A numerical taxonomic study of lactic acid bacteria isolated from irradiated pork and chicken packaged under various gas atmospheres

Ninety-four Gram-positive, catalase-negative bacteria, isolated from pork and chicken that had been packed in modified atmospheres and irradiated to 1·75 and 2·5 kGy respectively, were studied. The majority of the strains were Lactobacillus saké . Numerical taxonomy, with the group average clustering strategy, revealed the existence of six clusters at the 85% similarity level. The largest, Cluster 1, contained 78 (83%) of the test strains along with three Lact. saké strains. Cluster 2 contained three test strains and the type strains of Carnobacterium piscicola and Carn. divergens . Cluster 3 contained two chicken strains and Lact. curvatus . Cluster 4 contained a pork strain and Leuconostoc dextranicum . Clusters 5 and 6 contained four pork and two chicken strains respectively and no type strains. Four test strains remained unclustered as did the other reference strains included in the study.     

A numerical taxonomic study of lactic acid bacteria from vacuum-packed beef, pork, lamb and bacon

S haw B. G. & H arding C harmaigne D. 1984. A numerical taxonomic study of lactic acid bacteria from vacuum packed beef, pork, lamb and bacon. Journal of Applied Bacteriology 56 , 25–40.
A numerical taxonomic study using 79 unit characters has been performed on 100 isolates of lactic acid bacteria from refrigerated vacuum-packed beef, pork, lamb and bacon. Three clusters were observed at 78% S which contained all the strains apart from three unidentifiable streptobacteria, one Leuconostoc , and one strain of Pediococcus pentosaceus . One cluster (III) consisted of only one strain of Leuc. paramesenteroides and six unidentifiable Leuconostoc strains. The two largest clusters (I and II) were both composed entirely of streptobacteria. Cluster I contained 31 strains (G + C content 33–2–36–9 moles %) which were not identifiable with any described species. Cluster II contained 57 strains (G + C content 40–7–43–7 moles %) which were provisionally identified with Lactobacillus sake or Lact. bavaricus according to the lactic acid isomer produced. The division of nearly all the streptobacteria into two clearly defined clusters has resolved problems which have existed in the classification of lactic acid bacteria from vacuum-packed meat.     

Genetic (RAPD-PCR) and technological diversities among wild Lactobacillus helveticus strains

Diversity in 25 Lactobacillus helveticus strains isolated from natural whey cultures for Argentinian hard cheese production was studied by means of RAPD-PCR patterns and technological parameters (acidifying and proteolytic activities, salt tolerance, diacetyl, H₂O₂ and slime production, phage sensitivity). In the RAPD diversity study, 10 Lact. helveticus strains from the CNRZ collection were also included.
The clustering of RAPD patterns from the Argentinian strains revealed the existence of two Lact. helveticus biotypes. Cluster 1 contained 22 strains (15 wild and seven CNRZ collection strains), Cluster 2 grouped 10 wild strains and Cluster 3 contained only three CNRZ collection strains. RAPD groups could be related to specific cheese-making characteristics (cheese plants). Analysis of technological characteristics in the Argentinian strains showed the occurrence of different natural variants. According to their capacity for growing in milk, they were classified as 'fast', 'intermediate' and 'slow' variants. Among the strains, low salt tolerance and widespread phage resistance were demonstrated. The genetic diversity (RAPD-PCR clustering) did not show any clear relationship with phenotypical diversity. Study of genetic and technological diversity allows a better characterization of wild strains belonging to Lact. helveticus .     

Antibiotic-resistant strains of Enterococcus isolated from Swedish and Danish retailed chicken and pork

Two hundred and seventy-seven Enterococcus isolates representing the predominating enterococcal flora of retailed chicken and pork were identified by phenotypical features, and by random amplified polymorphic DNA. The resistance to nine different antibiotics was determined. Enterococcus faecium was the most frequently occurring species in both Swedish and Danish chicken. Enterococcus faecalis and unidentified groups of Enterococcus dominated in pork. Seventy-three per cent of the Enterococcus isolates from Swedish chicken were resistant to one or more of the tested antibiotics. The corresponding values for Swedish pork, Danish chicken and Danish pork were 9%, 55% and 14%, respectively. Tetracycline resistance was most frequent in isolates from Danish pork and Swedish chicken, while erythromycin resistance was most frequent in Danish chicken. Strains with acquired vancomycin resistance, mainly Ent. faecium , were found only on Danish chicken, except for one isolate from Danish pork. All vancomycin-resistant isolates contained the van A gene. Vancomycin resistance could be transferred to a vancomycin-sensitive Ent. faecium strain from four of nine tested donor strains.     

Genetic diversity in Swiss cheese starter cultures assessed by pulsed field gel electrophoresis and arbitrarily primed PCR

AIMS: To assess intraspecific genetic heterogeneity among commercial Swiss cheese starter culture strains of Lactobacillus helveticus, Streptococcus thermophilus and Propionibacterium freudenreichii and to compare the efficacy of two genetic typing methods. METHODS AND RESULTS: Two genetic typing methods, pulsed field gel electrophoresis (PFGE) and arbitrarily primed PCR (AP-PCR), were used. Nine Strep. thermophilus strains revealed eight PFGE and five AP-PCR genotypes. Seventeen Lactobacillus strains yielded 16 and five genotypes by PFGE and AP-PCR, respectively. Eleven Propionibacterium strains yielded 10 PFGE genotypes. Cluster analysis of PFGE profiles generated similarity coefficients for Strep. thermophilus, Lact. helveticus and Prop. freudenreichii strains of 29.5%, 60.3%, and 30.5%, respectively. Milk acidification rates for Strep. thermophilus and Lact. helveticus were determined. CONCLUSIONS: Pulsed field gel electrophoresis is more discriminatory than AP-PCR. The Lact. helveticus group is more homogeneous than the other species examined. Strains with > 87% similarity by PFGE consistently had the same acidification rate and AP-PCR profile. SIGNIFICANCE AND IMPACT OF THE STUDY: Bacterial strains sold for Swiss cheese manufacture in the United States are genetically diverse. Clustering of genetically related bacteria may be useful in identifying new strains with industrially relevant traits.     

Exclusion of vanA, vanB and vanC type glycopeptide resistance in strains of Lactobacillus reuteri and Lactobacillus rhamnosus used as probiotics by polymerase chain reaction and hybridization methods

Strains of Lactobacillus reuteri and Lact. rhamnosus are used as probiotics in man and animal. The aim of this study was to determine whether the glycopeptide resistance in these lactobacilli has a similar genetic basis as in enterococci. Five Lact. reuteri strains and one Lact. rhamnosus, as well as four Enterococcus control strains, were probed for the vanA gene cluster, the vanB gene and the vanC gene by PCR and Southern hybridization, and DNA/DNA hybridization. Their resistance and plasmid patterns were also investigated. All Lactobacillus strains were resistant to vancomycin but susceptible to a broad range of antibiotics. Four of the Lactobacillus strains (including the Lact. rhamnosus strain) did not harbour any plasmid and two of them contained five and 6 plasmid bands respectively. None of the Lactobacillus strains possessed the vanA, vanB or vanC gene. These findings indicate that the glycopeptide resistance of the Lactobacillus strains analysed is different from the enterococcal type. The study provides reassurance on the safety of the Lactobacillus strains used as probiotics with regard to their vancomycin resistance.     

Characterization and species recognition of Lactobacillus plantarum strains by restriction fragment length polymorphism (RFLP) of the 16S rRNA gene

Twenty-one strains, labelled Lactobacillus plantarum or Lact. plantarum -like, and isolated from different natural sources, were characterized by restriction fragment length polymorphism (RFLP) of the 16S rRNA gene using Hin dIII and Eco RI cleaved chromosomal DNA, together with Lact. plantarum ATCC 14917^T, Lact. pentosus ATCC 8041^T, Lact. plantarum ATCC 10776 and Lact. plantarum ATCC 8014. The fermentation patterns on API 50CH were recorded at 30°C and 37°C for all strains. The phenotypes were heterogeneous, and the ability to ferment 17 of the 49 carbohydrates varied. The fermentation of some carbohydrates, for example D-raffinose and D-arabitol, was temperature-dependent. Strains having identical API profiles were separated by the plasmid profile. All strains but one (affiliated to Lact. casei ) had identical 16S ribosomal DNA sequences ( Lact. plantarum/Lact. pentosus ). The RFLP study resulted in identical ribopatterns for 17 of the strains, including the type strain of Lact. plantarum (pattern A1). Four strains had related fragment patterns to that of Lact. plantarum sensu stricto; three of these strains had more than 60% DNA: DNA homology to the type strain of Lact. plantarum , and one had less than 50% DNA: DNA homology to Lact. plantarum ATCC 14917^T. Two strains had fragment patterns similar to the type strain of Lact. pentosus , and they had more than 80% DNA: DNA homology to Lact. pentosus ATCC 8041^T. One of the Lact. pentosus strains shared one band with the A1 pattern. The ribopatterns of Lact. plantarum were homogeneous (identical for 85% of the strains), irrespective of phenotype and source of isolation. RFLP of the 16S rRNA genes using Eco RI and Hin dIII might be used for species recognition of Lact. plantarum , but seems less suitable for strain typing.     

Cryogenic gamma irradiation of prototype pork and chicken and antagonistic effect between Clostridium botulinum types A and B.

Inoculated, irradiated pork (2,300 cans) and chicken (2,000 cans) pack studies were performed to establish the 12D dose for these foods. Each can was inoculated with a mixture of 10(6) spores of each of 10 strains of Clostridium botulinum (five type A and five type B), or a total of 10(7) spores. The cans received a series of increasing doses of gamma rays (60Co) at -30 +/- 10 degrees C; they were incubated for 6 months at 30 +/- 2 degrees C and examined for swelling, toxicity, and recoverable botulinal cells. The highest rate of swelling for both foods occurred within the first week of incubation, and maximum swelling was observed within 4 to 5 weeks. The minimal experimental sterilizing dose (ESD) based on flat, nontoxic sterile cans was 3.0 less than ESD less than or equal to 3.2 Mrad for pork and 4.0 less than ESD less than or equal to 4.2 Mrad for chicken. An analysis of the partial spoilage data by extreme-value statistics indicated with 90% confidence that the rate of spore death in the two foods was not a normal distribution, but appeared to favor a shifted exponential function. Based on the latter distribution, and assuming one most resistant strain in the mixture of 10 used, the 12D dose computed to 4.37 Mrad, with a shoulder of 0.11 Mrad, for pork and to 4.27 Mrad, with a shoulder of 0.51 Mrad, for chicken. An assumption that there were two or more most resistant strains in the inoculum progressively lowered the 12D dose. There was an apparent antagonism between the irradiated type A and B viable strains in the two foods. Cans with type B cells and toxin predominated over cans with type A cells and toxin, but cans with a mixture of type A and B toxins predominated over cans with a mixture of Type A and B cells. At the highest sublethal doses, only type A cells survived in pork, but in chicken there was a least one type B strain that was at least as resistant as type A strains.     

Cryogenic gamma irradiation of prototype pork and chicken and antagonistic effect between Clostridium botulinum types A and B.

Inoculated, irradiated pork (2,300 cans) and chicken (2,000 cans) pack studies were performed to establish the 12D dose for these foods. Each can was inoculated with a mixture of 10(6) spores of each of 10 strains of Clostridium botulinum (five type A and five type B), or a total of 10(7) spores. The cans received a series of increasing doses of gamma rays (60Co) at -30 +/- 10 degrees C; they were incubated for 6 months at 30 +/- 2 degrees C and examined for swelling, toxicity, and recoverable botulinal cells. The highest rate of swelling for both foods occurred within the first week of incubation, and maximum swelling was observed within 4 to 5 weeks. The minimal experimental sterilizing dose (ESD) based on flat, nontoxic sterile cans was 3.0 less than ESD less than or equal to 3.2 Mrad for pork and 4.0 less than ESD less than or equal to 4.2 Mrad for chicken. An analysis of the partial spoilage data by extreme-value statistics indicated with 90% confidence that the rate of spore death in the two foods was not a normal distribution, but appeared to favor a shifted exponential function. Based on the latter distribution, and assuming one most resistant strain in the mixture of 10 used, the 12D dose computed to 4.37 Mrad, with a shoulder of 0.11 Mrad, for pork and to 4.27 Mrad, with a shoulder of 0.51 Mrad, for chicken. An assumption that there were two or more most resistant strains in the inoculum progressively lowered the 12D dose. There was an apparent antagonism between the irradiated type A and B viable strains in the two foods. Cans with type B cells and toxin predominated over cans with type A cells and toxin, but cans with a mixture of type A and B toxins predominated over cans with a mixture of Type A and B cells. At the highest sublethal doses, only type A cells survived in pork, but in chicken there was a least one type B strain that was at least as resistant as type A strains.     

The presence of prtP proteinase gene in natural isolate Lactobacillus plantarum BGSJ3–18

Aims:  The study of proteolytic activity and examination of proteinase gene region organization in proteolytically active Lactobacillus plantarum strains from different natural sources.
Methods and Results:  A set of 37 lactobacilli was distinguished by using multiplex PCR assay. Results showed that 34 strains were Lact. plantarum and three of them were Lact. paraplantarum . The examination of proteolytic activity revealed that 28 Lact.   plantarum and two Lact.   paraplantarum hydrolyse β-casein. Further analyses of all proteolytically active Lact. plantarum with primers specific for different types of CEPs demonstrated that strain BGSJ3–18 has prtP catalytic domain as well as prtP – prtM intergenic region showing more than 95% sequence identity with the same regions present in Lact. paracasei , Lact. casei and L. lactis . No presence of prtB , prtH or prtR proteinase genes was detected in any of tested Lact. plantarum strains.
Conclusions:  One out of 28 analysed Lact. plantarum strains harbours the prtP -like gene. The other proteolytically active Lact. plantarum probably possesses a different type of extracellular proteinase(s).
Significance and Impact of the Study:  It is the first report about the presence of the prtP –like gene in Lact. plantarum , which illustrates the mobility of this gene and its presence in different species.     

Inhibition of Listeria monocytogenes in chicken cold cuts by addition of sakacin P and sakacin P-producing Lactobacillus sakei

AIMS: To evaluate the potential of sakacin P and sakacin P-producing Lactobacillus sakei for the inhibition of growth of Listeria monocytogenes in chicken cold cuts, by answering the following questions. (i) Is sakacin P actually produced in food? (ii) Is sakacin P produced in situ responsible for the inhibiting effect? (iii) How stable is sakacin P in food? METHODS AND RESULTS: Listeria monocytogenes, a Lact. sakei strain and/or the bacteriocin sakacin P were added to chicken cold cuts, vacuum packed and incubated at 4 or 10 degrees C for 4 weeks. Each of two isogenic Lact. sakei strains, one producing sakacin P and the other not, had an inhibiting effect on the growth of L. monocytogenes. The effect of these two isogenic strains on the growth of L. monocytogenes was indistinguishable, even though sakacin P was produced in the product by one of the two Lact. sakei strains. The addition of purified sakacin P had an inhibiting effect on the growth of L. monocytogenes. A high dosage of sakacin P (3.5 microg x g(-1)) had a bacteriostatic effect throughout the storage period of 4 weeks, while a low dosage (12 ng x g(-1)) permitted initial growth, but at a slow rate. After 4 weeks of storage, the number of L. monocytogenes in the samples with a low dosage of sakacin P was 2 logs below that in the untreated control. When using a high dosage of sakacin P, the bacteriocin was detected in samples stored for up to 6 weeks. CONCLUSIONS: (i) Sakacin P is produced by a Lact. sakei strain when growing on vacuum-packed chicken cold cuts. (ii) Inhibiting effects of Lact. sakei, other than sakacin P, are active in inhibiting the growth of L. monocytogenes growing on chicken cold cuts. (iii) Sakacin P is stable on chicken cold cuts over a period of 4 weeks. SIGNIFICANCE AND IMPACT OF THE STUDY: Both sakacin P and Lact. sakei were found to have potential for use in the control of L. monocytogenes in chicken cold cuts.     

Characterization of lactobacilli towards their use as probiotic adjuncts in poultry

AIMS: A total of 112 strains of lactic acid bacteria of duck origin were studied for their use as a probiotic feed supplement. METHODS AND RESULTS: In vitro studies included aggregation, co-aggregation, cell surface hydrophobicity and adhesion activities on poultry crop cells and human Hep2-cells. Additionally, growth with bile acids (chicken bile, ox gall and taurocholic acid) and tolerance to acidic pH were tested. Among all the isolates, two strains (Lactobacillus animalis TMW 1.972 and Lactobacillus salivarius TMW 1.992) were selected for a survival test in poultry. Monitoring and differentiation of these strains was achieved by selective detection as rifampicin and erythromycin double-resistant mutants. After a single feed administration, both micro-organisms were shown to persist in the crop and caecum of ducks for a period of 18 and 22 days, respectively. For identification of Lact. animalis and Lact. salivarius, two specific PCRs targeted against 16S rDNA were developed. CONCLUSIONS: Within the autochtoneous microflora of ducks, two strains of lactobacilli exhibited strong potential as probiotic adjuncts. The results indicate that the natural gut microflora of poultry serves as an excellent source for optimal strains. SIGNIFICANCE AND IMPACT OF THE STUDY: A general strategy for the selection of probiotic strains is presented. The suggested sequence of tests allows identification of the most promising candidates within complex ecosystems or large strain collections with minimal expenditure.     

Characterization and DNA homology of Lactobacillus strains isolated from pig intestine

Twenty strains of the genus Lactobacillus were isolated from pig intestines and characterized with phenotypic tests. Eleven of these strains, together with selected reference strains, were subjected to DNA homology studies. Three major groups could be distinguished by biochemical/physiological tests and by DNA homology studies: one homofermentative group of the subgenus Thermobacterium and two heterofermentative groups. The DNA homology studies revealed that strains from the homofermentative group were related to Lactobacillus acidophilus strain VPI 1754, but showed a low relationship to the type strain of Lact. acidophilus. One One group of the heterofermentative strains was related to the type strain of Lact. reuteri; the other group, consisting of three strains, showed a low relationship to all reference strains used. These three strains had the unusual property of producing succinic acid in large amounts.     

Characterization and DNA homology of Lactobacillus strains isolated from pig intestine

Twenty strains of the genus Lactobacillus were isolated from pig intestines and characterized with phenotypic tests. Eleven of these strains, together with selected reference strains, were subjected to DNA homology studies. Three major groups could be distinguished by biochemical/physiological tests and by DNA homology studies: one homofermentative group of the subgenus Thermobacterium and two heterofermentative groups. The DNA homology studies revealed that strains from the homofermentative group were related to Lactobacillus acidophilus strain VPI 1754, but showed a low relationship to the type strain of Lact. acidophilus . One group of the heterofermentative strains was related to the type strain of Lact. reuteri ; the other group, consisting of three strains, showed a low relationship to all reference strains used. These three strains had the unusual property of producing succinic acid in large amounts.     

Selection of lactobacilli for chicken probiotic adjuncts

During inhibitory activity screening of 296 strains of lactic acid bacteria from the gastro-intestinal tract of chicks, 77 strains showed inhibition against enteric indicator strains (Salmonella enteritidis and Escherichia coli). Eight different strains identified as Lactobacillus salivarius were selected for the following attributes: their ability to inhibit all the indicator strains; a high adhesion efficiency to the epithelial cells of chickens and also their resistance to a number of antibiotics, monensin, bile salts and pH 3.0. The inhibitory action was not affected by the addition of catalase and no inhibition was detected after neutralizing the supernatant culture fluid. The competitiveness of the most promising strains, Lact. salivarius CTC2183 and CTC2197, was assessed in chicken feed mixture and in vivo. It was concluded that both strains were capable of becoming predominant over the indigenous flora in the incubated chicken feed mixture. In vivo tests showed that Lact. salivarius CTC2197 was able to colonize and overcome Lact. salivarius CTC2183 and the indigenous flora in the crop and caecum of the inoculated chicks.     

Selection of lactobacilli for chicken probiotic adjuncts

During inhibitory activity screening of 296 strains of lactic acid bacteria from the gastro-intestinal tract of chicks, 77 strains showed inhibition against enteric indicator strains ( Salmonella enteritidis and Escherichia coli ). Eight different strains identified as Lactobacillus salivarius were selected for the following attributes: their ability to inhibit all the indicator strains; a high adhesion efficiency to the epithelial cells of chickens and also their resistance to a number of antibiotics, monensin, bile salts and pH 3·0. The inhibitory action was not affected by the addition of catalase and no inhibition was detected after neutralizing the supernatant culture fluid. The competitiveness of the most promising strains, Lact. salivarius CTC2183 and CTC2197, was assessed in chicken feed mixture and in vivo . It was concluded that both strains were capable of becoming predominant over the indigenous flora in the incubated chicken feed mixture. In vivo tests showed that Lact. salivarius CTC2197 was able to colonize and overcome Lact. salivarius CTC2183 and the indigenous flora in the crop and caecum of the inoculated chicks.     

Evaluation of numerical analyses of RAPD and API 50 CH patterns to differentiate Lactobacillus plantarum, Lact. fermentum, Lact. rhamnosus, Lact. sake, Lact. parabuchneri, Lact. gallinarum, Lact. casei, Weissella minor and related taxa isolated from kocho and tef

AIM: The study was carried out to assess the agreement of API 50 CH fermentation data of food lactobacilli with their RAPD profiles to determine whether the system could be used alone as a reliable taxonomic tool for this genus. METHODS AND RESULTS: API 50 CH, RAPD and DNA:DNA reassociation data for 42 lactobacilli from tef and kocho were compared with 30 type strains. Discrepancies were observed between the three methods in assigning strains of Lactobacillus plantarum, Lact. fermentum, Weissella minor and Lact. gallinarum, and Lact. fermentum, Lact. amylophilus, Lact. casei subsp. pseudoplantarum and Lact. rhamnosus. DNA reassociation data agreed well with RAPD results. CONCLUSIONS: API 50 CH profiles should be complemented with molecular genetic results for effective identification in Lactobacillus. SIGNIFICANCE AND IMPACT OF THE STUDY: The study suggested less dependability of metabolic data alone as an identification tool.     

Numerical analysis of fatty acid profiles in the identification of staphylococci

Representative strains of coagulase-positive and coagulase-negative staphylococci were degraded by acid methanolysis and the resultant fatty acid methyl esters analysed by gas chromatography. The quantitative data obtained were examined by cluster analysis. The coagulase-positive strains formed six major and one single-member cluster at the 90% S-level. The Staphylococcus intermedius aggregate cluster included the single-member cluster and major clusters 1 and 2. The four remaining clusters contained S. aureus strains and were homogeneous and distinct. The coagulase-negative strains were recovered in ten major and three single-member clusters at the 90% S-level. Five of the ten major clusters were reasonably homogeneous with respect to the existing classification. Thus, three S. capitis strains and five of the six S. epidermidis strains, two of the three S. hominis strains and five of the six S. simulans strains were recovered in separate clusters. Cluster 7 was divided into two subclusters; one contained five of the six S. hyicus strains and the other contained the two representatives of S. lentus. The remaining clusters were heterogeneous with regard to the named strains they contained.     

A numerical taxonomic study of non-motile non-fermentative gram-negative bacteria from foods

A numerical taxonomic study using 155 unit characters was performed on 63 strains of Gram-negative non-motile non-fermentative bacteria isolated from proteinaceous foods. Similar bacteria from other sources and several Pseudomonas strains from meat were included for reference purposes. Three clusters were observed at 76% S which contained all the food strains. Cluster 1 was composed entirely of Acinetobacter strains including 17 isolated from foods that were provisionally identified with Acinetobacter johnsonii. Cluster 2 contained 22 strains identified as Psychrobacter immobilis, of which 20 were from food. Cluster 3 contained all the Pseudomonas reference strains and 26 non-motile strains isolated from meat. These were shown to be non-motile variants of Pseudomonas fragi. A simple identification scheme, based on five tests, is presented for the distinction of the three types of bacteria.     

A numerical taxonomic study of non-motile non-fermentative Gram-negative bacteria from foods

A numerical taxonomic study using 155 unit characters was performed on 63 strains of Gram-negative non-motile non-fermentative bacteria isolated from proteinaceous foods. Similar bacteria from other sources and several Pseudomonas strains from meat were included for reference purposes. Three clusters were observed at 76% S which contained all the food strains. Cluster 1 was composed entirely of Acinetobacter strains including 17 isolated from foods that were provisionally identified with Acinetobacter johnsonii. Cluster 2 contained 22 strains identified as Psychrobacter immobilis , of which 20 were from food. Cluster 3 contained all the Pseudomonas reference strains and 26 non-motile strains isolated from meat. These were shown to be non-motile variants of Pseudomonas fragi. A simple identification scheme, based on five tests, is presented for the distinction of the three types of bacteria.