Staphylococcus aureus CodY negatively regulates virulence gene expression

Abstract

CodY is a global regulatory protein that was first discovered in Bacillus subtilis, where it couples gene expression to changes in the pools of critical metabolites through its activation by GTP and branched-chain amino acids. Homologs of CodY can be found encoded in the genomes of nearly all low-G+C gram-positive bacteria, including Staphylococcus aureus. The introduction of a codY-null mutation into two S. aureus clinical isolates, SA564 and UAMS-1, through allelic replacement, resulted in the overexpression of several virulence genes. The mutant strains had higher levels of hemolytic activity toward rabbit erythrocytes in their culture fluid, produced more polysaccharide intercellular adhesin (PIA), and formed more robust biofilms than did their isogenic parent strains. These phenotypes were associated with derepressed levels of RNA for the hemolytic alpha-toxin (hla), the accessory gene regulator (agr) (RNAII and RNAIII/hld), and the operon responsible for the production of PIA (icaADBC). These data suggest that CodY represses, either directly or indirectly, the synthesis of a number of virulence factors of S. aureus.

FIG. 1.

Effect of a ΔcodY mutation on the hemolytic activity of the culture fluid of S. aureus. The hemolytic activity of the culture fluid of each strain was tested on rabbit erythrocytes as described in Materials and Methods. One HU is defined as 50% total erythrocyte lysis per 50 μl of culture fluid. (A) The culture fluids of CDM7 (SA564 ΔcodY::ermC) (▪) and SA564 (wild type) (░⃞) were tested throughout growth for hemolytic activity. The growth of strains SA564 (▪) and CDM7 (▴) were measured as the OD₆₀₀. (B) The culture fluids of MS1 (UAMS-1 ΔcodY::ermC), UAMS-1 (wild type), MS2 (UAMS-1 containing pTL6936), and MS3 (MS1 containing pTL6936) were tested in the late stationary phase for hemolytic activity. Strains carrying the codY+ complementation plasmid, pTL6936, were grown with (I) or without (NI) IPTG to induce codY expression.

FIG. 2.

Effect of growth phase on alpha-toxin production in the culture supernatants of SA564, CDM7, UAMS-1, and MS1. Proteins in equal volumes of culture fluids of wild-type and codY mutant strains were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to an Immobilon-P membrane, and probed with rabbit antibody to alpha-toxin. Lanes 1 to 6 contained culture fluids from SA564 (wild type [WT]) and CDM7 (SA564 ΔcodY::ermC); lanes 7 to 8 contained culture fluids from UAMS-1(wild type) and MS1 (UAMS-1 ΔcodY::ermC). Culture fluids in lanes 1 and 2 were from exponentially growing cells (exp), culture fluids in lanes 3 and 4 were from postexponentially growing cells (pxp), and culture fluids in lanes 5 through 8 were from stationary-phase cells (stat). The bands appearing above 50 kDa are of unknown origin but appear to be unrelated to alpha-toxin.

FIG. 3.

Effect of a ΔcodY mutation on hla transcription. hla transcription values represent quantified reverse transcripts normalized to total 16S rRNA from primer extension analysis of the hla promoter. RNA was extracted from cells harvested in the exponential (E) or stationary (S) phases of growth. (A) SA564 (wild type) and CDM7 (SA564 ΔcodY::ermC); (B) UAMS-1 (wild type) and MS1 (UAMS-1 ΔcodY::ermC).

FIG. 4.

Effect of a ΔcodY mutation on transcription of the agr locus. Transcript analysis was done on RNA extracted from exponential-phase (E) or stationary-phase (S) cells. (A) RNAIII transcription values represent quantified reverse transcripts normalized to total 16S rRNA from primer extension analysis of the RNAIII promoter of the strains UAMS-1 (wild type [WT]) and MS1 (UAMS-1 ΔcodY::ermC). (B) RT-PCR analysis of RNAII, RNAIII, and 16S rRNA from strains UAMS-1, MS1, SA564 (wild type), and CDM7 (SA564 ΔcodY::ermC) harvested during the exponential phase (E) or the stationary phase (S).

FIG. 5.

In vitro static biofilm formation. All strains were grown in TSB medium supplemented with glucose and sodium chloride. The percent biofilm formation was determined as described in Materials and Methods after 24 h of growth of the strains UAMS-1 (wild type), MS1 (UAMS-1 ΔcodY::ermC), SA564 (wild type), and CDM7 (SA564 ΔcodY::ermC). The non-biofilm-producing strain S. carnosus TM300 was used as a negative control. One-hundred percent biofilm production was defined as that produced by S. epidermidis strain 1457.

FIG. 6.

Relative PIA accumulation. Samples were taken throughout growth and prepared for PIA release. A PIA immunoblot assay was performed on samples prepared from equal cell numbers, and the results were quantified. PIA accumulation of a wild-type sample at t = 2 h was set as a 100% standard for relative PIA accumulation. UAMS-1 (wild type) (░⃞) and MS1 (UAMS-1 ΔcodY::ermC) (▪).

FIG. 7.

Effect of a ΔcodY mutation on transcription of the ica locus. Transcripts corresponding to icaA, icaR, and 16S rRNA were analyzed by RT-PCR in samples extracted from exponential-phase (E) or stationary-phase (S) cells of strains UAMS-1 (wild type), MS1 (UAMS-1 ΔcodY::ermC), SA564 (wild type), and CDM7 (SA564 ΔcodY::ermC). The upper band that occurs in the RT-PCR analysis of icaA is likely to reflect nonspecific RT products. WT, wild type.