Leucine deprivation results in antidepressant effects via GCN2 in AgRP neurons

Abstract

Essential amino acids (EAAs) are crucial nutrients, whose levels change in rodents and patients with depression. However, how the levels of a single EAA affects depressive behaviors remains elusive. Here, we demonstrate that although deprivation of the EAA leucine has no effect in unstressed mice, it remarkably reverses the depression-like behaviors induced by chronic restraint stress (CRS). This beneficial effect is independent of feeding and is applicable to the dietary deficiency of other EAAs. Furthermore, the effect of leucine deprivation is suppressed by central injection of leucine or mimicked by central injection of leucinol. Moreover, hypothalamic agouti-related peptide (AgRP) neural activity changes during CRS and leucine deprivation, and chemogenetically inhibiting AgRP neurons eliminates the antidepressant effects of leucine deprivation. Finally, the leucine deprivation-regulated behavioral effects are mediated by amino acid sensor general control non-derepressible 2 (GCN2) in AgRP neurons. Taken together, our results suggest a new drug target and/or dietary intervention for the reduction of depressive symptoms.

Keywords: AgRP neurons; GCN2; amino acid sensing; depression; leucine deprivation.

Figure 1

Short-term leucine deprivation does not cause depression-like behavior. (a) Timeline of the mice fed with control (Cont) or leucine-deficient [(-) Leu] diet and behavioral tests. (b) Representative tracks of mice in OFT, travel distance, percentage of distance in center area, and percentage of time spent in center. (c) Representative tracks of mice in EPM, the number of entries into the open arms, and the time spent in the open arms. (d and e) Immobility time of TST and FST, respectively. Studies were conducted using 8–9-week-old male WT mice fed a Cont or (-) Leu diet for 4−7 days. Data are expressed as the mean ± SEM (n = 14−16 per group, as indicated), with individual data points. Data were analyzed via two-tailed unpaired Student’s t-test.

Figure 2

Leucine deprivation relieves CRS-induced depressive behaviors. (a) Schematic diagram illustrating the timeline of the restraint or control protocol and behavioral tests. (b) Representative tracks of mice in the OFT, travel distance, percentage of distance in center area, and percentage of time spent in center. (c) Representative tracks of mice in the EPM, the number of entries into the open arms, and the time spent in the open arms. (d and e) Immobility time in the TST and FST, respectively. Studies were conducted using 8–9-week-old male WT mice with or without CRS fed a Cont or (-) Leu diet for 4–7 days. Data are expressed as the mean ± SEM (n = 8 per group, as indicated), with individual data points. Data were analyzed via one-way ANOVA followed by the Student–Newman–Keuls (SNK) test. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 3

Effects of ICV injection of leucine and leucinol on depressive behaviors. (a and e) Schematic diagram illustrating the ICV injection of leucine or leucinol in mice with CRS fed a Cont or a (-) Leu diet. (b and f) Representative tracks of mice in OFT, travel distance, percentage of distance in center area, and percentage of time spent in center. (c and g) Representative tracks of mice in the EPM, the number of entries into the open arms, and the time spent in the open arms. (d and h) Immobility time of the TST. Studies for a–d were conducted using 7–8-week-old female WT mice with CRS fed a (-) Leu diet with ICV of saline (− ICV Leu) or leucine (+ ICV Leu) for 4–7 days; studies for e–h were conducted using 7–8-week-old female WT mice with CRS fed a Cont diet with ICV of saline (− ICV Leucinol) or leucinol (+ ICV Leucinol) for 4−7 days. Data are expressed as the mean ± SEM (n = 8–9 per group, as indicated), with individual data points. Data were analyzed via two-tailed unpaired Student’s t-test. *P < 0.05, **P < 0.01.

Figure 4

Inhibition of AgRP neural activity blocks leucine deprivation-induced antidepressant effects. (a) IF staining for tdTomato (red), c-Fos (green), and merge (yellow) in ARC (left), and quantification of c-Fos and tdTomato colocalized cell numbers (right); 3V, third ventricle. (b) IF staining for mCherry (red), c-Fos (green), and merge (yellow) in ARC (left), and quantification of c-Fos and mCherry colocalized cell numbers (right). (c) Representative tracks of mice in the OFT, travel distance, percentage of distance in center area, and percentage of time spent in center. (d) Representative tracks of mice in the EPM, the number of entries into the open arms, and the time spent in the open arms. (e) Immobility time in the TST. Studies for a were conducted using 8–9-week-old female AgRP-Ai9 mice with or without CRS fed a Cont or (-) Leu diet for 4–7 days; studies for b–e were conducted using 8–16-week-old male AgRP-Cre mice receiving AAVs expressing mCherry (− hM4Di) or hM4Di (+ hM4Di) under CRS and fed a Cont or (-) Leu diet for 4–7 days. Data are expressed as the mean ± SEM (n = 6–11 per group, as indicated), with individual data points. Data were analyzed via one-way ANOVA followed by the SNK test for a, or via two-tailed unpaired Student’s t-test for b, or via two-way ANOVA for c–e. *P < 0.05, **P < 0.01, ****P < 0.0001.

Figure 5

Knockdown of GCN2 in AgRP neurons disrupts leucine deprivation-induced antidepressant effects. (a) IF staining for tdTomato (red), p-GCN2 (green), and merge (yellow) in ARC (left), and quantification of p-GCN2 and tdTomato colocalized cell numbers (right); 3V, third ventricle. (b) Representative tracks of mice in the OFT, travel distance, percentage of distance in center area, and percentage of time spent in center. (c) Representative tracks of mice in the EPM, the number of entries into the open arms, and the time spent in the open arms. (d) Immobility time in the TST. Studies for a were conducted using 8–9-week-old female AgRP-Ai9 mice with or without CRS fed a Cont or (-) Leu diet for 4–7 days; studies for b–d were conducted using 10–12-week-old female AgRP-Cre mice receiving AAVs expressing GFP (− shGCN2) or shGCN2 (+ shGCN2) with CRS fed a (-) Leu diet for 4–7 days. Data are expressed as the mean ± SEM (n = 6–10 per group, as indicated), with individual data points. Data were analyzed via one-way ANOVA followed by the SNK test for a, or via two-way ANOVA for b–d. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6

Activation of AgRP neurons reverses the effects of GCN2 knockdown on depressive behaviors. (a) IF staining for GFP (green), c-Fos (red), and merge (yellow) in the ARC (left), and quantification of c-Fos and GFP colocalized cell numbers (right); 3V, third ventricle. (b) Representative tracks of mice in the OFT, travel distance, percentage of distance in center area, and percentage of time spent in center. (c) Representative tracks of mice in the EPM, the number of entries into the open arms, and the time spent in the open arms. (d) Immobility time in the TST. (e) Summary diagram illustrating that deprivation of the EAA leucine exhibits antidepressant effects under chronic stress by stimulating GCN2 and neural activity in AgRP neurons. Studies for a were conducted using 8–9-week-old female AgRP-Cre mice receiving AAV expressing GFP (− shGCN2) or shGCN2 (+ shGCN2) with CRS fed a (-) Leu diet for 4–7 days; studies for b–d were conducted using 25–30-week-old male AgRP-Cre mice receiving AAVs expressing GFP (− shGCN2) or shGCN2 (+ shGCN2) and AAVs expressing mCherry (− hM3Dq) or hM3Dq (+ hM3Dq) with CRS fed a (-) Leu diet for 4–7 days. Data are expressed as the mean ± SEM (n = 6–11 per group, as indicated), with individual data points. Data were analyzed via the two-tailed unpaired Student’s t-test for a, or via two-way ANOVA for b–d. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.