Supplement 1 Gao Lipid in serum Serum FFA

Supplement 1 (Gao) Lipid in serum Serum FFA in fed condition Fatty acids control butyrate 5. 48 9. 40* C 16: 0 745. 99 634. 49* C 18: 0 323. 65 294. 51* C 20: 3 n 6 75. 00 59. 13* C 22: 6 n 3 137. 20 107. 53* 3978. 879 3263. 78* C 4: 0 Total N=3, *, P<0. 05 Supplement 1. Fatty acids in serum were analyzed as methyl esters (FAMEs) according to the method of O’Fallon modified for small volumes [1]. In short, duplicate aliquots of 40 μl of serum were transferred to a 1. 5 ml amber glass vial along with triheptadecenoin surrogate standard. Methanol (530 μl) and 10 N KOH (70 μl) were added, samples were vortexed for 30 s and incubated in a shaking water bath at 55 ºC for 1. 5 hours. After incubation, samples were cooled and 58 μl of 24 N H 2 SO 4 was added, and samples were incubated for 1. 5 hours in shaking water bath. Fatty acid methyl esters were extracted into hexane (300 μl) and 90 μl was transferred into a vial with insert along with 10 μl of a C 23: 0 methyl ester internal standard. Fatty acid methyl esters were analyzed by gas chromatography with flame ionization detection (Model QP 2010, Shimadzu Co. , Columbia, MD). Samples containing methyl esters in hexane (1 l) were injected onto an HP-88 fused silica 100 m x 0. 25 mm column, 0. 20 μm film (Agilent Technologies, Palo Alto, CA). The injection port was maintained at 250ºC in the split mode, and the sample was split at a 10: 1 ratio with a 3. 0 ml/min purge flow. Hydrogen was used as the carrier gas at a linear velocity of 41. 1 cm/s. The temperature program was as follows: initial temperature 35°C and hold for 2 min, ramp at 40°C/min to 175ºC and hold for 4 min, ramp at 3. 5 ºC/min to 240ºC and hold for 25 min. The detector was operated at 250ºC and makeup gas was nitrogen 30 ml/min. Air and hydrogen flow to the detector was 450 ml/min and 40 ml/min, respectively. Total run time was 53. 07 min/sample. Peaks were identified according to retention time similarity to several standard FAME mixes (Nu. Chek Prep, Inc. , Elysian, MN). Raw peak areas were converted to concentration using response factors generated with standard FAME mixtures. [1] O'Fallon JV, Busboom JR, Nelson ML, Gaskins CT. A direct method for fatty acid methyl ester synthesis: application to wet meat tissues, oils, and feedstuffs. J Anim Sci 2007; 85(6): 1511 -21.

Supplement 2 (Gao) TSA effect B. ITT in TSA mice Glucose (mg/d. L) BW (g) A. Body weight * C. Insulin signaling in muscle Control TSA * 0 TSA p. IRS-1 Y 632 * * p. Akt Ser 473 Akt 30’ 60’ 120’ 180’ Time E. Type I fiber F. Immunoblot Control TSA Control PGC-1 Myosin TSA Myoglobin G. HDAC assay TSA HDAC activity Fold change D. Vastus laterais * Tubulin Supplement 2. TSA effect in mice. The mice were fed one HFD at 5 weeks in age to induce obesity. TSA was administrated at the dose of 0. 6 ug/kg/day. TSA treatment began together with HFD feeding. The experiments were conducted at 12 -13 weeks on HFD. A. Body weight was shown at the 12 week. B. ITT was done at 12 weeks on HFD (at 17 weeks of age). ITT was performed after 4 hour fast. C. Insulin signaling. The gastrocnemius muscle was isolated after insulin (0. 75 U/kg) injection for 30 minutes and used to prepare the whole cell lysate for immunoblot. IRS-1 and Akt were examined for tyrosine (Y 632) and serine (S 473) phosphorylation. D. Vastus laterais muscle. The muscle tissue was isolated from mice that were fed on HFD for 13 weeks. E. Type I fibers in serial cryostat sections of muscle. The muscle slides made from vastus lateralis were stained with antibody against the type I myosin heavy chain in the oxidative fiber and indicated by the brown color. The photograph was taken at 20 X magnification. F. Immunoblot. Whole cell lysate was prepared from gastrocnemius muscle and analyzed in an immunoblot. PGC-1 , type I myosin heavy chain (Myosin) and myoglobin were blotted with specific antibodies. G. HDAC activity in skeletal muscle. The results are the mean ± SE (n=8 in each group). *P<0. 05 (vs. control).

Supplement 3 (Gao) 2. 5% w/w butyrate-treated mice B. Fat content BW (g) A. Body weight * %BW * Control Butyrate C. ITT D. GTT Glucose (mg/dl) Control Butyrate * * * Supplement 3. Adiposity and insulin sensitivity in 2. 5% w/w butyrate-treated mice. A. Body weight at 12 weeks on HFD (at 16 weeks of age). B. Fat content at 12 weeks on HFD (at 16 weeks of age). C. ITT in butyrate-treated mice. ITT was done at 12 weeks on HFD (at 16 weeks of age). D. Glucose tolerance test (GTT). GTT was done at 13 weeks on HFD (at 17 weeks of age). After overnight fast, 2. 5 g/kg glucose was injected intraperitoneally. Blood glucose was measured at times as indicated. Values are the means ± SE (N=8 in each group). * P<0. 05.