Fig. 1. Reduced adiposity and enhanced beiging in Klf3^−/− mice.

a Lean and fat body mass composition (%) of WT (n = 12) and Klf3^−/− (n = 10) mice were assessed by EchoMRI analysis. b Weights of WT and Klf3^−/− AT depots were recorded as a percentage of total body weight (n = 6 mice for subcut and gonadal AT; n = 5 mice for brown AT). c Representative macroscopic images of WT and Klf3^−/− subcut AT pads showing relative size and complexion. d mRNA levels of thermogenic genes were assessed by qPCR in WT and Klf3^−/− subcut AT (n = 4–8 mice). Relative expression was normalised to 18S rRNA levels and the mean WT value for each gene was set to 1. e UCP1 protein expression was measured in WT and Klf3^−/− subcut AT by western blotting (n = 4 mice). 25 μg of whole-cell extract (WCE) was loaded alongside a Rainbow molecular weight marker. Nitrocellulose membranes were probed with anti-UCP1 antibody overnight. Diluted brown adipose tissue (BAT) extract included as a positive control. f Expression of mitochondrial oxphos proteins in WT and Klf3^−/− subcut AT WCE was assessed by western blotting (n = 4 mice). 25 μg of extract was loaded alongside rat heart mitochondria extract (Mito; positive control), and PVDF membranes were blocked overnight before being probed with the Total OXPHOS Rodent WB Antibody Cocktail for 2 h. Mitochondrial complexes are labelled by their Roman numerals (I–V). g VDAC protein expression was measured in WT and Klf3^−/− subcut AT (n = 4 mice) by western blotting. 20 μg of WCE was loaded and nitrocellulose membranes were probed overnight with anti-VDAC antibody. For e–g β-actin was included as a loading control. For a, b and d, error bars represent means ± SEM and one-sided non-parametric Mann–Whitney U tests were performed where *P < 0.05, **P < 0.01, ***P < 0.001. Source data are provided as a Source data file. kDa, kilodaltons. See also Supplementary Figs. 1 and 2.