Background The activation of the hypothalamic–pituitary–adrenal (HPA) axis is essential for metabolic adaptation in response to fasting. However, the neurocircuitry connecting changes in the peripheral energy stores to the activity of hypothalamic paraventricular corticotrophin-releasing factor (CRFPVN) neurons, the master controller of the HPA axis activity, is not completely understood. Our main goal was to determine if hypothalamic arcuate nucleus (ARC) POMC and AgRP neurons can communicate fasting-induced changes in peripheral energy stores, associated to a fall in plasma leptin levels, to CRFPVN neurons to modulate the HPA axis activity in mice. Results We observed increased plasma corticosterone levels associate with increased CRFPVN mRNA expression and increased CRFPVN neuronal activity in 36 h fasted mice. These responses were associated with a fall in plasma leptin levels and changes in the mRNA expression of Agrp and Pomc in the ARC. Fasting-induced decrease in plasma leptin partially modulated these responses through a change in the activity of ARC neurons. The chemogenetic activation of POMCARC by DREADDs did not affect fasting-induced activation of the HPA axis. DREADDs inhibition of AgRPARC neurons reduced the content of CRFPVN and increased its accumulation in the median eminence but had no effect on corticosterone secretion induced by fasting. Conclusion Our data indicate that AgRPARC neurons are part of the neurocircuitry involved in the coupling of PVNCRF activity to changes in peripheral energy stores induced by prolonged fasting.

No abstract

POMC neurons expressed in the ARC are essential for energy balance and glucose homeostasis. It has been suggested the involvement of these neurons in the control of endocrine axes, such as the HPA. During fasting, POMCARC neurons are silenced as an effort to reduce body weight loss and to avoid hypoglycemia. During this process glucocorticoid secretion and activation of enzymes involved in the hepatic gluconeogenesis take place in order to preserve the homeostasis. In this study, to clarify the contribution of POMCARC neurons to the adaptive changes in energy homeostasis, glucose metabolism and HPA axis activity induced by food deprivation we used DREADDs to specifically activate POMCARC. Bilateral injections of the AAV carrying the excitatory DREADD (hM3DGq) or only the reporter gene (mCherry) have been performed into the ARC of Pomc-ires-cre and WT mice. Two weeks later the animals were fasted for 36hr, treated with saline (5 i.p. injections each 8hrs) and blood samples were collected from the facial vein at 10am. Two weeks apart, the same animals were submitted to another period of fasting and treated with CNO (1mg/Kg, 5 i.p. injections each 8hrs). Four hours after the last injection of CNO, the mice were anesthetized, blood and the liver were collected and then the animals perfused for brain harvesting. Body weight measurements have been performed before and after the 36hrs period of fasting. Another set of Pomc-ires-cre (hM3DGq or mCherry) and WT animals were fasted (36hrs), treated with CNO (5X) and subjected to GTT. DREADD–induced activation of POMCARC neurons has been confirmed by the increased cFos/mCherry expression after CNO treatment only in Pomc-ires-cre animals expressing hM3DGq. We observed that the specific activation of POMCARC neurons did not change the fasting-induced activation of HPA axis. Surprisingly, we observed reduced body weight loss and higher plasma glucose in Pomc-ires-cre animals expressing the hM3DGq and treated with CNO. The GTT showed an impaired glucose tolerance after activation of POMCARC neurons. The increased fasting glucose plasma levels was associated with increased G6pc (Glucose-6-phosphatase) mRNA expression but with no effect on other hepatic gluconeogenic genes. The present study reveals that POMCARC neurons are not involved in the increased HPA axis activity in prolonged fasting conditions. Considering the classical anorexigenic/thermogenic and the glucose-lowering action of POMCARC neurons, the present data reveal an unpredicted reduced body weight loss and impaired glucose tolerance induced by activation of these neurons during fasting. These data reinforce the notion that POMCARC neurons are heterogeneous and might be playing dual effects on energy homeostasis. Of note, because part of ARC neurons shares a common progenitor, some of the functions ascribed to POMC neurons could be mediated by non-POMC neurons expressing the Cre transgene.

No abstract