Differential distribution and expression of Panton-Valentine leucocidin among community-acquired methicillin-resistant Staphylococcus aureus strains

Abstract

Community-acquired methicillin-resistant Staphylococcus aureus (CA-MRSA) is an emerging threat worldwide. CA-MRSA strains differ from hospital-acquired MRSA strains in their antibiotic susceptibilities and genetic backgrounds. Using several genotyping methods, we clearly define CA-MRSA at the genetic level and demonstrate that the prototypic CA-MRSA strain, MW2, has spread as a homogeneous clonal strain family that is distinct from other CA-MRSA strains. The Panton-Valentine leucocidin (PVL)-encoding genes, lukF and lukS, are prevalent among CA-MRSA strains and have previously been associated with CA-MRSA infections. To better elucidate the role of PVL in the pathogenesis of CA-MRSA, we first analyzed the distribution and expression of PVL among different CA-MRSA strains. Our data demonstrate that PVL genes are differentially distributed among CA-MRSA strains and, when they are present, are always transcribed, albeit with strain-to-strain variability of transcript levels. To directly test whether PVL is critical for the pathogenesis of CA-MRSA, we evaluated the lysis of human polymorphonuclear leukocytes (PMNs) during phagocytic interaction with PVL-positive and PVL-negative CA-MRSA strains. Unexpectedly, there was no correlation between PVL expression and PMN lysis, suggesting that additional virulence factors underlie leukotoxicity and, thus, the pathogenesis of CA-MRSA.

FIG. 1.

PFGE of strains with different spa types. STD, molecular size standard.

FIG. 2.

Differential expression of the lukF gene of Panton-Valentine leucocidin. The relative levels of lukF transcripts in CA-MRSA strains were determined by real-time PCR with a molecular beacon probe specific for the lukF gene. A molecular beacon directed against a Staphylococcus-specific region of 16S rRNA was used for normalization. †, Strain 9918 had greater than fivefold higher relative lukF transcript levels (1.29) than any of the other strains tested; *, the standard error for strain 11450 could not be reliably calculated due to the presence of an outlier.

FIG. 3.

PMN lysis during phagocytic interaction with PVL-negative (NEG) and PVL-positive (POS) S. aureus strains of closely related genetic backgrounds. (A) Percentage of PMN lysis over a period of 9 h. Ht648, heat-killed strain 648. (B) Genetic characteristics of the pairs used in the PMN assay.