Defining the Transcriptional Targets of Leptin Reveals a Role for Atf3 in Leptin Action

Abstract

Leptin acts via its receptor (LepRb) to modulate gene expression in hypothalamic LepRb-expressing neurons, thereby controlling energy balance and glucose homeostasis. Despite the importance of the control of gene expression in hypothalamic LepRb neurons for leptin action, the transcriptional targets of LepRb signaling have remained undefined because LepRb cells contribute a small fraction to the aggregate transcriptome of the brain regions in which they reside. We thus employed translating ribosome affinity purification followed by RNA sequencing to isolate and analyze mRNA from the hypothalamic LepRb neurons of wild-type or leptin-deficient (Lep^ob/ob) mice treated with vehicle or exogenous leptin. Although the expression of most of the genes encoding the neuropeptides commonly considered to represent the main targets of leptin action were altered only following chronic leptin deprivation, our analysis revealed other transcripts that were coordinately regulated by leptin under multiple treatment conditions. Among these, acute leptin treatment increased expression of the transcription factor Atf3 in LepRb neurons. Furthermore, ablation of Atf3 from LepRb neurons (Atf3^LepRbKO mice) decreased leptin efficacy and promoted positive energy balance in mice. Thus, this analysis revealed the gene targets of leptin action, including Atf3, which represents a cellular mediator of leptin action.

Figure 1

TRAP-seq analysis to identify leptin-regulated genes in hypothalamic LepRb neurons. A: Experimental design: LepRb^eGFP-L10a mice were subjected to a variety of perturbations in leptin levels before hypothalami were dissected and mRNA-containing ribosomes from LepRb neurons were TRAP purified. TRAP and supernatant mRNAs were subjected to RNA-seq; comparisons of TRAP and supernatant fractions revealed the identity of genes enriched in LepRb neurons under each condition, and comparing TRAP material among conditions revealed the regulation of these enriched genes by leptin. B: A heat map demonstrating changes in gene expression in LepRb neuron-derived TRAP samples across conditions. Fold change scale is from 4.1-fold increased (green) to –9.7-fold decreased (red). Euclidean distance among the conditions (maximum 50) is shown along the top. hr, hour; SUP, supernatant.

Figure 2

ATF3 is induced specifically in LepRb neurons following acute leptin stimulation. LepRb^eGFP-L10a mice were treated i.p. with vehicle or leptin (5 mg/kg i.p.) for 6 h. Brains were harvested, sectioned, and stained for ATF3 and/or GFP. Images are representative of 3–4 mice per treatment group. A and B: IHC detection of ATF3-IR (black nuclei) in the hypothalamus of control (A) and leptin-treated (B) mice. C–F: Dual IHC/immunofluorescence for ATF3 and eGFP in the DMH (C and D) and ARC (E and F) of control (C and E) and leptin-treated (D and F) LepRb^eGFP-L10a mice. White arrowheads indicate colocalized neurons. Scale bars = 100 μmol/L. cDMH, caudal DMH; VMH, ventromedial hypothalamus.

Figure 3

Ablation of ATF3, but not pSTAT3, induction by leptin in Atf3^LepRbKO mice. A: We crossed Lepr^cre mice onto the Atf3^flox background to generate Lepr^cre/cre;Atf3^flox/flox (Atf3^LepRbKO) mice and littermate controls. B and C: Control (B) and Atf3^LepRbKO (C) mice were fasted for 4 h prior to treatment with leptin (5 mg/kg i.p.) followed by perfusion 6 h later and processing for the IHC detection of ATF3 (black nuclei). D and E: Control (D) and Atf3^LepRbKO (E) mice were fasted overnight before treatment with leptin (5 mg/kg i.p.) 1 h before perfusion and processing for the IHC detection of pSTAT3 (black nuclei). B–E: Digital zoom of the boxed area is shown in the upper right corner.

Figure 4

Control of energy balance by leptin in Atf3^LepRbKO mice. We examined parameters of energy balance in control and Atf3^LepRbKO male (A–D, I) and female (E–H) mice. Longitudinal measurements of body weight (A and E) and food intake (B and F) are shown, along with measurements of adipose mass (C and G) and leptin (F and H). I: Male control and Atf3^LepRbKO mice were fasted for 24 h and then treated with vehicle (PBS) or leptin (5 mg/kg i.p.) before the return of food at the onset of the dark cycle. Food intake over the subsequent 24 h is shown. *P < 0.05 by unpaired t test.