APPLIED AND ENVIRONMENTAL MICROBIOLOGY, OCt. 1992, p. 3355-33590099-2240/92/103355-05$02.00/0Copyright ©) 1992, American Society for Microbiology

Vol. 58, No. 10

Novel Method To Extract Large Amounts of Bacteriocinsfrom Lactic Acid Bacteriat

RONGGUANG YANG, MONTY C. JOHNSON, AND BIBEK RAY*

Food Microbiology Laboratory, Animal Science Department,University of Wyoming, Laramie, Wyoming 82071

Received 4 May 1992/Accepted 21 July 1992

Antimicrobial peptides, bacteriocins, produced by lactic acid bacteria were adsorbed on the cells ofproducing strains and other gram-positive bacteria. pH was a crucial factor in determining the degree ofadsorption of these peptides onto cell surfaces. In general, between 93 and 100% of the bacteriocin moleculeswere adsorbed at pHs near 6.0, and the lowest (c5%) adsorption took place at pH 1.5 to 2.0. On the basis ofthis property, a novel isolation method was developed for bacteriocins from four genera of lactic acid bacteria.By using this method we made preparations of pediocin AcH, nisin, sakacin A, and leuconocin Lcml that werepotent and concentrated. This method produced a higher yield than isolation procedures, which rely onprecipitation of the bacteriocins from the cell-free culture liquor. It is simple and can be used to produce largequantities of bacteriocins from lactic acid bacteria to be used as food biopreservatives.

In the search for a food biopreservative, investigations oncertain antibacterial proteins (bacteriocins) from lactic acidbacteria have been very popular (11, 18, 25). To study theirchemical and antibacterial properties and to determine theireffectiveness in food systems, it is necessary to obtainrelatively large quantities of these peptides in a pure andconcentrated form. Currently, most methods rely on ammo-nium sulfate precipitation of the bacteriocins from cell-freeculture liquor. This method has been used to obtain bacte-riocins from Pediococcus spp. (4-6, 13), Lactobacillus spp.(17, 21, 23), Lactococcus spp. (12, 15, 16, 24), and Leu-conostoc spp. (10, 14). However, there is general agreementthat it does not yield a good product because many otherproteins from the medium can also be precipitated and theyield is not very high (5, 9, 12). For further purification ofprecipitated bacteriocins, especially in the determination ofthe amino acid composition and sequence, these researchershave used various column chromatography techniques.Commercial nisin preparations are available in highly

purified food-grade form. However, the methods used com-mercially are not known. Mattick and Hirsch (20) used acombination of acid treatment of the culture followed byremoval of the cells and then solvent extraction and precip-itation to obtain nisin with high potency. Later, otherworkers used several different methods based on propanol-NaCl or butanol-acetic acid extraction from culture super-natant (2, 9) and breaking of cells and extraction with acid (1,31). Although the nisin preparations had high potency, themethods were laborious and total yields were low.Bhunia et al. (7), while studying the mode of bactericidal

action of pediocin AcH from Pediococcus acidilactici H onsensitive cells of Lactobacillus plantarum NCDO 955, ob-served that adsorption of pediocin AcH to the cells is pHdependent. Maximum adsorption occurred at pH 6.0 to 6.5,but when the pH of the cell suspension was reduced to 2.0 orbelow, the pediocin AcH was not adsorbed. It is known that,in general, cells of a bacteriocin-producing bacterial strain

* Corresponding author.t Wyoming Agricultural Experiment Station journal article num-

ber 1687.

adsorb the bacteriocin molecules that they produce (7, 13,18, 25, 30). We hypothesized that if, after fermentation, theculture broth of a producer strain was adjusted to the pH formaximum adsorption of the bacteriocin onto the cell sur-faces, the cells with adsorbed bacteriocin molecules couldeasily be removed from the culture liquor by centrifugation.The peptide could then be selectively released from the cellsat pH 1.5 to 2.0, and the preparation could provide largequantities of purified bacteriocin. This hypothesis was testedwith bacteriocin-producing strains from four genera of lacticacid bacteria and the results are presented here.

MATERIALS AND METHODS

Bacterial cultures and media. The bacterial strains used inthis study are described in Table 1. Pediococcus acidilacticiLB 42-923 (27), used to produce pediocin AcH, was grown at370C for 18 h in TGE broth (8). Lactococcus lactis subsp.lactis ATCC 11454, used to produce nisin, was grown at30°C for 24 h in TGE buffered broth, which consisted of TGEbroth plus 0.5% sodium citrate, 0.1% sodium acetate, and0.05% dipotassium phosphate (pH 6.5). Leuconostoc carno-sum Lml, used to produce leuconocin Lcml, and Lactoba-cillus sake Lb 706, used to produce sakacin A (29), weregrown at 30°C for 24 h in MRS broth (Difco). All indicatorbacteria were grown in TGE broth and transferred daily.

Antimicrobial activity assay of bacteriocin. Each bacterio-cin preparation was serially diluted (1:10, 1:20, 1:30, etc.),and 5 ,ul from each dilution was spotted onto a lawn ofappropriate indicator bacteria. Activity units (AU) per mil-liliter were calculated as previously reported (7, 8). Theindicator bacteria were Lactobacillus plantarum NCDO 955for pediocin AcH and nisin, Leuconostoc mesenteroides Lyfor sakacin A, and Enterococcus faecalis MB1 for leucono-cin Lcml.

Preparation of cells for bacteriocin adsorption. A desiredbacterial strain (Fig. 1) was grown overnight in the appro-priate medium, and the cells were collected by centrifuga-tion, washed in 1/5 volume of sterile 5 mM sodium phosphate(pH 6), resuspended in 1/5 vol of sterile 1.0 M NaCl solution(pH 2.0, adjusted with 5% phosphoric acid), and mixed at4°C for 1 h. The cells were harvested by centrifugation,

3355

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 23

Dec

embe

r 20

21 b

y 38

.13.

23.1

34.

APPL. ENVIRON. MICROBIOL.

TABLE 1. Bacterial strains, bacteriocin sensitivity, and growth conditions

Bacteriocin Bacteriocin Growth GrowthStrain produced sensitivity medium conditions Source (reference)

P. acidilactici LB 42-923 Pediocin AcH Nisin TGE 37°C, 20 h Our laboratory (26)L. lactis subsp. lactis ATCC 11454 Nisin Nonea TGE 30°C, 24 h ATCCbL. sake Lb 706 Sakacin A Nisin MRS 25°C, 24 h F. K. Lucke (28)L. carnosum Lml Leuconocin Lcml Pediocin AcH, nisin MRS 25°C, 24 h Our laboratorycL. plantarum NCDO 955 Noned Pediocin AcH, nisin TGE 30°C, 18 h M. A. Daeschel (5-8)L. mesenteroides Ly None Pediocin AcH, nisin, TGE 30°C, 18 h Our laboratory (7)

sakacin AE. faecalis MB1 None Pediocin AcH, nisin, TGE 30°C, 18 h Our laboratorye

sakacin A, leuco-nocin Lcml

a Not sensitive to any of the bacteriocins tested.b ATCC, American Type Culture Collection, Rockville, Md.c Isolated from processed meat products in our laboratory.d Nonbacteriocin producer strains were used as indicators.e Originally from the Microbiology Department, University of Wyoming.

washed once with sterile deionized water (dH2O), and resus-pended in sterile dH2O to 20 times the original culturevolume.

Influence ofpH on bacteriocin adsorption by producing cellsand indicator cells. A cell suspension of a desired bacterialstrain (0.1 ml, about 107 cells) and a bacteriocin preparation(0.1 ml, about 105 AU) were added to 1.8 ml of 5 mM sodiumphosphate adjusted to a pH range from 1.5 to 10.0 (withphosphoric acid and NaOH). This mixture was incubated at4°C for 30 min and centrifuged at 17,000 x g for 5 min, andthe AU of the bacteriocin in the supernatant was assayed (7,8). Control I consisted of 0.1 ml of dH2O instead of bacte-riocin, and control II had 0.1 ml of dH2O instead of cellsuspension. The percentage of bacteriocin adsorbed wascalculated as follows: percentage of bacteriocin adsorbed =100 x [1 - (AU/ml in cell-free supernatant - AU/ml incontrol I)]/(AU/ml in control II).

Extraction of adsorbed bacteriocin from producer cells. Forextraction of pediocin AcH, nisin, leuconocin Lcml, andsakacin A, each producer strain was grown to early station-ary phase (about 5 x 108 cells per ml) in 1 liter of theappropriate medium. The culture broths were adjusted to pH6.5 for pediocin AcH, nisin, and sakacin A and pH 5.5 forleuconocin Lcml. Each culture broth was heated to 70°C for25 min to kill the cells (heating can be done before pHadjustment also), and the cells were harvested by centrifu-gation at 15,000 x g for 15 min. Supernatant activity wasassayed to calculate the amount of bacteriocin not adsorbedonto the cells. After the cells had been washed with 5 mMsodium phosphate (pH 6.5), they were resuspended in 50 mlof 100 mM NaCl at pH 2.0 (adjusted with 5% phosphoricacid) for pediocin AcH, leuconocin Lcml, and sakacin Aand pH 2.5 for nisin and mixed with a magnetic stirrer for 1h at 4°C. Cell suspensions were then centrifuged at 29,000 xg for 20 min, and the cells were resuspended in 5 mM sodiumphosphate (pH 6.5) and assayed to calculate the amount ofbacteriocin lost with the cells. The supematants were dia-lyzed in 1,000-molecular-weight-cutoff dialysis bags againstdH20 at 4°C for 24 h and then freeze-dried. The total drymatter, protein concentration (19), and bacteriocin activityof each product were determined. This process was per-formed twice for each bacteriocin.SDS-PAGE and identification of the activity band. Freeze-

dried extraction preparations were analyzed by sodiumdodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Methods previously reported (4, 5) were used with

a Tris-tricine system (28). This method used a 10-cm layer of16.5% acrylamide resolving gel and a 2-cm layer of 10%acrylamide spacer gel. The electrophoresis was run at 15°Cand 20 mA for the first 2 h and then at 30 mA for another 12h. Sample preparation, application of samples, and geldivision after electrophoresis have been described before(4). One-half of each gel was stained with Coomassie brilliantblue, and the other half was used for identification of thebacteriocin band by using growth inhibition of E. faecalisMB1 (4). For growth inhibition, the gel half was washed indeionized water for 2 h at 25°C, placed on a TGE agarprepoured plate, and overlaid with soft agar seeded with E.faecalis MB1. The plate was incubated at 30°C for 24 h,examined for the location of the zone of growth inhibition,and photographed. The stained gel half was destained andphotographed.

RESULTS

Sensitivity of bacterial strains to bacteriocins. The fourbacteriocin-producing strains and the three indicator strainswere tested for sensitivity to four bacteriocins (Table 1).Their sensitivities varied greatly: Lactococcus lactis subsp.lactis ATCC 11454 was not sensitive to any bacteriocin,whereas E. faecalis MB1 was sensitive to all four. Thesedata allowed us to select a specific strain for measuring theAU of each bacteriocin preparation and for measuring, inseparate experiments, the adsorption of each bacteriocin toa sensitive strain and the producer strain (Fig. 1).

Effect ofpH on adsorption of bacteriocin onto producer andindicator cells. The influence of pH on the adsorption ofpediocin AcH, nisin, sakacin A, and leuconocin Lcml ontotheir respective producer and indicator strains is presentedin Fig. 1. Adsorption of all four bacteriocins onto cells wasstrongly influenced by the pH of the suspending environ-ment. Pediocin AcH was adsorbed 100% at pH 6.0 to 6.5,while at pH below 1.5 it was not adsorbed to either P.acidilactici LB 42-923 or Lactobacillus plantarum NCDO955. Maximum adsorption of nisin to both producer andindicator bacteria occurred at pH 6.5, and complete loss ofadsorption was found at pH 3.0 and below. Maximumsakacin A adsorption occurred at pH 5.5 and above on boththe producer and indicator strains. At pH 2.0 no sakacin Awas adsorbed to the producer cells but about 75% remainedadsorbed onto E. faecalis MB1. Leuconocin Lcml had asharp adsorption curve on producer cells, with a maximum

3356 YANG ET AL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 23

Dec

embe

r 20

21 b

y 38

.13.

23.1

34.

BACTERIOCIN EXTRACTION FROM LACTIC ACID BACTERIA 3357

Extraction and purification of bacteriocin. On the basis ofthe influence of pH on the adsorption and release of eachbacteriocin, an extraction method for bacteriocin was devel-oped. The recovery and loss of each bacteriocin at differentstages of purification and in the final dried product arepresented in Table 2. The loss of bacteriocin in the supema-tant following harvesting of the cells ranged from 0 to 7.7%,and loss with the cells after extraction ranged from 0.6 to2.3%. Recovery in dried preparations ranged from 44.3% forsakacin A to 106.7% for pediocin AcH. The potency (AU pergram) of bacteriocins ranged from 1.8 x 10 to 5.0 x 10 ona dry-matter basis and from 1.1 x 108 to 4.9 x 108 on aprotein content basis (19). Although the loss of sakacin A insupernatant and cells was very low, the total yield in thedried preparation was only 44%. It was not clear why therewas such a discrepancy. The yields of dry matter and proteinfor both Lactobacillus sake Lb 706 and Leuconostoc camo-sum Lml were higher than for P. acidilactici LB 42-923 andLactococcus lactis subsp lactis ATCC 11454.SDS-PAGE analysis of the preparations. A Coomassie

blue-stained SDS-PAGE gel revealed a single sharp bandfrom both the pediocin AcH and nisin preparations, espe-cially when they were tested within 2 weeks of preparation(Fig. 2). After storage for 2 weeks or more, pediocin AcHand nisin preparations produced more than one band. How-ever, the sakacin A and leuconocin Lcml preparationsproduced several bands, some of which were diffused. Acomparison of the bands in the stained gel with the gel halfshowing zones of inhibition indicated that the lowest bandfor each preparation contained the active bacteriocin (Fig. 2and 3). Nisin, sakacin A, and leuconocin Lcml bands werevery close to 2.5 kDa, while pediocin AcH was just abovethe 2.5-kDa standards. Also, leuconocin Lcml gave twoactivity bands, one just below the other, and the pediocinAcH preparation showed a streak of slight activity above themain zone (Fig. 3).

% OVu X

A

d 60a0

r 40

b

d 20

1 2 3 4 5 6 7 8 9 10

pHFIG. 1. Adsorption of bacteriocins onto producing and indicator

bacteria. (A) Pediocin AcH adsorption on P. acidilactici LB42-923(*) and Lactobacillus plantarum NCDO 955 (E). (B) Nisin adsorp-tion on Lactococcus lactis subsp. lactis ATCC 11454 (*) andLactobacillusplantarum NCDO 955 (E). (C) Sakacin A adsorptionon Lactobacillus sake Lb706 (*) and E. faecalis MB1 (0). (D)Leuconocin Lcml adsorption on Leuconostoc carnosum Lml (*)

and Leuconostoc mesenteroides Ly (0).

at pH 5.5 and a minimum at pH 3.0 and below (acid range).Adsorption of leuconocin Lcml on Leuconostoc mesenteroi-des Ly was relatively low, about 50% between pH 4.5 and7.0, and the lowest was at 4.0 or below in the acid range. Inthe alkaline range the adsorption varied from 0% at pH 8.0for leuconocin Lcml to almost 100% at pH 10 for sakacin A.

DISCUSSION

Adsorption of bacteriocin molecules by the Pediococcus(7, 13), Lactobacillus (18), Lactococcus (16), and Leuconos-toc (10, 14) producer strains, as well as by other sensitiveand resistant gram-positive bacteria, has been reported.Also, the pH of optimum adsorption for nisin (16) andpediocin AcH (7) by gram-positive bacteria has been re-ported. Bhunia et al. (7) recently reported the conditions foradsorption of pediocin AcH onto sensitive Lactobacillusplantarum NCDO 955. This report was the basis for thepresent study on the influence of pHs between 1 and 10 onthe level of adsorption of four bacteriocins from four generaof lactic acid bacteria onto the respective producer strainsand lactic acid bacterium strains sensitive to a particularbacteriocin. For this purpose Lactobacillus plantarumNCDO 955 was used for pediocin AcH and nisin, Leuconos-toc mesenteroides Ly was used for sakacin A, and E.faecalis MB1 was used for leuconocin Lcml (Table 1; Fig.1).Methods for bacteriocin isolation that involved removal of

cells from culture broth, following adjusting the pH to 6.0(13), and/or precipitation of cell-free supernatant with(NH4)2SO4 will recover only a portion of the total bacterio-cin present in the culture broth, because the portion ad-sorbed on the cells is lost. Since adsorption of bacteriocins isnot affected by heating the cells, we used heat-killed cells toavoid any loss of activity by proteolytic enzymes of the cells(7, 8). Although the maximum and minimum adsorption with

10

Ad80rbed

Ada0rbed

Ad

0rbed

VOL. 58, 1992

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 23

Dec

embe

r 20

21 b

y 38

.13.

23.1

34.

APPL. ENvIRON. MICROBIOL.

TABLE 2. Recovery of bacteriocins at different stages of extraction and purification

AUa (%) of:Bacteriocin recovered

Pediocin AcH Nisin Sakacin A Leuconocin Lcml

Culture broth (1 liter)a 3.0 x 107 (100%) 3.0 x 107 (100%) 7.0 x 107 (100%) 2.6 x 107 (100%)Lost in broth supernatanta 8.0 x 105 (2.7%) 2.0 x 105 (0.7%) 0.0 (0.0%) 2.0 x 106 (7.7%)Lost with cells after extraction' 6.0 x 105 (2.0%) 2.0 x 105 (0.7%) 4.0 x 105 (0.6%) 6.0 x 105 (2.3%)In total dried preparationb 3.2 x 107 (106.7%) 2.8 x 107 (93.3%) 3.1 x 107 (44.3%) 2.5 x 107 (96.2%)

AU/g of dry mattet 5.0 x 107 5.0 x 107 3.1 x 107 1.8 x 107AU/g of proteinc 4.1 x 108 4.9 x 108 1.5 x 108 1.1 X 108

a Total AU in relation to 1-liter culture broths and supernatants and cells from the broths. Percentages are based on AU in culture broths as 100%.b AU in freeze-dried products was determined by resuspending the dry material in 50 mM SDS for better dispersion and prevention of clumping of bacteriocin

molecules.c After freeze-drying the values for total dry matter and total protein content, respectively, were as follows: pediocin AcH, 640 and 78 mg; nisin, 560 and 57.5

mg; sakacin A, 995 and 206 mg; leuconocin Lcml, 1,400 and 230 mg. The results are the average of two extractions.

respect to pH varied slightly among the bacteriocins, a singleprotocol could usually be used, namely pH 6.0 for highadsorption onto the cells and pH 2.0 for maximum releasefrom the cells. Extraction at alkaline pH was not usedbecause it is known that nisin (16), pediocin AcH (7), andsome other bacteriocins of lactic acid bacteria are inacti-vated at alkaline pHs. Phosphoric acid was used to adjust themixtures to the desired pH; other acids could also probablybe used. We used 100mM NaCl during extraction to preventthe bacteriocin molecules from clumping and thus beingremoved with the cells during centrifugation.The total recoveries of pediocin AcH, nisin, and leucono-

cin Lcml in the dried preparations were relatively high. Forpediocin AcH the recovery was over 100%. This may be aresult of the dilution method used for the assay of AU.Because only fixed dilutions are used, there may be differ-ences in the end point, which may introduce some error.However, for all three bacteriocins the recovery was over90%. For sakacin A the recovery was only 44.0%, eventhough losses in supernatant and cells were very low (Table2). The reason for this discrepancy is now being investi-gated. The dried preparations have very high bacteriocinactivity based on either dry-matter or protein concentration.Dry-matter and protein concentrations of both sakacin A andleuconocin Lcml were relatively high compared with thoseof pediocin AcH and nisin. Thus, it is possible that Lacto-

FIG. 2. SDS-PAGE gel half, stained with Coomassie blue. (A)Lanes: 1, Mr standards (a, Mr 16,900; b, Mr 14,400; c, Mr 8,200; d,Mr 6,200; e, Mr, 2,500); 2, pediocin AcH (old preparation); 3, nisin;4, sakacin A; 5, leuconocin Lcml. (B) Lanes: 1, pediocin AcH (freshpreparation); 2, Mr standards. Arrows show respective bacteriocinbands.

bacillus sake Lb 706 and Leuconostoc camnosum Lml havesome other proteins that were extracted along with thebacteriocins. Some Lactobacillus strains have been shownto possess easily extracted surface proteins (3, 26).The method described here produced dry preparations of

bacteriocins that were not only highly potent but, for pedi-ocin AcH and nisin, also pure (Fig. 2). In a separate studythe pediocin AcH band from the gel was transferred to amembrane by transblot and the membrane was used forpartial sequencing of amino acids (22). The formation ofmore than one band by stored preparations of pediocin AcHis now being studied. Both sakacin A and leuconocin Lcmlgave several bands with higher Mr than the bacteriocin.These bands are probably surface or cell wall proteins; thisis currently being investigated. In gel overlays there was asingle inhibition zone for each bacteriocin, except for leu-conocin Lcml, which gave two zones, suggesting that bothmonomer and dimer forms of leuconocin Lcml were present(Fig. 3).The actual Mr of the bacteriocins as determined from the

amino acid sequences and the Mr as estimated on the basis oftheir migration in relation to standards in the gel could bequite different. Nisin, with a known Mr of 3,500 (16), formsa band in the same line as the Mr standard of 2,500. PediocinAcH, with a known Mr of 4,600 (22), forms a band slightlyabove 2,500.

FIG. 3. SDS-PAGE gel half, showing the zone of growth inhibi-tion ofE. faecalis MB1 by the bacteriocin bands. Lanes: A, pediocinAcH (old preparation, showing a slight streak of activity above themain zone); B, nisin; C, sakacin A; D, leuconocin Lcml. Thesezones of inhibition correspond to the respective bacteriocin bands inFig. 1A. For further explanation, see the text.

3358 YANG ET AL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 23

Dec

embe

r 20

21 b

y 38

.13.

23.1

34.

BACTERIOCIN EXTRACTION FROM LACTIC ACID BACTERIA 3359

The method described here for the extraction of bacterio-cins from lactic acid bacterial strains produces products withhigh potency and in concentrated form. The total loss isquite low. This method could be an economical procedure toproduce large quantities of bacteriocins from lactic acidbacteria for use as food biopreservatives. The extractionprocedure described here can also be used for the extractionof proteins from biological systems in which adsorption toand release from specific receptors are pH dependent.

ACKNOWLEDGMENTS

This investigation was partially funded by the National Live Stockand Meat Board, Binational Agricultural Research Development(with Israel), and Wyoming Experimental Station funds.

REFERENCES1. Bailey, F. J., and A. Hurst. 1971. Preparation of a highly active

form of nisin from Streptococcus lactis. Can. J. Microbiol.17:61-67.

2. Berridge, N. J. 1949. Preparation of the antibiotic nisin. Bio-chem. J. 45:486-493.

3. Bhowmik, T., M. C. Johnson, and B. Ray. 1985. Isolation andpartial characterization of the surface layer proteins of Lacto-bacillus acidophilus. Int. J. Food Microbiol. 2:311-321.

4. Bhunia, A. K., M. C. Johnson, and B. Ray. 1987. Directdetection of an antimicrobial peptide ofPediococcus acidilacticiin sodium dodecyl sulfate-polyacrylamide gel electrophoresis, J.Ind. Microbiol. 2:319-322.

5. Bhunia, A. K., M. C. Johnson, and B. Ray. 1988. Purification,characterization and antimicrobial spectrum of a bacteriocinproduced by Pediococcus acidilactici H. J. Appl. Bacteriol.65:261-268.

6. Bhunia, A. K., M. C. Johnson, B. Ray, and E. L. Belden. 1990.Antigenic property of pediocin AcH produced by Pediococcusacidilactici H. J. Appl. Bacteriol. 69:211-215.

7. Bhunia, A. K., M. C. Johnson, B. Ray, and N. Kalchayanand.1991. Mode of action of pediocin AcH from Pediococcusacidilactici H on sensitive bacteria strains. J. Appl. Bacteriol.70:25-30.

8. Biswas, S. R., P. Ray, M. C. Johnson, and B. Ray. 1991.Influence of growth conditions on the production of a bacterio-cin, pediocin AcH, by Pediococcus acidilactici H. Appl. Envi-ron. Microbiol. 57:1265-1267.

9. Cheeseman, G. C., and N. J. Berridge. 1968. Observation onmolecular weight and chemical composition of nisin A. Bio-chem. J. 71:185-195.

10. Daba, H., S. Panadian, J. J. Gosselin, R. E. Simard, J. Huang,and C. Laeroix. 1991. Detection and activity of a bacteriocinproduced by Leuconostoc mesenteroides. Appl. Environ. Mi-crobiol. 57:3450-3455.

11. Daeschel, M. A. 1990. Application of bacteriocins in foodsystems, p. 91-104. In D. D. Bills and S. D. Kung (ed.),Biotechnology and food safety. Butterworth-Heinemann, Bos-ton.

12. Davey, G. P., and B. C. Richardson. 1981. Purification and someproperties of diplococcin from Streptococcus cremoris 346.Appl. Environ. Microbiol. 41:84-89.

13. Gonzalez, C. F., and B. S. Kunka. 1987. Plasmid-associatedbacteriocin production and sucrose fermentation in Pediococ-cus acidilactici. Appl. Environ. Microbiol. 53:2534-2538.

14. Hastings, J. W., M. Sailer, K. Johnson, K. L. Ray, J. C.Vederas, and M. E. Stiles. 1991. Characterization of leuconocinA-UAL 187 and cloning of the bacteriocin gene from Leuconos-toc gelidium. J. Bacteriol. 173:7491-7500.

15. Holo, H., 0. Nilssen, and I. F. Nes. 1991. Lactococcin A, a newbacteriocin from Lactococcus lactis subsp. cremoris: isolationand characterization of the protein and its gene. J. Bacteriol.173:3879-3887.

16. Hurst, A. 1981. Nisin. Adv. Appl. Microbiol. 27:85-123.17. Joerger, M. C., and T. R. Klaenhammer. 1986. Characterization

and purification of helveticin J and evidence for a chromoso-mally determined bacteriocin produced by Lactobacillus helve-ticus 481. J. Bacteriol. 167:439-446.

18. Klaenhammer, T. R. 1988. Bacteriocin from lactic acid bacteria.Biochimie 70:337-349.

19. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall.1951. Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193:265-275.

20. Mattick, A. T. R., and A. Hirsch. 1947. Further observations onan inhibitory substance (nisin) from lactic streptococci. Lancetii:5-8.

21. Mortvedt, C. I., J. Nissen-Meyer, K. Slettern, and I. F. Nes.1991. Purification and amino acid sequence of lactacin S, abacteriocin produced by Lactobacillus sake L45. Appl. Envi-ron. Microbiol. 57:1829-1834.

22. Motlagh, A. M., A. K. Bhunia, F. Szostek, T. R. Hansen, M. G.Johnson, and B. Ray. Nucleotide and amino acid sequence ofpap-gene (pediocin AcH production) in Pediococcus acidilacticiH. Lett. Appl. Microbiol., in press.

23. Muriana, P. M., and T. R. Klaenhammer. 1991. Purification andpartial characterization of lactacin F, a bacteriocin produced byLactobacillus acidophilus 11088. Appl. Environ. Microbiol.57:114-121.

24. Piard, J. C., P. M. Muniana, M. J. Desmazeand, and T. R.Klaenhammer. 1992. Purification and partial characterization oflacticin 481, a lanthionine-containing bacteriocin produced byLactococcus lactis subsp. lactis CNRZ 481. Appl. Environ.Microbiol. 58:249-284.

25. Ray, B. 1992. Bacteriocins of starter culture bacteria as bio-preservatives: an overview, p. 177-205. In B. Ray and M. A.Daeschel (ed.), Food biopreservatives of microbial origin. CRCPress, Inc., Boca Raton, Fla.

26. Ray, B., and M. C. Johnson. 1986. Freeze-drying injury ofsurface layer protein and its protection in Lactobacillus acido-philus. Cryo Lett. 7:210-217.

27. Ray, S. R., W. J. Kim, M. C. Johnson, and B. Ray. 1989.Conjugal transfer of a plasmid encoding bacteriocin productionand immunity in Pediococcus acidilactici H. J. Appl. Bacteriol.66:393-399.

28. Schagger, H., and G. von Jagow. 1987. Tricine-sodium dodecylsulfate-polyacrylamide gel electrophoresis for the separation ofproteins in the range from 1 to 100 kDa. Anal. Biochem.166:368-379.

29. Schillinger, U., and F.-K. Lucke. 1989. Antibacterial activity ofLactobacillus sake isolated from meat. Appl. Environ. Micro-biol. 55:1901-1906.

30. Tagg, H. G., A. S. Dajani, and L. W. Wannamaker. 1976.Bacteriocin of gram-positive bacteria. Bacteriol. Rev. 40:722-756.

31. White, R. J., and A. Hurst. 1968. The location of nisin in theproducer organism Streptococcus lactis. J. Gen. Microbiol.53:171-179.

VOL. 58, 1992

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/a

em o

n 23

Dec

embe

r 20

21 b

y 38

.13.

23.1

34.