An autoinhibitory switch of the LSD1 disordered region controls enhancer silencing.

Transcriptional coregulators and transcription factors (TFs) contain intrinsically disordered regions (IDRs) that are critical for their association and function in gene regulation. More recently, IDRs have been shown to promote multivalent protein-protein interactions between coregulators and TFs to drive their association into condensates. By contrast, here we demonstrate how the IDR of the corepressor LSD1 excludes TF association, acting as a dynamic conformational switch that tunes repression of active cis-regulatory elements. Hydrogen-deuterium exchange shows that the LSD1 IDR interconverts between transient open and closed conformational states, the latter of which inhibits partitioning of the protein's structured domains with TF condensates. This autoinhibitory switch controls leukemic differentiation by modulating repression of active cis-regulatory elements bound by LSD1 and master hematopoietic TFs. Together, these studies unveil alternative mechanisms by which disordered regions and their dynamic crosstalk with structured regions can shape coregulator-TF interactions to control cis-regulatory landscapes and cell fate.

KBTBD4 Cancer Hotspot Mutations Drive Neomorphic Degradation of HDAC1/2 Corepressor Complexes.

Cancer mutations can create neomorphic protein-protein interactions to drive aberrant function 1 . As a substrate receptor of the CULLIN3-RBX1 E3 ubiquitin ligase complex, KBTBD4 is recurrently mutated in medulloblastoma (MB) 2 , the most common embryonal brain tumor in children, and pineoblastoma 3 . These mutations impart gain-of-function to KBTBD4 to induce aberrant degradation of the transcriptional corepressor CoREST 4 . However, their mechanism of action remains unresolved. Here, we elucidate the mechanistic basis by which KBTBD4 mutations promote CoREST degradation through engaging HDAC1/2, the direct neomorphic target of the substrate receptor. Using deep mutational scanning, we systematically map the mutational landscape of the KBTBD4 cancer hotspot, revealing distinct preferences by which insertions and substitutions can promote gain-of-function and the critical residues involved in the hotspot interaction. Cryo-electron microscopy (cryo-EM) analysis of two distinct KBTBD4 cancer mutants bound to LSD1-HDAC1-CoREST reveals that a KBTBD4 homodimer asymmetrically engages HDAC1 with two KELCH-repeat propeller domains. The interface between HDAC1 and one of the KBTBD4 propellers is stabilized by the MB mutations, which directly insert a bulky side chain into the active site pocket of HDAC1. Our structural and mutational analyses inform how this hotspot E3-neo-substrate interface can be chemically modulated. First, our results unveil a converging shape complementarity-based mechanism between gain-of-function E3 mutations and a molecular glue degrader, UM171. Second, we demonstrate that HDAC1/2 inhibitors can block the mutant KBTBD4-HDAC1 interface, the aberrant degradation of CoREST, and the growth of KBTBD4-mutant MB models. Altogether, our work reveals the structural and mechanistic basis of cancer mutation-driven neomorphic protein-protein interactions and pharmacological strategies to modulate their action for therapeutic applications.

Asymmetric Engagement of Dimeric CRL3 KBTBD4 by the Molecular Glue UM171 Licenses Degradation of HDAC1/2 Complexes.

UM171 is a potent small molecule agonist of ex vivo human hematopoietic stem cell (HSC) self-renewal, a process that is tightly controlled by epigenetic regulation. By co-opting KBTBD4, a substrate receptor of the CULLIN3-RING E3 ubiquitin ligase complex, UM171 promotes the degradation of members of the CoREST transcriptional corepressor complex, thereby limiting HSC attrition. However, the direct target and mechanism of action of UM171 remain unclear. Here, we reveal that UM171 acts as a molecular glue to induce high-affinity interactions between KBTBD4 and HDAC1 to promote the degradation of select HDAC1/2 corepressor complexes. Through proteomics and chemical inhibitor studies, we discover that the principal target of UM171 is HDAC1/2. Cryo-electron microscopy (cryo-EM) analysis of dimeric KBTBD4 bound to UM171 and the LSD1-HDAC1-CoREST complex unveils an unexpected asymmetric assembly, in which a single UM171 molecule enables a pair of KBTBD4 KELCH-repeat propeller domains to recruit HDAC1 by clamping on its catalytic domain. One of the KBTBD4 propellers partially masks the rim of the HDAC1 active site pocket, which is exploited by UM171 to extend the E3-neo-substrate interface. The other propeller cooperatively strengthens HDAC1 binding via a separate and distinct interface. The overall neomorphic interaction is further buttressed by an endogenous cofactor of HDAC1-CoREST, inositol hexakisphosphate, which makes direct contacts with KBTBD4 and acts as a second molecular glue. The functional relevance of the quaternary complex interaction surfaces defined by cryo-EM is demonstrated by in situ base editor scanning of KBTBD4 and HDAC1. By delineating the direct target of UM171 and its mechanism of action, our results reveal how the cooperativity offered by a large dimeric CRL E3 family can be leveraged by a small molecule degrader and establish for the first time a dual molecular glue paradigm.

Base Editor Scanning Reveals Activating Mutations of DNMT3A.

DNA methyltransferase 3A (DNMT3A) is a de novo cytosine methyltransferase responsible for establishing proper DNA methylation during mammalian development. Loss-of-function (LOF) mutations to DNMT3A, including the hotspot mutation R882H, frequently occur in developmental growth disorders and hematological diseases, including clonal hematopoiesis and acute myeloid leukemia. Accordingly, identifying mechanisms that activate DNMT3A is of both fundamental and therapeutic interest. Here, we applied a base editor mutational scanning strategy with an improved DNA methylation reporter to systematically identify DNMT3A activating mutations in cells. By integrating an optimized cellular recruitment strategy with paired isogenic cell lines with or without the LOF hotspot R882H mutation, we identify and validate three distinct hyperactivating mutations within or interacting with the regulatory ADD domain of DNMT3A, nominating these regions as potential functional target sites for pharmacological intervention. Notably, these mutations are still activating in the context of a heterozygous R882H mutation. Altogether, we showcase the utility of base editor scanning for discovering functional regions of target proteins.

Systematic identification of anticancer drug targets reveals a nucleus-to-mitochondria ROS-sensing pathway.

Multiple anticancer drugs have been proposed to cause cell death, in part, by increasing the steady-state levels of cellular reactive oxygen species (ROS). However, for most of these drugs, exactly how the resultant ROS function and are sensed is poorly understood. It remains unclear which proteins the ROS modify and their roles in drug sensitivity/resistance. To answer these questions, we examined 11 anticancer drugs with an integrated proteogenomic approach identifying not only many unique targets but also shared ones-including ribosomal components, suggesting common mechanisms by which drugs regulate translation. We focus on CHK1 that we find is a nuclear H2O2 sensor that launches a cellular program to dampen ROS. CHK1 phosphorylates the mitochondrial DNA-binding protein SSBP1 to prevent its mitochondrial localization, which in turn decreases nuclear H2O2. Our results reveal a druggable nucleus-to-mitochondria ROS-sensing pathway-required to resolve nuclear H2O2 accumulation and mediate resistance to platinum-based agents in ovarian cancers.

Drug addiction unveils a repressive methylation ceiling in EZH2-mutant lymphoma.

Drug addiction, a phenomenon where cancer cells paradoxically depend on continuous drug treatment for survival, has uncovered cell signaling mechanisms and cancer codependencies. Here we discover mutations that confer drug addiction to inhibitors of the transcriptional repressor polycomb repressive complex 2 (PRC2) in diffuse large B-cell lymphoma. Drug addiction is mediated by hypermorphic mutations in the CXC domain of the catalytic subunit EZH2, which maintain H3K27me3 levels even in the presence of PRC2 inhibitors. Discontinuation of inhibitor treatment leads to overspreading of H3K27me3, surpassing a repressive methylation ceiling compatible with lymphoma cell survival. Exploiting this vulnerability, we show that inhibition of SETD2 similarly induces the spread of H3K27me3 and blocks lymphoma growth. Collectively, our findings demonstrate that constraints on chromatin landscapes can yield biphasic dependencies in epigenetic signaling in cancer cells. More broadly, we highlight how approaches to identify drug addiction mutations can be leveraged to discover cancer vulnerabilities.

Identification of chemotherapy targets reveals a nucleus-to-mitochondria ROS sensing pathway.

Multiple chemotherapies are proposed to cause cell death in part by increasing the steady-state levels of cellular reactive oxygen species (ROS). However, for most of these drugs exactly how the resultant ROS function and are sensed is poorly understood. In particular, it's unclear which proteins the ROS modify and their roles in chemotherapy sensitivity/resistance. To answer these questions, we examined 11 chemotherapies with an integrated proteogenomic approach identifying many unique targets for these drugs but also shared ones including ribosomal components, suggesting one mechanism by which chemotherapies regulate translation. We focus on CHK1 which we find is a nuclear H 2 O 2 sensor that promotes an anti-ROS cellular program. CHK1 acts by phosphorylating the mitochondrial-DNA binding protein SSBP1, preventing its mitochondrial localization, which in turn decreases nuclear H 2 O 2 . Our results reveal a druggable nucleus-to-mitochondria ROS sensing pathway required to resolve nuclear H 2 O 2 accumulation, which mediates resistance to platinum-based chemotherapies in ovarian cancers.

Chromatin complex dependencies reveal targeting opportunities in leukemia.

Chromatin regulators are frequently mutated in human cancer and are attractive drug targets. They include diverse proteins that share functional domains and assemble into related multi-subunit complexes. To investigate functional relationships among these regulators, here we apply combinatorial CRISPR knockouts (KOs) to test over 35,000 gene-gene pairings in leukemia cells, using a library of over 300,000 constructs. Top pairs that demonstrate either compensatory non-lethal interactions or synergistic lethality enrich for paralogs and targets that occupy the same protein complex. The screen highlights protein complex dependencies not apparent in single KO screens, for example MCM histone exchange, the nucleosome remodeling and deacetylase (NuRD) complex, and HBO1 (KAT7) complex. We explore two approaches to NuRD complex inactivation. Paralog and non-paralog combinations of the KAT7 complex emerge as synergistic lethal and specifically nominate the ING5 PHD domain as a potential therapeutic target when paired with other KAT7 complex member losses. These findings highlight the power of combinatorial screening to provide mechanistic insight and identify therapeutic targets within redundant networks.

Base editor scanning charts the DNMT3A activity landscape.

DNA methylation is critical for regulating gene expression, necessitating its accurate placement by enzymes such as the DNA methyltransferase DNMT3A. Dysregulation of this process is known to cause aberrant development and oncogenesis, yet how DNMT3A is regulated holistically by its three domains remains challenging to study. Here, we integrate base editing with a DNA methylation reporter to perform in situ mutational scanning of DNMT3A in cells. We identify mutations throughout the protein that perturb function, including ones at an interdomain interface that block allosteric activation. Unexpectedly, we also find mutations in the PWWP domain, a histone reader, that modulate enzyme activity despite preserving histone recognition and protein stability. These effects arise from altered PWWP domain DNA affinity, which we show is a noncanonical function required for full activity in cells. Our findings highlight mechanisms of interdomain crosstalk and demonstrate a generalizable strategy to probe sequence-activity relationships of nonessential chromatin regulators.

Gain-of-Function Genetic Alterations of G9a Drive Oncogenesis.

Epigenetic regulators, when genomically altered, may become driver oncogenes that mediate otherwise unexplained pro-oncogenic changes lacking a clear genetic stimulus, such as activation of the WNT/ß-catenin pathway in melanoma. This study identifies previously unrecognized recurrent activating mutations in the G9a histone methyltransferase gene, as well as G9a genomic copy gains in approximately 26% of human melanomas, which collectively drive tumor growth and an immunologically sterile microenvironment beyond melanoma. Furthermore, the WNT pathway is identified as a key tumorigenic target of G9a gain-of-function, via suppression of the WNT antagonist DKK1. Importantly, genetic or pharmacologic suppression of mutated or amplified G9a using multiple in vitro and in vivo models demonstrates that G9a is a druggable target for therapeutic intervention in melanoma and other cancers harboring G9a genomic aberrations. SIGNIFICANCE: Oncogenic G9a abnormalities drive tumorigenesis and the "cold" immune microenvironment by activating WNT signaling through DKK1 repression. These results reveal a key druggable mechanism for tumor development and identify strategies to restore "hot" tumor immune microenvironments.This article is highlighted in the In This Issue feature, p. 890.

Combining Glucose Units, m/z, and Collision Cross Section Values: Multiattribute Data for Increased Accuracy in Automated Glycosphingolipid Glycan Identifications and Its Application in Triple Negative Breast Cancer.

Glycan head-groups attached to glycosphingolipids (GSLs) found in the cell membrane bilayer can alter in response to external stimuli and disease, making them potential markers and/or targets for cellular disease states. To identify such markers, comprehensive analyses of glycan structures must be undertaken. Conventional analyses of fluorescently labeled glycans using hydrophilic interaction high-performance liquid chromatography (HILIC) coupled with mass spectrometry (MS) provides relative quantitation and has the ability to perform automated glycan assignments using glucose unit (GU) and mass matching. The use of ion mobility (IM) as an additional level of separation can aid the characterization of closely related or isomeric structures through the generation of glycan collision cross section (CCS) identifiers. Here, we present a workflow for the analysis of procainamide-labeled GSL glycans using HILIC-IM-MS and a new, automated glycan identification strategy whereby multiple glycan attributes are combined to increase accuracy in automated structural assignments. For glycan matching and identification, an experimental reference database of GSL glycans containing GU, mass, and CCS values for each glycan was created. To assess the accuracy of glycan assignments, a distance-based confidence metric was used. The assignment accuracy was significantly better compared to conventional HILIC-MS approaches (using mass and GU only). This workflow was applied to the study of two Triple Negative Breast Cancer (TNBC) cell lines and revealed potential GSL glycosylation signatures characteristic of different TNBC subtypes.

CRISPR-suppressor scanning reveals a nonenzymatic role of LSD1 in AML.

Understanding the mechanism of small molecules is a critical challenge in chemical biology and drug discovery. Medicinal chemistry is essential for elucidating drug mechanism, enabling variation of small molecule structure to gain structure-activity relationships (SARs). However, the development of complementary approaches that systematically vary target protein structure could provide equally informative SARs for investigating drug mechanism and protein function. Here we explore the ability of CRISPR-Cas9 mutagenesis to profile the interactions between lysine-specific histone demethylase 1 (LSD1) and chemical inhibitors in the context of acute myeloid leukemia (AML). Through this approach, termed CRISPR-suppressor scanning, we elucidate drug mechanism of action by showing that LSD1 enzyme activity is not required for AML survival and that LSD1 inhibitors instead function by disrupting interactions between LSD1 and the transcription factor GFI1B on chromatin. Our studies clarify how LSD1 inhibitors mechanistically operate in AML and demonstrate how CRISPR-suppressor scanning can uncover novel aspects of target biology.

Transcription elongation factors represent in vivo cancer dependencies in glioblastoma.

Glioblastoma is a universally lethal cancer with a median survival time of approximately 15 months. Despite substantial efforts to define druggable targets, there are no therapeutic options that notably extend the lifespan of patients with glioblastoma. While previous work has largely focused on in vitro cellular models, here we demonstrate a more physiologically relevant approach to target discovery in glioblastoma. We adapted pooled RNA interference (RNAi) screening technology for use in orthotopic patient-derived xenograft models, creating a high-throughput negative-selection screening platform in a functional in vivo tumour microenvironment. Using this approach, we performed parallel in vivo and in vitro screens and discovered that the chromatin and transcriptional regulators needed for cell survival in vivo are non-overlapping with those required in vitro. We identified transcription pause-release and elongation factors as one set of in vivo-specific cancer dependencies, and determined that these factors are necessary for enhancer-mediated transcriptional adaptations that enable cells to survive the tumour microenvironment. Our lead hit, JMJD6, mediates the upregulation of in vivo stress and stimulus response pathways through enhancer-mediated transcriptional pause-release, promoting cell survival specifically in vivo. Targeting JMJD6 or other identified elongation factors extends survival in orthotopic xenograft mouse models, suggesting that targeting transcription elongation machinery may be an effective therapeutic strategy for glioblastoma. More broadly, this study demonstrates the power of in vivo phenotypic screening to identify new classes of 'cancer dependencies' not identified by previous in vitro approaches, and could supply new opportunities for therapeutic intervention.

Characterization of H type 1 and type 1 N-acetyllactosamine glycan epitopes on ovarian cancer specifically recognized by the anti-glycan monoclonal antibody mAb-A4.

Cancer-specific glycans of ovarian cancer are promising epitopes for targeting with monoclonal antibodies (mAb). Despite their potential, structural characterization of these glycan epitopes remains a significant challenge in mAb preclinical development. Our group generated the monoclonal antibody mAb-A4 against human embryonic stem cells (hESC), which also bound specifically to N-glycans present on 11 of 19 ovarian cancer (OC) and 8 of 14 breast cancer cell lines tested. Normal cell lines and tissue were unstained by mAb-A4. To characterize the N-linked glycan epitopes on OC cell lines targeted by mAb-A4, we used glycosidases, glycan microarray, siRNA, and advanced high sensitivity matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). The mAb-A4 epitopes were found to be Fucα1-2Galß1-3GlcNAcß (H type 1) and Galß1-3GlcNAcß (type 1 LacNAc). These structures were found to be present on multiple proteins from hESC and OC. Importantly, endo-ß-galactosidase coupled with MALDI-MS allowed these two epitopes, for the first time, to be directly identified on the polylactosamines of N-glycans of SKOV3, IGROV1, OV90, and OVCA433. Furthermore, siRNA knockdown of B3GALT5 expression in SKOV3 demonstrated that mAb-A4 binding was dependent on B3GALT5, providing orthogonal evidence of the epitopes' structures. The recognition of oncofetal H type 1 and type 1 LacNAc on OC by mAb-A4 is a novel and promising way to target OC and supports the theory that cancer can acquire stem-like phenotypes. We propose that the orthogonal framework used in this work could be the basis for advancing anti-glycan mAb characterization.

Adaptive Chromatin Remodeling Drives Glioblastoma Stem Cell Plasticity and Drug Tolerance.

Glioblastoma, the most common and aggressive malignant brain tumor, is propagated by stem-like cancer cells refractory to existing therapies. Understanding the molecular mechanisms that control glioblastoma stem cell (GSC) proliferation and drug resistance may reveal opportunities for therapeutic interventions. Here we show that GSCs can reversibly transition to a slow-cycling, persistent state in response to targeted kinase inhibitors. In this state, GSCs upregulate primitive developmental programs and are dependent upon Notch signaling. This transition is accompanied by widespread redistribution of repressive histone methylation. Accordingly, persister GSCs upregulate, and are dependent on, the histone demethylases KDM6A/B. Slow-cycling cells with high Notch activity and histone demethylase expression are present in primary glioblastomas before treatment, potentially contributing to relapse. Our findings illustrate how cancer cells may hijack aspects of native developmental programs for deranged proliferation, adaptation, and tolerance. They also suggest strategies for eliminating refractory tumor cells by targeting epigenetic and developmental pathways.

Characterizing Protein Glycosylation through On-Chip Glycan Modification and Probing.

Glycans are critical to protein biology and are useful as disease biomarkers. Many studies of glycans rely on clinical specimens, but the low amount of sample available for some specimens limits the experimental options. Here we present a method to obtain information about protein glycosylation using a minimal amount of protein. We treat proteins that were captured or directly spotted in small microarrays (2.2 mm × 2.2 mm) with exoglycosidases to successively expose underlying features, and then we probe the native or exposed features using a panel of lectins or glycan-binding reagents. We developed an algorithm to interpret the data and provide predictions about the glycan motifs that are present in the sample. We demonstrated the efficacy of the method to characterize differences between glycoproteins in their sialic acid linkages and N-linked glycan branching, and we validated the assignments by comparing results from mass spectrometry and chromatography. The amount of protein used on-chip was about 11 ng. The method also proved effective for analyzing the glycosylation of a cancer biomarker in human plasma, MUC5AC, using only 20 µL of the plasma. A glycan on MUC5AC that is associated with cancer had mostly 2,3-linked sialic acid, whereas other glycans on MUC5AC had a 2,6 linkage of sialic acid. The on-chip glycan modification and probing (on-chip GMAP) method provides a platform for analyzing protein glycosylation in clinical specimens and could complement the existing toolkit for studying glycosylation in disease.

RBPJ maintains brain tumor-initiating cells through CDK9-mediated transcriptional elongation.

Glioblastomas co-opt stem cell regulatory pathways to maintain brain tumor-initiating cells (BTICs), also known as cancer stem cells. NOTCH signaling has been a molecular target in BTICs, but NOTCH antagonists have demonstrated limited efficacy in clinical trials. Recombining binding protein suppressor of hairless (RBPJ) is considered a central transcriptional mediator of NOTCH activity. Here, we report that pharmacologic NOTCH inhibitors were less effective than targeting RBPJ in suppressing tumor growth. While NOTCH inhibitors decreased canonical NOTCH gene expression, RBPJ regulated a distinct profile of genes critical to BTIC stemness and cell cycle progression. RBPJ was preferentially expressed by BTICs and required for BTIC self-renewal and tumor growth. MYC, a key BTIC regulator, bound the RBPJ promoter and treatment with a bromodomain and extraterminal domain (BET) family bromodomain inhibitor decreased MYC and RBPJ expression. Proteomic studies demonstrated that RBPJ binds CDK9, a component of positive transcription elongation factor b (P-TEFb), to target gene promoters, enhancing transcriptional elongation. Collectively, RBPJ links MYC and transcriptional control through CDK9, providing potential nodes of fragility for therapeutic intervention, potentially distinct from NOTCH.

Insulator dysfunction and oncogene activation in IDH mutant gliomas.

Gain-of-function IDH mutations are initiating events that define major clinical and prognostic classes of gliomas. Mutant IDH protein produces a new onco-metabolite, 2-hydroxyglutarate, which interferes with iron-dependent hydroxylases, including the TET family of 5'-methylcytosine hydroxylases. TET enzymes catalyse a key step in the removal of DNA methylation. IDH mutant gliomas thus manifest a CpG island methylator phenotype (G-CIMP), although the functional importance of this altered epigenetic state remains unclear. Here we show that human IDH mutant gliomas exhibit hypermethylation at cohesin and CCCTC-binding factor (CTCF)-binding sites, compromising binding of this methylation-sensitive insulator protein. Reduced CTCF binding is associated with loss of insulation between topological domains and aberrant gene activation. We specifically demonstrate that loss of CTCF at a domain boundary permits a constitutive enhancer to interact aberrantly with the receptor tyrosine kinase gene PDGFRA, a prominent glioma oncogene. Treatment of IDH mutant gliomaspheres with a demethylating agent partially restores insulator function and downregulates PDGFRA. Conversely, CRISPR-mediated disruption of the CTCF motif in IDH wild-type gliomaspheres upregulates PDGFRA and increases proliferation. Our study suggests that IDH mutations promote gliomagenesis by disrupting chromosomal topology and allowing aberrant regulatory interactions that induce oncogene expression.

A Multiplexed System for Quantitative Comparisons of Chromatin Landscapes.

Genome-wide profiling of histone modifications can provide systematic insight into the regulatory elements and programs engaged in a given cell type. However, conventional chromatin immunoprecipitation and sequencing (ChIP-seq) does not capture quantitative information on histone modification levels, requires large amounts of starting material, and involves tedious processing of each individual sample. Here, we address these limitations with a technology that leverages DNA barcoding to profile chromatin quantitatively and in multiplexed format. We concurrently map relative levels of multiple histone modifications across multiple samples, each comprising as few as a thousand cells. We demonstrate the technology by monitoring dynamic changes following inhibition of p300, EZH2, or KDM5, by linking altered epigenetic landscapes to chromatin regulator mutations, and by mapping active and repressive marks in purified human hematopoietic stem cells. Hence, this technology enables quantitative studies of chromatin state dynamics across rare cell types, genotypes, environmental conditions, and drug treatments.

Mediator kinase inhibition further activates super-enhancer-associated genes in AML.

Super-enhancers (SEs), which are composed of large clusters of enhancers densely loaded with the Mediator complex, transcription factors and chromatin regulators, drive high expression of genes implicated in cell identity and disease, such as lineage-controlling transcription factors and oncogenes. BRD4 and CDK7 are positive regulators of SE-mediated transcription. By contrast, negative regulators of SE-associated genes have not been well described. Here we show that the Mediator-associated kinases cyclin-dependent kinase 8 (CDK8) and CDK19 restrain increased activation of key SE-associated genes in acute myeloid leukaemia (AML) cells. We report that the natural product cortistatin A (CA) selectively inhibits Mediator kinases, has anti-leukaemic activity in vitro and in vivo, and disproportionately induces upregulation of SE-associated genes in CA-sensitive AML cell lines but not in CA-insensitive cell lines. In AML cells, CA upregulated SE-associated genes with tumour suppressor and lineage-controlling functions, including the transcription factors CEBPA, IRF8, IRF1 and ETV6 (refs 6-8). The BRD4 inhibitor I-BET151 downregulated these SE-associated genes, yet also has anti-leukaemic activity. Individually increasing or decreasing the expression of these transcription factors suppressed AML cell growth, providing evidence that leukaemia cells are sensitive to the dosage of SE-associated genes. Our results demonstrate that Mediator kinases can negatively regulate SE-associated gene expression in specific cell types, and can be pharmacologically targeted as a therapeutic approach to AML.