| |||||||||||||||||||||||||||||||
-adrenergic receptors.5 In this report, we provide evidence of the involvement of immune cells and one of their cytokine products in the formation of an adrenocortical adenoma in a patient with Cushing's syndrome. A striking lymphocytic infiltration of the patient's adrenocortical adenoma led us to examine the effects of interleukin-1 on the synthesis of cortisol by the adenoma cells and the expression of type I and II interleukin-1 receptors on these cells. Nearly all the biologic activity of interleukin-1 can be accounted for by type I receptors. Type II receptors are not signal-transducing but are probably cleaved enzymatically to yield soluble interleukin-1–binding protein, which inhibits interleukin-1 activity.6,7 We found that interleukin-1, but not corticotropin, increased the secretion of cortisol by the adenoma cells in vitro. In addition, type I interleukin-1–receptor protein and its messenger RNA (mRNA) were detected in the tumor cells.
Case Report
A 62-year-old woman was referred because of weight gain, hypertension, and hirsutism of three years' duration. Physical examination revealed an obese woman who weighed 75 kg and was 155 cm tall. Her resting pulse was 68 beats per minute, and her blood pressure was 160/95 mm Hg. She had facial plethora, hirsutism, telangiectasia, and ankle edema. The plasma cortisol concentration was 25 µg per deciliter (690 nmol per liter; normal, 7.2 to 18.1 µg per deciliter [200 to 500 nmol per liter]) in the morning and 19.8 µg per deciliter (547 nmol per liter) in the evening, and urinary cortisol excretion was 129 µg per day (357 nmol per day; normal, 7 to 36 µg per day [20 to 100 nmol per day]). The morning plasma corticotropin concentration was undetectable (<4.5 pg per milliliter [1.0 pmol per liter]; normal, 9 to 52 pg per milliliter [2 to 11 pmol per liter]). The plasma and urinary cortisol values did not decrease in response to dexamethasone. Computed tomography revealed a 2-cm mass in the right adrenal gland that was removed surgically, with no complications. The hypercortisolism and stigmata of Cushing's syndrome resolved after the removal of the tumor.
Methods
Tissue Processing
Tissue from the patient's tumor and from cortisol-producing adenomas from four other patients with corticotropin-independent Cushing's syndrome was transferred to prechilled phosphate-buffered saline, pH 7.6, and kept on ice until immunohistochemical studies (four tumors) and in vitro studies (two tumors) were performed. Tissue from three patients with adrenal carcinoma was also studied, along with normal adrenal glands from four patients who underwent nephrectomy for renal carcinoma.
Immunohistochemical Analysis
Adrenocortical, chromaffin, and immune cells were characterized immunohistochemically in paraffin-embedded sections of adrenal tissue with antibodies to chromogranin A (clone DAK-A3, Dako, Hamburg, Germany) and 17
-hydroxylase (courtesy of M.R. Waterman, Nashville), CD45 (clones 2B11 and PD7/26, Dako), and CD68 (clone KP1, Dako), as previously described.8 The presence of interleukin-1–receptor protein was evaluated by immunostaining with two mouse monoclonal antibodies to type I interleukin-1 receptors (clone C-20, Santa Cruz, Heidelberg, Germany, and Genzyme, Cambridge, Mass.). Bound antibodies were detected by the linked streptavidin–biotin–peroxidase method (Dako), and the enzyme reaction was visualized with 3-amino-ethylcarbazole (Dianova, Hamburg, Germany). Monoclonal mouse immunoglobulin was used as a negative control.
In Situ Hybridization
Single-stranded DNA antisense probes were amplified with the polymerase chain reaction (PCR) with 3' primers for type I or type II interleukin-1 receptors.9 Control sense probes were designed in the same manner, with 5' primers. All probes were purified with MicroSpin S-300 HR columns (Pharmacia Biotech, Heidelberg, Germany) and labeled with use of a reaction in which deoxyadenosine triphosphate and digoxigenin–deoxyuracil triphosphate were added to the 3' ends (Boehringer Mannheim, Mannheim, Germany). The digoxigenin-labeled probes were hybridized to cryostat-fixed sections of adrenal tumor tissue, and tissue-bound probe was stained with a monoclonal antibody to digoxigenin (Boehringer Mannheim). Sections of normal human adrenal tissue served as controls.
Isolation of RNA and Reverse-Transcription PCR
Total RNA was isolated from 5 million cells with the RNAeasy Miniprep Kit (Qiagen, Hilden, Germany). Complementary DNA (cDNA) was synthesized by reverse transcription (Ready-To-Go T-primed first-strand synthesis kit, Pharmacia Biotech) after treatment with deoxyribonuclease I (Boehringer Mannheim); cDNA for either type I or type II interleukin-1 receptor was amplified as previously described.9 The PCR products of the type I and type II interleukin-1 receptors were verified by digestion with MboII and MspI or MboII and EcoRI (Promega, Heidelberg, Germany), respectively. As a negative control, genomic DNA or RNA that was not reverse transcribed was processed in parallel.
Extraction of DNA and Analysis of Clonality
DNA was extracted from frozen tumor tissue and peripheral-blood lymphocytes from the patient as previously described.10 The human androgen-receptor gene locus A (HUMARA, GenBank accession number M21748) was used to identify the clonal origin of the adenoma.11
Cell-Culture Experiments
Specimens of adrenal tumors and normal adrenal tissue were mechanically dissected and digested enzymatically. After being depleted of macrophages with a monoclonal mouse antibody to CD68 (KP-1, Dako) and sheep antimouse immunoglobulin Dynabeads M-450 (Dynal, Oslo, Norway),12 the isolated cells were cultured in quadruplicate in 24-well plates (100,000 cells per well) for three days, as described previously.13 The cells were then washed and incubated with 10–8 M to 10–12 M human interleukin-1 (Endogen, Cambridge, Mass.) or 10–9 M corticotropin (Synacthen, Ciba–Geigy, Wehr, Germany) in serum-free medium for 6, 12, or 24 hours before cortisol concentrations were measured by radioimmunoassay (Biermann, Bad Nauheim, Germany).
Results
Histologic Examination
Microscopical examination of the patient's adrenal adenoma revealed a well-vascularized nodule that contained adenoma cells with no signs of malignancy and was heavily infiltrated by leukocytes.
Immunohistochemical Analysis
The patient's adenoma contained many cells that stained with antibodies to macrophage-specific CD45 and CD68 antigens, which are known to be a major source of interleukin-1 (Figure 1A). In contrast, few cells stained with these antibodies in the other four specimens of adrenal adenoma, the three samples of adrenal carcinoma, or the four samples of normal adrenal gland (Figure 1B and Figure 1C). Specimens from the patient's adenoma stained intensely with antibodies to type I interleukin-1 receptor (Figure 1D), whereas the tissues from the other adrenal adenomas and carcinomas and normal adrenal tissue did not stain with this antibody (Figure 1E and Figure 1F).
|
A strong signal indicating the presence of type I interleukin-1–receptor mRNA was detected throughout the patient's adrenal adenoma but not in normal adrenal tissue. In contrast, signals indicating the presence of type II interleukin-1 receptor mRNA were detected in perivascular regions of all samples of adrenal gland (data not shown).
Reverse-Transcription PCR
A reverse-transcription–PCR assay for the interleukin-1–receptor mRNA in cells isolated from the patient's adrenal tumor yielded the expected cDNA fragments of 299 bp for type I receptors and 391 bp for type II receptors, and enzymatic digestion of the PCR products confirmed the results (Figure 2). The results of this assay were negative in cells isolated from normal adrenal glands and the other adrenal adenomas. No PCR signals were obtained with these primers when either genomic DNA was used or RNA isolated from the patient's adenoma but not treated with reverse transcriptase was used (data not shown).
|
Hormonal Measurements
Corticotropin did not stimulate the secretion of cortisol in cultures of adenoma cells from the patient, whereas interleukin-1 caused a dose-dependent increase in cortisol secretion (P<0.001) (Figure 3A) that was also time dependent (Figure 3B). The combination of 10–9 M corticotropin and 10–8 M interleukin-1 stimulated cortisol production (P<0.001 for the comparison with basal secretion) no more than did 10–8 M interleukin-1 alone (P<0.001) (Figure 3A). In contrast, cultured normal adrenal cells responded more strongly to corticotropin (P<0.001 for the comparison with basal secretion) than to interleukin-1 (P=0.03 for the comparison with basal secretion) (Figure 3C). Similarly, cells from two other cortisol-secreting adenomas responded weakly to interleukin-1 (Figure 3C).
|
This patient's corticotropin-independent adrenal adenoma attracted our attention because of its massive infiltration with CD45- and CD68-containing leukocytes and macrophages, which are a major source of cytokines such as interleukin-1. In contrast to normal adrenal tissue and other adrenal adenomas or carcinomas, this adenoma expressed type I interleukin-1 receptors, and the cells of this tumor secreted cortisol in response to interleukin-1. In the normal adrenal tissue, interleukin-1 induced a slight increase in cortisol synthesis that was independent of corticotropin. Since interleukin-1 receptors were not detectable on normal adrenal cells, it is possible that this cytokine activates macrophages, fibroblasts, or adrenomedullary cells to secrete prostaglandins, corticotropin-releasing hormone, or other factors that then stimulate the synthesis and secretion of cortisol by adrenocortical cells.16
Infiltration of adrenal tissue by mononuclear cells has been noted in histologic sections from up to 15 percent of patients with Cushing's syndrome due to diffuse and nodular adrenal hyperplasia.17,18 A functional relation of these cells to adrenocortical hypersecretion was not suspected or studied, however.
In animals, interleukin-1 induces the proliferation and regeneration of adrenal tissue.19,20 Yet we did not find aberrant expression of interleukin-1 receptors on normal adrenal cells or cells from other cortisol-producing adrenal tumors. Therefore, it is unlikely that the expression of interleukin-1 receptors by our patient's tumor was merely the consequence of unrestrained tumor growth.
High local concentrations of interleukin-1 combined with aberrant expression of type I interleukin-1 receptors by a particular population of adrenocortical cells, followed by clonal expansion of the latter, may have provided the basis for tumor formation in this patient. The tumor cells were not responsive to corticotropin, and the strong expression of interleukin-1 receptors is consistent with the marked response to interleukin-1, leading to hypersecretion of cortisol. This case, in which aberrant expression of a cytokine receptor appears to be responsible for hormone secretion, expands the recent concept that corticotropin-independent adrenal Cushing's syndrome may be due to aberrant adrenocortical expression of receptors for hormones and neuropeptides.
Supported by grants from the Mildred Scheel Stiftung (10-1070-Re I; to Dr. Bornstein), Boehringer Ingelheim (to Dr. Willenberg), and the Deutsche Forschungsgemeinschaft (EH 161/1-1; to Dr. Ehrhart-Bornstein) and by a Heisenberg grant (to Dr. Bornstein).
Source Information
From the Department of Internal Medicine III, University of Leipzig, Leipzig, Germany (H.S.W., C.M.); and the National Institute of Child Health and Human Development (H.S.W., C.A.S., G.P.C., S.R.B.) and the National Institute of Mental Health (M.E.-B.), National Institutes of Health Clinical Center, Bethesda, Md.
Address reprint requests to Dr. Bornstein at the National Institutes of Health, NIH Clinical Center, NICHD, Bldg. 10, Rm. 10N262, Bethesda, MD 20892.
References
-adrenergic receptors in adrenal Cushing's syndrome. N Engl J Med 1997;337:1429-1434.
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
