Research ArticleCell Biology

The β4 subunit of Cav1.2 channels is required for an optimal interferon response in cardiac muscle cells

See allHide authors and affiliations

Science Signaling  11 Dec 2018:
Vol. 11, Issue 560, eaaj1676
DOI: 10.1126/scisignal.aaj1676

Channeling antiviral immunity?

The intracellular β4 subunit of the voltage-gated calcium channel Cav1.2 plays a major role in the proper localization and function of the pore-forming α subunit, but it has also been implicated in regulating gene expression in neurons and lymphocytes. Tammineni et al. found that β4 abundance and nuclear localization were increased in a cardiac cell line in response to treatment with type I interferon (IFN) or infection with dengue virus. The β4 subunit physically interacted with the transcriptional regulator IRF7 and was associated with the increased expression of IFN-stimulated genes. In cells overexpressing the β4 subunit, dengue virus replication was inhibited. Together, these data suggest that the β4 subunit may play a channel-independent role in the antiviral response.

Abstract

The auxiliary β4 subunit of the cardiac Cav1.2 channel plays a poorly understood role in gene transcription. Here, we characterized the regulatory effects of the β4 subunit in H9c2 rat cardiac cells on the abundances of Ifnb mRNA [which encodes interferon-β (IFN-β)] and of the IFN-β–related genes Ddx58, Ifitm3, Irf7, Stat2, Ifih1, and Mx1, as well as on the abundances of the antiviral proteins DDX58, IRF7, STAT2, and IFITM3. Knocking down the β4 subunit in H9c2 cells reduced the expression of IFN-β–stimulated genes. In response to inhibition of the kinase JAK1, the abundances of β4 subunit mRNA and protein were decreased. β4 subunit abundance was increased, and it translocated to the nucleus, in cells treated with IFN-β, infected with dengue virus (DENV), or transfected with poly(I:C), a synthetic analog of double-stranded RNA. Cells that surrounded the virus-infected cells showed translocation of β4 subunit proteins to nuclei in response to spreading infection. We showed that the β4 subunit interacted with the transcriptional regulator IRF7 and that the activity of an Irf7 promoter–driven reporter was increased in cells overexpressing the β4 subunit. Last, overexpressing β4 in undifferentiated and differentiated H9c2 cells reduced DENV infection and decreased the abundance of the viral proteins NS1, NS3, and E-protein. DENV infection and poly(I:C) also increased the concentration of intracellular Ca2+ in these cells. These findings suggest that the β4 subunit plays a role in promoting the expression of IFN-related genes, thereby reducing viral infection.

INTRODUCTION

Voltage-gated Cav1.2 channels, which conduct L-type Ca2+ currents, are ubiquitous membrane proteins that regulate various cellular processes, including the generation of action potentials, muscle fiber contraction, and synaptic transmission in excitable cells. In the heart, Cav1.2 channels play a fundamental role in excitation-contraction coupling by enabling a brief influx of extracellular Ca2+, leading to a transient increase in the concentration of cytosolic Ca2+ ([Ca2+]i), which in turn stimulates the release of Ca2+ from intracellular stores (1). The Cav1.2 channel is a heterodimeric complex composed of a pore-forming α1 subunit and an intracellular β subunit and is associated with other auxiliary subunits (28). The β subunit increases the abundance of the α1 subunit (Ca2+ pore) at the cell surface and modulates the Ca2+ current properties associated with channel opening (912). Four distinct β subunit types have been reported in excitable cells (β1 to β4) (1318), and all are present in the adult heart at the protein level (19). A previously uncharacterized signaling pathway was described in neurons in which the β4 subunit (hereafter simply named β4) translocates to the nucleus and directly regulates transcription of the gene encoding tyrosine hydroxylase (20). The Cav1.2 channel is also found in nonexcitable cells, such as T lymphocytes, in which it plays relevant physiological roles. β4-deficient T lymphocytes show reduced interferon (IFN) production and impaired nuclear translocation of the transcription factor NFAT (nuclear factor of activated T cells) after activation by anti-CD3 and anti-CD28 antibodies (21). Therefore, we hypothesized that β4 might have a relevant role in the nuclei of heart cells, and we tested this hypothesis in undifferentiated and differentiated H9c2 cells. This cell line has been used to investigate the ontogenic properties of Cav1.2 channels, and it acquires a cardiac phenotype upon differentiation. H9c2 cells are derived from rat cardiac tissue and preserve many elements of the electrical and hormonal pathway found in adult cardiac cells; thus, they are a good model for studying the regulation of cardiac Cav1.2 channels (22). These cells are also a good model in which to study viral infection in the heart. We previously showed that these cells are susceptible to infection by dengue virus (DENV) (23), a single-stranded positive RNA virus of the Flaviviridae family, the pathogenic agent that causes dengue fever, which can damage the heart, liver, lungs, and other organs and is potentially fatal (2426). Previous work showed that Ca2+ homeostasis is disturbed by viral proteins (27, 28). Furthermore, increased [Ca2+]i has been observed in skeletal muscle cells from DENV-infected patients (25), and critical virus-host interactions depend on cellular Ca2+-regulated proteins; however, the involvement of Cav1.2 channels in viral infection of the heart is unclear.

Here, we examined the role of β4 on the expression of several genes encoding key proteins with antiviral activity, including the genes that encode DDX58 (DEAD box polypeptide 58), a downstream signaling molecule in the retinoic acid (RA)–inducible gene I pathway; IFITM3 (IFN-induced transmembrane 3), which helps to confer antiviral immunity; STAT2 (signal transducer and activator of transcription 2), a transcription activator that forms homodimers and heterodimers in response to IFN and translocates to the nucleus; and IRF7 (IFN regulatory factor 7), a key mediator of IFN-β responses that plays a role in the transcriptional activation of virus-inducible cellular genes. We also showed that β4 interacted with IRF7 and that β4 increased the activity of an Irf7 promoter–driven reporter and that there were changes in the abundance and localization of β4 in H9c2 cells in response to IFN-β, poly(I:C) (polyinosinic:polycytidylic acid), and infection with DENV. Last, we examined the role of β4 on the susceptibility of cells to DENV infection and showed how overexpression of β4 affected the mRNA expression and protein abundances of the viral proteins NS1, NS3, and envelope protein (E-protein), as well as changes in resting [Ca2+]i during infection. To test whether β4 played an antiviral role, we overexpressed β4 in undifferentiated and differentiated H9c2 cells, analyzed the localization of overexpressed β4 by Western blotting and confocal microscopy, and then tested the effects of β4 overexpression on DENV infection, viral RNA synthesis, and viral protein abundance. The effects of knocking down β4 on the abundance of key antiviral proteins and of the inhibition of Janus kinase 1 (JAK1) on β4 actions were also examined. Digital gene expression sequencing analysis and quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis were used to examine the effects of β4 on the abundances of Ddx58, Irf7, Stat2, and Ifitm3 mRNAs, and Western blotting analysis was performed to detect the effects of β4 on the abundance of the proteins encoded by these genes. We hypothesized that the relative abundance and localization of β4 would be altered by viral infection and IFN and that β4 expression would affect infection response.

RESULTS

β4 is localized in the cytosol and nuclei of H9c2 cells

Because the nuclear localization of the β4 subunit would be expected if this channel subunit played a role in gene expression, we used Western blotting and confocal microscopy techniques to analyze its localization in H9c2 cells. Western blotting analysis showed that β4 protein was present in both the cytosolic and nuclear fractions of untransfected control cells (Fig. 1A). Overexpression of the β4 subunit led to a large increase in band density in both fractions. In β4-expressing cells, the antibody used recognized two protein bands in both fractions (Fig. 1A). The lower band had a molecular mass of slightly >55 kDa, as was expected for the β4 subunit (which has a molecular mass of 58 kDa), which was also detected in some control experiments. The second band, which was of higher molecular weight and was consistently detected in control experiments, was possibly due to the presence of β4 protein that had undergone posttranslational modifications. The relative purity of the cytosolic and nuclear fractions was verified with anti-tubulin and anti-histone antibodies, respectively (Fig. 1A). The identity of β4 was verified by Western blotting analysis of different amounts of whole-cell lysates with an antibody from a different source (Fig. 1B). As expected, band density increased as the amount of whole-cell lysate increased. When the amount of protein loaded was low, only the higher–molecular weight band was detected. At higher loads, the lower–molecular weight band became apparent. Consistent with the Western blotting experiments, confocal microscopy analysis demonstrated the presence of β4 in both the nuclear and cytosolic fractions of untransfected H9c2 cells (Fig. 1C) and in cells overexpressing the protein (Fig. 1C). β4 showed a distinct punctate labeling in cell nuclei similar to that observed in a study of Chinese hamster ovary cells transfected with plasmid encoding β4 (29).

Fig. 1 Analysis of the localization of the β4 subunit in H9c2 cells.

(A) H9c2 cells transfected with empty plasmid or with plasmid encoding the β4 subunit were processed to generate cytosolic and nuclear fractions that were analyzed by Western blotting with antibodies against the indicated proteins. Western blots are representative of three experiments. The anti-β4 antibody was obtained from Abcam. (B) H9c2 cells transfected with plasmid encoding the β4 subunit were lysed, and the indicated amounts of cell lysate were analyzed by Western blotting with antibodies against β4 (StressMark) and actin, which served as a loading control. Blots are representative of three independent experiments. (C) H9c2 cells that were left untransfected (top) or were transfected with plasmid expressing the β4 subunit (bottom) were analyzed by confocal microscopy to determine the cellular localization of β4. The β4 subunit was detected with the monoclonal antibody from Abcam (1:100 dilution, top; 1:200 dilution, bottom) and Alexa Fluor 488–conjugated secondary antibody (green). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) counterstained (blue). Images are representative of three independent experiments.

β4 stimulates the expression of genes encoding antiviral proteins

The presence of the β4 subunit in the nucleus suggested that it might play a role in gene transcription. To test this hypothesis, we performed an exploratory experiment overexpressing β4 in H9c2 cells with the aim of obtaining a list of possible genes whose expression was changed by the β4 subunit and performed digital gene expression sequencing analysis. As expected, Cacnb4, the gene encoding the β4 subunit, was highly expressed after transfection of the cells (table S1), as were several genes that encode proteins related to antiviral activity, such as Ddx58, Ifitm3, Irf7, Stat2, Mx1 (IFN-induced guanosine 5′-triphosphate binding protein), and Ifih1 (cytoplasmic sensor of viral nuclear acids), whose expression was also statistically significantly increased. These results then prompted us to validate our findings by qRT-PCR and Western blot analyses.

Consistent with the digital gene expression sequencing analysis, quantification of mRNA abundance by RT-PCR indicated that the abundances of mRNAs for Cacnb4, Irf7, Ifitm3, Ddx58, Stat2, Ifnb1, Mx1, and Ifih1 were statistically significantly increased in cells overexpressing β4 relative to that in control cells (Fig. 2A). Western blotting analysis showed that the abundances of DDX58, IRF7, IFITM3, and STAT2 were also statistically significantly increased in β4-overexpressing cells compared to those in control, vector-transfected cells (Fig. 2, B and C). The abundance of actin protein, which was used as a loading control, was not affected by β4 overexpression (Fig. 2B and fig. S1). Quantification of changes in protein abundance required estimation of the dynamic range of the intensities of the bands in the Western blots. This was assessed by Western blotting analysis of different amounts of the same sample from cells transfected with empty vector or β4-encoding plasmid (fig. S1).

Fig. 2 Effects of β4 on antiviral factors in H9c2 cells.

(A) H9c2 cells transfected with empty plasmid (open bars) or with plasmid expressing the β4 subunit (filled bars) were analyzed by qRT-PCR to determine the relative abundances of mRNAs for the indicated antiviral factors and of Cacnb4 mRNA. Data are means ± SEM of 10 to 14 experiments, and values are shown relative to those in control cells, which were set at 1. (B) Whole-cell lysates of H9c2 cells transfected with empty plasmid (left lanes) or with plasmid expressing the β4 subunit were analyzed by Western blotting with antibodies against the indicated proteins. Actin was used as a loading control. Blots are representative of three to five experiments. (C) Analysis of the relative abundances of DDX58, IRF7, IFITM3, and STAT2 proteins, normalized to that of actin, in H9c2 cells transfected with empty plasmid (empty bars) or with plasmid expressing the β4 subunit (filled bars) from the experiments shown in (B). Data are means ± SEM of three to seven experiments. *P < 0.05, **P < 0.01 compared to control.

IFN-β increases the abundances of β4 mRNA and protein

The antiviral agents regulated by the β4 subunit that we examined earlier (Fig. 2) are IFN-stimulated genes (ISGs) in other systems. Therefore, we tested the hypothesis that IFN-β also increased the abundance of the β4 subunit in these cells. First, we verified that IFN-β increased the abundance of proteins encoded by the ISGs. We found that IFN-β statistically significantly increased the abundances of two ISG-encoded proteins involved in the first line of defense against invading pathogens: IFITM3 (fig. S2, A and B) and STAT2 (fig. S2, A and C). Next, we showed that IFN-β also enhanced the abundance of the β4 subunit as assessed by confocal microscopy analysis (Fig. 3A). An increase in β4 abundance in response to IFN-β was confirmed by Western blotting analysis (Fig. 3B).

Fig. 3 Effect of IFN-β and DENV2 infection on β4 abundance and its nuclear localization in H9c2 cells.

(A) H9c2 cells were left untreated (top) or were treated with IFN-β (500 U/ml) for 48 hours (bottom) before being analyzed by confocal microscopy. Images show the localization of the β4 subunit (green) and DAPI-stained nuclei (blue). Scale bar, 20 μm. Images are representative of three independent experiments. (B) H9c2 cells were left untreated or were treated with IFN-β (500 U/ml) for 48 hours before whole-cell lysates were analyzed by Western blotting with antibodies specific for the β4 subunit or actin, as indicated (inset). The bar graph shows the relative abundance of the β4 subunit normalized to that of actin in the indicated cells. Data are means ± SEM of three experiments. *P < 0.05 compared to control, untreated cells. (C) H9c2 cells were left uninfected (0) or were infected with DENV2 at an MOI of 5 or 10, as indicated, for 48 hours. The cells were then separated into nuclear (left) and cytosolic (right) fractions, which were analyzed by Western blotting with antibodies against the β4 subunit or actin (insets). Bar graphs show the relative abundance of the β4 subunit normalized to that of actin in the indicated cells. Data are means ± SEM of three experiments. *P < 0.05 compared to control, uninfected cells. (D) H9c2 cells that were uninfected or infected with DENV2 at an MOI of 5 were analyzed by qRT-PCR to determine the relative abundance of Cacnb4 mRNA. Data are means ± SEM of three experiments. *P < 0.05 compared to control, uninfected cells. (E) H9c2 cells were left uninfected (top) or were infected with DENV2 at an MOI of 5 (middle) or 10 (bottom) for 48 hours. The cells were then analyzed by confocal microscopy to visualize the localization of the β4 subunit (green) and the NS3 viral protein (red), with DAPI-labeled nuclei (blue). Images are representative of three experiments. Scale bars, 30 μm. Mean fluorescence intensity data and analysis of β4 subunit localization in the cells labeled 1 to 4 are presented in table S2.

DENV infection increases β4 subunit abundance

The effect of β4 on antiviral factors and its regulation by IFN-β suggests that this subunit may be regulated during the cellular response to foreign RNA or DNA. To test this hypothesis, we analyzed the relative abundance and localization of the β4 subunit in DENV-infected H9c2 cells. Western blotting analysis revealed that DENV infection substantially increased the abundance of the β4 subunit in the nuclear and cytosolic fractions (Fig. 3C) in a dose-dependent manner compared with those in uninfected cells. The effect on β4 abundance in the cytosolic fraction was greater at a multiplicity of infection (MOI) of 10 than at an MOI of 5. The abundance of Cacnb4 mRNA in the infected cells was fourfold greater than that in the uninfected cells (Fig. 3D). Consistent with these results, β4 abundance and nuclear translocation increased when H9c2 cells were transfected with poly(I:C) (fig. S3).

To further examine the effects of DENV infection on the abundance of the β4 subunit, immunofluorescence analysis was performed (Fig. 3E). The viral protein NS3 was present in DENV2-infected cells at both tested MOIs (Fig. 3E). In mock-infected cells, β4-labeling was observed in both the cytosol and nucleus with a minor nucleus-referential distribution (Fig. 3E). In DENV-infected cells (Fig. 3E), β4 was predominantly observed in the nuclei of cells close to the infection site (Fig. 3E and table S2). Cells with abundant β4 in the nucleus had a decreased punctate labeling and were poorly infected. Infection not only brought about an overall increase in β4 abundance but also led to its translocation to the nucleus. We found that the nuclear/cytosolic ratio of the β4 subunit was about 2 in mock-infected cells but was much greater (that is, 6 to 11) in cells close to infected cells, whereas it was reduced in cells far from infection sites (table S2).

Knockdown of the β4 subunit decreases ISG expression

To further analyze the effects of the β4 subunit on antiviral factors, we performed β4-silencing experiments. First, we verified that the amounts of β4 mRNA and protein were substantially reduced in cells transfected with the silencing plasmid (Fig. 4, A and B). Next, we tested the effect of β4 knockdown on antiviral factors and found that the abundances of mRNAs for the ISGs Irf7, Stat2, Ifitm3, and Ifnb1 were statistically significantly decreased when the β4 subunit was knocked down (Fig. 4C), consistent with the results of our β4 overexpression experiments (Fig. 2). We also tested the possibility that silencing β4 blocked the increase in the abundance of antiviral proteins in H9c2 cells treated with IFN-β, infected with DENV or treated with poly(I:C). Western blotting analysis demonstrated that silencing β4 indeed decreased the abundances of IRF7, STAT2, and IFITM3 (Fig. 4, D to F).

Fig. 4 Knockdown of the β4 subunit results in the decreased abundance of IRF7, STAT2, IFITM3, and IFN-β mRNA and protein.

(A and B) H9c2 cells transfected with empty plasmid (open bars) or with plasmid encoding β4-specific small interfering RNA (siRNA; filled bars) were subjected to qRT-PCR analysis of the relative abundance of Cacnb4 mRNA (A) and to Western blotting analysis of the relative abundance of β4 subunit protein (B). Data are means ± SEM of three to five experiments. **P < 0.01 compared to control cells. (C) H9c2 cells transfected with empty plasmid (open bars) or with plasmid encoding β4-specific siRNA (filled bars) were subjected to qRT-PCR analysis of the relative abundances of the indicated mRNAs. Data are means ± SEM of three to five experiments. **P < 0.01 compared to control cells. (D to F) Top: H9c2 cells were transfected with empty plasmid (left lanes) or with plasmid encoding β4-specific siRNA (right lanes). Thereafter, the cells were treated with IFN-β (D), infected with DENV2 (E), or treated with poly(I:C) (F). Whole-cell lysates were then analyzed by Western blotting with antibodies against IRF7, STAT2, or IFITM3. Blots are representative of three to five experiments. Bottom: The relative abundances of the appropriate bands were determined by densitometry. Data are means ± SEM of three to five experiments. *P < 0.05, **P < 0.01.

β4 subunit interacts with IRF7 and regulates its promoter activity

We next examined whether β4 interacted with IRF7, which plays a role in the transcriptional activation of virus-inducible genes, including that encoding IFN-β. We overexpressed β4 in H9c2 cells and then immunoprecipitated either β4 or IRF7 from nuclear and cytosolic fractions with the appropriate specific antibodies, which was followed by Western blotting analysis. Two bands were observed upon immunoprecipitating IRF7 from nuclear lysates and blotting with the anti-β4 antibody (Fig. 5A): a 55-kDa band and a second more abundant band that migrated more slowly. A similar although fainter pattern was observed in the cytosolic lysates (Fig. 5A). After immunoprecipitating with the anti-β4 antibody and Western blotting with the anti-IRF7 antibody, a band of less than 55 kDa was detected, which corresponded to IRF7 in both the nuclear and cytosolic extracts, confirming a β4-IRF7 interaction (Fig. 5B). We verified that among the proteins precipitated by the anti-β4 antibody was the β4 protein itself (Fig. 5C). IRF7 is a positive regulator of IFN signaling, and it is autoregulated. This autoregulation contributes to Irf7 induction in response to viral infection; therefore, the increased expression of ISGs by β4 (Fig. 2 and table S1) might be due to the increased activity of the promoter of Irf7 as a result of the binding of β4 to IRF7. To test whether β4 affected the transcriptional activity of Irf7, H9c2 cells were cotransfected with a plasmid encoding a luciferase reporter construct under the control of the Irf7 promoter in combination with an expression vector for β4. The coexpression of β4 resulted in increased Irf7 promoter activity (Fig. 5D).

Fig. 5 β4 and IRF7 interactions.

(A) H9c2 cells transfected with empty plasmid (−) or with plasmid expressing the β4 subunit (+) were lysed, separated into nuclear and cytosolic fractions, and then subjected to immunoprecipitation (IP) with antibody against IRF7 and analyzed by Western blotting (IB) with antibody against the β4 subunit. (B) H9c2 cells transfected with empty plasmid (−) or with plasmid expressing the β4 subunit (+) were lysed, separated into nuclear and cytosolic fractions, and then subjected to immunoprecipitation with antibody against the β4 subunit and analyzed by Western blotting with antibody against IRF7. (C) H9c2 cells transfected with empty plasmid (−) or with plasmid expressing the β4 subunit (+) were lysed, separated into nuclear and cytosolic fractions, and then subjected to immunoprecipitation with antibody against the β4 subunit and analyzed by Western blotting with antibody against β4. Western blots in (A) to (C) are representative of four experiments. (D) H9c2 cells were transfected with the pGL2 vector expressing the Irf7 promoter and the pSG5 plasmid alone (empty bar) or in combination with plasmid expressing the β4 subunit (filled bar). The cells were then subjected to a luciferase reporter assay as described in Materials and Methods. Data are means ± SEM of eight experiments. *P < 0.05.

Inhibition of JAK decreases the abundance of β4

JAK-STAT signaling plays an important role in transducing signals stimulated by IFNs. In response to IFN, STAT proteins act as transactivators. When activated by IFN, the JAK-STAT pathway initiates a signaling cascade leading to the increased expression of ISGs. Therefore, we tested the hypothesis that inhibition of the JAK-STAT pathway might inhibit the effects of β4 on the expression of ISGs. To test this possibility, we assessed the effects of filgotinib, a specific inhibitor of JAK1, on the abundances of IFITM3 and STAT2 in control H9c2 cells and in cells overexpressing β4. We found that filgotinib decreased the abundance of both proteins in control cells, as expected (Fig. 6, A to D). Overexpressing β4 led to the increased abundance of IFITM3 and STAT2, confirming our earlier results (Fig. 2), and the inhibitor blocked this increase (Fig. 6, A to D). Next, we tested the hypothesis that β4 is itself an ISG by investigating the effects of filgotinib on the abundances of endogenous Cacnb4 mRNA and β4 protein. We found that filgotinib substantially decreased both β4 mRNA and protein abundance (Fig. 6, E to G).

Fig. 6 β4 and β4-regulated antiviral factors abundance is mediated by JAK1.

(A to D) H9c2 cells transfected with empty plasmid or with plasmid expressing the β4 subunit were left untreated or were treated with 1 μM filgotinib (GLPG0634) for 48 hours, as indicated. The cells were then lysed and subjected to Western blotting analysis with antibodies against IFITM3 (A) or STAT2 (C). Blots are representative of four to six experiments. The relative abundances of IFITM3 (B) and STAT2 (D) in the indicated samples were determined by densitometry. Data are means ± SEM of five to nine experiments. *P < 0.05, **P < 0.01. (E to G) H9c2 cells were left untreated or were treated with 1 μM filgotinib (GLPG0634) for 24 hours, as indicated. (E) The cells were then lysed and subjected to Western blotting analysis with antibodies against β4. Actin was used as a loading control. Blots are representative of six experiments. (F) The relative abundances of β4 in the indicated samples represented in (E) were determined by densitometry. Data are means ± SEM of five to nine experiments. *P < 0.05. (G) The same cells described in (E) were analyzed by qRT-PCR to determine the relative abundance of Cacnb4 mRNA. Data are means ± SEM of six experiments. **P < 0.01.

β4 inhibits DENV infection in H9c2 cells

The increase in the abundance of antiviral factors in cells expressing β4 (Fig. 2) suggests that this subunit may play a role against infection. To test this hypothesis, we infected H9c2 cells with DENV and then analyzed viral protein and viral RNA synthesis in control and β4-overexpressing cells. Enzyme-linked immunosorbent assay (ELISA) analysis demonstrated a decrease in the nonstructural viral protein NS1 in β4-overexpressing H9c2 cells infected with DENV2 or DENV4 (Fig. 7A). In addition, qRT-PCR experiments showed a reduction in viral RNA abundance in infected β4-overexpressing H9c2 cells (Fig. 7B). Similarly, Western blotting analysis revealed that the abundance of the related viral protein NS3 was also decreased in β4-overexpressing H9c2 cells. Western blotting analysis also showed that the abundance of NS3 was similar in untransfected H9c2 cells infected with DENV2 and in DENV-infected cells transfected with empty plasmid (Fig. 7C), suggesting that transfection had no major effects on NS3 abundance. In contrast, a statistically significant decrease in the density of the NS3 band was observed in cells transfected with plasmid expressing the β4 subunit (Fig. 7C). NS3 abundance largely decreased in β4-overexpressing cells. Consistent with these results, we found that the abundance of the E-protein of the virus also decreased in cells expressing β4 (Fig. 7D). To further characterize the actions of β4 on DENV infection in H9c2 cells, we tested the infective capacity of the supernatant of DENV2-infected H9c2 cells, which is expected to contain mature viral particles, on Vero cells, a well-characterized model of infection (Fig. 7E). The supernatant from DENV2-infected H9c2 cells (control) readily infected Vero cells. However, when the supernatant from β4-expressing, DENV2-infected H9c2 cells was used, the infection of Vero cell was substantially reduced, suggesting that any viral particles in this supernatant were less infective than those particles in the supernatant of the control cells (Fig. 7E). Last, we found that the number of β4-expressing H9c2 cells that were infected by DENV2 and DENV4 was statistically significantly lower than that of control H9c2 cells (Fig. 7, F and G).

Fig. 7 β4 reduces DENV infection.

(A) H9c2 cells transfected with empty plasmid (control) or plasmid expressing the β4 subunit were subjected to infection with the indicated viruses for 48 hours before the amount of NS1 secreted by the cells was determined by ELISA. (B) H9c2 cells transfected with empty plasmid (control) or plasmid expressing the β4 subunit were subjected to infection with DENV4 for 48 hours before the relative abundance of the viral genome was determined by qRT-PCR. (C and D) H9c2 cells transfected with empty plasmid (control) or plasmid expressing the β4 subunit were subjected to infection with DENV2 for 48 hours before the relative abundance of the NS3 protein (C) and viral E-protein (D) was determined by Western blotting (insets). Actin was used as a loading control. Bar graphs show the relative abundance of the indicated proteins as determined by densitometry. (E) Top: Scheme outlining the infection of Vero cells with culture medium from infected H9c2 cells. Bottom: Analysis of the number of focus-forming units of Vero cells 24 hours after the addition of culture medium from control or β4-expressing H9c2 cells. Data are means ± SEM of three to six experiments. *P < 0.05, **P < 0.01. (F and G) H9c2 cells transfected with empty vector plasmid (control) or plasmid expressing the β4 subunit were infected 24 hours later with DENV2 or DENV4 (F and G) at an MOI of 5. (F) Infected cells were examined by microscopy at the indicated magnifications. Images are representative of three experiments. (G) The numbers of DENV2-infected and DENV4-infected cells were then counted. Data are means ± SEM of three experiments. **P < 0.01.

β4 increases the expression of ISGs and inhibits DENV infection in differentiated H9c2 cells

The H9c2 cell line is derived from embryonic rat heart and achieves a more cardiac phenotype when RA is added to the culture medium. Therefore, we assessed whether the β4 subunit regulated ISG expression and played a protective role against infection in differentiated H9c2 cells. First, we confirmed that RA led to cellular differentiation under our experimental conditions and determined the effect of differentiation on β4 mRNA and protein abundance (fig. S4). Next, we tested whether differentiated H9c2 cells could be infected with DENV. In cells exposed to DENV2, we found in four experiments a mean value of the number of copies of the amplified RNA fragment of 11237.8 ± 88.4 SEM and detected NS1 protein with ELISA in two experiments with a mean absorbance value [in arbitrary units (a.u.)] of 0.44, indicating that differentiated cells could be infected with DENV2. In mock-infected cells, no values were detectable by qRT-PCR, and the ELISA mean background absorbance value of two experiments (in a.u.) was 0.05.

Next, we investigated the effects of β4 on antiviral factors and infection in differentiated cells (Fig. 8). Similar to our earlier experiments (Fig. 4D), we found that knocking down the β4 subunit decreased the abundances of STAT2, IFITM3, and IRF7 (Fig. 8, A and B). Similarly, as was observed with undifferentiated H9c2 cells (Fig. 2, B and C), overexpression of β4 substantially increased the abundances of these same proteins (Fig. 8, C and D). Furthermore, we also found that overexpression of β4 in differentiated H9c2 cells protected them from infection. Western blotting analysis revealed that overexpression of β4 resulted in a decrease in the abundance of the viral E-protein (Fig. 8, E and F), whereas ELISA analysis showed the decreased abundance of NS1 (Fig. 8G), and we found a reduced number of DENV2-infected cells in focus assays (Fig. 8H), similar to our findings from experiments with undifferentiated cells (Fig. 7E). Confocal microscopy analysis further confirmed these findings (fig. S5). In β4-silenced H9c2 cells, very little STAT2 was detected compared to that in cells in which the β4 subunit was not knocked down. This observation was made in both differentiated (fig. S5A) and undifferentiated (fig. S5B) cells.

Fig. 8 β4 reduces infection in differentiated H9c2 cells.

(A and B) Differentiated H9c2 cells were transfected with empty plasmid (left lanes and bars) or plasmid expressing β4-specific siRNA (right lanes and bars) and then were treated with 500 U/ml IFN-β for 42 hours. (A) Whole-cell lysates were analyzed by Western blotting with antibodies against the indicated proteins. (B) The relative abundances of the indicated proteins from the experiments represented in (A) were determined by densitometry. Data are means ± SEM of three or four experiments. *P < 0.05, **P < 0.01. (C and D) Differentiated H9c2 cells were transfected with empty plasmid (left lanes and bars) or plasmid expressing β4 (right lanes and bars). (C) Whole-cell lysates were analyzed by Western blotting with antibodies against the indicated proteins. (D) The relative abundances of the indicated proteins from the experiments represented in (C) were determined by densitometry. Data are means ± SEM of three or four experiments. *P < 0.05, **P < 0.01. (E and F) Differentiated H9c2 cells transfected with empty plasmid (left lanes and empty bar) or plasmid expressing β4 (right lanes and filled bar) were infected with DENV2 at an MOI of 5. (E) Whole-cell lysates were then analyzed by Western blotting to detect viral E-protein. Actin was used as a loading control. (F) The relative amounts of E-protein in the cells represented in (E) were determined by densitometry. Data are means ± SEM of four experiments. **P < 0.01. (G) Differentiated H9c2 cells transfected with empty plasmid (left) or plasmid expressing β4 (right) were infected with DENV2 at an MOI of 5 for 24 hours, and the relative amounts of NS1 secreted by the cells were determined by ELISA. Data are means ± SEM of four experiments. **P < 0.01. (H) Mean values of three experiments (± SEM) of focus-forming units in Vero cells were incubated with cell culture taken from DENV2-infected, differentiated H9c2 cells transfected with empty vector (empty bar) or with plasmid expressing the β4 subunit. The number of focus-forming units were then determined as described in Fig. 7E. **P < 0.01.

DISCUSSION

In this study, we demonstrated that the β4 subunit of Cav1.2 channels can act as an antiviral agent. We found that overexpression of β4 increased the amounts of four key antiviral proteins, namely, DDX58, IRF7, IFITM3, and STAT2, and contrasting results were obtained by silencing β4. These effects would be expected to be larger had the whole population of cells been transfected with the β4-expressing or β4-silencing plasmids, which was prevented by the constraints of the transfection procedures. Furthermore, we found that β4 abundance and nuclear translocation were increased when cells were challenged with DENV. As a result of increasing the abundance of β4, the amounts of the DENV viral proteins NS1, NS3, and E-protein were reduced, and we also observed a marked reduction in the number of DENV-infected cells. These actions most likely involve IFN and the classical JAK-STAT pathway (30). JAKs play a key role in downstream signaling when IFNs bind to their receptor (30). IFNs are known stimulators of the expression of genes encoding antiviral effectors and regulators, including DDX58, IRF7, IFITM3, and STAT2 (30, 31). In H9c2 cells, IFNs induce the expression of Ifitm3 (32), and we found an increase in β4 abundance in these cells in response to IFN-β but saw the opposite effect after JAK1 inhibition. A role for a diffusible factor such as IFN was also suggested by our observation of increased β4 abundance and nuclear translocation in noninfected cells that were in proximity to DENV-infected cells.

Several proteins encoded by ISGs reinforce the system by further inducing IFN production (30, 33). A knockout mouse study indicated that IRF7 is necessary for induction of genes encoding IFN-α and IFN-β (34). Thus, when DDX58 interacts with double-stranded RNA (a replication intermediate for RNA viruses), a signaling cascade is triggered, leading to the activation of transcription factors and production of IFNs (35, 36). The increased production of IFN-β after β4 overexpression, the regulation of β4 protein abundance by IFN-β, and the decrease in β4 mRNA and protein abundance observed after inhibition of JAK1 demonstrated in this study suggest a positive regulatory role for the β4 subunit in controlling pathogens. Consistent with this conclusion, we detected an interaction between β4 and IRF7, which plays a leading role in the transcriptional activation of virus-inducible cellular genes, including the gene encoding IFN-β. IRF7 exerts its function in cooperation with other factors (3437), and in this regard, it is interesting to note that we found increased activity of an Irf7 promoter by β4, possibly as a result of the interaction of β4 with IRF7, as shown in our coimmunoprecipitation experiments (Fig. 5). The transcriptional activity of Irf7 promoter is regulated by IRF7 protein, and β4 might enhance this autoregulation. It is likely that there is cooperation between these two proteins during the antiviral response. Together, our findings suggest that the β4 subunit plays a role in the response of cardiac cells to infection by optimizing IRF7 function. A role of β4 as an antiviral agent is consistent with previous work showing defective proliferation of CD4+ T cells in the absence of β4 and a reduction in the amount of IFN-γ produced by β4-deficient CD4+ T cells (21).

Previous roles of the β4 subunit include mediating the trafficking of the α1c subunit, the principal subunit of the Cav1.2 channels, to the plasma membrane in excitable cells and regulating the kinetic properties of Ca2+ currents that flow through these channels when bound to the α1c subunit (912, 38). The effects of the β4 subunit on antiviral factor production that we have demonstrated are likely independent of its involvement in regulating channel-gating properties and trafficking given that those functions involve interactions with the α1c subunit at the plasma membrane, not in the nucleus where the expression of genes is affected.

Although the detailed cellular mechanisms that underlie the triggering of the nuclear translocation of β4 after DENV infection remain to be resolved, it is reasonable to suggest that intracellular Ca2+ could play a role in this process. A Ca2+-dependent translocation of β4 to the nucleus was previously suggested by the experiments of Tadmouri et al. (20), who found that β4 translocation to the nucleus in hippocampal neurons depends on electrical activity. In these neurons, action potentials are accompanied by transient increases in the concentration of cytosolic Ca2+ that are controlled in part by a Ca2+-induced Ca2+ release mechanism, similar to that in heart muscle (39). In H9c2 cells that were infected with DENV or transfected with poly(I:C), we measured increases in [Ca2+]i; however, the transfection process by itself appeared to increase the basal concentration of Ca2+ compared to that in control cells (fig. S6). Our data are consistent with previous findings that, in human patients, DENV infection results in an increased [Ca2+]i in muscle cells, as measured with ion-sensitive microelectrodes (25). Furthermore, a relationship between β4 and [Ca2+]i was shown in immune cells. Badou et al. (21) showed that β4-deficient T cells have abnormally low Ca2+ responses when stimulated by anti-CD3 and anti-CD28 antibodies, as well as exhibiting reduced cytokine production and impaired nuclear translocation of NFAT after stimulation. In T cells, the influx of Ca2+ associated with stimulation occurs through voltage-independent Ca2+ channels (21). The source of the increased Ca2+ in H9c2 cells after DENV infection or poly(I:C) treatment was also voltage independent because we observed no changes in [Ca2+]i in response to electrical stimulation. Cav1.2 channels are nonfunctional in undifferentiated H9c2 cells (22).

The four known β subunits (β1, β2, β3, and β4) have different distributions in the cell (11, 19). β4 is the only one for which a predominant presence in cardiomyocyte nuclei has been described; this observation was reported by Colecraft et al. (11) in rat cardiomyocytes after recombinant adenoviral gene infection used to overexpress the green fluorescent protein (GFP)–fused β4 subunit. On the other hand, Foell et al. (19) observed native β4 expression in the plasma membrane of canine cardiomyocytes. The differing findings between these studies could be due to a species difference. Alternatively, the observation of Colecraft et al. (11) could have reflected nuclear translocation of β4 in response to adenoviral infection, similar to the post-infection translocation effects that we observed in the present study. Note that although we observed that native β4 was more abundant in the nucleus, it was also present in the cytosol of H9c2 cells. Moreover, when β4 was overexpressed, it increased in abundance in both fractions, suggesting that the transfection itself did not have a major effect on the cellular distribution of β4.

The β4 subunit has been linked to transcription in noncardiac cells previously. Hibino et al. (40) found that β4c, a short form of the subunit predominantly expressed in cochlear cells, is recruited to the nucleus and interacts directly with heterochromatin protein 1γ (HP1γ), a nuclear protein involved in gene silencing and transcriptional regulation. The interaction between the β4 subunit and HP1γ markedly reduced the silencing activity of this protein in Cos1 cells. Tadmouri et al. (20) found that β4 localizes predominantly in the nuclei of dentate gyrus neurons, and using a heterologous expression system, the authors showed that it associates with Ppp2r5d, a regulatory subunit of the phosphatase PP2A, which is followed by nuclear translocation of the complex, enabling repression of the Th (tyrosine hydroxylase) gene promoter by β4. The authors proposed that β4 may act as a repressor-recruiting platform to control neuronal gene expression. These two examples demonstrate the modulatory actions of β4 on repressive gene phenomena that are distinct from its stimulatory effects on antiviral factor expression reported here and thus are likely to be mediated by different targets.

Given that STAT2, IRF7, and IFITM3 are important antiviral factors in many cell types against various viruses (41), the anti-DENV protective effects of β4 demonstrated here may extend to other differentiated cell types and other viruses. We found that the β4 subunit also regulated the expression of these ISGs in differentiated H9c2 cells and protected the cells from DENV infection. Furthermore, the results from our experiments with poly(I:C) (fig. S3) suggest that β4 has a protective role not only against DENV but also against other pathogens. The β4 subunit is expressed in excitable cells, including neurons in diverse brain regions (42, 43) and cardiomyocytes (19, 44), as well as in nonexcitable cells, including T cells (45) and cells in the kidney and testis (46). Of the four β4 isoforms (β4a to β4d) that have been described (1216, 18), three are expressed in the heart, namely, isoforms β4a, β4b, and β4d (19), whereas β4c has not been detected. The β4d isoform is truncated because of a frameshift-mediated early stop codon, resulting in an absence of the domain that interacts with the α1c subunit (19, 44). The physiological relevance of β4d is unknown. The remaining β4 isoforms (β4a and β4b) have similar molecular weights, and we could not distinguish one from another in our Western blotting experiments. However, β4b is more predominant in the nucleus and affects gene expression to a greater extent than does the β4a isoform in cultured cerebellar granule cells (47), suggesting that it may have a preferential role in gene regulation. Our findings that expression of the β4b isoform led to a preferential localization of this subunit in the nucleus and to a positive regulation of IFN-related genes are consistent with this view. In conclusion, our experiments suggest that the β4 subunit of Cav1.2 channels in H9c2 cells plays an important role in the cellular response to viral infection. This cellular response involves IFN-β, IRF7, STAT2, and other key antiviral proteins. Our findings open exciting perspectives to understand the role of the β4 subunit of the voltage-dependent Ca2+ channel in the antiviral response.

MATERIALS AND METHODS

Virus stocks

Propagation of two DENV serotypes, DENV serotype 2 New Guinea strain and DENV serotype 4 H241 strain, was performed in CD1 suckling mouse brains (Harlan, México). Titers were determined by plaque assays in BHK-21cells (Instituto Pedro Kuri, Cuba). CD1 suckling mouse brains from mock-infected mice were used as control specimens. The experiments were performed according to protocols approved by the Division Laboratory Animal Unit (CICUAL) at Cinvestav, Mexico, in compliance with state law, federal statute, and Consejo Nacional de Ciencia y Tecnología (CONACYT) policy.

Cell lines and treatments

H9c2 cells, (passages 17 to 24, American Type Culture Collection) were cultured in monolayers in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco), sodium bicarbonate (1.5 g/liter), penicillin (50 IU), and streptomycin (50 μg/ml) under atmospheric conditions of 95% air and 5% CO2 at 37°C in a humidified incubator. Stocks of cell lines were propagated in 60-mm culture dishes for successive passage. Cells were used when they reached 80 to 90% confluence, usually within 24 to 48 hours. Vero cells (passages 18 to 23, American Type Culture Collection) were cultured in monolayers in advanced DMEM supplemented with 2 mM glutamine, penicillin (5 × 104 U/ml), streptomycin (50 mg/ml), 5% fetal calf serum, and amphotericin B (1 ml/liter; Fungizone) under the same atmospheric conditions as for H9c2 cells. Stocks of cell lines were propagated in culture flasks for successive passage. Cell culture medium was replaced with fresh medium every 48 hours. The specific inhibitor of JAK1 (a tyrosine kinase activated by IFN-β) filgotinib (GLPG0634, Med Chem) was dissolved in dimethyl sulfoxide (DMSO) to prepare a stock solution of 1 mM and was used at a final concentration of 1 μM in the culture medium. In β4 overexpression experiments, the inhibitor was applied at 6 hours after transfection and was incubated for a further 24 hours. In experiments investigating endogenous β4, cells were incubated with filgotinib for 48 hours. IFN-β from rat (Sigma-Aldrich) was dissolved in sterile water to prepare a stock solution of 100 U/μl and further dissolved at a final concentration of 500 U/ml in culture medium.

Differentiation of H9c2 cells

Differentiation of H9c2 cells to a more cardiac phenotype was achieved by decreasing the percentage of FBS in the medium to 1%, which was followed by RA supplementation as described elsewhere (48, 49). Briefly, cells were plated in 60-mm tissue culture dishes at a density of 35,000 cells/ml and cultured for 1 day in 10% FBS to allow cell attachment. Thereafter, cells were cultured for 6 days in 1% FBS and 1 μM RA. Replacement of medium was performed daily in the dark. All-trans-RA was prepared in DMSO and stored at −20°C in the dark to avoid degradation.

Transfection

H9c2 cells were transiently transfected with plasmid encoding the β4b subunit cloned by Castellano et al. (18) or empty vector pSG5 plasmid with Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. The medium was replaced after 4 to 6 hours, and cultures were maintained for a total of 48 hours. Cells were transfected with poly(I:C) (50 or 100 μg/ml; Sigma-Aldrich) with Lipofectamine 2000 and then incubated for 24 hours according to the manufacturer’s protocol. Poly(I:C) was also used after cells were transfected with other plasmids. In this case, poly(I:C) was added to the medium without Lipofectamine [poly(I:C) treatment]. Control experiments involved the use of mock transfections with a 0.9% NaCl solution. To knockdown the β4 subunit, cells were transfected as described earlier with SureSilencing shRNA plasmid 3 or 4 from Qiagen (336311) that also expressed GFP. Transfection of differentiated H9c2 cells with plasmids was performed with Lipofectamine 3000 (Invitrogen) according to the manufacturer’s instructions. Transfection efficiency was assessed by flow cytometry 24 hours after the cells were transfected with a plasmid expressing GFP. Transfection efficiencies were 48 and 62% for differentiated and nondifferentiated cells, respectively.

Luciferase reporter assay

A 1330–base pair (bp) fragment of rat genomic DNA corresponding to the human Irf7 promoter (50) was amplified by PCR. The PCR product was cloned into pGL2 Basic and sequenced. Primer sequences were as follows: (forward) 5′-GCCGCTCGAGTCTCCCTGGATGGTATATTCTTATCTGTAC-3′ and (reverse) 5′-CCCAAGCTTGCAAGCAGGGCAGCCATGCTAGGATTAGGG-3′. H9c2 cells were transfected with 500 ng of the Irf7 promoter construct pGL2 plasmid using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. Cotransfections were performed with 1 μg of pSG5 plasmid or constructs of the same plasmid expressing the β4 subunit. Luciferase activity assays were performed 48 hours after transfection using the Steady-Glo Luciferase Assay System (Promega) and a Modulus Single Tube Reader (Turner Biosystems). Luciferase activity was normalized to sample protein concentration. All reporter assays were performed in triplicate.

Infections

For all experiments, cells were washed three times in Hank’s balanced salt solution and infected by exposure to DENV (serotype 2 or 4) at an MOI of 5 or 10 in serum-free medium for 2 hours at 37°C. Cells were washed with acid glycine (pH 3) to inactivate noninternalized virus, washed three times with phosphate-buffered saline (PBS), and then, serum-supplemented DMEM was added. The infection was permitted to proceed for 24 or 48 hours at 37°C.

Immunofluorescence

H9c2 were grown on coverslips in 12-well plates for confocal microscopy. Cells were fixed with 1% formaldehyde 48 hours after infection, incubated for 20 min with permeabilizing solution (that is, PBS, 0.1% saponin, and 1% FBS), and incubated for 2 hours at room temperature with anti-NS3 primary antibody (1:10; GeneTex), followed by Alexa Fluor 555–conjugated goat anti-rabbit secondary antibody (Life Technologies) for 1 hour. To detect β4, monoclonal anti-CACNB4 (1:100 or 1:200; Abcam) primary antibody was applied, which was followed by Alexa Fluor 488–conjugated donkey anti-mouse (Life Technologies) secondary antibody for 1 hour. To detect STAT2, monoclonal anti-STAT2 (1:200; Abcam) primary antibody was applied, which was followed by Alexa Fluor 555–conjugated donkey anti-rabbit (Life Technologies) secondary antibody for 1 hour. Slides were mounted with VECTASHIELD, and labeled cells were observed through a Zeiss LSM700 imager.Z2 laser confocal microscope at room temperature. Images were acquired with a 40× objective (EC-Planar-Neofluar 40×/1.30 oil dichroic M27) and high-resolution camera Axiocam HRm. Image acquisition and analysis were performed with ZEN software version 2010. Confocal techniques were also performed using laser scanning microscopy (Leica TCS-SP8) with argon (488 nm), and helium/neon (543 nm) lasers. Both lasers were used with an optimized pinhole diameter. Confocal images were obtained as z-stacks of single optical sections. Stacks of optical sections were superimposed as a single image with the Leica LAS AF 2.6.0 build 7268 software. To acquire bright-field images, H9c2 cells were grown on coverslips and fixed with 4% formaldehyde for 20 min, washed, and incubated with Hoechst 33342 (1 mg/ml) for 10 min. Coverslips mounted with VECTASHIELD were observed under a Leica DMI6000 microscope equipped with a 700- to 2014-pixel camera, and bright-field images were acquired under a 63× objective.

Cytosolic Ca2+ measurements and fluorescence imaging

H9c2 cells grown on coverslips were loaded with the cell-permeant fluorescent Ca2+ indicator Fura 2-AM (Molecular Probes, Invitrogen) for 45 min at room temperature. Fura 2-AM was diluted in PBS with 1 mM Ca2+ to a final concentration of ~5 μM [from a DMSO stock solution containing 9 mM Fura 2-AM and 25% (w/v) Pluronic F127 (Molecular Probes)]. Cells were washed in PBS for 20 to 30 min at room temperature before performing [Ca2+]i measurements. Ratiometric images of Fura 2 fluorescence were monitored using an Eclipse TE300 microscope (Nikon) equipped with a Polychrome V (TILL Photonics, Germany), which changed excitation wavelengths rapidly between 340 and 380 nm. Fluorescence emissions were captured through a 510WB80 filter (Chroma Technology Corp.) using an iXon EM+DU885 digital camera (Andor Technology). Image acquisition and ratio analysis were performed with Imaging Workbench version 6.0 software (INDEC Biosystems, USA). Resting [Ca2+]i was estimated as Fura 2 fluorescence ratio (340/380).

qRT-PCR assays

Reverse transcription used 1 μg of total RNA from DENV-infected H9c2 cells in each experimental condition and was performed by using random primers (Promega) at a concentration of 0.025 μg/μl and the reverse transcriptase enzyme ImpromII (Promega) at 25°C for 5 min, 42°C for 60 min, and 70°C for 15 min. For real-time PCR amplification, SYBR Fast universal (Kapa) was used in the Eco Illumina System apparatus. A standard curve was generated as described previously (23). The number of copies of the 151-bp RNA fragment encoding the capsid was estimated by interpolation in the standard curve. Total RNA from H9c2 cells transfected with plasmid expressing the β4 subunit or with empty plasmid was isolated with an RNeasy Mini kit (Qiagen). Quantification was made by spectrophotometry (NanoDrop; Thermo Fisher Scientific). RT was performed with 500 ng of deoxyribonuclease-treated RNA in 20-μl reactions. Synthesis of complementary DNA was performed with Superscript III RT (Invitrogen) and random hexamers (250 ng) according to the manufacturer’s instructions. To quantify mRNA, we used TaqMan assays (Applied Biosystems) with an iCycler iQ (Bio-Rad) using the TaqMan Gene Expression Master Mix (4369016) and the following primer-probe sets: β4 (Cacnb4, Rn01449787_m1), α1c (Cacna1c, Rn00709287_m1), IRF7 (Irf7, Rn01450778_g1), IFITM3 (Ifitm3, Rn01479537_s1), DDX58 (Ddx58, Rn01439792_m1), STAT2 (Stat2, Rn01527200_m1), IFNB1 (Ifnb1, IFN-β1, fibroblast, Rn00569434_s1), MX1 [Mx1, Myxovirus (influenza virus) resistance 1, Rn00597360_m1], and IFIH1 (Ifh1; IFN induced with helicase C domain, Rn01514348_m1) were quantified by primers and were stored at −20°C. Eukaryotic 18S ribosomal RNA (Hs99999901_s1) was used as an internal control. Quantification was performed by the 2−ΔΔCt method. This procedure is valid if the amplification efficiencies of the target and reference genes are approximately equal, as was the case in our experiments.

Digital gene expression sequencing

Isolation of RNA from vector-transfected and β4-expressing transfected H9c2 cells was done as described earlier. Samples were processed by LC Sciences. Processing included the generation of a library and sequencing of transcripts and their identification in the rat genome (UCSC rn4). Mappable reads were aligned using Tophat_v1.4.1. Gene expression profiling was conducted with Illumina next-generation sequencing technology (LC Sciences) on the Illumina HiSeq 2000 platform (1 × 50 bp SE, rapid mode). Pairwise differential expression analysis at the isoform level was done, and the abundance was normalized and evaluated in fragments per kilobase of transcript per million mapped reads using the cuffdiff module of cufflinks_v2.02. Differences in the abundance of transcripts in both groups were evaluated using analysis of variance (ANOVA), and a q value < 0.05 was considered statistically significant (q = false discovery rate–adjusted P value) to generate the values in table S1. The analysis included log2 transformation.

Subcellular fractionation, Western blotting, and coimmunoprecipitation analysis

H9c2 cells grown in P100 plates were trypsinized in 0.25% trypsin solution, and cells were collected by centrifugation at 850g for 8 min. Nuclear and cytosolic fractions were separated by a nuclear complex coimmunoprecipitation kit (Active Motif) according to the manufacturer’s instructions. For whole-cell extracts, cells were lysed with lysis buffer [300 mM NaCl, 50 mM tris-HCl (pH 8.0), 1% Triton X-100, and 10% glycerol]. Lysates were stored in liquid nitrogen until further use. Isolated protein content was quantified by Bradford’s method. Equal amounts of proteins were resolved by SDS–polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane (Bio-Rad). In some experiments, we used Bio-Rad ready-made gels, which enable the detection of migration patterns of proteins in a broad range of molecular weights. Membranes were blocked with 4.5% nonfat milk in PBS. Membranes were incubated overnight with primary antibody at 4°C. Thereafter, they were washed in PBS containing 0.1% Tween 20 and incubated in horseradish peroxidase (HRP)–conjugated secondary antibody for 1 hour at room temperature. Chemiluminescence was detected with Immobilon Western reagent (Millipore Co.). For coimmunoprecipitation experiments, nuclear and cytosolic fractions were separated from β4-expressing transfected cells. Coimmunoprecipitation was performed with a nuclear complex coimmunoprecipitation kit (Active Motif) according to the manufacturer’s instructions with minor modifications. Protein extracts were precleared with 1% bovine serum albumin (BSA)–blocked A/G agarose beads (Thermo Fisher Scientific) for 1 hour at 4°C with mild agitation, and then, 1% BSA–blocked agarose beads were incubated with antibody and protein extracts. Some experiments were performed with the Invitrogen Dynabeads coimmunoprecipitation kit (14321D) according to the manufacturer’s instructions. The sources of antibodies were as follows: monoclonal anti-mouse or anti-goat anti-CACNB4 (1:500; Abcam), which were used in all experiments unless otherwise indicated in the figure legends, and monoclonal anti-rat CACNB4 (1:500; StressMark, Biosciences). Both recognize a region close to the C terminus between amino acid residues 458 and 474. This region is unique to the β4 subunit and is fully preserved in all four of its isoforms (51). We also used the following: monoclonal anti-actin (1:2000; Sigma-Aldrich); polyclonal anti-IRF7 (1:1000; Abcam); anti-IFITM3 (1:2000; Proteintech); polyclonal anti-DDX58 (1:1000; Abcam); monoclonal anti-STAT2 (1:5000; Abcam); polyclonal anti-histone H4 (1:1000; Abcam); polyclonal anti–β-tubulin (1:1000; Abcam); monoclonal anti-pCREB (1:1000; Abcam); polyclonal anti-NS3 (1:1000; GeneTex); monoclonal anti–E-protein (1:1000; GeneTex), and HRP-conjugated anti-mouse or anti-rabbit (Invitrogen) antibodies.

NS1 secretion analysis

The DENV nonstructural protein NS1 was detected in H9c2 infected cells by ELISAs with Platelia Dengue NS1 Ag (Bio-Rad), a one-step, sandwich format microplate ELISA for qualitative or semiquantitative detection in human plasma, which uses murine monoclonal antibodies for capture and revelation. The assay was performed with the reagents at room temperature, and samples and 50 μl of cell culture medium were incubated with the conjugate for 90 min at 37°C within microplate wells coated with anti-NS1 antibodies. After washing, binding was detected by 3,3′,5,5′-tetramethylbenzydine substrate development, which was stopped after a 30-min incubation at room temperature by the addition of an acidic solution. Last, the optical density (OD) was determined at 450 nm using an automatic ELISA plate reader (Multiskan EX, LabSystems). Negative, positive, and cutoff control reagents provided with the kit were run each time for validation. For simplicity, results are expressed directly as OD, rather than as a ratio OD sample/OD cutoff (the results were similar either way).

Focus assay with H9c2 and Vero cell lines

H9c2 cells were transfected and then infected as described earlier. Confluent monolayers of Vero cells grown in 96-well plates were inoculated with supernatant from the culture medium of DENV-infected H9c2 cells at the MOIs indicated in the figure legends (final volume, 0.05 ml). Viral absorption was allowed for 1 hour at 37°C. After the inoculum was removed, 0.2 ml of DMEM and 10% FBS (Sigma-Aldrich, St. Louis, MO) were added, and cell monolayers were washed once with Hank’s solution. The medium was removed 24 hours after infection, and the cells were fixed with 1% formaldehyde, incubated for 20 min with permeabilizing solution, incubated for 2 hours at room temperature with anti–E-protein (4G2 monoclonal antibody 1:100), and then incubated with anti-mouse fluorescein isothiocyanate secondary antibody for detection. Foci were observed by optical microscopy.

Statistical analysis

All results are representative of at least three independent experiments. Data are expressed as means ± SEM. Statistical analyses were performed in GraphPad Prism 4.0 (GraphPad Software) and Sigma Stat 2.0. For two-group comparisons, the Student’s t test was performed. For multiple comparisons, data with a normal distribution were analyzed by one-way ANOVAs followed by Dunn’s honest significant difference test. P < 0.05 was considered to be statistically significant.

SUPPLEMENTARY MATERIALS

www.sciencesignaling.org/cgi/content/full/11/560/eaaj1676/DC1

Fig. S1. Linearity range of Western blots.

Fig. S2. IFN-β increases the abundance of antiviral factors in H9c2 cells.

Fig. S3. The increase in β4 abundance and its nuclear localization in response to poly(I:C).

Fig. S4. Differentiation of H9c2 cells toward a cardiac phenotype.

Fig. S5. STAT2 expression and localization in β4-silenced differentiated and undifferentiated H9c2 cells.

Fig. S6. Effects of DENV2 infection and poly(I:C) transfection on cytosolic Ca2+.

Table S1. Effects of transfection with empty plasmid or β4-encoding plasmid on the expression of viral infection response–related genes.

Table S2. Summary of effects of DENV or mock infection on quantitated β4 localization.

REFERENCES AND NOTES

Acknowledgments: We thank V. Ortiz for relevant discussions; A. Hernández, G. Narasimhan, and A. Hurtado for their assistance in several experiments; and O. Ramírez and I. Lezama for technical assistance. Funding: This work was partially supported by CONACYT grant numbers 167946 and 284053 to J.A.S., 250937 to M.C.G., and 127447 to R.M.d.A. E.R.T., A.H.A.-A., and R.S.-A. were supported in part by fellowships from CONACYT. Author contributions: E.R.T., E.D.C., R.S.-A., A.H.A.-A., M.C.G., and P.B.-C. performed the experiments. E.R.T., E.D.C., M.C.G., and R.M.d.A. designed the experiments and analyzed the data. J.A.S. designed the experiments, analyzed the data, and wrote the paper. Competing interests: The authors declare that they have no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials. Gene expression analysis data were submitted to the Gene Expression Omnibus (GEO) with accession number GSE122950.
View Abstract

Stay Connected to Science Signaling

Navigate This Article