U.S. Department of Health and Human Services
Kai Ge

 Contact Info

Tel: 301-451-1998
Email: kai.ge@nih.gov

 Select Experience

  • Postdoctoral FellowThe Rockefeller University2000–2003
  • Postdoctoral FellowThe Wistar Institute1997–2000
  • Ph.D.Shanghai Institute of Biochemistry, Chinese Academy of Sciences1997
  • B.S.Fudan University1992

 Related Links

  • Cell Biology/Cell Signaling
  • Chromosome Biology/Epigenetics
  • Developmental Biology
  • Genetics/Genomics
  • Molecular Biology/Biochemistry
Research Summary/In Plain Language

Research Summary

Research Goal

We study epigenomic regulation of PPARγ and adipogenesis (generation of fat tissue).  We also use adipogenesis as a model system to understand epigenomic regulation of cell fate transition, with a focus on transcriptional enhancers.

Current Research

Epigenomic mechanisms, including histone modification, chromatin remodeling, and enhancer-promoter interaction, play critical roles in gene regulation.  Histone acetylation correlates with gene activation while methylation correlates with either activation or repression, depending on the specific lysine (K) residue that gets methylated.  Methylation on K4 of histone H3 (H3K4) correlates with gene activation, whereas methylation on K27 of histone H3 (H3K27) correlates with gene repression.

Enhancers control cell-type-specific gene expression and are critical for differentiation and development.  Primed enhancers are marked by H3K4 mono-methylation (H3K4me1).  Active enhancers are further marked by H3K27 acetylation (H3K27ac).  We identified CBP and p300 as H3K27 acetyltransferases (EMBO J 2011).  We also identified MLL3 (KMT2C) and MLL4 (KMT2D) as major enhancer H3K4me1 methyltransferases (eLife 2013).

PPARγ is a master regulator of adipogenesis. It is a nuclear receptor and thus a ligand-activated transcription factor (TF).  In search for novel PPARγ cofactors, we identified a nuclear protein complex that contains MLL3/MLL4 (MLL3/4) and the H3K27 demethylase UTX (JBC 2007, PNAS 2007).

Epigenomic regulation of enhancers

Combining mouse genetics with next generation sequencing and bioinformatics, we have shown:

  1. H3K4me1 methyltransferases MLL3/4 are required for enhancer activation and cell-type-specific gene expression in differentiation. MLL3/4 are essential for the development of embryos and adipose tissue, muscle, mammary gland, B cells, T cells, and heart (eLife 2013, reviewed in Gene 2017).
  2. Although enhancer priming by MLL3/4 is dispensable for cell-identity maintenance, it controls cell fate transition by orchestrating H3K27 acetyltransferases CBP/p300-mediated enhancer activation (PNAS 2016, Nucleic Acids Res 2017).  
  3. UTX protein, but not its H3K27 demethylase activity, is required for embryonic stem cell differentiation and mouse development (PNAS 2012). Interestingly, UTX demethylase activity is required for stem cell-mediated muscle regeneration (J Clin Invest 2016).
  4. The epigenomic reader Brd4 binds to active enhancers to control cell identity gene induction in adipogenesis and myogenesis (Nat Commun 2017).

Epigenomic and transcriptional regulation of adipogenesis

We have shown that several histone methyltransferases control expression of both positive and negative master regulators of adipogenesis [Reviewed in BBA 2012 and Cell Biosci 2014]:

  1. H3K4 methyltransferases MLL3/4 and associated PTIP directly control the induction of principal adipogenic TFs PPARγ and C/EBPα and are essential for adipogenesis (Cell Metab 2009eLife 2013).
  2. H3K27 methyltransferase Ezh2 uses its enzymatic activity to constitutively represses Wnt genes to facilitate adipogenesis (PNAS 2010).
  3. H3K9 methyltransferase G9a represses PPARγ expression and adipogenesis (EMBO J 2013).

Using conditional knockout mice and preadipocytes, we found that although ligand-bound glucocorticoid receptor (GR) accelerates adipogenesis in culture, endogenous GR is dispensable for adipogenesis in culture and in mice (MCB 2017a).  We also found that KLF4 and Krox20 are dispensable for adipogenesis in culture and in mice (MCB 2017b).  These unexpected results prompted us to study adipogenesis in vivo.

Epigenomic regulation of nuclear receptor target gene expression

We reported that two pairs of histone acetyltransferases (HATs), GCN5/PCAF and CBP/p300, are specifically required for H3K9 acetylation (H3K9ac) and H3K18/27 acetylation, respectively, in cells.  CBP/p300 and their HAT activities are essential, while GCN5/PCAF and associated H3K9ac are dispensable, for nuclear receptor target gene expression (EMBO J 2011).  GCN5/PCAF-mediated H3K9ac correlates well with, but is surprisingly dispensable for, the expression of endogenous interferon-β and the vast majority of active genes in fibroblasts.  Instead, GCN5/PCAF repress interferon-β production and innate antiviral immunity in cells in a HAT-independent and non-transcriptional manner (EMBO Rep 2014​).  We are interested in epigenomic regulation of PPARγ and GR target gene expression. We showed that ligand-activated GR accelerates expression of early adipogenic genes in cells by recruiting CBP/p300 to activate C/EBPb-primed enhancers (MCB 2017a).  We are investigating the role of H3K36 methylation in regulating PPARγ target gene expression and adipogenesis.

Applying our Research

Understanding how epigenomic mechanisms regulate PPARγ and adipogenesis may provide new ways to treat obesity and type 2 diabetes.  Synthetic PPARγ ligands have been used to treat millions of patients with type 2 diabetes.  Investigating how epigenomic mechanisms regulate ligand-induced nuclear receptor target gene expression will help us better understand how synthetic PPARγ ligands act as antidiabetic agents. ​​