Figure 1.
Detection of CovR phosphorylation in vitro and in vivo.
(A and B) Recombinant CovR and CovR-D53A proteins were incubated without (−) and with (+) acetyl-phosphate. Samples were run on a 12% native polyacrylamide gel (A) or a 12% Phos-tag-SDS polyacrylamide gel (B), respectively. In both cases the higher bands represent the phosphorylated species. The D53A change reduces the negative charge and leads to decreased migration of the protein under native conditions as shown by the higher bands for both lanes in panel (A) compared to unphosphorylated CovR [22]. For (B), samples marked with H were heated for 1 min at 100°C, which removes the heat-labile aspartate phosphorylation. (C) Western blot using anti-CovR antibody of indicated samples separated on a 10% Phos-tag-SDS polyacrylamide gel. CovR∼P indicates recombinant CovR incubated with acetyl-phosphate. Cell lysates were derived from cells grown to mid-exponential phase in THY. H indicates sample was heated whereas CIP indicates treatment with calf intestinal phosphatase. (D) Western blot using anti-CovR antibody as described for (C) except cells were grown in THY supplemented with 15 mM MgCl2. (E) Western blot using anti-CovR antibody of cell lysate derived from GBS cells grown to mid-exponential phase in THY and the respective controls as described for (C). For (C–E), cell lysates were loaded at 70 µg of total protein per sample.
Figure 2.
Mass spectra showing that CovR is phosphorylated on threonine 65.
Recombinant CovR was incubated with StkKD and then subjected to chymotrypsin digestion. (A) Mass spectra of the MS/MS fragmentation of a 2+ ion of the CovR-derived peptide EVTphosRRLQTEKTTY. Fragment marked with blue arrow shows the peptide that has lost phosphoric acid, thus decreasing the apparent mass from 852.92 to 804.27. The peptide marked with the red star corresponds to the y102+ fragment (RRLQTEKTTY) and has not lost phosphoric acid. Thus, T3 must be the position of phosphorylation. (B) Mass spectra of the MS/MS/MS fragmentation of the y102+ ion confirmed that the y102+ fragment is indeed RRLQTEKTTY. Fragments of corresponding b or y ions are indicated.
Figure 3.
CovR dual-site phosphorylation events influence each other.
(A) Western blot using anti-CovR antibody. Samples were separated on a 12% Phos-tag-SDS polyacrylamide gel which was loaded as follows: CovR phosphorylated with acetyl-phosphate as positive control, unphosphorylated CovR as negative control, StkKD alone, StkKD in Mg2+/ATP buffer or Mn2+/ATP buffer; CovR-T65A phosphorylated with StkKD in Mg2+/ATP or Mn2+/ATP buffer; CovR phosphorylated with StkKD in Mg2+/ATP or Mn2+/ATP buffer. Phosphorylated and unphosphorylated CovR species are labeled. (B) Recombinant wild type and variant CovR proteins were incubated with kinase buffer in the absence (−) and presence (+) of acetyl-phosphate, separated on a 12% Phos-tag SDS polyacrylamide gel and blotted with anti-CovR antibodies. Note that CovR-D53A and T65E variants cannot be phosphorylated by acetyl-phosphate. (C) Recombinant wild type and variant CovR proteins were incubated with kinase buffer in the absence (−) and presence (+) of acetyl-phosphate or StkKD, separated on a 12% Phos-tag SDS polyacrylamide gel and blotted with anti-CovR antibody. The retarded bands represent the phosphorylated CovR species. The * denotes unspecific background noted for lanes 4 and 5. Note that the CovR-T65A and T65E variants cannot be phosphorylated by StkKD.
Figure 4.
Growth curves and general characterization of GAS CovR/S variants.
Various GAS strains were created as described in Materials and Methods. (A) Photographs of indicated strains following overnight growth on 5% sheep blood agar plates. (B) Growth curves of various strains in THY. Arrow shows point where samples were obtained for RNA-Seq analysis. (C) Production of hyaluronic acid capsule following growth in THY. Significant elevation in hyaluronic acid production is indicated by * in comparison to strain MGAS10870 and by ** in comparison to strains MGAS10870, 10870ΔcovS, and CovR-T65A as measured by ANOVA followed by Bonferroni's post-hoc test. (D) Casein hydrolysis (marker of SpeB) activity. * indicates significant decrease in casein hydrolysis in strain 10870ΔcovS compared to the wild-type and CovR variant strains as measured by ANOVA followed by Bonferoni's post-hoc test. Strain 10870ΔspeB is included as a negative control. For (B–D), strains were grown in triplicate on three separate occasions. Data graphed are mean ± standard deviation.
Figure 5.
RNA-Seq analysis of various GAS strains.
Six strains were grown in quadruplicate to mid-exponential phase. (A) Heat-map of log2 transcript levels of selected genes for the indicated strains relative to strain MGAS10870. Color value scheme is shown below the figure. (B) Pairwise comparison of log2 transcript values relative to the wild-type strain MGAS10870 for the indicated strains. Each dot represents a particular gene. Values for strain 10870ΔcovR are plotted along the X-axis whereas values for strain CovR-D53A (red) and CovR-T65E (blue) are plotted along the Y-axis. The dotted lines indicate the 1.5-fold difference which was on e of our criteria for statistically significant transcript levels. (C) Transcript level data relative to strain MGAS10870 are plotted as in panel A with values for strain 10870ΔcovR plotted along with X-axis and values for strain 10870ΔcovS (red dots) and CovR-T65A (blue dots) plotted along the Y-axis.
Figure 6.
Strains MGAS10870 and CovR-T65A respond to changes in Mg2+.
For all panels, data shown are gene transcript levels as measured by TaqMan quantitative real-time PCR (qRT-PCR). Strains were grown in normal THY (low Mg2+) or in THY with 15 mM MgCl2 (high Mg2+) to mid-exponential phase in duplicate on two separate occasions and analyzed in duplicate. Data shown are mean ± standard deviation of 8 data points. (A) Log2 ratio of transcript levels of indicated genes during growth in low vs. high Mg2+ medium in strain MGAS10870. P value refers to Student's t-test comparing gene transcript levels in low vs. high Mg2+ medium. (B–F) Data shown are transcript levels of indicated genes in indicated strains relative to the endogenous control gene tufA. * indicates significant difference in gene transcript for the indicated strain between growth in a low Mg2+ vs. high Mg2+ conditions as determined by Student's t-test.
Figure 7.
CovR-D53A and CovR-T65E bind CovR-regulated promoters with similar affinity to unphosphorylated CovR.
Binding of recombinant CovR, CovR-D53A, and CovR-T65E proteins to the promoters of (A) hasA and (B) prtS. Increasing concentrations of unphosphorylated CovR, CovR-D53A, and CovR-T65E (monomer concentration given in µM) were used as indicated. Samples were incubated for 15 min at 37°C and electrophoresed on a 6% TBE polyacrylamide gel for 60 min at 120 V. The gels were stained with ethidium bromide. ns, non-specific DNA, f, free DNA, cI, lower molecular weight complex, cII, higher molecular weight complex. Gels shown are representatives of identical results obtained on two separate occasions.
Figure 8.
CovR phosphorylation affects GAS virulence and gene expression during infection.
(A) 20 CD-1 mice per strains were challenged intra-peritoneally with 1×107 colony forming units of the indicated strains and followed for near-mortality. Data graphed are survival. P values refer to Mantel-Cox (log-rank) test adjusted for multiple comparisons. For (B–F), RNA was isolated from infected animals as described in the Material and Methods and converted to cDNA. Data graphed are gene transcript levels of the indicated genes relative to the endogenous control gene tufA. Four animals were infected with each strain and data were analyzed in triplicate. Data graphed are mean ± standard deviation of 12 data points. * indicates that gene transcript level for that particular strain was significantly different compared to all other strains as determined by analysis of variance (ANOVA) and Bonferroni's post-hoc test. For panel (C), ** indicates that the transcript level of ska was significantly lower in strains 10870ΔcovS and CovR-T65A compared to strains 10870ΔcovR, CovR-D53A, and CovR-T65E.
Figure 9.
CovR levels are decreased in the CovR-T65A strain.
(A) Western blot analysis. Indicated strains were grown to mid-exponential phase in regular THY medium (−) and in THY medium supplemented with 15 mM MgCl2 (+). Cell lysates containing 70 µg of total protein were loaded on a 12% SDS-gel. The gel was divided into two parts; the upper part was blotted against anti-CovR antibody while the lower part was blotted against anti-HPr antibody which served as a loading control. (B) qRT-PCR analysis of covR transcript levels. For “Low MgCl2” and “High MgCl2” samples, the indicated strains were grown as for the Western blot analysis in duplicate on two separate occasions and analyzed in duplicate. Data shown are mean ± standard deviation of 8 data points. For the “During infection” samples, four animals were infected with each strain and data were analyzed in triplicate as described in Figure 8. Data graphed are mean ± standard deviation of 12 data points. For all three conditions the lack of a bar for 10870ΔcovR strain indicates no covR transcript level was detectable. * designates a significant difference in covR transcript level between strain MGAS10870 and CovR-T65A as determined by Student's t-test.
Table 1.
Strains and plasmids used in this work.