Computer-Assisted Identification of Lactic Acid Bacteria from Meats

Characterization of Lactobacillus sakei by the type of stereoisomers of lactic acid produced

Lactobacillus sakei strains were characterized by the shift of the type of stereoisomers of lactic acid produced in the presence of 50 mM sodium acetate in a medium. Of 27 Lactobacillus sakei strains studied, 20 strains showed high levels of DNA-DNA similarity with L. sakei NRIC 1071(T), and were confirmed as L. sakei. The three remaining strains were identified as Lactobacillus curvatus by DNA-DNA similarity, and three other strains were included in the cluster of Lactobacillus plantarum/Lactobacillus pentosus/Lactobacillus paraplantarum and one strain in the cluster of Lactobacillus paracasei on the basis of 16S rRNA gene sequences. Of the 20 L. sakei strains, 19 strains shifted the type of stereoisomers of lactic acid produced from the DL-type to the L-type in the presence of 50 mM sodium acetate. L. curvatus strains and strains included in the cluster of L. plantarum/L. pentosus/L. paraplantarum and in the cluster of L. paracasei did not shift the type of stereoisomers of lactic acid produced. The change of the type of stereoisomers of lactic acid from the DL-type to the L-type in the presence of sodium acetate was concluded to be species-specific for L. sakei and useful for identification of strains in this species.

Comparison of pattern recognition techniques for the identification of lactic acid bacteria

AIMS

The goal of this study was to evaluate three pattern recognition methods for use in the identification of lactic acid bacteria.

METHODS AND RESULTS

Lactic acid bacteria (21 unknown isolates and 30 well-characterized strains), including the Lactobacillus, Lactococcus, Streptococcus, Pediococcus and Oenococcus genera, were tested for 49 phenotypic responses (acid production on carbon sources). The results were scored in several ways. Three procedures, k-nearest neighbour analysis (KNN), k-means clustering and fuzzy c-means clustering (FCM), were applied to the data.

CONCLUSION

k-Nearest neighbour analysis performed better with five-point-scaled than with binary data, indicating that intermediate values are helpful to classification. k-Means clustering performed slightly better than KNN and was best with fuzzified data. The best overall results were obtained with FCM. Genus level classification was best with FCM using an exponent of 1.25.

SIGNIFICANCE AND IMPACT OF THE STUDY

The three pattern recognition methods offer some advantages over other approaches to organism classification.

Development of a rapid method for the identification of Lactobacillus spp. isolated from naturally fermented italian sausages using a polymerase chain reaction-temperature gradient gel electrophoresis

A rapid method for the identification of Lactobacillus spp. isolated from naturally fermented Italian sausages was developed. It is based on the amplification of a small fragment from the 16S rRNA gene followed by temperature gradient gel electrophoresis (TGGE). Lactobacillus sakei, L. curvatus, L. alimentarius, L. casei, L. plantarum and L. brevis, obtained from International Collections, were used to optimize the method. Thiry-nine strains of Lactobacillus spp. were isolated from naturally fermented sausages and, after traditional identification, were tested by the PCR-TGGE protocol developed. No differences were observed comparing the results obtained, apart from five strains identified as L. curvatus that showed a PCR-TGGE profile identical to L. sakei.

Amino acid decarboxylase capability of microorganisms isolated in Spanish fermented meat products

Enterobacteria, lactic acid bacteria (LAB) and Gram-positive cocci were isolated from Spanish meat products. The most frequent species in the meat products studied were identified as Lactobacillus sake, Lactobacillus plantarum and Lactobacillus curvatus from De Man-Rogosa-Sharpe agar; Staphylococcus xylosus, Staphylococcus saprophyticus and Micrococcus varians from mannitol salt phenol-red agar; and Hafnia alvei, Escherichia coli, Pseudomonas fluorescens, Enterobacter amnigenes and Enterobacter aerogenes from violet red bile dextrose agar. The amino acid decarboxylase activity of the microorganisms isolated was assayed. Enterobacteria had higher amino acid decarboxylase activity than the other groups. LAB did not show any significant amino acid decarboxylase capability in this study.

Characterisation of lactic acid bacteria isolated from naturally fermented Greek dry salami

A total of 348 lactic acid bacteria isolated from five batches of naturally fermented dry salami at various stages of ripening were characterised. The majority of the strains were assigned to two main phylogenetic groups of species: (i) the psychrotrophic, formerly called atypical, meat streptobacteria (169 strains) and (ii) a new genus Weissella (120), which was recently proposed (Collins et al., 1993) to include Leuconostoc paramesenteroides and some other closely related species. Meat streptobacteria were identified as Lactobacillus curvatus (88 strains) and L. sake (76), whereas 5 strains were indistinguishable and, thus designated L. sake/curvatus. Non-psychrotrophic streptobacteria were also isolated and identified as L. plantarum (34 strains), L. farciminis (10), L. coryniformis (1) and L. casei subsp. pseudoplantarum (1). The majority of the Weissella strains (86) were leuconostoc-like bacteria; four of them were identified as W. viridescens, 11 belonged to the newly described W. hellenica (Collins et al., 1993), another 11 resembled W. paramesenteroides, whereas 60 isolates were not classified to any species. The latter group comprised strains that produced D(L)-lactate. The remaining Weissella were gas-forming, arginine-positive rods assigned to W. minor (31) and W. halotolerans (3). Other species identified were Enterococcus faecium (10), Leuconostoc mesenteroides (1), L. brevis (1) and Pediococcus sp. (1). The main criteria used to distinguish between above species as well as their distribution on the five salami batches in relation to their succession with time and suitability as starters were discussed.

Numerical taxonomy and identification of lactic acid bacteria from spoiled, vacuum-packaged vienna sausages

Sixty-one lactic acid bacteria from spoiled vacuum-packaged vienna sausages and 15 reference strains were tested for 72 phenotypic characteristics. An identification key and a computer data base, both specific for lactic acid bacteria from meat sources, were used for identification and the results were compared. There was a high correlation (86.9%) between the two procedures in the identification of strains to genus level. However, only a 54.8% correlation was obtained in identifying strains to the species level. With numerical taxonomy (Ssm matching coefficient with average linkage clustering) 60 strains were recovered in six clusters at the 89% similarity level. While most Leuconostoc strains clustered separately from the Lactobacillus strains, the identity of many leuconostocs was not clarified. The presence of a heterogeneous cluster containing typical and 'atypical' strains of the Lactobacillus saké/curvatus group and a separate homogeneous Lact. curvatus cluster was noted. Closer examination of the data suggested that the 'atypical' lactobacilli were all strains of Lact. saké.