Human Cytomegalovirus UL84 Oligomerization and Heterodimerization Domains Act as Transdominant Inhibitors of oriLyt-Dependent DNA Replication: Evidence that IE2-UL84 and UL84-UL84 Interactions Are Required for Lytic DNA Replication

Cells and viruses.

Telomerized human foreskin fibroblasts (T-HFs) (31a) were used for transient-replication assays and were propagated in Dulbecco modified Eagle medium (DMEM) supplemented with 10% (vol/vol) fetal calf serum. The 293 packaging cell line (Clontech) that was used to produce UL84 recombinant adenovirus was propagated in DMEM supplemented with 10% (vol/vol) fetal calf serum. Cos7 cells and Vero cells were propagated in DMEM supplemented with 10% (vol/vol) newborn calf serum. AD169 (ATCC V-538) was propagated as described previously (25).

Construction of a recombinant adenovirus expressing UL84.

Ad84 virus was generated by using AdEasy Adenoviral Vector System (Stratagene). AD169 genomic DNA was used as a template for amplifying the UL84 ORF. PCR was performed with the following primers: 5′-CCGGTACCACACCAAGCATGCCACGCGTCGACCCCAACCT-3′ and 5′-CGCTCGAGCTTAGCCCTACTTATCGTCGTCATCCTTGTAATCGAGATCGCCGCAGACCATGGCTAA-3′.The amplified UL84 open reading frame (ORF) was generated such that the FLAG sequence was fused in the same reading frame context as UL84. The PCR product was cleaved with KpnI and XhoI and ligated into pShuttle-CMV vector (Stratagene). Recombinant adenovirus was generated as recommended by the manufacturer. Virus titers were determined with a 293 packaging cell line.

DNA constructs.

pUL84-TAG(1-586), pUL84-TAG(68-586), pUL84-TAG(83-586), pUL84-TAG(104-586), pUL84-TAG(141-586), pUL84-TAG(201-586), pUL84-TAG(351-586), pUL84-TAG(1-513), and pUL84-EGFP were described previously by Xu et al. (32). The pUL84-HA expression plasmid was made by PCR amplification of AD169 DNA with the primers 5′-CggcgcACACCAAGCATGCCACGCGTCGACCCCAA-3′ and 5′-GGCGTCTTATCAAGCGTAGTCTGGGACGTCGTATGGGTAGAGATCGCCGCAGACCATGGC-3′, which was subsequently ligated into pTarget (Promega).

The pUL84EXPRESS expression plasmid was made by PCR amplification of AD169 DNA with the primers 5′-GcgccgaattcCCAAGCATGCCACGCGTCGACCCC-3′ and 5′-CgcggtcctcgagcTCTGTTTAGAGATCGCCGCAG-3′, which was subsequently digested with EcoRI and XhoI and ligated into pCDNA3.1+ (Invitrogen).

All UL84 peptide subclones were made by PCR amplification of AD169 DNA followed by ligation of the resulting product into pTarget (Promega). PCR products were generated such that the FLAG or hemagglutinin (HA) epitope and the simian virus 40 (SV40) nuclear localization sequences (NLS) were fused in the same reading frame context as the UL84 ORF. The following primers were used for UL84 subclones: p84FR(1-100) (5′-CGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCCAAGCATGCCACGCGTCGACCCCAACCTTCGGAA-3′ and 5′-TTAGCCCTActtatcgtcgtcatccttgtaatcAGGCTGACGGATGTCGTGTTTATCCGT-3′), p84FR(101-200) (5′-CGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCCAAGCATGCGGGTGCACCGCGGCACCTACCATCT-3′ and 5′-TTAGCCCTActtatcgtcgtcatccttgtaatcGAAGGCCGGGCGCACGTCAGAATCCTC-3′), p84FR(201-300) (5′-CGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCCAAGCATGTCTCTCTTTCCCGCACGCCCAGGCTG-3′ and 5′-TTAGCCCTActtatcgtcgtcatccttgtaatcCGTGGGCGTCAGATCCGCGTCCGTTGA-3′), p84FR(301-400) (5′-CGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCTGACGGTGCGCGTACGACACGC-3′ and 5′-TTAGCCCTACTTATCGTCGTCATCCTTGTAATCTCGCACCGTTTCGCGTCGCAGCCGCAGGC-3′), p84FR(401-500) (5′-CGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCGACCCTTCTTTTCGGACGCGCC-3′ and 5′-TTAGCCCTACTTATCGTCGTCATCCTTGTAATCAAAATAGAGGCGCCCCAGGAGCGCGTC-3′), p84FR(501-586) (5′-CGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCCAAGCATGATCTCGTCAAAGCATACGCTGAATCG-3′ and 5′-TTAGCCCTActtatcgtcgtcatccttgtaatcGAGATCGCCGCAGACCATGGCTAAAGT-3′), and p84OD(151-300) (5′-CGGTGCCACCATGCCTCCAAAAAAGAAGAGAAAGGTCCcgaattcacaccaagcATGGAAGAAGAAACGGCG-3′ and 5′-TTAGCCCTActtatcgtcgtcatccttgtaatcCGTGGGCGTCAGATCCGCGTCCGTTGA-3′).

The IE2 cDNA expression vector, pON2206, was a gift from Edward Mocarski (Stanford University).

All IE2 subclones were made by PCR amplification of IE2 cDNA with pON2206 as a template and ligating the resulting product into pTarget vector. PCR products were generated such that the HA and SV40 NLS sequences were fused in the same reading frame context as the IE2 subclone. The following primers were used for IE2 subclones: pIE2-HA(1-215) (5′-CggtgccaccatgCCTCCAAAAAAGAAGAGAAAGGTCatgGAGTCCTCTGCCAAGAGAAA-3′and 5′-ggcgtcttatcaagcgtagtctgggacgtcgtatgggtaGGTTTCGACTTCTTCACCCTG-3′), pIE2-HA(216-386) (5′-cggtgccaccatgCCTCCAAAAAAGAAGAGAAAGGTCACCCGCGGTGCTACCGCGTCTTC-3′ and 5′-ggcgtcttatcaagcgtagtctgggacgtcgtatgggtaGGGAATCGTGAAGGGCAGGTT-3′), and pIE2-HA(387-580) (5′-cggtgccaccatgCCTCCAAAAAAGAAGAGAAAGGTCCCCAGTATGCACCAGGTGTTAGA-3′ and 5′-ggcgtcttatcaagcgtagtctgggacgtcgtatgggtaCTGAGACTTGTTCCTCAGGTC-3′).

The construct pUL112/113pr, containing the UL112/113 promoter upstream of the luciferase gene, was generated by PCR amplification with AD169 DNA as a template by using the primers 5′-CCGGTACCAGATCTCCGCGTCACCTTTCATCGAGT-3′ and 5′-CCAAGCTTGGCCGTGGAGCGAGTGCCGCCGCAGCC-3′. The 750-bp PCR product was then cleaved with KpnI and HindIII and subcloned into the luciferase reporter vector pGL2-Basic (Promega).

pEGFP-N1 (Clontech) and pFLAG-control (Stratagene) expression constructs were utilized according to manufacturer's recommendations.

Site-directed mutagenesis.

UL84 leucine zipper mutants were generated by using QuickChange Multi-Site-Directed mutagenesis kit (Stratagene). Construct pUL84 L1+L2, containing valine residues in place of leucine residues at aa 114 and 121, was made by using the mutagenesis primers 5′-ACCTACCATCTGATCCAGTTGCACGTCGACCTCCGACCCGAAGA-3′ and 5′-GACCTCCGACCCGAAGAAGTGCGGGATCCCTTCCAGATTCTG-3′. Construct pUL84 L2+L3, containing valine residues in the place of leucine residues at aa 121 and 128, was generated by using the mutagenesis primers 5′-GACCTCCGACCCGAAGAAGTGCGGGATCCCTTCCAGATTCTG-3′ and 5′-CGGGATCCCTTCCAGATTCTGGTCTCTACGCCGCTGCAA-3′. Construct pUL84 L3+L4, containing valine residues in place of leucine residues at aa 128 and 135, was generated by using the mutagenesis primers 5′-CGGGATCCCTTCCAGATTCTGGTCTCTACGCCGCTGCAA-3′ and 5′-TCTACGCCGCTGCAAGTGGGGGAAGCGAACGACGAGTCT-3′.

UL84 leucine zipper mutant subclone p84-2LZA was made by PCR amplification of a UL84 fragment with pUL84 L1+L2 DNA as a template and ligating the resulting product into pTarget vector. The primers used for the construction of p84-2LZA are 5′-CCAAGCATGCGGGTGCACCGCGGCACCTACCATCT-3′ and 5′-TTAGCCCTActtatcgtcgtcatccttgtaatcGAAGGCCGGGCGCACGTCAGAATCCTC-3′.

All constructs were confirmed by DNA sequence analysis.

Luciferase assay.

U373 cells were cotransfected with a construct containing the UL112-113 promoter, pUL112/113pr, the IE2 expression construct, pON2206, and pUL84-TAG(1-586) or UL84 leucine zipper mutants. Cells were lysed 2 days posttransfection, and the luciferase activity was determined with a luminometer according to the manufacturer's instructions.

UL84 protein purification.

Vero cells were plated to 70 to 90% confluency in 10 100-mm dishes. Cells were infected with Ad84 at a multiplicity of infection of 5. Cells were incubated for 48 h, at which time the cells were washed twice with phosphate-buffered saline (PBS; pH 7.4) and sonicated in 1 ml of lysis buffer A (50 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1 mM EDTA, 0.5% NP-40, 2 mM MgCl, 0.2 mM EDTA, 10 μl of protease inhibitor cocktail σ[/ml]) with shaking for 30 min at room temperature. Cells were removed from the plate by scraping and passed through a 22-gauge needle three times to shear DNA. Cellular debris was removed by centrifugation at 1,500 × g for 10 min. The resulting lysate was passed through a low-protein-binding 2-μm-pore-size filter (Millipore) before incubation with FLAG affinity beads (Sigma). Protein purification proceeded according to the manufacturer's recommendations for column purification. Protein was eluted with 100 μg of 3X FLAG peptide (Sigma)/ml. Fractions were analyzed by separation on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel and visualized by using Coomassie blue or reacted with antibody in a Western blot. Fractions shown to contain the most protein were concentrated by using an Amicon Ultra-4 30,000 MWCO (Millipore) and analyzed by Western blot before use.

Generation of a UL84 monoclonal antibody.

A UL84 monoclonal antibody (MAb84) was produced by using purified recombinant UL84 protein by the method described in reference 3. All work was performed at the University of Nevada Monoclonal Antibody Core Facility. Antibodies were screened by using purified UL84 recombinant protein as an antigen in an enzyme-linked immunosorbent assay. Several positive hybridoma clones were isolated, the clone used to produce purified antibody was chosen based on relative affinity to antigen and cell line viability. This clone was used to produce large quantities of secreted antibody that was purified by using a protein A column. After the antibody was isolated, the isotype of the antibody was determined to be immunoglobulin G2a.

Antibodies.

The IE2-specific antibody G13-12E2 (Vancouver Biotech), anti-FLAG M2 (Sigma), and anti-HA (Sigma) antibodies were used for coimmunoprecipitation assays.

Coimmunoprecipitation assay.

Cos7 cells were plated to 70 to 90% confluency in 100-mm dishes. Cells were transfected with 10 μg of the appropriate plasmids by using Transfectin reagent (Bio-Rad) per manufacturer's recommendations. At 48 h posttransfection cells were washed twice with PBS (pH 7.4) and lysed by using 1 ml of lysis buffer A by shaking for 30 min at room temperature. Cells were scraped from the plate and passed through a 22-gauge needle three times. Cellular debris was removed by centrifugation at 1,500 × g for 10 min. Lysates containing expressed proteins were mixed together (capture assay) for 1 h at 4°C before incubation with specific antibodies. Coimmunoprecipitations were carried out by using a conjugated or nonconjugated immunoprecipitation system.

(i) Antibody-conjugated coimmunoprecipitation.

Anti-FLAG M2 Affinity Gel Freezer-Safe beads σ were prepared according to the manufacturer's recommendations. Portions (50 μl) of the prepared beads were added for each coimmunoprecipitation and incubated at 4°C overnight. The complexes were washed according to the manufacturer's recommendations, and proteins were eluted by using 100 μg of 3X FLAG peptide/ml (and analyzed by Western blotting).

(ii) Antibody-nonconjugated coimmunoprecipitation.

Lysates were incubated with the appropriate monoclonal antibody for 1 h at 4°C, at which time 40 μl of protein G plus agarose beads were added. The coimmunoprecipitation was again incubated at 4°C overnight. The complexes were washed with ice-cold PBS (pH 7.4) four times. The protein complexes were removed from the beads by the addition of 2X Laemmli sample buffer (Bio-Rad) containing 2-mercaptoethanol and then heated to 95°C for 5 min. Samples were separated by SDS-PAGE and analyzed by Western blot.

For experiments involving full-length UL84 or IE2 expression, constructs were cotransfected into Cos7 cells, followed by incubation of cell lysates by either the conjugated or nonconjugated coimmunoprecipitation method.

Transdominant-negative inhibition of oriLyt-dependent DNA replication.

For transdominant-negative inhibition, telomerized HFs were cotransfected with 1 μg of each plasmid encoding each of the human herpesvirus 8 (HHV-8) core replication proteins under the control of the SV40 promoter (1) and 1 to 2 μg of plasmids expressing various UL84 or IE2 peptides and UL84 and IE2 expression constructs, pUL84-TAG(1-586) and pON2206, and 10 μg of the pGEM-oriLyt, which contained the HCMV cloned oriLyt (23). Total cellular DNA was harvested 5 days posttransfection, and replication products were analyzed as described previously (23, 32). Control transfections contained the same mixture as described above except that instead of including plasmids that expressed UL84 or IE2 peptides, we substituted the parent expression plasmids for these constructs.

For the analysis of UL84 leucine zipper site-directed mutants, cotransfections contained plasmids pUL84 L1+L2, pUL84 L2+L3, or pUL84 L3+L4 substituted for UL84 native expression constructs, and the replication assay was performed as described previously (32).

In vitro cross-linking.

UL84 protein fragments were expressed by using T_NT Quick-Coupled Transcription/Translation Kit (Promega). First, 10 μl of translated protein was incubated in a final concentration of 0.008% glutaraldehyde in 0.1 M NaCl for 30 min at room temperature. Then protein was separated through a SDS-12% PAGE gel, fixed, treated with Amplify (Amersham) for 30 min, and exposed to X-ray film for 15 h.

Expression and purification of UL84.

As a first step toward the characterization of UL84 and its interaction with other viral proteins, we developed a system to obtain relatively pure protein from mammalian cells. We chose to isolate UL84 from mammalian cells, as opposed to purification from a bacterial system, because it is not known if the lack of posttranslational modifications would effect protein-protein interactions. We generated a recombinant adenovirus expression construct such that the UL84 ORF was fused in frame with the FLAG epitope. This FLAG tag allowed for affinity purification of UL84 and also provided a mechanism to track the protein during the purification procedure.

High-titer UL84 recombinant adenovirus was produced by using 293 packaging cells. Virus was harvested and Vero cells were infected at a multiplicity of infection of ∼5. Infection was allowed to proceed for 48 h, at which time cells were harvested, and protein lysates were prepared. FLAG-tagged UL84 was purified by incubation of protein lysates with FLAG affinity beads, followed by a wash procedure. UL84 was then eluted by competition with 3X FLAG peptide. Initial protein lysates and column flowthrough samples are shown in Fig. 1 (Fig. 1, lanes 1 and 2, respectively). The column was washed until the flowthrough was cleared of almost all protein (Fig. 1, lane 3). UL84-FLAG protein was then eluted in five fractions (Fig. 1, lanes 4 to 8). The UL84-FLAG protein, which migrates at ca. 75 kDa, was collected and analyzed by Western blotting with the anti-FLAG antibody (data not shown). This procedure indicated that we could isolate UL84 recombinant protein that could be used in further studies.

Generation of a monoclonal antibody to UL84 purified protein.

In order to study UL84 and its interaction with itself and other virus-encoded proteins, we required a highly specific antibody. After purifying UL84 from Vero cells infected with an adenovirus expressing FLAG-tagged UL84 protein, we used this purified protein as antigen to produce a monoclonal antibody (see Materials and Methods). As an initial evaluation of this antibody, MAb84, we reacted it with various UL84 expression constructs and HCMV-infected cell protein [Fig. 2A, lanes 1-586(FL), 84-EGFP, and AD169]. MAb84 failed to interact with the FLAG control plasmid (Fig. 2A, lane TAG-Control). In order to determine the approximate location of the point of interaction of MAb84 with the UL84 peptide, we reacted MAb84 with protein lysates prepared from cells transfected with various UL84 truncation mutants (Fig. 2B). Western blot analysis showed that MAb84 was able to interact with four of the six UL84 truncation mutants (Fig. 2B and C). MAb84 failed to recognize mutants 4 (aa 201 to 586) and 5 (aa 351 to 586), indicating that the antibody interacted with an epitope present within the protein between mutant 3 (aa 141 to 586) and mutant 4 (aa 201 to 586) (Fig. 2B and C). All deletion constructs were shown to produce protein, as judged by Western blot analysis with anti-FLAG antibody (data not shown). To further characterize the MAb84 epitope, we expressed 100-aa FLAG-tagged segments of the UL84 protein and incubated cell extracts with MAb84. MAb84 recognized a UL84 fragment encoding aa 101 to 200, indicating that the epitope for this antibody was within this fragment (Fig. 2D). All fragments were shown to express protein when reacted with anti-FLAG antibody (Fig. 2D, anti-FLAG panel). Based on the results from the deletion mutant and the interaction with the expressed small peptides, MAb84 interacts with a protein domain mapping between aa 141 and 200.

UL84 interacts with itself in vivo.

Because UL84 contains a leucine zipper motif and these structures are implicated in protein-protein interactions, we wanted to investigate the possibility that UL84 forms a stable interaction with itself. We cotransfected COS7 cells with plasmid constructs that expressed two differentially tagged (enhanced green fluorescent protein [EGFP] and FLAG) UL84 recombinant proteins. Cells lysates were prepared from cotransfected cells, and proteins were immunoprecipitated with either anti-FLAG or anti-EGFP antibodies. As controls, we cotransfected a plasmid encoding either EGFP (pEGFP-N1) or FLAG control protein (pFLAG-Control), along with pUL84-TAG(1-586) or pUL84-EGFP. Immunoprecipitated proteins were separated by using SDS-PAGE, followed by analysis by Western blots to detect each UL84 protein species. Cell lysates that were immunoprecipitated with the anti-FLAG antibody were reacted with anti-EGFP. Western blot analysis showed that a band was observed in samples reacted with anti-EGFP, indicating that the anti-FLAG antibody pulled down the UL84-EGFP fusion protein (Fig. 3A, lane 1). However, no band was detected when the EGFP control plasmid was used in the cotransfection (Fig. 3A, lane 2). Likewise, when samples were immunoprecipitated with anti-EGFP antibody and reacted with anti-FLAG, a band was detected only in the UL84 cotransfected samples and not when one of the transfected plasmids was a control FLAG expression construct (Fig. 3A, lane 3 and 4, respectively).

In order to determine which region of the UL84 protein is involved in oligomerization, we transfected Cos7 cells with a series of UL84 deletion constructs (deletions 1 to 4 shown in Fig. 2B). Deletion constructs were transfected into Cos7 cells, and expressed proteins were captured by using FLAG affinity beads. Cell lysates containing full-length UL84-HA protein were then added to the deletion mutants previously captured on the beads. Beads were washed, and proteins were eluted by using 3X FLAG peptide. Full-length UL84 was able to interact with deletion mutants having removed up to aa 141. Mutant 4, which initiates at aa 201, was unable to interact with full-length UL84 (Fig. 3B, panel anti-HA, lane 4). Therefore, the protein domain involved in oligomerization is located in a region between aa 141 and 201. This region is downstream of the putative leucine zipper motif (aa 114 to 135), indicating that the leucine zipper is not involved in a UL84-UL84 interaction.

The UL84 oligomerization domain (OD) lies between aa 151 and 200.

In order to further define the protein domain involved in oligomerization, we initially expressed three FLAG-tagged 100-aa polypeptides [p84FR(1-100), p84FR(101-200), and p84FR(201-300) such that the first 300 aa of the UL84 protein were represented (Fig. 4A) ]. Three transfections were performed; each delivered one of the three plasmids encoding a 100-aa UL84 peptide. Additional transfections delivered full-length UL84 (untagged, pUL84EXPRESS). FLAG-tagged protein was captured on FLAG affinity beads, followed by the addition of cell lysates containing full-length UL84 protein. After incubation, the beads were washed, and samples were eluted with 3X FLAG peptide. Western blots of transfected cell lysates show that full-length UL84 was efficiently expressed in all transfections (Fig. 4B, top panel, UL84 full-length). The middle panel demonstrates that each of the small peptides were efficiently expressed and immunoprecipitated by the FLAG affinity beads in the capture assay (Fig. 4B, middle panel lanes 1, 2, and 3). Only peptides produced from the transfection of p84FR(101-200) were able to bind full-length UL84 in the capture assay (Fig. 4B, bottom panel, lane 2). These data indicated that UL84 interacts with itself by using a protein domain that is located in a region of the protein between aa 101 and 200 (consistent with deletion mapping results), and this expressed polypeptide is sufficient to mediate this interaction.

To further define the OD of UL84, we performed the capture assay using a fragment that encoded aa 151 to 300, fragment 4 in Fig. 4A [construct p84OD(151-300)] plus pUL84-HA. Since it is now known that the OD lies within fragment 2 (aa 101 to 200), if the polypeptide produced from fragment 4 captured full-length UL84, then this would indicate that the OD is located between aa 151 and 200. Figure 4C is a Western blot that shows the result of the capture assay with cell lysates from transfection of pUL84-HA and p84OD(151-300) (Fig. 4A, fragment 4). The polypeptide produced from fragment 4 [p84OD(151-300)] interacted with full-length UL84-HA as shown by reacting coimmunoprecipitated proteins with an anti-HA antibody on a Western blot (Fig. 4C, top panel, lane 2). Polypeptide 4 was produced in cell lysates and was efficiently captured by the FLAG beads, as shown by Western blotting with FLAG antibody (Fig. 4C, bottom panel, lanes 1 and 2, respectively). These results localized the UL84 OD between aa 151 and 200.

To confirm that UL84 forms a stable self-interaction, we performed an in vitro cross-linking experiment with fragment number 4. In vitro-translated protein was treated with glutaraldehyde, and protein species were separated on an SDS-PAGE gel. Chemical cross-linking resulted in the shift from a monomer protein species (∼40 kDa) to a band of ∼80 kDa, suggesting that two subunits of the expressed peptide can exist as a dimer complex (Fig. 4D, lanes 1 and 2, respectively). A control sample with a plasmid expressing amino acid region from aa 387 to 580 of IE2, which contains the oligomerization domain for this protein (6), also was used in a cross-linking experiment. Treatment of this expressed polypeptide gave the expected conversion from monomer to dimer as previously described (Fig. 4D, compare lanes 3 and 4) (6). No other expressed UL84 peptides were shown by these methods to cross-link (data not shown). These results strongly suggest that UL84 full-length protein can exist as a homodimer.

The UL84 putative leucine zipper motif is essential for complementation of oriLyt-dependent DNA replication but does not contribute to oligomerization.

The leucine residues that comprise the putative leucine zipper domain for UL84 are located at aa 114, 121,128, and 135, respectively (Fig. 5A). To confirm that the leucine heptad repeats did not contribute to oligomerization in the context of the full-length UL84 protein, we introduced base pair changes, by site-directed mutagenesis, such that the leucine residues were changed to valine residues. Three constructs were generated, each of which contained mutations in two of the four leucine residues. These constructs were used to determine whether mutations of the leucine residues would affect the ability of UL84 to oligomerize. Coimmunoprecipitation assays were performed with UL84-HA-tagged wild-type protein cotransfected with each of the FLAG-tagged leucine zipper mutants. Figure 5B shows that full-length UL84-HA (Fig. 5B, top panel) was still able to bind to all UL84 leucine zipper mutants L2+L3, L1+L2, and L3+L4 (Fig. 5B, bottom panel). This result showed that the mutation of the putative leucine zipper motif in the context of the full-length UL84 protein did not affect oligomerization. This result also confirms that the leucine zipper is not involved in an UL84-UL84 interaction.

We next chose to test the ability of leucine zipper mutants to complement oriLyt-dependent DNA replication in order to determine whether the leucine zipper was involved in the replication function of UL84. The replication assay was performed with the UL84 leucine zipper mutants in place of wild-type UL84. The core replication proteins for HHV-8 under the control of the SV40 promoter (1), along with HCMV oriLyt and pON2206, were used in the cotransfection replication assay. The HHV-8 core replication proteins, along with IE2 and UL84, efficiently amplified cloned oriLyt in T-HFs (Fig. 5C, lane control). The leucine zipper mutants L2+L3 and L1+L2 were unable to complement DNA replication, whereas the transfection of pUL84 L3+L4 resulted in a weak replication signal (Fig. 5C). Therefore, although UL84 oligomerization was unaffected by the changes in leucine residues, a finding consistent with the mapping of the OD between aa 151 and 200, UL84 leucine zipper mutants were no longer able to complement origin-dependent DNA replication. This suggests that the leucine zipper structure is important for the replication function of UL84. However, the failure of UL84 leucine zipper mutants to complement DNA replication is not due to the interruption of a UL84-UL84 interaction.

UL84 leucine zipper mutants no longer repress IE2-mediated transactivation.

It was reported previously that IE2 can transactivate the UL112/113 promoter and that UL84 expression can abolish this effect in U373 cells (13, 28, 29). The ability of UL84 to dampen the transactivation effects of IE2 was correlated to the first 104 aa and the last 73 aa of the UL84 protein (13). However, since we have now implicated the leucine zipper of UL84 to contribute to DNA replication, we wanted to investigate whether the loss of a replication function for UL84 leucine zipper mutants involved the ability of UL84 to interact with IE2. Consequently, we evaluated the ability of UL84 leucine zipper mutants to downregulate IE2 transactivation of the UL112/113 promoter. U373 cells were cotransfected with a luciferase reporter construct containing the UL112/113 promoter, along with IE2 and either wild-type UL84 or the UL84 leucine zipper mutant expression constructs. UL84 leucine zipper mutants were markedly decreased in their ability to repress the transactivation effect of IE2 on the UL112/113 promoter, whereas wild-type UL84 was fully capable of almost completely abolishing IE2 activation of the promoter (Fig. 6). Interestingly, UL84 leucine zipper mutant L3 +L4, which weakly complemented oriLyt-dependent DNA replication, had an intermediate effect on IE2 transactivation (Fig. 6). These results strongly suggest that the leucine zipper of the UL84 protein is involved in regulating the activity of IE2.

The leucine zipper motif of UL84 is involved in IE2-UL84 heterodimerization.

Since it was demonstrated that UL84 leucine zipper mutants had a diminished effect on the repression of transactivation of the UL112/113 promoter by IE2, we next investigated whether an intact UL84 leucine zipper was necessary to interact with IE2. We cotransfected Cos7 cells with both an IE2 expression plasmid and the UL84 leucine zipper mutant L1+L2. Protein lysates were prepared and immunoprecipitated with anti-FLAG antibody. Immunoprecipitated protein was separated on an SDS-PAGE gel, and Western blotting was performed to determine whether the UL84 leucine zipper mutants retained their ability to interact with IE2. As demonstrated previously (13), control coimmunoprecipitations from cells cotransfected with wild-type IE2 and pUL84-TAG(1-586) with the FLAG antibody resulted in the detection of both proteins on Western blots reacted with antibodies specific for FLAG and IE2 (Fig. 7, lane 1). However, no IE2 protein was detected when lysates from cells cotransfected with plasmids expressing the UL84 leucine zipper mutant and wild-type IE2 were immunoprecipitated (Fig. 7, lane 2). Both the UL84 leucine zipper mutant and the IE2 protein were shown to be present at approximately the same levels, as determined by Western blot analysis of protein lysates (Fig. 7, lane 3). These data indicated that the leucine zipper motif of UL84 is implicated in the interaction with IE2. This observation, coupled with the data that leucine zipper mutants of UL84 fail to complement oriLyt-dependent DNA replication, suggests that an IE2-UL84 interaction is necessary for DNA replication.

Identification of IE2 protein domain involved in IE2-UL84 heterodimerization.

Previous data showing the interaction between IE2 and UL84 in infected cells, coupled with the known effects of UL84 on IE2 transactivation, led us to investigate which region of the IE2 protein was involved in UL84 heterodimerization. We generated IE2 expression constructs that encoded three regions of the IE2 protein such that the resultant polypeptides had an in-frame HA epitope (Fig. 8A). These three IE2-HA expression constructs were transfected into Cos7 cells, and cell lysates were prepared. The expression of each IE2-HA fragment was confirmed by Western blot analysis performed on the cell lysates (Fig. 8B, top panel). IE2-HA cell lysates were incubated with UL84-FLAG immobilized on FLAG affinity beads. Bound UL84-FLAG protein was eluted with 3X FLAG peptide, and samples were subjected to Western blotting to identify which IE2-HA fragment interacted with UL84. Figure 8B, middle panel, shows that in all cases UL84 was expressed and efficiently eluted from the FLAG affinity beads. Western blots with anti-HA revealed that the IE2 fragment encoding the first 215 aa of the protein, pIE2-HA(1-215), interacted with full-length UL84 (Fig. 8B, bottom panel, lane 1). As a control for our capture system, we also pulled down full-length IE2 by using FLAG affinity bead-immobilized UL84. As expected, we efficiently captured full-length IE2 protein with immobilized UL84 (Fig. 8C). These data indicated that the region of IE2 involved in UL84-IE2 heterodimerization was located in the first 215 aa of the protein.

Expression of UL84 OD or LZ domain has a transdominant-negative effect on origin-dependent DNA replication.

Three possible protein-protein interactions can occur with respect to UL84 and IE2: (i) the previously described IE2-IE2 interaction (6), (ii) a UL84-UL84 interaction, or (iii) a UL84-IE2 association. Despite these possible interactions, it is not clear which of these possible scenarios contributes to DNA replication. It was previously shown that overexpression of UL84 in U373 cells can inhibit HCMV growth (13). However, the exact nature of this repression was not apparent in that report. We chose to investigate whether the expression of the UL84 oligomerization domain could act as a transdominant inhibitor of origin-dependent DNA replication. This would indicate whether a UL84-UL84 interaction participates in DNA replication. In addition, since we now have shown that a protein domain within the first 215 aa of the IE2 protein interacts with UL84 and that an intact leucine zipper motif of UL84 is required for an interaction with IE2, we also expressed polypeptides containing these domains in the transient replication assay as potential transdominant inhibitors.

We cotransfected T-HFs with the HHV-8 core replication proteins along with oriLyt and pUL84-TAG(1-586) and the IE2 expression plasmid pON2206. We chose to use HHV-8 replication proteins so that we could concentrate only on the UL84-IE2 interaction. Each cotransfection also included one of the expression constructs encoding 100-aa segments of the UL84 ORF fused in frame with the SV40 NLS. Total cellular DNA was harvested and cleaved with EcoRI and DpnI. Replicated plasmid was detected by Southern blot analysis as previously described (24, 32). Addition of the expression vector p84FR(101-200) resulted in the absence of a replication signal (Fig. 9A), lane p84FR(101-200). Replication of oriLyt in the presence of p84FR(1-100), p84FR(201-300), p84FR(301-400), p84FR(401-500), or p84FR(501-586) had no effect on oriLyt-dependent DNA replication (Fig. 9A). This result suggested that a UL84-UL84 and/or a UL84-IE2 interaction was required for DNA synthesis since the inhibitory polypeptide contained both the UL84 OD and an intact leucine zipper motif. In order to determine whether one or both (UL84-UL84 or UL84-IE2) of the protein-protein interactions were required, we repeated the transient replication assay by using an expression construct encoding small peptides of the IE2 protein. No replication signal was detected when pIE2-HA(1-215) was added to the replication mixture (Fig. 9B, lane 2). Transfections that contained plasmids that expressed IE2 fragments spanning aa 216 to 386 [pIE2-HA(216-386)] or aa 387 to 580 [pIE2-HA(387-580)] had no effect on the amplification of oriLyt in the transient assay (Fig. 9B, lanes 3 and 4). This suggests that the oligomerization of IE2 was not a requirement for oriLyt DNA amplification, since the oligomerization domain of IE2 lies between aa 388 and 542 of the IE2 protein (6). This result also corroborates the data that UL84-IE2 interaction is essential for HCMV lytic replication, since the IE2 protein domain expressing the IE2-UL84 heterodimerization domain had a lethal transdominant-negative effect on origin-dependent DNA replication.

In order to decipher if both UL84-UL84 and IE2-UL84 interactions contributed to DNA replication, we again repeated the cotransfection replication assay by using constructs that expressed an intact oligomerization domain and a disrupted leucine zipper motif. A construct expressing the same 101 to 200 aa was expressed, except that this new plasmid (p84-2LZA) has the same amino acid changes to the leucine residues within the leucine zipper as used previously(L1+L2), which no longer interacted with IE2. Transfection of this construct resulted in a loss of replication signal (Fig. 9C, lane 2). We next choose to investigate whether the expression of the UL84 oligomerization domain alone, that is, in the absence of the leucine zipper motif, could have a lethal effect on oriLyt-dependent DNA replication. p84OD(151-300) was cotransfected, along with all of the required replication plasmids, including wild-type UL84, pON2206, and the HCMV cloned oriLyt and a replication assay was performed. The addition of p84OD(151-300) to the transfection mixture was also shown to have a lethal effect on oriLyt-dependent DNA replication (Fig. 9C, lane 3). These results gave further evidence that the expression of a peptide containing the UL84 OD or the leucine zipper motif in transfected cells interfered with essential UL84-UL84 and UL84-IE2 interactions and consequently inhibited lytic DNA replication.