Role of the Arginine Cluster in the Disordered Domain of Herpes Simplex Virus 1 UL34 for the Recruitment of ESCRT-III for Viral Primary Envelopment

ABSTRACT

During the nuclear export of nascent nucleocapsids of herpesviruses, the nucleocapsids bud through the inner nuclear membrane (INM) by acquiring the INM as a primary envelope (primary envelopment). We recently reported that herpes simplex virus 1 (HSV-1) nuclear egress complex (NEC), which consists of UL34 and UL31, interacts with an endosomal sorting complex required for transport III (ESCRT-III) adaptor ALIX and recruits ESCRT-III machinery to the INM for efficient primary envelopment. In this study, we identified a cluster of six arginine residues in the disordered domain of UL34 as a minimal region required for the interaction with ALIX, as well as the recruitment of ALIX and an ESCRT-III protein CHMP4B to the INM in HSV-1-infected cells. Mutations in the arginine cluster exhibited phenotypes similar to those with ESCRT-III inhibition reported previously, including the mislocalization of NEC, induction of membranous invagination structures containing enveloped virions, aberrant accumulation of enveloped virions in the invaginations and perinuclear space, and reduction of viral replication. We also showed that the effect of the arginine cluster in UL34 on HSV-1 replication was dependent primarily on ALIX. These results indicated that the arginine cluster in the disordered domain of UL34 was required for the interaction with ALIX and the recruitment of ESCRT-III machinery to the INM to promote primary envelopment.

IMPORTANCE Herpesvirus UL34 homologs contain conserved amino-terminal domains that mediate vesicle formation through interactions with UL31 homologs during primary envelopment. UL34 homologs also comprise other domains adjacent to their membrane-anchoring regions, which differ in length, are variable in herpesviruses, and do not form distinguished secondary structures. However, the role of these disordered domains in infected cells remains to be elucidated. In this study, we present data suggesting that the arginine cluster in the disordered domain of HSV-1 UL34 mediates the interaction with ALIX, thereby leading to the recruitment of ESCRT-III machinery to the INM for efficient primary envelopment. This is the first study to report the role of the disordered domain of a UL34 homolog in herpesvirus infections.

INTRODUCTION

Viruses in the family Herpesviridae (herpesviruses) are subclassified into three subfamilies, Alphaherpesvirinae, Betaherpesvirinae, and Gammaherpesvirinae (1). Herpes simplex virus 1 (HSV-1), the subject of this study, is a member of the Alphaherpesvirinae subfamily and causes a variety of human diseases, including mucocutaneous diseases, keratitis, skin diseases, and encephalitis (2). Herpesviruses, including HSV-1, replicate their genomes and package the nascent progeny viral genomes into capsids in the nucleus, which are then transported to the cytoplasm, where they acquire a final envelope to produce infectious virions (3). The nuclear export of the nascent nucleocapsids of herpesviruses primarily depends on a unique nuclear egress mechanism. Thus, nucleocapsids acquire a primary envelope by budding through the inner nuclear membrane (INM) into the perinuclear space between the INM and the outer nuclear membrane (ONM) (primary envelopment). These enveloped nucleocapsids in the perinuclear space then fuse with the ONM to release nucleocapsids into the cytoplasm (de-envelopment) (3, 4).

The nuclear egress of HSV-1 nucleocapsids requires an HSV-1 heterodimeric complex termed the nuclear egress complex (NEC), which consists of UL31 and UL34 (5). UL31 and UL34 are conserved in all subfamilies of the family Herpesviridae, and their homologs have critical roles in the nuclear egress of nucleocapsids in other herpesvirus infections (3, 4, 6, 7). UL31 is a soluble nuclear protein (5), whereas UL34 is anchored to the INM by a carboxyl (C)-terminal transmembrane helix with several residues extending into the perinuclear space (5, 8), and its retention at the INM requires UL31 (5). Multiple sites in an amino (N)-terminal domain of UL34 interact with UL31, and this interaction appears to be critical for vesicle formation at the perinuclear space (9–11).

The endosomal sorting complex required for transport III (ESCRT-III) remodels membranes to function in a number of cellular processes, including extracellular microvesicle formation, enveloped virus budding, multivesicular body formation, the abscission stage of cytokinesis, repair of the plasma membrane and NM, and reformation of the NM (12, 13). ALIX, ESCRT-I/II complex, and/or charged multivesicular body protein 7 (CHMP7) are ESCRT-III adaptors that bind to regulators in various pathway-specific signals mediated by ESCRT-III machinery and recruit ESCRT-III proteins, including CHMP4A, -B, and -C, to their sites of action (13). In addition, Vps4, an AAA ATPase, works with ESCRT-III proteins during membrane scission by the disassembly and removal of ESCRT-III (13). Recently, we reported that HSV-1 infection induced the recruitment of CHMP4B and ALIX to the INM by an NEC-dependent mechanism (14). We also showed that HSV-1 NEC interacted with ALIX and that ALIX depletion impaired the recruitment of CHMP4B to the INM in HSV-1-infected cells (14). Furthermore, we demonstrated that the inhibition of ESCRT-III by the depletion of CHMP4A, -B, and/or -C; ALIX; or a Vps4 dominant negative mutant induced membranous structures that were invaginations of the INM into the nucleoplasm and contained enveloped virions and an aberrant accumulation of enveloped virions in the INM invaginations and perinuclear space, apparently due to impaired scission (14). These observations suggested a model whereby HSV-1 NEC interacts with ALIX to recruit the ESCRT-III machinery to the INM to promote viral primary envelopment. However, the detailed mechanism by which the HSV-1 NEC recruits ESCRT-III via ALIX to the INM, i.e., how the NEC-ALIX interaction is involved in the ESCRT-III recruitment to the INM as well as the promotion of primary envelopment, remains to be elucidated. To gain further insight into the mechanism of the NEC-mediated recruitment of ESCRT-III machinery to the INM and its significance in HSV-1 primary envelopment and replication, the current study mapped a region of the HSV-1 NEC required for the recruitment of CHMP4B. We then examined the effects of mutations in the identified region on the NEC interaction with ALIX and the recruitment of ALIX and CHMP4B to the INM in HSV-1-infected cells as well as viral primary envelopment and replication.

RESULTS

Generation of an HSV-1 UL34 mutant with a potentially impaired ability to recruit ESCRT-III.

We previously reported that CHMP4B fused to enhanced green fluorescent protein (EGFP) (CHMP4B-EGFP) was redistributed to the INM upon wild-type HSV-1 infection in HeLa-CHMP4B-EGFP cells, which stably expressed CHMP4-EGFP (14, 15). Notably, the distribution of CHMP4B-EGFP in HeLa-CHMP4B-EGFP cells infected with a UL34 deletion mutant virus was similar to that in mock-infected cells, whereas CHMP4B-EGFP was redistributed to the sites of UL34 accumulation in cells infected with a UL31 deletion mutant virus (14). These observations suggested that UL34, but not UL31, is involved in the recruitment of ESCRT-III in HSV-1-infected cells. Therefore, we focused on UL34 in this study and investigated whether UL34 itself had the ability to recruit ESCRT-III. To this end, UL34 fused to TagRFP (tRFP) (UL34-tRFP) at the C terminus (Fig. 1A and B) was ectopically expressed in HeLa-CHMP4B-EGFP cells, and the distribution of UL34-tRFP and CHMP4B-EGFP was examined by confocal microscopy. As reported previously, ectopically expressed UL34-tRFP was localized mainly within punctate structures in the cytoplasm and, to a lesser extent, was distributed at the nuclear rim (5, 16) (Fig. 1C). Notably, the punctate structures of UL34-tRFP were mostly colocalized with similar punctate structures of CHMP4B-EGFP in the cytoplasm (Fig. 1C). These results suggested that UL34 was able to recruit CHMP4B in the absence of any other HSV-1 proteins.

FIG 1.

Fluorescence microscopy of HeLa-CHMP4B-EGFP cells transfected with the UL34-tRFP expression plasmid. (A) Schematic diagrams of HSV-1 UL34. Amino acid sequence of the portion of the disordered domain (aa 170 to 206) of UL34 is shown. TM indicates the transmembrane domain. The arginine residues analyzed in this paper are highlighted by boxes. (B) Schematic diagrams of the wild-type and mutant UL34-tRFP expression plasmids. (C) HeLa-CHMP4B-EGFP cells were transfected with the indicated UL34-tRFP expression plasmid. At 24 h posttransfection, the cells were analyzed by confocal microscopy. Each image in the right column is the magnified image of the boxed area in the image to its left. Bars, 10 μm. (D) Percentage of cells (30 to 40 cells in each experiment) showing the colocalization of UL34-tRFP and CHMP4B-EGFP in experiments whose results are presented in panel C. Data are shown as the mean ± standard error from 3 independent experiments. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences between UL34-tRFP and tRFP, UL34-tRFP and UL34Δ170-248-tRFP, UL34-tRFP and UL34R6D-tRFP, UL34Δ207-248-tRFP and tRFP, UL34Δ207-248-tRFP and UL34Δ170-248-tRFP, and UL34Δ207-248-tRFP and UL34R6D-tRFP (*, P < 0.0001). (E) HeLa-CHMP4B-EGFP cells were transfected with the indicated expression plasmid. At 24 h posttransfection, cell lysates were analyzed by immunoblotting with anti-tRFP and anti-α-tubulin antibodies.

Next, we attempted to map the region(s) in UL34 required for ESCRT-III recruitment. To this end, we constructed two mutants of UL34-tRFP, UL34Δ170-248-tRFP and UL34Δ270-248-tRFP, in which a UL34 domain comprising amino acids 170 to 248 or the latter half of the domain, respectively, was deleted. We tested the mutants by confocal microscopy with HeLa-CHMP4B-EGFP cells. We focused on the domain comprising amino acids 170 to 248 in UL34 because it was reported that regions other than this UL34 domain contain multiple interfaces for interactions with UL31 (9, 11, 17), as well as sequences necessary for the anchoring of NEC to membranes (9, 17, 18). This suggests that mutations in these UL34 regions affect the essential nature of the NEC and, therefore, cannot be analyzed in these studies. As shown in Fig. 1C, ectopically expressed UL34Δ170-248-tRFP was diffusely distributed throughout the cytoplasm, unlike wild-type UL34-tRFP. In contrast, as observed with wild-type UL34-tRFP, ectopically expressed UL34Δ207-248-tRFP was localized mainly within punctate structures in the cytoplasm, and these structures were mostly colocalized with similar punctate structures of CHMP4B-EGFP (Fig. 1C). These results suggested that the UL34 domain comprising amino acids 207 to 248 was not required for ESCRT-III recruitment. We observed that the UL34 domain comprising amino acids 170 to 206 contained a cluster of arginine residues (Fig. 1A). It was reported that the nucleocapsid protein of human immunodeficiency virus (HIV), which promotes ESCRT-III-dependent viral budding, binds to ALIX through a cluster of basic amino acid residues (19, 20). Therefore, we generated an additional UL34-tRFP mutant (UL34R6D-tRFP) in which the arginine cluster comprising UL34 arginines 191, 192, 193, 194, 196, and 197 was replaced by aspartic acids (UL34R6D) (Fig. 1A and B), and we tested the mutant by confocal microscopy with HeLa-CHMP4B-EGFP cells. As shown in Fig. 1C, ectopically expressed UL34R6D-tRFP was localized mainly within punctate structures in the cytoplasm. However, unlike wild-type UL34-tRFP, these punctate structures of UL34R6D-tRFP showed low colocalization with punctate structures induced by CHMP4B-EGFP (Fig. 1C). The frequencies of cells with colocalization between UL34R6D-tRFP and CHMP4B-EGFP in the punctate structures were significantly lower than those with colocalization between UL34-tRFP and CHMP4B-EGFP (Fig. 1D). The expression of wild-type UL34-tRFP and its mutants in HeLa-CHMP4B-EGFP cells was confirmed by immunoblotting (Fig. 1E).

We also tested an additional mutant of UL34-tRFP, UL34Δ1-79-tRFP, in which an N-terminal UL34 domain comprising amino acids 1 to 79 was deleted (Fig. 2), together with UL34-tRFP and UL34R6D-tRFP. As observed with UL34R6D-tRFP, ectopically expressed UL34Δ1-79-tRFP was localized mainly within punctate structures in the cytoplasm, and these punctate structures of UL34Δ1-79-tRFP showed low colocalization with punctate structures induced by CHMP4B-EGFP (Fig. 2B). The frequencies of cells with colocalization between UL34Δ1-79-tRFP and CHMP4B-EGFP in the punctate structures and those with colocalization between UL34R6D-tRFP and CHMP4B-EGFP were significantly lower than those with colocalization between UL34-tRFP and CHMP4B-EGFP (Fig. 2C). Moreover, we tested UL34 fused to tRFP at the N terminus (tRFP-UL34) and its mutant tRFP-UL34R6D (Fig. 3A) and obtained results similar to those with UL34-tRFP and UL34R6D-tRFP (Fig. 3B and C). The expression of UL34-tRFP, tRFP-UL34, and their mutants in HeLa-CHMP4B-EGFP cells was confirmed by immunoblotting (Fig. 2D and Fig. 3D).

FIG 2.

Effects of N-terminal deletion of UL34-tRFP on localization of CHMP4B-EGFP and UL34-tRFP. (A) Schematic diagrams of the wild-type and mutant UL34-tRFP expression plasmids. (B) HeLa-CHMP4B-EGFP cells were transfected with the indicated UL34-tRFP expression plasmid. At 24 h posttransfection, the cells were analyzed by confocal microscopy. Each image in the right column is the magnified image of the boxed area in the image to its left. Bars, 10 μm. (C) Percentage of cells (30 to 40 cells in each experiment) showing the colocalization of UL34-tRFP and CHMP4B-EGFP in experiments whose results are presented in panel B. Data are shown as the mean ± standard error from 3 independent experiments. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences between UL34-tRFP and tRFP, UL34-tRFP and UL34Δ1-79-tRFP, and UL34-tRFP and UL34R6D-tRFP (*, P < 0.0001). (D) HeLa-CHMP4B-EGFP cells were transfected with the indicated expression plasmid. At 24 h posttransfection, cell lysates were analyzed by immunoblotting with anti-tRFP and anti-α-tubulin antibodies.

FIG 3.

Fluorescence microscopy of HeLa-CHMP4B-EGFP cells transfected with the tRFP-UL34 expression plasmid. (A) Schematic diagrams of the wild-type and mutant tRFP-UL34 expression plasmids. (B) HeLa-CHMP4B-EGFP cells were transfected with the indicated tRFP-UL34 expression plasmid. At 24 h posttransfection, the cells were analyzed by confocal microscopy. Each image in the right column is the magnified image of the boxed area in the image to its left. Bars, 10 μm. (C) Percentage of cells (30 to 40 cells in each experiment) showing the colocalization of tRFP-UL34 and CHMP4B-EGFP in experiments whose results are presented in panel B. Data are shown as the mean ± standard error from 3 independent experiments. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences between tRFP-UL34 and tRFP, and tRFP-UL34 and tRFP-UL34R6D (*, P < 0.0001). (D) HeLa-CHMP4B-EGFP cells were transfected with the indicated expression plasmid. At 24 h posttransfection, cell lysates were analyzed by immunoblotting with anti-tRFP and anti-α-tubulin antibodies.

Collectively, these results suggested that the UL34R6D mutant had an impaired ability to recruit ESCRT-III and that the UL34 disordered domain was not sufficient for the recruitment of ESCRT-III.

Construction of recombinant virus carrying mutations in the arginine cluster in UL34.

To investigate the role of the arginine cluster in UL34 in HSV-1-infected cells, we constructed a recombinant virus, YK745 (Strep-UL34R6D), encoding Strep-UL34 carrying R6D mutations in UL34 and its repaired virus YK746 (Strep-UL34R6D-repair) (Fig. 4A). As expected, HeLa cells infected with YK735 (Strep-UL34) (21), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) expressed Strep-tagged UL34, but cells infected with wild-type HSV-1(F) did not (Fig. 4B). HeLa cells infected with YK745 (Strep-UL34R6D) accumulated Strep-UL34 protein and/or viral proteins, including UL31, ICP4, and ICP8, at levels comparable with those in cells infected with wild-type HSV-1(F), YK735 (Strep-UL34), or YK746 (Strep-UL34R6D-repair) (Fig. 4B).

FIG 4.

Construction of recombinant HSV-1. (A) Schematic diagrams of the genome structure of wild-type HSV-1(F) and the relevant domains of the recombinant viruses used in this study. Line 1, wild-type HSV-1(F) genome; line 2, domain of the UL33 gene to the UL35 gene; line 3, domains of the UL34 gene; lines 4 to 6, recombinant viruses with mutations in UL34. (B) HeLa cells were mock infected or infected with wild-type HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 3 for 24 h. These cells were analyzed by immunoblotting with the indicated antibodies.

Effects of UL34R6D mutations on the interaction of the NEC with ALIX and on NEC formation in HSV-1-infected cells.

To examine the effect of the arginine cluster in UL34 on the interaction between the NEC and ALIX as well as NEC formation in HSV-1-infected cells, HeLa cells infected with YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) were lysed, precipitated with Strep-Tactin beads, and analyzed by immunoblotting. In agreement with our previous report (14), UL31 and ALIX were efficiently coprecipitated with Strep-UL34 from lysates of cells infected with YK735 (Strep-UL34) or YK746 (Strep-UL34R6D-repair). In contrast, ALIX was not coprecipitated with Strep-UL34R6D in YK745 (Strep-UL34-R6D)-infected cells, whereas UL31 was efficiently coprecipitated with Strep-UL34R6D in these infected cells as seen with Strep-UL34 and Strep-UL34R6D-repair in YK735 (Strep-UL34)- and YK746 (Strep-UL34R6D-repair)-infected cells, respectively (Fig. 5). These results indicated that the UL34R6D mutations abrogated the interaction between the NEC and ALIX without affecting NEC formation in HSV-1-infected cells and that the UL34 arginine cluster was required for the interaction of NEC with ALIX in HSV-1-infected cells.

FIG 5.

Effects of a mutation in the arginine cluster of UL34 on its interaction with ALIX in HSV-1-infected cells. HeLa cells were infected with wild-type HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 0.1 for 48 h, harvested, precipitated with Strep-Tactin Sepharose beads, and analyzed by immunoblotting with the indicated antibodies.

Effects of UL34R6D mutations on the distribution of ALIX and CHMP4B in HSV-1-infected cells.

To investigate the effect of the arginine cluster in UL34 on the recruitment of ALIX and CHMP4B to the INM in HSV-1-infected cells, HeLa-CHMP4B-EGFP and HeLa cells were infected with YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair), and the localization of CHMP4B-EGFP and ALIX in HeLa-CHMP4B-EGFP and HeLa cells, respectively, was analyzed by confocal microscopy. In agreement with the observations of wild-type HSV-1 in our previous report (14), ALIX and CHMP4B-EGFP were extensively redistributed to the nuclear rim and colocalized with lamin B1, a marker of the nuclear lamina that underlies the INM, and Strep-UL34, respectively, in cells infected with YK735 (Strep-UL34) or YK746 (Strep-UL34R6D-repair) (Fig. 6 and 7). In contrast, the redistribution of these cellular proteins to the nuclear rim was low in cells infected with YK745 (Strep-UL34R6D) (Fig. 6A and Fig. 7A). The frequencies of cells infected with YK735 (Strep-UL34) or YK746 (Strep-UL34R6D-repair) with localized ALIX and CHMP4B-EGFP at the nuclear rim were 70.5% and 70.5% for ALIX and 76.0% and 73.2% for CHMP4B-EGFP (Fig. 6B and Fig. 7B). In contrast, the frequencies decreased to 6.8% for ALIX and 13.6% for CHMP4B-EGFP in cells infected with YK745 (Strep-UL34R6D) (Fig. 6B and Fig. 7B). These results indicated that the UL34R6D mutant significantly impaired the ability to recruit ALIX and CHMP4B to the INM in HSV-1-infected cells and that the UL34 arginine cluster was required for the efficient recruitment of CHMP4B and ALIX to the INM in HSV-1-infected cells. We noted that the UL34R6D mutations mislocalized lamin B1 and Strep-UL34 in HSV-1-infected cells (Fig. 6A and Fig. 7A), and these results are described below in more detail.

FIG 6.

Effects of a mutation in the arginine cluster of UL34 on the localization of ALIX in HSV-1-infected cells examined by confocal microscopy. (A) HeLa cells were mock infected or infected with wild-type YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 3 for 20 h and then fixed, permeabilized, stained with anti-ALIX and anti-lamin B1 antibodies, and examined by confocal microscopy. Bars, 10 µm. (B) Percentage of cells (100 to 160 cells in each experiment) showing nuclear rim localization of ALIX in experiments whose results are presented in panel A. Data are shown as the mean ± standard error from 3 independent experiments. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences as follows: *, P < 0.0001.

FIG 7.

Effects of a mutation in the arginine cluster of UL34 on the localization of CHMP4B-EGFP in HSV-1-infected cells examined by confocal microscopy. (A) HeLa-CHMP4B-EGFP cells were mock infected or infected with wild-type YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 3 for 20 h and then fixed, permeabilized, stained with anti-strep antibodies, and examined by confocal microscopy. Each image in the right column is the magnified image of the boxed area in the image to its left. Dotted lines indicate the nucleus in infected cells. Bars, 10 µm. (B) Percentage of cells (100 to 160 cells in each experiment) showing nuclear rim localization of CHMP4B-EGFP in experiments whose results are presented in panel A. Data are shown as the mean ± standard error from 3 independent experiments. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences as follows: *, P < 0.0001.

Effects of UL34R6D mutations on the localization of the NEC in HSV-1-infected cells.

We also investigated the localization of UL31 and UL34 in HeLa cells infected with YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) by confocal microscopy. As reported previously with wild-type HSV-1 (5), Strep-UL34 and UL31 were smoothly distributed at the nuclear rim and colocalized with lamin B1 in cells infected with YK735 (Strep-UL34) or YK746 (Strep-UL34R6D-repair) (Fig. 8A and C). In contrast, in cells infected with YK745 (Strep-UL34R6D), Strep-UL34R6D and UL31 were not only smoothly distributed at the nuclear rim but also present in aberrant punctate structures induced at the nuclear rim and in the nucleus (Fig. 7 and 8). The frequencies of cells infected with YK735 (Strep-UL34) or YK746 (Strep-UL34R6D-repair) with punctate structures of Strep-UL34 and UL31 were 4.7% and 3.1% for Strep-UL34 and 18.2% and 14.2% for UL31 (Fig. 8B and D). In contrast, the frequencies increased to 90.0% for Strep-UL34 and 93.7% for UL31 in cells infected with YK745 (Strep-UL34R6D) (Fig. 8B and D). Furthermore, a fraction of lamin B1 proteins were also redistributed to the punctate structures of Strep-UL34R6D and UL31 (Fig. 8A and C). These localization patterns of UL34, UL31, and lamin B1 in HeLa cells infected with YK745 (Strep-UL34R6D) strongly resembled those of the viral and cellular proteins in wild-type HSV-1-infected HeLa cells in which CHMP4A, -B, and/or -C or ALIX was depleted (14). These results indicated that the arginine cluster in UL34 was required for the proper localization of the NEC and the integrity of the INM in HSV-1-infected cells.

FIG 8.

Effects of a mutation in the arginine cluster of UL34 on the localization of UL31 and UL34 in HSV-1-infected cells examined by confocal microscopy. (A and C) HeLa cells were infected with wild-type YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 3 for 18 h and then fixed, permeabilized, stained with anti-lamin B1 and anti-strep (A) or anti-UL31 (C) antibodies, and examined by confocal microscopy. Bars, 10 µm. (B and D) Percentage of cells (100 to 140 cells in each experiment) showing punctate structures of UL34 (B) or UL31 (D) in experiments whose results are presented in panels A and C. Data are shown as the mean ± standard error from 3 independent experiments. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences as follows: *, P < 0.0001.

It was reported that a mutation(s) in UL34, which abrogated its membrane anchoring, aberrantly redistributed UL34 and UL31 throughout the nucleoplasm in HSV-1-infected cells (9). As described above, Strep-UL34R6D did not show a diffuse distribution in the nucleoplasm, although it was still colocalized with lamin B1 at the nuclear rim and with nuclear punctate structures, suggesting that the mutant NEC carrying the R6D mutation in UL34 retained the ability to interact with the membrane in HSV-1-infected cells.

Effects of phosphomimetic substitutions at the phosphorylation sites in the UL34 arginine cluster region on the distribution of CHMP4B and the NEC in HSV-1-infected cells.

It has been reported that threonine 195 and serine 198 within the UL34 arginine cluster region (Fig. 9A) are phosphorylated by HSV-1 Us3 protein kinase in HSV-1-infected cells (22–24). It is known that replacement of the phosphorylation site with an acidic amino acid such as aspartic acid or glutamic acid mimics the negative charges produced by phosphorylation (25–28). Therefore, the results with the UL34R6D mutant above raised a possibility that the disruption of UL34 regulation by Us3 was required for the failure to recruit ESCRT-III to the INM. To test this possibility, we generated a recombinant virus YK747 (UL34T195/S198D) carrying phosphomimetic (aspartic acid) substitutions at the UL34 phosphorylation sites (UL34T195/S198D) (Fig. 9B). HeLa-CHMP4B-EGFP and HeLa cells were infected with wild-type HSV-1(F) or YK747 (UL34T195/S198D), and the localization of CHMP4B-EGFP, UL34m, and/or UL31 in HeLa-CHMP4B-EGFP and/or HeLa cells was analyzed by confocal microscopy. As shown in Fig. 9C and D, localization patterns of CHMP4B-EGFP, UL34, and UL31 in these cells infected with YK747 (UL34T195/S198D) were similar to those in the cells infected with wild-type HSV-1(F). These results indicated that, unlike the R6D mutations in the UL34 arginine cluster region, the 2 phosphomimetic substitutions at the UL34 phosphorylation sites in the UL34 arginine cluster region had no effect on the recruitment of ESCRT-III to the INM in HSV-1-infected cells and suggested that the possibility above was less likely.

FIG 9.

Effects of phosphomimetic substitutions at the UL34 phosphorylation sites on the localization of UL31, UL34, and CHMP4B-EGFP in HSV-1-infected cells. (A) Schematic diagrams of HSV-1 UL34. Amino acid sequence of the portion of the disordered domain (aa 170 to 206) of UL34 is shown. The UL34 phosphorylation sites by Us3 are indicated by arrowheads. TM indicates the transmembrane domain. The arginine residues analyzed in this paper are highlighted by boxes. (B) Schematic diagrams of the genome structure of wild-type HSV-1(F) and the relevant domains of the recombinant virus used in this study. Line 1, wild-type HSV-1(F) genome; line 2, domain of the UL33 gene to the UL35 gene; line 3, domains of the UL34 gene; lines 4, recombinant virus carrying phosphomimetic (aspartic acid) substitutions at the UL34 phosphorylation sites. (C) HeLa cells were infected with wild-type HSV-1(F) or YK747 (UL34T195/S198D) at an MOI of 3 for 18 h and then fixed, permeabilized, stained with anti-UL31 and anti-UL34 antibodies, and examined by confocal microscopy. (D) HeLa-CHMP4B-EGFP cells were infected with wild-type HSV-1(F) or YK747 (UL34T195/S198D) at an MOI of 3 for 20 h and then fixed, permeabilized, stained with anti-UL34 antibody, and examined by confocal microscopy. Bars, 10 µm.

Effects of UL34R6D mutations on HSV-1 virion morphogenesis.

To examine the effect of the arginine cluster in UL34 in HSV-1-infected cells at the ultrastructural level, HeLa cells infected with wild-type HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) were subjected to electron microscopy. As shown in Fig. 10A, membranous invagination structures, in which enveloped virions accumulated, were induced adjacent to the nuclear rim in cells infected with YK745 (Strep-UL34R6D).

FIG 10.

Effects of a mutation in the arginine cluster of UL34 on HSV-1 nuclear egress. (A) HeLa cells were infected with wild-type HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 3 for 20 h. Then, the infected cells were examined by transmission electron microscopy. Arrowheads indicate virions defective in the scission steps. N, nucleus; C, cytoplasm; NM, nuclear membrane. Bars, 500 nm. (B) The mean proportions of enveloped virions in the perinuclear space of 14 infected cells were determined. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences as follows: *, P < 0.001.

We also analyzed viral morphogenesis by quantitating the number of virus particles at different morphogenetic stages in these infected cells. As shown in Table 1 and Fig. 10B, 2.4%, 5.9%, and 5.7% of the total number of virus particles were enveloped virions in the perinuclear space in HeLa cells infected with wild-type HSV-1(F), YK735 (Strep-UL34), or YK746 (Strep-UL34R6D-repair), respectively. In contrast, the fraction of total virus particles that were in the perinuclear space in cells infected with YK745 (Strep-UL34R6D) increased to 19.5% (Table 1 and Fig. 10B). This represented 8.1-, 3.3-, and 3.4-fold increases compared with cells infected with wild-type HSV-1(F), YK735 (Strep-UL34), or YK746 (Strep-UL34R6D-repair), respectively (Table 1 and Fig. 10B).

TABLE 1.

Effect of the R6D mutation in UL34 on the distribution of virus particles in infected HeLa cells

Cell type	Percentage of virus particles in the morphogenetic stage (mean ± SE) (no. of virus particles)					Total (particles/cells)
	Nucleocapsids in the nucleus	Enveloped virions in the perinuclear space	Nucleocapsids in the cytoplasm	Enveloped virions in the cytoplasm	Extracellular enveloped virions
HSV-1(F)	63.0 ± 12.4 (731)	2.4 ± 1.0 (26)	7.8 ± 3.1 (91)	4.0 ± 2.0 (47)	22.8 ± 4.6 (265)	1161/14
Strep-UL34	54.6 ± 10.8 (678)	5.9 ± 2.4 (73)	8.5 ± 3.4 (106)	5.3 ± 2.2 (65)	25.8 ± 4.1 (320)	1242/14
Strep-UL34R6D	66.2 ± 12.2 (1064)	19.5 ± 4.0 (314)	3.4 ± 1.4 (54)	1.0 ± 0.7 (16)	10.0 ± 2.5 (160)	1608/14
Strep-UL34R6D-rep	50.4 ± 14.4 (746)	5.7 ± 3.3 (85)	9.5 ± 3.2 (141)	5.7 ± 2.0 (85)	28.6 ± 6.5 (424)	1481/14

Together, these results indicated that the UL34R6D mutations exhibited phenotypes similar to those by the depletion of CHMP4A, -B, and/or -C or ALIX with respect to the induction of membranous invagination structures containing enveloped virions adjacent to the nuclear rim and the aberrant accumulation of enveloped virions in the perinuclear space and the invagination structures shown in our previous report (14). In contrast to a report showing the depletion of CHMP4A, -B, and/or -C produced a significant accumulation of nucleocapsids in the cytoplasm (14), the accumulation of cytoplasmic nucleocapsids was not observed in cells infected with YK745 (Strep-UL34-R6D) (Table 1). Thus, these results indicated that the arginine cluster in UL34 was required for the efficient nuclear egress of nucleocapsids and the integrity of the INM in HSV-1-infected cells but not for final envelopment in the cytoplasm.

Effects of UL34 R6D mutations on HSV-1 replication.

To examine the effect of the arginine cluster in UL34 on HSV-1 replication, HeLa cells were infected with wild-type HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at a multiplicity of infection (MOI) of 3 or 0.01, and viral titers were assayed at various times postinfection. As shown in Fig. 11, YK745 (Strep-UL34R6D) growth was lower than that of wild-type HSV-1(F), YK735 (Strep-UL34), and YK746 (Strep-UL34R6D-repair) in these infected cells. The progeny virus titer of YK745 (Strep-UL34R6D) at an MOI of 3 at 24 h postinfection was significantly lower than that of wild-type HSV-1(F), YK735 (Strep-UL34), and YK746 (Strep-UL34R6D-repair) (7.6-, 7.6-, and 8.8-fold, respectively) (Fig. 11A). In cells infected at an MOI of 0.01 at 48 h postinfection, the progeny virus titer of YK745 (Strep-UL34R6D) was significantly lower than that of wild-type HSV-1(F), YK735 (Strep-UL34), and YK746 (Strep-UL34R6D-repair) (33-, 40-, and 31-fold, respectively) (Fig. 11B).

FIG 11.

Effects of a mutation in the arginine cluster of UL34 on HSV-1 growth. HeLa cells were infected with wild-type HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 3 (A) or 0.01 (B). The infected cells were harvested at the indicated times postinfection, and progeny viruses were assayed on Vero cells. Each data point is the mean ± standard error of 4 (MOI = 3) or 3 (MOI = 0.01) independent experiments. Statistical analysis was performed by the Tukey test. The asterisks indicate statistically significant differences between YK745 (Strep-UL34R6D) and HSV-1(F), YK745 (Strep-UL34R6D) and YK735 (Strep-UL34), and YK745 (Strep-UL34R6D) and YK746 (Strep-UL34R6D-repair) (*, P < 0.05).

Finally, we compared the effects of the arginine cluster in UL34 on HSV-1 replication in HeLa cells to those in HeLa-ALIX low cells expressing a significantly reduced level of ALIX protein (14). For these studies, HeLa or HeLa-ALIX low cells were infected with YK745 (Strep-UL34R6D) or YK746 (Strep-UL34R6D-repair) at an MOI of 0.01, and progeny virus titers were determined at 48 h postinfection. Whereas the progeny virus titers of YK745 (Strep-UL34R6D) were significantly lower than those of YK746 (Strep-UL34R6D-repair) (Fig. 12), in agreement with the results in Fig. 11, the progeny virus titers of YK745 (Strep-UL34R6D) in HeLa-ALIX low cells were similar to those of YK746 (Strep-UL34R6D-repair) (Fig. 12). These results suggested that the positive effects of the arginine cluster in UL34 on HSV-1 replication are dependent primarily on ALIX.

FIG 12.

Effects of a mutation in the arginine cluster of UL34 on HSV-1 growth in the presence or absence of ALIX. HeLa cells (A) or HeLa-ALIX low cells (B) were infected with YK745 (Strep-UL34R6D) or YK746 (Strep-UL34R6D-repair) at an MOI of 0.01. The infected cells were harvested at 48 h postinfection, and progeny viruses were titrated. Each data point is the mean ± standard error of 3 independent experiments. Statistical analysis was performed by the unpaired Student's t test. The asterisks indicate statistically significant differences as follows: *, P < 0.05.

DISCUSSION

In this study, we focused on the role of ESCRT-III in the budding step of HSV-1 at the INM and present data suggesting that the arginine cluster in the disordered domain of HSV-1 UL34 mediates the interaction with ALIX, thereby leading to the recruitment of ESCRT-III machinery to the INM to promote primary envelopment. The arginine cluster in the C-terminal disordered domain of HSV-1 UL34 is conserved in HSV-2 but not in other viruses in the subfamily Alphaherpesvirinae (Fig. 13A). Although the N-terminal half of UL34 homologs, which mediate vesicle formation via interactions with UL31 homologs (11, 29–31), is conserved in the family Herpesviridae, the C-terminal half of UL34 homologs is less conserved with different lengths (Fig. 13B) and does not form distinguished secondary structures. It is noteworthy that these C-terminal domains of UL34 homologs contain abundant basic residues, including arginine and lysine (Fig. 13). Similar to HSV-1, it was reported that ESCRT-III is involved in the nuclear egress of Epstein-Barr virus (EBV) that belongs to the subfamily Gammaherpesvirinae (32, 33). Thus, ESCRT-III inhibition led to an accumulation of viral genomes and capsids in the nucleus, and CHMP4B and ALIX were redistributed to the nucleus upon the lytic reactivation of latent EBV (33). BFRF1, a homolog of HSV-1 UL34, was sufficient to distribute ALIX and Vps4 to the NM, and the ectopic expression of BFRF1 induced vesicles derived from the NM in the cytoplasm by an ESCRT-III-dependent mechanism (33). Of note, multiple domains within BFRF1, including the C-terminal disordered domain, were required for vesicle formation and ALIX recruitment (33). Collectively, these observations raised the possibility that a cluster of basic amino acid residues in the C-terminal disordered domain of UL34 homologs commonly mediate the interaction with ESCRT-III. Although it has been well established that the N-terminal domains of UL34 homologs have conserved roles in vesicle formation during primary envelopment through interactions with UL31 homologs, function(s) of the C-terminal disordered domains of UL34 homologs remains largely unknown. This is the first report showing the role of the disordered domain of a UL34 homolog in herpesvirus-infected cells. We should note that a recent study proposed a model that highly basic clusters in membrane-proximal regions (MPRs) of NEC promote negative membrane curvature by inducing lipid ordering during membrane budding, based on the observations in in vitro budding assays (34). In that study (34), deletion of UL31 MPR reduced significantly (∼90%) membrane budding, whereas that of UL34 MPR, which lies within the UL34 disordered domain, reduced membrane budding only moderately (∼37%), suggesting that UL31 MPR plays an important role but UL34 MPR plays an additive role in this process. Although at present, it remains unknown whether the observations in the study, in fact, reflect membrane budding mediated by the NEC during the primary envelopment in HSV-1-infected cells, the UL34 arginine cluster may contribute to the primary envelopment not only by recruiting ESCRT-III to the INM for membrane scission but also by promoting membrane deformation in infected cells. Further studies to clarify the significance of these disordered domains of UL34 homologs in infected cells will be of importance and are underway. Interestingly, in the study (34), clusters of positive charges in the UL31 MPR were shown to be important for membrane deformation based on the observations that reducing positive charge or introducing negative charge in the UL31 MPR impaired membrane deformation ability of the NEC. In the present study, we showed that the replacement of all of the 6 positively charged arginines in the UL34 arginine cluster by negative charged aspartic acids inhibited the recruitment of ESCRT-III to the INM. Together, these observations raised a possibility that the shift in charge in the UL34 arginine cluster disrupts the interaction between the NEC and ESCRT-III rather than the loss of the arginines themselves. Furthermore, we cannot completely eliminate another possibility that the NEC carrying the R6D mutation in the UL34 arginine cluster might be defective in membrane scission mediated by the NEC by itself during primary envelopment as a result of the mutation.

FIG 13.

Alignment of amino acid sequences in the C-terminal regions of UL34 homologues. (A) Alignment of amino acid sequences in the disordered region of UL34 homologues in members of the Alphaherpesviridae, i.e., HSV-1 (GenBank accession no. GU734771.1), HSV-2 (GenBank accession no. Z86099.2), varicella-zoster virus (VZV; GenBank accession no. NC_001348), pseudorabies virus (PrV; GenBank accession no. JQ809328.1), equine herpesvirus 1 (EHV-1; GenBank accession no. AY665713.1), bovine herpesvirus 1 (BHV-1; GenBank accession no. NC_001847.1), and Gallid herpesvirus 2 (GaHV-2; GenBank accession no. AF147806.2). The residues conserved in at least four herpesviruses are shaded. (B) Alignment of the amino acid sequences in the C-terminal regions of UL34 homologues in representative members of three subfamilies in Herpesviridae. i.e., human cytomegalovirus (HCMV) (GenBank accession no. NC_006273.2) and EBV (GenBank accession no. AJ507799.2). The residues conserved in at least two herpesviruses are shaded. The arginine and lysine residues are shown in red. The arginine residues analyzed in this paper are indicated by arrowheads. Predicted transmembrane domains are highlighted by boxes.

Earlier studies on other enveloped viruses, including retroviruses and ebolavirus, revealed that short motif peptides called “L-domains” in viral structural proteins are, in principle, recognized by ESCRT-III adaptors for the budding of the viruses (35). These include PTAP, YPXL, and PPXY motifs that interact with ESCRT-I, ALIX, and NEDD4 ubiquitin ligases, respectively (35). In contrast, it was reported that some viruses have evolved viral proteins that interact with ESCRT-III adaptors through noncanonical motifs (36). For example, the FPIV motif in the matrix protein of parainfluenza virus 5 might mediate viral budding through interactions with ubiquitin (36, 37). UL34 does not contain any typical L-domains, and, instead, we showed in this study that the arginine cluster in the disordered domain of UL34 was required for its binding to ALIX. Similarly, the HIV nucleocapsid protein, which was shown to be critical for viral budding (19), interacts with ALIX through the basic residues in this viral protein (20) as described above, although HIV budding predominantly depends on PTAP and YPXL motifs in one of the HIV gag proteins, p6 protein (35). Moreover, NEDD4L ubiquitin ligase was reported to promote ESCRT-III-dependent HIV budding through its binding to angiomotin, which was shown to bind to a gag protein(s) (38), in a manner independent of the PTAP and YPXL motifs (38–40). Thus, our study reveals a novel mode of the interaction between a viral protein and ESCRT-III machinery and will provide further insight into the mechanisms of ESCRT-III-dependent viral budding.

ESCRT-III was reported to have multiple roles at the NM by reforming, resealing, and repairing the membrane (13). Therefore, in our previous report showing the involvement of ESCRT-III in HSV-1 primary envelopment by the inhibition of ESCRT-III (14), we cannot completely eliminate the possibilities that the recruitment of ESCRT-III proteins mediated by the NEC in HSV-1-infected cells was totally unrelated to viral primary envelopment, and the effects of ESCRT-III inhibition on the INM in HSV-1-infected cells were not a result of defects in viral nuclear egress. Rather, they were caused by problems with the reported roles of ESCRT-III in the regulation of NM reformation due to direct ESCRT-III inhibition. One approach to address this issue is to investigate the effects of a mutation(s) in the HSV-1 effector for ESCRT-III, NEC, which precludes its interaction with ALIX as well as the recruitment of ESCRT-III proteins to the INM and compare the phenotypes caused by the NEC mutation(s) to those observed with the inhibition of ESCRT-III reported previously (14). In this study, we demonstrated that mutations in the arginine cluster in the C-terminal disordered domain of UL34 (the UL34R6D mutations) abrogated the NEC interaction with ALIX and the recruitment of CHMP4B without affecting NEC formation and the ability of the NEC to localize to the NM in HSV-1-infected cells. We also showed that the effects of UL34R6D mutations on HSV-1 replication were dependent primarily on ALIX. Based on the observations in this study, it seems likely that the UL34R6D mutations inhibit only the effect of UL34 on ESCRT-III via ALIX but not other UL34 functions. Notably, the UL34R6D mutations exhibited phenotypes similar to those observed by the inhibition of ESCRT-III reported previously (14). In these experiments, the effects of the UL34R6D mutations were obtained without directly inhibiting any ESCRT-III proteins, and therefore, these results may eliminate the possibilities discussed above and further support the model that HSV-1 NEC interacts with ALIX and recruits the ESCRT-III machinery to the NM to promote viral primary envelopment. One distinct difference in the phenotypes between the UL34R6D mutations and ESCRT-III inhibition was the effect on the accumulation of cytoplasmic capsids. As described above, ESCRT-III functions in the HSV-1 final envelopment in the cytoplasm, and the inhibition of ESCRT-III was reported to accumulate cytoplasmic capsids (14, 41). In contrast, the UL34R6D mutations did not increase the frequencies of cytoplasmic capsids in this study, suggesting that the effect of UL34 on ESCRT-III via ALIX is not involved in final envelopment. This is in agreement with earlier studies showing that ALIX is not required for final envelopment (42) and that UL31 and UL34 were not detected in matured HSV-1 virions (43). The HSV-1 regulators for final envelopment reported to date are mostly incorporated into matured virions (44–55).

MATERIALS AND METHODS

Cells and viruses.

Vero, HeLa, HeLa-CHMP4B-EGFP, HeLa-ALIX-low, and rabbit skin cells were described previously (14, 56, 57). Wild-type HSV-1(F) and wild-type YK312 and YK735 (Strep-UL34) (Fig. 2) were described previously (21, 56).

Plasmids.

pTagRFP-N1 was constructed by substituting the EGFP coding region in pEGFP-N1 with a DNA fragment containing the TagRFP coding region, amplified by PCR from pTagRFP-C (Envrogen, Moscow, Russia). A plasmid encoding a fusion protein of the UL34 and Tag-RFP (pUL34-tRFP) was constructed by cloning the HSV-1 UL34 gene, amplified by PCR from pYEbac102, containing a full-length infectious HSV-1(F) clone (56), into pTagRFP-N1 in frame with TagRFP. Plasmid pUL34Δ170-248-tRFP, carrying a fusion protein of the structured and TM domains of UL34 (codons 1 to 169 and 249 to 273) and TagRFP, was constructed by amplifying and cloning UL34 codons 1 to 169 and 248 to 273 into pTagRFP-N1 in frame with TagRFP. Similarly, plasmid pUL34-Δ207-248-tRFP, carrying a fusion protein of the structured and TM domains of UL34 (codons 1 to 206 and 248 to 273) and TagRFP, was constructed. Plasmid pUL34R6D-tRFP was constructed by replacing the UL34 codons R191, R192, R193, R194 R196, and R197 in pUL34-tRFP with aspartic acid as described previously (54). Plasmid pUL34Δ1-79-tRFP, carrying a fusion protein of the UL34 lacking the sequence encoding the N-terminal domain (codons 80 to 275) and TagRFP, was constructed by amplifying and cloning UL34 codons 80 to 275 into pTagRFP-N1 in frame with TagRFP. A plasmid encoding UL34 or UL34R6D fused to tRFP at the N terminus (tRFP-UL34 or ptRFP-UL34R6D, respectively) was constructed by replacing the progerin gene in pTagRFP-progerin (15) with the HSV-1 UL34 gene or UL34R6D gene, amplified by PCR from pUL34-tRFP or pUL34R6D-tRFP, respectively.

Mutagenesis of viral genomes and the generation of recombinant HSV-1.

Recombinant virus YK745 (Strep-UL34R6D), encoding Strep-tagged UL34 and carrying the R191D, R192D, R193D, R194D, R196D, and R197D mutations in UL34 (Fig. 2), was generated by a two-step Red-mediated mutagenesis procedure using Escherichia coli GS1783 containing pYEbac102Cre (58, 59) and the primers 5′-CGGATCCTGTGCCGCGCCGCCGAGCAGGCTATTACCGATGACGACGATACCGATGATTCCAGGATGACGACGATAAGTAGGG-3′ and 5′-TCGGCCCCGTACGCCTCCCGGGAATCATCGGTATCGTCGTCATCGGTAATAGCCTGCTCGCAACCAATTAACCAATTCTGATTAG-3′, followed by 5′-CTCCCATCGCGGGCGCCATGTGGAGCCATCCGCAGTTTGAAAAGGCGGGACTGGGCAAGCAGGATGACGACGATAAGTAGGG-3′ and 5′-GGTTTACGCGGGCACGCACGCTCCCATCGCGGGCGCCATGTGGAGCCATCCGCAGTTTGAAAAGGCGGGACTGGGCAAGCCCTAAGGATGACGACGATAAGTAGGG-3′. Recombinant virus YK746 (Strep-UL34R6D-rep), in which the UL34-R6D mutations in YK745 were repaired (Fig. 2), was generated as described above except that the primers 5′-CGGATCCTGTGCCGCGCCGCCGAGCAGGCTATTACCCGTCGCCGCCGAACCCGGCGGTCCAGGATGACGACGATAAGTAGGG-3′ and 5′-TCGGCCCCGTACGCCTCCCGGGACCGCCGGGTTCGGCGGCGACGGGTAATAGCCTGCTCGCAACCAATTAACCAATTCTGATTAG-3′ were used. To verify the sequences of the gene encoding UL34 and those encoding UL31, UL47, ICP22, and US3, all of which were reported to interact with UL34 (5, 60, 61), Vero cells were infected with YK745 (Strep-UL34R6D) or its parental virus YK312 (wild type) (56) at an MOI of 3, harvested at 24 h postinfection, and lysed in 500 μl viral genome purification buffer (10 mM Tris-HCl [pH 7.5], 150 mM NaCl, 1.5 mM MgCl₂, 0.1% Nonidet P-40 [NP-40]). After a brief centrifugation, β-mercaptoethanol and EDTA were added to 400 μl of each supernatant to final concentrations of 50 mM and 1 mM, respectively. DNA was extracted with phenol-chloroform and precipitated with ethanol. The sequences of the genes encoding UL25, UL31, UL34, UL47, ICP22, and US3 were determined as described previously (21, 62). The sequences of these genes in the YK745 (Strep-UL34R6D) genome were identical to those in the YK312 (parental strain) genome except for the coding sequence of Strep-tag and the substitution mutation of interest (R6D in UL34). Recombinant virus YK747 (UL34T195/S197D), encoding UL34 and carrying the T195D and S197D mutations in UL34 (Fig. 9), was generated as described above except that the primers 5′-GTGCCGCGCCGCCGAGCAGGCTATTACCCGTCGCCGCCGAGACCGGCGGGACCGGGAGGCGTACGGGGCCGAAGGATGACGACGATAAGTAGGG-3′ and 5′-CGGCCACCCCCAGCCCGGCCTCGGCCCCGTACGCCTCCCGGTCCCGCCGGTCTCGGCGGCGACGGGTAATAGCAACCAATTAACCAATTCTGATTAG-3′ were used.

Antibodies.

Mouse monoclonal antibodies against α-tubulin (clone DM1A; Sigma), ICP4 (58S; ATCC), ICP8 (clone 10A3; Millipore), ALIX (catalog no. sc-53540; Santa Cruz Biotechnology), Strep-tag (clone 4F1; MBL), and a commercial rabbit polyclonal antibody against TagRFP (catalog no. AB233; Evrogen) were used in this study. Mouse polyclonal antibody against UL31 and rabbit polyclonal antibody against UL31 and UL34 were described previously (61). Rabbit polyclonal antibody against UL31 was used for immunoblotting, and mouse polyclonal antibody against UL31 was used for immunofluorescence.

Immunoblotting and immunofluorescence.

Immunoblotting and immunofluorescence were performed as described previously (14, 63).

Affinity precipitation.

HeLa cells were infected with HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 0.1 for 48 h, collected, and lysed with 0.5% NP-40 buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, and 0.5% NP-40) containing a protease inhibitor cocktail (Nacalai Tesque). After centrifugation, the supernatants were reacted with StrepTactin Sepharose beads (IBA) with rotation for 2 h at 4°C. The precipitates were collected by brief centrifugation, washed extensively with 0.5% NP-40 buffer, and analyzed by immunoblotting.

Electron microscopy.

HeLa cells infected with HSV-1(F), YK735 (Strep-UL34), YK745 (Strep-UL34R6D), or YK746 (Strep-UL34R6D-repair) at an MOI of 3 for 20 h were examined by ultrathin-section electron microscopy as described previously (64). The number of virus particles at different morphogenetic stages was quantitated in several randomly chosen HeLa cells.

Statistical analysis.

For the comparison of two groups, statistical analysis was performed using the unpaired Student's t test. Tukey’s test was used for multiple comparisons. A P value of >0.05 was considered not significant (n.s.).