ARTICLES o Mock●Ez 2,: 心 Housing temperature (C) Housing temperature (C) h Cyp7b1- WT·cvp7b °A° oWT●c7b7 O AAV-GFP· AAV-Cyp7b1 O AAV-GFP O AAV-GFP 6000 9401●AVcp 38 Housing temperature (C) z Figure 6 CYP7B1-derived bile acids promote adaptive thermogenesis. (a, b)Body temperature in Wt(n=8 per group)(a)and dbldb(30"C, s with cholesterol (n=8 mice per group).(d) Representative photographs (top)and histology images(bottom)of BaT from mice that were housed at cC(n=2 mice pet ar lel with n=5 images per mouse). Scale bars, 1 cm(top)and 100 um(bottom).(e, f) Relative expression of genes involved in the housekeeping protein AKT)(right)of UCPl expression in BAT and inguinal WAT from WT(n= 9)and Cyp 7b1--(n=8)mice housed at 6C One representative out of three technical replicates of nine(Wt, n=9 mice)or eight (Cyp 7b1-ki n=8 mice)biological replicates is shown. Uncropped estern blot images are shown in Supplementary Figure 15.(h, i)Oxygen consumption profiles(h)and quantification (i)in WT and Cyp- mice that ere housed at 6C and monitored during a 12-h light and a 12-h dark phase(n=9 mice per group).i) Body temperature in WT and Cyp7b1--mice, measured at the indicated ambient temperatures (n=9 mice per group).( k, I )Profiles(k)and quantification (I)of oxygen consumption in mice that wer infected with AAV-GFP or AAV-Cyp7bl and housed at 22C, as monitored during a 12-h light and a 12-h dark phase (n= 6 mice per group).(m)Body mperature in mice that were infected with AAv-GFP (n=7 mice)or AAV-Cyp7bl(n= 4 mice), as measured at the indicated ambient temperature Throughout, data are mean ts.e. m.P<0.05, *P<0.01, **P<0.00l; by unpaired two tailed Students t-test (a-c, e-m). intestinal microbiome are dependent on the synthesis and biliary thermogenesis, we observed by indirect calorimetry that Cyp7b1-- excretion of bile acids by the host mice consumed significantly less O2, which is indicative of dimin- Exogenous bile acids are known to stimulate BAT in mice and temperatures( Fig. response to gradually decreased ambient CYP7Bl-derived bile acids promote thermogenesis body tem gy expenditure,(Fig. 6h, i), and we observed a lower humans24. 25. To study the thermo-metabolic consequences of higher Moreover, AAV-mediated CyP7bl overexpression in the liver evels of endogenous bile acids after cold exposure, we investigated Supplementary Fig. 13a)caused slightly higher expression of genes whether dietary cholesterol uptake and its subsequent conversion via involved in thermogenesis in BAT and inguinal WAT, as compared EZ treatment in cold-housed mice resulted in lower plasma bile acid Fig. 13b, c). This. expressing the control AAV( Supplementary levels as compared to that in untreated controls(Supplementary (Fig. 6k, I), greater tail heat loss(Supplementary Fig. 13d)and higher WT and dbldb mice that were cold-housed(Fig 6a, b). Conversely, that cholesterol conversion to bile acids via CYP7B1 modulates adap dietary cholesterol supplementation led to higher plasma levels of bile tive thermogenesis by brown and beige adipocytes acids(Supplementary Fig 12c, d) and body temperature(Fig. 6c) To study the alternative bile acid pathway in this context, we ana- DISCUSSION lyzed thermogenic parameters in cold-exposed WT and Cyp7b1-l- Obesity and HFDs are known to be related to shifts in the gut micro- miceRelative to WT controls, CyP7b1-- mice showed higher BAT biota. These in turn have an effect on the development and progres- lipid content(Fig. 6d), lower thermogenic gene expression(Fig. 6e, f) sion of chronic metabolic diseases through alterations in circulating andlower amountsof mitochondrial uncoupling protein 1(UCP1), which metabolites and hormones2-6. Recent studies indicate that cold is essential for heat production via adaptive thermogenesis, ( Fig. 6g) exposure leading to the thermogenic activation of beige and brown in BAT and inguinal WAT. Consistent with impaired adaptive adipocytes is associated with changes in the gut microbiome DNLINE PUBLICATION© 2017 Nature America, Inc., part of Springer Nature. All rights reserved. a r t i c l e s nature medicine advance online publication  intestinal microbiome are dependent on the synthesis and biliary excretion of bile acids by the host. CYP7B1-derived bile acids promote thermogenesis Exogenous bile acids are known to stimulate BAT in mice and humans24,25. To study the thermo-metabolic consequences of higher levels of endogenous bile acids after cold exposure, we investigated whether dietary cholesterol uptake and its subsequent conversion via CYP7B1 would affect thermogenesis. Consistent with this concept, EZ treatment in cold-housed mice resulted in lower plasma bile acid levels as compared to that in untreated controls (Supplementary Fig. 12a,b), which was associated with lower body temperatures in WT and db/db mice that were cold-housed (Fig. 6a,b). Conversely, dietary cholesterol supplementation led to higher plasma levels of bile acids (Supplementary Fig. 12c,d) and body temperature (Fig. 6c). To study the alternative bile acid pathway in this context, we ana￾lyzed thermogenic parameters in cold-exposed WT and Cyp7b1−/− mice. Relative to WT controls, Cyp7b1−/− mice showed higher BAT lipid content (Fig. 6d), lower thermogenic gene expression (Fig. 6e,f) and lower amounts of mitochondrial uncoupling protein 1 (UCP1), which is essential for heat production via adaptive thermogenesis, (Fig. 6g) in BAT and inguinal WAT. Consistent with impaired adaptive thermogenesis, we observed by indirect calorimetry that Cyp7b1−/− mice consumed significantly less O2, which is indicative of dimin￾ished energy expenditure, (Fig. 6h,i), and we observed a lower body temperature in response to gradually decreased ambient temperatures (Fig. 6j). Moreover, AAV-mediated Cyp7b1 overexpression in the liver (Supplementary Fig. 13a) caused slightly higher expression of genes involved in thermogenesis in BAT and inguinal WAT, as compared to that in liver overexpressing the control AAV (Supplementary Fig. 13b,c). This was accompanied by higher O2 consumption (Fig. 6k,l), greater tail heat loss (Supplementary Fig. 13d) and higher body temperature (Fig. 6m). In conclusion, these data demonstrate that cholesterol conversion to bile acids via CYP7B1 modulates adap￾tive thermogenesis by brown and beige adipocytes. DISCUSSION Obesity and HFDs are known to be related to shifts in the gut micro￾biota. These in turn have an effect on the development and progres￾sion of chronic metabolic diseases through alterations in circulating metabolites and hormones2–6. Recent studies indicate that cold exposure leading to the thermogenic activation of beige and brown adipocytes is associated with changes in the gut microbiome that are 40 Mock * *** *** ** * * EZ 39 38 37 36 40 * * * Chow WT Cyp7b1–/– WT Cyp7b1–/– WT Cyp7b1–/– WT Cyp7b1–/– WT Cyp7b1–/– WT Cyp7b1–/– WT Cyp7b1 WT Cyp7b1 –/– –/– Chow + Chol 39 2.5 ** *** ** * *** Gene expression (fold) 2.0 1.5 1.0 0.5 0.0 2.5 UCP1 AKT UCP1 AKT BAT ingWAT BAT ** ** UCP1/AKT (a.u.)1.5 6,000 4,000 VO2 (ml/h/kg) VO2 (ml/h/kg) 2,000 0 6,000 AAV-GFP AAV-Cyp7b1 AAV-GFP AAV-Cyp7b1 AAV-GFP AAV-Cyp7b1 4,000 VO2 (ml/h/kg) VO2 (ml/h/kg) 2,000 0 6,000 8,000 4,000 2,000 0 6,000 4,000 2,000 Light phase Dark phase Light phase Dark phase * * 0 1.0 0.5 0.0 ing WAT Gene expression (fold) 2.0 1.5 1.0 0.5 0.0 38 37 36 40 db/db 30 °C db/db 16 °C db/db 30 °C EZ db/db 16 °C EZ 38 36 34 32 30 22 16 6 30 22 16 Ucp1 Dio2 Elovl3 Ppargc1a Prdm16 Ucp1 Dio2 Elovl3 Ppargc1a Prdm16 6 30 22 16 6 Rectal temperature (°C) 40 ** 39 * 38 37 36 Rectal temperature (°C) 30 22 16 40 39 38 37 36 Rectal temperature (°C) Rectal temperature (°C) Rectal temperature (°C) a b c d e f g h i j k l m Housing temperature (°C) Housing temperature (°C) Housing temperature (°C) Housing temperature (°C) Figure 6 CYP7B1-derived bile acids promote adaptive thermogenesis. (a,b) Body temperature in WT (n = 8 per group) (a) and db/db (30 °C, n = 7; 16 °C, n = 8; 30 °C +EZ, n = 7; 16 °C +EZ, n = 6) (b) mice that were treated without or with EZ and housed at the indicated ambient temperatures. (c) Body temperatures in mice that were housed at the indicated ambient temperatures and fed a chow diet supplemented without or with cholesterol (n = 8 mice per group). (d) Representative photographs (top) and histology images (bottom) of BAT from mice that were housed at 6 °C (n = 2 mice per group, with n = 5 images per mouse). Scale bars, 1 cm (top) and 100 µm (bottom). (e,f) Relative expression of genes involved in thermogenesis in BAT (e) and inguinal WAT (f (WT, n = 9 mice; Cyp7b1−/−, n = 8 mice)). (g) Western blot analysis (left) and quantification (relative to the housekeeping protein AKT) (right) of UCP1 expression in BAT and inguinal WAT from WT (n = 9) and Cyp7b1−/− (n = 8) mice housed at 6 °C. One representative out of three technical replicates of nine (WT, n = 9 mice) or eight (Cyp7b1−/−; n = 8 mice) biological replicates is shown. Uncropped western blot images are shown in Supplementary Figure 15. (h,i) Oxygen consumption profiles (h) and quantification (i) in WT and Cyp7b1−/− mice that were housed at 6 °C and monitored during a 12-h light and a 12-h dark phase (n = 9 mice per group). (j) Body temperature in WT and Cyp7b1−/− mice, measured at the indicated ambient temperatures (n = 9 mice per group). (k,l) Profiles (k) and quantification (l) of oxygen consumption in mice that were infected with AAV-GFP or AAV-Cyp7b1 and housed at 22 °C, as monitored during a 12-h light and a 12-h dark phase (n = 6 mice per group). (m) Body temperature in mice that were infected with AAV-GFP (n = 7 mice) or AAV-Cyp7b1 (n = 4 mice), as measured at the indicated ambient temperatures. Throughout, data are mean ± s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001; by unpaired two-tailed Student’s t-test (a–c,e–m)