Leptin signaling in the dorsomedial hypothalamus couples breathing and metabolism in obesity

Abstract

Mismatch between CO₂ production (Vco₂) and respiration underlies the pathogenesis of obesity hypoventilation. Leptin-mediated CNS pathways stimulate both metabolism and breathing, but interactions between these functions remain elusive. We hypothesized that LEPR^b+ neurons of the dorsomedial hypothalamus (DMH) regulate metabolism and breathing in obesity. In diet-induced obese Lepr^bCre mice, chemogenetic activation of LEPR^b+ DMH neurons increases minute ventilation (Ve) during sleep, the hypercapnic ventilatory response, Vco₂, and Ve/Vco₂, indicating that breathing is stimulated out of proportion to metabolism. The effects of chemogenetic activation are abolished by a serotonin blocker. Optogenetic stimulation of the LEPR^b+ DMH neurons evokes excitatory postsynaptic currents in downstream serotonergic neurons of the dorsal raphe (DR). Administration of retrograde AAV harboring Cre-dependent caspase to the DR deletes LEPR^b+ DMH neurons and abolishes metabolic and respiratory responses to leptin. These findings indicate that LEPR^b+ DMH neurons match breathing to metabolism through serotonergic pathways to prevent obesity-induced hypoventilation.

Keywords: CP: Metabolism; CP: Neuroscience; designer receptor exclusively activated by designer drugs; dorsal raphe nucleus; hypercapnic ventilatory response; inspiratory flow limitation; intranasal administration; leptin receptor; obesity hypoventilation; obstructive sleep apnea; serotonin; sleep disordered breathing.

Figure 1.. Activation of LEPR^b neurons in the dorsomedial hypothalamus (DMH) stimulates breathing during sleep in obese mice

(A) Cre-dependent designer receptor exclusively activated by designer drug (DREADD; AAV8-hSyn-DIO-hM3D[Gq]-mCherry) or control virus (AAV8-hSyn-DIO-mCherry) was deployed in the DMH of Lepr^b-Cre diet-induced obese (DIO) mice. 3V, third ventricle. (B) Coronal section of the mouse brain shows mCherry positivity in the DMH confirming successful DREADD transfection. Scale bar, 500 μM. (C) The outlined area is enlarged; scale bar, 100 μM. (D) Schematics of the mouse brain illustrating a site of injection of the DREADD or control virus in Lepr^b-Cre-GFP DIO mouse. (E) LEPR^b-positive neurons of the DMH (green) were transfected with DREADD (red). The orange color results from merging the red and green colors demonstrating DREADD expression in the LEPR^b-positive neurons. Scale bars, 300 μM. The outlined area is enlarged in the right panel (scale bar, 70 μM). (F) Representative high-resolution confocal image showing DREADD expression (red) in the LEPR^b-positive neurons (green) resulting in orange color. Scale bar, 50 μM. (G) Percentage of LEPR^b+ mCherry+ cells (n = 6 mice). The values are expressed as means ± SE. (H) A representative screen of REM sleep in an obese Lepr^b-Cre mouse with obstructive apnea. yInspiratory flow limitation. SpO₂, oxyhemoglobin saturation. (I) Representative traces of REM sleep recordings in the same mouse expressing DREADD in the DMH after treatment with saline versus J60. (J) Minute ventilation (VE) during NREM sleep (left) and REM sleep (center) and maximal inspiratory flow (V_Imax) (right) in DIO Lepr^b-Cre mice transfected with DREADD (n = 16 or 17) virus in DMH after saline and J60. (K) Number of apneas per hour (apnea index) in REM sleep. (L) Number of oxyhemoglobin desaturations ≥4% from baseline (oxygen desaturation index [ODI]) during sleep. Data are plotted using boxplots (median ± 1.5*interquartile range). *p ≤ 0.05, **p < 0.01, and ***p < 0.001 using Wilcoxon matched-pairs signed rank test.

Figure 2.. Activation of LEPR^b neurons in the dorsomedial hypothalamus (DMH) stimulates breathing out of proportion to increases in metabolism

(A) Individual and grouped data show the differences between saline (Sal) and J60 on the baseline minute ventilation (VE) (Ai), VE at 8% inspired CO₂ (Aii), and hypercapnic ventilatory response (HCVR) (Aiii) in diet-induced obese (DIO) Lepr^b-Cre mice transfected with control (n = 10) or DREADD (n = 18) virus. Representative traces of HCVR (Aiv). (B and C) Energy expenditure during the light phase in DIO Lepr^b-Cre-GFP mice transfected with control or DREADD virus. Total oxygen consumption (VO₂) (Bi), total carbon dioxide production (VCO₂) (Bii), respiratory exchange ratio (RER) (Biii), total motor activity (Ci), VE/VO₂ (Cii), and VE/VCO₂ (Ciii) in the control (n = 6 or 7) and DREADD (n = 6–13) groups. Data are plotted using boxplots (median ± 1.5*interquartile range). ^##p < 0.01 and ^###p < 0.001 for the interaction between the virus and J60 effects using repeated-measures two-way ANOVA. *p % 0.05, **p < 0.01, and ****p < 0.001 using Wilcoxonmatched-pairssignedrankorMann-Whitney test.

Figure 3.. Photostimulation of the LEPR^b+ DMH neurons evokes excitatory postsynaptic currents (EPSCs) in the dorsal raphe (DR) serotonergic (5-HT) neurons

(A) Paradigm of floxed ChR2 injection followed by photostimulation of the DMH and patch-clamp recording of the DR neurons. Sagittal section of the brain containing the DMH ChR2-YFP-labeled LEPR^b neurons, fibers, and DR serotonergic (5-HT) neurons. EPSCs and inhibitory postsynaptic currents (IPSCs) were measured in voltage-clamp configuration (B). LEPR^b+ DMH neurons (green) show projections to the DR nucleus (red 5-HT stain), coronal section; scale bar, 70 μm; (C and D) ChR2 photoactivated responses in LEPR^b+ DMH neurons (n = 3 from 3 animals). (E and F) Photoactivated EPSCs in 5-HT DR neurons upon activation of LEPR^b fibers in Lepr^b-Cre mice after injection of floxed ChR2-GFP virus into the DMH. (G and H) Photoactivated EPSCs in 5-HT DR neurons upon activation of LEPR^b DMH cell bodies in crossbred Lepr^b-Cre-ChR2 mice. In (F) and (H), top traces from are male and bottom traces from female mice. The values are expressed as means ± SE. Confocal images showing 5-HT-stained neurons in green (I), LEPR^b fibers in the DR in white (J), the biocytin (BCT)-filled patch recorded cell in the DR in red (K), and co-localization of the patched 5-HT neuron surrounded by LEPR^b fibers (L). Sagittal section; scale bars, 30 μm.

Figure 4.. The serotonin receptor antagonist methysergide prevented the increase minute ventilation during sleep and HCVR induced by the activation LEPR^b neurons in the DMH

(A) Experimental study design. (B–D) Ventilation during sleep in diet-induced obese Lepr^b-Cre mice transfected with DREADD virus and treated with a serotonin receptor antagonist, methysergide (Met). (B) Individual and grouped data showing the effects of J60 ligand or saline on maximal inspiratory flow (V_Imax) (Bi), minute ventilation (VE) (Bii), respiratory rate (RR) (Biii), and mean inspiratory flow (Biv) during nonflow-limited breathing in NREM sleep in DREADD group mice that received saline (Sal), J60, or J60 + Met. (C) V_Imax (n = 10–13), (Ci), VE (Cii), RR (Ciii), and mean inspiratory flow (Civ) during non-flow-limited breathing in REM sleep. (D) V_Imax, (Di), VE (Dii), RR (Diii), and mean inspiratory flow (Div) during flow-limited breathing in REM sleep. (E) Baseline minute ventilation (VE) (Ei), VE at 8% inspired CO₂ (Eii) and, HCVR (Eiii) in diet-induced obese (DIO) Lepr^b-Cre mice transfected with DREADD virus that received Sal, J60, or J60 + Met. Sal group is the same for Figure 1 for comparison purposes. Data are plotted using boxplots (median ± 1.5*interquartile range). *p ≤ 0.05 and **p < 0.01 using Wilcoxon matched-pairs signed rank or Mann-Whitney test.

Figure 5.. Elimination of LEPR^b positive DMH neurons projecting to the DR prevented previously evoked increase in ventilation and metabolism

(A) Experimental design, including overall design (Ai). Schematic overview of the hypothesis in Lepr^b-Cre mice transfected with DREADD into the DMH (Aii) or DREADD into the DMH plus retrograde Cre-dependent AAV harboring caspase to the DR (Aiii). (B) Number of mCherry+ cells per sections (n = 3–5) and (C) histology in the absence of caspase. DAPI represents nuclei (gray, Ci), DREADD in red in Lepr^b-Cre mice (Cii). Merged image (Ciii). Scale bars, 50 μM. The same at a higher magnification (Civ). Scale bars, 20 μM. (D) Histology in the presence of caspase. DAPI (gray Di), note a significant reduction in DREADD (red) in Lepr^b-Cre mice that received Cre-dependent AAV harboring caspase to the dorsal raphe (Dii). Merged image (Diii). Scale bars, 50 μM. The same at a higher magnification (Div). Scale bars, 20 μM. 3V, third ventricle. (E–I) Difference (delta) between respiratory and metabolic effects of J60 in Lepr^b-Cre mice transfected to the DMH with control virus only, DREADD virus only, or DREADD virus plus retrograde Cre-dependent caspase to the DR. Individual and grouped data show the delta between J60 ligand and saline on minute ventilation (V_E) during non-flow-limited breathing in NREM sleep (E), non-flow-limited breathing in REM sleep (F), flow-limited breathing in REM sleep (G), and hypercapnic ventilatory response (H). (I) Metabolic measurements, total oxygen consumption (VO₂) (Ii), and total carbon dioxide production (VCO₂) (Iii) (n = 10). Control and DREADD group are the same for Figures 1 and 2 for comparison purposes. Data are plotted using boxplots (median ± 1.5*interquartile range). *p ≤ 0.05, **p < 0.01, and ****p < 0.0001 using Mann-Whitney test.

Figure 6.. Elimination of LEPR^b-positive DMH neurons projecting to the DR prevented the increase in ventilation during sleep induced by intranasal leptin

(A) Experimental study design showing intranasal leptin or vehicle treatment in diet-induced obese (DIO) mice with intact LEPR^b-positive DMH neurons projecting to the dorsal raphe (DR) (Ai) (n = 9–11). Effects of intranasal leptin or vehicle in Lepr^b-Cre DIO mice after caspase-induced elimination of LEPR^b-positive DMH neurons projecting to the DR (Casp) (Aii) (n = 6 or 7). (B–D) Individual and grouped data show the effects of intranasal leptin or vehicle on maximal inspiratory flow (V_Imax) (Bi), minute ventilation (V_E) (Bii), tidal volume (V_T) (Biii), and mean inspiratory flow (Biv) during non-flow-limited breathing in NREM sleep. V_Imax, (Ci), V_E (Cii), V_T (Ciii), and mean inspiratory flow (Civ) show the effects of intranasal leptin during non-flow-limited breathing in REM sleep. V_Imax (Di), V_E (Dii), V_T (Diii), and mean inspiratory flow (Div) show the effects of intranasal leptin during flow-limited breathing in REM sleep. (E) Hypercapnic ventilatory response (HCVR) measurements while awake, including baseline V_E (Ei), V_E at 8% inspired CO₂ (Eii), and HCVR (Eiii). Data are plotted using boxplots (median ± 1.5*interquartile range). *p ≤ 0.05 and **p < 0.01 using Wilcoxon matched-pairs signed rank or Mann-Whitney test. ^†p % 0.05, ^††p < 0.01, and ^†††p < 0.001, effect of leptin using two-way ANOVA or mixed-effects model. ^§p % 0.05 and ^§§p < 0.01, effect of caspase virus using two-way ANOVA or mixed-effects model.

Figure 7.. The effect of activation of LEPR^b neurons in the dorsomedial hypothalamus (DMH) in Lepr^b-Cre DIO female mice

(A) Experimental study design. (B) Sex differences at baseline (Bi) and the effect of J60 vs. saline (Sal) in females (Bii) on the baseline maximal inspiratory flow (V_Imax) and sex differences at baseline (Biii) and the effect of J60 vs. Sal in females (Biv) on minute ventilation during non-flow-limited breathing in NREM sleep. (C) The same parameters as in (B) during non-flow-limited breathing in REM sleep. (D) The same parameters as in (B) during flow-limited breathing in REM sleep. (E) Sex differences and the effect of J60 vs. saline in females on V_E at 8% inspired CO₂ (Ei and Eii, respectively) and hypercapnic ventilatory response (HCVR) (Eiii and Eiv). n = 8–10. Data are plotted using boxplots (median ± 1.5*interquartile range). *p ≤ 0.05, **p < 0.01, and ***p < 0.001 using Wilcoxon matched-pairs signed rank or Mann-Whitney test.