epidermidis

spx mutant strain. We followed the same allelic exchange strategy (Bruckner, 1997) as that used in the construction of an S. epidermidis clpP mutant strain (Wang et al., 2007). More than 2000 clones were screened, but the desired double-crossover strain in which spx is replaced by an erythromycin-resistance cassette was not found, although we indentified single-crossover strains as determined by PCR amplifying the spx bordering regions (data not shown). The attempt to construct an spx mutant stain with a high-efficiency system through pKOR1 (Bae & Schneewind, 2006) also failed (data not shown). We further used a molecular epidemiological approach to examine the existence of spx in a collection of 80 S. epidermidis (Li et al., 2009) clinical isolates. All tested strains harbor the spx gene, indicating find more the possibility that spx could be an

essential gene (data not shown). Instead, we constructed an spx antisense knockdown plasmid PQG56 coding reversed spx mRNA to downregulate the expression of Spx. In a previous study, Nakano et al. (2003a) overexpressed Spx in B. subtilis to study its regulatory functions. Because the construction of an S. epidermidis spx mutant strain failed, we attempted to overexpress Spx in S. epidermidis to study its regulatory effect on biofilm formation. Attempts to overexpress Spx in B. subtilis were at first unsuccessful due to the rapid degradation of the protein by ClpP protease. Luminespib ic50 Successful overexpression of Spx was achieved when an spx mutant allele that codes for a protease-resistant form of Spx (C-terminal mutant) was constructed

and expressed in B. subtilis (Zuber, 2004). Thus, in addition to the expression plasmid pQG54, which carries a WT spx, we constructed another expression vector (PQG55) with an altered spx allele, with a substitution from Ala and Asn codons in the C-terminal to two Asp condons to encode a mutated DOCK10 SsrA peptide, in order to avoid the SsrA peptide-tagged proteolysis by ClpXP. These three plasmids were transformed into S. epidermidis. To prevent the resistance from being degraded, we compared the expression level of Spx in strains carrying PQG53, PQG54 and PQG 55 separately. As a result, little Spx protein was detected in the vector control stain harboring PQG53 and the WT expression allele harboring PQG54, whereas Spx accumulated in the strain harboring PQG55 (Fig. 1). Biofilm formations of S. epidermidis strains harboring different plasmids were compared using semi-quantitative assays. Biofilm formation of the strain harboring pQG54 was comparable with that of the vector control strain harboring pQG53, whereas biofilm formation of the strain harboring pQG55 decreased drastically (Fig. 2). The Spx levels in these strains were examined by Western blot. The result that Spx accumulated in the strain harboring pQG55, but not in the strain harboring pQG54, indicates that Spx had a negative effect on the biofilm formation of S. epidermidis.