ABSTRACT
During nuclear egress of nascent progeny herpesvirus nucleocapsids, the nucleocapsids acquire a primary envelope by budding through the inner nuclear membrane of infected cells into the perinuclear space between the inner and outer nuclear membranes. Herpes simplex virus 1 (HSV-1) UL34 and UL31 proteins form a nuclear egress complex (NEC) and play critical roles in this budding process, designated primary envelopment. To clarify the role of NEC binding to progeny nucleocapsids in HSV-1 primary envelopment, we established an assay system for HSV-1 NEC binding to nucleocapsids and capsid proteins in vitro. Using this assay system, we showed that HSV-1 NEC bound to nucleocapsids and to capsid protein UL25 but not to the other capsid proteins tested (i.e., VP5, VP23, and UL17) and that HSV-1 NEC binding of nucleocapsids was mediated by the interaction of NEC with UL25. UL31 residues arginine-281 (R281) and aspartic acid-282 (D282) were required for efficient NEC binding to nucleocapsids and UL25. We also showed that alanine substitution of UL31 R281 and D282 reduced HSV-1 replication, caused aberrant accumulation of capsids in the nucleus, and induced an accumulation of empty vesicles that were similar in size and morphology to primary envelopes in the perinuclear space. These results suggested that NEC binding via UL31 R281 and D282 to nucleocapsids, and probably to UL25 in the nucleocapsids, has an important role in HSV-1 replication by promoting the incorporation of nucleocapsids into vesicles during primary envelopment.
IMPORTANCE Binding of HSV-1 NEC to nucleocapsids has been thought to promote nucleocapsid budding at the inner nuclear membrane and subsequent incorporation of nucleocapsids into vesicles during nuclear egress of nucleocapsids. However, data to directly support this hypothesis have not been reported thus far. In this study, we have present data showing that two amino acids in the membrane-distal face of the HSV-1 NEC, which contains the putative capsid binding site based on the solved NEC structure, were in fact required for efficient NEC binding to nucleocapsids and for efficient incorporation of nucleocapsids into vesicles during primary envelopment. This is the first report showing direct linkage between NEC binding to nucleocapsids and an increase in nucleocapsid incorporation into vesicles during herpesvirus primary envelopment.
INTRODUCTION
Herpesviruses in the family Herpesviridae 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, e.g., mucocutaneous diseases, keratitis, skin diseases, and encephalitis (2). Herpesviruses, including HSV-1, replicate their DNA genomes and package the nascent progeny virus genomes into capsids in the nuclei of infected cells (3–5). Herpesvirus capsids have some common structural features (3–5), as follows: an icosahedral shape formed by 161 capsomeres (150 hexons and 11 pentons), a portal complex with an axial channel through which viral genomic DNA enters and exits the capsid, 320 triplexes that connect the capsomeres and the portal complex, 150 hexameric rings of small capsomere-interacting proteins (SCPs) that cover the outer surface of each hexon, and 5 rod-shaped structures of capsid vertex-specific components (CVSCs) that project radially outward from each penton. In HSV-1 capsids, the pentons and hexons are composed of 5 and 6 VP5 molecules, respectively, the CVSCs are composed of 1 molecule of UL17 and 1 molecule of UL25, the triplexes are composed of 1 molecule of VP19C and 2 molecules of VP23, the portal complex is composed of 12 molecules of UL6, and the SCPs are VP26. Three types of capsids have been detected in HSV-1-infected cells. The A and B capsids are incomplete structures resulting from problems in viral genome packaging (3–6). The C capsids are mature capsids containing viral genomes (nucleocapsids) and on which CVSCs are specifically enriched (7–12).
Nascent nucleocapsids in the nucleus are translocated to the cytoplasm where final envelopment of progeny herpesviruses takes place (13, 14). The nucleocapsids acquire a primary envelope during nuclear export by budding through the inner nuclear membrane (INM) into the perinuclear space between the INM and the outer nuclear membrane (ONM) (primary envelopment). The enveloped nucleocapsids in the perinuclear space then fuse with the ONM to release nucleocapsids into the cytoplasm (deenvelopment) (13, 14). This vesicle-mediated nucleocytoplasmic transport is primarily mediated by two viral proteins, UL31 and UL34, in HSV-1; both are thought to be conserved in all members of the Herpesviridae family (13–18). UL31, a nucleophosphoprotein, and UL34, a type II membrane protein, are recruited to the NM, where they form a heterodimeric complex designated the nuclear egress complex (NEC) (18–22). In primary envelopment, herpesvirus nucleocapsids need to circumvent the nuclear lamina to engage the INM, which then deforms to wrap around the nucleocapsids, and vesiculation is finalized by abscission of the INM (13, 14, 23). The HSV-1 NEC has been reported to play multiple roles in these primary envelopment steps, including deformation of the INM and recruitment of host cell factors, such as members of the protein kinase C family and components of the ESCRT-III machinery, that are thought to dissolve the nuclear lamina by phosphorylation of the lamin proteins and to mediate abscission of the INM, respectively (15, 16, 23–25).
For HSV-1 nucleocapsid recruitment to budding sites at the INM for primary envelopment, it has been suggested that UL31 binds to nucleocapsids in the nucleoplasm and recruits them to the INM (26). Of particular interest, HSV-1 C capsids are preferentially enveloped at the INM compared to A and B capsids (13). It has been reported that capsid association of HSV-1 UL31 required UL25 but not UL17 (27, 28), although both UL25 and UL17 are components of the CVSCs and are enriched on C capsids, as described above. It also has been shown that UL31 associates with UL25 and UL17 in HSV-1-infected cells (27, 28) and that the NEC contacts the nucleocapsid, predominantly via the CVSCs in HSV-1 primary enveloped virions (29). Based on these observations, it has been proposed that UL31 (i) binds to capsids via UL25 in the CVSCs, which are already on capsids or subsequently associate with capsids, (ii) recruits the capsids to the INM, and (iii) eventually forms a complex with UL34 at the INM to initiate nucleocapsid budding, leading to selective primary envelopment of mature C capsids (29, 30). However, although UL31 association with capsids and UL25 has been reported as described above (27, 28), the interaction of the NEC with nucleocapsids and UL25 has not been reported thus far. In addition, there is a lack of information on the significance of the interactions between NECs and nucleocapsids and between NECs and UL25 in nucleocapsid incorporation into primary enveloped virions.
The solved structure of the NECs of HSV-1 and of a porcine alphaherpesvirus, pseudorabiesvirus (PRV), predicted the surface patches at the membrane-distal ends of the NECs, located in helix 9 and helix 10 of HSV-1 UL31 and the PRV UL31 homolog, respectively, as potential capsid binding sites (22) (Fig. 1A). The amino acid sequence of HSV-1 UL31 helix 9 is well conserved in the PRV UL31 homolog and in other alphaherpesvirus UL31 homologs (Fig. 1B). In agreement with the prediction based on the solved structure, it has been reported that an amino acid substitution for lysine at residue 242 (K242) in helix 10 of the PRV UL31 homolog, which corresponds to the K279 residue in HSV-1 UL31 helix 9, induced the accumulation of empty vesicles resembling primary envelopes in the perinuclear space (31). These observations suggested that the potential capsid binding site in the PRV NEC (residue K242 in the PRV UL31 homolog) was required for capsid incorporation into vesicles during primary envelopment. However, in that study, it was not determined whether the mutation in PRV UL31 K242 impaired the interaction between the PRV NEC and nucleocapsids. Therefore, it remains to be shown whether binding of the PRV NEC to a nucleocapsid is required for nucleocapsid incorporation into vesicles during primary envelopment. Furthermore, in other herpesviruses, including HSV-1, the site in NECs responsible for nucleocapsid incorporation has not been identified.
Location of potential capsid binding sites in HSV-1 UL31 and comparison of the amino acid sequences of HSV-1 UL31 helix 9 and the corresponding domains of other alphaherpesvirus UL31 homologs. (A) In the HSV-1 NEC structure (22), UL31 is shown in light blue, and UL34 is shown in pale green. Helix 9 in UL31 is shown in red. The boxed area is enlarged to show the side chains of UL31 helix 9, with the amino acids in green (D275), red (R281 and D282), and blue (K279). Molecular graphics and analyses were performed with the PyMOL molecular graphics system, version 2.0.6 (Schrödinger, LLC). (B) Alignment of the amino acid sequences in HSV-1 UL31 helix 9 and the corresponding domains in alphaherpesvirus UL31 homologs, i.e., HSV-1 (GenBank accession no. CAA32324); herpes simplex virus 2 (HSV-2; GenBank accession no. CAB06756); VZV, varicella-zoster virus (NCBI RefSeq accession no. NP_040150); PrV (GenBank accession no. AFI70796); bovine herpesvirus 1 (BHV-1; NCBI RefSeq accession no. NP_045327); equine herpesvirus 1 (EHV-1; GenBank accession no. AAT67286); and gallid herpesvirus 2 (GaHV-2; GenBank accession no. AAF66766). The mutations investigated in this study are in red (HSV-1 residues R281 and D282) and blue (HSV-1 residue K279).
Therefore, to elucidate the mechanism for nucleocapsid incorporation into vesicles during herpesvirus primary envelopment in this study, we first tried to examine the effect of an amino acid substitution at residue K279 in HSV-1 UL31 helix 9 on the interaction of NECs with nucleocapsids. However, we found that the mutation in HSV-1 UL31 K279 abrogated the interaction of UL31 with UL34 in our assay system. Therefore, we investigated the effects of other amino acid substitutions in HSV-1 UL31 helix 9 that had no effect on the interaction of UL31 with UL34 and impaired the interaction of NECs with nucleocapsids on viral primary envelopment and replication.
RESULTS
HSV-1 NEC binding to capsid proteins and nucleocapsids in vitro.To examine the linkage between NEC binding to nucleocapsids and nucleocapsid incorporation into vesicles during HSV-1 primary envelopment, we needed to establish an assay system to measure binding of the HSV-1 NEC and its mutants to nucleocapsids and capsid proteins. For this assay system, we constructed a truncated UL31 fused to a polyhistidine tag at its carboxyl terminus (UL31Δ50-His) and a truncated UL34 fused to glutathione S-transferase (GST) at its amino terminus (GST-UL341–185) (Fig. 2A). These were coexpressed in Escherichia coli, and GST-UL341–185 was purified with glutathione-Sepharose beads, as described previously (19). UL31Δ50-His encoded UL31 amino acids 51 to 306, which included the amino acids in helix 9 and contained the potential capsid binding sites, and GST-UL341–185 encoded UL34 amino acids 1 to 185 (Fig. 2A). As previously reported (19), UL31Δ50-His copurified with GST-UL341–185 (Fig. 2B), indicating that these proteins formed a stable complex. The purified complex was designated GST-NEC185-Δ50.
Purification of recombinant NECs. (A) Schematic diagrams of the wild-type and recombinant HSV-1 UL31 and UL34 viral proteins used in this study. Line 1, wild-type HSV-1 NEC UL31 and UL34; line 2, GST-NEC185-Δ50 fusion proteins; lines 3 and 4, GST-NEC185-Δ50 fusion proteins carrying a single substitution mutation in residue K279 (line 3) or a double mutation in residues R281 and D282 (line 4) of UL31Δ50. (B) GST, GST-NEC185-Δ50, and the two GST-NEC185-Δ50 mutants were expressed in E. coli, lysed, and precipitated using glutathione-Sepharose beads. The lysates and beads were analyzed by electrophoresis in a denaturing gel and then immunoblotted with anti-UL31 and anti-UL34 antisera or stained with CBB.
The purified GST-NEC185-Δ50 was then used for three series of GST pulldown experiments. In the first series of experiments, purified GST-NEC185-Δ50 immobilized on glutathione-Sepharose beads was reacted with lysates of 293FT cells ectopically expressing either Flag-tagged UL25 (Flag-UL25), Flag-tagged VP23 (Flag-VP23), Flag-tagged UL17 (Flag-UL17), or Flag-tagged VP5 (Flag-VP5). As shown in Fig. 3, GST-NEC185-Δ50 pulled down Flag-UL25 from lysates of cells expressing Flag-UL25, although GST alone did not. In contrast, GST-NEC185-Δ50 and GST alone did not pull down Flag-VP23, Flag-UL17, or Flag-VP5 from lysates of cells expressing each of these capsid proteins (Fig. 3). These results indicated that this assay system was able to detect the binding of GST-NEC185-Δ50 to a specific capsid protein and that the NEC specifically interacted with UL25 in vitro.
Effect of the mutations in UL31 R281/D282 on binding of recombinant NEC to capsid proteins from cells expressing each capsid protein. (A to D) 293FT cells were transfected with plasmids expressing either Flag-UL25 (A), Flag-VP23 (B), Flag-UL17 (C), or Flag-VP5 (D) for 24 h. These cells then were lysed and reacted with GST, GST-NEC185-Δ50, or GST-NEC185-Δ50R281A/D282A31 that was immobilized on glutathione-Sepharose beads for 1 h at 4°C. After extensive washing, the beads were divided into two parts. One part was analyzed by electrophoresis in a denaturing gel and immunoblotted with anti-Flag antibody (top gels), and the other part was analyzed by electrophoresis in a denaturing gel and stained with CBB (bottom gels).
In the second series of experiments, purified GST-NEC185-Δ50 immobilized on glutathione-Sepharose beads was reacted with lysates of Vero cells infected with a recombinant HSV-1 YK497 (UL17-Myc/Flag-UL25) (32) expressing Myc-tagged UL17 (UL17-Myc) and Flag-UL25 (Fig. 4). As shown in Fig. 5, GST-NEC185-Δ50 pulled down all the capsid proteins tested (i.e., VP5, UL17-Myc, Flag-UL25, and VP23) from lysates of Vero cells infected with YK497 (UL17-Myc/Flag-UL25), although GST alone did not. These results, together with those in the first series of experiments, showing that NEC interacted with UL25 but not with UL17, VP5, or VP23 in the absence of other viral proteins, suggested that GST-NEC185-Δ50 bound to UL25 in capsids pulled down VP5, UL17-Myc, and VP23 in capsids from lysates of Vero cells infected with YK497 (UL17-Myc/Flag-UL25).
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 UL30 gene to the UL34 gene; line 3, domains of the UL31 gene and the UL34 gene; lines 4 to 6, recombinant viruses with mutations in UL31; line 7, recombinant virus encoding Strep-tagged UL34, lines 8 and 9; recombinant viruses encoding Strep-tagged UL34 and carrying mutations in UL31; line 10, domains of the UL17 gene and the UL25 gene; line 11, recombinant virus encoding Myc-tagged UL17 and Flag-tagged UL25; and line 12, recombinant virus in which A106 in UL25 was substituted with a stop codon.
Effect of the UL31 R281/D282 mutations on recombinant NEC binding to capsid proteins from HSV-1-infected cells. The GST fusion proteins shown in Fig. 2 were immobilized on glutathione-Sepharose beads and reacted with lysates of Vero cells that had been infected with YK497 (UL17-Myc/Flag-UL25) at an MOI of 5 for 18 h. The beads were washed extensively and divided into two parts. One part was analyzed by electrophoresis in a denaturing gel and immunoblotted with anti-VP5, anti-Myc, anti-Flag, and anti-VP23 antibodies (top gels), and the other part was analyzed by electrophoresis in a denaturing gel and stained with CBB (bottom gel).
In the third series of experiments, intranuclear capsids were isolated from lysates of nuclear fractions from Vero cells infected with YK497 (UL17-Myc/Flag-UL25) and purified on sucrose gradients (Fig. 6A and B). The gradients were fractionated, and fractions containing C capsids were obtained. As shown in Fig. 6B, ICP8, an HSV-1 nuclear protein that is not incorporated into virions, was easily detectable in lysates of the nuclear fractions from YK497 (UL17-Myc/Flag-UL25)-infected cells. However, when lysates were separated by sucrose gradient analysis, ICP8 was not found in the gradient fractions containing C capsids. In contrast, all of the other capsid proteins tested (i.e., VP5, UL17-Myc, Flag-UL25, and VP23) were detected both in the lysates of the nuclear fractions of infected cells and in the sucrose gradient fractions containing C capsids (Fig. 6B). Purified GST-NEC185-Δ50 immobilized on glutathione-Sepharose beads was then reacted with the fractions containing C capsids. In agreement with the results described in Fig. 5, GST-NEC185-Δ50 pulled down all of the capsid proteins tested (i.e., VP5, UL17-Myc, Flag-UL25, and VP23), but GST alone did not (Fig. 6C). Furthermore, to examine whether an anti-Flag antibody that can recognize Flag-UL25 blocked binding of GST-NEC185-Δ50 to nucleocapsids purified GST-NEC185-Δ50 immobilized on glutathione-Sepharose beads was reacted with the fractions containing C capsids from wild-type HSV-1(F)- or YK497 (UL17-Myc/Flag-UL25)-infected cells in the presence of anti-Flag monoclonal antibody or an IgG isotype control. As shown in Fig. 7, whereas GST-NEC185-Δ50 pulled down all the YK497 (UL17-Myc/Flag-UL25) capsid proteins tested (i.e., VP5 and VP23) in the presence of the IgG isotype control, these capsid proteins pulled down by GST-NEC185-Δ50 in the presence of anti-Flag antibody were barely detectable. In contrast, GST-NEC185-Δ50 was able to pull down VP5 and VP23 of wild-type HSV-1(F) in the presence of anti-Flag antibody as efficiently as it did in the presence of the IgG isotype control. These results indicated that this assay system was able to detect GST-NEC185-Δ50 binding to nucleocapsids and were in agreement with the possibility noted above, that GST-NEC185-Δ50 pulled down nucleocapsids by binding to UL25 in the nucleocapsids.
Effect of the UL31 R281 and D282 mutations on recombinant NEC binding to nucleocapsids. (A) Vero cells were infected with YK497 (UL17-Myc/Flag-UL25) at an MOI of 3 and harvested at 18 h postinfection. Nuclear lysates were isolated and layered onto 20% to 50% sucrose gradients and ultracentrifuged. The positions of type A, B, and C capsid bands in the sucrose gradient are indicated. (B) Proteins in the gradient fractions, shown in panel A, containing type A, B, or C capsids were analyzed by immunoblotting with the indicated antibodies. (C) Fractions containing C capsids were reacted with GST, GST-NEC185-Δ50, or GST-NEC185-Δ50R281A/D282A31 immobilized on glutathione-Sepharose beads for 1 h at 4°C. Beads were then extensively washed and divided into two parts. One part was analyzed by electrophoresis in a denaturing gel and immunoblotted with anti-VP5, anti-Myc, anti-Flag, and anti-VP23 antibodies (top gels), and the other was analyzed by electrophoresis in a denaturing gel and stained with CBB (bottom gel).
Blocking effect of anti-Flag antibody on recombinant NEC binding to nucleocapsids of the recombinant virus encoding Flag-tagged UL25. (A) Fractions containing C capsids prepared as described in Fig. 6A and B were reacted with GST or GST-NEC185-Δ50 immobilized on glutathione-Sepharose beads in the presence of anti-Flag antibody or an IgG isotype control and analyzed as described Fig. 6C.
Effect of mutations in potential HSV-1 NEC capsid binding sites on NEC binding to UL25 and nucleocapsids.The surface patches at the membrane-distal end of the HSV-1 NEC (Fig. 1) are located in HSV-1 UL31 helix 9 and may be NEC capsid binding sites (22). Therefore, to investigate NEC binding to nucleocapsids and UL25, we first focused on residue K279 in HSV-1 UL31 helix 9. As described above, HSV-1 UL31 K279 corresponds to residue K242 in the PRV UL31 homolog (Fig. 1B). An amino acid substitution at this PRV residue significantly impaired capsid incorporation into primary enveloped virions in PRV-infected cells, although the effect of this mutation on NEC binding to capsids was not examined (31). As shown in Fig. 2B, UL31Δ50-His, which carried an alanine substitution in UL31 K279 (UL31Δ50-His-K279A) (Fig. 2A), and GST-UL341–185 were coexpressed in bacteria at levels comparable to those of UL31Δ50-His and GST-UL341–185. However, unlike UL31Δ50-His, UL31Δ50-His-K279A was not copurified with GST-NEC185-Δ50 (Fig. 2B). These results indicated that the K279A mutation in UL31 abrogated NEC formation in this assay system and, therefore, we could not study the role of HSV-1 UL31 K279 in NEC binding to capsids and UL25.
It has been suggested that binding of the HSV-1 and PRV NECs to nucleocapsids is mediated by electrostatic interactions (20). In view of this possibility, we noted that there were three charged amino acids (D275, R281, and D282) in addition to K279 in HSV-1 UL31 helix 9 (Fig. 1B). Therefore, we focused on UL31 R281 and D282, because the side chains of these amino acids protrude from the NEC in a direction opposite that of the side chains of K279 and D275 (Fig. 1A) (22). As shown in Fig. 2B, UL31Δ50-His carrying alanine substitutions at UL31 residues R281 and D282 (UL31Δ50-His-R281A/D282A) (Fig. 2A) and GST-UL341–185 were coexpressed in E. coli at levels similar to those of UL31Δ50-His and GST-UL341–185. In addition, UL31Δ50-His-R281/D282 copurified with GST-UL341–185 as efficiently as did UL31Δ50-His (Fig. 2B). These results indicated that the R281A/D282A double mutation in UL31 had no obvious effect on NEC formation in this assay system; this NEC double mutant was designated GST-NEC185-Δ50-R281A/D282A31.
We then examined the effect of the R281A/D282A mutations in GST-NEC185-Δ50 on its binding to UL25 and C capsids in GST pulldown assays. As shown in Fig. 3A, the amount of Flag-UL25 that was pulled down by GST-NEC185-Δ50-R281A/D282A31 from lysates of 293FT cells expressing Flag-UL25 was smaller than that of Flag-UL25 pulled down by GST-NEC185-Δ50. However, neither GST-NEC185-Δ50 nor GST-NEC185-Δ50-R281A/D282A31 was able to pull down Flag-VP23, Flag-UL17, or Flag-VP5 from lysates of 293FT cells expressing each of the proteins (Fig. 3). GST-NEC185-Δ50-R281A/D282A31 pulled down smaller amounts of VP5, UL17-Myc, Flag-UL25, and VP23 from lysates of YK497 (UL17-Myc/Flag-UL25)-infected Vero cells than did GST-NEC185-Δ50 (Fig. 5). In agreement with these results (Fig. 5), GST-NEC185-Δ50-R281A/D282A31 also pulled down VP5, UL17-Myc, Flag-UL25, and VP23 from sucrose gradient fractions containing C capsids at lower levels than did GST-NEC185-Δ50 (Fig. 6). These results indicated that UL31 residues R281 and D282 were required for efficient binding of NEC to UL25 and nucleocapsids.
Construction and characterization of recombinant viruses carrying mutations in UL31 R281 and D282.To investigate the role of UL31 residues R281 and D282 in HSV-1-infected cells, we constructed and characterized recombinant virus HSV-1 YK731 (UL31-R281A/D282A) carrying the R281A/D282A mutations in UL31, its repaired virus YK732 (UL31-R281A/D282A-repair), YK735 (Strep-UL34) encoding UL34 fused to a Strep tag (Strep-UL34), YK736 (Strep-UL34/UL31-R281A/D282A) encoding Strep-UL34 and carrying the R281A/D282A mutations in UL31, and YK737 (Strep-UL34/UL31-R281A/D282A-repair) in which the UL31 R281A/D282A mutations in YK736 (Strep-UL34/UL31-R281A/D282A) were repaired (Fig. 4). As shown in Fig. 8A, Vero cells infected with YK731 (UL31-R281A/D282A) at a multiplicity of infection (MOI) of 5 for 18 h accumulated the UL31 protein and viral proteins UL34 and HSV-1 dUTPase (vdUTPase) at levels similar to those in cells infected with wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair). As expected, Vero cells infected with YK735 (Strep-UL34), YK736 (Strep-UL34/UL31-R281A/D282A), or YK737 (Strep-UL34/UL31-R281A/D282A-repair) expressed tagged UL34, but cells infected with wild-type HSV-1(F) did not (Fig. 8B). In agreement with the results in Fig. 8A described above, Vero cells infected with YK736 (Strep-UL34/UL31-R281A/D282A) at an MOI of 5 for 18 h accumulated the UL31 protein and viral proteins UL34 and HSV-1 dUTPase (vdUTPase) at levels similar to those in cells infected with wild-type HSV-1(F), YK735 (Strep-UL34), or YK737 (Strep-UL34/UL31-R281A/D282A-repair) (Fig. 8B).
Effects of the UL31 R281/D282 mutations and/or the tagging of UL34 with Strep tag on accumulation of viral proteins in HSV-1-infected cells. (A and B) Vero cells were mock infected or infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) (A), or mock-infected or infected with wild-type HSV-1(F), YK735 (Strep-UL34), YK736 (Strep-UL34/UL31-R281A/D282A), or YK737 (Strep-UL34/UL31-R281A/D282A-repair) (B), at an MOI of 5 for 18 h. These cells then were analyzed by immunoblotting with the indicated antibodies.
To examine the interaction between NEC and capsids and the effect of UL31 residues R281 and D282 on the interaction in HSV-1-infected cells, Vero cells infected with YK735 (Strep-UL34), YK736 (Strep-UL34/UL31-R281A/D282A), or YK737 (Strep-UL34/UL31-R281A/D282A-repair) were lysed, precipitated with Strep-Tactin beads, and analyzed by immunoblotting. As shown in Fig. 9, not only UL31 but also VP5 and VP23 were coprecipitated with Strep-UL34 from lysates of cells infected with YK735 (Strep-UL34) or YK737 (Strep-UL34/UL31-R281A/D282A-repair). In contrast, the amounts of VP5 and VP23 that were pulled down by Strep-UL34 from the lysates of cells infected with YK736 (Strep-UL34/UL31-R281A/D282A) were smaller than those of VP5 and VP23 pulled down by Strep-UL34 from lysates of cells infected with YK735 (Strep-UL34) or YK737 (Strep-UL34/UL31-R281A/D282A-repair) (Fig. 9). UL31 was pulled down by Strep-UL34 from lysates of cells infected with YK735 (Strep-UL34) or YK737 (Strep-UL34/UL31-R281A/D282A-repair) at levels comparable to those of UL31-R282A/D282A pulled down by Strep-UL34 from lysates of cells infected with YK736 (Strep-UL34/UL31-R281A/D282A) (Fig. 9). These results were in agreement with those obtained with the GST pulldown experiments using HSV-1-infected cell lysates and fractions containing C capsids described above (Fig. 5 and 6) and suggested that UL31 residues R281 and D282 were required for efficient interaction of NEC with capsids in HSV-1-infected cells.
Effect of the UL31 R281/D282 mutations on UL31 interaction with UL34 and capsid proteins in HSV-1-infected cells. Vero cells infected with YK735 (Strep-UL34), YK736 (Strep-UL34/UL31-R281A/D282A), or YK727 (Strep-UL34/UL31-R281A/D282A-repair) at an MOI of 5 for 18 h were lysed, precipitated with Strep-Tactin Sepharose beads, and analyzed by immunoblotting with the indicated antibodies.
We next examined the localization of UL31 and UL34 in Vero cells infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) by confocal microscopy. As shown in Fig. 10, in Vero cells infected with YK731 (UL31-R281A/D282A) at an MOI of 5 for 18 h, UL31 and UL34 appeared to be smoothly distributed and colocalized around the nuclear rim. These localization patterns were similar to those in cells infected with wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair) (Fig. 10). However, when these infected cells were examined with the Airyscan system (Zeiss), a superresolution confocal imaging system, the pattern of UL34 localization in cells infected with YK731 (UL31-R281A/D282A) was different from that in cells infected with wild-type HSV-1(F) or YK734 (UL31-R281A/D282A-repair) (Fig. 11). In cells infected with YK731 (UL31-R281A/D282A), UL34 colocalized with lamin A/C along the nuclear rim, but there was a small separation between UL34 and lamin A/C, as the fine punctate dots of UL34 were detectable outside lamin A/C along the nuclear rim (Fig. 11). In contrast, UL34 colocalized almost completely with lamin A/C along the nuclear rim in cells infected with wild-type HSV-1(F) or YK734 (UL31-R281A/D282A-repair) (Fig. 11). The Airyscan superresolution confocal imaging system also showed good colocalization of UL31 and UL34 in Vero cells infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) (Fig. 12). Taken together, these results indicated that the R281A/D282A mutations in UL31 had no obvious effect on the accumulation of NEC components and vdUTPase, and that UL31 residues R281 and D282 were required for the proper localization of NEC components in HSV-1-infected cells.
Effect of the UL31 R281/D282 mutations on localization of UL31 and UL34 in HSV-1-infected cells examined by confocal microscopy. Vero cells were infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) at an MOI of 5 for 18 h and then fixed, permeabilized, stained with anti-UL31 and anti-UL34 antibodies, and examined by confocal microscopy. Scale bars = 5 μm.
Effect of the UL31 R281/D282 mutations on localization of lamin A/C and UL34 in HSV-1-infected cells examined by superresolution confocal microscopy. Vero cells were infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) at an MOI of 5 for 18 h and then fixed, permeabilized, stained with anti-lamin A/C and anti-UL34 antibodies, and analyzed with the Airyscan system. Each image in the lower panels is the magnified image of the boxed area in the image in the upper panels. Fluorescence line scans along the dotted lines of the Airyscan images are shown on the right of each image. Scale bars = 2 μm.
Effect of the UL31 R281/D282 mutations on localization of UL31 and UL34 in HSV-1-infected cells examined by superresolution confocal microscopy. Vero cells were infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) at an MOI of 5 for 18 h and then fixed, permeabilized, stained with anti-UL31 and anti-UL34 antibodies, and analyzed with the Airyscan system. Each image in the lower panels is the magnified image of the boxed area in the image in the upper panels. Fluorescence line scans along the dotted lines of the Airyscan images are shown on the right of each image. Scale bars = 2 μm.
We then examined progeny virus yields at various times postinfection in Vero cells infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), YK732 (UL31-R281A/D282A-repair), YK720 (ΔUL31), YK735 (Strep-UL34), YK736 (Strep-UL34/UL31-R281A/D282A), or YK737 (Strep-UL34/UL31-R281A/D282A-repair) at an MOI of 5 or 0.01. As shown in Fig. 13, the progeny virus yield of YK731 (UL31-R281A/D282A) was less than those of wild-type HSV-1(F) and YK732 (UL31-R281A/D282A-repair) but was more than that of YK720 (ΔUL31). In particular, the progeny virus yield of YK731 (UL31-R281A/D282A) in Vero cells infected at an MOI of 5 (Fig. 13A) at 24 h postinfection and at an MOI of 0.01 (Fig. 13B) at 48 h postinfection were significantly less than those of wild-type HSV-1(F) and YK731 (UL31-R281A/D282A-repair) (11.6- and 12.9-fold at 24 h postinfection, and 14.7- and 20.0-fold at 48 h postinfection, respectively). In contrast, the progeny virus yield of YK731 (UL31-R281A/D282A) in Vero cells infected at an MOI of 5 at 24 h postinfection and at an MOI of 0.01 at 48 h postinfection were significantly higher than those of YK720 (ΔUL31) (200-fold at 24 h postinfection and 85.7-fold at 48 h postinfection, respectively) (Fig. 13A and B). In agreement with these results, the progeny virus yields of YK736 (Strep-UL34/UL31-R281A/D282A) in Vero cells infected at an MOI of 5 (Fig. 13C) at 24 h postinfection and at an MOI of 0.01 (Fig. 13D) at 48 h postinfection were significantly less than those of YK735 (Strep-UL34) and YK737 (Strep-UL34/UL31-R281A/D282A-repair) (9.71- and 7.98-fold at 24 h postinfection, and 27.2- and 42.7-fold at 48 h postinfection, respectively). Notably, the growth properties of YK735 (Strep-UL34) were almost identical to those of wild-type HSV-1(F), pointing out that the tagging of UL34 with the Strep epitope had no effect on viral growth in Vero cells (Fig. 13C and D). These results indicated that UL31 residues R281 and D282 were required for efficient HSV-1 replication in cell cultures, but the effect of the R281A/D282A mutations in UL31 on HSV-1 replication was significantly lower than that of the UL31-null mutation.
Effect of the UL31 R281/D282 mutations and/or the tagging of UL34 with Strep tag on HSV-1 growth. (A and B) Vero cells were infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), YK732 (UL31-R281A/D282A-repair), or YK720 (ΔUL31) at an MOI of 5 (A) or 0.01 (B). The infected cells were harvested at the indicated times postinfection, and progeny viruses were assayed on UL31-CV1 cells. (C and D) Vero cells were infected with YK735 (Step-UL34), YK736 (Strep-UL34/UL31-R281A/D282A), or YK737 (Strep-UL34/UL31-R281A/D282A-repair) at an MOI of 5 (C) or 0.01 (D). 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 the results from 5 independent experiments. Statistical analysis was performed by the Student’s t test, and P values of <0.0083 (0.05/6), <0.001 (0.05/5), and <0.00125 (0.05/4) were considered significant after Holm’s sequentially rejective Bonferroni multiple-comparison adjustment. The asterisks indicate statistically significant differences between YK731 (UL31-R281A/D282A) and HSV-1(F), YK731 (UL31-R281A/D282A) and YK732 (UL31-R281A/D282A-repair), YK731 (UL31-R281A/D282A) and YK720 (ΔUL31), YK736 (Strep-UL34/UL31-R281A/D282A) and HSV-1(F), YK736 (Strep-UL34/UL31-R281A/D282A), and YK735 (Strep-UL34), and YK736 (Strep-UL34/UL31-R281A/D282A), and YK737 (Strep-UL34/UL31-R281A/D282A-repair).
Effect of the UL31 R281 and D282 mutations on HSV-1 virion morphogenesis.To determine the step(s) at which UL31 R281 and D282 act during HSV-1 replication, we analyzed viral morphogenesis by quantitating the number of virus particles at different morphogenetic stages by electron microscopy of Vero cells infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) at an MOI of 5 for 18 h. As shown in Table 1, in Vero cells infected with wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair), 10.4 and 9.7%, respectively, of the total number of virus particles were enveloped virions in the perinuclear space. In contrast, the fraction of total virus particles that were in the perinuclear space in cells infected with YK731 (UL31-R281A/D282A) decreased significantly to 4.1%; this represented 2.5- and 2.4-fold reductions compared to cells infected with wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair), respectively (Table 1). Although 55.8 and 56.1% of the total number of virus particles were capsids in the nuclei of cells infected with wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair), respectively, the fraction of total particles that were capsids in the nucleus of cells infected with YK731 (UL31-R281A/D282A) increased significantly to 80.5% (Table 1). The differences between the fractions of virions in the perinuclear space and nucleus of YK731 (UL31-R281A/D282A)-infected cells and the fractions in wild-type HSV-1(F)- or YK732 (UL31-R281A/D282A-repair)-infected cells were statistically significant (Table 1). These results indicated that the R281A/D282A mutations in UL31 resulted in a decrease in the fraction of virus particles that were primary enveloped virions in the perinuclear space and in an increase in the fraction of virus particles that were capsids in the nucleus. Since the UL31 R281A/D282A mutations produced an increase in the fraction of virus particles in capsids in the nuclei of YK731 (UL31-R281A/D282A)-infected cells, there was a decrease in virus particles in the cytoplasm and extracellular space in cells infected with YK731 (UL31-R281A/D282A), from 33.8 and 34.0% in wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair)-infected cells, respectively, to 15.5% in YK731 (UL31-R281A/D282A)-infected cells (Table 1).
Effect of the mutation in UL31 helix 9 on the distribution of virus particles in infected Vero cellsa
We noted that membranous structures containing empty vesicles, formed by evaginations of the ONM into the cytoplasm, were observed in most cells infected with YK731 (UL31-R281A/D282A) (Fig. 14). In addition, membranous structures containing empty vesicles, formed by invaginations of the INM into the nucleoplasm, were also observed in a small fraction of these infected cells (Fig. 14). These results were in agreement with those by Airyscan confocal microscopy, as described above (Fig. 11 and 12), that fine punctate dots of UL31 and UL34 were detected outside the lamin A/C along the nuclear rim. In addition, the empty vesicles in the membranous structures in cells infected with YK731 (UL31-R281A/D282A) were similar in size and morphology to the primary envelopes in the perinuclear space of cells infected with wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair) (Fig. 14). These empty vesicles in the perinuclear space were barely detectable in cells infected with wild-type HSV-1(F) or YK732 (UL31-R281A/D282A-repair), and the differences between the number of empty vesicles in the perinuclear space in YK731 (UL31-R281A/D282A)-infected cells and the numbers in wild-type HSV-1(F)- or YK732 (UL31-R281A/D282A-repair)-infected cells were statistically significant (Fig. 14 and 15). We also examined whether Vero cells infected with YK720 (ΔUL31) or a UL25-null mutant virus YK738 (ΔUL25), the latter of which was generated in this study (Fig. 4 and 16), produced empty vesicles in the perinuclear space like those infected with YK731 (UL31-R281A/D282A). Similarly to wild-type HSV-1(F)- and YK732 (UL31-R281A/D282A-repair)-infected cells (Fig. 14 and 15), YK720 (ΔUL31)- and YK738 (ΔUL25)-infected cells barely produced empty vesicles in the perinuclear space (Fig. 15 and 17).
Effect of the UL31 R281A/D282A mutations on HSV-1 nuclear egress. (A to C) Vero cells were infected with wild-type HSV-1(F) (A), YK732 (UL31-R281A/D282A-repair) (B), or YK731 (UL31-R281A/D282A) (C) at an MOI of 5. At 18 h postinfection, the cells were fixed, embedded, sectioned, stained, and examined by electron microscopy. (A and B) An enlarged image of an enveloped virion in the perinuclear space is shown in the upper right of each image. (C) Each image on the right is the magnified image of the boxed area on the left. Membranous structures containing empty vesicles formed by evaginations of ONM into the cytoplasm are shown in the top images. Membranous structures containing empty vesicles formed by invaginations of INM into the nucleoplasm are shown on the bottom. Arrowheads indicate empty particles in membranous structures. Nu, nucleus; NM, nuclear membrane; Cy, cytoplasm. Scale bars = 300 nm.
Quantification of empty particles in the perinuclear space of HSV-1-infected cells. (A) Vero cells were infected with wild-type HSV-1(F), YK731 (UL31-R281A/D282A), or YK732 (UL31-R281A/D282A-repair) at an MOI of 5 for 18 h. The infected cells then were examined by electron microscopy as described in Fig. 14, and the numbers of empty particles in the perinuclear spaces of 10 infected cells were quantitated. Statistical analysis was performed by the Student's t test, and P values of <0.0167 (0.05/3), <0.025 (0.05/2), and <0.05 (0.05/1) were sequentially considered significant after Holm’s sequentially rejective Bonferroni multiple-comparison adjustment. The horizontal bars indicate the means ± standard errors. The asterisk indicates a statistical difference between YK731 (UL31-R281A/D282A) and HSV-1(F), and between YK731 (UL31-R281A/D282A) and YK732 (UL31-R281A/D282A-repair). (B and C) Vero cells infected with wild-type HSV-1(F) or YK720 (ΔUL31) (B) or YK738 (ΔUL25) (C) at an MOI of 5 for 18 h. The infected cells were then examined by electron microscopy as described in Fig. 17, and the numbers of empty particles in the perinuclear spaces of 10 infected cells were quantitated.
Effect of the UL25-null mutation on accumulation of viral proteins in HSV-1-infected cells. Vero cells were mock infected or infected with wild-type HSV-1(F) or YK738 (ΔUL25) at an MOI of 5 for 18 h. These cells were then analyzed by immunoblotting with the indicated antibodies.
Effect of the UL31-null or UL25-null mutation on HSV-1 nuclear egress. (A and B) Vero cells were infected with wild-type HSV-1(F) (A) or YK720 (ΔUL31) (B) at an MOI of 5. At 18 h postinfection, the cells were fixed, embedded, sectioned, stained, and examined by electron microscopy. (C and D) Vero cells infected with wild-type HSV-1(F) (C) or YK738 (ΔUL25) (D) at an MOI of 5 for 18 h were analyzed as described in panels A and B. Nu, nucleus; NM, nuclear membrane; Cy, cytoplasm. Scale bars = 300 nm.
DISCUSSION
HSV-1 NEC binding to nucleocapsids has been proposed to promote nucleocapsid budding at the INM and subsequent incorporation of nucleocapsids into primary enveloped virions during nuclear egress (13, 14, 23). However, direct evidence to support this hypothesis has not been reported thus far, e.g., mutation(s) of a capsid binding site(s) in the NEC that impairs nucleocapsid incorporation into primary enveloped virions. To investigate this issue, a system to detect NEC binding to nucleocapsids and capsid proteins was required. Therefore, for this study, we established a system combining bacterial expression of recombinant NEC, previously used for characterization of HSV-1 NEC-mediated membrane deformation (19, 22), and GST pulldown assays. UL31 residues R281 and D282, which are in helix 9 in the UL31 structure (22), have been shown to be at the NEC membrane-distal face and are proposed to be putative NEC capsid binding sites (20, 22, 33). From the NEC structure (22) and with the NEC binding system in this study, we showed that alanine substitutions at UL31 residues R281 and D282 impaired NEC binding to nucleocapsids in vitro. These results suggested that, as predicted from the solved structure (22), surface patches, such as UL31 R281 and D282, at the membrane-distal face of NEC were a nucleocapsid binding site. This is the first identification of a site in a herpesvirus NEC shown to be required for NEC binding to nucleocapsids. In cells infected with a recombinant virus carrying substitution mutations in UL31 R281 and D282, empty vesicles accumulated in evaginations of the ONM and, to a lesser extent, in invaginations of the INM. The empty vesicles were similar in size and morphology to primary virus envelopes, which suggested that they were primary envelopes that did not contain capsids and that the mutations in UL31 R281 and D282 impaired the incorporation of capsids into vesicles during primary envelopment, as previously reported (31). These observations provided direct evidence supporting the model that NEC binding to nucleocapsids was required for efficient incorporation of nucleocapsids into vesicles during primary envelopment. In agreement with this conclusion, the mutations in UL31 R281 and D282 produced an increase in nuclear capsids and a decrease in primary enveloped virions in the perinuclear space, which are typical results of impairment of primary envelopment, as reported previously (34, 35).
Data from biochemical analyses of HSV-1 UL31 association with capsids and capsid proteins (27, 28) and of cryo-electron microscopy and tomography of HSV-1 primary enveloped virions (29) have suggested that UL25 may be a bridge between NECs and nucleocapsids. In support of this hypothesis, we have shown here that purified NECs specifically pulled down UL25 from lysates of cells ectopically expressing this capsid protein. The NECs also pulled down VP5, VP23, and UL17, in addition to UL25, from lysates of HSV-1-infected cells and from sucrose gradient fractions containing C capsids, but VP5, VP23, and UL17 were not pulled down by NECs from lysates of cells expressing each of these capsid proteins in the absence of UL25. These results suggested that only when NEC was bound to UL25 in nucleocapsids was the NEC able to pull down VP5, VP23, and UL17 from lysates of HSV-1-infected cells and from sucrose gradient fractions containing C capsids. In agreement with this, the anti-Flag antibody that can recognize Flag-UL25 of YK497 (UL17-Myc/Flag-UL25) specifically blocked the affinity precipitation of VP5 and VP23 in C capsids of YK497 (UL17-Myc/Flag-UL25) by the NECs. In addition, the R281A/D282A mutations in UL31 reduced the amount of UL25 pulled down by the NECs from lysates of cells expressing UL25 and reduced the amounts of UL25, VP5, VP23, and UL17 pulled down by the NECs from lysates of HSV-1-infected cells and from sucrose gradient fractions containing C capsids. Collectively, these results suggested that the NEC interacted with UL25 in the nucleocapsid via UL31 residues R281 and D282 in the NEC membrane-distal face. This supported the model that UL25 in nucleocapsids was a bridge between the NEC and nucleocapsids, leading to selective envelopment of C capsids in primary envelopment.
The R281A/D282A mutations in UL31 may not perturb the interaction of NECs with nucleocapsids and UL25, but instead, it may cause misfolding of UL31 to negatively affect NEC function. Although we cannot completely eliminate this possibility, it seems unlikely because UL31 residues R281 and D282 are solvent-exposed residues, based on the reported structure of the HSV-1 NEC, which should minimize misfolding of UL31 with the R281A/D282A mutations (22). In addition, NEC with UL31 carrying the R281A/D282A mutations appeared to retain many of the properties of wild-type NEC in HSV-1-infected cells. Thus, we have shown that the R281A/D282A mutations in UL31 had no obvious effect on NEC complex formation in vitro and that NEC with UL31 carrying the R281A/D282A mutations was efficiently recruited to the NM in HSV-1-infected cells. Formation of the empty vesicles in the perinuclear space of cells infected with YK731 (UL31-R281A/D282A) suggested that, although nucleocapsid incorporation into vesicles during primary envelopment was impaired by the R281A/D282A mutations as described above, vesicle formation and scission took place during primary envelopment despite the mutations. However, although the incorporation of nucleocapsids into vesicles during primary envelopment is essential for HSV-1 replication, the R281A/D282A mutations in UL31 reduced HSV-1 replication only 11.6-fold. This may be due to incomplete abrogation of NEC binding to nucleocapsids due to the mutations. In agreement with this possibility, NECs carrying UL31 with the double mutation sometimes pulled down small amounts of UL25 from lysates of cells ectopically expressing UL25 alone or small amounts of UL25 and/or another capsid protein from lysates of HSV-1-infected cells in our assay systems. The reduction in progeny virus yield in cells infected by HSV-1 carrying UL31 with the R281A/D282A mutations was significantly less than in cells infected by HSV-1 carrying the UL31-null mutation. Alternatively, compensatory mutation(s) outside the UL31 gene might have a role(s) to play in diminishing the severity of the growth defect observed for the UL31 mutant HSV-1, as previously reported (36–39). We verified that sequences of the genes encoding UL31 and those encoding HSV-1 proteins interacting with UL31, including UL25, UL34, UL47, ICP22, and US3, in the YK731 (UL31-R281A/D282A) genome were identical to those in parental strain genome except the introduced mutation of interest (R281A/D282A in UL31). However, we could not completely eliminate the possibility that secondary mutation(s) outside these genes might have an effect on the phenotypes of YK731 (UL31-R281A/D282A) in infected cells.
A phenotype like that of YK731 (UL31-R281A/D282A), including growth defects and induction of membranous structures containing empty vesicles similar to primary envelopes, was also reported from complementation assays with an alanine substitution mutation in PRV UL31 K242, which corresponds to HSV-1 UL31 K279 (31). In this study, we showed that the K279A mutation in HSV-1 UL31 abrogated NEC formation in vitro, unlike the UL31 R281A/D282A mutations. These results suggested that, although the amino acid sequences at the NEC membrane-distal faces and their structures are well conserved between HSV-1 and PRV, the effects of amino acid substitutions in the conserved residues in these regions were different. In particular, the structures of the NEC domains close to the membrane-distal faces appear to be different in HSV-1 and PRV (22). Thus, the domain of the PRV NEC adjacent to the membrane-distal face forms an α-helix, but it has not yet been possible to solve the structure of the corresponding domain of the HSV-1 NEC (22). Structures that are difficult to solve often have flexible conformations (40, 41), and therefore, the structural dynamics of the HSV-1 and PRV domains adjacent to their membrane-distal faces may be different. This may account for the different effects of mutations in the conserved residues in the membrane-distal faces of HSV-1 and PRV on the functionality of their NECs. PRV UL31 carrying the K242A mutation may still be able to form a complex with UL34, but the resulting NEC may have an impaired ability to interact with nucleocapsids, similar to the effects of the R281A/D282A mutations in HSV-1 UL31. Therefore, it would be interesting to investigate whether the K242A mutation in PRV UL31 impairs the interaction of the PRV NEC with nucleocapsids but has no effect on NEC formation. If this is the case, those results would further support the conclusions of this study.
MATERIALS AND METHODS
Cells and viruses.Vero, rabbit skin, 293FT, Plat-GP, and UL31-CV1 cells were described previously (42–45). Wild-type HSV-1(F), wild-type YK312, YK497 (UL17-Myc/Flag-UL25), and UL31-null mutant virus YK720 (ΔUL31) (Fig. 4) were described previously (32, 34, 42, 46).
Plasmids.The VP5, VP23, UL17, and UL25 open reading frames (ORFs) were amplified by PCR from the YK497 genome and cloned into pcDNA3.1(+) (Invitrogen) in frame with the Flag epitope to generate pcDNA3.1-Flag-VP5, pcDNA3.1-Flag-VP23, pcDNA3.1-Flag-UL17, and pcDNA3.1-Flag-UL25, respectively. pMxs-UL25 was generated by cloning the UL25 ORF, which had been amplified by PCR from the HSV-1(F) genome, into pMxs-puro (47). The pGEX-UL341-185 plasmid encoding a fusion protein of glutathione S-transferase (GST) and UL341–185 (GST-UL341-185) was constructed by cloning PCR-amplified UL34 codons 1 to 185 into pGEX-4T-1 (GE Healthcare) in frame with GST (Fig. 2A). UL31 codons 51 to 306 were amplified by PCR from pcDNA-UL31 (48) and cloned into pET-24b (Novagen) in frame with a 6×His tag to generate pET-UL31Δ50 (Fig. 2A). UL31 codons 51 to 306 were amplified by PCR from the HSV-1 genome carrying the K279A mutation in UL31 and cloned into pET-24b in frame with a 6×His tag to generate pET-UL31Δ50-K279A (Fig. 2A). UL31 codons 51 to 306 were amplified by PCR from YK731 (R281A/D282A) and cloned into pET-24b in frame with a 6×His tag to generate pET-UL31Δ50-R281A/D282A (Fig. 2A).
Establishment of Vero cells stably expressing UL25.Plat-GP cells were cotransfected with pMDG (49) and pMxs-UL25, and the supernatants were harvested. Vero cells were transduced with the supernatants and selected with puromycin (5 μg/ml) as described previously (49). Resistant cells were cloned from a single colony and designated UL25-Vero cells.
Mutagenesis of viral genomes and generation of recombinant HSV-1.Recombinant virus YK731 (UL31-R281A/D282A) carrying the R281A/D282A mutations in UL31 (Fig. 4) was generated by the two-step Red-mediated mutagenesis procedure using Escherichia coli GS1783 containing pYEbac102Cre (46, 50, 51) and primers 5′-CCGACCGTGTCGGCCGCAGACATTTATTGTAAAATGGCGGCCATCAGCTTCGACGGGGGGCAGGATGACGACGATAAGTAGGG-3′ and 5′-CCTTTGATACTCTAGCATGAGCCCCCCGTCGAAGCTGATGGCCGCCATTTTACAATAAATCAACCAATTAACCAATTCTGATTAG-3′. Recombinant virus YK732, in which the UL31-R281A/D282A mutations in YK731 were repaired (Fig. 4), was generated as described above, except that primers 5′-CCGACCGTGTCGGCCGCAGACATTTATTGTAAAATGAGGGACATCAGCTTCGACGGGGGGCAGGATGACGACGATAAGTAGGG-3′ and 5′-CCTTTGATACTCTAGCATGAGCCCCCCGTCGAAGCTGATGTCCCTCATTTTACAATAAATCAACCAATTAACCAATTCTGATTAG-3′ were used. To verify the sequences of the gene encoding UL31 and those encoding UL25, UL34, UL47, ICP22, and US3, all of which were reported to interact with UL31 (18, 28, 34, 35), Vero cells were infected with YK731 (UL31-R281A/D282A) or its parental virus YK312 (wild type) (46) at an MOI of 5, 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 (42). The sequences of these genes in the YK731 (UL31-R281A/D282A) genome were identical to those in the YK312 (parental strain) genome except for the substitution mutation of interest (R281A/D282A in UL31). Recombinant virus YK735 (Strep-UL34) encoding Strep-tagged UL34 in which the Strep epitope was inserted at the amino terminus of UL34 (Fig. 4) was generated as described above except using primers 5′-CTCCCATCGCGGGCGCCATGTGGAGCCATCCGCAGTTTGAAAAGGCGGGACTGGGCAAGCAGGATGACGACGATAAGTAGGG-3′ and 5′-GGTTTACGCGGGCACGCACGCTCCCATCGCGGGCGCCATGTGGAGCCATCCGCAGTTTGAAAAGGCGGGACTGGGCAAGCCCTAAGGATGACGACGATAAGTAGGG-3′. Recombinant virus YK736 (Strep-UL34/UL31-R281A/D282A), encoding Strep-tagged UL34 and carrying the R281A/D282A mutations in UL31 (Fig. 4), was generated as described above, except using Escherichia coli GS1783 carrying the YK731 (UL31-R281A/D282A) genome. Recombinant virus YK737 (Strep-UL34/UL31-R281A/D282A-repair), in which the UL31-R281A/D282A mutations in YK735 (Strep-UL34/UL31-R281A/D282A) were repaired (Fig. 4), were generated as described above, except using Escherichia coli GS1783 carrying the YK736 (Strep-UL34/UL31-R281A/D282A) genome. Recombinant virus YK738 (ΔUL25), in which UL25 codon 106 was substituted with stop codon, was generated as described above except using primers 5′-TCGCAGGCGCCCTGGAGGCGCTGGAGACGGCGGCCTAGCGCCGAAGAGGCgGATGCCAGGATGACGACGATAAGTAGGG-3′ and 5′-CGCCGGCTCATCCCCGCGCGCGGCATCCGCTAGCTCTTCGGCGGCCGCCGTCTCCAGCGCCTCCAACCAATTAACCAATTCTGATTAG-3′ and UL25-Vero cells. The HSV-1 genome carrying the K279A mutation in UL31 was generated as described above except using primers 5′-CGCGGAGATTCCGACCGTGTCGGCCGCAGACATTTATTGTGCAATGAGGGACATCAGCTTCGAGGATGACGACGATAAGTAGGG-3′ and 5′-CTAGCATGAGCCCCCCGTCGAAGCTGATGTCCCTCATTGCACAATAAATGTCTGCGGCCGCAACCAATTAACCAATTCTGATTAG-3′. The viruses used in this study, except YK720 (ΔUL31) and YK738 (ΔUL25), were propagated and titrated in Vero cells. In cases where YK720 (ΔUL31) was used, this recombinant virus as well as wild-type HSV-1(F), YK731 (UL31-R281A/D282A), and YK732 (UL31-R281A/D282A-repair) were propagated and titrated in UL31-CV1 cells. In cases where YK738 (ΔUL25) was used, this recombinant virus and wild-type HSV-1(F) were propagated and titrated in UL25-Vero cells.
Antibodies.Mouse monoclonal antibodies against α-tubulin (antibody DM1A; Sigma), Flag (M2; Sigma), Myc (PL14; MBL), VP5 (3B6; Virusys), and Strep tag (4F1, M211-3; MBL), mouse IgG1 isotype control (Sigma), and commercial rabbit polyclonal antibody against VP23 (CAC-CT-HSV-UL18; Cosmo Bio) were used in this study. Mouse polyclonal antibody against UL31 and UL25 and rabbit polyclonal antibody against UL31, UL34, vdUTPase, and VP16 were described previously (35, 52, 53). 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 (54, 55). For superresolution imaging, image acquisition was performed using an LSM800 microscope with Airyscan (Zeiss), and image reconstruction was carried out using the ZEN2.3 software (Zeiss).
Purification of fusion proteins.E. coli BL21 was transformed with pGEX-4T-1 or cotransformed with pGEX-UL341-185 and either pET-UL31Δ50, pET-UL31Δ50-K279A, or pET-UL31Δ50-R281A/D282A to express GST, GST-NEC185-Δ50, GST-NEC185-Δ50K279A31, or GST-NEC185-Δ50R281A/D282A31, respectively. After the transformed E. coli cultures reached an optical density at 600 nm of 0.6 to 0.8, the proteins were induced with 0.1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) for 16 h at 18°C. The cells were lysed by a brief sonication on ice in lysis buffer (50 mM HEPES [pH 7.0], 500 mM NaCl, 10 μM ZnSO4, 0.5% NP-40). After cell debris was removed by centrifugation (20,400 × g for 10 min at 4°C), the GST fusion proteins were captured on glutathione-Sepharose beads (GE Healthcare Bio-Sciences) and washed with lysis buffer three times, and the beads were divided into two parts. Both parts were analyzed by electrophoresis in a denaturing gel; one part then was immunoblotted, and the other was stained with Coomassie brilliant blue (CBB).
GST pulldown assays.GST, GST-NEC185-Δ50, and GST-NEC185-Δ50R281A/D282A31 were expressed and then purified on glutathione-Sepharose beads, as described above. 293FT cells were transfected with pcDNA3.1-Flag-VP5, pcDNA3.1-Flag-VP23, pcDNA3.1-Flag-UL17, or pcDNA3.1-Flag-UL25 using polyethylenimine (PEI) Max (PSI) for 24 h. Vero cells then were infected with YK497 (UL17-Myc/Flag-UL25) at an MOI of 5 for 18 h, lysed in lysis buffer containing a proteinase inhibitor cocktail (Nacalai Tesque), and incubated with GST proteins immobilized on glutathione-Sepharose beads for 1 h at 4°C. After extensive washing of the beads with lysis buffer, the proteins on the beads were analyzed by electrophoresis in a denaturing gel and then immunoblotted or stained with CBB.
Purification of nuclear capsids.Nuclear capsids were purified as described previously (32, 56, 57). Briefly, Vero cells were infected with wild-type HSV-1(F) or YK497 (UL17-Myc/Flag-UL25) at an MOI of 3 for 18 h. The cells then were harvested, resuspended in hypotonic buffer (10 mM Tris-HCl [pH 7.5], 10 mM KCl, 3 mM MgCl2, 0.05% NP-40, 1 mM EDTA, 1 mM dithiothreitol [DTT], 10 mM NaF, 1% protease inhibitor cocktail), and incubated on ice for 15 min. The nuclei were pelleted using low-speed centrifugation (250 × g for 8 min at 4°C), washed with the hypotonic buffer, resuspended in TNE buffer (500 mM NaCl, 1 mM EDTA, 20 mM Tris-HCl [pH 7.5]), and sonicated on ice three times for 15 s each. Cell debris was removed by centrifugation (2,200 × g for 5 min at 4°C), and nuclear capsids in the supernatants were pelleted through a 35% (wt/wt) sucrose cushion (in TNE buffer) in a P40ST rotor (Hitachi) by centrifugation at 110,000 × g for 1 h at 4°C. The pellets were resuspended in TNE buffer by sonication on ice three times for 15 s each, layered onto a 20% to 50% (wt/wt) sucrose gradient (in TNE buffer), and centrifuged at 110,000 × g for 1 h at 4°C in a P40ST rotor. After centrifugation, 0.3-ml fractions were collected from the top of each tube. Each fraction was divided into two parts. Trichloroacetic acid (TCA) was added to one part of each fraction to a final concentration of 10% TCA, and the fractions were incubated overnight at 4°C. The precipitated proteins were pelleted by centrifugation at 21,000 × g for 20 min at 4°C, washed two times with ethanol, dried, and analyzed by immunoblotting. The other part of each fraction, containing C capsids, was incubated with GST proteins immobilized on glutathione-Sepharose beads as described above for 1 h at 4°C. After extensive washing of the beads with wash buffer (20 mM Tris-HCl [pH 7.5], 500 mM NaCl, 10 μM ZnSO4), the beads were analyzed by electrophoresis in a denaturing gel and then immunoblotted or stained with CBB.
Analyses of blocking effects of antibodies on binding of GST-NEC185-Δ50 to nucleocapsids.The collected fractions containing C capsids were incubated with an IgG isotype control (10 μg/ml) or anti-Flag monoclonal antibody (10 μg/ml) for 1 h at 4°C, followed by incubation with GST proteins immobilized on glutathione-Sepharose beads for 1 h at 4°C. After extensive washing of the beads with wash buffer, the beads were analyzed by electrophoresis in a denaturing gel and then immunoblotted or stained with CBB as described above.
Affinity precipitation.Vero cells were infected with YK735 (Strep-UL34), YK736 (Strep-UL34/UL31-R281A/D282A), or YK727 (Strep-UL34/UL31-R281A/D282A-repair) at an MOI of 5 for 18 h and lysed with lysis buffer containing a protease inhibitor cocktail (Nacalai Tesque). After centrifugation, the supernatants were reacted with Strep-Tactin Sepharose beads (IBA) with rotation for 1 h at 4°C. The precipitates were collected by brief centrifugation, washed extensively with lysis buffer, and analyzed by immunoblotting.
Electron microscopy.Vero cells infected with HSV-1(F), YK731 (UL31-R281A/D282A), YK732 (UL31-R281A/D282A-repair), YK720 (ΔUL31), or YK738 (ΔUL25) at an MOI of 5 for 18 h were examined by ultrathin-section electron microscopy, as described previously (58). The number of virus particles at different morphogenetic stages was quantitated in a number of randomly chosen Vero cells. “Empty particles,” which were morphologically similar to enveloped virions but did not contain a capsid, in the perinuclear space were also quantitated.
Statistical analysis.Statistical analysis was performed by the Student’s t test. For the three-comparison analyses, P values of <0.0167 (0.05/3), <0.025 (0.05/2), and <0.05 (0.05/1) were sequentially considered significant after Holm’s sequentially rejective Bonferroni multiple-comparison adjustment. For the four-comparison analyses, P values of <0.00833 (0.05/6), <0.01 (0.05/5), and <0.0125 (0.05/4) were sequentially considered significant, as described above.
ACKNOWLEDGMENTS
We thank Risa Abe and Hiroshi Sagara for their excellent technical assistance.
This study was supported by grants for scientific research from the Japan Society for the Promotion of Science (JSPS), grants for scientific research on innovative areas from the Ministry of Education, Culture, Science, Sports and Technology of Japan (grants 16H06433, 16H06429, 16K21723, 19H05286, and 19H05417), contract research funds from the Program of the Japan Initiative for the Global Research Network on Infectious Diseases (J-GRID) (grant JP18fm0108006) and the Research Program on Emerging and Re-emerging Infectious Diseases (grant 19fk018105h0001) from the Japan Agency for Medical Research and Development (AMED), a grant from the Joint Research Project of the Institute of Medical Science, the University of Tokyo, and grants from the Takeda Science Foundation, GSK Japan Research Grant 2017, and the Uehara Memorial Foundation.
FOOTNOTES
- Received 5 August 2019.
- Accepted 6 August 2019.
- Accepted manuscript posted online 7 August 2019.
- Copyright © 2019 American Society for Microbiology.
