VirulenceFinder for Enterococcus faecium and Enterococcus lactis : an enhanced database for detection of putative virulence markers by using whole-genome sequencing data

Abstract

ABSTRACT Enterococcus faecium ( Efm ) is a leading cause of hospital-associated (HA) infections, often enriched in putative virulence markers (PVMs). Recently, the Efm clade B was assigned as Enterococcus lactis ( Elts ), which usually lack HA- Efm infection markers. Available databases for extracting PVM are incomplete and/or present an intermix of genes from Efm and Enterococcus faecalis , with distinct virulence profiles. In this study, we constructed a new database containing 27 PVMs [ acm, scm, sgrA, ecbA, fnm, sagA, hylEfm, ptsD, orf1481, fms15, fms21-fms20 (pili gene cluster 1, PGC-1), fms14-fms17-fms13 (PGC-2), empA-empB-empC (PGC-3), fms11-fms19-fms16 (PGC-4), ccpA, bepA, gls20-glsB1, and gls33-glsB ] from nine reference genomes (seven Efm + two Elts ). The database was validated against these reference genomes and further evaluated using a collection of well-characterized Efm ( n = 43) and Elts ( n = 7) control strains, by assessing PVM presence/absence and its variants together with a genomic phylogeny constructed as single-nucleotide polymorphisms. We found a high concordance between the phylogeny and in silico findings of the PVM, with Elts clustering separately and mostly carrying Elts- specific PVM gene variants. Based on our validation results, we recommend using the database with raw reads instead of assemblies to avoid missing gene variants. This newly constructed database of 27 PVMs will enable a more comprehensive characterization of Efm and Elts based on WGS data. The developed database exhibits scalability and boasts a range of applications in public health, including diagnostics, outbreak investigations, and epidemiological studies. It can be further used in risk assessment for distinguishing between safe and unsafe enterococci. IMPORTANCE The newly constructed database, consisting of 27 putative virulence markers, is highly scalable and serves as a valuable resource for the comprehensive characterization of these closely related species using WGS data. It holds significant potential for various public health applications, including hospital outbreak investigations, surveillance, and risk assessment for probiotics and feed additives.