Staphylococcus aureus CcpA affects virulence determinant production and antibiotic resistance

Abstract

Carbon catabolite protein A (CcpA) is known to function as a major regulator of gene expression in different gram-positive organisms. Deletion of the ccpA homologue (saCOL1786) in Staphylococcus aureus was found to affect growth, glucose metabolization, and transcription of selected virulence determinants. In liquid culture, deletion of CcpA decreased the growth rate and yield; however, the effect was only transient during the exponential-growth phase as long as glucose was present in the medium. Depletion of glucose and production of lactate was delayed, while the level of excretion of acetate was less affected and was even higher in the mutant culture. On solid medium, in contrast, growth of the DeltaccpA mutant resulted in smaller colonies containing a lower number of CFU per colony. Deletion of CcpA had an effect on the expression of important virulence factors of S. aureus by down-regulating RNAIII, the effector molecule of the agr locus, and altering the transcription patterns of hla, encoding alpha-hemolysin, and spa, encoding protein A. CcpA inactivation markedly reduced the oxacillin resistance levels in the highly methicillin-resistant S. aureus strain COLn and the teicoplanin resistance level in a glycopeptide-intermediate-resistant S. aureus strain. The presence of CcpA in the capsular polysaccharide serotype 5 (CP5)-producing strain Newman abolished capsule formation and decreased cap operon transcription in the presence of glucose. The staphylococcal CcpA thus not only is involved in the regulation of carbon metabolism but seems to function as a modulator of virulence gene expression as well.

FIG. 1.

Schematic representation of the ccpA region of S. aureus and of the strategy used to obtain MST23 (COLn ΔccpA). The genetic organization of the S. aureus COL ccpA region (A), pMR1 (B), pMR2 (C), and MST23 (D) is shown. Open reading frame notations and nucleotide numbers correspond to those of the respective genomic regions of strain COL (GenBank accession no. NC_002951). Primers used to amplify the ccpA region (A), the ccpA-flanking regions including the backbone of pEC1 (B), and the respective PCR products (dotted lines) are indicated.

FIG. 2.

Growth characteristics of COLn (squares) and its ΔccpA mutant MST23 (triangles) in LB supplemented with 10 mM glucose. (A) Absorbances at 600 nm (A₆₀₀) and CFU over the growth cycle. (B) Cfu per colony of COLn and MST23 grown on sheep blood agar for 48 h at 35°C. (C) Glucose, acetate, and lactate concentrations in the culture supernatants corresponding to panel A. The data presented are mean values of three independent experiments. (D) Scanning electron microscopy images of COLn and MST23 adhering to polyethylene terephthalate (Thermanox) disks.

FIG. 3.

Susceptibility of COLn and NM143 and their isogenic ΔccpA mutants. (A) Effects of ccpA inactivation and complementation on oxacillin resistance. The methicillin-resistant strain COLn, its ΔccpA mutant MST23, control strains MST31 (COLn pAW17) and MST35 [COLn ΔccpA::tet(L) complemented with control plasmid pAW17], and strain MST36 [COLn ΔccpA::tet(L) complemented with plasmid pMST1] were swabbed along a plate containing an oxacillin gradient. The corresponding MICs of oxacillin (in micrograms per milliliter as determined by E-Test and broth microdilution) for the strains are indicated. (B) Population analysis profiles of COLn (squares) and MST23 [COLn ΔccpA::tet(L); triangles] on oxacillin. (C) Population analysis profiles of the step-selected glycopeptide-intermediate-resistant S. aureus derivative NM143 (squares) and its ΔccpA mutant KS30 (triangles) on teicoplanin.

FIG. 4.

Northern blot analyses of COLn and its ΔccpA mutant MST23 during growth. (A) Growth characteristics of COLn (squares) and MST23 (triangles) grown in HEPES-buffered LB (open symbols) and HEPES-buffered LB supplemented with 10 mM glucose (closed symbols). At 1-h intervals, an aliquot (2 ml) was removed, the absorbance at 600 nm was measured, and the pH in the culture supernatants (circles) was determined. The results presented are mean values of at least three independent experiments. Time points of sampling for the Northern blot analyses are indicated by arrows. (B) Transcription of ccpA, hla, pckA, RNAIII, and spa in COLn and MST23 during growth in HEPES-buffered LB (−Glc) and in HEPES-buffered LB supplemented with 10 mM glucose (+Glc). Relevant transcript sizes and time points of sampling are indicated. Ethidium bromide-stained 16S rRNA patterns are shown as an indication of RNA loading.

FIG. 5.

Effect of glucose addition on the gene expression of COLn and its ΔccpA mutant MST23. (A) Northern blot analyses of ccpA, hla, pckA, and spa. Cells were grown in HEPES-buffered LB to midexponential-growth phase (A₆₀₀ = 1), cultures were split in half, and 10 mM glucose was added to one half (+Glc) while the other half was left unchanged (−Glc). Relevant transcript sizes and time points of sampling are indicated. Ethidium bromide-stained 16S rRNA patterns are shown as an indication of RNA loading. (B) Putative CREs identified upstream of ccpA, hla, pckA, and spa. Nucleotides fitting with the CRE consensus of B. subtilis (38) are highlighted in bold type. Inverted repeats are indicated by arrows. Ambiguity codes are as follows: W denotes A or T, R denotes G or A, and N denotes A, C, G, or T.

FIG. 6.

Capsule production and cap expression of Newman and its ΔccpA mutant MST14 in response to glucose. (A) CP5 expression determined by indirect immunofluorescence of strain Newman and its isogenic ΔccpA mutant grown for 24 h at 37°C in HEPES-buffered LB (−Glc), or in HEPES-buffered LB supplemented with 10 mM glucose (+Glc). Bacteria were stained with 4′,6′-diamidino-2-phenylindole (DAPI), and marked with CP5-specific monoclonal antibodies and stained with Cy3-conjugated anti-mouse antibodies (CY3). (B) Quantitative transcript analysis of capA by LightCycler RT-PCR of strain Newman and its isogenic ΔccpA mutant grown for 8 h at 37°C in HEPES-buffered LB (−Glc) or in HEPES-buffered LB supplemented with 10 mM glucose (+Glc). Transcripts were quantified in reference to the transcription of gyrase (in copies per copy of gyrB). Values from two separate RNA isolations and two independent RT-PCRs each were used to calculate the mean expression (±standard errors of the mean). Glc, glucose; asterisk, P < 0.05 for Newman without Glc, MST14 without Glc, and MST14 with Glc.