Applied Microbiology and Biotechnology 30, 5–10.

Bailey, M.J., Biely, P. and Poutanen, K. (1992) Interlaboratory testing of methods for assay ofxylanase activity. Journal of Biotechnology 23, 257–270.

Barr, B.K., Hsieh, Y.-L., Ganem, B. and Wilson, D.B. (1996) Identification of two function-ally different classes of exocellulases. Biochemistry 35, 586–592.





Barras, F., Bortoli-German, I., Bauzan, M., Rouvier, J., Gey, C., Heyraud, A. and Henrissat,B. (1992) Stereochemistry of the hydrolysis reaction catalysed by endoglucanase Z fromErwinia chrysanthemi. FEBS Letters 300, 145–148.

Barrett, T., Suresh, C.G., Tolley, S.P., Dodson, E.J. and Hughes, M.A. (1995) The crystalstructure of a cyanogenic β-glucosidase from white clover, a family 1 glycosyl hydrolase.Structure 3, 951–960.

Bayer, E.A., Morag, E. and Lamed, R. (1994) The cellulosome – a treasure-trove for biotech-nology. Trends in Biotechnology 12, 379–386.

Bayer, E.A., Morag, E., Wilchek, M., Lamed, R., Yaron, S. and Shoham, Y. (1995)Cellulosome domains for novel biotechnological application. In: Petersen, S.B., Svens-son, B. and Pedersen, S. (eds) Carbohydrate Bioengineering. Progress in Biotechnology,Vol. 10. Elsevier, Amsterdam, The Netherlands, pp. 251–260.

Bayer, E.A., Morag, E., Lamed, R., Yaron, S. and Shoham, Y. (1998) Cellulosome structure:four-pronged attack using biochemistry, molecular biology, crystallography andbioinformatics. In: Claeyssens, M., Nerinckx, W. and Piens, K. (eds) Carbohydrases fromTrichoderma reesei and Other Microorganisms – Structure, Biochemistry, Genetics andApplications. Proceedings of Tricel ’97 Meeting, The Royal Society of Chemistry,Cambridge, UK, pp. 39–72.

Bedford, M.R. and Classen, H.L. (1992) The influence of dietary xylanase on intestinalviscosity and molecular weight distribution of carbohydrates in rye-fed broiler chicks.In: Visser, J., Beldman, G., Kusters-van Someren, M.A. and Voragen, A.G.J. (eds) Xylansand Xylanases. Progress in Biotechnology, Vol. 7, Elsevier, Amsterdam, The Netherlands,pp. 361–370.

Beguin, P. and Alzari, P.M. (1998) The cellulosome of Clostridium thermocellum. BiochemicalSociety Transactions 26, 178–184.

Beguin, P. and Aubert, J.-P. (1994) The biological degradation of cellulose. FEMS Micro-biology Review 13, 25–58.

Beguin, P. and Lemaire, M. (1996) The cellulosome: an exocellular, multiprotein complexspecialised in cellulose degradation. Critical Reviews in Biochemistry and Molecular Biology31, 201–236.

Beguin, P., Millet, J., Chauvaux, S., Salamitou, S., Tokatlidis, K., Navas, J., Fujino, T.,Lemaire, M., Raynaud, O., Daniel, M.-K. and Aubert, J.-P. (1992) Bacterial cellulases.Biochemical Society Transactions 20, 42–46.

Beguin, P., Chauvaux, S., Chaveroche, M.K., Gugliemi, G., Kataeva, I., Leibovitz, E. andMiras, I. (1998) The cellulosome: a versatile system for coupling cellulolytic enzymes andattaching them to the cell surface. In: Claeyssens, M., Nerinckx, W. and Piens, K. (eds)Carbohydrases from Trichoderma reesei and other Microorganisms – Structure, Biochemis-try, Genetics and Applications. Proceedings of Tricel ’97 Meeting, The Royal Society ofChemistry, Cambridge, UK, pp. 66–72.

Bennett, N.A., Ryan, J., Biely, P., Vrsanska, M., Kremnicky, L., Macris, B.J., Kekos, D.,Christakopoulos, P., Katapodis, P., Claeyssens, M., Nerinckx, W., Ntauma, P. and Bhat,M.K. (1998) Biochemical and catalytic properties of an endoxylanase purified from theculture filtrate of Thermomyces lanuginosus ATCC 46882. Carbohydrate Research 306,445–455.

Bhat, M.K. and Bhat, S. (1997) Cellulose degrading enzymes and their potential industrialapplications. Biotechnology Advances 15, 583–620.

Bhat, M.K. and Bhat, S. (1998) Clostridium thermocellum cellulosome: dissociation, isolationand characterisation of subunits and the potential biotechnological implications. In:





Pandalai, S.G. (ed.) Recent Research Developments in Biotechnology and Bioengineering,Vol. 1, Part I. Research Signpost, Trivandrum, India, pp. 59–84.

Bhat, K.M. and Maheshwari, R. (1987) Sporotrichum thermophile: growth, cellulose degrada-tion and cellulase activity. Applied and Environmental Microbiology 53, 2175–2182.

Bhat, K.M. and Wood, T.M. (1989) Multiple forms of endo-1,4-β-glucanase in theextracellular cellulase of Penicillium pinophilum. Biotechnology and Bioengineering 33,1242–1248.

Bhat, K.M., McCrae, S.I. and Wood, T.M. (1989) The endo-(1-4)-β-D-glucanase system ofPenicillium pinophilum cellulase: isolation, purification and characterisation of five majorendoglucanase components. Carbohydrate Research 190, 279–297.

Bhat, K.M., Hay, A.J., Claeyssens, M. and Wood, T.M. (1990) Study of the mode of actionand site specificity of the endo-(1-4)-β-D-glucanases of the fungus Penicilliumpinophilum with normal, 1-3H labelled, reduced and chromogenic cello-oligosaccharides.Biochemical Journal 266, 371–378.

Bhat, K.M., Gaikwad, J.S. and Maheshwari, R. (1993) Purification and characterisation of anextracellular β-glucosidase from the thermophilic fungus Sporotrichum thermophile andits influence on cellulase activity. Journal of General Microbiology 139, 2825–2832.

Bhat, S., Goodenough, P.W., Bhat, M.K. and Owen, E. (1994) Isolation of four major sub-units from Clostridium thermocellum cellulosome and their synergism in the hydrolysis ofcrystalline cellulose. International Journal of Biological Macromolecules 16, 335–342.

Biely, P. (1985) Microbial xylanolytic enzymes. Trends in Biotechnology 3, 286–290.Biely, P. (1993) Biochemical aspects of the production of microbial hemicellulases. In:

Coughlan, M.P. and Hazlewood, G.P. (eds) Hemicellulose and Hemicellulases. PortlandPress, London, pp. 29–51.

Biely, P., Mislovicova, D. and Toman, R. (1985) Soluble chromogenic substrates for the assayof endo-1,4-β-glucanases and endo-1,4-β-xylanases in gels. Analytical Biochemistry 144,142–146.

Biely, P., MacKenzie, C.R., Puls, J. and Schneider, H. (1986) Cooperativity of esterases andxylanases in the enzymatic degradation of acetyl xylan. Bio/Technology 4, 731–733.

Biely, P., Vrsanska, M. and Kucar, S. (1992) Identification and mode of action ofendo-(1->4)-β-xylanases. In: Visser, J., Beldman, G., Kusters-van Someren, M.A. andVoragen, A.G.J. (eds) Xylans and Xylanases. Progress in Biotechnology, Vol. 7. Elsevier,Amsterdam, The Netherlands, pp. 81–95.

Biely, P., Vrsanska, M., Tenkanen, M. and Kluepfel, D. (1997) Endo-β1,4-xylanase families:differences in catalytic properties. Journal of Biotechnology 57, 151–166.

Black, G.W., Hazlewood, G.P., Millward-Sadler, S., Laurie, J. and Gilbert, H.J. (1995)A modular xylanase containing a novel non-catalytic xylan-specific binding domain.Biochemical Journal 307, 191–195.

Black, G.W., Rixon, J.E., Clarke, J.H., Hazlewood, G.P., Ferreira, L.M.A., Bolam, D.N. andGilbert, H.J. (1997) Cellulose binding domains and linker sequences potentiate theactivity of hemicellulases against complex substrates. Journal of Biotechnology 57, 59–69.

Brown, A.J., Collin, S.A. and Wood, T.M. (1987) Development of a medium for highcellulase, xylanase and β-glucosidase production by a mutant strain (NTG III/6) of thecellulolytic fungus Penicillium pinophilum. Enzyme Microbial Technology 9, 355–360.

Caldini, C., Bonomi, F., Pifferi, P.G., Lanzarini, G. and Galante, Y.M. (1994) Kinetic andimmobilization studies on fungal glycosidases for aroma enhancement in wine. Enzymeand Microbial Technology 16, 286–291.

Campbell, R.L., Rose, D.R., Wakarchuck, W.W., To, R., Sung, W. and Yaguchi, M. (1993)A comparison of the structures of the 20 kDa xylanases from Trichoderma harzianum and





Bacillus circulans. In: Suominen, P. and Reinikainen, T. (eds) Proceeding of the 2nd TricelSymposium on Trichoderma reesei and other hydrolases. Foundations for Biotechnical andIndustrial Fermentation Research 8, Helsinki, pp. 63–72.

Christakopoulos, P., Goodenough, P.W., Kekos, D., Macris, B.J., Claeyssens, M. and Bhat,M.K. (1994a) Purification and characterisation of an extracellular β-glucosidase withtransglycosylation and exoglucosidase activities from Fusarium oxysporum. EuropeanJournal of Biochemistry 224, 379–385.

Christakopoulos, P., Bhat, M.K., Kekos, D. and Macris, B.J. (1994b) Enzymatic synthesis oftrisaccharides and alkyl β-D-glucosides by the transglycosylation reaction of β-glucosidasefrom Fusarium oxysporum. International Journal of Biological Macromolecules 16,331–334.

Christakopoulos, P., Goodenough, P.W., Kekos, D., Macris, B.J. and Bhat, M.K. (1994c)β-Glucosidase catalysed synthesis of trisaccharides. Biotechnology Letters 16, 587–592.

Ciruela, A., Gilbert, H.J., Ali, B.R.S. and Hazlewood, G.P. (1998) Synergistic interaction ofthe cellulosome integrating protein (CipA) from Clostridium thermocellum with acellulosomal endoglucanase. FEBS Letters 422, 221–224.

Claeyssens, M. and Henrissat, B. (1992) Specificity mapping of cellulolytic enzymes:classification into families of structurally related proteins confirmed by biochemicalanalysis. Protein Science 1, 1293–1297.

Claeyssens, M., van Tilbeurgh, H., Tomme, P., Wood, T.M. and McCrae, S.I. (1989)Comparison of the specificities of the CBH isolated from Penicillium pinophilum andTrichoderma reesei. Biochemical Journal 261, 819–825.

Claeyssens, M., van Tilbeurgh, H., Kamerling, J.P., Berg, J., Vrsanska, M. and Biely, P.(1990a) Studies of the cellulolytic system of the filamentous fungus Trichoderma reeseiQM 9414. Substrate specificity and transfer activity of endoglucanase I. BiochemicalJournal 270, 251–256.

Claeyssens, M., Tomme, P., Brewer, C.F. and Hehre, E.J. (1990b) Stereochemical course ofhydrolysis and hydration reactions catalysed by CBHs I and II from Trichoderma reesei.FEBS Letters 263, 89–92.

Claeyssens, M., Nerinckx, W. and Piens, K. (1998) Carbohydrases from Trichodermareesei and other microorganisms – structure, biochemistry, genetics and applications.Proceedings of Tricel ’97 Meeting, The Royal Society of Chemistry, Cambridge, UK,351 pp.

Clarke, J.H., Davidson, K., Gilbert, H.J., Fontes, C.M.A. and Hazlewood, G.P. (1996) Amodular xylanase from mesophilic Cellulomonas fimi contains the same cellulose-bindingand thermostabilizing domains as xylanases from thermophilic bacteria. FEMS Microbiol-ogy Letters 139, 27–35.

Comtat, J. and Joseleau, J.P. (1981) Mode of action of a xylanase and its significance for thestructural investigation of the branched L-arabino-D-glucurono-D-xylan from redwood(Sequoia sempervirens). Carbohydrate Research 95, 101–112.

Coughlan, M.P. (1985) Cellulases: production, properties and applications. BiochemicalSociety Transactions 13, 405–406.

Coughlan, M.P. (1988) Staining techniques for the detection of the individual componentsof cellulolytic enzyme systems. In: Wood, W.A. and Kellogg, S.T. (eds) Methods inEnzymology, Vol. 160. Academic Press, London, pp. 135–144.

Coughlan, M.P. (1992) Towards an understanding of the mechanism of action of mainchain-hydrolysing xylanases. In: Visser, J., Beldman, G., Kusters-van Someren, M.A. andVoragen, A.G.J. (eds) Xylans and Xylanases. Progress in Biotechnology, Vol. 7. Elsevier,Amsterdam, The Netherlands, pp. 111–139.





Coughlan, M.P. and Hazlewood, G.P. (1993a) β-1,4-D-Xylan-degrading enzyme systems:biochemistry, molecular biology and applications. Biotechnology and Applied Biochemistry17, 259–289.

Coughlan, M.P. and Hazlewood, G.P. (1993b) Hemicellulose and Hemicellulases. PortlandPress, London, 152 pp.

Coughlan, M.P. and Ljungdahl, L.G. (1988) Comparative biochemistry of fungal andbacterial cellulolytic enzyme systems. In: Aubert, J.-P., Beguin, P. and Millet, J. (eds) Bio-chemistry and Genetics of Cellulose Degradation. FEMS Symposium No. 43. AcademicPress, London, pp. 11–30.

Coughlan, M.P., Tuohy, M.A., Filho, E.X.F., Puls, J., Claeyssens, M., Vrsanska, M. andHughes, M.M. (1993) Enzymological aspects of microbial hemicellulases with emphasison fungal systems. In: Coughlan, M.P. and Hazlewood, G.P. (eds.) Hemicellulose andHemicellulases. Portland Press, London, pp. 53–84.

Coutinho, J.B., Gilkes, N.R., Kilburn, D.G., Warren, R.A.J. and Miller, R.C. Jr (1993) Thenature of the cellulose-binding domain affects the activities of a bacterial endoglucanaseon different forms of cellulose. FEMS Microbiology Letters 113, 211–218.

Davies, G.J. (1998) Structural studies on cellulases. Biochemical Society Transactions 26,167–172.

Davies, G.J. and Henrissat, B. (1995) Structures and mechanisms of glycosyl hydrolases.Structure 3, 853–859.

Davies, G.J., Dodson, G.G., Hubbard, R.E., Tolley, S.P., Dauter, Z., Wilson, K.S., Hjort, C.,Mikkelsen, J.M., Rasmussen, G. and Schulein, M. (1993) Structure and function ofendoglucanase V. Nature 365, 362–364.

Davies, G.J., Tolley, S.P., Henrissat, B., Hjort, C. and Schulein, M. (1995) Structures ofoligosaccharides-bound forms of the endoglucanase V from Humicola insolens at 1.9Åresolution. Biochemistry 34, 16210–16220.

Dekker, R.F.H. (1985) Biodegradation of hemicelluloses. In: Higuchi, T. (ed.) Biosynthesisand Biodegradation of Wood Components. Academic Press, Orlando, Florida, pp.505–533.

Derewenda, U., Swenson, L., Green, R., Wei, Y., Morosoli, R., Shareck, F., Kluepfel, D. andDerewenda, Z.S. (1994) Crystal structure, at 2.6 Å resolution, of the Streptomyces lividansxylanase A, a member of the F family of β-1,4-D-glycanases. Journal of BiologicalChemistry 269, 20811–20814.

Deshpande, M.V., Eriksson, K.E. and Pettersson, L.G. (1984) An assay of selective determina-tion of exo-1,4-β-glucanase in a mixture of cellulolytic enzymes. Analytical Biochemistry138, 481–487.

Deshpande, V., Lachke, A., Mishra, C., Keskar, S. and Rao, M. (1986) Mode of actionand properties of xylanase and β-xylosidase from Neurospora crassa. Biotechnology andBioengineering 28, 1832–1837.

Din, N., Gilkes, N.R., Tekant, B., Miller, R.C., Warren, R.A.J. and Kilburn, D.G. (1991)Non-hydrolytic disruption of cellulose fibres by the binding domain of a bacterialcellulase. Bio/Technology 9, 1096–1099.

Divne, C., Stahlberg, J., Reinikainen, T., Ruohonen, L., Pettersson, G., Knowles, J.K.C.,Teeri, T.T. and Jones, T.A. (1994) The three-dimensional crystal structure of thecatalytic core of CBH I from Trichoderma reesei. Science 265, 524–528.

Dominguez, R., Souchon, H., Spinelli, S., Dauter, Z., Wilson, K.S., Chauvaux, S., Beguin, P.and Alzari, P.M. (1995) A common protein fold and similar active site in two distinctfamilies of β-glycanases. Nature Structure Biology 2, 569–576.





Ducros, V., Czizek, M., Belaich, A., Gaudin, C., Fierobe, H.P., Belaich, L.P., Davies, G.J.and Haser, R. (1995) Crystal structure of the catalytic domain of a bacterial cellulasebelonging to family 5. Structure 3, 939–949.

El-Gogary, S., Leite, A., Crivellaro, O., Eveleigh, D.E. and El-Dorry, H. (1989) Mechanismby which cellulose triggers CBH I gene expression in Trichoderma reesei. Proceedings of theNational Academy of Sciences USA 86, 6138–6141.

Eriksson, K.E. and Goodell, E.W. (1974) Pleiotropic mutants of the wood-rotting fungusPolyporus adustus lacking cellulase, mannanase, and xylanase. Canadian Journal ofMicrobiology 20, 371–378.

Eriksson, K.E. and Wood, T.M. (1985) Biodegradation of cellulose. In: Higuchi, T. (ed.)Biosynthesis and Biodegradation of Wood Components. Academic Press, New York,pp. 469–503.

Fan, L.T., Lee, Y.-H. and Beardmore, D.H. (1980) Major chemical and physical features ofcellulosic materials as substrates for enzymatic hydrolysis. In: Fiechter, A. (ed.) Advancesin Biochemical Engineering, Vol. 14. Springer-Verlag, Berlin, pp. 101–117.

Farkas, V., Liskova, M. and Biely, P. (1985) Novel media for detection of microbial producersof cellulase and xylanase. FEMS Microbiology Letters 28, 137–140.

Faulds, C.B. and Williamson, G. (1991) The purification and characterisation of 4-hydroxy-3-methoxycinnamic (ferulic) acid esterase from Streptomyces olivochromogenes. Journal ofGeneral Microbiology 137, 2339–2345.

Fierobe, H.-P., Bagnara-Tardif, C., Gaudin, C., Guerlesquin, F., Sauve, P., Belaich, A. andBelaich, J.-P. (1993) Purification and characterisation of endoglucanase C fromClostridium cellulolyticum: catalytic comparison with endoglucanase A. European Journalof Biochemistry 217, 557–565.

Flint, H.J. and Zhang, J.-X. (1992) Molecular biology of xylanases from the rumen cellulolyticanaerobe Ruminococcus flavefaciens 17-multiple genes and bifunctional enzymes. In:Visser, J., Beldman, G., Kusters-van Someren, M.A. and Voragen, A.G.J. (eds) Xylanand Xylanases. Progress in Biotechnology, Vol. 7. Elsevier, Amsterdam, The Netherlands,pp. 301–308.

Flint, H.J., Martin, J., McPherson, C.A., Daniel, A.S. and Zhang, J.X. (1993) A bifunctionalenzyme with separate xylanase and β(1,3-1,4)-glucanase domains encoded by the xynDgene of Ruminococcus flavefaciens. Journal of Bacteriology 175, 2943–2951.

Fontes, C.M.G.A., Hazlewood, G.P., Morag, E., Hall, J., Hirst, B.M. and Gilbert, H.J.(1995) Evidence for a general role for non-catalytic thermostabilizing domains inxylanases from thermophilic bacteria. Biochemical Journal 307, 151–158.

Frederick, M.M., Kiang, C.-H., Frederick, J.R. and Reilly, P.J. (1985) Purification andcharacterisation of endo-xylanases from Aspergillus niger. I. Two isozymes active on xylanbackbones near branch points. Biotechnology and Bioengineering 27, 525–528.

Fritscher, C., Messner, R. and Kubicek, C.P. (1990) Cellobiose metabolism and CBH Ibiosynthesis by Trichoderma reesei. Experimental Mycology 14, 405–415.

Gebler, J.C., Gilkes, N.R., Claeyssens, M., Wilson, D.B., Beguin, P., Wakarchuk, W.,Kilburn, D.G., Miller, R.C., Warren, R.A.J. and Withers, S.G. (1992) Stereospecifichydrolysis catalysed by related β-1,4-glycanases and β-1,4-xylanases. Journal of BiologicalChemistry 267, 12559–12561.

Gilbert, H.J. and Hazlewood, G.P. (1993) Bacterial cellulases and xylanases. Journal of GeneralMicrobiology 139, 187–194.

Giligan, W. and Reese, E.T. (1954) Evidence for multiple components in microbial cellulases.Canadian Journal of Microbiology 1, 90–107.





Gilkes, N.R., Langsford, M.L., Kilburn, D.G., Miller, R.C. Jr and Warren, R.A.J. (1984)Mode of action and substrate specificities of cellulases from cloned bacterial genes.Journal of Biological Chemistry 259, 10455–10459.

Gilkes, N.R., Claeyssens, M., Aebersold, R., Henrissat, B., Meinke, A., Morrison, H.D., Kil-burn, D.G., Warren, R.A.J. and Miller, R.C. (1991a) Structural and functional relation-ships in two families of β-1,4-glycanases. European Journal of Biochemistry 202, 367–377.

Gilkes, N.R., Henrissat, B., Kilburn, D.G., Miller, R.C. Jr and Warren, R.A.J. (1991b)Domains in microbial β-1,4-glycanases: sequence conservation, function, and enzymefamilies. Microbiology Review 55, 303–315.

Gilkes, N.R., Kwan, E., Kilburn, D.G., Miller, R.C. and Warren, R.A.J. (1997) Attack ofCM-cellulose at opposite ends by two CBHs from Cellulomonas fimi. Journal of Biotech-nology 57, 83–90.

Gong, C.S., Ladisch, M.R. and Tsao, T.G. (1979) Biosynthesis, purification and mode ofaction cellulases of Trichoderma reesei. Advances in Chemistry Series 181, 261–287.

Groleau, D. and Forsberg, C.W. (1981) Cellulolytic activity of the rumen bacteriumBacteroides succinogenes. Canadian Journal of Microbiology 27, 517–530.

Gruber, K., Klintschar, G., Hayn, M., Schlacher, A., Steiner, W. and Kratky, C. (1998)Thermophilic xylanase from Thermomyces lanuginosus: high resolution X-ray structureand modelling studies. Biochemistry 37, 13475–13485.

Hall, J., Black, G.W., Ferreira, L.M.A., Millward-Sadler, S.J., Ali, B.R.S., Hazlewood, G.P.and Gilbert, H.J. (1995) The non-catalytic cellulose binding domain of a novel cellulasefrom Pseudomonas fluorescens subsp. cellulosa is important for the efficient hydrolysis ofAvicel. Biochemical Journal 309, 749–756.

Haltrich, D., Sebesta, B. and Steiner, W. (1995) Induction of xylanase and cellulase inSchizophilum commune. In: Saddler, J.N. and Penner, M.H. (eds) Enzymatic Degradationof Insoluble Carbohydrates. ACS Symposium Series 618, American Chemical Society,Washington, DC, pp. 305–318.

Harris, G.W., Jenkins, J.A., Connerton, I., Cummings, N., Lo Leggio, L., Scott, M.,Hazlewood, G.P., Laurie, J.I., Gilbert, H.J. and Pickersgill, R.W. (1994) Structure of thecatalytic core of the family F xylanase from Pseudomonas fluorescens and identification ofthe xylopentaose-binding site. Structure 2, 1107–1116.

Hartley, R.D. and Ford, C.W. (1989) Phenolic constitutents of plant cell walls and wallbiodegradability. In: Lewis, N.G. and Paice, M.G. (eds) Plant Cell Wall Polymers,Biogenesis and Biodegradation. ACS Symposium Series 399, American Chemical Society,Washington, DC, pp. 135–145.

Havukainen, R., Torronen, A., Laitinen, T. and Rouvinen, J. (1996) Covalent binding ofthree epoxyalkyl xylosides to the active site of endo-1,4-xylanase II from Trichodermareesei. Biochemistry 35, 9617–9624.

Hayashi, H., Takagi, K., Fukumura, M., Kimura, T., Karita, S., Sakka, K. and Ohmiya, K.(1997) Sequence of xynC and properties of xynC, a major component of the Clostridiumthermocellum. Journal of Bacteriology 179, 4246–4253.

Hazlewood, G.P. and Gilbert, H.J. (1998a) Structure and function analysis of Pseudomonasplant cell wall hydrolases. Progress in Nucleic Acid Research and Molecular Biology 61,211–241.

Hazlewood, G.P. and Gilbert, H.J. (1998b) Structure–function relationships in thecellulase–hemicellulase system of anaerobic fungi. In: Claeyssens, M., Nerinckx, W.and Piens, K. (eds) Carbohydrases from Trichoderma reesei and Other Microorganisms –Structure, Biochemistry, Genetics and Applications. Proceedings of Tricel ’97 Meeting, TheRoyal Society of Chemistry, Cambridge, UK, pp. 147–155.





Hazlewood, G.P. and Gilbert, H.J. (1998c) Structure and function analysis of Pseudomonasplant cell wall hydrolases. In: Claeyssens, M., Nerinckx, W. and Piens, K. (eds) Carbo-hydrases from Trichoderma reesei and Other Microorganisms – Structure, Biochemistry,Genetics and Applications. Proceedings of Tricel ’97 Meeting, The Royal Society ofChemistry, Cambridge, UK, pp. 156–165.

Hazlewood, G.P. and Gilbert, H.J. (1998d) Structure and function analysis of Pseudomonasplant cell wall hydrolases. Biochemical Society Transactions 26, 185–189.

Hazlewood, G.P., Romaniec, M.P.M., Davidson, K., Grepinet, O., Beguin, P., Millet, J.,Raynaud, O. and Aubert, J.-P. (1988) A catalogue of Clostridium thermocellumendoglucanase, β-glucosidase and xylanase genes cloned in Escherichia coli. FEMSMicrobiology Letters 51, 231–236.

Henrissat, B. (1998) Glycosidase families. Biochemical Society Transactions 26, 153–156.Henrissat, B. and Bairoch, A. (1993) New families in the classification of glycosyl hydrolase

based on amino acid sequence similarities. Biochemical Journal 293, 781–788.Henrissat, B. and Bairoch, A. (1996) Updating the sequence-based classification of glycosyl

hydrolases. Biochemical Journal 316, 695–696.Henrissat, B. and Davies, G. (1997) Structural and sequence-based classification of glycosyl

hydrolases. Current Opinion in Structural Biology 7, 637–644.Henrissat, B., Claeyssens, M., Tomme, P., Lemesle, L. and Mornon, J.P. (1989) Cellulase

families revealed by hydrophobic cluster analysis. Gene 81, 83–95.Hoshino, E., Kanda, T., Sasaki, Y. and Nisizawa, K. (1992) Adsorption, mode of action

of exo- and endo-cellulases from Irpex lacteus (Polyporus tulipiferae) on cellulose withdifferent crystallinities. Journal of Biochemistry 111, 600–605.

Ilmen, M., Saloheimo, A., Onnela, M.-L. and Penttila, M.E. (1997) Regulation of cellulasegene expression in the filamentous fungus Trichoderma reesei. Applied and EnvironmentalMicrobiology 63, 1298–1306.

Irwin, D., Jung, E. and Wilson, D. (1994) Characterisation of a Thermomonospora fuscaxylanase. Applied and Environmental Microbiology 60, 763–770.

Johnson, P.E., Joshi, M.D., Tomme, P., Kilburn, D.G. and McIntosh, L.P. (1996) Structureof the N-terminal CBD of Cellulomonas fimi CenC determined by nuclear magneticresonance spectroscopy. Biochemistry 35, 14381–14394.

Juy, M., Amit, A.G., Alzari, P.M., Poljak, R.J., Claeyssens, M., Beguin, P. and Aubert, J.-P.(1992) Three-dimensional structure of a thermostable bacterial cellulase. Nature 357,89–91.

Kalogiannis, S., Owen, E., Beever, D.E., Claeyssens, M., Nerinckx, W. and Bhat, M.K.(1996) Characterisation of the xylanolytic system of the thermophilic fungusThermoascus aurantiacus. SCI Lecture Paper Series 78, pp. 1–11.

Kataeva, I., Gugliemi, G. and Beguin, P. (1997) Interaction between Clostridium thermocellumendoglucanase CelD and polypeptides derived from the cellulosome integrating proteinCipA: stoichiometry and cellulolytic activity of the complexes. Biochemical Journal 326,617–624.

Khandke, K.M., Vithayathil, P.J. and Murthy, S.K. (1989) Purification of xylanase,β-glucosidase, endocellulase, and exocellulase from a thermophilic fungus, Thermoascusaurantiacus. Archives of Biochemistry and Biophysics 274, 491–500.

Kluepfel, D. (1988) Screening of prokaryotes for cellulose- and hemicellulose-degradingenzymes. In: Wood, W.A. and Kellogg, S.T. (eds) Methods in Enzymology, Vol. 160.Academic Press, London, pp.180–186.

Klyosov, A. (1990) Trends in biochemistry and enzymology of cellulose degradation.Biochemistry 29, 10577–10585.





Kormelink, F.J.M., Gruppen, H., Vietor, R.J. and Voragen, A.G.J. (1993) Mode of action ofthe xylan-degrading enzymes from Aspergillus awamori on alkali-extractable cerealarabinoxylans. Carbohydrate Research 249, 355–367.

Kraulis, P., Clore, M., Nilges, M., Jones, T.A., Pettersson, G., Knowles, J. and Gronenborn,A. (1989) Determination of the three-dimensional solution structure of the C-terminaldomain of CBH I from Trichoderma reesei. Biochemistry 28, 7241–7257.

Kubackova, M., Karacsonyi, S., Bilisics, L. and Toman, R. (1978) Some properties ofan endo-1,4-β-D-xylanase from the ligniperdous fungus Trametes hirsuta. FoliaMicrobiologica 23, 202–209.

Kubicek, C.P. (1992) The cellulase proteins of Trichoderma reesei: structure, multiplicity,mode of action and regulation of formation. Advances in Biochemical Engineering 45,1–27.

Kubicek, C.P., Muhlbauer, G., Krotz, M., John, E. and Kubicek-Pranz, E.M. (1988)Properties of a conidial bound cellulase enzyme system from Trichoderma reesei. Journalof General Microbiology 134, 1215–1222.

Kubicek, C.P., Messner, R., Gruber, F., Mach, R.L. and Kubicek-Pranz, E.M. (1993) TheTrichoderma reesei cellulase regulatory puzzle – from the interior life of a secretory fungus.Enzyme and Microbial Technology 15, 90–99.

Lachke, A.H. and Deshpande, M.V. (1988) Sclerotium rolfsii: status in cellulase research.FEMS Microbiology Review 54, 177–194.

Ladisch, M.R., Lin, K.W., Voloch, M. and Tsao, G.T. (1983) Process considerations in theenzymatic hydrolysis of biomass. Enzyme Microbial Technology 5, 82–100.

Lamed, R. and Bayer, E.A. (1987) Cellulose degradation by thermophilic anaerobic bacteria.In: Weimer, P.J. and Haigler, C.A. (eds) Biosynthesis and Biodegradation of Cellulose andCellulose Materials. Marcel Dekker, New York, pp. 1–20.

Lamed, R. and Bayer, E.A. (1988) The cellulosome of Clostridium thermocellum. Advances inApplied Microbiology 33, 1–46.

Lamed, R., Naimark, J., Morgenstern, E. and Bayer, E.A. (1987) Specialised cell surfacestructures of cellulolytic bacteria. Journal of Bacteriology 169, 3792–3800.

Lee, H., Rebecca, J.B.To., Latta, R.K., Biely, P. and Schneider, H. (1987) Some properties ofextracellular acetylxylan esterase produced by the yeast Rhodotorula mucilaginosa. Appliedand Environmental Microbiology 53, 2831–2834.

Lewis, G.E., Hunt, C.W., Sanchez, W.K., Treacher, R., Pritchard, G.T. and Feng, P. (1996)Effect of direct-fed fibrolytic enzymes on the digestive characteristics of a forage-baseddiet fed to beef steers. Journal of Animal Science 74, 3020–3028.

Linder, M. and Teeri, T.T. (1997) The roles and function of cellulose-binding domains.Journal of Biotechnology 57, 15–28.

Linder, M., Lindeberg, G., Reinikainen, T., Teeri, T.T. and Pettersson, G. (1995) Thedifference in affinity between two fungal CBDs is dominated by a single amino acidsubstitution. FEBS Letters 372, 96–98.

Linder, M., Salovuori, I., Ruohonen, L. and Teeri, T.T. (1996) Characterisation of a doubleCBD: synergistic high-affinity binding to cellulose. Journal of Biological Chemistry 271,21268–21272.

Maat, J., Roza, M., Verbakel, J., Stam, H., Santos da Silva, M.J., Bosse, M., Egmond, M.R.,Hagemans, M.L.D., van Gorcom, R.F.M., Hessing, J.G.M., van den Hondel,C.A.M.J.J. and van Rotterdam, C. (1992) Xylanases and their applications in bakery.In: Visser, J., Beldman, G., Kusters-van Someren, M.A. and Voragen, A.G.J. (eds) Xylansand Xylanases. Progress in Biotechnology, Vol. 7. Elsevier, Amsterdam, The Netherlands,pp. 349–360.





Mackenzie, L.F., Davies, G.J., Schulein, M. and Withers, S.G. (1997) Identification of thecatalytic nucleophile of endoglucanase I from Fusarium oxysporum by mass spectroscopy.Biochemistry 36, 5893–5901.

Mandels, M. (1985) Applications of cellulases. Biochemical Society Transactions 13, 414–415.Marsh, C.B., Merola, G.V. and Simpson, M.E. (1953) Experiments with an alkali

swelling-centrifuge test applied to cotton fibre. Textile Research Journal 23, 831.Marx-Figini, M. and Schultz, G.V. (1966) Zur biosynthese der cellulose. Naturwissenschaften

53, 466–474.Matte, A. and Forsberg, C.W. (1992) Purification, characterisation, and mode of action of

endoxylanases 1 and 2 from Fibrobacter succinogenes S85. Applied and EnvironmentalMicrobiology 58, 157–168.

McHale, A. and Coughlan, M.P. (1980) Synergistic hydrolysis of cellulose by components ofthe extracellular cellulase system of Talaromyces emersonii. FEBS Letters 117, 319–322.

Meinke, A., Gilkes, N.R., Kwan, E., Kilburn, D.G., Warren, R.A.J. and Miller, R.C. (1994)Cellobiohydrolase A (cbhA) from the cellulolytic bacterium Cellulomonas fimi is aβ-1,4-exocellobiohydrolase analogous to Trichoderma reesei CBHII. Molecular Micro-biology 12, 413–422.

Miller, G.L. (1959) Use of dinitrosalicylic acid reagent for determination of reducing sugars.Analytical Chemistry 31, 426–428.

Millward-Sadler, S.J., Poole, D.M., Henrissat, B., Hazlewood, G.P., Clarke, J.H. andGilbert, H.J. (1994) Evidence for a general role for high-affinity non-catalytic cellulosebinding domains in microbial plant cell wall hydrolases. Molecular Microbiology 11,375–382.

Millward-Sadler, S.J., Davidson, K., Hazlewood, G.P., Black, G.W., Gilbert, H.J. andClarke, J.H. (1995) Novel cellulose-binding domains, Nod B homologues and conservedmodular architecture in xylanases from the aerobic soil bacteria Pseudomonas fluorescenssubsp. cellulosa and Cellvibro mixtus. Biochemical Journal 312, 39–48.

Moloney, A.P., Considine, P.J. and Coughlan, M.P. (1983) Cellulose hydrolysis by thecellulases produced by Talaromyces emersonii when grown on different inducingsubstrates. Biotechnology and Bioengineering 25, 1169–1173.

Nishitani, K. and Nevins, D.J. (1991) Glucuronoxylan xylanohydrolase: a unique xylanasewith the requirement for appendant glucuronosyl units. Journal of Biological Chemistry266, 6539–6543.

Nisizawa, T., Suzuki, H. and Nisizawa, K. (1972) Catabolite repression of cellulase formationin Trichoderma viride. Journal of Biochemistry (Tokyo) 71, 999–1007.

Nummi, M., Fox, P.C., Niku-Paavola, M.L. and Enari, T.M. (1981) Nephelometric andturbidometric assays of cellulase activity. Analytical Biochemistry 116, 133–136.

Nummi, M., Perrin, J.M., Niku-Paavola, M.L. and Enari, T.M. (1985) Measurement ofxylanase activity with insoluble xylan substrate. Biochemical Journal 226, 617–620.

Petre, D., Millet, J., Longuin, R., Beguin, P., Girard, H. and Aubert, J.-P. (1986) Purificationand properties of the endoglucanase C of Clostridium thermocellum produced inEscherichia coli. Biochimie 68, 687–695.

Philippidis, G.P. and Smith, T.K. (1995) Limiting factors in the simultaneoussaccharificiation and fermentation process for conversion of cellulosic biomass to fuelethanol. Applied Biochemistry and Biotechnology 51/52, 117–124.

Pickersgill, R., Harris, G., Lo Leggio, L., Mayans, O. and Jenkins, J. (1998) Superfamilies: the4/7 superfamily of βα-barrel glycosidases and the right-handed parallel β-helixsuperfamily. Biochemical Society Transactions 26, 190–197.





Puls, J. and Schuseil, J. (1993) Chemistry of hemicellulose: relationship between hemicellulosestructure and enzymes required for hydrolysis. In: Coughlan, M.P. and Hazlewood, G.P.(eds) Hemicellulose and Hemicellulases. Portland Press, London, pp. 1–27.

Puls, J., Schmidt, O. and Ganzow, C. (1987) α-Glucuronidase in two microbial xylanolyticsystems. Enzyme and Microbial Technology 9, 83–88.

Reese, E.T., Si, R.G.H. and Levinson, H.S. (1950) The biological degradation of solublecellulose derivatives and its relationship to the mechanism of cellulose hydrolysis. Journalof Bacteriology 59, 485–497.

Reinikainen, T., Ruohonen, L., Nevanen, T., Laaksonen, L., Kraulis, P., Jones, T.A.,Knowles, J.K.C. and Teerei, T.T. (1992) Investigation of the function of mutated CBDsof Trichoderma reesei CBH I. Proteins 14, 475–482.

Rivard, C.J., Nieves, R.A., Nagle, N.J. and Himmel, M.E. (1994) Evaluation of discretecellulase enzyme activities from anaerobic digester sludge fed a municipal solid wastefeedstock. Applied Biochemistry and Biotechnology 45/46, 453–462.

Rouvinen, J., Bergfors, T., Teeri, T.T., Knowles, J.K.C. and Jones, T.A. (1990) Three-dimensional structure of CBH II from Trichoderma reesei. Science 249, 380–386.

Ryan, S.E., Jong, E. de, Gubitz, G.M., Tuohy, M. and Saddler, J.N. (1998) The potentialof hydrolytic enzymes to modify douglas-fir derived bleached mechanical pulps.In: Claeyssens, M., Nerinckx, W. and Piens, K. (eds) Carbohydrases from Trichodermareesei and Other Microorganisms – Structure, Biochemistry, Genetics and Applications.Proceedings of Tricel ’97 Meeting, The Royal Society of Chemistry, Cambridge, UK,pp. 214–226.

Ryu, D.D. and Mandels, M. (1980) Cellulases: biosynthesis and applications. Enzyme andMicrobial Technology 2, 91–101.

Sadana, J.C., Patil, R.V. and Shewale, J.G. (1988) β-D-Glucosidases from Streptomyces rolfsii.In: Wood, W.A. and Kellogg, S.T. (eds) Methods in Enzymology, Vol. 160. AcademicPress, London, pp. 424–431.

Sakka, K., Furuse, S. and Shimada, K. (1989) Cloning and expression in Escherichia coliof thermophilic Clostridium sp. F1 genes related to cellulose hydrolysis. AgriculturalBiological Chemistry 53, 905–911.

Sakon, J., Irwin, D., Wilson, D.B. and Karplus, P.A. (1997) Structure and mechanismof endo/exocellulase E4 from Thermomonospora fusca. Nature Structural Biology 4,810–818.

Saloheimo, A., Ilmen, M., Aro, N., Margolles-Clark, E. and Penttila, M. (1998) Regulatorymechanisms involved in expression of extracellular hydrolytic enzymes from Trichodermareesei. In: Claeyssens, M., Nerinckx, W. and Piens, K. (eds) Carbohydrases fromTrichoderma reesei and Other Microorganisms – Structure, Biochemistry, Genetics andApplications. Proceedings of Tricel ’97 Meeting, The Royal Society of Chemistry,Cambridge, UK, pp. 267–273.

Scott, T.C., Losgrove, J.M., Coon, C.L., Kenney, J.A. and Scott, C.D. (1995) Use of cellulasefrom Pseudomonas fluorescens for the hydrolysis of waste paper in an attrition bioreactor.Applied Biochemistry and Biotechnology 51/52, 537–543.

Selmer-Olsen, I., Henderson, A.R., Robertson, S. and McGinn, R. (1993) Cell wall degradingenzymes for silage. 1. The fermentation of enzyme-treated rye grass in laboratory silos.Grass and Forage Science 48, 45–54.

Sengupta, S., Khowala, S. and Goswami, P.K. (1987) Assay of endo-β-D-xylanase activity witha soluble O-(carboxymethyl) derivative of larch-wood D-xylan. Carbohydrate Research168, 156–161.





Shen, H., Gilkes, N.R., Kilburn, D.G., Miller, R.C. and Warren, R.A.J. (1995)Cellobiohydrolase B, a second exo-CBH from the cellulolytic bacterium Cellulomonasfimi. Biochemical Journal 311, 67–74.

Simpson, H.D., Haufler, U.R. and Daniel, R.M. (1991) An extremely thermostable xylanasefrom the thermophilic eubacterium Thermotoga. Biochemical Journal 277, 413–417.

Sinitsyn, A.P., Gusakov, A.V. and Yu Vlasenko, E. (1990) Effect of structural andphysico-chemical features of cellulosic substrates on the efficiency of enzymatichydrolysis. Applied Biochemistry and Biotechnology 30, 43–59.

Sinnott, M.L. (1990) Catalytic mechanisms of enzymic glycosyl transfer. Chemical Review 90,1171–1202.

Somogyi, M. (1952) Notes on sugar determination. Journal of Biological Chemistry 195,19–23.

Spezio, M., Wilson, D.B. and Karplus, P.A. (1993) Crystal structure of the catalytic domain ofa thermophilic endocellulase. Biochemistry 32, 9906–9916.

Sprey, B. and Lambert, C. (1983) Titration curves of cellulases from Trichoderma reesei:demonstration of cellulase–xylanase-β-glucosidase containing complex. FEMS Micro-biology Letters 18, 217–222.

Srisodsuk, M., Lehtio, J., Linder, M., Margolles-Clark, E., Reinikainen, T. and Teeri, T.T.(1997) Trichoderma reesei CBH I with an endoglucanase CBD: action onmicrocrystalline cellulose. Journal of Biotechnology 57, 49–58.

Stephen, A.M. (1983) Other plant polysaccharides. In: Aspinall, G.O. (ed.) The Polysaccha-rides, Vol. 2. Academic Press, London, pp. 98–194.

Sternberg, D. and Mandels, G.R. (1979) Induction of cellulolytic enzymes in Trichodermareesei by sophorose. Journal of Bacteriology 139, 761–769.

Sulzenbacher, G., Driguez, H., Henrissat, B., Schulein, M. and Davies, G.J. (1996) Structureof the Fusarium oxysporum endoglucanase I with a non-hydrolysable substrate analogue:substrate distortion gives rise to a pseudo-axial orientation for the leaving group.Biochemistry 35, 15280–15287.

Sulzenbacher, G., Shareck, F., Morosoli, R., Dupont, C. and Davies, G.J. (1997) TheStreptomyces lividans family 12 endoglucanase: construction of the catalytic core,expression, and X-ray structure at 1.75 Å resolution. Biochemistry 36, 16032–16039.

Teeri, T.T. (1997) Crystalline cellulose degradation: new insight into the function of CBHs.Trends in Biotechnology 15, 160–167.

Tenkanen, M., Schuseil, J., Puls, J. and Poutanen, K. (1991) Production, purification andcharacterisation of an esterase liberating phenolic acids from lignocellulosics. Journal ofBiotechnology 18, 69–94.

Timell, T.E. (1967) Recent progress in the chemistry of wood and hemicelluloses. WoodScience Technology 1, 45–70.

Tomme, P., van Tilbeurgh, H., Pettersson, G., van Damme, J., Vandekerckhove, J., Knowles,J., Teeri, T.T. and Claeyssens, M. (1988) Studies of the cellulolytic system ofTrichoderma reesei QM 9414. Analysis of domain function in two CBHs by limitedproteolysis. European Journal of Biochemistry 170, 575–581.

Tomme, P., Warren, R.A.J., Miller, R.C., Kilburn, D.G. and Gilkes, N.R. (1995a)Cellulose-binding domains: classification and properties. In: Saddler, J.N. and Penner,M.H. (eds) Enzymatic Degradation of Insoluble Carbohydrates. ACS Symposium Series168, American Chemical Society, Washington, DC, pp. 143–163.

Tomme, P., Warren, R.A.J. and Gilkes, N.R. (1995b) Cellulose hydrolysis by bacteria andfungi. Advances in Microbial Physiology 37, 1–81.





Tomme, P., Driver, D., Amandoron, E., Miller, R.C., Warren, R.A.J. and Kilburn, D.G.(1995c) Comparison of a fungal (family I) and bacterial (family II) CBD. Journal ofBacteriology 177, 4356–4363.

Tomme, P., Creagh, L.A., Kilburn, D.G. and Haynes, C.A. (1996) Interaction ofploysaccharides with the N-terminal cellulose-binding domains of Cellulomonas fimiCenC. 1. Binding specificity and calorimetric analysis. Biochemistry 35, 13885–13894.

Tormo, J., Lamed, R., Chirino, A.J., Morag, E., Bayer, E.A., Shoham, Y. and Steitz, T.A.(1996) Crystal structure of a bacterial family III CBD: a general mechanism forattachment to cellulose. EMBO Journal 15, 5739–5751.

Torronen, A. and Rouvinen, J. (1997) Structural and functional properties of low molecularweight endo-1,4-β-xylanases. Journal of Biotechnology 57, 137–149.

Torronen, A., Harkki, A. and Rouvinen, J. (1994) Three-dimensional structure ofendo-1,4-β-xylanase II from Trichoderma reesei: two conformational states in the activesite. EMBO Journal 13, 2493–2501.

Tuohy, M.G., Puls, J., Claeyssens, M., Vrasanska, M. and Coughlan, M.P. (1992) Thexylan-degrading enzyme system of Talaromyces emersonii: novel enzymes with activityagainst aryl β-D-xylosides and unsubstituted xylans. Biochemical Journal 290, 515–523.

van Tilbeurgh, H., Loontiens, F.G., De Bruyne, C.K. and Claeyssens, M. (1988) Fluorogenicand chromogenic glycosides as substrates and ligands of carbohydrases. In: Wood, W.A.and Kellogg, S.T. (eds) Methods in Enzymology, Vol. 160. Academic Press, London,pp. 45–59.

Viikari, L., Tenkanen, M., Buchert, J., Ratto, M., Bailey, M., Siika-Apo, M. and Linko, M.(1993) Hemicellulases for industrial applications. In: Saddler, J.N. (ed.) Bioconversionof Forest and Agricultural Plant Residues. CAB International, Wallingford, UK,pp. 131–182.

Vrsanska, M. and Biely, P. (1992) The cellobiohydrolase I from Trichoderma reesei QM 9414:action on cello-oligosaccharides. Carbohydrate Research 227, 19–27.

Warren, R.A.J. (1998) Structure and function in β-1,4-glycanases. In: Claeyssens, M.,Nerinckx, W. and Piens, K. (eds) Carbohydrases from Trichoderma reesei and OtherMicroorganisms – Structure, Biochemistry, Genetics and applications. Proceedings of Tricel’97 meeting, The Royal Society of Chemistry, Cambridge, UK, pp. 115–123.

Warkarchuck, W.W., Campbell, R.L., Sung, W.L., Davodi, J. and Yaguchi, M. (1994)Mutational and crystallographic analysis of the active site residues of the Bacillus circulansxylanase. Protein Science 3, 467–475.

Whistler, R.L. and Richards, E.L. (1970) Hemicelluloses. In: Pigman, W. and Horton, D.(eds) The Carbohydrates – Chemistry and Biochemistry, 2nd edn, Vol. 2A. Academic Press,New York, pp. 447–469.

White, A., Withers, S.G., Gilkes, N.R. and Rose, D.R. (1994) Crystal structure of thecatalytic domain of the β-1,4-glycanase cex from Cellulomonas fimi. Biochemistry 33,12546–12552.

Winterhalter, C. and Liebl, W. (1995) Two extremely thermostable xylanases of thehyperthermophilic bacterium Thermotoga maritima MSB8. Applied and EnvironmentalMicrobiology 61, 1810–1815.

Withers, S.G., Dombroski, D., Berven, L.A., Kilburn, D.G., Miller, R.C., Warren, R.A.J andGilkes, N.R. (1988) Direct 1H NMR determination of the stereochemical course ofhydrolysis catalysed by glucanase components of the cellulase complex. Biochemical andBiophysical Research Communications 139, 487–494.

Wong, K.K.Y and Saddler, J.N. (1992) Trichoderma xylanases, their properties andapplications. In: Visser, J., Beldman, G., Kusters-van Someren, M.A. and Voragen,





A.G.J. (eds) Xylans and Xylanases. Progress in Biotechnology, Vol. 7. Elsevier,Amsterdam, The Netherlands, pp. 171–186.

Wong, K.K.Y. and Saddler, J.N. (1993) Applications of hemicellulases in the food, feed, andpulp and paper industries. In: Coughlan, M.P. and Hazlewood, G.P. (eds) Hemicelluloseand Hemicellulases. Portland Press, London, pp. 127–143.

Wong, K.Y., Tan, L.U.L. and Saddler, J.N. (1986) Functional interactions among threexylanases from Trichoderma harzianum. Enzyme and Microbial Technology 8, 617–622.

Wong, K.Y., Tan, L.U.L. and Saddler, J.N. (1988) Multiplicity of β-1,4-xylanase inmicroorganisms: functions and applications. Microbiological Reviews 52, 305–317.

Wood, T.M. (1969) The cellulase of Fusarium solani. Resolution of the enzyme complex.Biochemical Journal 115, 457–464.

Wood, T.M. (1985) Properties of cellulolytic enzyme systems. Biochemical Society Transactions13, 407–410.

Wood, T.M. (1988) Preparation of crystalline, amorphous, and dyed cellulase substrates.In: Wood, W.A. and Kellogg, S.T. (eds) Methods in Enzymology, Vol. 160. AcademicPress, London, pp. 19–25.

Wood, T.M. (1992a) Microbial enzymes involved in the degradation of the cellulosecomponent of plant cell walls. In: Rowett Research Institute Annual Report, pp. 10–24.

Wood, T.M. (1992b) Fungal cellulases. Biochemical Society Transactions 20, 46–53.Wood, T.M. and Bhat, K.M. (1988) Methods for measuring cellulase activities. In: Wood,

W.A. and Kellogg, S.T. (eds) Methods in Enzymology, Vol. 160. Academic Press, London,pp. 87–112.

Wood, T.M. and McCrae, S.I. (1972) The purification and properties of the C1 component ofTrichoderma koningii cellulases. Biochemical Journal 128, 1183–1192.

Wood, T.M. and McCrae, S.I. (1978) The cellulase of Trichoderma koningii. BiochemicalJournal 171, 61–72.

Wood, T.M. and McCrae, S.I. (1982) Purification and some properties of a (1,4)-β-D-glucanglucohydrolase associated with the cellulase from the fungus Penicillium funiculosum.Carbohydrate Research 110, 291–303.

Wood, T.M. and McCrae, S.I. (1986) The cellulase of Penicillium pinophilum: synergismbetween enzyme components in solubilizing cellulose with special reference to theinvolvement of two immunologically distinct CBHs. Carbohydrate Research 234, 93–99.

Wood, T.M., McCrae, S.I., Wilson, C., Bhat, K.M. and Gow, L. (1988) Aerobic andanaerobic fungal cellulases with special reference to their mode of attack on crystallinecellulose. In: Aubert, J.-P., Beguin, P. and Millet, J. (eds) Biochemistry and Genetics ofCellulose Degradation. FEMS Symposium No. 43, Academic Press, London,pp. 31–52.

Wood, T.M., McCrae, S.I. and Bhat, K.M. (1989) The mechanism of fungal cellulase action.Synergism between enzyme components of Penicillium pinophilum cellulase insolubilizing hydrogen bond-ordered cellulose. Biochemical Journal 260, 37–43.

Wood, T.M., Wilson, C.A. and McCrae, S.I. (1994) Synergism between components of thecellulase system of the anaerobic rumen fungus Neocallimastix frontalis and those of theaerobic fungus Penicillium pinophilum and Trichoderma koningii in degrading crystallinecellulose. Applied Microbiology and Biotechnology 41, 257–261.

Woodward, J. and Wiseman, A. (1982) Fungal and other β-D-glucosidases – their propertiesand applications. Enzyme and Microbial Technology 4, 73–79.

Xu, G.Y., Ong, E., Gilkes, N., Kilburn, D., Muhandiram, D.R., Harris-Brandts, M., Carver,J., Kay, L. and Harvey, T. (1995) Solution structure of a CBD from Cellulomonas fimi bynuclear magnetic resonance spectroscopy. Biochemistry 34, 6993–7009.





Yoshioka, H., Nagato, N., Chavanich, S., Nilubol, N. and Hayashida, S. (1981) Purificationand properties of thermostable xylanase from Talaromyces byssochlamydoides. AgriculturalBiological Chemistry 45, 2425–2432.

Yu, E.K.C., Tan, L.U.L., Chan, M.K.-H., Deschatelets, L. and Saddler, J.N. (1987)Production of thermostable xylanase by a thermophilic fungus, Thermoascus aurantiacus.Enzyme and Microbial Technology 9, 16–24.





D.D. MaenzCharacteristics of Phytases3

Introduction

Substantial quantities of phytic acid occur in plant-based diets fed to productionanimals throughout the world. Phytic acid is poorly digested by monogastric animals.Disposing of manure containing excess phytate-phosphorus can damage freshwaterand other ecosystems. The poor digestibility of phytate-P increases the cost ofproduction since additional sources of available P are needed in ration formulation tomeet the nutritional requirements of the animal. Further, undigested phytate isknown to have a negative impact on mineral and protein digestibility. In recent yearsconsiderable research has been directed toward understanding the process of phytatedigestion and developing methods to improve phytate-P utilization by animals. Thedevelopment of enzyme technology based on supplementing diets with sources ofmicrobial phytase has proven to be a practical and effective method of improvingphytate digestibility in animal diets.

Several excellent reviews have appeared covering the environmental andnutritional consequences of phytic acid (Cheryan, 1980; Torre et al., 1991;Ravindran et al., 1995; Rickard and Thompson, 1997), and the application,structure and kinetic properties of phytase enzymes (Dvorakova, 1998; Liu et al.,1998). The purpose of this chapter will not be to revisit material that has beenextensively covered in other publications. Rather, the intentions are to provide ageneral overview of the subject area in combination with a specific focus on factorsthat affect the efficacy of phytate hydrolysis and methods to reduce further oreliminate indigestible phytate in plant-based diets fed to production animals.



Enzymatic Characteristics ofPhytases as they Relate to Their

Use in Animal FeedsD.D. MAENZ

Prairie Feed Resource Centre (Canada) Inc., Department of Animal andPoultry Science, University of Saskatchewan, Saskatoon,

Saskatchewan, Canada S7N 0W0

3



Phytic Acid in Animal Diets

Structure of phytic acid

Structurally, phytic acid consists of a fully phosphorylated myo-inositol ring (Fig.3.1) that exists in a chair conformation in dilute solution (Johnson and Tete, 1969).The molecule contains a total of 12 proton dissociation sites. Six of these sites arestrongly acidic, with pKa values approximating 1.5; three sites are weakly acidic, withpKa values of 5.7, 6.8 and 7.6; and the remaining three sites are very weakly acidic,with pKa values > 10 (Costello et al., 1976). The structure of the molecule is consis-tent with a high chelation capacity for multivalent cations. In solution, the stabilityof phytate–mineral complexes can be measured by the pH-drop method wherebymetals displace H+ for binding to phosphate groups on the phytate molecule(Maddaiah et al., 1964; Vohra and Kratzer, 1965). In summarizing the results ofseveral studies using the pH-drop method, Cheryan (1980) concluded that phyticacid readily forms complexes with multivalent cations, with Zn2+ forming the moststable complexes followed by Cu2+, Ni2+, Co2+, Mn2+, Ca2+ and Fe2+ in decreasingorder of stability. The phytate–mineral complex can exist as either soluble chelates oras insoluble complexes that precipitate out of solution, depending on the concentra-tions of phytic acid and mineral and the pH of the solution (Cheryan, 1980).

Some information is available on the structure of soluble mineral–phytatecomplexes. Champagne et al. (1990) and Champagne and Fisher (1990) studiedshifts in 31P NMR spectra of phytic acid in solution at pH 7 with a relativelylow molar ratio of minerals to phytate. These workers proposed that, under theconditions used in their experiment, a single Zn2+ ion binds to the P5 phosphategroups and thereby forms a bridge between two molecules of phytate. In comparison,Ca2+ and Cu2+ showed more of a tendency to associate with the P4 and P6 phosphategroups and thus was likely to form soluble complexes by electrostatic binding withina single phytate molecule. Mineral–phytate precipitants formed as the molar ratio ofmineral to phytate increased at neutral pH (Champagne et al., 1990).

62 D.D. Maenz


Fig. 3.1. Structure of the fully deprotonated form of phytic acid (myo-inositolhexakis phosphate).



Insoluble phytate–mineral complexes tend to form at neutral and basic pHwhen the concentration of mineral exceeds the concentration of phytate. At mediumpH < 5, phytate does not precipitate out of solution in the presence of Ca2+

(Grynspan and Cheryan, 1983) or Mg2+ (Cheryan et al., 1983) even atmineral : phytate ratios of 12 : 1. These results imply that one or more of the weakacid phosphate groups on the molecule has a higher affinity for protons than Ca2+

or Mg2+. In both of these studies, as the pH of the medium was increased, themineral : phytate molar ratio required for phytate precipitation decreased. Underconditions of excess mineral at neutral pH, it was suggested that precipitates con-sisted of pentacalcium (Grynspan and Cheryan, 1983) or pentamagnesium (Cheryanet al., 1983) salts of phytate. At acidic pH, partial protonation of phytic acid willdecrease net association of minerals to the molecule and thus prevent the formationof insoluble precipitates. As described later, the structure of phytic acid as affected byminerals and pH has profound effects on the process of phytate hydrolysis.

Occurrence of phytic acid in plants

Phytic acid serves as the storage form of P in plant seeds. Ravindran et al. (1995)provided an extensive review of the occurrence of phytic acid in a wide variety ofplants and plant materials. Cereals (maize, barley, wheat, oats) and grain legumes(field peas, chickpeas) that are commonly used as feed ingredients all have similarphytate levels, approximating 0.25% of dry matter. In general, oilseed meals tendto have higher levels of phytate-P as indicated by the percentage of dry mattercomposition for soybean (0.39%), rapeseed meal (0.70%), cottonseed meal (0.84%)and sunflower meal (0.89%). On average, 70% of the total P in the feed ingredient isfound as phytate-P.

In plants, phytic acid complexes K+ and Mg2+, and to a lesser extent Ca2+, toform phytin. Phytin is deposited within the proteinaceous matrix of membrane-bound protein vacuoles or protein bodies. Under the electron microscope, phytin isoften visible within the protein bodies as insoluble globoid crystal structures. In ricegrains, globoid crystals are composed of 67% phytic acid, 19% K, 11% Mg and 1%Ca on a dry weight basis (Ogawa et al., 1975). The size of the globoid crystals isdependent upon the ratio of divalent cations/K in the protein bodies. Ratios thatfavour divalent cations result in large insoluble crystals (Lott et al., 1985). Legumessuch as soybeans and peas have higher levels of K and thus smaller globoid crystalstructures and a proportionally greater amount of phytin as predominantly K-boundsoluble forms of phytic acid (Lott and Ockenden, 1986). This soluble phytin tendsto exist as phytin–protein complexes that are evenly dispersed within the proteinbodies.

The location of phytin-containing protein bodies in the seed varies considerablybetween plants. In wheat and rice phytin is found primarily in the aleurone layer andthe bran, while in maize phytin is concentrated within the endosperm (O’Dell et al.,1972; de Boland et al., 1975). Phytin tends to be distributed evenly throughout the

Characteristics of Phytases 63




seed in oilseeds and legume grains (Erdman, 1979). The form and location of phytinwithin the seed are likely to be important variables that impact on the efficiency ofphytate hydrolysis. One could hypothesize that phytin as insoluble globoid crystals inthe fibrous layers of the seed coat would be relatively indigestible in comparison withphytin that exists predominantly as soluble forms in the germ portion of the seed.

Effects of phytic acid in animal diets

Environmental effectsIn the United States, agriculture is the leading source of contaminants that have anegative impact on water quality (Parry, 1998). An estimated 60% of river miles and50% of lakes that are classified as impaired are negatively affected by agriculture(Daniel et al., 1998). Animal wastes are the primary source of P pollution fromagricultural practices (Daniel et al., 1998). In areas of intensive livestock production,P added to land from animal waste can readily exceed the capacity of the crop toincorporate the mineral. This, in turn, can lead to excess P accumulation in thesoil and the potential for runoff into lakes and rivers. Problems associated withagriculture and P pollution are greatest in areas of intensive livestock productionwith close proximity to freshwater systems. In The Netherlands there is a N, P and Kimbalance and livestock production contributes over 80% of the excess mineralsaccumulating in the environment (de Boer et al., 1997). The manure problem in TheNetherlands has led to legislation that limits mineral application to soils from animalwastes (de Boer et al., 1997).

Undigested phytate-P is the primary source of P in the manure. Any measuresthat decrease unavailable phytate-P in animal diets will be effective in decreasing theP load on the environment arising from animal production. Further, for mixedfarming operations, decreasing the P content of swine manure may benefit neteconomic returns in areas that are generally not classified as environmentallysensitive. Bosch et al. (1998) determined that reducing manure P decreases the landarea required for application based on crop P requirements which, in turn, decreasescrop production costs.

Effects on mineral utilizationUndigested phytic acid is known to decrease mineral bioavailability. The reader isreferred to the reviews of Cheryan (1980) and Torre et al. (1991) on the formation ofmineral–phytate complexes and the effects of phytate on mineral digestibility. Ingeneral an inverse relationship exists between the phytate content of the diet anddigestibility of multivalent cations. As previously described, phytate can complexwith multivalent cations, forming insoluble complexes at neutral pH. Thesecomplexes appear to be resistant to the digestive–absorptive process and the net resultis that a portion of the dietary mineral pool is unavailable to the animal.

In rats, Zn bioavailability is markedly affected by the phytate content of the diet(Forbes et al., 1984; Zhou et al., 1992; Saha et al., 1994). These results are consistentwith in vitro work indicating a high stability of Zn–phytate complexes (Cheryan,

64 D.D. Maenz




1980). Other minerals, such as Ca, when added at increasing levels in the presence ofa constant level of phytate, further reduced Zn bioavailability in rats (Ellis et al.,1982; Forbes et al., 1984). The mechanism of this effect is likely to involve increasedformation of insoluble phytate–mineral complexes containing both Ca and Zn thatare resistant to the digestive–absorptive process.

In humans, dephytinization of preparations of soybean protein isolates doubledthe mean fractional Fe incorporation into red cells (Davidsson et al., 1994) andresulted in 2.3-fold increase in Mn absorption (Davidsson et al., 1995). Fe solubilityunder simulated physiological conditions in vitro was increased from 3 to 53% withdephytinization of wheat bran and from 5 to 21% with dephytinization of rye flour(Sandberg and Svanberg, 1991).

Han et al. (1994) compared the effects of IP6 (phytic acid), IP5, IP4 and IP3 onFe and Zn uptake by the human adenocarcinoma cell line Caco-2 and found thatmineral solubility and uptake rates decreased with the degree of phosphorylation ofthe inositol ring. Lower forms of inositol phosphates that are formed during thehydrolysis of the parent phytate have fewer mineral-binding sites, a diminishedchelation capacity, and thus less of an effect on mineral bioavailability than is the casefor phytic acid (Tao et al., 1986).

Interestingly, certain compounds with chelation capacity have been shown toincrease mineral availability when included in plant-based diets fed to animals.Supplementation of Zn-deficient diets with various chelators such as EDTApromoted growth and prevented symptoms of Zn deficiency in chicks (Nielsen et al.,1966) and turkey poults (Vohra and Kratzer, 1964). Pectins characterized by ahigh degree of esterification and a low molecular weight are known to complex withminerals and have been shown to improve Fe solubility and absorption in rats(Kim and Atallah, 1993). Ascorbic acid and citric acid were shown to enhance Febioavailability in vegetarian diets (Anand and Seshadri, 1995) and ascorbic acidincreased Fe bioavailability in soy-based infant formulas (Davidsson et al., 1994). Amechanism of competitive chelation could account for the effect of chelators onimproving mineral bioavailability. In this model the chelator binds minerals andthereby decreases the pool of minerals forming insoluble phytate complexes. Themineral–chelator complex exists in a soluble form and thus could be absorbed intactor could release minerals to binding sites on the brush border membrane of theintestinal epithelium. The concept of competitive chelation as a mechanism forimproved phytate hydrolysis will be developed in a later section of this review.

Effects on protein digestibilityThe charged phosphate groups on phytic acid can form electrostatic associations withthe terminal amino group on proteins or with the free amino group on lysine andarginine residues within protein molecules (Cheryan, 1980). In addition, phytate–mineral–protein complexes can form with multivalent cations acting as a bridgebetween phosphate groups on the phytate molecule and the terminal carboxyl groupof proteins or free carboxyl group on aspartate and glutamate residues within proteinmolecules (Cheryan, 1980). In theory, protein in the phytate-bound form may beless susceptible to protease activity during intestinal passage. In addition, phytate





binding to proteins and minerals in the digesta has the potential to impair the activityof digestive enzymes. Phytate has been shown to inhibit trypsin activity (Singh andKrikorian, 1982; Caldwell, 1992) and in vitro starch digestion with human saliva(Yoon et al., 1983). The mechanism may involve mineral chelation and thus removalof cofactors required for optimum enzyme activity and/or the formation of lessreactive phytate–enzyme complexes.

With a few notable exceptions, a review of the literature generally supports theconcept that phytic acid does impair in vivo protein digestibility. Thompson andSerraino (1986) reported that phytic acid did not affect rapeseed protein digestibilityand amino acid absorption. Chitra et al. (1995) found a negative correlation betweenphytic acid content and in vitro protein digestibility in genotypes of mung bean,urd bean and soybean. Phytic acid–protein interactions have been reported todecrease the availability of protein in legumes (Camovale et al., 1988). Completedephytinization of a diet based on maize and soybean meal increased dialysableprotein by 12–29% (Zyla et al., 1995). Dephytinization of soy-protein concentrateimproved protein digestibility in fish (Storebakken et al., 1998). One couldhypothesize that the effect of phytate on protein and amino acid digestibility mightvary considerably between diets based on plant ingredients and the mineral status ofthe diet. Phytate that is largely in the form of insoluble mineral-bound complexesmay be less active as an inhibitor of protein digestion. Clearly, interactions betweenphytate, minerals and proteins and effects of these interactions on proteindigestibility are a complex issue that deserves further research.

Digestion and Utilization of Phytate

In diet formulation, plant P is often classified as available and unavailable and a figureof 30% is routinely assigned as an estimate of available plant P. This exercise is usefulin that it permits a simple calculation of supplemental P that must be added to thediet to meet the nutritional requirements of the animal. However, an assumptionthat 30% of total plant P is available and thus 70% is unavailable is simplistic and, inmany circumstances, may be grossly inaccurate. As described in detail by Ravindranet al. (1995), a review of the literature shows considerable variation in the digestibilityof dietary P of plant origin. As an example, Nelson (1976) reported 0% phytate-Pdigestibility in chicks fed maize-based diets and 8% phytate-P digestibility withwheat-based diets. In comparison, other studies have shown that chicks can retain upto 60% of dietary phytate-P (Temperton and Cassidy, 1964a,b). Ballam et al. (1984)found that phytate hydrolysis ranged from 3 to 42% in the chick and was dependentupon the level of dietary Ca. Phytate-P retention values of 37–56% were reported forchicks fed suboptimal levels of non-phytate-P (Edwards, 1983).

Interestingly, phytate-P when added to the diets in acidic form or as a sodiumsalt is readily available to chicks and will support growth and bone mineralization tothe same extent as P originating from dicalcium phosphate (Harms et al., 1962;Waldroup et al., 1964a). P originating from supplemental calcium phytate couldsupport growth and bone mineralization at levels up to 0.2% dietary P and at a

66 D.D. Maenz




Ca : P ratio of 0.8 : 1 (Waldroup et al., 1964b). In this study, increasing the dietaryCa : P ratio to 1.4 : 1 or 2 : 1 decreased the apparent utilization of P from calciumphytate (Waldroup et al., 1964). These studies provide strong support for theconcept that the form and location of phytate within the plant material is animportant factor defining phytate-P utilization.

Phytate digestibility is ultimately dependent upon the efficacy of hydrolysis ofphytate prior to feeding and during digestive passage. Hydrolysis is dependent uponphytase activity originating from plant, animal or microbial sources. The efficiency ofthis hydrolytic process is dependent upon the total phytase activity in combinationwith other factors such as the form and location of phytin within the plant ingredientand the physical conditions of the reaction.

Sources of phytase activity

Dvorakova (1998) and Liu et al. (1998) provided extensive reviews of the structureand kinetic characteristics of phytase from microbial, plant and animal sources.There will be no attempt to elaborate further or to assemble an extensive review ofthis area here. The discussion in this chapter will be restricted to key points on what isknown of phytase structure and kinetic function.

Phytases can be broadly categorized into 6-phytase and 3-phytase classes. Thisdesignation refers to the site of initial hydrolysis on the phytate molecule. The6-phytases are commonly found in plants, while 3-phytases are produced by fungi(Dvorakova, 1998). The full sequence of hydrolysis of phytic acid down tomyo-inositol has yet to be determined. However, it would appear that no singleenzyme is capable of fully dephosphorylating phytic acid and thus a combinationof phytase and non-specific phosphatase activities are involved in the process. Noinformation is available as to the mechanism of phytate hydrolysis by the smallintestinal phytase activity in animals.

Animal phytaseOften it is assumed, and frequently stated, that animals do not possess the capacity tohydrolyse phytic acid. However, a specific phytase activity has been demonstrated inthe brush border membrane of the chick small intestine (Maenz and Classen, 1998).Earlier studies demonstrated phytate hydrolysis with small intestinal mucosalpreparations from chick (Davies et al., 1970; Bitar and Reinhold, 1972; Davies andMotzok, 1972), human (Bitar and Reinhold, 1972; Iqbal et al., 1994), calf (Bitar andReinhold, 1972) and rat (Bitar and Reinhold, 1972; Yang et al., 1991; Iqbal et al.,1994). The contribution of mucosal phytase to phytate-P utilization by animals isnot known, but animals are known to utilize a portion of the total phytate-P with nosupplemental phytase in the diet. Further, a substantial body of literature indicatesthat dietary supplementation with 1,25-dihydroxycholecalciferol improves phytate-Pdigestibility in poultry (Ravindran et al., 1995). Rats (Yang et al., 1991) and chickens(Davies et al., 1970) fed phosphorus-deficient diets show an increase in intestinalphytase activity and in the case of the rat this correlated with an increase in a 90 kDa





protein isolated from the intestinal mucosa (Yang et al., 1991). Taken together, thesestudies indicate that an intestinal phytase probably does contribute to the utilizationof phytate-P and the activity of the enzyme may well be regulated by the mineral andvitamin status of the animal.

Plant phytaseAlmost all plants contain some phytase activity, though the level of phytase and thesignificance of the enzyme in hydrolysing phytate within the seed varies considerablybetween plants. Eeckhout and De Paepe (1994) evaluated phytase levels in 51feedstuffs used in Belgian feed mills and concluded that substantial phytaseactivity occurs in cereal grains such as rye, triticale, wheat and barley while all otherfeedstuffs, including soybean meal, contained marginal levels of phytase. Phosphorusavailability in wheat fed to poultry ranged from 45 to 70% (Barrier-Guillot et al.,1996b). Barrier-Guillot et al. (1996a) measured phytase activity in 56 wheat samplesgrown in France in 1992 and found a substantial variation that ranged from 206 to775 mU g−1. In a follow-up study (Barrier-Guillot et al., 1996b), the same groupfound a linear correlation between wheat phytase activity and P retention in broilers.Pointillart (1994) pooled results on plant P availability in pigs fed cereal-based diets.Availability ranged from less than 20% for diets with little phytase up to 60% fordiets rich in plant phytase activity and a clear linear relationship was found betweenthe plant phytase activity in the diet and P digestibility. Kemme et al. (1998)measured gastric degradation of phytic acid in pigs as 3% when fed a maize-baseddiet (91 phytase units kg−1), 31% when fed a maize–wheat-based diet (342 phytaseunits kg−1) and 47% when fed a wheat–barley-based diet (1005 phytase units kg−1).Clearly, the high phytase activity found in wheat and barley is significant and doescontribute to phytate digestibility. These studies argue against a simple value foravailable P based solely on the total and phytate-P content of the diet. As such, thepotential exists to refine diet formulation to meet the P requirements of the animalbased on a realistic estimate of available P in a given plant-based diet.

Microbial phytaseEnzymes with hydrolytic activity directed toward phytic acid are produced by avariety of microorganisms. Dvorakova (1998) listed 29 fungal, bacterial and yeastspecies known to produce a phytase activity. Of the 29 species listed, 21 producedan extracellular phytase. Strains of the filamentous fungi Aspergillus niger producea high-activity extracellular phytase (Volfova et al., 1994). Currently, commercialphytase products are based on the phytase encoding gene originating from A. niger.

The enzymes produced by A. niger var. ficuum are by far the most extensivelystudied phytase. Originally a single activity was described with an optimum pHof 5.0 plus a secondary optimum pH at 2.5 (Van Hartingsveldt et al., 1993).Subsequently, a second distinct phytase from A. niger var. ficuum, with a pHoptimum of 2.5 and no activity at pH 5.0, was described (Ullah and Phillippy,1994). The pH 5 optimum enzyme has been termed PhyA and the pH 2.5 optimumenzyme has been termed PhyB. The amino acid composition of the unglycosylatedA. niger var. ficuum PhyA indicates a molecular weight of 62 kDa (Ullah, 1988).

68 D.D. Maenz




The protein is 27.3% glycosylated with a molecular weight of 84 kDa in the nativeform (Ullah and Gibson, 1987). PhyA has a high affinity for phytic acid with a Km

that approximates 40 µM (Ullah and Gibson, 1987). Variations in temperature andpH were found to have little effect on the affinity of PhyA for phytic acid (Ullah,1988). PhyA has a very low turnover number of 216 mol P formed per mol phytasesec−1 at optimal conditions of pH 5.0 and 58°C (Ullah and Gibson, 1987). Acomparison of amino acid sequences for the protein products of expression of clonedphyA (Van Hartingsveldt et al., 1993) and phyB (Ehrlich et al., 1993) genes revealed a23.5% homology with four proposed catalytic sites on phyA and a single proposedcatalytic site on phyB. These differences are likely to account for the multiple pHoptimum reported for phyA and the single pH optimum of phyB.

As reviewed by Dvorakova (1998), the industrial importance of phytase fromA. niger has fostered intensive investigation and numerous patent applications todevelop cost-effective microbial expression systems for production of the enzyme.With simple gene amplification, overexpression is repressed by inorganic P (VanHartingsveldt et al., 1993) and significant overexpression requires strong P-resistantpromoters (Piddington et al., 1993). Effective expression systems for phytasecDNA originating from A. niger var. ficuum are currently described in a variety ofmicroorganisms (Brocades, 1991, 1993; Berka et al., 1995a,b). Efficient hydrolysis ofphytic acid down to myo-inositol and P requires phytase plus non-specificphosphatase activities. Expression systems for DNA sequences encoding for bothphytase and acid phosphatase have been described (Alko, 1994; Panlabs, Alko, 1994).

Extracellular phytase enzymes from microbial sources are moderately heatstable. A step-wise increase in the temperature of the reaction medium from roomtemperature to 58°C resulted in a corresponding increase in the rate of phytatehydrolysis by A. ficuum phytase (Ullah and Gibson, 1987). Further increases inmedium temperature resulted in a dramatic drop in activity, with no detectablephytate hydrolysis at 68°C (Ullah and Gibson, 1987). The temperature sensitivity ofthe enzyme must be considered in any processing situation that involves heat afterenzyme application to the feed ingredient.

Thermostable phytase activities have been described from Bacillus sp. DS11(Kim et al., 1998) and Aspergillus fumigatus (Pasamontes et al., 1997). TheA. fumigatus phytase has a broad pH activity range and can withstand extremetemperatures of 100°C for 20 min or 90°C for 120 min (Pasamontes et al., 1997).The gene for the enzyme has been cloned and overexpressed in A. niger. Phytase fromA. fumigatus has commercial potential in that it can withstand conditions of feedpelleting without significant loss of activity.

Factors affecting phytate hydrolysis during intestinal passage

MineralsSeveral studies have shown that phytate-P digestibility varies with the mineralcontent of the diet. Increasing the level of dietary calcium results in a decreasedphytate-P digestibility in rats (Nelson and Kirby, 1979), poultry (Scheideler and Sell,





1987) and pigs (Sandberg et al., 1993). Presumably, increasing the concentration ofa multivalent cation such as calcium will increase the formation of insolublemineral-bound phytin crystals, which may be resistant to hydrolysis by phytaseactivities. Mohammed et al. (1991) found that decreasing the calcium level in dietslow in P from 1% down to 0.5% resulted in a 15% increase in phytate-P digestibilityin chicks. Ballam et al. (1984) found similar results in that phytate-P digestibility inchicks was increased when dietary calcium levels were reduced from 1.0% to 0.85%of the diet.

Mineral binding may well have negative effects on phytate hydrolysis byendogenous phytase of animal or plant origin in the diet. In general, Ca : phytateratios of greater than 2 : 1 are thought to reduce intrinsic phytate digestibility (Wise,1983). In pigs, supplementing a diet based on barley, rapeseed meal and peas with1.25% calcium carbonate decreased phytate digestibility (Skoglund et al., 1997). In aseparate study with pigs, total gastrointestinal hydrolysis of phytate decreased withincreasing levels of dietary calcium carbonate (Sandberg et al., 1993).

In vitro studies have demonstrated that minerals impair the efficiency of phytatehydrolysis by phytase of microbial origin. At neutral pH, Maenz et al. (1999) rankedZn2+ >> Fe2+ > Mn2+ > Fe3+ > Ca2+ > Mg2+ in terms of potency as inhibitors ofphytate hydrolysis by microbial phytase. Decreasing the pH resulted in a substantialdrop in the inhibitory effects of minerals. These results are consistent with the modelof pH-dependent mineral–phytate interaction whereby protonation of weakacid phosphate groups displaces minerals and thereby converts phytate fromphytase-resistant mineral-bound forms to phytase-susceptible forms.

Other in vitro studies have demonstrated that minerals have varying effects onthe rate of phytic acid hydrolysis by animal phytase. Maenz and Classen (1998)found that inclusion of 25 mM MgCl2 doubled the rate of phytate hydrolysis bychick small intestinal brush border phytase. Rat mucosal phytase activity is increased40% by low concentrations of Zn (Bitar and Reinhold, 1972). At Zn concentrationsgreater than 1 mM, hydrolysis of phytic acid by chick small intestinal brush borderphytase was markedly decreased (Maenz and Classen, 1998). Animal brush borderphytase is likely to have a requirement for minerals as cofactors for optimal activity.The activation of activity may be confounded by mineral binding to phytate and adecrease in the susceptibility of the substrate to hydrolysis.

pHVery little information exists describing phytate hydrolysis during intestinal passage.In comparing pH conditions down the length of the gut one could speculate that themajority of phytate hydrolysis by dietary plant or microbial phytase will occur duringgastric retention. The low pH in the glandular stomach of monogastric animals willfavour protonation of the weak-acids phytate groups. As such, after consumption of ameal, one would anticipate substantial formation of partially protonated forms ofphytate that are susceptible to hydrolysis during residence in the stomach or crop.The optimum pH value for phytase activity approximates 5, which is below theeffective pH for the formation of phytase-resistant mineral-bound complexes. In theupper regions of the small intestine the pH of the digesta will increase and thus

70 D.D. Maenz




favour the reformation of mineral-bound phytase-resistant forms of phytate. In vitro,0.5 M Ca was ineffective as an inhibitor of microbial phytase at pH 5, while 0.005 MCa caused complete inhibition when the pH value of the medium was increased to7.5 (Maenz et al., 1999). The extent of hydrolysis during gastric residence may wellbe limited by the amount of enzyme and the physical conditions of retention time,pH, mineral concentration, temperature, moisture, mixing, etc., that influence reac-tion rates. Kemme et al. (1998) determined average gastric phytate-P degradability inpigs as 47% for wheat–barley-based diets (1005 plant phytase units kg−1), 3% formaize-based diets (91 plant phytase units kg−1) and 47% for microbial phytase-supplemented maize-based diets (1565 microbial phytase units kg−1). The authorsevaluated the effect of diurnal variations in gastric retention time and pH of thestomach. At optimum conditions of prolonged retention time, phytate degradationapproached 90% in the stomach of pigs fed diets with high levels of plant phytase.They concluded that the efficacy of phytase degradation by plant and microbialsources of phytase is to a large extent determined by conditions of pH and retentiontime in the stomach.

A model of phytate hydrolysis by microbial and plant phytase occurringprimarily in the stomach is largely hypothetical. Research quantifying phytatehydrolysis down the full length of the gut with various dietary constituents and levelsof dietary phytase is needed for a better understanding of the physiology of lumenalphytate hydrolysis. In the caecum, bacterial populations may provide an additionalsource of phytase activity, but the significance of gut bacteria in the process ofphytate hydrolysis is unknown.

The process of phytate hydrolysis by mucosal phytase found on the brush bordermembrane of the small intestine of the animal is distinct from lumenal hydrolysis byplant and microbial phytases. The mucosal enzyme is fixed on the membrane andthe microenvironment in the unstirred water layer immediately adjacent to thebrush border is maintained at a moderately acidic pH of 6 (Lucas, 1983) down thelength of the gut. The pH of the unstirred water layer approximates the pH optimumof intestinal phytase activity and will be less than optimal for the formation ofphytase-resistant mineral complexes. As such, brush border phytate hydrolysis willdepend upon a combination of the degree of enzyme expression and substrate accessto the binding site on the enzyme within the unstirred water layer. In broiler chicksand laying hens the specific and total brush border phytase activity was highest in theduodenum and decreased progressively down the length of the small intestine(Maenz and Classen, 1998). The contribution of brush border phytase to the processof phytate hydrolysis may well be limited by poor mixing of insoluble mineral–phytate complexes relative to soluble nutrients in the digesta.

Temperature, duration, moisture and mixingThe rate of any enzymatic reaction is affected by the physical conditions of the reac-tion mixture. In vitro, the activity of microbial phytase near the body temperature ofthe animal is less than half of that obtained under optimal temperatures conditions of55°C (Ullah and Gibson, 1987). Insoluble mineral-bound phytin associated with thefibrous portions of the seed is likely to require prolonged extensive mixing in a slurry





with excess water to optimize enzyme access to the substrate. Physical conditionsof the digesta, especially in the small and large intestine, may well limit the efficacy oflumenal phytate hydrolysis.

Improving Phytate Digestibility

Decreasing plant phytate levels

Selecting for low-phytate plantsA viable low-phytate mutant of maize (lpa1-1) has been described by Ertl et al.(1998). When compared with conventional maize, this mutant is characterized by asimilar total P content, a 60% reduction in phytate, and a corresponding molarincrease in the inorganic P content of the seed. The inorganic P in the lpa1-1 mutantmaize supports growth and bone mineralization in broiler chicks. The lpa1-1 mutanthas the potential to provide genetic stock for production of low-phytate maize withacceptable agronomic traits. Given the success obtained with low-phytate maize,development of other low-phytate plants with higher levels of available P for animalfeeding can be envisaged.

Germination of plants to induce phytase activityThe function of plant phytase is to convert phosphorus from the storage form asphytate-P to readily available inorganic phosphate during development and growthof the plant. Germination is associated with increased phytase activity, decreasedphytate and increased inorganic phosphate in canola (Lu et al., 1987), rapeseed(Mahajan and Dua, 1997), triticale (Niziolek, 1995) and soybean (Bau et al., 1997).For some feed ingredients a short period of controlled germination prior to furtherprocessing has potential as a practical and effective method of improving plant Pavailability.

Maugenest et al. (1997) cloned the DNA encoding for the phytase expressedin germinating maize seedlings. Further isolation and characterization of associatedregulatory genes may eventually allow for controlled expression of phytase in plants.

Steeping to reduce phytate levelsSteeping (water soaking) a barley–rapeseed cake–pea-based diet for 9 h caused a 50%reduction in phytate levels and a corresponding increase in lower-level inositolphosphates and free inorganic phosphate (Skoglund et al., 1997). The authorsspeculate that prolonged soaking of the meal provided conditions whereby asubstantial portion of the phytate was susceptible to hydrolysis by plant phytase.

Enzyme pretreatment of plant mealsPretreatment of plant meals with microbial phytase under controlled conditions haspotential as an alternative method of achieving efficient and measurable hydrolysisof phytase. Thus far, published work on controlled dephytinization during meal

72 D.D. Maenz




pretreatment has been restricted to the laboratory investigation of solid-statefermentation of plant meals with phytase-producing organisms and enzyme cocktailapplication under controlled conditions in vitro.

Solid-state fermentation involves inoculation of the plant meal with anextracellular phytase-producing organism, followed by batch incubation undercontrolled conditions to maximize growth of the organism, production of phytaseand subsequent hydrolysis of phytic acid. Inoculation of canola meal with A. ficuumresulted in a massive increase in phytase, a linear decrease in phytic acid and completeelimination of phytate after 144 h of incubation (Nair et al., 1991). The sameresearch group further characterized the effects of phosphate, glucose and surfactantson the efficacy of solid-state fermentation of canola meal with Aspergillus carbonarius(Al-Ashed and Duvnjak, 1994a,b, 1995a,b) and A. ficuum (Ebune et al., 1995a,b).From this work, predictive models that relate biomass, enzyme production andreduction of phytate content to reaction conditions have been developed.Refinement of fermentation conditions has increased the efficiency of enzymeproduction such that under some conditions complete dephytinization of canolameal occurs within 48 h (Al-Ashed and Duvnjak, 1994a, 1995a; Ebune et al.,1995a,b). The economics and practicality of solid-state fermentation as a method forthe commercial production of dephytinized plant meals has yet to be determined.

Batch dephytinization of plant meals and mixed feed using phytase-containingenzyme cocktails has the advantage over in-feed enzyme application in that thephysical conditions of the reaction can be set to optimize the process of phytatehydrolysis. Zhu et al. (1990) achieved 81% phytate hydrolysis of a maize–soybeanmeal–wheat bran mixture using a two-step procedure that involved a 2 h initialconditioning of the maize–soybean meal with citric acid at pH 3.1 and a 2 htreatment with wheat bran at pH 5.1. Presumably citric acid acted as a competitivechelator to increase phytate susceptibility to hydrolysis by the wheat phytase duringthe second stage. Zyla et al. (1995a,b) developed a cocktail of commercial fungalphytase, acid phosphatase, citric acid and pectinase which, under simulated intestinalconditions of the turkey, resulted in complete dephytinization of a maize–soybeanmeal-based feed. Citric acid and cofactor enzymes probably functioned to improvesubstrate susceptibility to hydrolysis by the phytase and non-specific phosphatases inthe cocktail. Dephytinization resulted in a 12–29% increase in dialysable protein andutilization of plant P for growth and bone mineralization in turkey poults (Zyla et al.,1995a, 1996). Zyla and Koreleski (1993) found that rapeseed meal could bedephytinized during a 4 h incubation at 40°C with fungal phytase and acidphosphatase in a 3 : 1 water : meal slurry at acidic pH. The practicalities ofcommercial batch dephytinization of plant feed ingredients are not known at present.The cost of drying the meal after dephytinization of a high-moisture slurry presents aconsiderable obstacle to the application of the technology. However, if practical andcost-effective, batch dephytinization of plant meals has a distinct advantage overconventional in-feed enzyme application in that, by controlling reaction conditionsand measuring inorganic P, it should be possible to produce a consistent, highlydephytinized feed ingredient.





Transgenic plants expressing foreign phytase activityTransgenic tobacco (Verwoerd et al., 1995) and soybeans (Li et al., 1997) expressingphytase originating from A. niger have been developed. At present, transgenic plantsexpressing phytase are viewed not as a feed ingredient but as another potentialmethod for manufacturing the enzyme. However, the potential does exist fordeveloping transgenic plants with controlled expression of a foreign phytase thatcould be fed directly to animals. Cereal grains that do not undergo extensive heattreatment prior to diet formulation are the more likely targets for developing suchtransgenic plants.

Increasing phytate digestibility through in-feed supplementation

Vitamin DNumerous studies have indicated that vitamin D (Mohammed et al., 1991), theactive vitamin D metabolite 1,25-dihydroxycholecalciferol (Edwards, 1993; Mitchelland Edwards, 1996) and various active analogues of vitamin D (Biehl and Baker,1997; Biehl et al., 1998) are effective in increasing phytate-P availability when addedto diets fed to poultry. The magnitude of the effect of vitamin D on phytate-Pdigestibility varies considerably between studies and appears to depend upon themineral content of the diet. Mohammed et al. (1991) measured phytate-P availabilityin broiler chicks as 50.1% for control diets, 58.5% for normal Ca, high vitamin Ddiets, and 76.5% for low Ca, high vitamin D diets.

The mechanism of phytate-P digestibility enhanced by vitamin D is poorlyresolved. One of the physiological functions of the vitamin is to enhance Ca and Pabsorption. Increased Ca absorption could decrease Ca–phytate formation in thedigesta, which could increase the efficacy of phytate hydrolysis (Ravindran et al.,1995). Vitamin D has been shown to induce chick intestinal mucosal phytase activity(Davies et al., 1970). In the same study, gut phytase activity was threefold greater inbirds on P-deficient diets (Davies et al., 1970). As such, a part of the mechanism ofvitamin D action in promoting absorption of dietary P may well involve activationof intestinal mucosal phytase. Mitchell and Edwards (1996) found thatsupplementation with vitamin D3 and microbial phytase had an additive effect onincreasing phytate utilization in broiler chicks and they suggested a separatemechanism of action. With both microbial phytase and vitamin D3 in the diet, onecould hypothesize increased lumenal and brush border hydrolysis of the substrate.

Supplementation with microbial phytaseDietary supplementation with sources of microbial phytase is well established as aneffective and practical method of improving phytate digestibility in productionanimals. Ravindran et al. (1995) summarized a large body of literature in statingthat microbial phytase supplementation generally results in a 20–45% improvementin phytate-P utilization in poultry. A similar range for the effect of phytasesupplementation on P availability has been reported for plant-based diets fed topigs. Phytase supplementation of a barley–soybean meal diet resulted in a 40%

74 D.D. Maenz




improvement in P absorption in comparison with a control diet and a 23% improve-ment in P absorption in comparison with a diet supplemented with dicalciumphosphate (Nasi and Helander, 1994). Harper et al. (1997) found a linear improve-ment in P digestibility with increasing phytase in maize–soybean meal-based dietswith low dicalcium phosphate fed to growing pigs. In the same study, faecal P excre-tion was decreased by 21.5% with optimal phytase and low dicalcium phosphate(Harper et al., 1997). In weanling pigs, phytase supplementation is associated with a50% increase in P retention (Lei et al., 1993; Pallauf et al., 1994) and a 42% decreasein faecal P (Lei et al., 1993). Similar improvements in P digestibility were reportedfor sows (Kemme et al., 1997a) and growing–finishing pigs (Kemme et al., 1997b)fed phytase supplemented tapioca–soybean-based diets. Bruce and Sundstol (1995)found that ileal P digestibility increased from 19.4% to 26.8% with phytasesupplementation of an oat-based diet fed to growing pigs. In piglets, Mroz et al.(1994) reported that apparent ileal phytate digestibility increased from 31% to 79%with 800 units phytase kg−1 of a maize–tapioca–soybean meal-based diet.

The efficacy of lumenal hydrolysis of phytate is affected by the mineral contentof the diet. In turkeys, an increase in the dietary Ca : P ratio from 1.1 : 1 to 2.0 : 1decreased the efficacy of microbial phytase as an enhancer of P digestibility (Qianet al., 1996). Increasing Ca levels from low (0.4%) to normal (0.8%) levels in low Pmaize–soybean meal-based diets markedly reduced the efficacy of microbial phytasein improving the performance of weanling pigs (Lei et al., 1994).

Not surprisingly, the negative effect of phytic acid on mineral digestibilityis ameliorated by dietary supplementation with microbial phytase. Dietarysupplementation with microbial phytase improved Zn utilization in rats (Rimbachand Pallauf, 1993) and broiler chicks (Sebastian et al., 1996a). In pigs, phytasesupplementation increased the apparent absorption of Mg, Zn, Cu and Fe (Adeola,1995). Microbial phytase supplementation of maize–soybean meal-based dietsimproved Ca availability in poultry (Sebastian et al., 1996a,b). Increasing the Ca : Pratio while maintaining a constant level of phytase supplementation in maize–soybean meal-based diets correlated with a decrease in Ca retention (Yi et al., 1996a).This result is consistent with the formation of phytase-resistant insolublemineral–phytate complexes as the mineral content the diet increases.

Overall, studies on phytase supplementation of diets fed to production animalstend to indicate improvements in protein and amino acid digestibility (Ravindranet al., 1995). Microbial phytase supplementation of maize–tapioca–soybeanmeal-based diets increased the ileal digestibility of crude protein, and most aminoacids in pigs (Mroz et al., 1994). In turkey poults fed maize–soybean meal-baseddiets, microbial phytase tended to increase the ileal digestibility of N and most aminoacids (Yi et al., 1996b). Phytase supplementation of maize–soybean-based diets hadlittle effect on amino acid digestibilities in male broiler chicks but improved aminoacid digestibility in females, with an optimum effect on low P/low Ca diets(Sebastian et al., 1997).

The improvement of plant P and Ca digestibility with microbial phytasesupplementation of diets fed to production animals has led to the concept of definingmicrobial phytase as a dietary ingredient that contributes available P and Ca. Qian





et al. (1996) determined that 652 units of phytase kg−1 of maize–soybean meal-baseddiet was equivalent to 1 g of non-phytate P to turkey poults when the diet contained0.27% non-phytate P, while 963 units where required to supply 1 g of non-phytate Pwhen the diet contained 0.36% non-phytate P. With pigs, Yi et al. (1996a) foundthat 1146 units kg−1 of phytase provided 1 g of available P when added to soybeanmeal-based semipurified diets, while 785 units kg−1 provided 1 g of available P inmaize–soybean meal-based diets. Harper et al. (1997) estimated that 500 units ofphytase kg−1 of maize–soybean-based diets provided 0.87–0.96 g of available P togrower–finisher pigs.

Based on the efficacy of the enzyme in improving plant P and Ca availability inproduction animals, suppliers of commercially available sources of microbial phytasehave developed a set of recommendations for the P and Ca feed equivalence of theenzyme. These recommendations are listed as levels of phytase required for a givenage and species of animal that allow for a 0.1% reduction in available P and Ca levelsin the diet. Recommendation for starter diets are in the order of 500 units kg−1 andconsiderably lower for grower and finisher diets. These guidelines are useful andprovide some general indication of the sparing effect of phytase on the need fordietary supplementation with P and Ca. However, caution should be exercised inadapting these recommendations to a wide range of diets. As indicated in thischapter, the efficacy of microbial phytase will vary considerably with the ingredientcomposition of the diet. Maximum efficacy of phytase would be expected for a dietcharacterized by a high level of plant phytate in combination with low intrinsic(plant) phytase activity and a low multivalent mineral content. One couldhypothesize that supplemental phytase is likely to be largely ineffective in a meatmeal-based diet and of lower efficacy in a wheat-based diet with high intrinsicphytase. Clearly, research is needed to establish accurate recommendations foreffective levels of phytase for specific types of diets.

Supplementation with mineral chelatorsChelating agents such as EDTA and EDTA-like compounds are known to increaseZn availability to chicks (Nielsen et al., 1966) and turkey poults (Vohra and Kratzer,1964). Citrate and ascorbate improve Fe availability in vegetarian meals (Anand andSeshadri, 1995) and certain pectins improve Fe absorption in rats (Kim and Atallah,1993). The mechanism of the effect of chelators on mineral absorption may involvecompetitive chelation whereby soluble mineral–chelates form in the digesta andthereby decrease mineral binding to phytic acid. In this model, minerals boundto chelators are readily absorbed either as intact complexes or after dissociationand rebinding to mineral-binding transport sites on the intestinal brush bordermembrane. In theory, chelators may well be effective in converting phytate frommineral-bound phytase-resistant forms to phytase-susceptible forms in the digesta.Indeed Zyla et al. (1995) found that citrate, when added to a phytase-containingcocktail, increased phytate hydrolysis of maize–soybean meal under simulatedintestinal conditions. With pigs fed a maize–soybean meal-based diet, citrate tendedto enhance the efficacy of phytase in promoting P and Ca digestibility (Li et al.,

76 D.D. Maenz




1998). Further investigation is required to determine the potential for chelators asdietary supplements to promote phytate digestion.

Summary and Conclusions

Application of commercial products containing microbial phytase activity tofeed ingredients has proved to be a practical and effective method for improvingphytate-P utilization by monogastric animals and thereby reducing P output in themanure. However, the process of phytate-P digestion is a complex issue and variesconsiderably with the ingredient composition of the diet and the physical conditionsin the digesta. Feed ingredients such as wheat with high levels of intrinsic phytaseare characterized by relatively high utilization of phytate-P in comparison with feedingredients such as oilseed meals that contain low levels of intrinsic phytase.Multivalent minerals can complex with phytate at neutral pH and form phytase-resistant complexes. At present, oversimplistic guidelines are in place defining theavailable P content for a broad spectrum of feed ingredients. The opportunityexists to measure actual phytate-P availability for specific feed ingredients and dietformulations and to measure the true efficacy of microbial phytase with differentfeeding scenarios. This information can then be used for more precise dietformulation based on a true reflection of available P requirements. More precisediet formulation will improve the overall efficiency of dietary P utilizationand thereby reduce P levels in the manure. Finally, methods of reducing the phytatecontent of the feed ingredient, such as developing low-phytate plants, controlledgermination, or enzyme pretreatment of feed ingredients under controlledconditions in vitro, have the potential to produce feed ingredients characterizedby low levels of phytate and high levels of readily available inorganic P.

References

Adeola, O. (1995) Digestive utilization of minerals by weanling pigs fed copper- and phytase-supplemented diets. Canadian Journal of Animal Science 75, 603–610.

Al-Asheh, S. and Duvnjak, Z. (1994a) Effects of glucose concentration on the biomassand phytase productions and reduction of the phytic acid content in canola meal byAspergillus carbonarius during a solid-state fermentation process. Biotechnology Program10, 353–359.

Al-Asheh, S. and Duvnjak, Z. (1994b) The effects of surfactants on the phytase productionand the reduction of the phytic acid content in canola meal by Aspergillus carbonariusduring a solid state fermentation process. Biotechnology Letters 16, 183–188.

Al-Asheh, S. and Duvnjak, Z. (1995a) The effect of phosphate concentration on phytaseproduction and the reduction of phytic acid content in canola meal by Aspergilluscarbonarius during a solid-state fermentation process. Applied Microbiology andBiotechnology 43, 25–30.





Al-Asheh, S. and Duvnjak, Z. (1995b) Phytase production and decrease of phytic acid contentin canola meal by Aspergillus carbonarius in solid-state fermentation. World Journal ofMicrobiology and Biotechnology 11, 228–231.

Alko (1994) Phytate-degrading enzyme, e.g. phytase and acid phosphatase over-production.Patent WO 9 403 612.

Anand, A.N. and Seshadri, S. (1995) A quantitative model for prediction of ironbioavailability from Indian meals: an experimental study. International Journal of FoodSciences and Nutrition 46, 335–342.

Ballam, G.C., Nelson, T.S. and Kirby, L.K. (1984) Effect of fiber and phytate source and ofcalcium and phosphorus level on phytate hydrolysis in the chick. Poultry Science 63,333–338.

Barrier-Guillot, B., Casado, P., Maupetit, P., Jondreville, C. and Gatel, F. (1996a) Wheatphosphorus availability. 1. In vitro study; factors affecting endogenous phytasic activityand phytic phosphorus content. Journal of the Science of Food and Agriculture 70, 62–68.

Barrier-Guillot, B., Casado, P., Maupetit, P., Jonderville, C. and Gatel, F. (1996b) Wheatphosphorus availability. 2. In vivo study in broilers and pigs; relationship withendogenous phytasic activity and phytic phosphorus content in wheat. Journal of theScience of Food Agriculture 70, 69–74.

Bau, H.-M., Villaume, C., Nicolas, J. and Mejean, L. (1997) Effect of germination onchemical composition, biochemical constituents and antinutritional factors of soya bean(Glycine max) seeds. Journal of the Science of Food and Agriculture 73, 1–9.

Berka, R.M., Yoder, W., Takagi, S. and Boominathan, K.C. (Novo-Nordisk-Biotech) (1995a)Aspergillus foetidus cells expressing heterologous enzyme. The fungal promotor isAspergillus niger phytase. Patent WO 9 515 390.

Berka, R.M., Yoder, W., Takagi, S. and Boominathan, K.C. (Novo-Nordisk-Biotech)(1995b) Aspergillus japonicum-type cells expressing heterologous enzyme. The fungalpromotor is Aspergillus niger phytase. Patent WO 9 515 391.

Biehl, R.R. and Baker, D.H. (1997) Utilization of phytate and nonphytate phosphorus inchicks as affected by source and amount of vitamin D3. Journal of Animal Science 75,2986–2993.

Biehl, R.R., Baker, D.H. and Deluca, H.F. (1998) Activity of various hydroxylated vitamin D3

analogs for improving phosphorus utilization in chicks receiving diets adequate invitamin D3. British Poultry Science 39, 408–412.

Bitar, K. and Reinhold, J.G. (1972) Phytase and alkaline phosphatase activities in intestinalmucosae of rat, chicken, calf and man. Biochimica et Biophysica Acta 268, 442–452.

Bosch, D.J., Zhu, M. and Kornegay, E.T. (1998) Net returns from microbial phytase whencrop applications of swine manure are limited by phosphorus. Journal of ProductionAgriculture 11, 157–213.

Brocades (1991) DNA sequence encoding phytase. Patent EP 420 358.Brocades (1993) A gene expression and protein vector secretion system for a fungus. Patent EP

549 062.Bruce, J.A.M. and Sundstol, F. (1995) The effect of microbial phytase in diets of pigs on

apparent ileal and faecal digestibility, pH and flow of digesta measurements in growingpigs fed a high-fibre diet. Canadian Journal of Animal Science 75, 121–127.

Caldwell, R.A. (1992) Effect of calcium and phytic acid on the activation of trypsinogen andthe stability of trypsin. Journal of Agriculture and Food Chemistry 40, 43–46.

Camovale, E., Lugaro, E. and Lombardi-Boccia, G. (1988) Phytic acid in faba bean and pea.Effect on protein availability. Cereal Chemistry 65, 114–117.

78 D.D. Maenz




Champagne, E.T. and Fisher, M.S. (1990) Binding diverences of Zn(II) and Cu(II) ions withphytate. Journal of Inorganic Biochemistry 38, 217–223.

Champagne, E.T., Fisher, M.S. and Hinojosa, O. (1990) NMR and ESR studies of interac-tions among divalent cations, phytic acid, and N-acetyl-amino acids. Journal of InorganicBiochemistry 38, 199–215.

Cheryan, M. (1980) Phytic acid interactions in food systems. CRC Critical Reviews in FoodScience and Nutrition 13, 297–335.

Cheryan, M., Anderson, F.W. and Grynspan, F. (1983) Magnesium–phytate complexes:effect of pH and molar ratio on solubility characteristics. Cereal Chemistry 60, 235–237.

Chitra, U., Vimala, V., Singh, U. and Geervani, P. (1995) Variability in phytic acidcontent and protein digestibility of grain legumes. Plant Foods for Human Nutrition 47,163–172.

Costello, A.J.R., Glonek, T. and Myers, T.C. (1976) 31P-nuclear magnetic resonance-pHtitrations of myo-inositol hexaphosphate. Carbohydrate Resource 46, 159–171.

Daniel, T.C., Sharpley, A.N. and Lemunyon, J.L. (1998) Agricultural phosphorus andeutrophication: a symposium overview. Journal of Environmental Quality 27, 251–257.

Davidsson, L., Galan, P., Kastenmayer, P., Cherouvrier, F., Juillerat, M.A., Hercberg, S. andHurrell, R.F. (1994) Iron bioavailability studied in infants: the influence of phyticacid and ascorbic acid in infant formulas based on soy isolate. Pediatric Research 36,816–822.

Davidsson, L., Almgren, A., Juillerat, M.A. and Hurrell, R.F. (1995) Manganese absorption inhumans: the effect of phytic acid and ascorbic acid in soy formula. American Journal ofClinical Nutrition 62, 984–987.

Davies, M.I. and Motzok, I. (1972) Properties of chick intestinal phytase. Poultry Science 51,494–501.

Davies, M.I., Ritcey, G.M. and Motzok, I. (1970) Intestinal phytase and alkaline phosphataseof chicks: influence of dietary calcium, inorganic and phytate phosphorus and vitaminD3. Poultry Science 49, 1280–1286.

de Boer, I.J.M., Peters, H.T.A., Grossman, M. and Koops, W.J. (1997) Nutrient flows inagriculture in the Netherlands with special emphasis on pig production. Journal ofAnimal Science 75, 2054–2063.

de Boland, A.R., Garner, G.B. and O’Dell, B.L. (1975) Identification and properties of‘phytate’ in cereal grains and oilseed products. Journal of Agriculture and Food Chemistry23, 1186–1189.

Dvorakova, J. (1998) Phytase: sources, preparation and exploitation. Folia Microbiology 43,323–338.

Ebune, A., Al-Asheh, S. and Duvnjak, Z. (1995a) Production of phytase during solid statefermentation using Aspergillus ficuum NRRL 3135 in canola meal. Bioresource Technology53, 7–12.

Ebune, A., Al-Asheh, S. and Duvnjak, Z. (1995b) Effects of phosphate, surfactants andglucose on phytase production and hydrolysis of phytic acid in canola meal by Aspergillusficuum during solid state fermentation. Bioresource Technology 54, 241–247.

Edwards, H.M. Jr (1983) Phosphorus. I. Effect of breed and strain on utilization ofsuboptimal levels of phosphorus in the ration. Poultry Science 62, 77–84.

Edwards, H.M. Jr (1993) Dietary 1,25-dihydroxycholecalciferol supplementation increasesnatural phytate phosphorus utilization in chickens. Journal of Nutrition 123, 567–577.

Eeckhout, W. and De Paepe, M. (1994) Total phosphorus, phytate-phosphorus and phytaseactivity in plant feed stuffs. Animal Feed Science and Technology 47, 19–29.





Ehrlich, K.C., Montalbano, B.G., Mullaney, E.J., Dischinger, H.C. Jr and Ullah, A.H.J.(1993) Identification and cloning of a second phytase gene (phyB) from Aspergillus niger(ficuum). Biochemical and Biophysical Research Communications 195, 53–57.

Ellis, R., Morris, E.R. and Hill, A.D. (1982) Bioavailability to rats of iron and zinc in calcium–iron–phytate and calcium–zinc–phytate complexes. Nutrition Research 2, 319–322.

Erdman, J.W. Jr (1979) Oilseed phytates: nutritional implications. Journal of the American OilChemists Society 56, 736–741.

Ertl, D.S., Young, K.A. and Raboy, V. (1998) Plant genetic approaches to phosphorusmanagement in agricultural production. Journal of Environmental Quality 27, 299–304.

Forbes, R.M., Parker, H.M. and Erdman, J.W. Jr (1984) Effects of dietary phytate, calciumand magnesium levels on zinc bioavailability to rats. Journal of Nutrition 114,1421–1425.

Grynspan, F. and Cheryan, M. (1983) Calcium phytate: effect of pH and molar ratio on invitro solubility. Journal of the American Oil Chemists Society 60, 1761–1764.

Han, O., Failla, M.L., Hill, A.D., Morris, E.R. and Smith, J.C. (1994) Inositol phosphatesinhibit uptake and transport of iron and zinc by human intestinal cell line. Journal ofNutrition 124, 580–587.

Harms, R.H., Waldroup, P.W., Shirley, R.L. and Ammerman, C.B. (1962) Availability ofphytic acid phosphorus for chick. Poultry Science 63, 1189–1191.

Harper, A.F., Kornegay, E.T. and Schell, T.C. (1997) Phytase supplementation of low-phosphorus growing–finishing pig diets improves performance, phosphorus digestibility,and bone mineralization and reduces phosphorus excretion. Journal of Animal Science 75,3174–3186.

Iqbal, T.H., Lewis, K.O. and Cooper, B.T. (1994) Phytase activity in the human and rat smallintestine. Gut 35, 1233–1236.

Johnson, L.F. and Tete, M.E. (1969) Structure of ‘phytic acids’. Canadian Journal ofChemistry 47, 67–73.

Kemme, P.A., Radcliffe, J.S., Jongbloed, A.W. and Mroz, Z. (1997a) The effects of sow parityon digestibility of proximate components and minerals during lactation as influenced bydiet and microbial phytase supplementation. Journal of Animal Science 75, 2147–2153.

Kemme, P.A., Radcliffe, J.S., Jongbloed, A.W. and Mroz, Z. (1997b) Factors affectingphosphorus and calcium digestibility in diets for growing finishing pigs. Journal ofAnimal Science 75, 2139–2146.

Kemme, P.A., Jongbloed, A.W., Mroz, Z. and Beynen, A.C. (1998) Diurnal variation indegradation of phytic acid by plant phytase in the pig stomach. Livestock ProductionScience 54, 33–44.

Kim, M. and Atallah, M.T. (1993) Intestinal solubility and absorption of ferrous iron ingrowing rats are affected by different dietary pectins. Journal of Nutrition 123, 117–124.

Kim, Y., Kim, H., Bae K.-S., Yu, J.-H. and Oh, T.-K. (1998) Purification and properties of athermostable phytase from Bacillus sp. DS11. Enzyme and Microbial Technology 22, 2–7.

Lei, X.G., Ku, P.K., Miller, E.R. and Yokoyama, M.T. (1993) Supplementing corn–soybeanmeal diets with microbial phytase linearly improves phytate phosphorus utilization byweanling pigs. Journal of Animal Science 71, 3359–3367.

Lei, X.G., Ku, P.K., Miler, E.R., Yokoyama, M.T. and Ullrey, D.E. (1994) Calcium levelaffects the efficacy of supplemental microbial phytase in corn–soybean meal diets ofweanling pigs. Journal of Animal Science 72, 139–143.

Li, D., Che, X., Wang, Y., Hong, C. and Thacker, P.A. (1998) Effect of microbial phytase,vitamin D3, and citric acid on growth performance and phosphorus, nitrogen andcalcium digestibility in growing swine. Animal Feed Science and Technology 73, 173–186.

80 D.D. Maenz




Li, J., Hegeman, C.E., Hanlon, R.W., Lacy, G.H., Denbow, M.D. and Grabau, E.A. (1997)Secretion of active recombinant phytase from soybean cell-suspension cultures. PlantPhysiology 114, 1103–1110.

Liu, B.-L., Rafiq, A., Tzeng, Y.-M. and Rob, A. (1998) The induction and characterization ofphytase and beyond. Enzyme and Microbial Technology 22, 415–424.

Lott, J.N.A. and Ockenden, I. (1986) The fine structure of phytate-rich particles in plants.In: Graf, E. (ed.) Phytic Acid Chemistry and Applications. Pilatus Press, Minneapolis,pp. 43–55.

Lott, J.N.A., Randall, P.J., Goodchild, D.J. and Craig, S. (1985) Occurrence of globoidcrystals in cotyledonary protein bodies of Pisum sativum as influenced by experimentallyinduced changes in Mg, Ca and K contents of seeds. Australian Journal of Plant Physiology12, 341–353.

Lu, S.-Y., Kim, H., Eskin, N.A.M., Latta, M. and Johnson, S. (1987) Changes in phytaseactivity and phytate during the germination of six canola cultivars. Journal of Food Science52, 173–175.

Lucas, M. (1983) Determination of acid surface pH in vivo in rat proximal jejunum. Gut 24,734–739.

Maddaiah, V.T., Kurnick, A. and Reid, B.L. (1964) Phytic acid studies. Proceedings of theSociety for Experimental Biology and Medicine 115, 391–393.

Maenz, D.D. and Classen, H.L. (1998) Phytase activity in the small intestinal brush bordermembrane of the chicken. Poultry Science 77, 557–563.

Maenz, D.D., Engele-Schaan, C.M., Newkirk, R.W. and Classen, H.L. (1999) The effect ofminerals and mineral chelators on the formation of phytase-resistant and phytase-susceptible forms of phytic acid in solution and in a slurry of canola meal. Animal FeedScience and Tecnology 81, 177–192.

Mahajan, A. and Dua, S. (1997) Nonchemical approach for reducing antinutritional factors inrapeseed (Brassica campestris var. toria) and characterization of enzyme phytase. Journal ofAgricultural and Food Chemistry 45, 2504–2508.

Maugenest, S., Martinez, I. and Lescure, A.M. (1997) Cloning and characterization of acDNA encoding a maize seedling phytase. Biochemical Journal 322, 511–517.

Mitchell, R.D. and Edwards, H.M. Jr (1996) Effects of phytase and 1,25-dihydroxysholecalciferol on phytate utilization and the quantitative requirement forcalcium and phosphorus in young broiler chickens. Poultry Science 75, 95–110.

Mohammed, A., Gibney, M.J. and Taylor, T.G. (1991) The effects of dietary levels ofinorganic phosphorus, calcium and cholecalciferol on the digestibility of phytate-P by thechick. British Journal of Nutrition 66, 251–259.

Mroz, Z., Jongbloed, A.W. and Kemme, P.A. (1994) Apparent digestibility and retention ofnutrients bound to phytate complexes as influenced by microbial phytase and feedingregime in pigs. Journal of Animal Science 72, 126–132.

Nair, V.C., Laflamme, J. and Duvnjak, Z. (1991) Production of phytase by Aspergillus ficuumand reduction of phytic acid content in canola meal. Journal of the Science of Food andAgriculture 54, 355–365.

Nasi, M. and Helander, E. (1994) Effects of microbial phytase supplementation and soakingof barley–soybean meal on availability of plant phosphorus for growing pigs. ActaAgriculture Scandinavian Section A, Animal Science 44, 79–86.

Nelson, T.S. (1976) The hydrolysis of phytate phosphorus by chicks and laying hens. PoultryScience 55, 2262–2264.

Nelson, T.S. and Kirby, L.K. (1979) Effect of age and diet composition on the hydrolysis ofphytate phosphorus by rats. Nutrition Reports International 20, 729–734.





Nielsen, F.H., Sunde, M.L. and Hoekstra, W.G. (1966) Effect of some dietary syntheticand natural chelating agents on the zinc-deficiency syndrome in the chick. Journal ofNutrition 89, 35–42.

Niziolek, S. (1995) Changes in phytic acid, phytase activity and inorganic phosphorus duringmaturation of triticale varieties resistant and susceptible to pre-harvest sprouting. ActaPhysiologicae Plantarum 17, 269–274.

O’Dell, B.L., de Boland, A.R. and Koirtyohann, S.R. (1972) Distribution of phytate andnutritionally important elements among morphological components of cereal grains.Journal of Agricultural and Food Chemistry 20, 718–721.

Ogawa, M., Tanaka, K. and Kasai, Z. (1975) Isolation of high phytin containing particlesfrom rice grains using an aqueous polymer two-phase system. Agricultural BiologicalChemistry 39, 695–700.

Pallauf, J., Rimbah, G., Pippig, S., Schindler, B., Hohler, D. and Most, E. (1994) Dietaryeffect of phytogenic phytase and an addition of microbial phytase to diet based on fieldbeans, wheat, peas and barley on utilization of phosphorus, calcium, magnesium, zincand protein in piglets. Zeitschrift für Ernahrungswissenschaft 33, 128–135.

Panlabs, Alko (1994) Nucleic acid encoding phytase and acid 2.5 phosphatase. Patent WO 9403 072.

Parry, R. (1998) Agricultural phosphorus and water quality: a US Environmental ProtectionAgency perspective. Journal of Environmental Quality 27, 258–261.

Pasamontes, L., Haiker, M., Wyss, M., Tessier, M. and Van Loon, A.P.G.M. (1997) Genecloning, purification and characterization of a heat-stable phytase from fungus Aspergillusfumigatus. Applied and Environmental Microbiology 63, 1696–1700.

Piddington, C.S., Houston, C.S., Paloheimo, M., Cantrell, M., Muttinen Oinonen, A.,Nevalainen, H. and Rambosek, J. (1993) The cloning and sequencing of the genesencoding phytase (phy) and 2.5-optimum acid phosphatase (aph) from Aspergillus nigervar. awamori. Gene 133, 55–62.

Pointillart, A. (1994) The importance of cereal phytases. Feed Mix 2, 12–15.Qian, H., Kornegay, E.T. and Denbow, D.M. (1996) Phosphorus equivalence of microbial

phytase in turkey diets as influenced by calcium to phosphorus ratios and phosphoruslevels. Poultry Science 75, 69–81.

Ravindran, V., Bryden, W.L. and Kornegay, E.T. (1995) Phytates: occurrence, bioavailabilityand implications in poultry nutrition. Poultry and Avian Biology Reviews 6, 125–143(5.1.9).

Rickard, S.E. and Thompson, L.U. (1997) Interactions and biological effects of phytic acid.In: Shadidi, F. (ed.) Antinutrients and Phytochemicals in Food. American ChemicalSociety, Washington DC, pp. 294–312.

Rimbach, G. and Pallauf, J. (1993) Enhancement of zinc utilization from phytate-rich soyprotein isolate by microbial phytase. Zeitschrift für Ernahrungswissenschaft 32, 308–315.

Saha, P.R., Weaver, C.M. and Mason, A.C. (1994) Mineral bioavailability in rats fromintrinsically labeled whole wheat flour of various phytate levels. Journal of Agriculturaland Food Chemistry 44, 2531–2535.

Sandberg, A.-S. and Svanberg, U. (1991) Phytate hydrolysis by phytase in cereals; effects on invitro estimation of iron availability. Journal of Food Science 56, 1330–1333.

Sandberg, A.-S., Larsen, T. and Sandstrom, B. (1993) High dietary calcium level decreasescolonic phytate degradation in pigs fed rapeseed diet. Journal of Nutrition 123, 559–566.

Scheideler, S.E. and Sell, J.L. (1987) Utilization of phytate phosphorus in laying hens asinfluenced by dietary phosphorus and calcium. Nutrition Reports International 35,1073–1081.

82 D.D. Maenz




Sebastian, S., Touchburn, S.P., Chavez, E.R. and Lague, P.C. (1996a) The effects ofsupplemental microbial phytase on the performance and utilization of dietary calcium,phosphorus, copper and zinc in broiler chickens fed corn–soybean diets. Poultry Science75, 729–736.

Sebastian, S., Touchburn, S.P., Chavez, E.R. and Lague, P.C. (1996b) Efficacy of supple-mental microbial phytase at different dietary calcium levels on growth performance andmineral utilization of broiler chickens. Poultry Science 75, 1516–1523.

Sebastian, S., Touchburn, S.P., Chavez, E.R. and Lague, P.C. (1997) Apparent digestibility ofprotein and amino acids in broiler chickens fed a corn–soybean diet supplemented withmicrobial phytase. Poultry Science 76, 1760–1769.

Singh, M. and Krikorian, A.D. (1982) Inhibition of trypsin activity in vitro by phytate. Journalof Agricultural and Food Chemistry 30, 799–800.

Skoglund, E., Larsen, T. and Sandberg, A.-S. (1997) Comparison between steeping andpelleting a mixed diet at different calcium levels on phytate degradation in pigs. CanadianJournal of Animal Science 77, 471–477.

Storebakken, T., Shearer, K.D. and Roem, A.J. (1998) Availability of protein phosphorus andother elements in fish meal, soy-protein concentrate and phytase-treated soy-protein-concentrate-based diets to Atlantic salmon Salmo salar. Aquaculture 161, 365–379.

Tao, S.H., Fox, M.R.S., Phillippy, B.Q., Fry, B.E., Johnson, M.L. and Johnson, M.R. (1986)Effects of inositol phosphates on mineral utilization. Federation of American Societies forExperimental Biology. Federation Proceedings 45, 819–826.

Temperton, H. and Cassidy, J. (1964a) Phosphorus requirements of poultry. I. The utilizationof phytin phosphorus by the chick as indicated by balance experiments. British PoultryScience 5, 75–80.

Temperton, H. and Cassidy, J. (1964b) Phosphorus requirements of poultry. II. Theutilization of phytin phosphorus by the chick for growth and bone formation. BritishPoultry Science 5, 81–86.

Thompson, L.U. and Serraino, M.R. (1986) Effect of phytic acid reduction on rapeseedprotein digestibility and amino acid absorption. Journal of Agricultural and FoodChemistry 34, 468–469.

Torre, M., Rodriguez, A.R. and Saura-Calixto, F. (1991) Effects of dietary fiber and phyticacid on mineral availability. Critical Reviews in Food Science and Nutrition 1, 1–22.

Ullah, A.H.J. (1988) Aspergillus ficuum phytase: partial primary structures, substrateselectivity, and kinetic characterization. Preparative Biochemistry 18, 459–471.

Ullah, A.H.J. and Gibson, D.M. (1987) Extracellular phytase (E.C. 3.1.3.8) from Aspergillusficuum NRRL 3135: purification and characterization. Preparative Biochemistry 17,63–91.

Ullah, A.H.J. and Phillippy, B.Q. (1994) Substrate selectivity in Aspergillus ficuum phytaseand acid phosphatase using myo-inositol phosphates. Journal of Agricultural and FoodChemistry 42, 423–425.

Van Hartingsveldt, W., van Zeijl, C.M., Harteveld, G.M., Gorika, R.J., Suykerbuyk, M.E.,Luiten, R.G., van Paridon, P.A., Selten, G.C., Veenstra, A.E. and van Gorcom, R.F.(1993) Cloning, characterization and over expression of the phytase-encoding gene(phyA) of Aspergillus niger. Gene 127, 87–94.

Verwoerd, T.C., van Paridon, P.A., van Ooen, A.J., van Lent, L.W., Hockems, A. and Pen, J.(1995) Stable accumulation of Aspergillus niger phytase in transgenic tobacco leaves. PlantPhysiology 109, 1199–1205.

Vohra, P. and Kratzer, F.H. (1964) Influence of various chelating agents on the availability ofzinc. Journal Nutrition 82, 249–256.





Vohra, P. and Kratzer, F.H. (1965) Phytic acid mineral complexes. Proceedings of the Society forExperimental Biology and Medicine 120, 447–449.

Volfova, O., Dvorakova, J., Hanzlikova, A. and Jandera, A. (1994) Phytase from Aspergillusniger. Folia Microbiology 39, 481–484.

Waldroup, P.W., Ammerman, C.B. and Harms, R.H. (1964a) The availability of phytic acidphosphorus for chicks. 2. Comparison of phytin phosphorus sources. Poultry Science 43,426–432.

Waldroup, P.W., Ammerman, C.B. and Harms, R.H. (1964b) The availability of phytic acidphosphorus for chicks. 3. Effect of calcium and vitamin D3 levels on the utilization ofcalcium phytate. Poultry Science 43, 926–931.

Wise, A. (1983) Dietary factors determining the biological activities of phytate. NutritionAbstracts and Reviews 53, 791–806.

Yang, W.-J., Matsuda, Y., Inomata, M. and Nakagawa, H. (1991) Developmental and dietaryintroduction of the 90K subunit of rat intestinal phytase. Biochemica et Biophysica Acta1075, 83–87.

Yi, Z., Kornegay, E.T., Ravindran, V. and Denbow, D.M. (1996a) Improving phytatephosphorus availability in corn and soybean meal for broilers using microbial phytase andcalculation of phosphorus equivalency values for phytase. Poultry Science 75, 240–249.

Yi, Z., Kornegay, E.T. and Denbow, D.M. (1996b) Effect of microbial phytase on nitrogenand amino acid digestibility and nitrogen retention of turkey poults fed corn–soybeanmeal diets. Poultry Science 75, 979–990.

Yoon, J.H., Thompson, L.U. and Jenkins, D.J.A. (1983) The effect of phytic acid on in vitrorate of starch digestibility and blood glucose response. American Journal of ClinicalNutrition 38, 835–842.

Zhou, J.R., Fordyce, E.J., Raboy, V., Dickinson, D.B., Wong, M.S., Burns, R.A. andErdman, J.W. Jr (1992) Reduction of phytic acid in soybean products improves zincbioavailability in rats. Journal of Nutrition 122, 2466–2473.

Zhu, X.S., Seib, P.A., Allee, G.L. and Liang, Y.T. (1990) Preparation of a low-phytatefeed mixture and bioavailability of its phosphorus to chicks. Animal Feed Science andTechnology 27, 341–351.

Zyla, K. and Koreleski, J. (1993) In-vitro and in-vivo dephosphorylation of rapeseed meal bymeans of phytate-degrading enzymes derived from Aspergillus niger. Journal of the Scienceof Food and Agriculture 61, 1–6.

Zyla, K., Ledoux, D.R., Garcia, A. and Veum, T.L. (1995a) An in vitro procedure for studyingenzymic dephosphorylation of phytate in maize–soyabean feeds for turkey poults. BritishJournal of Nutrition 74, 3–17.

Zyla, K., Ledoux, D.R. and Veum, T.L. (1995b) Complete enzymic dephosphorylation ofcorn–soybean meal feed under simulated intestinal conditions of the turkey. Journal ofAgricultural and Food Chemistry 43, 288–294.

Zyla, K., Ledoux, D.R., Kujawski, M. and Veum, T.L. (1996) The efficacy of an enzymiccocktail and a fungal mycelium in dephosphorylating corn–soybean meal-based feeds fedto growing turkeys. Poultry Science 75, 381–387.

84 D.D. Maenz




B.V. McClearyAnalysis of Feed Enzymes4

Introduction

Enzymes are added to animal feed to increase its digestibility, to remove anti-nutritional factors, to improve the availability of components, and for environmentalreasons (Campbell and Bedford, 1992; Walsh et al., 1993). A wide variety ofcarbohydrase, protease, phytase and lipase enzymes find use in animal feeds. Inmonogastric diets, enzyme activity must be sufficiently high to allow for the relativelyshort transit time. Also, the enzyme employed must be able to resist unfavourableconditions that may be experienced in feed preparation (e.g. extrusion and pelleting)and that exist in the gastrointestinal tract. Measurement of trace levels of enzymes inanimal feed mixtures is difficult. Independent of the enzyme studied, many of theproblems experienced are similar; namely, low levels of activity, extraction problems,inactivation during feed preparation, non-specific binding to other feed componentsand inhibition by specific feed-derived inhibitors, e.g. specific xylanase inhibitors inwheat flour (Debyser et al., 1999).

In this chapter, some of the procedures used to assay for β-glucanase, β-xylanase,α-amylase, α-galactosidase, phytase and endo-protease enzymes will be described. Inparticular, some of the problems and limitations of current assay procedures will bediscussed.

b-Glucanase

The anti-nutritional properties of β-glucans (1,3 : 1,4-β-D-glucans, mixed-linkageβ-glucans) have been known for many years (Aastrup, 1979). In animal feeds themajor source of β-glucan is barley grain. Although levels are also high in oats, theseare rarely fed to chickens and pigs. The anti-nutritional properties of β-glucan areattributed to their viscosity-inducing properties, which significantly affect the rate ofmovement of barley-based diets through the digestive tract of chickens, and reducethe rate of nutrient absorption. This problem is effectively ‘dissolved’ by thejudicious use of β-glucan-degrading enzymes.



Analysis of Feed EnzymesB.V. McCLEARY

Megazyme International Ireland Limited, Bray BusinessPark, Bray, County Wicklow, Ireland

4



A wide range of enzymes are active on β-glucan and these include the endo-acting fungal cellulases [endo-1,4-β-D-glucanase (EC 3.2.1.4)] and the bacterialβ-glucanase [lichenase; endo-1,3 : 1,4-β-D-glucanase (EC 3.2.1.73)]. Both groups ofenzymes cleave within the main chain of mixed-linkage β-glucan, although theirpoint of action is different (Fig. 4.1). Fungal cellulases also hydrolyse cellulose(1,4-β-D-glucan) in an endo-hydrolytic pattern, but the bacterial 1,3 : 1,4-β-glucanases have no action on cellulose. This point of difference is important indesigning specific substrates for the assay of these groups of enzymes.

A range of substrates and assay procedures is available for the measurement ofbacterial β-glucanase and cellulase enzymes, some of which are listed in Table 4.1.Bacterial 1,3 : 1,4-β-D-glucanase is assayed using substrates based on a 1,3 : 1,4-β-D-glucan polysaccharide, i.e. oat or barley β-glucan or lichenan. Procedures for thepurification of barley and oat β-glucan have been developed and are reported in theliterature (McCleary, 1988), and high purity polysaccharides are commerciallyavailable. However, the purification of lichenan is much more difficult. Lichenan ispurified from Icelandic moss, but it occurs together with isolichenan (Chandra et al.,1957), from which it is hard to separate. Other glucose-containing polysaccharides(starch and cellulose) are absent from purified lichenan.

86 B.V. McCleary


Fig. 4.1. Schematic representation of the mode of action of cellulase and1,3 : 1,4-β-glucanase on mixed-linkage β-glucan.

Substrate Nature Assay procedure

CellulaseBarley β-glucanLichenanCMC-7MCMC-4MAzo-CMCAzo-barley glucanCellazyme C tabletsCellazyme T tablets

(tamarind xyloglucan)Beta-Glucazyme tablets1,3 : 1,4-b-GlucanaseBarley β-glucanAzo-barley glucanBeta-Glucazyme tablets

SolubleSolubleSolubleSoluble/gelSolubleSolubleGel particlesGel particles

Gel particles

SolubleSolubleGel particles

Reducing-sugar or viscometricReducing-sugar or gel plateReducing-sugar or viscometricReducing-sugarChromogenic substrateChromogenic substrateChromogenic substrateChromogenic substrate

Chromogenic substrate

Reducing-sugar or viscometricChromogenic substrateChromogenic substrate

Table 4.1. Substrates for the assay of cellulase and β-glucanase.



Reducing-sugar methods

Pure β-glucan (barley or oat) can be used directly to assay cellulase or β-glucanaseemploying either a reducing-sugar or viscometric assay procedure. Of the reducing-sugar procedures available, the only commonly used method that gives astoichiometric colour response with homologous oligosaccharides of varyingdegrees of polymerization (DP) is the Nelson/Somogyi procedure (Somogyi, 1960).Consequently, it is the only method that gives a true measure of glycosidic bondscleaved, and thus of enzyme activity. For this reason, this is the method of choice foruse as a reference method against which all other assay procedures can be comparedand standardized. Details of a suggested format for the assay of β-glucanase usingbarley β-glucan and the Nelson/Somogyi reducing-sugar procedure are given inAppendix 4.1.

Several cellulose-based substrates are available for the assay of endo-1,4-β-glucanase (cellulase). These substrates range from highly crystalline cotton cellu-lose to amorphous cellulose and lightly or heavily substituted chemical derivativessuch as carboxymethyl-cellulose 4M (CMC-4M) and CMC-7M. CMC-4M is amedium viscosity (300–600 cP at 2%, 25oC) cellulose with a degree of substitutionwith carboxymethyl groups of 0.4 (i.e. over ten sugar residues, only four of the 30available hydroxyl groups are substituted by carboxymethyl groups). CMC-7M iscompletely soluble, whereas CMC-4M at 1% concentration in water is somewherebetween a solution and a colloidal suspension. Many cellulase enzymes hydrolyseCMC-7M at the same rate as CMC-4M, but for some the rate is substantially lower.Consequentially, CMC-4M is the substrate of choice. Cellulase enzymes act onother polysaccharide substrates, such as mixed-linkage β-glucan and tamarind-seedxyloglucan (which is composed of a 1,4-β-D-glucan backbone, to which D-xyloseis attached in a relatively regular manner, some of which is further substitutedby D-galactose). Both of these polysaccharides serve as the base for excellent dyedsubstrates for the assay of cellulases (Table 4.1).

Traditionally, cellulase activity has been standardized using a reducing-sugarassay with either CMC-7M or barley β-glucan as substrate. In most cases thedinitrosalicyclic acid (DNSA) reducing-sugar method (Bailey, 1988) was used withcellobiose as the standard. The DNSA method does not give a stoichiometric colourresponse with homologous oligosaccharides of varying degrees of polymerization;thus it is not ideal as a reference method. The Nelson/Somogyi reducing-sugarmethod (Somogyi, 1960) is preferable. Also, since some cellulase enzymes hydrolyseCMC-4M more rapidly than CMC-7M, then CMC-4M or barley β-glucan are thesubstrates of choice.

Viscometric methods

The Nelson/Somogyi reducing-sugar procedure (Somogyi, 1960) is an excellentmethod for the assay and standardization of relatively pure β-glucanase preparations,i.e. preparations essentially devoid of other enzyme activities active on the substrate,

Analysis of Feed Enzymes 87




and for preparations containing low levels of reducing-sugars (which interferewith the assay). However, the method cannot be applied to the measurement ofβ-glucanase in animal feed mixtures. For such materials, other procedures need to beemployed, such as viscometric methods and methods based on the use of dyedpolysaccharide substrates. Both procedures specifically measure endo-hydrolaseactivity and can be used in the presence of high levels of reducing sugars.

In industry, the most commonly used viscometric assay for polysaccharideendo-hydrolase activity is the Institute of Brewing procedure for the measurementof β-glucanase in malt (Bathgate, 1979). For reproducibility, accuracy and reliability,a β-glucan preparation of high purity and a defined viscosity range is essential. Inprinciple, the enzyme preparation is mixed with buffered β-glucan solution(10 mg ml−1) under controlled temperature conditions. The viscosity (as the time toflow between two marked points on a standardized type C U-tube viscometer) ismeasured at various time intervals after the addition of the enzyme to the substrate.The inverse reciprocal viscosity is calculated and plotted against time of incubation(see Appendix 4.2). The slope of the curve is a direct measure of enzyme activity. Analternative substrate is CMC-7M. This is completely soluble in water and is rapidlyhydrolysed by some endo-cellulases. A procedure for the measurement of β-glucanaseor xylanase in animal feeds, based on the Institute of Brewing IRVU method, is givenin Appendix 4.2.

Agar plate diffusion procedures

Several semi-quantitative procedures have been developed for the assay of enzymeactivity based on the rate and degree of diffusion of the enzyme through agar plates.The enzyme is applied in a central well and diffuses through agar containing aspecific substrate. The level of activity is then monitored by staining the plate with adye specific for the particular polysaccharide substrate. In one such procedure, Walshet al. (1995) detected β-glucanase using an agar plate impregnated with lichenan, andactivity was detected by staining with Congo Red. These procedures do work, but atbest are semi-quantitative.

Soluble, dye-labelled substrates

The two general types of dye-labelled substrates available for the assay of cellulase andβ-glucanase are soluble dye-labelled substrates and insoluble dyed substrates. Thesoluble dyed substrates include azo-CM-cellulose (azo-CMC) and azo-barley glucan(McCleary and Shameer, 1987). These substrates find widespread application in theassay of β-glucanase in feeds (Rotter et al., 1990; Cosson et al., 1999). In Fig. 4.2, astandard curve relating enzyme activity to colour released on hydrolysis of azo-CMC(CMC-4M dyed with Remazolbrilliant Blue R dye) is shown. The curves obtainedfor the two unfractionated commercial enzyme preparations and for a purifiedcellulase from a Trichoderma sp. are relatively similar. Differences can be attributed

88 B.V. McCleary




to the range of activities in the crude preparations, as well as to subtle differencesin the action patterns of the individual endo-cellulase enzyme components. Thisprocedure has been adopted by the UK silage industry for the standardizationof cellulase activity in silage additive preparations. The major limitation in theapplication of this procedure to measurement of cellulase activity in animal feedpreparations is the lack of sensitivity. A second concern is the fact that the basesubstrate is CM-cellulose rather than β-glucan, which is the target substrate in animalfeeds. The action of pure Trichoderma sp. cellulase on azo-barley glucan (barleyglucan dyed with Remazolbrilliant Blue R dye) is shown in Fig. 4.3. The sensitivitywith azo-barley glucan is about three times that with azo-CMC, but unfortunately isstill not adequate for the measurement of trace levels of cellulase in animal feeds.However, it has been found that action on azo-barley glucan is a good predictor ofin vivo response to cellulase supplementation of barley-based diets in young chickens(Rotter et al., 1990).

Insoluble dye-labelled substrates

Insoluble dyed substrates are prepared by dyeing and cross-linking soluble polysac-charides to give dyed cross-linked polysaccharide gel matrices. Such substrates arereadily hydrolysed and include AZCL-HE-cellulose (Cellazyme C tablets), AZCL-β-glucan (Beta-Glucazyme tablets) and AZCL-xyloglucan (Cellazyme T tablets).Assays based on the use of these substrates are three to ten times more sensitivethan assays using azo-barley glucan or azo-CMC. In Fig. 4.4, standard curves for



Fig. 4.2. Standard curves for the action of cellulase preparations onazo-CMC (lot 60601): (A) a crude Trichoderma viride preparation; (B) a crudeT. longibrachiatium preparation; (C) a pure cellulase from a T. longibrachiatiumpreparation.



90 B.V. McCleary


Fig. 4.3. Standard curve relating the activity of a pure T. longibrachiatiumcellulase on CMC-4M (Nelson/Somogyi assay) to action on: (A) azo-barley glucan(lot 60602) (using EGME/zinc acetate-based precipitant) and (B) azo-CMC (lot60601) (using ethanol/zinc acetate-based precipitant).

1.2

1.1

1.0

0.9

0.8

0.70.6

0.5

0.4

0.3

0.2

0.1

00 10 20 30 40 50 60 70 80 90 100

Cellulase, milliUnits/assay (i.e. per 0.5 ml)on barley β-glucan

A

B

C

Abs

orba

nce,

590

nm

Fig. 4.4. Standard curves relating the activity of a pure T. longibrachiatiumcellulase on barley β-glucan (Nelson/Somogyi assay) to activity on: (A) CellazymeT tablets (lot 70801); (B) Beta-Glucazyme tablets (lot 50901); and (C) Cellazyme Ctablets (lot 80201).



Trichoderma sp. cellulase on dyed cross-linked HE-cellulose (Cellazyme C), dyedcross-linked β-glucan (Beta-Glucazyme) and dyed cross-linked tamarind xyloglucan(Cellazyme T tablets) are compared. It is evident that, of the substrates compared, theCellazyme T tablets give the greatest sensitivity, and thus are the substrate of choicefor the detection and measurement of trace levels of cellulase in animal feed. Thereason for this greater sensitivity must be due to the regular distribution of xyloseresidues along the cellulose backbone of the polysaccharide, compared with theirregular to random distribution of hydroxypropyl residues in HE-cellulose.

Measurement of cellulase and β-glucanase enzymes in feed is complicated byseveral factors, including inactivation of the enzyme during feed pelleting, adsorptionof the enzymes to feed components, competition with endogenous substrate andpossible enzyme inhibitors. These same problems are experienced in the assay ofxylanase and protease enzymes in feed.

Endo-1,4-b-D-Xylanase (Xylanase)

The major endosperm cell-wall polysaccharide in wheat and rye grain isarabinoxylan, which represents about 2–5% of the grain weight. About two-thirds ofwheat-flour arabinoxylan is water insoluble (Amado and Neukom, 1985). However,both the soluble and the insoluble components have high water-absorbing properties.Both can absorb about ten times their weight of water (Kulp, 1968). Consequently,arabinoxylans in feed cause problems similar to those experienced with β-glucans(Annison and Choct, 1991). The development and application of a range of xylanaseenzymes have resolved these problems. Measurement of xylanase in feeds iscomplicated by the fact that the levels added are very low, and this enzyme canbe lost through inactivation during pelleting, adsorption to feed components orinactivation by specific xylanase inhibitors (Debyser et al., 1999).

Several assay procedures have been developed for the assay of xylanase, and theseinclude reducing-sugar assays, viscometric assays, and assays based on the use ofchromogenic substrates (Table 4.2).

Reducing-sugar and viscometric assays

Pure wheat arabinoxylan forms the basis of both reducing-sugar and viscometricassays for xylanase. Reducing sugar assays are not specific and cannot be used for themeasurement of xylanase in trace levels in feed mixtures. However, the Nelson/Somogyi reducing-sugar procedure (Somogyi, 1960) is the method of choice for thestandardization of pure enzyme preparations. These enzyme preparations can then beused to standardize other assay procedures that use less defined substrates (e.g. thedyed xylan and arabinoxylan substrates) and assay procedures that are expressedin non-conventional units (e.g. viscometric assays). Viscometric assays are usedin some laboratories, and the substrate generally employed is wheat arabinoxylan(20–30 cSt). Other xylan substrates (from oat and birchwood) are available, but the





viscosity of commercially available preparations is so low that they are of little use assubstrates in viscometric assays. While viscometric assays allow specific measurementof endo-xylanase, the assay procedure is very tedious. With manual viscometricequipment (type C U-tube viscometer) a single operator can assay just five samples in1 day. This compares with about five to ten samples by a single operator in 2 h withthe newer chromogenic substrates.

Soluble chromogenic substrates

Various dye-labelled substrates have been prepared for the assay of endo-1,4-β-D-xylanase (xylanase), including dyed oat, birchwood and beechwood xylans anddyed wheat arabinoxylan (Table 4.2). In most cases, the dye used is RemazolbrilliantBlue. The reason for this is that it is possible to get good colour intensity with thisdye; and because the dye is hydrophobic, it is easy to separate reaction productsfrom unhydrolysed substrate, by alcohol precipitation. In the preparation ofsoluble dyed substrates for the measurement of xylanase there are several importantconsiderations, including: (i) the final substrate must be readily hydrolysed by theenzyme, with good sensitivity and a good standard curve covering approximately oneabsorbance unit (and preferably linear); (ii) the substrate must be homogeneous andpreferably completely soluble; (iii) the substrate must be specific for the measurementof xylanase, i.e. the base polysaccharide should have a xylan backbone and not becontaminated by other polysaccharides; and (iv) the substrate must form the base of asimple assay procedure.

The relative sensitivities of three soluble, chromogenic xylan substrates arecompared in Fig. 4.5. Of the three, azo-wheat arabinoxylan is the most sensitive, andas such is the substrate of choice in the assay of trace levels of xylanase in animal feeds.This substrate is completely soluble and gives a good (essentially linear) standardcurve over one absorbance unit.

In the measurement of xylanase in feeds, a further set of problems is experienced.The levels of enzyme added to the feed are usually low (just enough to depolymerizethe endogenous arabinoxylan), which causes analytical problems, especially sincesome of this activity is lost due to pelleting or adsorption to feed components and

92 B.V. McCleary


Substrate Nature Assay procedure

Wheat arabinoxylanWheat arabinoxylanXylazyme AX tabletsAzo-wheat arabinoxylanAzo-xylan (oat spelts)Azo-xylan (birchwood)

SolubleSolubleGel particlesSolubleSolubleSoluble

Reducing-sugarViscometricChromogenic substrateChromogenic substrateChromogenic substrateChromogenic substrate

Table 4.2. Substrates for the assay of β-xylanase.



some is inhibited by endogenous xylanase inhibitors. Assays are also complicatedby the fact that the feed material contains arabinoxylan that acts as a competingsubstrate. Consequently, the analyst is left with a choice of either determining thedirectly measurable activity (i.e. a simple assay on feed extracts) or trying to estimatethe level of active enzyme in the feed. In the latter case, it is essential to get someestimate of the degree of adsorption and inhibition of the enzyme. This can beachieved in one of two ways. In the first and preferable approach, if a sample ofessentially identical non-enzyme-treated feed is available, a standard curve can beconstructed by adding increasing quantities of enzyme to feed slurries to give finalanalytical values that cover the range of the activity extracted from the feed. Then,by reference to this standard curve, it is possible to calculate the actual amountof enzyme in the feed. Alternatively, and perhaps more realistically, recoveryexperiments can be performed. In this case, samples of non-enzyme-treated feedare not available. Instead, a sample of the feed is extracted with buffer, or withbuffer to which set amounts of enzyme have been added. From the recovery of theadded enzyme, it is possible to calculate the actual amount of enzyme in the originalfeed. Since this format can be used to estimate any enzyme in a feed mixture, it isdescribed in detail in Appendix 4.3.



Fig. 4.5. Standard curves relating the activity of a pure A. niger xylanase onwheat arabinoxylan (Nelson/Somogyi assay) to action on: (A) azo-wheatarabinoxylan (lot 50201); (B) azo-xylan oat (lot 60701); and (C) azo-xylanbirchwood (lot 30702).



Insoluble chromogenic substrates

In an attempt to improve the sensitivity of chromogenic substrates for the assay ofendo-xylanase, insoluble dyed substrates have been prepared by dyeing andcross-linking soluble xylan substrates, such as oat and birchwood xylans and wheatarabinoxylan (McCleary, 1995). Of these, the preferred starting material is wheatarabinoxylan, because it has a high molecular weight (which is preferable forcross-linking) and stable gel formation. Dyed, cross-linked arabinoxylan (AZCL-arabinoxylan) is available in tablet form (Xylazyme AX) (McCleary, 1992).

Assays for the measurement of xylanase using Xylazyme AX tablets areapproximately three to four times more sensitive than assays employing azo-wheat arabinoxylan (Fig. 4.6). Thus, where enzyme activity is present in very lowconcentrations, Xylazyme AX tablets are the preferred substrate. However, in the useof these tablets, some problems are experienced that are not experienced in the use ofazo-wheat arabinoxylan. The tablet substrate (AZCL-arabinoxylan) is quite sensitiveto salt concentration. High salt concentrations affect the swelling of the gel particleswhich results in the substrate being less susceptible to hydrolysis. The effect of saltconcentration on the rate of release of coloured products from Xylazyme AX tabletsis shown in Fig. 4.7. To minimize this effect and maximize the sensitivity of thesubstrate, feed samples can be extracted with 0.1 M acetic acid (instead of 0.1 Msodium acetate buffer). The final pH in the extract must be in line with the pHactivity profile of the enzyme under evaluation. The major limitation with this

94 B.V. McCleary


Fig. 4.6. Standard curve relating the activity of pure A. niger xylanase on wheatarabinoxylan (Nelson/Somogyi assay) to action on Xylazyme AX tablets (lot 40602).



approach is that it is not possible to assay the pure enzyme in the extracting solution(0.1 M acetic acid), as this will inactivate the enzyme.

On the basis of the above findings, the preferred substrate for the measurementof xylanase in animal feeds is azo-wheat arabinoxylan. However, if extra sensitivity isessential, then Xylazyme AX tablets should be used.

Enzyme-linked sorbent assays

An enzyme-linked sorbent assay (ELSA) has been developed for the assay ofβ-glucanase and β-xylanase in animal feeds (Wong et al., 1999). This assay offers theadvantage of greater sensitivity than any of the procedures previously described.Adoption of the method by industry will depend on numerous factors, includingassay reproducibility, accuracy and reliability, which remain to be ascertained.

a-Amylase

Several methods are available for the measurement of α-amylase in pure enzymepreparations and in cereal flours or animal feeds. Procedures for assaying pureenzyme preparations usually employ soluble starch as substrate and measure theincrease in reducing-sugar levels by either the DNSA (Bailey, 1988) or the Nelson/Somogyi (Somogyi, 1960) method. Traditional methods for the assay of α-amylasein cereal flours are based on the reaction of iodine with starch, or particularly theβ-limit dextrin of starch (Sandsted et al., 1939; Farrand, 1964; European BrewingConvention, 1987). Hydrolysis of the starch results in a decrease in the purple-bluecolour formed on reaction with iodine.



Fig. 4.7. Effect of buffer salt concentration on the activity of A. niger xylanase onXylazyme AX tablets (lot 40602).



Several newer and simpler procedures for the assay of α-amylase are nowavailable. Of these, the preferred method employs ‘end-blocked’ p-nitrophenylmaltoheptaoside as substrate (in the presence of a thermostable α-glucosidase). Theprinciple of this assay is shown in Fig. 4.8 (McCleary and Sheehan, 1987). Thesubstrate is specific for endo-acting α-amylase. On hydrolysis of the oligosaccharidechain by α-amylase, the thermostable α-glucosidase immediately hydrolyses thenitrophenyl-oligosaccharide fragment to glucose and p-nitrophenol. The thermo-stable α-glucosidase (in place of an amyloglucosidase/yeast maltase mixture used inprevious formulations) allows the reagent to be used in the pH range of 5.2–7.5, andat temperatures up to 60°C. The major advantage of this method over other methodsis that the substrate is a defined oligosaccharide. With all other methods, thesubstrate is starch or modified starch, and thus is ill defined.

In situations where the levels of α-amylase are very low, it may be necessary toemploy a more sensitive substrate such as Amylazyme tablets (McCleary, 1991). Thissubstrate is prepared by dyeing and cross-linking amylose to give gel particles that aredehydrated and incorporated into tablets. On hydrolysis by α-amylase, soluble dyedamylose fragments are released into solution and the non-hydrolysed material isremoved by filtration. Assays based on this substrate can be performed directly onfeed slurries, allowing an increase in sensitivity of at least tenfold. These assays are

96 B.V. McCleary


Fig. 4.8. Schematic representation of the assay of α-amylase with Amylase HRreagent (Ceralpha method).



standardized in international units (U) of activity against either the Nelson/Somogyireducing-sugar assay or the Ceralpha assay.

a-Galactosidase

α-Galactosidase enzymes act on galactosyl-sucrose oligosaccharides (raffinose,stachyose and verbascose) producing galactose and sucrose, which removes theanti-nutritional properties of these oligosaccharides (Gitzelmann and Auricchio,1965; Liebmann, 1991). To be effective, these enzymes must rapidly hydrolysegalactosyl-sucrose oligosaccharides, and must be able to work under conditions oflimited water availability.

α-Galactosidase is readily assayed with p-nitrophenyl α-D-galactoside (10 mM)as substrate. However, for a particular α-galactosidase, the relative rates of hydrolysisof p-nitrophenyl α-D-galactoside, raffinose, stachyose and verbascose should bedetermined to ensure that the enzyme being used is actually active on the componentof interest. Activity of α-galactosidase on galactosyl-sucrose oligosaccharides can bedetermined using the particular oligosaccharide (10 mg ml−1) and the Nelson/Somogyi reducing-sugar assay (Somogyi, 1960).

Phytase

Phytase is successfully used as a feed additive to improve phosphorus availabilityin feed. This enzyme is found in plant material and is produced by many micro-organisms, especially moulds of the Aspergillus type. Commercially availablemicrobial phytases are non-specific phosphomonoesterases (EC 3.1.3.8) belongingto the group of acid phosphatases (Jongbloed et al., 1993). These enzymes catalysedephosphorylation of myo-inositol hexakisphosphates in a step-wise manner,producing five classes of intermediate products (myo-inositol pentakis-, tetrakis-,tris-, bis- and monophosphates) of variable stereochemistry (Frolich et al., 1986). Aschematic illustration of dephosphorylation of phytate is shown in Fig. 4.9.

Phytase activity in industrial enzyme preparations and animal feeds is usuallyassayed by measuring the amount of orthophosphates released from phytic acid



Fig. 4.9. Schematic representation of the hydrolysis of phytic acid by phytase.

107

within a period of linear release with time. One unit of phytase activity is defined asthe amount of enzyme required to release 1 µmol of orthophosphate from phytic acidunder standard assay conditions (pH 5.5, 37°C) in 1 min. However, for generalacceptance of phytase enzyme as a feed additive, there is a need for validation of thistype of assay. To satisfy this need, a major international evaluation of an assay formathas been performed under the auspices of AOAC International. The results of thesestudies are currently under evaluation (P. Randsdorp, personal communication,1999).

endo-Protease

Several substrates and assay procedures are available for the measurement of protease.Many of those based on the use of native proteins such as casein, albumin andhaemoglobin do not distinguish between endo- and exo-protease (peptidase) activity.Assays based on trichloroacetic acid (TCA) precipitation of non-hydrolysed substrateand measurement of the absorption (235 nm) of the supernatant solution oncentrifugation (Kunitz, 1947; AACC, 1985) are likely to be more specific thanprocedures that measure the release of free amino groups (Lin et al., 1969). However,even TCA precipitation methods are not specific. A greater specificity can beobtained with dye-labelled proteins (Charney and Tomarelli, 1947), or dye-labelledand cross-linked proteins. Action of exo-acting peptidases on these substrates will behindered or stopped by the dye molecule, or at the point of cross-linking.

Several dyed protein substrates are commercially available, including albumin,casein and collagen dyed with sulphanilic acid (azo-albumin, azo-casein and azo-collagen), and collagen dyed with Remazolbrilliant Blue (Hide powder azure). Theseare useful substrates and form the basis of simple assay procedures, but the quality ofcommercially available substrates varies significantly, limiting their value as analyticalreagents. Standard curves for Subtilisin A on two commercially available azo-caseinsubstrates are shown in Fig. 4.10 and regression equations for several proteases onone of the substrates is shown in Table 4.3 (McCleary and Monaghan, 1999). It isevident that one of the preparations (ex. Megazyme) is more useful than the other,with better sensitivity and linearity of the standard curve. This preparation is usefulin the assay of industrial enzyme preparations, but problems are experienced in themeasurement of enzymes in feed. Whether this is due to binding of the enzyme to thefeed, inhibition by endogenous protease inhibitors or competition by other proteinsubstrates is not clear.

An alternative endo-protease substrate, and one that finds considerable use, iscowhide azure (basically collagen dyed with Remazolbrilliant Blue). This substrateis very fibrous and difficult to weigh accurately. Consequently, an alternativecollagen-based substrate was prepared by dyeing and cross-linking collagen from theswim bladders of fish. The structure of this collagen is very similar to that in cowhide.The major difference in the final product is that the new product from fish collagen(AZCL-collagen) can be milled to a fine powder and incorporated into tablets(Protazyme OL). The suitability of this material as a general or selective substrate was

98 B.V. McCleary




determined by comparing the activity of a wide range of proteases on this substrateto a similar material prepared from casein (AZCL-casein; Protazyme AK). Theconclusion derived from this study was that a similar relative rate of hydrolysis of thetwo substrates was obtained for every protease tested. Consequently, AZCL-collagenhad no particular advantage over AZCL-casein, and since the standard curvesobtained with AZCL-casein (Protazyme AK tablets) were consistently more linear,we consider that this is the substrate of choice.

Assays with Protazyme AK substrate are four to five times more sensitivethan those with Azo-casein, and the standard curves obtained for all of the proteasesstudied are linear (Table 4.4). However, the substrate tablets do not hydrate asrapidly as do tablets containing dyed and cross-linked polysaccharide substrates.Thus, for good reproducibility, the tube contents should be stirred during the



Fig. 4.10. Standard curve relating the activity of Subtilisin A on casein to actionon two commercially available azo-casein preparations: (A) Megazyme (lot 81001);(B) Sigma Chemical Co. (lot 74H7165).

EnzymeProtease (mU ml−1)a,b

(R = 0.99)Linear

absorbance range

Papain (from Papaya latex)Bromelain (from pineapple stem)Ficin (from figs)Subtilisin A (from Bacillus licheniformis)Bacterial protease (from Bacillus subtilis)Proteinase K (from Tritirachium album)Fungal protease (A. niger; from Sigma Chemical Co.)

270 × Abs. (440 nm) + 7460 × Abs. (440 nm) − 13190 × Abs. (440 nm) + 3129 × Abs. (440 nm) + 4250 × Abs. (440 nm) − 8140 × Abs. (440 nm) − 4146 × Abs. (440 nm) − 4

0.1–1.00.1–0.90.1–1.10.1–1.00.1–1.00.1–1.00.1–1.0

aOne protease unit is defined as the amount of enzyme that will produce the equivalant of 1 µmoltyrosine min−1 from soluble casein at pH 7.0 and at 40°C.bAbs. = absorbance.

Table 4.3. Action of protease enzymes on azo-casein (lot 81001).



reaction period. This can easily be achieved but it does make this assay a bit morecomplicated than that using Azo-casein.

Even with this greater sensitivity, it is still difficult to assay protease activity infeed mixtures accurately. Obviously, this area of analyses requires further input. Therequired sensitivity may be obtained with fluorimetric substrates but, even then, theapparent problems of adsorption of the enzyme to the feed and/or specific proteaseinhibitors will still need to be resolved.

Appendix 4.1 Modified Nelson/Somogyi (Somogyi, 1960)Reducing-sugar Method for Measurement of b-Glucanaseand Xylanase Activity

Nelson/Somogyi reagents

1. Reagent A. Dissolve 25 g of anhydrous sodium carbonate, 25 g of sodiumpotassium tartrate and 200 g of sodium sulphate in 800 ml of water and dilute to 1 l.Filter if necessary.2. Reagent B. Dissolve 30 g of copper sulphate pentahydrate in 200 ml of watercontaining four drops of concentrated sulphuric acid.3. Solution C. Dissolve 50 g of ammonium molybdate in 900 ml of water andcarefully add 42 ml of concentrated sulphuric acid; separately, dissolve 6 g of sodiumarsenate heptahydrate in 50 ml of water and add this to the above solution. Dilute thewhole to 1 l. Warm to 55°C to dissolve completely (if necessary).4. Solution D. Add 1 ml of reagent B to 25 ml of reagent A.5. Solution E. Dilute solution C fivefold (50 ml to 250 ml) with distilled waterbefore use (stable at 4°C for about 1 week).

100 B.V. McCleary


Enzyme Protease (mU ml−1)a,b (R = 0.99)

Papain (from Papaya latex)Bromelain (from pineapple stem)Ficin (from figs)Subtilisin A (from Bacillus licheniformis)Fungal protease (A. niger; from Quest International)

76.2 × Abs. (590 nm) + 7.6304 × Abs. (590 nm) + 10132 × Abs. (590 nm) + 3.534.2 × Abs. (590 nm) + 0.642.0 × Abs. (590 nm) + 2.9

aOne protease unit is defined as the amount of enzyme that will produce the equivalant of 1 µmoltyrosine min−1 from soluble casein at pH 7.0 and at 40°C.bAbs. = absorbance.

Table 4.4. Action of protease enzymes on Protazyme AK tablets (regression equations).



Substrates

Dissolve 1 g of pure barley β-glucan (20–30 cSt viscosity), CM-cellulose 4 M orwheat arabinoxylan (20–30 cSt) in 90 ml of hot deionized water. Cool to roomtemperature and add 5 ml of 2 M sodium acetate buffer (pH 4.6). Adjust the pH to4.6 with 1 M sodium hydroxide or hydrochloric acid, and adjust the volume to100 ml. Store in a well-sealed bottle at room temperature. Add two drops of tolueneto prevent microbial contamination. If pH values other than pH 4.6 are required,use an appropriate buffer to obtain the correct pH at a final concentration in thesubstrate of 100 mM.

Procedure

1. Enzyme solution (0.2 ml) is incubated with pre-equilibrated substrate solution(0.5 ml) at 40°C for 0, 3, 6, 9 and 12 min. The reaction is terminated by adding 0.5 mlof solution D with vigorous stirring.2. These tubes, along with reaction blank solutions (to which solution D is addedbefore the enzyme) and glucose or xylose standard solutions (which contain substrateand glucose or xylose at 0–50 µg per tube) are incubated in a boiling-water bath for20 min.3. The tubes are cooled to room temperature and 3.0 ml of solution E is added. Thetubes are stirred well (10 s on vortex mixer) and allowed to stand for 10 min beforemixing again. Centrifuge (3000 rpm for 10 min) if necessary.4. The absorbance of all solutions is measured at 520 nm against the reaction blank.A standard curve is prepared using appropriate amounts of glucose or xylose(0–50 µg).5. Enzyme activity is calculated as µmol of glucose (or xylose) reducing-sugarequivalents released per minute ml−1 or g−1 of the original enzyme preparation.

Appendix 4.2 Measurement of b-Glucanase and XylanaseActivity by Viscometry

This method describes a viscometric assay for the measurement of endo-xylanase orβ-glucanase in microbial preparations, animal feeds and bread improver mixtures.

Principle

The activity is measured by incubating suitably diluted enzyme extract with asolution of β-glucan or wheat-flour arabinoxylan (1% w/v, 20–30 cSt) in appropriatebuffer (sodium acetate or sodium phosphate, 100 mM) and measuring the rate ofdecrease in viscosity at 40°C.





Unit definition

One inverse reciprocal viscosity unit (IRVU) is the increase in reciprocal viscosityh−1 ml−1 (or g−1) of enzyme (or plant material), under standard assay conditions.

Preparation of b-glucan or arabinoxylan substrate

1. Barley β-glucan (1 g, pure, 20–30 cSt) or wheat arabinoxylan (1 g, pure,20–30 cSt) is accurately weighed into a 120 ml dry pyrex beaker.2. The sample is wetted with 6 ml of 95% ethanol and then with 80 ml of coldwater.3. A magnetic stirrer bar is added to the beaker and the beaker is placed on amagnetic stirrer-hotplate and heated at a setting of 120°C with vigorous stirring. Thebeaker is loosely covered with aluminium foil and stirred and heated for about 15 min.4. The polysaccharide should completely dissolve. If dissolution is not complete,continue stirring for a further 30 min with the heating turned off.5. Add 10 ml of acetate buffer (1 M, pH 4.6) or 20 ml of phosphate buffer (0.5 M,pH 6.0) and adjust the pH to the desired value.6. Adjust the volume to 100 ml and store in a well-sealed glass bottle. Preventmicrobial contamination by adding a few drops of toluene. The solution can be storedat room temperature for several weeks.

Enzyme extraction and dilution

1. Plant samples are milled to pass through a 0.5 mm sieve on a Retsch centrifugalmill (or similar).2. Accurately weigh 1.0 g of flour or powdered enzyme preparation into a 250 mlErlemeyer flask and add 100 ml of appropriate extraction buffer; stir and extract over15 min at room temperature.3. Filter an aliquot of the slurry through a Whatman No. 1 filter circle, or centrifugeat 3000 rpm for 10 min. Store the filtrate in an ice bath.4. Dilute the filtrate (or supernatant) with the appropriate buffer (by sequentialdilution of 1 ml to 10 ml with dilution buffer) to give an enzyme concentrationsuitable for assay (i.e. a slope value ‘A’ of 0.06 to 0.60).5. Prepare liquid enzyme samples by dilution of 1.0 ml to 100 ml with appropriatebuffer. This solution is further diluted as for extracts of dry samples.

Viscometric assay of activity

1. Pre-equilibrate the enzyme preparation at 40°C for 5 min and then pipette 1 mlof this solution into 12 ml of pre-equilibrated, buffered substrate solution (wheat

102 B.V. McCleary




arabinoxylan or β-glucan; 1% w/v) in a type C U-tube viscometer (in a water bath at40°C), and mix the contents by blowing air into the viscometer tube. Immediatelystart a stop clock and leave this running throughout the entire assay to recordincubation time (in minutes).2. Using a second stop clock, take five falling time readings (in seconds) over a periodof approximately 30 min. Take the time for each reading as the elapsed time frommixing the enzyme/substrate solutions to the mean of the falling time.3. The viscosity (n) of the reaction digest is proportional to the falling numberaccording to the following equation:

ndigest = (tdigest − tsolvent)/tsolvent

where ndigest = specific viscosity of the digest; tdigest = falling time in seconds of thedigest; and tsolvent = falling time in seconds of the buffer (100 mM sodium acetate).

Calculations

The reciprocal viscosity (1/n) (Table 4.5) is calculated and plotted against incubationtime (in minutes) (Fig. 4.11). The slope (A) is determined from the linear graph interms of increase in reciprocal viscosity per hour.

IRV units = A × 100 × Dilution

where A = slope from graph in terms of increase in reciprocal viscosity per hour;100 = 1 g of original enzyme preparation extracted with 100 ml of buffer (100 mM),or 1 ml of liquid enzyme preparation diluted to 100 ml with extraction/dilutionbuffer; and Dilution = further dilution of the extract or diluted liquid enzymeconcentrate required to get an appropriate activity for assay.

Appendix 4.3 Extraction and Assay of Xylanase Enzymes inAnimal Feeds

Example: Trichoderma sp. xylanase



Incubation time (t) (min) Time to flow (s) Specific viscosity n = (t − to)/to 1/n

05

10141721

302.75247.12211.34190.76174.47162.45

11.759.417.907.046.355.84

0.0850.1060.1270.1420.1570.171

Table 4.5. Determination of reciprocal viscosity values.



Extraction

1. Mill a feed sample (approximately 100 g) to pass a 0.5 mm screen and mixthoroughly.2. Weigh samples of the above feed (0.5 ± 0.01 g in quadruplicate) into glass testtubes (16 × 120 mm).3. Treat each sample with 5 ml of 0.1 M sodium acetate buffer (pH 4.7) and stir on avortex mixer. To two of these tubes, add water (0.2 ml) with stirring, and to the othertwo tubes add control Trichoderma sp. xylanase (0.2 ml, 1360 mU) with vigorous andimmediate stirring on a vortex mixer.4. Leave the slurries at room temperature with occasional stirring on a vortex mixerover the following 20 min.5. Centrifuge tubes (3000 rpm, 10 min) in a bench centrifuge and use thesupernatant directly. Assays should be initiated within 30 min of obtaining theseextracts to minimize loss of enzyme activity.

Assay

1. Accurately transfer 0.5 ml aliquots of supernatant solutions (in duplicate) to glasstest tubes (16 × 100 mm) at room temperature.2. Add 0.5 ml of Azo-WAX to each tube and stir the tube vigorously. Immediatelyplace the tube in a water bath set at 50°C ± 0.1°C and incubate for 30 min.3. After exactly 15 or 30 min (depending on the level of enzyme activity), add 2.5 mlof industrial methylated spirits (IMS) and stir the tube vigorously on a vortex mixer.Store the tube at room temperature for 5 min. This treatment terminates the reactionand precipitates non-depolymerized dyed substrate.4. Centrifuge the tubes at 3000 rpm for 10 min and measure the absorbance of thesupernatant solutions at 590 nm against a reaction blank.

104 B.V. McCleary


Fig. 4.11. Plot of reciprocal viscosity (1/n) against incubation time.

114

The reaction blank is prepared by adding 2.5 ml of IMS to a mixture of 0.5 ml ofazo-wheat arabinoxylan and 0.5 ml of 0.1 M sodium acetate buffer (pH 4.6). Theslurry is stirred and stored at room temperature for 5 min before centrifugation(3000 rpm, 10 min). A single reaction blank is required for each feed sample.

Calculation of activity

The level of xylanase in the flour sample is calculated as follows.

Activity in feed sample (0.5 g) = Added activity × SA / (TA − SA)

where Added activity is the amount of xylanase added to the feed slurry at the time ofassay (in the control xylanase solution; 0.2 ml) (e.g. 1360 mU); SA is the reactionabsorbance obtained for extracts of the feed to which no control xylanase was added;TA is the total absorbance (i.e. the absorbance of extracts of the sample to which thecontrol xylanase was added).

Example calculation

SampleAbsorbance (590 nm)/30 min incubation

1. Feed A2. Feed A containing Trichoderma sp. xylanase (SA)3. SA + 1360 mU xylanase (in the assay) (TA)

0.0000.5020.908

Activity in 0.5 g of feed A = Added activity × SA / (TA − SA)

where SA = absorbance of extract of sample A assayed by the standard format (e.g.Abs = 0.502); TA = the total absorbance, i.e. the absorbance of extracts of sample Ato which the additional xylanase (0.2 ml; 1360 mU) was added (e.g. Abs = 0.908).

Thus:

Activity in the feed = (1360 / 1000) U × [0.502 / (0.908 − 0.502)](U per 0.5 g)

= 1.682 U.

U g−1 (or kU ton−1) = 1.682 × 2000 = 3363.

References

Aastrup, S. (1979) The effect of rain on β-glucan content in barley grains. Carlsberg ResearchCommunication 44, 381–393.





Amado, R. and Neukom, H. (1985) Minor constituents of wheat flour: the pentosans. In:Hill, R.D. and Munk, L. (eds) New Approaches to Research on Cereal Carbohydrates.Elsevier Science Publishers BV, Amsterdam, pp. 241–251.

AACC (American Association of Cereal Chemists) (1985) Approved Methods. Method22-60. ‘Proteolytic activity of flour’.

Annison, G. and Choct, M. (1991) Anti-nutritive activities of cereal non-starch polysaccha-rides in broiler diets and strategies minimising their effects. World’s Poultry ScienceJournal 47, 232–242.

Bailey, M.J. (1988) A note on the use of dinitrosalicylic acid for determining the products ofenzymatic reactions. Applied Microbiology and Biotechnology 29, 494–496.

Bathgate, G.N. (1979) The determination of endo-glucanase activity in malt. Journal of theInstitute of Brewing 85, 92–94.

Campbell, G.L. and Bedford, M.R. (1992) Enzyme applications for monogastric feeds: areview. Canadian Journal of Animal Science 72, 449–466.

Chandra, N.B., Hirst, E.L. and Manners, D.J. (1957) A comparison of isolichenan andlichenan from icelandic moss (Cetraria islandica). Journal of the Chemical Society,1951–1958.

Charney, J. and Tomarelli, R.M. (1947) A colorimetric method for the determination ofproteolytic activity of duodenal juice. Journal of Biological Chemistry 171, 501–514.

Cosson, T., Vendrell, A.M.P., Teresa, B.G., Rene, D., Taillade, P. and Brufau, J. (1999)Enzymatic assays for xylanase and β-glucanase feed enzymes. Animal Feed ScienceTechnology 77, 345–353.

Debyser, W., Peumans, W.J., Van Damme, E.J.M. and Delcour, J.A. (1999) Triticumaestivum xylanase inhibitor (TAXI), a new class of enzyme inhibitor affectingbreadmaking performance. Journal of Cereal Science 30, 39–43.

European Brewing Convention, Analytica EBC (fourth edition; 1987). Method 4.12.3.‘Alpha-Amylase (Colorimetric Method)’.

Farrand, E.A. (1964) Flour properties in relation to the modern bread process in the UnitedKingdom with specific reference to alpha-amylase and starch damage. Cereal Chemistry41, 98–111.

Frolich, W., Drakenberg, T. and Asp, N.G. (1986) Enzymatic degradation of phytate(myo-inositol) hexaphosphate in whole grain flour suspensions and dough. A comparisonbetween 31P NMR spectroscopy and a ferric ion method. Journal of Cereal Science 4,325–334.

Gitzelmann, R. and Auricchio, S. (1965) The handling of soya alpha-galactosides by a normaland galactosemic child. Pediatrics 36, 347–387.

Jongbloed, A.W., Kemme, P.A. and Mroz, Z. (1993) The role of phytases in pig production.In: Wenk, C. and Boessinger, M. (eds) Enzymes in Animal Nutrition. Proceedings of the1st Symposium, Switzerland, pp. 173–180.

Kulp, K. (1968) Enzymolysis of flour pentosans. Cereal Chemistry 45, 339–350.Kunitz, M. (1947) Crystalline soybean trypsin inhibitors. II. General properties. Journal of

General Physiology 30, 291–310.Liebmann, B. (1991) Out of gas. Nutrition Action Health Letter, March.Lin, Y., Means, G.E. and Feeney, R.E. (1969) The action of proteolytic enzymes on

N,N-dimethyl proteins. Journal of Biological Chemistry 244, 789–793.McCleary, B.V. (1988) Purification of (1-3)(1-4)-β-D-glucan from barley flour. In: Wood,

W.A. and Kellogg, S.J. (eds) Methods in Enzymology. Academic Press, New York, 160,511–514.

106 B.V. McCleary




McCleary, B.V. (1991) Measurement of polysaccharide degrading enzymes usingchromogenic and colorimetric substrates. Chemistry in Australia 58, 398–401.

McCleary, B.V. (1992) Measurement of endo-1,4-β-D-xylanase. In: Visser, J., Beldman, G.,Kusters-van Someren, M.A. and Voragen, A.G.J. (eds) Xylans and Xylanases. ElsevierScience Publishers, Amsterdam, pp. 161–170.

McCleary, B.V. (1995) Measurement of polysaccharide degrading enzymes in plants usingchromogenic and colourimetric substrates. In: Skerritt, J.H. and Appels, R. (eds) NewDiagnostics in Crop Sciences. CAB International, Wallingford, UK, pp. 277–301.

McCleary, B.V. and Monaghan, D. (1999) New developments in the measurement ofα-amylase, endo-protease, β-glucanase and β-xylanase. 2nd European Symposium onEnzymes in Grain Processing, in press.

McCleary, B.V. and Shameer, I. (1987) Assay of malt β-glucanase using Azo Barley Glucan:an improved precipitant. Journal of the Institute of Brewing 93, 87–90.

McCleary, B.V. and Sheehan, H. (1987) Measurement of cereal α-amylase: a new assayprocedure. Journal of Cereal Science 6, 237–251.

Rotter, B.A., Marquardt, R.R., Guenter, W. and Crow, G.H. (1990) Evaluation of threeenzymic methods as predictors of in vivo response to enzyme supplementation ofbarley-based diets when fed to young chicks. Journal of the Science of Food and Agriculture50, 19–27.

Sandsted, R.M., Kneen, E. and Blish, M.J.A. (1939) A standardised Wohlgemuth procedurefor alpha-amylase activity. Cereal Chemistry 16, 712–723.

Somogyi, M. (1960) Modifications of two methods for the assay of amylase. Clinical Chemistry6, 23–35.

Walsh, G.A., Power, R.F. and Headon, D.R. (1993) Enzymes in the animal-feed industry.TIBTECH 11, 424–430.

Walsh, G.A., Murphy, R.A., Killeen, G.F., Headon, D.R. and Power, F.P. (1995) Detectionand quantification of supplemental fungal β-glucanase activity in animal feed. Journal ofAnimal Science 73, 1074–1076.

Wong, G.J., Marquardt, R.R., Xiao, H. and Zhang, Z.Q. (1999) Development of a 96-wellenzyme-linked solid-phase assay for beta-glucanase and xylanase. Journal of Agriculturaland Food Chemistry 47, 1262–1267.





J.D. SummersDigestibility and Feed Value of Maize5

Introduction

On a world basis, maize is the most important cereal in livestock feeding. Having ahigh available energy content and a relatively low protein content, it has usually beenlooked on mainly as an energy source in a diet. However, since the levels used inpoultry and swine diets can approach 70%, many diets have in excess of 20% of theirprotein content contributed by maize. Maize has always been considered to be avery uniform product with slight variations in protein and energy. Thus, feedcompounders have spent little time monitoring its nutrient composition. There isgrowing evidence today that the feeding value of maize samples can vary significantly,based on animal performance. As there is little data available to suggest the reasonsfor these differences in nutritive value, it is of interest to review briefly some of thework reported on the composition of maize by the milling and food industries.

Composition and Physical Characteristics of Maize

Starch is a unique and omnipresent polysaccharide found in many foodstuffs.Most of the energy in maize is derived from the starch fraction that is found in theendosperm of the kernel. While many consider starch a homogeneous product it canin fact vary considerably in composition. Regular or normal maize (NM) has most ofits starch present as amylopectin, while waxy maize (which is used mainly in the foodindustry) has little amylose present as starch sources. Normal maize, often referred toas dent maize, and waxy maize are the two most important maize varieties. Flintmaize has similar composition to NM, but it has a hard, starchy layer entirelysurrounding the outer part of the kernel. Other types of maize are important in smallspecialized markets: sweet maize (sweetcorn), for example, has a high level of sugarsin its endosperm; pop maize (popcorn), used in the confectionery industry, has ahigh proportion of hard starch which, on heating, expands rapidly, resulting in aexplosive rupture of the epidermis and starch granules. While these varieties differ in



Maize: Factors Affecting itsDigestibility and Variability in

its Feeding ValueJ.D. SUMMERS

Poultry Industry Centre, RR#2, Guelph, Ontario, Canada N1G 6H8

5



their composition compared with NM, there is little information available as to theirexact make-up.

Starch granules can differ significantly in size and composition, depending ontype and variety of cereal (South et al., 1991). Granule size and composition are alsodependent on the age of the cells in the developing endosperm. Three distinct typesof starch granules have been reported which appear to be developmentally regulated.These vary in their content of associated protein (Wasilick et al., 1994).

Not all starch is readily digested. Some starch and products of starch digestionhave been shown not to be readily absorbed in the small intestine of healthy humans.Indeed, large amounts of starch compounds have been found in the large bowelcontents of people who have succumbed to sudden death syndrome (Brown, 1996).Starch has been classified as three types: (i) starch gelatinized by cooking or heating,which is rapidly digested; (ii) native starch granules from many cereals that may beslowly but completely digested; and (iii) starch that is resistant to digestion. Threesubcategories of resistant starch have been identified: (i) granules that are physicallyinaccessible, e.g. trapped in food matrix or in partially milled or whole grain, so thatthe size or composition of food particles prevents or delays the action of digestiveenzymes; (ii) native starch granules where resistance is related to structure andconformation of granules, e.g. susceptibility to gelatinization – the gelatinizationtemperature of some starches is 70°C, while for high amylose maize the completegelatinization temperature has to be around 170°C; and (iii) the formation ofretrograde starch material during processing. Starch composed of amylose andamylopectin can be dispersed in water by heating beyond the gelatinizationtemperature. Upon cooling, the dispersed molecules of amylose and amylopectinspontaneously reassociate and can form crystallites that resist enzymatic hydrolysis.Retrograde and resistant starch can vary significantly with regard to enzymesusceptibility (Eerlingen et al., 1994). This type of starch usually passes undigested tothe lower bowel (Brown, 1996).

Amylose content increases with age and size of granule. Some of the maizemutants differ in amylose concentration and it is this characteristic that is mainlyresponsible for the wide variations seen in some of the responses with these mutants(e.g. susceptibility to enzyme degradation, gelatinization, etc.). Maize varieties veryhigh in amylose are high in resistant starch and fibre and are used in a number ofnovel and innovative foods.

While it is beyond the scope of this article to delve into the physical andbiochemical properties of maize and its constituents, the above short review gives

110 J.D. Summers


Grain Gross energy (kcal g−1) Metabolizable energy (kcal g−1) Percentage available

MaizeWheatBarleyOats

4.544.464.394.64

3.943.523.042.74

86.878.969.259.1

Table 5.1. Energy availability for poultry from cereal grains.



some insight as to how maize quality, and thus its nutritive value, can be altered withdifferent growing conditions, which could alter starch cell structure, and variations indrying conditions, especially time and temperature, which could affect the physicalproperties of the starch granules.

Nutrient availability of maize

Values of apparent metabolizable energy (AME) and true metabolizable energy(TME), determined for the most part by the faecal excretion method in adult roost-ers, have been used as a measure of the available energy in maize. The ratio of avail-able energy to gross energy (AME : GE) is the highest for maize compared with othercommonly used feedstuffs (Table 5.1). This could well be the reason that little efforthas been expended in trying to improve the digestibility of the nutrients in maize.

A report from Indiana (Maier, 1995) shows the type of information usuallyavailable to the purchaser of maize. A number of samples were taken from differentparts of the state and analysed for various nutrients. A wide range in these nutrientscan be noted (Table 5.2). There is little in the nutrient composition data, exceptperhaps kernel density, to give an estimate of the available energy in the samples.

Several reported studies have looked at the composition of various samples ofmaize. Most of these studies have investigated maize samples that, due to the type ofgrowing season, had a significant range in density or weight per bushel. Leeson andSummers (1976) compared maize samples of different moisture and bushel weightsthat were harvested during a wet, cold autumn and thus stage of maturity would beexpected to differ. There was a marked range in bushel weights and also compositionvalues. While bushel weights varied by approximately 40%, ME as determined withadult roosters varied by only 12%. Composition data for three of the highest, lowestand middle range bushel weight samples are shown in Table 5.3. The lower bushelweight samples had a higher level of protein than did the more normal heavier weightsamples. Lilburn and Dale (1989) also reported higher levels of protein with lowerbushel weight maize; however, levels of lysine and methionine were not increased inthe same proportion as was total protein.

Whitacre (1987) had reported that as protein concentration in different maizesamples increased, the relative increase in non-essential amino acids was greater thanthat of the essential amino acids (EAAs). A paper by Lilburn et al. (1991), in whichthey studied variation in the composition of maize harvested after a season of severe

Digestibility and Feed Value of Maize 111


Protein Oil Starch Density

% Range % Range % Range % Range

7.7 5.7–9.7 3.3 2.6–4.9 61.7 59.9–64.8 1.26 1.20–1.31

Table 5.2. Average composition of maize from nine areas of Indiana (values based on 15%moisture). (Selected data from Maier, 1995.)



drought, again demonstrated the lower EAA values in the higher protein maizesamples. In a further paper Leeson et al. (1993) studied 26 different maize samplesproduced in Ontario after a wet, cold growing season that resulted in a significantpercentage of low bushel weight maize. A selected summary of their results is shownin Table 5.4. The composition values were quite variable and for the most part didnot correlate with bushel weight of the maize. Indeed, little difference in energy orprotein content could be traced to the bushel weight of the maize. These workersconcluded, as did Dale (1994) with a similar study, that bushel weight is a poorestimator of the feeding value of maize. It is of interest that the low bushel weightmaize of Dale (1994) had lower protein levels as compared with the higher proteinlevels reported for low bushel weight maize from a crop exposed to droughtconditions or to a wet harvesting season. Hsu and Sell (1995) studied low and highbushel weight maize in feeding trials with young poults. They found little or nodifference in the energy value of the different maize samples. The low bushel weight

112 J.D. Summers


Bushel weight(lb)

Metabolizable energy(kcal kg−1)

Crude protein(%)

Fat(%)

Fibre(%)

Starch(%)

Sugar(%)

575554494948444241

330032323410339532473388305430843051

7.68.98.98.59.18.89.1

10.910.0

3.43.34.04.33.73.53.24.34.3

2.73.23.13.12.22.93.83.22.5

55.358.157.656.957.958.256.355.356.6

5.14.34.34.86.84.36.27.76.6

Table 5.3. Metabolizable energy and composition of maize samples (values expressed on an85% moisture level). (Selected data from Leeson and Summers, 1976.)

Bushel weight(lb)

Metabolizable energy(kcal kg−1)

Crude protein(%)

Crude fat(%)

Soluble fibre(%)

Insoluble fibre(%)

58.555.358.553.153.153.450.250.845.2

327033063275316329263363347130652944

9.08.98.87.57.17.78.17.57.9

3.22.73.32.52.72.93.32.62.3

1.21.40.61.10.40.71.21.20.9

9.110.6

9.28.38.48.7

10.38.78.9

Table 5.4. Bushel weight, metabolizable energy and proximate components of maize. (Selecteddata from Leeson et al., 1993.)



maize in their study also had a lower protein content at 6.63% versus 8.53% formaize of normal bushel weight.

Since maize quality can be reduced by the amount of broken kernels or presenceof foreign matter, Dale and Jackson (1994) took samples from various shipments ofmaize and separated them into whole kernels, broken kernels and foreign matter(which ranged from 2 to 7% in the samples). The broken kernels averaged about2.5% and the foreign matter contained 115 kcal kg−1 less nitrogen corrected truemetabolizable energy (TMEn) than did the whole grain maize.

Potential for Improving Maize Nutritive Value

It is estimated that 60–70% of the maize produced in the world ends up in livestockfeed. Until very recently, seed companies were essentially interested in producingvarieties that increase the yield per hectare. The last number of years have seen ashift in the goals of the plant breeders as they now are considering altering thecomposition of the maize in order to improve its nutritional level.

With gene transfer technology firmly established, the stage is set for a rapidadvancement in nutrient quality improvements of cereal grains. While maize compo-sition and quality, on a dry matter (DM) basis, has always been assumed to be fairlyconstant, marked differences in composition (as shown in previous tables) have beenknown for years. Because maize makes up such a large portion of swine and poultrydiets, minor changes in composition can reflect on animal performance or carcass quality.

With maize being a major contributor of energy to diets, increasing the availableenergy in maize is an avenue being actively pursued. There are two strategies availableto increase energy density of maize. One is to increase the gross energy value ofthe grain and, assuming that the ratio of GE : ME remains constant, an increase inavailable energy will take place. A second way is to increase the percentage of GE thatcan be utilized by the animal.

Oil

The approach that has received the most attention to date is to increase the GE ofmaize by increasing its oil content. This has proved to be successful and there are nowmaize types on the market with oil levels in excess of 8% on a dry matter basis, incontrast to around 3–4% for NM. The oil is found almost exclusively in the germportion of the kernel and thus an increase in germ size brings about a correspondingdecrease in endosperm and starch content. For every 1% increase in oil content,starch decreases by 1.3%. Since the germ is relatively high in protein, it also increasesand will be higher by approximately 0.3% with a 1% increase in oil content. Forevery 0.1% increase in oil content, the available energy content of the maize isestimated to increase by approximately 4.5 kcal kg−1. An increase in the ratio ofME : GE from 85 to 89% would have to see an increase in the oil content of themaize by three percentage points to equal the increased available energy produced.





A survey by Maier and Briggs (1997) also showed that high oil maize (HOM)can vary significantly in composition (Table 5.5). Han et al. (1987) investigated thefeeding value of HOM for poultry in low protein diets in order to achieve dietscontaining high levels of the maize samples being tested. There was no differencein protein quality of the HOM or NM as measured by a net protein ratio test. In atest with broilers to 22 days of age, weight gain was equal while feed efficiency wassignificantly improved when the HOM replaced NM in the basal diet (Table 5.6).They concluded that the magnitude of the response was highly correlated with the oilcontent of the maize. In a 15-week test with laying hens, feed to egg mass output wassignificantly better when HOM replaced regular maize in a 17% protein diet whenthe maize was substituted on an isonitrogenous basis. There were no differences forother parameters measured between the two maize samples (Table 5.6).

Adams et al. (1994) evaluated HOM in a broiler study simulating commercialconditions, where male birds were reared to 42 days of age. The experiment wasdesigned to demonstrate the ability of the HOM to increase the energy level inbroiler diets without going to high levels of added dietary fat. A range of diets con-taining up to a level of 6% added poultry oil were compared with similar maize/soyadiets where the ratio of energy to protein was kept constant. Thus the HOM dietscontained a higher level of energy and protein. Broilers fed the HOM diets were

114 J.D. Summers


Protein Oil Starch Density

% Range % Range % Range % Range

8.7 6.6–10.2 6.9 6.1–7.5 55.6 53.3–57.4 1.26 1.23–1.28

Table 5.5. Average composition of high oil maize from 44 samples produced in Indiana (valuesbased on 15% moisture). (Selected data from Maier and Briggs, 1997.)

Maize type Gain (g) Gain : Feed

RegularHigh oila

471.5478.3

0.6720.687

aHOM added at same level as regular maize.

(b) Layers (23–38 weeks)

TreatmentProduction

(%)Egg weight

(g)Feed : egg

(g g−1)Body weight

gain (g)

1. Regular maize (17% CP)2. As (1), with HOM (18% CP)3. As (2), diet kept isonitrogenous with (1)

83.784.486.8

54.254.154.7

2.212.142.08

123.2183.9180.0

Table 5.6. Comparison of regular maize and high oil maize for broilers and laying hens.(Selected data from Han et al., 1987.)(a) Broilers (8–22 days)



significantly heavier and had an improved feed : gain ratio than birds fed the NMdiets. They also had less abdominal fat than the NM-fed birds (Table 5.7). Theseresults show the possibility of formulating higher density diets with the use of HOM.In a further publication from Arkansas (Saleh et al., 1997) a HOM sample was com-pared with a NM variety. The HOM contained 2.32% more oil, 156 kcal kg−1 moreTMEn and 0.38% more crude protein than the regular maize sample. These were fedin commercial-type diets, with equal levels of protein and energy, under commercialconditions, to 42 days of age. There were no differences in body weight, feed conver-sion or dressing percentage between the various diets. The HOM-fed birds had sig-nificantly less abdominal fat. The authors again pointed out the advantage of beingable to use higher energy diets without reverting to higher levels of added dietary fat.

Bartov and Bar-Zur (1995) compared a HOM variety produced in Israel withlocal NM. The composition of the maize types, HOM versus regular, was: oil6.7–2.8%; protein 9.8–7.2%; AMEn 3658–3437 kcal kg−1, respectively. One wouldexpect to find improved performance of the HOM type based on its composition.However, a broiler feeding trial from 7 to 28 days resulted in no difference in birdperformance when the HOM replaced an equal proportion of NM. Supplementingthe NM diet with protein and fat to equal levels of protein and energy in the HOMdiet resulted in a significant improvement in performance of the birds. The authorshad no ready explanation for the failure of the HOM diets to give superior



% Supplemental oil

Maize type 0 2 4 6 Mean

HOMYDM

21242020

21222056

22572125

22362158

21852090

(b) Feed : gain

% Supplemental oil


HOMYDM

1.711.80

1.691.75

1.611.72

1.571.66

1.651.73

(c) Abdominal fat (% of body weight)

% Supplemental oil


HOMYDM

2.082.26

1.962.48

2.092.35

2.162.24

2.072.33

Table 5.7. Effect of high oil (HOM) vs regular yellow dent maize (YDM) in diets with variouslevels of supplemental oil (male broilers to 42 days). (Selected data from Adams et al., 1994.)(a) Body weight (g)



performance, but they did point out that some of the reported responses to HOMhave been variable.

Vieira et al. (1997) reported on a feeding trial with a HOM variety producedand grown in Brazil. Their HOM contained 5.7% oil compared with 3.8% in theirNM sample. However interesting, unlike most other HOM samples reported, theirsample contained slightly less protein than their NM (9.17% versus 9.39%, DMbasis). Similar to the report from Bartov and Bar-Zur (1995), their data showed thatthe NM sample gave superior performance when supplemental oil was added tomake the diets equal in energy. Rand et al. (1998) reported that under commercialconditions HOM was particularily advantageous in enhancing broiler performanceduring the hot summer months. This was probably due to the more efficient use ofthe higher oil diet.

Parsons et al. (1998) studied the availability of the amino acids in HOM ascompared with an NM sample. Digestibility studies with adult cockerels suggestedthat digestibility for the amino acids in both samples of maize were the same. Using achick feeding study they demonstrated that the bioavailability of the lysine in HOMwas equal to or slightly better than that in NM.

One would have to conclude that most of the HOM varieties on the markettoday are superior in feeding value to NM samples. Their greater nutrient densityand superior nutritive value, along with their advantage in producing higher densitydiets without increased dietary fat inclusion, make them a potential competitor forthe maize market of the future.

Protein

Several maize varieties developed in the 1960s showed promise due to their improvedlevel of protein and the essential amino acids lysine and methionine. Opaque-2 maizehad a higher level of protein and significantly more lysine than NM. When comparedwith NM, growth response of chicks was not as good until the opaque-2 maize dietwas supplemented with methionine to balance the higher level of lysine. Floury-2maize had increased levels of protein as well as lysine and methionine. This maizegave an improvement in performance when compared with NM without EAAsupplementation. Cromwell et al. (1968) compared the feeding value of a sample ofopaque-2 and floury-2 maize with that of NM. These were tested in a 21-day feedingtrial with chicks fed 15% protein diets where equal amounts of maize protein weremaintained by varying the amount of maize added. The results, shown in Table 5.8,demonstrate the superiority of the higher protein maize samples with proper EAAsupplementation. These maize types never did gain favour with producers, due topoor storage problems as well as low yields.

An increase in the proportion of glutelin as compared with zein protein in thekernel is obviously the reason for the changes noted for the altered composition of thehigh protein maize varieties.

Bond et al. (1991) tested two new high protein maize varieties that werereported to be free of the problems that plagued the earlier samples. These varieties

116 J.D. Summers




contained approximately 11% protein as against 9% for the NM that they used forcomparison. In a 6-month test with layers, where the diets were formulated to beisocaloric but not isonitrogenous, production and egg size did not differ; however,feed per kilogram of egg mass produced was significantly lower with one of thenew maize varieties (Table 5.9). In a test with broilers to 6 weeks of age, diets wereformulated with the HPM-1 sample. The diets were formulated to be isonitrogenous



Sulphur amino acids (%)

Maize type 0.45 0.60 0.80 Average

NormalOpaque-2Floury-2Normal + lysineAverage

377299397367351

332389380368368

330382384377367

333357387371

(b) Feed : gain

Sulphur amino acids (%)

Maize type 0.45 0.60 0.80 Average

NormalOpaque-2Floury-2Normal + lysineAverage

1.741.811.651.701.72

1.771.631.661.651.68

1.741.621.651.661.67

1.751.691.651.67

Table 5.8. Effect of adding methionine to normal, opaque-2 and floury-2 maize, and to normalmaize plus lysine, in diets containing 15% protein. (From Cromwell et al., 1968.)(a) 20 day weight (g)

NM HPM-1 HPM-2

Production (%)Egg weight (g)Feed intake (g per bird day−1)Feed : egg mass

69.164.1

109.572.56

70.764.7

106.572.35

68.563.9

109.572.57

(b) Broilers

NM HPM-1

Body-weight gain (g)Feed : gain

1633.002.00

1749.001.96

Table 5.9. Comparison of normal maize and high-protein maize. (Selected data from Bond et al.,1991.)(a) Layers



as well as isocaloric. An improvement in weight gain and feed : gain ratio was notedwith the higher protein maize sample. The authors suggested that higher levels ofmethionine, as well as several other EAA present, could account for the betterperformance of this particular sample.

Waxy maize, while not used to any extent in commercial feeding operations, attimes does find its way into diets for poultry. Waxy maize contains amylopectin as itsonly source of starch, while NM contains approximately 73% amylopectin and 27%amylose as starch sources. Dinn et al. (1982) reported that the feeding value of thewaxy maize was similar to that of NM when fed to broilers. Ertl and Dale (1997)compared the available energy content of normal and waxy maize of similar geneticbackgrounds, and grown under similar conditions, using adult roosters. Their resultsshowed that the TMEn values of the two maize varieties were quite similar, thussuggesting that they could be substituted in diets without any fear of reducedperformance.

Increasing the Available Energy of Maize by ImprovingNutrient Digestibility

The digestibility and availability of amino acids, starch and oil and the presence ofpoorly digested components of a grain determine the fraction of GE which shows upas energy available to an animal. Imbalances of EAA in relation to an animal’s needsobviously also reduce the nutritive value of a cereal. Most of the improved maizevarieties in the past have resulted in increases or decreases in some of the major com-ponents such as protein, starch and oil with little or no change in the composition ofthese components.

Little headway has been achieved in increasing the nutritive value of maize byimproving the digestibility or utilization of its various components. Recent workquestions whether the AME and TMEn values used today accurately predict theenergy derived by the bird, especially the young bird, from maize. Mahagna et al.(1995) reported that the AME of maize differed from the first week post-hatch to thethird week by around 200–250 kcal kg−1. Collins et al. (1998) also reported that ageof bird was an important consideration when assigning an available energy value tomaize. It has been shown that measuring available energy with ileal rather than faecalcontents indicates that starch digestion is not complete in the small intestine. Noyand Sklan (1995) reported low ileal digestibility of both starch and fat in youngbroiler chicks fed a maize/soybean-meal diet. It was shown that digestibility for starchin chicks from 4 to 21 days of age, up to the end of the small intestine, was as lowas 82%, with no evidence of any increase as the birds got older. This suggests that aportion of the maize starch is reaching the hind gut, where it undergoes fermentationwith poor energetic utilization. Microscopic examination of the ileal digesta supportsthis conclusion as large maize endosperm particles have been found to be undigested.Persia and Lilburn (1998), on studying the AME and starch digestibility in the smallintestine of turkey poults, reported that there was a significant difference in AME forwheat and maize when measured with small intestine versus colon contents. Their

118 J.D. Summers




data showed 5.21% of undigested starch entering the colon from maize versus 2.67%from wheat. Pack et al. (1998) reviewed much of the recent work in the area of maizestarch availability and reported that not only is starch not completely digested inthe small intestine but also there can be a significant amount of protein – and inparticular amino acids – entering the hind gut, depending on age of bird and maizesample. It has been reported that amino acid digestibility is significantly lower whenmeasured with ileal as compared with faecal determinations. The reason for thesedifferences is the contribution by the hind gut in digesting starch, which is notreadily available to the animal, and also the error involved in determining amino aciddigestibility values when a correction is made for endogenous losses. There is also asignificant amount of data to indicate that overall digestion is low in the chickto around 10 days and then increases to a plateau around 4 weeks of age. Thusthe values used for available energy for the young chick show significant room forimprovement by way of altering maize composition or the use of specific enzymes.Noy and Sklan (1995) have shown that amylase, trypsin and lipase are low at 4 daysof age and that by 21 days of age they increase by 100-, 50- and 20-fold, respectively.They suggest that only 82–89% of fatty acids and starch are digested in the smallintestine. While nitrogen digestion is also not complete in the small intestine, itincreased from around 78% at 4 days to 92% at 21 days of age.

Enzyme Supplementation

The action of enzymes on various starches and factors influencing starch degradationhas been well documented (Knutson et al., 1982; Planchot et al., 1995). There isalso a wealth of data reported on enhancing the feeding value of the coarse cereals bysupplementing diets with various enzymes. These enzymes degrade the non-starchpolysaccharides present in cereals such as wheat, barley, rye and oats, thussignificantly improving their available energy content. Maize with a very low NSPlevel does not respond to such enzymes. However, in recent years it has been shownthat a mixture of enzymes containing amylase, xylanase and protease is effective inenhancing the nutritive value of a maize/soybean-meal diet. Much of the recent workin this area is covered in the reports of Pack and Bedford (1997) and Pack et al.(1998). These reports, as well as others, suggests a 2–5% improvement in availableenergy with enzyme supplementation of a maize/soybean-meal diet. More uniformbroiler weights at marketing and reduced mortality have also been noted with theenzyme-supplemented diets. These have been attributed to the possible removal ofanti-nutritional factors present in the feed ingredients, along with more effectivenutrient digestibility in the small intestine and hence an influence on gut microbialactivity. It is of interest that Yan et al. (1998) also reported reduced mortality as oneof their consistent observations when phytase was added to a maize/soybean diet forbroilers.

Maize/soybean-meal diets have also been improved in nutritive value by theaddition of phytase enzyme. While there is some information to indicate that theoverall improvement of the diet has been enhanced with phytase supplementation, it





is mainly an improvement in phytate-phosphorus availability that has been noted.Simons et al. (1990) showed an improvement in dietary phosphorus availability forpoultry of approximately 65% with microbial phytase supplementation of diets. VanDer Klis et al. (1997) reported a marked improvement in phytate degradation withlaying hens, from 8 to 50%, with the addition of phytase to the laying diet of the hen(Table 5.10).

Recently, low phytic acid maize varieties have been identified and they holdpromise in markedly enhancing the utilization of the phosphorus from maize as wellas reducing faecal phosphorus excretion. Two papers from Arkansas at a recentpoultry science meeting showed that the total phosphorus in the maize was similar toNM and that the non-phytic acid phosphorus was equal to the availability ofinorganic phosphorus supplements (Kersey et al., 1998; Li et al., 1998; Yan et al.,1998).

Physical Properties of Maize

Another factor to consider in looking at the response to diets containing maize isthe type of grinding to which it is subjected. One of the major ways to improve thenutritive value of cereals for livestock feeding is to grind the material. Grindingincreases the surface area and thus allows for better enzyme action in the gut. Therehas been renewed interest in looking at type and degree of grinding in order toenhance feeding value as well as to reduce energy costs for grinding. Deaton et al.(1995) reported that hammer-milling maize through a 9.53 mm screen gave as goodperformance with broilers as did maize processed through a 6.35 and a 3.18 mmscreen (Table 5.11). Work reported by Wasburn (1991) suggested that birds underheat stress increased the rate of food passage through the gut, thus altering digestionand subsequent feed utilization, but this was not confirmed in the study by Deatonet al. (1995). There are a number of papers looking at particle size and dietutilization. Nir et al. (1994) looked at particle size of maize and its influence ondiet utilization (Table 5.12) and their work suggests that degree of grinding andparticle size affect diet utilization.

120 J.D. Summers


Inorganic phosphorus(g kg−1)

Phytase(FTU kg−1)

Ileal P absorption(g kg−1 diet)

Phytate degradation(%)

0.50.51.00.50.50.5

000

250250500

0.721.251.411.721.641.71

21.721.620.954.259.071.7

Table 5.10. Effect of dietary P and microbial phytase on the ileal absorption of P in laying hens.(Selected data from Van Der Klis et al., 1997.)



An area that has received passing attention from time to time is the influence offertilizer on composition of maize. Bird and Olsen (1972) reported that fertilizingwith N, P and K resulted in increased crude protein content of the maize and anincrease in some of the EAA. Diets formulated taking into account the higher level ofprotein in the maize, and without EAA supplementation, gave performance resultsfor broilers equal to the control normal maize diet. However, use of the higherprotein maize resulted in a 12% saving in soybean meal (Table 5.13).



Screen size(mm)

Body weight(g) Feed : gain

Mortality(%)

3.186.359.53

244224242434

1.941.941.93

3.23.03.5

Average values for birds reared at 21 and 31°C.

Table 5.11. Effect of hammer-mill screen size of maize grind on male broiler weights to 49 daysof age. (Selected and rearranged data from Deaton et al., 1995.)

Particle sizea

Parameter Coarse Medium Originalb Fine

Weight : volumeGeometric mean diameter (mm)Calculated specific surface (cm2 g−1)

0.6422.01

24.102

0.5450.897

54.102

0.6301.102

54.102

0.4820.525

102.102

aCoarse = 1.41 mm; fine = 0.64 mm; medium = mix of fine and coarse particles.bRegular maize ground with an 8 mm screen size.

(b) Performance of chicks fed the above fractions in maize/soybean meal diets. (Selected datafrom Nir et al., 1994.)

Particle size

Parameter Coarse Medium OriginalOriginal +18% fines

Original +30% fines

Experiment 1 (1–7 days)Weight gain (g)Gain : feed

Experiment 2 (7–21 days)Weight gain (g)Gain : feed

86.1020.806

473.1020.662

100.1020.893

522.1020.725

96.1020.847

463.1020.649

100.1020.870

477.1020.667

94.1020.855

474.1020.662

Table 5.12.(a) Particle size of maize ground to various degrees of fineness.



Potential Immune Response

An interesting paper was published by Williams (1992), who investigated why birdsfed on maize diets were much less susceptible to coccidiosis than birds fed on wheatdiets. While no explanation could be given for the apparent immune response of themaize diet, differences in some of the vitamins in wheat versus maize were discussedas possible contributing factors.

In recent years the biotechnologists have produced several new plant varietiesthat offer protection against pests. The first commercial planting of insect-resistantmaize hybrids, commonly referred to as ‘Bt’ maize, took place in 1996. These hybridspossess a gene that enables the plants to produce an insecticidal protein that is toxicto the larvae of the pests. Brake and Vlachos (1998) compared this type of maize withNM in a broiler feeding test. Their results suggest that the maize is equal to NM in itsnutritive value.

There are several potential avenues for the plant breeders to investigate in theirsearch for novel ways of improving the nutritive value of various cereals. With maizebeing such an important cereal for human as well as animal consumption, it willreceive increased attention in the years ahead as energy becomes more of a limitingfactor in producing food for an expanding world population.

References

Adams, M.H., Watkins, S.E., Waldroup, A.L., Waldroup, P.W. and Fletcher, D.L. (1994)Utilization of high-oil corn in diets for broilers. Journal of Applied Poultry Research 3,146–156.

Bartov, I. and Bar-Zur, A. (1995) The nutritional value of high-oil corn for broiler chicks.Poultry Science 74, 517–522.

Bird, H.R. and Olsen, D.W. (1972) Effect of fertilizer on the protein and amino acid contentof yellow corn and implications for feed formulation. Poultry Science 51, 1353–1358.

Bond, P.L., Sullivan, T.W., Douglas, J.H., Robsen, L.G. and Baier, J.G. (1991) Compositionand nutritive value of an experimental high-protein corn in the diets of broilers andlaying hens. Poultry Science 70, 1578–1584.

Brake, J. and Vlachos, D. (1998) Evaluation of transgenic event 176 ‘Bt’ corn in broilerchicken. Poultry Science 77, 648–653.

122 J.D. Summers


Plot number

Fertilizer level S2 S3 S6 S9 S10

OX2X

7.969.42

10.21

6.958.458.76

7.559.058.98

8.109.36

11.20

8.367.549.80

X = 50 kg each of N, P2O5 and K2O per hectare.

Table 5.13. Fertilizer level and % crude protein in maize (dry matter basis) for various plots.(From Bird and Olson, 1972.)



Brown, I. (1996) Complex carbohydrates and resistant starch. Nutrition Reviews 54(11),S115–S119.

Collins, N.E., Moran, E.T. Jr and Stilborn, H.L. (1998) Corn hybrid and bird maturity affectapparent metabolizable energy values. Poultry Science 77 (Suppl. 1), 42.

Cromwell, G.L., Rogler, J.C., Featherston, W.R. and Cline, T.R. (1968) A comparison of thenutritive value of Opaque-2, Floury-2 and normal corn for the chick. Poultry Science 47,840–847.

Dale, N. (1994) Relationship between bushel weight, metabolizable energy and proteincontent of corn from an adverse growing season. Journal of Applied Poultry Research 3,83–86.

Dale, N. and Jackson, D. (1994) True metabolizable energy of corn fractions. Journal ofApplied Poultry Research 3, 179–183.

Deaton, J.W., Lott, B.D. and Branton, S.L. (1995) Corn grind size and broilers reared undertwo temperature conditions. Journal of Applied Poultry Research 4, 402–406.

Dinn, Z.Z., Bird, H.R. and Sunde, M.L. (1982) Nutritional value of waxy corn for chicks.Poultry Science 61, 998–1000.

Eerlingen, R.C., Jacobs, H. and Delcour, J.A. (1994) Enzyme-resistant starch. V. Effect ofretrogradation of waxy maize starch on enzyme susceptibility. Cereal Chemistry 71(4),351–355.

Ertl, D. and Dale, N. (1997) The metabolizable energy of waxy vs. normal corn for poultry.Journal of Applied Poultry Research 6, 432–435.

Han, Y., Parsons, C.M. and Alexander, D.E. (1987) Nutritive value of high-oil corn forpoultry. Poultry Science 66, 103–111.

Hsu, Li-Wen and Sell, J.L. (1995) Nutritional value for growing turkeys of corn of light testweight. Poultry Science 74, 1703–1707.

Kersey, J.H., Saleh, E.A., Stilborn, H.L., Crum, R.C. Jr, Raboy, V. and Waldroup, P.W.(1998) Effects of dietary phosphorus level, high available phosphorus corn, andmicrobial phytase on performance and fecal phosphorus content. Poultry Science 77(Suppl.), 268.

Knutson, C.A., Khoo, U., Cluskey, J.E. and Inglett, G.E. (1982) Variation in enzymedigestibility and gelatinization behavior of corn starch granule fractions. Cereal Chemistry59(6), 512–515.

Leeson, S. and Summers, J.D. (1976) Effect of adverse growing conditions on corn maturityand feeding value for poultry. Poultry Science 55, 588–593.

Leeson, S., Yersen, A. and Volker, L. (1993) Nutritive value of the 1992 corn crop. Journal ofApplied Poultry Research 2, 208–213.

Li, Y.C., Ledoux, D.R., Venum, T.L., Raboy, V. and Ertl, D.S. (1998) Low phytic acidcorn improves performance and bone mineralization in chicks. Poultry Science 77(Suppl.), 24.

Lilburn, M.S. and Dale, N. (1989) Characterization of two samples of corn that vary in bushelweight. Poultry Science 68, 857–860.

Lilburn, M.S., Ngidi, E.M., Ward, N.E. and Llames, C. (1991) The influence of severedrought on selected nutritional characteristics of commercial corn hybrids. PoultryScience 70, 2329–2334.

Mahagna, M., Said, N., Nir, I. and Nitsan, Z. (1995) The development of digestibility ofsome nutrients and of energy utilization in young broiler chickens. World’s Poultry ScienceAssociation Proceedings, 10th European Symposium on Poultry Nutrition, 250–251.

Maier, D.E. (1995) Indiana corn quality survey – composition data. Grain quality task forceno. 24, Purdue University Extension Service.





Maier, D.E. and Briggs, J.L. (1997) High-oil corn composition. Grain quality task forceNo. 33, Purdue University Extension Service.

Nir, I., Shefet, G. and Aaroni, Y. (1994) Effect of particle size on performance. 1. Corn.Poultry Science 73, 45–49.

Noy, Y. and Sklan, D. (1995) Digestion and absorption in the young chick. Poultry Science 74,366–373.

Pack, M. and Bedford, M. (1997) Feed enzymes for corn–soybean broiler diets, a new conceptto improve nutritional value and economics. AFMA Matrix, December, 18–21.

Pack, M., Bedford, M., Harker, A. and Le Ny, P. (1998) Alleviation of Corn Variability withPoultry Feed Enzymes: Recent Programmes in Development and Practical Application.Finnfeeds International Ltd, Marlborough, UK.

Parsons, C.M., Zhang, Y.E. and Araba, M. (1998) Availability of amino acids in high-oil corn.Poultry Science 77, 1016–1019.

Persia, M.E. and Lilburn, M.S. (1998) The availability of dietary energy from wheat or cornbased diets in growing turkeys. Poultry Science 77 (Suppl.), 43.

Planchot, V., Colonna, P., Gallant, D.J. and Bouchet, B. (1995) Extensive degradation ofnative starch granules by alpha-amylase from Aspergillus fumigatus. Journal of CerealScience 21, 163–171.

Rand, N., Noy, Y., Dvorin, A. and Viola, S. (1998) Effectiveness of high-oil corn (HOC) insummer broiler feeds. Poultry Science 77 (Suppl.), 25.

Saleh, E.A., Watkins, S.E., England, J.A. and Waldroup, P.W. (1997) Utilization of high-oilcorn in broiler diets varying in energy content. Journal of Applied Poultry Research 74,517–522.

Simons, P.C.M., Versteegh, H.A.J., Jongbloed, A.W., Kemme, P.A., Slump, P., Bos, K.D.,Walters, M.G.E., Beudeker, R.F. and Verschoor, G.J. (1990) Improvement of phos-phorus availability by microbial phytase in broilers and pigs. British Journal of Nutrition64, 525–540.

South, J.B., Morrison, W.R. and Nelson, O.E. (1991) A relationship between the amylose andlipid contents of starches from various mutants for amylose content in maize. Journal ofCereal Science 14, 267–278.

Van Der Klis, J.D., Versteegh, H.A.J., Simons, P.C.M. and Kies, A.K. (1997) The efficacy ofphytase in corn–soybean meal-based diets for laying hens. Poultry Science 76, 1535–1542.

Vieira, S.L., Penz, A.M. Jr, Kessler, A.M. and Ludke, J.V. (1997) Broiler utilization of dietsformulated with high oil corn and energy from fat. Journal of Applied Poultry Research 6,404–409.

Wasburn, K.W. (1991) Efficiency of feed utilization and rate of feed passage through thedigestive system. Poultry Science 70, 447–452.

Wasilick, K.R., Fulcher, R.G., Jones, R.J. and Gengenback, B.G. (1994) Characterization ofstarch granules in maize using microspectrophotometry. Starch 46, 369–373.

Whitacre, M.E. (1987) Current research by Degussa. In: Proceedings of DegussaTechnical Symposium, Indianapolis, Indiana. Degussa Corporation Teterboro, NewJersey, pp. 90–122.

Williams, R.B. (1992) Differences between the anticoccidial potencies of monensen inmaize-based or wheat-based chicken diets. Veterinary Research Communications 16,147–152.

Yan, F., Kersey, J.H., Stilborn, H.L., Crum, R.C. Jr, Rice, D.W., Raboy, V. and Waldroup,P.W. (1998) Effects of dietary phosphorus level, high available phosphorus corn, andmicrobial phytase on performance and fecal phosphorus content. 2. Broilers grown tomarket weights in litter floor pens. Poultry Science 77 (Suppl.), 269.

124 J.D. Summers




J. Thorpe and J.D. BealVegetable Protein Meals6

Introduction

The cereal grains that constitute the bulk of animal feedstuffs also provide 30–60%of dietary amino acids (National Research Council (NRC), 1998). However, thisprotein is inadequate, not only in sufficiency but also in amino acid balance. Toobtain optimum animal performance, whether it be growth, reproduction orlactation, protein must be added to the diet to provide both an adequate amount oftotal protein and a balance of amino acids that approaches the ideal amino acidpattern for the species and developmental stage of the animal. This chapter willconcentrate on the use of vegetable protein meals (VPM) and exogenous enzymes inmonogastric farm animals, as exogenous enzyme treatments are not in general usein ruminant diets.

On a global scale the majority of protein ingredients incorporated into animalfeeds are supplied by vegetable proteins, with the oilseeds and legumes being themain providers. In the UK the demand for high quality VPMs has increased due tothe ban on the use of meat and bone meal in animal feeds in the wake of the BSEcrisis. The vegetable sources of protein vary both in crude protein concentrationand in the amino acid composition of the protein. The proximate analyses of some ofthe protein supplements used in animal feeds are outlined in Table 6.1. Amino acidprofiles for these meals can be obtained from various sources, for example Ewing(1997) and NRC (1998).

Anti-nutritional Factors Found in VPMs

The availability of nutrients in feedstuffs is often limited by the presence ofanti-nutritional factors (ANFs), which may limit the use of these feedstuffs inanimal diets. ANFs can be classified in different ways. Huisman and Tolman (1992)classified them on the basis of their effects on the nutritional value of feedstuffs, and



Vegetable Protein Meals andthe Effects of Enzymes

J. THORPE1 AND J.D. BEAL2

1University of Bristol, Department of Clinical Veterinary Science,Langford House, Langford, Bristol BS40 5DU, UK;

2University of Plymouth, Seale-Hayne Faculty,Newton Abbot, Devon TQ12 6NQ, UK

6



the biological response in the animal as: (i) factors with a depressive effect on proteindigestion and on the utilization of protein (protease inhibitors, lectins, phenoliccompounds, saponins); (ii) factors with a negative effect on the digestion of carbo-hydrates (amylase inhibitors, phenolic compounds, flatulence factors); (iii) factorswith a negative effect on the utilization of minerals (glucosinolates, phytic acid, oxalicacid, gossypol); (iv) factors that inactivate vitamins, or cause an increase in theanimal’s vitamin requirement; (v) factors stimulating the immune system that maycause a damaging hypersensitivity reaction (antigenic proteins); and (vi) factors infeed that have a toxic effect (e.g. lectins, cyanide-containing compounds).

Table 6.2 shows the distribution and physiological effects of the various ANFsfound in VPMs. The toxicity and effects of the different ANFs may vary considerablybetween VPMs and these toxic effects may also vary in potency between animalspecies. ANFs appear to play a role in the protection of plants against predation frommoulds, bacteria, insects and birds by disturbing the digestive processes of theseorganisms, and presumably act in a similar manner in domestic animals (Birk, 1989).

Reduction and Elimination of ANFs in VPMs

The ANFs in VPMs can be reduced or eliminated by processing procedures.Maximum destruction of ANFs may require different processing treatments, due todifferences in ANF structure and their biological effects. Processing can be suppliedby physical, chemical, thermal or bacterial means. Currently, various methods can beapplied in the reduction and elimination of ANFs in legume seeds (Table 6.3).

Processing technology is conventionally applied to soybean, the most commonbeing some form of heat treatment, which has proved most effective at reducinglevels of trypsin inhibitors and soybean lectin. Commercial heat processing is

126 J. Thorpe and J.D. Beal


Feedstuff

DE MJ(kg−1)pigs

ME MJ(kg−1)poultry

DM(g kg−1)

CP(g kg−1)

Oil (s.e.)(g kg−1)

CF(g kg−1)

Ash(g kg−1)

Starch andsugars(g kg−1)

Cottonseed meal (s.e.)Faba beansLupin mealPeasPhaseolus beans*Soybean meal (s.e.)Soybeans (FF)Sunflower meal (s.e.)Rapeseed meal (s.e.)

11.115.813.215.4

–14.819.310.212.0

9.113.511.013.0

–10.716.9

7.110.5

900860860860870880890880880

410290320260238470400360385

171853161620

2052032

15990

13570358260

23011

603437353972557070

90455130495541145120

75145

*Adapted from van der Poel (1991).s.e. = Solvent extracted.FF = Full fat.

Table 6.1. Proximate analysis of some VPMs used in animal feeds. (From Ewing, 1997.)



carefully controlled: underheating can result in inadequate inactivation of ANFswhilst overheating may reduce availability through the occurrence of Maillardreactions (reviewed by van der Poel, 1989, and Huisman and Tolman, 1992).

Little attention has been paid to the heat processing of other legumes andoilseeds (post oil extraction). The insufficiencies of some processing techniques haveled to the search for new methods for the elimination of residual ANFs in VPMs.For example, new breeding technologies have led to the development of lowglucosinolate varieties of rapeseed, and although this has made rapeseed a more

Vegetable Protein Meals 127


Anti-nutritional factor Distribution Physiological effect

ProteinsProtease inhibitors

Lectins

Amylase inhibitors‘Antigenic’ proteins

PolyphenolsTannins

GlycosidesVicin / convicin

SaponinsGlucosides

Glucosinolates

Cyanogens

AlkaloidsQuinolizidine

Other ANFsPhytate

Gossypol

SinapinsOligosaccharides(NSP)

Most legumes

Most legumes

Kidney beansSoybean, kidneybeans

Most legumes

Faba beans

Soybean

Rapeseed

Linseed, traces inkidney beans and peas

Lupin

Most legumes

Cottonseed

RapeseedSoybeans, peas, fababeans, Phaseolus beans

Reduction of (chymo)trypsin activity, impairedgrowth, pancreatic hypertrophy, pancreascarcinogenGut wall damage, immune response, increasedendogenous nitrogen lossImpaired digestion of starchImmune response, interference with gut wallintegrity

Interference with protein and carbohydratedigestibility by formation of protein–carbohydratecomplexes

Haemolytic anaemia, adverse effect on eggproductionHaemolysis, effects on intestinal permeability

Impaired iodine utilization (goitogrenicity),reduced palatability and growthRespiratory failure

Neural disturbances, depressed growth, reducedpalatability

Formation of complexes with minerals andproteins, depresses mineral absorptionAnaemia due to formation of iron complexes,reduced egg weight‘Fish’ taint in eggsFlatulence, diarrhoea, discomfort

Table 6.2. Distribution and physiological effects of ANFs found in VPMs. (Adapted from Huismanand Tolman, 1992; Liener, 1994.)



attractive ingredient in animal feed it has not totally solved the ANF problem. Recentresearch has focused on the development of biotechnological approaches to thisproblem, mainly by the application of exogenous enzymes to feedstuffs containingANFs.

The Use of Exogenous Enzymes

Until recently, the majority of feed enzyme research has been performed on cereal-based diets, for example barley and wheat, particularly for poultry diets (Dierick,1989). The type of feed enzymes currently available are outlined in Table 6.4.

There are many factors evolving in the animal feed industry which suggest thatthe use of exogenous enzymes will become more important in the future (reviewed byJohnson et al., 1993). These include: (i) an increasing shift in the use of ‘alternative’feedstuffs in formulating diets; (ii) the use of enzymes known to be effective againstparticular dietary components; (iii) the production of novel by-products (forexample, linseed meal derived after linseed oil production) that have a depressingeffect on growth; (iv) the increased availability of free amino acids which may reducethe requirement for high quality protein supplements; (v) novel feed systems, such asliquid feed systems and those that take advantage of pretreatment, which support theuse of in-feed enzymes; (vi) the introduction of pollution control, for example ofnitrogen and phosphorus (enzymes that reduce excretion of these substances wouldbe advantageous); (vii) the possibility of indirect physiological actions on problems ofcommercial importance, for example the reduction of sticky litter in poultry andpost-weaning diarrhoea in pigs (Inborr and Ogle, 1988); and (viii) the possiblereduction in animal performance due to the restricted use of growth-promotingantibiotics.

Presentation of Exogenous Enzymes to the Animal

The most common method of feeding enzymes is with dry feed, the enzymes usuallybeing added to the feed during blending, often prior to processing treatments such aspelleting. However, most of the enzymes currently of interest to the animal feed



No. Technology

12

3

Breeding and genetic manipulationFeed formulation: selection of ingredients, supplementation with amino acids, othersupplements, e.g. probiotics, enzymes, etc.Primary processing: chemical treatments, enzymatic treatment, physical treatments(fractionation, heat treatments)Secondary processing: conditioning/pelleting

Table 6.3. Processing techniques for reduction and elimination of ANFs in legume seeds (vander Poel, 1989).



industry, such as β-glucanase, amylases, proteases and phytase, suffer temperature-related reduction in activity due to heat treatments (Inborr, 1990; Cowan, 1992;Graham and Inborr, 1993; Bedford and Pack, 1998). This problem can be reducedby the application of the enzyme in a liquid base, post-pelleting, for example sprayingor by use in liquid feed.

Action of Feed Enzymes in the Animal

The application of exogenous enzymes to dry diets presupposes that the enzyme willbe active in the digestive tract of the animal, and it must therefore fulfil a number ofcriteria. The enzyme must be active under the physiological conditions prevailingin the animal’s digestive tract; it must be able to resist proteolysis by the animal’sendogenous proteases and supplement rather than antagonize the animal’s digestiveenzymes. Species differences in the anatomy and physiology of the digestive tract arelikely to affect exogenous enzyme activity in this respect. For example, between pigs



Enzyme Action Target substrate Type of feed Expected benefits

β-Glucanases

Amylases

Cellulases

Pentosanases(Xylanases)

α-Galactosidase

Phytases

Proteases

Lipases

β-Glucans tooligosaccharidesand glucoseDegrade cerealstarch to dextrinsand sugarsCellulose to lowmolecular weightproducts and glucoseArabinoxylans to lowmolecular weightproducts and sugarsDegradesoligosaccharidesand ANFsIncreases availabilityof phosphorus fromphytic acidProtein to peptidesand amino acids

Fats to fatty acids

Barley, oatsand rye baseddietsHigh starchcereal diets

High fibrediets

Rye, barleyand wheat

Soybean andother legumes

Many differentdiets

Wheatby-products,legume proteinsAnimal andvegetable fats

Poultry andpig diets

Early pig/calfdiets

Poor-gradeforages

Pig andpoultry diets

Pig diets

Pig andpoultry diets

Milk replacerusing soybean orsoybean proteinPet diets/broilerdiets

Reduction of stickydroppings, improvedfeed utilizationIncreased availabilityof cereals in weanerfeedsImproved energyavailability

Improved litterquality, improvedfeed utilizationImproved energyavailability, reducedscours in pigletsReduces needfor inorganicphosphorusHigher proteindigestibility, lowernitrogen excretionImproved digestibilityof fat and enhancedenergy retention asa result

Table 6.4. Feed enzymes in use today. (From Ogden, 1995, modified from Rotter et al., 1989;Cowan, 1992.)



and poultry these differences include the following (Partridge, 1993; Dierick andDecuypere, 1994).

1. Anatomical: in poultry, feed passes into the crop, where any added enzymescan act for several hours at a pH of approximately 6.0 before passing into the acidenvironment of the gizzard, whereas in the pig, feed passes directly into the acidenvironment of the stomach immediately after ingestion.2. Digestive capacity: poultry have a shorter small intestine and thus reducedpossibilities for enzyme inactivation by the microflora, a shorter mean retention timein the small intestine (1–2 h in poultry versus 4–5 h in the pig) and a lower watercontent in the upper part of the gastrointestinal tract.3. Bacterial activity: the importance of the microflora in the gut of poultry is muchless than in the pig.4. Fibre fermentation: lower in poultry than in pigs, due to their widely differenthind gut capacities.

Survivability of enzymes in the digestive tract varies widely; for example,Thacker and Baas (1996) and Baas and Thacker (1996) demonstrated that 84% ofpentosanase and 26% of β-glucanase activity was recovered in the duodenal digestaof pigs 4 h after feeding diets supplemented with these enzymes. Approximately75% of exogenous protease has been detected in the ileal digesta of young pigs fedprotease-supplemented diets.

An alternative to the use of exogenous enzymes in dry diets is the pretreatmentof the diet or its components prior to secondary processing. Adding enzymes tocomplete compound feeds may be appropriate when the enzyme is targeted at cereals,which make up the bulk of the diet. However, VPMs only constitute a smallproportion of the total diet and therefore target substrates are considerably diluted.The dilution effect may be overcome by pretreating the VPM prior to inclusion inthe diet. This approach has been adopted in much of the recent research into theeffect of enzyme application to VPMs, much of which has focused on the effects ofproteases on soybean meals.

Pretreatment of Diets Containing VPMs

Pretreatment of soybean with proteases

The prime target of proteases in soybean are the anti-nutritional factors, such asresidual trypsin inhibitors, lectins and ‘antigenic’ protein. Pretreatment of specificfeed components allows the effect of any enzyme addition to be ascertained prior tofeeding the animal. This is a pertinent point as far as the inclusion of proteasesin animal feeds is concerned, as it is impossible to assess exogenous protease activityin ileal digesta. Immunochemical techniques are being developed that will detect thepresence of an exogenous protease in ileal digesta; however, these give no indicationof functionality (Thorpe, 1999).





Pretreatment of soybeans with enzymes has been used, albeit inadvertentlythrough the processes of fermentation, to improve the nutritional quality of legumesin human foods for many years (Campbell-Platt and Cook, 1991). It is not the inten-tion of this chapter to review fermentation processes extensively as little work hasbeen undertaken on the effect of fermenting protein meals for inclusion in animalfeeds. However, Zamora and Veum (1979) fed heated soybeans fermented withAspergillus oryzae or Rhizopus oligosporus to growing pigs. Table 6.5 summarizes theirresults. Fermentation significantly improved feed conversion ratio (FCR) with bothorganisms and average daily gain with A. oryzae, and showed an improvement(although not significant) in nitrogen retention with both organisms. This workindicated that fermentation improves the performance of growing pigs, possibly bythe action of fungal enzymes on the soybean present in the feed.

In the 1950s, work by Lewis et al. (1955) and Baker et al. (1956) examined theeffects of the addition of pepsin, pancreatin, a fungal protease, a diastatic proteaseand papain to various soybean diets in a number of trials using pigs from 6 to 67 daysold. Pepsin and pancreatin supplementation showed some beneficial effects on aver-age daily gain (ADG) and FCR, particularly in younger pigs. Papain and the diastaticprotease were as effective as pepsin and pancreatin, but the fungal protease showedno beneficial effects. This work was aimed at supplementing the insufficiencies ofproteolytic and amylolytic digestive enzymes in the baby pig, as opposed to thecurrent work, which targets ANFs in protein meals.

More recent work has concentrated on the use of isolated enzymes producedfrom large-scale microbial fermentations. This research can be divided into eitherin vitro or in vivo studies. In vitro studies have been used by a number of workers inthis field to determine the effect of exogenous enzymes on VPMs prior to expensivein vivo studies. Assessment of enzyme activity in vitro may indicate any potentialbenefits the enzyme may have in vivo. A number of studies have been undertakenutilizing proteases on soybean meals in vitro, analysing different parameters.

Huo et al. (1993) showed that one fungal and four bacterial protease enzymescould inactivate trypsin inhibitors and lectin in raw soybean (RSB) and low-temperature extruded soybean (LTES) in vitro, to various extents. The most effectiveof these reduced trypsin inhibitor levels in RSB and LTES by 96% after 12 hincubation at an inclusion level of 1%. Apart from being described as fungal orbacterial proteases, there were no other details provided about the enzymes. Thebacterial proteases appeared to be more effective at breaking down trypsin inhibitorsthan the fungal protease. Meijer and Spekking (1993) isolated microorganisms



Soybean treatment ADG (kg) FCR N retention (%)

UnfermentedFermented with A. oryzaeFermented with R. oligosporus

0.52a0.60a

0.56

2.33a

2.04a

1.96a

50.655.755.0

aP > 0.05.

Table 6.5. Summary of results of Zamora and Veum (1979).



producing enzymes that could utilize purified sources of Kunitz soybean trypsininhibitor (KSTI) and Bowman-Birk inhibitor (BBI) as sole carbon, nitrogen andsulphur sources, and inactivate them over a time course of several days. Thereforethese microorganisms must all possess enzymes capable of hydrolysing the trypsininhibitors in soybeans. This does not necessarily mean that trypsin inhibitors wouldbe utilized in preference to the other sources of C, N or S in whole soybeans. Theseenzymes may have potential activity against trypsin inhibitors in other VPMs.

Rooke et al. (1996) aimed to assess whether protease treatment of soybean meal(SBM) could reduce its in vitro antigenicity and improve its nutritional value whenfed to newly weaned piglets. Table 6.6 (Ref. 1) describes the protease treatment ofSBM in this study. Antigenic soybean proteins were measured by a competitiveenzyme-linked immunosorbent assay (ELISA) technique. All diets, including the



Ref.no.

Vegetableprotein Enzyme Treatment Reference

1

2

3

4

5

6

7

8

9

10

11

SBM

SBM

SBM

SBM

FFSBM(autoclaved)RawsoybeanMicronizedFFSBMRawsoybean

Acid protease(P2)

Acid protease(P2)

Alkalineprotease (P1)

Acid protease(P2)

Alkalineprotease (P1)

Protease(B.subtilis:subtilisin)

Protease (P4)

Protease (P4)

Protease (P3)

Protease (P3)

0.1% protease added to SBM (800 g kg−1

moisture) pH 4.5. Incubated for 3 h at 50°C,neutralized, dried at 65°C0.1% protease added to SBM (800 g kg−1

moisture) pH 4.5. Incubated for 3 h at 50°C,dried at 55°C0.1% protease added to SBM (800 g kg−1

moisture) pH 8.5. Incubated for 3 h at 50°C,dried at 55°C0.1% protease added to SBM (800 g kg−1

moisture) pH 4.5. Incubated for 2 h at 50°C,fed as a wet mash0.1% protease added to SBM (800 g kg−1

moisture) pH 8.5. Incubated for 2 h at 50°C,fed as a wet mash0.1% protease added to SBM (1 : 2 wt : vol water)pH 4.5. Incubated for 16 h at 50°C, freeze dried50 ml of enzyme solution (pH 4.5) to give finalenzyme concentration of 0.1% sprayed onSBM, air dried at ambient temperature for 24 h0.25% added to soybean meal (1 : 3 wt : volwater) incubated for 24 h at 20°C, fed as liquid0.25% added to soybean meal (1 : 3 wt : volwater) incubated for 24 h at 20°C, fed as liquid0.5% added to soybean meal (1 : 3 wt : volwater) incubated for 24 h at 20°C, fed as liquid0.5% added to soybean meal (1 : 3 wt : volwater) incubated for 24 h at 20°C, fed as liquid

Rooke et al.(1996)

Hessinget al. (1996)

Hessinget al. (1996)Rooke et al.(1998)

Caine et al.(1997)

Beal et al.(1998b)

Beal et al.(1999)

SBM = soybean meal – method of processing unstated. FFSBM = full fat soybean meal. Proteases(refs 1–11) supplied by Finnfeeds International Ltd (Marlborough, UK).

Table 6.6. Summary of recent studies utilizing protease pretreatment of SBM.



skim milk diet, contained antigenic soybean proteins, indicating a possible problemwith the technique. The SBM treated with protease contained fewer antigenicproteins than the other soybean-containing diets. It is assumed here, as in manystudies, that reduction in levels of antigenic proteins as measured by ELISA indicatesprotein denaturation by exogenous protease treatment. This is not necessarily anaccurate assumption as, for example, antiserum from animals immunized withsoybean protein may not recognize any protein components of an enzyme-treatedsoybean product, but the treated product may still contain antigenic epitopes.

Hessing et al. (1996) examined the ability of two microbial proteases (P1 andP2) to degrade ANFs, and to determine whether enzymatically hydrolysed SBMcould improve the productive performance of newly weaned piglets or broiler chicks.The SBM was pretreated with the protease before feeding. Table 6.6 (Refs 2–5)describes the protease pretreatments. SDS-PAGE and Western blotting analysisdemonstrated that P1 could significantly hydrolyse the storage proteins glycinin andβ-conglycinin, and to a certain extent KSTI at inclusion levels of 1000–10,000 U g−1

material, but there were no effects on soybean lectin. It was concluded that thepotential of proteases to improve the nutritional value of soybean meal waspromising. It was cautioned that in vitro immunochemical analysis of ANFs mustbe interpreted with care, as the hydrolysis of proteins such as trypsin inhibitorsmay expose more antigenic epitopes, but the inhibitor itself may not be functional,resulting in misleading results.

Caine et al. (1997) determined the optimum conditions for Bacillus subtilissubtilisin pretreatment of SBM by measuring increases in protein solubility. Theyfound the optimum conditions to be incubation at 50°C and pH 4.5.

Beal et al. (1998a) used an in vitro technique (Boisen and Fernandez, 1997) asan initial evaluation of the potential of three proteases to improve the in vitronitrogen digestibility of RSB and four processed SBMs. They found significantincreases (P < 0.05) of approximately 5–12% over control with the three proteases.These results indicate the possibility of increases of protein digestibility of these mealsin vivo. Beal et al. (1998c) also examined the effects of one of these proteases (P4)using SDS-PAGE, and found a reduction in the number and density of proteinbands with apparent molecular weights greater than 66 kDa, indicating thehydrolysis of storage proteins.

Rooke et al. (1998) also used SDS-PAGE to examine the effects of proteases P1and P2 (Table 6.6, Refs 4–5) on SBM. Pretreatment changed the composition ofSBM, and soluble α-amino nitrogen concentrations were increased by treatmentwith P1 and P2. P1 reduced antigenic protein concentration (as determined bySDS-PAGE) to a greater extent than P2.

Table 6.6 summarizes exogenous protease pretreatments of soybean mealsin recent studies described above. The in vivo results obtained from these trials aresummarized in Table 6.7. The results of these trials are variable. In some studies,enzyme pretreatment produced positive effects on some of the parameters measuredwhilst in others there were no significant effects. As with all such studies, it is difficultto draw direct comparisons, due to differences in age, weight and genotype of pig andin diet formulation.





Pretreatment of other VPMs with exogenous enzymes

Castanon and Marquardt (1989) investigated the effects of Celluclast (cellulase),SP249 (polygalactouronase), Alcalase (protease) and BAN (α-amylase), in variouscombinations, on raw, autoclaved, or steeped faba beans for young chickens. The



Crossref.

Animal age/weightand trial duration Summary of results, enzyme treatment (T) compared with control (C)

1

2

3

4

5

6

7

8

9

10

11

Newly weaned pigs21 days

7 day old chicks141 g, 27 days

7 day old chicks141 g, 27 daysNewly weaned pigs7.5 kg, 14 days

Newly weaned pigs7.7 kg, 14 days

Newly weaned pigs6.4 kg, 2 × 9 daychangeoverNewly weaned pigs6.4 kg, 2 × 9 daychangeoverGrowing pigs33.5 kg, 6 weeksGrowing pigs33.5 kg, 6 weeks

Growing pigs28.7 kg to slaughter(90 kg)Growing pigs28.7 kg to slaughter(60 kg)

Significant (P > 0.01) increase in ADG for 7 days post weaning155 g day−1(T) vs. 95 g day−1 (C). Significant increase in ADFI for 21days post weaning 392 g day−1(T) vs. 347 g day−1(C). No significantincrease in FCRSignificant increase (P < 0.01) in ADG; 41 g day−1(T) vs. 30 gday−1(C), ADFI; 79 g day−1(T) vs. 64 g day−1(C), apparent ileal Ndigestibility 0.85(T) vs. 0.76(C)No significant increases in ADG, ADFI or apparent ileal NdigestibilitySignificant increases (P < 0.01) in ADG over first 7 dayspostweaning; 120 g day−1(T) vs. 64 g day−1(C) and (P < 0.05) inADG over 14 days postweaning; 181 g day−1(T) vs.128 g day−1(C).No significant difference in FCRSignificant decrease (P < 0.01) in ADG over first 7 dayspostweaning; 78 g day−1(T) vs. 143 g day−1(C) and (P < 0.05) inADG over 14 days postweaning; 153 g day−1(T) vs. 199 g day−1(C).No significant difference in FCRNo significant differences in ADFI, ADG, FCR or apparent ilealamino acid digestibilities

No significant differences in ADFI, ADG, FCR or apparent ilealamino acid digestibilities

No significant differences in ADG, ADFI or FCR

Significant (P < 0.05) increases in ADFI; 1.37 kg day−1(T) vs. 1.24 kgday−1(C), ADG 0.606 kg day−1(T) vs. 0.529 kg day−1(C). Nosignificant difference in FCRNo significant difference in ADG, ADFI, FCR or lean : fat ratio.Significant reduction of 4 ± 1.86 days to reach slaughter weight dueto protease additionSignificant (P < 0.05) improvement in ADG; 0.606 kg day−1(T) vs.0.515 kg day−1(C) and FCR; 3.042(T) vs. 3.512(C) in finisher period.Significant (P < 0.05) increase of 0.423 ± 0.158 in lean : fat ratio dueto protease addition

CP = crude protein; ADG = average daily liveweight gain; ADFI = average daily feed intake;FCR = feed conversion ratio (feed/gain).

Table 6.7. Summary of the results of recent trials utilizing protease pretreatment of SBM.



authors found that cellulase and protease, and a combination of the two, significantlyincreased weight gain and FCR (P < 0.05) by approximately 5–10% in both raw andautoclaved beans, and the combination significantly improved the weight gain(P < 0.05) and FCR (P < 0.01) of steeped beans. The other enzymes had no effect.

Näsi (1991) investigated the effects of processing, enzyme pretreatmentand multi-enzyme supplementation (cellulase, protease and β-glucanase) of SBMand rapeseed meal (RSM) diets on the nutrient digestibility, protein utilization andperformance in growing pigs. Extrusion and the addition of enzyme premixsignificantly improved the organic matter and protein digestibilities of SBM(P < 0.05). No significant differences in growth rate, FCR or carcass quality wereseen between pigs fed diets containing differently treated SBM, relative to anuntreated control. No significant effects of enzyme treatment on any parameters wereobserved with RSM.

Schulze et al. (1997) examined the effects of germination or pancreatinpretreatment of Phaseolus vulgaris (white kidney bean) on the ANF activities andapparent ileal digestibility of nitrogen in young pigs. The authors found germinationsignificantly decreased (P < 0.05) the flow of amino acids at the terminal ileum byapproximately 50% compared with pancreatin or control. Trypsin inhibitor andlectin activity were reduced by germination, although the significance of thesereductions was not reported.

Direct Addition of Enzymes to Complete Diets ContainingVPMs

A number of pig and poultry trials have been performed using the addition ofexogenous enzyme premixes to dry diets. Pluske and Lindemann (1998) haveextensively reviewed the use of Vegpro, a mixture of protease, cellulase, pentosanase,α-galactosidase and amylase, in pig and poultry trials using various VPMs. Cowanet al. (1996) reviewed the use of Biofeed Plus CT (main activity protease) andEnergex MG (main activities β-glucanase and pectinase). Some other work using thesame enzymes is reviewed below.

Enzymes targeted to SBM dietary components

De Koning and van der Wel (1996) supplemented starter broiler maize/SBM dietswith Vegpro at 0.1 and 0.05%. Significant increases (P < 0.05) in weight gain after21 days were observed with both enzyme inclusion levels, but no significantdifferences in FCR were observed. Similar results were obtained by Sefton andPerdok (1996). Swift et al. (1996) examined the effects of Vegpro on expandedand pelleted maize/SBM diets for broilers. Enzyme treatment significantlyimproved (P < 0.05) FCR in the expanded diet over the 35-day feeding period.Vegpro significantly improved (P < 0.05) nitrogen and energy digestibility in growerdiets but not starter diets. Schang et al. (1997) evaluated the effects of Vegpro in





maize/SBM and maize/full-fat SBM (FFSBM) diets for broilers, using high andlow nutrient density formulations. There were no significant effects of addingVegpro to high density diets. Significant improvements (P < 0.05) in weight gainwere seen with maize/FFSBM at low nutrient digestibilities. Spring et al. (1998)examined the effect of Vegpro on a grain/SBM diet with reduced protein con-centration, in weaned piglets. Significant improvements (P < 0.05) in FCR due toVegpro treatment were observed in diets in which the lysine concentration wasreduced to 0.97%.

Enzymes targeted to other VPM dietary components

Viveros et al. (1993) examined the effect of addition of a combination of BiofeedPro, a mixture of protease, amylase, hemicellulase, pentosanase and xylanase, andBAN, a mixture of amylase and β-glucanase to diets containing 10 and 15% fababean hulls in diets for chickens. Addition of the enzyme mixtures significantlyimproved (P < 0.05) weight gain in 10 and 15% hull diets, and FCR in 15% hulldiets. However, enzyme addition did not improve the feed efficiency of the hull dietsto the level of a maize/SBM diet. The same workers investigated the use of tannase ina 10% hull diet. Tannase addition significantly improved (P < 0.05) weight gain, butnot FCR.

Pfirter et al. (1993) investigated the effects of exogenous protease supple-mentation on broiler feeds containing crude or extracted lupins. No significantimprovements over control, crude lupin diets were observed.

Hadron et al. (1993) evaluated the effects of exogenous carbohydrases andproteases on growing-pig diets containing peas. No effects of enzymes were observedon nitrogen or energy digestibility, or on production characteristics.

Bedford and Morgan (1995) examined the effects of the addition of xylanase,xylanase plus alkaline protease, xylanase plus neutral protease, and xylanase plusacid protease on canola meal (CM) diets for broilers. Addition of xylanase alonesignificantly improved (P < 0.05) weight gain over the CM control diet. Additionof xylanase and alkaline protease significantly improved (P < 0.05) FCR over aSBM-based control diet. The addition of acid protease also significantly (P < 0.05)improved FCR, but the neutral protease had no effect.

Annison et al. (1996) examined the effects of two commercial enzymepreparations on the nutritive value of dehulled lupins for broiler diets. Enzyme Acontained xylanase, pentosanase, and hemicellulase; Enzyme B containedβ-glucanase, hemicellulase and pectinase. Enzyme A increased the AME of lupins(P < 0.05); Enzyme B had no significant effects.

Stanley et al. (1996) examined the effect of Vegpro addition on various levelsof cottonseed meal (CSM) in diets for broiler chicks. The inclusion of Vegprosignificantly improved (P < 0.05) FCR in diets containing 7.5, 15 and 30% levels ofCSM. Schang and Azcona (1998) evaluated the effects of Vegpro on maize/sunflowerdiets as a replacement for maize/SBM diets. Sunflower meal significantly (P < 0.05)decreased egg production, and the addition of Vegpro did not compensate for this.





Discussion

Trials conducted to date on the effects of exogenous enzyme supplementation ofVPMs show variable results, with some showing significant changes with enzymeinclusion, and others showing trends towards improvement, which are not actuallysignificant. There are a number of possible explanations for these observations. Inmany cases it is impossible to draw direct comparisons between individual studiesdue to differences in enzyme type, activity and inclusion level. Where enzymeactivities are actually stated, there are often differences in units and the method ofdetermining activity is not always stated.

The majority of trials have been performed under experimental conditions, withhigh health status animals, which may not necessarily reflect commercial situations.Basal diets can vary widely between different trials depending on the availability ofdietary components. It must be remembered that modern diets are formulated toachieve maximum performance from an animal. Any increase in nutrient availabilitydue to enzyme supplementation may not elicit a response from the animal unless thedietary factor in question is reduced to suboptimal levels. This is not necessarilya consideration when substituting proportions of a diet with inferior feedstuffs,provided nutrient specifications do not exceed recommended levels for that species,age and genotype. For example, if the effect of protease inclusion in the diet is beinginvestigated, the protein : energy ratio needs to be reduced to suboptimal levels inorder for the animal to respond to any increase in amino acid availability due toenzyme treatment. In the studies outlined in Tables 6.6 and 6.7, Rooke et al.(1996, 1998), Hessing et al. (1996) and Beal et al. (1998b, 1999) indicated thatlysine : energy ratios were reduced below recommended levels, whereas Caine et al.(1997) stated that the diets used were formulated to exceed NRC recommendations.This could explain the lack of response to enzyme inclusion in their study. Otherstudies corroborate this, such as Schang et al. (1997) and Spring et al. (1998), whoshowed significant differences with diets formulated with nutrient densities belowrecommendations, but not with diets formulated to meet or exceed nutrientrecommendations.

It is difficult to assess the efficacy of enzyme preparations on target substrateswhen applied to complete diets, as all plant feedstuffs in a diet are likely to containsubstrates on which enzymes can act. This does not present a major problem withenzymes designed to target substrates such as NSPs, which can present problemsin other major dietary constituents, such as cereals. However, when targetingproteinaceous ANFs in VPMs with exogenous proteases, a number of factors need tobe considered. The VPM in a compound feed represents only a fraction of the totalfeed, typically 20–30%. The target substrate for the enzyme will be even less thanthis. For example, if full-fat soybean meal with a crude protein content of 40%(see Table 6.1) is included in a diet at a rate of 25%, then the actual soybean proteincontent of that compound feed will be 10%, and the ANF content even less, at 6–7%of the total protein (Nielsen, 1983). This represents a considerable dilution of thesubstrate by other feed components; the competition from other protein substrates ina complete diet may result in little or no ANF degradation. Pretreatment of VPMs





prior to inclusion in a diet has two advantages. Firstly, the dilution effect of otherdietary components is reduced. Secondly, pretreatment allows the effects of enzymeactivity on the specific substrate to be evaluated. Analysis of in vitro results ofpretreatment have to be interpreted with caution; for example, ELISA techniquesmay demonstrate changes in protein structure, but not necessarily functionality.However, the development of immunochemical assays which measure KSTI, BBIand soybean lectin (SBL), both by aspects of protein structure and by biologicalactivity, in feed and digesta samples may allow this problem to be resolved (Thorpeet al., 1997; Miller et al., 1998). Techniques used to measure in vitro digestibilitycannot necessarily be extrapolated to in vivo situations; however, they are useful toolsin the initial screening of the suitability of enzyme preparations.

In designing exogenous enzyme studies, the predicted application of the enzymeneeds to be considered to allow the transfer of technologies to commercial situations.This is of particular importance with pretreatment strategies. Some of the studiespreviously described use pretreatments with relatively energy-intensive processes,such as high incubation temperatures followed by drying (Hessing et al., 1996;Rooke et al., 1996; Caine et al., 1997). This may not be economically viable whenextrapolated to a commercial situation.

New developments in feeding systems, particularly of those for pigs, haveincreased scope for effective commercial enzymatic applications. One such develop-ment is the increased use of liquid feeding systems, which provide excellentconditions for feed enzymes. On many UK farms, liquid feed systems for pigs areused to enable farmers to take advantage of liquid by-products from food processors;thus diets are often mixed on-farm (Geary, 1997). Enzyme preparations couldbe added directly into the feeding system, thereby eliminating the risk of enzymedenaturation during processing procedures such as pelleting. These systems providean ideal opportunity for pretreatment of individual raw materials, such as VPMs,in situ, in a commercial environment, before addition of other dietary components.Pretreatment of VPMs prior to incorporation into poultry diets does not appear tohave attracted so much attention. Pretreatment may not be so relevant to the poultryindustry, as liquid feed systems are generally not used and anatomical features of thefowl digestive system effectively provide a pretreatment chamber in the crop. Theactivity in the crop may explain the greater response of poultry to exogenous enzymesupplementation in general.

Pretreatment of SBM with exogenous enzymes appears to have the greatesteffect in young animals, particularly noted in early weaned pigs (Hessing et al.,1996; Rooke et al., 1996, 1998; Caine et al., 1997). This effect is usually greatestin the 7 days immediately post-weaning, possibly helping to overcome the transientdeleterious effect of feeding SBM observed by workers such as Li et al. (1991a,b) andDreau et al. (1994). At this time the animal is often performing suboptimally, givinga window of opportunity for the use of exogenous enzyme supplementation. Effectsof enzyme supplementation do not appear to be as great in grower/finisher pigs.Possible reasons for this include the maturity of the digestive and immune systemand the formulation of diets to high nutrient specifications. The major VPM in usein grower/finisher pig diets at present is SBM. This is often highly processed to





maximize digestibility (van der Poel, 1989). Processes used are often expensive and itis possible that the use of enzyme supplementation of less digestible SBMs or otherVPMs may allow higher inclusion rates of less expensive ingredients in diets withoutaffecting performance.

Future Research Requirements

To date, the majority of research has been aimed towards the improvement ofSBMs. SBM is considered to be a high quality protein source and other, possiblylower quality, VPMs may offer more scope for improvement by exogenous enzymesupplementation. This could allow the use of VPMs from indigenous rather thanimported sources.

The pretreatment of individual dietary components with exogenous enzymeshas not been fully explored. In particular, the application of exogenous enzymepretreatments in liquid feed systems in the pig industry provides a potentialopportunity for the improvement of the nutritional status of VPMs.

There is little published data on the optimum inclusion levels of exogenousenzymes in animal diets. Inclusion levels are often extrapolated between speciesand, considering the differences in anatomy and physiology of the digestive tract,these levels may be incorrect. Maximum improvements of VPMs may not beachieved unless optimum enzyme inclusion levels are used. These levels need to bedetermined.

The implications of the recent European ban of antibiotic growth promotersin animal diets on performance and health status have yet to be fully explored. It ispossible that this ban may result in decreased animal performance and exogenousenzymes could be utilized to help to alleviate this.

References

Annison, G., Hughes, R.J. and Choct, M. (1996) Effects of enzyme supplementation on thenutritive value of dehulled lupins. British Poultry Science 37, 157.

Baas, T.C. and Thacker, P.A. (1996) Impact of gastric pH on dietary enzyme activity andsurvivability in swine fed β-glucanase supplemented diets. Canadian Journal of AnimalScience 36, 245–252.

Baker, R.O., Lewis, C.J., Wilbur, R.W., Hartman, P.A., Speer, V.C., Ashton, G.C. andCatron, D.V. (1956) Supplementation of baby pig diets with enzymes. Journal of AnimalScience 15, 1245.

Beal, J.D., Brooks, P.H. and Schulze, H. (1998a) The effect of pre-treatment with differentproteases on the in vitro digestibility of nitrogen in raw soya bean and four different fullfat soya bean meals. In: van Arendonk, J.A.M. (ed.) Book of Abstracts of the 49th AnnualMeeting of the European Association for Animal Production, Warsaw, Poland. WageningenPers, Wageningen, The Netherlands. PN 2.7 (Abstr.) p. 264.

Beal, J.D., Brooks, P.H. and Schulze, H. (1998b) The effect of the addition of a proteaseenzyme to raw or autoclaved soya bean on the growth performance of liquid fed





grower/finisher pigs. British Society of Animal Science Winter Meeting, Scarborough, UK,p. 161.

Beal, J.D., Brooks, P.H. and Schulze, H. (1998c) The hydrolysation of protein in raw andautoclaved soyabean meals by a microbial protease. British Society of Animal ScienceWinter Meeting, Scarborough, UK, p. 167.

Beal, J.D., Brooks, P.H. and Schulze, H. (1999) The effect of protease pre-treatment of rawor micronized soyabean meal on the growth performance and carcass composition inliquid fed grower and finisher pigs. British Society of Animal Science Winter Meeting,Scarborough, UK, p. 170.

Bedford, M.R. and Morgan, A.J. (1995) The use of enzymes in canola-based diets. In:van Hartingsveldt, W., Hessing, M., van der Lugt, J.P. and Somers, W.A.C. (eds) 2ndEuropean Symposium on Feed Enzymes. Proceedings of ESFE2, Noordwijkerhout, TheNetherlands, pp. 125–131.

Bedford, M. and Pack, M. (1998) Thermostability of enzymes in feed processing: re-thinkingfeed assays. Feed International June, 18–21.

Birk, Y. (1989) Protein protease inhibitors of plant origin and their significance in nutrition.In: Huisman, J., van der Poel, A.F.B. and Liener, I.E. (eds) Recent Advances of Researchin Antinutritional Factors in Legume Seeds. Proceedings of the First InternationalWorkshop on ‘Antinutritional Factors (ANFs) in Legume Seeds’. Pudoc, Wageningen,The Netherlands, pp. 83–94.

Boisen, S. and Fernandez, J.A. (1997) Prediction of the total tract digestibility of energy infeedstuffs and pig diets by in vitro analyses. Animal Feed Science and Technology 68,277–286.

Caine, W.R., Sauer, W.C., Tamminga, S., Verstegen, M.W.A. and Schulze, H. (1997) Appar-ent ileal digestibilities of amino acids in newly weaned pigs fed diets with protease-treatedsoybean meal. Journal of Animal Science 75, 2962–2969.

Campbell-Platt, G. and Cook, P.E. (1991) Enzymes in fermented vegetable and legumeproducts. In: Fox, P.F. (ed.) Food Enzymology, Vol. 2. Elsevier Applied Science, NewYork, pp. 455–478.

Castanon, J.I.R. and Marquardt, R.R. (1989) Effect of enzyme addition, autoclave treatmentand fermenting on the nutritive value of field beans (Vicia faba L.). Animal Feed Scienceand Technology 26, 71–79.

Cowan, D. (1992) Advances in feed enzyme technology. Agro Food Industry Hi Tech 3(3), 9.Cowan, W.D., Korsbak, A., Hastrup, T. and Rasmussen, P.B. (1996) Influence of added

microbial enzymes on energy and protein availability of selected feed ingredients. AnimalFeed Science and Technology 60, 311–319.

de Koning, W. and van der Wel, C. (1996) Effect of inclusion of vegetable protein enzymein broiler starter diets. Worlds Poultry Congress, New Delhi, India, 2–5 September, vol. IV,p. 250.

Dierick, N.A. (1989) Biotechnology aids to improve feed and feed digestion: enzymes andfermentation. Archives on Animal Nutrition Berlin 3, 241–261.

Dierick, N.A. and Decuypere, J.A. (1994) Enzymes and growth in pigs. In: Cole, D.J.A.,Wiseman, J. and Varley, M. (eds) Principles of Pig Science. Nottingham University Press,Nottingham, pp. 169–195.

Dreau, D., Lalles, J.P., Philouze-Rome, V., Toullec, R. and Salmon, H. (1994) Local andsystemic immune response to soybean protein ingestion in early weaned pigs. Journal ofAnimal Science 72, 2090–2098.

Ewing, W.N. (1997) The Feeds Directory. Context, Coalville, Leicestershire, UK, p. 118.





Geary, T.M. (1997) Improving performance of weaner pigs through developments in liquidfeeding. PhD thesis, University of Plymouth, Plymouth, UK.

Graham, H. and Inborr, J. (1993) Stability of enzymes during processing. Feed Mix 3(1),18–35.

Hadron, R., Jeger, P., Messikommer, R., Yverneau, A. and Wenk, C. (1993) Effect of fieldpea (Pisum sativum L.) and supplemental enzymes in diets for growing pigs. In: Enzymesin Animal Nutrition, Proceedings of the 1st Symposium. Kartause Ittingen, Switzerland,pp. 157–159.

Hessing, M., van Laarhoven, H., Rooke, J.A. and Morgan, A. (1996) Quality of soyabeanmeals (SBM) and effect of microbial enzymes in degrading soya antinutritionalcompounds (ANC). In: 2nd International Soyabean Processing and Utilization Conference.Bangkok, Thailand, pp. 8–13.

Huisman, J. and Tolman, G.H. (1992) Antinutritional factors in the plant proteins of dietsfor non-ruminants. In: Garnsworthy, P.C., Haresign, W. and Cole, D.J.A. (eds) RecentAdvances in Animal Nutrition. Butterworth-Heinemann, Oxford, pp. 3–32.

Huo, G.C., Fowler, V.R., Inborr, J. and Bedford, M. (1993) The use of enzymes to denatureantinutritive factors in soyabean. In: van der Poel, A.F.B., Huisman, J. and Saini, H.S.(eds) Recent Advances of Research in Antinutritional Factors in Legume Seeds. Proceedingsof the Second International Workshop on ‘Antinutritional Factors (ANFs) in LegumeSeeds’. Wageningen Pers, Wageningen, The Netherlands, pp. 517–521.

Inborr, J. (1990) Practical application of feed enzymes. Feed Compounder 10(1), 41–49.Inborr, J. and Ogle, R.B. (1988) Effect of enzyme treatment of piglet feeds on performance

and post weaning diarrhoea. Swedish Journal of Agricultural Research 18, 129–133.Johnson, R., Williams, P. and Campbell, R. (1993) Use of enzymes in pig production.

In: Enzymes in Animal Nutrition, Proceedings of the 1st Symposium. Kartause Ittingen,Switzerland, pp. 49–60.

Lewis, C.J., Catron, D.V., Liu, C.H., Speer, V.C. and Ashton, G.C. (1955) Enzymesupplementation of baby pig diets. Agricultual Food Chemistry 3(12), 1047–1050.

Li, D.F., Nelssen, J.L., Reddy, P.G., Blecha, F., Klemm, R. and Goodband, R.D. (1991a)Interrelationship between hypersensitivity to soybean proteins and growth performancein early-weaned pigs. Journal of Animal Science 69, 4062–4069.

Li, D.F., Nelssen, J.L., Reddy, P.G., Blecha, F., Klemm, R., Giesting, D.W., Hancock, J.D.,Allee, G.L. and Goodband, R.D. (1991b) Measuring suitability of soybean productsfor early weaned pigs with immunological criteria. Journal of Animal Science 69,3299–3307.

Liener, I.E. (1994) Implications of antinutritional components in soybean foods. CriticalReviews in Food Science and Nutrition 34(1), 31–67.

Meijer, M.M.T. and Spekking, W.T.J. (1993) Enzymatic inactivation of proteinaceousprotease inhibitors. In: van der Poel, A.F.B., Huisman, J. and Saini, H.S. (eds) RecentAdvances of Research in Antinutritional Factors in Legume Seeds. Proceedings of the SecondInternational Workshop on ‘Antinutritional Factors (ANFs) in Legume Seeds’.Wageningen Pers, Wageningen, The Netherlands, pp. 529–533.

Miller, B.G., Thorpe, J., Patel, D. and Schulze, H. (1998) Technique to determine levelsof soya bean lectin in feed samples and digesta samples collected in vivo. In: Jansman,A.J.M., Hill, G.D., Huisman, J. and van der Poel, A.F.B. (eds) Recent Advances ofResearch in Antinutritional Factors in Legume Seeds. Proceedings of the Third Inter-national Workshop on ‘Antinutritional Factors (ANFs) in Legume Seeds’. WageningenPers, Wageningen, The Netherlands, pp. 29–32.





Näsi, M. (1991) Digestibility and protein utilization responses of soybean and rape seed mealto physical and enzymatic treatments in diets for growing pigs. Journal of AgriculturalScience, Finland 63, 465–474.

Nielsen, N.C. (1983) The chemistry of legume storage proteins. In: A Royal Society DiscussionMeeting on Seed Storage Proteins. The Royal Society, London, pp. 13–22.

NRC (1998) Nutrient Requirements of Swine. National Academic Press, Washington, DC,190pp.

Ogden, J.M. (1995) Feeding trial to evaluate the effects of in-feed enzyme inclusion upongrowth, food conversion ratio and carcass composition of pigs. MSc thesis, University ofBristol, UK.

Partridge, G.G. (1993) In feed enzymes and antibodies. Pig Veterinary Journal 31, 34–50.Pfirter, H.P., Hellal, H., Messikommer, R. and Wenk, C. (1993) Effect of a protease in broiler

feed containing crude or extruded lupins. In: Enzymes in Animal Nutrition, Proceedings ofthe 1st Symposium. Kartause Ittingen, Switzerland, pp. 283–286.

Pluske, J.R. and Lindemann, M.D. (1998) Maximising the response in pig and poultry dietscontaining vegetable proteins by enzyme supplementation. In: Lyons, T.P. and Jacques,K.A. (eds) Passport to the Year 2000, Biotechnology in the Feed Industry. Proceedingsof Alltech’s 14th Annual Symposium. Nottingham University Press, Nottingham, UK,pp. 375–392.

Rooke, J.A., Fraser, H., Shanks, M. and Morgan, A. (1996) The potential for improvingsoya-bean meal in diets for weaned piglets by protease treatment: comparison with otherprotein sources. British Society of Animal Science Winter Meeting. Scarborough, UK, Paper136.

Rooke, J.A., Slessor, M., Fraser, H. and Thomson, J.R. (1998) Growth performance and gutfunction of piglets weaned at four weeks of age and fed protease-treated soyabean meal.Animal Feed Science and Technology 70, 175–190.

Rotter, B.A., Marquardt, R.R. and Guenter, W. (1989) Enzymes in feed – they really can bemade to work. Feed Compounder 9, 32–35.

Schang, M.J. and Azcona, J.O. (1998) Performance of laying hens fed a corn–sunflowerdiet supplemented with enzymes. In: Lyons, T.P. and Jacques, K.A. (eds) Passport to theYear 2000, Biotechnology in the Feed Industry. Proceedings of Alltech’s 14th AnnualSymposium. Nottingham University Press, Nottingham, UK, pp. 405–410.

Schang, M.J., Azcona, J.O. and Arias, J.E. (1997) Effects of a soya enzyme supplement onperformance of broilers fed corn/soy or corn/soy/full-fat soy diets. Poultry Science 76(Suppl. 1), 132 (abstract).

Schulze, H., Savelkoul, F.H.M.G., Verstegen, M.W.A., van der Poel, A.F.B., Tamminga, S.and Groot Nibbelink, S. (1997) Nutritional evaluation of biologically treated whitekidney beans (Phaseolus vulgaris L.) in pigs: ileal and amino acid digestibility. Journal ofAnimal Science 75, 3187–3194.

Sefton, A.E. and Perdok, H. (1996) Effect of inclusion of Allzyme Vegpro in a broiler starterdiet. 12th Annual Symposium of Biotechnology in the Feed Industry, Lexington, Kentucky.

Spring, P., Wenk, C., Lemme, A., Bee, G. and Gerbert, S. (1998) Effect of an enzymecomplex targeting soybean meal on nutrient digestibility and growth performance inweanling piglets. 14th Annual Symposium on Biotechnology in the Feed Industry.Lexington, Kentucky. Supplement 1, enclosure code UL 5.5.

Stanley, V.G., Gray, C., Chukwu, H. and Thompson, D. (1996) Effects of enzyme(treatment) in enhancing the feeding value of cottonseed meal and soybean mealin broiler chick diets. 12th Annual Symposium on Biotechnology in the Feed Industry.Lexington, Kentucky. Supplement 1, enclosure code UL 2.2.





Swift, M.L., von Keyserlingk, M.A.G., Leslie, A. and Teltge, D. (1996) The effect of AllzymeVegpro supplementation and expander processing on the nutrient digestibility andgrowth of broilers. 12th Annual Symposium on Biotechnology in the Feed Industry.Lexington, Kentucky. Supplement 1, enclosure code UL 2.1.

Thacker, P.A. and Baas, T.C. (1996) Effects of gastric pH on the activity of exogenouspentosanase and the effect of pentosanase supplementation of the diet on the perfor-mance of growing–finishing pigs. Animal Feed Science and Technology 63, 187–200.

Thorpe, J. (1999) The effects of two in feed protease enzymes on the digestibility of soyaproteins in the early weaned pig. PhD thesis, University of Bristol, UK.

Thorpe, J., Patel, D. and Miller, B.G. (1997) Technique to determine levels of trypsininhibitor in gut samples collected in vivo. In: Laplace, J.P., Fevrier, C. and Barbeau, A.(eds) Digestive Physiology in Pigs. Proceedings of the 7th International Symposium onDigestive Physiology in Pigs, INRA, St Malo, France, pp. 382–385.

van der Poel, A.F.B. (1989) Effects of processing on antinutritional factors (ANF) andnutritional value of legume seeds for non-ruminant feeding. In: Huisman, J., van derPoel, A.F.B. and Liener, I.E. (eds) Recent Advance of Research in Antinutritional Factors inLegume Seeds. Proceedings of the First International Workshop on ‘AntinutritionalFactors (ANF) in Legume Seeds’. Pudoc, Wageningen, The Netherlands, pp. 213–227.

van der Poel, A.F.B. (1991) Effects of processing on bean (Phaseolus vulgaris L.) proteinquality. PhD thesis, Agricultural University Wageningen, Wageningen, TheNetherlands.

Viveros, A., Brenes, A., Elices, R. and Canales, R. (1993) Effect of enzyme addition (proteaseplus amylase and tannase) on the nutritive value of faba bean hulls in chickens. In:van der Poel, A.F.B., Huisman, J. and Saini, H.S. (eds) Recent Advances of Researchin Antinutritional Factors in Legume Seeds. Proceedings of the Second InternationalWorkshop on ‘Antinutritional Factors (ANFs) in Legume Seeds’. Wageningen Pers,Wageningen, The Netherlands, pp. 523–527.

Zamora, R.G. and Veum, T.L. (1979) Whole soybeans fermented with Aspergillus oryzae andRhizopus oligosporus for growing pigs. Journal of Animal Science 48(1), 63–68.





M. ChoctEnzyme Supplementation of Poultry Diets7

Introduction

Feed enzymes have come to be regarded by many nutritionists as being a necessary‘ingredient’ for formulating poultry rations. This has occurred mainly since the1990s, but the concept of enhancing animal performance using enzymes is by nomeans new. For example, in the mid 20th century, various preparations of amylasewere used in an attempt to overcome poor performance of chicks fed barley dietsby increasing the availability of starch (Hastings, 1946; Fry et al., 1957). The earlywork focused on the hydrolysis of specific substrates to their simple constituentsfor absorption (Fry et al., 1957; Moran et al., 1968), but this approach was notsuccessful. Appreciable advances have since been achieved in the use of enzymes inpoultry diets with a clear understanding of the target substrates and the developmentof microbiological technology. The prime example of this is the use of β-glucanasesin barley diets and xylanases (pentosanases) in rye or wheat diets.

Enhancing Bird Performance by Enzyme Supplementation

Barley

The macro-nutrient contents of barley and maize are very similar, but their nutritivevalue for poultry is vastly different. This led scientists to use various treatments,including enzyme supplementation (Hastings, 1946; Fry et al., 1957). These workersused an α-amylase, which significantly increased the liveweight gain and feedconversion efficiency (FCE) of chickens fed barley diets. It is now well establishedthat the starch in barley is totally digestible by the amylase secreted by chickens.Therefore the reported improvements with amylase supplementation were probablydue to the impurities in the enzymes used, i.e. the crude enzyme preparation



Enzyme Supplementation ofPoultry Diets Based on

Viscous CerealsM. CHOCT

School of Rural Science and Natural Resources,Department of Animal Science, The University of New England,

Armidale, NSW 2351, Australia

7



contained β-glucanase activity. β-Glucans are glucose polymers containing a mixtureof β1,3 and β1,4 linkages that make their physicochemical properties totally differentfrom cellulose that is a straight-chain glucose polymer with only β1,4 linkages. Barleycontains a high level of mixed-linked β-glucan (3–4%), which is responsible forits poor nutritive value in chickens (Burnett, 1966). Since this significant finding,there have been numerous studies on the use of enzymes, in particular β-glucanasesin barley-based poultry diets (Hesselman and Åman, 1986; Campbell et al., 1989),with outstanding success. The enzymes increase growth performance and feedconversion efficiency. Increases of up to 17% in liveweight gain (Broz and Frigg,1986) and 19% in feed conversion efficiency (Newman and Newman, 1987) havebeen reported for broiler chickens fed barley diets supplemented with β-glucanases.Barley also contains an appreciable amount of soluble NSP other than β-glucansand thus the majority of enzymes for barley diets have both β-glucanase andarabinoxylanase activities.

Rye

The poor feeding value of rye was reported more than 60 years ago (Halpin et al.,1936). In the search for an answer to the problem, Fernandez et al. (1973) extractedrye grain with water and freeze-dried the extract. When this extract was added to amaize-based diet, it depressed the growth of the birds and caused sticky droppings,whereas the water-extracted rye was markedly better than normal rye. Thiswater-extractable factor is now known to be the soluble arabinoxylan (Fengler andMarquardt, 1988a,b). Thus, addition of arabinoxylanases to rye-based broiler dietssignificantly improves the growth performance and feed conversion efficiency(Bedford et al., 1991; Bedford and Classen, 1992). Supplementation with increasinglevels (0.11, 0.22, 0.44 and 0.88 g kg−1) of an enzyme preparation havingarabinoxylanase and β-glucanase activities to a rye–wheat-based diet improved theweight gain of broilers up to 27% and FCE up to 10% (Pettersson and Åman, 1988,1989). Although enzymes always substantially improve the performance of birds fedrye diets, they do not seem to have as large an effect on the litter quality.

‘Low-ME’ wheat

Although the large variability (up to 4 MJ kg−1 dry matter) of the nutritive quality ofwheat was reported by numerous researchers (Sibbald and Slinger, 1962; Schumaierand McGinnis, 1967), the significance of the problem had not been widelyappreciated until recently. Connor et al. (1976) noticed that the AME value obtainedfor wheats was 7–25% lower than that of sorghum, and Payne (1976) postulated thatsome wheats may contain a ‘slightly toxic inhibitor’. Subsequently, two major studiesfound that approximately 25% of the Australian wheats had AME values below13 MJ kg−1 dry matter (Mollah et al., 1983; Rogel et al., 1987). When these‘low-ME’ wheats are included at above 50% in the ration, chickens have sticky and

146 M. Choct




watery droppings accompanied by poor growth and feed efficiency. At very highlevels of inclusion (up to 90% of the diet), young broilers can only utilize a verysmall portion of the energy (2.90 MJ kg−1) (Wiseman and McNab, 1998). It is nowgenerally conceded that the occurrence of low-ME wheats is due to an increased levelof NSP, in particular the arabinoxylan (Choct and Annison, 1990; Annison, 1991).Choct et al. (1995) demonstrated that supplementation of a low-ME wheat diet witha commercial glycanase preparation increased the AME by 24% and the FCE by25% in 3–4 week old broiler chickens.

Mechanisms of Action

The term non-starch polysaccharide (NSP) represents a diversity of compoundspossessing different physicochemical properties; thus their nutritional effects inpoultry are also diverse and, in some cases, extreme. It is, however, generallyconceded that the detrimental effect of NSP in viscous grains is associated withthe viscous nature of these polysaccharides, their physiological and morphologicaleffects on the digestive tract and the interaction with the microflora of the gut. Themechanisms include altered intestinal transit time and modification of the intestinalmucosa, as well as changes in hormonal regulation due to a varied rate of nutrientabsorption. To elucidate the action of NSP in poultry diets, Choct and Annison(1990, 1992a) isolated the cell wall polysaccharides from wheat and added them toa semi-purified diet. Two fractions were prepared. Fraction 1 was extracted usingwater only and hence named water-extractable arabinoxylans (WEP). Fraction 2 wasobtained by extraction with NaOH (0.2 mol l−1) and called alkali-extractablearabinoxylans (AEP), which were also soluble in water. Addition of these solublearabinoxylan preparations at an equivalent of 30 g pure arabinoxylans kg−1 dietdepressed weight gain by 24.6%, feed conversion ratio by 11.2%, DM digestibilityby 8.4% and AME of sorghum by 9.9%. The anti-nutritive effects of wheatarabinoxylans were also demonstrated in a practical type of broiler diet (Choctand Annison, 1992a). The arabinoxylans severely depressed bird performance andnutrient digestibilities when added at above 2% to the diet. Part of the data is shownin Tables 7.1 and 7.2.

The wheat arabinoxylans caused a general inhibition of nutrient digestionaffecting starch, fat and protein, which indicated that they acted in a similar manneras the anti-nutritive NSP of rye and barley. The arabinoxylans of rye have beenreported to depress fat, protein and DM retention (Ward and Marquardt, 1987;Fengler and Marquardt, 1988b). Depressed ileal digestibilities of starch and proteinhave been observed in chickens fed barley diets which had high levels of extractableviscous β-glucans (Hesselman and Åman, 1986). Variability in the experimental dataincreased as greater amounts of arabinoxylans were added to the diets. This mimicsthe low-AME wheat phenomenon. When broilers were fed low-AME wheats, greaterbetween-bird variation was seen than when normal wheats were fed (Rogel et al.,1987). Classen et al. (1988) demonstrated that supplementation of barley diets with

Enzyme Supplementation of Poultry Diets 147




β-glucanase reduced the variability from 11.9 to 3.3% for weight gain and from 5.2to 2.7% for FCR in broilers.

Viscous nature of soluble NSP

Increased bulk and viscosity of the intestinal contents decrease the rate of diffusion ofsubstrates and digestive enzymes and hinder their effective interaction at the mucosalsurface (Edwards et al., 1988; Ikegami et al., 1990). Studies in vitro with guar gumshowed that soluble, indigestible polysaccharides interact with the glycocalyx of the

148 M. Choct


Diet1Arabinoxylan2

(g kg−1 DM)AME3

(MJ kg−1 DM)NR

(g d−1)FCR

(g : g)WG

(g 6d−1)FI

(g 6d−1)

ControlWEP(20)AEP(5)AEP(10)AEP(25)AEP(40)

25.943.830.834.948.065.7

15.05a

13.90b

15.00a

14.70a

13.34b

12.48c

3.05a

2.79a

2.89a

2.86a

2.42b

1.96c

1.91ab

2.08ab

1.94ab

1.95ab

2.49bc

2.70cb

349ab

312ab

348ab

352ab

268bc

216cb

661a

637a

675a

686a

640a

552b

SE (pooled) 0.22 0.13b 0.12bb 15b 19

1Values in parentheses are added arabinoxylans (g kg−1 DM).2Determined values for total pentosans in diets.3Columns sharing the same superscripts are not significantly different (P > 0.05).WEP = water extractable arabinoxylans; AEP = alkali extractable arabinoxylans.

Table 7.1. Effects of soluble arabinoxylans from wheat on AME, nitrogen retention (NR), feedconversion ratio (FCR), weight gain (WG) and feed intake (FI) in broilers (n = 8). (Data from Choctand Annison, 1992a.)

Diet1 Arabinoxylan2 Starch3 Protein Lipid

ControlWEP(20)AEP(5)AEP(10)AEP(25)AEP(40)

25.943.830.834.948.065.7

0.96ab

0.91bb

0.96ab

0.95ab

0.92ab

0.82cb

0.75ab

0.70ab

0.75ab

0.73ab

0.69ab

0.61bb

0.93ab

0.87ab

0.93ab

0.92ab

0.76ab

0.69bb

SE (pooled) 0.02bb 0.03bb 0.06bb

1Values in parentheses are added arabinoxylans (g kg−1 DM).2Determined values for total pentosans in diets.3Columns sharing the same superscripts are not significantly different (P > 0.05).WEP = water extractable arabinoxylans; AEP = alkali extractable arabinoxylans.

Table 7.2. Effects of wheat arabinoxylans on ileal digestibility coefficients of starch, protein andlipid (n = 3). (Data from Choct and Annison, 1992a.)



intestinal brush border, producing a thickening of the rate-limiting unstirred waterlayer, which results in decreased nutrient absorption (Johnson and Gee, 1981).Germination of barleys can lower the viscosity of their aqueous extracts and suchbarleys are of better nutritive value in poultry compared with normal barleys (Fengleret al., 1990). Barleys with high extract viscosity are more detrimental to chickensthan those with low extract viscosity (Campbell et al., 1989). These results lendsupport to the hypothesis that viscosity is involved in the anti-nutritive effect ofNSP in poultry diets. Choct and Annison (1992b) compared the effects of a highmolecular weight NSP isolate (Intact-NSP) (854 g arabinoxylans and 42 g glucankg−1 DM; molecular weight = 758,000) and a partially depolymerized NSP isolate(Depol-NSP; 194,000) (Annison et al., 1992), and pentoses on broiler performanceand nutrient digestibility (Table 7.3). When included at a level of 30 g kg−1 diet, thepartially depolymerized NSP increased digesta viscosity significantly compared withpentose sugars added at the same level, but it was lower than that in birds fed intactNSP. Bird performance was not significantly affected by the depolymerized NSP,indicating that birds can tolerate small increases in digesta viscosity without adetrimental effect on performance.

The anti-nutritive effects of the arabinoxylans appear largely dependent on thepolymeric nature of the polysaccharides and their viscous property. This is supportedby studies demonstrating that supplementation with enzymes capable of degradingβ-glucans and arabinoxylans increases the nutritive value of barley and rye in poultrydiets and this coincides with a significant decrease in the digesta viscosity of thechickens (White et al., 1981; Bedford et al., 1991).

Watery and sticky droppings alone may not be a sufficient indicator of theanti-nutritive effect of the arabinoxylans in poultry, as excreta from birds fed the dietcontaining the depolymerized arabinoxylans were watery and sticky, and those frombirds given the diet containing pentose sugars also appeared more moist comparedwith controls. In the rat, dietary components that are not completely digested orabsorbed in the small intestine give rise to an increased amount of osmotically activematerials in the gut content. If these materials are utilized by the hindgut bacteria,



DietLevel added

(g kg−1)2Digestaviscosity

WG(g wk−1)3

FCR(g : g)

FI(g wk−1)

AME(MJ kg−1 DM)

ControlIntact-NSPPentosesDepol-NSP1

0303030

1.2a

3.0c

1.2a

2.2b

430ac

325bc

394ac

404ac

1.589ac

1.960bc

1.695ac

1.649ac

681622658661

16.13a

14.53b

16.23a

15.74a

1Partially depolymerized NSP.2Levels added at equivalent to pure NSP.3Columns sharing the same superscripts are not significantly different (P > 0.05).

Table 7.3. Effects of high molecular weight NSP, partially depolymerized NSP and pentoses onweight gain (WG), feed conversion ratio (FCR), feed intake (FI) and AME in broilers (n = 8). (FromChoct and Annison, 1992b.)



production of low molecular weight metabolites that are not readily absorbed mayresult. This can lead to further accumulation of osmotically active substances whichattract a large amount of water. This may well relate to the inefficient utilization offive-carbon sugars by chickens (Wagh and Waibel, 1966). The significant improve-ment in performance of birds fed barley or rye supplemented with β-glucanase orarabinoxylanase is not due to a complete hydrolysis of the polysaccharides and asubsequent absorption of the released sugars (White et al., 1981), but is due to thedepolymerization of the polysaccharides into smaller polymers which do not greatlyelevate the viscosity of digesta (De Silva et al., 1983).

Modification of secretory response of the gut

Viscous polysaccharides cause physiological and morphological changes to thedigestive system of rats, pigs and humans (Brown et al., 1979; Cassidy et al., 1981;Morgan et al., 1985; Ide et al., 1989; Low, 1989). The endogenous secretion ofwater, proteins, electrolytes and lipids can be increased markedly by NSPsupplementation of the diet (Low, 1989). The metabolic cost of such processes canbe considerable. Prolonged consumption of diets containing viscous polysaccharidesis associated with significant adaptive changes in the digestive system in rats (Ikegamiet al., 1990). The changes in the gastrointestinal tract are characterized by enlarge-ment of digestive organs and increased secretion of digestive juices, accompanied bydecreases in nutrient digestion. The depressed apparent ileal protein digestibilitycaused by the wheat arabinoxylans, therefore, may be due to an inhibition of proteinbreakdown and/or a reduction in amino acid absorption. It may also result from anincrease in the secretion of endogenous proteins, which can be derived from gutsecretions and losses of intestinal cells. In the following study (Angkanaporn et al.,1994), the effect of soluble wheat arabinoxylans on endogenous protein (sum ofamino acids) secretions, in comparison to that of pure cellulose and an inert nutrientdiluent (polythene powder), was investigated. Endogenous and exogenous aminoacids were distinguished using the homoarginine marker technique (Hagemeisterand Erbersdobler, 1985). The apparent protein digestibility was depressed to asimilar extent by addition of wheat arabinoxylans at levels of 15 and 30 g kg−1, butthe true protein digestibility was significantly inhibited only at the higher level ofinclusion. This indicates that the anti-nutritive effect of wheat arabinoxylans on theapparent protein digestibility is mainly due to increased endogenous secretionsof amino acids at low levels of inclusion, while at high levels a direct inhibition ofprotein breakdown and/or absorption occurs. Addition of cellulose and polythenedid not have an effect although the levels of inclusion were much higher than that ofwheat arabinoxylans (Table 7.4).

The biological effects of NSP differ considerably due to their diverse physio-chemical properties. Southon et al. (1985) reported that a diet containing 75 gnon-cellulosic polysaccharides kg−1 and 24 g cellulose kg−1 fed to rats induced higherrates of protein synthesis in the jejunum and ileum, and more rapid mucosal celldivision than rats given a semi-purified diet containing cellulose as the only source of

150 M. Choct




NSP. The lack of action of cellulose on the endogenous amino acid secretions wasfurther demonstrated in chickens by Parsons (1984), who found that when anitrogen-free diet containing 400 g raw potato starch kg−1 and 100 g pectin kg−1

was fed to laying hens, the total endogenous amino acid secretion was increased by40%; whereas the diet containing 500 g cellulose kg−1 elicited little effect. Theseobservations indicate that the action of NSP in modification of the physiologicalfunctions of the gastrointestinal tract is not a mere mechanical stimulation of themucosa. It is possible that the soluble NSPs interact with the gut wall in a way thatmodifies the endocrine regulation.

Encapsulated nutrients

Although the insoluble NSPs have mainly been regarded as a nutrient diluent in thediet, they can also affect digesta transit time and gut motility. Another facet of therole of insoluble NSP in poultry diets that is worthy of reiteration is the possibilityof them acting as a physical barrier to digestive enzymes, such as amylase andproteases, thus reducing their efficient digestion in the upper part of the gut. Thisis probably particularly important in non-viscous grains, such as sorghum. There isevidence that enzymes with affinity for insoluble NSP can also elicit a positiveresponse in growth performance of broilers (Cowan, 1995; Choct, 1998). Thisindicates that breakdown of cell wall matrix, especially the insoluble components,may facilitate easier access of digestive enzymes to their substrates within the shortfeed transit time in birds. Wiseman and McNab (1998) also showed that the rate ofin vitro starch digestion correlates closely with the AME values in wheat, indicatingthat the accessibility of amylotytic enzymes to starch granules is vastly different.

NSP and gut microflora

The gut harbours a highly evolved and complex microbial ecosystem containing avast number of diverse populations. For example, microbes make up approximately



PentosanCellulose92 g kg−1

Polythene92 g kg−1Diet Control 15 g kg−1 30 g kg−1

APTPEP

0.770a

0.969a

30.2a

0.595b

0.946a

53.9b

0.600b

0.885b

40.2a

0.701a

0.955a

37.4a

0.695a

0.966a

34.9a

Rows sharing the same superscripts are not significantly different (P > 0.05).

Table 7.4. Effects of wheat pentosans, cellulose and polythene on apparent (AP) and trueprotein (TP) digestibility coefficients and endogenous protein secretions (EP); g kg−1 DM intake(n = 2). (From Angkanaporn et al., 1994.)



600 g kg−1 of the wet weight of poultry excreta. The proper feeding of poultry shouldtherefore consider the provision of ‘correct’ substrates for the microflora to keepit stable. The consequences of altered rate of nutrient digestion in the gut maybe manifested in the number and type of microorganisms present in the gut(Vahjen et al., 1998). An increased amount of NSP in the digesta has beendemonstrated to influence the gut microflora in a negative manner to birds (Choctet al., 1996). The overall digestibilities of protein and long-chain fatty acids,especially the saturated ones, were more severely depressed in intact birds than incaecectomized birds when wheat arabinoxylans were added to the diets (Choct et al.,1992). Bile salts are essential for emulsification of fats and activation of lipase (Tso,1985), especially for saturated fatty acids (Johnston, 1977) as opposed to short-chainand unsaturated fatty acids, which are more easily absorbed in the absence of bileacids (Garrett and Young, 1975). Excessive deconjugation of bile salts is a possibledeleterious effect of gut microflora (Feighner and Dashkevicz, 1988) and it maybe exacerbated by the presence of large amounts of undigested carbohydratesin the hindgut. This is supported in studies by Campbell et al. (1983) wheresupplementation of the conjugated bile salt, sodium taurocholate, to chickens fed ryediets significantly improved the fat retention. Furthermore, utilization of nutrientsthrough microbial conversion of digestible carbohydrates, such as starch, to VFAs isnot efficient compared with a direct absorption of glucose released from enzymaticdigestion (Müller et al., 1989; Carré et al., 1995). In a recent study, Choct et al.(1996) demonstrated that addition of soluble NSP in broiler chicken dietssignificantly elevated fermentation in the small intestine. Subsequent in vivodepolymerization of the soluble NSP using glycanases almost totally overcamethis problem.

Another factor that may contribute to the negative effect of a high microbialload in the small intestine is increased turnover of intestinal cells. According toLeBond and Walker (1956), a 100 g rat gaining 5 g a day synthesizes 1 g mucosalcells daily, which represents a 20% additional tissue synthesis without showingup as weight gain. Extrapolating this to a 2 kg bird gaining 60 g daily, the birdwould synthesize 12 g of mucosal tissue to maintain the integrity of its smallintestine. Increased microbial load can exacerbate this loss (Abrams et al., 1963;Lesher et al., 1964), since some of their fermentation products, e.g. putrescine,have been shown to enhance small intestinal and colonic mucosal growth ratessignificantly (Seidel et al., 1985; Osborne and Seidel, 1989). Furthermore,antibiotics can significantly improve performance of birds fed high-NSP diets(MacAuliffe and McGinnis, 1971; Misir and Marquardt, 1978a,b). Whatever thereason, excess fermentation in the small intestine of the chicken may be detrimentalto nutrient digestion and absorption and enzyme supplementation seemed toalleviate this problem. Perhaps reducing the viscosity of the digesta in the smallintestine hastens digesta passage and nutrient digestion rate (through removal ofthe diffusional constraint of viscous gums), thereby giving less substrate and timefor the fermentative organisms to proliferate. This may in turn restore thenormal and efficient digestion (enzymatically) of starch and protein in the smallintestine.

152 M. Choct




Predicting Enzyme Response

The concept of predicting the effect of enzymes in a particular feed is attractivebecause the producer could then adjust the amount of enzymes to be added or adjustthe nutrient specifications in diet formulations. Although it is extremely difficult topredict the specific effects of enzymes accurately in practice, some approaches haveproved to provide useful estimates of enzyme responses in viscous grains such as rye,barley and wheat.

The NSP content of the diet

The negative correlation between NSP content of the diet and its nutritive value hasbeen clearly demonstrated in poultry (Choct and Annison, 1990; Annison, 1991), inpigs (King and Taverner, 1975) and in pet dogs and cats (Earle et al., 1998). Thisindicates that by measuring the NSP content of a diet, it is possible to predict theamount of enzymes to add to the diet. However, the practicality of such an approachis questionable. Firstly, a least-cost diet contains a number of plant ingredients,which contain different forms of NSP. It is no way to know the amounts and types ofdifferent NSP substrates present in a diet by a quantitative measurement of the NSPlevel. Thus the responses to various types of enzyme activities added to the diet aredifficult to predict. Secondly, the NSP assay is tedious and complicated and, hence, isnot ideal for use in quality control laboratories. Thirdly, the anti-nutritive effect ofsoluble NSP is related, to a large extent, to the viscous nature of the polymers, whichin turn depends on their molecular sizes and structures. Again a qualitative or acombination of qualitative and quantitative measurements is required.

Gut viscosity

The relationship between gut viscosity and the nutritive value of barley in poultrywas first described by Burnett (1966). Only during the past 10 years has the impor-tance of his work been recognized and a wealth of information on viscosity of the gutcontents and its effect on nutrient digestion and absorption has since emerged (Bed-ford et al., 1991; Bedford and Classen, 1992; Choct and Annison, 1992a; Steenfeldtet al., 1998). Whether gut viscosity can be used to predict enzyme response dependson a number of factors, including the section of the intestine where the digesta issampled, the age of the bird, and antibiotic growth promoters in the diet. Firstly,different sections of the gut will influence the viscosity value, because nutrients aredigested and absorbed as digesta moves down the gut, thus leaving the indigestibleportion (mainly NSP) to accumulate. On a relative basis, the soluble NSP contentper ml of digesta supernatant therefore increases; secondly, digesta viscosity decreasesas the bird gets older (Dusel et al., 1998; Steenfeldt et al., 1998). This also relates tothe finding that older birds cope better with low-ME wheats (Rogel et al., 1987) orwith barley (Salih et al., 1991). In addition, the relationship between gut viscosity





and bird performance is not always apparent (Wiseman and McNab, 1998). Forexample, Hughes and Zviedrans (1999) reported that two groups of birds fed wheathad ileal viscosity value (mPa s) of 12.5 and 49, but the AME contents (MJ kg−1

DM) for the wheat were 10.6 and 12.0. This does not necessarily mean that theviscosity has no effect on nutrient digestion and absorption; rather, it highlights thecomplex nature of the interaction between digesta viscosity and the gut microflora. Itis hypothesized that as the bird ages, its gut microflora becomes more established andit can adapt to the environment – for example, to cope with the viscous digesta betterby producing small amounts of NSP-degrading enzymes (Choct and Kocher, 2000).This highlights the third point: that addition of antibiotic growth promoters in thediet can have a major effect on gut viscosity and hence enzyme response. Fourthly, asdiscussed earlier, removal of cell wall barriers using enzymes with an affinity for theinsoluble NSP can also benefit the bird by releasing encapsulated nutrients. Theresponse of such an action cannot be predicted from gut viscosity measurements.

Generally, gut viscosity values in birds fed viscous grains can be used to predictthe response to enzyme supplementation with strong viscosity-reducing propertywithin a grain type under set conditions. However, measuring gut viscosity is neitherrapid nor cheap.

Extract viscosity

A ground sample of ingredients or compound feed can be extracted in water or in abuffer to measure extract viscosity. The viscosity of the extract comes predominantlyfrom soluble NSP in the feed (Izydorczyk et al., 1991; Saulnier et al., 1995). Sincethe soluble NSP content of a diet is related to its nutritive value, it is possiblethat extract viscosity can be used to predict enzyme response. Rotter et al. (1989)demonstrated that the growth responses of chickens fed barley diets were accuratelypredicted by an extract viscosity method. Choct et al. (1993) developed an extractviscosity assay to distinguish wheats with very low AME from normal samples. Moresystematic studies with promising outcomes have since been reported on the use ofextract viscosity as a predictor of nutritive value for poultry (Carré and Melcion,1995; Choct and Hughes, 1997; Dusel et al., 1997, 1998; Wiseman and McNab,1998). The extract viscosity method is simple and rapid and, more importantly, itgives qualitative information on the characteristics of the NSP. However, since theextract viscosity assay is based on the amount of soluble NSP content of the ingredi-ent, it can only be used for predicting the response of viscosity-reducing enzymes. Inaddition, the reliability of the assay may be influenced by the extraction method andthe endogenous enzyme activities in the diet or the individual ingredient.

Rate of starch digestion

Wiseman and McNab (1998) used an in vitro method to measure the rate of starchdigestion by amylase. They demonstrated a close correlation between in vitro and

154 M. Choct




in vivo starch digestion for wheat (R2 = 0.65) to distinguish between wheats withhigh and low metabolizable energy values in poultry. With validation, the assay maybe used to characterize wheat for poultry diets; thus it offers potential as a means ofpredicting enzyme response.

Conclusion

The use of xylanase for wheat and β-glucanase for barley in poultry is a well acceptedpractice today. Generally, most of the enzymes effectively depolymerize the solubleNSP into smaller polymers, though some products with affinity for both soluble andinsoluble NSP are also used. The soluble NSPs elicit anti-nutritive activities inpoultry diets, which are closely related to their polymeric nature and ability toincrease digesta viscosity. Extract viscosity has great potential to be used as a simpleand rapid method to predict enzyme response in poultry diets. Other assays such asthe in vitro starch digestibility measurement may prove useful. Further studies onNSPs need to address their effect on nutrient utilization in terms of relative efficiencyof enzymatic digestion vs. fermentation, and muscle growth vs. gut cell turnover.

References

Abrams, G.D., Bauer, H. and Sprinz, H. (1963) Influence of the normal flora on mucosal cellmorphology and cellular renewal. Laboratory Investigations 12, 355–363.

Angkanaporn, K., Choct, M., Bryden, W.L., Annison, E.F. and Annison, G. (1994) Effects ofwheat pentosans on endogenous amino acid losses in chickens. Journal of the Science ofFood and Agriculture 66, 399–404.

Annison, G. (1991) Relationship between the levels of soluble nonstarch polysaccharidesand the apparent metabolizable energy of wheats assayed in broiler chickens. Journal ofAgricultural and Food Chemistry 39, 1252–1256.

Annison, G., Choct, M. and Cheetham, N.W. (1992) Analysis of arabinoxylans from alarge-scale isolation. Carbohydrate Polymers 19, 151–159.

Bedford, M.R. and Classen, H.L. (1992) Reduction of intestinal viscosity through manipula-tion of dietary rye and arabinoxylanase concentration is effected through changes inthe carbohydrate composition of the intestinal aqueous phase and results in improvedgrowth rate and food conversion efficiency of broiler chicks. Journal of Nutrition 122,560–569.

Bedford, M.R., Classen, H.L. and Campbell, G.L. (1991) The effect of pelleting, salt, andpentosanase on the viscosity of intestinal contents and the performance of broilers fed rye.Poultry Science 70, 1571–1577.

Brown, R.C., Kelleher, J. and Losowsky, M.S. (1979) The effect of pectin on the structure andfunction of the rat small intestine. British Journal of Nutrition 42, 357–365.

Broz, J. and Frigg, M. (1986) Effects of beta-glucanase on the feeding value of broiler dietsbased on barley or oats. Archiv für Geflügelkunde 50, 41–47.

Burnett, G.S. (1966) Studies of viscosity as the probable factor involved in the improvement ofcertain barleys for chickens by enzyme supplementation. British Poultry Science 7, 55–75.





Campbell, G.L., Campbell, L.D. and Classen, H.L. (1983) Utilisation of rye by chickens:effect of microbial status, diet gamma irradiation and sodium taurocholate supple-mentation. British Poultry Science 24, 191–203.

Campbell, G.L., Rossnagel, B.F., Classen, H.L. and Thacker, P.A. (1989) Genotypic andenvironmental differences in extract viscosity of barley and their relationship to itsnutritive value for broiler chickens. Animal Feed Science and Technology 26, 221–230.

Carré, B. and Melcion, J.P. (1995) Effects of feed viscosity on water excretion in meat-turkeypoults. In: Rowe, J.B. and Nolan, J.V. (eds) Recent Advances in Animal Nutrition inAustralia. University of New England, Armidale, Australia, pp. 1–6.

Carré, B., Gomez, J. and Chagneau, A.M. (1995) Contribution of oligosaccharide andpolysaccharide digestion, and excreta losses of lactic acid and short chain fatty acids, todietary metabolisable energy values in broiler chickens and adult cockerels. British PoultryJournal 36, 611–629.

Cassidy, M.M., Lightfoot, F.G., Grau, L.E., Story, J.A., Krichevsky, D. and Vahouny, G.V.(1981) Effect of chronic intake of dietary fibers on the ultrastructural topography of ratjejunum and colon: a scanning electron microscopy study. American Journal of ClinicalNutrition 34, 218–228.

Choct, M. (1998) The effect of different xylanases on carbohydrate digestion and viscosityalong the intestinal tract in broilers. Proceedings of the Australian Poultry ScienceSymposium 10, 111–115.

Choct, M. and Annison, G. (1990) Anti-nutritive activity of wheat pentosans in broiler diets.British Poultry Science 31, 811–822.

Choct, M. and Annison, G. (1992a) The inhibition of nutrient digestion by wheat pentosans.British Journal of Nutrition 67, 123–132.

Choct, M. and Annison, G. (1992b) Anti-nutritive effect of wheat pentosans in broilerchickens: roles of viscosity and gut microflora. British Poultry Science 33, 821–834.

Choct, M. and Hughes, R.J. (1997) The nutritive value of new season grains for poultry.In: Corbett, J.L., Choct, M., Nolan, J.V. and Rowe, J.B. (eds) Recent Advances in AnimalNutrition in Australia. pp. 146–150.

Choct, M. and Kocher, A. (2000) Excreta viscosity as an indicator of microbial enzymeactivity in the hindgut and as a predictor of between-bird variation in AME in broilers.Proceedings of the Australian Poultry Science Symposium 12, 211.

Choct, M., Annison, G. and Trimble, R.P. (1992) Soluble wheat pentosans exhibit differentanti-nutritive activities in intact and cecectomized broiler chickens. Journal of Nutrition122, 2457–2465.

Choct, M., Annison, G. and Trimble, R.P. (1993) Extract viscosity as a predictor of thenutritive quality of wheat in poultry. Proceedings of the Australian Poultry ScienceSymposium 5, 78.

Choct, M., Hughes, R.J., Trimble, R.P., Angkanaporn, K. and Annison, G. (1995)Non-starch polysaccharide-degrading enzymes increase the performance of broilerchickens fed wheat of low apparent metabolizable energy. Journal of Nutrition 125,485–492.

Choct, M., Hughes, R.J., Wang, J., Bedford, M.R., Morgan, A.J. and Annison, G. (1996)Increased small intestinal fermentation is partly responsible for the anti-nutritive activityof non-starch polysaccharides in chickens. British Poultry Science 37, 609–621.

Classen, H.L., Campbell, G.L. and Grootwassink, J.W.D. (1988) Improved feeding valueof Saskatchewan-grown barley for chickens with dietary enzyme supplementation.Canadian Journal of Animal Science 68, 1253–1259.

156 M. Choct




Connor, J.K., Neil, A.R. and Barram, K.M. (1976) The metabolizable energy content forthe chicken of maize and sorghum grain hybrids grown at several geographical regions.Australian Journal of Experimental Agriculture and Animal Husbandry 16, 699–703.

Cowan, W.D. (1995) The relevance of intestinal viscosity on performance of practical broilerdiets. Proceedings of the Australian Poultry Science Symposium 7, 116–120.

De Silva, S., Hesselman, K. and Åman, P. (1983) Effects of water and β-glucanase treatmenton non-starch polysaccharides in endosperm of low and high viscous barley. SwedishJournal of Agricultural Research 13, 211–219.

Dusel, G., Kluge, H., Glaser, K., Simon, O., Hartmann, G., Lengerken, J. and Jeroch, H.(1997) An investigation into the variability of extract viscosity of wheat – relationshipwith the content of non-starch-polysaccharide fractions and metabolisable energy forbroiler chickens. Archiv Tierernahr. 50, 121–135.

Dusel, G., Kluge, H. and Jeroch, H. (1998) Xylanase supplementation of wheat-based rationsfor broilers: influence of wheat characteristics. Journal of Applied Poultry Science 7,119–131.

Earle, K.E., Opitz, B., Kienzle, E., Smith, P.M. and Maskell, I.E. (1998) Effect of fiber ondigestibility of organic matter and energy in pet foods. www.waltham.com/vets/pubs/vp02_14.htm

Edwards, C.A., Johnson, I.T. and Read, N.W. (1988) Do viscous polysaccharides slowabsorption by inhibiting diffusion or convection? European Journal of Clinical Nutrition42, 306–312.

Feighner, S.D. and Dashkevicz, M.P. (1988) Effect of dietary carbohydrates on bacterialcholyltaurine hydrolase in poultry intestinal homogenates. Applied and EnvironmentalMicrobiology 54, 337–342.

Fengler, A.I. and Marquardt, R.R. (1988a) Water-soluble arabinoxylans from rye. I. Isolation,partial purification, and characterisation. Cereal Chemistry 65, 291–297.

Fengler, A.I. and Marquardt, R.R. (1988b) Water-soluble arabinoxylans from rye. II. Effectsof rate of dialysis and on the retention of nutrients by the chick. Cereal Chemistry 65,298–302.

Fengler, A.I., Aherne, F.X. and Robblee, A.R. (1990) Influence of germination of cereals onviscosity of their aqueous extracts and nutritive value. Animal Feed Science and Technology28, 243–253.

Fernandez, R., Lucas, E. and McGinnis, J. (1973) Fractionation of a chick growth depressingfactor from rye. Poultry Science 52, 2252–2259.

Fry, R.E., Allred, J.B., Jensen, L.S. and McGinnis, J. (1957) Influence of cereal grain compo-nents of the diet on the response of chicks and poults to dietary enzyme supplements.Poultry Science 36, 1120 (abstract).

Garrett, R.L. and Young, R.J. (1975) Effect of micelle formation on the absorption of neutralfat and fatty acids by the chicken. Journal of Nutrition 105, 827–838.

Hagemeister, E. and Erbersdobler, H. (1985) Chemical labelling of dietary protein bytransformation of lysine to homoarginine: a new technique to follow intestinal digestionand absorption. Proceedings of the Nutrition Society 44, 133a.

Halpin, J.G., Holmes, C.E. and Hart, E.B. (1936) Rye as a feed for poultry. Poultry Science 15,3–8.

Hastings, W.H. (1946) Enzyme supplements for poultry feeds. Poultry Science 25, 584–586.Hesselman, K. and Åman, P. (1986) The effect of β-glucanase on the utilization of starch and

nitrogen by broiler chickens fed on barley of low or high viscosity. Animal Feed Scienceand Technology 15, 83–93.





Hughes, R.J. and Zviedrans, P. (1999) Influence of dietary inclusion rate of wheat on AME,digesta viscosity and enzyme response. Proceedings of the Australian Poultry ScienceSymposium 11, 101–104.

Ide, T., Horii, M., Kawashima, K. and Yamamota, T. (1989) Bile acid conjugation andhepatic taurine concentration in rats fed on pectin. British Journal of Nutrition 62,539–550.

Ikegami, S., Tsuchihashi, F., Harada, H., Tsuchihashi, N., Nishide, E. and Innami, S. (1990)Effect of viscous indigestible polysaccharides on pancreatic-biliary secretion and digestiveorgans in rats. Journal of Nutrition 120, 353–360.

Izydorczyk, M., Biliaderis, C.G. and Bushuk, W. (1991) Comparison of the structure andcomposition of water-soluble pentosans from different wheat varieties. Cereal Chemistry68, 139–144.

Johnson, I.T. and Gee, M. (1981) Effect of gel-forming gums on the intestinal unstirred layerand sugar transport in vitro. Gut 22, 398–403.

Johnston, J.M. (1977) Gastrointestinal tissue. In: Snyder, F. (ed.) Metabolism in Mammals.Plenum Press, New York, pp. 151–187.

King, R.H. and Taverner, M.R. (1975) Prediction of digestible energy in pig diets fromanalyses of fibre contents. Animal Production 21, 275–284.

Lebond, C.P. and Walker, B.E. (1956) Renewal of cell populations. Physiology Reviews 36,255–276.

Lesher, S., Wallbury, H.E. Jr and Sacher, G.A. (1964) Generation cycle in the duodenal cryptcells of germ-free and conventional mice. Nature 202, 884–886.

Low, A.G. (1989) Secretory response of the pig gut to non-starch polysaccharides. Animal FeedScience and Technology 23, 55–65.

MacAuliffe, T. and McGinnis, J. (1971) Effect of antibiotic supplements to diets containingrye on chick growth. Poultry Science 50, 1130–1134.

Misir, R. and Marquardt, R.R. (1978a) Factors affecting rye (Secale cereale L.) utilizationin growing chicks. I. The influence of rye level, ergot and penicillin supplementation.Canadian Journal of Animal Science 58, 691–701.

Misir, R. and Marquardt, R.R. (1978b) Factors affecting rye (Secale cereale L.) utilisation ingrowing chicks. II. The influence of protein type, protein level and penicillin. CanadianJournal of Animal Science 58, 703–715.

Mollah, Y., Bryden, W.L., Wallis, I.R., Balnave, D. and Annison, E.F. (1983) Studies on lowmetabolisable energy wheats for poultry using conventional and rapid assay proceduresand the effects of processing. British Poultry Science 24, 81–89.

Moran, E.T. Jr, Lall, S.P. and Summers, J.D. (1968) The feeding value of rye for the growingchick: effect of enzyme supplements, antibiotics, autoclaving and geographical area ofproduction. Poultry Science 48, 939–949.

Morgan, L.M., Tredger, J.A., Madden, A., Wasowski, P.K. and Marks, V. (1985) The effectof guar gum on carbohydrate-, fat- and protein-stimulated gut hormone secretion:modification of postprandial gastric inhibitory polypeptide and gastric responses. BritishJournal of Nutrition 53, 467–475.

Müller, H.L., Kirchgessner, M. and Roth, F.X. (1989) Energy utilisation of intracaicallyinfused carbohydrates and casein in sows. In: van der Honing, Y. and Close, W.H. (eds)Energy Metabolism of Farm Animals. Pudoc, Wageningen, The Netherlands, pp.123–126.

Newman, R.K. and Newman, C.W. (1987) Beta-glucanase effect on the performance ofbroiler chicks fed covered and hulless barley isotypes having normal and waxy starch.Nutrition Reports International 36, 693–699.

158 M. Choct




Osborne, D.L. and Seidel, E.R. (1989) Microflora derived polymines modulate obstructioninduced colonic mucosal hypertrophy. American Journal of Physiology 256,G1049–G1057.

Parsons, C.M. (1984) Influence of caecectomy and source of dietary fibre or starch onexcretion of endogenous amino acids by laying hens. British Journal of Nutrition 51,541–548.

Payne, C.G. (1976) Nutritional syndromes of poultry in relation to wheat-based diets. In:Haresign, W., Swan, H. and Lewis, D. (eds) Nutrition and the Climatic Environment.Butterworths, London, pp. 155–173.

Pettersson, D. and Åman, P. (1988) Effects of enzyme supplementation of diets based onwheat, rye or triticale on their productive value for broiler chickens. Animal Feed Scienceand Technology 20, 313–324.

Pettersson, D. and Åman, P. (1989) Enzyme supplementation of a poultry diet containing ryeand wheat. British Journal of Nutrition 62, 139–149.

Rogel, A.M., Annison, E.F., Bryden, W.L. and Balnave, D. (1987) The digestion of wheatstarch in broiler chickens. Australian Journal of Agricultural Research 38, 639–649.

Rotter, B.A., Marquardt, R.R., Guenter, W., Biliaderis, C. and Newman, C.W. (1989) Invitro viscosity measurements of barley extracts as predictors of growth responses in chicksfed barley-based diets supplemented with a fungal enzyme preparation. Canadian Journalof Animal Science 69, 431–439.

Salih, M.E., Classen, H.L. and Campbell, G.L. (1991) Response of chickens fed on hull-lessbarley to dietary α-glucanase at different ages. Animal Feed Science and Technology 33,139–149.

Saulnier, L., Peneau, N. and Thibault, J.F. (1995) Variability in grain extract viscosityand water soluble arabinoxylan content in wheat. Journal of Cereal Science 22,259–264.

Schumaier, G. and McGinnis, J. (1967) Metabolizable energy values of wheat and someby-product feedstuffs for growing chicks. Poultry Science 46, 79–82.

Seidel, E.R., Haddox, M.K. and Johnson, L.R. (1985) Ileal mucosal growth duringintraluminal infusion of ethylamine or putrescine. American Journal of Physiology 2498,G434–G438.

Sibbald, I.R. and Slinger, S.J. (1962) The metabolizable energy of materials fed to growingchicks. Poultry Science 41, 1612–1613.

Southon, S., Livesey, G., Gee, J.M. and Johnson, I.T. (1985) Differences in intestinal proteinsynthesis and cellular proliferation in well-nourished rats consuming conventionallaboratory diets. British Journal of Nutrition 53, 87–95.

Steenfeldt, S., Hammershoj, M., Mullertz, A. and Jensen, J.F. (1998) Enzyme supple-mentation of wheat-based diets for broilers. 2. Effect on apparent metabolisable energycontent and nutrient digestibility. Animal Feed Science and Technology 75, 45–64.

Tso, P. (1985) Gastrointestinal digestion and absorption of lipid. In: Paoletti, R. andKritchevsky, D. (eds) Advances in Lipid Research Vol. 21. Academic Press, Orlando,Florida, pp. 132–186.

Vahjen, W., Gläser, K., Schäfer, K. and Simon, O. (1998) Influence of xylanase-supplementedfeed on the development of selected bacterial groups in the intestinal tract of broilerchicks. Journal of Agricultural Science 130, 489–500.

Wagh, P.V. and Waibel, P.E. (1966) Metabolizability and nutritional implications ofL-arabinose and D-xylose for chicks. Journal of Nutrition 90, 207–211.

Ward, A.T. and Marquardt, R.R. (1987) Antinutritional activity of a water-solublepentosan-rich fraction from rye grain. Poultry Science 66, 1665–1674.





White, W.B., Bird, H.R., Sunde, M.L., Burger, W.C. and Marlett, J.A. (1981) The viscosityinteraction of barley beta-glucan with Trichoderma viride cellulase in the chick intestine.Poultry Science 60, 1043–1048.

Wiseman, J. and McNab, J.M. (1998) Nutritive Value of Wheat Varieties Fed toNon-ruminants. HGCA Project Report No. 111. Home Grown Cereals Authority,Nottingham University, UK.

160 M. Choct




G.G. PartridgeCarbohydrase Enzymes in Pig Nutrition8

Introduction

The history of use of carbohydrase enzymes into pig diets based on ‘viscous’ grainssuch as wheat and barley is virtually as long as that of poultry but, until recent years,far less convincing. There are many reasons for this, and certainly not all are relatedto issues of cost effectiveness, which is always the ultimate arbitrator in the adoptionof any new feed technology. The basic structure of the pig and poultry industriesin many countries (i.e. ‘highly integrated’ versus ‘semi-extensive’, respectively) hastended to work against the rapid adoption of novel feed technologies into the pigsector. This situation is changing rapidly in swine markets in some parts of the worldand, in terms of the potential rapid uptake of new ideas, will undoubtedly be a movefor the better.

This chapter examines the role and efficacy of carbohydrase enzymes ingrain-based diets. Phytases and enzymes directed against vegetable protein meals areassessed elsewhere in this book. It is recognized that, at the time of writing, thepenetration of feed enzymes into the pig sector still lags behind that of poultry butthat knowledge has accrued rapidly, particularly in the past 5 years, bringing bothmore effective enzymes for pig applications and considerably more applicationsexpertise. Due to this rapid change the emphasis in this chapter is placed on the last10 years of published information.

Enzyme Responses: Pigs vs. Poultry

The physiological reasons why differing responses to exogenous carbohydrases mightbe expected in pigs and poultry have been covered by a number of authors on anumber of occasions, and for this reason will only be summarized here. Readers arereferred to a number of excellent papers that have been systematically published overthe past 10 years on these and other aspects of enzyme use into pig diets. A numberof these compared and contrasted responses in poultry and pigs (e.g. Dierick, 1989;



The Role and Efficacy ofCarbohydrase Enzymes in

Pig NutritionG.G. PARTRIDGE

Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UK

8



Chesson, 1993; Dierick and Decuypere, 1994, 1996; Graham and Balnave, 1995;Bedford and Schulze, 1998; Danicke et al., 1999).

The main factors that may influence the potential for enzyme response in pigscompared with poultry include gastrointestinal anatomy, digestive capacity anddigesta characteristics.

The presence of the crop in the broiler may, depending on the pH profile ofthe exogenous enzyme, allow for some enzyme activation at a relatively high pH(approximately 6) prior to passing into the acid environment of the gizzard.Generally in the pig a much lower pH (< 3) will be more rapidly experienced by theenzyme and its substrates in the stomach, though this will be highly dependent uponpig age, the degree of ingredient buffering, and the extent of microbial fermentationin the stomach leading to organic acid production (e.g. lactic acid). The pig is noted(Moran, 1982) for its relatively gentle gastric motility, leading to regional variationsacross the stomach in terms of pH and endogenous enzyme activity. All of thesefactors can potentially impinge on the activity of an exogenous enzyme, dependingon its pH and temperature characteristics and its inherent rate of reaction (Km).

Mean retention times (MRTs) of solid and liquid digesta in poultry also tendto be reduced relative to that of the pig (Moran, 1982). Solute markers in the upperpart of the gastrointestinal tract of broilers showed mean retention times in the crop,gizzard and small intestine of 2.8, 0.3 and 1.0 h, respectively (i.e. about 4 h total tothe end of the small intestine: Dierick and Decuypere, 1994). In contrast, Dierickand Decuypere (1994) cite a total MRT of 4–5 h through the stomach and smallintestine of the pig and Mahan (1982) quotes studies showing approximately 7.5 hthrough this same part of the gastrointestinal tract. The absolute figure will clearlyvary in both species depending on diet composition, soluble/insoluble fibre level,plane of feeding, age, etc., but the fact remains that the broiler, having less time thanthe pig for effective digestion and nutrient absorption, will be potentially disruptedmore by certain anti-nutritional factors that interfere with this process (e.g. viscous,high molecular weight soluble fibres). This may in turn have implications for enzymeaction and the relative magnitude of its effects in the two species, but it shouldequally be remembered that both species lack the appropriate enzymes to break downthese anti-nutrients, irrespective of the length of the digestive tract and residencetime. Both species, therefore, will potentially suffer the same consequences ofdisturbed digestion – not only less nutrients absorbed but also a rise in bacterialnumbers in the small intestine, resulting in increased competition for nutrientsbetween the host and its microflora.

The greater bacterial proliferation in the gut of the pig, compared with thebroiler, is often quoted as a reason why the magnitude of exogenous enzyme responsemight be expected to be lower in the pig. Certainly in the pig’s hindgut the longresidence time and voluminous caecum and colon offer an ideal opportunity forbacterial fermentation of fibrous residues to yield substantial quantities of volatilefatty acids, which will contribute to the animal’s maintenance requirement. In thesmall intestine, however, as mentioned above, such proliferation will always be apotentially negative factor. Caution should also be observed in interpretation of somestudies reporting substantial breakdown of fibre fractions in the small intestine of

162 G.G. Partridge




the pig in the absence of exogenous enzymes. Some of these studies used cannulatedpigs with T-pieces inserted at the terminal ileum and, occasionally, the duodenum(Graham et al., 1986, 1988). The opportunities for an unnatural microbialcolonization in the region of the cannula could influence the results seen.

It is well known that soluble, high molecular weight, non-starch polysaccharides(NSPs) from viscous cereal grains interfere with nutrient digestion and absorption inpoultry, with consequences for microbial proliferation in the gut (Apajalahti andBedford, 1998; Bedford and Schulze, 1998). The naturally higher water content ofpig digesta (about 10 percentage units higher than poultry on similar diets) leads todilution effects that will negate, to some degree, this viscosity problem in the pig’s gut(Fig. 8.1) (Danicke et al., 1999). This is not to say that, in pigs, viscosity effects aretotally irrelevant to potential carbohydrase responses in viscous grain diets, but ratherthat they are likely to be of a lower order of importance to the responses seen inbroilers. Reductions in digesta viscosity have been seen in many trials in pigs offeredviscous cereal-based diets (Fig. 8.2) but it is clear that starting viscosity levels arealways considerably lower than equivalent trials in poultry. It is of interest, however,that soluble fibre levels per se seem to be the key factor involved in negative bacterialchanges in the gut of pigs under disease challenge and fed specific types of diet, asreviewed later in this chapter, so viscosity should certainly not be underestimatedin the pig as an influence on productive performance. It should be recognized,therefore, that its influence may be more subtle than direct effects on nutrientdigestibility and, rather, mediated indirectly through its effects on the gut microflora.

Digesta dry matter flow rates from the stomach of young and grower pigs fedwheat or barley-based diets have been found to be increased by certain exogenousenzymes (Sudendey and Kamphues, 1995) (Fig. 8.3). In these trials, viscosityreduction was apparent immediately in the stomach of the animals offered enzymesupplemented diets. Ellis et al. (1996) noted that an increase in gastric viscosity in

Carbohydrase Enzymes in Pig Nutrition 163


2.71.8 1.7

8.7

5.7

2.4

Wheat var. 1 Wheat var. 2

Jejunum

6.03.8

1.8

36.3

11.0

3.0

Wheat var. 1 Wheat var. 2

Ileum*

Fig. 8.1. Effect of wheat variety on intestinal viscosity (mPa s) in broilers, turkeysand piglets (adapted form Danicke et al., 1999). % Broilers (3 weeks) wheatinclusion 73%; $ turkey (4 weeks) wheat inclusion 73%; ( pig (8 weeks) wheatinclusion 86%. * = small intestine, lower two-thirds for the pig. Wheat variety1 = ‘Ibis’ 11 g kg−1 DM soluble arabinoxylan; wheat variety 2 = ‘Alidos’17 g kg−1 DM soluble arabinoxylan. DM = dry matter.



the pig can lead to a failure of the sieving mechanism whereby large digesta particlesnormally fall to the bottom of the stomach, a feature seemingly unique to the pig. Ina viscous environment these large particles stay suspended and are thereby morelikely to exit the pylorus before being exposed to gastric secretions or particle sizereduction, which may have implications for subsequent digestion. Further down thegastrointestinal tract, viscosity is also known to disturb peristalsis, creating laminarrather than disturbed flow, and also to lead to disrupted pancreatic secretion via itseffects on gastric inhibitory polypeptide. Viscous non-starch polysaccharides mayalso physically coat starch granules, further reducing the rate of digestion (Ellis et al.,1996).

In successful pig trials using enzyme supplementation, a stimulation of volun-tary food intake is frequently an important contributor to the observed improve-ments in daily liveweight gain (Haberer et al., 1998). The fact that gastric emptyingrate and distension of the stomach are two of a number of factors involved in satietysignals in the pig (Forbes, 1995) is unlikely to be coincidental and it is proposed thatsome of the positive responses to exogenous enzymes on voluntary food intake couldbe due to various influences on digesta flow rate, as well as improvements in nutrientavailability and digestibility in the small intestine. Feedback loops involvinggut hormones could also be contributors to carbohydrase effects, as outlined byBedford and Schulze (1998), who proposed that caecal fermentation, enhanced

164 G.G. Partridge


3.3

2.2*

Ref. 1

1.7 1.5*

Ref. 2

4.2

1.7*

Ref. 2

6.4

3.1*

Ref. 3

2.2 1.8(*)

Ref. 4

Control +Enzyme complex *P < 0.05 (*)P < 0.10

Fig. 8.2. Effects of enzymes on viscosity in the small intestine of piglets.Ref. 1. Inborr (1994); ref. 2. Sudendey and Kamphues (1995) wheat/piglet andbarley/grower; ref. 3. Dusel et al. (1997); ref. 4. Partridge et al. (1998a).

2.272.62

Wheat, piglet

3.22 3.36

Barley, piglet

1.702.26

Barley, grower

(*) *

Fig. 8.3. Effect of enzyme addition to wheat or barley-based diets on dry matteroutflow rate from the stomach (Sudendey and Kamphues, 1995). Values are g drymatter leaving the stomach h−1 kg−1 body weight. ( Control; % +Enzyme complex(amylase, xylanase, β glucanase). *P < 0.05, (*)P < 0.10.



by breakdown products from xylanase addition (fermentable oligomers), couldinfluence enteroglucagon concentrations, which in turn influence stomach motilityby depressing gastrin concentrations. These many possibilities suggest exogenousenzymes could be potent influencers of gastrointestinal physiology through a varietyof different mechanisms, but these aspects have been little studied to date.

The relatively slow passage rate of digesta through the gut of the pig, whencompared with the broiler, may offer some advantages to exogenous enzymes when itcomes to the opportunity for insoluble cell wall breakdown, or thinning, coupledwith physicochemical changes to fibre (e.g. reduced water-holding capacity) asdigesta makes its way down the small intestine. Kyriazakis and Emmans (1995)found that the water-holding capacity of feedstuffs for young pigs (12–25 kg) was amajor influence on voluntary food intake (Fig. 8.4). As carbohydrase enzymes areknown to reduce the water-holding capacity of feedstuffs in vitro (Fig. 8.5), it followsthat some of the potential benefits of exogenous enzyme addition in the pig maybe related to disruption of water and soluble nutrient entrapment, by both solubleand insoluble fibre residues in the gut. Whilst the importance of soluble, viscouspolysaccharides to digestive disruption in the broiler seems to be a major mode ofaction in wheat and barley-based diets (Bedford and Schulze, 1998), in the pig itsimportance, as discussed, is clearly less profound. The conclusion must be that abroader range of mechanisms is likely to be relevant to the pig, not least the ability ofan exogenous enzyme to influence both soluble and, more particularly, insolublefibre in the gut. This has implications for product design, as enzymes will differ intheir affinities and rates of reaction against various fibre structures and it follows that



Fig. 8.4. The effect of water-holding capacity of the feed on voluntary food intakein pigs (12–25 kg liveweight). Scaled feed intake = g feed kg−1 liveweight day−1

(Kyriazakis and Emmans, 1995).



enzymes designed for maximum efficacy in the pig may involve both different levelsand/or different sources than those ideal for poultry. Xylanases, for example, evenderived from the same source organism, can vary widely in their catalytic activities onvarious xylan substrates (Bedford and Schulze, 1998) and the active concentration ofxylanase will also be an important determinant of the extent of cell wall hydrolysis(Tervila-Wilo et al., 1996).

Such issues become particularly important as commercial emphasis switchestowards carbohydrase products designed for non-viscous grains such as maizeand sorghum, which are the predominant cereals globally. In these grains, factorsassociated with insoluble fibre such as the packaging of nutrients inside cell wallmaterial, together with starch structure and composition, seem to be more relevantand may require a new range of enzyme solutions.

The rest of this chapter will systematically consider the major grain andgrain by-product sources available to pigs globally, examine their feeding value, itsvariability and reasons for it, and then review recent work on their potential fornutritional upgrading with appropriate exogenous enzymes.

Barley as a Grain Source for Pigs

The feeding value of conventional (hulled) barley for pigs was reviewed byvan Barneveld (1999). Estimates of DE in 16 cited studies ranged from 11.7 MJ kg−1

dry matter (DM) to 16.0 MJ kg−1 DM. Energy digestibility, rather than grossenergy content (a range of about 2 MJ kg−1 DM), had the greatest influence on thisvalue. In the recent studies of Fairbairn et al. (1999) the DE content of 20 barleysamples, from three locations and representing five varieties, ranged from 11.2 to13.1 MJ kg−1 (90% DM), with less variation between varieties than within varieties

166 G.G. Partridge


Maiz

e

Whe

at flo

ur

Barley

Oats

Maiz

e ge

rmSBM

Bread

Sunflo

wer

Rapes

eed

mea

l

Grass

mea

l

Potat

o fla

kes

−12% −29% −31% −19% −12% −17% −52% −8% −14% −75%−5%

10

8

6

4

2

0

Fig. 8.5. Effect of an enzyme complex on water-holding capacity (g g−1) of rawmaterials. ( No enzyme; % +Enzyme complex (xylanase; β-glucanase; protease).(Finnfeeds, personal communication, 1995.)



(5.8%, 0.7 MJ kg−1 versus 8.4%, 1.0 MJ kg−1). This re-emphasizes the importanceof growing conditions to proximate analysis, carbohydrate composition and,ultimately, feeding value for both swine and poultry (Aman et al., 1985; Amanand Newman, 1986; Bach Knudsen et al., 1987).

The last decade has seen an increasing interest in the potential of hull-less barleyas a feed ingredient for monogastrics. Hull-less barley differs from conventionalbarley grain in that the hull is less firmly attached to the kernel and consequentlybecomes detached during threshing. Despite, consequently, having lower insolublefibre levels and higher levels of protein, its feeding value for pigs and poultry hasoften failed to live up to its theoretical potential (Baidoo et al., 1998b; Thacker,1999). The reasons for this are most likely associated with the soluble fibre fractions(particularly β-glucans), which are concentrated in the endosperm of the barleygrain. Removing the hull fraction effectively concentrates these soluble componentsin hull-less barley, with comparative levels of β-glucans in the two grains being3.5–4.5% (hulled) versus 4.5–7% (hull-less) (Baidoo and Liu, 1998). The negativeeffects on growth seen after feeding micronized, hull-less barley to growing/finishingpigs are almost certainly due to further solubilization of this fibre during thermalprocessing of the grain, effectively negating the potentially positive effects of theprocess on starch gelatinization (Thacker, 1999). The latter study re-emphasizedthe limitations of measuring total tract digestibility in the pig in isolation fromperformance studies, whereby ‘apparent’ improvements in digestibility of nutrientscan be more than offset by reduced voluntary food intake coupled with implicationsfor microbial proliferation/competition for nutrients in the gut. This aspect will beexamined in more detail later in this chapter.

Differences in feeding value for hulled and hull-less barley varieties will alsoundoubtedly be influenced by other components of the grain apart from fibre,notably the amylose : amylopectin ratio of the starch fraction. Pettersson andLindberg (1997) investigated ileal and faecal digestibility of hull-less versus hulledbarleys, with varying amylose : amylopectin ratios, in growing/finishing pigs. Theyfound that the poorest energy and starch digestibility was in hulled/high amylosegrain and the best in hull-less/high amylopectin, at both the ileal and faecal levels.Hull-less barley tended to give higher nutrient digestibilities overall and showedincreased hindgut digestion compared with hulled material. This would presumablyhave implications for rate and pattern of microbial fermentation, but these were notinvestigated in this study. Equally, productive performance was not measured on thedifferent barley types to see if the digestibility differences seen were actually reflectedin growth, feed intake and feed utilization. Miller et al. (1994) also describeddigestibility experiments looking at the interaction between fibre (ADF – aciddetergent fibre) level in barley, β-glucan content and amylose : amylopectin ratio.They concluded that the energy content of barley for both poultry and swine wasdecreased greatly by ADF content and to a significant but lesser extent by higher totalβ-glucans. They also found that β-glucans had a larger negative effect on energydigestibility in poultry than in swine and that high amylopectin (‘waxy’) barley starchoffset some of the reductions in energy digestibility caused by a high ADF level. Thisillustrates the complex interrelationships between various fractions in the grain





which can all have potential implications for feeding value and, importantly to thisdiscussion, subsequent response to exogenous enzymes.

Response of Hulled and Hull-less Barley to ExogenousEnzymes in Pigs

The surface of the aleurone cell walls of barley contains high levels of xylan,surrounding a core of mixed-linked β-glucans (Autio et al., 1996). For this reason,optimal responses in rations based on barley seem to be found when both xylanaseand β-glucanase activities are supplied. This is illustrated in the studies of vanLunen and Schulze (1996a) and Ramaswamy et al. (1996) where clear responses,particularly in hulled barley, are only achieved in the presence of significant xylanaseactivity. Some studies using exogenous enzymes in hulled barley-based rations(Graham et al., 1989; Thacker et al., 1992b) have tended to emphasize theβ-glucanase component for reducing levels of soluble β-glucans, which are wellknown to exacerbate digesta viscosity problems in broilers. In the pig these effects arebelieved to be of a much lower magnitude, due to pig digesta having a much lowerdry matter content than the broiler (Bedford and Schulze, 1998), effectively dilutingthe effects of these viscous fibres and partially negating their detrimental effects onnutrient digestion and absorption. However, viscous fibre can still be relevant for itsimpact on the microbial population of the gut and thereby indirectly influenceenzyme response, an aspect that will be examined in more detail later in this chapter.

Table 8.1 shows a summary of a number of published trials over the past10 years into both hulled and hull-less barley-based diets, using a range of enzymeadditions. Although an overview is made more difficult by potential influencesof a number of factors (e.g. growth versus digestibility trials; faecal versus ilealdigestibility measurements; pig age/weight used; choice of enzyme activities;pelleting versus meal/mash, i.e. potential for denaturation with some products;hulled versus hull-less barley), some trends do emerge.

Responses to these various products in young pigs (< 25 kg) are strong andquite consistent, particularly in the hull-less barley-based diets that have tended todominate recent publications. Baidoo et al. (1998a) describe a declining response inpigs over the weight range 40–60 kg but an increasing numerical response to theenzyme used as animals moved from the 9–20 kg weight range (daily gain +12%;feed : gain −3%) to the 20–40 kg weight range (daily gain +17%; feed : gain −10%).It should be noted that even in this trial the animals in the heaviest weight category(40–60 kg) showed improvements in both growth and feed : gain (+4%; −12%,respectively) after addition of the enzyme but that the large standard errors associatedwith the data meant they failed to achieve statistical significance. Data from studieswith grower/finisher pigs (> 25 kg) range from no response or small responses(Graham et al., 1989; Thacker et al., 1992a,b – but see comment above) to thosewhere clear, significant effects on digestibility and/or growth rate were seen (vanLunen and Schulze, 1996a; McCracken et al., 1996; Ramaswamy et al., 1996;Baidoo et al., 1998a,b). Even in some apparent ‘no-response’ trials (Thacker et al.

168 G.G. Partridge




1992a) significant, positive effects on faecal digestibility of protein and numericaleffects on energy were seen after enzyme addition. This is a reminder, particularly intrials with grower/finisher pigs, that products potentially liberating more nutrientsmust be tested in conditions that allow this nutrient release to be expressed in termsof improved energy utilization, lean gain, etc. In trials where the products are addedon top of an existing formulation, which may already be in excess of an animal’s dailynutrient needs for its lean protein deposition potential, carcass information is afundamental requirement to aid trial interpretation.

The studies of Baidoo et al. (1998a), McCracken et al. (1996) and Han andFroseth (1993) illustrate that interpretation of productive and economic responses toadded feed enzymes in barley-based rations is highly dependent on an appraisal ofgrain ‘quality’ in terms of energy and amino acid availability. Emphasis thus switchesto simple predictors of feeding value in barley and, in particular, their rapid estima-tion. Fairbairn et al. (1999) described a series of predictive equations for hulled barleyfor pigs in which ADF content of the barley described DE, with an R2 value of 0.85:

DE (kcal kg−1, 90% dry matter) = 3526 − (92.8 × ADF%).

More complex equations, incorporating further factors, increased the R2 value to0.90; for example:

DE (kcal kg−1, = 3964 − (81.3 × ADF%) − (273.9 × acid90% dry matter) detergent lignin (%)) − (56.4 × total β-glucans (%)).

Near-infrared spectroscopy (NIRS) appears currently to offer the best potentialfor rapid estimation of barley DE (Zijlstra et al., 1998) and amino acid acid content(Jaikaran et al., 1998) with sufficient accuracy to allow for its subsequent use as a toolto maximize the cost efficacy of enzyme addition. Emphasis on DE estimation aloneas a predictor of potential enzyme response must, however, be tempered with somecaution. Experiences with wheat in pigs (Cadogan et al., 1999) have indicated thatthe factors controlling voluntary food intake of different samples do not appear to beclosely related to measured DE, hence any screening method must ultimately takeaccount of this anomaly and understand the key factors responsible.

Wheat, Triticale and Rye as Grain Sources For Pigs

A review of the literature on wheat feeding value for pigs (van Barneveld, 1999)found a range of 3.7 MJ DE kg−1 DM across all cultivars studied. The lowestreported DE was 13.3 MJ kg−1 DM and the highest 17.0 MJ kg−1 DM. As for othergrains, the variability seen is a function of methodology, cultivars, growing regions,sites and years. Most studies saw differences between ‘poor’ and ‘good’ wheats in therange 0.8–1.4 MJ kg−1 DM.

The importance of cultivar to feeding value was illustrated in a study in the UK(Schulze et al., 1997) where six varieties of wheat were grown on one site, to negateenvironmental influences, and then used as the sole grain source in diets for pigletsfrom 9 kg body weight (Table 8.2). Variations in growth rate and feed intake of





170 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d,le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Gra

ham

etal

.(19

88)

Barle

y;w

heat

polla

rd;m

eal/m

ash

19–2

5kg

ileal

cann

ulat

edpi

gsβ-

gluc

anas

e(1

6U

g−1)

xyla

nase

(130

0U

g−1)

2kg

t−1

Ileal

dige

stib

ility

ofpr

otei

n(%

):C

:64.

5+E

:70.

1*

Ileal

dige

stib

ility

ofen

ergy

(%):

C:5

5.9

+E:5

8.2

( P<

0.10

)

Gra

ham

etal

.(19

89)

Hul

l-les

sba

rley;

soyb

ean

mea

l;m

eal/m

ash

vers

uspe

llete

d(9

3 °C

)80

kgile

alca

nnul

ated

pigs

β-gl

ucan

ase

(20

Ug−1

)5

kgt−1

Ileal

dige

stib

ility

ofpr

otei

n(%

):C

(mea

l):52

.2+E

(mea

l):53

.3C

(pel

let):

54.9

+E(p

elle

t):52

.3

Ileal

dige

stib

ility

ofen

ergy

(%):

C(m

eal):

60.3

+E(m

eal):

61.9

C(p

elle

t):63

.3+E

(pel

let):

63.9

Bohm

e(1

990)

Barle

y;w

heat

;soy

bean

mea

l;pe

llete

d11

–25

kgC

ellu

lase

, β-g

luca

nase

,α-

amyl

ase;

gluc

oam

ylas

e1

kgt−1

Gro

wth

perfo

rman

ce:

C:3

81g

day−1

Feed

:gai

n1.

88+E

:440

gda

y−1Fe

ed:g

ain

1.70

*

Thac

ker e

tal.

(199

2a)

Barle

y;so

ybea

nm

eal;

pelle

ted

(<60

°C)

C1

=no

grow

thpr

omot

erC

2=

+sa

linom

ycin

22–8

6kg

β-gl

ucan

ase

(750

Ug−1

)pe

ntos

anas

e(6

50U

g−1)

2.5

kgt−1

Gro

wth

perfo

rman

ce:

C1:

840

gda

y−1Fe

ed:g

ain

2.53

+E:8

60g

day−1

Feed

:gai

n2.

59C

2:87

0g

day−1

Feed

:gai

n2.

53+E

:840

gda

y−1Fe

ed:g

ain

2.56

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C1:

72.2

+E:7

3.7

C2:

75.3

+E:8

1.3*

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

1:74

.2+E

:75.

6C

2:73

.5+E

:78.

5

Tabl

e8.

1.So

me

publ

ishe

dst

udie

son

the

effe

cts

ofen

zym

epr

epar

atio

nson

barle

y-ba

sed

diet

s.(N

ote

that

whe

rebo

thile

alan

dfa

ecal

dige

stib

ility

mea

sure

men

tsw

ere

mad

e,da

tais

pres

ente

dpr

efer

entia

llyat

the

ileal

leve

l.)





Bedf

ord

etal

.(19

92)

Hul

l-les

sba

rley;

soyb

ean

mea

l;m

eal/m

ash

12–1

8kg

β-gl

ucan

ase

(108

6U

g−1)

2kg

t−1W

eigh

tgai

n:C

:5.2

4kg

+E:6

.14

kg*

Feed

:gai

nC

:1.7

5+E

:1.5

5( P

=0.

10)

Thac

kere

tal.

(199

2b)

Barle

y;so

ybea

nm

eal;

pelle

ted

(<60

°C)

C1

=no

acid

C2

=+

prop

rioni

cac

id25

–98

kg

β-gl

ucan

ase

(750

Ug−1

)2.

5kg

t−1G

row

thpe

rform

ance

:C

1:83

0g

day−1

Feed

:gai

n3.

09+E

:840

gda

y−1Fe

ed:g

ain

3.04

C2:

860

gda

y−1Fe

ed:g

ain

2.89

+E:7

90g

day−1

Feed

:gai

n3.

00Fa

ecal

dige

stib

ility

ofpr

otei

n(%

):C

1:78

.6+E

:75.

3C

2:74

.3+E

:79.

2Fa

ecal

dige

stib

ility

ofen

ergy

(%):

C1:

76.3

+E:7

9.3

C2:

78.7

+E:7

8.1

Thac

ker e

tal.

(199

2b)

Hul

l-les

sba

rley;

soyb

ean

mea

l;pe

llete

d(<

60°C

)C

1=

noac

idC

2=

+fu

mar

icac

id8–

18kg

β-gl

ucan

ase

(750

Ug−1

)2.

5kg

t−1G

row

thpe

rform

ance

:C

1:25

4g

day−1

Feed

:gai

n2.

20+E

:274

gda

y−1Fe

ed:g

ain

2.03

C2:

275

gda

y−1Fe

ed:g

ain

1.99

+E:2

75g

day−1

Feed

:gai

n1.

91Fa

ecal

dige

stib

ility

ofpr

otei

n(%

):C

1:78

.4+E

:77.

8C

2:77

.6+E

:81.

7Fa

ecal

dige

stib

ility

ofen

ergy

(%):

C1:

82.6

+E:8

1.9

C2:

81.0

+E:8

2.3



172 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d,le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Han

and

Fros

eth

(199

3)N

ine

barle

ycu

ltiva

rsof

‘hig

h’an

d‘lo

w’q

ualit

y‘h

igh’

:<7.

6%AD

F;<

4.7%

β-gl

ucan

s(n

=4)

‘low

’:>

9.8%

ADF;

>5.

4%β-

gluc

ans

(n=

5)70

kgdu

oden

alca

nnul

ae–

mob

ileny

lon

bag

tech

niqu

e

β-gl

ucan

ase,

cellu

lase

;xyl

anas

e;pe

ctin

ase

1kg

t−1

Faec

alD

E(k

calk

g−1dr

ym

atte

r):‘h

igh’

cont

rol:

3455

‘hig

h’+E

:353

2*(+

2.2%

)‘lo

w’c

ontro

l:32

96‘lo

w’+

E:34

14*(

+3.6

%)

Inbo

rr(1

994)

Hul

led

orhu

ll-le

ssba

rley;

soyb

ean

mea

l;m

eal/m

ash

9.5–

14kg

β-gl

ucan

ase

2.5

kgt−1

togi

ve89

–95

mU

g−1fe

edG

row

thpe

rform

ance

:C

(hul

l-les

s):2

23g

day−1

Feed

:gai

n1.

60+E

(hul

l-les

s):2

29g

day−1

Feed

:gai

n1.

52C

(hul

led)

:200

gda

y−1Fe

ed:g

ain

1.71

+E(h

ulle

d):2

08g

day−1

Feed

:gai

n1.

65Ef

fect

ofen

zym

eon

grow

th( P

=0.

07)

Effe

ctof

enzy

me

onfe

ed:g

ain

( P=

0.06

)

van

Lune

nan

dSc

hulz

e(1

996a

)Ba

rley;

rape

seed

mea

l;pe

llete

d35

kgE1

:β-g

luca

nase

E2:x

ylan

ase;

β-gl

ucan

ase

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C:8

4.6

+E1:

85.3

+E2:

86.7

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

:80.

9+E

1:81

.8+E

2:83

.4*

Tabl

e8.

1.C

ontin

ued.

182Z:\Customer\CABI\A3882 - Bedford + Partridge - Enzymes\A3957 - Bedford - Enzymes #Q.vp24 November 2000 17:41:18




Ram

asw

amy

etal

.(1

996)

Hul

l-les

sor

hulle

dba

rley;

soyb

ean

mea

l;ca

nola

mea

l47

kg

E1:x

ylan

ase

E2:x

ylan

ase;

β-gl

ucan

ase

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C(h

ull-l

ess)

:82.

0+E

1:81

.0+E

2:81

.0C

(hul

led)

:76.

8+E

1:81

.5+E

2:77

.3En

zym

eP

<0.

05Ba

rley

×en

zym

eP

=0.

06

Mea

sure

dD

E(M

Jkg

−1):

C(h

ull-l

ess)

:14.

40+E

1:14

.73

+E2:

14.9

4C

(hul

led)

:13.

5+E

1:15

.3+E

2:13

.5En

zym

eP

<0.

001

Barle

y×

enzy

me

P=

0.00

8

Pow

eret

al.(

1996

)H

ulle

dba

rley;

whe

at;w

heat

mid

dlin

gs;

pelle

ted

(~70

°C)

6–15

kg

β-gl

ucan

ase

(0,5

00,1

000,

2000

p.p.

m.)

Gro

wth

perfo

rman

cean

ddi

gest

ibilit

yat

optim

umin

clus

ion

leve

l(2

000

p.p.

m.):

C:2

01g

day−1

Feed

:gai

n2.

11+E

:227

gda

y−1*

Feed

:gai

n1.

95*

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C:7

7.4

+E:8

0.3

Liet

al.(

1996

b)H

ull-l

ess

barle

y;so

ybea

nm

eal;

mea

l/mas

h6–

11kg

β-gl

ucan

ase

(100

0U

g−1)

Ado

se-re

spon

sest

udy

(not

eth

em

axim

umre

spon

sew

assh

own

atth

ehi

ghes

tlev

elof

incl

usio

n,2

kgt−1

,so

only

this

data

ispr

esen

ted)

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C:8

1.6

+E:8

8.5*

(sig

nific

antl

inea

reffe

cts

ofdo

sera

tew

ere

seen

)

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

:85.

2+E

:89.

5*(s

igni

fican

tlin

eare

ffect

sof

dose

rate

wer

ese

en)



174 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d,le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Liet

al.(

1996

a)H

ull-l

ess

barle

y;so

ybea

nm

eal;

mea

l/mas

h7–

11kg

β-gl

ucan

ase

(100

0U

g−1)

2kg

t−1Ile

aldi

gest

ibilit

yof

prot

ein

(%):

C:6

5.2

+E:7

3.5*

Ileal

dige

stib

ility

ofen

ergy

(%):

C:6

4.9

+E:7

1.1*

Baid

ooet

al.(

1998

b)H

ull-l

ess

barle

y(v

arie

ties

‘Con

dor’

and

‘Buc

k’);

cano

lam

eal;

mea

l/mas

h14

kgile

al-c

annu

late

dpi

gs

Cel

lula

se;β

-glu

cana

se;x

ylan

ase

(11

Ug−1

;27

Ug−1

;43

Ug−1

)10

0g

t−1

Ileal

dige

stib

ility

ofpr

otei

n(%

):C

:57.

6+E

:61.

8*(e

ffect

sw

ere

seen

onbo

thva

rietie

s,he

nce

thes

ear

eov

eral

lval

ues)

Ileal

dige

stib

ility

ofen

ergy

(%):

C:5

7.1

+E:6

3.3*

(effe

cts

wer

ese

enon

both

varie

ties,

henc

eth

ese

are

over

allv

alue

s)

Baid

ooet

al.(

1998

b)H

ull-l

ess

barle

y;so

ybea

nm

eal;

cano

lam

eal;

mea

l/mas

hor

pelle

t(80

°C)

9–60

kg

Cel

lula

se;β

-glu

cana

se;x

ylan

ase

(11

Ug−1

;27

Ug−1

;43

Ug−1

)10

0g

t−1

Gro

wth

perfo

rman

ce(9

–60

kg):

C:7

21g

day−1

Feed

:gai

n2.

01+E

:782

gda

y−1*

Feed

:gai

n1.

84*

Posi

tive

effe

cts

toen

zym

ead

ditio

nw

ere

seen

inbo

thm

ash

and

pelle

ted

feed

,hen

ceov

eral

lval

ues

are

show

n

Baid

ooet

al.(

1998

a)H

ull-l

ess

barle

y(’F

alco

n’);

mea

l/mas

h10

–14

kgile

al-c

annu

late

dpi

gsE1

:xyl

anas

eE2

: β-g

luca

nase

E3:x

ylan

ase;

β-gl

ucan

ase;

prot

ease

Ileal

dige

stib

ility

ofpr

otei

n(%

):C

:59.

6+E

1:60

.1+E

2:66

.7*

+E3:

67.2

*

Ileal

dige

stib

ility

ofen

ergy

(%):

C:6

2.1

+E1:

64.6

+E2:

69.8

*+E

3:71

.2*

Tabl

e8.

1.C

ontin

ued.





Baid

ooet

al.(

1998

a)H

ull-l

ess

barle

y(’B

uck’

,‘C

ondo

r’or

‘Fal

con’

);ca

nola

mea

l;m

eal/m

ash

14kg

ileal

-can

nula

ted

pigs

Xyla

nase

;β-g

luca

nase

;pro

teas

eIle

aldi

gest

ibilit

yof

prot

ein

(%):

C(b

uck)

:60.

3+E

:64.

5C

(con

dor):

54.8

+E:

60.7

C(fa

lcon

):67

.6+E

:67.

2

Ileal

dige

stib

ility

ofen

ergy

(%):

C(b

uck)

:57.

7+E

:67.

0C

(con

dor):

56.2

+E:

66.3

C(fa

lcon

):60

.6+

E:61

.3

Yin

etal

.(20

00)

Hul

led

barle

y(h

igh

orno

rmal

bush

elw

eigh

t–H

BWor

NBW

);pe

llete

dH

BW=

70kg

hl−1

NBW

=64

kghl

−1

23kg

ileal

-can

nula

ted

pigs

Xyla

nase

; β-g

luca

nase

(400

;400

Ug−1

)1

kgt−1

Ileal

dige

stib

ility

ofpr

otei

n(%

):C

(HBW

):74

.8+E

:76.

1C

(NBW

):72

.6+E

:76.

8*

Ileal

dige

stib

ility

ofen

ergy

(%):

C(H

BW):

67.8

+E:6

8.5

C(N

BW):

65.8

+E:7

0.2*

Gill

etal

.(20

00)

Hul

led

barle

y;so

ybea

nm

eal;

pelle

ted

(75°

C)

8–18

kgβ-

gluc

anas

e;xy

lana

se;a

myl

ase

1kg

t−1G

row

thpe

rform

ance

:C

:329

gda

y−1Fe

ed:g

ain

1.60

+E:3

50g

day−1

Feed

:gai

n1.

53D

E(M

Jkg

−1)

C:1

4.86

+E:1

5.08

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C:7

6.8

+E:7

5.1



70–80 g day−1 were seen in this trial over a 3-week period. Voluntary feed intake andapparent DE intake (feed intake × measured faecal DE) appeared to be influenced bythe water-holding capacity (WHC) of the finished feed after pelleting and wasdescribed by the equation:

DEI (kcal day−1) = 3325.1 − 663.64 * WHC (g g−1); (P = 0.015, r2 = 0.81),

which in turn suggests that the fibre content of the diet (soluble plus insoluble) was acontributory factor. This agrees with the observations of Kyriazakis and Emmans(1995), who found that varying fibre types influenced voluntary food intake in theyoung pig (12–25 kg) by increasing the WHC of the final feed (Fig. 8.4).

Cadogan et al. (1999) described similar weaner growth studies (7.5–15.6 kg) inwhich ten Australian cultivars of wheat were compared immediately after harvest,and again after 10 months storage. Growth performance across the ten cultivarsvaried from 233 to 447 g day−1 (mean 388, SD 62.9) and was heavily influencedby voluntary food intake (mean 438 g day−1, SD 70.0, range 271–514 g day−1).Apparent DM digestibility of the ten wheats showed no relationship to pig growthperformance, illustrating the limitations of faecal digestibility in predicting pigperformance, without further knowledge of the factors influencing feed intake. After10 months of storage the marked differences between the varieties remained,although feed intake and daily gain were, on average, 11.3% and 9.8% higher than inthe same samples immediately post-harvest.

Variation in triticale and rye feeding value has been studied relatively little inpigs. Van Barneveld (1999) cited data from Charmley and Greenhalgh (1987), whofound triticale feeding value was lower than that of wheat (15.1 versus 16.0 MJ kg−1

DM), but that differences between the three triticale cultivars studied were relativelysmall. It is likely that the feeding value of both triticale and rye will be at least asvariable as that of wheat, and likely more so, allowing for the fact that both rawmaterials will have higher total levels of non-starch polysaccharides and, particularly,soluble arabinoxylans (Cyran et al., 1995). Rye-based diets offered to poultry are aconsiderable challenge to the bird, due to their negative effects on digesta viscosity,

176 G.G. Partridge


VarietyaTest wt(lb bu−1)

Water holdingcapacity

(filter method)Daily gain

(g)Daily feed

(g) Feed : gainDEb

(kcal kg−1)DE intake

(kcal day−1)

HussarHunterBrigadierBeaverDynamoRibandSEM

62.461.660.960.660.660.5

–

1.921.852.042.221.922.07

–

488470457435419415

20

636608609558614554

22

1.311.301.341.291.491.380.063

3365.53363.53246.53308.53298.53384.5

21.5

214020451977184620251875

–

aAll varieties grown in one location.bMeasured digestible energy.

Table 8.2. Effect of variety on the feeding value of wheat (Schulze et al., 1997).



leading to poorer nutrient digestibility and bird growth. In pigs this may be less of aphysiological problem than to the bird, but the feeding value of rye is still recognizedto be well below that of equivalent wheat-based diets and its fibre content seems to bethe main contributory factor (Bayzlo, 1990).

Response of Wheat, Triticale and Rye to Exogenous Enzymesin Pigs

The predominant non-starch polysaccharides present in the endosperm and aleuronelayers of wheat, triticale and rye are arabinoxylans, which comprise 50–60% of thetotal NSPs (Dierick and Decuypere, 1994; Evers et al., 1999). In contrast to barleyand oats, β-glucan is a minor component in these grains. The arabinoxylans in thethin cell walls within the endosperm are water-soluble to varying degrees (dependingon variety, growing conditions, etc.) whereas those in the thick cell walls of thealeurone layer are predominantly insoluble. Xylanases are therefore the main type ofexogenous enzyme used for these grain types, as well as for grain by-products derivedfrom them (e.g. wheat bran, middlings, etc.) Table 8.3 shows a summary of anumber of published trials over the past 10 years into wheat, rye and triticale-baseddiets, using a range of enzyme additions.

As with the barley data in Table 8.1, there is a greater emphasis on trials withyoung pigs (< 25 kg) than in grower/finishers. This is based on the assumption thatpotential responses to enzymes will be greater in the younger animal where daily nutri-ent intake, in terms of both energy and amino acids, is often the limitation to achievingthe animal’s lean gain potential. The success rate in showing positive effects of enzymeson pig performance, measured as either growth rate, feed : gain or digestibility, seemsto be higher in the younger animal but this trend is influenced by the fact that veryfew studies have made an attempt to predetermine wheat/triticale/ rye feeding value,before adding an appropriate enzyme. In the few recent studies where this has beenattempted (Choct et al., 1999; Partridge et al., 1999) the ability of a xylanase to raisesignificantly the productive performance of a ‘poor’ wheat to the level of that of ‘good’wheat has been clearly demonstrated, irrespective of the age of the pig (Table 8.4).

The relative importance of soluble pentosans to the response to xylanase in thepig is thrown into question by a number of studies where rye was the predominantgrain (Thacker et al., 1991, 1992a; Bedford et al., 1992; Thacker and Baas,1996). Dusel et al. (1997b) showed that the use of a mixed-grain diet based on acombination of wheat (high extract viscosity), rye and barley gave responses toenzyme addition that went beyond the apparent effects on viscosity reduction (Table8.5). As in the studies of Choct et al. (1999) and Partridge et al. (1999), addition ofxylanase or the xylanase/protease combination released an apparent limitation onvoluntary food intake by the animal in these diets such that the growth response seenwas a cumulative effect of both increased nutrient availability and intake. Other,more subtle, changes in the microbial flora may also have influenced the response tothe enzyme. Pigs in the xylanase-supplemented group appeared to show decreasesin bacterial proliferation in the small intestine (Fig. 8.6) when compared with both





178 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d,le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Bohm

e(1

990)

Whe

at;b

arle

y;m

aize

;soy

bean

mea

l;pe

llete

d11

–25

kg

Cel

lula

se, β

-glu

cana

se,α

-am

ylas

e;gl

ucoa

myl

ase

1kg

t−1

Gro

wth

perfo

rman

ce:

C:4

11g

day−1

Feed

:gai

n1.

81+E

:458

gda

y−1Fe

ed:g

ain

1.56

*

Thac

kere

tal.

(199

1)R

ye;s

oybe

anm

eal;

mea

l(e

xpt1

)orp

elle

ts(5

5 °C

,exp

t2)

20–9

8kg

Pent

osan

ase

(650

Ug−1

)2.

5kg

t−1G

row

thpe

rform

ance

:C

(mea

l):73

0g

day−1

Feed

:gai

n3.

48+E

:760

gda

y−1Fe

ed:g

ain

3.12

*C

(pel

let):

760

gda

y−1Fe

ed:g

ain

2.64

+E:7

80g

day−1

Feed

:gai

n2.

64Fa

ecal

dige

stib

ility

ofpr

otei

n(%

):C

(mea

l):79

.2+E

:80.

0C

(pel

let):

77.0

+E:7

7.3

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

(mea

l):80

.8+E

:80.

1C

(pel

let):

81.2

+E:8

0.1

Thac

ker e

tal.

(199

2a)

Rye

;soy

bean

mea

l;pe

llete

d(<

60°C

)C

1=

nogr

owth

prom

oter

C2

=+

salin

omyc

in20

–84

kg

β-gl

ucan

ase

(750

Ug−1

)pe

ntos

anas

e(6

50U

g−1)

2.5

kgt−1

Gro

wth

perfo

rman

ce:

C1:

740

gda

y−1Fe

ed:g

ain

2.40

+E:7

10g

day−1

Feed

:gai

n2.

48C

2:72

0g

day−1

Feed

:gai

n2.

38+E

:750

gda

y−1Fe

ed:g

ain

2.43

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C1:

72.5

+E:7

6.5

C2:

78.9

+E:7

9.8

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

1:78

.9+E

:80.

6C

2:82

.7+E

:82.

5

Tabl

e8.

3.So

me

publ

ishe

dst

udie

son

the

effe

cts

ofen

zym

epr

epar

atio

nson

whe

at,r

yeor

tritic

ale-

base

ddi

ets.

(Not

eth

atw

here

both

ileal

and

faec

aldi

gest

ibilit

ym

easu

rem

ents

wer

em

ade,

data

are

pres

ente

dpr

efer

entia

llyat

the

ileal

leve

l.)





Bedf

ord

etal

.(19

92)

Rye

;soy

bean

mea

l;m

ash/

mea

l11

–15

kgXy

lana

se(8

50U

g−1)

2kg

t−1W

eigh

tgai

n:C

:4.2

4kg

+E:4

.24

kgFe

ed:g

ain:

C:1

.64

+E:1

.57

McL

ean

etal

.(19

92)

Whe

at;s

oybe

anm

eal

Wea

nerp

igs

E1:a

myl

ase;

xyla

nase

;cel

lula

se;

pect

inas

e(7

00g

t−1)

E2: β

-glu

cana

se;p

ectin

ase;

cellu

lase

;he

mic

ellu

lase

(1kg

t−1)

Ileal

dige

stib

ility

ofpr

otei

n(%

):C

:74.

0+E

1:76

.6+

E2:7

3.1

Ileal

dige

stib

ility

ofen

ergy

(%):

C:7

0.8

+E1:

73.5

*+E

2:72

.9

Inbo

rret

al.(

1993

)W

heat

:bar

ley

(50

:50)

;soy

bean

mea

l;m

eal/m

ash

~6–1

0kg

E1:β

-glu

cana

se#1

;xyl

anas

e;am

ylas

e(3

5;59

0;33

00U

g−1)

1kg

t−1

E2:β

-glu

cana

se#2

;xyl

anas

e;am

ylas

e(4

1;74

0;33

00U

g−1)

950

gt−1

Gro

wth

perfo

rman

ce:

C:2

04g

day−1

Feed

:gai

n1.

38+E

1:22

3g

day−1

Feed

:gai

n1.

38+E

2:20

5g

day−1

Feed

:gai

n1.

39Ile

aldi

gest

ibilit

yof

prot

ein

(%):

C:7

6.8

+E1:

78.4

+E2:

77.9

Ileal

dige

stib

ility

ofst

arch

(%):

C:9

4.9

+E1:

97.7

*+E

2:96

.8

Offi

cer(

1995

)W

heat

;fis

hmea

l;m

eatm

eal;

soyb

ean

mea

l;m

eal/m

ash

~6.5

–14

kg

E1:p

rote

ase;

lipas

e;β-

gluc

anas

e;am

ylas

e;ce

llula

se(2

kgt−1

)E2

:pro

teas

e;lip

ase;

β-gl

ucan

ase;

amyl

ase;

cellu

lase

;hem

icel

lula

se;

pent

osan

ase

(2kg

t−1)

E3:a

myl

ase;

xyla

nase

; β-g

luca

nase

;pe

ctin

ase

(2kg

t−1)

Gro

wth

perfo

rman

ce(a

fterw

eani

ng):

C:3

09g

day−1

Feed

:gai

n1.

54+E

:319

gda

y−1Fe

ed:g

ain

1.40

( P<

0.10

)C

:299

gda

y−1Fe

ed:g

ain

1.76

+E2:

282

gda

y−1Fe

ed:g

ain

1.82

C:3

61g

day−1

Feed

:gai

n1.

56+E

3:33

2g

day−1

Feed

:gai

n1.

49( P

<0.

10)



180 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d,le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Cam

pbel

l eta

l.(1

995)

Whe

at;l

upin

(W/L

)orw

heat

;soy

bean

mea

l(W

/S);

pelle

ted

(70 °

C)

6–13

kg

Xyla

nase

s;pe

ntos

anas

es;

β-gl

ucan

ases

400

gt−1

Gro

wth

perfo

rman

ce:

C(W

/L):

355

gda

y−1Fe

ed:g

ain

1.21

+E:3

88g

day−1

Feed

:gai

n1.

15C

(W/S

):29

1g

day−1

Feed

:gai

n1.

29+E

:353

gda

y−1Fe

ed:g

ain

1.23

Ove

rall

effe

cts

ofen

zym

eon

grow

thra

te( P

<0.

05)a

ndfe

ed:g

ain

( P=

0.08

)

Thac

kera

ndBa

as(1

996)

Rye

;soy

bean

mea

lPe

llete

d(<

65°C

)23

–83

kg(T

rial1

)28

–92

kg(T

rial2

)

‘Mul

ti-en

zym

epr

epar

atio

ns’E

1–E4

:E1

1kg

t−1

E220

0g

t−1

E350

0g

t−1

E470

0g

t−1

Incl

usio

nra

tes

supp

lied

the

sam

epe

ntos

anas

eac

tivity

,acc

ordi

ngto

the

auth

ors’

assa

ym

etho

d

Gro

wth

perfo

rman

ce(T

rial1

):C

:750

gda

y−1Fe

ed:g

ain

2.52

+E1:

780

gda

y−1Fe

ed:g

ain

2.42

+E2:

820

gda

y−1Fe

ed:g

ain

2.39

+E3:

800

gda

y−1Fe

ed:g

ain

2.37

+E4:

800

gda

y−1Fe

ed:g

ain

2.44

Gro

wth

perfo

rman

ce(T

rial2

):C

:880

gda

y−1Fe

ed:g

ain

2.42

+E1:

940

gda

y−1Fe

ed:g

ain

2.33

+E2:

930

gda

y−1Fe

ed:g

ain

2.37

+E3:

920

gda

y−1Fe

ed:g

ain

2.37

+E4:

950

gda

y−1Fe

ed:g

ain

2.37

Tabl

e8.

3.C

ontin

ued.





Schu

lze

etal

.(19

96b)

Whe

at;s

oybe

anm

eal;

pelle

ted

8–22

kgE1

:xyl

anas

e#1

(500

0U

g−1)1

kgt−1

E2:x

ylan

ase

#1/p

rote

ase

(500

0/50

0U

g−1)

1kg

t−1

E3:x

ylan

ase

#2(5

000

Ug−1

)1kg

t−1

Gro

wth

perfo

rman

ce:

C:3

04g

day−1

Feed

:gai

n1.

69+E

1:39

4g

day−1

*Fe

ed:g

ain

1.49

*+E

2:38

5g

day−1

*Fe

ed:g

ain

1.49

*+E

3:40

2g

day−1

*Fe

ed:g

ain

1.45

*

Gill

and

Schu

lze

(199

6)W

heat

;soy

bean

mea

l;m

eal/m

ash

(M)o

rpe

llete

d(P

)33

–94

kg

Xyla

nase

(400

0U

g−1)

1kg

t−1G

row

thpe

rform

ance

:C

(M):

859

gda

y−1Fe

ed:g

ain

2.46

+E:8

64g

day−1

Feed

:gai

n2.

54C

(P):

836

gda

y−1Fe

ed:g

ain

2.43

+E:8

71g

day−1

Feed

:gai

n2.

30Pe

nfo

ulin

g‘g

reat

lyre

duce

d’in

pelle

ted

diet

s+

enzy

me

van

Lune

nan

dSc

hulz

e(1

996b

)W

heat

:cor

n(6

0:0

);(40

:20)

;(20

:40)

;w

heat

mid

dlin

gs;s

oybe

anm

eal;

pelle

ted

22–8

0kg

Xyla

nase

(400

0U

g−1)

1kg

t−1O

vera

llgr

owth

perfo

rman

ce:

C:9

80g

day−1

Feed

:gai

n2.

78+E

:107

0g

day−1

*Fe

ed:g

ain

2.63

*

Liet

al.(

1996

a)W

heat

;soy

bean

mea

l;m

eal/m

ash

7–11

kgβ-

gluc

anas

e(1

000

Ug−1

)2

kgt−1

Ileal

dige

stib

ility

ofpr

otei

n(%

):C

:68.

8+E

:75.

9

Ileal

dige

stib

ility

ofen

ergy

(%):

C:6

6.9

+E:7

1.2

Liet

al.(

1996

b)W

heat

;soy

bean

mea

l(W

)R

ye;s

oybe

anm

eal(

R)

mea

l/mas

h6–

11kg

β-gl

ucan

ase,

1000

Ug−1

(ado

se-re

spon

sest

udy

–th

enu

mer

ical

lym

ax.r

espo

nses

are

show

n)2

kgt−1

(W)

1kg

t−1(R

)

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C(W

):85

.1+E

:89.

1C

(R):

87.0

+E:8

9.3

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

(W):

86.8

+E:8

8.4

C(R

):87

.2+E

:88.

1



182 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d,le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Hab

erer

etal

.(19

97)

Rye

:bar

ley

(50

:50)

;whe

atbr

an;

soyb

ean

mea

l;m

eal/m

ash,

wet

fed

26–4

0kg

Xyla

nase

(800

FXU

);β-

gluc

anas

e(7

5FBG

)75

0g

t−1

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C:7

4.9

+E:7

6.3

Mea

sure

dD

E(M

Jkg

−1dr

ym

atte

r):C

:14.

33+E

:14.

83*

Yin

etal

.(19

97)

Whe

at;w

heat

/whe

atm

iddl

ings

;w

heat

/whe

atbr

an~2

0kg

ileal

-can

nula

ted

pigs

Xyla

nase

(400

0U

g−1)

1kg

t−1Ile

aldi

gest

ibilit

yof

prot

ein

(%):

C:7

6.5

+E:7

7.8*

Ileal

dige

stib

ility

ofen

ergy

(%):

C:7

0.4

+E:7

1.4

Dus

elet

al.

(199

7b)

and

Jero

chet

al.

(199

9)

Whe

at(h

igh

extra

ctvi

scos

ity);

rye;

barle

y;so

ybea

nm

eal;

pelle

ted

10–2

5kg

E1:x

ylan

ase

(500

0U

g−1)1

kgt−1

or E2:x

ylan

ase

(500

0U

g−1)

Prot

ease

(500

Ug−1

)1

kgt−1

Gro

wth

perfo

rman

ce:

C:3

54g

day−1

Feed

:gai

n1.

92+E

1:46

2g

day−1

*Fe

ed:g

ain

1.66

*+E

2:48

0g

day−1

*Fe

ed:g

ain

1.65

*D

E(M

Jkg

−1dr

ym

atte

r):C

:13.

87+E

1:15

.08*

Rat

tay

etal

.(19

98)

Barle

y;w

heat

;whe

atbr

an;s

oybe

anm

eal

mea

l/mas

h−/

+Av

ilam

ycin

(A;4

0m

gkg

−1)

23–4

5kg

Xyla

nase

(400

0U

g−1)

1kg

t−1Fa

ecal

dige

stib

ility

ofpr

otei

n(%

):C

:77.

9+A

:82.

3*+E

:81.

6*+A

+E:8

2.4*

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

:80.

5+A

:83.

3*+E

:83.

1*+A

+E:8

3.5*

Partr

idge

etal

.(1

998a

)W

heat

:bar

ley

(50

:50)

;soy

bean

mea

l–C

1W

heat

:mai

ze(5

0:5

0);w

heat

polla

rd;

soyb

ean

mea

l–C

2Bo

thdi

ets

pelle

ted

8–30

kg

Xyla

nase

; β-g

luca

nase

;am

ylas

e;pe

ctin

ase

(400

0;15

0;10

00;2

5U

g−1)

1kg

t−1

Gro

wth

perfo

rman

ce:

Cl:

575

gda

y−1Fe

ed:g

ain

1.67

+E:5

81g

day−1

Feed

:gai

n1.

59*

C2:

512

gda

y−1Fe

ed:g

ain

1.69

+E:5

57g

day−1

*Fe

ed:g

ain

1.62

*

Tabl

e8.

3.C

ontin

ued.





Cal

lese

n(1

998)

Whe

at;b

arle

y;w

heat

bran

;soy

bean

mea

lpe

llete

d(8

1 °C

min

.)30

–97

kg

Xyla

nase

(400

0U

g−1)

1kg

t−1G

row

thpe

rform

ance

:C

:824

gda

y−1FU

kg−1

gain

2.66

+E:8

36g

day−1

FUkg

−1ga

in2.

59Pr

otei

nef

ficie

ncy

valu

e(p

rote

inco

nsum

ptio

n/kg

lean

mea

tgai

n):

C:9

66g

+E:9

29g*

Gro

ssm

argi

n/pe

npl

ace/

year

:C

:658

DKK

+E:7

06D

KK*

Partr

idge

etal

.(1

999)

Whe

at–

high

(H)v

ersu

sm

ediu

m(M

)ver

sus

low

(L)p

erfo

rman

ce*;

pelle

ted

28–6

0kg

*Det

erm

ined

inw

eane

rgro

wth

trial

s

Xyla

nase

(400

0U

g−1)

1kg

t−1G

row

thpe

rform

ance

:H

:960

gda

y−1

M:9

18g

day−1

+E:9

45g

day−1

L:87

8g

day−1

*+E

:952

gda

y−1

Feed

:gai

n:H

:1.8

4M

:1.8

0+E

:1.8

1L:

1.76

+E:1

.81

Dre

sche

l eta

l.(1

999)

Triti

cale

;rye

;bar

ley;

whe

at25

–110

kgXy

lana

se;β

-glu

cana

se,3

00an

d20

0g

t−1in

grow

eran

dfin

ishe

rpha

ses,

resp

ectiv

ely

Gro

wth

perfo

rman

ce:

C:8

65g

day−1

+E:8

93g

day−1

Ener

gyef

ficie

ncy:

C:4

4.0

MJ

ME

kg−1

gain

+E:4

2.4

MJ

ME

kg−1

gain

Jero

chet

al.(

1999

)W

heat

(86%

ofdi

et),

low

(L)o

rhig

h(H

)ex

tract

visc

osity

;pel

lete

d10

–25

kg

Xyla

nase

(500

0U

g−1)

1kg

t−1G

row

thpe

rform

ance

:C

(L):

376

gda

y−1Fe

ed:g

ain

1.96

+E(L

):38

3g

day−1

Feed

:gai

n1.

93C

(H):

398

gda

y−1Fe

ed:g

ain

1.90

+E(H

):40

0g

day−1

Feed

:gai

n1.

80D

E(m

easu

red,

MJ

kg−1

dry

mat

ter):

C(L

):16

.23

+E:1

6.29

C(H

):15

.91

+E:1

6.07



184 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d,le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Cho

ctet

al.(

1999

)H

igh

(H),

med

ium

(M)a

ndlo

w(L

)fee

din

take

whe

ats

–pr

edet

erm

ined

7–16

kg

Xyla

nase

;β-g

luca

nase

;cel

lula

se12

0g

t−1G

row

thpe

rform

ance

:C

(L):

230

gda

y−1Fe

ed:g

ain

1.38

+E(L

):46

6g

day−1

*Fe

ed:g

ain

1.23

C(M

):42

5g

day−1

Feed

:gai

n1.

27+E

(M):

445

gda

y−1Fe

ed:g

ain

1.20

C(H

):46

0g

day−1

Feed

:gai

n1.

14+E

(H):

479

gda

y−1Fe

ed:g

ain

1.20

Gill

etal

.(20

00)

Whe

at/s

oybe

anm

eal(

W);

Whe

at/b

eetp

ulp/

soyb

ean

mea

l(W

SBP)

;pe

llete

d(7

5 °C

)8–

18kg

Xyla

nase

;am

ylas

e;pe

ctin

ase

(W)

β-gl

ucan

ase;

amyl

ase;

pect

inas

e(W

SBP)

1kg

t−1

Gro

wth

perfo

rman

ce:

C(W

):33

8g

day−1

Feed

:gai

n1.

53+E

(W):

360

gda

y−1Fe

ed:g

ain

1.48

C(W

SBP)

:345

gda

y−1Fe

ed:g

ain

1.56

+E(W

SBP)

:363

gda

y−1Fe

ed:g

ain

1.48

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C(W

SBP)

72.1

+E(W

SBP)

73.9

DE

(MJ

kg−1

)C

(WSB

P)14

.10

+E(W

SBP)

14.3

3

Tabl

e8.

3.C

ontin

ued.



control rations, suggesting that some degree of nutrient ‘sparing’ may also havecontributed to the productive response seen. These studies illustrate the difficulty ininterpreting responses, or apparent lack of responses, to exogenous enzymes in thepig unless many parameters are examined synchronously.

Maize and Sorghum as Grain Sources for Pigs

Systematic studies on variability in the feeding value of maize for pigs are relativelydifficult to find in the scientific literature. Considering the global importance ofmaize as the key feed raw material in pig and poultry diets, this apparent dearth ofinformation is surprising but has been remarked upon by others (e.g. Dale, 1994).The oft-held perception that maize, assuming it is mycotoxin-free, is a grain withconsistent feeding value is clearly erroneous and this has been demonstrated in anumber of poultry trials (Leeson et al., 1993; Dale, 1994) and alluded to in the pig(Leigh, 1994) (Table 8.6). It is equally clear that this observed variability in animalperformance on maize-based diets is a function of many interacting factors, includingbushel weight, drying conditions, particle size after grinding, variety (e.g. waxy versusdent) and amino acid and oil content. Research in recent years has concentrated



Wheatperformance

−/+Xylanase

Finish weight(kg)

Daily gain(g)

Daily feedintake (g) Feed : gain

HighMediumMediumLowLowSEM

−−+−+

61.960.261.258.361.5

(NS)0.651(NS)

960ab

918ab

945ab

878bb

952ab

12.4*

1772ab

1622ab

1710ab

1576bb

1728ab

33*

1.841.801.811.761.81

(NS)0.021(NS)

abValues within columns with different superscripts are significantly different (P < 0.05).*P < 0.05. NS = not significant.

Table 8.4. Effect of xylanase addition to wheats preselected for productive performance(determined in prior growth studies with weaner pigs), 28–60 kg (Partridge et al., 1999).

Wheat extract viscosity*Daily gain

(g)Daily feed

(g) Feed : gainViscosity in the small

intestine (mPa s)

Low (positive control)HighHigh + xylanaseHigh + xylanase + protease

388ab

354bb

462ab

480ab

717727766794

1.85b

1.92b

1.66a

1.65a

3.3a

6.4b

3.9a

3.1a

abValues within columns with different superscripts are significantly different (P < 0.05).*Low = 1.3 mPa s; high = 3.3 mPa s.

Table 8.5. Effect of enzyme addition to a wheat-based diet with a high extract viscosity, piglets10–25 kg. Dusel et al. (1997b) and personal communication (viscosity values).



less on maize variability as an issue and more on the so-called ‘designer’ versions ofthe maize crop – e.g. high-lysine, high-oil, low-phytate – all with potentially highagronomic value for the future.

For sorghum, van Barneveld (1999) reviewed a number of digestibility studieswith pigs and found that DE estimates ranged from 15.8 to 17.4 MJ kg−1 DM, withenergy digestibility coefficients ranging from 0.72 to 0.92. Kopinski (1997) foundthat the largest variation between cultivars at a single study site was 0.71 MJ kg−1

DM and within a cultivar at different sites it was 0.5 MJ kg−1 DM. Kemm and Brand(1996), summarizing trials with samples of low- and high-tannin sorghum, foundthat high-tannin varieties had reduced DE content (10–16%) compared withlow-tannin varieties. Feed conversion was correspondingly poorer on high-tanninsorghum when compared with low-tannin sorghum and maize meal.

Response of Maize and Sorghum to Exogenous Enzymes inPigs

The NSP levels in maize and sorghum are similar (9–10%; Dierick and Decuypere,1994) with insoluble arabinoxylans comprising over 40% of this total. It is known

186 G.G. Partridge


Bushel weight Daily gain (lb) Daily feed (lb) Feed : gain

Low (41 lb, 9.7% CP)Medium (48 lb, 8.6% CP)Heavy (54 lb, 8.7% CP)

1.661.691.74

5.896.446.21

3.563.823.56

Table 8.6. Effect of corn bushel weight on performance of growing pigs, 63–200 lb (Iowa StateUniversity, 1975, cited by Leigh, 1994).

4

3

2

1

0Stomach Duodenum Jejunum Ileum Caecum

High viscous + xylanase/protease High viscous Low viscous

Fig. 8.6. Effect on microbial populations in digesta when piglets are fed dietsbased on high viscous wheat −/+ xylanase/protease, or low viscous wheat.Microscopical analysis of digesta (VTT Biotechnology and Food Research, Finland):0 = few microbes present, 4 = high levels of microbes present. Samples from thestudies of Dusel et al. (1997a).



that standard milling procedures (e.g. roller milling or hammer milling throughscreens of varying diameter) still result in intact packages of cell wall material,with enclosed nutrients, entering the stomach and small intestine. As the cell wallstructure in these particular grains is predominantly insoluble fibre, and unresponsiveto endogenous enzymes, varying quantities of packaged material have been observedat the end of the small intestine, entering the hindgut (M.R. Bedford, personalcommunication, 2000). Exogenous enzyme activities of particular relevance to thesefibre components in maize and sorghum are xylanase and, to a lesser extent, cellulasedesigned to break down this cell wall ‘packaging’ and disrupt its associated negativefibre properties, e.g. water-holding capacity.

Maize starch digestibility per se will be most influenced by factors such asthe amylose : amylopectin ratio, which will vary particularly when comparing waxy(high amylopectin) versus dent maize. The recent interest in high amylose maizestarch as an ingredient in human foods, to stimulate potentially beneficial butyratefermentation in the hindgut, is a reminder of the importance of this factor. Toppinget al. (1997) fed pigs on diets containing either standard maize starch or a highamylose material and observed large differences in the amounts of starch reaching thecaecum (Fig. 8.7). This situation has implications, not only for nutrient digestibilityand availability in the small intestine, but also with respect to substrate-relatedchanges in the microflora in the hindgut. The relevance to exogenous enzymes is thatvarious amylase sources, having different properties to endogenous amylase, couldhave a potential role to play in improving starch digestibility in the small intestineand thereby indirectly influence the microbial population in the gut.

Unfortunately, relatively few studies have been reported on the effects ofexogenous enzymes on ‘simple’ maize/soybean or sorghum/soybean diets inpigs (Table 8.7). Lindemann et al. (1997) found positive responses in the growerphase (26–64 kg) after addition of a complex enzyme blend (protease, cellulase,pentosanase, galactosidase and amylase) to a maize/soybean meal-based ration. Thesame enzyme combination gave no response when 20% wheat mill by-product was



8.4

39.1

3.5

3.9

58.0

5.5Time after feeding (h)

Fig. 8.7. Concentration of starch (mg g−1 dry matter) in caecal digesta of pigsfed diets containing maize starch or high amylose starch; mean of two pigs/treatment. (, maize starch; %, high amylose starch. (Topping et al., 1997.)



188 G.G. Partridge


Ref

eren

ceBa

sald

iet–

mai

nin

gred

ient

s;ag

e/w

eigh

tof

pig;

pelle

ted

orm

eal/m

ash

(whe

rest

ated

)En

zym

e(s)

adde

d;le

vels

and

incl

usio

nra

tes

(whe

rest

ated

)R

espo

nses

seen

(C=

cont

rol;

+E=

plus

enzy

me;

* P<

0.05

)

Liet

al.(

1996

b)M

aize

;soy

bean

mea

l;m

eal/m

ash

6–11

kgβ-

gluc

anas

e,10

00U

g−1(a

dose

resp

onse

stud

y–

the

num

eric

ally

max

imum

resp

onse

sar

esh

own)

2kg

t−1

Faec

aldi

gest

ibilit

yof

prot

ein

(%):

C:8

4.4

+E:8

2.7

Faec

aldi

gest

ibilit

yof

ener

gy(%

):C

:85.

8+E

:85.

7

Schu

lze

etal

.(1

996a

)M

aize

;soy

bean

mea

l;ric

ebr

an;m

eal/m

ash

46–9

3kg

E1:x

ylan

ase

#1E2

:xyl

anas

e#1

+am

ylas

eE3

:xyl

anas

e#2

Gro

wth

perfo

rman

ce:

C:7

00g

day−1

Feed

:gai

n3.

10+E

1:74

1g

day−1

Feed

:gai

n2.

82*

+E2:

725

gda

y−1Fe

ed:g

ain

2.93

+E3:

763

gda

y−1*

Feed

:gai

n2.

97

Lind

eman

net

al.

(199

7)C

H:m

aize

;soy

bean

mea

l(‘h

igh’

ener

gy)

CL:

mai

ze;s

oybe

anm

eal;

whe

atby

-pro

duct

s(‘l

ow’e

nerg

y)26

–109

kg

Prot

ease

;cel

lula

se;p

ento

sana

se;

α-ga

lact

osid

ase;

amyl

ase

1kg

t−1

Gro

wth

perfo

rman

ce:

CH

:801

gda

y−1Fe

ed:g

ain

2.97

+E:8

60g

day−1

Feed

:gai

n2.

75C

L:77

1g

day−1

Feed

:gai

n3.

48+E

:748

gda

y−1Fe

ed:g

ain

3.42

Enzy

me

×en

ergy

leve

lP<

0.05

Schu

lze

and

Cam

pbel

l(19

98)

Mai

ze;s

oybe

anm

eal;

whe

atm

iddl

ings

;pe

llete

d42

–74

kg

E1:x

ylan

ase

#1E2

:xyl

anas

e#2

Gro

wth

perfo

rman

ce:

C:7

06g

day−1

Feed

:gai

n2.

30+E

1:78

1g

day−1

Feed

:gai

n1.

83*

+E2:

765

gda

y−1Fe

ed:g

ain

2.40

Tabl

e8.

7.So

me

publ

ishe

dst

udie

son

the

effe

cts

ofen

zym

epr

epar

atio

nson

mai

zeor

sorg

hum

-bas

eddi

ets

−/+

milli

ngby

-pro

duct

s(N

B:w

here

both

ileal

and

faec

aldi

gest

ibilit

ym

easu

rem

ents

wer

em

ade

then

data

are

pres

ente

dpr

efer

entia

llyat

the

ileal

leve

l).





Partr

idge

etal

.(1

998b

)M

aize

;whe

atpo

llard

;soy

bean

mea

l;m

eal/m

ash

37–7

8kg

Xyla

nase

(400

0U

g−1)

1kg

t−1G

row

thpe

rform

ance

attw

osi

tes:

C:6

59g

day−1

Feed

:gai

n3.

07+E

:694

gda

y−1Fe

ed:g

ain

2.76

*C

:704

gda

y−1Fe

ed:g

ain

3.25

+E:7

51g

day−1

*Fe

ed:g

ain

3.11

Kim

etal

.(19

98)

Sorg

hum

,soy

bean

mea

l;m

eal/m

ash

51–1

18kg

Cel

lula

se50

0g

t−1G

row

thpe

rform

ance

:C

:930

gda

y−1Fe

ed:g

ain

3.51

+E:9

60g

day−1

Feed

:gai

n3.

52Fa

ecal

dige

stib

ility

ofpr

otei

n(%

):C

:65.

1+E

:66.

8Fa

ecal

dige

stib

ility

ofen

ergy

(%):

C:8

0.6

+E:8

1.3

Evan

gelis

taet

al.

(199

8)M

aize

;soy

bean

mea

l;m

eal/m

ash

28–8

5kg

Amyl

ase;

β-gl

ucan

ase;

xyla

nase

;pe

ctin

ase

(100

0;15

0;40

00;2

5U

g−1)1

kgt−1

Gro

wth

perfo

rman

ce:

Feed

:gai

n:C

:792

gda

y−12.

83+E

:853

gda

y−1*

2.82

Liet

al.(

1999

)C

:Mai

ze;s

oybe

anm

eal

CA:

Mai

ze;s

oybe

anm

eal+

acid

ifier

;m

eal/m

ash

10–2

1kg

Amyl

ases

; β-g

luca

nase

;pec

tinas

e;pr

otea

ses;

cellu

lase

700

gt−1

Gro

wth

perfo

rman

ce:

C:3

80g

day−1

Feed

:gai

n1.

78+E

:412

gda

y−1Fe

ed:g

ain

1.62

CA:

399

gda

y−1Fe

ed:g

ain

1.72

+E:4

01g

day−1

Feed

:gai

n1.

67

Partr

idge

etal

.(1

999)

Mai

ze;w

heat

by-p

rodu

ct;p

elle

ted

28–6

0kg

Xyla

nase

(400

0U

g−1)

1kg

t−1G

row

thpe

rform

ance

:C

:886

gda

y−1Fe

ed:g

ain

1.88

+E:9

38g

day−1

Feed

:gai

n1.

82



added to a lower energy-dense version of the same ration. Li et al. (1999) found nosignificant effects of adding an enzyme blend (amylases, β-glucanase, pectinase,proteases, cellulase) to maize-based diets for weaner pigs, with or without an acidifierin the ration. Similarly Kim et al. (1998) failed to show benefits of adding cellulase toa sorghum/soybean diet fed to grower/finishers. In contrast, Evangelista et al. (1998)described significant responses in growth rate and feed intake to an enzyme blend(amylase, β-glucanase, xylanase, pectinase) offered to grower/finisher pigs.

Van Lunen and Schulze (1996b) reported significant responses to xylanaseindependent of the maize : wheat ratio in diets for grower/finishers. However, thediet containing the highest level of maize (400 g kg−1) still contained 20% wheat and10% wheat by-products, making interpretation of the response in relation to specificsubstrates more difficult.

When maize/soybean-based rations are supplemented with various millingby-products (e.g. wheat middlings/pollard/bran; extracted and full-fat rice bran)a number of trials have shown a beneficial effect of exogenous enzyme supple-mentation. Schulze et al. (1996a), Schulze and Campbell (1998) and Partridge et al.(1998b) all showed significant benefits to xylanase supplementation on growth rateand/or feed conversion ratio in growing/finishing pigs.

Overall, it is clear that the reported responses to enzymes in pure maize/soybeanor sorghum/soybean diets are, so far, limited and more work is needed to understandthe underlying reasons for variability in these raw materials before consistentlycost-effective carbohydrase enzyme solutions can be offered for pigs. At the sametime it is clear that when such diets are supplemented with certain fibrousby-products, containing significant quantities of insoluble cell wall material, thenthe potential for a response to an effective enzyme source (particularly xylanase) isconsiderably magnified. In commercial practice such by-products are frequentlyadded to maize-based diets for grower/finisher pigs to offer some savings in costsof production, so the economic opportunities for cost-effective enzyme additionbecome a realistic opportunity in these situations.

Implications for Enzyme Use in Diet-induced DiseaseSyndromes

Recent studies have highlighted the important interactions between diet compositionand microflora changes in the pig gut under both chronic and acute diseasechallenge. In Australia, Pluske and co-workers have elegantly illustrated the rolethat diet composition plays in the development of swine dysentery, provoked bythe anaerobic spirochaetal bacterium Serpulina hyodysenteriae. Both soluble,fermentable fibre and resistant starch from various feed raw materials have beenfound to be particularly provocative to this disease (Pluske et al., 1996, 1998)(Fig. 8.8). To date, attempts to influence this particular condition by the strategic useof carbohydrase enzymes have given equivocal results (Durmic et al., 1997) but theconcept is receiving increased research attention with the continued emphasis onprophylaxis, rather than medication therapy, in modern pig production systems.

190 G.G. Partridge




Undigested soluble fibre has already been identified as one of the key influenceson the non-specific colitis syndrome affecting, particularly, pigs in the 15–40 kgweight range offered pelleted, wheat-based rations (Taylor, 1989). Adding anappropriate xylanase has been shown to have a positive influence on this syndrome(Hazzledine and Partridge, 1996), negating the need for meal feeding or potentiallycostly diet reformulation on affected units.



Fig. 8.8. Incidence of swine dysentery, in relation to soluble non-starchpolysaccharide concentration and resistant starch concentration in the diet(Pluske et al., 1996).

201

Other studies have similarly reported positive interactions between the use ofenzymes and diet-induced diarrhoea, particularly in the post-weaned pig (Inborrand Ogle, 1988; Florou-Paneri et al., 1998; Kantas et al., 1998). These, togetherwith reports on the potential synergies between enzymes and therapeutic andsubtherapeutic antibiotic addition (Florou-Paneri et al., 1998; Kantas et al., 1998;Gollnisch et al., 1999), offer interesting possibilities for future production methodswhere strategic rather than routine antibiotic use will become the norm.

Future Trends in Enzyme Research for Pig Applications?

The review of many trials published in the last 10 years, across a range of dietarysubstrates (Tables 8.1, 8.3 and 8.7), suggests that the question to be asked in the nextfew years is not so much: ‘Do enzymes work in pig diets?’ but far more: ‘When doproven enzymes for pig application give their most cost-effective responses?’ Enzymetechnology has moved apace over the last decade to bring, amongst others, a rangeof carbohydrase activities that have the potential to have a major impact on theeconomics of pig production. However, some key challenges remain.

1. To identify quickly, cheaply and easily, the key parameters in grains and theirby-products that appear to limit both voluntary food intake and nutrient availability inthe growing pig (e.g. Schulze et al., 1997; Cadogan et al., 1999; Partridge et al., 1999).Then to be able to relate these to responses to specific enzyme additions, given thatdifferent sources of carbohydrases (e.g. xylanases, β-glucanases, amylases) will havediffering characteristics and relative efficacies both in vitro and in vivo.2. To ensure, contrary to a number of published studies, that scientific trialsinvolving enzymes are set up in the most appropriate manner to have the opportunityto show an effect in a marginally nutrient-limited animal. This will almost invariablyinvolve some prior response modelling in animals whose lean deposition rateshave been predetermined. At the same time in such trials, to superimpose somepredetermined aspects of raw material feeding value, determined in vitro (e.g. Chenet al., 1997; Cadogan et al., 1999; Moughan et al., 1999). More attention to detail indescribing diet characteristics, particularly fibre analysis and properties, as well asenzyme sources and levels, will also aid interpretation across a range of studies.3. To understand better the role that enzymes have to play in gut physiology,particularly aspects such as digesta passage rate and patterns of fermentation indifferent parts of the digestive tract. The effects of these on gut hormone secretionand the physiological control of voluntary food intake would also merit further study.4. To understand better the potential role for feed enzymes in manipulating thegut microflora with a view to their future use with synergistic additives in productionsystems where there is far less reliance on prophylactic and therapeutic medication.Fundamental work on how enzymes can potentially influence fermentation (e.g.volatile fatty acid production) in the fore- and hindgut will be particularly valuable inthe pig and offers good opportunities to manipulate the gut microflora in a positiveway.

192 G.G. Partridge




5. To understand more clearly the interactions between different enzymes,particularly carbohydrases, phytases and proteases, in order to maximize their costeffectiveness across a range of dietary substrates and situations.6. To maximize the use of appropriate enzymes in novel pig feeding applicationsinvolving preprocessing of raw materials, with or without liquid feeding technology.

References

Aman, P. and Newman, C.W. (1986) Chemical composition of some different types of barleygrown in Montana, USA. Journal of Cereal Science 4, 133–141.

Aman, P., Hessleman, K. and Tilly, A.-C. (1985) The variation in chemical composition ofSwedish barleys. Journal of Cereal Science 3, 73–77.

Apajalahti, J. and Bedford, M.R. (1998) Nutrition effects on the microflora of the GI tract.Proceedings of the 19th Western Nutrition Conference. Saskatoon, Canada, pp. 60–68.

Autio, K., Mannonen, L., Pietila, K., Kosinken, M., Siika-aho, M. and Linko, M. (1996)Incubation of barley kernel sections with purified cell wall degrading enzymes. Journal ofthe Institute of Brewing 102, 427–432.

Bach Knudsen, K.E., Aman, P. and Eggum, B.O. (1987) Nutritive values of Danish-grownbarley varieties. I. Carbohydrates and other major constituents. Journal of Cereal Science6, 173–186.

Baidoo, S.K. and Liu, Y.G. (1998) Hulless barley for swine: ileal and faecal digestibility ofproximate nutrients, amino acids and non-starch polysaccharides. Journal of the Science ofFood and Agriculture 76, 397–403.

Baidoo, S.K., Liu, Y.G. and Grandhi, R.R. (1998a) Exogenous enzymes and hulless barleyutilization by young pigs. In: Proceedings of the 1998 Feed Grain Quality Conference.8th–10th November, Edmonton, Alberta, Canada, pp. 67–71.

Baidoo, S.K., Liu, Y.G. and Yungblut, D. (1998b) Effect of microbial enzyme supple-mentation on energy, amino acid digestibility and performance of pigs fed hulless barleybased diets. Canadian Journal of Animal Science 78, 625–631.

Bayzlo, R.B. (1990) Rye. In: Thacker, P.A. and Kirkwood, R.N. (eds) Nontraditional FeedSources for Use in Swine Production. Butterworths, Stoneham, Maryland, pp. 363–372.

Bedford, M.R. and Schulze, H. (1998) Exogenous enzymes in pigs and poultry. NutritionResearch Reviews 11, 91–114.

Bedford, M.R., Patience, J.F., Classen, H.L. and Inborr, J. (1992) The effect of dietaryenzyme supplementation of rye- and barley-based diets on digestion and subsequentperformance in weanling pigs. Canadian Journal of Animal Science 72, 97–105.

Bohme, H. (1990) Untersuchungen zur Wirksamkeit von Enzymzusatzen in derFerkelaufzucht. Landbauforschung Volkenrode 3, 213–217.

Cadogan, D.J., Choct, M., Campbell, R.G. and Kershaw, S. (1999) Effects of new season’swheat on the growth performance of young male pigs. In: Cranwell, P.D. (ed.)Manipulating Pig Production. Proceedings of the 7th Conference of the AustralasianPig Science Association, Adelaide, p. 40.

Callesen, J. (1998) Porzyme 9300 for finishers. Landsudvalget For Svin, Danske SlagterierReport no. 403, 7 pp.

Campbell, R.G., Whitaker, A., Hastrup, T. and Rasmussen, P.B. (1995) Interrelationshipbetween protein source and Biofeed Plus on the performance of weaned pigs from 23 to





42 days of age. Proceedings of the 5th Conference of the Australasian Pig Science Association.Canberra, p. 194.

Charmley, E. and Greenhalgh, J.F.D. (1987) Nutritive values of three cultivars of barley forsheep, pigs and poultry. Animal Feed Science and Technology 18, 19–35.

Chen, J., Boisen, S., Pearson, G., Morel, P.C.H. and Moughan, P.J. (1995) An in vitromethod for predicting digestible energy applied to milling by-products. In: Hennessy,D.P. and Cranwell, P.D. (eds) Manipulating Pig Production. Proceedings of the 5thConference of the Australasian Pig Science Association, Canberra, p. 40.

Chesson, A. (1993) Feed enzymes. Animal Feed Science and Technology 45, 65–79.Choct, M., Cadogan, D.J., Campbell, R.G. and Kershaw, S. (1999) Enzymes can eliminate

the difference in the nutritive value of wheats for pigs. In: Cranwell, P.D. (eds.) Manipu-lating Pig Production. Proceedings of the 7th Conference of the Australasian Pig ScienceAssociation, Adelaide, p. 39.

Cyran, M., Rakowska, M., Wasilewko, J. and Buraczewska, L. (1995) Degradation of dietaryfibre polysaccharides of rye in the intestinal tract of growing pigs used as a model animalfor studying digestion in humans. Journal of Animal Feed Science 4, 217–227.

Dale, N. (1994) Matching corn quality and nutritional value. Feed Mix 2, 26–29.Danicke, S., Dusel, G., Jeroch, H. and Kluge, H. (1999) Factors affecting efficiency of

NSP-degrading enzymes in rations for pigs and poultry. Agribiological Research 52,1–24.

Dierick, N.A. (1989) Biotechnology aids to improve feed and feed digestion: enzymes andfermentation. Archives of Animal Nutrition, Berlin 39, 241–261.

Dierick, N.A. and Decuypere, J.A. (1994) Enzymes and growth in pigs. In: Cole, D.J.A.,Wiseman, J. and Varley, M.A. (eds) Principles of Pig Science. Nottingham UniversityPress, Nottingham, UK, pp. 169–195.

Dierick, N. and Decuypere, J. (1996) Mode of action of exogenous enzymes in growing pignutrition. Pig News Information 17, 41N–48N.

Dreschel, H., Hackl, W., Matthes, W., Stolken, B. and Hennig, U. (1999) Investigations onsupplements of NSP-hydrolysing enzymes to diets for pigs. Landbauforschung Volkenrode193, 343–348.

Durmic, Z., Pethick, D.W., Mullan, B.P., Schulze, H. and Hampson, D.J. (1997) The effectsof extrusion and enzyme addition in wheat based diets on fermentation in the largeintestine and expression of swine dysentery. In: Cranwell, P.D. (ed.) Manipulating PigProduction, Proceedings of the 6th Biennial Conference of the Australasian Pig ScienceAssociation. Canberra 7th–10th December 1997, p.180.

Dusel, G., Kluge, H., Simon, O., Jeroch, H. and Schulze, H. (1997a) [Investigation onthe effect of NSP-degrading enzymes in wheat-based diets on intestinal viscosityand nutrient digestibility of piglets.] Proceedings of the Society of Nutrition and Physiology6 (in German).

Dusel, G., Schulze, H., Kluge, H., Simon, O. and Jeroch, H. (1997b) The effect of wheatquality measured by extract viscosity and dietary addition of feed enzymes onperformance of young pigs. Journal of Animal Science 75 (Suppl. 1), 200.

Ellis, P.R., Rayment, P. and Wang, Q. (1996) A physio-chemical perspective of plantpolysaccharides in relation to glucose absorption, insulin secretion and the entero-insularaxis. Proceedings of the Nutrition Society 55, 881–898.

Evangelista, J.N.B., Sales Bastos, F.J. and Amarante, V. (1998) [Use of enzymes in maize–soyadiets for pigs.] Anaporc 177, 89–97.

Evers, A.D., Blakeney, A.B. and O’Brien, L.O. (1999) Cereal structure and composition.Australian Journal of Agricultural Research 50, 629–650.

194 G.G. Partridge




Fairbairn, S.L., Patience, J.F., Classen, H.L. and Zijlstra, R.T. (1999) The energy content ofbarley fed to growing pigs: characterising the nature of its variability and developingprediction equations for its estimation. Journal of Animal Science 77, 1502–1512.

Florou-Paneri, P., Kantas, D., Alexopoulos, C., Tsinas, A.C., Vassilopoulos, V. and Kyriakis,S.C. (1998) A comparative study of the effect of a dietary multi-enzyme system and/orvirginiamycin on weaned piglet performance. Proceedings of the 15th IPVS Congress.Birmingham, UK, p. 35.

Forbes, J.M. (1995) Voluntary Food Intake and Diet Selection in Farm Animals. CABInternational, Wallingford, UK.

Gill, B.P. and Schulze, H. (1996) Effect of diet processing and enzyme supplementation ofwheat-based diets on growing pig performance. Animal Science 62, 662.

Gill, B.P., Mellange, J. and Rooke, J.A. (2000) Growth performance and apparent nutrientdigestibility in weaned piglets offered wheat-, barley- or sugar-beet pulp-based dietssupplemented with food enzymes. Animal Science 70, 107–118.

Gollnisch, K., Vahjen, W., Simon, O. and Schulz, E. (1999) [Influence of an antimicrobial(avilamycin) and an enzymatic (xylanase) feed additive alone or in combination onpathogenic microorganisms in the intestine of pigs (E. coli, C. perfringens).] Landbau-forschung Volkenrode 193, 337–342.

Graham, H. and Balnave, D. (1995) Dietary enzymes for increasing energy availability. In:Wallace, R.J. and Chesson, A. (eds) Biotechnology in Animal Feeds and Animal Feeding.VCH, Weinheim, Germany, pp. 295–309.

Graham, H., Hessleman, K., Jonsson, E. and Aman, P. (1986) Influence of β-glucanasesupplementation on digestion of a barley-based diet in the pig gastrointestinal tract.Nutrition Reports International 34, 1089–1096.

Graham, H., Lowgren, W., Petterson, D. and Aman, P. (1988) Effect of enzymesupplementation on digestion of a barley/pollard-based pig diet. Nutrition ReportsInternational 38, 1073–1079.

Graham, H., Fadel, J.G., Newman, C.W. and Newman, R.K. (1989) Effect of pelleting andβ-glucanase supplementation on the ileal and fecal digestibility of a barley-based diet inthe pig. Journal of Animal Science 67, 1293–1298.

Haberer, B., Schulz, E. and Flachowsky, G. (1997) [The effect of non-starch polysaccharidedegrading enzymes on the overall digestibility of nutrients and the content ofmetabolizable energy in growing pigs.] Landbauforschung Volkenrode Heft 2, 56–61.

Haberer, B., Schulz, E. and Flachowsky, G. (1998) Effects of β-glucanase and xylanasesupplementation in pigs fed a diet rich in nonstarch polysaccharides: disappearance anddisappearance rate of nutrients including the nonstarch polysaccharides in stomachand small intestine. Journal of Animal Physiology and Animal Nutrition 78, 95–103.

Han, M.S. and Froseth, J.A. (1993) Effect of enzyme supplementation on the utilisationof high and low quality barley in pigs as measured by the mobile nylon bag technique.Journal of Animal Science 71, (Supplement 1), 361A, p. 180.

Hazzledine, M. and Partridge, G.G. (1996) Enzymes in animal feeds – application technologyand effectiveness. Proceedings of the 12th Annual Carolina Swine Nutrition Conference.pp. 12–33.

Inborr, J. (1994) Supplementation of pig starter diets with carbohydrate-degrading enzymes –stability, activity and mode of action. Agricultural Science in Finland 3 (Suppl. no. 2).

Inborr, J. and Ogle, R.B. (1988) Effect of enzyme treatment of piglet feeds on performanceand post weaning diarrhoea. Swedish Journal of Agricultural Research 18, 129–133.

Inborr, J., Schmitz, M. and Ahrens, F. (1993) Effect of adding fibre and starch degradingenzymes to a barley/wheat based diet on performance and nutrient digestibility in





different segments of the small intestine of early weaned pigs. Animal Feed Science andTechnology 44, 113–127.

Jaikaran, S., Recinos Diaz, G., Hsu, H. and Prommer, E.A. (1998) NIRS predictions foramino acids in whole and ground barley. In: 1998 Feed Grain Quality Conference.8th–10th November, Edmonton, Alberta, Canada, pp. 82–92.

Jeroch, H., Dusel, D., Kluge, H. and Nonn, H. (1999) The effectiveness of microbial xylanasein piglet rations based on wheat or wheat, rye and barley. Landbauforschung Volkenrode193, 223–228.

Kantas, D., Florou-Paneri, P., Vassilopoulos, V. and Kyriakis, S.C. (1998) The effect of adietary multi-enzyme system on piglet post-weaning performance. Proceedings of the 15thIPVS Congress. Birmingham, UK, p. 36.

Kemm, E.H. and Brand, T.S. (1996) Grain sorghum as an energy source for growing pigs.Pig News Information 17, 87N–89N.

Kim, I.H., Hancock, J.D., Hines, R.H. and Kim, C.S. (1998) Effects of cellulase enzymesand bacterial feed additives on the nutritional value of sorghum grain for finishing pigs.American Journal of Animal Science 11(5), 538–544.

Kyriazakis, I. and Emmans, G.C. (1995) The voluntary feed intake of pigs given feeds basedon wheat bran, dried citrus pulp and grass meal, in relation to measurements of feed bulk.British Journal of Nutrition 73, 191–207.

Leeson, S., Yersin, A. and Volker, L. (1993) Nutritive value of the 1992 corn crop. Journal ofApplied Poultry Research 2, 208–213.

Leigh, K. (1994) The unpredictable nature of corn. Pigs December, pp. 37–39.Li, D., Liu, S.D., Qiao, S.Y., Yi, G.F., Liang, C. and Thacker, P. (1999) Effect of feeding

organic acid with or without enzyme on intestinal microflora, intestinal enzyme activityand performance of weaned pigs. Asian-Australian Journal of Animal Science 12, 411–416.

Li, S., Sauer, W.C., Huang, S.X. and Gabert, V.M. (1996a) Effect of β-glucanase supple-mentation to hulless barley or wheat–soybean meal diets on the digestibilities of energy,protein, β-glucans and amino acids in young pigs. Journal of Animal Science 74,1649–1656.

Li, S., Sauer, W.C., Mosenthin, R. and Kerr, B. (1996b) Effect of β-glucanase supple-mentation of cereal-based diets for starter pigs on the apparent digestibilities of drymatter, crude protein and energy. Animal Feed Science and Technology 59, 223–231.

Lindemann, M.D., Gentry, J.L., Monegue, H.J., Cromwell, G.L. and Jacques, K.A. (1997)Determination of the contribution of an enzyme combination (Vegpro) to performancein grower–finisher pigs. Proceedings of the 6th Conference of the Australasian Pig ScienceAssociation. Canberra, p. 247.

McClean, D., McEvoy, J. and McCracken, K.J. (1992) Effects of processing and feed enzymeson nutrient digestibility in diets for weaned pigs. Proceedings of the Nutrition Society 52,211A.

McCracken, K.J.D., Yin, Y.-L., McEvoy, J.D. and Schulze, H. (1996) Effects of dietcomposition, food enzymes and method of cannulation on faecal and ileal digestibilitymeasurements in growing pigs. Animal Science 62, 662.

Miller, M.C., Froseth, J.A., Wyatt, C.L. and Ullrich, S.E. (1994) Effect of starch type, totalβ-glucans and acid detergent fibre levels on the energy content of barley (Hordeumvulgare L.) for poultry and swine. Canadian Journal of Animal Science 74, 679–686.

Moran, E.T. (1982) Comparative Nutrition of Fowl and Swine. The Gastrointestinal Systems.University of Guelph, Canada.

Moughan, P.J. (1999) In vitro techniques for the assessment of the nutritive value of feedgrains for pigs: a review. Australian Journal of Agricultural Research 50, 871–879.

196 G.G. Partridge




Officer, D.I. (1995) Effect of multi-enzyme supplements on the growth performance of pigletsduring the pre- and post-weaning periods. Animal Feed Science and Technology 56,55–65.

Partridge, G.G., de la Fuente, J.M., Flores, A., Sanz, M. and Tan, M.Y. (1998a) The use of amulti-enzyme product in mixed grain diets for young pigs. Proceedings of the 8th WorldConference on Animal Production. Seoul, Korea, pp. 116–117.

Partridge, G.G., Alcantara, P.F. and Creswell, D. (1998b) Effect of xylanase addition tocorn/soybean meal/wheat pollard diets for grower/finisher pigs. Proceedings of the 8thWorld Conference on Animal Production. Seoul, Korea, pp. 626–627.

Partridge, G.G., Simmins, P.H. and Cadogan, D.J. (1999) Influence of xylanase additionto diets containing wheat co-products and nutritionally defined wheat on growing pigperformance. Proceedings of the 7th Conference of the Australasian Pig Science Association.Adelaide, p. 37.

Pettersson, A. and Lindberg, J.E. (1997) Ileal and total tract digestibility in pigs of naked andhulled barley with different starch composition. Animal Feed Science and Technology 66,97–109.

Pluske, J.R., Siba, P.M., Pethick, D.W., Durmic, Z., Mullan, B.P. and Hampson, D.J. (1996)The incidence of swine dysentery in pigs can be reduced by feeding diets that limit theamount of fermentable substrate entering the large intestine. Journal of Nutrition 126,2920–2933.

Pluske, J.R., Durmic, Z., Pethick, D.W., Mullan, B.P. and Hampson, D.J. (1998) Confirma-tion of the role of rapidly fermentable carbohydrates in the expression of swine dysenteryin pigs after experimental infection. Journal of Nutrition 128, 1737–1744.

Power, R., de Koning, W. and Fremant, D. (1996) The effect of β-glucanase supplementationon production parameters in piglets receiving barley-based diets. Animal Science 62,662–663.

Ramaswamy, C.M., Van Lunen, T.A. and Schulze, H. (1996) Effect of xylanase andβ-glucanase on nutrient digestibility in pigs fed hull-less or hulled barley based diets.Proceedings of the Eastern Nutrition Conference. Canada, 15–17 May.

Rattay, D., Gollnisch, K., Schulz, E. and Flachowsky, G. (1998) [The effect of an antimicro-bial feed additive, a non-starch polysaccharide (NSP)-hydrolysing enzyme and theircombination on the digestibility of nutrients and the content of metabolizable energyin growing pigs.] Landbauforschung Volkenrode Heft 4, 169–175.

Schulze, H. and Campbell, R.G. (1998) Effect of exogenous xylanase on performance of pigsfed corn/soya based diets. Journal of Animal Science 76, Supplement 1, p. 179.

Schulze, H., Partridge, G.G. and Creswell, D. (1996a) The effect of feed enzymesupplementation to corn/soya based diets on performance of finisher pigs from 46 to92 kg. Journal of Animal Science 74, Suppl. 1, 191.

Schulze, H., Partridge, G.G. and Gill, B.P. (1996b) The use of different xylanase sources and aprotease in wheat-based diets for weaner pigs. Animal Production 62, 663.

Schulze, H., Pearce, A.N. and Rose, S.P. (1997) The effect of six different wheat cultivars ofsimilar bushel weight on growth performance of young pigs. Journal of Animal Science 75,Supplement 1, 197.

Sudendey, C. and Kamphues, J. (1995) [Effect of enzyme addition (amylase, xylanase andβ-glucanase) on digestive processes in the intestinal tract of force fed weaner pigs.]Proceedings of the Society for Nutrition and Physiology 3, 88A.

Taylor, D.J. (1989) Pig Diseases, 5th edn. Burlington Press, Cambridge, UK, pp. 269–271.



207

Tervila-Wilo, A., Parkkonen, T., Morgan, A., Hopeakoski-Nurminen, M., Poutanen, K.,Heikkinen, P. and Autio, K. (1996) In vitro digestion of wheat microstructure withxylanase and cellulase from Trichoderma reesei. Journal of Cereal Science 24, 215–225.

Thacker, P.A. (1999) Effect of micronisation on the performance of growing/finishing pigs feddiets based on hulled and hulless barley. Animal Feed Science and Technology 79, 29–41.

Thacker, P.A. and Baas, T.C. (1996) Effects of gastric pH on the activity of exogeneouspentosanase and the effect of pentosanase supplementation of the diet on the perfor-mance of growing–finishing pigs. Animal Feed Science and Technology 63, 187–200.

Thacker, P.A., Campbell, G.L. and GrootWassink, J. (1991) The effect of enzyme supple-mentation on the nutritive value of rye-based diets for swine. Canadian Journal of AnimalScience 71, 489–496.

Thacker, P.A., Campbell, G.L. and GrootWassink, J.W.D. (1992a) Effect of salinomycin andenzyme supplementation on nutrient digestibility and the performance of pigs fed barley-or rye-based diets. Canadian Journal of Animal Science 72, 117–125.

Thacker, P.A., Campbell, G.L. and GrootWassink, J.W.D. (1992b) The effect of organicacids and enzyme supplementation on the performance of pigs fed barley-based diets.Canadian Journal of Animal Science 72, 395–402.

Topping, D.L., Gooden, J.M., Brown, I.L., Biebrick, D.A., McGrath, L., Trimble, R.P.,Choct, M. and Ilman, R.J. (1997) A high amylose (amylomaize) starch raises proximallarge bowel starch and increases colon length in pigs. Journal of Nutrition 127, 615–622.

van Barneveld, R.J. (1999) Chemical and physical characteristics of grains related to variabilityin energy and amino acid availablility in pigs: a review. Australian Journal of AgriculturalResearch 50, 667–687.

van Lunen, T.A. and Schulze, H. (1996a) Evaluation of enzyme treatment of barley/rapeseedmeal diets on digestibility and availability of nutrients for pigs from 40 to 90 kg liveweight. Proceedings of the 14th International Pig Veterinary Congress. Bologna, Italy, 7–10July, p. 466.

van Lunen, T.A. and Schulze, H. (1996b) Influence of Trichoderma longibrachiatum xylanasesupplementation of wheat and corn based diets on growth performance of pigs. CanadianJournal of Animal Science 76, 271–273.

Yin, Y.-L., McEvoy, J., Schulze, H. and McCracken, K.J. (1997) Effects of levels of dietarynon-starch polysaccharides (NSP) and feed enzymes on digestibility and ilealconcentration of volatile fatty acids (VFA) in growing pigs. In: Laplace, J.-P., Février, C.and Barbeau, A. (eds) Digestive Physiology in Pigs. Proceedings of the VIIth InternationalSymposium on Digestive Physiology in Pigs, St Malo, France. EAAP Publication no.88,pp. 502–505.

Yin, Y.-L., McEvoy, J.D.G., Schulze, H. and McCracken, K.J. (2000) Studies on cannulationmethod and alternative indigestible markers and the effects of food enzyme supple-mentation in barley-based diets on ileal and overall apparent digestibility in growing pigs.Animal Science 70, 63–72.

Zijlstra, R.T., Scott, T.A., Edney, M.J., Swift, M.L. and Patience, J.F. (1998) Measurementsto predict swine DE of barley. In: 1998 Feed Grain Quality Conference. 8th–10thNovember, Edmonton, Alberta, Canada, pp. 78–81.

198 G.G. Partridge




S. DänickeUse of Feed Enzymes in Broilers9

Introduction

The use of dietary fat of plant or animal origin as a feed ingredient is a commonpractice in broiler feeding, especially in intensive production systems where dietaryenergy concentrations of 13 MJ ME kg−1 and greater are required. Under suchconditions, its proportion might reach 5–10% of the diet, which corresponds toapproximately 40–85% of total dietary fat. The actual proportion of a supplementedfat in a broiler diet depends on several factors, such as local availability, relative priceand the effect on the feed manufacturing process, in addition to physiologicallyderived constraints for a particular fat. The last point in particular requiresconsideration as several factors modify or interfere with an effective use of a particulartype of dietary fat. For example, fat digestibility in a broiler diet depends not only onfat type but also on the particular cereal with which it is used. Antoniou et al. (1980)and Ward and Marquardt (1983) showed that the combination of beef tallow withrye depressed fat digestibility much more than when wheat was the dominant cereal.In contrast, when both cereals were tested with a vegetable oil only slight differencesin fat digestibility were found.

The interaction between cereal identity and dietary fat type became even moreinteresting with the widespread use of exogenous feed enzymes, especially xylanasesand β-glucanases. These endo-enzymes are capable of partial hydrolysis of anti-nutritive pentosans (which are more concentrated in rye than in wheat) and mixedlinked β-glucans (most concentrated in barley), respectively, thereby reducing theirviscous properties in vivo. This in turn allows the use of higher dietary proportions ofsuch ‘critical’ cereals. The fact that greater enzyme effects were reported for fat digest-ibility in rye- than in wheat-containing diets (Friesen et al., 1992) further confirms thefat by cereal identity interaction, and moreover suggests that the viscous properties of



Interaction between CerealIdentity and Fat Quality

and Content in Response toFeed Enzymes in Broilers

S. DÄNICKEInstitute of Animal Nutrition, Federal Agricultural Research Centre,

Braunschweig (FAL), Bundesallee 50, D-38116 Braunschweig,Germany

9



these cereals is the reason for their interaction with fat digestibility. In addition, theobserved enzyme effect was greater with increasing proportion of dietary rye.

In the following sections an attempt will be made to contribute to the under-standing of the underlying mechanisms and to draw some conclusions for practicalpurposes.

Principles of Fat Digestion in Broilers

Digestion and absorption of fat in poultry was extensively reviewed by Krogdahl(1985). Briefly, the initiation of fat digestion seems to start with a possible onset of fatemulsification in the upper digestive tract but mainly with the entrance of digesta intothe duodenum, where it is mixed with the secretions of the exocrine pancreas and gallbladder, which contain lipase, co-lipase and conjugated bile salts, respectively. Bile saltsand phospholipids facilitate progressive fat emulsification. Binding of co-lipase at thesurface of these relatively large aggregates is the precondition for the action of the lipasethat hydrolyses the triglycerides at the 1 and 3 positions of the glycerol molecule.Released fatty acids, the remaining 2-monoglycerides, phospholipids and bile saltsform spontaneously into mixed micelles, which are much smaller in diameter than theemulsified fat droplets. Micelles are able to solubilize non-polar lipid compounds suchas fat-soluble vitamins or long-chain saturated fatty acids (LCSFA) within their hydro-phobic core. The role of the micelle is to act as a shuttle between the bulk intestinalcontents and the aqueous–microvillus interface by overcoming unstirred water-layerresistance (Westergaard and Dietschy, 1976). At this point, micelles release theircontents to the luminal membranes, from where they passively diffuse into the cell.Under the conditions of a high fat digestibility these processes take place quantitativelyin the proximal regions of the small intestine. Bile salts released from the micellesdiffuse back to the lumen, and from this point might be used again for micelleformation in a lower part of the small intestine. Ultimately the majority are reabsorbedby the distal small intestine. They are transported back to the liver, where they are againmade available for bile production, thus conserving the metabolic bile salt pool.

A long-chain fatty acid that diffuses through the luminal membrane will onlyenter the intracellular pool if it is bound to the intracellular fatty acid binding protein(FABP) (Ockner et al., 1972a; Ockner and Manning, 1974; Katongole and March,1979, 1980; for review, see Hohoff and Spener, 1998). The FABP–fatty acid complextranslocates the fatty acid to other intracellular compartments where it is normally re-esterified to a triacylglycerol and further processed. It has been shown for broilers thatthe greatest concentration of this inducible intestinal FABP is found at the main siteof fatty acid absorption, i.e. the proximal small intestine (Katongole and March, 1980).

Factors Affecting Fat Digestibility

From the above section it becomes clear that a variety of events have to take place in acoordinated manner within a relatively short time (see below) if high quantities of

200 S. Dänicke




ingested fat are to be absorbed. Hence, it is useful to take note of several factors thatinfluence fat digestibility in broilers before dealing with the effects of intestinalviscosity caused by ingestion of cereal grain soluble non-starch polysaccharides(NSPs) – and consequently the effects of the intestinal viscosity reducingcarbohydrases – in modification of fat digestibility. The first and probably the mostimportant factor is the chemical nature of the dietary fat itself. For example, beeftallow is characterized by a lower fat digestibility and a lower ME content than soyaoil. The reasons are the differences in fatty acid composition of both dietary fatsources. From their fatty acid profile (Table 9.1) it can be deduced that beef tallowis mainly composed of saturated fatty acids (palmitic and stearic acid) and ofmono-unsaturated oleic acid, whereas linoleic acid, linolenic acid and oleic acidare the principal fatty acids in soya oil. Differences in fat digestibility arise fromdifferences in absorbability of saturated and unsaturated fatty acids and can besummarized from the literature as follows.

Degree of unsaturation and polarity of fatty acids

LCSFAs (C16 : 0, C18 : 0) are more non-polar than unsaturated long-chain fattyacids of the C18 family and among these fatty acids the polarity increases as thenumber of double bonds increases. Among the saturated fatty acids (C8 : 0 . . .C18 : 0), the polarity increases as the chain length decreases. The more non-polar afatty acid is, the more it relies on an adequate presence of bile salts and phospholipidsfor adequate emulsification. It has been shown that medium-chain saturated fattyacids (MCSFA) and long-chain unsaturated fatty acids (LCUSFA) can be absorbedin large quantities even in the absence of bile salts (Garret and Young, 1975;Westergaard and Dietschy, 1976). The differences in polarity of fatty acids arereflected by differences in their absorbability. In general figures, absorbabilityincreases in the following order: C18 : 0 < C16 : 0 < C14 : 0 < C18 : 1, C18 : 2,C18 : 3 (Renner and Hill, 1961; Young, 1961; Garret and Young, 1975; Ketels et al.,1987; Blanch et al., 1995; Vila and Esteve-Garcia, 1996a,b; Dänicke et al., 1997b).This order is generally at a higher level for fatty acids derived from soybean oil thanfor beef tallow, and differences in absorbability between particular fatty acids aremore pronounced for the latter. Palmitic and stearic acid from soya oil are found tobe better absorbable than those originating from beef tallow. It is thought that this isdue to the synergistic effects of LCUSFAs, which support the micellar solubilizationof LCSFAs. In this respect, monoglycerides or glycerol seem to be necessary for opti-mized micellar dissolution (Garret and Young, 1975; Sklan, 1979). The synergisticeffects of the ratio of unsaturated to saturated fatty acids in relation to fatty acidabsorption and ME values of the fats have been described by several regressionapproaches (e.g. Wiseman and Lessire, 1987; Ketels and De Groote, 1987, 1989;Wiseman and Salvador, 1991). It should be noted, however, that the order ofdigestibility of fats in addition to the absolute digestibility value might be modifiedby other factors, such as dietary fat inclusion level (Ketels et al., 1987; Wisemanand Lessire, 1987), age of the birds (Wiseman and Lessire, 1987) and free fatty

Use of Feed Enzymes in Broilers 201




202 S. Dänicke


Fatty

acid

(C-a

tom

s:d

oubl

ebo

nds)

Lard

Soyb

ean

oil

Beef

tallo

wLi

nsee

dPa

lmM

aize

oil

Coc

onut

oil

Min

Max

Min

Max

Min

Max

Min

Max

Min

Max

Min

Max

Min

Max

8:0

10:0

12:0

14:0

16:0

16:1

18:0

18:1

18:2

18:3

0.1

1.3

21.2 2.1

11.1

40.3 8.0

0.1

0.1

1.7

26.6 5.3

17.7

51.8

10.4 2.1

0.1

0.7

0.0

1.8

19.1

44.0 3.3

0.1

13.5 0.1

7.4

27.3

59.4 8.2

0.0

2.0

23.7 0.1

13.5

24.5 0.9

0.1

0.2

4.9

35.4 6.1

36.5

46.4 5.4

1.1

5.0

0.1

3.1

18.2

13.9

49.4

10.0 0.1

7.8

21.4

17.9

66.2

0.8

40.7 0.1

3.9

36.7 7.9

0.5

1.0

48.8 0.3

5.2

41.6

12.7 0.5

9.4

0.1

1.7

25.8

42.0 1.0

13.6 0.1

6.7

30.1

58.6 1.0

2.6

3.0

28.4

12.1 8.5

2.0

6.0

1.5

7.5

7.0

49.5

19.5

12.8 8.1

13.7

23.1

No.

ofre

fere

nces

7.1

9.1

11.1

4.1

3.1

3.1

5.1

Ref

eren

ces

used

:You

ng(1

961)

,Vee

net

al.(

1974

),Si

ckin

ger(

1975

),Si

bbal

dan

dKr

amer

(197

8),W

isem

anet

al.(

1986

),Ke

tels

etal

.(19

87),

Huy

gheb

aert

etal

.(1

988)

,Frit

sche

etal

.(19

91),

Satc

hith

anan

dam

etal

.(19

93),

Blan

chet

al.(

1995

,199

6),H

alle

(199

6),V

ilaan

dEs

teve

-Gar

cia

(199

6a,b

),D

änic

keet

al.(

1999

a).

Tabl

e9.

1.Fa

ttyac

idco

mpo

sitio

nof

som

edi

etar

yfa

ttyp

es(%

).



acid content (Young, 1961; Wiseman et al., 1991; Blanch et al., 1995; Vila andEsteve-Garcia, 1996c).

Bile salt availability

Bile salt availability in young broilers plays a central role in determining the extentto which fats differ in their digestibility. It has been suggested frequently that bileavailability in the very young chick is insufficient, particularly if high proportions oftallow are provided as dietary fat. It was demonstrated that addition of bile or severalbile salts to broiler diets containing beef tallow resulted in an improvement in fatdigestibility (Fedde et al., 1960; Gomez and Polin, 1976; Katongole and March,1980; Polin et al., 1980; Polin and Hussein, 1982). It might be concluded thatthe addition of bile salts would be beneficial in the presence of large amounts ofundigested or non-absorbed fatty acids within the small intestine. The reason forthis bile salt deficiency in tallow-fed birds might be due to an increase in bile saltexcretion whereby bile salts escape the enterohepatic bile salt cycle. This in turnchanges hepatic cholesterol metabolism and the profile of synthesized bile salts andultimately decreases the intestinal bile acid pool, as shown by Monsma et al. (1996)in an experiment with rats fed different dietary fats. These authors demonstrated astrong linear relationship between dietary intake of stearic acid and muricholicacid-derived bile salt excretion. Furthermore, intestinal microflora might alsocontribute to bile salt metabolism, as shown by Kussaibati et al. (1982) usingconventional and germ-free chickens. They found that the addition of bile salt to thediet markedly improved the digestibility of LCSFAs in conventional birds, whereasonly slight effects were observed in germ-free chickens. The benefit of adding bilesalts to either bird type was marginal for digestibility of LCUSFAs. The authorsconcluded that intestinal microbes in conventional birds deconjugate bile salts,which are poorly absorbed and consequently escape the enterohepatic bile salt cycle.

Kritchevsky and Story (1974) demonstrated that non-nutritive fibres from dif-ferent feedstuffs are able to bind bile salts to a different degree in vitro. Consequently,such binding could contribute to changes in the intestinal bile salt pool.

Triglyceride structure – fatty acid positional effects

Another important point that adversely influences the digestibility of dietary fats isthe positional distribution of the fatty acids at the glycerol molecule. Generally, highproportions of the LCSFAs at the 2-position are associated with an increased micellarsolubility of these fatty acids in the form of the respective monoglycerides comparedwith their relative solubility when released as free fatty acids from the 1- and3-positions by the action of the lipase. It has been shown that a relatively highproportion of the LCUSFAs in beef tallow (mainly oleic acid) are bound to the2-position, and the majority of palmitic and stearic acid are esterified at the 1- and3-positions (Sibbald and Kramer, 1977; Ketels et al., 1987; Bracco, 1994). Such





positional effects are presumed to be detrimental to fat digestion and weredemonstrated to influence TME values of fats in broilers (Sibbald and Kramer,1977), energy absorption in rats (Brink et al., 1995), and fat absorption in rats(Mattson et al., 1979) and human infants (Filer et al., 1969).

Lipase availability

Lipase activity was reported to be low in the very young turkey (Krogdahl andSell, 1989) and chick (Nir et al., 1993) and lipase addition to a tallow-containing dietwas shown to improve fat digestibility (Polin et al., 1980). Noy and Sklan (1995) didnot detect any increase in ileal fat digestibility in chicks between 4 and 21 days ofage (6% unsaturated fat in the diet) despite an age-related increase in lipase secretioninto the duodenum. At present, no final conclusion can be drawn under whichcircumstances lipase secretion might be a limiting factor in fat digestion.

Mineral soaps

Finally, it is evident from the literature that calcium or magnesium might reduce fatabsorption by forming indigestible soaps. This is of special importance if either fatabsorption is low, i.e. high intestinal amounts of fat as in tallow digestion, or higherlevels of those minerals are present in the diet (Fedde et al., 1960; Mattson et al.,1979; Antoniou et al., 1980; Brink et al., 1995).

Although the above sections focus mainly on beef tallow and soybean oil, it should benoted that the effects of the other dietary fats listed in Table 9.1 might be deducedfrom their fatty acid profile and other physical and chemical characteristics in asimilar manner.

Role of Intestinal Viscosity in Interference of Fat Digestion

General remarks

Increases in intestinal viscosity of broilers are mainly caused by feeding higherproportions of soluble NSPs such as the soluble pentosans in rye and wheat andsoluble mixed linked β-glucans in barley. Although both rye and barley arecomparable in soluble NSP content, the relationship between water extract viscosityprofile and soluble NSP concentration is quite different (Fig. 9.1). It has beenemphasized that, in addition to chemical analysis of polysaccharides, their physicalproperties have to be considered in evaluation of their physiological effects (Morris,1992). Molecular weight and degree of branching of the polysaccharide chain mainlydetermine the so-called intrinsic viscosity, i.e. the fractional increase in viscosity perunit concentration of polymer (Morris, 1992). Bedford and Classen (1992) reported

204 S. Dänicke




a positive correlation between high molecular weight carbohydrate complexes(molecular weight > 500,000) and viscosity in the intestine of chickens fed dietsvarying in pentosan content. Hence, the distinct structural properties might accountfor differences between rye and barley with respect to in vitro and in vivo viscosity.Annison et al. (1995) compared soluble isolates from wheat and rice bran withrespect to chemical and physical properties and found that, although both isolateswere composed mainly of arabinoxylans, they produced extract viscosities of 64 and1.6 mPa s, respectively. This was associated with arabinose : xylose ratios of 0.58and 1.23, respectively. These authors suggested that rice bran xylan backbone carriesmore arabinose side-chains, whereas the unsubstituted sections of the wheat xylanmain chain facilitated the formation of inter-chain hydrogen bonds. Consequently,the degree of branching within wheat varieties could be a contributing factor todifferences in extract viscosity.

It has been suggested that a viscosity of a polysaccharide solution ofapproximately 10 mPa s indicates a critical polysaccharide concentration abovewhich viscosity responds more steeply to increasing polysaccharide concentrations.This is thought to be the result of the onset of formation of an entangled networkfrom individual polysaccharide coils (Morris et al., 1981). In spite of the use ofcomplete cereal matrices for in vitro viscosity measurements as shown in Fig. 9.1, itcan be clearly seen that rye solution viscosity started to rise more steeply in the rangebetween 1 : 6 and 1 : 5 dilution, which corresponded to viscosities of approximately5–10 mPa s. From the same graph it might be concluded that such an entanglementwas not initiated in the case of barley. The corresponding ileal viscosities in broilersfed diets containing rye and barley (69% dietary inclusion) were measured as 480.4and 10.2 mPa s, respectively (Dänicke et al., 1999a).



Fig. 9.1. Water extract viscosity of rye (v) and barley (r) as influenced byincubation volume (Dänicke et al., 1999a).



Feed passage through the digestive tract, gut motility and unstirredwater-layer resistance

Carbohydrase supplementation of diets based on barley, rye or wheat has been shownto accelerate the total gastrointestinal tract transit time in chickens (Jeroch et al.,1990; Salih et al., 1991; Almirall and Esteve-Garcia, 1994; Dänicke et al., 1997a)and in turkeys (Großer and Jeroch, 1997), which in turn suggests that an increasedintestinal viscosity would result in the opposite effect and would consequently reducefeed intake and performance. Table 9.2 summarizes some results on measurement offeed transit time, which is expressed as the time required for 50% of marker excretion(T50). From the results of Van der Klis and Van Voorst (1993) it would appear thatdietary soluble NSP concentration is closely related to feed transit time. The effects of

206 S. Dänicke


Reference1. Bird type2. Age Remarks Carbohydrase

T50

(h)

Salih et al. (1991) 1. Broiler2. 2 weeks

2..4 weeks

2..6 weeks

2..8 weeks

High-viscosity barley based diets

WithoutWithWithoutWithWithoutWithWithoutWith

7.596.316.86.456.596.76.847.21

Van der Klis andVan Voorst (1993)

1. Broiler2. 5 weeks

Without carboxy methyl cellulose10 g carboxy methyl cellulose30 g carboxy methyl cellulose

4.585.357.2

Almirall andEsteve-Garcia(1994)

1. Broiler2. 3 weeks1. Cocks2. 1 year

High-viscosity barley

High-viscosity barley

WithoutWithWithoutWith

8.895.493.394.81

Dänicke et al.(1997a)

1. Broiler2. 4 weeks

Rye-based diet, 10% soybean oilRye-based diet, 10% soybean oilRye-based diet, 10% beef tallowRye-based diet, 10% beef tallow

WithoutWithWithoutWith

8.386.687.966.88

Großer and Jeroch(1997)

1. Turkey2. 3 weeks

Wheat-based diet, 8% soybean oilWheat-based diet, 8% soybean oilWheat-based diet, 8% beef tallowWheat-based diet, 8% beef tallowWheat-based diet, 8% coconut oilWheat-based diet, 8% coconut oil

WithoutWithWithoutWithWithoutWith

7.98.36

14.529.45.586.07

Table 9.2. Mean total gastrointestinal tract transit time of feed, expressed as time required for50% of excretion of a marker.



different fat types and xylanase supplementation on transit times were investigated byDänicke et al. (1997a,b) in broiler chickens and by Großer and Jeroch (1997) inyoung turkeys. Xylanase addition resulted in an accelerated transit time for broilersfed on soya oil and beef tallow. A fat effect was not obvious in broilers. In contrast,different fat and enzyme effects were observed in turkeys. Xylanase addition to dietscontaining soya oil or coconut oil changed the transit time only slightly, whereas theeffect was dramatic (reduced transit by 5 h) in the diet containing beef tallow. Pas-sage rate was fastest with the coconut oil diet followed by the diets containing soya oiland finally beef tallow. A comparison between species makes it clear that the youngturkey seems to be more responsive to different fat types than the growing chick.

In a more detailed study, Dänicke et al. (1999b) examined the mean retentiontime (MRT) of digesta in different segments of the digestive tract of broilers fedrye-based diets with differing levels of added dietary fat type and xylanase (Fig. 9.2).Again, no significant fat effect was observed. The majority of the reduction in MRTon addition of xylanase was observed in the jejunum and ileum. Experiments withhumans and pigs revealed that an increased digesta viscosity delayed gastric emptying(Furuya et al., 1978; Blackburn et al., 1984), thereby reducing the overall feed transittime through the entire digestive tract. Sudendey and Kamphues (1995) observedan accelerated rate of gastric emptying when diets based on wheat or barley fed topiglets were supplemented with a carbohydrase. Furthermore, Meyer et al. (1986)concluded from their experiment with dogs that meal viscosity affects gastric sieving,which in turn could influence both gastric and intestinal digestion of nutrients. Sucheffects of viscosity on gastric emptying rate have not been observed with broilers and



Fig. 9.2. Effect of fat type and xylanase supplementation on mean retention timein successive segments of the gastrointestinal tract of male chickens (day 24 ofage, average of four birds per treatment ± standard error) (Dänicke et al., 1999b).Crop, ]; proventriculus plus gizzard, &; duodenum, $; jejunum, *; ileum, \;rectum, *.



the effect of different fat types on the contribution of gastric emptying to overall feedtransit changes (Fig. 9.2) has proved inconclusive. In this regard the chicken appearsto differ substantially from other monogastrics.

However, a general effect of fat as such on the MRT of digesta in the smallintestine cannot be excluded completely since its presence modifies gut motility viastimulation of gastrointestinal neurotransmitters or hormones. Neurotensin has beenshown to decrease the frequency and strength of contractions of the gizzard andpartially of the duodenum and ileum in broilers (Degolier et al., 1997). The presenceof lipids in the duodenum was found to inhibit gastric motility in turkeys (Dukeand Evanson, 1972). Degolier et al. (1997) reported that neurotensin was releasedinto the hepatic-portal circulation in response to the presence of oleic acid in theduodenum. It was suggested that, by a parallel decrease in gut motility, the efficiencyof digestion could be enhanced by a longer MRT of digesta in the intestine.Moreover, fats differing in their contents of unsaturated and saturated fatty acidsare not equivalent in their stimulation of release of neurotensin and other gastro-intestinal peptides. According to the results of Sagher et al. (1991) obtained with rats,maize oil and olive oil increased neurotensin and substance P, whereas butterfat didnot, when compared with the low fat control group.

It might be concluded from these findings that both intestinal viscosity andintestinal fatty acid composition contribute to differences in the MRT of digesta inintestinal segments and consequently the overall gastrointestinal tract transit time,albeit by different mechanisms. Hence, an interaction between both factors cannotbe excluded in interpreting the results shown in Fig. 9.2.

A reduction in gut motility, caused by increased viscosity, decreases the mixingof digesta with pancreatic and biliary secretions, which is of particular importancefor emulsification of saturated fatty acids and formation of micelles. It has beensuggested by Smulikowska (1998) that gut motility-mediated shuttling of digestabetween the duodenum and gizzard increases the time of feed exposed to digestiveenzymes and favours enhanced digestion and absorption in the upper parts of thesmall intestine. At the same time as intestinal viscosity is increased and mixing isdecreased, the thickness of the unstirred water layer (UWL), an extracellular barriercovering the intestinal microvilli, is proportionally increased (Johnson and Gee,1981, 1982; Flourie et al., 1984; Lund et al., 1989). Westergaard and Dietschy(1976) proposed that the principal role of the micelle is to overcome the resistance ofthat UWL. Consequently, an increased thickness of the UWL would increase thisresistance and decrease the transfer rate between intestinal bulk and the absorptivesite, which is, again, of particular importance for saturated fatty acids. In addition,Smulikowska (1998) suggested that, especially in the young chicken, an increase inthe UWL would decrease the effective intraluminal diameter and as a result the rateof digesta transit and mixing would be further impeded. However, this concept of theUWL is open to debate. Edwards et al. (1988) and Read and Eastwood (1992)pointed out that a reduction in mixing and an increase in the UWL are two ways ofdescribing the same thing. Nevertheless, strong relationships between the thicknessof the UWL and the malabsorption of sugars and of cholesterol have been reported(e.g. Lund et al., 1989). Cholesterol absorption by everted intestinal sacs was

208 S. Dänicke




significantly reduced even at low concentrations of oat gum and continued to fall asthe concentration increased, but proportionately less with each increment (Lundet al., 1989). The authors concluded that this initial decrease could indicate a sievingeffect of dispersed β-glucan polymer on the micelles, which have relatively highmolecular volume (Phillips, 1986). Interestingly, this decrease coincided with arelatively low in vitro viscosity of approximately 5–20 mPa s, which is the rangewhere polysaccharides in solution start to form an entangled network from individualpolysaccharide coils (Morris et al., 1981; see above section), associated with a steepincremental increase in viscosity. Moreover, higher intestinal lipid concentrations areparalleled by a corresponding increase in the size of the formed micelles (Ockneret al., 1972b). This would explain why tallow feeding is more dramatically sensitiveto small increases in intestinal viscosity when compared with soybean oil feeding.

Availability of endogenous lipase and bile salts

Experiments dealing with the effects of viscous carbohydrates on lipase activity in theintestinal contents and pancreas in relation to fat absorption are often inconclusiveand the underlying mechanisms are not fully understood. Isaksson et al. (1983a)found an increased lipase output in rats fed a pectin-containing diet, whereas lipaseconcentration per gram wet material was not changed. Total fat excretion in thefaeces was increased at the same time. These results show that pectin obviouslystimulated pancreatic lipase output but do not suggest that specific lipase activitywas responsible for the increased fat excretion. In contrast, feeding of neither lowmethylated or high methylated pectin nor of wheat bran changed specific lipaseactivity (units per mg) or total lipase output in rats (Isaksson et al., 1983b), whereasprotein concentration and protein content of the pancreas were altered. These resultswere interpreted as adaptive changes in pancreatic enzyme production. In anotherexperiment, Isaksson et al. (1984) found a decrease in specific lipase activity (unitsper ml) after consumption of a pectin-containing test meal in pancreatectomizedpatients treated with a granulated pancreatic enzyme preparation. This decrease wasaccompanied by a decrease in 14CO2 excretion after 14C triolein ingestion, whichwould indicate that lipase activity indeed became a limiting factor. However, theeffect of pancreatic regulation due to pectin ingestion could not be investigated withsuch an experimental design.

The decrease of in vitro lipase activity and activity of other pancreatic enzymes inhuman duodenal juice was shown to be directly related to an increased viscosity(Isaksson et al., 1982). It was also suggested that adsorption of enzymes by specificprotein enzyme inhibitors could play a role in activity depression in the case of wheatbran. Moreover, it was shown that inhibition in lipase activity due to increasedviscosity occurred in a relatively short period of time and it remained at thesedepressed levels thereafter. Therefore, it was suggested that these inhibitory effectswould be relatively independent of intestinal passage time. Finally, lipase activitydecreased as the pH was lowered from 5.7 to 5. Dänicke et al. (1997a) reported asignificant decrease in pH in the small intestine of broilers fed tallow-containing





diets. Whether such a decrease could be a contributing factor in lower tallowdigestibility remains to be clarified.

Almirall et al. (1995) reported the specific lipase activity to be decreased inbroiler chickens and cocks after feeding barley, compared with maize. Activity waspartially restored after β-glucanase supplementation of the barley-based diets. Inaddition, chickens had higher specific lipase activities and seemed more susceptibleto dietary treatments than cocks.

Dänicke et al. (1999c) demonstrated that the provision of a xylanase decreasedspecific lipase activity (units per gram wet weight) in the jejunal contents and in thepancreas of broilers fed rye-based diets supplemented with beef tallow, whereas noeffect was noted in diets supplemented with soybean oil. Total pancreatic lipaseactivity remained unchanged but was significantly increased on body weight basis inbirds fed the unsupplemented tallow-containing diet. In addition, relative pancreaticlipase activity was linearly dependent on relative pancreas weight. It would appear,therefore, that birds fed the unsupplemented tallow-containing diet had attemptedto increase lipase activity to compensate for such poor fat digestibility and that thiswas reflected in low body weights and the highest specific intraluminal lipase activity.A decreased pancreatic lipase activity and an increased intestinal activity in rats fedguar gum was reported by Poksay and Schneeman (1983), which suggests a similarresponse. It was concluded that the observed reduction in fat digestibility followingguar gum addition to the diet could not be the result of an insufficient presenceof lipase in the intestine. Furthermore, Silva et al. (1997) found intestinal lipaseactivity (corrected for body weight and weight of small intestine and contents) to besignificantly decreased in broilers fed xylanase-added rye-based diets. Interestingly,when intestinal contents where examined in vitro, lipase activity was slightlyincreased after the addition of the xylanase when compared with measured activitiesbefore xylanase administration. It was suggested that the xylanase released somelipase activity, which had been adsorbed by dietary fibre in the intestines of birds fedthe unsupplemented diets. Adsorption of lipase and bile salts to different fibre typeswas examined by Lairon et al. (1985). It was found that binding capacity of cellulose,purified xylan and pectin was negligible, whereas wheat bran was shown to have amoderate capacity. Furthermore, fine wheat bran bound more lipase and bile saltsthan coarse wheat bran. Pasquier et al. (1996) demonstrated, by in vitro experimentsusing reconstituted duodenal medium, that the amount of emulsified lipids wasreduced and the size of emulsified droplets was increased as the concentration ofviscous fibre was increased. In the range of 0–20 mPa s the correlations to bothparameters were r = −0.79 and r = 0.88, respectively. In addition, triglyceridehydrolysis was reduced by approximately 30% over a similar range of viscosity.

Intestinal viscosity also influences bile salt metabolism in rats. Inclusion ofpsyllium, a dietary fibre rich in soluble components, was shown to increase theexcretion of bile salts and total steroids in rats significantly when compared withinclusion of cellulose at a similar level (Buhman et al., 1998). The authors suggestedintestinal viscosity itself to be responsible for a decreased reabsorption of bile-saltsrather than bile-salt binding per se, since psyllium does not bind bile salts in vitro.Furthermore, the hepatic cholesterol metabolism was up-regulated, as suggested by

210 S. Dänicke




elevated activities of cholesterol 7α-hydroxylase and cholesterol 7α-hydroxylasemRNA as a result of psyllium feeding. Ikegami et al. (1990) reported a linearrelationship between intestinal viscosity and the secretion of pancreatic enzymes andtotal bile salts in rats which was accompanied by an increased pancreatic weight. Itwas suggested that the animal compensates for the inefficiency of nutrient digestionwith an enlargement of digestive organs and increased secretion of digestive juices.

Gut morphology and microbial implications

Feeding soluble NSP was shown to increase the length and the absolute and relativeweight of the small intestine in rats (Johnson et al., 1984; Johnson and Gee, 1986)and broilers (Simon, 1998) (Table 9.3). It was suggested by Johnson et al. (1984)that the presence of large amounts of unabsorbed material in the intestine exertstrophic effects on the intestinal mucosa. Feeding of guar gum and carboxy methylcellulose was shown to increase the cell division rate in the small intestine, colon andcaecum of rats, which was accompanied by reduced activity of some mucosalenzymes (Johnson et al., 1984; Johnson and Gee, 1986). It was hypothesized thatincreased cellular proliferation gives rise to a reduction in lifespan of mucosal cellsas a result of an increased rate of shedding of immature (with respect to enzymeexpression) cells at the apex of the villi. Guar gum and carboxy methyl cellulosediffered, however, in that the latter induced a greater rate of cell division and in themore distal regions of the small intestine. The authors concluded that factors othersthan intestinal viscosity per se could be involved, such as microbial activity in thesmall intestine. Rakowska et al. (1993) found severe damage of intestinal villi andmucous membranes of the duodenum and small intestine in broilers after feeding arye-based diet. Microbial involvement was concluded from this experiment since theantibiotic Nisine protected the intestinal villi. An impact of intestinal viscosity ongut morphology was also shown by Drakley et al. (1997) in that feeding of a highlyviscous wheat resulted in a significantly increased villus height and width whencompared with a low viscosity wheat. Crypt cell proliferation rate is also increasedwith increasing intestinal viscosity in broilers, as demonstrated by a decrease afterfeeding of xylanase-supplemented rye-based diets (Silva and Smithard, 1997).

Dietary fat type might also contribute to changes in intestinal morphology,as shown by Sagher et al. (1991), who found that feeding of maize oil and olive oilsignificantly increased villus height compared with a low-fat control diet, whereasfeeding of butterfat significantly decreased villus height. The question of whetherchanges in intestinal viscosity interact with such a response remains unanswered.

Dänicke et al. (1999b) examined the effect of xylanase supplementation of rye-based diets containing either soybean oil or beef tallow offered to broilers on tissue-associated bacterial groups in several segments of the digestive tract (Table 9.4).Increased intestinal viscosity is largely influenced by feeding of such diets as shown inrelated experiments (Dänicke et al., 1997a,c) (Table 9.5; Fig. 9.3) and might havecaused the described intestinal morphological changes, which lead to reduced coloni-zation surface, altered adhesion structures and an increased release of epithelial cells





212 S. Dänicke


Fatt

ype

Duo

denu

mJe

junu

mIle

umPa

ncre

asEn

doge

nous

dige

stiv

eN

-loss

esm

gda

y−1pe

rLW

(kg)

0.67

g10

0g−1

LWk S

(%da

y−1)*

g10

0g−1

LWk S

(%da

y−1)*

g10

0g−1

LWk S

(%da

y−1)*

g10

0g−1

LWk S

(%da

y−1)*

S− S+ T − T+

1.5a

1.4a

1.9b

1.4a

56a

64a

84b

61a

2.1a

2.0a

2.6b

1.9a

51a

52a

75b

58a

1.3a

1.2a

1.6b

1.2a

53ab

56ab

68bb

50ab

0.4ab

0.4ab

0.6bb

0.5ab

52ab

45ab

118bb

39ab

87 69 244 81

*Fra

ctio

nalr

ates

ofpr

otei

nsy

nthe

sis,

i.e.d

aily

prot

ein

synt

hesi

sas

perc

enta

geof

tissu

epr

otei

n.ab

Valu

esw

ithdi

ffere

ntsu

pers

crip

tsw

ithin

the

colu

mns

are

sign

ifica

ntly

diffe

rent

(P<

0.05

).

Tabl

e9.

3.Ef

fect

sof

diet

ary

fatt

ype

(S,1

0%so

ybea

noi

l;T,

10%

beef

tallo

w)a

nden

zym

esu

pple

men

tatio

n(−

,with

out;

+,w

ith1

gAv

izym

e13

00kg

−1di

et)i

na

rye-

base

ddi

et(5

6%di

etar

yin

clus

ion)

onpr

otei

nm

etab

olis

min

mal

ebr

oile

rs(D

änic

keet

al.,

2000

a,b)

.



(and adhering bacteria) into the lumen as a result of stimulation of the proliferationrate of mucosal cells. The observed effects on microbial colonization were mostpronounced in duodenal and jejunal samples, which exhibited short digesta retentiontimes and thus low steady-state concentrations of soluble arabinoxylans.

The response of cocci bacteria to enzyme supplementation seemed less drastic,which indicates that this group of bacteria is less affected by morphological changesor nutrient composition. On the other hand, supplementation with tallow resultedin an increased number of cocci in the jejunum and ileum, but seemed to depressgrowth of total anaerobic bacteria and enterobacteria. Significant higher pH valueswere measured in most intestinal segments when tallow was used as dietary fat



Fat XylanaseIntestinalsegment

Entero-bacteria

Gram-positivecocci

Entero-cocci

Totalanaerobes

Soybean oilSoybean oilTallowTallow

Without3000 IU kg−1 dietWithout3000 IU kg−1 diet

CropCropCropCrop

6.58ba

5.38aa

5.53aa

5.94ab

5.04aa

5.59ab

6.22ba

5.14aa

4.18ab

4.46ab

4.00ab

4.28ab

5.89ab

6.02bb

6.76bb

4.82ab



DuodenumDuodenumDuodenumDuodenum

6.26ba

3.54aa

4.96ca

5.19ca

3.96ab

4.26ab

5.01ab

4.72ab

3.16aa

3.88ba

4.09ba

3.08aa

5.74ab

4.29ab

5.48ab

4.66ab



JejunumJejunumJejunumJejunum

6.35ba

4.42aa

5.73ba

5.71ba

4.62ba

4.06aa

5.26ca

4.82ba

3.43ab

2.95ab

4.01ab

3.88ab

6.20ab

4.72ab

5.35ab

5.30ab



IleumIleumIleumIleum

6.13ba

6.13ba

6.49ba

5.03aa

4.25aa

4.85ab

6.35ca

5.06ba

3.80aa

4.29ab

5.90ba

3.58aa

6.50ab

5.97ab

6.49ab

5.52ab

Probability:FatEnzymeIntestinal segmentFat × enzymeFat × intestinal segmentEnzyme × intestinal segmentFat × enzyme × intestinal segment

0.7967< 0.001< 0.001< 0.001

0.12430.0869

< 0.001

< 0.0010.0147

< 0.001< 0.001

0.08490.38080.0067

0.05550.1083

< 0.0010.00800.06030.04100.0012

0.3749< 0.001

0.01690.67420.97650.93280.0805

Pooled standard error of means 0.29 0.25 0.38 0.52

a–cValues with no common superscript within columns differ significantly (P < 0.05).

Table 9.4. Tissue-associated bacterial counts in different segments of digestive tract of chickens(day 16 of age) as influenced by dietary fat type and xylanase supplementation (CFU, log10)(Dänicke et al., 1999b).



214 S. Dänicke


1.Bi

rdty

pe2.

Age

(bal

ance

exp.

)3.

Age

(gro

wth

exp.

)4.

ANO

VA( P

)

Die

tary

cere

alin

clus

ion

(%)

Die

tary

fati

nclu

sion

(%)

Ref

no.

RT

WM

BS

TL

ABC

EBW

GFC

RVi

s1FD

2PD

2

[1]

1.Br

oile

r2.

Day

s7–

11

1.D

ays

21–2

5

1.D

ays

32–3

6

3.D

ays

1–28

4.Fa

t1.

Enzy

me

1.Ag

e1.

Fat ×

enzy

me

1.Fa

t ×ag

e

61 61 61 61 61 61 61 61 61 61 61 61 61 61 61 61

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10.0

10 10 10 10 10 10 10 10

− + − + − + − + − + − + − + − +

1130

1275 13

810

64 ***

***

***

1.46

41.

331

4.25

41.

573

***

***

***

250.

015

0.0

865.

018

9.0

82.3

87.3

34.0

51.0

74.1

83.1

23.1

50.2

62.7

79.8

39.2

73.6

***

*** * ***

***

90.9

91.3

88.8

89.9

91.5

91.1

87.0

93.0

89.3

89.2

79.9

88.9

***

***

***

*** *

Tabl

e9.

5.Ef

fect

ofdi

etar

yfa

ttyp

ean

dca

rboh

ydra

sesu

pple

men

tatio

nof

broi

lera

ndtu

rkey

diet

son

body

-wei

ghtg

ain

(BW

G,g

),fe

ed-to

-gai

nra

tio(F

CR

,gg−1

),in

test

inal

visc

osity

(mPa

s),a

ppar

entf

atdi

gest

ibilit

y(F

D%

)and

appa

rent

prot

ein

dige

stib

ility

(PD

%).





1.En

zym

e×

age

1.Fa

t ×en

zym

e×

age

* NS

* *

1.Br

oile

r2.

21da

ys3.

Day

s1–

35

4.Fa

t4.

Enzy

me

4.Fa

t ×en

zym

e

56 56 56 56

10.0

10.0

10 10

− + − +

749

761

673

714

***

NS

NS

1.34

81.

352

1.48

01.

397

*** ** **

17.0 9.3

21.9

12.4

** ***

NS

91.7

93.0

45.3

62.8

*** ** **

87.8

87.9

86.2

88.4

NS

NS

NS

[2]

1.Br

oile

r2.

21da

ys3.

Day

s1–

35

4.Fa

t4.

Enzy

me

4.Fa

t ×en

zym

e

56 56 56 56

10.0

10.0

10 10

− + − +

826

846

778

824

** ** NS

1.26

21.

240

1.36

61.

291

*** * NS

8.8

4.7

10.2 4.6

NS ***

NS

87.2

90.0

67.4

66.7

***

NS

NS

85.2

88.7

88.3

89.2

NS * NS

[3]

1.Tu

rkey

s2.

Day

s15

–20

3.D

ays

15–4

2

4.Fa

t4.

Enzy

me

4.Fa

t ×en

zym

e

48 48 48 48 48 48

8.0

8.0

8 88 8

− + − + − +

ac17

98ac

c 1847

a

c 1651

b

ac18

01ac

c 1745

c

c 1867

a

** ***

NS

ac1.

53ac

b 1.39

b

b 1.64

c

bb1.

52ac

bb1.

56ac

bb1.

48ab

* ***

NS

a 87a .0

a 86a .0

a 54b .0

a 62c .0

a 82a .0

a 79a .0

***

NS

NS

a 87a .0

a 91c .0

ab88

ab.0

a 91c .0

a 91c .0

ab90

bc.0

NS

NS

NS



216 S. Dänicke


1.Bi

rdty

pe2.

Age

(bal

ance

exp.

)3.

Age

(gro

wth

exp.

)4.

ANO

VA( P

)

Die

tary

cere

alin

clus

ion

(%)

Die

tary

fati

nclu

sion

(%)

Ref

no.

RT

WM

BS

TL

ABC

EBW

GFC

RVi

s1FD

2PD

2

[2]

1.Br

oile

r2.

21da

ys3.

Day

s1–

35

4.Fa

t4.

Enzy

me

4.Fa

t ×en

zym

e

56 56 56 56

10.0

10.0

10 10

− + − +

748

742

685

705

***

NS

NS

1.25

31.

236

1.35

31.

307

*** ** NS

3.9

2.9

3.2

2.4

***

***

NS

92.3

93.1

65.2

63.4

***

0.05 NS

91.3

90.9

92.0

90.6

NS

NS

NS

[2]

1.Br

oile

r2.

21da

ys3.

Day

s1–

35

4.Fa

t4.

Enzy

me

4.Fa

t ×en

zym

e

56 56 56 56

10.0

10.0

10 10

− + − +

760

797

641

755

***

*** *

1.27

61.

257

1.46

61.

344

***

*** **

13.8 5.0

23.8 7.6

** ***

NS

86.3

90.2

51.3

59.4

***

NS

NS

83.0

88.3

83.8

88.7

NS ***

NS

Tabl

e9.

5.C

ontin

ued.

226



[4]

1.Br

oile

r2.

Day

s21

–24

3. 4.Fa

t4.

Enzy

me

4.Fa

t ×en

zym

e

30 30 30 30 30 30 30 30

29 29 29 29 29 29 29 29

6.0

6.0

3.0

3.0

3 3 6 66 6

− + − + − + − +

730

783

712

765

681

777

690

804

NS ***

NS

2.06

1.76

2.12

1.81

2.33

1.93

2.07

1.83 **

***

*N

S

49.8

16.5

41.2

16.4

30.7

20.5

46.2

17.9

NS ***

***

75.9

84.8

61.7

73.9

45.5

63.1

66.4

77.7

***

***

NS

83.8

85.7

86.3

86.5

85.2

86.1

85.6

85.7 ** NS

NS

[5]

1.Br

oile

r2.

Day

s24

–28

3.D

ays

1–21

4.

10 10 10 10

50 50 50 50

6.5

6.5

0.5

0.5

6.0

6.0

− + − +

b 636b

b 638b

b 567a

b 621b

bb1.

610ab

b 1.55

6ab

b 1.74

8cb

b 1.64

4b b

a 78.3

c

a 80.2

c

a 60.5

a

a 69.9

b

[1]D

änic

keet

al.(

1997

b),[

2]D

änic

keet

al.(

1999

g),[

3]G

roße

rand

Jero

ch(1

997)

,[4]

Smul

ikow

ska

and

Mie

czko

wsk

a(1

996)

,[5]

Lang

hout

etal

.(19

97).

1In

test

inal

visc

osity

mea

sure

din

dige

sta

ofje

junu

mpl

usile

um(re

fere

nce

[4])

orin

dige

sta

ofile

um(re

fere

nces

[1]t

o[3

]).2

Dig

estib

ility

was

mea

sure

dat

the

faec

alle

vel(

refe

renc

es[1

],[3

],[4

],[5

])or

atth

ete

rmin

alile

um(re

fere

nce

[2]).

Abbr

evia

tions

:R,r

ye;W

,whe

at;T

,trit

ical

e;M

,mai

ze;B

,bar

ley;

S,so

yaoi

l;T,

beef

tallo

w;L

,lar

d;AB

,ani

mal

fatb

lend

;C,c

ocon

utoi

l;E,

enzy

me

prep

arat

ion.

a–c Va

lues

with

noco

mm

onsu

pers

crip

tare

sign

ifica

ntly

diffe

rent

with

inco

lum

ns(P

<0.

05).

Prob

abilit

y:* P

<0.

05;*

*P<

0.01

;***

P<

0.00

1;N

S,no

tsig

nific

ant.

227

instead of soybean oil (Dänicke et al., 1997a). Thus, fat type clearly influences themicrobial composition in the intestine independently of viscosity. It is not knownif cocci growth was actively stimulated by tallow addition or if it was merely a resultof inhibition of other bacteria, possibly due to increased secretion of bile acids bythe host. Bile acids have a bacteriostatic effect, but this effect varies with each speciesof interest. Cocci can be distinguished by their ability to grow in the presence ofrelatively high concentrations of bile salts; thus the most resistant cocci, such asEnterococci, may become dominant under higher bile acid concentrations. Activede-conjugation of conjugated bile acids, as carried out by some lactobacilli, alsooccurs in the intestinal tract, but it is not known if the generated bile salt hydrolaseactivity is sufficient to interfere quantitatively with the bile salt metabolism.

Interactions between Dietary Fat Type and CarbohydraseAddition in Broiler Diets Containing Different Cereals

General remarks

In the previous section an attempt was made to explain why exogenously suppliedcarbohydrases capable of reducing intestinal viscosity could result in different effectsdepending upon, amongst other factors, the chemical composition of the dietary fatsemployed.

Several experiments have demonstrated that carbohydrase supplementation ofbroiler and young turkey diets results in far greater responses in nutrient digestibilityand performance when the fat source contains more LCSFAs. Additionally, theperformance and nutrient digestibility responses are more closely related to changes

218 S. Dänicke


Fig. 9.3. Effects of dietary proportion of rye, dietary fat type and xylanasesupplementation on jejunal and ileal supernatant viscosity of broilers (Dänickeet al., 1999c). (Without xylanase, –––; with xylanase, -----; jejunum, q; ileum, r.)

228

in intestinal viscosity in animals fed such fats. Therefore, at a comparable total fatsupply, the response to carbohydrase supplementation would increase in the ordermaize < wheat < wheat/rye, triticale, barley < rye/wheat < rye (Table 9.5) since thisis the general order of increasing viscosity. Whilst this order will remain regardlessof dietary fat type, the relative differences between grains will be much smaller iffat blends (Table 9.5) or lower dietary fat proportions are used. Under practicalconditions, one can expect a fat blend in most cases rather than pure fats in broilerdiets, for several reasons. Full-fat seeds (especially soybean) are often used as fat orenergy sources for feed-manufacture reasons and to decrease the proportion of‘free’ supplemented fat. The use of maize or oats will itself increase the dietary plantoil proportion substantially. Thus, the scale of response seen on enzyme supplement-ation is mediated by total dietary fat concentration, fatty acid profile and triglyceridestructure, mineral content and age of the birds (see above sections). The interactionsare considered in more detail in the following sections.

Intestinal viscosity

It is now well established that the anti-nutritive effects of soluble NSP inbroilers are mediated by an increase in intestinal viscosity. It can be seen from theresults presented in Table 9.5 that intestinal viscosity was markedly decreased bycarbohydrase supplementation. Significant interactions between dietary fat type andcarbohydrase addition were especially observed in rye-fed birds in the experimentsby Dänicke et al. (1999c,d) (Fig. 9.3) in that enzyme effects were more pronouncedin tallow-fed birds. It was concluded that tallow itself, and other unabsorbedmaterials, contributed to intestinal viscosity. These results are in contrast to thosereported by Smulikowska and Mieczkowska (1996), who found the xylanase-dependent decrease in intestinal viscosity to be more pronounced for birds fedsoybean oil than for those fed tallow (Table 9.5). For excreta, however, a substantiallyhigher viscosity was measured in broilers fed unsupplemented diets containingtallow or lard, whereas xylanase supplementation reduced excreta viscosity tocomparable levels for all fat types tested that caused the significant interactionsbetween fat type and enzyme addition. It was suggested that more insoluble NSPdissolved in the last part of the digestive tract, due to a decrease in MRT mediatedby undigested fat and viscosity that could have contributed to the increased excretaviscosity. It should be borne in mind that the described steep response of viscosityto small incremental changes in polysaccharide concentration above a viscosity of10–20 mPa s might cause a remarkable variation in viscosity even in the sameexperimental groups.

Digestibility of fat and fatty acids

Carbohydrase supplementation of broiler diets based on rye, rye/wheat, triticale orbarley resulted in larger improvements in fat digestibility if the dietary fat was of





animal origin compared with plant. In simple terms, it could be concluded that thereis more scope for improvement in fat digestibility when it is low to begin with. Thisinterpretation is probably the reverse of how such data should be viewed. An increasein intestinal viscosity depresses fat digestibility more in animal fat-based diets, due tothe mechanisms explained above. Consequently, a reduction in viscosity due toenzyme addition in such diets should exert a more pronounced effect. Digestibilityof different dietary fat types clearly demonstrated what was expected, i.e. a relativelyhigh digestibility of soybean oil, which was less drastically modified over a wide rangeof dietary soluble NSP concentrations, and consequently intestinal viscosity, thanbeef tallow. Intermediate effects are observed for lard and animal and vegetable fatblends. That enzyme effect on fat digestibility strongly depends on dietary pentosanconcentration was clearly demonstrated in dose–response studies by Smulikowskaand Mieczkowska (1996) where significant interactions described by greater enzymeresponses with higher rye proportions were observed (Fig. 9.4). Moreover, Dänickeet al. (1999c) showed that this relationship is further modified by dietary fat type.The use of beef tallow in place of of soybean oil resulted on the one hand in a greaterreduction in fat digestibility with increasing proprtions of rye in the diet, andconsequently on the other hand in greater enzyme effects in such diets (Fig. 9.5) –hence the significant interactions between fat type and rye proportion, and betweenrye proportion and enzyme supplementation. However, in evaluation of such effectsit should be stressed that the effects of enzyme on fat digestibility decrease as birdsage (Fig. 9.6). That such age-related effects were not observed in the experimentsby Dänicke et al. (1997b) (Table 9.5) might be explained by the restricted feeding

220 S. Dänicke


Fig. 9.4. Effect of dietary rye proportion and xylanase addition (1.5 g AvizymeTx per kg of each diet) on apparent fat digestibility and metabolizability ofenergy in broilers (each diet contained 4.25% soybean oil and 4.25% lard;data from Smulikowska and Mieczkowska, 1996). (Apparent fat digestibility, –––;metabolizability of engergy, -----; without xylanase, r; with xylanase, v.)



regimen applied in these tests, which might modify the relationship between feedsupply, body weight, endogenous fat loss and mean retention time.

The differences in crude fat digestibility can be explained by differences indigestibility of the individual fatty acids (Table 9.6, Figs 9.7 and 9.8). Unsaturatedfatty acids were more efficiently absorbed than saturated. The digestibility ofsaturated fatty acids decreased with increasing chain length. Within the C18 family,digestibility increased with increasing number of double bonds. These general



Fig. 9.5. Effect of dietary proportion of rye, dietary fat type and xylanasesupplementation on apparent crude fat digestibility at the terminal ileum(from Dänicke et al., 1999c). Without xylanase, q; with xylanase, w.

Fig. 9.6. Effect of age and β-glucanase addition on fat digestibility in broilers fedbarley-based diets (60% and 70% barley and 5% and 6% tallow inclusion in starterand grower diets, respectively; data from Salih et al., 1991). Without β-glucanase,r; with β-glucanase, v.



222 S. Dänicke


1.Bi

rdty

pe2.

Age

3.AN

OVA

( P)

Die

tary

cere

alin

clus

ion

(%)

Die

tary

fati

nclu

sion

(%)

Car

bo-

hydr

ase

Ref

no.

Rye

Whe

atSo

ybea

noi

lBe

efta

llow

Coc

onut

oil

C12

:0C

16:0

C18

:0C

18:1

C18

:2C

18:3

[1]

1.Br

oile

r2.

Day

s21

–25

3.Fa

t3.

Enzy

me

3.Fa

t ×en

zym

e

61 61 61 61

10 1010 10

− + − +

74.3

81.2

39.0

46.5

***

NS

NS

54.1

61.8

23.4

28.8

***

NS

NS

81.0

86.6

59.5

71.2

*** * NS

88.6

92.2

64.3

74.7

*** * NS

92.1

94.9

72.3

80.5

***

NS

NS

[2]

1.Tu

rkey

s2.

Day

s15

–20

48 48 48 48 48 48

8 88 8

8 8

67.3

b

65.8

b

b 79.6

ab

90.8

a

94.6

a

93.8

a

80.9

a

78.2

a

40.1

c

c 49.6

bc

58.1

b

56.9

b

69.0

a

66.9

a

24.4

c

c 34.7

bc

46.3

b

44.6

b

86.2

a

86.6

a

74.0

b

84.6

a

c 79.1

ab

c 78.6

ab

88.7

a

c 87.9

ab

76.1

c

c 81.8

bc

cc82

.4ab

c

80.9

c

91.2

a

c 89.1

ab

83.1

c

cc87

.0ab

c

cc86

.5ab

c

c 84.1

bc

[1]D

änic

keet

al.(

1997

b),[

2]G

roße

rand

Jero

ch(1

997,

pers

onal

com

mun

icat

ion)

.a–

c Valu

esw

ithno

com

mon

supe

rscr

ipta

resi

gnifi

cant

lydi

ffere

ntw

ithin

colu

mns

(P<

0.05

).Pr

obab

ility:

* P<

0.05

;***

P<

0.00

1;N

S,no

tsig

nific

ant.

Tabl

e9.

6.Ef

fect

ofdi

etar

yfa

ttyp

ean

dca

rboh

ydra

sesu

pple

men

tatio

non

faec

aldi

gest

ibilit

yof

fatty

acid

sin

broi

lers

and

turk

eys.



relationships were shown to be influenced by dietary rye proportion and dietary fattype, as shown in Figs 9.7 and 9.8. However, the relative decrease in digestibility ofsaturated fatty acids as rye replaced wheat was greater for the tallow-supplementedrations. Replacing wheat with rye reduced digestibility of palmitic and stearic acidby approximately 24% and 48%, respectively, in birds fed soybean oil, whilst for



Fig. 9.7. Effect of dietary proprotion of rye, dietary fat type and xylanasesupplementation on apparent fatty acid digestibility at the terminal ileum (fromDänicke et al., 1999c). Without xylanase, –––; with xylanase, -----; stearic acid, r;oleic acid, q; linoleic acid, v.

Fig. 9.8. Apparent fatty acid digestibility as influenced by dietary fat type andxylanase supplementation at different measurement sites (from Dänicke et al.,1999c). Without xylanase, –––; with xylanase, -----; stearic acid, r; oleic acid, q;linoleic acid, v.



tallow-fed birds this increased to 44% and 59%, respectively. In contrast totallow-fed birds, digestibility of unsaturated fatty acids in those fed soybean oil wasless affected by intestinal viscosity. In all cases, increased viscosity reduced saturatedfatty acid digestibility more dramatically compared with unsaturated fatty acids,regardless of source. However, the original source of the fatty acids had a markedinfluenced on the absolute response in fatty acid digestibility to increased intestinalviscosity. Increased viscosity reduced digestibility of a given fatty acid to a greaterextent when it was derived from tallow than when it was derived from soybean oil.

Significant interactions between fat type and xylanase supplementation arecaused by higher enzyme effects in tallow-fed birds and significant interactionsbetween rye level and enzyme supplementation reflect increasing xylanase effectswith increasing rye proportions, especially in tallow-fed birds.

The effects of a xylanase addition to a turkey diet containing coconut oil(Table 9.6) have to be evaluated in a different way. Whereas both soybean oil andbeef tallow are composed of fatty acids with a chain length of 16–18 C-atoms, morethan 50% of coconut oil is saturated fatty acids with fewer than 14 C-atoms. Theseshorter fatty acids are less dependent on bile salts and micelle formation than theirlonger-chained counterparts, which is clearly indicated by greater absorption of lauricacid in poults fed coconut oil (Table 9.6). The digestibility of palmitic and stearicacid from coconut oil lies between that of soybean oil and tallow fed to poults. Therewere virtually no differences in the digestibility of unsaturated fatty acids from turkeysfed tallow or coconut oil. Such results suggest a relative deficiency of bile salts in tallow-fed poults, whereas sufficient bile salts were available in turkeys fed coconut oil to aidin emulsification and absorption of greater amounts of palmitic and stearic acid.

With respect to a species comparison, it should be noted that the young turkeyseemed to be more sensitive to dietary treatments than the broiler. Comparabledigestibility values were observed between species but the diets differed in that theturkeys received diets containing less pentosans and less supplemented fat.

There were also differences in main sites of absorption observed betweensaturated and unsaturated fatty acids. The latter disappeared to a greater extent fromthe proximal segments of the small intestine, whereas the former were digested andabsorbed in the more distal segments (Fig. 9.8).

Digestibility of protein and amino acids

The effect of intestinal viscosity on protein and amino acid digestibility is generallyless dramatic than on digestibility of fat and fatty acids (Table 9.5). Digestibility ofcrude protein and that of some amino acids at the terminal ileum was decreasedas dietary pentosan content was increased and significantly improved by xylanaseaddition (Fig. 9.9). No fat effect or interactions were detected at this site. Incontrast, measurements made over the whole gastrointestinal tract showed signifi-cantly lower protein and amino acid digestibility values for tallow-fed birds, andsignificant greater enzyme effects in tallow-containing diets, especially in dietswith higher pentosan concentrations (Fig. 9.9). Langhout et al. (1997) reported the

224 S. Dänicke




improvement in faecal digestibility of amino acids after xylanase addition to a wheat/rye-containing diet to be greater in birds fed blended animal fat compared with thosefed soybean oil. The improvements for the individual amino acids varied between−0.5% and 1.5% in the soya oil group and 0.9% and 4.3% in the blended animalfat group. Moreover, Rotter et al. (1990) combined a highly viscous barley witheither 4% maize oil or 4 % tallow and tested these diets without or with an enzymeaddition. A significant enzyme-related improvement was found but there was noeffect of dietary fat type. Smulikowska and Mieczkowska (1996) found that proteindigestibility did not differ significantly between diets containing either 6% tallow or6% soybean oil in the presence of 30% rye. Xylanase addition improved digestibilityonly slightly. No interactions were observed in that study. Furthermore, Großerand Jeroch (1997) found a xylanase-related increase in apparent faecal proteindigestibility but failed to demonstrate fat effects in turkeys fed either soybean oil orbeef tallow containing wheat-based diets.

Energy metabolism

Increased intestinal viscosity results in an increase in the absolute and relative weightof the gastrointestinal tract, especially of the small intestine. Increasing the relativeproportion of gut and liver in rats results in a simultaneous decrease in utilization ofME for energy retention and an increase in energy maintenance requirement (MEm)(Ferrell and Koong, 1986). Furthermore, Pekas and Wray (1991) detected a highcorrelation between fasting heat production and several gut measurement variables inpigs, such as mass and length but also gastrointestinal fill. In contrast, only weak orno correlations at all were observed between heat production and empty body mass



Fig. 9.9. Praecaecal and faecal protein digestibility as influenced by ryeproportion, dietary fat type and xylanase supplementation (Dänicke et al., 1999e).Without xylanase, –––; with xylanase, -----; praecaecal, r; faecal, q.



or carcass. These results would suggest that changes in the proportions of the gut andgut fill in relation to liveweight due to an increased intestinal viscosity shouldalso affect heat production or energy metabolism. Simon (1998) calculated that a30% increase in intestines as a proportion of body weight would increase the heatproduction of a broiler of 1 kg by approximately 5%. Jørgensen et al. (1996) reportedan increase in heat production as a proportion of ME in broilers after feeding of peafibre containing high amounts of soluble NSP, whereas feeding of wheat or oat brandid not increase heat production. Addition of xylanase to rye-based broiler diets wasshown to increase the caloric efficiency (Table 9.7), i.e. the proportion between netenergy and ME, the partial efficiency of ME for net energy above MEm (kpf) but alsoMEm itself. Caloric efficiency includes both maintenance heat and heat of energyaccretion, thus the proportion between heat production and ME was decreased byxylanase supplementation. An additional effect of dietary fat was mostly obvious formetabolizability of energy, which was probably caused by differences in fatdigestibility.

Generally, improved energy utilization might be expected followingcarbohydrase addition as a result of an improvement in both metabolizability ofenergy (fat digestion) and an increased metabolic utilization of absorbed nutrients(increased caloric efficiency).

Nitrogen-corrected apparent metabolizable energy (AMEN) content as a measureof the overall nutrient effect on energy metabolism decreased with increasing dietarypentosan content. This measure was significantly improved by addition of xylanaseand was lower in tallow-fed birds (Dänicke et al., 1999e). Again, xylanase effects werefound to be greater for tallow-fed birds and at higher pentosan concentrations.

Protein metabolism

Expected effects on protein metabolism may also be expected since increases inintestinal viscosity result in an altered pancreatic function and an increased gut

226 S. Dänicke


Metabolizability ofgross energy

Energy retention(kJ day−1 W−0.75)

Caloricefficiency(kJ kJ−1)

Fatretention

(% ofWG)

kpf

(kJ kJ−1)

MEm

(kJ day−1

W−0.75)Fattype kJ kJ−1

Protein(NEp)

Fat(NEf)

Total(NEpf)

S −S +T −T +

0.643ab

0.706bb

0.606ab

0.656ab

299ab

315ab

301ab

354bb

242ab

295bc

286ab

340cb

540a

611a

588a

694b

0.421ab

0.449ab

0.426ab

0.462bb

11.0ab

12.1ab

12.4ab

13.6bb

0.665ab

0.729ab

0.662ab

0.886bb

468521482720

a–cValues with different superscripts within the columns are significantly different (P < 0.05).

Table 9.7. Effects of dietary fat type (S, 10% soybean oil; T, 10% beef tallow) and enzymesupplementation (−, without, +, with 1 g Avizyme 1300 kg−1 diet) in a rye-based diet (56% dietaryinclusion) on energy metabolism in male broilers (Dänicke et al., 1999f).



weight and consequently an increased contribution of gut protein mass to the overallprotein mass of the bird. It was shown by Dänicke et al. (2000a) that tissue fractionalprotein synthesis, i.e. the daily protein synthesis expressed as percentage of total tissueprotein, is significantly greater in small intestinal tissues and in the pancreas of birdsfed an unsupplemented tallow-containing rye-based diet (Table 9.3) compared withtheir soybean oil-fed counterparts. Xylanase supplementation reduced intestinaland pancreatic weights and also the fractional protein synthesis rates of these tissuesto a level comparable to birds fed the soybean oil diets. Increased intestinal proteinsynthesis might be related to increased intestinal endogenous N-losses, since it wasdemonstrated by Dänicke et al. (2000b) that endogenous N-losses, as measured by anisotope dilution technique, were threefold greater in birds fed the tallow-containingrye-based diet (Table 9.3). Moreover, Langhout et al. (1999) measured an increasednumber of goblet cells per 100 villus cells when high or low methylated pectin wasadded to a maize-based broiler diet, which could result in an increased secretion ofmucin.

The lack of an enzyme effect on intestinal parameters in soybean oil-fed birdswould suggest that soybean oil may have exerted a protective effect on the intestinalvilli and cell proliferation rate, since intestinal viscosity is also reduced by enzymesupplementation in these birds.

Protein synthesis in the pancreas is closely associated with synthesis of secretoryproteins, i.e. hydrolases such as lipase, protease and amylase. Therefore, it mightbe concluded that increased pancreatic protein synthesis rate in birds fed theunsupplemented tallow-containing diet might have been related to increaseddigestive enzyme secretion. Further support for involvement of both intestinal andpancreatic contributions to digestive endogenous N-losses comes from Dänicke et al.(2000b), who reported that the [15N]-enrichment of digesta and faeces of broilerswas equally as a result of [15N]-enrichments of pancreas and intestinal tissues of[15N]-labelled animals.

Performance and carcass characteristics

Performance is the most important criterion from a practical point of view. It corre-lates more or less with fat digestibility but reflects also the described interactionsbetween fat type and enzyme supplementation. In most experiments, liveweight gainand feed-to-gain ratio were improved to a greater extent by carbohydrase addition inbroilers fed diets containing animal-originated fats (Table 9.5, Fig. 9.10). Althoughthe interactions were not always significant they should not be neglected, since thetendency is obvious in most cases. As already shown for fat digestibility, carbohydraseeffects depend on dietary concentration of soluble NSP (Fig. 9.5). Enzyme effectscan be expected to be greater in birds fed animal fat and at higher concentrations ofdietary soluble NSP. Total mortality (including losses by culling) depends also on fattype, carbohydrase addition and rye proportion (Dänicke et al., 1999d). Mortalityincreased significantly with the introduction of rye into the diets and was muchgreater in groups fed the unsupplemented tallow-containing diets, where mortality





of 23–26% was observed. The viscosity causing stickiness of excreta was furtheramplified in tallow-fed birds due to large amounts of undigested fat in the litterand around the cloaca of the birds. Their feathers appeared dirty and wet, whichundoubtedly increases pathogenic microbial stress.

228 S. Dänicke


Fig. 9.10. Interactions between rye percentage (pentosan level), fat type andxylanase supplementation for cumulative feed to gain ratio of male broilers. Day1–35 of age: without xylanase, w; with xylanase, q. (From Dänicke et al., 1999d.)

Fig. 9.11. Interactions between rye percentage (pentosan level), fat type andxylanase supplementation for abdominal plus visceral fat of male broilers. Day 35of age: without xylanase, w; with xylanase, q. (From Dänicke et al., 1999d.)



It should be stressed that interactions between fat type and carbohydrasesupplementation were detected both in semi-purified diets and in practical diets.

These quantitative changes in performance as a result of enzyme and fat inter-actions are paralleled by changes in carcass characteristics and liver lipid composition.Increased fat digestibility was shown to increase broiler fatness (Fig. 9.11), which isclosely related to energy availability and energy metabolism and affects skin colour,since the major pigments are fat soluble. It can be seen from the relationships shownin Fig. 9.12 that increasing fatness, as a result of either xylanase supplementation orreduced rye levels, is associated with a more yellow skin colour (increasing b-values),more light (increasing L-values) and less red (decreasing b-values).

The liver concentrations of the fat-soluble vitamins A and E were also enhancedafter enzyme addition as a result of improved fat digestibility (Dänicke et al., 1997b,1999c). Alterations in liver fatty acid composition were also explained by bothdietary fat type and xylanase addition (Dänicke et al., 1999c).

Concluding Remarks

There is a strong body of evidence from a number of sources which indicates thatdietary fat type is a factor that modifies the overall effect of carbohydrase supple-mentation in broilers. Generally, greater enzyme effects can be expected if a fat ofanimal origin is used rather than one of plant origin. The importance of suchinteractions increases: (i) with increasing dietary proportions and absolute amountsof an animal fat; (ii) with increasing dietary concentrations of soluble NSP; and



Fig. 9.12. Relationship between abdominal plus visceral fat and parameters ofdrumstick skin colour. Minolta L, q; a, w; b, v; S, soybean oil; T, tallow; −, withoutxylanase supplementation; +, with xylanase supplementation; 0, 33, 66, 100%,respectively, rye of total dietary cereals. (From Dänicke et al., 1999d.)



(iii) in early fattening periods. The detection of such interactions even in practicaldiets leads to the conclusion that animal fat-supplemented diets that are based on rye,triticale, barley or wheat, or on a mixture thereof, should be supplemented with anappropriate carbohydrase preparation to avoid adverse effects on performance.

References

Almirall, M. and Esteve-Garcia, E. (1994) Rate of passage of barley diets with chromiumoxide: influence of age and poultry strain and effect of β-glucanase supplementation.Poultry Science 73, 1433–1440.

Almirall, M., Francesch, M., Perez-Vendrell, A.M., Brufau, J. and Esteve-Garcia, E. (1995)The differences in intestinal viscosity produced by barley and β-glucanase alter digestaenzyme activities and ileal nutrient digestibilities more in broiler chicks than in cocks.Journal of Nutrition 125, 947–955.

Annison, G., Moughan, P.J. and Thomas, D.V. (1995) Nutritive activity of soluble rice branarabinoxylans in broiler diets. British Poultry Science 36, 479–488.

Antoniou, T., Marquardt, R.R. and Misir, R. (1980) The utilization of rye by growing chicksas influenced by calcium, vitamin D3, and fat type and level. Poultry Science 59, 758–769.

Bedford, M.R. and Classen, H.L. (1992) Reduction of intestinal viscosity through manipula-tion of dietary rye and pentosanase concentration is effected through changes in thecarbohydrate composition of the intestinal aqueous phase and results in improved growthrate and food conversion efficiciency of broiler chicks. Journal of Nutrition 122, 560–569.

Blackburn, N.A., Holgate, A.M. and Read, N.W. (1984) Does guar gum improve post-prandial hyperglycaemia in humans by reducing small intestinal contact area? BritishJournal of Nutrition 52, 197–204.

Blanch, A., Barroeta, A.C., Baucells, M.D. and Puchal, F. (1995) The nutritive value ofdietary fats in relation to their chemical composition. Apparent fat availability andmetabolizable energy in two-week-old chicks. Poultry Science 74, 1335–1340.

Blanch, A., Barroeta, A.C., Baucells, M.D., Serrano, X. and Puchal, F. (1996) Utilization ofdifferent fats and oils by adult chickens as a source of energy, lipid and fatty acids. AnimalFeed Science and Technology 61, 335–342.

Bracco, U. (1994) Effect of triglyceride structure on fat absorption. American Journal ofClinical Nutrition 60, 1002S–1009S.

Brink, E.J., Haddeman, E. and de Fouw, N.J. (1995) Positional distribution of stearic acidand oleic acid in a triacylglycerol and dietary calcium concentration determines theapparent absorption of these fatty acids in rats. Journal of Nutrition 125, 2379–2387.

Buhman, K.K., Furumoto, E.J., Donkin, S.S. and Story, J.A. (1998) Dietary psylliumincreases fecal bile acid excretion, total steroid excretion and bile acid biosynthesis in rats.Journal of Nutrition 128, 1199–1203.

Dänicke, S., Simon, O., Jeroch, H. and Bedford, M. (1997a) Interactions between dietary fattype and xylanase supplementation when rye-based diets are fed to broiler chickens. 1.Physico-chemical chyme features. British Poultry Science 38, 537–545.

Dänicke, S., Simon, O., Jeroch, H. and Bedford, M. (1997b) Interactions between dietary fattype and xylanase supplementation when rye-based diets are fed to broiler chickens. 2.Performance, nutrient digestibility and the fat-soluble vitamin status of livers. BritishPoultry Science 38, 546–556.

230 S. Dänicke




Dänicke, S., Simon, O. and Jeroch, H. (1999a) Effects of supplementation of xylanase orβ-glucanase containing enzyme preparations to either rye- or barley-based broiler diets onperformance and nutrient digestibility. Archiv für Geflügelkunde 63, 252–259.

Dänicke, S., Vahjen, W., Simon, O. and Jeroch, H. (1999b) Effects of dietary fat type andxylanase supplementation to rye-based broiler diets on selected bacterial groups adheringto the intestinal epithelium, on transit time of feed and on nutrient digestibility. PoultryScience 78, 1292–1299.

Dänicke, S., Jeroch, H., Böttcher, W., Bedford, M.R. and Simon, O. (1999c) Effects ofdietary fat type, pentosan level and xylanases on digestibility of fatty acids, liver lipids andvitamin E in broilers. Fett/Lipid 101, 90–100.

Dänicke, S., Jeroch, H., Dusel, G., Simon, O. and Bedford, M. (1999d) Effects of fat type,pentosan level and xylanase supplementation on performance and carcass characteristicsof male broilers. Archiv für Geflügelkunde 63, 194–203.

Dänicke, S., Simon, O., Jeroch, H., Keller, K., Gläser, K., Kluge, H. and Bedford, M.R.(1999e) Effects of dietary fat type, pentosan level and xylanase supplementation ondigestibility of nutrients and metabolizability of energy in male broilers. Archives ofAnimal Nutrition 52, 245–261.

Dänicke, S., Franke, E., Strobel, E., Jeroch, H. and Simon, O. (1999f) Effects of dietary fattype and xylanase supplementation in rye containing diets on energy metabolism in malebroilers. Journal of Animal Physiology and Animal Nutrition 81, 90–102.

Dänicke, S., Jeroch, H., Simon, O. and Bedford, M.R. (1999g) Interactions between dietaryfat type and exogenous enzyme supplementation of broiler diets based on maize, wheat,triticale or barley. Journal of Animal and Feed Science 8, 467–483.

Dänicke, S., Böttcher, W., Jeroch, H., Thielebein, J. and Simon, O. (2000a) Replacement ofsoybean oil with tallow in rye-based diets without xylanase increases protein synthesis insmall intestine of broilers. Journal of Nutrition 130, 827–834.

Dänicke, S., Jeroch, H. and Simon, O. (2000b) Endogenous N-losses in broilers estimated bya [15N]-isotope dilution technique: effect of dietary fat type and xylanase addition.Archives of Animal Nutrition 53, 75–97.

Degolier, T.F., Duke, G.E. and Carraway, R.E. (1997) Neurotensin decreases pepsin outputand gastrointestinal motility in chickens. Poultry Science 76, 1435–1439.

Drakley, C., Wiseman, J. and Bedford, M.R. (1997) Changes in apparent metabolizableenergy, digesta viscosity and small intestinal morphology in broilers fed diets based onwheat. British Poultry Science, S27–S28.

Duke, G.E. and Evanson, O.A. (1972) Inhibition of gastric motility by duodenal contents inturkeys. Poultry Science 51, 1625–1636.

Edwards, C.A., Johnson, I.T. and Read, N.W. (1988) Do viscous polysaccharides slowabsorption by inhibiting diffusion or convection? European Journal of Clinical Nutrition42, 307–312.

Fedde, M.R., Waibel, P.E. and Burger, R.E. (1960) Factors affecting the absorbability ofcertain dietary fats in the chick. Journal of Nutrition 70, 447–452.

Ferrell, C.L. and Koong, K.J. (1986) Influence of plane of nutrition on body composition,organ size and energy utilization of Sprague-Dawley rats. Journal of Nutrition 116,2525–2535.

Filer, L.J., Mattson, F.H. and Fomon, S.J. (1969) Triglyceride configuration and fatabsorption by the human infant. Journal of Nutrition 99, 293–298.

Flourie, B., Vidon, N., Florent, C. and Bernier, J.J. (1984) Effect of pectin on jejunal glucoseabsorption and unstirred layer thickness in normal man. Gut 25, 936–941.





Friesen, O.D., Guenter, W., Marquardt, R.R. and Rotter, B.A. (1992) The effect of enzymesupplementation on the apparent metabolizable energy and nutrient digestibilities ofwheat, barley, oats, and rye for the young broiler chicks. Poultry Science 71, 1710–1721.

Fritsche, K.L., Cassity, N.A. and Huang, S.-C. (1991) Effect of dietary fat source on antibodyproduction and lymphocyte proliferation in chickens. Poultry Science 70, 611–617.

Furuya, S., Sakamoto, K., Asano, T., Takahashi, S. and Kameoka, K. (1978) Effects of addeddietary sodium polyacrylate on passage rate of markers and apparent digestibility bygrowing swine. Journal of Animal Science 47, 159–165.

Garrett, R.L. and Young, R.J. (1975) Effect of micelle formation on the absorption of neutralfat and fatty acids by the chicken. Journal of Nutrition 105, 827–838.

Gomez, M.X. and Polin, D. (1976) The use of bile salts to improve absorption of tallow inchicks, one to three weeks of age. Poultry Science 55, 2189–2195.

Großer, K. and Jeroch, H. (1997) Effektivität einer Xylanase-Supplementation beiweizenbetonten Rationen unter Zulage verschiedener Futterfette bei jungen Puten.In: Schubert, R., Flachowsky, G., Bitsch, R. and Jahreis, G. (eds) Proc. 6. Symp. Vitamineund Zusatzstoffe in der Ernährung von Mensch und Tier. pp. 341–346.

Halle, I. (1996) Untersuchungen zum Einfluß unterschiedlicher Nahrungsfette aufLeistungsparameter und Dotterfettsäurespektrum bei Legehennen. Archiv fürGeflügelkunde 60, 65–72.

Hohoff, C. and Spener, F. (1998) Fatty acid binding proteins and mammary-derived growthinhibitor. Fett/Lipid 100, 252–263.

Huyghebaert, G., de Munter, G. and de Groote, G. (1988) The metabolisable energy (AMEn)of fats for broilers in relation to their chemical composition. Animal Feed Science andTechnology 20, 45–58.

Ikegami, S., Tsuchihashi, F., Harada, H., Tsuchihashi, N., Nishide, E. and Innami, S. (1990)Effect of viscous indigestible polysaccharides on pancreatic–biliary secretion and digestiveorgans in rats. Journal of Nutrition 120, 353–360.

Isaksson, G., Lundquist, I. and Ihse, I. (1982) Effect of dietary fiber on pancreatic enzymeactivity in vitro, the importance of viscosity, pH, ionic strength, adsorption, and time ofincubation. Gastroenterology 82, 918–924.

Isaksson, G., Asp, N.-G. and Ihse, I. (1983a) Effects of dietary fiber on pancreatic enzymeactivities of ileostomy evacuates and on excretion of fat and nitrogen in the rat. Scandina-vian Journal of Gastroenterology 18, 417–423.

Isaksson, G., Lilja, P., Lundquist, I. and Ihse, I. (1983b) Influence of dietary fiber on exocrinepancreatic function in the rat. Digestion 27, 57–62.

Isaksson, G., Lundquist, I., Akesson, B. and Ihse, I. (1984) Effects of pectin and wheat branon intraluminal pancreatic enzyme activities and on fat absorption as examined withthe triolein breath test in patients with pancreatic insufficiency. Scandinavian Journal ofGastroenterology 19, 467–472.

Jeroch, H., Müller, A. and Gebhardt, G. (1990) Influence of beta-glucanase supplementationon digestion of different barley varieties in chickens. In: Proceedings VIII European PoultryConference. Barcelona, Spain, 25 June to 28 June 1990, pp. 259–262.

Johnson, I.T. and Gee, J.M. (1981) Effect of gel-forming gums on the intestinal unstirredlayer and sugar transport in vitro. Gut 22, 398–403.

Johnson, I.T. and Gee, J.M. (1982) Influence of viscous incubation media on the resistance todiffusion of the intestinal unstirred water layer in vitro. Pflügers Archiv 393, 139–143.

Johnson, I.T. and Gee, J.M. (1986) Gastrointestinal adaptation in response to solublenon-available polysaccharides in the rat. British Journal of Nutrition 55, 497–505.

232 S. Dänicke




Johnson, I.T., Gee, J.M. and Mahoney, R.R. (1984) Effect of dietary guar gum and celluloseon intestinal cell proliferation, enzyme levels and sugar transport in the rat. BritishJournal of Nutrition 52, 477–487.

Jørgensen, H., Zhao, X.-Q., Bach Knudsen, K.E. and Eggum, B.O. (1996) The influence ofdietary fibre source and level on the development of the gastrointestinal tract, digestibilityand energy metabolism in broiler chickens. British Journal of Nutrition 75, 379–395.

Katongole, J.B.D. and March, B.E. (1979) Fatty acid binding protein in the intestine of thechicken. Poultry Science 58, 372–375.

Katongole, J.B.D. and March, B.E. (1980) Fat utilization in relation to intestinal fatty acidbinding protein and bile salts in chicks of different ages and different genetic sources.Poultry Science 59, 819–827.

Ketels, E. and De Groote, G. (1987) The nutritional value of commercial fat blends in broilerdiets. 2. Effect of fat source and level of fat inclusion on the utilization of fatty acids inbroiler diets. Archiv für Geflügelkunde 51, 127–132.

Ketels, E. and De Groote, G. (1989) Effect of ratio of unsaturated to saturated fatty acids ofthe dietary lipid fraction on utilization and metabolizable energy of added fats in youngchicks. Poultry Science 68, 1506–1512.

Ketels, E., Huyghebaert, G. and De Groote, G. (1987) The nutritional value of commercialfat blends in broiler diets. 1. Effect of the incorporation level on the metabolizable energycontent. Archiv für Geflügelkunde 51, 59–64.

Kritchevsky, D. and Story, J.A. (1974) Binding of bile salts in vitro by nonnutritive fiber.Journal of Nutrition 104, 458–462.

Krogdahl, A. (1985) Digestion and absorption of lipids in poultry. Journal of Nutrition 115,675–685.

Krogdahl, A. and Sell, J.L. (1989) Influence of age on lipase, amylase, and protease activitiesin pancreatic tissue and intestinal contents of young turkeys. Poultry Science 68,1561–1568.

Kussaibati, R., Guillaume, J., Leclercq, B. and Lafont, J.P. (1982) Effects of the intestinalmicroflora and added bile salts on the metabolisable energy and digestibility of saturatedfats in the chicken. Archiv für Geflügelkunde 46, 42–46.

Lairon, D., Lafont, H., Vigne, J.-L., Nalbone, G., Leonardi, J. and Hauton, J.C. (1985)Effects of dietary fibers and cholestyramine on the activity of pancreatic lipase in vitro1–3. The American Journal of Clinical Nutrition 42, 629–638.

Langhout, D.J., Schutte, J.B., Geerse, C., Kies, A.K., De Jong, J. and Verstegen, M.W.A.(1997) Effects on chick performance and nutrient digestibility of an endo-xylanase addedto a wheat- and rye-based diet in relation to fat source. British Poultry Science 38,557–563.

Langhout, D.J., Schutte, J.B., Van Leeuwen, P., Wiebenga, J. and Tamminga, S. (1999)Effect of dietary high- and low-methylated citrus pectin on activity of the ileal microfloraand morphology of the small intestinal wall of broiler chicks. British Poultry Science 40,340–347.

Lund, E.K., Gee, J.M., Brown, J.C., Wood, P.J. and Johnson, I.T. (1989) Effect of oat gumon the physical properties of the gastrointestinal contents and on the uptake ofd-galactose and cholesterol by rat small intestine in vitro. British Journal of Nutrition 62,91–101.

Mattson, F.H. and Volpenhein, R.A. (1962) Rearrangement of glyceride fatty acids duringdigestion and absorption. Journal of Biological Chemistry 237, 53–55.





Mattson, F.H., Nolen, G.A. and Webb, M.R. (1979) The absorbability by rats of varioustriglycerides of stearic and oleic acid and the effect of dietary calcium and magnesium.Journal of Nutrition 109, 1682–1687.

Meyer, J.H., Gu, Y., Elashoff, J., Reedy, T., Dressman, J. and Amidon, G. (1986) Effects ofviscosity and fluid outflow on postcibal gastric emptying of solids. American Journal ofPhysiology 250, G161–G164.

Monsma, C.C., Gallaher, D.D. and Ney, D.M. (1996) Reduced digestibility of beef tallowand cocoa butter affects bile acid excretion and reduces hepatic esterified cholesterol inrats. Journal of Nutrition 126, 2028–2035.

Morris, E.R. (1992) Physico-chemical properties of food polysaccharides. In: Schweizer, T.F.and Edwards, C.A. (eds) Dietary Fibre – a Component of Food. Nutritional Function inHealth and Disease. Springer-Verlag, London, pp. 41–56.

Morris, E.R., Cutler, A.N., Ross-Murphey, S.B., Rees, D.A. and Price, J. (1981) Concentra-tion and shear rate dependence of viscosity in random coil polysaccharide solutions.Carbohydrate Polymers 1, 5–21.

Nir, I., Nitsan, Z. and Mahagna, M. (1993) Comparative growth and development of thedigestive organs and some enzymes in broiler and egg type chicks after hatching. BritishPoultry Science 34, 523–532.

Noy, Y. and Sklan, D. (1995) Digestion and absorption in the young chick. Poultry Science 74,366–373.

Ockner, R.K. and Manning, J.A. (1974) Fatty acid-binding protein in small intestine –identification, isolation, and evidence for its role in cellular fatty acid transport. Journal ofClinical Investigation 54, 326–338.

Ockner, R.K., Manning, J.A., Poppenhausen, R.B. and Ho, W.K.L. (1972a) A bindingprotein for fatty acids in cytosol of intestinal mucosa, liver, myocardium, and othertissues. Science 177, 56–58.

Ockner, R.K., Pittmann, J.P. and Yager, J.L. (1972b) Differences in the intestinal absorptionof saturated and unsaturated long chain fatty acids. Gastroenterology 62, 981–992.

Pasquier, B., Armand, M., Guillon, F., Castelain, C., Borel, P., Barry, J.-L., Pieroni, G. andLairon, D. (1996) Viscous soluble dietary fibers alter emulsification and lipolysis oftriacylglycerols in duodenal medium in vitro. Nutritional Biochemistry 7, 293–302.

Pekas, J.C. and Wray, J.E. (1991) Principal gastrointestinal variables associated with metabolicheat production in pigs: statistical cluster analysis. Journal of Nutrition 121, 231–239.

Phillips, D.R. (1986) The effect of guar gum in solution on diffusion of cholesterol mixedmicelles. Journal of Science of Food and Agriculture 37, 548–552.

Poksay, K.S. and Schneeman, B.O. (1983) Pancreatic and intestinal response to dietary guargum in rats. Journal of Nutrition 113, 1544–1549.

Polin, D. and Hussein, T.H. (1982) The effect of bile acid on lipid and nitrogen retention,carcass composition, and dietary metabolizable energy in very young chicks. PoultryScience 61, 1697–1707.

Polin, D., Wing, T.L., Ki, P. and Pell, K.E. (1980) The effects of bile acids and lipase onabsorption of tallow in young chicks. Poultry Science 59, 2738–2743.

Rakowska, M., Rek-Ciepty, B., Sot, A., Lipínska, E., Kubínski, T., Barcz, I. and Afanasjew, B.(1993) The effect of rye, probiotics and nisine on faecal flora and histology of the smallintestine of chicks. Journal of Animal and Feed Sciences 2, 73–81.

Read, N.W. and Eastwood, M.A. (1992) Gastro-intestinal physiology and function. In:Schweizer, T.F. and Edwards, C.A. (eds) Dietary Fibre - a Component of Food. NutritionalFunction in Health and Disease. Springer-Verlag, London, pp. 103–117.

234 S. Dänicke

A3882: A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-9



Renner, R. and Hill, F.W. (1961) Factors affecting the absorbability of saturated fatty acids inthe chick. Journal of Nutrition 74, 254–258.

Rotter, B.A., Friesen, O.D., Guenter, W. and Marquardt, R.R. (1990) Influence of enzymesupplementation on the bioavailable energy of barley. Poultry Science 69, 1174–1181.

Sagher, F.A., Dodge, J.A., Johnston, C.F., Shaw, C., Buchanan, K.D. and Carr, K.E. (1991)Rat small intestinal morphology and tissue regulatory peptides: effects of high dietary fat.British Journal of Nutrition 65, 21–28.

Salih, M.E., Classen, H.L. and Campbell, G.L. (1991) Response of chickens fed on hull-lessbarley to dietary β-glucanase at different ages. Animal Feed Science and Technology 33,139–149.

Satchithanandam, S., Reicks, M., Calvert, R.J., Cassidy, M.M. and Kritchevsky, D. (1993)Coconut oil and sesame oil affect lymphatic absorption of cholesterol and fatty acids inrats. Journal of Nutrition 123, 1852–1858.

Sibbald, I.R. and Kramer, J.K.G. (1977) The true metabolizable energy values of fats and fatmixtures. Poultry Science 56, 2079–2086.

Sibbald, I.R. and Kramer, J.K.G. (1978) The effect of the basal diet on the true metabolizableenergy value of fat. Poultry Science 57, 685–691.

Sickinger, K. (1975) Clinical aspects and therapy of fat malassimilation with particularreference to the use of medium-chain triglycerides. In: Vergroesen, A.J. (ed.) The Role ofFats in Human Nutrition. Academic Press, London, pp. 115–209.

Silva, S.S.P. and Smithard, R.R. (1997) Digestion of protein, fat and energy in rye-basedbroiler diets is improved by the addition of exogenous xylanase and protease. BritishPoultry Science 38, S38–S39.

Silva, S.S.P., Gilbert, H.J. and Smithard, R.R. (1997) Exogenous polysaccharidases do notimprove digestion of fat and protein by increasing trypsin or lipase activities in the smallintestine. British Poultry Science 38, S39–S40.

Simon, O. (1998) The mode of action of NSP hydrolysing enzymes in the gastrointestinaltract. Journal of Animal and Feed Sciences 7, 115–123.

Sklan, D. (1979) Digestion and absorption of lipids in chicks fed triglycerides or free fattyacids: synthesis of monoglycerides in the intestine. Poultry Science 58, 885–889.

Smulikowska, S. (1998) Relationship between the stage of digestive tract development inchicks and the effect of viscosity reducing enzymes on fat digestion. Journal of Animal andFeed Sciences 7, 125–134.

Smulikowska, S. and Mieczkowska, A. (1996) Effect of rye level, fat source and enzymesupplementation on fat utilization, diet metabolizable energy, intestinal viscosity andperformance of broiler chickens. Journal of Animal and Feed Sciences 5, 379–393.

Sudendey, C. and Kamphues, J. (1995) Effekte einer Enzymzulage (alpha-Amylase, Xylanase,β-Glucanase) auf Verdauungsvorgänge im Magen-Darmtrakt von Absetzferkeln unterden Bedingungen einer forcierten Futteraufnahme. Proceedings of Society of Nutrition andPhysiology 4, 108.

Van der Klis, J.D. and Van Voorst, A. (1993) The effect of carboxy methyl cellulose (a solublepolysaccharide) on the rate of marker excretion from gastrointestinal tract of broilers.Poultry Science 72, 503–512.

Veen, W.A.G., Grimbergen, A.H.M. and Stappers, H.P. (1974) The true digestibility andcaloric value of various fats used in feeds for broilers. Archiv für Geflügelkunde 6,213–220.

Vila, B. and Esteve-Garcia, E. (1996a) Studies on acid oils and fatty acids for chickens. I.Influence of age, rate of inclusion and degree of saturation on fat digestibility andmetabolisable energy of acid oils. British Poultry Science 37, 105–117.


A3882:A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-9



Vila, B. and Esteve-Garcia, E. (1996b) Studies on acid oils and fatty acids for chickens. II.Effect of free fatty acid content and degree of saturation of free fatty acids and neutral faton fatty acid digestibility. British Poultry Science 37, 119–130.

Vila, B. and Esteve-Garcia, E. (1996c) Studies on acid oils and fatty acids for chickens. III.Effect of chemical composition on metabolisable energy of by-products of vegetable oilrefining. British Poultry Science 37, 131–144.

Ward, A.T. and Marquardt, R.R. (1983) The effect of saturation, chain length of puretriglycerides, and age of bird on the utilization of rye diets. Poultry Science 62,1054–1062.

Westergaard, H. and Dietschy, J.M. (1976) The mechanism whereby bile acid micellesincrease the rate of fatty acid and cholesterol uptake into the intestinal mucosal cell.The Journal of Clinical Investigation 58, 97–108.

Wiseman, J. and Lessire, M. (1987) Interactions between fats of differing chemical content:apparent metabolisable energy values and apparent fat availability. British Poultry Science28, 663–676.

Wiseman, J. and Salvador, F. (1991) The influence of free fatty acid content and degree ofsaturation on the apparent metabolizable energy value of fats fed to broilers. PoultryScience 70, 573–582.

Wiseman, J., Cole, D.J.A., Perry, F.G., Vernon, B.G. and Cooke, B.C. (1986) Apparentmetabolisable energy values of fats for broiler chicks. British Poultry Science 27, 561–576.

Wiseman, J., Salvador, F. and Craigon, J. (1991) Prediction of the apparent metabolizableenergy content of fats fed to broiler chickens. Poultry Science 70, 1527–1533.

Young, R.J. (1961) The energy value of fats and fatty acids for chicks: 1. Metabolizable energy.Poultry Science 40, 1225–1233.

236 S. Dänicke




E.T. KornegayRole of Phytases in Nutrient Digestion10

Introduction

Phytase may well be the miracle enzyme of the 1990s, as soybeans were describedas the miracle crop for producing high quality plant protein in the 1940s. Dietaryaddition of microbial phytase or the inclusion of high phytase ingredients in pig,poultry and fish diets is now well documented to release a large portion of thenaturally occurring phytate P, and thus greatly reduce the amount of inorganic P thatmust be added to meet the animal’s P requirement. The net result is a reduction in Pexcretion that can range from 20 to 50%.

Microbial phytase was initially used as a tool to reduce P excretion becausemodern commercial production of pigs and poultry had led to large amounts ofmanure which, when applied to the land in excess, resulted in a build-up of nutrientsin and on the soil. This had a potential for environmental pollution that continues tolead to legislation in many countries that require nutrient management plans formanure from these farms.

It is becoming increasingly clear that the use of adequate amounts of phytase inmost pig and poultry diets results in improved availability of Ca, Zn, protein/aminoacids (AA) and energy. This is because seeds or products from seeds, which are themajor ingredients in pig and poultry diets, contain 60–80% of the P in the form ofphytic acid or phytate. Phytate is known to complex with other nutrients. Thus, theunavailable phytate P and nutrients complexed with it cannot be utilized and areexcreted.

This chapter will provide a brief review of phytate, phytase, and the effectivenessof microbial phytase in pig, poultry and fish diets for enhancing the utilization of P,


Digestion of Phosphorus andOther Nutrients: the Roleof Phytases and Factors

Influencing Their ActivityE.T. KORNEGAY*

Department of Animal and Poultry Sciences,Virginia Polytechnic Institute and State University,

3060 Litton-Reaves Hall, Blacksburg, VA 24061, USA

10

*Deceased.

Ca, Zn, amino acids/protein and energy. Factors that influence phytase activity willbe discussed.

Phytate

Occurrence and structure of phytate

The major ingredients in commercial pig and poultry diets are seeds (cereal grains) orproducts from seeds (oilseed meal and grain by-products). A large portion (60–80%)of the P in these ingredients occurs in the form of phytates, the salts of phytic acid(Table 10.1). Phytic acid, myo-inositol 1, 2, 3, 4, 5, 6 hexakis dihydrogen phosphate,is an essential component of all seeds. Seeds rapidly accumulate phytic acid duringthe ripening period (Asada et al., 1969). The location of phytate in the seed varies(Pallauf and Rimbach, 1995). In small grains, phytate lies mainly in the bran(aleurone layer, testa and pericarp), and in the case of maize it is found mainly inthe germ. Generally, with legume seeds, phytate accumulates in the cotyledon. Insoybeans, phytic acid is located in protein bodies distributed throughout the seed(Baker, 1991). Rapeseed phytates are more concentrated in seed tissues than soybeanphytates, which appear to have no specific site location (Kratzer and Vohra, 1986).Detailed information on the phytic acid content of various foods and feedstuffs canbe found in the reviews by Oberleas and Harland (1981), Ingelmann et al. (1993),Eeckhout and DePaepe (1994) and Ravindran et al. (1994, 1995b).

238 E.T. Kornegay

IngredientPhytate Pa

(%)Phytate Pa

(% of total P)Phytase activityb

(units kg−1)

Cereals and by-productsMaizeWheatSorghumBarleyOatsWheat bran

Oilseed mealsSoybean mealCanola mealSunflower mealGroundnut mealCottonseed meal

0.240.270.240.270.290.92

0.390.700.890.480.84

726966646771

6059778070

151193

24582

402957

81660

3NA

aData adapted from Ravindran (1996) and Ravindran et al. (1994, 1995b).bData from Eeckhout and De Paepe (1994). One unit is defined as that amount of phytase whichliberates inorganic phosphorus from a 5.1 mM Na-phytate solution at a rate of 1 µmol min−1 at pH5.5 and 37°C (98.6°F).

Table 10.1. Phytate phosphorus content and phytase activity of some common feed ingredients.

Bioavailability of phytate P

The bioavailability of P in cereal grains and products from oilseeds is generally verylow for pigs and poultry, because they have limited capability to utilize phytate P(Table 10.2). Bioavailability estimates of P in maize and soybean meal for pigsand poultry range from 10 to 30% (Nelson, 1967; Calvert et al., 1978; Jongbloedand Kemme, 1990; Cromwell, 1992). This low availability of phytate P poses twoproblems for producers: (i) the need to add inorganic P supplements to diets; and(ii) the excretion of large amounts of P in the manure.

Potential for binding nutrients

The phytic acid molecule has a high P content (28.2%) and chelating potential(Fig. 10.1) to form a wide variety of insoluble salts with di- and trivalent cations atneutral pH (Vohra et al., 1965; Oberleas, 1973; Cheryan, 1980). One mole of phyticacid can bind an average of 3–6 mol of Ca to form insoluble phytates at the pH ofthe small intestine. Formation of insoluble phytate makes both Ca and P unavailable.Zn, Cu, Co, Mn, Fe and Mg can also be complexed, but Zn and Cu have thestrongest binding affinity (Maddaiah et al., 1964; Vohra et al., 1965). This bindingpotentially renders these minerals unavailable for intestinal absorption. Zinc may bethe trace element whose bioavailability is most influenced by phytate (Pallauf andRimbach, 1995). Phytic acid may have a negative influence on dietary proteinand amino acids (O’Dell and de Borland, 1976; Knuckles et al., 1985) and inhibitsproteolytic enzymes such as pepsin and trypsin under gastrointestinal conditions

Role of Phytases in Nutrient Digestion 239

FeedstuffBioavailability of P for pigsa,b

(%)Nonphytate-P for poultryc

(% of total)

Cereal grainsMaizeOatsBarleyTriticaleWheatMaize, high moisture

High protein meals – plant originGroundnut mealCanola mealSoybean meal, dehulledSoybean meal, 44% protein

122331465053

12212535

2833363331–

21263540

aAdapted from Cromwell (1992).bRelative to the availability of phosphorus in monosodium phosphate, which is given a value of 100.cNRC (1994) – poultry.

Table 10.2. Bioavailability of phosphorus for pigs and nonphytate phosphorus for poultry.

(Singh and Krikorian, 1982). Under acidic conditions, the basic phosphate groupsof phytic acid may complex with amino groups such as lysyl, histidyl and arginine(De Rham and Jost, 1979; Fretzdorff et al., 1995). Under neutral conditions, thecarboxyl groups of some amino acids may bind to phytate through a divalent ortrivalent mineral. Phytate–protein or phytate–mineral–protein complexes mayreduce the utilization of protein.

Starch is also known to be complexed by phytate. The in vitro hydrolysis ofeither wheat or bean starch incubated with human saliva was retarded when Naphytate was included in the mixture, but digestion was restored when Ca was addedwith the Na phytate (Yoon et al., 1983; Thompson, 1986; Thompson et al., 1987).As will be reviewed later, Ravindran (1999) reported studies with broilers that clearlyshow that apparent metabolizable energy is improved when high-phytate diets aresupplemented with microbial phytase.

Phytases

Phytases are known to occur widely in microorganisms, plants and certain animaltissues (Nayini and Markakis, 1986; Nys et al., 1996). Phytase of microbial origin(3-phytase, EC 3.1.3.8) hydrolyses the phosphate group at the C3 position first,whereas phytase of plant origin (6-phytase, EC 3.1.3.26) acts first at the C6 position

240 E.T. Kornegay

Fig. 10.1. Structure of phytate and possible bonds (after Thompson, 1988).

(Pallauf and Rimbach, 1995). Phytase produced by Aspergillus has two pH optima:one at pH 2.5 and the other at pH 5.5. Wheat phytase has only one pH optimum: at5.2. Aspergillus phytase has been shown to be more effective per unit of activity thanwheat phytase, probably due to these differences (Eeckhout and De Paepe, 1996).Phytase activity is generally defined as follows: one unit of phytase activity is theamount of enzyme that liberates 1 µmol of inorganic phosphorus in 1 minfrom 5.1 mmol solution of sodium phytate at 37°C and pH 5.5. At least threeabbreviations are used in the literature for phytase activity: FTU, PU and U. Thelatter (U kg−1) will be used in this chapter unless otherwise noted.

Four possible sources of phytase could breakdown phytate within the digestivetract of pigs and poultry: (i) intestinal phytase in digestive secretions; (ii) endogenousphytase present in some feed ingredients; (iii) phytase originating from residentbacteria; or (iv) phytase produced by exogenous microorganisms (Nys et al., 1996).Contents of the stomach and intestine of pigs (Jongbloed et al., 1992; Yi andKornegay, 1996) and crop, stomach and small intestine of chickens (Liebert et al.,1993) have negligible phytase activity. Although endogenous phytase activity ofintestinal mucosae has been reported for pigs (Pointillart, 1993) and poultry (Maenzand Classen, 1998), its contribution toward dietary phytate P digestibility is notknown. A higher total of small intestinal brush-border phytase activity was reportedfor 50-week-old hens compared with 4-week-old broilers, but the specific activity perunit of brush-border tissue was similar for broilers and mature hens. Phytase activityhas been detected in some animal tissue (McCollum and Hart, 1908; Bitar andReinhold, 1972; Pointillart et al., 1985; Yang et al., 1991); however, this activity, ifpresent, is negligible for improving the availability of phytate P in non-ruminantanimals (Pallauf and Rimbach, 1995). Additionally, the significance of phytase pro-duced by resident bacteria in non-ruminants has not been demonstrated and is prob-ably negligible, considering the lack of significant absorption in the large intestines.

Plant phytase

It has been known for more than 50 years that plant phytase has the ability tohydrolyse phytate (McCance and Widdowson, 1944; Hill and Tyler, 1954) and itseffectiveness for improving P digesitibility in pigs and poultry has been clearly shown(Nelson, 1967; Newton et al., 1983; Bagheri and Gueguen, 1985). Phytase activityhas been reported in a wide range of seeds, such as rice, wheat, barley, maize, rye,soybean and oil seeds (Reddy et al., 1982; Gibson and Ullah, 1990; Eeckhout and DePaepe, 1994); however, phytase activity of seeds varies greatly among species of plants(Table 10.1). With the exception of wheat, rye and triticale, most dormant seedscontain very low phytase activity. Phytase activity in maize and soybean meal is solow that it is not of practical importance. Barley may or may not contain nutrition-ally significant levels of phytase activity (Pointillart, 1993). The majority of thephytase activity in wheat, rye and triticale is in the bran. Diets formulated usingingredients having high phytase activity, such as wheat bran, wheat, triticale, rye branand wheat middlings, promote greater absorption of phytate P (Pointillart, 1991,


1993). Eeckhout and De Paepe (1991) reported that microbial phytase was 74%more efficient in vivo (fed to pigs) than phytase in wheat middlings when added at anequal in vitro activity level (500 U kg−1). They showed that genetically modifiedmicrobial phytase was active over a wider pH range than wheat middlings phytase.They suggest that microbial phytase may be more active at pH levels present in thestomach than wheat phytase.

Although some feed ingredients contain native phytase activity, steam-pelletingused in the manufacture of many commercial pig and poultry feeds results insignificant losses of this intrinsic phytase activity. Because of variation of phytaseactivity among and within plant species, damaging effects of pelleting during feedmanufacturing and the lack of availability of feed ingredients of high phytase activity,the presence of residual phytase activity often may not be considered in dietformulation when feeds are pelleted.

Microbial phytase

Microbial phytases are found in numerous bacteria, yeast and fungi (Harlandand Morris, 1995) but they have been detected most frequently in media with theAspergillus genus of ascomycetous fungi (Irving and Cosgrove, 1974; Nair andDuvnjak, 1991). Hydrolysis of dietary phytate by extrinsic microbial phytase wasprobably investigated first by Nelson et al. (1971) who observed improved Putilization by chicks fed maize–soybean meal diets containing preparations ofAspergillus. However, the costs of adding the enzyme were very high and there waslittle environmental pressure to reduce the P excretion until the late 1980s.

Two major factors led to the development of commercial phytase that couldbe economically used in pig and poultry diets. The first was the advance inbiotechnology that led to techniques for genetically modifying fungi; second wasmandate that Dutch livestock farmers had to reduce P excretion. Certainly, the factthat the phytase gene had been identified and isolated made possible the geneticmodification of A. niger. Also, an improvement in fermentation technology was acontributing factor.

Sources of phytase

Currently, there are at least four commercially available microbial phytases, twoobtained by fermentation of a genetically modified Aspergillus (Natuphos® andNovo phytase), and two obtained by extraction of media with Aspergillus (Finase™and Alltech phytase). Patents influence the availability of these in some countries.Most products are available in a powder or granular form and as a liquid. The geneticinformation for both Natuphos® and Novo phytase originated from A. ficuum andwas transferred to either A. niger or A. oryzae.

Research is also underway expressing the A. niger phytase gene in seeds fromtobacco and rapeseed (Beudeker and Pen, 1995). Zhang et al. (1998a,b) reported

242 E.T. Kornegay

that Phytaseed® (rapeseed genetically modified with the A. ficuum phytase gene)and Natuphos® were equally effective in low P pig and broiler diets for improving Putilization. Similar findings were reported for turkeys by Ledoux et al. (1998). Otherproducts may be under development. Van Loon et al. (1998) reported fromshort-term pig trials that a heat-stable phytase enzyme, cloned from A. fumigatus andoverexpressed in A.niger (Pasamontes et al., 1997), was more effective in reducingP excretion than a commercially available genetically modified phytase using theA. ficuum phytase gene. The A. fumigatus phytase was reported to have retained morethan 90% of its enzymatic activity after 20 min heating at 90°C.

Safety of genetically modified phytase

Genetically modified microbial and plant phytases were fed to pigs and poultry atlevels 5–20-fold higher than recommended levels (500 U kg−1). In a piglet (5 weeks)and a broiler (4 weeks) trial, the performance, P retention coefficients and bonemineralization values continued to increase (at a decreasing rate) to 2500 U kg−1, thehighest level fed in these two trials (Zhang et al., 1998a,b). General necropsy andhistological examination of liver, kidney and tibial tissues revealed no adverse effects.In trials for both a pig (29 kg initially for 6 weeks) (Harper et al., 1999) and a turkey(4 weeks after 7 days adjustment) (Kornegay et al., 1999), there were no differencesin mortality among phosphorus or microbial phytase treatments. It was evident thatthe addition of 10,000 U phytase kg−1 did not cause any negative effects based ona general necropsy and histological examination of liver, kidney and tibial bonesbut, rather, continued to provide small positive improvements in performance, bonemineralization and digestibility of P.

Site of phytase activity

Results from several post-slaughter and cannulation experiments with pigs haveshown that dietary phytase activity, whether from plants (Gueguen et al., 1968;Schulz and Oslage, 1972; Lantzsch et al., 1992; Kemme et al., 1998) or fungalsources (Jongbloed et al., 1992; Mroz et al., 1994; Yi and Kornegay, 1996), waspredominantly active in the stomach. In pigs, phytase activity or breakdown ofphytate P was generally very low or was not observed in the lower small intestine.Significant amounts of phytate P (40–50% of the total) would be passed into thelarge intestine, where most of it would be hydrolysed, but little of the hydrolysedP would be absorbed (Jongbloed et al., 1992; Lantzsch et al., 1992; Kemme et al.,1998). The low phytase activity in the small intestine may be due to a much morebasic pH (6.5–7.6) of the jejunum and ileum (van der Meulen and Bakker, 1991)that is less favourable for high phytase activity, or due to proteolytic enzymes in thesmall intestine that may degrade phytase.

Liebert et al. (1993) reported in chickens that 69–86% of added microbialphytase activity was detected in the crop and that 31–38% of added phytase activity


was detected in the proventriculus. No phytase activity was detected in the smallintestine. The disappearance of phytate P in the crop and proventriculus supports theobservation that the crop and proventriculus are the main sites of phytase activity inpoultry.

Effectiveness of Microbial Phytase in Poultry and Swine Diets

Supplemental microbial phytase is well known for its effectiveness in improving theavailability of P from plant ingredients containing high levels of phytate P. Somereports have also suggested that the availabilities of Ca, Zn, amino acids and, in somecases, energy are also improved when phytase is supplemented in the diet. However,the use of microbial phytase in poultry and swine diets will depend upon a number offactors, including the cost of P and phytase, the overall effectiveness of phytase torelease P and other nutrients, and the environmental need to reduce the excretion ofP and other elements which translate into disposal costs. The remainder of thischapter will discuss the use of microbial phytase in pig and poultry diets to enhancethe utilization of P, Ca, Zn, amino acids and energy so that excretion of thesenutrients can be reduced. Factors to consider that can influence the efficiency ofmicrobial phytase in these diets will be discussed.

Effects of phytase on improving the bioavailability of phosphorus

Phosphorus digestibility response equations were generated using 52 pig experimentsrepresenting 32 references in a review reported by Kornegay et al. (1998a). Also, 23poultry experiments representing 13 references were used to generate P retentionequations (Kornegay, 1999). Studies that had very high dietary P levels (NRC orabove) for the negative control diets were excluded. Because of limited data, noattempt was made to separate and evaluate data by commercial inorganic phosphorussources. Several equations were generated for the pig and poultry data sets with theupper levels of phytase varying from 600 to 800, 1000, 1200 or 1500 U kg−1 of diet.An evaluation was made of estimates of digestibility (or retention) of P calculatedusing each of these equations. Within the range of 100–600 U kg−1 for pigs and200–800 U kg−1 for poultry, similar values were generally obtained for the variousequations. It was decided to use the wider range of data, but equations for the upperlevel at 600 or 800 U kg−1 are also given in the figures. Unless otherwise noted, alldigestibility or retention coefficients are apparent.

Phosphorus digestibilityA nonlinear response of supplemental phytase on P digestibility for pigs and P reten-tion for poultry was observed (Fig. 10.2). The nonlinear response of phytase on Pdigestibility is in agreement with a report by Beers (1992), who found an exponentialand logistic relationship between the dosage of microbial phytase and apparent P

244 E.T. Kornegay

digestibility, and other reports where multiple levels of phytase were fed (Denbowet al., 1995; Ravindran et al., 1995a; Kornegay and Qian, 1996; Yi et al., 1996a).

Our response curve was remarkably similar to the nonlinear response curve(corrected for the digestibility of the negative controls) of added phytase on


Fig. 10.2. Increase of P digestibility of pigs (top) and poultry (bottom) fed low P,low phytase activity plant-based diets supplemented with microbial phytase.

digestibility of P reported by Dungelhoef and Rodehutscord (1995) for pigs[Y = 28.1(1 − e−0.0024X), r2 = 0.38]. For example, in their data set there was a 19.6%unit increase in digestibility of P when 500 U phytase kg−1 was added, and a value of19.7% was obtained from our pig data set (Table 10.3).

The magnitude of the response per unit of phytase was much greater at the lowerphytase levels. For example, in pigs with 400–500 U phytase kg−1 added, there was a2.17% unit increase in P digestibility; whereas for 900–1000 U kg−1 there was only a0.58% unit increase.

In poultry, the increase in P retention was 7.8% units for 500 U phytase kg−1,less than one-half that observed for pigs (19.7%). Similar to pigs, the magnitude ofthe response per unit of phytase was greater at the lower phytase levels. For example,for 400–500 U phytase kg−1 there was a 1.18% unit increase; whereas for900–1000 U kg−1 there was only a 0.60% unit increase. This relationship of adecrease in the magnitude of the response to phytase as the amount of phytase addedincreased was previously described in pigs (Kornegay and Qian, 1996; Yi et al.,1996a), broilers (Denbow et al., 1995; Kornegay et al., 1996a) and turkeys(Ravindran et al., 1995a) for growth and bone mineralization.

Digested phosphorusDigested P resulting from microbial phytase supplementation was calculated by mul-tiplying the total P content of the low P diet (negative control) by the increase in Pdigestibility or P retention resulting from phytase supplementation. The data areshown in Table 10.3 for pigs and Table 10.4 for poultry.

For pigs at a dietary phytase level of 500 U kg−1, 0.075% units (0.75 g) of Pwere digested. In the pig data set, the average total dietary P level of the low P diet(negative control) was 0.381 ± 0.008% (mean ± SEM) and the Ca level was0.588 ± 0.020. The average ratio of total Ca to total P was 1.54.

For poultry, 0.037% units (0.37 g) of P were estimated to be digested when500 U phytase kg−1 was added to the diet. The average total P level of the low P diet(negative control) was 0.478 ± 0.010% (mean ± SEM) and the Ca level was0.780 ± 0.036%. The average ratio of total Ca to total P was 1.63. Although total Plevels fed are higher and retention coefficients are higher for negative control diets forpoultry than for pigs, the increase of retention coefficients due to microbialsupplementation was lower for poultry than the increase of digestibility coefficientsfor pigs. As will be discussed later, the percentage utilization of inorganic phosphateappears to be less for poultry than for pigs.

Excretion of phosphorus

Based on calculations using the digestibility equation from the pig data set, Pexcretion can be reduced by 33.2% when 500 U phytase kg−1 is added to a low Pdiet, compared with a positive control diet (0.481% P) containing 0.1% units moreP (Table 10.3). When the P retention equation from the poultry data set was used tomake similar estimates, P excretion was reduced by 31.9% when 500 U phytase kg−1

246 E.T. Kornegay

is added to a low P diet, compared with a positive control diet (0.57% dietary P)which would be 0.1% units higher in P (Table 10.4). Simply lowering the level ofdietary P by 0.1% will decrease P excretion by about 8.3% in pigs and 18.4% inpoultry. Unless very high levels of P (well above the minimum requirement of the


Supplementalphytase(U kg−1)

Total Pdigestibilitya

(%)

P digestibilityb

by phytase(%)

P digestedby phytasec

(%)

Total Pexcretedd

(%)

Decreased Pexcretione

(%)

0100200250300350400450500550600650700750800900

100011001200130014001500

27.934.238.940.942.644.145.546.647.648.549.350.050.651.151.652.352.953.453.754.054.254.3

06.2

11.013.014.716.217.518.719.720.621.422.122.723.223.624.425.025.425.826.026.226.4

00.0240.0420.0490.0560.0620.0670.0710.0750.0780.0810.0840.0860.0880.0900.0930.0950.0970.0980.0990.1000.101

0.2750.2510.2330.2250.2190.2130.2080.2030.2000.1960.1930.1910.1880.1860.1840.1820.1790.1780.1760.1750.1750.174

8.316.322.324.827.028.930.632.133.234.535.536.437.137.838.439.440.140.741.141.541.741.9

aGenerated from the equation given in Fig. 10.2 [54.86(1 − 0.4908e−0.00263X)].bCalculated by subtracting P digestibility of basal diet without phytase from coefficients at eachphytase level.cCalculated by multiplying the P digestibility due to phytase by the average P content (0.381% totalP) of the basal diet. The equation of this data is = 0.1026(1 − e−0.00263X), where X = phytase level.dCalculated by subtracting total P digestibility coefficients from 100 and multiplying the product bythe average P content (0.381% total P) of the basal diet.eBased on an inorganic P digested equation (Y = −0.167 + 0.755X, r2 = 0.24, where X = % P frominorganic source); 0.0245% P was excreted for the 0.1% unit of added inorganic P above the basaldiet making the total P excreted by the positive control diet (0.481% total P) equal to 0.2995% P(0.275 + 0.0245). Decreased P excretion was calculated by subtracting the total P excreted by thephytase supplemented diets from the P excreted by the positive control diet (0.2995%) and thendividing by the P excreted by positive control diet and multiplying by 100. For example, at500 U phytase kg−1, 33.2% = [(0.2995 − 0.200)/(0.2995)]*100.

Table 10.3. Predicted P digestibility, P digested, and percentage reduction in P excretion basedon data generated from the pig data set in this study.

pig) are fed, the amount of P excreted in pig urine will generally be minimal and wasnot considered in the calculations in this chapter; this is consistent with conclusionsreached by Jongbloed (1987).

248 E.T. Kornegay

Supplementalphytase(U kg−1)

Total Pdigestibilitya

(%)

P digestibilityb

by phytase(%)

P digestedby phytasec

(%)

Total Pexcretedd

(%)

Decreased Pexcretione

(%)

0100200250300350400450500550600650700750800900

100011001200130014001500

52.154.155.956.757.458.158.859.460.060.561.061.561.962.362.763.464.364.765.065.465.766.0

0.02.03.84.65.36.06.77.37.88.48.99.39.8

10.210.611.211.812.412.813.213.613.9

0.0100.0100.0180.0220.0250.0290.0320.0350.0370.0400.0420.0450.0470.0490.0500.0540.0570.0590.0610.0630.0650.066

0.2290.2190.2110.2070.2030.2000.1970.1940.1910.1890.1860.1840.1820.1801.1780.1750.1720.1700.1680.1660.1640.162

18.421.824.826.227.528.629.730.831.932.733.534.335.035.736.437.638.639.540.341.041.642.1

aGenerated from the equation given in Fig. 10.1 [68.18(1 − 0.2354e−0.00134X)].bCalculated by subtracting P digestibility of basal diet without phytase from coefficients at eachphytase level.cCalculated by multiplying the P digestibility due to phytase by the average P content (0.478% totalP) of the basal diet. The equation of this data (%) = 0.07672(1 − e−0.00134X), where X = phytase level.dCalculated by subtracting total P digestibility coefficients from 100 and multiplying the product bythe average P content (0.478% total P) of the basal diet.eBased on an inorganic P digested (retained) equation (Y = −0.000936 + 0.484X, r2 = 0.69, whereX = % P from inorganic source); 0.0516% was excreted for 0.1% added inorganic P above thebasal diet making the total P excreted by the positive control diet (0.578% total P) equal to0.2804% P (0.2288 + 0.0516). Decreased P excretion was calculated by subtracting the total Pexcreted by the phytase supplemented diets from the P excreted by the positive control diet(0.2804%) and then dividing by the P excreted by positive control diet and multiplying by 100.For example, at 500 U phytase kg−1, 31.9% =[(0.2804 − 0.191)/(0.2804)]*100.

Table 10.4. Predicted P digestibility, P digested, and percentage reduction in P excretion basedon data generated from the poultry data set in this study.

Digested P vs. P equivalency value of phytase

Phosphorus equivalency value, a term used to describe the replacement orsubstitution value of phytase, is defined as the amount of inorganic P that can beremoved by a given amount of added or intrinsic phytase. For direct comparison ofequivalency value of phytase for P and digested P, equivalency values must beadjusted by the estimated digestibility of the inorganic P sources that phytasereplaces. Kornegay et al. (1998a) estimated that the digestibility of P in severalfeed-grade P sources was 76.7% for pigs. The retention of P from several feed-gradeP sources was estimated to be 46.2% for broilers and turkeys (Kornegay, 1999).Equivalency values (or equations) are usually obtained from non-linear or linearequations generated from body-weight gain, bone mineralization, and sometimesdigested P data obtained by feeding multiple levels of P without phytase addition andmultiple levels of added phytase to a low P diet; these equations are set equaland solved. This procedure was described in detailed by Denbow et al. (1995).

Using a small data set of four pig studies from Virginia Tech, Radcliffe et al.(1998a) reported an average P equivalency value of 1.0 g inorganic P for 500 Uphytase kg−1 based on average daily gain, bone mineralization and P digestibility.Using growing and finishing pig data reported by Shih and Hsu (1997), Veum(1996), O’Quinn et al. (1997) and Harper and Kornegay (1997), non-linear andlinear equations were generated for the response to phytase, and the average Pequivalency value of 500 U phytase kg−1 was 1.0 g inorganic P.

For example, if the average equivalency value of 1.0 g P is multiplied by 76.7%(0.767 g P digested per 1 g of inorganic P fed), the product, 0.767 g P (1 × 0.767) issimilar to a digested P value of 0.75 g (0.075%) obtained from the pig data set shownin Table 10.3. Remember that equivalency values must be adjusted by the apparentdigestibility of the inorganic P source phytase that is being replaced. Jongbloed et al.(1996a) estimated that 500 U phytase kg−1 was equivalent to 0.8 g digestible P,which was equivalent to 1.0 g P from monocalcium phosphate.

Yi and Kornegay (Virginia Polytechnic Institute and State University, 1998,unpublished data) developed and organized equivalency values from severalpublished broiler studies. For broilers, usually from hatch to 3 weeks of age, theaverage mean P equivalency of 500 U phytase kg−1 was 0.70 g inorganic P based onbody-weight (BW) gain and toe ash percentage. In a 3-week starter turkey trial, theaverage P equivalency of 500 U phytase kg−1 was 0.82 g inorganic P based on BWgain and toe ash. If the 0.70 g P and 0.82 g P are multiplied by 46.2% (0.462 g Pretained per 1 g inorganic P), the products are 0.32 g P (0.7 × 0.462) and 0.38 g P(0.082 × 0.462). A value of 0.37 g retained P (0.037%) was obtained in the poultrydata set for 500 U phytase kg−1 (Table 10.4).

Higher P equivalency values for BW gain and tibia ash percentage were reportedby Kornegay et al. (1998a) for growing–finishing broilers: 1.0 and 0.9 g P,respectively, for 400 U phytase kg−1, the highest level of phytase fed in this study.However, the calculated digested P in this study was only slightly more (0.036 vs.0.032 g P) than the calculated digested P value obtained for 400 U phytase kg−1

from the poultry data set (Table 10.4).


Although studies have reported that the digestibility of inorganic P sourcesvary, our research results (Kornegay and Radcliffe, 1996; Kornegay et al., 1996b)with young pigs and turkeys suggest little to no difference in the bioavailabilityof monocalcium/dicalcium, dicalcium/monocalcium and defluorinated phosphatefrom ‘good quality’ commercial sources in the United States. Several commercialfeed phosphate sources were used in the studies included in our pig and poultrydata sets shown in Tables 10.3 and 10.4, but calculations could not be made for thedifferent P sources, because of inadequate data and variation observed.

Based on the similarity of digested P values calculated from P equivalencyestimates, and digested P values derived from equations generated in the pig andpoultry data sets, the estimates of P excretion calculated from these data sets shouldbe accurate for a range of situations. However, a larger response than observed inthese data sets is possible if careful attention is given to ingredient composition anddiet formulation (optimal Ca and P levels), and if quality processing procedures arefollowed, as will be discussed later.

Response of layers to phytase supplementation

In a series of trials reported by Simons et al. (1992), Peter and Jeroch (1993), Simonsand Versteegh (1993), Vahl et al. (1993), Van der Klis et al. (1997) and Gordon andRoland (1997), phytase supplementation of a low P diet for layers was very effectiveas a replacement for inorganic P. A range of P equivalencies, 0.5–1.2 g P as mono-calcium phosphate, have been reported for 200–300 U phytase kg−1. Van der Kliset al. (1997) reported that the effect of phytase supplementation (250 and 500 Ukg−1) on ileal P absorption was 12% units greater when added to a low P basal dietcontaining 3.0% Ca compared with a low P basal diet containing 4.0% Ca. Leskeand Coon (1998) reported that phytate P retention was 36.7, 29.0 and 14.8% unitsgreater with phytase supplementation (300 U kg−1 of diet), respectively, for soybeanmeal, maize and rice bran, but total P retention was only 16.6, 16.1 and 7.1% unitsgreater. As has been reported for pigs, broilers and turkeys, the response of phytasesupplementation was greater at the lower levels of total P. Phytase supplementation isvery effective at releasing P in layer diets, which can result in reduced dietary P levelsand reduced P excretion. The efficiency appears to be greater for layers than forbroilers and turkeys. Phytase supplementation of layer diets is also simplified becausemost diets are fed in a mash form and thus heat stability issues are averted.

Factors influencing phytase response

The dietary level of P influences the response to phytase (Dungelhoef andRodehutscord, 1995; Kornegay and Qian, 1996; Kornegay, 1996a,b). Results ofstudies in pigs (Kornegay and Qian, 1996; Yi et al., 1996a), broilers (Denbow et al.,1995; Kornegay et al., 1996a) and turkeys (Ravindran et al., 1995a) clearlydemonstrate that the response to microbial phytase is influenced by dietary levels of

250 E.T. Kornegay

available or non-phytate P, and that digested P per unit of phytase decreases as theamount of phytase increases per unit of diet. Extremely high and low levels of dietaryP should be avoided, to optimize the response to phytase.

Vitamin D probably influences phytase activity indirectly by increasing Caabsorption, therefore limiting the formation of insoluble Ca-phytates, which aremore resistant for phytase hydrolysis (Pointillart, 1993). High levels of Ca or wideCa : P ratios will reduce the response for phytase (discussed later).

Effects of phytase on calcium bioavailability

Broilers and turkeysSchoner et al. (1991, 1993, 1994) reported improved Ca retention in broilers fedsupplemental phytase. In a broiler study designed to measure the effect of phytase onCa availability, Schoner et al. (1994) reported that 500 U microbial phytase wereequivalent to 0.46 g Ca based on BW gain and phalanx ash. Calcium retention andDM digestibility were improved when phytase was added to broiler diets (Kornegayet al., 1996a, 1998c; Yi et al., 1996b). In our study with turkey poults (Kornegayet al., 1996c), estimates based on BW gain, gain : feed and digested Ca suggest that500 U phytase are equivalent to 0.87 g Ca. Both P and Ca retention were sensitiveto the addition of phytase at two non-phytate (nP) levels and four Ca : tP ratiosin turkeys (Qian et al., 1996b), or two vitamin D3 levels and four Ca : tP ratios inbroilers (Qian et al., 1997). Calcium retention, as well as P retention, linearlyincreased as the amount of supplemented phytase increased, and decreased as theCa : tP ratios became wider in both broilers and turkeys.

PigsRadcliffe et al. (1995) reported results of two pig trials conducted to determine theeffectiveness of microbial phytase for improving the bioavailability of Ca. Based onthis data, Kornegay et al. (1996c) estimated Ca equivalency values of 1.08 and 0.38 gCa per 500 U of microbial phytase for pigs in Trials 1 and 2, respectively, with anaverage of 0.73 g Ca. Calcium equivalency estimates for phytase were based on dailygain during weeks 3 and 4, digestible Ca and tenth-rib ash percentages.

The interaction between dietary Ca and microbial phytase were reported byJongbloed et al. (1996c) in two trials. In both trials, the apparent total tract digestibil-ity of P linearly decreased with increasing Ca level. No significant interaction couldbe demonstrated between dietary Ca level and microbial phytase. Microbial phytaseenhanced not only the apparent total tract digestibility of P, but also the apparenttotal tract digestibility of Ca. They estimated that an extra 0.8 g of P and between 0.4and 0.7 g of Ca were absorbed with 500 U supplemental phytase kg−1.

Using primarily cereal phytase, Pointillart (1993) reported that improved Putilization was generally accompanied by improved Ca retention. Simecek et al.(1995) also reported that P as well as Ca digestibility was improved when two formsof microbial phytase were supplemented in a barley–soybean meal diet for growingpigs. In other studies, the apparent absorption of Ca and N (Kornegay and Qian,


1996; Yi et al., 1996a) in pigs was improved when phytase was added to the diet.Eeckhout and De Paepe (1991) reported a highly positive correlation between Caand P digestibilities and phytase supplementation to a low P pig diet. They suggestedthat this relationship could be explained by the fact that phytic acid acts as aCa-binding agent in the proximal small intestine. Hydrolysis of phytate in thestomach as a result of phytase activity results in increased digestibility, not only of Pbut, indirectly, of Ca.

Influence of phytase on zinc bioavailability

Zinc is probably the most vulnerable mineral to phytate complexation. In earlyresearch in chicks, swine, rats and other species it was observed that dietary Znrequirement was affected by dietary phytate (Prasad, 1966). Oberleas and Harland(1996) suggest that it is imperative that much of the endogenously secreted mineralsneeds to be reabsorbed lest perpetual deficiencies result. Oberleas and Harland(1996) reported results that clearly showed that phytate was a significant factor in thedevelopment of zinc deficiency. The presence of Ca aggravated the effect of phytateon zinc utilization, probably through an insoluble Ca–phytate–Zn complex that isformed which prevents the absorption of Zn.

Broilers and turkeysThe addition of 800 U phytase kg−1 to a diet containing 27 mg Zn kg−1 increasedthe retention of Zn and decreased Zn excretion of chicks (Thiel and Weigand, 1992).Thiel et al. (1993) reported that the femoral Zn content of chicks fed a dietcontaining 30 mg Zn kg−1 plus 700 U phytase kg−1 of diet was equal to that of chicksfed a diet containing 39 mg Zn kg−1 without phytase. Using chicks fed a glucose–soybean concentrate diet (13 mg Zn kg−1 diet) with multiple levels of added Zn,phytase or 1,25-dihydroxycholecalciferol (di-OH-D3), Biehl et al. (1995) reportedthat phytase and di-OH-D3 supplementation both increased growth rate and tibialZn to a similar extent. Based on Biehl et al. (1995) estimates using tibia Zn, the Znequivalency of 600 and 1200 U phytase kg−1 was 3.8 and 5.5 mg, respectively. Incontrast, Roberson and Edward (1994) did not consistently observe an improvementin Zn absorption or retention in broilers when 600–750 U phytase kg−1 was addedto a maize–soybean diet containing 32 mg Zn kg−1 diet.

In our laboratory, day-old male broilers were fed a maize–soybean isolate basaldiet containing 20 mg Zn kg−1 alone and supplemented with multiple levels of Znand phytase for 21 days (Yi et al., 1996d). Non-linear or linear response equations ofthe effects of Zn and phytase levels were generated for BW gain, feed intake, Znretention, Zn concentration of toes, tibia and liver and used to calculate an averageZn equivalency of 5.4 mg kg−1 for 600 U phytase kg−1.

PigsBased on enhanced growth, increased plasma Zn concentration and alkalinephosphatase activity, the bioavailability of Zn for pigs was improved when phytase

252 E.T. Kornegay

was added to a low P (0.3%) and Zn (30 mg kg−1) maize–soybean meal diet (Leiet al., 1993). Adding microbial phytase to the diets of young pigs significantlyimproved apparent absorption of Zn and Mg (Pallauf et al., 1992; Nasi andHelander, 1994). Using a low Zn diet (23 mg kg−1) containing 0.8% Ca and 0.62%P, Adeola et al. (1995) reported that growth rate and the retention of Zn, Cu, Ca andP were increased when 1500 U phytase kg−1 diet was fed. A Zn equivalency forphytase has not been established for pigs.

Influence of microbial phytase on amino acids and nitrogenbioavailability

Phytate can bind with protein/amino acid (AA) at low and neutral pH (De Rhamand Jost, 1979; Cosgrove, 1980; Anderson, 1985; Thompson, 1986; Fretzdorff et al.,1995). Phytate–protein/AA complexes may occur in foodstuffs in the native state,and may be formed in the upper gastrointestinal tract. Complexing of phytatewith proteolytic enzymes may also occur in the upper gastrointestinal tract. Thesepotential phytate–protein complexes thus may reduce the utilization of proteinsand amino acids.

If phytate is hydrolysed, then its inhibitory effects are reduced. Most of the earlyresearch with microbial phytase was conducted to measure the effects of phytase onP utilization, and results were inconsistent when total tract N digestibility wasmeasured. Amino acid digestibility was rarely evaluated. Kemme (1998) provides agood review in his dissertation.

The opportunity to show improvements in protein/AA utilization is influencedby the dietary level of protein/AA. If the protein/AA retention is maximum, then thepotential to show an improvement is greatly reduced. Furthermore, the use of totaltract (faecal) digestibility may not be reliable because of the influence of the microbialpopulation in the large intestine. Ileal digestibility is a more appropriate method ofevaluating the influence of phytase on protein/AA utilization.

Binding of free amino acids (particularly lysine) by phytate has been demon-strated in vitro (Rutherfurd et al., 1997). Approximately 20% of free lysine-HCl wasbound following incubation with rice pollards (rich in phytate). Half of that amountwas liberated after the addition of phytase. In broilers, Ravindran and Bryden (1997)and Ravindran et al. (1999a) reported that dietary addition of phytic acid as ricepollards reduced ileal digestibility of N and AA (lysine, threonine, isoleucine, leucine,valine, phenylalanine and histidine). These adverse effects of phytic acid wereeffectively overcome by supplemental phytase.

Broilers and turkeysThe apparent N retention of broilers was improved when phytase was added to a23% crude protein maize–soybean diet (Kornegay et al., 1996b). Yi et al. (1996c),using Large White turkey female poults fed a maize–soybean meal diet, reported thatapparent and true ileal digestibilities of N and AA were generally improved when750 U phytase kg−1 was added to both 22.5 and 28.0% CP diets that contained


0.45% nP. Improvements were only observed for birds fed 22.5% CP that contained0.60% nP. Ravindran and Bryden (1997) similarly observed that improvements inAA digestibility were higher in low P wheat–sorghum–soybean meal diets.

Using young broilers, Kornegay (1996b) reported that ileal digestibilities of allamino acids was linearly decreased (P < 0.001) as the CP level increased (17, 20 and23%). Ileal digestibilities of all amino acids except methionine and proline werelinearly increased (P < 0.05 to 0.001) as phytase (250, 500 and 750 U kg−1) wasadded to the 17 and 20% CP diets. In a broiler finishing study, Kornegay et al.(1998b) reported that when dietary protein/AA levels were reduced from 95 to85% of the NRC (1994) recommendation (1.5–2.0% units of CP), additions of300–450 U phytase kg−1 diet prevented the decrease in performance (slightly lower),breast meat yield, and ileal CP/AA digestibilities observed.

Supplemental phytase (1200 U kg−1) was reported by Ravindran et al. (1999b)to improve ileal digestibilities of protein/AA of three cereals (maize, sorghum andwheat), four oilseed meals (soybean meal, canola meal, cottonseed meal andsunflower meal) and two cereal by-products (wheat middlings and rice polishings).The magnitude of the response varied among feedstuffs and individual AA.

Using a low lysine (85% NRC) wheat–sorghum–soybean meal diet withmultiple levels of crystalline lysine or microbial phytase and based on BW gainand feed : gain ratios of broilers, Ravindran and Bryden (1998) reported that500 U phytase kg−1 diet was equivalent to 0.7 g lysine kg−1 diet. Apparentmetabolizable energy was also linearly increased as phytase was added, but not aslysine was added. Ravindran and Bryden (1998) suggest that BW gain and feedefficiency responses to phytase may have resulted from both lysine and energy effects.In a similar study using broilers fed a single level (600 U kg−1) of phytase or multiplelevels of crystalline lysine added to a low lysine (0.8%) maize–soybean meal diet,Johnston and Southern (1999) reported an average equivalency, based on BW gainand feed efficiency, of 0.23 g lysine kg−1 diet for 600 U phytase kg−1 diet.

PigsSimilar findings have been reported in pigs. Officer and Batterham (1992) and Khanand Cole (1993) reported that microbial phytase improved the apparent ileal andtotal tract digestibility of crude protein by 2.3–12.8% units. Using cannulated pigsin two studies, Mroz et al. (1994) and Kemme (1998) reported that microbialphytase (800 and 900 U kg−1, respectively) enhanced the apparent ileal digestibilityof N and most amino acids. Added microbial phytase generally enhanced the totaltract digestibility of N and the retention of N. Li et al. (1998) reported improvedDM digestibility and N retention when phytase was added to a maize–soybean mealdiet fed to weaned pigs. In a literature review, Jongbloed et al. (1996d) reported in 17experiments using microbial phytase that the average apparent total tract digestibilityof protein was enhanced by 0.85 ± 1.70% units.

Faecal N digestibility, ileal AA and N digestibilities and N retention weredetermined in a feeding trial and N-balance study, and a cannulation study usingfinishing pigs fed diets containing: (i) NRC (1998) CP levels (12% CP); (ii) 7.5%

254 E.T. Kornegay

CP reduction (11% CP); (iii) 15.0% CP reduction (10% CP); (iv) diet 3 (10% CP)plus 250 U phytase kg−1; and (v) diet 3 plus 500 U phytase kg−1 (Kornegay et al.,1998a). Apparent digestion coefficients for N and AA increased as the dietary levelsof CP increased and they increased as microbial phytase was added to the 10% CPdiet. Coefficients were generally equal to or greater than values for pigs fed the 11%CP diet (a 7.5% reduction in CP). Using a conservative reduction of CP of 1% unit(7.1% reduction), and assuming a N excretion (faeces and urine) value of 40% of Nintake, N excretion was estimated to be reduced by 7.1% when phytase was added to10–11% CP pig diets at a level of 500 U kg−1. Therefore, phytase supplementationhas the potential to reduce P and N/AA excretion when P and N/AA levels arereduced appropriately.

Influence of microbial phytase on energy metabolism

Thompson and Yoon (1984) indicated that, in the native state, phytate couldcomplex with starch. Rajas and Scott (1969) reported that the apparentmetabolizable energy (AME) of cottonseed meal and soybean meal for chicks wasimproved following treatment of the meals with a crude phytase preparation fromAspergillus ficuum. Later, Miles and Nelson (1974), using chicks and a similarproduct, reported improvements in the AME value of cottonseed meal and wheatbran, but not soybean meal, when treated with the phytase preparation.

More recent studies, primarily in Australia, used a genetically modified phytase(reviewed by Ravindran, 1999). Small but significant improvements in AMEwere observed for broilers fed sorghum–soybean meal-based diets (Farrell et al.,1993; Selle et al., 1999), and sorghum–soybean meal–canola meal–cottonseedmeal–wheat middling diets (Selle et al., 1999) when phytase was supplemented to thediets. The AME values of a sorghum–soybean meal-based diet and in diets with 60%rice bran fed to ducks was improved when phytase was added (Martin and Farrell,1994).

Findings from three very recent studies (Bryden and Ravindran, 1998;Ravindran et al., 1999a,b), which were designed to determine the influence ofmicrobial phytase on protein/AA and energy utilization in poultry, clearly show thatsupplemental phytase improves the AME value of wheat- and sorghum-based poultrydiets (reviewed by Ravindran, 1999). In agreement, Kornegay and Denbow (VirginiaPolytechnic Institute and State University, 1998, unpublished data) observed thatphytase supplementation (600 U kg−1) improved AME coefficients in 4-week-oldbroilers fed maize–soybean meal, maize–wheat middlings–soybean meal, maize–meat meal–soybean meal or wheat–maize–soybean meal diets. It was suggested,based on literature reported by Ravindran (1999), that phytic acid may exert itseffects on starch digestion in one or more ways: (i) by binding α-amylase or bychelating Ca2+, which is needed for the normal activity of amylase; and (ii) bybinding starch directly through a protein linkage. Of course, hydrolysis of phytateP would release the enzyme or free the starch.


Influence of dietary calcium and calcium : phosphorus ratio on theeffectiveness of phytase

The response to a given level of supplemental phytase will be influenced by dietaryCa level and/or Ca : P ratio, dietary P level and dietary phytate level (Lei et al., 1994;Dungelhoef and Rodehutscord, 1995; Kornegay, 1996b). A high molar ratio of Cato phytate in the diet can lead to the formation of extremely insoluble Ca–phytatecomplexes under intestinal conditions, making the phytate molecule inaccessible tophytase. The presence of such strong complexes could explain the apparent inactivityof phytase in calcium-rich diets rather than a direct inhibition of the enzyme by Ca2+

(Wise, 1983; Bos, 1988). The importance of maintaining a narrow ratio of totalCa to total P (or for that matter available Ca to available P) has recently beendemonstrated in broilers (Qian et al., 1997), turkeys (Qian et al., 1996b) and pigs(Qian et al., 1996a; Liu et al., 1998).

In broilers and turkeys, a total Ca : total P ratio of 1.1 : 1 to 1.4 : 1 appeared tobe equally effective. Feeding diets with wider ratios reduces performance, P utiliza-tion and bone mineralization. In young broilers, using data reported by Qian et al.(1997), a 14.5, 8.5 and 8.3% decrease was calculated for BW gain, P retention andtoe ash percentage, respectively, when the Ca : total P ratio was increased from1.1 : 1 to 2.0 : 0. In young turkeys, using data reported by Qian et al. (1996b), an8.7, 10.8 and 6.6% decrease was calculated for BW gain, P retention and toe ashpercentage, respectively, when Ca : total P ratio was increased from 1.1 : 1 to 2.0 : 1.

A Ca : nP (weight/weight) ratio of 2 : 1 is suggested for poultry with theexception of laying birds (NRC, 1994). A high level of dietary Ca was shown to affectadversely the availability of phytate P; whereas a high level of vitamin D was shown tohave a positive effect (Mohammed et al., 1991; Edwards et al., 1992). Schoner et al.(1993) reported in broilers that feeding high levels of Ca with a constant level of P(0.35%) reduced the increase in BW gain, feed intake and P and Ca retention thatwas observed when phytase was added. From their lowest (0.6%) to highest (0.9%)levels of Ca, the Ca : total P ratio varied from 1.7 : 1 to 2.57 : 1.

In pig studies, a greater response to phytase (better utilization of P) is obtainedwhen the total Ca : total P ratio is maintained between 1 : 1 to 1.1 : 1 (Qian et al.,1996a). For example, in growing–finishing pigs, Liu et al. (1998) reported a 4.5, 8.2and 9.7% decrease in ADG, P digestibility and metacarpal breaking strength, respec-tively, when the Ca : total P ratio was increased from 1.0 : 1 to 1.5 : 1. Lei et al.(1994) reported for pigs fed a maize–soybean meal diet with no added inorganic P(0.31% tP) that the ability of phytase to improve phytate–P availability was greatlyreduced at 0.92% dietary Ca (Ca : tP ratio was 3.1) compared with the 0.50%dietary Ca (Ca : tP ratio was 1.6 : 1). Dungelhoef and Rodehutscord (1995) alsoobserved from this review of the literature that high Ca levels reduce the absorptionof P and the utilization of phytate. Similar findings have also been reported byJongbloed et al. (1996c,d).

The decrease in the effectiveness of phytase as the Ca : tP ratio becomes widercould be explained as follows: (i) phytate P utilization in maize–soybean diets fed tobroilers is influenced by Ca and P levels in the diet (Edwards and Veltmann, 1983;

256 E.T. Kornegay

Ballam et al., 1984); (ii) the extra Ca can bind with phytate to form an insolublecomplex that is less accessible to phytase; or (iii) the extra Ca may directly repressphytase activity by competing for the active sites of the enzyme (Wise, 1983;Pointillart et al., 1985). The negative effect of wide Ca : tP ratios is greater at lowerlevels of supplemental phytase and at lower levels of aP or nP, because less P would bereleased as a result of reduced phytase activity.

Phytase and organic acids

The use of organic acids in combination with phytase to enhance the effectiveness ofphytase has produced mixed results. It is known that the inclusion of organic acidsin the diet will generally lower dietary pH (Kirchgessner and Roth, 1982; Giestingand Easter, 1985; Risley et al., 1992) with a resultant smaller effect on lowering thestomach pH (Risley et al., 1992; Radcliffe et al., 1998b). Also, it is well documentedthat there are two optimal pH values for maximizing microbial phytase activity – 2.5and 5.5 (Shieh et al., 1969; Simons et al., 1990) – with the activity about 50% greaterat pH 5.5. Organic acids (fumaric and citric) have been reported to promote someimprovement in the total tract apparent digestibility of minerals (Hohler and Pallauf,1993; Ravindran and Kornegay, 1993). Complexes are known to be formed betweenorganic acids and various cations such as minerals, and these may result in increasedintestinal absorption of minerals (Ravindran and Kornegay, 1993).

Jongbloed (1987) pointed out that a lower intestinal pH enhanced the solubilityof P and phytate, and thus increased P absorption. Jongbloed et al. (1996e) reportedtwo trials (cannulation and grower feeding trial) in which a combination of phytaseand lactic acid was fed in trial 1 and a combination of phytase and lactic acid,formic acid and propionic acid was fed in trial 2. In trial 1, lactic acid enhanced theeffectiveness of phytase on P digestibility but not in trial 2. In trial 2, only formic acidenhanced the effectiveness of phytase. In two trials, Radcliffe et al. (1998b) reportedthat the addition of citric acid to weanling pig diets lowered gastric pH and improvedpig performance. However, the lowered gastric pH did not enhance the efficacy ofmicrobial phytase. In fact, when high levels of phytase were supplemented to dietscontaining citric acid, there was a decrease in the amount of phytase activityrecovered from the stomach after slaughter (Yi and Kornegay, 1996). This did notaffect phytase efficacy, because no citric acid interaction with phytase was observedon growth performance, feed efficiency, dry matter, Ca and P digestibility or bonemeasurements.

Valencia and Chavez (1997) reported that 1.0% acetic acid addition to amaize–soybean meal diet enhanced the effectiveness of microbial phytase asmeasured by performance and apparent mineral utilization of young pigs. In abalance trial with young pigs, Li et al. (1998) reported that microbial phytasesupplementation of a maize–soybean meal diet improved digestibilities of P, Ca, Nretention and dry matter, and only non-significant trends for further improvementswere observed when 1.5% citric acid was added to the diet. Boling et al. (1998)


observed only additive effects of dietary addition of 6% citrate and 1450 Uphytase kg−1 using bone ash and weight gain of young chickens.

Further research is needed to understand the relationship between dietacidification and phytase effectiveness.

Effects of age and physiological status

The effects of age and physiological status of the pig on the efficacy of phytase inrendering phytate P available for absorption have been inconsistent. Kemme et al.(1997a) reported that the efficacy of the phytase for improving digestible P decreasedin the order of lactating sows, growing–finishing pigs, sows at the end of pregnancy,piglets and sows at mid-pregnancy. Efficacy of phytase was particularly low in sowson day 60 of pregnancy and high in lactating sows, with the growing–finishingpigs intermediate (Kemme et al., 1997b). They suggested that the differences amongthe categories probably relate to differences in gastric retention time, but also therequirement for digestible P by the specific category has to be taken into account.Harper and Kornegay (1997) reported slightly higher apparent digestibilitycoefficients for finisher pigs compared with grower pigs in one trial, but the oppositeeffect was observed in a second trial. Rodehutscord et al. (1998) recently comparedthe efficacy of microbial phytase fed in combination with a low P semipurified dietin growing pigs of either 16 or 39 kg body weight (BW). Digestibility of P was verysimilar for both BW groups. From these studies there is no indication that theefficacy of microbial phytase is dependent on BW of young pigs.

Effect of soaking on phosphorus availability

As pointed out by Helander (1995) in her dissertation review, the positive effect ofsoaking on P availability has been reported in many experiments involving differentfeedstuffs (Irving, 1980). The P in high-moisture maize or grain sorghum is knownto be considerably more available than in dry grain (Trotter and Allee, 1979a,b; Boydet al., 1983; Ross et al., 1983; Jongbloed et al., 1991). Although phytase is present inmature seeds, it appears to have little effect on phytate in dry or dormant seeds(Maga, 1982). Irving (1980) reported that water was necessary for the hydrolysis ofphytate.

Kemme and Jongbloed (1993) reported that the efficacy of microbial phytasewas enhanced by soaking the diet in water in one experiment, but in a secondexperiment, soaking had no effect on P digestibility. A pelleted diet with a differentcomposition was used in the second experiment and the soaking time was muchshorter; these factors, together, may explain why soaking had no effect on P utiliza-tion. A higher incubation temperature may also have increased phytate hydrolysis(Hill and Tyler 1954; Han, 1988). The pH of the diet is also most important.Tossenberger et al. (1993) reported that the degradation of sodium phytate was 81%

258 E.T. Kornegay

and 93%, respectively, for 2.5 and 5.0 h of soaking in a pH 5.0 solution. Therespective values were only 38% and 47% when the pH was increased to 6.0.

Beneficial effects of soaking on P utilization with whey were observed for abarley–rapeseed meal diet fed to growing pigs (Nasi et al., 1994), whereas no effect ofsoaking a pelleted barley–soybean meal diet was observed by Nasi and Helander(1994). Liu et al. (1997) reported improved P utilization when a soybean meal-baseddiet was soaked. A soaking by phytase quadratic interaction was observed; soakingthe 250 U kg−1 diet increased P absorption to a level similar to that obtained with the500 U kg−1 diet fed dry.

Wet feeding of growing–finishing pigs has increased, especially in Europe andhas the potential of enhancing phytase effectiveness and improving P digestibility.Reports from the industry are that phytase is more efficient in these feeding systems.

Effectiveness of Microbial Phytase in Fish Diets

Feed ingredients of plant origin are being used in greater amounts in intensive fishfarming. Like pigs and poultry, several species of fish utilize only one-third of thephytate P in plant feedstuffs. The addition of inorganic P is necessary to meet the Pneeds of the fish for normal growth. It has also been observed that high dietaryphytate levels lead to depressed growth, feed intake and protein utilization, whichmay be a result of phytate complexing with cations in the gastrointestinal tract so thatZn, protein and energy bioavailability are reduced (Spinelli et al., 1983; Richardsonet al., 1985, 1986; McClain and Gatlin, 1988).

A number of studies with striped bass (Hughes and Soares, 1998), carp (Schaferet al., 1995), catfish (Jackson et al., 1996; Eya and Lovell, 1997; Li and Robinson,1997), rainbow trout (Cain and Garling, 1995; Rodehutscord and Pfeffer, 1995;Lanari et al., 1998; Vielma et al., 1998) and Atlantic salmon (Storebakken et al.,1998) have shown that microbial phytase is very effective in enhancing P digestibilityand utilization. Generally, only single and usually high levels of phytase have beenfed, but multiple levels have been used in a few studies (Schafer et al., 1995; Jacksonet al., 1996; Li and Robinson, 1997). Reported improvements in P absorption andretention range from 15 to 45%, and are influenced by phytase level and sources ofingredients. Reduction in P excretion is reported to range from 30 to 88%.

In summary, microbial phytase is very effective in improving P availability ofdiets based on plant ingredients and can significantly reduce P excretion and accu-mulation in the water when dietary P levels are reduced. Because of very limited datawhere multiple levels of phytase were fed, it is not possible to derive a response curvefor any measurement so as to determine equivalency value based on good data sets.

Summary

When pig and poultry diets are formulated using significant amounts of plant-basedingredients that are low in native phytase, microbial phytase supplementation is very


effective for improving the availability of phytate P. Because phytate is known tocomplex a number of other minerals, amino acids/protein and even starch, theirbioavailability is enhanced when phytate is hydrolysed by phytase. Thus, theexcretion of P, Ca, Zn and N can be reduced significantly when diets are properlyformulated using phytase. The dose–response curve of phytase for improving Putilization is non-linear for pigs and poultry (broilers and turkeys) and the responsehas been described over a wide range of phytase levels. The dose–response curves forthe effects of phytase on Ca, Zn, amino acid/N and energy utilization are not asclearly understood as they are for P.

The dose response of phytase will, however, depend upon: (i) the level of phytaseused; (ii) the level of total P in the diet; (iii) the level of phytate P in the diet; (iv) thelevel of Ca and the Ca : P ratio; (v) the intrinsic level of phytase in foodstuffs; and (vi)processing and pelleting methods. The effects of physiological age and the influenceof organic acids are not clearly understood, if they exist.

Microbial supplementation of diets based on plant ingredients for several speciesof fish is very effective in improving P availability, which leads to reduced P excretionand accumulation in the water.

References

Adeola, O., Lawrence, B.V., Sutton, A.L. and Cline, T.R. (1995) Mineral utilization inphytase-and-zinc supplemented diets for pigs. Journal of Animal Science 73 (Suppl. 1), 73(abstract).

Anderson, P.A. (1985) Interactions between proteins and constituents that affect proteinquality. In: Finley, G.W. and Hopkins, D.T. (eds) Digestibility and Amino AcidAvailability in Cereals and Oilseeds. American Association of Cereal Chemists, St Paul,Minnesota, p. 31.

Asada, K., Tanaka, K. and Kasai, Z. (1969) Formation of phytic acid in cereal grains. Annals ofthe New York Academy of Science 165, 801–814.

Bagheri, S. and Gueguen, L. (1985) Effect of wheat bran and pectin on the absorption andretention of phosphorus, calcium, magnesium and zinc by the growing pig. ReproductiveNutrition Development 25, 705–716.

Baker, D.H. (1991) Bioavailability of minerals and vitamins. In: Miller et al. (eds) SwineNutrition, 1st edn. Butterworth-Heinemann, Stoneham, pp. 341–359.

Ballam, G.C., Nelson, T.S. and Kirby, L.K. (1984) Effect of fiber and phytate source and ofcalcium and P level on phytate hydrolysis in the chick. Poultry Science 63, 333–338.

Beers, S. (1992) Relative tussen dosering microbieel fytase en de verteerbaarheid van fosfor intwee verschillende startvoeders voor varkens. Rapport I.V.V.O. Nr. 228, Lelystad.

Beudeker, R.F. and Pen, J. (1995) Development of plant seeds expressing phytase as a feedadditive. Proceedings of 2nd European Symposium on Feed Enzymes. Noordwijkerhout,The Netherlands, October 25, p. 225.

Biehl, R.R., Baker, D.H. and DeLuca, H.F. (1995) 1a-Hydroxylated cholecalciferolcompounds act additively with microbial phytase to improve phosphorus, zinc andmanganese in chicks fed soy-based diets. Journal of Nutrition 125, 2407.

Bitar, K. and Reinhold, J.G. (1972) Phytase and alkaline phosphatase activities in intestinalmucosal of rat, chicken, calf and man. Biochimestry et Biophysiology Acta 268, 442.

260 E.T. Kornegay

Boling, S.D., Parsons, C.M. and Baker, D.H. (1998) Citric acid improves phytate phosphorusutilization in broiler chicks fed corn–soybean meal diets. Poultry Science 77 (Suppl. 1),S31 (abstract).

Bos, K.D. (1988) Degradation of phytate by wheat phytase. In: Proceedings of AnimalNutrition and Feeding Workshop. Wageningen, November 23–25, 3 pp.

Boyd, R.D., Hall, D. and Wu, J.F. (1983) Plasma alkaline phosphatase as a criterion fordetermining biological availability of phosphorus for swine. Journal of Animal Science 57,396–401.

Bryden, W.L. and Ravindran, V. (1998) Effects of enzyme combinations on apparentmetabolisable energy and ileal nitrogen digestibility in broiler chickens. Proceedings of theNutrition Society of Austria 22, 180.

Cain, K.D. and Garling, D.L. (1995) Pretreatment of soybean meal with phytase for salmoniddiets to reduce phosphorus concentrations in hatchery effluents. Progressive Fish-Culturist57, 114–119.

Calvert, C.C., Besecker, R.J., Plumlee, M.P., Cline, T.R. and Forsyth, D.M. (1978) Apparentdigestibility of phosphorus in barley and corn for growing swine. Journal of AnimalScience 47, 420.

Cheryan, M. (1980) Phytic acid interactions in food system. CRC Critical Review of FoodScience and Nutrition 13, 297–335.

Cosgrove, D.J. (1980) Inositol Phosphates: Their Chemistry, Biochemistry and Physiology.Elsevier, New York.

Cromwell, G.L. (1992) The biological availability of phosphorus in feedstuffs for pigs. PigNews and Information 13(2), 75N.

De Rham, O. and Jost, T. (1979) Phytate protein interactions in soybean extracts andlow-phytate soy protein products. Journal of Food Science 44, 596.

Denbow, D.M., Ravindran, V., Kornegay, E.T., Yi, Z. and Hulet, R.M. (1995) Improvingphosphorus availability in soybean meal for broilers by supplemental phytase. PoultryScience 74, 1831–1842.

Dungelhoef, M. and Rodehutscord, M. (1995) Wirkung von phytasen auf die verdaulichkeitdes phosphorus beim schwein. Übersichten zur Tierernährung 23, 133–157.

Edwards, H.M. Jr and Veltmann, J.R. Jr (1983) The role of calcium and phosphorus inthe etiology of tibial dyschondroplasia in young chicks. Journal of Nutrition 113,1568–1575.

Edwards, H.M. Jr, Elliot, M.A. and Sooncharernying, S. (1992) Effect of dietarycalcium on tibial dyschondroplasia. Interaction with light, cholecalciferol,1,25-dihydroxycholecalciferol, protein and synthetic zeolite. Poultry Science 71,2041–2055.

Eeckhout, W. and De Paepe, M. (1991) The quantitative effects of an industrial microbialphytase and wheat phytase on the apparent phosphorus absorbability of a mixed feed bypiglets. Med. Fac. Landbouww. Rijksuniv. Gent, 56, 1643–47.

Eeckhout, W. and De Paepe, M. (1994) Total phosphorus, phytate–phosphorus and phytaseactivity in plant feedstuffs. Animal Feed Science and Technology 47, 19.

Eeckhout, W. and De Paepe, M. (1996) In vitro and in vivo comparison of microbialand plant phytase, In: Coelho, M.B. and Kornegay, E.T. (eds) Phytase in AnimalNutrition and Waste Management. BASF Corporation, Mount Olive, New Jersey,pp. 237–241.

Eya, J.C. and Lovell, R.T. (1997) Net absorption of dietary phosphorus from variousinorganic sources and effect of fungal phytase on net absorption of plan phosphorus bychannel catfish Ictalurus punctatus. Journal World Aquaculture Society 28, 386–391.


Farrell, D.J., Martin, E.A., Du Preez, J.J., Bongarts, M., Betts, M., Sudaman, A. andThomson, E. (1993) The beneficial effects of a microbial phytase in diets of broilerchickens and ducklings. Journal of Animal Physiology and Animal Nutrition 69, 278–283.

Fretzdorff, B., Brummer, J.M., Rocken, W., Greiner, R., Konietzny, U. and Jany, K.-D.(1995) Reduktion des Phytinsaure-Gehaltes bei der Herstellung von Backwaren undGetreidenahrmittein. AID-Verbraucherdienst 40, 12.

Gibson, D.M. and Ullah, A.B.J. (1990) Phytase and their actions on phytic acid. In: Morre,D.J., Boss, W.F. and Loewus, F.A. (eds) Inositol Metabolism in Plants. Wiley-Liss, Inc.,New York, pp. 77–92.

Giesting, D.W. and Easter, R.A. (1985) Response of starter pigs to supplementation ofcorn–soybean meal diets with organic acids. Journal of Animal Science 60, 1288–1294.

Gordon, R.W. and Roland, D.A. Sr (1997) Performance of commercial laying hens fedvarious phosphorus levels, with and without supplemental phytase. Poultry Science 76,1172–1177.

Gueguen, L., Besancon, P. and Rerat, A. (1968) Utilisation digestive, cinétique de l’absorptionet éfficacité de retention du phosphore phytique chez le porc. Annals of Biology, AnimalBiochemistry and Biophysiology 8(2), 273.

Han, Y.W. (1988) Phytate hydrolysis in soybean and cottonseed meals by Aspergillus ficuumphytase. Journal of Agricultural and Food Chemistry 36, 259–262.

Harland, B.F. and Morris, E.R. (1995) Phytate: a good or a bad food component. NutritionResearch 15, 733–754.

Harper, A.F. and Kornegay, E.T. (1997) Phytase supplementation of low-phosphorusgrowing–finishing pig diets improves performance, phosphorus digestibility and bonemineralization and reduces phosphorus excretion. Journal of Animal Science 75,3174–3186.

Harper, A.F., Skaggs, J.H., Veit, H.P. and Kornegay, E.T. (1999) Efficacy and safety of NovoSP938 microbial phytase supplementation of a corn–soybean meal diet fed to growingpigs. Journal of Animal Science (abstract) (in press).

Helander, E. (1995) Efficiency of microbial phytases on phosphorus utilization ingrowing–finishing pigs. Academic Dissertation, University of Helsinki.

Hill, R. and Tyler, C. (1954) The influence of time, temperature, pH and calcium carbonateon the activity of the phytase of certain cereals. Journal of Agricultural Science 44,306–310.

Höhler, D. and Pallauf, J. (1993) Untersuchungen zum einfluss von zitronensaure aufdie mineralstoffverwertung bveim ferkel anhand einer mais-soja-diat mit und ohneZn-erganzung. Journal of Animal Physiology and Animal Nutrition 69, 133.

Hughes, K.P. and Soares, J.H. Jr (1998) Efficacy of phytase on phosphorus utilization inpractical diets fed to striped bass Morone saxatilis. Aquaculture Nutrition 4, 133–140.

Ingelmann, H.-J., Rimbach, G. and Pallauf, J. (1993) Phytinsaure-ein antinutritiver Faktor?Ernahrungsumschau 40, 400.

Irving, G.C.J. (1980) Phytates. In: Cosgrove, D.J. (ed.) Inositol Phytates. Elsevier, Amsterdam,pp. 85–127.

Irving, G.C.J. and Cosgrove, D.J. (1974) Inositol phosphate phosphatases of microbiologicalorigin. Some properties of the partially purified phosphatases of Aspergillus ficuum NRRL3135. Australian Journal of Biological Science 27, 361–368.

Jackson, S.L., Li, M.H. and Robinson, E.H. (1996) Use of microbial phytase in channelcatfish Icatalurus punctatus diets to improve utilization of phytate phosphorus. Journal ofWorld Aquaculture Society 27, 309–313.

262 E.T. Kornegay

Johnston, S.L. and Southern, L.L. (1999) The effect of phytase addition or the reduction ofdietary calcium and available phosphorus on lysine bioavailability in growing chicks.Southern Poultry Science Society Meeting. 18 January 1999, Atlanta, Georgia, p. 13.

Jongbloed, A.W. (1987) Phosphorus in the feeding of pigs. Effect of the diet on the absorptionand retention of phosphorus by growing pigs. PhD thesis, University of Wageningen,The Netherlands.

Jongbloed, A.W. and Kemme, P.A. (1990) Apparent digestible phosphorus in the feeding ofpigs in relation to availability, requirement and environment. 1. Digestible phosphorus infeedstuffs from plant and animal origin. Netherlands Journal of Agriculture Science 38,567.

Jongbloed, A.W., Everts, H. and Kemme, P.A. (1991) Phosphorus availability andrequirements in pigs. In: Haresign, W. and Cole, D.J.A. (eds) Recent Advances in AnimalNutrition. Butterworths, London, pp. 65–80.

Jongbloed, A.W., Mroz, Z. and Kemme, P.A. (1992) The effect of supplementary Aspergillusniger phytase in diets for pigs on concentration and apparent digestibility of dry matter,total phosphorus and phytic acid in different sections of the alimentary tract. Journal ofAnimal Science 70, 1159–1168.

Jongbloed, A.W., Kemme, P.A. and Mroz, Z. (1996a) Effectiveness of Natuphos phytase inimproving the bioavailabilities of phosphorus and other nutrients for growing–finishingpigs. In: Coelho, M.B. and Kornegay, E.T. (eds) Phytase in Animal Nutrition and WasteManagement. BASF Corporation, Mount Olive, New Jersey, pp. 259–274.

Jongbloed, A.W., Kemme, P.A. and Mroz, Z. (1996b) The influence of dietary calcium levelson efficacy of microbial phytase. In: Coelho, M.B. and Kornegay, E.T. (eds) Phytase inAnimal Nutrition and Waste Management. BASF Corporation, Mount Olive, New Jersey,pp. 393–400.

Jongbloed, A.W., Kemme, P.A. and Mroz, Z. (1996c) Phytase in swine rations: impact onnutrition and environment. BASF Technical Symposium. Des Moines, Iowa, January 29,pp. 44–69.

Jongbloed, A.W., Kemme, P.A., Mroz, Z. and Jongbloed, R. (1996d) The effect of organicacids in diets for growing pigs on the efficacy of microbial phytase. In: Coelho, M.B.and Kornegay, E.T. (eds) Phytase in Animal Nutrition and Waste Management. BASFCorporation, Mount Olive, New Jersey, pp. 515–524.

Kemme, P.A. (1998) Phytate and phytases in pig nutrition. Impact on nutrient digestibilityand factors affecting phytase efficacy. PhD thesis, University of Wageningen, TheNetherlands.

Kemme, P.A. and Jongbloed, A.W. (1993) Effect of Aspergillus niger phytase and soakingon the apparent digestibility of P in diets for pigs. Journal of Animal Science 71(Supplement 1), 181.

Kemme, P.A., Jongbloed, A.W., Mroz, Z. and Beynen, A.C. (1997a) The efficacy ofAspergillus niger phytase in rendering phytate phosphorus available for absorption in pigsis influenced by pig physiological status. Journal of Animal Science 75, 2129–2138.

Kemme, P.A., Radcliffe, J.S., Jongbloed, A.W. and Mroz, Z. (1997b) The effect of sow parityon digestibility of proximate components and minerals during lactation as influencedby diet and microbial phytase supplementation. Journal of Animal Science 75,2147–2153.

Kemme, P.A., Jongbloed, A.W., Mroz, Z. and Beynen, A.C. (1998) Diurnal variation indegradation of phytic acid by plant phytase in the pig’s stomach. Livestock ProductionScience 54, 33–44.


Khan, N. and Cole, D.J.A. (1993) The effect of dietary inclusions of phytase and yeast onapparent phosphorus digestibility in pigs. Proceedings of Winter Meeting of the BritishSociety of Animal Production. Scarborough, UK, p. 2.

Kirchgessner, M. and Roth, F.X. (1982) Fumaric acid as a feed additive in pig nutrition. PigNews and Information 3, 259–264.

Knuckles, B.E., Kuzmicky, D.D. and Betschart, A.A. (1985) Effect of phytate and partiallyhydrolyzed phytate on in vitro protein digestibility. Journal of Food Science 50,1080–1082.

Kornegay, E.T. (1996a) Effectiveness of Natuphos phytase in improving the bioavailability ofphosphorus and other nutrients in corn–soybean meal diets for young pigs. In: Coelho,M.B. and Kornegay, E.T. (eds) Phytase in Animal Nutrition and Waste Management.BASF Corporation, Mount Olive, New Jersey, pp. 249–258.

Kornegay, E.T. (1996b) Natuphos™ phytase in swine diets: digestibility, bone and carcasscharacteristics. In: Proceedings of BASF Technical Symposium/Carolina Swine NutritionConference. 12 November 1996, pp. 28–67.

Kornegay, E.T. (1999) A review of phosphorus digestion and excretion as influenced bymicrobial phytase in poultry. BASF Technical Symposium following the southern PoultryScience meetings (18 January 1999), Atlanta, Georgia.

Kornegay, E.T. and Qian, H. (1996) Replacement of inorganic phosphorus by microbialphytase for young pigs fed a corn soybean meal diet. British Journal of Nutrition 76,563–578.

Kornegay, E.T. and Radcliffe, J.S. (1996) Bioavailability of inorganic sources of phosphorusfor swine. Animal and Poultry Science Research Report 12, 20–21.

Kornegay, E.T., Denbow, D.M., Yi, Z. and Ravindran, V. (1996a) Response of broilers tograded levels of microbial phytase added to maize–soyabean meal-based diets containingthree levels of non-phytate phosphorus. British Journal of Nutrition 75, 839–852.

Kornegay, E.T., Denbow, D.M. and Radcliffe, J.S. (1996b) Bioavailability of inorganicsources of phosphorus for turkeys. Animal and Poultry Science Research Report 12, 22–23.

Kornegay, E.T., Radcliffe, J.S. and Denbow, D.M. (1996c) Influence of Natuphos phytase oncalcium bioavailability in plant ingredients and development of calcium equivalencyvalues for swine and poultry. In: Coelho, M.B. and Kornegay, E.T. (eds) Phytase inAnimal Nutrition and Waste Management. BASF Corporation, Mount Olive, New Jersey,pp. 419–434.

Kornegay, E.T., Radcliffe, J.S. and Zhang, Z. (1998a) Influence of phytase and dietcomposition on phosphorus and amino acid digestibilities, and phosphorus and nitrogenexcretion in swine. BASF Technical Symposium, Carolina Swine Nutrition Conference.16 November, pp. 125–155.

Kornegay, E.T., Zhang, Z. and Denbow, D.M. (1998b) Influence of microbial phytasesupplementation of a low protein/amino acid diet on performance, ileal digestibility ofprotein and amino acids, and carcass measurements of finishing broilers. Poultry Science77 (Suppl. 1), S33.

Kornegay, E.T., Radcliffe, J.S. and Zhang, Z. (1998c) Non-phosphorus benefits of feedingmicrobial phytase to pigs and poultry. Animal Production Systems and the Environment.Des Moines, Iowa, July 19–22, 1998, pp. 79–84.

Kornegay, E.T., Skaggs, J.H., Denbow, D.M., Larsen, C.T. and Veit, H.P. (1999) Efficacyand safety of Novo SP938 microbial phytase supplementation of a corn–soybean mealdiet fed to turkeys. Poultry Science 78, 15.

Kratzer, F.H. and Vohra, P. (1986) Role of phytic acid and other phosphates as chelatingagent. In: Chelates in Nutrition. CRC Press, Boca Raton, Florida, pp. 49–61.

264 E.T. Kornegay

Lanari, D., D’Agaro, E. and Turri, C. (1998) Use of nonlinear regression to evaluate theeffects of phytase enzyme treatment of plant protein diets for rainbow trout(Oncorhynchus mykiss). Aquaculture 161, 343–354.

Lantzsch, H.J., Hillenbrand, S., Scheuermann, S.E. and Menke, K.H. (1992) Comparativestudy of phosphorus utilization from wheat, barley and corn diets by young rats and pigs.Journal of Animal Physiology and Animal Nutrition 67, 123–132.

Ledoux, D.R., Broomhead, J.N., Firman, J.D. and Bermudez, A.J. (1998) Efficacy ofPhytaseed®, a phytase containing transgenic canola, to improve phytate phosphorusutilization from corn–soybean meal diets fed to turkey poults from day 1 to 35. PoultryScience 77 (Suppl. 1), 54.

Lei, X.G., Ku, P.U., Miller, E.R., Ullrey, D.E. and Yokoyama, M.T. (1993) Supplementalmicrobial phytase improves bioavailability of dietary zinc to weanling pigs. Journal ofNutrition 123, 1117.

Lei, X.G., Ku, P.K., Miller, E.R., Yokoyama, M.T. and Ullrey, D.E. (1994) Calcium levelaffects the efficacy of supplemental microbial phytase in corn–soybean meal diets ofweanling pigs. Journal of Animal Science 72, 139–143.

Leske, K.L. and Coon, C.N. (1998) Effect of phytase on total and phytate phosphorusretention of feed ingredients as determined with broilers and laying hens. Poultry Science77 (Suppl. 1), 54 (abstract).

Li, D., Che, X., Wang, Y., Hong, C. and Thacker, P.A. (1998) Effect of microbial phytase,vitamin D3, and citric acid on growth performance and phosphorus, nitrogen andcalcium digestibility in growing swine. Animal Feed Science and Technology 73, 173–186.

Li, M.H. and Robinson, E.H. (1997) Microbial phytase can replace inorganic phosphorussupplements in channel catfish Ictalurus punctatus diets. Journal of World AquacultureSociety 28, 402–406.

Liebert, F., Wecke, C. and Schoner, F.J. (1993) Phytase activities in different gut contents ofchickens are dependent on level of phosphorus and phytase supplementations. In: Wenk,C. and Boessinger, M. (eds) Enzymes in Animal Nutrition. Proceedings of 1st SymposiumKartause Ittingen, Switzerland, October 13–16, p. 202.

Liu, J., Bollinger, D.W., Ledoux, D.R., Ellersieck, M.R. and Veum, T.L. (1997) Soakingincreases the efficacy of supplemental microbial phytase in a low-phosphorus corn–soybean meal diet for growing pigs. Journal of Animal Science 75, 1292–1298.

Liu, J., Bollinger, D.W., Ledoux, D.R. and Veum, T.L. (1998) Lowering the dietary calciumto total phosphorus ratio increased phosphorus utilization in low-phosphorus corn–soybean meal diets supplemented with microbial phytase for growing–finishing pigs.Journal of Animal Science 76, 808–813.

Maddaiah, V.T., Kurnick, A.A. and Reid, B.L. (1964) Phytic acid studies. Proceedings of theSociety of Experimental Biology and Medicine 115, 391.

Maenz, D.D. and Classen, H.L. (1998) Phytase activity in the small intestinal brush bordermembrane of the chicken. Poultry Science 77, 557–563.

Maga, J.A. (1982) Phytate: its chemistry, occurrence, food interactions, nutritional signifi-cance and methods of analysis. Journal of Agricultural Food Chemistry 30, 1–9.

Martin, E. and Farrell, D.J. (1994) The effect of a microbial phytase in rice bran based dietsfed to grower–finisher diets. Proceedings of the Australian Poultry Science Symposium 6,88–91.

McCance, R.A. and Widdowson, E.M. (1944) Activity of the phytase in different cereals andits resistance to dry heat. Nature 153, 660.

McClain, W.R. and Gatlin, D.M. (1988) Journal of World Aquaculture Society 19, 103.


McCollum, E.V. and Hart, E.B. (1908) On the occurrence of a phytin-splitting enzyme inanimal tissue. Journal of Biological Chemistry 4, 497.

Meulen, J. van der and Bakker, J.G.M. (1991) Effect of various sources of dietary fibre onchemico-physical characteristics of digesta in the stomach and the small intestine of thepig. Proceedings from the Fifth International Symposium on Digestive Physiology in the Pig,pp. 440–445.

Miles, R.D. Jr and Nelson, T.S. (1974) The effect of enzymatic hydrolysis of phytate on theavailable energy content of feed ingredients for chicks and rats. Poultry Science 53,1714–1717.

Mohammed, A., Gibney, M.J. and Taylor, T.G. (1991) The effects of dietary levels ofinorganic phosphorus, calcium and cholecalciferol on the digestibility of phytate-P by thechick. British Journal of Nutrition 66, 251.

Mroz, Z., Jongbloed, A.W. and Kemme, P.A. (1994) Apparent digestibility and retention ofnutrients bound to phytate complexes as influenced by microbial phytase and feedingregimen in pigs. Journal of Animal Science 72, 126–132.

Nair, V.C. and Duvnjak, Z. (1991) Phytic acid content reduction in canola meal by variousmicroorganisms in a solid state fermentation process. Acta Biotechnology 11, 211–218.

Nasi, M. and Helander, E. (1994) Effects of microbial phytase supplementation and soakingof barley–soybean meal on availability of plant phosphorus for growing pigs. ActaAgriculturae Scandinavica 44, 79.

Nasi, M., Helander, E. and Partanen, K. (1994) The effects of Aspergillus niger phytaseand soaking with whey on phosphorus utilization in growing pigs on barley–rapeseedmeal diet. Proceedings of VI International Symposium on Digestive Physiology in Pigs. BadDoberan, 10/4–6/94, Vol. I, 89–92.

Nayini, N.R. and Markakis, P. (1986) Phytases. In: Graf, E. (ed.) Phytic Acid: Chemistry andApplications. Pilatus Press, Minneapolis, p. 101.

Nelson, T.S. (1967) The utilization of phytate P by poultry – a review. Poultry Science 46,862.

Nelson, T.S., Shieh, T.J., Wodzinski, R.J. and Ware, J.H. (1971) Effect of supplementalphytase on the utilization of phytate phosphorus by chicks. Journal of Nutrition 101,1289–1293.

Newton, G.L., Hale, O.M. and Plank, C.O. (1983) Effect of wheat bran in practical diets onmineral absorption by pigs at two ages. Canadian Journal of Animal Science 63, 399–408.

NRC (1994) Nutrient Requirements of Poultry, 9th edn. National Academy Press, Washington,DC.

NRC (1998) Nutrient Requirements of Swine, 10th edn. National Academy Press, Washington,DC.

Nys, Y., Frapin, D. and Pointillart, P. (1996) Occurrence of phytase in plants, animals andmicroorganisms. In: Coelho, M.B. and Kornegay, E.T. (eds) Phytase in Animal Nutritionand Waste Management. BASF Corporation, Mount Olive, New Jersey, pp. 213–240.

Oberleas, D. (1973) Phytates. In: Toxicants Occurring Naturally in Foods. National Academyof Sciences, Washington, DC, pp. 363–371.

Oberleas, D. and Harland, B.F. (1981) Phytate content of foods: effect on dietary zincbioavailability. Journal of the American Dietetics Association 79, 443.

Oberleas, D. and Harland, B.F. (1996) Impact of phytic acid on nutrient availability.In: Coelho, M.B. and Kornegay, E.T. (eds) Phytase in Animal Nutrition and WasteManagement. BASF Corporation, Mount Olive, New Jersey, pp. 77–84.

O’Dell, B.L. and de Borland, A.R. (1976) Complexation of phytate with proteins and cationsin corn germ and oilseed meals. Journal of Agricultural Food Chemistry 24 (4), 804–808.

266 E.T. Kornegay

Officer, D.I. and Batterham, E.S. (1992) Enzyme supplementation of Linola meal. Proceedingsof Wollongbar Pig Industry Seminar on Feed Enzymes. Queensland, pp. 56–57.

Ogino, C., Takeuchi, L., Takeda, H. and Watanabe, T. (1979) Availability of dietaryphosphorus in carp and rainbow trout. Bulletin Japanese Society Science and Fisheries 45,1527–1532.

O’Quinn, P.R., Knabe, D.A. and Gregg, E.J. (1997) Digestible phosphorus needs ofterminal-cross growing–finishing pigs. Journal of Animal Science 75, 1308–1318.

Pallauf, J. and Rimbach, G. (1995) Recent results on phytic acid and phytase. Proceedings of5th Forum Animal Nutrition. May 4–5. BASF, p. 43.

Pallauf, V.J., Holer, D. and Rimbach, G. (1992) Effect of microbial phytase supplementationto a maize–soya-diet on the apparent absorption on Mg, Fe, Cu, Mn and Zn and para-meters of Zn-status in piglets. Journal of Animal Physiology and Animal Nutrition 68, 1–9.

Pasamontes, L., Haiker, M., Wyss, M., Tessier, M. and Loon, A.P.G.M. van (1997) Genecloning, purification, and characterization of a heat-stable phytase from the fungusAspergillus fumigatus. Applied and Environmental Microbiology 63, 1696–1700.

Peter, W. and Jeroch, H. (1993) The effectiveness of microbial phytase addition to layerrations on maize and wheat basis. In: Wenk, C. and Boessinger, M. (eds) Enzymes inAnimal Nutrition. Proceedings from First Symposium Kartause Ittingen, Switzerland,October 13–16, pp. 206–209.

Pointillart, A. (1991) Enhancement of phosphorus utilization in growing pigs fed phytate-richdiets by using rye bran. Journal of Animal Science 69, 1109–1115.

Pointillart, A. (1993) Importance of phytates and cereal phytases in the feeding of pigs.In: Wenk, C. and Boessinger, M. (eds) Enzymes in Animal Nutrition. Proceedings of 1stSymposium Kartause Ittingen, Switzerland, 13–16 October, pp. 192–199.

Pointillart, A., Fontaine, N., Thomasset, M. and Jay, M.E. (1985) Phosphorus utilization,intestinal phosphatases and hormonal control of calcium metabolism in pigs fed phyticphosphorus: soyabean or rapeseed diets. Nutrition Reports International 32, 155–166.

Prasad, A.S. (1966) Metabolism of zinc and its deficiency in human subjects. In: Prasad, A.S.(ed.) Zinc Metabolism. C.C. Thomas, Springfield, Illinois, p. 250.

Qian, H., Kornegay, E.T. and Conner, D.E. Jr (1996a) Adverse effects of wide calcium : phos-phorus ratios on supplemental phytase efficacy for weanling pigs fed two dietaryphosphorus levels. Journal of Animal Science 74, 1288–1297.

Qian, H., Kornegay, E.T. and Denbow, D.M. (1996b) Phosphorus equivalence of microbialphytase in turkey diets as influenced by calcium to phosphorus ratios and phosphoruslevels. Poultry Science 75, 69–81.

Qian, H., Kornegay, E.T. and Denbow, D.M. (1997) Utilization of phytate phosphorusand calcium as influenced by microbial phytase, cholecalciferol and the calcium : totalphosphorus ratio in broiler diets. Poultry Science 76, 37–46.

Radcliffe, J.S., Kornegay, E.T. and Conner, D.E. Jr (1995) The effect of phytase on calciumrelease in weanling pigs fed corn–soybean meal diets. Journal of Animal Science 73 (Suppl.1), 173 (abstract).

Radcliffe, J.S., Kornegay, E.T. and Zhang, Z. (1998a) Effectiveness of microbial phytase as afeeding strategy for reducing phosphorus excretion in pigs and poultry. Animal Produc-tion Systems and the Environment. Des Moines, Iowa, July 19–22, 1998, pp. 97–102.

Radcliffe, J.S., Zhang, Z. and Kornegay, E.T. (1998b) The effects of microbial phytase, citricacid and their interaction in a corn–soybean meal based diet for weanling pigs. Journal ofAnimal Science 76, 1880–1886.

Rajas, S.W. and Scott, J.L. (1969) Factors affecting the nutritive value of cottonseed meal as aprotein source for chick diets. Poultry Science 48, 819–835.


Ravindran, V. (1996) Occurrence of phytic acid in plant feed ingredients. In: Coelho, M.B.and Kornegay, E.T. (eds) Phytase in Animal Nutrition and Waste Management. BASFCorporation, Mount Olive, New Jersey, pp. 85–92.

Ravindran, V. (1999) Protein and energy effects of microbial phytase in poultry diets. Paperpresented at the BASF Technical Symposium following the southern Poultry Sciencemeetings, Atlanta, Georgia, 19 January 1999.

Ravindran, V. and Bryden, W.L. (1997) Influence of dietary phytic acid and available phospho-rus levels on the response of broilers to supplemental Natuphos®. University of Sydney, Reportto BASF, 1–54.

Ravindran, V. and Bryden, W.L. (1998) Lysine replacement value of Natuphos“ phytase inbroiler diets. University of Sydney, Report to BASF, 1–17.

Ravindran, V. and Kornegay, E.T. (1993) Acidification of weaner pig diet: a review. Journal ofthe Science of Food and Agriculture 62, 313–322.

Ravindran, V., Ravindran, G. and Sivalogan, S. (1994) Total and phytate phosphoruscontents of various foods and feedstuffs of plant origin. Food Chemistry 50, 133–136.

Ravindran, V., Kornegay, E.T., Denbow, D.M., Yi, Z. and Hulet, R.M. (1995a) Response ofturkey poults to tiered levels of Natuphos® phytase added to soybean meal-basedsemi-purified diets containing three levels of nonphytate phosphorus. Poultry Science 74,1843–1854.

Ravindran, V., Bryden, W.L. and Kornegay, E.T. (1995b) Phytates: occurrence, bioavail-ability and implications in poultry nutrition. Poultry Avian Biological Review 6, 125–143.

Ravindran, V., Cabahug, S., Ravindran, G. and Bryden, W.L. (1999) Influence of microbialphytase on apparent ileal amino acid digestibility of feedstuffs for broilers. Poultry Science78, 699–706.

Ravindran, V., Cabahug, S., Ravindran, G., Selle, P.H. and Bryden, W.L. (2000) Response ofbroiler chickens to microbial phytase supplementation as influenced by dietary phyticacid and non-phytate phosphorus levels. II. Effects on apparent metabolisable energy,nutrient digestibility and nutrient retention. British Poultry Science 41, 193–300.

Reddy, N.R., Sathe, S.K. and Salunkhe, D.K. (1982) Phytates in legumes and cereals.In: Chichester, C.O., Mrak, E.M. and Stewart, G.F. (eds) Advances in Food Research.Academic Press, New York, p. 1.

Richardson, N.L., Higgs, D.A., Beames, R.M. and McBride, J.R. (1985) Influence of dietarycalcium, phosphorus, zinc, and sodium phytate level on cataract incidence, growthand histopathology in juvenile Chinook salmon (Oncorhynchus tshawytscha). Journal ofNutrition 115, 553–567.

Richardson, N.L., Higgs, D.A. and Beames, R.M. (1986) The susceptibility of juvenilechinook salmon (Oncorhynchus tshawytscha) to cataract formation in relation to dietarychanges in early life. Aquaculture 52, 237–243.

Risley, C.R., Kornegay, E.T., Lindemann, M.D., Wood, C.M. and Eigel, W.N. (1992) Effectof feeding organic acids on selected intestinal content measurements at varying timespostweaning in pigs. Journal of Animal Science 70, 196–206.

Roberson, K.D. and Edwards, H.M. Jr (1994) Effects of 1,25-dihydroxycholecalciferol andphytase on zinc utilization in broiler chicks. Poultry Science 72, 1312–1326.

Rodehutscord, M. and Pfeffer, E. (1995) Effects of supplemental microbial phytase onphosphorus digestibility and utilization in rainbow trout (Oncorhynchus mykiss). WaterScience and Technology 31, 143–147.

Rodehutscord, M., Krause, G. and Pfeffer, E. (1998) The effect of phytase on the availabilityof phosphorus in different ingredients in swine. Proceedings of BASF TechnicalSymposium/Carolina Swine Nutrition Conference, November 12 1996, pp. 32–45.

268 E.T. Kornegay

Ross, R.D., Cromwell, G.L. and Stahly, T.S. (1983) Biological availability of the phosphorusin regular and dehulled soybean meal for growing pigs. Journal of Animal Science 55(Suppl. 1), 93 (abstract).

Rutherfurd, S.M., Edwards, A.C. and Selle, P.H. (1997) Effect of phytase on lysine–ricepollard complexes. In: Cranwell, P.D. (ed.) Manipulating Pig Production VI. AustralasianPig Science Association, Werribee, Victoria, Australia, p. 248.

Schafer, A., Koppe, W.M., Meyer-Burgdorff, K.-H. and Gunther, K.D. (1995) Effects ofa microbial phytase on the utilization of native phosphorus by carp in a diet based onsoybean meal. Water Science and Technology 31, 149–155.

Schoner, F.J., Hoppe, P.P. and Schwarz, G. (1991) Comparative effects of microbial phytaseand inorganic phosphorus on performance and on retention of phosphorus, calcium andcrude ash in broilers. Journal of Animal Physiology and Animal Nutrition 66, 248–255.

Schoner, F.J., Hoppe, P.P., Schwarz, G. and Wiesche, H. (1993) Effects of microbial phytaseand inorganic phosphate in broiler chickens: performance and mineral retention atvarious calcium levels. Journal of Animal Physiology and Animal Nutrition 69, 235–244.

Schoner, F.J., Schwarz, G., Hoppe, P.P. and Wiesche, H. (1994) Effect of microbial phytaseon Ca-availability in broilers. Third Conference of Pig and Poultry Nutrition. Halle,Germany, November 29–December 1.

Schulz, E. and Oslage, H.J. (1972) Untersuchungen zur intestinalen hydrolyse von inosit-phosphorsaureester und zur absorption von phytinphosphor beim schwein. 2. Mitteilunguntersuchungen zur hydrolyse der inositphosphorsaureester im verdauungstrakt beimschwein. Z. Tierphysiol., Tierernahrg. u. Futtermittelkde 30, 76–91.

Selle, P.H., Ravindran, V., Pittolo, P.H. and Bryden, W.L. (1999) An evaluation of microbialphytase in sorghum-based broiler diets. Proceedings of Australian Poultry Symposium 11(in press).

Shieh, T.R., Wodzinski, R.J. and Ware, J.H. (1969) Regulation of the formation of acidphosphatases by inorganic phosphate in Aspergillus ficuum. Journal of Bacteriology 100,1161.

Shih, B.L. and Hsu, A. (1997) Effects of dietary phytase supplementation on the growthperformance, bone mechanical properties and phosphorus excretion of finishing pigs.Journal of Taiwan Livestock Research 30, 1–11.

Simecek, K., Zobac, P. and Kumprecht, I. (1995) The effect of microbial phytasesupplementation in feed mixture on phosphorus and calcium utilization in growing pigs.Zivocisna Vyroba 40, 403–406.

Simons, P.C.M. and Versteegh, H.A.J. (1993) The effect of the addition of low doses of microbialphytase to layer feed on the technical results and skeleton and eggshell quality. Spelderholtpublication no. 589 (NL), Beekbergen, The Netherlands.

Simons, P.C.M., Versteegh, H.A.J., Jongbloed, A.W., Kemme, P.A., Slump, P., Bos, K.E.,Wolters, M.G.E., Beudeker, R.F. and Verschoor, G.J. (1990) Improvement ofphosphorus availability by microbial phytase in broilers and pigs. British Journal ofNutrition 64, 525–540.

Simons, P.C.M., Jongbloed, A.W., Versteegh, H.A.J. and Kemme, P.A. (1992) Improvementof phosphorus availability by microbial phytase in poultry and pigs. Georgia NutritionConference (19 November), pp. 100–109.

Singh, M. and Krikorian, A.D. (1982) Inhibition of trypsin activity by phytate. Journal ofAgricultural and Food Chemistry 30, 799–800.

Spinelli, J., Houle, C.R. and Wekell, J.C. (1983) The effect of phytates on the growth ofrainbow trout (Salmo gairdneri) fed purified diets containing varying quantitites ofcalcium and magnesium. Aquaculture 30, 71–83.


Storebakken, T., Shearer, K.D. and Roem, A.J. (1998) Availability of protein, phosphorusand other elements in fish meal, soy-protein concentrate and phytase-treated soy-protein-concentrate-based diets to Atlantic salmon, Salmo salar. Aquaculture 161,363–377.

Thiel, U. and Weigand, E. (1992) Influence of dietary zinc and microbial phytasesupplementation on Zn retention and Zn excretion in broiler chicks. XIX World’s PoultryCongress, Amsterdam, The Netherlands, 20–24 September.

Thiel, U., Weigand, E., Schoner, F.J. and Hoppe, P.P. (1993) Zinc retention of broilerchicken as affected by dietary supplementation of zinc and microbial phytase. In: Anke,M., Meissner, D. and Mills, C.F. (eds) Proceedings of 8th International Symposium, TraceElements in Man and Animals (TEMA). Dresden, Gersdorf, pp. 658–659.

Thompson, L.U. (1986) Phytic acid: a factor influencing starch digestibility and blood glucoseresponse. In: Graf, E. (ed.) Phytic Acid: Chemistry and Applications. Pilatus Press,Minneapolis, p. 173.

Thompson, L.U. and Yoon, J.H. (1984) Starch digestibility as affected by polyphenols andphytic acid. Journal of Food Science 49, 1228–1229.

Thompson, L.U., Button, C.L. and Jenkins, D.J.A. (1987) Phytic acid and calcium affectthe in vitro rate of navy bean starch digestion and blood glucose response in humans.American Journal of Clinical Nutrition 46, 467–473.

Tossenberger, J., Liebert, F. and Schulz, E. (1993) Zum einfluss von phytase auf de bauvon phytaten verschiedener herkunfte. In: Vitamine und Weitere Zusatzstoffe bei MenscTier, 4th Symposium. pp. 365–370.

Trotter, M. and Allee, C.L. (1979a) Availability of phosphorus in corn, soybean meal andwheat. Journal of Animal Science 49 (Suppl. 1), 255 (abstract).

Trotter, M. and Allee, C.L. (1979b) Availibility of phosphorus in dry and high-moisture grainfor pigs and chicks. Journal of Animal Science 49 (Suppl. 1), 98 (abstract).

Vahl, H.A., Borggreve, G.J. and Stappers, H.P. (1993) The Effect of Microbial Phytase in LayerFeed. Experimental Report No. 374, CLO-Schothorst (NL), Lelystad, The Netherlands.

Valencia, Z. and Chavez, E.R. (1997) Phytase and acetic acid supplementation of piglet diets:effect on performance and apparent nutrient digestibility. Canadian Journal of AnimalScience 77, 742.

Van der Klis, J.D., Versteegh, H.A.J., Simons, P.C.M. and Kies, A.K. (1997) The efficacy ofphytase in corn–soybean meal-based diets for laying hens. Poultry Science 76, 1535–1542.

Van Loon, A.P.G.M., Simoes-Nunes, C., Wyss, M., Tomschy, A., Hug, D., Vogel, K. andPasamontes, L. (1998) A heat resistant phytase of Aspergillus fumigatus with superiorperformance in animal experiments. Phytase optimisation and natural variability. TheBiochemistry of Phytate and Phytases (in press).

Veum, T.L. (1996) Use of microbial phytase in corn–soybean meal and grain sorghum–canolameal diets for growing–finishing swine. In: Coelho, M.B. and Kornegay, E.T. (eds)Phytase in Animal Nutrition and Waste Management. BASF Corporation, Mount Olive,New Jersey, pp. 365–380.

Vielma, J., Lall, S.P., Koskela, J., Schoner, F.J. and Mattila, P. (1998) Effects of dietaryphytase and cholecalciferol on phosphorus bioavailability in rainbow trout (Oncorhynchusmykiss). Aquaculture 163, 307–321.

Vohra, P., Gray, G.A. and Kratzer, F.H. (1965) Phytic acid–metal complexes. Proceedings ofthe Society of Experimental Biology and Medicine 120, 447.

Wise, A. (1983) Dietary factors determining the biological activities of phytase. NutritionAbstract Review (Section B) 53, 791–806.

270 E.T. Kornegay

Yang, W.J., Matsuda, Y., Sano, S., Masutani, H. and Nakagawa, H. (1991) Purification andcharacterization of phytase from rat intestinal mucosa. Biochimica Biophysica Acta 1075,75.

Yi, Z. and Kornegay, E.T. (1996) Sites of phytase activity in the gastrointestinal tract of youngpigs. Animal Feed Science and Technology 61, 361–368.

Yi, Z., Kornegay, E.T., Lindemann, M.D., Ravindran, V. and Wilson, J.H. (1996a) Effective-ness of Natuphos® phytase in improving the bioavailabilities of phosphorus andother nutrients in soybean meal-based semipurified diets for young pigs. Journal ofAnimal Science 74, 1601–1611.

Yi, Z., Kornegay, E.T., Ravindran, V. and Denbow, D.M. (1996b) Improving phytatephosphorus availability in corn and soybean meal for broilers using microbial phytaseand calculation of phosphorus equivalency values for phytase. Poultry Science 75,240–249.

Yi, Z., Kornegay, E.T. and Denbow, D.M. (1996c) Effect of microbial phytase on nitrogenand amino acid digestibility and nitrogen retention of turkey poults fed corn–soybeandiets. Poultry Science 75, 979–990.

Yi, Z., Kornegay, E.T. and Denbow, D.M. (1996d) Supplemental microbial phytase improveszinc utilization in broilers. Poultry Science 75, 540–546.

Yoon, J.H., Thompson, L.U. and Jenkins, D.J.A. (1983) The effect of phytic acid on invitro rate of starch digestibility and blood glucose response. American Journal of ClinicalNutrition 38, 835–842.

Zhang, Z.B., Kornegay, E.T. and Veit, H.P. (1998a) Comparison of genetically engineeredmicrobial and plant phytase for young broilers. Poultry Science 77 (Suppl. 1), 71.

Zhang, Z.B., Kornegay, E.T. and Veit, H.P. (1998b) Comparison of genetically engineeredmicrobial and plant phytase for young pigs. Journal of Animal Science 76 (Suppl. 1), 178.


T.A. McAllister et al.Enzymes in Ruminant Diets11

Introduction

Exogenous enzymes have been used extensively to remove anti-nutritional factorsfrom feeds, to increase the digestibility of existing nutrients, and to supplement theactivity of the endogenous enzymes of poultry (Classen et al., 1991; Bedford, 1993).Researchers in the 1960s examined the use of exogenous enzymes in ruminants(Burroughs et al., 1960; Rovics and Ely, 1962; Rust et al., 1965), but responses werevariable and no effort was made to determine the mode of action of these products.Furthermore, production of exogenous enzymes was expensive at the time and it wasnot economically feasible to apply these preparations at the concentrations necessaryto elicit a positive animal response. Recent reductions in fermentation costs, togetherwith more active and better defined enzyme preparations, have prompted researchersto re-examine the role of exogenous enzymes in ruminant production (Chen et al.,1995; Beauchemin et al., 1997; McAllister et al., 1999). Several studies haveattempted to define possible modes of action of these additives (Judkins and Stobart,1988; Feng et al., 1996; Hristov et al., 1998a,b; Yang et al., 1999). Exogenousenzymes could exert a number of effects, both on the gastrointestinal microflora andon the ruminant animal itself. It is highly probable, therefore, that physiologicalresponses to exogenous enzymes are multifactorial in origin.

This review will summarize production responses to supplementary exogenousenzymes obtained to date in ruminants. Possible mechanisms by which theseproducts may improve nutrient utilization by ruminants will be discussed andsuggestions will be made with regard to strategies that may further enhance theefficacy of these products for ruminants.


Enzymes in Ruminant DietsT.A. McALLISTER,1 A.N. HRISTOV,1

K.A. BEAUCHEMIN,1 L.M. RODE1

AND K.-J. CHENG2

1Agriculture and Agri-Food Canada, Lethbridge Research Centre,PO Box 3000, Lethbridge, Alberta, Canada, T1J 4B1;

2Faculty of Agricultural Sciences, University of British Columbia,Vancouver, British Columbia, Canada, V6T 1Z4

11

Sources of Enzymes

Although enzyme products marketed for livestock number in the hundreds, they arederived primarily from only four bacterial (Bacillus subtilis, Lactobacillus acidophilus,L. plantarum and Streptococcus faecium) and three fungal (Aspergillus oryzae,Trichoderma reesei and Saccharomyces cerevisiae) species (Muirhead, 1996). Otherfungal species, including Humicola insolens and Thermomyces anuginosus, are beingmarketed to a lesser extent. It is unlikely that this list of source organisms will expandsubstantially, given that no petition to the Food and Drug Administration to add anew organism has been successful (Pendleton, 1996).

Enzymes are naturally occurring biocatalysts produced by living cells to bringabout specific biochemical reactions. In the context of feed additives for ruminants,enzymes are employed to catalyse the degradative reactions by which substrates(i.e. feedstuffs) are digested into their chemical components (e.g. simple sugars,amino acids, fatty acids). These are in turn used for cell growth, either by ruminalmicroorganisms or by the host animal.

Complete digestion of complex feeds such as hay or grain requires literallyhundreds of enzymes. Enzyme preparations for ruminants are marketed primarilyon the basis of their capacity to degrade plant cell walls and, as such, are often referredto as cellulases or xylanases. However, none of these commercial products is a prepa-ration of single enzymes; secondary enzyme activities such as amylases, proteasesor pectinases are invariably present. Degradation of cellulose and hemicellulosealone requires a number of enzymes, and differences in the relative proportions andactivities of these individual enzymes impacts the efficacy of cell wall degradation bythe marketed products. Even within a single microbial species, the types and activityof enzymes produced can vary widely, depending on the strain selected and thegrowth substrate and culture conditions employed (Considine and Coughlan, 1989;Gashe, 1992).

The diversity of enzyme activities present in commercially available enzymepreparations is advantageous, in that a wide variety of substrates can be targeted by asingle product, but it presents problems in terms of quality control and extrapolationof research findings among different preparations. For ruminants, enzyme productsare usually standardized by blending crude enzyme extracts to obtain specifiedlevels of one or two defined enzyme activities, such as xylanase and/or cellulase.These products are not currently standardized for secondary activities. In fact, theseactivities, which may well be affecting the overall effectiveness of a given product, areseldom even measured.

Measurement of Enzyme Activity

Enzyme activity is assayed by measuring over time either the disappearance of adefined substrate or the generation of a product from the biochemical reactioncatalysed by the enzyme. Activities of enzymes for use in the feed industry are mostcommonly measured using the latter approach, and are expressed as the amount of

274 T.A. McAllister et al.

product produced per unit time. These measurements must be conducted underconditions closely defined with respect to temperature, pH, ionic strength, substrateconcentration and substrate type, as all of these factors can affect the activity of anenzyme (Headon, 1993). For example, the relative ranking of cellulase activity ofthree enzyme preparations differs depending on the test substrate (cellulose, carboxy-methylcellulose (CMC) or β-glucan) selected for analysis (Table 11.1). Furthermore,release of reducing-sugars from CMC or xylan is not directly proportional to enzymeconcentration. Consequently, the ratio of enzyme to substrate will affect enzymeactivity estimates (Fig. 11.1) (Hristov and McAllister, unpublished data).

Enzyme activity can also be assessed using synthetic substrates, which usuallyconsist of chromophores linked to molecules chemically similar to the naturalsubstrate. Enzyme activity is measured as the release of the dye or chromophore(Biely et al., 1985). These synthetic substrates offer uniformity among assays, but aresubject to criticism in that they do not represent the substrate found in intact feedssuch as cereal grains or forages. Furthermore, the assays used to assess enzyme activityare not representative of the conditions in the digestive tract, where ultimately thelevel and persistence of enzyme activity may be most important. For these reasons,measurement of enzyme activity using traditional assay techniques may have littlerelevance to the potential efficacy of an enzyme as a feed additive for ruminants.

Researchers have attempted to develop biological assays that may be moreindicative of the value of a given enzyme preparation for ruminants. These methodsusually involve in vitro incubation of enzyme and feed with ruminal contents, andmeasurement of the disappearance of substrates (e.g. cereal grain, straw, hay)representative of those consumed by the animal (Forwood et al., 1990; Varel et al.,1993; Hristov et al., 1996b, 1998a). Alternatively, the amount of gas produced bythe mixed culture can be used as an indication of digestion (Iwaasa et al., 1998),which enables rapid screening of different enzyme products and application rates.These procedures may provide valuable information on the extent to whichexogenous enzymes complement the digestive activity of ruminal microorganisms.However, extrapolation of information from these procedures to whole-animal

Enzymes in Ruminant Diets 275

Substrate1

Product % Protein Cellulose2 CMC3 β-Glucan

123

11.521.315.6

102.1126.1225.6

737.9727.1699.7

1016.41042.11638.3

1For all substrates, activity is expressed as nmol reducing-sugar released mg per product min−1,after a 3 h incubation of the substrate with the enzyme product in 0.2 M sodium phosphate (pH 6.0)buffer at 39°C.2Microcrystalline cellulose.3Carboxymethylcellulose.

Table 11.1. Relative activities of three commercial enzyme preparations on substrates thatreflect cellulase activity. (Adapted from Hristov et al., 1996b.)

situations is limited by: (i) variations in microbial composition among inocula fromdifferent donor animals; (ii) differences in growth of microbial populations in the invitro system versus in the rumen; and (iii) artefactual accumulation of end productsthat alter enzyme activity. Additionally, these assays do not consider the possibleimpact of exogenous enzymes on biological parameters such as feed intake, rate ofpassage or post-ruminal digestion of nutrients.

Because viscosity of intestinal digesta is closely correlated with growth and feedefficiency in poultry, viscosity measurements have been used as a standard for assess-ing the biological value of exogenous carbohydrases for poultry (Sabatier and Fish,1996). For ruminants, however, the value of enzymes can presently be assessed onlythrough expensive and time-consuming production experiments with beef or dairycattle, which makes screening large numbers of products impractical. This lack of anadequate bioassay for assessing the value of exogenous enzymes is perhaps the greatestimpediment to the development of more efficacious enzyme products for ruminants.


Fig. 11.1. Release of reducing-sugars (RS) from xylan and carboxymethylcellulose(CMC) during a 5 h incubation with three commercially available enzymepreparations. Note that the relationship between reducing-sugar release andenzyme concentration is not linear, indicating that concentration of enzymeselected can affect estimates of activity. (Hristov and McAllister, unpublished.)

Production Responses to Exogenous Enzymes

Beef cattle

Evidence that exogenous enzymes could improve average daily gain and feedefficiency in beef cattle was first recorded in a series of ten feeding trials reported 40years ago (Burroughs et al., 1960). When given diets of ground ear maize, oat silage,maize silage or lucerne hay treated with an enzyme cocktail containing amylolytic,proteolytic and cellulolytic activities (Agrozyme , Merck Sharp and DohmeResearch Laboratories), cattle gained 6.8–24.0% more weight and exhibited feedefficiencies improved by 6.0–21.2%, relative to cattle fed untreated control diets.In the same year, four different enzyme preparations (Agrozyme , Zymo-Pabst ,Rhozyme and Takamine ; Merck and Company, Rahway, New Jersey), given incombination with diethylstilbesterol, were shown to increase gain by cattle fed amaize–lucerne hay diet by an average of 14.0% (Nelson and Damon, 1960).

Further studies confirmed that enzyme supplements could improve averagedaily gain (ADG) and feed efficiency in cattle fed high silage diets (Rovics andEly, 1962), but not all responses by feedlot cattle to enzyme supplementation werepositive. Leatherwood et al. (1960) added a fungal enzyme to a grain supplementfor calves fed a lucerne hay-based diet and found no improvement in ADG or feedefficiency. Two enzyme preparations containing primarily amylase and proteaseactivities also failed to increase ADG by cattle given a diet comprising 80%concentrate and 20% chopped lucerne hay (Clark et al., 1961). In a separate study,Agrozyme even reduced the ADG of cattle by 20.4% when it was fed with a maizecarrier to beef cattle given a maize silage diet (Perry et al., 1960). Similarly, Kercher(1960) found that ADG was reduced when Zymo-Pabst was fed with a maizecarrier to cattle given a diet of steam-rolled barley, lucerne hay and maize silage. Perryet al. (1966) attributed an 18.2% decline in ADG observed in cattle fed Agrozyme

with maize cob diets to a 6.8% reduction in feed intake, because the enzyme hadbeen shown to enhance fibre digestibility in metabolism experiments.

Although these early studies provided valuable information on the potentialbenefits of enzymes for beef cattle, they did little to address the impact on animalresponses of factors such as the composition of the diet, types and levels of enzymeactivities present, or the method of enzyme application. More recent studies havebeen designed specifically to address these issues. Different feed types (Beaucheminet al., 1995, 1997), application levels (Beauchemin and Rode, 1996; Michal et al.,1996; McAllister et al., 1999), enzyme products (Pritchard et al., 1996) and enzymeapplication methods (Beauchemin et al., 1998; Hristov et al., 1998b; Yang et al.,1999) have been compared under controlled conditions. Application of differentlevels (0.25–4.0 l t−1) of a mixture of xylanase and cellulase products and cellulaseincreased ADG of steers fed lucerne hay or timothy hay cubes by 30 and 36%,respectively, but had no effect on ADG when applied to barley silage (Beaucheminet al., 1995). When this same mixture was applied to a 95% grain diet, feed efficiencyof cattle fed barley was improved by 11% but performance of cattle fed maize wasunaffected (Beauchemin and Rode, 1996). Application of a different mix of fungal


enzyme preparations at rates up to 5.0 l t−1, however, increased the final weightand ADG of feedlot cattle given diets based on lucerne silage (Michal et al., 1996;Pritchard et al., 1996) or barley silage (McAllister et al., 1999). Treatment of 82.5%maize diets with ‘multiple stabilized enzymes’ increased ADG and feed conversion byfeedlot cattle by 10 and 7.5%, respectively (Weichenthal et al., 1996), and similarimprovements in feed efficiency have been reported for cattle fed sorghum-baseddiets treated with amylase (Krause et al., 1989; Boyles et al., 1992).

Dairy cattle

The effect of exogenous enzymes on milk production in dairy cows was firstexamined in the mid-1990s (Chen et al., 1995; Lewis et al., 1995; Stokes andZheng, 1995) and recently there has been a flurry of research activity in this area(Luchini et al., 1997; Nussio et al., 1997; Kung et al., 1998; Yang et al., 1998, 1999;Beauchemin et al., 1999). As in studies using beef cattle, production responses bydairy cattle to exogenous enzymes have also been variable. Application of enzymemixtures on to sorghum did not improve milk production by Holstein cows (Chenet al., 1995). Similarly, milk responses to exogenous enzymes depended on themethod of application to barley-based diets (Beauchemin et al., 1999). Applying acellulase/xylanase mixture to diets containing 45–50% concentrate and lucernesilage, lucerne hay or a mixture of lucerne silage, maize silage and lucerne hay alsofailed to increase milk yield (Luchini et al., 1997; Nussio et al., 1997). In contrast,spraying two similar enzyme preparations on to maize silage in a 50% concentratediet increased milk production by 2.5 kg day−1 without altering milk composition(Kung, 1996). Although it was demonstrated that these enzyme preparations couldincrease milk production in cows fed lucerne hay/silage-based total mixed rations(Lewis et al., 1995; Stokes and Zheng, 1995; Sanchez et al., 1996), positive responsesin milk production were highly dependent on the level of enzyme applied (Sanchezet al., 1996). For example, research at the Lethbridge Research Centre has shown thatincreasing the amount of enzyme applied to lucerne cubes from 1 g kg−1 to 2 g kg−1

increased milk production from 23.7 kg day−1 (control) to 24.6 kg day−1 and25.6 kg day−1, respectively (Table 11.2) (Yang et al., 1999). Responses to thisenzyme preparation were even more dramatic in early lactation, when treatment of adiet comprising 24% maize silage, 15% lucerne hay and 61% barley-based concen-trate increased milk production by 4 kg day−1. The efficacy of this enzyme was appar-ently dependent upon the method of its application, as spraying the enzyme on to thetotal mixed ration did not affect milk production, whereas applying it to the concen-trate did (4 kg day−1) (Beauchemin et al., 1998).

Lambs

It was shown in the 1960s that feeding a mixture of amylolytic, cellulolytic andproteolytic enzymes (Agrozyme ; 1.5, 3 and 6 g day−1), as well as a potent


proteolytic enzyme (Ficin , Merck and Company; 5, 10 and 20 mg day−1) did notalter feed conversion or the ADG of fattening lambs fed ground maize or lucerne hay(Theurer et al., 1963). Recently, McAllister et al. (2000) have also found thatfibrolytic enzymes did not increase feed intake or ADG by lambs fed lucerne hay- orbarley-based diets.

Learning from animal experiments

The positive effects of exogenous enzymes on growth production both by beef and bydairy cattle have been demonstrated definitively, but the information required toimprove the consistency and increase the magnitude of these responses is unfortu-nately still lacking. Comparisons between experiments are exceedingly difficult,because many enzyme products are poorly defined. Further, several studies haveshown that over-application of enzyme is possible, such that increased applicationcosts are not recovered by corresponding improvements in animal performance(Beauchemin et al., 1996; McAllister et al., 1999). Thus, application of one enzymepreparation at a given concentration provides little information with regard to thepotential effect on animal performance of a different application level, let alone adifferent product. Method of application also influences production responses; theyhave been shown to differ between dry forage, fresh forage and silage (Beauchemin


Diet1

Item Control LH HH HT SE2

Dry matter intake (DMI)kg day−1

% of BWBody weight (kg)Milk yield3 (kg day−1)

Actual4% FCMSCM

Milk composition (%)FatProteinLactose

Milk DMI−1 (kg kg−1)

20.43.29

621.11

b23.7b

b22.4b

b22.2b

3.793.36

b4.56b

1.2

20.73.39

623.11

ab24.6ab

ab22.9ab

ab23.2ab

3.703.41

ab4.61ab

1.22

20.73.32

626.11

b25.6a

b24.6a

b24.4a

3.783.48

ab4.60ab

1.29

20.83.42

619.11

ab25.3ab

b24.2a

b24.2a

3.763.49

b4.62a

1.25

0.70.143.11

0.60.70.7

0.110.110.110.1

abWithin a row, means bearing unlike superscripts differ (P < 0.05).1LH = low level of enzyme added to cubed lucerne hay; HH = high level of enzyme added to cubedlucerne hay; HT = enzyme added to both cubed lucerne hay and concentrate.2SE = standard error.3FCM = fat-corrected milk; SCM = solids-corrected milk.

Table 11.2. Dry matter intake and yield and composition of milk from lactating cows fedenzyme-treated diets.

et al., 1995; Feng et al., 1996), and if the enzyme is infused directly into the rumen,or applied to the complete diet or to the concentrate component only (Lewis et al.,1996; Beauchemin et al., 1998; McAllister et al., 1999). It is obvious that manyfactors may influence enzyme efficacy in ruminants. Therefore, an understanding ofthe modes of action by which enzymes improve nutrient utilization in ruminants iskey to obtaining consistent positive responses to exogenous enzymes over a broadrange of diets and animal types.

Modes of Action

Upon initial consideration, exogenous enzymes might be expected to alter feed utili-zation in ruminants either through their effects on the feed prior to consumption, orthrough their enhancement of digestion in the rumen and/or in the post-ruminaldigestive tract (Fig. 11.2). In actuality, all of these possible modes of action are inter-twined and enzyme-mediated alteration of the feed prior to consumption likely hasramifications on ruminal and post-ruminal digestion of nutrients. Preconsumptiveeffects of exogenous enzymes may be as simple as the release of soluble carbohydrateor as complex as the removal of structural barriers (Fig. 11.2A) that limit themicrobial digestion of feed in the rumen. Within the rumen, exogenous enzymescould act directly on the feed or could indirectly stimulate digestive activity throughsynergistic effects on ruminal microorganisms (Fig. 11.2B). Exogenous enzymes mayremain active in the lower digestive tract, contributing to the post-ruminal digestionof fibre, or they could indirectly improve nutrient absorption in the lower tract byreducing viscosity of intestinal digesta (Fig. 11.2C). They may also supplementenzyme activity in the faeces, thereby contributing by accelerating decomposition ofwaste (Fig. 11.2D). Ultimately, the goal of enzyme supplementation is to improvethe efficiency of feed utilization in ruminants and reduce waste production.Undoubtedly, the mode of action of exogenous enzymes in ruminants is exceedinglycomplex and continues to be a major focus of the research presently being conductedwith these additives.

Preconsumption effects

There is ample evidence that exogenous enzymes can release reducing-sugarsfrom feedstuffs prior to consumption (Beauchemin and Rode, 1996; Hristov et al.,1996a,b). The degree of sugar release depends on both the type of feed and thetype of enzyme. For example, only two of 11 enzyme preparations tested releasedsignificant amounts of reducing-sugars from barley silage (Hristov et al., 1996a).Moreover, the preparations most effective at releasing reducing-sugars from lucernehay were not those that released the most reducing-sugars from barley silage. There isevidence that exogenous enzymes may be more effective when applied to dry forageas opposed to wet forage (Feng et al., 1996; Beauchemin et al., 1998). At first thisseems improbable, given that the role of water in the hydrolysis of soluble sugars


from complex polymers is a fundamental biochemical principle (Lehninger, 1982).However, feed offered to ruminants is seldom absolutely dry; even feeds thatnutritionists would describe as ‘dry’ (e.g. grain, hay) contain 6–10% moisture.Release of soluble sugars from these ‘dry’ feeds suggests that their water content issufficient to enable hydrolysis.

Release of sugars from feeds arises at least partially from the solubilization ofNDF and ADF (Hristov et al., 1996a; Gwayumba and Christensen, 1997). This isconsistent with observed increases in the soluble fraction and rate of in situ digestion(Feng et al., 1996; Hristov et al., 1996a; Dong, 1998; Hristov et al., 1998a; Yanget al., 1999). However, most studies have not found that exogenous enzymesimprove the extent of in situ or in vitro DM digestion (Feng et al., 1996; Hristovet al., 1996a). These results suggest that enzyme additives only degrade substratesthat would be naturally digested by the endogenous enzymes of the rumenmicroflora. Using scanning electron microscopy, we have observed that highconcentrations of fibrolytic enzymes can cause digestive pits in the cell walls of bar-ley straw (Fig. 11.3). However, no similar degradation was observed when enzymes


Fig. 11.2. Possible modes of action of exogenous enzymes in ruminants. (A) Priorto consumption, exogenous enzymes may partially digest feed or weaken structuralbarriers that impede microbial digestion in the rumen. (B) In the rumen, exogenousenzymes may hydrolyse feed directly or work synergistically with ruminal micro-organisms to enhance feed digestion. (C) In the small intestine, exogenous enzymesmay improve nutrient absorption by reducing intestinal viscosity, or by hydrolysingsubstrates that escape ruminal digestion. (D) In the faeces, exogenous enzymes mayincrease the rate of decomposition.

were applied at the concentration recommended by the manufacturer (McAllisteret al., unpublished data).

Although exogenous enzymes do effect release of soluble carbohydrates, theamount liberated represents only a minute portion of the total carbohydrate presentin the diet. It is difficult to attribute observed enzyme-associated productionresponses solely to the generation of soluble carbohydrates prior to consumption,given that comparable increases in yield were not seen when up to 9% of total dietaryDM was supplied as molasses (Wing et al., 1988). Additionally, there is ample evi-dence that through associative effects soluble carbohydrates can actually depress fibredigestion in ruminants (Huhtanen, 1991). Some exogenous enzyme preparationscontain soluble carbohydrates, which can affect in vitro gas production (Varel et al.,1993), but it is unlikely that they contribute significant levels of soluble carbohydrateto the feed at practical application levels. Thus, it is most likely that at least a portionof the positive production responses observed to accompany enzyme supplements isdue to alterations in ruminal or post-ruminal digestion.

Ruminal effects

Direct hydrolysisUntil recently it was assumed that, upon introduction into the rumen, exogenousenzymes would be rapidly degraded by the array of proteases produced by ruminal


Fig. 11.3. Scanning electron micrographs of barley straw, (a) untreated or (b)incubated in a 1 : 10 dilution of concentrated enzyme product containingcellulases and xylanases. Note that this exceedingly high concentration of theenzyme caused visible degradation of the barley straw. In other studies, enzymeswere applied at levels recommended by the manufacturer (e.g. 2.0 l t−1) and novisible damage of the surface of the straw was observed (data not shown).Bars = 250 µm. (McAllister, Bae and Cheng, unpublished.)

microorganisms (Kung, 1996). Indeed, fungal cellulases incubated with ruminalfluid were rapidly degraded to the extent that after 6 h of incubation, less than25% of their original activity remained (Kopency et al., 1987; Vandevoorde andVerstraete, 1987). However, experimentation with other enzyme products showedthat CMCase activity and xylanase activity remained constant after 6 h of incubationwith ruminal fluid (Hristov et al., 1998b). Furthermore, exogenous enzymesincreased xylanase and cellulase activity in the rumen (Hristov et al., 1998a,b, 2000)and there is evidence that declining exogenous enzyme activity in ruminal fluid isassociated both with inactivation of the enzymes and with their outflow with thefluid phase of ruminal contents (Fig. 11.4) (Hristov et al., 1996b).

The fact that exogenous enzymes remain active in the rumen raises thepossibility that they may improve digestion through the direct hydrolysis of ingestedfeed. Several researchers have shown that exogenous enzymes can enhance fibre


Fig. 11.4. Decline in endo-cellulase and β-glucanase activities in the rumenduring a 15 h period following administration of two enzyme products directly intothe rumen. Note that decline of both enzyme activities differs between the twoproducts, and that the declines are related to the passage of fluid from the rumen,as measured by the decline in Cobalt-EDTA. RS, reducing-sugars. (Adapted fromHristov et al., 1996b.)

degradation by ruminal microorganisms in vitro (Forwood et al., 1990; Varel et al.,1993; Feng et al., 1996; Hristov et al., 1996a; Dong et al., 1999) and in situ (Lewiset al., 1996). This effect has been confirmed in some (Beauchemin et al., 1999; Yanget al., 1999) but not in all (Firkins et al., 1990; Varel and Kreikemeier, 1994) studiesconducted using cattle with ruminal and duodenal cannulae. Although addingexogenous enzymes may increase the activity of xylanases and cellulases in ruminalfluid, enzyme activity in the fluid usually represents less than 30% of the totalenzyme activity in the rumen, the remainder being associated with the feed particles(Minato et al., 1966; Brock et al., 1982). For example, applying fibrolytic enzymes toa grass hay diet for sheep prior to consumption increased endo-glucanase activity andxylanase activity in ruminal fluid, but this activity accounted for only 0.5% of thetotal endo-glucanase activity in the rumen (Table 11.3) (Dong, 1998). Given thatexogenous enzymes represent only a small fraction of the ruminal enzyme activity,and that the ruminal microbiota are inherently capable of readily digesting fibre(McAllister et al., 1994), it is difficult to envision how exogenous enzymes wouldenhance ruminal fibre digestion through direct hydrolysis.

Synergism with ruminal microorganismsEnhancement of fibre digestion in the rumen would seem more feasible if theseproducts are working synergistically with ruminal microbes. Logically, this conceptimplies that exogenous enzyme preparations contain enzymatic activities that wouldnormally be limiting to digestion of plant cell walls by ruminal microorganisms.


% Total ruminal activity

CON ENZ CON ENZ

Activity in liquid phase (units ml−1 h−1)EndoglucanaseXylanase

Activity in particulate phase (units g DM−1 h−1)EndoglucanaseXylanase

Total activity in liquid phase (units × 103)EndoglucanaseXylanase

Total activity in particulate phase (units × 103)EndoglucanaseXylanase

a0.026a

a0.315a

3.142.37

0.141.76

3.7930.5

a0.029b

a0.344b

3.1228.27

0.1651.95

3.4528.54

––

––

3.65.4

96.494.6

––

––

4.66.4

95.493.6

a,bWithin a row, means bearing unlike superscripts differ (P < 0.05).1Endoglucanase activity was standardized against a commercial enzyme preparation from P.funiculosum (EC 3.2.1.4, Sigma Chemical Co., St Louis, Missouri) and xylanase activity wasstandardized against a commercial xylanase activity from A. niger (EC 3.2.18, Sigma Chemical).Incubations were conducted in sodium phosphate buffer (pH 6.5) at 39°C for 30 min.

Table 11.3. Effect of exogenous enzymes on endoglucanase and xylanase activities1 in therumen of sheep fed grass hay. (Adapted from Dong, 1998.)

Limitations to plant cell wall digestion in the rumen could result from insufficientquantities or types of enzyme production by the ruminal microbes, from an inabilityof degradative enzyme(s) to interact with the target substrates, or from conditions inthe rumen not being optimal for enzyme activity (e.g. low ruminal pH). At least 21different enzymatic activities have been identified as being involved in the hydrolysisof the structural polysaccharides of the plant cell wall, all of which are produced bya normally functioning ruminal microflora (White et al., 1993). Researchers haveshown that extracts from Aspergillus oryzae can increase the number of ruminalbacteria (Newbold et al., 1992a,b) and can work synergistically with extracts fromruminal microorganisms to enhance release of soluble sugars from hay (Newbold,1995). The extent of cross-linking by p-coumaryl and feruloyl groups toarabinoxylans has been identified as one factor that limits the digestion of plantcell walls (Hatfield, 1993). A. oryzae has been shown to produce an esterase capableof breaking the ester bridges formed between ferulic and p-coumaric acids andarabinoxylan (Tenkanen et al., 1991); and these activities have been proposed tofunction synergistically with ruminal microorganisms (Varel et al., 1993). However,many of the ruminal fungi (e.g. Neocallimastix spp.: Borneman et al., 1990) andsome species of ruminal bacteria (e.g. Fibrobacter succinogenes: McDermid et al.,1990; Butyrivibrio fibrisolvens: Dalrymple et al., 1996) produce esterases capable ofhydrolysing linkages with phenolic acids. The fact that exogenous enzymes usuallyonly increase the rate and not the extent of digestion (Varel et al., 1993; Feng et al.,1996; Hristov et al., 1996a) suggests that the activities contributed by these additivesare not novel to the ruminal environment. Recent work in which oligonucleotide16S rRNA probes are used to study rumen ecology suggests that there may be severalspecies of fibrolytic ruminal bacteria (e.g. clostridia, ruminococci) that have yet to becultured in the laboratory (Forster and Whitford, 1998). If this is the case, thesemicroorganisms may also be contributing to the diverse array of enzyme activitiesrequired for efficient digestion of plant cell walls.

Hydrolysis of cellulose and hemicellulose is accomplished either by free enzymesor by cellusomal structures comprising multiple enzymes bound non-covalently toform an organized complex (Teeri, 1997). Aerobic fungi, the principal commercialsource of exogenous enzymes, hydrolyse plant cell walls by means of free enzymes,whereas hydrolysis of plant cell walls by Clostridium spp. and the anaerobic ruminalfungi involves cellulosomal structures (Beguin et al., 1998; Ljungdahl et al., 1998).There is also evidence that the ruminococci may rely on a cellulosome-likemultienzyme complex for fibre degradation (Flint et al., 1998; Ohara et al., 1998).Destruction of these multienzyme complexes during the extraction process mayexplain why enzymes from mixed ruminal microorganisms failed to release as muchsoluble sugar from hay and straw as extracts from A. oryzae (Newbold, 1995).

Evidence has also been presented that cellulosomes may play a role in adhesionof microbial cells to their substrates (Pell and Schofield, 1993; Beguin et al., 1998).The process of adhesion is absolutely essential for efficient digestion of foragesand cereal grains in the rumen (McAllister et al., 1994; McAllister and Cheng, 1996).Cellulose-binding domains may be involved in the attachment of ruminal bacteria tocellulose (Pell and Schofield, 1993). Presently it is not known if exogenous enzymes


block microbial adhesion sites on the surface of feeds, or expose additional ones.Administering an aqueous solution of mixed fibrolytic enzymes (3.3% vol/vol)directly into the rumen of cannulated sheep actually lowered DM digestion(McAllister et al., 1999), and similar results were reported when enzymes wereinfused into beef steers (Lewis et al., 1996). These studies suggest that enzymes arenot as efficacious when ruminally infused as when applied directly to the feed.Alternatively, this response may be enzyme dependent, given that administration ofother enzymes directly into the rumen did not affect DM digestion (Hristov et al.,2000). Treating lucerne cubes with exogenous enzymes prior to consumptionincreased bacterial colonization and in situ DM disappearance of forage between 3 hand 24 h of ruminal incubation (Yang et al., 1999). These responses were supportedby a concurrent numerical increase in digestibility of fibre in the rumen, and bysignificantly increased digestibility of fibre in the total gastrointestinal tract (Yanget al., 1999).

Considering the low fibre content of high concentrate diets, it is surprisingthat fibrolytic enzymes have improved feed digestion (Krause et al., 1998) andperformance of cattle fed high cereal grain diets (Beauchemin et al., 1997; Iwaasaet al., 1997). An explanation of this phenomenon may come from comparing thepH optima of the fibrolytic enzymes produced by ruminal microorganisms withthe pH optima of exogenous fibrolytic enzymes produced by aerobic fungi. It is welldocumented that growth of fibrolytic bacteria is inhibited (Russell and Dombrowski,1980), and that fibre digestion is severely compromised (Hoover et al., 1984) whenpH falls below 6.2. Most of the fibrolytic enzymes produced by ruminal micro-organisms function optimally at pH values above 6.2 (Greve et al., 1984; Matteand Forsberg, 1992). In contrast, the pH optima of fibrolytic enzymes produced byaerobic fungi typically range from 4.0 to 6.0 (Gashe, 1992; Muzakhar et al., 1998).This point is illustrated by an experiment in which the extent to which Trichodermalongibrachiatum enzymes enhanced gas production was shown to increase as the pHdeclined from 6.5 to 5.5 (Table 11.4) (Morgavi et al., unpublished data). Further,although a decline in pH from 6.5 to 5.5 reduced DM disappearance from maizesilage in mixed ruminal cultures supplemented with T. longibrachiatum enzymes, thenegative effect of low pH on DM disappearance was more pronounced in the absenceof added enzyme (Table 11.4). Ruminal pH can be below 6.0 for a significantportion of the day in dairy cattle (Nocek, 1998; Yang et al., 1999) and in feedlotcattle (Krause et al., 1998). Under these conditions, exogenous enzymes could makea significant contribution to ruminal fibre digestion. The higher fibre content ofbarley, as compared with maize, may explain why exogenous enzymes improved feedconversion in finishing cattle fed barley grain but did not affect feed conversion byfinishing cattle fed maize (Beauchemin et al., 1997).

Some evidence also suggests that non-enzymatic factors in crude enzyme extractsmay work synergistically with ruminal microorganisms. In early experiments withfungal cellulases, boiling the extract for 20 min failed to reduce its effect on cellulosedigestion in vitro (Bowden and Church, 1959). In that study, adding valine orproline to the incubation medium enhanced cellulose digestion to the same extentas adding fungal cellulases. A more recent study showed that autoclaved A. oryzae


extract (8% vol/vol) enhanced cell wall degradation in vitro (Varel et al., 1993). Theresearchers attributed the effect to the presence of soluble substrate in the extract,but concluded that it would not be relevant at the concentration of extract(i.e. 0.067 mg ml−1) expected at recommended in vivo dosages. The effects of non-enzymatic factors are likely expressed more in vitro than in vivo, because microbialpopulations have limited adaptive capabilities in vitro, and they are exposed to higherconcentrations of the additive due to the absence of flow or dilution. Enzyme extractsoften contain preservatives to prolong their shelf life, as well as emulsifying agents(e.g. surfactants) that maintain the enzymes in suspension and facilitate applicationof the product to the feed. Work in our laboratory has shown that surfactants canalter microbial activity and feed degradation in the rumen (McAllister et al., 2000).Unfortunately, few enzyme manufacturers list the non-enzymatic componentsof their products and the consequent difficulty in distinguishing between non-enzymatic and enzymatic effects of enzyme products continues to hamper progresstoward defining the mode of action of these products.

Digesta flowEnzymes have been shown to enhance (Feng et al., 1996; Dong, 1998) and to exertno effect on (Beauchemin et al., 1999; Yang et al., 1999) the rate of passage ofparticulate matter from the rumen. Increases in passage rate can be associated with afaster rate of particle size reduction in the rumen and a corresponding increase in feedintake (Mertens et al., 1984). Although exogenous enzymes have improved bothintake and digestion in some experiments (Stokes and Zheng, 1995; Sanchez et al.,1996), in other srudies enhanced digestion is not associated with an increase in feedconsumption or particulate passage rate (Judkins and Stobart, 1988; Krause et al.,1998; Kung et al., 1998; Beauchemin et al., 1999). These studies suggest that someportion of the effects of exogenous enzymes may be post-ruminal.


Enzyme

Item pH No enzyme Autoclaved Unautoclaved

Gas production (ml)

DM disappearance (%)

6.56.05.56.56.05.5

9.4a

a7.3a

a6.4a

32.1a

a23.6a

a23.2a

10.4a

8.0a

7.1a

31.5a

23.5a

22.8a

11.9a

9.8b

9.0b

36.2b

31.8b

32.7b

a,bWithin a row, means bearing unlike superscripts differ (P < 0.05).1Consecutive batch culture techniques were used to adapt mixed ruminal microorganisms to eachrespective pH prior to incubation (Morgavi et al., unpublished data).

Table 11.4. Effect of pH and T. longibrachiatum enzyme preparations on in vitro gas productionand dry matter disappearance from maize silage during 48 h of incubation with mixed ruminalcultures.1

Post-ruminal effects

Experiments in our laboratory have shown that exogenous enzymes not onlyheighten fibrolytic activity in the rumen, but also increase fibrolytic activity in thesmall intestine (Hristov et al., 1998a,b, 2000). This phenomenon was particularlyevident with xylanase activity. Supplementary enzymes increased duodenal xylanaseactivity by 30% (Hristov et al., 1998a). In the same study, enzyme supplementationincreased cellulase activity at the small intestine by only 2–5%, because theseenzymes were largely inactivated by the low pH and pepsin in the abomasum. Otherresearchers have reported that xylanases from both mesophilic and thermophilicmicroorganisms resist proteolysis (Fontes et al., 1995), possibly due to their highdegree of glycosylation (Gorbacheva and Rodionova, 1977), but our work has shownthat acidity, not pepsin, is the primary factor responsible for the inactivation ofxylanases in the abomasum (Hristov et al., 1998a). At pH 2.0–2.6 (Sturkie, 1970)gastric fluids in the gizzard of poultry are typically less acidic than digesta in thestomach of swine or in the abomasum of ruminants, and they have shorter residencetimes. This difference in gastrointestinal physiology may explain why positiveresponses to enzyme supplementation are more consistent in poultry than in swine orruminants (Campbell and Bedford, 1992; Baas and Thacker, 1996).

Administering large quantities of enzymes (400 g day−1) directly into the rumencan significantly increase cellulase activity in the small intestine (Hristov et al., 2000).Exogenous enzymes supplemented in this manner flow mainly with the fluidphase of ruminal contents (Hristov et al., 2000) and at these elevated feeding levelsa portion of the cellulases even escapes inactivation by the low pH and pepsin in theabomasum. We have observed that increased xylanase activity in the small intestine isassociated with a decline in intestinal viscosity (Table 11.5) (Hristov et al., 1998b,2000). Because viscosity of duodenal digesta increases with increasing levels of grainin the diet (Mir et al., 1998), enzyme-mediated reductions in viscosity could improvenutrient absorption in the small intestine of cattle fed grain diets. Reduced intestinalviscosity was associated with 1.2% and 1.5% increases in total tract digestibilityof DM when enzymes were applied to the feed or infused into the abomasum,respectively (Hristov et al., 1998a). However, intestinal viscosity in cattle is onlybetween 1 and 2 cP (Mir et al., 1998) whereas intestinal viscosity in poultrymay exceed 400 cP (Bedford, 1993). Improved growth performance in poultrysupplemented with enzymes is often associated with tenfold reductions in intestinalviscosity (Bedford, 1993; Graham, 1996). Consequently, it is difficult to compre-hend how the relatively modest declines in intestinal viscosity observed in ruminantssupplemented with high levels of enzymes results in a substantial improvement innutrient absorption in the small intestine.

In studies with dairy cows fed barley grain diets, improvements in total tractdigestion were attributed primarily to an improvement in the digestibility of fibreand starch in the lower tract (Beauchemin et al., 1999). Hydrolysis of complex carbo-hydrates by exogenous enzymes in the small intestine and subsequent absorption ofreleased sugars would offer energetic and nitrogen balance benefits to the animal thatwould not be accessible if these substrates remained undigested or were fermented by


microbial populations residing in the large intestine. It is possible that exogenousenzymes work synergistically with the microbes even in the large intestine, giventhat we have determined that xylanase activity in the faeces increases linearly withincreasing levels of enzyme supplementation (Table 11.5) (Hristov et al., 2000). Thisheightened fibrolytic activity may have important ramifications with regard to therate at which faecal material decomposes in the environment.

Steps toward Improving Exogenous Enzymes for Ruminants

Matching the enzyme to the feed

Not all exogenous enzymes are equally effective at digesting complex substratessuch as lucerne and barley grain (Table 11.6). Feedstuffs are exceedingly complexstructurally, and lack of knowledge of the factors that limit the rate and extent of feed


Experiment 12 Experiment 23

Item Control EF EA Control EFL EFM EFH P4

Ruminal contentsCMCaseXylanaseAmylaseViscosity (mPa s)

Duodenal contentsCMCaseXylanaseAmylaseViscosity (mPa s)

FaecesCMCaseXylanaseAmylase

Total tract digestibility (%)

42.7215.2

a190.1a

AB3.16AB

NDa3.6a

0.981.73

NMNMNM80.2

51.9298.1

ab160.5ab

B2.99B

0.75a45.1b

0.821.56

NMNMNM81.4

41.1208.1b95.9b

A3.34A

2.20b190.1b

0.771.54

NMNMNM81.9

43.3179.1144.1

3.33

0.18ND0.921.75

30.7658.5447.3

74.9

48.0237.3138.1

3.06

0.796.720.871.53

19.6686.0479.2

76.2

48.6264.0137.9

2.90

1.8913.7

0.981.73

26.5832.1512.1

75.3

52.9299.1144.0

2.74

6.3237.4

2.881.49

46.0900.0450.2

74.5

******NS***

************

NS***NSNS

1Expressed as nmol reducing-sugars released ml−1 min−1.2EF, enzyme applied to feed; EA, enzyme infused into abomasum. Adapted from Hristov et al.(1998).3Enzyme introduced directly into the rumen at rates of 0 (Control) or 100, 200 and 400 g day−1

(EFL, EFM and EFH, respectively). Adapted from Hristov et al. (2000).4Linear effect of enzyme supplementation on enzyme activity (P < 0.001).abWithin a row and experiment, means bearing unlike superscripts differ (P < 0.05).ABWithin a row and experiment, means bearing unlike superscripts differ (P < 0.10).NS, not significant (P < 0.05).

Table 11.5. Enzyme activity1 in duodenal digesta and faeces of cattle with and without enzymesupplementation.

digestion impedes the engineering of enzyme preparations designed to overcomeconstraints to feed digestion. With some feeds, specific targets can be identified. Inmaize, for example, the protein matrix surrounding the starch granules, as opposed tothe properties of the starch itself, dictates the extent and rate of starch digestion inthat grain (McAllister et al., 1993). Thus, exogenous enzymes designed to improvethe utilization of maize should contain proteases capable of digesting the proteinmatrix and exposing starch granules to digestion by endogenous ruminal or hostenzymes. An exogenous preparation containing amylase but not protease activitywould not be expected to substantially improve utilization of maize by ruminants. Instraw, the major barriers to microbial digestion are apparently silica, wax and cutin(Bae et al., 1997), whereas in unprocessed grain the pericarp dictates the extent ofgrain digestion (Wang et al., 1998). Enzyme preparations that digest the structuralcomponents limiting the extent of feed digestion in the rumen would be expected tobe more efficacious than preparations that just increase the rate of feed degradation inthe rumen. Many enzyme preparations are currently in use, with no attempt beingmade to define the types or activities of the enzymes they contain. Such randomemployment of enzymes on feeds, without consideration for specific substratetargets, will only discourage or delay adoption of exogenous enzymes for morestandard use in the animal production industry. Ultimately, enzyme cocktailsshould be designed specifically to overcome the constraints limiting digestion ofa particular type of feed. Component enzymes in such cocktails might vary evenfor a given forage, targeting particular maturity levels and structural barriers. Recentdevelopments in biotechnology make it feasible to engineer such enzyme cocktailscontaining xylanase and β-glucanase activities, but we presently lack the technologyfor specific production of many other potentially important enzymes (e.g. cutinase,ferulic acid esterase, acetylxylan esterase, arabinofuranosidase).


Substrate

Product Lucerne hay Hull-less barley

ABCDEFGHIJ

379.0102.8122.3106.4

30.231.0

134.7156.1170.7

62.8

4980.321384.41785.51201.8

387.0489.5

1808.72780.32424.92558.3

1Activity is expressed as p.p.m. of glucose released by the product (present at 0.250 mg ml−1)under standardized test conditions (Hristov and McAllister, unpublished observations).

Table 11.6. Release of reducing-sugars1 from lucerne hay and hull-less barley by a number ofcommercially available fibrolytic enzyme products.

Lowering enzyme cost

Once the mechanisms of action and specific targets for degradative exogenousenzymes have been identified, steps can be taken to optimize the application of thesepreparations. Application concentrations can be minimized by ensuring that thepreparations contain the enzyme activities most likely to elicit an improvement infeed utilization. In some instances, the activity of crude enzyme preparations may beincreased by including specific enzymes most likely to overcome the constraints tofeed digestion. Recently, application of recombinant DNA technology has enabledmanufacturers to increase the volume and efficiency of enzyme production, and tocreate new products (Ward and Conneely, 1993; Hodgson, 1994). Genes encodingsuperior enzymes can be transferred from organisms such as anaerobic bacteria andfungi, typically impractical for commercial production, into well-characterizedindustrial microbial production hosts (e.g. Aspergillus and Bacillus spp.). Expressionof genes encoding novel enzymes in plants such as canola and potato may be a partic-ularly effective means of lowering the cost of enzyme production (Pen et al., 1993;van Rooijen and Moloney, 1994). At the Lethbridge Research Centre, a β-glucanasegene from the ruminal bacterium F. succinogenes has been expressed in a number oflines of potato, and a tenfold difference in level of expression has been observedamong lines (Table 11.7). The line exhibiting the highest expression levels is nowbeing evaluated in diets for poultry. It is possible that similar technologies (e.g.grasses expressing cutinase, esterase) may have applications in ruminant production.

Conclusion

The use of feed enzymes in the monogastric animal production industry hasincreased dramatically in recent years and numerous commercial products arepresently being marketed. In many instances, the mechanisms and modes of action


Glucanase (as % of total protein)

Line Leaf tissue Tuber tissue

1234567

0.0110.0160.0200.0310.0670.1050.047

0.0040.0050.0110.023

–0.0500.021

1β-glucanase from F. succinogenes was expressed in potato using the cauliflower mosaic virus(CAMV) 35S promoter (Armstrong et al., unpublished observations).

Table 11.7. Accumulation of glucanase in the leaf and tuber tissue of seven transgenic lines1 ofpotato (Solanum tuberosum).

of these preparations have been defined. In contrast, few commercial preparations ofexogenous enzymes exist for ruminants and many of these just now are entering themarketplace. Although positive responses in animal performance have been observed,results have been inconsistent. Characteristics of the ruminant digestive tract(e.g. complex microbial populations producing numerous endogenous enzymes)complicate the elucidation of the mechanisms of exogenous enzyme action inruminants. There is evidence that exogenous enzymes initiate digestion of feeds priorto consumption and that they can improve feed digestion in the rumen and lowerdigestive tract. The challenge for researchers is to determine the modes of action,singly or in combination, that enable exogenous enzymes to improve feed efficiencyand increase growth and milk production.

References

Baas, T.C. and Thacker, P.A. (1996) Impact of gastric pH on dietary enzyme activity andsurvivability in swine fed β-glucanase supplemented diets. Canadian Journal of AnimalScience 76, 245–252.

Bae, H.D., McAllister, T.A., Kokko, E.G., Leggett, F.L., Yanke, L.J., Jakober, K.D., Ha, J.K.,Shin, H.T. and Cheng, K.-J. (1997) Effect of silica on the colonization of rice straw byruminal bacteria. Animal Feed Science and Technology 65, 165–181.

Beauchemin, K.A. and Rode, L.M. (1996) Use of feed enzymes in ruminant nutrition. In:Rode, L.M. (ed.) Animal Science Research and Development – Meeting Future Challenges.Minister of Supply and Services Canada, Ottawa, Ontario, pp. 103–131.

Beauchemin, K.A., Rode, L.M. and Sewalt, V.J.H. (1995) Fibrolytic enzymes increase fiberdigestibility and growth rate of steers fed dry forages. Canadian Journal of Animal Science75, 641–644.

Beauchemin, K.A., Jones, S.D.M., Rode, L.M. and Sewalt, V.J.H. (1997) Effects of fibrolyticenzyme in corn or barley diets on performance and carcass characteristics of feedlot cattle.Canadian Journal of Animal Science 77, 645–653.

Beauchemin, K.A., Rode, L.M., Yang, Z. and McAllister, T.A. (1998) Use of feed enzymesin ruminant nutrition. Proceedings of the 33rd Pacific Northwest Nutrition Conference.Vancouver, British Columbia, pp. 121–135.

Beauchemin, K.A., Yang, W.Z. and Rode, L.M. (1999) Effects of grain source and enzymeadditive or grain source on site and extent of nutrient digestion in dairy cows. Journal ofDairy Science 82, 378–390.

Bedford, M.R. (1993) Mode of action of feed enzymes. Journal of Applied Poultry Research 2,85–92.

Beguin, P., Chauvaux, S., Guglielmi, G., Matuschek, M., Leibovitz, E., Chaveroche, M.-K.,Miras, I., Alzari, P. and Gounon, P. (1998) The Clostridium thermocellum cellulosome:organization and mode of attachment to the cell. MIE Bioforum, Genetics, Biochemistryand Ecology of Cellulose Degradation. Suzuka, Japan, p. 107.

Biely, P., Mislovicova, D. and Toman, R. (1985) Soluble chromogenic substrates for theassay of endo-1,4-β-xylanases and endo-1,4-β-glucanases. Analytical Biochemistry 144,142–146.

Borneman, W.S., Hartley, R.D., Morrison, W.H., Akin, D.E. and Ljundahl, L.G. (1990)Feruloyl and p-coumaroyl esterase from anaerobic fungi in relation to plant cell walldegradation. Applied Microbiology and Biotechnology 33, 345–351.


Bowden, D.W. and Church, D.C. (1959) Effect of several enzyme concentrates on cellulosedigestion in the artificial rumen. Journal of Animal Science 18, 1523.

Boyles, D.W., Richardson, C.R., Robinson, K.D. and Cobb, C.W. (1992) Feedlot perfor-mance of steers fed steam-flaked grain sorghum with added enzymes. Proceedings of theWestern Section American Society of Animal Science. Fort Collins, Colorado, 43, 502–505.

Brock, F.M., Forsberg, C.W. and Buchanan-Smith, J.G. (1982) Proteolytic activity of rumenmicroorganisms and effects of proteinase inhibitors. Applied Environmental Microbiology44, 561–569.

Burroughs, W., Woods, W., Ewing, S.A., Greig, J. and Theurer, B. (1960) Enzyme additionsto fattening cattle rations. Journal of Animal Science 19, 458–464.

Campbell, G.L. and Bedford, M.R. (1992) Enzyme applications for monogastric feeds.A review. Canadian Journal of Animal Science 72, 449–466.

Chen, K.H., Huber, J.T., Simas, J., Theurer, C.B., Yu, P., Chan, S.C., Santos, F., Wu, Z. andSwingle, R.S. (1995) Effect of enzyme treatment or steam-flaking of sorghum grain onlactation and digestion in dairy cows. Journal of Dairy Science 78, 1721–1727.

Clark, J.D., Dyer, I.A. and Templeton, J.A. (1961) Some nutritional and physiological effectsof enzymes for fattening cattle. Journal of Animal Science 20, 928.

Classen, H.L., Graham, H., Inborr, J. and Bedford, M.R. (1991) Growing interest in feedenzymes to lead to new products. Feedstuffs 63, 22–24.

Considine, P.J. and Coughlan, M.P. (1989) Production of carbohydrate-hydrolyzingenzyme blends by solid-state fermentation. In: Coughlan, M.P. (ed.) Enzyme Systems forLignocellulose Degradation. Elsevier Applied Science, New York, pp. 273–281.

Dalrymple, B.P., Swadling, Y., Cybinski, D.H. and Xue, G.P. (1996) Cloning of a geneencoding cinnamoyl ester hydrolase from the ruminal bacterium Butyrivibrio fibrisolvensE14 by a novel method. FEMS Microbiology Letters 143, 115–120.

Dong, Y. (1998) Reducing methane emissions from ruminant animals. PhD dissertation,University of Alberta, Edmonton, Alberta.

Dong, Y., Bae, H.D., McAllister, T.A., Mathison, G.W. and Cheng, K.-J. (1999) The effectof exogenous fibrolytic enzymes, α-bromoethanesulfonate and monensin on digestibilityof grass hay and methane production in the Rusitec. Canadian Journal of Animal Science79, 491–498.

Feng, P., Hunt, C.W., Pritchard, G.T. and Julien, W.E. (1996) Effect of enzyme preparationson in situ and in vitro degradation and in vivo digestive characteristics of maturecool-season grass forage in beef steers. Journal of Animal Science 74, 1349–1357.

Firkins, J.L., Weiss, W.P., Eastridge, M.L. and Hull, B.L. (1990) Effects of feeding fungalculture extract and animal–vegetable fat on degradation of hemicellulose and on ruminalbacterial growth in heifers. Journal of Dairy Science 73, 1812–1822.

Flint, H.J., Aurillia, V., Kirby, J., Miyazaki, K., McCrae, S.I., Rincon, M.T. and Martin, J.C.(1998) Organization of plant cell wall degrading enzymes in the ruminal anaerobicbacteria Ruminococcus flavefaciens and Prevotella bryantii. MIE Bioforum, Genetics,Biochemistry and Ecology of Cellulose Degradation, Suzuka, Japan, p. 115.

Fontes, C.M.G.A., Hall, J., Hirst, B.H., Hazlewood, G.P. and Gilbert, H.J. (1995) Theresistance of cellulases and xylanases to proteolytic inactivation. Applied MicrobialBiotechnology 43, 52–57.

Forster, R.J. and Whitford, M.F. (1998) Investigations into rumen microbial diversity usingmolecular cloning and probing techniques. MIE Bioforum, Genetics, Biochemistry andEcology of Cellulose Degradation. Suzuka, Japan, p. 93.

Forwood, J.R., Sleper, D.A. and Henning, J.A. (1990) Topical cellulase application effects ontall fescue digestibility. Agronomy Journal 82, 900–913.


Gashe, B.A. (1992) Cellulase production and activity by Trichoderma sp. A-001. Journal ofApplied Bacteriology 73, 79–82.

Gorbacheva, I. and Rodionova, N.A. (1977) Studies on xylan degrading enzymes. I.Purification and characterization of endo-1,4-β-xylanase from Aspergillus niger str. 14.Biochemistry, Biophysics Acta 484, 79–93.

Graham, H. (1996) Feed enzymes influence intestinal microflora. Zootechnica International19, 88–92.

Greve, L.C., Labavitch, J.M. and Hungate, R.E. (1984) α-L-Arabinofuranosidase fromRuminococcus albus. 8: Purification and possible role in the hydrolysis of alfalfa cell wall.Applied and Environmental Microbiology 47, 1135–1140.

Gwayumba, W. and Christensen, D.A. (1997) The effect of fibrolytic enzymes on protein andcarbohydrate degradation fractions in forages. Canadian Journal of Animal Science 77,541–542.

Hatfield, R.D. (1993) Cell wall polysaccharide interactions and degradability. Phenolicconstituents of plant cell walls and wall biodegradability. In: Jung, H.G., Buxton, D.R.,Hatfield, R.D. and Ralph, J. (eds) Forage Cell Wall Structure and Digestibility. AmericanSociety of Agronomy, Crop Science Society of America, Soil Science Society of America,Madison, Wisconsin, pp. 285–313.

Headon, D.R. (1993) Activity analysis of enzyme under field conditions. In: Wenk C. andBoessinger, M. (eds) Enzymes in Animal Nutrition. Proceedings of the 1st Symposium.Kartause, Switzerland, pp. 233–240.

Hill, K.J. (1970) Salivary and gastric secretions. In: Swenson, M.J. (ed.) Duke’s Physiology ofDomestic Animals. Cornell University Press, New York, p. 385.

Hodgson, J. (1994) The changing bulk biocatalyst market. Bio/Technology 12, 789–790.Hoover, W.H., Kincaid, C.R., Varga, G.A., Thayne, W.V. and Junkins, L.L. Jr (1984) Effects

of solids and liquid flows of fermentation in continuous cultures. IV. pH and dilutionrates. Journal of Animal Science 58, 692–699.

Hristov, A.N., Rode, L.M., Beauchemin, K.A. and Wuerfel, R.L. (1996a) Effect of acommercial enzyme preparation on barley silage in vitro and in sacco dry matterdegradability. Proceedings of the Western Section American Society of Animal Science. RapidCity, South Dakota, 47, 282–284.

Hristov, A.N., McAllister, T.A. and Cheng, K.-J. (1996b) Exogenous enzymes for ruminants.Proceedings of 17th Western Nutrition Conference. Edmonton, Alberta, pp. 51- 61.

Hristov, A.N., McAllister, T.A. and Cheng, K.-J. (1998a) Stability of exogenouspolysaccharide-degrading enzyme in the rumen. Animal Feed Science and Technology 76,165–172.

Hristov, A.N., McAllister, T.A. and Cheng, K.-J. (1998b) Effect of dietary or abomasalsupplementation of exogenous polysaccharide-degrading enzyme supplementation onrumen fermentation and nutrient digestibility. Journal of Animal Science 76, 3146–3156.

Hristov, A.N., McAllister, T.A. and Cheng, K.-J. (2000) Intraruminal supplementation withincreasing levels of exogenous polysaccharide-degrading enzymes: effects on nutrientdigestion in cattle fed barley grain diets. Journal of Animal Science 78, 477–487.

Huhtanen, P. (1991) Associative effect of feeds in ruminants. Norwegian Journal ofAgricultural Sciences, Suppl. 5, 37–57.

Iwaasa, A.D., Rode, L.M., Beauchemin, K.A. and Eivemark, S. (1997) Effect of fibrolyticenzymes in barley-based diets on performance of feedlot cattle and in vitro gas produc-tion. In: Evolution of the Rumen Microbial Ecosystem, Joint RRI-INRA Rumen MicrobiologySymposium. Aberdeen, Scotland, Poster 39.


Iwaasa, A.D., Rode, L.M. and Beauchemin, K.A. (1998) Cumulative gas production of alfalfaforage treated with different cell wall-degrading enzymes. Journal of Dairy Science 81,Suppl. 1, 291.

Judkins, M.B. and Stobart, R.H. (1988) Influence of two levels of enzyme preparation onruminal fermentation, particulate and fluid passage rate and cell wall digestion in wetherlambs consuming either a 10% or 25% grain diet. Journal of Animal Science 66,1010–1015.

Kercher, J. (1960) Value of diallylstilbesterol and Zymo-pabst enzyme preparation forfattening yearling steers. Journal of Animal Science 19, 966.

Kopency, J., Marounek, M. and Holub, K. (1987) Testing the suitability of the addition ofTrichoderma viride cellulases to feed rations for ruminants. Zivocisna vyroba 32, 587–592.

Krause, M., Beauchemin, K.A., Rode, L.M., Farr, B.I. and Norgaard, P. (1998) Fibrolyticenzyme treatment of barley grain and source of forage in high grain diets fed to growingcattle. Journal of Animal Science 96, 1010–1015.

Krause, O.G., Richardson, R. and Cobb, C.O. (1989) Biological responses of steers andrats fed grains with an added microbial enzyme mixture. Journal of Animal Science 67,Suppl. 1, 280.

Kung, L. Jr (1996) Direct-fed microbial and enzyme feed additives. In: Muirhead S. (ed.)Direct-Fed Microbial, Enzyme and Forage Additive Compendium. The Miller PublishingCompany, Minetonka, Minnesota, pp. 15–20.

Kung, L. Jr, Treacher, R.J. and Cohen, M.A. (1998) Enzyme-treated forages for lactatingcows. Journal of Animal Science 76, (Suppl. 1), 196.

Leatherwood, J.M., Mochrie, R.D. and Thomas, W.E. (1960) Some effects of a supplemen-tary cellulase preparation on feed utilization by ruminants. Journal of Dairy Science 43,1460–1464.

Lehninger, A.L. (1982) Principles of Biochemistry. Worth Publishers, New York.Lewis, G.E., Sanchez, W.K., Treacher, R., Hunt, C.W. and Pritchard, G.T. (1995) Effect of

direct-fed fibrolytic enzymes on lactational performance of midlactation Holstein cows.Proceedings of the Western Section American Society of Animal Science. Lethbridge, Alberta,46, 310–313.

Lewis, G.E., Hunt, C.W., Sanchez, W.K., Treacher, R., Pritchard, G.T. and Feng, P. (1996)Effect of direct-fed fibrolytic enzymes on the digestive characteristics of a forage-baseddiet fed to beef steers. Journal of Animal Science 74, 3020–3028.

Ljungdahl, L.G., Blum, D.L., Chen, H., He, Y., Kataeva, I., Li, X. and Ximenes, E.A. (1998)The cellulase/hemicellulase system of the anaerobic fungus Orpinomyces and aspects offurther cellulase research. MIE Bioforum, Genetics, Biochemistry and Ecology of CelluloseDegradation. Suzuka, Japan, p. 19.

Luchini, N.D., Broderick, G.A., Hefner, D.L., Derosa, R., Reynal, S. and Treacher, R. (1997)Production response to treating forage with fibrolytic enzymes prior to feeding tolactating cows. Journal of Dairy Science 80 (Suppl. 1), 262.

Matte, A. and Forsberg, C.W. (1992) Purification, characterization, and mode of action ofendoxylanases 1 and 2 from Fibrobacter succinogenes S85. Applied and EnvironmentalMicrobiology 58, 157–168.

McAllister, T.A. and Cheng, K.-J. (1996) Microbial strategies in the ruminal digestion ofcereal grains. Animal Feed Science and Technology 62, 29–36.

McAllister, T.A., Phillippe, R.C., Rode, L.M. and Cheng, K.-J. (1993) Effect of the proteinmatrix on the digestion of cereal grains by ruminal microorganisms. Journal of AnimalScience 71, 205–212.


McAllister, T.A., Bae, H.D., Jones, G.A. and Cheng, K.-J. (1994) Microbial attachment andfeed digestion in the rumen. Journal of Animal Science 72, 3004–3018.

McAllister, T.A., Oosting, S.J., Popp, J.D., Mir, Z., Yanke, L.J., Hristov, A.N., Treacher, R.J.and Cheng, K.-J. (1999) Effect of exogenous enzymes on digestibility of barley silage andgrowth performance of feedlot cattle. Canadian Journal of Animal Science 79, 353–360.

McAllister, T.A., Stanford, K., Bae, H.D., Treacher, R.J., Hristov, A.N., Baah, J., Shelford,J.A. and Cheng, K.-J. (2000) Effect of a surfactant and exogenous enzymes ondigestibility of feed and on growth performance and carcass traits of lambs. CanadianJournal of Animal Science 80, 35–44.

McDermid, K.P., McKenzie, C.R. and Forsberg, C.W. (1990) Esterase activities ofFibrobacter succinogenes subsp. succinogenes S85. Journal of Bacteriology 170, 2914–2922.

Mertens, D.R., Strawn, T.L. and Cardoza, R. (1984). Modelling ruminal particle sizereduction: its relationship to particle size description. In: Kennedy, P.M. (ed.) Techniquesin Particle Size Analysis of Feed and Digesta in Ruminants. Canadian Society of AnimalScience Publication No 1, Edmonton, Alberta, pp. 134–141.

Michal, J.J., Johnson, K.A., Treacher, R.J., Gaskins, C.T. and Sears, O. (1996) The impact ofdirect fed fibrolytic enzymes on the growth rate and feed efficiency of growing beef steersand heifers. Journal of Animal Science 74 (Suppl. 1), 296.

Minato, H., Endo, A., Ootomo, Y. and Uemura, T. (1966) Ecological treatise on the rumenfermentation. II. The amylolytic and cellulolytic activities of fractionated bacterialportions attached to the rumen solids. Journal of General Applied Microbiology 12, 53–69.

Mir, P.S., Mears, G.J., Mir, Z. and Morgan Jones, S.D. (1998) Effects of increasing dietarygrain on viscosity of duodenal digesta and plasma hormone and glucose concentrations insteers. Journal of Animal Science 76 (Suppl. 1), 247.

Muirhead, S. (1996) Direct Fed Microbial, Enzyme and Forage Additive Compendium, 3rd edn.The Miller Publishing Company, Minetonka, Minnesota, 391 pp.

Muzakhar, K., Hayashii, H., Kawaguchi, T., Sumitani, J. and Arai, M. (1998) Purificationand properties of α-L-arabinofuranosidase and endo-β-D-1,4-galactanase from Aspergillusniger KF-267 which liquefied the okara. MIE Bioforum, Genetics, Biochemistry and Ecologyof Cellulose Degradation. Suzuka, Japan, p. 133.

Nelson, F. and Damon, V. (1960) Comparison of different supplemental enzymes with andwithout diethylstilbestrol for fattening steers. Journal of Animal Science 19, 1279.

Newbold, C.J. (1995) Microbial feed additives for ruminants. In: Wallace, R.J. and Chesson,H.C. (eds) Biotechnology in Animal Feeds and Animal Feeding. VCH Publishers, NewYork, pp. 259–278.

Newbold, C.J., Brock, R. and Wallace, R.J. (1992a) The effect of Aspergillus oryzaefermentation extract on the growth of fungi and ciliate protozoa in the rumen. Letters inApplied Microbiology 15, 109–112.

Newbold, C.J., Frumholtz, P.P. and Wallace, R.J. (1992b) Influence of Aspergillus oryzaefermentation extract on rumen fermentation and blood constituents in sheep given dietsof grass hay and barley. Journal of Agricultural Science Cambridge 119, 423–427.

Nocek, J.W. (1998) Carbohydrates in dairy rations: managing subclinical acidosis. Proceedingsof the 33rd Annual Pacific Northwest Animal Nutrition Conference. Vancouver, BritishColumbia, pp. 17–31.

Nussio, L.G., Huber, J.T., Theurer, C.B., Nussio, C.B., Santos, J., Tarazon, M., Lima-Filho,R.O., Riggs, B., Lamoreaux, M. and Treacher, R.J. (1997) Influence of a cellulase/xylanase complex (C/X) on lactational performance of dairy cows fed alfalfa hay baseddiets. Journal of Dairy Science 80 (Suppl. 1), 220.


Ohara, H., Karita, S., Kimura, T., Sakka, K. and Ohmiya, K. (1998) Multicellulase complexof Ruminococcus albus F40. MIE Bioforum, Genetics, Biochemistry and Ecology of CelluloseDegradation. Suzuka, Japan, p. 140.

Pell, A.N. and Schofield, P. (1993) Microbial adhesion and degradation of plant cell walls. In:Jung, H.G., Buxton, D.R., Hatfield, R.D. and Ralph, J. (eds) Forage Cell Wall Structureand Digestibility. American Society of Agronomy, Crop Science Society of America, SoilScience Society of America, Madison, Wisconsin, pp. 397–423.

Pen, J., Verwoerd, T.C., van Paridon, P.A., Beudeker, R.F., van den Elzen, P.J.M., Geerse, K.,van der Klis, J.D., Versteegh, H.A.J., van Ooyen, A.J.J. and Hoekema, A. (1993)Phytase-containing transgenic seeds as a novel feed additive for improved phosphorusutilization. Bio/Technology 11, 811–814.

Pendleton, B. (1996) The regulatory environment. In: Muirhead, S. (ed.) Direct-FedMicrobial, Enzyme and Forage Additive Compendium. The Miller Publishing Company,Minetonka, Minnesota, pp. 47–52.

Perry, T.W., Cope, D.D. and Beeson, W.M. (1960) Low vs high moisture shelled corn withand without enzymes and stilbestrol for fattening steers. Journal of Animal Science 19,1284.

Perry, T.W., Purkhiser, E.D. and Beeson, W.M. (1966) Effects of supplemental enzyme onnitrogen balance, digestibility of energy and nutrients and on growth and feed efficiencyof cattle. Journal of Animal Science 25, 760–764.

Pritchard, G., Hunt, C., Allen, A. and Treacher, R. (1996) Effect of direct-fed fibrolyticenzymes on digestion and growth performance in beef cattle. Journal of Animal Science74, Suppl. 1, 296.

Rovics, J.J. and Ely, C.M. (1962) Response of beef cattle to enzyme supplement. Journal ofAnimal Science 21, 1012.

Russell, J.B. and Dombrowski, D.B. (1980) Effect of pH on the efficiency of growth bypure cultures of rumen bacteria in continuous culture. Applied and EnvironmentalMicrobiology 39, 606–610.

Rust, J.W., Jacobsen, N.L., McGilliard, A.D. and Hotchkiss, D.K. (1965) Supplementationof dairy calf diets with enzymes. II. Effect on nutrient utilization and on composition ofrumen fluid. Journal of Animal Science 24, 156–160.

Sabatier, A.M. and Fish, N.M. (1996) Method of analysis for feed enzymes: methodologicalproblems? Journal of Applied Poultry Research 5, 408–413.

Sanchez, W.K., Hunt, C.W., Guy, M.A., Pritchard, G.T., Swanson, B.I., Warner, T.B. andHiggins, J.M. (1996) Effect of fibrolytic enzymes on lactational performance of dairycows. Journal of Dairy Science 79, Suppl. 1, 183.

Stokes, M.R. and Zheng, S. (1995) The use of carbohydrase enzymes as feed additives forearly lactation cows. Proceedings of the 23rd Conference on Rumen Function. Chicago,Illinois, p. 34.

Sturkie, P.D. (1970) Avian digestion. In: Swenson, M.J. (ed.) Duke’s Physiology of DomesticAnimals. Cornell University Press, New York, pp. 530–533.

Teeri, T.T. (1997) Crystalline cellulose degradation: new insight into the function ofcellobiohydrolases. Trends in Biotechnology 15, 160–167.

Tenkanen, M., Schuseil, J., Puls, J. and Poutanen, K. (1991) Production, purification andcharacterization of an esterase liberating phenolic acids from lignocellulosics. Journal ofBiotechnology 18, 69–84.

Theurer, B., Woods, W. and Burroughs, W. (1963) Influence of enzyme supplements in lambfattening rations. Journal of Animal Science 22, 150–154.


van Rooijen, G.J.H. and Moloney, M.M. (1994) Plant seed oil-bodies as carriers for foreignproteins. Bio/Technology 13, 72–77.

Vandevoorde, L. and Verstraete, W. (1987) The effect of aerobic cellulases on rumenfermentation. Medical Faculty Landbouww Rijksuniversity Gent 52, 1647–1654.

Varel, V.H. and Kreikemeier, K. (1994) Influence of feeding Aspergillus oryzae fermentationextract (Amarferm) on the in situ fiber degradation, ruminal fermentation, and microbialprotein synthesis in nonlactating cows fed alfalfa or bromegrass hay. Journal of AnimalScience 72, 1814–1822.

Varel, V.H., Kreikemeier, K.K., Jung, H.G. and Hatfield, R.D. (1993) In vitro stimulation offorage fiber degradation by ruminal microorganisms with Aspergillus oryzae fermentationextract. Applied and Environmental Microbiology 59, 3171–3176.

Wang, Y., McAllister, T.A., Xu, Z.J., Gruber, M.Y., Skadhauge, B., Jende-Strid, B. andCheng, K.-J. (1998) Effects of proanthocyanidins, dehulling and removal of pericarp ondigestion of barley grain by ruminal microorganisms. Journal of the Science of Food andAgriculture 79, 929–938.

Ward, P.P. and Conneely, O.M. (1993) Using biotechnology to improve enzyme yields:from DNA to the marketplace. In: Wenk, C. and Boessinger, M. (eds) Enzymes inAnimal Nutrition, Proceedings of the 1st Symposium. Schriftenreihe aus dem Institut fürNutztierwissenschaften Gruppe Ernährung, Kartause Ittingen, Switzerland, pp. 16–21.

Weichenthal, B., Rush, I. and Van Pelt, B. (1996) An enzyme-microbial feed product forfinishing steers. In: Nebraska Beef Cattle Report. University of Nebraska, Lincoln,Nebraska, pp. 68–69.

White, B.A., Mackie, R.I. and Doerner, K.C. (1993) Enzymatic hydrolysis of forage cellwalls. In: Jung, H.G., Buxton, D.R., Hatfield, R.D. and Ralph, J. (eds) Forage Cell WallStructure and Digestibility. American Society of Agronomy, Crop Science Society ofAmerica, Soil Science Society of America, Madison, Wisconsin, pp. 455–484.

Wing, J.M., Van Horn, H.H., Sklare, S.D. and Harris, B. Jr (1988) Effects of citrus molassesdistillers solubles and molasses on rumen parameters and lactation. Journal of DairyScience 71, 414–420.

Yang, W.Z., Rode, L.M. and Beauchemin, K.A. (1998) Effects of fibrolytic enzyme additiveson milk production of dairy cows. Journal of Animal Science 76 (Suppl. 1), 320.

Yang, W.Z., Beauchemin, K.A. and Rode, L.M. (1999) Effects of enzyme feed additives onextent of digestion and milk production of lactating dairy cows. Journal of Dairy Science82, 391–403.


M.R. Bedford and J. ApajalahtiMicrobial Interactions with Exogenous Enzymes12

Introduction

The intestinal microflora are an integral part of the digestive system of all animals.The fact that they are living organisms means that they have nutritional and spatialrequirements. This implies that the ability of the digestive system to digest andabsorb nutrients is, in part, dependent upon the species distribution and totalpopulation of resident microbes. The species distribution determines the range ofmetabolic processes that can be found and the total population the degree or extentof such processes. The microfloral population is itself very much dependent upon thediet as the ultimate source of substrates for metabolism (Wagner and Thomas, 1987;Savory, 1992). As a result, changes in dietary composition or nutrient density canhave dramatic effects on the intestinal microfloral populations (Gibson et al., 1996;Hillman, 1999; Reid and Hillman, 1999), which in turn influence the ability of theanimal to digest and absorb nutrients therein. Exogenous enzymes have frequentlybeen shown to improve nutrient digestibility and some authors have identifiedconcomitant changes in the intestinal flora (Hock et al., 1997; Vahjen et al., 1998).Whether such changes are merely an adaptation to the new environment or whetherthese changes play a large role in the physiological response seen in the host animalon use of exogenous enzymes is the focus of this chapter. Due to the relative paucityof data in pigs, the majority of this chapter will focus on poultry.

Physicochemical Environment of the Intestinal Tract

There are various estimates of the numbers of different strains of bacteria that inhabitthe gut of the broiler, but many settle in the range of 400–500. Most strains areunknown, many unculturable and thus knowledge of this ecosystem is at best rudi-mentary. The conditions in the intestine range from the relatively aerobic environ-ments of the crop and duodenum to progressively anaerobic conditions of the ileum


Microbial Interactions inthe Response to Exogenous

Enzyme UtilizationM.R. BEDFORD1 AND

J. APAJALAHTI2

1Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UK;2Danisco-Cultor Innovation, Kantvik, FIN-02460, Finland

12

and ultimately the caecum and colon where the redox potential is often less than−400 mV (J. Apajalahti, unpublished data), equivalent to that found in the rumen.Differences exist in the pH of these environments, either as a result of the digestiveactivity of the host (e.g. acid secreted in the stomach/proventriculus or neutralizationby secretion of sodium bicarbonate in the duodenum) or the microbes themselves(production of volatile fatty acids, amines, etc.). The substrates available for fermen-tation range from complex plant and animal polymeric compounds such as cellulose,xylan and proteins to the simplest monomeric sugars, amino acids and fatty acids.The conditions existing in the compartment of the intestine where such substratesfind themselves limits and dictates the metabolic processes through which they canbe utilized. Anaerobic fermentation dominates in the large intestine whilst facultativeanaerobes and aerobes are more frequent in younger animals (Naqi et al., 1970) andin the upper reaches of the intestine. Bacterial fermentation may provide resources tothe host in terms of vitamins, essential amino acids and fatty acids, energy in the formof volatile fatty acids (VFAs) and many growth factors, which include VFAs andpolyamines (Annison et al., 1968; Just et al., 1983; Dierick et al., 1989, 1990; Choctet al., 1996; Noack et al., 1996). On the other hand they can produce toxins andcarcinogens, compete for nutrients directly and create disease conditions directly orindirectly (Muramatsu et al., 1994; Choct et al., 1996; Johnston, 1999; Schutte andLanghout, 1999; Titball et al., 1999; Uchida et al., 1999). Ultimately the microfloralpopulation resident in the gut is dependent upon the resources provided (principallyderived from the diet and subsequent metabolites, and from the intestine itself –mucin, endogenous losses, etc.) and the initiating flora (which depends upon envi-ronmental exposure and species interactions). Since diet, sex, age and housing, whichrepresent a fraction of all the parameters involved in determining the subsequentmicrofloral ecosystem, vary with each experiment, it is not surprising to find discrep-ant data in the literature. The microbial population permutations are potentially lim-itless. It is therefore currently not possible to describe consistently and adequately agiven microfloral population on the basis of the diet offered to the host and it is evenmore difficult to describe its contribution to the nutritional or health status of thehost. Whilst some of the effects and capabilities of individual species may be under-stood in isolation, in a living gut the roles that each species plays and the potentialinteractions with others, be it beneficial or detrimental to the host, are not clear at all.

Given these constraints, however, it is still clear that performance responses toexogenous enzyme addition are likely mediated to be in part through changes in themicroflora species distribution and total population, and therefore it is importantthat their role should be determined.

Microbial Responses to Use of Enzymes

Enzyme effects in the ileum

Much of the early work with exogenous enzymes addressed the problematic grains,rye and barley. The response to enzymes targeting these grains was considered to be

300 M.R. Bedford and J. Apajalahti

due to their ability to increase nutrient digestibility and hence assimilation throughdepolymerization of the viscous cell wall carbohydrates in these grains (discussedin more detail in Chapter 7). Viscous digesta was blamed for reduced rates ofconvection and diffusion, which slowed the rate of nutrient extraction from theintestines (White et al., 1981, 1983; Annison and Choct, 1991; Bedford andClassen, 1992; Classen et al., 1995). Enzymatic depolymerization reduced viscosityand improved nutrient extraction rate and the result was more rapid growth.However, even in such early work it was quickly established that the detrimentaleffects of feeding rye- and barley-based rations could be partially, if not largely,overcome by use of antibiotics (Moran and McGinnis, 1968; MacAuliffe et al., 1976;Korelski and Rys, 1985). More recent work, in which germ-free and conventionalbirds were fed highly viscous diets, demonstrated that the negative effects of viscositywere virtually absent in the germ-free bird (Schutte and Langhout, 1999). Suchobservations suggest that viscosity per se is not relevant unless taken in context withthe microbial status of the bird. Subsequent work with wheat and more recentlymaize and sorghum-based diets has identified additional and alternative mechanismsof direct enzyme action (cell wall mechanisms, starch digestion mechanisms) which,again, may only be relevant when considered along with the microfloral status of thehost. The proposition of this chapter is that exogenous enzymes in use today (withthe notable exception of phytase) improve animal performance largely throughinteraction with the intestinal microflora. It is proposed that the exogenous cellwall-degrading enzymes have two types of activity that need consideration: removaland provision of substrates for microbial fermentation. The former is more likely tobe an effect seen in the small intestine and the latter in both lower small and largeintestine. Each activity and its consequences are discussed in detail below.

Ileal phase – substrate removal

The small intestine of monogastrics, particularly chickens, was for a long timethought to be largely devoid of bacteria, and any counts were assumed to be contami-nants rather than established flora. It was later established that in fact there was a veryrapid colonization of the small intestine of both turkeys and chickens within thefirst few days post hatch (Lev et al., 1957; Naqi et al., 1970; Vahjen et al., 1998).Subsequent work has established that there is a specific progression from bacteriastaining Gram-positive to those staining Gram-negative in the jejunum of turkeys asthey age from 1–21 days old. This evolutionary pattern varies, depending upon thesource flock and farm on which the birds were raised (Morishita et al., 1992).

The direct source of substrate for much of this flora is the diet of the host, withpoorly digested diets leaving far more substrate for microbial fermentation thanhighly digestible diets. Dramatically different ileal flora are evident when comparingrye with maize-based diets (Wagner and Thomas, 1987), the rye diet being verypoorly digested and supporting much larger numbers of anaerobes, and in particularspore-forming anaerobes, than the maize-based diet. Supplementation of the samemaize diet with a small percentage of pectin resulted in a dramatic increase in

Microbial Interactions with Exogenous Enzymes 301

numbers of anaerobes and in particular the spore formers. This was likely broughtabout through two properties of pectin that are relevant to this chapter, namely therheological properties of pectin, which would slow down the rate of digestibility andreduce oxygen tension through reduced digesta mixing, and its role as a substrate forbacterial fermentation. Reducing oxygen tension and provision of extra substrate willevidently favour increases in the numbers of anaerobes, and the fact that the pectinitself is fermentable will select for those strains capable of using the constituenturonic acids.

Such observations are supported by more recent work, where addition ofexogenous enzymes reduces viscosity of wheat- and rye-based diets and as a resultreduces populations of small-intestinal flora (Choct et al., 1996, 1999; Hock et al.,1997). Presumably, the addition of the enzyme increases the rate at which starch andprotein are removed from the small intestine, which effectively leads to less materialbeing available for microbial fermentation.

In many respects, the numerous studies that have reported significantimprovements in ileal digestibility of nutrients on enzyme addition (Pettersson et al.,1991; Bruce and Sundstol, 1994; Jondreville et al., 1995; Dänicke et al., 1997b;Kemme et al., 1999; Ravindran et al., 1999; Wyatt et al., 1999) have underestimatedthe significance of their results. Whilst extra nutrients captured by the terminal ileumdo indeed contribute to a greater extent to the productive energy of the animalthan those captured in the large intestine, the effect of such a response on the small-intestinal microflora is ignored by most. For example, if starch digestibility is 80%at the terminal ileum, then there is 20% remaining which has the potential to feedthe microflora in the large intestine. If addition of an enzyme accelerates the rateof digestion of the starch such that 85% has been digested by the terminal ileum,then the classic response of a nutritionist would be that the animal has extracted5% more starch and as such will gain more productive energy from the diet. Themicrobiologist will recognize that there is now only 75% of the fermentable starch ofthe control diet entering the large intestine. Moreover, that proportion of the starchthat has been removed by enzyme addition was likely the most easily degradable ofthe residual starch. With less substrate available to support starch fermenters in thelower small intestine, populations fall (Fig. 12.1). Similar conditions will apply forprotein digestibility and protein utilizers.

Highly indigestible diets will support greater intestinal microfloral populations,and population reductions will be greatest with the most effective enzyme applica-tion. Figure 12.2 demonstrates that in birds fed a reasonable quality wheat-baseddiet, the application of a xylanase-based enzyme results in a 60% reduction in micro-bial numbers. The antibiotic evidently reduces populations by direct activity againstthe target species, and the numbers are reduced more dramatically than that seenwith enzyme use. Whilst both reduce microfloral populations, their modes of actionare evidently quite different. The performance gain on use of an antibiotic is thoughtto be due to many factors, such as reduced competition for nutrients in the smallintestine, reduced local inflammation due to control of pathogens, and reducedintestinal thickening and length as a result of improved digestibility and pathogenloading (Thomke and Elwinger, 1998a,b). The latter two mechanisms result in a


more efficient digestion and reduced maintenance energy requirement. As a result ofthe total population changes and species shifts, many of these benefits apply to use ofenzymes (Brenes et al., 1992, 1993; Gollnisch et al., 1996; Hock et al., 1997; Simon,1998; Vahjen et al., 1998). Whilst the issues discussed to date relate to the ability ofenzymes to remove fermentable substrate from the small intestine, it is apparent thatthis is not the only mechanism by which they interact with the microflora.

Caecal phase – substrate provision

By their very nature, enzymes that reduce the size of carbohydrates produce smaller,more fermentable products. Xylanases and cellulases ultimately produce xyloseand glucose, respectively, and oligomers of various chain length and degrees of


Fig. 12.1. Micrograph of the terminal ileum contents of broilers fed wheat-baseddiets without (top panel and bottom left) and with enzyme addition. Top panelstained for starch, bottom autofluorescing bacteria in digesta sample.

substitution (in the case of xylans). The small intestinal concentration of carbohy-drates with an apparent molecular weight of 40,000 Da or less was increased almosttwofold on addition of high levels of a xylanase-based enzyme (Bedford and Classen,1992). More recent work (Apajalahti and Bedford, 1999), shown in Fig. 12.3, breakssuch data into additional categories of smaller molecular weight fragments. The effectof the enzyme is to increase dramatically the concentration of all xylo-oligomerfractions, but it is much more pronounced for those that are less than ten xylosesugars long (fivefold increase compared with 2.3-fold for the DP < 500). Theseoligomers are not, by their very nature, either digested or absorbed by the host andas a result remain in the aqueous phase of the intestinal digesta. Whilst inert tohost enzymes, such oligomers are broken down by bacterial enzymes that are presentin the caeca (Dänicke et al., 1999a) into their constituent sugars, which can then beutilized by many organisms inhabiting the digesta. Xylo-oligomers fed to diabeticrats were shown to increase caecal acetate levels more than threefold (Imaizumi et al.,1991), indicating their potential in supplying substrate to the large intestine. As aresult there is a reduction in the number of starch and protein users (Vahjen et al.,1998) and an increase in xylose users on incorporation of a xylanase enzyme in the


Fig. 12.2. Effect of addition of either an antibiotic (Avoparcin) or a xylanase onthe total population of bacteria inhabiting the small intestine of 21 day broilers.

Fig. 12.3. Addition of xylanase to a wheat-based diet increases the concentrationof small, soluble xylo-oligomers in the small intestinal tract.

ration (Fig. 12.4., Apajalahti and Bedford, 1999). In this work, broilers were fed awheat-based diet with or without supplementation with a xylanase enzyme. Theentire small intestinal digesta was removed and washed in buffer, and the extractedbacteria were grown in a medium in which xylose provided the only carbohydratesource. Microscopic enumeration of the total bacteria in each treatment prior toincubation revealed that the enzyme-treated birds yielded only 40% of the numberof organisms found in the control. Despite this fact, the results demonstrated aconsiderable degree of adaptation of the microflora towards utilization of xylose inenzyme-treated animals. Clearly the continued generation of this sugar in situ byexogenous enzymes exerts a significant selection pressure on the inhabitants of theintestine. This is seen in alterations in total numbers of bacteria and/or speciesdominance on feeding xylanase to pigs and poultry (Gollnisch et al., 1996; Hock etal., 1997; Vahjen et al., 1998). As a result of alterations in the intestinal flora profilein both the small and large intestines, it is likely there will be concomitant alterationsin intestinal morphology, digestive enzyme secretions, and stability of thesesecretions and local and systemic immune function.

Some effects are clearly direct. Some species, particularly enterococci andlactobacilli, numbers of which are often reduced on feeding enzymes, havebeen shown to secrete hydrolases that deconjugate taurine and cysteine frombile acids, rendering them less effective in fat digestion (Campbell et al., 1983;Bovee-Oudenhoven et al., 1999; Schutte and Langhout, 1999; Tanaka et al., 1999).As a result, the digestion of fat, particularly saturated fats, is compromised (Dänickeet al., 1997a, 1999b) and with it the digestibility of many fat-soluble compounds,


Fig. 12.4. Influence of previous exposure to xylanase enzyme on ability of totalsmall-intestinal flora isolated from broiler intestines to utilize xylose as the onlycarbon source. Incubation in media containing only xylose as source of carbon.Total bacterial contents of digesta washed and collected and used as starter culturein each experiment. Since 60% fewer bacteria were harvested and hence used perincubation test for the xylanase-treated birds, the differences between treatmentsfor each parameter is actually much larger on a per bacterial count.

including vitamins and pigments (Roxas et al., 1996). Application of an enzymeresults in fewer of these organisms and hence improved digestibility of these nutrients(Hock et al., 1997).

Indirect Interactions between Host and Microflora Mediatedby Enzyme Use

Some effects of enzyme-mediated changes in the microflora have a less direct effecton host nutrition. For example, use of enzymes has been reported to alter theconcentrations of polyamines and volatile fatty acids (Choct et al., 1996, 1999;Jansman et al., 1996; Silva and Smithard, 1996), compounds that are known toinfluence many processes in intestinal growth and development (Seidel et al., 1985;Osborne and Seidel, 1990; Shinki et al., 1991; Gee et al., 1996; Noack et al., 1996;Raul and Schleiffer, 1996). Furthermore, interaction of the microflora with theintestine will have far-reaching effects on digestion, due to modulations in villusstructure and turnover rate, mucin production and growth rates of the animal asa whole (Seidel et al., 1985; Smith, 1990; Bhatt et al., 1991; Furuse et al., 1991;Sharma and Schumacher, 1995; Gee et al., 1996; Noack et al., 1996; Silva andSmithard, 1996; Sharma et al., 1997). The interaction of the microflora with theintestines also has effects on the immune status of the animal which will inevitablyinteract with and alter the microbial community profile in itself (Ouwehand et al.,1999) and the susceptibility of the host to antigen damage and some nutrientdeficiencies (Fukushima et al., 1999; Mengheri et al., 1999). The effects of shiftsin fermentation in the caeca can also influence activity elsewhere in the digestivetract. In rats, the provision of fermentable carbohydrates has been shown to haveprofound effects on secretion of enteroglucagon from the colon, which in turninfluences gastric secretions and emptying rates (Gee et al., 1996). As a result themicrofloral status of the large intestine can influence the gastric phase of digestion,the extent of which in turn will influence the provision of substrates into the smalland large intestines.

In many respects it appears that the nutrition of the animal cannot be divorcedfrom the process of microfloral interaction with the diet. Digestion and absorption ofnutrients by the host seem to be accomplished through a two-way negotiated processbetween host and intestinal microflora that establishes a sustainable partition of theavailable resources. Use of exogenous enzyme seems to move this equilibrium morein favour of the host, hence the animal benefits.

The absence of a microflora will thus reduce much of the benefit of additionof an enzyme. It is the very fact that the populations of intestinal flora are probablyso different between experiments that results vary so much between experiments.The next section focuses on some of the factors that need to be considered wheninterpreting data from enzyme trials.


Microbial Modulators with Potential to Interact withResponse to Enzyme Usage

The species diversity and density within the intestinal tract of an animal will vary,depending upon the environment and diet. Environment evidently influences thecomposition and density of the resident flora – ranging from the germ-free to thedeliberately challenged or seeded environment (Muramatsu et al., 1994; Droleskeyet al., 1995; Noack et al., 1996; Hofacre et al., 1998; Schutte and Langhout, 1999).Nipple drinkers are cleaner than bell drinkers (Belyavin, 1995), and floor pens withlitter will provide for greater recycling and build-up of bacterial numbers thanwire-floor cages. Even litter material can significantly influence intestinal flora andhealth (Corrier et al., 1993; Bilgili et al., 1999).

The diet is of course a major source of the variation in intestinal flora diversityand total population. Provision of specific fermentable sugars can significantly influ-ence composition of the flora, with some (for example, the fructo-oligosaccharides)being put forward as carbohydrates that confer a favourable fermentation pattern(Oyarzabal and Conner, 1995; Grizard and Barthomeuf, 1999), whilst others (suchas the raffinose, stachyose and verbascose series of oligosaccharides) have a morequestionable value (Brenes et al., 1989; Krause et al., 1994; Iji and Tivey, 1998). Inany given study, the dietary concentrations of fermentable compounds derived fromingredients such as soybean meal, canola, peas, etc. are mostly unmeasured.

Antibiotics and coccidiostats with antimicrobial activity markedly influence thepopulation and composition of the resident flora (Ohya and Sato, 1983; Gollnisch etal., 1996; Hock et al., 1997). There are also many dietary nutrients, particularly met-als such as zinc and copper, which, when fed at elevated levels, have a pronouncedinfluence on the microfloral populations either directly or through interaction withthe immune system (Kidd et al., 1996; Yenigun et al., 1996; Leeson et al., 1997).Many nutrients interact with the immune system, either positively or negatively(Korver and Klasing, 1997; Klasing, 1998), which can significantly alter the popula-tions of the intestines.

When reviewing the literature, it is of value to consider the microbiologicalchallenge of each study as determined by chick handling, caging and rearing, dietaryingedients, antimicrobial compounds present, and the status of the immune system.All such factors, and many more, contribute to the condition and description of theresident flora. Since enzymes alter substrate flow into the small and large intestines,the subsequent response to their use will vary according to the populations presentat time of administration and their reaction to such changes. It is not surprising,therefore, that given the huge range in microfloral conditions likely to exist betweenstudies, the responses to enzyme use, rather than being absolute, are a continuumor population of responses from detrimental to highly positive. Determination ofthe key bacterial species involved in enzyme response will allow for more thoroughinterpretation of future research where such measurements are made.


Conclusions

Many of the factors that govern the scale of response to enzyme addition are just asrelevant to those influencing the response to antibiotics. The response to antibioticsvery much depends on the conditions under which the animals are grown, withresponses being much more marked in conventional than germ-free animals(Bywater, 1998). Indeed, in conventional birds, the general response to antibioticadministration is very much influenced by the relative performance of thenon-supplemented controls. The greatest responses are observed in those trials wherethe controls performed particularly poorly, and virtually no benefit is reported inthose trials where the controls performed very well (Rosen, 1995). The response toexogenous enzymes is subject to the same constraints, the greatest responses beingobserved with the poorest performing controls (Barrier-Guillot et al., 1995; Classenet al., 1995; Scott et al., 1998). Scale of response to exogenous enzymes seems todepend upon not only initial cereal quality but also the microfloral status of the bird.In studies where both antibiotics and enzymes have been used, in general theresponse to enzyme in the presence of the antibiotic is reduced in comparison withthat observed in its absence (Elwinger and Teglof, 1991; Broz et al., 1994; Vranjesand Wenk, 1996, 1997; Hock et al., 1997). Such observations suggest that theresponse to enzymes is reduced when the microflora is controlled by the presence ofan antibiotic. In those studies where the response to enzyme is just as great in thepresence or absence of antibiotic, it is interesting to note that the response tothe antibiotic alone is not particularly great (Allen et al., 1996; Esteve-Garcia et al.,1997). Reduced intestinal microbial populations probably limit the scope forimprovements in performance due to enzyme supplementation. When digestion isslowed through addition of carboxymethyl cellulose (CMC), which is a viscousnon-fermentable carbohydrate, the result is more apparent in conventional comparedwith germ-free chicks (Smits and Annison, 1996). In essence, addition of an enzymeto a wheat- or barley-based diet has much the same rheological effect as removal ofthe CMC from the diet in the work of Smits and Annison (1996). The implicationsof this work and that discussed earlier in this section are that the response to enzymeaddition probably depends on the initial microfloral status of the animals, which inturn depends on the digestibility of the ration and microfloral challenge. Nutrition,enzymes, antibiotics, prebiotics and probiotics are in effect all tools for negotiationfor nutrient abstraction from the intestines between the host and its residents.

References

Allen, C.M., Bedford, M.R. and McCracken, K.J. (1996) Interactions between rate of wheatinclusion, variety, antibiotic and enzyme addition in the responses of broilers toheat-treated, pelleted diets. British Poultry Science 37 (Suppl.), pp. S45–S46.

Annison, E.F., Hill, K.J. and Kenworthy, R. (1968) Volatile fatty acids in the digestive tract ofthe fowl. British Journal of Nutrition 22, 207–216.


Annison, G. and Choct, M. (1991) Anti-nutritive activities of cereal non-starch polysaccha-rides in broiler diets and strategies minimizing their effects. World’s Poultry ScienceJournal 47, 232–242.

Apajalahti, J. and Bedford, M.R. (1999) Improve bird performance by feeding its microflora.World Poultry 15, 20–23.

Barrier-Guillot, B., Bedford, M.R., Metayer, J.P. and Gatel, F. (1995) Effect of xylanaseon the feeding value of wheat-based diets from different wheat varieties for broilers.Proceedings of the WPSA 10th European Symposium on Poultry Nutrition, pp. 324–325.

Bedford, M.R. and Classen, H.L. (1992) Reduction of intestinal viscosity through manipula-tion of dietary rye and pentosanase concentration is effected through changes in thecarbohydrate composition of the intestinal aqueous phase and results in improvedgrowth rate and food conversion efficiency of broiler chicks. Journal of Nutrition 122,560–569.

Belyavin, C. (1995) Water – the significant input. International Poultry Production March,7–11.

Bhatt, R.S., Sharma, M. and Katoch, B.S. (1991) Effect of supplementation of diet with fibredegrading enzyme on performance and nutrient utilization in broilers. Indian Journal ofAnimal Nutrition 8, 135–138.

Bilgili, S.F., Montenegro, G.I., Hess, J.B. and Eckman, M.K. (1999) Sand as litter for rearingbroiler chickens. Journal of Applied Poultry Research 8, 345–351.

Bovee-Oudenhoven, I.M., Wissink, M.L., Wouters, J.T. and Van der Meer, R. (1999)Dietary calcium phosphate stimulates intestinal lactobacilli and decreases the severity of asalmonella infection in rats. Journal of Nutrition 129, 607–612.

Brenes, A., Trevi, J., Cente, C. and Yuste, P. (1989) Influence of peas (Pisum sativum) as adietary ingredient and flavomycin supplementation on the performance and intestinalmicroflora of broiler chicks. British Poultry Science 30, 81–89.

Brenes, A., Marquardt, R.R., Guenter, W. and Slominski, B.A. (1992) Broiler chickperformance, gastrointestinal size, and digestibility of non-starch polysaccharides (NSP)and oligosaccharides as affected by enzyme addition to diets containing whole anddehulled lupin. Conference Européenne sur les Proteagineux, Angers, France.

Brenes, A., Smith, M., Guenter, W. and Marquardt, R.R. (1993) Effect of enzyme supple-mentation on the performance and digestive tract size of broiler chickens fed wheat andbarley based diets. Poultry Science 72, 1731–1739.

Broz, J., Oldale, P. and Perrin-Volz, A.-H. (1994) Effects of Trichoderma viride enzymecomplex on performance of broiler chickens receiving pelleted diets. Archiv fürGeflugelkunde 58, 182–185.

Bruce, J.A.M. and Sundstol, F. (1994) The effect of microbial phytase in diets for pigs onapparent ileal and faecal digestibility, pH and flow of digesta measurements in growingpigs fed a high-fibre diet. Canadian Journal of Animal Science 55, 121–127.

Bywater, R.J. (1998) Benefits and microbial risks of feeding additive antibiotics. InternationalFeed Ingredient Forum, 1–5.

Campbell, G.L., Classen, H.L. and Goldsmith, K.A. (1983) Effect of fat retention on therachitogenic effect of rye fed to broiler chicks. Poultry Science 62, 2218–2223.

Choct, M., Hughes, R.J., Wang, J., Bedford, M.R., Morgan, A.J. and Annison, G. (1996)Increased small intestinal fermentation is partly responsible for the anti-nutritive activityof non-starch polysaccharides in chickens. British Poultry Science 37, 609–621.

Choct, M., Hughes, R.J. and Bedford, M.R. (1999) Effects of a xylanase on individual birdvariation, starch digestion throughout the intestine, and ileal and caecal volatile fatty acidproduction in chickens fed wheat. British Poultry Science 40, 419–422.


Classen, H.L., Scott, T.A., Irish, G.G., Huck, P., Swift, M. and Bedford, M.R. (1995)The relationship of chemical and physical measurements to the apparent metabolisableenergy (AME) of wheat when fed to broiler chickens with and without a wheat enzymesource. Proceedings of the 2nd European Symposium on Feed Enzymes, Antalya, Turkey,pp. 65–69.

Corrier, D.E., Hargis, B.M., Deloach, J.R. and Hinton, A. (1993) Protective effect ofused poultry litter and lactose in the feed ration on salmonella enteritidis colonizationof leghorn chicks and hens. Avian Diseases 37, 47–52.

Dänicke, S., Simon, O., Jeroch, H. and Bedford, M.R. (1997a) Interactions between dietaryfat type and xylanase supplementation when rye-based diets are fed to broiler chickens. 2.Performance, nutrient digestibility and the fat-soluble vitamin status of livers. BritishPoultry Science 38, 546–556.

Dänicke, S., Simon, O., Jeroch, H. and Bedford, M.R. (1997b) Interactions between dietaryfat type and xylanase supplementation when rye-based diets are fed to broiler chickens. 1.Physico-chemical chyme features. British Poultry Science 38, 537–545.

Dänicke, S., Vahjen, W., Simon, O. and Jeroch, H. (1999a) Effects of dietary fat type andxylanase supplementation to rye-based broiler diets on selected bacterial groups adheringto the intestinal epithelium, on transit time of feed, and on nutrient digestibility. PoultryScience 78, 1292–1299.

Dänicke, S., Jeroch, H., Bottcher, W., Bedford, M.R. and Simon, O. (1999b) Effect of dietaryfat type, pentosan level and xylanases on digestibility of fatty acids, liver lipids andvitamin E in broilers. Fett/Lipid 101, S90–S100.

Dierick, N.A., Vervaeke, I.J., Demeyer, D.I. and Decuypere, J.A. (1989) Approach to theenergetic importance of fibre digestion in pigs. I. Importance of fermentation in theoverall energy supply. Animal Feed Science and Technology 23, 141–167.

Dierick, N.A., Vervaeke, I.J., Decuypere, J.A. and Henderickx, H.K. (1990) Bacterial proteinsynthesis in relation to organic matter digestion in the hindgut of growing pigs; contribu-tion of hindgut fermentation to total energy supply and growth performances. Journal ofAnimal Physiology and Animal Nutrition 63, 220–235.

Droleskey, R.E., Corrier, D.E., Nisbet, D.J. and Deloach, J.R. (1995) Colonization of cecal(caecal) mucosal epithelium in chicks treated with a continuous flow culture of 29characterised bacteria: confirmation by scanning electron microscopy. Journal of FoodProtection 58, 837–842.

Elwinger, K. and Teglof, B. (1991) Performance of broiler chickens as influenced by a dietaryenzyme complex and without antibiotic supplementation. Archiv für Geflugelkunde 55,69–73.

Esteve-Garcia, E., Brufau, J., Perez-Vendrell, A.M., Miquel, A. and Duven, K. (1997)Bioefficacy of enzyme preparations containing beta-glucanase and xylanase activities inbroiler diets based on barley or wheat, in combination with Flavomycin. Poultry Science76, 1728–1737.

Fukushima, Y., Kawata, Y., Mizumachi, K., Kurisaki, J. and Mitsuoka, T. (1999) Effect ofbifidobacteria feeding on fecal flora and production of immunoglobulins in lactatingmouse. International Journal of Food Microbiology 46, 193–197.

Furuse, M., Yang, S.I., Niwa, N. and Okumura, J. (1991) Effect of short chain fatty acidson the performance and intestinal weight in germ-free and conventional chicks. BritishPoultry Science 32, 159–165.

Gee, J.M., Lee-Finglas, W. and Johnson, I.T. (1996) Fermentable carbohydrate modulatespostprandial enteroglucagon and gastrin release in rats. British Journal of Nutrition 75,757–766.


Gibson, G.R., Willems, A., Reading, S. and Collins, M.D. (1996) Fermentation of non-digestible oligosaccharides by human colonic bacteria. Proceedings of the Nutrition Society55, 899–912.

Gollnisch, K., Vahjen, W., Simon, O. and Schulz, E. (1996) Influence of an antimicrobial(avilamycin) and an enzymatic (xylanase) feed additive alone or in combinationon pathogenic micro-organisms in the intestine of pigs (E. coli, C. perfringens).Landbauforschung Volkenrode 193, 337–342.

Grizard, D. and Barthomeuf, C. (1999) Non-digestible oligosaccharides used as prebioticagents: mode of production and beneficial effects on animal and human health.Reproduction, Nutrition, Development 39, 563–588.

Hillman, K. (1999) Manipulation of the intestinal microflora for improved health and growthin pigs. Proceedings of the World’s Poultry Science Association Spring Meeting Scarborough,22–24 March, pp. 59–61.

Hock, E., Halle, I., Matthes, S. and Jeroch, H. (1997) Investigations on the compositionof the ileal and caecal microflora of broiler chicks in consideration to dietary enzymepreparation and zinc bacitracin in wheat-based diets. Agribiological Research – Zeitschriftfür Agrarbiologie Agrikulturchemie Ökologie 50, 85–95.

Hofacre, C.L., Froyman, R., Gautrias, B., George, B. and Goodwin, M.A. (1998) Use ofAviguard and other intestinal bioproducts in experimental Clostridium perfringens-associated necrotizing enteritis in broiler chickens. Avian Diseases 42, 579–584.

Iji, P.A. and Tivey, D.R. (1998) Natural and synthetic oligosaccharides in broiler chickendiets. World’s Poultry Science Journal 54, 129–143.

Imaizumi, K., Nakatsu, Y., Sato, M., Sedarnawati, Y. and Suga, M. (1991) Effects ofxylooligosaccharides on blood glucose, serum and liver lipids and cecum short-chain fattyacids in diabetic rats. Agricultural and Biological Chemistry 55, 199–205.

Jansman, A.J.M., Schulze, H., Bosch, M.W. and Leuvenink, H. (1996) Effect of xylanasesupplementation to a wheat based diet on the portal fluxes of glucose and volatile fattyacids in pigs. Animal Science 74, Suppl. 1.

Johnston, K.L. (1999) Small intestinal bacterial overgrowth. Veterinary Clinics of NorthAmerica – Small Animal Practice 29, 523.

Jondreville, C., Brongniart, I., Gatel, F. and Grosjean, F. (1995) Effect of enzymesupplementation on the ileal digestibility of barley-based diets in growing pigs. JournéesRecherche Porcine 27, 217–222.

Just, A., Fernandez, J.A. and Jorgensen, H. (1983) The net energy value of diets for growthin pigs in relation to the fermentative processes in the digestive tract and the site ofabsorption of the nutrients. Livestock Production Science 10, 171–186.

Kemme, P.A., Jongbloed, A.W., Mroz, Z., Kogut, J. and Beynen, A.C. (1999) Digestibilityof nutrients in growing–finishing pigs is affected by Aspergillus niger phytase, phytateand lactic acid levels. 1. Apparent ileal digestibility of amino acids. Livestock ProductionScience 58, 107–117.

Kidd, M.T., Ferket, P.R. and Qureshi, M.A. (1996) Zinc metabolism with special reference toits role in immunity. World’s Poultry Science Journal 52, 309–324.

Klasing, K.C. (1998) Nutritional modulation of resistance to infectious diseases. PoultryScience 77, 1119–1125.

Korelski, J. and Rys, R. (1985) Preliminary investigation on the effect of enzyme and anti-biotic supplementation on the feeding value of rye containing diets in broiler feeding.Biuletyn Informacying Przemyslu Paszowego 4, 35–41.

Korver, D.R. and Klasing, K.C. (1997) Dietary fish oil alters specific and inflammatoryimmune responses in chicks. Journal of Nutrition 127, 2039–2046.


Krause, D.O., Easter, R.A. and Mackie, R.I. (1994) Fermentation of stachyose and raffinoseby hind-gut bacteria of the weanling pig. Letters in Applied Microbiology 18, 349–352.

Leeson, S., Zubair, A.K., Squires, E.J. and Forsberg, C. (1997) Influence of dietary levelsof fat, fiber and copper sulfate and fat rancidity on cecal activity in the growing turkey.Poultry Science 76, 59–66.

Lev, M., Briggs, C.A.E. and Coates, M.E. (1957) The gut flora of the chick. 3. Differences incaecal flora between infected, uninfected and penicillin-fed chicks. British Journal ofNutrition 11, 364–373.

MacAuliffe, T., Pietraszek, A. and McGinnis, J. (1976) Variable rachitogenic effect of grainand alleviation by extraction or supplementation with vitamin D, fat and antibiotics.Poultry Science 55, 2142–2147.

Mengheri, E., Nobili, F., Vignolini, F., Pesenti, M., Brandi, G. and Biavati, B. (1999)Bifidobacterium animalis protects intestine from damage induced by zinc deficiency inrats. Journal of Nutrition 129, 2251–2257.

Moran, E.T. and McGinnis, J. (1968) Growth of chicks and turkey poults fed western barleyand corn-based rations: effect of autoclaving on supplemental enzyme requirement andassymetry of antibiotic response between grains. Poultry Science 47, 152–158.

Morishita, T.Y., Lam, K.M. and McCapes, R.H. (1992) The microbiologic ecology of theturkey poult jejunum. Preventive Veterinary Medicine 14, 233–240.

Muramatsu, T., Nakajima, S. and Okumura, J. (1994) Modification of energy metabolismby the presence of the gut microflora in the chicken. British Journal of Nutrition 71,709–717.

Naqi, S.A., Lewis, D.H. and Hall, C.F. (1970) The intestinal microflora of turkeys. AvianDiseases 14, 620–625.

Noack, J., Kleessen, B., Lorenz, A. and Blaut, M. (1996) The effect of alimentary polyaminedepletion on germ-free and conventional rats. Nutritional Biochemistry 7, 560–566.

Ohya, T. and Sato, S. (1983) Effects of dietary antibiotics on intestinal microflora in broilerchickens. National Institute of Animal Health Quarterly (Tokyo) 23, 49–60.

Osborne, D.L. and Seidel, E.R. (1990) Gastrointestinal luminal polyamines: cellularaccumulation and enterohepatic circulation. American Journal of Physiology 258,G576–G584.

Ouwehand, A.C., Kirjavainen, P.V., Shortt, C. and Salminen, S. (1999) Probiotics:mechanisms and established effects. International Dairy Journal 9, 43–52.

Oyarzabal, O.A. and Conner, D.E. (1995) Environment and health: in vitro fructo-oligosaccharide utilisation and inhibition of salmonellas by selected bacteria. PoultryScience 74, 1418–1425.

Pettersson, D., Graham, H. and Aman, P. (1991) The nutritive value for broiler chickens ofpelleting and enzyme supplementation of a diet containing barley, wheat and rye. AnimalFeed Science and Technology 33, 1–14.

Raul, F. and Schleiffer, R. (1996) Intestinal adaptation to nutritional stress. Proceedings of theNutrition Society 55, 279–289.

Ravindran, V., Hew, L.I., Ravindran, G., Gill, R.J., Pittolo, P.H. and Bryden, W.L. (1999)Influence of xylanase supplementation on the apparent metabolisable energy and ilealamino acid digestibility in a diet containing wheat and oats, and on the performance ofthree strains of broiler chickens. Australian Journal of Agricultural Research 50,1159–1163.

Reid, C.-A. and Hillman, K. (1999) The effects of retrogradation and amylose/amylopectinratio of starches on carbohydrate fermentation and microbial populations in the porcinecolon. Animal Science 68, 503–510.


Rosen, G.D. (1995) Antibacterials in poultry and pig nutrition. In: Wallace, R.J. andChesson, A. (eds) Biotechnology in Animal Feeds and Animal Feeding. VCHVerlagsgesellschaft mbH, Weinheim, Germany, pp. 143–172.

Roxas, E.S., Luis, E.S., Roxas, N.P. and Basilio, N.L. (1996) Enzyme supplementation of layerdiet. Proeedings of the 2nd Asian Broiler Association Congress. Manila, Philippines,pp. 127–138.

Savory, C.J. (1992) Enzyme supplementation, degradation and metabolism of threeU-14C-labelled cell-wall substrates in the fowl. British Journal of Nutrition 67, 91–102.

Schutte, J.B. and Langhout, D.J. (1999) Influence of the intestinal microflora on health andperformance of broiler chicks. Proceedings of the WPSA Spring Meeting. Scarborough,22–24 March, pp. 57–58.

Scott, T.A., Silversides, F.G., Classen, H.L., Swift, M.L. and Bedford, M.R. (1998) Effect ofcultivar and environment on the feeding value of Western Canadian wheat and barleysamples with and without enzyme supplementation. Canadian Journal of Animal Science78, 649–656.

Seidel, E.R., Haddox, M.K. and Johnson, L.R. (1985) Ileal mucosal growth duringintraluminal infusion of ethylamine or putrescine. American Journal of Physiology 249,G434–G438.

Sharma, R. and Schumacher, U. (1995) Morphometric analysis of intestinal mucins underdifferent dietary conditions and gut flora in rats. Digestive Diseases and Sciences 40,2532–2539.

Sharma, R., Fernandez, F., Hinton, M. and Schumacher, U. (1997) The influence of diet onthe mucin carbohydrates in the chick intestinal tract. Cellular and Molecular Life Sciences53, 935–942.

Shinki, T., Tanaka, H., Takito, J., Yamaguchi, A., Nakamura, Y., Yoshiki, S. and Suda, T.(1991) Putrescine is involved in the Vitamin D action in chick intestine. Gastroenterology100, 113–122.

Silva, S.S.P. and Smithard, R.R. (1996) Exogenous enzymes in broiler diets: crypt cellproliferation, digesta viscosity, short chain fatty acids and xylanase in the jejunum. BritishPoultry Science 37, S77–S79.

Simon, O. (1998) The mode of action of NSP hydrolysing enzymes in the gastrointestinaltract. Journal of Animal Feed Science 7, 115–123.

Smith, T.K. (1990) Effect of dietary putrescine on whole body growth and polyamine meta-bolism. Proceedings of the Society of Experimental Biology and Medicine 194, 332–336.

Smits, C.H.M. and Annison, G. (1996) Non-starch plant polysaccharides in broiler nutrition– towards a physiologically valid approach to their determination. World’s Poultry ScienceJournal 52, 203–221.

Tanaka, H., Doesburg, K., Iwasaki, T. and Mierau, I. (1999) Screening of lactic acid bacteriafor bile salt hydrolase activity. Journal of Dairy Science 82, 2530–2535.

Thomke, S. and Elwinger, K. (1998a) Growth promotants in feeding pigs and poultry. I.Growth and feed efficiency responses to antibiotic growth promotants. AnnalsZootechnica 47, 85–97.

Thomke, S. and Elwinger, K. (1998b) Growth promotants in feeding pigs and poultry. II.Mode of action of antibiotic growth promotants. Annals Zootechica 47, 153–167.

Titball, R.W., Naylor, C.E. and Basak, A.K. (1999) The Clostridium perfringens alpha-toxin.Anaerobe 5, 51–64.

Uchida, K., Satoh, T., Narushima, S., Itoh, K., Takase, H., Kuruma, K., Nakao, H., Yamaga,N. and Yamada, K. (1999) Transformation of bile acids and sterols by clostridia(fusiform bacteria) in Wistar rats. Lipids 34, 269–273.


Vahjen, W., Glaser, K., Schafer, K. and Simon, O. (1998) Influence of xylanase-supplementedfeed on the development of selected bacterial groups in the intestinal tract of broilerchicks. Journal of Agricultural Sciences 130, 489–500.

Vranjes, M.V. and Wenk, C. (1996) Influence of Trichoderma viride enzyme complex onnutrient utilisation and performance of laying hens in diets with and without antibioticsupplementation. Poultry Science 75, 551–555.

Vranjes, M.V. and Wenk, C. (1997) Influence of dietary enzyme complex on broiler perfor-mance in diets with and without antibiotic supplementation. British Poultry Science 36,265–275.

Wagner, D.D. and Thomas, O.P. (1987) Influence of diets containing rye or pectin on theintestinal flora of chicks. Poultry Science 57, 971–975.

White, W.B., Bird, H.R., Sunde, M.L., Prentice, N.A., Burger, W.C. and Marlett, J.A. (1981)The viscosity interaction of barley beta-glucan with Trichoderma viridae cellulase in thechick intestine. Poultry Science 60, 1043–1048.

White, W.B., Bird, H.R., Sunde, M.L., Marlett, J.A., Prentice, N.A. and Burger, W.C. (1983)Viscosity of β-D-glucan as a factor in the enzymatic improvement of barley for chicks.Poultry Science 62, 853–862.

Wyatt, C.L., Bedford, M.R. and Waldron, L.A. (1999) Role of enzymes in reducing variabilityin nutritive value of maize using the ileal digestibility method. Proceedings of theAustralian Poultry Science Symposium 11, 108–111.

Yenigun, O., Kizilgun, F. and Yilmazer, G. (1996) Inhibition effects of zinc and copper onvolatile fatty acid production during anaerobic digestion. Environmental Technology 17,1269–1274.


K. Clarkson et al.Development of New Enzyme Products13

Introduction

Enzymes were discovered in the latter part of the 19th century and have been used inindustrial and food processes since the early 1900s. Examples include the use ofamylases in textiles for removal of the starch sizing, bovine rennin in cheese making,pancreatic enzymes in production of leather and degumming of raw silk, andproteases for beer stabilization. As a result of the advances in biotechnology over thepast 10–20 years, especially in the areas of genetics and protein engineering, thereis now widespread use of enzymes in detergent, textile, grain wet milling, food,and pulp and paper applications. Incorporation of enzyme additives in feed rationsis more recent. The benefits of enzymes in feed applications include (Annison,1993; Bedford and Schulze, 1998; Simon, 1998): (i) balancing deficiencies in theendogenous enzyme system of the animal; (ii) elimination of anti-nutritionals; and(iii) increased utilization of the feed. As the value of enzymes continues to grow andnew markets are developed, technology, which can provide methods for more rapididentification and development of new or improved enzymes, will be a key elementfor success. This chapter is a review of current and emerging technologies used inthe discovery, research and development of new enzyme products with a focus onmicrobial production hosts.

Background

Enzyme classes and sources

Enzymes are extremely efficient and highly specific catalysts. They also exhibit lowtoxicity and act over wide pH and temperature ranges. There are several classes ofenzymes (International Union of Biochemistry and Molecular Biology, 1992):


Enzymes: Screening,Expression, Design

and ProductionK. CLARKSON, B. JONES, R. BOTT, B. BOWER,

G. CHOTANI AND T. BECKERGenencor International Inc., 925 Page Mill Road, Palo Alto,

CA 94304-1013, USA

13

1. Oxidoreductases catalyse the transfer of hydrogen from a donor to an acceptormolecule. These enzymes are also referred to as dehydrogenases or reductases. In caseswhere the hydrogen acceptor is oxygen, the enzymes are referred to as oxidases.2. Transferases carry out the transfer of a group, such as a glucosyl, from onecompound to another.3. Hydrolases catalyse the cleavage of specific bonds including carbon–oxygen,carbon–nitrogen and carbon–carbon bonds, and oxygen–phosphorus bonds.Essentially these are transfer reactions where water is the acceptor molecule. Many ofthe enzymes currently used in feed applications, such as amylases, xylanases, cellulases,proteases and phosphatases, are hydrolases.4. Lyases can cleave similar bonds as hydrolases but do so by a mechanism ofelimination resulting in the formation of double bonds or rings. They can also carryout the reverse reaction and add groups to double bonds.5. Isomerases carry out structural changes within a molecule.6. Ligases catalyse the coupling of two molecules in conjunction with the hydrolysisof a high-energy phosphate bond such as ATP.

The majority of enzyme products to date have been derived from mesophilicand, to a lesser extent, alkaliphilic microorganisms. While these sources continue toserve as valuable reservoirs of commercial enzymes, other microbial systems, suchas extremophiles, are rapidly gaining attention (Kristjansson and Hreggvidsson,1995; van der Oost et al., 1996; Vielle and Zeikus, 1996; Adams and Kelly, 1998;Davis, 1998; Horikoshi and Grant, 1998). Extremophiles are organisms that residein extreme environments and they include thermophiles (> 70°C and often in excessof 100°C), psychrophiles (< 20°C), barophiles (> 1 atm) and halophiles (high salt).Plant-derived enzymes as well as genetic engineering of plant cells and transgenicplants are also being explored (Kahl and Weising, 1993, Willmitzer, 1993; Flavell,1995; Ma et al., 1995; Mazur, 1995; Moffat, 1995; Evangelista et al., 1998; Galeand Devos, 1998; Kusnadi et al., 1998).

Enzyme properties

It is not unusual for a microorganism to produce an array of enzyme representativesfrom each class, subclass and category of enzyme. Many have even evolved multi-enzyme complexes or systems. The cellulolytic enzyme system, which is a mixture ofcellulases that act synergistically to degrade cellulose, is one example (Wood andKellogg, 1988). The cellulolytic system produced by each microorganism is uniqueand often comprises several types of cellulases, which can differ in their mode ofaction and specificity (substrate preference). Endoglucanases (endocellulases) attackthe internal glucosidic bonds of the cellulose polymer in a random fashion andproduce multiple cellulose chains of varying length as well as oligosaccharides.Another type of cellulase, cellobiohydrolases, attacks the cellulose chain ends andproduces primarily cellobiose. The more chain ends produced by the endoglucanases,the more substrate there is for the cellobiohydrolases. Build-up of cellobiose, a potent

316 K. Clarkson et al.

end-product inhibitor of cellobiohydrolases, is alleviated by β-glucosidase, whichconverts cellobiose to glucose. Therefore, the complete or whole cellulase system isessential for driving the degradation of cellulose to glucose, an important energysource for many microorganisms. Of further interest is the fact that cellulolyticmicroorganisms often generate a variety of endoglucanases and/or cellobiohydrolases.These enzymes can differ in their physical and biochemical properties such as molec-ular weight, isoelectric point (pI), pH and temperature activity profile, stability,and substrate specificity. Production of multiple xylanases, amylases, proteases,phosphatases, esterases, etc., is also not uncommon. This arsenal of enzymes allowsthe microorganism to access nutrients from a variety of sources and to adapt to a widerange of environmental conditions (pH, temperature, etc.). Through sophisticatedcontrol mechanisms, which have evolved over millions of years, the cell can react tothese types of changes and turn on genes in order to express the required enzymes andproteins or shut down non-essential protein synthesis and metabolic pathways. Inthis way, energy is conserved and directed to growth, reproduction and survival of theorganism.

Although many of the enzymes that are of value in feed applications can befound in nature, the levels produced by wild-type organisms are relatively low,making the economics unsuitable for most commercial applications. An added issueis the fact that many microorganisms produce a mixture of enzymes. While thesemixtures may show a benefit in the application, only one or two of the enzymespresent may actually be responsible. Also, while certain enzymes produced bya microorganism may be of benefit, others may exert a negative effect in theapplication. An example is an enzyme mixture containing both an endo-typexylanase, which is useful in reducing the viscosity of xylan, and a xylosidase, whichcatalyses the hydrolysis of short xylooligosaccharides to xylose. Given that xylose isknown to cause cataracts, diarrhoea and anorexia in some animals, this side activity isnot desirable (Schutte, 1990; Schutte et al., 1990, 1991). Therefore, before designinga production organism or host, it is important to know which enzyme(s) provides thebenefit in order to develop a host and process that are optimized for its production,thus reducing undesirable side activities while providing improved economics.

Identification and Development of Novel Enzymes

Enzyme identification

The development of enzyme products often relies on screening a large numberof organisms for an enzyme activity with a specific set of biochemical and physicalcharacteristics that suit the targeted application. Traditionally, microbial screeningprocesses have involved evaluation of purified enzymes isolated from: (i) pure cul-tures obtained from in-house or accessible culture collections throughout the world;(ii) microorganisms isolated from environments rich in the substrate of interest; and(iii) microorganisms that have a history of being good enzyme producers, such asTrichoderma, Aspergillus and Bacillus. While these approaches have met with success

Development of New Enzyme Products 317

over the years, there are limitations in that they provide access to only a very smallportion of the immense microbial population or diversity. Indeed it is believed that,to date, only a very small number of microbial species, possibly less than 1%, hasbeen cultured (Pace et al., 1986; Ward et al., 1992; Amann et al., 1996; Pace, 1997).Part of the problem lies in the fact that the majority of microbes are not culturableusing standard techniques. While traditional sources and approaches are still valid,several methods for handling unculturables and for speeding up their analysis haverecently become available (Olsen et al., 1986; Pace et al., 1986; Saiki et al., 1988;Weller and Ward, 1989; Ward et al., 1992; Amann et al., 1995, 1996). Most ofthese involve nucleic acid approaches in which the need to culture an organism iseliminated. One such method is sequencing of the total genome of specific organismsin order to screen for DNA sequences that are related to known DNA sequences,which express the desired type of enzyme activity. Other methods include extractionand amplification of DNA from unculturables or mixed culture populations. Theseapproaches and others, including use of RNA, will be discussed in more detail laterin this chapter. To complement these techniques much effort is also being investedin development of automated, miniaturized, highly sensitive, accurate and rapidmethods for obtaining and analysing nucleic acids and enzymes from various sources.

Despite the diversity of enzymes available in nature, it is sometimes the case thatthese enzymes do not possess all of the features necessary for a targeted application.Instead of rescreening for an enzyme that possesses one additional property, it ispossible to protein-engineer a deficient enzyme by modifying its sequence suchthat the lacking feature is incorporated into the enzyme without loss of the existingdesirable features.

Protein engineering of an enzyme can be carried out by site-directed and/orrandom mutagenesis. To a limited extent chemical methods, such as acetylation oramidation, can also be used. Although the chemical approach has met with some suc-cess, overall it is less predictable. Site-directed mutation involves modification of theenzyme properties by specific changes in the DNA code through the use of recombi-nant technology (Carter and Wells, 1987; Bott et al., 1988; Estell, 1993; Clelandet al., 1996). These changes can be selective amino acid substitutions, deletions orinsertions, or a selective combination of specific structural units of two enzymes togenerate hybrid or chimeric enzymes (Schneider et al., 1981; Mas et al., 1986). Aprotein engineering approach can be carried out by elucidating and studying thethree-dimensional structure of the protein by X-ray diffraction, nuclear magneticresonance (NMR) and other methods. Molecular graphics and modelling studies canbe used to predict the effect of single or multiple amino acid substitutions. In the past40 years or so, over 3000 protein structures have been elucidated (PDB, 1996;Godzik, 1997). Even so, it is still difficult to predict protein structures based onamino acid sequences. Fortunately some alternative approaches exist, such as randommutagenesis and, more recently, directed evolution, which has emerged as a powerfulalternative to site-directed protein design. These techniques can be applied toproduce designer catalysts without much knowledge of the protein structure.

Random mutagenesis has been used for decades as a tool for increasing geneticvariability and improving the properties of microbial strains as well as plant cultivars.


Improved strains or enzymes can be generated by random modification of the nucleicacid sequence followed by screening for the desired properties. Modification ormutation of the DNA sequence, which encodes for the targeted enzyme, can beachieved by physical (heat, freezing, UV, etc.) or chemical methods. With randommutagenesis there is less control over the number and types of mutations madecompared with site-directed engineering, which tends to generate modifications inlocalized regions of the amino acid sequence. This means that a huge number ofmutants must be generated and screened. These types of approaches require highlyefficient and rapid screening techniques.

Directed evolution involves multiple cycles of mutagenesis with selection orscreening. Mutants that show improved properties after each cycle are subjectedto further treatment. An accumulation of beneficial mutations is then achieved(Kuchner and Arnold, 1997). A suitable host for expression of the gene, appropriatemethods for generating genetic variability, and suitable selection and screeningmethods are all needed. Again, to ensure a high success rate, a large but manageablenumber of clones need to be screened. Another approach that can be applied isDNA shuffling. This technique involves random fragmentation of DNA sequencesfollowed by assembly of the fragments into genes by various recombination methods.

By combining enzyme screening with modern techniques such as proteinengineering and directed evolution, hybrid molecules with improved performanceunder specific application conditions can be generated.

Enzyme development and production

Once the targeted enzyme has been identified, the gene which encodes the enzyme isthen inserted into a suitable production host. As mentioned previously, many naturalisolates, from which an enzyme can be derived, are usually not acceptable productionhosts. Although it may be possible, to some extent, to alter the type and amountof protein produced by natural isolates or mutants through modification of thefermentation conditions, in most cases this is not sufficient to achieve the desiredpurity, process reproducibility and economics. Of equal importance are safetyconcerns and the fact that production hosts are usually limited to those with a historyof safe use. Therefore, recombinant methods are often employed. Given that the genesequence in a recombinant host is well characterized, there should be no safety issuesbeyond the normal procedures, which regulate all enzymes derived from geneticallyengineered organisms.

Attractive production systems or hosts include those microorganisms thathave the ability to: (i) produce and secrete large amounts of intact protein; (ii) carryout proper folding of the protein and post-translational modifications such assignal sequence cleavage, proteolytic processing, glycosylation, disulphide bridgeformation, etc.; (iii) generate stable recombinant cells or transformants; and (iv)comply with regulatory requirements such as ‘generally regarded as safe’ (GRAS)status (van den Hombergh et al., 1997).


Regardless of the nature of the production system developed, it needs to be com-patible with industrial-scale fermentation and recovery processes. These processesshould be reliable, meet all safety and quality assurance requirements, and providea reproducible source of high quality product. Ideally, the strain developmentand fermentation development efforts should be highly interactive. In turn, thefermentation development effort needs to be synchronized with the recovery efforts,since the nature of the fermentation broth exerts a large impact on downstreamprocesses such as cell separation and product recovery. The fermentation andrecovery processes must together generate a liquid concentrate that can be formulatedas a stable liquid or solid product without significant cost. Successful fermentationdevelopment requires: (i) a good understanding of the physiology of the microbialhost; (ii) development of a suitable medium; and (iii) optimization of the interactionsbetween the microorganism and its chemical and physical environment in thefermentor (Arbige et al., 1993). The chemical and physical environment, oncedefined, can be controlled by computers and automated processes. Given all thefactors that can exert an effect upon the productivity, statistical–mathematicalapproaches are often employed to optimize the various parameters.

The type of medium employed as well as by-products produced by the cells canhave a large impact on the protein recovery yields and the cost of the downstreamprocesses. Even enzyme products that are sold as crude preparations require down-stream recovery after fermentation. Once the enzyme has been isolated from thebroth, it can then be concentrated by ultrafiltration, precipitation, extraction or othermethods prior to formulation. Where purity is of high concern, further processinginvolving one or several fractionation steps is needed. The yield and product puritycriteria need to be balanced with the economics associated with each step, which isincorporated into the downstream process. Also, the nature of the ingredients used inthe fermentation and recovery processes can have a significant effect on the cost ofwaste disposal.

After the enzyme has been recovered and concentrated, it is usually formulatedin order to meet the product stability specifications suited to the application andhandling practices employed by the end-user. Food-grade ingredients are required forfood and feed applications. For solid products, the enzyme concentrate or formulatedproduct, if suitable, can be granulated or dried in a manner that provides a solidproduct that meets health and safety requirements, such as low dusting.

Screening

Screening in essence means searching for a particular gene coding for the targetenzyme. In the past this involved exclusively the screening of living microorganisms(‘classical microbial screening’), but today, by the application of modern tools ofmolecular biology, a screening may be performed without the need to culture theorganisms involved. As with all methods, there are limitations and a combination oftechniques (‘combinatorial screening’) often provides the most success.


The most rewarding starting point for the process of search and discovery is therich diversity of microorganisms in nature (Bull, 1992). It is seldom appreciated thatfor 85% of the earth’s history, life was restricted to microbial forms. The metabolicdiversity this has created is truly immense. During more than 4 billion years ofevolution, enzymes have emerged that are perfectly adapted for maintaining cellularprocesses. However, many enzymes are required to perform in a milieu that is farremoved from the natural physiological conditions and may even involve exposure tochemicals known to degrade or denature enzymes, which leads to loss of activity.Usually, enzyme screening and selection strategies are based on knowledge of theapplication and the physical and chemical conditions under which the enzymemust operate. Therefore, selection for enzyme performance under the applicationconditions is an essential early step in the screening process. Access to a large genepool is a prerequisite for a successful screening programme. The approaches caninvolve the indiscriminate examination of heterogeneous material (e.g. soil) gatheredfrom the environment, or an ecological approach with a targeted examination ofspecific habitats where a particular activity is likely to be found. Strategies can beadapted to search specifically in poorly explored environments in order to accessa greater variation in genomes from novel microbes. In the search for new genes,exploration of extreme environments such as volcanic hot springs, hypersaline lakesand polar soils has yielded many novel microorganisms with peculiar properties(Horikoshi, 1995; Kristjansson and Hreggvidsson, 1995; Horikoshi and Grant,1998). Alternatively, a taxonomic-based approach involving a systematic study ofknown organisms can also be rewarding.

Natural isolate screening

The source material can be plant or animal matter, or microbes – both prokaryotes(e.g. bacteria and archaea) and eukaryotes (e.g. yeasts and fungi). Currently, micro-organisms are the major source for industrial enzymes in the feed sector. The classicalmethod of screening natural microbial isolates is a well established process having itsroots in antibiotic discovery (e.g. penicillins, streptomycin) in the decades after 1945.It involves the examination of thousands of samples of soil, plant material, etc., andthe random isolation and screening of the resident microbial flora. Although capableof significant automation, the throughput capacity largely determines the speedof progress. In the past, screening strategies using soil samples as a starting pointwere highly empirical but much can be done to reduce the need for extensiveexperimentation. These processes involve subjecting an environmental sample (soil,etc.) to selective pressure using the principles of culture enrichment establishedearly in the 20th century by microbiologists such as Beyerinck and Winogradsky.By manipulating the growth conditions, only those microorganisms expressing aparticular enzyme are able to grow and survive. For example, by providing onlyxylan as the source of carbon and energy, only microbes producing a xylanase eitherconstitutively or by induction will grow and the culture will become enriched inthese organisms. The system is capable of considerable refinement; for example,


should an acidic xylanase be required, then the pH of the enrichment culture can bedesigned to permit growth of microbes producing the enzyme at pH 4–6. The out-come can be improved by a judicious choice of environment for examination – forexample, acidic soils, peat bogs and forest litter.

The time between collection of the environmental sample and examination inthe laboratory can be critical since changes will inevitably occur. Even in the best-preserved samples chemical and physical changes may occur, leading to the deathof some organisms and the reduced viability of others. The longer the time lapseinvolved, the more serious these changes can be. There are many advantages totreating the sample immediately in the field since this can lead to the isolation of atotally different population of microbes. It can be useful to provide ‘bait’ for specificmicrobes or types of microbial activity. An example of this in situ enrichment isplacing cotton linter in an environment such as lake sediment and retrieving thesample at a later date. The sample can then be taken back to the laboratory andscreened for cellulolytic activity.

The first problem in processing of samples is often one of numbers. Compostsand soils rich in organic matter contain enormous numbers of microbes. Some formof dilution is required, or targeting of specific groups. Selective techniques caninvolve, for instance, careful air drying of the (soil) sample, which reduces thenumbers of Gram-negative bacteria, or a brief heat treatment at 80°C, which willkill most microorganisms but leave the resistant spores of bacteria such as Bacillusintact. Antibiotics may be incorporated into the growth medium. For example,Gram-positive bacteria are more sensitive towards penicillin than Gram-negativebacteria. Many techniques have been developed empirically, such as the use of water-or chitin-agar for isolation of slow-growing actinomycetes, and such ideas arelimitless (Williams et al., 1984).

An ecological approach takes advantage of the natural selective pressure ofthe environment. For example, an alkaline environment such as a soda lake at pH 9is inhabited only by alkaliphilic microorganisms. It is likely that these organismshave extracellular enzymes that are stable and perform optimally at a high pH,which is indeed the case (Jones et al., 1998). For screening of feed enzymes suchconsiderations are often less clear-cut. What, for example, is the natural habitat fora phytase-producing microorganism? Phytate, a storage phosphate, is often a majorcomponent of some seeds and screening microorganisms in intimate associationwith the roots of germinating seeds (rhizosphere) would be an appropriate niche toexamine.

Whatever the screening strategy adopted, all of these approaches rely onsome form of selection criteria for the organism–activity combination being sought.The most widely applied methods involve growing the microbes on an agar platecontaining the appropriate enzyme substrate for detecting the desired enzymeactivity. Often the methods employed are proprietary and frequently constitute moreof an art than an exact science. Some methods, however, are well known. Forinstance, incorporating casein as sole carbon source into an agar medium providingan opaque surface will detect protease-secreting organisms as a clear halo ofhydrolysed casein around the growing colony. Specific protease cleavage reactions


can be detected by dye release from appropriate chemicals carrying a chromophorethat is only visible after the specific hydrolysis reaction has taken place.

The selection of microorganisms is a very critical stage in the screening processand can be quite subjective, especially if the numbers of microorganisms involved arevery large. Even though the screening has followed some form of selection process,it is often not easy to distinguish the different types of microorganisms growing onan agar medium. This is particularly true of bacteria, which, unless they are clearlycoloured, all look much the same. This gives rise to one of the biggest drawbacks ofthe random approach to isolation and screening of naturally occurring strains: theproblem of redundancy. The same microbe will recur time and again from differentsamples collected from different locations, which may explain why so many of thecurrent products come from a very limited range of microbes. Until relativelyrecently most screening strategies were fairly conservative. The tendency to use richmedia which favour fast-growing, nutritionally undemanding organisms has led toan overemphasis on particular groups of microorganism. This is coupled to the factthat certain fungal genera such as Aspergillus and Trichoderma, and Bacillus andPseudomonas species among the bacteria, have traditionally been a rich and provensource of commercial enzymes.

To ensure greater diversity and as a measure of how extensive a screening hasbeen, some estimate is required of the biodiversity of the population screened.A decade ago this was a daunting task involving a detailed examination of eachorganism, but nowadays fast molecular methods are available. A range of techniqueswill rapidly group collections of unknown strains into clusters of related organisms.Now, techniques such as total cell protein patterns, amplified fragment length poly-morphism, ribotyping, amplified ribosomal DNA restriction analysis, inter-nally transcribed spacer region–polymerase chain reaction (PCR), single-strandconformational polymorphism combined with sophisticated electrophoresis meth-ods, image analysis for data acquisition and computation can provide information onthe diversity of a strain collection. The ease with which small subunit ribosomal RNA(ssu rRNA) gene sequencing can be performed today, combined with a large andgrowing database for comparison, means that specific organisms can often be quicklyassigned to recognized taxonomic groups, and that novel microorganisms can berecognized when found. The clustering of organisms into related groups of strainscan be combined with screening selection data, thus permitting the identification ofspecific attributes with recognized groups of organisms. This information allowsparticular procedures to be devised for the isolation of a specific group and theimposition of more stringent selection criteria. Even so, it is still too easy to overlookpotential novel isolates, especially when confronted by many hundreds of culturedishes, each containing several hundred similar-looking microbial colonies.Nowadays, the task can be performed by colony-picking robots and visual imagingsystems coupled to computers.

As indicated earlier, one problem often encountered early in the screeningprocess is that natural microbial isolates usually produce commercially importantenzymes in exceedingly low concentrations. Environmental manipulation thatoptimizes the growth and production conditions is limited by the intrinsic maximum


ability of the wild-type organism. Classical genetic manipulation by selectingmutants with improved properties can lead to an increase of potential yield. Theclassical techniques of mutation and selection for microorganisms with improvedattributes have been applied for many decades. Modern methods of molecularbiology have in recent years accelerated the possibilities of creating in the laboratorymicroorganisms with new or improved activities. However, there is increasingevidence suggesting that many of the techniques associated with ‘genetic engineering’do, in fact, take place in the natural environment between groups of quite diversemicroorganisms. It is part of the opportunistic nature of microbes that allows themto cope with changing environmental conditions. This can lead directly to theacquisition of new activities within a microbial community. New activities are oftendue to selection pressure, especially where recalcitrant chemicals, such as a new herbi-cide, are introduced into the environment. This can be mimicked in the laboratoryby microbial selection in a continuous culture chemostat. This should not beconfused with the idea of the creation of a new microbe. Microorganisms react in amanner to preserve their genetic individuality, and in any case, the sudden emergenceof an entirely new organism is not an event we are ever likely to be able to witness.

Molecular screening

The last decade has witnessed spectacular progress in the technology enabling thedirect sequencing of nucleotides of a DNA strand. Already there are thousands ofgene sequences in public databases, although most relate to genes involved in thetranslation of the genetic code. The complete sequencing of microbial genomes hasalso accelerated quickly in recent years. At present some 35 prokaryotic (Bacteria andArchaea) and three eukaryotic microbial genomes have been reported in their entiretyand the sequencing of some 70 other microbial genomes is currently in progress.

There are several ways this can be done. If a suitable ‘lead’ enzyme (for example,an acid xylanase) has been identified in a particular species, then a search can be madefor homologous enzymes in related species. The process requires the application of‘reverse genetics’. Enzymes are polymers of amino acids covalently bonded in adefined sequence. The order in which the amino acids are arranged in the enzymecan be analysed by a process known as N-terminal sequencing. Since the genetic codeis universal it is possible to predict the probable sequence of nucleotide bases inthe gene coding for the specific enzyme. Using this information a probe can beconstructed consisting of 15–20 nucleotides (primer set) in the correct sequence,which because of the double helix nature of DNA will bind to the complementarystrand. By using the primer set in PCR on chromosomal DNA from relatedorganisms, a homologous gene can be amplified and transferred to a host organismsuch as Escherichia coli for the expression of the gene product – a homologous enzyme(Dalborge and Lange, 1998). In practice the procedure can be a little more complexthan is described here but it is possible to track down enzymes with identicalfunctionality but with a significantly different sequence of amino acids, where onlythe crucial elements of the enzyme, such as the active site, have been conserved


through evolution. In application tests these ‘new’ enzymes may have significantlydifferent properties, such as improved temperature or pH stability, compared withthe ‘lead’ molecule. In this way, whole families of related organisms can be screened.Even though this is a very powerful technique it is limited by the ability to grow andextract suitable DNA from the target organisms. It is also confined to known micro-organisms that have been isolated, described and deposited in culture collections. Ithas been estimated that fewer than 1% of all microorganisms that exist in nature havebeen isolated and even fewer have been adequately characterized and described(Amann et al., 1995). It is thought that fewer than 25% of the microbes in a sampletaken from the environment can actually be cultured in the laboratory. There may bemany reasons for this ‘unculturable’ portion of the microbial community. Some cellsmay be in a resting phase, or be damaged or moribund; some may not be trueresidents and not form part of the active population; others may be opportunistswaiting for a shift in environmental conditions in their favour; while others may haveunknown requirements for culture that remain unmet in the laboratory.

Fortunately, recent developments have permitted access to at least a portion ofthese uncultivated microbes (Hugenholtz and Pace, 1996). It is possible to extractDNA directly from the microbial community present in a sample taken from theenvironment. Although there are many techniques for extracting chromosomal DNAfrom natural microbial communities, not all cells are equally susceptible to lysis andthe DNA recovered is never 100%. Another concern is the detection of organismsthat form only a minor proportion of the total community. Their DNA willbe under-represented unless efforts are made to normalize the sample. A recentexamination of sequences of ssu rRNA genes in environmental DNA clones derivedfrom habitats as diverse as polar seawater from the Southern Ocean and hot springsin Yellowstone National Park revealed a vast biodiversity of ‘yet to be cultivated’microorganisms unsuspected a few years ago (Hugenholtz and Pace, 1996; DeLong,1997). It has also revealed that the cultivated microorganisms may not be thedominant flora in these environments, which provides a further challenge to thescreening-by-enrichment culture methods. The environmental DNA can be testedfor known genes using the PCR techniques described above. This requires goodquality DNA since the PCR screening is particularly sensitive to disturbance bycontaminating substances in the environment. Or, the DNA can be probed usingDNA carrying a radioactive or fluorescent label by a technique called hybridization.Alternatively, the environmental DNA can be cut into smaller fragments usingrestriction enzymes and the fragments cloned into an expression host such as E. coli.The clones can be screened for new enzymes’ genes by techniques similar to many ofthose described for the screening of natural microbial isolates.

However, all these molecular methods based on similarity screening have onebig drawback. They rely on a comparison with data that has already been recorded.Since more than 70% of the putative genes revealed by total genome sequencingprojects (Bacillus subtilis, for example) have no known function, these data indicatethat the resources for screening of novel enzymes are still largely untapped. It alsoremains true that the greatest diversity of life forms on this planet is microscopic. Itprobably makes little difference whether the exploration and screening of microbial


and gene diversity for novel enzymes for food and feed application use the techniquesof the ‘hunters and gathers’ (conventional microbial screening) or those of the‘breeders and selectors’ (molecular screening); the resources available are trulyimmense and the ultimate performance of the enzyme under application conditionsstill remains the crucial factor in selection.

Protein Engineering

The challenges for feed enzymes often require modification of enzymes to functionin a new environment or to possess a new property. Through protein engineeringtechniques, tailored enzymes for cleaning and starch processing have been developedand shown to outperform the best naturally occurring enzymes. The enzymes usedin cleaning and starch processing face quite unnatural environments: detergentformulations contain surfactants and buffers to stabilize highly alkaline conditionsand starch processing involves liquefaction of the raw starch at temperatures oftennear 105°C. The challenge has been to improve enzymes taken from quite differentenvironments in nature. Examples include subtilisin (protease) from Bacillus lentus(a microorganism growing on plants and in soil) for cleaning, and α-amylase fromBacillus licheniformis for starch processing. While neither enzyme would be expectedto function under commercial conditions, both perform the best of any foundin nature thus far. Although isolated from a mesophilic organism, the α-amylasefrom B. licheniformis is equal or superior to the α-amylase from the thermophile,B. sterathermophilus, in starch processing; and B. lentus subtilisin, while normallyoperating in near neutral environments, seems to excel in alkaline detergentenvironments. Thus natural enzymes can often be found that are already surprisinglyrobust. These enzymes have evolved over time and as a consequence possess relativelystable structures that can accommodate the accumulation of amino acid changesnecessary for the process of evolution. What has been undertaken in the process of‘engineering’ enzymes is nothing more than a directed evolution of the naturalenzyme to be better suited for the commercial task and conditions.

There are a number of complementary approaches to develop an enzyme withimproved characteristics. Generally, as previously noted, the first step is to surveynaturally occurring enzymes and identify the best starting enzyme. Sampling theavailable natural diversity and comparing functional properties versus the amino acidsequences may provide clues as to what properties are associated with each desirablemolecular trait. The difficulty is that, over the evolutionary process, enzyme aminoacid sequences have diverged considerably. With so much variation, the problemof identifying which sequence changes confer the desired trait can be extremelydifficult. For example, subtilisin Carlsberg differs from subtilisin BPN′ in 87 of apossible 275 amino acid residues (Siezen et al., 1991). These two enzymes possessquite different kinetic parameters, yet, as will be described below, changes in as few asone to three of the possible 87 amino acids appear to be able to shift the kineticparameters. The protein tertiary structure plays an important role in identifying thelikely structural elements that result in these functional differences.


Subtilisin engineering

Enzymes are synthesized as linear polypeptide chains. The amino acid sequence isdictated by the DNA sequence of the structural gene. After being synthesized, theenzyme folds into a stable three-dimensional structure that brings residues farremoved in the linear sequence into close juxtaposition. The juxtaposition of theseresidues and the creation of a substrate binding surface confer the enzyme moleculewith its catalytic properties. As an example, the structure of subtilisin BPN′, a serineprotease, is shown in Fig. 13.1. In this figure, three residues – aspartic acid (Asp) 32,histidine (His) 64 and serine (Ser) 221 – are brought into position to form the‘catalytic triad’ that is common to all serine proteases, not only subtilisin type butenzymes belonging to two other classes as well. There is also a wide depression wherea segment of the polypeptide chain, from the substrate, binds to position a peptidebond to be attacked by the activated Ser 221 hydroxyl oxygen.

Knowledge of what residues form the substrate binding surface is invaluablewhen deciphering which of the multitude of possible changes may contribute to an


Fig. 13.1. Three-dimensional structure of subtilisin BPN′ (Bott et al., 1988). Thelinear sequence adopts a stable three-dimensional fold to juxtapose the catalytictriad residues: Asp32, His64 and Ser221. The catalytic triad is situated in adepression that forms the substrate binding region.

enzyme having a desired property. Focusing on which residues differ betweensubtilisin BPN′ from B. amyloliquefaciens and subtilisin Carlsberg from B.licheniformis allowed the identification of three of 87 possible substitutions that werelikely to contribute to the difference in catalytic activity between the enzymes (Wellset al., 1987). Subtilisin Carlsberg has a turnover number, kcat, that is approximatelytenfold greater than subtilisin BPN′ and a different specificity (substrate preference)as well. The three-dimensional structures of subtilisin Carlsberg and subtilisin BPN′share a common folding pattern despite the differences in their sequences. It is possi-ble to superpose the two enzymes and relate residues that, while chemically different,are structurally homologous. This homology in tertiary structure along with therelative restricted fraction of residues forming the catalytic site and substrate bindingregion allowed the identification of three amino acid residues out of the 87 that werelikely to affect the kinetic parameter and specificity of the enzymes. By introducingthree substitutions – Ser for glutamic acid (Glu) 156 (E156S), alanine (Ala) forglycine (Gly) 169 (G169A) and leucine (Leu) for tyrosine (Tyr) 217 (Y217L) – asubtilisin BPN′ variant enzyme having the kinetic parameters and specificity corre-sponding to subtilisin Carlsberg was generated (Wells et al., 1987). Of these changes,a single substitution was found to exert a major effect on the difference seen for theturnover numbers, kcat. Even at this site as shown in Fig. 13.2 for the Y217L variant,only subtle rearrangement is necessary to accommodate the substituted residue (notethe superimposed leucine and tryosine residues at amino acid position 217).


Fig. 13.2. Comparison in three-dimensional structures of native subtilisin BPN′and a variant having the substitution of Leu for Tyr at position 217.

Amylase engineering

A similar strategy has been successfully employed with amylase. The α-amylase fromB. licheniformis consists of 458 amino acid residues, which in three dimensionsconsist of three domains as shown in Fig. 13.3. The active site and substrate bindingregion are situated between domains A and B.

The engineering goal for amylase was to increase the stability of the enzyme.Numerous variants with increased stability were found by exploiting the diversityfound in related α-amylases from different species. Several clues were initiallyobtained from variants derived from a screen of randomly created variants. These ledto a strategy of exchanging key amino acids with larger structurally homologousresidues observed in different α-amylases. It was proposed that bulkier residues,known to be accommodated in homologous locations in these other α-amylases,would fill surface cavities and thus stabilize the enzyme. An example is presented inFig. 13.4, where the substitution of Ser for Ala at position 379 has been modelled.This method has been used to create variants having increased pH and thermalstability (Day et al., 1998).

The strategy is based on the close homology seen in the structures of differentα-amylases. A residue found in one position in α-amylase from one species can often


Fig. 13.3. Three-dimensional structure of α-amylase from B. licheniformis. Themolecule consists of three domains A, B and C. Tha catalytic and substrate bindingregions are formed between domains A and B. (Courtesy of Dr Andrew Shaw,Genencor International.)

be accommodated in other α-amylases, which are homologous. The strategy thusrelies on the diversity of nature to provide clues such as use of crevice-filling residuesin the common α-amylase fold. These crevice-filling residues were then recruitedinto the B. lichenformis α-amylase. What is striking about the strategy was that thehomologous residues selected for substitution were found in the sequences of othermesophilic α-amylases having lower stability than B. licheniformis. These residueswere presumed to be very compatible with the α-amylase fold.

The overall process of selecting substitutions resulting in the most stable variantincludes rational site-specific substitutions, recruiting changes from variants isolatedfrom random mutagenesis and screening, and from the strategy of homologousrecruitment described above. The common thread is that all diversity can beexploited successfully for clues as to how to make an enzyme better suited for aparticular function and environment.

Combinatorial and other methods

The most recent advance in designing enzymes is the realization that, with advancedrecombinant DNA technology, it is possible to entertain the possibility of accelerat-ing evolution to a point where thousands of mutations can be created in successivegenerations that are subject to either direct selection or to indirect selection by way of


Fig. 13.4. Space filling representation of the substitution of Ser for Ala at position379. At position 379 the additional hydroxyl atom fills an external crevice andleads to a more stable variant. (Coordinates of the modelled variant provided byDr Anthony Day, Genencor International.)

a rapid screening strategy. This approach has been termed ‘directed evolution’. Todate there are numerous examples in the literature when either enzyme functionalityor stability has been improved using this approach. Subtilisin E variants have beenproduced to perform in radically different non-aqueous environments usingthese techniques (Chen and Arnold, 1993). While these demonstrations may beacademically successful, what is required for an industrial success is the ability toscreen or select for relevant properties and conditions.

These techniques generally produce numerous incrementally beneficial variants.A leapfrog technique that involves recombining the individual substitutions by shuf-fling their DNA has also been exploited. All of these techniques share the commonelement of exploiting the range of diversity found in homologous functioningenzymes. We see that a multitude of different but homologous sequences produceenzymes having a common tertiary structure and function. Different combinationsof substitutions from just a single amino acid substitution to substitutions rangingover the entire molecule result in enzymes with detectable differences but which areable to perform a common function.

It may well be an inherent property of natural enzyme structures to have a robusttertiary fold that is tolerant of considerable sequence variation. Thus evolution isallowed to sample a relatively broad range of variation, providing sufficient diversityfor enzymes better suited for specific environmental factors and functionalrequirements facing a particular organism.

Engineering Microorganisms for Enzyme Production

Classical mutagenesis

Classical mutagenesis is one of the most powerful techniques used to increase enzymeyields from microorganisms. Dramatic productivity improvements in the orderof tenfold, or even 100-fold, are common. Classical mutagenesis techniques wereoriginally developed by the pharmaceutical industry to improve the yields ofantibiotics. The same techniques are used to improve enzyme yields.

Classical mutagenesis techniques are not unlike the process of evolution. Bothrequire mutations in the DNA of an organism to bring about changes in genetictraits. While evolution selects for traits that confer greater reproductive fitness uponan organism, classical mutagenesis can select for traits chosen by the experimenter.Cells or spores are treated with mutagens to increase the frequency of mutation abovethe natural rate. The mutagens can include ultraviolet light, gamma ray irradiation orchemicals reactive with DNA. After treatment with mutagens, a diverse populationof cells is obtained. A screening process is used to identify cells having mutationsconferring desirable traits. The enzyme productivity of a microorganism is a trait thatcan be altered through mutagenesis. While many mutations are detrimental or lethalto an organism, a few can lead to increased enzyme productivity.

Classical mutagenesis was used to improve the yields of α-amylase enzymeproduction by the bacterium B. subtilis. The enzyme α-amylase is used for liquefying


maize starch for the production of high fructose maize syrup and for increasing thedigestibility of maize starch in animal feed. However, natural isolates of B. subtilisproduce very low levels of α-amylase. Classical mutagenesis and screening of awild-type strain resulted in the isolation of a more productive and thus commerciallyviable strain. Figure 13.5 shows a comparison of the wild-type strain and theimproved mutant strain growing on a nutrient agar supplemented with insolublestarch. Degradation and clearing of the starch provides an indication of the presenceof α-amylase. Large clear haloes surrounding the cells of the improved mutated strainindicate that the cells are producing and secreting α-amylase. Practically no halois apparent surrounding the wild-type cells, indicating low levels of α-amylaseproduction.

Genetic engineering to alter enzyme expression

Genetic engineering allows the in vitro manipulation of DNA with the ability toreintroduce the modified DNA back into host cells. The genetic elements of genesas well as their regulating elements can be altered to achieve changes in enzymeexpression. The typical bacteria or fungi used in enzyme production can havebetween two and 10,000 genes. The bacterium B. subtilis encodes its genes in 4.2million bases of DNA (Kunst et al., 1997) on one chromosome (one continuouslength of DNA). The fungus Aspergillus niger uses approximately 35 million bases ofDNA to encode its genes (Debets et al., 1990) on eight chromosomes. Each genespecifies the amino acid composition of a protein. One gene confers the sequence for


Fig. 13.5. Comparison of α-amylase production from Bacillus sp. strains growingon starch agar. The plate on the left shows an industrial overproducing α-amylasestrain generated using classical mutagenesis; the plate on the right shows a wildisolate strain. Haloes surrounding the colonies result from α-amylase hydrolysis ofthe insoluble starch.

one protein. The length of a typical bacterial enzyme gene, encoding a protein of 300amino acids, is in the order of 1000 nucleotides. In addition to segments of DNAthat encode genes, there are also segments of DNA called promoters that control thefunction of genes. Only a fraction of the genes of an organism are ‘on’ at any giventime. Coordination of the expression of genes is controlled by promoter elements.These regulatory segments of DNA can be thought of as light switches for genes, ableto turn a gene on, off or to a state in between.

A key breakthrough in the 1970s that made genetic engineering possible was thediscovery of specialized enzymes called restriction endonucleases. These enzymes arecapable of cutting DNA in precise locations, in effect acting as molecular knives.Another group of enzymes, called DNA ligases, can join cut ends of DNA together,in effect acting as molecular splicers. Segments of DNA can thus be excised fromchromosomes and recombined.

Genetic engineering requires the ability to transfer recombined DNA into a hostorganism, a process referred to as transformation. The bacteria E. coli and B. subtiliscan be transformed fairly easily: mixing cells with calcium chloride renders the cellscompetent (susceptible) to take up DNA. When genetically engineered DNA isadded to such competent cells, the DNA adheres to the cell walls and is transportedinside the cell. Though transformation of fungi is more complex, a method wasfound in the early 1980s to transform the fungus Aspergillus nidulans (Balance et al.,1983). Minor variations soon followed to allow transformation (and hence geneticengineering) of the fungi A. niger and Trichoderma reesei, both of which arefrequently used for enzyme manufacturing.

To enable high expression levels of enzymes, DNA can be recombined into anoverexpression vector. An overexpression vector is a specialized assembly of DNAthat when placed into a host cell causes a specific enzyme to be expressed (produced)above normal levels. Several DNA elements are essential to create a suitable vector.An example of an overexpression vector is one that has been used for production offerulic acid esterase A from A. niger (the enzyme is abbreviated as FAEA and the geneencoding this enzyme is abbreviated as faeA). FAEA (de Vries et al., 1997) can cleavethe ester link between ferulic acid and arabinose present in the cell walls of many feedsubstrates. For this overexpression vector (Fig. 13.6), recombination of four elementswas required: a strong promoter glaA, the faeA gene, the glaA terminator and thepyrG selectable marker. The glaA promoter regulates the faeA gene. The glaApromoter is known to result in high-level expression of genes that it regulates, andfor this reason was chosen for the vector. When the overexpression vector depictedin Fig. 13.6 was transferred into an A. niger host, production of the FAEA enzymeincreased significantly.

The selectable marker gene allows for a way to determine if a host organism hasbeen transformed with the vector DNA. Before transforming the vector into the A.niger host strain, the A. niger cells are initially mutated to make the native pyrG genenon-functional. The pyrG gene produces an enzyme required to synthesize the DNAbuilding block, pyrimidine. If pyrimidine cannot be made, the cell cannot synthesizeDNA, which abolishes any possibility of cell growth. After transformation with thevector, the cells are placed on selective nutrient media that lacks any pyrimidine. In


order to grow, cells must take up the overexpression vector DNA containing a pyrGgene. Those cells that take up the vector can synthesize pyrimidine and DNA.

Genetic engineering to delete enzyme genes

The use of genetically engineered overexpression vectors does not prevent non-targetenzymes from also being produced by a microorganism. Ideally a productionmicroorganism maximizes the expression of the enzyme of interest while minimizingproduction of inconsequential or deleterious enzymes such as proteases or enzymeswith interfering activities. Deletion of genes coding for undesirable enzymes can beperformed when necessary to prevent their production.

T. reesei is a filamentous fungus found in warm moist environments growing onwood. The original isolate was put through classical mutagenesis and screeningprogrammes which resulted in higher-producing cellulase strains. However, T. reeseiproduces at least six different cellulase enzymes, each with its own unique enzymaticproperties. Additionally, these enzymes are not produced at equal ratios. TypicalT. reesei cellulase strains produce an enzyme mixture consisting of approximately40–60% cellobiohydrolase I (CBHI), about 5–20% each of cellobiohydrolase II(CBHII), endoglucanase I (EGI) and endoglucanase II (EGII) and less than a fewpercent of EGIII (see Fig. 13.7, lane B). While this enzymatic ratio may be useful for


Fig. 13.6. A. niger expression vector for ferulic acid esterase A (faeA). Thevector shows required elements to produce high level faeA expression: a strongglucoamylase (glaA) promoter, the expressed gene, faeA and a selectable marker,pyrG.

T. reesei to convert the cellulose in plant material to glucose as an energy source, it isproblematic for applications where only a single cellulase enzyme is required.

In order to tailor the cellulase enzyme composition to a specific application, astrain of T. reesei was genetically engineered to prevent the production of CBHI,CBHII, EGI and EGII. Deletion of the genes that encode for the non-desiredcellulase enzymes resulted in an altered enzyme profile. An example of the genedeletion method, using the cbhI gene as a model, is illustrated in Fig. 13.8. The cbhIdeletion vector DNA cannot replicate independently upon entering the cell but mustintegrate itself into an existing chromosome, otherwise the vector DNA will not bereplicated. This integration can happen at random. However, since identical DNAhas an affinity for itself, the deletion vector was made with the selectable marker, thepyr4 gene, flanked on either side by cbhI DNA. The pyr4 gene is used as a selectablemarker in the equivalent manner that the pyrG gene is used in A. niger. The cbhIportions of the vector can then recombine with the cbhI portions of the chromosomeof the host strain, resulting in the replacement of the cbhI gene with the pyr4 gene.Upon deletion of the cbhI gene, production of the CBHI enzyme is not possible.Thus, presence of the selectable marker can be used to identify those cells that nolonger contain the cbhI gene.


Fig. 13.7. Progression of genetically engineered T. reesei strains. The SDS-PAGEgel shows the fermentation products produced by a series of strains starting fromthe parent production strain and ending with the EGIII overproducing strain. LaneA, molecular weight markers; B, parent strain; C, double deleted strain (Dcbh1,Dcbh2); D, quad deleted strain (Dcbh1, Dcbh2, Deg11, Deg12); E, eg13overexpression in quad deleted strain, late fermentation sample.

Deletion vectors were also made for the three other major cellulase genes:cellobiohydrolase II (cbhII ), endoglucanase I (eglI ) and endoglucanase II (eglII ).These deletion vectors were used to delete these three genes sequentially. The effectof gene deletions upon the enzymatic profile of various T. reesei strains can be seen ina protein gel (Fig. 13.7). The protein gel (SDS-PAGE) separates proteins on the basisof size (molecular weight). Fermentation broth samples were taken from variousgenetically engineered strains of T. reesei. When the cbhI and cbhII genes are deleted,the enzyme bands corresponding to these proteins (CBHI, CBHII) disappear (com-pare lanes B and C in Fig. 13.7). When the eglI and eglII genes are deleted, the EGIand EGII enzyme bands are no longer present (compare lanes C and D, Fig. 13.7).

The resulting deleted strain was further engineered to achieve higher productionof endoglucanase III (eglIII ) using an eglIII overexpression vector. This vectorcontained the strong cbhI promoter to regulate high expression of the eglIII gene. Thevector was transformed into the strain that had been deleted for cbhI, cbhII, eglI andeglII. The resulting strain showed an enhanced production of EGIII as compared


Fig. 13.8. Deletion of the T. reesei cellobiohydrolase I gene (cbhI).

with all the strains that preceded it (compare lane E with lanes B, C and D inFig. 13.7).

Classical mutagenesis and the genetic engineering techniques of gene deletionand gene overexpression can result in dramatic changes in the enzyme production ofa microorganism. By using such techniques, the enzymatic production of a microor-ganism can be tailored to fit specific applications needs.

Fermentation Process – Design and Development

The main goal of fermentation process development is to maximize the productivityand minimize the cost. Also, the process needs to be reproducible and mustmeet all regulatory, health and safety, and quality assurance requirements. Ideally,fermentation and strain development should take place in an integrated manner.Strain development must take into account the fermentation process design optionsand limitations, and the fermentation development must address critical strainrequirements.

Fermentor design

Fermentors are operated in batch, fed-batch, or continuous mode (Fig. 13.9).They are controlled by monitoring and estimating several critical parameters, such astemperature, pH, dissolved oxygen, redox potential, respiration rate, foam level andmass transfer coefficient (Fig. 13.10). During batch fermentation, the productionrate increases and then declines. This is often associated with the cell state: exponen-tial growth, stationary or sporulating. In fed-batch fermentation, by careful design ofbatch and feeding medium, cell growth is regulated and production is maximized.During continuous culture, production is affected by continuous growth, repression,low or non-producing mutants, and dilution. A continuous cell recycle systemdecouples growth and production and maximizes cell density, which results in highproductivity. If extracellular enzyme production under minimal growth is sustain-able, then an immobilized viable cell fermentor design would obviate the need for cellseparation and result in even higher productivity.

When cells are cultured on a rapidly utilizable carbon source (e.g. glucose andrich nutrients such as amino acids) the production of many enzymes is repressed.This is referred to as catabolite repression. Regardless of how catabolite repression iscontrolled inside a cell, either production strains must be able to make products inthe presence of carbon/nitrogen in the fermentor, or fermentors must be operatedunder carbon/nitrogen limitation. The maximum rate of product synthesis of manyextracellular enzymes occurs during the late exponential or early stationary phaseof growth, as the cells become committed to sporulation (Doi, 1989). The useof sporulation-deficient strains is therefore common in the industry (Priest, 1977).For production, the default industrial mode of operation is batch/fed-batchfermentation, with production cycles ranging from 12 to 300 h. Production


338K

. Clarkson

et al.

Fig. 13.9. Fermentor modes.

maximization with this type of fermentation usually entails optimization of thestrain, the medium, the carbon feed and the run conditions (temperature, pH,aeration, agitation, pressure). What exactly determines the limited period ofproduction is not known. The impact of other control systems (e.g. related tonutrient starvation, cell cycle events) is likely to be involved.

Fermentation process design is a highly interdisciplinary effort, requiringintegrated use of concepts and methodologies of both chemical engineering andmicrobial physiology, to accomplish scale-up. Fermentor ‘scale’ commonly refers tofermentor volume. Scale-up is a procedure for designing and building a large-scalesystem based on small-scale models. In practice, it involves the study of the effects ofscale on physiochemical and biological phenomena in fermentation process design.Scale-up effects are more pronounced for aerobic (carried out in aerated and agitatedvessels) than for anaerobic fermentations (Atkinson and Mavituna, 1991). Therefore,as a ‘rule of thumb’, in an aerobic fermentor, the constant oxygen transfer rate anddissolved oxygen concentration are generally maintained in the scale-up. Whilethermodynamic and intrinsic kinetic phenomena are independent of scale,momentum, mass and heat transfer are dependent on scale. For example, mixing,aeration and cooling are well controlled and uniform at the 10 l scale. However, notall transport parameters can be maintained in this manner at a greater than 10,000 lscale. Further scale-up complications arise from cell growth, adaptation and decay.


Fig. 13.10. Parameters affecting fermentors.

The development of entirely new processes or the improvement of existingprocesses requires the evaluation of a wide range of strains and cultivation conditionsin a short period of time. Shake-flask fermentation studies have a cost advantagebut lack the process control options (pH, nutrient addition, aeration). This oftenleads to use of laboratory scale (about 10 l size) agitated-aerated fermentors, withadequate instrumentation and control. Almost all fermentation processes can betranslated to production scale by use of laboratory-scale fermentors. However,pilot-scale fermentors are often necessary for the downstream process scale-up.

Process development

A considerable amount of published information is available on laboratory-scalefermentation processes but commercial-scale process information is for the most partproprietary and is generally not published in detail. Mutation, genetic manipulationand strain selection contribute the most to the industrially improved processes.

Once the host of choice has been decided, development of a culture, which canbe used to inoculate the seed flask, begins with the process of correctly storingand maintaining the organism. Usual methods include use of culture stocks stored:(i) in 10–20% glycerol or DMSO at liquid nitrogen temperature (−70°C); (ii)as freeze-dried cells; or (iii) on agar nutrient plates/slants kept refrigerated andtransferred periodically. In the process of scaling up a fermentation from a flask tothe production stage, a proper seed for inoculating the production fermentor needsto be generated. The requirement of the seed is to reach high cell densities in thepre-production fermentor so as to minimize the lag phase in the production vessel,without compromising the cells. A typical seed train for inoculation of a productionfermentor is shown in Fig. 13.11.

Seed flask inocula are prepared from frozen, freeze-dried or plated stock cultures.Erlenmeyer flasks, containing nutrient broth designed for optimal growth, are first


Fig. 13.11. Production fermentor: typical seed train for inoculation.

inoculated with a stored culture and incubated on an orbital shaker for about 4–48 h.Both the time of transfer and the cycle stage of the cells are critical factors. Thepurpose of the seed is also to adapt the cells for growth in the production media.Therefore the seed medium must be designed based on the physiological needs of theorganism.

All microorganisms have basic requirements, such as water, a source of energy,carbon, nitrogen, salts, trace metals and possibly growth factors. The challenge is todevelop a medium and conditions that optimize growth, by meeting the organism’sbasic needs, while producing the desired product economically. The media usedto isolate and screen production hosts are not necessarily those that can be usedin production fermentors. Development of an industrial process requires mediumcomponents that provide carbon, nitrogen, phosphorus, potassium, magnesium,sulphur and trace nutrients. The carbon source also provides energy for synthesisof cell mass and product. The objective of fermentation-medium development istherefore to maximize productivity, minimize side-products and costs, and ensuresteady product quality. Generally, large-scale fermentation media are made up ofcomplex natural materials supplemented with inorganic/organic salts, alkali andacids. Development of a balanced fermentation medium is based on knowledge ofphysiology as well as empirical work. Medium development combines growth andproduction data obtained with agar plate, microtitre plate, shake-flask cultureand fermentor studies. Statistical methods (Plackett–Burman, Box–Benkhen) areused for media screening and optimization (Monaghan and Koupal, 1990). Themedia and conditions used may change from shake-flask through the fermentorstages. Although many models have been developed for providing the basic needsof the organism (Pavlou and Fredrickson, 1989; Priest and Sharp, 1989; Arbige et al.,1993), the investigator is usually left to rely on either traditional recipes or mediadevelopment based on stochiometric requirements (growth, maintenance andproduction) and statistical design of experiments.

When choosing raw materials for a production-scale process, many inexpensivenutrients are available that provide good sources of the basic requirements, and manycontain inducers for product gene expression. Among these are soy flours, cottonseedmeal, cornsteep liquor and yeast and protein hydrolysates for nitrogen sources; andglucose, molasses, starch, cellulose and peat for carbohydrate sources. The macro-elements can be added as the salts (e.g. magnesium sulphate, potassium phosphate,sodium phosphate) while the trace elements can be added separately as salts, or maybe sufficiently available in the complex raw materials. Among metal ions, onlypotassium and magnesium are required as bulk salts, since these are involved mostlyas osmo- and energy-regulators, while the other trace elements have more specificphysiological roles (Hughes and Poole, 1989). Yeast extract usually contains all themetals required for growth, especially copper, iron, magnesium and zinc.

When designing media for product formation, many factors affecting proteinexpression and secretion need to be considered. In recombinant production hosts,vector selection plays an important part in determining the medium and the process.Also, when choosing a carbon source, the phenomenon of catabolite repression mustbe considered. Regulation of nitrogen and phosphorus metabolism is an additional


consideration. It has been observed on numerous occasions that some organicand inorganic nitrogen substrates have negative effects on exocellular proteaseproduction. While oligopeptides and proteins seem to be the inducers, salts and freeamino acids seem to be strong inhibitors (Chu et al., 1992). Repression by aminoacids appears to be due not to any particular one, but rather by the supply of anumber of amino acids in sufficiently high concentrations. Production can also beaffected by high levels of carbon. In amylase production, where the composition ofthe medium has a profound effect on product yields, the presence of phosphates canhave a significant effect on enzyme production (Thirunavukkarasu and Priest, 1980).It has also been shown that production is enhanced by using more slowlymetabolized carbon sources, such as starch or lactose, or by slow feeding of glucoseto maintain a very low concentration.

When developing media formulations optimized for growth and production,one area often neglected is the elimination of unwanted by-products, either comingfrom a raw ingredient or produced via metabolism. Common metabolites producedby bacilli are acetate, propionate and the metabolites of the butanediol cycle. Theseacids not only represent a waste of carbon; they can also lead to inhibition of growthor even cell death at high levels of accumulation.

Preparation of the fermentor is the next step after designing the basic mediafor growth and production. Some general texts have been written on fermentorpreparation (Rehm and Reed, 1993), but because of the advances in genetic, enzymeand pathway engineering, the investigator may also consider special requirements ofthe production host. Selection and regulation of ingredients can make the differencebetween a good and bad fermentation run. Another factor is the method ofsterilization, which refers to the physical or chemical processes of inactivation orelimination of contaminants present in liquid and gas phases before the fermentationtakes place. Besides the initial sterilization procedure, it is also important to preventcontamination during the run. There are several conditions that affect the probabilityof contamination: temperature, pH, sugar concentration, defined/complex nutrients,oxygen, back pressure, osmotic strength, and antibiotics present in the fermentor(Bader, 1986). Typical fermentation media can support contaminating moulds,yeasts and bacteria but generally fermentations are more susceptible tocontamination by spore-forming bacilli and bacteriophages. The sterilization kineticsvary dramatically between vegetative cell and bacterial spore, and between dry andwet heat sterilization. The lower the initial contaminants and particulate matter, themore effective the sterilization procedure. Good housekeeping helps in loweringthe bioburden and eliminating difficult-to-sterilize residues. Steam sterilization, asopposed to dry heat, is often preferred because of more efficient heat transfer andhigher reactivity, resulting in greater cell kill, most probably by rapid denaturation ofcellular protein. A rule of thumb is that all surfaces be exposed to saturated but notsuperheated steam, for time periods of 10–60 min at 121°C. These conditions aresufficient to inactivate all organisms and heat-resistant bacterial spores. Germinationand pasteurization (heating to 63°C for 30 min, or 80°C for 15 s) can also inactivateheat-resistant spores of the Bacillaceae family. On the other hand, phagecontaminants are difficult to contain and tend to spread rapidly throughout the


plant. The only effective way to prevent phage attack is to develop resistant cultures.In addition to sterilization, high-solids media are sometimes pretreated by enzymesor acid to avoid mass transfer limitations before seeding the fermentor with cells,since the cells utilize salts and trace nutrients only if available in the right form. As themedium components are utilized by the cells, the broth rheology changes, CO2 isproduced and foaming occurs, making the use of anti- or de-foaming agentsnecessary. Besides sterilizing all liquids and surfaces in contact with fermenting cells,it is equally important to sterilize the air used for cultivation. Air is first compressedto increase its pressure to as high as 100 psi, resulting in temperatures up to 260°C.Heat of compression reduces contaminants, including heat-resistant spores andphages, but is generally not solely relied upon for sterilization. So, filtration isconsidered the primary method of sterilization of dry air. The standard air ‘inletfilter’ is membrane based. These membranes are able to eliminate spores as well ascells, and are also able to withstand steam sterilization conditions.

Once the media handling and fermentor sterilization are completed, and theprocess variables are optimized, the performance of the host production strain in itscontrolled environment is studied. At this stage, the primary goal is to obtain themaximum biomass or product per unit of substrate in the minimal amount of time.Optimizing growth yields while maintaining cell viability is key to the success of theprocess development. Avoiding nutrient and O2 depletion and understandingby-product formation are also important for maximizing the yield and minimizingstress responses.

Downstream Processing and Formulation of Enzymes

The overall goal of downstream processing and formulation is to recover the enzymeof interest cost effectively at sufficiently high yield, purity and concentration, and in aform that is stable, safe and easy to use in a target application.

The variety of separations technologies available is large and the possible orderand permutation of sequential processing steps is vast. Several general considerations,however, limit the range of practical choices for a process design (Becker, 1995).Intracellular and membrane-bound enzymes are more difficult to recover than thosethat are secreted, requiring physical or chemical disruption of cells and the challengeof separating the enzyme from viscous or entraining substances such as nucleic acidsand cell wall materials. Fortunately, xylanases, proteases, amylases, cellulases andphytases – virtually all feed enzymes of commercial interest today – are extracellular,allowing for a cleaner first-stage separation. Most microbial fermentations are batchor fed-batch, and are typically not harvested until the cell mass, enzyme and othersolids are quite concentrated, often requiring some post-harvest dilution to avoidentrainment losses and low fluxes.

Typically, very high purity is not required of feed and food enzymes as it wouldbe of, say, human pharmaceutical proteins for injection. On the other hand, rawmaterials must be food grade or otherwise shown to be safe and suitable, and newenzyme formulations must be tested for toxicological properties and shown to be safe


for the intended use. Finally, economic considerations constrain both the rangeof raw materials and separation processes that can be used. While multiplechromatographic steps are routinely included in the downstream processing ofhuman therapeutic proteins, few feed or food enzymes include even a singlechromatographic step. Purification, if any, is accomplished by less expensivetechniques such as extraction or crystallization, as described below.

Harvest and cell separation

In general, cells should be removed from a fermentation broth within hours afterharvest in order to prevent cell lysis. After cell separation, the clarified fermentationbroth is usually stable and can be stored refrigerated for days. Upon harvest, the brothmay need to be cooled, pH adjusted, and certain stabilizers added in order to mini-mize enzyme degradation. For the more labile enzymes such as proteases, control oftemperature, pH, oxidants, inhibitors and activators will be important throughoutthe recovery process. For example, it is important to maintain a molar excess ofcalcium ions to ensure the thermal stability of certain Bacillus proteases and amylasesthat contain calcium-binding sites (Lloyd et al., 1970; Letton and Yunker, 1982).If minor proteases and other side activities cannot be deleted from the strain itself,finding ways to remove or neutralize them early in recovery becomes imperative.

Cell separation is usually accomplished by means of either depth filtration orcentrifugation. Fungal fermentations, such as those of Trichoderma or Aspergillus,lend themselves particularly well to cell separation by means of filtration through arotary drum vacuum filter (RDVF) because of the ease with which fungal mats can bethinly shaved off across the drum’s knife, renewing the filter cake surface to maintainhigh filtration flux. High-capacity filter presses can be a reasonable alternative toRDVFs. Bacterial fermentation broths can usually be processed by either filtration orcentrifugation, but the much smaller size of bacteria generally requires the additionof a polymeric flocculant. Cell flocculants are generally cationic, and function bybridging the negative surface charges on neighbouring cells to increase the particlesize and facilitate either sedimentation rate or filtration flux and clarity. In addition,the choice of flocculant and optimization of dosage is usually a delicate balancebetween considerations of separation quality, yield, cost and minimization of residualexcess polymer in the product.

The main challenge in centrifugation is to find flocculant and machine-operating conditions such that cell mass or sludge is easily conveyed or dischargedfrom the centrifuge bowl without break-up or carry-over in the centrate. Whilecentrifugation can handle a high concentration of cell solids, filtration may providemore complete removal of the most finely suspended solids, which can interfere withdownstream concentration or purification steps. On the other hand, the presence ofhigh solids and sometimes slimy polysaccharide fermentation by-products can limitfiltration fluxes without the judicious use of filter aids – usually diatomaceous earth,perlite, or other mined materials. The presence of flocculants and filter aids in cellwastes often gives rise to added disposal costs.


Microfiltration – the use of tangential flow anisotropic membranes to permeateenzyme while retaining suspended solids – would appear to be an attractive cellseparation technique because it does not require the use of flocculants or filteraids. However, microfiltration yields are typically low due to progressive fouling ofmembranes. Equipment and membrane replacement costs are currently higher thancapital and operating costs of filters and centrifuges of equal production capacity.Clarified filtrates or centrates are usually too dilute for use in feed applications, sosubstantial amounts of water must be removed.

Concentration

Industrial concentration methods, such as evaporation and solvent extraction, areunsuitable for dewatering enzymes because of their potential for thermal or chemicaldenaturation, and evaporation gives rise to high energy costs. Chromatography isgenerally uneconomical for concentrating feed enzymes, for reasons detailed below.The most common concentration technique in use today for industrial enzymes isultrafiltration (UF), using hydrophilic tangential flow membranes with molecularweight cutoffs in the range of 10,000–100,000 Da. UF fluxes and yields are oftensignificantly enhanced by removal – or omission – upstream of potential membranefoulants such as certain polysaccharides or anti-foams. Precipitation, crystallizationand extraction can also be used for concentration, but are more typically utilized aspurification techniques, and therefore will be discussed below.

Purification, decolorization and finishing

The main ‘purity’ consideration for feed enzymes is not the removal of proteinaceousimpurities, but rather ensuring that only food-grade raw materials and non-pathogenic, non-toxigenic production organisms are used. In some cases, anundesirable side activity must be removed through some means of purification,particularly when it is not feasible to delete the side activity genetically. Even ifminor proteases co-secreted by the host organism do not cause significantdegradation of the product enzyme during processing, they are more likely to wreakhavoc over several months of storage in the final product formulation. In other cases,side activities must be removed because they interfere in the ultimate application.Although chromatography offers the greatest potential and diversity of mechanismsfor fractionation, it is usually not cost effective in feed enzyme manufacture due tothe expense and low protein-binding capacity of most chromatography resins, andthe complexity and control required for reproducible operation.

Purification can often be accomplished with surprising effectiveness by two ofthe oldest known protein purification technologies: fractional precipitation andcrystallization. Fractional precipitation is most useful when the protein of interest isonly one of several enzymes or other proteins present. Chaotropic salts, such asammonium sulphate, or water-binding polymers, such as polyethylene glycol, are


added in increments to a concentrate containing the enzyme, and those fractional‘cuts’ are combined that have the most favourable balance of net enzyme purity andrecovered yield. When a higher level of purity is desired than can be attained throughfractional precipitation, crystallization is an attractive purification technique. Tothe surprise of those unfamiliar with it, it is also a highly scalable, reproducibleand cost-effective technique (Becker and Lawlis, 1991; Visuri, 1992). The purityattainable is comparable to what can be achieved through multiple chromatographicsteps. The major challenge is the often time-consuming screening of precipitants, pHand temperature optimization, and mapping of the solubility phase behaviour. Oncethe development work is completed, however, the process can be linearly scaled tomultiple tonnes. It should be noted that highly pure forms of certain proteases andother enzymes are in some cases less stable than the impure forms, which oftencontain stabilizing peptides and other impurities that inhibit degradation.

In the last one to two decades, aqueous two-phase extraction has become anattractive concentration technique. It provides the selectivity of classical solventextraction without its denaturing potential. Utilizing incompatible two-polymer andpolymer–salt combinations and adjustments in pH and ionic strength, this techniqueseparates proteins based on differences in hydrophobicity, surface charge andmolecular weight. One outstanding application of the technique has been thepurification of genetically engineered chymosin from multiple side activitiesproduced by the fungal host (Heinsohn et al., 1992).

Decolorization is sometimes preferred in certain applications, mostly as an aes-thetic preference. Colour can often be minimized by choice of fermentation mediumcomponents and control of the sterilization cycle so as to minimize the browningreaction between nitrogen and carbon sources. Colour can also be reduced by treat-ment with activated carbon, use of antioxidants, and diafiltration through membranes.

‘Finishing’ refers to one or more filtration steps at the end of downstreamprocessing which have the goal of clarifying and reducing the microbial load ofthe product. Solids are most conveniently removed using a filter press loaded withcellulosic pads, or precoated with filter aid; disposable submicron-gauge filtercartridges can be used to remove microbes, to the point of sterility if required.

Liquid formulation

The majority of feed enzymes are sold as formulated liquid concentrates, which maybe further formulated as liquid or solid products suitable for the end-user. The majorrequirements for a liquid formulation are enzymatic stability and preservation againstmicrobial growth. It is sometimes not appreciated that the dominant factor affectingenzyme stability is the intrinsic stability of the enzyme itself; formulation can dovery little to correct for a structurally labile protein. It is advisable, therefore, to makestability an important criterion of the initial screening process.

As most feed enzymes are hydrolases, they are subject to the three principalmeans of deactivation: denaturing or unfolding, catalytic site inactivation, andproteolysis (Becker et al., 1997). Denaturation is best minimized by controlling


temperature and pH and avoiding the presence of chemical denaturants. Catalyticsite inactivation is prevented by supplying sufficient levels of cofactor, typically ametal cation, and preventing oxidation of the active site – for example, by formulat-ing with antioxidants. Alternatively, oxidative resistance can be enzyme engineeredinto the protein structure. Finally, proteolysis – or autodigestion, in the case ofproteases – can be minimized by upstream deletion of proteases, by reducingwater activity through the use of water-sequestering compounds, or by addition ofinhibitors. Useful water-sequestering compounds include sugars and other polyols,such as glucose, glycerol, sorbitol and propylene glycol. Useful inhibitors includesubstrate analogues such as peptides and acid salts.

Microbial growth is fairly easy to prevent by the addition of food-gradeantimicrobials such as sodium benzoate, methyl paraben and various commercialpreparations. The prevention of precipitation and hazes is often a highly empiricalchallenge and one that is very specific to the enzyme and the specific process rawmaterials used. The best advice is to use ingredients with high solubilities, and toscreen potential formulations by extended storage at high and low temperatures.

Solid formulation

Most feed enzymes are supplied as liquid formulations at the end-user level, as theseare convenient to use. It is mostly dry enzyme premixes that are being sold to the feedmanufacturer. However, solid formulations can provide some significant advantages,such as enhanced stability, delayed or controlled release, and protection against deac-tivation during harsh applications. One example of the latter is the use of granules toencapsulate enzymes against deactivation in the steam pelleting process. The use ofeffective stabilizers and coatings can prevent or retard exposure to moisture underhigh heat, yet allow release of the enzyme in the subsequent feed application.

Dry enzymes, in most industries, are almost universally supplied as encapsulatedgranules in order to meet strict industry standards for control of airborne dust. Theprimary technologies for enzyme granulation include prilling, extrusion andspheronization, high-shear granulation, and fluidized-bed spray coating. The lattertwo techniques are superior in producing low dust granules. The advantage of thefluidized-bed method is that the entire granulation process can be carried out withina single, contained piece of equipment. In addition, the spray-coating processinvolved in this method allows the sequencing of layers of different thicknesses andcompositions with almost infinite flexibility.

Regulatory and Environmental Aspects

Along with the transfer of the fermentation, recovery and formulation processes tothe manufacturing plant, validated quality control methods and assays, which aresuitable for assuring product quality, must also be provided. Various agenciesthroughout the world are involved in the regulation of enzyme products for feed. In


the USA, all ingredients used in feed, including silage ingredients, need to be listed inthe official publication of the Association of American Feed Control Officials(AAFCO Manual ), and registered with the states in which they are sold, as well astonnage fees paid to the states. Any enzyme not listed in the AAFCO Manual requiresan assessment of safety and efficacy by the Center of Veterinary Medicine at the Foodand Drug Administration (FDA), who will then recommend listing in the Manual.In the EU, all enzymes require submission of a full dossier and specific approvalby trade name. Although the regulations can vary from country to country, in gen-eral most require (Wohlleben et al., 1993; Chapman, 1996): (i) the use of well-char-acterized non-pathogenic non-toxigenic organisms; (ii) good manufacturing prac-tices (GMPs) involving well-documented, safe and reproducible fermentation andrecovery processes in which only approved raw materials are used; (iii) the absence oftoxins or pathogens; and (iv) low levels of heavy metals or other contaminants. Notonly are genetically engineered organisms well characterized; they have also beenengineered for poor survival outside a controlled environment. This, along with thefact that recombinant organisms are grown under contained good industriallarge-scale practices (GILSPs), minimizes the safety issues relating to environmentalrelease. Biotechnology has also had a favourable environmental impact comparedwith traditional techniques. For example, enzymes derived from recombinantorganisms usually have higher fermentation yields and therefore reduce theconsumption of raw materials used for their production.

Conclusions

Advances in biotechnology over the past few decades have had an immense impactnot only in the areas of therapeutics, pharmaceuticals and health care but also inindustrial, food and feed applications. Recombinant DNA technology has proved tobe an effective means for increasing the production of enzymes, including foreign orminor enzyme components, and for generating tailored enzyme compositions fornumerous applications. Recent breakthroughs in screening methods and genomicshave allowed greater and faster access to unique enzymes from culturable as well asunculturable organisms. The ability to genetically engineer additional features intothese naturally occurring enzymes and to accelerate the evolutionary process in adirected manner has provided routes to the design of unique enzymes that aretailored to a specific application. These technologies, coupled with the developmentof well-characterized generic hosts and optimized and controlled fermentation,recovery and formulation processes, have assured a reliable flow of safe, stable andcost-effective products.

References

Adams, M.J. and Kelly, R.M. (1998) Finding and using hyperthermophilic enzymes. Trends inBiotechnology 16, 329–332.


Amann, R.I., Ludwig, W. and Schleifer, K.H. (1995) Phylogenetic identification and in situdetection of individual microbial cells without cultivation. Microbiology Reviews 59,143–169.

Amann, R., Ludwig, W. and Schleifer, K.H. (1996) Identification of uncultured bacteria: achallenging task for molecular taxonomists. ASM News 60, 360–365.

Annison, G. (1993) The role of wheat non-starch polysaccharides in broiler nutrition.Australian Journal of Agricultural Research 44, 405–422.

Arbige, M.V., Bulthuis, B.A., Schultz, J. and Crabb, D. (1993) Fermentation of Bacillus. In:Sonenshein, A.L. (ed.) Bacillus subtilus and Other Gram-Positive Bacteria: Biochemistry,Physiology and Molecular Genetics. ASM, Washington, DC, pp. 871–895.

Atkinson, B. and Mavituna, F. (1991) Biochemical Engineering and Biotechnology Handbook.Stockton Press, New York, 1271 pp.

Bader, F.G. (1986) Sterilization: prevention of contamination. In: DeMain, A.L. andSolomon, N.A. (eds) Manual of Industrial Microbiology and Biotechnology. AmericanSociety for Microbiology, Washington, DC, pp. 345–362.

Balance, D.J., Buxton, E.P. and Turner, G. (1983) Transformation of Aspergillus nidulansby the orotidine-5-phosphate decarboxylase gene of Neurospora crassa. Biochemical andBiophysical Research Communications 112, 284–289.

Becker, T. (1995) Separation and purification processes for recovery of industrial enzymes.In: Singh, R.K. and Rizvi, S.S.H. (eds) Bioseparation Processes in Foods. Marcel Dekker,New York, pp. 427–445.

Becker, T. and Lawlis, V.B. (1991) Subtilisin crystallization process. US Patent 5,041,377 toGenencor International.

Becker, T., Park, G. and Gaertner, A.L. (1997) Formulation of detergent enzymes. In: Van Ee,J.H., Misset, O. and Baas, E. (eds) Enzymes in Detergency. Marcel Dekker, New York,pp. 299–325.

Bedford, M.R. and Schulze, H. (1998) Exogenous enzymes for pigs and poultry (Review).Nutrition Research Reviews 11, 91–114.

Beecher, J.E., Andrews, A.T. and Vulfson, E.N. (1990) Glycosidases in organic solvents. II.Transglycosidase. Enzyme Microbial Technology 12, 955–959.

Bott, R.R., Ultsch, M., Kossiakoff, A., Graycar, T., Katz, B. and Power, S. (1988) The three-dimensional structure of Bacillus amyloliquefaciens subtilisin at 1.8 Å and an analysis ofthe structural consequences of peroxide inactivation. Journal of Biological Chemistry 263,7895–7906.

Bull, A.T. (1992) Isolation and screening of industrially important organisms. In:Vardar-Sukan, F. and Sukan, S.S. (eds) Recent Advances in Biotechnology. KluwerAcademic, Boston, pp. 1–17.

Carter, P. and Wells, J.A. (1987) Engineering enzyme specificity by ‘substrate-assistedcatalysis’. Science 237, 394–399.

Chapman, M.J. (1996) Toxicological considerations and regulatory aspects. In: Godfrey, T.and West, S. (eds) Industrial Enzymology. Stockton Press, New York, pp. 414–423.

Chen, K. and Arnold, F.H. (1993) Tuning the activity of an enzyme for unusualenvironments. Sequential random mutagenesis of subtilisin E for catalysis of indimethylformamide. Proceedings of the National Academy of Sciences USA 90, 5618–5622.

Chu, I.-M., Lee, C. and Li, T. (1992) Production and degradation of alkaline protease inbatch cultures of Bacillus subtilis ATCC 14416. Enzyme Microbial Technology 14,755–761.

Cleland, J.L., Jones, A.J. and Craik, C.S. (1996) Introduction to protein engineering. In:Cleland, J.L. and Craik, C.S. (eds) Protein Engineering. Wiley-Liss, New York, pp. 1–32.


Dalborge, H. and Lange, L. (1998) Using molecular techniques to identify new microbialbiocatalysts. Trends in Biotechnology 16, 265–272.

Davis, M.C. (1998) Making a living at the extremes. Trends in Biotechnology 16, 102–104.Day, A.G., Mitchinson, C. and Shaw, A. (1998) Mutant alpha-amylase enzymes. WO Patent

98/26078 awarded to Genencor International.Debets, A.J., Holub, E.F., Swart, K., van den Broek, H.W. and Bos, C.J. (1990) An electro-

phoretic karyotype of Aspergillus niger. Molecular Genes and Genetics 224(2), 264–268.DeLong, E.F. (1997) Marine microbial diversity: the tip of the iceberg. Trends in Biotechnology

15, 203–207.Doi, R.H. (1989) Sporulation and germination. In: Harwood, C.R. (ed.) Bacillus Biotechnol-

ogy Handbooks 2. Plenum Press, New York, pp. 169–215.Estell, D.A. (1993) Engineering enzymes for improved performance in industrial applications.

Journal of Biotechnology 28, 25–30.Evangelista, R.L., Kusnadi, A.R., Howard, J.A. and Nikolov, Z.L. (1998) Process and eco-

nomic evaluation of the extraction and purification of recombinant beta-glucuronidasefrom transgenic corn. Biotechnological Progress 14, 607–614.

Flavell, R.B. (1995) Plant biotechnology R&D – the next ten years. Trends in Biotechnology13, 313–319.

Gale, M.D. and Devos, K.M. (1998) Plant comparative genetics after 10 years. Science 282,656–661.

Godzik, A. (1997) Counting and classifying possible protein folds. Trends in Biotechnology 15,147–150.

Heinsohn, H.G., Lorch, J.D., Hayenga, K.J. and Arnold, R.E. (1992) Processes for therecovery of microbially produced chymosin. US Patent 5,139,943 to GenencorInternational.

Horikoshi, K. (1995) Discovering novel bacteria, with an eye to biotechnological applications.Current Biology 6, 292–297.

Horikoshi, K. and Grant, W.D. (1998) Extremophiles: Microbial Life in Extreme Environments.John Wiley & Sons, New York.

Hugenholtz, P. and Pace, N.R. (1996) Identifying microbial diversity in the naturalenvironment: a molecular phylogenetic approach. Trends in Biotechnology 14, 190–197.

Hughes, M.N. and Poole, R.K. (1989) Metals and Micro-organisms. Chapman and Hall, NewYork, 412 pp.

International Union of Biochemistry and Molecular Biology (1992) Enzyme Nomenclature.Academic Press, San Diego, 862 pp.

Jones, B.E., Grant, W.D., Duckworth, A.W. and Owenson, G.G. (1998) Microbial diversityof soda lakes. Extremophiles 2, 191–200.

Kahl, G. and Weising, K. (1993) Genetic engineering of plant cells. In: Rehm, H.-J. and Reed,G. (eds) Biotechnology 2. VCH, Weinheim, pp. 547–625.

Kristjansson, J.K. and Hreggvidsson, G.O. (1995) Ecology and habitats of extremophiles.World Journal of Microbiology & Biotechnology 11, 17–25.

Kuchner, O. and Arnold, F.H. (1997) Directed evolution of enzyme catalysts. Trends inBiotechnology 15, 523–530.

Kunst, F., Ogasawara, N., Moszer, I., Albertini, A.M., Alloni, G., Azevedo, V., Bertero, M.G.,Bessieres, P., Bolotin, A., Borchert, S., Borriss, R., Boursier, L., Brans, A., Braun, M.,Brignell, S.C., Bron, S., Brouillet, S., Bruschi, C.V., Caldwell, B., Capuano, V., Carter,N.M., Choi, S.K., Codani, J.J., Connerton, I.F., Danchin, A. et al. (1997) The completegenome sequence of the gram-positive bacterium Bacillus subtilis. Nature 390, 249–256.


Kusnadi, A.R., Hood, E.E., Witcher, D.R., Howard, J.A. and Kikolov, Z.L. (1998) Produc-tion and purification of two recombinant proteins from transgenic corn. BiotechnologicalProgress 14, 149–155.

Letton, J.C. and Yunker, M.J. (1982) US Patent 4,318,818, to The Procter & GambleCompany.

Lloyd, N.E., Lodge, R.E. and Waynes, R.A. (1970) US Patent 3,524,789 to Standard Brands.Ma, J.K.-C., Hiatt, A., Hein, M., Vine, N.D., Wang, F., Stabila, P., van Dolleweerd, C.,

Mostov, K. and Lehner, T. (1995) Generation and assembly of secretory antibodies inplants. Science 268, 716–719.

Mas, M.T., Chen, C.Y., Hitzeman, R.A. and Riggs, A.D. (1986) Active human–yeastchimeric phosphoglycerate kinases engineered by domain exchange. Science 233,788–790.

Mazur, B.J. (1995) Commercializing the products of plant biotechnology. Trends inBiotechnology 13, 319–323.

Moffat, A.S. (1995) Exploring transgenic plants as a new vaccine source. Science 268,658–660.

Monaghan, R. and Koupal, L. (1990) Use of the Plackett and Burman technique in adiscovery program for new natural products. In: Demain, A.L. (ed.) Novel MicrobialProducts for Medicine and Agriculture. Society for Industrial Microbiology, Arlington,Virginia, pp. 94–116.

Olsen, G.J., Lane, D.J., Giovannoni, S.J., Pace, N.R. and Stahl, D.A. (1986) Microbialecology and evolution: a ribosomal RNA approach. Annual Review of Microbiology 40,337–365.

Pace, N.R. (1997) A molecular view of microbial diversity and the biosphere. Science 276,734–735.

Pace, N.R., Stahl, D.A., Lane, D.J. and Olsen, G.J. (1986) The analysis of natural microbialpopulations by ribosomal RNA sequences. In: Marshall, K.C. (ed.) Advanced MicrobialEcology. Plenum Press, New York, 9, 1–55.

Pavlou, S. and Fredrickson, A.G. (1989) Growth of microbial populations in nonminimalmedia: some considerations for modeling. Biotechnology and Bioengineering 43, 971–989.

PDB (1996) Protein Databank, Brookhaven National Laboratory. http://www.pdb.bnl.govPriest, F. (1977) Extracellular enzyme synthesis in the genus Bacillus. Bacteriological Reviews

41, 711–753.Priest, F.G. and Sharp, R.J. (1989) Fermentation of Bacilli. In: Neway, J.O. (ed.) Fermenta-

tion Process Development of Industrial Organisms. Marcel Dekker, New York, pp. 73–132.Rehm, H.J. and Reed, G. (1993) The Fundamentals of Biochemical Engineering: Biotechnology

2. VCH, Weinheim, 880 pp.Robison, K., Gilbert, W. and Church, G.M. (1994) Large scale bacterial gene discovery by

similarity search. Nature Genetics 7, 205–214.Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B. and

Ehrlich, H.A. (1988) Primer-directed enzymatic amplification of DNA with athermostable DNA polymerase. Science 239, 487–491.

Schneider, W.P., Nichols, B.P. and Yanofsky, C. (1981) Procedure for production of hybridgenes and proteins and its use in assessing significance of amino acid differences inhomologous tryptophan synthetase α polypeptides. Proceedings of the National Academyof Sciencies USA 78, 2169–2174.

Schutte, J.B. (1990) Nutritional implications and metabolizable energy value of D-xylose andL-arabinose in chicks. Poultry Science 69, 1724–1730.


Schutte, J.B., van Leeuwen, P. and Lichtendonk, W.J. (1990) Ileal digestibility and urinaryexcretion of D-xylose and L-arabinose in ileostomized adult rooster. Poultry Science 70,884–891.

Schutte, J.B., de Jong, J., Polziehn, R. and Verstegen, M.W.A. (1991) Nutritionalimplications of D-xylose in pigs. British Journal of Nutrition 66, 83–93.

Siezen, R.J., de Vos, W.M., Leunissen, J.A.M. and Dijkstra, B.W. (1991) Homology model-ling and protein engineering strategy of subtilases, the family of subtilisin-like serineproteinases. Protein Engineering 4, 719–737.

Simon, O. (1998) The mode of action of NSP hydrolysing enzymes in the gastrointestinaltract. Journal of Animal Feed Science 7, 115–123.

Thirunavukkarasu, M. and Priest, F. (1980) Regulation of amylase synthesis in Bacilluslicheniformis NCIB-6346. FEMS Microbiology Letters 7, 315–318.

van den Hombergh, J.P.T.W., van de Vondervoort, P.J.I., Fraissinet-Tachet, L. and Visser, J.(1997) Aspergillus as a host for heterologous protein production: the problem ofproteases. Trends in Biotechnology 15, 256–263.

van der Oost, J., de Vos, W.M. and Antranikian, G. (1996) Extremophiles. Trends inBiotechnology 14, 415–417.

Vielle, C. and Zeikus, J.G. (1996) Thermozymes: identifying molecular determinants ofprotein structural and functional stability. Trends in Biotechnology 14, 183–190.

Visuri, K.V. (1992) Method for producing crystalline glucose isomerase. US Patent5,120,650.

Vries, R.P. de, Michelsen, B., Poulsen, C.H., Kroon, P.A., van den Heuvel, R.H., Faulds,C.B., Williamson, G., van den Hombergh, J.P. and Visser, J. (1997) The faeA genes fromAspergillus niger and Aspergillus tubingensis encode ferulic acid esterases involved indegradation of complex cell wall polysaccharides. Applied Environmental Microbiology63(12), 4638–4644.

Ward, D.M., Bateson, M.M., Weller, R. and Ruff-Roberts, A.L. (1992) Ribosomal RNAanalysis of microorganisms as they occur in nature. In: Marshall, K.C. (ed.) AdvancedMicrobial Ecology. Plenum Press, New York, 12, pp. 219–286.

Weller, R. and Ward, D.M. (1989) Selective recovery of 16S rRNA sequences from naturalmicrobial communities in the form of cDNA. Applied Environmental Microbiology 55,1818–1822.

Wells, J.A., Cunningham, B.C., Graycar, T.P. and Estell, D.A. (1987) Recruitment ofsubstrate-specificity properties from one enzyme into a related one by proteinengineering. Proceedings of the National Academy of Sciences USA 84, 5167–5171.

Williams, S.T., Goodfellow, M. and Vickers, J.C. (1984) New microbes from old habitats?In: Kelley, D.P. and Carr, N.G. (eds) The Microbe 1984, part II: Prokaryotes andEukaryotes. Cambridge University Press, UK, pp. 219–256.

Willmitzer, L. (1993) Transgenic plants. In: Rehm, H.-J. and Reed, G. (eds) Biotechnology 2.VCH, Weinheim, pp. 627–659.

Wohlleben, W., Muth, G. and Kalinowski, J. (1993) Genetic engineering of Gram positivebacteria. In: Rehm, H.-J. and Reed, G. (eds) Biotechnology 2. VCH, Weinheim,pp. 455–505.

Wood, W.A. and Kellogg, S.T. (1988) Biomass, part A. In: Abelson, J.N. and Simon, M.I.(eds) Methods in Enzymology. Academic Press, San Diego, 774 pp.


P. SteenLiquid Application Systems for Feed Enzymes14

Introduction

Enzymes have found widespread application in those countries where animal dietsare based principally upon wheat, barley, oats, triticale or rye. The rapid commercialuptake of this new technology in recent years has resulted from extensive research,yielding a greater understanding of the anti-nutritive properties of these cereals andof the enzyme types and dosages required to reduce their negative effects. This in turnhas stimulated the commercial availability of a number of feed enzyme products thatcan offer consistent and economic benefits to the user.

All animals use enzymes in the digestion of food. These are produced either bythe animal itself or by the microbes naturally present in its digestive tract. However,the digestive process is less than 100% efficient. Therefore, the supplementation ofanimal feeds with enzymes to increase the efficiency of digestion can be viewed as anextension of the animal’s own digestive process.

Recent developments in feed processing technology have resulted in increasedprocessing temperatures (e.g. expanders). This has provided benefits such asimprovements in feed hygiene, production efficiency and pellet durability. However,the increases in processing temperature have also focused attention on thermo-stability of some of the more heat-sensitive feed ingredients, such as vitamins, aminoacids and enzymes. This has presented an interesting dilemma for the feed industry instriking a compromise between what these ingredients can endure and what processconditions will allow. The situation is confounded by the fact that there are no setstandards worldwide for feed processing. In effect there are almost as many variationsin feed conditioning retention times, temperatures and pellet die sizes as there arefeed mills (Pickford, 1992).

Enzymes are proteins and can therefore be susceptible to degradation by externalfactors such as pH, hydrothermal conditioning and frictional forces and by the heavymetals that are added to certain animal feeds. Stability of granular enzymes can beimproved through the selection of activities with inherent high thermal stability,selection of specific carriers and through novel manufacturing techniques such ascoating. Consequently, some granular feed enzymes have been shown to maintain


Liquid Application Systemsfor Feed Enzymes

P. STEENFinnfeeds, PO Box 777, Marlborough, Wiltshire SN8 IXN, UK

14

efficacy after exposure to conditioning temperatures of up to 90°C. However, whereprocessing conditions exceed this temperature it is generally advisable to applyenzymes as liquids, post-pelleting, thereby avoiding exposure to high temperatures.

Application of Liquid Enzymes

Use of enzymes in the feed mill

As various pressures mount on the global animal feed industry, so the demand toproduce ‘hygienic food’ for farm animals has increased. This has resulted in theadoption of new processing technologies that have elevated processing temperaturesbeyond those achieved with conventional steam conditioning. The expander is onesuch example of a new processing technique. Typically, expanders are capable ofprocessing feed at 105–130°C and at pressures of 10–40 bar (145–580 psi). Hydro-thermal processes – expanders or extruders in particular – are often regarded as, atbest, neutral or even potentially destructive to certain feed additives (Pickford, 1992).

As feed-processing temperatures increase in areas of the world where viscouscereals are employed, so the intestinal viscosity of the targeted animal (broiler, layer,etc.) fed untreated diets will increase (Table 14.1). The application of enzymesaddresses this dilemma, although when feed-processing temperatures exceed 90°Cthe application of heat-sensitive additives such as enzymes, vitamins, probiotics andcertain antibiotics, etc., favours liquid addition at the end of the manufacturingprocess.

Addition of liquid enzymes

A major consideration for successful use of liquids is the method and accuracyof dosing. Whilst other micro-ingredients, such as amino acids, are routinely

354 P. Steen

Pelleted Expanded and pelleted

7–28 days Control + Granular enzyme Control + Liquid enzyme

Liveweight gain (g)Feed : gain1

Foregut viscosity (mPa s)(CV)

1042ab.371.37bc

6.9ab

(22)

1060aa.371.34ab

4.3ba

(13)

1010b.371.39c

20.0c

(44)

1064a.371.31a

3.8b

(11)

a–cP < 0.05.1Feed intake corrected to dry matter.60% wheat, 20% soybean meal, 5% sunflower oil; pelleting = 75°C, expanding = 105°C/AmandusKahl; batch-type liquid dosing system.

Table 14.1. Effect of (i) pelleting and (ii) expanding, followed by pelleting, on broiler performanceand foregut viscosity (Harker, 1996).

incorporated as liquids, they are dosed directly into the mixer as opposed to post-pelleting. Dosing into the mixer is a relatively simple process typically involvingonly a dosing pump, flow meter and spray nozzle. Some micro-ingredients, such asflavours/spices, are sprayed on to pelleted feed via low-cost and relatively inaccuratespray systems. However, the relative value of these products and importance ofdosing accuracy is typically low compared with feed enzymes. Feed enzymes requirea different approach due to the role they play in the feed formulation and thenutritional value they add to the feed. Ineffective application of the enzyme to thefeed can result in an energy-deficient diet leaving the feed mill, and subsequent poorperformance of the targeted animal.

There are two recognized positions within the feed manufacturing processfor the application of enzymes: in process (post-pellet sieving) or at bulk outloadingas the finished feed is loaded into the delivery truck.

The advantage of the installation in the loading line is that only one post-pelleting application (PPA) device is required, though it must be correspondinglylarge. In the other case, several spray systems are necessary, depending on the numberof pelleting lines. Application directly before loading additionally offers the option ofmanufacturing compound feed of one particular composition in large quantities andthen adjusting it to individual customer wishes directly prior to loading. Further-more, unwanted dispersal or cross-contamination with critical components such asliquid medication can be largely ruled out in this way. This simplifies the productionprocess and improves accuracy (Schwarz, 1998).

Application of Enzymes in the Feed Manufacturing Process

In any post-pelleting liquid application, we must first consider the upstream anddownstream processing equipment and how it operates. We must also consider themaximum processing capacity, which will not necessarily be the pelleting capacity.These two factors are essential to the successful design and operation of theequipment.

Today, many feed mills have installed counter-flow coolers. These coolersoperate in a semi-continuous mode, retaining the product in the cooler for apredetermined time, then discharging the product by opening the discharge gate,allowing product to flow en masse from the cooler. This discharge rate will behigher than the output capacity of the pellet press. Therefore, if a PPA system isinstalled to match the pelleting capacity, the plant will be bottlenecked at thePPA system, due to the reduction in capacity at this point in the process. In order toovercome this problem, the capacity of the PPA system should be calculated againstthe maximum transfer capacity (volumetric flow rate × maximum feed bulk density)of the mechanical handling device immediately below the cooler. In feed mills wherethe horizontal cooler is still employed, the feed rate from the cooler is consistent withthe pelleting capacity. It follows, therefore, that if the pelleting capacity changes thenthis upstream fluctuation will affect the operation of any downstream processingequipment unless adequate precautions are taken.

Liquid Application Systems for Feed Enzymes 355

Another critical aspect in the effective application of liquid enzymes is thelocation of the PPA system. If the feed is to be screened, for the removal of fines,before being transferred to the finished product bin(s), then it is essential that theapplication of the enzymes be applied to the feed post-screening. If not, it has beendemonstrated that 24–50% of the enzyme activity can be recovered from the fines,which typically amounts to between 5 and 10% of the feed (Gunther et al., 1997; vanLaarhoven, 1998). Fines will be particularly high in enzyme activity, due to theirlarge surface area and absorptive capacity. If these fines are removed from the finishedproduct by sieving after the PPA of the enzymes, then the enzyme activity will returnto the pelleting system for reprocessing. As the thermal stability of liquid enzymes isfar less than their granulated equivalents, then most of this activity risks denaturationduring subsequent re-processing of the feed. This will have a major influence on theoverall economics of enzyme use.

From Table 14.2 it can be seen that the xylanase activity level of the fines ismore than three times higher than the xylanase activity in the pellets. The fines areformed by attrition of the pellets and it can be concluded from these experimentsthat the absorption of the enzyme into the centre of the pellet is very limited, withthe result that the enzyme activity is confined to the outer layers (Engelen, 1998).Consequently the better the pellet quality, in terms of durability and hardness, thehigher will be the enzyme recovery during subsequent enzyme analysis.

356 P. Steen

Xylanase activity in samples(U kg−1)

Relative proportion of pellets andfines (%)* Total xylanase

activity in asample (U kg−1)Sample ID1 Pellets2 Fines Pellets Fines

A1A2A3A4A5A6A7A8A9A10

6210.76410.76380.75960.75430.77480.75480.75995.75490.75720.7

19,580.019,260.019,720.019,690.019,550.019,670.019,240.019,800.019,470.019,140.0

89.689.885.588.393.492.794.391.591.394.8

10.410.214.411.6

6.67.35.78.58.65.2

7600.37717.48308.37557.36362.38369.46259.27168.16698.56414.2

AverageSTDCV

6056.7587.5

9.7

19,512.019,216.019,261.1

91.1––

8.8––

7245.5747.5

10.3

1Water activity of feed 0.86 at 25°C.2Pellet quality: Holmen 76.4%; Kahl 2.0 ± 0.5.*Xylanase is sprayed on the feed in a batch mixer and the resultant pellets and fines are analysedseparately for enzyme activity.

Table 14.2. Xylanase activity in pellets and fines for different samples (Engelen, 1998).

Eliminating fines from the feed-manufacturing process may be virtuallyimpossible, but there are devices on the market, in addition to the pellet sieve,to improve pellet quality. Although pellet quality will not directly improve theefficiency of the PPA system, it will reduce the analytical variation of the assay byreducing the percentage of fines in the final sample.

Post-pelleting Application of Liquid Enzymes

Measurement and control of the liquid and dry flow rates

The accuracy of any liquid application system is based upon a number of essentialfactors; the most fundamental of these is the measurement and control of the liquidand dry flow rates. To achieve high homogeneity of application (low coefficient ofvariation) and precision dosing of the enzyme on the feed it is essential to control,and not simply measure, the dry flow rate. This is because fluctuations in the dryflow rate can result from upstream processing equipment, changes in manufacturingcapacity, variations in formulation, etc. Without the necessary control logic tomanage these immediate changes, long feedback times will render any advantagesexpected from measuring the dry flow an expensive exercise.

There are principally two methods for controlling the dry and liquid flowrates, namely by volumetric or gravimetric means (Figs 14.1 and 14.2). With thevolumetric system, both the liquid and dry elements are dosed independently ofeach other. No control or adjustment is made for variations in specific gravity or bulkdensity. Therefore, both elements are dosed on a theoretical capacity. Accuracy of thevolumetric system is based upon the variations in the manufacturing process for thepellets (bulk density and product form) and repeatability of the dosing equipment forthe liquids, plus any changes in specific gravity of the liquid.

The gravimetric method of measurement is when both the liquid and dryelements are gravimetrically weighed or measured in mass terms (mass/time). Thedry flow can be measured in a number of different ways, e.g. impact weigher, beltweigher, batch weight, etc. The liquid flow rate has fewer options for the gravimetricmeasurement of the flow, being limited to the batch weigher in the case of a batchsystem or as a loss-in-weigh system for measurement of presumed use. The morecommon option is the use of a mass flow meter based on the Coriolis measuringprinciple.

The major benefit of the gravimetric (mass rate of flow) system has to be theaccuracy that is achievable. With the gravimetric system, both the dry flow and liquidflow rates are gravimetrically measured, with the flow rate being equated to mass. Ifthe flow rate of either the dry or liquid changes, then the corresponding change ismade to the other component to offset any inaccuracies. This type of system is notaffected by changes in bulk density or specific gravity, flow rates or flow profile,although the more constant the flow rate, the greater is the degree of accuracy.Economically the level of automation needed for such systems makes it feasible toincorporate a number of ‘production aids’ to the benefit of the feed mill operator


and mill manager. It also makes it possible to link the application control systemwith the mill formulation computer so that the required inclusion rate for a specificformulation can be set by the nutritionist within the formulation computer.

Volumetric systems are lower in specification and therefore lower in cost. Forthe same reasons the installation of the equipment is much simpler. This type ofsystem comprises a dosing pump without the use of a flow meter and relies solely onthe calibrated displacement and repeatability of the dosing pump to deliver a knownvolume of liquid. The other component of the volumetric system is the volumetric

358 P. Steen

Fig. 14.1. Typical mill schematic arrangement for a gravimetric system.

flow of the pellets. The major deficiency of the volumetric system is the non-relationship between the two flow rates, dry or liquid, i.e. if the capacity of pellet flowincreases, the liquid flow rate does not. It is relatively easy, therefore, to appreciatehow the capacity of this type of system can change when manufacturing a differentproduct form or simply a change in bulk density or specific gravity of the liquid.


Fig. 14.2. Typical mill schematic showing the arrangement for a volumetricsystem.

Application System Types

The three commonest systems for the application of liquids are the semi-continuous,continuous and batch type.

Semi-continuous system

The semi-continuous system is the most commonly adopted system for the post-pelleting addition of liquids. This type of system was introduced for economicreasons, balanced by its flexibility and accuracy. Initially used as a volumetric systemfor the addition of fat, in recent years the system has been further developed toincorporate gravimetric measurement for the addition of aqueous-based liquids.

The installation of a surge hopper to accumulate the flow of dry product willeliminate any flow fluctuations from upstream processing equipment. The hopperhas sufficient capacity to accommodate a volume of product, predetermined tomaximize the operation of the dosing system, therefore limiting the frequency withwhich the complete system has to stop and start. The hopper is fitted with a dischargedevice that is automatically sequenced to operate by the level of the product in thesurge hopper. This is achieved by the inclusion in the hopper of a working-level andlow-level probe. The control logic of the system is that when the low level and thenthe working level are covered by product, the discharge device will operate. When the

360 P. Steen

Fig. 14.3. Sprout-Matador Micro Fluid System (Sprout Matador Ltd).

working level and then the low level are uncovered, the discharge device will stop.The position of the low-level probe is such that a residue of product will alwaysremain in the feeder and the surge hopper. This mode of operation ensures that whenthe discharger next operates the volumetric discharge capacity is as the previousoperation, eliminating any surging effect.

Advances in technology have seen the introduction of the gravimetric weighfeeder. Although the system operates in the same way as detailed above, the feeder issuspended on a fulcrum immediately under the outlet of the surge hopper. Theopposing end of the feeder is suspended on a load cell, which measures the productflow through the feeder.

The volumetric feeder ensures a constant volumetric flow rate from thefeeder without compensating for changes in bulk density. This can be a distinctdisadvantage if a number of different products are to be manufactured. However, it ispossible to compensate for these changes, either manually or automatically, if thechanges in the bulk density fall within a range. Manually, the mill operator can adjusta feed gate, precalibrated, at the feeder inlet, and the capacity of the system can bealtered by either increasing or decreasing the bed-depth in the feeder. The same canbe completed automatically by the use of a variable speed drive fitted to the feeder.This degree of control can provide the flexibility to maintain a constant throughputcapacity, given the variations of feed bulk density.

Continuous system

The processing of friable products is best suited to the continuous system, due to theability of the system to follow the manufacturing rate. As the operating capacity of theapplication system is matched to the upstream processing equipment, it is possible tooptimize the system with only a relatively small surge hopper, so preventing degrada-tion of the product but also optimizing the dry and liquid flow rates (Fig. 14.3).

For the continuous system to operate effectively, it must be capable of adjustingto upstream fluctuations. For this to be possible, it must be capable of determiningthe upstream capacity and adjust accordingly the operating capacity of the dosingsystem. This can be achieved by the installation of an ultrasonic level measurement,positioned in the surge hopper before the dosing system. The device is calibratedwith an empty hopper and with a full hopper, the scale in between being calibratedagainst the known manufacturing rates for the given products. The capacity of thedosing system increases as the hopper fills and decreases as the level in the hopperdiminishes. The objective is to maintain a consistent capacity through the dosingsystem by minimizing any large fluctuations in the capacity of the system.

The control of the dosing system must be such that it can respond to thenecessary changes in capacity. A continuous liquid dosing system requires extraaccuracy in determining the pellet flow and the liquid flow (Engelen, 1998). Thismeans that both the dry and liquid flow rates have to be changed proportionally. Forthis to be possible, both elements have to be gravimetrically measured to ensure theaccuracy of the system.


Batch system

The batch system is a discontinuous system that operates (as the name implies) on abatch-by-batch basis. The system provides the flexibility to apply multiple liquidsand combine multiple dry product forms. There are several different arrangementsfor the batch system and these can be defined in two categories, as either a pre-weighing system or a combined weigher–mixer.

The weigher–mixer combination is when the batch mixer is supported onload cells and performs the dual function of both weighing and mixing. The maindisadvantage of this system is that the capacity of the mixer needs to be greater for thesame manufacturing capacity as a pre-weighing system, as the batch cycle time isincreased. This will increase the cost of the mixer, but this cost may be offset againstthe cost of the weigh hopper(s) and associated control items. It may also rule out therequirement for additional processing equipment such as conveyors and elevators, asthe height required for the installation is reduced.

The pre-weighing system shown in Fig. 14.4 comprises a twin weigh hopperarrangement. This system is configured for the batch weighing of both the dry mate-rial and liquid additives. The principle of the batch system is to operate using theactual batch weight and not the theoretical batch weight. For this reason, the systemis extremely accurate, proportioning the correct quantity of liquid additives to thecorresponding weight of dry material. For this principle to work, the batch weight ofthe dry material must be known before weighing the liquid additives. Therefore,when the initial batch of dry material is weighed, the corresponding batch weight forthe liquid additives is then weighed. For the succeeding batches, the weighing of boththe dry material and liquids is completed before the finish of the preceding batch.

Batch-weighing multiple liquids can have its disadvantages. The principle forthe addition of micro-liquids to the batch mixers differs from the addition of fat andother high-inclusion liquids. If these liquids are weighed in the same weigher, then acompromise may have to be made regarding the addition method to the mixer, i.e.addition time, nozzle type, nozzle capacity, etc.

The more commonly adopted method for the addition of micro-liquids to thebatch mixer is to meter each liquid individually. This provides finite control andflexibility. Unlike other systems, the batch system does not require continuousmeasurement and control of the liquid flow. The liquid addition is calibrated to dosethe theoretical quantity of liquid to the mixer in a predetermined time, the additiontime being important to the performance of the system. The dosing capacity of thepump being constant, the enzyme is metered until the set point is achieved.

Measurement and Control of the Liquid Flow Rate

The dosing precision of liquid enzymes is critical, due to the known dose–responserelationship that exists between enzyme dose and animal performance (Fig. 14.5).Under-dosing can result in significantly reduced efficacy and bird response,

362 P. Steen

ultimately reducing feed quality, whilst over-dosing can be costly in terms of wastageof feed enzymes and reduced animal performance under certain circumstances.

Soon, enzymes may not be blanket-fed throughout the animal’s life but targetedwith various dose levels at different stages of growth or reproduction (Overfield,1999a). In addition, some enzyme suppliers are developing modelling tools able topredict the optimum dose rate based on cereal quality. For this to be of practical usewithin the feed mill, the dosing system needs to be such that the application rate of


Fig. 14.4. Typical mill arrangement for a batch system.

the enzymes can be varied easily, without being recalibrated, and yet still maintain itsaccuracy.

Flow meters

The measurement and control of the liquid flow rate can be achieved with theincorporation of a flow meter in the liquid line. The process signal from the flowmeter is used to vary the speed of the drive motor on the dosing pump, therebyaltering the liquid flow rate. The process signal from the flow meter is proportionalto the required application rate and dry flow capacity.

There are three main types of meters used in the measurement of liquids and abrief explanation of each of these follows.

Positive displacement or turbine metersThese are mechanically operated flow meters. The internal arrangement of the metercontains moving parts – either oval gears or a Pelton wheel type arrangement. Therotation of the rotors is sensed by an electronic system, using the interaction of aferrite material imbedded in the rotor tips. A sensing device is mounted in the top ofthe meter body to pick up the rotation of the rotors. The sensor will then generateand transmit the resulting signal. The speed of the rotor and the resulting output aredirectly proportional to the linear flow rate of the liquid through the meter.

The main concern with this type of meter is the low flow rate used in theapplication of enzymes and the possible build-up of material on the gears of themeter. Either of these parameters can affect the accuracy of the meter or prevent itfrom operating.

364 P. Steen

Fig. 14.5. Non-linear (exponential) dose–response curve for five trials combined(M. Bedford, Finnfeeds, personal communication, 1999).

Electromagnetic metersThis type of meter has no moving parts, which eliminates the possibility ofentrapment of material in the sensing unit. The measuring principle is basedon Faraday’s law of magnetic induction. This is when a voltage is induced into aconductor, which moves in a magnetic field, the flowing liquid being the conductor.The induced voltage is proportionally related to the flow velocity, and throughknowledge of the cross-sectional area of the meter sensor, the volumetric flowrate can be calculated. The application of the meter is limited to materials that areelectrically conductive above 5 µS cm−1.

Mass flow metersThe measuring principle for these meters is based on the controlled generation ofCoriolis forces. These forces are always present when both transitional (straight-line)and rotational movement occur simultaneously. When there is no flow, themeasuring tubes both oscillate in phase. When there is flow, the tubes’ oscillation isdecelerated at the inlet and accelerated at the outlet. As the flow rate increases, thephase difference also increases. The oscillation of the measuring tubes is determinedusing electrodynamic sensors at the inlet and outlet. The measuring principle of themass flow meter is independent of temperature, pressure, viscosity, conductivity orflow profile.

Dosing pumps

The other essential element in controlling the application of liquid enzymes isthe correct selection of the dosing pump. There are numerous pumps availableand capable of dispensing liquid enzymes. The most commonly adopted typeof pump for this application is the diaphragm pump, which can be mechanically orelectronically coupled. One of the benefits of the diaphragm pump is its abilityto meter volumes of liquid precisely. It is also self-priming and can handleparticulate-filled and shear-sensitive liquids.

The pump (Fig. 14.6) operates in two stages, with a positive and negative stroke.For the pump to operate efficiently, the suction valve (1) has to allow liquid intothe head on the suction stroke and allow it through the discharge valve (2) onthe delivery stroke. This is achieved due to the motion of the diaphragm. As thediaphragm moves backward (negative) on the suction stroke, a partial vacuum iscaused in the pump head. This lifts the suction ball valve, allowing liquid to floodinto the pump head. The vacuum created also ensures that the delivery ball valveremains tightly seated. When the diaphragm moves forward (positive) on the deliverystroke, the suction ball valve is forced back on to its seat. The pressure in the pumphead rises until it is greater than the system pressure. At this point, the ball in thedelivery valve lifts and the liquid is displaced. Once the diaphragm has reached itslimit of travel, the system pressure on the delivery ball valve forces the ball back on toits seat. The cycle is then repeated.


Flow oscillation is common to all diaphragm pumps. To overcome this and toachieve a consistent and even flow through the flow meter, it is necessary to install apulsation damper after the pump installation. The gas in the pulsation damper iscompressed until the system pressure is overcome, resulting in low-pulsation flowfrom the discharge of the damper.

Enzyme Application to the Feed Pellets

Feed compounders need reassurance that the products they incorporate into theirfeed are not just present, but present in the desired quantity. It is impossible forenzymes to achieve repeatedly the perfect application criteria of 100% recoveryand a coefficient of variation of zero, given the external influencing factors, suchas assay procedure, sample collection, sample size, product quality, etc. To achieve ahomogeneous distribution, enzyme should be sprayed uniformly on to the feed, the

366 P. Steen

Fig. 14.6. Cross-section of a diaphragm pump head.

number of droplets per gram of feed must be sufficiently large, and the pelletssprayed with enzyme must be thoroughly distributed (Gunther et al., 1997).

There are two accepted methods for the application of enzymes on to feed. Oneinvolves spraying into a free-falling curtain of material, or a mechanically formedcurtain (mist-coater); the other is by spraying into a fluidized area of pellets (batchtype mixer, etc.). Fodge et al. (1997) conducted an extensive survey of the differentapplication points and equipment options for the PPA of enzymes. These data will beequally applicable to any heat-labile micro-ingredients, across a range of mill systems,and for this reason are detailed below.

Figure 14.7 shows the application of the enzyme at the mist-coater, followed bysubsequent mixing in distribution augers. The left-hand Y-axis indicates the level ofenzyme activity in millions of units per ton. The right-hand Y-axis indicates theuncorrected coefficient of variation (CV) (%).

Of the seven systems depicted in Fig. 14.7, system G yielded an uncorrectedcoefficient of variation (CV) of less than 10% and the smallest spread of averageactivities. The next best uniformities were observed with systems F, C and D.Systems A, B and E were the least uniform.

Both G and F were continuously fed systems in which an impact-style load cellis positioned in the feed line just prior to the roto-coaters. Feed strikes the impactplate, a proportional voltage signal from the impact plate is forwarded to a variable


Fig. 14.7. Application of enzyme at roto-coaters followed by mixing in distribu-tion augers (Fodge et al., 1997). Lower dotted line is the uncorrected CV, obtainedfrom spraying Hemicell into a batch mixer (use right y-axis). Upper dotted line isthe desired amount of enzyme per ton applied to the feed (use left y-axis). Aboveeach letter are ten data points where each point represents the average enzymeactivity for ten feed samples collected in approximately 60 s. The bar above eachletter is the average CV obtained over a period of several weeks from ten samplesets.

frequency drive that controls the output of the fat and enzyme pumps in synchronywith feed flowing into the roto-coaters. Feed slides off the impact plate in a curtain-shape, roughly 50 mm thick and 410 mm wide, directly above the roto-coater. Onespray nozzle was positioned on either side of this curtain and sprayed the enzyme onthe feed rather than applying it inside the roto-coater (Fodge et al., 1997).

A and B roto-coaters were semi-continuous fed systems that were very old andin a feed mill designed for about 50% of the current output. Both systems oftenhad problems and had never performed smoothly. The feed-flow rate varied by 10%,or more, and the mixing of the feed was poor. Systems C and D were batch-fedroto-coaters in a well-maintained old feed mill, also producing feed at greater thanthe design capacity. Preventive maintenance programmes for C and D were welldeveloped and effective, and the feed mill did not experience unusual problems thatslowed their performance. Roto-coater E was constructed by a feed mill maintenanceshop. It was fed semi-continuously, and although the accuracy was reasonable, theuniformity of mixing was poor (Fodge et al., 1997).

Applying enzymes to the feed at an existing fat coater is a widely adopted andaccurate solution. The above figures help to support this, but they also indicate that,if the correct operating practices are not adopted, then poor results can also occur,as demonstrated by systems A, B and E. In the case of volumetric systems it isvital that changes can be made for variations in product bulk density to compensatefor increase or decrease in the system capacity. Roto-coater type systems have beendeveloped primarily for the application of fat; however, testing with enzymes hasbeen conducted and has resulted in excellent uniform coating (Decksheimer, 1998).

Figure 14.8 shows the results of spraying enzyme on to pelleted feed in fourdifferent augers. The accuracy of enzyme application was undesirable in three of thefour and the CV was slightly less than 20% on two systems and worse than 20% onthe third system. In system A, enzyme application was linked to feed flow rate, whichwas measured, whereas in systems B–D it was not measured. The main problemswith using augers are: (i) the need for accurate measurement of feed flow rates inthem; and (ii) the difficulty of keeping liquid product from building up on either thewalls or the drive shaft in the auger (Fodge et al., 1997).

The results shown for system A are not typical for a measured flow rate, with allsamples indicating an over-application of enzyme. It may be the case here that theenzyme and dry flow rate were not correctly calibrated. Either the dry flow rate waslow for the given enzyme set point or the enzyme set point was high for the dry flowrate.

The application of enzyme in conveying augers is not generally recommended,due to (i) above. However, in applications such as a fat coater complete with a blend-ing screw, this configuration can provide good opportunities for enzyme application.Additionally, with care and diligence regarding the nozzle selection, positioning andair pressure, it is possible to overcome the associated problems with (ii) above.

The data collected from four feed lines where the enzyme was applied byspraying it on to curtains of feed falling from the end of weigh-belt coaters are shownin Fig. 14.9. Spray nozzles for spraying fat were positioned on either side of thiscurtain. For adding enzymes, nozzles were positioned on both sides of the curtain of

368 P. Steen

feed, either above or below the fat nozzles, and occasionally included the enzymewith the fat. System A operates at a constant rate, systems B–D at varying rates(Fodge et al., 1997).

Feed mills may not have roto-coaters, augers or horizontal weigh-belt coaters.To develop an alternative application site the product was applied beneath the coolerjust before the feed drops into a chain drag. The transition piece between the coolerexit and the chain drag was replaced with a redesigned transition piece that collectedthe feed forming a curtain of feed 25–50 mm thick. The new transition piececomprised two compartments, each with two internal baffles that helped to form thefeed into a curtain-like shape, and spray nozzles were inserted into the transitionbehind these curtains. Positioning the nozzles behind the curtain of feed helped toprotect the spray patterns from the airflow turbulence that is prevalent at these sites(Fodge et al., 1997).

Figure 14.10 shows the results obtained at four such transition systems installedin poultry feed mills. Accuracy and uniformity were quite good in A and B, accept-able in C but still problematic in D. System D had more airflow turbulence problemsthan the others. It was concluded that these modified systems were an acceptablemeans to apply heat-labile products and were cost effective (Fodge et al., 1997).

Spraying at transitions and spouts is a very contentious topic. From the dataabove it can be seen to work, but also in other circumstances to fail. It is a very


Fig. 14.8. Enzyme application in conveying augers (Fodge et al., 1997). Lowerdotted line is the uncorrected CV, obtained from spraying Hemicell into a batchmixer. Upper dotted line is the desired amount of enzyme per ton applied to thefeed. Above each letter are ten data points where each point represents the averageenzyme activity for ten feed samples collected in approximately 60 s. The barabove each letter is the average CV obtained over a period of several weeks fromten sample sets.

demanding application point, in that it needs constant inspection and continualmeasurement to ensure the system stays within the predefined limits of acceptance.

Proportion of sprayed pellets

The homogeneous distribution of the additive (liquid) on the feed is criticallydetermined by the proportion of the sprayed pellets. The system should work insuch a way that the CV is minimized. A value of less than 20% should be aimed for.The more pellets that are sprayed, the sooner this value is achieved (Schwarz, 1998).

The volume of liquid applied to the feed in a PPA system is relatively small (typi-cally 500–1000 g tt−1). To disperse this quantity of liquid across the feed, the use ofair atomizing nozzles is required. Air is used to shear the liquid; this mechanism pro-duces a broad spectrum of droplet sizes, often varying from submicron up to severalhundred micrometres in the same spray. A perspective of droplet diameters can begained by realizing that there are 1000 µm mm−1 and the required range of dropletsto be applied to the feed would need to have a mean of 300–400 µm. It is critical thatthe liquid is not over-atomized by the use of too high an air pressure. This will resultin a micro-mist, with many of the micro-droplets condensing on the internal surfacesof the application equipment. In addition, the smaller the droplet size, the smaller is

370 P. Steen

Fig. 14.9. Enzyme application results with weigh-belt coaters using spray nozzles(Fodge et al., 1997). Lower dotted line is the uncorrected CV, obtained fromspraying Hemicell into a batch mixer. Upper dotted line is the desired amount ofenzyme per ton applied to the feed. Above each letter are ten data points whereeach point represents the average enzyme activity for ten feed samples collectedin approximately 60 s. The bar above each letter is the average CV obtained overa period of several weeks from ten sample sets.

the mass and, as a result, droplets can be affected by the airflow in the applicationequipment, with the droplets being drawn away from the application point.

To maximize the exposure of enzyme and feed pellets, it is critical to presentboth elements so that the pellets come into contact with the enzyme spray pattern foras long as possible, and that the pellets are in suspension during this period. Thenozzle design, application or orientation should be such that, if used in a continuousor semi-continuous system of varying capacities, the varying flow rate through thenozzle would not in any way affect the exposure of the enzyme and pellets, reducingthe efficiency of the system.

Addition of water with enzymes

The inclusion rate of some concentrated enzymes can be as little as 50–100 g tt−1.This does not present problems for the metering or measurement of the liquid, but itdoes make it a considerable task to apply the liquid to the feed accurately. These con-centrated enzymes need to be diluted to increase the volume of liquid to feed,improving the exposure of the enzyme to the pellets.

Although the objective is a minimum level of water inclusion, a 1000 g tt−1

total inclusion has proved effective (Decksheimer, 1998). Figure 14.11 demonstrates


Fig. 14.10. Spraying in transitions beneath coolers, before augers (Fodge et al.,1997). Lower dotted line is the uncorrected CV, obtained from spraying Hemicellinto a batch mixer. Upper dotted line is the desired amount of enzyme per tonapplied to the feed. Above each letter are ten data points where each point repre-sents the average enzyme activity for ten feed samples collected in approximately60 s. The bar above each letter is the average CV obtained over a period of severalweeks from ten sample sets.

that acceptable coefficients of variation were achievable with enzymes sprayedon to samples of feed using different production rates and application points(Decksheimer, 1998).

Predilution of liquid enzyme products to aid application also raises potentialissues of product stability. In practice many companies advise against prolongedstorage of diluted products, in order to avoid the risk of microbial contamination(e.g. Schwarz, 1998). Any dilution with water should take place immediately prior tothe spray nozzle and mixing should be with a static mixer, without moving parts. Themixer, nozzle and connecting parts should also be automatically flushed after eachaddition (Overfield, 1999b).

Pre- and post-fat addition of enzymes

There are conflicting opinions regarding the application of enzymes at the fat coater.Should enzymes be applied to the feed before the fat and then be subjected to thehigh fat temperature, or applied post-fat application, with concern then focusing onthe enzyme’s ability to penetrate the fat coating on the pellets?

Assay work carried out with samples from commercial feed mills has shownthere to be no clear interaction between enzyme recovery and stability and pre- orpost-application of enzymes at the fat coater. Research cited by Engelen (1998)(Tables 14.3 and 14.4) also supports this view. The recovery of xylanase LC wasshown to be unchanged irrespective of the fat temperature, the fat level and thespraying order.

372 P. Steen

Fig. 14.11. Water/enzyme concentration at 1 kg tt−1 (Decksheimer, 1998).Samples 1–2, pelleted hog grower, 15 t h−1, spray chamber. Samples 3–6, pelleted17% layer, 22 t h−1, spray chamber. Samples 7–8, pelleted broiler, 45 t h−1,roto-coater. Samples 9–12, pelleted broiler, 15 t h−1, mixing auger.

Liquid Enzyme Stability

Liquid enzymes are inherently less storage stable than their granular counterparts. Inliquid form, water activity is sufficiently high for the component enzymes to remainactive, whereas in the granular form the low water activity renders the componentenzymes relatively inactive prior to ingestion by the animal. Commercially availableenzymes are typically less than 100% pure and often contain side enzyme activities inaddition to the desired main activity or activities. These side activities can includetrace levels of proteases which, even at low activity, can cause some degradation of themain activities (e.g. xylanase). An aqueous environment also introduces a significantrisk of microbial contamination and proliferation, which can cause rapid declinein enzyme activity in liquid products. The problem of storage stability is increasedfurther when different liquid enzymes are mixed to produce ‘multi-enzyme’ liquidproducts and when liquid enzyme products are stored at ambient temperatures indifferent parts of the world.

Enzyme suppliers generally guarantee the shelf-life of liquid enzyme productsfor between 3 and 6 months following the date of manufacture, if stored at or below22°C (e.g. Fig. 14.12). For shipment to tropical countries, where the ambientconditions exceed 22°C, suppliers should ship and store enzymes in refrigeratedcontainers that maintain the product at a constant temperature below 5°C priorto delivery to the feed mill. Once liquid enzymes have been applied to the feed,


Enzymeb recovery (%)

Dieta ControlEnzyme sprayedbefore fat spray

Enzyme sprayed immediatelyafter fat spraying

Enzyme sprayed 30min after fat spraying

StarterGrower

00

100100

90104

9999

a50% barley based.b2 l xylanase LC t−1 of feed.

Table 14.3. Effect of spraying fat before or after enzyme addition on measured enzyme recovery(Engelen, 1998).

Enzyme recovery (%) in feeda,b

Fat temperature (°C) Fat level: 1.5% Fat level: 3.0%

648088

114110105

118117117

a50% barley based.b2 l xylanase LC t−1 of feed.

Table 14.4. Effect of fat level and fat temperature on the recovery of a liquid enzyme (Engelen,1998).

the component enzymes are rendered inactive and stable during the period prior toconsumption by the animal.

Delivery of Enzymes

Liquid enzymes can be delivered in a range of different sized containers to the feedproducer’s requirements. The more commonly adopted sizes are either a 1000 lintermediate bulk container (IBC) or a 210 l barrel. Containers are manufacturedfrom high-density polyethylene with UV stabilizers.

Discharge from the 210 l container is from the 50 mm valve at the top of thecontainer. The discharge from the IBC is via a 50 mm valve located at the bottom ofthe IBC. The IBC has a 150 mm screw cap, which incorporates a degas/regas vent forbalancing internal and external pressure. The 1000 l container is self-emptying, dueto the bottom discharge and sloped bottom of the IBC. The IBC is mounted on apallet for handling by forklift truck.

The IBC offers greater flexibility, ease of handling and self-emptying, andminimizes risk of exposure to operators, etc. The 210 l barrel offers flexibility forinstallations when access/handling of IBCs is not possible.

Handling of Liquids

During the period prior to dosing, liquid enzymes must be stored carefully in the feedmill to maintain stability. Ideally the product should remain in its original containerand be stored in cool dry conditions. If transfer to another container or tank is neces-sary, then this must be sterilized before use and at regular intervals. These containersmust also be ‘closed’ to prevent ingress of dust and other sources of contamination.

374 P. Steen

Fig. 14.12. Storage stability of a Trichoderma xylanase, product stored for 18months at 22°C (Cultor Technology Centre, Finland).

Microbial contamination of liquid enzymes may not only reduce their stability priorto dosing, but also recontaminate the feed following heat treatment.

Appropriate health and safety precautions must be observed whilst handlingliquid enzymes. These precautions will be found in the supplier’s Material SafetyData Sheet.

Measurement of Dosing Accuracy vs. Bird Performance

Measurement of the accuracy of dosing enzymes on to feed pellets relies upon asensitive assay procedure. Currently, feed enzymes stand alone from many otherhigh-value feed additives due to the absence of universally accepted assays fortheir determination, both as products and when incorporated in-feed. The assayprocedures used will be influenced by the properties of the enzyme concerned, e.g.pH, temperature characteristics. The assay is undertaken to measure the effectivenessof the application of the enzymes to the feed, the main criteria for measurementbeing homogeneity (coefficient of variation) and absolute recovery of enzyme activityfrom the feed. It can be seen that, provided an accurate dosing system is employed,productive performance is similar whether the product is applied post-pelleting(liquid) or pre-pelleting (granular, Fig. 14.13).

Summary

The choices of liquid application system vary both in cost and in accuracy. The use ofthe simple volumetric dosing system to apply enzymes is an economical choice in


Fig. 14.13. Effect of temperature on response in broiler performance (0–42 days)to enzyme supplementation in granular (Az 1300) vs. liquid (Az 1310) products(Finnfeeds Int. Technical Report 1300.UK.98.29).

terms of the equipment only. Frequently such systems do not offer the accuracyrequired to exploit the economic relationship that exists between enzyme dose, targetsubstrate and resultant animal performance.

The installation of the dosing equipment must be at the right location in theprocess (post-screening) and be correctly sized for the feed mill capacity, which maynot necessarily be the same as the pelleting capacity. The measurement and control ofthe dry flow rate are as vital to accuracy of the system as the measurement and controlof the liquid flow rate.

The choice of system must include the present and future requirements ofthe feed producer, provide the flexibility for the system to dose different numbersof liquids at varied application rates, and handle the different number of productforms manufactured by the feed mill.

References

Decksheimer, D. (1998) Accurate application of liquid ingredients. Proceedings of the EasternNutrition Conference. Montreal, Canada, pp. 111–126.

Engelen, G.M.A. (1998) Technology of Liquid Additives in Post Pelleting Applications.Wageningen Institute of Animal Science, The Netherlands, 101 pp.

Fodge, D., Smith, A., Redman, C., Humg-Yu, H. and Treidel, B. (1997) Post-pelletingapplication of heat-labile products explored. Feedstuffs, 29 September, 18–26.

Gunther, C., Beudeker, R.F. and Vahl, J.L. (1997) Choosing a liquid enzyme system. FeedMix 5(5), 24–27.

Harker, A.J. (1996) Application of liquid enzymes in poultry feeds. Proceedings of the CarolinaNutrition Conference. Charlotte, North Carolina, 12 pp.

Overfield, N. (1999a) New standards with liquid application system. Agricultural SupplyIndustry 29(38), 8.

Overfield, N. (1999b) New liquid enzyme application system sets high standards. FeedCompounder, June/July, 16–20.

Pickford, J.R. (1992) Effects of processing on stability of heat labile nutrients in animal feeds.In: Garnsworthy, P.C., Haresign, W. and Cole, D.J.A. (eds) Recent Advances in AnimalNutrition. Butterworth-Heinemann, Oxford, UK, pp. 177–192.

Schwarz, G. (1998) Liquid feed additives, an alternative in modern animal nutrition (Part 2).Kraftfutter 11/98, 506–511.

van Laarhoven, H. (1998) Future developments in liquid applications of enzymes. FeedExpo 98, UK. The European Feed Seminar, 6 pp.

376 P. Steen

M.R. Bedford et al.Stability and Detection of Feed Enzymes15

Introduction

Enzymes are now routinely added to monogastric diets in many parts of the world. Itis clear that their use is based on determined benefits in animal performance, butproving their presence in manufactured animal feeds is a daunting task and has beenthe focus of much research. Enzymes are proteins and as such are susceptible to therigours of feed manufacture, as are all other feed proteins. Whilst the structuralconformation need not be maintained for feed proteins to deliver their benefit interms of amino acid content, it is critical that feed enzymes are not irreversiblydenatured during feed manufacture or they will no longer function. It is clear thatdetection of their activity in complete feeds is therefore important for confirmationof survival. Whilst it is not within the remit of this chapter to discuss the methodsavailable for their detection (this is covered in Chapters 4 and 15), the idiosyncrasiesof each major class of enzyme used in animal feed are of interest and will be covered.Issues addressed will include differences between source enzyme stability to thermaldenaturation in vitro, problems with analysis due to interaction with the feed matrix(and methods of limiting this effect), data from in-feed processing trials, and futuretrends.

The vast majority of mixed feed for poultry and swine undergoes some sortof processing. For many years, feed has been pelleted, a process whereby thefeed mixture is conditioned by adding steam, then pressed through a die. Pelletingincreases nutrient density, improves the handling characteristics of the feed, andreduces the microbial load. Pelleting commonly uses temperatures between 65 and95°C (Gibson, 1995), which can be damaging to heat-sensitive nutrients, includingenzymes.



Process Stability andMethods of Detection

of Feed Enzymes inComplete Diets

M.R. BEDFORD,1 F.G. SILVERSIDES2 ANDW.D. COWAN3

1Finnfeeds, PO Box 777, Marlborough, Wiltshire SN8 1XN, UK;2Denman Island, British Columbia, Canada; and 3Novo Nordisk,

Krogshoevej 36, Bagsvaerd, Denmark 2880

15



In the past few years, concern for feed-borne pathogens and factors of pelletquality have caused feed manufacturers to increase the processing temperature, timeand pressure, and feed has been submitted to double pelleting or expansion(Pickford, 1992). The increase in the severity of the processing treatment underlinesthe importance of enzyme stability. Several approaches have been taken to overcomethe difficulties, including avoidance of the treatment altogether by adding theenzymes as a liquid after the pellets have cooled. Despite the possibility ofpost-pelleting supplementation, feed enzymes are often added to the mash beforetreatment. The effects of heat may be reduced by protection of the enzymes withhydrophobic coatings or selection of heat-resistant enzymes.

Until recently, there had been limited research published on the survivability offeed enzymes (Chesson, 1993). Yet enzyme stability is of critical importance for theproducers of feed enzymes and we must assume that in-house evaluations precededthe marketing of enzyme preparations. Since 1993, several research trials have beenreported in the popular press or in symposia proceedings, and a few trials havebeen described in the peer-reviewed scientific literature. Clearly, measuring enzymeactivity in vitro, either in solution or in the feed, is of critical importance. Recentresearch suggests that measurements of in vitro activity must be tempered by theresults of tests on in vivo efficacy.

Phytase

A surprising amount of research has been reported on the temperature stability ofphytase, considering that phytase makes up less than 20% of commercial enzymeusage (Bedford and Schulze, 1998). The interest is likely because phytic acid, whichrenders phosphorus and other nutrients unavailable, is present in many plant sourcesof feed ingredients (Cheryan, 1980; Eeckhout and dePaepe, 1994; Ravindran et al.,1995) yet endogenous phytase activity in monogastric animals is minimal or absent(Pallauf and Rimbach, 1997). To complicate the situation, many of the same plantsources of phytic acid also contain significant amounts of phytase and the nutritionalproblem of phosphorus digestion is compounded by the environmental problem ofphosphorus deposition on the land. Phosphorus pollution has become a limitation toproduction in areas with intensive animal production.

Phytase enzymes are diverse in both their sources and their characteristics.Liu et al. (1998) summarized the literature to 1998 and showed that phytases frombacteria, fungi, yeast and plants have optimal temperatures that range from 45 to77°C, a difference of 32°C. A phytase isolated from Aspergillus niger was described byDvorakova et al. (1997) as having activity at temperatures between 25 and 65°C withan optimum at 55°C. Incubation of the enzyme for 10 min at 60°C destroyed 5% ofits original activity, while incubation at 80°C destroyed 80% of the activity. As partof an effort to find heat-stable enzymes, Wyss et al. (1998) studied heat denaturationof purified phytases isolated from A. fumigatus and A. niger. Both enzymes weredenatured at temperatures as low as 55°C. However, even when submitted to a tem-perature of 90°C, the enzyme from A. fumigatus refolded into an active conformation

378 M.R. Bedford et al.




on cooling, while that from A. niger did not. Some of the heat-resistant phytases willundoubtedly become available commercially in the near future.

Heat inactivation of enzyme in solution does not translate directly into heatinactivation in the feed, because of an interaction of the enzyme with the feed matrix.In effect, the feed ingredients can protect the enzyme from steam or elevatedtemperatures for a short period of time (Chesson, 1993). Measuring phytase activityin pelleted feed provides a more accurate evaluation of the commercial importance ofinactivation. Simons et al. (1990) added phytase to a ‘common pig feed’, which theyheated to either 50 or 65°C before pelleting. Heating to 50°C resulted in a pellettemperature of 78 or 81°C but did not lower the enzyme activity. However, heatingto 65°C produced a pellet temperature of 84 or 87°C, resulting in inactivation of17 and 54%, respectively, of the original enzyme activity. Gibson (1995) added threephytase preparations to a wheat-based feed and pelleted the feed at temperaturesof 65–95°C. Two of the preparations suffered inactivation even at 65°C and onlya commercially available stabilized preparation retained substantial activity withpelleting at 85°C or above. In addition to studying enzyme stability in solution,Wyss et al. (1998) added phytase enzymes to commercial feed before pelleting at75 or 85°C. Recovery of the activity was similar for the two enzymes after pelletingat 75°C, but pelleting at 85°C resulted in a greater inactivation of the enzyme fromA. niger than that from A. fumigatus, supporting their results on denaturationkinetics. Eeckhout et al. (1995) added a commercial phytase to feed and reportedthat pelleting temperatures between 69 and 74°C destroyed 50–65% of the enzymeactivity.

Inactivation is not limited to microbial enzymes that may be added to thefeed, but also affects the enzymes naturally contained in feed ingredients. Gibson(1995) found that pelleting at temperatures of 85°C or above largely inactivatedthe endogenous phytase in wheat. Eeckhout and dePaepe (1994), in their survey ofphytase activity in different feedstuffs, reported that pelleted samples of wheatbran, which is particularly rich in phytase, had only 56% of the phytase activity ofunpelleted wheat bran. In three experiments, Jongbloed and Kemme (1990) foundthat pelleting at temperatures approaching 80°C decreased enzyme activity in pigfeeds that were based on feedstuffs that were rich or poor in phytase activity. Theycarried their experimentation a step further by measuring the effect of pelleting onthe apparent absorption of phosphorus. In two experiments, they found thatpelleting of phytase-rich diets decreased phosphorus absorption, supporting the dataon endogenous phytase inactivation.

Research interest in phytase stability by both the research community and thefeed industry is due to the higher processing temperatures currently being usedand the increased importance of phosphorus absorption for both nutritionaland environmental reasons. Thermal protection of available exogenous enzymes byencapsulation or granulation provides a solution at present. A more fundamentalsolution may be the isolation of enzymes that are heat-resistant or that can reformtheir active structure after denaturation. Unfortunately, neither of these solutionsprevents high processing temperatures from destroying endogenous enzymescontained in feed sources.

Stability and Detection of Feed Enzymes 379




b-Glucanase

The commercial availability of enzyme preparations containing β-glucanase activityhas allowed barley to be incorporated into poultry diets without the depressedperformance and sticky droppings that are associated with high β-glucan levels(Campbell and Bedford, 1992). The use of β-glucanase is now widespread in areas ofthe world where barley is commonly grown. There are a limited number of reports onthe effect of heat treatment on β-glucanase enzyme that has been added to feed.

Eeckhout et al. (1995) measured the activity of a β-glucanase included in acommercial piglet feed conditioned at temperatures between 50 and 95°C andpelleted at temperatures between 72 and 91°C. Even at the lowest temperatures,40% of the original activity was lost, and at the highest temperatures, only 7% of theoriginal activity remained after processing. More than two thirds of enzyme inactiva-tion occurred in the conditioner. On the other hand, Esteve-Garcia et al. (1997) didnot find major inactivation of enzyme in broiler feed with conditioning and pelletingtemperatures approaching 80°C. The enzyme that they used was described as amicrogranulate, suggesting that it was incorporated into the feed in a stabilized form.

At least two trials have measured the performance of broiler chicks fedenzyme-supplemented diets that were submitted to heat treatment. McCracken et al.(1993) supplemented a barley-based diet with a stabilized commercial enzymemixture containing β-glucanase and xylanase activities and heated it to 85°C for15 min before pelleting. Heat treatment of the feed without exogenous enzymedecreased the apparent digestibility of nutrients, increased the viscosity of theintestinal contents, and reduced faecal dry matter content. The addition of theenzyme improved nutrient digestibility and eliminated the negative effects ofheat treatment, demonstrating clearly that the enzymes remained active at thistemperature. Vukic-Vranjes et al. (1994) tested a commercial enzyme mixture,including β-glucanase, xylanase, amylase and pectinase activities, in two diets, oneof which contained 20% barley. The feed was conditioned at 70–75°C, and pelletedor extruded with the temperature in the extruder barrel reaching 110–120°C.Compared with pelleting, extrusion produced negative effects on chick performance.Extrusion also resulted in increased in vitro viscosity of the feed, suggesting thatthe higher temperature increased the solubility of non-starch polysaccharides. Theenzyme mixture improved chick performance whether the diet was pelleted orextruded, likely indicating that there was enzyme activity even after exposure to theextrusion process. Vukic-Vranjes et al. (1994) also showed that supplementationwith the enzyme mixture reduced the viscosity of the feed extract, demonstrating thatthe enzyme was active in the feed before ingestion by the bird. They did not rule outenzyme action in the feed even before processing.

Inborr and Bedford (1994) measured both enzyme activity in the feed andbird performance after supplementation and heat treatment. These authors foundsignificant inactivation of added β-glucanase (the same product used by McCrackenet al., 1993) in a barley-based diet after conditioning at a temperature as low as 75°Cfollowed by pelleting. Conditioning at 95°C for 30 s and 15 min destroyed 84 and91%, respectively, of the original β-glucanase activity. Inborr and Bedford (1994)





wondered if measuring enzyme activity in vitro or in vivo was the most accuratemethod of evaluating the effect of processing on enzyme activity. While recovery ofenzyme activity was highest at the lowest processing temperatures and lowest atthe highest temperatures, this was not accurately reflected in bird performance. Thehighest value for weight gain and the lowest value for the feed-to-gain ratio wereobtained with the intermediate conditioning temperature of 85°C. Clearly, theenzyme remained active at this temperature, and the positive effects of heat treatmentoutweighed the negative effects of enzyme inactivation.

Pickford (1992) showed that stabilization of β-glucanase protected it from apelleting temperature of 75 but not 95°C. Cowan and Rasmussen (1993) reportedthat a commercial β-glucanase lost significant activity with pelleting at a temperatureabove 65°C, but that coating the same product provided protection to 75°C. Boththese reports are brief and there is very little published research relating specifically toincreasing the temperature stability of β-glucanase.

Xylanase

Xylanase represents by far the greatest share of worldwide enzyme sales (Bedford andSchultz, 1998) and its use allows high-viscosity wheat to be incorporated into poultrydiets without negative effects. Perhaps because of the economic importance, thereis considerably more research reported in the scientific literature on the stability ofxylanases than there is on the stability of β-glucanases.

Xylanase enzymes are produced by many species of fungi and bacteria, and varia-tion in the optimum conditions for activity of these is not surprising because eachorganism is best adapted to a specific environment. In a review, Bedford and Schulze(1998) show that the optimum temperature for a number of bacterial and fungalxylanases can range from 30 to 105°C and the optimum pH can range from 2.0 to10. Pickford (1992) compared the stability of three commercial enzyme preparationswithout providing information on the type of activities that they contained. At apelleting temperature of 80°C, 85% of the original activity was retained by one prep-aration, 55% by a second, and only 33% by the third. No enzyme activity was shownfor any preparation with pelleting at 95°C. Expansion, with a higher temperature butshorter exposure, caused greater loss with similar differences between enzymes.

Pettersson and Rasmussen (1997) demonstrated differences in temperaturestability of xylanase enzymes isolated from Thermomyces, Humicola and Trichoderma.At a conditioning temperature of 75°C, the Trichoderma enzyme lost significantactivity, while at 85°C both the Thermomyces and Humicola enzymes retained morethan 80% of their original activity. Even with conditioning at 95°C the Thermoycesxylanase retained 70% of its activity. Gibson (1995) observed variation between ninexylanase preparations, seven of which were commercially available at the time of thetrial. Activity remaining after treatment at 90°C was above 90% for one preparationand lower than 10% for several of the others. It is not clear from the report howmuch of this variation was due to the source of the enzyme and how much was due tostabilization of the enzyme. Perez-Vendrell et al. (1999a) tested the in vitro stability





of eight commercially available preparations, including coated or encapsulatedproducts, after feed processing at 65–70, 75–80 and 85–90°C before pelleting.Measurement in vitro showed that, even at the lowest temperatures, most productslost at least 30% of their activity, while at the highest temperatures, activity lossreached 90%. As they found for β-glucanase, Esteve-Garcia et al. (1997) reportedthat the microgranulated xylanase preparation that they studied was not inactivatedby conditioning and pelleting temperatures that approached 80°C.

In the preceding paragraph reference was made to several methods of thermalprotection. The most fundamental protection is that suggested by the study ofPettersson and Rasmussen (1997) in which heat-resistant enzymes are preparedfrom heat-resistant organisms. At the other extreme, the problem can be avoidedaltogether by spraying the enzyme on the finished pellets (Perez-Vendrell et al.,1999b), though this requires some additional equipment and an extra step in themanufacturing process.

Dry enzyme preparations can also be stabilized to increase heat resistance andallow their addition to the feed before pelleting without undue inactivation. Methodsof stabilization were the subject of a previous chapter. Much of the inactivation ofenzymes is due to the steam used for conditioning. Thus, stabilizing a preparationrelates largely to protecting the enzyme from the steam by absorbing it on to acarrier or by coating it with a hydrophobic substance (Cowan, 1996). Pickford(1992) provides an example whereby stabilization increased activity remainingafter pelleting at 75°C from 48 to 76%, and after pelleting at 95°C from 12 to 34%.Clearly, even protected enzymes have a limit of temperature stability; Steen (1999)suggests that if a feed is to be processed at a temperature above 90°C, even enzymesthat have been stabilized should be added after processing.

Tests of in vitro enzyme activity are valuable tools in determining enzymeactivity loss, and they suggest that exposure to even relatively low temperaturesresults in significant destruction. However, in vitro tests are but one tool amongmany. Clearly, enzyme activity in a buffer solution at the optimum pH provides onlya very limited view of activity under the conditions of feed processing and digestion.In fact, most investigators (Vukic-Vranjes et al., 1994; Pettersson and Rasmussen,1997; Perez-Vendrell et al., 1999a; and others) have recognized that interactionsbetween the enzyme and the feed are important and they have measured activity incomplete feeds. Measurement of the viscosity of feed extracts (Spring et al., 1996;Bedford et al., 1997) can provide a measurement of the effects of enzyme actionbefore the feed is ingested by the bird. Preston et al. (1999) demonstrated this actionby showing that even with apparently complete enzyme inactivation in the finishedfeed, supplementation resulted in improved bird performance.

To complete the story of enzyme action in relation to processing, the effectsof enzyme addition and heat treatment of the feed must be measured by birdperformance. Enzyme inactivation does not always translate directly into perfor-mance differences (Bedford et al., 1997; Perez-Vendrell et al., 1999a; Silversides andBedford, 1999). With processing temperatures between 65 and 105°C, Bedford et al.(1997) found that apparent inactivation of enzyme was linear, whereas performance(body-weight gain, feed conversion ratio) was best when the feed was processed at





81–83°C. Silversides and Bedford (1999) also demonstrated this effect of heattreatment, as is shown in Fig. 15.1. As the processing temperature increases, enzymeactivity decreases in a linear fashion (R2 = 0.97) while the change in broiler bodyweight with increasing processing temperatures was quadratic (R2 = 0.84). Graphingthe results for the feed conversion ratio produces very similar results (R2 = 0.98).The highest body weight and lowest feed conversion ratio were at a processingtemperature between 80 and 85°C. When such a performance response is achievedbut is juxtaposed with apparent enzyme inactivation in an in vitro assay, it suggeststwo possibilities: either the enzyme completes its function almost entirely in theconditioner, making post-pelleting analysis largely irrelevant, or enzyme activity isnot effectively measured by the assays used.

Data from enzymes derived from coated Thermomyces species do not seem to fallinto the same category as those from uncoated Trichoderma or Aspergillus, probablydue to a lack of interaction with the feed matrix of the former. Under such conditionsthe enzyme is more easily extracted and a correlation between performance andrecovered enzyme activity has been demonstrated (Cowan and Rasmussen, 1993;Pettersson and Rasmussen, 1997; Andersen and Dalbøge, 1999). These contrastingobservations clearly indicate that there is no single method that can be applied fordetermining subsequent bird performance from analysis of a feed sample whenidentity of the enzyme is unknown.

Improved performance with a decline in in vitro enzyme activity underlinesthe danger of relying solely on measuring activity, and it is not as contradictoryas it appears. Heat affects many things in addition to enzymes (Pickford, 1992).Moderate heat increases starch gelatinization and cell wall destruction, increasingthe availability of nutrients and subsequently improving performance as well. Thisis, after all, one of the principal reasons for pelleting feed. Higher processing



Fig. 15.1. Changes in broiler body weight (q) and xylanase activity (x) withincreasing processing temperatures. (From Silversides and Bedford, 1999.)



temperatures also cause increased solubilization of non-starch polysaccharides,leading to higher intestinal viscosity and reduced performance, and actuallyincreasing the need for exogenous enzymes. Silversides and Bedford (1999) showedthat without exogenous xylanase in a wheat-based diet, the viscosity of the intestinalcontents of broilers increased dramatically as processing temperature increased(Fig. 15.2). Addition of xylanase produced low digesta viscosity even with highprocessing temperatures, and the actual reduction in viscosity was greatest at thehighest temperatures. The greater enzyme action with high processing temperaturesis likely because more substrate is available for enzyme action. The enzyme may beactive before or during processing, reducing the viscosity measured either in vitro orin vivo. High processing temperatures reduce performance not only due to increaseddigesta viscosity and decreased exogenous enzyme action, but also because vitaminsand other enzymes are also inactivated, and starch and protein digestibility isdecreased. With heat-resistant or stabilized enzyme preparations available, heatdamage to vitamins and nutrients other than exogenous enzymes will likely limit theprocessing temperatures that will be used commercially.

Conclusions and Trends

The activity of enzymes in solution can be reduced at temperatures of 60°C or lower.Enzymes in mixed feed are somewhat protected by the feed ingredients and mostare likely to be reasonably stable at a slightly higher temperature. However, with



Fig. 15.2. Changes in the viscosity of the intestinal contents of broilers given feedsupplemented (w) or unsupplemented (z) with xylanase and processed at differenttemperatures. (From Silversides and Bedford, 1999.)



processing temperatures as high as 95°C there is a serious loss of enzyme activity.Several approaches have been used or suggested to reduce the negative effects ofheat treatment, including protecting the enzymes from steam penetration, usingheat-resistant enzymes or simply adding the enzymes as a liquid after processing.Feed enzymes may have some activity before the animal ingests the feed, so that totalloss of in vitro activity may not always mean total loss of benefit. This seems to be atleast partially dependent on the enzyme and the assay method. In some cases in vitroanalysis is misleading, whereas in others there is a clear link between enzyme contentand in vivo performance. Common processing temperatures cannot increaseindefinitely because of the increased damage to vitamins, proteins and starch, whichmay be even more susceptible to heat damage than exogenous feed enzymes. Thefurther development of enzymes selected for heat resistance and the utilization ofenzyme activity before processing or ingestion by the animal may be promisingapproaches to obtaining the maximum benefit of added enzymes for the feedindustry.

References

Andersen, L.D. and Dalbøge, H. (1999) Future trends in feed enzymes. Proceedings of theAmerican Soybean Association Joint Seminar, Bangkok, Thailand, pp. 1–9.

Bedford, M.R. and Schulz, H. (1998) Exogenous enzymes for pigs and poultry. NutritionResearch Reviews 11, 91–114.

Bedford, M.R., Pack, M. and Wyatt, C.L. (1997) Relevance of in feed analysis of enzymeactivity for prediction of bird performance in wheat-based diets. Poultry Science 76(Suppl. 1), 39.

Campbell, G.L. and Bedford, M.R. (1992) Enzyme applications for monogastric feeds: areview. Canadian Journal of Animal Science 72, 449–466.

Cheryan, M. (1980) Phytic acid interactions in food systems. CRC Critical Reviews in FoodScience and Nutrition 13, 297–335.

Chesson, A. (1993) Feed enzymes. Animal Feed Science and Technology 45, 65–79.Cowan, W.D. (1996) Animal feed. In: Godfrey, T. and West, S. (eds) Industrial Enzymology,

2nd edn. MacMillan Press, London, pp. 71–86.Cowan, W.D. and Rasmussen, P.B. (1993) Thermostability of microbial enzymes in expander

and pelleting process and application systems for post-pelleting addition. In: Wenk, C.and Boessinger, M. (eds) Proceedings of the 1st Symposium on Enzymes in AnimalNutrition. Kartause Ittingen, Switzerland, pp. 263–268.

Dvorakova, J., Volfova, O. and Kopeck, J. (1997) Characterisation of phytase produced byAspergillus niger. Folia Microbiology 42, 349–352.

Eeckhout, W. and dePaepe, M. (1994) Total phosphorus, phytate phosphorus and phytaseactivity in plant feedstuffs. Animal Feed Science and Technology 47, 19–29.

Eeckhout, M., De Schrijver, M. and Vanderbeke, E. (1995) The influence of processparameters on the stability of feed enzymes during steam-pelleting. In: vanHartingsveldt, W., Hessing, M., van der Lugt, J.T. and Somers, W.A.C. (eds) Proceedingsof the 2nd European Symposium on Feed Enzymes. Noordwijkerhout, The Netherlands,pp. 163–169.





Esteve-Garcia, E., Brufau, J., Pérez-Vendrell, A., Miquel, A. and Duven, K. (1997)Bioefficiency of enzyme preparations containing β-glucanase. Poultry Science 76,1728–1737.

Gibson, K. (1995) The pelleting stability of animal feed enzymes. In: van Hartingsveldt, W.,Hessing, M., van der Lugt, J.T. and Somers, W.A.C. (eds) Proceedings of the2nd European Symposium on Feed Enzymes. Noordwijkerhout, The Netherlands,pp. 157–162.

Inborr, J. and Bedford, M.R. (1994) Stability of feed enzymes to steam pelleting during feedprocessing. Animal Feed Science and Technology 46, 179–196.

Jongbloed, A.W. and Kemme, P.A. (1990) Effect of pelleting mixed feeds on phytase activityand the apparent absorbability of phosphorus and calcium in pigs. Animal Feed Scienceand Technology 28, 233–242.

Liu, B.-L., Rafiq, A., Tzeng, Y.-M. and Rob, A. (1998) The induction and characterization ofphytase and beyond. Enzyme and Microbial Technology 22, 415–424.

McCracken, K.J., Urquhart, R. and Bedford, M.R. (1993) Effect of heat treatment andenzyme supplementation of barley-based diets on performance of broiler chicks. Proceed-ings of the Nutrition Society 50A.

Pallauf, J. and Rimbach, G. (1997) Nutritional significance of phytic acid and phytase.Archives of Animal Nutrition 50, 301–319.

Perez-Vendrell, A.M., Fish, N.M., Sabatier, A.M., Frapin, D. and Geraert, P.A. (1999a)Thermostability of powder enzymes: in vitro recoveries and in vivo efficiencies.Proceedings of the Australian Poultry Science Symposium. Australian Poultry Science,Sydney, p. 172.

Perez-Vendrell, A.M., Brufau, J., Uzu, G. and Geraert, P.A. (1999b) Spraying enzyme beforeor after fat coating: in vitro recoveries and in vivo efficiencies. Proceedings of the AustralianPoultry Science Symposium. Australian Poultry Science, Sydney, pp. 105–107.

Pettersson, D. and Rasmussen, P.B. (1997) Improved heat stability of xylanases. Proceedingsof the Australian Poultry Science Symposium. Australian Poultry Science, Sydney,pp. 119–121.

Pickford, J.R. (1992) Effects of processing on the stability of heat labile nutrients in animalfeeds. In: Garnsworthy, P.C., Haresign, W. and Cole, D.J.A. (eds) Advances in AnimalNutrition. Butterworth Heinemann, Nottingham, pp. 177–192.

Preston, C.M., Ellis, S., Bedford, M.R. and McCracken, K.J. (1999) Effects of feed enzymesduring heat processing on diet characteristics and subsequent broiler performance.World’s Poultry Science Association in press.

Ravindran, V., Bryden, W.L. and Kornegay, E.T. (1995) Phytates: occurrence bioavailabilityand implications in poultry nutrition. Poultry and Avian Biology Reviews 6, 125–143.

Silversides, F.G. and Bedford, M.R. (1999) Effect of pelleting temperature on the recoveryand efficiency of a xylanase enzyme in wheat based diets. Poultry Science 78, 1184–1190.

Simons, P.C.M., Versteegh, H.A.J., Jongbloed, A.W., Kemme, P.A., Slump, P., Bos, K.D.,Wolters, M.G.E., Beudeker, R.F. and Verschoor, G.J. (1990) Improvement ofphosphorus availability by microbial phytase in broilers and pigs. British Journal ofNutrition 64, 525–540.

Spring, P., Newman, K.E., Wenk, C., Messikommer, R. and Vukic Vranjes, M. (1996) Effectof pelleting temperature on the activity of different enzymes. Poultry Science 75,357–361.

Steen, P. (1999) Modern milling requires thermostabile granular enzymes. Feed Technology3(3), 43–47.





Vukic-Vranjes, M., Pfirter, H.P. and Wenk, C. (1994) Influence of processing treatment andtype of cereal on the effect of dietary enzymes in broiler diets. Animal Feed Science andTechnology 46, 261–270.

Wyss, M., Pasamontes, L., Rémy, R., Kohler, J., Kusznir, E., Gadient, M., Müller, F. andvanLoon, A.P.G.M. (1998) Comparison of the thermostability properties of three acidphosphatases from molds: Aspergillus fumigatus phytase, A. niger phytase, and A. niger pH2.5 acid phosphatase. Applied and Environmental Microbiology 64, 4446–4451.





R.R. Marquardt and M.R. BedfordFuture Horizons16

Introduction

To date, the use of exogenous enzymes in animal diets has had a great impact on thelivestock industry. They are routinely used to improve the nutritive quality of dietscontaining cereals such as barley, wheat, rye and oats, especially for poultry, and as aresult they have improved the quality of the environment by reducing the output ofexcreta and pollutants, such as phosphate and nitrogen, including ammonia. Sinceenzymes tend to increase performance of animals fed low apparent metabolizableenergy (AME) cereals more so than those fed cereals with higher AME values, afurther benefit that results is improved animal uniformity within and betweenexperiments. In commercial situations this results in reduced variance betweenflocks.

Enzymes have also been shown to decrease the size of the gastrointestinal (GI)tract, which increases partitioning of nutrients into edible tissue. Their use has alsobeen shown to alter microbial fermentation, which again may affect the availability ofnutrients and the health status of the animal. Most of the research on enzymes as feedadditives has been with poultry but studies with pigs, especially young pigs, have alsodemonstrated a positive response to enzyme supplements. The recent appearance onthe market of recombinant enzymes, especially phytase, should further accelerate theadoption of enzymes by the feed industry. For some recent reviews on the use ofenzymes in the feed industry, see Annison and Choct (1991), Bedford (1995) andJeroch et al. (1995).

Although enzymes have proved to be highly beneficial, their use is still in itsinfancy. Many problems need to be solved before their full potential is reached. Thepurpose of this chapter is to highlight areas of further research that should enhancenot only the use but also the value of exogenous enzymes in animal nutrition.



Future HorizonsR.R. MARQUARDT1 AND

M.R. BEDFORD2

1Department of Animal Science, University of Manitoba,Winnipeg, Manitoba, Canada R3T 2N2; 2Finnfeeds, PO Box 777,

Marlborough, Wiltshire SN8 1XN, UK

16



Areas Requiring Additional Research

Improved enzyme assays

Of great concern to current users of enzymes is not only how to select an enzymefrom the multitude on offer, but also how to assay for the presence of the product incomplete animal feed. Currently, there is no single standard procedure for assessingthe quality or quantity of all commercial enzyme products, nor has a satisfactory assaybeen developed for monitoring quantities of enzymes present in complete diets. Ifxylanases are taken as an example, the current commercial products are derived fromat least four different genera of organisms. Even within one genus, some productscontain one, two, three or even more xylanases of differing characteristics (pH andtemperature optimum, substrate specificity, etc.). As a result it is not valid to assignone assay for determination of the ‘relative activity’ of all products since any oneproduct can be made to appear superior by judicious choice of assay conditions.A further problem is that, once added to a complete feed, some xylanases exhibit asignificant binding to the feed matrix, and so estimation of activity is difficult toaccomplish (Bedford, 1993).

Some of the commonly used enzyme assays include: measurements of thereducing sugars liberated by the enzymes; the use of coloured substrates;immunological methods, such as the enzyme-linked immunosorbent assay (ELISA);and assays based on the ability of an enzyme to reduce the viscosity of water-solublenon-starch polysaccharides (WSNSPs) (Cowan and Rasmussen, 1993; Hendon andWalsh, 1993). Cowan and Rasmussen (1993) reported that, among the differentmethods, the release of dye from dyed substrate is one of the easiest and mostsensitive methods but it is not sensitive enough to detect enzymes in feeds readily.The ELISA procedure can detect enzymes in feeds, but antisera are not availablefor all of the different enzymes, and ELISA sometimes shows a weak reaction toinactivated enzymes. Measurement of the reducing-sugars liberated by enzymes in afeed matrix is inaccurate because of the high background level of reduced sugarsin feed. A viscosity-based assay has been reported but its use is not common incomplete feeds (Vahjen et al., 1997). McCleary discusses the strengths andweaknesses of each of these methods and others in more detail in Chapter 4. Thelimitations of the current situation do point to a major goal for future researchwhich, if successful, will significantly improve the trust that feed manufacturers placein enzymes. This goal is to produce a meaningful assay, or series of assays, that willdetermine not only the amount of enzyme present in the complete feed, but also itsbiological relevance. Clearly, there should be an association between the activitydetermined in the feed and its activity in vivo.

Government regulatory bodies, enzyme manufacturers, the feed industry andprofessional societies should therefore benefit from development of standardizedassays for feed enzymes. This is important, as standardized assays would: (i) allow themanufacturers and purchasers of enzymes to compare enzymes based on their activityvalues; (ii) enable livestock producers to determine the amount of active-enzymeproduct that was added to a diet or the amount that survived the rigours of the

390 R.R. Marquardt and M.R. Bedford




environment, including heat treatment during processing; (iii) provide a means ofassessing how well the enzymes survive in different sections of the GI tract, especiallyin the section where they are most efficacious (the benefits of this are discussed in thenext section); and (iv) bear some relation to the physiological effects that the enzymesare supposed to have (for example, it may be possible to run an assay at pH 8 and60°C, but the results would be irrelevant to the feed industry).

Site of action of enzymes

Fundamental information is lacking on where in the GI tract enzymes produce mostof their beneficial effects. It is not known, for example, whether the main site ofaction of enzymes in chickens is in the crop, proventriculus, gizzard, duodenum,ileum, rectum, or all or part of the GI tract. Data presented in Chapter 15 suggestthat enzyme activity may be as much as 60–70% completed during the conditioningprocess of a wheat-based diet at 85°C. Confirmation of such information may assistin the selection of enzymes appropriate for the target substrate and the conditions atthe site where they are most efficacious. The type of enzymes selected for use in poul-try, for example, could be very different if the major site of action turned out to bethe crop rather than the lower sections of the GI tract. The ability of the enzymes toresist proteolysis, their optimal operating temperature (particularly if most activityis expended during feed manufacture) and pH would be considerations for theirselection. Sites of activity may also vary considerably between animal species, sincethe conditions in the GI tract vary considerably between pigs, poultry and ruminants.As a result, whereas today most enzymes destined for use in pig diets are derived fromthose used in products designed for poultry, it is likely that future ruminant, swineand poultry products will be designed specifically for the requirements of the targetspecies. Indeed, differences clearly exist between young and older animals of the samespecies, and within the grouping of poultry there are significant differences betweenturkeys and chickens, for example. Determination of the sites and modes of action ofspecific enzymes in each of these species will ultimately lead to better products thatare more cost efficient for the industry as a whole.

Production of new enzymes

Whilst the current generation of enzymes are highly beneficial, their usefulnesswill undoubtedly be surpassed when new activities with greater activity are madeavailable. Sources other than those currently used, such as those produced bymicroorganisms in the rumen of cattle, from thermophilic organisms or from otherextremophiles, will probably provide enzymes with the properties desired formaximal activity. Such properties will include: (i) highly specific activities (substrateturnover rates per unit of protein) under the conditions where they act; (ii) highlevels of resistance to inactivation by heat treatment, low pH and proteolyticenzymes; (iii) low production costs; (iv) long shelf-life under ambient storage

Future Horizons 391




conditions; (v) lack of apparent binding to feed matrix to facilitate the quantitativedetermination of enzyme in finished feed; and (vi) specificity to suit the purpose forwhich they are designed.

Substrate and product identification

The focus so far has been on the developments in enzyme properties that are likely tooccur in the next few years. Without a doubt the advances will only be made possiblethrough better identification of the target substrates and the desired products ofenzyme activity. Such knowledge will permit more efficient, rapid and focuseddevelopment of enzymes that will have a biological effect, and will hopefully supplantthe current empirical approach that is largely used for product development.

Many researchers have hypothesized that the principal mode of action of mostxylanase and β-glucanase enzymes is the destruction of gel-forming polysaccharidesleached from cereal cell walls in amounts sufficient to depress performance (Annisonand Choct, 1991; Bedford, 1993; Chesson, 1993). Hesselman and Aman (1986)proposed an alternative explanation: namely, that the β-glucans and arabinoxylansthat form the endosperm wall of cereals restrict the access of enzymes to nutrients.They postulated that the disruption of intact walls and the release of entrappednutrients are the major factors in the improvement of the nutritive value of dietswith exogenous enzymes. Chesson (1993) postulated that a single enzyme shouldbe effective if the beneficial effects of enzymes are attributable solely to reducedviscosity. The reason is that viscosity is partially a function of chain length and so it isonly necessary to break the chain in a few sites to reduce substantially or destroy itsgel-forming capacity. However, if the beneficial effects of added enzymes are due todisruption of intact cell walls and release of entrapped nutrients, rather than reducedviscosity, then many more enzymes would be required. Regardless of mechanism, it isfair to say that the actual structure of the anti-nutritive factor in the intestinal lumenis not known. If the structure were known, then in vitro work using such a substratewould enable rapid screening of new enzyme candidates for use in new productdesign. With modern discovery and mutagenic techniques it is possible to supplymany hundreds, if not thousands, of candidates for such screening methods.High-throughput robotic methods allow many thousands of assays to be completedin very short periods of time. As a result, the more faithful the screening method isin reproducing the ‘real life’ conditions and substrate, the more likely it is that theprocess will lead to provision of successful and cost-effective research.

Of further interest are the saccharides of various degrees of polymerizationreleased in the process of cell wall carbohydrate hydrolysis, i.e. the products ofenzymatic activity. Recent work discussed in this book is beginning to show that theresponse to addition of cell wall hydrolysing enzymes is due in large part to changesin the intestinal microfloral populations that accompany their use. The feeding ofa xylanase selects for ileal populations of bacteria that can utilize xylose and





concomitantly alters caecal populations dramatically (e.g. Apajalahti and Bedford,1999). Such shifts can be seen as a response to the addition of the enzyme andperhaps not a route through which they act on animal performance. Work withgerm-free animals, however, tends to suggest that the animal performance responsesobserved on addition of enzymes may in fact be due to microfloral responses to thealteration in the availability of a number of substrates. Some substrates, such as starchand protein, are more rapidly digested in the presence of an effective enzyme and as aresult bacteria that rely on such substrates are less prevalent. Others, such as cell walloligo- and polysaccharides, increase in concentration and as a result populations thatcan use such substrates thrive. The areas for future investigation must thereforeinclude a detailed understanding of the microflora of the gut. Desirable andundesirable species need to be identified and their nutritional requirementsdetermined so that enzymes can be designed to supply the correct balance ofoligosaccharides and peptides for beneficial organisms, whilst at the same time denythe pathogens their nutrition. Clearly such work is in its infancy but is likely to yieldsignificant benefits.

Raw material pretreatment

Whilst partial hydrolysis of cell wall NSPs reduces intestinal viscosity and mayaccelerate digestion through permeation of endosperm cell walls, it is also likely thatin many cases there is some degree of complete hydrolysis to monosaccharides.Glucose from cellulose and galactose from the α-galactosides can be readily utilizedby most animals, whereas the pentoses cannot be readily utilized by all classes oflivestock, especially poultry (Schutte et al., 1991, 1992). Most poultry, and evenswine to some degree, however, do not have a long enough digesta residence time toenable complete cell wall hydrolysis. The development of raw material pretreatmentprocedures for the complete hydrolysis of the hemicellulose fraction cereals, andeven less digestible feedstuffs such as straw, would be of great economic importance,as it could provide a basis for using products that are present in vast quantitiesbut cannot be utilized to any appreciable extent by monogastric animals. Thepretreatment of high-phytate ingredients such as rapeseed meal and cottonseed mealwould allow a far greater degree of phytate hydrolysis than is currently achievablein vivo. Pretreatment would yield the added benefit that the content of anti-nutrientscan be monitored throughout the process and controlled so that there is consistencyfrom batch to batch. Currently there is variation in efficacy of enzymes added tocomplete feed, due to variations in dosage rate, survival through processing andintestinal tract conditions. Control of enzyme addition rates, pH, temperatureand moisture during processing would likely increase uniformity of product andhence of animal nutrition. Such procedures, in order to be economic, would need tobe at minimal cost, which inevitably means low moisture and temperature. Suchrequirements set the challenge for enzyme discovery projects.

Future Horizons 393




Targeted enzymes

Knowledge of the exact conformation of the substrate to be degraded may conferadditional benefits to future enzyme discovery programmes. Often, the target may bea protein, and as a result enzymes are selected on their ability to degrade theanti-nutrient. Blanket administration of a such an enzyme to a complete diet has twoproblems. The first is that the target protein substrate may represent only a few percent of the total protein in the diet. In order to be effective, large amounts of proteaseneed to be added in order to provide enough activity to seek out the desired targetand render it ineffective. In such a scenario, much of the total protein in the diet isdegraded to a lesser or greater extent compared with the target, the extent dependingupon its relative susceptibility to the enzyme. Whilst this may be of benefit inspeeding up digestion of protein in general, it may also produce bitter peptides inthe process, which are thought to decrease diet palatability. As a result, it wouldbe desirable if the added enzyme hydrolysed only the targeted protein and left otherproteins intact. Not only would the enzyme be more specific, but in all likelihood farless would be needed. The development of such enzymes, sometimes referred to ascatalytic antibodies, is a key area of research. Once the identity of the substrate is wellcharacterized, this technology will lead to the production of antibodies withenzyme-like properties designed to meet the requirements of specific feedstuffs andclasses of livestock and poultry (Lerner et al., 1992; Mayforth, 1993). Modificationsof these procedures will undoubtedly develop. Alternatively, antibodies could belinked to specific proteases to concentrate the enzyme on its target substrate.

Alternative sources of enzymes

Enzymes are not only produced by fungi that have been improved using traditionalmethods but also expressed in microorganisms, such as bacteria, and in plants,such as in canola seed. Production in plants is seen as a far less costly method ofmanufacture. An abundant supply of enzymes in canola seed, for example, woulddramatically reduce the cost of enzyme production and hence costs to the livestockproducer. Production of phytase in canola or xylanase in rye, if successful, would alsoresult in a higher value feedstuff. Whilst enzyme selection and discovery are still intheir infancy, however, it is likely that production techniques need to be able to reactrapidly to new discoveries. At present it is quicker to bring a new product to marketthrough fermentation methods than it is through production in plants. Once therate of improvement in next-generation enzymes slows down, and the base enzymeactivities effectively become commodities, then considerations of costs of productionwill likely drive enzyme production through plant ‘biofactories’. Whilst emphasishere is on reducing costs and economics, it must be born in mind that by reducingcosts the scope for enzyme usage increases dramatically. Ten years ago, xylanases weretoo expensive to consider as viable additives in pig feed. Current prices have allowedrapid expansion into this sector in recent years. Use of enzymes in ruminant feeds ismostly constrained by the cost : efficacy ratio, a ratio that will improve over the next





few years and as a result lead to greater enzyme use in this, as yet, untapped area ofanimal feed.

An alternative strategy for the hydrolysis of anti-nutritive compounds in animalfeeds is to construct transgenic monogastric animals able to digest cellulose,β-glucans, xylans or phytic acid. Forsberg et al. (1993) concluded that it should befeasible to use bacterial DNA constructs to express and secrete glycanases in ratpancreatic acinar cell lines. The major challenge will be to obtain sufficiently highlevels of the expression of glucanase genes and other genes in pancreatic cells to effectadequate hydrolysis of the glucans, xylans, etc., in the intestine. Again, such a strategywould require that the enzymes of interest were commodities and not likely tobe improved upon by genetic manipulation. Only under these circumstances wouldthe gene inserted into the animal be competitive with the bacterial, fungal, orplant-derived products freely available on the market at the time of introduction ofthe animal to the marketplace. The animal breeders would have to be certain of sucha state of affairs, since the multiplication time required from germ-line breedingstock to food-producing animals can take as long as 5 years, which is more thansufficient time for development of superior products through fermentation methods.

Enzyme dosage and synergies

Most enzymes present on the market are simple xylanases, β-glucanases or phytases.It is fair to say that whilst there are some data describing the dose–responsecharacteristics of many of these products, this relationship is for the most part poorlydescribed, particularly when compared with the data available for several essentialamino acids, for example. As a result the current usage of such products is far fromoptimized. Dose-titration studies and linkage of the enzyme dose to the substratequantity present in any complete diet will certainly help to improve on theeconomics of current enzyme usage. Zhang et al. (1996) recently showed that theycould predict the response of poultry to a particular feed enzyme by using a simplelinear model. The model predicts that the response to enzyme supplementation is afunction of enzyme concentration. When this is converted to a logarithmic value, adoubling or halving of the response to enzyme treatment can be achieved only byvarying the amount of enzyme by a factor of 10, not 2 as may be expected. Clearly, asthe costs of feed and enzyme change in proportion to one another, the economicallyoptimum dose of enzyme will change also. The ability to match feed and enzymecosts with substrate concentration and thus predict optimum enzyme dosage viacomputer models is currently available in wheat- and barley-based rations. The intro-duction of near-infrared techniques for prediction of substrate concentrations, henceallowing for an ‘on-line’ enzyme dosage optimization at the level of the feed mill, is aclear goal for optimizing the economy of enzyme usage. The advantages of titrationof the correct amount of enzyme are realized in terms of not only more cost-efficientenzyme usage, but also greater uniformity of feed and hence animal production.

Nearly all the research done on the response of poultry to enzymes (with theexception of phytase) has used crude fungal extracts with high levels of the desired

Future Horizons 395




activities, such as of β-glucanase or xylanase, but also with considerable amounts ofother side activities or contaminating enzymes, including proteases. Research carriedout using different combinations of pure enzymes (that is, devoid of contaminatingand potentially synergistic enzyme activities) has and will allow determination ofwhether the principal enzymes have synergistic, antagonistic or additive effects. Forexample, the ability of enzymes to reduce the viscosity of the water-solublearabinoxylans in wheat or rye may depend on the amount not only of endo-xylanasebut also of arabinofuranosidase, and possibly β-glucanase, acetylxylan esterase andferuloyl esterase, in the preparation (Forsberg et al., 1993). Recent work suggests thatthere may be synergies between phytases and xylanases (Ravindran et al., 1999).Detailed descriptions of substrates will allow for more focused research into enzymemixtures based on likely synergistic interactions.

Studies with other animals

Although studies with chickens have clearly established the benefits of enzymes asfeed additives, only a limited number of studies have been carried out with otherspecies of poultry, such as turkeys, ducks, geese and ostriches. Also, very few studieshave been carried out using fish, eels, alligators, turtles, other exotic animals, pets(such as dogs and cats) and fur-bearing animals that have simple stomachs. Enzymesmay be particularly beneficial in the diets of animals that tend to be carnivorous,as they also tend to have GI tracts with smaller large intestines and therefore donot house a large population of anaerobic microorganisms capable of hydrolysingcomplex carbohydrates.

Conclusions

In the past several years, the use of enzymes as supplements in feeds has expandeddramatically. The research that led to this was mostly carried out at universities andresearch institutions and funded in large part by the industry, in cooperation withgovernment agencies. Although dramatic progress has been made in the past decadeon the use of enzymes in the diets of poultry and livestock, additional research will beneeded in many areas to exploit the full potential of this very powerful and beneficialtechnology. Areas of research and development should include the following:

• Development of more sensitive, accurate and (most importantly) biologicallymeaningful assays.

• Precise identification of the catalytic properties of enzymes most suited todifferent classes of livestock and poultry through characterization of thesubstrates and environments in which they reside.

• A better understanding of the effects that enzymes have on the physiological,microbiological and endocrine responses of animals fed cereal-based diets.





• Development of methods to predict substrate concentrations and use of modelsto predict responses to enzyme treatment and hence optimize economy andefficacy of usage.

It is clear from the above that, in order to be successful, research into this field needsto move from what has been a relatively simple approach to one that is moremultidisciplinary.

Whilst the roadmap for development of new products from a research viewpointseems clear, it is by no means clear when viewing the current climate of animal feedproduction. Bovine spongiform encephalopathy, dioxins in fat sources, sewerage inanimal feed and, most importantly, the use of antibiotics in animal production havein recent times increased consumer, regulatory and media interest in food productionprocesses. Genetically modified organisms are currently suffering acute attentionand rejection in some parts of the world. Such attention may well increase the timerequired to introduce new products into some parts of the world, which serves as areminder that the identification, improvement and production of these products isonly one step in the process of their introduction into the marketplace.

References

Annison, G. and Choct, M. (1991) Antinutritive activities of cereal non-starch polysaccharidesin broiler diets and strategies for minimizing their effects. World’s Poultry Science Journal47, 232–242.

Apajalahti, J. and Bedford, M.R. (1999) Improve bird performance by feeding its microflora.World Poultry 15, 20–23.

Bedford, M.R. (1993) Matching enzymes to application. Feed Management 44, 14–18.Bedford, M.R. (1995) Mechanism of action and potential environmental benefits from the use

of feed enzymes. Animal Feed Science and Technology 53, 145–155.Chesson, A. (1993) Feed enzymes. Animal Feed Science and Technology 45, 65–79.Cowan, W.D. and Rasmussen, P.B. (1993) Thermostability of microbial enzymes in expander

and pelleting processes and application systems for post-pelleting addition. In: Wenk, C.and Boessinger, M. (eds) Enzymes in Animal Nutrition. Kartause Ittingen, Thurgau,Switzerland, pp. 263–268.

Dänicke, S., Simon, O., Jeroch H. and Bedford, M.R. (1995) Effect of fat source and xylanasesupplementation on the performance and intestinal viscosity of rye-fed birds. In: vanHartingsveldt, W., Hessing, M., van der Lugt, P. and Somers, W.A.C. (eds) Proceedingsof the Second European Symposium on Feed Enzymes, 23–27 October, Noordwijkerhout, TheNetherlands. TNO Nutrition and Food Research Institute, Zeist, The Netherlands,pp. 102–107.

Forsberg, C.A.W., Cheng, K.-J., Knell, J. and Phillips, J.P. (1993) Establishment of rumenmicrobial gene pools and their manipulation to benefit fibre digestion by domesticanimals. Proceedings, 7th World Conference on Animal Production, Edmonton, AB,Canada. University of Alberta Press, Edmonton, Alberta, Canada, vol. 1, pp. 281–316.

Hendon, D.R. and Walsh, G.A. (1993) Activity analysis of enzymes under field conditions.In: Wenk, C. and Boessinger, M. (eds) Enzymes in Animal Nutrition. Kartause Ittingen,Thurgau, Switzerland, pp. 233–240.

Future Horizons 397




Hesselmann, K. and Aman, P. (1986) The effect of β-glucanase on the utilization of starch andnitrogen by broiler chickens fed on low- or high-viscosity barley. Animal Feed Science andTechnology 15, 83–93.

Hutchison, C.A. III, Philips, S., Edgell, M.H., Gillam, S., Jahnke, P. and Smith, M. (1978)Mutagenesis at a specific position in a DNA sequence. Journal of Biological Chemistry253, 6551–6560.

Jeroch, H., Dänicke S. and Brufau, J. (1995) The influence of enzyme preparations on thenutritional value of cereals for poultry: a review. Journal of Animal Feed Science 4,263–285.

Lerner, R.A., Kang, A.S., Bain, J.D., Burton, D.R. and Barbas, C.F. III (1992) Antibodieswithout immunization. Science 258, 1313–1314.

Mayforth, R.D. (1993) Designing Antibodies. Academic Press, San Diego, California, 207 pp.Ravindran, V., Selle, P.H. and Bryden, W.L. (1999) Effects of phytase supplementation,

individually and in combination with glycanase, on the nutritive value of wheat andbarley. Poultry Science 78, 1588–1595.

Schutte, J.B., Van Leeuwen, P. and Lichtendonk, W.J. (1991) Real digestibility and urinaryexcretion of D-xylose and L-arabinose in ileostomised adult roosters. Poultry Science 70,884–891.

Schutte, J.B., de Jong, J. and Van Weerden, E.J. (1992) Nutritional implications ofL-arabinose in pigs. British Journal of Nutrition 68, 195–207.

Schutte, J.B., de Jong, J. and Langhout, D.J. (1995) Effect of endo-xylanase preparation inwheat-based rations for broilers. Feed Compounder 15, 20–24.

Vahjen, W., Glaser, K., Froeck, M. and Simon, O. (1997) Non starch polysaccharidehydrolyzing enzymes as feed additives – detection of enzyme activities and problemsencountered with quantitative determination in complex samples. Archives AnimalNutrition 50, 331–345.

Ward, P.P. and Conneely, O.M. (1993) Using biotechnology to improve enzyme yields: fromDNA to the market place. In: Wenk, C. and Boessinger, M. (eds) Enzymes in AnimalNutrition. Kartause Ittingen, Thurgau, Switzerland, pp. 17–21.

Zhang, Z., Marquardt, R.R., Wang, G., Guenter, W., Crow, G.H., Han, Z. and Bedford,M.R. (1996) A simple model for predicting the response of chicks to dietary enzymesupplementation. Journal of Animal Science 74, 394–402.





IndexIndexIndexAcetivibrio 14Alpha-galactosidase

assay methods 97Amylase

assay methods 95–97engineering 329–330pH and thermal stability 329–330three-dimensional structure 329–330

Amylopectin 109Amylose 109Antibiotic growth promoters

interaction with enzymes 192Antibiotics 9, 301, 307, 308Antigenicity

effects of enzymes on, in soybean132–133

Arabinan 12Arabinofuranosidase 14Arabinoxylans 5, 13

effects in poultry 147–148properties 91see also Xylans

Aspergillus 14Assay methods for enzymes

agar plate diffusion 88Avicel 15, 23coloured substrates 18Congo Red 18DNS 18ELSA (enzyme-linked sorbent) 95general 16, 17

insoluble, dye-labelled substrates89–91, 94–95

interfering factors 91, 92–93, 94RBB xylan 18reducing-sugar methods 87, 91–92,

100–101soluble, dye-labelled substrates 88–89,

92–93Somogyi–Nelson 18viscometric methods 87–88, 91–92,

101–103see also individual enzymes, e.g.

β-glucanase; Xylanase

Bacterial metabolites 300Bacterial proliferation

effects of enzymes 162–163, 185–186Barley 5, 7

amylose : amylopectin ratio 167effects of enzymes in pigs 166–169,

170–175effects of enzymes in poultry 145–146energy variation in pigs 166–167prediction of energy value for pigs 169

Beef cattleresponse to enzymes 277

Bile salts 203deconjugation 203effects of NSPs on 152

Bioavailability of phosphorus 239

399

A3882:Bedford:AMA:First Proof: 04/09/00 CHAP-Index

Index



Calcium to phosphorus ratio 70Caloric efficiency

influence of enzyme addition 226Catalytic mechanisms 31–33, 34–35Cell wall

degradation 282, 285encapsulation of nutrients-effects of

enzymes on 151Cellobiohydrolase 19

assay 15mode of action 22–23stereospecificity 28–29

Cellulase 11, 316–317adsorption 38applications 45architecture 39assay 15catalytic domain 41fungal vs. bacterial sources, properties

86mode of action 22–24physicochemical properties 37regulation 33, 36synergy with CBH 28three-dimensional structure 41, 42, 43variety of forms 37–38

Cellulose 11, 12, 14structure 12

Cellulose binding domain 24, 38, 39, 40Cellulosomes 24, 285

effect on bacterial adhesion 285Cereal

growing conditions 5variety 5

Chelatorsvalue in phytic acid rich diets 65

Citric acid 73, 257Classical mutagenesis 331–332Clostridium 14Coccidiostats 307Copper 307Corn see MaizeCottonseed meal-effects of enzymes on 136

Dairy cattleinfluence of enzyme activity 278–279influence of method of enzyme

application 278

response to enzymes 278Diet digestibility

influence on intestinal microflora 302Dietary fat

influence on gut morphology 211Digestion

efficiency 4simulation methods 275variability 4

Directed evolution 319, 326, 330–331Disease

role of non-starch polysaccharides andenzymes

colitis in pigs 191–192swine dysentery 190–191

DNAexpression and overexpression 333ligase 333promotors 333restriction endonuclease 333transformation 333

Dockerin 24

Emulsification of fat 200, 209Endocrine system

effect of enzymes on 151Endogenous losses

effects of NSPs and enzymes 150–152Endoglucanase 19

stereospecificity 28–29Energy

availability from cereal 110Enzyme

activityduring feed processing 380effect of substrate concentration

275–276of exogenous in small intestine 289influence of amino acid changes 326

adsorption 38assays 390augmentation of ruminal activity 284characteristics of value to feed industry

391–392classes 316concentration from broth 345crystallization 346dosage 395

400 Index




dose–response 279expression in animals 395expression in plants 291, 394factors influencing recovery from

fermentation broth 320, 340families 34–35and fat

influence on broiler performanceand digestion 214–217, 227,228

influence on rate of passage inruminants 287

influence on substrate availability tomicroflora 302

inhibition 21interaction with antibiotics 308kinetics 20liquid formulation 346–347market size 3, 4, 6, 8mediated provision of substrate for

microflora 303–304methods for targeting to substrate 394mode of action 392multiple forms 317non catalytic effects in ruminant 286,

287post-ruminal effects 288pre-processing of raw materials 393processing from fermentation broth

343–344products 25

and effects 392–393purification 345–346response

factors involved 307influence of dietary ingredients 277

screening 321–325, 332enrichment techniques 321factors limiting discovery 318molecular techniques 324, 325,

330–331limitations 325

organism identification bysequencing techniques 323, 324

redundancy 322selective environments/criteria 321

site of action 391solid formulation 347sources 274

elimination of need to culture sourceorganism 318

substrate ratio 275survival in rumen 283synergy 28–31, 395

exogenous with endogenous 284three-dimensional structure effect on

activity 327–331uses in industry 315

Enzymes (exogenous)survival in the gut 130

Esterase 285Exogenous enzyme

influence on endogenous enzymes 284as proportion of endogenous 284

Exoglucanase 19see also Cellobiohydrolase

Expression vector 334Extrusion

effect on enzyme activity 380

Faba beanseffects of proteases on 134

Faecal moistureeffects of enzymes 149–150, 192

Faeces decomposition by enzymes 289Fat digestibility

effects of NSPs on 152influence of barley, fat source and

β-glucanase 221influence on fat soluble vitamin storage

229influence on pigmentation 229influence of rye, fat source and xylanase

221, 222, 223influence of wheat, fat source and

xylanase 222Fat source

effect on intestinal protein turnover227

effect on protein and amino aciddigestibility 225

fatty acid composition 202interaction with cereal 199

Fatty acid binding protein 200Fatty acid digestibility

influence of cereal, fat source andenzyme 221–224

Index 401




Fatty acidsdegree of saturation, effects on

digestibility 201site of absorption 224

Feed passage rateeffects of enzymes and fat type 207,

208influence of viscosity 206

Feed processing 377, 383effect on intestinal viscosity 384effect on nutrient availability 384implications for enzymes 353–354

Feed production 1, 2, 3Fermentation

bacteriophage control 342–343common substrates 341control 337, 338, 339inhibitors 341–342maximizing yield 337, 338, 341preparation of fermentor 342–343process 320, 337–340scale-up 340–341separation of cells from enzymes

344–345types 338

Fibre digestionpH effect 286, 287role of surfactants 287

Fibrobacter 14Fibrolytic activity post rumen 288Filtration 344–345Fines in pellet production

implications for liquid enzymeapplication 356

Formic acid 257Fractional precipitation 345–346Fructo-oligosaccharides 307Fungi 14Fusarium 14

Galactan 12Gastric digestion

effects on enzyme function 162Gene deletion 334–337Germ free chickens 301β-Glucan 5

properties 85, 145–146

β-Glucanaseassay methods 85–91correlation of in feed activity with in

vivo performance 381effects of coating on

thermostability 381microgranulate thermostability 380recovery in feed 381thermostability 380

Glucomannan 12β-Glucosidase 19

assay 15α-Glucuronidase 14Grain

by-products (rice and wheat)effects of enzymes in pigs 188–190

extract viscosityas predictor of enzyme response in

poultry 154Granulation 347Grinding of cereals 120Grist size 120Gut

effects of NSPs on the mucosa 150microflora

effects of NSPs on 151–152transit time

effects of enzymes 152, 163–165

Hammer milling see Grinding of cerealsHemicellulase 11Hemicellulose 11, 12Hind gut fermentation 119Humicola 14

Immune system 307Immunochemical techniques for protease

detection 130In-feed activity of enzymes 280Intestinal

damagemicrobial involvement 211

microbial populationsinfluence of enzyme and fat type 213

microfloraadaptation to diet 305

402 Index




changes with age 301effect on bile acids and fat digestion

305effects on gut physiology 306interactions between species 300numbers 299response to viscosity 302

phytase 241substrates 300viscosity

ruminants 288see also viscosity

Intestinesenvironment 299–300

Lactic acid 257Lambs 279Lectins 5

protease effects on 131Lichenase function 86Lipase in turkeys 204Liquid enzymes 9

application systems 355–363coefficient of variation in different

application systems 367–371desired droplet size at spraying

370–371dilution and application rate 371–372dosing pumps 365flow meters 364handling 374importance of dosing accuracy 375pre- or post fat addition 372–373stability in storage 373

Liquid feeding of pigsapplication of enzymes 138

Lupinseffects of enzymes on 136

Maize 6amino acids 116amylose : amylopectin ratio 187breeding 113, 122bushel weight 111, 112dent 109digestibility 118effects of enzymes in pigs 185–190

effect of fertilizers on nutritive quality121, 122

flint 109floury 116, 117genetic modification 122germ 113high oil varieties 113, 114, 115, 116high protein varieties 116, 117ileal digestible energy 118immunity 122low phytic acid varieties 120normal 109nutrient composition 111, 112nutritive value 114, 115, 116oil 113opaque 116, 117protein 112, 116

effects on starch digestion inruminant 290

true metabolizable energy 118use of enzymes 119waxy 109, 118

Manure 5, 6Marker genes 333Meat consumption 2Micelles 201Mineral soaps 204Mode of action of enzymes

ruminants 280–281Mortality 119

influence of fat source and enzymeaddition 228

Neocallimastix 14NIRS (near-infrared spectroscopy)

prediction of raw material feedingvalue 169

Oligosaccharides 307

Particle size 120Penicillium 14Phenyl esterase 14Phytase

animal sources 67Aspergillus, structure 69

Index 403




Phytase continuedassay methods 97–98brush border see Phytase, intestinalcompetitive chelation 65, 73, 76dose–response relationship with P

availability 246effect of Ca on activity 75effect on

Ca availability 251Ca and P digestibilty 76energy metabolism 255phosphorus availability 244–245phosphorus pollution 75, 246–248protein and amino acid availability

252–255protein and amino acid digestibilty

75zinc availability 252

factors influencing response 250–251,254, 256

in fish diets 259in genetically modified plants 74,

242–243influence of

age on response 258Ca to P ratio on activity 256feed matrix on thermostability 379

interaction withEDTA see Phytase, competitive

chelationorganic acids 257phosphatases 67vitamin D 256

intestinal 67, 70, 71bacteria 71

in laying hen rations 250microbial 68

sources 242optimum

conditions 71temperature 378

pH optima 241, 259phosphorus equivalence values

246–249plant 68, 70, 71

sources 241pretreatment of plant meals 72–73safety 243site of action in animal 69–71

site of activity in animal 243sources 67, 68temperature sensitivity 378, 379thermostability 69

of plant sources 379through coating 379

units of activity 241units equivalence for Ca and P

replacement 76variability in response 76

Phytic acidbinding of minerals 239complexes with metal ions 62, 63, 70content in feed ingredients 63, 238digestion and utilization 66effect of calcium on digestibility 70effect on

digestive enzyme activity 66mineral utilization 64–65phosphorus pollution 64protein digestibility 65–66

germination effects on content 72hydrolysis pattern 67location in ingredients 238pKa values 62selection for low phytate varieties 72solubility of complexes 63structure 62, 238, 240

Phytin 63location in feed ingredients 63–64

Pig vs. poultry-comparative responses toenzymes 129–130, 161–165

Piromyces 14Production organisms 319Protease 5

assay methods 98–100bacterial vs. fungal, efficacy 131effects on

animal performance 131, 133soybean 130–134

Protein engineering 318–319, 326–331

Random mutagenesis 319Rapeseed

antinutrients 127–128effects of enzymes on 135

Regulatory authorities 347–348Rennet 3

404 Index




Rice bran see Grain, by-productsRuminal effects of enzymes 282Ruminal infusion 280Ruminococcus 14Rye

effects of enzymes in pigs 176–185effects of enzymes in poultry 146

Scaffoldin 24Soaking of cereals

effect on P availability 258influence on phytase response 258

Sorghum-effects of enzymes in pigs186–190

Soybean mealantinutrients and their elimination

130effects of proteases 130–134, 138

Sporotrichum 14Stabilization of enzymes 320Starch 109

granule 110resistant 110

Substrateavailability

influence on microflora 301identification of relevant target for new

enzymes 392influence of enzyme activity 290

Substrates 16, 17Subtilisin

effects of amino acid substitution onactivity 328

three-dimensional structure 327–328turnover number 328

Talaromyces 14Thermomyces 14Thermostabilizing domains 39Thermostability 9Thermostable phytase 378–379Transferase reactions 21Trichoderma 14Triglycerides

fatty acid positional effects 203–204Triticale

effects of enzymes in pigs 176–185

energy variation in pigs 176Trypsin Inhibitor 5

protease effects on 131

Uniformitybody weights 119

Unstirred water layer 208

Vegetable protein mealsantinutrients and their elimination

125–128, 134effects of enzymes on animal

performance 135–136see also individual raw materials, e.g.

Soybean meal, RapeseedViscosity 5, 204–205

changing effect on fat digestibility withage 220

effect of fat type, enzyme and ryeconcentration 218

effect onbile salt metabolism 210fat digestibility 219–221, 224lipase activity 209–210pancreatic lipase secretion 209

individual bird variability 219influence of

concentration of solute 205molecular weight of solute 205

influence ongut morphology 211maintenance energy 225–226protein and amino acid digestibility

224, 225relevance for microflora 301

Viscosity in the gutphysiological effects in poultry and pigs

148–151, 164pig vs. poultry

implications for enzymes 163–164as predictor of enzyme response in

poultry 153–154Vitamin D

influence on phytic acid digestibility74

Voluntary food intake-effects of enzymes164–165, 177

Index 405




Water holding capacity of the feed-effects ofenzymes 165–166, 176

Wet feeding systemsopportunity for phytase 259

Wheat 5effects of enzymes in pigs 169–185energy variation

in pigs 169in poultry 146–147, 154–155

starch digestibility 154–155varietal effects on feeding value in pigs

169, 176Wheat bran see Grain, by-productsWhite kidney bean (Phaseolus vulgaris)

effects of enzymes on 135

Xylanase 12, 13, 14, 19antisynergy 31applications 45assay methods 91–95, 103–105catalytic domain 41correlation of in feed activity with

in vivo performance 382–383effects of coating 382, 383

family 10 25–27family 11 25–27and fat type

influence on protein metabolism 212influence of feed matrix on

thermostability 382influence on

endogenous N losses 227lipase activity 210

mode of action 22–25physicochemical properties 37post-pelleting application 382recovery in feed 381regulation 33, 36stabilization 382synergy with other activities 30–31thermostability 381, 382three-dimensional structure 41, 43, 44variety of forms 37–38

Xylans 13Xylo-oligomers 304β-xylosidase 14

Zinc 307

406 Index




Documents

ENZYMES IN FARM ANIMAL NUTRITION...ENZYMES IN FARM ANIMAL NUTRITION 1 Z:\Customer\CABI\A3882 - Bedford + Partridge - Enzymes\A3957 - Bedford - Enzymes #P.vp 17 November 2000 14:57:14