Hypothalamic eIF2α signaling regulates food intake

Abstract

The reversible phosphorylation of the α subunit of eukaryotic initiation factor 2 (eIF2α) is a highly conserved signal implicated in the cellular adaptation to numerous stresses such as the one caused by amino acid limitation. In response to dietary amino acid deficiency, the brain-specific activation of the eIF2α kinase GCN2 leads to food intake inhibition. We report here that GCN2 is rapidly activated in the mediobasal hypothalamus (MBH) after consumption of a leucine-deficient diet. Furthermore, knockdown of GCN2 in this particular area shows that MBH GCN2 activity controls the onset of the aversive response. Importantly, pharmacological experiments demonstrate that the sole phosphorylation of eIF2α in the MBH is sufficient to regulate food intake. eIF2α signaling being at the crossroad of stress pathways activated in several pathological states, our study indicates that hypothalamic eIF2α phosphorylation could play a critical role in the onset of anorexia associated with certain diseases.

Figure 1. A ΔLeu Meal Activates GCN2 in the MBH

(A) Expression of GCN2 mRNA was detected in the MBH by ISH on mice brain sections. Brains from GCN2 knockout (GCN2^−/−) mice were used as a negative control. Scale bar, 100 μm. (B) Western blot analysis showed a GCN2-dependent increase in MBH phospho-eIF2α level in response to a ΔLeu meal, as compared to a balanced meal. After overnight starvation, wild-type (GCN2^+/+) and GCN2^−/− mice were fed either with a control (Ctr) or a ΔLeu diet for 40 min, then animals were sacrificed for analysis. Four male mice were used per group. Signal intensity was quantified using ImageJ software. Results are given as mean ± SEM. The statistical difference between groups was assessed by a two-way ANOVA. **p < 0.01. (C) IHC showed phospho-eIF2α labeling in the ARC. A representative mouse for each group (four male mice) is shown. Scale bar, 100 μm. (D) Western blot analysis showed increased ATF4 protein expression in the ARC following a ΔLeu meal. After overnight starvation, mice were fed either with a Ctr or a ΔLeu diet for 4 hr, then animals were sacrificed for analysis. Nuclear protein extracts were prepared from pooled ARC tissues of four mice per group. Histone H3 served as a loading Ctr. See also Figures S1 and S2.

Figure 2. GCN2 Knockdown in the ARC Markedly Blunts the Aversive Response to a ΔLeu Meal

Lentivectors encoding GCN2-specific shRNA or scramble sequence were delivered bilaterally into the ARC of wild-type male mice (see Figure S3 for technical details). (A) GCN2-shRNA delivery in the ARC resulted in a loss of the aversive response to a ΔLeu meal (*p < 0.05, unpaired Student’s t test; six to seven males per group). The relative consumption of a ΔLeu versus Ctr diet during a 1 hr meal was expressed as the ratio of consumption (Δ% ± SEM) of the ΔLeu diet compared to the consumption of the Ctr diet by the same animal. We verified that GCN2 knockdown did not affect the intake of Ctr diet (see Table S1). At the end of the experiment, mice were sacrificed, then the ARC was dissected to extract total RNA. (B) GCN2-shRNA lentiviral delivery in the ARC resulted in a 40% decrease of the mean level of GCN2 mRNA expression in the ARC. Results are given as mean ± SEM (***p < 0.001, unpaired Student’s t test; six to seven males per group). (C) The strength of association between levels of food intake inhibition and GCN2 expression in the ARC was analyzed by the Pearson correlation test. See also Figure S3 and Table S1.

Figure 3. GCN2 Activation in the MBH Is Sufficient to Inhibit Food Intake

A permanent cannula had been first placed into the third ventricle of wild-type (n = 10) or GCN2^−/− (n = 12) male mice. After overnight starvation, mice were injected either with 1 μl L-leucinol (10 mM in 0.9% NaCl) or vehicle just before giving the diet (standard chow). (A) L-leucinol administration into the third ventricle induced a strong inhibition of food intake in wild-type (GCN2^+/+) mice, whereas it had no effect in GCN2^−/− mice. Results are given as Δ% ± SEM of food intake level after L-leucinol injection to food intake level after vehicle injection in the same animal. The statistical difference between groups was assessed by a two-way ANOVA: effect of L-leucinol, *p < 0.05 and **p < 0.01; effect of genotype, #p < 0.05 and §p < 0.01. We verified that GCN2 knockout did not affect the intake of control diet (see Table S1). (B) Representative IHC analysis of phosho-eIF2α (blue labeling) shows that L-leucinol administration (1 hr treatment) resulted in GCN2 activation in the MBH of the GCN2^+/+ mice. Following IHC, sections were stained with hematoxylin (pink labeling). We checked that L-leucinol injection in the third ventricle did not induce eIF2α phosphorylation in a distal site, such as the APC. L-leucinol treatment did not increase phospho-eIF2α labeling in the MBH of GCN2^−/− mice. (C) Western blot analysis showed increased ATF4 protein expression in the ARC following L-leucinol administration into the third ventricle. After overnight starvation, mice were treated for 2 hr with L-leucinol, and ARC tissues were harvested. Both cytoplasmic and nuclear protein extracts were prepared from pooled ARC tissues of four mice per group. See also Table S1.

Figure 4. Increasing Phospho-eIF2α Level in the MBH Leads to Food Intake Inhibition

A permanent cannula had been placed into the third ventricle of wild-type male mice (n = 10). After a 6 hr starvation period, mice were injected either with 2.5 μl salubrinal (100 μM in 0.5% DMSO/0.9% NaCl) or vehicle. Meal (standard chow) was given 2 hr after injection. (A) Salubrinal administration into the third ventricle resulted in a marked food intake inhibition as measured in the first 3 hr after giving diet. Results are given as Δ% ± SEM of food intake level after salubrinal injection to food intake level after vehicle injection in the same animal. Paired Student’s t tests were performed to evaluate the significance of salubrinal effect on food intake: **p ≤ 0.01. (B) Western blot analysis showed that, 2 hr after salubrinal injection in the third ventricle, eIF2α phosphorylation was increased in the ARC, but not in the APC. Signal intensity was quantified using ImageJ software. Results are given as mean ± SEM. (C) Western blot analysis showed increased ATF4 protein expression in the ARC following salubrinal administration into the third ventricle. After a 6 hr starvation period, mice were treated for 2 hr, and ARC tissues were harvested. Nuclear protein extracts were prepared from pools of ARC tissues of four mice per group.