Randomly amplified polymorphic DNA (RAPD) has been used for rapid typing of Lactobacillus plantarum strains. RAPD was used with either purified chromosomal DNA serving as template in the polymerase chain reaction, or with crude cell extracts, and using a 9-mer primer with 80% G+C content. Amplified DNA was visualized by ethidium bromide staining after separation on agarose gels. Patterns from 20 Lact. plantarum strains and two Lact. pentosus strains were analysed using the Pearson products moment correlation coefficient (r) and the unweighted pair group method with arithmetic averages (UPGMA). With some exceptions, the two sources of template DNA gave the same clusters and subclusters of strains at the similarity level of 50%. About 50% of the strains could be individually separated from all the other tested strains. The buffer brand, the amount of primer and crude cell extract used in the PCR-step were crucial for the final pattern.

Broad-range 16s rDNA PCR (BR-PCR) applied to DNA from 32 clinical enterococcal isolates and 12 other enterococci from a clinical reference collection followed by species-specific hybridization analysis identified 25 strains of Enterncoccus faecalis and 19 Enterncoccus species. Randomly amplified polymorphic DNA (RAPD) analysis using UPGMA clustering on the same material revealed four different clusters at a similarity level of 49%. Based on partial 165 rDNA sequence analysis of variable regions V4 and V9, it was possible to divide the 19 type strains specifying the genus Enterncoccus into 12 different 16s rDNA species groups. The type strain distribution then served as a template for the analysis of the other 44 strains which were assigned to four different species groups (it-d) based on their 165 rDNA motifs. There was good agreement with the RAPD clusters. Species group a was an individual species line containing 25 strains that were identified as E. faecalis. Group b also represented an individual species line of 12 strains identified as Em faecium. The remaining seven strains that formed species groups c and d could not be fully identified to species by this analysis. It was concluded that BR-PCR of 16s rDNA followed by partial sequence analysis of the PCR products is a reliable technique for the identification and classification of enterococci. Further division of unresolved species groups should be achievable if regions other than V4 and V9 of 165 rDNA are also analysed.

The use of randomly amplified polymorphic DNA (RAPD) for rapid, reliable, and easily interpreted identification of enterococci was evaluated. Nineteen type strains of Enterococcus, 12 reference strains, and 114 clinical isolates of Enterococcus were analyzed. Discrimination was obtained between most type strains, the exceptions being Ent. casseliflavus and Ent. flavescens, which had relatively similar RAPD-profiles. Ent. faecalis and Ent. faecium were readily separated, and Ent. gallinarum and Ent. durans could also be identified. Extracts to be used in the polymerase chain reaction (PCR) were prepared directly from agar plate colonies, which made it possible to complete the identification procedure in one day. RAPD was proved to be a fast and reliable method for identification of most Enterococcus spp. of clinical significance.

A total of 17 Lactobacillus plantarum strains that originated from different environments and 24 reference strains were classified by performing a restriction endonuclease analysis of total chromosomal DNAs digested with EcolRI, HindIII, and ClaI, and the resulting patterns were visualized after the fragments were separated according to size by agarose gel electrophoresis. The patterns were analyzed by using the Pearson product moment correlation coefficient and the unweighted pair group algorithm with arithmetic averages. All but two L. plantarum strains formed a cluster that was separated from the reference strains at a similarity level of 2% (cluster 1). The two remaining L. plantarum strains (cluster 2) merged with cluster 1 at a level of similarity of 28%. The reference strains formed four additional clusters, and four reference strains were stragglers. Cluster 3 (three Lactobacillus pentosus strains) and cluster 4 (including Pediococcus acidilacti CCUG 32235T [T = type strain] and Lactobacillus fermentum ATCC 14931T) merged with cluster 1 at a level of similarity of 25%. Cluster 5 comprised 10 Lactobacillus reuteri strains, and cluster 6 contained the type strains of Lactobacillus amylovorus, Lactobacillus gasseri, and Lactobacillus vaginalis. Cluster 1 (L. plantarum) was divided into three subclusters, and this subdivision reflected to some extent certain phenotypic features of presumed ecological significance, including the ability to adhere to intestinal mucosa (subcluster lc) and starch and glycogen degradation (subcluster la). A principal-component analysis significantly distinguished the strains belonging to different species. Also, the subcluster l c strains could be separated from the rest of the L. plantarum strains. The results of restriction endonuclease analysis of total chromosomal DNA were found to be reproducible, and this method can be used to (i) differentiate between similar strains belonging to the same Lactobacillus species and group strains within a species, (ii) distinguish between strains of different Lactobacillus species, and (iii) place strains in specific Lactobacillus species.The genus Lactobacillus is heterogeneous, and the relationships among the different species have been studied by reverse transcriptase sequencing of 16s rRNAs (2). Larger evolutionary distances between species have been revealed clearly, but the relationships among more closely related species are muddled. Furthermore, rRNA gene sequencing is useless for classification below the species level; for example, the 16s rRNA gene sequences of Lactobacillus plantarum and Lactobacillus pentosus are identical, and the same is true for the sequences of Lactobacillus casei and Lactobacillus rhamnosus (2, 12).Workers have developed a method in which the genomes of Lactobacillus strains are characterized by performing a restriction endonuclease analysis (REA) of chromosomal DNAs and then visualizing the resulting fragment patterns after agarose gel electrophoresis is performed (14). This method has been used to classify...

No abstract