Role of the accessory gene regulator agr in community-associated methicillin-resistant Staphylococcus aureus pathogenesis

Abstract

The molecular basis underlying the pathogenic success of community-associated methicillin-resistant Staphylococcus aureus (CA-MRSA) is not completely understood, but differential gene expression has been suggested to account at least in part for the high virulence of CA-MRSA strains. Here, we show that the agr gene regulatory system has a crucial role in the development of skin infections in the most prevalent CA-MRSA strain USA300. Importantly, our data indicate that this is due to discrepancies between the agr regulon of CA-MRSA and those of hospital-associated MRSA and laboratory strains. In particular, agr regulation in strain USA300 led to exceptionally strong expression of toxins and exoenzymes, upregulation of fibrinogen-binding proteins, increased capacity to bind fibrinogen, and increased expression of methicillin resistance genes. Our findings demonstrate that agr functionality is critical for CA-MRSA disease and indicate that an adaptation of the agr regulon contributed to the evolution of highly pathogenic CA-MRSA.

Fig. 1.

Effect of agr on virulence in a mouse subcutaneous infection model. (A) Abscess sizes. Bacteria of the indicated strains were injected subcutaneously in hairless mice at ∼3 × 10⁷ CFU (except RN6390 and RN6911, 1 × 10⁶ CFU), and abscess dimensions were measured every day. Statistical significance of differences between abscess sizes on each day for each of the three wild-type/agr mutant comparisons was determined using unpaired t tests. *, P < 0.05; **, P < 0.01; ***, P < 0.001; #, all values achieved with the agr mutant strain were 0. (B) Development of dermonecrotic lesions. Mice infected with the LAC and RN6390 strains, but not the other strains, commonly developed open dermonecrotic lesions as shown in the upper panel. The lower panel shows a characteristic closed abscess formed by the LAC agr strain. (C) Histopathological evaluation of abscesses formed by the LAC and LAC agr strains. Histopathology of abscesses formed by RN6390 was similar to that of abscesses formed by strain LAC, and histopathologies of abscesses formed by the 252, 252 agr, and RN6911 strains were similar to those of abscesses formed by the LAC agr strain.

Fig. 2.

Growth-dependent expression of agr. Expression of agr was determined using qRT-PCR of RNAIII during the growth of S. aureus strains and their isogenic agr mutants. Lines represent OD₆₀₀, plotted on the left y axis. Bars represent the relative expression of RNAIII compared to that of the housekeeping gene gyrB (control), plotted on the right y axis. Expression of RNAIII in agr-negative strains was not detectable in any strain at any time point.

Fig. 3.

Effect of agr on PSM expression in various strain backgrounds. (A) Expression of the psmα and psmβ operons was determined using qRT-PCR as previously described (35). (B) Concentrations of all S. aureus PSMs (see legend) in culture filtrates were measured as described previously (35). Samples for measurements were taken from 8-h (strains LAC and RN6390) or 10-h (strain 252) cultures. Striped parts of bars represent the N-deformylated fraction of a particular PSM. Note that strain 252 contains the psm-mec gene in contrast to strains LAC and RN6390.

Fig. 4.

Effect of agr on the expression of selected surface protein and regulatory genes in various strain backgrounds. The graphs show expression of selected genes as determined by qRT-PCR at maximal expression of agr in the LAC, LAC agr, RN6390, and RN6911 strains for 8 h and the 252 and 252 agr strains for 10 h. Data shown on the y axis represent expression relative to that of the housekeeping gene gyrB (control). *, P < 0.05; **, P < 0.01; ***, P < 0.001 (t tests comparing agr-negative to wild-type strain samples). Expression levels of alpha-toxin were confirmed on the protein level by Western blots with specific alpha-toxin antiserum and are shown above the qRT-PCR data.

Fig. 5.

Effect of agr on methicillin resistance and fibrinogen binding phenotypes. (A) MIC assays with oxacillin. Strains were inoculated from precultures and grown with the addition of oxacillin at different concentrations in microtiter plates for 24 h. (B) Fibrinogen binding capacity. Bacteria were grown for 16 h and assayed for fibrinogen binding capacity on fibrinogen-coated microtiter plates after an incubation period of 2 h. *, P < 0.05; **, P < 0.01; ***, P < 0.001.