AgRP neurons trigger long-term potentiation and facilitate food seeking

Abstract

Sufficient feeding is essential for animals' survival, which requires a cognitive capability to facilitate food seeking, but the neurobiological processes regulating food seeking are not fully understood. Here we show that stimulation of agouti-related peptide-expressing (AgRP) neurons triggers a long-term depression (LTD) of spontaneous excitatory post-synaptic current (sEPSC) in adjacent pro-opiomelanocortin (POMC) neurons and in most of their distant synaptic targets, including neurons in the paraventricular nucleus of the thalamus (PVT). The AgRP-induced sEPCS LTD can be enhanced by fasting but blunted by satiety signals, e.g. leptin and insulin. Mice subjected to food-seeking tasks develop similar neural plasticity in AgRP-innervated PVT neurons. Further, ablation of the majority of AgRP neurons, or only a subset of AgRP neurons that project to the PVT, impairs animals' ability to associate spatial and contextual cues with food availability during food seeking. A similar impairment can be also induced by optogenetic inhibition of the AgRP→PVT projections. Together, these results indicate that the AgRP→PVT circuit is necessary for food seeking.

Fig. 1. Stimulation of an AgRP neuron triggers sEPSC LTD in its downstream POMC neuron.

A Representative microscopic images showing GFP-labelled AgRP neurons (left), tdTOMATO-labelled POMC neurons (middle), and merge (right) from NPY-GFP/POMC-CreERT2/Rosa26-LSL-tdTOMATO mice. B, C Representative action potential traces from an AgRP neuron and a POMC neuron that do not form a synapse (B) or from a pair that forms a synapse (C). D, E Representative traces for IPSCs recorded from a POMC neuron upon a positive current injected into its pre-synaptic AgRP neuron, in the absence (D) or the presence (E) of 4-AP and TTX. F, G Quantifications of IPSC amplitude (F) and latency (G). Data are mean ± SEM. n = 7 per group. ***P < 0.001 vs. 1st group; ###P < 0.001 vs. 2nd group in one-way ANOVA analyses followed by Tukey’s tests. H Schematic protocol to induce sEPSC LTD at the AgRP→POMC synapse. I Upper panel: representative sEPSC traces at the baseline and after AgRP neuron stimulation recorded from the POMC neuron that was synapsed by the AgRP neuron. Lower panel: quantitative changes in sEPSC amplitudes (as % of the baseline) in POMC neurons before and after the 10-min AgRP neuron stimulation. Data are mean ± SEM. n = 6 per group. J Upper panel: representative sEPSC traces at the baseline and after AgRP neuron stimulation recorded from the POMC neuron that was not synapsed by the AgRP neuron. Lower panel: quantitative changes in sEPSC amplitudes (as % of the baseline) in POMC neurons before and after the 10-min AgRP neuron stimulation. Data are mean ± SEM. n = 6 per group. Also see Fig. S1.

Fig. 2. Characterization of the sEPSC LTD in AgRP-innervated POMC neurons.

A Temporal changes in sEPSC amplitudes in AgRP-innervated POMC neurons from mice that were fed ad libitum or fasted for 24 h. B The average sEPSC amplitude after AgRP stimulation in (A). Data are mean ± SEM. n = 6 per group. ***P < 0.001 in two-tailed unpaired t tests. C Temporal changes in sEPSC amplitudes in AgRP-innervated POMC neurons (from fasted mice) after the brain slices were pre-incubated with vehicle, leptin, or insulin. D The average sEPSC amplitude after AgRP stimulation in (C). Data are mean ± SEM. n = 5 or 6 per group. ***P < 0.001 vs. the no treatment group in one-way ANOVA analyses followed by post hoc Tukey’s tests. E, F Amplitudes of NMDA (E) and AMPA (F) sEPSC in POMC neurons before (baseline) and after (LTD) the 10-min AgRP neuron stimulation. Data are mean ± SEM. n = 6 per group. **P < 0.01 and ***P < 0.001 in two-tailed paired t tests. G Temporal changes in sEPSC amplitudes in POMC neurons before and after the 10-min AgRP neuron stimulation (from fasted mice) after the brain slices were pre-incubated with various blockers. H The average sEPSC amplitude after AgRP stimulation in (G). Data are mean ± SEM. n = 6 or 7 per group. **P < 0.01 and ***P < 0.001 vs. the no blocker group in one-way ANOVA analyses followed by post hoc Tukey’s tests. Also see Fig. S1.

Fig. 3. AgRP neurons trigger sEPSC LTDs in distant synaptic targets.

A Schematic illustration to identify distant AgRP-originated synapses. B, D, F, H, J, L Temporal changes in sEPSC amplitudes before and after the 10-min photostimulation in AgRP-innervated PVH (B), PBN (D), PVT (F), BNST (H), MeA (J), CeA (L) neurons from mice that were fed ad libitum or fasted for 24 h. C, E, G, I, K The average sEPSC amplitude after AgRP stimulation in AgRP-innervated neurons in (B, D, F, H, J), respectively. Data are mean ± SEM. n = 4, 5, 6, or 7 per group. *P < 0.05, **P < 0.01, and ***P < 0.001 vs. fed group in two-tailed unpaired t tests. Also see Figs. S2 and S3.

Fig. 4. Loss of AgRP neurons impairs food seeking.

A A photo of the Y-maze. B Number of entries to food-coupled arm by control and AgRP^neoAblation mice during pre-test and test in the Y-maze assay. n = 14 or 15 per group. **P < 0.01 between pre-test and test in two-way ANOVA repeated measurement followed by post-hoc-Sidak tests. Note that the number of data points may appear fewer than the n values because multiple mice showed the same number of entries during the pre-test or test. C Schematic presentation of the hole-board with food baits. D, E Incorrect (D) and correct pokes (E) of control and AgRP^neoAblation mice in the food-baited hole board test. Data are mean ± SEM. n = 16 or 17 per group. *P < 0.05 vs. control in two-tailed unpaired t tests. Also see Fig. S4.

Fig. 5. Y-maze conditioning induced neural plasticity in AgRP-innervated PVT neurons.

A, C, E NMDA sEPSC in the AgRP-innervated neurons in the CeA (A), BNST (C), and PVT (E) from naive mice or Y-maze-trained mice. B, D, F AMPA sEPSC in the AgRP-innervated neurons in the CeA (B), BNST (D), and PVT (F) from naive mice or Y-maze-trained mice. Data are mean ± SEM. n = 5 or 10 per group. ***P < 0.001 vs. naive group in two-tailed unpaired t tests. G, H Firing frequency (G) and resting membrane potential (H) of AgRP neurons from naive mice or Y-maze-trained mice. Data are mean ± SEM. n = 12 per group. Also see Fig. S5.

Fig. 6. The AgRP→ PVT circuit regulates food seeking.

A Schematic illustration to selectively ablate AgRP neurons projecting to the PVT. B Number of entries to food-coupled arm by control and AgRP^PVT-Ablation mice during pre-test and the test in the Y-maze assay. n = 7 or 12 per group. *P < 0.05 between pre-test and the test in two-way ANOVA repeated measurement followed by post hoc Sidak tests. Note that the number of data points may appear fewer than the n values because multiple mice showed the same number of entries during the pre-test or the test. C, D Incorrect (C) and correct pokes (D) of control and AgRP^PVT-Ablation mice in the food-baited hole board test. Data are mean ± SEM. n = 12 or 15 per group. *P < 0.05 vs. control in one-tailed unpaired t tests. E Schematic illustration to selectively inhibit AgRP→PVT projections. F Number of entries to the food-coupled arm in the pre-test and the test in the Y-maze assay by control mice and mice with inhibited AgRP→PVT projections. n = 10 per group. *P < 0.05 between the pre-test and the test in two-way ANOVA repeated measurement followed by post hoc Sidak tests. Note that the number of data points may appear fewer than the n values in because multiple mice showed the same number of entries during the pre-test or the test. G Schematic illustration to selectively activate AgRP→PVT projections. H Number of entries to the food-coupled arm in the pre-test and the test in the Y-maze assay by control mice and mice with activated AgRP→PVT projections. n = 10 per group. Note that the number of data points may appear fewer than the n values in because multiple mice showed the same number of entries during the pre-test or the test. Also see Figs. S6, S7.