PPARγ binds SETD6 promoter and activates its expression.

(A) Top: Sequence logo of the PPARγ response element (PPRE) from JASPAR. Bottom: Schematic representation of PPARγ binding to a predicted PPRE site within the SETD6 promoter (B) Capture of a genome browser showing the enrichment of PPARγ at the SETD6 promoter in HepG2 cells with open chromatin state represented by H3K4me3, H3K27ac, and ATAC-seq tracks. (C) ChIP assay with Flag-PPARγ antibody or beads as negative control in HepG2 cells followed by qPCR with primers flanking the predicted binding site at the SETD6 promoter. Graphs show % input of the quantified DNA. (D) RNA was extracted from HepG2 cells transfected with control or Flag-PPARγ WT. Transcript levels of SETD6 were determined by qPCR. Error bars are SEM. Statistical analysis was performed for three experimental repeats using one-way ANOVA (*p < 0.05, ****p < 0.0001). (E)- dual-luciferase assay in HepG2 cell transfect with an increasing amount of Flag-PPARγ. About 24 h post-transfection, the whole cell lysates were subjected to dual-luciferase assay (Promega). Relative luminescence was calculated after normalization of the firefly luciferase signal over Renilla luciferase control. Error bars are SD. Statistical analysis was performed for three experimental repeats. *p ≤ 0.03

Physical interaction between PPARγ and SETD6 in-vitro and in cells

(A) ELISA-based analysis of the interaction between recombinant GST-SETD6 and the indicated recombinant proteins. ****p < 0.0001. (B) Endogenous PPARγ was immunoprecipitated from RIPA lysates of HepG2 cells followed by western blot analysis using the indicated antibodies. (C) Endogenous SETD6 was immunoprecipitated using chromatin fraction isolated from HepG2 followed by western blot analysis using the indicated antibodies. (C) Endogenous SETD6 was immunoprecipitated from chromatin isolated from HepG2 cells followed by western blot with the indicated antibodies. (D) Left, representative images of proximity ligation assays (PLA) between Flag-PPARγ and GFP-SETD6 WT or the catalytic inactive mutant GFP-SETD6 Y285A in HeLa cells. The negative control (NC) was performed in the absence of one primary antibody. Red dots indicate positive PLA signals. Scale bar = 10 μm. Right – Quantification of PLA signals for each sample. Statistical analysis was performed using Student’s t-test (***p < 0.001).

SETD6 methylates PPARγ at K170 in-vitro and in cells.

(A) In vitro methylation assay in the presence of 3H-labeled SAM and the indicated purified proteins. Coomassie stain of the recombinant proteins used in the reactions is shown at the bottom. Schematic representation of PPARγ domain structure. The methylated residue (K170) identified by mass spectrometry is shown in red. DBD- DNA binding domain; LBD- ligand binding domain. (B) Endogenous PPARγ was immunoprecipitated from HepG2 cells followed by WB with the indicated antibodies (C) Flag-PPARγ was over-expressed followed by immunoprecipitation using pan-methyl antibody in control (CT) and SETD6 KO HeLa cells followed by western blot with indicated antibodies. (D) 0.25 ug of PPARγ peptides (un-modified and K170me1) were spotted on a nitrocellulose membrane followed by incubation with anti- PPARγ K170me1 antibody or Streptavidin-HRP. (E) Endogenous PPARγ was immunoprecipitated from control and KO SETD6 HepG2 cells followed by western blot with the indicated antibodies.

PPARγ binds and activates SETD6 expression in a K170 methylation dependent manner

(A) - dual-luciferase assay in Control and SETD6 KO HepG2 cells. 24 h post-transfection, the whole cell lysates were subjected to dual-luciferase assay (Promega). Relative luminescence was calculated after normalization of the firefly luciferase signal over Renilla luciferase control. Error bars are SD. Statistical analysis was performed for three experimental repeats. ***p < 0.001. (B) Left- WB analysis with the indicated antibodies for stable cells expressing Flag-WT or Flag-K170R PPARγ. Right- Transcript levels of the SETD6 were determined by qPCR of stably expressing HepG2 cells- Empty, Flag-PPARγ WT, and Flag-PPARγ K170R mutant. mRNA levels were normalized to GAPDH and then to Empty. (C) Chromatin immunoprecipitation (ChIP) assay. The chromatin fraction of HepG2 SETD6 CRISPR CT, KO1, and KO2 cells were immunoprecipitated with magnetic beads conjugated with anti-PPARγ antibody. The bound DNA was purified and amplified by qPCR. (D) Same as C for cells stably expression of Empty, Flag-PPARγ WT, and Flag-PPARγ K170R mutant. Graphs for C and D show the percent input of the quantified DNA. Two-way ANOVA analysis was performed; error bars are S.E.M. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

SETD6 positively regulates lipid droplets formation:

(A) WB analysis with the indicated antibodies for HepG2 control (2 clones) and SETD6 KO cells (3 clones) (B) Heatmap showing upregulated and downregulated genes identified by RNA-sequencing analysis of two SETD6 control (CT) and three SETD6 KO HepG2 independent clones. Gene expression values are displayed as normalized Z-scores. Yellow and blue colors represent relatively high and low expression levels, respectively. (C+D) KEGG pathway enrichment analysis of differentially expressed genes identified by RNA-sequencing analysis of SETD6 control (CT) and SETD6 KO HepG2 cells. (C) pathways enriched among genes downregulated in SETD6 KO cells. (D) pathways enriched among genes upregulated in SETD6 KO cells. Circle size represents the number of differentially expressed genes associated with each pathway, and color intensity indicates statistical significance expressed as −log10(FDR). The x-axis represents the enrichment score. (E) Illustration of the system to monitor lipid droplets accumulation over time: HepG2 cells are treated with oleic acid (OA) and stained with Hoechst (nucleus) and BODIPY(neutral lipids), followed by live cell imaging. (F) Top-Representative images of HepG2 CRISPR CT (control), KO1, and KO2 challenged with 300 mM of OA (20 H time point) and stained with Hoechst (nucleus) and BODIPY (neutral lipids). Bottom- Mean green fluorescence was calculated as fluorescence signal divided by cell count. Data is analyzed from three beacons per well in three wells. Statistical analysis was performed using one-way ANOVA. ns: non-significant; *p < 0.05; **p < 0.01. (G) WB analysis for CT and SETD6 KO with or without over-expression of HA-SETD6. (H) Top- Representative images of HepG2 CRISPR CT (control) and SETD6 KO without or with over-expression of HA-SETD6, challenged with 300 mM of OA (20 H time point) and stained with Hoechst (nucleus) and BODIPY (neutral lipids). Bottom- Mean green fluorescence was calculated as fluorescence signal divided by cell count. Data is analyzed from three beacons per well in three wells. Statistical analysis was performed using one-way ANOVA. ns: non-significant; ***p < 0.001; ****p < 0.0001.

PPARγ methylation at K170 positively regulates gene expression and lipid droplets formation.

(A) Heatmap showing upregulated and downregulated genes identified by RNA-sequencing analysis of HepG2 independent clones stably expressing PPARγ WT or the PPARγ K170R mutant. Gene expression values are displayed as normalized Z-scores. Yellow and blue colors represent relatively high and low expression levels, respectively. (B+C) KEGG pathway enrichment analysis of differentially expressed genes identified by RNA-sequencing analysis of HepG2 cells stably expressing PPARγ WT or the PPARγ K170R mutant. (B) pathways enriched among genes downregulated in the PPARγ K170R mutant cells. (C) pathways enriched among genes upregulated in the PPARγ K170R mutant cells. Circle size represents the number of differentially expressed genes associated with each pathway, and color intensity indicates statistical significance expressed as −log10(FDR). The x-axis represents the enrichment score. (D) ChIP analysis for HepG2 control and SETD6 KO cells (two clones) that were immunoprecipitated with PPARγ antibody. The bound DNA was purified and amplified by qPCR using specific primers to MOGAT1 and PLIN2 gene promoter regions. Graphs show the percent input of the quantified DNA. Two-way ANOVA analysis was performed; error bars are S.E.M. ****p < 0.0001. (E) Same as for panel D with HepG2 cells stably expressing Empty, Flag-PPARγ2 WT, and Flag-PPARγ2 K170R mutant that were immunoprecipitated with FLAG conjugated magnetic beads. error bars are S.E.M. ns: non-significant; *p < 0.05; **p < 0.01; ****p < 0.0001. (F) Structural model in cartoon representation of the PPARγ–RXRα heterodimer bound to DNA based on the published co-crystal structure (PDB: 3DZY). PPARγ is shown in blue, RXRα in cyan, DNA in orange, and NCOA2 peptides in gray. The boxed region indicates K170 within the PPARγ DNA-binding domain. Enlarged views in stick representation of the K170 side chain (top), the modeled K170me1 side chain (middle), and multiple modeled K170me1 rotamers (bottom) are shown. (G) Top- Representative images of HepG2 cells stably expressing Empty, PPARγ WT or PPARγ K170R mutant, challenged with 300 mM of OA (20 H time point) and stained with Hoechst (nucleus) and BODIPY (neutral lipids). Bottom- Mean green fluorescence was calculated as the fluorescence signal divided by cell count. Data is analyzed from three beacons per well in three wells. Statistical analysis was performed using one-way ANOVA. ****p < 0.0001. (H) A Schematic representation of our proposed working model.

Primers for cloning and mutagenesis

Primers for qPCR

Primers for ChIP – qPCR

Interaction between overexpressed SETD6 and PPARγ in chromatin fractions.

HEK293T cells were transfected with the indicated plasmids followed by immunoprecipitation with Flag antibody of the chromatin fraction. Samples were then subjected to WB analysis using the indicated antibodies.

Structural modeling of the SETD6–PPARγ complex.

(A) Structural model in cartoon representation of the predicted SETD6–PPARγ complex generated using AlphaFold. PPARγ is shown in blue, SETD6 is shown in green, and the SET domain is highlighted in dark green. Yellow boxes indicate the predicted interaction interface between the SET domain of SETD6 and the DNA-binding domain of PPARγ, shown at higher magnification in (B). (B) Enlarged view of the predicted interaction interface comparing wild-type SETD6 (top) and the catalytically inactive mutant SETD6 Y285A (bottom). The catalytic residue Y285 (or A285 in the mutant) and PPARγ K170 are indicated in stick representation.

MS/MS spectra showing monomethylation of recombinant PPARγ at lysine-170 (LKme1LIYDR, m/zobserved =467.789 (z=+2)) after in vitro methylation by SETD6.

The MS spectra was visualized with PDV software (v.2.0.0; (60)). The y-and b-ions are annotated and displayed as red, and blue, respectively.

Top: OA accumulation curve of HepG2 cells over 20 hours challenged with 300 µM, 600 µM, and 900 µM of OA. 900 µM DMSO served as a control treatment. Mean green fluorescence was calculated as fluorescence signal divided by cell count. Data is analyzed from five beacons per well, with three wells per OA or DMSO treatment. Bottom- Representative images of last time point (20 H) with three concentrations of OA. For each image, a magnified area of interest is shown (black boxes).

Transcript levels of the indicated genes were determined by qPCR of stably expressing HepG2 cells- Empty, Flag-PPARγ WT, and Flag-PPARγ K170R mutant.

mRNA levels were normalized to GAPDH and then to Empty. error bars are S.E.M. **p < 0.01; ****p < 0.0001.

Structural modeling of T166 phosphorylation within the PPARγ DNA-binding domain.

Structural model in cartoon representation of the PPARγ–RXRα heterodimer bound to DNA based on the published co-crystal structure (PDB: 3DZY). PPARγ is shown in blue, RXRα in cyan, DNA in orange, and NCOA2 peptides in gray. The boxed region highlights threonine 166 (T166) within the PPARγ DNA-binding domain. Enlarged views of the unmodified T166 residue (top) and the modeled phosphorylated T166 (pT166; bottom) are shown in stick representation. Predicted intramolecular steric clashes introduced by the phosphate group are indicated by green dashed lines.