Sensory neurons expressing calcitonin gene-related peptide α regulate adaptive thermogenesis and diet-induced obesity

Abstract

Objectives: Heat-sensory neurons from the dorsal root ganglia (DRG) play a pivotal role in detecting the cutaneous temperature and transmission of external signals to the brain, ensuring the maintenance of thermoregulation. However, whether these thermoreceptor neurons contribute to adaptive thermogenesis remains elusive. It is also unknown whether these neurons play a role in obesity and energy metabolism.

Methods: We used genetic ablation of heat-sensing neurons expressing calcitonin gene-related peptide α (CGRPα) to assess whole-body energy expenditure, weight gain, glucose tolerance, and insulin sensitivity in normal chow and high-fat diet-fed mice. Exvivo lipolysis and transcriptional characterization were combined with adipose tissue-clearing methods to visualize and probe the role of sensory nerves in adipose tissue. Adaptive thermogenesis was explored using infrared imaging of intrascapular brown adipose tissue (iBAT), tail, and core temperature upon various stimuli including diet, external temperature, and the cooling agent icilin.

Results: In this report, we show that genetic ablation of heat-sensing CGRPα neurons promotes resistance to weight gain upon high-fat diet (HFD) feeding and increases energy expenditure in mice. Mechanistically, we found that loss of CGRPα-expressing sensory neurons was associated with reduced lipid deposition in adipose tissue, enhanced expression of fatty acid oxidation genes, higher exvivo lipolysis in primary white adipocytes, and increased mitochondrial respiration from iBAT. Remarkably, mice lacking CGRPα sensory neurons manifested increased tail cutaneous vasoconstriction at room temperature. This exacerbated cold perception was not associated with reduced core temperature, suggesting that heat production and heat conservation mechanisms were engaged. Specific denervation of CGRPα neurons in intrascapular BAT did not contribute to the increased metabolic rate observed upon global sensory denervation.

Conclusions: Taken together, these findings highlight an important role of cutaneous thermoreceptors in regulating energy metabolism by triggering counter-regulatory responses involving energy dissipation processes including lipid fuel utilization and cutaneous vasodilation.

Keywords: CGRP; Energy expenditure; Lipolysis; Obesity; Spinal sensory ganglion; Thermoregulation.

Figure 1

DT-injected CGRPα-DTR were resistant to diet-induced obesity. (A) Strategy to ablate CGRPα sensory neurons adapted from Mc Coy et al. Advillin-Cre was used to removed floxed GFP and drive human DTR in CGRPα sensory neurons. (B) Weight gain of CGRPα⁻ and littermate controls (WT) on a normal chow diet (NC) and high-fat diet (HFD). (C) Fat and lean mass of the CGRPα⁻ and control littermates after 12 weeks of HFD feeding. (D) VO₂ measured in 12-week-old HFD-fed mice. (E) Oxygen consumption in 12-week-old NC-fed mice. (F) ANCOVA analysis of energy expenditure with body weight as a covariate in 12-week-old HFD-fed mice, p < 0.005. (G) Cumulative food intake of 12-week-old HFD-fed mice measured over the course of 3 days. (H) Hematoxylin and eosin (H&E) sections of liver biopsies from the HFD-fed CGRPα⁻ and WT mice, scale 300 μm. (I) Blood glucose (BG) levels during glucose tolerance testing and area under the curve analysis of 16-week-old HFD-fed mice. (J) Insulin tolerance testing and area under the curve analysis of 17-week-old HFD-fed mice. N = 9–11, ∗∗∗∗p < 0.0001, ∗∗∗p < 0.0001, ∗∗p < 0.001, and ∗p < 0.05. All the values denote means ± SEM.

Figure 2

Loss of CGRPα neurons resulted in higher energy expenditure on high-fat diet feeding. (A) VO₂ in NC-fed animals before and after injection of the β-3 adrenergic agonist CL316,243 (CL). (B) Immediate response to CL monitored over 3 h. (C) Peak VO₂ values of the dark cycle obtained pre- and post-injection of CL. (D) RER in NC-fed animals before and after injection of CL. (E) VO₂ in HFD-fed animals on DIO for 2 weeks pre- and post-injection of CL. (F) Immediate response to CL monitored over 3 h. (G) Peak VO₂ values of the dark cycle obtained before and after CL delivery. (H) RER in HFD-fed animals before and after injection of CL. N = 5, ∗∗p < 0.001, and ∗p < 0.05. All of the values denote means ± SEM.

Figure 3

Increased lipid utilization in iBAT of the mice lacking CGRPα neurons. (A) Hematoxylin and eosin staining of iBAT from the HFD-fed CGRPα⁻ and WT controls littermates after 17 weeks of DIO. Scale bar 40 μm. (B) Basal oxygen consumption rate (OCR) of brown fat explants collected from the NC and HFD-fed mice measured by XF-Seahorse analyzer (N = 5). (C) Differential gene expression between the HFD-fed CGRPα⁻ and WT controls showing the distribution of shortlisted genes (p adj < 0.05, log₂ fold change > 0.7 or < −0.7, N = 4). (D and E) Heat maps representing differentially expressed genes in iBAT involved in fat metabolism, thermogenesis (D), and inflammatory chemokines (E) between HFD CGRPα⁻ and WT. (F) RT-qPCR validation of the differentially expressed genes (N = 6). ∗∗∗∗p < 0.0001, ∗∗∗p < 0.0001, ∗∗p < 0.001, and ∗p < 0.05. All of the values denote means ± SEM.

Figure 4

Higher energy expenditure of the CGRPα⁻mice was linked with increased fatty acid utilization in white adipose tissue. (A) Hematoxylin and eosin staining of gonadal WAT from littermates after 17 weeks of DIO. Scale bar 60 μm. (B–D) RT-qPCR of fatty acid oxidation genes in gonadal WAT (B), inguinal WAT (C), and intrascapular BAT collected from the HFD-fed mice treated with acute CL injection (1 mg/kg) or vehicle (Saline). (E) Exvivo lipolysis in gWAT biopsies treated with vehicle (Veh) or isoproterenol (Iso) in the HFD-fed mice that received acute vehicle (Saline) injections a week prior to sacrifice. (F) Exvivo lipolysis in gWAT biopsies treated with vehicle (Veh) or isoproterenol (Iso) in the HFD-fed mice that received acute CL injections a week prior to sacrifice. N = 5–6, ∗∗∗∗p < 0.0001, ∗∗∗p < 0.0001, ∗∗p < 0.001, and ∗p < 0.05. All of the values denote means ± SEM.

Figure 5

Loss of CGRPα neurons resulted in tail vasoconstriction and increased iBAT temperature on the high-fat diet. (A and B) Infrared imaging of tail temperatures at room temperature (22 ^оC) in (A) the NC-fed mice and (B) HFD-fed mice. (C) Representative images of dark cycle tail recordings from the HFD-fed mice at 22 ^оC. (D and E) Tail temperature recordings at thermoneutrality (28 ^оC) in the NC-fed (D) and HFD-fed (E) mice. (F) Representative images of dark cycle tail recordings from the HFD-fed mice at 28 ^оC. (G–H) Cutaneous iBAT temperature at 22 ^оC (G) and 28 ^оC (H) in the HFD-fed mice. (I) Infrared imaging of iBAT in the NC-fed mice upon administration of icilin (0.6 mg/kg). N = 5, ∗∗p < 0.001, and ∗p < 0.05. All of the values denote means ± SEM.

Figure 6

Intrascapular BAT-specific deletion of CGRPα neurons did not regulate energy metabolism and tail vasoconstriction. (A) Immunostaining of GFP (green) and TH (magenta) in cleared iBAT sections isolated from the Calca^{−<lox−GFP-lox-hDTR>} mice on NC showing CGRPα-GFP nerves lining TH⁺ fibers in the vasculature. Left panel (scale 50 μm) and right panel (scale 40 μm). (B) Weight gain of 22-week-old mice placed on the HFD after receiving DT microinjections in iBAT. (C) Fat mass and lean mass after 5 weeks of DIO. (D and E) VO₂ (D) and food intake (E) measurements obtained 4 weeks after switching the mice to DIO. (F–H) Infrared imaging of tail temperatures (F), rectal probe measurement of body core temperatures (G), and infrared imaging of iBAT (H) obtained at room temperature (22 ^оC) in the HFD-fed iBAT-DT mice.