|Home | About | Journals | Submit | Contact Us | Français|
The death-associated protein Daxx found in PML (promyelocytic leukemia protein) nuclear bodies (PML-NBs) is involved in transcriptional regulation and cellular intrinsic antiviral resistence against incoming viruses. We found that knockdown of Daxx in a nontransformed human hepatocyte cell line using RNA interference (RNAi) techniques results in significantly increased adenoviral (Ad) replication, including enhanced viral mRNA synthesis and viral protein expression. This Daxx restriction imposed upon adenovirus growth is counteracted by early protein E1B-55K (early region 1B 55-kDa protein), a multifunctional regulator of cell-cycle-independent Ad5 replication. The viral protein binds to Daxx and induces its degradation through a proteasome-dependent pathway. We show that this process is independent of Ad E4orf6 (early region 4 open reading frame 6), known to promote the proteasomal degradation of cellular p53, Mre11, DNA ligase IV, and integrin α3 in combination with E1B-55K. These results illustrate the importance of the PML-NB-associated factor Daxx in virus growth restriction and suggest that E1B-55K antagonizes innate antiviral activities of Daxx and PML-NBs to stimulate viral replication at a posttranslational level.
PML (promyelocytic leukemia protein) nuclear bodies (PML-NBs) represent cellular multiprotein complexes assembling in large distinct foci within the interchromosomal space of the nucleus (56). Previous studies identified PML as the scaffold protein of the PML-NBs, required for the assembly and recruitment of the major components Daxx, Sp100, ATRX, and SUMO (53). These nuclear structures have been implicated in processes such as transcriptional regulation, genome stability, apoptosis, and tumor suppression (2, 8, 45, 47, 70, 75, 89). Recent data indicate that PML-NBs and their components contribute to an intrinsic cellular defense mechanism repressing virus replication (17, 18, 79).
The 740-amino-acid protein Daxx (death-domain-associated protein) is a multifunctional phosphoprotein, which contains a coiled-coil domain, an acidic region, and a C-terminal serine/proline/threonine-rich region (69). Daxx was described as a regulatory factor in FAS-dependent apoptosis originally (87). It has been well established that Daxx interacts and thereby represses several transcription factors: e.g., Pax3, ETS1, E2F1, NF-κB, p53, and p73 (15, 26, 31, 32, 40, 48, 54). Additionally, a transcriptionally repressive function of Daxx was reported to be mediated by interaction with ATRX, histone deacetylase II (HDAC II), core histones, and the chromatin-associated protein DEK (31, 37, 60, 76, 86).
Several studies showed that viruses have evolved multiple mechanisms to counteract cellular antiviral response by encoding proteins that target PML-NBs (18, 79). Herpes simplex virus type 1 (HSV-1), human cytomegalovirus (HCMV), simian virus 40 (SV40), and adenovirus type 5 (Ad5) infection leads to colocalization of viral genomes with PML-NBs and gives rise to viral replication centers in close proximity (35, 51, 52). Interestingly, Ad5 E1A and E4orf3 (early region 4 open reading frame 3) colocalize with PML-NBs after infection, resulting in track-like rearrangement of these nuclear structures (14, 63, 78). ICP0 (intracellular protein 0) of HSV-1 was shown to be sufficient for the disruption of PML-NBs via proteasomal degradation of PML, and HCMV IE2 (immediate-early protein 2) appears to be necessary for disruption of PML-NBs consequently to abrogation of PML SUMOylation (17). Daxx was recently identified as an additional target for proteasomal degradation after HCMV infection mediated by interaction with HCMV tegument protein pp71 (34).
We and others identified E1B-55K (early region 1B 55-kDa protein) as an Ad5 protein interacting with Daxx (72, 88). E1B-55K is a multifunctional phosphoprotein, promoting efficient viral replication via a number of different mechanisms. In the early phase of Ad5 infection, E1B-55K protein counteracts antiproliferative processes induced by the host cell, including activation of p53-dependent and -independent apoptosis, induction of cell cycle arrest, and stimulation of cellular DNA damage response (82, 84). In the late phase, E1B-55K controls efficient late viral protein production by stimulating the preferential cytoplasmic accumulation and translation of the viral late mRNAs (13, 21). These multiple functions of E1B-55K require interaction with E4orf6 (early region 4 open reading frame 6). Recent work demonstrates that E4orf6 connects E1B-55K to a variety of cellular proteins termed cullin-5, Rbx1/RCO1/Hrt1, and elongins B and C to assemble an SCF (Skp, cullin, F-box)-like E3-ubiquitin-ligase complex allowing the proteasomal degradation of interacting factors like p53, Mre11, DNA ligase IV, and integrin α3 subunit (1, 12, 64, 74).
In this study, we demonstrate that Daxx represses Ad5 replication in infected nontransformed human hepatocytes (HepaRG). Biochemical approaches determined enhanced total virus growth as well as synthesis of early viral proteins, viral mRNA, and DNA in Daxx-depleted human cells. These data provide evidence for a major role of Daxx in an intrinsic defense mechanism of the host cell against Ad5. We observe that Daxx steady-state concentrations are antagonized by an E1B-55K-mediated proteasomal degradation via a cullin-5-dependent E3-ubiquitin-ligase late in Ad5 productive infection.
HepaRG cells (28), HEK293 cells (27), H1299 cells (55), U2OS (61), SAOS-2 (23), and A549 cells (DSMZ ACC 107; Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany) were grown as monolayer cultures in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS), 100 U of penicillin, and 100 μg of streptomycin per ml in a 5% CO2 atmosphere at 37°C.
For HepaRG cells, the medium was supplemented with 5 μg/ml of bovine insulin and 0.5 μM hydrocortisone. To generate Daxx (HAD) and PML (HAP) knockdown cell lines, the nontransformed HepaRG cells were transduced with lentiviral vectors expressing short hairpin RNA (shRNA) targeted to the coding strand sequence 5′-GGAGTTGGATCTCTCAGAA-3′ located at nucleotides (nt) 626 to 643 in Daxx and PML as described previously (19, 49, 54). Transduced cells were selected and maintained in medium containing puromycin (1 μg/ml).
Adenoviral proteins examined in this study were expressed from their respective complementary DNAs under the control of the cytomegalovirus (CMV) immediate-early promoter, derived from the pcDNA3 vector (Invitrogen) to express Ad5 wild-type E1B-55K and E4orf6 (16, 58, 68). For transient transfection, subconfluent H1299 cells were treated with a transfection mixture of DNA and 25-kDa linear polyethylenimine (PEI; Polysciences). Prior to transfection, the culture medium was removed from the cells and replaced by DMEM supplemented with 10% FCS without antibiotics. The transfection solution was prepared by incubating a mixture of DNA, PEI, and DMEM in a ratio of 1 μg DNA to 10 μl PEI to 100 μl DMEM for 10 min at room temperature. After application of transfection solution, cells were incubated for 6 to 8 h in a 5% CO2 atmosphere at 37°C before replacement of the medium with DMEM supplemented with 10% FCS, 100 U of penicillin, and 100 μg of streptomycin per ml.
H5pg4100 served as the wild-type Ad5 parent virus in these studies (29). The mutant viruses H5pm4149 and H5pm4154 were generated as described recently (5, 42). Both viruses carry mutations in the E1B-55K (H5pm4149) or in the E4orf6 (H5pm4154) open reading frame and do not express the respective viral protein (5, 42). Viruses were propagated and titrated in HEK293 monolayer cultures. Infections were performed as described previously (42).
To measure virus growth, infected cells were harvested at 24 and 48 h postinfection (p.i.) and lysed by three cycles of freeze-thawing. The cell lysates were serially diluted in DMEM for infection of HEK293 cells, and virus yield was determined by quantitative E2A immunofluorescence staining at 24 h after infection. Viral DNA replication was monitored by quantitative PCR. HepaRG parental and HAD cells were infected with wild-type virus H5pg4100. Cells were harvested at 48 h postinfection, and total cell extracts were prepared and treated with proteinase K. Quantitative real-time PCR was performed using hexon-specific primers (hexon-qPCR-fw, 5′-CGCTGGACATGACTTTTGAG-3′; hexon-qPCR-rev, 5′-GAACGGTGTGCGCAGGTA-3′). Ad5 H5pg4100 bacmid DNA was used as a control to obtain a standard curve.
For protein analysis, cells were resuspended in NP-40 lysis buffer (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 5 mM EDTA, 0.15% Nonidet P-40) supplemented with a protease inhibitor cocktail containing 1% (vol/vol) phenymethylsulfonyl fluoride (PMSF), 0.1% (vol/vol) aprotinin, 1 μg/ml leupeptin, 1 μg/ml pepstatin, and 1 mM dithiothreitol (DTT). After 1 h on ice, the lysates were sonicated and the insoluble debris was pelleted at 15,000 × g at 4°C.
Primary antibodies specific for Ad proteins used in this study included E1A mouse monoclonal antibody (MAb) M73 (30), E1B-55K mouse MAb 2A6 (71), E2A-72K mouse MAb B6-8 (66), E4orf6 mouse MAb RSA3 (50), L4-100K rat MAb 6B-10 (44), E4orf3 rat MAb 6A11 (57), E4orf6 rabbit polyclonal antibody 1807 (6), and Ad5 rabbit polyclonal serum L133 (41). Primary antibodies specific for cellular proteins included Daxx rabbit polyclonal antibody (Upstate/Millipore), PML rabbit polyclonal antibody H-238 (Santa Cruz), Mre11 rabbit polyclonal antibody pNB 100-142 (Novus Biologicals, Inc.), p53 mouse MAb DO-1 (Santa Cruz Biotechnology, Inc.) (81), and β-actin mouse MAb AC-15 (Sigma-Aldrich, Inc.).
Secondary antibodies conjugated to horseradish peroxidase (HRP) for detection of proteins by immunoblotting were anti-rabbit IgG, anti-mouse IgG and anti-rat IgG (Jackson/Dianova). Fluorescent secondary antibodies were affinity-purified fluorescein isothiocyanate (FITC)-conjugated donkey anti-mouse IgG and Texas Red-conjugated donkey anti-rat IgG (Invitrogen). These were used at a 1:100 dilution in all immunofluorescence experiments.
For immunoblotting, equal amounts of total protein were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes (Schleicher & Schuell/Whatman). Membranes were incubated as described previously (42). The bands were visualized by enhanced chemiluminescence as recommended by the manufacturer (Pierce) on X-ray films (CEA RP new medical X-ray film). Autoradiograms were scanned and cropped using Adobe Photoshop CS4, and figures were prepared using Adobe Illustrator CS4 software.
For indirect immunofluorescence, cells were grown on glass coverslips as described recently (42). At the indicated times, cells were fixed in ice-cold methanol at −20°C for 15 min and permeabilized in phosphate-buffered saline (PBS)-0.5% Triton X-100 for 30 min at room temperature. After 1 h of blocking in Tris-buffered saline-bovine serum albumin-glycine (TBS-BG) buffer, the cells were treated for 1 h with the primary antibody diluted in PBS and washed three times in PBS-0.1% Tween 20, followed by incubation with the corresponding FITC- or Texas Red-conjugated secondary antibodies (Invitrogen). Coverslips were washed three times in PBS-0.1% Tween 20 and mounted in Glow medium (Energene), and digital images were acquired on a DMRB fluorescence microscope (Leica) with a charge-coupled device camera (Diagnostic Instruments). Images were cropped using Adobe Photoshop CS4 and assembled with Adobe Illustrator CS4.
Subconfluent HepaRG parental and shRNA-expressing (shDaxx) cells (HAD) were infected with wild-type virus at 20 focus-forming units (FFU)/cell. Cells were harvested 12 and 36 h postinfection (p.i.). Total RNA was isolated with Trizol reagent (Invitrogen) as described by the manufacturer. The amount of total RNA was measured, and 1 μg of RNA was reverse transcribed using the Transcriptor high-fidelity cDNA synthesis sample kit from Roche, including anchored-oligo(dT)18 primer specific to the poly(A)+ RNA. To amplify specific viral genes, primers were designed as shown in Table Table1.1. Quantitative reverse transcription (RT)-PCR was performed with a first-strand method in a Rotor-Gene 6000 (Corbett Life Sciences, Sydney, Australia) in 0.5-ml reaction tubes containing a 1/100 dilution of the cDNA template, 10 pmol/μl of each synthetic oligonucleotide primer (Table (Table1),1), and 12.5 μl/sample Power SYBR green PCR master mix (Applied Biosystems). The PCR conditions were as follows: 10 min at 95°C, followed by 55 cycles of 30 s at 95°C, 30 s at 55 to 62°C (depending upon the primer set), and 30 s at 72°C. The average threshold cycle (CT) value was determined from triplicate reactions, and levels of viral mRNA relative to cellular 18S rRNA were calculated as described recently (83). The identities of the products obtained were confirmed by melting curve analysis.
PML-NBs participate in intrinsic antiviral resistance activities, which involve cellular components antagonized by a variety of different viral proteins often compromising the integrity of PML-NBs. As part of our study to analyze the role of Daxx and PML during Ad5 infection, experiments were performed in a nontransformed human hepatocyte cell line termed HepaRG (28) expressing shRNAs depleting either Daxx (this study) or PML (19). To determine whether Ad5 infection of HepaRG cells is equivalent to already published data, total virus yield was performed in wild-type (H5pg4100)-infected hepatocytes and H1299 cells (Fig. (Fig.1A).1A). We observed a reduction of virus growth in HepaRG cells compared to the human tumor cell line at different time points after infection. This indicates that Ad5 infection is supported in HepaRG cells but occurs with a delay compared to wild-type-infected H1299 cells (Fig. (Fig.1A).1A). Further support was obtained by indirect immunofluorescence analysis (Fig. (Fig.1B).1B). Consistent with previous observations in human cell lines, E1B-55K protein is diffusely distributed in wild-type-infected HepaRG cells (Fig. 1Bb, Be, and Bh) (24, 46, 59). E1B-55K forms some distinct cytoplasmic aggregates and localizes in a limited number of globular condensations, which are mostly in the vicinity of ring-like structures containing E2A protein (Fig. 1Ba). The viral E4orf3 protein colocalizes and reorganizes the PML-NBs from nuclear punctate structures into track-like structures in infected hepatocytes (Fig. 1Bk). As already published, the viral E4orf6 protein shows a diffuse staining pattern in the nucleus (Fig. 1Ad) (25, 59). On the basis of these data, we conclude that the human HepaRG cell line is susceptible and permissive for Ad5 infection and sufficient to support the adenoviral life cycle.
To reveal the effect of Daxx and PML depletion, indirect immunofluorescence analysis demonstrated that HAD (shDaxx cell line) and HAP (shRNA-expressing PML cell line) cells exhibit substantially reduced levels of nuclear Daxx (Fig. (Fig.22 Ac) and PML protein (Fig. 2Bc) compared to the control HepaRG parental cells (Fig. 2Aa, Ab, Ba, and Bb). The reduction in Daxx and PML RNA expression was confirmed by real-time PCR analysis (data not shown) and immunoblotting (Fig. (Fig.2C).2C). Interestingly, the steady-state concentration of the Daxx protein was decreased in the HAP cell line (Fig. (Fig.2C,2C, lane 3), indicating that PML may contribute to Daxx stability.
To reveal the effect of Daxx and PML depletion on the production of adenoviral progeny, we determined virus growth in the HepaRG cell lines (Fig. (Fig.33 A). Consistent with recent work from Ullman and Hearing, knockdown of PML increased the production of wild-type virus particles approximately 1.5-fold 48 h p.i. (Fig. (Fig.3A)3A) (80). In contrast, infection of the shDaxx cell line HAD resulted in a 4-fold increase of progeny wild-type virions already 24 h p.i. and a 3.5-fold increase after 48 h p.i. (Fig. (Fig.3A).3A). We note that this negative effect of Daxx on virus progeny production could not be diminished at later times of infection (data not shown). In line with previous reports, virus mutant H5pm4149, which lacks E1B-55K and its related proteins (42) was defective for growth in all hepatocyte cell lines compared to the wild-type virus H5pg4100 (Fig. (Fig.3A).3A). A similar effect was observed when we monitored viral DNA replication in the parental and HAD cells. In line with the results from the virus growth experiments, viral DNA synthesis was more efficient in the Daxx knockdown cell line after 48 h postinfection (Fig. (Fig.3B).3B). Identical results for virus growth were obtained from analysis using Daxx-depleted H1299 and A549 cells (data not shown). These data strongly indicate that Daxx is a negative regulator of Ad5 replication in human cells.
Data from the replication assays (Fig. (Fig.3A)3A) suggest that Daxx mediates its repressive effect during the early phase of the infectious cycle. To test this possibility, expression of viral early and late proteins was monitored at different time points after infection (Fig. (Fig.44 A). Expression of early proteins (E1B-55K, E2A, and E4orf6) was substantially increased in HAD cells compared to the parental cell line (Fig. (Fig.4A).4A). Consistent with this, viral early mRNA production for E1A, E1B, E2A, and E4orf6 was decreased in the parental cells compared to that in the Daxx knockdown line (Fig. (Fig.4B).4B). Furthermore we could also observe an increase in the steady-state concentration of late proteins in infected HAD cells (Fig. (Fig.4A).4A). Taken together, these data strongly suggest that Daxx is a negative regulator of Ad5 replication during early times of the infectious cycle.
To assess the effect of Daxx on overall virus growth properties, total virus yield was determined in HepaRG parental, HAD, and HAD-derived cells (HD-ED) (49) expressing an EYFP-Daxx fusion protein which was reintroduced into HAD cells using lentiviral vectors. The sequence was altered to 5′-GGAGTTAGATCTGAGCGAA-3′ in the engineered cDNA to render an mRNA resistant to the Daxx shRNA, while maintaining the encoded amino acid sequence (49). As anticipated, the negative effect of Daxx expression was partially rescued in the HD-ED cell line (Fig. (Fig.5).5). Altogether, Daxx reconstitution of HAD cells results in a substantially reduced virus growth in nontransformed human hepatocytes.
On the basis of these data, we examined Daxx protein levels in infected HepaRG and H1299 cells (Fig. (Fig.6).6). Daxx protein steady-state levels were reproducibly reduced at late times of infection. This decrease was even more pronounced at a higher multiplicity of infection and was comparable to the reduction of p53 (Fig. (Fig.6)6) known to be a target of the E1B-55K/E4orf6 E3-ubiquitin-ligase complex (9, 64, 67).
Additionally, as part of our studies to identify new host cell substrates of the Ad5 E3-ubiquitin-ligase complex, we infected different human tumor cell lines A549, U2OS, and SAOS-2 with wild-type Ad5 and monitored steady-state levels of several cellular proteins (Fig. (Fig.7).7). Whole-cell extracts were prepared at different times after infection and analyzed for expression of cellular E1B-55K interaction partners, including Mre11, p53, and the PML-NB component Daxx. Consistent with previous publications, we observed significantly reduced steady-state concentrations of Mre11 and p53 during lytic infection (Fig. (Fig.7)7) (64, 74). We could detect reduction of Daxx protein at late times of Ad5 infection in all cell lines tested (Fig. (Fig.77).
Furthermore, we determined Daxx protein concentrations in wild-type- and mutant virus-infected cells lacking either E1B-55K (H5pm4149) or E4orf6 (H5pm4154) (Fig. (Fig.8).8). As before, Daxx was dramatically reduced in cells infected with the wild-type virus H5pg4100 (Fig. (Fig.8A,8A, lane 2), whereas Daxx accumulated to levels comparable to noninfected cells in infected cells lacking E1B-55K (H5pm4149) (Fig. (Fig.8A,8A, lane 3). In contrast, Daxx was also reduced in cells infected with the E4orf6-minus virus mutant H5pm4154 (Fig. (Fig.8A,8A, lane 4). In line with previous work, Mre11 concentrations were reduced in wild-type-infected cells (Fig. (Fig.8A,8A, lane 2), while no reduction was observed in the absence of E1B-55K (H5pm4149; Fig. Fig.8A,8A, lane 3) or E4orf6 (H5pm4154; Fig. Fig.8A,8A, lane 4). The decrease of Mre11 seen in wild-type-infected cells and Daxx additionally observed in H5pm4154-infected cells was abolished when treated with the proteasome inhibitor MG-132 (Fig. (Fig.8B,8B, lanes 2 and 4).
Reduction of Daxx and Mre11 is dependent on cullin-5 as no decrease of Daxx was observed in cullin-5-negative H1299 cells (12) infected with wild-type (Fig. (Fig.8D,8D, lane 2) and mutant (Fig. (Fig.8D,8D, lanes 3 and 4) viruses. These data demonstrate that similar to p53 and Mre11, Daxx protein is degraded via a proteasomal pathway, which requires E1B-55K, but occurs at least in part independent from E4orf6. Consistent with our results, we observed E4orf6-independent degradation of Daxx in other cell lines (e.g., A549 cells) (data not shown).
To test our hypothesis, we analyzed the levels of Daxx protein in transfected cells (Fig. (Fig.9).9). Expression of wild-type E4orf6 alone had no effect on the steady-state concentration of endogenous Daxx protein (Fig. (Fig.9,9, lane 3), while expression of E1B-55K alone (Fig. (Fig.9,9, lane 2) or in combination with E4orf6 (Fig. (Fig.9,9, lane 4) diminished the Daxx steady-state concentration similar to virus-infected cells. The transient assays reflect degradation rather than an effect on protein synthesis as the decrease of Mre11 seen in E1B-55K/E4orf-transfected cells (Fig. (Fig.9,9, lane 4) and Daxx additionally observed in E1B-55K-transfected cells (Fig. (Fig.9,9, lane 2) was abolished, when treated with the proteasome inhibitor (Fig. (Fig.9,9, lanes 5 and 7). Our results show clearly that the transfection assays mimic viral infection and support an E4orf6-independent reduction of Daxx protein concentrations.
Taken together, these data suggest that the proteasomal degradation of Daxx in virus-infected and transfected cells occurs via a cullin-5-dependent mechanism, which requires E1B-55K but not E4orf6, at least in the cell lines tested.
Several lines of evidence indicate that PML-NBs contribute to an intrinsic cellular defense antagonizing virus replication (17, 18, 20). These subnuclear structures are known to assemble more than 30 proteins: e.g., Daxx, Sp100, ATRX, and a variety of enzymes involved in chromatin remodeling, including histone deacetylases (HDACs) and histone methyltransferases (31, 56). Predominantly, recruitment to these nuclear aggregates requires SUMO modification of the scaffold protein PML (56). PML and Daxx can associate with HDACs exhibiting a transcriptionally repressive function (31, 48). Different viruses, including human papillomavirus (HPV), HSV-1, HCMV, Ad5, varicella-zoster virus (VZV), and also complex retroviruses like HIV-1, were shown to reorganize or disrupt PML-NBs and likely evolved mechanisms to counteract PML-associated antiviral defense, resulting in efficient viral progeny production (20, 22, 33, 43, 79, 80, 85).
In this study, we focused on the importance of the cellular transcription factor Daxx during productive Ad5 infection. Daxx is a transcriptional repressor, appearing as a predominantly nuclear protein associated with PML-NBs and also diffusely throughout the nucleoplasm (32, 48, 54). In the cytoplasm, it has been described to interact with proteins involved in cell death regulation (69). Originally, Daxx was reported to be involved in Fas-induced apoptotic signaling. Upon interaction with the death domain of an activated Fas receptor (CD 95), Daxx binds and activates apoptosis signal-regulating kinase 1 (ASK-1), which in turn activates the c-Jun N-terminal kinase (JNK) pathway and promotes apoptosis (10, 11, 87). Here, we demonstrate that depletion of Daxx results in a significantly increased Ad5 replication (Fig. (Fig.33 and and4).4). This is consistent with results obtained with HCMV, where Daxx is linked to negative regulation of productive viral infection. In this context, Daxx was shown to repress the HCMV major immediate-early promoter (MIEP) and consequently reduces viral early gene expression (62). In fact, it is evident that Daxx recruits HDACs to the viral promoter and additionally modulates HDAC-associated chromatin modification (31, 48, 77). Similar to this, Daxx negatively regulates Ad5 mRNA (Fig. (Fig.4B),4B), DNA (Fig. (Fig.3B),3B), and protein synthesis (Fig. (Fig.4A),4A), correlating with downregulation of virus growth properties (Fig. (Fig.3A).3A). It is therefore tempting to speculate that Daxx negatively influences HCMV and Ad5 replication via similar mechanisms. So far, the mechanism of Daxx-mediated Ad5 repression is still unclear, but our studies suggest transcriptional events due to direct repression of viral promoters via Daxx binding or establishment of a closed chromatin structure.
It has already been suggested that Daxx may be depleted from the PML-NBs in the presence of E1B-55K, being dynamically distributed between E1B-55K-p53 protein complexes and PML-NBs (88). Moreover, Daxx localization in the PML-NBs appears to correlate with a prosurvival function (54, 90). In addition, treatment of cells with stimuli like interferon and arsenic trioxide (As2O3) leads to recruitment of Daxx to PML-NBs (39), which might be influenced by Ad5 E1B-55K-Daxx interaction. Taken together, Daxx proapoptotic functions might be inhibited by interaction with antiapoptotic E1B-55K protein and degradation of the cellular factor (Fig. (Fig.66 to to9).9). Further support for this hypothesis is obtained from the fact that Daxx-mediated activation of JNK-dependent apoptosis is suppressed by members of the Bcl-2 family of proteins (87). Daxx disassembly from PML-NBs could be considered to be a basic step in prevention of the death of the host cell due to Ad5 infection. As Ad5 E1B-19K protein is known to be a Bcl-2-homologue, we suggest a close connection between antiapoptotic functions of Ad5 E1B-proteins and functional inactivation of Daxx.
Recent work has also shown that E1B-55K and E4orf6 form a ubiquitin-dependent E3-ligase complex that plays an important role in targeting the cellular factors p53, Mre11, DNA ligase IV, and integrin α3 for proteasomal degradation (1, 4, 12, 65, 67, 73, 74). Daxx steady-state concentrations are also reduced during lytic Ad5 infection (Fig. (Fig.8).8). Remarkably, in contrast to the known cellular targets of the Ad5 E3-ligase, E4orf6 is dispensable for Daxx reduction (Fig. (Fig.88 and and9).9). This indicates that proteasomal degradation of Daxx is likely independent from the activity of the E1B-55K-E4orf6 E3-ligase complex. In fact our data suggest that E1B-55K alone is sufficient to assemble components that form an E3-ubiquitin-ligase complex. In line with this hypothesis, a highly conserved BC-box motif, known to mediate the interaction with elongins B and C, containing the consensus sequence (A,P,S,T)LxxxCxxx(A,I,L,V) in which the leucine and cysteine residues are almost invariable, is present in the Ad5 E1B polypeptide (179ALRPDCKYKI188) (4). Also, E1B-55K contains a Cys/His-rich region which may be part of a RING-like domain in its central part (3) and thus may play a critical role in mediating ubiquitin transfer to heterologous substrates (38). A similar situation has been described for the tegument protein pp71 from HCMV, which promotes immediate-early viral gene expression through binding to and proteasomal degradation of Daxx in the context of PML-NBs (7, 34, 36).
In summary, these results together with recent findings from Hearing and coworkers (80) demonstrate that Daxx is a negative regulator of Ad5 replication likely operating at the level of transcriptional repression during the early phase of the infection. Furthermore, our data show that this negative effect is counteracted by E1B-55K, which induces the proteasomal degradation of Daxx via a novel E4orf6-independent E3-ubiquitin-ligase activity.
We thank Philippe Gripon for providing the HepaRG cell line.
This work was supported by grants from the DFG and Stiftung für neuronale Erkrankungen. Peter Wimmer was supported by the “Studienstiftung des Deutschen Volkes e.V,” Bonn, Germany. The Heinrich-Pette-Institute is supported by the Freie und Hansestadt Hamburg and the Bundesministerium für Gesundheit (BMG).
Published ahead of print on 19 May 2010.