Insights into the CtrA regulon in development of stress resistance in obligatory intracellular pathogen Ehrlichia chaffeensis

Abstract

Ehrlichia chaffeensis is an obligate intracellular bacterium that causes human monocytic ehrlichiosis. Ehrlichiae have a biphasic developmental cycle consisting of dense-cored cells (DCs) and reticulate cells (RCs). Isolated DCs are more stress resistant and infectious than RCs. Here, we report that a response regulator, CtrA was upregulated in human monocytes at the late growth stage when DCs develop. E. chaffeensis CtrA bound to the promoters of late-stage transcribed genes: ctrA, ompA (peptidoglycan-associated lipoprotein), bolA (stress-induced morphogen) and surE (stationary-phase survival protein), which contain CtrA-binding motifs, and transactivated ompA, surE and bolA promoter-lacZ fusions in Escherichia coli. OmpA was predominantly expressed in DCs. E. chaffeensis binding to and subsequent infection of monocytes were inhibited by anti-OmpA IgG. E. chaffeensis BolA bound to the promoters of genes encoding outer surface proteins TRP120 and ECH_1038, which were expressed in DCs, and transactivated trp120 and ECH_1038 promoter-lacZ fusions. E. chaffeensis bolA complemented a stress-sensitive E. coli bolA mutant. E. coli expressing E. chaffeensis SurE exhibited increased resistance to osmotic stress. Our results suggest that E. chaffeensis CtrA plays a role in co-ordinating development of the stress resistance for passage from the present to the next host cells through its regulon.

Fig. 1. DCs are more resistant to oxidative stress than RCs

The percentage of viable bacteria in two size groups: >0.5 μm and <0.5 μm corresponding to RCs and DCs, respectively after the oxidative stress induced by paraquat treatment. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Images show the viability of individual bacteria of diverse sizes: green, live bacteria; red, dead bacteria. White arrowheads point to live DCs (<0.5 μm). Bar = 1 μm.

Fig. 2. CtrA is upregulated at the late stage of E. chaffeensis intracellular growth and in DCs

A. RNA samples prepared from synchronously cultured E. chaffeensis in THP-1 cells at different time points were subjected to quantitative RT-PCR analysis. The value reflects bacterial 16S rRNA normalized against human gapdh mRNA, relative to the amount determined at 0 h p.i. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. B. Upper panel: RNA samples prepared from synchronously cultured E. chaffeensis in THP-1 cells at different time points were subjected to quantitative ctrA RT-PCR analysis and normalized against bacterial 16S rRNA. Relative values to the amount at 0 h p.i. are shown. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) compared to the value at 0 h p.i. by analysis of variance. Lower panel: CtrA amounts in samples at different time points were determined by Western blot analysis. All samples were normalized against bacterial 16S rRNA as determined by quantitative RT-PCR. The sample at 72 h p.i. was diluted 10 times. The numbers below the panel indicate relative densities of the protein bands. ^#Estimated band density without dilution. C. The percentage of CtrA-positive bacteria in two size groups: >0.5 μm and <0.5 μm. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Immunofluorescence images show labeled individual bacteria in different sizes. Host cell-free bacteria were double labeled with dog anti-E. chaffeensis serum (Ech, red) and rabbit anti-CtrA serum (CtrA, green). The merged image (merge) is viewed with green and red filters. White arrowheads point to CtrA-positive DCs (<0.5 μm). Bar = 1 μm. D. Immunogold labeling of CtrA of E. chaffeensis in the inclusions of THP-1 cells. Ultrathin sections of infected cells were labeled with rabbit anti-CtrA or pre-immune serum followed by gold-conjugated goat anti-rabbit IgG. DCs and RCs are indicated by arrowheads and arrows respectively. Bar = 1 μm.

Fig. 3. Comparison of E. chaffeensis CtrA with CtrAs from other members of α-proteobacteria

Alignment of CtrA of E. chaffeensis (E. cha) (GenBank accession number, YP_507798), B. abortus (B.abo) (AAL86376), C. cresentus (C. cre) (NP_421829), R. capsulatus (R. cap) (ADE85408), R. centenum (R. cen) (YP_002297962), and S. meliloti (S. mel) (AAG01017). Identical amino acids are shaded. The helix-turn-helix DNA-binding motif is boxed by a dashed line and the sequence for DNA-sequence-recognition α3 helix is boxed by a solid line. Amino acid numbers of each CtrA are shown on the right.

Fig. 4. ompA, bolA, and surE are upregulated at the late stage of intracellular growth

RNA samples prepared from synchronously cultured E. chaffeensis in THP-1 cells at different time points were subjected to quantitative RT-PCR analysis and normalized against bacterial 16S rRNA. Relative values to the amount at 0 h p.i. are shown. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) compared to the value at 0 h p.i. by analysis of variance.

Fig. 5. E. chaffeensis CtrA binds to the promoter regions of ctrA, ompA, bolA, and surE, and transactivates the promoter-lacZ fusions in E. coli

A. EMSA for GST-CtrA binding to the promoter regions of ompA, bolA, surE, and ctrA. The length (bp) of the probe is shown above each panel. For each panel: biotinylated DNA probe (2 nM) was incubated alone (lane 1), with GST-CtrA (20 nM, lane 2), or GST-CtrA in the presence of 50-fold excess of the corresponding unlabeled DNA competitor (lane 3). Shifted bands are indicated by arrowheads. B. EMSA for GST-CtrA binding to the consensus CtrA-binding motifs in the E. chaffeensis ctrA promoter. The name of each probe (24 bp) is shown above. For each probe: biotinylated DNA probe (2 nM) was incubated alone (lane 1), with GST (20 nM, lane 2), GST-CtrA (20 nM, lane 3), or GST-CtrA in the presence of 50-fold excess of the corresponding unlabeled DNA competitor (lane 4). Shifted bands are indicated by an arrowhead. C. EMSA for dose-dependent binding of GST-CtrA to the bolA promoter. The length of the probe derived from the bolA promoter region is 171 bp. Biotinylated DNA probe (2 nM) was incubated alone (lane 1), with GST (200 nM, lane 2), or GST-CtrA at different concentrations (lanes 3-8, 1, 2, 6, 20, 60, and 200 nM, respectively). Shifted band is indicated by an arrowhead. D. Alu I-digested E. chaffeensis DNA was pulled down with GST-CtrA. The PCR was performed using the pulled down DNA as a template. The promoter region and the length (bp) amplified by specific primers are shown above each panel. For each panel: bacterial total DNA (lane 1, positive control), DNA pulled down with GST-CtrA (lane 2), and DNA pulled down with GST-BolA (lane 3, negative control) were used as templates. E. E. coli strains containing pET-33b(+) encoding His-CtrA and His-NtrYHKD were transformed with the ompA, bolA, surE, or glnA-lacZ fusions. The fusions contain the promoter regions of ompA (368 bp), bolA (171 bp), surE (430 bp), or glnA (243 bp). After inducing the His-CtrA or His-NtrYHKD with IPTG, β-galactosidase activity was measured. Upper panel: β-galactosidase activity (Miller units). His-NtrYHKD, E. coli strain expressing His-NtrYHKD; His-CtrA, E. coli strain expressing His-CtrA. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Lower panel: Western blot analysis of samples from the β-galactosidase assays was performed using monoclonal antibody against the His-tag to verify the expression of His-CtrA and His-NtrYHKD. A representative of three independent experiments is shown. Each lane received 1 μg protein.

Fig. 6. OmpA expression is upregulated at the late stage of intracellular growth and in DCs

A. OmpA amounts in synchronously cultured E. chaffeensis in THP-1 cells at different time points were determined by Western blot analysis. All samples were normalized against bacterial 16S rRNA as determined by quantitative RT-PCR. B. The percentage of OmpA-positive bacteria in two size groups: >0.5 μm and <0.5 μm. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Immunofluorescence images show labeled individual bacteria of different sizes. Host cell-free bacteria were double labeled with dog anti-E. chaffeensis serum (Ech, red) and rabbit anti-OmpA IgG (OmpA, green). The merged image (merge) is viewed with green and red filters. White arrowheads point to OmpA-positive DCs (<0.5 μm). Bar = 1μm. C. Immunogold labeling of OmpA in E. chaffeensis DCs in THP-1 cells. Ultrathin section of infected cells was labeled with rabbit anti-OmpA IgG followed by gold-conjugated goat anti-rabbit IgG. Bar = 1 μm.

Fig. 7. E. chaffeensis animals develop antibodies against OmpA, and anti-OmpA IgG reduces bacterial binding and infection

A. E. chaffeensis infection induces antibody against OmpA in vivo. Purified His-OmpA (1 μg) was subjected to SDS-PAGE (12% acrylamide) and transferred to a PVDF membrane, which was then incubated with rabbit anti-OmpA IgG (lane 1), sera from E. chaffeensis-infected dogs (3918815, CTUALJ, 1425, and OH001) (lane 2-5, respectively) or normal dog serum (lane 6). His-OmpA is indicated by an arrow. B. Inhibition of E. chaffeensis binding to THP-1 cells with rabbit anti-OmpA IgG. Host-cell free bacteria were pre-incubated with culture medium, or 20, 100, 500 μg ml^-1 of pre-immune IgG or anti-OmpA IgG, and then incubated with THP-1 cells. Bound bacteria were counted in 100 THP-1 cells. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) compared to medium alone or pre-immune IgG controls by analysis of variance. C. Inhibition of E. chaffeensis infection of THP-1 cells with rabbit anti-OmpA IgG. The value reflects bacterial 16S rRNA under the indicated conditions relative to culture medium. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) compared to medium alone or pre-immune IgG controls by analysis of variance.

Fig. 8. E. chaffeensis bolA complements an E. coli bolA mutant

Upper panel: the percentage of bacterial growth (CFU ml^-1) after 24 h treatment with 0.4% SDS relative to the detergent-less culture, and the proportion of round (length to width ratio < 1.5) bacteria in total 100 bacteria after the detergent stress. wt, wild-type strain; +bolA, strain overexpressing E. coli BolA; ΔbolA, bolA knockout mutant. pUC19-Ech bolA, pUC19 containing E. chaffeensis bolA gene downstream of the E. coli bolA promoter. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Lower panels: bacteria after detergent stress stained with Diff-Quik. Bar = 2 μm.

Fig. 9. The expression of TRP120 and ECH_1038 is upregulated at the late stage of intracellular growth and in DCs

A. The percentage of TRP120-positive bacteria in two size groups: >0.5 μm and <0.5 μm. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Immunofluorescence images show labeled individual bacteria of different sizes. Host cell-free bacteria were double labeled with dog anti-E. chaffeensis serum (Ech, red) and rabbit anti-TRP120 serum (TRP120, green). The merged image (merge) is viewed with green and red filters. White arrowheads point to TRP120-positive DCs (<0.5 μm). Bar = 1 μm. B. RNA samples prepared from synchronously cultured E. chaffeensis in THP-1 cells at different time points were subjected to quantitative ECH_1038 RT-PCR analysis and normalized against bacterial 16S rRNA. Relative values to the amount at 0 h p.i. are shown. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) compared to the value at 0 h p.i. by analysis of variance. C. Purified His-ECH1038N and antibody specificity. Purified His-ECH1038N (0.5 μg) was subjected to SDS-PAGE (10% acrylamide) followed by Coomassie brilliant blue staining (lane 1) and Western blot analysis using rabbit anti-ECH1038N serum (lane 2). His-ECH1038N is indicated by an arrowhead. The lysates prepared from uninfected (lane 3) and E. chaffeensis-infected (Lane 4) THP-1 cells (7.5 × 10⁵) were subjected to SDS-PAGE (6% acrylamide) followed by Western blot analysis using rabbit anti-ECH1038N serum. Native ECH_1038 is indicated by an arrow. D. The percentage of ECH_1038-positive bacteria in two size groups: >0.5 μm and <0.5 μm. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Immunofluorescence images show labeled individual bacteria in different sizes. Host cell-free bacteria were double labeled with dog anti-E. chaffeensis serum (Ech, red) and rabbit anti-ECH1038N serum (ECH_1038, green). The merged image (merge) is viewed with green and red filters. White arrowheads point to ECH_1038-positive DCs (<0.5 μm). Bar = 1μm.

Fig. 10. E. chaffeensis BolA binds to the promoter regions of trp120 and ECH_1038, and transactivates the promoter-lacZ fusions in E. coli

A. EMSA for GST-BolA binding to the promoter regions of trp120 and ECH_1038. The length (bp) of the probe is shown above each panel. For each panel: biotinylated DNA probe (2 nM) was incubated alone (lane 1), with GST-BolA (2 nM, lane 2), or GST- BolA in the presence of 50-fold excess of the corresponding unlabeled DNA competitor (lane 3). Shifted bands are indicated by arrowheads. B. EMSA for dose-dependent binding of GST-BolA to the ECH_1038 promoter. The length of the probe derived from the ECH_1038 promoter region is 196 bp. Biotinylated DNA probe (2 nM) was incubated alone (lane 1), with GST (200 nM, lane 2), or GST-BolA in different concentrations (lanes 3-8: 1, 2, 6, 20, 60, and 200 nM, respectively). Shifted band is indicated by an arrowhead. C. Alu I-digested E. chaffeensis DNA was pulled down with GST-BolA. The PCR was performed using the pulled down DNA as a template. The promoter region and the length (bp) amplified by specific primers are shown above each panel. For each panel: bacterial total DNA (lane 1, positive control), DNA pulled down with GST-BolA (lane 2), and DNA pulled down with GST-CtrA (lane 3, negative control) were used as templates. D. E. coli strains containing pET-33b(+) encoding His-BolA or His-NtrYHKD were transformed with the trp120 and ECH_1038-lacZ fusions. The fusions contain the promoter regions of trp120 (245 bp) and ECH_1038 (199 bp). After inducing His-BolA and His-NtrYHKD with IPTG, β-galactosidase activity was measured. Upper panel: β-galactosidase activity (Miller units). His-NtrYHKD, E. coli strain expressing His-NtrYHKD; His-BolA, E. coli strain expressing His-BolA. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Lower panel: Western blot analysis of samples from the β-galactosidase assays was performed using monoclonal antibody against the His-tag to verify the expression of His-BolA and His-NtrYHKD. A representative of three independent experiments is shown. Each lane received 1 μg protein.

Fig. 11. E. chaffeensis SurE confers stress resistance to E. coli and has phosphatase activity

A. Upper panel: the percentage of bacteria growth (CFU ml^-1) after 3 h treatment with 0.5 M NaCl relative to that without NaCl. His-NtrYHKD, E. coli strain expressing His-NtrYHKD; His-SurE, E. coli strain expressing His-SurE. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. *Significantly different (P < 0.05) by Student's t-test. Lower panel: Western blot analysis of samples under stress treatment was performed using monoclonal antibody against the His-tag to verify the expression of His-SurE and His-NtrYHKD. A representative of three independent experiments is shown. Each lane received 1 μg protein. B. pH dependence of His-SurE phosphatase activity with 5′-AMP as a substrate. The reaction mixture contained 1 mM 5′-AMP, 1 mM Mg²⁺, 0.1 μg His-SurE. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. C. Metal dependence of His-SurE phosphatase activity with 5′-AMP as a substrate. The reaction mixture contained 1 mM 5′-AMP, 0.1 μg His-SurE, 1 mM Mg²⁺ or 0.1 mM other metal as indicated, 0.1 mM Tris-HCl, pH 6.8. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates. D. Phosphatase activity of His-SurE and His-NtrYHKD (a negative control) dose response with 5′-AMP as a substrate. The reaction mixture contained 1 μg His-SurE or His-NtrYHKD, 1 mM Mg²⁺, 0.1 mM Tris-HCl, pH 6.8. Data indicate mean values ± standard deviations from three independent experiments performed in triplicates.