Thermoneutral Housing Accelerates Metabolic Inflammation to Potentiate Atherosclerosis but Not Insulin Resistance

Abstract

Chronic, low-grade inflammation triggered by excess intake of dietary lipids has been proposed to contribute to the pathogenesis of metabolic disorders, such as obesity, insulin resistance, type 2 diabetes, and atherosclerosis. Although considerable evidence supports a causal association between inflammation and metabolic diseases, most tests of this link have been performed in cold-stressed mice that are housed below their thermoneutral zone. We report here that thermoneutral housing of mice has a profound effect on the development of metabolic inflammation, insulin resistance, and atherosclerosis. Mice housed at thermoneutrality develop metabolic inflammation in adipose tissue and in the vasculature at an accelerated rate. Unexpectedly, this increased inflammatory response contributes to the progression of atherosclerosis but not insulin resistance. These findings not only suggest that metabolic inflammation can be uncoupled from obesity-associated insulin resistance, but also point to how thermal stress might limit our ability to faithfully model human diseases in mice.

Figure 1. Thermoneutral housing accelerates onset of metabolic inflammation in adipose tissue

Kinetics of infiltration of epididymal WAT by immune cells in C57BL/6J mice fed normal chow (NC) or high fat diet (HFD) and housed at T_a of 22°C or 30°C (n=4–5 per temperature/diet/time point). (A, B) Total numbers of CD45⁺ hematopoietic cells at 22°C (A) and 30°C (B). (C, D) Macrophages at 22°C (C) and 30°C (D). (E and F) CD11c⁺ CD206⁻ macrophages at 22°C (E) and 30°C (F). (G and H) CD169⁺ macrophages at 22°C (G) and 30°C (H). (I and J) CD11c⁻ CD206⁺ cells at 22°C (I) and 30°C (J). (K and L) CD301⁺ cells at 22°C (K) and 30°C (L). Data are represented as mean ± SEM.

Figure 2. Increase in metabolic inflammation does not potentiate insulin resistance in thermoneutral mice

(A, B) Quantification of infiltrating neutrophils (A) and monocytes (B) in the model of zymosan-induced peritonitis in C57BL/6J mice housed at T_a 22°C or 30°C (n=10 per group). (C) Quantification of monocytes following zymosan-induced peritonitis in WT and Ccr2⁻⁻ mice housed at T_a of 30°C (n=4–5 per genotype). (D) Changes in body mass of C57BL/6J mice housed at T_a of 22°C or 30°C and fed normal chow (NC) or high fat diet (HFD) starting at 6 weeks of age. (E–G) Mass of adipose tissues of C57BL/6J mice housed at T_a of 22°C or 30°C and fed NC or HFD; eWAT (E), scWAT (F) and BAT (G). (H–L) Assessment of glucose homoeostasis in C57BL/6J mice housed at T_a of 22°C or 30°C and fed NC or HFD. Glucose (H) and insulin (I) tolerance tests were performed 9 and 10 weeks after initiation of HFD, respectively (n=5 per temperature and diet). Serum insulin (J) and insulin-stimulated AKT signaling in eWAT (K) and liver (L) in C57BL/6J mice housed at T_a of 22°C or 30°C and fed HFD. Data are represented as mean ± SEM.

Figure 3. CCR2 contributes to adipose tissue inflammation but not insulin resistance in thermoneutral mice

(A) Changes in total body mass of WT and Ccr2⁻⁻ fed HFD for 15 weeks at T_a of 30°C. (B) Adipose tissue mass of WT and Ccr2⁻⁻ fed a HFD for 15 weeks at 30°C. (C–H) Quantification of macrophages (C), CD11c⁺ cells (D), monocytes (E), Ly6C^hi monocytes (F), Ly6C^lo monocytes (G) and FoxP3⁺ (H) in scWAT and eWAT of thermoneutral WT and Ccr2⁻⁻ mice fed HFD. (I–M) Assessment of glucose homoeostasis in thermoneutral WT and Ccr2⁻⁻ mice fed HFD. Serum insulin (I), glucose (J) and insulin (J) tolerance tests, and insulin-stimulated AKT signaling in eWAT (L) and liver (M). Data are represented as mean ± SEM, n=4–5 mice per genotype.

Figure 4. Thermoneutral housing exacerbates atherosclerosis

(A, B) Quantification of en face atherosclerotic lesions in Apoe⁻⁻ mice fed normal chow (NC) or western (WD) for 16 weeks, and housed at T_a of 22°C or 30°C (n=17–20 per temperature and diet). (C) En face aorta preparations from Apoe⁻⁻ mice housed at T_a of 22°C or 30°C and fed NC or WD were stained with Sudan IV. Representative images are shown. (D) Representative sections of aortic lesions from WD fed Apoe⁻⁻ mice housed at T_a of 22°C (left panels) and 30°C (right panels) were stained with Oil Red O for neutral lipids (top), CD68 for macrophages (middle), and smooth muscle actin (SMA) for smooth muscle cells (bottom). (E, F) Quantification of serum cholesterol (E) and triglycerides (F) Apoe⁻⁻ mice fed NC or WD at the two different ambient temperatures (n=5–19 per diet and temperature). (G) Body mass of Apoe⁻⁻ mice housed at T_a of 22°C or 30°C that were fed NC or WD for 16 weeks (n=14–15 per group). Data are represented as mean ± SEM.

Figure 5. Increased vascular wall inflammation in thermoneutral mice

(A–K) Quantification of immune cells present in aortas of Apoe⁻⁻ mice fed WD and housed at T_a of 22°C or 30°C (n=6 per temperature). Macrophages (A), Ly6C^hi macrophages (B), Ly6C^mid macrophages (C), Ly6C^lo macrophages (D), resident macrophages (E), CD301 MFI in macrophages (F), CD301 MFI in Ly6C^lo macrophages (G), CD103⁻CD11b⁺ DCs (H), CD103⁺CD11b⁻ DCs (I), CD301 MFI in CD103⁻CD11b⁺ DCs (J), and CD301 MFI in CD103⁺CD11b⁻ DCs (K), (n=10–12 per group for A–E and H, I; n=5–6 per group for F, G, J, K). (L and M) Quantitative RT-PCR analysis of mRNA markers of M1 (L) and M2 (M) macrophages in aortas of Apoe⁻⁻ mice fed the WD. Data are represented as mean ± SEM.

Figure 6. Thermoneutral housing increases inflammation in thoracic perivascular fat

(A) Thoracic perivascular fat mass of Apoe⁻⁻ mice fed WD at 22°C and 30°C. (B, C) Gross histology (B) and hematoxylin eosin (C) staining of thoracic perivascular fat sections from Apoe⁻⁻ mice fed WD at 22°C and 30°C. (D–F) Quantitative RT-PCR analysis of brown fat (D), M1 inflammation (E), and M2 genes (F) in the thoracic perivascular fat of Apoe⁻⁻ mice fed WD at 22°C and 30°C. (G to I) Flow cytometric quantification of macrophages (G), Ly6c^hi macrophages (H), and Ly6C^lo macrophages (I) in perivascular fat of Apoe⁻⁻ mice fed WD at 22°C and 30°C. (J) CD301 median fluorescence intensity (MFI) of thoracic perivascular fat macrophages from Apoe⁻⁻ mice fed WD at 22°C and 30°C. (K) BrdU incorporation in thoracic perivascular fat macrophages of Apoe⁻⁻ mice fed WD at 22°C and 30°C. Data are represented as mean ± SEM, n=6 per temperature.

Figure 7

Model for differential effects of metabolic inflammation in thermoneutral mice.