Glucose and insulin tolerance tests in BCATm null and wild-type mice
A) Blood glucose (left panel) and plasma insulin concentrations (right panel) during glucose tolerance test. * P < 0.05, ** P < 0.01 and *** P < 0.001, n=7–8.
B) Blood glucose expressed as a percent of basal level during insulin tolerance test. Left panel: ITT in male mice fed a choice of normal chow and −BCAA diets as described in Figure 1. *** P < 0.001, n=7–9 for each group. Right panel: ITT in male mice fed a 60% fat-containing diet for 10 week as described Figure 3. * P < 0.05 and *** P < 0.001, n=7–8.

Elevated energy expenditure in BCATm null mice is associated with food consumption and is partially blunted by Rapamycin
A) Oxygen consumption (VO₂, left panel), respiratory quotient (RQ, middle panel), and food intake (FI, right panel). Male mice fed a choice of amino acid purified BCAA-containing (+BCAA) and BCAA-free diet for 4 weeks were placed in indirect calorimetry chambers at the age of ~16 weeks. Food intake was measured for 3 weeks and calculated as average daily values. It was also adjusted for body weight (25g). * P < 0.05, ** P < 0.01 and *** P < 0.001, n=8.
B) VO₂ (left panel) and RQ (right panel) during fasted-refed. Male mice fed a mix of NC and −BCAA diets were fasted for 21 h and refed with a choice of NC and BCAA diets. VO₂ and RQ were measured during fasting (light and dark phases) and a 3-h refeeding period. * P < 0.05, ** P < 0.01 and *** P < 0.001, n=6.
C). VO₂ after treated with rapamycin during fasted-refed. 12–15 week old mice were i.p. injected with 0.75 mg/kg of rapamycin at 11 a.m. and fasted for 21 h and injected again with the same dose of rapamycin. Food was provided 1-h after second injection for 3 h. VO₂ was measured during the dark phase and refeeding. One way ANOVA was used to compare VO₂ among groups under each nutritional condition. No difference was found between groups during fasting. * P < 0.05 between groups during refed, n=8–11.

Elevated protein turnover in vivo and mTOR signaling in BCATm null mice
A) In vivo protein synthesis rates measured by flooding does of L-[³H]phenylalanine. ** P < 0.01, n=10–13.
B) Amount of 24-h urinary 3-methyhistidine and molar ratio of urine 3-methyhistidine to creatinine. Eight male mice from each genotype were placed in individual metabolic cages (Nalegene, Rochester, NY) for 2 days. Daily urine was collected for analysis of 3-methyhistidine and creatinine. ** P < 0.01 and *** P < 0.001, n=16.
C) Western blot analysis for 4E-BP1 and pT37/46 4E-BP1 in gastrocnemius of BCATm^−/− mice fed a choice of NC/−BCAA diets. ** P < 0.01, n=8.
D) Western blot analysis for 4E-BP1, pS235/236 S6, S6, pT389 S6K1, and S6K1 in gastrocnemius of fasted (F) and fasted-refed (R) BCATm^−/− mice. * P<0.05, ** P < 0.01 and *** P < 0.001, n=4 for fasted and 8 for fasted-refed.

Growth curve, food intake, organ weight, and fat size
A) Immunoblots of BCATm and BCATc in selected organs from wild-type and BCATm null mice. Equal amounts of protein (20μg) were loaded in each lane of tissue samples examined.
B) Growth curve of BCATm null and wild-type mice. Male mice were weighed weekly, ** P < 0.01, n=6–12.
C) Food intake in BCATm^−/− and wild-type mice. Mice were fed with a choice of normal chow (NC) and amino acid purified BCAA-free (−BCAA) diets at weaning. Food intake (FI) was measured for 3 weeks at the age of ~10 weeks and calculated as average daily values. It was also adjusted for body weight (25g). * P < 0.05 and *** P < 0.001, n=8–10.
D) Relative organ weights in BCATm null and wild-type mice. Values are expressed as percent of +/+ mice and adjusted for body weight (BW). EF, epididymal fat; L, liver; H, heart; K, kidney; G, gastrocnemius muscle; B, brain, BF, brown fat. * P < 0.05 and ** P < 0.001, n=6, male mice at the age of ~16 weeks.
E) H.E. staining of paraformaldehyde fixed section of epididymal fat.

BCATm null mice are protected from high fat diet induced obesity
A) Growth curve (right panel) and representative picture of +/+ and −/− mice after high-fat-diet feeding (left panel). ** P < 0.01, −/− vs. +/+ males at each time point, n=7–8 for each group.
B) Food intake (FI) measured during high fat diet feeding. Food consumption was measured for 1 week; and average food intake was calculated and also normalized to body weigh. * P < 0.05, n=7–9.
C) Representative MRI of mice after high fat diet feeding. Transverse images set at the same distance from the anus were taken for both mice. Abdominal and subcutaneous fat shown as white was separated by the peritoneal membrane.

Common factors associated with thermogenesis.
A) Locomotor activity measured in the indirect calorimetry and B) Rectal core temperature in BCATm null and wild-type mice fed with a choice of NC and −BCAA diet. C) mRNA expression of selected genes in brown fat (left panel) and gastrocnemius muscle (right panel) in mice fed with a choice of +BCAA and −BCAA diet.