Previous Article | Next Article ![]()
Journal of Virology, January 2008, p. 795-804, Vol. 82, No. 2
0022-538X/08/$08.00+0 doi:10.1128/JVI.01722-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Tomohiko Sadaoka,1,
Hironori Yoshii,1,3
Pranee Somboonthum,1
Takayoshi Imazawa,4
Kazuhiro Nagaike,3
Keiichi Ozono,2
Koichi Yamanishi,1 and
Yasuko Mori1*
Laboratory of Virology and Vaccinology,1 Laboratory of Toxicogenomics, Division of Biomedical Research, National Institute of Biomedical Innovation, 7-6-8, Saito-Asagi, Ibaraki, Osaka 567-0085,4 Department of Pediatrics, Graduate School of Medicine, Osaka University, 2-2, Yamadaoka, Suita, Osaka 565-0871,2 Kanonji Institute, The Research Foundation for Microbial Diseases of Osaka University, Kannonji, Kagawa, Japan3
Received 7 August 2007/ Accepted 16 October 2007
|
|
|---|
|
|
|---|
The genome of herpesviruses encodes many glycoproteins, which are involved in various aspects of the viral life cycle, such as entry into the host cell (43) and the envelopment and maturation of progeny virions (29, 30), and are major immunological determinants for host defense mechanisms. VZV encodes at least seven proteins that are glycosylated—glycoproteins B, C, E, H, I, K, and L (gB, gC, gE, gH, gI, gK, and gL)—and contains the sequences for gM and gN, although it is not certain whether VZV gM and gN are glycosylated (7).
Although herpesviruses encode several glycoproteins, the genes for only five are conserved among all herpesviruses: gB, gH, gL, gM, and gN. The conservation of these glycoproteins in herpesvirus evolution suggests that they have important roles in virus life cycles. While gB and gH/gL are essential for viral replication in cell culture (43), disruption of gM in HSV-1 (2, 26), pseudorabies virus (PRV) (10), or equine herpesvirus 1 (EHV-1) (36) shows 1 to 10% of the wild-type growth. The gM genes encoded by human cytomegalovirus (HCMV) (16) and Marek's disease virus serotype 1 (48) are reported to be essential for viral replication. Therefore, the functional roles of the gM protein during viral replication may differ among herpesviruses. The gM homologs in PRV (17), bovine herpesvirus 1 (24), infectious laryngotracheitis virus (13), and HCMV (25) all form a disulfide-linked complex with a gN homolog.
Although the gM/gN complex is dispensable for alphaherpesvirus replication in cell culture, a null mutation of the gM gene or null mutations of both the gM and gN genes give rise to some interesting phenotypes. The deletion of gM in PRV (10), EHV-1 (36), or EHV-4 (51) impairs not only the cell-to-cell spread of infection but also the penetration of cell-free virus into cells. The gM of HSV-1 or PRV expressed in cell culture inhibits the cell-cell fusion induced by the herpesvirus glycoproteins gB, gD, and the gH-gL complex (8, 21) and by another viral protein, human respiratory syncytial virus F protein (8). The expression of gM causes other viral membrane proteins to relocate to the trans-Golgi network (TGN), suggesting that gM functions in the maturation of viral particles and the cell-to-cell spread of infection (8). The gM/gN complex is found in the viral particles of PRV (17) or HCMV (25). Recently, antibodies against the gM/gN complex were shown to have a neutralizing function in HCMV infection (42). These findings suggest that the gM/gN complex is involved in virus entry into host cells.
In contrast to the information available for other herpesviruses, the gN homolog encoded by VZV has only been reported to be nonessential for viral growth in cell culture (38), and the VZV homolog of gM has not been investigated at all. Therefore, the aim of this study was to identify and characterize the VZV gM homolog and to investigate its function in cell culture. For these analyses, we produced monospecific antibodies (Abs) against VZV gM and constructed a gM-deficient mutant virus. We found that the VZV gM homolog is a membrane glycoprotein expressed on virions and that the gM-deficient mutant exhibited impaired cell-to-cell spread compared with the parental virus.
|
|
|---|
Construction of rpOka and gM deletion mutant viruses.
The recombinant pOka virus (rpOka) was constructed as previously reported (33). In brief, rpOka was reconstituted by transfecting MRC-5 cells with the pOka bacterial artificial chromosome (BAC) genome (pOka-BAC) and then infecting them with a recombinant adenovirus harboring the Cre recombinase, AxCANCre, to delete the BAC sequence from the reconstituted virus. AxCANCre was kindly provided by Y. Kawaguchi, Tokyo University (18, 46). The gM deletion mutant virus (rpOka
gM) was constructed using pOka-BAC. To generate the gM-deficient mutant virus, a kanamycin resistance (Kmr) gene flanked by partial sequences of gM was constructed. The DNA fragments containing the Kmr gene were amplified from pCR2.1-TOPO (Invitrogen) by PCR with primers VZ50KMF200 (5'-TGCGTTTCTGTCGATGATACACATATGCACGCATCAATCTGAGAAGCAACGACAGCAAGCGAACCGGAATTGC-3') and VZ50KMR980 (5'-CCATGGTATATACACTGATCACGGCCGCGATTATACACCGAGAAACGCCTTTTTCAATTCAGAAGAACTC-3'). The DNA fragments were introduced by electroporation into Escherichia coli DH10B containing pOka-BAC and pGETrec, which was a kind gift from P. A. Ioannou (34), and the transformants were incubated and restreaked onto plates containing chloramphenicol and kanamycin. The presence of the expected recombination in the resulting colonies was confirmed by PCR amplification of the appropriate regions and restriction enzyme analysis.
Purification of virus particles. The cell-free virus samples were loaded onto a Histodenz (Sigma-Aldrich) gradient and subjected to ultracentrifugation at 27,000 rpm for 1 h at 4°C (P40ST-1689 rotor; Hitachi High Technologies). Fractions were collected from the top. The presence of virions in each fraction was analyzed by PCR with primers VZ50F80 and VZ50R1230. The presence of gM in each fraction was examined by Western blotting. The fractions containing both the viral genome and the gM protein were used as suspensions of purified virions (data not shown).
Endoglycosidase digestion. For endoglycosidase digestion, endoglycosidase H (endo H) and peptide N-glycosidase F (PNGase F) were purchased from New England Biolabs. The rpOka-infected cells were harvested 48 h after infection. rpOka- or mock-infected MRC-5 cells were lysed with radioimmunoprecipitation assay buffer (0.01 M Tris-HCl [pH 7.4]), 0.15 M NaCl, 0.1% sodium deoxycholate, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate [SDS], 1 mM EDTA) and spun at 70,000 rpm for 1 h in a TLA110 rotor (Beckman Coulter). The supernatants were then resuspended in digestion buffer (New England Biolabs), and the samples were digested with endo H or PNGase F in accordance with the manufacturer's protocol.
Abs.
Monospecific Abs were produced against synthetic peptides [(C)EDELLYERSNGWE] of the gM protein (Sigma-Aldrich, Japan). These peptides were used to immunize rabbits, and antisera were obtained (Sigma-Aldrich). Polyclonal Abs against open reading frame 16 (ORF16), ORF49, and gB were produced as described previously (40). Mouse monoclonal antibodies (MAbs) against
-tubulin (clone B-5-1-2; Sigma-Aldrich), GM130 (clone 35; BD Biosciences), p230-TGN (clone 15; BD Biosciences), EEA1 (clone 14; BD Biosciences), cathepsin D (clone BC011; EMD Biosciences), and nucleoporin p62 (clone 53; BD Biosciences) were purchased. Tetramethyl rhodamine isocyanate (TRITC)-conjugated swine anti-rabbit immunoglobulin G (IgG) (Dako) and an Alexa Fluor 488-conjugated F'(ab')2 fragment of goat anti-mouse IgG (Invitrogen) were also obtained.
Western blotting. MRC-5 cells were infected with VZV at a multiplicity of infection (MOI) of 0.01 and incubated for 48 h at 37°C. MeWo and MeWo-gM cells were inoculated with virus-infected cells and incubated for 72 h at 37°C. Samples were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and electrotransferred to polyvinylidene difluoride membranes. The membranes were incubated first with primary Abs and then with peroxidase-conjugated secondary Abs. The bound Abs were detected by chemiluminescence using Western blotting detection reagents (GE Healthcare Biosciences) and were visualized on X-ray film.
Immunofluorescence microscopy. Indirect immunofluorescence assays were performed as described previously (1, 40). All samples were analyzed on a model DMIRE2 Leica confocal microscope (Leica Microsystems).
RNA extraction and cDNA preparation for RT-PCR. Total RNA was extracted from virus-infected MRC-5 cells by using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. cDNA was synthesized with the SuperScript II First Strand cDNA synthesis kit (Invitrogen) in accordance with the manufacturer's instructions. PCR amplification was performed with the primer pairs described below, using the cDNA as the template. The primer pairs were as follows: for ORF51, ORF51F2041 (5'-ACCTTAAACCGCCCGGTTTGGCGAACAACC-3') and ORF51R (5'-ACCTTATAAACTTTCAAAATTTACCGC-3'); for cellular elongation factor, EF-F (5'-GCTCCAGCATGTTGTCACCATTC-3') and EF-R (5'-GGTGAATTTGAAGCTGGTATCTC-3'). PCR of the extracted RNA was performed without the reverse transcription (RT) reaction in order to rule out DNA contamination.
Southern blotting.
pOka-BAC DNA, pOka-BAC
gM DNA, or DNA extracted from rpOka- or rpOka
gM-infected cells (viral DNA) was digested with BamHI and separated by electrophoresis in a 0.5% agarose gel. Viral DNA was extracted from infected cells as previously described (33). After transfer to nylon membranes, the blots were probed with horseradish peroxidase-bound nucleotide probes using the ECL direct nucleic acid labeling and detection system (GE Healthcare Biosciences) in accordance with the manufacturer's instructions. The probes for the gM gene, ORF62 derived from pOka-BAC, and the Kmr gene derived from pCR2.1-TOPO (Invitrogen) were made by PCR with primers VZ50F80 and VZ50R1280, IE62fWA (5'-GCTTTTGGTAGAACTGGTTC-3') and IE62RvA (5'-TTCAACCAGAACCCAGAACG-3'), and KMF (5'-ATGATTGAACAAGATGGATTG-3') and KMR (5'-AAGAAGGCGATAGAAGGCGATG-3'), respectively. Bound probes were detected by chemiluminescence using ECL detection reagents and were visualized on X-ray film.
Determination of plaque size. Cells growing on six-well plates were infected in duplicate with cell-free virus at an MOI of 0.005 and incubated for 1 h to permit virus adsorption. The medium was then replaced with fresh medium, and the infected cells were incubated at 37°C for 10 days. The cells were then stained with 1% crystal violet-70% ethanol. The plaques were counted and their areas measured by ImageJ software (http://rsb.info.nih.gov/ij/).
Analysis of virus spread. To analyze the cell-to-cell spread of the recombinant virus, an infectious center assay was performed as described previously (15). Briefly, 24 h after seeding, cells were infected at an MOI of 0.005 with cell-free viruses as described above. The infected cells were harvested at the indicated time points by trypsinization. The cells were serially diluted, added to uninfected cells on six-well plates, incubated at 37°C for 7 days, and stained with 1% crystal violet-70% ethanol. The plaques were counted, and the number of infectious centers was evaluated as the increase over the value on day zero.
Electron microscopy.
rpOka- and rpOka
gM-infected MRC-5 cell cultures in culture flasks with a 75-cm2 growth surface (Greiner) were scraped off the plate and pelleted by low-speed centrifugation 36 h after cell-to-cell infection. The cells were immediately fixed for 120 min at 4°C with 2.5% glutaraldehyde and 4% paraformaldehyde in 0.05 M sodium cacodylate buffer (pH 7.4). Small pieces were postfixed in sodium cacodylate-buffered 1.5% osmium tetroxide for 60 min at 4°C, block stained in 0.5% uranyl acetate, dehydrated through a series of ethanol concentrations, and embedded in Epon resin (TAAB). Ultrathin sections were stained with uranyl acetate and lead citrate. Viral particles, purified by Histodenz gradients as described above, were negatively stained using 3% uranyl acetate. The samples were examined under a Hitachi (Tokyo, Japan) electron microscope (H-7650).
|
|
|---|
![]() View larger version (47K): [in a new window] |
FIG. 1. Western blotting of VZV- and mock-infected MRC-5 cells and purified virions. (A) (a) Cells and purified virions were lysed, separated by SDS-PAGE, and subjected to Western blotting with an anti-gM monospecific Ab, an anti-gB Ab, an anti-ORF16 Ab, or an anti-EEA1 MAb. Molecular masses are given to the left of Western blots. Infected cell, rpOka-infected MRC-5 cells; virion, rpOka virions; mock, mock-infected MRC-5 cells. (b) The purified virions used for Western blotting were confirmed by electron microscopy and negative staining. Scale bar, 200 nm. (B) Lysates were digested with endo H or PNGase F, resolved by SDS-PAGE under reducing conditions, and electrotransferred to polyvinylidene difluoride membranes. The blots were reacted with an anti-gM or anti-gB Ab. Open arrowheads indicate the precursor of gM.
|
VZV gM proteins were localized mainly to the Golgi apparatus and TGN in VZV-infected cells. In HSV-1, gM has been reported to localize with a marker of the Golgi apparatus and partially with a marker of the TGN in infected cells (4). To examine the localization of gM in VZV-infected cells, mock- or rpOka-infected MRC-5 cells were harvested at 48 h postinfection (hpi), fixed, and double stained with an anti-gM Ab and with Abs against several cellular proteins. Since gM was expressed abundantly at 48 hpi in VZV-infected cells and the localization was clearly observed, this time point was chosen for the study. As shown in Fig. 2, strong gM expression was observed in the perinuclear region. gM colocalized with GM130 (Fig. 2A), a marker of the Golgi matrix, and p230 (Fig. 2B), a marker of the TGN, and the localizations of GM130 and p230 were altered by VZV infection (Fig. 2A and B). gM also partially colocalized with LAMP1, a marker of late endosomes and lysosomes (Fig. 2C), but did not colocalize with cathepsin D, a marker of lysosomes (Fig. 2D), indicating that gM localized slightly to late endosomes but not to lysosomes. However, gM scarcely overlapped with EEA1, a marker of early endosomes (Fig. 2E), and did not overlap with nucleoporin p62, a marker of the nuclear membrane (Fig. 2F). These results suggest that VZV gM localized mainly to the Golgi apparatus and TGN.
![]() View larger version (32K): [in a new window] |
FIG. 2. Subcellular localization of gM in rpOka-infected MRC-5 cells. MRC-5 cells were either mock infected or infected with rpOka at an MOI of 0.01, fixed 48 h later, and stained with a rabbit anti-gM monospecific Ab and a mouse anti-GM130 (A), anti-p230 (B), anti-LAMP1 (C), anti-cathepsin D (D), anti-EEA1 (E), or anti-nucleoporin p62 (F) MAb. Nuclei were stained with Hoechst 33342. The anti-gM Ab was visualized with TRITC-conjugated anti-rabbit IgG, and the anti-GM130, anti-p230, anti-LAMP1, anti-cathepsin D, anti-EEA1, and anti-nucleoporin p62 mouse MAbs were visualized with an Alexa Fluor 488-conjugated F'(ab')2 fragment of goat anti-mouse IgG. Localization was analyzed by confocal laser scanning microscopy. Costained areas appear yellow in the merged images. Bars, 10 µm.
|
gM DNA or rpOka
gM viral DNA. The ORF62 probe hybridized with the predicted fragments (4.8 kb) in all samples (Fig. 4C and E).
![]() View larger version (13K): [in a new window] |
FIG. 3. Structure of the pOka-BAC genome and construction of the gM deletion mutant virus. (A) Schematic map of the pOka-BAC genome showing the long (UL) and short (US) unique regions, the inverted-repeat sequences (TRL, IRL, IRS, and TRS), the BAC sequence, the gM gene, and the vicinity of these sequences. (B) The relevant region of the genome is enlarged. Pointed rectangles represent ORFs in their transcriptional orientation. The indicated nucleotide fragment containing the kanamycin resistance gene (Kmr) was used for deletion of the gM gene by homologous recombination in E. coli. (C) The gM gene deletion mutant was constructed from the pOka-BAC genome (pOka-BAC gM). (D) The BAC sequence was removed from the reconstituted recombinant virus by Cre recombinase (rpOka gM). The resulting mutant virus contained a single loxP site. Black rectangles with a white letter L represent loxP sites.
|
![]() View larger version (74K): [in a new window] |
FIG. 4. Restriction enzyme digestion and Southern blot analyses. (A) pOka-BAC and pOka-BAC gM were digested with BamHI and separated on a 0.8% agarose gel. Molecular size markers are shown at the left. (B through E) Southern blot analyses of BamHI-digested pOka-BAC gM DNA and pOka-BAC DNA (B and C) and of rpOka gM and rpOka viral DNA (D and E) were performed. The blots were treated with labeled probes for the Kmr gene (B and D) and the gM and ORF62 genes (C and E).
|
gM-infected cells was verified by Western blot analysis. As expected, gM protein was detected in lysates of rpOka-infected MRC-5 cells but not in those of mock- or rpOka
gM-infected cells (Fig. 5A), confirming that the gM gene had been deleted from the VZV genome. Levels of protein or mRNA expression from the neighboring ORF49 or ORF51 gene were similar in cells infected with the two types of virus, indicating that gM's neighboring genes were not affected by the mutagenesis (Fig. 5B and D).
-Tubulin or elongation factor was used as a positive control for protein or mRNA expression of the cell lysates and was detected in all the samples (Fig. 5C and E).
![]() View larger version (46K): [in a new window] |
FIG. 5. Absence of gM expression in rpOka gM-infected cells. rpOka-, rpOka gM-, or mock-infected MRC-5 cells were harvested at 48 hpi. (A, B, and C) Cells were lysed, separated by SDS-PAGE, and subjected to Western blotting with an anti-gM (A) or anti-ORF49 (B) Ab or with an anti- -tubulin MAb (C). (D and E) RNA was extracted, and RT-PCRs of ORF51 (D, lanes 3 and 6) and elongation factor (E, lanes 3 and 6) were performed using specific primer pairs. Lanes 2 and 5, PCRs without RT, performed as a control for contamination with genomic DNA. Lanes 1 and 4, PCRs of viral DNA, performed as a positive control. Sizes of molecular mass markers are given on the left.
|
gM, plaque formation and infectious center assays were performed. The infection of MRC-5 cells by rpOka
gM caused plaques (0.35 mm2 ± 0.01 mm2 [average ± standard error]) about 90% smaller than those caused by the wild-type virus, rpOka (4.09 mm2 ± 0.37 mm2) (Fig. 6), and the same result was obtained when MeWo cells were used (data not shown), suggesting that plaque formation was impaired in rpOka
gM-infected cells. The plaques in rpOka
gM-infected MeWo-gM cells, stably expressing gM protein, were approximately 33% larger than those in rpOka
gM-infected MeWo cells (Fig. 7A), indicating that the phenotype of the mutant was due to the deletion of the gM gene. The expression of gM in MeWo-gM cells was confirmed by Western blotting. As shown Fig. 7B, the 42- to 48-kDa form of gM was detected in rpOka
gM-infected MeWo-gM cells, while only the 37-kDa form, a precursor of gM, was detected in mock-infected MeWo-gM cells, indicating that gM expression was found in MeWo-gM cells; however, the formation of the mature form of gM is required for the expression of the other virus gene(s). In addition, the expression of gM was lower in rpOka
gM-infected MeWo-gM cells than in rpOka-infected MeWo cells. This may be due to the lower copy numbers of the gM gene in MeWo-gM cells.
![]() View larger version (75K): [in a new window] |
FIG. 6. Comparison of plaque sizes of rpOka gM and rpOka. MRC-5 cells were infected with cell-free rpOka or rpOka gM virus at an MOI of 0.005 and were cultured for 10 days. The cells were fixed and stained with 1% crystal violet-70% ethanol.
|
![]() View larger version (28K): [in a new window] |
FIG. 7. Comparison of plaque sizes of rpOka gM in MeWo and MeWo-gM cells. (A) MeWo or MeWo-gM cells were infected with cell-free rOka gM virus at an MOI of 0.005 and were cultured for 7 days. The cells were fixed and stained with 1% crystal violet-70% ethanol. Plaque sizes were scanned and measured using ImageJ. Error bars, standard errors. Statistical significance was determined by Student's t test. (B) MeWo cells infected with rpOka (lane 1) or rpOka gM (lane 2) or mock infected (lane 3) and MeWo-gM cells infected with rpOka gM (lane 4) or mock infected (lane 5) were harvested at 72 hpi. The cells were lysed, separated by SDS-PAGE, and subjected to Western blotting with an anti-gM Ab and an anti-tubulin MAb.
|
gM, and the number of infected cells from 0 to 5 days was titrated. In this assay, rpOka exhibited an obviously faster spread than the mutant virus in MRC-5 cells. After 5 days of cell spread, the number of cells infected by rpOka
gM was about half the number infected by rpOka (Fig. 8). These results suggested that viral growth was marginally affected and that the cell-to-cell spread of viral progeny was impaired by the deletion of gM from the VZV genome.
![]() View larger version (11K): [in a new window] |
FIG. 8. Infectious center assay of rpOka and rpOka gM. MRC-5 cells in six-well plates were infected independently with similar titers of cell-free rpOka or rpOka gM virus and were then treated with trypsin from day 1 to day 5 postinfection. The treated cells were diluted and overlaid onto newly prepared, uninfected MRC-5 cells in six-well plates. Seven days later, the overlaid cells were fixed and stained with 1% crystal violet-70% ethanol, and the number of infected cells was assessed by counting the number of VZV plaques. The number of infected cells was normalized to the initial viral titer per dish; the increase (n-fold) is expressed as the number of infected cells from one initial infected cell on day zero. Error bars, standard errors. Statistical significance was determined by a paired Student t test.
|
gM or rpOka by inoculating them with virus-infected MRC-5 cells. At 36 h after cell-to-cell infection, cells were fixed, stained, and analyzed. As shown in Fig. 9B, after infection with rpOka
gM, numerous aberrant vacuoles containing electron-dense materials were found in the cytoplasm. Enveloped particles termed L particles (28, 44, 45) lining the outer surfaces of the cells (Fig. 9C and F) and capsid assembly in the nucleus (Fig. 9A and D) were observed in cells infected with either type of virus. L particles are noninfectious virus-like particles that lack both capsids and virus DNA (28, 45).
![]() View larger version (152K): [in a new window] |
FIG. 9. Electron microscopy of rpOka gM- or rpOka-infected cells. MRC-5 cells were infected with rpOka gM (A, B, and C) or rpOka (D, E, and F) and analyzed at 36 h after cell-to-cell infection. (A and D) Nucleocapsids were observed in the nuclei of both rpOka gM- and pOka-infected MRC-5 cells. (B and E) Aberrant vacuoles containing relatively electron dense materials (arrows) were found in the cytoplasm of rpOka gM-infected cells but not in that of pOka-infected cells. (C and F) Numerous extracellular L particles (arrowheads) lining the surface of the plasma membrane were observed in both rpOka gM- and rpOka-infected cells. Bars, 1 µm.
|
|
|
|---|
In this study, abundant gM proteins with a molecular mass of 42 to 48 kDa or 37 kDa were detected in VZV-infected cells, and only the 42- to 48-kDa form was present in virions as a glycoprotein with a complex N-linked oligosaccharide. Generally, gM proteins encoded by herpesviruses are highly hydrophobic proteins that are predicted to contain six to eight transmembrane domains. They have a potential N-glycosylation site within the first extracellular domain and a conserved cysteine residue within the same loop that is predicted to form a disulfide bond with the relevant gN homolog (10, 17). VZV gM is also predicted to be an integral membrane protein with eight transmembrane domains, a putative N-glycosylation site, and a cysteine residue that could form a disulfide bond in the first ectodomain (Fig. 10).
![]() View larger version (27K): [in a new window] |
FIG. 10. Potential topology of VZV gM. The topology of VZV gM was predicted using the TMHMM transmembrane topology prediction server (http://www.cbs.dtu.dk/services/TMHMM-2.0/). The rectangle at the top is a schematic representation of the predicted gM protein. aa, amino acids. Dashed lines indicate the region of the gM gene replaced with the kanamycin resistance gene. The approximate locations of an N-linked glycosylation site (NAT, asparagine-alanine-threonine; branched structure, N-linked sugar chain) and a cysteine (C) that could form a disulfide bond (-S-S-) are indicated.
|
Because VZV gM was found to be an envelope glycoprotein, we next examined its localization in infected cells. VZV gM was localized mainly to the Golgi apparatus and TGN of infected cells, indicating that gM together with other envelope glycoproteins entered the TGN for secondary envelopment, because VZV budding into TGN-derived vesicles has been reported (14). Furthermore, gM was localized partially to the late endosomes of infected cells, which might have reflected gM incorporation into virions, because mature or immature virions have been reported to be transported to late endosomes (14). In addition, gM was scarcely localized to early endosomes, indicating that gM may not traffic to early endosomes and may traffic directly to the TGN for secondary envelopment.
Recently, Baines et al. reported that HSV-1 gM was localized at the perinuclear region near the nuclear membrane and that gM was incorporated into virions during budding at the inner nuclear membrane (4). In the present study, no VZV gM expression was observed on the nuclear membrane, showing that there may be different biological functions of gM among alphaherpesviruses. However, the detection of gM on the nuclear membrane may depend on the stage of infection or the recognition site of the antibody used for detection. Further study would be required to analyze this issue more.
To analyze the function of VZV gM more precisely, we performed an electron microscopic examination, and the morphological features of rpOka
gM- and rpOka-infected MRC-5 cells were compared. Whereas cell-to-cell spread and plaque formation were impaired in rpOka
gM infection, L particles lining the outer surfaces of the cells and capsid assembly in the nucleus were observed in mutant virus-infected cells. This showed that the single deletion of gM of VZV did not cause significant impairment of particle formation, although the formation of mature virions was thought to be partially impaired. Ultrastructural studies of PRV (5) and EHV-1 (41) indicate that the deletion of gM impairs secondary envelopment, and this phenotype is more severe when the gE/gI complex is deleted together with gM. Deletion of the C-terminal region of gE in a gM deletion mutant of PRV showed a phenotype similar to that of a gE/gI/gM triple-deletion mutant (6). gM may have supplemental roles in secondary envelopment, which were not affected in the gM single-deletion mutant. rpOka
gM showed impairment of cell-to-cell spread of progeny but no impairment of L particle formation. Aberrant vacuoles containing electron-dense materials were found in rpOka
gM-infected MRC-5 cells, while they were scarcely found in rpOka-infected cells. In light of these data, the deletion of gM is thought to partially impair virion maturation and the related protein sorting, although the detailed mechanism is unknown. In PRV-infected cells, gM interacted with UL49, and a gE/gI/gM deletion mutant failed to incorporate UL49 into viral particles (11). As well, VZV gM may be more important for reciprocal functions with other gene products.
Here, by deleting VZV gM, we found that, although VZV gM was nonessential for virus growth, its loss caused impairments in plaque size and cell-to-cell spread, indicating that VZV gM may play an important role in virus spread to neighboring cells. In contrast, a gM gp2 double-negative mutant of EHV-1 (39) shows a defect in viral egress from infected cells, but only a slight impairment in cell-to-cell spread, compared with the parental virus and a gp2 single-deletion mutant. Therefore, the roles of gM may be redundant and/or have variable importance among the herpesviruses.
VZV gE is the most abundant viral glycoprotein in VZV-infected cells (32), and its deletion from VZV abolishes the virus's ability to reproduce (31). Recently, insulin-degrading enzyme was found to be a cellular receptor for VZV, and gE was shown to be the viral ligand for the receptor (23). In PRV, triple deletion of the gM, gE, and gI genes significantly impairs viral growth and secondary envelopment (5), indicating a redundancy between gM and the gE/gI complex. The gE/gI complex has an essential role in the cell-to-cell spread of VZV in vitro, and VZV gM may assist the gE/gI complex in this function.
Complementation of the growth of the rpOka
gM virus in MeWo-gM cells was partial. It appeared that not all of the virus particles produced in rpOka
gM-infected MeWo-gM cells were complemented with gM, in view of the lower expression of gM in MeWo-gM cells than in wild-type virus-infected MeWo cells (Fig. 7B).
In summary, we showed that VZV gM is a complex-type N-linked glycoprotein, that it is present in viral particles, and that the gM deletion mutant of VZV shows impaired viral growth and cell-to-cell spread. This is the first report to demonstrate that VZV gM is an envelope glycoprotein and plays a role in viral cell-to-cell spread.
This study was supported in part by a grant-in-aid for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan.
Published ahead of print on 31 October 2007. ![]()
Yoshiaki Yamagishi and Tomohiko Sadaoka contributed equally to this work. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»