Expression of progesterone and estradiol receptors in normal adrenal ...

Endocrine-Related Cancer (2008) 15 465–474

Expression of progesterone and estradiol receptors in normal adrenal cortex, adrenocortical tumors, and primary pigmented nodular adrenocortical disease Patricia de Cremoux1, Dan Rosenberg2, Jacques Goussard 3, Catherine Bre´mont-Weil 4, Fre´de´rique Tissier 5,8,9,10,11, Carine Tran-Perennou1, Lionnel Groussin8,9,10,11, Xavier Bertagna4,8,9,10,11, Je´roˆme Bertherat 4,8,9,10,11 and Marie-Laure Raffin-Sanson6,7,8,9,10,11 1

Unite´ de Pharmacologie, Institut Curie, 26 rue d’Ulm, 75005 Paris, France Ecole Nationale Ve´te´rinaire d’Alfort, 94704 Maisons-Alfort Cedex, France Laboratoire d’Oncobiologie, Centre Francois Baclesse, Caen, France 4 Department of Pathology and 5Department of Endocrinology, Hoˆpital Cochin, Assistance Publique-Hoˆpitaux de Paris, Paris, France 6 Department of Endocrinology, Hoˆpital Ambroise Pare´, Assistance Publique-Hoˆpitaux de Paris, 9 Av Charles de Gaulle, Boulogne 92100, France 7 Faculte´ de Me´decine Paris-Ile de France Ouest, Universite´ de Versailles SQY, Versailles, France 8 INSERM, U567, Paris F-75014, France 9 De´partement d’Endocrinologie Me´tabolisme and Cancer, Institut Cochin, Paris F-75014, France 10 CNRS, UMR8104, Paris F-75014, France 11 Faculte´ de Me´decine Rene´ Descartes, Universite´ Paris 5, Paris F-75014, France 2 3

(Correspondence should be addressed to M-L Raffin-Sanson, Department of Endocrinology, Hoˆpital Ambroise Pare´, 9 Av Charles de Gaulle, Boulogne, 92100, France; Email: [email protected])

Abstract Adrenal tumors occur more frequently in women and are the leading cause of Cushing’s syndrome during pregnancy. We aimed to evaluate the potential role of sex steroids in the susceptibility of women to adrenocortical tumors. We evaluated the presence of the progesterone receptor (PR), estradiol receptors (ERs), and aromatase in 5 patients with primary pigmented nodular adrenal disease (PPNAD), 15 adrenocortical adenomas (ACAs) and adjacent normal tissues, 12 adrenocortical carcinomas (ACCs), and 3 normal adrenal glands (NA). The expression of PR and ERa was evaluated by enzyme immunoassays, real-time RT-PCR, immunohistochemistry, and cytosol-based ligand-binding assays. ERb and aromatase levels were evaluated by real-time RT-PCR. ERa concentrations were low in NA, in adrenal tissues adjacent to ACA (51G33), in ACC (53G78), and lower in ACA (11G11 fmol/mg DNA). Conversely, PR concentrations were high in NA and adrenal tissues adjacent to ACA, at 307G216 fmol/mg DNA, and were even higher in tumors – 726G706 fmol/mg DNA in ACA and 1154G1586 fmol/mg DNA in ACC – and in isolated PPNAD nodules. Binding study results in four tumors were compatible with binding to a steroid receptor. In patients with PPNAD, a strong positive immunohistochemical signal was associated with the sole isolated nodular regions. ERb transcript levels were very high in all samples except those for two ACCs, whereas aromatase levels were low. PR and ERb are clearly present in normal adrenal glands and adrenal tumors. Further studies may shed light on the possible pathogenic role of these receptors in adrenal proliferation. Endocrine-Related Cancer (2008) 15 465–474

Endocrine-Related Cancer (2008) 15 465–474 1351–0088/08/015–465 q 2008 Society for Endocrinology Printed in Great Britain

DOI: 10.1677/ERC-07-0081 Online version via http://www.endocrinology-journals.org

P de Cremoux et al.: Estradiol receptors in adrenocortical tumors

Introduction Unilateral tumors account for most cases of adrenocorticotrophin (ACTH)-independent hypercortisolism and may be classified into benign adrenal adenomas (ACAs) and malignant adrenocortical carcinomas (ACCs). ACCs have an extremely poor prognosis, with a survival rate of only about 20% at 5 years (Luton et al. 1990). Macronodular adrenal hyperplasia and primary pigmented nodular adrenocortical disease (PPNAD) result in tumoral proliferation affecting the entire gland. There is some clinical evidence to suggest that adrenocortical tumors are more prevalent in women than in men. For example, a sex ratio of 4.2 was reported for a series of 199 adrenal tumors (Luton et al. 1990). Moreover, ACC generally affects women at a younger age than men (Luton et al. 1990). The prevalence of adrenocortical tumors is particularly high during pregnancy, accounting for 70% of all cases of Cushing’s syndrome (CS) diagnosed in pregnant women (Guilhaume et al. 1992, Lindsay & Nieman 2006), versus only about 25% of cases of CS diagnosed in non-pregnant patients. A few cases of pregnancydependent CS have been reported elsewhere (Caticha et al. 1993, Close et al. 1993, Wallace et al. 1996, Kasperlik-Zaluska et al. 2000). We recently observed a worsening of hypercortisolism during pregnancy in a patient with PPNAD, followed by a clear improvement in this condition after spontaneous abortion. PPNAD is characterized by adrenal glands of the normal size containing small cortical pigmented nodules. It may occur in isolation or be diagnosed as a component of Carney complex (CNC) – a multiple neoplasia syndrome with spotty skin pigmentation, myxomas, and endocrine overactivity. PPNAD, like CNC, is transmitted in an autosomal dominant manner. Germline inactivating mutations of the gene encoding the protein kinase A (PKA) regulatory subunit type I (PRKAR1A) are found in some patients (Kirschner et al. 2000, Groussin et al. 2002c, Libe & Bertherat 2005), whereas other kindreds bear mutations in the gene encoding the phosphodiesterase (PDE11A4; Horvath et al. 2006). Like adrenal tumors, PPNAD occurs more frequently in female gene carriers than in male gene carriers: of the 64 patients diagnosed with PPNAD at the Department of Endocrinology and/or the Oncogenetics Laboratory of Cochin Hospital, 50 were females and 14 were males (sex ratio O3/1; unpublished results). The nature of the link between being female and adrenal proliferation and secretion is unknown. The ectopic expression of hormone receptors (including receptors for gastric inhibitory polypeptide, catecholamines, serotonin, and vasopressin type 1A) has already been 466

implicated in cortisol production by adrenal tumors and the macronodular adrenal hyperplasia responsible for CS. Such expression is responsible for food-dependent cortisol secretion and unexpected responses to posture or pharmacological agents (Groussin et al. 2002b, Bertagna et al. 2003). Luteinizing hormone (LH) receptor is present in normal adrenal glands, but is found mostly in the zona reticularis (Pabon et al. 1996). Overexpression of the LH receptor has recently been reported in some Conn adenomas (Saner-Amigh et al. 2006). A few cases of gonadotropin-dependent hypercortisolism have been reported (Bertherat et al. 2005). The stimulation of cortisol production by estradiol in cultured adrenal cells from one patient with pregnancyinduced CS has also been reported (Caticha et al. 1993). Estrogen and progesterone receptors (PRs) belong to the steroid nuclear receptor superfamily and commonly function as ligand-regulated transcription factors. They are involved in the normal developmental and reproductive functions and in disease processes, such as the development of cancers of the breast and other female reproductive organs (Graham & Clarke 1997, Rossouw et al. 2002). Estrogen signals through two subtypes of estradiol receptor (ER), ERa and ERb encoded by two different genes. The expression of PR and ER has been demonstrated in tissues known to be responsive to gonadal steroids: the uterus and the ovary, normal and neoplastic breast tissues, the brain, the pituitary gland, and the hypothalamus. It has also been described in tissues, in which the action of these hormones is less well established, including the vascular endothelium and osteoblast-like cells. ER is even more widely expressed. ERb and ERa seem to have different distributions in normal human tissues, but are frequently coexpressed in breast cancers, with ERb mRNA frequently less abundant than that for ERa. However, the respective roles of these receptors in cellular responses to estrogen remain unclear. ER expression has been demonstrated in the adrenal glands of rodents (Kuiper et al. 1997) and monkeys (Hirst et al. 1992), whereas PR was not detected in these species. During fetal development, large amounts of ERb mRNA are present in the definitive zone of the human adrenal cortex. In these tissues, ERb mRNA levels have been shown to be higher than those of ERa mRNA levels (Brandenberger et al. 1997). However, the expression of steroid receptors in normal human adult or tumoral adrenocortical tissues has not been investigated. We investigated the possible role of steroid sex hormones in favoring the development of adrenocortical neoplasms and/or controlling secretion from these tumors, by evaluating the presence of estradiol and PRs and of aromatase in benign and malignant www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 465–474 adrenocortical tumors and in PPNAD. We demonstrated the presence of high levels of PR in normal and tumoral adrenal tissues, and in isolated nodes from patients with PPNAD. ERb was strongly expressed in all samples tested, whereas ERa and aromatase were detected only in small amounts.

Patients and methods Patients and tissue collection

In this study, 5 patients with PPNAD and 27 patients with sporadic adrenocortical tumors were included. The hormonal investigations were performed as described elsewhere (Luton et al. 2000, Bertherat et al. 2005). For patient PPNAD 5, whose disease was discovered during pregnancy, paradoxical responses to LH and other hormones were sought 1 month after delivery. Adrenal tumors were classified by the same pathologist in all cases, according to the McFarlane classification, Weiss criteria, and molecular markers (Macfarlane 1958, Weiss et al. 1989, Gicquel et al. 2001). Three normal adrenal samples, consisting of tissue adjacent to an adrenal cyst or another non-secreting benign tumor, were also studied. Adrenal tissue was obtained during surgery and immediately frozen in liquid nitrogen. Informed consent was obtained for analysis of the PRKAR1A gene and adrenal tissue collection, as part of a COMETE network protocol approved by the Institutional Review Board and the Ethics Committee of Cochin Hospital. The NCIH295R cell line

The H295R human steroid-producing adrenocortical tumor cell line, which was derived from a malignant tumor, was cultured as described previously (Groussin et al. 2002a). RNA extraction and cDNA synthesis

Adrenal tissue collection, DNA and total RNA extraction, and reverse transcription were performed as described previously (Gicquel et al. 2001, de Cremoux et al. 2002). The cDNA Cycle kit (Invitrogen, Groningen, The Netherlands) was used for reverse transcription (RT). For mutation analysis, the 12 exons and the flanking intronic sequences of the PRKAR1A gene were amplified separately by PCR and directly sequenced on an automated sequencer, as described previously (Groussin et al. 2002c).

(PE Applied Biosystem, Foster City, CA, USA). PR, ERa ERb and aromatase primers, probe, and PCR conditions were as described previously (de Cremoux et al. 2002, 2004). RPLPO transcripts were also quantified as an endogenous reference RNA (Gicquel et al. 2001, de Cremoux et al. 2004). Accurate quantification was achieved by generating calibration curves, by serial dilutions of cDNA from the T47D and MDA-MB 231 human breast cancer cell lines, making it possible to quantify PCR efficiency for each run. Quantitative values were obtained from the cycle threshold (Ct) number, as described previously, using Applied Biosystems analysis software). The DCt values of the sample were determined by subtracting the average of duplicate Ct values for the target gene from the average of duplicate Ct values for the reference gene. The relative gene expression level was also normalized with respect to a positive calibrator, consisting of one of the samples used to generate the calibration curve (T47D dilutions) for the assay. Results are expressed as ‘N target’ values, determined as follows: Ntarget Z 2ðDCt sampleKDCt calibratorÞ . Breast tumors were considered ‘positive’ for PR and ERa (i.e., high probability of response to hormonal treatment) when ‘Ntarget’ values exceeded 50 arbitrary units (AU) and 250 AU respectively. They were considered positive for ERb when ‘Ntarget’ values exceeded 4.5 AU (corresponding to the ERb content of normal breast tissue; de Cremoux et al. 2004). ERa and PR protein assays

ER and PR protein assays were routinely performed on frozen samples, using an EIA (ER EIA, PR EIA; Abbott Laboratories), as described previously for breast tumor samples (Bernoux et al. 1998). The EIA kit for ER can detect Era, but not the ERb subtype. The EIA kit for PR can detect both the PRa and PRb isoforms. ER and PR levels are expressed with respect to tissue DNA status (fmol/mg DNA). The presence of significant amounts of ERa or PR in adrenal tissue was arbitrarily defined using a cutoff value determined for breast samples, 250 fmol/mg DNA. If this threshold is exceeded, breast tumors are considered ‘positive’ for ER or PR. This cutoff point is higher than the detection limit of the assay and, for breast cancer, corresponds to a clinically relevant threshold below which responses to endocrine treatment are poor (Bernoux et al. 1998). Immunohistochemistry for ER and PR

Real-time PCR

Real-time PCR amplification was carried out with the ABI Prism 7700 Sequence Detection System www.endocrinology-journals.org

Immunostaining for ER and PR was performed on paraffin-embedded adrenal tissue sections. The techniques and antibodies used (Nichols Institute 467

P de Cremoux et al.: Estradiol receptors in adrenocortical tumors Diagnosis, San Juan Capistrano, CA, USA) are routinely used for breast tumor analysis (Balaton et al. 1996). Antibodies against ER recognized ERa whereas antibodies against PR reacted with both isoforms of PR. Cytosol-based ligand-binding assays for ER and PR

Data are reported as meansG1S.D. Non-parametric statistical methods were used (Wilcoxon and Mann–Whitney tests) due to the non-normal distribution of variables and the small number of subjects in certain groups of interest.

Results

Scatchard curves were generated with tritiated [(16alpha-ethyl-21-hydroxy-19-nor-pregn-4-ene-3,20dione)-6,7-3H] (ORG-2058) for PR and tritiated estradiol for ER, as described previously (Goussard et al. 1989). Statistical analysis

Clinical and biological data were input into a Microsoft Access 98 computerized database.

Clinical and biological features

The clinical and pathological characteristics of the patients are shown in Table 1. All five patients with PPNAD were female and were referred for CS. Two had isolated adrenal disease and three were diagnosed with CNC on the basis of family history or typical clinical features (lentigine, heart myxoma, cafe´-au-lait spots). In one patient (PPNAD 5), CS was discovered during pregnancy, due to the onset of central obesity, edema,

Table 1 Clinical and biological features at presentation in the patients with primary pigmented nodular adrenocortical disease (PPNAD) and sporadic adrenocortical tumors

PPNAD 1 PPNAD 2 PPNAD 3 PPNAD 4 PPNAD 5 ACA 1 ACA 2 ACA 3 ACA 4 ACA 5 ACA 6 ACA 7 ACA 8 ACA 9 ACA 10 ACA 11 ACA 12 ACA 13 ACA 14 ACA 15 ACC 1 ACC 2 ACC 3 ACC 4 ACC 5 ACC 6 ACC 7 ACC 8 ACC 9 ACC 10 ACC 11 ACC 12

Sex

Age

Secretion

Pregnancy

Clinical features

Germ PKAR1A mutation

F F F F F F F F F F F F F F F M M M M M F F F F M F F F M M M M

15 11 10 27 29 69 46 50 44 30 46 71 35 34 69 67 41 56 65 56 30 80 53 45 54 28 69 27 73 39 67 68

Cortisol Cortisol Cortisol Cortisol Cortisol Cortisol Cortisol Cortisol Cortisol Cortisol Cortisol None Cortisol Cortisol Cortisol Cortisol None None Cortisol Cortisol Mineralo Cortisol/androgen Cortisol/androgen Cortisol/androgen Cortisol Cortisol/androgen None Cortisol/androgen Cortisol Cortisol None None

No No No No Yes No No No No Yes No No No No No – – – – – No No No No – No No No – – – –

CNC PPNAD CNC CNC PPNAD it it it it it it it it it it it it it it it it it it it it it it it it it it it

C.753 del AT C.846 ins A C.502C1 GOT C.708C1 GOT C.109 COT – – – – – – – – – – – – – – – – – – – – – – – – – – –

ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma; Adj Adr, tissue adjacent to ACA; CNC, carney complex; and i t, isolated tumor. Genetic data (i.e., mutations found in the coding sequence or splice sites of the cAMP-dependent protein kinase A (PKA) regulatory subunit 1A (PRKAR1A) gene) are also given for PPNAD.

468

www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 465–474 diabetes, and high blood pressure. Only plasma cortisol determinations were available and these confirmed the high levels of cortisol hypersecretion. At 19 weeks of pregnancy, plasma cortisol concentration at 0800 h was 42 mg/dl (normal values for this term: 25G2.1 mg/dl), and plasma cortisol concentration at 2000 h was 44 mg/dl. At 23 weeks of pregnancy, plasma cortisol concentration at 0800 h was 55 mg/dl (normal values for this term: 35G1.9 mg/dl). The patient suffered a spontaneous abortion at 24 weeks of pregnancy. After 6 weeks, morning plasma cortisol concentration was within the normal range, whereas urinary cortisol excretion (UC) demonstrated the persistence of mild ACTH-independent hypercortisolism: UCZ109 mg/24 h (normal range: 25– 90 mg/24 h), ACTH concentration at 0800 h !5 pg/ml (normal range: 20–60 pg/ml). No abnormal response of cortisol secretion to food intake, human chorionic gonadotrophin (hCG), gonadotrophin-releasing hormone, or thyrotrophin-releasing hormone administration was observed. Abdominal CT scan showed adrenal glands of normal size. Bilateral adrenalectomy was performed a few weeks later, and histological features confirmed the diagnosis of PPNAD. The sporadic adrenocortical tumors studied consisted of 15 ACAs (10 in women, 5 in men, with a mean age of 51.9G13.9 years) and 12 ACCs (7 in women and 5 in men, with a mean age of 52.7G18.8 years). The hypersecretion of cortisol, cortisol plus androgens, or mineralocorticoid was observed for all but three ACAs and three ACCs. In one patient (ACA5), ACA was diagnosed during pregnancy and the tumor was removed by surgical resection of the adrenal tumor at the time of Cesarean section.

Figure 1 Estradiol receptor a (ERa) protein in normal adrenal tissue (NA), adrenocortical tumors, and primary pigmented nodular adrenocortical disease (PPNAD). ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma; and Adj Adr, tissue adjacent to ACA. Male patients are represented by gray symbols and female by black % symbols. NS, the difference between ACC and non-tumoral tissues was not statistically significant. The shaded line indicates the threshold for ERa positivity. This cutoff point exceeds the detection limit of the assay but, for breast cancer, corresponds to a clinically relevant threshold below which the response to hormonal treatment is poor.

PR expression in tumor cells further by determining PR protein levels in the H295R cell line: 2713 and 1455 fmol/mg DNA in two samples from two different batches. We investigated the site of PR expression in the adrenal glands of PPNAD patients, by comparing PR expression in the total adrenal gland with samples isolated from the pigmented nodules of two patients (PPNAD 4 and PPNAD 5), in which the size of some

Detection of ERa and PR protein

We used EIA to assess levels of the ERa subtype and PR in 5 cases of PPNAD, 11 ACAs and adjacent normal tissues, 8 ACCs, and 3 normal adrenal gland samples (NA). ERa was detectable, but present at only low concentrations in NA and adrenal tissue adjacent to ACA (51G33). The concentration of this receptor was also low in ACC (53G78) and was significantly lower in the ACA itself (11G11 fmol/mg DNA (meanGS.D.; PZ0.012; Fig. 1). Conversely, PR concentrations were high in NA and adrenal tissue adjacent to ACA: 307G 216 fmol/mg DNA, and were even higher in benign tumors, 726G706 fmol/mg DNA (PZ0.036). PR expression was also very strong in malignant ACC, at 1154G1586 fmol/mg DNA, but this level of expression was not significantly different from that in the normal tissues (PZ0.12; Fig. 2). We investigated www.endocrinology-journals.org

Figure 2 Progesterone receptor (PR) protein in normal adrenal tissues (NAs), adrenocortical tumors, and primary pigmented nodular adrenocortical disease (PPNAD). ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma; and Adj Adr, tissue adjacent to ACA. Male patients are represented by gray symbols and female by black % symbols. NS, the difference between ACC and non-tumoral tissues was not statistically significant. The shaded line indicates the threshold for positivity for PR. This cutoff point exceeds the detection limit of the assay, but corresponds, for breast cancer, to a clinically relevant threshold below which the response to hormonal treatment is poor.

469

P de Cremoux et al.: Estradiol receptors in adrenocortical tumors nodules allowed separate sampling. The PR concentration in the isolated nodules was high, at 1583 and 564 fmol/mg DNA, whereas much lower levels were found in adjacent micronodular/atrophic adrenal tissues, 112 and 140 fmol/mg DNA. The PR expression was also assessed by immunohistochemistry in the same tissues: immunostaining for PR was strong in the nodular regions with a nuclear and cytoplasmic localization, whereas no signal was detected in the adjacent atrophic tissue (Fig. 3A). Immunohistochemistry was performed in adrenal tissues from patients with high PR levels. Cytoplasmic and nuclear immunostaining of PR was observed in adrenal cells, but not in vessels (Fig. 3B). ERa, ERb and PR transcript levels

Consistent with the data obtained for the corresponding protein, ERa mRNA was detectable, but present at only low levels in NA (84, 111) and adrenal tissue adjacent to ACA (262G108; nZ5). The concentration

Figure 3 A(i) PR immunostaining in the nodular region and adjacent adrenal tissue in a patient with PPNAD (PPNAD 5). (ii) The same zone with he´matoxyline–eosine– safran staining. B(i) PR immunostaining in an ACC with high PR content. (ii) ACC with low levels of PR expression.

470

Figure 4 Estradiol receptor a (ERa) mRNA in normal adrenal tissues (NAs) and adrenocortical tumors. ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma; Adj Adr, tissue adjacent to ACA. Male patients are represented by gray symbols and female by black % symbols. NS, the difference between ACC and non-tumoral tissues was not statistically significant. The shaded line indicates the threshold for positivity for ERa expression defined in breast cancer: below this limit, no response to hormonal treatment is expected.

of this mRNA was also low in ACC (149G244; nZ14) and was significantly lower in the ACA samples (60G37 AU; nZ11; PZ0.043; Fig. 4). ERb mRNA levels were very high in all samples, with the exception of three ACC samples. Mean ERb mRNA levels seemed to be higher in ACA (31.4G11.8 AU; nZ14) than in adjacent tissues (19.6G4.5; nZ5), but this difference was not statistically significant (PZ0.07). ERb mRNA levels were also high in ACC (35.43G44.4 AU; nZ11) and were not significantly different from those in the two other groups (Fig. 5). For comparison, ERb mRNA levels assayed with the same technique were much lower in normal breast tissue (4.91G4.77; nZ22) and even lower in breast tumors (0.89G0.70; nZ52; personal data). An RT-PCR quantification of PR mRNA confirmed PR expression to be strong in most tissues. The Ntarget was 18 and 19 AU in two normal adrenal glands, 76.7G60.1 and 45.5G17.5 AU in ACA (nZ14) and adjacent tissue (nZ5) respectively, 117.7G90.9 AU in 11 ACCs. By comparison, breast tumors are considered ‘positive’ for PR when ‘Ntarget’ exceeds 50 AU. The difference between tumoral and non-tumoral tissues was not statistically significant (Fig. 6). The concentrations of mRNA for these receptors did not differ in men and women. The very high levels of ER-b and PR in some tumors from male patients were striking. However, it is difficult to draw a conclusion because of the large dispersion of values. Aromatase levels were low in ACA (1.8G1.1 AU; nZ10) and adjacent normal tissues 2.7G2.6 AU (nZ5), and in ACC (2.9G3.1AU (nZ8)). One normal www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 465–474 all but one tumor (6.66 nM). The human malignant adrenocortical cell line H295R, which strongly expresses PR, was also studied for progesterone binding. The calculated kd for this line was 0.35 nM.

Discussion

Figure 5 Estradiol receptor b (ERb) mRNA in normal adrenal tissues (NAs), adrenocortical tumors, and primary pigmented nodular adrenocortical disease (PPNAD). ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma; and Adj Adr, tissue adjacent to ACA. Male patients are represented by gray symbols and female by black % symbols. NS, the difference between tumoral and non-tumoral tissues was not statistically significant. The shaded line indicates the threshold for positivity for ERb expression defined in breast tissue. This cutoff point exceeds the detection limit of the assay, but corresponds to mean values in normal breast tissue.

adrenal sample had a high aromatase content (17.4 AU), whereas the other did not (2.85 AU). Cytosol-based ligand-binding assays for ER and PR

Scatchard plots, used to assess PR binding in four tumors and ER binding in one tumor, yielded kd values compatible with a steroid receptor (0.35–1.99 nM) for

Figure 6 PR mRNA levels, as determined by real-time PCR assays, in normal adrenal tissues (NAs), adrenocortical tumors, and primary pigmented nodular adrenocortical disease (PPNAD). ACA, adrenocortical adenoma; ACC, adrenocortical carcinoma; and Adj Adr, tissue adjacent to ACA. The mRNA levels for each tissue are expressed in arbitrary units. Male patients are represented by gray symbols and female by black % symbols. NS, the difference between tumoral and non-tumoral tissues was not statistically significant. The shaded line indicates the threshold for PR positivity defined in breast cancer: below this limit, no response to hormonal treatment is expected.


The particular susceptibility of women to adrenocortical tumors is striking and remains unexplained. The high frequency of these tumors among hypercortisolic patients during pregnancy (Guilhaume et al. 1992) and the description of pregnancy-dependent ACTH-independent CS (Caticha et al. 1993, Close et al. 1993, Wallace et al. 1996, Kasperlik-Zaluska et al. 2000) suggest that adrenocortical proliferation and/or secretion may be sensitive to female hormones. We demonstrate strong PR expression in normal adrenal tissues, adrenal tumors, and pigmented nodules from PPNAD. The ERa was weakly detected in normal adrenal tissues and its expression was downregulated in benign tumors. In contrast, ERb transcript levels were higher than those in normal breast tissues and breast carcinoma (de Cremoux et al. 2002, 2004), in both normal adrenal tissues and adrenal tumors. This is the first demonstration of ERb expression in human adult adrenal tissues. The ERb does not seem to be downregulated in ACC, in contrast to what has been reported in the breast, in which ERb levels are consistently higher in the normal tissue than in tumoral tissue. Few studies have investigated the expression of ER and PR in the adrenal gland. A previous study based on immunohistochemistry and gradient shift assays assessed the levels of steroid receptors in adult rhesus monkey adrenal gland and concluded that ER was present in adult adrenal gland whereas PR was absent (Hirst et al. 1992). A recent RT-PCR-based study showed that ERa was expressed in the rat adrenal gland, whereas ERb was not detected and PR was not studied (Kuiper et al. 1997). The difference between the strong expression of PR in normal and tumoral adrenal tissues observed here and the results obtained by Hirst et al. (1992) in normal monkey adrenal glands may be due to differences in the species studied or in the techniques used. Immunohistochemistry and studies of the H295R adrenocortical cell line showed that, in tumors, PR is expressed in the tumoral cells themselves, rather than in the stroma or the vessels. Cytoplasmic and nuclear localization observed here may suggest the presence of the isoform B of PR, although the quantitative techniques used here do not allow to distinguish between the two isoforms. Antibody against PR also specifically stained cortical cells in PPNAD nodules – the site of adrenal proliferation in this condition. In the surrounding cortex, 471

P de Cremoux et al.: Estradiol receptors in adrenocortical tumors PR staining was weak as in normal adrenal tissue. These findings were confirmed by EIA on separate samples from isolated PPNAD nodules and adjacent atrophic/micronodular zones. Various factors may influence the regulation of PR expression in adrenal tissues. In reproductive organs, the induction of PR gene expression by estradiol involves mostly the ERa subtype. However, in some regions of the central nervous system in male rodents, the induction of PR by estradiol depends on ERb (Kudwa et al. 2004). This may be the case in adrenal cells. PR may also be upregulated through activation of the cAMP/PKA pathway, as observed in preovulatory ovarian granulosa cells. In these cells, LH via the production of cAMP, but not estradiol, induces PR mRNA expression (Clemens et al. 1998). In patients with PPNAD, inactivating mutations of PDE11A4, which encodes a phosphodiesterase, and of PRKAR1A, which encodes the RI subunit of the cAMP-dependent PKA, are frequently identified (Cazabat et al. 2006). Somatic mutations of this gene are also observed in a small number of cortisol-secreting ACA, unrelated to CNC. In those tissues, cAMP-stimulated kinase activity is increased (Bertherat et al. 2003). Given the strong and specific expression of PR observed in nodules from PPNAD, in which the PKA pathway is activated, the PR expression may be induced by cAMP stimulation in tumoral tissue. The physiopathological relevance of steroid receptors in adrenal cell proliferation is unknown. By comparison with the ERa protein levels determined with the same tools in normal and tumoral mammary glands, we considered the level of expression of this receptor in normal adrenal tissues to be low. The even lower level of expression observed in adrenal tumors suggests that this receptor plays a more minor role in adrenal tumorigenesis. In contrast, ERb levels were high, suggesting a more probable role. Montanaro et al. (2005) demonstrated that the H295R adrenal carcinoma cell line produces more mRNA for ERb than for ERa. They also showed that 4-OH-tamoxifen (a selective estrogen receptor modulation (SERM)) and ICI 182 780 (a pure anti-estrogen) upregulate ERb and inhibit cell proliferation. The physiopathological relevance of the high levels of PR expression in adrenal cells remains unclear. However, as progesterone is a precursor of most of the steroids synthesized in adrenal glands, local production may activate PR in an autocrine/paracrine manner. The PR may also be activated by intracellular signaling pathways: the production of cAMP following PKA activation can induce the ligand-independent activation of chicken PR, enhancing its steroid-dependent 472

activation (Beck et al. 1992). Could PR activation favor adrenal tumorigenesis. Transgenic mice expressing the simian virus 40 T antigen under control of the murine inhibin-a subunit promoter develop gonadal tumors. If early gonadectomy is performed, adrenal tumors appear. In these mice, adrenal tumorigenesis is dependent on steroidogenesis: the proliferation of C a1 cells (derived from these murine adrenal tumors) is stimulated by progesterone at physiological concentrations (Rilianawati et al. 1998). Progestin activation of Src/MAPK may occur outside the nucleus with the B isoform of PR, which is distributed between the cytoplasm and nucleus, and induce the transcription of target genes involved in cell cycle progression (Boonyaratanakornkit et al. 2007). Functional studies are required to evaluate the role of PR in human adrenal gland. Although the most convincing evidence would come from neoplasms isolated from human patients, the H295 cell line could be a useful tool to address the biological effect of progesterone on adrenal tumors. However, available data on animal models together with the presence of the high levels of ERb and PR in human adrenal tumors reported here provide new insight into the potential therapeutic benefit of anti-estrogens and/or anti-progesterone for adrenal neoplasm. We conclude that PR is present in human adrenal gland and specifically overexpressed in some adrenocortical tumors and in the nodules of PPNAD patients. This overexpression occurs in a context of very low levels of ERa expression and high levels of ERb expression. These findings raise questions about the molecular mechanisms underlying steroid receptor regulation in adrenal tumors and their potential role in adrenal tumorigenesis.

Acknowledgements We thank the patients and all participants in this study. We particularly thank E Boudou and C Barbaroux for their excellent technical assistance. This work was supported in part by the Plan Hospitalier de Recherche Clinique to the COMETE network (AOM 02068), the Ligue National Contre le Cancer (04-7571) and the GIS-INSERM Institut des Maladies Rares for the Carney Complex network and the European Network for the study of Adrenal Tumours (ENSAT). F Tissier, L Groussin, J Bertherat, X Bertagna and M-L RaffinSanson are members of the COMETE Network. We declare there to be no conflict of interest prejudicing the impartiality of this work. www.endocrinology-journals.org

Endocrine-Related Cancer (2008) 15 465–474

References Balaton AL, Coindre JM, Collin F, Ettore F, Fiche M, Jacquemier J, Le Doussal V, Mandard AM, Mathieu MC, Migeon C et al. 1996 Recommendations for the immunohistochemical evaluation of hormone receptors on paraffin sections of breast cancer. Study Group on Hormone Receptors using Immunohistochemistry FNCLCC/AFAQAP. National Federation of Centres to Combat Cancer/French Association for Quality Assurance in Pathology. Annales de Pathologie 16 144–148. Beck CA, Weigel NL & Edwards DP 1992 Effects of hormone and cellular modulators of protein phosphorylation on transcriptional activity, DNA binding, and phosphorylation of human progesterone receptors. Molecular Endocrinology 6 607–620. Bernoux A, de Cremoux P, Laine-Bidron C, Martin EC, Asselain B & Magdelenat H 1998 Estrogen receptor negative and progesterone receptor positive primary breast cancer: pathological characteristics and clinical outcome. Institut Curie Breast Cancer Study Group. Breast Cancer Research and Treatment 49 219–225. Bertagna X, Groussin L, Luton JP & Bertherat J 2003 Aberrant receptor-mediated Cushing’s syndrome. Hormone Research 59 (Supplement 1) 99–103. Bertherat J, Groussin L, Sandrini F, Matyakhina L, Bei T, Stergiopoulos S, Papageorgiou T, Bourdeau I, Kirschner LS, Vincent-Dejean C et al. 2003 Molecular and functional analysis of PRKAR1A and its locus (17q22-24) in sporadic adrenocortical tumors: 17q losses, somatic mutations, and protein kinase A expression and activity. Cancer Research 63 5308–5319. Bertherat J, Contesse V, Louiset E, Barrande G, Duparc C, Groussin L, Emy P, Bertagna X, Kuhn JM, Vaudry H et al. 2005 In vivo and in vitro screening for illegitimate receptors in adrenocorticotropin-independent macronodular adrenal hyperplasia causing Cushing’s syndrome: identification of two cases of gonadotropin/gastric inhibitory polypeptide-dependent hypercortisolism. Journal of Clinical Endocrinology and Metabolism 90 1302–1310. Boonyaratanakornkit V, McGowan E, Sherman L, Mancini MA, Cheskis BJ & Edwards DP 2007 The role of extranuclear signaling actions of progesterone receptor in mediating progesterone regulation of gene expression and the cell cycle. Molecular Endocrinology 21 359–375. Brandenberger AW, Tee MK, Lee JY, Chao V & Jaffe RB 1997 Tissue distribution of estrogen receptors alpha (ER-alpha) and beta (ER-beta) mRNA in the midgestational human fetus. Journal of Clinical Endocrinology and Metabolism 82 3509–3512. Caticha O, Odell WD, Wilson DE, Dowdell LA, Noth RH, Swislocki AL, Lamothe JJ & Barrow R 1993 Estradiol stimulates cortisol production by adrenal cells in estrogen-dependent primary adrenocortical nodular dysplasia. Journal of Clinical Endocrinology and Metabolism 77 494–497.


Cazabat L, Ragazzon B, Groussin L & Bertherat J 2006 PRKAR1A mutations in primary pigmented nodular adrenocortical disease. Pituitary 9 211–219. Clemens JW, Robker RL, Kraus WL, Katzenellenbogen BS & Richards JS 1998 Hormone induction of progesterone receptor (PR) messenger ribonucleic acid and activation of PR promoter regions in ovarian granulosa cells: evidence for a role of cyclic adenosine 3 0 ,5 0 -monophosphate but not estradiol. Molecular Endocrinology 12 1201–1214. Close CF, Mann MC, Watts JF & Taylor KG 1993 ACTHindependent Cushing’s syndrome in pregnancy with spontaneous resolution after delivery: control of the hypercortisolism with metyrapone. Clinical Endocrinology 39 375–379. de Cremoux P, Tran-Perennou C, Elie C, Boudou E, Barbaroux C, Poupon MF, De Rycke Y, Asselain B & Magdelenat H 2002 Quantitation of estradiol receptors alpha and beta and progesterone receptors in human breast tumors by real-time reverse transcription-polymerase chain reaction. Correlation with protein assays. Biochemical Pharmacology 64 507–515. de Cremoux P, Bieche I, Tran-Perennou C, Vignaud S, Boudou E, Asselain B, Lidereau R, Magdelenat H, Becette V, Sigal-Zafrani B et al. 2004 Inter-laboratory quality control for hormone-dependent gene expression in human breast tumors using real-time reverse transcription-polymerase chain reaction. Endocrine-Related Cancer 11 489–495. Gicquel C, Bertagna X, Gaston V, Coste J, Louvel A, Baudin E, Bertherat J, Chapuis Y, Duclos JM, Schlumberger M et al. 2001 Molecular markers and long-term recurrences in a large cohort of patients with sporadic adrenocortical tumors. Cancer Research 61 6762–6767. Goussard J, Lechevrel C & Roussel G 1989 Immunological and physiological analysis of human breast cancer progesterone receptor heterogeneity, following KCl dissociation and size exclusion chromatography. Journal of Steroid Biochemistry 33 1063–1072. Graham JD & Clarke CL 1997 Physiological action of progesterone in target tissues. Endocrine Reviews 18 502–519. Groussin L, Jullian E, Perlemoine K, Louvel A, Leheup B, Luton JP, Bertagna X & Bertherat J 2002a Mutations of the PRKAR1A gene in Cushing’s syndrome due to sporadic primary pigmented nodular adrenocortical disease. Journal of Clinical Endocrinology and Metabolism 87 4324–4329. Groussin L, Perlemoine K, Contesse V, Lefebvre H, Tabarin A, Thieblot P, Schlienger JL, Luton JP, Bertagna X & Bertherat J 2002b The ectopic expression of the gastric inhibitory polypeptide receptor is frequent in adrenocorticotropinindependent bilateral macronodular adrenal hyperplasia, but rare in unilateral tumors. Journal of Clinical Endocrinology and Metabolism 87 1980–1985. Groussin L, Kirschner LS, Vincent-Dejean C, Perlemoine K, Jullian E, Delemer B, Zacharieva S, Pignatelli D, Carney JA, Luton JP et al. 2002c Molecular analysis of the cyclic

473

P de Cremoux et al.: Estradiol receptors in adrenocortical tumors AMP-dependent protein kinase A (PKA) regulatory subunit 1A (PRKAR1A) gene in patients with Carney complex and primary pigmented nodular adrenocortical disease (PPNAD) reveals novel mutations and clues for pathophysiology: augmented PKA signaling is associated with adrenal tumorigenesis in PPNAD. American Journal of Human Genetics 71 1433–1442. Guilhaume B, Sanson ML, Billaud L, Bertagna X, Laudat MH & Luton JP 1992 Cushing’s syndrome and pregnancy: aetiologies and prognosis in twenty-two patients. European Journal of Medicine 1 83–89. Hirst JJ, West NB, Brenner RM & Novy MJ 1992 Steroid hormone receptors in the adrenal glands of fetal and adult rhesus monkeys. Journal of Clinical Endocrinology and Metabolism 75 308–314. Horvath A, Boikos S, Giatzakis C, Robinson-White A, Groussin L, Griffin KJ, Stein E, Levine E, Delimpasi G, Hsiao HP et al. 2006 A genome-wide scan identifies mutations in the gene encoding phosphodiesterase 11A4 (PDE11A) in individuals with adrenocortical hyperplasia. Nature Genetics 38 794–800. Kasperlik-Zaluska AA, Szczupacka I, LeszczynskaBystrzanowska J & Drus-Przybyszewska G 2000 Pregnancy-dependent Cushing’s syndrome in three pregnancies. British Journal of Obstetrics and Gynaecology 107 810–812. Kirschner LS, Carney JA, Pack SD, Taymans SE, Giatzakis C, Cho YS, Cho-Chung YS & Stratakis CA 2000 Mutations of the gene encoding the protein kinase A type I-alpha regulatory subunit in patients with the Carney complex. Nature Genetics 26 89–92. Kudwa AE, Gustafsson JA & Rissman EF 2004 Estrogen receptor beta modulates estradiol induction of progestin receptor immunoreactivity in male, but not in female, mouse medial preoptic area. Endocrinology 145 4500–4506. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S & Gustafsson JA 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 138 863–870. Libe R & Bertherat J 2005 Molecular genetics of adrenocortical tumours, from familial to sporadic diseases. European Journal of Endocrinology 153 477–487. Lindsay JR & Nieman LK 2006 Adrenal disorders in pregnancy. Endocrinology and Metabolism Clinics of North America 35 1–20. Luton JP, Cerdas S, Billaud L, Thomas G, Guilhaume B, Bertagna X, Laudat MH, Louvel A, Chapuis Y, Blondeau P

474

et al. 1990 Clinical features of adrenocortical carcinoma, prognostic factors, and the effect of mitotane therapy. New England Journal of Medicine 322 1195–1201. Luton JP, Martinez M, Coste J & Bertherat J 2000 Outcome in patients with adrenal incidentaloma selected for surgery: an analysis of 88 cases investigated in a single clinical center. European Journal of Endocrinology 143 111–117. Macfarlane DA 1958 Cancer of the adrenal cortex; the natural history, prognosis and treatment in a study of fiftyfive cases. Annals of the Royal College of Surgeons of England 23 155–186. Montanaro D, Maggiolini M, Recchia AG, Sirianni R, Aquila S, Barzon L, Fallo F, Ando S & Pezzi V 2005 Antiestrogens upregulate estrogen receptor beta expression and inhibit adrenocortical H295R cell proliferation. Journal of Molecular Endocrinology 35 245–256. Pabon JE, Li X, Lei ZM, Sanfilippo JS, Yussman MA & Rao CV 1996 Novel presence of luteinizing hormone/ chorionic gonadotropin receptors in human adrenal glands. Journal of Clinical Endocrinology and Metabolism 81 2397–2400. Rilianawati PT, Kero J, Zhang FP, Rahman N, Kananen K & Huhtaniemi I 1998 Direct luteinizing hormone action triggers adrenocortical tumorigenesis in castrated mice transgenic for the murine inhibin alpha-subunit promoter/simian virus 40 T-antigen fusion gene. Molecular Endocrinology 12 801–809. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC et al. 2002 Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. Journal of the American Medical Association 288 321–333. Saner-Amigh K, Mayhew BA, Mantero F, Schiavi F, White PC, Rao CV & Rainey WE 2006 Elevated expression of luteinizing hormone receptor in aldosterone-producing adenomas. Journal of Clinical Endocrinology and Metabolism 91 1136–1142. Wallace C, Toth EL, Lewanczuk RZ & Siminoski K 1996 Pregnancy-induced Cushing’s syndrome in multiple pregnancies. Journal of Clinical Endocrinology and Metabolism 81 15–21. Weiss LM, Medeiros LJ & Vickery AL Jr 1989 Pathologic features of prognostic significance in adrenocortical carcinoma. American Journal of Surgical Pathology 13 202–206.
