Advertisement

Evolving insights on histone methylome regulation in human acute myeloid leukemia pathogenesis and targeted therapy

  • Liberalis Debraj Boila
    Affiliations
    Stem Cell & Leukemia Laboratory, Cancer Biology & Inflammatory Disorder Division, CSIR–Indian Institute of Chemical Biology, Jadavpur, Kolkata, West Bengal, India

    Translational Research Unit of Excellence, Salt Lake, Kolkata, West Bengal, India
    Search for articles by this author
  • Amitava Sengupta
    Correspondence
    Offprint requests to: Amitava Sengupta, Cancer Biology & Inflammatory Disorder Division, Indian Institute of Chemical Biology (CSIR-IICB), CN-6, Sector V, Salt Lake, Kolkata, West Bengal 700091, India
    Affiliations
    Stem Cell & Leukemia Laboratory, Cancer Biology & Inflammatory Disorder Division, CSIR–Indian Institute of Chemical Biology, Jadavpur, Kolkata, West Bengal, India

    Translational Research Unit of Excellence, Salt Lake, Kolkata, West Bengal, India
    Search for articles by this author
Open ArchivePublished:September 17, 2020DOI:https://doi.org/10.1016/j.exphem.2020.09.189

      Highlights

      • Histone methylome regulates chromatin accessibility and transcriptional output.
      • Aberrant histone methylation in AML is associated with epigenetic plasticity.
      • Derangements in histone methyl-editing enzymes are implicated in AML pathogenesis.
      • Histone methylation can influence chemotherapy response and AML-immune ecosystem.
      • Immuno-epigenetic reprogramming shows significant promise for AML-targeted therapy.
      Acute myeloid leukemia (AML) is an aggressive, disseminated hematological malignancy associated with clonal selection of aberrant self-renewing hematopoietic stem cells and progenitors and poorly differentiated myeloid blasts. The most prevalent form of leukemia in adults, AML is predominantly an age-related disorder and accounts for more than 10,000 deaths per year in the United States alone. In comparison to solid tumors, AML has an overall low mutational burden, albeit more than 70% of AML patients harbor somatic mutations in genes encoding epigenetic modifiers and chromatin regulators. In the past decade, discoveries highlighting the role of DNA and histone modifications in determining cellular plasticity and lineage commitment have attested to the importance of epigenetic contributions to tumor cell de-differentiation and heterogeneity, tumor initiation, maintenance, and relapse. Orchestration in histone methylation levels regulates pluripotency and multicellular development. The increasing number of reversible methylation regulators being identified, including histone methylation writer, reader, and eraser enzymes, and their implications in AML pathogenesis have widened the scope of epigenetic reprogramming, with multiple drugs currently in various stages of preclinical and clinical trials. AML methylome also determines response to conventional chemotherapy, as well as AML cell interaction within a tumor-immune microenvironment ecosystem. Here we summarize the latest developments focusing on molecular derangements in histone methyltransferases (HMTs) and histone demethylases (HDMs) in AML pathogenesis. AML-associated HMTs and HDMs, through intricate crosstalk mechanisms, maintain an altered histone methylation code conducive to disease progression. We further discuss their importance in governing response to therapy, which can be used as a biomarker for treatment efficacy. Finally we deliberate on the therapeutic potential of targeting aberrant histone methylome in AML, examine available small molecule inhibitors in combination with immunomodulating therapeutic approaches and caveats, and discuss how future studies can enable posited epigenome-based targeted therapy to become a mainstay for AML treatment.

      Graphical abstract

      Acute myeloid leukemia (AML) is characterized by uncontrolled proliferation of clonal hematopoietic precursor cells. This interrupts normal hematopoiesis and may lead to bone marrow failure. It occurs predominantly in older adults, with the average age at diagnosis being 68 years [
      • Appelbaum FR
      • Gundacker H
      • Head DR
      • et al.
      Age and acute myeloid leukemia.
      ]. Intensive chemotherapy combined with hematopoietic stem cell (HSC) transplantation has considerably improved outcomes in younger adults. However, about 80% of older adults still succumb to the disease or to the associated therapeutic toxicity [
      • Versluis J
      • Hazenberg CL
      • Passweg JR
      • et al.
      Post-remission treatment with allogeneic stem cell transplantation in patients aged 60 years and older with acute myeloid leukaemia: a time-dependent analysis.
      ]. Recent advancements in molecular and cytogenetic analyses have helped identify genetic abnormalities that contribute to AML initiation and maintenance [
      • Welch JS
      • Ley TJ
      • Link DC
      • et al.
      The origin and evolution of mutations in acute myeloid leukemia.
      ]. Approximately 55% of AML cases harbor recurrent chromosomal translocations, which have been considered one of the most important prognostic factors for clinical outcome prediction. However, the remaining 45% cases harbor normal karyotypes, indicating that chromosomal rearrangement is not the only determinant of the disease. Compared with other cancers, AML has a low mutational load, but is highly heterogeneous in terms of genetic background and clinical presentation [
      • Osorio FG
      • Rosendahl Huber A
      • Oka R
      • et al.
      Somatic Mutations Reveal Lineage Relationships and Age-Related Mutagenesis in Human Hematopoiesis.
      ]. Though mutations are few, they frequently occur in genes encoding epigenetic regulators [
      • Glass JL
      • Hassane D
      • Wouters BJ
      • et al.
      Epigenetic identity in AML depends on disruption of nonpromoter regulatory elements and is affected by antagonistic effects of mutations in epigenetic modifiers.
      ]. Some of the common examples are DNMT3A, TET2, IDH1, and IDH2 (regulating DNA methylation), CBP and P300 (regulating histone acetylation), and EZH2, ASXL1, and KDMs (regulating histone methylation). Additionally, mutations in the CTCF and cohesin complex have also been identified that regulate three-dimensional chromatin conformation.
      Unlike genetic mutations, which are hardwired, epigenetic modifiers work by transcriptional regulation of their downstream target genes and are reversible. DNA methylation was one of the first characterized epigenetic regulations. In addition to DNA methylation, histone tail modifications, including acetylation, methylation, and ubiquitination, have been identified that significantly contribute to transcriptional output. Many of these modifications are frequently dysregulated in AML. Histone methylation, unlike other epigenetic modifications, was initially considered irreversible. This perception changed dramatically with the discovery of two families of enzymes capable of demethylating histone lysine residues [
      • Shi Y
      • Lan F
      • Matson C
      • et al.
      Histone demethylation mediated by the nuclear amine oxidase homologue LSD1.
      ,
      • Chen Z
      • Zang J
      • Whetstine J
      • et al.
      Structural insights into histone demethylation by JMJD2 family members.
      ]. Thereafter, histone methylation has come to be recognized to play a key role in the initiation and maintenance of several cancers including AML. In this review, we discuss the importance of histone methylation to epigenetic dysregulation in AML. We also try to understand the correlation between histone methylation and clinical outcome and discuss available epigenetic therapies and their limitations and issues, which remain unaddressed.

      HMTs in AML development

      Histone methyltransferases (HMTs) catalyze addition of methyl groups to specific histone residues. Depending on the position and nature of the methylated residues, histone methylation can either promote or repress transcription. In general, methylation on H3K4, H3K36, and H3K79, as well as asymmetric dimethylation of H4R3 activates gene expression, whereas methylation on H3K9, H3K27, and H4K20 and symmetric dimethylation of H4R3 are associated with transcription repression [
      • Zhang Y
      • Reinberg D
      Transcription regulation by histone methylation: interplay between different covalent modifications of the core histone tails.
      ,
      • Greer EL
      • Shi Y
      Histone methylation: a dynamic mark in health, disease and inheritance.
      ].

      MLL1

      MLL1, also known as KMT2A, is a member of the family of SET domain-containing enzymes. It methylates H3K4, resulting in transcription activation. Several studies have reported that Mll1 is required for definitive hematopoiesis, and regulates repopulating ability of both fetal and adult HSCs [
      • McMahon KA
      • Hiew SY
      • Hadjur S
      • et al.
      Mll has a critical role in fetal and adult hematopoietic stem cell self-renewal.
      ,
      • Yagi H
      • Deguchi K
      • Aono A
      • Tani Y
      • Kishimoto T
      • Komori T
      Growth disturbance in fetal liver hematopoiesis of Mll-mutant mice.
      ]. MLL1 is involved in translocations in approximately 5% to 10% of AML cases [
      • Meyer C
      • Burmeister T
      • Groger D
      • et al.
      The MLL recombinome of acute leukemias in 2017.
      ]. MLL1 translocations result in fusion proteins that lack the wild-type SET domain, hence its transformation capacity is not attributed to methyltransferase activity. This is comparable to its function in normal hematopoiesis wherein H4K16ac, through MLL1 interacting partners, is more important than its HMT activity for maintaining expression of downstream targets [
      • Mishra BP
      • Zaffuto KM
      • Artinger EL
      • et al.
      The histone methyltransferase activity of MLL1 is dispensable for hematopoiesis and leukemogenesis.
      ]. MLL1 fusions in leukemia occur with about six common partner genes. These usually encode super-elongation-complex nuclear proteins, such as AF4/6/9/10, ELL, and ENL, with MLL1–AF9 (MLL–AF9) being the most common and accounting for about 30% of all MLL1 fusions in AML [
      • Meyer C
      • Burmeister T
      • Groger D
      • et al.
      The MLL recombinome of acute leukemias in 2017.
      ]. The MLL1 fusion genes arising from these translocations have been characterized as potent oncogenes. Though MLL1 fusions themselves are devoid of HMT activity, several methylation-related enzymes, such as EZH2, LSD1, and DOT1L, have been found to be essential components of this leukemogenic program.
      Additionally, cells harboring MLL fusions may retain a copy of the wild-type MLL1. Several studies have indicated dependence of MLL fusions on the intact MLL1 allele for their functioning. Studies in mouse fibroblasts demonstrated recruitment of MLL–AF9 to the HoxA9 locus is mediated by wild-type Mll1 [
      • Milne TA
      • Kim J
      • Wang GG
      • et al.
      Multiple interactions recruit MLL1 and MLL1 fusion proteins to the HOXA9 locus in leukemogenesis.
      ]. Similarly in MLL–AF9 murine AML cells, menin-mediated recruitment of both wild-type Mll1 and MLL–AF9 fusion is required for Hox gene activation and leukemia progression [
      • Thiel AT
      • Blessington P
      • Zou T
      • et al.
      MLL-AF9-induced leukemogenesis requires coexpression of the wild-type Mll allele.
      ]. However, other studies indicate that HMT activity of wild-type MLL1 is dispensable for hematopoiesis and leukemogenesis of MLL–AF9-driven leukemia [
      • Mishra BP
      • Zaffuto KM
      • Artinger EL
      • et al.
      The histone methyltransferase activity of MLL1 is dispensable for hematopoiesis and leukemogenesis.
      ]. Moreover, not Mll1, but its closest orthologue Mll2, another H3K4 HMT, is required for MLL fusion-driven leukemogenesis [
      • Chen Y
      • Anastassiadis K
      • Kranz A
      • et al.
      MLL2, not MLL1, plays a major role in sustaining MLL-rearranged acute myeloid leukemia.
      ]. In contrast, loss of Mll3 or Mll4 did not influence H3K4 methylation on Hox loci or their expression [
      • Wang P
      • Lin C
      • Smith ER
      • et al.
      Global analysis of H3K4 methylation defines MLL family member targets and points to a role for MLL1-mediated H3K4 methylation in the regulation of transcriptional initiation by RNA polymerase II.
      ]. These contradictory findings indicate cell type specificity, redundancy, and interdependence among the MLL family of genes. The menin–MLL interaction is critical for MLL-fusion binding to target loci [
      • Krivtsov AV
      • Evans K
      • Gadrey JY
      • et al.
      A menin-MLL inhibitor induces specific chromatin changes and eradicates disease in models of MLL-rearranged leukemia.
      ], and this dependency has been exploited to develop drugs that specifically disrupt this interaction. VTP50469, an orally bioavailable small-molecule inhibitor, displaces menin from the MLL-containing protein complex and reduces MLL1 and DOT1L binding to target loci, resulting in downregulation of key genes of the MLL-rearranged (MLL-r) transcriptional program, such as MEIS1, MEF2C, KDM3C, and PBX3 [
      • Krivtsov AV
      • Evans K
      • Gadrey JY
      • et al.
      A menin-MLL inhibitor induces specific chromatin changes and eradicates disease in models of MLL-rearranged leukemia.
      ]. VTP50469 treatment successfully eradicated disease in patient-derived xenograft (PDX) models of MLL-r AML. Menin–MLL interaction plays a key role in both MLL-r and NPM1-mutant (NPM1mut) AMLs, which frequently also harbor activating FLT3 mutations. FLT3 is a downstream target of MEIS1, one of the transcriptional factors downregulated on VTP50469 treatment. Combining menin–MLL inhibition with quizartinib, a potent and highly selective FLT3 inhibitor, reduced FLT3 phosphorylation, suppressed its target genes, and synergistically enhanced apoptosis and differentiation in models of human and murine NPM1mut and MLL-r leukemias harboring an FLT3 mutation [
      • Dzama MM
      • Steiner M
      • Rausch J
      • et al.
      Synergistic targeting of FLT3 mutations in AML via combined menin-MLL and FLT3 inhibition.
      ].
      Since the last decade, researchers around the globe have been trying to address initiation and evolution of AML from premalignant clones. Targeting such clones could effectively lead to preventive therapies. Npm1c/Dnmt3a mutant mice exhibit a period of extended myeloid progenitor cell proliferation and self-renewal before leukemia, presenting a premalignant model of AML development. Menin–MLL inhibition using VTP50469 abrogated self-renewal of these myeloid progenitor cells, suggesting its potential as a preventive therapy [
      • Uckelmann HJ
      • Kim SM
      • Wong EM
      • et al.
      Therapeutic targeting of preleukemia cells in a mouse model of NPM1 mutant acute myeloid leukemia.
      ]. MLL-r AML is associated with increased activation of Rac GTPases, through elevated Frat expression [
      • Walf-Vorderwulbecke V
      • de Boer J
      • Horton SJ
      • et al.
      Frat2 mediates the oncogenic activation of Rac by MLL fusions.
      ]. Frat1 and Frat2 are associated with canonical and noncanonical Wnt signaling respectively. Thus, through Frat upregulation, MLL fusions promote integration of canonical and noncanonical Wnt signaling, which somewhat justifies the paradoxical requirement of both canonical Wnt signaling and GSK3 activity in MLL-r leukemia [
      • Wang Y
      • Krivtsov AV
      • Sinha AU
      • et al.
      The Wnt/beta-catenin pathway is required for the development of leukemia stem cells in AML.
      ]. Another study has shown RUNX1 to be a direct target of MLL–AF4 and critical for its transformation potential [
      • Wilkinson AC
      • Ballabio E
      • Geng H
      • et al.
      RUNX1 is a key target in t(4;11) leukemias that contributes to gene activation through an AF4-MLL complex interaction.
      ].

      PRC2

      Polycomb repressive complex 2 (PRC2), which promotes gene repression, consists of four core subunits: EED, SUZ12, RBBP4, and either of the two catalytic subunits EZH1 and EZH2. Being the more common catalytic subunit, EZH2 regulates expression of numerous genes critical for stem cell renewal by controlling H3K27 methylation at “poised” promoters [
      • Voigt P
      • Tee WW
      • Reinberg D
      A double take on bivalent promoters.
      ]. Ezh2 is essential for fetal, but not adult, HSC function [
      • O'Carroll D
      • Erhardt S
      • Pagani M
      • Barton SC
      • Surani MA
      • Jenuwein T
      The polycomb-group gene Ezh2 is required for early mouse development.
      ]. Ezh2 deletion in adult bone marrow compromises specifically lymphopoiesis [
      • Su IH
      • Basavaraj A
      • Krutchinsky AN
      • et al.
      Ezh2 controls B cell development through histone H3 methylation and Igh rearrangement.
      ]. Unlike Ezh2, which is ubiquitously expressed, Ezh1 is highly expressed in HSCs in the bone marrow, compared with those in the fetal liver, indicating that Ezh1 complements Ezh2 in the adult bone marrow, but not in the fetal liver. EZH2 is frequently mutated in different leukemia types. In contrast to B-cell lymphoma, where EZH2 mutations are gain of function [
      • Morin RD
      • Johnson NA
      • Severson TM
      • et al.
      Somatic mutations altering EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas of germinal-center origin.
      ], the majority of EZH2 mutations found in myeloid disorders are inactivating mutations, resulting in loss of its H3K27 methyltransferase activity [
      • Ernst T
      • Chase AJ
      • Score J
      • et al.
      Inactivating mutations of the histone methyltransferase gene EZH2 in myeloid disorders.
      ]. EZH2 is found in the 7q region, which is frequently deleted in myeloid neoplasms. Deletion of Ezh2 in a mouse model induced MDS-like disease, suggesting tumor suppressor function [
      • Sashida G
      • Harada H
      • Matsui H
      • et al.
      Ezh2 loss promotes development of myelodysplastic syndrome but attenuates its predisposition to leukaemic transformation.
      ]. However, a recent report indicated mutations in EZH2 rather exert stage-dependent and opposing effects on AML. When Ezh2 was deleted before transformation with MLL–AF9 or AML1–ETO9a oncogenes, it accelerated AML progression and shortened survival, indicating a tumor suppressor role. In contrast, when Ezh2 was deleted in secondary recipients during the maintenance phase of AML, disease severity was attenuated and survival enhanced [
      • Basheer F
      • Giotopoulos G
      • Meduri E
      • et al.
      Contrasting requirements during disease evolution identify EZH2 as a therapeutic target in AML.
      ]. Other members of the PRC2 complex, including ASXL1, JARID2, and SUZ12, have also been implicated in the development of AML, with most of the mutations in these genes resulting in loss of function of the complex.

      Other HMTs and reader proteins

      Apart from MLL and PRC2, there are methyltransferases regulating methylation at the other common residues, H3K9, H3K36, and H3K79. SUV39H1 methylates H3K9. MECOM, a potent proto-oncogene known to be involved in stem cell self-renewal and leukemogenesis, physically interacts with SUV39H1 and has been implicated in disease progression of AML [
      • Goyama S
      • Nitta E
      • Yoshino T
      • et al.
      EVI-1 interacts with histone methyltransferases SUV39H1 and G9a for transcriptional repression and bone marrow immortalization.
      ]. Loss of G9a, another H3K9 HMT, was reported to suppress leukemogenesis in a mouse model of leukemia induced by HOXA9 [
      • Lehnertz B
      • Pabst C
      • Su L
      • et al.
      The methyltransferase G9a regulates HoxA9-dependent transcription in AML.
      ]. H3K36 trimethylation, which promotes transcription elongation, is catalyzed by SETD2. Homozygous Setd2 deficiency during early hematopoietic development causes pancytopenia, splenomegaly, and overall bone marrow hypocellularity, along with a reduced total number of HSCs [
      • Haihua Chu S
      • Chabon JR
      • Minehart J
      • et al.
      Loss of lysine histone methyltransferase Setd2 disrupts normal hematopoiesis, lineage commitment and reveals a novel role for H3K36me3 in immunoglobulin VDJ recombination.
      ]. SETD2 loss of function mutation is common and has been reported in more than 20% of MLL-r leukemias [
      • Zhu X
      • He F
      • Zeng H
      • et al.
      Identification of functional cooperative mutations of SETD2 in human acute leukemia.
      ]. Like EZH2, SETD2 may have a context-dependent tumor suppressor or oncogenic function. Partial SETD2 loss enhances leukemogenesis and leads to drug resistance [
      • Mar BG
      • Chu SH
      • Kahn JD
      • et al.
      SETD2 alterations impair DNA damage recognition and lead to resistance to chemotherapy in leukemia.
      ], while complete SETD2 loss delays leukemia progression, suggesting a possible gene dosage effect [
      • Mar BG
      • Chu SH
      • Kahn JD
      • et al.
      SETD2 alterations impair DNA damage recognition and lead to resistance to chemotherapy in leukemia.
      ,
      • Skucha A
      • Ebner J
      • Schmollerl J
      • et al.
      MLL-fusion-driven leukemia requires SETD2 to safeguard genomic integrity.
      ].
      CBX7, a chromodomain-containing member of PRC1 recognizing H3K27me3, has been reported to promote self-renewal of human normal and AML stem cells and progenitors, which involves nonhistone protein interactions with H3K9 methyltransferases SETDB1, EHMT1, and EHMT2 [
      • Jung J
      • Buisman SC
      • Weersing E
      • et al.
      CBX7 induces self-renewal of human normal and malignant hematopoietic stem and progenitor cells by canonical and non-canonical interactions.
      ]. Contribution of plant homeodomain (PHD) motif-containing proteins, which are able to read H3K4me3 marks, to AML development has also been suggested [
      • Gough SM
      • Lee F
      • Yang F
      • et al.
      NUP98-PHF23 is a chromatin-modifying oncoprotein that causes a wide array of leukemias sensitive to inhibition of PHD histone reader function.
      ]. Additionally, HMTs and HDMs are erratic for nonhistone targets as well as methylation-independent roles. These complexities warrant further investigation. DOT1L is the only lysine methyltransferase known to be responsible for H3K79 methylation in mammalian systems. Dot1l-null mice are embryonic lethal, with disrupts erythroid development and causes severe anemia [
      • Feng Y
      • Yang Y
      • Ortega MM
      • et al.
      Early mammalian erythropoiesis requires the Dot1L methyltransferase.
      ]. In MLL-r leukemia, aberrant recruitment of DOT1L occurs on promoters and gene bodies of MLL targets, aiding their expression [
      • Bernt KM
      • Zhu N
      • Sinha AU
      • et al.
      MLL-rearranged leukemia is dependent on aberrant H3K79 methylation by DOT1L.
      ]. DOT1L is critical to the MLL-r oncogenic transcriptional program, and DOT1L inhibition suppresses this program and leukemia development [
      • Chen CW
      • Koche RP
      • Sinha AU
      • et al.
      DOT1L inhibits SIRT1-mediated epigenetic silencing to maintain leukemic gene expression in MLL-rearranged leukemia.
      ]. Interestingly DOT1L also presents as a therapeutic target for the treatment of DNMT3A-mutant AML [
      • Rau RE
      • Rodriguez BA
      • Luo M
      • et al.
      DOT1L as a therapeutic target for the treatment of DNMT3A-mutant acute myeloid leukemia.
      ]. Methylation of arginine residues within histone tails (H3 and H4), though less common than lysine methylation, is regulated by protein arginine methyltransferases (PRMTs). PRMT1, the founding member of the PRMTs, has been reported to be required for leukemia initiation by the fusion proteins MLL–GAS7 and MOZ–TIF2, and its silencing was able to block leukemia transformation [
      • Cheung N
      • Fung TK
      • Zeisig BB
      • et al.
      Targeting aberrant epigenetic networks mediated by PRMT1 and KDM4C in acute myeloid leukemia.
      ]. PRMT4 and PRMT5 enzymes also play important roles in AML development [
      • Greenblatt SM
      • Man N
      • Hamard PJ
      • et al.
      CARM1 is essential for myeloid leukemogenesis but dispensable for normal hematopoiesis.
      ,
      • Hamard PJ
      • Santiago GE
      • Liu F
      • et al.
      PRMT5 Regulates DNA repair by controlling the alternative splicing of histone-modifying enzymes.
      ].

      HDMs in AML pathogenesis

      Histone demethylases (HDMs) catalyze removal of methyl groups from specific residues on histone tails, counteracting the function of methyltransferases. A dynamic balance between these two groups of enzymes ensures efficient histone methylome regulation, which is often disrupted in most malignancies, including AML (Figure 1).
      Figure 1
      Figure 1Schema representing molecular derangements and crosstalk between HMTs and HDMs impinging on an aberrant histone methylome and their implications in AML pathogenesis and targeted therapy.

      LSD family

      HDMs consist of two families of enzymes, LSD and JMJC. LSD1 (also known as KDM1A), the first demethylase discovered, is an amine oxidase that demethylates di- and monomethylated H3K4 and H3K9 residues. H3K4 methylation and H3K9 methylation play antagonistic roles; thus, LSD1 may function as either a repressor or an activator of transcription depending on its interacting partners. It is usually associated with gene repression and is a critical component of transcription repressor complexes such as CoREST and NuRD. LSD1, along with CoREST, associates with the transcriptional repressors Gfi-1/1b during hematopoietic differentiation [
      • Saleque S
      • Kim J
      • Rooke HM
      • Orkin SH
      Epigenetic regulation of hematopoietic differentiation by Gfi-1 and Gfi-1b is mediated by the cofactors CoREST and LSD1.
      ]. Lsd1 loss de-represses Gfi-1/1b lineage-specific transcriptional programs, hampering terminal differentiation of erythroid, megakaryocytic, and granulocytic cells [
      • Thambyrajah R
      • Mazan M
      • Patel R
      • et al.
      GFI1 proteins orchestrate the emergence of haematopoietic stem cells through recruitment of LSD1.
      ]. LSD1 is frequently overexpressed in AML. LSD1 loss impairs proliferation and increases differentiation and apoptosis in MLL- and AML1-rearranged leukemias [
      • Harris WJ
      • Huang X
      • Lynch JT
      • et al.
      The histone demethylase KDM1A sustains the oncogenic potential of MLL-AF9 leukemia stem cells.
      ].

      JMJC family

      The JMJC family encompasses most of the known demethylases. They have a catalytic C-terminal Jumonji (JmJC) domain and, unlike the LSD family, can demethylate mono-, di-, and trimethylated residues. The JMJC demethylases consist of many subfamilies with multiple members implicated in normal and malignant hematopoiesis. KDM2B, which demethylates H3K36, is overexpressed in leukemia stem cells (LSCs) and is required for their neoplastic transformation [
      • He J
      • Nguyen AT
      • Zhang Y
      KDM2b/JHDM1b, an H3K36me2-specific demethylase, is required for initiation and maintenance of acute myeloid leukemia.
      ]. Kdm2b-deleted mice exhibit a reduced number of long-term HSCs as well as defective lymphopoiesis, with a resultant upregulation of myeloid differentiation [
      • Andricovich J
      • Kai Y
      • Peng W
      • Foudi A
      • Tzatsos A
      Histone demethylase KDM2B regulates lineage commitment in normal and malignant hematopoiesis.
      ]. KDM2B also functions as the DNA binding subunit of the noncanonical PRC1.1 complex. Downregulation of KDM2B suppressed in vitro AML cell growth, as well as abrogated leukemogenesis in vivo, in PDX models, revealing the role of the PRC1.1 complex in regulating leukemia progression [
      • van den Boom V
      • Maat H
      • Geugien M
      • et al.
      Non-canonical PRC1.1 targets active genes independent of H3K27me3 and is essential for leukemogenesis.
      ]. KDM3A, specific for H3K9, has been reported to maintain myeloma cell survival. The 5q31 genomic region, which contains a portion of the KDM3B gene, is frequently deleted in myelodysplastic syndromes (MDS) and AML. Therefore, KDM3B is thought to play a role in tumor suppression. In contrast, another study reported that KDM3B is involved in transcriptional activation of the LMO2 oncogene in leukemia [
      • Kim JY
      • Kim KB
      • Eom GH
      • et al.
      KDM3B is the H3K9 demethylase involved in transcriptional activation of lmo2 in leukemia.
      ]. Though KDM3C is dispensable for healthy adult hematopoiesis as well as for JAK2V617F-driven MPN disease initiation, it has recently implicated in the progression of AML1–ETO- and HOXA9-mediated leukemias [
      • Chen M
      • Zhu N
      • Liu X
      • et al.
      JMJD1C is required for the survival of acute myeloid leukemia by functioning as a coactivator for key transcription factors.
      ]. KDM3C was identified as a co-activator in AETFC, a complex formed by AML1–ETO, where KDM3C maintained low level of H3K9me2, hence enhancing gene expression of AML1–ETO targets.
      KDM4, which targets both H3K9 and H3K36, has been characterized primarily as proto-oncogenic. KDM4A, KDM4B, and KDM4C in conjunction have been reported to mediate survival of leukemia cells by enhancing the expression of interleukin (IL)-3 receptor-α, a key initiator in the JAK–STAT pathway, in MLL–AF9-translocated AML [
      • Agger K
      • Miyagi S
      • Pedersen MT
      • Kooistra SM
      • Johansen JV
      • Helin K
      Jmjd2/Kdm4 demethylases are required for expression of Il3ra and survival of acute myeloid leukemia cells.
      ]. Additionally, conditional Kdm4a/Kdm4b/Kdm4c triple-knockout mice exhibit high H3K9me3 at transcription start sites and concomitant repression of several genes involved in HSC maintenance [
      • Agger K
      • Nishimura K
      • Miyagi S
      • Messling JE
      • Rasmussen KD
      • Helin K
      The KDM4/JMJD2 histone demethylases are required for hematopoietic stem cell maintenance.
      ]. KDM6 enzymes are exclusively H3K27 demethylases. KDM6A (UTX) has been found to direct migration of HSCs in response to SDF-1/CXCR4 signaling [
      • Thieme S
      • Gyarfas T
      • Richter C
      • et al.
      The histone demethylase UTX regulates stem cell migration and hematopoiesis.
      ]. Though KDM6A is mutated in many cancers and classified as a tumor suppressor, we had observed that it is also overexpressed in some cases of AML [
      • Boila LD
      • Chatterjee SS
      • Banerjee D
      • Sengupta A
      KDM6 and KDM4 histone lysine demethylases emerge as molecular therapeutic targets in human acute myeloid leukemia.
      ]. We found that KDM6A interacts and cooperates with an MBD3-deficient nucleosome remodeler histone deacetylase (NuRD) complex to promote CBP recruitment and H3K27 acetylation at DOCK5/8 loci, thereby inducing Rac GTPase activation and AML cell migration [
      • Biswas M
      • Chatterjee SS
      • Boila LD
      • Chakraborty S
      • Banerjee D
      • Sengupta A
      MBD3/NuRD loss participates with KDM6A program to promote DOCK5/8 expression and Rac GTPase activation in human acute myeloid leukemia.
      ]. KDM6A, being encoded by the X chromosome, has sex-specific effects; homozygous female Kdm6a-null mice exhibit MDS and suppressed erythro-megakaryocytopoiesis, whereas males have normal hematopoietic development [
      • Gozdecka M
      • Meduri E
      • Mazan M
      • et al.
      UTX-mediated enhancer and chromatin remodeling suppresses myeloid leukemogenesis through noncatalytic inverse regulation of ETS and GATA programs.
      ]. KDM6B (JMJD3) regulates transcriptional elongation, and overexpression of KDM6B is reported in MDS hematopoietic stem and progenitor cells (HSPCs) [
      • Wei Y
      • Zheng H
      • Bao N
      • et al.
      KDM6B overexpression activates innate immune signaling and impairs hematopoiesis in mice.
      ]. In addition, recent reports have also identified it as playing an oncogenic role in AML.

      Arginine demethylases

      Arginine demethylation is a relatively unexplored avenue. Although many of the lysine demethylases also exhibit in vitro arginine demethylation capability, to date, only two bona fide arginine demethylases (PAD4 and JMJD6) have been discovered. PAD4 catalyzes conversion of monomethylated arginine to citrulline on H3R17 and H4R3, while JMJD6 directly converts methylarginine to arginine by removing the methyl group. PAD4, along with LEF1 and HDAC1, represses c-myc expression and regulates proliferation of lineageSca-1+c-Kit+ (LSK) mouse bone marrow multipotent progenitor cells [
      • Nakashima K
      • Arai S
      • Suzuki A
      • et al.
      PAD4 regulates proliferation of multipotent haematopoietic cells by controlling c-myc expression.
      ]. Furthermore, PAD4, acting as a co-activator by influencing H3R2me2a, facilitates expression of the Tal1 target IL6ST [
      • Kolodziej S
      • Kuvardina ON
      • Oellerich T
      • et al.
      PADI4 acts as a coactivator of Tal1 by counteracting repressive histone arginine methylation.
      ]. This provides control over IL6ST expression during lineage differentiation of HSPCs. Recent studies have revealed involvement of PAD4 in ATRA-mediated differentiation of AML cells. On ATRA exposure, PAD4 translocates into the nucleus and downregulates SOX4 expression, which in turn relieves transcriptional repression of PU.1 [
      • Song G
      • Shi L
      • Guo Y
      • et al.
      A novel PAD4/SOX4/PU.1 signaling pathway is involved in the committed differentiation of acute promyelocytic leukemia cells into granulocytic cells.
      ]. Thus, PAD4 controls ATRA-mediated differentiation in a SOX4-dependent manner. JMJD6 removes dimethyl groups from H3R2 and H4R3. Though JMJD6 has not yet been implicated in AML pathogenesis, reports have indicated that it can influence AML cell sensitivity to BET protein inhibitors.

      Histone mutations in AML

      Apart from alterations in histone-modifying enzymes, mutations in the histone gene itself can alter methylation dynamics. A key example is diffuse intrinsic pontine glioma (DIPG), wherein heterozygous mutations (K27M and G34R/V) in H3F3A, the gene encoding the histone variant H3.3, are identified [
      • Schwartzentruber J
      • Korshunov A
      • Liu XY
      • et al.
      Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma.
      ]. Apart from DIPG, it also occurs in up to 60% of pediatric glioblastoma multiforme patients, in whom the presence of K27M–H3.3 predicts poor survival [
      • Wu G
      • Broniscer A
      • McEachron TA
      • et al.
      Somatic histone H3 alterations in pediatric diffuse intrinsic pontine gliomas and non-brainstem glioblastomas.
      ]. Primary human samples, as well as cell lines containing the K27M–H3.3 mutation, exhibit reduced levels of H3K27me2/3 and DNA hypomethylation at many loci. Similarly, H3K27 methylation also plays a critical role in AML [
      • Lehnertz B
      • Zhang YW
      • Boivin I
      • et al.
      H3(K27M/I) mutations promote context-dependent transformation in acute myeloid leukemia with RUNX1 alterations.
      ]. Loss of the H3K27me3 mark is associated with the multidrug resistance phenotype in AML [
      • Gollner S
      • Oellerich T
      • Agrawal-Singh S
      • et al.
      Loss of the histone methyltransferase EZH2 induces resistance to multiple drugs in acute myeloid leukemia.
      ]. Like glioblastoma, sequencing of the 16 histone H3 genes in AML identified 7 recurrent mutations, with H3K27 mutations having the highest frequency. These mutations occur with a higher frequency in secondary AML and have been found to exist in preleukemic HSCs and major leukemic clones [
      • Boileau M
      • Shirinian M
      • Gayden T
      • et al.
      Mutant H3 histones drive human pre-leukemic hematopoietic stem cell expansion and promote leukemic aggressiveness.
      ]. A previous study reported reduced H3K27me2/3 in patients with H3K27M/I mutations and accelerated disease progression in an AML1-ETO mouse model [
      • Lehnertz B
      • Zhang YW
      • Boivin I
      • et al.
      H3(K27M/I) mutations promote context-dependent transformation in acute myeloid leukemia with RUNX1 alterations.
      ]. In vivo functional assays using mutant human HSCs (CD34+CD38) revealed that H3.1K27M/I mutations increased stem cell-enriched population and engraftment potential in secondary recipients, along with a blockage in erythroid differentiation [
      • Boileau M
      • Shirinian M
      • Gayden T
      • et al.
      Mutant H3 histones drive human pre-leukemic hematopoietic stem cell expansion and promote leukemic aggressiveness.
      ]. Whether H3K27 mutations or methylation levels provide a clonal advantage, whether this is associated with drug resistance/relapse, and whether this alteration can be exploited to specifically target LSCs are important questions that need to be addressed.

      Crosstalk between HMTs and HDMs in AML

      Although histone-modifying enzymes appear to individually regulate their targets, epigenetic regulation is rather a concerted effort and extensive functional crosstalk exists between HMTs and HDMs to regulate locus-specific gene expression, global chromatin architecture, and cellular states. Emerging studies, using state-of-the-art genetic models, single-cell analyses, and modern technologies in chromatin biology, are helping us to further explore this crosstalk and appreciate the intricate molecular regulation of histone methylation in AML. Recently, it was reported that DOT1L antagonizes recruitment of SIRT1 and SUV39H1 to their targets [
      • Chen CW
      • Koche RP
      • Sinha AU
      • et al.
      DOT1L inhibits SIRT1-mediated epigenetic silencing to maintain leukemic gene expression in MLL-rearranged leukemia.
      ]. On DOT1L inhibition, SIRT1 and SUV39H1 bound to MLL targets and repressed their expression. Thus MLL-r AML sensitivity to DOT1L inhibitors depends on SIRT1 and SUV39H1 levels, indicating the importance of the functional crosstalk between these enzymes. Similarly, in MLL-AF9 AML, KDM4C together with PRMT1 co-regulates transcription of MLL downstream targets [
      • Cheung N
      • Fung TK
      • Zeisig BB
      • et al.
      Targeting aberrant epigenetic networks mediated by PRMT1 and KDM4C in acute myeloid leukemia.
      ]. Another interesting example is that KDM4A inhibition restores H3K36me3 and sensitizes SETD2-mutant AML to cytarabine treatment [
      • Mar BG
      • Chu SH
      • Kahn JD
      • et al.
      SETD2 alterations impair DNA damage recognition and lead to resistance to chemotherapy in leukemia.
      ]. Though not yet demonstrated in AML, KDM6A mutation sensitizes multiple myeloma cells to EZH2 inhibition [
      • Ezponda T
      • Dupere-Richer D
      • Will CM
      • et al.
      UTX/KDM6A loss enhances the malignant phenotype of multiple myeloma and sensitizes cells to EZH2 inhibition.
      ]. Studies have also found that MLL fusion oncoproteins promote EZH2 transcription, indicating a role for PRC2 in MLL-r leukemia. PRC2, in turn, represses genes that are critical to the myeloid differentiation program [
      • Neff T
      • Sinha AU
      • Kluk MJ
      • et al.
      Polycomb repressive complex 2 is required for MLL-AF9 leukemia.
      ]. In MLL-r AML, both EZH2 and EZH1 are expressed and compensate each other to promote leukemogenesis; thus, simultaneous disruption of both enzymes is required to inhibit growth of leukemia carrying MLL–AF9.

      Chromatin accessibility and AML stem cell function

      Chromatin architecture plays a key role in determining the accessibility of transcription factors to gene regulatory regions. This controls cell type- as well as development stage-specific gene expression. Chromatin markers define stem cell function, and orchestrated changes in chromatin accessibility ensure efficient progression of HSCs through maturational stages to form the differentiated blood cells [
      • Ng SW
      • Mitchell A
      • Kennedy JA
      • et al.
      A 17-gene stemness score for rapid determination of risk in acute leukaemia.
      ,
      • Shlush LI
      • Zandi S
      • Mitchell A
      • et al.
      Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia.
      ,
      • Radzisheuskaya A
      • Shliaha PV
      • Grinev V
      • et al.
      PRMT5 methylome profiling uncovers a direct link to splicing regulation in acute myeloid leukemia.
      ]. Aberrant chromatin structure is often associated with disruption of this regulation, leading to malignancies [
      • Glass JL
      • Hassane D
      • Wouters BJ
      • et al.
      Epigenetic identity in AML depends on disruption of nonpromoter regulatory elements and is affected by antagonistic effects of mutations in epigenetic modifiers.
      ,
      • Chatterjee SS
      • Biswas M
      • Boila LD
      • Banerjee D
      • Sengupta A
      SMARCB1 deficiency integrates epigenetic signals to oncogenic gene expression program maintenance in human acute myeloid leukemia.
      ,
      • Sinha S
      • Biswas M
      • Chatterjee SS
      • Kumar S
      • Sengupta A
      Pbrm1 steers mesenchymal stromal cell osteolineage differentiation by integrating PBAF-dependent chromatin remodeling and BMP/TGF-beta signaling.
      ]. Profiling of genomewide histone methylation marks has revealed that LSCs in MLL-r leukemia are characterized by high H3K4me3 and low H3K79me2. KDM5B, the H3K4-specific demethylase, negatively regulates LSC potential, demonstrating importance of the H3K4 methylome in determining LSC fate [
      • Wong SH
      • Goode DL
      • Iwasaki M
      • et al.
      The H3K4-methyl epigenome regulates leukemia stem cell oncogenic potential.
      ]. H3K4me3 ensures an open chromatin structure and accessibility of leukemogenic MLL to its downstream targets Hoxa9 and Meis1. In addition to MLL-r AML, the HoxA cluster is upregulated in AML1–ETO-induced AML as well. HMGN1, a chromatin modulator, is frequently amplified in AML and is associated with high H3K27ac and increased accessibility and expression of HoxA cluster genes [
      • Cabal-Hierro L
      • van Galen P
      • Prado MA
      • et al.
      Chromatin accessibility promotes hematopoietic and leukemia stem cell activity.
      ]. HMGN1 overexpression decreases quiescence and induces HSC proliferation. It functions through cooperation with AML1–ETO oncoprotein, blocking myeloid differentiation and enhancing LSC activity. Inhibition of the H3K27 histone acetyltransferases CBP/p300, with concomitant reduction in HMGN1-associated H3K27ac, relieves differentiation block [
      • Cabal-Hierro L
      • van Galen P
      • Prado MA
      • et al.
      Chromatin accessibility promotes hematopoietic and leukemia stem cell activity.
      ]. Thus, balance between H3K27 methylation and acetylation states regulates HMGN1-mediated LSC potential. H3K9me3, another repressive mark, also differentiates gene expression between normal HSCs and LSCs.
      ALKBH5 is an m6A demethylase required for LSC function and is regulated by H3K9me3 levels at its promoter [
      • Wang J
      • Li Y
      • Wang P
      • et al.
      Leukemogenic chromatin alterations promote AML leukemia stem cells via a KDM4C-ALKBH5-AXL signaling axis.
      ]. KDM4C, which is upregulated in AML, removes the repressive H3K9me3 at the ALKBH5 promoter, increasing chromatin accessibility. This facilitates recruitment of MYB and Pol II, increasing ALKBH5 expression. It has also been observed that chromatin architecture differs for AMLs with different underlying genetic signatures, which in turn affect their stemness. Patients harboring NPM1 mutations or MLL-fusion genes have greater chromatin accessibility of HOX-family genes and have high self-renewal capacity and stemness and a poorer prognosis [
      • Kuhn MW
      • Song E
      • Feng Z
      • et al.
      Targeting chromatin regulators inhibits leukemogenic gene expression in NPM1 mutant leukemia.
      ,
      • Yi G
      • Wierenga ATJ
      • Petraglia F
      • et al.
      Chromatin-based classification of genetically heterogeneous AMLs into two distinct subtypes with diverse stemness phenotypes.
      ]. In contrast, patients with RUNX1 or spliceosome mutations depend mainly on IRF family regulators for the downstream transcriptional program [
      • Yi G
      • Wierenga ATJ
      • Petraglia F
      • et al.
      Chromatin-based classification of genetically heterogeneous AMLs into two distinct subtypes with diverse stemness phenotypes.
      ]. Thus, histone methylation states play an indispensable role in regulating chromatin architecture, transcriptional accessibility, and AML stem cell function.

      Histone methylome in AML therapy and response

      As the role of the histone methylome in AML pathophysiology becomes more evident, new therapeutic strategies to target these aberrations are being increasingly explored (Table 1). Among the HMTs, DOT1L and EZH2 have emerged as promising targets. DOT1L inhibitors have extensively been used to reduce leukemia burden in a variety of MLL-r AML models. Similarly, DZNep, an EZH2 inhibitor, caused accumulation of reactive oxygen species (ROS) and induced apoptosis in MLL-r AML cells, reducing the frequency of leukemia-initiating cells (LICs) [
      • Zhou J
      • Bi C
      • Cheong LL
      • et al.
      The histone methyltransferase inhibitor, DZNep, up-regulates TXNIP, increases ROS production, and targets leukemia cells in AML.
      ]. However, in many cases, simultaneous inhibition of both EZH1 and EZH2 is required [
      • Neff T
      • Sinha AU
      • Kluk MJ
      • et al.
      Polycomb repressive complex 2 is required for MLL-AF9 leukemia.
      ,
      • Shi J
      • Wang E
      • Zuber J
      • et al.
      The polycomb complex PRC2 supports aberrant self-renewal in a mouse model of MLL-AF9;Nras(G12D) acute myeloid leukemia.
      ]. UNC1999, an orally bioavailable dual EZH1 and EZH2 inhibitor, has emerged as a promising candidate in MLL-r leukemia [
      • Xu B
      • On DM
      • Ma A
      • et al.
      Selective inhibition of EZH2 and EZH1 enzymatic activity by a small molecule suppresses MLL-rearranged leukemia.
      ]. Combining DOT1L and EZH2 deletion demonstrated synergy in vivo in an MLL–AF9 leukemia model [
      • Lenard A
      • Xie HM
      • Pastuer T
      • et al.
      Epigenetic regulation of protein translation in KMT2A-rearranged AML.
      ]. However, the same combination had either a synergistic effect or an antagonistic effect when investigated in a panel of human AML cell lines. DOT1L inhibition suppressed ribosomal biogenesis and protein translation. Consequently, combination with homoharringtonine, a protein translation inhibitor, revealed an additive effect in MLL-r leukemias [
      • Lenard A
      • Xie HM
      • Pastuer T
      • et al.
      Epigenetic regulation of protein translation in KMT2A-rearranged AML.
      ]. Additionally, PRMT1 inhibition using AMI-408 has also proved effective in MLL–GAS7 and MOZ–TIF2 fusion-carrying mouse models [
      • Cheung N
      • Fung TK
      • Zeisig BB
      • et al.
      Targeting aberrant epigenetic networks mediated by PRMT1 and KDM4C in acute myeloid leukemia.
      ]. The dosage effect of SETD2 in AML, and its associated tumor vulnerability, also provides an excellent therapeutic opportunity to be explored.
      Table 1Histone methylation-modifying enzymes implicated in AML pathogenesis
      GeneTargetAlterationFunctionInhibitorsClinical trial identifier
      MLL1H3K4→H3K4me1/2/3Fusion of N-terminal DNA binding domain with C-terminal of transcription elongation factorsOncogeneMM-401, MM-102, KO-539 (menin-MLL inhibitor), SNDX-5613 (menin-MLL inhibitor)NCT04067336 NCT04065399
      EZH1H3K27→H3K27me1/2/3UpregulatedOncogeneCPI-360, UNC1999, DS-3201
      EZH2H3K27→H3K27me1/2/3Loss of function mutation, upregulatedOncogene, tumor suppressorEl1, GSK2816126, EPZ-6438, CPI-1205, GSK343
      SUV39H1H3K9me1→H3K9me3NoneOncogeneChaetocin
      G9AH3K9me→H3K9me1/2NoneOncogeneUNC0638, UNC0642
      SETD2H3K36me2→H3K36me3Loss of function mutationTumor suppressorEPZ-040414
      DOT1LH3K79→H3K79me1/2/3NoneOncogeneEPZ004777, EPZ-5676, SGC0946NCT01684150 NCT02141828
      PRMT1H4R3→H4R3me1, H4R3me2aUpregulatedOncogeneAMI-408, C7280948
      PRMT5H3R8→H3R8me2s

      H4R3→H4R3me2s
      UpregulatedOncogeneGSK3326595NCT03614728
      LSD1H3K4me1/2→H3K4

      H3K9me1/2→H3K9
      UpregulatedOncogeneTCP, GSK-LSD1, ORY1001, IMG7289, GSK2879552NCT02177812 NCT02273102

      NCT02261779 NCT02842827
      KDM2BH3K4me3→H3K4me2

      H3K36me2→H3K36me1
      UpregulatedOncogeneNone
      KDM3AH3K9me1/2→H3K9NoneOncogene, tumor suppressorNone
      KDM3CH3K9me1/2→H3K9NoneOncogeneNone
      KDM4AH3K9me2/3→H3K9me1

      H3K36me3→H3K36me2
      UpregulatedOncogeneCP2
      KDM4BH3K9me2/3→H3K9me1

      H3K36me3→H3K36me2
      UpregulatedOncogeneNSC636819
      KDM4CH3K9me2/3→H3K9me1

      H3K36me3→H3K36me2
      UpregulatedOncogeneSD-70
      KDM6AH327me2/3→H3K27me1Loss of function mutation, upregulatedOncogene, tumor suppressorGSK-J4
      KDM6BH327me2/3→H3K27me1UpregulatedOncogeneGSK-J4
      PADI4H3R17me→H3R17ci

      H4R3me→H4R3ci
      DownregulatedTumor suppressorGSK199, GSK484
      JMJD6H3R2me1/2/2a→H3R2

      H4R3me1/2/2a→H4R3
      NoneTumor suppressorSKLB325
      Among HDMs, LSD1 inhibition has shown promise in MLL-r leukemia [
      • Schenk T
      • Chen WC
      • Gollner S
      • et al.
      Inhibition of the LSD1 (KDM1A) demethylase reactivates the all-trans-retinoic acid differentiation pathway in acute myeloid leukemia.
      ]. The LSD1 inhibitor tranylcypromine (TCP), either alone or in combination with ATRA, disrupted the MLL oncogenic program and induced expression of myeloid differentiation genes in MLL-r AML cells [
      • Cusan M
      • Cai SF
      • Mohammad HP
      • et al.
      LSD1 inhibition exerts its antileukemic effect by recommissioning PU.1- and C/EBPalpha-dependent enhancers in AML.
      ]. Overall, this has resulted in a proof-of-concept phase I/II pilot trial with relapsed/refractory (r/r) AML patients ineligible for intensive therapy and another phase I/II trial for MDS patients. Among 18 AML patients, the overall response rate was a meager 20%, which included 2 complete remissions and 1 partial response [
      • Wass M
      • Gollner S
      • Besenbeck B
      • et al.
      A proof of concept phase I/II pilot trial of LSD1 inhibition by tranylcypromine combined with ATRA in refractory/relapsed AML patients not eligible for intensive therapy.
      ]. However, there was no hematological recovery, with the median overall survival being 3.3 months. This pilot trial indicates that the TCP/ATRA combination does induce differentiation and response in r/r AML, but only slightly so. The trial on MDS patients was subsequently terminated as risk–benefit analysis did not favor continuation of the study. Therefore, effects observed in vitro may not always translate into suitable drugs in vivo, advocating for improving efficacy and establishing better preclinical models. A recent study further strengthened this notion, wherein irreversible LSD1 inhibition was sufficient to induce cytotoxicity in vitro in a model of CEBPA/CSF3R mutant AML, but failed to do so in vivo. However, when combined with ruxolitinib, an inhibitor of JAK/STAT signaling, LSD1 inhibition synergized to normalize peripheral blood WBC counts and double median survival in vivo [
      • Braun TP
      • Coblentz C
      • Smith BM
      • et al.
      Combined inhibition of JAK/STAT pathway and lysine-specific demethylase 1 as a therapeutic strategy in CSF3R/CEBPA mutant acute myeloid leukemia.
      ]. Similarly, in MLL-r leukemias, apart from ATRA, combination with mTORC1 inhibition has also been found to enhance the differentiation potential of LSD1 inhibition [
      • Deb G
      • Wingelhofer B
      • Amaral FMR
      • et al.
      Pre-clinical activity of combined LSD1 and mTORC1 inhibition in MLL-translocated acute myeloid leukaemia.
      ]. Leukemias derived from HSCs exhibit resistance to LSD1 treatment compared with those initiated from myeloid progenitor cells [
      • Cai SF
      • Chu SH
      • Goldberg AD
      • et al.
      Leukemia cell of origin influences apoptotic priming and sensitivity to LSD1 inhibition.
      ]. Elevated Evi1 expression in HSCs attenuates p53 apoptotic response, which can be overcome by combining LSD1 inhibition with BCL2 inhibition. Among the JMJC demethylases, the KDM4C inhibitor SD70 has so far had an excellent therapeutic effect on AML expressing MOZ–TIF2 and MLL fusions [
      • Cheung N
      • Fung TK
      • Zeisig BB
      • et al.
      Targeting aberrant epigenetic networks mediated by PRMT1 and KDM4C in acute myeloid leukemia.
      ]. We and others have found that KDM6 inhibition using GSKJ4 also has antileukemic potential both in vitro and in vivo [
      • Boila LD
      • Chatterjee SS
      • Banerjee D
      • Sengupta A
      KDM6 and KDM4 histone lysine demethylases emerge as molecular therapeutic targets in human acute myeloid leukemia.
      ,
      • Li Y
      • Zhang M
      • Sheng M
      • et al.
      Therapeutic potential of GSK-J4, a histone demethylase KDM6B/JMJD3 inhibitor, for acute myeloid leukemia.
      ].
      In addition to their being therapeutic targets, HMTs and HDMs often predict treatment response in AML. As previously mentioned, H3K27me3 loss promotes multidrug resistance in AML [
      • Gollner S
      • Oellerich T
      • Agrawal-Singh S
      • et al.
      Loss of the histone methyltransferase EZH2 induces resistance to multiple drugs in acute myeloid leukemia.
      ]. Low H3K27me3 levels predict a poor prognosis with significantly decreased overall (median 11.06 vs. 38.6 months) and disease-free survival (median 10.0 vs. 31.2 months) compared with patients with high H3K27me3. Loss of EZH2, the enzyme responsible for H3K27me3 levels, also leads to acquired drug resistance to tyrosine kinase inhibitors (TKIs) and cytotoxic drugs in AML [
      • Gollner S
      • Oellerich T
      • Agrawal-Singh S
      • et al.
      Loss of the histone methyltransferase EZH2 induces resistance to multiple drugs in acute myeloid leukemia.
      ]. Surprisingly, though KDM6A plays an antagonistic role to EZH2, its loss of function has also been found to result in a similar phenomenon. In 45.7% of CN-AML patients, relapse-specific loss of KDM6A was observed [
      • Stief SM
      • Hanneforth AL
      • Weser S
      • et al.
      Loss of KDM6A confers drug resistance in acute myeloid leukemia.
      ]. KDM6A loss caused reduced H3K27ac at the nucleoside membrane transporter ENT1 locus, repressing its expression, which led to cytarabine resistance. Similarly, LSD1 expression predicts response to ATRA-mediated differentiation in MLL-r leukemia, and LSD1 inhibition appears synergistic with ATRA therapy [
      • Schenk T
      • Chen WC
      • Gollner S
      • et al.
      Inhibition of the LSD1 (KDM1A) demethylase reactivates the all-trans-retinoic acid differentiation pathway in acute myeloid leukemia.
      ]. KDM7B, an H3K9 demethylase, is also implicated to define response to ATRA-mediated differentiation of RARα fusion-driven acute promyelocytic leukemia [
      • Arteaga MF
      • Mikesch JH
      • Qiu J
      • et al.
      The histone demethylase PHF8 governs retinoic acid response in acute promyelocytic leukemia.
      ].

      Immunomodulation and combination therapy

      In patients unresponsive to conventional chemotherapy or in elderly individuals in whom induction therapy is not an option, immunomodulatory strategies, including immune checkpoint inhibition (ICI) therapy, are being increasingly utilized. Though immunotherapy has shown promise in selected solid tumors, its effectiveness in AML remains limited. Remission achieved through allogenic HSC transplantation in AML patients was a kind of first successful “adoptive immunotherapy” cancer treatment, suggesting that AML was amenable to immunomodulation [
      • Walter RB
      • Gooley TA
      • Wood BL
      • et al.
      Impact of pretransplantation minimal residual disease, as detected by multiparametric flow cytometry, on outcome of myeloablative hematopoietic cell transplantation for acute myeloid leukemia.
      ,
      • Curran E
      • Chen X
      • Corrales L
      • et al.
      Sting pathway activation stimulates potent immunity against acute myeloid leukemia.
      ]. However, there are several challenges that make AML a difficult target for immunotherapy.
      First, AML is a disseminated and systemic malignancy, which itself arises within the core of hematopoietic hierarchy that generates a spectrum of immune cells. Second, as AML is present predominantly in individuals with age-related clonal hematopoiesis with an expansion of the myeloid compartment, an adaptive immune system is intrinsically prone to immunosenescence. Third, AML has a relatively lower mutational burden, thus reducing neo-antigens for elimination by cytotoxic T lymphocytes (CTLs) [
      Cancer Genome Atlas Research Network
      Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia.
      ,
      • Lawrence MS
      • Stojanov P
      • Polak P
      • et al.
      Mutational heterogeneity in cancer and the search for new cancer-associated genes.
      ]. Fourth, the immune-evasive strategies of AML blasts include inefficient cross-priming by antigen presenting cells, downregulation of antigen presentation by MHC class I and II molecules, expansion of immunosuppressive regulatory T lymphocytes (Tregs), and upregulation of checkpoint inhibitors such as PD-L1, CTLA-4, TIM-3, and LAG-3, among others [
      • Ustun C
      • Miller JS
      • Munn DH
      • Weisdorf DJ
      • Blazar BR
      Regulatory T cells in acute myelogenous leukemia: is it time for immunomodulation?.
      ,
      • Le Dieu R
      • Taussig DC
      • Ramsay AG
      • et al.
      Peripheral blood T cells in acute myeloid leukemia (AML) patients at diagnosis have abnormal phenotype and genotype and form defective immune synapses with AML blasts.
      ,
      • Vago L
      • Perna SK
      • Zanussi M
      • et al.
      Loss of mismatched HLA in leukemia after stem-cell transplantation.
      ,
      • Curran EK
      • Godfrey J
      • Kline J
      Mechanisms of immune tolerance in leukemia and lymphoma.
      ]. Overall, these immune suppressive circuits make AML blasts resistant to immunotherapy, which may be improved by rationally combining targeted, monoclonal, or immune-activating approaches with epigenetic or cytotoxic therapies.
      Several studies have revealed epigenetic regulators to have immunomodulatory function, thus making them ideal targets for combination therapy. Immunomodulation can be attained by the action of the drugs either on the target cells itself, on the immune cells, or on the tumor microenvironment. Control of CIITA expression, the master regulator of MHC-II, upon interferon (IFN)-γ treatment by EZH2 was first demonstrated in cervical cancer cells. EZH2 suppresses CIITA, reducing surface MHC-II, thus restricting antigen presentation to CD4+ T cells, which can be reversed by EZH2 inhibition [
      • Mehta NT
      • Truax AD
      • Boyd NH
      • Greer SF
      Early epigenetic events regulate the adaptive immune response gene CIITA.
      ]. Furthermore, combining EZH2 and DNMT1 inhibition in ovarian cancer derepresses Th-1-type chemokines CXCL9 and CXCL10, increasing CD8+ T-cell infiltration; enhances PD-L1 checkpoint inhibition; and reduces tumor burden [
      • Peng D
      • Kryczek I
      • Nagarsheth N
      • et al.
      Epigenetic silencing of TH1-type chemokines shapes tumour immunity and immunotherapy.
      ]. Similarly, KDM1A inhibition synergizes with HDAC1 inhibition, increasing pro-inflammatory cytokine expression, effectively rewiring the tumor from an immunologically “cold” to a “hot” microenvironment [
      • Janzer A
      • Lim S
      • Fronhoffs F
      • Niazy N
      • Buettner R
      • Kirfel J
      Lysine-specific demethylase 1 (LSD1) and histone deacetylase 1 (HDAC1) synergistically repress proinflammatory cytokines and classical complement pathway components.
      ]. Histone methylome also plays a pivotal part in determining the plasticity of multipotent memory progenitor (MP) cells and their differentiation into CD8+ terminal effector (TE) cells. MP cells exhibit low H3K27me3 at pro-memory and pro-survival genes, indicative of permissive chromatin [
      • Gray SM
      • Amezquita RA
      • Guan T
      • Kleinstein SH
      • Kaech SM
      Polycomb repressive complex 2-mediated chromatin repression guides effector CD8(+) T cell terminal differentiation and loss of multipotency.
      ]. EZH2-mediated deposition of H3K27me3 at pro-memory genes occurs during differentiation of MP cells into CD8+ TE cells. Additionally, Th-cell differentiation is regulated by methylation status at H3K9 and H3K27 residues [
      • Yang XP
      • Jiang K
      • Hirahara K
      • et al.
      EZH2 is crucial for both differentiation of regulatory T cells and T effector cell expansion.
      ]. The SUV39H1–H3K9me3–HP1α pathway is essential for silencing antitumorigenic Th1-specific gene loci, favoring the protumorigenic Th2 subtype CD4+ T cells [
      • Allan RS
      • Zueva E
      • Cammas F
      • et al.
      An epigenetic silencing pathway controlling T helper 2 cell lineage commitment.
      ], whereas EZH2 loss results in enhanced Th1 and Th2 cell polarization [
      • Tumes DJ
      • Onodera A
      • Suzuki A
      • et al.
      The polycomb protein Ezh2 regulates differentiation and plasticity of CD4(+) T helper type 1 and type 2 cells.
      ]. Among HDMs, KDM6 plays an integral function in inducing pro-inflammatory cytokines and has been reported to be important for Th1 lineage commitment as well as M2 macrophage activation [
      • Liu PS
      • Wang H
      • Li X
      • et al.
      Alpha-ketoglutarate orchestrates macrophage activation through metabolic and epigenetic reprogramming.
      ].
      AMG-330 is a bispecific anti-CD33/CD3 T cell-engaging construct reported to be effective in favorable-risk AML patients [
      • Friedrich M
      • Henn A
      • Raum T
      • et al.
      Preclinical characterization of AMG 330, a CD3/CD33-bispecific T-cell-engaging antibody with potential for treatment of acute myelogenous leukemia.
      ]. It is currently in a phase I clinical trial (NCT02520427). AMG-330, though effective, is susceptible to resistance mechanisms such PD-L1 upregulation and Treg expansion [
      • Laszlo GS
      • Gudgeon CJ
      • Harrington KH
      • et al.
      Cellular determinants for preclinical activity of a novel CD33/CD3 bispecific T-cell engager (BiTE) antibody, AMG 330, against human AML.
      ], highlighting the need to combine it with ICI therapy [
      • Krupka C
      • Kufer P
      • Kischel R
      • et al.
      Blockade of the PD-1/PD-L1 axis augments lysis of AML cells by the CD33/CD3 BiTE antibody construct AMG 330: reversing a T-cell-induced immune escape mechanism.
      ]. DNA hypomethylation has been found to upregulate expression of immune checkpoint inhibitors. Treatment with hypomethylating agents (HMAs) such as azacitidine in AML often induces PD-L1 expression, which blocks CTL activity and has been associated with azacitidine resistance. This has led to clinical trials combining PD-1/PD-L1 inhibitors with azacitidine in AML and MDS [
      • Daver N
      • Boddu P
      • Garcia-Manero G
      • et al.
      Hypomethylating agents in combination with immune checkpoint inhibitors in acute myeloid leukemia and myelodysplastic syndromes.
      ]. In a phase II clinical trial combining nivolumab and 5-azacitidine, 11 of the 53 patients treated (21%) achieved CR/CRi, and 7 (14%) had hematologic improvement [
      • Daver N BS G-MG
      • Cortes JE
      • et al.
      Phase IB/II study of nivolumab with azacytidine (AZA) in patients (pts) with relapsed AML [Abstract].
      ]. Apart from DNA methylation, the repressive histone marks H3K9me3 and H3K27me3 also play a crucial role in determining expression of immune checkpoint inhibitors. In breast cancer PD-1, expression of CTLA-4, TIM-3, and LAG-3 is elevated on removal of the repressive histone modifications [
      • Sasidharan Nair V
      • El Salhat H
      • Taha RZ
      • John A
      • Ali BR
      • Elkord E
      DNA methylation and repressive H3K9 and H3K27 trimethylation in the promoter regions of PD-1, CTLA-4, TIM-3, LAG-3, TIGIT, and PD-L1 genes in human primary breast cancer.
      ]. Unlike HMAs, HMT and KDM inhibitors appear to be more specific, and combining histone methylome drugs with checkpoint inhibitors presents a promising therapeutic avenue. Apart from directly modulating checkpoint molecules, histone methylation has been described to play a pivotal role in determining expression of key cytokines, thus regulating the cytokine milieu in the tumor microenvironment, which in turn affects immune cell infiltration. Thus, epigenetic reversal of aberrant histone methylation can provide for enhanced recruitment of cytotoxic immune cells, efficient targeting of AML blasts, and improved patient survival.

      Discussion and future perspective

      Characterization of mutational landscape in AML, owing to the advent of genome sequencing studies over the past several years, has significantly increased our knowledge of alterations involving epigenetic modifiers and the importance of epigenetic contribution in AML pathogenesis. Despite significant progress in understanding the molecular basis of AML, treatment has remained stagnant for the past four decades. Epigenetic plasticity and transcriptional dysregulation are key hallmarks in leukemogenesis, which can contribute immensely to treatment responses. In this effort we have highlighted critically emerging regulators of histone methylation and the alterations they undergo during AML transformation, as well as their potential to serve as therapeutic targets. Epigenetic therapy is an emerging proposition with an immense potential. However, there are multiple aspects and caveats that need to be considered before it can synergize with, or even substitute for, conventional chemotherapy as the mainstay of AML treatment. Specific clinical patterns of sensitivity to epigenetic therapies are associated with molecularly defined subtypes and genetic signatures. Therefore, a major challenge in determining appropriate epigenetic drugs will be based on individual genetic barcoding of the cohort. Indeed, this would demand robust sensitivity studies involving in silico and in vitro small molecule inhibitor and genetic screening, combined with an in-depth characterization, to identify specific “epigenetic lesions” and their respective drivers.
      Epigenetic inhibitors often do not destroy the malignant clone, but rather promote differentiation of leukemia cells. To achieve complete remission, they need to be combined with other small molecule inhibitors, immunomodulatory drugs, or conventional chemotherapy. Thus, design of newer trials should involve an understanding of the synergism and antagonism of epigenetic drugs with current therapeutic modalities. As witnessed by recent clinical trial failures, new models to test epigenetic therapies must be developed, possibly employing patient samples and patient-derived xenografts, as well as genetically engineered mouse models that recapitulate the entire spectrum of AML mutations. Another significant challenge to AML therapy is therapy resistance and subsequent relapse. With the recent technological advancements, combining epi/genomic, transcriptomic, proteomic, and mass cytometry analysis, even at the single-cell level, with an unsupervised systems learning approach matched with in vivo clonal repopulation assays, it is now plausible to trace the evolution of individual clones to better understand AML pathophysiology and identify molecular vulnerabilities.
      As AML is a heterogeneous disease, individual subclones are phenotypically and functionally different. Proteomics-based assays have identified 50 AML-enriched plasma membrane proteins, permitting isolation of individual clones from an oligoclonal patient [
      • de Boer B
      • Prick J
      • Pruis MG
      • et al.
      Prospective isolation and characterization of genetically and functionally distinct AML subclones.
      ]. This is integral to tracing the individual clones in a heterogeneous setting, and can be used for diagnosis, treatment, and survival prediction. Tracing of individual clones is a robust evaluation of minimal residual disease. A clone-specific strategy employing next-generation sequencing and fluorescent in situ hybridization in 69 AMLs with known clonal architecture revealed that the presence of two or more lesions in more than 0.4% of remission cells correlates with lower overall and disease-free survival [
      • Hirsch P
      • Tang R
      • Abermil N
      • et al.
      Precision and prognostic value of clone-specific minimal residual disease in acute myeloid leukemia.
      ]. This contrasts with recent studies indicating that residual mutations in less than 5% of cells in complete remission correlates with a better survival. Lineage tracing has also enhanced our understanding of mechanisms of chemoresistance and how it can be overcome. A recent study using lentivirus-mediated DNA barcoding in human AML cells identified DNMT inhibition as preventing outgrowth of chemoresistant clones with enhanced stemness [
      • Caiado F
      • Maia-Silva D
      • Jardim C
      • et al.
      Lineage tracing of acute myeloid leukemia reveals the impact of hypomethylating agents on chemoresistance selection.
      ]. Currently, a limited number of selective inhibitors against epigenetic targets are available, and efforts to develop a broader arsenal are underway. Future studies focusing on discerning the molecular regulation of pathologically relevant histone methyl writers, readers, and erasers, their potential crosstalk, and the pivotal role of critical histone residues will be instrumental for the development of next-generation AML-targeted therapy.

      Conflict of interest disclosure

      The authors declare no conflicts of interest.

      Acknowledgments

      This study is supported by funding from the Council for Scientific & Industrial Research (CSIR) (NWP/HCP-0008 to AS), Department of Biotechnology (DBT) (BT/RLF/RE-ENTRY/06/2010), Ramalingaswami Fellowship (to AS), DBT (BT/PR13023/MED/31/311/2015) (to AS), and SERB-Department of Science & Technology (DST) (SB/SO/HS-053/2013), Government of India (to AS). AS is a recipient of the Indian Council of Medical Research–Department of Health Research (ICMR-DHR) International Fellowship for Indian Biomedical Scientists (INDO/FRC/452/S-11/2019-20-lHD). LDB acknowledges support from CSIR and the Shyama Prasad Mukherjee Fellowship.
      We regret for not citing several other original references because of space constraints.

      References

        • Appelbaum FR
        • Gundacker H
        • Head DR
        • et al.
        Age and acute myeloid leukemia.
        Blood. 2006; 107: 3481-3485
        • Versluis J
        • Hazenberg CL
        • Passweg JR
        • et al.
        Post-remission treatment with allogeneic stem cell transplantation in patients aged 60 years and older with acute myeloid leukaemia: a time-dependent analysis.
        Lancet Haematol. 2015; 2: e427-e436
        • Welch JS
        • Ley TJ
        • Link DC
        • et al.
        The origin and evolution of mutations in acute myeloid leukemia.
        Cell. 2012; 150: 264-278
        • Osorio FG
        • Rosendahl Huber A
        • Oka R
        • et al.
        Somatic Mutations Reveal Lineage Relationships and Age-Related Mutagenesis in Human Hematopoiesis.
        Cell Rep. 2018; 25 (e4): 2308-2316
        • Glass JL
        • Hassane D
        • Wouters BJ
        • et al.
        Epigenetic identity in AML depends on disruption of nonpromoter regulatory elements and is affected by antagonistic effects of mutations in epigenetic modifiers.
        Cancer Discov. 2017; 7: 868-883
        • Shi Y
        • Lan F
        • Matson C
        • et al.
        Histone demethylation mediated by the nuclear amine oxidase homologue LSD1.
        Cell. 2004; 119: 941-953
        • Chen Z
        • Zang J
        • Whetstine J
        • et al.
        Structural insights into histone demethylation by JMJD2 family members.
        Cell. 2006; 125: 691-702
        • Zhang Y
        • Reinberg D
        Transcription regulation by histone methylation: interplay between different covalent modifications of the core histone tails.
        Genes Dev. 2001; 15: 2343-2360
        • Greer EL
        • Shi Y
        Histone methylation: a dynamic mark in health, disease and inheritance.
        Nat Rev Genet. 2012; 13: 343-357
        • McMahon KA
        • Hiew SY
        • Hadjur S
        • et al.
        Mll has a critical role in fetal and adult hematopoietic stem cell self-renewal.
        Cell Stem Cell. 2007; 1: 338-345
        • Yagi H
        • Deguchi K
        • Aono A
        • Tani Y
        • Kishimoto T
        • Komori T
        Growth disturbance in fetal liver hematopoiesis of Mll-mutant mice.
        Blood. 1998; 92: 108-117
        • Meyer C
        • Burmeister T
        • Groger D
        • et al.
        The MLL recombinome of acute leukemias in 2017.
        Leukemia. 2018; 32: 273-284
        • Mishra BP
        • Zaffuto KM
        • Artinger EL
        • et al.
        The histone methyltransferase activity of MLL1 is dispensable for hematopoiesis and leukemogenesis.
        Cell Rep. 2014; 7: 1239-1247
        • Milne TA
        • Kim J
        • Wang GG
        • et al.
        Multiple interactions recruit MLL1 and MLL1 fusion proteins to the HOXA9 locus in leukemogenesis.
        Mol Cell. 2010; 38: 853-863
        • Thiel AT
        • Blessington P
        • Zou T
        • et al.
        MLL-AF9-induced leukemogenesis requires coexpression of the wild-type Mll allele.
        Cancer Cell. 2010; 17: 148-159
        • Chen Y
        • Anastassiadis K
        • Kranz A
        • et al.
        MLL2, not MLL1, plays a major role in sustaining MLL-rearranged acute myeloid leukemia.
        Cancer cell. 2017; 31 (e6): 755-770
        • Wang P
        • Lin C
        • Smith ER
        • et al.
        Global analysis of H3K4 methylation defines MLL family member targets and points to a role for MLL1-mediated H3K4 methylation in the regulation of transcriptional initiation by RNA polymerase II.
        Mol Cell Biol. 2009; 29: 6074-6085
        • Krivtsov AV
        • Evans K
        • Gadrey JY
        • et al.
        A menin-MLL inhibitor induces specific chromatin changes and eradicates disease in models of MLL-rearranged leukemia.
        Cancer Cell. 2019; 36 (e11): 660-673
        • Dzama MM
        • Steiner M
        • Rausch J
        • et al.
        Synergistic targeting of FLT3 mutations in AML via combined menin-MLL and FLT3 inhibition.
        Blood. 2020; ([Online ahead of print.])https://doi.org/10.1182/blood.2020005037
        • Uckelmann HJ
        • Kim SM
        • Wong EM
        • et al.
        Therapeutic targeting of preleukemia cells in a mouse model of NPM1 mutant acute myeloid leukemia.
        Science. 2020; 367: 586-590
        • Walf-Vorderwulbecke V
        • de Boer J
        • Horton SJ
        • et al.
        Frat2 mediates the oncogenic activation of Rac by MLL fusions.
        Blood. 2012; 120: 4819-4828
        • Wang Y
        • Krivtsov AV
        • Sinha AU
        • et al.
        The Wnt/beta-catenin pathway is required for the development of leukemia stem cells in AML.
        Science. 2010; 327: 1650-1653
        • Wilkinson AC
        • Ballabio E
        • Geng H
        • et al.
        RUNX1 is a key target in t(4;11) leukemias that contributes to gene activation through an AF4-MLL complex interaction.
        Cell Rep. 2013; 3: 116-127
        • Voigt P
        • Tee WW
        • Reinberg D
        A double take on bivalent promoters.
        Genes Dev. 2013; 27: 1318-1338
        • O'Carroll D
        • Erhardt S
        • Pagani M
        • Barton SC
        • Surani MA
        • Jenuwein T
        The polycomb-group gene Ezh2 is required for early mouse development.
        Mol Cell Biol. 2001; 21: 4330-4336
        • Su IH
        • Basavaraj A
        • Krutchinsky AN
        • et al.
        Ezh2 controls B cell development through histone H3 methylation and Igh rearrangement.
        Nat Immunol. 2003; 4: 124-131
        • Morin RD
        • Johnson NA
        • Severson TM
        • et al.
        Somatic mutations altering EZH2 (Tyr641) in follicular and diffuse large B-cell lymphomas of germinal-center origin.
        Nat Genet. 2010; 42: 181-185
        • Ernst T
        • Chase AJ
        • Score J
        • et al.
        Inactivating mutations of the histone methyltransferase gene EZH2 in myeloid disorders.
        Nat Genet. 2010; 42: 722-726
        • Sashida G
        • Harada H
        • Matsui H
        • et al.
        Ezh2 loss promotes development of myelodysplastic syndrome but attenuates its predisposition to leukaemic transformation.
        Nat Commun. 2014; 5: 4177
        • Basheer F
        • Giotopoulos G
        • Meduri E
        • et al.
        Contrasting requirements during disease evolution identify EZH2 as a therapeutic target in AML.
        J Exp Med. 2019; 216: 966-981
        • Goyama S
        • Nitta E
        • Yoshino T
        • et al.
        EVI-1 interacts with histone methyltransferases SUV39H1 and G9a for transcriptional repression and bone marrow immortalization.
        Leukemia. 2010; 24: 81-88
        • Lehnertz B
        • Pabst C
        • Su L
        • et al.
        The methyltransferase G9a regulates HoxA9-dependent transcription in AML.
        Genes Dev. 2014; 28: 317-327
        • Haihua Chu S
        • Chabon JR
        • Minehart J
        • et al.
        Loss of lysine histone methyltransferase Setd2 disrupts normal hematopoiesis, lineage commitment and reveals a novel role for H3K36me3 in immunoglobulin VDJ recombination.
        Blood. 2016; 128: 423
        • Zhu X
        • He F
        • Zeng H
        • et al.
        Identification of functional cooperative mutations of SETD2 in human acute leukemia.
        Nat Genet. 2014; 46: 287-293
        • Mar BG
        • Chu SH
        • Kahn JD
        • et al.
        SETD2 alterations impair DNA damage recognition and lead to resistance to chemotherapy in leukemia.
        Blood. 2017; 130: 2631-2641
        • Skucha A
        • Ebner J
        • Schmollerl J
        • et al.
        MLL-fusion-driven leukemia requires SETD2 to safeguard genomic integrity.
        Nature Commun. 2018; 9: 1983
        • Jung J
        • Buisman SC
        • Weersing E
        • et al.
        CBX7 induces self-renewal of human normal and malignant hematopoietic stem and progenitor cells by canonical and non-canonical interactions.
        Cell Rep. 2019; 26 (e8): 1906-1918
        • Gough SM
        • Lee F
        • Yang F
        • et al.
        NUP98-PHF23 is a chromatin-modifying oncoprotein that causes a wide array of leukemias sensitive to inhibition of PHD histone reader function.
        Cancer Discov. 2014; 4: 564-577
        • Feng Y
        • Yang Y
        • Ortega MM
        • et al.
        Early mammalian erythropoiesis requires the Dot1L methyltransferase.
        Blood. 2010; 116: 4483-4491
        • Bernt KM
        • Zhu N
        • Sinha AU
        • et al.
        MLL-rearranged leukemia is dependent on aberrant H3K79 methylation by DOT1L.
        Cancer Cell. 2011; 20: 66-78
        • Chen CW
        • Koche RP
        • Sinha AU
        • et al.
        DOT1L inhibits SIRT1-mediated epigenetic silencing to maintain leukemic gene expression in MLL-rearranged leukemia.
        Nat Med. 2015; 21: 335-343
        • Rau RE
        • Rodriguez BA
        • Luo M
        • et al.
        DOT1L as a therapeutic target for the treatment of DNMT3A-mutant acute myeloid leukemia.
        Blood. 2016; 128: 971-981
        • Cheung N
        • Fung TK
        • Zeisig BB
        • et al.
        Targeting aberrant epigenetic networks mediated by PRMT1 and KDM4C in acute myeloid leukemia.
        Cancer Cell. 2016; 29: 32-48
        • Greenblatt SM
        • Man N
        • Hamard PJ
        • et al.
        CARM1 is essential for myeloid leukemogenesis but dispensable for normal hematopoiesis.
        Cancer Cell. 2018; 34: 868
        • Hamard PJ
        • Santiago GE
        • Liu F
        • et al.
        PRMT5 Regulates DNA repair by controlling the alternative splicing of histone-modifying enzymes.
        Cell Rep. 2018; 24: 2643-2657
        • Saleque S
        • Kim J
        • Rooke HM
        • Orkin SH
        Epigenetic regulation of hematopoietic differentiation by Gfi-1 and Gfi-1b is mediated by the cofactors CoREST and LSD1.
        Mol Cell. 2007; 27: 562-572
        • Thambyrajah R
        • Mazan M
        • Patel R
        • et al.
        GFI1 proteins orchestrate the emergence of haematopoietic stem cells through recruitment of LSD1.
        Nat Cell Biol. 2016; 18: 21-32
        • Harris WJ
        • Huang X
        • Lynch JT
        • et al.
        The histone demethylase KDM1A sustains the oncogenic potential of MLL-AF9 leukemia stem cells.
        Cancer Cell. 2012; 21: 473-487
        • He J
        • Nguyen AT
        • Zhang Y
        KDM2b/JHDM1b, an H3K36me2-specific demethylase, is required for initiation and maintenance of acute myeloid leukemia.
        Blood. 2011; 117: 3869-3880
        • Andricovich J
        • Kai Y
        • Peng W
        • Foudi A
        • Tzatsos A
        Histone demethylase KDM2B regulates lineage commitment in normal and malignant hematopoiesis.
        J Clin Invest. 2016; 126: 905-920
        • van den Boom V
        • Maat H
        • Geugien M
        • et al.
        Non-canonical PRC1.1 targets active genes independent of H3K27me3 and is essential for leukemogenesis.
        Cell Rep. 2016; 14: 332-346
        • Kim JY
        • Kim KB
        • Eom GH
        • et al.
        KDM3B is the H3K9 demethylase involved in transcriptional activation of lmo2 in leukemia.
        Mol Cell Biol. 2012; 32: 2917-2933
        • Chen M
        • Zhu N
        • Liu X
        • et al.
        JMJD1C is required for the survival of acute myeloid leukemia by functioning as a coactivator for key transcription factors.
        Genes Dev. 2015; 29: 2123-2139
        • Agger K
        • Miyagi S
        • Pedersen MT
        • Kooistra SM
        • Johansen JV
        • Helin K
        Jmjd2/Kdm4 demethylases are required for expression of Il3ra and survival of acute myeloid leukemia cells.
        Genes Dev. 2016; 30: 1278-1288
        • Agger K
        • Nishimura K
        • Miyagi S
        • Messling JE
        • Rasmussen KD
        • Helin K
        The KDM4/JMJD2 histone demethylases are required for hematopoietic stem cell maintenance.
        Blood. 2019; 134: 1154-1158
        • Thieme S
        • Gyarfas T
        • Richter C
        • et al.
        The histone demethylase UTX regulates stem cell migration and hematopoiesis.
        Blood. 2013; 121: 2462-2473
        • Boila LD
        • Chatterjee SS
        • Banerjee D
        • Sengupta A
        KDM6 and KDM4 histone lysine demethylases emerge as molecular therapeutic targets in human acute myeloid leukemia.
        Exp Hematol. 2018; 58 (e7): 44-51
        • Biswas M
        • Chatterjee SS
        • Boila LD
        • Chakraborty S
        • Banerjee D
        • Sengupta A
        MBD3/NuRD loss participates with KDM6A program to promote DOCK5/8 expression and Rac GTPase activation in human acute myeloid leukemia.
        FASEB J. 2019; 33: 5268-5286
        • Gozdecka M
        • Meduri E
        • Mazan M
        • et al.
        UTX-mediated enhancer and chromatin remodeling suppresses myeloid leukemogenesis through noncatalytic inverse regulation of ETS and GATA programs.
        Nat Genet. 2018; 50: 883-894
        • Wei Y
        • Zheng H
        • Bao N
        • et al.
        KDM6B overexpression activates innate immune signaling and impairs hematopoiesis in mice.
        Blood Adv. 2018; 2: 2491-2504
        • Nakashima K
        • Arai S
        • Suzuki A
        • et al.
        PAD4 regulates proliferation of multipotent haematopoietic cells by controlling c-myc expression.
        Nat Commun. 2013; 4: 1836
        • Kolodziej S
        • Kuvardina ON
        • Oellerich T
        • et al.
        PADI4 acts as a coactivator of Tal1 by counteracting repressive histone arginine methylation.
        Nat Commun. 2014; 5: 3995
        • Song G
        • Shi L
        • Guo Y
        • et al.
        A novel PAD4/SOX4/PU.1 signaling pathway is involved in the committed differentiation of acute promyelocytic leukemia cells into granulocytic cells.
        Oncotarget. 2016; 7: 3144-3157
        • Schwartzentruber J
        • Korshunov A
        • Liu XY
        • et al.
        Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma.
        Nature. 2012; 482: 226-231
        • Wu G
        • Broniscer A
        • McEachron TA
        • et al.
        Somatic histone H3 alterations in pediatric diffuse intrinsic pontine gliomas and non-brainstem glioblastomas.
        Nat Genet. 2012; 44: 251-253
        • Lehnertz B
        • Zhang YW
        • Boivin I
        • et al.
        H3(K27M/I) mutations promote context-dependent transformation in acute myeloid leukemia with RUNX1 alterations.
        Blood. 2017; 130: 2204-2214
        • Gollner S
        • Oellerich T
        • Agrawal-Singh S
        • et al.
        Loss of the histone methyltransferase EZH2 induces resistance to multiple drugs in acute myeloid leukemia.
        Nat Med. 2017; 23: 69-78
        • Boileau M
        • Shirinian M
        • Gayden T
        • et al.
        Mutant H3 histones drive human pre-leukemic hematopoietic stem cell expansion and promote leukemic aggressiveness.
        Nat Commun. 2019; 10: 2891
        • Ezponda T
        • Dupere-Richer D
        • Will CM
        • et al.
        UTX/KDM6A loss enhances the malignant phenotype of multiple myeloma and sensitizes cells to EZH2 inhibition.
        Cell Rep. 2017; 21: 628-640
        • Neff T
        • Sinha AU
        • Kluk MJ
        • et al.
        Polycomb repressive complex 2 is required for MLL-AF9 leukemia.
        Proc Natl Acad Sci USA. 2012; 109: 5028-5033
        • Ng SW
        • Mitchell A
        • Kennedy JA
        • et al.
        A 17-gene stemness score for rapid determination of risk in acute leukaemia.
        Nature. 2016; 540: 433-437
        • Shlush LI
        • Zandi S
        • Mitchell A
        • et al.
        Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia.
        Nature. 2014; 506: 328-333
        • Radzisheuskaya A
        • Shliaha PV
        • Grinev V
        • et al.
        PRMT5 methylome profiling uncovers a direct link to splicing regulation in acute myeloid leukemia.
        Nat Struct Mol Biol. 2019; 26: 999-1012
        • Chatterjee SS
        • Biswas M
        • Boila LD
        • Banerjee D
        • Sengupta A
        SMARCB1 deficiency integrates epigenetic signals to oncogenic gene expression program maintenance in human acute myeloid leukemia.
        Mol Cancer Res. 2018; 16: 791-804
        • Sinha S
        • Biswas M
        • Chatterjee SS
        • Kumar S
        • Sengupta A
        Pbrm1 steers mesenchymal stromal cell osteolineage differentiation by integrating PBAF-dependent chromatin remodeling and BMP/TGF-beta signaling.
        Cell Rep. 2020; 31107570
        • Wong SH
        • Goode DL
        • Iwasaki M
        • et al.
        The H3K4-methyl epigenome regulates leukemia stem cell oncogenic potential.
        Cancer Cell. 2015; 28: 198-209
        • Cabal-Hierro L
        • van Galen P
        • Prado MA
        • et al.
        Chromatin accessibility promotes hematopoietic and leukemia stem cell activity.
        Nat Commun. 2020; 11: 1406
        • Wang J
        • Li Y
        • Wang P
        • et al.
        Leukemogenic chromatin alterations promote AML leukemia stem cells via a KDM4C-ALKBH5-AXL signaling axis.
        Cell Stem Cell. 2020; 27 (e8): 81-97
        • Kuhn MW
        • Song E
        • Feng Z
        • et al.
        Targeting chromatin regulators inhibits leukemogenic gene expression in NPM1 mutant leukemia.
        Cancer Discov. 2016; 6: 1166-1181
        • Yi G
        • Wierenga ATJ
        • Petraglia F
        • et al.
        Chromatin-based classification of genetically heterogeneous AMLs into two distinct subtypes with diverse stemness phenotypes.
        Cell Rep. 2019; 26 (e6): 1059-1069
        • Zhou J
        • Bi C
        • Cheong LL
        • et al.
        The histone methyltransferase inhibitor, DZNep, up-regulates TXNIP, increases ROS production, and targets leukemia cells in AML.
        Blood. 2011; 118: 2830-2839
        • Shi J
        • Wang E
        • Zuber J
        • et al.
        The polycomb complex PRC2 supports aberrant self-renewal in a mouse model of MLL-AF9;Nras(G12D) acute myeloid leukemia.
        Oncogene. 2013; 32: 930-938
        • Xu B
        • On DM
        • Ma A
        • et al.
        Selective inhibition of EZH2 and EZH1 enzymatic activity by a small molecule suppresses MLL-rearranged leukemia.
        Blood. 2015; 125: 346-357
        • Lenard A
        • Xie HM
        • Pastuer T
        • et al.
        Epigenetic regulation of protein translation in KMT2A-rearranged AML.
        Exp Hematol. 2020; 85: 57-69
        • Schenk T
        • Chen WC
        • Gollner S
        • et al.
        Inhibition of the LSD1 (KDM1A) demethylase reactivates the all-trans-retinoic acid differentiation pathway in acute myeloid leukemia.
        Nat Med. 2012; 18: 605-611
        • Cusan M
        • Cai SF
        • Mohammad HP
        • et al.
        LSD1 inhibition exerts its antileukemic effect by recommissioning PU.1- and C/EBPalpha-dependent enhancers in AML.
        Blood. 2018; 131: 1730-1742
        • Wass M
        • Gollner S
        • Besenbeck B
        • et al.
        A proof of concept phase I/II pilot trial of LSD1 inhibition by tranylcypromine combined with ATRA in refractory/relapsed AML patients not eligible for intensive therapy.
        Leukemia. 2020; ([Online ahead of print.])https://doi.org/10.1038/s41375-020-0892-z
        • Braun TP
        • Coblentz C
        • Smith BM
        • et al.
        Combined inhibition of JAK/STAT pathway and lysine-specific demethylase 1 as a therapeutic strategy in CSF3R/CEBPA mutant acute myeloid leukemia.
        Proc Natl Acad Sci USA. 2020; 117: 13670-13679
        • Deb G
        • Wingelhofer B
        • Amaral FMR
        • et al.
        Pre-clinical activity of combined LSD1 and mTORC1 inhibition in MLL-translocated acute myeloid leukaemia.
        Leukemia. 2020; 34: 1266-1277
        • Cai SF
        • Chu SH
        • Goldberg AD
        • et al.
        Leukemia cell of origin influences apoptotic priming and sensitivity to LSD1 inhibition.
        Cancer Discov. 2020; ([Online ahead of print.])https://doi.org/10.1158/2159-8290.CD-19-1469
        • Li Y
        • Zhang M
        • Sheng M
        • et al.
        Therapeutic potential of GSK-J4, a histone demethylase KDM6B/JMJD3 inhibitor, for acute myeloid leukemia.
        J Cancer Res Clin Oncol. 2018; 144: 1065-1077
        • Stief SM
        • Hanneforth AL
        • Weser S
        • et al.
        Loss of KDM6A confers drug resistance in acute myeloid leukemia.
        Leukemia. 2020; 34: 50-62
        • Arteaga MF
        • Mikesch JH
        • Qiu J
        • et al.
        The histone demethylase PHF8 governs retinoic acid response in acute promyelocytic leukemia.
        Cancer Cell. 2013; 23: 376-389
        • Walter RB
        • Gooley TA
        • Wood BL
        • et al.
        Impact of pretransplantation minimal residual disease, as detected by multiparametric flow cytometry, on outcome of myeloablative hematopoietic cell transplantation for acute myeloid leukemia.
        J Clin Oncol. 2011; 29: 1190-1197
        • Curran E
        • Chen X
        • Corrales L
        • et al.
        Sting pathway activation stimulates potent immunity against acute myeloid leukemia.
        Cell Rep. 2016; 15: 2357-2366
        • Cancer Genome Atlas Research Network
        Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia.
        N Engl J Med. 2013; 368: 2059-2074
        • Lawrence MS
        • Stojanov P
        • Polak P
        • et al.
        Mutational heterogeneity in cancer and the search for new cancer-associated genes.
        Nature. 2013; 499: 214-218
        • Ustun C
        • Miller JS
        • Munn DH
        • Weisdorf DJ
        • Blazar BR
        Regulatory T cells in acute myelogenous leukemia: is it time for immunomodulation?.
        Blood. 2011; 118: 5084-5095
        • Le Dieu R
        • Taussig DC
        • Ramsay AG
        • et al.
        Peripheral blood T cells in acute myeloid leukemia (AML) patients at diagnosis have abnormal phenotype and genotype and form defective immune synapses with AML blasts.
        Blood. 2009; 114: 3909-3916
        • Vago L
        • Perna SK
        • Zanussi M
        • et al.
        Loss of mismatched HLA in leukemia after stem-cell transplantation.
        N Engl J Med. 2009; 361: 478-488
        • Curran EK
        • Godfrey J
        • Kline J
        Mechanisms of immune tolerance in leukemia and lymphoma.
        Trends Immunol. 2017; 38: 513-525
        • Mehta NT
        • Truax AD
        • Boyd NH
        • Greer SF
        Early epigenetic events regulate the adaptive immune response gene CIITA.
        Epigenetics. 2011; 6: 516-525
        • Peng D
        • Kryczek I
        • Nagarsheth N
        • et al.
        Epigenetic silencing of TH1-type chemokines shapes tumour immunity and immunotherapy.
        Nature. 2015; 527: 249-253
        • Janzer A
        • Lim S
        • Fronhoffs F
        • Niazy N
        • Buettner R
        • Kirfel J
        Lysine-specific demethylase 1 (LSD1) and histone deacetylase 1 (HDAC1) synergistically repress proinflammatory cytokines and classical complement pathway components.
        Biochem Biophys Res Commun. 2012; 421: 665-670
        • Gray SM
        • Amezquita RA
        • Guan T
        • Kleinstein SH
        • Kaech SM
        Polycomb repressive complex 2-mediated chromatin repression guides effector CD8(+) T cell terminal differentiation and loss of multipotency.
        Immunity. 2017; 46: 596-608
        • Yang XP
        • Jiang K
        • Hirahara K
        • et al.
        EZH2 is crucial for both differentiation of regulatory T cells and T effector cell expansion.
        Sci Rep. 2015; 5: 10643
        • Allan RS
        • Zueva E
        • Cammas F
        • et al.
        An epigenetic silencing pathway controlling T helper 2 cell lineage commitment.
        Nature. 2012; 487: 249-253
        • Tumes DJ
        • Onodera A
        • Suzuki A
        • et al.
        The polycomb protein Ezh2 regulates differentiation and plasticity of CD4(+) T helper type 1 and type 2 cells.
        Immunity. 2013; 39: 819-832
        • Liu PS
        • Wang H
        • Li X
        • et al.
        Alpha-ketoglutarate orchestrates macrophage activation through metabolic and epigenetic reprogramming.
        Nat Immunol. 2017; 18: 985-994
        • Friedrich M
        • Henn A
        • Raum T
        • et al.
        Preclinical characterization of AMG 330, a CD3/CD33-bispecific T-cell-engaging antibody with potential for treatment of acute myelogenous leukemia.
        Mol Cancer Ther. 2014; 13: 1549-1557
        • Laszlo GS
        • Gudgeon CJ
        • Harrington KH
        • et al.
        Cellular determinants for preclinical activity of a novel CD33/CD3 bispecific T-cell engager (BiTE) antibody, AMG 330, against human AML.
        Blood. 2014; 123: 554-561
        • Krupka C
        • Kufer P
        • Kischel R
        • et al.
        Blockade of the PD-1/PD-L1 axis augments lysis of AML cells by the CD33/CD3 BiTE antibody construct AMG 330: reversing a T-cell-induced immune escape mechanism.
        Leukemia. 2016; 30: 484-491
        • Daver N
        • Boddu P
        • Garcia-Manero G
        • et al.
        Hypomethylating agents in combination with immune checkpoint inhibitors in acute myeloid leukemia and myelodysplastic syndromes.
        Leukemia. 2018; 32: 1094-1105
        • Daver N BS G-MG
        • Cortes JE
        • et al.
        Phase IB/II study of nivolumab with azacytidine (AZA) in patients (pts) with relapsed AML [Abstract].
        J Clin Oncol. 2017; 35: 7026
        • Sasidharan Nair V
        • El Salhat H
        • Taha RZ
        • John A
        • Ali BR
        • Elkord E
        DNA methylation and repressive H3K9 and H3K27 trimethylation in the promoter regions of PD-1, CTLA-4, TIM-3, LAG-3, TIGIT, and PD-L1 genes in human primary breast cancer.
        Clin Epigenet. 2018; 10: 78
        • de Boer B
        • Prick J
        • Pruis MG
        • et al.
        Prospective isolation and characterization of genetically and functionally distinct AML subclones.
        Cancer Cell. 2018; 34 (e8): 674-689
        • Hirsch P
        • Tang R
        • Abermil N
        • et al.
        Precision and prognostic value of clone-specific minimal residual disease in acute myeloid leukemia.
        Haematologica. 2017; 102: 1227-1237
        • Caiado F
        • Maia-Silva D
        • Jardim C
        • et al.
        Lineage tracing of acute myeloid leukemia reveals the impact of hypomethylating agents on chemoresistance selection.
        Nat Commun. 2019; 10: 4986