Defective autophagy in Sf1 neurons perturbs the metabolic response to fasting and causes mitochondrial dysfunction

Abstract

Objective: The ventromedial nucleus of the hypothalamus (VMH) is a critical component of the forebrain pathways that regulate energy homeostasis. It also plays an important role in the metabolic response to fasting. However, the mechanisms contributing to these physiological processes remain elusive. Autophagy is an evolutionarily conserved mechanism that maintains cellular homeostasis by turning over cellular components and providing nutrients to the cells during starvation. Here, we investigated the importance of the autophagy-related gene Atg7 in Sf1-expressing neurons of the VMH in control and fasted conditions.

Methods: We generated Sf1-Cre; Atg7^loxP/loxP mice and examined their metabolic and cellular response to fasting.

Results: Fasting induces autophagy in the VMH, and mice lacking Atg7 in Sf1-expressing neurons display altered leptin sensitivity and impaired energy expenditure regulation in response to fasting. Moreover, loss of Atg7 in Sf1 neurons causes alterations in the central response to fasting. Furthermore, alterations in mitochondria morphology and activity are observed in mutant mice.

Conclusion: Together, these data show that autophagy is nutritionally regulated in VMH neurons and that VMH autophagy participates in the control of energy homeostasis during fasting.

Keywords: Atg7; Energy balance; Fasting; Hypothalamus; Mitochondria; Ventromedial nucleus.

Graphical abstract

Figure 1

Fasting promotes autophagy in the ventromedial nucleus of the hypothalamus. (A) Representative images and (B) quantification of LC3-GFP immunofluorescence in the dorsomedial (dm) and ventrolateral (vl) parts of the ventromedial nucleus of the hypothalamus (VMH) of adult fed or 12 h-fasted mice (n = 3 per group). (C) Autophagy-related gene mRNA levels in the VMH of adult fed or 12 h-fasted mice (n = 3–5 per group). V3, third ventricle. Scale bar, 10 μm. Values are shown as mean ± SEM. ∗P ≤ 0.05 and ∗∗P ≤ 0.01 versus fed.

Figure 2

Altered energy homeostasis in mice lacking autophagy in SF1 neurons. (A) Body weight and (B) average food intake of Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP (n 4–6 per group) male mice. (C) Food intake of refed of adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP male mice after fasting (n = 5 for each group). (D) Body composition and (E) leptin sensitivity of adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP male mice (n = 5–6 for each group). (F) Serum leptin levels in fed and 12 h-fasted adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP (n = 3–5 per group) male mice. (G–I) Oxygen (O2) consumption and (J–L) carbon dioxide (CO2) production in (G, J) fed and (H, K) 12 h-fasted Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP male mice (n = 5 for each group). Values are shown as mean ± SEM. ∗P ≤ 0.05, and P ≤ 0.01 versus each group.

Figure 3

Glucose metabolism in Sf1-Cre; Atg7^loxP/loxPmice. (A) Serum glucose and (B) insulin levels and (C) VMH insulin receptor (InsR) mRNA expression in fed and 12h-fasted adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP (n = 4–7 per group) male mice. (D) Glucose and (E) insulin tolerance tests of adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP (n = 4–6 per group) male mice. Values are shown as mean ± SEM. ∗P ≤ 0.05, ∗∗P ≤ 0.01, and ∗∗P ≤ 0.001 versus each group.

Figure 4

Fasting-induced hypothalamic neuronal activation is impaired in Sf1-Cre; Atg7^loxP/loxPmice. (A–C) Representative images and quantification of cFos-immunoreactive cells in the (A) ventromedial (VMH), (B) arcuate (ARH), and (C) dorsomedial (DMH) nuclei of the hypothalamus of fed and fasted adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP male mice (n = 3–7 per group). (D) Vglut2, (E) Pomc, and (F) Npy mRNA expression in the ARH of adult fed and fasted Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP male mice (n = 3–4 per group). Values are shown as mean ± SEM. ∗P ≤ 0.05 versus each group.

Figure 5

Loss of autophagy in SF1 neurons causes altered mitochondrial morphology and function. (A) Representative electron microscopy images of VMH neurons, percentage of autophagy-mitochondria contacts, and mitochondria aspect ratio in adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP male mice (n = 4 per group). (B) Representative images and quantification of MitoTracker labeling in primary cultures of hypothalamic neurons 1 day after transfection with Atg7 siRNA or scrambled siRNA (n = 3–4 per group). (C) Mitochondria-related gene mRNA levels in the VMH of adult (10-week-old) fed or 12 h-fasted mice (n = 3–5 per group). (D) Measurement of mitochondrial respiration in the mediobasal hypothalamus of adult Atg7^loxP/loxP and Sf1-Cre; Atg7^loxP/loxP male mice (n = 4–8 per group). Values are shown as mean ± SEM. ∗P ≤ 0.05 versus each group.