CPEB2-activated Prdm16 translation promotes brown adipocyte function and prevents obesity

Figures

• 

  Figure 1. Obesogenic phenotype of CPEB2-knockout (KO) female mice is caused by hypometabolism. (A) Body weight of CPEB2-WT and -KO female littermates fed a 9% fat diet. (B) Photograph of WT and KO female littermate at 10 months (mo) old. (C) Homecage activity of 5-mo-old females monitored for 48 h under 12-h light/dark cycles. The time spent (min) per hour in spontaneous activities, including drinking, eating, grooming, rearing, hanging and walking, is indicated. (D) The average daily food consumed (in g) per gram body weight by 3.5-mo-old or 8.5-mo-old female mice were recorded for consecutive 10 days or 21 days, respectively (3.5-mo-old, n = 21 for both groups of mice; 8.5-mo-old, 28 WT and 26 KO mice). (E) Body weight of UCP1-WT, -heterozygous and -KO female littermates fed a 9% fat diet. (F) Body weight of WT and Ucp1ΔL female littermates fed a 9% fat diet. Data are mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, two-way ANOVA with Fisher's LSD post-hoc test. Numbers in parentheses denote the number of mice in each group.

• 

  Figure 2. Adaptive thermogenesis-evoked transcriptional upregulation is normal in CPEB2-KO mice. (A) Western blot analysis of CPEB2 level in hypothalamus (hypoTH) and BAT (n = 3 male mice). (B) RT-qPCR analysis of expression of genes associated with thermogenesis and lipid metabolism in BAT of WT and KO mice fed 9% fat diet or a high-fat diet (HFD; n = male (M)/female (F), WT: 4/10; WT_HFD: 3/3; KO: 3/10; KO_HFD: 3/3). Annotations indicate significant effects of a = diet and b = diet × genotype interaction, two-way ANOVA. (C) RT-qPCR analysis of expression of genes associated with thermogenesis in BAT of WT and KO male mice after 5-h cold exposure (CE, n = 3 per group). Annotations denote significant effects of a = temperature and b = temperature × genotype interaction, by two-way ANOVA. β3AR, β3 adrenergic receptor; Ucp1, uncoupling protein 1; Dio2, iodothyronine deiodinases II; Ampk, AMP-activated protein kinase; Pparγ, peroxisome proliferator-activated receptor γ; Srebp1, sterol regulatory element-binding transcription factor 1; Hsl, hormone-sensitive lipase. Data are mean ± SEM. (D) Body weight of cWT (Cpeb2f/f) and neuron/glia-specific cKO (Cpeb2f/f, Nes-cre/+) female mice (cWT/cKO, n…

• 

  Figure 3. Ectopic expression of CPEB2 in the KO BAT increases UCP1 synthesis and reduces weight gain. (A) Body weight of cWT and adipose-specific cKO (Cpeb2f/f, aP2-cre/+) female mice. Data are mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001, two-way ANOVA. (B) Western blot analysis of BAT UCP1 protein level in cWT and cKOaP2 mice at 10 mo old. ∗P < 0.05, Student's t test. (C) CPEB2 WT and KO mice were transduced with adeno-associated virus (AAV) expressing full-length (AAV-CP2) or RNA-binding domain (AAV-RBD) of CPEB2. Immunostaining of CPEB2 and UCP1 in KO BAT transduced with AAV-CP2 for 2 weeks, with Hoechst staining of nuclei. Scale, 10 μm. (D) Body weight gain of WT and KO female mice (20-22-week-old) after transduction with AAV-CP2 or AAV-RBD. Data are mean ± SEM. ∗P < 0.05, two-way ANOVA. with Fisher's LSD post-hoc test. KO + RBD compared with KO + CP2 (blue∗); with WT + RBD (red∗); and with WT + CP2 (green∗). KO + CP2 compared with WT + RBD (red#); with WT + CP2 (green #). ∗ #P < 0.05, ∗∗ ##P < 0.01, ∗∗∗P < 0.001, two-way ANOVA with Fisher's LSD post-hoc test. Numbers in parentheses denote the number of mice in each group. (For interpretation of the references to color in this…

• 

  Figure 4. CPEB2-KO mice show impaired BAT thermogenesis and increased body fat composition. (A) Body weight of CPEB2-cWT and -cKOaP2 male littermates fed a 9% fat diet. (B) Infrared thermography images of CPEB2 WT and KO male mice at 5 mo old (n = 6–8). The thermal images were used to calculate surface back temperature. (C) The core body temperature of mice used in (B) determined by a rectal thermometer. (D) Fat and lean mass in WT and KO mice of both sexes determined in mice at 7 mo old by MRI (female/male, n = 5–6/6-7). (E) Body weight and (F) weight gain of CPEB2-WT and -KO male littermates fed a 9% fat diet. Data are mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, two-way ANOVA for (A, E, F) and Student's t test for (B–D).

• 

  Figure 5. CPEB2 deficiency impairs Prdm16 translation and fails to suppress muscle gene expression in BAT. (A) BATs isolated from CPEB2-WT and -KO mice (n = 5) were used for RNA-seq. The principal component analysis (PCA) plot in the left shows different distribution between CPEB2-WT and -KO BATs. The volcano plot shows the red dots and blue dots respectively representing upregulated and downregulated transcripts in CPEB2-KO BATs. (B) The top 10 gene ontology (GO) categories are listed in the right ranking with their P values. The numbers of differentially expressed transcripts in each category are also listed. (C) PRDM16 switches the cell fates of Myf5+ precursors differentiating to brown adipoblasts or myoblasts. (D) CPEB2-WT and -KO BATs from young (≤5 mo, n = 5 per group) and old (7–13 mo, n = 4 per group) male mice were used for western blotting and quantified relative to α-tubulin level. (E) RT-qPCR analysis of Acta1, Myh1, Myl1 and Prdm16 mRNA levels (relative to Gapdh) in BATs collected from mice of 2-mo-old (1M+1F), 5-mo-old (2M) and 11-mo-old (2M+1F). (F) RNA immunoprecipitation (RNA-IP) was conducted using BAT lysates from adult male mice with control (Ctrl) or CPEB2 (CP2) IgG.…

• 

  Figure 6. Ectopic expression of PRDM16 in CPEB2-KO BATs reduces weight gain and muscle gene expression. (A) The immunostaining images of CPEB2-WT and -KO BAT infected with AAV-GFP or AAV-flag-PRDM16. Scale, 50 μm. (B) The accumulated body weight gains after AAV infection in CPEB2-WT and -KO BATs. Numbers in parentheses denote the number of mice in each group. Color asterisks, KO + GFP compared with KO + PRDM16 (magenta); with WT + GFP (green); with WT + PRDM16 (red), and #, WT + PRDM16 vs KO + PRDM16. ∗ #P < 0.05, ∗∗ ##P < 0.01, ∗∗∗ <0.001, two-way ANOVA with Fisher's LSD post-hoc test. (C) CPEB2-WT and -KO BATs injected with AAV-GFP or AAV-flag-PRDM16 (3 mice for each group) were isolated for RNA-Seq. The PCA plot shows specified distribution between each group. (D) The top 1500 genes with the most variable expression were processed for k-means clustering analysis and grouped into 2 GO categories. The heat map is shown on the left, where shades of red and blue represent the relative level of upregulation and downregulation, respectively, among the 12 samples. The detailed pathways in the 2 categories are listed in the right panel, with the number of genes indicated for each pathway. (E)…