Comparative genome analysis, predicted lifestyle and antimicrobial strategies of Lactococcus carnosus and Lactococcus paracarnosus isolated from meat.

Strains of the psychrotrophic bacterium Lactococcus piscium have gained increasing attention as potentially bioprotective cultures due to their assertiveness against fish and meat spoilage bacteria as well as pathogenic bacteria. Recently, we have described two novel species within the genus Lactococcus (Lc.) namely Lc. carnosus (TMW 2.1612) and Lc. paracarnosus (TMW 2.1615) isolated from modified atmosphere packaged meat. Within this study, we compared the genomes of two Lc. carnosus strains, two Lc. paracarnosus strains and 16 Lc. piscium strains from our laboratory and five publicly available genomes previously affiliated to the species Lc. piscium. Our phylogenetic analysis supports reclassification of 20 of the strains to either Lc. carnosus or Lc. paracarnosus, so far limiting the Lc. piscium type strain (DSM 6634) as sole representative of this species. Comparative genomics approach was conducted to predict underlying mechanisms involved in interspecies competition strategies of Lc. carnosus and Lc. paracarnosus against meat spoilers and predict their lifestyle in meat environments. In general, strains of the three species were highly similar regarding metabolic pathways for most of the relevant meat-derived substrates. In silico analyses enabled prediction of homolactic hexose fermentation by Lc. carnosus, Lc. paracarnosus and Lc. piscium. Further, genes required for the heterofermentative metabolism of hexoses and pentoses were only found in the Lc. pisicum type strain (DSM 6634). We predict a low spoilage potential for Lc. carnosus and Lc. paracarnosus strains. No genes for decarboxylation of amino acids yielding biogenic amines were found in the genomes. Regarding their antimicrobial mechanisms against spoilers, we found a strain-specific putative polymorphic toxin system predictively delivered by the type VIIb secretion system, enabling cell-to-cell contact-dependent growth inhibition. Furthermore, we found additional genes predictively involved to the suppression of spoilers within the food microbiome (prophages, lytic domains, bacteriocins, metabolites).