Deletion of PPARα in mouse brown adipocytes increases their De Novo Lipogenesis

Abstract

Objective: Peroxisome Proliferator-Activated Receptors (PPARs) are nuclear receptors involved in the control of lipid metabolism. The PPARα isoform is highly expressed in brown adipose tissue (BAT). However, its precise role in BAT remains unclear. Here, we aimed to investigate the role of PPARα in BAT of high fat diet-induced obese mice in a thermoneutral environment.

Methods: We used tamoxifen-inducible-BAT specific PPARα knockout mice (PPARαBATKO) that were housed at thermoneutrality to minimize BAT basal activation, fed a high-fat diet for 20 weeks and challenged with a β₃-adrenergic agonist (CL316,243) during the last week. Both male and female mice were studied.

Results: Body weight and glucose tolerance tests were similar in both sexes and genotypes. However, BAT morphology was altered in PPARαBATKO mice, with more unilocular and larger lipid droplets compared to control mice, suggesting BAT impaired function. Indeed, when treated with CL316,243, both male and female mice had increased De Novo Lipogenesis (DNL), reflected by an increased expression of ChREBPβ and lipogenic enzymes ACLY, ACC1, FASN and SCD1. These changes were accompanied by an increase in fatty acids in triglycerides, and thus an increase in lipid storage. Moreover, lipid profiles in phospholipids were different, suggesting a modification in the membrane content with an increase of palmitoleate.

Conclusions: Altogether, our results reveal a key role for PPARα in DNL in BAT and in the regulation of lipid metabolism in HFD-induced obesity.

Keywords: Brown adipose tissue; High fat diet-induced obesity; Inducible UCP1-CRE; Lipid metabolism; Peroxisome proliferator-activated receptor alpha; β(3)-adrenergic stimulation.

Graphical abstract

Figure 1

Generation of PPARαBATKO mice showing deletion of PPARα specifically in brown adipose tissue. (A) Schematic representation for the generation of PPARαBATKO. (B) Experimental design. Control (UCP1-WT-PPARalphalox/lox) and PPARαBATKO (UCP1-CRE^ERT2-PPARalphalox/lox) were housed at thermoneutrality and fed a high fat diet (HFD) for 20 weeks and daily injected with 1 mg/kg of CL316,243 or NaCl the last week of the experiment. To induce CRE^ERT2 activity, mice were injected intraperitoneally with tamoxifen at a dosage of 60 mg/kg the first three days of the HFD, then every three weeks. (C) Relative mRNA levels of Pparα in brown adipose tissue (BAT) of female and male adult mice (n = 5–6 per group). (D) Relative mRNA levels of Pparα in subcutaneous (scWAT) of female and male mice (n = 5–6 per group). (E) Relative mRNA levels of Pparα in liver of female and male mice (n = 3–6 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test (C, E for female) or Mann–Whitney test (D, E for male); ∗p < 0.05, vs Ctrl.

Figure 2

BAT specific deletion of PPARα does not alter glucose tolerance or body weight gain in HFD-induced obese mice. (A, H) Weight curves of control (Ctrl) and PPARαBATKO female (A) and male (H) mice (n = 12 per group) on high fat diet (HFD) during 20 weeks. To induce CRE^ERT2 activity, mice were injected intraperitoneally with tamoxifen (TAM) at a dosage of 60 mg/kg the first three days of the HFD, then every three weeks. (B, I) Glucose tolerance test (GTT) after 13 weeks of HFD in female (B) and male (I) mice, and corresponding area under curves (AUC) (n = 8–12 per group). (C, J) Total Body weight (TBW) in female (E) and male (G) mice (n = 4–6 per group). (D, K) Interscapular brown adipose tissue (BAT) and subcutaneous white adipose tissue (scWAT) weights in female (D) and male (K) mice (n = 4 = 6 per group). (E, L) AT volume estimation assessed by tomography in female (E) and male (L) adult mice (n = 3–6 per group). (F, M) Relative mRNA levels of Adiponectin and Leptin in BAT of female (F) and male (M) adult mice (n = 4–6 per group). (G, N) Plasma levels of Adiponectin and Leptin in female (G) and male (N) mice (n = 4–6 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test, Mann–Whitney test (E for TBW, J for Adiponectin) or student t-test for AUC. ∗p < 0.05, vs Ctrl. £ p < 0.05, vs NaCl. NaCl: vehicle; CL: CL316,243 (β3-adrenergic agonist).

Figure 3

BAT specific deletion of PPARα alters BAT morphology and slightly changes the β3-adrenergic agonist-induced BAT activation in HFD-induced obese female mice. (A) Representative H&E staining of brown adipose tissue (BAT) in female mice fed a HFD treated with vehicle (NaCl) or β3-adrenergic agonist (scale = 40 μm). (B) Lipid droplet (LD) diameter distribution in female mice treated with vehicle (NaCl) or β3-adrenergic agonist (n = 4 per group). (C) LD average diameter (μm) in female mice treated with vehicle (NaCl) or β3-adrenergic agonist (n = 4 per group). (D) Class distribution of LD diameters (in μm) to the total number of lipid droplets in female mice fed a HFD and treated with NaCl (left panel) or β3-adrenergic agonist (right panel) (n = 4 per group). (E) Representative Picro-Sirius red staining of BAT in female mice (scale = 40 μm) and corresponding quantification. (F) Relative mRNA levels of Col1a1 and Col6a1 in BAT of female mice (n = 5–6 per group). (G) Relative mRNA levels of Ucp1 in (BAT) of female mice (n = 4–6 per group). (H) Western blot of UCP1 protein content in BAT of female mice and corresponding quantification (n = 3 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using Mann–Whitney test (F) or 2-way ANOVA with Šídák’s post hoc test (E, G and H) or Fisher’s LSD test (C and D). ∗p < 0.05, vs Ctrl. £p < 0.05, vs NaCl. NaCl: vehicle; CL: CL316,243 (β3-adrenergic agonist). Ctrl: control mice; KO: PPARαBATKO mice.

Figure 4

BAT specific deletion of PPARα increases BAT De Novo Lipogenesis for fat storage in HFD-induced obese mice upon β3-adrenergic stimulation. (A, G) Relative mRNA levels of De Novo Lipogenesis (DNL) genes in brown adipose tissue (BAT) of female (A) and male (G) mice (n = 4–6 per group). (B) Western blots of DNL proteins in BAT of female mice and corresponding quantifications (C) (n = 3 per group). (D, H) Relative abundancy of palmitate and lipogenic index in BAT triglycerides (TG) of female (D) and male (H) mice (n = 4 per group). (E) Western blot of PKM2 protein in BAT of female mice and corresponding quantification (n = 3 per group). (F) Western blot of AKT2 protein in BAT of female mice and corresponding quantification (n = 3 per group). (I) Relative fatty acid content in BAT triglycerides compared to internal standard C19 of male mice (n = 4 per group). (J) Relative mRNA levels of Pkm2 in BAT of male mice (n = 4–6 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test or Mann–Whitney test (A for Chrebpb, Acly, Fasn, and G, Chrebpb, Acly, Acaca, Fasn). ∗p < 0.05, vs Ctrl. £p < 0.05, vs NaCl. NaCl: vehicle; CL: CL316,243 (β3-adrenergic agonist). Ctrl: control mice; KO: PPARαBATKO mice.

Figure 5

BAT specific deletion of PPARα increases FA desaturation in BAT in HFD-induced obese mice upon β3-adrenergic stimulation. (A, G) Relative mRNA levels of Scd1 and Elovl6 in BAT of female (A) and male (G) mice (n = 4–6 per group). (B) SCD1 protein content in BAT of female mice and corresponding quantification (n = 3 per group). (C, H) Lipogenic indexes in BAT triglycerides (TG) of female (C) and male (H) mice (n = 4 per group). (D, I) SCD1 and ELOVL6 ratios in BAT TG of female (D) and male (I) mice (n = 4 per group). (E, J) Lipogenic indexes in BAT phospholipids (PL) of female (E) and male (J) mice (n = 4 per group). (F, K) SCD1 and ELOVL6 ratios in BAT PL of female (F) and male (K) mice (n = 4 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test or Mann–Whitney test (G for Scd1). ∗p < 0.05, vs Ctrl. £p < 0.05, vs NaCl. Ctrl: control mice; KO: PPARαBATKO mice.

Figure 6

BAT specific deletion of PPARα induces glucose intolerance in mice fed a chow diet. (A) Experimental design. Control (UCP1-WT-PPARalphalox/lox) and PPARαBATKO (UCP1-CRE^ERT2-PPARalphalox/lox) were housed at thermoneutrality and fed a standard chow diet (CD) for 20 weeks and daily injected with 1 mg/kg of CL316,243 or NaCl the last week of the experiment. To induce CRE^ERT2 activity, mice were injected intraperitoneally with tamoxifen (TAM) at a dosage of 60 mg/kg the first three days of the CD, then every three weeks. (B, K) Relative mRNA levels of Pparα in brown adipose tissue (BAT) of female (B) and male (K) adult mice (n = 6 per group). (C, L) Relative mRNA levels of Pparα subcutaneous (scWAT) of female (C) and male (L) adult mice (n = 6 per group). (D, M) Weight curves of control (Ctrl) and PPARαBATKO female (D) and male (M) mice (n = 11–12 per group) on standard chow diet (CD) during 20 weeks. (F, O) Glucose tolerance test (GTT) after 13 weeks of CD in female (F) and male (O) mice, and corresponding area under curves (AUC) (n = 12 for (F), n = 9 for (O)). (G, P) Plasma levels of Adiponectin and Leptin in female (G) and male (P) mice (n = 4–6 per group). (H, Q) AT volume estimation assessed by tomography in female (H) and male (Q) adult mice (n = 3–5 per group). (I, R) Subcutaneous white adipose tissue (scWAT) weights in female (I) and male (R) mice (n = 5–6 per group). (J, S) Intrascapular brown adipose tissue (BAT) weights in female (J) and male (S) mice (n = 5–6 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test or Mann–Whitney test (B and K). ∗p < 0.05, vs Ctrl. £p < 0.05, vs NaCl. Ctrl: control mice; KO: PPARαBATKO mice.

Figure 7

BAT specific deletion of PPARα does not modulate BAT De Novo Lipogenesis in mice fed a chow diet but increases fat storage. (A, C) Relative mRNA levels of De Novo Lipogenesis (DNL) genes in brown adipose tissue (BAT) of female (A) and male (C) mice (n = 6 per group). (B, D) Representative H&E staining of BAT in female (B) and male (D) mice fed a HFD (scale = 50 μm). (E) Lipogenic index in in BAT triglycerides (TG) and phospholipids (PL) of male mice (n = 4 per group). (F) Relative fatty acid content in BAT triglycerides compared to internal standard C19 of male mice (n = 4 per group). (G) Relative triglycerides to phospholipids fatty acid content in BAT of male mice (n = 4 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test or Mann–Whitney test (A for Elovl6, C for Chrebpb, Acly, Acaca, Fasn and Scd1). ∗p < 0.05, vs Ctrl. £p < 0.05, vs NaCl. Ctrl: control mice; KO: PPARαBATKO mice.

figs1

Supplementary F1, related to Figures 1, 4 and 7. (A) Relative mRNA levels of Pparγ and Pparδ in brown adipose tissue (BAT) of female adult mice fed a standard chow diet (CD) or high fat diet (HFD) (n = 5–6 per group). (B) Plasma level of FGF21 in female mice fed a HFD (n = 4 per group). (C) Relative mRNA levels of Pparγ and Pparδ in BAT of male adult mice (n = 5–6 per group). (D) Relative mRNA levels of Chrebpa and Srebp1 in BAT of female adult mice (n = 5–6 per group). (E) Relative mRNA levels of Chrebpa and Srebp1 in BAT of male adult mice (n = 5–6 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test. NaCl: vehicle; CL: CL316,243 (β3-adrenergic agonist). Ctrl: control mice; KO: PPARαBATKO mice.

figs2

Supplementary F2, related to Figure 3. (A) Relative mRNA levels of Ucp1 in brown adipose tissue (BAT) of female mice fed a standard chow diet (CD) or a high fat diet (HFD) (n = 4–6 per group); UCP1 protein content in BAT of female mice fed a standard chow diet. (B) Relative mRNA levels of Cpt1m, Ppargc1α, Atgl and Lpl in BAT of female adult mice fed a standard chow diet or a high fat diet (n = 5–6 per group). (C) Western bots of DRP1 and GK1 protein in BAT of female mice fed a high fat diet (n = 3 per group). (D) Relative mitochondrial DNA content in BAT of female mice fed a high fat diet (n = 4–6 per group). (E) OXPHOS protein content in BAT of female mice fed a high fat diet (n = 3 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test. ∗p < 0.05, vs Ctrl. £ p < 0.05, vs NaCl. NaCl: vehicle; CL: CL316,243 (β3-adrenergic agonist). Ctrl: control mice; KO: PPARαBATKO mice, CD: chow diet.

figs3

Supplementary F3, related to Figure 3. (A) Relative mRNA levels of Ucp1, Cpt1m, Atgl, Lpl and Ppargc1α in BAT of male adult mice fed a standard chow diet or a high fat diet (n = 5–6 per group). (B) Relative mitochondrial DNA content in BAT of female mice fed a high fat diet (n = 3 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test. ∗p < 0.05, vs Ctrl. £ p < 0.05, vs NaCl. NaCl: vehicle; CL: CL316,243 (β3-adrenergic agonist). Ctrl: control mice; KO: PPARαBATKO mice, CD: chow diet.

figs4

Supplementary F4, related to Figures 4 and 5. (A, B) Triglyceride fatty acid profile in BAT of female (A) and male (B) mice fed a HFD and challenged with CL316,243 (CL) or vehicle (NaCl) (n = 4 per group). (C, D) Phospholipid fatty acid profile in BAT of female (C) and male (D) mice fed a HFD and challenged with CL316,243 (CL) or vehicle (NaCl) (n = 4 per group). Data are displayed as mean ± SEM. Statistical analysis was performed using 2-way ANOVA with Šídák’s post hoc test. ∗p < 0.05, vs Ctrl. Ctrl: control mice; KO: PPARαBATKO mice.

figs5

Supplementary F5, related to Figure 3. (A, B) Representative H&E staining of BAT in female (A) and male (B) mice fed a HFD and housed at 22 °C (scale = 20 μm).