Search

Background Koumiss is a naturally fermented mare’s milk. Over recent decades, numerous studies have revealed the diversity of lactic acid bacteria in koumiss. However, there is limited information available regarding its secondary major component yeast profile. Results A total of 119 bacterial and 36 yeast species were identified among the 14 koumiss samples. The dominant bacterial species in koumiss were Lactobacillus helveticus, Lactobacillus kefiranofaciens, Lactococcus lactis, Lactococcus raffinolactis, and Citrobacter freundii. The main yeast species were Dekkera anomala, Kazachstania unispora, Meyerozyma caribbica, Pichia sp.BZ159, Kluyveromyces marxianus, and uncultured Guehomyces. The bacterial and yeast Shannon diversity of the Xilinhaote-urban group were higher than those of the Xilingol-rural group. The most dominant organic acids were lactic, acetic, tartaric, and malic acids. Lactic acid bacteria species were mostly responsible for the accumulation of those organic acids, although Kazachstania unispora, Dekkera anomala, and Meyerozyma caribbica may also have contributed. Redundancy analysis suggested that both bacteria and yeast respond to koumiss flavor, such as Lactobacillus helveticus and Dekkera anomala are associated with sourness, astringency, bitterness, and aftertaste, whereas Lactococcus lactis and Kazachstania unispora are associated with umami. Conclusions Our results suggest that differences were observed in koumiss microbiota of Xilinhaote-urban and Xilingol-rural samples. The biodiversity of the former was higher than the latter group. Positive or negative correlations between bacteria and yeast species and taste also were found.

Background The genomic diversity of different species within the genus Lactococcus and the relationships between genomic differentiation and environmental factors remain unclear. In this study, type isolates of ten Lactococcus species/subspecies were sequenced to assess their genomic characteristics, metabolic diversity, and phylogenetic relationships. Results The total genome sizes varied between 1.99 (Lactococcus plantarum) and 2.46 megabases (Mb; L. lactis subsp. lactis), and the G + C content ranged from 34.81 (L. lactis subsp. hordniae) to 39.67% (L. raffinolactis) with an average value of 37.02%. Analysis of genome dynamics indicated that the genus Lactococcus has an open pan-genome, while the core genome size decreased with sequential addition at the genus and species group levels. A phylogenetic dendrogram based on the concatenated amino acid sequences of 643 core genes was largely consistent with the phylogenetic tree obtained by 16S ribosomal RNA (rRNA) genes, but it provided a more robust phylogenetic resolution than the 16S rRNA gene-based analysis. Conclusions Comparative genomics indicated that species in the genus Lactococcus had high degrees of diversity in genome size, gene content, and carbohydrate metabolism. This may be important for the specific adaptations that allow different Lactococcus species to survive in different environments. These results provide a quantitative basis for understanding the genomic and metabolic diversity within the genus Lactococcus, laying the foundation for future studies on taxonomy and functional genomics. Electronic supplementary material The online version of this article (10.1186/s12866-017-1120-5) contains supplementary material, which is available to authorized users.

Background Lactobacillus plantarum is a lactic acid bacterium (LAB) of considerable industrial interest since it has an important role in the production of fermented food. In the present study, the genetic diversity and population structure within 186 L. plantarum isolates was determined based on a novel MLST scheme employing eight housekeeping genes. These isolates had originated from different sources and geographic regions: 179 isolates were from our own culture collection and originated from China and Mongolia and seven isolates were type or reference isolates from other collections. Results The results showed that 179 isolates and seven reference isolates could be assigned to 73 different sequence types (STs), forming ten clonal complexes (CCs) and 23 singletons. There were 158 polymorphic sites detected in total, and the nucleotide diversity per site varied from 0.00401 in clpX to 0.03220 in groEL. The minimum spanning tree analyses suggested that the evolution of L. plantarum isolates have little relationship with ecological sources have similar nucleotide diversity. Phylogenetic trees and structure indicated that there were six lineages in the L. plantarum isolates used in our study. Split-decomposition and ClonalFrame analysis indicated that recombination had occurred throughout the population of L. plantarum, but it occurred at a low frequency in these eight loci. Conclusion We deduced that L. plantarum isolates from the same ecological niches have similar genetic diversity and population structure. The MLST scheme presented in this study provides abundant sequence data for L. plantarum and enabled global comparisons of isolates associated with various environmental origins to be made. This will further advance our understanding of the microbial ecology of this industrially important LAB. Electronic supplementary material The online version of this article (doi:10.1186/s12866-015-0584-4) contains supplementary material, which is available to authorized users.

Background Lactobacillus fermentum is economically important in the production and preservation of fermented foods. A repeatable and discriminative typing method was devised to characterize L. fermentum at the molecular level. The multilocus sequence typing (MLST) scheme developed was based on analysis of the internal sequence of 11 housekeeping gene fragments (clpX, dnaA, dnaK, groEL, murC, murE, pepX, pyrG, recA, rpoB, and uvrC). Results MLST analysis of 203 isolates of L. fermentum from Mongolia and seven provinces/ autonomous regions in China identified 57 sequence types (ST), 27 of which were represented by only a single isolate, indicating high genetic diversity. Phylogenetic analyses based on the sequence of the 11 housekeeping gene fragments indicated that the L. fermentum isolates analyzed belonged to two major groups. A standardized index of association (IAS) indicated a weak clonal population structure in L. fermentum. Split decomposition analysis indicated that recombination played an important role in generating the genetic diversity observed in L. fermentum. The results from the minimum spanning tree strongly suggested that evolution of L. fermentum STs was not correlated with geography or food-type. Conclusions The MLST scheme developed will be valuable for further studies on the evolution and population structure of L. fermentum isolates used in food products. Electronic supplementary material The online version of this article (doi:10.1186/s12866-015-0447-z) contains supplementary material, which is available to authorized users.

Background Economically, Leuconostoc lactis is one of the most important species in the genus Leuconostoc. It plays an important role in the food industry including the production of dextrans and bacteriocins. Currently, traditional molecular typing approaches for characterisation of this species at the isolate level are either unavailable or are not sufficiently reliable for practical use. Multilocus sequence typing (MLST) is a robust and reliable method for characterising bacterial and fungal species at the molecular level. In this study, a novel MLST protocol was developed for 50 L. lactis isolates from Mongolia and China. Results Sequences from eight targeted genes (groEL, carB, recA, pheS, murC, pyrG, rpoB and uvrC) were obtained. Sequence analysis indicated 20 different sequence types (STs), with 13 of them being represented by a single isolate. Phylogenetic analysis based on the sequences of eight MLST loci indicated that the isolates belonged to two major groups, A (34 isolates) and B (16 isolates). Linkage disequilibrium analyses indicated that recombination occurred at a low frequency in L. lactis, indicating a clonal population structure. Split-decomposition analysis indicated that intraspecies recombination played a role in generating genotypic diversity amongst isolates. Conclusions Our results indicated that MLST is a valuable tool for typing L. lactis isolates that can be used for further monitoring of evolutionary changes and population genetics.

Lactobacillus fermentum is economically important in the production and preservation of fermented foods. A repeatable and discriminative typing method was devised to characterize L. fermentum at the molecular level. The multilocus sequence typing (MLST) scheme developed was based on analysis of the internal sequence of 11 housekeeping gene fragments (clpX, dnaA, dnaK, groEL, murC, murE, pepX, pyrG, recA, rpoB, and uvrC).MLST analysis of 203 isolates of L. fermentum from Mongolia and seven provinces/ autonomous regions in China identified 57 sequence types (ST), 27 of which were represented by only a single isolate, indicating high genetic diversity. Phylogenetic analyses based on the sequence of the 11 housekeeping gene fragments indicated that the L. fermentum isolates analyzed belonged to two major groups. A standardized index of association (I A (S)) indicated a weak clonal population structure in L. fermentum. Split decomposition analysis indicated that recombination played an important role in generating the genetic diversity observed in L. fermentum. The results from the minimum spanning tree strongly suggested that evolution of L. fermentum STs was not correlated with geography or food-type.The MLST scheme developed will be valuable for further studies on the evolution and population structure of L. fermentum isolates used in food products.

Background Streptococcus thermophilus is a major dairy starter used for manufacturing of dairy products. In the present study, we developed a multilocus sequence typing (MLST) scheme for this important food bacterium. Sequences of 10 housekeeping genes (carB, clpX, dnaA, murC, murE, pepN, pepX, pyrG, recA, and rpoB) were obtained for 239 S. thermophilus strains, which were isolated from home-made fermented dairy foods in 18 different regions of Mongolia and China. Methods All 10 genes of S. thermophilus were sequenced, aligned, and defined sequence types (STs) using the BioNumerics Software. The nucleotide diversity was calculated by START v2.0. The population structure, phylogenetic relationships and the role of recombination were inferred using ClonalFrame v1.2, SplitsTree 4.0 and Structure v2.3. Results The 239 S. thermophilus isolates and 18 reference strains could be assigned into 119 different STs, which could be further separated into 16 clonal complexes (CCs) and 38 singletons. Among the 10 loci, a total of 132 polymorphic sites were detected. The standardized index of association (IAS = 0.0916), split-decomposition and ρ/θ (relative frequency of occurrence of recombination and mutation) and r/m value (relative impact of recombination and mutation in the diversification) confirms that recombination may have occurred, but it occurred at a low frequency in these 10 loci. Phylogenetic trees indicated that there were five lineages in the S. thermophilus isolates used in our study. MSTree and ClonalFrame tree analyses suggest that the evolution of S. thermophilus isolates have little relationship with geographic locality, but revealed no association with the types of fermented dairy product. Phylogenetic analysis of 36 whole genome strains (18 S. thermophilus, 2 S. vestibularis and 16 S. salivarius strains) indicated that our MLST scheme could clearly separate three closely related species within the salivarius group and is suitable for analyzing the population structure of the other two species in the salivarius group. Conclusions Our newly developed MLST scheme improved the understanding on the genetic diversity and population structure of the S. thermophilus, as well as provided useful information for further studies on the genotyping and evolutionary research for S. thermophilus strains with global diversity. Electronic supplementary material The online version of this article (doi:10.1186/s12866-015-0551-0) contains supplementary material, which is available to authorized users.