Abstract 3013: Identification of PHF5A as a common cellular target of splicing-modulating chemical probes

Recent discoveries that splicing factors such SF3B1, U2AF1, SRSF2 are frequently mutated in multiple hematological malignancies including chronic lymphocytic leukaemia and myelodysplastic syndromes have generated interest in therapeutic approaches to target the splicesome dependency in cancer cells bearing mutations in splicing factors. Previously, several structurally unrelated natural compounds including pladienolide, herboxidiene, and FR901464 have been shown to exert potent anti-proliferative effects in cancer cells grown in vitro. Further optimization has led to the discovery of natural product analogs (e.g. E7107) with anti-tumor efficacy in vivo in xenograft models. Target identification has revealed the SF3B complex of the splicesome as the common action site for these compounds. Recent work has demonstrated biological and genetic evidence that single amino acid substitution (R1074H) in SF3B1 completely abolished the anti-proliferative effect of pladienolide derivative E7107, suggesting that SF3B1 is the direct binding partner for pladienolides. However, the same SF3B1 R1074H mutation does not provide equal level of protections for cells treated with herboxidiene derivatives, indicating differential mechanism of action for these two classes of splicing modulators. To identify targets for herboxidiene-like compounds, we have generated resistant HCT116 clones upon continuous administration of herboxidiene derivative H3B-37045 in vitro. Whole exome sequencing from 6 resistant clones revealed a common Y36C mutation in SF3B subunit component PHF5A (SF3B14b). Over-expression of PHF5A Y36C but not the wild-type form in parental HCT116 cells confirmed the protective effect of this mutation to H3B-37045. Surprisingly, PHF5A Y36C expression also conferred resistance to the pladienolide derivative E7107, which indicates that, unlike the SF3B1 R1074H mutation, PHF5A resides within a common node of action site among different splicing modulators. RNA-seq, biochemical and structure homology-modeling analysis suggested that PHF5A Y36C mutation disrupted the action of splicing modulators through interfering with the compounds’ interaction with the SF3B complex. Detailed analysis of the function of the Y36C mutant and wild-type PHF5A in the SF3B complex is currently ongoing. Understanding the function of PHF5A in splicing and the molecular mechanism of Y36C mutation shall provide new insights of the biological role of splicesome, and guide the development of next generation splicesome inhibitors. Citation Format: Teng Teng, Xiaoling Puyang, Shouyong Peng, Jacob Feala, Betty Chan, Jennifer Tsai, Benjamin Caleb, Craig Karr, Eunice Park, Laura Corson, Yoshiharu Mizui, Peter Smith, Nicholas Larsen, Lihua Yu, Markus Warmuth, Ping Zhu, Agustin Chicas. Identification of PHF5A as a common cellular target of splicing-modulating chemical probes. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3013.