Relocalization of the Mre11-Rad50-Nbs1 Complex by
http://www.100md.com
病菌学杂志 2005年第10期
Department of Molecular Genetics and Microbiology, Stony Brook University School of Medicine, Stony Brook, New York
ABSTRACT
Adenovirus replication is controlled by the relocalization or modification of nuclear protein complexes, including promyelocytic leukemia protein (PML) nuclear domains and the Mre11-Rad50-Nbs1 (MRN) DNA damage machinery. In this study, we demonstrated that the E4 ORF3 protein effects the relocalization of both PML and MRN proteins to similar structures within the nucleus at early times after infection. These proteins colocalize with E4 ORF3. Through the analysis of specific viral mutants, we found a direct correlation between MRN reorganization at early times after infection and the establishment of viral DNA replication domains. Further, the reorganization of MRN components may be uncoupled from the ability of E4 ORF3 to rearrange PML. At later stages of infection, components of the MRN complex disperse within the nucleus, Nbs1 is found within viral replication centers, Rad50 remains localized with E4 ORF3, and Mre11 is degraded. The importance of viral regulation of the MRN complex is underscored by the complementation of E4 mutant viruses in cells that lack Mre11 or Nbs1 activity. These results illustrate the importance of nuclear organization in virus growth and suggest that E4 ORF3 regulates activities in both PML nuclear bodies and the MRN complex to stimulate the viral replication program.
INTRODUCTION
Adenovirus (Ad) replication involves a number of events that must be temporally and spatially organized within the host cell in order to lead to optimal productive infection. The 36-kbp Ad genome encodes at least 25 early gene products (9). Previous studies showed that early region 4 (E4) encodes two proteins of particular importance to viral replication, the products of open reading frames (ORFs) 3 and 6 (E4 ORF3 and E4 ORF6) (4, 15). In the absence of E4 gene products, early gene expression occurs normally, but a profound defect in viral DNA replication is evident (12, 30). E4 ORF3 and E4 ORF6 have compensatory functions to stimulate efficient viral DNA replication independently of one another (4, 15). E4 ORF3 and E4 ORF6 proteins both bind to the E1B 55-kDa protein and play roles in the regulation of late virus RNA transcription and splicing, inhibition of Ad genome concatenation, and stimulation of viral growth (28).
The E4 ORF3 protein plays an important role in modifying the host nucleus during the beginning stages of viral infection. E4 ORF3 interacts with and reorganizes distinct nuclear structures known as nuclear domain 10, promyelocytic leukemia protein (PML) oncogenic domains, or PML nuclear bodies (6, 8). These domains are of unknown function but have been implicated in a number of cellular processes, including transcriptional regulation, cellular growth control, response to DNA damage, apoptosis, and multiple cellular defense mechanisms (2, 14, 22). PML nuclear domains are highly sensitive to heat, stress, and interferon treatment. PML nuclear bodies are targeted by numerous DNA virus early gene products (21, 32). Deregulation of PML nuclear domains has been linked to a severe hematopoietic disease, acute promyelocytic leukemia (20). This form of leukemia is the result of chromosomal translocation resulting in the production of a fusion protein that consists of PML linked to retinoic acid receptor . PML nuclear domains are aberrantly localized as a result of the expression of the PML-retinoic acid receptor fusion. Treatment of acute promyelocytic leukemia patients with all-trans retinoic acid or arsenic trioxide leads to disease remission and restoration of PML nuclear body integrity, suggesting the need for a particular organization of PML and PML domain resident proteins for normal cell function. It has been shown that PML is the nucleating protein of the nuclear domains (17). Without PML, the other resident proteins are redistributed throughout the nucleus. E4 ORF3 is necessary and sufficient to direct the redistribution of PML nuclear bodies into a track-like structure that localizes to the periphery of Ad DNA replication centers (6, 8).
The Mre11-Rad50-Nbs1 (MRN) complex plays a critical role in the cellular response to double-strand breaks and DNA damage (7, 26). Mutations in the Mre11 and Nbs1 genes have been implicated in the human diseases ataxia-telangiectasia-like disorder and Nijmegen breakage syndrome, respectively. These genetic diseases result in a higher incidence of cancer and growth defects in afflicted individuals (7, 26). Cell lines founded from patients with these diseases have been shown to be sensitive to ionizing radiation. The MRN complex plays critical roles in the detection and repair of DNA damage. The E4 ORF3 and E4 ORF6 proteins interfere with the cellular DNA damage response to Ad infection (3, 27, 29). Without E4 gene expression, the linear double-stranded Ad DNA genome activates the host DNA damage response, and viral genomes are concatenated by cellular repair pathways (27). This event effectively blocks viral DNA replication, since the Ad terminal repeats, which contain the origins of replication, are buried and mutated within genomic concatemers. Ad genome concatenation requires the cellular MRN complex and DNA ligase IV as well as the checkpoint signaling protein kinases DNA protein kinase (PK) and ATM (3, 27). The E4 ORF3 and E4 ORF6 proteins independently interact with the catalytic subunit of DNA PK (3). The E4 ORF6-E1B 55-kDa complex targets p53 and Mre11 for ubiquitin-mediated proteasome-dependent degradation, effectively interfering with the cellular DNA damage response checkpoint to block viral genome concatenation (13, 25, 27).
The interaction of E4 ORF3 with the DNA damage response machinery is intriguing. The expression of E4 ORF3 was shown to effect the relocalization of nuclear MRN components, and E4 ORF3 was found to colocalize with MRN proteins (27), but no direct link between these observations and E4 ORF3 function was known. MRN foci localized to the periphery of viral DNA replication centers (27). This finding led to the speculation that E4 ORF3 may sequester MRN components from viral DNA to prevent cellular repair of the viral genome. However, this idea is at odds with previous work in which we found that several E4 ORF3 mutant viruses showed wild-type growth kinetics yet accumulated a large number of viral genome concatemers (10). One such mutant effectively rearranged MRN components, drawing into question the role of MRN reorganization in the viral life cycle.
The aim of this study was to elucidate the function of E4 ORF3 relocalization of MRN complex proteins during Ad infection. We found that E4 ORF3 reorganizes PML and MRN complex proteins in a similar manner within the nuclei of Ad-infected cells. The relocalization of MRN proteins precisely correlated with the growth properties of mutant viruses. Further, cells harboring mutations in Mre11 and Nbs1 proteins but not several other effectors in the cellular DNA damage response were able to complement the growth of viruses that do not express functional E4 ORF3 and E4 ORF6 proteins. These findings emphasize the importance of the MRN complex as a target of Ad E4 function and suggest that a very early step in MRN function plays the most important role in regulating the Ad replication cycle.
MATERIALS AND METHODS
Viruses, cells, and growth assays. Ad mutant viruses dl355 (E4 ORF6 negative [E4 ORF6–]), inORF3 (E4 ORF3 negative [E4 ORF3–]), dl355/inORF3 (E4 ORF3 negative and E4 ORF6 negative [E4 ORF3/6–]), and dl355/pmD105L106 (E4 ORF6–/ORF3 D105A/L106A point mutations) were described previously (10, 12, 15). Mutants dl355/pmN82 (27) and dl355/pmL103 contain alanine substitution mutations at E4 ORF3 amino acids 82 and 103, respectively, in a dl355 virus background and were constructed as previously described (10). A549 and W162 cells were maintained in Dulbecco's modified minimal essential medium containing 10% calf serum. ATLD1 (Mre11 negative [Mre11–]; gift from John Petrini, Memorial Sloan Kettering Institute), GM07166 (Nbs1 negative [Nbs1–]; Coriel Institute for Medical Research), M059J (DNA PK negative [DNA PK–]; American Type Culture Collection), M059K (DNA PK positive [DNA PK+]; American Type Culture Collection), BLM (Bloom helicase negative [BLM–]; gift from Norma Neff, New York Blood Center), 180BRM (DNA ligase IV negative [DNA ligase IV–]; gift from Huichen Wang, Jefferson Medical College), and IMR90 (American Type Culture Collection) were maintained in Dulbecco's modified minimal essential medium supplemented with 15% fetal bovine serum at 37°C in 5.0% CO2.
All viral infections were carried out in an identical manner. Purified virus particles were prepared by CsCl equilibrium gradient centrifugation. Cells were infected with 200 virus particles/cell for 1 h at 37°C, the virus inoculum was removed, the cells were washed twice, and fresh medium was added. The infection mixtures were incubated for various times at 37°C in 5.0% CO2. Single-step growth curves were used to measure virus growth in A549 cells and cells with mutant DNA repair proteins. For A549 cells, cells were harvested at 4, 20, 32, and 44 h postinfection. For cells with mutant DNA repair proteins, cells were harvested at 4, 72, and 120 h postinfection. Cellular lysates were prepared by four cycles of freezing/thawing, and infectious virus yields were determined by plaque assays with E4-complementing cell line W162 (31).
Immunofluorescence analysis. A549 cells were placed on coverslips and infected as described above. At 6, 9, 12, and 15 h postinfection, the cells were fixed with –20°C methanol for 5 min. The cells were washed with phosphate-buffered saline (PBS) and blocked in PBS containing 10% goat serum for 1 h. Primary antibodies were diluted in PBS containing 1.5% goat serum and incubated on coverslips for 1 h at room temperature. Antibodies used were anti-PML mouse monoclonal antibody (PG-M3; Santa Cruz Biotechnology), anti-E4 ORF3 rat monoclonal antibody (6A-11; gift from Thomas Dobner, Regensberg Universitat), anti-DNA binding protein (DBP) monoclonal antibody (gift from Arnold Levine, Princeton University), anti-DBP polyclonal antibody (gift from Peter van der Vleit, University Medical Centre Utrecht), anti-Mre11 rabbit polyclonal antibody (100-142; Novus Biologicals), anti-Rad50 mouse monoclonal antibody (100-147; Novus Biologicals), and anti-Nbs1 rabbit polyclonal antibody (100-143; Novus Biologicals). The coverslips were washed three times with PBS; incubated with fluoroscein isothiocyanate (FITC)-conjugated goat anti-mouse immunoglobulin G (IgG) (Zymed), Texas Red isothiocyanate (TRITC)-conjugated goat anti-rat IgG (Zymed), and Alexa 350-labeled goat anti-rabbit IgG (Molecular Probes) for 45 min at room temperature in the dark; and washed again three times with PBS. The coverslips were mounted on slides in Fluoromount G (Southern Biotechnology Associates). Immunofluorescence was viewed and photographed by using a Zeiss microscope with Axiovision software.
Viral DNA replication, pulsed-field gel electrophoresis (PFGE), and Western blot analysis. Viral DNA replication was determined by Southern blot analysis. Total nuclear DNA was isolated at 6, 12, 18, and 24 h after infection as previously described (10). The samples were diluted 1:10, 1:30, 1:100, and 1:200 in Tris-EDTA, respectively, and the DNA was denatured and prepared for hybridization as previously described. The DNA was bound to a Hybond-N membrane (Amersham) by the slot blot method followed by UV cross-linking. The probe was an Ad5 (positions 1 to 194) PCR fragment fluorescently labeled with the Atto-Phos system (Promega). The probe was hybridized to the membrane and detected according to the manufacturer's instructions. The signal was visualized and quantitated by using a Molecular Dynamics Storm 800 PhosphorImager and ImageQuant software.
For PFGE, infected Nbs1– cells were harvested at 72 h postinfection and prepared as previously described (10). Briefly, cells were harvested by trypsinization, washed once with PBS, and resuspended in 100 μl of PBS with 125 mM EDTA. Next, 100 μl of 1.25% low-melting-temperature agarose in 50 mM Tris (pH 7.4) and 125 mM EDTA was added to the cells. The solidified plugs were treated with 1.2% sodium dodecyl sulfate (SDS), 0.125 mM EDTA, and 1 mg/ml proteinase K overnight at 50°C. The plugs were washed three times over 8 h with 50 mM EDTA. A 1.0% agarose gel was loaded with 30 μl of melted plug. The gel was run at 6 V/cm with a switching time of 6 s for 20 h. The DNA was transferred to a Hybond-N nylon membrane (Amersham), probed with a 32P-labeled Ad total genome probe, and visualized by autoradiography as previously described (10).
For Western blot analysis of wild-type and mutant E4 ORF3 proteins, A549 cells were infected with E1 replacement viruses expressing hemagglutinin (HA)-tagged E4 ORF3 wild-type and mutant proteins for 12 h as previously described (10). Infected cells were harvested and lysed in radioimmunoprecipitation buffer (150 mM NaCl, 1.0% NP-40, 0.5% deoxycholic acid, 0.1% SDS, 50 mM Tris [pH 8.0]) supplemented with a protease inhibitor cocktail.
Proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to a Hybond-P membrane (Amersham) by standard methods. The membrane was probed with anti-HA mouse monoclonal antibody (Roche), and proteins were visualized with enhanced chemiluminescence reagents (Amersham).
RESULTS
Analysis of E4 ORF3 mutant viruses. A group of recombinant viruses was constructed and analyzed that expressed wild type and mutant E4 ORF3 proteins from the natural promoter in the absence of E4 ORF6 (dl355). Under these conditions, efficient viral DNA replication and growth are dependent on E4 ORF3 function (4, 10, 15). Single and double alanine substitution mutations were introduced that spanned E4 ORF3. Viral growth properties were analyzed in a single-step growth curve (Fig. 1) and by analyzing viral DNA replication (Fig. 2). In total, sixteen E4 ORF3 mutants were analyzed. Only three mutant viruses (dl355/pmN82 [27], dl355/pmL103, and dl355/pmD105L106) were severely defective for virus growth; pmD105L106 was previously reported (10) and dl355/pmN82 and dl355/pmL103 are newly described viral mutants. In single-step growth curves in noncomplementing A549 cells, these viruses replicated to levels equally reduced compared to a virus that lacks both E4 ORF3 and E4 ORF6 proteins (dl355/inORF3) (Fig. 1). These E4 ORF3 mutant viruses were decreased 100-fold for growth when compared to the parent virus dl355. This reduction was reflected in comparably reduced viral DNA replication levels (Fig. 2 and data not shown). Comparable levels of wild type and mutant proteins were detected by Western blot analysis (Fig. 2A, wild type versus N82 and L103; for D105/L106, see reference 10). These viruses provide a number of mutants with which to study E4 ORF3 function in the context of natural infection.
E4 ORF3 relocalizes MRN complex components. Recently, the DNA damage protein MRN complex was shown to localize to foci correlating to areas of Adenovirus replication in the absence of E4 proteins during virus infection (27). Also, E4 ORF3 was able to reorganize the MRN proteins to track-like structures that partially overlapped with rearranged PML in a transient transfection assay (27), but the functional consequences of this rearrangement was not known. Here we show that E4 ORF3 reorganization of MRN is critical for virus replication.
Immunofluorescence was used to examine Mre11, Rad50, Nbs1, and PML localization following infection with wild type Ad5 (Fig. 3). In uninfected cells, MRN components were diffusely localized throughout the nucleus while PML was localized to discreet nuclear bodies (Fig. 3A, E, I, and M). At 6 h postinfection, wild-type E4 ORF3 caused Mre11, Rad50, and Nbs1 to relocalize to track-like structures throughout the nuclei that showed almost complete colocalization with E4 ORF3 (Fig. 3B to D, F to H, and J to L). Rearranged MRN proteins formed structures that were similar to those observed with PML and E4 ORF3 (Fig. 3N to P). Such a rearrangement of the MRN proteins was found whether or not E4 ORF6 protein was expressed (wild type dl309, E4 ORF6+ [Fig. 3]; dl355, E4 ORF6– [data not shown]). MRN relocalization was completely dependent on E4 ORF3 since no rearrangement was observed with any viral mutant that did not express this protein (data not shown).
As the infection progressed, the pattern of MRN reorganization continued to change (Fig. 4). When E4 ORF6 was expressed, Mre11 levels declined (Fig. 4A), as previously reported (27). Rad50 was maintained in association with E4 ORF3 in track-like structures that surrounded, but did not overlap with, viral DNA replication centers (Fig. 4E to H); viral DNA binding protein (DBP) staining was used to identify replication centers. Nbs1 localization significantly changed as the infection progressed and almost complete colocalization of Nbs1 with viral replication centers was evident by 15 h postinfection (Fig. 4J to L). In the absence of E4 ORF6 protein, identical patterns were observed for Rad50 and Nbs1 (data not shown), whereas Mre11 was still evident in association with E4 ORF3 (Fig. 4M to P). We conclude that E4 ORF3 significantly alters the subnuclear localization and organization of MRN complex components. Mre11, Rad50 and Nbs1 are all targeted to E4 ORF3-containing tracks early after infection, whereas the MRN complex is disrupted at later times after infection to disperse the components.
MRN reorganization is required for efficient viral DNA replication. Using mutant viruses that were either replication competent or defective, we analyzed the ability of E4 ORF3 mutants to reorganize MRN components and PML nuclear domains at 6 h postinfection (Fig. 5). Mutants dl355/pmN82 and dl355/pmL103 were defective for growth and viral DNA replication (Fig. 1 and 2) and were unable to reorganize Rad50 (Fig. 5C and E), PML (Fig. 5D and F), or Mre11 and Nbs1 (data not shown). The E4 ORF3 proteins produced by these viruses were diffusely localized throughout the cell (most obvious in Fig. 5D and F). Mutant dl355/pmD105L106 also was defective for replication and was unable to reorganize Rad50 (Fig. 5G) or Mre11 and Nbs1 (data not shown). The D105L106 mutant protein, however, was able to form tracks itself (red fluorescence in Fig. 5G) and efficiently reorganized PML early after infection (Fig. 5H). All three of these mutant viruses were unable to establish DBP-containing replication centers consistent with their replication defect, but produced DBP to levels equivalent as wild type virus (data not shown). Identical results were found at 15 h postinfection. Viruses expressing wild-type and mutant E4 ORF3 proteins that grew efficiently directed Rad50 and PML relocalization (Fig. 5A, B, and I to L) as well as Mre11 and Nbs1 (data not shown). The entire panel of E4 ORF3 mutant viruses (sixteen mutants) was examined for their abilities to reorganize Rad50 and PML. In every case, a direct correlation was found between the ability of the mutant proteins to reorganize Rad50 early after infection with the abilities of the mutant viruses to replicate in infected cells (data not shown). Thus, we conclude that reorganization of the MRN complex early after Ad infection by the E4 ORF3 protein is required for efficient viral DNA replication and to establish viral replication centers in the absence of E4 ORF6 expression.
Loss of Mre11 and Nbs1 complements the growth of E4 mutant viruses. To determine if the MRN complex or other DNA repair protein components play a role in virus replication and growth, wild type Ad and several E4 mutant viruses were used in single-step growth curves in cells lacking different DNA damage repair proteins (Table 1). The growth of wild-type Ad5 (dl309) was compared to dl355 (E4 ORF6–), dl355/inORF3 (E4 ORF6–/ORF3–) and dl355/DL (E4 ORF6–/E4 ORF3 mutant D105L106). IMR90 normal human fibroblasts were used as a positive control for cells with active DNA repair pathways. Efficient viral growth was observed in IMR90 cells with wild type Ad5 (dl309). Loss of E4 ORF6 (dl355) reduced virus growth 3 to 5-fold (similar to the reduction in growth observed comparing these viruses in A549 cells). Loss of functional E4 ORF3 in addition to E4 ORF6 significantly reduced virus growth in IMR90 cells (average decrease 50-fold) similar to that observed in A549 cells (Fig. 1). Growth of these mutant viruses was restored in human cells hypomorphic for Mre11 (ATLD1) or Nbs1 (GM07166) where the E4 ORF3/ORF6 double mutants grew nearly as well as the parent virus dl355 (2-fold average reduction). Thus, lack of functional Mre11 or Nbs1 rescued the growth defect observed with E4 mutant viruses. The cells that are defective for Nbs1 and Mre11 used in our studies are primary human fibroblasts derived from diseased patients that are at relatively early passage after isolation (10 to 15 passages from isolation). Since these are not clonal cell line derivatives, the chances that the population of cells contains additional mutations is low. Rescue of E4 mutant virus growth, however, was not observed in cells lacking other components of the cellular DNA repair pathways including DNA PK catalytic subunit (M059J), DNA ligase IV (180BRM) or Bloom helicase (BLM) (Table 1). To extend these studies, viral DNA replication and genome concatenation was analyzed in Nbs1– cells infected with the different viruses (Fig. 6). As expected from the growth properties of the wild type and mutant viruses in these cells, all the mutant viruses replicated viral DNA as well as the wild type (Fig. 6A) and minimal and comparable levels of viral genome concatenation were observed (Fig. 6B). We conclude from these analyses that the loss of two components of the MRN complex (Mre11 and Nbs1) rescues the replication and growth properties of E4 mutant viruses. These results argue that a specific aspect of the cellular DNA damage response to Ad infection is the target of E4 function and not simply that the DNA damage response is globally modified by viral infection.
DISCUSSION
PML nuclear domains are a common target for regulation by many eukaryotic DNA viruses (11, 32). Early after infection with papovaviruses (e.g., simian virus 40 and human papillomavirus), Ads, and herpesviruses (e.g., herpes simplex virus type 1 and cytomegalovirus), viral genomes are found adjacent to PML bodies (19). Whether this is a cellular response to incoming viral genomes or an active viral mechanism to direct viral DNAs to a specific subnuclear location is not known, but PML reorganization appears to be a prerequisite for viral DNA replication to commence. The diversity of functions associated with proteins that localize to PML nuclear bodies (2, 14, 22) has made it difficult to attribute PML reorganization to any specific cellular or viral functions. However, the role of PML bodies in the cellular DNA damage response (1, 5, 21, 34) and the specific targeting of DNA damage components by Ad E4 products (3, 27, 29) strongly indicates that this PML-associated activity is a major target of E4 function. It is well established that the E4 ORF3 and E4 ORF6 proteins have compensatory functions in viral infection and are important for efficient viral DNA replication, late gene expression, and virus production (28). Our results and those reported by Stracker and colleagues (27) show that the cellular MRN complex is an important target for E4 ORF3 and E4 ORF6 function. We find that E4 ORF3 directs the relocalization of MRN components into nuclear track-like structures that resemble rearranged PML structures at early times after Ad infection (Fig. 3 and 4). We speculate that such reorganization of MRN precludes the recognition of the ends of the Ad genome by MRN and, thus, blocks the function of MRN as a sensor of DNA damage. Stracker et al. (27) demonstrated that the E4 ORF6-E1B 55-kDa protein complex targets Mre11 for ubiquitin-mediated proteasome-dependent degradation, thus interfering with MRN function in a different yet redundant manner with E4 ORF3.
The analysis of E4 ORF3 mutant D105L106 demonstrates that PML reorganization may be uncoupled from the ability of the protein to reorganize MRN components since this mutant effectively reorganized PML but had no effect on MRN distribution (Fig. 5). No E4 ORF3 mutant has been identified that has the converse activity, i.e., MRN reorganization with no effect on PML localization. Since this mutant is completely defective for E4 ORF3 activity, as are mutants N82 and L103 (Fig. 1 and 2), we conclude that MRN relocalization is required for efficient viral DNA replication. This does not preclude a role for PML rearrangement in the viral replication process. In fact, the ability of evolutionarily diverse Ads to modify PML localization supports the likely role that the reorganization of PML nuclear bodies plays in promoting virus infection (10). However, we believe that MRN relocalization represents a very early event that must take place for viral DNA replication to ensue. There is compelling evidence that the MRN complex acts as a sensor to recognize damaged DNA (18, 23, 26), such as the ends of the linear, double-stranded Ad genome may appear within an infected cell. Mre11 is a nuclease with both endonucleolytic and exonucleolytic activities (7). The termini of the Ad genome may be processed by the nucleolytic activities of the MRN complex which may cause the genome to suffer two adverse events. First, the viral terminal protein is covalently linked to the ends of the genome and acts as a primer to initiate each round of DNA replication. Removal of the terminal protein by MRN activity would effectively block the very first event in the replication process. Second, further nucleolytic activity of MRN on the exposed ends of the genome may irreversibly inactivate the replication origins by removal of critical sequences necessary for recognition by the terminal protein-DNA polymerase complex. The combination of these two processes may eliminate functional origins of replication from the input Ad genomes before replication can even begin and, therefore, would represent a very early and essential target of E4 protein inactivation.
It is interesting that the Nbs1 protein localizes to viral replication domains in cells infected with both wild type Ad5 and a mutant that lacks E4 ORF6. A very similar pattern of colocalization of Nbs1 with viral DNA replication centers was recently described in herpes simplex virus type 1-infected cells (33). With HSV-1, other cellular repair and recombination proteins are found in viral replication domains which may reflect their roles in viral DNA replication, genome concatenation and/or intergenomic recombination (33). Similar models apply with Ad-infected cells. The more important role of Nbs1 in homologous recombination compared to nonhomologous end joining has been revealed in several different systems (16) suggesting that the localization of Nbs1 within viral replication domains in Ad- and herpes simplex virus type 1-infected cells may be indicative of the active recombination processes that take place in the nuclei of cells infected with these viruses. Our results differ from those reported by Stracker et al. (27) with respect to Nbs1 localization in wild-type Ad-infected cells. We found significant colocalization of Nbs1 both within and surrounding viral replication centers, whereas Stracker et al. (27) only observed Nbs1 localized neighboring, but outside, of viral replication compartments. This may reflect different cell lines used for these studies: A549 cells that were used in our experiments contain an intact p53 pathway, whereas HeLa cells used in the studies by Stracker et al. (27) do not. Several lines of evidence support the conclusion that the MRN complex stimulates the protein kinase activity of ATM which in turn phosphorylates a number of substrates including Nbs1 (24). This may promote a positive feedback loop to enhance different aspects of the DNA damage response including MRN activity and cell cycle checkpoint controls, and may involve p53 activity. Consistent with this possibility, several reports demonstrated colocalization of p53 with MRN proteins following DNA damage as well as a dependence on p53 for PML redistribution following DNA damage.
We previously reported that a number of viruses with E4 ORF3 mutations were capable of efficient DNA replication and virus production yet were unable to inhibit concatenation of viral genomes (10). Thus, the function of E4 ORF3 in Ad DNA replication is genetically separable from its ability to inhibit viral genome concatenation. Significant concatenation of viral genomes only is found at very late stages after infection, generally after the bulk of viral DNA replication has taken place, and not all genomes are found concatenated in these assays (29). We suggest that the inhibition of genomic concatenation in cell culture systems does not reflect the true nature of E4 function, but rather may represent a dead-end pathway that occurs late in the viral life cycle using residual genomes that have not been packaged into virions. Perhaps Ad E4 function is designed to forestall the cellular DNA damage response long enough to allow sufficient genome replication and encapsidation to further the infection process.
Our data suggest a model of infection that requires the control and sequestration of the MRN complex for efficient virus growth. At early times after infection, wild-type E4 ORF3 reorganizes MRN components in a manner similar to that of PML, while several mutants that are defective for DNA replication cannot (Fig. 4 to 6). The inability of these mutants to reorganize MRN proteins correlates with decreased DNA replication and growth capabilities (Fig. 1 and 2). Further, cell lines lacking functional MRN components complement the growth defects of E4 mutant viruses (Table 1). These results further strengthen the argument that the MRN complex represents a primary target of E4 ORF3 and E4 ORF6 functions. The mechanism by which E4 ORF3 relocalizes the MRN complex or PML nuclear domains remains unknown. E4 ORF3 has not been seen to interact directly with PML or any of the MRN proteins in in vitro assays (unpublished results). Perhaps an interaction with another protein enables E4 ORF3 to manipulate the distribution of these activities. The results presented here highlight the complex interplay between viruses and cells during infection. More importantly, these results demonstrate for the first time that the sequestration of the DNA damage repair proteins Mre11, Nbs1, and Rad50 by Ad E4 ORF3 is necessary for efficient virus replication to occur in vivo.
ACKNOWLEDGMENTS
We thank our colleagues for many helpful discussions and Jihong Yang for excellent technical assistance. We are very grateful to colleagues who provided antibodies and cells used in these studies. We thank Thomas Dobner, Arnold Levine, Norma Neff, John Petrini, Peter van der Vleit, and Huichen Wang for the generous gifts of reagents.
This research was supported by Public Health Service grant CA28146 from the National Institutes of Health to P.H. J.D.E. was supported by NIH training grant CA09176.
REFERENCES
Bischof, O., S. Kim, J. Irving, S. Beresten, N. Ellis, and J. Campisi. 2001. Regulation and localization of the Bloom syndrome protein in response to DNA damage. J. Cell Biol. 153:367-380.
Borden, K. L. 2002. Pondering the promyelocytic leukemia protein (PML) puzzle: possible functions for PML nuclear bodies. Mol. Cell. Biol. 22:5259-5269.
Boyer, J., K. Rohleder, and G. Ketner. 1999. Adenovirus E4 34k and E4 11k inhibit double strand break repair and are physically associated with the cellular DNA-dependent protein kinase. Virology 263:307-312.
Bridge, E., and G. Ketner. 1989. Redundant control of adenovirus late gene expression by early region 4. J. Virol. 63:631-638.
Carbone, R., M. Pearson, S. Minucci, and P. Pelicci. 2002. PML NBs associate with the hMre11 complex and p53 at sites of irradiation induced DNA damage. Oncogene 21:1633-1640.
Carvalho, T., J. S. Seeler, K. Ohman, P. Jordan, U. Pettersson, G. Akusjarvi, M. Carmo-Fonseca, and A. Dejean. 1995. Targeting of adenovirus E1A and E4-ORF3 proteins to nuclear matrix-associated PML bodies. J. Cell Biol. 131:45-56.
D'Amours, D., and S. P. Jackson. 2002. The Mre11 complex: at the crossroads of DNA repair and checkpoint signalling. Nat. Rev. Mol. Cell. Biol. 3:317-327.
Doucas, V., A. M. Ishov, A. Romo, H. Juguilon, M. D. Weitzman, R. M. Evans, and G. G. Maul. 1996. Adenovirus replication is coupled with the dynamic properties of the PML nuclear structure. Genes Dev. 10:196-207.
Evans, J. D., and P. Hearing. 2002. Adenovirus replication, p. 39-70. In D. T. Curiel and J. T. Douglass (ed.), Adenoviral vectors for gene therapy. Elsevier Science Publishing, Inc., New York, N.Y.
Evans, J. D., and P. Hearing. 2003. Distinct roles of the adenovirus E4 ORF3 protein in viral DNA replication and inhibition of genome concatenation. J. Virol. 77:5295-5304.
Everett, R. D. 2001. DNA viruses and viral proteins that interact with PML nuclear bodies. Oncogene 20:7266-7273.
Halbert, D. N., J. R. Cutt, and T. Shenk. 1985. Adenovirus early region 4 encodes functions required for efficient DNA replication, late gene expression, and host cell shutoff. J. Virol. 56:250-257.
Harada, J. N., A. Shevchenko, A. Shevchenko, D. C. Pallas, and A. J. Berk. 2002. Analysis of the adenovirus E1B-55K-anchored proteome reveals its link to ubiquitination machinery. J. Virol. 76:9194-9206.
Hofmann, T., and H. Will. 2003. Body language: the function of PML nuclear bodies in apoptosis regulation. Cell Death Differ. 10:1290-1299.
Huang, M. M., and P. Hearing. 1989. Adenovirus early region 4 encodes two gene products with redundant effects in lytic infection. J. Virol. 63:2605-2615.
Iijima, K., K. Komatsu, S. Matsuura, and H. Tauchi. 2004. The Nijmegen breakage syndrome gene and its role in genome stability. Chromosoma 113:53-61.
Ishov, A. M., A. G. Sotnikov, D. Negorev, O. V. Vladimirova, N. Neff, T. Kamitani, E. T. Yeh, J. F. Strauss III, and G. G. Maul. 1999. PML is critical for ND10 formation and recruits the PML-interacting protein daxx to this nuclear structure when modified by SUMO-1. J. Cell Biol. 147:221-234.
Lisby, M., J. H. Barlow, R. C. Burgess, and R. Rothstein. 2004. Choreography of the DNA damage response: spatiotemporal relationships among checkpoint and repair proteins. Cell 118:699-713.
Maul, G. G. 1998. Nuclear domain 10, the site of DNA virus transcription and replication. Bioessays 20:660-667.
Melnick, A., and J. D. Licht. 1999. Deconstructing a disease: RARalpha, its fusion partners, and their roles in the pathogenesis of acute promyelocytic leukemia. Blood 93:3167-3215.
Mirzoeva, O., and J. Petrini. 2001. DNA damage-dependent nuclear dynamics of the Mre11 complex. Mol. Cell. Biol. 21:281-288.
Negorev, D., and G. Maul. 2001. Cellular proteins localized at and interacting within ND10/PML nuclear bodies/PODs suggest functions of a nuclear depot. Oncogene 20:7234-7242.
Petrini, J., and T. Stracker. 2003. The cellular response to DNA double-stranded breaks: defining the sensors and mediators. Trends Cell Biol. 13:458-462.
Petrini, J. H. 2000. The Mre11 complex and ATM: collaborating to navigate S phase. Curr. Opin. Cell Biol. 12:293-296.
Querido, E., P. Blanchette, Q. Yan, T. Kamura, M. Morrison, D. Boivin, W. G. Kaelin, R. C. Conaway, J. W. Conaway, and P. E. Branton. 2001. Degradation of p53 by adenovirus E4orf6 and E1B55K proteins occurs via a novel mechanism involving a Cullin-containing complex. Genes Dev. 15:3104-3117.
Stracker, T., J. Theunissen, M. Morales, and J. Petrini. 2004. The Mre11 complex and the metabolism of chromosome breaks: the importance of communicating and holding things together. DNA Repair 3:845-854.
Stracker, T. H., C. T. Carson, and M. D. Weitzman. 2002. Adenovirus oncoproteins inactivate the Mre11-Rad50-NBS1 DNA repair complex. Nature 418:348-352.
Tauber, B., and T. Dobner. 2001. Adenovirus early E4 genes in viral oncogenesis. Oncogene 20:7847-7854.
Weiden, M. D., and H. S. Ginsberg. 1994. Deletion of the E4 region of the genome produces adenovirus DNA concatemers. Proc. Natl. Acad. Sci. USA 91:153-157.
Weinberg, D. H., and G. Ketner. 1986. Adenoviral early region 4 is required for efficient viral DNA replication and for late gene expression. J. Virol. 57:833-838.
Weinberg, D. H., and G. Ketner. 1983. A cell line that supports the growth of a defective early region 4 deletion mutant of human adenovirus type 2. Proc. Natl. Acad. Sci. USA 80:5383-5386.
Weitzman, M., C. Carson, R. Schwartz, and C. Lilley. 2004. Interactions of viruses with the cellular DNA repair machinery. DNA Repair 3:1165-1173.
Wilkinson, D., and S. Weller. 2004. Recruitment of cellular recombination and repair proteins to sites of herpes simplex virus type 1 DNA replication is dependent on the composition of viral proteins within prereplicative sites and correlates with the induction of the DNA damage response. J. Virol. 78:4783-4796.
Xu, Z., A. Timanova-Atanasova, R. Zhao, and K. Chang. 2003. Promyelocytic leukemia protein colocalizes with and stabilizes the DNA damage response protein TopBP1. Mol. Cell. Biol. 23:4247-4256.(Jared D. Evans and Patric)
ABSTRACT
Adenovirus replication is controlled by the relocalization or modification of nuclear protein complexes, including promyelocytic leukemia protein (PML) nuclear domains and the Mre11-Rad50-Nbs1 (MRN) DNA damage machinery. In this study, we demonstrated that the E4 ORF3 protein effects the relocalization of both PML and MRN proteins to similar structures within the nucleus at early times after infection. These proteins colocalize with E4 ORF3. Through the analysis of specific viral mutants, we found a direct correlation between MRN reorganization at early times after infection and the establishment of viral DNA replication domains. Further, the reorganization of MRN components may be uncoupled from the ability of E4 ORF3 to rearrange PML. At later stages of infection, components of the MRN complex disperse within the nucleus, Nbs1 is found within viral replication centers, Rad50 remains localized with E4 ORF3, and Mre11 is degraded. The importance of viral regulation of the MRN complex is underscored by the complementation of E4 mutant viruses in cells that lack Mre11 or Nbs1 activity. These results illustrate the importance of nuclear organization in virus growth and suggest that E4 ORF3 regulates activities in both PML nuclear bodies and the MRN complex to stimulate the viral replication program.
INTRODUCTION
Adenovirus (Ad) replication involves a number of events that must be temporally and spatially organized within the host cell in order to lead to optimal productive infection. The 36-kbp Ad genome encodes at least 25 early gene products (9). Previous studies showed that early region 4 (E4) encodes two proteins of particular importance to viral replication, the products of open reading frames (ORFs) 3 and 6 (E4 ORF3 and E4 ORF6) (4, 15). In the absence of E4 gene products, early gene expression occurs normally, but a profound defect in viral DNA replication is evident (12, 30). E4 ORF3 and E4 ORF6 have compensatory functions to stimulate efficient viral DNA replication independently of one another (4, 15). E4 ORF3 and E4 ORF6 proteins both bind to the E1B 55-kDa protein and play roles in the regulation of late virus RNA transcription and splicing, inhibition of Ad genome concatenation, and stimulation of viral growth (28).
The E4 ORF3 protein plays an important role in modifying the host nucleus during the beginning stages of viral infection. E4 ORF3 interacts with and reorganizes distinct nuclear structures known as nuclear domain 10, promyelocytic leukemia protein (PML) oncogenic domains, or PML nuclear bodies (6, 8). These domains are of unknown function but have been implicated in a number of cellular processes, including transcriptional regulation, cellular growth control, response to DNA damage, apoptosis, and multiple cellular defense mechanisms (2, 14, 22). PML nuclear domains are highly sensitive to heat, stress, and interferon treatment. PML nuclear bodies are targeted by numerous DNA virus early gene products (21, 32). Deregulation of PML nuclear domains has been linked to a severe hematopoietic disease, acute promyelocytic leukemia (20). This form of leukemia is the result of chromosomal translocation resulting in the production of a fusion protein that consists of PML linked to retinoic acid receptor . PML nuclear domains are aberrantly localized as a result of the expression of the PML-retinoic acid receptor fusion. Treatment of acute promyelocytic leukemia patients with all-trans retinoic acid or arsenic trioxide leads to disease remission and restoration of PML nuclear body integrity, suggesting the need for a particular organization of PML and PML domain resident proteins for normal cell function. It has been shown that PML is the nucleating protein of the nuclear domains (17). Without PML, the other resident proteins are redistributed throughout the nucleus. E4 ORF3 is necessary and sufficient to direct the redistribution of PML nuclear bodies into a track-like structure that localizes to the periphery of Ad DNA replication centers (6, 8).
The Mre11-Rad50-Nbs1 (MRN) complex plays a critical role in the cellular response to double-strand breaks and DNA damage (7, 26). Mutations in the Mre11 and Nbs1 genes have been implicated in the human diseases ataxia-telangiectasia-like disorder and Nijmegen breakage syndrome, respectively. These genetic diseases result in a higher incidence of cancer and growth defects in afflicted individuals (7, 26). Cell lines founded from patients with these diseases have been shown to be sensitive to ionizing radiation. The MRN complex plays critical roles in the detection and repair of DNA damage. The E4 ORF3 and E4 ORF6 proteins interfere with the cellular DNA damage response to Ad infection (3, 27, 29). Without E4 gene expression, the linear double-stranded Ad DNA genome activates the host DNA damage response, and viral genomes are concatenated by cellular repair pathways (27). This event effectively blocks viral DNA replication, since the Ad terminal repeats, which contain the origins of replication, are buried and mutated within genomic concatemers. Ad genome concatenation requires the cellular MRN complex and DNA ligase IV as well as the checkpoint signaling protein kinases DNA protein kinase (PK) and ATM (3, 27). The E4 ORF3 and E4 ORF6 proteins independently interact with the catalytic subunit of DNA PK (3). The E4 ORF6-E1B 55-kDa complex targets p53 and Mre11 for ubiquitin-mediated proteasome-dependent degradation, effectively interfering with the cellular DNA damage response checkpoint to block viral genome concatenation (13, 25, 27).
The interaction of E4 ORF3 with the DNA damage response machinery is intriguing. The expression of E4 ORF3 was shown to effect the relocalization of nuclear MRN components, and E4 ORF3 was found to colocalize with MRN proteins (27), but no direct link between these observations and E4 ORF3 function was known. MRN foci localized to the periphery of viral DNA replication centers (27). This finding led to the speculation that E4 ORF3 may sequester MRN components from viral DNA to prevent cellular repair of the viral genome. However, this idea is at odds with previous work in which we found that several E4 ORF3 mutant viruses showed wild-type growth kinetics yet accumulated a large number of viral genome concatemers (10). One such mutant effectively rearranged MRN components, drawing into question the role of MRN reorganization in the viral life cycle.
The aim of this study was to elucidate the function of E4 ORF3 relocalization of MRN complex proteins during Ad infection. We found that E4 ORF3 reorganizes PML and MRN complex proteins in a similar manner within the nuclei of Ad-infected cells. The relocalization of MRN proteins precisely correlated with the growth properties of mutant viruses. Further, cells harboring mutations in Mre11 and Nbs1 proteins but not several other effectors in the cellular DNA damage response were able to complement the growth of viruses that do not express functional E4 ORF3 and E4 ORF6 proteins. These findings emphasize the importance of the MRN complex as a target of Ad E4 function and suggest that a very early step in MRN function plays the most important role in regulating the Ad replication cycle.
MATERIALS AND METHODS
Viruses, cells, and growth assays. Ad mutant viruses dl355 (E4 ORF6 negative [E4 ORF6–]), inORF3 (E4 ORF3 negative [E4 ORF3–]), dl355/inORF3 (E4 ORF3 negative and E4 ORF6 negative [E4 ORF3/6–]), and dl355/pmD105L106 (E4 ORF6–/ORF3 D105A/L106A point mutations) were described previously (10, 12, 15). Mutants dl355/pmN82 (27) and dl355/pmL103 contain alanine substitution mutations at E4 ORF3 amino acids 82 and 103, respectively, in a dl355 virus background and were constructed as previously described (10). A549 and W162 cells were maintained in Dulbecco's modified minimal essential medium containing 10% calf serum. ATLD1 (Mre11 negative [Mre11–]; gift from John Petrini, Memorial Sloan Kettering Institute), GM07166 (Nbs1 negative [Nbs1–]; Coriel Institute for Medical Research), M059J (DNA PK negative [DNA PK–]; American Type Culture Collection), M059K (DNA PK positive [DNA PK+]; American Type Culture Collection), BLM (Bloom helicase negative [BLM–]; gift from Norma Neff, New York Blood Center), 180BRM (DNA ligase IV negative [DNA ligase IV–]; gift from Huichen Wang, Jefferson Medical College), and IMR90 (American Type Culture Collection) were maintained in Dulbecco's modified minimal essential medium supplemented with 15% fetal bovine serum at 37°C in 5.0% CO2.
All viral infections were carried out in an identical manner. Purified virus particles were prepared by CsCl equilibrium gradient centrifugation. Cells were infected with 200 virus particles/cell for 1 h at 37°C, the virus inoculum was removed, the cells were washed twice, and fresh medium was added. The infection mixtures were incubated for various times at 37°C in 5.0% CO2. Single-step growth curves were used to measure virus growth in A549 cells and cells with mutant DNA repair proteins. For A549 cells, cells were harvested at 4, 20, 32, and 44 h postinfection. For cells with mutant DNA repair proteins, cells were harvested at 4, 72, and 120 h postinfection. Cellular lysates were prepared by four cycles of freezing/thawing, and infectious virus yields were determined by plaque assays with E4-complementing cell line W162 (31).
Immunofluorescence analysis. A549 cells were placed on coverslips and infected as described above. At 6, 9, 12, and 15 h postinfection, the cells were fixed with –20°C methanol for 5 min. The cells were washed with phosphate-buffered saline (PBS) and blocked in PBS containing 10% goat serum for 1 h. Primary antibodies were diluted in PBS containing 1.5% goat serum and incubated on coverslips for 1 h at room temperature. Antibodies used were anti-PML mouse monoclonal antibody (PG-M3; Santa Cruz Biotechnology), anti-E4 ORF3 rat monoclonal antibody (6A-11; gift from Thomas Dobner, Regensberg Universitat), anti-DNA binding protein (DBP) monoclonal antibody (gift from Arnold Levine, Princeton University), anti-DBP polyclonal antibody (gift from Peter van der Vleit, University Medical Centre Utrecht), anti-Mre11 rabbit polyclonal antibody (100-142; Novus Biologicals), anti-Rad50 mouse monoclonal antibody (100-147; Novus Biologicals), and anti-Nbs1 rabbit polyclonal antibody (100-143; Novus Biologicals). The coverslips were washed three times with PBS; incubated with fluoroscein isothiocyanate (FITC)-conjugated goat anti-mouse immunoglobulin G (IgG) (Zymed), Texas Red isothiocyanate (TRITC)-conjugated goat anti-rat IgG (Zymed), and Alexa 350-labeled goat anti-rabbit IgG (Molecular Probes) for 45 min at room temperature in the dark; and washed again three times with PBS. The coverslips were mounted on slides in Fluoromount G (Southern Biotechnology Associates). Immunofluorescence was viewed and photographed by using a Zeiss microscope with Axiovision software.
Viral DNA replication, pulsed-field gel electrophoresis (PFGE), and Western blot analysis. Viral DNA replication was determined by Southern blot analysis. Total nuclear DNA was isolated at 6, 12, 18, and 24 h after infection as previously described (10). The samples were diluted 1:10, 1:30, 1:100, and 1:200 in Tris-EDTA, respectively, and the DNA was denatured and prepared for hybridization as previously described. The DNA was bound to a Hybond-N membrane (Amersham) by the slot blot method followed by UV cross-linking. The probe was an Ad5 (positions 1 to 194) PCR fragment fluorescently labeled with the Atto-Phos system (Promega). The probe was hybridized to the membrane and detected according to the manufacturer's instructions. The signal was visualized and quantitated by using a Molecular Dynamics Storm 800 PhosphorImager and ImageQuant software.
For PFGE, infected Nbs1– cells were harvested at 72 h postinfection and prepared as previously described (10). Briefly, cells were harvested by trypsinization, washed once with PBS, and resuspended in 100 μl of PBS with 125 mM EDTA. Next, 100 μl of 1.25% low-melting-temperature agarose in 50 mM Tris (pH 7.4) and 125 mM EDTA was added to the cells. The solidified plugs were treated with 1.2% sodium dodecyl sulfate (SDS), 0.125 mM EDTA, and 1 mg/ml proteinase K overnight at 50°C. The plugs were washed three times over 8 h with 50 mM EDTA. A 1.0% agarose gel was loaded with 30 μl of melted plug. The gel was run at 6 V/cm with a switching time of 6 s for 20 h. The DNA was transferred to a Hybond-N nylon membrane (Amersham), probed with a 32P-labeled Ad total genome probe, and visualized by autoradiography as previously described (10).
For Western blot analysis of wild-type and mutant E4 ORF3 proteins, A549 cells were infected with E1 replacement viruses expressing hemagglutinin (HA)-tagged E4 ORF3 wild-type and mutant proteins for 12 h as previously described (10). Infected cells were harvested and lysed in radioimmunoprecipitation buffer (150 mM NaCl, 1.0% NP-40, 0.5% deoxycholic acid, 0.1% SDS, 50 mM Tris [pH 8.0]) supplemented with a protease inhibitor cocktail.
Proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to a Hybond-P membrane (Amersham) by standard methods. The membrane was probed with anti-HA mouse monoclonal antibody (Roche), and proteins were visualized with enhanced chemiluminescence reagents (Amersham).
RESULTS
Analysis of E4 ORF3 mutant viruses. A group of recombinant viruses was constructed and analyzed that expressed wild type and mutant E4 ORF3 proteins from the natural promoter in the absence of E4 ORF6 (dl355). Under these conditions, efficient viral DNA replication and growth are dependent on E4 ORF3 function (4, 10, 15). Single and double alanine substitution mutations were introduced that spanned E4 ORF3. Viral growth properties were analyzed in a single-step growth curve (Fig. 1) and by analyzing viral DNA replication (Fig. 2). In total, sixteen E4 ORF3 mutants were analyzed. Only three mutant viruses (dl355/pmN82 [27], dl355/pmL103, and dl355/pmD105L106) were severely defective for virus growth; pmD105L106 was previously reported (10) and dl355/pmN82 and dl355/pmL103 are newly described viral mutants. In single-step growth curves in noncomplementing A549 cells, these viruses replicated to levels equally reduced compared to a virus that lacks both E4 ORF3 and E4 ORF6 proteins (dl355/inORF3) (Fig. 1). These E4 ORF3 mutant viruses were decreased 100-fold for growth when compared to the parent virus dl355. This reduction was reflected in comparably reduced viral DNA replication levels (Fig. 2 and data not shown). Comparable levels of wild type and mutant proteins were detected by Western blot analysis (Fig. 2A, wild type versus N82 and L103; for D105/L106, see reference 10). These viruses provide a number of mutants with which to study E4 ORF3 function in the context of natural infection.
E4 ORF3 relocalizes MRN complex components. Recently, the DNA damage protein MRN complex was shown to localize to foci correlating to areas of Adenovirus replication in the absence of E4 proteins during virus infection (27). Also, E4 ORF3 was able to reorganize the MRN proteins to track-like structures that partially overlapped with rearranged PML in a transient transfection assay (27), but the functional consequences of this rearrangement was not known. Here we show that E4 ORF3 reorganization of MRN is critical for virus replication.
Immunofluorescence was used to examine Mre11, Rad50, Nbs1, and PML localization following infection with wild type Ad5 (Fig. 3). In uninfected cells, MRN components were diffusely localized throughout the nucleus while PML was localized to discreet nuclear bodies (Fig. 3A, E, I, and M). At 6 h postinfection, wild-type E4 ORF3 caused Mre11, Rad50, and Nbs1 to relocalize to track-like structures throughout the nuclei that showed almost complete colocalization with E4 ORF3 (Fig. 3B to D, F to H, and J to L). Rearranged MRN proteins formed structures that were similar to those observed with PML and E4 ORF3 (Fig. 3N to P). Such a rearrangement of the MRN proteins was found whether or not E4 ORF6 protein was expressed (wild type dl309, E4 ORF6+ [Fig. 3]; dl355, E4 ORF6– [data not shown]). MRN relocalization was completely dependent on E4 ORF3 since no rearrangement was observed with any viral mutant that did not express this protein (data not shown).
As the infection progressed, the pattern of MRN reorganization continued to change (Fig. 4). When E4 ORF6 was expressed, Mre11 levels declined (Fig. 4A), as previously reported (27). Rad50 was maintained in association with E4 ORF3 in track-like structures that surrounded, but did not overlap with, viral DNA replication centers (Fig. 4E to H); viral DNA binding protein (DBP) staining was used to identify replication centers. Nbs1 localization significantly changed as the infection progressed and almost complete colocalization of Nbs1 with viral replication centers was evident by 15 h postinfection (Fig. 4J to L). In the absence of E4 ORF6 protein, identical patterns were observed for Rad50 and Nbs1 (data not shown), whereas Mre11 was still evident in association with E4 ORF3 (Fig. 4M to P). We conclude that E4 ORF3 significantly alters the subnuclear localization and organization of MRN complex components. Mre11, Rad50 and Nbs1 are all targeted to E4 ORF3-containing tracks early after infection, whereas the MRN complex is disrupted at later times after infection to disperse the components.
MRN reorganization is required for efficient viral DNA replication. Using mutant viruses that were either replication competent or defective, we analyzed the ability of E4 ORF3 mutants to reorganize MRN components and PML nuclear domains at 6 h postinfection (Fig. 5). Mutants dl355/pmN82 and dl355/pmL103 were defective for growth and viral DNA replication (Fig. 1 and 2) and were unable to reorganize Rad50 (Fig. 5C and E), PML (Fig. 5D and F), or Mre11 and Nbs1 (data not shown). The E4 ORF3 proteins produced by these viruses were diffusely localized throughout the cell (most obvious in Fig. 5D and F). Mutant dl355/pmD105L106 also was defective for replication and was unable to reorganize Rad50 (Fig. 5G) or Mre11 and Nbs1 (data not shown). The D105L106 mutant protein, however, was able to form tracks itself (red fluorescence in Fig. 5G) and efficiently reorganized PML early after infection (Fig. 5H). All three of these mutant viruses were unable to establish DBP-containing replication centers consistent with their replication defect, but produced DBP to levels equivalent as wild type virus (data not shown). Identical results were found at 15 h postinfection. Viruses expressing wild-type and mutant E4 ORF3 proteins that grew efficiently directed Rad50 and PML relocalization (Fig. 5A, B, and I to L) as well as Mre11 and Nbs1 (data not shown). The entire panel of E4 ORF3 mutant viruses (sixteen mutants) was examined for their abilities to reorganize Rad50 and PML. In every case, a direct correlation was found between the ability of the mutant proteins to reorganize Rad50 early after infection with the abilities of the mutant viruses to replicate in infected cells (data not shown). Thus, we conclude that reorganization of the MRN complex early after Ad infection by the E4 ORF3 protein is required for efficient viral DNA replication and to establish viral replication centers in the absence of E4 ORF6 expression.
Loss of Mre11 and Nbs1 complements the growth of E4 mutant viruses. To determine if the MRN complex or other DNA repair protein components play a role in virus replication and growth, wild type Ad and several E4 mutant viruses were used in single-step growth curves in cells lacking different DNA damage repair proteins (Table 1). The growth of wild-type Ad5 (dl309) was compared to dl355 (E4 ORF6–), dl355/inORF3 (E4 ORF6–/ORF3–) and dl355/DL (E4 ORF6–/E4 ORF3 mutant D105L106). IMR90 normal human fibroblasts were used as a positive control for cells with active DNA repair pathways. Efficient viral growth was observed in IMR90 cells with wild type Ad5 (dl309). Loss of E4 ORF6 (dl355) reduced virus growth 3 to 5-fold (similar to the reduction in growth observed comparing these viruses in A549 cells). Loss of functional E4 ORF3 in addition to E4 ORF6 significantly reduced virus growth in IMR90 cells (average decrease 50-fold) similar to that observed in A549 cells (Fig. 1). Growth of these mutant viruses was restored in human cells hypomorphic for Mre11 (ATLD1) or Nbs1 (GM07166) where the E4 ORF3/ORF6 double mutants grew nearly as well as the parent virus dl355 (2-fold average reduction). Thus, lack of functional Mre11 or Nbs1 rescued the growth defect observed with E4 mutant viruses. The cells that are defective for Nbs1 and Mre11 used in our studies are primary human fibroblasts derived from diseased patients that are at relatively early passage after isolation (10 to 15 passages from isolation). Since these are not clonal cell line derivatives, the chances that the population of cells contains additional mutations is low. Rescue of E4 mutant virus growth, however, was not observed in cells lacking other components of the cellular DNA repair pathways including DNA PK catalytic subunit (M059J), DNA ligase IV (180BRM) or Bloom helicase (BLM) (Table 1). To extend these studies, viral DNA replication and genome concatenation was analyzed in Nbs1– cells infected with the different viruses (Fig. 6). As expected from the growth properties of the wild type and mutant viruses in these cells, all the mutant viruses replicated viral DNA as well as the wild type (Fig. 6A) and minimal and comparable levels of viral genome concatenation were observed (Fig. 6B). We conclude from these analyses that the loss of two components of the MRN complex (Mre11 and Nbs1) rescues the replication and growth properties of E4 mutant viruses. These results argue that a specific aspect of the cellular DNA damage response to Ad infection is the target of E4 function and not simply that the DNA damage response is globally modified by viral infection.
DISCUSSION
PML nuclear domains are a common target for regulation by many eukaryotic DNA viruses (11, 32). Early after infection with papovaviruses (e.g., simian virus 40 and human papillomavirus), Ads, and herpesviruses (e.g., herpes simplex virus type 1 and cytomegalovirus), viral genomes are found adjacent to PML bodies (19). Whether this is a cellular response to incoming viral genomes or an active viral mechanism to direct viral DNAs to a specific subnuclear location is not known, but PML reorganization appears to be a prerequisite for viral DNA replication to commence. The diversity of functions associated with proteins that localize to PML nuclear bodies (2, 14, 22) has made it difficult to attribute PML reorganization to any specific cellular or viral functions. However, the role of PML bodies in the cellular DNA damage response (1, 5, 21, 34) and the specific targeting of DNA damage components by Ad E4 products (3, 27, 29) strongly indicates that this PML-associated activity is a major target of E4 function. It is well established that the E4 ORF3 and E4 ORF6 proteins have compensatory functions in viral infection and are important for efficient viral DNA replication, late gene expression, and virus production (28). Our results and those reported by Stracker and colleagues (27) show that the cellular MRN complex is an important target for E4 ORF3 and E4 ORF6 function. We find that E4 ORF3 directs the relocalization of MRN components into nuclear track-like structures that resemble rearranged PML structures at early times after Ad infection (Fig. 3 and 4). We speculate that such reorganization of MRN precludes the recognition of the ends of the Ad genome by MRN and, thus, blocks the function of MRN as a sensor of DNA damage. Stracker et al. (27) demonstrated that the E4 ORF6-E1B 55-kDa protein complex targets Mre11 for ubiquitin-mediated proteasome-dependent degradation, thus interfering with MRN function in a different yet redundant manner with E4 ORF3.
The analysis of E4 ORF3 mutant D105L106 demonstrates that PML reorganization may be uncoupled from the ability of the protein to reorganize MRN components since this mutant effectively reorganized PML but had no effect on MRN distribution (Fig. 5). No E4 ORF3 mutant has been identified that has the converse activity, i.e., MRN reorganization with no effect on PML localization. Since this mutant is completely defective for E4 ORF3 activity, as are mutants N82 and L103 (Fig. 1 and 2), we conclude that MRN relocalization is required for efficient viral DNA replication. This does not preclude a role for PML rearrangement in the viral replication process. In fact, the ability of evolutionarily diverse Ads to modify PML localization supports the likely role that the reorganization of PML nuclear bodies plays in promoting virus infection (10). However, we believe that MRN relocalization represents a very early event that must take place for viral DNA replication to ensue. There is compelling evidence that the MRN complex acts as a sensor to recognize damaged DNA (18, 23, 26), such as the ends of the linear, double-stranded Ad genome may appear within an infected cell. Mre11 is a nuclease with both endonucleolytic and exonucleolytic activities (7). The termini of the Ad genome may be processed by the nucleolytic activities of the MRN complex which may cause the genome to suffer two adverse events. First, the viral terminal protein is covalently linked to the ends of the genome and acts as a primer to initiate each round of DNA replication. Removal of the terminal protein by MRN activity would effectively block the very first event in the replication process. Second, further nucleolytic activity of MRN on the exposed ends of the genome may irreversibly inactivate the replication origins by removal of critical sequences necessary for recognition by the terminal protein-DNA polymerase complex. The combination of these two processes may eliminate functional origins of replication from the input Ad genomes before replication can even begin and, therefore, would represent a very early and essential target of E4 protein inactivation.
It is interesting that the Nbs1 protein localizes to viral replication domains in cells infected with both wild type Ad5 and a mutant that lacks E4 ORF6. A very similar pattern of colocalization of Nbs1 with viral DNA replication centers was recently described in herpes simplex virus type 1-infected cells (33). With HSV-1, other cellular repair and recombination proteins are found in viral replication domains which may reflect their roles in viral DNA replication, genome concatenation and/or intergenomic recombination (33). Similar models apply with Ad-infected cells. The more important role of Nbs1 in homologous recombination compared to nonhomologous end joining has been revealed in several different systems (16) suggesting that the localization of Nbs1 within viral replication domains in Ad- and herpes simplex virus type 1-infected cells may be indicative of the active recombination processes that take place in the nuclei of cells infected with these viruses. Our results differ from those reported by Stracker et al. (27) with respect to Nbs1 localization in wild-type Ad-infected cells. We found significant colocalization of Nbs1 both within and surrounding viral replication centers, whereas Stracker et al. (27) only observed Nbs1 localized neighboring, but outside, of viral replication compartments. This may reflect different cell lines used for these studies: A549 cells that were used in our experiments contain an intact p53 pathway, whereas HeLa cells used in the studies by Stracker et al. (27) do not. Several lines of evidence support the conclusion that the MRN complex stimulates the protein kinase activity of ATM which in turn phosphorylates a number of substrates including Nbs1 (24). This may promote a positive feedback loop to enhance different aspects of the DNA damage response including MRN activity and cell cycle checkpoint controls, and may involve p53 activity. Consistent with this possibility, several reports demonstrated colocalization of p53 with MRN proteins following DNA damage as well as a dependence on p53 for PML redistribution following DNA damage.
We previously reported that a number of viruses with E4 ORF3 mutations were capable of efficient DNA replication and virus production yet were unable to inhibit concatenation of viral genomes (10). Thus, the function of E4 ORF3 in Ad DNA replication is genetically separable from its ability to inhibit viral genome concatenation. Significant concatenation of viral genomes only is found at very late stages after infection, generally after the bulk of viral DNA replication has taken place, and not all genomes are found concatenated in these assays (29). We suggest that the inhibition of genomic concatenation in cell culture systems does not reflect the true nature of E4 function, but rather may represent a dead-end pathway that occurs late in the viral life cycle using residual genomes that have not been packaged into virions. Perhaps Ad E4 function is designed to forestall the cellular DNA damage response long enough to allow sufficient genome replication and encapsidation to further the infection process.
Our data suggest a model of infection that requires the control and sequestration of the MRN complex for efficient virus growth. At early times after infection, wild-type E4 ORF3 reorganizes MRN components in a manner similar to that of PML, while several mutants that are defective for DNA replication cannot (Fig. 4 to 6). The inability of these mutants to reorganize MRN proteins correlates with decreased DNA replication and growth capabilities (Fig. 1 and 2). Further, cell lines lacking functional MRN components complement the growth defects of E4 mutant viruses (Table 1). These results further strengthen the argument that the MRN complex represents a primary target of E4 ORF3 and E4 ORF6 functions. The mechanism by which E4 ORF3 relocalizes the MRN complex or PML nuclear domains remains unknown. E4 ORF3 has not been seen to interact directly with PML or any of the MRN proteins in in vitro assays (unpublished results). Perhaps an interaction with another protein enables E4 ORF3 to manipulate the distribution of these activities. The results presented here highlight the complex interplay between viruses and cells during infection. More importantly, these results demonstrate for the first time that the sequestration of the DNA damage repair proteins Mre11, Nbs1, and Rad50 by Ad E4 ORF3 is necessary for efficient virus replication to occur in vivo.
ACKNOWLEDGMENTS
We thank our colleagues for many helpful discussions and Jihong Yang for excellent technical assistance. We are very grateful to colleagues who provided antibodies and cells used in these studies. We thank Thomas Dobner, Arnold Levine, Norma Neff, John Petrini, Peter van der Vleit, and Huichen Wang for the generous gifts of reagents.
This research was supported by Public Health Service grant CA28146 from the National Institutes of Health to P.H. J.D.E. was supported by NIH training grant CA09176.
REFERENCES
Bischof, O., S. Kim, J. Irving, S. Beresten, N. Ellis, and J. Campisi. 2001. Regulation and localization of the Bloom syndrome protein in response to DNA damage. J. Cell Biol. 153:367-380.
Borden, K. L. 2002. Pondering the promyelocytic leukemia protein (PML) puzzle: possible functions for PML nuclear bodies. Mol. Cell. Biol. 22:5259-5269.
Boyer, J., K. Rohleder, and G. Ketner. 1999. Adenovirus E4 34k and E4 11k inhibit double strand break repair and are physically associated with the cellular DNA-dependent protein kinase. Virology 263:307-312.
Bridge, E., and G. Ketner. 1989. Redundant control of adenovirus late gene expression by early region 4. J. Virol. 63:631-638.
Carbone, R., M. Pearson, S. Minucci, and P. Pelicci. 2002. PML NBs associate with the hMre11 complex and p53 at sites of irradiation induced DNA damage. Oncogene 21:1633-1640.
Carvalho, T., J. S. Seeler, K. Ohman, P. Jordan, U. Pettersson, G. Akusjarvi, M. Carmo-Fonseca, and A. Dejean. 1995. Targeting of adenovirus E1A and E4-ORF3 proteins to nuclear matrix-associated PML bodies. J. Cell Biol. 131:45-56.
D'Amours, D., and S. P. Jackson. 2002. The Mre11 complex: at the crossroads of DNA repair and checkpoint signalling. Nat. Rev. Mol. Cell. Biol. 3:317-327.
Doucas, V., A. M. Ishov, A. Romo, H. Juguilon, M. D. Weitzman, R. M. Evans, and G. G. Maul. 1996. Adenovirus replication is coupled with the dynamic properties of the PML nuclear structure. Genes Dev. 10:196-207.
Evans, J. D., and P. Hearing. 2002. Adenovirus replication, p. 39-70. In D. T. Curiel and J. T. Douglass (ed.), Adenoviral vectors for gene therapy. Elsevier Science Publishing, Inc., New York, N.Y.
Evans, J. D., and P. Hearing. 2003. Distinct roles of the adenovirus E4 ORF3 protein in viral DNA replication and inhibition of genome concatenation. J. Virol. 77:5295-5304.
Everett, R. D. 2001. DNA viruses and viral proteins that interact with PML nuclear bodies. Oncogene 20:7266-7273.
Halbert, D. N., J. R. Cutt, and T. Shenk. 1985. Adenovirus early region 4 encodes functions required for efficient DNA replication, late gene expression, and host cell shutoff. J. Virol. 56:250-257.
Harada, J. N., A. Shevchenko, A. Shevchenko, D. C. Pallas, and A. J. Berk. 2002. Analysis of the adenovirus E1B-55K-anchored proteome reveals its link to ubiquitination machinery. J. Virol. 76:9194-9206.
Hofmann, T., and H. Will. 2003. Body language: the function of PML nuclear bodies in apoptosis regulation. Cell Death Differ. 10:1290-1299.
Huang, M. M., and P. Hearing. 1989. Adenovirus early region 4 encodes two gene products with redundant effects in lytic infection. J. Virol. 63:2605-2615.
Iijima, K., K. Komatsu, S. Matsuura, and H. Tauchi. 2004. The Nijmegen breakage syndrome gene and its role in genome stability. Chromosoma 113:53-61.
Ishov, A. M., A. G. Sotnikov, D. Negorev, O. V. Vladimirova, N. Neff, T. Kamitani, E. T. Yeh, J. F. Strauss III, and G. G. Maul. 1999. PML is critical for ND10 formation and recruits the PML-interacting protein daxx to this nuclear structure when modified by SUMO-1. J. Cell Biol. 147:221-234.
Lisby, M., J. H. Barlow, R. C. Burgess, and R. Rothstein. 2004. Choreography of the DNA damage response: spatiotemporal relationships among checkpoint and repair proteins. Cell 118:699-713.
Maul, G. G. 1998. Nuclear domain 10, the site of DNA virus transcription and replication. Bioessays 20:660-667.
Melnick, A., and J. D. Licht. 1999. Deconstructing a disease: RARalpha, its fusion partners, and their roles in the pathogenesis of acute promyelocytic leukemia. Blood 93:3167-3215.
Mirzoeva, O., and J. Petrini. 2001. DNA damage-dependent nuclear dynamics of the Mre11 complex. Mol. Cell. Biol. 21:281-288.
Negorev, D., and G. Maul. 2001. Cellular proteins localized at and interacting within ND10/PML nuclear bodies/PODs suggest functions of a nuclear depot. Oncogene 20:7234-7242.
Petrini, J., and T. Stracker. 2003. The cellular response to DNA double-stranded breaks: defining the sensors and mediators. Trends Cell Biol. 13:458-462.
Petrini, J. H. 2000. The Mre11 complex and ATM: collaborating to navigate S phase. Curr. Opin. Cell Biol. 12:293-296.
Querido, E., P. Blanchette, Q. Yan, T. Kamura, M. Morrison, D. Boivin, W. G. Kaelin, R. C. Conaway, J. W. Conaway, and P. E. Branton. 2001. Degradation of p53 by adenovirus E4orf6 and E1B55K proteins occurs via a novel mechanism involving a Cullin-containing complex. Genes Dev. 15:3104-3117.
Stracker, T., J. Theunissen, M. Morales, and J. Petrini. 2004. The Mre11 complex and the metabolism of chromosome breaks: the importance of communicating and holding things together. DNA Repair 3:845-854.
Stracker, T. H., C. T. Carson, and M. D. Weitzman. 2002. Adenovirus oncoproteins inactivate the Mre11-Rad50-NBS1 DNA repair complex. Nature 418:348-352.
Tauber, B., and T. Dobner. 2001. Adenovirus early E4 genes in viral oncogenesis. Oncogene 20:7847-7854.
Weiden, M. D., and H. S. Ginsberg. 1994. Deletion of the E4 region of the genome produces adenovirus DNA concatemers. Proc. Natl. Acad. Sci. USA 91:153-157.
Weinberg, D. H., and G. Ketner. 1986. Adenoviral early region 4 is required for efficient viral DNA replication and for late gene expression. J. Virol. 57:833-838.
Weinberg, D. H., and G. Ketner. 1983. A cell line that supports the growth of a defective early region 4 deletion mutant of human adenovirus type 2. Proc. Natl. Acad. Sci. USA 80:5383-5386.
Weitzman, M., C. Carson, R. Schwartz, and C. Lilley. 2004. Interactions of viruses with the cellular DNA repair machinery. DNA Repair 3:1165-1173.
Wilkinson, D., and S. Weller. 2004. Recruitment of cellular recombination and repair proteins to sites of herpes simplex virus type 1 DNA replication is dependent on the composition of viral proteins within prereplicative sites and correlates with the induction of the DNA damage response. J. Virol. 78:4783-4796.
Xu, Z., A. Timanova-Atanasova, R. Zhao, and K. Chang. 2003. Promyelocytic leukemia protein colocalizes with and stabilizes the DNA damage response protein TopBP1. Mol. Cell. Biol. 23:4247-4256.(Jared D. Evans and Patric)