IQGAP1 Mediates Hcp1-Promoted Escherichia coli Meningitis by Stimulating the MAPK Pathway

Abstract

Escherichia coli-induced meningitis remains a life-threatening disease despite recent advances in the field of antibiotics-based therapeutics, necessitating continued research on its pathogenesis. The current study aims to elucidate the mechanism through which hemolysin-coregulated protein 1 (Hcp1) induces the apoptosis of human brain microvascular endothelial cells (HBMEC). Co-immunoprecipitation coupled with mass spectrometric (MS) characterization led to the identification of IQ motif containing GTPase activating protein 1 (IQGAP1) as a downstream target of Hcp1. IQGAP1 was found to be up-regulated by Hcp1 treatment and mediate the stimulation of HBMEC apoptosis. It was shown that Hcp1 could compete against Smurf1 for binding to IQGAP1, thereby rescuing the latter from ubiquitin-dependent degradation. Subsequent study suggested that IQGAP1 could stimulate the MAPK signaling pathway by promoting the phosphorylation of ERK1/2, an effect that was blocked by U0126, an MAPK inhibitor. Furthermore, U0126 also demonstrated therapeutic potential against E. coli meningitis in a mouse model. Taken together, our results suggested the feasibility of targeting the MAPK pathway as a putative therapeutic strategy against bacterial meningitis.

Keywords: Escherichia coli; HCP1; IQGAP1; MAPK pathway; meningitis.

Figure 1

Transport of Hcp1 into the cytoplasm of HBMEC cells leads to increased apoptosis. (A) TUNEL assays were conducted on HBMEC cells treated with PBS solution (Mock), wild-type RS218 (RS218), Δhcp1 mutant of RS218 (Δhcp1), Hcp1 (Hcp1+), or Hcp2 (Hcp2+). Left panel: TUNEL-positive cells were indicated by the green fluorescence. DAPI was used as a counterstain to mark the nuclear regions of the cells. Right panel: Relative mean integrated density value (IDV) of TUNEL signal (green) from three independent experiments. Scale bar, 50 μm. ^*P < 0.05, ^**P < 0.01. (B) Fluorescence microscopy showing the internalization of Hcp1 by HBMEC cells. The presence of the Hcp proteins (left: Hcp1; right: Hcp2) was detected by rabbit anti-His antibody and indicated by the red fluorescence. Scale bar, 50 μm. (C) Localization of Hcp1 (top) and Hcp2 (bottom) in the treated HBMEC cells as determined by Western blotting. UAP56 and β-actin were used as nuclear and cytosolic marker, respectively.

Figure 2

IQGAP1 and ARHGAP24 can interact with Hcp1 in HBMEC cells. (A) Putative binding partners of Hcp1 as identified by MS analysis. Proteins that could potentially bind to Hcp1 were co-immunoprecipitated by anti-His antibody, separated by SDS-PAGE and identified by MS. The matched epitope IgG was used as a negative control in the co-immunoprecipitation assay. (B) Detection of IQGAP1 (middle left) and ARHGAP24 (bottom left) in the Hcp1 immune complexes by anti-IQGAP1 and anti-ARHGAP24, respectively. IgG was used as a negative control. No binding of Hcp1 to either MYH9 (top right) or VIM (middle right) was observed.

Figure 3

IQGAP1 was involved in the modulation of HBMEC cell apoptosis by Hcp1. (A) Up-regulation of both IQGAP1 (top) and ARHGAP24 (middle) in HBMEC cells incubated with wild-type RS218, RS218 mutant, or Hcp1 as demonstrated by Western blotting. β-actin was used as an internal standard. (B,C) Knockdown of IQGAP1 (left) and ARHGAP24 (right) by specific siRNAs. Quantitation of the relative mRNA and protein expression levels of IQGAP1 and ARHGAP24. (D) The expression levels of Hcp1 (panel 1), IQGAP1 (panel 2), ARHGAP24 (panel 3), and cleaved caspase 8 (panel 4) were measured by Western blotting (E) Immunofluorescence assay showing the uptake level of Hcp1 in HBMEC cells transfected with Scr-siR, IQGAP1-siRNA, or ARHGAP21-siRNA. Scale bar, 50 μm. (F) The expression levels of IQGAP1 (panel 2) and cleaved caspase 8 (panel 3) in Hcp1-treated HBMEC cells that were not transfected, transfected with Scr-siR, with IQGAP1-siRNA, or co-transfected with IQGAP1-siRNA and FLAG-tagged IQGAP1. (G) TUNEL assay showing the effect of IQGAP1 knockdown on reducing Hcp1-promoted apoptosis of HBMEC cells. Co-transfection of the IQGAP1-overexpression plasmid was shown to inhibitory effect of IQGAP1 knockdown on cell apoptosis. The average value ± s.d. of three separate experiments was plotted. ^**P < 0.01. Scale bar, 50 μm.

Figure 4

Hcp1 can inhibit ubiquitin-dependent degradation of IQGAP1 by preventing it from binding to Smurf1 through substrate competition. (A) mRNA expression levels of IQGAP1 in untreated and Hcp1-treated HBMEC cells. (B) HBMEC cells incubated with MG132, a proteasome-inhibiting agent, exhibited a higher cellular level of IQGAP1 compared to the untreated control. (C) Western blot showing that both MG132 and Smurf1 knockdown could significantly increase the protein level of IQGAP1 compared to the untreated control (left). The decrease of Smurf1 level as a result of siRNA transfection was verified by anti-Smurf1 antibody (right). (D) Co-immunoprecipitation experiment confirming the interaction between Smurf1 and IQGAP1. Proteins precipitated by anti-Smurf1 antibody were separated by SDS-PAGE and detected using anti-IQGAP1 antibody. IgG was used as an isotype control. (E) Hcp1 can compete against Smurf1 for binding to IQGAP1. HBMEC cells were incubated with increasing concentrations of Hcp1. Cell lysate samples (Input, right) were subjected to co-immunoprecipitation by anti-Smurf1 antibody, followed by Western blot detection using anti-IQGAP1 antibody (left). The fraction of Smurf1 binding IQGAP1 decreased in a dose-dependent manner with regard to the concentration of Hcp1 used. The increasing percentages of IQGAP1 binding to Hcp1 were verified by co-immunoprecipitation with anti-His antibody, followed by Western blot detection using anti-IQGAP1 antibody (left). Scale bar, 50 μm.

Figure 5

IQGAP-promoted phosphorylation of ERK1/2 is responsible for Hcp1-induced apoptosis of HBMEC cells. (A) Determination of the protein levels of total and phosphorylated P38 (panel 1–2), JNK (panel 3–4), and ERK1/2 (panel 5–6) in untreated (left) and Hcp1-treated (right) HBMEC cells. (B) IQGAP1 knockdown inhibited the Hcp1-induced up-regulation of cleaved caspase 8 and ERK1/2 phosphorylation compared to the Hcp1-treated control or Scr-siRNA-transfected cells. (C) TUNEL assay indicating that U0126, a MEK inhibitor, could significantly reduce cell apoptosis resulting from Hcp1 treatment (top). HBMEC cells treated with U0126 clone showed no significant apoptosis compared to the untreated control. The results are also shown in a column chart (P < 0.05) (bottom). (D) Western blot showing that both phosphorylated ERK1/2 and cleaved caspase 8 levels decreased in cells co-incubated with Hcp1 and U0126, compared to the ones treated with Hcp1 alone. The average value ± s.d. of three separate experiments was plotted. ^*P < 0.05, ^***P < 0.001. Scale bar, 50 μm.

Figure 6

U0126 can protect murine brain from E. coli meningitis. (A) Schematic diagram illustrating the experimental design. (B) Inflammatory cell infiltration was evaluated by anti-Ly6G antibody(red) and H&E staining (black arrow). Injection of E. coli stimulated neutrophil infiltration in murine brain tissues (E. coli +); however, U0126 was shown to be able to effectively reduce E. coli-induced brain infection. IF scale bar, 100 μm, H&E staining scale bar, 10 μm. (C) Measurement of the concentrations of six inflammatory cytokines in murine brain tissues of four groups. (D) A mechanistic model illustrating the role of IQGAP1 in Hcp1-induced apoptosis of HBMEC cells. Hcp1 competes against Smurf1 for binding to IQGAP1, thereby rescuing the latter from ubiquitin-dependent degradation. On the other hand, IQGAP1 can promote the phosphorylation of ERK1/2, a key factor in the MAPK pathway responsible for cell apoptosis. The average value ± s.d. of three separate experiments was plotted. ^*P < 0.05, ^**P < 0.01.