Alternative titles; symbolsHLA-B-ASSOCIATED TRANSCRIPT 8; BAT8G9ANG36HGNC Approved Gene Symbol: EHMT2Cytogenetic location: 6p21.33 Genomic coordinates (GRCh3...
Alternative titles; symbols
HGNC Approved Gene Symbol: EHMT2
Cytogenetic location: 6p21.33 Genomic coordinates (GRCh38): 6:31,879,758-31,897,697 (from NCBI)
The EHMT2 gene encodes an enzyme that, with EHMT1 (607001), comprises a histone methyltransferase complex that methylates histone H3 (see 602812) on lysine 9 (H3K9me2). This methylation is associated with gene silencing in euchromatin (summary by Schaefer et al., 2009).
▼ Cloning and Expression
The class III region of the human MHC on chromosome 6 contains at least 36 genes, most of which are unrelated to each other or to the histocompatibility antigens. By screening a monocytic cell line cDNA library with a genomic cosmid insert, Milner and Campbell (1993) obtained a full-length cDNA, called G9A, corresponding to the BAT8 gene localized to 6p21.3 by Spies et al. (1989). The G9A gene encodes an apparently intracellular protein of 1,001 amino acids with 24 contiguous glutamate residues in the N terminus and 6 contiguous copies of a 33-amino acid ankyrin repeat in the C terminus. Northern blot analysis revealed expression of a 3.4-kb G9A transcript in monocytic, macrophage, hepatocyte, T-cell, B-cell, epithelial, and promyelocytic cell lines. Western blot analysis showed that G9A is expressed as an approximately 98-kD protein. The presence of the ankyrin repeats suggested that G9A is involved in intracellular protein-protein interaction.
Sampath et al. (2007) identified a conserved methylation motif at lys165 (K165) in the noncatalytic N-terminal half of G9A similar to the histone H3 (see 602812) lys9 (H3K9) methylation motif. Following the ankyrin repeats in its catalytic C-terminal half, G9A has a SET domain (see 604396) flanked by pre- and post-SET domains.
Nishida et al. (2007) reported that the 1,210-amino acid G9A protein contains an evolutionarily conserved G9A homology domain (GHD) in its N-terminal half that shares low homology with a RING finger motif. Western blot analysis detected endogenous G9A at an apparent molecular mass of 180 kD in human embryonic kidney cells. Epitope-tagged G9A colocalized with ZNF200 (603231) in most areas of the nucleus as discrete speckles. During metaphase, the 2 proteins colocalized outside of mitotic chromosomes.
▼ Gene Function
Tachibana et al. (2001) demonstrated that G9A is a lysine-preferring histone methyltransferase. Like Suv39h1 (300254), G9A transfers methyl groups to the lysine residues of histone H3, but with a 10- to 20-fold higher activity. G9A was able to add methyl groups to lys27 as well as lys9 in histone H3, compared with Suv39h1, which was able only to methylate lys9. Tachibana et al. (2001) demonstrated that G9A is localized to the nucleus but not in the repressive chromatin domains of centromeric loci, in which Suv39h1 family proteins were localized.
Using mass spectrometry, Ueda et al. (2006) identified Wiz (619715) as a component of the G9a/Glp complex in mouse embryonic stem cells. Wiz interacted with both G9a and Glp in the complex. The interactions were mediated by the SET domains of G9a and Glp and by zinc finger motif-6 of Wiz. Knockout analysis revealed that binding of Wiz to both G9a and Glp was more stable in the G9a/Glp heteromeric complex and that Wiz, in turn, contributed to the stability of G9a in the complex. The authors also identified 2 PxDLS-like putative CTBP (see 602618)-binding motifs in Wiz, which connected the G9a/Glp heteromeric complex with CTBP corepressors.
Sampath et al. (2007) found that G9A methylated itself at K165. Automethylation at K165 was necessary and sufficient to mediate in vivo interaction of G9A with heterochromatin protein-1 (HP1; see 604478), and this interaction was reversed by phosphorylation of the adjacent residue in G9A, thr166. NMR analysis indicated that the chromodomain of HP1 recognized methyl-G9A through a binding mode similar to that used in recognition of methyl-H3K9. Methyltransferase activity of G9A was necessary for H3K9 dimethylation, HP1-gamma (CBX3; 604477) localization to euchromatin, and proper in vivo control of gene expression. A mutation of G9A K165 that interfered with methylation caused only a mild increase in the amount of HP1-gamma redistributed from euchromatin to pericentric heterochromatin and significant dysregulation of only the HEY1 gene (602953).
By yeast 2-hybrid analysis of a fetal brain cDNA library, Nishida et al. (2007) showed that the N-terminal GHD of G9A interacted with ZNF200. Immunoprecipitation analysis confirmed interaction between endogenous G9A and ZNF200 in human embryonic kidney cells. Mutation analysis indicated that at least 2 of the 5 zinc finger domains of ZNF200 were required for its interaction with G9A. Mouse G9a strongly repressed transcription of a reporter gene, and ZNF200 had no effect on this repression.
The Air noncoding RNA (604893) is imprinted, being monoallelically expressed from the paternal allele. Air is required for allele-specific silencing of the cis-linked Slc22a3 (604842), Slc22a2 (602608), and Igf2r (147280) genes in mouse placenta. Nagano et al. (2008) demonstrated that Air interacts with the Slc22a3 promoter chromatin and the H3K9 histone methyltransferase G9a in placenta. Air accumulates at the Slc22a3 promoter in correlation with localized H3K9 methylation and transcriptional repression. Genetic ablation of G9a results in nonimprinted, biallelic transcription of Slc22a3. Truncated Air fails to accumulate at the Slc22a3 promoter, which results in reduced G9a recruitment and biallelic transcription. Nagano et al. (2008) concluded that Air, and potentially other large noncoding RNAs, target repressive histone-modifying activities through molecular interaction with specific chromatin domains to epigenetically silence transcription.
Using a yeast 2-hybrid assay, Ding et al. (2008) showed that the mediator subunit MED12 (300188) interacted directly with G9A in a human fetal brain cDNA library. Mutation analysis showed that the pro-glu-leu (PQL) region of the C-terminal domain of MED12 interacted with an ankyrin-repeat domain on G9A and, more weakly, with a cys-rich domain on G9A. Ding et al. (2008) found that purified HeLa cell mediator complexes that included MED12 interacted directly with G9A and REST (600571), a gene repressor that functions through repressor element-1 (RE1). Endogenous REST in HEK293 cells suppressed expression of a reporter gene bearing RE1 sites, and knockdown of either MED12 or G9A abrogated the suppression. Depletion of MED12 significantly reduced the association of G9A with RE1 elements and decreased the level of H3K9 dimethylation by G9A without influencing RE1 site occupancy by REST.
Maze et al. (2010) identified an essential role for H3K9 dimethylation and the lysine dimethyltransferase G9a in cocaine-induced structural and behavioral plasticity in mouse. Repeated cocaine administration reduced global levels of H3K9 dimethylation in the nucleus accumbens. This reduction in histone methylation was mediated through the repression of G9a in this brain region, which was regulated by the cocaine-induced transcription factor delta-FosB (164772). Using conditional mutagenesis and viral-mediated gene transfer, Maze et al. (2010) found that G9a downregulation increased the dendritic spine plasticity of nucleus accumbens neurons and enhanced the preference for cocaine, thereby establishing a crucial role for histone methylation in the long-term actions of cocaine.
Using mouse embryonic stem cells (ESCs), Maier et al. (2015) confirmed interaction between 2 major repressive histone methyltransferase complexes, PRC2 (see EZH1, 601674) and G9a-Glp (EHMT1). Moreover, the complexes shared several interaction partners, including Znf518a (617733) and Znf518b (617734). In vitro, Znf518b interacted directly with G9a and with the 2 alternative PRC2 methyltransferase subunits, Ezh1 and Ezh2 (601573). Knockdown of Znf518b in mouse ESCs reduced global H3K9 dimethylation. Maier et al. (2015) concluded that ZNF518B may mediate association between PRC2 and G9A-GLP and regulate G9A-GLP activity.
Pan et al. (2015) found that G9A expression was unregulated in human pancreatic cancer cells and that upregulation of G9A increased methylation of H3K9 and H3K27 on the promoter of E-cadherin (CDH1; 192090) to downregulate its expression. Specifically, G9A bound directly to the promoter of PCL3 (PHF19; 609740), a component of polycomb repressive complex-2 (PRC2), and increased PCL3 expression. By increasing PCL3 expression, G9A increased recruitment of PRC2 to the E-cadherin promoter to downregulate its expression by repressing expression of KDM7A (619640). Bioinformatic analysis and in vivo study with mice supported these results and indicated that G9A likely orchestrated PCL3 and KDM7A to inhibit E-cadherin in pancreatic cancer.
Using mass spectrometry, Bian et al. (2015) identified ZNF644 (614159) and WIZ as subunits of the G9A/GLP complex in 293T cells. The N terminus of ZNF644 and the C terminus of WIZ interacted with the complex through the transcription activation domains of G9A and GLP, respectively. ZNF644 and WIZ bound to chromatin and facilitated localization of the G9A/GLP complex to chromatin. Chromatin immunoprecipitation-sequencing analysis showed that WIZ and ZNF644 associated with G9A at the promoter regions of specific loci and targeted G9A and GLP to genomic loci for transcriptional repression.
Using tissue recombination assays, Dutta et al. (2016) showed that loss of function of Nkx3.1 (602041) in mouse prostate resulted in downregulation of genes essential for prostate differentiation and upregulation of genes not normally expressed in prostate. Gain of function of Nkx3.1 in fully differentiated nonprostate mouse epithelium was sufficient for respecification to prostate in grafts placed under the kidney capsule. In human prostate cells, these activities required interaction of NKX3.1 with G9A methyltransferase via the NKX3.1 homeodomain and were mediated by activation of target genes such as UTY (400009). Dutta et al. (2016) proposed that the NKX3.1-EHMT2-UTY transcriptional regulatory network is essential for prostate differentiation and that disruption of such a network predisposes to prostate cancer.
▼ Gene Structure
The BAT8 gene contains 3 exons (Milner and Campbell, 1993).
By chromosome walking studies and screening of cDNAs with cosmid probes, Spies et al. (1989) mapped the BAT8 gene to chromosome 6p21.3.
▼ Animal Model
Tachibana et al. (2002) developed G9a-deficient mice and embryonic stem cells. G9a-null embryos displayed severe growth retardation and early lethality. Embryos that were recovered at day 9.5 resembled wildtype embryos at days 8.0 to 8.5, and no organ-specific abnormalities were apparent. H3-K9 methylation was drastically decreased in G9a-deficient embryos, leading to accumulation of acetylated H3-K9 and methylated H3-K4. G9a-deficient embryonic stem cells also exhibited reduced H3-K9 methylation, indicating that G9a is a dominant histone methyltransferase in vivo. Loss of G9a abolished methylated H3-K9 mostly in euchromatic regions. G9a exerted a transcriptionally suppressive function that depended on its histone methyltransferase activity.
Schaefer et al. (2009) found that conditional ablation of Ehmt2 (G9a) or Ehmt1 (607001) in postnatal mouse forebrain neurons caused a reduction in euchromatic H3K9 methylation in the forebrain and upregulation/derepression of neuronal and nonneuronal genes (e.g., AFP; 104150), including those involved in developmental stage-dependent gene expression (e.g., Dach2; 300608). These changes were not associated with alterations in neuronal architecture or electrophysiologic features. Mice with postnatal knockout of Ehmt1 or Ehmt2 in the forebrain showed decreased exploratory behavior in a novel environment and had a decrease in the preference for sucrose solution compared to wildtype mice, the latter of which may indicate an underlying dysfunction in motivation/reward. Both Ehmt2-null and Ehmt1-null mice became obese. Mice lacking Ehmt2 specifically in Drd1 (126449)- or Drd2 (126450)-expressing neurons in the striatum showed altered locomotor and behavioral responses to Drd-specific receptor agonist or antagonists, reflecting reductions of cell type-specific activity. The changes in these mice resembled the features of the human chromosome 9q34.3 deletion syndrome (610253), which is due to mutations in EHMT1. Schaefer et al. (2009) concluded that Ehmt1 and Ehmt2 are key regulators of cognition and adaptive behavior in adult mice through regulation of transcriptional homeostasis.