DAX-1 EXPRESSION IN HUMAN ADRENOCORTICAL NEOPLASMS: IMPLICATIONS FOR STEROIDOGENESIS
M. REINCKE, F. BEUSCHLEIN*, E. LALLI*, W. ARLT*, S. VAY*, P. SASSONE- CORSI*, B. ALLOLIO*
Division of Endocrinology, Department of Medicine II, University of Freiburg, 79106 Freiburg, FRG; Department of Medicine, University of Würzburg, 97080 Würzburg *; FRG; Institut de Genetique et de Biologie Moleculaire et Cellulaire, CNRS-INSERM-ULP, 67404 Illkirch, Strasbourg, France*
ABSTRACT: The DAX-1 gene encodes an orphan nuclear hormone receptor essential for normal fetal development of the adrenal cortex. Recently, DAX-1 has been shown to act as a transcriptional repressor of steroidogenic acute regulatory protein gene expression (StAR), suppressing steroidogenesis. We, therefore, investigated the expression of DAX-1 in a variety of adrenocortical tumors and compared the results with StAR mRNA expression. We found low or absent DAX-1 expression in aldosterone-producing adenomas (n=11: 35±11%; normal adrenals: 100±17%) and in aldosterone-producing adrenocortical carcinomas (n=2: 24 and 36 %). Cortisol-producing adenomas showed intermediate DAX-1 expression (n=8; 92±16), as did 3 non-aldosterone-producing carcinomas (72, 132 and 132 %). High DAX-1 expression was present in nonfunctional adenomas (n=3; 160±17%). In contrast to DAX-1, StAR mRNA expression did not show significant variations between groups. We did not detect the expected negative correlation between DAX-1 and StAR mRNA in adrenocortical tumors. These data suggest that high DAX-1 expression in adrenocortical tumors is associated with a non-functional phenotype whereas low DAX-1 expression favors mineralocorticoid secretion. These effects on steroidogenesis are mediated by mechanisms other than re- pression of StAR gene expression. Our results indicate that DAX-1 may be one of the factors influencing the steroid biosyn- thesis of adrenocortical neoplasms.
Mutations in the DAX-1 gene are responsible for congenital X-linked adrenal hypoplasia (1). Affected pa- tients suffer from adrenal insufficiency associated with hypogonadotropic hypogonadism (2). DAX-1 is an unusual member of the nuclear receptor superfamily of transcrip- tion factors. Its tissue-specific expression in the adrenal cortex, the testis, the pituitary and the hypothalamus (3) is highly regulated during fetal development suggesting a pivotal role in adrenal and gonadal maturation and func- tion (4, 5). DAX-1 binds DNA and acts as a transcriptional repressor of other nuclear hormone receptors (1). Recently, it has been shown that overexpression of DAX-1 in the Y-1 mouse adrenocortical tumor cell line suppresses StAR gene expression and thereby drastically decreases basal and forskolin stimulated steroid secretion (6). This suggests a prominent role for DAX-1 in the control of steroidogene- sis. In order to further elucidate the role of DAX-1 in adre- nal steroidogenesis, we investigated the expression of DAX-1 and StAR in adrenocortical tumors and correlated these data with the endocrine profiles of the tumors.
PATIENTS AND METHODS
We studied 4 normal adrenals, 3 non-functional adenomas (NFA), 8 cortisol-producing adenomas (CPA), 11 aldosterone-producing adenomas (APA) and 5 adreno- cortical carcinomas (ACC)(Table 1). The clinical and pathological diagnosis was made according to established criteria. Patients with non-functional adrenal adenomas had no signs or symptoms of hormone excess, normal serum potassium levels and a normal suppression of serum cortisol (below 3 µg/dl) after low-dose dexamethasone (2 mg). Patients with cortisol-producing adenomas had Cushing’s syndrome with cortisol concentrations non- suppressible by dexamethasone in the presence of sup- pressed plasma ACTH concentrations. Patients with al- dosterone-producing tumors had hypertension, hypoka- lemic alkalosis, elevated aldosterone concentrations and suppressed plasma renin activity.
Neoplastic adrenal tissue was collected with the approval of the Ethical Committee of the University Hospi-
tal of Würzburg. Only central portions of the tumors with- out hemorrhage or necrosis were used for RNA and protein extraction. Normal adrenal tissue was collected from the adjacent normal cortex of aldosterone-producing adenomas (n=4). After removing adjacent fat, the tissues were snap- frozen and immediately stored at -80 ℃ until processing. Poly-A-RNA was isolated from tissues using the Oligotex Direct Kit (Qiagene, Hilden, FRG). After electrophoresis hybridization was performed using PCR-generated 32P- CTP labeled human DAX-1 and StAR cDNA’s (1, 7).
| Tissue | N | Age (years) | Sex | Size (cm) |
|---|---|---|---|---|
| NFA | 3 | 35-70 | 3F | 5-7 |
| APA | 11 | 46-62 | 5F, 6M | 1-3 |
| CPA | 8 | 1-62 | 6F, 2M | 2-6 |
| ACC | 5 | 14-79 | 4F,1M | 7-12 |
| aldosterone-secreting: | 2 | |||
| cortisol-producing: | 1 | |||
| androgen-secreting: | 2 |
Resulting bands were quantified by scanning densitometry. For standardization the blots were rehybrid- ized with a mouse ß-actin cDNA probe. Whole cell protein extracts of normal and tumorous adrenal tissue (20 µg) were separated on SDS-PAGE and transferred to nitrocel- lulose. After blocking with dried milk blots were incubated with the monoclonal antibody 2F4 generated against a synthetic peptide corresponding to amino acid 135-166 of human DAX-1 (1:4000 dilution) and with the polyclonal antibody against Ad4BP/Steroidogenic factor-1 (8, 9). The immunoreactive protein was detected by the ECL method (Amersham, Braunschweig, FRG). All values are ex- pressed as mean±SEM. Significance of differences between groups was determined with a two-way ANOVA using the StatView 4.5 software on a PowerMacintosh computer.
RESULTS
DAX-1 mRNA expression normalized for ß-actin expression showed significant variations between the groups studied (p<0.01 by ANOVA). Its expression (normal adrenals: 100±17%) was high in the non- functional tumors (160±17%). In contrast, its expression was low or undetectable in APA (Fig. 1 and 2) and inter- mediate in CPA. As a group, ACC showed heterogeneous DAX-1 expression: Androgen-secreting carcinomas had high DAX-1 expression, whereas aldosterone-producing carcinomas had low DAX-1 expression similar to APA (Fig. 2).
Western blot confirmed these results showing increased expression of DAX-1 protein in NFA and re- duced expression in aldosterone-producing tumors. In two
APA, tumor tissue and the normal adjacent cortex could be studied in parallel (Fig. 3). DAX-1 mRNA and protein was clearly lower in the tumor tissue than in the normal adre- nal.
DAX-1 protein
180 114
84
58
48
38.5
26.6
DAX-1 mRNA
-28 S
18 S
B-actin
CARC
APC
APA
APA
CPA
NFA
NFA
StAR mRNA expression was detected, although at variable levels, in all tumors. We did not observe signifi- cant changes between tumor groups (Fig. 2, middle panel), and no correlation was observed between DAX-1 and StAR mRNA expression (Fig. 2, lower panel). Ad4BP/SF-1 expression was detected in tumor tissue and normal adrenals (data not shown) but did not correlate with endo- crine activity or DAX-1 expression, in accordance with previous data (10).
DISCUSSION
The main finding of our study is, that the orphan nuclear receptor DAX-1 is overexpressed in non-functional adrenal tumors, whereas its expression is low in aldoster- one-producing adenomas and carcinomas. This suggests
that DAX-1 is a key regulatory factor of steroidogenesis in the tumorous adrenal gland. Its expression may, therefore, influence the functional state and the endocrine profile of these tumors.
DAX-1 mRNA (%)
DAX-1 mRNA
200
150
p<0.01
100
50
0
NFA
CPA
APA
ACC
StAR mRNA (%)
StAR mRNA
120
100
80
p=ns
60
T
40
T
20
0
NFA
CPA
APA
ACC
140
StAR mRNA (%)
120
100
80
60
40
20
0
-10
40
90
DAX-1 mRNA (%)
140
190
In vitro DAX-1 acts as a transcriptional repressor suppressing the transcriptional activity of other nuclear receptors like RAR (1). More recently, we have shown that DAX-1 binds to hairpin secondary structures and blocks steroidogenesis via transcriptional repression of the StAR promoter in the Y1 mouse adrenocortical carcinoma cell line (6). The molecular mechanisms of DAX-1 mediated repression requires two cooperating domains, one located within helix H3 and the other within H12 (11). StAR pro- tein is essential for the cholesterol transport from the outer to the inner mitochondrial membrane, the rate-limiting step of steroidogenesis (7). This suggests that DAX-1 -
besides its role in the development and zonation of the adrenal cortex (3-5) - may be a major regulator of steroi- dogenesis in the adult adrenal gland. The Y-1 cell line is a functional rodent adrenocortical tumor cell line naturally deficient in 17a-hydroxylase and thus, it produces multiple steroids, but not cortisol. DAX-1 overexpression in this system suppressed basal and forskolin stimulated steroid synthesis (6). Our data are in accordance with a suppres- sive role of DAX-1 on steroidogenesis. We found high DAX-1 mRNA and protein levels in clinically and bio- chemically non-functional adrenal adenomas. Conversely, in aldosterone-producing adenomas and carcinomas, but not in cortisol-producing tumors, DAX-1 mRNA and pro- tein was low or absent supporting the concept that low DAX-1 expression is associated with preferential mineral- ocorticoid secretion. The mechanisms involved are not yet clear, but may involve specific interactions with steroi- dogenic enzymes of the mineralocorticoid pathway.
We did not find evidence for DAX-1 regulation of StAR expression in adrenocortical tumors. StAR mRNA was neither downregulated in NFA nor upregulated in functional adrenal tumors. The apparent discrepancy to the data obtained with the Y-1 cell line
DAX-1 mRNA
B-actin
Adenoma
Norm.
Adenoma
Norm. Adren.
Adren.
in vitro may be explained by the experimental model used in (6). Y1 cells lack endogenous DAX-1. Expression of DAX-1 was achieved by transfection with the pSG.DAX-1 expression plasmid (1) which may have generated higher DAX-1 levels than those present in the adrenal gland in vivo. Since DAX-1 upregulation in non-functional tumors was only moderate, we believe that DAX-1 modulate ster- oidogenesis in tumor tissue by repressing gene transcrip-
tion of other genes than StAR. In fact, recent evidence suggests that DAX-1 blocks steroidogenesis at multiple levels. For example, DAX-1 expression in Y-1 cells results also in the inhibition of the activity of the P450 side chain cleavage enzyme and 3ß-hydroxysteroid-dehydrogenase promoters (unpublished observation). The non-functional status of tumors with high DAX-1 expression is, therefore, most likely explained by potent effects on other key ele- ments of the steroid biosynthesis.
Several reasons for low DAX-1 expression in aldosterone-producing tumors should be considered. First, somatic mutations in the DAX-1 gene like deletions and rearrangements could result in reduced or missing expres- sion of this gene. More likely, low DAX-1 expression is induced by regulatory changes associated with tumori- genesis. Although DAX-1 gene regulation is still poorly defined one of the transcription factors implicated in DAX- 1 expression is Ad4BP/SF-1 (8, 9). Ablation of the Ad4BP/SF-1 gene by homologous recombination in mice resulted in a similar phenotype of adrenocortical insuffi- ciency and deficiency of pituitary gonadotropins than dele- tion of the DAX-1 gene (9). In addition, RT-PCR analysis indicates a tissue distribution pattern of mouse DAX-1 similar to that reported for mouse SF-1, including expres- sion in the hypothalamus, the pituitary and adrenal gland, the ovary and testis (3, 12). Recently, a putative SF-1 re- sponse element was identified in the promoter of the DAX- 1 gene (13). Coexpression of the DAX-1 promoter coupled with a luciferase reporter gene together with SF-1 into JEG-3 cells resulted in a 7-fold increase in DAX-1 pro- moter activity (14). These data suggest that SF-1 can be an important regulator of DAX-1 gene expression. On the other hand, promoter activity of the DAX-1 5’-flanking sequences did not require this potential SF-1 responsive element, and DAX-1 expression was unimpaired in knock- out mice lacking SF-1 (15), establishing that SF-1 is not required for DAX-1 gene expression in this setting. In adrenocortical tumors, SF-1 does not seem to be a regulator of DAX-1 since SF-1 gene expression was identified with equal intensity in all tumors independent of their endocrine activity and their respective DAX-1 expression in our series and in the series of Sasano et al. (10).
ACKNOWLEDGMENTS
This work was supported by a grant from the Wil- helm-Sander-Stiftung, München. We thank Dr. Morohashi for the kind gift of the Ad4BP antibody.
REFERENCES
1. Zanaria E, Muscatelli F, Bardoni B, et al. 1994. An unusual member of the nuclear hormone receptor su- perfamily responsible for X-linked adrenal hypoplasia congenita. Nature 372: 635-41.
2. Muscatelli F, Strom TM, Walker AP, et al. 1994 Mutations in the DAX-1 gene give rise to both X- linked adrenal hypoplasia congenita and hypogonado-
tropic hypogonadism. Nature 372: 672-6.
3. . Guo W, Burris TP, McCabe ERB. 1995. Expression of DAX-1, the gene responsible for X-linkd adrenal hypoplasia congenita and hypogonadotropic hypgo- nadism, in the hypothalamic-pituitary-adrenal/gonadal axis. Biochem Mol Med 56: 8-13.
4. Tamai KT, Monaco L, Alastalo T-P, Lalli E, Parvinen M, Sassone-Corsi P. 1996. Hormonal and developmental regulation of DAX-1 expression in sertoli cells. Mol Endocrinol 10: 1561-69.
5. Majdic G, Saundrs PTK. 1995. Differential pattern of expression of DAX-1 and SF-1 in the fetal rat testis. Endocrinology 137: 3586-89.
6. Zazopoulos E, Lalli E, Stocco DM, Sassone-Corsi P. 1997. DNA binding and transcriptional repression by DAX-1 blocks steroidogenesis. Nature 390: 311-315.
7. Clark BJ, Wells J, King SR, Stocco DM. 1994 The purification, cloning, and expression of a novel lu- teinizing hormone-induced mitochondrial protein in MA-10 mouse leydig tumor cells. J Biol Chem 269: 28314-22.
8. Morohashi K, Honda S, Inomato Y, Handa H, Omura T. 1992. A common trans-acting factor, Ad4- binding protein, to the promoters of steroidogenic P- 450s. J Biol Chem 267: 17913-17919.
9. Luo X, Ikeda Y, Parker KL. 1994 A cell-specific nuclear receptor is essential for adrenal and gonadal development and sexual differentiation. Cell 77: 481- 90.
10. Sasano H, Shizawa S, Suzuki T, et al. 1995 Ad4BP in the human adrenal cortex and its disorders. J Clin Endocrinol Metab 80: 2378-80.
11. Lalli E, Bardoni B, Zazopoulos E, Wurtz J-M, Strohm TM, Moras D, Sassone-Corsi P. 1997. A transcriptional silencing domain in DAX-1 whose mutation causes adrenal hypolasia congenita. Mol En- docrinol 11: 1950-60.
12. Ikeda Y, Shen WH, Ingraham HA, Parker KL. 1994. Developmental expression of mouse SF-1, an es- sential regulator of steroid hydroxylases. Mol Endocri- nol 8: 654-62.
13. Burris TP, Guo W, Le T, McCabe ER. 1995 Identi- fication of a putative steroidogenic factor-1 response element in the DAX-1 promoter. Biochem Biophys Res Comm 214: 576-81.
14. Yu RN, Ito M, Jameson JL. 1996 Transcriptional regulation of the mouse DAX-1 promoter by steroi- dogenic factor-1. The International Society of Endo- crinology, San Francisco, Program & Abstracts, Vol. II, 798.
15. Ikeda Y, Swain A, Weber TJ, Hentges KE, Zanaria E, Lalli E, Tamai KT, Sassone-Corsi P, Lovell- Badge R, Camerino G, Parker KL. 1996 Steroi- dogenic factor 1 and DAX-1 colocalize in multiple cell lineages: Potential links in endocrine development. Mol Endocrinol 10: 1261-72.