Endothelin-1 and Its Receptors A and B in Human Aldosterone-Producing Adenomas
Abstract Endothelin-1 stimulates aldosterone secretion by interacting with specific receptors. Accordingly, we wished to investigate endothelin-1, endothelin-A (ETA) receptor, and endothelin-B (ETB) receptor gene expression, localization, and properties in aldosterone-producing adenomas and in the normal human adrenal cortex. We carried out 125I–endothelin-1 displacement studies with cold endothelin-1, endothelin-3, the specific ETA antagonist BQ-123, and the specific ETB weak agonist sarafotoxin 6 C and coanalyzed data with the nonlinear iterative curve-fitting program ligand. We also studied gene expression with reverse transcription–polymerase chain reaction with specific primers for endothelin-1, ETA, and ETB complementary DNA. Normal adrenal cortices from consenting kidney cancer patients (n=2) and aldosterone-producing adenomas (n=4) were studied; for the latter, surrounding normal cortex and kidney biopsy tissue served as controls. To further localize the receptor subtypes, tissue sections were studied by autoradiography in the presence and absence of 500 nmol/L BQ-123, 100 nmol/L sarafotoxin 6 C, and 1 μmol/L cold endothelin-1. In all tissues examined, endothelin-1, ETA, and ETB messenger RNAs were easily detected. However, in aldosterone-producing adenomas, both receptors’ genes were expressed at a higher level than in the kidney. In aldosterone-producing adenomas (F=9.49, P<.01) as well as in the normal adrenal cortex (F=8.57, P<.01), but not in adrenocortical tissue surrounding aldosterone-producing adenomas (F=5.08, P=NS), the significantly best fitting of binding data was provided by a two-site model indicating the presence of two receptor subtypes with density (Bmax) and affinity (Kd) similar to those previously found in other tissues. Autoradiography confirmed the presence of both ETA and ETB receptors on normal zona glomerulosa cells as well as on aldosterone-producing adenoma cells. Thus, the genes of endothelin-1 and of its receptors, ETA and ETB, are actively transcribed in the human adrenal cortex, and both receptor subtypes are translated into proteins in zona glomerulosa and aldosterone-producing adenoma cells. These data are consistent with an autocrine-paracrine role of endothelin-1 in the regulation of aldosterone secretion, both under normal conditions and in aldosterone-producing adenomas.
The very potent 21–amino acid vasoconstrictor peptide endothelin-1 (ET-1) has recently been shown to stimulate aldosterone secretion, both in vitro and in vivo,1 2 3 4 5 6 7 and to enhance corticotropin- and angiotensin II–stimulated aldosterone secretion.8 9 Because endothelial damage can turn on ET-1 synthesis and secretion, the peptide is a likely mediator of the hyperaldosteronism of several conditions in which enhanced ET-1 synthesis and endothelial damage coexist, including liver cirrhosis, congestive heart failure, preeclampsia, liver transplantation, endotoxemic shock, and primary and malignant hypertension.10 11 12 13 14 15 16 17 18 Of interest, the chronic infusion of ET-1 was found to raise plasma aldosterone concentration in rats and cause a notable hypertrophy of the zona glomerulosa (ZG) cells; furthermore, ZG cells isolated from the infused animals exhibited an enhanced basal production of aldosterone.19 20 These findings raise the hypothesis that ET-1 may be implicated in causing cell growth and enhanced aldosterone secretion in primary aldosteronism, a condition in which no known stimulus of ZG cell growth and enhanced aldosterone secretion has yet been identified.21
The physiological effects of ET-1 appear to be mediated by two different ET-1–specific receptors, ETA and ETB, which have been pharmacologically characterized.22 Autoradiographic evidence of ET-1–specific binding to the rat, porcine, and human adrenal cortex, as well as to cultured calf adrenal ZG cells, has been reported.2 23 24 25 26 Northern blot analysis demonstrated the expression of ETA and ETB receptors in homogenates of rat adrenals; furthermore, in hybridization experiments in situ, localization of the messenger RNA (mRNA) of ETA to the corticomedullary junction was observed, whereas ETB was found to be diffusely distributed throughout the adrenal cortex and medulla.27 However, in another study, the ETB receptor was detected immunochemically on the endothelial lining of capillaries around the ZG and in the zona fasciculata but not on the ZG steroidogenic cells of bovine adrenals.28 By taking advantage of the recent development of the specific ETA antagonist BQ-123 and of the ETB weak agonist sarafotoxin 6 C, we have recently provided evidence of the existence of both ETA and ETB receptors in the normal human ZG.29 This finding was further confirmed by autoradiography and gene expression studies, both with a reverse transcription–polymerase chain reaction (RT-PCR) on normal adrenal cortices and aldosterone-producing adenoma (APA) tissue30 and with Northern blot analysis of human adrenal glands of three patients with APA.31 However, the finding of ET-1 receptor expression on homogenates of APA tissue was not consistent with the functional observation that ET-1 stimulated in a dose-dependent fashion the secretion of aldosterone in vitro from normal adrenal cortices and from the cortex surrounding APA, but not from the tumors in patients with primary aldosteronism.32
Thus, our purpose was to investigate whether and where ETA and ETB receptors are expressed in APA and to assess their anatomic distribution and their binding properties compared with the normal adrenal cortex.
Preparation of Adrenal Tissues
We studied histologically normal adrenal glands, obtained at surgery from consenting patients undergoing unilateral nephrectomy for kidney cancer (n=2), and adenomatous adrenal glands of patients with Conn’s disease (n=4). For the latter tissues, the surrounding normal cortex and renal cortex biopsy tissues served as controls for the gene expression studies. After excision, tissues were immediately frozen in liquid nitrogen and stored at −195°C until they were used for ET-1 binding studies, nucleic acid extraction, and autoradiography. The protocol followed our institutional guidelines for the use of human tissue.
After homogenization, centrifugation, and resuspension of the cells in a Tris-HCl buffer, protein concentration was measured with a modified Lowry method. Membrane suspensions (15 to 25 μg protein) were then incubated with 25 pmol/L 125I–ET-1 (Amersham Laboratories; specific activity, 2000 Ci/mmol) in the absence and presence of increasing concentrations of unlabeled ET-1, ET-3, sarafotoxin 6 C33 (Sigma Aldrich), and BQ-123 (Peninsula Laboratories Inc), as already reported. The binding experiments were analyzed by the nonlinear iterative curve-fitting program ligand34 35 (Ligand, Biosoft) to establish the model that provided the significantly best fit (P<.05) by use of the F test and to obtain final parameter estimates of the dissociation constant (Kd) and receptor density (Bmax) values.
Gene Expression Studies
Total RNA was checked for integrity by gel electrophoresis and ultraviolet absorbance as reported.28 After reverse transcription, PCR amplification (GeneAmp RNA PCR Kit, Perkin-Elmer) was carried out, as previously reported in detail.30 To rule out the possibility of genomic DNA amplification, in some experiments the PCR was performed without prior reverse transcription of the RNA. The digoxigenin-labeled amplification products underwent size-fractionation on 1.5% agarose gel electrophoresis stained with ethidium bromide, followed by Southern blotting onto a nylon membrane, ultraviolet cross-linking (Stratagene UV-Crosslinker 1800, Stratagene-Duotech), and detection by chemiluminescence (DIG, Boehringer Mannheim), as previously reported.30
Frozen 10- to 15-μm sections of APA and normal adrenal cortices, immediately frozen in isopentane cooled in liquid nitrogen in the operating room, were cut in a cryostat (Leitz 1720 Digital) at −20°C and processed as reported previously.29 30 After preincubation, sections were labeled in vitro by incubation for 120 minutes with 100 pmol/L 125I–ET-1 at room temperature; nonspecific binding was determined by adding 1 μmol/L cold ET-1. Selective displacement of 125I–ET-1 was studied by adding 500 nmol/L BQ-123 or 100 nmol/L sarafotoxin 6 C. Reaction was terminated by washing of the samples three times in 50 mmol/L Tris-HCl buffer. After being rinsed in distilled water, the sections were rapidly dried, fixed in paraformaldehyde vapors at 80°C for 120 minutes, and then coated with NTB2 Kodak Nuclear emulsion (Eastman Kodak). The autoradiograms were exposed for 2 weeks at 4°C and were then developed with undiluted D19 Kodak developer. They were stained with hematoxylin-eosin and observed and photographed with a Leitz Laborlux microscope.
Results are reported as mean±SEM. Comparison between groups was performed by Student’s t test for unpaired data and the Mann-Whitney nonparametric test. Statistical analysis was performed with the spss/pc+ statistical package (SPSS Inc).
125I–ET-1 Displacement Binding
Cold ET-1 was the most potent in displacing 125I–ET-1, followed by sarafotoxin 6 C, ET-3, and BQ-123. In normal and APA tissue, both sarafotoxin 6 C and BQ-123 produced biphasic competition binding curves, suggesting the presence of two binding sites. Analysis of the saturation and displacement experiments showed Hill and pseudo-Hill coefficient values less than unity for all ligands, suggesting the presence of multiple binding sites. Further analysis revealed that the best fit was provided by a two-site model, as shown by the significant values of the F ratio34 both in normal adrenal cortices and in APA tissue but not in the normal adrenal cortex surrounding the tumors (Table⇓). The estimated Bmax values were similar in APA and normal cortices and were within the ranges reported for human tissues35 ; the estimated Kd values were similar to those measured in vitro in cells transfected with human ETA and ETB complementary DNA36 but were higher in normal cortices than in APA.
The RT-PCR consistently allowed detection of ET-1, ETA, and ETB mRNA in all adrenal specimens examined. An example of an ethidium bromide–stained 1.5% agarose gel is shown in Fig 1⇓. As can be seen, amplified complementary DNA fragments of the expected size for both the ETA and the ETB receptors and for the control β-actin gene were easily detected in the normal adrenal cortex and in the APA tissue. In the latter tissue, a notable difference is evident in the expression of the ETA and ETB receptors between both normal and APA tissue and the renal cortex, despite no evident difference in the expression of the β-actin gene.
Specific 125I–ET-1 binding was evident in all adrenals examined, both in the normal cortices and in the cortex surrounding tumors (not shown). In the tumors, it was mainly located in the capillaries and arterioles running among tumor cells. A more intense labeling of areas made of compact ZG-like cells than of areas of larger (lipid-laden) cells was observed (Figs 2⇓ and 3A⇓). The addition of an excess of cold ET-1 virtually displaced all 125I–ET-1 binding (Figs 2⇓ and 3B⇓). BQ-123 completely eliminated labeling in the vascular tunica muscularis (Fig 2C⇓ and 2E⇓), without apparently affecting 125I–ET-1 binding of areas of compact tumor ZG-like cells and of capillary endothelium (Fig 2C⇓ and 2E⇓). However, binding to larger clear zona fasciculata–type tumor cells was not affected (Fig 3C⇓ and 3E⇓). Sarafotoxin 6 C determined either a moderate or a marked attenuation of labeling of compact ZG-like tumor cells (Fig 2D⇓ and 2F⇓) and of light zona fasciculata–type tumor cells (Fig 3D⇓ and 3F⇓), respectively; in addition, it completely displaced binding to endothelium, while tunica muscularis remained well labeled (Figs 2D⇓, 2F⇓, 3D⇓, and 3F⇓).
Soon after their discovery, endothelins were deemed to play an important role in blood pressure regulation because of their multiple biological actions, including very potent and long-lasting vasoconstriction, mitogenesis, and stimulation of endothelium-derived relaxing factor, atrial natriuretic peptide, arginine vasopressin, and aldosterone release.1 4 In anesthetized dogs, the infusion of ET-1, aside from causing widespread hemodynamic effects, consistently increased plasma renin activity and aldosterone secretion.4 The latter seemed to be independent of the former, because it was also observed in vitro in dispersed ZG cells2 3 6 and even after blockade of angiotensin II formation.36 It appears to involve a Ca2+-dependent mechanism7 37 38 and prostaglandin synthesis, as shown in perfused slices of frog adrenal gland.7 ET-1 was also found to potentiate the aldosterone response to angiotensin II and corticotropin.8 9 26 39 Of interest, corticotropin has been found to increase the release of ET-1 from the adrenals, thereby suggesting that ET-1 is a mediator of corticotropin-stimulated aldosterone secretion.40 In vitro binding experiments on calf adrenal cultured glomerulosa cells have shown the presence of specific, saturable endothelin binding sites.2 23 24 Further studies on the same experimental model with the endothelin analogue sarafotoxin 6 B and with ET-3 have suggested the presence of a high-affinity and a low-affinity receptor.26 In an immunochemical study, however, no staining of steroidogenic cells of ZG with an ETB-specific antiserum was detected.27 With the use of specific antagonists for both the ETA and the ETB receptor subtypes and of RT-PCR with specific primers for each of the human endothelin receptor genes, we have recently demonstrated that both receptors are expressed as transcription and translation products in the normal human adrenal cortex.30 We were also able to localize them autoradiographically in the human ZG as well as to detect the ETA and ETB subtypes on the arteriolar tunica media and the endothelial lining of the sinusoids, respectively. This was in keeping with the results of Northern blot analysis and in situ hybridization studies of rat adrenal glands.28 The reasons for the discrepancy between immunohistochemistry and molecular studies is likely to be due to the lower sensitivity of the former compared with the latter methodologies.27
In this study, APA tissue was investigated in order for us to gain some insight into the possible autocrine-paracrine role of ET-1 in the excess aldosterone secretion and cell growth of this condition. The results of 125I–ET-1 saturation binding displacement experiments with a specific ETA antagonist and a specific ETB ligand show that both receptor subtypes are detectable in APA tissue (Table⇑). This is further confirmed by the results of gene expression studies, which in addition provide evidence of translation of the ET-1 gene in the tumor tissue (Fig 1⇑). However, autoradiographic studies demonstrate a marked heterogeneity of expression of the two receptor subtypes in the different tumors as well as in different areas and structures of the same tumor (Fig 2⇑). In fact, both ETA and ETB receptors were detected on compact tumor cells of one patient (Fig 2A⇑ through 2D), whereas only ETB receptors were found in other tumors made of large lipid-rich cells (Fig 2E⇑ through 2H). This finding has obvious implications both for gene expression and binding experiments and for functional studies. Of interest, in five patients with primary aldosteronism, Zeng et al32 reported that ET-1 stimulated the secretion of aldosterone in vitro from normal adrenal cortex as well as from the cortex surrounding APA, but not from tumor slices; they therefore suggested that the latter might be lacking in ET-1 receptors. Our finding of a marked heterogeneity of receptor distribution might explain the discrepancy between those negative functional results and the demonstration of both receptors with 125I–ET-1 displacement binding and gene expression studies. The latter are carried out on tissue homogenates and therefore provide averaged information on a bulk of different cells and structures of the tissue examined, including endothelium and arterioles’ tunica media. In contrast, functional studies on tissue slices are likely to be critically dependent on the topographical location of the section being investigated. Thus, although the concept of a physiological role of ET-1 in the paracrine regulation of aldosterone secretion is supported by the finding that genes for ET-1 and its receptors, ETA and ETB, are consistently expressed and translated into protein in the normal human adrenal ZG cells, as well as in the cortex surrounding APA, the picture may be different in the context of APA tissue. The mRNAs for the ETA and ETB receptors are easily detectable, and these receptors can be functionally measured in this tissue. However, the marked heterogeneity of expression of the two receptor subtypes between different tumors and even among different areas of the same tumor suggests the possibility of an autocrine-paracrine downregulation of ET-1 receptors due to local activation of ET-1 synthesis; this hypothesis needs to be further explored.
This study was supported by The Italian National Research Council (CNR)—Targeted Project “Prevention and Control of Disease Factors (FATMA)”: Subproject 8, Contract No. 91.00.218 PF41 115.06.654, and by Regione Veneto, Giunta Regionale, Ricerca Finalizzata, Venezia, Italia.
Presented, in part, at the 15th Scientific Meeting of the International Society of Hypertension, Melbourne, Australia, March 20-24, 1994.
Cozza EN, Gomez Sanchez CE, Foecking MF, Chiou S. Endothelin binding to cultured calf adrenal zona glomerulosa cells and stimulation of aldosterone secretion. J Clin Invest. 1989;84:1032-1035.
Miller WL, Redfield MM, Burnett JCJ. Integrated cardiac, renal, and endocrine actions of endothelin. J Clin Invest. 1989;83:317-320.
Delarue C, Delton I, Fiorini F, Homo-Delarche F, Fasolo A, Braquet P, Vaudry H. Endothelin stimulates steroid secretion by frog adrenal gland in vitro: evidence for the involvement of prostaglandins and extracellular calcium in the mechanism of action of endothelin. Endocrinology. 1990;127:2001-2008.
Cozza EN, Chiou S, Gomez Sanchez CE. Endothelin-1 potentiation of angiotensin II stimulation of aldosterone production. Am J Physiol. 1992;262:R85-R89.
Stewart DJ, Cernacek P, Costello KB, Rouleau JL. Elevated endothelin-1 in heart failure and loss of normal response to postural change. Circulation. 1992;85:510-517.
Hensen J, Levenson B, Schroder K, Jereczek M, Spielberg C, Schwenn K, Gross P. Plasma endothelin is increased in heart failure: no effect of a short infusion of atrial natriuretic factor [in German]. Z Kardiol. 1991;80(suppl 8):101-102.
Clark BA, Halvorson L, Sachs B, Epstein FH. Plasma endothelin levels in preeclampsia: elevation and correlation with uric acid levels and renal impairment. Am J Obstet Gynecol. 1992;66:962-968.
Widimsky JJ, Horky K, Dvorakova J. Plasma endothelin-1,2 levels in mild and severe hypertension. J Hypertens. 1991;9:S194-S195.
Sakurai T, Yanagisawa M, Masaki T. Molecular characterization of endothelin receptors. TIPS. 1992;13:103-108.
Koseki C, Imai M, Hirata Y, Yanagisawa M, Masaki T. Autoradiographic distribution in rat tissues of binding sites for endothelin: a neuropeptide? Am J Physiol. 1989;256:R858-R866.
Davenport AP, Nunez DJ, Hall JA, Kaumann J, Brown MJ. Autoradiographical localization of binding sites for porcine [125I]endothelin-1 in humans, pigs and rats: functional relevance in humans. J Cardiovasc Pharmacol. 1989;13(suppl 5):S166-S170.
Hagiwara H, Nagasawa T, Yamamoto T, Lohdi KM, Ito T, Takemura N, Hirose S. Immunochemical characterization and localization of endothelin ETB receptor. Am J Physiol. 1993;33:R777-R783.
Belloni AS, Rossi GP, Zanin L, Pessina AC, Nussdorfer GG. In vitro autoradiographic demonstration of endothelin-1 binding sites in the human adrenal cortex. Biomed Res. 1994;15:95-99.
Rossi GP, Albertin G, Belloni AS, Zanin L, Biasolo MA, Prayer-Galetti T, Bader M, Nussdorfer GG, Palù G, Pessina AC. Gene expression, localization and characterization of endothelin A and B receptors in the human adrenal cortex. J Clin Invest. 1994;94:1226-1234.
Molenaar P, O’Reilly G, Sharkey A, Kuc RE, Harding DP, Plumpton C, Gresham GA, Davenport AP. Characterization and localization of endothelin receptor subtypes in the human atrioventricular conducting system and myocardium. Circ Res. 1993;72:526-538.
Cozza EN, Gomez Sanchez CE. Mechanisms of ET-1 potentiation of angiotensin II stimulation of aldosterone. Am J Physiol. 1993;265:E179-E183.
Hinson JP, Vinson GP, Kapas S, Teja R. The role of endothelin in the control of adrenocortical function: stimulation of endothelin release by ACTH and the effects of endothelin-1 and endothelin-3 on steroidogenesis in rat and human adrenocortical cells. J Endocrinol. 1991;128:275-280.