Regulation of Herpes Simplex Virus 2 Protein Kinase UL13 by Phosphorylation and Its Role in Viral Pathogenesis

Based on studies on cellular protein kinases, it is obvious that the regulatory mechanisms of protein kinases are as crucial as their functional consequences. Herpesviruses each encode at least one protein kinase, but the mechanism by which these kinases are regulated in infected cells remains to be elucidated, with a few exceptions, although information on their functional effects has been accumulating. In this study, we have shown that phosphorylation of the HSV-2 UL13 protein kinase at Ser-18 regulated its function in infected cells, and this regulation was critical for HSV-2 replication and pathogenesis in vivo. UL13 is conserved in all members of the family Herpesviridae, and this is the first report clarifying the regulatory mechanism of a conserved herpesvirus protein kinase that is involved in viral replication and pathogenesis in vivo. Our study may provide insight into the regulatory mechanisms of the other conserved herpesvirus protein kinases.

ABSTRACT

UL13 proteins are serine/threonine protein kinases encoded by herpes simplex virus 1 (HSV-1) and HSV-2. Although the downstream effects of the HSV protein kinases, mostly those of HSV-1 UL13, have been reported, there is a lack of information on how these viral protein kinases are regulated in HSV-infected cells. In this study, we used a large-scale phosphoproteomic analysis of HSV-2-infected cells to identify a physiological phosphorylation site in HSV-2 UL13 (i.e., Ser-18) and investigated the significance of phosphorylation of this site in HSV-2-infected cell cultures and mice. Our results were as follows. (i) An alanine substitution at UL13 Ser-18 (S18A) significantly reduced HSV-2 replication and cell-to-cell spread in U2OS cells to a level similar to those of the UL13-null and kinase-dead mutations. (ii) The UL13 S18A mutation significantly impaired phosphorylation of a cellular substrate of this viral protein kinase in HSV-2-infected U2OS cells. (iii) Following vaginal infection of mice, the UL13 S18A mutation significantly reduced mortality, HSV-2 replication in the vagina, and development of vaginal disease to levels similar to those of the UL13-null and the kinase-dead mutations. (iv) A phosphomimetic substitution at UL13 Ser-18 significantly restored the phenotype observed with the UL13 S18A mutation in U2OS cells and mice. Collectively, our results suggested that phosphorylation of UL13 Ser-18 regulated UL13 function in HSV-2-infected cells and that this regulation was critical for the functional activity of HSV-2 UL13 in vitro and in vivo and also for HSV-2 replication and pathogenesis.

IMPORTANCE Based on studies on cellular protein kinases, it is obvious that the regulatory mechanisms of protein kinases are as crucial as their functional consequences. Herpesviruses each encode at least one protein kinase, but the mechanism by which these kinases are regulated in infected cells remains to be elucidated, with a few exceptions, although information on their functional effects has been accumulating. In this study, we have shown that phosphorylation of the HSV-2 UL13 protein kinase at Ser-18 regulated its function in infected cells, and this regulation was critical for HSV-2 replication and pathogenesis in vivo. UL13 is conserved in all members of the family Herpesviridae, and this is the first report clarifying the regulatory mechanism of a conserved herpesvirus protein kinase that is involved in viral replication and pathogenesis in vivo. Our study may provide insight into the regulatory mechanisms of the other conserved herpesvirus protein kinases.

INTRODUCTION

Protein phosphorylation by protein kinases is one of the most common and effective posttranslational modifications by which viruses regulate the function of their target proteins (1, 2). Interestingly, herpesviruses each encode at least one protein kinase, and these viral protein kinases have been reported to regulate their own viral proteins and their host cell proteins to produce a cellular environment and evade the host immune response for efficient viral replication (3–8). Examples of such herpesvirus protein kinases are herpes simplex virus 1 (HSV-1) Us3, pseudorabies virus Us3, varicella-zoster virus (VZV) open reading frame 66 (ORF66), HSV-1 UL13, VZV ORF47, human cytomegalovirus (HCMV) UL97, Epstein-Barr virus (EBV) BGLF4, and Kaposi's sarcoma-associated herpesvirus (KSHV) ORF36. UL13 protein kinases encoded by HSV-1 and HSV-2 are serine/threonine protein kinases with amino acid sequences that are conserved throughout the Herpesviridae family (7–9), and these conserved viral protein kinases, including HCMV UL97 and EBV BGLF4, have been designated conserved herpesvirus protein kinases (CHPKs). CHPKs share common cellular substrates, especially those involved in the DNA damage response (10–14). In addition, CHPKs are structurally similar to the cellular cyclin-dependent kinase cdk2 (15) and have a function that mimics the cyclin-dependent kinases (cdk's) (13, 16, 17).

The HSV-1 UL13 protein kinase activity has been shown to promote viral replication and cell-to-cell spread in cell cultures in a cell type-dependent manner (18–20). The mechanism(s) by which UL13 functions in viral replication and cell-to-cell spread remains unclear. However, UL13 has been shown to promote the expression of a subset of viral proteins, including ICP0, UL26, UL26.5, UL38, UL41, and Us11, in a cell type-dependent manner, suggesting that UL13 promoted viral replication and cell-to-cell spread by regulating the expression of these viral proteins. Recently, it was reported that UL13 kinase activity promoted the evasion of HSV-1-specific CD8⁺ T cell infiltration in the central nervous system (CNS) in mice following ocular infection and that this UL13-mediated immune evasion was critical for viral replication and pathogenicity in the mouse CNS (21). Although information on the activity of HSV-1 UL13 has been accumulating, little is known regarding the regulation of HSV-1 UL13 protein kinase in infected cells. HSV-2 UL13, the subject of this study, has a high degree of homology to HSV-1 UL13 at the amino acid level (86.3%): the HSV-2 UL13 gene encodes the same number of amino acids (518 amino acids) as the HSV-1 UL13 gene (8, 9). These features of HSV-2 UL13 suggest that it acts like HSV-1 UL13 in infected cells. However, unlike HSV-1 UL13, there has been no report on the role(s) of HSV-2 UL13 in infected cells and in vivo. Furthermore, like HSV-1 UL13, there is a lack of information on the regulation of HSV-2 UL13 in infected cells.

In general, protein kinases are tightly regulated by phosphorylation, binding of the regulatory subunit(s), and the presence of a small molecule(s) and/or by their subcellular localization. Data on their regulatory mechanism(s) are crucial for understanding the overall features of protein kinases as well as the functional consequence(s) of their manipulation, as has been done for the EBV BGLF4 CHPK (22). Thus, it has been reported that the small ubiquitin-related modifier (SUMO) binds to BGLF4 and that this binding is required for tethering the viral kinase in the nucleus, thereby regulating its nuclear functions, which are critical for efficient viral replication (22). In this study, we identified UL13 Ser-18 as a novel phosphorylation site in HSV-2-infected cells and investigated the effects of this phosphorylation on UL13 regulation and on HSV-2 replication and pathogenesis.

RESULTS

Modification of HSV-2-BAC (pYEbac356).

pYEbac356 contains a full-length HSV-2 strain 186 genome cloned into a bacterial artificial chromosome (BAC). Recombinant virus YK356 reconstituted from pYEbac356 has been shown to retain wild-type (WT) expression levels of the viral proteins, wild-type replication kinetics in cell culture, and a wild-type level of virulence in mice (23). When we started analyzing HSV-2 UL13, we found that YK356-infected cells produced UL13 that migrated differently than UL13 from wild-type HSV-2 186-infected cells in denaturing gels. UL13 in HSV-2 186-infected cells was detected as two bands by immunoblotting, in agreement with a previous study that detected hemagglutinin (HA)-tagged UL13 in HSV-2 strain 333-infected cells (24). However, YK356-infected cells produced forms of UL13 that migrated more slowly in denaturing gels than did those produced by HSV-2 186-infected cells (Fig. 1C). Sequence analysis showed that nucleotide A at position 1557 (relative to the first nucleotide of the UL13 gene) in the HSV-2 186 UL13 gene was replaced with G in pYEbac356 (Fig. 1A and B), changing the TGA stop codon in HSV-2 186 UL13 to a TGG Trp codon in YK356 UL13 (Fig. 1A and B). Therefore, YK356 encodes a larger UL13 polypeptide of 552 amino acids than the UL13 polypeptide of 518 amino acids encoded by wild-type HSV-2 186. We then introduced a G-to-A mutation at position 1557 in pYEbac356 to produce pYEbac861 encoding a UL13 polypeptide of 518 amino acids (Fig. 1A and B). We confirmed that the sequence of the UL13 gene in pYEbac861 was identical to that in HSV-2 186 (data not shown) and that YK861 (UL13-WT)-infected cells produced UL13 that migrated in denaturing gels as fast as UL13 produced by wild-type HSV-2 186-infected cells (Fig. 1C). In addition, the growth of recombinant virus YK861 (UL13-WT), which was reconstituted from pYEbac861, in Vero cells infected at multiplicities of infection (MOIs) of 3 and 0.01 was similar to that in cells infected with wild-type HSV-2 186 or YK356 (Fig. 1D and E).

FIG 1.

Construction of HSV-2-BAC (pYEbac861). (A) Schematic diagrams of the genome structures of wild-type HSV-2 186, YK356, and YK861 (UL13-WT). Line 1, wild-type HSV-2 186 genome; line 2, domain from the UL12 gene to the UL15 gene; lines 3 and 4, recombinant viruses YK356 and YK861 (UL13-WT). aa, amino acids. (B) DNA (top line) and amino acid (bottom line) sequences of the 3′- and carboxyl-terminal domains of UL13 from wild-type HSV-2 186, YK356, and YK861 (UL13-WT). Asterisks, nt, and a.a. denote the stop codon, nucleotide, and amino acid, respectively. (C to E) Expression of UL13 proteins and growth curves from cells infected with wild-type HSV-2 186, YK356, and YK861 (UL13-WT). (C) Vero cells were mock infected (lane 1) or infected with wild-type HSV-2 186 (lane 2), YK356 (lane 3), or YK861 (UL13-WT) (lane 4) at an MOI of 3 and harvested at 24 h postinfection, and lysates were analyzed by immunoblotting with the indicated antibodies. β-Actin and UL37 were used as a loading control and an infection indicator, respectively. Molecular mass markers are shown on the left. (D and E) Vero cells were infected at an MOI of 3 (D) or an MOI of 0.01 (E) with wild-type HSV-2 186, YK356, or YK861 (UL13-WT). Total virus from the cell culture supernatants and infected cells was harvested at the indicated times and assayed on Vero cells. Each value represents the mean ± standard error of the mean (SEM) of the results of three independent experiments.

Identification of phosphorylation sites in UL13 in HSV-2-infected cells.

It has been reported that HSV-2 UL13 was phosphorylated in infected cells (25). To identify the physiological phosphorylation site(s) in UL13 in HSV-2-infected cells, we carried out a large-scale phosphoproteomic analysis of titanium dioxide affinity chromatography-enriched phosphopeptides from HSV-2-infected human osteosarcoma U2OS cells by high-accuracy mass spectrometry (MS). Briefly, U2OS cells were infected at an MOI of 3 with wild-type HSV-2 186, harvested at 24 h postinfection, denatured, alkylated, and digested in solution with a combination of lysyl endopeptidase (Lys-C) and trypsin. The resulting peptide mixtures were subjected to phosphopeptide enrichment, using titanium dioxide affinity chromatography, and enriched samples were analyzed using a high-accuracy mass spectrometer. Peptide spectra were searched in sequence databases and validated as described in Materials and Methods. Using this approach, we identified serines 11, 18, 47, 50, 90, and 91 in UL13 as physiological phosphorylation sites in HSV-2-infected cells (Table 1). In this study, we focus on the phosphorylation of Ser-18 and Ser-91. We noted that these two serines were conserved in the UL13 proteins of various HSV-1 and HSV-2 strains (Fig. 2) but not in the UL13 homologs of other alphaherpesviruses, such as VZV, pseudorabies virus, and bovine herpesvirus 1 (data not shown), and have been identified as physiological phosphorylation sites in HSV-1-infected cells by previously reported large-scale phosphoproteomic analyses (26, 27).

TABLE 1.

UL13 phosphopeptides and phosphorylation sites identified by MS/MS

Phosphorylation site(s)	Peptide sequencea	Peptide count
Ser-11 and Ser-18	(q)RPA(p)SHVAADI(p)SPQGAHR	1
Ser-18	QRPASHVAADI(p)SPQGAHR	3
Ser-18	QRPASHVAADI(p)SPQGAHRR	1
Ser-18	(q)RPASHVAADI(p)SPQGAHRR	1
Ser-47 and Ser-50	ASGRP(p)SGP(p)SPR	1
Ser-50	ASGRPSGP(p)SPR	2
Ser-90 and Ser-91	RR(p)S(p)SPEAPGPAAK	2
Ser-91	RS(p)SPEAPGPAAK	1

^aIn the peptide sequences, (p)S indicates phosphorylated serine and (q) indicates an N-terminal glutamate-to-pyroglutamate conversion.

FIG 2.

Sequence alignment of the UL13 protein homologs from HSV-1(F), HSV-1(17), HSV-1(McKrae), HSV-1(KOS), HSV-2(186), HSV-2(HG52), and HSV-2(SD90e). The residues near UL13 Ser-18 (A) and UL13 Ser-91 (B) of HSV-2 186 and other HSV-1 and HSV-2 strains are shown.

Construction and characterization of recombinant viruses.

To investigate the significance of the phosphorylation of UL13 at Ser-18 and Ser-91 in HSV-2-infected cells, we constructed a series of recombinant viruses, including UL13-null mutant virus YK862 (ΔUL13), recombinant virus YK864 (UL13-K176M) encoding an enzymatically inactive mutant of UL13, recombinant virus YK866 (UL13-S18A) encoding a mutant UL13 in which Ser-18 was replaced with alanine, recombinant virus YK867 (UL13-S18D) carrying a phosphomimetic mutation at Ser-18, and recombinant virus YK869 (UL13-S91A) encoding a mutant UL13 in which Ser-91 was replaced with alanine (Fig. 3). In addition, we constructed viruses in which these mutations were repaired: YK863 (ΔUL13-repair), YK865 (UL13-K176M-repair), and YK868 (UL13-S18A/D-repair) (Fig. 3). We also constructed YK868 (UL13-S18A/D-repair) in which the mutations in both YK866 (UL13-S18A) and YK867 (UL13-S18D) were repaired (Fig. 3).

FIG 3.

Schematic diagrams of the genome structures of wild-type HSV-2 186 and the relevant domains of the recombinant viruses used in this study. Line 1, wild-type HSV-2 186 genome; line 2, domain from the UL12 gene to the UL15 gene; lines 3 to 11, recombinant viruses with mutations in the UL13 gene.

The recombinant viruses were characterized, with the following results. (i) Vero cells infected with wild-type HSV-2 186, YK864 (UL13-K176M), YK866 (UL13-S18A), YK867 (UL13-S18D), YK869 (UL13-S91A), YK863 (ΔUL13-repair), YK865 (UL13-K176M-repair), or YK868 (UL13-S18A/D-repair) expressed UL13 proteins, whereas cells infected with YK862 (ΔUL13) did not (Fig. 4A to C and 5), confirming that the null mutation in UL13 inactivated the expression of the UL13 gene. (ii) As described above (Fig. 1C), UL13 proteins from Vero cells infected with wild-type HSV-2 186 or YK865 (UL13-K176M-repair) were detected as two bands with mobility differences in denaturing gels (Fig. 4A to C and 5). After phosphatase treatment of the lysates from wild-type HSV-2 186-infected cells, the slower-migrating UL13 band disappeared (Fig. 4D), indicating that the slower-migrating band was a phosphorylated form of UL13. In cells infected with YK864 (UL13-K176M), UL13 was detected as a single band in denaturing gels, with a mobility similar to that of the faster-migrating UL13 band in wild-type HSV-2 186-infected cells and YK865 (UL13-K176M-repair)-infected cells (Fig. 4B). These results indicated that UL13 kinase activity was required for phosphorylation of the UL13 protein itself in infected cells. (iii) In Vero cells infected with YK866 (UL13-S18A) or YK867 (UL13-S18D), UL13 proteins were detected as two bands with mobility differences in denaturing gels, as observed with cells infected with wild-type HSV-2 186 or YK868 (UL13-S18A/D-repair). However, the slower-migrating form of UL13 in YK866 (UL13-S18A)-infected cells and in YK867 (UL13-S18D)-infected cells appeared to migrate slightly faster than the slower-migrating band of UL13 in cells infected with wild-type HSV-2 186 or YK868 (UL13-S18A/D-repair) (Fig. 5). In contrast, the electrophoretic pattern of the two UL13 bands in YK869 (UL13-S91A)-infected cells could not be differentiated from the two UL13 bands in cells infected with wild-type HSV-2 186 (Fig. 4C).

FIG 4.

Effect of UL13 mutations on expression of UL13 proteins. (A) Vero cells were mock infected (lane 1) or infected with wild-type HSV-2 186 (lane 2), YK862 (ΔUL13) (lane 3), or YK863 (ΔUL13-repair) (lane 4) at an MOI of 3 and harvested at 24 h postinfection, and lysates were analyzed by immunoblotting with the indicated antibodies. β-Actin and UL37 were used as a loading control and an infection indicator, respectively. (B) Vero cells were mock infected (lane 1) or infected with wild-type HSV-2 186 (lane 2), YK864 (UL13-K176M) (lane 3), or YK865 (UL13-K176M-repair) (lane 4) and harvested, and lysates were analyzed as described above for panel A. (C) Vero cells were mock infected (lane 1) or infected with wild-type HSV-2 186 (lane 2) or YK869 (UL13-S91A) (lane 3) and harvested, and lysates were analyzed as described above for panel A. (D) Vero cells infected with wild-type HSV-2 186 at an MOI of 3 were harvested at 24 h postinfection and lysed, and cell lysates were mock treated (lane 1) or treated with CIP (lane 2) and then analyzed as described above for panel A. Molecular mass markers are shown on the left.

FIG 5.

Effect of the UL13 S18A and S18D mutations on expression of UL13 proteins. Vero cells were mock infected (lane 1) or infected with wild-type HSV-2 186 (lane 2), YK866 (UL13-S18A) (lane 3), YK867 (UL13-S18D) (lane 4), or YK868 (UL13-S18A/D-repair) (lane 5) at an MOI of 3 and harvested at 24 h postinfection, and lysates were analyzed by immunoblotting with the indicated antibodies. β-Actin and UL37 were used as a loading control and an infection indicator, respectively. Molecular mass markers are shown on the left.

Production and characterization of the monoclonal antibody to EF-1δ-S133^P.

Elongation factor 1δ (EF-1δ) is one of the cellular substrates of HSV-1 UL13 (10). It has been reported that infection of cells with HSV-1 or HSV-2 caused extensive hyperphosphorylation of EF-1δ (11), and HSV-1 UL13 was shown to be responsible for EF-1δ hyperphosphorylation. Furthermore, EF-1δ Ser-133 was identified as a phosphorylation site: phosphorylation of this site by HSV-1 UL13 or by cdk1 was associated with EF-1δ hyperphosphorylation (13). Indeed, phosphorylation of EF-1δ at Ser-133 in HSV-2-infected cells was detected in the phosphoproteomic analysis in this study (data not shown).

There are two predominant forms of EF-1δ, a hypophosphorylated form and a hyperphosphorylated form, that are detected as a faster-migrating band and a slower-migrating band, respectively, in denaturing gels (10). In mock-infected cells, the hypophosphorylation form of EF-1δ was dominant (10). In agreement with previous observations of HSV-1 UL13, infection of U2OS cells with wild-type HSV-2 186, YK863 (ΔUL13-repair), or YK865 (UL13-K176M-repair) significantly increased the hyperphosphorylation form of EF-1δ, but infection of cells with YK862 (ΔUL13) or YK864 (UL13-K176M) did not (Fig. 6B and C). These results strongly suggested that EF-1δ was a physiological substrate of UL13 in HSV-2-infected cells as it is in HSV-1-infected cells.

FIG 6.

Characterization of the anti-EF-1δ-S133^P monoclonal antibody. (A) HEK293T cells were transfected with a plasmid expressing EGFP-EF-1δ(F) (lane 1 and 2) or a plasmid expressing EGFP-EF-1δS133A(F) (lane 3) and harvested at 48 h posttransfection. Cell lysates were mock treated (lanes 1 and 3) or treated with CIP (lane 2) and then analyzed by immunoblotting with anti-Flag monoclonal antibody (top), anti-EF-1δ-S133^P monoclonal antibody (middle), or anti-β-actin monoclonal antibody (bottom). β-Actin was used as a loading control. (B) U2OS cells were mock infected (lane 1) or infected with wild-type HSV-2 186 (lane 2), YK862 (ΔUL13) (lane 3), or YK863 (ΔUL13-repair) (lane 4) at an MOI of 3 and harvested at 24 h postinfection, and lysates were analyzed by immunoblotting with the indicated antibodies. α-Tubulin and UL37 were used as a loading control and an infection indicator, respectively. (C) U2OS cells were mock infected (lanes 1) or infected with wild-type HSV-2 186 (lane 2), YK864 (UL13-K176M) (lane 3), or YK865 (UL13-K176M-repair) (lane 4) and harvested, and lysates were analyzed as described above for panel B. Molecular mass markers are shown on the left.

To monitor UL13-mediated phosphorylation of EF-1δ in HSV-infected cells more easily, we developed a monoclonal antibody (anti-EF-1δ-S133^P antibody) that reacted with EF-1δ-S133^P. As shown in Fig. 6A, the anti-EF-1δ-S133^P antibody reacted with wild-type Flag-tagged EF-1δ fused with enhanced green fluorescent protein (EGFP) [EGFP-EF-1δ(F)] in lysates of HEK293T cells transfected with the expression plasmid for EGFP-EF-1δ(F) by immunoblotting but did not react with a mutant of EGFP-EF-1δ(F) carrying an S133A mutation. Phosphatase treatment of lysates of HEK293T cells transfected with the EGFP-EF-1δ(F) expression plasmid abolished the anti-EF-1δ-S133^P antibody reactivity. Furthermore, in agreement with the results with the anti-EF-1δ polyclonal antibody described above, infection of cells with wild-type HSV-2 186, YK863 (ΔUL13-repair), or YK865 (UL13-K176M-repair) increased EF-1δ-S133^P reactivity, as detected by the anti-EF-1δ-S133^P monoclonal antibody, but infection with YK862 (ΔUL13) or YK864 (UL13-K176M) did not (Fig. 6B and C). Taken together, these results confirmed that the anti-EF-1δ-S133^P monoclonal antibody was able to specifically detect EF-1δ-S133^P in HSV-infected cells.

Effect of phosphorylation of HSV-2 UL13 Ser-18 on viral cell-to-cell spread and replication.

To examine the effect of phosphorylation of UL13 Ser-18 and Ser-91 on HSV-2 replication in cell cultures, we analyzed progeny virus yields in U2OS and Vero cells infected with wild-type HSV-2 186, YK862 (ΔUL13), YK863 (ΔUL13-repair), YK864 (UL13-K176M), YK865 (UL13-K176M-repair), YK866 (UL13-S18A), YK867 (UL13-S18D), YK868 (UL13-S18A/D-repair), or YK869 (UL13-S91A) at an MOI of 0.01 for 24 h or at an MOI of 3 for 12 h. As shown in Fig. 7A, progeny virus yields in U2OS cells infected with YK862 (ΔUL13) or YK864 (UL13-K176M) at an MOI of 0.01 were similar but lower (i.e., 18.9- or 14.4-fold, respectively) than those in cells infected with wild-type HSV-2 186. Similarly, progeny virus yields in U2OS cells infected with YK866 (UL13-S18A) at an MOI of 0.01 were 21.6-fold lower than those in cells infected with wild-type HSV-2 186 (Fig. 7A). In contrast, the progeny virus yield in U2OS cells infected with YK869 (UL13-S91A) at an MOI of 0.01 was similar to that in cells infected with wild-type HSV-2 186. Thus, the UL13 S18A mutation reduced HSV-2 replication to a level similar to that observed with the UL13-null and kinase-dead mutations. The wild-type level of progeny virus yields in U2OS cells at an MOI of 0.01 was restored in cells infected with YK863 (ΔUL13-repair), YK865 (UL13-K176M-repair), YK868 (UL13-S18A/D-repair), or YK867 (UL13-S18D) carrying a phosphomimetic mutation at UL13 Ser-18 (Fig. 7A). In contrast, progeny virus yields in U2OS cells infected with each of the viruses described above at an MOI of 3 were similar (Fig. 7B). Furthermore, progeny virus yields in Vero cells infected with each of the viruses at MOIs of 0.01 and 3 were similar (Fig. 8A and B).

FIG 7.

Effect of UL13 mutations on progeny virus yields and virus plaque formation in U2OS cells. (A and B) U2OS cells were infected with either wild-type HSV-2 186, YK862 (ΔUL13), YK863 (ΔUL13-repair), YK864 (UL13-K176M), YK865 (UL13-K176M-repair), YK866 (UL13-S18A), YK867 (UL13-S18D), YK868 (UL13-S18A/D-repair), or YK869 (UL13-S91A) at an MOI of 0.01 (A) or an MOI of 3 (B). Total virus from the cell culture supernatants and infected cells was harvested at 24 h (A) or at 12 h (B) postinfection and assayed on Vero cells. Each value is the mean ± SEM for four experiments. Statistical analysis was performed by ANOVA with the Tukey test. Asterisks indicate statistically significant values (*, P < 0.05; **, P < 0.01). n.s., not significant. (C) U2OS cells were infected with either wild-type HSV-2 186, YK862 (ΔUL13), YK863 (ΔUL13-repair), YK864 (UL13-K176M), YK865 (UL13-K176M-repair), YK866 (UL13-S18A), YK867 (UL13-S18D), YK868 (UL13-S18A/D-repair), or YK869 (UL13-S91A) at an MOI of 0.0001 under plaque assay conditions. The diameters of 20 single plaques for each of the indicated viruses were measured at 48 h postinfection. Each data point is the mean ± SEM of the measured plaque sizes. Statistical analysis was performed by ANOVA with the Tukey test. Asterisks indicate statistically significant values (*, P < 0.0001). Data are representative of results from three independent experiments.

FIG 8.

Effect of each UL13 mutation on progeny virus yields and virus plaque formation in Vero cells. (A and B) Vero cells were infected with either wild-type HSV-2 186, YK862 (ΔUL13), YK863 (ΔUL13-repair), YK864 (UL13-K176M), YK865 (UL13-K176M-repair), YK866 (UL13-S18A), YK867 (UL13-S18D), YK868 (UL13-S18A/D-repair), or YK869 (UL13-S91A) at an MOI of 0.01 (A) or an MOI of 3 (B). Total virus from the cell culture supernatants and infected cells was harvested at 24 h (A) or at 12 h (B) postinfection and assayed on Vero cells. Each value is the mean ± SEM of the results of three independent experiments. Statistical analysis was performed by ANOVA with the Tukey test. n.s., not significant. (C) Vero cells were infected with either wild-type HSV-2 186, YK862 (ΔUL13), YK863 (ΔUL13-repair), YK864 (UL13-K176M), YK865 (UL13-K176M-repair), YK866 (UL13-S18A), YK867 (UL13-S18D), YK868 (UL13-S18A/D-repair), or YK869 (UL13-S91A) at an MOI of 0.0001 under plaque assay conditions. The diameters of 20 single plaques for each of the indicated viruses were measured at 48 h postinfection. Each data point is the mean ± SEM of the measured plaque sizes. Statistical analysis was performed by ANOVA with the Tukey test. Data are representative of results from three independent experiments.

We also analyzed plaque size on U2OS and Vero cells infected with wild-type HSV-2 186, YK862 (ΔUL13), YK863 (ΔUL13-repair), YK864 (UL13-K176M), YK865 (UL13-K176M-repair), YK866 (UL13-S18A), YK867 (UL13-S18D), YK868 (UL13-S18A/D-repair), or YK869 (UL13-S91A) under plaque assay conditions. In agreement with the growth properties of these viruses described above (Fig. 7A and B and 8A and B), YK866 (UL13-S18A) produced plaques that were smaller than those of wild-type HSV-2 186 and similar in size to those of YK862 (ΔUL13) and YK864 (UL13-K176M) in U2OS cells (Fig. 7C). The wild-type plaque size was restored in U2OS cells infected with YK863 (ΔUL13-repair), YK865 (UL13-K176M-repair), YK868 (UL13-S18A/D-repair), and YK867 (UL13S18D) (Fig. 7C). YK869 (UL13-S91A) produced plaques with sizes similar to those of wild-type HSV-2 186 in U2OS cells (Fig. 7C), and all these viruses produced plaques with sizes similar to those on Vero cells (Fig. 8C).

Taken together, these results suggested that UL13 itself, its protein kinase activity, and phosphorylation of UL13 Ser-18 were required for efficient HSV-2 replication and cell-to-cell spread in a manner dependent on the cell type and MOI. Since the loss of phosphorylation of UL13 Ser-91, due to the S91A mutation in UL13, had no effect on HSV-2 replication and cell-to-cell spread, we focused on the biological significance of UL13 Ser-18 phosphorylation.

Effect of UL13 Ser-18 phosphorylation on UL13-mediated phosphorylation of EF-1δ in HSV-2-infected cells.

To examine the effect of UL13 Ser-18 phosphorylation on UL13 kinase function in HSV-2-infected U2OS cells, we assayed the level of phosphorylation of EF-1δ Ser-133, which, as described above, was mediated by UL13 in HSV-2-infected cells (Fig. 6), using the phosphospecific monoclonal antibody (anti-EF-1δ-S133^P antibody) developed for this study. As shown in Fig. 9A and B, phosphorylation of EF-1δ Ser-133 was decreased significantly in YK866 (UL13-S18A)-infected U2OS cells compared to phosphorylation in wild-type HSV-2 186-infected cells. The wild-type phenotype was restored in cells infected with YK868 (UL13-S18A/D-repair) or YK867 (UL13S18D) (Fig. 9). In agreement with these results using the anti-EF-1δ-S133^P antibody, the amount of the hyperphosphorylation form of EF-1δ relative to that of total EF-1δ decreased significantly in U2OS cells infected with YK866 (UL13-S18A), compared to the level in cells infected with wild-type HSV-2 186, YK868 (UL13-S18A/D-repair), or YK867 (UL13S18D) (Fig. 9A and C). These results suggested that phosphorylation of UL13 Ser-18 was required for optimal UL13 activity in HSV-2-infected cells.

FIG 9.

Effects of the UL13 S18A and S18D mutations on phosphorylation of EF-1δ Ser-133 in U2OS cells. (A) U2OS cells were mock infected (lane 1) or infected with wild-type HSV-2 186 (lane 2), YK866 (UL13-S18A) (lane 3), YK867 (UL13-S18D) (lane 4), or YK868 (UL13-S18A/D-repair) (lane 5) at an MOI of 3 and harvested at 24 h postinfection, and lysates were analyzed by immunoblotting with the indicated antibodies. α-Tubulin and UL37 were used as a loading control and an infection indicator, respectively. Molecular mass markers are shown on the left. (B) Amount of EF-1δ-S133^P protein detected with anti-EF-1δ-S133^P monoclonal antibody (A, top panel) relative to that of α-tubulin protein detected with anti-α-tubulin antibody (A, bottom panel) in HSV-2-infected cells. The data were normalized to the value for mock-infected cells in panel A, lane 1. Each value is the mean ± SEM for seven experiments. (C) Amount of the hyperphosphorylated form of EF-1δ protein detected with anti-EF-1δ polyclonal antibody (A, top band in the second panel) relative to that of total EF-1δ protein (hyperphosphorylated and hypophosphorylated forms of EF-1δ) detected with anti-EF-1δ polyclonal antibody (A, both bands in the second panel) in HSV-2-infected cells. Each value is the mean ± SEM for seven experiments and is expressed relative to the mean value of wild-type HSV-2 186-infected cells, which was normalized to 1. Statistical analysis was performed by ANOVA with the Tukey test. Asterisks indicate statistically significant values (*, P < 0.05; **, P < 0.01). n.s., not significant.

Effects of UL13 Ser-18 phosphorylation on HSV-2 replication and pathogenesis in mice.

To examine the effect of UL13 Ser-18 phosphorylation on HSV-2 replication in vivo and on viral pathogenesis, 6-week-old female ICR mice pretreated with medroxyprogesterone were infected vaginally with YK862 (ΔUL13), YK863 (ΔUL13-repair), YK864 (UL13-K176M), YK865 (UL13-K176M-repair), YK866 (UL13-S18A), YK867 (UL13-S18D), or YK868 (UL13-S18A/D-repair). The survival of the infected mice was monitored for 18 days, virus titers in vaginal secretions were determined at 2 and 4 days postinfection, and disease progression was observed at 9 and 12 days postinfection. As shown in Fig. 10, YK862 (ΔUL13), YK864 (UL13-K176M), and YK866 (UL13-S18A) had similar phenotypes. The null, kinase-dead, and S18A mutations in UL13 all significantly reduced mortality in infected mice, development of vaginal disease in mice at 9 and 12 days postinfection at similar levels, and viral replication in the mouse vagina at 2 days postinfection. One aspect specific to the UL13 S18A mutation was that this mutation had no effect on HSV-2 replication in the vagina at 4 days postinfection, although the null and kinase-dead mutations in UL13 significantly reduced HSV-2 replication in the vagina at 4 days postinfection (Fig. 10I). For the phosphomimetic S18D mutation in UL13, the HSV-2 UL13-S18D mortality rate was intermediate between those of UL13-S18A and UL13-S18A/D-repair, and the development of vaginal disease at 9 and 12 days postinfection was intermediate between those of UL13-S18A and UL13-S18A/D-repair (Fig. 10G and H). The differences in mortality and development of vaginal disease between UL13-S18D and UL13-S18A were statistically significant. However, although HSV-2 UL13-S18D replication was intermediate between those of UL13-S18A and UL13-S18A/D-repair at 2 days postinfection and the difference in HSV-2 replication between UL13-S18D and UL13-S18A was statistically significant, at 4 days postinfection, UL13-S18D replication was not statistically different than those of UL13-S18A and UL13-S18A/D-repair (Fig. 10I). These results suggested that UL13 itself, its protein kinase activity, and phosphorylation of UL13 Ser-18 were required for efficient HSV-2 replication and pathogenesis in mice following vaginal infection.

FIG 10.

Effects of UL13 mutations on mortality, viral pathogenesis in the vagina, and viral replication in the vagina of mice following intravaginal infection. Fifteen (A to F) or 20 (G to I) 6-week-old female ICR mice were pretreated with medroxyprogesterone, and the vagina of each mouse was infected with 1 × 10⁴ PFU YK862 (ΔUL13) or YK863 (ΔUL13-repair) (A to C); YK864 (UL13-K176M) or YK865 (UL13-K176M-repair) (D to F); or YK866 (UL13-S18A), YK867 (UL13-S18D), or YK868 (UL13-S18A/D-repair) (G to I). (A, D, and G) Survival of mice was monitored for 18 days postinfection. Statistical significance according to a log rank test (A and D) or a log rank test with Bonferroni adjustment for the three-comparison analyses (G) is shown. (B, E, and H) The clinical scores of infected mice at 9 and 12 days postinfection were monitored. Each data point is the clinical score for one mouse. The horizontal bars indicate the means for each group. The statistical significance values were analyzed by the Mann-Whitney U test (B and E) or Dunn's multiple-comparison test (H). (C, F, and I) The vaginal secretions of infected mice at 2 and 4 days postinfection were harvested, and virus titers were assayed. Each data point is the virus titer in the vaginal secretions of one mouse. The horizontal bars indicate the means for each group. The statistical significance values were analyzed by the Mann-Whitney U test (C and F) or Dunn's multiple-comparison test (I). Asterisks indicate statistically significant values (*, P < 0.05; **, P < 0.01; ***, P < 0.0001). n.s., not significant.

DISCUSSION

Numerous reports have shown that protein kinases encoded by herpesviruses play multiple roles in the viral life cycle by regulating various aspects of viral replication and pathogenesis and the host cell response (3). However, there have been only a few reports indicating how these viral protein kinases are regulated in infected cells (22), although studies of cellular protein kinases show that understanding the regulation of a protein kinase is as important as understanding its downstream effects (28, 29). Since the HSV-1 and HSV-2 UL13 genes were predicted and shown to encode protein kinases in 1989 and 1997, respectively (8, 9), data on the downstream effects of these HSV protein kinases, mostly of HSV-1 UL13, have been accumulating (13, 18, 19). However, there have been no reports on the mechanism(s) by which these HSV protein kinases are regulated in infected cells. The key finding reported here is that phosphorylation of UL13 Ser-18 regulated UL13 activity in HSV-2-infected cells. In particular, a mutation that blocked phosphorylation of UL13 Ser-18 (i.e., the UL13 S18A mutation) significantly reduced HSV-2 replication and cell-to-cell spread in cell cultures and reduced HSV-2 replication and pathogenesis in mice to levels similar to those with the UL13-null and kinase-dead mutations. These results suggested that UL13 regulation was critical for the functional effects of the UL13 viral protein kinase in vitro and in vivo and also for HSV-2 replication and pathogenesis in vivo. Interestingly, UL13 Ser-18 has been reported to be phosphorylated in HSV-1-infected cells (26, 27), suggesting that this regulatory mechanism and its significance in HSV-2 infection, clarified in this study, may be conserved in HSV-1 infection.

In this study, we showed that the slowly migrating band of HSV-2 UL13 detected in denaturing gels was abolished by either phosphatase treatment or the UL13 kinase-dead mutation, indicating that UL13 autophosphorylated itself in HSV-2-infected cells, as reported previously (25), and that the slowly migrating band was an autophosphorylated form of UL13. Interestingly, a mutation that blocked the phosphorylation of UL13 Ser-18 (i.e., the UL13 S18A mutation) caused the autophosphorylated form of mutant UL13 to migrate faster in denaturing gels than wild-type UL13, and it has been reported that the amino acid sequences around UL13 Ser-18 (SPQG) can be phosphorylation targets of cdk's (30–33). As described above, UL13 was shown to share phosphorylation target sites with cdk's (13). Taken together, these observations suggested that Ser-18 was an autophosphorylation site in UL13. In general, protein kinases often autophosphorylate themselves to regulate their catalytic activities (34, 35). It seemed likely that phosphorylation of UL13 Ser-18 regulated the catalytic activity of HSV-2 UL13, based on the observations that blocking HSV-2 Ser-18 phosphorylation (i.e., by the UL13 S18A mutation) reduced phosphorylation of the UL13 substrate EF-1δ, and that the phosphomimetic mutation in the UL13 phosphorylation site (i.e., UL13 S18D) restored the wild-type level of EF-1δ phosphorylation. However, we cannot completely eliminate the possibility that phosphorylation of UL13 Ser-18 did not regulate the catalytic activity of HSV-2 UL13, since phosphorylation of UL13 Ser-18 might regulate substrate specificity and/or subcellular localization of UL13, which could affect the phosphorylation of its substrates. Various herpesvirus protein kinases (e.g., HSV-1 UL13, HSV-1 Us3, EBV BGLF4, and HCMV UL97) have been shown to autophosphorylate themselves (13, 36–38), and it has been reported that autophosphorylation of HSV-1 Us3 regulated its catalytic activity and that this regulation was critical for viral replication in vivo and pathogenesis (36, 39), as observed for phosphorylation of HSV-2 UL13 Ser-18 in this study. In particular, the autophosphorylation sites in the CHPKs, including HCMV UL97 and EBV BGLF4, that have been reported are in their amino-terminal domains (37, 40). Therefore, autophosphorylation in the amino-terminal domains of CHPKs may be involved in their regulation.

Phosphorylation of HSV-2 UL13 Ser-18 did not appear to be critical for the regulatory activity of HSV-2 UL13 but appeared to optimize UL13 regulatory activity in infected cells, since blocking the phosphorylation of HSV-2 UL13 Ser-18 by the UL13 S18A mutation reduced phosphorylation of EF-1δ, a cell substrate of UL13, to only 74 to 81% of that in wild-type HSV-2-infected cells. In contrast, as described above, the UL13 S18A mutation that blocked the phosphorylation of UL13 Ser-18 reduced HSV-2 replication and pathogenesis to levels similar to those of the UL13-null and kinase-dead mutations. These results suggested that HSV-2 UL13 activity was precisely regulated by phosphorylation of Ser-18 and that this regulation was critical for the UL13 functions in HSV-2 replication and pathogenesis. In agreement with these observations, it has been reported that HSV-1 Us3 protein kinase phosphorylated and tightly regulated the activity of a dUTPase encoded by HSV-1 and that this strict regulation was critical for the function of HSV-1 dUTPase in viral replication and pathogenesis (41). In this case, the phosphorylation-dead mutation in HSV-1 dUTPase reduced its activity to only 68% of that of the wild-type dUTPase, but this mutation reduced viral replication and pathogenesis in mice to levels similar to those of the HSV-1 dUTPase-null and enzyme-dead mutations (42).

As described above, studies of HSV UL13 have concentrated on HSV-1 UL13. There have been no reports on the construction and characterization of HSV-2 UL13-null and kinase-dead mutant viruses, meaning that the role(s) of HSV-2 UL13 in viral replication and pathogenesis in vitro and in vivo has not been elucidated. However, this study showed that HSV-2 UL13 itself and its kinase activity were critical for viral replication and cell-to-cell spread in cell cultures in a manner dependent on the cell type and MOI and for viral replication and pathogenesis in mice following vaginal infection. These features of HSV-2 UL13 were similar to those of HSV-1 UL13, as described above (19–21), and therefore, these data suggested that the HSV-1 and HSV-2 UL13 protein kinases may have functions in viral replication and pathogenesis that are conserved in these viruses. At present, the mechanism(s) by which the requirements of HSV UL13 kinases and phosphorylation of HSV-2 at Ser-18 in viral replications and cell-to-cell spread are dependent on the cell type is unknown. A cellular activity(ies) that can compensate for the activity of the UL13 kinases may be available in Vero cells but not in U2OS cells. As described above, CHPKs mimic the cdk's (13, 16, 17), and therefore, it is of interest to investigate the activities of these cellular kinases in these HSV-infected cells.

In this study, we developed a monoclonal antibody that specifically recognized EF-1δ-S133^P. This monoclonal antibody enabled us to detect the small reduction in UL13 phosphorylation of EF-1δ in HSV-2-infected cells, due to the block in the phosphorylation of HSV-2 UL13 protein kinase by the UL13 S18A mutation. Thus, using this phosphorylation-specific monoclonal antibody, the change in the level of HSV-2 UL13 phosphorylation activity could be easily and accurately monitored. It has been reported that various alpha-, beta-, and gammaherpesviruses induced the hyperphosphorylation of EF-1δ and that some CHPKs, including HSV-1 UL13, HCMV UL97, and EBV BGLF4, mediated the hyperphosphorylation of EF-1δ (10–12). These observations indicated that the anti-EF-1δ-S133^P antibody is generally useful for monitoring the activity of various CHPKs in infected cells.

MATERIALS AND METHODS

Cells and viruses.

Vero, U2OS, HEK293T, SP2/O, and rabbit skin cells were described previously (39, 43–45), as was HSV-2 wild-type strain 186 (46). Recombinant virus YK356, which was reconstituted from pYEbac356 containing a full-length HSV-2 186 genome with a BAC sequence inserted into the intergenic region between UL50 and UL51, was described previously (23).

Plasmids.

To generate pMEF-UL13-2, the entire HSV-2 UL13 coding sequence was amplified by PCR from the HSV-2 186 genome, isolated as described previously (43), and cloned into the EcoRI and XbaI sites of pMEF (47). To generate pMEF-UL13-2-KanS, for generating recombinant virus YK863 (ΔUL13-repair) in which the mutation in YK862 (ΔUL13) was repaired, we amplified an I-SceI site, a kanamycin resistance gene, and 52 bp of the HSV-2 UL13 sequence by PCR from pEP-KanS (48) with primers 5′-CGAGCGGCCGCACCCACGACATCCGCGGCTTTATCACCCCGCTCGGGTTCTCGCTTAGGGATAACAGGGTAATCGATTT-3′ and 5′-GCAGCGGCCGCCATATGCTAGCCAGTGTTACAACCAATTAACC-3′ and cloned the amplicons into the NotI site of pMEF-UL13-2. To generate pEGFP-EF-1δ(F), the entire EF-1δ coding sequence was amplified by PCR from pME-EF-1δ(F) (13), in which EF-1δ was tagged with the Flag epitope, and cloned into the EcoRI and BamHI sites of pEGFP-C1 (Clontech). To generate pEGFP-EF-1δS133A(F), in which EF-1δ Ser-133 was replaced with alanine, the entire EF-1δ coding sequence was amplified by PCR from pME-EF-1δS133A(F) (13) and cloned into the EcoRI and BamHI sites of pEGFP-C1 (Clontech).

Mutagenesis of viral genomes and generation of recombinant HSV-2.

Recombinant virus YK861 (UL13-WT), carrying a stop codon at UL13 residue 519 in YK356 (Fig. 1A and B), was constructed by the two-step Red-mediated mutagenesis procedure using Escherichia coli GS1783 containing pYEbac356 (23), as described previously (36), except using the primers listed in Table 2. Recombinant virus YK862 (ΔUL13), in which the UL13 gene was disrupted by deleting UL13 codons 159 to 417 (Fig. 3); YK864 (UL13-K176M), encoding an enzymatically inactive UL13 mutant in which the lysine at UL13 residue 176 was replaced with methionine (K176M) (Fig. 3); YK866 (UL13-S18A), encoding UL13 with alanine substituted for serine at residue 18 (S18A) (Fig. 3); and YK869 (UL13-S91A), encoding UL13 with alanine substituted for serine at residue 91 (S91A) (Fig. 3), were generated by the two-step Red-mediated mutagenesis procedure using E. coli strain GS1783 carrying the YK861 genome (pYEbac861), as described as above, except using the primers listed in Table 2. Recombinant virus YK867 (UL13-S18D), encoding UL13 with aspartic acid substituted for alanine at residue 18 (S18D), was generated by the two-step Red-mediated mutagenesis procedure using E. coli strain GS1783 carrying the YK866 (UL13-S18A) genome, as described previously (36), except using the primers listed in Table 2. Recombinant viruses YK865 (UL13K176M-repair) and YK868 (UL13-S18A/D-repair), in which the mutations in YK864 (UL13K176M) and YK867 (UL13-S18D) were repaired, respectively (Fig. 3), were generated as described previously (36), except using the primers listed in Table 2. YK868 (UL13-S18A/D-repair) is the repaired virus for both YK866 (UL13-S18A) and YK867 (UL13-S18D). Recombinant virus YK863 (ΔUL13-repair), in which the mutation in YK862 (ΔUL13) was repaired (Fig. 3), was generated as described previously (36), except using the primers listed in Table 2, pMEF-UL13-2-KanS, and E. coli GS1783 containing the YK862 (ΔUL13) genome (Fig. 3).

TABLE 2.

Primer sequences used for the construction of recombinant viruses in this study

Sample preparation for MS.

U2OS cells (∼1.25 × 10⁷) were infected at an MOI of 3 with wild-type HSV-2 186, harvested at 24 h postinfection, and suspended in 8 M urea containing PhosSTOP phosphatase inhibitor cocktail (Roche Diagnostics) and Benzonase nuclease (Novagen). The mixture was kept on ice for 1 h, and cellular debris was then pelleted by centrifugation at 15,000 rpm for 30 min. The cell lysate was reduced with 1 mM dithiothreitol (DTT) for 90 min and then alkylated with 5.5 mM iodoacetamide (IAA) for 30 min. After digestion with Lys-C (1:50 [wt/wt]) (Wako) at 37°C for 3 h, the resulting peptide mixture was diluted with 10 mM Tris-HCl (pH 8.2) to a final urea concentration of <2 M and then digested with modified trypsin (1:50 [wt/wt]) (sequencing grade; Promega) at 37°C for 3 h. An equal amount of trypsin was then added for overnight digestion. Phosphopeptides were enriched using a Titansphere Phos-TiO kit (GL Sciences) as described previously (41).

Mass spectrometric analysis, protein identification, and determination of phosphorylated sites.

Shotgun proteomic analyses of the Titansphere eluates were performed by using a linear ion trap-Orbitrap mass spectrometer (LTQ-Orbitrap Velos; Thermo Fisher Scientific) coupled with a nanoflow liquid chromatography (LC) system (Dina-2A; KYA Technologies). Peptides were injected into a 75-μm reversed-phase C₁₈ column at a flow rate of 10 μl/min and eluted with a linear gradient of solvent A (2% acetonitrile and 0.1% formic acid in H₂O) to solvent B (40% acetonitrile and 0.1% formic acid in H₂O) at 300 nl/min. Peptides were sequentially sprayed from a nanoelectrospray ion source (KYA Technologies) and analyzed by collision-induced dissociation (CID). The analyses were carried out in the data-dependent mode, switching automatically between MS and tandem MS (MS/MS) acquisition. For CID analyses, full-scan MS spectra (from m/z 380 to 2,000) were acquired in the Orbitrap with a resolution of 100,000 at m/z 400 after ion count accumulation to the target value of 1,000,000. The 20 most intense ions at a threshold above 2,000 were fragmented in the linear ion trap with a normalized collision energy of 35% for an activation time of 10 ms. The Orbitrap analyzer was operated with the “lock mass” option to perform shotgun detection with high accuracy (49). Protein identification was conducted by analyzing the MS and MS/MS data with Mascot (Matrix Science). Carbamidomethylation of cysteine residues was set as a fixed modification, whereas methionine oxidation, protein N-terminal acetylation, pyroglutamination for N-terminal glutamine, and phosphorylation (Ser, Thr, and Tyr) were set as variable modifications. A maximum of two missed cleavages was allowed in the database search. The tolerances for mass deviation were set at 3 ppm for peptide masses and 0.8 Da for MS/MS peaks. For peptide identification, we conducted decoy database searching with Mascot and applied a filter for a false-positive rate of <1%. Determination of phosphorylated sites in the peptides was performed using the Proteome Discoverer, version 1.3, software program (Thermo Fisher Scientific). We performed this large-scale phosphoproteomic analysis once.

Immunoblotting.

Immunoblotting was performed as described previously (50). The amount of protein in immunoblot bands was quantified using the ImageQuant LAS 4000 system with ImageQuant TL7.0 analysis software (GE Healthcare Life Sciences).

Antibodies.

Rabbit polyclonal antibody against EF-1δ was described previously (51). Mouse monoclonal antibody against UL13 was described previously (52). Mouse monoclonal antibodies against the Flag epitope (M2), α-tubulin (DM1A), and β-actin (AC15) were purchased from Sigma. Rabbit polyclonal antibody against UL37 (CAC-CT-HSV-UL37) was purchased from CosmoBio. Mouse monoclonal antibody that recognizes EF-1δ with phosphorylated Ser-133 (EF-1δ-S133^P) was generated as described previously (39), except that peptides corresponding to EF-1δ residues 128 to 140 (QTQHVSPMRQVEP) with and without phosphorylated Ser-133 were used (purchased from GL Biochem).

Phosphatase treatment.

Lysates of Vero cells that had been infected with wild-type HSV-2 186 at an MOI of 3 for 24 h and lysates of HEK293T cells that had been transfected with pEGFP-EF-1δ(F) were treated with calf intestinal alkaline phosphatase (CIP) (New England BioLabs) as described previously (53).

Determination of plaque size.

Vero and U2OS cells were infected with each of the recombinant viruses at an MOI of 0.0001, and plaque sizes were determined as described previously (20).

Animal studies.

Female ICR mice were purchased from Charles River. For intravaginal infections, 5-week-old mice were each injected subcutaneously in the neck ruff with 1.67 mg medroxyprogesterone (Depo-Gestin; ANB Laboratories) in 200 μl phosphate-buffered saline (PBS) 7 days prior to viral infection. The treated mice were then infected intravaginally with 1 × 10⁴ PFU of each of the indicated viruses (54). Mice were monitored daily until 18 days postinfection for survival and the severity of vaginal disease using a scoring system of 0 for no sign of disease, 1 for slight genital erythema and edema, 2 for moderate genital inflammation, 3 for purulent genital lesions, 4 for hind-limb paralysis, and 5 for death, as described previously (54). Virus titers in the vaginal secretions of mice were determined as described previously (54). All animal experiments were carried out in accordance with guidelines for proper conduct of animal experiments of the Science Council of Japan. The protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of the Institute of Medical Science, The University of Tokyo (IACUC protocol approvals 19-26, PA11-81, and PA16-69).

Statistical analysis.

Differences in viral replication and plaque size in cell cultures and the relative amounts of phosphorylated EF-1δ were statistically analyzed using analysis of variance (ANOVA) followed by Tukey's post hoc test. Differences in the survival of infected mice were statistically analyzed by the log rank test or the log rank test with Bonferroni adjustment for the three-comparison analyses. Differences in viral replication in vaginal wash specimens and in the disease scores of mice were statistically analyzed by the Mann-Whitney U test or Dunn's multiple-comparison test. A P value of ≤0.05 was considered statistically significant, except for the log rank test with Bonferroni adjustment for the three-comparison analyses, in which a P value of ≤0.0167 was considered statistically significant.