The N6-methyladenosine (m 6 A) is an abundant internal RNA modification 1,2 catalysed predominantly by the METTL3-METTL14 methyltransferase complex 3,4 . The m 6 A writer METTL3 has been linked to the initiation and maintenance of acute myeloid leukaemia (AML), but its true therapeutic importance is still unknown [5][6][7] . Here we present the identification and characterisation of a highly potent and selective first-in-class catalytic inhibitor of METTL3 (STM2457) and its co-crystal structure bound to METTL3/METTL14. Treatment with (Extended Data Fig. 2f). These data demonstrate that STM2457 is a highly potent, specific and bioavailable inhibitor of METTL3, suitable for in vivo investigations. Cellular and molecular effects of STM2457To study the anti-leukaemic potential of STM2457 we examined the proliferation of a panel of human AML cell lines post-treatment and detected significant growth reduction in a concentration-dependent manner (Fig. 2a) while we found that STM2457 did not affect the colony-forming ability of normal human cord blood CD34 + cells (Extended Data Fig. 3a). We also observed no impact on the proliferation of MOLM-13 cells treated with the control small molecule STM2120, unlike our observations with STM2457 (Extended Data Fig. 3b). Additionally, treatment with STM2457 significantly reduced the clonogenic potential of primary murine AML cells (Fig. 2b and Extended Data Fig. 3c), while having no effect on normal haematopoietic stem and progenitor cells (HSPCs) (Fig. 2c). Pharmacological inhibition of METTL3 also caused significant myeloid differentiation 6,11 and cell cycle arrest in MOLM-13 and primary murine AML cells (Fig. 2d, e). In contrast, the same effects were not identified using the non-leukaemic haemopoietic cell line HPC7 (Fig. 3e and Extended Data Fig. 3d). Moreover, treatment with STM2457 induced apoptosis in human and mouse AML models but not in normal non-leukaemic haemopoietic cells (Fig. 2f and Extended Data Fig. 3e). To assess the impact of pharmacological inhibition of METTL3 on two known METTL3 biomarkers associated with AML, SP1 6,12 and BRD4 13,14 , we treated MOLM-13 cells with STM2457 and observed a dose-dependent reduction of SP1 and BRD4 protein levels (Fig. 2g). Notably, ectopic expression of SP1 significantly reduced the sensitivity of MOLM-13 cells to STM2457 (Extended Data Fig. 3f, g). These data establish that the catalytic function of METTL3 is important for leukaemia growth, in line with previous findings 6,7,15 . We next sought to investigate the molecular mechanism by which STM2457 affects AML. RNAseq analysis of MOLM-13 cells treated with STM2457 revealed 1,338 up-regulated and 489 down-regulated genes (Extended Data Fig. 4a and Supplementary Table 1). Gene ontology (GO) analysis of the differentially expressed genes showed enrichment in pathways related to myeloid differentiation, cell cycle and leukaemia progression (Extended Data Fig. 4b, c) in close agreement with our phenotypic observations (Supplementary Table 2). To examine the impact of the pharmac...

No abstract

A major cause of failed drug discovery programs is suboptimal target selection, resulting in the development of drug candidates that are potent inhibitors, but ineffective at treating the disease. In the genomics era, the availability of large biomedical datasets with genome-wide readouts has the potential to transform target selection and validation. In this study we investigate how computational intelligence methods can be applied to predict novel therapeutic targets in oncology. We compared different machine learning classifiers applied to the task of drug target classification for nine different human cancer types. For each cancer type, a set of “known” target genes was obtained and equally-sized sets of “non-targets” were sampled multiple times from the human protein-coding genes. Models were trained on mutation, gene expression (TCGA), and gene essentiality (DepMap) data. In addition, we generated a numerical embedding of the interaction network of protein-coding genes using deep network representation learning and included the results in the modeling. We assessed feature importance using a random forests classifier and performed feature selection based on measuring permutation importance against a null distribution. Our best models achieved good generalization performance based on the AUROC metric. With the best model for each cancer type, we ran predictions on more than 15,000 protein-coding genes to identify potential novel targets. Our results indicate that this approach may be useful to inform early stages of the drug discovery pipeline.

Acute myeloid leukemia (AML) is an aggressive cancer with a poor prognosis, for which the therapeutic landscape has changed little for decades. New evidence has revealed an important role for RNA modifications in cancer development and maintenance via the catalytic function of RNA-modifying enzymes. We and others have recently shown that METTL3, the RNA methyltransferase responsible for the deposition of N-6-methyl groups on adenosine (m6A) in mRNA, is a promising therapeutic target for AML1,2. Here we present the in vitro and in vivo characterization of novel small molecule inhibitors of METTL3 as an effective therapeutic strategy in AML. Recently, we generated a comprehensive catalogue of RNA-modifying enzymes that are essential for AML cells using CRISPR-Cas9 recessive screens and characterised METTL3 as a novel therapeutic candidate through its effects on mRNA translational efficiency of key leukemia oncogenes1. Using a structure-guided medicinal chemistry platform we developed and optimised small molecule inhibitors of METTL3 from 2 distinct chemical series. Here we demonstrate that compounds 1 and 2 show biochemical inhibition of METTL3 enzyme with single digit nanomolar potency, while direct binding to METTL3 was confirmed by Surface Plasmon Resonance (SPR) analysis with comparable potency between compounds. Additionally, we developed compound 3 as an inactive analog which was confirmed inactive in enzyme assays (>50 µM IC50). Importantly, we verified that compounds 1 and 2 are selective for METTL3 and do not inhibit a panel of other RNA, DNA or protein methyltransferases tested (>10 µM IC50). Cellular target engagement was confirmed by demonstrating that compounds 1 and 2 reduced m6A levels and inhibited the protein expression of METTL3-dependent m6A substrates in mouse and human AML models, including SP1, with nanomolar potency. Furthermore, treatment of MOLM13 cells with compounds 1 and 2 inhibited their proliferation with comparable potency to SP1 inhibition. The same anti-proliferative effect was observed using a large panel of human AML cell lines. In addition, polyribosome profiling in MOLM13 cells treated with compounds 1 and 2 revealed enhanced blocking of mRNA translation, mirroring the effects derived from the genetic inhibition of METTL3. Notably, all of the above effects were not observed when the inactive analog (compound 3) was used, further highlighting the specificity and sensitivity of our active candidates. We subsequently performed in vivo characterisation of compound 1. This compound exhibited excellent bioavailability after oral or intraperitoneal administration with good dose-proportional exposure in mice and a half-life of 3.5 hours. It also appeared to be well-tolerated with no body weight loss or clinical signs of toxicity. We also evaluated its anti-tumor effects in patient derived xenotransplantation experiments (PDX) as well as transplantation experiments using an MLL-AF9 driven primary murine AML model. Daily dosing of 30 mg/kg significantly inhibited AML expansion and reduced spleen weight compared to vehicle control, indicating a pronounced anti-tumor effect in vivo. Target engagement was confirmed in bone marrow and spleen as measured by the reduction of METTL3-dependent m6A targets. Importantly, we went on to demonstrate that, while the pharmacological inhibition of METTL3 is required for AML cell survival, it was dispensable for normal hematopoiesis. Collectively, we describe the detailed characterization of potent and selective inhibitors of the METTL3 RNA methyltransferase, and demonstrate their activity and utility using biochemical, cellular and in vivo systems. We show that inhibition of METTL3 by small molecules in vivo leads to strong anti-tumor effects in physiologically and clinically relevant models of AML. To our knowledge, this is the first study demonstrating in vivo activity of inhibitors of an RNA methyltransferase, hence providing proof of concept that RNA modifying enzymes represent a new target class for anti-cancer therapeutics. References Barbieri, I. et al. Promoter-bound METTL3 maintains myeloid leukaemia by m(6)A-dependent translation control. Nature552, 126-131, doi:10.1038/nature24678 (2017). Vu, L. P. et al. The N(6)-methyladenosine (m(6)A)-forming enzyme METTL3 controls myeloid differentiation of normal hematopoietic and leukemia cells. Nat Med23, 1369-1376, doi:10.1038/nm.4416 (2017). Disclosures Yankova: STORM THERAPEUTICS: Employment. Fosbeary:STORM THERAPEUTICS: Employment. Hendrick:STORM THERAPEUTICS: Employment. Leggate:STORM THERAPEUTICS: Employment. Ofir-Rosenfeld:STORM THERAPEUTICS: Employment. Sapetschnig:STORM THERAPEUTICS: Employment. Albertella:STORM THERAPEUTICS: Employment. Blackaby:STORM THERAPEUTICS: Employment. Rausch:STORM THERAPEUTICS: Employment. Vassiliou:Kymab Ltd: Consultancy, Other: Minor Stockholder; Oxstem Ltd: Consultancy; Celgene: Research Funding. Kouzarides:STORM THERAPEUTICS: Equity Ownership.

METTL3 is an RNA methyltransferase which is responsible for the deposition of N-6-methyladenosine (m6A) on mRNA targets such as SP1, to modulate their stability and expression. METTL3 was identified as an essential gene for the growth of AML cells and proposed as a novel target for cancer therapy (Barbieri 2017). We present the in vitro and in vivo characterization of novel small molecule inhibitors of METTL3, which recapitulate the genetic validation of METTL3 as a novel cancer target using a pharmacological audit trail. Small molecule inhibitors from 2 distinct chemical series were identified and optimised using a structure-guided medicinal chemistry platform. Compounds 1 and 2 are from different series, and both showed biochemical inhibition of METTL3 enzyme with single digit nanomolar potency. Direct binding to METTL3 was confirmed by SPR with comparable potency. Compound 3 is an inactive analog which was confirmed inactive in the enzyme assay (IC50 >50microM). Compounds 1-2 are selective for METTL3 and did not inhibit a panel of other RNA, DNA or protein methyltransferases tested (>10microM IC50). Cellular target engagement was confirmed by the demonstration that compounds 1 and 2 inhibited SP1 and Brd4 protein expression with submicromolar potency, whereas the inactive analog compound 3 had no effect. Compounds 1 and 2 treatment of MOLM13 cells inhibited their proliferation which correlated with SP1 inhibition, and compound 3 had no effect, demonstrating that their activity was METTL3-dependent. Compound 1 has excellent oral bioavailability with good dose-proportional exposure in mice and a half-life of 3.5 hrs, and was well-tolerated with no body weight loss or clinical signs. Compound 1 was evaluated for anti-tumor effects in an MLL-AF9 driven primary murine AML model. 30 mg/kg daily oral dosing of compound 1 gave a significant reduction in AML expansion and a reduction in spleen weight compared to vehicle control, indicating a pronounced anti-tumor effect in vivo. Target engagement was confirmed in bone marrow and spleen as measured by reduction of METTL3-dependent m6A targets. We have described the comprehensive characterization of potent and selective inhibitors of the METTL3 RNA methyltransferase, and demonstrated their activity and utility using biochemical, cellular and in vivo systems. We have demonstrated that inhibition of METTL3 by small molecules in vivo leads to a pronounced anti-tumor effect in a physiologically relevant model of acute myeloid leukemia. To our knowledge, this is the first demonstration of in vivo activity of inhibitors of an RNA methyltransferase and providing proof of concept that RNA modifying enzymes are a new target class for the development of novel cancer therapeutics. Citation Format: Mark Albertella, Wesley Blackaby, Richard Fosbeary, Alan Hendrick, Dan Leggate, Yaara Ofir-Rosenfeld, Alexandra Sapetschnig, Konstantinos Tzelepis, Eliza Yankova, Tony Kouzarides, Oliver Rausch. A small molecule inhibitor of the RNA m6A writer METTL3 inhibits the development of acute myeloid leukemia (AML) in vivo [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics; 2019 Oct 26-30; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2019;18(12 Suppl):Abstract nr B126. doi:10.1158/1535-7163.TARG-19-B126