Selective inactivation of Socs3 in SF1 neurons improves glucose homeostasis without affecting body weight

Abstract

Suppressor of cytokine signaling 3 (Socs3) has been identified as a mediator of central leptin resistance, but the identity of specific neurons in which Socs3 acts to suppress leptin signaling remains elusive. The ventromedial hypothalamus (VMH) was recently shown to be an important site for leptin action because deleting leptin receptor within VMH neurons causes obesity. To examine the role of VMH Socs3 in leptin resistance and energy homeostasis, we generated mice lacking Socs3 specifically in neurons positive for steroidogenic factor 1 (SF1), which is expressed abundantly in the VMH. These mice had increased phosphorylation of signal transducer and activator of transcription-3 in VMH neurons, suggesting improved leptin signaling, and consistently, food intake and weight-reducing effects of exogenous leptin were enhanced. Furthermore, on either chow or high-fat diets, these mice had reduced food intake. Unexpectedly, energy expenditure was reduced as well. Mice lacking Socs3 in SF1 neurons, despite no change in body weight, had improved glucose homeostasis and were partially protected from hyperglycemia and hyperinsulinemia induced by high-fat diets. These results suggest that Socs3 in SF1 neurons negatively regulates leptin signaling and plays important roles in mediating leptin sensitivity, glucose homeostasis, and energy expenditure.

Figure 1

Reduced Socs3 expression and enhanced Stat3 phosphorylation in the VMH of Socs3^flox/flox, Sf1-Cre mice. A, Socs3 mRNA expression in the VMH/DMH (n = 3 per group). B, Average numbers of leptin-induced phosphorylated Stat3 (P-Stat3)-positive neurons within the VMH per section (n = 3 per group). Data are represented as mean ± sem. P-Stat3 immunohistochemistry from representative saline-injected Socs3^flox/flox (C), Sf1-Cre and Socs3^flox/flox (D) mice. P-Stat3 immunohistochemistry from representative leptin-injected Socs3^flox/flox, Sf1-Cre (E) and Socs3^flox/flox (F) mice. G, SF1 immunohistochemistry for Socs3^flox/flox mice. Arrows indicate VMH regions.

Figure 2

Enhanced weight-reducing and food intake-reducing effects of exogenous leptin in Socs3^flox/flox, Sf1-Cre mice. A, Body weight change of 20-wk-old Socs3^flox/flox, Sf1-Cre mice and Socs3^flox/flox mice after leptin infusion. B, Cumulative food intake of 20-wk-old Socs3^flox/flox, Sf1-Cre mice and Socs3^flox/flox mice after leptin infusion. Wild-type C57B6 mice were infused with saline as a control. Mouse numbers were six and six for Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice, respectively. Data are expressed as mean ± sem. *, P < 0.05.

Figure 3

Reduced food intake, blood glucose levels, leptin levels, and improved glucose homeostasis in Socs3^flox/flox, Sf1-Cre mice. Body weight (A) and average weekly food intake (B) of Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice are shown. Fed and fasting blood glucose levels at wk 10 (C), leptin levels at wk 12 (D), and insulin levels at wk 12 (E) for Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice are shown. F, Glucose tolerance test at wk 13 (1.5 mg glucose per gram body weight). G, Insulin tolerance test at wk 15 (1 mU human Insulin R per gram body weight; n = 6 and 7 for Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice, respectively). Data are expressed as mean ± sem. *, P < 0.05; **, P < 0.01.

Figure 4

On HFD, Socs3^flox/flox, Sf1-Cre mice had reduced food intake, blood glucose levels, insulin levels, and improved glucose homeostasis. Body weight (A) and average weekly food intake (B) of Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice are shown. Fed and fasting blood glucose levels at wk 10 (C), leptin levels at wk 12 (D), and insulin levels at wk12 (E) for Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice are shown. F, Glucose tolerance test at wk 13 (1.5 mg glucose per gram body weight). G, ITT at wk 16 (1 mU human Insulin R per gram body weight; n = 6 and 9 for Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice, respectively). Data are expressed as mean ± sem. *, P <0.05.

Figure 5

Reduced energy expenditure of Socs3^flox/flox, Sf1-Cre mice on chow and HFD. Oxygen consumption (A), heat production (B), RER (VO₂ to VCO₂) (C), and locomotor activity (D), of mice on chow diets are shown (n = 6 and 7 for Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice, respectively). Oxygen consumption (E), heat production (F), RER (VO₂ to VCO₂) (G), and locomotor activity (H) of mice on HFD are shown (n = 6 and 9 for Socs3^flox/flox, Sf1-Cre and Socs3^flox/flox mice, respectively). Data are expressed as mean ± sem. **, P <0.01.