Strategy for Overcoming Crebbp and EP300 Mutations in Lymphoma: Development of First-in-Class HAT Activators

Epigenetic dysregulation is a hallmark of germinal center derived B cell lymphomas. In particular, the monoallelic mutations in histone acetyltransferases (HATs), p300 and CBP are present in approximately 40% of DLBCL and FL patients. The haploinsufficiency of CBP leads to loss of MHC which contributes to both immune escape and lymphomagenesis. The intact wild-type allele offers an opportunity to leverage the normal enzyme to overcome the pathogenic impact of the mutated allele. We reported our lead compound YF2 out of a library (n=70) of HAT activators led to significant cytotoxic effect in lymphoma cell lines compared to its parent compound CTPB, which is not cell membrane permeable. YF2 induced p300-mediated histone acetylation as well as p53 acetylation in cell-free acetylation assay. YF2 also induced the acetylation of H3K14 and H3K27 at the cellular level. Since CREBBP mutations are more frequent than EP300 and are mutually exclusive in patients, it is of high insterest to determine if YF2 affects CBP-mediated acetylation. We measured CBP mediated histone acetylation in a cell free assay by combining the catalytic core of CBP, histone H3 and acetyl-CoA. YF2 increased the level of CBP-mediated Ac-H3K27 (EC50=175.4nM) and Ac-H3K18 (EC50=1330 nM). Studies have shown that in mice lacking Crebbp there is reduced MHC II surface expression on GC cells impairing antigen presentation. To determine if YF2 affects the expression of MHC class I, II and related genes in B cell lymphoma cells, qPCR based gene expression profiling was performed pre and post YF2 exposure. YF2 significantly induces mRNA expression level of HLA-DMA (2.9 fold), PSMB10 (2 fold), and reduced the expression of HLA-G (10 fold). Encode ChIP-seq data confirms enrichement of CBP and p300 in the promotor regions of HLA-DMA and PSMB10 in germinal center B cells, indicating a regulatory effect of YF2 on these genes via HAT modulation. HLA‐G is known to inhibit NK‐cell- and CTL‐mediated cytotoxicity, therefore a reduction in HLA-G may release the inhibition of NK-cell and CTL recognition. Taken together, YF2 may modulate p300 and CBP mediated HLA and related gene expression to induce improved immune surveillance. YF2 is selectively cytotoxic in EP300 mutated B cell lymphoma lines. Since HDAC inhibitors have been approved in the clinic for patients with T cell lymphoma (TCL), we further tested the cytotoxic effect of YF2 in 6 TCL cell lines. Interestingly, YF2 is equivalently cytotoxic in TCL cell lines with IC50 ranging from 4.4 to 14 uM compared to the IC50 values in EP300 mutated BCL lines (5 to 11 uM). Additionally, YF2 induces apoptosis and G1 arrest in a dose dependent manner demonstrated by flow cytometry. We collected three primary lymphoma patient samples including one FL sample carrying CREBBP and p53 mutations and observed that YF2 induced cell death in patient samples, but not in PBMCs from 3 healthy donors, indicating a good therapeutic window. Robust synergy with combination of YF2 and romidepsin was observed in 100% (6/6) EP300 mutated BCL, 40% (2/5) EP300 wild-type BCL and 60% (3/5) TCL cells (EOB = 28). In vivo, SCID‐beige mice were xenografted with SUDHL6 and treated with either vehicle, YF2 40mg/kg, romidepsin 2mg/kg or the combination. Single agent YF2 led to equivalent tumor growth delay as FDA-approved drug romidepsin. The YF2 plus romidepsin combination led to significant tumor growth delay compared to the single agents and control. The efficacy of YF2 and romidepsin in PDX model will be determined in an ongoing study. Given the fact that the RING domain of p300 functions as an inhibitory domain to regulate p300 activity, we hypothesized that YF2 may bind to the RING domain, change the confirmation to facilitate access to the HAT substrate-binding pocket. As predicted by LBias, small molecule binding sites on p300 are on the RING domain and bromodomain. We further checked the potential binding site by docking YF2-like small molecules on the RING domain. We discovered the potential binding site of YF2 in the pocket of RING domain. Kinome screening showed YF2 has no off-target effects on 97 kinases. Ongoing experiments will determine the binding of YF2 to the p300 and CBP catalytic cores. In summary, our lead compound YF2 induces HAT enzymes by activating not only p300 but also CBP mediated acetylation of histone. Modulating HLA genes by YF2 may potentially offer an opportunity to reverse immune evasion adopted by lymphoma cells as a mechanism for survival. Download : Download high-res image (299KB) Download : Download full-size image Disclosures O'Connor: Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Mundipharma: Consultancy, Membership on an entity's Board of Directors or advisory committees; ADCT Therapeutics, Affimed, Agensys, Merck, Seattle Genetics, Spectrum, Trillium, and Verastem Oncology.: Research Funding; TG Therapeutics: Other: Travel Support, Research Funding. Arancio: Appia Pharmaceuticals: Research Funding; Neurokine Therapeutics: Other: Founder. Amengual: Epizyme: Speakers Bureau; Appia Pharmaceuticals: Research Funding.