Mammalian transient receptor potential (TRP) channels are major components of Ca2+ signaling pathways and control a diversity of physiological functions. Here, we report a specific role for TRPC1 in the entry of herpes simplex virus type 1 (HSV-1) into cells. HSV-1–induced Ca2+ release and entry were dependent on Orai1, STIM1, and TRPC1. Inhibition of Ca2+ entry or knockdown of these proteins attenuated viral entry and infection. HSV-1 glycoprotein D interacted with the third ectodomain of TRPC1, and this interaction facilitated viral entry. Knockout of TRPC1 attenuated HSV-1–induced ocular abnormality and morbidity in vivo in TRPC1−/− mice. There was a strong correlation between HSV-1 infection and plasma membrane localization of TRPC1 in epithelial cells within oral lesions in buccal biopsies from HSV-1–infected patients. Together, our findings demonstrate a critical role for TRPC1 in HSV-1 infection and suggest the channel as a potential target for anti-HSV therapy.
Herpes simplex virus 1 (HSV-1) is a ubiquitous and contagious human virus that remains a tremendous worldwide health care burden and has the potential to cause substantial morbidity (1). Initial infection with HSV-1 starts with binding to and then entry into host cells. The entry of HSV-1 is essential for subsequent infection and occurs by a complicated mechanism that involves interactions between viral components and host cell surface receptors (2, 3). For example, viral glycoprotein D (gD) interacts with cell surface receptors, including HVEM (herpesvirus entry mediator), nectin-1 (4), and nectin-2 (5) in the early step of entry (6, 7).
HSV-1 entry into cells has been shown to induce the release of Ca2+ from endoplasmic reticulum and cause intracellular Ca2+ ([Ca2+]i) elevation, which contributes to HSV-1 infection (8). However, there is little information regarding the role of Ca2+ influx in HSV-1 infection. Notably, Ca2+ entry is important in infection by several types of virus, such as Sindbis virus, West Nile (9), human immunodeficiency virus, filovirus, and arenavirus (10, 11).
In this study, we aimed to investigate the involvement and relevance of host cell Ca2+ influx pathways during HSV-1 entry and infection using the HEp-2 cell line, mouse models, and buccal biopsies from healthy and HSV-1–infected patients. Our data suggest that during HSV-1 entry, Ca2+-entry mediated by Orai1/STIM1 (Ca2+ release–activated Ca2+ channel protein 1/stromal interaction molecule 1) causes translocation of TRPC1 (transient receptor potential canonical 1). We further demonstrated a specific role for TRPC1 in binding HSV-1 and facilitating its entry into cells. These findings reveal a previously unidentified mechanism that contributes to HSV-1 infection and pathology.
RESULTS AND DISCUSSION
Involvement of store-operated Ca2+ entry during HSV-1 infection
HSV-1 infection of HEp-2 cells elicited intracellular Ca2+ release, measured as an increase in [Ca2+]i in Ca2+-free external medium, as well as Ca2+ influx, seen as a [Ca2+]i increase when extracellular Ca2+ was provided (Fig. 1A). The Ca2+ influx component was blocked by the well-established store-operated Ca2+ entry (SOCE) inhibitor GSK-7975A and 2-aminoethoxydiphenyl borate (2-APB) (Fig. 1A) (12–16). GSK-7975A did not inhibit either thapsigargin (TG)– or HSV-1–induced Ca2+ release (fig. S1A). Furthermore, GSK-7975A did not affect 1-oleoyl-2-acetyl-sn-glycerol (OAG)–induced Ca2+ influx, which is not regulated by endoplasmic reticulum Ca2+ store depletion (fig. S1B) (17). The HSV-1–induced intracellular Ca2+ release was abolished by pretreating cells with TG (fig. S1C), suggesting that HSV-1 induced the release of Ca2+ from a TG-sensitive Ca2+ store. Together, these findings demonstrate that HSV-1 induces SOCE in HEp-2 cells. When HEp-2 cells were preinfected with HSV-1 for 15 min at 37°C, the TG treatment induced a threefold greater Ca2+ influx than in mock-infected control cells (Fig. 1B), suggesting that HSV-1 infection enhances Ca2+ influx in host cells.
(A) HSV-1 infection elicits SOCE. Fluo-4–loaded HEp-2 cells were infected with 0.5 MOI of HSV-1 in the absence of extracellular Ca2+ (first arrow) and then treated with 50 μM 2-APB or 10 μM GSK-7975A before Ca2+ was added back (second arrow). Traces show relative fluorescence intensity. Bar graph to the right shows statistical analysis of peak fluorescence increase due to Ca2+ influx over the resting levels. For each condition, data were obtained from three replications, each of which included 10 cells, meaning a total of 30 cells per condition [same for (B), (C), and (E)]. (B) SOCE is increased in HSV-1–infected cells. HEp-2 cells were infected with HSV-1 at 0.5 MOI for 15 min and then treated with TG (left arrow). Ca2+ was then returned to the cells (right arrow). (C and D) SOCE modulates HSV-1 entry. HEp-2 cells were treated with 50 μM 2-APB or 10 μM GSK-7975A and then infected with HSV-1. (C) Representative immunofluorescence images showing the entry of HSV-1. DiI (red) indicates the PM; soluble heparin (10 μg/ml) was used to block binding as an additional negative control. Scale bar, 10 μm. DAPI, 4′,6-diamidino-2-phenylindole. (D) Viral entry analyzed with β-Gal assays. (E) HSV-1 entry is associated with SOCE modulation. SOCE was analyzed during HSV-1 binding or entry into HEp-2 cells as described in (B). Alternatively, uninfected cells were treated with 2-APB or GSK-7975A as negative controls. *P < 0.05, **P < 0.01, and ***P < 0.001 by unpaired t test (B), one-way analysis of variance (ANOVA) (A and E), and two-way ANOVA (D). Graphs show the mean ± SD.
We assessed the contribution of Ca2+ entry to HSV-1 infection by determining the HSV-1 localization in control cells and those treated with the SOCE inhibitors 2-APB and GSK-7975A. Blocking SOCE significantly inhibited the cytoplasmic presence of HSV-1 antigen, as shown by the strong peripheral signal of HSV-1 in cells treated with 2-APB or GSK-7975A, indicating that viral entry into host cells was decreased. In contrast, control cells showed evident HSV-1 antigen in the cytosol in a discrete vesicular pattern, an indicator of viral entry (Fig. 1C). Note that 2-APB and GSK-7975A did not affect HSV-1 binding (fig. S1D). The viral entry was also quantitatively assessed by infecting cells with HSV-1 (KOS) gL86 (18–20). GL86 is a recombinant virus whose gL gene has been replaced by the lacZ gene encoding β-galactosidase (β-Gal). Production of β-Gal in host cells indicates HSV-1 entry into the host cell and transport to the nucleus where gene expression is elicited. The results showed that both 2-APB and GSK-7975A significantly suppressed the production of β-Gal (Fig. 1D). Together, these data show that SOCE plays a role in HSV-1 entry.
To determine whether SOCE is also involved in HSV-1 binding, cells were maintained at 4°C for 1 hour, conditions under which HSV-1 entry is inhibited. Alternatively, HSV-1 entry was assessed by incubating HEp-2 cells with the virus at 37°C for 15 min (see Materials and Methods). TG-stimulated SOCE was increased by HSV-1 entry into HEp-2 cells but not by binding (Fig. 1E). Consistent with this, a mutant HSV-1 that does not express gD (21) did not enter HEp-2 cells and failed to enhance SOCE (fig. S1E). Together, the data shown in Fig. 1 suggest that (i) SOCE is activated as a consequence of HSV-1–induced intracellular Ca2+ release, (ii) SOCE is enhanced by HSV-1 entry, and (iii) SOCE is required for HSV-1 entry.
Involvement of TRPC1 during HSV-1 infection
To further elucidate the mechanism of HSV-1–induced SOCE, major cellular components of SOCE, STIM1, Orai1, and TRPC1 (22, 23) were examined for their role in HSV-1 infection. We then found that HSV-1–induced SOCE, but not intracellular Ca2+ release, was abrogated by knockdown of STIM1, Orai1, and TRPC1 using small interfering RNAs (siRNAs) (siSTIM1, siOrai1, and siTRPC1; Fig. 2A and fig. S1F). Unlike siOrai1 and siSTIM1, siTRPC1 only caused a partial reduction of Ca2+ influx, consistent with previous findings (22, 23). On the other hand, knockdown of TRPC3, TRPC4, TRPC5, and TRPC6 did not significantly affect HSV-1–stimulated Ca2+ entry in HEp-2 cells (fig. S1G).
(A) STIM1, Orai1, and TRPC1 are involved in HSV-1–induced SOCE. HEp-2 cells treated with siOrai1, siTRPC1, or siSTIM1 were infected with HSV-1 at 0.5 MOI for 15 min, and SOCE was analyzed as described in Fig. 1B. (B) STIM1, Orai1, and TRPC1 are involved in HSV-1 entry. Entry of HSV-1 into HEp-2 cells treated with siOrai1, siTRPC1, or siSTIM1 was analyzed with β-Gal assays. n = 6 for each treatment, same in (C) to (E). (C) TRPC1 modulates HSV-1 entry. HSV-1 entry into TRPC1−/− MEFs assessed with β-Gal assays. (D) TRPC1 modulates HSV-1 replication. HSV-1 replication in TRPC1−/− MEFs (left) or siTRPC1-treated HEp-2 cells (right) was analyzed by ELISA. (E) TRPC1 enables HSV-1 entry into CHO cells. CHO cells were transfected with TRPC1 overexpression vector (TRPC1-OE), and HSV-1 entry was assessed with β-Gal assays. (F) HSV-1 entry triggers TRPC1 translocation. Increased TRPC1 (green) in the PM during HSV-1 binding or entry into HEp-2 cells was visualized by TIRF microscope. Scale bar, 10 μm. For each condition, data were obtained from three replications, each of which included 5 cells, meaning a total of 15 cells per condition. (G) HSV-1 entry triggers TRPC1 surface expression. Left, representative blots; right, quantitation of TRPC1 surface expression. n = 3 blots for each treatment. *P < 0.05, **P < 0.01, and ***P < 0.001 by one-way ANOVA (A and F), two-way ANOVA (B to E), or unpaired t test (G). Graphs show the mean ± SD.
Unlike the results for SOCE, siOrai1, siSTIM1, and siTRPC1 equally inhibited virus entry (Fig. 2B), suggesting that the presence of TRPC1 protein is crucial for HSV-1 entry. To further investigate the crucial role of TRPC1 in HSV-1 infection, we assessed the HSV-1 infection in mouse embryonic fibroblasts (MEFs) derived from TRPC1−/− mice (24). HSV-1 entry was significantly decreased in these cells following HSV-1 infection (Fig. 2C), and furthermore, the cells displayed a reduced level of HSV-1 antigen after 18 hours of viral replication (Fig. 2D), similar to that seen in siTRPC1-treated HEp-2 cells (Fig. 2D). In contrast, overexpression of TRPC1 enhanced TG-induced Ca2+ entry in HSV-1–infected HEp-2 cells (fig. S2A) together with an increase in HSV-1 entry and replication (fig. S2B). Knockdown or overexpression of TRPC1 did not affect the amount of HSV-1 antigen in cells under conditions of binding (fig. S2C).
In addition, we followed the “gain of function” protocol demonstrated by Montgomery et al. (18). Here, Chinese hamster ovary (CHO) cells, which lack the known entry receptor for HSV-1 and are resistant to HSV-1 entry (18, 25), were transfected with TRPC1 overexpression vector and then inoculated with HSV-1 to determine whether the overexpressed TRPC1 could serve as a bona fide receptor for HSV-1 entry. Figure 2E shows that HSV-1 infection was significantly increased in TRPC1-expressing CHO cells, thus suggesting that TRPC1 is an entry receptor for HSV-1.
The localization of TRPC1 in the plasma membrane (PM) is a dynamic process and can affect the total Ca2+ influx following store depletion (15, 26). We examined TRPC1 localization in the PM by using total internal reflection fluorescence (TIRF) microscopy. First, TG treatment induced marked TRPC1 translocation in HEp-2 cells (fig. S3A), suggesting that SOCE is able to trigger TRPC1 translocation in HEp-2 cells. Then, the TRPC1 translocation was assessed in HSV-1–infected HEp-2 cells. Viral entry increased the TRPC1 signal within the TIRF microscopy plane (Fig. 2F), and this was not seen during viral binding. Consistent with this, gD-negative HSV-1 did not cause significant translocation of TRPC1 (fig. S3A). Viral entry also caused an increase in the surface expression of TRPC1 (Fig. 2G). Whole-cell expression of TRPC1, STIM1, and Orai1 was not affected under these conditions (Fig. 2G for TRPC1, and fig. S3B for STIM1 and Orai1).
The TRPC1 mutant TRPC1D581K, which carries a mutation at the ion permeation pore region of TRPC1 (27), was used to transfect HEp-2 cells, and SOCE, TRPC1 translocation, and viral entry were monitored. Consistent with previous studies and the notion that this pore-dead mutant is dominant negative to TRPC1-mediated SOCE (15, 27), the HSV-1–induced increase in SOCE as seen in Fig. 1B was diminished in TRPC1D581K-transfected HEp-2 cells (fig. S3C). However, the HSV-1–induced PM translocation of TRPC1D581K protein was comparable to that of wild-type (WT) TRPC1 (fig. S3D). The entry of HSV-1 remained unchanged and comparable to that of WT TRPC1 (fig. S3E). Therefore, the results suggest that translocation of TRPC1 to the PM is an important factor for HSV-1 entry. Notably, this experiment does not rule out the importance of TRPC1-mediated SOCE during HSV-1 entry. This SOCE elicits downstream pathways, i.e., nuclear factor κB pathways (28, 29), which are closely associated with viral replication processes after entry (30, 31).
Because Orai1-mediated Ca2+ entry has been shown to control the PM expression of TRPC1, we studied the effect of HSV-1 entry on Orai1. We found that Orai1 was diffusely localized in control cells but showed punctate signals under conditions of HSV-1 entry (fig. S3F). Furthermore, TRPC1 was colocalized with Orai1 in response to HSV-1 entry, and a close association between the two channels was shown by fluorescence resonance energy transfer (FRET) measurements (fig. S3F) (32, 33). In addition, we studied HSV-1–induced TRPC1 translocation using TIRF microscope in HEp-2 cells loaded with the intracellular Ca2+ chelator BAPTA-AM [1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid tetra(acetoxymethyl ester)]. BAPTA-AM did not inhibit the TRPC1 TIRF signal after HSV-1 entry (fig. S3G). Furthermore, BAPTA, which chelates the extracellular Ca2+, decreased HSV-1–induced TRPC1 translocation (fig. S3G). These results indicates that Ca2+ entry from extracellular space is critical for the translocation of TRPC1, and the site(s) of the Ca2+ action must be very close to the passage of the Ca2+ influx. Otherwise, the calcium ions that entered the cell would have been captured by the intracellular BAPTA before they are able to exert the effect on TRPC1.
In Fig. 1, we have shown that HSV-1 entry enhances SOCE and, conversely, SOCE promotes HSV-1 entry in HEp-2 cells. Together with the data in Fig. 2, we further demonstrated that HSV-1 increased the surface localization of TRPC1, which likely accounts for the increase in SOCE.
Interaction between TRPC1 and HSV-1 gD
We next determined whether HSV-1 directly interacts with TRPC1. FRET measuring TRPC1 and HSV-1 gD interaction on the PM showed a significant enhancement of the FRET signal during HSV-1 entry (Fig. 3A). Other viral glycoproteins, such as gB, did not interact with TRPC1 (Fig. 3A). Consistently, after infection with HSV-1 for 15 min, host cell TRPC1 coimmunoprecipitated with HSV-1 gD (Fig. 3B) but not with gB and gH (fig. S4A). Nectin-1, a known gD receptor (4), was present in the immunoprecipitated fraction together with gD and TRPC1, with a time course similar to the gD-TRPC1 interaction (fig. S4A), indicating that gD associates with both TRPC1 and nectin-1. In addition, TRPC1 coimmunoprecipitated with gD in CHO cells that lack nectin-1 (Fig. 3B) (18, 34). These data suggest that the gD-TRPC1 interaction does not require nectin-1. Whether the presence of nectin-1 interferes with the efficacy of the gD-TRPC1 interaction, and whether there is competition or synergy between TRPC1 and nectin-1 in binding gD, deserves further evaluation.
(A) Representative images and statistics of interaction between TRPC1 and gD. Colocalization of TRPC1 with gD or gB was analyzed by FRET at different time points of infection. For each condition, data were obtained from three replications, each of which included 5 cells, meaning a total of 15 cells per condition [same for (C)]. (B) TRPC1 coimmunoprecipitates with gD. HEp-2 cells (upper panels) or TRPV1-OE–transfected CHO cells (lower panels) were infected with HSV-1, and the interaction between TRPC1 and HSV-1 gD was analyzed by coimmunoprecipitation. Preimmune IgG served as a control. A converse coimmunoprecipitation was also performed for each treatment. Input was from the whole-cell extract as a positive control; supernatant and pellet were obtained after immunoprecipitation (n = 3 blots for each condition). (C) Effects of TRPC1 mutants on the gD-TRPC1 interaction. Five mutations were introduced on the three ectodomains of TRPC1 as shown in the cartoon. Middle and right panels show representative images and statistics for gD-TRPC1 interactions measured by FRET in HEp-2 cells transfected with vectors overexpressing mutant TRPC1 (S1 to S5). (D) Effect of TRPC1 mutants on HSV-1 entry. HEp-2 cells were transfected with mutant TRPC1 (S1 to S5) vectors, and viral entry was analyzed with β-Gal assays. Scale bars, 10 μm. **P < 0.01 and ***P < 0.001 by one-way ANOVA (A and C) or two-way ANOVA (D). Graphs show the mean ± SD.
Five sites (S1 to S5) were mutated in the three ectodomains of TRPC1 (see cartoons in Fig. 3C and fig. S4B). These mutants were expressed individually in HEp-2 cells along with HSV-1 infection. Similar TG-stimulated Ca2+ entry was seen in cells expressing the different TRPC1 mutants (fig. S4C). FRET was then used to measure interaction between viral gD and different mutant TRPC1s in HEp-2 cells. Those expressing TRPC1 with a mutation in S5 (third ectodomain) displayed a lower gD-TRPC1 interaction than cells expressing the other TRPC1 mutants (Fig. 3C). HEp-2 cells expressing S5-mutated TRPC1 also showed less β-Gal enzyme activity (Fig. 3D). Furthermore, when CHO cells were transfected with the S1-S5 TRPC1 vectors and then infected with HSV-1 gL86, overexpression of S1-S4 TRPC1 mutants enabled viral entry. However, for the S5 mutant, the HSV-1 entry was comparable to that of control vector–transfected cells (fig. S4D). Therefore, we concluded that the TRPC1 S5 site is important for both the gD-TRPC1 interaction and HSV-1 entry. Together, the findings suggest that HSV-1–induced SOCE leads to enhancement of TRPC1 in the PM and direct interaction between TRPC1 and viral gD at the cell membrane promotes HSV-1 entry. Notably, TRPC1 is present at a low level in the PM before HSV-1 infection (35), so we hypothesized that the initial entry of HSV-1 into cells is mediated by these TRPC1 proteins. SOCE, activated subsequent to HSV-1 entry and intracellular Ca2+ release, further enhances the level of TRPC1 in the PM, which can then bind to extracellular HSV-1 and promote entry of the virus.
Interaction between TRPC1 and HSV-1 gD in mice and human samples
To further examine the role of the TRPC1–HSV-1 interaction under physiological conditions of HSV-1 infection, we established an ocular infection model in WT and TRPC1−/− mice using two HSV-1 strains, 17 syn+ and McKrae. Note that the HSV-1 strain KOS used in in vitro experiments display low virulence in mouse models. In WT mice, both 17 syn+ and McKrae HSV-1 strains induced marked edema in the eyeballs with hypercellularity of the sclera and detachment of the retina. In contrast, TRPC1−/− mice displayed mild ocular abnormality, with relatively little retinal detachment (Fig. 4A). Quantitatively, TRPC1−/− mice had significantly lower herpes stromal keratitis (HSK) scores, which were measured on day 5 after infection (Fig. 4B). HSK development (i.e., the time taken to induce 50% corneal opacity) was significantly delayed in TRPC1−/− mice (Fig. 4C). TRPC1−/− mice had significantly higher survival rates after HSV-1 infection than infected WT mice (Fig. 4D).
(A) Effect of TRPC1 on in vivo HSV-1 infection. The mice were infected with 104 plaque-forming units (pfu) per mouse of 17 syn+ or 103 pfu per mouse of McKrae HSV-1 [same for (B) to (D)], and the eyeballs were stained with hematoxylin and eosin after 5 days of infection. Figures show representative eyeball morphology caused by McKrae infection; magnification, ×40; n = 15 for each infected group. (B) HSK scores in WT and TRPC1−/− groups on day 5 of infection. n = 15 for each infected group. (C) Days to HSK causing 50% corneal opacity. n = 5 for each infected group. (D) Survival curves of infected mice. Kaplan-Meier curves for TRPC1−/− and WT mice infected with HSV-1 for 15 days (n = 10 for each infected group). (E) Percentage of HSV-1 gD–positive cells in buccal biopsies from each patient. The percentages of HSV-1 gD–positive epithelial cells were calculated for 100 randomly chosen oral epithelial cells in biopsies from three slides for each patient (p1 to p25). (F) Features of TRPC1 fluorescence distribution in human samples. The TRPC1 fluorescence on the PM versus that in the cytoplasm (PM/Cyto; in pixel) in HSV-1–infected or uninfected epithelial cells from buccal biopsies of 25 patients (p1 to p25) and 3 healthy candidates (h1 to h3). SUM, overall PM/Cyto value difference between patients and healthy candidates (t test). (G) Representative images of healthy and HSV-1–infected oral epithelial cells. Scale bar, 10 μm; n = 25 patients, n = 3 healthy individuals; 10 cells from each individual were imaged; DiI (red) indicates PM. (H) Fluorescence intensity of TRPC1 calculated for Pearson correlation with gD on the PM (n = 100 HSV-1–infected cells). ***P < 0.001 by unpaired t test (B, C, and F). Graphs show the mean ± SD.
In humans, the common symptoms of HSV-1 infection are pharyngitis, tonsillitis, and gingivostomatitis (36). Buccal biopsies from patients diagnosed with herpes oral lesions and healthy individuals were used to examine the expression and location of TRPC1 and HSV-1 gD in epithelial cells. In samples from 25 patients, 80.8 ± 3.95% of the oral epithelial cells were infected by HSV-1 (presence of fluorescent HSV-1 gD antigen signals) (Fig. 4E). In healthy uninfected epithelial cells, the fluorescent signal of TRPC1 was dispersed in the cytoplasm. The PM versus cytoplasm (PM/Cyto) value of TRPC1 fluorescence in these uninfected cells was significantly lower in cells positive for HSV-1 gD antigen (Fig. 4, F and G). A few cells from patients with herpes-induced oral lesions were not positive for HSV-1 antigen; they might have been from healthy tissue near the lesion site. These gD-negative cells did not show PM expression of TRPC1 and thus had a lower TRPC1 PM/Cyto ratio (Fig. 4F). There was strong correlation between the fluorescence intensity of TRPC1 and that of gD on the PM in HSV-1–infected cells (Pearson correlation coefficient = 0.767, P < 0.05; Fig. 4H).