Epigenetic silencing via SP140 and H3K27me3 restricts immunogenic viral antigen expression in EBV+ lymphomas Free

Epstein-Barr virus (EBV) infection is implicated in the pathogenesis of a significant subset of human lymphomas. EBV+ tumors exhibit distinct latency programs: latency I (expressing only EBNA1) is minimally immunogenic, whereas latency II/III expresses additional viral antigens (e.g., LMP1/2 and EBNA2/3) that are recognised by EBV-specific cytotoxic T cells (EBV-CTLs). Consequently, allogeneic EBV-CTLs are effective cellular therapy in latency II/III tumors but are largely ineffective against latency I tumors such as EBV+ Burkitt lymphoma (BL) and subsets of diffuse large B-cell lymphoma (DLBCL).

We previously showed that DNMT1 inhibition with decitabine (DCB) can partially induce latency II/III gene expression in latency I BL, sensitizing tumors to EBV-CTLs. One limitation of DNMT1 inhibition is that the induction of latency II/III is only observed in a subset of cells. Here, we investigated the epigenetic mechanisms underlying resistance to latency conversion in EBV+ tumours.

To understand the transcriptional programs limiting latency conversion, we performed RNA-sequencing on EBV+ BL cells that are sensitive vs. resistant to DCB-induced latency conversion.

Transcription factor enrichment analysis revealed SP140, a chromatin reader linked to the repressive mark H3K27me3, as enriched in DCB-resistant cells. To examine whether SP140 is responsible for the resistance to DCB-induced latency conversion, we examined the effect of the SP140 inhibitor, GSK761 on viral latency with or without DCB in latency I EBV+ BL cells (Kem I). When combined with DCB, GSK761 significantly increased the proportion of cells converting to latency II/III cells compared to either drug alone (EBNA2+/LMP1+ cells after treatment with DCB vs DCB+GSK761: 15.4% vs 52.2% adj-p<0.0001). Enhanced latency conversion with DCB+GSK761 was also observed in latency I EBV+ DLBCL cells (Val). Given SP140's role in regulating H3K27me3, we next performed ChIP-qPCR to determine if H3K27me3 at key viral promoters of latency (Cp and LMP1p) could be mediating latency restriction. We observed elevated H3K27me3 in resistant vs. sensitive Kem I cells at Cp (2-fold; adj-p=0.0009) and LMP1p (2.45-fold; adj-p=0.0086). The elevation in H3K27me3 at Cp and LMP1p was confirmed in an additional BL cell line (Rael) and in DLBCL (Val).

Based on this, we evaluated the EZH2 inhibitor, tazemetostat (TAZ), which reduces H3K27me3. Combined treatment with DCB+TAZ increased EBNA2+ LMP1+ cells compared to DCB alone (DCB vs DCB+TAZ: 15.4% vs 38.7% adj-p<0.0001) in Kem I cells. Increased latency conversion with the combination was also observed in DLBCL (Val). The addition of both GSK761 and TAZ to DCB did not further enhance latency conversion, suggesting SP140 promotes resistance via H3K27me3. Given the efficacy of combined DNMT1 and EZH2 inbition in vitro, we evaluated the combination in vivo in BL xenografts. Mice were engrafted with Kem I and then treated with either DCB, TAZ, or the combination. At day 15, tumors were evaluated for latency profiling by flow cytometry. Combined treatment with DCB+TAZ significantly increased the proportion of LMP1-expressing tumor cells in comparison to DCB alone (adj-p≤0.0001).

We next evaluated if enhanced latency conversion improves T-cell mediated killing of BL by allogeneic EBV-CTLs. EBV-CTLs were generated using partially HLA-matched allogeneic peripheral blood mononuclear cells exposed to an EBV peptide mix including latent antigens. EBV-CTLs were co-cultured with BL cells pre-treated with DCB, TAZ, or the combination. T cells exposed to BL pretreated with the combination of DCB and TAZ showed a 1.46-fold increase in IFNg production compared to those exposed to DCB only treated BL (p=0.0032), indicating enhanced tumor immunogenicity.Together, these findings identify SP140 and H3K27me3 as key epigenetic barriers to latency conversion in EBV+ latency I lymphomas. Dual inhibition of DNMT1 and EZH2 enhances expression of immunogenic viral antigens, improving tumor susceptibility to EBV-CTLs. This strategy offers a promising approach for treating otherwise immune-refractory EBV+ lymphomas.