Martina P. Pasillas

NUP98-NSD1 links H3K36 methylation to Hox-A gene activation and leukaemogenesis.

Nuclear receptor-binding SET domain protein 1 (NSD1) prototype is a family of mammalian histone methyltransferases (NSD1, NSD2/MMSET/WHSC1, NSD3/WHSC1L1) that are essential in development and are mutated in human acute myeloid leukemia (AML), overgrowth syndromes, multiple myeloma and lung cancers. In AML, the recurring t(5;11)(q35;p15.5) translocation fuses NSD1 to nucleoporin-98 (NUP98). Here, we present the first characterization of the transforming properties and molecular mechanisms of NUP98–NSD1. We demonstrate that NUP98–NSD1 induces AML in vivo, sustains self-renewal of myeloid stem cells in vitro, and enforces expression of the HoxA7, HoxA9, HoxA10 and Meis1 proto-oncogenes. Mechanistically, NUP98–NSD1 binds genomic elements adjacent to HoxA7 and HoxA9, maintains histone H3 Lys 36 (H3K36) methylation and histone acetylation, and prevents EZH2-mediated transcriptional repression of the Hox-A locus during differentiation. Deletion of the NUP98 FG-repeat domain, or mutations in NSD1 that inactivate the H3K36 methyltransferase activity or that prevent binding of NUP98–NSD1 to the Hox-A locus precluded both Hox-A gene activation and myeloid progenitor immortalization. We propose that NUP98–NSD1 prevents EZH2-mediated repression of Hox-A locus genes by colocalizing H3K36 methylation and histone acetylation at regulatory DNA elements. This report is the first to link deregulated H3K36 methylation to tumorigenesis and to link NSD1 to transcriptional regulation of the Hox-A locus.

Quantitative production of macrophages or neutrophils ex vivo using conditional Hoxb8

Differentiation mechanisms and inflammatory functions of neutrophils and macrophages are usually studied by genetic and biochemical approaches that require costly breeding and time-consuming purification to obtain phagocytes for functional analysis. Because Hox oncoproteins enforce self-renewal of factor-dependent myeloid progenitors, we queried whether estrogen-regulated Hoxb8 (ER-Hoxb8) could immortalize macrophage or neutrophil progenitors that would execute normal differentiation and normal innate immune function upon ER-Hoxb8 inactivation. Here we describe methods to derive unlimited quantities of mouse macrophages or neutrophils by immortalizing their respective progenitors with ER-Hoxb8 using different cytokines to target expansion of different committed progenitors. ER-Hoxb8 neutrophils and macrophages are functionally superior to those produced by many other ex vivo differentiation models, have strong inflammatory responses and can be derived easily from embryonic day 13 (e13) fetal liver of mice exhibiting embryonic-lethal phenotypes. Using knockout or small interfering RNA (siRNA) technologies, this ER-Hoxb8 phagocyte maturation system represents a rapid analytical tool for studying macrophage and neutrophil biology.

Hoxa9 immortalizes a granulocyte-macrophage colony-stimulating factor-dependent promyelocyte capable of biphenotypic differentiation to neutrophils or macrophages, independent of enforced meis expression.

ABSTRACT The genes encoding Hoxa9 and Meis1 are transcriptionally coactivated in a subset of acute myeloid leukemia (AML) in mice. In marrow reconstitution experiments, coexpression of both genes produces rapid AML, while neither gene alone generates overt leukemia. Although Hoxa9 and Meis1 can bind DNA as heterodimers, both can also heterodimerize with Pbx proteins. Thus, while their coactivation may result from the necessity to bind promoters as heterodimers, it may also result from the necessity of altering independent biochemical pathways that cooperate to generate AML, either as monomers or as heterodimers with Pbx proteins. Here we demonstrate that constitutive expression of Hoxa9 in primary murine marrow immortalizes a late myelomonocytic progenitor, preventing it from executing terminal differentiation to granulocytes or monocytes in the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF) or interleukin-3. This immortalized phenotype is achieved in the absence of endogenous or exogenous Meis gene expression. The Hoxa9-immortalized progenitor exhibited a promyelocytic transcriptional profile, expressing PU.1, AML1, c-Myb, C/EBP alpha, and C/EBP epsilon as well as their target genes, the receptors for GM-CSF, G-CSF, and M-CSF and the primary granule proteins myeloperoxidase and neutrophil elastase. G-CSF obviated the differentiation block of Hoxa9, inducing neutrophilic differentiation with accompanying expression of neutrophil gelatinase B and upregulation of gp91phox. M-CSF also obviated the differentiation block, inducing monocytic differentiation with accompanying expression of the macrophage acetyl-low-density lipoprotein scavenger receptor and F4/80 antigen. Versions of Hoxa9 lacking the ANWL Pbx interaction motif (PIM) also immortalized a promyelocytic progenitor with intrinsic biphenotypic differentiation potential. Therefore, Hoxa9 evokes a cytokine-selective block in differentiation by a mechanism that does not require Meisgene expression or interaction with Pbx through the PIM.

Nup98-HoxA9 immortalizes myeloid progenitors, enforces expression of Hoxa9, Hoxa7 and Meis1, and alters cytokine-specific responses in a manner similar to that induced by retroviral co-expression of Hoxa9 and Meis1.

The association between acute myeloid leukaemia (AML) and the aberrant expression of Hoxa9 is evidenced by (1) proviral activation of Hoxa9 and Meis1 in BXH-2 murine AML, (2) formation of the chimeric Nup98-HoxA9 transactivator protein as a consequence of the t(7;11) translocation in human AML, and (3) the strong expression of HoxA9 and Meis1 in human AML. In mouse models, enforced retroviral expression of Hoxa9 alone in marrow is not sufficient to cause rapid AML, while co-expression of Meis1 and Hoxa9 induces rapid AML. In contrast, retroviral expression of Nup98-HoxA9 is sufficient to cause rapid AML in the absence of enforced Meis1 expression. Previously, we demonstrated that Hoxa9 could block the differentiation of murine marrow progenitors cultured in granulocyte-macrophage colony-simulating factor (GM–CSF). These progenitors lacked Meis1 expression, could not proliferate in stem cell factor (SCF), but could differentiate into neutrophils when switched into granulocyte colony-simulating factor (G-CSF). Ectopic expression of Meis1 in these Hoxa9 cells suppressed their G-CSF-induced differentiation, permitted proliferation in SCF, and therein offered a potential explanation of cooperative function. Because Meis1 binds N-terminal Hoxa9 sequences that are replaced by Nup98, we hypothesized that Nup98-HoxA9 might consolidate the biochemical functions of both Hoxa9 and Meis1 on target gene promoters and might evoke their same lymphokine-responsive profile in immortalized progenitors. Here we report that Nup98-HoxA9, indeed mimicks Hoxa9 plus Meis1 coexpression – it immortalizes myeloid progenitors, prevents differentiation in response to GM–CSF, IL-3, G-CSF, and permits proliferation in SCF. Unexpectedly, however, Nup98-Hoxa9 also enforced strong transcription of the cellular Hoxa9, Hoxa7 and Meis1 genes at levels similar to those found in mouse AMLs generated by proviral activation of Hoxa9 and Meis1. Using Hoxa9−/− marrow, we demonstrate that expression of Hoxa9 is not required for myeloid immortalization by Nup98-HoxA9. Rapid leukaemogenesis by Nup98-HoxA9 may therefore result from both the intrinsic functions of Nup98-HoxA9, as well as of those of coexpressed HOX and MEIS1 genes.

An environment-dependent transcriptional network specifies human microglia identity

Of mice and mens microglia Microglia are immune system cells that function in protecting and maintaining the brain. Gosselin et al. examined the epigenetics and RNA transcripts from single microglial cells and observed consistent profiles among samples despite differences in age, sex, and diagnosis. Mouse and human microglia demonstrated similar microglia-specific gene expression profiles, as well as a shared environmental response among microglia collected either immediately after surgery (ex vivo) or after culturing (in vitro). Interestingly, those genes exhibiting differences in expression between humans and mice or after culturing were often implicated in neurodegenerative diseases. Science, this issue p. eaal3222 Single-cell sequencing of brain microglia reveals ex vivo and in vitro differences in transcription. INTRODUCTION Microglia play essential roles in central nervous system homeostasis and influence diverse aspects of neuronal function, including refinement of synaptic networks and elaboration of neuromodulatory factors for memory and motor learning. Many lines of evidence indicate that dysregulation of microglial functions contributes to the pathogenesis of neurodegenerative diseases, including Alzheimer’s disease and Parkinson’s disease. Emerging evidence from mouse and human studies also suggests that microglia influence neurodevelopmental and psychiatric disorders such as schizophrenia and depression. Most disease risk alleles associated with neurodegenerative diseases reside in noncoding regions of the genome, requiring the delineation of functional genomic elements in the relevant human cell types to establish mechanisms of causation. The recent observation that mouse brain environment strongly influences microglia-specific gene expression has implications for understanding pathogenic responses of microglia in diseases and disorders and modeling their phenotypes in vitro. RATIONALE Although dysregulation of microglial activity is genetically linked to neurodegenerative diseases and psychiatric disorders, no systematic evaluations of human microglia gene expression or regulatory landscapes are currently available. In addition, the extent to which mice provide suitable models for human microglia is unclear. The major goals of this study were to define the transcriptomes and DNA regulatory elements of human microglia ex vivo and in vitro in comparison to the mouse and to systematically relate these features to expression of genes associated with genome-wide association study (GWAS) risk alleles or exhibiting altered expression in neurodegenerative diseases and psychiatric disorders. RESULTS We used RNA sequencing, chromatin immunoprecipitation sequencing, and assay for transposase-accessible chromatin sequencing to characterize the transcriptomes and epigenetic landscapes of human microglia isolated from surgically resected brain tissue in excess of that needed for diagnosis. Although some effects of underlying disease cannot be excluded, the overall pattern of gene expression was markedly consistent. Microglia-enriched genes were found to overlap significantly with genes exhibiting altered expression in neurodegenerative diseases and psychiatric disorders and with genes associated with a wide spectrum of disease-specific risk alleles. Human microglia gene expression was well correlated with mouse microglia gene expression, but numerous species-specific differences were also observed that included genes linked to human disease. More than half of the genes associated with noncoding GWAS risk alleles for Alzheimer’s disease are preferentially expressed in microglia. DNA recognition motifs enriched at active enhancers and expression of the corresponding lineage-determining transcription factors were very similar for human and mouse microglia. Transition of human and mouse microglia from the brain to tissue culture revealed remodeling of their respective enhancer landscapes and extensive down-regulation of genes that are induced in primitive mouse macrophages following migration into the fetal brain. Treatment of microglia in vitro with transforming growth factor β1 (TGF-β1) had relatively modest effects in maintaining the ex vivo pattern of gene expression. A significant subset of the genes up- or down-regulated in vitro exhibited altered expression in neurodegenerative diseases and psychiatric disorders. CONCLUSION These studies identify core features of human microglial transcriptomes and epigenetic landscapes. Intersection of the microglia-specific gene signature with GWAS and transcriptomic data supports roles of microglia as both responders and contributors to disease phenotypes. The identification of an environment-sensitive program of gene expression and corresponding regulatory elements enables inference of a conserved and dynamic transcription factor network that maintains microglia identity and function. The combinations of signaling factors in the brain necessary to maintain microglia phenotypes remain largely unknown. In concert, these findings will inform efforts to generate microglia-like cells in simple and complex culture systems and understand gene-environment interactions that influence homeostatic and pathogenic functions of microglia in the human brain. Brain environment specifies gene expression in microglia. Human microglia transcriptomes and enhancer landscapes were defined ex vivo following purification from surgically resected brain tissue and in vitro after transfer to a tissue culture environment. Dynamic changes in these features enabled delineation of transcription factors controlling an environment-dependent program of gene expression that overlaps with genes that are dysregulated in brain pathologies. Microglia play essential roles in central nervous system (CNS) homeostasis and influence diverse aspects of neuronal function. However, the transcriptional mechanisms that specify human microglia phenotypes are largely unknown. We examined the transcriptomes and epigenetic landscapes of human microglia isolated from surgically resected brain tissue ex vivo and after transition to an in vitro environment. Transfer to a tissue culture environment resulted in rapid and extensive down-regulation of microglia-specific genes that were induced in primitive mouse macrophages after migration into the fetal brain. Substantial subsets of these genes exhibited altered expression in neurodegenerative and behavioral diseases and were associated with noncoding risk variants. These findings reveal an environment-dependent transcriptional network specifying microglia-specific programs of gene expression and facilitate efforts to understand the roles of microglia in human brain diseases.

Meis1a suppresses differentiation by G-CSF and promotes proliferation by SCF: Potential mechanisms of cooperativity with Hoxa9 in myeloid leukemia

Hoxa9 and Meis1a are homeodomain transcription factors that heterodimerize on DNA and are down-regulated during normal myeloid differentiation. Hoxa9 and Meis1a cooperate to induce acute myeloid leukemia (AML) in mice, and are coexpressed in human AML. Despite their cooperativity in leukemogenesis, we demonstrated previously that retroviral expression of Hoxa9 alone—in the absence of coexpressed retroviral Meis1 or of expression of endogenous Meis genes—blocks neutrophil and macrophage differentiation of primary myeloid progenitors cultured in granulocyte–macrophage colony-stimulating factor (GM-CSF). Expression of Meis1 alone did not immortalize any factor-dependent marrow progenitor. Because HoxA9-immortalized progenitors still execute granulocytic differentiation in response to granulocyte CSF (G-CSF) and monocyte differentiation in response to macrophage CSF (M-CSF), we tested the possibility that Meis1a cooperates with Hoxa9 by blocking viable differentiation pathways unaffected by Hoxa9 alone. Here we report that Meis1a suppresses G-CSF-induced granulocytic differentiation of Hoxa9-immortalized progenitors, permitting indefinite self-renewal in G-CSF. Meis1a also reprograms Hoxa9-immortalized progenitors to proliferate, rather than die, in response to stem cell factor (SCF) alone. We propose that Meis1a and Hoxa9 are part of a molecular switch that regulates progenitor abundance by suppressing differentiation and maintaining self-renewal in response to different subsets of cytokines during myelopoiesis. The independent differentiation pathways targeted by Hoxa9 and Meis1a prompt a “cooperative differentiation arrest” hypothesis for a subset of leukemia, in which cooperating transcription factor oncoproteins block complementary subsets of differentiation pathways, establishing a more complete differentiation block in vivo.

Persistent Transactivation by Meis1 Replaces Hox Function in Myeloid Leukemogenesis Models: Evidence for Co-Occupancy of Meis1-Pbx and Hox-Pbx Complexes on Promoters of Leukemia-Associated Genes

ABSTRACT Homeobox transcription factors Meis1 and Hoxa9 promote hematopoietic progenitor self-renewal and cooperate to cause acute myeloid leukemia (AML). While Hoxa9 alone blocks the differentiation of nonleukemogenic myeloid cell-committed progenitors, coexpression with Meis1 is required for the production of AML-initiating progenitors, which also transcribe a group of hematopoietic stem cell genes, including Cd34 and Flt3 (defined as Meis1-related leukemic signature genes). Here, we use dominant trans-activating (Vp16 fusion) or trans-repressing (engrailed fusion) forms of Meis1 to define its biochemical functions that contribute to leukemogenesis. Surprisingly, Vp16-Meis1 (but not engrailed-Meis1) functioned as an autonomous oncoprotein that mimicked combined activities of Meis1 plus Hoxa9, immortalizing early progenitors, inducing low-level expression of Meis1-related signature genes, and causing leukemia without coexpression of exogenous or endogenous Hox genes. Vp16-Meis1-mediated transformation required the Meis1 function of binding to Pbx and DNA but not its C-terminal domain (CTD). The absence of endogenous Hox gene expression in Vp16-Meis1-immortalized progenitors allowed us to investigate how Hox alters gene expression and cell biology in early hematopoietic progenitors. Strikingly, expression of Hoxa9 or Hoxa7 stimulated both leukemic aggressiveness and transcription of Meis1-related signature genes in Vp16-Meis1 progenitors. Interestingly, while the Hoxa9 N-terminal domain (NTD) is essential for cooperative transformation with wild-type Meis1, it was dispensable in Vp16-Meis1 progenitors. The fact that a dominant transactivation domain fused to Meis1 replaces the essential functions of both the Meis1 CTD and Hoxa9 NTD suggests that Meis-Pbx and Hox-Pbx (or Hox-Pbx-Meis) complexes co-occupy cellular promoters that drive leukemogenesis and that Meis1 CTD and Hox NTD cooperate in gene activation. Chromatin immunoprecipitation confirmed co-occupancy of Hoxa9 and Meis1 on the Flt3 promoter.

HoxB8 requires its Pbx-interaction motif to block differentiation of primary myeloid progenitors and of most cell line models of myeloid differentiation

HoxB8 was the first homeobox gene identified as a cause of leukemia. In murine WEHI3B acute myeloid leukemia (AML) cells, proviral integration leads to the expression of both HoxB8 and Interleukin (IL-3). Enforced expression of HoxB8 blocks differentiation of factor-dependent myeloid progenitors, while IL-3 co-expression induces autocrine proliferation and overt leukemogenicity. Previously, we demonstrated that HoxB8 binds DNA cooperatively with members of the Pbx family of transcription factors, and that HoxB8 makes contact with the Pbx homeodomain through a hexameric sequence designated the Pbx-interaction motif (PIM). E2a-Pbx1, an oncogenic derivative of Pbx1, both retains its ability to heterodimerize with Hox proteins and arrest myeloid differentiation. This observation prompts the question of whether E2a-Pbx1 and Hox oncoproteins use endogenous Hox and Pbx proteins, respectively, to target a common set of cellular genes. Here, we use four different models of neutrophil and macrophage differentiation to determine whether HoxB8 needs to bind DNA or Pbx cofactors in order to arrest myeloid differentiation. The ability of HoxB8 to bind DNA or to bind Pbx was essential (1) to block differentiation of factor-dependent myeloid progenitors from primary marrow; (2) to block IL-6-induced monocytic differentiation of M1-AML cells; and (3) to block granulocytic differentiation of GM-CSF-dependent ECoM-G cells. However, while DNA-binding was required, the HoxB8 Pbx-interaction motif was unnecessary for preventing macrophage differentiation of ECoM-M cells. We conclude that HoxB8 prevents differentiation by directly influencing cellular gene expression, and that the genetic context within a cell dictates whether the effect of HoxB8 is dependent on a physical interaction with Pbx proteins.

NSD1 PHD domains bind methylated H3K4 and H3K9 using interactions disrupted by point mutations in human sotos syndrome

Sotos syndrome is a human developmental and cognitive disorder caused by happloinsufficiency of transcription factor NSD1. Similar phenotypes arise from NSD1 gene deletion or from point mutations in 9 of 13 NSD1 domains, including all 6 PHD domains, indicating that each NSD1 domain performs an essential role. To gain insight into the biochemical basis of Sotos syndrome, we tested the ability of each NSD1 PHD domain to bind histone H3 when methylated at regulatory sites Lys4, Lys9, Lys27, Lys36, and Lys79, and histone H4 at regulatory Lys20, and determined whether Sotos point mutations disrupted methylation site‐specific binding. NSD1 PHD domains 1, 4, 5, and 6 bound histone H3 methylated at Lys4 or Lys9. Eleven of 12 Sotos mutations in PHD4, PHD5, and PHD6 disrupted binding to these methylated lysines, and 8 of 9 mutations in PHD4 and PHD6 severely compromised binding to transcription cofactor Nizp1. One mutation in PHD1 did not alter binding to specific methylated histone H3, and one mutation in PHD4 did not alter binding to either methylated histone or Nizp1. Our data suggests that Sotos point mutations in NSD1 PHD domains disrupt its transcriptional regulation by interfering with its ability to bind epigenetic marks and recruit cofactors. Hum Mutat 32:292–298, 2011.

EB-1, a tyrosine kinase signal transduction gene, is transcriptionally activated in the t(1;19) subset of pre-B ALL, which express oncoprotein E2a-Pbx1

The t(1;19) translocation of pre-B cell acute lymphocytic leukemia (ALL) produces E2a-Pbx1, a chimeric oncoprotein containing the transactivation domains of E2a joined to the homeodomain protein, Pbx1. E2a-Pbx1 causes T cell and myeloid leukemia in mice, blocks differentiation of cultured myeloid progenitors, and transforms fibroblasts through a mechanism accompanied by aberrant expression of tissue-specific and developmentally-regulated genes. Here we investigate whether aberrant gene expression also occurs specifically in the t(1;19)-containing subset of pre-B cell ALL in man. Two new genes, EB-1 and EB-2, as well as Caldesmon were transcriptionally activated in each of seven t(1;19) cell lines. EB-1 expression was extremely low in marrow from patients having pre-B ALL not associated with the t(1;19), and elevated more than 100-fold in marrow from patients with pre-B ALL associated with the t(1;19). Normal EB-1 expression was strong in brain and testis, the same tissues exhibiting the highest levels of PBX1 expression. EB-1 encodes a signaling protein containing a phosphotyrosine binding domain homologous to that of dNumb developmental regulators and two SAM domains homologous to those in the C-terminal tail of Eph receptor tyrosine kinases. We conclude that aberrant expression of tissue-specific genes is a characteristic of t(1;19) pre-B ALL, as was previously found in fibroblasts transformed by E2a-Pbx1. Potentially, EB-1 overexpression could interfere with normal signaling controlling proliferation or differentiation.