Tyrosine 116 of the Herpes Simplex Virus Type 1 IEα22 Protein Is an Ocular Virulence Determinant and Potential Phosphorylation Site

Abstract

purpose. To determine whether tyrosine 116 of the HSV-1 α22 protein is involved in virulence and is a potential phosphorylation site.

methods. Site-directed mutagenesis was used to revert the Y116C mutation in the α22 gene of the strain OD4 to wild type (C116Y), and the effect of virulence was tested by using a marker transfer–infection protocol in mice. Immunoblot analysis, tryptic phosphopeptide mapping, and phosphotyrosine pulldown–immunoblot protocols were used to assess the OD4 α22 isoforms.

results. Reversion of the Y116C mutation resulted in a significant increase in the severity of ocular disease compared with the OD4 virus alone. Reversion of the Y116C and a previously identified mutation (S34A) together did not alter the severity of virulence compared with either mutation alone. Immunoblot analysis revealed a loss or reduction in α22 isoforms in the OD4 virus compared with wild type (CJ394 virus). The OD4 virus had numerous alterations in the α22 tryptic phosphopeptide pattern, including loss of specific peptides and shifts in the position of several peptides. Phosphotyrosine pulldowns revealed a loss of one or more isoforms and shifts in the apparent size of others.

conclusions. The data indicate that Y116 is a determinant of peripheral virulence in mice and that mutations at S34 and Y116 affect virulence independently. The data also show that the S34 and Y116 mutations substantially alter phosphorylation of the α22 protein, that Y116 is a potential phosphorylation site, and that the α22 protein contains at least two phosphotyrosines. These results are the first to show that mutation of a specific tyrosine in the α22 protein is associated with virulence.

Herpes simplex virus (HSV) is an major pathogen, infecting up to 90% of the population and causing mucocutaneous ulcers at many sites in the body. HSV is also the leading cause of sporadic viral encephalitis and blindness in developed countries. ¹ Despite the availability of several antivirals, HSV infections remain a serious problem. 

Animal studies have clearly shown that the strain of virus plays an important role in determining disease severity, indicating that genes in the virus are critical. Although the HSV genome was sequenced in 1988, ² little is known about the functions of most HSV genes in the viral life cycle, and even less is known about the role of genes in virulence. Most studies have focused on the role of HSV genes in central nervous system (CNS) infection and several genes that function in encephalitis have been identified. ^{3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20} Genes influencing peripheral disease are less well characterized, but some have been identified. ^{4 5 21 22}  

Recently, we demonstrated that multiple genes can act in concert to determine the virulence phenotype of a given strain of HSV and that altering the combinations of genes could alter the disease. ²³ During these studies, we not only identified new HSV virulence genes, but also revealed several novel mutations in previously identified virulence genes. In particular, we found two mutations, S34A and Y116C, in the US1 gene of viral strain OD4, which encodes the immediate early α22 (IEα22) protein. Using site-directed mutagenesis, we reverted the mutation at position 34 to wild type and showed that this restores ocular virulence. ²³  

The US1 gene contains an open reading frame of 420 amino acids that encodes a family of proteins that migrate with an apparent molecular mass ranging from 68 to 85 kDa. ^{24 25 26} Deletion mutants without the US1 gene replicate in cultured cells, ^{27 28 29} but display a host range phenotype, in that they replicate less well in certain cell types. These mutant viruses also have reduced capacity to establish latency, at least in mice. ^{30 31} The α22 protein performs several functions, including regulating the expression of a subset of HSV-1 genes including ICP0, as well as certain β and γ genes. ^{30 32 33 34} The α22 protein is also involved in regulating the phosphorylation state and localization of RNA polymerase II, ^{35 36 37} and this function may be involved in the regulation of viral gene expression. The α22 protein has also been shown to bind to or affect the localization of other viral (UL3, UL4) and cellular proteins (p60, p78) within cells. ^{38 39 40 41}  

The α22 protein is heavily posttranslationally modified, being tyrosine and serine phosphorylated, guanylated, and adenylated. ^{26 32 42 43 44 45} At least seven isoforms of the α22 protein have been identified in infected cells. ²⁶ One important unknown concerning the α22 protein and its function is the location of the modifications. Mutation of a tyrosine kinase consensus site (Y193A) has been shown to alter the isoform pattern, but to date none of the posttranslational processing sites have been unambiguously identified. We do not know which isoforms have which modifications, nor which isoforms are responsible for performing the numerous functions ascribed to the α22 protein. 

Given that the α22 protein is tyrosine phosphorylated, it was of interest to ask two questions regarding the Y116C mutation we had identified. ²³ First, is tyrosine 116 a virulence determinant, and, second, is tyrosine 116 a phosphorylation site in the protein? Using site-directed mutagenesis and marker transfer–infection assays, we showed in the current study that Y116 is involved in virulence and that the S34A and Y116C mutations in the strain OD4 US1 gene act independently to promote ocular virulence. We also showed that an α22 protein from strain OD4, carrying both the S34A and Y116C mutations, displays an altered phosphorylation pattern and, using phosphotyrosine pulldown assays, we showed that Y116C is a possible phosphorylation site in at least some of the α22 isoforms. 

Materials and Methods

Cell Culture and Viruses

Vero cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) containing 5% serum and antibiotics, as we described previously. ⁴⁶ High-titer stocks of the HSV-1 strains CJ394 and OD4 were prepared by infecting Vero cells at a low multiplicity of infection (MOI 0.1), as described elsewhere. ⁴⁶ The virulence properties of CJ394 and OD4 have been described. ^{46 47} Briefly, CJ394 causes moderate keratitis after ocular infection and is not neurovirulent in mice aged 4 to 6 weeks or more. Strain OD4 is highly attenuated for peripheral and ocular virulence, carrying numerous attenuating mutations. ²³ The sequence of the CJ394 and OD4 US1 genes has been reported. ²³ Briefly, the CJ394 α22 gene has the same sequence as strain 17. ² Strain OD4 has four amino acid changes—S34A, D46E, D68E, and Y116C (Fig. 1) —in the region known to transfer virulence. ²³  

Site-Directed Mutagenesis of the OD4 US1 Gene

The reversion of the C116Y mutation was performed with two consecutive PCR reactions. In the first step, which introduced the desired change (C to Y), two primers, 5′-AAGCCACTCTGGTCCTCCGA-3′ (forward) and 5′-GAGGGAATGTCATAAGACCAGTCG-3′ (reverse), were used to generate a megaprimer. The first reaction consisted of 1× reaction buffer, 200 ng of the plasmid pOD4HE, ²³ 300 ng of each primer, 100 μM of dNTPs (final concentration), 1% dimethyl sulfoxide (DMSO), and 2.5 units of pfu polymerase (Promega, Madison, WI) in a total volume of 100 μL. The cycling conditions were 1 cycle at 95°C for 5 minutes; 35 cycles of 95°C for 1 minute, 47°C for 1 minute, and 72°C for 17 seconds; and 1 cycle at 72°C for 7 minutes. After electrophoresis in a 1.5% agarose gel, the 145-bp megaprimer was isolated by gel extraction (QIAquick Gel Extraction Kit; Qiagen, Inc., Valencia, CA). The same reaction conditions were used in the second PCR step, except the megaprimer and the reverse (AAACAAGAAAGCTTGCACACG) primer were used. The cycling conditions were 1 cycle of 95°C for 5 minutes; 30 cycles of 95°C for 1 minute, 47°C for 2 minutes, and 72°C for 1 minute 43 seconds; and a final cycle at 72°C for 7 minutes. The product was gel purified and cloned into the pZErO-1 vector (Invitrogen, Carlsbad, CA). We also engineered the loss of a Tsp45I restriction site in the mutagenic primer to follow the mutation. The resultant C116Y clone was verified by sequencing at the University of Wisconsin-Madison Biotechnology Center, as we have described previously. ²³ The method for reversion of the S34A mutation has also been described. ²³  

Marker Transfer of the C116Y Revertant US1 Gene

The gel-purified C116Y US1 gene fragment or a fragment containing reversions at both sites (34 and 116) was mixed with purified OD4 DNA ⁴⁸ and cotransfected into Vero cells using the modified calcium-phosphate method we described previously. ²³ Briefly, the ends of the US1 fragment were filled in with dideoxynucleotides, the CaPO₄ precipitates were allowed to form for only 60 seconds at room temperature, the cells were split 1:1 12 to 18 hours before transfection, and finally the cells were trypsinized, mixed with the precipitates, and replated. The cells were shocked with 10% DMSO in HEPES buffer (140 mM NaCl, 5 mM KCl, 0.7 mM Na₂HPO₄, 6 mM glucose, 20 mM HEPES [pH 6.9]), rinsed twice with medium, and refed 4 to 6 hours after transfection. The next day, the medium was changed to 2% serum and incubated at 37°C. When the cultures reached 100% cytopathic effect (cpe), the cells were collected, centrifuged at 2000g for 10 minutes, resuspended in 250 μL of medium with 2% serum, and frozen and thawed three times. After centrifugation at 2000g for 10 minutes, the supernatants were stored at −80°C. The titers of the marker transfer stocks ranged from 1.0 × 10⁸ to 5.0 × 10⁸ plaque-forming units (pfu)/mL. 

Animal Infection and Disease Scoring

Four- to six-week-old female BALB/c mice were infected as we have described. ⁴⁹ Briefly, the mice were anesthetized with halothane (3%–5%) and the corneas were scratched with a 30-gauge sterile needle. A 5-μL drop of each marker transfer stock (5 × 10⁵ to 2.5 × 10⁶ pfu) was then placed on the scarified cornea, and the mice were returned to their cages. Five independent transfections were tested for each gene with five mice in each of the five groups (25 mice in total). The severity of ocular disease was scored as we have described in several publications. ^{23 46 47 50 51 52} We have shown that strain OD4 causes mild blepharitis, but no corneal neovascularization or stromal disease. ⁴⁶ Thus, for the mice to be scored positive for virulence, they had to have a minimum score of 1 for vascularization or 2 for stromal keratitis or had to die of encephalitis. The presence of blepharitis was used as a marker of infection, particularly in OD4-infected mice, but was not counted for virulence. For a mutation to be considered positive for virulence, a minimum of 20% of the mice had to score positive (P < 0.05) and at least two of the five independent transfections had to score positive. ²³ The use of animals in this study was approved by the University of Wisconsin Medical School IACUC and conforms to the guidelines of the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Immunoblot Analysis

Vero cells were infected at an MOI of 5 in DMEM containing 2% serum. When the cytopathic effect reached 100% (24 hours after infection), the cells were harvested and centrifuged at 2000g for 10 minutes The medium was removed and the cells were resuspended in Laemmli buffer with 2-mercaptoethanol at a concentration of 8 × 10⁷ cells/mL. The samples were sonicated at 30% duty cycle for 10 pulses with a cell disrupter (model 200; Branson Ultrasonics, Danbury, CT), boiled for 5 minutes, and centrifuged at 14,000g for 2 minutes. The supernatants were then electrophoresed in a 10% polyacrylamide gel ⁵³ for 18 hours at a 15-mA constant current. The samples were transferred electrophoretically to polyvinylidene difluoride (PVDF) membranes (Osmonics, Minnetonka, MN). The filters were then probed with a 1:2500 dilution of RGST22 antibody (kindly provided by John Blaho, Mount Sinai School of Medicine, New York, NY) for 1.5 hours. After they were washed, the filters were treated with a 1:2500 dilution of goat anti-rabbit horseradish peroxidase (HRP)–conjugated secondary antibody (Sigma-Aldrich; St. Louis, MO) for 1.5 hours. The filters were then washed and developed with a chemiluminescence detection kit (ECL; Amersham Biosciences, Piscataway, NJ) and exposed to x-ray film (Fuji Photograph Film, Ltd., Tokyo, Japan). 

Two-Dimensional Phosphopeptide Assay

A confluent 10-cm plate of Vero cells was infected with either CJ394 or OD4 at an MOI of 10. At 50 minutes after attachment (1 hour 50 minutes after exposure to virus) 500 μCi of ³²[P]-orthophosphate in phosphate-free DMEM-12% serum was added, and the labeling was allowed to proceed for 5 hours. The cells were then harvested and centrifuged at 2000g for 10 minutes. The cell pellet was resuspended in 300 μL of Laemmli buffer and processed as described for immunoblot analysis. The sample was electrophoresed in a 12% polyacrylamide gel for 18 hours at 15 mA and then electrophoretically transferred to nitrocellulose. 

The filter was wrapped in plastic and exposed to x-ray film (Fuji) for 2 hours with the protein side toward the film. The band corresponding to the α22 protein was excised and placed into a microfuge tube with 200 μL of polyvinylpyrilidone solution (PVP 360,000, 0.5% vol/vol) and 100 mM acetic acid and then incubated with shaking for 30 minutes at 37°C. The gel slice was washed five times with deionized distilled (dd)H₂O (1 mL each), and twice with 1 mL freshly prepared 50 mM NH₄CO₃ solution. Tryptic peptides were prepared by incubating the gel slice in 150 μL of a solution containing 100 μg sequencing grade trypsin (V511A; Promega) in 50 mM NH₄CO₃ for 5 hours at 37°C with shaking. The sample was mixed and centrifuged at 12,000g for 30 seconds, and the supernatant was transferred to a new tube. The gel slice was rinsed with 300 μL of water, and the supernatants were combined. The sample was then lyophilized and resuspended in 50 μL of ice-cold performic acid solution (98% performic acid [vol/vol], 10% H₂O₂ [vol/vol]), and the radioactivity measured in a scintillation counter (model LS5801; Beckman, Inc., Fullerton, CA). The samples were then applied to a cellulose thin-layer chromatography (TLC) plate (JT Baker, Inc., Phillipsburg, NJ) and electrophoresed with a commercial apparatus (Multiphore-II; Amersham Biosciences) for 25 minutes at 100 V, 50 mA, and 50 W. The TLC plates were rotated 90° and then subjected to TLC in n-butanol, pyridine, and acetic acid (15:10:12) for 5 hours. The plates were then exposed to x-ray film overnight. 

Phosphotyrosine Pulldown Assays

Vero cells (2 × 10⁷) were infected with either OD4 or CJ394 virus at an MOI of 10 and incubated for 8 hours at 37°C. The cells were rinsed once with PBS, 1 mL of lysis buffer (PBS with 1% NP-40 [vol/vol], 1% deoxycholate [wt/vol], 10 μM tolylsulfonylphenalanyl chloromethyl ketone, and 10 μM α-tosyl-l-lysine chloromethyl ketone) was added, and the plates were incubated at 4°C for 30 minutes. The lysate was scraped and transferred to a microcentrifuge tube sonicated at 30% duty cycle, 10 pulses, with the cell disrupter (model 200; Branson). The lysate was then centrifuged for 10 minutes at 10,000g at 4°C. The supernatant was transferred to a new tube and 10 μL of agarose beads coupled to anti-phosphotyrosine antibody (A-1806; Sigma-Aldrich) was added, and the lysate incubated with mixing for 1 hour at 4°C. The lysate was then centrifuged at 2000g for 5 minutes and rinsed with 1 mL of ice-cold PBS. The washes were repeated twice more and the pellet was resuspended in 50 μL of Laemmli buffer. After they were boiled for 5 minutes and centrifuged at 2000g for 5 minutes, the supernatants were electrophoresed in 10% SDS-PAGE gels as described earlier. After transfer to nitrocellulose, the filters were immunoblotted using α22 specific antiserum (RGST22), as described earlier. 

Results

Effect of Reversion of the Y116C Mutation on Ocular Virulence

To determine whether tyrosine 116 of the US1 gene was involved in virulence, we used site-directed mutagenesis to revert the Y116C mutation in the OD4 gene back to wild type (C116Y; Fig. 1 ). We then tested for restoration of virulence using a marker transfer–infection assay previously validated by us and others. ^{23 54} As shown in Figure 2A , 40% of the mice showed significant ocular disease with the reverted mutation (C116Y) compared with none showing disease with the OD4 transfection alone. None of the infected mice died, suggesting that reversion of the Y116C mutation affects only peripheral disease. Figure 2B shows the mean peak disease scores (MPDSs) for OD4 and the revertant virus. Vascularization and stromal disease were restored to moderate levels only with the reverted US1 gene, and the increases in disease severity with the C116Y revertant were highly significant (Table 1) . 

The OD4 US1 gene also carries a mutation at position 34 (S34A), which we have shown is involved in peripheral virulence. ²³ Because reversion of either the S34A or Y116C mutations restored virulence separately, we asked how reversion of both mutations would affect disease. As shown in Figure 2 , reversion of both mutations together resulted in virulence that was indistinguishable from either mutation alone, with 40% of the mice showing that disease, stromal keratitis, and vascularization were restored to the same levels. The differences in the MPDSs between OD4 and the double reversion were highly significant (Table 1) . Mortality was also not increased. These results suggest that the two mutations independently affect peripheral virulence. 

Altered α22 Isoform Pattern in Strain OD4

Because the α22 protein is known to be phosphorylated on both serine and tyrosine residues and because we have identified virulence-related serine and tyrosine mutations in strain OD4, it was of interest to see whether the OD4 α22 protein displays an altered isoform pattern. Strain CJ394, which causes moderate ocular disease and in which the US1 gene has the wild-type sequence at positions 34 and 116, was used as the control. Vero cells were infected with each virus, and cell extracts were analyzed by immunoblot. To optimize visualization of the various isoforms, the samples were serially diluted twofold before electrophoresis. As shown in Figure 3 , several isoforms were seen in the CJ394 sample with what are two clusters of proteins migrating with apparent molecular masses of 85 kDa (open arrows) and 75 kDa (closed arrows) and a single form at 68 kDa (small arrow, CJ394, lane 1). Because the samples were serially diluted, the 68-kDa band was lost first in the OD4-infected cells, followed by the cluster at 85 kDa, suggesting these isoforms were present at lower levels in infected cells. In the OD4 samples, the isoforms at 68 and 85 kDa in the undiluted samples were either absent or present in much smaller amounts than in the CJ394 sample. Because the OD4 sample was diluted, only the 75-kDa isoforms were seen. In addition, it also appeared that the number of the 75-kDa forms was increased in the OD4-infected cells (Fig. 3 , compare lanes 3 and 7). These results suggested that posttranslational processing of the OD4 α22 protein was altered. 

To confirm that phosphorylation of α22 was altered in OD4, we performed a two-dimensional tryptic phosphopeptide mapping analysis. The data in Figure 4A show that at least 17 tryptic phosphopeptides were present in the CJ394 α22 sample. Analysis of the OD4 pattern (Fig. 4B) showed that overall, the number of peptides was much reduced, that numerous tryptic phosphopeptides were missing, and that others had shifted position. To highlight the differences, the OD4 and CJ394 data were colorized and then the images were merged (Fig. 4C) . Several peptides that differ between the two viruses are marked in Figure 4C . Note, in particular, peptides marked 394-2b and 394-5 in the CJ394 protein, which are missing in the OD4 α22 protein. Also tryptic peptides OD4-2b and OD4-4 occupy positions similar to those of peptides 394-7 and 394-10; however, the OD4 peptides are shifted toward the anode to a lesser degree. In addition, peptide intensity also changes. For example, peptide 7-394 shows greater intensity than the similar peptide OD4-2b. These results confirm the immunoblot analysis data indicating that posttranslational processing of the OD4 α22 protein is substantially altered. 

A Potential Phosphorylation Site

We next wanted to determine whether either S34 or Y116 was a possible phosphorylation site. Reasoning that if either residue were phosphorylated, the isoforms without S34 or Y116 would not be precipitated, we used agarose-coupled monoclonal antibodies to either phosphoserine or phosphotyrosine in pulldown assays. To date, we have been unable to pull down the serine-phosphorylated α22 protein. However, Figure 5 shows the results from the phosphotyrosine pulldown–α22 immunoblot analyses. At least one isoform migrating at 85 kDa in CJ394-infected cells was clearly absent in the OD4 samples. In addition, the material at 75 kDa, which is composed of a doublet, appeared to have shifted in the OD4 samples to a lower apparent size, as would be expected if these isoforms lost one or more phosphate groups. The loss of the 85-kDa isoform indicates that in this isoform, tyrosine phosphorylation occurs only at position 116 in strain CJ394. A faint band migrating at 68 kDa, which is difficult to see in Figure 5 , was also seen in the CJ394 and OD4 phosphotyrosine pulldown assays, suggesting a small amount of this isoform may be tyrosine phosphorylated. 

Discussion

The HSV-1 α22 protein has been shown to be a host range determinant, because deletion of the entire gene ³¹ or the carboxyl terminus ³⁰ reduces the capacity of the mutant viruses to cause disease in animal models, alters the latency phenotype of the mutant viruses, results in growth defects in certain cell lines, and reduces viral replication in vivo. Recently, we demonstrated that the virulence phenotype of HSV-1 was dependent on the functioning of multiple genes. ²³ One of the genes we identified as being required for peripheral virulence (keratitis) was the US1 gene encoding the α22 protein. We also reported that S34 of the α22 protein was required for virulence and noted the presence of a second mutation, Y116C, in the same protein. ²³ In the present work, we show that the tyrosine at position 116 is also required for ocular virulence. 

We previously showed that reversion of the S34A mutation restores virulence. ²³ Our observation that reversion of the Y116C mutation also restored virulence led us to ask whether S34 and Y116 had any effect on each other. We found that reverting both mutations to wild type in the same construct resulted in virulence scores identical with those obtained with either mutation alone, suggesting that S34 and Y116 apparently have independent functions in virulence. This observation has important implications for the studies on the genetics of viral virulence. Considering that both mutations are involved in virulence, if either residue in the wild-type protein had been mutated alone, we would have failed to see any effect. These results demonstrate the value of gain-of-function studies for the identification of virulence factors. Presently, we do not know whether S34 and Y116, independently, affect a similar step or whether two independent virulence functions are affected. 

Our data also indicate that posttranslational processing of the OD4 α22 protein is substantially altered based on the changes in the isoform pattern (Fig. 3) , the significant differences in the tryptic phosphopeptide patterns between CJ394 and OD4 viruses (Fig. 4) , and the differences in results of the pulldown experiments (Fig. 5) . Because it has been shown that multiple mutations in widely dispersed sites in the α22 protein can affect posttranslational processing, ⁴⁴ we cannot as yet ascribe the altered processing to any single mutation. The OD4 US1 gene also contains two conservative mutations, D46E and D68E; however, because these changes were present in all the constructs we tested, they would not affect the interpretation of our data. 

We found that at least one isoform migrating with an apparent size of 85 kDa was missing in the OD4 phosphotyrosine pulldown study, indicating that this isoform has a single phosphotyrosine modification that is lost in the OD4 protein. The observation that additional isoforms were seen in the OD4 phosphotyrosine pulldown studies shows that some isoforms must contain at least two phosphotyrosines. O’Toole et al. ²⁶ have recently shown that mutation of tyrosine 193 alters the processing patterns of the α22 protein, and this may be the second tyrosine phosphorylation site. 

Two possible explanations exist for the altered posttranslational processing of the OD4 α22 protein with respect to tyrosine phosphorylation. The Y116C mutation in the OD4 protein could alter the ability of tyrosine kinases to modify the protein, or the Y116C mutation could remove a phosphorylation site. A comparison of the relative amounts of the α22 protein in OD4- and CJ394-infected cells (Fig. 3) shows that the total amount of protein is essentially equal. Data from the phosphotyrosine pulldown assays showed that the OD4 protein is tyrosine phosphorylated in at least one additional site, suggesting that any potential conformational change is not severe enough to affect tyrosine phosphorylation at this site. 

Our analysis of isoform patterns was performed at several times after infection, ranging from 8 hours to as late as 24 hours. Although the α22 gene is classified as an immediate early gene, ²⁹ the function of α22 is required for efficient expression of a subset of late genes ³² ; thus, the protein functions throughout the life cycle and analysis of isoforms at late times is warranted. It will be interesting to analyze the kinetics of processing the α22 protein at various times after infection, once viruses containing only the S34A and Y116C mutations are available. The role of α22 in function is also dependent on the cell type, ³⁰ and it will be of interest to analyze α22 processing in various cells with appropriate mutant viruses. 

Some of the modifications to the α22 protein depend on the UL13 kinase and to a lesser extent on the US3 kinase, ^{32 55} and deletion of the UL13 kinase gene generates a virus with a phenotype very similar, if not identical, to a US1 deletion, ³² raising the possibility that mutations in the OD4 UL13 or US3 genes could affect virulence through altered phosphorylation of the α22 protein. A previous genome screen of OD4 virulence mutations ²³ showed that the CJ394 UL13 and US3 genes were not involved in transfer of virulence to the OD4 virus. Moreover, the reversion of the Y116C mutation in the OD4 α22 protein restored ocular virulence, localizing the effect to the α22 protein itself. 

In 1996, Carter and Roizman identified a gene (denoted US1.5) that overlaps the US1 gene. ⁵⁶ This second gene encodes a protein containing the carboxyl terminal 273 residues of the α22 protein. This does not affect our findings, because both S34 and Y116 are contained only in the portion encoding the α22 protein. 

In summary, we have shown that tyrosine 116 of the HSV-1 α22 protein is required for peripheral virulence. In addition, alterations in virulence may be associated with altered tyrosine phosphorylation of the α22 protein. The data also suggest that at least two tyrosine phosphorylation sites exist in the protein and that Y116C may be one of these sites. Further studies of the role of tyrosine 116 of the α22 protein will provide critical information on the role of this residue on the biology of the virus and the impact of tyrosine phosphorylation on viral virulence. 

 Supported by National Eye Institute Grant EY07336 (CRB) and an unrestricted grant to the Department of Ophthalmology and Visual Sciences from Research to Prevent Blindness, Inc.

 Submitted for publication June 10, 2003; revised July 15, 2003; accepted July 17, 2003.

 Disclosure: C.R. Brandt, None; A.W. Kolb, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Diagram of the HSV-1 genome showing the location of the α22 and US1.5 open reading frames and the relevant mutations in the OD4 α22 protein. Line A: schematic of the HSV-1 genome denoting the locations of the US1/1.5 gene in the unique short (U_S) region. Line B: US1.5 and α22 open reading frames. Line C: location and nature of the revertant mutations in the OD4 α22 protein and the corresponding wild-type amino acids in CJ394. TR_L, terminal repeat long; U_L, unique long region; IR_L, internal repeat long; IR_S, internal repeat short; TR_S, terminal repeat short.

Figure 2.

 

Reversion of the OD4 C116Y and A34S/C116Y together restores ocular virulence. Site-directed mutagenesis was used to specifically correct the OD4 116 mutation or both the 34 and 116 mutations and the effect on ocular disease was tested in a marker transfer–infection assay. (A) The percentage of mice showing evidence of increased disease severity (20% is P = 0.05). (B) The MPDSs for corneal neovascularization (▪) and stromal keratitis ( ).

Table 1.

 

Statistical Analysis of Mean Peak Disease Scores for Marker Transfer/Infections

	Vascularization	Stromal Keratitis
OD4 vs. C116Y	0.000246	0.000086
OD4 vs. A34S+C116Y	0.000218	0.000160
C116Y vs. A34S+C116Y	0.924350	0.933913

Figure 3.

 

Immunoblot analysis of α22 protein isoforms in CJ394- and OD4-infected Vero cells. Vero cells were infected at an MOI of 5, and at 24 hours after infection the cells were harvested and lysed in Laemmli buffer. The samples were then serially diluted twofold (lanes 1-4), electrophoresed in a 10% denaturing polyacrylamide gel, and immunoblotted with RGST22 antibody. Extracts from equal numbers of cells were loaded in the comparable lanes 1 to 4 in the CJ394 and lanes 5 to 8 in the OD4 samples. Large open arrows: cluster of isoforms at 85 kDa. Large filled arrowheads: isoforms at 75 kDa; smaller filled arrows: 68-kDa isoform.

Figure 4.

 

Tryptic phosphopeptide mapping of α22 proteins. Infected cells were labeled with ³²[P]-orthophosphate from 1 hour 50 minutes to 5 hours after virus was added and harvested, and the proteins were electrophoresed in a 12% denaturing polyacrylamide gel. After exposure to x-ray film, the region containing the α22 protein was removed and digested with trypsin, and the resultant radiolabeled peptides were analyzed by two-dimensional electrophoresis/TLC. (A) CJ394 α22 protein. The peptides were given increasing numbers from top to bottom. Peptides were labeled with lowercase letters moving from left to right. (B) OD4 α22 peptides numbered as in (A). (C) shows the merged image and several peptides that differed between the two samples are labeled with the corresponding numbers from (A) and (B).

Figure 5.

 

Phosphotyrosine pulldown immunoblot analysis of α22 proteins. Infected cells were lysed and incubated with agarose coupled anti-phosphotyrosine antibody. After extensive washing, the proteins were electrophoresed and blotted with the α22-specific RGST22 antibody. The top band in the CJ394 sample migrated as a doublet of 85 kDa, with the top band being much more intense. The second band in the CJ394 lane is actually a doublet migrating at approximately 75 kDa. Note that this doublet is visible in the OD4 lane, but migrates slightly faster, suggesting the loss of at least one phosphate group.

The authors thank Inna Larsen for administrative assistance, and members of the laboratory for helpful suggestions concerning the studies and comments on the manuscript.