Isolate IMAU20185 belonging to ST9 was a six-locus variant of ST1 to which it was connected by a dotted line. Isolate IMAU80137 belonging to ST19 was a six-locus variant of ST14 to which it was also connected by a dotted line. UPGMA tree based on MLST data Genetic relatedness amongst the L. lactis isolates investigated in this study showed they were well clustered within two major groups, A and B. Group A was comprised of 34 isolates and group B of only 16 isolates. Group A was the better supported

group and included two subgroups. Group B was a weakly supported group that included four Erastin in vivo subgroups (Figure  3). With the exception of ST19, isolates in group A were closely related only differing in PI3K inhibitor two out of the eight loci from the primary founder, ST14. The isolate that belonged to ST19 was a six-locus variant of the primary founder. Isolates in Group B were distantly related and differed in between two and six of the eight loci from the primary founder ST1. Figure 3 UPGMA dendrogram showing the genetic relationships between the 20 STs that belong to L . lactis through MM-102 MLST typing in this study. The Phylogenetic tree was produced using START 2.0 software and the UPGMA method.

The numbering in the figure refers to the ST. Two major phylogroups were designated as A and B. Discussion MLST is considered to be the best method for studying molecular epidemiology and population structure of bacteria [29–31]. Although this approach has been developed for several LAB, such as Lb. plantarum, Lb. delbrueckii, Lb. casei, and O. oeni[25, 26, 32], until this study there had been no MLST protocol used for L. lactis. In this study, we used MLST with eight housekeeping genes on 50 L. lactis isolates from a relatively large geographic area including Mongolia, a number of Chinese Provinces and an Autonomous region. These representative isolates are unique in their diversity of sources and provide the relevant information required for a better understanding of genetic diversity, persistence and movement. The first step in development of a MLST typing

method required analysis of the sequence diversity of eight housekeeping genes from the 50 L. lactis isolates under evaluation, to ensure that the MLST protocol had Dichloromethane dehalogenase the discriminatory power to type isolates within a single species. The two loci that had low polymorphism, contained three and four polymorphic sites in the recA and carB loci respectively (Table  1). The low level of biodiversity in recA and carB suggested they had similar sequences at the species level and would, therefore, have a lower discriminatory ability than the other housekeeping loci used in this study. The remaining six loci, groEL, pheS, uvrC, rpoB, pyrG, murC had more polymorphic sites (between five and nine), suggesting that they would have a good discriminatory ability when used in MLST. A total of 47 polymorphic sites were detected in the eight loci giving a polymorphism rate of 0.88% of the 5,325 nucleotides present.