A two-component regulatory system, CsrR-CsrS, represses expression of three Streptococcus pyogenes virulence factors, hyaluronic acid capsule, streptolysin S, and pyrogenic exotoxin B

Abstract

Certain Tn916 insertions in the chromosome of an M1-type, nonmucoid Streptococcus pyogenes isolate (MGAS166) were previously shown to result in stable mucoidy with increased expression of the capsular synthetic genes. The transposon insertions in these strains are directly upstream of an apparent operon encoding a two-component regulatory system, designated csrR-csrS. Compared with MGAS166, these mucoid mutants are more hemolytic and cause significantly more tissue damage in a murine model of skin infection. To extend these observations, we constructed an in-frame deletion in the gene encoding the response regulator, csrR, and we evaluated the expression of other known S. pyogenes virulence factors. We discovered that csrR mutants have enhanced transcription of sagA, a gene associated with streptolysin S (SLS) and speB, the gene encoding pyrogenic exotoxin B (SpeB). The mutants also express substantially higher SLS activity and SpeB antigen in late-exponential-phase cultures. There is no change in expression of emm, scpA, sic, or cpa (genes encoding other S. pyogenes virulence factors). CsrR- strains but not the wild-type parental strain produce necrotizing lesions in a mouse model of subcutaneous infection. A double mutant with deletions in both csrR and the capsular synthesis genes caused fewer and smaller necrotic skin lesions than the csrR mutants. However, this nonmucoid csrR strain was more likely than the wild type to yield necrotic lesions, suggesting that mucoidy contributes to virulence in this model of infection but that there are other csrR-regulated factors involved in the production of necrotic lesions.

FIG. 1

Northern hybridization analysis of virulence-associated genes in CsrR⁻ streptococcal strains. Bacterial RNAs from six strains were isolated in late exponential phase and hybridized with four different PCR-generated probes as indicated to the left of each panel. The source of each RNA sample is indicated above the respective lane. The fifth and sixth lanes in each panel contain RNA from derivatives of SBmuc5 and SBmuc7, respectively, into which a csrRS-bearing plasmid (pNLB2223) has been introduced. wt, wild type.

FIG. 2

Northern hybridization analysis of virulence-associated genes in CsrR⁻ streptococcal strains. Samples from early-stationary-phase cultures of three strains were hybridized to probes specific for sagA, speB, and cpa. Samples included RNA from MGAS166, UMAA2392 (ΔcsrR csrS), UMAA2526 (ΔcsrR csrS ΔhasAB), and UMAA2392 with glucose added to a final concentration of 0.2% 2 h before RNA isolation.

FIG. 3

Hemolytic activities of streptococcal supernatants during 11 h of broth culture. (A) Absorbance of cultures from which aliquots were assayed. (B) Assays for SLS, performed in the presence of cholesterol to inhibit SLO. (C) Assays for SLO, performed in the presence of trypan blue to inhibit SLS. Note the different scales used in panels B and C. Data points for strain MGAS166 (wild type) are represented by the circles; those for strain UMAA2392 (ΔcsrR csrS) are represented by squares. White (open) and black (closed) data points indicate that assays were done in the absence or presence of horse serum, respectively. Assays were also performed without cholesterol and trypan blue and were within ±10% of the sum of the SLS and SLO activities (data not shown).

FIG. 4

Immunoblot analysis of streptococcal strains with a polyclonal antisera against SpeB. The three panels represent samples taken at different time intervals during growth in broth culture: mid-exponential (6 h of culture) (A), late exponential (12 h of culture) (B), and late stationary phase (48 h of culture) (C). Lanes were loaded with samples from the following strains: lane 1, MGAS166; lane 2, SBmuc5 (csrR::Tn916); lane 3, SBmuc7 (csrR::Tn916); lane 4, UMAA2392 (ΔcsrR csrS); lane 5, SBmuc5 (pNLB2223); and lane 6, SBmuc7 (pNLB2223). Lane M, molecular weight markers. The arrows indicate the mobility of 28-kDa proteins, corresponding to the molecular mass of fully processed SpeB. The slower-migrating bands that coexist in specimens containing specific SpeB bands are thought to represent unprocessed or complexed SpeB.

FIG. 5

Comparison of the size of necrotic lesions during the course of experiments II and IV (Table 3). In these experiments, necrotic lesions developed in all infected mice except in those inoculated with the wild-type strain, MGAS166 (no lesions), and in three of the mice inoculated with strain UMAA2526 (ΔcsrR csrS ΔhasAB). One of six mice inoculated with UMAA2526 in experiment II (A) and two of six mice in experiment IV (B) had no skin necrosis and are excluded from the analysis of lesion size in this figure. Lesions were measured, and the area was calculated 24, 48, and 72 h after inoculation. The infecting strains were SBmuc5 (solid black columns), SBmuc7 (white columns), UMAA2392 (cross-hatched columns), and UMAA2526 (shaded columns).