Role of AgRP neurons in regulating branched-chain amino acids and its implication on glucose homeostasis
Background The hypothalamus is considered as one of the major homeostatic regulatory centers in the brain and is known to play an essential role in controlling energy balance and nutrient metabolism. Agouti-related protein (AgRP)-expressing neurons residing in the hypothalamus are shown to disrupt insulin sensitivity and glucose homeostasis. However, the underlying mechanism is not well understood. Branched-chain amino acids (BCAAs) are essential amino acids as we need to obtain through diet. However, recent studies reveal that circulating BCAAs are elevated in obesity and diabetes. Chronic BCAA supplementation leads to abnormal blood glucose levels and insulin resistance, whereas BCAA restriction reverses this abnormality. Evidence suggests that the dysfunctional breakdown of BCAAs in insulin-sensitive tissues may contribute to elevated circulating BCAAs and their metabolites in obesity. In parallel with hypothalamic insulin resistance in obese individuals, animal studies have shown that a short-term high-fat diet feeding and activation of AgRP neurons lead to whole-body insulin resistance and failure to suppress circulating glucose and BCAAs by insulin, pointing towards the involvement of activated AgRP neurons in faulty BCAA catabolism, impaired glucose homeostasis, and insulin sensitivity. Our lab has recently demonstrated that AgRP neurons can raise plasma BCAA levels and suppress their breakdown in the liver. To determine if this is causally associated with the ability of these neurons to impair glucose metabolism, in this study, we examined the acute effects of BCAAs on glycemic control and systemic insulin sensitivity and tested if defective glucose homeostasis by AgRP neurons requires BCAAs. Methods First, we determined the acute effect of elevated BCAAs on blood glucose by infusing BCAAs or saline in the jugular vein catheter of C57BL/6 mice. Using catheters enabled us to frequently sample blood and measure glucose in stress-free, consciously moving mice. Next, we performed hyperinsulinemic-euglycemic clamps, a gold-standard method to assess whole-body insulin sensitivity, with a continuous infusion of saline or BCAAs in mice to sustain elevated BCAAs. Next, we used BT2, a small molecule that stimulates BCAA catabolism and lowers BCAAs, and tested if the acute reduction in circulating BCAAs improves glucose tolerance and insulin sensitivity in diet-induced obese (DIO) mice. Next, we tested if acute chemogenetic stimulation of AgRP neurons impairs glycemic control through BCAAs. We performed a bilateral AAV-mediated delivery of hM3D gene construct (stimulatory DREADD) in the arcuate nucleus of the hypothalamus of male AgRP-Ires-Cre or C57BL/6 (WT) mice. After recovery, these mice underwent a glucose tolerance test (GTT), and we tested if BT2-mediated reduction in BCAA levels improves glycemic control in the mice with stimulated AgRP neurons. Next, we performed an insulin tolerance test (ITT) to test if BT2-induced lowering of BCAAs improves insulin sensitivity in mice with stimulated AgRP neurons. Finally, we tested if AgRP neuronal inhibition lowers the BCAAs and improves glucose tolerance in chow-fed lean or DIO mice. We performed a bilateral AAV-mediated delivery of the hM4Di gene construct (inhibitory DREADD) in the arcuate nucleus of the hypothalamus of lean chow-fed or 6-week HFD-fed WT and AgRP-Ires-Cre mice. Three weeks post-recovery, these mice underwent overnight fasting followed by AgRP neuronal inhibition using GTT or ITT. Results BCAA, but not saline, infusion acutely raised blood glucose starting at t=30 min and ended with a 20% increase compared to baseline in male mice but not in female mice. The rise in blood glucose occurred despite elevated plasma insulin, suggesting possible insulin resistance. To examine this further in male mice, we performed hyperinsulinemic-euglycemic clamps, a gold-standard method to assess whole-body insulin sensitivity. Mice infused with BCAAs required a higher glucose infusion rate to maintain euglycemia, indicating reduced insulin sensitivity. Next, we used a pharmacological approach to reduce BCAAs in circulation acutely by inducing its catabolism. We observed that a single injection of BT2 before the glucose tolerance test (GTT) improved glucose tolerance in diet-induced obese (DIO) male mice, whereas chow-fed lean male and DIO female mice remained unaffected. In our next step, we assessed the effect of stimulation of AgRP neurons on glucose tolerance, insulin sensitivity, and systemic BCAA levels. Stimulation of AgRP neurons increased BCAAs and significantly impaired glucose tolerance. Interestingly, BT2 injection in mice with stimulated AgRP neurons prevented BCAA rise and markedly improved glucose tolerance. Next, we also observed that AgRP neuronal activation impaired insulin sensitivity, but BT2 failed to improve insulin sensitivity in the mice with stimulated AgRP neurons. Next, we tested if AgRP neuronal inhibition lowers the BCAAs and improves glucose tolerance in chow-fed lean or DIO mice. We did not observe any improvement in glucose tolerance, insulin sensitivity, and reduction in systemic BCAA levels after inhibiting AgRP neurons. Conclusions Collectively, these findings suggest that acute changes in circulating BCAAs can disturb glucose homeostasis and insulin action. Impaired glycemic control by AgRP neurons can be reversed by lowering circulating BCAAs, suggesting a critical role of BCAAs in glucose homeostasis. Our study sheds light on a novel function of BCAAs in the brain control of glucose metabolism, and if confirmed in humans, it may facilitate the development of strategies to maintain low BCAAs to mitigate glucose impairment and associated metabolic disorders such as obesity and diabetes.