A PPARγ/long noncoding RNA axis regulates adipose thermoneutral remodeling in mice

Abstract

Interplay between energy-storing white adipose cells and thermogenic beige adipocytes contributes to obesity and insulin resistance. Irrespective of specialized niche, adipocytes require the activity of the nuclear receptor PPARγ for proper function. Exposure to cold or adrenergic signaling enriches thermogenic cells though multiple pathways that act synergistically with PPARγ; however, the molecular mechanisms by which PPARγ licenses white adipose tissue to preferentially adopt a thermogenic or white adipose fate in response to dietary cues or thermoneutral conditions are not fully elucidated. Here, we show that a PPARγ/long noncoding RNA (lncRNA) axis integrates canonical and noncanonical thermogenesis to restrain white adipose tissue heat dissipation during thermoneutrality and diet-induced obesity. Pharmacologic inhibition or genetic deletion of the lncRNA Lexis enhances uncoupling protein 1-dependent (UCP1-dependent) and -independent thermogenesis. Adipose-specific deletion of Lexis counteracted diet-induced obesity, improved insulin sensitivity, and enhanced energy expenditure. Single-nuclei transcriptomics revealed that Lexis regulates a distinct population of thermogenic adipocytes. We systematically map Lexis motif preferences and show that it regulates the thermogenic program through the activity of the metabolic GWAS gene and WNT modulator TCF7L2. Collectively, our studies uncover a new mode of crosstalk between PPARγ and WNT that preserves white adipose tissue plasticity.

Keywords: Adipose tissue; Cardiology; Metabolism; Molecular genetics; Noncoding RNAs.

Figure 1. Lexis is regulated by PPARγ in adipose depots.

(A) Fold change of lncRNAs in murine adipocytes after differentiation day 5 (Diff) compared with day 0 (non-diff) (from GEO GSE94654). (B) Fold change of lncRNAs in differentiated adipocytes (Diff) compared with nondifferentiated human mesenchymal stem cells (hMSCs) (from GEO GSE151324). (C) Lexis expression (human orthologue is putative) in differentiated versus nondifferentiated based on RNA-Seq data in A and B. (D) qPCR analysis in human adipose-derived mesenchymal stem cells (ADMSCs) (day 0) or differentiated adipocytes induced at day 8 or day 12 (n = 3 per group). (E) qPCR analysis of in C3H10T1/2 and 3T3L1 cells treated with differentiation cocktail (n = 3 per group). (F) qPCR analysis of C3H10T1/2 and 3T3L1 treated with PPARγ agonist GW1929 (20 nM) (n = 3 per group). (G) Single-molecule RNA-FISH targeting Lexis in C3H10T1/2 cells treated with vehicle or PPARγ agonist GW1929 (20 nM). Nuclear DNA was labeled with DAPI. Scale bars: 10 μm. (H) ChIP-Seq peaks of PPARγ iWAT and eWAT from GEO GSM2433426 (iWAT, low fat), GSM2433425 (iWAT, HFD), GSM2433449 (eWAT, low fat), and GSM2433453 (eWAT, HFD). (I) qPCR in iWAT from 10-week-old male mice placed on WD or HFD for 2 weeks (n = 5 per group). (J) qRT-PCR in iWAT from 8- to 10-week-old male mice under different thermal conditions, 7 days thermoneutrality (TN), 4 days cold exposure (CE), 7 days CE or room temperature (RT). n = 7 (TN, RT, and 7 days CE); n = 8 (4 days CE). Data are represented as mean ± SD (D, E, and F) and mean ± SEM (I and J). P values were calculated either by 1-way ANOVA (D, I, and J) or 2-way ANOVA (E and F). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 2. Loss of Lexis leads to lean phenotype.

(A) Body mass after 3 weeks of WD feeding of Lexis WT or Lexis-KO mice (WT, n = 9; KO, n = 8). (B) Gross appearance and iWAT depot from WT or Lexis-KO mice. (C) Body fat composition determined by EchoMRI (WT, n = 9; KO, n = 8). (D) H&E staining of iWAT and eWAT from WT mice or Lexis-KO mice. Scale bars: 100 μm. (E) Glucose tolerance test (GTT) performed on male mice (n = 6 per group). (F) Energy expenditure from WT or Lexis-KO mice measured by indirect calorimetry (P = 0.0106, n = 6 per group). (G) Body weight and percentage changes of body mass from baseline of male mice treated with ASO control (Ctrl) or ASO Lexis placed on WD (Ctrl, n = 8; ASO Lexis, n = 7). (H) Body fat composition of mice in G determined by EchoMRI (Ctrl, n = 8; ASO Lexis, n = 7). (I) Fat depot mass after WD feeding (Ctrl, n = 8; ASO Lexis, n = 7). (J) GTT performed on mice after WD feeding (n = 9 per group). (K) Energy expenditure in ASO Ctrl or ASO Lexis mice using indirect calorimetry after WD feeding (n = 9 per group, P < 0.05 using either total body mass or lean body mass as covariates). Data in A, C, E, and G–J are represented as mean ± SD. Data in F and K are represented as mean ± SEM. P values were calculated by unpaired t test (A, C, G, H, and I) or by 2-way ANOVA (E and J). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Analysis of covariance (ANCOVA) was used for F and K.

Figure 3. Adipose-specific deletion of Lexis counteracts DIO by increasing energy expenditure.

(A) Gross appearance of Lexis adipose tissue–specific KO mice (AdKO) or littermate control mice (AdWT) and their adipose depots after 12 weeks of WD. (B) Body weight of AdKO or AdWT on WD (male, n = 8). (C) Fat pad mass after 12-week WD feeding (n = 8). (D) Body fat composition of 12-week WD-fed mice determined by EchoMRI (n = 8). (E) H&E staining of iWAT from mice in B. Scale bars: 100 μm. (F) Intraperitoneal glucose tolerance test administered on WD (male, n = 8). (G) Intraperitoneal insulin tolerance test on WD (Male, n = 8). (H) Energy expenditure (EE) using indirect calorimetry in male mice after 3 weeks on a WD, showing the mean value per hour ± SEM (n = 10, P < 0.05 by ANCOVA using either total body mass or lean body mass as covariates). (I and J) Average OCR in electron flow (I) and coupling Figure 3 (J) assays of mitochondria isolated from differentiated SVF from iWAT of AdKO or AdWT mice (I, n = 5 for AdWT, n = 4 for AdKO; J, n = 5 per group). (K) Schematic of experiment and H&E staining of iWAT. Scale bars: 100 μm. (L) Core body temperature measured from K (AdWT, n = 7, 5 male, 2 female; AdKO, n = 8, 5 male, 3 female). Data are represented as mean ± SD (B–D, F, G, and L) or mean ± SEM in Figure 2 (H, I, and J). P values were calculated by unpaired t test (C, D, I, J, and L) or 2-way ANOVA (B, F, and G). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 4. Loss of Lexis from adipose enriches for thermogenic populations.

(A) t-Distributed stochastic neighbor embedding (tSNE) map of adipocyte-related subpopulation of adipose nuclei isolated from the iWAT of Lexis-AdWT mice (WT) and Lexis-AdKO mice (KO). (B) Fraction of WT and KO cells across each adipocyte-related subpopulation relative to the total number of adipocyte nuclei. (C) Annotation of adipocyte-related subpopulation derived from cluster-specific gene expression analysis (Supplemental Figure 3 G and Supplemental Table 1). (D) Heatmap showing average expression of genes as population markers identified by Schwalie et al., (43) in our identified adipocyte subpopulations. (E) tSNE plots highlighting the expression of representative thermogenic genes in adipocyte subclusters. (F) Function annotation of C6 subcluster in B. Top 10 functional terms (GO Biological Process) shown. (G) Normalized expression in different adipocyte subclusters under WT or KO group.

Figure 5. Systemic identification of Lexis “interactome” shows direct contact with Atp2a2 promoter region.

(A) Volcano plot of RNA-Seq data in the iWAT of mice fed 12-week WD (n = 3 per group). Red dots indicate the gene highly expressed in AdWT or AdKO group (Cutoff: fold change [FC] > 1.5 and P < 0.05). (B) Gene expression by qRT-PCR in the iWAT of 12-week WD-fed mice in Figure 3 (B) (n = 8 per group). (C) Gene expression by qRT-PCR in the iWAT of mice given temperature stress in Figure 3 (I) (n = 6 for Lexis-AdWT; n = 8 for Lexis-AdWT). (D) Experimental schematic Lexis chromatin-affinity assay. (E) Top enriched pathways analyzed by Metascape using the data from Lexis RNA-Seq in A. (F) Top enriched pathways analyzed by Metascape using the data from D. (G) Representative peaks of Lexis, H3K27ac ChIP-Seq, H3K4me1 ChIP-Seq, and H3K4me2 ChIP-Seq on Atp2a2 promoter region. Lexis peaks are from this experiment and other ChIP-Seq from data sets under GEO GSE56872 and GSE95533. (H) Lexis ChIRP-qPCR using primers targeting binding sites of Lexis at Atp2a2 in 10T1/2 cells treated with 24 hours of GW1929 (n = 4 per group). (I) Motif analysis based on Lexis-enriched contact sites. The values in the circles revealed the −log₁₀ (P value). (J) TCF7L2 ChIP-qPCR performed in TCF7L2 WT or TCF7L2 KO preadipocytes (left) or in iWAT of Lexis WT (Lexis AdWT) or Lexis-KO (Lexis-AdKO) mice with 1 week of thermoneutrality (n = 4 per group). Data are represented as mean ± SD (H and J) or mean ± SEM (B and C). P values were calculated by unpaired t test (B, C, and J). *P < 0.05; **P < 0.01; ****P < 0.0001.

Figure 6. TCF7L2 is required for the effects of Lexis on thermogenesis.

(A) Relative mRNA levels of Serca1, Serca2 (Serca2a and Serca2b, coded by Atp2a2), and Serca3 in iWAT of Lexis-AdWT or Lexis-AdKO mice 12-week WD-fed detected by qRT-PCR (n = 7 for Lexis-AdWT; n = 8 for Lexis-AdKO). (B) Images of fluorescent intracellular Ca²⁺ levels in SVF from Lexis-AdWT or Lexis-AdKO mice with NE stimulation. (C) Fluorescence intensity quantification of B (n = 5 per group). (D) Average OCR using iWAT from Lexis-AdWT or Lexis-AdKO mice (n = 4 per group). OCR was normalized to mitochondrial content. (E) Ucp1-KO male mice (10–12 weeks) treated with ASO Ctrl or ASO Lexis (n = 6 per group). (F) Gene expression by qRT-PCR from iWAT of mice from E. (G) Schematic for generation of Atp2a2-KO preadipocytes. Preadipocytes in each condition were induced by browning cocktail for 5 days and OCR was detected (n = 10 per group). Basal and maximal respiration shown. (H) Gene expression of Atp2a2 and Ucp1 detected by qRT-PCR from TCF7L2 WT preadipocytes treated with ASO Ctrl or ASO Lexis (n = 3 per group). (I) Gene expression of Atp2a2 and Ucp1 detected by qRT-PCR from TCF7L2-KO treated with ASO-Ctrl or ASO-Lexis (n = 3 per group). Data are represented as mean ± SEM (A, D, and F) or mean ± SD (C, E, G, H, and I). P values were calculated by either unpaired t test (A, C, D, F–I) or 2-way ANOVA (E). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.