The U69 Gene of Human Herpesvirus 6 Encodes a Protein Kinase Which Can Confer Ganciclovir Sensitivity to Baculoviruses

Abstract

The protein encoded by the U69 open reading frame (ORF) of human herpesvirus 6 (HHV-6) has been predicted to be a protein kinase. To investigate its functional properties, we have expressed the U69 ORFs from both HHV-6 variants, A and B, by using recombinant baculoviruses (BV6AU69 and BV6BU69). Nickel agarose and antibody affinity chromatography was used to purify the proteins to homogeneity and when incubated with [γ-³²P]ATP, both U69 proteins became phosphorylated on predominantly serine residues. These data strongly suggest that U69 is a protein kinase which autophosphorylates. The phosphorylation reaction was optimal at physiological pH and low NaCl concentrations. It required the presence of Mg²⁺ or Mn²⁺, and Mg²⁺ was able to support phosphorylation over a wider range of concentrations than Mn²⁺. Both ATP and GTP could donate phosphate in the protein kinase assay and the former was more efficient. U69 was capable of phosphorylating histone and casein (serine/threonine kinase substrates) but not enolase (a tyrosine kinase substrate). For the autophosphorylation reaction, the Michaelis constants for ATP of baculovirus-expressed HHV-6A and HHV-6B U69 were calculated to be 44 and 11 μM, respectively. U69 is a homologue of the UL97 gene encoded by human cytomegalovirus which has been shown to phosphorylate the antiviral drug ganciclovir (GCV). We analyzed whether the U69 ORF alone was capable of conferring GCV sensitivity on baculoviruses BV6AU69 and BV6BU69. In plaque reduction experiments, these baculoviruses displayed a GCV-sensitive phenotype compared to a control baculovirus (BVLacZ). The 50% inhibitory concentrations (IC₅₀) of BV6AU69 and BV6BU69 were calculated to be 0.35 and 0.26 mM, respectively, whereas the control baculovirus had an IC₅₀ of >1.4 mM. This shows that the U69 gene product is the only one required to confer GCV sensitivity on baculovirus.

Human herpesvirus 6 (HHV-6) was first isolated from the peripheral blood of patients with lymphoproliferative disorders (1, 44). The DNA sequence of HHV-6 (strain U1102) has been published (20), and analysis has revealed that the overall arrangement of genes is similar to that of human cytomegalovirus (HCMV), with 66% of the DNA sequence showing homology (30); therefore, HHV-6 has been classified as a betaherpesvirus. As with all other human herpesviruses, primary infection is followed by lifelong persistence in the host. HHV-6 is widespread throughout the world’s population, with about 90% seroconverting within the first 2 years of life (40, 45, 52).

HHV-6 can be further subdivided into two distinct groups, variants A and B (17), and in infants, primary HHV-6B infection causes febrile illness including exanthem subitum (41, 51). The epidemiology and clinical consequences of HHV-6A infections are not as fully defined; however, primary HHV-6A infection has been reported in one case of exanthem subitum (25). HHV-6 infections have also been associated with encephalitis (4), hepatitis (3), multiple sclerosis, chronic fatigue syndrome, and some lymphomas, although a formal causal association has not been proven (5, 7, 31, 48). In individuals who are immunosuppressed by drugs, such as transplant patients, or individuals infected with human immunodeficiency virus, HHV-6 is emerging as an important pathogen. In bone marrow transplant recipients, both HHV-6A and HHV-6B have been associated with marrow suppression, pneumonitis, graft-versus-host disease, and encephalitis (9, 10, 18, 24, 28, 29, 43, 53). In individuals infected with human immunodeficiency virus, HHV-6 has been implicated as being a cofactor in disease progression (34).

Ganciclovir (GCV), a nucleoside analogue, is the first-choice antiviral agent used to treat HCMV infections. However, no drugs have been tested in a controlled manner to determine efficacy in treating HHV-6 infections. In vitro, the overall antiviral susceptibility of HHV-6 to a range of antiviral drugs is similar to that of HCMV (2, 8), and an interesting feature of one study is that GCV was found to be more effective against HHV-6A than against HHV-6B (50). If this result is confirmed, then it may have important implications for the treatment of HHV-6 infections.

The mechanism of action of GCV depends on the formation of the triphosphate form via sequential phosphorylation at the 5′-hydroxyl position. The UL97 protein kinase encoded by HCMV is responsible for the monophosphorylation of GCV, and resistance to the drug has been mapped to this gene (12, 22, 32, 33, 35, 47, 49). Sequence analysis revealed that HHV-6 codes for a homologue of UL97, the U69 gene, which has been implicated as a functional homologue (32, 11, 16). The U69 gene shows homology to protein kinases and belongs to a family of genes encoded by all herpesviruses, many of which have been shown to catalyze the transfer of phosphates onto target molecules or proteins (11, 13, 14, 16, 23, 26, 37–39, 42). The precise biological function performed by the herpesvirus kinases in the virus life cycle remains to be elucidated; however, their ability to phosphorylate proteins, many of which may have important regulatory roles in virus replication, makes them potential targets for antiviral intervention. Therefore, further information about this class of protein may facilitate the discovery of new and more effective antiherpesvirus compounds.

In order to characterize the U69 genes of both HHV-6A and HHV-6B, we have expressed them in heterologous expression systems and purified them to homogeneity by using polyclonal monospecific antisera. The data presented here demonstrate that the U69 gene can confer GCV sensitivity on baculovirus, is a protein kinase that autophosphorylates predominately on serine residues, and can catalyze the phosphorylation of exogenous substrates. In addition, we present evidence that the U69 proteins of HHV-6A and HHV-6B have differential protein kinase activities.

MATERIALS AND METHODS

Cells and viruses.

Spodoptera frugiperda 21 (Sf21) cells (Invitrogen) were maintained in TC100 medium (Life Technologies) supplemented with 10% fetal bovine serum, 50 IU of penicillin per ml, and 50 μg of streptomycin per ml. Wild-type linearized Autographa californica multiple nuclear polyhedrosis virus (AcMNPV; Invitrogen) was used to construct the recombinant baculoviruses. HHV-6A (AJ) and HHV-6B (Z29) were propagated in the J-Jhan and Molt-3 continuous T-cell lines, respectively. Both cell lines were maintained in RPMI medium (Life Technologies) supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 100 IU of penicillin per ml, and 100 μg of streptomycin per ml in 5% CO₂.

Cloning and sequencing of the HHV-6A and HHV-6B U69 ORFs.

Total DNA was extracted from HHV-6A- and HHV-6B-infected cells in culture by using the Puregene DNA isolation kit and used as a PCR template to generate full-length U69 open reading frames (ORFs). The primers (IP1-BamHI, dGATCGATGGATCCGAATAATTATGGACAACGGTGTGGA; IP2-HindIII, dTGCAGTCAAGCTTTCCATTACTATATCACATATGAAAG) were designed such that the PCR would generate an amplicon with a BamHI site at the 5′ position and a HindIII site at the 3′ position. To minimize Taq-generated errors, Bio-X-Act (Bioline) proofreading enzyme was used in the PCR as recommended by the manufacturer. The PCR generated a single DNA fragment approximately 1.7 kb long which was purified by standard phenol chloroform extraction and ethanol precipitation. The amplicons were then restricted with BamHI and HindIII and cloned into appropriately digested baculovirus transfer vector pBluBacHis (Invitrogen) downstream of a 4-kDa tag, generating plasmids pblu6AU69 and pblu6BU69 (Fig. 1). The N-terminal tag consists of six tandem histidine residues and an antibody recognition site (RGSHis). The clones were identified and distinguished by restriction enzyme analysis. To confirm that the U69 ORFs had been inserted downstream and in frame with respect to the fusion, the plasmid clones were DNA sequenced by the Sequenase version 2.0 protocol (United States Biochemical). In order to allow for expression in Escherichia coli, pblu6AU69 and pblu6BU69 were restricted with BamHI and HindIII and the DNA fragments were separated by agarose gel electrophoresis. The U69 ORFs were excised from the gel, purified via the Gene Clean method (Bio 101), and cloned into bacterial expression vector pTrcHisC (Invitrogen), producing the plasmids pTrc6AU69 and pTrc6BU69. Proteins expressed from these E. coli-based vectors are N-terminal fusions identical to the baculovirus transfer vectors described above.

FIG. 1.

Cloning of the HHV-6A and HHV-6B U69 proteins. (A) Schematic organization of the HHV-6 genome with terminal direct repeats (■) and a unique region (U). The scale is marked in kilobase pairs. (B) PCR primers IP1-BamHI and IP2-HindIII generated HHV-6A and HHV-6B U69 amplicons which had restriction endonuclease recognition sites at either end to allow subsequent cloning of the amplicons into appropriately digested baculovirus transfer vector pBluBacHis. (C) Recombinant baculoviruses (BV6AU69 and BV6BU69) were constructed containing the HHV-6 U69 gene driven by the polyhedrin promoter (➞). The U69 protein was expressed as an N-terminal antibody epitope (RGSHis) and polyhistidine [(His)₅] fusion.

Construction of recombinant baculoviruses.

From each of the transfer vectors (pblu6AU69 and pblu6BU69), a recombinant baculovirus was constructed by cotransfecting them individually with linear, wild-type AcMNPV DNA using Insectin-Plus liposomes (Invitrogen). By standard methodologies, recombinant baculoviruses BV6AU69 and BV6BU69 were isolated by two rounds of plaque purification and expanded into high-titer virus stocks (27). The insertion of the U69 gene into the baculovirus genome and the absence of contaminating wild-type AcMNPV were confirmed by PCR.

Preparation of U69-specific IgG.

A 5-liter Luria-Bertani (LB) broth (Sigma) culture of E. coli BL21 (Novagene) harboring pTrc6AU69 was grown to an A₆₀₀ of 0.6, and expression of the recombinant protein was induced by addition of isopropyl-β-d-thiogalactopyranoside (IPTG) to a final concentration of 0.5 mM. Cells were harvested by centrifugation after a 6-h postinduction period and resuspended in ice-cold lysis buffer A (50 mM NaH₂PO₄ [pH 8.0], 300 mM NaCl, 10 mM β-mercaptoethanol, 1 mM pefabloc [Boehringer Mannheim], 10 mg of lysosome per ml, 1% Triton X-100). The bacteria were incubated on ice for 1 h and sonicated (3 × 15 s), and soluble material was removed by ultracentrifugation at 30,000 × g. The E. coli-expressed U69 protein cosedimented with the insoluble fraction, which was then resuspended in ice-cold buffer B (50 mM NaH₂PO₄ [pH 8.0], 300 mM NaCl, 10 mM β-mercaptoethanol, 1 mM pefabloc [Boehringer Mannheim], 1.5% sarkosyl) and sonicated as described above. The supernatant was clarified by centrifugation at 30,000 × g and loaded onto a nickel-nitrilotriacetic acid (Ni²⁺-NTA [Qiagen]) column. The column was washed with excess ice-cold buffer B containing 10 mM imadizole, and bound protein was eluted with ice-cold buffer B containing 500 mM imadizole. This method resulted in a single species of protein upon analysis of the eluate by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie brilliant blue staining. The sheep antisera were prepared by the Scottish Antibody Production Unit. The initial immunization used 250 μg of antigen in Freund’s complete adjuvant. The animal was subsequently boosted a further three times with the same amount of antigen at 28-day intervals, and the final antiserum was collected on day 112. The serum was found to contain a low level of antibodies reactive against bacterial proteins, which were subsequently removed by incubating the antisera with immobilized protein extracts from E. coli cells harboring pTrcHis. The immunoglobulin G (IgG) component of the antisera was captured from the serum by using a protein G column (Pharmacia), eluted, and immobilized on CNBr-activated Sepharose 4B (Pharmacia) in accordance with the manufacturer’s methods.

Western analysis of protein expression.

Proteins were separated by SDS-PAGE and electrophoretically transferred onto a polyvinylidene difluoride (PVDF; Bio-Rad) membrane by using a semidry blotter (Pharmacia Biotech). The membrane was incubated for 1 h at room temperature in blocking buffer consisting of 3% bovine serum albumin (Sigma) in TBS (10 mM Tris-HCl [pH 7.5], 150 mM NaCl) and then washed with 0.05% Tween 20–0.1% Triton X-100 in TBS (wash buffer). The membrane was then incubated with the RGSHis (Qiagen) murine primary antibody or with U69-specific antisera for 1 h at room temperature, diluted to 1:1,000 or 1:10,000, respectively, in blocking buffer, and then further washed. Goat anti-mouse IgG (Bio-Rad) or goat anti-sheep IgG (Sigma) conjugated to alkaline phosphatase, diluted to 1:6,000 in blocking buffer, was used as the secondary antibody and incubated with the membrane for 1 h at room temperature. The membrane was further washed, and immunoreactive bands were visualized by incubation of the membrane in a staining solution consisting of one tablet of 5-bromo-4-chloro-3-indolylphosphate–nitroblue tetrazolium chloride (Sigma) dissolved in 10 ml of distilled H₂O.

Purification of baculovirus-expressed U69 protein.

Sf21 cells (6 × 10⁶) were infected with BV6AU69 or BV6BU69 at a multiplicity of infection (MOI) of 5 and incubated at 28°C. Infected cells were harvested at 72 h postinfection, and the culture medium was removed by centrifugation at 1,000 × g. The cell pellet was washed twice with 10 ml of phosphate-buffered saline and resuspended in 0.5 ml of ice-cold buffer C (50 mM Tris-HCl [pH 7.6], 100 mM NaCl, 5 mM MgCl₂, 0.1% Nonidet P-40, 10% glycerol, 10 μg of apoprotinin per ml, 10 μg of leupeptin per ml, 1 mM pefabloc). The cells were sonicated on ice for 10 s three times with 30 s of cooling in between. All subsequent procedures were carried out at 4°C. Insoluble material was removed by centrifugation at 13,000 × g in a desktop microcentrifuge for 10 min, and 50 μl of a 50% slurry of Ni²⁺-NTA pre-equilibrated with buffer C was added to the soluble extract and allowed to bind for 1 h with constant gentle agitation. The Ni²⁺-NTA was then pelleted in a benchtop microcentrifuge for 12 s at 13,000 × g and washed three times with excess buffer C containing 10 mM imadizole. Bound proteins were eluted by incubating the Ni²⁺-NTA in 100 μl of buffer C containing 500 mM imadizole. The enzyme was further purified by mixing the eluate with 100 μl (50% slurry pre-equilibrated with buffer C) of immobilized U69-specific IgG and allowed to bind for 1 h. The Sepharose beads were then washed three times with buffer C containing 0.5 M NaCl and twice with buffer C. To measure the amount of specific U69 complexed with the antibody, equal volumes of immunoprecipitate were subjected to SDS-PAGE alongside standards of known amounts of bovine serum albumin. The gel was silver stained and scanned by using a Bio-Rad Fluoroimager, the densities of the bands corresponding to the protein standards were measured by using Multianalyst software, and a standard curve was generated from this. Because the density of the band corresponding to the U69 protein was known, the amount of U69 protein that complexed with the antibody could be extrapolated from the standard curve.

Protein kinase assays.

Protein samples were added to 20 μl of kinase buffer (50 mM Tris-HCl [pH 9.0], 1 M NaCl, 10 mM MgCl₂, 2 mM dithiothreitol, 5 μM ATP, 5 μCi of [γ-³²P]ATP at >5,000 Ci/mmol [Amersham]) and incubated for 30 min, unless stated otherwise, with or without 1 mg of exogenous substrate per ml at 37°C. The reaction was terminated by the addition of 20 μl of 2 × SDS sample buffer (125 mM Tris-HCl [pH 8.0], 2.0% SDS, 10% sucrose, 0.01% bromophenol blue), and the mixture was boiled for 3 min. The reaction mixture was then subjected to SDS-PAGE, after which the gels were dried and exposed to Hyperfilm (Amersham). To quantify the amount of label incorporated, autoradiographs were digitized by using a Bio-Rad Fluoroimager, and the densities of appropriate bands were measured by using Multianalyst software (Bio-Rad). The Michaelis constant (K_m) for ATP was calculated by standard Lineweaver-Burke analysis. Briefly, 2-min protein kinase assays were performed in the presence of various total ATP concentrations. The reactions were terminated, and the reaction mixtures were separated by SDS-PAGE. The gel was exposed to film, and the amount of radiolabel incorporated for each ATP concentration was measured by densitometry. A Lineweaver-Burke graph of the inverse of the amount of radiolabel incorporated was plotted against the inverse of the corresponding ATP concentration, and the extrapolated intercept on the x axis (−1/K_m) determined the K_m value.

Phosphoamino acid analysis.

The recombinant protein was partially purified, and 100 ng of protein was phosphorylated as described above. The reaction mixture was subjected to SDS-PAGE and electrophoretically transferred onto PVDF. The membrane was wrapped in Saran film (to prevent drying), and phosphorylated U69 protein was located by exposing the membrane to X-ray film. The corresponding band was excised from the membrane, and bound protein was hydrolyzed by heating at 110°C in the presence of 6 M HCl for 60 min. The hydrolysate was transferred to a fresh microcentrifuge tube and concentrated in a Speedvac evaporator. After the liquid had evaporated, the dessicate was resuspended in 6 μl of water and vortexed vigorously for 5 min. Half of the sample was spotted onto a cellulose thin-layer chromatographic plate in 0.5-μl aliquots, which were allowed to dry between loadings. Samples (0.3 ng each) of phosphoserine, phosphotyrosine, and phosphothreonine (Sigma) were also spotted as standards. The plate was subjected to two-dimensional electrophoresis at 1.5 kV for 20 min in pH 1.9 (88% formic acid-acetic acid-water, 25:78:897) electrophoresis buffer in the first dimension and in pH 3.5 buffer (acetic acid-pyridine-100 mM EDTA-water, 50:5:5:940) for 20 min at 1.3 kV in the second dimension. The plate was developed with 0.2% ninhydrin in ethanol.

Plaque reduction assay.

To test whether the U69 protein can confer GCV sensitivity on AcMNPV, plaque reduction experiments were performed as follows. Sf21 cells were infected at an MOI of 0.3. After an absorption period of 1 h, the inoculum was removed and replaced with culture medium containing GCV (concentrations between 0 and 1.4 mM). After incubation at 28°C for 72 h, the culture medium was harvested and the virus was titrated by plaque assay. As a control, BVLacZ, which is a baculovirus containing the lacZ gene, was used. This was considered an appropriate control because BV6AU69 and BV6BU69 were constructed such that they also express β-galactosidase.

RESULTS

Recombinant-baculovirus expression of U69.

In order to produce recombinant U69 protein, the gene was cloned into the pBluBacHis baculovirus transfer vector and recombinant baculoviruses BV6AU69 and BV6BU69 were generated. ORFs cloned into the above baculovirus transfer vector are fused at the N terminus to a 4-kDa tag consisting of five tandem histidine residues and an antibody recognition site. In initial experiments, there was no antiserum available for U69 therefore expressed U69 was detected via RGSHis (Qiagen), which is an antibody directed to the N-terminal fusion. Although the U69 gene was placed under the control of the strong polyhedrin promoter, we found that U69 was expressed relatively poorly, even under optimized conditions (MOI of 5 and harvesting at 72 h postinfection [p.i.]). Nevertheless, under these conditions, a novel ∼67-kDa protein accumulated that had polyhedrin promoter kinetics and reacted with the anti-RGSHis antibody. This protein was not present in either uninfected or BVLacZ (baculovirus expressing β-galactosidase)-infected cells (Fig. 2A). Despite the use of a number of insect cell lines and conditions, U69 expression could not be enhanced.

FIG. 2.

Detection and phosphorylation of baculovirus-expressed U69. Insect cells were infected with recombinant baculovirus BV6AU69, BV6BU69, or BVLacZ and harvested at 24, 48, 60, and 72 h p.i. Protein extracts were prepared, incubated in a standard protein kinase assay (see Materials and Methods), and subjected to SDS-PAGE. The gels were then either analyzed by Western blotting (A) or exposed to autoradiography (B). The major phosphorylated species accumulated with polyhedrin promoter kinetics and comigrated with recombinant U69 as detected by Western blotting. The images shown here and in other figures are continuous two-tone images obtaining by scanning with a UMAX Astra 600S and Adobe Photoshop 4.0.

Protein kinase activity of U69.

To investigate whether HHV-6 U69 could catalyze the transfer of radiolabelled phosphates from ATP, BV6AU69- and BV6BU69-infected Sf21 cells were harvested at 24, 48, 60, and 72 h p.i. and U69 was partially purified. The protein preparations were subjected to a protein kinase assay containing [γ-³²P]ATP, followed by SDS-PAGE and autoradiography. A 67-kDa protein was found to be the major labelled species, which accumulated with polyhedrin promoter kinetics (Fig. 2B). The phosphorylated species comigrated with U69 detected by Western blotting (Fig. 2A), and no labelled species was apparent at this molecular mass when protein extracts from BVLacZ- and mock-infected cells were similarly assayed (Fig. 2B), indicating that the U69 protein was being phosphorylated.

To eliminate the possibility of the 4-kDa tag participating in the phosphorylation reaction, we similarly analyzed chloroamphenicol acetyltransferase (26 kDa) expressed by baculovirus (BV26kDa) as an identical fusion. A crude extract containing this protein was either used directly in a protein kinase assay or mixed with a partially purified sample of recombinant U69 and coincubated in a protein kinase assay. Neither situation resulted in a labelled product of 30 kDa (data not shown).

Transfer of GCV susceptibility to baculovirus.

To test whether introduction of HHV-6 U69 could confer GCV susceptibility on baculovirus, BV6AU69, BV6BU69, and BVLacZ (control) were cultured with increasing amounts of GCV. In preliminary experiments, concentrations of GCV above 2.0 mM were found to be cytotoxic (data not shown). Therefore, GCV concentrations between 0 and 1.4 mM were used. The amount of virus replication was measured by plaque assay after a 72-h incubation period, and titers were plotted against GCV concentrations as percentages of the amount of virus produced in the absence of GCV. The data shown in Fig. 3 illustrate that baculoviruses expressing HHV-6 U69 were more sensitive to GCV growth inhibition than was the control baculovirus. The 50% inhibitory concentrations (IC₅₀s) for BV6AU69 and BV6BU69 were calculated to be 0.35 and 0.26 mM, respectively, whereas the IC₅₀ of the control virus was estimated to be 2.2 mM by extrapolation.

FIG. 3.

Baculovirus plaque reduction assays. Baculoviruses expressing the U69 ORFs of HHV-6A and HHV-6B (BV6AU69, ■; BV6BU69, ●) were individually cultured in increasing concentrations of GCV (0 to 1.4 mM), and the effect of the drug on virus replication was measured by plaque assay. The inhibitory effect of GCV was greater for the U69-expressing baculoviruses than for the control (BVlacZ, ▴). A slight reduction in plaque formation was observed for BvlacZ; however, this was at GCV concentrations near the cytotoxic concentration. The IC₅₀s were calculated to be 0.35 and 0.26 mM for BV6AU69 and BV6BU69, respectively. The IC₅₀ for the control virus could not be determined within this data set. The values shown are means of three independent experiments, each performed in duplicate.

Purification of U69.

Although we increased the specific activity of U69 in a partially purified protein preparation compared to the specific activity of U69 in a crude extract (data not shown), we were unable to purify it further by Ni²⁺-NTA chromatography. Consequently, we expressed HHV-6 U69 in E. coli by using the pTrcHisC vector and found that U69 was expressed almost exclusively as insoluble inclusion bodies. Although large amounts of U69 were purified from this system, the enzyme was not active in the protein kinase assay and did not phosphorylate GCV in vitro (data not shown). However, the protein was used to immunize sheep and produce antisera which could recognize the baculovirus-expressed protein. The IgG component from the antisera was immobilized on CNBr-activated Sepharose 4B and used to affinity purify baculovirus-expressed U69 from partially purified preparations. This approach (using first Ni²⁺-NTA and then immunoaffinity chromatography) allowed us to produce highly purified baculovirus-expressed U69 as demonstrated by silver staining (Fig. 4).

FIG. 4.

Purification of BV6AU69- and BV6BU69-expressed U69. Aliquots of protein samples were taken from each step of the purification procedure and analyzed by SDS-PAGE followed by silver staining. Lanes 1 and 5 represent crude lysates of insect cells infected with BV6AU69 or BV6BU69, respectively. Eluates of the Ni²⁺-NTA affinity purification step (lanes 2 and 6) were then immobilized by immunoaffinity chromatography, and samples were loaded on lanes 3 and 7. Lanes 4 and 8 represent aliquots of immobilized IgG. The molecular sizes of the protein markers (lane M) are indicated on the left.

To test for a difference in phosphotransferase activity between the HHV-6A and HHV-6B U69 proteins, equal amounts of specific U69 (10 ng) were subjected to a protein kinase assay and phosphorylation was detected by autoradiography (Fig. 5). This analysis revealed that (i) the homogeneous U69 protein product incorporated radiolabelled phosphates, (ii) the HHV-6A and HHV-6B U69 proteins had a subtle difference in electrophoretic mobility, and (iii) the protein band corresponding to HHV-6B U69 appeared to be more dense than the band corresponding to HHV-6A U69. Enzyme kinetic analysis of these proteins with respect to ATP revealed that HHV-6B U69 had a higher affinity for ATP than did U69 of HHV-6A (K_ms, 11 μM for HHV-6B U69 and 44 μM for HHV-6A U69).

FIG. 5.

Phosphorylation of purified recombinant U69. Equal amounts of U69, purified from BV6AU69- or BV6BU69-infected insect cells, were incubated with [γ-³²P]ATP in a protein kinase assay and subjected to SDS-PAGE (lanes 2 and 3, respectively). The gel was dried and exposed to film. The resulting autoradiograph is shown. Lane 4 is immobilized U69-specific IgG used for immunoprecipitation of BVLacZ-infected insect cells and subjected to a protein kinase assay as described above. The molecular sizes of the protein markers (lane M) are indicated on the left.

As a final test to determine if U69 was being phosphorylated by a baculovirus or insect cell protein kinase, purified U69 was inactivated by exposure to low pH and coincubated in a standard protein kinase assay with protein preparations from uninfected or BV26kDa-infected cells. Neither situation resulted in a phosphorylated product with a molecular weight corresponding to that of U69 (data not shown).

Phosphoamino acid analysis.

From sequence analysis of the U69 ORF and protein kinases, we have located the catalytic domains of U69 (data not shown). Of interest are domains VI (D₃₁₃ISPMN) and VIII (F₃₇₃NPGFRPL), which resemble the consensus sequence of a protein kinase that has specificity for serines and threonines more closely rather than a protein kinase that has specificity for tyrosine residues (21). To determine U69 hydroxyamino acid specificity experimentally, partially purified U69 was phosphorylated in a kinase assay and subjected to SDS-PAGE. The proteins were transferred onto a PVDF membrane, and the band corresponding to U69 was excised and subjected to phosphoamino acid analysis. The results suggested that the HHV-6A and HHV-6B U69 proteins are predominantly phosphorylated on serine residues with a smaller degree of phosphorylation on threonine residues (Fig. 6). No labelled species corresponding to phosphotyrosine was observed.

FIG. 6.

Phosphoamino acid analysis of recombinant U69. U69 was partially purified with Ni²⁺-NTA and phosphorylated in vitro in the presence of [γ-³²P]ATP. The proteins were separated by SDS-PAGE and transferred electrophoretically onto a PVDF membrane, and the band corresponding to phosphorylated U69 was excised and acid hydrolyzed. The hydrolysate was mixed with unlabelled phosphoamino acids, phosphoserine (P-SER), phosphothreonine (P-THR), and phosphotyrosine (P-TYR) and subjected to two-dimensional electrophoretic thin-layer chromatography. The unlabelled phosphoamino acids were visualized with ninhydrin, and the autoradiograph of the thin-layer chromatography plate for the phosphoaminoacid analysis of HHV-6A U69 is shown. Phosphoamino acid analysis of HHV-6B U69 revealed a similar autoradiograph. The labelled phosphoamino acids comigrated with P-SER and P-THR, and the migration of P-TYR is indicated by the dotted circle.

Biochemical analysis of autophosphorylation.

The velocity of the autophosphorylation reaction with respect to time, by using first ATP as the phosphate donor and then GTP, was investigated. Our results indicate that U69 could use both nucleoside triphosphates as the phosphate donor, although ATP was favored (Fig. 7A and B). The autophosphorylation was found to be linear for approximately 15 min and reached a maximum after approximately 40 min. The subsequent biochemical analysis (with respect to salt dependence, divalent cation specificity and concentration, and pH) were therefore performed by using 2-min reactions in order to measure initial rates. Phosphorylation was maximal at physiological pH (7 to 7.5) (Fig. 8A) and tolerated a wide range of Mg²⁺ concentrations (0.2 to 100 mM were tested) with 60 to 80% of the activity still remaining at 100 mM (Fig. 8B). Mn²⁺ could substitute as a divalent cation (Fig. 8C) with a more restricted range (0.2 to 10 mM) and an optimum concentration of 2 mM. Ca²⁺ concentrations between 1 and 10 mM did not support phosphorylation. The optimum NaCl concentration required was approximately 0.5 M (Fig. 8D), with 50% activity still remaining at 1.5 M. The HHV-6A and HHV-6B U69 proteins had similar biochemical characteristics.

FIG. 7.

Velocity of autophosphorylation of recombinant HHV-6A (A) and HHV-6B (B) U69 with respect to time. Protein kinase assays were performed for the times indicated by using either ATP (■) or GTP (○) as the radiolabelled phosphate donor. The reaction was terminated and subjected to SDS-PAGE. The amount of radiolabelled phosphate incorporation was measured by densitometry of the autoradiographs.

FIG. 8.

Biochemical characteristics of HHV-6A (■) and HHV-6B (○) U69. Baculovirus-expressed U69 was subjected to a standard protein kinase assay using radiolabelled ATP as the phosphate donor, except that the pH (A), divalent cation concentration (B and C), and NaCl concentration (D) were varied as indicated. In order to measure initial rates, 2-min reactions were performed. Autophosphorylation was measured for each condition by densitometry.

Phosphorylation of exogenous substrates.

To test the ability of the protein kinase associated with HHV-6 U69 to catalyze the phosphorylation of exogenous substrates, histones, casein (serine/threonine protein kinase substrates), or enolase (a tyrosine protein kinase substrate) was mixed individually with purified U69 and incubated with [γ-³²P]ATP in a protein kinase assay. The fact that only histones and casein were phosphorylated (Fig. 9) is further evidence that U69 is a protein kinase that has specificity for serine and threonine residues.

FIG. 9.

Phosphorylation of exogenous substrates. Purified U69 was coincubated in a protein kinase assay with exogenous proteins. The reaction was terminated, samples were subjected to SDS-PAGE, and the gel was exposed to film. The autoradiograph shows purified U69 from BV6AU69-infected insect cells incubated either alone (lane 2) or with histone casein or enolase (lanes 3 to 5) and purified U69 from BV6BU69-infected insect cells incubated either alone (lane 6) or with the exogenous substrates indicated (lanes 7 to 9). The positions and molecular sizes of marker proteins (lane M) are shown on the left side, and the migration of the exogenous substrates is shown on the right side. Lanes 10 to 12 are the exogenous substrates incubated in the absence of U69 in a standard protein kinase assay.

DISCUSSION

It has been known for a number of years that HCMV infection in individuals undergoing organ transplantation is associated with significant morbidity, and there is increasing evidence that HHV-6 is also as an important pathogen in such individuals (48). HCMV infections are usually treated with GCV, which has been shown to cause a rapid reduction in viremia (6). It has been found (2, 8, 50), albeit in vitro, that HHV-6 replication can also be controlled with GCV; however, the precise molecular mechanism which governs the susceptibility of HHV-6 to the drug has yet to be investigated. The UL97 protein of HCMV has been shown to direct the phosphorylation of GCV in HCMV-infected cells. We describe here for the first time the molecular characterization of the protein encoded by the U69 ORFs of both HHV-6A and HHV-6B, sequence homologues of the HCMV UL97 ORF. Our results indicate that U69 is a protein kinase that autophosphorylates on serine or threonine residues and can also confer GCV sensitivity on baculovirus.

When partially purified preparations of baculovirus-expressed U69 were incubated with [γ-³²P]ATP, a radiolabelled species was apparent which comigrated with U69 detected by Western analysis, indicating that U69 may have autophosphorylating properties. It was possible that this activity was either inherent within the U69 protein itself or it was a substrate for a cellular or baculovirus-encoded protein kinase. We favored the former for the following reasons. U69 is a sequence homologue of a family of proteins encoded by all known herpesviruses, many of which have been shown to have phosphotransferase activity. For example, herpes simplex virus type 2 UL13 has been purified and shown to phosphorylate casein (15), and herpes simplex virus type 1 UL13 has been reported by several investigators to exhibit similar properties (14, 26, 42). HCMV UL97 and pseudorabies virus UL13 have been expressed in heterologous expression systems, and both have been shown to catalyze autophosphorylation in an in vitro kinase assay (16, 22). Although the varicella-zoster virus ORF 47 showed no phosphotransferase activity when produced in a heterologous expression system, it has been reported to autophosphorylate when immunoprecipitated from varicella-zoster virus-infected cells (37, 38). Cellular kinases, with the exception of casein kinase, are specific in the type of nucleotide triphosphate they utilize to catalyze a phosphotransferase reaction. The herpesvirus kinases studied to date, like casein kinase, are unique in that they can utilize both ATP and GTP as a phosphate donor, and our data show that U69 can also utilize both as phosphate donors. Nevertheless, it remains that U69 was serving as a substrate for phosphorylation by a cellular or baculovirus-encoded protein kinase. Subsequently, to investigate whether U69 could incorporate radiolabelled phosphates in the absence of contaminating proteins, we highly purified U69 and found that the phosphotransferase activity remained. Albeit unlikely, it may be possible that a contaminating kinase was copurified with preparations of U69. To eliminate this possibility, purified U69 was inactivated and used as a substrate for phosphorylation by cellular or baculovirus-encoded proteins, and neither situation resulted in the phosphorylation of U69. Since protein kinases recognize a linear sequence as the target for phosphorylation and synthetic peptides are often used as substrates in experimental procedures, it follows that if the phosphorylation of U69 was being carried out by a copurified kinase, then radiolabelling of U69 would be expected, irrespective of the physical state of U69. Taken together, these data show that U69 is a protein kinase that autophosphorylates and is a member of the family of kinases encoded by all known herpesviruses.

Despite the fact that they are members of the same family, there were significant differences in the biochemical properties of U69 and those reported for UL97 (22). Firstly, we found the U69 had a preference for low NaCl concentration (0.5 M), whereas the HCMV UL97 has optimal activity at 1.5 M NaCl. Secondly, maximal U69 activity was observed at physiological pH, contrasting with HCMV UL97, which has a pH optimum of 9.5. The only biochemical characteristic shared by U69 and UL97 was that both magnesium and manganese could support phosphorylation. A possible explanation for these apparent differences is that the proteins may play different roles in the viral life cycle, including distinct viral or cellular substrates.

The majority of protein kinases can be classified into two categories, depending on their hydroxyamino acid specificity: (i) kinases that generate phosphate monoesters utilizing protein alcohol groups (on serine or threonine) and (ii) ones that utilize protein phenolic groups (on tyrosine) as phosphate acceptors. By analyzing the sequences of subdomains VI and VIII, we were able to predict the hydroxyamino acid specificity of U69 to be serine and threonine. We have confirmed this experimentally and shown that the majority of the autophosphorylation occurs at serine residues, with threonine residues being phosphorylated to a much lesser extent. While the overall conservation of domains VI and VIII between the herpesvirus kinases is low (11), all of the herpesvirus U69 homologues studied to date have been shown to be serine/threonine protein kinases (15, 22, 37), with the exception of pseudorabies virus UL13, which has specificity for only serine residues (16). This points to the relatively few amino acids that are conserved within these domains. Of particular interest are the invariant proline (within domain VI) which is 100% conserved in serine/threonine protein kinases and the phenylalanine (within domain VIII) which is highly conserved in serine/threonine protein kinases, both of which can be found in the U69 ORF at amino acid positions 313 and 373, respectively. These amino acids may be considered essential in determining the hydroxyamino acid specificity of U69 and are therefore candidates for site-directed mutagenesis to investigate functional domains of the protein.

We compared the activities of the U69 proteins from HHV-6A and HHV-6B by using equal amounts of the proteins and observed that autophosphorylation of HHV-6B U69 was more extensive than that of HHV-6A U69. HHV-6B U69 also had increased electrophoretic mobility, which has been reported to be a direct consequence of protein phosphorylation (20). These data suggest that HHV-6A U69 was either less efficient at autophosphorylation or had fewer phosphorylation sites than HHV-6B U69. To investigate this enzymologically, enzyme kinetic analysis of the rate of phosphorylation showed that the K_m of HHV-6A U69 for ATP was fourfold higher than that of HHV-6B U69. Therefore, these observations indicate that U69 from HHV-6A is less efficient at autophosphorylation than U69 from HHV-6B at the level of the U69-ATP interaction. It is not inconceivable that a difference in protein kinase activity exists between the U69 proteins because HHV-6A and HHV-6B are distinct in biology (reviewed in the introduction). DNA sequencing of the HHV-6A and HHV-6B U69 ORFs did reveal differences, but surprisingly, mutations were not found within the catalytic domains of the U69 ORFs. The catalytic domains of protein kinases are important in substrate specificity and binding (21), but other regions that lie outside those may be important in stabilizing the overall structure of the protein and hence contribute to the catalytic activity. Therefore, it follows that the difference in protein kinase activity observed between the HHV-6A and HHV-6B U69 proteins is likely due to subtle differences in structural stability.

The expression of U69 using a recombinant baculovirus which is otherwise relatively insensitive to GCV at noncytotoxic concentrations renders viral replication sensitive to the drug. This suggests that the U69 gene product can confer GCV sensitivity on baculovirus independently of other HHV-6 proteins. The simplest explanation for the mechanism of GCV phosphorylation is that U69 phosphorylates the drug by a direct molecular interaction between the drug and the protein. Our data alone cannot exclude the alternative, which is that U69 in some way activates a cellular or baculovirus protein which results in GCV phosphorylation as an end product. Nevertheless, our data do demonstrate that U69 controls GCV phosphorylation in insect cells and, by implication, may carry out a similar function in HHV-6-infected cells. The system described here will be very helpful to test the influence of U69 mutations on the phosphorylation of not only GCV but also other thymidine kinase-dependent drugs.

Since the herpesvirus kinases have been postulated as being required for efficient viral replication (4, 11, 12, 36, 38, 42), further studies are required to identify the cellular or viral protein targets for U69 phosphorylation. Identification of the catalytic residues required for its protein kinase and GCV kinase activities would be important, since independent catalytic sites may facilitate the development of novel U69 protein kinase inhibitors which could be used in conjunction with GCV as a therapeutic approach for HHV-6.