|Home | About | Journals | Submit | Contact Us | Français|
In a step toward clarifying how acute viral infections provoke the host DNA damage response, Tarakanova et al. (2007) characterized a γ-herpesvirus protein, which phosphorylates histone H2AX during infection, suggesting that the virus actively initiates and benefits from the host DNA damage response.
An interesting front in the war between virus and host is the host cell’s encounter with the viral genome itself. Some viruses infect the host as linear double-stranded (ds) DNA or generate such structures during infection. The infected cell may react by mounting a “classical” double-strand break (DSB)/DNA damage response, with activation of the ATM kinase, phosphorylation of the variant histone H2AX on serine 139 of its C-terminal tail (to form “γ-H2AX”), and phosphorylation of other ATM targets. Such responses could have deleterious consequences for the virus. For example, the mammalian Mre11/Rad50/Nbs1 (MRN) DSB-binding complex triggers concatemerization of the adenoviral genome, rendering it unpackageable (Stracker et al., 2002). Certain adenoviral gene products disrupt MRN, suppress concatemerization, and promote viral replication. DNA damage-induced host cell apoptosis might also limit viral replication; some viral gene products actively suppress apoptosis. In other settings, DNA damage signaling may benefit the virus (Dahl et al., 2005). During polyomavirus infection, ATM activation is accompanied by phosphorylation of H2AX and of the cohesin subunit SMC1. Importantly, viral replication is diminished in ATM−/− or SMC1 phosphomutant cells. How ATM and SMC1 assist viral replication is unknown, but it has been suggested that ATM- and SMC1-mediated checkpoint activation might promote viral replication by prolonging S phase.
Although viral DNA structures might mimic DNA damage, the initiating signal(s) of the damage response in infected cells has not been identified unequivocally. Work published recently in Cell Host and Microbe intriguingly demonstrates that a virally encoded kinase helps to activate the host DNA damage response during murine γ-herpesvirus 68 (γHV68) infection (Tarakanova et al., 2007) (Figure 1). Tarakanova et al. observed marked phosphorylation of H2AX during acute γHV68 infection of mouse myeloid cells—the disease-relevant target tissue. γ-H2AX was distributed in a diffuse nuclear pattern, distinct from the γ-H2AX foci typical of the response to broken mammalian chromosomes (Rogakou et al., 1999). A mutagenesis screen identified a γHV68 mutant that is profoundly defective for induction of H2AX phosphorylation. This mutant harbored a transposon insertion in viral orf36, which encodes a protein kinase. Transfection of uninfected mouse fibroblasts with wild-type, but not kinase-inactive, orf36 triggered robust H2AX phosphorylation. Immunoprecipitated wild-type, but not kinase-inactive, orf36 phosphorylated recombinant H2AX in vitro on serine 139, suggesting that orf36 may phosphorylate H2AX directly—although this remains to be shown with purified components. The Epstein-Barr virus orf36 homolog BGLF4 performed similarly, but other herpesviral orf36 homologs failed to phosphorylate H2AX. The functional significance of H2AX for γHV68 infection is shown by the fact that, 7 days after infection, γHV68 titer in H2AX−/− macrophages was almost 100-fold lower than in H2AX+/+ hosts.
How might H2AX promote γHV68 replication? H2AX accounts for a small fraction of the mammalian histone H2A pool but is the major H2A species in yeast. H2A(X) mutants in yeast and mammals reveal defective sister chromatid recombination (SCR)-mediated DSB repair, suggesting some evolutionary conservation of H2A(X) function (Unal et al., 2004; Xie et al., 2004). S. cerevisiae γ-H2A recruits cohesin to chromatin near the break, promoting sister chromatid cohesion, a function required for efficient SCR (Shroff et al., 2004; Strom et al., 2004). Mouse H2A(X) mutants show additional defects in nonhomologous end joining, transcriptional silencing, and checkpoint function (Fernandez-Capetillo et al., 2004). γ-H2AX in mammalian chromatin flanking the DSB promotes assembly of a specialized chromatin domain containing MDC1 (a direct γ-H2AX-binding protein), the MRN complex, 53BP1, the BRCA1 breast/ovarian tumor suppressor, and other DNA damage response proteins (Stucki and Jackson, 2006). Clearly, γHV68 orf36-mediated phosphorylation of H2AX could influence a number of cellular processes.
γHV68 orf36 is an “early-late” gene, required for expression of certain other viral genes and for efficient viral replication in macrophages, but not in fibroblasts. orf36 phosphorylates H2AX in either cell type. This implies that orf36-mediated H2AX phosphorylation does not have a detectable impact on viral replication in all cell types. Further, in H2AX−/− macrophages, an orf36 mutation further impairs viral replication, suggesting that orf36 has functionally important targets in macrophages additional to H2AX.
It is not clear whether orf36 triggers a “normal” DNA damage response in target chromatin—although it does promote ATM activation and might therefore trigger DNA damage checkpoints. Nor is it yet clear whether, during viral infection, orf36 phosphorylates H2AX predominantly in viral chromatin, in host chromatin, or in the nucleoplasm. If host chromatin is the key target, H2AX/ATM activation might enhance viral replication by prolonging S phase, as discussed previously (Dahl et al., 2005). Alternatively, the key target of orf36 during infection may be viral chromatin. Viral chromatin is hastily assembled upon a naked DNA molecule and presumably lacks the organized histone tail modification patterns and chromatin domains characteristic of mature chromatin. This “epigenetic immaturity” may enhance fragility of the viral chromosome—a problem accentuated by the fact that the viral chromosome is replicating. Replication is a major source of chromosome breakage, and efficient viral “SCR” might be as important for error-free viral replication as conventional SCR is for the host. γ-H2AX induction within viral chromatin might facilitate viral/sister chromatid viral “sister chromatid” cohesion and lead to more successful viral replication cycles, analogous to the deduced role of H2AX in mammalian SCR (Xie et al., 2004). This hypothesis predicts that, in H2AX−/− cells, a higher proportion of viral replication attempts would result in aberrant replication products. γ-H2AX induction might also promote chromatin remodeling or histone replacement, accelerating the organization of viral chromatin into more mature, less fragile epigenetic states. Alternatively, orf36-mediated H2AX phosphorylation in viral chromatin might promote relocation to viral replication centers, or assist other processes such as viral genome circularization or viral integration for latent infection. Perhaps the key effects are limiting for γHV68 replication only in certain cell types (such as the macrophage). Whatever the final picture, γHV68 orf36 promises to be a valuable tool for studying the virus, the host, and the unending duel between the two.