BMP-ACVR1 Axis is Critical for Efficacy of PRC2 Inhibitors in B-Cell Lymphoma.

EZH2 is the catalytic subunit of the histone methyltransferase Polycomb Repressive Complex 2 (PRC2), and its somatic activating mutations drive lymphoma, particularly the germinal center B-cell type. Although PRC2 inhibitors, such as tazemetostat, have demonstrated anti-lymphoma activity in patients, the clinical efficacy is not limited to EZH2-mutant lymphoma. In this study, Activin A Receptor Type 1 (ACVR1), a type I Bone Morphogenetic Protein (BMP) receptor, is identified as critical for the anti-lymphoma efficacy of PRC2 inhibitors through a whole-genome CRISPR screen. BMP6, BMP7, and ACVR1 are repressed by PRC2-mediated H3K27me3, and PRC2 inhibition upregulates their expression and signaling in cell and patient-derived xenograft models. Through BMP-ACVR1 signaling, PRC2 inhibitors robustly induced cell cycle arrest and B cell lineage differentiation in vivo. Remarkably, blocking ACVR1 signaling using an inhibitor or genetic depletion significantly compromised the in vitro and in vivo efficacy of PRC2 inhibitors. Furthermore, high levels of BMP6 and BMP7, along with ACVR1, are associated with longer survival in lymphoma patients, underscoring the clinical relevance of this study. Altogether, BMP-ACVR1 exhibits anti-lymphoma function and represents a critical PRC2-repressed pathway contributing to the efficacy of PRC2 inhibitors.

EZH2 W113C is a gain-of-function mutation in B-cell lymphoma enabling both PRC2 methyltransferase activation and tazemetostat resistance.

Polycomb repressive complex 2 (PRC2) suppresses gene transcription by methylating lysine 27 of histone H3 (H3K27) and plays critical roles in embryonic development. Among the core PRC2 subunits, EZH2 is the catalytic subunit and EED allosterically activates EZH2 upon binding trimethylated H3K27 (H3K27me3). Activating mutations on Y641, A677, and A687 within the enzymatic SET (Su(Var)3 to 9, Enhancer-of-zeste, and Trithorax) domain of EZH2 have been associated with enhanced H3K27me3 and tumorigenicity of many cancers including B-cell lymphoma and melanoma. To tackle the critical residues outside the EZH2 SET domain, we examined EZH2 mutations in lymphoma from cancer genome databases and identified a novel gain-of-function mutation W113C, which increases H3K27me3 in vitro and in vivo and promotes CDKN2A silencing to a similar level as EZH2 Y641F. Different from other gain-of-function mutations, this mutation is located in the SET-activation loop at the EZH2 N terminus, which stabilizes the SET domain and facilitates substrate binding. This may explain how the W113C mutation increases PRC2 activity. Tazemetostat is a Food and Drug Administration-approved EZH2-binding inhibitor for follicular lymphoma treatment. Intriguingly, the W113C mutation leads to tazemetostat resistance in both H3K27 methylation and tumor proliferation. Another class of allosteric PRC2 inhibitor binding EED overcomes the resistance, effectively decreases H3K27me3, and blocks tumor proliferation in cells expressing EZH2 W113C. As this mutation is originally identified from lymphoma samples, our results demonstrated its activating characteristic and the deleterious consequence, provide insights on PRC2 regulation, and support the continued exploration of treatment optimization for lymphoma patients.