Prohibitin-1 Contributes to Cell-to-Cell Transmission of Herpes Simplex Virus 1 via the MAPK/ERK Signaling Pathway

Herpesviruses are ubiquitous pathogens of various animals, including humans. These viruses primarily pass through cell junctions to spread to uninfected cells.

ABSTRACT

Viral cell-to-cell spread, a method employed by several viral families for entrance via cell junctions, is highly relevant to the pathogenesis of various viral infections. Cell-to-cell spread of herpes simplex virus 1 (HSV-1) is known to depend greatly on envelope glycoprotein E (gE). However, the molecular mechanism by which gE acts in HSV-1 cell-to-cell spread and the mechanisms of cell-to-cell spread by other herpesviruses remain poorly understood. Here, we describe our identification of prohibitin-1 as a novel gE-interacting host cell protein. Ectopic expression of prohibitin-1 increased gE-dependent HSV-1 cell-to-cell spread. As observed with the gE-null mutation, decreased expression or pharmacological inhibition of prohibitin-1 reduced HSV-1 cell-to-cell spread without affecting the yield of virus progeny. Similar effects were produced by pharmacological inhibition of the mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) pathway, wherein prohibitin-1 acts as a protein scaffold and is required for induction of this pathway. Furthermore, artificial activation of the MAPK/ERK pathway restored HSV-1 cell-to-cell spread impaired by the gE-null mutation. Notably, pharmacological inhibition of prohibitins or the MAPK/ERK pathway reduced viral cell-to-cell spread of representative members in all herpesvirus subfamilies. Our results suggest that prohibitin-1 contributes to gE-dependent HSV-1 cell-to-cell spread via the MAPK/ERK pathway and that this mechanism is conserved throughout the Herpesviridae, whereas gE is conserved only in the Alphaherpesvirinae subfamily.

IMPORTANCE Herpesviruses are ubiquitous pathogens of various animals, including humans. These viruses primarily pass through cell junctions to spread to uninfected cells. This method of cell-to-cell spread is an important pathogenic characteristic of these viruses. Here, we show that the host cell protein prohibitin-1 contributes to HSV-1 cell-to-cell spread via a downstream intracellular signaling cascade, the MAPK/ERK pathway. We also demonstrate that the role of the prohibitin-1-mediated MAPK/ERK pathway in viral cell-to-cell spread is conserved in representative members of every herpesvirus subfamily. This study has revealed a common molecular mechanism of the cell-to-cell spread of herpesviruses.

INTRODUCTION

Viruses use two distinct modes to spread from a donor-infected cell to an uninfected target cell: cell-free spread and cell-to-cell spread (1, 2). In cell-free spread, viruses released from infected cells spread distantly, followed by diffusion through the extracellular environment and, finally, entry into a new cell. However, this method is vulnerable to interference from barriers against viral infections associated with the extracellular environment and the donor and/or target cell, which include neutralizing antibodies, intrinsic immunity, and factors that interfere with viral entry or postentry steps (2). In cell-to-cell spread, virions are sorted to the junctional surfaces of cells, which enables the rapid and efficient infection of neighboring cells while avoiding the barriers against viral infections encountered in cell-free spread (2). Notably, cell-to-cell spread is considered the predominant method of viral spread in vivo and contributes to the pathogenesis of various viral infections (2, 3). However, the specific molecular mechanisms used by most viruses for cell-to-cell viral spread remain poorly understood.

Herpesviruses, belonging to the family Herpesviridae and subdivided into 3 subfamilies, Alphaherpesvirinae, Betaherpesvirinae, and Gammaherpesvirinae, generally infect mammals, birds, and reptiles by cell-to-cell spread (4). These viruses are ubiquitous in nature and unique in that they establish lifelong infections in their hosts (4). Herpes simplex virus 1 (HSV-1) belongs to the Alphaherpesvirinae subfamily, and is one of the most extensively studied herpesviruses (4). It causes various mucocutaneous and skin diseases in humans, including herpes labialis, genital herpes, herpetic whitlow, keratitis, and life-threatening encephalitis (5). Following primary HSV-1 infection at peripheral mucosal sites, the virus is transported via innervating sensory neurons to a sensory ganglion, where it establishes lifelong latency (5). HSV-1 is periodically reactivated and travels back through sensory neurons to the peripheral site to form lesions (5).

The cell-to-cell transmission of HSV-1 and the porcine alphaherpesvirus pseudorabies virus (PRV) was shown to be critical for viral spread between mucosal epithelial cells, from mucosal epithelial cells to sensory neurons, and from sensory neurons to mucosal epithelial cells (6, 7). An HSV-1 heterodimer of the envelope glycoprotein E (gE), its chaperone gI, and the PRV heterodimer of their homologs were previously reported to play central roles in viral cell-to-cell spread and, in the case of HSV-1, to potentially facilitate the transport of enveloped virions to the junctional cell surface (3, 8–10). However, although more than a quarter of a century has passed since gE was first identified as a specific viral factor for viral cell-to-cell spread in herpesviruses, the molecular mechanisms involved are still poorly understood.

In this study, we identified host cell proteins that interacted with gE and cDNAs that encoded host cell proteins that promoted HSV-1 cell-to-cell spread. Then, to investigate the mechanism of HSV-1 cell-to-cell spread, we characterized the host cell proteins that interacted with gE and contributed to HSV-1 cell-to-cell spread.

RESULTS

Identification of prohibitin-1 (PHB1) as a host cell protein that interacts with gE and promotes gE-dependent HSV-1 cell-to-cell spread.

We used tandem affinity purification coupled with mass spectrometry (MS)-based proteomics to screen for host cell proteins that interact with gE in HSV-1-infected HaCaT cells and identified approximately 700 host cell proteins that could be potential interactors with gE (data not shown).

HSV-1 cell-to-cell spread can be analyzed in vitro by measuring the plaques produced in the presence of neutralizing antibodies (9, 11). Our preliminary observations showed that the sizes of plaques produced by HSV-1 varied, depending on the infected cell line. HSV-1 produces very tiny plaques in HeLa cells in comparison to those it produces in Vero cells. Although HSV-1 efficiently yields progeny viruses in both cell lines, the very small plaques produced in HeLa cells suggest that the host cell protein(s) required for HSV-1 cell-to-cell spread is dysfunctional in that cell line. To screen for cDNAs that encode host cell proteins promoting HSV-1 cell-to-cell spread, we transduced HeLa cells with cDNAs prepared from Vero cells; infected the transduced HeLa cells with enhanced green fluorescent protein (EGFP)-expressing HSV-1 (HSV-1 EGFP), which is phenotypically the same as wild-type HSV-1(F) in cell cultures (12) (Fig. 1) in the presence of neutralizing antibodies; analyzed plaque formation in the transduced cells; and determined the sequences of transduced cDNAs from cell pools containing cells that showed increased levels of plaque formation of fluorescent foci of infected cells (Table 1). Among 11 host cell proteins encoded by the identified cDNAs that potentially promote HSV-1 cell-to-cell spread in HeLa cells, only PHB1 was also identified as the potential interactor with gE.

FIG 1.

Schematic diagram of the genome structures of HSV-1(F) and recombinant viruses used in this study. Line 1, wild-type HSV-1(F) genome and domains of UL3 to UL4, UL22 to UL24, UL50 to UL51, and Us7 (gI) to Us9 genes; lines 2 to 9, recombinant viruses with insertions between the UL3 and UL4 genes or UL23 (vTK) gene or between the UL50 and UL51 genes or Us8 (gE) gene. “Stop” denotes a stop codon.

TABLE 1.

List of cDNAs identified by expression screening

Gene IDa	Gene designation	Full name
5245	PHB	Prohibitin
5528	TMEM140	Transmembrane protein140
28988	DBNL	Drebrin-like
10399	RACK1	Receptor for activated C kinase 1
23541	SEC14L2	SEC14-like lipid binding 2
81605	URM1	Ubiquitin-related modifier 1
6187	RPS2	Ribosomal protein S2
10491	CRTAP	Cartilage-associated protein
71	ACTG1	Actin gamma 1
6050	RNH1	Ribonuclease/angiogenin inhibitor 1
100653233	LOC100653233	Uncharacterized LOC100653233

^aID, identifier.

The interaction between gE and PHB1 in HSV-1-infected cells was verified by coimmunoprecipitation, wherein PHB1 specifically formed a coprecipitate with gE, which had been tagged by a Strep-tag peptide (gE-Strep) from lysates of HaCaT cells infected with gE-Strep-expressing HSV-1 (HSV-1 gE-Strep) (Fig. 1; see also Fig. 2A). In contrast, HSV-1 thymidine kinase (vTK) tagged with a Strep-tag (vTK-Strep) was unable to pull down PHB1 from lysates of HaCaT cells infected with vTK-Strep-expressing HSV-1 (HSV-1 vTK-Strep) (Fig. 1; see also Fig. 2B). Reciprocally, a carboxyl-terminal domain of PHB1 encoded by PHB1 amino acids 219 to 272 fused to glutathione S-transferase (GST-PHB1-CTD), but not GST alone or an amino-terminal domain of PHB1 encoded by PHB1 amino acids 1 to 218 fused to GST (GST-PHB1-NTD), pulled down gE from the lysates of Vero cells infected with wild-type HSV-1(F) (Fig. 2C and D). Furthermore, ectopic expression of PHB1 tagged with Myc-tag peptide (PHB1-Myc) in HeLa cells (Fig. 3A) led to significantly increased sizes of plaques produced in those HeLa cells that had been infected by HSV-1 EGFP (Fig. 3B). In contrast, ectopic expression of PHB1 had little effect on the yield of progeny virus from these infected HeLa cells (Fig. 3C).

FIG 2.

Interaction between gE and PHB1. (A and B) HaCaT cells were infected with wild-type HSV-1(F) or HSV-1 gE-Strep (A) or with wild-type HSV-1(F) or HSV-1 vTK-Strep (B) at an MOI of 5. At 36 h postinfection, the cells were harvested, captured by Strep-Tactin Sepharose beads, and analyzed by immunoblotting (IB) with the indicated antibodies. (C) Schematic diagrams of wild-type PHB1 and its domains used in the GST pulldown experiments. Line 1, wild-type PHB1; line 2, the amino-terminal domain used for the GST-PHB1-NTD fusion protein; line 3, the carbonyl-terminal domain used for the GST-PHB1-CTD fusion protein. Transmembrane (TM) and PHB domains are shown. (D) GST, GST-PHB1-NTD, or GST-PHB1-CTD fusion proteins were expressed in E. coli, immobilized on glutathione-Sepharose beads, and reacted with lysates of Vero cells infected with wild-type HSV-1(F) at an MOI of 5 for 24 h. The beads were washed extensively and divided into two parts. One part was analyzed by electrophoresis in a denaturing gel and immunoblotted with anti-gE antibody (top gel), and the other part was analyzed by electrophoresis in a denaturing gel and stained with Coomassie brilliant blue (CBB) (bottom gel). Images are representative of results of 3 independent experiments. WCE, whole-cell extract; α, anti-.

FIG 3.

Effects of ectopically expressed PHB1 on HSV-1 infection. (A) HeLa/puro cells or 2 clones of HeLa/PHB1-Myc cells were analyzed by immunoblotting with the indicated antibodies. Images are representative of results of 3 independent experiments. (B) HeLa/puro cells or 2 clones of HeLa/PHB1-Myc cells were infected with HSV-1 EGFP at an MOI of 0.001 under plaque assay conditions. The areas of 15 individual plaques of each infected cell culture were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Images are representative of results of 3 independent experiments. The asterisks indicate statistical differences as follows: ***, P < 0.001 (by Tukey test). (C) HeLa/puro or HeLa/PHB1-Myc cells were infected by wild-type HSV-1(F) at an MOI of 10 or 0.05. Total virus from the cell culture supernatants and the infected cells was harvested at 6, 12, 18, and 36 h postinfection and at 6, 12, 24, 48, and 72 h postinfection, respectively, and quantified on Vero cells. Each data point represents the mean ± standard error of data from 3 independent experiments. n.s., not significant by Student's t test. (D) HeLa/puro and HeLa/PHB1-Myc cells were infected with HSV-1 EGFP, HSV-1 ΔgE/Venus, or HSV-1 ΔgE-repair/Venus at an MOI of 0.001 under plaque assay conditions. The areas of 15 individual plaques for each of the indicated viruses were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. ***, P < 0.001 (by Tukey test); n.s., not significant.

In HeLa cells infected with a gE-null mutant HSV-1 that expressed Venus fluorescent protein (HSV-1 ΔgE/Venus) (Fig. 1), the plaque sizes were as small as those produced in HeLa cells infected with HSV-1 EGFP or with a recombinant HSV-1 in which the gE-null mutation in HSV-1 ΔgE/Venus had been repaired (HSV-1 ΔgE-repair/Venus) (Fig. 1; see also Fig. 3D). Importantly, unlike the effect of ectopic expression of PHB1 on plaque sizes produced in HeLa cell cultures infected with HSV-1 EGFP or HSV-1 ΔgE-repair/Venus, ectopic expression of PHB1 in HeLa cells infected with HSV-1 ΔgE/Venus had little effect on plaque sizes (Fig. 3D). These results suggest that PHB1 interacts with gE in HSV-1-infected cells and that this host cell protein has the ability to promote HSV-1 cell-to-cell spread (without affecting viral replication) in a manner dependent on gE.

We then constructed A549/PHB1-low cells that expressed markedly reduced levels of PHB1 by transfection with a plasmid expressing Cas9 and a synthetic single-guide RNA targeting PHB1 to examine the effect of PHB1 depletion on HSV-1 cell-to-cell spread (Fig. 4). A549/PHB1-low cells infected with HSV-1 EGFP produced significantly smaller plaques than parental A549 cells infected with HSV-1 EGFP or HSV-1 ΔgE-repair/Venus (Fig. 5A). In contrast, as observed with the gE-null mutation (Fig. 5B) and reported previously (13, 14), low PHB1 expression levels had little effect on viral replication (Fig. 5C) in A549 cells at multiplicities of infection (MOIs) of 5 and 0.05. We note that, in agreement with earlier reports (13, 14), inactivation of an important factor for HSV-1 cell-to-cell spread such as gE does not necessarily depress progeny virus yields at low MOIs. The reduction in plaque sizes associated with PHB1 depletion was significantly restored by ectopic expression of PHB1-Myc (Fig. 5D and E). Consistent with the findings in HeLa cells, the plaques produced in A549/PHB1-low cells infected by HSV-1 ΔgE/Venus were as small as those produced in these cells infected by HSV-1 EGFP or HSV-1 ΔgE-repair/Venus, and PHB1 depletion did not affect the sizes of plaques produced in A549 cells infected with HSV-1 ΔgE/Venus (Fig. 5A). We also examined the effect of an inhibitor for PHB1, rocaglamide (RocA) (15), on HSV-1 cell-to-cell spread. The RocA treatment of HSV-1-infected A549 cells led to a plaque phenotype similar to that resulting from PHB1 depletion, i.e., reduced plaque sizes without obvious effects on yield of progeny virus (Fig. 6).

FIG 4.

Characterization of A549/PHB1-low cells. (A) The targeted PHB1 mutation sequences and the parental sequence in A549/PHB1-low cells are shown. (B) Lysates of A549 and A549/PHB1-low cells were analyzed by immunoblotting with the indicated antibodies. Images are representative of results of 3 independent experiments. (C) Percentages of PHB1 protein in the cells relative to α-tubulin, as determined from intensity measurements of the corresponding immunoblots as described for panel B. Each percentage value represents the mean ± standard error of results from 3 independent experiments. *, P < 0.05 (by Student's t test). (D) Viability of A549 and A549/PHB1-low cells. Data are shown as means ± standard errors of the results of 3 independent experiments and are expressed relative to the mean determined for A549 cells, which was normalized to 100%. n.s., not significant by Student's t test.

FIG 5.

Effects of decreased PHB1 expression on HSV-1 infection in A549 cells. (A) A549 and A549/PHB1-low cells were infected with HSV-1 EGFP, HSV-1 ΔgE/Venus, or HSV-1 ΔgE-repair/Venus at an MOI of 0.001 under plaque assay conditions. The areas of 15 individual plaques for each of the indicated infected cell lines were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Images are representative of results of 3 independent experiments. ***, P < 0.001 (by Tukey test); n.s., not significant. (B) A549 cells were infected with wild-type HSV-1(F), HSV-1 ΔgE, or HSV-1 ΔgE-repair at an MOI of 5 or 0.05. Total virus from the cell culture supernatants and the infected cells was harvested at 6, 12, 18, and 36 h and at 6, 12, 24, 48, and 72 h postinfection, respectively, and assayed on Vero cells. n.s., not significant by Student's t test. (C) A549 or A549/PHB1-low cells were infected by wild-type HSV-1(F) at an MOI of 5 or 0.05. Total virus from the cell culture supernatants and the infected cells was harvested at 6, 12, 18, and 36 h and at 12, 24, 48, and 72 h postinfection, respectively, and assayed on Vero cells. n.s., not significant by Student's t test. (D) Lysates of A549/puro, A549/PHB1-Myc, A549/PHB1-low/puro, and A549/PHB1-low/PHB1-Myc cells were analyzed by immunoblotting with the indicated antibodies. Images are representative of results of 3 independent experiments. (E) A549/puro, A549/PHB1-Myc, A549/PHB1-low/puro, and A549/PHB1-low/PHB1-Myc cells were infected by wild-type HSV-1 EGFP at an MOI of 0.001 under plaque assay conditions. The areas of 15 individual plaques of each of the indicated infected lines were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. **, P < 0.01; ***, P < 0.001 (by Tukey test).

FIG 6.

Effects of treatment by the PHB1 inhibitor RocA on HSV-1 infection in A549 cells. (A) A549 cells were infected by HSV-1 EGFP at an MOI of 0.001. At 1 h postinfection, the cells were incubated under plaque assay conditions with or without 20 nM RocA. The areas of 15 individual plaques of each of the infected cell cultures with or without RocA were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Images are representative of results of 3 independent experiments. ***, P < 0.001 (by Student's t test). (B) A549 cells were infected with wild-type HSV-1(F) at an MOI of 5 or 0.05 with or without 20 nM RocA. At 18 h and 48 h postinfection, respectively, total virus from the supernatants and the infected cells was harvested and assayed on Vero cells. Each bar represents the mean ± standard error of results from 3 independent experiments. n.s., not significant by Student's t test.

Altogether, these findings indicate that PHB1 acts as an important host cell factor for HSV-1 cell-to-cell spread in a gE-dependent manner.

We also examined the effects of PHB1 on HSV-1 cell-to-cell spread in other cell lines. In agreement with the results seen using A549 cells, HaCaT/PHB1-low cells infected with HSV-1 EGFP produced significantly smaller plaques than parental HaCaT cells infected with HSV-1 EGFP and low PHB1 expression had little effect on viral replication as observed with the gE-null mutation (Fig. 7). Furthermore, knockdown of PHB1 by each of the two different small interfering RNAs (siRNAs) in Vero cells significantly decreased the size of plaques produced by HSV-1 EGFP (Fig. 8). These results further support our conclusion that PHB1 is an important host cell factor for HSV-1 cell-to-cell spread.

FIG 7.

Effects of decreased PHB1 expression on HSV-1 infection in HaCaT cells. (A) The targeted PHB1 mutation sequences and the parental sequence in HaCaT/PHB1-low cells are shown. (B) Lysates of HaCaT and HaCaT/PHB1-low cells were analyzed by immunoblotting with the indicated antibodies. Images are representative of results of 3 independent experiments. (C) Percentages of PHB1 protein in cells relative to α-tubulin, as determined from intensity measurements of the corresponding immunoblots as described for panel B. Each percentage value represents the mean ± standard error of results from 3 independent experiments. *, P < 0.05 (by Student's t test). (D) Viability of HaCaT and HaCaT/PHB1-low cells. Data are shown as means ± standard errors of the results of 3 independent experiments and are expressed relative to the mean determined for HaCaT cells, which was normalized to 100%. n.s., not significant by Student's t test. (E) HaCaT or HaCaT/PHB1-low cells were infected by wild-type HSV-1 EGFP at an MOI of 0.0001 under plaque assay conditions. The areas of 15 individual plaques of each of the indicated infected lines were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. **, P < 0.01 (by Student's t test). (F) HaCaT or HaCaT/PHB1-low cells were infected by wild-type HSV-1(F) at an MOI of 5 or 0.05. Total virus from the cell culture supernatants and the infected cells was harvested at 18 h and at 48 h postinfection, respectively, and assayed on Vero cells. n.s., not significant by Student's t test. (G) HaCaT cells were infected by wild-type HSV-1(F) at an MOI of 5 or 0.05. Total virus from the cell culture supernatants and the infected cells was harvested at 18 h and at 48 h postinfection, respectively, and assayed on Vero cells. Each data point represents the mean ± standard error of data from 3 independent experiments. n.s., not significant by Tukey test.

FIG 8.

Effects of decreased PHB1 expression on HSV-1 infection in Vero cells. (A) Lysates of Vero cells treated with control siRNA (Vero/si control) or either of the two different siRNAs to PHB1 for 48 h were analyzed by immunoblotting with the indicated antibodies. Images are representative of results of 3 independent experiments. (B) Vero cells treated with control siRNA or each of the siRNAs to PHB1 were infected with wild-type HSV-1 EGFP at an MOI of 0.0001 under plaque assay conditions. The areas of 23 individual plaques of each of the indicated infected lines were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 23 plotted plaque areas. Data are representative of results of 3 independent experiments. ***, P < 0.001 (by Tukey test).

PHB1 facilitates the transport of nascent enveloped virions in the cytoplasm.

HSV-1 gE has been reported to contribute to viral cell-to-cell spread by promoting the transport of enveloped virions in the cytoplasm (10). To clarify the gE-dependent mechanism by which PHB1 acts in HSV-1 cell-to-cell spread, we used electron microscopy to investigate the effects of PHB1 depletion and the gE-null mutation on viral morphogenesis in A549 cells. The decreased expression of PHB1 in infected A549/PHB1-low cells led to the accumulation of enveloped virions in the cytoplasm and to reduced levels of enveloped virions in the extracellular space (Fig. 9; see also Table 2). Thus, 8.6% of the viral particles were enveloped virions in the cytoplasm of wild-type HSV-1(F)-infected A549 cells, whereas the proportion of virus particles that were enveloped virions in the cytoplasm of HSV-1(F)-infected A549/PHB1-low cells increased to 33.0% (Fig. 9B; see also Table 2). Likewise, 19.2% of viral particles occurred in the extracellular space of HSV-1(F)-infected A549 cells, whereas the proportion of viral particles in the extracellular space of HSV-1(F)-infected A549/PHB1-low cells decreased to 8.1% (Fig. 9B; see also Table 2). Similar results were also seen with RocA treatment (Fig. 10A and B; see also Table 3) and HSV-1 gE-null-mutant-infected A549 cells (Fig. 10C and D; see also Table 4). These phenotypes were reported to reflect the impaired transport of enveloped virions to the cell surface (10, 16, 17), and our findings were in agreement with those obtained with a gE-null mutant virus in human epithelial cells, as described previously (10). In addition, effects similar to those of PHB1 depletion and gE-null mutation in A549 cells on HSV-1 morphogenesis were also observed in HaCaT cells (Fig. 11; see also Tables 5 and 6). Our results indicate that PHB1 and gE are required for efficient transport of enveloped virions in the cytoplasm in these cells and suggest that PHB1 contributes to gE-dependent cell-to-cell spread by promoting the transport of enveloped virions in the cytoplasm.

FIG 9.

Effect of decreased expression of PHB1 on HSV-1 morphogenesis in A549 cells. (A) A549 and A549/PHB1-low cells were infected by wild-type HSV-1(F) at an MOI of 15. At 20 h postinfection, the cells were embedded, sectioned, stained, and examined by electron microscopy. The images labeled a to d on the right are magnifications of the corresponding boxed areas on the left. Images are representative of results of 3 independent experiments. Bars, 500 nm. (B) The mean proportions of enveloped virions in the cytoplasm and extracellular space of 10 infected cells were determined. ***, P < 0.001 by Student's t test.

TABLE 2.

Effect of depression of PHB on the distribution of virus particles in A549 cells

Cell line	Mean ± SE of % of virus particles in each morphogenetic stagea					Total no. of counted particles/no. of cells
	Nucleocapsids in nucleus	Enveloped virions in perinuclear space	Nucleocapsids in cytoplasm	Enveloped virions in cytoplasm	Extracellular enveloped virions
A549	70.4 ± 4.0 (1,201)	1.2 ± 0.6 (13)	0.6 ± 0.3 (11)	8.6 ± 2.2 (150)	19.2 ± 3.4 (258)	1,633/10
A549/PHB1-low	57.2 ± 5.3 (1,249)	0.7 ± 0.3 (16)	1.0 ± 0.4 (18)	33.0 ± 4.6b (622)	8.1 ± 1.6c (140)	2,045/10

^aNumbers in parentheses are the numbers of virus particles.

^bStatistically significant difference from A549 cells (P = 1.4 × 10⁻⁴ by Student's t test).

^cStatistically significant difference from A549 cells (P = 0.009 by Student's t test).

FIG 10.

Effect of pharmacological inhibition of PHB1 and gE-null mutation on HSV-1 morphogenesis in A549 cells. (A) A549 cells were infected by wild-type HSV-1(F) at an MOI of 15. At 1 h postinfection, the infected cells were treated with or without 20 nM RocA. At 20 h postinfection, these cells were fixed, embedded, sectioned, stained, and examined by transmission electron microscopy. The images labeled a to d on the right are magnifications of the corresponding boxed areas on the left. Images are representative of results of 3 independent experiments. Scale bars, 500 nm. (B) The percentages of enveloped virions in the cytoplasm and enveloped extracellular virions of 20 infected cells were determined. *, P < 0.05; **, P < 0.01 by Student's t test. (C) A549 cells were infected with HSV-1 ΔgE or HSV-1 ΔgE-repair at an MOI of 15. At 20 h postinfection, the cells were fixed embedded, sectioned, stained, and examined by transmission electron microscopy. The images labeled a to d on the right are magnifications of the corresponding boxed areas on the left. Images are representative of results of 3 independent experiments. Scale bars, 500 nm. (D) The percentages of enveloped virions in the cytoplasm and enveloped extracellular virions of 12 infected cells were determined. ***, P < 0.001 by Student's t test.

TABLE 3.

Effect of treatment of a PHB inhibitor on the distribution of virus particles in A549 cells

Cell line	Mean ± SE of % of virus particles in each morphogenetic stagea					Total no. of counted particles/ no. of cells
	Nucleocapsids in nucleus	Enveloped virions in perinuclear space	Nucleocapsids in cytoplasm	Enveloped virions in cytoplasm	Extracellular enveloped virions
DMSOb	61.4 ± 4.2 (1,537)	0.5 ± 0.2 (10)	1.7 ± 0.5 (35)	18.3 ± 3.3 (451)	18.1 ± 2.3 (446)	2,479/20
RocA	49.7 ± 4.2 (347)	0.5 ± 0.3 (3)	3.7 ± 1.0 (24)	34.2 ± 3.9c (239)	11.8 ± 1.3d (80)	693/20

^aNumbers in parentheses are the numbers of virus particles.

^bDMSO, dimethyl sulfoxide.

^cStatistically significant difference from the DMSO-treated cells (P = 0.0034 by Student's t test).

^dStatistically significant difference from the DMSO-treated cells (P = 0.023 by Student's t test).

TABLE 4.

Effect of gE null mutation on the distribution of virus particles in A549 cells

Treatment	Mean ± SE of % of virus particles in each morphogenetic stagea					Total no. of counted particles/ no. of cells
	Nucleocapsids in nucleus	Enveloped virions in perinuclear space	Nucleocapsids in cytoplasm	Enveloped virions in cytoplasm	Extracellular enveloped virions
ΔgE	60.1 ± 4.9 (766)	1.1 ± 0.4 (19)	2.1 ± 0.3 (27)	28.8 ± 4.0 (375)	7.9 ± 1.4 (84)	1,271/12
ΔgE-repair	59.3 ± 4.3 (688)	0.7 ± 0.3 (12)	1.5 ± 0.4 (14)	8.4 ± 0.9b (97)	30.0 ± 4.5c (261)	1,072/12

^aNumbers in parentheses are the numbers of virus particles.

^bStatistically significant difference from the ΔgE-infected cells (P = 6.1 × 10⁻⁵ by Student's t test).

^cStatistically significant difference from the ΔgE-infected cells (P = 1.8 × 10⁻⁴ by Student's t test).

FIG 11.

Effect of decreased PHB1 expression and gE-null mutation on HSV-1 morphogenesis in HaCaT cells. (A) HaCaT or HaCaT/PHB1-low cells were infected by wild-type HSV-1(F) at an MOI of 15. At 20 h postinfection, the cells were embedded, sectioned, stained, and examined by electron microscopy. The images labeled a to d on the right are magnifications of the corresponding boxed areas on the left. Images are representative of results of 3 independent experiments. Scale bars, 500 nm. (B) The mean proportions of enveloped virions in the cytoplasm and extracellular space of 10 infected cells were determined. (C) HaCaT cells were infected with HSV-1 ΔgE or HSV-1 ΔgE-repair at an MOI of 15. At 20 h postinfection, the cells were fixed embedded, sectioned, stained, and examined by transmission electron microscopy. The images labeled a to d on the right are magnifications of the corresponding boxed areas on the left. Images are representative of results of 3 independent experiments. Scale bars, 500 nm. (D) The percentages of enveloped virions in the cytoplasm and enveloped extracellular virions of 15 infected cells were determined. ***, P < 0.001 (by Student's t test).

TABLE 5.

Effect of depression of PHB on the distribution of virus particles in HaCaT cells

Cell line	Mean ± SE of % of virus particles in each morphogenetic stagea					Total no. of counted particles/ no. of cells
	Nucleocapsids in nucleus	Enveloped virions in perinuclear space	Nucleocapsids in cytoplasm	Enveloped virions in cytoplasm	Extracellular enveloped virions
HaCaT	41.2 ± 2.1 (531)	2.6 ± 0.3 (33)	3.9 ± 0.4 (48)	10.8 ± 0.7 (136)	41.4 ± 1.6 (521)	1,269/10
HaCaT/PHB1-low	37.2 ± 1.2 (469)	3.6 ± 0.3 (46)	4.7 ± 0.4 (58)	34.8 ± 1.2b (441)	19.7 ± 1.2c (248)	1,262/10

^aNumbers in parentheses are the numbers of virus particles.

^bStatistically significant difference from HaCaT cells (P = 9.9 × 10⁻¹³ by Student's t test).

^cStatistically significant difference from HaCaT cells (P = 2.5 × 10⁻⁹ by Student's t test).

TABLE 6.

Effect of gE null mutation on the distribution of virus particles in HaCaT cells

Treatment	Mean ± SE of % of virus particles in each morphogenetic stagea					Total no. of counted particles/ no. of cells
	Nucleocapsids in nucleus	Enveloped virions in perinuclear space	Nucleocapsids in cytoplasm	Enveloped virions in cytoplasm	Extracellular enveloped virions
ΔgE	59.3 ± 1.2 (905)	4.3 ± 0.6 (67)	5.3 ± 0.4 (80)	21.3 ± 1.0 (327)	9.9 ± 0.7 (150)	1,529/15
ΔgE-repair	44.9 ± 1.1 (756)	4.4 ± 0.3 (74)	4.1 ± 0.4 (69)	8.3 ± 0.7b (139)	38.3 ± 1.1c (644)	1,682/15

^aNumbers in parentheses are the numbers of virus particles.

^bStatistically significant difference from the ΔgE-infected cells (P = 1.9 × 10⁻¹¹ by Student's t test).

^cStatistically significant difference from the ΔgE-infected cells (P = 3.7 × 10⁻¹⁹ by Student's t test).

The MAPK/ERK signaling pathway contributes to gE-dependent HSV-1 cell-to-cell spread.

One of the physiological roles of PHB1 is that of acting as a protein scaffold required for the Ras/Raf-induced mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) signaling pathway, which mediates various biological functions such as cell proliferation, survival, and differentiation (18). To investigate whether the MAPK/ERK pathway is involved in gE-dependent HSV-1 cell-to-cell spread, we first examined the effect of U0126 (19), an inhibitor of MEK, which is an upstream kinase of ERK, on HSV-1 cell-to-cell spread in A549 cells. As observed in infected A549 and HaCaT cells with decreased PHB1 expression, the U0126 treatment of HSV-1-infected A549 and HaCaT cells led to reductions in plaque sizes, cytoplasmic accumulation of enveloped virions, and reduced levels of enveloped virions in the extracellular space, without affecting yield of progeny virus (Fig. 12 and 13; see also Tables 7 and 8). These results suggest that the MAPK/ERK signaling cascade is required for HSV-1 cell-to-cell spread and the transport of enveloped virions in the cytoplasm.

FIG 12.

Effects of treatment by the MEK inhibitor U0126 on HSV-1 infection in A549 cells. (A) A549 cells were infected by HSV-1 EGFP at an MOI of 0.001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with or without 10 mM U0126. The areas of 15 individual plaques of each of the infected cell cultures with or without U0126 were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. ***, P < 0.001 (by Student's t test). (B) A549 cells were infected with wild-type HSV-1(F) at an MOI of 5 or 0.05 with or without 10 μM U0126. At 18 h and 48 h postinfection, respectively, total virus from the cell culture supernatants and the infected cells was harvested and assayed on Vero cells. Each bar represents the mean ± standard error of results from 3 independent experiments. n.s., not significant by Student's t test. (C) A549 cells were infected with wild-type HSV-1(F) at an MOI of 15. At 1 h postinfection, the infected cells were incubated with or without 10 mM U0126. These cells were fixed at 20 h postinfection, embedded, sectioned, stained, and examined by transmission electron microscopy. The images labeled a to d on the right are magnifications of the corresponding boxed areas on the left. Images are representative of results of 3 independent experiments. Scale bars, 500 nm. (D) The mean percentages of enveloped virions in the cytoplasm and extracellular space of 8 infected cells were determined. *, P < 0.05; **, P < 0.01 by Student's t test.

FIG 13.

Effects of treatment with the MEK inhibitor U0126 on HSV-1 infection in HaCaT cells. (A) HaCaT cells were infected by HSV-1 EGFP at an MOI of 0.0001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with or without 10 mM U0126. The areas of 15 individual plaques of each of the infected cell cultures with or without U0126 were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. ***, P < 0.001 (by Student's t test). (B) HaCaT cells were infected with wild-type HSV-1(F) at an MOI of 5 or 0.05 with or without 10 μM U0126. At 18 h or 48 h postinfection, total virus from the cell culture supernatants and the infected cells was harvested and assayed on Vero cells. Each bar represents the mean ± standard error of results from 3 independent experiments. n.s., not significant by Student's t test. (C) HaCaT cells were infected with wild-type HSV-1(F) at an MOI of 15. At 1 h postinfection, the infected cells were incubated with or without 10 mM U0126. These cells were fixed at 20 h postinfection, embedded, sectioned, stained, and examined by transmission electron microscopy. The images labeled a to d on the right are magnifications of the corresponding boxed areas on the left. Images are representative of results of 3 independent experiments. Scale bars, 500 nm. (D) The mean percentages of enveloped virions in the cytoplasm and extracellular space of 10 infected cells were determined. ***, P < 0.001 (by Student's t test).

TABLE 7.

Effect of treatment by the MEK inhibitor on the distribution of virus particles in A549 cells

Treatment	Mean ± SE of % of virus particles in each morphogenetic stagea					Total no. of counted particles/ no. of cells
	Nucleocapsids in nucleus	Enveloped virions in perinuclear space	Nucleocapsids in cytoplasm	Enveloped virions in cytoplasm	Extracellular enveloped virions
DMSO	65.2 ± 7.1 (760)	0.7 ± 0.3 (8)	1.1 ± 0.5 (13)	7.3 ± 2.6 (86)	25.6 ± 4.9 (280)	1,147/8
U0126	62.1 ± 4.9 (660)	1.1 ± 0.3 (12)	2.6 ± 1.1 (27)	22.9 ± 3.6b (241)	11.2 ± 2.2c (114)	1,054/8

^aNumbers in parentheses are the numbers of virus particles.

^bStatistically significant difference from the DMSO-treated cells (P = 0.0032 by Student's t test).

^cStatistically significant difference from the DMSO-treated cells (P = 0.018 by Student's t test).

TABLE 8.

Effect of treatment by the MEK inhibitor on the distribution of virus particles in HaCaT cells

Treatment	Mean ± SE of % of virus particles in each morphogenetic stagea					Total no. of counted particles/ no. of cells
	Nucleocapsids in nucleus	Enveloped virions in perinuclear space	Nucleocapsids in cytoplasm	Enveloped virions in cytoplasm	Extracellular enveloped virions
DMSO	42.8 ± 2.5 (556)	3.0 ± 0.3 (39)	3.6 ± 0.3 (47)	10.6 ± 0.6 (138)	39.9 ± 2.7 (519)	1,299/10
U0126	59.4 ± 1.8 (716)	3.6 ± 0.1 (44)	4.6 ± 0.4 (56)	23.3 ± 1.7b (284)	9.1 ± 0.7c (111)	1,211/10

^aNumbers in parentheses are the numbers of virus particles.

^bStatistically significant difference from the DMSO-treated cells (P = 1.2 × 10⁻⁶ by Student's t test).

^cStatistically significant difference from the DMSO-treated cells (P = 2.1 × 10⁻⁹ by Student's t test).

Epidermal growth factor (EGF) stimulates various intracellular signaling pathways, including the MAPK/ERK, phosphatidylinositol-3-kinase/protein kinase B (P13K/Akt), and Janus kinase/signal transducers and activators of transcription (JAK/STAT) pathways (20). We next investigated whether EGF restored the impaired HSV-1 cell-to-cell spread seen in cultures of A549 cells infected by HSV-1 ΔgE. The treatment of A549 cells with EGF led to significantly increased plaque sizes produced by HSV-1 ΔgE/Venus, although it had little effect on plaque sizes produced by HSV-1 EGFP or HSV-1 ΔgE-repair/Venus (Fig. 14A). Furthermore, treatment of PHB1-depleted or RocA-treated A549 cells with EGF had little effect on plaque sizes produced by HSV-1 EGFP and/or HSV-1 ΔgE-repair/Venus (Fig. 14A and B). Notably, the increase in plaque sizes of HSV-1 ΔgE/Venus resulting from the EGF treatment of A549 cells was blocked by RocA and U0126 treatments but not by treatment with an inhibitor (MK-2206) (21) of the PI3K/Akt pathway or either inhibitor (ruxolitinib [22] or tofacitinib [23]) of the JAK/STAT pathway (Fig. 14A and C). We verified that treatment of infected A549 cells with U0126, MK-2206, and ruxolitinib or tofacitinib led to reduced phosphorylation of ERK, AKT, and STAT3, respectively (Fig. 14D), indicating that each of these inhibitors in fact blocked the MAPK/ERK, PI3K/AKT, and JAK/STAT pathways, respectively, in infected A549 cells. Similar results were also obtained with HaCaT cells (Fig. 15). These results indicate that the EGF-stimulated MAPK/ERK signaling pathway is able to compensate for the absence of gE function in HSV-1 cell-to-cell spread in a manner dependent on PHB1. Altogether, these results suggest that the PHB1-dependent MAPK/ERK pathway specifically contributes to gE-dependent HSV-1 cell-to-cell spread.

FIG 14.

Effects of EGF treatment on HSV-1 cell-to-cell spread in A549 cells. (A) A549 cells were infected with HSV-1 EGFP, HSV-1 ΔgE/Venus, or HSV-1 ΔgE-repair/Venus at an MOI of 0.001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with 20 nM RocA and/or 20 ng/ml of EGF, as indicated. The areas of 15 individual plaques corresponding to each of the indicated viral infections/culture conditions were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. (B) A549 or A549/PHB1-low cells were infected with HSV-1 EGFP at an MOI of 0.001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with or without 20 ng/ml of EGF. The areas of 15 individual plaques of each of the indicated infected cell lines with or without EGF were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. (C) A549 cells were infected with HSV-1 EGFP, HSV-1 ΔgE/Venus, or HSV-1 ΔgE-repair/Venus at an MOI of 0.001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with or without 10 mM U0126, MK-2206, ruxolitinib, or tofacitinib combined with 20 ng/ml of EGF as indicated. The areas of 15 individual plaques corresponding to each of the indicated viral infections/culture conditions were measured 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. **, P < 0.01; ***, P < 0.001 by Tukey test; n.s., not significant. (D) A549 cells were incubated with or without 10 mM U0126, MK-2206, ruxolitinib, or tofacitinib combined with 20 ng/ml of EGF for 24 h, and then harvested and analyzed by immunoblotting with the indicated antibodies. Images are representative of results of 3 independent experiments.

FIG 15.

Effects of EGF treatment on HSV-1 cell-to-cell spread in HaCaT cells. (A) HaCaT cells were infected with HSV-1 EGFP, HSV-1 ΔgE/Venus, or HSV-1 ΔgE-repair/Venus at an MOI of 0.0001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with 20 nM RocA and/or 20 ng/ml of EGF, as indicated. The areas of 15 individual plaques corresponding to each of the indicated viral infections/culture conditions were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. (B) HaCaT or HaCaT/PHB1-low cells were infected with HSV-1 EGFP at an MOI of 0.0001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with or without 20 ng/ml of EGF. The areas of 15 individual plaques of each of the indicated infected cell lines with or without EGF were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. Data are representative of results of 3 independent experiments. (C) HaCaT cells were infected with HSV-1 EGFP, HSV-1 ΔgE/Venus, or HSV-1 ΔgE-repair/Venus at an MOI of 0.0001. At 1 h postinfection, the infected cells were incubated under plaque assay conditions with or without 10 mM U0126, MK-2206, ruxolitinib, or tofacitinib combined with 20 ng/ml of EGF as indicated. The areas of 15 individual plaques corresponding to each of the indicated viral infections/culture conditions were measured 48 h postinfection. Each plot shows the mean area ± standard error of 15 plotted plaque areas. **, P < 0.01; ***, P < 0.001 (by Tukey test; n.s., not significant). (D) HaCaT cells were incubated with or without 10 mM U0126, MK-2206, ruxolitinib, or tofacitinib combined with 20 ng/ml of EGF for 24 h and were harvested and analyzed by immunoblotting with the indicated antibodies. Images are representative of results of 3 independent experiments.

Effects of inhibition of PHB1 or the MAPK/ERK signaling pathway on cell-to-cell spread by various herpesviruses.

To investigate whether PHB1 and the MAPK/ERK signaling pathway are also involved in the cell-to-cell spread of cells infected by other herpesviruses, we examined the effects of RocA and U0126 on viral plaque formation by various herpesviruses, including HSV-2 and PRV, the murine cytomegalovirus (MCMV), and murine gammaherpesvirus 68 (MHV-68), which belong to the Alphaherpesvirinae, Betaherpesvirinae, and Gammaherpesvirinae subfamilies, respectively. As shown in Fig. 16 and 17, RocA or U0126 treatment of A549 cells infected by EGFP-expressing or monomeric GFP (mGFP)-expressing recombinants HSV-2, PRV, MCMV, and MHV-68 led to significant reductions in plaque sizes without affecting the yields of progeny virus. These results suggest that representative members of every Herpesviridae subfamily commonly use PHB1 and the MAPK/ERK pathway for efficient viral cell-to-cell spread.

FIG 16.

Effects of RocA or U0126 treatment on cell-to-cell spread of various herpesviruses. A549, NIH 3T3, or BHK cells were infected with HSV-2 EGFP, PRV EGFP, MCMV EGFP, or MHV-68 EGFP at an MOI of 0.001. The infected cell cultures were incubated under plaque assay conditions with 20 nM RocA or 10 mM U0126. The diameters of 15 individual plaques for HSV-2 EGFP, PRV EGFP, and MHV-68 insGFP or of 20 individual plaques for MCMV EGFP were measured at 48 h postinfection. Each plot shows the mean area ± standard error of 15 or 20 plotted plaque areas. Data are representative of results of 3 independent experiments. ***, P < 0.001 by Student's t test.

FIG 17.

Effects of RocA or U0126 treatment on progeny virus yields of various herpesviruses. A549, NIH 3T3, or BHK cells were infected with HSV-2 EGFP, PRV EGFP, MCMV EGFP, or MHV-68 insGFP at an MOI of 5 or 0.05. At 18 h and 48 h postinfection, respectively, total virus from the cell culture supernatants and the infected cells was harvested and assayed on Vero cells. n.s., not significant by Student's t test.

DISCUSSION

Virus factors involved in viral cell-to-cell spread often control viral entry and/or replication; therefore, identifying the mechanisms specific to viral cell-to-cell spread based on evaluation of these viral factors is difficult. However, viral and host cell proteins that contribute to viral cell-to-cell spread without any obvious role in viral replication have been identified, including the gEs of HSV-1 and PRV (1–3, 8, 13, 14). Two viral proteins identified in vaccinia virus infection, F11L and vaccinia growth factor (VGF), were previously shown to stimulate the mobility of infected cells by hijacking the activity of the Ras homolog gene family, member A (RhoA-), and of the epidermal growth factor receptor (EGFR)/MAPK/ERK signaling pathway, respectively, thereby promoting viral cell-to-cell spread (24–26). However, the mechanisms by which viral proteins other than these vaccinia proteins promote viral cell-to-cell spread are poorly understood. In this report, we present evidence suggesting that host cell protein PHB1 specifically facilitates gE-dependent HSV-1 cell-to-cell spread by a MAPK/ERK signaling pathway-dependent mechanism. Importantly, pharmacological inhibition of PHB1 or the MAPK/ERK pathway impaired the cell-to-cell spread of a representative member of each Herpesviridae subfamily without affecting viral replication, suggesting that the role of the PHB1-dependent MAPK/ERK pathway in viral cell-to-cell spread is conserved throughout the Herpesviridae family. As described above, herpesviruses and unrelated vaccinia virus commonly expropriate the MAPK/ERK pathway for efficient viral cell-to-cell spread; however, the downstream effects of the signaling pathway appear to differ between herpesviruses and vaccinia virus (26).

Although two series of experiments in this study demonstrated an interaction between gE and PHB1 in HSV-1-infected cells and in vitro, how the gE/PHB1 axis is involved in MAPK/ERK-mediated HSV-1 cell-to-cell spread remains unclear. It was previously reported that gE targets the plasma membrane in HSV-1-infected cells (9). We showed that EGF, a stimulator of Ras/Raf-induced MAPK/ERK signaling, compensated for the functions of gE in HSV-1 cell-to-cell spread in cells infected with a gE-null mutant virus. Together, these observations suggested that gE interacts with PHB1 to activate the MAPK/ERK signaling pathway by recruiting Raf to the plasma membrane (18) in the absence of any upstream stimulators, such as vaccinia virus VGF (26), thereby promoting MAPK/ERK-mediated HSV-1 cell-to-cell spread. Further studies designed to investigate the effects of the interaction between gE and PHB1 on Ras/Raf-induced MAPK/ERK signaling in HSV-1-infected cells and viral cell-to-cell spread are needed.

RocA, a PHB1 inhibitor used in this study, also inhibits an important translational factor, eukaryotic initiation factor 4A (eIF4A), and downregulates global cellular protein synthesis, leading to reduced cell proliferation (27, 28). Therefore, the effects of RocA in HSV-1-infected cells observed in this study might be associated with the inhibition of eIF4A and not PHB1. However, this is unlikely because the effects of RocA treatment on HSV-1 cell-to-cell spread and viral morphogenesis in the cytoplasm were also observed after treatment with U0126, a specific inhibitor of MEK, and PHB1 depletion and/or gE-null mutation, all of which appear not to be involved in the inhibition of eIF4A. Furthermore, RocA at the concentration (20 nM) used in this study was previously reported to inhibit cellular translation only partly (28) and, in agreement with this, RocA treatment impaired HSV-1 cell-to-cell spread but not virus progeny yields.

We found that ectopic expression of PHB1 in HeLa cells promoted HSV-1 cell-to-cell spread. In these experiments, ectopically expressed PHB1 levels were lower than endogenous PHB1 levels. Unexpectedly, a marginal increase in PHB1 expression mediated via ectopic expression resulted in increased HSV-1 cell-cell spread in HeLa cells. However, similarly to our observations, previous studies reported that a small amount of ectopically expressed PHB1 relative to the amount of endogenous PHB1 induced various phenotypes (29–31). Together, these observations suggest that tight regulation of PHB1 expression may be critical for its function and that endogenous expression of PHB1 in HeLa cells might be not quite sufficient for the formation of plaques by HSV-1 as well as other functions reported in previous studies (29–31). Therefore, a comparatively small amount of ectopically expressed PHB1 was able to increase the levels of these PHB1 functions.

Results of electron microscopic analyses performed in this study showed that PHB1 depletion as well as the gE-null mutation caused the accumulation of enveloped virions in the cytoplasm and reduced levels of virions on the cell surface, suggesting that PHB1 is required for the efficient transport of nascent enveloped virions to the cell surface. Currently, whether PHB1, similarly to gE, has a role in HSV-1 cell-to-cell spread by promoting the transport of nascent enveloped virions to the junctional cell surface is unclear. However, it seems likely, based on the following points. Our results suggest two possibilities: (i) PHB1, similarly to gE, contributes to the trafficking of enveloped virions specifically to the junctional cell surface for cell-to-cell spread or (ii) PHB1 has a role in the general movement of enveloped virions to the cell surface. In the latter case, it may be expected that PHB1 depletion would decrease virus titers in the supernatants of infected cells and increase virus titers associated with infected cells. In contrast, in the former case, PHB1 depletion may not necessarily decrease virus titers in the supernatants of infected cells or increase cell-associated virus titers because virions at the junctional cell surfaces of infected cells might be detected as virus titers associated with infected cells. We observed that PHB1 depletion, similarly to results seen with the gE-null mutation, had no obvious effect on virus titers in supernatants or those associated with infected cells (Fig. 18). Thus, PHB1 depletion resulted in a phenotype similar to that seen with the gE-null mutation. Taken together, these results suggest a role for PHB1 in the transport of enveloped virions specifically to the junctional cell surface, similarly to gE, thereby facilitating HSV-1 cell-to-cell spread.

FIG 18.

Effect of decreased PHB1 expression and gE-null mutation on progeny HSV-1 yields in supernatants and those associated with infected-cells. A549 or A549/PHB1-low cells (A and B) or HaCaT or HaCaT/PHB1-low cells (C and D) were infected with wild-type HSV-1(F), HSV-1 ΔgE, or HSV-1 ΔgE-repair at an MOI of 5. Progeny virus from the cell culture supernatants (left panels) and that from the infected cells (right panels) were separately harvested at the indicated times postinfection and assayed on Vero cells. Each data point represents the mean ± standard error of data from 3 independent experiments. n.s., not significant by Student's t test.

HSV-1 gE is conserved in alphaherpesviruses but not in betaherpesviruses or gammaherpesviruses (4). The functional counterparts for HSV-1 gE in betaherpesviruses and gammaherpesviruses might have evolved to allow efficient viral cell-to-cell spread, and PHB1 might contribute to gE-counterpart–dependent viral cell-to-cell spread through the MAPK/ERK pathway. Reports that PHB1 also interacts with the envelope proteins of various viruses unrelated to herpesviruses (32–35) and that various betaherpesviruses and gammaherpesviruses modulate the MAPK/ERK pathway during infection support this hypothesis (36, 37). However, PRV upregulated ERK activity in infected porcine epithelial cells, and the PRV gE homolog was required for ERK activation in these infected cells (38). Those previous findings, as well as the results of our study showing that the pharmacological inhibition of PHB1 and the MAPK/ERK pathway impaired PRV cell-to-cell spread, support our conclusion that gE-mediated HSV-1 cell-to-cell spread is regulated by the PHB1-dependent MAPK/ERK pathway and suggest that this HSV-1 mechanism is conserved in alphaherpesviruses. Similarly to PRV, HSV-2 activates ERK in infected human cell lines; however, ERK activation, which occurs very early in HSV-1 infections of human cell lines, was transient and was suppressed in late HSV-1 infections of these cell lines (39). Therefore, the transient and/or spatially finite activation of the PHB1-mediated MAPK/ERK pathway might be required for gE-dependent HSV-1 cell-to-cell spread. In agreement with this hypothesis, we demonstrated that treatment with EGF, which transiently stimulates the MAPK/ERK pathway (40), restored the gE-null-mutation-mediated impairment of HSV-1 cell-to-cell spread.

The MAPK/ERK pathway regulates the transport of various cytoplasmic vesicles (41–43). For example, in response to EGF signaling, ERK phosphorylates and activates Rabin8, a major activator of Rab8, and promotes Rab8-mediated vesicular trafficking from the trans-Golgi network (TGN) (expropriated by herpesviruses to transport enveloped virions to the cell surface for secretion) to the plasma membrane (41, 44). Interestingly, vesicles containing enveloped PRV virions were associated with Rab6, Rab8, and Rab11 in the cytoplasm (45). Therefore, it would be of interest to identify the specific vesicular trafficking mechanism of enveloped herpesvirus virions that is responsible for viral cell-to-cell spread based on the findings of our study indicating the mechanism is downstream of the MAPK/ERK pathway.

Protein tyrosine phosphatase 1B (PTP1B, also known as tyrosine-protein phosphatase nonreceptor type 1 [PTPN1]) was recently identified as the host cell protein specifically required for HSV-1 cell-to-cell spread, although the molecular mechanism involved is unknown (11). Interestingly, protein tyrosine phosphatase interacting protein 51 (PTPIP51), a binding partner and substrate of PTP1B (46), activated the MAPK/ERK pathway by interacting with Raf (46–48). Furthermore, the interaction between PTPIP51 and Raf was negatively regulated by phosphorylation of the tyr-176 residue on PTPIP51 (46, 49), and pharmacological inhibition of PTP1B impaired the interaction between PTPIP51 and Raf because of the higher level of tyr-176 phosphorylation on PTPIP51 (46, 50). These observations suggest that PTP1B might be a signaling molecule in the PHB1-mediated MAPK/ERK pathway in HSV-1-infected cells, thereby promoting HSV-1 cell-to-cell spread. Supporting this hypothesis, our proteomic analyses showed an association between PTP1B and PHB1, based on the observation that gE pulled down PTP1B (PTPN1) together with PHB1 from the lysates of HSV-1-infected HaCaT cells (data not shown).

In addition to the roles of PHB1 in herpesviruses shown in this study, other viruses, including hepatitis C virus (HCV), dengue virus 2, Chikungunya virus, influenza virus, lymphocytic choriomeningitis virus, and enterovirus 71, were reported to take advantage of PHB1 in their infected host cells (33, 34, 51–55). In most cases, PHB1 interacts with viral envelope proteins such as HSV-1 gE and contributes to the viral entry of the host cell and/or replication (33, 34). We demonstrated that PHB1 is required by HSV-1 to induce signaling pathways in HSV-1-infected cells, and this might also be true for HCV, which is unrelated to HSV-1. HCV exploits PHB1 for viral entry by regulating the EGFR/Ras/Raf pathway, which is mediated by an HCV envelope glycoprotein, upstream of the MAPK/ERK pathway (52). Thus, PHB1 and these signaling pathways can be targeted by many viruses for the expropriation of host cellular machinery for efficient viral infection. This suggests that PHB1 and the pathways that it regulates might be attractive targets for novel broad-spectrum antiviral drugs directed against RNA and DNA viruses.

MATERIALS AND METHODS

Cells and viruses.

The Vero, A549, HeLa, HaCaT, RSC, Plat-GP, NIH 3T3, BHK, and P3U1 cell lines used in the study were described previously (56–61). Wild-type HSV-1(F), gE-null HSV-1 YK640 (ΔgE), and the repaired virus HSV-1 YK641 (ΔgE-repair) were described previously (13). The recombinant virus HSV-1 YK333 (HSV-1 EGFP) carried an EGFP expression cassette in the UL3-UL4 intergenic region, which is under the control of the Egr-1 promoter and is phenotypically the same as that carried by wild-type HSV-1(F) in cell cultures (12). The recombinant viruses HSV-2 YK381 (HSV-2 EGFP), PRV152 (PRV EGFP), and MCMV EGFP were described previously (60, 62–64).

Construction of MHV-68 insGFP.

To construct plasmid pBeloBAC11MHV68insGFP, fragments containing MHV68 open reading frame 11 (ORF11) (nucleotides [nt] 23932 to 24279) and ORF12 (nt 24923 to 25330) were amplified from the MHV68 genome (MHV4 WMUS strain; ATCC VR1465) and cloned into Flippase recognition target (FRT)-franked pBeloBAC11 along with a fragment containing the human EF1α promoter (EF1) and the coding sequence of the mutant mGFP, both of which were flanked by the CTCF recognition sequence. Plasmid pBeloBAC11MHV68insGFP was cotransfected with MHV-68 DNA into BHK21 cells by electroporation. The resulting homologous recombinant MHV-68 genomic DNA (pBAC-MHV-68-insGFP) was isolated from the transfected BHK21 cells and then electroporated into Escherichia coli DH10B, which was cultured with chloramphenicol. To reconstitute recombinant virus, pBAC-MHV-68-insGFP was isolated from the bacteria and transfected into BHK21 cells expressing Flip E to excise the pBeloBAC11 backbone fragments without the mGFP expression cassette. Recombinant virus MHV68-insGFP was selected from GFP-positive plaques and expanded by infecting the P3U1 mouse myeloma cell line.

Plasmids.

pMXs-PHB1-Myc-puro, a retrovirus vector that expresses PHB1, which is fused to the Myc tag, was constructed by PCR amplification of the PHB1 ORF sequence from the cDNA synthesized from the total RNA of A549 cells, as described previously (59), and was cloned into pMxs-puro.

To produce pX458-PHB1, sense and antisense oligonucleotides were designed for insertion into the BbsI site in the pX458 bicistronic expression vector, which expresses Cas9 and a synthetic single-guide RNA (Addgene), as follows: 5′-CACCGTGAACTCTGCCTTATATAA-3′ and 5′-AAACTTATATAAGGCAGAGTTCAC-3′. The DNA oligonucleotides were annealed and incorporated into the pX458 vector linearized with the BbsI restriction enzyme.

Plasmid pBS-Venus-KanS was constructed by amplifying the kanamycin resistance cassette and I-SceI recognition site from pEPkan-S (65) and cloned into the PstI site of pBS-Venus (57). Plasmid pRB-EGRp-Venus-KanS-polyA was constructed by amplifying the ORF of Venus with the kanamycin resistance cassette and I-SceI recognition site from plasmid pBS-Venus-KanS into plasmid pRB5160 (66).

Plasmid pGEX-PHB1-NTD or plasmid pGEX-PHB1-CTD, encoding a fusion protein of GST and the amino-terminal domain of PHB1 (GST-PHB1-NTD) or the carboxyl-terminal domain of PHB1 (GST-PHB1-CTD), was constructed by cloning PCR-amplified PHB1 codons 1 to 218 or codons 219 to 272, respectively, into pGEX-4T-1 (GE Healthcare) in frame with GST (Fig. 2C).

Mutagenesis of viral genomes and generation of recombinant viruses.

Recombinant viruses (Fig. 1) were constructed by a two-step Red-mediated mutagenesis procedure using E. coli GS1783, which carries pYEbac102, a full-length infectious HSV-1(F) clone, as described previously (67). Recombinant virus YK644 (HSV-1 gE-Strep), encoding Strep-tagged gE with the Strep epitope at its carboxyl terminus, was constructed using the following primers: 5′-CGCCGTTACTCCCAGGCCTCCGATTCGTCCGTCTTCTGGGGTGGAGGTTGGAGCCACCCGCAGTTCGAGAAATAAAGGATGACGACGATAAGTAGGG-3′ and 5′-GTTCGGCGACCGACGTGGGGCCTCGGGATGGGGCGCCTTATTTCTCGAACTGCGGGTGGCTCCAACCTCCACCCCAGAAGCAACCAATTAACCAATTCTGATTAG-3′.

Recombinant virus YK648 (Strep-vTK), encoding Strep-tagged UL23 (vTK) with the Strep epitope at its C terminus, was generated by two-step Red-mediated mutagenesis procedures using the pYEbac102Cre genome (68), primers 5′-CCTGGCGCGCACGTTTGCCCGGGAGATGGGGGAGGCTAACATGGCTAGCTGGAGCCACCC-3′ and 5′-AGCGCGGGTTCCTTCCGGTATTGTCTCCTTCCGTGTTTCAACCCTGAAAATACAAATTCT-3′, and pEP-KanS-SEM (69, 70).

Recombinant virus YK647, which encodes gE tagged with Flag-TEV-Myc (gE-FEM) at its carboxyl terminus, was constructed using the following primers: 5′-GAAGGCCCGATCGCCGTTACTCCCAGGCCTCCGATTCGTCCGTCTTCTGGGACTACAAGGACGACGATGA-3′ and 5′-CGGTCGCCCAGTTCGGCGACCGACGTGGGGCCTCGGGATGGGGCGCCTTACAAGTCCTCTTCAGAAATGA-3′.

gE null HSV-1 YK645 (HSV-1 ΔgE/Venus), in which the Venus expression cassette had been inserted into the intergenic region between UL50 and UL51, was generated by the Red-mediated mutagenesis procedure using the YK640 (HSV-1 ΔgE) genome, primers 5′-TATCTCATCTTTCCTGTGTGTAGTTGTTTCTGTTGGAGGCCTGTGGGTTATGCGCCGACCCGGAAACGCC-3′ and 5′-TTCATCCAACCCGTGTGTTCTGTGTTTGTGGGATGGAGGGGCGGGTTAATGGACAAGTGTCCCGTTTTTT-3′, and pRB-EGRp-Venus-KanS-polyA.

The recombinant virus YK646 (HSV-1 ΔgE-repair/Venus), in which the gE-null mutation in YK645 (HSV-1 ΔgE/Venus) was repaired, was constructed using the following primers: 5′-TATCGTGCTTGGCGGGAACGCCCAAAACGTCCTGGAGACGGGTGAGTGTCGGCGAGGACGAGGATGACGACGATAAGTAGGG-3′ and 5′-CGGAGCTGGAAGCAACGAAACGTCCTCGCCGACACTCACCCGTCTCCAGGACGTTTTGGGCAACCAATTAACCAATTCTGATTAG-3′.

Identification of proteins that interact with gE.

HaCaT cells were infected with YK647 (HSV-1 gE-FEM) at a multiplicity of infection (MOI) of 3. At 36 h postinfection, the infected cells were lysed in a digitonin buffer (0.5% digitonin, 50 mM Tris-HCl [pH 8.0], 120 mM NaCl, 50 mM 0.1% NaF) containing protease and phosphatase inhibitor cocktails (Nacalai Tesque). After centrifugation of the lysate, the supernatant was immunoprecipitated with an anti-Myc monoclonal antibody (MBL), and the immunoprecipitation product was incubated with AcTEV protease (Invitrogen) for 1 h at room temperature. After another centrifugation step, the supernatant was immunoprecipitated with an anti-Flag M2 affinity gel (Sigma), and the immunoprecipitation products were washed 2 times with digitonin buffer. The products were then eluted by rotation with a Flag elution buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.5 mg Flag peptide/ml) for 2 h at 4°C. The eluted proteins were digested into peptides based on the filter-aided sample preparation (FASP) method (71). Briefly, the eluted protein solution was loaded onto a centrifugal filter (Amicon Ultra 0.5-ml centrifugal filter 30K; Merck Millipore), and the detergent was removed by washing 4 times with a filter buffer (50 mM Tris-HCl [pH 8.2], 0.15 M NaCl). The proteins were then reduced by incubation with 1 mM dithiothreitol for 90 min, alkylated with 5.5 mM iodoacetamide for 30 min, treated by buffer exchange with 50 mM ammonium bicarbonate, and digested with trypsin for 18 h at 37°C. The peptide solutions were acidified by the use of 1% trifluoroacetic acid, desalted with ZipTip C18 resin (Merck Millipore), and evaporated to a maximum volume of 10 μl by a vacuum concentrator. Peptides were analyzed by the use of a Dina-2A nanoflow liquid chromatography (LC) system (KYA Technologies) coupled with a LTQ-Orbitrap Velos mass spectrometer (Thermo Fisher Scientific).

Proteins were identified by analysis of the MS and the tandem MS (MS/MS) signals, using the Mascot algorithm (Matrix Science), with reference to the 68,711 protein sequences in the RefSeq human protein database (National Center for Biotechnology Information). The following parameters were used: for variable modifications, oxidation (Met), protein N-terminal acetylation, pyroglutamination for N-terminal glutamine, and phosphorylation (Ser, Thr, and Tyr); for fixed modifications, carbamidomethylation of cysteine residues, maximum number of missed cleavages of 2, peptide mass tolerance of 3 ppm, and MS/MS tolerance of 0.8 Da. Protein identification was based on the criterion of at least 1 MS/MS data signal with a Mascot score greater than the threshold (P < 0.01). Phosphorylation sites were identified by Proteome Discoverer ver. 1.3 (Thermo Fisher Scientific).

Expression screening.

Murine leukemia virus (MLV) retroviral vectors carrying the cDNA library from Vero cells were prepared as described previously (72). MLVs were used to infect HeLa cells. Library-transduced HeLa cells were seeded into multiwell plates and infected with YK333 (HSV-1 EGFP) under plaque assay conditions. Fluorescent microscopy was used to examine the infected cultures, and plaques showing increased numbers of EGFP-positive regions were collected. The plaque material was treated with a lysis buffer (150 mM NaCl, 10 mM [Tris pH 7.4], 1.5 mM MgCl₂, 1% SDS), and genomic DNA was isolated and subjected to PCR and direct sequencing by the use of primers 5′-GGTGGACCATCCTCTAGACT-3′ and 5′-TCCCCCCTTTTTCTGGAGACTAAATA-3′ to identify transduced cDNAs.

Establishment of HeLa cells expressing PHB1-Myc.

Plat-GP cells, a 293T-derived MLV packaging-based cell line, were cotransfected with pMxs-PHB1-Myc or pMxs-Puro and with pMDG, which encodes vesicular stomatitis virus envelope protein G, using polyethylenimine. The cell culture supernatants were harvested at 48 h posttransfection. HeLa cells then were transduced with the retrovirus-containing supernatants from the transfected Plat-GP cells and selected with 2 μg puromycin/ml. Resistant cells transduced by each recombinant retrovirus derived from the pMxs vectors were cloned from single colonies and designated HeLa/PHB1-Myc or HeLa/puro.

Affinity precipitation.

HaCaT cells were infected with wild-type HSV-1(F), YK644 (gE-strep), or YK648 (vTK-Strep) at an MOI of 5 for 36 h and were then harvested and lysed by the use of digitonin buffer containing a protease inhibitor cocktail. After centrifugation of the lysate, the supernatant was mixed with StrepTactin Sepharose beads (IBA) and incubated with mixing for 1 h at 4°C. The precipitates were collected by brief centrifugation, washed extensively with digitonin buffer, and analyzed by immunoblotting.

GST pulldown.

GST, GST-PHB1-NTD, and GST-PHB1-CTD were each expressed in E. coli and purified on glutathione-Sepharose beads (GE Healthcare Bio-Sciences) as described previously (73). Vero cells were infected with wild-type HSV-1(F) at an MOI of 5 for 24 h and were then lysed in digitonin buffer containing a protease inhibitor cocktail and incubated with purified GST proteins immobilized on glutathione-Sepharose beads for 1 h at 4°C with rotation. Then, the beads were washed extensively with digitonin buffer and analyzed by immunoblotting.

Establishment of A549/PHB1-low and HaCaT/PHB1-low cells.

A549 or HaCaT cells were transfected with pX458-PHB1 by electroporation using a NEPA21 electroporator (Nepa Gene). At 48 h posttransfection, GFP-positive cells were sorted by the use of a FACSAria cell sorter (BD Biosciences) and transferred to a 96-well plate. Further growth of the transfected cells was observed in several wells wherein single colonies were grown. The single colonies were individually picked for further analysis to identify A549/PHB1-low and HaCaT/PHB1-low cells. To determine the genotypes of each allele from A549/PHB1-low or HaCaT/PHB1-low cells, genomic DNA from these cells was amplified by PCR and sequenced directly. In the case of A549/PHB1-low cells, the sequenced PCR products showed mixed sequencing patterns, indicating that the sequences of PHB1 were derived from more than 4 PHB1 alleles. Therefore, the PCR products were cloned into plasmids, and their sequences were determined. We finally obtained 4 sequence patterns that represented PHB1 sequences of the 4 PHB1 alleles and a single pattern that represented a wild-type PHB1 protein from A549/PHB1-low cells (Fig. 4). In the case of HaCaT/PHB1-low cells, sequencing of PCR products showed a single pattern (Fig. 7).

Plat-GP cells were cotransfected with pMxs-puro or pMxs-PHB1-Myc-puro and pMDG. At 48 h after transfection, the supernatants of the transfected cells were harvested, and A549/PHB1-low cells were transduced with these supernatants and selected with 2 μg/ml of puromycin. Resistant cells were designated A549/PHB1-low/puro and A549/PHB1-low/PHB1-Myc. Similarly, A549 cells were transduced with the supernatants described above to obtain A549/puro or A549/PHB1-Myc cells.

Assay for cell viability.

The viability of A549, A549 PHB1-low, HaCaT, or HaCaT-low cells was evaluated using cell-counting kit-8 (Dojindo), according to the manufacturer’s instructions.

Immunoblotting.

Immunoblotting was performed as described previously (69).

Antibodies.

The commercial antibodies used in this study were mouse monoclonal antibodies to gE (9H3; Virusys), Myc (PL14; MBL), Strep-tag (M211-3; MBL), α-tubulin (DM1A; Sigma), STAT3 (CST), phosphor-STAT3 (CST), Akt (CST), phospho-Akt (CST), ERK (CST), and phospho-ERK (CST) and rabbit polyclonal antibodies to PHB1 (ab28172; Abcam).

Determination of plaque areas.

Cells were infected with each of the indicated HSV-1 strains at an MOI of 0.001. After adsorption was performed for 1 h, the inoculum was removed, and the cell monolayers were overlaid with medium 199 containing 1% fetal calf serum (FCS) and 160 g/ml of pooled human immunoglobulin (Sigma) with or without each of the indicated inhibitors. For infections with PRV EGFP, MHV-68 insGFP, or MCMV GFP, the infected cell monolayers were overlaid with medium 199 containing 1% FCS and 0.8% methylcellulose with or without each of the indicated inhibitors. At 2 days postinfection, 15 plaques in each culture were examined under a fluorescence microscope equipped with a digital DP80 camera (Olympus) and CellSens software (Olympus).

Knockdown experiments using siRNAs.

Small interfering RNAs (siRNAs) with target sequences corresponding to PHB1 (5′-UGUCAACAUCACACUGCGCdTdT-3′ and 5′-AGCCAGCUUCCUCGCAUCUdTdT-3′) and a control sequence were purchased from Dharmacon. Vero cells were treated with 10 nM siRNA for 48 h and then infected with wild-type HSV-1 EGFP at an MOI of 0.0001 under plaque assay conditions.

Assay for progeny virus yields.

To examine replication kinetics, cells were infected with each virus of interest at the indicated MOI and incubated for 1 h at 37°C to allow for adsorption. Thereafter, the cell medium was replaced with fresh medium containing 1% fetal calf serum (FCS) with or without each of the indicated inhibitors. Cells and supernatants were harvested at the indicated times after infection. Virus progeny were titrated by plaque assay on Vero cells (for HSV-1, HSV-2 and PRV), NIH 3T3 cells (for MCMV), or BHK cells (for MHV-68).

Inhibitor treatment.

The indicated cells were infected with each of the indicated viruses at the indicated MOI for 1 h. The inoculum was then removed, and the cells were incubated with medium 199 containing 1% FCS and 20 nM RocA (Enzo Life Sciences), 10 μM U0126 (Selleck Chemicals), 10 μM MK-2206 (Selleck Chemicals), 10 μM ruxolitinib (Selleck Chemicals), or 10 μM tofacitinib (Selleck Chemicals) and/or 20 ng/ml of EGF (Gibco) in the presence or absence of 160 g/ml of pooled human immunoglobulin for further analyses.

Electron microscopy.

A549, A549/PHB1-low, HaCaT, or HaCaT/PHB1-low cells were infected with HSV-1(F) at an MOI of 15 for 20 h, and A549 or HaCaT cells were infected with HSV-1 ΔgE or HSV-1 ΔgE-repair at an MOI of 15 for 20 h. The infected cells were examined by ultra-thin-section electron microscopy, as described previously (74). For inhibitor treatment, A549 or HaCaT cells infected with wild-type HSV-1(F) at an MOI of 15 were treated with or without 20 nM RocA or 10 μM U0126 at 1 h postinfection. The infected cells were fixed and examined by ultra-thin-section electron microscopy at 20 h postinfection.

Statistical analysis.

The unpaired Student's t test was used to compare 2 groups. The Tukey test was used for multiple comparisons. A P value of >0.05 was considered not significant (n.s.). Statistical analysis was performed by the use of GraphPad Prism 7 (GraphPad Software).