Fig 2: Role of micro-RNA (miR)-27b in liraglutide-induced browning of white adipocytes: (a) quantitative real-time polymerase chain reaction (qRT-PCR) analysis of relative miR-27b levels in epididymal white adipose tissue (WAT) in high-fat diet (HF)-induced obese rats following administration of different doses of liraglutide for 12 weeks; HF-NC, high-fat diet control group; HF + liraglutide-L group, HF + 0.4 mg/kg liraglutide; HF + liraglutide-M group, HF + 0.6 mg/kg liraglutide; liraglutide-H group, HF + 0.8 mg/kg liraglutide; *P < 0.05, **P < 0.01 versus the control group; ##P < 0.01 versus the HF-NC group; one-way analysis of variance followed by Tukey’s post hoc test. All data are shown as the mean ± SD of eight rats; (b) qRT-PCR analysis of relative miR-27b levels in mature white adipocytes after treatment with liraglutide (1 µmol/l or 10 µmol/l) for 2 days; **P < 0.01 versus the blank group; (c) transfection efficiency was measured using qRT-PCR in mature white adipocytes transfected with mimic-NC or miR-27b mimics treated with/without 1 µmol/l liraglutide for 48 h; (d–h) relative mRNA levels of uncoupling protein 1 (UCP1), PR domain containing 16 (PRDM16), CCAAT enhancer binding protein ß (CEBPß), cell death-inducing DFFA-like effector A (CIDEA) and peroxisome proliferator-activated receptor-?-coactivator 1a (PGC-1a) in mature white adipocytes transfected with mimic-NC or miR-27b mimics treated with/without 1 µmol/l liraglutide for 48 h; (i–n) Relative protein levels of UCP1, PRDM16, CEBPß, CIDEA and PGC-1a in mature white adipocytes transfected with mimic-NC or miR-27b mimics treated with/without 1 µmol/l liraglutide for 48 h as measured by Western blotting. *P < 0.05, **P < 0.01 versus the mimic-NC group; #P < 0.05, ##P < 0.01 versus the miR-27b mimics group; one-way analysis of variance followed by Tukey’s post hoc test. All data are shown as the mean ± SD of three independent experiments.

Fig 3: Abolishing CIDEA-mediated lipid storage impairs morphogenesis.a Z-stack images of BODIPY 493/503-stained LDs in spheroids formed by WT and CIDEA KO (clone 19.7) ESCs (n = 3). Apical domains of spheroids are marked by concentrated F-ACTIN labelled with phalloidin. Similar data were obtained with clones 16.1 and 19.9. b Cidea transcript expression (RT-qPCR) along with the pluripotency Pou5f1 (also known as Oct4), Esrrb (naive), and Otx2 (primed) factors in WT (black) and CIDEA KO (red) ESCs prior to induction (0 h) and post-induction of spheroid differentiation (24, 48 and 72 h). Error bars, means ± s.e.m. (n = 3) (see also Supplementary Fig. 5b–e for additional RNA-sequencing analysis). Quantification of c LD volume (µm3) and d neutral lipid content (BODIPY fluorescence intensity normalised to spheroid volume) in spheroids from (a). Each symbol in (d) represents one single spheroid at 24 h (green) and 48 h (magenta) post-induction. 30 spheroids per condition and time-point were analysed across independent experiments (n = 3). Error bars, means ± s.e.m.; two-sided Mann–Whitney U test; *p < 0.05; ***p < 0.001. Exact p-valued: c WT 24 h vs CIDEA KO 24 h, p = 2.1 × 10-13; WT 24 h vs WT 48 h, p = 1.2 × 10-8, and d WT 24 h vs CIDEA KO 24 h, p = 0.0002; WT 48 h vs CIDEA KO 48 h, p = 0.0242; WT 24 h vs WT 48 h, p = 3.5 × 10-8; CIDEA KO 24 h vs CIDEA KO 48 h, p = 0.0364. e Quantification of the phenotype of WT and CIDEA KO spheroids 72 h post-induction of differentiation, scored as central-lumened (CL), multi-lumened (ML) and disorganised (D) (see also Supplementary Fig. 6a,b). n \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\ge$$\end{document}= 137 spheroids; error bars, means ± s.e.m. (n = 3); two-sided Chi-square (Fisher exact) test; ***p < 0.001 (p = 8.6 × 10-13). f Quantification of the phenotype of spheroids formed by CIDEA KO ESCs stably transfected with control (Empty), wild-type (WT) and point-mutated (R47E and R171E) V5-tagged Cidea constructs (see also Supplementary Fig. 3b; Supplementary Fig. 6g). n \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\ge$$\end{document}= 177 spheroids; error bars, means ± s.e.m. (n = 3); two-sided Chi-square (Fisher exact) test; ***p < 0.001 (p = 2 × 10-11). g Experimental set-up. LifeAct RFP-expressing WT and CIDEA KO ESCs were injected in E2.5 embryos and transferred back into surrogate females. Embryo chimaeras were harvested and analysed at E5.5 or E7.5. h Immunostaining of representative highly colonized E5.5 ESC (labelled by RFP) embryo chimaeras (n = 10–13 embryos). Epiblast is labelled by the lineage marker OCT4. White arrows, foci of lumen formation marked by PODXL (see also Supplementary Movies 2, 3). i Quantification of the epiblast phenotype of E5.5 embryo chimaeras from (h) scored as central-lumened (CL), multi-lumened (ML) and disorganised (D). Embryo numbers: WT, CL = 9/13, ML = 0/13, D = 4/13; CIDEA KO, CL = 1/10, ML = 7/10, D = 2/10. Two-sided Chi-square (Fisher exact) test; ***p < 0.001 (p = 0.00095). Source data are provided as a Source data file.

Fig 4: Effects of liraglutide treatment on the browning of white adipocytes in vitro: (a) representative images of pre-adipocytes and mature adipocytes using Oil Red O staining (scale bar 100 µm); (b) representative images of mature adipocytes treated with different concentrations of liraglutide (1 µmol/l or 10 µmol/l) using Oil Red O staining (scale bar 100 µm); (c) relative TG levels in each group; (d–h) relative mRNA levels of browning-related markers, including uncoupling protein 1 (UCP1), PR domain containing 16 (PRDM16), CCAAT enhancer binding protein ß (CEBPß), cell death-inducing DFFA-like effector A (CIDEA) and peroxisome proliferator-activated receptor-?-coactivator 1a (PGC-1a) in mature white adipocytes; (i–n) relative protein levels of UCP1, PRDM16, CEBPß, CIDEA and PGC-1a in mature white adipocytes were measured by Western blotting. *P < 0.05, **P < 0.01 versus the blank group; one-way analysis of variance followed by Tukey’s post hoc test. All data are shown as the mean ± SD of three independent experiments. The colour version of this figure is available at: http://imr.sagepub.com.

Fig 5: CIDEA-mediated LD enlargement confers protection against promiscuous degradation by cytosolic lipases.a Time-lapse live imaging of BODIPY 493/503 pulse-labelled LDs in spheroids formed by LifeACT-RFP-expressing WT and CIDEA KO ESCs 24 h post-induction of differentiation (see also Supplementary Movies 4–6). White arrows indicate small LDs being degraded in CIDEA KO spheroids. Representative images of ~9 spheroids across independent experiments (n = 7). b Representative images of BODIPY-stained LDs in CIDEA KO spheroids treated with DMSO (CIDEA KODMSO) and the cytosolic lipase inhibitor DEUP (CIDEA KODEUP) for 24 h upon induction of differentiation (n = 3). DEUP, Diethylumbelliferyl phosphate. c Quantification of neutral lipid content (BODIPY fluorescence intensity normalised to spheroid volume) in spheroids from (b). n = 30 spheroids per condition and time-point. Error bars, means ± s.e.m. (n = 3); two-sided Mann–Whitney U test, **p < 0.01 (p = 0.0053). d Representative images of CIDEA KODMSO and CIDEA KODEUP spheroids at indicated time-points. Foci of lumen formation are marked by PODXL, PARD3 and F-ACTIN (n = 3). e Quantification of the phenotype of spheroids from (d). n \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\ge$$\end{document}= 168 spheroids; error bars, means ± s.e.m. (n = 3); two-sided Chi-square (Fisher exact) test, **p < 0.01 (48 h, p = 0.0016); ***p < 0.001 (72 h, p = 8 × 10-7). R, rosette; NL, nascent lumen; CL, central-lumened; ML, multi-lumened; D, disorganised. Source data are provided as a Source data file.