Enterococci are normal flora in the human intestinal tract, and also one of the leading causes of nosocomial infections, with most of the clinical isolates being Enterococcus faecalis and Enterococcus faecium. To identify potential virulence factors based on their antigenicity during infection, E. faecium genomic libraries were constructed and screened using sera from patients with E. faecium endocarditis. A number of E. faecium antigens were identified, most of which hit to surface proteins involved in transport, stress response, and metabolism in enterococci or other organisms. One of the antigens, SagA, is a prominent antigen that is secreted into the extracellular fluid, and the predicted sagA gene product was composed of three domains: a coiled-coil N-terminal domain, a repeat domain in the middle and a C-terminal domain that is also found in various proteins by BLAST searches, including cell wall hydrolases P60 and P45 of Listeria monocytogenes. The SagA proteins may form aggregates, based on expression and purification of the His-SagA protein and crosslinking experiments. The sagA gene was found to be essential for E. faecium growth and could not be disrupted. A sagA homolog, sagB, was found in Enterococcus hirae, and the sagB gene product showed similar N-terminal and C-terminal domains as the SagA protein, but with different repeats between these domains. The E. hirae SagB protein was also a secreted antigen that reacted with anti-SagA antibody. Two sag-like genes (salA and salB) were identified in a strain of Enterococcus faecalis, and the sal gene products had similar N-terminal domains as the Sag proteins, but differed in the C-terminus. The E. coli clones expressing Sal proteins reacted with sera from patients with E. faecalis endocarditis and the Sal proteins in E. faecalis appeared to be cell-associated. Characterization of salB mutant suggests that the salB gene may be involved in cell wall metabolism or cell division.

A similar immunoscreening approach has been used previously in our laboratory to identify potential virulence factors in E. faecalis, and a gene cluster involved in polysaccharide biosynthesis (epa) was identified and shown to be involved in E. faecalis virulence. As one of my projects, total polysaccharides were extracted from E. faecalis OG1RF and from two epa mutants constructed previously, TX5179 and TX5180, and western blots with patient sera showed that an immuno-reactive polysaccharide present in wild type OG1RF was not produced by either of the two epa mutants. The epa mutants were more sensitive to ethanol stress, neutrophil killing, and neutrophil phagocytosis than the wild type OG1RF, suggesting that the epa genes are involved in neutrophil phagocytosis and killing of E. faecalis.

Expression of virulence factors is commonly regulated by two component systems. A BLAST search was performed to identify potential two component systems in the E. faecalis V583 genome database using Bacillus subtilis PhoP/PhoS as query sequences. Eleven gene pairs were identified, seven of which were disrupted in E. faecalis OG1RF. A mouse peritonitis model was used to evaluate the virulence of these mutants and two sets of genes, etaRS and etbRS, were found to be involved in stress response and E. faecalis virulence.

Search pubmed for papers by F Teng and GM Weinstock

Research Info

Identification and Characterization of Antigens and Potential Virulence Factors in Enterococcus