485

Journal of Food Protection, Vol. 56, No. 6, Pages 485-488, 492 (June 1993) Copyright©, International Association of Milk, Food and Environmental Sanitarians

Inhibition of Clostridium botulinum Growth and Toxigenesis in a Model Gravy System by Coinoculation With Bacteriocin-Producing Lactic Acid Bacteria

ALLISON D. CRANDALL and THOMAS J. MONTVILLE* Downloaded from http://meridian.allenpress.com/jfp/article-pdf/56/6/485/1659916/0362-028x-56_6_485.pdf by guest on 26 September 2021 Department of Food Science, New Jersey Agricultural Experiment Station, Cook College, Rutgers - The State University, New Brunswick, New Jersey 08903

(Received for publication August 14, 1992)

ABSTRACT naturally in soil (28) and may contaminate raw meat (12). In addition, contamination from spices or other ingredients should The ability of several lactic acid bacteria (LAB) to inhibit not be overlooked (26). As a result, spores of C. botulinum Clostridium botulinum toxigenesis was investigated. Acidification may find their way into minimally processed foods. studies identified the bacteriocinogenic strains Lactococcus lactis ATCC 11454 and Pediococcus pentosaceus ATCC 43200 as the Proteolytic type A and B strains of C. botulinum have a most promising based on their ability to rapidly acidify a model minimum growth temperature of 10°C (27) and form fairly gravy system. These two strains, a third bacteriocinogenic strain heat-resistant spores (20). As a result, spores of proteolytic Lactobacillus plantarum BN, and nonbacteriocinogenic strains as strains might survive a minimal heat processing and outgrow controls were then coinoculated along with C. botulinum type A and produce toxin at temperatures only slightly above normal and B spores into a model gravy system to determine if bacteriocin refrigeration conditions (33). Nonproteolytic type B strains, production and acidification are effective in preventing C. botuli­ the spores of which are considerably less heat resistant num growth and toxin production. Triplicate tubes of gravy-like (20,25), grow and produce toxin at temperatures as low as media containing either 0 or 0.5% glucose were coinoculated with 3.3°C (29,30). Spores of nonproteolytic strains would pose a the LAB at 104 CFU/ml and with the pool of heat-shocked C. botulism hazard even under proper refrigeration temperatures botulinum spores at 102, 104, and 106 CFU/ml and incubated in the event of insufficient heat processing. Notermans et al. anaerobically at 15, 25, or 35°C. The media were monitored for C. botulinum growth, toxin production, and acidity. At 15°C, both the (21) showed that at 8°C nonproteolytic strains can produce bacteriocinogenic and nonbacteriocinogenic strains of L. lactis and toxin within 3 weeks. The same study also demonstrated that L. plantarum prevented toxigenesis in gravy containing glucose at heating minimally processed foods before consumption can­ all C. botulinum inocula levels. The bacteriocinogenic and not be considered a safety factor to reduce botulinum risk. In nonbacteriocinogenic strains of P. pentosaceus prevented toxin addition, the minimal heat treatment delivered to these new production by C. botulinum at 102 and 10" CFU/ml in the presence generation foods, in combination with the anaerobic environ­ of glucose. P. pentosaceus 43200 was the only strain tested show­ ment created by vacuum or modified atmosphere packaging, ing inhibition in the absence of glucose, preventing toxigenesis by 2 may in fact create a greater botulism hazard by destroying C. botulinum at 10 CFU/ml. At 25 and 35°C, none of the lactic and inhibiting competing spoilage microorganisms. As a acid bacteria tested prevented toxigenesis. The results suggest that result, an organoleptically acceptable food may contain botu­ acid production by these strains of LAB may afford some protec­ linal toxin (3). The most important factor in controlling C. tion against mild temperature abuse and that bacteriocin production had little if any additional effect. The biopreservation system was botulinum growth in minimally processed foods is probably ineffective at temperatures of 25 and 35°C. storage of the finished product at temperatures below 5°C (9,17,21,26). However, since temperature abuse is a common Minimally processed refrigerated meats, which are be­ occurrence at both the retail and consumer levels coming increasingly popular with consumers, require special (2,7,14,35,36), the Refrigerated Foods and Microbiological attention to assure their microbiological safety (4,15). These Criteria Committee of the National Food Processors Associa­ foods often receive only a mild heat treatment, contain little tion has recommended that additional safety barriers be or no preservatives, are vacuum or modified atmosphere incorporated into refrigerated foods (24). packaged, rely heavily on refrigeration for preservation, and This laboratory has previously demonstrated the require little or no heat for preparation. However, the result­ antibotulinal activities of Pediococcus pentosaceus ATCC ing "fresh" product with an extended shelf life may also pose 43200, Lactococcus lactis ATCC 11454, Lactobacillus an increased botulism hazard (3). plantarum Lb75, L. plantarum Lb592, and L. plantarum BN Clostridium botulinum, an anaerobic sporeformer, is the (22). In addition, we demonstrated the bacteriocinogenicity of causative agent of botulinal food poisoning which, although L. lactis 11454, L. plantarum BN, and P. pentosaceus 43200 rare, has a high mortality rate. C. botulinum spores occur at refrigeration and abuse temperatures against proteolytic

JOURNAL OF FOOD PROTECTION, VOL. 56, JUNE 1993 486 CRANDALL AND MONTVILLE and nonproteolytic C. botulinum spores (23). In the present determined. Appearance of gas bubbles in the Durham tube was study, we investigated the ability of these strains to grow, monitored as an indicator of outgrowth. Growth was confirmed to produce lactic acid and bacteriocins, and inhibit botulinal be C. botulinum by assaying the supernatant fluid for toxin. Toxin growth and toxin production in a model gravy system. determination was by enzyme-linked immunosorbent assay (ELISA) and confirmed by the mouse bioassay. MATERIALS AND METHODS Enzyme-linked immunosorbent assay (ELISA) for botulinal toxin Bacterial cultures and media Toxin assays were carried out according to the ELISA method Lactococcus lactis ATCC 11454, L. plantarum BN (19), L. of Dezfulian and Bartlett (8) as modified by Wei H. Lee (U.S. plantarum Lb75, L. plantarum Lb592, and P. pentosaceus ATCC Department of Agriculture, Food Safety and Inspection Service, 43200 [also designated FBB61 (6) prior to its deposition with the Beltville, MD). Although this assay utilizes a Granite goat anti- American Type Culture Collection] were used as the bacteriocin- botulinum type A antiserum and a rabbit anti-botulinum type A producing strains. L lactis ATCC 21053, L. plantarum ATCC antiserum (kindly provided by Dr. Wei H. Lee), it is used to detect 8014, and P. pentosaceus 43NP1 [a bar derivative obtained (18) both A and B toxins because of the immunogenicity of the hemag­ by curing strain ATCC 43200 of its 13.6 MDa pediocin-coding glutinin molecules common to both type A and type B toxin plasmid (6)], strains which do not produce bacteriocin, were used complexes (16,31). Sterile ELISA plates (Corning) were coated (100 ulAvell) with 10 ul of the goat anti-botulinum antiserum in 11 as negative controls. Lactic acid bacteria (LAB) stocks were main­ Downloaded from http://meridian.allenpress.com/jfp/article-pdf/56/6/485/1659916/0362-028x-56_6_485.pdf by guest on 26 September 2021 tained at -80°C in Lactobacilli MRS broth (Difco) supplemented ml of coating buffer (0.4 g Na2C03, 0.73 g NaHC03, 0.05 g NaN3, with 20% glycerol. Working cultures were made as slants on MRS 250 ml H20, pH adjusted to 9.6 with 1 N HC1 prior to autoclaving, broth supplemented with 1.5% Bacto-agar (Difco), stored at 4°C, stored refrigerated) and incubated in a humidity container at 45°C and transferred bimonthly for 6 months before a new working overnight. Plates were washed three times with phosphate-Tween culture was made. Spores of C. botulinum proteolytic type A 20 buffer (PBS; 0.4 g KC1, 16 g NaCl, 0.4 g NaN3, 0.4 g KH2P04, strains 17 409A, 25763A, 62A, Clovis-A, and 56A; proteolytic type 2.3 g Na2HP04, 2.0 g Tween 20, 2 L H20, pH 7.4 prior to B strains 213B, 53B, 999B, 7949B, 169B, 642B; and Aphis 4B autoclaving, stored refrigerated) with 1 % fetal bovine serum (FBS, were used as a single pool. The individual C. botulinum strains Sigma) and pounded dry. Gravy supernatants were obtained by were maintained at -80°C as spore suspensions, with the working centrifugation, and 100 ul were added to duplicate wells. Plates pool stored at 4°C. Aliquots were heat shocked for 10 min at 80°C were incubated in a humidity container at 45°C for 1 h and then prior to use. washed three times with PBS-FBS and pounded dry. The rabbit anti-botulinum antiserum was added (2.5 ul) to 11 ml of PBS-FBS and dispensed 100 ul/well. The plates were incubated at 45°C for Gravy formulation 1 h in a humidity container, washed, and pounded dry as before. The model gravy contained 1.8% Trypticase peptone (BBL), Anti-rabbit IgG alkaline phosphatase conjugate from goat (Sigma) 1.2% beef extract (Difco), 0.6% yeast extract (Difco), 0.2% was prepared by adding 20 ul to 11 ml of PBS with 1% horse carregeenan type II (Sigma), and 2% starch (National Starch and serum (Sigma), and 100 ul were dispensed into each well. The Chemical Corporation, Bridgewater, NJ). For some experiments, plates were incubated, washed, and pounded dry as before. The 0.5, 1.0, or 3.0% glucose (Fisher) was added. Medium was auto- substrate was prepared by adding 20 mg p-nitrophenyl phosphate claved for 15 min at 15 psi prior to use. (Sigma) to 11 ml of substrate buffer (1.88 g glycine, 0.02 g MgCL,, 0.035 g ZnCl2, 250 ml H20, pH adjusted to 10.4 with 1 N NaOH Acidification of model gravy system by lactic acid bacteria prior to autoclaving, stored refrigerated) previously warmed to Lactococcus lactis 11454, L plantarum Lb592, L. plantarum room temperature, and then 100 ul were added to each well. The Lb75, P. pentosaceus 43200, and L. plantarum BN were inocu­ 6 plates were incubated in a humidity chamber for 1 h before reading lated at 10 CFU/ml into triplicate 20-ml tubes of model gravy at 405 nm. A net absorbance for each sample was calculated by containing 0, 0.5, 1.0, or 3.0% glucose and incubated at 37°C for subtracting the mean absorbance of empty wells from the mean 24 h. Uninoculated tubes were left as controls. The tubes were absorbance of wells containing gravy supernatant. For screening sampled at selected times to determine LAB populations and pH. purposes, as recommended by the U.S. Department of Agriculture LAB were enumerated by spiral plating (Spiral System, Inc.) on protocol, a net absorbance of 0.2 or higher was considered positive brain heart infusion agar (Difco), incubating at 30°C for 48 h, and for botulinum toxin. then counting using a Bacteria Colony Counter (Model 500A, Spiral System Instruments, Inc.). The pH of approximately 0.2 ml of gravy pipeted onto aluminum foil was determined using a flat- Mouse bioassay for botulinal toxin surface pH electrode (Microcomputer pH-Vision 6071 pH meter, All samples that had ELISA absorbances less than 0.4 (2-fold Markson Science, Inc.). higher than the suggested cutoff value) were also tested for toxin using a modification of the Bacteriological Analytical Manual Challenge of gravy model system Mouse Bioassay method (IT). Six randomly selected samples with Test tubes containing inverted Durham tubes were filled with ELISA absorbances greater than 0.4 were also assayed. Gravy 6 ml of gravy containing either 0 or 0.5% glucose and sealed with supernatants were obtained by centrifugation. For qualitative toxin Hungate style screw caps. All manipulations were carried out in a determination, two 1-cc 26 gauge 3/8 in. tuberculin syringes (Becton flexible anaerobic chamber (Coy Laboratory Products, Inc., Ann and Dickinson) were loaded with 0.5 cc of untreated supernatant. Arbor, MI) with an atmosphere of N2 (85%), FL, (10%), and C02 Then, for each sample a portion of the supernatant was boiled for (5%). Triplicate tubes of gravy were coinoculated (at 10" CFU/ml) 10 min and 0.5 cc loaded into a third syringe. For quantitative toxin with one of the three bacteriocin-producing LAB (L. lactis 11454, determination, the gravy supernatant was serially diluted 1/9, 1/75, P. pentosaceus 43200, or L. plantarum BN) or a nonbacteriocin- 1/650, and 1/7510. Duplicate syringes were loaded with 0.5 cc of producing control (L. lactis 21053, P. pentosaceus 43NP1, or L each dilution. White mice (Techonic Farms) were injected intrap- plantarum 8014) and with a pool of heat-shocked C. botulinum eritoneally and observed for symptoms of botulism poisoning spores (102, 104, and 106 CFU/ml). Aliquots inoculated with LAB (ruffling of fur, labored breathing, weakness of limbs, total paraly­ alone, C. botulinum alone, or left uninoculated served as controls. sis with gasping for breath, and death due to respiratory failure) for Incubation was anaerobic at 15, 25, or 35°C for 44, 8, or 4 d, 72 h. A sample was concluded to be toxin positive if both mice respectively. At the end of each incubation period the gravy pH was injected with untreated supernatant developed symptoms of botu-

JOURNAL OF FOOD PROTECTION, VOL. 56, JUNE 1993 INHIBITION OF C. BOTULINUM IN GRAVY 487 lism poisoning and if the mouse receiving boiled supernatant the bacteriocinogenic and nonbacteriocinogenic strains of L. remained symptomless. A sample was toxin negative if all mice lactis and L. plantarum inhibited C. botulinum toxin produc­ remained symptomless. In situations where the results of the mouse tion at all inoculation levels in gravy that contained glucose. bioassay contradicted those of the ELISA method, the toxin status The bacteriocinogenic and nonbacteriocinogenic strains of P. of the sample was reported for the more sensitive mouse bioassay. pentosaceus were able to prevent toxin production by C. botulinum at 102 and 104 CFU/ml in the presence of glucose. RESULTS All combinations which prevented toxigenesis in the presence Growth and pH results for the acidification of the model of glucose had terminal pH values below 5.0. P. pentosaceus gravy are illustrated in Fig. 1. Increasing the fermentable 43200 was the only strain tested showing inhibition in the absence of glucose, preventing toxigenesis by C. botulinum at carbohydrate above 0.5% had little effect on the terminal pH 2 or on the rate of acidification (data not shown). All strains 10 CFU/ml. The terminal pH of this combination was 5.74. grew to at least 108 CFU/ml within 8 h of incubation at 37 °C Detection of botulinal toxin in the challenge study (Table regardless of glucose level (Fig. 1A, C). In the absence of 1) was by ELISA with the results confirmed by the more glucose, none of the strains dropped the pH below 5.8 (Fig. sensitive mouse bioassay. All of the randomly selected samples with ELISA absorbances greater than 0.4 were shown to IB). In the presence of glucose, L. lactis 11454 was best able Downloaded from http://meridian.allenpress.com/jfp/article-pdf/56/6/485/1659916/0362-028x-56_6_485.pdf by guest on 26 September 2021 to rapidly lower the pH of the system, achieving a pH of 4.7 contain toxin by the mouse bioassay. Of the samples with within 6 h and a terminal pH of 4.5 (Fig. ID). L. plantarum ELISA absorbances less than 0.4, the results of the mouse Lb75 and L. plantarum Lb592 acidified the gravy to a lower assay contradicted those of the ELISA for five out of 40 pH (4.3) than L. lactis 11454, but were unable to do so as samples. In four samples, the ELISA method failed to detect rapidly. P. pentosaceus 43200 acidified to a terminal pH of toxin. The ELISA absorbances for these samples were 0.099, 4.6. L. plantarum BN was the least effective in acidifying the 0.102, 0.127, and 0.135. According to the mouse bioassay, gravy, reaching a terminal pH of 4.9. the sample with an absorbance of 0.135 contained more than 75 but less than 650 mouse lethal doses. In one sample, toxin to io was concluded to be present based on an ELISA absorbance of 0.204 when in fact no toxin was present.

DISCUSSION

Lactobacillus lactis 11454 and P. pentosaceus 43200 were chosen for use in the coinoculation study based on their ability to rapidly acidify a gravy system and on their ability to produce bacteriocins under refrigeration temperatures (23). L. plantarum BN, a strain originally isolated from meat (79), was also selected because, in addition to its ability to produce bacteriocin under refrigeration (23), it might have a competi­ tive advantage in a meat system. For toxin determination, a combination of the ELISA method and a modified mouse bioassay were used. The ELISA method functioned to screen out obviously positive samples and thus allowed us to drastically reduce the number of mice needed for this study from 780 to 135. The modified mouse assay was then used to confirm negative samples and those samples whose ELISA absorbances where close to the

12 16 20 24 0 4 8 12 16 20 cutoff absorbance value. All of the samples for which the Time (hours) Time (hours) ELISA method failed to accurately indicate toxin status had Figure 1. Growth of lactic acid bacteria and pH values in a modelELIS A absorbances close to the cutoff absorbance of 0.2. The gravy system containing no glucose (panels A and B) or 0.5% samples for which the ELISA gave false negatives probably glucose (panels C and D) incubated at 37°C for 24 h. contained amounts of toxin at the lower limit of detection by the ELISA method. The sample for which toxin was deter­ C In the challenge study at 25 and 35 C, at least one tube mined quantitatively had an ELISA absorbance of 0.135 and in each triplicate for the coinoculated combinations produced contained only between 75 and 650 mouse lethal doses. gas within 3 d (data not shown). None of the coinoculated These results are consistent with those of Huhtanen et al. (13) combinations delayed C. botulinum outgrowth longer than who found, depending on the strain of C. botulinum, the the C. botulinum alone controls. All of the coinoculated mouse bioassay to be one to 32 times more sensitive than the combinations at 25 and 35 °C contained botulinal toxin (Table ELISA in detecting botulinum toxin. 1). At 15°C, we were unable to use gas as an indicator of At the severely abusive temperatures of 25 and 35°C, outgrowth. Gas did not appear in many tubes, including those none of the LAB tested were able to prevent C. botulinum shown to contain botulinal toxin. This was probably a result growth and toxin production at the tested inoculation levels. 2 4 6 of higher gas solubility and higher gravy viscosity at the At 15°C, inhibition of C. botulinum at 10 , 10 , and 10 lower temperature. In the coinoculated combinations, both CFU/ml by L. lactis and L. plantarum and inhibition of C.

JOURNAL OF FOOD PROTECTION, VOL. 56, JUNE 1993 488 CRANDALL AND MONTVILLE

TABLE 1. Presence or absence of toxin and final pH of gravy coinoculated with lactic acid bacteria and C. botulinum. Toxin determination was by ELISA and confirmed by mouse bioassay. Lactic acid bacteria alone and uninoculated controls were all toxin negative. C. botulinum alone controls were all toxin positive (data not shown).

Incubation temperature \5°C 25°C 35°C C botulinum inoculum (CFU/ml) KF 10* W UP 104 10* 10* W 10" Lactic acid bacteria % Strain Glucose Toxin pH Toxin pH Toxin pH Toxin pH Toxin pH Toxin pH Toxin pH Toxin pH Toxin pH

L. lactis 11454toc+ 0 + 6.02 + 5.91 + 6.28 + 6.35 + 7.34 + 7.39 + 5.94 + ND + 6.99 0.5 - 4.48 - 4.54 - 4.71 + 4.92 + 4.94 + 4.93 + 5.08 + 4.92 + 5.07 L. lactis 21053toc- 0 + 5.80 + 5.78 + 5.75 + 6.44 + 6.37 + 6.40 + 6.95 + 7.05 + 6.95 0.5 - 4.55 - 4.65 - 4.70 + 4.87 + 4.75 + 5.03 + 4.85 + 5.68 + 5.80 L. plantarum BNbac+ 0 + 64(3 + 6/34 + 6^59 + 6^22 + 6.15 + 6/78 + 5^ + 6^59 + 6^86 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/56/6/485/1659916/0362-028x-56_6_485.pdf by guest on 26 September 2021 0.5 - 4.78 - 4.87 - 4.85 + 5.08 + 5.29 + 5.32 + 5.71 + 6.44 + 6.19 L. plantarum %0\A^: 0 + 648 + 5/78 + 5^82 + 6^50 + 6.47 + 645 + 6^58 + 7/14 + 7J8" 0.5 - 4.60 - 4.70 - 4.57 + 4.58 + 4.67 + 4.82 + 5.27 + 5.75 + 6.16 pVPentosaceus 43200bac+ 0 - 5?74 + 5^90 + 5^81 + 6.32 ND ND + 6S7 + 61® + 6^95 + 1W 0.5 - 4.81 - 4.82 + 4.75 + 4.86 + 4.98 + 4.96 + 5.49 + 5.18 + 5.39 P. Pentosaceus 43NP7b"~0 + 6/09 + 5^93 + 6T6 + 6^52 + 6.44 + 6^58 + 740 + 7(36 + 7J36 0.5 - 4.92 - 4.94 + 4.89 + 4.97 + 5.14 + 5.04 + 5.55 + 5.63 + 6.56

bac+ Bacteriocin producer. bac" Bacteriocin nonproducer. ND - Not done.

botulinum at 102 and 104 CFU/ml by P. pentosaceus in the including food composition, type and level of carbohydrate, presence of glucose were caused solely by acidification as product buffering capacity, incubation temperature, inocula­ evidenced by the inhibitory action of both bacteriocinogenic tion level, and lactic strain bacteriocinogenicity all interact and nonbacteriocinogenic strains and the lack of inhibition in and can influence the success of the system. Berry et al. (7) gravy which did not contain a fermentable carbohydrate. demonstrated that Listeria monocytogenes inhibition by Under these conditions, P. pentosaceus was, however, unable Pediococcus acidilactici was dependent upon LAB concen­ 6 to inhibit C. botulinum at 10 CFU/ml. tration, temperature, and packaging atmosphere. Our findings All of the coinoculated combinations incubated at 15°C further suggest that the safety of minimally processed refrig­ with pH values below 5.0 were nontoxic. However, at 25 and erated meats depends on multiple barriers to inhibit patho­ 35°C, all of the combinations, including those with pH values gens. Any such preservation system must undergo extensive below 5.0, were toxic. This may be due to the greater inoculated pack trials to validate its efficacy. effectiveness of acid against C. botulinum at lower tempera­ ACKNOWLEDGMENTS tures. Also, at the lower temperature, the LAB may grow and drop the pH to an inhibitory level before the C. botulinum This is manuscript D-10564-2-92 of the New Jersey Agricul­ spores outgrow. At the higher temperatures, the C. botulinum tural Experiment Station. This research was supported by state may outgrow and begin toxin production before the LAB appropriations, U.S. Hatch Act funds, and a grant from the drop the pH. The observation that all of the nontoxic combi­ Cattlemen's Beef Promotion and Research Board which is admin­ nations in the presence of glucose had terminal pH levels istered in cooperation with the Beef Industry Council. We thank Dr. below 5.0 is consistent with the observations of Tanaka et al. Amechi Okereke, Karen Winkowski, and Yaroslaw Hrywna for their assistance and helpful discussions. (34). The only case where bacteriocin production appeared to REFERENCES play a role was with P. pentosaceus 43200 in the absence of 1. Berry, E. D„ R. W. Hutkins, and R. W. Mandigo. 1991. The use of glucose against C. botulinum at 102 CFU/ml. However, since bacteriocin-producing Pediococcus acidilacitci to control postprocessing Listeria monocytogenes contamination of frankfurters. this combination had a lower pH (5.74) than the correspond­ J. Food Prot. 54:681-686. ing combination with P. pentosaceus 43NP1 (pH 6.09), the 2. Bryan, F. L., L. A. Seabolt, R. W. Peterson, and L. M. Roberts. 1978. effect cannot be conclusively attributed to bacteriocin produc­ Time-temperature observations of food and equipment in airline tion. catering operations. J. Food Prot. 41:80-92. 3. Conner, D. E., V. N. Scott, and D. T. Bernard. 1989. Potential Biopreservation systems in which LAB inhibit the growth Clostridium botulinum hazards associated with extended shelf-life of pathogens such as C. botulinum are extremely complex. refrigerated foods: a review. J. Food Safety 10:131-153. Commercial applications of LAB and/or their bacteriocins as 4. Corlett, Jr., D. A. 1989. Refrigerated foods and use of hazard analysis antibotulinal agents include Wisconsin processed bacon (34) and critical control point principles. Food Technol. 43(2):91-94. 5. Daeschel, M. A. 1989. Antimicrobial substances from lactic acid and nisin in pasteurized cheese spreads (10,32). A combina­ bacteria for use as food preservatives. Food Technol. 43(1): 164-167. tion of Pediococcus acidilactici and dextrose can prevent 6. Daeschel, M., and T. D. Klaenhammer. 1985. Association of a 13.6- botulinal toxigenesis in chicken salad (14). Numerous factors cont. on p. 492

JOURNAL OF FOOD PROTECTION. VOL. 56, JUNE 1993 492 KNUDTSON AND HARTMAN processing. J. Food Prot. 56:6-9. gism in Streptococcus faecium and Streptococcus faecalis. J. Infect. 9. Langlois, B. E., K. A. Dawson. I. Leak, and D. K. Aaron. 1988. Dis. 140:203-208. Antimicrobial resistance of fecal coliforms from pigs in a herd not 12. Muirhead, S. (ed.). 1992. 1993 Feed additive compendium. Miller exposed to antimicrobial agents for 126 months. Vet. Microbiol. Publishing Co., Minnetonka, MN. 18:147-153. 13. Murray, B. E. 1990. The life and times of the Enterococcus. Clin. 10. Louie, M., A. E. Simor, S. Szeto, M. Patel, B. Kreiswirth, and D. E. Microbiol. Rev. 3:46-65. Low. 1992. Susceptibility testing of clinical isolates of Enterococcus 14. Neu, H. C. 1992. The crisis in antibiotic resistance. Science 257:1064- faecium and Enterococcus faecalis. J. Clin. Microbiol. 30:41-45. 1072. 11. Moellering, R. C, Jr., O. M. Korzeniowski, M. A. Sande, and C. B. 15. Rouff, K. L. 1990. Recent taxonomic changes in the genus Entero­ Wennersten. 1979. Species-specific resistance to antimicrobial syner- coccus. Eur. J. Clin. Microbiol. Infect. Dis. 9:75-79.

Crandall and Montville, cont. from p. 488

megadalton plasmid in Pediococcus pentosaceus with bacteriocin 21. Notermans, S., J. Dufrenne, and B. M. Lund. 1990. Botulism risk of activity. Appl. Environ. Microbiol. 50:1538-1541. refrigerated, processed foods of extended durability. J. Food Prot. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/56/6/485/1659916/0362-028x-56_6_485.pdf by guest on 26 September 2021 7. Daniels, R. W. 1991. Applying HACCP to new-generation refriger­ 53:1020-1024. ated foods at retail and beyond. Food Technol. 45(6): 122, 124. 22. Okereke, A., and T. I. Montville. 1991. Bacteriocin inhibition of 8. Dezfulian, M., and J. G. Bartlett. 1984. Detection of Clostridium Clostridium botulinum spores by lactic acid bacteria. J. Food Prot. botulimtm type A toxin by enzyme-linked immunosorbent assay with 54:349-353. antibodies produced in immunologically tolerant animals. J. Clinical 23. Okereke, A., and T. ]. Montville. 1991. Bacteriocin-mediated inhibi­ Microbiol. 19:645-648. tion of Clostridium botulinum spores by lactic acid bacteria at 9. Eklund, M. W. 1982. Significance of Clostridium botulinum in refrigeration and abuse temperatures. Appl. Environ. Microbiol. fishery products preserved short of sterilization. Food Technol. 36:107- 57:3423-3428. 112, 115. 24. Refrigerated Foods and Microbiological Criteria Committee of the 10. Federal Register. 1988. Nisin preparation: Affirmation of GRAS National Food Processors Association. 1988. Factors to be considered status as a direct human food ingredient. 21CFR Part 184. 53:11247- in establishing good manufacturing practices for the production of 11251. refrigerated foods. Dairy Food Sanit. 8:288-291. 11. Food and Drug Administration. 1984. The bacteriological analytical 25. Scott, V. N„ and D. T. Bernard. 1982. Heat resistance of spores of manual, 6th ed. Association of Official Analytical Chemists, Arling­ non-proteolytic type B Clostridium botulinum. J. Food Prot. 45:909- ton, VA. 912. 12. Greenberg, R. A., R. B. Tompkin, B. O. Bladel, R. S. Kittaka, and A. 26. Simunovic, J., J. L. Oblinger, and J. P. Adams. 1985. Potential for Anellis. 1966. Incidence of mesophilic Clostridium spores in raw growth of nonproteolytic types of Clostridium botulinum in pasteur­ pork, beef, and chicken in processing plants in the United States and ized restructured meat products: a review. J. Food Prot. 48:265-276. Canada. Appl. Microbiol. 14:789-793. 27. Smelt, J. P. P. M., and H. Haas. 1978. Behaviour of proteolytic 13. Huhtanen, C. N„ R. C. Whiting, A. J. Miller, and J. E. Call. 1992. Clostridium botulinum types A and B near the lower temperature Qualitative correlation of the mouse neurotoxin and enzyme-linked limits of growth. Eur. ]. Appl. Microbiol. Biotechnol. 5:143-154. immunoassay for determining Clostridium botulinum types A and B 28. Smith, G. R., and A. M. Young. 1980. Clostridium botulinum in toxins. J. Food Safety 12:119-127. British soil. J. Hyg. 85:271-274. 14. Hutton, M. T., P. A. Chehak, and J. H. Hanlin. 1991. Inhibition of 29. Solomon, H. M., D. A. Kautter, and R. K. Lynt. 1982. Effect of low botulinum toxin production by Pediococcus acidilactici in tempera­ temperatures on growth of nonproteolytic Clostridium botulinum ture abused refrigerated foods. J. Food Safety 11:255-267. types B and F and proteolytic type G in crabmeat and broth. J. Food 15. Klaenhammer, T. R. 1988. Bacteriocins of lactic acid bacteria. Prot. 45:516-518. Biochemie. 10:337-349. 30. Solomon, H. M„ R. K. Lynt, T. Lilly, Jr., and D. A. Kautter. 1977. 16. Lamanna, C, and J. P. Lowentahl. 1951. The lack of identity between Effect of low temperatures on growth of Clostridium botulinum hemagglutinin and the toxin of type A botulinal organism. J. Bacteriol. spores in meat of the blue crab. J. Food Prot. 40:5-7. 61:751-752. 31. Somers, E., and B. R. DasGupta. 1991. Clostridium botulinum types 17. Lerke, P., and L. Farber. 1971. Heat pasteurization of crab and A, B, C , and E produce proteins with or without hemagglutinating shrimp from the pacific coast of the United States: public health activity: do they share common amino acid sequences and gene? J. aspects. J. Food Sci. 36:277-279, Prot. Chem. 10:415-425. 18. Lewus, C. B. 1991. Characterization of bacteriocins produced by 32. Somers, E. B., and S. L. Taylor. 1987. Antibotulinal effectiveness of lactic acid bacteria isolated from meat. Ph. D. dissertation. Rutgers, nisin in pasteurized process cheese spreads. J. Food Prot. 50:842-848. the State University. New Brunswick, NJ. 33. Sperber, W. H. 1982. Requirements of Clostridium botulinum for 19. Lewus, C. B„ A. Kaiser, and T. J. Montville. 1991. Inhibition of growth and toxin production. Food Technol. 36(12):89-94. food-borne bacterial pathogens by bacteriocins from lactic acid bac­ 34. Tanaka, N., E. Traisman. M. H. Lee, R. G. Cassens, and E. M. Foster. teria isolated from meat. Appl. and Environ. Microbiol. 57:1683- 1980. Inhibition of botulinum toxin formation in bacon by acid 1688. development. J. Food Prot. 43:450-457. 20. Lynt, R. K., D. A. Kautter, and H. M. Solomon. 1982. Differences 35. Van Garde, S. J., and M. J. Woodbum. 1987. Food discard practices and similarities among proteolytic and nonproteolytic strains of of householders. J. Am. Dietet. Assoc. 87:322-329. Clostridium botulinum types A, B, E and F: a review. J. Food Prot. 36. Wyatt, L. D. and V. Guy. 1980. Relationships of microbial quality or 45:466-474. retail meat samples and sanitary conditions. J. Food Prot. 43:385-389.

JOURNAL OF FOOD PROTECTION, VOL. 56, JUNE 1993