ARTICLES ●cold owam●cold 2,500 2,000 200 Cold Cyp7b1-- o WT cold 4.000 Cold 250- Cold 81001头鲁 UBA CBA Housing temperature (C) Sic 10a2 UBA CBA O Warm k O War●cod● Warm AB● Cold AB =1 80 是0001 actvity Figure 3 Cold exposure promotes fecal excretion of CYP7Bl-derived bile acids. (a, b)Relative levels of bile acid species (a)and the sum of the amounts of UBA and CBa species(b)in the feces of cold- housed and warm-housed mice(n=5 mice per group). (c, d)Heat map representation of the fecal levels of bile acid species in cold-housed WT and Cyp/- mice, relative to those in the respective warm-housed controls, (c)and total levels of UBA g and CBA species in warm-housed and cold-housed WT and Cyp 7b1--mice(d)(n= 3 mice per group). (e)Bile acid levels in the feces of mice that were 9 of S/c10a2, which encodes apical sodium-dependent bile transporter(ASBT)(n=8 mice per group), (f)and ASBT protein (n=3 mice per group)(g)in arm-housed and cold- housed mice. One representative out of two technical replicates of three biological replicates is shown. Uncropped western blot images are shown in Supplementary Figure 15.(h, i) Bile acid concentrations in plasma from portal vein blood (warm-housed, n=8 mice; cold-housed 9 mice)(h)and in plasma from systemic blood (n= 4 mice per group)(i)of warm-housed and cold- housed mice. gj, k)Heat map showing fecal bile acid species (relative to those in warm-housed controls)(), as well as total UBA and CBa levels (k), from warm-housed and cold-housed mice that were not or were treated with antibiotics(AB)(warm-housed: no AB (warm), n=l0 mice; +AB, n= 10 mice; cold- housed: no AB(cold), n=10 mice: +AB n=9 mice).(1)Cecal BSH activity in warm-housed and cold-housed mice that were not or were treated with antibiotics (warm-housed: no AB (warm), n=7 mice: +AB, n= 6 mice; cold- housed: no AB(cold), n=6 mice; +AB, n=5 mice).(m) Fecal taurine concentration in warm- housed and cold- housed mice(n= 5 mice per group). Throughout, data are mean +s.e. m 'P<0.05, P<0.01, P<0.00l; by unpaired two tailed Students t-test (a, b, d-f, h, i, m)or two-way ANOVA (k, i) warm-housed control mice(Fig. 5f-h), indicating that dietary cho- Bile acids shape the gut microbiome in cold-housed mice lesterol uptake determined cold-induced fecal bile acid excretion. It is conceivable that cold-related endogenous metabolites generated Treatment with EZ resulted in a higher fecal cholesterol content with- by the host, such as bile acids, have an effect on gut bacteria27.We ut affecting food intake or fecal mass( Supplementary Fig. 9d-f). therefore postulated that cold-induced bile acid levels would allow Elevated dietary cholesterol intake induced by cold exposure was selection of intestinal bacteria and thus determine the microbiome in (Fig. 1j, k), we calculated the absorption of cholesterol and its conver- housed and cold-housed mice that were or were not treated with EZ sion to bile acids, which we determined from feces(Supplementary MDS analysis revealed a clear separation of the cold-housed and Table 1b), and observed 17. 1% conversion in warm-housed mice, warm-housed groups( Fig. 5i), whereas EZ-treated cold-housed mice whereas cold exposure caused a 70% increase in the conversion rate; clustered similarly to the untreated warm-housed group( Fig. 5i). thus, 29.2% of absorbed cholesterol was converted to bile acids in The EZ-treated warm-housed group clustered in a position opposite cold-housed mice. Taking into account the higher amount of choles- to the untreated cold-housed group, which is consistent with lower terol absorption, cold-housed mice produced 5 8-fold more amounts fecal bile acid levels observed in the EZ-treated warm-housed mice, of bile acids than warm-housed controls compared to that in the untreated warm-housed group( Fig. 5f)© 2017 Nature America, Inc., part of Springer Nature. All rights reserved. Ar t i c l e s  advance online publication nature medicine warm-housed control mice (Fig. 5f–h), indicating that dietary cho￾lesterol uptake determined cold-induced fecal bile acid excretion. Treatment with EZ resulted in a higher fecal cholesterol content with￾out affecting food intake or fecal mass (Supplementary Fig. 9d–f). Elevated dietary cholesterol intake induced by cold exposure was accompanied by higher fecal content of cholesterol (Supplementary Table 1b). On the basis of metabolic studies using a radioactive tracer (Fig. 1j,k), we calculated the absorption of cholesterol and its conver￾sion to bile acids, which we determined from feces (Supplementary Table 1b), and observed 17.1% conversion in warm-housed mice, whereas cold exposure caused a 70% increase in the conversion rate; thus, 29.2% of absorbed cholesterol was converted to bile acids in cold-housed mice. Taking into account the higher amount of choles￾terol absorption, cold-housed mice produced 5.8-fold more amounts of bile acids than warm-housed controls. Bile acids shape the gut microbiome in cold-housed mice It is conceivable that cold-related endogenous metabolites generated by the host, such as bile acids, have an effect on gut bacteria27. We therefore postulated that cold-induced bile acid levels would allow selection of intestinal bacteria and thus determine the microbiome in cold-housed mice. To test this possibility, we performed sequencing of the gene encoding the 16S rRNA using fecal samples from warm￾housed and cold-housed mice that were or were not treated with EZ. MDS analysis revealed a clear separation of the cold-housed and warm-housed groups (Fig. 5i), whereas EZ-treated cold-housed mice clustered similarly to the untreated warm-housed group (Fig. 5i). The EZ-treated warm-housed group clustered in a position opposite to the untreated cold-housed group, which is consistent with lower fecal bile acid levels observed in the EZ-treated warm-housed mice, as compared to that in the untreated warm-housed group (Fig. 5f). 1,000 Warm Cold Warm Cold Warm Cold 55 Mr (kDa) 40 Mr (kDa) 250 200 150 100 50 0 * * * * * * ** * * * 800 2,500 2,000 1,500 1,000 500 0 2,000 2,000 3,000 4,000 2.0 1.5 ASBT β-actin Gene expression (fold) 1.0 0.5 0.0 Slc10a2 * * * * WT cold AAV-GFP AAV-Cyp7b1 1,500 1,000 500 1,000 0 0 30 22 16 UBA CBA UBA CBA UBA CBA Warm AB Cold AB UBA 6 4 2 * 0 CBA Housing temperature (°C) Cold wild type 0 1 >7 0 1 >19 Cold Cyp7b1–/– Cyp7b1–/– cold Bile acids (ng/mg) Bile acids (ng/mg) Bile acids (ng/mg) Bile acids (µM) Bile acids (ng/mg) Bile acids (µM) 600 400 200 0 CA β-MCA DCA/CDCA UDCA α/ω-MCA T-α/β-MCA THDCA TUDCA TCDCA/TDCA TLCA TCA GCDCA GCA CA β-MCA α-MCA CDCA/DCA UDCA ω TDCA -MCA T-α/β-MCA THDCA TUDCA TCDCA TLCA TCA GCDCA GCA CA β-MCA α-MCA CDCA/DCA UDCA ω-MCA T-β-MCA T-α-MCA THDCA TUDCA TDCA TLCA TCA GCDCA GCA Warm Cold Warm Cold Warm Cold Warm Cold Warm AB Cold AB UBA CBA Bile acids (ng/mg) 8,000 0.08 100 80 60 40 20 0 0.06 0.04 0.02 0.00 BSH activity Taurine (mmol/10 min) *** Concentration (ng/mg) *** *** *** *** *** *** ** Taurine * *** *** ** 6,000 4,000 2,000 0 a b c d e f g h i j k l m Figure 3 Cold exposure promotes fecal excretion of CYP7B1-derived bile acids. (a,b) Relative levels of bile acid species (a) and the sum of the amounts of UBA and CBA species (b) in the feces of cold-housed and warm-housed mice (n = 5 mice per group). (c,d) Heat map representation of the fecal levels of bile acid species in cold-housed WT and Cyp7b1−/− mice, relative to those in the respective warm-housed controls, (c) and total levels of UBA and CBA species in warm-housed and cold-housed WT and Cyp7b1−/− mice (d) (n = 3 mice per group). (e) Bile acid levels in the feces of mice that were housed at the indicated ambient temperatures after infection with either AAV-GFP or AAV-Cyp7b1 (n = 7 mice per group). (f,g) Ileal mRNA expression of Slc10a2, which encodes apical sodium-dependent bile transporter (ASBT) (n = 8 mice per group), (f) and ASBT protein (n = 3 mice per group) (g) in warm-housed and cold-housed mice. One representative out of two technical replicates of three biological replicates is shown. Uncropped western blot images are shown in Supplementary Figure 15. (h,i) Bile acid concentrations in plasma from portal vein blood (warm-housed, n = 8 mice; cold-housed, n = 9 mice) (h) and in plasma from systemic blood (n = 4 mice per group) (i) of warm-housed and cold-housed mice. (j,k) Heat map showing fecal bile acid species (relative to those in warm-housed controls) (j), as well as total UBA and CBA levels (k), from warm-housed and cold-housed mice that were not or were treated with antibiotics (AB) (warm-housed: no AB (warm), n = 10 mice; +AB, n = 10 mice; cold-housed: no AB (cold), n = 10 mice; +AB, n = 9 mice). (l) Cecal BSH activity in warm-housed and cold-housed mice that were not or were treated with antibiotics (warm-housed: no AB (warm), n = 7 mice; +AB, n = 6 mice; cold-housed: no AB (cold), n = 6 mice; +AB, n = 5 mice). (m) Fecal taurine concentration in warm-housed and cold￾housed mice (n = 5 mice per group). 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,b,d–f,h,i,m) or two-way ANOVA (k,i)