Exploiting synthetic lethality to kill H3K36me3-deficient cancers by WEE1 inhibition
1CRUK MRC Oxford Institute for Radiation Oncology, Department of Oncology, University of Oxford, Oxford, OX3 7DQ, UK, Oxford, UK,2Laboratory of Cancer Biology, Department of Oncology, University of Oxford, Oxford, OX3 7DQ, UK, Oxford, UK,33Department of Respiratory and Critical Care Medicine of the Second Affiliated Hospital, and Department of Pharmacology, Zhejiang University School of Medicine, Hangzhou 310058, China., Hangzhou, China
Loss of the histone mark H3K36 trimethylation (H3K36me3) is found in multiple cancer types and is associated with poor patient prognosis, making it a potential therapeutic target. However, there are currently no therapies targeting H3K36me3-deficient cancers. Here we show that these cancers are killed by inhibiting the checkpoint kinase WEE1, and provide insights into the molecular mechanisms.
H3K36me3-deficient isogenic cancer cell lines were generated by three methods: CRISPR knockout of SETD2 trimethyltransferase; overexpression of KDM4A demethylase; or expression of histone H3.3K36M. DNA replication was analysed by DNA fibre assay and iPOND. Deoxynucleotide (dNTP) pools were measured by polymerase-catalysed incorporation of radioactive dNTP.
We have identified a synthetic lethal interaction between loss of both H3K36me3 and WEE1. Accordingly, cancer cells exhibiting reduced or depleted H3K36me3 levels displayed a striking sensitivity to the WEE1 inhibitor AZD1775 compared to wild type. Sensitivity was associated with DNA replication fork arrest, high levels of MUS81-dependent DNA damage, and apoptosis. Consistent with these findings, we found that WEE1 inhibition or H3K36me3 depletion resulted in significantly reduced dNTP levels, which were further critically reduced following treatment of H3K36me3-deficient cells. Xenograft experiments showed oral treatment with AZD1775 induced a robust regression in H3K36me3-deficient tumours (tumour size on day 12 = 50mm3 vs. 291mm3, p<0.0001) with no effect on H3K36me3-proficient tumours
We show that WEE1 inhibition selectively kills H3K36me3-deficient cells by replication stress and suggest that patients with H3K36me3-deficient cancers could benefit from treatment with WEE1 inhibitors.