CRISPR-Cas9 base editor screening identifies the spectrum of MEN1 mutations impacting menin inhibitors in clinical trials [CRISPRseq]

Menin inhibitors have entered clinical trials for KMT2A-rearranged and NPM1 mutant acute leukemias and are demonstrating promising activity. CRISPR-Cas9 base editor screening previously predicted several MEN1 mutations that have arisen in patients receiving SNDX-5613 and confer menin inhibitor resistance. The extent to which MEN1 mutations will impact each menin inhibitor is currently unknown. We leveraged advances in CRISPR-Cas9 base editing technology to predict the MEN1 mutation profile that will impact five different menin inhibitors in clinical trials. We found key similarities (M327) and differences (C334, E368, V372) in the profile of MEN1 mutations that affect each compound. The co-crystal structure of menin bound to each compound suggests resistance mechanisms related to how each menin inhibitor binds in the KMT2A binding pocket of menin. Our in vitro and in vivo validation suggests that the MEN1 mutations identified and validated with this approach are likely to arise and impact all menin inhibitors. Overall design: This screen was performed in two different MV4;11 cell lines, each constitutively overexpressing a SpCas9 nickase fused to a deaminase enzyme capable of making either A>G edits or C>T edits (A>G editor: ABE8e (V106W), SpG-SpyoCas9 (D10A), pRDA_867, addgene plasmid #231109; C>T editor: TadA-CDd, SpG-SpyoCas9 (D10A), pRDB_092, addgene plasmid # #231110), hereafter called MV4;11 A>G and MV4;11 C>T. With the genomic perturbation platform at the Broad institute, we designed and cloned a guide RNA (sgRNA) library tiling the coding exons of MEN1 (target transcript: NM_000244.4). We then transduced our MEN1-targeting library into MV4;11 A>G and MV4;11 C>T cell lines. After 48 hours, the transduction efficiency was 15%, suggesting a low frequency of multiple guide infection of the same cell. Cells were selected with puromycin and expanded, and drug treatment was initiated 10 days after library transduction. We treated with roughly the 75% growth inhibitory concentration at day 5 and the 97.5% growth inhibitory concentration at day 9 and used this as the concentration during the screen: 13.8 nM DS-1594, 17.3 nM JNJ-6617, 23.4 nM KO-539, 51.6 nM SNDX-5613, 25.2 nM DSP-5336. During the screen, cells were treated in three technical replicates for each condition (DMSO and each of the 5 menin inhibitors). We maintained 2,000x coverage of each sgRNA during the duration of the 21-day screen, during which time cells were split and drug replenished every 3-to-4 days. The screen was terminated after 21 days, at which point the cells exhibited resistance based on their accelerating growth towards the end of the screen. Genomic DNA isolation, library preparation, and sequencing were then performed.