| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology and Pathology (I.B., A.L., W.S., T.A.), Hôtel-Dieu du Centre Hospitalier de lUniversité de Montréal and Department of Biochemistry, Université de Montréal, Montréal, Canada H2W 1T7; Department of Endocrinology (P.C.), Centre Hospitalier Universitaire (CHU) Rangueil, 31403 Toulouse Cedex, France; and Unit of Genetics and Endocrinology (I.B., C.A.S.), Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Dr. Constantine A. Stratakis, Unit on Genetics and Endocrinology, Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Building 10, Room 10N262, 10 Center Drive, MSC1862, Bethesda, Maryland 20892-1862. E-mail: stratakc{at}mail.nih.gov.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Recently we described the usefulness of the Liddle test in diagnosing PPNAD (3). This test, as originally described by Liddle, consists of 24-h urine collection for the determination of 17-hydroxysteroids at baseline and after dexamethasone administration at doses of 0.5 mg every 6 h (low dose) for 2 d and 2 mg every 6 h (high dose) for another 2 d (4). When we performed this test in patients with PPNAD we found an unexpected, paradoxical rise in urinary glucocorticoids (both 17-hydroxysteroids and urinary free cortisol), which is rarely, if ever, seen in patients with other primary adrenocortical disorders (3). We have been using this response as a diagnostic test for the differentiation of PPNAD from other forms of bilateral adrenocortical tumors. A more than 50% rise of urinary free cortisol excretion on the second day of the high dose dexamethasone administration identifies PPNAD and distinguishes it from, for example, ACTH-independent macronodular adrenal hyperplasia (AIMAH) (3). To date, it is unclear whether this in vivo phenomenon could be replicated in vitro and thus directly be attributed to dexamethasone.
The molecular or pathophysiological mechanisms underlying these paradoxical responses to dexamethasone in PPNAD remain unknown. A recent report described that immunostaining for synaptophysin (SYN) distinguishes PPNAD nodules from the surrounding adrenal cortex (5). SYN is a marker for neuroendocrine cells, suggesting that perhaps affected cortical cells in PPNAD tissues assume unusual neuroendocrine properties (5), not unlike the situation in AIMAH. In the latter, cortisol secretion often results from the aberrant adrenocortical expression of neuroendocrine molecules, such as the receptors for gastric inhibiting polypeptide (GIP), vasopressin, catecholamines, LH, serotonin, angiotensin II, and leptin (6, 7, 8). Thus, the expression of ectopic or aberrantly regulated markers may be an explanation for some of the unexpected features of several adrenocortical tumors, including those of PPNAD, some common adenomas and adrenocortical cancer. For example, ectopic expression of the GIP receptor was recently demonstrated not only in AIMAH associated with food-dependent CS, but also in unilateral adrenocortical adenomas (9, 10).
Among the factors that have been proposed to regulate glucocorticoid synthesis in PPNAD are steroid receptors. This suggestion was supported by the study of a 33-yr-old woman with PPNAD, dexamethasone-induced cortisol rises, and periodic CS, which was exacerbated by pregnancy and the use of contraceptive pills; in vitro her adrenal cells were indeed stimulated by estradiol (11). Glucocorticoid receptors (GR) have been demonstrated by binding assays in the cytosols from adrenocortical carcinomas (and pheochromocytomas); their presence in normal tissue, cortisol-producing adenomas, and ACTH-dependent hyperplasias is, however, debated (12).
In the present investigation, we sought to replicate in vitro the phenomenon of dexamethasone-induced cortisol secretion in PPNAD and examine the presence of GR at the mRNA and protein levels in tissues affected by this disorder. Indeed, after studying a total of eight patients, we confirmed the direct stimulation by dexamethasone of cortisol secretion from adrenal PPNAD slices in vitro. GR mRNA amplification showed predominant expression of GR in PPNAD not only in the above tissues, but also in additional specimens (that were studied only by message analysis), and inconsistent presence in other adrenal lesions; these data were confirmed by immunohistochemistry. Immunoreactive GR was specifically present in PPNAD nodules and absent from the surrounding cortical tissue. Other normal or neoplastic adrenal samples showed weak, if any, immunoreactivity, suggesting that the GR is at least in part responsible for the dexamethasone-related increase in glucocorticoid secretion in this condition.
| Patients and Methods |
|---|
|
|
|---|
The review committees of the participating institutions approved this study, and written informed consent was obtained from all patients. The clinical profile of the investigated patients is given in Table 1
.
|
The protocol used to detect the presence of aberrant adrenal hormone receptors in patients 6 and 7 was described previously (13). In addition, urinary free cortisol levels were measured before and after the administration of ethinyl estradiol (50 µg/d for 4 d) in patient 7.
Screening for the presence of PRKAR1A mutations
Six patients have been screened for the presence of a mutation in the PRKAR1A gene (Table 1
), as we have described previously (14, 15); for two patients, no DNA was available for sequencing. The families included here and their mutations have been reported previously (15); the family designations are given in Table 1
.
In vitro dexamethasone studies
For these studies we collected adrenal tissues from eight patients: five PPNAD, two cortisol-producing adenomas, and one AIMAH (left and right sides). Investigators and the case surgical pathologist separated fragments of tissue (immediately after the diagnostic frozen biopsy); pathological samples were carefully obtained after separation from normal cortex and/or periadrenal fat and fibrous tissue and placed in culture medium for transport to the laboratory. There, adrenal tissues were minced and cultured in DMEM supplemented with 15% heat-inactivated FBS and 1% glutamine and antibiotics (Life Technologies, Inc., Gaithersburg, MD). The tissues were kept at 37 C in 5% CO2 humidified atmosphere. After cellular attachment (45 d), the medium was removed, and dexamethasone (10-6 M) was added to a total of 15 ml newly made medium, which was supplemented with 10% cortisol-free, heat-inactivated, bovine serum. Cortisol was measured in duplicate in 1 ml medium taken from the culture at 60 min, 18 h, and 24 h on the first day and then every 24 h over a total of 7296 h. At each sampling time the volume taken was replaced with an equal volume of the same medium (supplemented with 10-6 M dexamethasone). Control flasks were established that were treated the same way, but the medium was not supplemented with dexamethasone. The viability of the tissue was tested at the end of each experiment by washing the attached slices with 1x PBS twice and reincubating under the same conditions (without dexamethasone) for 2448 h; cortisol was measured again at baseline (24 h after the wash) and after this second incubation. Tissues that had undetectable levels of cortisol during that second incubation were not used in the present study.
RT-PCR
Total RNA was isolated from frozen tissues using TRIzol reagent (Invitrogen, Carlsbad, CA) and was further purified using RNeasy maxi kits (Qiagen, Valencia, CA) according to the manufacturers instructions. Single-stranded cDNA was synthesized by RT reaction using AMV reverse transcriptase (Roche, Indianapolis, IN). Briefly, 510 µg RNA were reverse transcribed using 400 pmol oligo(deoxythymidine)15 primer, 1x RT buffer, 10 mM dithiothreitol, 25 U ribonuclease inhibitor, 1 mM of each deoxy-NTP, and 50 U AMV reverse transcriptase enzyme at 42 C for 60 min. An aliquot of the reaction mix was heated to 95 C for 5 min and added to a PCR mixture containing 2.5 U Taq polymerase (Pierce Chemical Co., Rockford, IL) and 20 pmol of sense and antisense primers in a thermal cycler (PerkinElmer/Cetus, Norwalk, CT). The amplification of GR was performed with the following primers encompassing the ligand-binding domain of the receptor: sense, 5'-GCAACGTTACCACAACTCACC-3'; and antisense, 5'-TAGCTCTTGGCTCTTCAGACC-3'). The ß-actin control primers were: sense, 5'-TCACCCACACTGTGCCCATCTACGA-3'; and antisense, 5'-CAGCGGAACCGCTCATTGCCAATGG-3'. PCR reactions were performed for 35 cycles under these respective conditions: GR, 94 C for 60 sec, 54 C for 60 sec, and 72 C for 60 sec; and ß-actin, 94 C for 60 sec, 66 C for 60 sec, and 72 C for 60 sec. PCR products were electrophoresed in agarose gel and visualized by ethidium bromide staining. To confirm their GR origin, the bands were extracted from the gel and digested with the restriction endonuclease PstI (Roche, Indianapolis, IN), which generated the two expected fragments of 220 and 268 bp according to the GR sequence NM_000176 (data not shown).
Immunohistochemistry (IHC)
A total of 11 paraffin-embedded and frozen adrenal tissues were studied by IHC: 7 PPNAD tumors, 1 normal adrenal gland, 1 pheochromocytoma, 1 cortisol-producing adrenal adenoma, and 1 AIMAH. The tissues were immunostained using the labeled streptavidin-biotin system (LSAB-S, Dako, Carpinteria, CA). We used an antiserum to the GR raised against a 14-amino acid peptide of the amino-terminal domain of the human GR (16). Antibodies against SYN (mouse antihuman, Dako Corp., Hamburg, Germany) were used, also employing the LSAB-S. Tissues were screened by other markers (S-100, chromogranin-A, tyrosine hydroxylase) for diagnostic reasons and for the differentiation of medullary structures as described previously (5) (data not shown).
| Results |
|---|
|
|
|---|
All patients were diagnosed with CS and PPNAD by standard diagnostic testing. Their clinical characteristics are given in Table 1
. In vivo screening for the presence of aberrant regulation of cortisol secretion by membrane hormone receptors in PPNAD adrenals failed to show any significant variation of plasma cortisol to upright posture, mixed meal, TRH, GnRH, glucagon, vasopressin, or cisapride (data not shown). In addition, urinary cortisol secretion did not increase after oral administration of ethinyl estradiol in patient 7.
In vitro stimulation with dexamethasone
Dexamethasone caused significant stimulation of cortisol secretion in tissues from five patients with PPNAD who were studied after they had bilateral adrenalectomy. In four cases, maximum cortisol secretion was obtained with the last cortisol measurement of the experiment at 72 or 96 h. In patient 3, left and right adrenals were available and showed similar increases in cortisol levels at 72 h: an increase in the left adrenal from a baseline of 3.3 µg/dl to a peak of 45.7 µg/dl, and in the right adrenal, an increase from baseline of 3.2 to 37.5 µg/dl, respectively. In contrast, in the absence of dexamethasone no significant variation in cortisol levels was present: 0.9 µg/dl to a maximum value of 3.2 µg/dl (which occurred at 18 h over a 72-h observation period; Fig. 1A
). In addition, no cortisol increase was observed in two single cortisol-producing adenomas and in left and right adrenals of AIMAH tissues under dexamethasone stimulation (Fig. 1B
).
|
The expression of GR was examined by semiquantitative RT-PCR in five of the eight patients described here and in eight additional PPNAD tissues and five AIMAH samples (part of the data shown in Fig. 2
). The expected band of 488 bp was present in all PPNAD samples (Fig. 2
). After digestion with the restriction endonuclease PstI, all amplicons produced the expected fragments of 220 and 268 bp according to the GR published sequence (data not shown).
|
IHC was performed in seven of the eight patients studied (adequate paraffin-fixed tissue was not available from one; Fig. 3
). In all PPNAD cases, cytoplasmic granular expression of SYN was observed in cells of PPNAD nodules, as previously described (Fig. 3C
) (5). IHC for the GR was specifically concentrated in the cytoplasm of cells of the PPNAD nodules and not in nonnodular cortical cells in all patients (Fig. 3
, A, B, D, and E). Interestingly, like SYN (5), GR stained all nodules, including some that were very small and were otherwise not differentiated from the surrounding normal cortical tissue. GR staining was weak in adrenomedullary cells. In normal adrenal tissue and tissues from a cortisol-producing adenoma and one case of AIMAH, GR staining was seen mostly in the medulla and not in the cortex (Fig. 4
). In a pheochromocytoma, stained as a control adrenomedullary tumor tissue, SYN was highly expressed, whereas GR immunoreactivity was only weakly present (data not shown).
|
|
| Discussion |
|---|
|
|
|---|
The presence of GR staining in normal medulla and pheochromocytoma was not surprising; Kontula et al. (12) reported measurable glucocorticoid-binding activity in pheochromocytomas. It should be noted that the antibody we used in this study does not cross-react with other steroid receptors; it is against the N-terminal domain of the human GR and has been characterized previously (16). This antibody appears to recognize steroid-bound and steroid-free GR receptors as well as the cytoplasmic and nuclear forms.
After synthetic glucocorticoid administration, adrenocortical suppression is mostly due to inhibition of the central part of the hypothalamic-pituitary-adrenal axis. Some studies have suggested that glucocorticoids may also directly inhibit the adrenal cortex (18, 19). The observation that dexamethasone administration suppresses steroidogenesis in human fetal adrenal before the hypothalamic-pituitary adrenal axis is established supports this hypothesis (20). It is unclear, however, whether this effect is through GR receptors, because dexamethasone may act directly on steroidogenesis as previously suggested by Brentano et al. (21). GR receptors have been identified by binding assays in normal adrenal cortical cells of rat (19), bovine (22), and sheep (23), but not in those from human adult cortex (12). In vivo studies conducted in patients with hypopituitarism showed that the response of cortisol to ACTH infusion under dexamethasone administration was not altered (12), suggesting that glucocorticoids may not directly suppress human adrenocortical tissue (12, 24).
Various factors may influence the regulation of GR expression. Previous data demonstrated GR down-regulation in adult rat liver tissue after the administration of glucocorticoids and its 2-fold up-regulation after the withdrawal of glucocorticoids following adrenalectomy (16, 25, 26). A study of patients with endogenous CS showed that GRs in peripheral mononuclear blood leukocytes have lower ligand affinity, but no differences in the number of binding sites (27). These data, if one were to generalize from other tissues, make quite noteworthy the observation in the present paper of an increase in immunoreactive GR in all PPNAD tissues despite the concurrent presence of hypercortisolism in all patients. Previous studies reported that endogenous and exogenous glucocorticoids may modulate steroidogenesis differently (28, 29). Indeed, the dexamethasone response observed in vivo may correspond to the difference between the effect of dexamethasone and the chronic exposure to endogenous glucocorticoids, whereas the affinity and action of the ligand may vary depending on its nature (endogenous or synthetic).
It could be postulated that the event involved in the pathophysiology of PPNAD occurs during embryogenesis when GR is highly expressed, and this phenotype may be maintained during progression of the disease. Indeed, Condon et al. (30), using cRNA probes against the human GR, found that GR mRNA was distributed ubiquitously in fetal adrenal tissues between 8 and 12 wk. In fetal sheep adrenals the number of binding sites for GR varied with gestational age, being highest on d 100 and at the end of the term, but then decreased in newborn and adult animals (23). Accordingly, during fetal rat lung development, GRs are refractory to down-regulation by dexamethasone until just before term (31).
It has been suggested that the cAMP signaling pathway may influence GR expression and the biological response to glucocorticoids (31). Indeed, cAMP increases cellular glucocorticoid-binding capacity, and GR mRNA and GR protein levels in rat hepatoma cells increase after treatment with forskolin (31). CNC has been mapped to chromosomes 2p16 (CNC2) (32) and 17q24 (CNC1) (33), and mutations of the PRKAR1A gene, coding for the type 1A regulatory subunit of protein kinase A, were identified in familial as well as sporadic CNC cases (14, 15, 34, 35). These mutations lead to an increase in protein kinase A activity in CNC tumors after cAMP stimulation (14). Although not all patients with CNC have protein kinase A-activating mutations, it is possible that cAMP signaling abnormalities underlie the pathophysiology of the disease in all affected kindreds. However, in the present study PPNAD and/or CNC patients with and without PRKAR1A mutations demonstrated in vitro and/or in vivo paradoxical dexamethasone responses and GR activation, indicating that PRKAR1A alterations are not necessary for these phenomena.
Other reasons for the GR-mediated increased glucocorticoid responses in PPNAD may be increased GR mRNA half-life due to cAMP and/or other factors, up-regulation of GR transcriptional coregulation, nuclear trafficking, and/or the effect of other steroid hormones (36). In one patient we explored the possibility of estrogen-dependent cortisol secretion, as previously described in a woman with CS due to PPNAD (11). This patient showed no significant variation in urinary cortisol secretion after receiving 4 d of ethinyl estradiol. The aberrant regulation of cortisol production by GIP, vasopressin, ß-adrenergic agonists, hCG/LH, or serotonin has been described extensively in AIMAH, but not in PPNAD (7). Indeed, two patients evaluated in this study failed to demonstrate any response to different stimuli tested in a protocol that included mixed meal, glucagon, posture, TRH, GnRH, cisapride, and vasopressin testing, such as that used in AIMAH (7).
We conclude that the previously described increase in glucocorticoid secretion after the administration of dexamethasone in vivo is reproducible in vitro, suggesting that it is a direct effect of this steroid and is not mediated by another hormone. The increased presence of GR mRNA and immunoreactivity mainly in the hormonally active PPNAD nodules suggests that this is a GR-specific effect, although further studies need to be undertaken to unravel the molecular etiology of GR up-regulation.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations: AIMAH, ACTH-independent macronodular adrenal hyperplasia; CNC, Carney complex; CS, Cushings syndrome; GIP, gastric inhibiting polypeptide; GR, glucocorticoid receptor; IHC, immunohistochemistry; PPNAD, primary pigmented nodular adrenocortical disease; SYN, synaptophysin.
Received December 18, 2002.
Accepted April 24, 2003.
| References |
|---|
|
|
|---|
regulatory subunit in patients with the Carney complex. Nat Genet 26:8992[CrossRef][Medline]
This article has been cited by other articles:
![]() |
L. Groussin, A. Horvath, E. Jullian, S. Boikos, F. Rene-Corail, H. Lefebvre, F.-L. Cephise-Velayoudom, M.-C. Vantyghem, P. Chanson, B. Conte-Devolx, et al. A PRKAR1A Mutation Associated with Primary Pigmented Nodular Adrenocortical Disease in 12 Kindreds J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1943 - 1949. [Abstract] [Full Text] [PDF] |
||||
![]() |
A J Bauer and C A Stratakis The lentiginoses: cutaneous markers of systemic disease and a window to new aspects of tumourigenesis J. Med. Genet., November 1, 2005; 42(11): 801 - 810. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. F. Gunther, I. Bourdeau, L. Matyakhina, D. Cassarino, D. E. Kleiner, K. Griffin, N. Courkoutsakis, M. Abu-Asab, M. Tsokos, M. Keil, et al. Cyclical Cushing Syndrome Presenting in Infancy: An Early Form of Primary Pigmented Nodular Adrenocortical Disease, or a New Entity? J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3173 - 3182. [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 |