Altered Microbiota Contributes to Reduced Diet-Induced Obesity upon Cold Exposure

Abstract

Maintenance of body temperature in cold-exposed animals requires induction of thermogenesis and management of fuel. Here, we demonstrated that reducing ambient temperature attenuated diet-induced obesity (DIO), which was associated with increased iBAT thermogenesis and a plasma bile acid profile similar to that of germ-free mice. We observed a marked shift in the microbiome composition at the phylum and family levels within 1 day of acute cold exposure and after 4 weeks at 12°C. Gut microbiota was characterized by increased levels of Adlercreutzia, Mogibacteriaceae, Ruminococcaceae, and Desulfovibrio and reduced levels of Bacilli, Erysipelotrichaceae, and the genus rc4-4. These genera have been associated with leanness and obesity, respectively. Germ-free mice fed a high-fat diet at room temperature gained less adiposity and improved glucose tolerance when transplanted with caecal microbiota of mice housed at 12°C compared to mice transplanted with microbiota from 29°C. Thus, a microbiota-liver-BAT axis may mediate protection against obesity at reduced temperature.

Graphical abstract

Figure 1

Energy Balance Phenotype in Response to Reduced Ambient Temperature and Thermoneutrality in Mice Fed HFD or CHD for 4 Weeks (A–D) Change in fat mass (A); adiposity index (B); cumulative food intake (C); total energy expenditure (D). (E–H) Ucp1 mRNA (E) and protein (F) by immunoblot in iBAT. Ucp1 mRNA (G) and protein (H) by immunoblot in ING. (I–K) Hepatic induction of pAMPK (I), pACC (J), and Cpt1a mRNA level (K). (L) Glucose tolerance test. Data are presented as mean ± SEM. n = 8–10 (GTT and qRT-PCR) or 5–6 (immunoblot). FM, fat mass; LM, lean mass; cFI, cumulative food intake; TEE, total energy expenditure. ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001. See also Figure S1.

Figure 2

The Gut Microbiota Composition and Microbial Metabolism Shaped by Ambient Temperature and Diet (A–H) Rarefaction curves calculated for phylogenetic distance between the microbiota of mice fed a HFD and a CHD at 12°C, 17°C, and 29°C. Rarefaction depth used is 60,000 sequences per sample (A). Principal coordinate analysis (PCoA) of unweighted UniFrac revealed clustering of the gut microbiota after diet and temperatures. The percentage of the variation explained by the plotted principal coordinates is indicated in the axis labels. Each dot represents a caecal community (B). Relative abundance at phylum (C) and family (D) level in caecal community of mice fed a HFD and a CHD at 12°C, 17°C, and 29°C. Hepatic expression of genes involved in BA synthesis (E). Plasma BA profile in mice fed a HFD (F), a CHD (G), and a ration of t-conjugated to unconjugated BAs (H). The data are given as mean ± SEM. n = 8. p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001. See also Figure S2.

Figure 3

Time-Dependent Effects of Acute Cold Exposure on Energy Balance Phenotypes and Gut Microbiota Composition of Mice with Diet-Induced Obesity (A–H) Fat mass (A), adiposity (B), food intake (C), and Ucp1 expression in iBAT and ING (D) of mice fed a HFD and housed at 29°C for 4 weeks, then transferred to 12°C for 6 days. Change in BA composition measured in plasma (E), phylogenetic diversity comparison of caecal microbial communities (F), phylum level abundance (G), and ratio of Bacteroidetes to Firmicutes (H) of the microbiota of mice with DIO exposed to 6 days of cold exposure. Data are presented as mean ± SEM or ±SD (graph F). n = 6. ^∗p < 0.05; ^∗∗p < 0.01. FM, fat mass; LM, lean mass. See also Figure S3.

Figure 4

Transfer of Cold-Adapted Microbiota Improves Metabolic Phenotype (A–J) Fat mass (A), adiposity index (B), and changes in mRNA levels of thermogenic genes (C) and protein levels of UCP1 (D) and DIO2 (E) in iBAT and mRNA levels of thermogenic genes in ING (F) in GF mice colonized with microbiota harvested from mice kept at 12°C or 29°C. Glucose tolerance test (G). BA profile of recipient mice (H). pAMPK (I) and pACC (J) protein levels in liver and hepatic Cpt1a (K) expression of recipient mice. The data are given as mean ± SEM. n = 4–5 per group. ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001. FM, fat mass; LM, lean mass. See also Figure S4.