SIAH-1 Interacts with the Kaposi's Sarcoma-Associated Herpesvirus-Encoded ORF45 Protein and Promotes Its Ubiquitylation and Proteasomal Degradation

Abstract

Kaposi's sarcoma-associated herpesvirus (KSHV), also referred to as human herpesvirus 8, is a potentially tumorigenic virus implicated in the etiology of Kaposi's sarcoma, primary effusion lymphoma, and some forms of multicentric Castleman's disease. The open reading frame 45 (ORF45) protein, encoded by the KSHV genome, is capable of inhibiting virus-dependent interferon induction and appears to be essential for both early and late stages of infection. In the present study, we show, both in yeast two-hybrid assays and in mammalian cells, that the ORF45 protein interacts with the cellular ubiquitin E3 ligase family designated seven in absentia homologue (SIAH). We provide evidence that SIAH-1 promotes the degradation of KSHV ORF45 through a RING domain-dependent mechanism and via the ubiquitin-proteasome system. Furthermore, our data indicate the involvement of SIAH-1 in the regulation of the expression of ORF45 in KSHV-infected cells. Since the availability of KSHV ORF45 is expected to influence the course of KSHV infection, our findings identify a novel biological role for SIAH proteins as modulators of virus infection.

Kaposi's sarcoma-associated herpesvirus (KSHV), also known as human herpesvirus 8, is a DNA tumor virus that is etiologically related to infrequent endothelial and lymphoid tumors, namely Kaposi's sarcoma, primary effusion lymphoma, and plasmablastic multicentric Castleman's disease. KSHV is classified as a gamma-2-herpesvirus and shares significant homology with several primate and nonprimate mammalian viruses. Among human herpesviruses, KSHV is related most closely to the Epstein-Barr virus (EBV), which is known as a ubiquitous lymphotrophic virus that is associated causally with several human cancers, including certain types of lymphomas (4, 5, 7, 8, 38).

Open reading frame 45 (ORF45) is a conserved gammaherpesvirus gene (45), yet its critical role in virus infection has been recognized only recently (21, 22, 47). The disruption of KSHV ORF45 expression has no effect on viral lytic DNA replication or on late gene expression during virus reactivation; still, it causes a drastic decrease in the yield of progeny viruses, suggesting a function of ORF45 in viral maturation or egress. Furthermore, ORF45 deficiency results in a considerably reduced virus infectivity, indicating its requirement during early stages of infection (47). Similarly, an ORF45 null mutant of murine gammaherpesvirus 68 is incapable of virion production (21). The KSHV-encoded ORF45 is expressed as an immediate-early lytic gene (45). It is a phosphorylated protein that is localized predominantly to the cytoplasm of infected cells and is tightly associated with purified KSHV virions, probably through the inner layer of the tegument (48). The functional characterization of KSHV ORF45 established its role as an inhibitor of the induction of type I interferon (IFN) genes upon infection through an interaction with cellular IFN regulatory factor 7 (IRF-7). Since the IFN-induced cellular response is the primary defense mechanism against viral infection, KSHV ORF45 has been classified as a viral immune evasion protein (46). Still, given that ORF45-deficient viruses are impaired in the transport of capsid and ingress and in virion assembly and egress, it is likely that ORF45 performs additional functions besides blocking IRF-7 activation (47).

Members of the seven in absentia/seven in absentia homologue (SINA/SIAH) family of proteins are evolutionarily conserved from plants to mammals and function primarily as ubiquitin E3 ligases (30, 36). By means of direct and specific interactions with substrates, the ubiquitin E3 ligases provide the specificity of the ubiquitin conjugation system and are responsible for forming polyubiquitin chains on substrate proteins. In general, these chains function as a tag for proteasomal degradation. The human SIAH proteins, SIAH-1 and SIAH-2, are highly homologous and differ mostly in their N terminus, which encodes a RING domain that confers their E3 ubiquitin ligase activity (16, 17). The SIAH C terminus encodes a domain implicated in mediating binding to various substrate proteins, some of which are degraded. Mammalian substrates targeted for degradation by SIAH are quite diverse; examples include the netrin-1 receptor/deleted in colorectal cancer (18); the nuclear receptor corepressor (43); the motor protein Kid (10); the transcriptional activator OBF-1 (3, 39); the neural transmitter protein synaptophysin (41); the presynaptic proteins synphilin-1 and α-synuclein (27); the transcriptional repressor TIEG-1 (23); and the HIF-1α degradation regulators propyl-hydroxylating domain containing 1 and 3 (PHD1 and PHD3) (33, 34). SIAH proteins also limit their own availability through efficient self ubiquitylation and degradation (17). In the examples listed above, the SIAH protein functions as a single-target subunit E3 ligase. However, SIAH has been shown to facilitate the degradation of β-catenin as part of an SCF-type complex, which includes Skp1, Ebi, and SIAH interacting protein (SIP) (29, 32). In this complex, Skp-1 and SIP act as a molecular bridge linking SIAH to Ebi, which is the subunit that directly binds β-catenin.

To better understand the function and regulation of KSHV ORF45, we undertook a systematic search for novel cellular interacting partners. In this report, we describe the interaction and functional relationship between KSHV ORF45 and the SIAH E3 ubiquitin ligases. Our results provide explicit evidence that SIAH-1 promotes the degradation of KSHV ORF45 through a RING domain-dependent mechanism and via the ubiquitin-proteasome system. Since the availability of KSHV ORF45 is expected to influence the course of KSHV infection, our findings suggest a novel biological role for SIAH proteins as modulators of virus infection.

MATERIALS AND METHODS

Chemical reagents and antibodies.

The proteasome inhibitor MG132 was purchased from Calbiochem. Cycloheximide, n-butyrate, and 12-O-tetradecanoylphorbol-13-acetate (TPA) were obtained from Sigma. Mouse anti-Flag, anti-hemagglutinin (HA.11) (Covance Research Products), anti-green fluorescent protein (anti-GFP) (Covance Research Products), anti-myc (myc-Tag 9B11) (Cell Signaling Technology), anti-SIAH-1 (FL-282) (Santa Cruz biotechnology), and anti-ORF45 (2D4A5; Abcam) antibodies were used with either peroxidase-conjugated or rhodamine-conjugated goat anti-mouse immunoglobulin G (IgG) secondary antibodies (Jackson ImmunoResearch Laboratories, Inc.).

Plasmids.

To create a yeast expression plasmid encoding the full-length KSHV ORF45, PCR amplification was performed using the primers 5′CCCGAATTCATGGCGATGTTTGTGAGGACC3′ and 5′AAAGGATCCTCAGTCCAGCCACGGCCAGTT3′ (restriction enzyme recognition sites are underlined). The PCR product was subcloned into pAS2.1 (Clontech) following digestion of the plasmid and PCR product with EcoRI and BamHI. A mammalian expression plasmid encoding HA-ORF45 was obtained by PCR amplification using the outer sense primer 5′TACGGATCCATGGCGATGTTTGTG3′ and outer antisense primer 5′AAAACTCGAGTCAGTCCAGCCACGGCCA3′, followed by cloning the BamHI- and XhoI-digested PCR product into similarly digested pcDNA-HA. Deletion variants of KSHV ORF45 were expressed from pcDNA-HA and were created by PCR using the following internal primers: ORF45(1-141) internal antisense, 5′-AAACTCGAGTCATTTGGGCGTATGGGCCCG-3′; ORF45(1-166) internal antisense, 5′-AAACTCGAGTCAAACCCATCCCATGGACGC-3′; ORF45(250-407) internal sense, 5′-AAAGGATCCTCCAACTCCCGGACGTG-3′; and ORF45 (167-407) internal sense, 5′-AAAGGATCCAGTCAGGATGACGGATTTTCC −3′, and the flanking outer sense and antisense primers described above that contained BamHI and XhoI sites. The mammalian GFP-ORF45 expression plasmid was obtained by PCR amplification of ORF45 with 5′CCCGAATTCATGGCGATGTTTGTGAGGACC3′ and 5′ AAAGGATCCTCCAGCCACGGCCAGTTATAT 3′ and cloning the PCR product into pEGFP-N1 (Clontech) subsequent to its digestion with EcoRI and BamHI. All constructs were verified by sequencing. Yeast vectors expressing subregions of SIAH-1 were kindly provided by David J. Elliott (University of Newcastle, Newcastle, United Kingdom) (40). Mammalian expression vectors containing myc-tagged wild-type SIAH-1 and SIAH-1 mutated at the RING finger (C55A/C59H/C72S) were obtained from Simone Engelender (Technion, Israel) (27). The Flag-tagged mammalian expression vector containing the RING mutant (H99A/C102A) form of SIAH-2 was kindly provided by Ze'ev Ronai (The Burnham Institute, La Jolla, CA) (12).

Site-directed mutagenesis.

Point mutants in the pcDNA-HA-ORF45 plasmid were created by two rounds of PCR. During the first round of PCR, we used the flanking outer primers described above that contained BamHI and XhoI sites, together with overlapping inner sense and antisense primer pairs that contained the mutations. This round produced amplification products that overlapped at their 3′ and 5′ ends. A second-round amplification used these products as amplification templates with the outer primers described above. The mutated full-length PCR fragments of ORF45, generated by the second round of PCR, were inserted into pcDNA-HA. The oligonucleotide primers used were the following: HA-ORF45-(A144G/V146G) inner sense, 5′CAAACCGGTAGGAGTGGGAGCGGGCCGCG3′; HA-ORF45-(A144G/V146G) inner antisense, 5′ACGCGGCCCGCTCCCACTCCTACCGGTTTG3′; HA-ORF45-(A236G/V238G) inner sense, 5′CGCACCCACCGGGATCGTGGACCTGACATC3′; and HA-ORF45-(A236G/V238G) inner antisense, 5′GATGTCAGGTCCACGATCCCGGTGGGTGCG3′ (mutated residues are underlined).

Yeast two-hybrid screen.

Full-length KSHV ORF45 in pAS2.1 was used to screen a Matchmaker pretransformed human bone marrow cDNA library (Clontech) according to the manufacturer's protocol. Large colonies that grew on −Trp/−Leu/−His plates were patched on full selection medium (−Trp/−Leu/−His/−Ade), and a standard β-galactosidase filter lift assay was performed. Blue colonies were recultured on +Trp medium to select for the library plasmids. Crude yeast plasmid preparations were then transformed into Escherichia coli DH5α. Interacting clones were identified by sequencing pGADT7 (prey) library inserts. The interaction was confirmed following cotransformation into Y190 yeast cells of SIAH-1 fragments cloned in pGBKT7 with the KSHV ORF45 bait plasmid.

Cell culture and transfection.

Human kidney 293T cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS) (Biological Industries, Kibbutz Beit Haemek, Israel) and antibiotics. Bacterial artificial chromosome 36 (BAC36) was kindly provided by Shou-Jiang Gao (The University of Texas Health Science Center at San Antonio, San Antonio Cancer Institute, TX) (44), and human kidney 293T cells that carry infection were maintained under selection with 75 μg/ml of hygromycin. For transient transfections, the calcium phosphate precipitation method was employed using 4 μg of each plasmid DNA per 10-cm plate or a total of 3 μg plasmid DNA per well of a 6-well dish. Transient transfection into BAC36-infected 293T cells was carried out with Lipofectamine (Invitrogen). In all cases, the total amount of DNA was normalized by the addition of control plasmids.

Western blot analysis.

Cells were washed twice in cold phosphate-buffered saline (PBS), suspended in lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.5% sodium deoxycholate, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate [SDS], 1 mM phenylmethylsulfonyl fluoride, 50 μg/ml aprotinin, 50 μM leupeptin, 0.2 mM Na₃VO₄, 50 mM NaF) and incubated on ice for 30 min. Cell debris then were removed by centrifugation at 12,000 × g for 15 min at 4°C. Loading buffer (2×; 2% SDS, 20% glycerol, 125 mM Tris [pH 6.8], 0.02% bromophenol blue, and 10% β-mercaptoethanol) was added, and the samples were boiled for 5 min. Protein lysates were resolved by SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to nitrocellulose membranes (Schliecher & Schuell). The protein contents of different samples were verified to be similar by Ponceau staining. The nitrocellulose membranes were blocked with 5% dry milk in Tris-buffered saline (TBS) or 1% dry milk and 1% bovine serum albumin (BSA) in TBS and subsequently incubated with primary antibody. Specific reactive bands were detected using goat anti-rabbit IgG or goat anti-mouse antibody conjugated to horseradish peroxidase. Immunoreactive bands were visualized using the enhanced chemiluminescence Western blotting detection kit (Amersham, Arlington Heights, IL). The quantification of the protein bands was performed using the Java-based public domain image-processing and analysis program ImageJ (http://rsb.info.nih.gov/ij/).

Immunoprecipitation assays.

Cell extracts were clarified by centrifugation at 14,000 × g for 5 min, and the resulting supernatant was incubated overnight at 4° C with anti-HA, anti-Flag, or anti-myc and then with protein A/G plus agarose beads (Santa Cruz). After intensive washing with buffer containing 50 mM Tris (pH 7.4), 140 mM NaCl, 0.1% Triton X-100, and protease inhibitor cocktail (Complete; Boehringer) and centrifugation, immune complexes were separated by SDS-PAGE and probed by Western blotting.

The detection of ORF45 ubiquitin conjugates in cells was carried out as previously described (1) under strong denaturing conditions that inhibit deubiquitylating enzymes.

Immunofluorescence staining.

293T cells were fixed in PBS with 4% paraformaldehyde and then permeabilized and blocked for 20 min in PBS with 0.2% Triton X-100 and 1% BSA. Slides were incubated with primary antibody overnight at 4°C, followed by incubation with anti-mouse antibody coupled to rhodamine for 30 min at room temperature. The slides were mounted in antifading medium (1% n-propyl gallate, 90% glycerol in PBS) and visualized by indirect immunofluorescence.

Protein turnover analysis.

Cycloheximide and MG132 experiments were performed on cells transfected in duplicate as described above. Cells were cultured for 6 h after transfection, pooled together, and split into 12-well dishes. Cycloheximide (150 μg/ml) or MG132 (10 μM) was added to the cultures 24 h after transfection (time zero). Proteins were prepared at the indicated times, and equal amounts were subjected to immunoblot analysis. The expression of cotransfected GFP was used to monitor equal loading.

In vitro ubiquitylation assay.

Substrates (wild-type ORF45, mutant ORF45, and HIS-Ring1B) were translated using a TNT quick coupled transcription-translation kit from Promega with [³⁵S]methionine (Amersham Pharmacia). In vitro-translated proteins were incubated in ubiquitin reaction medium containing 40 mM Tris (pH 7.6), 5 mM MgCl₂, 1 mM dithiothreitol, 2 mM ATP, 5 μg of ubiquitin, 100 ng UbA1, 250 ng of E1, and 200 ng of His-UbcH5b, in the absence or presence of bacterially purified recombinant glutathione S-transferase-SIAH-1 (GST-SIAH-1) (kindly provided by Simone Engelender [Technion, Israel]). Reactions were incubated at 37°C for 1 h and resolved on SDS-PAGE. Ring1B E3, which mediates its own polyubiquitylation (2), was used to control for the specificity of SIAH-1. GST was used as a negative control for ubiquitylation. ³⁵S-labeled proteins were detected by PhosphorImager analysis (ubiquitin aldehyde, purified E1, UbcH5b, and Ring1B plasmid were kindly provided by Aaron Ciechanover [Technion, Israel]).

RESULTS

ORF45 protein binds SIAH in yeast.

In order to identify proteins that interact with KSHV ORF45, we performed a yeast two-hybrid screen using full-length KSHV ORF45 as the bait and a human bone marrow cDNA library as the prey. Of the interacting clones, two contained the siah-1 gene and three contained the siah-2 gene. Each of these positive clones varied in length, suggesting independent interaction events. Of note, sequence analysis indicated that none of the positive clones could produce the SIAH protein fused to the GAL4 activation domain unless translational frameshift events allowed the production of chimeric SIAH proteins. Since five independent clones of SIAH interacted with ORF45, we hypothesized that it is likely that an authentic protein-protein interaction indeed takes place between these proteins.

To confirm the interaction and to delineate the domains of SIAH mediating its interaction with ORF45, three SIAH-1 fragments, including the N-terminal RING finger domain (amino acids [aa] 1 to 99), the central zinc finger region (aa 99 to 153), and the C-terminal 130-aa (aa 152 to 282) substrate binding domain were tested in a directed yeast two-hybrid assay. The specificity of the interaction was verified by the coexpression of KSHV ORF45 with unrelated plasmids (the pGAD4 activation domain alone or fused to p53). Only the C-terminal substrate binding domain and none of the other subregions of SIAH-1 interacted specifically with KSHV ORF45, as indicated by growth on selective medium (−Trp/−Leu/−His) and by a β-galactosidase assay (Table 1). This result corroborates the interaction of KSHV ORF45 and SIAH in yeast and maps the interaction site to the C-terminal substrate binding domain of SIAH-1.

TABLE 1.

Interaction between ORF45 and SIAH-1 in yeast

GAL4-binding plasmid	β-Galactosidase and HIS reactions for GAL4 activation domain plasmid alone or with:
	Alone	p53	SIAH-1 (aa 1-99)	SIAH-1 (aa 99-153)	SIAH-1 (aa 152-282)
ORF45	−	−	−	−	+
Simian virus 40 large T antigen	−	+	−	−	−
None	−	−	−	−	−

ORF45 protein physically associates with SIAH in mammalian cells.

To verify that KSHV ORF45 interacts with SIAH proteins in human cells in vivo, full-size ORF45 and myc-tagged RING domain-mutated SIAH-1 or Flag-tagged RING domain-mutated SIAH-2 expression vectors [pcDNA-HA-ORF45, myc-SIAH-1-(C55A/C59H/C72S), and Flag-SIAH-2-(H99A/C102A), respectively] were transfected into 293T cells. Since SIAH proteins mediate their own ubiquitylation and degradation, we used the dominant-negative (DN) RING mutants that retain the ability to bind most substrates, lack E3 activity, and tend to accumulate in the cells (17). As shown in Fig. 1, SIAH-1 and SIAH-2 coimmunoprecipitated with HA-ORF45 but not with the unrelated protein HA-vBcl-2. Similar results were obtained when a reciprocal immunoprecipitation was performed (data not shown).

FIG. 1.

KSHV ORF45 interacts with SIAH in 293T cells. To detect the association between SIAH and ORF45, expression vectors containing the myc-tagged RING mutant form of SIAH-1 [myc-SIAH-1-(C55A/C59H/C72S)] or the Flag-tagged RING mutant form of SIAH-2 [Flag-SIAH-2-(H99A/C102A)] and HA-tagged ORF45 (pcDNA-HA-ORF45) were transfected into 293T cells (lanes 4 and 5). Untransfected cells (lane 1) and cells transfected with the HA-vBcl-2 expression vector were used as negative controls (lanes 2 and 3). After 24 h, whole-cell extracts (WCE) were prepared, and 30-μg aliquots were analyzed for protein expression. The overexpression of HA- and Flag-tagged proteins was sequentially assayed on the same blot, which was first reacted with anti-HA and then with anti-Flag antibody (upper panel). The expression of myc-tagged proteins was assayed on a separate blot (middle panel). To this end, 400 μg of the lysates was subjected to immunoprecipitation (IP) with anti-HA, followed by sequential Western blotting (WB) with anti-myc and anti-Flag antibodies (lower panel). myc-SIAH-1-(C55A/C59H/C72S) and Flag-SIAH-2-(H99A/C102A) coprecipitated with HA-ORF45 but not with the unrelated protein HA-vBcl-2. NS, nonspecific reacting bands.

Because SIAH-1 and SIAH-2 are highly similar to one another in both structure and function, and since the regulation and functional characteristics of SIAH-1 have been studied in more detail, all of our further studies concentrated on the interaction with SIAH-1. As already described (17), due to the rapid degradation of SIAH-1 it was not possible to detect its endogenous protein expression; we therefore employed the ectopic expression of tagged and mutated SIAH-1 in most of the experiments described in the present study.

ORF45 colocalizes with SIAH-1 in mammalian cells.

To establish the interaction between ORF45 and SIAH-1 in human cells, we first determined whether ORF45 and SIAH-1 colocalize. Expression vectors for enhanced GFP (EGFP)-ORF45 (pEGFPN1-ORF45) and myc-tagged SIAH-1 or RING domain-mutated SIAH-1 [myc-SIAH-1-(C55A/C59H/C72S)] were cotransfected into 293T cells. Immunofluorescence analysis of these cells showed that EGFP-ORF45 was localized predominantly to the cytoplasm (Fig. 2). Wild-type SIAH-1 also was present in the cytoplasm, mainly in large aggregates of unknown identity. These results are consistent with previous reports on the cellular localization of ORF45 and SIAH-1 (16, 17, 48). The subcellular expression patterns of KSHV ORF45 and SIAH-1 almost completely overlapped in the majority of cotransfected cells. As previously reported (17, 18), DN RING-mutated SIAH-1 [myc-SIAH-1-(C55A/C59H/C72S)] appeared in a much more uniform distribution throughout the cytoplasm. KSHV ORF45 partially overlapped with the mutated SIAH-1 in the cytoplasm and also exhibited some nuclear localization. These findings provide further evidence in support of a physical interaction between ORF45 and SIAH-1.

FIG. 2.

KSHV ORF45 and SIAH-1 colocalize in 293T cells. 293T cells were transfected with plasmids encoding EGFP-ORF45 (pEGFPN1-ORF45) and myc-tagged SIAH-1 or RING domain-mutated SIAH-1 [myc-SIAH-1-(C55A/C59H/C72S)]. Cells were stained with mouse monoclonal anti-myc antibody to detect SIAH-1. Rhodamine-conjugated goat anti-mouse IgG (Texas Red) was used as a secondary antibody. All cell nuclei in the fields are shown by Hoechst staining (blue), and overlays of the fluorescence micrographs are shown on the right.

SIAH-1 promotes the degradation of ORF45 protein via the ubiquitin proteasome pathway.

SIAH proteins function primarily as E3 ubiquitin ligases that target substrates for degradation; however, they also bind proteins without inducing their degradation (31, 42, 13, 33, 35). To determine if SIAH-1 regulates ORF45 expression, we examined the ability of transfected SIAH-1 to influence the steady-state expression of cotransfected ORF45. As shown in Fig. 3A, the coexpression of SIAH-1 led to a significant reduction in the amount of recovered ORF45 protein compared to that of control cells cotransfected with control vector. This reduction was specific for ORF45, as the expression of cotransfected GFP was constant and, therefore, served as an internal control for similar transfection efficiencies. In contrast, the expression of myc-SIAH-1-(C55A/C59H/C72S), which has DN activity, resulted in elevated steady-state levels of ORF45. This probably is due to the ability of the SIAH-1 mutant to inhibit the activity of the limiting amounts of wild-type endogenous SIAH-1 through its accumulation and binding to target substrates (17). These data indicate that SIAH-1 influences ORF45 protein levels and suggests that an intact RING domain of SIAH-1 is required to reduce ORF45 stability.

FIG. 3.

Cotransfected SIAH-1 protein reduces the levels of ORF45 protein in a proteasome-dependent manner. (A) 293T cells were transfected with the HA-ORF45 expression plasmid together with empty myc vector, myc-SIAH-1, or myc-SIAH-1-(C55A/C59H/C72S). The steady-state levels of ORF45 were assayed by Western blotting using anti-HA antibody 36 h after transfection. The expression of GFP was used to control for similar transfection efficiencies and loading. (B) 293T cells were transfected as described above. To avoid variation in protein levels as a result of different transfection efficiencies, the cells were incubated for 6 h after transfection and then pooled and split into 12-well dishes. MG132 (10 μM) or dimethylsulfoxide (solvent control) were added to the cultures 24 h after transfection (time zero) and 2 h before harvesting. Equal amounts of protein extracts (30 μg) were analyzed by using anti-HA antibodies. GFP was used to control for transfection efficiencies and loading. The results shown are representative of those from two similar experiments.

We next used an inhibitor of proteasome function to examine whether the proteasomal degradation pathway mediates the effects of SIAH-1 on ORF45 protein levels. As shown in Fig. 3B, treatment with the proteasome inhibitor MG132 induced the accumulation of HA-ORF45 in 293T cells that were cotransfected with SIAH-1, while only a minor accumulation was evident when the HA-ORF45 expression vector was transfected alone. The coexpression of RING-mutated SIAH-1-(C55A/C59H/C72S) with HA-ORF45 resulted in the steady expression of HA-ORF45 upon proteasomal inhibition. These results indicate that the proteasomal degradation mediates the effects of SIAH-1 on KSHV ORF45 protein levels.

To confirm that the reduction of ORF45 protein in SIAH-1-cotransfected cells is due to protein degradation, we examined the relative stability of the ORF45 protein 4 and 8 h after the addition of cycloheximide to block all new protein synthesis. As shown in Fig. 4, the degradation of HA-ORF45 was accelerated in cells overexpressing SIAH-1. In contrast, the degradation of HA-ORF45 was reduced in cells expressing RING-mutated SIAH-1-(C55A/C59H/C72S). Thus, it appears that SIAH-1 targets ORF45 for proteasomal degradation.

FIG. 4.

SIAH-1 reduces the stability of KSHV ORF45 protein. (A) 293T cells were transfected as described in the legend to Fig. 3B, and cycloheximide was added at 24 h. Standard whole-cell extracts were prepared at 4 and 8 h after treatment and immunoblotted with anti-HA antibodies. GFP was used to control for transfection efficiency and loading. (B) The plot shows the quantification of the remaining HA-ORF45 from three independent experiments.

To confirm that SIAH-1 promotes the degradation of ORF45 through the ubiquitin-mediated pathway, exogenous Flag-ORF45 and HA-ubiquitin were cotransfected in the presence or absence of SIAH-1 into 293T cells for intracellular ubiquitylation assays. A reduced level of ORF45 protein was detected in cells overexpressing SIAH-1 compared to that of cells transfected with ORF45 and HA-ubiquitin alone (Fig. 5). However, the level of the slow-migrating high-molecular-weight smear of polyubiquitylated ORF45 was much higher in cells cotransfected with SIAH-1 than in cells transfected with Flag-ORF45 and HA-ubiquitin alone. This suggests that the ectopic expression of SIAH-1 induced a greatly enhanced proteasome-mediated degradation of polyubiquitylated ORF45. Taken together, these findings suggest that SIAH-1 mediates the ubiquitylation and consequent proteasomal degradation of ORF45, as reported for other substrates of SIAH.

FIG. 5.

SIAH-1 promotes the ubiquitylation of ORF45 in vivo. 293T cells were transiently cotransfected with plasmids encoding Flag-ORF45 and HA-ubiquitin with or without myc-SIAH-1. After 24 h, protein extracts were prepared under strong denaturing conditions, and the expression of Flag-ORF45, myc-SIAH-1, and HA-ubiquitin was assayed by Western blotting (WB) using 10% of the total extract (Input). Flag-ORF45 was immunoprecipitated (IP) from the remaining cell extracts with anti-Flag, separated by SDS-PAGE, and transferred to Western blots. The immunoprecipitates were probed with anti-HA antibody to identify Flag-ORF45, to which the HA-ubiquitin polypeptide had been attached.

Identification of ORF45 sequences required for SIAH-1 interaction.

A short consensus peptide motif comprising PxAxVxP with the less conserved flanking residues RPVAxVxPxxR has been reported recently to mediate the interaction between SIAH-1 and a range of protein partners (15). The inspection of the ORF45 amino acid sequence revealed two semiconserved potential SIAH-1 binding motifs. The first includes KPVAVVAGRVR (aa 140 to 151), and the second, which is less conserved, includes APTAIVDLTSD (aa 232 to 243) (boldface indicates a consensus SIAH-1 binding site) (Fig. 6A). A semiconserved potential SIAH-1 binding site also was found in the ORF45 protein sequence encoded by Rhesus monkey rhadinovirus (RPVAVVTGQHR; aa 146 to 157). This site is located at a position corresponding to the first site found on residues 140 to 151 of KSHV ORF45.

FIG. 6.

Mapping the SIAH-1 interaction sites. (A) Alignment of the consensus SIAH-1 binding motif with potential SIAH-1 binding sites in KSHV ORF45. The positions of the point mutations A144G/V146G and A236G/V238G are shown by arrows. Also shown is a scheme of ORF45 deletion fragments, with the potential SIAH-1 binding motifs marked by gray boxes. (B) Western blots showing the coprecipitation of HA-ORF45 mutants and SIAH-1. Extracts of 293T cells cotransfected with plasmids encoding RING-mutated myc-SIAH-1-(C55A/C59H/C72S) and HA-ORF45 deletion mutants [HA-ORF45(1-141), HA-ORF45(1-166), and HA-ORF45(250-407) HA-ORF45(167-407)] or (C) or point mutants [HA-ORF45-(A144G/V146G) and HA-ORF45-(A236G/V238G)] were immunoprecipitated with anti-HA antibody, and the presence of RING-mutated myc-SIAH-1-(C55A/C59H/C72S) in the precipitates was examined by probing with anti-myc monoclonal antibody as described for Fig. 1. WCE, whole-cell extracts; IP, immunoprecipitation; WB, Western blotting.

To delineate the region(s) responsible for the interaction of ORF45 with SIAH-1, we first constructed four deletion variants of pcDNA-HA-ORF45 (Fig. 6A). The different HA-tagged ORF45 constructs were cotransfected into 293T cells with myc-tagged RING-mutated SIAH-1 expression vector, and cell extracts were immunoprecipitated with anti-HA antibody followed by Western blotting with anti-myc antibody. The results revealed that HA-ORF45(1-141), HA-ORF45(250-407), and HA-ORF45(167-407) fragments lost the ability to coprecipitate with SIAH-1, and the only fragment that retained the ability to coprecipitate with SIAH-1 was HA-ORF45(1-166). These results suggest that the critical residues in KSHV ORF45 for SIAH-1 binding map between residues 142 and 166 (Fig. 6B).

Furthermore, to ensure that the lack of interaction between ORF45 mutants and SIAH-1 was not an artifact of the truncations, we mutated the conserved potential SIAH-1 binding sites in ORF45 and investigated the ability of the ORF45 mutants [HA-ORF45-(A144G/V146G) and HA-ORF45-(A236G/V238G)] to interact with SIAH-1 by coimmunoprecipitation. As shown in Fig. 6C, point mutations of the site located at aa 232 to 243 did not affect the interaction of ORF45 with SIAH-1, whereas point mutations of the more conserved site located at aa 140 to 151 almost completely abolished the interaction of ORF45 with SIAH-1. These experiments confirm that SIAH-1 interacts with a conserved SIAH-1 binding motif located at aa 140 to 151.

We next assessed the ability of transfected SIAH-1 to influence the steady-state expression of mutated ORF45, similarly to the experiment performed previously with wild-type ORF45. Unlike wild-type ORF45, the coexpression of the ORF45 mutant HA-ORF45-(A144G/V146G), which lacks the SIAH-1 binding motif, with SIAH-1 or with the DN RING mutant of SIAH-1-(C55A/C59H/C72S) did not affect the steady-state level of HA-ORF45 (Fig. 7).

FIG. 7.

Cotransfected SIAH-1 proteins fail to alter the levels of mutated ORF45 [HA-ORF45-(A144G/V146G)] protein. 293T cells were transfected and assayed as described in the legend to Fig. 3A.

SIAH-1 functions as a ubiquitin E3 ligase for ORF45 in vitro.

To confirm that SIAH-1 itself functions directly to add polyubiquitin to ORF45 through a conserved SIAH-1 binding motif, the E3 ubiquitin ligase activity of SIAH-1 was assessed by performing in vitro cell-free ubiquitylation assays. The coincubation of a bacterially expressed purified form of GST-SIAH-1 and in vitro-translated ³⁵S-labeled ORF45 with ubiquitin system components, including the Ubc5b E2 enzyme, produced a slow-migrating smear representing ubiquitin-conjugated ORF45 (Fig. 8A, lane 3). The omission of GST-SIAH-1 (lane 2) or Ubc5b (lane 1) failed to yield any polyubiquitin-conjugated bands. No similar high-molecular-weight smear was produced when mutated ORF45 [HA-ORF45-(A144G/V146G)] that fails to interact with SIAH-1 was used as a substrate and incubated with all necessary ubiquitin components (lanes 4 to 6). The controls, including nonrelevant HIS-Ring1B E3 ligase (lanes 7 and 8) or GST alone (lanes 9 and 10), failed to ubiquitylate either wild-type or mutant ORF45. Western blotting with anti-ubiquitin mouse monoclonal antibodies was used to confirm the self ubiquitylation activity of HIS-Ring1B (Fig. 8B). These results provide unequivocal evidence that ORF45 is directly ubiquitylated by SIAH-1 through a conserved SIAH-1 interaction site.

FIG. 8.

SIAH-1 targets the ubiquitylation of ORF45 in vitro. (A) In vitro-translated ³⁵S-labeled wild-type (WT) and mutant ORF45 [HA-ORF45 and HA-ORF45-(A144G/V146G)] were incubated at 37°C in ubiquitin reaction medium in the absence or presence of bacterially expressed purified GST-SIAH-1. HIS-Ring1B was used to control for the specificity of SIAH-1, and GST was used as a negative control. Reactions were incubated for 1 h and resolved by SDS-PAGE. ³⁵S-labeled proteins were detected by PhosphorImager analysis. (B) The E3 ligase autoubiquitylation activity of HIS-Ring1B was confirmed by Western blotting (WB) with anti-ubiquitin.

SIAH-1 coimmunoprecipitates with ORF45 and regulates the endogenous expression of ORF45 in KSHV-infected cells.

Finally, we asked whether there are any functional consequences of SIAH-1 and ORF45 interactions on virus infection. To this end, we used 293T cells that carry KSHV BAC36 infection (44). These cells are predominantly latently infected, with a small fraction of cells undergoing sporadic virus reactivation, which can be further augmented by TPA and butyrate treatment (26, 44). As shown in Fig. 9A, the increased expression of ORF45 was evident in protein extracts from cells undergoing lytic induction by combined treatment with TPA and butyrate for 24 h. The ectopic expression of SIAH-1 resulted in a small reduction in the level of ORF45 compared to that of cells transfected with a control vector, whereas significant increases in levels of ORF45 were seen in cells expressing DN RING mutant SIAH-1-(C55A/C59H/C72S) with or without lytic induction. Therefore, these data suggest that the endogenous SIAH-1 protein is active in 293T cells that are KSHV infected and has the potential to regulate the expression of ORF45. The endogenous activity of SIAH-1, also suggested by our findings with uninfected 293T cells (Fig. 3A), may explain the minor effect of ectopic SIAH-1 expression on the endogenous ORF45, which is expressed in relatively low levels compared to that of overexpressed ORF45. Nonetheless, similarly to previous reports (33), we were unable to detect endogenous SIAH-1 protein expression in uninfected and KSHV-infected 293T cells even after treatment with a proteasome inhibitor.

FIG. 9.

SIAH-1 interacts with ORF45 and alters its expression levels in KSHV-infected cells. (A) Latently infected 293T cells were transfected with plasmids expressing myc-tagged control vector, myc-tagged SIAH-1, or RING domain-mutated SIAH-1 [myc-SIAH-1-(C55A/C59H/C72S)]; left untreated; or treated with TPA (20 ng/ml) and n-butyrate (0.3 mM). After 24 h, whole-cell extracts were prepared and 30-μg aliquots were analyzed for the expression of ORF45 and SIAH-1. (B) Anti-tubulin antibody was used as a loading control. Lysates (400 μg) from control vector and RING domain-mutated SIAH-1-transfected cells were subjected to immunoprecipitation (IP) with anti-myc antibodies followed by Western blotting (WB) with anti-ORF45 antibodies.

To confirm that the interaction between ORF45 and SIAH-1 also occurs in KSHV-infected cells, a coimmunoprecipitation assay was performed with extracts of 293T cells infected with KSHV, induced with n-butyrate and TPA, and transfected with the myc-SIAH-1-(C55A/C59H/C72S) expression plasmid as described above. Western blotting of the immunoprecipitated proteins with anti-ORF45 antibody revealed the presence of ORF45 in the precipitates generated with anti-myc antibody (Fig. 9B). Overall, these results suggest that ORF45 interacts with SIAH-1 during virus reactivation and that SIAH-1 operates during this phase.

DISCUSSION

ORF45 is an essential regulator of the gammaherpesvirus infections; hence, its availability is likely to influence the fate of virus infection (21, 22, 45-48). ORF45 is expressed as a lytic viral gene that is critical for efficient and productive virus infection. Although the latent, but not the lytic, infection characterizes cells associated with KSHV-related malignancies, several studies demonstrated the requirement for lytic viral infection to maintain the malignant condition (11). Therefore, ORF45 may play an important role in the pathogenesis of KSHV and its potential transforming activities.

Here, we demonstrate that the potent E3 ligase SIAH-1 is a novel partner of KSHV ORF45 and provide evidence for an important role of SIAH-1 in regulating the stability of KSHV ORF45. Using the full-length KSHV ORF45 as the bait in yeast two-hybrid screens, we identified SIAH-1 and SIAH-2 as ORF45-interacting proteins; these results then were confirmed by coimmunoprecipitation. Like several other cellular proteins that are targeted for degradation by SIAH (36, 40), we found that this interaction is mediated by the substrate binding domain of SIAH-1. SIAH-1 and SIAH-2 contain similar substrate recognition sites and share many substrates. Still, some SIAH partners are specifically degraded by only one isoform; for example, β-catenin and protein inhibitor of activated STAT (Pias) are uniquely targeted for degradation by SIAH-1 and SIAH-2, respectively (6, 32), suggesting certain differential functions for SIAH proteins with their partners. Furthermore, the transcription of SIAH-1 and SIAH-2 is regulated by different signaling pathways. Here we demonstrated, both in yeast and in mammalian cells, the association of both SIAH-1 and SIAH-2 with ORF45. The functional association between ORF45 and SIAH-1 was more fully characterized; however, we obtained preliminary indications for a similar functional association between ORF45 and SIAH-2 (data not shown). Whether both SIAH-1 and SIAH-2 or one of them is most critical for the regulation of ORF45 stability during virus infection is a matter for future studies. Moreover, it will be informative to decipher the transcription, regulation, and expression of SIAH during different phases of KSHV infection.

Several lines of evidence establish functional interactions between SIAH-1 and ORF45. First, the steady-state level of ORF45 decreased in SIAH-1-overexpressing cells, whereas it increased in cells that overexpress a RING mutant DN SIAH-1 that inhibits the endogenous SIAH proteins. This decrease was reversed upon proteasomal inhibition with MG132, and ORF45 protein accumulated with a larger relative increase in cells overexpressing SIAH-1. The inhibition of protein synthesis by cycloheximide confirmed that ORF45 is regulated by SIAH-1 at the protein level, since the overexpression of SIAH-1 reduced the stability of ORF45, whereas a DN SIAH-1 mutant increased the stability of ORF45 protein. In addition, in agreement with the above findings, purified components of the ubiquitylation pathway were shown to mediate efficient in vitro ubiquitylation of ORF45 in an SIAH-1-dependent manner. This finding also was confirmed in mammalian cells. Furthermore, the stability of an ORF45 mutant that did not interact with SIAH-1 was resistant to manipulations of SIAH-1, and it was not targeted to ubiquitylation, as shown by the in vitro ubiquitylation assay. Finally, endogenous ORF45 coimmunoprecipitated with SIAH-1 protein and was increased in cells that overexpressed a DN SIAH-1 mutant, suggesting that the SIAH-mediated proteasomal degradation pathway is active in KSHV-infected cells. Thus, our data suggest that ORF45 is targeted by SIAH-1 for efficient ubiquitylation and subsequent degradation.

Through their interactions with a range of cellular targets, SIAH proteins appear to play critical roles in a variety of cellular processes, such as transcription regulation, mitosis, apoptosis, and tumor progression (36). The activities of SIAH also have been suggested to participate in hypoxia responses. In particular, following exposure to hypoxia, the transcription of siah-2 is induced, which enhances the degradation of the prolyl-hydroxylase proteins and consequently increases the abundance of HIF-1α (33, 34). Recently, the involvement of SIAH-1 in a cell death cascade was identified. Cell stress activates nitric oxide, leading to the s-nitrosylation of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), which activates its binding to SIAH-1 and stimulates the nuclear translocation of GAPDH-SIAH-1 complexes, facilitating the degradation of SIAH-1 nuclear substrates (13, 14). Some studies suggest that SIAH-1 is induced in response to DNA damage and acts as a downstream effector of p21 and p53 (9, 19, 37). SIAH activity in this case is linked to cell cycle arrest via concomitant ubiquitylation and degradation of β-catenin. Alternatively, such degradation can be mediated by another ubiquitylation pathway that requires glycogen-synthase kinase 3β (GSK3β) to phosphorylate β-catenin (9, 28, 29, 32). A similar type of dual differential regulation of β-catenin levels was demonstrated by the hepatitis B virus X protein (HBx). In the presence of p53, HBx downregulates β-catenin through the induction of SIAH-1 expression at the transcriptional level, whereas in the absence of p53, HBx stabilizes β-catenin through the inhibition of GSK3β (24). Notably, it was recently reported that the principal EBV oncoprotein, latent membrane protein 1 (LMP-1), has distinct effects on the expression of SIAH-1 in different cell types (20, 25). These effects in turn regulate the expression of β-catenin and HIF-1α in EBV-infected cells.

In conclusion, the ability of SIAH to mediate the ubiquitylation of ORF45 implicates SIAH proteins as regulators of virus infection and hints, for the first time, at an important role for SIAH in the pathogenesis and tumorigenesis caused by KSHV infection. At present, our working hypothesis suggests that SIAH functions as a two-edged cellular sword. On the one hand, SIAH proteins may inhibit viral infection through the degradation of ORF45; such degradation may reduce virus ingress and egress and enhance the cellular IFN antiviral response. On the other hand, if a reduction in the expression of ORF45 is necessary for the coordinated progression of the infectious cycle, then SIAH activity also could serve the interests of the virus as opposed to those of the host cell. Moreover, besides the regulation of ORF45 stability, the pleiotropic effects of SIAH proteins suggest their participation in other pathways during KSHV infection. Extensive studies will be needed to determine whether any KSHV-encoded proteins or transcripts control the expression and function of SIAH proteins during different phases of infection.