RNA-binding protein NF90 mediates Polycomb-independent transactivation by EZH2 to promote cancer growth

Background: EZH2 is an enzymatic subunit of the Polycomb Repressive Complex 2 (PRC2) that catalyzes histone H3 lysine 27 trimethylation to suppress gene expression, a role that can be effectively targeted by enzymatic EZH2 inhibitors. Recent evidence suggests that EZH2 plays PRC2-independent roles in activating gene expression. Our preliminary results showed AR as such a direct target of EZH2-mediated transcriptional activation. Nevertheless, critical gaps remain regarding how EZH2 activates AR in prostate cancer (PCa). Methods: EZH2-immunoprecipiated protein lysate was analyzed by mass spectrometry (IP-MS). NF90 or EZH2 co-immunoprecipitation (Co-IP) was also performed in PCa cells to verify their interaction. Western blot and RT-qPCR were conducted to further assess AR mRNA and protein levels in EZH2- and NF90-knockdown (KD) PCa cells. Additionally, ChIP-qPCR was used to investigate the occupancy of EZH2 and NF90 at the AR locus. WST-1 and colony formation assays were used to assess the impact of NF90 KD on PCa cells growth. The TCGA datasets were used to examine gene expression in patient PCa samples. Immunohistochemical (IHC) analyses of tissue microarrays in primary and CRPC samples were performed to evaluate NF90 protein levels. Results: IP-MS in LNCaP cells overexpressing EZH2 WT identified that NF90 as a potential interacting partner of EZH2. Co-IP further revealed that NF90 interacts with EZH2 through their RNA-binding domains, dependent on RNA. Accordingly, like EZH2, NF90 also induced AR gene transcription and downstream signaling. Subsequent ChIP-qPCR analysis revealed that NF90 and EZH2 mutually recruit each other to the AR promoter, where they cooperatively activate AR transcription and enhance downstream AR signaling. This NF90-EZH2 complex is essential for PCa cell growth: depletion of either factor abolishes proliferation, an effect rescued by AR re-expression. Similar to EZH2, NF90 promotes cell cycle gene expression, is upregulated in advanced PCa, and is associated with poor clinical outcomes. Conclusion: Our study demonstrates that RNA binding is essential for EZH2's function as a transcriptional activator and identifies NF90 as a key co-activator of solo-EZH2, collaboratively inducing AR expression and cell growth. Significantly, our findings uncover RNA-mediated protein interactions as a central mechanism underlying PRC2-independent transcriptional activation by EZH2 and establish NF90 as a major EZH2 coactivator, a master regulator of the cell cycle, and a promising therapeutic target in advanced PCa.