New Results

Reversal of obesogenic feeding and hypometabolism by a bifurcating GABAergic neural circuit

Yong Han, Yang He, Lauren Harris, Yong Xu, Qi Wu
doi: https://doi.org/10.1101/2022.01.23.477433

Abstract

Diet-induced obesity is characterized by unsatiated consumption of energy-dense diets and impaired metabolism, whereby anti-obesity effect of the high-level of circulating leptin is unknowingly blunted. Emerging evidence suggests that the leptin receptor (LepR) signaling system, residing within the agouti-related protein (AgRP) neurons of the hypothalamus, critically contributes to obesogenic feeding, nutrient partitioning, and energy metabolism. However, the neural circuit mechanism underlying the leptin-dependent control of obesogenic feeding and energy balance remains largely elusive. Here, we show that two distinct subgroups of LepR-expressing AgRP neurons send non-collateral, GABAergic projections to the dorsomedial hypothalamic nucleus (DMH) and to the medial part of the medial preoptic nucleus (MPO) for the differential control of metabolic homeostasis and obesogenic feeding, respectively. We found that the AgRPLepR-DMH neural circuit plays a significant role in leptin-dependent control of metabolic homeostasis through the α3-containing GABAA receptor signaling on the melanocortin 4 receptor neurons within the DMH (MC4RDMH). In contrast, the AgRPLepR-MPO neural circuit elicits dominant effects on the appetitive response to high-fat diet through the α2-containing GABAA receptors on the MC4RMPO neurons. Consistent with these behavioral results, the post-synaptic GABAA neurons located within the DMH and MPO displayed differentiated firing responses under various feeding and nutrient conditions. Our results demonstrate that these novel GABAergic neural circuits exert differentiated control of metabolic hemostasis and obesogenic feeding via distinct post-synaptic targets of leptin-responsive AgRP neurons. The findings of two genetically and anatomically distinct GABAA receptor signaling pathways within the DMH and MPO would undoubtedly accelerate the development of targeted, individualized, anti-obesity therapy.

INTRODUCTION

Diet-induced obesity (DIO) drastically increases susceptibility to the development of metabolic, cardiovascular, and neurological diseases, as well as the recent coronavirus disease 1–5. Leptin, a hormone secreted by white adipocytes, interacts with the brain to communicate peripheral fuel status, suppress appetite following a meal, promote energy expenditure, and maintain blood glucose stability6–12. One of the key pathological characteristics in DIO is the impaired metabolic homeostasis13,14. There is much evidence the central leptin signaling plays a vital role in the regulation of body weight, primarily via action upon the arcuate nucleus (ARC)7,8,15–20. It has been found that the long-form leptin receptors are abundantly expressed in the AgRP neurons of the ARC18,21–24. Previous studies demonstrate that, under the physiological condition, leptin directly inhibits the LepR-expressing AgRP neurons25,26. However, these dynamic effects are blunted in obesity mice occurring within primary targeting neurons of the leptin signaling 27. On the other hand, the neural circuit and transmitter signaling mechanism underlying leptin-mediated body weight and energy balance have not yet been explored.

AgRP neurons play fundamental roles in regulating feeding behavior and body weight by releasing inhibitory NPY, AgRP and GABA transmitters into broad downstream brain areas28–30. The neural circuits comprised of AgRP neurons, the neural and hormonal signaling afferent to AgRP neurons, and their postsynaptic targets have been identified as key players in the regulation of energy balance and systemic insulin sensitivity31–36. Leptin acts to decrease food intake and promote energy expenditure by suppressing the activity of AgRP neurons25,26,37,38. Selective ablation of LepR in AgRP neurons gives rise to an obese phenotype and diabetes 18,39. Although the LepR signaling within AgRP neurons has been implicated as the dominant component for the regulation of metabolic homeostasis, the underlying neural circuit mechanism is so far poorly understood. We suggest that identification of the critical transmitter signaling components underlying the leptin-responsive neural circuit is crucial for the development of more efficient therapeutics for obesity.

Emerging data suggest a critical role of the GABA signaling on the control of feeding behavior and energy homeostasis 20,40–55. Numerous studies suggest that central GABAA and GABAB receptor signaling exert prominent influences on feeding behavior under various metabolic states 56–62. For example, pharmacological activation of GABAA receptor signaling in the hindbrain parabrachial nucleus (PBN) enhances the positive hedonic perception of tastes and foods, thereby promoting food intake and motivational response to food reward 56,57,60,63–66. We have previously shown that loss of GABA from AgRP neurons resulted in glucose intolerance51. Our recent work shows that the α5-containing GABAA receptor signaling within the bed nucleus of the stria terminalis neurons reciprocally regulates mental disorders and obesity, implicating that the GABAA receptor signaling exerts a role in controlling obesity and comorbid diseases 30.

In this report, we examined the neural circuit mechanism underlying the leptin action upon the AgRP neurons using a newly established and robust method by rapid inactivation of the LepR signaling within AgRP neurons51. We found that the LepR in the AgRP neurons plays a pivotal role in the control of obesity and metabolic homeostasis. Moreover, a subset of leptin-responsive AgRP neurons sent GABAergic projections to a group of α3-GABAA-expressing neurons in the DMH for regulation of leptin-mediated metabolic homeostasis. Another subset of LepR-expressing AgRP neurons innervated a group of α2-GABAA-expressing neurons in the MPO, which mediates leptin’s control on obesogenic feeding. Taken together, these findings suggest that the identification of key GABAA signaling pathways within two distinct post-synaptic targets of a novel leptin-responsive GABAergic neural circuit will fundamentally accelerate the development of efficient treatments for obesity.

RESULTS

To test the role of leptin receptor (LepR) in AgRP neurons, we applied our newly developed conditional knockout approach by generating two different lines of mice: AgrpnsCre/+::Leprlox/lox::Jax2356NeoR/+::Rosa26tdTomato mice, termed the knockout group (Agrp::LeprKO), and Agrp+/+::Leprlox/lox::Jax2356NeoR/+::Rosa26tdTomato mice, termed the control group51. Immunohistochemistry and qPCR results showed the expression of tdTomato and deletion of Lepr in AgRP neurons 4 days after treatment of NB124, a synthetic nonsense-suppressor that can rapidly restore the functions of Cre recombinase within the AgRP neurons (Fig.1a-f). Agrp::LeprKO mice showed a significant increase in feeding and body weight one week after deletion of LepR signaling from AgRP neurons (Fig.1g,h). Furthermore, along with the increase of body weight and food intake, Agrp::LeprKO mice exhibited impaired glucose tolerance (Fig.1i). Intriguingly, chronic infusion of bicuculline (4 ng, a GABAA receptor antagonist) into the 3rd ventricle obviously abolished the hyperphagia responses in Agrp::LeprKO mice (Fig.1j), suggesting that facilitating the post-synaptic GABAA signaling to the AgRP neural circuit is important in regulating leptin-mediated overfeeding. Meanwhile, we calculated the feeding efficiency which is presented as mg of body weight gain/kcal consumed in the mice. The feeding efficiency was significantly higher in Agrp::LeprKO mice but can be rescued by infusion of bicuculline (Fig.1k). We further showed that the expansion of white adipose tissue (WAT) accounted for the entire body weight gain in Agrp::LeprKO mice (Fig.1l). Respiratory quotient (RQ) is defined as the volume of CO2 released over the volume of O2 absorbed during respiration (defined as ratio of VCO2/VO2), which provides an indication of the nature of the substrate being used by an organism (i.e., RQ=1 for glucose utilization RQ=0.7 for lipid utilization). We observed a significant decrease in RQ from the Agrp::LeprKO mice that can be further normalized by potentiating the GABAA signaling (Fig.1m). In line with other studies, these results suggest that GABA is a crucial signaling molecule by which AgRP neurons control adiposity and nutrient utilization 47,67,68. Further, the control and Agrp::LeprKO mice were analyzed for their glucose tolerance. The results showed that ablation of LepR in the AgRP neurons impaired glucose tolerance, which can be fully restored by infusion of GABAA antagonist (Fig.1n). These results indicate that leptin regulates appetite and energy metabolism via the post-synaptic GABAA signaling within the AgRP neural circuit.

Fig. 1:
Fig. 1: Enhancing the GABAA signaling reverses obesogenic feeding and metabolic dysfunction induced by leptin deficiency in AgRP neurons.

a,b, Expression of tdTomato in the arcuate nucleus after i.c.v. injection of vehicle (control mice, a) or NB124 (Agrp::LeprKO mice, b) into the 3rd ventricle of AgrpnsCre::Leprlox/lox::Rosa26tdTomato/+::Jax2356Neo mice. Scale bar in b for a and b, 200 μm. c-e, Colocalization of tdTomato (c) and AgRP (d) in ARC. Scale bar in e for c-e, 200 μm. f, Real-time qPCR analysis of Lepr transcript levels within the AgRP neurons of the control and Agrp::LeprKO mice. (n = 8 per group; *p < 0.05). g-I, Daily calorie intake (g), body weight (h), and GTT (i) of the control and Agrp::LeprKO mice. (n = 8 per group; *p < 0.05). j, The 4-hr food intake by the control and Agrp::LeprKO mice after chronic infusion of bicuculline into the 3rd ventricle for 4 weeks. k, Feeding efficiency is presented as mg of body weight gain/kcal consumed in the mice as described in j. l, Weight of adipose tissues in the mice as described in j (Visc WAT: visceral white adipose tissue; BAT: brown adipose tissue). m, Average value for RQ (defined as ratio of VCO2/VO2) tested on Day 28 in the mice as described in j. n, The GTT was performed in the mice as described in j. (n = 8 per group in j-n; *p < 0.05, #p < 0.05, ‡p < 0.05). Error bars represent mean ± SEM. unpaired two-tailed t test in f; one-way ANOVA and followed by Tukey comparisons test in j-l; two-way ANOVA and followed by Bonferroni comparisons test in g-i, and n.

As in our initial step to identify the key downstream targets of AgRP neurons that contribute to the leptin-mediated responses, we examined the Npas4 expression profile in those neurons projected by AgRP neurons. Traditional immediate early genes are regulated by neuromodulators via cAMP, neurotrophins and other paracrine factors, and their kinetics are relatively slow. In contrast, Npas4 has a more dynamic response to activity-dependent signaling via Ca2+, which expressed in an activity-dependent manner not only in excitatory neurons but also in all inhibitory neurons69. The qPCR results showed that rapid deletion of LepR signaling from AgRP neurons resulted in a significant decrease of neural activities within the DMH and MPO (Fig.2a). Next, we devised an assay to further identify the functional relevance of each downstream target to leptin signaling. To achieve this goal, we administered DT into neonatal AgrpDTR mice to ablate all AgRP neurons and examined the neural activities of downstream targets of AgRP neurons in response to leptin (Fig.2b)70,71. Without ablation of AgRP neurons, we found that leptin induced robust Fos induction in almost all major targets of AgRP neurons47,48,72. However, neonatal ablation of AgRP neurons significantly diminished the leptin-mediated Fos induction within post-synaptic neurons residing in the MPO and DMH (Fig.2c-f). These results indicate that the DMH and MPO could be the key downstream targets that critically contribute to the regulation of leptin-mediated appetitive and metabolic responses.

Fig. 2:
Fig. 2: Identification of downstream targets of the leptin-responsive AgRP neurons.

a, Real-time qPCR analysis of transcript levels of Npas4 expressed in the MPO, DMH, PVH, PVT, and PBN in the control and Agrp::LeprKO mice. (n = 8 per group; *p < 0.05). b, The schematic diagram showing Fos activities in downstream targets of AgRP neurons after treatment of leptin combined with or without neonatal ablation of AgRP neurons. c-f, Expression profiling of Fos (green) in the MPO, DMH, PVH, PVT, DRN, and PBN under vehicle (d), leptin (i.p.) (e), leptin (i.p.) and DT (s.c. neonates) (f). Scale bar in d for d-f, 200 μm. (n = 8 per group; *p < 0.05). Error bars represent mean ± SEM; unpaired two-tailed t test in a; one-way ANOVA and followed by Tukey comparisons test in c.

To establish a role of the DMH and MPO within the leptin-responsive AgRP neural circuit, we applied a transsynaptic tracer to characterize the neurons in the DMH and MPO that are innervated by AgRP neurons (Fig.3a,b)29,30,73,74. The results derived from RNA in situ hybridization combined with transsynaptic tracing suggest that AgRP neurons make extensive connections with MC4R neurons in the DMH and MPO (Fig.3a-d and Extended Data Fig.1). Next, we performed whole-cell, patch-clamp recordings to investigate the synaptic connectivity between channelrhodopsin-2 (ChR2)-expressing AgRP axons and ZsGreen-labeled postsynaptic neurons within the DMH (Fig.3a). To test whether this connectivity was monosynaptic input from AgRP axon terminals, we perfused tetrodotoxin (TTX) and 4-aminopyridine (4-AP) into the bath to remove any network activity. We observed that inhibitory postsynaptic currents (IPSCs) in the DMH neurons triggered by photostimulation of ChR2-expressing axonal terminals of AgRP neurons were fully blocked by bicuculline (Fig.3e,f), confirming that these terminals were releasing GABA. In the presence of DNQX (a competitive AMPA/kainate receptor antagonist), AP5 (a selective NMDA receptor antagonist) and bicuculline, photostimulation of the ChR2-expressing AgRP terminals resulted in robust inhibition of action potential in postsynaptic ZsGreen neurons of the DMH in a reversible manner with significant reduction of firing rate and resting membrane potential (Fig.3g-j). We further examined the effect of leptin on the firing of postsynaptic ZsGreen neurons in the DMH. We found that leptin treatment significantly enhanced neural activities of DMH neurons (Fig.3k,l). To reveal the hierarchical structure of this neural circuit, results derived from HSV-based dual retrograde tracing showed that two distinct subgroups of AgRP neurons project to the DMH and MPO, respectively (Fig.3m-p). Taken together, these results implicated potential functional segregation of the bifurcating AgRP→DMH/MPO neural circuit.

Fig. 3:
Fig. 3: Leptin-responsive AgRP neurons directly innervate GABAergic neurons within the DMH and MPO.

a, Diagram of transsynaptic tracing downstream targets from AgRP neurons. AAV9-DIO-WGA-ZsGreen, a Cre-dependent tracer that labels neurons with synaptic connections, was injected into the ARC of AgrpCre::Ai32 mice. Ai32 mice express ChR2/EYFP fusion protein conditioned Cre-mediated recombination. b, ChR2-EYFP-labeled AgRP neurons in the ARC. Scale bar, 200 μm. c,d, Double-labeling of MC4R neurons with transsynaptic WGA-ZsGreen in the DMH (c, green; WGA-ZsGreen; red, Mc4r RNA in situ) and MPO (d, green; WGA-ZsGreen; red, Mc4r RNA in situ). Scale bars in c and d, 200 μm. e, The response of IPSCs recorded from an ZsGreen+ neuron upon photostimulation of AgRP terminals within the DMH (10 ms pulse) with a pretreatment of vehicle or TTX+4-AP, and in slices pretreated with either TTX+4-AP or TTX+4-AP+bicuculline. f, Statistical amplitude analysis of IPSCs from neurons recorded in e. (n = 3 neurons; *P < 0.05 between TTX+4-AP and TTX+4-AP+bicuculline). g, Representative spikes of ZsGreen-labeled DMH neurons before and after blue light pulses (20 ms/pulse, 20 Hz) shined onto AgRP axonal fibers. h-j, Firing frequency (h), resting membrane (i), and statistical analysis (j) of neurons recorded in g. (n < 3 neurons; *P < 0.05). k,l, Firing frequency (k), and resting membrane (l) of ZsGreen-labeled DMH neurons with or without leptin i.p. treatment. (n < 3 neurons; *P < 0.05). m, Diagram shows dual retrograde tracing of AgRPARC→DMH neurons and AgRPARC→MPO neurons by simultaneous injection of HSV-LSL-mCherry into the DMH and HSV-LSL-EYFP into the MPO of AgrpCre mice. n-p, The fluorescence in the DMH (n) and MPO (o), and the ARC (p; red, DMH-projecting AgRP neurons; green, MPO-projecting AgRP neurons). Scale bar in n for n and o, 200 μm; scale bar in p, 100 μm. Error bars represent mean ± SEM. paired two-tailed t test in h, i, k, and l; one-way ANOVA and followed by Tukey comparisons test in f.

To examine the role of the AgRP → DMH circuit in regulation of feeding and metabolism, we performed optogenetic manipulation in AgrpCre::Ai32 mice where ChR2-eYFP was selectively expressed in AgRP neurons and axonal terminals (Fig.4a)29,30. Photostimulation of AgRP fibers in the DMH promoted feeding of chow diet but not high-fat diet (HFD), coupled with disrupted glucose intolerance (Fig.4b,c). Consistently, photostimulation of post-synaptic MC4RDMH neurons resulted in a significant decrease in feeding of chow but not HFD and improved glucose intolerance (Fig.4d,e). Bilateral infusion of bicuculline into the DMH of Agrp::LeprKO mice resulted in reduced chow feeding and a significant reduction of body weight, coupled with improved glucose tolerance (Fig.4f-i). Moreover, our studies showed that blockage of GABAA receptor in the DMH significantly rescued the feeding efficiency, expansion of WAT, and the RQ profiles in Agrp::LeprKO mice (Fig.4j-l). These results suggest that GABAA receptor signaling within the AgRP→DMH circuit is critical for the regulation of chow diet intake, adiposity, and nutrient utilization.

Fig. 4:
Fig. 4: The leptin-responsive AgRP-DMH circuit regulates glucose tolerance, and nutrient partitioning.

a, Diagram showing the photostimulation of AgRP axonal terminals within the DMH of AgrpCre::Ai32 mice. b,c, 1-hr food intake of LFD and HFD (b) and GTT (c) after photostimulation of the AgRP→DMH circuit. (n = 8 per group; *p < 0.05; p = 0.11 between Control and ChR2 in HFD group). d,e, 1-hr food intake of LFD and HFD (d) and GTT (e) after photostimulation of the MC4RDMH neurons. (n = 8 per group; *p < 0.05). f-h, Daily food intake (f), body weight (g) and GTT (h) in control mice, Agrp::LeprKO mice with bicuculline or vehicle injected into the DMH. The saline or bicuculline (0.1 ul/min, 5min) was chronically infused into the DMH via the osmotic minipump. (n = 8 per group; *p < 0.05, #p < 0.05). I, 4-hrs food intake of LFD and HFD (f) in the mice described in f-h. j, Feed efficiency in the mice described in f-h. k, Weight of adipose tissues in the mice described in f-h. l, Average value for RQ (VCO2/VO2) in the mice described in f-h. (n = 8 per group in i-l; *p < 0.05, #p < 0.05). m-o, 4-hrs food intake of LFD and HFD (m), body weight (n) and GTT (o) in control mice, GABA KO mice with or without treatment of leptin (1 μg/g body weight). The AAV9-fDIO-WGA-nsCre-mCitrine virus or control AAV9-fDIO-WGA-mCitrine virus was injected into the ARC of NpyFlp::Gad1lox/lox::Gad2lox/lox::Rosa26tdTomato mice followed with injection of NB124 into the DMH. (n = 8 per group in m-o; *p < 0.05, #p < 0.05, ‡p < 0.05). Error bars represent mean ± SEM. unpaired two-tailed t test in b and d; one-way ANOVA and followed by Tukey comparisons test in b, d, f, k; two-way ANOVA and followed by Bonferroni comparisons test in c-h, n, and o.

To better understand the contribution of the AgRP→DMH circuit to the control of leptin-mediated functions, we bilaterally injected the AAV9-fDIO-WGA-nsCre virus into the ARC of NpyFlp::Gad1lox/lox::Gad2lox/lox::Rosa26tdTomato mice followed with treatment of NB124 into the DMH, a strategy which can specifically inactivate the GABA signaling from those DMH neurons directly innervated by AgRP neurons. We showed that acute ablation of GABA signaling from the AgRP-projected DMH neurons significantly increased feeding of chow diet without affecting intake of HFD (Fig.4m). Under the treatment of chow diet, these GABA signaling-deficient mice displayed a moderate but significant increase in body weight coupled with impaired glucose intolerance (Fig.4n,o). More importantly, inactivation of post-synaptic GABA signaling in the AgRP → DMH circuit blunted the actions of systemically administered leptin on chow diet feeding, body weight, and glucose tolerance (Fig.4m-o). Overall, these data demonstrate that the GABA signaling within the downstream targets of the AgRP→DMH neural circuit plays a significant role in control of leptin-mediated metabolic homeostasis.

To better understand the physiological responses of these GABAergic neurons in the context of feeding and metabolic regulation, we employed in vivo opto-tetrode system to reveal the dynamic activities of MC4RDMH neurons73. With the injection of AAV2-DIO-ChR2-GFP into the DMH of Mc4rCre mice, the ChR2 neurons can be identified by the latencies of evoked spikes accurately following high-frequency photostimulation, as well as the identical waveforms of evoked and spontaneous spikes73. We investigated the activities of MC4RDMH neurons and non-MC4RDMH neurons under low-fat diet (LFD), HFD, and hyperglycemia condition. A total 13 MC4RDMH neurons and 9 non-MC4RDMH neurons were identified through optogenetic-invoked spikes. The results showed that 7 out of 13 identified MC4R neurons responded to the feeding with LFD, with a reduction of firing rate from 22.4 Hz to 12.7 Hz (Fig.5a,c). We also observed 6 out of 13 MC4R neurons that respond to the enhancement of blood glucose with a reduction of firing rate from 21.9 Hz to 13.9 Hz (Fig.5b,d). This result indicates that the MC4RDMH neurons mediate feeding and glucose tolerance.

Fig. 5:
Fig. 5: The α3-GABAA receptor signaling within the MC4RDMH mediates glucose tolerance and nutrient partitioning.

a-d, Raster plots and firing rate of representative MC4RDMH neurons and non-MC4RDMH neurons when mice under baseline, feeding with LFD and HFD (a), and under glucose i.p. injection (b). The statistical analysis of the firing rate under baseline, feeding with LFD and HFD (c), and under glucose treatment (d) were calculated. (n = 6 or 7 in each group; *p < 0.05, #p < 0.05). e, Expression level of Gabra1-Gabra6 in the DMH neurons of control and Agrp::LeprKO mice. f, Immunostaining showing the expression of GABAA α3 in the DMH. Scale bar, 200 μm. g-I, Colocolization of GABAA α3 and transsynaptically labeled neurons by WGA-ZsGreen in the DMH after injection of AAV9-DIO-WGA-ZsGreen into the ARC of AgrpCre mice. j-l, The intake of LFD (j), body weight (k) and GTT (l) were performed in Mc4rCre::Rosa26Cas9 mice with a bilateral injection of vehicle, AAV9-DIO-Gabra3sgRNA-tdTomato (GABAA-α3 KO), or AAV9-DIO-Gabra3cDNA-tdTomato (GABAA-α3 OE) into the DMH. m, Feeding efficiency in the mice described in j-l. n, Weight of adipose tissues in the mice described in j-l. o, Average value for RQ (VCO2 /VO2) in the mice described in j-l. (n = 8 per group in j-o; *p < 0.05, #p < 0.05). Error bars represent mean ± SEM. unpaired two-tailed t test in d and e; one-way ANOVA and followed by Tukey comparisons test in c, m, and n; two-way ANOVA and followed by Bonferroni comparisons test in j-l.

Our results showed that GABAA receptors in the DMH are involved in the regulation of feeding and nutrient partitioning; thus, we attempted to identify the key GABAA receptor subunits that are functionally relevant to glucose and feeding regulation. Results of qPCR analysis showed that, among all major regulatory α subunits, the transcript level of Gabra3 (encoding GABAA receptor α3 subunits) in the DMH neurons of Agrp::LeprKO mice was robustly enhanced (Fig.5e). Immunostaining data showed the α3-GABAA signaling was abundantly expressed within the DMH (Fig.5f). Our transsynaptic tracing study further confirmed that the α3-GABAA signaling was highly co-localized within the post-synaptic targets of the AgRP→DMH neural circuit (Fig.5g-i). To understand the functional roles of the α3-GABAA signaling with the downstream DMH neurons in relevancy to leptin-mediated feeding and metabolism, Mc4rCre::Rosa26Cas9 mice were injected with AAV9-Gabra3sgRNA-mCherry into DMH. We found that knockout of Gabra3 signaling in the MC4RDMH neurons reduces feeding of chow diet and body weight while the glucose tolerance was significantly improved (Fig.5j-l and Extended Data Fig.2). Meanwhile, genetic deficiency in α3-GABAA signaling significantly decreased feeding efficiency and WAT (Fig.5m-o). On the other hand, the gain-of-function study showed that overexpression of α3-GABAA within the same MC4RDMH neurons manifested the exact opposite phenotypes, including moderately increased chow diet feeding and body weight, significantly increased feeding efficiency and WAT adiposity, and profoundly decreased RQ (Fig.5m-o). These results suggest that the MC4RDMH neurons play a significant role in control of glucose tolerance, nutrient partitioning, and feeding efficiency through the α3-GABAA signaling.

Next, we characterized the functional role of the AgRP→MPO circuit in leptin-mediated feeding and metabolism. Contrasting to the role of the AgRP→DMH circuit, photostimulation of AgRP fibers within the MPO led to a significant increase in feeding of HFD without affecting the normal response to LFD or glucose intolerance (Fig.6a, and fig. S3A). Similarly, photostimulation of post-synaptic MC4RMPO neurons specifically suppressed intake of HFD but not LFD while glucose tolerance remained intact (Fig. 6B and Extended Data Fig.3b). Further, by employing a similar genetic strategy described in Fig.4, M-O, we showed that acute ablation of GABA signaling from the post-synaptic MC4RMPO neurons within the AgRP→MPO circuit significantly increased HFD feeding while chow diet intake and glucose tolerance were unaffected (Fig.6c and Extended Data Fig.3c). Notably, genetic disruption of post-synaptic GABA signaling within the MPO abolished the anorexic effect of leptin on HFD feeding, while the leptin-induced effect on LFD feeding remained intact (Fig.6c). Taken together, these data demonstrate that AgRPLepR→MPO neural circuit and post-synaptic GABAergic signaling play an exclusive role in control of leptin-mediated obesogenic feeding.

Fig. 6:
Fig. 6: The leptin-responsive AgRP-MPO circuit regulates obesogenic feeding via the α2-GABAA signaling.

a,b, 1-hr food intake of LFD and HFD after photostimulation of the AgRP→MPO circuit (a) or MC4RMPO neurons (b). (n = 8 per group; *p < 0.05). (c) 4-hrs food intake of LFD and HFD in the control, Agrp::LeprKO mice with leptin (i.p.) or vehicle. (n = 8 per group; *p < 0.05, #p < 0.05, ‡p < 0.05). d, Expression level of Gabra1 - Gabra6 in the MPO neurons of control and Agrp::LeprKO mice. e-g, The 4-hrs food intake of HFD (e), daily food intake of HFD (f), and body weight (g) were performed in Mc4rCre::Rosa26Cas9 mice with bilateral injection of either vehicle, AAV9-DIO-Gabra2sgRNA-tdTomato (GABAA-α2 KO), or AAV9-DIO-Gabra2cDNA-tdTomato (GABAA-α2 OE) into the MPO. h, Feeding efficiency in the mice described in e-g. i, Weight of adipose tissues in the mice described in e-g. j, Average value for RQ (VCO2/VO2) of the mice described in e-g. (n = 8 per group in e-j; *p < 0.05, #p < 0.05). k-n, Raster plots and firing rate of representative MC4RMPO neurons and non-MC4RMPO neurons under well-fed baseline, feeding with LFD and HFD (k), and under glucose i.p. injection (l). The statistical analysis of the firing rate under baseline, feeding with LFD and HFD (m), and under glucose treatment (n) were calculated. (n = 9 per group; *p < 0.05). Error bars represent mean ± SEM. paired two-tailed t test in a, b; unpaired two-tailed t test d; one-way ANOVA and followed by Tukey comparisons test in c, h, i, and m; two-way ANOVA and followed by Bonferroni comparisons test in e-g.

To identify the critical GABAA receptor components contributed to these phenotypes, we examined the expression profile of the major regulatory α1-α6 subunits within the MPO. Our statistical results showed that Gabra2 (encoding GABAA receptor α2 subunits) in the MPO neurons of Agrp::LeprKO mice was significantly enhanced (Fig.6d).

To establish the physiological roles of the α2-GABAA signaling in the context of leptin-mediated behavioral and metabolic phenotypes, Mc4rCre::Rosa26Cas9 mice were injected with AAV9-Gabra2sgRNA-mCherry into the MPO. We found that knockout of Gabra2 signaling in the MC4RMPO neurons reduces feeding of HFD and body weight while the glucose tolerance was not affected (Fig.6e-g and Extended Data Fig.4). Meanwhile, genetic deficiency in α2-GABAA signaling significantly decreased feeding efficiency and WAT (Fig.6h-j). On the other hand, the gain-of-function study showed that overexpression of α2-GABAA within the same MC4RMPO neurons manifested the exact opposite phenotypes, including moderately increased HFD feeding and body weight, significantly increased feeding efficiency and WAT adiposity (Fig.6h-j). Together, these results suggest that the α2-GABAA signaling within the MC4RMPO neurons exerts a key role in control of obesogenic feeding and obesity.

To better understand the functional role of MC4RMPO neurons in the regulation of feeding of HFD, we employed opto-tetrode recording in the MPO to investigate the activities of MC4RMPO neurons and non-MC4RMPO neurons under LFD, HFD, and hyperglycemia condition. A total 9 MC4RDMH neurons and 8 non-MC4RDMH neurons were identified through optogenetic-invoked spikes. The results showed that 5 out of 9 identified MC4R neurons responded to the feeding with HFD, with a reduction of firing rate from 20.5 Hz to 15.2 Hz (Fig.6k,m). These 9 MC4R neurons did not respond to the enhancement of blood glucose (Fig.6l,n). In conclusion, these data indicate that the post-synaptic MC4RMPO neurons within the AgRPLepR-MPO neural circuit mediate leptin-dependent control of hyperphagic response to the obesogenic diet.

DISCUSSION

Leptin exerts its behavioral and metabolic effects by dynamic influences onto the neural signaling within the hypothalamic AgRP neurons that are otherwise susceptible to the obesity-induced leptin resistance. In this report, we explored the neural circuit and transmitter signaling mechanism underlying leptin-mediated feeding and energy metabolism (Fig.7). We identified and characterized a unique bifurcating GABAergic neural circuit bearing distinct functional significance: the AgRPLepR→DMH circuit play a critical role in control of metabolic homeostasis through the α3-GABAA receptor signaling, whereas the AgRPLepR→MPO circuit dominantly regulates high-fat diet intake through the α2-GABAA receptor signaling. These leptin-responsive neural circuits play a fundamental role in regulation of hyperphagia and metabolic dysfunction in obesity. Further, enhancement of these GABAA receptor signaling systems found within the distinct post-synaptic targets exerts potent suppression of obesogenic feeding and restoration of metabolic homeostasis, a combinatorial effect which prevents obesity. We suggest that manipulation of these neural circuits and associated GABAA pathways can benefit novel obesity therapeutics.

Fig. 7:
Fig. 7: Diagram showing a GABAergic neural circuit reverses obesity by overriding central leptin resistance.

The GABAergic AgRPLepR→DMH circuit plays a critical role in control of leptin-mediated metabolic homeostasis through the α3-GABAA signaling within the MC4RDMH neurons, whereas the AgRPLepR→MPO circuit dominantly regulates feeding of obesogenic diet through α2-GABAA signaling within the MC4RMPO neurons. Together, this neural circuit overrides hypothalamic leptin resistance by exerting differentiated actions onto the post-synaptic targets, a combinatorial effect which prevents obesity.

This study utilized our newly established inducible knockout strategy to achieve rapid, post-developmental, targeted inactivation of LepR signaling which precisely illuminates the pathophysiological roles of central leptin signaling on the control of nutrient partitioning, and feeding efficiency. Compared with the mild effects of the non-inducible manipulation of the brain LepR signaling, our Lepr-KO model showed a robust disruption in various feeding and metabolism parameters, culminating into severe obesity which can only be achieved by global genetic deletion or the CRISPR–Cas9 technique 18,39,75,76. Our technique boasts many attractive features, such as the large repertoire of conditional mouse lines, transient and non-BBB-crossing inducer, and easy combination with numerous viral tools, that can be feasible to apply for many other neurological and endocrine questions.

The profiling assay using neonatal ablation of AgRP neurons revealed key neural population responding to leptin. Among various downstream targets of the AgRP circuit, the DMH and MPO neurons are most sensitive to the AgRP-dependent leptin signaling. Acute stimulation of post-synaptic MC4R neurons within the DMH but not MPO blunted glucose tolerance and feeding with LFD not HFD. Acute stimulation of MC4R neurons within the MPO specifically blunted the feeding with HFD. Further, we developed a novel viral-mediated, circuit-dependent, gene-editing technique to specifically ablate the GABA signaling from a subpopulation of AgRP downstream targets. Our results indicated that the GABA signaling within the AgRP→DMH circuit plays a significant role in metabolic homeostasis, whereas the GABAergic signaling within the AgRP→MPO circuit specially regulate leptin-dependent obesogenic feeding.

It has been demonstrated that GABAA receptor antagonists can prevent abnormalities in leptin actions on paraventricular hypothalamic neurons, which indicates GABAA signaling seems to contribute to a persistently reduced negative feedback of adiposity signals in obese models. Moreover, acute bicuculline administration seems to be able to suppress food intake and prevent the obesity phenotype77. Further studies identified the differential roles of subunits of GABAA receptors in the regulation of feeding and body weight. Studies using genetic knock-in mice revealed a strong correlation between individual GABAA subunits and specific phenotypes in BDZ treatment: the α1 subunit with sedative and amnesic effects; the α2 subunit with myorelaxant effects, the α3 subunit with anxiolytic effects, and the α5 subunit with neural plasticity and metabolic effects 57,78,79. Another study implicated that β2/β3-containing GABAA receptors contribute to feeding behavior in a hypothalamic circuit 80. Therefore, it is important to discover how the GABAA subunit in the hypothalamus is involved in the regulation of feeding. We employed the CRISPR-Cas9 gene-editing method to comprehensively evaluate the functions of GABAA subunits α3 and α2 within a leptin-responsive neural circuit. Genetic deletion of α3 in the MC4R neurons in the DMH suppressed nutrient partitioning and feeding efficiency. Genetic deletion of α2 in the MC4R neurons in the MPO inhibited obesogenic feeding. These studies demonstrate that the GABAA subunits α3 and α2 play differentiated roles in the control of leptin-mediated hypometabolism and obesogenic dietary intake.

Utilizing the in vivo optrode recording system, we had the unique advantage to perform stable intracellular recordings from MC4R neurons in the DMH and MPO in an intact and awake animal, allowing us to study complex behaviors and physiological responses. Here, we focused on MC4R neurons in the DMH, which are the post-synaptic targets of AgRP neurons. Our results showed that these groups of neurons are the most relevant to hyperglycemia due to their suppressed neural activity associated with high glucose levels. This study implicated that MC4R neurons in the DMH is dynamically involved in glucose homeostasis and LFD intake. On the other hand, the MC4R neurons in the MPO could respond to HFD during feeding behavior tests. This reveals the different roles of MC4R neurons in the various downstream targets of AgRP neurons and their regulatory effects on feeding and metabolism. In conclusion, MC4R signaling located within the DMH and MPO displayed differentiated firing responses under various feeding and nutrient conditions.

In conclusion, LepR in the AgRP neurons and the associated neural circuits and receptors are the primary mediators of leptin action in obesogenic feeding and metabolic homeostasis. We suggest that novel therapeutics targeting the α2/α3-GABAA receptor signaling, and associated pathways identified within distinct post-synaptic targets of the leptin-responsive GABAergic neural circuit would prevent obesity.

Funding

This project was supported by funding from a Shared Instrumentation grant from the NIH (S10 OD016167) and the NIH Digestive Diseases Center PHS grant P30 DK056338 to Cecilia Ljungberg. This work was supported by NIH grants (R01DK109194, R56DK109194, R01DK131596) to Q. Wu, the Pew Charitable Trust awards to Q. Wu (0026188), American Diabetes Association awards (#7-13-JF-61) to Q. Wu, Baylor Collaborative Faculty Research Investment Program grants to Q. Wu, USDA/CRIS grants (3092-5-001-059) to Q. Wu, the Faculty Start-up grants from USDA/ARS to Q. Wu, NIH grants (R01DK093587, R01DK101379, and R01DK117281) to Y. Xu, USDA/CRIS grants (3092-5-001-059) to Y. Xu, American Heart Association awards (17GRNT32960003) to Y. Xu,. Q. Wu is the Pew Scholar of Biomedical Sciences and the Kavli Scholar.

Author contributions

Q.W. and Y.Han conceived and designed the experiments. Y.Han performed the surgery, in vivo electrophysiology, immunohistological imaging, and relevant data analysis. Y.He performed slice electrophysiology. Y.Han performed the behavioral tests and relevant data analysis. L.H. performed the colony management, genotyping. Q.W. and Y.Han wrote the manuscript with inputs from all authors on the manuscript. Y.X. provided comments for manuscript. Q.W. supervised the project.

Competing interests

The authors declare that they have no competing interests.

Data and materials availability

All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Extended Data Fig. 1 Transsynaptic tracing from AgRP neurons to the MC4R neurons in the DMH and MPO. a-f, Representative images showing the expression of WGA-ZsGreen and MC4R in the DMH of the AgrpCre::Ai32 mice with AAV9-DIO-WGA-ZsGreen injected into the ARC. Scale bar in a for a-c, 200 μm; Scale bar in d for d-f, 100 μm. g-l, Representative images showing the expression of WGA-ZsGreen and MC4R in the MPO. Scale bar in g for g-i, 200 μm; Scale bar in j for j-l, 100 μm.

Extended Data Fig. 2 GABAA receptor α3 subunits in the DMH do not mediate HFD intake. a-c, The food intake of HFD (a), body weight (b) and GTT (c) were performed in Mc4rCre::Rosa26Cas9 mice with bilateral injection of either vehicle, AAV9-DIO-Gabra3sgRNA-tdTomato (a3 KO) into the DMH. (n = 8 per group; *p < 0.05). d, Real-time qPCR analysis of transcript levels of a3 expressed in MC4RDMH neurons isolated by fluorescence-activated cell sorting in the control and a3 KO mice. (n = 8 per group; *p < 0.05). e-g, The food intake of HFD (e, body weight (f) and GTT (g) were performed in Mc4rCre mice with bilateral injection of either vehicle, AAV9-DIO-Gabra3cDNA-tdTomato (a3 OE) into the DMH. (n = 8 per group; *p < 0.05). h, Real-time qPCR analysis of transcript levels of a3 expressed in MC4RDMH neurons isolated by fluorescence-activated cell sorting in the control and a3 OE mice. (n = 8 per group; *p < 0.05). Error bars represent mean ± SEM. one-way ANOVA and followed by Tukey comparisons test in d and h; two-way ANOVA and followed by Bonferroni comparisons test in c and g.

Extended Data Fig. 3 AgRP→MPO neural pathway dose not regulate glucose tolerance. a,b, GTT after photostimulation of the AgRP→MPO circuit (a) or MC4RMPO neurons. (n = 8 per group; *p < 0.05). c, GTT in control mice, GABA KO mice and leptin injected mice. The saline or leptin (4 mg/kg, i.p.) was injected respectively. Food intake or GTT was tested 30 minutes later. (n = 8 per group; *p < 0.05). Error bars represent mean ± SEM.

Extended Data Fig. 4 GABAA receptor α2 subunits in the MPO do not mediate LFD intake and glucose tolerance. a-c, The food intake of LFD (a), body weight (b) and GTT (c) were performed in Mc4rCre::Rosa26Cas9 mice with bilateral injection of either vehicle, AAV9-DIO-Gabra2sgRNA-tdTomato (a2 KO) into the MPO. (n = 8 per group; *p < 0.05). d, Real-time qPCR analysis of transcript levels of a2 expressed in MC4RMPO neurons isolated by fluorescence-activated cell sorting in the control and a2 KO mice. (n = 8 per group; *p < 0.05). e-g, The food intake of LFD (e), body weight (f) and GTT (g) were performed in Mc4rCre mice with bilateral injection of either vehicle, AAV9-DIO-Gabra2cDNA-tdTomato (a2 OE) into the MPO. (n = 8 per group; *p < 0.05). h, Real-time qPCR analysis of transcript levels of a2 expressed in MC4RMPO neurons isolated by fluorescence-activated cell sorting in the control and a2 OE mice. (n = 8 per group; *p < 0.05). Error bars represent mean ± SEM. one-way ANOVA and followed by Tukey comparisons test in d and h.

ACKNOWLEDGEMENT

The Metabolomics Core and the Mouse Metabolism Core at Baylor College of Medicine provided various technical support. Some AAV vectors were packaged by the Optogenetics and Viral Design/Expression Core at Baylor College of Medicine. Cecilia Ljungberg with the RNA In Situ Hybridization Core facility at Baylor College of Medicine provided technical support on the in situ hybridization.

Reference

  1. ↵
    Kahn, S. E., Hull, R. L. & Utzschneider, K. M. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature 444, 840–846, (2006).
    OpenUrlCrossRefPubMedWeb of Science
  2. Ren, J., Wu, N. N., Wang, S., Sowers, J. R. & Zhang, Y. Obesity cardiomyopathy: evidence, mechanisms, and therapeutic implications. Physiol Rev 101, 1745–1807, (2021).
    OpenUrl
  3. Clemmensen, C., Petersen, M. B. & Sorensen, T. I. A. Will the COVID-19 pandemic worsen the obesity epidemic? Nature reviews. Endocrinology 16, 469–470, (2020).
    OpenUrl
  4. Xia, G. et al. Reciprocal control of obesity and anxiety-depressive disorder via a GABA and serotonin neural circuit. Mol Psychiatry 26, 2837–2853, (2021).
    OpenUrl
  5. ↵
    Benito-Leon, J., Mitchell, A. J., Hernandez-Gallego, J. & Bermejo-Pareja, F. Obesity and impaired cognitive functioning in the elderly: a population-based cross-sectional study (NEDICES). Eur J Neurol 20, 899–906, e876-897, (2013).
    OpenUrlCrossRefPubMed
  6. ↵
    Friedman, J. M. Leptin at 14 y of age: an ongoing story. Am J Clin Nutr 89, 973S–979S, (2009).
    OpenUrlAbstract/FREE Full Text
  7. ↵
    Friedman, J. 20 years of leptin: leptin at 20: an overview. J Endocrinol 223, T1–8, (2014).
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Flak, J. N. & Myers, M. G., Jr.. Minireview: CNS Mechanisms of Leptin Action. Mol Endocrinol 30, 3–12, (2016).
    OpenUrlCrossRefPubMed
  9. Friedman, J. M. Leptin and the endocrine control of energy balance. Nat Metab 1, 754–764, (2019).
    OpenUrl
  10. Knight, Z. A., Hannan, K. S., Greenberg, M. L. & Friedman, J. M. Hyperleptinemia is required for the development of leptin resistance. PloS one 5, e11376, (2010).
    OpenUrlCrossRefPubMed
  11. Berglund, E. D. et al. Direct leptin action on POMC neurons regulates glucose homeostasis and hepatic insulin sensitivity in mice. J Clin Invest 122, 1000–1009, (2012).
    OpenUrlCrossRefPubMedWeb of Science
  12. ↵
    Caron, A. et al. POMC neurons expressing leptin receptors coordinate metabolic responses to fasting via suppression of leptin levels. eLife 7, (2018).
  13. ↵
    Considine, R. V. et al. Evidence against either a premature stop codon or the absence of obese gene mRNA in human obesity. J Clin Invest 95, 2986–2988, (1995).
    OpenUrlCrossRefPubMedWeb of Science
  14. ↵
    Maffei, M. et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1, 1155–1161, (1995).
    OpenUrlCrossRefPubMedWeb of Science
  15. ↵
    Minokoshi, Y. et al. Leptin stimulates fatty-acid oxidation by activating AMP-activated protein kinase. Nature 415, 339–343, (2002).
    OpenUrlCrossRefPubMedWeb of Science
  16. Huynh, F. K. et al. Disruption of hepatic leptin signaling protects mice from age- and diet-related glucose intolerance. Diabetes 59, 3032–3040, (2010).
    OpenUrlAbstract/FREE Full Text
  17. Pereira, S. et al. Metabolic effects of leptin receptor knockdown or reconstitution in adipose tissues. Sci Rep 9, 3307, (2019).
    OpenUrl
  18. ↵
    Xu, J. et al. Genetic identification of leptin neural circuits in energy and glucose homeostases. Nature 556, 505–509, (2018).
    OpenUrlCrossRefPubMed
  19. Varela, L. & Horvath, T. L. Leptin and insulin pathways in POMC and AgRP neurons that modulate energy balance and glucose homeostasis. EMBO Rep 13, 1079–1086, (2012).
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Vong, L. et al. Leptin action on GABAergic neurons prevents obesity and reduces inhibitory tone to POMC neurons. Neuron 71, 142–154, (2011).
    OpenUrlCrossRefPubMedWeb of Science
  21. ↵
    Baskin, D. G., Breininger, J. F. & Schwartz, M. W. Leptin receptor mRNA identifies a subpopulation of neuropeptide Y neurons activated by fasting in rat hypothalamus. Diabetes 48, 828–833, (1999).
    OpenUrlAbstract
  22. Xu, B. & Xie, X. Neurotrophic factor control of satiety and body weight. Nat Rev Neurosci 17, 282–292, (2016).
    OpenUrlCrossRefPubMed
  23. Scott, M. M. et al. Leptin targets in the mouse brain. J Comp Neurol 514, 518–532, (2009).
    OpenUrlCrossRefPubMedWeb of Science
  24. ↵
    Scott, M. M., Williams, K. W., Rossi, J., Lee, C. E. & Elmquist, J. K. Leptin receptor expression in hindbrain Glp-1 neurons regulates food intake and energy balance in mice. J Clin Invest 121, 2413–2421, (2011).
    OpenUrlCrossRefPubMedWeb of Science
  25. ↵
    Cowley, M. A. et al. Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus. Nature 411, 480–484, (2001).
    OpenUrlCrossRefPubMedWeb of Science
  26. ↵
    van den Top, M., Lee, K., Whyment, A. D., Blanks, A. M. & Spanswick, D. Orexigen-sensitive NPY/AgRP pacemaker neurons in the hypothalamic arcuate nucleus. Nat Neurosci 7, 493–494, (2004).
    OpenUrlCrossRefPubMedWeb of Science
  27. ↵
    Matheny, M., Shapiro, A., Tumer, N. & Scarpace, P. J. Region-specific diet-induced and leptin-induced cellular leptin resistance includes the ventral tegmental area in rats. Neuropharmacology 60, 480–487, (2011).
    OpenUrlCrossRefPubMedWeb of Science
  28. ↵
    Kishi, T. & Elmquist, J. K. Body weight is regulated by the brain: a link between feeding and emotion. Mol Psychiatry 10, 132–146, (2005).
    OpenUrlCrossRefPubMedWeb of Science
  29. ↵
    Han, Y. et al. Deciphering an AgRP-serotoninergic neural circuit in distinct control of energy metabolism from feeding. Nature communications 12, 3525, (2021).
    OpenUrl
  30. ↵
    Xia, G. et al. Reciprocal control of obesity and anxiety-depressive disorder via a GABA and serotonin neural circuit. Mol Psychiatry, (2021).
  31. ↵
    Myers, M. G., Jr.., Munzberg, H., Leinninger, G. M. & Leshan, R. L. The geometry of leptin action in the brain: more complicated than a simple ARC. Cell Metab 9, 117–123, (2009).
    OpenUrlCrossRefPubMedWeb of Science
  32. Gropp, E. et al. Agouti-related peptide-expressing neurons are mandatory for feeding. Nat Neurosci 8, 1289–1291, (2005).
    OpenUrlCrossRefPubMedWeb of Science
  33. Luquet, S., Perez, F. A., Hnasko, T. S. & Palmiter, R. D. NPY/AgRP neurons are essential for feeding in adult mice but can be ablated in neonates. Science 310, 683–685, (2005).
    OpenUrlAbstract/FREE Full Text
  34. Aponte, Y., Atasoy, D. & Sternson, S. M. AGRP neurons are sufficient to orchestrate feeding behavior rapidly and without training. Nat Neurosci 14, 351–355, (2011).
    OpenUrlCrossRefPubMed
  35. Konner, A. C. et al. Insulin action in AgRP-expressing neurons is required for suppression of hepatic glucose production. Cell Metab 5, 438–449, (2007).
    OpenUrlCrossRefPubMedWeb of Science
  36. ↵
    Steculorum, S. M. et al. AgRP Neurons Control Systemic Insulin Sensitivity via Myostatin Expression in Brown Adipose Tissue. Cell 165, 125–138, (2016).
    OpenUrlCrossRefPubMed
  37. ↵
    Halaas, J. L. et al. Physiological response to long-term peripheral and central leptin infusion in lean and obese mice. Proc Natl Acad Sci U S A 94, 8878–8883, (1997).
    OpenUrlAbstract/FREE Full Text
  38. ↵
    Friedman, J. M. & Halaas, J. L. Leptin and the regulation of body weight in mammals. Nature 395, 763–770, (1998).
    OpenUrlCrossRefPubMedWeb of Science
  39. ↵
    van de Wall, E. et al. Collective and individual functions of leptin receptor modulated neurons controlling metabolism and ingestion. Endocrinology 149, 1773–1785, (2008).
    OpenUrlCrossRefPubMedWeb of Science
  40. ↵
    Cone, R. D. Anatomy and regulation of the central melanocortin system. Nat Neurosci 8, 571–578, (2005).
    OpenUrlCrossRefPubMedWeb of Science
  41. Meister, B. Neurotransmitters in key neurons of the hypothalamus that regulate feeding behavior and body weight. Physiol Behav 92, 263–271, (2007).
    OpenUrlCrossRefPubMedWeb of Science
  42. Morton, G. J., Cummings, D. E., Baskin, D. G., Barsh, G. S. & Schwartz, M. W. Central nervous system control of food intake and body weight. Nature 443, 289–295, (2006).
    OpenUrlCrossRefPubMedWeb of Science
  43. Saper, C. B., Chou, T. C. & Elmquist, J. K. The need to feed: homeostatic and hedonic control of eating. Neuron 36, 199–211, (2002).
    OpenUrlCrossRefPubMedWeb of Science
  44. Wu, Q. & Palmiter, R. D. GABAergic signaling by AgRP neurons prevents anorexia via a melanocortin-independent mechanism. Eur J Pharmacol 660, 21–27, (2011).
    OpenUrlCrossRefPubMed
  45. Liu, T., Wang, Q., Berglund, E. D. & Tong, Q. Action of Neurotransmitter: A Key to Unlock the AgRP Neuron Feeding Circuit. Frontiers in neuroscience 6, 200, (2012).
    OpenUrl
  46. Wu, Q., Zheng, R., Srisai, D., McKnight, G. S. & Palmiter, R. D. NR2B subunit of the NMDA glutamate receptor regulates appetite in the parabrachial nucleus. Proc Natl Acad Sci U S A 110, 14765–14770, (2013).
    OpenUrlAbstract/FREE Full Text
  47. ↵
    Wu, Q., Boyle, M. P. & Palmiter, R. D. Loss of GABAergic signaling by AgRP neurons to the parabrachial nucleus leads to starvation. Cell 137, 1225–1234, (2009).
    OpenUrlCrossRefPubMedWeb of Science
  48. ↵
    Wu, Q., Clark, M. S. & Palmiter, R. D. Deciphering a neuronal circuit that mediates appetite. Nature 483, 594–597, (2012).
    OpenUrlCrossRefPubMedWeb of Science
  49. Rossi, M. A. & Stuber, G. D. Overlapping Brain Circuits for Homeostatic and Hedonic Feeding. Cell Metab, (2017).
  50. Richard, D. Cognitive and autonomic determinants of energy homeostasis in obesity. Nature reviews. Endocrinology 11, 489–501, (2015).
    OpenUrl
  51. ↵
    Meng, F. et al. New inducible genetic method reveals critical roles of GABA in the control of feeding and metabolism. Proc Natl Acad Sci U S A 113, 3645–3650, (2016).
    OpenUrlAbstract/FREE Full Text
  52. Garfield, A. S. et al. Dynamic GABAergic afferent modulation of AgRP neurons. Nat Neurosci 19, 1628–1635, (2016).
    OpenUrlCrossRefPubMed
  53. Pool, A. H. et al. Four GABAergic interneurons impose feeding restraint in Drosophila. Neuron 83, 164–177, (2014).
    OpenUrlCrossRefPubMed
  54. Cai, H., Haubensak, W., Anthony, T. E. & Anderson, D. J. Central amygdala PKC-delta(+) neurons mediate the influence of multiple anorexigenic signals. Nat Neurosci 17, 1240–1248, (2014).
    OpenUrlCrossRefPubMed
  55. ↵
    Krashes, M. J., Shah, B. P., Koda, S. & Lowell, B. B. Rapid versus delayed stimulation of feeding by the endogenously released AgRP neuron mediators GABA, NPY, and AgRP. Cell Metab 18, 588–595, (2013).
    OpenUrlCrossRefPubMedWeb of Science
  56. ↵
    Berridge, K. C. ‘Liking’ and ‘wanting’ food rewards: brain substrates and roles in eating disorders. Physiol Behav 97, 537–550, (2009).
    OpenUrlCrossRefPubMedWeb of Science
  57. ↵
    Cooper, S. J. Palatability-dependent appetite and benzodiazepines: new directions from the pharmacology of GABA(A) receptor subtypes. Appetite 44, 133–150, (2005).
    OpenUrlCrossRefPubMedWeb of Science
  58. Duke, A. N. et al. Enhanced sucrose pellet consumption induced by benzodiazepine-type drugs in squirrel monkeys: role of GABAA receptor subtypes. Psychopharmacology (Berl) 187, 321–330, (2006).
    OpenUrlPubMed
  59. Martire, S. I., Parkes, S. L. & Westbrook, R. F. The effects of FG 7142 on sensory-specific satiety in rats. Behav Brain Res 209, 131–136, (2010).
    OpenUrlCrossRefPubMedWeb of Science
  60. ↵
    Pecina, S. & Berridge, K. C. Brainstem mediates diazepam enhancement of palatability and feeding: microinjections into fourth ventricle versus lateral ventricle. Brain Res 727, 22–30, (1996).
    OpenUrlPubMedWeb of Science
  61. Strader, C. D., MacIntyre, D. E., Candelore, M. R. & Parker, E. M. Molecular approaches to the discovery of new treatments for obesity. Curr Opin Chem Biol 1, 204–209, (1997).
    OpenUrlPubMed
  62. ↵
    Berner, L. A., Bocarsly, M. E., Hoebel, B. G. & Avena, N. M. Baclofen suppresses binge eating of pure fat but not a sugar-rich or sweet-fat diet. Behav Pharmacol 20, 631–634, (2009).
    OpenUrlCrossRefPubMed
  63. ↵
    Berridge, K. C. & Pecina, S. Benzodiazepines, appetite, and taste palatability. Neurosci Biobehav Rev 19, 121–131, (1995).
    OpenUrlCrossRefPubMedWeb of Science
  64. Higgs, S. & Cooper, S. J. Hyperphagia induced by direct administration of midazolam into the parabrachial nucleus of the rat. Eur J Pharmacol 313, 1–9, (1996).
    OpenUrlCrossRefPubMedWeb of Science
  65. Soderpalm, A. H. & Berridge, K. C. The hedonic impact and intake of food are increased by midazolam microinjection in the parabrachial nucleus. Brain Res 877, 288–297, (2000).
    OpenUrlCrossRefPubMedWeb of Science
  66. ↵
    Berridge, K. C. & Treit, D. Chlordiazepoxide directly enhances positive ingestive reactions in rats. Pharmacol Biochem Behav 24, 217–221, (1986).
    OpenUrlCrossRefPubMedWeb of Science
  67. ↵
    Wu, Q., Howell, M. P., Cowley, M. A. & Palmiter, R. D. Starvation after AgRP neuron ablation is independent of melanocortin signaling. Proc Natl Acad Sci U S A 105, 2687–2692, (2008).
    OpenUrlAbstract/FREE Full Text
  68. ↵
    Tong, Q., Ye, C. P., Jones, J. E., Elmquist, J. K. & Lowell, B. B. Synaptic release of GABA by AgRP neurons is required for normal regulation of energy balance. Nat Neurosci 11, 998–1000, (2008).
    OpenUrlCrossRefPubMedWeb of Science
  69. ↵
    Shepard, R., Heslin, K., Hagerdorn, P. & Coutellier, L. Downregulation of Npas4 in parvalbumin interneurons and cognitive deficits after neonatal NMDA receptor blockade: relevance for schizophrenia. Translational psychiatry 9, 99, (2019).
    OpenUrl
  70. ↵
    Luquet, S., Phillips, C. T. & Palmiter, R. D. NPY/AgRP neurons are not essential for feeding responses to glucoprivation. Peptides 28, 214–225, (2007).
    OpenUrlCrossRefPubMedWeb of Science
  71. ↵
    Zimmer, M. R., Fonseca, A. H. O., Iyilikci, O., Pra, R. D. & Dietrich, M. O. Functional Ontogeny of Hypothalamic Agrp Neurons in Neonatal Mouse Behaviors. Cell, (2019).
  72. ↵
    Wu, Q., Howell, M. P. & Palmiter, R. D. Ablation of neurons expressing agouti-related protein activates fos and gliosis in postsynaptic target regions. J Neurosci 28, 9218–9226, (2008).
    OpenUrlAbstract/FREE Full Text
  73. ↵
    Han, Y. et al. A hindbrain dopaminergic neural circuit prevents weight gain by reinforcing food satiation. Sci Adv 7, (2021).
  74. ↵
    Xu, P. et al. Activation of Serotonin 2C Receptors in Dopamine Neurons Inhibits Binge-like Eating in Mice. Biol Psychiatry 81, 737–747, (2017).
    OpenUrlCrossRefPubMed
  75. ↵
    Goncalves, G. H., Li, W., Garcia, A. V., Figueiredo, M. S. & Bjorbaek, C. Hypothalamic agouti-related peptide neurons and the central melanocortin system are crucial mediators of leptin’s antidiabetic actions. Cell reports 7, 1093–1103, (2014).
    OpenUrl
  76. ↵
    Egan, O. K., Inglis, M. A. & Anderson, G. M. Leptin Signaling in AgRP Neurons Modulates Puberty Onset and Adult Fertility in Mice. J Neurosci 37, 3875–3886, (2017).
    OpenUrlAbstract/FREE Full Text
  77. ↵
    Meena, H., Nakhate, K. T., Kokare, D. M. & Subhedar, N. K. GABAA receptors in nucleus accumbens shell mediate the hyperphagia and weight gain following haloperidol treatment in rats. Life Sci 84, 156–163, (2009).
    OpenUrlPubMed
  78. ↵
    Cooper, S. J. Endocannabinoids and food consumption: comparisons with benzodiazepine and opioid palatability-dependent appetite. Eur J Pharmacol 500, 37–49, (2004).
    OpenUrlCrossRefPubMedWeb of Science
  79. ↵
    Cooper, S. J. Bidirectional control of palatable food consumption through a common benzodiazepine receptor: theory and evidence. Brain Res Bull 15, 397–410, (1985).
    OpenUrlPubMed
  80. ↵
    Sheng, G. et al. Hypothalamic huntingtin-associated protein 1 as a mediator of feeding behavior. Nat Med 12, 526–533, (2006).
    OpenUrlCrossRefPubMedWeb of Science
Back to top
Posted January 24, 2022.
Download PDF

Supplementary Material

Email
Reversal of obesogenic feeding and hypometabolism by a bifurcating GABAergic neural circuit
Yong Han, Yang He, Lauren Harris, Yong Xu, Qi Wu
bioRxiv 2022.01.23.477433; doi: https://doi.org/10.1101/2022.01.23.477433
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area