Regulation of Aldosterone Secretion by Substance P and the Neurokinin Type 1 Receptor in Aldosterone‐Producing Adenomas

Abstract

Background

Aldosterone‐producing adenoma (APA) is a major cause of primary aldosteronism, the most frequent form of secondary hypertension. Although somatic mutations in ion channels within APA have been shown to activate Ca²⁺ signaling and drive aldosterone production, the pathophysiology of primary aldosteronism remains partially understood. SP (Substance P), encoded by the TAC1 gene, is a neuropeptide of the tachykinin family, known for its role in stimulating aldosterone production through activation of the neurokinin 1 receptor (NK1R) in the human adrenal cortex. The aim of our work was to investigate the presence of SP nerve fibers and the NK1R in a large series of APA to assess the potential role of tachykinins in the pathophysiology of primary aldosteronism.

Methods

Using molecular, immunohistochemical, and functional techniques, 56 APA tissues were analyzed to assess the expression of SP and NK1R and their impact on aldosterone secretion.

Results

SP‐positive nerve fibers were detected in 90% of the APA tissues, localized both within and around the adenomas, which also showed strong NK1R expression. Functional studies revealed that SP stimulated aldosterone secretion in 6 of 10 APA cultures. The NK1R antagonist aprepitant inhibited SP‐induced aldosterone secretion in 3 of the 4 SP‐responsive APA cultures on which the antagonist was tested. Additionally, in perifused APA explants, SP influenced aldosterone pulsatility, resulting in enhanced mineralocorticoid secretion.

Conclusions

These findings suggest that the SP‐NK1R signaling pathway may contribute to APA pathophysiology and represent a novel potential target for the pharmacological treatment of PA in a subset of patients.

Nonstandard Abbreviations and Acronyms

APA

aldosterone‐producing adenoma

NK1R

neurokinin type 1 receptor

PA

primary aldosteronism

SP

substance P

Clinical Perspective.

What Is New?

• SP (Substance P), a neuropeptide of the tachykinin family, exerts a stimulatory action on aldosterone secretion from the human normal adrenal cortex, but its role in aldosterone‐producing adenomas had not been investigated.

• Aldosterone‐producing adenoma tissues contain SP‐positive nerve fibers and express tachykinin receptors, the neurokinin 1receptor (NK1R) being the predominant tachykinin receptor transcript; in a subset of aldosterone‐producing adenomas, SP stimulates aldosterone secretion via NK1R, as indicated by the inhibitory effect of the NK1R antagonist aprepitant on SP‐evoked aldosterone response.

What Are the Clinical Implications?

• SP–NK1R signaling represents a novel, renin–angiotensin system–independent mechanism regulating aldosterone secretion in a large subset of aldosterone‐producing adenomas; This mechanism may represent a valuable target for new pharmacological treatments of primary aldosteronism—especially NK1R antagonists such as aprepitant, which is currently approved for the treatment of chemotherapy‐induced nausea and may emerge as promising candidates for this novel therapeutic approach.

Primary aldosteronism (PA) is a common and potentially curable cause of secondary hypertension, characterized by autonomous aldosterone production independent of the renin–angiotensin system. ¹ Its 2 major causes are aldosterone‐producing adenomas (APAs) and bilateral adrenal hyperplasia, both associated with increased cardiovascular, renal, and cerebrovascular risks that can be reduced through surgical or pharmacological treatment. ¹ , ² , ³ To better define unilateral forms of PA and guide treatment, the international histopathology of PA consensus proposed a histopathological classification in 2021. Classical forms are defined by a solitary CYP11B2‐positive adenoma or a dominant aldosterone‐producing nodule, and nonclassical forms include multiple aldosterone‐producing nodules, micronodules, or diffuse hyperplasia. ⁴

Over the past decade, genomic studies have identified both somatic mutations in genes encoding membrane ion channels and ATPases in up to 90% of APAs and germline mutations in the less frequent familial forms of PA. ⁵ , ⁶ The affected genes that play a role in aldosterone hypersecretion include KCNJ5, ⁷ ATP1A1, ⁸ ATP2B3, ⁸ CACNA1D, ⁹ , ¹⁰ CACNA1H, ¹¹ CLCN2, ¹² , ¹³ and SLC30A1. ¹⁴ These molecular alterations result in Ca²⁺ signaling activation in adenoma cells with subsequent increases in both CYP11B2 expression and aldosterone production. Conversely, the role of the aldosterone‐driver gene mutations in APA development remains unclear, leading to the hypothesis that yet unidentified intracellular pathways may be involved in APA expansion. A possible role of the Wnt/β‐catenin pathway has been suggested, based on the observation that β‐catenin activation is common in APAs, although somatic β‐catenin mutations, which are usually responsible for activation of the pathway in many tumor types, are less frequently observed in APAs. ¹⁵ , ¹⁶ , ¹⁷

Numerous studies have shown that aldosterone production by APAs is not only activated by somatic driver gene mutations but is also regulated by abnormally expressed membrane receptors which confer to adenoma tissues an abnormal sensitivity to various hormones, neuropeptides and conventional neurotransmitters. ¹⁸ , ¹⁹ These illicit receptors include both overexpressed eutopic receptors which are physiologically present in the zona glomerulosa of the normal adrenal cortex like the serotonin type 4 receptor and ectopic receptors such as the luteinizing hormone/chorionic gonadotrophin receptor. ²⁰ , ²¹ In addition, aldosterone secretion often remains partly sensitive to physiological regulators such as angiotensin II and Adrenocorticotropic hormone, contributing to interindividual variability. ²² , ²³

SP (Substance P) is a member of the tachykinin family, which also includes NKA and NKB (neurokinins A and B), hemokinin‐1, and endokinins. These neuropeptides are involved in pain modulation, emesis, and gonadotropic function and participate in the pathogenesis of menopausal hot flushes. ²⁴ , ²⁵ , ²⁶ , ²⁷ We have recently observed the presence of SP in nerve fibers located in the subcapsular region of the normal human adrenal gland. ²⁸ After its release, SP can stimulate aldosterone production through a paracrine mechanism involving the activation of neurokinin type 1 receptor (NK1R), which is expressed by aldosterone‐producing cells in the zona glomerulosa. In agreement with this mechanism, administration of the NK1R antagonist aprepitant to healthy volunteers leads to a significant decrease in aldosterone production. ²⁸ The stimulatory effect of SP appears to be independent and complementary to renin–angiotensin system activation, with SP stimulating basal aldosterone production while the renin–angiotensin system triggers aldosterone synthesis in response to upright position, hypovolemia, and arterial hypotension. ²⁸ It is thus conceivable that SP could play a role in the regulation of APA‐associated aldosterone secretion which is independent of the renin‐angiotensin system.

Earlier immunohistochemical analyses reported sparse distribution of SP‐positive nerve fibers in APAs, a finding that has been interpreted as evidence against the involvement of SP in the pathophysiology of these tumors. ²⁹ However, more recent integrative omics studies have shown hypomethylation of the promoter region of the TACR1 gene, which encodes the NK1R, along with increased TACR1 mRNA expression in APA tissues, suggesting an enhanced sensitivity of adenoma cells to SP. ³⁰ , ³¹ Considering these contradictory results, we have decided to investigate whether the SP‐NK1R signaling pathway is present and active in APA tissues as a first step before clinical trials aimed at investigating the effect of NK1R antagonists on plasma aldosterone levels in patients with APA.

METHODS

The data that support the findings of this study are available from the corresponding author on reasonable request.

Patient Samples

We have investigated in vitro 56 APAs surgically removed from patients with unilateral PA referred to 2 French tertiary care centers specializing in adrenal diseases and hypertension, the Departments of Endocrinology and Nephrology of the University Hospital of Rouen, France, and the Hypertension Unit of European Hospital Georges Pompidou, Paris, France. The diagnosis of PA was established according to current guidelines and recommendations. ¹ , ² , ³ Clinical, biological, and genetic patient characteristics are summarized in Table S1. Adrenal samples were obtained at surgery and immediately immersed in culture medium for perifusion and cell culture experiments, frozen at −80 °C for reverse transcription‐polymerase chain reaction (PCR) analyses, or fixed in formalin for histological studies. Patients included in this study were recruited within the COMETE (COrtico‐ et MEdullo‐surrénale, les Tumeurs Endocrines) – HEGP (Hôpital Européen Georges Pompidou) protocol (authorization CPP 2012‐A00508‐35). Written informed consent was obtained from all patients for scientific study of adrenal specimens including genetic analyses.

Real‐Time Reverse Transcription‐PCR

Total RNA was extracted from adrenal tissue using Tri Reagent (Sigma–Aldrich, Saint‐Quentin‐Fallavier, France) and purified on Nucleospin RNAII (Macherey–Nagel, Hoerdt, France). Control specimens included polyA mRNAs from human adrenal, spinal cord, small intestine, and placenta (Clontech, Ozyme, Montigny‐le‐Bretonneux, France), as well as RNA from LAD2 and Caco2 cell lines (kindly provided by Dr D Metcalfe, National Institute of Allergy and Infectious Disease, National Institutes of Health, Bethesda, MD; and Dr Moïse Coeffier, Normandie Univ, UNIROUEN, INSERM, Rouen, France; respectively). cDNA synthesis was performed using ImProm‐II RT System (SensiFast, Cincinnati, USA). Real‐time PCR amplifications were performed with SYBR Green I Master Mix (Applied Biosystem, Courtaboeuf, France) on a QuantStudio 3 System (ThermoFisher, Illkirch, France) using gene‐specific primers (Table S2). Each sample was analyzed in duplicates, and cDNA quantification was normalized to PPIA (cyclophilin) using the ΔCt method and standard curves generated from polyA mRNAs or TACR1 human quantitative PCR template (HK210362, Origene, Rockville, USA).

DNA Isolation and KCNJ5 Sequencing

Tumor DNA from CYP11B2‐positive areas was extracted using QIAamp DNA midi kit (Qiagen, Courtaboeuf Cedex, France) or Maxwell 16 FFPE Plus LEV DNA Kit (Promega). KCNJ5 was amplified using intron‐spanning primers (Table S3). PCR was performed on 100 ng of DNA in a 25 μL reaction volume, containing 0.75 mmol/L MgCl₂, 400 nmol/L of each primer, 200 μM deoxynucleotide triphosphate, and 1.25 U Invitrogen Platinum Taq DNA Polymerase (ThermoFisher). The cycling conditions were initial denaturation at 95 °C for 5 minutes, followed by 30 cycles at 95 °C for 30 seconds, annealing at 60 °C, and extension at 72 °C for 1 minute with a final extension at 72 °C for 7 minutes. Direct sequencing was performed using the ABI Prism Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) on a Hitachi 3500 xL system (Rouen samples) and an ABI Prism 3700 DNA Analyzer (Paris samples).

Immunohistochemistry

Formalin‐fixed, paraffin‐embedded adrenal tissue sections were deparaffinized and subjected to antigen retrieval by heating at 95 °C for 20 minutes in either 10 mmol/L citrate buffer (pH 6) or Tris EDTA (pH 9). Tissue sections were treated with peroxidase blocking reagent (Dako Corporation, Les Ulis, France). Sections were then successively incubated with primary antibodies (Table S4) and anti‐immunoglobulin antibodies coupled to peroxidase. Immunoreactivities were revealed with diaminobenzidine (Dako Corporation, Carpinteria, CA, USA), and sections were counterstained with hematoxylin. Imaging was performed using an AxioScope 7 microscope (Zeiss) at PRIMACEN, the Cell Imaging Platform of Normandie, University of Rouen Normandie. NK1R expression was semiquantitatively evaluated by estimating the proportion of stained cells within each sample. An H score was assigned to each tissue based on the percentage of stained cells: undetectable (0, no expression), low (1, <30% positive cells), moderate (2, 30–70% positive cells), and high (3, >70% positive cells).

Primary Cell Culture

Ten different aldosteronomas were processed for cell culture experiments. After dissociation with collagenase type 1A and desoxyribonuclease 1 type 4 (Sigma–Aldrich), adrenocortical cells were cultured at a density of 10⁶ cells/ml in culture medium (50% in Dulbecco’s modified Eagle’s medium, DMEM; 50% Ham–F12; ThermoFisher Scientific, Illkirch, France) supplemented with 1% antibiotic–antimycotic solution, 1% insulin–transferrin–selenium solution (ThermoFisher), 10% fetal bovine serum (Sigma–Aldrich), and 10% horse serum (Eurobio‐Scientific). To prepare cells for stimulation, the medium was changed 24 hours before treatment, reducing fetal calf serum to 1%. Cells were then incubated in fresh DMEM (basal conditions) or DMEM containing increasing concentrations of SP (10⁻⁹ to 10⁻⁶ M; Sigma‐Aldrich) or NKA (10⁻⁷ M; Enzo Life Sciences, Lyon, France). SP was administered to cultured cells either alone or in combination with the NK1R antagonist, aprepitant (10⁻⁹ M; Selleck Chemicals; Houston, TX, USA) a concentration exceeding the reported IC₅₀ for NK1R (0.1–0.2 nM), thus ensuring effective receptor blockade under our conditions. ³² A detailed list of the substances tested in vitro on aldosterone secretion by APA explants is provided in Table S5. Incubation experiments were conducted in quadruplicate at 37 °C in a 5% CO₂–95% air atmosphere with 100% relative humidity for 24 hour.

Perifusion Experiments

Perifusion experiments were conducted on 5 different aldosteronomas using a previously described technique. ³³ Tumor explants were dissected into small fragments (1–2 mm³), rinsed 3 times with fresh DMEM and layered into perifusion chambers. Tissue slices were perifused at a constant flow rate of 300 μL/min with DMEM (pH 7.4, 37 °C), continuously gassed with a 95% O₂‐5% CO₂ mixture. Tissues stabilized for 2 hours before the administration of SP (10⁻⁶ M for 20 minutes; Sigma‐Aldrich). SP was dissolved in gassed DMEM and infused into the perifusion chambers at the same flow rate as DMEM alone by means of a multichannel peristaltic pump. Effluent perifusate fractions were collected every 5 minutes and immediately frozen until aldosterone assay.

Aldosterone Assay

Aldosterone concentrations in culture supernatants and perifusate fractions were quantified using a previously described radioimmunoassay procedure with specific antibodies and tritiated hormone ³⁴ (Perkin Elmer, Villebon‐sur‐Yvette, France) and a homogeneous time resolved fluorescence method (Cisbio, Bedford, MA, USA) according to the manufacturer’s protocol. Radioactivity was quantified by using a Tri‐Carb 4910TR scintillation counter (Perkin Elmer). Fluorescence was measured by using an Infinite F200 Pro microplate reader (Tecan, Switzerland). Aldosterone concentration was calculated using the sigmoidal standard curve interpolation with Prism 6.0 software (GraphPad Software, Dotmatics; San Diego, CA, USA). For perifusion experiments, aldosterone secretion was normalized to basal level to investigate the action of SP irrespective of spontaneous fluctuations of secretory activity. In this context, basal levels were defined as the average aldosterone secretion measured before SP stimulation, serving as a reference for evaluating changes in secretion. Relative changes were expressed as percentages of basal level. Radioimmunoassay and homogeneous time resolved fluorescence sensitivities were 80 pg/mL and 25 pg/mL, respectively. Cross‐reactivity of aldosterone antibodies with corticosterone, cortisol, testosterone, and Δ4‐androstenedione were < 0.05% for both assay methods.

Aldosterone Pulse Analysis

Aldosterone secretion pulse analysis was performed as previously described. ³⁵ , ³⁶ To evaluate dynamic changes in aldosterone secretion, key parameters were analyzed over 2 time periods: the basal phase before SP administration (basal, 0–60 minutes) and the SP‐stimulated phase after SP administration (SP 10⁻⁶ mol/L, 65–190 minutes). To account for differences in duration between these periods, the area under the curve was normalized per time unit and calculated using Prism (GraphPad Software). Mean aldosterone levels were determined as the average of all measured values. Aldosterone pulse peaks were defined as values exceeding the preceding nadir by ≥10%. Nadir aldosterone levels were calculated as the average of all nonpeak values. Pulse frequency was expressed as the number of pulses per hour. Mean pulse interval represented the average time (in minutes) between consecutive peaks. Pulse amplitude was calculated as the difference between each peak and the preceding nadir.

Statistical Analysis

Data were expressed as median±interquartile range, minima and maxima values, mean±SEM, or mean±SD, as appropriate. All statistical analyses were performed using Prism software (GraphPad Software). Statistical significance was assessed by t test, Mann–Whitney U test, or Kruskal–Wallis test and Dunn’s posttest after 1‐way ANOVA (with Welch’s correction when appropriate) or 2‐way ANOVA including interaction terms followed when relevant by Tukey’s or Dunn’s post hoc tests. Contingency analysis was carried out using the Fisher’s test. Univariate correlations were established using Pearson correlations. Exact p‐values are reported in the text and values <0.05 were considered statistically significant.

RESULTS

Expression of Tachykinins in APA

We have investigated the expression of the TAC1, TAC3 and TAC4 genes that respectively encode SP, NKB, and endokinins, by means of quantitative reverse transcription‐PCR analysis in a series of 42 APA samples. Our results revealed that TAC1 mRNA levels were significantly higher, compared with those of TAC3 and TAC4 mRNAs (P<0.0001), which were either undetectable or barely measurable (Figure 1A). Interestingly, we observed that the expression of the TAC1 gene was similar among both classical and nonclassical APAs (P=0.41), and irrespective of the presence of KCNJ5 somatic mutations (P=0.33). In addition, statistical analysis of the data shows a moderate positive correlation between APA diameter and TAC1 mRNA levels, with larger APAs being associated with higher TAC1 mRNA levels (R=0.41, P=0.0078) (Figure 1B). Among the series of 42 APAs, 7 tissues exhibited high TAC1 expression levels (TAC1 h) compared with the 35 other samples (TAC1 l) (P<0.0001; Figure S1). The 2 subgroups of tissues did not differ in age at surgery, sex distribution, KCNJ5 mutation status, nodule size, plasma aldosterone levels, or the proportion of classical versus nonclassical APAs (Figure S1). As depicted in Figure 1C through 1G, ∼90% of the resected tissues exhibited SP‐immunopositive fibers. SP‐positive nerve fibers were identified in adenomas and in the adrenocortical tissues adjacent to the adenomas. The distribution pattern of SP‐positive nerve fibers was similar in classical and nonclassical APAs. Granular SP‐immunoreactivity was present in the vicinity of adrenocortical cells, suggesting a potential neurocrine control of endocrine cells. Moreover, a granular cytoplasmic SP staining was detected within a subset of steroidogenic cells. Both SP‐positive cells and nerve fibers were observed in CYP11B2‐positive areas within adenomas and the adjacent tissues. Additionally, SP‐positive fibers and cells were observed within nerve ganglia and along the walls of blood vessels, respectively (Figure 1H).

Figure 1. Expression of tachykinins in APA.

A, Quantitative RT‐PCR analysis of TAC1, TAC3, and TAC4 mRNAs in 42 independent APAs. Each dot indicates 1 adenoma. Data are presented as mean±SEM. P<0.0001. B, Comparison of the expression levels of the TAC1 gene between APA subgroups: C, classical APAs and nonclassical APAs (left panel); NKD and KCNJ5 (middle panel), and correlation of the expression levels of the TAC1 gene with APA diameter (right panel). C through G. Immunohistochemical analysis of substance P in adenoma and adjacent adrenal tissue representative of n=56 independent APAs. C through F, Microphotographs showing a classical APA. C, SP‐positive longitudinal nerve fibers (arrow) and granulation in the immediate vicinity of steroidogenic cells in an APA region positive for CYP11B2. D, High magnification view showing granular SP staining at the periphery (arrow) and within steroidogenic cells (arrowhead). E, CYP11B2 immunostaining. The boxed areas highlight the zones of interest shown in (C) and (F) panels, exhibiting positive SP staining in both the adenoma (C) and adjacent tissues (F), respectively. F, SP‐positive fibers (arrows) surrounding a blood vessel in the adjacent adrenal tissue. G, SP‐positive fibers (arrows) in close contact with steroidogenic cells in the zona glomerulosa and zona fasciculata in the adrenal tissue adjacent to another classical APA. H, SP‐staining in the adjacent tissue to the adenoma detected in nerve fibers (arrows) and ganglia (G) in a nonclassical APA. APA indicates aldosterone‐producing adenoma; C, classical; KCNJ5, KCNJ5 mutation detected; NC, nonclassical; NKD, no KCNJ5 mutation detected; RT‐PCR, reverse transcription polymerase chain reaction; SP, Substance P; V, vessel; ZF, zona fasciculata; and ZG, zona glomerulosa.

Expression of Tachykinin Receptors in APA and Correlation With CYP11B2 Expression

Next, we investigated the potential effect of SP on aldosterone secretion in APA. The biological actions of tachykinins are known to be mediated by 3 types of G protein‐coupled receptors named NK1, NK2, and NK3, encoded respectively by the TACR1, TACR2, and TACR3 genes. Quantitative reverse transcription‐PCR analysis revealed various levels of TACR1, TACR2, and TACR3 mRNA expression. APAs displayed significantly higher expression of TACR1 mRNA compared with TACR2 and TACR3 mRNAs, TACR3 mRNA levels being nearly undetectable (P=0.012 and P<0.0001, respectively, compared with TACR1 mRNA levels). In addition, TACR1 gene expression was significantly higher in nonclassical APAs compared with classical APAs (P=0.045). TACR1 mRNA expression levels were similar in KCNJ5‐mutated and nonmutated adenomas (P=0.79). Within the nonclassical APA subgroup, TACR1 mRNA expression levels did not differ significantly between cases with 2 or more CYP11B2‐positive nodules (P=0.07) (Figures 2A and 2B). In order to evaluate potential different impacts of TACR1 and TACR2 on the pathophysiology of APA, we have examined the characteristics of different APA subgroups based on their receptor expression profiles. First, we did not observe any significant correlation between TACR1 and TACR2 mRNA levels in both classical and nonclassical APAs (Figure S2). We have also defined 3 APA subsets according to the TACR1/TACR2 mRNA ratio: TACR1 predominant expression (TACR1 p) ratio > 2, TACR2 predominant expression (TACR2 p) ratio<0.5, and balanced expression (TACR1 ~ 2) ratio between 0.5 and 2, respectively (P=0.0004; Kruskal–Wallis test) (Figure S3). The TACR1 p subset was exclusively composed of KCNJ5 nonmutated APAs. In contrast, the 3 subgroups of tissues did not differ in age at surgery, sex distribution, nodule size, plasma aldosterone levels, or the proportion of classical versus nonclassical APAs. The expression and distribution of NK1R in adrenal tissues were examined using immunohistochemistry in 56 APAs. Heterogeneous NK1R staining, observed in either the adenoma or the adrenal cortex adjacent to the adenoma, was present in 90% of cases, across both classical and nonclassical subtypes (Figure 2C through 2F and Figure S4). Out of the 56 APAs, 29 showed high NK1R staining, 9 had moderate staining, and 12 exhibited low staining in the adenoma, according to the H score based on the percentage of stained cells (Figure S5). In the adrenocortical tissues adjacent to the adenomas, NK1R‐positive staining was observed in the zona glomerulosa and nerve ganglia, as well as in the membranes and cytoplasm of the adrenocortical cells. Additionally, adjacent adrenal micronodules exhibited NK1R immunoreactivity in 27 of the 56 APAs. Interestingly, the comparison between the respective distributions of NK1 and CYP11B2 immunoreactivities within the tissues revealed that both proteins were detected in the same regions in 33 out of 56 cases. Among these adenomas, 22 were classified as classical APAs and 11 as nonclassical APAs, as shown in Figure 3 and Figure S6. However, we were unable to clearly define the APA subgroup showing similar distribution of NK1R and CYP11B2, because it was not associated with any specific genotype, histological form, or distinct clinical profile.

Figure 2. Expression of tachykinin receptors in APA.

A, Quantitative RT‐PCR analysis of TACR1, TACR2, and TACR3 mRNAs in 42 independent APAs. Each dot indicates 1 adenoma. Data are presented as mean±SEM. P=0.012; P<0.0001. B, Comparison of the expression levels of the TACR1 gene between different APA subgroups. classical APAs; nonclassical APAs (left panel) P=0.045; NKD and KCNJ5 (middle panel); CYP11B2‐positive nodules in nonclassical APAs (<2 nodules vs ≥2 nodules, right panel). C through F, Distribution of NK1R immunostaining in APA and adjacent adrenal tissue representative of n=56 independent APAs. C, E, F, NK1R staining in a classical APA and its adjacent adrenal cortex. C, Overall NK1R staining pattern in an APA and the adjacent adrenal cortex. D, Closer view of the NK1R staining in a different classical APA, highlighting areas of staining present in the zona glomerulosa or organized in micronodules within the adrenal tissue adjacent to the adenoma (arrows). E, Close‐up view of NK1R‐positive staining in the ZG and nerve ganglia in the adrenal cortex adjacent to the adenoma). F, High magnification view illustrating the distribution of NK1R staining in a group of adenoma cells. Arrow and arrowhead show membrane and cytoplasmic NK1R staining, respectively. APA indicates aldosterone‐producing adenoma; C, classical; Ca, capsule; G, nerve ganglia; KCNJ5, KCNJ5 mutation detected; NC, nonclassical; NKD, no KCNJ5 mutation detected; NK1R, neurokinin type 1 receptor; RT‐PCR, reverse transcription polymerase chain reaction; and ZG, zona glomerulosa.

Figure 3. Comparison of NK1R and CYP11B2 expressions in APA by immunohistochemistry.

A‐F, Comparison of NK1R and CYP11B2 immunostainings in the adenoma and adjacent adrenal tissue, illustrated by microphotographs representative of n=33 independent APAs. All panels depict a classical APA. A and B, Low magnification views of NK1R (A) and CYP11B2 (B) immunostainings in the same APA tissue. C and D: Closer view of NK1R (C) and CYP11B2 (D) staining patterns in adenoma and adrenal cortex adjacent to adenoma (Adjacent). E‐F: Higher magnification views showing NK1R (E) and CYP11B2 (F) staining in the zona glomerulosa and adrenal capsule in the adjacent tissue. APA indicates aldosterone‐producing adenoma; Ca, capsule; NK1R, neurokinin type 1 receptor; and ZG, zona glomerulosa.

Effect of Tachykinins on Aldosterone Production From APA Tissues

We investigated the effects of SP, which primarily binds to and activates the NK1 receptor, and NKA, which preferentially binds to and activates the NK2 receptor, on aldosterone production by APA cells in primary culture and perifused APA explants (Figure 4). The impact of SP on aldosterone production was examined in 10 independent APA cell cultures (Table S5). SP (10⁻⁷ M) stimulated aldosterone production by cultured cells derived from 6 APA (Patients: P1, P4, P5, P6, P8, P10; mean±SEM: basal: 100±4.9% [95% CI, 87.4–113.6] versus SP, 147.2±13.3% [95% CI, 97–173]; P=0.003) (Figure 4A). Among them, SP dose‐dependently stimulated aldosterone secretion in 1 case (P1; EC₅₀ [half maximal effective concentration] =1.71±0.3 nM, maximum efficiency 228.9±14.3% of basal level) (Figure 4B).

Figure 4. Effect of tachykinins on aldosterone production by APA cells in primary culture.

A, Effect of SP on aldosterone production by cultured APA cells (n=10 independent APA cell cultures, each performed in quadruplicate) (P=0.003). SP (10⁻⁷ M) significantly increased aldosterone production in 6 APA samples. B, Effect of increasing concentrations of SP (10⁻¹⁰ M to 10⁻⁶ M) on aldosterone production in 1 of the 6 SP‐sensitive APAs (from patient 1 [P1]), showing a dose‐dependent increase in aldosterone secretion (EC₅₀=1.71±0.3 nM, Emax 228.9±14.3% of basal level). C, Effect of SP (10⁻⁷ M) alone or in the presence of the NK1R antagonist aprepitant (10⁻⁹ M) on aldosterone production by cultured APA cells (n=3 independent cultures). The stimulatory effect of SP was blunted by AP (P=0.008). D, Effect of neurokinin A (10⁻⁷ M) on aldosterone production by cultured APA cells (n=2 independent cultures from patients 7 [P7] and 8 [P8]). NKA significantly increased aldosterone production in the APA cell culture derived from P7 (P=0.008) but did not significantly modify aldosterone release in the cell culture derived from P8 (P=0.084). Data represent the mean±SEM of the values obtained in independent experiments and are expressed as % of basal level. In all culture experiments, aldosterone secretion was normalized to basal levels. AP indicates aprepitant; APA, aldosterone‐producing adenoma; EC₅₀, half maximal effective concentration; NKA, neurokinin A; NK1R, neurokinin type 1 receptor; and SP, Substance P.

The effect of the NK1R antagonist aprepitant on SP‐induced aldosterone secretion was evaluated in 4 SP‐responsive APAs. The aldosterone response to SP was inhibited by aprepitant (10⁻⁹ M) in 3 cases (P1, P4, P5; basal: 100±9.5% [95% CI, 75.7–142.7] versus SP 10⁻⁷ M, 189.7±21.3% [95% CI, 107.5–249.4], P=0.002; SP (10⁻⁷ M) + aprepitant (10⁻⁹ M), 119±14.4% [95% CI, 75.2–180.3]; P=0.008 versus SP alone) (Figure 4C). The effect of NKA was studied in 2 independent APA cell cultures. NKA stimulated aldosterone secretion in 1 APA, which was unresponsive to SP (P7; basal: 100±9% [95% CI%, 77–120] versus NKA (10⁻⁷ M), 205.2±12.3% [95% CI, 180.6–233.7], P=0.008) (Figure 4D). In the APA unresponsive to NKA, SP alone significantly increased aldosterone secretion (P8; basal: 100±7.7% [95% CI, 82.6–113.6] versus NKA (10⁻⁷ M), 153.8±17.4% [95% CI, 122.9–200.2], P=0.08). The effect of aprepitant was not studied in this case. Among APAs, the responsiveness to SP was not associated with any particular histological form or clinical profile.

The kinetics of the aldosterone response to SP was further investigated using the perifusion technique in tissues derived from 5 different APAs. In all cases, aldosterone secretion spontaneously exhibited a pulsatile pattern (Figure S7). Analysis of the data showed that the mean aldosterone level after SP administration was not different from that observed in basal conditions (basal=100±5 versus SP=128±16.8% basal level; P=0.148) (Figure 5A). The nadir aldosterone levels remained similar during the 2 conditions (basal=100±4.9 versus SP=112.7±6.9% basal level; P=0.173) (Figure 5B). There were no statistically significant changes in pulse frequency (basal=2.7±0.67 versus SP=4±0.58 pulses/h; P=0.167) (Figure 5C), mean aldosterone pulse interval (basal=9.3±2.8 versus SP=11.6±1.9 minutes; P=0.52) (Figure 5D), or aldosterone pulse amplitude (basal=100±31.5 and SP=219.7±70.2% basal level; P=0.158) (Figure 5E). Conversely, the integrated aldosterone response, calculated as the area under the curve, showed a significant increase after SP administration (basal=100±22.8 and SP=252.8±46.9% basal level × minutes; P=0.032) (Figure 5F).

Figure 5. Effect of SP on aldosterone production by perifused APA tissues.

Aldosterone production kinetics in basal state and in response to SP (10⁻⁶ M, administered for 20 minutes) in 5 independent perifused APA tissues. Perifusion experiments were conducted over a total duration of 250 minutes. A, Integrated aldosterone response using the area under the curve. B, Mean aldosterone levels calculated as the average of all values during each condition. C, Nadir aldosterone levels, representing the lowest nonpeak values in each condition. D, Aldosterone pulse frequency expressed as the number of pulses per h. E, Mean aldosterone pulse interval, defined as the average time (in min) between consecutive pulse peaks. F, Pulse amplitude, calculated as the difference between each peak and its preceding nadir. For each tissue, the data were normalized to basal secretion levels to allow evaluation of the aldosterone response to SP irrespective of the variability of the spontaneous aldosterone production among APAs. Error bars represent the SEM. Data are expressed as % basal level. APA indicates aldosterone‐producing adenoma; and SP, Substance P.

DISCUSSION

Our team has recently demonstrated that SP, a neuropeptide localized in subcapsular nerve fibers of the human adrenal gland, stimulates aldosterone production via activation of the NK1 receptor, this regulatory mechanism being independent of the renin–angiotensin system. Based on these findings, we have hypothesized that SP and the NK1R may play a role in the control of aldosterone secretion by APA.

The predominant expression of TAC1 mRNA together with the low levels of TAC3 and TAC4 mRNA observed in APAs suggests that TAC1 is the primary potential source of SP production in these tumors. ³⁷ This pattern of tachykinin gene expression is similar to that observed in the normal adrenal gland. ²⁸ SP‐positive fibers were first visualized in the normal adrenal cortex by Heym et al. but were not precisely characterized. ³⁸ , ³⁹ Our previous work showed that intracortical SP‐positive fibers are independent of adrenergic and cholinergic fibers, adding further complexity to adrenal innervation. This finding is consistent with the known role of SP as a key neuromediator of the nonadrenergic, noncholinergic system, which is often considered as a third component of the autonomic nervous system although usually operating in synergy with the sympathetic system. ²⁸ , ⁴⁰ We now show the presence of SP‐containing nerve fibers in APAs, located both in direct contact with adenoma cells as well as around peripheral blood vessels. These localizations suggest that SP may regulate aldosterone secretion through a dual mechanism, including a direct action on steroidogenic cells and an indirect effect through modulation of adrenal blood flow. Indeed, as SP is a known vasodilator peptide, ⁴¹ its release near adrenal blood vessels may enhance adenoma perfusion. In addition, it is well established that the adrenal blood flow is a crucial determinant of steroidogenesis. As a matter of fact, increased adrenal blood flow, induced by vasodilatory agents like histamine, has been shown to enhance corticosteroid production in rats. ⁴² , ⁴³ , ⁴⁴ SP immunoreactivity was also detected in the cytoplasm of APA adrenocortical cells. The granular appearance of the staining suggests that the peptide is stored in secretory granules in agreement with the well‐established neuroendocrine differentiation of APA tissues. ⁴⁵ , ⁴⁶ , ⁴⁷ Collectively, these results provide a histological basis for an autocrine and neurocrine/paracrine regulation of aldosterone secretion by SP in APAs.

We found that TACR1 mRNA is the most abundantly expressed tachykinin receptor transcript in APAs, whereas TACR2 expression level is lower but consistently detectable and TACR3 mRNAs appear undetectable in the majority of samples. This contrasts with the normal adrenal gland, where TACR2 mRNA levels are barely detectable ²⁸ indicating that NK2R may be considered as an aberrantly expressed receptor in APAs. ⁴⁸ , ⁴⁹ Our findings are supported by a transcriptomic study conducted by our team, which revealed overexpression of TACR1 and TACR2 in 48 APAs compared with normal adrenal tissues. ³¹ Consistently, Itcho et al. identified TACR1 among the top G‐protein coupled receptor genes overexpressed in APAs relative to nonfunctioning adrenal adenomas, along with other receptors known to regulate aldosterone production such as melanocortin‐2 receptor, 5‐hydroxytryptamine receptor 4, and prostaglandin E receptor 1. ³⁰ This supports the hypothesis that the SP‐NK1R pathway may contribute to the complex landscape of aberrant G protein‐coupled receptor signaling in APA pathophysiology. Our immunohistochemical analysis of NK1R distribution corroborated the mRNA expression data by showing diffuse and heterogeneous NK1R labeling in both adenomas and adjacent adrenal cortex in most tissues. The proximity of NK1R‐expressing adrenocortical cells to SP‐containing nerve fibers, combined with the widespread distribution of NK1R within tumor tissue, suggests that SP may exert a direct influence on aldosterone production via NK1R. We have thus investigated the spatial distribution of NK1R and CYP11B2 in APA tissues. Immunohistochemical data show that the 2 proteins are detected in the same tissue areas in a subset of APAs, which represents the majority of the tumors studied. However, this subgroup of APAs does not seem to be associated with a specific genotype, distinct histological form, or any specific clinical characteristics.

The effect of SP on aldosterone secretion was then investigated using both primary cultures of APA‐derived cells and perifusion of APA fragments. The observed variability in the aldosterone response to SP probably results from the heterogeneous expression of the NK1R within APAs. As circulating SP levels in human plasma generally do not exceed 10⁻¹¹ M ⁵⁰ , a concentration insufficient to stimulate aldosterone secretion in vitro, SP regulation of aldosterone production by APAs likely involves peptide release from adrenal nerve endings or adenomatous cells. Under our experimental conditions, the maximum efficiency of SP on aldosterone secretion reaches ∼+50% of the basal level, a lower response compared with the 70% increase observed in normal adrenal tissue. ²⁸ This reduced responsiveness of APA tissues to SP might be due to receptor desensitization, ⁵¹ or partial saturation by endogenous SP in APA cells. ²⁷ The EC₅₀ of SP to stimulate aldosterone secretion was consistent with the known affinity of SP for NK1R, supporting the involvement of this receptor in SP‐induced aldosterone production in APA cells. ³² Accordingly, the NK1R antagonist aprepitant inhibited SP‐induced aldosterone secretion in the majority of the APAs studied, confirming NK1R involvement in a subpopulation of APAs, as in normal adrenocortical physiology. ²⁸ Remarkably, NKA, which binds to NK2R, significantly stimulated aldosterone secretion in one APA cell culture that was unresponsive to SP. This suggests that NKA, produced by the TAC1 gene alongside with SP, may regulate aldosterone production via the NK2R in some APAs. In addition, it cannot be excluded that SP may also activate the NK2R expressed by some APAs. In fact, although it is known that SP binds the NK2R only at high concentrations (≥10⁻⁶ M), the combined production of the peptide by nerve fibers and some adenoma cells may generate local concentrations capable of stimulating the NK2R signaling pathway. ⁵²

Our findings reveal that aldosterone secretion by perifused APA tissues exhibits spontaneous pulsatility, occurring thus independently of any external stimuli. Interestingly, Siragy et al. had demonstrated in vivo that plasma aldosterone concentration in APA patients shows regular pulses, including predominant peaks every 2 hours superimposed with additional micropulses appearing approximately every 30 minutes, a frequence relatively close to that of aldosterone pulses observed in our in vitro model (1 pulse every 20 minutes). ⁵³ It seems therefore likely that this microcyclicity reflects the intrinsic pulsatility of the APA secretory activity, as evidenced by the perifusion technique. The mechanisms underlying the APA intrinsic pulsatility may be multiple, including the expression of clock genes by adrenal cortical cells. ⁵⁴ , ⁵⁵ Local regulatory loops within the tissue, potentially governed by autocrine or paracrine signaling, may also contribute to this secretory pattern. Although the precise regulation of the in vivo aldosterone rhythmicity has not been extensively studied, similarities can be drawn with cortisol regulation. ⁵⁶ Cortisol rhythmicity is mainly controlled by the pituitary gland and the suprachiasmatic nucleus which plays a central role in the response to stress and light. There is also evidence suggesting that sympathetic innervation of the adrenal cortex contributes to this synchronization, functioning independently of pituitary control to coordinate central and adrenal clocks. ⁵⁷ , ⁵⁸ Thus, it is plausible that the nervous adrenal command mediated by SP plays a role in the secretory rhythm of APAs. Supporting this hypothesis, our perifusion experiments suggest that SP administration may influence the pulsatility of aldosterone secretion as evidenced by the increase of the area under the curve which reflects the global SP‐induced production of aldosterone. The fact that the different parameters of rhythmicity, that is, mean aldosterone level, nadir, pulse frequency, amplitude, and peak intervals, did not exhibit any significant changes after SP administration could be related to the limited number of perifused APAs together the well‐known heterogeneity of the samples.

Conclusions

In conclusion, our data indicate that tachykinins, especially SP expressed in APAs, may contribute to the stimulation of aldosterone production by adenoma tissues through an autocrine or neurocrine/paracrine mechanism mainly involving the NK1 receptor. The formal demonstration that SP actually exerts a stimulatory tone on aldosterone secretion in primary aldosteronism will necessitate in vivo studies. However, there is currently no validated animal model that fully recapitulates the distinctive pattern of innervation and paracrine regulation observed in APA. Especially, in rodents, at variance with the direct action of SP on mineralocorticoid production by the human adrenal cortex, the tachykininergic control of aldosterone secretion by the adrenal cortex is a complex indirect mechanism involving the medulla. ⁵⁹ Therefore, the next step will be to initiate clinical trials aimed at investigating the action of NK1R antagonists like aprepitant on plasma aldosterone levels in patients with APA.

PERSPECTIVES

Our findings open exciting avenues for exploring the SP‐NK1R pathway as a potential target in a subset of APA. NK1R antagonists such as aprepitant, which is currently approved for the treatment of chemotherapy‐induced nausea, emerge as promising candidates for a novel therapeutic approach of APA‐related primary aldosteronism. Although clinical validation is still needed, this approach offers a compelling direction that could ultimately expand the therapeutic arsenal for managing this condition.

Sources of Funding

This work was supported by the Institut National de la Santé et de la Recherche Médicale, the University of Rouen Normandy, and the Programme Hospitalier de Recherche Clinique which partially funded the COMETE network (Grant AOM95201). Additional support was provided by the Conseil Régional de Normandie, the European Regional Development Fund (Steroids project), and the Société Française d’Endocrinologie through the 2022 Research Award in Endocrinology.

Disclosures

None.

Supporting information

Tables S1–S5

Figures S1–S7