Serotype-specific inactivation of the cellular DNA damage response during adenovirus infection

Adenovirus type 5 (Ad5) inactivates the host cell DNA damage response by facilitating the degradation of Mre11, DNA ligase IV, and p53. In the case of p53, this is achieved through polyubiquitylation by Ad5E1B55K and Ad5E4orf6, which recruit a Cul5-based E3 ubiquitin ligase. Recent evidence indicates that this paradigm does not apply to other adenovirus serotypes, since Ad12, but not Ad5, causes the degradation of TOPBP1 through the action of E4orf6 alone and a Cul2-based E3 ubiquitin ligase. We now have extended these studies to adenovirus groups A to E. While infection by Ad4, Ad5, and Ad12 (groups E, C, and A, respectively) cause the degradation of Mre11, DNA ligase IV, and p53, infection with Ad3, Ad7, Ad9, and Ad11 (groups B1, B1, D, and B2, respectively) only affects DNA ligase IV levels. Indeed, Ad3, Ad7, and Ad11 cause the marked accumulation of p53. Despite this, MDM2 levels were very low following infection with all of the viruses examined here, regardless of whether they increase p53 expression. In addition, we found that only Ad12 causes the degradation of TOPBP1, and, like Ad5, Ad4 recruits a Cul5-based E3 ubiquitin ligase to degrade p53. Surprisingly, Mre11 and DNA ligase IV degradation do not appear to be significantly affected in Ad4-, Ad5-, or Ad12-infected cells depleted of Cul2 or Cul5, indicating that E1B55K and E4orf6 recruit multiple ubiquitin ligases to target cellular proteins. Finally, although Mre11 is not degraded by Ad3, Ad7, Ad9, and Ad11, no viral DNA concatemers could be detected. We suggest that group B and D adenoviruses have evolved mechanisms based on the loss of DNA ligase IV and perhaps other unknown molecules to disable the host cell DNA damage response to promote viral replication.     

p53 Status Does Not Determine Outcome of E1B 55-Kilodalton Mutant Adenovirus Lytic Infection

The ability of the adenovirus type 5 E1B 55-kDa mutants dl1520 and dl338 to replicate efficiently and independently of the cell cycle, to synthesis viral DNA, and to lyse infected cells did not correlate with the status of p53 in seven cell lines examined. Rather, cell cycle-independent replication and virus-induced cell killing correlated with permissivity to viral replication. This correlation extended to S-phase HeLa cells, which were more susceptible to virus-induced cell killing by the E1B 55-kDa mutant virus than HeLa cells infected during G₁. Wild-type p53 had only a modest effect on E1B mutant virus yields in H1299 cells expressing a temperature-sensitive p53 allele. The defect in E1B 55-kDa mutant virus replication resulting from reduced temperature was as much as 10-fold greater than the defect due to p53 function. At 39°C, the E1B 55-kDa mutant viruses produced wild-type yields of virus and replicated independently of the cell cycle. In addition, the E1B 55-kDa mutant viruses directed the synthesis of late viral proteins to levels equivalent to the wild-type virus level at 39°C. We have previously shown that the defect in mutant virus replication can also be overcome by infecting HeLa cells during S phase. Taken together, these results indicate that the capacity of the E1B 55-kDa mutant virus to replicate independently of the cell cycle does not correlate with the status of p53 but is determined by yet unidentified mechanisms. The cold-sensitive nature of the defect of the E1B 55-kDa mutant virus in both late gene expression and cell cycle-independent replication leads us to speculate that these functions of the E1B 55-kDa protein may be linked.     

STAT1 Interaction with E3-14.7K in Monocytes Affects the Efficacy of Oncolytic Adenovirus

Oncolytic viruses based on adenovirus type 5 (Ad5) have been developed as a new class of therapeutic agents for cancers that are resistant to conventional therapies. Clinical experience shows that these agents are safe, but virotherapy alone has not achieved long-term cure in cancer patients. The vast majority of oncolytic adenoviruses used in clinical trials to date have deletion of the E3B genes. It has been demonstrated that the antitumor potency of the E3B-deleted mutant (dl309) is inferior to adenovirus with E3B genes intact. Tumors treated with dl309 show markedly greater macrophage infiltration than E3B-intact adenovirus. However, the functional mechanisms for this were not previously known. Here, we demonstrate that deletion of E3B genes increases production of chemokines by monocytes after adenovirus infection and increases monocyte migration. The E3B 14,700-Da protein (E3B-14.7K) inhibits STAT1 function by preventing its phosphorylation and nuclear translocation. The STAT1 inhibitor, fludarabine, rescues the effect of E3B-14.7K deletion by downregulating target chemokine expression in human and murine monocytes and results in an enhanced antitumor efficacy with dl309 in vivo. These findings have important implications for clinical use of E3B-deleted oncolytic adenovirus and other E3B-deleted adenovirus vector-based therapy.     

Downregulation of Mdm2 and Mdm4 enhances viral gene expression during adenovirus infection

Successful viral replication entails elimination or bypass of host antiviral mechanisms. Here, we show that shRNA-mediated knockdown of murine double minute (Mdm2) and its paralog Mdm4 enhanced the expression of early and late viral gene products during adenovirus (HAdV) infection. Remarkably, whereas the expression of HAdV genes was low in p53-deficient mouse embryonic fibroblasts (p53KO MEFs), the HAdV early gene products were efficiently expressed in Mdm2/p53 double-knockout (DKO) and Mdm4/p53 DKO MEFs, and viral capsid proteins were produced in Mdm2/p53 DKO MEFs. Thus, Mdm2 and Mdm4 seem to have potent antiviral property. In cells infected with wt HAdV or a mutant virus lacking the E1B-55K gene (dl1520), both Mdm2 and Mdm4 were rapidly depleted, whereas replication-deficient mutant viruses (Ad-GFP) or ΔpTP with deletions within the coding sequence of preterminal binding protein failed to induce their downregulation. Reduced expression of Mdm2 and Mdm4 was not due to general shutoff of host protein synthesis. Additionally, expression of a dominant-negative mutant of Cul5 did not affect Mdm2/Mdm4 downregulation. Thus, viral replication but not the presence of E1B-55K is required for Mdm2/Mdm4 degradation. Surprisingly, treatment of HAdV-infected cells with proteasome inhibitor MG132 only partially restored the protein levels of Mdm2 and Mdm4, suggesting that they may also be downregulated through an additional mechanism independent of proteasome. Interestingly, cyclin D1 and p21 appear to be downregulated similarly during HAdV infection. Collectively, our work provides the first biochemical evidence for antiviral function of Mdm2 and Mdm4 and that viruses employ efficient countermeasure to ensure viral replication.Key words: adenovirus (HAdV), antiviral mechanism, virus-host interaction, Mdm2, Mdm4, mouse embryonic fibroblast (MEF), DNA-damage response, cell cycle, p21, cyclin D1     

Adenovirus-Mediated Sensitization to the Cytotoxic Drugs Docetaxel and Mitoxantrone Is Dependent on Regulatory Domains in the E1ACR1 Gene-Region

Oncolytic adenoviruses have shown promising efficacy in clinical trials targeting prostate cancers that frequently develop resistance to all current therapies. The replication-selective mutants AdΔΔ and dl922–947, defective in pRb-binding, have been demonstrated to synergise with the current standard of care, mitoxantrone and docetaxel, in prostate cancer models. While expression of the early viral E1A gene is essential for the enhanced cell killing, the specific E1A-regions required for the effects are unknown. Here, we demonstrate that replicating mutants deleted in small E1A-domains, binding pRb (dl1108), p300/CBP (dl1104) and p400/TRRAP or p21 (dl1102) sensitize human prostate cancer cells (PC-3, DU145, 22Rv1) to mitoxantrone and docetaxel. Through generation of non-replicating mutants, we demonstrate that the small E1A12S protein is sufficient to potently sensitize all prostate cancer cells to the drugs even in the absence of viral replication and the E1A transactivating domain, conserved region (CR) 3. Furthermore, the p300/CBP-binding domain in E1ACR1 is essential for drug-sensitisation in the absence (AdE1A1104) but not in the presence of the E1ACR3 (dl1104) domain. AdE1A1104 also failed to increase apoptosis and accumulation of cells in G2/M. All E1AΔCR2 mutants (AdE1A1108, dl922–947) and AdE1A1102 or dl1102 enhance cell killing to the same degree as wild type virus. In PC-3 xenografts in vivo the dl1102 mutant significantly prolongs time to tumor progression that is further enhanced in combination with docetaxel. Neither dl1102 nor dl1104 replicates in normal human epithelial cells (NHBE). These findings suggest that additional E1A-deletions might be included when developing more potent replication-selective oncolytic viruses, such as the AdΔCR2-mutants, to further enhance potency through synergistic cell killing in combination with current chemotherapeutics.     

Adenovirus type 5 E1A immortalizes primary rat cells expressing wild-type p53

Adenovirus (Ad) E1A induces apoptosis in cells expressing wild-type p53, and stable transformation by Ad E1A requires the co-introduction of an anti-apoptotic gene such as Ad E1B 19K. Thus, cells immortalized by Ad E1A alone might have lost functional p53. In order to analyze the p53 in rat cells expressing Ad E1A, we established rat cell lines by transfecting primary rat embryo fibroblast (REF) and baby rat kidney (BRK) cells with cloned Ad5 E1A. By using a yeast functional assay, we analyzed p53 in six primary REF and three BRK cell lines immortalized by Ad5 E1A as well as five spontaneously immortalized rat cell lines (REF52, NRK, WFB, Rat-1 and 3Y1). The yeast functional assay revealed that all of the spontaneously and Ad5 ElA-immortalized rat cell lines except for 3Y1 expressed wild-type p53. All of the Ad5 E1A-immortalized rat cell lines contained p53 detectable by immunoprecipitation. Recombinant adenovirus expressing rat p53 cloned from a REF cell line immortalized by Ad5 E1A, as well as that expressing murine wild-type p53, induced apoptosis in p53-null cells in collaboration with E1A. Thus, it is suggested that the mutation of p53 appears to be not frequent in the spontaneous immortalization of primary rat cells, and that the functional loss of wild-type p53 is not a prerequisite of E1A-mediated immortalization.     

Large E1B proteins of adenovirus types 5 and 12 have different effects on p53 and distinct roles in cell transformation.

The formation of complexes between oncoproteins of DNA tumor viruses and the cellular protein p53 is thought to result in inactivation of the growth suppressor function of p53. In cells transformed by nononcogenic human adenovirus type 5 (Ad5), the 55-kDa protein encoded by E1B forms a stable complex with p53 and sequesters it in the cytoplasm. However, the homologous 54-kDa protein of highly oncogenic Ad12 does not detectably associate with p53. Yet in Ad12-transformed cells, p53 is metabolically stable, is present at high levels in the nucleus, and contributes to the oncogenicity of the cells. Such properties have previously been described for mutant forms of p53. Here, we show that stable p53 in Ad12-transformed cells is wild type rather than mutant and that stabilization of p53 is a direct consequence of the expression of the Ad12 E1B protein. We also compared the effects of the E1B proteins on transformation of rodent cells by different combinations of oncogenes. A synergistic interaction was observed for the gene encoding the 54-kDa E1B protein of Ad12 with myc plus ras oncogenes, resembling the effect of mutant p53 on myc plus ras. In contrast, the Ad5 55-kDa E1B protein strongly inhibited transformation by myc plus ras but stimulated transformation by E1A plus ras. The data are explained in terms of different interactions of the two E1B proteins with endogenous p53. The results suggest that in cultured rat cells, endogenous wild-type p53 plays an essential role in cell proliferation, even in the presence of myc plus ras. The dependence on p53 is lost, however, when the adenovirus E1A oncogene is present.     

In vivo potential effects of adenovirus type 5 E1A and E1B on lung carcinogenesis and lymphoproliferative inflammation

Triggering uncontrolled cellular proliferation, chronic inflammation, and/or disruption of p53 activity is critical for tumorigenesis initiated by latent viral oncogenes. The adenovirus type 5 (Ad5) early genes E1A and E1B can maintain lifelong latency in the lungs of patients with chronic pulmonary diseases. To determine the in vivo effects of the latent Ad5 E1A and E1B oncogenes, we have examined the influence of Ad5 E1A and E1B gene products on mouse lung carcinogenesis and inflammation by generation and characterization of lung-specific transgenic mouse models. Here, we show that either the Ad5 E1A 243-amino-acid (aa) protein or the E1B 58-kDa protein was dominantly expressed in the transgenic lung. Preferential expression of Ad5 E1A 243-aa protein alone was not sufficient to induce lung carcinogenesis but resulted in low-grade cellular proliferation and high-grade lymphoproliferative inflammation in the lung. The presence of Ad5 E1B dramatically enhanced the expression of the E1A 243-aa protein, in addition to impairing p53 and apoptosis response, resulting in uncontrolled cellular proliferation, lymphoproliferative inflammation, and metastatic carcinomas in the lung after a period of latency. Our studies may provide clues to understanding the potential in vivo biological effects of Ad5 E1A and E1B latent in the lung and a new scope for assessing in vivo functions of viral genes latent in the infection target tissue.     

The E4orf6/E1B55K E3 ubiquitin ligase complexes of human adenoviruses exhibit heterogeneity in composition and substrate specificity

Although human adenovirus type 5 (Ad5) has been widely studied, relatively little work has been done with other human adenovirus serotypes. The Ad5 E4orf6 and E1B55K proteins form Cul5-based E3 ubiquitin ligase complexes to degrade p53, Mre11, DNA ligase IV, integrin α3, and almost certainly other targets, presumably to optimize the cellular environment for viral replication and perhaps to facilitate persistence or latency. As this complex is essential for the efficient replication of Ad5, we undertook a systematic analysis of the structure and function of corresponding E4orf6/E1B55K complexes from other serotypes to determine the importance of this E3 ligase throughout adenovirus evolution. E4orf6 and E1B55K coding sequences from serotypes representing all subgroups were cloned, and each pair was expressed and analyzed for their capacity to assemble the Cullin-based ligase complex and to degrade substrates following plasmid DNA transfection. The results indicated that all formed Cullin-based E3 ligase complexes but that heterogeneity in both structure and function existed. Whereas Cul5 was present in the complexes of some serotypes, others recruited primarily Cul2, and the Ad16 complex clearly bound both Cul2 and Cul5. There was also heterogeneity in substrate specificity. Whereas all serotypes tested appeared to degrade DNA ligase IV, complexes from some serotypes failed to degrade Mre11, p53, or integrin α3. Thus, a major evolutionary pressure for formation of the adenovirus ligase complex may lie in the degradation of DNA ligase IV; however, it seems possible that the degradation of as-yet-unidentified critical targets or, perhaps even more likely, appropriate combinations of substrates plays a central role for these adenoviruses.     

Anti-adenovirus type 5 cytotoxic T lymphocytes: immunodominant epitopes are encoded by the E1A gene.

Virus specific, major histocompatibility complex-restricted, cytotoxic T lymphocytes (CTL) generated in Fischer strain rats infected with human adenovirus type 5 (Ad5) were found to recognize antigenic determinants encoded within the Ad5 early region 1A (E1A) gene. Preliminary mapping studies suggest that the E1A CTL epitopes are encoded within the regions between bp 625 to 810 and 916 to 974 in the first exon of this gene. These epitope-coding regions occur within subregions of E1A that are conserved functionally, and to some extent structurally (approximately 50% sequence homology), among adenoviruses of different groups. Nevertheless, Ad5-specific CTL lysed only targets infected with adenoviruses of the same group (group C; e.g., Ad2) and not targets infected with adenoviruses of different groups (groups A, B, and E). These results suggest that virus-specific CTL may limit adenoviral dissemination by destroying virus-infected cells at an early stage in the viral replicative cycle, during E1A gene expression. Expression of other adenovirus genes does not appear to be required to target infected cells for elimination by CTL.     

Modulation of the level of expression of cellular genes in adenovirus 12-infected and transformed human cells.

The level of expression of thymidine kinase (TK), heat shock protein 70 (HSP70), beta-tubulin and p53 was assessed in human embryo kidney cells (HEKs) infected with adenovirus type 12 (Ad 12) and Ad 12 early region 1 (E1) mutants. HSP70, beta-tubulin and p53 levels were unchanged but TK activity was dramatically increased following wild-type infection. The initial activation of TK required the expression of the product of the E1A 13S mRNA but sustained expression only occurred with those viruses expressing the E1B proteins as well. A number of human cell lines transformed with either Ad 12 or Ad 5 E1 DNA were also assessed for the level of expression of HSP70, beta-tubulin and p53. Both HSP70 and beta-tubulin levels were greatly increased compared with primary human cells although there was considerable variation between lines. p53 was only expressed at high levels in Ad 12-transformed lines expressing E1A and E1B proteins.     

Association between the cellular p53 and the adenovirus 5 E1B-55kd proteins reduces the oncogenicity of Ad-transformed cells.

The association of the cellular p53 protein with the E1B-55kd protein of adenovirus 5 (Ad5) is thought to result in inactivation of the p53 recessive oncogene product. Here we show that Ad5 E1-transformed 3Y1 rat cells which express low levels of the 55 kd E1B protein do not contain the p53-E1B 55kd complex. These cells have nuclearly located p53 and are highly oncogenic in nude mice. In 3Y1 cells expressing the E1B protein at a sufficiently high level, association between p53 and E1B-55kd occurs, resulting in an almost complete trapping of p53 into a discrete cytoplasmic body. These cells only form tumors after a very long latency period and in the tumors that eventually appear selection has occurred in favor of cells lacking the complex and containing free nuclear p53. Comparable results were found when highly oncogenic Ad12-transformed cells were supertransfected with the Ad5 E1B region. In none of the Ad-transformed mouse, rat and human cell lines examined, could we detect p53 of abnormal molecular weight or association with hsc70, neither could we immunoprecipitate p53 by the mutant specific antibody PAb240. These data suggest that a high level of nuclear p53 with a wild-type conformation contributes to the oncogenicity of Ad transformed cells.     

Transforming genes among three different oncogenic subgroups of human adenoviruses have similar replicative functions.

We have examined the functional similarity of the transforming genes for replicative functions among three different subgroups of human adenoviruses (A, B, and C), using mutant complementation as an assay. A host range deletion mutant (dl201.2) of Ad2 (nononcogenic subgroup C) lacking about 5% of the viral DNA covering two early gene blocks (E1a and E1b) involved in cellular transformation was isolated and tested for its ability to replicate in nonpermissive KB cells in the presence of Ad7 (weakly oncogenic group B) or ad12 (highly oncogenic group A). The complementation of the mutant defect was demonstrated by cleaving the viral DNA extracted from mixed infected cells or the DNA extracted from purified virions from mixed infected cells with restriction endonuclease BamHI, which produces a different cleavage pattern with the DNA of each serotype. It was found that the defects in E1a plus E1b of dl201.2 could be complemented by Ad7 and Ad12, indicating that these genes in Ad2, Ad7, and Ad12 have similar functions during productive infection.     

Adenovirus serotype determines association and localization of the large E1B tumor antigen with cellular tumor antigen p53 in transformed cells.

The distribution and stability of the cellular tumor antigen p53 were studied in baby rat kidney cells transformed by region E1 sequences of nononcogenic adenovirus (Ad) type 5 (Ad5) or oncogenic type 12 (Ad12). In transformed cells expressing the large E1B T antigen of Ad5, p53 was associated with this T antigen. The complexed proteins were concentrated in a cytoplasmic body, which has been shown to consist of a cluster of 8-nm filaments (A. Zantema et al., Virology 142:44-58, 1985). In transformed cells expressing the E1B region of Ad12, however, no association between the viral large T antigen and p53 was detectable. In the latter case, both proteins were found almost exclusively in the nucleus. The stability of p53 in both Ad5- and Ad12-transformed cells was increased relative to that in primary cells or cells immortalized by the E1A region only. Thus, the increased stability of p53 in Ad-transformed cells is not caused by association with a viral T antigen, but it correlates with expression of E1B and with morphological transformation.     

Adenovirus ubiquitin-protein ligase stimulates viral late mRNA nuclear export

Theadenovirus type 5 (Ad5) E1B-55K and E4orf6 proteins are required together to stimulate viral late nuclear mRNA export to the cytoplasm and to restrict host cell nuclear mRNA export during the late phase of infection. Previous studies have shown that these two viral proteins interact with the cellular proteins elongins B and C, cullin 5, RBX1, and additional cellular proteins to form an E3 ubiquitin-protein ligase that polyubiquitinates p53 and probably one or more subunits of the MRE11-RAD50-NBS1 (MRN) complex, directing their proteasomal degradation. The MRN complex is required for cellular DNA double-strand break repair and induction of the DNA damage response by adenovirus infection. To determine if the ability of E1B-55K and E4orf6 to stimulate viral late mRNA nuclear export requires the ubiquitin-protein ligase activity of this viral ubiquitin-protein ligase complex, we designed and expressed a dominant-negative mutant form of cullin 5 in HeLa cells before infection with wild-type Ad5 or the E1B-55K null mutant dl1520. The dominant-negative cullin 5 protein stabilized p53 and the MRN complex, indicating that it inhibited the viral ubiquitin-protein ligase but had no effect on viral early mRNA synthesis, early protein synthesis, or viral DNA replication. However, expression of the dominant-negative cullin 5 protein caused a decrease in viral late protein synthesis and viral nuclear mRNA export similar to the phenotype produced by mutations in E1B-55K. We conclude that the stimulation of adenovirus late mRNA nuclear export by E1B-55K and E4orf6 results from the ubiquitin-protein ligase activity of the adenovirus ubiquitin-protein ligase complex.     

Downregulation of Mdm2 and Mdm4 enhances viral gene expression during adenovirus infection

Successful viral replication entails elimination or bypass of host antiviral mechanisms. Here, we show that shRNA-mediated knockdown of murine double minute (Mdm2) and its paralog Mdm4 enhanced the expression of early and late viral gene products during adenovirus (HAdV) infection. Remarkably, whereas the expression of HAdV genes was low in p53-deficient mouse embryonic fibroblasts (p53KO MEFs), the HAdV early gene products were efficiently expressed in Mdm2/p53 double-knockout (DKO) and Mdm4/p53 DKO MEFs, and viral capsid proteins were produced in Mdm2/p53 DKO MEFs. Thus, Mdm2 and Mdm4 seem to have potent antiviral property. In cells infected with wt HAdV or a mutant virus lacking the E1B-55K gene (dl1520), both Mdm2 and Mdm4 were rapidly depleted, whereas replication-deficient mutant viruses (Ad-GFP) or ΔpTP with deletions within the coding sequence of preterminal binding protein failed to induce their downregulation. Reduced expression of Mdm2 and Mdm4 was not due to general shutoff of host protein synthesis. Additionally, expression of a dominant-negative mutant of Cul5 did not affect Mdm2/Mdm4 downregulation. Thus, viral replication but not the presence of E1B-55K is required for Mdm2/Mdm4 degradation. Surprisingly, treatment of HAdV-infected cells with proteasome inhibitor MG132 only partially restored the protein levels of Mdm2 and Mdm4, suggesting that they may also be downregulated through an additional mechanism independent of proteasome. Interestingly, cyclin D1 and p21 appear to be downregulated similarly during HAdV infection. Collectively, our work provides the first biochemical evidence for antiviral function of Mdm2 and Mdm4 and that viruses employ efficient countermeasure to ensure viral replication.