SpyA, a C3-like ADP-ribosyltransferase, contributes to virulence in a mouse subcutaneous model of Streptococcus pyogenes infection

Abstract

Streptococcus pyogenes is an important human pathogen with an expansive repertoire of verified and putative virulence factors. Here we demonstrate that a mutant deficient in the production of the streptococcal ADP-ribosyltransferase SpyA generates lesions of reduced size in a subcutaneous mouse infection model. At early stages of infection, when the difference in lesion size is first established, inflamed tissue isolated from lesions of mice infected with spyA mutant bacteria has higher levels of mRNA encoding the chemokines CXCL1 and CCL2 than does tissue isolated from mice infected with wild-type bacteria. In addition, at these early times, the mRNA levels for the gene encoding the intermediate filament vimentin are higher in the mutant-infected tissue. As wound resolution progresses, mRNA levels of the gene encoding matrix metallopeptidase 2 are lower in mutant-infected tissue. Furthermore, we demonstrate that the spyA mutant is internalized more efficiently than wild-type bacteria by HeLa cells. We conclude that SpyA contributes to streptococcal pathogenesis in the mouse subcutaneous infection model. Our observations suggest that the presence of SpyA delays wound healing in this model.

Fig. 1.

Deletion of spyA. (A) Chromosomal location of spyA. The white bar indicates deleted region. Arrows below chromosome show locations of primers used for PCR analysis. (B) SpyA protein. Conserved residues required for ADP-ribosyltransferase activity are shown. The gray bar shows protein; the white area represents the deleted region. Numbers above protein indicate residues retained in the deletion construct. (C) PCR using primers shown in panel A. Size standards are shown on the left; sizes of PCR fragments are shown on the right. (D) Western blots to detect SpeA, streptokinase, and Mac in GAS supernatants. Lanes 1 and 2 are mid-log; lanes 3 and 4 are stationary phase. Lanes 1 and 3 are supernatants from the wild type, and lanes 2 and 4 are from the mutant.

Fig. 2.

Skin lesions formed by S. pyogenes in Crl:SKH1-Hr^hr mice. (A) Mice were infected subcutaneously with 2 × 10⁷ bacteria, and resulting lesions were measured daily for 2 weeks (“*” indicates P < 0.05 using the Mann-Whitney test). The plot shows median, quartiles, minimum, and maximum. Clear boxes represent wild-type-infected mice (n = 15), and gray boxes represent spyA mutant-infected mice (n = 13). (B) CFU per gram of lesion tissue were determined at 24, 48, and 72 h postinfection in mice infected with 8 × 10⁶ bacteria. No significant differences were detected using a two-tailed, unpaired Student t test after confirming equal variance by F test. Each circle represents a single mouse. Open circles represent wild-type CFU, and filled circles represent spyA CFU.

Fig. 3.

Cxcl1 and Ccl2 expression in GAS-infected tissue. RNA was isolated from inflamed tissue at 24 and 72 h postinfection, and levels of Cxcl1 (A) or Ccl2 (B) transcripts were quantified in relation to that of Gapdh using qRT-PCR. Values reported are fold change compared to values for wild-type-infected tissue at 24 h. Graphs show means and SEM of results for 3 mice per group at each time, 2 to 3 technical replicates per mouse. (“*” indicates P < 0.05 by two-tailed, unpaired Student's t test; “**,” P < 0.01). Clear bars represent mice infected with wild-type bacteria; black bars represent mice infected with spyA mutant bacteria.

Fig. 4.

Expression of Mmp2 and Vim in infected tissue. RNA was isolated from inflamed tissue, and using qRT-PCR levels of Mmp2 (A) or Vim (B), transcripts were determined in relation to Gapdh transcripts and normalized to levels found in wild-type-infected tissue at 24 h. Graphs show means and SEM of 3 mice per group (2 to 3 replicates per mouse) at each time. *, P < 0.05; **, P < 0.01. A two-tailed, unpaired Student's t test was used to compare the wild type to the mutant, and ANOVA followed by Tukey's posttest was used to compare across time points. White bars represent wild-type bacteria, and black bars represent spyA mutant bacteria.

Fig. 5.

Internalization of GAS into HeLa cells. Wild-type and spyA mutant bacteria were incubated with HeLa cells for 3 h, at which point associated bacteria were quantified, or cells were treated with antibiotics for 1 h. No difference was detected in associated bacteria at 3 h. The graph shows the percentage of associated bacteria that remain viable after antibiotic treatment. “**” indicates P < 0.01 using a two-tailed, paired Student t test. Bars show means and standard errors of results from 7 independent experiments.