Epstein-Barr virus nuclear antigen 3C recruits histone deacetylase activity and associates with the corepressors mSin3A and NCoR in human B-cell lines

Selective efficacy of depsipeptide in a xenograft model of Epstein-Barr virus-positive lymphoproliferative disorder

BACKGROUND

Immune-compromised individuals are at increased risk for developing aggressive Epstein-Barr virus (EBV)-associated lymphoproliferative disorders after primary EBV infection or for reactivation of a preexisting latent EBV infection. We evaluated the effect of depsipeptide, a histone deacetylase inhibitor, on EBV-positive lymphoblastoid cell lines (LCLs) and Burkitt lymphoma cell lines in a mouse model and explored its mechanism of action in vitro.

METHODS

We studied EBV-transformed LCLs, which express a latent III (Lat-III) viral gene profile, as do some EBV-positive lymphoproliferative malignancies, and Burkitt lymphoma cell lines, which express a Lat-I viral gene profile. Cell lines were used to characterize depsipeptide-induced apoptosis, which was evaluated by flow cytometry. Flow cytometry, western blot analyses, and histone deacetylase inhibitors were used to investigate components of prodeath and survival pathways in vitro. We studied depsipeptide's effects on survival with a mouse xenograft model of EBV-positive human B-cell tumors (groups of 10 mice). All statistical tests were two-sided.

RESULTS

Depsipeptide (5 mg/m2 of body surface area) treatment was associated with statistically significantly improved survival of mice carrying Lat-III EBV-positive LCL tumors, compared with that of control-treated mice (day 30: for depsipeptide-treated mice, 90% survival, 95% confidence interval [CI] = 73.2% to 100%; for control-treated mice, 20% survival, 95% CI = 5.79% to 69.1%; P<.001), but it was not associated with survival of mice carrying Lat-I EBV-positive Burkitt lymphoma tumors. Depsipeptide induced apoptosis in 64% of LCLs and in 14% of EBV-positive Burkitt lymphoma cells in vitro. Depsipeptide-treated LCL cultures had two distinct cell populations--one sensitive and one resistant to depsipeptide. Depsipeptide-mediated apoptosis was associated with a 12-fold increased level of active caspase 3, but some apoptosis persisted despite z-VAD-fmk treatment to inhibit caspase activity. Depsipeptide-resistant LCLs expressed higher levels of latent membrane protein 1 (LMP1; P = .017), BCL2 (P = .032), and nuclear factor kappaB (NF-kappaB) (P<.001) than depsipeptide-sensitive LCLs; this resistance was circumvented by treatment with PS-1145, an inhibitor of NF-kappaB activation (P<.001).

CONCLUSIONS

Apoptosis is induced by depsipeptide via caspase-dependent and -independent pathways in Lat-III EBV-positive LCLs and is enhanced by inhibiting NF-kappaB activity. Depsipeptide as a treatment for Lat-III EBV-associated lymphoproliferative disorders should be explored further in clinical trials.

Biophysical and mutational analysis of the putative bZIP domain of Epstein-Barr virus EBNA 3C

Epstein-Barr virus nuclear antigen 3C (EBNA 3C) is essential for B-cell immortalization and functions as a regulator of viral and cellular transcription. EBNA 3C contains glutamine-rich and proline-rich domains and a region in the N terminus consisting of a stretch of basic residues followed by a run of leucine residues spaced seven amino acids apart. This N-terminal domain is widely believed to represent a leucine zipper dimerization motif (bZIP). We have performed the first structural and functional analysis of this motif and demonstrated that this domain is not capable of forming stable homodimers. Peptides encompassing the EBNA 3C zipper domain are approximately 54 to 67% alpha-helical in solution but cannot form dimers at physiologically relevant concentrations. Moreover, the EBNA 3C leucine zipper cannot functionally substitute for another homodimerizing zipper domain in domain-swapping experiments. Our data indicate, however, that the EBNA 3C zipper domain behaves as an atypical bZIP domain and is capable of self-associating to form higher-order alpha-helical oligomers. Using directed mutagenesis, we also identified a new role for the bZIP domain in maintaining the interaction between EBNA 3C and RBP-Jkappa in vivo. Disruption of the helical nature of the zipper domain by the introduction of proline residues reduces the ability of EBNA 3C to inhibit EBNA 2 activation and interact with RBP-Jkappa in vivo by 50%, and perturbation of the charge on the basic region completely abolishes this function of EBNA 3C.