VMAT2-Mediated Neurotransmission from Midbrain Leptin Receptor Neurons in Feeding Regulation

A subset of VTA LepR neurons are glutamatergic or GABAergic. Immunostaining of pSTAT3 was performed on brain sections of mice that received i.c.v. injection of leptin. These mice also received stereotaxic injections to the VTA of AAV-FLEX-GFP vectors to allow labeling of Cre-expressing neurons with GFP. Expression of GFP (A, D) and p-STAT3 (B, E) was found in the VTA of Vgat-Ires-Cre mice (A–C) and Vglut2-IRES-Cre mice (D–F). p-STAT3 positive neurons were found in a subset of GFP-expressing neurons in both Vgat-Ires-Cre mice (C, arrows) and Vglut2-IRES-Cre mice (F, arrows). IP, interpeduncular nucleus; ml, medial lemniscus. Scale bars, 25 μm.

DA accumulation in vesicles was specifically disrupted from midbrain LepR neurons in LIC::Vmat2^fl°^x/fl°^x mice. In the VTA, majority of LepR neurons (red) were costained by DA (green) in LIC::VMAT2^+/+::Ai9 mice (A, C). However, no DA immunoreactivity was observed in VTA LepR neurons of LIC::VMAT^fl°^x/fl°^x::Ai9 mice (B–D). Arrows pointed neurons with strong DA immunoreactivity. IP, interpeduncular nucleus; ml, medial lemniscus; SNC, SN compact. Scale bars, 100 μm (A, C) and 25 μm (B, D).

Leptin inhibits midbrain LepR neurons. Electrophysiological recording were performed on LepR neurons identified by tdTomato (red) expression in VTA brain sections of LIC::Ai9 mice. In total, 16 out of 34 neurons responded to leptin. Representative images showing VTA brain sections with one LepR neuron identified for recording (A, arrow), immunostaining of injected biocytin during recording (B, blue), immunostaining of TH (C, green) and the merged (D). A representative recording trace showing changes in actual firing frequency (E) and the histogram of firing frequency (F) in response to leptin and washout. G, Summary of firing frequency changes in response to leptin of all neurons that responded to leptin. Scale bars, 100 μm.

Effects of VMAT2 deletion in VTA LepR neurons on synaptic neurotransmitter release to the amygdala. Recordings of oEPCSc and oIPSCs were made in amydalar neurons elicited by blue laser stimulation of ChR2-expressing fibers from VTA LepR neurons in brain slices. A–A’, Representative pictures showing ChR2-expressing fibers (green) in the central amygdala in low (A) and high magnification (A’). Representative recording traces of light evoked oEPSCs and oIPSCs from brain slices of control (B) and LIC::Vmat2^fl°^x/fl°^x (KO) mice (C). Responses of oEPSCs and oIPSCs to TTX/4-AP were also shown in controls (B). D, Quantitative analyses illustrate the latency for light-evoked oIPSCs (red) and oEPSCs (green) to reach 10% peak and 100% peak of the evoked currents from the onset of laser stimulation. E, Mean amplitudes of oIPSC (red) and oEPSC (green) in controls and LIC::Vmat2^fl°^x/fl°^x mice. Numbers in columns D, E showed frequency of successful detection of indicated currents in all recorded neurons. BLA, basolateral amygdalar nucleus; opt, optic tract. Scale bars, 100 μm (A) and 50 μm (A’). Data were presented as mean ± SEM; *p < 0.05, ***p < 0.001, unpaired Student’s t tests.

Effects of VMAT2 deletion in VTA LepR neurons on synaptic neurotransmitter release to the accumbens. Recordings of oEPCSc and oIPSCs were made in accumbens neurons elicited by blue laser stimulation of ChR2-expressing fibers from VTA LepR neurons in brain slices. Representative pictures showing ChR2-expressing fibers in the accumbens in low (A) and high magnification (A’). Representative recording traces of light evoked oEPSCs and oIPSCs from brain slices of control (B) and LIC::Vmat2^fl°^x/fl°^x (KO) mice (C). D, Representative traces showing responses of light-evoked oEPSCs and oIPSCs to TTX/4-AP in controls. E, Mean amplitudes of light evoked oIPSC (red) and oEPSC (green) in controls and LIC::Vmat2^fl°^x/fl°^x mice; numbers in the bar showed frequency of successful detection of the indicated currents in recorded neurons. Data were presented as mean ± SEM; ac, anterior commissure; LV, lateral ventricle. Scale bars, 100 μm (A) and 50 μm (A’).

VMAT2 deletion in midbrain LepR neurons led to resistance to diet-induced obesity. Littermate mice were fed chow or HFD after weaning and weekly body weight and food intake were measured. Weekly body weight was measured in mice fed regular chow in males (A, n = 5–12) and females (B, n = 5–11), or HFD in males (C, n = 13–25) and females (D, n = 12–16). E, F, Body composition measurements in males (E, n = 12) and females (F, n = 12) fed with HFD at 18 weeks old. G, H, Energy expenditure assessed by O₂ consumption (G, n = 6) and locomotion assessed by beam breaks (H, n = 6) measured by CLAMS in control and LIC::Vmat2^fl°^x/fl°^x mice fed HFD diet at the age of 8–10 weeks. I, Measurements of accumulated HFD food intake at ages of 8–12 weeks (n = 6–7). Data were presented as mean ± SEM; *p < 0.05, **p < 0.005 versus control mice, unpaired Student’s t tests on each measurement time point.

Acute activation of midbrain LepR neurons increased food intake. LepR-Ires-Cre and KO mice received stereotaxic injections of AAV-DIO-hM3D(Gq)-mCherry vectors to the VTA and food intake was measured four weeks after the injection. A–C, Expression pattern of AAV-DIO-h3MD(Gq)-mCherry vectors (A), c-Fos (B), and their colocalization (C) in the VTA of LepR-Ires-Cre mice treated with saline. D–F, Expression pattern of AAV-DIO-hM3D(Gq)-mCherry vectors (D), c-Fos (D), and their colocalization (F, arrows) in the VTA of LepR-Ires-Cre mice treated with CNO. G, Food intake measured in early morning during the indicated time periods from mice fed chow treated with saline or CNO. H, Food intake measured in early morning during the indicated time periods from mice fed HFD treated with saline or CNO. Scale bars, 50 μm; data were presented as mean ± SEM; n = 5–6 (F, G), two-way ANOVA tests.