Melanocortin System Activates Carotid Body Arterial Chemoreceptors in Hypertension

Abstract

BACKGROUND:

The internal milieu of the body is controlled by a system of interoceptors coupled to motor outflows that drive compensatory adaptive responses. These include the arterial chemoreceptors, best known for sensing arterial oxygen. In cardiometabolic diseases, such as essential hypertension, the carotid bodies (CB) exhibit heightened reflex sensitivity and tonic activity without an apparent stimulus. The mechanisms behind CB sensitization in these conditions are not well understood.

METHODS:

Guided by functional genomics, a range of functional assays is used to interrogate downstream intracellular and interorgan signaling pathways involved in arterial chemosensory function.

RESULTS:

Here, we report the presence of the MC4R (melanocortin 4 receptor) in the mammalian CB and show its elevated expression in experimental hypertension. We demonstrate that melanocortin agonists activate arterial chemosensory cells, modulating CB chemosensory afferent drive to influence chemoreflex-evoked sympathetic and ventilatory activity. Transcriptional analysis of hypertensive CB implicates the activation of the Mash1 (mammalian achaete-scute homolog 1; Ascl1) regulatory network in driving elevated Mc4r expression.

CONCLUSIONS:

Collectively, our data indicate a primarily pathophysiological role of melanocortin signaling in arterial chemosensation, contributing to excess sympathetic activity in cardiometabolic disease.

NOVELTY AND SIGNIFICANCE.

What Is known?

• Chronically elevated sympathetic nerve activity (eSNA) is a poorly controlled yet clinically relevant symptom of essential hypertension and comorbid type 2 diabetes.

• Aberrant activity of carotid body (CB) arterial chemoreceptors has been implicated in the generation of eSNA in essential hypertension and type 2 diabetes.

• The mechanisms underlying increased CB activity in cardiovascular and metabolic diseases are poorly understood.

What New Information Does This Article Contribute?

• MC4R (melanocortin 4 receptor) are expressed in hypoxia-sensitive glomus cells of the CB, and their expression is markedly elevated in hypertension.

• α-MSH (alpha melanocyte–stimulating hormone), which can be released into the circulation during stress or inflammation, activates MC4R in the CB, enhancing chemoreflex motor responses including eSNA.

• Melanocortin signaling in the CB appears primarily as a pathological adaptation to hypertension that may result in further amplification of eSNA.

The CB is an emerging therapeutic target for reducing eSNA in cardiorespiratory and metabolic disease. Here, we demonstrate for the first time a functional role of MC4R in CB arterial chemosensation and in mediating chemoreflex-evoked sympathoexcitation. We also present evidence of MC4R expression in human CB. By testing multiple melanocortin receptor agonists, we show agonist-specific bias in chemoreflex-evoked motor response activation. Although some MC4R agonists induce hypertension, Setmelanotide, a clinically approved MC4R agonist, activates arterial chemoreflex arc without eliciting sympathoexcitation. We further demonstrate that MC4R upregulation in hypertension is driven by dysregulated MASH1 (mammalian achaete-scute homolog 1; Ascl1) expression in the CB, revealing a mechanistic link between developmental transcriptional networks and pathological sympathoexcitation. The significance of our findings is 2-fold—they explain the absence of a pressor effect observed with Setmelanotide in contrast to other melanocortin agonists; and suggest that the arterial chemosensory arc may recruit new sensory modalities in disease states, linking pathological eSNA in hypertension with dysregulated metabolism.

Meet the First Author, see p 931

Editorial, see p 983

Neurogenic hypertension is a form of elevated blood pressure resulting from chronically elevated sympathetic nerve activity (eSNA).^1,2 Untreated eSNA exacerbates cardiovascular and all-cause mortality risk and represents an important clinical target in cardiometabolic disease management.^1,3,4 One known driver of eSNA in essential hypertension is the aberrant carotid body (CB) arterial chemoreceptor afferent activity.⁵ Both surgical removal and experimental pharmacological blockade of CB chemotransduction lower eSNA and arterial pressure.^5–10 However, the etiological mechanisms underlying arterial chemoreflex sensitization in the hypertensive state remain poorly understood. In this context, the CB is now recognized as a multimodal sensor that integrates various humoral signals in arterial circulation.^11,12 For example, recent evidence links disrupted signaling by GLP-1 (glucagon-like peptide-1) and leptin peptides in the CB to arterial chemoreflex sensitization, driving eSNA and elevated blood pressure in experimental hypertension.^13,14 This highlights the contribution of a largely unexplored arterial chemosensory-endocrine niche in the development of eSNA in cardiometabolic disease.

In the context of cardiometabolic disease, the melanocortin system represents a complex neuroendocrine network governing energy homeostasis and metabolism.¹⁵ It relies on α-MSH (alpha-Melanocyte–stimulating hormone) as the principal endogenous peptide acting on hypothalamic MC3R (melanocortin 3) and MC4R (melanocortin 4) receptors.^16,17 Due to their anorexigenic effect, MC4R agonists have emerged as promising antiobesity treatments.^18,19 However, the development of MC4R medication has been hindered by the side effects of sympathetically mediated cardiovascular stress.^20,21 A novel selective MC4R agonist, Setmelanotide, induced weight-loss in syndromic obesity caused by pathogenic POMC (pro-opiomelanocortin), PCSK1 (proprotein convertase subtilisin/kexin type 1), or LEPR (leptin receptor) deficiency without producing notable changes in blood pressure or heart rate.²² However, the absence of a hemodynamic pressor effect from Setmelanotide is not well understood.

In this study, we characterize Mc4r expression in the mammalian CB. We demonstrate that α-MSH acting on the CB evokes tonic afferent chemosensory activity and potentiates CB response to hypoxia via chemosensory cell activation. We further show that the small molecule MC4R agonist tetrahydroisoquinoline, but not setmelanotide, potentiates arterial chemoreflex-mediated sympathetic activity. Lastly, we demonstrate Mc4r upregulation in hypertensive CB is linked to transcriptional reprogramming of the Mash1 (mammalian achaete-scute homolog 1; Ascl1) transcriptional regulatory network.

Methods

Data Availability

Detailed description of Materials and Methods and Statistical Analysis performed is available in the Supplemental Methods. All supporting data are available from the corresponding authors on request.

Animal Studies

Procedures involving use of laboratory animals were carried out at the University of Bristol (United Kingdom) and the University of Auckland (New Zealand) with strict adherence to local rules and regulations. Procedures were approved by the University of Bristol Animal Welfare and Ethical Review Board and performed under a Home Office UK license in accordance with the provision of the UK Animals (Scientific Procedures) Act (1986), and the University of Auckland Animal Ethics Committee, respectively. All procedures also conform to the guidelines of animal care by the Division of Experimental Animals, Nagoya University Graduate School of Medicine and were approved by the Nagoya University Animal Experiment Committee. All animal procedures performed were in accordance with the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes.

Human Samples

All participants provided informed consent and voluntarily donated their post-mortem remains for medical studies at the Institute of Anatomy, Lithuanian University of Health Sciences, legislated under the Republic of Lithuania People Burial Act (2007, No. X-1404; Article 17). All procedures involving human tissue were carried out in accordance with local rules and regulations protecting participant anonymity and keeping prior medical history confidential.

Statistical Analysis

Hypothesis testing was performed using mixed regression models in R using Jamovi (Version 2.4.8; retrieved from https://www.jamovi.org). The criteria to choose the final model was based, first, on the analysis of residuals and, second, on the value of Akaike Information Criterion (AIC) goodness of fit. As a general approach, we first fitted our data using a linear distribution model (i.e., assuming normal distribution). If the analysis of residuals exposed a violation of the assumption of normality and/or heteroskedasticity of residuals, then data was fit a generalized Gaussian or Gamma distribution model, accordingly. Statistical tests performed for each dataset are indicated in the corresponding figure legends. Differences were considered statistically significant when p<0.05 and data are presented as mean ± Standard Error of the Mean (SEM).

Results

Our recent RNA-sequencing screen identified Mc4r as the most highly upregulated (log2 fold-change [LFC]=2.57, P_adj=3.60E-39) GPCR (G-protein–coupled receptor) in the CB of spontaneously hypertensive rats (SHR) compared with normotensive Wistar-Kyoto controls (Figure 1A and 1B).¹³ As Mc4r expression has not been previously described in arterial chemoreceptors, we mined public transcriptome data and confirmed conserved gene expression across species (Figure S1A).²³ Coexpression of melanocortin receptor accessory proteins (Mrap, Mrap2), Pomc, and Agrp further supported active melanocortin signaling in the CB (Figure S1C). To further validate our discovery, we measured Mc4r expression in an independent cohort of animals and found it consistently higher in the CB of SHRs across different sexes, age groups, and colonies (Figure 1C; Figure S1D through S1F). Notably, Mc4r expression was higher in inbred Wistar-Kyoto compared with outbred Wistar rats (Figure S1D), highlighting a phenotypic distinction between 2 commonly used SHR control strains.²⁴ Wistar controls were thus selected for all further analyses to better reflect changes in SHR. Next, we found Mc4r mRNA to be located inside clusters marked by high Th expression within the CB (Figure 1D). Quantifying Mc4r expression localized to TH (tyrosine hydroxylase)-immunoreactive chemosensory (glomus) cell clusters revealed a 36% higher expression (linear mixed-effect regression_Strain: estimate=1.36 [95% CI, 1.13–1.64]; P=0.015) in SHR compared with Wistar rats (Figure 1E and 1F; Figure S1G through S1I). Mc4r localization in isolated CB chemosensory cells was further corroborated in published single-cell RNA-sequencing data (Figure S1B).²⁵

Figure 1.

Mc4r expression is increased in arterial chemoreceptors in hypertension. A, Situated at the common carotid artery (CCA) bifurcation into internal carotid artery (ICA) and external carotid artery (ECA) branches, the carotid body is the primary arterial chemosensory organ in the mammalian body. B, Melanocortin receptor expression in the carotid bodies (CB) of Wistar-Kyoto (WKY) and spontaneously hypertensive rats (SHR) detected by RNA-sequencing (RNA-seq), as published originally.¹³ n=6. DEseq2: Wald test, Benjamini-Hochberg correction. C, Quantitative reverse transcription PCR (RT-qPCR) validation of Mc4r expression in the CB of male (13 weeks, n=4), dioestrus female (12–13 weeks, n=4), and prehypertensive male (4 weeks, n=3) SHR/NHsd and WKY/NHsd rats. Mc4r expression normalized to Eif4b—a housekeeping gene and presented as change relative to 13-week-old male WKY. Mean±SEM. Kruskal-Wallis test, Dunn post hoc test (Benjamini-Hochberg correction). D, Mc4r and Th mRNA in transverse CB section of Wistar rats. Dotted line marks CB chemosensory cell clusters visualized by dense clustering of Th mRNA puncta. Arrowheads indicate Mc4r mRNA located within chemosensory cell clusters. Images represent n=1. E, Quantification of Mc4r expression in TH⁺ chemosensory (glomus) cell clusters as shown in F. SHR (n=5; N=55), Wistar (n=4; N=54). P value indicates the fixed effect of strain on response variable tested using linear mixed-effects model. For details on the statistical models applied, see Supplemental Material. F, Paired fluorescence (left) and brightfield (right) images showing RNAscope in situ hybridization (ISH) signal for Mc4r expression TH⁺ chemosensory cells of SHR and Wistar rats. Images displayed using identical settings to ensure consistency and comparability across samples. Images representative n=5 (SHR), n=4 (Wistar). AU indicates arbitrary units; DAPI, 4′,6-diamidino-2-phenylindole; ΔΔCt, delta-delta cycle threshold; IHC, immunohistochemistry; LFC, log2 fold-change; padj, p-value corrected for multiple comparisons; Rn, Rattus norvegicus; SCG, superior cervical ganglion; and Th, tyrosine hydroxylase.

To validate MC4R functionality in arterial chemoreceptors, we measured Ca²⁺ transients in dissociated CB (glomus) cells using live-cell imaging (Figure 2A). Typically, 30 s superfusion of α-MSH (100 nmol/L) evoked a single Ca²⁺ event (Ca²⁺ response frequency: 1.1±0.04 events per α-MSH stimulus) measured 1.8±0.06 ΔF/F₀ in amplitude, 54.3±2.45 s in duration, and required 19.5±0.9 s to reach peak fluorescence (time to peak [TTP]; Figure 2B through 2I). Event mass, an estimation of the total amount of Ca²⁺ released, was calculated to be 96.9±5.4 arbitrary units (AU) (Figure 2H). Some cells exhibited a multievent Ca²⁺ response with amplitudes decreasing with time (Figure 2D and 2I). There were no statistically significant differences in event amplitude, duration, or TTP between strains; however, response frequency was greater in the SHR (1.8±0.15 versus 1.1±0.04 events/stimulus; Figure 2I), giving rise to an overall 36% reduction in event mass (Figure 2H). Defined by morphology and responsiveness to cyanide (CN⁻), 32% of chemosensory glomus cells in Wistar rats and 68% in SHR responded to α-MSH (Figure 2J). SHU-9119 (10 nmol/L), a potent melanocortin MC3 and MC4 receptor antagonist, abolished α-MSH–mediated increase in intracellular Ca²⁺ in 48% of cells that previously responded to α-MSH (Figure 2K). Cells responding to α-MSH in the presence of SHU-9119 exhibited diminished amplitude; however, neither frequency, duration, nor TTP was statistically different (Figure S2). Together, these data demonstrate that CB chemosensory cells are sensitive to an endogenous melanocortin receptor agonist, whereas upregulation of Mc4r in the CB of the SHR was associated with a greater number of α-MSH–sensitive chemosensory cells.

Figure 2.

α-MSH (alpha melanocyte–stimulating hormone) evokes transient Ca2⁺ events in dissociated carotid body (CB) chemosensory cells. A, Protocol used for live-cell Ca²⁺ imaging using Fluor-4 AM indicator dye. Chemosensory cells were first identified by their sensitivity to cyanide (potassium cyanide [KCN]). Subsequently, α-MSH (100 nmol/L) was administered in the absence and later in the presence of SHU-9119 (10 nmol/L). B, Representative morphology of α-MSH–sensitive chemosensory cells. Roman numerals correspond to the time points shown in D. Bar=10 µm. C, Representative trace of an α-MSH–evoked change in fluorescence intensity in dissociated chemosensory cells. α-MSH–evoked Ca²⁺ response was blocked by coadministration of SHU-9119 in 48% of chemosensory cells. D, Example trace of multievent Ca²⁺ oscillations evoked by α-MSH in spontaneously hypertensive rat (SHR). Note, majority of chemosensory cells exhibited a single Ca²⁺ event to 30 s superfusion of α-MSH. E, Ca²⁺ response amplitude, (F) duration, and (G) time to peak fluorescence. H, Event mass calculated as a function of response amplitude and duration, representing the total amount of Ca²⁺ released. Each dot in E through H represents an individual cell. I, Average number of Ca²⁺ events evoked by α-MSH stimulus. Dot size represents the number of cells exhibiting the indicated number of Ca²⁺ events. J and K, Proportion of cyanide (CN⁻)–sensitive cells that also exhibited Ca²⁺ response to α-MSH. Each dot in J represents the ratio of cells in a recording frame used to record Ca²⁺ responses. Transparent lines and dots in K represent ratio of cells in individual recording frame and average value across all recorded frames, respectively. Line thickness in K represents the number of cells in a recording frame. P values in E through J indicate the fixed effect of Strain on response variable tested using generalized mixed-effects models. P values in K show false discovery rate (FDR)-adjusted post hoc pairwise comparison of estimated marginal means between α-MSH and SHU within each strain, based on a generalized mixed-effect model testing the interaction between treatment (α-MSH, SHU-9119) and strain while accounting for paired responses within each recording frame. For details on the statistical models applied, see Supplemental Material. Wistar n=6. SHR n=10.

To validate the presence of MC4R at the protein level and visualize the chemosensory cell heterogeneity indicated by live-cell imaging, we devised 3 independent MC4R detection strategies (Figure 3A). First, we performed immunolabeling using a commercial polyclonal antibody targeting the extracellular N terminus of human MC4R (Alomone Labs, AMR-024), previously validated for specificity by site-specific MC4R knockdown.²⁸ AMR-024–labeled CB chemosensory cells outlined by S100-immunoreactive sustentacular (type II) cells (Figure 3B through 3D). Notably, some chemosensory cells displayed no immunoreactivity to AMR-024 (Figure 3C). AMR-024 labeling of dissociated chemosensory cells similarly indicated a mixed population of immunoreactive cells (Figure 3E). As an alternative strategy to investigate MC4R localization and accessibility, we synthesized a novel fluorescent label—HS-014–Sulfo549—based on HS-014, a potent and selective melanocortin MC4R antagonistic peptide (Figure 3A).²⁹ Supporting the binding of endogenous and/or exogenously administered melanocortin receptor ligands within the CB, HS-014-Sulfo549 (5nmol; s.c. injection) resulted in punctate fluorescence signal within chemosensory cell clusters compared with sham-injected controls (Figure 3F through 3G). Notably, strong fluorescent signal labeled the endothelial cells within the CB (Figure S7), indicating that HS-014-Sulfo549 was partly retained in the vasculature (Figure 3F). Applied to dissociated chemosensory cells HS-014-Sulfo549 (100 nmol/L) exhibited bright punctate signal (Figure 3H). Here, HS-014-Sulfo549 labeled 30.6% of chemosensory glomus cells, of which 45.3% subsequently responded to α-MSH, suggesting sustained receptor blockade by HS-014 binding (Figure 3J). Overall, the percentage of HS-014-Sulfo549 labeled chemosensory cells was concordant with the expected ratio of α-MSH–sensitive chemosensory cells (Figure 2I). However, we observed HS-014-Sulfo549 bound to cells nonresponsive to CN⁻ (Figure 3I); thus we could not solely attribute label binding to chemosensory glomus cells. Lastly, we used 2 different primary polyclonal antibodies targeting N and C termini of rat MC4R validated using Mc4r-deficient mice.²⁶ Immunohistochemistry with these antibodies in the CB of the SHR revealed weak background signals, which were not preabsorbed with their antigenic peptides (Figure 3K; data not shown for C terminus antibody). The same antibodies visualized MC4R-bearing neuronal primary cilia in the paraventricular hypothalamic nucleus (Figure S3).²⁶ Contrasting results between AMR-024, custom antibodies, and HS-014–Sulfo549 antagonistic label, performed by 2 independent laboratories, indicated that MC4R cannot be unequivocally detected in the CB at the protein level with experimental strategies used in this study. These findings likely relate to overall sparsity of the protein and the differences in antibody specificity, site-specific MC4R modifications, and receptor availability to exogenous ligands.

Figure 3.

MC4R (melanocortin 4 receptor) localization in carotid body arterial chemoreceptors. A, Strategies used to detect MC4R included (1) immunolabeling with a commercial polyclonal antibody targeting human (hs) MC4R (AMR-024; Alomone); (2) custom antagonistic peptide label HS-014-Sulfo549; (3) immunolabeling with custom polyclonal antibodies targeting N and C termini of rat (M) MC4R as described originally.²⁶ B and C, Immunolabeling of transverse carotid body (CB) sections using AMR-024 antibody. AMR-024 signal was localized to CB chemosensory cells enveloped by S100-immunoreactive sustentacular cells. Asterisks mark nuclei within glomus cell cluster inert to both S100 and AMR-024 immunolabeling. Images represent n=2. D, Fluorescence intensity profile for S100 and AMR-024 signals across line shown in C1 and C2. E, Immunolabeling of dissociated chemosensory cells using AMR-024 antibody. PNA (peanut agglutinin) was used to visualize chemosensory cells.²⁷ Asterisks mark chemosensory cells not immunoreactive to AMR-024. F, Representative HS-014–Sulfo549 labeling of transverse CB sections in the spontaneously hypertensive rat (SHR) and (G) sham-injected Wistar controls. Subcutaneous injection of HS-014–Sulfo549 resulted in punctate fluorescence signal within glomus cells visualized by tyrosine hydroxylase (TH) immunoreactivity. Arrowheads mark capillaries labeled by tomato (Lycopersicon esculentum) lectin (TL/LEL). HS-014–Sulfo549 fluorescence signal uniformly labeled the microvasculature with the CB and overlapped with TL/LEL labeling. Maximum intensity projection of 7 confocal optical sections. Images represent n=2. H and I, Punctate HS-014–Sulfo549 labeling (yellow) of dissociated chemosensory cells loaded with Fluor-4 (blue). Cells outlined during peak Ca²⁺ response. Black asterisks indicate cyanide (CN⁻) and α-MSH (alpha melanocyte–stimulating hormone) responsive cells. White asterisk indicates CN⁻-responsive cells. Arrowheads mark nonchemosensory cells. J, Proportion of chemosensory cells (CN⁻ sensitive) that responded to α-MSH and colabeled by HS-014–Sulfo549. Diagram depicts the protocol used to derive the proportion. n=4 (Wistar). Whiskers indicate SEM. Dots represent ratio of cells in individual recording frame. K, Immunolabeling of transverse CB sections in SHR using the N terminus targeting anti-MC4R antibody and (L) preabsorbed antibody control.²⁶ Images represent n=3. See also Figure S3. AU indicates arbitrary units; DAPI 4′,6-diamidino-2-phenylindole; and KCN, potassium cyanide.

To explore the functional role of melanocortin agonism on CB excitability, we devised an ex vivo CB–carotid sinus nerve (CSN) preparation allowing us to measure isolated CB afferent discharge in response to arterially delivered stimuli (Figure 4A). Administration of incremental concentrations of α-MSH (0.1, 1, 10 µM; Figure 4B) evoked a cumulative increase in tonic CSN firing frequency that was dose dependent (Wald-χ²_{Strain*α-MSH[2]}=13.056, P=0.001; Figure 4C and 4D). Notably, α-MSH administration did not evoke a transient burst of CSN activity (Figure S4A) such as occurs in response to mimetic hypoxia (Figure 4G). The increase in tonic CSN firing evoked by 10 µM α-MSH was blocked by 100 nM SHU-9119 (Wald-χ²_{strain×stimulus[1]}=0.4347, P=0.0037; Figure 4C and 4D) and persisted in normoxic conditions (P=0.257; Figure S4B through S4E). In this protocol (Figure 4B), we also assessed the effects of α-MSH on CB chemosensitivity using a suprathreshold, submaximal CN⁻ dose (12 mmol, 100 µL bolus). We found that α-MSH potentiated the CB afferent response to mimetic hypoxia (F_{condition[1, 8]}=6.617, P=0.033; Figure 4E through 4G). Notably, α-MSH–evoked changes in CB chemosensitivity were augmented in the SHR (F_{SHR[1, 8]}=11.299, P=0.01), but not in normotensive Wistar rats (F_{Wistar[1, 8]}=0.0467, P=0.514; Figure 4E and 4F). To further validate if melanocortin signaling had a physiological effect on arterial chemosensory function in vivo, we assessed the chemoreflex-evoked pressor response in unanaesthetized, freely moving SHR and Wistar rats after intravenous α-MSH administration. In a paired design, each animal served as its own control where the exposure to α-MSH (1 nmol), α-MSH+SHU-9119 (5 µg kg⁻¹), or vehicle was delivered separately on consecutive days, and resting arterial pressure and pressor motor responses to CN⁻ (0.04% (m/v); 20 µL IV) were assessed at baseline and after treatment (Figure 4H through 4I; Figure S5). Here, the peak systolic BP response to CN⁻ differed depending on treatment (F_{treatment:time[2]}=9.92; P=8.47×10⁻⁴) and was augmented by α-MSH (P_adj=0.0173), diminished in presence of SHU-9119 (P_adj=0.0001), whereas no statistically significant change was observed with vehicle control (P_adj=0.8884; Figure 4J). The same trend was observed on resting mean arterial pressure (F_{treatment:time[2]}=9.992, P=8.18×10⁻⁴; Figure 4L). An inverse trend was observed for chemoreflex-evoked bradycardia (F_{treatment:time[2]}=8.977; P=7.16×10⁻⁴; Figure 4K). Notably, for all measures there was a significant interaction between strain and treatment effect (pressor response: F_{strain:treatment:time[2]}=6.89, P=4.74×10⁻³; bradycardia: F_{strain:treatment:time[2]}=4.07, P=0.0256; resting mean arterial pressure: F_{strain:treatment:time[2]}=11.33, P=4.13×10⁻⁴), where neither treatment had a statistically significant effect in normotensive Wistars (Figure S5). For all readings, no significant differences (P>0.05) were observed at baseline between different days (Figure 4H; Figure S5). Collectively, these data show melanocortin system is acting on the CB to potentiate both tonic chemoafferent activity and response sensitivity to hypoxia.

Figure 4.

α-MSH (alpha melanocyte–stimulating hormone) activates arterial chemoreceptors and potentiates carotid body (CB)–evoked pressor response. A, diagram of the ex vivo CB–carotid sinus nerve (CSN) preparation. B, Recording protocol: After a 30-minute acclimatization period, 2 cyanide (CN⁻) responses were recorded at baseline. Increasing doses of α-MSH were then delivered as a concentrated bolus in carbogenated artificial cerebrospinal fluid. After a cumulative dose-response to α-MSH, another CN⁻ challenge was administered to measure changes in CB sensitivity. Subsequently, α-MSH was delivered together with SHU-9119 (100 nmol/L). At the end of the recording, CN⁻ was used to assess CB viability, and lignocaine (0.2 mL bolus; Nopaine 20 mg mL⁻¹) was delivered to record the baseline noise level for subsequent subtraction for CSN analysis. C, Representative traces of 10 µmol/L α-MSH evoked increase in tonic CB activity in spontaneously hypertensive rat (SHR). D, CSN firing frequency change evoked by α-MSH. Data represent the maximum increase in the average CSN discharge frequency after each α-MSH stimulus. Percentage change was calculated in comparison to the average CSN discharge frequency preceding each α-MSH stimulus. Hypothesis testing was performed using mixed regression models with random intercept, excluding the SHU-9119 time point. Data shown for SHR. n=7. E, Change in area under curve (AUC) of the integrated CSN activity response to CN⁻ challenge before and after α-MSH administration. Two initial CN⁻ responses were averaged to represent baseline (BL). Hypothesis testing was performed using mixed regression models with random intercept. SHR n=7; Wistar n=5. P values in D through E indicate the fixed effect of α-MSH treatment on response variable tested using a linear mixed-effects model. F, Same as E expressed as percentage change compared with baseline response. G, Representative traces of CSN activity response to CN⁻ challenge before and after α-MSH administration in SHR. H, Protocol used to assess the effect of intravenous α-MSH (1nmol) on arterial chemoreflex in conscious, instrumented SHR. Black arrows denote time of CN⁻ administration. White arrowheads denote time of treatment administration. Roman numerals correspond to traces shown in I. I, Representative traces of arterial pressor response to CN⁻ intravenous bolus are compared between baseline and different treatments. All traces originate from the same animal. Representative n=6. J, Chemoreflex-evoked pressor response measured as the peak systolic blood pressure (SBP) reading after CN⁻ bolus. K, Chemoreflex-evoked bradycardia measured as the lowest drop in heart rate (HR) after CN⁻ bolus. L, Resting mean arterial pressure (AP) measured for periods indicated by the double horizontal line in H. In J through L, points and thick lines represent Mean±SEM across all animals in each group. Thin lines represent individual animals. α-MSH n=6. α-MSH+SHU-9119 n=6. vehicle n=4. P values in J through L show FDR-adjusted post hoc pairwise comparisons of estimated marginal means between baseline and treatment within each group, based on a generalized mixed-effect model testing the fixed effect of treatment (α-MSH, α-MSH+SHU-9119, or vehicle) while accounting for paired responses within animals. For details on the statistical models applied, see Supplemental Material. CCA indicates common carotid artery; ECA, external carotid artery; and ICA, internal carotid artery.

As the arterial pressor response is primarily mediated via chemoreflex-evoked reflex sympathoexcitation, we used an in situ working heart-brainstem preparation (WHBP; Figure 5A).³⁰ In this way, we were able to validate the contribution of CB melanocortin signaling to arterial chemoreflex-mediated motor activity. WHBP allowed focal drug delivery to the CB via a cannulated internal carotid artery (Figure 5B) and simultaneous recording of phrenic and thoracic sympathetic activity (tSNA) in response to mimetic hypoxia (6 mmol CN⁻, 100 µL bolus). Drug effects were compared with averaged control responses (Figure 5C). For these experiments, we chose tetrahydroisoquinoline, a potent and selective small molecule MC4R agonist, due to its enhanced selectivity and stability compared with α-MSH. Tetrahydroisoquinoline delivered to the CB progressively increased resting respiratory rate (F_{time[3, 15]}=13.95, P=4.82×10⁻⁶; Figure 5D and 5F). This change was linked to diminished inspiratory time (Wald-χ²_[3]Time=31.75, P=5.9×10⁻⁷; Figure 5G) and phrenic nerve activity discharge amplitude (Wald-χ²_[3]Time=8.76, P=0.033; Figure 5H), resulting in a strain-dependent (Wald-χ²_{[3] interaction} =19.99, P=1.7×10⁻⁴) change in central inspiratory drive (Wald-χ²_[3]Time=13.24, P=0.004; Figure 5I) with no statistically significant effect on resting tSNA (Wald-χ²_[3]Time=2.62, P=0.454; Figure 5E). No significant difference in the effect of tetrahydroisoquinoline on resting parameters was observed between SHR and normotensive Wistar controls. Because our previous data showed α-MSH to potentiate the CSN afferent and reflex pressor responses to hypoxia (Figure 4), we tested if tetrahydroisoquinoline agonism potentiates chemoreflex-evoked reflex phrenic and sympathetic nerve activity. Tetrahydroisoquinoline augmented the chemoreflex-evoked tSNA response (F_{time[4, 40]}=5.62, P=0.001) and exhibited no statistically significant effect on chemoreflex-evoked bradycardia (F_{time[4, 33]}=1.203, P=0.328) and tachypnoea (Wald-χ²_time[4]=5.684, P=0.224; Figure 5K through 5L).

Figure 5.

Tetrahydroisoquinoline (THIQ) potentiates chemoreflex-evoked sympathetic and respiratory motor responses. A, Diagram of the working heart-brainstem preparation (WHBP) preparation used to investigate sympathetic and respiratory effects of THIQ. In decerebrated WHBP (no hypothalamus), observed responses were evoked by focal arterial drug injections to carotid bodies (CB). B, THIQ was administered unilaterally to the CB via a cannula inserted into the internal carotid artery (ICA). After accessing the CB, drugs were washed out via the external carotid artery (ECA). C, Protocol used to examine THIQ effects on resting and chemoreflex-evoked tSNA and respiratory drive. Initially, the functional arterial chemoreflex was confirmed using a single cyanide (CN⁻) challenge administered via the abdominal aorta perfused retrogradely. Before administering THIQ, 2 CN⁻ control responses were recorded. THIQ effect on arterial chemoreflex sensitivity was tested through a series of CN⁻ challenges. Data shown in E through I represent a different set of animals, where no CN⁻ challenges were used besides the initial reflex viability assessment. Arrows denote CN⁻ bolus. D, Representative traces of chemoreflex-evoked phrenic (respiratory) and tSNA responses before and 30 min after THIQ administration in spontaneously hypertensive rats (SHR). Arrows denote CN⁻ bolus. E through I, THIQ effect on resting tSNA (E) and phrenic discharge rate (F), inspiration time (TI) measured as the duration of the PN burst (G), peak PN amplitude (Amp) (H), and central inspiratory drive (I) defined as a function of peak phrenic Amp and inspiration time. n=7 SHR. n=6 Wistar. J through L, THIQ effect on chemoreflex-evoked motor tSNA (J), tachypnoeic (K), and bradycardia (L) responses. n=6. P values in E through L show the fixed effect of THIQ (time) and strain on response variable tested using a generalized mixed-effect model accounting for paired responses within animals. For details on the statistical models applied, see Supplemental Material. AU indicates arbitrary units; BL, baseline; CCA, common carotid artery; PNA, phrenic nerve activity; PP, perfusion pressure; and tSNA, thoracic sympathetic nerve activity.

To validate our findings in a more clinically pertinent scenario, we tested an Food and Drug Administration–approved MC4R agonist, Setmelanotide, delivered systemically (10 µmol/L) in the circulating solution using WHBP (Figure 6A and 6B). We focused this analysis on the SHR, as the chemoafferent and motor effects of melanocortin agonism were predominantly observed in the hypertensive condition. Setmelanotide in the perfusate had no significant effect on the resting tSNA tone (Wald-χ²_[5]=6.15, P=0.292; Figure 6C). However, Setmelanotide significantly increased the phrenic nerve activity burst rate from an average of 28±2.7 (mean±SEM) burst/min at baseline to 36±1.8 burst/min 30 minutes after administration (Wald-χ²_[5]=16.0, P=0.007; Figure 6D). This was accompanied by shortened inspiratory time (Wald-χ²_[5]=14.9, P=0.011), whereas no statistically significant effect was observed on phrenic amplitude (Wald- χ ²_[5]=8.24, P=0.143). Collectively, this resulted in an increase in central inspiratory drive (Wald-χ²_[5]=16.3, P=0.006; Figure 6E through 6G). Setmelanotide had no significant effect on respiratory-sympathetic coupling (7). At the end of the protocol, CB denervation (CBX) led to a reduction in resting sympathetic tone and inspiratory drive, underscoring peripheral chemosensory input as the mediator of the observed effects. In contrast to tetrahydroisoquinoline, Setmelanotide had no statistically significant effect on chemoreflex-evoked sympathoexcitation (F_time[5,18]=0.909, P=0.497; Figure 6H), whereas tachypnoea response appeared markedly augmented. Setmelanotide augmented the chemoreflex-evoked increase in respiratory rate (F_time[5,18]=3.64, P=0.019) while phrenic nerve activity amplitude was significantly reduced (Wald-χ²_[5]=14, P=0.016) resulting in a net increase in chemoreflex-evoked central inspiratory drive (Wald-χ²_[5]=17.8, P=0.003; Figure 6J through 6L). Collectively, our data provides considerable evidence that melanocortin agonism in the CB contributes to arterial chemoreflex-mediated respiratory (tetrahydroisoquinoline and Setmelanotide) and sympathetic (tetrahydroisoquinoline only) motor responses in experimental hypertension.

Figure 6.

Setmelanotide augments chemoreflex-evoked respiratory but not sympathetic response. A, Working heart-brainstem preparation (WHBP) setup used to investigate the sympathetic and respiratory effects of setmelanotide (SMT). Cyanide (CN⁻) was administered unilaterally via the internal carotid artery (ICA) to evoke arterial chemoreflex motor responses. B, Protocol used to examine chemoreflex-evoked thoracic sympathetic nerve activity (tSNA) and respiratory drive. Initially, resting activity was recorded, followed by a pair of CN⁻ responses at baseline. Next, 10 µmol/L SMT was added to the circulating Ringer’s solution and arterial chemoreflex sensitivity tested through a series of CN⁻ challenges. At the end of the protocol, carotid bodies (CB) were denervated (CBX) to assess peripheral chemosensory drive. CBX was confirmed by the absence of CN⁻ response as shown previously.⁵ C through G, Setmelanotide effect on resting tSNA (C) and phrenic nerve (PN) discharge rate (D), inspiration time (TI) measured as the duration of the PN burst (E), peak PN amplitude (F) and central inspiratory drive (G) defined as a function of peak phrenic amplitude and inspiration time. Data presented represented average value over a 1-minute interval preceding the administration of CN⁻ and SMT, and after CBX. H through L, Setmelanotide effect on chemoreflex-evoked motor tSNA (H) and tachypnoeic (I through L) responses. Data presented as the area under curve (AUC) (H), mean (J), or max (I, K, and L) value during the CN⁻ response in relation to the background (bg) level preceding the CN⁻ stimulus. Delta values (gray) represent changes in chemoreflex-evoked motor responses normalized for changes in background activity. P values in C through L show the fixed effect of Setmelanotide (time) on response variable tested using a generalized mixed-effect models accounting for paired responses within animals. For details on the statistical models applied, see Supplemental Material. Data shown were collected in spontaneously hypertensive rats. n=7. Amp indicates amplitude; AU, arbitrary units; CCA, common carotid artery; ECA, external carotid artery; KCN, potassium cyanide; and PP, perfusion pressure.

To assess whether the identified sympathoexcitatory and ventilatory effects may be pertinent to human disease, we validated MC4R expression in human CB samples (Figure 7A). MC4R mRNA was detected in 4 of 5 subjects included in the analysis and presented in relation to the housekeeping (GAPDH) and chemosensory cell marker (TH) genes. Additionally, we performed immunolabeling of fixed human CB tissue using the AMR-024 antibody. Punctate AMR-024 signal was contained within chemosensory glomus cells colabeled by UCHL1 (Ubiquitin carboxy-terminal hydrolase L1) (Figure 7B).³¹ Concordant with AMR-024 labeling results in animal tissues (Figure 2B and 2C), AMR-024 signal was not limited to the cell membrane, suggesting constitutive activity within the CB.

Figure 7.

MC4R (melanocortin 4 receptor) expression in human carotid bodies (CBs). A, MC4R mRNA expression in human CB detected by quantitative reverse transcription PCR (RT-qPCR). Expression presented as cycle threshold (Ct) values in comparison to housekeeping (GAPDH) and reference (TH) genes. All reactions were performed on the same plate. A and B, Contralateral sides from the same individual. n=5. B through E, AMR-024 immunolabeling in human CB. C and E correspond to areas outline in B and E, respectively. Chemosensory glomus cells are marked by UCHL1 (Ubiquitin carboxy-terminal hydrolase L1) immunoreactivity. Arrowheads indicate glomus cell clusters. n=1. Representative images were selected where chemosensory cells are best defined. Amp indicates amplitude; and DAPI, 4′,6-diamidino-2-phenylindole.

Collectively, our data indicated that in normotensive rats, Mc4r in the CB is sparsely expressed and exhibits marginal effects but exhibits a more dominant role in hypertension. To understand the molecular drivers of Mc4r upregulation in the SHR, we performed transcriptional regulatory network analysis on published CB transcriptome data.^13,32 A RTF (regulon transcription factor) can positively or negatively interact with its effector genes (Figure 8A). Hence, a minor alteration in RTF expression can significantly impact a specific function by influencing multiple downstream mediators. Out of 61 active regulatory networks identified (Data Set 1 in Supplemental Material), 24 regulons were found enriched in the CB of SHR by both 2-tailed GSEA and Master Regulator Analysis.³³ Notably, Ascl1 (Mash1) regulon, containing 64 effector genes was among the top enriched transcriptional regulatory networks (Figure 8B and 8C). Here, Mc4r emerged as a downstream target of Mash1 and ranked as the top upregulated gene within its transcriptional network (Figure 8D).

Figure 8.

Mash1 (mammalian achaete-scute homolog 1; Ascl1) transcriptional regulatory network is upregulated in hypertensive carotid bodies (CBs). A, Diagram illustrating a hypothetical regulatory transcriptional networks formed of a TF (transcription factor) and its effector genes. The TF representing the regulon can either have a positive or a negative interaction with its target genes, acting via direct (TF-target) or indirect (TF-TF-target) regulation. B, Twenty-four enriched regulon networks identified by 2-tailed gene set enrichment analysis (GSEA-2T) in the spontaneously hypertensive rat (SHR) CB. Color gradient represents regulon activity scores (dES) as described originally.^32,33 For each gene in a sample, differential gene expression is calculated from its expression in the sample relative to its average expression in the cohort; the genes are then ordered as a ranked list representing a differential gene expression signature in that sample, which is used to run the GSEA-2T returning dES for each regulon in each sample. C, GSEA-2T results showing Ascl1 (Mash1) regulon’s positive or negative targets (red/blue vertical bars) ranked by differential gene expression. D, Ascl1 (Mash1) regulon effector genes differentially expressed (DEG) in SHR CB as reported originally.¹³ Color gradient represents direction of gene expression change. Size of the point indicates gene expression levels based on average normalized reads across all samples. Ascl1 is plotted in red at the top. Alternating shapes (diamond, circle) indicate the order of genes; symbols listed on the left. WKY indicates Wistar-Kyoto Adj. p-value indicates p-value corrected for multiple comparisons; diff, significance of the difference between the pos and neg enrichment scores; LFC, log2 fold-change; neg, significance of negative enrichment; and pos, significance of positive enrichment.

Discussion

Our study demonstrates that Mc4r is expressed in the mammalian CB and that its expression is elevated in experimental hypertension. Consistent with these observations we show endogenous melanocortin agonist, α-MSH, activates CB chemosensory glomus cells to control CB chemosensory afferent output and motor reflex responses. Further, we show synthetic melanocortin agonists acting on the CB elevate resting and chemoreflex-evoked sympathetic and ventilatory activity. Supported by its regulation within the Mash1 (Ascl1) transcriptional network, we propose that melanocortin signaling may primarily represent a pathophysiological process that occurs in cardiometabolic disease states, contributing to autonomic imbalance.

We found a robust and consistent upregulation of Mc4r in the CB of the SHR compared with age-matched normotensive controls (Figure 1). This was linked to an increased number of α-MSH–sensitive CB chemosensory cells in the SHR (Figure 2J). This suggests a distinct chemosensory cell subpopulation involved in sensing α-MSH within the CB, consistent with the ribbon cable hypothesis of CB connectivity.¹² This is supported by an independent single-cell transcriptome analysis of mouse CB chemosensory cells, where a subset of chemosensory cells had detectable Mc4r expression (Figure S1B).²⁵ At the systems level, we found melanocortin agonism to primarily affect ventilatory and sympathetic responses, having no statistically significant effect on chemoreflex-evoked bradycardia (Figures 5 and 6). This again suggests a distinct CB afferent line to the nucleus tractus solitarius regulating selective reflex motor outputs. Moreover, this provides evidence that in disease states, such as essential hypertension, the abundance of CB chemosensory cells mediating a specific modality may be subject to alteration and, hence, druggable for therapeutic benefit.

We were unable to unequivocally visualize a distinct subpopulation of chemosensory cells defined by MC4R reactivity in intact tissue sections, which we deemed critical for definitive evidence (Figure 3). The contrasting results obtained by the MC4R localization experiments may primarily relate to (1) low protein abundance within the CB, and (2) limitations of existing tools available globally. At the transcript level, although consistently detectable within the CB, Mc4r exhibits low expression in the normotensive condition,^13,23,25 as expected for a transmembrane GPCR.³⁴ Notwithstanding, our functional data demonstrated that melanocortin agonists to evoke profound effects on the arterial chemosensory function despite the apparent low protein abundance. We used 3 different knockout/knockdown validated, primary polyclonal antibodies targeting MC4R. Results obtained using these tools should be considered in terms of their affinity-specificity trade-offs. It is known that properties enhancing antibody affinity negatively impact antigen specificity.³⁵ Whereas antigens of antibodies exhibiting higher specificity are significantly longer.³⁶ Although both AMR-024 and the N-terminal polyclonal antibodies target the N-terminal extracellular domain of the MC4R protein sequence, the N-terminal antibodies were designed against a considerably longer antigen sequence (15 versus 34 amino acid residues). This modification may translate to reduced affinity, impacting the ability of the antibody to effectively detect low protein levels, despite its demonstrated efficacy in the paraventricular hypothalamic nucleus (Figure S3).²⁶ Another possibility for contrasting AMR-024 and custom antibody results may be related to different MC4R posttranslational modification (glycosylation or proteolytic cleavage of the N terminus; phosphorylation of the intracellular domains) between tissues, impeding antibody-epitope access. Importantly, by using the antagonistic peptide label, HS-014-Sulfo549, we demonstrate that exogenous melanocortin ligands enter and accumulate within the CB when administered systemically (Figure 3F). Sulfonated rhodamine³⁷ used to generate HS-014-Sulfo549 is impermeable to the cell membrane; therefore, punctate fluorescence signal obtained likely corresponds to extracellular visualization of the bound protein. Notably, we could not assign HS-014-Sulfo549 signal specificity solely to α-MSH responsive dissociated CB chemosensory cells in vitro (Figure 3H through 3J). At the concentration used, HS-014-Sulfo549 may bind other targets when used on primary cells, notwithstanding gene expression data indicating sole expression of Mc4r in the CB (Figure 1B; Figure S1A). In sum, we deemed it critical to report the full set of MC4R detection results, because they may inform future studies on MC4R availability and tool specificity.

We have demonstrated that melanocortin agonism promotes sympathoexcitation by activating peripheral autonomic circuits. Our results demonstrate that exogenous α-MSH peptide and a small molecule tetrahydroisoquinoline, acting on the CB, not only promoted ventilation but also potentiated chemoreflex-evoked reflex sympathoexcitation (Figure 5). In contrast, Setmelanotide produced no statistically significant effect on sympathethic activity (Figure 6). This finding was unexpected, as the arterial chemoreflex has a powerful sympathoexcitatory effect sufficient to induce and maintain systemic hypertension.^5,9,10,14 MC4R is known to exhibit several unusual pharmacological properties, including agonist-dependent biased activation of downstream signaling pathways.³⁸ Setmelanotide exhibits a 100-fold higher activation of Gα_q-PLC (phospholipase C) pathway compared with α-MSH classically associated with Gα_s–adenylyl cyclase signaling.¹⁹ Adverse effects of melanocortin agonism, such as target-mediated tachycardia and pressor response, are classically associated with the hypothalamic Gα_s signaling.^20,21 The fact that Setmelanotide and tetrahydroisoquinoline lead to divergent sympathoexcitatory motor responses suggests biased activation of Gα_q-PLC pathway within the CB; this remains to be validated. Importantly, the WHBP used to investigate melanocortin agonism on arterial chemosensory function does not contain a hypothalamus ruling out sympathoexcitatory effects mediated by this structure. Furthermore, the action of Setmelanotide in our study was primarily confined to respiratory effects. Setmelanotide was previously shown to promote inspiratory flow, respiratory rate, and minute ventilation in obese mice.^39,40 We provide further evidence that Setmelanotide influences respiration in hypertension that involves peripheral action on the CB in addition to any aforementioned central effects. Importantly again, in the WHBP experimental setting, these effects were limited to enhanced peripheral chemoreceptor drive and changes in brainstem respiratory rhythm generation.

The pathophysiological mechanisms leading to CB sensitization in hypertension are not well understood but are known to involve disruption of hormonal and metabolic signaling pathways linked to energy metabolism. Both hypoglycemia and hyperglycemia significantly enhance the hypoxic ventilatory response in healthy volunteers⁴¹ while CB desensitization with hyperoxia attenuates hypoxic ventilatory response, highlighting CB involvement in sensing blood glucose.^42,43 CB chemoreceptors also respond to insulin, whereas hyperinsulinemia leads to CB overactivity in a model of metabolic syndrome.⁴⁴ Furthermore, CB denervation alleviates high-calorie diet–induced cardiovascular stress in the same model. Similarly, leptin-induced hypertension was shown to be abolished by CB denervation in C57BL/6J mice.¹⁴ Our recent study highlighted the role of the GLP-1R (glucagon-like peptide-1 receptor) in controlling arterial chemoreflex-mediated sympathetic drive.¹³ To explain the physiological significance of melanocortin signaling in the CB, we consider that it might be an integral part of the aforementioned endocrine and metabolic arterial chemosensory niche. Considering low expression of endogenous Pomc (Figure S1C), we suggest that the endogenous ligand α-MSH, which activates the system, likely originates from the intermediate lobe of the pituitary gland in response to stress,⁴⁵ and in the control of energy homeostasis.⁴⁶ Leptin administration to ob/ob mice has been demonstrated to stimulate the release of α-MSH into the circulation.⁴⁶ This suggests that leptin and α-MSH, acting on the CB, may converge to enhance lipolysis via increased sympathetic activity mediated via the arterial chemoreflex pathway, autonomous and independent from the central melanocortin system. Interestingly, the lack of a hemodynamic pressor effect from Setmelanotide in this context may be partly due to the lack of sympathoexcitatory effect evoked by the arterial chemoreflex axis reported in this study.

Although we primarily focused on arterial chemosensory cells, Mc4r expression in other cell types within the CB cannot be ruled out by our steady-state gene expression data (Figure 1; Figure S1).

Interestingly, we found HS-014-Sulfo549 fluorescent label accumulated in the vascular component; however, we think this reflects nonspecific labeling of the vasculature. Moreover, our live-cell Ca²⁺ imaging of dissociated CB chemosensory cells showed that, in the presence of SHU-9119, the intracellular response to α-MSH was retained in over 50% of chemosensory glomus cells where SHU-9119 led to a partial reduction in event mass (Figure S2). These data suggest that effects produced by melanocortin agonism within the CB may involve multiple targets.

Mash1 (Ascl1) is a classical pioneering transcription factor, meaning it can bind to condensed chromatin and open it up, making the DNA accessible for transcription.⁴⁷ This ability to remodel chromatin allows Mash1 to activate gene expression programs previously shown critical for the neogenesis and maturation of CB chemosensory cells⁴⁸ and neurogenesis of POMC and NPY neurons in murine ventral hypothalamus.⁴⁹ Furthermore, upregulation of Ascl1 was previously implicated in hypoxia-induced proliferation of CB glomus cells.⁵⁰ Our data suggest a primarily pathophysiological role of melanocortin signaling in arterial chemosensation. In the SHR, ectopic expression of Mc4r may be a transcriptional adjunct to Mash1 transcriptional reprogramming linked to CB expansion and arterial chemoreflex sensitization.^6,51 Demonstration of MC4R expression in human CB samples (Figure 7) further suggests that this mechanism may be pertinent to human disease concordant with CB expansion.^52,53 Together, this identifies molecular machinery involved in both arterial chemotransduction and arterial chemoreceptor neogenesis (expansion) for their potential therapeutic significance in managing aberrant sympathetic activity in hypertension. In our study, we identified 28 regulon networks enriched in the SHR (Figure 8B), which we anticipate will guide future research into the mechanisms underlying arterial chemoreceptor sensitization in cardiometabolic disease.

ARTICLE INFORMATION

Author Contributions

A.G. Pauza, D. Murphy, and J.F.R. Paton conceived and supervised the project. A.G. Pauza made the initial discovery, performed and analyzed the molecular and imaging (Figures 1, 3, and 8), and afferent carotid sinus nerve activity experiments (Figure 4), wrote the manuscript, and generated the figures with editorial input from D. Murphy and J.F.R. Paton. X. Shen performed and analyzed all Ca²⁺ imaging experiments (Figure 2). I. Felippe, P. Thakkar, A.G. Pauza, and F.D. McBryde performed in vivo experiments (Figure 4). P. Thakkar performed and analyzed working heart-brainstem preparation (WHBP) experiments using tetrahydroisoquinoline (Figure 5). I.S.A. Felippe and A.G. Pauza performed and analyzed WHBP experiments using setmelanotide (Figure 6). D.H. Pauza and A.G. Pauza oversaw human carotid body tissue collection and analysis (Figure 7). J. Broichhagen, K. Roßmann, and D.J. Hodson designed and synthesized HS-014–Sulfo549. M. Oya and K. Nakamura performed immunolabeling experiments using custom MC4R (melanocortin 4 receptor) antibodies (Figure 3; Figure S3).

Sources of Funding

The research support of the British Heart Foundation (FS/17/60/33474), the National Heart Foundation of New Zealand (NHFNZ) grant no. 2021 supported by Ernest Hyam Davis & Ted & Mollie Carr Legacies and No. 2026 supported by G.R. Winn Trust, and the Health Research Council of New Zealand (19/687) are acknowledged. P. Thakkar was supported by HRC-NZ explorer grant (22/629/A), NHFNZ grants (1976/3728668, 1959/3728667). I.S.A. Felippe was supported by NHFNZ grant nos. 2016 and 2025. M. Oya and K. Nakamura were supported by JST Moonshot R&D (JPMJMS2023 to K. Nakamura); AMED-CREST (JP23GM1910003 to K. Nakamura); and Ministry of Education, Culture, Sports, Science and Technology (MEXT)/Japan Society for the Promotion of Science (JSPS) KAKENHI (JP21K15343 to M. Oya and JP23H00398 to K. Nakamura). D.J. Hodson was supported by the Medical Research Council (MRC) (MR/S025618/1), Diabetes UK (17/0005681 and 22/0006389), and UK Research and Innovation (UKRI) - European Research Council (ERC) Frontier Research Guarantee (EP/X026833/1) Grants. This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (Starting Grant 715884 to D.J. Hodson). This project has received funding from the European Union’s Horizon Europe Framework program (deuterON, grant agreement no. 101042046 to J. Broichhagen). This work was supported on behalf of the “Steve Morgan Foundation Type 1 Diabetes Grand Challenge” by Diabetes UK and Steve Morgan Foundation (SMF) (grant number 23/0006627 to D.J. Hodson and J. Broichhagen). The research was funded by the National Institute for Health Research (NIHR) Oxford Biomedical Research Center. The views expressed are those of the author(s) and not necessarily those of the National Health Service (NHS), the NIHR, or the Department of Health. The project involves an element of animal work not funded by the NIHR but by another funder, as well as an element focused on patients and people appropriately funded by the NIHR.

Disclosures

D.J. Hodson and J. Broichhagen receive licensing revenue from Celtarys Research for provision of chemical probes. D.J. Hodson and J. Broichhagen have filed patents on cardiometabolic disease The other authors report no conflicts.

Supplemental Material

Supplemental Methods

Tables S1–S7

Figures S1–S7

Data Set 1

Reference 54

ARRIVE Guidelines

Major Resources Table

Supplementary Material

res-137-967-s001.pdf ^{(18.5MB, pdf)}

res-137-967-s002.pdf ^{(426.1KB, pdf)}

res-137-967-s003.xlsx ^{(17.8KB, xlsx)}