MLL4 and MLL3 complexes regulate expression of PPARG target genes in adipogenesis and hepatic steatosis

PPARG isoform PPARG2-positive adipocyte nuclei isolated from visceral adipose tissue show significantly higher expression level of KMT2C, the catalytic subunit of the MLL3 complex, and PAXIP1, a cofactor of MLL3 and KMT2D (MLL4) complexes, than PPARG2-negative nuclei. Based on mouse studies, PAXIP1 (PTIP), an accessory subunit of MLL3 and MLL4 complexes, is required for adipogenesis in mouse embryonic fibroblasts (MEFs) and primary preadipocytes. PAXIP1-deficient MEFs show significant defects in both PPARG- and CEBPA-stimulated adipogenesis. Knockout of Paxip1 gene in brown adipose tissue (BAT) leads to significant decrease of BAT in knockout mice, and a significant decrease of expression of markers shared between white adipose tissue and BAT, such as Pparg, Cebpa, and Fabp4, as well as BAT-specific/prevalent markers Prdm16, Cidea, Mpzl2 (Eva1), Ntrk3, Ucp1, Ppargc1a (Pgc1a), Cox5b and Cox8b. Paxip1 BAT knockout mice are cold intolerant, with impaired cold-mediated induction of genes involved in fatty acid catabolism, such as Cpt1a, Lpl, and Mlycd (Mcd).

In prostate cancer, KMT2D and PPARG are overexpressed at the protein level relative to the normal tissue. KMT2D knockdown significantly reduces the lipid droplet content in prostate cancer cell lines. In prostate cancer tumors, KMT2D mRNA expression significantly correlates with mRNA expression of lipid metabolism genes FASN, ACC, SCD, and ACLY. KMT2D knockdown in prostate cancer cell lines leads to significant decrease in the mRNA levels of ACC, ACLY, and FASN. Stimulation of PPARG by the synthetic agonist rosiglitazone stimulates lipid synthesis in prostate cancer cell lines, but the effect of rosiglitazone is diminished upon KMT2D knockdown.

In addition to regulating genes involved in lipid metabolism, the PPARG:RXRA complex and MLL3/MLL4 complexes (Jang et al. 2019: supplementary information) may also regulate expression of some of the genes involved in glucose metabolism and the tricarboxylic acid (TCA) cycle.

Hepatic steatosis represents the synthesis and accumulation of triglycerides in hepatocytes which can, if prolonged, lead to the development of non-alcoholic fatty liver disease (NAFLD) that can progress to non-alcoholic steatohepatitis (NASH), ultimately resulting in liver cirrhosis. Like Kmt2c (Mll3) delta/delta mice, which express catalytically inactive Kmt2c (Lee, Saha et al. 2008; Lee S., Lee J. et al. 2008), Kmt2d (Mll4)+/- mice, with deletion of one allele of Kmt2d, are resistant to high fat diet-induced hepatic steatosis, with Kmt2d+/- livers accumulating much less fat relative to wild type littermate controls in response to high fat diet feeding. Bulk transcriptomic analysis of Kmt2d+/- mouse livers shows that the expression of a large portion of high fat diet controlled genes requires Kmt2d. Among the defined hepatic steatotic transcription factors, which include MLXIPL (ChREBP), SREBF1 (SREBP1) isoform SREBP1c (SREBP 1C), the liver X receptors (LXRs) – NR1H3 (LXRA) and NR1H2 (LXRB), and PPARG, KMT2D has been reported to associate with LXRs (Lee S., Lee J. et al. 2008) and PPARG (Lee, Saha et al. 2008). No association between mouse orthologs of KMT2D and MLXIPL or SREBP1c could be detected.