The role of pituitary adenylate cyclase-activating polypeptide neurons in the hypothalamic ventromedial nucleus and the cognate PAC1 receptor in the regulation of hedonic feeding

Abstract

Obesity is a health malady that affects mental, physical, and social health. Pathology includes chronic imbalance between energy intake and expenditure, likely facilitated by dysregulation of the mesolimbic dopamine (DA) pathway. We explored the role of pituitary adenylate cyclase-activating polypeptide (PACAP) neurons in the hypothalamic ventromedial nucleus (VMN) and the PACAP-selective (PAC1) receptor in regulating hedonic feeding. We hypothesized that VMN PACAP neurons would inhibit reward-encoding mesolimbic (A10) dopamine neurons via PAC1 receptor activation and thereby suppress impulsive consumption brought on by intermittent exposure to highly palatable food. Visualized whole-cell patch clamp recordings coupled with in vivo behavioral experiments were utilized in wildtype, PACAP-cre, TH-cre, and TH-cre/PAC1 receptor-floxed mice. We found that bath application of PACAP directly inhibited preidentified A₁₀ dopamine neurons in the ventral tegmental area (VTA) from TH-cre mice. This inhibitory action was abrogated by the selective knockdown of the PAC1 receptor in A₁₀ dopamine neurons. PACAP delivered directly into the VTA decreases binge feeding accompanied by reduced meal size and duration in TH-cre mice. These effects are negated by PAC1 receptor knockdown in A₁₀ dopamine neurons. Additionally, apoptotic ablation of VMN PACAP neurons increased binge consumption in both lean and obese, male and female PACAP-cre mice relative to wildtype controls. These findings demonstrate that VMN PACAP neurons blunt impulsive, binge feeding behavior by activating PAC1 receptors to inhibit A₁₀ dopamine neurons. As such, they impart impactful insight into potential treatment strategies for conditions such as obesity and food addiction.

Keywords: A10 dopamine neurons; PAC1 receptor; food addiction; hedonic feeding; hypothalamic ventromedial nucleus; pituitary adenylate cyclase-activating polypeptide; ventral tegmental area.

Figure 1

Validation of PAC1R knockdown in TH-cre/PAC1R^fl/fl mice. Photomicrographs on the left represent TH immunostaining (visualized with Alexa Fluor 488) in sections from TH-cre (A) and TH-cre/PAC1R^fl/fl (D) mice. Photomicrographs in the middle illustrate PAC1R immunostaining (visualized with Alexa Fluor 546) in the same sections from TH-cre (B) and TH-cre/ PAC1R^fl/fl (E) mice. The photomicrographs in panels (C,F) denote merged images depicting retention of PAC1R in A₁₀ DA neurons (denoted by the arrows) from TH-cre but not TH-cre/PAC1R^fl/fl mice. (G) Percent of cells positive for colocalization of TH and PAC1R in TH-cre and TH-cre/PAC1R^fl/fl mice. Mann–Whitney U-test, ^*p < 0.05, TH-cre n = 4, TH-cre/PAC1R^fl/fl n = 4. Bars represent means and lines 1 SEM.

Figure 2

Lack of PAC1R attenuates PACAP induced outward current in VTA A₁₀ DA neurons in TH-cre/PAC1R^fl/fl mice. (A) Low powered (4X) photomicrograph of eYFP-fluorescing A₁₀ DA neurons localized to the VTA. (B) High powered (40X) photomicrograph illustrating a representative A₁₀ DA neuron about to undergo electrophysiologic recording. (C) Corresponding DIC image of the recorded TH neuron. (D) Representative outward current elicited in A₁₀ DA neurons in TH-cre mice upon bath application of PACAP_1-38. (E) I/V plot showing a −90 mV reversal potential corresponding with the Nernst equilibrium potential for potassium conductance. (F) Representative voltage clamp trace depicting no significant change in membrane current in A₁₀ DA neurons recorded from TH-cre/PAC1R^fl/fl animals. (G) I/V relationship showing no significant change in slope conductance compared to baseline. (H,I) Composite data depicting slope conductance and membrane current are lower in TH-cre/PAC1R^fl/fl animals following bath application of PACAP. Student’s t-test, ^*p < 0.05, TH-cre n = 6 animals, 9 slices, 9 cells, TH-cre/PAC1R^fl/fl n = 5 animals, 5 slices, 9 cells. Bars represent means and lines 1 SEM. (J,K) Pie charts reflecting the distribution of A₁₀ dopamine neurons from TH-cre and TH-cre/PAC1Rfl/fl animals that are inhibited, excited or unresponsive to PACAP.

Figure 3

Intra-VTA PACAP_1-38 significantly reduces binge intake, meal frequency, and bout duration in TH-cre but not TH-cre/PAC1R^fl/fl mice. (A) Schematic illustrating the procedural timeline for the execution of these experiments. (B–D) Composite data depicting PACAP_1–38 significantly decreases binge intake (B), meal frequency (C), and bout duration (D) in TH-cre mice, effects which are largely negated in the absence of PAC1R in VTA TH neurons. ^*p < 0.05 relative to vehicle, ^#p < 0.5 relative to genotype. Repeated measures, multifactorial ANOVA/LSD, TH-cre vehicle n = 10, TH-cre PACAP_1–38 n = 9, TH-cre/PAC1R^fl/fl vehicle n = 30, and TH-cre/PAC1R^fl/fl n = 26. Bars represent means and lines 1 SEM.

Figure 4

VMN PACAP neuron ablation in PACAP-cre mice increases binge consumption in males and females. (A) Schematic illustrating the procedural timeline for the execution of these experiments. (B,C) Composite data depicting caspase ablation of VMN PACAP neurons in PACAP-cre mice elicits a significant increase in binge consumption in both lean and obese male (B) as well as female (C) mice. ^*p < 0.05 relative to wildtype controls. Repeated measures, multifactorial ANOVA/LSD, (B) lean wildtype n = 30, lean PACAP-cre n = 30, DIO wildtype n = 10, DIO PACAP-cre n = 10; (C) lean wildtype n = 10, lean PACAP-cre n = 10, DIO wildtype n = 10, DIO PACAP-cre n = 10. Bars represent means and lines 1 SEM.