Functional brown adipose tissue limits cardiomyocyte injury and adverse remodeling in catecholamine-induced cardiomyopathy

Abstract

Brown adipose tissue (BAT) has well recognized thermogenic properties mediated by uncoupling protein 1 (UCP1); more recently, BAT has been demonstrated to modulate cardiovascular risk factors. To investigate whether BAT also affects myocardial injury and remodeling, UCP1-deficient (UCP1(-/-)) mice, which have dysfunctional BAT, were subjected to catecholamine-induced cardiomyopathy. At baseline, there were no differences in echocardiographic parameters, plasma cardiac troponin I (cTnI) or myocardial fibrosis between wild-type (WT) and UCP1(-/-) mice. Isoproterenol infusion increased cTnI and myocardial fibrosis and induced left ventricular (LV) hypertrophy in both WT and UCP1(-/-) mice. UCP1(-/-) mice also demonstrated exaggerated myocardial injury, fibrosis, and adverse remodeling, as well as decreased survival. Transplantation of WT BAT to UCP1(-/-) mice prevented the isoproterenol-induced cTnI increase and improved survival, whereas UCP1(-/-) BAT transplanted to either UCP1(-/-) or WT mice had no effect on cTnI release. After 3 days of isoproterenol treatment, phosphorylated AKT and ERK were lower in the LV's of UCP1(-/-) mice than in those of WT mice. Activation of BAT was also noted in a model of chronic ischemic cardiomyopathy, and was correlated to LV dysfunction. Deficiency in UCP1, and accompanying BAT dysfunction, increases cardiomyocyte injury and adverse LV remodeling, and decreases survival in a mouse model of catecholamine-induced cardiomyopathy. Myocardial injury and decreased survival are rescued by transplantation of functional BAT to UCP1(-/-) mice, suggesting a systemic cardioprotective role of functional BAT. BAT is also activated in chronic ischemic cardiomyopathy.

Keywords: Brown adipose tissue; Cardioprotection; Heart failure; Isoproterenol; Uncoupling protein 1.

Fig. 1

Flow chart of experimental procedures.

Fig. 2

Effect of 14 days of isoproterenol infusion on echocardiographic and pathologic parameters in male WT and UCP1^−/− mice. The UCP1^−/− mice developed more LV hypertrophy and fibrosis than WT mice and decreased LV fractional shortening after isoproterenol. Panel A: Posterior wall thickness (PWT) measured with echocardiography. Panel B: Echocardiographically-derived thickness/radius ratio (H/R). Panel C: Ratio of LV mass measured at necropsy to bodyweight (LVW/BW). Panel D: Echocardiographically-derived fractional shortening (FS). (Echocardiographic data acquired in WT SAL N = 6, WT ISO N = 8, KO SAL N = 11, KO ISO N = 8) Panels E, F, G, H, I: Myocardial fibrosis detected by PicroSirius staining. Representative slices in a WT mouse (E) and UCP1^−/− mouse (G) after 14 days of saline infusion and in a WT mouse (F) and UCP1^−/− mouse (H) after 14 days of isoproterenol infusion. Panel I: Quantification of myocardial fibrosis in WT and UCP1^−/− mice, after 14 days of saline or isoproterenol infusion (WT SAL N = 6, WT ISO N = 8, KO SAL N = 5, KO ISO N = 5). *: p < 0.05; **: p < 0.01; ***: p < 0.001 vs. sham (saline infusion) of the same strain at 14 days; $$: p < 0.01; $$$: p < 0.001 vs. baseline (D0) in the same strain; #: p < 0.05; ###: p < 0.001 vs. isoproterenol in WT mice. WT: wild-type; KO: UCP1^−/− mice; D0: baseline; D14: 14 days of isoproterenol; SAL: saline infusion; ISO: isoproterenol infusion.

Fig. 3

Markers of myocardial injury in WT and UCP1^−/− mice at 3 days and survival of WT and UCP1^−/− mice with saline or isoproterenol infusion after 14 days. Panel A–C: Cardiac troponin I (cTnI, Panel A), fatty acid binding protein 3 (FABP3, Panel B), myosin light chain 3 (Myl3, Panel C) were measured in EDTA plasma after 3 days of isoproterenol. Both cTnI and Myl3 were higher in WT and UCP1^−/− mice, however, the increase was greater in UCP1^−/−mice. FABP3 only increased in UCP1^−/− mice. (WT SAL N = 12, WT ISO N = 12, KO SAL N = 11, KO ISO N = 10). Panel D: Kaplan–Meier plot of survival during 14 days of follow-up in WT and UCP1^−/− mice after saline or isoproterenol infusion (WT SAL N = 25, WT ISO N = 35, KO SAL N = 25, KO ISO N = 35). **: p < 0.01; ****: p < 0.0001 vs. saline infusion of the same strain; ##: p < 0.01; ###: p < 0.001 vs. isoproterenol in WT mice. WT: wild-type; KO: UCP1^−/− mice; SAL: saline infusion; ISO: isoproterenol infusion.

Fig. 4

Gene expression levels of uncoupling proteins 1 (Panel A), 2 (Panel B) and 3 (Panel C) in the brown adipose tissue and left ventricle of WT and UCP1^−/− mice, after 3 days of saline or isoproterenol infusion. Panel A: UCP1 expression increased in WT BAT after isoproterenol and was not detected in the LV of WT mice. Panel B–C: No compensatory increase in the expression levels of UCP2 and 3 in BAT was noted in UCP1^−/− mice. Panel B: Levels of UCP2 in the LV were similar between WT and UCP1^−/− mice at baseline but in UCP1^−/− mice UCP2 in the LV was increased by isoproterenol. Panel C: The changes in UCP3 in the LV induced by isoproterenol were similar between WT and UCP1^−/− mice. (N = 6 for all groups). *: p < 0.05; **: p < 0.01; ***: p < 0.001 vs. saline infusion of the same strain at 3 days; #: p < 0.05; ###: p < 0.001 vs. same condition (saline or isoproterenol) in WT mice. WT: wild-type; KO: UCP1^−/− mice; SAL: saline infusion; ISO: isoproterenol infusion; BAT: brown adipose tissue; LV: left ventricle.

Fig. 5

Body weight and metabolic parameters in WT and UCP1^−/− mice. Panel A: Body weight of WT and UCP1^−/− mice (WT N = 12, KO N = 14). Panel B–D: Plasma levels of glucose (Panel B) (WTN = 12, KON = 14), insulin (Panel C) (WTN = 8, KO N = 6) and triglycerides (Panel D) (WTN = 10, KON = 8) in mice fasted for 16 h were similar in the WT and UCP1^−/− mice. Panel E–F: Glucose tolerance test (WTN = 12, KO N = 14). (Panel E) and insulin tolerance test (WT N = 8, KO N = 6). (Panel F) in WT and UCP1^−/− mice. The UCP1^−/− mice were glucose-intolerant, however, were not insulin resistant compared to WT mice. **: p < 0.01; ***: p < 0.001 vs. WT at the same time point. TG: triglycerides; WT: wild-type; KO: UCP1^−/− mice; GTT: glucose tolerance test; ITT: insulin tolerance test.

Fig. 6

Effects of the transplantation of brown adipose tissue (BAT) on the glucose tolerance (Panel A), and plasma cardiac troponin I levels (Panel B) and survival (Panel C) after isoproterenol infusion. Panel A–B: The transplantation of WTBAT attenuated the glucose intolerance in UCP1^−/− mice 8 weeks post-transplantation (Panel A) (UCP1^−/− BAT to WTN = 10, WT BAT to UCP1^−/− N = 10, UCP1^−/− BAT to UCP1^−/− N = 10) and prevented the increased plasma cardiac troponin I level detected 3 days after isoproterenol infusion (Panel B) (UCP1^−/− BAT to WT: SAL N = 3, ISO N = 6; WT BAT to UCP1^−/−: SAL N = 3, ISO N = 3; UCP1^−/− BAT to UCP1^−/− SAL N = 3, ISO N = 3). Panel C: Mortality was higher in UCP1^−/− mice transplanted with UCP1^−/− BAT after 14 days of isoproterenol infusion. (SAL groups N = 5, ISO groups N = 6) *: p < 0.05; **: p < 0.01; ***: p < 0.001 vs. baseline of the same group; #: p < 0.05; ###: p < 0.001 vs. WT to KO at the same time point; $$: p < 0.01 vs. KO to WT at the same time point; ††: p < 0.01; †††: p < 0.001 vs. KO to WT at the same time point: wild-type; KO: UCP1^−/− mice; SAL: saline infusion; ISO: isoproterenol infusion; KO to WT: KO BAT to WT mouse; WT to KO: WT BAT to KO mouse; KO to KO: KO BAT to KO mouse. WT: wild-type; KO: UCP1^−/− mice.

Fig. 7

Effect of isoproterenol infusion on the phosphorylation of AKT and ERK. Panel A–B: Representative immunoblots and quantification of P-AKT on Ser473 and total AKT (Panel A) and P-ERK1/2 on Thr202/Tyr204 and Thr185/Tyr187, respectively, and total ERK 1/2 (Panel B) on LV homogenates of WT and KO mice after 3 days of saline or isoproterenol infusion. Densitometry is presented as an average of 3 technical repeats, with 4 biological repeats per group. *: p < 0.05; **: p < 0.01 vs. saline of the same group; #: p < 0.05 vs. isoproterenol in WT mice. WT: wild-type; KO: UCP1^−/− mice.

Fig. 8

Activation of brown adipose tissue (BAT) in heart failure. Panel A: Activation of BAT, assessed by the gene expression level of UCP1, was noted 8 weeks after myocardial infarction and was correlated to the degree of LV dysfunction. Gene expression levels of UCP1 in WT mice, either sham-operated or subjected to left anterior descending coronary artery ligation (SHAM N = 11; MI N = 13). Panel B: Correlation between UCP1 gene expression levels and LV ejection fraction. The straight line represents the correlation, the dashed line the 95% confidence interval of the correlation. MI: myocardial infarction, LVEF: left ventricular ejection fraction. *: p < 0.05 vs. SHAM.