A SUMOylation-dependent pathway mediates transrepression of inflammatory response genes by PPAR-gamma

Abstract

Peroxisome proliferator-activated receptor-gamma (PPAR-gamma) has essential roles in adipogenesis and glucose homeostasis, and is a molecular target of insulin-sensitizing drugs. Although the ability of PPAR-gamma agonists to antagonize inflammatory responses by transrepression of nuclear factor kappa B (NF-kappaB) target genes is linked to antidiabetic and antiatherogenic actions, the mechanisms remain poorly understood. Here we report the identification of a molecular pathway by which PPAR-gamma represses the transcriptional activation of inflammatory response genes in mouse macrophages. The initial step of this pathway involves ligand-dependent SUMOylation of the PPAR-gamma ligand-binding domain, which targets PPAR-gamma to nuclear receptor corepressor (NCoR)-histone deacetylase-3 (HDAC3) complexes on inflammatory gene promoters. This in turn prevents recruitment of the ubiquitylation/19S proteosome machinery that normally mediates the signal-dependent removal of corepressor complexes required for gene activation. As a result, NCoR complexes are not cleared from the promoter and target genes are maintained in a repressed state. This mechanism provides an explanation for how an agonist-bound nuclear receptor can be converted from an activator of transcription to a promoter-specific repressor of NF-kappaB target genes that regulate immunity and homeostasis.

Figure 1

PPARγ prevents LPS-induced dissociation of the NCoR/HDAC3 complex from the iNOS promoter. a, Northern blot analysis indicating that siRNAs directed against NCoR abolish rosiglitazone (Ro) and GW0072-dependent repression of LPS-induced iNOS expression. b, Ro and GW0073 inhibit release of NCoR from the iNOS promoter as demonstrated by ChIP assay. c, siRNAs directed against TBL1, TBLR1 or Ubc5c prevent LPS induction of the iNOS promoter in transiently transfected RAW264.7 macrophages. Error bars in this and succeeding figures represent standard deviations of triplicate, independent determinations. *indicates p<0.001 compared to Ro repression; ** refers to p<0.01 compared to LPS induction. d, Rosiglitazone prevents LPS-dependent recruitment of Ubc5 to the iNOS promoter as detected by ChIP assay. e, PPARγ binds to the CD36 and iNOS promoters in a ligand-dependent manner as detected by ChIP assay. f, Knockdown of NCoR expression prevents PPARγ recruitment to the iNOS promoter as detected by ChIP assay.

Figure 2

PIAS1 interacts with PPARγ and is required for transrepression of iNOS. a, Immunoblots for FLAG-tagged PIAS1 and HA-tagged PPARγ in transfected RAW264.7 macrophages (left panel) or endogenous PIAS1 and PPARγ in primary macrophages (right panel) indicate PIAS1 interaction with immunoprecipitated PPARγ. b, siRNAs directed against NCoR and PIAS1 in RAW264.7 macrophages reverse PPARγ transrepression of the iNOS promoter. c, Primary macrophages transfected with control siRNA, PIAS1 or Ubc9-specific siRNA show reversal of PPARγ transrepression by semi-quantitative PCR for endogenous GAPDH and iNOS expression. d, PPARγ recruitment to the iNOS but not the CD36 promoter in primary macrophages requires PIAS1 as detected by ChIP assays. e, PPARγ recruitment to the iNOS promoter in primary macrophages requires PIAS1 and Ubc9 as detected by ChIP assays.

Figure 3

Ligand-dependent sumoylation of PPARγ is required for transrepression. a, Upper right panel shows a schematic representation of PPARγ with two consensus sumoylation sites at K77 and K365. Upper left panel illustrates the PPARγ ligand binding domain, indicating the location of K365 in helix 7. The lower panels illustrate a close-up of this region of the apo PPARγ ligand-binding domain (LBD) surface (left), with the arrow pointing to a hydrogen atom in K365 side chain and liganded PPARγ LBD (right) with the arrow pointing to solvent-exposure of the nitrogen atom of the primary amine of K365 (shown in blue). b, Sumoylation of PPARγ at K365 occurs in a ligand-dependent manner. NT refers to non-transfected cells. c, d, Sumoylation of PPARγ at K365 but not K77 is required for transrepression of the iNOS promoter but not transactivation of the AoxTK promoter in transfected RAW264.7 macrophages. e, f, Sumoylation of PPARγ at K365 is required for recruitment to the iNOS promoter but not the CD36 promoter in RAW264.7 cells as demonstrated by ChIP assays.

Figure 4

Sumoylation of PPARγ promotes interaction with the NCoR/HDAC3 complex. a, Mammalian two hybrid assay in RAW264.7 cells indicating that VP-16 PPARγWT interacts with GalDBD-NCoR ΔIDC (NCoR a.a. 1-2277 without IDC) but not GalDBD-NCoR IDC (C-terminal nuclear receptor interaction motif) in a ligand-dependent and Ubc9-dependent manner. b, siRNAs directed against HDAC3 reduce but do not abolish recruitment of PPARγ to the iNOS promoter in primary macrophages as demonstrated by ChIP assays. c, Model for mechanisms of LPS activation and PPARγ-dependent repression of the iNOS gene. See text for details.