Regulatory role of endoplasmic reticulum resident chaperone protein ERp29 in anti-murine β-coronavirus host cell response

Abstract

Gap junctional intercellular communication (GJIC) involving astrocytes is important for proper CNS homeostasis. As determined in our previous studies, trafficking of the predominant astrocyte GJ protein, Connexin43 (Cx43), is disrupted in response to infection with a neurotropic murine β-coronavirus (MHV-A59). However, how host factors are involved in Cx43 trafficking and the infection response is not clear. Here, we show that Cx43 retention due to MHV-A59 infection was associated with increased ER stress and reduced expression of chaperone protein ERp29. Treatment of MHV-A59-infected astrocytes with the chemical chaperone 4-sodium phenylbutyrate increased ERp29 expression, rescued Cx43 transport to the cell surface, increased GJIC, and reduced ER stress. We obtained similar results using an astrocytoma cell line (delayed brain tumor) upon MHV-A59 infection. Critically, delayed brain tumor cells transfected to express exogenous ERp29 were less susceptible to MHV-A59 infection and showed increased Cx43-mediated GJIC. Treatment with Cx43 mimetic peptides inhibited GJIC and increased viral susceptibility, demonstrating a role for intercellular communication in reducing MHV-A59 infectivity. Taken together, these results support a therapeutically targetable ERp29-dependent mechanism where β-coronavirus infectivity is modulated by reducing ER stress and rescuing Cx43 trafficking and function.

Keywords: Connexin43; ER stress; ERp29; MHV-A59; astrocytes; gap junctional intercellular communication (GJIC); murine coronavirus (mCoV); β-coronavirus infection.

Figure 1

Retention of Cx43, decrease in ERp29 and increase in BiP, ATF6 in response to MHV-A59 infection by primary astrocytes. A–D, localization of Cx43 in mock (A) and MHV-A59 infected primary mouse astrocytes at MOI 2 (B), 5 (C), and 10 (D), 24 h p.i. analyzed by double immunolabeling followed by confocal immunofluorescence microscopy with anti-Cx43 (red) and antinucleocapsid (N) antibodies (green). Small Arrows and large arrows in insets show punctate Cx43 labeling at the cell–cell interfaces (A) or intracellular Cx43 retention in MHV-A59–infected astrocytes (B–D). E, scatter plot of ERp29 mRNA in mock or MHV-A59–infected astrocytes measured by qPCR. F and G, representative immunoblot and corresponding scatter plot of ERp29 protein in mock or MHV-A59–infected astrocytes. H, scatter plot of BiP mRNA in mock or MHV-A59 infected astrocytes measured by qPCR. I and J, representative immunoblot and corresponding scatter plot of BiP protein in mock or MHV-A59–infected astrocytes. K and L, representative immunoblot and corresponding scatter plot of ATF6 protein in mock and MHV-A59–infected astrocytes. γ-Actin was used as a loading control. Values are mean ± SEM analyzed by unpaired Student’s t test or one-way ANOVA (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, ns – not significant, N = 3). ATF6, activating transcription factor 6; BiP, binding immunoglobulin protein; MHV-A59, mouse hepatitis virus strain A59; MOI, multiplicities of infection; p.i., postinfection.

Figure 2

4-PBA rescues Cx43 trafficking to the cell surface in MHV-A59–infected primary astrocytes. A–H, representative confocal photomicrographs of primary murine astrocytes infected with MHV-A59 at MOI 5 (A–D) or MOI 10 (E–H) in either the presence or absence of 4-PBA for 24 h. The cells were fixed, immunolabeled for Cx43 (red) and viral N protein (green) and counterstained with DAPI (blue). Arrows indicate Cx43 localization in the intracellular compartment in absence of 4-PBA (A–H: left panel and inset) and surface localization of Cx43 in presence of 4-PBA (A–H: right panel and inset). I and J, scatter plots showing quantification of punctate areas in infected astrocytes with and without 4-PBA. Data from five independent experiments were used for quantification. Values are mean ± SEM, analyzed by unpaired Student’s t test (∗p < 0.05, ∗∗∗p < 0.001, N = 5). 4-PBA, 4-phenylbutyrate; MHV-A59, mouse hepatitis virus strain A59; MOI, multiplicities of infection.

Figure 3

4-PBA upregulates ERp29 and reduces BiP expression in primary astrocytes infected with MHV-A59.A, scatter plot of ERp29 mRNA in MHV-A59–infected astrocytes (MOI 5) in the absence or presence of 4-PBA as measured by qPCR. B and C, representative immunoblot and corresponding scatter plot of ERp29 protein in MHV-A59–infected astrocytes (MOI 5) in the absence or presence of 4-PBA. γ-Actin was used as a loading control. D, scatter plot of BiP mRNA in MHV-A59–infected astrocytes (MOI 5) in the absence or presence of 4-PBA as measured by qPCR. E and F, representative immunoblot and corresponding scatter plot of BiP protein in MHV-A59–infected astrocytes (MOI 5) in the absence or presence of 4-PBA. G, scatter plot of ERp29 mRNA in MHV-A59–infected astrocytes (MOI 10) in the absence or presence of 4-PBA as measured by qPCR. H and I, representative immunoblot and corresponding scatter plot of ERp29 protein in MHV-A59–infected astrocytes (MOI 10) in the absence or presence of 4-PBA. γ-Actin was used as a loading control. J, scatter plot of BiP mRNA in MHV-A59–infected astrocytes (MOI 10) in the absence or presence of 4-PBA as measured by qPCR. K and L, representative immunoblot and corresponding scatter plot of BiP protein in MHV-A59–infected astrocytes (MOI 10) in the absence or presence of 4-PBA. Values are mean ± SEM analyzed by unpaired Student’s t test (∗ p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, ns – not significant, N = 3). 4-PBA, 4-phenylbutyrate; BiP, binding immunoglobulin protein; MHV-A59, mouse hepatitis virus strain A59; MOI, multiplicities of infection.

Figure 4

4-PBA increases ERp29 and Cx43 GJIC in DBT-WT cells. A and B, immunoblot and respective scatter plot showing significant upregulation of ERp29 protein level (relative intensity) in control and 4-PBA-treated DBT-WT cells. C and D, immunoblot and respective scatter plot showing significant upregulation of Cx43 protein in control and 4-PBA-treated DBT-WT cells. E, representative images of control and 4-PBA-treated DBT-WT cells immunolabeled with anti-Cx43 (red) and counterstained DAPI, showing cellular localization of Cx43 (arrows). F, fluorescence photomicrographs showing Cx43 gap junction assembly in control (upper panels) and 4-PBA-treated DBT-WT cells (lower panels) following in-situ Triton X-100 solubilization and immunolabeling with anti-Cx43 (green). Arrow shows Triton X-100 insoluble Cx43 gap junction plaques. G and H, photomicrographs and scatter plot showing a significant increase in functional GJIC in 4-PBA-treated DBT cells compared to untreated controls. Arrow indicates the direction and extent of Lucifer yellow dye transfer (green). Values are mean ± SEM analyzed by unpaired student’s t test (∗∗p < 0.01, ∗∗∗∗p < 0.0001, N = 3 or 5). 4-PBA, 4-phenylbutyrate; DAPI, 4, 6-diamidino-2-phenylindole, DBT, delayed brain tumor; GJIC, gap junctional intercellular communication.

Figure 5

MHV-A59–induced ER stress and downregulated ERp29 in DBT-WT cells. A, representative microscopic image of DBT-WT cells infected with MHV-A59 at MOI 2 for 6, 9, and 12 h, immunostained with antinucleocapsid (N) (red) counterstained with DAPI, showing viral infection and profuse syncytia formation with increasing time p.i. B–E, scatter plots showing relative mRNA expression of BiP (B), both Xbp-total and Xbp-spliced normalized with Gapdh (C), Xbp-spliced normalized with Xbp-total (D), and viral nucleocapsid (N) (E) in MHV-A59 infected DBT-WT cells at 6 and 9 h p.i. compared to mock infected controls. F and G, representative immunoblot and corresponding scatter plot showing increased BiP protein in MHV-A59 infected DBT cells at 6 and 9 h p.i. compared to mock infected controls. H, scatter plot showing relative ERp29 mRNA expression in DBT-WT cells infected with MHV-A59 at 6 h and 9 h p.i. compared to mock infected controls. I and J, representative immunoblot and scatterplot showing significant downregulation of ERp29 protein level at 6 h p.i. only. K–M, representative immunoblot and scatterplot showing significant upregulation of ATF6 and p-eIF2α protein level at both 6 h p.i. and 9 h p.i. γ-Actin was used as a loading control. Values are mean ± SEM analyzed by unpaired Student’s t test or one-way ANOVA (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, ns – not significant, N = 3). ATF6, activating transcription factor 6; BiP, binding immunoglobulin protein DAPI, 4, 6-diamidino-2-phenylindole; DBT, delayed brain tumor; MHV-A59, mouse hepatitis virus strain A59; MOI, multiplicities of infection.

Figure 6

4-PBA reduces MHV-A59 infectivity in DBT-WT cells. A–F, representative fluorescence photomicrographs and scatter plot showing MHV-A59 induced host cell–cell fusion in DBT-WT cells in the absence or presence of 4-PBA at 6 h (A and B), 9 h (C and D), and 12 h p.i. (E and F). Arrows indicate the area of syncytia in control and 4-PBA-treated cells. Scatter plots (B, D, and F) represent a fusion index calculated from 10 fields each from three independent experiments. G, kinetics of virus induced cell-cell fusion at 6, 9, and 12 h p.i. in DBT-WT cells with (red line) or without (black line) 4-PBA. H, viral load in MHV-A59–infected DBT-WT cells in the absence or presence of 4-PBA as determined by qPCR for viral N mRNA at 6 h and 9 h p.i. I, differential growth kinetics of MHV-A59 in DBT-WT cells with (red line) or without (black line) 4-PBA treatment determined by calculating viral titer at 0 h, 1.5 h, 3 h, 6 h, and 9 h p.i. Values are mean ± SEM analyzed by unpaired Student’s t test or one-way ANOVA (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ns – not significant, N = 3). 4-PBA, 4-phenylbutyrate; MHV-A59, mouse hepatitis virus strain A59; DAPI, 4, 6-diamidino-2-phenylindole; DBT, delayed brain tumor.

Figure 7

Exogenous ERp29 expression increases Cx43 trafficking to the cell surface and attenuates ER stress. A, representative images of DBT-ERp29 and DBT-WT cells immunolabeled with anti-Cx43 (red) and counterstained with DAPI. Arrows indicate intracellular staining of Cx43 in DBT-WT cells (upper panel), and arrowheads indicate punctate Cx43 staining at cell surface in EGFP-positive DBT-ERp29 cells (lower panel). B and C, immunoblot and respective scatter plot showing significant upregulation of Cx43 protein in DBT-ERp29 and DBT-WT cells. D and E, immunoblot and respective scatter plot showing Triton X-100 soluble (Sol) and insoluble (InSol) fractions of Cx43. F–H, scatter plots showing relative expression of BiP (F), Xbp-total, and Xbp-spliced normalized with Gapdh (G) and Xbp-spliced normalized with Xbp-total (H) mRNA in DBT-WT and DBT-ERp29 cells upon tunicamycin treatment. Tun, tunicamycin. Values are mean ± SEM and analyzed by unpaired Student’s t test or one-way ANOVA (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, ns – not significant, N = 3). DAPI, 4, 6-diamidino-2-phenylindole; DBT, delayed brain tumor.

Figure 8

Exogenous ERp29 expression reduces MHV-A59 infectivity and replication in DBT cells.A–F, representative confocal images and scatter plots showing host cell–cell fusion in DBT-WT and DBT-ERp29 cells following MHV-A59 infection at 6 h (A and B), 9 h (C and D), and 12 h (E and F) p.i. The perimeter of viral syncytia is indicated by the dashed area, arrows show restricted syncytia formation by DBT-ERp29. Scatter plots (B, D, and F) represent fusion index calculated from 10 fields, each from three independent experiments. G, plot represents kinetics of virus induced cell–cell fusion across 6, 9, and 12 h p.i. in DBT-WT versus DBT-ERp29 cells. H, viral load in MHV-A59–infected DBT-WT and DBT-ERp29 cells was examined by qPCR for viral N mRNA levels at 6 h and 9 h p.i. I, differential growth kinetics of MHV-A59 in DBT-WT (black line) versus DBT-ERp29 cells (red line), estimated by calculating viral titer at 0 h, 1.5 h, 3 h, 6 h, and 9 h p.i. Values are mean ± SEM analyzed by unpaired Student’s t test or one-way ANOVA (∗p < 0.05, ∗∗∗p < 0.001, ∗∗∗∗p < 0.001, ns – not significant N = 3). DAPI, 4, 6-diamidino-2-phenylindole; DBT, delayed brain tumor; MHV-A59, mouse hepatitis virus strain A59.

Figure 9

Exogenous ERp29 expression show reduced ER stress in response to MHV-A59 infection.A–C, scatter plot showing relative expression of BiP (A), Xbp-total and Xbp-spliced normalized with Gapdh (B), and Xbp-spliced normalized with Xbp-total (C) in DBT-WT and DBT-ERp29 cells infected with MHV-A59 at 6 h p.i. BiP, Xbp(t) and Xbp(s) mRNA levels show significant downregulation in DBT-ERp29 cells with respect to DBT-WT cells upon MHV-A59 infection. D and E, immunoblot and respective scatter plot of BiP protein in MHV-A59 infected DBT-ERp29 cells compared to DBT-WT cells. F, G, and H, immunoblot and respective scatter plot of ATF6 and p-eIF2α protein in MHV-A59–infected DBT-ERp29 cells compared to DBT-WT cells. γ-Actin was used as a loading control. Values are mean ± SEM analyzed by unpaired Student’s t test (∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, N = 3). ATF6, activating transcription factor 6; BiP, binding immunoglobulin protein; DBT, delayed brain tumor; MHV-A59, mouse hepatitis virus strain A59.

Figure 10

Inhibition of Cx43 mediated gap junction channels increased viral susceptibility of DBT-ERp29 cells. Representative confocal images and corresponding scatter plots showing host cell–cell fusion in DBT-ERp29 cells. A–I, the cells were either untreated controls (A, D, and G) or treated with Cx43 mimetic peptide GAP 26 (B, E, and H) or GAP 27 (C, F, and I) followed by MHV-A59 infection for 9 h (A–C), 12 h (D,E,F), and 24 h (G–I) immunolabeled with anti-N (red) and counterstained with DAPI in DBT cell line expressing exogenous EGFP-ERp29 (green). J–L, scatter plots (J–L) represent fusion index at 9 h (J), 12 H (K), and 24 h (L), respectively, calculated from 10 fields, each from three independent experiments. Values are mean ± SEM analyzed by unpaired Student’s t test (∗p < 0.05, ∗∗p < 0.01, ns – not significant, N = 3). DBT, delayed brain tumor.