Reviews

Bacterial Polyesters: Biosynthesis, Biodegradable Plastics andBiotechnology

Robert W. Lenz*,† and Robert H. Marchessault‡

Polymer Science & Engineering Department, University of Massachusetts,Amherst, Massachusetts 01003-4530, and Department of Chemistry, McGill University,

3420 University St., Montreal, QC, H3A 2A7 Canada

Received May 21, 2004; Revised Manuscript Received September 23, 2004

The discovery and chemical identification, in the 1920s, of the aliphatic polyester: poly(3-hydroxybutyrate),PHB, as a granular component in bacterial cells proceeded without any of the controversies which markedthe recognition of macromolecules by Staudinger. Some thirty years after its discovery, PHB was recognizedas the prototypical biodegradable thermoplastic to solve the waste disposal challenge. The developmenteffort led by Imperial Chemical Industries Ltd., encouraged interdisciplinary research from genetic engineeringand biotechnology to the study of enzymes involved in biosynthesis and biodegradation. From the simplePHB homopolyester discovered by Maurice Lemoigne in the mid-twenties, a family of over 100 differentaliphatic polyesters of the same general structure has been discovered. Depending on bacterial species andsubstrates, these high molecular weight stereoregular polyesters have emerged as a new family of naturalpolymers ranking with nucleic acids, polyamides, polyisoprenoids, polyphenols, polyphosphates, andpolysaccharides. In this historical review, the chemical, biochemical and microbial highlights are linked topersonalities and locations involved with the events covering a discovery timespan of 75 years.

In 1982, Imperial Chemical Industries Ltd. (ICI) inEngland announced a product development program on anew type of thermoplastic polyester which was totallybiodegradable and could be melt processed into a widevariety of consumer products including plastics, films, andfibers.1 The polymer was to be manufactured by a large-scale fermentation process not unlike the brewing of beerbut which, in this case, involved the production of thepolymer inside the cells of bacteria grown in high densitiesand containing as much as 90% of their dry weight aspolymer. The bacterium capable of performing this feat wasAlcaligenes eutrophus, since renamedRalstonia eutropha(more recently changed again toWautersia eutropha) and

the commercial polyester product, tradenamed “Biopol”,was a copolyester containing randomly arranged units of[R]-3-hydroxybutyrate, HB, and [R]-3-hydroxyvalerate,HV:2

Discovery of Bacterial Polyesters

That bacteria could produce polyesters was unknown topolymer chemists before 1960 and even to most biochemistsand microbiologists before 1958, although their presence inbacterial cells in isolable amounts, their chemical composi-tion, and even the fact that they were polymers, were reported

* To whom correspondence should be addressed. Tel.: 1-413-545-3060.Fax: 1-413-545-0082. E-mail: rwlenz@polysci.umass.edu.

† University of Massachusetts.‡ McGill University.

© Copyright 2005 by the American Chemical Society

January/February 2005 Published by the American Chemical Society Volume 6, Number 1

10.1021/bm049700c CCC: $30.25 © 2005 American Chemical SocietyPublished on Web 11/23/2004

in the literature as early as 1926. These natural polyestersremained unknown to a wider scientific community for solong because their discoverer, Maurice Lemoigne, publishedhis results in little-read French journals at a time whenmicrobiologists had no interest in lipids, as they were referredto by Lemoigne, that were not ether soluble, and manyorganic chemists refused to believe that there were suchthings as polymers. Lemoigne (Figure 1) was a bacteriologistwith training in analytical chemistry, and his series ofpapers published over the five-year period from 1923 to1927 are remarkable for their breadth of research andprescience.3-8

The polyester that Lemoigne isolated and characterizedwas poly-3-hydroxybutyrate, PHB, shown below. PHB is thereserve polymer found in many types of bacteria, which cangrow in a wide variety of natural environments and who have

the ability to produce and polymerize the monomer, [R]-3-hydroxybutyric acid:

As indicated in this structure, the repeating unit of PHBhas a chiral center, and Lemoigne reported that the polymeris optically active.4 In fact, PHB is only the parent memberof a family of natural polyesters having the same three-carbonbackbone structure but differing in the type of alkyl groupat the â or 3 position. These polymers are referred to ingeneral as polyhydroxyalkanoates, PHAs, and all such naturalpolyesters have the same configuration for the chiral center

Figure 1. Photograph of Maurice Lemoigne, Head of Services de Fermentation at Institut Pasteur, Paris 1949. Courtesy of Institut PasteurArchives, with permission.

2 Biomacromolecules, Vol. 6, No. 1, 2005 Lenz and Marchessault

at the 3 position, which is very important both for theirphysical properties and for the activities of the enzymesinvolved in their biosynthesis and biodegradation.1,9

At the time of his discovery of PHB in bacteria, Lemoignewas the Director of the Fermentation Laboratory of thePasteur Institute in Lille, France. He became involved withPHB in an attempt to determine the cause of the acidificationof aqueous suspensions of the bacteriumBacillus megateriumwhen it was kept under an oxygen-free atmosphere. In 1923,Lemoigne reported that the acid produced by the bacteriawas 3-hydroxybutyric acid,3 and in 1927, he described theisolation of a solid material obtained from the cell which hecharacterized as a polymer of 3-hydroxybutyric acid.8 Hecame to that conclusion by carefully hydrolyzing the solidinto a series of water-soluble oligomers of 3-hydroxybutyricacid, which he characterized for molecular weight andmelting point. He named the source of the acidlipide-â-hydroxybutyrique, and, remarkably for his time, he evensuggested that the polymer was produced within the cell bya “dehydration polymerization”.7

Lemoigne published these observations and interpretationsat the time when Herman Staudinger at the University ofFreiburg, Germany was being ridiculed by his colleagues inorganic chemistry in Europe for proposing the existence ofhigh molecular weight molecules or polymers, which hetermed “macromolecules”. Fortunately, Lemoigne was freeof such prejudices, and he was probably familiar with thework of Emil Fischer, who demonstrated as early as 1906that proteins are large molecules of “enchained” amino acidunits or “polypeptides”, a term he originated.10 EventuallyStaudinger’s concepts about synthetic polymers won out, butnot until the 1930s and the publication of the definitiveresearch of Wallace Carothers at duPont ExperimentalStation, Wilmington, Delaware, on the synthesis and char-acterization of aliphatic polyesters and polyamides. In 1953,Staudinger was awarded the Nobel Prize in Chemistry forhis work on polymer synthesis and for his staunch defenseof the concept of macromolecules.11 There is no indicationthat either Staudinger or Carothers were ever aware ofLemoigne’s discovery of nature’s polyesters, which remainedhidden from organic and polymer chemists for over 30 yearseven though PHB was described in biochemistry textbooks,where, however, it was referred to as a “lipid” not a polyester.

Lemoigne and co-workers reported on their PHB studiesin 27 publications from 1923 until 1951, and in their laterwork they found that the cells ofB. megateriumcould containas much as 44% of their dry weight of PHB depending ongrowth conditions.12,63 Lemoigne was the first to describean analytical method for quantifying PHB, and he showedthat PHB could be cast into a transparent film like the thenwell-known cellulose nitrate material, collodion.7 In follow-up studies, he and co-workers also reported that a variety ofbacteria could produce PHB, but apparently he never becameinvolved in determining the function of such polyesters incell metabolism even though he labeled it as a “reservematerial”. It was not until microbial physiologists recognized,in the late 1950s, the important role that PHB played in theoverall metabolism of bacterial cells that the significance ofLemoigne’s earlier discoveries was realized.

Rediscovery of PHB

The rediscovery of PHB occurred simultaneously and waspublished independently in 1957 and 1958 by microbiologistsin Great Britain and the United States. At the University ofEdinburgh in Scotland, Wilkinson and co-workers becameinterested in the relationship between the presence of theintracellular lipid granules in bacteria, which had been knownsince 1901, and the large amounts of PHB found in somespecies as reported by Lemoigne and co-workers, and thefunction of PHB in the cells.13 During the same time periodat the University of California in Berkeley, Stanier,Doudoroff and co-workers found that PHB was the primaryproduct of the oxidative and photosynthetic assimilation oforganic compounds by phototropic bacteria, and they at-tempted to detail the biosynthesis and breakdown mechanismof PHB in the cells.14

Well before these studies, Weibull in 1953 had isolatedthe granules ofB. megateriumby dissolution of the cell wallwith a lysozyme, and he confirmed the claim made byLemoigne in 1944 that PHB was the major constituent ofthe granules.15 A typical example of such granules inside acell is shown in Figure 2 for the bacteriumA. Chrococcum.In 1958, Wilkinson and co-workers obtained morphologicallyintact granules fromBacillus cereusby disrupting the cellswith alkaline hypochlorite solution and determined theamount of PHB in the granules, but this reagent, the well-known “eau de javelle” laundry bleach, was later shown byLundgren and co-workers at Syracuse University to degradethese polyesters and yield, only low molecular weightpolymers.16,17 Only cells that were chloroform extractedyielded high molecular weight PHB when reliable methodswere used later, starting in 1965.17a

In 1961, Doudoroff and Merrick isolated what theydescribed as “native” PHB granules of two chemohetero-tropic bacteria,Rhodospirillum rubrumandB. megaterium.“Native” granules are intact granules which are carefullyisolated from the cell and purified to retain the activesynthase.18 R. rubrum “native” granules also retain thedepolymerase enzyme that can degrade PHB to the monomer,but the B. megaterium“native” granules have only theassociated synthase. Doudoroff and co-workers also studiedthe enzyme-catalyzed hydrolysis of PHB extracted from thecell and free of all proteins by the extracellular depolymeraseswhich are excreted by a variety of bacteria that can use thepolymer as a carbon source as discussed below.19

Biosynthesis of PHB

Stanier and Wilkinson and their co-workers determinedthat the PHB granules in bacteria serve as an intracellularfood and energy reserve and that the polymer is producedby the cell in response to a nutrient limitation in theenvironment in order to prevent starvation if an essentialelement becomes unavailable. The nutrient limitation acti-vates a metabolic pathway, which shunts acetyl units fromthe tricarboxylic cycle into the production of PHB. The latteris ideal as a carbon-storage polymer because it is waterinsoluble, chemically and osmotically inert, and can bereadily reconverted to acetic acid by a series of enzymaticreactions inside the cell.20

Historical Review of Bacterial Polyesters Biomacromolecules, Vol. 6, No. 1, 2005 3

The reactions involved in the metabolic pathway respon-sible for the biosynthesis of PHB from acetic acid were firstidentified by Stanier and co-workers in 1959 in their studieson PHB formation inR. rubrum. However, the specificenzymes which catalyzed the reactions for the synthesis of3-hydroxybutyric acid, the monomer for PHB, were notidentified until 1973, when Schlegel at the University ofGottingen, Germany and Dawes at the University of Hull,England, working independently, were able to isolate andcharacterize those enzymes.21,22 Schlegel and co-workerscarried out their investigations on the metabolic cycle forPHB production inAlcaligenes eutrophuswhile Dawes andco-workers studied the cycle for PHB production inAzoto-bacter beijerinckii.

Schlegel first began his research onA. eutrophusin thelate 1950s as part of a study of the oxidation of molecularhydrogen by Hydrogenomonasbacteria. In 1961, theyobserved thatA. eutrophus, a member of this group, couldaccumulate very large amounts of PHB during growth innitrogen-limited media.23 Coincidentally, Dawes becameinterested in PHB while preparing a review on microbialmetabolism in 1962 before joining the University of Hull.He began his research program there with a study of theaccumulation of PHB inA. beijerinckii, which he foundcapable of accumulating as much as 70% of its dry weightof polymer.24 In the same 1973 issue ofBiochemical Journal,Schlegel and Dawes published simultaneously their discover-ies on the identification of the two enzymes involved in thereactions for converting acetic acid to 3-hydroxybutyric acidin the two different bacteria.21,22 For both bacteria, theenzymes were a ketothiolase (1), which catalyzes thedimerization of the Coenzyme A derivative of acetic acid,acetyl-CoA, to acetoacetyl-CoA, and a reductase (2), whichcatalyzes the hydrogenation of the latter to [R]-3-hydroxy-butyryl-CoA, the monomer that is polymerized to PHBby a synthase (3), as shown in the following reactionscheme:

As discussed above, this cycle becomes activated whenacetyl-CoA is restricted from entering the tricarboxylic acidcycle because of a deficiency in nutrients (generally eitherphosphorus, nitrogen or oxygen) needed by the cell to furthermetabolize acetyl-CoA for cell growth.

Although the first two enzymes, the ketothiolase (1) andthe reductase (2) which are responsible for monomersynthesis, were not fully identified and characterized untilthe investigations of Schlegel and Dawes in 1973, the enzymeresponsible for the polymerization process, the synthase orpolymerase (3), was initially recognized by Doudoroff,Merrick and co-workers as early as 1964, and it wascharacterized by Merrick and co-workers in 1968 in theirstudies on the production of PHB in bothR. rubrumandB.megaterium.25,26 They made their initial recognition of the

existence and role of the synthase in their work on theisolation and characterization of the active “native” PHBgranules from these bacteria. These granules could be usedin an aqueous suspension for the “in vitro” polymerizationof [R]-3-hydroxybutyryl-CoA to PHB. With their “native”PHB granules, Merrick and co-workers were also able tocarry out kinetic studies on the polymerization reaction todetermine the Michaelis-Menten constants for the reaction.They even proposed in their 1968 studies that the active siteof the synthase contains a cysteine unit which provides athiol group that covalently bonds to the growing polymerchain as a thioester.26

A detailed mechanism for the polymerization reaction,which was based on Merrick’s suggestion, was proposed byBallard and co-workers at ICI in 1987 and further elaboratedby Doi and co-workers at the RIKEN Institute in Japan in1992.27,28 They proposed a mechanism in which two thiolgroups are involved in the active site for both the initiationand propagation reactions of the polymerization. For initia-tion, the two thiol groups form thioesters with two moleculesof monomer, which then undergo a thioester-oxyesterinterchange reaction at the active site to form a dimer andrelease one of the thiol groups for the propagation reaction,as follows:

Propagation ensues by bonding another monomer to thefree thiol group of the active site followed by anotherthioester-oxyester exchange reaction to form the trimer, andso on. These reactions are thermodynamically favorablebecause of the higher bond strength of the oxyester comparedto the thioester. The synthase, therefore, both initiates andcatalyzes the polymerization process, which proceeds by acontinuous series of insertion reactions in much the samemanner as in the stereoregular polymerization of olefins byZiegler-Natta catalysts. In this case, the enzyme is specificfor monomers with the [R] configuration and will notpolymerize identical compounds having the [S] configurationas initially reported by Dawes and co-workers in 1989,29 soas a result, all natural PHAs are completely isotactic.

Biotechnology

As mentioned at the start of this review, bacterialpolyesters became an article of commerce when ICI begantheir production of “Biopol” in 1982, but “Biopol” was notPHB. PHB has a high melting point (180°C) and formshighly crystalline solids which crystallize slowly and formlarge spherulitic structures that impart poor mechanicalproperties in molded plastics and films, although, additionof nucleating agents and suitable posttreatment after extrusion

4 Biomacromolecules, Vol. 6, No. 1, 2005 Lenz and Marchessault

or casting can lead to much improved properties.30 Becauseof its high melting point, PHB is also susceptible to thermaldegradation during melt processing by ester pyrolysis of thealiphatic secondary esters of the repeating units. Thesedeficiencies were partly eliminated when it was found that,when A. eutrophusis grown on a mixture of glucose andpropionic acid, the storage polyester formed is a randomcopolyester of HB and HV units which has a lower meltingpoint.30 As a result, the copolymers have better processingcharacteristics and considerably improved mechanical prop-erties for use as plastics as shown in Figure 3. Nevertheless,like PHB, the copolymer is fully biodegradable in a widevariety of natural environments as well as in waste disposalfacilities, especially in municipal compost sites.

The ability of bacteria to produce storage polyesters withcompositions other than PHB was not realized until 1974when Wallen and Rohwedder at the USDA NorthernRegional Research Laboratory reported that a polyesterisolated from activated sludge contained both HB and HVunits, but they were not able to identify the microbial speciesin sludge which produced the polyester.31 In 1983, Whiteand co-workers at Florida State University demonstrated thatthe hydroxyalkanoic acid units present in polyesters extractedfrom bacteria in marine sediments included even more thanHB and HV units.32 They analyzed the ethyl esters of theunits, which were obtained by ethanolysis of the polyester,by gas chromatography and showed the presence of at least11 types of repeating units, including both linear andbranched 3-hydroxyalkanoic acids with compositions varyingfrom four to eight carbon atoms. White and co-workers alsoshowed that Lemoigne’s original bacteriumB. megateriumcould produce polyesters containing at least six differenttypes of units, although HB units still comprised ap-proximately 95% of the contents. In 1983, Witholt and co-workers at the University of Groningen, The Netherlands,found that Pseudomonas oleoVorans grown on alkanesproduced a large number of granules containing polyesterswith units of 6-10 carbon atoms.33 These bacterial polyestershave low glass transition temperatures and much lowercrystallinities than PHB, and as a result, they displayelastomeric properties. In 1988, Doi and co-workers obtainedPHAs with 4-hydroxybutyric acid repeat units from bacteriagrown on carbon substrates having these structures.34

ICI became involved in the commercial development ofbacterial polyesters after terminating a program on the large-scale production of single cell proteins, SCP, by bacteria foruse as fodder.35 After evaluating a variety of methods forthat purpose, they had concentrated on the use of methylo-tropic bacteria to produce SCPs from methanol, but theproject was terminated in 1976 because of consumerresistance. The company then turned instead to the possiblelarge-scale production of PHB by the same bacteria for theirentree into industrial biotechnology and bioprocessing.36 Inthis case, they were partially motivated by the majorpetroleum crisis of the 1970s, which made the productionof plastics from renewable resources economically attrac-tive.35

During the 1950s, Schlegel had also studied the productionof SCPs by bacteria, eventually selecting theHydrogenomas

bacteriumA. eutrophus for that purpose. In those investiga-tions, he and co-workers observed that this bacterium wascapable of producing very large amounts of PHB underselected growth conditions, and in 1979, Schlegel providedsamples of several strains ofA. eutrophusto ICI for theirPHB process development program.23 ICI selected one ofthese strains for intensive study after Holmes and co-workersin their research laboratory found that the bacterium couldproduce up to 80% of its dry weight of the HB/HVcopolymer when grown on a mixture of glucose andpropionic acid as mentioned above.35 Furthermore, thecomposition of the copolymer could be varied over a widerange by varying the composition of the feed mixture.

ICI was not the first company to consider the commercialdevelopment of bacterial polyesters for consumer products.Their efforts were preceded by a much earlier attempt at theW. R. Grace Company in Maryland, where Baptist andWerber initiated a similar program in 1960.37 Baptist hadjoined the Research Division of Grace in 1959 after apostdoctoral in biochemistry at the University of Michigan,where he had learned about bacterial production of PHB.Baptist and Werber recognized that PHB was a stereoregularpolymer with a melting point close to that of polypropylene,which suggested to them that PHB might be able to competewith polyolefins as a thermoplastic but with the addedadvantage of being biodegradable. Baptist used a sample ofRhizobiumobtained from Hayward and co-workers at theColonial Microbiological Research Institute in Trinidad, whoreported in 1958 that their strain of that bacterium couldproduce PHB to 58% of its dry cell weight.38 With thisbacterium Baptist was able to produce large quantities ofPHB for evaluation, initially for molded plastics and laterfor absorbable sutures. The latter subject was of interestbecause PHB, as a natural polyester, was assumed to be abiocompatible polymer in humans, which it has been foundto be in more recent studies, but it is very slowly resorbable.They improved the mechanical properties of PHB plasticswhen they found that PHB was compatible with a variety ofplasticizers, which greatly improved its processing andsolid-state properties. Nevertheless, the project was termi-nated in 1962 because of the poor thermal stability of PHB,and their work on the synthesis and properties of PHB wasreported in 1964 in theTransactionsof the Society ofPlastics Engineers.37 However,Chemistry and EngineeringNewsin the March 18, 1963, issue published an extensiveresearch report titled: “Bacteria Produce Polyester Thermo-plastic”, which was apparently the first time that mostpolymer chemists became aware of this thermoplasticbiopolyester, which occurred as inclusions in bacterial cells(Figure 2).

Coincidentally, in 1962, Marchessault and co-workersbegan a program at the State University of New York inSyracuse on characterization of the structure and propertiesof PHB both in the solid state and in solution.12,39 Samplesof the polymer were provided to them by Lundgren, whohad been studying the presence of PHB in bacteria for severalyears in the Microbiology Department of Syracuse Univer-sity. Issues such as obtaining high molecular weight poly-mers, optical rotation, and X-ray crystal structure were settled

Historical Review of Bacterial Polyesters Biomacromolecules, Vol. 6, No. 1, 2005 5

in definitive experiments. Also, at about the same time,Merrick joined Lundgren’s Department and continued hisstudies on the composition and activity of the “native” PHBgranules ofB. megaterium.26,40 They observed that thecarefully purified, or native, granules (cf. Figure 2) weresurrounded by a protein membrane. They also continued andexpanded the earlier studies of Merrick and Doudoroff onthe depolymerization of PHB by soluble depolymerases,which they obtained from a variety of bacteria that produceand release these enzymes and are capable of utilizing PHBas a sole carbon source.19,41,42 The protein membranescovering the PHA granules ofP. oleVoranswere studied indetail by Fuller and co-workers at the University of Mas-sachusetts, Amherst. In 1995, they reported that the granulesare enclosed in two separate protein membranes, and thesynthase is associated with the inner membrane.42

Biodegradable Polymers

Because PHB is stored by bacteria for eventual breakdownand utilization as a carbon source when extracellular carbonis no longer available, there must be an effective and rapidmechanism within the cell for the biodegradation of this highmolecular weight polyester into simple organic compounds.As discussed above, Lemoigne had been led into his studyof PHB by finding that [R]-3-hydroxybutyric acid is releasedby B. megateriumin an aqueous environment. In 1958,Wilkinson and co-workers also observed the release of bothacetoacetic acid and acetic acid during the utilization of PHBreserves by that bacterium.43 Subsequently, in 1962, Merrickand co-workers demonstrated that “native” granules fromB.rubrumwere self-hydrolyzing, and they isolated the enzymeresponsible for this reaction which they referred to as adepolymerase or hydrolase (1).44 In 1967, Williamson andco-workers identified a specific dehydrogenase (2) thatconverted [R]-3-hydroxybutyric acid to acetoacetic acid,45

and in 1973, Dawes and co-workers identified an enzymefor the conversion of acetoacetic acid to acetic acid (3),22

so the entire intracellular pathway for the reconversion ofPHB to acetic acid was established to include the followingsteps:

PHB can also be rapidly hydrolyzed to the monomer byextracellular depolymerase enzymes secreted by a widevariety of bacteria and fungi that can utilize this compoundafter it is liberated by the death and lyses of bacteria in whichit is stored. Initial observations made by Schlegel andChowdhury in 1963 with strains ofPseudomonasobtainedfrom soil and compost samples established this concept,46

and in 1965 Delafield, Doudoroff and co-workers isolatedand characterized a number of pseudomonads capable ofutilizing extracellular PHB as their sole source of carbonand energy.47

It is now known that microorganisms exist in all naturalenvironments that are capable of degrading PHB andmetabolizing [R]-3-hydroxybutyric acid by enzyme-catalyzedreactions, so by definition, PHB is a biodegradable polymer.In more recent studies, depolymerases have also been foundfor the PHAs with long alkyl chains.48 As mentioned above,these polyesters have much different physical and mechanicalproperties, and they can also be utilized as biodegradablepolymers in applications such as elastomers and adhesives.Many different types of intracellular and extracellularpolyester depolymerases have now been isolated and char-acterized. As reported by Doi and co-workers, all of theseenzymes consist of a single polypeptide chain in themolecular weight range of approximately 40 000-60 000.49

The structural genes of a large number of extracellulardepolymerases of different microorganisms have been iso-lated and analyzed, and they appear to have three charac-teristics in common along the polypeptide chain, including(1) a catalytic domain (termed a “lipid” box), (2) a substrate-binding domain, and (3) a linking region connecting thesetwo domains. In that manner, they have the same featuresas the depolymerizing enzymes for insoluble polysaccharidessuch as cellulose and chitin.

Genetic Engineering

In 1988, Dennis and co-workers at James MadisonUniversity cloned the entire set of genes inR. eutrophaforthe three enzymes involved in the synthesis of PHB fromacetyl CoA as described above.50 The three genes areclustered in one operon, and Dennis and co-workers wereable to introduce this operon intoE. coli. The geneticallyengineeredE. coli containing the operon can express all threeenzymes and can synthesize PHB in large quantities from awide range of organic compounds. Some recombinant strainsof E. coli can also produce the HB/HV copolymer,51 oralternatively as reported by Sinskey and co-workers at theMassachusetts Institute of Technology in 1994, strainscontaining only the synthase gene can express this proteinin sufficiently large quantities for isolation and purification.52

Figure 2. Transmission electron micrograph of ultrathin section ofAzotobacter chroococcum cell treated with phenylacetic acid. FromNuti et al., ref 17a, herein, with permission.

6 Biomacromolecules, Vol. 6, No. 1, 2005 Lenz and Marchessault

The purified enzyme is stable in aqueous solution and hasbeen used for in vitro polymerization reactions of a widevariety of 3- and 4-hydroxyalkanoate-CoA monomers.53 Lenzand co-workers at the University of Massachusetts reportedin 2000 that these in vitro polymerization reactions can form“living polymers”, which means that the polymerizationprocess has no polymer chain termination reaction, so thepropagating end group remains active indefinitely and veryhigh molecular weight polymers can be prepared in vitro.54

In another application of genetic engineering for bacterialpolyester synthesis, Somerville and co-workers at MichiganState University reported in 1992 that the reductase andsynthase genes ofA. eutrophuscan be inserted into a plant,Arabidopsis thaliana, which can also produce acetoacetyl-CoA, and the transgenic plant can then accumulate PHBgranules, Figure 4, to approximately 14% of its dry weight.55

PHB and the Macromolecule Controversy

Reserve polymers also played a role in the controversybetween Staudinger and his colleagues in organic chemistryin Germany during the 1920s over the very existence of“macromolecules”. In his book on the history of polymerscience, Morawetz discusses how organic chemists at thattime considered starch, which is a reserve polymer for plants,and cellulose to be colloidal aggregates of glucose moleculesrather than long chain polymers.56 A leading German organicchemist at that time, Karrer, reasoned that, because starch is

utilized by plants as a food and energy reserve, “one has tobe surprised that the view of hundreds or thousands ofglucose molecules joined together by glucosidic bonds intolong chains could have remained unchallenged” because “itis improbable that a plant in converting sugar to a reservesubstance from which it might soon have to be recoveredwould perform such complex work as would be required inthe build-up of a polyglucoside”.56 Had he known aboutPHB, Karrer would undoubtedly have made the sameargument against Lemoigne’s contention of the existence ofhigh molecular weight reserve polyesters in bacteria. Somuch for conventional wisdom.

Outlook

Despite the 75 years, on and off, of research on PHAsand 20 years of intense industrial interest, PHAs still appearto be far removed from large scale production. At thiswriting, two development programs on these biopolymersare receiving attention, namely (1) a joint program by theProctor & Gamble Co. and Kaneka Corp. on a family ofshort and medium chain copolymers, especially on poly(3-hydroxybutyrate-co-3-hydroxyhexanoate), and (2) a programat Metabolix Inc. on PHAs for medical applications. Thelack of commercialization of the initially promising bacterialpoly(3-hydroxybutyrate-co-3-hydroxyvalerate) copolymershas been generally attributed to the high investment for thefermentation and product recovery processes on a large scaleand to the cost of the substrates. To reduce the latterlimitation, alternative substrates are receiving much attention,including starch and vegetable oils, but no major break-throughs in this area have been announced. Nevertheless, inthe long run, it is possible that advances in our understandingand control of the genetic pathways involved in the biosyn-thesis of PHAs in microorganisms and plants could makethe industrial scale production of these biopolymers competi-tive with oil-based synthetic polymers.

As for the agricultural production of PHAs, the feasibilityof this route has been demonstrated in small plants such asArabidopsis thaliana,57 but the transfer of this technologyinto crops such as canola with acceptable production levelsis still in the research stage. On the other hand, the chemicalmodification of medium chain PHAs produced by bacteriais a promising approach to the commercialization of high-value polymers for specialty applications.58-60 Indeed, byeither direct bacterial synthesis or by the chemical modifica-tion of bacterially produced PHAs, polyesters with more thanone hundred different types of repeating units have beenidentified and characterized.61 Very recently, it was evenfound possible to produce a thioester analogue of the PHAswith bacteria,62 so it is apparent that there is still much moreto be discovered about the synthesis of bacterial polyesters.

Acknowledgment. We wish to recognize the importantcontributions by Professors Schlegel, Dawes, Merrick, andFuller for the factual details herein. In addition, Dr. BernardHautecoeur and Archives of the Institut Pasteur providedhistorical background concerning Professor Maurice Lem-oigne as surveyed by Rene´ Dujarric de la Riviere.63 Dr.Francis Werber kindly informed us on the development of

Figure 3. Moulded PHB objects for various applications. In soil burialor composting experiments, such objects biodegrade in about threemonths.

Figure 4. PHB granules in the choloroplast of Arabidopsis thaliana.With permission from Yves Poirier.

Historical Review of Bacterial Polyesters Biomacromolecules, Vol. 6, No. 1, 2005 7

PHB at W.R. Grace Company, where he was V.P. research.Finally, we are grateful to Professor Alexander Steinbu¨chelwho provided a critical review of our manuscript.

References and Notes

(1) Anderson, A. J.; Dawes. E. A.Microbiol. ReV. 1990, 54, 450.(2) Holmes, P. A. InDeVelopments in Crystalline Polymers Vol. 2;

Basset, D. C., Ed.; Elsevier Applied Science: London, 1988; pp1-65.

(3) Lemoigne, M.C. R. Acad. Sci.1923, 176, 1761.(4) Lemoigne, M.C. R. Acad. Sci.1924, 178, 1093.(5) Lemoigne, M.C. R. Acad. Sci.1924, 179, 253.(6) Lemoigne, M.C. R. Acad. Sci.1925, 180, 1539.(7) Lemoigne, M.Bull. Soc. Chim. Biol.1926, 8, 770.(8) Lemoigne, M.Bull. Soc. Chim. Biol.1927, 9, 446.(9) Steinbu¨chel, A. InBiotechnology, 2nd ed.; Vol. 6; Rehm, H. J., Reed,

G., Puhler, A., Stadler, P., Eds.: VCH Publishers: New York, 1996;pp 405-464.

(10) Morawetz, H. InPolymers: The Origin and Growth of a Science;John Wiley: New York, 1985; pp 41-416.

(11) Furukawa, Y. InInVenting. Polymer Science; University of Penn-sylvania Press: Philadelphia, 1998.

(12) Marchessault, R. H.; Yu, G. InBiopolymers: Polyesters II; Doi, Y.,Steinbuchel, A., Eds.; Wiley-VCH: Weinheim, 2002, pp 157-202.

(13) Williamson, D. H.; Wilkinson, J. F.J. Gen. Microbiol.1958, 19,198.

(14) Doudoroff, M.; Stanier, R. Y.Nature1959, 183, 1440.(15) Weibull, C.J. Bacteriol.1953, 66, 696.(16) Macrae, R. M.; Wilkinson, J. F.J. Gen. Microbiol.1958, 19, 210.(17) Lundgren, D. C.; Alper, R.; Schnaitman, C.; Marchessault, R. H.J.

Bacteriol.1965, 89, 245. (a) Nuti, M. P.; de Bertoldi, M.; Lepidi, A.A. Can. J. Microbiol. 1972, 18, 1257.

(18) Merrick, J. M.; Doudoroff, M.Nature1961, 189, 890.(19) Lusty, C. J.; Doudoroff, M.Proc. N. A. S.1966, 56, 960.(20) Stanier, R. Y.; Doudoroff, M.; Kunisawa, R.; Contopoulou, R.Proc.

N. A. S.1959, 45, 1246.(21) Oeding, V.; Schlegel, H. G.Biochem. J.1973, 134, 239.(22) Senior, P. J.; Dawes, E. A.Biochem. J.1973, 134, 225.(23) (a) Schlegel, H.; Gottschalk, G.; Von Bartha, R.Nature1961, 191,

463. (b) Schlegel, H. G., InNoVel Biodegradable Microbial Polymers;Dawes, E. A., Ed.: Kluwer: Dordrecht, 1990; pp 133-141.

(24) Stockdale, H.; Ribbons, D. W.; Dawes, E. A.J. Bacteriol.1968, 95,1798.

(25) Merrick, J. M.; Doudoroff, M.J. Bacteriol.1964, 88, 60.(26) Griebel, R.; Smith, Z.; Merrick, J. M.Biochemistry1968, 7, 3676.(27) Ballard, D. G. H.; Holmes, P. A.; Senior, P. J. InRecent AdVances

in Mechanistic and Synthesis Aspects of Polymers; Fontanille, M.,Guyot, A., Eds: Reidel (Kluwer) Pub., Lancaster, U.K., 1987; pp293-314.

(28) Kawaguchi, Y.; Doi, Y.Macromolecules1992, 25, 2324.(29) Haywood, G. W.; Anderson, A. J.; Dawes, E. A.FEMS Microbiol.

Lett. 1989, 57, 1.(30) Doi, Y. InMicrobial Polyesters; VCH Publishers: Weinheim, 1990.(31) Wallen, L. L.; Rohwedder, W. K.EnViron. Sci. Technol.1974, 8,

576.(32) Findlay, R. H.; White, D. C.Appl. EnViron. Microbiol. 1983, 45,

71.(33) de Smet, M. J.; Eggink, G.; Witholt, B.; Kingma, J.; Wynberg, H.J.

Bacteriol.1983, 154, 870.

(34) Doi, Y.; Kunioka, M.; Nakamura, Y.; Soga, K.Macromolecules1988,21, 2722.

(35) Braunegg, G.; Lefebvre, G.; Genser, K. F.J. Biotechnol.1998, 65,127.

(36) Byrom, D.Trends Biotechnol.1987, 5, 246.(37) Baptist, J. N.; Werber, F. X.SPE Transactions1964, 4, 245.(38) Forsyth, W. G. C.; Hayward, A. C.; Roberts, J. B.Nature1958, 182,

800.(39) Alper, R.; Lundgren, D. G.; Marchessault, R. H.; Cote, W. A.

Biopolymers1963, 1, 545.(40) (a) Merrick, J. M.; Lundgren, D. G.; Pfister, R. M.J. Bacteriol.1965,

80, 234. (b) Lundgren, D. G.; Pfister, R. M.; Merrick, J. M.J. Gen.Microbiol. 1964, 34, 441.

(41) Stinson, M. W.; Merrick, J. M.J. Bacteriol.1974, 119, 152.(42) Stuart, E. S.; Lenz, R. W.; Fuller, R. C.Can. J. Microbiol.1995, 41,

84.(43) Macrae, R. M.; Willkinson, J. F.J. Gen. Microbiol. 1958, 19,

210.(44) Merrick, J. M.; Delafield, F. P.; Doudoroff, M.Federation Proc.

1962, 21, 228.(45) Bergmeyer, H. V.; Gawehu, K.; Klotzach, H.; Krebs, H. A.;

Willkinson, D. H. Biochem. J.1967, 102, 423.(46) Chowdhury, A. A.Arch. Microbiol. 1963, 47, 167.(47) Delafield, F. P.; Doudoroff, M.; Palleroni, N. J.; Lusty, C. J.;

Contopoulou, R.J. Bacteriol.1965, 90, 1455.(48) Schirmer, A.; Matz, C.; Jendrossek, D.Can. J. Microbiol.1995, 41,

170.(49) Sudesh, K.; Abe, H.; Doi, Y.Prog. Polym. Sci.2000, 25, 1503.(50) Slater, S. C.; Volge, W. H.; Dennis, D. E.J. Bacteriol.1988, 170,

4431.(51) Slater, S.; Gallaher, T.; Dennis, D.Appld. EnViron. Microbiol. 1992,

58, 1089.(52) Gerngross, T. V.; Snell, K. D.; Peoples, O. P.; Sinskey, A. J.; Cauhai,

E.; Masamune, S.; Stubbe, J.Biochemistry1994, 33, 9311.(53) Zhang, S.; Lenz, R. W.; Goodwin, S. InBiopolymers: Polyesters I;

Doi, Y., Steinbuchel, A., Eds.: Wiley-VCH: Weinheim, 2002, pp353-372.

(54) Su, L.; Lenz, R. W.; Takagi, Y.; Zhang, S.; Goodwin, S.; Zhong, L.;Martin, D. P.Macromolecules2000, 33, 229.

(55) Poirier, Y.; Dennis, D.; Klompareus, K.; Nawrath, C.FEMSMicrobiol. ReV. 1992, 103, 237.

(56) Morawetz, H. InPolymers: The Origin and Growth of a Science;John Wiley: New York, 1985, Chapter 10.

(57) Poirier, Y.; Nawrath, C.; Somerville, C.Biotechnology1995, 13, 142.(58) Gagnon, K. D.; Lenz, R. W.; Farris, R. J.; Fuller, R. C.Polymer

1994, 35, 4358.(59) Dufresne, A.; Reche, L.; Marchessault, R. H.; Lacroix, M.Int. J.

Biol. Macromol.2001, 29, 73.(60) Hany, R.; Bo¨hlen, C.; Geiger, T.; Hartmann, R.; Kawada, J.; Schmid,

M.; Zinn, M.; Marchessault, R. H.Macromolecules2004, 37, 385.(61) Steinbu¨chel, A.; Valentin, H. E.FEMS Microbiol. Lett.1995, 128,

219.(62) Lutke-Eversloh, T.; Fischer, A.; Remminghorst, U.; Kawada, J.;

Marchessault, R. H.; Bo¨gershausen, A.; Kalwei, M.; Eckert, H.;Reichelt, R.; Liu, S.-J.; Steinbu¨chel, A. Nat. Mater.2002, 1, 236.

(63) Rene´ Dujarric de la Riviere “Notice Necrologique sur M. MauriceLemoigne (1883-1967) C. R. Acad. des Sci., Paris, t. 264 (12 juin1967).

BM049700C

8 Biomacromolecules, Vol. 6, No. 1, 2005 Lenz and Marchessault