Nisin A extends the shelf life of high-fat chilled dairy dessert, a milk-based pudding

ORIGINAL ARTICLE

Nisin A extends the shelf life of high-fat chilled dairydessert, a milk-based puddingS. Oshima1,2, A. Hirano2, H. Kamikado2, J. Nishimura1, Y. Kawai3 and T. Saito1

1 Laboratory of Animal Products Chemistry, Graduate School of Agricultural Science, Tohoku University, Sendai, Miyagi, Japan

2 Food Technology Research Laboratories, R&D Division, Meiji Co., Ltd., Odawara, Kanagawa, Japan

3 Laboratory of Milk Science, College of Bioresource Sciences, Nihon University, Fujisawa, Kanagawa, Japan

Keywords

Bacillus cereus, Bacillus thuringiensis, chilled

dessert, nisin, pudding, spoilage.

Correspondence

Tadao Saito, Laboratory of Animal Products

Chemistry, Graduate School of Agricultural

Science, Tohoku University, 1-1 Tsutsumidori-

Amamiyamachi, Aoba-ku, Sendai, Miyagi

981-8555, Japan.

E-mail: [email protected]

2013/2165: received 27 October 2013,

revised 24 December 2013 and accepted 13

January 2014

doi:10.1111/jam.12454

Abstract

Aims: The aims of this study were to evaluate the effectiveness of nisin A to

control the growth of spore-forming bacteria, Bacillus and Paenibacillus, in

chilled high-fat, milk pudding and to reduce heat treatment to improve aroma

and flavour.

Methods and Results: Nisin A was added to milk pudding containing 5�0 and

7�5% fat to final concentrations of 40, 80, 120 and 240 IU ml�1. Spores from

Bacillus thuringiensis, Bacillus cereus and Paenibacillus jamilae were inoculated

into samples at 10 spores ml�1 prior to pasteurization at 130°C for 2 s. Milk

pudding without inoculation was pasteurized using less heat condition (100,

110 and 120°C for 2 s) to measure the effect of adjusting the ingredients to

prevent naturally occurring bacteria. The viable cells during storage at 15, 20

and 30°C showed nisin A inhibited spiked bacteria to varying degrees

depending on species, sensitivities to nisin A concentration and fat content,

and inhibited natural populations at 80 IU g�1 nisin A in 5�0% fat and at

120 IU g�1 in 7�5% fat milk pudding. An aroma compound analysis and

organoleptic assessment showed processing at 110 and 120°C decreased the

temperature-dependent unpleasant odours, for example, reduced dimethyl

sulfide and dimethyl disulfide by 1�2–1�5 times and increased rankings in taste

tests compared with 130°C treated pudding.

Conclusions: Nisin A was found to be effective as a natural preservative to

control spoilage bacteria in high-fat milk pudding and extend its shelf life,

when using reduced heat treatments to improve the flavour and aroma

without compromising food safety.

Significance and Impact of the Study: This is the first report showing nisin A

is effective in reducing spoilage bacteria in high-fat, chilled dessert, milk

pudding. Therefore, nisin A can be used to improve milk puddings to satisfy

both industry and consumer demand for food quality and safety.

Introduction

Bacteriocins are antimicrobial peptides produced by vari-

ous lactic acid bacteria (LAB) including Lactococci, Lacto-

bacillus and Pediococci (Klaenhammer 1988; Jack et al.

1995). Many LAB bacteriocins have a relatively broad

antimicrobial spectrum against important food-borne

pathogens, for example Listeria monocytogens and Staphy-

lococcus aureus as well as spoilage bacteria(H�echard et al.

1992; Sobrino-L�opez and Mart�ın-Belloso 2008 ). LAB

bacteriocins may be used as natural food preservatives

(Gonzalez et al. 1994). Recent studies show bacteriocins

in food preservation offer several benefits (Thomas et al.

2000): (i) extended shelf life of foods, (ii) extra protec-

tion during temperature abuse conditions, (iii) decreased

risk of transmission of food-borne pathogens, (iv)

reduced economic loss due to food spoilage, (v) reduced

use of chemical preservatives and (vi) use of less severe

heat treatments without compromising food safety.

Despite the many benefits of bacteriocins, to date, only

Journal of Applied Microbiology 116, 1218--1228 © 2014 The Society for Applied Microbiology1218

Journal of Applied Microbiology ISSN 1364-5072

Nisin A is allowed as a food additive (E234, 1983), recog-

nized as safe (WHO 1969), and accepted by the American

Food and Drug Administration (FDA/HHS 1998).

Nisin A is a polypeptide produced by strains of Lacto-

coccus lactis subsp. lactis. Structurally, it is a 34-amino

acid peptide and is cationic due to a combination of free

lysine residues and one or more histidine residues (Cleve-

land et al. 2001). Nisin A is bactericidal against many

Gram-positive organisms, including the vegetative cells of

spore formers, and is bacteriostatic against spores. The

antimicrobial mechanism of nisin A is to bind to the

outer membrane receptors by conjugation with other cell

components (i.e. phospholipid) or aggregation with other

proteins (i.e. glycoprotein). This creates ion channels

through the cytoplasmic membrane, rendering the bacte-

rial cell permeable (Delves 1990a). Commercially available

nisin A is a dried concentrated powder celled Nisaplin�

(Danisco A/S, Grindsted, Denmark) that is heat stable,

soluble in water and suitable for dairy products. It is used

in many dairy products to prevent proliferation of unde-

sired bacteria, for example processed cheese, chilled des-

serts, pasteurized milk, heat-sterilized milk, flavoured

milk, clotted cream and canned evaporated milk (Hurst

and Hoover 1993; Thomas et al. 2000).

Controlling food-borne pathogenic bacteria and ensur-

ing the safety of food products are the most important

issues for food processors. With globalization of the food

market, it is difficult for testing authorities, consumers, as

well as food processors to control the microbial contami-

nants in raw materials and thus the appropriate storage

temperature for the final products during handing and

transportation (O’Mahony et al. 2001; Martinez Viedma

et al. 2009 ). In addition to food safety issues associated

with food-borne pathogens, food processors have the risk

of economic loss due to spoilage resulting from microbial

contamination and propagation in improperly stored

products.

Nisin A is expected to be a tool for food biopreserva-

tion and improve food technology; however, nisin A use

is limited because its greatest antimicrobial activity occurs

under acid conditions (Chung et al. 1989; Liu and Han-

sen 1990). In a pH rise from 3 to 7, the nisin A molecule

becomes increasingly vulnerable to heat (Delves 1990a).

Further studies show the effectiveness of nisin A is often

decreased by food-related factors, especially fat or pro-

tein, because nisin A binds to fat or protein surfaces,

resulting in reduced accessibility to bacterial cells (Laridi

et al. 2003). Therefore, the effectiveness of nisin A as a

biopreservative has to be evaluated for each new product.

Among the dairy products, chilled desserts are nutrient-

rich, high-fat, and high-protein products, and therefore,

the efficacy of nisin A may be decreased by the interaction

between the food matrix and nisin A, shortening the

microbial shelf life. Previous studies with chilled desserts

using nisin A showed controlled spoilage and extended

shelf life in ‘Cr�eme Caramel’, ‘Chocolate Dairy Dessert’

and ‘Quarg Dessert’ (Anon 1985; Plockov�a et al. 1998).

However, these studies are of limited value because nisin

A acts differently in various food products, and its con-

centration was unavailable to high-fat milk pudding.

Further, nisin A may not progress to practical use in

dairy desserts because transition from laboratory to food

production (from pilot scale to commercial scale)

has been relatively unsuccessful (Jones 1974; Jung et al.

1992).

The purpose of this study was to evaluate the effective-

ness of nisin A as a biopreservative in a thermized, milk-

based dessert pudding, one of the most popular chilled

dairy desserts available in Japan, and explore the possibil-

ity of improving the organoleptic quality using lower pas-

teurizing temperatures without compromising food

safety.

Materials and methods

Bacteria strains and spore preparation

Three spore-forming bacterial strains from the Meiji Co.,

Ltd. (Tokyo, Japan) bacterial collection, Bacillus thuringien-

sis B-1695, Bacillus cereus B-344 and Paenibacillus jamilae

S-34, which were isolated from spoiled pudding, pasteurized

milk, and raw milk, respectively, were grown in Trypticase

soy broth at 30°C for 16 h. A portion (0�3 ml) of growth

medium was spread onto Schaeffer’s sporulation medium

agar (Schaeffer et al. 1965) (8 g l�1 nutrient broth,

10 ml l�1 of 10% (w/v) KCl, 10 ml l�1 of 1�2% (w/v)

MgSO4�7H2O, 0�5 ml l�1 of 1 mol l�1 NaOH, 1�0 ml l�1

of 1 mol l�1 Ca (NO3)2, 1�0 ml l�1 of 0�01 mol l�1 MnCl2,

1�0 ml l�1 of FeSO4, and 15 g l�1 agar for plates). These

cultures were incubated at 30°C, and sporulation was moni-

tored by microscopy following staining with malachite

green. Spores were harvested using phosphate-buffered sal-

ine (PBS: 10 mmol l�1 phosphate, 150 mmol l�1 NaCl, pH

7�4). The spore culture was washed with PBS, harvested by

centrifugation at 2200 g for 10 min. The pellet was washed

three times with sterile distilled water, resuspended in 5 ml

of distilled water and then heated at 80°C for 10 min to

inactivate nonsporulated vegetative cells. The concentration

of endospores was estimated using pour plating on plate

count agar (PCA; Eiken, Tokyo, Japan). The spores were

stored at 4°C.

Nisin A and bacteriocin assay

Nisin A was added as Nisaplin�, a commercial prepara-

tion that is 2�5% active nisin A (Danisco A/S, Grindsted,

Journal of Applied Microbiology 116, 1218--1228 © 2014 The Society for Applied Microbiology 1219

S. Oshima et al. Application of nisin A for milk-based pudding

Denmark), to the milk-based pudding at least 1 h before

processing. Nisaplin� has a standard activity of

1 9 106 IU g�1, whereas pure nisin A has an activity of

40 9 106 IU g�1 (Beard et al. 1999). The activity of nisin

A was assessed using the agar well diffusion assay with

Lactobacillus delbrueckii subsp. bulgaricus JCM1002T as

the indicator strain as described previously (Arakawa

et al. 2009). MRS agar inoculated with a 0�1% overnight

culture of indicator strain was poured into a Petri dish.

After solidification, a 6-mm well was made in the plates

with a sterile borer. Fifty microlitre of serially diluted

sample was introduced into the well, and plates were

incubated at 37°C for 18 h. The unit of bacteriocin activ-

ity (Arbitrary Unit (AU)) was defined as the highest dilu-

tion producing a zone of inhibition 8 mm in diameter.

The bacteriocin activity of Nisaplin� using the JCM1002T

strain is 8 192 000 AU g�1.

Preparation of milk-based dessert pudding

Two different fat concentrations (5�0%, 7�5%) of milk-based

pudding (milk pudding) were prepared. The basic composi-

tion of milk pudding was skimmilk (83 g kg�1), cream (dou-

ble cream with a fat content of 47%: 33 g or 52 g kg�1),

vegetable oil (34 g or 50 g kg�1), lactose (3 g kg�1), sugar

(62 g kg�1), starch (5 g kg�1), dextrin (13 g kg�1), gelatin

(2 g kg�1), agar (2 g kg�1), sodium bicarbonate (1 g kg�1),

palm-hydrogenated oil monoglyceride (2 g kg�1) and deion-

ized water. All ingredients except for cream and vegetable

oil were mixed warm to 40°C in a mixing tank. Cream and

vegetable oil were added and heated to 70°C, and subse-

quently, the preparation was homogenized for 10 min at

4000 g avoiding cream plug formation, and then, the prepa-

ration was maintained at 60–70°C until after the processing

tests.

Effect of nisin A on Bacillus and Paenibacillus species

spores

Nisin A was added to the preparation mixture of milk pud-

ding at the different concentration of 40, 80, 120 and

240 IU g�1. After thoroughly mixing, they were pasteurized

at 130°C for 2 s using a pilot-plant-scale plate heat exchan-

ger 50 LPH (50 l h�1; SPX APV A/S, Kolding, Denmark)

with pressure homogenization at 15 MPa using a two-stage

homogenizer (1st stage 10 MPa, 2nd stage 5 MPa; Sanwa

Machine, Shizuoka, Japan). The product mixtures were dis-

pensed into sterilized containers with precautions to mini-

mize postprocessing contamination and then chilled to

below 10°C. Bacillus, and Paenibacillus spores diluted in

PBS were added to the product mixtures to a final concen-

tration of 10 CFU g�1 and incubated at 15, 20 and 30°Cfor 29 days. To determine the antimicrobial effect of nisin

A, viable cell numbers in these products were determined

on PCA using pour plating or the spiral inoculation tech-

nique using an Eddy Jet Spiral Plater (Iul S/A, Barcelona,

Spain). Colonies were counted after incubation at 30°C for

24 h. Sample product mixtures with no nisin A were used

as the control.

Effect of nisin A on the natural microbiota in the

ingredients

We determined the inhibitory effect of nisin A with

ingredient-derived bacteria in product mixtures that were

processed using mild heat treatment. Milk pudding prod-

uct mixtures with or without nisin A were prepared and

treated at 100, 110 and 120°C as described above. The

samples were stored at 15, 20 and 30°C for 29 days.

Microbial counts were determined during storage. Sepa-

rate colonies of micro-organisms that survived the pas-

teurization were isolated and clone purified transferring

at least three times on PCA, and then we identified

species using 16S rRNA gene analysis.

Aroma compounds analysis

To identify the flavour characteristics of milk pudding with

nisin A (240 IU g�1) processed using different pasteuriza-

tion temperatures from 100 to 130°C, an aroma compound

analysis was performed using the head space solid-phase

microextraction (HS-SPME) method (Kataoka et al.

2000). Milk pudding samples stored in the refrigerator for

3 days were used for the analysis. Samples were diluted

twice with saturated saline, and 10 g of solution was placed

in a vial (20 ml) and held in water bath at 60°C for

40 min. After equilibration, a SPME fibre (50/30 lmDivinylbenzene/CarboxenTM/Polydimethylsiloxane (DVB/

CAR/PDMS), Stableflex (2 cm); Sigma Aldrich, St. Louis,

MO) was inserted into the vial, and the aroma constituents

were extracted. The aromatic compounds were analysed

using a GC/MS system (Agilent Technologies, Santa Clara,

CA) and identified by mass spectrometry.

Organoleptic assessment

An organoleptic assessment of the 5�0% fat milk pudding

with and without nisin A (240 IU g�1) stored at 10°Cfor 2 days after processing at 100, 110 120 and 130°C for

2 s was made using 15 taste panelists with high taste

senses. Panelists evaluated ‘milk taste’, ‘sweetness’, ‘rich-

ness’, ‘sulphur-like odour’,’ smoothness’, ‘good finish’,

‘clean after taste’ and ‘overall evaluation’. They rated each

sample for preference and intensity on a scale between 1

and 5, where a score of 5 corresponded to very good and

a score of 1 corresponded to very poor.


Application of nisin A for milk-based pudding S. Oshima et al.

Statistical analysis

Microbiological spike tests were repeated three to four

times. The average data � standard deviations were

determined with Excel program. For organoleptic assess-

ment, ANOVA was performed on the sensory data obtained

from assessor. Significant differences between treatments

determined with Student’s t-test using Microsoft Excel

2003 software (Microsoft Corp., Redmond, WA) were

considered significant when P < 0�01 and P < 0�05.

Results

The microbiological spike tests using three species of

Bacillus and Paenibacillus genus were conducted assuming

the origin ingredients and postprocessing contamination

by heat-resistant spore formers. In parallel, milk pudding

processed at 130°C for 2 s without nisin A was incubated

at 30°C for 48 h to verify the existence of spoilage bacte-

ria. No microbial activity could be detected in all the

nisin A-free samples, suggesting that there was no bacte-

rial contamination (data not shown). Bacterial popula-

tions larger than 107 CFU g�1 in the food are generally

considered to show the food spoilage in the case that

contaminated bacteria have the capacity of producing

acids, gas and enzyme such as protease and lipase with

the growing population.

Effect of nisin A against Bacillus thuringiensis in milk

pudding

The growth of B. thuringiensis B-1695 in milk pudding

where the fat content was 5�0 and 7�5% is shown in

Fig. 1 and 2, respectively. In 5�0 and 7�5% milk pudding

without nisin A, the B. thuringiensis B-1695 population

reached more than 7 log CFU g�1 in 2 days at 15°C, in2 days at 20°C and in 1 day at 30°C. However, depend-

ing on the concentration of nisin A, the growth of

B. thuringiensis B-1695 in milk pudding was delayed

except for milk pudding containing 7�5% fat stored at

30°C. Considering that the initial number of B. thuringi-

ensis in milk pudding is approximately 10 CFU g�1 to

assure shelf life, the microbial quantity must be below

107 CFU g�1 with added nisin A. When milk pudding

was held at incorrect storage temperatures of 15 or 20°C,the addition of nisin A at 80 IU g�1 and upper concen-

trations in 5�0% fat milk pudding increased the shelf life

to 20 days.

Effect of nisin A against Bacillus cereus in milk pudding

The growth of B. cereus B-344 in milk pudding in 5�0and 7�5% fat is shown in Fig. 3. In the control sample

without nisin A, the B. cereus reached over 107 CFU g�1

after 3, 2 and 1 day at 15, 20 and 30°C, respectively. In5�0% fat pudding, nisin A at 80 IU g�1 and upper con-

centrations had bacteriostatic effects on B. cereus B-344

for 29 days (below detection levels), whereas addition of

40 IU g�1 had no growth at 15°C, and a similar growth

pattern compared with control (without nisin A) at 20

and 30°C (Fig. 3a–c). With a fat content of 7�5% in the

milk pudding, 240 IU g�1, the largest amount of nisin A

had little or no effect in controlling the growth of

B. cereus B-344 at 20 and 30°C. At 15°C, milk pudding

with 40 IU g�1 of nisin A exhibited a decline in the

growth rate and with ≥80 IU g�1 inhibited the growth of

123456789

0 4 8 12 16 20 24 28

Log

cfu

g–1

Time (days)

(a)

1

2

3

4

5

6

7

8

9

0 4 8 12 16 20 24 28

Log

cfu

g–1

Time (days)

(b)

1

2

3

4

5

6

7

8

9

0 4 8 12 16 20 24 28

Log

cfu

g–1

Time (days)

(c)

Figure 1 Effect on nisin A on the growth of Bacillus thuringiensis

B-1695 in milk-based pudding containing 5�0% fat at storage

temperatures of 15°C (a), 20°C (b) and 30°C (c). (■) Controls. Samples

were treated with nisin A at concentrations of 40 IU g�1 (◊), 80 IU g�1

(Δ), 120 IU g�1 (○) or 240 IU g�1 (9).



B. cereus B-344 to below detection levels for 29 days

(Fig. 3d–f).

Effect of nisin A against Paenibacillus jamilae in milk

pudding

The control samples without nisin A attained c.

108 CFU g�1 in both 5�0% fat and 7�5% fat after 5 days

at 15°C, 3 days at 20°C and 2 days at 30°C, whereas milk

pudding including more than 40 IU g�1 of nisin A pre-

vented the growth of inoculated P. jamilae S-34 for

29 days (Fig. 4). Paenibacillus jamilae is most sensitive

species among the spiked three spore-forming bacteria.

Nisin A activity during storage

Residual antimicrobial activity of nisin A in milk pudding

immediately after pasteurization and during storage at

15°C was evaluated. The results are shown in Table 1. All

samples, at concentrations of 40, 80, 120 and 240 IU g�1,

maintained the activity for up to 9 days after pasteuriza-

tion. The residual activity levels observed indicated high

revel retention, with only 25–50% loss in nisin A activity

for up to 27 days at 15°C.

Control of microbial content in mild-heated milk

pudding

Bacterial species of isolates that survived through pasteur-

ization are listed in Table 2. Each living cell number in

the control milk puddings without nisin A attained a

count 106 CFU g�1 after 2 days at 30°C in both 5�0% fat

and 7�5%. Especially, B. cereus strains were isolated from

every control sample and also detected from samples in

7�5% fat with nisin A (40 IU g�1) treated at 100 or

110°C for 2 s, at storage temperatures of 15 and 30°C,respectively. The genera Brevibacillus strains, Br. parabre-

vis and Br. agri, were isolated from pudding with

40 IU g�1 of nisin A in 5�0% fat (100°C 2 s) and

80 IU g�1 in 7�5% fat (100, 110°C 2 s). With addition of

120 and 240 IU g�1 of nisin A, growth of spoilage bacte-

ria was not observed for 29 days at 15, 20 and 30°C.

Aroma evaluation of milk pudding

Seven substances in milk puddings with 240 IU g�1 of

nisin A were analysed as indicators of unpleasant aroma

change. Dimethyl sulfide (DMS) and dimethyl disulfide

(DMDS) are derived from thermal degradation of pro-

tein. 2-Pentanone, 2-heptanone and 2-nonanone are

derived from thermal degradation of milk fat. 2-Furfural

and 2-furanmethanol are substances caused by the Mail-

lard reaction. The peak area ratio of these substances

given in the GC/MS chromatogram is shown in Fig. 5.

The sum of DMS and DMDS generated was temperature

dependent and increased by 1�29 at 110°C, 1�59 at

120°C, 1�89 at 130°C relative to 100°C (Fig. 5a). Con-

versely, indicator substances of milk fat degradation

(Fig. 5b) and Maillard reaction (Fig. 5c) showed little or

no change from 100 to 120°C and comparatively

increased at 130°C relative to 100°C by 2�09 and 1�39,

respectively.

Organoleptic properties of lower heated milk pudding

The average scores of 7 attribute evaluation terms of

5�0% fat thermalized milk pudding with 240 IU g�1 of

1

2

3

4

5

6

7

8

9

0 4 8 12 16 20 24 28

Log

cfu

g–1

Time (days)

(b)

(a)

1

2

3

4

5

6

7

8

9

0 4 8 12 16 20 24 28

Log

cfu

g–1

Time (days)

(c)

123456789

0 4 8 12 16 20 24 28

Log

cfu

g–1

Time (days)

Figure 2 Effect of nisin A on the growth of Bacillus thuringiensis

B-1695 in milk-based pudding containing 7�5% fat at storage tem-

peratures of 15°C (a), 20°C (b) and 30°C (c). (■) Controls. Samples

were treated with nisin A at concentrations of 40 IU g�1 (◊),80 IU g�1 (Δ), 120 IU g�1 (○) or 240 IU g�1 (9).



nisin A are shown in Fig. 6. ‘Milk taste’ and ‘overall eval-

uation’ were significantly higher at 110 and 120°C in

comparison with 130°C (Fig. 6b,c), and also ‘“richness”

at 110°C showed significantly higher scores (P < 0�05)(Fig. 6b).’ Sulphur-like odour’ decreased with a reduction

in pasteurizing temperature (Fig. 6a–c). There was no

significant difference (P ≥ 0�05) between with and with-

out nisin A (data not shown). These results indicate it

intends to be a preference for pasteurized milk pudding

at 110 and 120°C for 2 s.

Discussion

Nisin A is the only bacteriocin that has been approved by

the World Health Organization for use as a food preser-

vative and is permitted currently for use in over 50 coun-

tries. Previous studies demonstrated nisin A is capable of

preventing the outgrowth of bacterial spores that are sig-

nificant in a wide range of heat-processed foods (Delves

1990a). However, there are several limitations in using

nisin A in dairy products. First, the nisin A molecule is

active and exhibits its greatest solubility and stability

under acidic conditions. Delves (1990a) shows raising the

pH from 3 to 7, the nisin A molecule becomes increas-

ingly more vulnerable to the effect of heat decreasing to

0�5% at pH 7 compared with 100% at pH 3. Second,

food components can decrease the antimicrobial activity

of nisin A due to proteolytic degradation (Jung

et al.1992; Bhatti et al. 2004; Chollet et al. 2008) or bind-

ing of the peptide to fats or proteins, resulting in reduced

accessibility of nisin to the target bacterial cells (Laridi

et al. 2003). Among the dairy products, chilled desserts

are one of the highest nutrient-rich, high-fat, and high-

protein products where the efficacy of nisin A may be

1

2

3

4

5

6

7

8

9

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(b)

1

2

3

4

5

6

7

8

9

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(c)

123456789

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(a)

1

2

3

4

5

6

7

8

9

0 2 4 6 8

Log

cfu

g–1

Time (days)

(e)

1

2

3

4

5

6

7

8

9

0 2 4 6 8

Log

cfu

g–1

Time (days)

(f)

1

2

3

4

5

6

7

8

9

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(d)

Figure 3 Effect of nisin A on the growth of

Bacillus cereus B-344 in milk-based pudding

containing 5�0% fat at storage temperatures

of 15°C (a), 20°C (b) and 30°C (c) and

containing 7�5% fat at 15°C (d), 20°C (e)

and 30°C (f). (■) Controls. Samples were

treated with nisin A at concentrations of

40 IU g�1 (◊), 80 IU g�1 (Δ), 120 IU g�1 (○)

or 240 IU g�1 (9).



123456789

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(b)

123456789

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(c)

123456789

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(a)

123456789

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(e)

123456789

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(f)

123456789

0 2 4 6 8 26 28

Log

cfu

g–1

Time (days)

(d)

Figure 4 Effect of nisin A on the growth of

Paenibacillus jamilae S-34 in milk-based

pudding containing 5�0% fat at storage

temperatures of 15°C (a), 20°C (b) and 30°C

(c) and containing 7�5% fat at 15°C (d), 20°C

(e) and 30°C (f). (■) Controls. Samples were

treated with nisin A at concentrations of

40 IU g�1 (◊), 80 IU g�1 (Δ).

Table 1 Nisin A activity in the milk-based pudding with a 5�0% fat content processed at 130°C for 2 s during storage at 15°C

Concentration

of nisin

A (IU g�1)

Before

treatment

After

treatment

Storage time (day) (AU ml�1)

1 5 9 11 17 20 21 27

40 5120 5120 5120 5120 5120 2560 (50) 1280 (25) 2560 (50) 1280 (25) 1280 (25)

80 5120 5120 5120 5120 5120 5120 2560 (50) 2560 (50) 5120 2560 (50)

120 10 240 10 240 10 240 10 240 10 240 10 240 10 240 10 240 5120 (50) 5120 (50)

240 20 480 20 480 20 480 20 480 20 480 20 480 20 480 20 480 10 240 (50) 10 240 (50)

Residual nisin A activity expressed as % scale in parentheses.

Table 2 Identification of isolated bacteria from milk-based pudding with or without nisin A

Concentration

of nisin A

Fat content

in sample (%)

Process

condition

Storage

temperature (°C)

Storage

time (days)*

Bacterial

species

Control (0 IU g�1) 5�0 100°C 2 s 30 2 Bacillus cereus

Control (0 IU g�1) 7�5 100°C 2 s 30 1 B. cereus

40 IU g�1 5�0 100°C 2 s 20 14 Brevibacillus parabrevis

40 IU g�1 7�5 110°C 2 s 15 13 B. cereus

40 IU g�1 7�5 100°C 2 s 30 13 B. cereus

80 IU g�1 7�5 110°C 2 s 30 4 Brevibacillus agri

80 IU g�1 7�5 100°C 2 s 30 4 Br. agri

*Storage times to attain a bacterial count of 106 CFU g�1.



decreased by the interaction between the food matrix and

nisin A, and therefore, the shelf life may be shortened

(Anon 1985).

We used nisin A in milk pudding as a preservative for

chilled dairy dessert because nisin A is approved in the

food category of western confectionery including milk

pudding by Japanese regulations in 2009 when added at a

maximum concentration of 6�25 mg kg�1 (250 IU g�1).

The raw material and composition of milk pudding used

in this study were determined by reference to commer-

cially available chilled milk puddings in the Japanese

market. Fat components of the samples were 5�0% (milk

fat 1�5%) and 7�5% (milk fat 2�4%). The pasteurization

temperature and holding time at the final stage of the

heat exchanger was 130°C for 2 s, which is widely used

to extend the shelf of dairy products in Japan.

Nisin A is a relatively heat-stable peptide. Previous

studies evaluated the thermo stability of bacteriocins in

0·0

2·0

4·0

6·0

8·0

100°C 110°C 120°C 130°C

Pea

k ar

ea r

atio

Process temperature

(a)

0·0

2·0

4·0

6·0

8·0

100°C 110°C 120°C 130°C

Pea

k ar

ea r

atio

Process temperature

(b)

0·0

2·0

4·0

6·0

8·0

100°C 110°C 120°C 130°C

Pea

k ar

ea r

atio

Process temperature

(c)

Figure 5 Comparison of aroma compounds from milk pudding pro-

duced using different processing temperatures. (a) Aroma substances

derived from the thermal degradation of protein, dimethyl sulfide

(DMS) ( ) and dimethyl disulfide (DMDS) (■). (b) Aroma substances

derived from the thermal degradation of milk fat, 2-pentanone (□), 2-heptanone ( ) and 2-nonanone (■). (c) Aroma substances caused by

the Maillard reaction, 2-furfural ( ) and 2-furanmethanol (■).

Milk

Taste**

Sweetness

Richness*

Sulfur-like

odor

Smoothness

Good finish

Clean

aftertaste

Overall*

evaluation

–1

–0·5

0

0·5

1

1·5

110°C 130°C

(b)

Milk taste

Sweetness

Richness

Sulfur-like

odorSmoothness

Good finish

Clean

aftertaste

Overall

evaluation

–1

–0·5

0

0·5

1

1·5

130°C 100°C

(a)

Milk taste*

Sweetness

Richness

Sulfur-like

odorSmoothness

Good finish

Clean

aftertaste

Overall**

evaluation

–1

–0·5

0

0·5

1

1·5

130°C 120°C

(c)

Figure 6 Organoleptic evaluation of mild thermized milk pudding

(grey line) in comparison with control pudding treated at 130°C for

2 s (black line). (a) 100°C for 2 s vs 130°C for 2 s, (b) 110°C for 2 s

vs 130°C for 2 s, (c) 120°C for 2 s vs 130°C for 2 s. *P < 0�05,**P < 0�01.



broth, skim milk or whole milk by pasteurization at

60–120°C for 5–20 min (Hata et al. 2010; Rehaiem et al.

2010), or high-temperature, short-time (HTST) pasteuri-

zation at 72, 90 and 115°C for 15, 15 and 2 s, respectively

(Wirjantoro and Jewis 1996). As holding–heating gives a

higher heat load than continuous heating, antimicrobial

peptides treated using holding–heating are more vulnera-

ble to thermal load. However, nisin is generally thermo-

protected by food ingredients and does not lose activity

during pasteurization or sterilization (Henning et al.,

1986). Therefore, we needed to confirm the antimicrobial

effect of nisin A in a real product as well as under manu-

facturing conditions. Our data show nisin A in milk pud-

ding at neutral pH 7�04 including a fat component of 5�0or 7�5% maintained its initial activity values after heat

processing at 130°C for 2 s and retained a high revel

activity (25–50% loss) up to 27 days under an abuse tem-

perature of 15°C. This is the first report showing the

thermostability of nisin A in chilled pudding under the

ultrapasteurization conditions and suggesting commercial

application of nisin A in neutral pH chilled desserts.

The antimicrobial effect of nisin A against B. thuringi-

ensis B-1695, B. cereus B-344 and P. jamilae S-34 spore

spiked into milk pudding having 5�0 or 7�5% fat was

determined. The data show B. thuringiensis B-1695 has

the highest resistance to nisin A; in contrast, P. jamilae

S-34 has sensitivity to nisin A in milk pudding. Paeniba-

cillus jamilae S-34 was bactericidally inhibited by nisin A

at a concentration of 40 IU g�1 for at least 29 days at 15,

20 and 30°C.For B. thuringiensis B-1695 and B. cereus B-344, the

efficacy of nisin A in 7�5% fat pudding was shown a gen-

eral decrease compared with 5�0% fat. Previous reports

show the activity of nisin against Listeria monocytogenes

decreased as the milk fat concentration increased (Jung

et al. 1992), and the maximum antilisterial effect in skim

milk was reduced in milk with ≥2�0% fat (Bhatti et al.

2004). Further, homogenization reduced the antilisterial

activity of nisin, which may be due to the binding of

nisin to milk fat globules. This may be prevented by add-

ing emulsifiers such as polyoxyethylene sorbitan monool-

eate (Tween 80) (Bhatti et al. 2004). We clarified the

activity of nisin A against Bacillus spp. decreases in 2�4%milk fat in milk pudding (total fat 7�5%) as previously

reported for Listeria spp. The milk fat globules in milk

pudding are reduced to approximately 1 lm in diameter

by homogenization (15 MPa), increasing the surface area

to which nisin A can bind. However, further palm-hydro-

genated oil monoglyceride was added to products as an

emulsifier, consequently nisin A was prevented from

adsorption to the gloubules’ surface.

One of the benefits of adding nisin A is an extended

shelf life of food products sold at ambient temperature.

In general, the shelf life of foods is determined to multi-

ply the maximum tolerable period regarding physics,

chemistry and microbial quality by safe factor of 0�7.Therefore, microbiological shelf life of milk pudding with

nisin A was estimated using the growth curve of the resis-

tant strain, B. thuringiensis B-1695. These calculation

results suggested that nisin A is effective in extending

shelf life of milk pudding several-fold compared with

control products without nisin A. Especially, nisin A at

levels of 240 IU g�1 in milk pudding with standard pas-

teurization provided an extension of shelf life to 4 weeks.

Nisin A that exerts a bacteriostatic effect on spiked

B. thuringiensis B-1695 and B. cereus B-344 has been tested

for strain sensitivity by agar well diffusion assay using

differently originated B. thuringiensis (six strains) and

B. cereus (eight strains), and it exhibited antimicrobial

activity similar to spiked strains (data not shown). There-

fore, both strains of B-1695 and B-344 could be appropriate

for indicator strains used as representative growth curve.

With regard to the effect of nisin A on spore-forming

bacteria in food products, previous studies reported that

nisin A exerted a bacteriostatic/or bactericidal effect on

B. cereus in flour-based products with 100 IU g�1 (Jenson

et al. 1994), liquid eggs with 200 IU g�1 (Delves et al.

1992), Alicyclobacillus acidoterrestris in acidic drinks with

25–50 IU g�1 (Yamazaki et al. 2000), Bacillus species

(B. pumilus, B. licheniformis, B. cereus and B. thuringien-

sis) in creams with 5–100 IU g�1 (Nissen et al. 2001),

Clostridium botulinum in processed cheese spreads with

500 IU g�1 (Somers and Taylor 1987), Cl. pasteurianum

and Paenibacillus macerans in canned tomatoes with 100–200 IU g�1 (Delves 1990b), and natural contaminant

Bacillus species, Clostridium sporogenes, Clostridium tyro-

butyricum in mashed potatoes with 250 IU g�1 (Thomas

et al. 2002). The activity of nisin A is not uniform and

depends on the chemical composition and physical condi-

tion of foods (Balciunas et al. 2013). High-fat milk pud-

ding requires high levels of nisin A to prevent outgrowth

of spore-forming bacteria by comparison with other liquid

and semi-fluid foods. Therefore, these finding in this

study might provide additional information for control-

ling microbiological quality and extending shelf life of

milk-based high-fat chilled desserts.

Nisin A does not cause objectionable flavours when

added to a variety of foods (Heinemann et al. 1965).

Therefore, nisin A may improve the organoleptic qualities

by allowing less severe pasteurizing conditions without

compromising food safety. Preservation tests of the milk

pudding treated at 100, 110 and 120°C for 2 s at nisin A

concentrations of 40, 80, 120 and 240 IU g�1 elucidate

that food spoilage bacteria in normal ingredients could

be prevented to propagate in the samples with ≥80 and

≥120 IU g�1 nisin A at 5�0 and 7�5% fat milk pudding,



respectively. Bacillus and Brevibacillus species detected

from the samples of control, 40 IU and 80 IU g�1, were

confirmed nisin sensitivity using agar well diffusion assay

(data not shown), suggesting that further levels of nisin A

need to control the natural contaminant heat-resistant

bacterial spores.

By analysis of the aroma compounds using HS-SPME

with a GC/MS system and an organoleptic assessment

with 15 trained panelists, the organoleptic qualities of

mild heat processing at 110 and 120°C ranked signifi-

cantly high for ‘milk taste’, and the overall, and ‘sulphur-

like odour’ decreased compared with 130°C processing.

These results suggested that the combination of nisin A

with reduced thermal treatments (preferred 110–120°C)may diminish the typical ‘cooked flavour’ of dairy ingre-

dients and cause minimal alteration to the sensory prop-

erties of milk components.

Nowadays, consumers are of particularly high interest

to the health concerns regarding food additives (Balciun-

as et al. 2013). The food products without any addition

of chemical preservatives are strongly expected. The use

of nisin A as a natural food preservative satisfies both

consumer and industrial demands for food quality and

food safety.

Acknowledgements

Thanks to Takashi Osada, Takefumi Ichimura and Saori

Yonehata (Meiji Co., Ltd.) for technical assistance in pilot

heat exchanger trial and aroma compounds analysis.

Conflict of Interests

There are no conflict of interests.

References

Anon (1985) Nisin preservation of chilled desserts. Dairy Ind

Int 50, 41–43.

Arakawa, K., Kawaii, Y., Nisimura, J., Kitazawa, H. and Saito, T.

(2009) Negative effect of divalent metal cations on production

of gassericin T, a bacteriocin produced by Lactobacillus gasseri,

in milk-basedmedia. Int Dairy J 19, 612–616.

Balciunas, E.M., Martinez, F.A.C., Franco, S., Franco,

D.T.B.D.M. and Oliveira, A.C.R.P.S. (2013) Novel

biotechnological applications of bacteriocins: a review.

Food Control 32, 134–142.

Beard, B.M., Sheldom, B.W. and Foegeding, P.M. (1999)

Thermal resistance of bacterial spores in milk-based

beverage supplemented with nisin. J Food Prot 62,

484–491.

Bhatti, M., Veeramachaneni, A. and Shelef, L.A. (2004) Factors

affecting the antilisterial effects of nisin in milk. Int J Food

Microbiol 97, 215–219.

Chollet, E., Sebti, I., Martial-Gros, A. and Degraeve, P. (2008)

Nisin preliminary study as a potential preservative for

sliced ripened cheese: NaCl, fat and enzymes influence on

nisin concentration and its antimicrobial activity. Food

Control 19, 982–989.

Chung, K.T., Dickson, J.S. and Crouse, J.D. (1989) Effects of

nisin on growth of bacteria attached to meat. Appl Environ

Microbiol 55, 1329–1333.

Cleveland, J., Montville, T., Nes, I. and Chikindas, M.L. (2001)

Bacteriocins: safe, natural antimicrobials for food

preservation. Int J Food Microbiol 71, 1–20.

Delves, B.J. (1990a) Nisin and its application as a food

preservative. J Soc Dairy Technol 43, 73–76.

Delves, B.J. (1990b) Nisin and its uses as a food preservative.

Food Technol 44, 100–117.

Delves, B.J., Williams, G.C. and Wilkinson, S. (1992) The use

of the bacteriocin, nisin, as a preservative in pasteurized

liquid whole egg. Lett Appl Microbiol 15, 133–136.

FDA/HHS (1998) Direct food substances affirmed as generally

recognized as safe: nisin preparation. Fed Reg 53, 11247–

11251.

Gonzalez, B., Arca, P., Mayo, N. and Suarez, J.E. (1994)

Detection, purification, and partial characterization of

plantaricin C, a bacteriosin produced by a Lactobacillus

plantarum strain of dairy origin. Appl Environ Microbiol

61, 2873–2878.

Hata, T., Tanaka, R. and Ohmono, S. (2010) Isolation and

characterization of plantaricin ASM1: a new bacteriocin

produced by Lactobacillus plantarum A-1. Int J Food


H�echard, Y., Derijard, B., Letellier, G. and Cenatiempo, Y.

(1992) Characterization and purification of mesentericin

Y105, an anti-Listeria bacteriocin from Leuconostoc

mesenteroides. J Gen Microbiol 138, 2725–2731.

Heinemann, B., Voris, L. and Stumbo, C.R. (1965) Use of

nisin in processing food products. Food Technol 19,

160–164.

Henning, S., Metz, R. and Hammes, W.P. (1986) New aspects

for the application of nisin to foods based on its mode of

action. Int J Food Microbiol 3, 135–141.

Hurst, A. and Hoover, D.G. (1993) Nisin. In Antimicrobials in

Foods ed. Davidson, P.M. and Branen, A.L. pp. 369. New

York: M. Dekker.

Jack, R.W., Tagg, J.R. and Ray, B. (1995) Bacteriocin of Gram

positive bacteria. Microbiol Rev 59, 171–200.

Jenson, I., Baird, L. and Delves-Broughton, J. (1994) The use of

nisin as a preservative in crumpets. J Food Prot 57, 874–877.

Jones, L.W. (1974) Effect of butterfat on inhibition of

Staphylococcus aureus by nisin. Can J Microbiol 20, 1257–1260.

Jung, D.S., Bodyfelt, F.W. and Daeschel, M.A. (1992) Influence

of fat and emulsifiers on the efficacy of nisin in inhibiting

Listeria monocytogenes in fluid milk. J Dairy Sci 75, 387–393.

Kataoka, H., Load, H.L. and Pawliszyn, J. (2000) Applications

of solid-phase microfiltration in food analysis.

J Chromatogr A, 880, 35–62.



Klaenhammer, T.R. (1988) Bacteriocins of lactic acid bacteria.

Biochimie 70, 337–349.

Laridi, R., Kheadr, E.E., Benech, R.-O., Vuillemard, J.C.,

Lacroic, C. and Fliss, I. (2003) Liposome encapsulated

nisin Z: optimisation, stability and release during milk

fermentation. Int Dairy J 13, 325–336.

Liu, W. and Hansen, J.N. (1990) Some chemical and physical

properties of nisin, a small-protein antibiotic produced by

Lactococcus lactis. Appl Environ Microbiol 56, 2551–2558.

Martinez Viedma, P., Abriouel, H., Ben Omar, N., Lucas

L�opez, R., Valdivia, E. and G�alvez, A. (2009) Assay of

enterocin AS-48 for inhibition of foodborne pathogens in

desserts. J Food Prot 72, 1654–1659.

Nissen, H., Axelson, L., Holo, H. and Blom, H. (2001)

Characterization and growth of Bacillus spp. in heat-

treated cream with and without nisin. J Appl Microbiol 90,

530–534.

O’Mahony, T., Rekhif, N., Cavadini, C. and Fitzgerald, G.F.

(2001) The application of a fermented food ingredient

containing ‘variacin’, a novel antimicrobial produced by

Kocuria varians, to control the growth of Bacillus cereus in

chilled dairy products. J Appl Microbiol 90, 106–114.

Plockov�a, ́M., Rih�akova, Z. and ́Kova, E. (1998) The efficacyof nisin and nisin producing strain Lactococcus lactis

subsp. lactis in thermisized quarg desserts. Adv Food Sci

20, 17–22.

Rehaiem, A., Mart�ınez, B., Manai, M. and Rodr�ıguez, A.

(2010) Production of enterocin A by Enterococcus faecium

MMRA isolated from ‘Rayeb’, a traditional Tunisian dairy

beverage. J Appl Microbiol 108, 1685–1693.

Schaeffer, P., Miller, J. and Aubert, J. (1965) Catabolic

repression of bacterial sporulation. Proc Natl Acad Sci

USA 54, 704–711.

Sobrino-L�opez, A. and Mart�ın-Belloso, O. (2008) Use of nisin

and other bacteriocins for preservation of dairy products.

Int Dairy J 18, 329–343.

Somers, E.B. and Taylor, S.L. (1987) Antibotulinal

effectiveness of nisin in pasteurized process cheese spreads.

J Food Prot 50, 842–848.

Thomas, L., Clarlkson, M. and Delves-Broughton, J. (2000)

Nisin. In Natural Food Antimicrobial Systems ed. Naidu,

A.S. pp. 463–524. Boca Raton, FL: CRC Press.

Thomas, L.V., Ingram, R.E., Bevis, H.E., Davies, E.A., Milne,

C.F. and Delves, B.J. (2002) Effective use of nisin to

control Bacillus and Clostridium spoilage of a

pasteurized mashed potato product. J Food Prot 65,

1580–1585.

WHO (1969) Specifications for identity and purity of some

antibiotics. WHO/Food Addit 69, 53.

Wirjantoro, T. and Jewis, M.J. (1996) Effect of nisin and high

temperature pasteurization on the shelf-life of whole milk.

Int J Dairy Technol 49, 99–102.

Yamazaki, K., Murakami, M., Kawai, Y., Inoue, N. and

Matsuda, T. (2000) Use of nisin for inhibition of

Alicyclobacillus acidoterrestris in acidic drinks. Food




Documents

Nisin A extends the shelf life of high-fat chilled dairy dessert, a milk-based pudding