| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Third Department of Internal Medicine (T.M., Y.I., T.T., H.D., K.Y.), Department of General Medicine (H.M.), and Department of Urology (Y.T., T.D.), Gifu University School of Medicine, Gifu 500-8705, Japan; Matsunami General Hospital (N.Y.), Gifu 501-6062, Japan; Department of Pathology, Tohoku University School of Medicine (T.S., H.S.), Sendai 980-8575, Japan; and Division of Pediatric Endocrinology, University of Texas Southwestern Medical Center (P.C.W.), Dallas, Texas 75235-9063
Address all correspondence and requests for reprints to: Dr. Tomoatsu Mune, Third Department of Internal Medicine, Gifu University School of Medicine, 40 Tsukasa-machi, Gifu 500-8705, Japan. E-mail: mune{at}cc.gifu-u.ac.jp.
| Abstract |
|---|
|
|
|---|
Adrenal HSD11B2 may regulate levels of active and inactive glucocorticoids in the systemic circulation under these conditions, presumably by acting in an autocrine or paracrine manner. Nonfunctioning adenomas and those causing preclinical and overt Cushings syndrome may represent a continuum with clinical manifestations depending mainly on tumor size and HSD11B2 expression levels.
| Introduction |
|---|
|
|
|---|
Two isozymes of 11ß-hydroxysteroid dehydrogenase, HSD11B1 and HSD11B2, interconvert cortisol (F) and its inactive metabolite, cortisone (E), thus modulating glucocorticoid actions. HSD11B1 (5) is a widely distributed NADP(H)-dependent enzyme that is reversible in vitro, but predominantly catalyzes conversion of E to F in vivo, so that it potentially enhances glucocorticoid action. HSD11B2 was initially identified in mineralocorticoid target tissues and has only dehydrogenase activity (conversion of F to E); hence, it has been thought to be a protective mechanism to keep mineralocorticoid receptors from being occupied by F instead of aldosterone (6, 7, 8).
As the adrenal cortex expresses HSD11B1 (9) and HSD11B2 (10), we determined whether there were any variations in expression of these isozymes in adrenal adenomas that might affect circulating levels of F and E (i.e. active and inactive glucocorticoids) and thus explain some of the variation in clinical presentation of these tumors.
| Patients and Methods |
|---|
|
|
|---|
We examined 61 patients with adrenal tumors admitted to our department or related hospitals who underwent unilateral adrenalectomy during 19922000. Adenomas were classified ad hoc based on clinical characteristics, and their clinical and hormonal data are summarized in Table 1
with operative findings. Patients with clinically nonfunctioning adenomas had no signs or symptoms of hormonal excess, normal serum potassium levels, and plasma F levels suppressible after 1 mg (<83 nM; 3 µg/dl) and 8 mg (<28 nM; 1 µg/dl) dexamethasone. Patients with preclinical Cushings syndrome had no signs or symptoms of Cushings syndrome and normal basal plasma F levels, but had plasma F levels incompletely suppressible by 1 and 8 mg dexamethasone with low normal or suppressed plasma ACTH levels. These patients consented to unilateral adrenalectomy after being informed about possible malignancies in tumors bigger than 4 cm in diameter and the potential for complications. All patients with adrenal adenomas causing overt Cushings syndrome had signs or symptoms of Cushings syndrome, suppressed plasma ACTH levels, and high plasma F levels that were not suppressed below 138 nM (5 µg/dl) with 1 mg dexamethasone. Patients with aldosterone-producing adenomas had hypertension, hypokalemic alkalosis, elevated plasma aldosterone concentrations, and suppressed plasma renin activity. Two patients had aldosterone-producing adenomas with autonomous F secretion. All adrenal tissues were obtained at surgery, and the expected pathological findings were confirmed. Tumor volumes were calculated from measured three-dimensional diameters by estimating each tumor as an ellipsoid. As controls, normal adult adrenal (n = 7) and kidney (n = 5) tissues were obtained during surgical removal for renal cell carcinomas. These tissues were collected after obtaining written informed consent and stored at -80 C immediately after tumor resection until extraction of total RNA.
|
Because only a small amount of RNA could be obtained from some adrenal tissues, we employed quantitative competitive RT-PCR to measure the mRNA levels of HSD11B1 and HSD11B2 as described previously (11). Representative results of the competitive RT-PCR derived from a control kidney sample are shown in Fig. 1
. To minimize the effects of variations in deoxyribonuclease digestion and RT between samples, mRNA values were normalized for those of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a housekeeping gene, by presenting the mRNA levels of HSD11B1 and HSD11B2 as ratios to those of GAPDH in each sample.
|
Hormone and biochemical measurements
Peripheral blood samples for hormone examinations were drawn from fasting patients in the supine position after resting for at least 30 min in the morning between 08000900 h. Twenty-four-hour urine samples were collected in plain plastic containers without preservatives. Plasma levels of ACTH, F, renin activity, or aldosterone and urinary excretion of aldosterone or free F were measured by commercially available RIA kits.
In addition, plasma levels of cortisol and E were measured by ELISA in some subjects. Briefly, 100 µl plasma were extracted with 4 ml dichloromethane (mean recovery rate, 8590% for [3H]F or [3H]E), reconstituted in 50 µl buffer, and subjected to the following ELISA. Sheep polyclonal anti-F and anti-E sera were used at final dilutions of 1:30,000 and 1:100,000, respectively, and the avidin-biotin-peroxidase method was employed. The cross-reactivities of anti-F and anti-E serum were 4.2% and 3.4% against E and F, respectively. The intra- and interassay coefficients of variation, respectively, were: F, 8.7% and 12.5%; and E, 6.9% and 14.9%. The means of triplicate determinations were used in all calculations.
Analysis of data
Results were expressed as the mean ± SD. The Bonferroni/Dunn test was used to evaluate differences in mRNA levels among six adrenal tissue groups. P value less than 0.0033 was considered significant. The Wilcoxon signed rank test was used to compare mRNA levels between each tumor and the adjacent adrenal gland. The Mann-Whitney U test was used to evaluate the differences in the ratio of plasma E to F between groups. A linear regression analysis was employed where appropriate. In the multiple regression analysis, standardized regression coefficients (ß) and squared multiple correlation coefficients (r2) were calculated. P value less than 0.05 was considered significant.
| Results |
|---|
|
|
|---|
We first compared HSD11B2 mRNA levels (the ratio of HSD11B2/GAPDH x 10-3; mean ± SD) in adrenal glands and different types of tumors (Fig. 2
). The HSD11B2 mRNA level was 4.2 ± 1.9 in control adrenals (n = 7). This corresponds to approximately 5% of the expression levels in human kidney (81.7 ± 16.7; n = 5). Levels of HSD11B2 mRNA were 10.9 ± 11.0 in aldosterone-producing adenomas (n = 26), but only 1.1 and 2.1 in two tumors that secreted both aldosterone and F. Among all other adenomas, there was a striking trend toward increasing expression of HSD11B2 with lower plasma F levels: 12.1 ± 11.1 in F-secreting adenomas resulting in overt Cushings syndrome (n = 18), 26.1 ± 11.8 in adenomas with clinically autonomous F secretion resulting in preclinical Cushings syndrome (n = 8), and 35.0 ± 19.7 in clinically nonfunctioning adenomas (n = 7; Fig. 2
).
|
HSD11B1 mRNA content in adrenal tissues
The HSD11B1 mRNA level (expressed as the ratio of HSD11B1/GAPDH x 10-3; mean ± SD) was 4.0 ± 2.4 in control adrenals, which corresponds to about 15% of the expression level in human kidney (27.2 ± 25.5). There were no differences in HSD11B1 mRNA levels among normal glands and the different groups of adrenal tumors (3.1 ± 3.6 in aldosterone-producing adenomas, 2.0 and below the detection limit in the two adenomas secreting both F and aldosterone, 2.9 ± 2.9 in Cushings syndrome adenomas, 4.6 ± 4.6 in preclinical Cushings syndrome, and 2.4 ± 0.8 in clinically nonfunctioning adenomas).
HSD11B1 and HSD11B2 immunohistochemistry in adrenal tissues
HSD11B2 immunoreactivity was not detected in control adrenals. However, HSD11B2 immunoreactivity was detected in the cytoplasm of both compact and some clear cells in four of six preclinical Cushings syndrome adenomas examined (Fig. 3A
). HSD11B2 immunoreactivity was also detected in the normal adjacent adrenal cortex (this was especially marked in the zona glomerulosa) in two of three preclinical Cushings syndrome cases (Fig. 3B
). In addition, all three clinically nonfunctioning adenomas and one of two adjacent adrenal glands were positive for HSD11B2 immunoreactivity. Further, the HSD11B2 mRNA levels of the same tissue samples examined were higher in tissues with positive staining than in those with negative staining (36.7 ± 14.3 vs. 8.7 ± 9.2 x 10-3; P = 0.0003). HSD11B1 immunoreactivity was not detected in control adrenals or in any of the cases of preclinical Cushings syndrome or clinically nonfunctioning adenomas examined, but weak HSD11B1 immunoreactivity was detected in the adjacent nonneoplastic adrenal cortex, especially in the zona reticularis, in both clinically nonfunctioning adenomas.
|
The differences in HSD11B2 levels between the different groups of adenomas were intriguing. Because so-called nonfunctioning, preclinical and overt Cushings adenomas do not differ consistently in their histopathology (3, 4), even by immunohistochemical analysis of steroidogenic enzymes (4, 14), we hypothesized that these types of tumors form a continuum and that differences in HSD11B2 expression partially explain the differences in F secretion between these tumors. In other words, we postulated that F secretion would be inversely related to HSD11B2 level when all patients with these tumors were considered as a single group. Inversely correlated variables cannot be fitted to a linear equation unless the data are transformed either logarithmically or reciprocally (e.g. 1/y = ax + b, where a and b are constants); we chose the latter transformation.
Indeed, both plasma and urinary F levels (Fig. 4
, A and B; plotted as their reciprocals against HSD11B2 mRNA levels) were inversely correlated with tumor HSD11B2 mRNA levels in patients with these adenomas. Because the size of each tumor was directly related to plasma F (Fig. 4C
), but did not correlate with tumor HSD11B2 mRNA levels (Fig. 4D
), we employed a multiple regression analysis of plasma or urinary F (plotted as the reciprocal) vs. HSD11B2 mRNA levels and tumor volume (plotted as the reciprocal). Taken together, HSD11B2 level and tumor volume were able to predict more than 50% of the interindividual variation in plasma F levels (ß = 0.51 for HSD11B2 and 0.49 for tumor volume; r2 = 0.54; P < 0.0001).
|
|
| Discussion |
|---|
|
|
|---|
The initial report of cloning of HSD11B2 (10) noted the presence of HSD11B2 in the sheep adrenal cortex. HSD11B2 was also detected in rat (15, 16) and mouse (17) adrenal cortex. In humans, HSD11B2 immunoreactivity was demonstrated in the fetal zone of human fetal adrenal glands (18). Other evidence for HSD11B2 activity in the human adrenal cortex includes the secretion of both F and E from decapsulated normal human adrenocortical slices, NAD+-dependent dehydrogenase activity in adrenal microsomal preparations (19), and higher levels of E in adrenal vein than in inferior vena cava (20). However, previous studies (21, 22) measuring both E and F in adrenal, renal, hepatic, and systemic venous effluents suggest that the kidney is the main organ metabolizing F under normal conditions. In the present study, among patients with clinically nonfunctioning, preclinical, or overt Cushings adenomas, the pE/F (considered to be a measure of net whole body activity of HSD11B1 and HSD11B2) was highly correlated with tumor HSD11B2 mRNA levels. The simplest explanation of this finding is that when HSD11B2 is expressed at relatively high levels in the adrenal gland, it accounts for most conversion of F to E in an autocrine or paracrine manner due to the very high concentration of F within the adrenal cortex. However, we cannot rule out the possibility that a genetic mutation or an unknown humoral factor increasing HSD11B2 expression in the adrenal cortex adjacent to the tumor (see below) might also increase HSD11B2 expression in the kidney. The most interesting finding in our study was the high predictive power of a simple model for F secretion using just two factors: the volume of the tumor and the level of HSD11B2 expression. Thus, some adenomas might inactivate F at the site of production, resulting in apparently nonfunctioning adenomas. Similarly, the signs and symptoms of F excess in patients with adenomas causing preclinical Cushings syndrome might be masked due to the immediate inactivation of F, resulting in a prolonged clinical course until diagnosis. Considering the lack of histopathological differences between these tumors (3, 4, 14), one can even speculate that most of clinically nonfunctioning adenomas and those causing preclinical and overt Cushings syndrome are essentially of the same origin and differ mainly in HSD11B2 expression. If so, the term nonfunctioning adenoma is a misnomer in at least in some cases.
It is notable that the adrenal cortex samples adjacent to clinically nonfunctioning adrenocortical adenomas or to adenomas causing preclinical Cushings syndrome had higher HSD11B2 mRNA levels than control adrenals, suggesting a mutation of an as yet unidentified factor(s) common to both the adenoma and the remaining cortex. Alternatively, the involvement of an as yet unidentified humoral factor(s) up-regulating HSD11B2 expression might be considered.
In this context, understanding the factors regulating HSD11B2 expression might provide insight into the etiology and pathogenesis of these adrenal tumors. Restriction of dietary sodium was reported to decrease rat kidney HSD11B2 activity (23) and mRNA expression (24), whereas sodium loading (25, 26) and angiotensin II infusion (27) have been reported to have no effect on expression. There has been only one study of the regulation of adrenal HSD11B2, which showed up-regulation of HSD11B2 mRNA by forskolin, dibutyryl cAMP, or ACTH and down-regulation by phorbol ester in primary cultures of rat adrenocortical cells (28). With regard to the effect of glucocorticoids, the results of in vitro studies are conflicting; glucocorticoids increased HSD11B2 expression in a human endometrial cell line (29), but not in the T47D breast cancer cell line (30). Determining the importance of any of these observations for regulation of HSD11B2 in adrenal tumors requires further study.
With regard to HSD11B1, a previous report (9) described immunohistochemical detection in the zona reticularis > glomerulosa > fasciculata of human adrenal cortex. Probably due to the different antiserum used, we could not confirm immunoreactivity in normal adrenal cortex, but some positive staining was observed in adrenal cortex adjacent to nonfunctioning adenomas. When we analyzed the expression of HSD11B1 mRNA, absolute levels were lower than those of HSD11B2, and no difference was observed among normal adrenal glands or adrenal tumors. Considering the much higher Km value for F of HSD11B1 compared with HSD11B2, HSD11B1 expression appears to be less functionally important than HSD11B2 in adrenal tumors.
In conclusion, our results suggest that increased expression of HSD11B2 in clinically nonfunctioning or preclinical Cushings adrenocortical adenomas is associated with an absent or diminished phenotype of F excess through the immediate conversion of F to E within such tumors.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: E, Cortisone; F, cortisol; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HSD11B1, 11ß-hydroxysteroid dehydrogenase type 1; pE/F, ratio of plasma E to F.
Received August 13, 2001.
Accepted October 21, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. H. Bassett, B. Mayhew, K. Rehman, P. C. White, F. Mantero, G. Arnaldi, P. M. Stewart, I. Bujalska, and W. E. Rainey Expression Profiles for Steroidogenic Enzymes in Adrenocortical Disease J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5446 - 5455. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |