The H3K4 methyltransferase KMT2D is an essential cofactor for GATA1 at erythroid gene enhancers

The H3K4 methyltransferase KMT2D is an essential cofactor for GATA1 at erythroid gene enhancers | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article The H3K4 methyltransferase KMT2D is an essential cofactor for GATA1 at erythroid gene enhancers Peng Xu, Jianxiang Zhang, Ye Xin, Li Cheng, Yuan Xing, Mengli Zhang, and 7 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7853319/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Gene expression during cellular differentiation is coordinated by combinatorial interactions between transcription factors (TFs) and cofactors at promoters and enhancers. The “master TF” GATA1 coordinates gene transcription in a subset of hematopoietic lineages, including erythroid, megakaryocytic, mast, and eosinophil, while repressing the development of other blood lineages. However, the specific cofactors required for GATA1-activated gene expression during hematopoiesis are incompletely defined. We identified the cofactor KMT2D, an H3K4 methyltransferase that collaborates with H3K27 acetyltransferases to activate transcription, in an unbiased CRISPR/Cas9 screen for epigenetic regulators of erythropoiesis. Loss of KMT2D in human erythroid precursors caused developmental arrest with impaired expression of numerous erythroid genes. Mechanistically, KMT2D colocalized with GATA1 on more than one thousand erythroid enhancers associated with over two hundred erythroid genes. In general, co-occupancy of GATA1 and KMT2D at erythroid enhancers was associated with stronger transcriptional activity than occupancy by GATA1 alone. Acute depletion of KMT2D in erythroid precursors caused rapid reductions of H3K4me1 and H3K27ac on a subset of GATA1-bound enhancers and impaired the expression of canonical erythroid genes, including ZFPM1, SLC4A1 , and EPOR . Moreover, acute depletion of GATA1 or KMT2D individually caused downregulation of overlapping gene sets. Thus, KMT2D controls erythropoiesis by selectively activating GATA1-dependent erythroid enhancers. Our studies identify KMT2D as a novel cofactor for transcriptional activation by GATA1 during erythropoiesis. More generally, our findings demonstrate how a lineage-specific TF cooperates with a ubiquitous epigenic regulator to drive lineage-specific gene expression during cellular differentiation. Biological sciences/Genetics/Gene regulation Biological sciences/Developmental biology/Haematopoiesis/Erythropoiesis/Haematopoietic stem cells Biological sciences/Molecular biology/Epigenetics KMT2D GATA1 enhancer transcription regulation hematopoiesis erythropoiesis epigenetics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction In general, cellular differentiation is coordinated by specialized transcriptional regulatory networks consisting of lineage-specific and general transcription factors (TFs) and cofactors, including epigenetic regulators 1 , 2 . Red blood cells are generated from hematopoietic stem and progenitor cells (HSPCs) through a stepwise, highly controlled differentiation process known as erythropoiesis 3 – 7 . During erythropoiesis, the master TF GATA1 binds to enhancers and recruits additional TFs and chromatin modifiers to initiate RNA polymerase II (Pol II)-mediated transcription at promoters 8 – 15 . How GATA1 and other erythroid TFs modulate enhancer activity remains incompletely understood. Active enhancers are marked by the histone modifications H3K4me1 and H3K27ac, which are catalyzed by specific chromatin “writers” 16 . The Complex of Proteins Associated with Set1 (COMPASS)-like KMT2C/KMT2D complexes install H3K4me1 at enhancers to activate transcription in collaboration with the histone acetyltransferases p300/CBP, which catalyze H3K27 acetylation 17 – 24 . While previous studies have uncovered essential roles for several epigenetic modifiers in erythropoiesis 25 – 30 , the role of the KMT2C/KMT2D COMPASS-like complexes in erythropoiesis remains unexplored. We performed a CRISPR/single guide RNA (sgRNA) screen to investigate the functions of 496 epigenetic modifiers in human erythropoiesis. Our data revealed that Cas9-sgRNA targeting genes related to histone H3K4 methylation caused impaired erythroid maturation. Specifically, KMT2D, the enzymatic subunit of the KMT2D COMPASS-like complex, was identified as a top positive regulator of erythroid maturation. We validated that loss of KMT2D delays erythroid maturation in the immortalized erythroblast progenitor line HUDEP-2 and in primary human CD34 + HSPCs. Transcriptome analysis revealed that KMT2D is required for the dynamic activation of approximately 200 erythroid signature genes. To define the immediate consequences of acute KMT2D loss, we established auxin-inducible degron 2 (AID2) systems in two cellular models for erythropoiesis 31 , 32 and showed that hundreds of KMT2D-bound genes were suppressed within 6 hours of KMT2D depletion. Remarkably, more than 7,000 KMT2D-bound regions were enriched for the cognate GATA1 binding motif, and KMT2D colocalized with GATA1 at over 1,000 active enhancers. More importantly, enhancers bound by GATA1 and KMT2D exhibited stronger H3K4me1 and H3K27ac signals and stronger binding of core erythroid TFs GATA1, KLF1, and TAL1 relative to KMT2D-independent enhancers. After KMT2D depletion, H3K4me1 and H3K27ac marks on those enhancers were significantly reduced, resulting in transcriptional repression of the associated target genes. Thus, we propose that KMT2D is essential for GATA1-dependent enhancer activation at a subset of key erythroid genes. Results An unbiased CRISPR/Cas9 screen identifies KMT2D as a critical regulator of erythropoiesis To comprehensively interrogate the functions of epigenetic modifiers in human erythropoiesis, we conducted a CRISPR knockout screen in HUDEP-2 cells, an immortalized erythroblast line that can be induced to undergo terminal maturation (Fig. 1 A) 3 , 31 . HUDEP-2 cells stably expressing Cas9 were transfected with a lentiviral vector library encoding 5,082 sgRNAs targeting a total of 496 genes encoding epigenetic regulators, grown in maintenance medium for 5 days, then switched to erythroid maturation medium for 5 days. Cells were fractionated by immune flow cytometry for surface expression of the erythroid maturation marker Band3 and analyzed by next-generation DNA sequencing for sgRNA representation in Band3 High and Band3 Low cells ( Supplementary table 1 ). Guide RNA sequences enriched in the Band3 Low population are predicted to target positive regulators of erythroid maturation. As validation, sgRNAs targeting the previously reported positive regulator of erythropoiesis FBXO11 were enriched in Band3 Low cells (Fig. 1 B, Supplementary table 1 ) 3 . Notably, KMT2D and NCOA6, members of the KMT2C/D COMPASS-like complexes, were among the top three positive regulators (Fig. 1 B, C; Supplementary table 1 ). Additionally, SETD1A/SETD1B COMPASS components (SETD1A, SETD1B, and CXXC1) and core subunits of COMPASS (RBBP5 and DPY30) were identified as significant positive regulators. Gene ontology (GO) analysis of the 28 top positive regulators of erythroid maturation revealed histone H3 lysine 4 (H3K4) methylation as the highest-scoring biological process, reflecting the importance of COMPASS and COMPASS-like subunits in erythropoiesis ( Supplementary Fig. 1A ). We next analyzed the expression profiles of KMT2C/D and SETD1A/SETD1B complex components during in vitro erythroid differentiation of CD34 + HSPCs. Most subunits started to become upregulated at the CFU-E or proerythroblast stages, with expression of KMT2C/D components maintained at higher levels into later stages of erythroid maturation (Fig. 1 C, D; Supplementary Fig. 1B) . Transfection of Cas9-expressing HUDEP-2 cells with two different KMT2D -targeting sgRNAs caused marked reductions in the expression of KMT2D and the KMT2C/D COMPASS complex-specific subunit KDM6A/UTX (Fig. 1 E). Disruption of KMT2D had no effect on the expansion of HUDEP-2 cells grown in maintenance medium ( Supplementary Fig. 1C ). However, after switching to erythroid maturation medium, KMT2D-depleted cells exhibited reduced expression of cell surface Band3 and less hemoglobinization (Fig. 1 F, G). Next, we transfected HUDEP-2 cells with ribonucleoprotein complexes (RNPs) consisting of Cas9 and one of two different KMT2D -targeting sgRNAs or non-targeting sgRNA control. Editing efficiencies were maintained over six days of growth in expansion medium, indicating that KMT2D -disrupted cells were not outgrown by non-targeted cells ( Supplementary Fig. 1D ). However, during induced erythroid maturation, KMT2D -disrupted cells exhibited reduced expression of mRNAs encoding the erythroid markers CD235a ( GYPA ), Band3 ( SLC4A1 ) and adult globins ( HBB , HBA ) (Fig. 1 H; Supplementary Fig. 1E ). KMT2D is required for activation of erythroid signature genes and proper chromatin accessibility of GATA1 To understand how KMT2D loss impairs erythroid maturation, we performed RNA sequencing (RNA-seq) of HUDEP-2 cells expressing Cas9 and control non-targeting sgRNA (Ctrl) or KMT2D-sg2 (KMT2D KO) in the immature state and after three days of induced maturation (Fig. 2 A). Principal component analysis (PCA) distinguished all four groups ( Supplementary Fig. 2A ). Moreover, the differences between control and KMT2D KO cells before maturation were less than those after induced maturation, indicating a greater effect of KMT2D disruption on the transcriptome of the latter group ( Supplementary Fig. 2A ). In agreement, the number of differentially expressed genes (DEGs), defined as fold change > 1.5 and FDR < 0.05, was greater between control and KMT2D KO cells after maturation vs. before maturation (Fig. 2 B; Supplementary table 2 ). The DEGs identified in immature cells were largely different from those identified during maturation (Fig. 2 C), suggesting a dynamic role for KMT2D in erythroid development. Gene set enrichment analysis (GSEA), gene ontology (GO), and enrichR 33 analysis showed downregulation of genes associated with erythroid differentiation and heme metabolism, and GATA1 target genes in KMT2D KO cells, before and after induced maturation (Fig. 2 D; Supplementary Fig. 2B–D; Supplementary table 3 ). In contrast, HSC or lymphocyte lineage signatures were enriched in gene sets that were upregulated after KMT2D KO ( Supplementary table 3 ). In total, 11,400 genes were expressed in control HUDEP-2 cells after maturation. Of those, 2,237 genes and 1,739 genes were significantly downregulated and upregulated, respectively, relative to immature cells (Fig. 2 E). Among the 1,739 upregulated genes, 367 were KMT2D-dependent and significantly enriched for erythropoiesis-related pathways and GATA1 occupancy (Fig. 2 F–H). Erythroid genes that were downregulated by KMT2D disruption before and after induced maturation include SLC25A37, GYPA, EPOR, SLC4A1, ALAS2 (Fig. 2 I). In contrast, several erythroid genes were not regulated by KMT2D, including WDR26 , JAK2 , ANK1 , and BCL11A ( Supplementary table 2 ). These findings support the conclusion that KMT2D facilitates erythroid maturation by activating selective erythroid signature genes. Given that KMT2D-dependent genes were significantly enriched for GATA1 occupancy, we hypothesized that KMT2D maintains appropriate chromatin accessibility for specific transcription factors (TFs) in erythroid progenitors. To test this hypothesis, we performed genome-wide ATAC-seq in both control and KMT2D KO HUDEP-2 cells. We identified differential accessibility regions (DARs) from a total of 27,919 reproducible peaks using a stringent cutoff (false discovery rate (FDR) controlled P -value 2-fold change). Notably, 659 chromatin accessibility regions were significantly decreased, while 119 regions were significantly increased ( Supplementary Fig. 2E; Supplementary table 2 ). We then predicted the TF occupancy profiles of these DARs using the TRANSFAC and Homer motif database as a reference and scored the enrichment frequency 34 . Regions with decreased chromatin accessibility after KMT2D KO were most significantly enriched for GATA TF binding motifs ( Supplementary Fig. 2F, G; Supplementary table 2 ). These results indicate that KMT2D is selectively required for maintaining proper chromatin accessibility for GATA1 in erythroid progenitor cells. Acute depletion of KMT2D selectively suppresses the expression of GATA1 targets Transcriptome analysis of KMT2D KO HUDEP-2 cells at steady state cannot distinguish direct from indirect effects on gene expression. We established an AID2-based acute protein degradation system 35 , 36 for KMT2D to study its direct target genes. We fused an EGFP-AID degron cassette to the amino (N)-terminus of both endogenous KMT2D loci in HUDEP-2 and K562 cells that were engineered to constitutively express OsTIR1 F74G , a substrate receptor for the Skp1, Cullin, F-box (SCF) ubiquitin ligase complex ( Supplementary Fig. 3A, B ). In the presence of the small molecule 5-Ph-IAA, OsTIR1 F74G mediates the ubiquitination and proteasomal degradation of AID-tagged proteins (Fig. 3 A). Thus, addition of 5-Ph-IAA to AID-KMT2D-HUDEP-2 or -K562 cells resulted in depletion of KMT2D within six hours, which was reversible after drug washout (Fig. 3 B; Supplementary Fig. 3C, 3D ). To determine whether EGFP-AID-tagged KMT2D assembles into the KMT2D complex, we performed immunoprecipitation in nuclear lysates with an anti-KMT2D antibody followed by mass spectrometry. We identified 212 KMT2D-interacting proteins relative to an IgG control ( Supplementary table S4) . Notably, all nine subunits of the KMT2D complex were substantially enriched ( Supplementary Fig. 3E ). Additionally, protein network analysis confirmed known KMT2D-interacting proteins, such as the Mediator and SWI/SNF complexes 37 , 38 ( Supplementary Fig. 3F) . These findings suggest that the N-terminal GFP-AID fusion did not interfere with the proper assembly of the KMT2D complex. The addition of 5-Ph-IAA to AID-KMT2D HUDEP-2 cells had no effect on proliferation in either expansion or maturation medium, but caused a block to terminal maturation, including reduced hemoglobinization and failure to upregulate numerous erythroid marker genes ( Supplementary Fig. 3G–J ). Thus, drug-induced depletion of AID-KMT2D in HUDEP-2 cells phenocopies the effects of KMT2D depletion, suggesting that the fusion protein retains biological function. We next examined transcriptome changes at 6, 12, and 24 hours after 5-Ph-IAA treatment of AID-KMT2D-HUDEP-2 and AID-KMT2D-K562 cells in expansion medium (Fig. 3 C). Principal component analysis revealed distinct gene expression profiles at each timepoint after addition of 5-Ph-IAA ( Supplementary Fig. 3K ), with progressive increases in the number of DEGs from 6 to 24 hours and common downregulated genes at three timepoints (Fig. 3 C, D; Supplementary tables 5 and table 6 ). The DEGs were strongly enriched for occupancy of GATA1 (Fig. 3 E, F; Supplementary table 7 ). To further confirm that GATA1 and KMT2D share common targets, we generated a dTAG degradation system 39 . Specifically, we used lentiviral vectors to overexpress the FKBP F36V -GATA1 fusion gene while disrupting the endogenous GATA1 gene in AID-KMT2D-HUDEP-2 cells (Fig. 3 G). Addition of 5-Ph-IAA or dTAG13 caused acute degradation of KMT2D or GATA1, respectively (Fig. 3 H), and resulted in the downregulation of overlapping gene sets, including SLC25A37 , SLC22A4 , ZFPM1 and EPOR (Fig. 3 I). Taken together, these results suggest that KMT2D directly activates selective GATA1 target genes to facilitate human erythropoiesis. KMT2D and GATA1 co-occupy erythroid-expressed genes We next performed ChIP-seq analysis of AID-KMT2D-HUDEP-2 cells to identify erythroid genes that bound KMT2D. In the absence of 5-Ph-IAA, 7,288 KMT2D occupancy peaks were detected reproducibly in three replicate experiments (Fig. 4 A; Supplementary Fig. 4A, B; Supplementary table 8 ). Occupancy of KMT2D was markedly reduced at 24 hours after the addition of 5-Ph-IAA. Notably, KMT2D occupancy predominated at distal intergenic regions (46%) and introns (32%), compared to promoters (17%) (Fig. 4 B). Homer analysis revealed enrichment of GATA/SCL, GATA, and KLF binding motifs (Fig. 4 C; Supplementary Fig. 4C; Supplementary table 8 ). To test whether KMT2D co-localizes with erythroid transcription factor binding on cis-regulatory elements, we performed an integrated analysis using available ChIP-seq data for GATA1, KLF1, TAL1 occupancy, and histone marks in wild-type HUDEP-2 cells (GSE115357, GSE157311) 3 , 40 . Approximately 35% of KMT2D-bound regions (2,559 peaks) were co-localized with GATA1 (Fig. 4 D). Over half of KMT2D-GATA1 co-bound regions were at enhancers (Fig. 4 D). Moreover, the binding signals of histone marks H3K4me1 and H3K27ac, as well as the erythroid TFs KLF1 and TAL1, were significantly higher at KMT2D-GATA1 co-occupied enhancers compared to enhancers bound by GATA1 alone (Fig. 4 E, F). In addition, genes that were co-occupied by KMT2D and GATA1 were more rapidly suppressed after acute KMT2D degradation than genes that were bound by GATA1 only (Fig. 4 G). Moreover, a co-immunoprecipitation assay identified interactions between GATA1 and the key KMT2D/COMPASS component KDM6A ( Supplementary Fig. 4D and E ). These results suggest that KMT2D/COMPASS complex and GATA1 co-localize on active enhancers to activate gene expression. Next, we investigated whether KMT2D binding occurs at genes that are transcriptionally deregulated after KMT2D depletion. Notably, KMT2D binding was enriched at genes that were downregulated after acute depletion of KMT2D in AID-KMT2D HUDEP-2 cells, shown to be KMT2D-dependent (see Fig. 2 F), or designated as erythroid differentiation signature genes 41 (Fig. 4 H). Additionally, genes bound by KMT2D exhibited higher basal expression levels compared to those unbound by KMT2D, both in HUDEP-2 cells and in primary erythroblasts derived from primary HSPCs ( Supplementary Fig. 4F ). Thus, KMT2D and GATA1 co-occupy selective erythroid target genes, thereby enhancing their expression. KMT2D is required for enhancer activation of select GATA1 target genes We next asked whether KMT2D is required for GATA1-mediated enhancer activation during erythroid maturation by performing ChIP-seq analysis for H3K4me1, H3K27ac, and GATA1 in control and KMT2D KO HUDEP-2 cells. We utilized the GATA1 binding pattern and histone marks to annotate active enhancers (AEs: H3K4me1+, GATA1+, H3K27ac+) 42 , then separated the total annotated AEs into KMT2D-dependent (1,348) and KMT2D-independent AEs (9,643) based on the KMT2D binding signal (Fig. 5 A; Supplementary table 9 ). Importantly, both AE histone marks and GATA1 binding signals were stronger on KMT2D-dependent AEs relative to KMT2D-independent AEs. Depletion of KMT2D resulted in a significant reduction of the histone marks H3K4me1 and H3K27ac at KMT2D-dependent AEs, while the GATA1 signals did not change (Fig. 5 B). Compared with KMT2D-independent enhancers, KMT2D–GATA1 co-dependent enhancers are enriched in erythroid and myeloid differentiation pathways and harbor core erythroid-specific TF motifs, including those for GATA1 and KLF1( Supplementary Fig. 5A, B; Supplementary table 9 ). We next asked how enhancer activity is associated with gene expression. In AID-KMT2D HUDEP-2 cells treated with 5-Ph-IAA, a greater proportion of downregulated genes was associated with KMT2D-dependent AEs compared to upregulated genes or unchanged genes (Fig. 5 C). In addition, KMT2D-dependent AE-associated genes exhibited higher basal expression levels and were more likely to be suppressed by KMT2D depletion compared to genes with KMT2D-independent AEs in HUDEP-2 cells (Fig. 5 D). Examples of KMT2D and GATA1 co-dependent target genes include ZFPM1, SLC4A1 and EPOR , which were directly occupied by both GATA1 and KMT2D in their AE regions (Fig. 5 E, F; Supplementary Fig. 5C ). Similar to what we observed after disruption of endogenous KMT2D genes in HUDEP-2 cells, acute depletion of KMT2D in AID-KMT2D cells led to significant reductions of the active histone marks (H3K4me1 and H3K27ac), whereas GATA1 binding signals were stable (Fig. 5 G; Supplementary Fig. 5D, E ). Thus, KMT2D is essential for the activation of GATA1-dependent enhancers during erythropoiesis but is dispensable for its chromatin occupancy. KMT2D is required for the survival and maturation of primary human erythroblasts. To investigate further whether KMT2D is required for erythropoiesis, we transfected normal donor peripheral blood-mobilized CD34 + HSPCs with RNPs consisting of Cas9 and KMT2D -targeting or control non-targeting sgRNAs followed by expansion in stem cell/progenitor medium or in vitro differentiation toward erythroid or myeloid lineages (Fig. 6 A) 3 , 25 . Transfection of CD34 + cells with KMT2D -targeting RNP caused marked depletion of the corresponding protein compared to control cells (Fig. 6 B). KMT2D-disrupted HSPCs that were maintained in stem cell/progenitor medium expanded normally for up to seven days, with no dropout of KMT2D indels or apoptosis (Fig. 6 C, D; Supplementary Fig. 6A, B ). However, KMT2D suppression via Cas9 in CD34 + HSPCs caused significant reductions in erythroid, myeloid and mixed lineage colonies (Fig. 6 E). We further analyzed the effect on erythroid differentiation by suspension culture system. In contrast to findings in immature HUDEP-2 cells grown in expansion medium, disruption of KMT2D in CD34 + cells was associated with significant dropout of KMT2D indels and impaired cell expansion during in vitro erythroid differentiation (Fig. 6 F). Consistent with findings in HUDEP-2 cells, depletion of KMT2D caused impaired erythroid maturation as evidenced by reduced expression of CD235a and CD105 (Fig. 6 G, H), reduced enucleation, immature morphology, and reduced hemoglobinization (Fig. 6 I–K). Similarly, in vitro myeloid differentiation of KMT2D -disrupted CD34 + cells was associated with reduced cell expansion, indel dropout, and mildly reduced expression of the myeloid maturation marker CD11b ( Supplementary Fig. 6C, D ). To test the effects of KMT2D depletion using an orthologous approach, we transduced CD34 + HSPCs with a lentiviral vector encoding one of two different KMT2D shRNAs followed by in vitro erythroid maturation. Both targeting shRNAs caused approximately 40% reduction in KMT2D mRNA relative to luciferase shRNA control ( Supplementary Fig. 6E ), which resulted in reduced erythroid, myeloid and mixed lineage colony formation ( Supplementary Fig. 6F ) and impaired erythroid maturation, evidenced by characteristic cell surface markers, immature morphology and reduced hemoglobinization ( Supplementary Fig. 6G-I ). These results suggest that KMT2D is dispensable for CD34 + HSPCs but is required for the proliferation and/or survival of committed erythroid and myeloid lineages. We next investigated the dynamics of gene expression levels and chromatin accessibility signals associated with KMT2D-dependent and KMT2D-independent enhancers during in vitro erythroid differentiation of CD34 + HSPCs. Notably, the expression levels of genes associated with KMT2D-dependent AEs were consistently higher in erythroid progenitors of all maturation stages (Fig. 6 L). Moreover, KMT2D-dependent AEs exhibited significantly higher chromatin accessibility signals across the same developmental stages (Fig. 6 M). These results indicate that KMT2D stimulates the expression of erythroid genes by activating their enhancers during the differentiation of primary HSPCs. Discussion In this study, we identified KMT2D as a crucial cofactor for the master TF GATA1 in human erythropoiesis. Specifically, we demonstrated that KMT2D co-localizes with GATA1 on a subset of erythroid enhancers and is essential for their activation during terminal erythroid maturation. More generally, our findings reveal how a lineage-specific TF (GATA1) utilizes a general histone methyltransferase complex (KMT2D/COMPASS) to facilitate gene expression. We propose a model in which the pioneer TF GATA1 43 binds to distal enhancers during early erythropoiesis and recruits the KMT2D/COMPASS and CBP/p300 complexes, which deposit active histone marks H3K4me1 and H3K27ac, respectively, thereby driving transcription (Fig. 7 ). Supporting this model: 1) loss of KMT2D in erythroblasts impairs terminal maturation and erythroid gene expression, which resembles the loss of GATA1; 2) the genome-wide occupancy of KMT2D is highly enriched for the GATA motif and KMT2D co-occupies with GATA1 on thousands of erythroid enhancers that regulate highly expressed erythroid signature genes; 3) Induced KMT2D degradation resulted in enhancer inactivation with decreased H3K4me1 and H3K27ac marks, despite stable GATA1 binding; 4) GATA1 is physically associated with KMT2D/COMPASS complex through KDM6A. A key question is how to distinguish KMT2D-dependent enhancers from those that are KMT2D-independent. Our findings suggest that KMT2D-dependent enhancers are more likely to be erythroid-specific, as they are occupied by the erythroid transcription factor GATA1 and associated with high expression levels of their target genes. This aligns with recent studies showing that KMT2D is crucial for dynamic enhancer activity during neuronal differentiation 44 . Additionally, chromatin looping interactions may differentiate KMT2D-dependent and KMT2D-independent enhancers, given KMT2D’s role in facilitating chromatin looping 38 . Supporting this, KMT2D interactome data identified components of the Cohesin and Mediator complexes, which are essential for chromatin looping 38 . Future studies should investigate these features to elucidate the regulatory mechanisms of KMT2D-dependent enhancers, potentially uncovering additional complexities in the interplay between KMT2D and other regulatory factors to enhance our understanding of erythropoiesis. KMT2D lacks intrinsic DNA-binding ability, prompting the question of how it is recruited to enhancers or promoters. Typically, KMT2D is recruited by transcription factors (TFs). For example, KMT2D has been shown to colocalize with pluripotency TFs such as Oct4, Sox2, and Nanog on active enhancers in mouse ESCs through physical interactions 18 . Notably, recent studies suggest that KMT2D-dependent dynamic enhancers may be driven by GATA family members during embryonic stem cell differentiation 44 . During erythropoiesis, GATA1 binds the chromatin modifiers CBP/P300, the Med1 subunit of the Mediator complex, and the SWI/SNF catalytic subunit BRG1 to promote erythroid differentiation 45 – 48 . Recently, a study mapping human TF interaction networks identified KMT2D, KDM6A, and PAXI1 as interaction partners for both GATA1 and GATA2 using proximity-dependent biotinylation (BioID) 49 . Consistent with this, our KMT2D IP-mass dataset from K562 cells identified the known GATA1 cofactors, including several SWI/SNF and Mediator complex components and CBP/P300. More importantly, the GATA1 pulldown assay confirmed the presence of KDM6A, suggesting that GATA1 recruits the KMT2D/COMPASS complex to erythroid enhancers and promoters via KDM6A. Another possibility is that KMT2D/COMPASS is recruited via GATA2, which shares many binding sites with GATA1. During erythropoiesis, GATA1 displaces GATA2 from chromatin and represses GATA2 expression 50 , 51 . Future studies will be necessary to fully elucidate the detailed biochemical mechanisms between GATA1 and the KMT2D/COMPASS complex. Our study elucidated KMT2D’s role in human erythroid lineage differentiation and erythroid signature gene expression. A genetic screen in mice identified over 140 chromatin factors involved in myeloid and megakaryocytic-erythroid lineage specification, highlighting that Kmt2d and Kdm6a, but not Kmt2c, are crucial for myeloid progenitor identities and early myeloid priming 52 . In our study, an unbiased genetic screen targeting 496 human epigenetic modifiers underscored the importance of COMPASS-like components, including KMT2D, NCOA6, and KDM6A, in definitive erythroid maturation. Notably, KMT2D is essential for the survival of erythroid and myeloid progenitors but not hematopoietic stem and progenitor cells (HSPCs). Collectively, our data and previous reports suggest that KMT2D dynamically influences hematopoietic lineage specification, affecting both myelopoiesis and erythropoiesis. This role likely modulates the activity of various lineage-specific transcription factors. However, the precise regulatory mechanisms of KMT2D in hematopoiesis, particularly during transitions between myeloid lineages, warrant further investigation via in vivo models. KMT2D and its Drosophila homolog Trithorax-related are major H3K4 methyltransferases for H3K4me1 and H3K4me3 on enhancers during cell fate transitions in development and disease 53 – 56 . Using the KMT2D-AID system in human erythroid cell lines, we identified over 100 downregulated genes whose regulatory elements were directly occupied by both KMT2D and GATA1. A limitation of our study is that we could not determine the extent to which the observed transcriptional changes depended on KMT2D’s enzymatic activity, given its recently recognized noncatalytic functions 57 . Additionally, other H3K4 methyltransferases like KMT2B or KMT2C might compensate for KMT2D in some contexts 44 , which we could not fully rule out. Thus, comprehensive genetic studies involving triple or double knockouts, and catalytically inactive mutants of KMT2D, are required to better define its roles in erythropoiesis. In summary, we identified KMT2D as a crucial cofactor for GATA1 in human erythropoiesis. Our findings provide a foundation for further investigating KMT2D’s role and its clinical implications in hematopoiesis and blood cancers. Methods Culture and induced maturation of HUDEP-2 and K562 cells Immature HUDEP-2 cells were expanded in the StemSpan serum-free medium (SFEM; STEMCELL Technologies) supplemented with 1 µM dexamethasone, 1 µg/mL doxycycline, 50 ng/mL human stem cell factor (SCF), 3 units/mL erythropoietin (EPO), and 1% penicillin–streptomycin 31 . To induce erythroid maturation, HUDEP-2 cells were cultured in a differentiation medium composed of IMDM base medium (Invitrogen) supplemented with 2% FBS, 3% human serum albumin, 3 units/mL EPO, 10 µg/mL insulin, 1000 µg/mL holo-transferrin, and 3 units/mL heparin. Erythroid maturation was monitored by flow cytometry, using FITC-conjugated anti-CD235a (BD Biosciences, clone GA-R2), APC-conjugated anti-Band3 (from New York Blood Center), APC-conjugated anti-CD105 (BioLegend, clone SN6h) and Violet Blue–conjugated anti-CD49d (Miltenyi Biotec, clone MZ18-24A9) antibodies. Band3 high and Band3 low cell populations from the CD235a + cell fraction was purified by fluorescence-activated cell sorting (FACS). The human K562 cell line was maintained in RPMI-1640 medium (Hycita) containing 10% fetal bovine serum (FBS) (Hyclone), 2 mM glutamine (Sigma) and 1% penicillin/streptomycin (Thermo Fisher Scientific). To induce erythroid maturation of K562 cells, 10uM Hemin (HY-19424) can be added to the expansion medium. CD34 + cell culture and manipulation CD34 + hematopoietic stem and progenitor cells (HSPCs) were mobilized from normal subjects by granulocyte colony-stimulating factor, collected by apheresis, and enriched by immunomagnetic bead selection using an autoMACS Pro Separator (Miltenyi Biotec), in accordance with the manufacturer’s protocol. At least 95% purity was achieved, as assessed by flow cytometry using a PE-conjugated anti-human CD34 antibody (Miltenyi Biotec, clone AC136, catalog #130-081- 002). A two-phase culture protocol was used to promote erythroid differentiation and maturation 25 . In phase 1 (days 0–6), cells were cultured at a density of 2 * 10 5 cells/mL in SFEM with 10% FBS, 1% penicillin/streptomycin, 50 ng/mL SCF, 1.6U EPO and 10 ng/mL IL-3. In phase 2 (days 6–14), IL-3 and SCF was omitted from the medium and additional addition to 30% FBS. Erythroid differentiation and maturation were monitored by flow cytometry, using FITC-conjugated anti-CD235 (BD Biosciences, clone GA-R2, catalog #561017), APC-conjugated anti-CD105 (BioLegend, clone SN6h) antibodies. CRISPR library construction and screening A lentiviral vector library encoding 5,080 sgRNAs targeting 496 epigenetic modifier genes was purchased from Transomic Technologies (CAHV9002). Approximately 1 × 10 7 HUDEP-2 cells stably expressing Cas9 were transduced at a multiplicity of infection of ∼0.3 to minimize the introduction of > 1 vector particles and to achieve a 1000-fold library coverage. After 24 hours, transduced cells were selected in puromycin (1 µg/mL) for 2 days, then maintained in expansion medium for 5 days. Erythroid maturation was monitored by flow cytometry using fluorescein isothiocyanate–conjugated anti-CD235a (clone GA-R2; BD Biosciences) and allophycocyanin-conjugated anti-Band3 (gift from Xiuli An, New York Blood Center, New York, NY). Band3 high and Band3 low cell populations from the CD235a + cell fractions were purified by fluorescence-activated cell sorting. The representation of lentiviral vector-encoded sgRNAs in Band3 High and Band3 Low populations was determined by next-generation sequencing and compared. Data analysis of CRISPR screening The raw FASTQ data obtained after HiSeq sequencing were demultiplexed and mapped to the original reference sgRNA library for data analysis. The read counts for each sgRNA were normalized against the total read counts across all samples. The differentially enriched sgRNAs were defined by comparing normalized counts between sorted cells in the top 10% and those in the bottom 10% of Band3 low expressing bulk populations. Two independent screenings were performed with the HUDEP-2 cell line stably expressing Cas9. The sgRNA rank was displayed based on a P-value and log 2 fold change by DESeq2. The gene ranking analysis for significant genes and related GO (gene ontology) analysis were conducted using the MAGeCK, MAGeCK Flute, and EnrichR 33 . Generation of a GFP-P2A-miniAID-KMT2D endogenous knock-in cell lines For GFP-P2A-miniAID-KMT2D knock-in delivery, 500 ng of the donor plasmid and sgRNA/Cas9 ribonucleoprotein complexes (RNPs) were used for 2 × 10 6 cells. Both K562 and HUDEP-2 cells were electroporated with Neon NxT (Invitrogen MPK5000S). Seventy-two hours after electroporation, the cells were sorted for the expression of the GFP fluorescent marker to enrich the knock-in cell population. After the sorted cells recovered in culture for up to 3 weeks, a second round of sorting was performed to select GFP + cells for successful knock-in events, followed by clonal generation. CRISPR-Cas9-mediated gene targeting and vector construction The Cas9-expression vector, Lenti-Cas9-Blast, was purchased from Addgene (#52,962). Cas9 protein was introduced to human K562 and HUDEP-2 cell lines by lentiviral transduction and selected with 10 µg/mL blasticidin (Gibco, A11139-03) to generate Cas9-stable cell lines. The sgRNA sequences were selected from the CRISPR library or using a CRISPR design tool ( http://www.crisprscan.org/ ) and generated as oligonucleotide pairs. After annealing, constructs were cloned into the Lenti-guide-puro vector (a sgRNA-expression vector), which encodes the sgRNA. Cells were transduced with sgRNA lentivirus and puromycin (Gibco, A11138-03) selection for 48 h after transduction. The oligonucleotides encoding sgRNAs are listed in Supplementary table 10 . For the endogenous mini-AID-KMT2D (auxin-induced degron) system, the cDNA of KMT2D were cloned into the pGL3 vector using the Gibson Assembly. The KMT2D-N-terminal-mini-AID cassette was PCR amplified from the previous established pGL3 vector with AID-GFP fragment. For the GATA1 Tet-on system, the cDNA of GATA1 were cloned into the pGL3 vector using the Gibson Assembly. The Tet-on system cassette was PCR amplified from the previous established Tet-on system vector. SnapGene software was used to design all primers used for cloning. The PCR amplifications of products for cloning were performed using PrimeSTAR Max Premix (TaKaRa, R045). A Gibson Assembly Cloning Kit (Abclonal, RK21020) was used in accordance with the manufacturer’s instructions. ShRNA-mediated gene knockdown Oligonucleotides encoding short hairpin RNA (shRNA) constructs were designed by the RNAi Consortium of the Broad Institute, obtained from IDT, and cloned into the lentiviral vector pLKO.1-PURO, and luciferase shRNA-encoding pLKO.1-NT (shNT, non-targeting) was used as a control. Cells were transduced with shRNA lentivirus and then selected with 2 µg/ml puromycin for 48 h. The oligonucleotides encoding shRNAs are listed in Supplementary table 10 . Virus production and transduction Lentivirus was produced in HEK293T cells by transfecting lentiviral plasmids with helper packaging plasmids (VSVG and psPAX2) using the polyethyleneimine (PEI 40000; MAOKANGBIO, 49,553–93 − 7) transfection reagent. HEK293T cells were plated in 10-cm culture dishes and were transfected when confluency reached ~ 80–90%. For one 10-cm dish of HEK293T cells, 12 µg of plasmid DNA, 4 µg of pVSVG and 8 µg psPAX2, and 96 µL of 1 mg/mL PEI were mixed, incubated at room temperature for 20 min, and then added to the cells. The fresh medium was changed 6–8 h post-transfection. Lentivirus soup was collected at 48h and 72h post-transfection. The collected virus was filtered through a 0.45-µM non-pyrogenic filter. KMT2D editing in normal donors derived CD34 + HSPCs and HUDEP-2 cell CD34 + HSPCs were electroporated using the Neon™ NxT Electroporation (Neon) with the Neon™ Transfection Kit (Invitrogen by Thermo Fisher Scientific, MPK1096B). The electroporation program is 1600 V, 10 ms, and 3 pulse. HUDEP-2 were electroporated using electroporation program is 1200 V, 40 ms, and 1 pulse. For ribonucleoprotein (RNP) complex delivery, 50 µM of KMT2D sgRNA#1/2 (synthesized by GenScript) and 5 µg of Cas9 protein (Integrated DNA Technologies, 1,081,058) (molar ratio of sgRNA:Cas9 is 1:3) were used for 0.2 million CD34 + HSPCs. Nontarget sgRNA (NT) was used as a negative control. After 24 h of electroporation, cells were washed using PBS and transferred to a fresh growth medium. After 48 h of electroporation, the cells will be used for functional experiments and knockout efficiency assessment. Quantitative PCR (qPCR) analysis for gene expression Total RNA was extracted from K562 and HUDEP-2 cells using the FastPure Cell/Tissue Total RNA Isolation Kit (Vazyme, RC112-01). cDNA was generated using PrimeScript™ RT Master Mix (Takara, RR036Q) from 1 µg RNA and diluted 1:200 for qPCR analysis. qPCR was performed using 1 µL diluted cDNA with biological and technical replicates using SYBR Green Master Mix (Vazyme, Q511-03) with QuantStudio 6 real-time PCR system, and results were normalized to the expression of ACTB. Primer sequences utilized for qPCR are in Supplementary table 10 . Immunoblotting Cells were washed with PBS and lysed in Cell Lysis Buffer for RIPA (Beyotime, # P0013B) with a protease inhibitor cocktail. Lysates were heated to 95°C in SDS sample buffer, separated by SDS-PAGE, and transferred to a nitrocellulose membrane. Membranes were blocked in 5% nonfat milk in PBS with 0.1% Tween-20, probed with indicated primary antibodies, and followed by incubation with HRP-coupled secondary antibody for 1 h at room temperature. Blots were visualized using enhanced chemiluminescence detection reagents and exposed to X-ray film. All antibodies used in this study are listed in Supplementary table 10 . Nucleoprotein extraction and identification of KMT2D-interacting proteins Collect 20M cells from each group of samples and use the nuclear and cytoplasmic extraction kit (Abbkine, #KTP3001) to collect nuclear extracts. Briefly, cells were harvested and then centrifuged at 500 g for 5 min. After washing with ice-cold 1× PBS, 200 µL of ice-cold hypotonic buffer [10 mM HEPES, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol (DTT), 1 mM EDTA, pH 7.9] containing 1× protease inhibitor cocktail (Sigma-Aldrich, P8849) was added to burst the cell pellet, and incubation was carried out on ice for 10 min. Then, 11 µL of ice-cold detergent (0.05% NP-40) was added, and the incubation was prolonged on ice for an additional 1 min. After centrifugation (5 min at 16,000 g), the supernatant (cytoplasmic extract) was transferred to a pre-chilled tube while the pellet fraction containing the nuclei was suspended in 100 µL of ice-cold nuclear extraction buffer (5 mM HEPES, 1.5 mM MgCl 2 , 300 mM NaCl, 0.2 mM EDTA, 0.5 mM DTT, 26% glycerol, pH 7.9) containing 1× protease inhibitor cocktail. The samples were placed on ice and vortexed for 15 s every 10 min for a total of 40 min. After centrifugation (10 min at 16, 000 g), the supernatant (nuclear extract) was collected into a pre-chilled tube. The collected samples were sent to PTM Biolabs Inc for proteomic sequencing and WB validation. RNA-seq sample preparation and analysis RNA samples of two biological replicates were extracted from cultured cells, using TRIzol (Ambion by Life technologies, 343,911), following the manufacturer’s instructions. RNA was then sent out for library preparation and next-generation sequencing to a commercial company, Novogene (CA, USA). Raw counts of gene transcripts were derived from raw fastq files using the alignment-independent quantification tool, Salmon ( https://combine- lab.github.io/salmon/) with standard settings. The raw count matrix was then imported into RStudio and utilized as input for Limma-voom analysis following the vignette of the package for normalization, differential gene expression analysis, and unbiased clustering analysis, including principal component analysis. The output of Limma-voom was used as the input for pre-ranked-based GSEA to enrich functional pathways and gene signatures. The RNA-seq datasets during primary erythropoiesis from the normal CD34 + HSPCs were available under accession GSE53983 58 . ChIP-seq sample preparation and data analysis ChIP experiments were performed as previously described with at least two biological replicates for each study 59 . First, 2 × 10 7 HUDEP-2 cells were suspended in 50 mL of PBS and processed according to previous detailed method. Two percent of the mixture was set aside as input. For each ChIP, chromatin from 2 × 10 7 cells was mixed with 20 ng spike-in chromatin (Active Motif, 53083) and incubated with 2–8 µg primary target antibody and 2 µg spike-in antibody (Active Motif, 61686) overnight at 4°C. For ChIP–seq data analysis, raw sequencing data were aligned to the human genome GRCh37 and the drosophila genome dm6 using Bowtie2 60 . For ChIP–seq of KMT2D, the window size of 50 bp, the gap size of 50 bp and the false discovery rate (FDR) threshold of 0.05 were used by SICER 61 . For ChIP–seq of histone modifications (H3K4me1 and H3K27ac), the window size of 200 bp, the gap size of 200 bp and the FDR threshold of 10 − 3 were used. Reads on indicated regions were collected to calculate reads per kilobase million as a measure of signal intensity. ATAC-seq sample preparation and data analysis The ATAC-seq library was prepared according to the published omni-ATAC protocol 62 . For the HUDEP-2 cells, 50,000 live cells were used per sample. After centrifugation at 500 rpm for 5 min at 4°C (Eppendorf 5417R refrigerated centrifuge), the cell pellets were resuspended in cold lysis buffer supplemented with protease inhibitors (10 mM Tris with a pH 7.4, 10 mM NaCl, 3 mM MgCl 2 , and 0.1% IGEPAL), followed by centrifugation. The pellets were resuspended in 25 µL of tagment DNA buffer (Nextera, FC-121–1030) and used directly in the transposition reaction. Nextera Tn5 (Nextera, FC-121–1030) was added to the resuspended nuclei, and the transposition reaction mixture was incubated at 37°C for 30 min. After transposition, the DNA was purified using a Qiagen MinElute PCR purification kit (Qiagen, 28,004). Indexing PCR was conducted for 12 cycles using the NEB Next HiFi 2X PCR Master Mix (NEB, M0541S) and indexing primers. The PCR products were purified at a 1:3 ratio of Agencourt AMPure XP beads (Beckman Coulter, A63881). Libraries were paired-end 100-bp sequenced using an Illumina HiSeq 4000 system. The ATAC-seq datasets were analyzed via the previous methods 35 . The publicly available ATAC-seq datasets during primary erythropoiesis from the normal CD34 + HSPCs were obtained from the GSE128266 63 . First, DeepTools (v3.5.6) was used to plot the average heatmap of peak signals in 10-bp bins. Subsequently, boxplot analysis was conducted using R. The P value was determined by the Wilcoxon test. Declarations Competing interests M.J.W. is a consultant for Glaxo SmithKline, Cellarity Inc, Graphite Bio, Fulcrum Therapeutics and Dyne Therapeutics, and owns equity in Cellarity Inc. Dr. Li Cheng is currently affiliated with a commercial company (GenAssist Therapeutics Incorporation, Suzhou) and declares that she has no competing interests. Dr. Chunliang Li is currently an editorial board member at Genome Biology. The other authors declare that they have no other competing interests. Author contributions P.X., and J.-X. Z. planned the experimental design, analyzed data, draw the Fig. s and wrote the manuscript; J.-X. Z., Y. X. and L.C. performed RNA-seq, ATAC-Seq, and ChIP-seq with the help from Y. X., and M. -L.Z.; J.-X. Z. established the AID2 model for KMT2D; Y. X., L.C. performed bioinformatics analyses under the guidance of B.-S.X., C.-L. L., and P.X.; R. -P. F., X.-H. Q. helped to set up the KMT2D ChIP-seq under the guidance of H.-M. H., Y. C., and M.J.W.. The whole project administration, supervision and funding acquisition: P.X.. The authors have read, discussed the results and approved the final manuscript. Acknowledgments Ryo Kurita and Yukio Nakamura (Cell Engineering Division, RIKEN BioResource Center, Tsukuba, Japan) provided HUDEP-2 cells. Xiuli An (Laboratory of Membrane Biology, New York Blood Center) provided the anti-Band3 antibody. We thank the insightful discussion and comments of members from the Xu lab. This research was supported by the National Natural Science Foundation of China (82170119), by Jiangsu Province National Science and Technology grant BK20243008, by the High-Level Personnel Project of Jiangsu Province (JSSCTD202353), by Interdisciplinary Basic Frontier Innovation Program of Suzhou Medical College of Soochow University, The Pediatric Hematology& Oncology Key Laboratory of Higher Education Institutions in Jiangsu Province, by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) and Collaborative Innovation Center of Hematology (all to P.X.). Data availability All data are available in the main text or the Supplementary materials. The RNA-seq and ChIP-seq raw datasets generated from this study were deposited to GEO under the accession number: GSE293567 (reviewer token: sjmtooiwlbepbej) 64 . The H3K4me1/3, H3K27ac and GATA1 ChIP-seq datasets in HUDEP-2 cells were obtained from the GSE115357 3 , while the TAL1 and KLF1 ChIP-seq datasets in HUDEP-2 cells were obtained from the GSE157311 40 . The ATAC-seq and RNA-seq datasets during primary erythropoiesis from the normal CD34 + HSPCs were obtained from the GSE128266 63 and GSE53983 58 , respectively. Code repositories collected at https://doi.org/10.6084/m9.figshare.c.6186670 65 . References Cramer P (2019) Organization and regulation of gene transcription. Nature 573:45–54 Stadhouders R, Filion GJ, Graf T (2019) Transcription factors and 3D genome conformation in cell-fate decisions. Nature 569:345–354 Xu P et al (2021) FBXO11-mediated proteolysis of BAHD1 relieves PRC2-dependent transcriptional repression in erythropoiesis. Blood 137:155–167 Karayel Ö et al (2020) Integrative proteomics reveals principles of dynamic phosphosignaling networks in human erythropoiesis. Mol Syst Biol 16:e9813 An X, Schulz VP, Mohandas N, Gallagher PG (2015) Human and murine erythropoiesis. Curr Opin Hematol 22:206–211 Nandakumar SK, Ulirsch JC, Sankaran V (2016) G. Advances in understanding erythropoiesis: evolving perspectives. Br J Haematol 173:206–218 Sankaran VG, Weiss MJ (2015) Anemia: progress in molecular mechanisms and therapies. Nat Med 21:221–230 Pop R et al (2010) A Key Commitment Step in Erythropoiesis Is Synchronized with the Cell Cycle Clock through Mutual Inhibition between PU.1 and S-Phase Progression. PLoS Biol 8:e1000484 Nguyen AT et al (2017) UBE2O remodels the proteome during terminal erythroid differentiation. Sci (80-). 357 Wadman IA et al (1997) The LIM-only protein Lmo2 is a bridging molecule assembling an erythroid, DNA-binding complex which includes the TAL1, E47, GATA-1 and Ldb1/NLI proteins. EMBO J 16:3145–3157 Tallack MR et al (2010) A global role for KLF1 in erythropoiesis revealed by ChIP-seq in primary erythroid cells. Genome Res 20:1052–1063 Orkin SH (1995) Transcription factors and hematopoietic development. J Biol Chem 270:4955–4958 Cheng Y et al (2009) Erythroid GATA1 function revealed by genome-wide analysis of transcription factor occupancy, histone modifications, and mRNA expression. Genome Res 19:2172–2184 Oudelaar AM, Higgs DR (2021) The relationship between genome structure and function. Nat Rev Genet 22:154–168 Murphy ZC et al (2021) Regulation of RNA polymerase II activity is essential for terminal erythroid maturation. Blood 138:1740–1756 Heintzman ND et al (2009) Histone modifications at human enhancers reflect global cell-type-specific gene expression. Nature 459:108–112 Miller T et al (2001) COMPASS: A complex of proteins associated with a trithorax-related SET domain protein. Proc. Natl. Acad. Sci. U. S. A. 98, 12902–12907 Wang C et al (2016) Enhancer priming by H3K4 methyltransferase MLL4 controls cell fate transition. Proc. Natl. Acad. Sci. 113, 11871–11876 Ge K (2019) Enhancer regulation by H3K4 methyltransferases MLL3/MLL4. FASEB J 33 Wang SP et al (2017) A UTX-MLL4-p300 Transcriptional Regulatory Network Coordinately Shapes Active Enhancer Landscapes for Eliciting Transcription. Mol Cell 67:308–321e6 Boileau RM, Chen KX, Blelloch R (2023) Loss of MLL3/4 decouples enhancer H3K4 monomethylation, H3K27 acetylation, and gene activation during embryonic stem cell differentiation. Genome Biol 24:41 Wang Z, Ren B (2024) Role of H3K4 monomethylation in gene regulation. Curr Opin Genet Dev 84:102153 Morgan MAJ, Shilatifard A (2020) Reevaluating the roles of histone-modifying enzymes and their associated chromatin modifications in transcriptional regulation. Nat Genet 52:1271–1281 Cenik BK, Shilatifard A (2021) COMPASS and SWI/SNF complexes in development and disease. Nat Rev Genet 22:38–58 Yan H et al (2017) Distinct roles for TET family proteins in regulating human erythropoiesis. Blood 129:2002–2012 Wang Y et al (2021) Impairment of human terminal erythroid differentiation by histone deacetylase 5 deficiency. Blood 138:1615–1627 Li M et al (2023) Stage-specific dual function: EZH2 regulates human erythropoiesis by eliciting histone and non-histone methylation. Haematologica 108:2487–2502 Malik J, Getman M, Steiner LA (2015) Histone Methyltransferase Setd8 Represses Gata2 Expression and Regulates Erythroid Maturation. Mol Cell Biol 35:2059–2072 Malik J, Lillis JA, Couch T, Getman M, Steiner LA (2017) The Methyltransferase Setd8 Is Essential for Erythroblast Survival and Maturation. Cell Rep 21:2376–2383 Myers JA et al (2020) The histone methyltransferase Setd8 alters the chromatin landscape and regulates the expression of key transcription factors during erythroid differentiation. Epigenetics Chromatin 13 Kurita R et al (2013) Establishment of immortalized human erythroid progenitor cell lines able to produce enucleated red blood cells. PLoS ONE 8:e59890 Andersson LC, Jokinen M, Gahmberg CG (1979) Induction of erythroid differentiation in the human leukaemia cell line K562. Nature 278:364–365 Kuleshov MV et al (2016) Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 44:W90–W97 Matys V (2006) TRANSFAC(R) and its module TRANSCompel(R): transcriptional gene regulation in eukaryotes. Nucleic Acids Res 34:D108–D110 Hyle J et al (2023) Auxin-inducible degron 2 system deciphers functions of CTCF domains in transcriptional regulation. Genome Biol 24:1–30 Yesbolatova A et al (2020) The auxin-inducible degron 2 technology provides sharp degradation control in yeast, mammalian cells, and mice. Nat Commun 11:5701 Lee J-E et al (2017) Brd4 binds to active enhancers to control cell identity gene induction in adipogenesis and myogenesis. Nat Commun 8:2217 Yan J et al (2018) Histone H3 lysine 4 monomethylation modulates long-range chromatin interactions at enhancers. Cell Res 28:204–220 Nabet B et al (2018) The dTAG system for immediate and target-specific protein degradation. Nat Chem Biol 14:431–441 Cheng L et al (2021) Single-nucleotide-level mapping of DNA regulatory elements that control fetal hemoglobin expression. Nat Genet 53:869–880 Giani FC et al (2016) Targeted Application of Human Genetic Variation Can Improve Red Blood Cell Production from Stem Cells. Cell Stem Cell 18:73–78 Dorighi KM et al (2017) Mll3 and Mll4 Facilitate Enhancer RNA Synthesis and Transcription from Promoters Independently of H3K4 Monomethylation. Mol Cell 66:568–576e4 Kadauke S et al (2012) Tissue-specific mitotic bookmarking by hematopoietic transcription factor GATA1. Cell 150:725–737 Kubo N et al (2024) H3K4me1 facilitates promoter-enhancer interactions and gene activation during embryonic stem cell differentiation. Mol Cell 84:1742–1752e5 Boyes J, Byfield P, Nakatani Y, Ogryzko V (1998) Regulation of activity of the transcription factor GATA-1 by acetylation. Nature 396:594–598 Kim S, Il, Bultman SJ, Kiefer CM, Dean A, Bresnick EH (2009) BRG1 requirement for long-range interaction of a locus control region with a downstream promoter. Proc. Natl. Acad. Sci. U. S. A. 106 Letting DL, Rakowski C, Weiss MJ, Blobel GA (2003) Formation of a Tissue-Specific Histone Acetylation Pattern by the Hematopoietic Transcription Factor GATA-1. Mol Cell Biol 23 Stumpf M et al (2006) The mediator complex functions as a coactivator for GATA-1 in erythropoiesis via subunit Med1/TRAP220. Proc. Natl. Acad. Sci. U. S. A. 103 Göös H et al (2022) Human transcription factor protein interaction networks. Nat Commun 13:766 Grass JA et al (2003) GATA-1-dependent transcriptional repression of GATA-2 via disruption of positive autoregulation and domain-wide chromatin remodeling. Proc. Natl. Acad. Sci. U. S. A. 100, 8811–8816 Bresnick EH, Lee H-YY, Fujiwara T, Johnson KD, Keles S (2010) GATA switches as developmental drivers. J Biol Chem 285:31087–31093 Lara-Astiaso D et al (2023) In vivo screening characterizes chromatin factor functions during normal and malignant hematopoiesis. Nat Genet 55:1542–1554 Herz HM et al (2012) Enhancer-associated H3K4 monomethylation by trithorax-related, the drosophila homolog of mammalian MLL3/MLL4. Genes Dev 26:2604–2620 Hu D et al (2013) The MLL3/MLL4 Branches of the COMPASS Family Function as Major Histone H3K4 Monomethylases at Enhancers. Mol Cell Biol 33:4745–4754 Dhar SS et al (2012) Trans-tail regulation of MLL4-catalyzed H3K4 methylation by H4R3 symmetric dimethylation is mediated by a tandem PHD of MLL4. Genes Dev 26:2749–2762 Dhar SS et al (2018) MLL4 Is Required to Maintain Broad H3K4me3 Peaks and Super-Enhancers at Tumor Suppressor Genes. Mol Cell 70:825–841 Van HT, Xie G, Dong P, Liu Z, Ge K (2024) KMT2 Family of H3K4 Methyltransferases: Enzymatic Activity-dependent and -independent Functions. J Mol Biol 436:168453 An X et al (2014) Global transcriptome analyses of human and murine terminal erythroid differentiation. Blood 123:3466–3477 Landt SG et al (2012) ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia. Genome Res 22:1813–1831 Langmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9 Zang C et al (2009) A clustering approach for identification of enriched domains from histone modification ChIP-Seq data. Bioinformatics 25:1952–1958 Corces MR et al (2017) An improved ATAC-seq protocol reduces background and enables interrogation of frozen tissues. Nat Methods 14:959–962 Schulz VP et al (2019) A Unique Epigenomic Landscape Defines Human Erythropoiesis. Cell Rep 28:2996–3009e7 Zhang J, Xin Y, Cheng L, Yang X, Xing Y, Xu B, Li C (2025) X. P. The H3K4 methyltransferase KMT2D is an essential cofactor for GATA1 at erythroid gene enhancers. https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE293567 https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE293567 (2025) Xu B, Djekidel MN, Li C (2022) W. J. St Jude Center for Applied Bioinformatics General Pipelines. figshare. Collection. https://doi.org/10.6084/m9.figshare.c.6186670 Additional Declarations Yes there is potential Competing Interest. M.J.W. is a consultant for Glaxo SmithKline, Cellarity Inc, Graphite Bio, Fulcrum Therapeutics and Dyne Therapeutics, and owns equity in Cellarity Inc. Dr. Li Cheng is currently affiliated with a commercial company (GenAssist Therapeutics Incorporation, Suzhou) and declares that she has no competing interests. Dr. Chunliang Li is currently an editorial board member at Genome Biology. The other authors declare that they have no other competing interests. Supplementary Files Figure1S20250925updated.pdf Supplementary Fig.1 TableS1.xlsx Supplementary Table 1 TableS3.xlsx Supplementary Table 3 Figure6S20250926updated.pdf Supplementary Fig.6 Figure2S20250928updated.pdf Supplementary Fig.2 TableS7.xlsx Supplementary Table 7 TableS8.xlsx Supplementary Table 8 SupplementalFigurelegends20251013updated.pdf Supplemental Figure legends Figure4S20251013updated.pdf Supplementary Fig.4 TableS9.xlsx Supplementary Table 9 TableS4.xlsx Supplementary Table 4 TableS2.xlsx Supplementary Table 2 TableS6.xlsx Supplementary Table 6 TableS10.xlsx Supplementary Table 10 Figure5S20250925updated.pdf Supplementary Fig.5 TableS5.xlsx Supplementary Table 5 Figure3S20251013updated.pdf Supplementary Fig.3 Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7853319","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":532894970,"identity":"a6d0f2ed-9d3f-43c8-9dac-e08c1926918f","order_by":0,"name":"Peng Xu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0UlEQVRIiWNgGAWjYPACGyBOAGI24rWkka7lMAlazNl7D7/mbTtvb3A8+QHDh7LDDPyzG/Brsew5l2Y5s+124oYzzwwYZ5w7zCBx5wB+LQY3cswMPrbdTjC4kWDAzNt2mMFAIoGAlvtvzAwS287ZG9xI/8D8lygtN3iMH3xsO8C44UaOATMjMVose3LMgF5ITpx55k3BwZ5z6TwSNwhoMWc/Y/yZp8zOnu94+sYHP8qs5fhnEHIYMC4kYJwDQMyDXz1EC/MHgqpGwSgYBaNgZAMA2hFGtlkxTqQAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-7573-4813","institution":"Soochow University","correspondingAuthor":true,"prefix":"","firstName":"Peng","middleName":"","lastName":"Xu","suffix":""},{"id":532894971,"identity":"e883ca29-5f9d-45bd-a7e5-f0df195e636b","order_by":1,"name":"Jianxiang Zhang","email":"","orcid":"","institution":"Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Jianxiang","middleName":"","lastName":"Zhang","suffix":""},{"id":532894972,"identity":"b422d90c-702a-46ef-a526-3ed80cd50a1e","order_by":2,"name":"Ye Xin","email":"","orcid":"","institution":"Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Ye","middleName":"","lastName":"Xin","suffix":""},{"id":532894973,"identity":"d6c81272-cbae-4b12-8f44-de321c99ca10","order_by":3,"name":"Li Cheng","email":"","orcid":"","institution":"St. Jude Children's Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Cheng","suffix":""},{"id":532894974,"identity":"9af286f6-b621-41f0-8a28-81bdecd180fe","order_by":4,"name":"Yuan Xing","email":"","orcid":"","institution":"Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Yuan","middleName":"","lastName":"Xing","suffix":""},{"id":532894975,"identity":"2d4ae54d-9aab-49a6-9448-5aaed16c90f8","order_by":5,"name":"Mengli Zhang","email":"","orcid":"","institution":"Soochow University","correspondingAuthor":false,"prefix":"","firstName":"Mengli","middleName":"","lastName":"Zhang","suffix":""},{"id":532894976,"identity":"7f7cc771-3401-4b43-b688-136d566ed91d","order_by":6,"name":"Ruopeng Feng","email":"","orcid":"","institution":"St. Jude Children's Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ruopeng","middleName":"","lastName":"Feng","suffix":""},{"id":532894977,"identity":"3cb9ee7b-da41-43ab-a616-3bd280b8c43b","order_by":7,"name":"Xiaohui Qiu","email":"","orcid":"https://orcid.org/0000-0002-5767-0422","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Xiaohui","middleName":"","lastName":"Qiu","suffix":""},{"id":532894978,"identity":"ab68dc8a-b06e-4619-a5f6-9904dccc3e0f","order_by":8,"name":"Beisi Xu","email":"","orcid":"https://orcid.org/0000-0003-0099-858X","institution":"St Jude Children's Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"Beisi","middleName":"","lastName":"Xu","suffix":""},{"id":532894979,"identity":"4d6b129a-a01f-4634-bd10-51f303fe5c75","order_by":9,"name":"Yong Cheng","email":"","orcid":"https://orcid.org/0000-0002-9314-0971","institution":"St. Jude Children's Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yong","middleName":"","lastName":"Cheng","suffix":""},{"id":532894980,"identity":"b6d64f91-846e-4368-9348-c3918f676a60","order_by":10,"name":"CHUNLIANG LI","email":"","orcid":"https://orcid.org/0000-0002-5938-5510","institution":"St Jude Children's Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"CHUNLIANG","middleName":"","lastName":"LI","suffix":""},{"id":532894981,"identity":"074d7c88-aa6e-419a-bf9b-16a87491ae22","order_by":11,"name":"Hans-Martin Herz","email":"","orcid":"https://orcid.org/0000-0003-4780-9176","institution":"St. Jude Children's Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"Hans-Martin","middleName":"","lastName":"Herz","suffix":""},{"id":532894982,"identity":"efa41f36-f836-4c93-96e1-418bf851668f","order_by":12,"name":"Mitchell Weiss","email":"","orcid":"https://orcid.org/0000-0002-6734-4858","institution":"St. Jude Children's Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"Mitchell","middleName":"","lastName":"Weiss","suffix":""}],"badges":[],"createdAt":"2025-10-14 02:50:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7853319/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7853319/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94097417,"identity":"a17cc0d0-c36a-429f-aa0a-2aaa60e7fe54","added_by":"auto","created_at":"2025-10-22 10:15:12","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":194073,"visible":true,"origin":"","legend":"","description":"","filename":"Maintext20251014updated.docx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/ffa86fab2f82867d3e5fea30.docx"},{"id":94096599,"identity":"34b23511-d3b9-4c72-9dd6-74614d341b63","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"json","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13561,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS2581973.json","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/3a468c6fb952d3bbd3ae3064.json"},{"id":94096607,"identity":"bba89f8a-b5bc-4cca-9129-c517d2ba3484","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157805,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS25819730enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/cf4f583a3fb11ed89fa14245.xml"},{"id":94096606,"identity":"0c33d1a5-1e4b-4ca3-a32d-6a158f41bfd1","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"pdf","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3158753,"visible":true,"origin":"","legend":"","description":"","filename":"Figure120251001updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/6e3f8caf806ed434878b97d6.pdf"},{"id":94097420,"identity":"65f07208-2102-4a7a-a2e6-61477c99a8f3","added_by":"auto","created_at":"2025-10-22 10:15:12","extension":"pdf","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1775801,"visible":true,"origin":"","legend":"","description":"","filename":"Figure220250912updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/a3a94c34712f523913275b3d.pdf"},{"id":94097568,"identity":"21c3f8f2-ace2-49f7-93cd-c26a9b6e1e24","added_by":"auto","created_at":"2025-10-22 10:23:12","extension":"pdf","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1971345,"visible":true,"origin":"","legend":"","description":"","filename":"Figure320251013updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/613dcf9fb3942b6d130cbb4f.pdf"},{"id":94096614,"identity":"fc577575-257f-4951-8685-b44cbd94a474","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"pdf","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2798023,"visible":true,"origin":"","legend":"","description":"","filename":"Figure420251008updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/d4de23d6e35f1bf11bda4b47.pdf"},{"id":94096618,"identity":"dc8bdf87-1b93-4407-83da-d6d36e068472","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"pdf","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2877810,"visible":true,"origin":"","legend":"","description":"","filename":"Figure520250925updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/d951999f877e3432e7f5629a.pdf"},{"id":94096615,"identity":"d04566ab-aa2d-4c17-9e70-75717bd073fc","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1661693,"visible":true,"origin":"","legend":"","description":"","filename":"Figure620251001updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/d69d74296e4dea6191edb662.pdf"},{"id":94096613,"identity":"573ae6ff-68e5-4d48-9ef4-363297f7de7d","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"pdf","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":538941,"visible":true,"origin":"","legend":"","description":"","filename":"Figure720251013updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/0352a9b27f108d41572b9678.pdf"},{"id":94098234,"identity":"5ecd0908-9bae-44bf-a4aa-ce5b57dedb78","added_by":"auto","created_at":"2025-10-22 10:31:12","extension":"xml","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":156554,"visible":true,"origin":"","legend":"","description":"","filename":"NCOMMS25819730structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/68a3b4b0d72ca7198838cd6b.xml"},{"id":94096622,"identity":"fa08937d-b209-49a6-8e6d-3b5fb06d8372","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"html","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":170531,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/369c4a469f84af89e2a276d1.html"},{"id":94096588,"identity":"6082dadc-5d91-4b53-a67d-21cf5c57c2e5","added_by":"auto","created_at":"2025-10-22 10:07:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":228122,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAn unbiased CRISPR/Cas9 screen for epigenetic regulators identifies KMT2D as a critical regulator of erythropoiesis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e. The CRISPR/sgRNA screening strategy. The erythroid cell line HUDEP-2 was transfected with lentiviral vectors encoding Cas9 and 5,082 single guide RNAs (sgRNAs) targeting 496 human epigenetic modifier genes. After five days of induced differentiation, the effect of individual sgRNAs on erythroid maturation was assessed by immuno-flow cytometry for cells expressing the 10% highest and lowest levels of the maturation marker Band3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e. Rank plot of sgRNA enrichment in Band3+ and Band3- cells, calculated as Log2[Fold Change (Band3\u003csup\u003eHigh\u003c/sup\u003e/Band3\u003csup\u003eLow\u003c/sup\u003e)].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e. Components of the KMT2D- and SETD1B-COMPASS H3K4 methyltransferase complexes showing shared (WDR5, RBBP5, ASH2L and DPY30) and complex-specific subunits.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e. Heat map showing the mRNA levels of KMT2C/KMT2D and SETD1A/SETD1B complex subunits in HSCs and defined erythroid progenitors of increasing maturation stage during \u003cem\u003ein vitro\u003c/em\u003e differentiation of peripheral blood-derived CD34\u003csup\u003e+\u003c/sup\u003e hematopoietic stem and progenitor cells (HSPCs) according to data from a published study (GSE53983)\u003csup\u003e58\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e. Western blot showing KMT2D, KDM6A (UTX) and ACTINB protein expression in nuclear extracts of HUDEP-2 cells transfected with lentiviral vectors encoding Cas9 and control (Ctrl, non-targeting) or KMT2D-targeting sgRNAs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e. Representative flow cytometry plots and graphical summary of CD49d and Band3 expression in KMT2D-disrupted HUDEP-2 cells, as described in panel \u003cstrong\u003eE\u003c/strong\u003e. Cells were analyzed after 5 days of induced erythroid maturation. *P \u0026lt; 0.05, **P \u0026lt; 0.01, unpaired Student’s t-test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e. Cell pellets showing hemoglobinization of HUDEP-2 cells described in panels \u003cstrong\u003eE \u003c/strong\u003eand \u003cstrong\u003eF\u003c/strong\u003e after five days of induced erythroid maturation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e. RT-qPCR (real-time quantitative PCR) of HUDEP-2 cells treated with the indicated sgRNAs showing mRNA expression levels of erythroid marker genes \u003cem\u003eGYPA, SLC4A1, HBA\u003c/em\u003e and\u003cem\u003e HBB \u003c/em\u003eon day five of induced maturation. Expression levels are normalized to β-Actin mRNA. Data are displayed as the mean ± standard error of the mean (SEM) of three independent replicates. ****P \u0026lt; 0.0001, unpaired Student’s t-test.\u003c/p\u003e","description":"","filename":"Figure120251001updated1.png","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/ef02d6e8703ba92b20b6bd59.png"},{"id":94097414,"identity":"a345c30a-e2aa-458e-915d-c3cfdc6836d9","added_by":"auto","created_at":"2025-10-22 10:15:12","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":171739,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKMT2D is required for activation of erythroid signature genes and proper chromatin accessibility of GATA1.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e. Experimental strategy for RNA-seq analysis. HUDEP-2 cells were transduced with lentiviral vectors expressing Cas9 and Ctrl sgRNA or KMT2D sgRNA1 (sg2), subsequently referred to as KMT2D KO cells. RNA was purified from cells grown in expansion medium (day 0) and three days after induced erythroid maturation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e. MA plot\u003cstrong\u003e \u003c/strong\u003eshowing differentially expressed genes (DEGs) in KMT2D KO versus control HUDEP-2 cells in mature and immature states. Red and blue dots designate upregulated and downregulated mRNAs, respectively, in immature and mature KMT2D KO versus. control cells (fold change \u0026gt; 1.5, FDR \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e.\u003cstrong\u003e \u003c/strong\u003eVenn diagrams depicting the\u003cstrong\u003e \u003c/strong\u003eoverlap of up- and downregulated genes in the mature and immature states of KMT2D KO versus control HUDEP-2 cells.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e. Gene set enrichment analysis (GSEA) of RNA-seq data showing enrichment for erythroid differentiation markers in immature and mature KMT2D KO versus control HUDEP-2 cells. Erythroid differentiation signature gene sets are from a previous study\u003csup\u003e41\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e. MA plot\u003cstrong\u003e \u003c/strong\u003eshowing transcriptional changes in control HUDEP-2 cells in the mature versus immature state (fold change \u0026gt; 1.5, FDR \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e. Pie charts displaying the numbers of up- and downregulated genes in control HUDEP-2 cells in the mature versus immature state from the total number of expressed genes and the number of KMT2D-dependent genes in the upregulated category.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e. Gene ontology (GO) analysis of the KMT2D-dependent genes defined in\u0026nbsp;panel\u003cstrong\u003e F\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e. Transcription factor (TF) enrichment analysis of the KMT2D-dependent genes defined in\u0026nbsp;panel \u003cstrong\u003eF\u003c/strong\u003e. The relative P-value was calculated using enrichR online software.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(I)\u003c/strong\u003e. Heatmap showing significantly downregulated erythroid signature genes in KMT2D KO versus control HUDEP-2 cells in the immature and mature states.\u003c/p\u003e","description":"","filename":"Figure120251001updated2.png","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/fae93587cf6d32d7e9c9243e.png"},{"id":94096589,"identity":"2aec2b0b-17c4-4dc9-97d3-f3a8d59d5fc8","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":258084,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAcute depletion of KMT2D selectively suppresses the expression of GATA1 targets.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e. Diagram showing the auxin-inducible degron 2 (AID2) system for acute depletion of KMT2D in HUDEP-2 and K562 cells. An in-frame EGFP-AID tag was inserted at the N-terminus of endogenous KMT2D gene in cells constitutively expressing OsTIR1\u003csup\u003eF74G\u003c/sup\u003e. The small molecule 5-Ph-IAA induces binding of the AID to OsTIR1\u003csup\u003eF74G\u003c/sup\u003e within the Skp1, Cullin, and F-box (SCF) ubiquitin ligase complex, resulting in the ubiquitination and proteasomal degradation of AID-tagged KMT2D. Transcriptome changes were analyzed at the indicated time points after treatment with 5-Ph-IAA.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e. Western blot showing KMT2D, KDM6A (UTX) and ACTINB protein expression in nuclear extracts of AID-KMT2D HUDEP-2 cells at the indicated time points after treatment with 5-Ph-IAA.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e. MA plots\u003cstrong\u003e \u003c/strong\u003eshowing the transcriptional changes in AID-KMT2D HUDEP-2 and K562 cells after 6, 12, and 24 hours of 5-Ph-IAA treatment (fold change \u0026gt; 1.5, P \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e. Venn diagrams displaying the overlap of downregulated genes in AID-KMT2D HUDEP-2 and K562 cells after 6, 12, and 24 hours of 5-Ph-IAA treatment (fold change \u0026gt; 1.5, P \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e. TF enrichment analysis of the gene groups defined in\u0026nbsp;panel \u003cstrong\u003eD\u003c/strong\u003e. The relative P value was calculated from enrichR online software.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e. Heat map showing GATA1 targets from the genes in panel \u003cstrong\u003eE.\u003c/strong\u003e Arrows indicated selected GATA1 target genes analyzed in greater detail in panel \u003cstrong\u003eI\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e. The GATA1 and KMT2D dual-degradation system. HUDEP-2 cells expressing KMT2D-AID2 fusion protein were constructed as described in panel A. Cells with or without this fusion protein were transduced with lentiviral vector encoding FKBP12\u003csup\u003eF36V\u003c/sup\u003e fused to the GATA1 N-terminus, followed by Cas9 disruption of the endogenous \u003cem\u003eGATA1\u003c/em\u003e gene. Addition of the small molecule dTAGV-13 recruits the CRBN E3 ubiquitin ligase complex to FKBP12\u003csup\u003eF36V\u003c/sup\u003e-GATA1, leading to its ubiquitylation and subsequent proteasomal degradation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e. Western blot showing KMT2D, KDM6A(UTX) and lentiviral vector-encoded FKBP12\u003csup\u003eF36V\u003c/sup\u003e-GATA1 proteins in HUDEP-2 cells, with or without addition of IAA or dTAG13 to induce degradation of KMT2D-AID or FKBP12\u003csup\u003eF36V\u003c/sup\u003e-GATA1, respectively.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(I)\u003c/strong\u003e. RT-qPCR showing \u003cem\u003eSLC25A37\u003c/em\u003e, \u003cem\u003eZFPM1 (FOG1), SLC22A4\u003c/em\u003e, and\u003cem\u003e EPOR\u003c/em\u003e mRNA levels normalized to β-Actin in AID-KMT2D HUDEP-2 cells treated as described for panel \u003cstrong\u003eH\u003c/strong\u003e. Data are expressed as the mean ± SEM of three replicates. ***P \u0026lt; 0.001, ****P \u0026lt; 0.0001, unpaired Student’s t-test.\u003c/p\u003e","description":"","filename":"Figure120251001updated3.png","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/27c9f6323c77ccd2ca680f26.png"},{"id":94096592,"identity":"b9e54aaa-eed3-46d9-bcc1-3018c3e7cc4d","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":357449,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKMT2D and GATA1 co-occupy erythroid-expressed genes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e. Genomic heatmaps from three replicate ChIP-seq experiments showing KMT2D binding peaks in undifferentiated AID-KMT2D HUDEP-2 cells + 1 mM 5-Ph-IAA or vehicle for 24 hours.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e. Pie chart depicting the genomic location of all KMT2D-bound regions (7,288) shown in panel \u003cstrong\u003eA\u003c/strong\u003e, analyzed by ChIPseeker.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e. Homer motif enrichment analysis of the top 5 TF motifs at KMT2D-bound regions shown in panel \u003cstrong\u003eA\u003c/strong\u003e. P values were determined by the hypergeometric test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e. Co-occupancy of GATA1 and KMT2D on chromatin. Upper pie charts show the number of KMT2D-bound regions with or without co-occupancy of GATA1. Lower pie chart shows the distribution of GATA1/KMT2D co-bound regions in transcriptional regulatory regions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e. Genomic heatmaps showing epigenetic marks and TF factor binding in undifferentiated HUDEP-2 cells, centered on KMT2D and GATA1 co-localized enhancer regions (1,411), or enhancer regions that bind GATA1 but not KMT2D (6,754), centered on the GATA1-bound region. ChIP-seq analysis of KMT2D was performed in AID-KMT2D HUDEP-2 cells (without IAA treatment) as shown in panel A. ChIP-seq data for histone marks and GATA1 in HUDEP-2 cells is from GSE115357\u003csup\u003e3\u003c/sup\u003e. ChIP-seq data for KLF1 and TAL1 in HUDEP-2 cells is from GSE157311\u003csup\u003e40\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e. Metagene representation of data shown in panel \u003cstrong\u003eE\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e. Empirical Cumulative distribution function (ECDF) of all expressed genes in AID-KMT2D HUDEP-2 cells, genes with GATA1/KMT2D co-occupied enhancers, and genes with enhancers occupied by GATA1 only, as defined in panel\u003cstrong\u003e E\u003c/strong\u003e. The X-axis represents the log2 fold change (FC) in gene expression level at 24 hours after addition of 5-Ph-IAA to eliminate KMT2D; the Y-axis represents cumulative probability. *** P \u0026lt; 0.001, Wilcoxon test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e. The proportions of KMT2D-bound genes among those that are downregulated or upregulated after treatment of AID-KMT2D HUDEP-2 cells with 5-Ph-IAA for 24 hours (left pie charts), those that are KMT2D-dependent or independent, as shown in \u003cstrong\u003eFig. \u0026nbsp;2F \u003c/strong\u003e(middle pie charts)\u003cstrong\u003e,\u003c/strong\u003e and erythroid signature genes\u003csup\u003e41\u003c/sup\u003e (right pie chart). **** P \u0026lt; 0.0001 according to the Wilcoxon test.\u003c/p\u003e","description":"","filename":"Figure120251001updated4.png","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/36dea49567b494f80eda34df.png"},{"id":94096591,"identity":"5cf09d52-eaae-4239-ad4d-ad6fa6e45f97","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":630320,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKMT2D is required for enhancer activation of select GATA1 target genes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e. Genomic heatmaps showing the chromatin landscape at KMT2D-dependent and KMT2D-independent active enhancers (AEs) in AID-KMT2D HUDEP-2 cells, based onKMT2D ChIP-seq signal intensity before 5-Ph-IAA treatment. The analysis includes histone modification profiles (H3K4me1, H3K27ac), ATAC-seq, and GATA1 binding patterns in the non-targeting sgRNA control (Ctrl) and KMT2D-sg2 mediated knockout (KO) HUDEP-2 cells. The AEs were defined by enrichment for H3K4me1, H3K27ac, ATAC-seq, and GATA1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e. Violin plot of the normalized H3K4me1, H3K27ac and GATA1 binding signals (fragments per kilobase of peaks per million reads mapped, FPKM) at KMT2D-dependent AEs in the non-targeting sgRNA control (Ctrl) and KMT2D-sg2 mediated knockout (KO) HUDEP-2 cells. **** P \u0026lt; 0.0001, Wilcoxon test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C).\u003c/strong\u003e Pie charts depicting the proportion of KMT2D-dependent AEs among the downregulated, upregulated, and unchanged genes in AID-KMT2D HUDEP-2 cells treated with 5-Ph-IAA for 24 hours. **** P \u0026lt; 0.0001, Wilcoxon test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e. Empirical Cumulative distribution function (ECDF) of all expressed genes, KMT2D-dependent AE-associated genes and KMT2D-independent AE-associated genes (left panel). The X-axis represents the log2 fold change (FC) in gene expression level at 24 hours after 5-Ph-IAA-induced KMT2D loss in AID-KMT2D HUDEP-2 cells. The Y-axis shows cumulative probability. * P \u0026lt; 0.05, *** P \u0026lt; 0.001, Wilcoxon test.Violin plot (right panel) shows the basal expression levels of genes associated with KMT2D-dependent and KMT2D-independent AEs in HUDEP-2 cells.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e. Genome browser tracks of the \u003cem\u003eZFPM1\u003c/em\u003e (FOG1) locus showing ChIP-seq profiles of H3K4me1, H3K27ac, and GATA1 in the non-targeting sgRNA control (Ctrl) and KMT2D-sg2 treated (KO) HUDEP-2 cells, and the KMT2D signal in AID-KMT2D HUDEP-2 cells at 24 hours after treatment with 5-Ph-IAA or vehicle. The grey rectangles highlight two AEs (Z1, Z2).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e. Genome browser tracks of the \u003cem\u003eEPOR\u003c/em\u003e locus, annotated as described for panel \u003cstrong\u003eE\u003c/strong\u003e. The grey rectangles highlight two active AEs (E1, E2).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e. RT-qPCR showing the relative fold enrichment of H3K4me1 and H3K27ac signal on AEs of \u003cem\u003eZFPM1\u003c/em\u003e (Z1, Z2) and \u003cem\u003eEPOR\u003c/em\u003e (E1, E2) loci, determined by CUT\u0026amp;RUN analysis in AID-KMT2D HUDEP-2 cells at 24 hours after treatment with 5-Ph-IAA or vehicle. *** P \u0026lt; 0.001, **** P \u0026lt; 0.0001, unpaired Student’s t-test.\u003c/p\u003e","description":"","filename":"Figure120251001updated5.png","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/90be46df1b6e01e57561bca1.png"},{"id":94097415,"identity":"48e405c8-cb70-4f9d-aada-19a7547137eb","added_by":"auto","created_at":"2025-10-22 10:15:12","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":346918,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKMT2D is required for the survival and maturation of primary human erythroblasts.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A)\u003c/strong\u003e. Schematic of the experimental process for expansion and differentiation of CD34\u003csup\u003e+ \u003c/sup\u003eHSPCs. HSPCs were electroporated on day 0 with ribonucleoprotein complexes (RNPs) consisting of Cas9 and either non-targeting control (Ctrl) or KMT2D (KMT2D-sg2) sgRNAs, followed by growth in expansion medium or induction of myeloid or erythroid differentiation. Cells were analyzed at various timepoints, as shown in panels \u003cstrong\u003eB-M\u003c/strong\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B)\u003c/strong\u003e. Western blot showing KMT2D, KDM6A and ACTINB protein expression in nuclear extracts on day five of erythroid differentiation.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(C)\u003c/strong\u003e. Expansion of edited HSPCs grown in maintenance medium (left panel). Right panel shows indel frequencies after editing with Cas9-KMT2D-sg2 RNP after two and seven days of expansion. n.s., not significant, unpaired Student’s t-test for left panel; paired Student’s t-test for the right panel.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(D)\u003c/strong\u003e. Flow cytometry analysis showing CD34 expression in edited HSPCs after seven days of expansion in maintenance medium. n.s., not significant, unpaired Student’s t-test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(E)\u003c/strong\u003e. Hematopoietic colonies generated by edited HSPCs in methylcellulose containing multi-lineage cytokines. N = three biological replicate experiments. **\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.01, ***P \u0026lt; 0.001, ****P \u0026lt; 0.0001, unpaired Student’s\u0026nbsp;\u003cem\u003et\u003c/em\u003e-test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e. Expansion of edited\u003csup\u003e \u003c/sup\u003eHSPCs cells during induced erythroid maturation (left panel). The graph on the right shows indel frequencies after KMT2D disruption vs. day of erythroid differentiation. *P \u0026lt; 0.05, **P \u0026lt; 0.01, unpaired Student’s t-test (left panel) or paired Student’s t-test vs. indel frequency on day 2 (right panel).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e. Expression of the erythroid maturation marker CD235a on day 7 of erythroid differentiation. Left panels show representative flow cytometry plots. Right panel shows a summary of three biological replicate experiments. ***P \u0026lt; 0.001, unpaired Student’s t-test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e. Expression of maturation markers CD235a and CD105 at the indicated time points after induced erythroid maturation. Representative flow cytometry plots are shown on the left. The right panels show the summary of three biological replicate experiments. *P \u0026lt; 0.05, ***P \u0026lt; 0.001, unpaired Student’s\u0026nbsp;\u003cem\u003et\u003c/em\u003e-test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(I)\u003c/strong\u003e. Enucleation of CD235a\u003csup\u003e+\u003c/sup\u003e erythroblasts measured by loss of staining with the nuclear dye Hoechst33258 after fourteen days of induced erythroid maturation. *P \u0026lt; 0.05, unpaired Student’s t-test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(J)\u003c/strong\u003e.\u003cstrong\u003e \u003c/strong\u003eMay-Grünwald-Giemsa–stained erythroblasts at day 14 after induced erythroid maturation. Images were acquired with Olympus DP73 camera. Red arrows denote immature erythroblasts. Scale bar, 10 mm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(K)\u003c/strong\u003e. Cell pellets at day 10 of induced erythroid maturation.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(L)\u003c/strong\u003e. Box plots showing the basal expression levels of genes linked to KMT2D-dependent and KMT2D-independent AEs at different stages of erythropoiesis during \u003cem\u003ein vitro\u003c/em\u003e differentiation of CD34⁺ HSPCs. n.s., not significant, * P \u0026lt; 0.05, ** P \u0026lt; 0.01, **** P \u0026lt; 0.0001, Wilcoxon test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(M)\u003c/strong\u003e. Box plots showing the chromatin accessibility signal (ATAC-seq) associated with KMT2D-dependent and KMT2D-independent AEs at different stages of erythropoiesis during \u003cem\u003ein vitro\u003c/em\u003e differentiation of CD34⁺ HSPCs. n.s., not significant, *** P \u0026lt; 0.001, **** P \u0026lt; 0.0001 according to the Wilcoxon test.\u003c/p\u003e","description":"","filename":"Figure120251001updated6.png","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/39e787f66cf7a6d0b9089325.png"},{"id":94097564,"identity":"514b8230-896d-4dbd-9239-aab0b5fa3cf0","added_by":"auto","created_at":"2025-10-22 10:23:12","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":80003,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eModel for selective KMT2D-dependent GATA1 activation of erythroid enhancers.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring early erythropoiesis, the pioneer transcription factor (TF) GATA1 binds to distal enhancers to activate gene expression. On a subset of these enhancers, GATA1 recruits the KMT2D/COMPASS and CBP/p300 complexes, which deposit active histone marks H3K4me1 and H3K27ac, respectively, to drive transcription. Loss of KMT2D causes a reduction of activating histone marks, thereby diminishing the transcriptional activity of GATA1 on a subset of its erythroid target genes. The net result is reduced cell expansion and impaired terminal maturation. GATA1 may recruit KMT2D/COMPASS through direct interactions or by interacting with P300/CBP, which also binds the KMT2D/COMPASS complex.\u003c/p\u003e","description":"","filename":"Figure120251001updated7.png","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/ee3e704c6c0267fb857b8277.png"},{"id":94290503,"identity":"ab180315-01ae-457b-8865-92f20cba2014","added_by":"auto","created_at":"2025-10-27 11:20:15","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3936776,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/8ac74477-612b-4c35-a675-4193aa8ba25b.pdf"},{"id":94096587,"identity":"e7c8fe87-fc0d-4960-8c17-ea530bcbc979","added_by":"auto","created_at":"2025-10-22 10:07:11","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":541798,"visible":true,"origin":"","legend":"Supplementary Fig.1","description":"","filename":"Figure1S20250925updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/a5ec7db4a144a5fbd61c28fa.pdf"},{"id":94096597,"identity":"3a788409-1c34-474f-ba56-661fecbb8270","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":251997,"visible":true,"origin":"","legend":"Supplementary Table 1","description":"","filename":"TableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/8eaf9b4323066bb98e0d7978.xlsx"},{"id":94097563,"identity":"38c0ff39-eb90-483c-bcd9-8cb5c9500af4","added_by":"auto","created_at":"2025-10-22 10:23:12","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":304340,"visible":true,"origin":"","legend":"Supplementary Table 3","description":"","filename":"TableS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/a865a8979a1cf96efd0f2773.xlsx"},{"id":94096593,"identity":"4053d77e-5640-4ecd-bdb4-9310daeb0ec2","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":2147865,"visible":true,"origin":"","legend":"Supplementary Fig.6","description":"","filename":"Figure6S20250926updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/dad6481f47e4c526a6815716.pdf"},{"id":94097569,"identity":"01a512ff-0464-4e06-bdc6-18256e58d7ed","added_by":"auto","created_at":"2025-10-22 10:23:12","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":2603710,"visible":true,"origin":"","legend":"Supplementary Fig.2","description":"","filename":"Figure2S20250928updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/a1f51b9d76b431609314311c.pdf"},{"id":94096598,"identity":"61b65ffc-e447-4727-a292-16fe2c337cdf","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":41977,"visible":true,"origin":"","legend":"Supplementary Table 7","description":"","filename":"TableS7.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/64bd54333f3a85e6f708288b.xlsx"},{"id":94097565,"identity":"725f25ac-446a-43ab-a199-ae7cdc51216c","added_by":"auto","created_at":"2025-10-22 10:23:12","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":1470732,"visible":true,"origin":"","legend":"Supplementary Table 8","description":"","filename":"TableS8.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/1add21f994ecb60c8c5cfbc3.xlsx"},{"id":94097567,"identity":"393e4daf-414b-477c-956a-e1b4d127500c","added_by":"auto","created_at":"2025-10-22 10:23:12","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":227014,"visible":true,"origin":"","legend":"Supplemental Figure legends","description":"","filename":"SupplementalFigurelegends20251013updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/4843792cd54fda64156dcaf2.pdf"},{"id":94096610,"identity":"37b4969e-a306-4a34-988c-71b3f0741fbc","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":5334819,"visible":true,"origin":"","legend":"Supplementary Fig.4","description":"","filename":"Figure4S20251013updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/0319a7b7d201339a12ade9e1.pdf"},{"id":94096604,"identity":"b042517c-1ca8-43b9-ad5c-a3b8ae2dd1b5","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"xlsx","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":2005178,"visible":true,"origin":"","legend":"Supplementary Table 9","description":"","filename":"TableS9.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/b651f6dfd6e7ef8a64863c12.xlsx"},{"id":94096602,"identity":"f973f269-8aa4-4834-ad85-0155b03660e0","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"xlsx","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":28066,"visible":true,"origin":"","legend":"Supplementary Table 4","description":"","filename":"TableS4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/c5c8cc26c5ab75da384eef26.xlsx"},{"id":94097427,"identity":"e48c639b-71a0-48e5-a2ed-f7c993db422f","added_by":"auto","created_at":"2025-10-22 10:15:12","extension":"xlsx","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":5544417,"visible":true,"origin":"","legend":"Supplementary Table 2","description":"","filename":"TableS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/28ddfecf7a580fc638c1f647.xlsx"},{"id":94096623,"identity":"1f95ae79-5b1b-4498-b127-f201cf74fa60","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"xlsx","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":5777919,"visible":true,"origin":"","legend":"Supplementary Table 6","description":"","filename":"TableS6.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/8293965ddd93aaee0f8319a7.xlsx"},{"id":94097422,"identity":"ba978e91-4200-4027-adb7-eee7fd969627","added_by":"auto","created_at":"2025-10-22 10:15:12","extension":"xlsx","order_by":14,"title":"","display":"","copyAsset":false,"role":"supplement","size":20147,"visible":true,"origin":"","legend":"Supplementary Table 10","description":"","filename":"TableS10.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/5e1949e89f36216a38d9f01b.xlsx"},{"id":94098823,"identity":"b60bde91-bbd1-4e05-a15c-3b40d7a91cb8","added_by":"auto","created_at":"2025-10-22 10:39:12","extension":"pdf","order_by":15,"title":"","display":"","copyAsset":false,"role":"supplement","size":1608919,"visible":true,"origin":"","legend":"Supplementary Fig.5","description":"","filename":"Figure5S20250925updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/0e3090347dd1e32e2ae78218.pdf"},{"id":94096608,"identity":"9bd6c8c6-840b-4073-8f9c-190549ebd17e","added_by":"auto","created_at":"2025-10-22 10:07:12","extension":"xlsx","order_by":16,"title":"","display":"","copyAsset":false,"role":"supplement","size":5145109,"visible":true,"origin":"","legend":"Supplementary Table 5","description":"","filename":"TableS5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/696175e77c13bd37510a4226.xlsx"},{"id":94097424,"identity":"067b29e2-f761-4dde-8d1c-63f19ab9e63d","added_by":"auto","created_at":"2025-10-22 10:15:12","extension":"pdf","order_by":17,"title":"","display":"","copyAsset":false,"role":"supplement","size":1766620,"visible":true,"origin":"","legend":"Supplementary Fig.3","description":"","filename":"Figure3S20251013updated.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7853319/v1/ca63c8dfdc73553b1b8118b4.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nM.J.W. is a consultant for Glaxo SmithKline, Cellarity Inc, Graphite Bio, Fulcrum Therapeutics and Dyne Therapeutics, and owns equity in Cellarity Inc. Dr. Li Cheng is currently affiliated with a commercial company (GenAssist Therapeutics Incorporation, Suzhou) and declares that she has no competing interests. Dr. Chunliang Li is currently an editorial board member at Genome Biology. The other authors declare that they have no other competing interests.","formattedTitle":"The H3K4 methyltransferase KMT2D is an essential cofactor for GATA1 at erythroid gene enhancers","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIn general, cellular differentiation is coordinated by specialized transcriptional regulatory networks consisting of lineage-specific and general transcription factors (TFs) and cofactors, including epigenetic regulators\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Red blood cells are generated from hematopoietic stem and progenitor cells (HSPCs) through a stepwise, highly controlled differentiation process known as erythropoiesis\u003csup\u003e\u003cspan additionalcitationids=\"CR4 CR5 CR6\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. During erythropoiesis, the master TF GATA1 binds to enhancers and recruits additional TFs and chromatin modifiers to initiate RNA polymerase II (Pol II)-mediated transcription at promoters\u003csup\u003e\u003cspan additionalcitationids=\"CR9 CR10 CR11 CR12 CR13 CR14\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. How GATA1 and other erythroid TFs modulate enhancer activity remains incompletely understood. Active enhancers are marked by the histone modifications H3K4me1 and H3K27ac, which are catalyzed by specific chromatin \u0026ldquo;writers\u0026rdquo;\u003csup\u003e16\u003c/sup\u003e. The Complex of Proteins Associated with Set1 (COMPASS)-like KMT2C/KMT2D complexes install H3K4me1 at enhancers to activate transcription in collaboration with the histone acetyltransferases p300/CBP, which catalyze H3K27 acetylation\u003csup\u003e\u003cspan additionalcitationids=\"CR18 CR19 CR20 CR21 CR22 CR23\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. While previous studies have uncovered essential roles for several epigenetic modifiers in erythropoiesis\u003csup\u003e\u003cspan additionalcitationids=\"CR26 CR27 CR28 CR29\" citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, the role of the KMT2C/KMT2D COMPASS-like complexes in erythropoiesis remains unexplored.\u003c/p\u003e\u003cp\u003eWe performed a CRISPR/single guide RNA (sgRNA) screen to investigate the functions of 496 epigenetic modifiers in human erythropoiesis. Our data revealed that Cas9-sgRNA targeting genes related to histone H3K4 methylation caused impaired erythroid maturation. Specifically, KMT2D, the enzymatic subunit of the KMT2D COMPASS-like complex, was identified as a top positive regulator of erythroid maturation. We validated that loss of KMT2D delays erythroid maturation in the immortalized erythroblast progenitor line HUDEP-2 and in primary human CD34\u003csup\u003e+\u003c/sup\u003e HSPCs. Transcriptome analysis revealed that KMT2D is required for the dynamic activation of approximately 200 erythroid signature genes. To define the immediate consequences of acute KMT2D loss, we established auxin-inducible degron 2 (AID2) systems in two cellular models for erythropoiesis\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e and showed that hundreds of KMT2D-bound genes were suppressed within 6 hours of KMT2D depletion. Remarkably, more than 7,000 KMT2D-bound regions were enriched for the cognate GATA1 binding motif, and KMT2D colocalized with GATA1 at over 1,000 active enhancers. More importantly, enhancers bound by GATA1 and KMT2D exhibited stronger H3K4me1 and H3K27ac signals and stronger binding of core erythroid TFs GATA1, KLF1, and TAL1 relative to KMT2D-independent enhancers. After KMT2D depletion, H3K4me1 and H3K27ac marks on those enhancers were significantly reduced, resulting in transcriptional repression of the associated target genes. Thus, we propose that KMT2D is essential for GATA1-dependent enhancer activation at a subset of key erythroid genes.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eAn unbiased CRISPR/Cas9 screen identifies KMT2D as a critical regulator of erythropoiesis\u003c/h2\u003e\u003cp\u003eTo comprehensively interrogate the functions of epigenetic modifiers in human erythropoiesis, we conducted a CRISPR knockout screen in HUDEP-2 cells, an immortalized erythroblast line that can be induced to undergo terminal maturation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA)\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. HUDEP-2 cells stably expressing Cas9 were transfected with a lentiviral vector library encoding 5,082 sgRNAs targeting a total of 496 genes encoding epigenetic regulators, grown in maintenance medium for 5 days, then switched to erythroid maturation medium for 5 days. Cells were fractionated by immune flow cytometry for surface expression of the erythroid maturation marker Band3 and analyzed by next-generation DNA sequencing for sgRNA representation in Band3\u003csup\u003eHigh\u003c/sup\u003e and Band3\u003csup\u003eLow\u003c/sup\u003e cells (\u003cb\u003eSupplementary table 1\u003c/b\u003e). Guide RNA sequences enriched in the Band3\u003csup\u003eLow\u003c/sup\u003e population are predicted to target positive regulators of erythroid maturation. As validation, sgRNAs targeting the previously reported positive regulator of erythropoiesis FBXO11 were enriched in Band3\u003csup\u003eLow\u003c/sup\u003e cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB, \u003cb\u003eSupplementary table 1\u003c/b\u003e)\u003csup\u003e3\u003c/sup\u003e. Notably, KMT2D and NCOA6, members of the KMT2C/D COMPASS-like complexes, were among the top three positive regulators (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB, C; \u003cb\u003eSupplementary table 1\u003c/b\u003e). Additionally, SETD1A/SETD1B COMPASS components (SETD1A, SETD1B, and CXXC1) and core subunits of COMPASS (RBBP5 and DPY30) were identified as significant positive regulators. Gene ontology (GO) analysis of the 28 top positive regulators of erythroid maturation revealed histone H3 lysine 4 (H3K4) methylation as the highest-scoring biological process, reflecting the importance of COMPASS and COMPASS-like subunits in erythropoiesis (\u003cb\u003eSupplementary Fig.\u0026nbsp;1A\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe next analyzed the expression profiles of KMT2C/D and SETD1A/SETD1B complex components during \u003cem\u003ein vitro\u003c/em\u003e erythroid differentiation of CD34\u003csup\u003e+\u003c/sup\u003e HSPCs. Most subunits started to become upregulated at the CFU-E or proerythroblast stages, with expression of KMT2C/D components maintained at higher levels into later stages of erythroid maturation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC, D; \u003cb\u003eSupplementary Fig.\u0026nbsp;1B)\u003c/b\u003e. Transfection of Cas9-expressing HUDEP-2 cells with two different \u003cem\u003eKMT2D\u003c/em\u003e-targeting sgRNAs caused marked reductions in the expression of KMT2D and the KMT2C/D COMPASS complex-specific subunit KDM6A/UTX (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE). Disruption of \u003cem\u003eKMT2D\u003c/em\u003e had no effect on the expansion of HUDEP-2 cells grown in maintenance medium (\u003cb\u003eSupplementary Fig.\u0026nbsp;1C\u003c/b\u003e). However, after switching to erythroid maturation medium, KMT2D-depleted cells exhibited reduced expression of cell surface Band3 and less hemoglobinization (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eF, G). Next, we transfected HUDEP-2 cells with ribonucleoprotein complexes (RNPs) consisting of Cas9 and one of two different \u003cem\u003eKMT2D\u003c/em\u003e-targeting sgRNAs or non-targeting sgRNA control. Editing efficiencies were maintained over six days of growth in expansion medium, indicating that \u003cem\u003eKMT2D\u003c/em\u003e-disrupted cells were not outgrown by non-targeted cells (\u003cb\u003eSupplementary Fig.\u0026nbsp;1D\u003c/b\u003e). However, during induced erythroid maturation, \u003cem\u003eKMT2D\u003c/em\u003e-disrupted cells exhibited reduced expression of mRNAs encoding the erythroid markers CD235a (\u003cem\u003eGYPA\u003c/em\u003e), Band3 (\u003cem\u003eSLC4A1\u003c/em\u003e) and adult globins (\u003cem\u003eHBB\u003c/em\u003e, \u003cem\u003eHBA\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH; \u003cb\u003eSupplementary Fig.\u0026nbsp;1E\u003c/b\u003e).\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eKMT2D is required for activation of erythroid signature genes and proper chromatin accessibility of GATA1\u003c/h3\u003e\n\u003cp\u003eTo understand how KMT2D loss impairs erythroid maturation, we performed RNA sequencing (RNA-seq) of HUDEP-2 cells expressing Cas9 and control non-targeting sgRNA (Ctrl) or KMT2D-sg2 (KMT2D KO) in the immature state and after three days of induced maturation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Principal component analysis (PCA) distinguished all four groups (\u003cb\u003eSupplementary Fig.\u0026nbsp;2A\u003c/b\u003e). Moreover, the differences between control and KMT2D KO cells before maturation were less than those after induced maturation, indicating a greater effect of KMT2D disruption on the transcriptome of the latter group (\u003cb\u003eSupplementary Fig.\u0026nbsp;2A\u003c/b\u003e). In agreement, the number of differentially expressed genes (DEGs), defined as fold change\u0026thinsp;\u0026gt;\u0026thinsp;1.5 and FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.05, was greater between control and KMT2D KO cells after maturation vs. before maturation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB; \u003cb\u003eSupplementary table 2\u003c/b\u003e). The DEGs identified in immature cells were largely different from those identified during maturation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC), suggesting a dynamic role for KMT2D in erythroid development. Gene set enrichment analysis (GSEA), gene ontology (GO), and enrichR\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e analysis showed downregulation of genes associated with erythroid differentiation and heme metabolism, and GATA1 target genes in KMT2D KO cells, before and after induced maturation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD; \u003cb\u003eSupplementary Fig.\u0026nbsp;2B\u0026ndash;D; Supplementary table 3\u003c/b\u003e). In contrast, HSC or lymphocyte lineage signatures were enriched in gene sets that were upregulated after KMT2D KO (\u003cb\u003eSupplementary table 3\u003c/b\u003e). In total, 11,400 genes were expressed in control HUDEP-2 cells after maturation. Of those, 2,237 genes and 1,739 genes were significantly downregulated and upregulated, respectively, relative to immature cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). Among the 1,739 upregulated genes, 367 were KMT2D-dependent and significantly enriched for erythropoiesis-related pathways and GATA1 occupancy (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF\u0026ndash;H). Erythroid genes that were downregulated by KMT2D disruption before and after induced maturation include \u003cem\u003eSLC25A37, GYPA, EPOR, SLC4A1, ALAS2\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eI). In contrast, several erythroid genes were not regulated by KMT2D, including \u003cem\u003eWDR26\u003c/em\u003e, \u003cem\u003eJAK2\u003c/em\u003e, \u003cem\u003eANK1\u003c/em\u003e, and \u003cem\u003eBCL11A\u003c/em\u003e (\u003cb\u003eSupplementary table 2\u003c/b\u003e). These findings support the conclusion that KMT2D facilitates erythroid maturation by activating selective erythroid signature genes.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eGiven that KMT2D-dependent genes were significantly enriched for GATA1 occupancy, we hypothesized that KMT2D maintains appropriate chromatin accessibility for specific transcription factors (TFs) in erythroid progenitors. To test this hypothesis, we performed genome-wide ATAC-seq in both control and KMT2D KO HUDEP-2 cells. We identified differential accessibility regions (DARs) from a total of 27,919 reproducible peaks using a stringent cutoff (false discovery rate (FDR) controlled \u003cem\u003eP\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and \u0026gt;\u0026thinsp;2-fold change). Notably, 659 chromatin accessibility regions were significantly decreased, while 119 regions were significantly increased (\u003cb\u003eSupplementary Fig.\u0026nbsp;2E; Supplementary table 2\u003c/b\u003e). We then predicted the TF occupancy profiles of these DARs using the TRANSFAC and Homer motif database as a reference and scored the enrichment frequency\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Regions with decreased chromatin accessibility after KMT2D KO were most significantly enriched for GATA TF binding motifs (\u003cb\u003eSupplementary Fig.\u0026nbsp;2F, G; Supplementary table 2\u003c/b\u003e). These results indicate that KMT2D is selectively required for maintaining proper chromatin accessibility for GATA1 in erythroid progenitor cells.\u003c/p\u003e\n\u003ch3\u003eAcute depletion of KMT2D selectively suppresses the expression of GATA1 targets\u003c/h3\u003e\n\u003cp\u003eTranscriptome analysis of KMT2D KO HUDEP-2 cells at steady state cannot distinguish direct from indirect effects on gene expression. We established an AID2-based acute protein degradation system\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e for KMT2D to study its direct target genes. We fused an EGFP-AID degron cassette to the amino (N)-terminus of both endogenous KMT2D loci in HUDEP-2 and K562 cells that were engineered to constitutively express OsTIR1\u003csup\u003eF74G\u003c/sup\u003e, a substrate receptor for the Skp1, Cullin, F-box (SCF) ubiquitin ligase complex (\u003cb\u003eSupplementary Fig.\u0026nbsp;3A, B\u003c/b\u003e). In the presence of the small molecule 5-Ph-IAA, OsTIR1\u003csup\u003eF74G\u003c/sup\u003e mediates the ubiquitination and proteasomal degradation of AID-tagged proteins (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Thus, addition of 5-Ph-IAA to AID-KMT2D-HUDEP-2 or -K562 cells resulted in depletion of KMT2D within six hours, which was reversible after drug washout (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB; \u003cb\u003eSupplementary Fig.\u0026nbsp;3C, 3D\u003c/b\u003e). To determine whether EGFP-AID-tagged KMT2D assembles into the KMT2D complex, we performed immunoprecipitation in nuclear lysates with an anti-KMT2D antibody followed by mass spectrometry. We identified 212 KMT2D-interacting proteins relative to an IgG control (\u003cb\u003eSupplementary table S4)\u003c/b\u003e. Notably, all nine subunits of the KMT2D complex were substantially enriched (\u003cb\u003eSupplementary Fig.\u0026nbsp;3E\u003c/b\u003e). Additionally, protein network analysis confirmed known KMT2D-interacting proteins, such as the Mediator and SWI/SNF complexes\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e (\u003cb\u003eSupplementary Fig.\u0026nbsp;3F)\u003c/b\u003e. These findings suggest that the N-terminal GFP-AID fusion did not interfere with the proper assembly of the KMT2D complex. The addition of 5-Ph-IAA to AID-KMT2D HUDEP-2 cells had no effect on proliferation in either expansion or maturation medium, but caused a block to terminal maturation, including reduced hemoglobinization and failure to upregulate numerous erythroid marker genes (\u003cb\u003eSupplementary Fig.\u0026nbsp;3G\u0026ndash;J\u003c/b\u003e). Thus, drug-induced depletion of AID-KMT2D in HUDEP-2 cells phenocopies the effects of KMT2D depletion, suggesting that the fusion protein retains biological function.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe next examined transcriptome changes at 6, 12, and 24 hours after 5-Ph-IAA treatment of AID-KMT2D-HUDEP-2 and AID-KMT2D-K562 cells in expansion medium (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Principal component analysis revealed distinct gene expression profiles at each timepoint after addition of 5-Ph-IAA (\u003cb\u003eSupplementary Fig.\u0026nbsp;3K\u003c/b\u003e), with progressive increases in the number of DEGs from 6 to 24 hours and common downregulated genes at three timepoints (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC, D; \u003cb\u003eSupplementary tables 5 and table 6\u003c/b\u003e). The DEGs were strongly enriched for occupancy of GATA1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE, F; \u003cb\u003eSupplementary table 7\u003c/b\u003e). To further confirm that GATA1 and KMT2D share common targets, we generated a dTAG degradation system\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e. Specifically, we used lentiviral vectors to overexpress the FKBP\u003csup\u003eF36V\u003c/sup\u003e-GATA1 fusion gene while disrupting the endogenous GATA1 gene in AID-KMT2D-HUDEP-2 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG). Addition of 5-Ph-IAA or dTAG13 caused acute degradation of KMT2D or GATA1, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH), and resulted in the downregulation of overlapping gene sets, including \u003cem\u003eSLC25A37\u003c/em\u003e, \u003cem\u003eSLC22A4\u003c/em\u003e, \u003cem\u003eZFPM1\u003c/em\u003e and \u003cem\u003eEPOR\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI). Taken together, these results suggest that KMT2D directly activates selective GATA1 target genes to facilitate human erythropoiesis.\u003c/p\u003e\n\u003ch3\u003eKMT2D and GATA1 co-occupy erythroid-expressed genes\u003c/h3\u003e\n\u003cp\u003eWe next performed ChIP-seq analysis of AID-KMT2D-HUDEP-2 cells to identify erythroid genes that bound KMT2D. In the absence of 5-Ph-IAA, 7,288 KMT2D occupancy peaks were detected reproducibly in three replicate experiments (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA; \u003cb\u003eSupplementary Fig.\u0026nbsp;4A, B; Supplementary table 8\u003c/b\u003e). Occupancy of KMT2D was markedly reduced at 24 hours after the addition of 5-Ph-IAA. Notably, KMT2D occupancy predominated at distal intergenic regions (46%) and introns (32%), compared to promoters (17%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). Homer analysis revealed enrichment of GATA/SCL, GATA, and KLF binding motifs (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC; \u003cb\u003eSupplementary Fig.\u0026nbsp;4C; Supplementary table 8\u003c/b\u003e). To test whether KMT2D co-localizes with erythroid transcription factor binding on cis-regulatory elements, we performed an integrated analysis using available ChIP-seq data for GATA1, KLF1, TAL1 occupancy, and histone marks in wild-type HUDEP-2 cells (GSE115357, GSE157311)\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Approximately 35% of KMT2D-bound regions (2,559 peaks) were co-localized with GATA1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). Over half of KMT2D-GATA1 co-bound regions were at enhancers (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). Moreover, the binding signals of histone marks H3K4me1 and H3K27ac, as well as the erythroid TFs KLF1 and TAL1, were significantly higher at KMT2D-GATA1 co-occupied enhancers compared to enhancers bound by GATA1 alone (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE, F). In addition, genes that were co-occupied by KMT2D and GATA1 were more rapidly suppressed after acute KMT2D degradation than genes that were bound by GATA1 only (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG). Moreover, a co-immunoprecipitation assay identified interactions between GATA1 and the key KMT2D/COMPASS component KDM6A (\u003cb\u003eSupplementary Fig.\u0026nbsp;4D and E\u003c/b\u003e). These results suggest that KMT2D/COMPASS complex and GATA1 co-localize on active enhancers to activate gene expression. Next, we investigated whether KMT2D binding occurs at genes that are transcriptionally deregulated after KMT2D depletion. Notably, KMT2D binding was enriched at genes that were downregulated after acute depletion of KMT2D in AID-KMT2D HUDEP-2 cells, shown to be KMT2D-dependent (see Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF), or designated as erythroid differentiation signature genes\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH). Additionally, genes bound by KMT2D exhibited higher basal expression levels compared to those unbound by KMT2D, both in HUDEP-2 cells and in primary erythroblasts derived from primary HSPCs (\u003cb\u003eSupplementary Fig.\u0026nbsp;4F\u003c/b\u003e). Thus, KMT2D and GATA1 co-occupy selective erythroid target genes, thereby enhancing their expression.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eKMT2D is required for enhancer activation of select GATA1 target genes\u003c/h3\u003e\n\u003cp\u003eWe next asked whether KMT2D is required for GATA1-mediated enhancer activation during erythroid maturation by performing ChIP-seq analysis for H3K4me1, H3K27ac, and GATA1 in control and KMT2D KO HUDEP-2 cells. We utilized the GATA1 binding pattern and histone marks to annotate active enhancers (AEs: H3K4me1+, GATA1+, H3K27ac+)\u003csup\u003e42\u003c/sup\u003e, then separated the total annotated AEs into KMT2D-dependent (1,348) and KMT2D-independent AEs (9,643) based on the KMT2D binding signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA; \u003cb\u003eSupplementary table 9\u003c/b\u003e). Importantly, both AE histone marks and GATA1 binding signals were stronger on KMT2D-dependent AEs relative to KMT2D-independent AEs. Depletion of KMT2D resulted in a significant reduction of the histone marks H3K4me1 and H3K27ac at KMT2D-dependent AEs, while the GATA1 signals did not change (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Compared with KMT2D-independent enhancers, KMT2D\u0026ndash;GATA1 co-dependent enhancers are enriched in erythroid and myeloid differentiation pathways and harbor core erythroid-specific TF motifs, including those for GATA1 and KLF1(\u003cb\u003eSupplementary Fig.\u0026nbsp;5A, B; Supplementary table 9\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe next asked how enhancer activity is associated with gene expression. In AID-KMT2D HUDEP-2 cells treated with 5-Ph-IAA, a greater proportion of downregulated genes was associated with KMT2D-dependent AEs compared to upregulated genes or unchanged genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). In addition, KMT2D-dependent AE-associated genes exhibited higher basal expression levels and were more likely to be suppressed by KMT2D depletion compared to genes with KMT2D-independent AEs in HUDEP-2 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD). Examples of KMT2D and GATA1 co-dependent target genes include \u003cem\u003eZFPM1, SLC4A1 and EPOR\u003c/em\u003e, which were directly occupied by both GATA1 and KMT2D in their AE regions (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE, F; \u003cb\u003eSupplementary Fig.\u0026nbsp;5C\u003c/b\u003e). Similar to what we observed after disruption of endogenous \u003cem\u003eKMT2D\u003c/em\u003e genes in HUDEP-2 cells, acute depletion of KMT2D in AID-KMT2D cells led to significant reductions of the active histone marks (H3K4me1 and H3K27ac), whereas GATA1 binding signals were stable (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG; \u003cb\u003eSupplementary Fig.\u0026nbsp;5D, E\u003c/b\u003e). Thus, KMT2D is essential for the activation of GATA1-dependent enhancers during erythropoiesis but is dispensable for its chromatin occupancy.\u003c/p\u003e\u003cp\u003e\u003cb\u003eKMT2D is required for the survival and maturation of primary human erythroblasts.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo investigate further whether KMT2D is required for erythropoiesis, we transfected normal donor peripheral blood-mobilized CD34\u003csup\u003e+\u003c/sup\u003e HSPCs with RNPs consisting of Cas9 and \u003cem\u003eKMT2D\u003c/em\u003e-targeting or control non-targeting sgRNAs followed by expansion in stem cell/progenitor medium or \u003cem\u003ein vitro\u003c/em\u003e differentiation toward erythroid or myeloid lineages (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA)\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Transfection of CD34\u003csup\u003e+\u003c/sup\u003e cells with \u003cem\u003eKMT2D\u003c/em\u003e-targeting RNP caused marked depletion of the corresponding protein compared to control cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB). KMT2D-disrupted HSPCs that were maintained in stem cell/progenitor medium expanded normally for up to seven days, with no dropout of \u003cem\u003eKMT2D\u003c/em\u003e indels or apoptosis (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC, D; \u003cb\u003eSupplementary Fig.\u0026nbsp;6A, B\u003c/b\u003e). However, \u003cem\u003eKMT2D\u003c/em\u003e suppression via Cas9 in CD34\u003csup\u003e+\u003c/sup\u003e HSPCs caused significant reductions in erythroid, myeloid and mixed lineage colonies (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE). We further analyzed the effect on erythroid differentiation by suspension culture system. In contrast to findings in immature HUDEP-2 cells grown in expansion medium, disruption of \u003cem\u003eKMT2D\u003c/em\u003e in CD34\u003csup\u003e+\u003c/sup\u003e cells was associated with significant dropout of \u003cem\u003eKMT2D\u003c/em\u003e indels and impaired cell expansion during \u003cem\u003ein vitro\u003c/em\u003e erythroid differentiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF). Consistent with findings in HUDEP-2 cells, depletion of KMT2D caused impaired erythroid maturation as evidenced by reduced expression of CD235a and CD105 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eG, H), reduced enucleation, immature morphology, and reduced hemoglobinization (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eI\u0026ndash;K). Similarly, \u003cem\u003ein vitro\u003c/em\u003e myeloid differentiation of \u003cem\u003eKMT2D\u003c/em\u003e-disrupted CD34\u003csup\u003e+\u003c/sup\u003e cells was associated with reduced cell expansion, indel dropout, and mildly reduced expression of the myeloid maturation marker CD11b (\u003cb\u003eSupplementary Fig.\u0026nbsp;6C, D\u003c/b\u003e). To test the effects of KMT2D depletion using an orthologous approach, we transduced CD34\u003csup\u003e+\u003c/sup\u003e HSPCs with a lentiviral vector encoding one of two different \u003cem\u003eKMT2D\u003c/em\u003e shRNAs followed by \u003cem\u003ein vitro\u003c/em\u003e erythroid maturation. Both targeting shRNAs caused approximately 40% reduction in \u003cem\u003eKMT2D\u003c/em\u003e mRNA relative to luciferase shRNA control (\u003cb\u003eSupplementary Fig.\u0026nbsp;6E\u003c/b\u003e), which resulted in reduced erythroid, myeloid and mixed lineage colony formation (\u003cb\u003eSupplementary Fig.\u0026nbsp;6F\u003c/b\u003e) and impaired erythroid maturation, evidenced by characteristic cell surface markers, immature morphology and reduced hemoglobinization (\u003cb\u003eSupplementary Fig.\u0026nbsp;6G-I\u003c/b\u003e). These results suggest that KMT2D is dispensable for CD34\u003csup\u003e+\u003c/sup\u003e HSPCs but is required for the proliferation and/or survival of committed erythroid and myeloid lineages.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWe next investigated the dynamics of gene expression levels and chromatin accessibility signals associated with KMT2D-dependent and KMT2D-independent enhancers during \u003cem\u003ein vitro\u003c/em\u003e erythroid differentiation of CD34\u003csup\u003e+\u003c/sup\u003e HSPCs. Notably, the expression levels of genes associated with KMT2D-dependent AEs were consistently higher in erythroid progenitors of all maturation stages (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eL). Moreover, KMT2D-dependent AEs exhibited significantly higher chromatin accessibility signals across the same developmental stages (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eM). These results indicate that KMT2D stimulates the expression of erythroid genes by activating their enhancers during the differentiation of primary HSPCs.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we identified KMT2D as a crucial cofactor for the master TF GATA1 in human erythropoiesis. Specifically, we demonstrated that KMT2D co-localizes with GATA1 on a subset of erythroid enhancers and is essential for their activation during terminal erythroid maturation. More generally, our findings reveal how a lineage-specific TF (GATA1) utilizes a general histone methyltransferase complex (KMT2D/COMPASS) to facilitate gene expression.\u003c/p\u003e\u003cp\u003eWe propose a model in which the pioneer TF GATA1\u003csup\u003e43\u003c/sup\u003e binds to distal enhancers during early erythropoiesis and recruits the KMT2D/COMPASS and CBP/p300 complexes, which deposit active histone marks H3K4me1 and H3K27ac, respectively, thereby driving transcription (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). Supporting this model: 1) loss of KMT2D in erythroblasts impairs terminal maturation and erythroid gene expression, which resembles the loss of GATA1; 2) the genome-wide occupancy of KMT2D is highly enriched for the GATA motif and KMT2D co-occupies with GATA1 on thousands of erythroid enhancers that regulate highly expressed erythroid signature genes; 3) Induced KMT2D degradation resulted in enhancer inactivation with decreased H3K4me1 and H3K27ac marks, despite stable GATA1 binding; 4) GATA1 is physically associated with KMT2D/COMPASS complex through KDM6A.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eA key question is how to distinguish KMT2D-dependent enhancers from those that are KMT2D-independent. Our findings suggest that KMT2D-dependent enhancers are more likely to be erythroid-specific, as they are occupied by the erythroid transcription factor GATA1 and associated with high expression levels of their target genes. This aligns with recent studies showing that KMT2D is crucial for dynamic enhancer activity during neuronal differentiation\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. Additionally, chromatin looping interactions may differentiate KMT2D-dependent and KMT2D-independent enhancers, given KMT2D\u0026rsquo;s role in facilitating chromatin looping\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Supporting this, KMT2D interactome data identified components of the Cohesin and Mediator complexes, which are essential for chromatin looping\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Future studies should investigate these features to elucidate the regulatory mechanisms of KMT2D-dependent enhancers, potentially uncovering additional complexities in the interplay between KMT2D and other regulatory factors to enhance our understanding of erythropoiesis.\u003c/p\u003e\u003cp\u003eKMT2D lacks intrinsic DNA-binding ability, prompting the question of how it is recruited to enhancers or promoters. Typically, KMT2D is recruited by transcription factors (TFs). For example, KMT2D has been shown to colocalize with pluripotency TFs such as Oct4, Sox2, and Nanog on active enhancers in mouse ESCs through physical interactions\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Notably, recent studies suggest that KMT2D-dependent dynamic enhancers may be driven by GATA family members during embryonic stem cell differentiation\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. During erythropoiesis, GATA1 binds the chromatin modifiers CBP/P300, the Med1 subunit of the Mediator complex, and the SWI/SNF catalytic subunit BRG1 to promote erythroid differentiation\u003csup\u003e\u003cspan additionalcitationids=\"CR46 CR47\" citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e. Recently, a study mapping human TF interaction networks identified KMT2D, KDM6A, and PAXI1 as interaction partners for both GATA1 and GATA2 using proximity-dependent biotinylation (BioID)\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. Consistent with this, our KMT2D IP-mass dataset from K562 cells identified the known GATA1 cofactors, including several SWI/SNF and Mediator complex components and CBP/P300. More importantly, the GATA1 pulldown assay confirmed the presence of KDM6A, suggesting that GATA1 recruits the KMT2D/COMPASS complex to erythroid enhancers and promoters via KDM6A. Another possibility is that KMT2D/COMPASS is recruited via GATA2, which shares many binding sites with GATA1. During erythropoiesis, GATA1 displaces GATA2 from chromatin and represses GATA2 expression\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. Future studies will be necessary to fully elucidate the detailed biochemical mechanisms between GATA1 and the KMT2D/COMPASS complex.\u003c/p\u003e\u003cp\u003eOur study elucidated KMT2D\u0026rsquo;s role in human erythroid lineage differentiation and erythroid signature gene expression. A genetic screen in mice identified over 140 chromatin factors involved in myeloid and megakaryocytic-erythroid lineage specification, highlighting that Kmt2d and Kdm6a, but not Kmt2c, are crucial for myeloid progenitor identities and early myeloid priming\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. In our study, an unbiased genetic screen targeting 496 human epigenetic modifiers underscored the importance of COMPASS-like components, including KMT2D, NCOA6, and KDM6A, in definitive erythroid maturation. Notably, KMT2D is essential for the survival of erythroid and myeloid progenitors but not hematopoietic stem and progenitor cells (HSPCs). Collectively, our data and previous reports suggest that KMT2D dynamically influences hematopoietic lineage specification, affecting both myelopoiesis and erythropoiesis. This role likely modulates the activity of various lineage-specific transcription factors. However, the precise regulatory mechanisms of KMT2D in hematopoiesis, particularly during transitions between myeloid lineages, warrant further investigation via \u003cem\u003ein vivo\u003c/em\u003e models.\u003c/p\u003e\u003cp\u003eKMT2D and its Drosophila homolog Trithorax-related are major H3K4 methyltransferases for H3K4me1 and H3K4me3 on enhancers during cell fate transitions in development and disease\u003csup\u003e\u003cspan additionalcitationids=\"CR54 CR55\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. Using the KMT2D-AID system in human erythroid cell lines, we identified over 100 downregulated genes whose regulatory elements were directly occupied by both KMT2D and GATA1. A limitation of our study is that we could not determine the extent to which the observed transcriptional changes depended on KMT2D\u0026rsquo;s enzymatic activity, given its recently recognized noncatalytic functions\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e. Additionally, other H3K4 methyltransferases like KMT2B or KMT2C might compensate for KMT2D in some contexts\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e, which we could not fully rule out. Thus, comprehensive genetic studies involving triple or double knockouts, and catalytically inactive mutants of KMT2D, are required to better define its roles in erythropoiesis.\u003c/p\u003e\u003cp\u003eIn summary, we identified KMT2D as a crucial cofactor for GATA1 in human erythropoiesis. Our findings provide a foundation for further investigating KMT2D\u0026rsquo;s role and its clinical implications in hematopoiesis and blood cancers.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003eCulture and induced maturation of HUDEP-2 and K562 cells\u003c/h2\u003e\u003cp\u003eImmature HUDEP-2 cells were expanded in the StemSpan serum-free medium (SFEM; STEMCELL Technologies) supplemented with 1 \u0026micro;M dexamethasone, 1 \u0026micro;g/mL doxycycline, 50 ng/mL human stem cell factor (SCF), 3 units/mL erythropoietin (EPO), and 1% penicillin\u0026ndash;streptomycin\u003csup\u003e31\u003c/sup\u003e. To induce erythroid maturation, HUDEP-2 cells were cultured in a differentiation medium composed of IMDM base medium (Invitrogen) supplemented with 2% FBS, 3% human serum albumin, 3 units/mL EPO, 10 \u0026micro;g/mL insulin, 1000 \u0026micro;g/mL holo-transferrin, and 3 units/mL heparin. Erythroid maturation was monitored by flow cytometry, using FITC-conjugated anti-CD235a (BD Biosciences, clone GA-R2), APC-conjugated anti-Band3 (from New York Blood Center), APC-conjugated anti-CD105 (BioLegend, clone SN6h) and Violet Blue\u0026ndash;conjugated anti-CD49d (Miltenyi Biotec, clone MZ18-24A9) antibodies. Band3\u003csup\u003ehigh\u003c/sup\u003e and Band3\u003csup\u003elow\u003c/sup\u003e cell populations from the CD235a\u003csup\u003e+\u003c/sup\u003e cell fraction was purified by fluorescence-activated cell sorting (FACS). The human K562 cell line was maintained in RPMI-1640 medium (Hycita) containing 10% fetal bovine serum (FBS) (Hyclone), 2 mM glutamine (Sigma) and 1% penicillin/streptomycin (Thermo Fisher Scientific). To induce erythroid maturation of K562 cells, 10uM Hemin (HY-19424) can be added to the expansion medium.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eCD34\u003csup\u003e+\u003c/sup\u003e cell culture and manipulation\u003c/h2\u003e\u003cp\u003eCD34\u003csup\u003e+\u003c/sup\u003e hematopoietic stem and progenitor cells (HSPCs) were mobilized from normal subjects by granulocyte colony-stimulating factor, collected by apheresis, and enriched by immunomagnetic bead selection using an autoMACS Pro Separator (Miltenyi Biotec), in accordance with the manufacturer\u0026rsquo;s protocol. At least 95% purity was achieved, as assessed by flow cytometry using a PE-conjugated anti-human CD34 antibody (Miltenyi Biotec, clone AC136, catalog #130-081- 002). A two-phase culture protocol was used to promote erythroid differentiation and maturation\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. In phase 1 (days 0\u0026ndash;6), cells were cultured at a density of 2 * 10\u003csup\u003e5\u003c/sup\u003e cells/mL in SFEM with 10% FBS, 1% penicillin/streptomycin, 50 ng/mL SCF, 1.6U EPO and 10 ng/mL IL-3. In phase 2 (days 6\u0026ndash;14), IL-3 and SCF was omitted from the medium and additional addition to 30% FBS. Erythroid differentiation and maturation were monitored by flow cytometry, using FITC-conjugated anti-CD235 (BD Biosciences, clone GA-R2, catalog #561017), APC-conjugated anti-CD105 (BioLegend, clone SN6h) antibodies.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eCRISPR library construction and screening\u003c/h2\u003e\u003cp\u003eA lentiviral vector library encoding 5,080 sgRNAs targeting 496 epigenetic modifier genes was purchased from Transomic Technologies (CAHV9002). Approximately 1 \u0026times; 10\u003csup\u003e7\u003c/sup\u003e HUDEP-2 cells stably expressing Cas9 were transduced at a multiplicity of infection of \u0026sim;0.3 to minimize the introduction of \u0026gt;\u0026thinsp;1 vector particles and to achieve a 1000-fold library coverage. After 24 hours, transduced cells were selected in puromycin (1 \u0026micro;g/mL) for 2 days, then maintained in expansion medium for 5 days. Erythroid maturation was monitored by flow cytometry using fluorescein isothiocyanate\u0026ndash;conjugated anti-CD235a (clone GA-R2; BD Biosciences) and allophycocyanin-conjugated anti-Band3 (gift from Xiuli An, New York Blood Center, New York, NY). Band3\u003csup\u003ehigh\u003c/sup\u003e and Band3\u003csup\u003elow\u003c/sup\u003e cell populations from the CD235a\u003csup\u003e+\u003c/sup\u003e cell fractions were purified by fluorescence-activated cell sorting. The representation of lentiviral vector-encoded sgRNAs in Band3\u003csup\u003eHigh\u003c/sup\u003e and Band3\u003csup\u003eLow\u003c/sup\u003e populations was determined by next-generation sequencing and compared.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eData analysis of CRISPR screening\u003c/h2\u003e\u003cp\u003eThe raw FASTQ data obtained after HiSeq sequencing were demultiplexed and mapped to the original reference sgRNA library for data analysis. The read counts for each sgRNA were normalized against the total read counts across all samples. The differentially enriched sgRNAs were defined by comparing normalized counts between sorted cells in the top 10% and those in the bottom 10% of Band3\u003csup\u003elow\u003c/sup\u003e expressing bulk populations. Two independent screenings were performed with the HUDEP-2 cell line stably expressing Cas9. The sgRNA rank was displayed based on a P-value and log\u003csub\u003e2\u003c/sub\u003e fold change by DESeq2. The gene ranking analysis for significant genes and related GO (gene ontology) analysis were conducted using the MAGeCK, MAGeCK Flute, and EnrichR\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eGeneration of a GFP-P2A-miniAID-KMT2D endogenous knock-in cell lines\u003c/h2\u003e\u003cp\u003eFor GFP-P2A-miniAID-KMT2D knock-in delivery, 500 ng of the donor plasmid and sgRNA/Cas9 ribonucleoprotein complexes (RNPs) were used for 2 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e cells. Both K562 and HUDEP-2 cells were electroporated with Neon NxT (Invitrogen MPK5000S). Seventy-two hours after electroporation, the cells were sorted for the expression of the GFP fluorescent marker to enrich the knock-in cell population. After the sorted cells recovered in culture for up to 3 weeks, a second round of sorting was performed to select GFP\u003csup\u003e+\u003c/sup\u003e cells for successful knock-in events, followed by clonal generation.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003eCRISPR-Cas9-mediated gene targeting and vector construction\u003c/h2\u003e\u003cp\u003eThe Cas9-expression vector, Lenti-Cas9-Blast, was purchased from Addgene (#52,962). Cas9 protein was introduced to human K562 and HUDEP-2 cell lines by lentiviral transduction and selected with 10 \u0026micro;g/mL blasticidin (Gibco, A11139-03) to generate Cas9-stable cell lines. The sgRNA sequences were selected from the CRISPR library or using a CRISPR design tool (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.crisprscan.org/\u003c/span\u003e\u003cspan address=\"http://www.crisprscan.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and generated as oligonucleotide pairs. After annealing, constructs were cloned into the Lenti-guide-puro vector (a sgRNA-expression vector), which encodes the sgRNA. Cells were transduced with sgRNA lentivirus and puromycin (Gibco, A11138-03) selection for 48 h after transduction. The oligonucleotides encoding sgRNAs are listed in \u003cb\u003eSupplementary table 10\u003c/b\u003e. For the endogenous mini-AID-KMT2D (auxin-induced degron) system, the cDNA of KMT2D were cloned into the pGL3 vector using the Gibson Assembly. The KMT2D-N-terminal-mini-AID cassette was PCR amplified from the previous established pGL3 vector with AID-GFP fragment. For the GATA1 Tet-on system, the cDNA of GATA1 were cloned into the pGL3 vector using the Gibson Assembly. The Tet-on system cassette was PCR amplified from the previous established Tet-on system vector. SnapGene software was used to design all primers used for cloning. The PCR amplifications of products for cloning were performed using PrimeSTAR Max Premix (TaKaRa, R045). A Gibson Assembly Cloning Kit (Abclonal, RK21020) was used in accordance with the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003eShRNA-mediated gene knockdown\u003c/h2\u003e\u003cp\u003eOligonucleotides encoding short hairpin RNA (shRNA) constructs were designed by the RNAi Consortium of the Broad Institute, obtained from IDT, and cloned into the lentiviral vector pLKO.1-PURO, and luciferase shRNA-encoding pLKO.1-NT (shNT, non-targeting) was used as a control. Cells were transduced with shRNA lentivirus and then selected with 2 \u0026micro;g/ml puromycin for 48 h. The oligonucleotides encoding shRNAs are listed in \u003cb\u003eSupplementary table 10\u003c/b\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\u003ch2\u003eVirus production and transduction\u003c/h2\u003e\u003cp\u003eLentivirus was produced in HEK293T cells by transfecting lentiviral plasmids with helper packaging plasmids (VSVG and psPAX2) using the polyethyleneimine (PEI 40000; MAOKANGBIO, 49,553\u0026ndash;93\u0026thinsp;\u0026minus;\u0026thinsp;7) transfection reagent. HEK293T cells were plated in 10-cm culture dishes and were transfected when confluency reached\u0026thinsp;~\u0026thinsp;80\u0026ndash;90%. For one 10-cm dish of HEK293T cells, 12 \u0026micro;g of plasmid DNA, 4 \u0026micro;g of pVSVG and 8 \u0026micro;g psPAX2, and 96 \u0026micro;L of 1 mg/mL PEI were mixed, incubated at room temperature for 20 min, and then added to the cells. The fresh medium was changed 6\u0026ndash;8 h post-transfection. Lentivirus soup was collected at 48h and 72h post-transfection. The collected virus was filtered through a 0.45-\u0026micro;M non-pyrogenic filter.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\u003ch2\u003eKMT2D editing in normal donors derived CD34\u003csup\u003e+\u003c/sup\u003e HSPCs and HUDEP-2 cell\u003c/h2\u003e\u003cp\u003eCD34\u003csup\u003e+\u003c/sup\u003e HSPCs were electroporated using the Neon\u0026trade; NxT Electroporation (Neon) with the Neon\u0026trade; Transfection Kit (Invitrogen by Thermo Fisher Scientific, MPK1096B). The electroporation program is 1600 V, 10 ms, and 3 pulse. HUDEP-2 were electroporated using electroporation program is 1200 V, 40 ms, and 1 pulse. For ribonucleoprotein (RNP) complex delivery, 50 \u0026micro;M of KMT2D sgRNA#1/2 (synthesized by GenScript) and 5 \u0026micro;g of Cas9 protein (Integrated DNA Technologies, 1,081,058) (molar ratio of sgRNA:Cas9 is 1:3) were used for 0.2\u0026nbsp;million CD34\u003csup\u003e+\u003c/sup\u003e HSPCs. Nontarget sgRNA (NT) was used as a negative control. After 24 h of electroporation, cells were washed using PBS and transferred to a fresh growth medium. After 48 h of electroporation, the cells will be used for functional experiments and knockout efficiency assessment.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\u003ch2\u003eQuantitative PCR (qPCR) analysis for gene expression\u003c/h2\u003e\u003cp\u003eTotal RNA was extracted from K562 and HUDEP-2 cells using the FastPure Cell/Tissue Total RNA Isolation Kit (Vazyme, RC112-01). cDNA was generated using PrimeScript\u0026trade; RT Master Mix (Takara, RR036Q) from 1 \u0026micro;g RNA and diluted 1:200 for qPCR analysis. qPCR was performed using 1 \u0026micro;L diluted cDNA with biological and technical replicates using SYBR Green Master Mix (Vazyme, Q511-03) with QuantStudio 6 real-time PCR system, and results were normalized to the expression of ACTB. Primer sequences utilized for qPCR are in \u003cb\u003eSupplementary table 10\u003c/b\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\u003ch2\u003eImmunoblotting\u003c/h2\u003e\u003cp\u003eCells were washed with PBS and lysed in Cell Lysis Buffer for RIPA (Beyotime, # P0013B) with a protease inhibitor cocktail. Lysates were heated to 95\u0026deg;C in SDS sample buffer, separated by SDS-PAGE, and transferred to a nitrocellulose membrane. Membranes were blocked in 5% nonfat milk in PBS with 0.1% Tween-20, probed with indicated primary antibodies, and followed by incubation with HRP-coupled secondary antibody for 1 h at room temperature. Blots were visualized using enhanced chemiluminescence detection reagents and exposed to X-ray film. All antibodies used in this study are listed in \u003cb\u003eSupplementary table 10\u003c/b\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\u003ch2\u003eNucleoprotein extraction and identification of KMT2D-interacting proteins\u003c/h2\u003e\u003cp\u003eCollect 20M cells from each group of samples and use the nuclear and cytoplasmic extraction kit (Abbkine, #KTP3001) to collect nuclear extracts. Briefly, cells were harvested and then centrifuged at 500 g for 5 min. After washing with ice-cold 1\u0026times; PBS, 200 \u0026micro;L of ice-cold hypotonic buffer [10 mM HEPES, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol (DTT), 1 mM EDTA, pH 7.9] containing 1\u0026times; protease inhibitor cocktail (Sigma-Aldrich, P8849) was added to burst the cell pellet, and incubation was carried out on ice for 10 min. Then, 11 \u0026micro;L of ice-cold detergent (0.05% NP-40) was added, and the incubation was prolonged on ice for an additional 1 min. After centrifugation (5 min at 16,000 g), the supernatant (cytoplasmic extract) was transferred to a pre-chilled tube while the pellet fraction containing the nuclei was suspended in 100 \u0026micro;L of ice-cold nuclear extraction buffer (5 mM HEPES, 1.5 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 300 mM NaCl, 0.2 mM EDTA, 0.5 mM DTT, 26% glycerol, pH 7.9) containing 1\u0026times; protease inhibitor cocktail. The samples were placed on ice and vortexed for 15 s every 10 min for a total of 40 min. After centrifugation (10 min at 16, 000 g), the supernatant (nuclear extract) was collected into a pre-chilled tube. The collected samples were sent to PTM Biolabs Inc for proteomic sequencing and WB validation.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec22\" class=\"Section2\"\u003e\u003ch2\u003eRNA-seq sample preparation and analysis\u003c/h2\u003e\u003cp\u003eRNA samples of two biological replicates were extracted from cultured cells, using TRIzol (Ambion by Life technologies, 343,911), following the manufacturer\u0026rsquo;s instructions. RNA was then sent out for library preparation and next-generation sequencing to a commercial company, Novogene (CA, USA). Raw counts of gene transcripts were derived from raw fastq files using the alignment-independent quantification tool, Salmon (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://combine-\u003c/span\u003e\u003cspan address=\"https://combine-\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e lab.github.io/salmon/) with standard settings. The raw count matrix was then imported into RStudio and utilized as input for Limma-voom analysis following the vignette of the package for normalization, differential gene expression analysis, and unbiased clustering analysis, including principal component analysis. The output of Limma-voom was used as the input for pre-ranked-based GSEA to enrich functional pathways and gene signatures. The RNA-seq datasets during primary erythropoiesis from the normal CD34\u003csup\u003e+\u003c/sup\u003e HSPCs were available under accession GSE53983\u003csup\u003e58\u003c/sup\u003e.\u003c/p\u003e\u003cdiv id=\"Sec23\" class=\"Section3\"\u003e\u003ch2\u003eChIP-seq sample preparation and data analysis\u003c/h2\u003e\u003cp\u003eChIP experiments were performed as previously described with at least two biological replicates for each study\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. First, 2 \u0026times; 10\u003csup\u003e7\u003c/sup\u003e HUDEP-2 cells were suspended in 50 mL of PBS and processed according to previous detailed method. Two percent of the mixture was set aside as input. For each ChIP, chromatin from 2 \u0026times; 10\u003csup\u003e7\u003c/sup\u003e cells was mixed with 20 ng spike-in chromatin (Active Motif, 53083) and incubated with 2\u0026ndash;8 \u0026micro;g primary target antibody and 2 \u0026micro;g spike-in antibody (Active Motif, 61686) overnight at 4\u0026deg;C.\u003c/p\u003e\u003cp\u003eFor ChIP\u0026ndash;seq data analysis, raw sequencing data were aligned to the human genome GRCh37 and the drosophila genome dm6 using Bowtie2\u003csup\u003e60\u003c/sup\u003e. For ChIP\u0026ndash;seq of KMT2D, the window size of 50 bp, the gap size of 50 bp and the false discovery rate (FDR) threshold of 0.05 were used by SICER\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. For ChIP\u0026ndash;seq of histone modifications (H3K4me1 and H3K27ac), the window size of 200 bp, the gap size of 200 bp and the FDR threshold of 10\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e were used. Reads on indicated regions were collected to calculate reads per kilobase million as a measure of signal intensity.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Sec24\" class=\"Section2\"\u003e\u003ch2\u003eATAC-seq sample preparation and data analysis\u003c/h2\u003e\u003cp\u003eThe ATAC-seq library was prepared according to the published omni-ATAC protocol\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. For the HUDEP-2 cells, 50,000 live cells were used per sample. After centrifugation at 500 rpm for 5 min at 4\u0026deg;C (Eppendorf 5417R refrigerated centrifuge), the cell pellets were resuspended in cold lysis buffer supplemented with protease inhibitors (10 mM Tris with a pH 7.4, 10 mM NaCl, 3 mM MgCl\u003csub\u003e2\u003c/sub\u003e, and 0.1% IGEPAL), followed by centrifugation. The pellets were resuspended in 25 \u0026micro;L of tagment DNA buffer (Nextera, FC-121\u0026ndash;1030) and used directly in the transposition reaction. Nextera Tn5 (Nextera, FC-121\u0026ndash;1030) was added to the resuspended nuclei, and the transposition reaction mixture was incubated at 37\u0026deg;C for 30 min. After transposition, the DNA was purified using a Qiagen MinElute PCR purification kit (Qiagen, 28,004). Indexing PCR was conducted for 12 cycles using the NEB Next HiFi 2X PCR Master Mix (NEB, M0541S) and indexing primers. The PCR products were purified at a 1:3 ratio of Agencourt AMPure XP beads (Beckman Coulter, A63881). Libraries were paired-end 100-bp sequenced using an Illumina HiSeq 4000 system. The ATAC-seq datasets were analyzed via the previous methods\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. The publicly available ATAC-seq datasets during primary erythropoiesis from the normal CD34\u003csup\u003e+\u003c/sup\u003e HSPCs were obtained from the GSE128266\u003csup\u003e63\u003c/sup\u003e. First, DeepTools (v3.5.6) was used to plot the average heatmap of peak signals in 10-bp bins. Subsequently, boxplot analysis was conducted using R. The P value was determined by the Wilcoxon test.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting interests\u003c/h2\u003e\u003cp\u003eM.J.W. is a consultant for Glaxo SmithKline, Cellarity Inc, Graphite Bio, Fulcrum Therapeutics and Dyne Therapeutics, and owns equity in Cellarity Inc. Dr. Li Cheng is currently affiliated with a commercial company (GenAssist Therapeutics Incorporation, Suzhou) and declares that she has no competing interests. Dr. Chunliang Li is currently an editorial board member at Genome Biology. The other authors declare that they have no other competing interests.\u003c/p\u003e\u003ch2\u003eAuthor contributions\u003c/h2\u003e\u003cp\u003eP.X., and J.-X. Z. planned the experimental design, analyzed data, draw the Fig. s and wrote the manuscript; J.-X. Z., Y. X. and L.C. performed RNA-seq, ATAC-Seq, and ChIP-seq with the help from Y. X., and M. -L.Z.; J.-X. Z. established the AID2 model for KMT2D; Y. X., L.C. performed bioinformatics analyses under the guidance of B.-S.X., C.-L. L., and P.X.; R. -P. F., X.-H. Q. helped to set up the KMT2D ChIP-seq under the guidance of H.-M. H., Y. C., and M.J.W.. The whole project administration, supervision and funding acquisition: P.X.. The authors have read, discussed the results and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgments\u003c/h2\u003e\u003cp\u003eRyo Kurita and Yukio Nakamura (Cell Engineering Division, RIKEN BioResource Center, Tsukuba, Japan) provided HUDEP-2 cells. Xiuli An (Laboratory of Membrane Biology, New York Blood Center) provided the anti-Band3 antibody. We thank the insightful discussion and comments of members from the Xu lab. This research was supported by the National Natural Science Foundation of China (82170119), by Jiangsu Province National Science and Technology grant BK20243008, by the High-Level Personnel Project of Jiangsu Province (JSSCTD202353), by Interdisciplinary Basic Frontier Innovation Program of Suzhou Medical College of Soochow University, The Pediatric Hematology\u0026amp; Oncology Key Laboratory of Higher Education Institutions in Jiangsu Province, by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) and Collaborative Innovation Center of Hematology (all to P.X.).\u003c/p\u003e\u003cdiv id=\"Sec25\" class=\"Section3\"\u003e\u003ch2\u003eData availability\u003c/h2\u003e\u003cp\u003eAll data are available in the main text or the Supplementary materials. The RNA-seq and ChIP-seq raw datasets generated from this study were deposited to GEO under the accession number: GSE293567 (reviewer token: sjmtooiwlbepbej)\u003csup\u003e\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. The H3K4me1/3, H3K27ac and GATA1 ChIP-seq datasets in HUDEP-2 cells were obtained from the GSE115357\u003csup\u003e3\u003c/sup\u003e, while the TAL1 and KLF1 ChIP-seq datasets in HUDEP-2 cells were obtained from the GSE157311\u003csup\u003e40\u003c/sup\u003e. The ATAC-seq and RNA-seq datasets during primary erythropoiesis from the normal CD34\u003csup\u003e+\u003c/sup\u003e HSPCs were obtained from the GSE128266\u003csup\u003e63\u003c/sup\u003e and GSE53983\u003csup\u003e58\u003c/sup\u003e, respectively. Code repositories collected at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.6084/m9.figshare.c.6186670\u003c/span\u003e\u003cspan address=\"10.6084/m9.figshare.c.6186670\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003csup\u003e65\u003c/sup\u003e.\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCramer P (2019) Organization and regulation of gene transcription. Nature 573:45\u0026ndash;54\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStadhouders R, Filion GJ, Graf T (2019) Transcription factors and 3D genome conformation in cell-fate decisions. Nature 569:345\u0026ndash;354\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu P et al (2021) FBXO11-mediated proteolysis of BAHD1 relieves PRC2-dependent transcriptional repression in erythropoiesis. Blood 137:155\u0026ndash;167\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKarayel \u0026Ouml; et al (2020) Integrative proteomics reveals principles of dynamic phosphosignaling networks in human erythropoiesis. Mol Syst Biol 16:e9813\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAn X, Schulz VP, Mohandas N, Gallagher PG (2015) Human and murine erythropoiesis. Curr Opin Hematol 22:206\u0026ndash;211\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNandakumar SK, Ulirsch JC, Sankaran V (2016) G. Advances in understanding erythropoiesis: evolving perspectives. Br J Haematol 173:206\u0026ndash;218\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSankaran VG, Weiss MJ (2015) Anemia: progress in molecular mechanisms and therapies. Nat Med 21:221\u0026ndash;230\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePop R et al (2010) A Key Commitment Step in Erythropoiesis Is Synchronized with the Cell Cycle Clock through Mutual Inhibition between PU.1 and S-Phase Progression. PLoS Biol 8:e1000484\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNguyen AT et al (2017) UBE2O remodels the proteome during terminal erythroid differentiation. Sci (80-). 357\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWadman IA et al (1997) The LIM-only protein Lmo2 is a bridging molecule assembling an erythroid, DNA-binding complex which includes the TAL1, E47, GATA-1 and Ldb1/NLI proteins. EMBO J 16:3145\u0026ndash;3157\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTallack MR et al (2010) A global role for KLF1 in erythropoiesis revealed by ChIP-seq in primary erythroid cells. Genome Res 20:1052\u0026ndash;1063\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOrkin SH (1995) Transcription factors and hematopoietic development. J Biol Chem 270:4955\u0026ndash;4958\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheng Y et al (2009) Erythroid GATA1 function revealed by genome-wide analysis of transcription factor occupancy, histone modifications, and mRNA expression. Genome Res 19:2172\u0026ndash;2184\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eOudelaar AM, Higgs DR (2021) The relationship between genome structure and function. Nat Rev Genet 22:154\u0026ndash;168\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMurphy ZC et al (2021) Regulation of RNA polymerase II activity is essential for terminal erythroid maturation. Blood 138:1740\u0026ndash;1756\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHeintzman ND et al (2009) Histone modifications at human enhancers reflect global cell-type-specific gene expression. Nature 459:108\u0026ndash;112\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMiller T et al (2001) COMPASS: A complex of proteins associated with a trithorax-related SET domain protein. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e 98, 12902\u0026ndash;12907\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang C et al (2016) Enhancer priming by H3K4 methyltransferase MLL4 controls cell fate transition. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 113, 11871\u0026ndash;11876\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGe K (2019) Enhancer regulation by H3K4 methyltransferases MLL3/MLL4. FASEB J 33\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang SP et al (2017) A UTX-MLL4-p300 Transcriptional Regulatory Network Coordinately Shapes Active Enhancer Landscapes for Eliciting Transcription. Mol Cell 67:308\u0026ndash;321e6\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoileau RM, Chen KX, Blelloch R (2023) Loss of MLL3/4 decouples enhancer H3K4 monomethylation, H3K27 acetylation, and gene activation during embryonic stem cell differentiation. Genome Biol 24:41\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang Z, Ren B (2024) Role of H3K4 monomethylation in gene regulation. Curr Opin Genet Dev 84:102153\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMorgan MAJ, Shilatifard A (2020) Reevaluating the roles of histone-modifying enzymes and their associated chromatin modifications in transcriptional regulation. Nat Genet 52:1271\u0026ndash;1281\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCenik BK, Shilatifard A (2021) COMPASS and SWI/SNF complexes in development and disease. Nat Rev Genet 22:38\u0026ndash;58\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYan H et al (2017) Distinct roles for TET family proteins in regulating human erythropoiesis. Blood 129:2002\u0026ndash;2012\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang Y et al (2021) Impairment of human terminal erythroid differentiation by histone deacetylase 5 deficiency. Blood 138:1615\u0026ndash;1627\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi M et al (2023) Stage-specific dual function: EZH2 regulates human erythropoiesis by eliciting histone and non-histone methylation. Haematologica 108:2487\u0026ndash;2502\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMalik J, Getman M, Steiner LA (2015) Histone Methyltransferase Setd8 Represses Gata2 Expression and Regulates Erythroid Maturation. Mol Cell Biol 35:2059\u0026ndash;2072\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMalik J, Lillis JA, Couch T, Getman M, Steiner LA (2017) The Methyltransferase Setd8 Is Essential for Erythroblast Survival and Maturation. Cell Rep 21:2376\u0026ndash;2383\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMyers JA et al (2020) The histone methyltransferase Setd8 alters the chromatin landscape and regulates the expression of key transcription factors during erythroid differentiation. Epigenetics Chromatin 13\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKurita R et al (2013) Establishment of immortalized human erythroid progenitor cell lines able to produce enucleated red blood cells. PLoS ONE 8:e59890\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAndersson LC, Jokinen M, Gahmberg CG (1979) Induction of erythroid differentiation in the human leukaemia cell line K562. Nature 278:364\u0026ndash;365\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKuleshov MV et al (2016) Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 44:W90\u0026ndash;W97\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMatys V (2006) TRANSFAC(R) and its module TRANSCompel(R): transcriptional gene regulation in eukaryotes. Nucleic Acids Res 34:D108\u0026ndash;D110\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHyle J et al (2023) Auxin-inducible degron 2 system deciphers functions of CTCF domains in transcriptional regulation. Genome Biol 24:1\u0026ndash;30\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYesbolatova A et al (2020) The auxin-inducible degron 2 technology provides sharp degradation control in yeast, mammalian cells, and mice. Nat Commun 11:5701\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee J-E et al (2017) Brd4 binds to active enhancers to control cell identity gene induction in adipogenesis and myogenesis. Nat Commun 8:2217\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYan J et al (2018) Histone H3 lysine 4 monomethylation modulates long-range chromatin interactions at enhancers. Cell Res 28:204\u0026ndash;220\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNabet B et al (2018) The dTAG system for immediate and target-specific protein degradation. Nat Chem Biol 14:431\u0026ndash;441\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheng L et al (2021) Single-nucleotide-level mapping of DNA regulatory elements that control fetal hemoglobin expression. Nat Genet 53:869\u0026ndash;880\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGiani FC et al (2016) Targeted Application of Human Genetic Variation Can Improve Red Blood Cell Production from Stem Cells. Cell Stem Cell 18:73\u0026ndash;78\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDorighi KM et al (2017) Mll3 and Mll4 Facilitate Enhancer RNA Synthesis and Transcription from Promoters Independently of H3K4 Monomethylation. Mol Cell 66:568\u0026ndash;576e4\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKadauke S et al (2012) Tissue-specific mitotic bookmarking by hematopoietic transcription factor GATA1. Cell 150:725\u0026ndash;737\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKubo N et al (2024) H3K4me1 facilitates promoter-enhancer interactions and gene activation during embryonic stem cell differentiation. Mol Cell 84:1742\u0026ndash;1752e5\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBoyes J, Byfield P, Nakatani Y, Ogryzko V (1998) Regulation of activity of the transcription factor GATA-1 by acetylation. Nature 396:594\u0026ndash;598\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKim S, Il, Bultman SJ, Kiefer CM, Dean A, Bresnick EH (2009) BRG1 requirement for long-range interaction of a locus control region with a downstream promoter. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e 106\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLetting DL, Rakowski C, Weiss MJ, Blobel GA (2003) Formation of a Tissue-Specific Histone Acetylation Pattern by the Hematopoietic Transcription Factor GATA-1. Mol Cell Biol 23\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStumpf M et al (2006) The mediator complex functions as a coactivator for GATA-1 in erythropoiesis via subunit Med1/TRAP220. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e 103\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eG\u0026ouml;\u0026ouml;s H et al (2022) Human transcription factor protein interaction networks. Nat Commun 13:766\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrass JA et al (2003) GATA-1-dependent transcriptional repression of GATA-2 via disruption of positive autoregulation and domain-wide chromatin remodeling. \u003cem\u003eProc. Natl. Acad. Sci. U. S. A.\u003c/em\u003e 100, 8811\u0026ndash;8816\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBresnick EH, Lee H-YY, Fujiwara T, Johnson KD, Keles S (2010) GATA switches as developmental drivers. J Biol Chem 285:31087\u0026ndash;31093\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLara-Astiaso D et al (2023) In vivo screening characterizes chromatin factor functions during normal and malignant hematopoiesis. Nat Genet 55:1542\u0026ndash;1554\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHerz HM et al (2012) Enhancer-associated H3K4 monomethylation by trithorax-related, the drosophila homolog of mammalian MLL3/MLL4. Genes Dev 26:2604\u0026ndash;2620\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHu D et al (2013) The MLL3/MLL4 Branches of the COMPASS Family Function as Major Histone H3K4 Monomethylases at Enhancers. Mol Cell Biol 33:4745\u0026ndash;4754\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDhar SS et al (2012) Trans-tail regulation of MLL4-catalyzed H3K4 methylation by H4R3 symmetric dimethylation is mediated by a tandem PHD of MLL4. Genes Dev 26:2749\u0026ndash;2762\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDhar SS et al (2018) MLL4 Is Required to Maintain Broad H3K4me3 Peaks and Super-Enhancers at Tumor Suppressor Genes. Mol Cell 70:825\u0026ndash;841\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVan HT, Xie G, Dong P, Liu Z, Ge K (2024) KMT2 Family of H3K4 Methyltransferases: Enzymatic Activity-dependent and -independent Functions. J Mol Biol 436:168453\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAn X et al (2014) Global transcriptome analyses of human and murine terminal erythroid differentiation. Blood 123:3466\u0026ndash;3477\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLandt SG et al (2012) ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia. Genome Res 22:1813\u0026ndash;1831\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLangmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZang C et al (2009) A clustering approach for identification of enriched domains from histone modification ChIP-Seq data. Bioinformatics 25:1952\u0026ndash;1958\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCorces MR et al (2017) An improved ATAC-seq protocol reduces background and enables interrogation of frozen tissues. Nat Methods 14:959\u0026ndash;962\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchulz VP et al (2019) A Unique Epigenomic Landscape Defines Human Erythropoiesis. Cell Rep 28:2996\u0026ndash;3009e7\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang J, Xin Y, Cheng L, Yang X, Xing Y, Xu B, Li C (2025) X. P. The H3K4 methyltransferase KMT2D is an essential cofactor for GATA1 at erythroid gene enhancers.\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE293567\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE293567\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE293567 (2025)\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu B, Djekidel MN, Li C (2022) W. J. St Jude Center for Applied Bioinformatics General Pipelines. figshare. Collection. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.6084/m9.figshare.c.6186670\u003c/span\u003e\u003cspan address=\"10.6084/m9.figshare.c.6186670\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"KMT2D, GATA1, enhancer, transcription regulation, hematopoiesis, erythropoiesis, epigenetics","lastPublishedDoi":"10.21203/rs.3.rs-7853319/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7853319/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGene expression during cellular differentiation is coordinated by combinatorial interactions between transcription factors (TFs) and cofactors at promoters and enhancers. The \u0026ldquo;master TF\u0026rdquo; GATA1 coordinates gene transcription in a subset of hematopoietic lineages, including erythroid, megakaryocytic, mast, and eosinophil, while repressing the development of other blood lineages. However, the specific cofactors required for GATA1-activated gene expression during hematopoiesis are incompletely defined. We identified the cofactor KMT2D, an H3K4 methyltransferase that collaborates with H3K27 acetyltransferases to activate transcription, in an unbiased CRISPR/Cas9 screen for epigenetic regulators of erythropoiesis. Loss of KMT2D in human erythroid precursors caused developmental arrest with impaired expression of numerous erythroid genes. Mechanistically, KMT2D colocalized with GATA1 on more than one thousand erythroid enhancers associated with over two hundred erythroid genes. In general, co-occupancy of GATA1 and KMT2D at erythroid enhancers was associated with stronger transcriptional activity than occupancy by GATA1 alone. Acute depletion of KMT2D in erythroid precursors caused rapid reductions of H3K4me1 and H3K27ac on a subset of GATA1-bound enhancers and impaired the expression of canonical erythroid genes, including \u003cem\u003eZFPM1, SLC4A1\u003c/em\u003e, and \u003cem\u003eEPOR\u003c/em\u003e. Moreover, acute depletion of GATA1 or KMT2D individually caused downregulation of overlapping gene sets. Thus, KMT2D controls erythropoiesis by selectively activating GATA1-dependent erythroid enhancers. Our studies identify KMT2D as a novel cofactor for transcriptional activation by GATA1 during erythropoiesis. More generally, our findings demonstrate how a lineage-specific TF cooperates with a ubiquitous epigenic regulator to drive lineage-specific gene expression during cellular differentiation.\u003c/p\u003e","manuscriptTitle":"The H3K4 methyltransferase KMT2D is an essential cofactor for GATA1 at erythroid gene enhancers","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-22 10:07:07","doi":"10.21203/rs.3.rs-7853319/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"e012c3b1-edb7-4756-8a4c-76fcf366679c","owner":[],"postedDate":"October 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":56649553,"name":"Biological sciences/Genetics/Gene regulation"},{"id":56649554,"name":"Biological sciences/Developmental biology/Haematopoiesis/Erythropoiesis/Haematopoietic stem cells"},{"id":56649555,"name":"Biological sciences/Molecular biology/Epigenetics"}],"tags":[],"updatedAt":"2026-04-11T02:15:17+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-22 10:07:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7853319","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7853319","identity":"rs-7853319","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}