INTRODUCTION

Herpes simplex virus 1 and 2 (HSV-1 and -2) are human pathogens that cause various diseases, including encephalitis, keratitis, and mucocutaneous and skin diseases such as herpes labialis, genital herpes, and herpetic whitlow.¹ Similar to other enveloped viruses, membrane fusion is a critical step in the replication of HSV. Thus, infection of host cells with HSV is initiated by fusion of the viral envelope with the host cell membrane, followed by delivery of the nucleocapsid into the host cell.² Envelope glycoprotein B (gB), gD, and a complex of gH and gL are thought to be sufficient for fusion upon HSV entry, based on cell-based fusion assays showing that co-transfection of genes encoding gB, gD, gH, and gL into cells expressing an HSV entry receptor(s) results in massive fusion of the cells with one another.³ Infection with envelope viruses such as measles virus, human respiratory syncytial virus, severe acute respiratory syndrome coronavirus 1 and 2, and human immunodeficiency virus causes extensive cell–cell fusion in cell cultures, resulting in large multinucleated cells.⁴ Similarly, HSV infection induces cell–cell fusion, whereas cell–cell fusion caused by wild-type HSV strains is infrequent in cell cultures, and large multinucleated cells are barely observed.^5-7 However, wild-type HSV strains induce low but consistent cell–cell fusion,^{6, 7} and large multinucleated cells are frequently observed in herpetic lesions in humans.⁴ Although these observations suggest that cell–cell fusion contributes to HSV pathogenesis, the relevance of HSV-induced cell–cell fusion in viral replication and pathogenesis in vivo remains largely unknown. Furthermore, infrequent cell–cell fusion induced by wild-type HSV strains has barely been focused and studied. In this study, we established a system to quantify cell–cell fusion induced by wild-type HSV strains. Using this system, we demonstrated that the N-glycan of HSV-1 gB, which was previously reported to be significant for viral neurovirulence and replication in the central nervous system in mice,⁸ was required for cell–cell fusion induced by wild-type HSV-1 in cell culture.

MATERIALS AND METHODS

Cells and viruses

Vero, and Plat-GP cells have been described previously.^{9, 10} Wild-type HSV-1(F), wild-type HSV-2(186), YK695(ΔgB), YK696(ΔgB-repair), YK683(gB-N141Q), and YK684(gB-N141Q-repair) (Figure 1) have also been previously described.^{8, 11, 12}

RESULTS

Establishment of a quantitative system for wild-type HSV-induced cell–cell fusion

To quantify the infrequent cell–cell fusion caused by wild-type HSV strains in cell culture, we generated Vero cells that stably expressed the monomeric fluorescent protein VenusA206K¹⁷ or TagRFP¹⁸ and designated them as Vero-Venus or Vero-TagRFP cells, respectively (Figure 2). The Vero-Venus and Vero-TagRFP cells were mixed in a 1:1 ratio. One day after cocultivation, the cells were mock infected or infected with wild-type HSV-1(F) or wild-type HSV-2(186) at an MOI of 1 for an additional 24 h and then subjected to flow cytometry. As the cells positive for both Venus and TagRFP fluorescence resulted from cell–cell fusion between Vero-Venus and Vero-TagRFP cells, virus-induced cell–cell fusion was evaluated by analyzing the frequency of TagRFP+/Venus+ cells using flow cytometer (Figure 3a). As shown in Figure 3b,c, although TagRFP+/Venus+ cells were barely detectable (0.11%) in mock-infected cells, the frequency of TagRFP+/Venus+ cells significantly increased to 5.3% and 7.7% in cells infected with wild-type HSV-1(F) and wild-type HSV-2(186), respectively. These results indicate that this established system could detect infrequent cell–cell fusion caused by wild-type HSV-1 and HSV-2.

Effect of gB on HSV-1–induced cell–cell fusion

It has been reported that cell–cell fusion induced by other envelope viruses involves virus-encoded fusion proteins.⁴ HSV-1 gB is a class III fusion glycoprotein that plays an essential role in the fusion of the envelope with host cell membrane during virus entry.¹⁹ However, whether gB is required for infrequent cell–cell fusion induced by wild-type HSV strains remains unclear. To address this issue, cocultured Vero-Venus and Vero-TagRFP cells were mock-infected or infected with wild-type HSV-1(F), a gB-null mutant ΔgB,⁸ or its repaired virus ΔgB-repair⁸ and subjected to flow cytometry. As shown in Figure 4, the frequency of TagRFP+/Venus+ cells in ΔgB infection was significantly lower than that in wild-type HSV-1(F) or ΔgB-repair infection. These results indicate that gB is required for efficient cell–cell fusion caused by wild-type HSV-1. Notably, the frequencies of cells in ΔgB infection were as low as those in mock-infected cells, suggesting that gB is critical for cell–cell fusion caused by wild-type HSV-1.

Effect of the N-glycan at gB Asn-141 on cell–cell fusion induced by HSV-1

We previously reported that the N-glycan at gB Asn-141 is required for efficient cell surface expression of gB, without affecting its subcellular localization.⁸ To examine whether N-glycans at gB Asn-141 play a role in the cell–cell fusion induced by wild-type HSV-1, cocultured Vero-Venus and Vero-TagRFP cells were mock infected or infected with wild-type HSV-1(F), a recombinant virus gB-N141Q in which gB Asn-141 was substituted with glutamine,⁸ or the repaired virus gB-N141Q-repair, followed by flow cytometry. As shown in Figure 5, the frequency of TagRFP+/Venus+ cells in gB-N141Q infection was significantly lower than that in wild-type HSV-1(F) or gB-N141Q-repair infection as observed with ΔgB infection. These results suggest that N-glycosylation at gB Asn-141 is required for efficient cell–cell fusion caused by wild-type HSV-1.

DISCUSSION

Virus-mediated cell–cell fusion is conserved among certain envelope viruses belonging to different families, suggesting its importance in various viral infections.⁴ HSV-induced cell–cell fusion has long been investigated using syncytial variants of laboratory stocks, which aberrantly cause extensive cell–cell fusion in cell cultures, resulting in large multinucleated cells.^{4, 5} However, syncytial mutants rarely occur in clinical samples of HSV and emerge during the propagation of cultured cells, suggesting that these mutations artificially occur in cell cultures and reduce viral replication and pathogenesis in vivo. Considering these findings, the significance of cell–cell fusion in HSV infection should be investigated using wild-type strains. However, cell–cell fusion induced by wild-type HSV strains remains largely unstudied, probably owing to the low frequencies of cell–cell fusion.

In this study, we established a system to quantify infrequent cell–cell fusion induced by wild-type HSV strains. This system appeared to be easy to operate and sensitive to detect cell–cell fusion. The reporter cells (Vero-Venus and Vero-TagRFP cells) generated for this system were derived from Vero cells, which are susceptible to a variety of viruses. Therefore, the established system with reporter cells could be applicable for quantitative analyses of cell–cell fusion induced by other viruses.

Using the established system, we demonstrated that gB and its N-glycosylation at Asn-141 are important for cell–cell fusion induced by wild-type HSV-1 in cell culture. Notably, we previously reported that glycosylation at gB Asn-141 is required for the proper cell surface expression of gB in HSV-1–infected Vero cells.⁸ Furthermore, N-glycosylation was significant for HSV-1 neurovirulence and replication in the central nervous system of mice, although it played no obvious role in viral replication in Vero and neuroblastoma SK-N-SH cells.⁸ Collectively, these observations suggest that N-glycosylation at HSV-1 gB Asn-141 acts in cell–cell fusion induced by wild-type HSV-1 by regulating proper cell surface expression of gB and that this regulation might be involved in HSV-1 pathogenesis in vivo. This study sheds light on the role of cell–cell fusion induced by wild-type HSV-1 and provides a link between HSV-1–mediated cell–cell fusion and viral pathogenesis in vivo. Further studies to clarify the significance and mechanisms of cell–cell fusion induced by wild-type HSV strains are necessary and interesting and are currently under way in this laboratory.

DISCLOSURE

The authors have no conflict of interest to disclose regarding this work.