Abstract B40: WEE1 inhibition selectively kills histone H3K36me3-deficient cancers by dNTP starvation

Abstract

Abstract Background: Loss of the histone mark H3 Lysine 36 trimethylation (H3K36me3) is highly prevalent and is associated with poor prognosis. However, there is currently no therapy targeting H3K36me3-deficient cancers. Recently we found a novel synthetic lethal interaction in which H3K36me3-deficient cancers are acutely sensitive to inhibition of WEE1 (a checkpoint kinase). Here we provide our latest and detailed understanding of the underlying mechanism and report our findings in biomarker development. Methods: SETD2 is the sole methyltransferase responsible for generating H3K36me3. We used CRISPR-Cas9 methodology to knockout expression of SETD2 in order to create isogenic cancer cell lines that are negative for H3K36me3. Survival of these isogenic cell lines was measured after treatment with the WEE1 inhibitor AZD1775. Recruitment of transcription factors to the chromatin was analyzed by chromatin immunoprecipitation (ChIP). Rescue experiments were performed by expressing a mutant ribonucleotide reductase (RRM2) that cannot be phosphorylated by CDK (Cyclin-dependent kinase) and is resistant to subsequent degradation by SCF(CyclinF). Immunohistochemistry (IHC) was performed on patient tissue microarrays (>100 samples per cancer type) using a monoclonal antibody against H3K36me3. Results: We showed that SETD2 CRISPR knockout cells are 12-fold more sensitive to the WEE1 inhibitor AZD1775 than parental cells (IC50 = 10 nM vs. 128 nM, p<0.0001) and established that RRM2 (a ribonucleotide reductase subunit that produces deoxynucleotides (dNTPs)) is the target of this synthetic lethal interaction. RRM2 is regulated by two pathways in this context: first, H3K36me3 facilitates RRM2 transcription by recruiting transcription initiation factors to the promoter of RRM2; second, WEE1 inhibition degrades RRM2 through activation of CDK and untimely phosphorylation of RRM2. Therefore WEE1 inhibition in H3K36me3-deficient cells results in critical depletion of RRM2 and dNTPs (70% reduction in dNTP, p<0.01), collapsed replication forks and apoptosis. Accordingly, lethality in SETD2 knockout cells is suppressed by increasing RRM2 transcription or inhibiting RRM2 degradation. Encouraged by a robust regression in xenograft tumors, we tested the clinical potential of treating H3K36me3-deficient cancers with the WEE1 inhibitor. To this end we developed an immunohistochemistry assay for patient tissue microarrays, and found that 46% of renal, 16% of colorectal and 10% of non-small cell lung cancers exhibited markedly reduced levels of H3K36me3. Conclusions: Our study has uncovered the mechanism for the selective killing of H3K36me3-deficient cancers by inhibition of WEE1. We identified novel roles for histone H3K36me3 in promoting gene transcription and DNA replication, and for WEE1 in maintaining nucleotide pools by controlling RRM2 degradation. Our proposed treatment of H3K36me3-deficient cancers is based on synthetic lethality, which provides a less toxic and more effective treatment as it specifically targets cancer cells. Our biomarker analyses revealed that H3K36me3 loss is highly prevalent in renal, colorectal and lung cancers, suggesting that it could be an important therapeutic target in these cancers. H3K36me3 loss can be used as a predictive biomarker for WEE1 inhibitor treatment, and may enable patient selection through immunohistochemistry. Finally, as the WEE1 inhibitor AZD1775, for which we describe a new target, is in clinical trials, we anticipate that these findings will be of immediate clinical relevance. Citation Format: Sophia X. Pfister, Enni Markkanen, Yanyan Jiang, Sovan Sarkar, Mick Woodcock, Lykourgos-Panagiotis Zalmas, Giulia Orlando, Neele Drobnitzky, Grigory Dianov, Songmin Ying, Nicholas B. La Thangue, Vincenzo D'Angiolella, Anderson Ryan, Timothy C. Humphrey. WEE1 inhibition selectively kills histone H3K36me3-deficient cancers by dNTP starvation. [abstract]. In: Proceedings of the AACR Special Conference on Chromatin and Epigenetics in Cancer; Sep 24-27, 2015; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2016;76(2 Suppl):Abstract nr B40.