Ectopic Expression of the Gastric Inhibitory Polypeptide Receptor Gene Is a Sufficient Genetic Event to Induce Benign Adrenocortic
http://www.100md.com
《内分泌学杂志》
Institut National de la Sante et de la Recherche Medicale Equipe Mixte 01-05 (T.L.M., O.C., J.-J.F., M.T.)
Commissariat a l’Energie Atomique, Departement Reponse et Dynamique Cellulaires, Laboratoire ANGIO (T.L.M., O.C., J.-J.F., M.T.), 38054 Grenoble, France
Centre Hospitalier Regional Universitaire de Grenoble, Departement de Diabetologie, Urologie, Nephrologie, et Endocrinologie, Service d’Endocrinologie (T.L.M., O.C.)
Departement d’Anatomie et de Cytologie Pathologique, Laboratoire de Pathologie Cellulaire (N.S.), 38043 Grenoble, France
Abstract
Aberrant expression of ectopic G protein-coupled receptors (GPCRs) in adrenal cortex tissue has been observed in several cases of ACTH-independent macronodular adrenal hyperplasias and adenomas associated with Cushing’s syndrome. Although there is clear clinical evidence for the implication of these ectopic receptors in abnormal regulation of cortisol production, whether this aberrant GPCR expression is the cause or the consequence of the development of an adrenal hyperplasia is still an open question. To answer it, we genetically engineered primary bovine adrenocortical cells to have them express the gastric inhibitory polypeptide receptor. After transplantation of these modified cells under the renal capsule of adrenalectomized immunodeficient mice, tissues formed had their functional and histological characteristics analyzed. We observed the formation of an enlarged and hyperproliferative adenomatous adrenocortical tissue that secreted cortisol in a gastric inhibitory polypeptide-dependent manner and induced a mild Cushing’s syndrome with hyperglycemia. Moreover, we show that tumor development was ACTH independent. Thus, a single genetic event, inappropriate expression of a nonmutated GPCR gene, is sufficient to initiate the complete phenotypic alterations that ultimately lead to the formation of a benign adrenocortical tumor.
Introduction
DEFECTS IN THE expression or activity of G protein-coupled receptors (GPCRs) have been identified in a wide variety of endocrine disorders (1). Several natural mutations of GPCRs have been characterized that result in receptor loss-of-function, whereas other mutations constitutively stimulate G protein coupling or alter the specificity of the binding domain, resulting in gain-of-function (1). Aberrant expression of a GPCR in a tissue that normally does not express or poorly expresses it is another mechanism leading to endocrine diseases, which is observed in particular in several variants of Cushing’s syndrome (CS). CS results from lasting exposure to excess glucocorticoids and is either ACTH dependent or independent. Recent studies have clearly established that some cortisol-producing ACTH-independent macronodular adrenal hyperplasia (AIMAH) and adenomas are controlled by the aberrant adrenocortical expression of ectopic or hyperactive eutopic GPCRs (2). In such cases, in vitro studies have demonstrated that cortisol secretion becomes driven by an unusual hormone that activates the adenylyl cyclase (AC)/cAMP signaling cascade normally triggered by the ACTH receptor. During the last decade, the GPCRs for gastric inhibitory polypeptide (GIP), catecholamines, TSH, vasopressin, serotonin, and LH have been shown to be overexpressed in several cases of unusual AIMAH and adenomas (2). However, although there is clear clinical evidence for the implication of these ectopic receptors in the dysregulated cortisol production observed in these pathologies, whether this aberrant GPCR expression is the cause or the consequence of the development of an adrenal benign tumor is still an unanswered question. We chose to address this question in the present work using the GIP receptor as a model GPCR, because it appears to be one of the most commonly encountered receptors in vitro among abnormally expressed GPCR-dependent CS (3).
GIP is a gastrointestinal hormone that is released during meals. GIP acts on pancreatic -cells to stimulate insulin release through binding to a seven-transmembrane domain GPCR. Once bound by GIP, this pancreatic -cell receptor activates AC, increases cAMP production, and ultimately leads to the elevation of intracellular calcium and insulin exocytosis (4, 5). Some studies have indicated that GIP may contribute to -cell growth (6, 7). Moreover, the GIP receptor (GIPR) gene has been found to be expressed at a lower level in brain, intestine, adipose tissue, and heart but not in normal adrenal cortex (8, 9, 10, 11, 12, 13, 14). Ectopic GIPR expression in the adrenal cortex is observed in food-dependent CS associated with benign adrenal tumors (15). The pathophysiology of this syndrome is characterized by a hypercortisolism induced by physiological postprandial increase in circulating GIP concentrations (16, 17). Abnormal GIPR expression has also been described in other cases of benign adrenal tumors (8, 9, 11, 15, 18, 19, 20, 21) but was not observed in a small cohort of malignant adrenal tissues (3). No mutation in GIPR gene coding sequence and promoter could be identified so far (12, 22), and the molecular mechanisms of this ectopic GIPR expression remain unclear.
To investigate the role of ectopic GIPR expression in the development of benign adrenocortical tumors, we used an in vivo model of cell transplantation and tissue reconstruction (23). In this model, primary bovine adrenocortical cells are transplanted under the renal capsule of adrenalectomized immunodeficient mice reconstituting a vascularized and functional adrenocortical tissue, which secretes cortisol and avoids the otherwise lethal effect of adrenalectomy (24). We have shown previously that retroviral introduction of a limited number of transforming genes (hTERT, SV40 large T, and ras val12) in the primary cells before transplantation leads to the formation of an invasive tumor (25). The introduction of single genes or combinations, other than all three, leads to the formation of tissue with various degrees of abnormality, which is neither continuously expanding nor invasive (25). In the present work, we genetically engineered primary bovine adrenocortical cells to have them abnormally express the GIPR and analyzed the functional and proliferative characteristics of the tissue formed after cell transplantation. In this way, we could directly evaluate whether the unique genetic alteration that results in ectopic expression of the GIPR in adrenal cortex is the cause of both benign adrenocortical mass and aberrant cortisol production observed in the food-dependent CS. Our results clearly validate this hypothesis.
Materials and Methods
GIPR vector construction and retrovirus production
The rat GIPR cDNA initially cloned in a modified pCMV vector (13) was amplified by PCR with primers designed for creation of HindIII and ClaI restriction sites at 5' and 3' ends, respectively. The 1.4-kb DNA fragment was subcloned into the pGEM-T Easy cloning vector (Promega, Madison WI); it corresponds to the GIPR cDNA coding sequence 163-1534, which includes part of exon 2 and exons 3–14 (according to sequence number L19660, EMBL database). The sequence-verified GIPR cDNA was transferred from construct rGIPR-pGEM-T Easy to a pLNCX2 retroviral expression vector (Clontech, Palo Alto, CA), downstream of its immediate-early cytomegalovirus promoter. Retroviral vector without insert was used as a negative control. The packaging PT67 cells (Clontech) were transfected with vectors and used for the production of replication-incompetent viruses. After 7 d of 400 μg/ml G418 antibiotic selection, viral supernatant was passed through a 0.45-μm-syringe filter to obtain cell-free viruses for adrenocortical cell infection.
Growth of bovine adrenocortical cells in culture and retroviral transduction
Primary adrenocortical cells were prepared by enzymatic digestion of adrenal glands from 2-yr-old steers, as previously described (26). Primary cell suspensions were stored frozen in liquid nitrogen. Frozen cells were thawed and plated in DMEM/Ham’s F-12 1:1 supplemented with 10% fetal calf serum, 10% horse serum, and 1% (vol/vol) UltroSer G (Biosepra, Villeneuve-la-Garenne, France). Adrenocortical cells were infected for 6 h by adding the filtered medium containing retroviruses at a ratio of 1:1 with fresh culture medium. Infected cells were selected with G418 for 7 d generating either control cells or GIPR cells.
In vitro studies
cAMP production and cortisol secretion by transduced cells were measured after 2 h of incubation with 10 μM forskolin (Sigma, Saint Quentin Fallavier, France), 100 nM ACTH (1–24) (Neosystem, Strasbourg, France), 100 nM GIP (Bachem, Voisins-le-Bretonneux, France), or fresh medium. Cortisol secretion was assayed directly on culture medium by a specific RIA with a cortisol antiserum (Endocrine Sciences, Calabasas, CA). cAMP production was measured from the supernatant of homogenized cells in the presence of 1 mmol/liter IBMX using an ELISA kit (Neogen Corp., Lansing, MI) following the manufacturer’s recommendations.
For cell cycle analysis by fluorescence-activated cell sorting (FACS), cells were grown for 72 h in serum-free medium (basal condition). During the last 24 h, 100 nM GIP or ACTH was added in some plates. A positive control of proliferation was performed by adding 20% serum to the culture medium. Before FACS analysis, cells were incubated with 20 μg/ml propidium iodide and 100 μg/ml RNase A. In each assay, 25,000 propidium iodide-positive cells were analyzed by FACScan using the CellQuest program (Becton Dickinson, Franklin Lakes, NJ). This experiment was repeated three times.
Gene expression analysis was assessed after total RNA extraction from cultured cells (RNeasy Mini Kit; QIAGEN, Courtaboeuf, France) and cDNA synthesis from 1 μg RNA with ImProm-II reverse transcriptase (Promega) using random primers. Semiquantitative PCR was performed with specific primers for human and bovine GIPR (forward primer, 5'-ATCTGCTGGTGGTTGTGAGACG-3'; reverse primer, 5'-ACGGTTCCTACGCCTATTTC-3'), bovine melanocortin 2 receptor (MC2R) (forward primer, 5'-GGGGTTTTGGAGAACCTGATGG-3'; reverse primer, 5'-GGCGACTGGCGATGTAGTGTTA-3'), human MC2R (forward primer, 5'-GACTGTCCTCGTGTGGTTTTGC-3'; reverse primer, 5'-CTTTGGTGTCGGCTACTGTAGTA-3'), bovine 3--hydroxy-5-steroid dehydrogenase/isomerase-1 (3HSD) (forward primer, 5'-GAGACCATCATGAACGTCAA-3'; reverse primer, 5'-GTCCTTCGAGTGATAAAGGT-3'), bovine 17-hydroxylase (CYP17) (forward primer, 5'-GGCCCCATCTATTCCTTTCGT-3'; reverse primer, 5'-CGTCTGTTATTGTTACGACCG-3'), bovine Bax (forward primer, 5'-TGCTTCAGGGTTTCATCCAG-3'; reverse primer, 5'-GGAGAGGATGAAACCCTGTG-3'), bovine Bcl-xL (forward primer, 5'-GGATAGCCCTGCTGTGAATGG-3'; reverse primer, 5'-CACACCAAGACGACCCGAGT-3'), bovine Col-1A2 (forward primer, 5'-CAACCATGCCTCTCAGAACA-3'; reverse primer, 5'-TGTACCTACTCCTTTGACCG-3'), and RP-L27 (forward primer, 5'-GAACATTGATGATGGCACCTC-3'; reverse primer, 5'-GGGGATATCCACAGAGTACC-3') genes. Total RNA extraction from transplanted tissues was also performed to analyze GIPR gene expression by RT-PCR. Amplification products were separated by 2% agarose gel electrophoresis, scanned with a FluorImager (Molecular Dynamics, Sunnyvale, CA), and normalized to RP-L27. An apoptosis index was calculated based on the ratio of expression levels of proapoptotic Bax and antiapoptotic Bcl-xL (27).
Cell transplantation and animal experimentation
Immunodeficient RAG 2 –/– mice originally purchased from CDTA (Orleans, France) were maintained in our animal facility as a breeding colony. All animal studies were approved by the institutional guidelines and the European Community for the Use of Experimental Animals. Under Avertin anesthesia, male and female mice at an age between 8 and 12 wk (22 g body weight) were adrenalectomized, and 2 x 106 control or GIPR adrenocortical cells were transplanted under the renal capsule (23, 24) together with 4 x 105 fibroblast growth factor (FGF)-1-secreting 3T3 cells pretreated by 2 μg/ml mitomycin C (Sigma) to prevent their further proliferation The 3T3 cell line stably expressed FGF-1 fused in frame with a signal peptide from hst/KS3 gene, yielding a highly angiogenic secreted product (28). Postoperative animal care consisted of administration of an analgesic and a mixture of antibiotics in the drinking water for 7 d. After surgery, the mouse body weight was surveyed on a daily basis during the first week and then twice a week until euthanasia.
After 4 wk, tail blood samples were taken at time 0 and 15 min after the injection of ACTH (1–39) (Neosystem, 42 pmol/g body weight) or GIP (50 pmol/g body weight). Plasma cortisol and ACTH were measured by RIA and immunoradiometric assay, respectively. Blood glucose values were determined from whole blood using an automatic glucose monitor (Accu-Chek Active, Roche, France). Blood samples were collected in the afternoon, during the physiological fasting period and nadir of cortisol level. After 8 wk, the animals were killed and the kidneys bearing the adrenocortical transplant were excised for histological analysis.
In some experiments, 1 μg/g·d dexamethasone (Sigma) was delivered to animals by sc osmotic mini-pump (Alzet, Cupertino, CA) implanted at the time of cell transplantation to suppress pituitary ACTH secretion. Mice were killed after 14 d of dexamethasone or excipient (cyclodextrin) treatment. Cardiac blood samples were collected, and transplants were processed for histological analysis.
Histological studies
Paraformaldehyde-fixed tissues were paraffin embedded and sectioned using standard methods. Sections were stained with hematoxylin and eosin. Picro-Sirius Red staining was used to localize collagen. Tissues were immunostained with antibodies to CYP17, 3HSD, GIPR (29), and Ki-67 (Dako A/S, Trappes, France), with biotinylated antimouse or antirabbit secondary antibodies as required, followed by detection using an avidin-biotin-peroxidase complex and diaminobenzidine (Dako A/S). For 3HSD-antibody detection we used the Fast Red chromogen as a substrate. DNA fragmentation associated with apoptosis was detected by nick end labeling of sections using the TdT-FragEL kit (Oncogene Research Products, Boston, MA).
The number of Ki-67-positive nuclei per 100 adrenocortical cell nuclei was used as the proliferation index. Counting was performed manually, using two nonconsecutive tissue sections per tissue sample, selected at random in both groups (control and GIPR tissues); n = 10 sections of each group. Overall comparison was performed with graphical statistical summary and Student’s t test.
Results
GIPR expression in adrenocortical cells triggers cortisol secretion in vitro
We cloned the coding sequence of the rat GIPR cDNA in a retroviral vector and transduced primary cultures of bovine adrenocortical cells to induce stable ectopic expression in these cells (Fig. 1A). The rat GIPR nucleotide sequence displays 85.5% identity with the human GIPR coding sequence, and the rat and human proteins display similar affinities for GIPR (13). We first compared the phenotypes and steroidogenic capacities of the infected cells in vitro. Uninfected primary adrenocortical cells, selected stable infected cells transduced with an empty vector encoding only the drug resistance gene (named control cells hereafter), and selected stable infected cells transduced with a GIPR-expressing vector (named GIPR cells hereafter) all presented a similar morphology (Fig. 1B) and identical growth rates (data not shown) and presented contact inhibition at confluence. We analyzed also the GIPR cells grown in vitro under basal conditions and after either ACTH or GIP addition, as regards their cell cycle distribution. DNA labeling and FACS analysis revealed a similar distribution in the S-G2/M phases of the cell cycle (10.8 ± 1.2, 11.5 ± 0.3, and 10.2 ± 0.3%, respectively). The control cells exhibited a similar S-G2/M distribution as the GIPR cells under the same conditions (data not shown). Under stimulation by a 20% serum-rich culture medium, both control and GIPR cells exhibited a significant increase in the percentage of cells in S-G2/M phases (28.4 ± 1.0%, P < 0.001; 29.7 ± 0.6%, P < 0.001, respectively) compared with basal conditions and after either ACTH or GIP incubation. Thus, the expression of a functional GIPR in adrenocortical cells did not induce morphological or proliferative changes in vitro but triggered cortisol secretion in response to GIP.
As shown by RT-PCR analysis, GIPR gene expression was observed only in the GIPR-infected cell population and in a human adrenocortical adenoma from a patient presenting a food-dependent CS (Fig. 1C). We confirmed the absence of eutopic GIPR gene expression in primary adrenocortical cells. The expression of MC2R (the ACTH receptor) was preserved in all three types of cells, even after the antibiotic selection steps undergone by the infected cells (Fig. 1C). We also observed that genes encoding the steroid-converting enzymes type II 3HSD and steroid CYP17, involved in cortisol biosynthesis, were unaffected by retroviral infection-mediated gene transduction and were equally expressed in all cell types (Fig. 1C).
Cortisol is the main corticosteroid produced by bovine and human adrenocortical cells, whereas mouse cells, which lack the steroid CYP17 enzyme, secrete corticosterone. However, long-term in vitro culture of adrenocortical cells may result in loss of their steroidogenic capacities, despite detectable expression of 3HSD and CYP17 genes (30). We thus ascertained that our infected cells were still hormonally responsive after the 2 wk of culture necessary for the infection and antibiotic selection steps. We measured cAMP production in response to addition of either ACTH or GIP. Upon ACTH (100 nM) stimulation, we observed a strong (7-fold) elevation of cAMP in both infected cell types (control and GIPR cells), whereas upon GIP (100 nM) stimulation, only the GIPR cells responded (16-fold cAMP elevation) (Fig. 1D). We then measured the cortisol secretion in response to ACTH or forskolin (10 μM), a direct activator of AC, which appeared to be preserved in all transduced cells (2-fold over basal, Fig. 1D). In contrast, under GIP treatment, only the GIPR cells showed a 2-fold elevation of cortisol secretion (P < 0.05) (Fig. 1D). Thus, adrenocortical steroid synthesis appears to be correlated with receptor activation and cAMP production in both control and GIPR cells, and ectopic GIPR gene expression clearly confers to adrenocortical cells the new capacity to secrete cortisol in response to GIP.
Transplantation of GIPR cells into mice induces the formation of a functional GIP-responsive transgenic adrenocortical tissue
We then implanted GIPR cells under the renal capsule of adrenalectomized immunodeficient mice, allowed the grafted cells to form a neo-organ, and first analyzed the physiological responses of the grafted mice to hormonal challenges 6 wk after transplantation. In a previous study, we have shown that implantation of unmodified primary bovine adrenocortical cells leads to the formation of a vascularized and cortisol-secreting organ that saves mice from the lethal effect of adrenalectomy (24). To assess the functional status of the tissue formed from the transplanted GIPR cells, we first intended to check the direct effect of food intake on plasma cortisol levels. However, because mice eat several times over a long period of time during darkness, it was technically difficult to monitor cortisol levels as a function of food intake. It also turned out to be impossible to check the effect of food deprivation on the plasma cortisol levels of GIPR mice because, after one night of starving, mice had significantly lost weight, making the conclusions of such an experiment extremely uncertain. We thus decided to measure plasma cortisol levels in response to injection of a physiological postprandial concentration (50 pmol/g ip) of GIP. The mice transplanted with GIPR cells did show elevated cortisol in response to GIP injection (+44%), whereas the mice transplanted with control cells did not (Fig. 2, A and B). In contrast, as expected, injection of ACTH (2 pmol/g ip) resulted in a mild but reproducible elevation (+26%) of plasma cortisol levels in GIPR mice as well as in control mice (Fig. 2B). Because GIPR mice exhibited elevated cortisol levels in response to GIP, we screened some physical and biological features that might occur in a situation of cortisol hypersecretion such as CS. Eight weeks after transplantation, the plasma ACTH levels in GIPR mice (21.4 ± 6.3 pmol/liter) were decreased to 48% of those in age-matched control mice (44.6 ± 23.5 pmol/liter). This difference did not reach the threshold of statistical significance (P > 0.05). This partial ACTH suppression is consistent with a discontinuous inhibition of pituitary corticotrope cells due to fluctuating cortisol levels, as seen in some atypical clinical presentations of CS (mild hypercortisolism, cyclical CS, and subclinical CS) (31). Because these forms of hypercortisolism can alter glucose metabolism, resulting in glucose intolerance or overt diabetes mellitus (31), we assessed the insulin sensitivity of transplanted mice by measuring glycemia during their physiological fasting period. In a series of measurements performed between d 28 and 53 after transplantation, glycemia was always higher in GIPR mice than in control mice, and the mean values were significantly different (P < 0.001; n = 9) (Fig. 2C). We also noticed growth attenuation along with weight loss in mice transplanted with GIPR cells compared with mice transplanted with control cells (–15% on d 53 after surgery). Similar growth retardation is encountered in other mouse models of CS, as observed in glucocorticoid-treated mice (our unpublished observations) and in different transgenic mouse models of CS (32, 33, 34). No remarkable changes in distribution of body fat distribution such as the presence of a buffalo hump or intraabdominal fat deposits were observed at necropsy.
Mice transplanted with GIPR cells develop a hyperplasic adrenocortical tissue
We analyzed the histology and proliferative status of the neo-organs formed in the renal capsule niche. At the time of killing (8 wk after transplantation), tissues formed from transplanted adrenocortical cells were fixed and processed for conventional histology and histochemistry. The pale tissue originating from control cells was spread over the kidney parenchyma and occupied a thin space beneath the kidney capsule (Fig. 3A). In contrast, transplanted GIPR cells gave rise to a voluminous mass that was prominently yellow in its center and white on its periphery. Cross-sections of this tissue were about 15 times larger than those of the control neo-organ, and compression of the renal parenchyma was visible (Fig. 3B). Control tissue presented a uniform structure of regular eosinophilic adrenocortical cells in close contact with the kidney parenchyma on its lower side and the renal capsule on its upper side (Fig. 3B). In contrast, GIPR tissue had an aspect of an encapsulated mass losing the contact between adrenocortical cells and the kidney surface; the tissue showed an irregular architecture with cellular pleiomorphism and nuclear atypia (Fig. 3B). Both clear, lipid-laden (fasciculata-type) cells and eosinophilic lipid-depleted (reticularis-type) cells were dispersed in a stromal reaction, without any sign of necrosis. All 14 GIPR tumors collected up to 53 d after transplantation showed a clear border between the benign tumor and the kidney without any sign of invasion. Comparatively, the histological appearance of a human GIPR adenoma is shown in the right panels (Fig. 3B) to illustrate their irregular cellular distribution consisting in sheets of fasciculata-type cells interspersed with reticularis-type cells. This human adenoma was encapsulated and had some areas composed of tumor stroma, as revealed by Picro-Sirius Red staining (Fig. 3B). A pronounced fibrous stroma was similarly observed in GIPR transplants, where a framework of connective tissue is delineated by red-stained collagen fibers. In marked contrast, only the renal capsule was stained in control tissue (Fig. 3B). The stromal reaction was a result of the in vivo development of GIPR tissue because no difference in the levels of 2 type I collagen (Col-I) gene expression was observable by RT-PCR in GIPR cells in vitro (data not shown).
Preservation of the steroidogenic phenotype of the developing neo-organs was demonstrated by immunohistochemistry using antibodies against the steroid-converting enzymes 3HSD and CYP17 (Fig. 3C). Interestingly, in the GIPR tumor, all cells with adrenocortical histological features appeared to be steroidogenic, forming islets or nodular regions between unstained stromal cells. The expression of the GIPR gene and the detection of the GIPR protein were confirmed in 8-wk-old tissues formed from transplanted GIPR cells by RT-PCR (data not shown) and by immunohistochemistry (Fig. 3D). GIPR gene and protein expressions were not detected in tissues formed from control cells (data not shown).
To further characterize the enlargement of GIPR tissue mass compared with control tissue, we checked whether this hyperplasia resulted from increased proliferation or decreased apoptosis. Regarding proliferation, we used the antibody Ki-67, which binds a nonhistone nuclear protein expressed in all cell cycle phases, except G0/early G1. Ki-67 staining was observed throughout the transplant in 17.7 ± 1.3% of cells in GIPR tissue, a value significantly higher than that observed in control tissue (3.5 ± 1.0%, P < 0.001; Fig. 4, A and B). There was no difference between the apoptotic rates of control and GIPR cells before transplantation, as demonstrated by the proapoptotic index based on expression of BCL2-like 1 gene (Bcl-xL) and BCL2-associated X gene (Bax) (Fig. 4C). The detection of apoptotic cells by terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling showed no significant differences between both types of neo-organs (data not shown), indicating that decreased apoptosis was certainly not the cause for the observed hyperplasia. Thus, cells encoding GIPR gene showed a growth advantage after transplantation in mice, leading to an enlarged hyperproliferative adrenocortical mass.
To investigate whether the growth of the GIPR transplants was ACTH independent, we used a mouse model of synthetic glucocorticoid-induced ACTH suppression. In a previous study (35), we have shown that dexamethasone treatment induces adrenal atrophy as a consequence of ACTH suppression and that this tissue regression does not result from a direct glucocorticoid effect on the adrenal cortex. Adrenalectomized mice were transplanted with control or GIPR cells and chronically perfused with dexamethasone via sc osmotic mini-pumps from the onset of transplantation. Dexamethasone dramatically reduced plasma ACTH concentration in both groups of mice (1.3 ± 0.8 pmol/liter in control mice; 2.4 ± 1.2 pmol/liter in GIPR mice) as soon as 24 h after dexamethasone treatment. When dexamethasone-treated animals transplanted with control adrenocortical cells were killed on d 14 after surgery, an atrophic transplant tissue was present at the site of injection, with reduced cellularity and disorganized architecture compared with the one in untreated control mice (Fig. 5). In contrast, GIPR transplants had developed to a similar large size in the absence as well as in the presence of ACTH (Fig. 5). The ACTH-independent development of GIPR tumors thus suggests that physiological GIP may act as a trophic factor through activation of the adrenal ectopic GIPR.
Discussion
We report here on the generation of a GIP-dependent adrenocortical mass associated with a mild CS through the transplantation of genetically engineered bovine adrenocortical cells into immunodeficient mice. Very interestingly, the endocrine adrenocortical cells giving rise to these tumors were minimally modified by the retroviral introduction into their genome of a single nonmutated gene encoding the GIPR. This clearly demonstrates that in our animal model, the ectopic adrenal expression of this AC-activating receptor (physiologically expressed in the pancreatic -cells but not in adrenocortical cells) is sufficient to induce not only aberrant cortisol secretion but also hyperproliferation and benign cell transformation. These findings give strength to the hypothesis that aberrant expression of GPCRs plays a direct role in the development of these hyperplasic tissues. This is a novel and important contribution to the understanding of food-dependent CS in humans and more generally to all cases of benign adrenal tumors associated with aberrant GPCR expression. The clinical observations collected so far in the description of food-dependent CS, vasopressin-responsive CS, catecholamine-dependent CS, postmenopausal CS, and serotonin-dependent CS have clearly demonstrated the link between these endocrine disorders and the aberrant adrenal expression of GIPR, arginine vasopressin V1-3 receptors, -adrenergic receptors, LH-human chorionic gonadotropin receptor, and serotonin 5-HT4 and 5-HT7 receptors, respectively (2). In particular, in vitro studies using primary cell cultures derived from these tumors have established that the aberrantly expressed GPCRs confer steroidogenic response to the corresponding ligands, suggesting that these receptors directly couple at the cell surface with the AC-G-proteins complex and substitute for the MC2R function. Because cortisol overproduction exerts a feedback control on pituitary ACTH secretion but does not control the circulating levels of the ectopic receptor ligand, steroidogenesis becomes uncontrolled. It is noteworthy that the transplanted animals with GIPR cells had detectable ACTH levels. This lack of suppression of pituitary ACTH had previously been reported in the documented cases of food-dependent CS and might indicate that the discontinuous elevation of plasma GIP levels inducing cortisol secretion does not always result in complete suppression of the hypothalamic-pituitary-adrenal axis (18, 21, 36).
As concerns proliferation, only very few in vitro studies have addressed this question in benign adrenal tumor-derived cells. Interestingly, in our experiment, the percentage of genetically engineered cell population expressing GIPR in the S-G2/M phases was not modified by the addition of either GIP or ACTH to the medium. Our group reported previously that GIP could weakly stimulate MAPK activity and thymidine incorporation in adrenocortical cells from an adenoma of a patient with food-dependent CS (8). Conversely, GIP has been shown to have an inhibitory action on thymidine incorporation in adrenocortical cells isolated from a food-dependent adenoma (9). This type of analysis has unfortunately not been performed in other cases of CS secondary to aberrant GPCR expression. How does stimulation of GIPR in adrenocortical cells lead to cellular hyperplasia and pleiomorphism with stromal reaction, as obtained in GIPR transplants The most evident but quite vague answer to this question is that GIP acts through the same pathways as ACTH. It is long known that in vivo, an excess of endogenous or exogenous ACTH causes adrenal hypertrophy, rapidly followed by hyperplasia (37), whereas in vitro studies using adrenal cells from several species have shown that ACTH behaves as antimitogenic leading to the notion that ACTH is an indirect mitogen in intact animals (38). Although known for several decades, the trophic action of ACTH in vivo is still poorly deciphered at the molecular level because of this paradoxical effect on proliferation. It has been proposed that FGF-2, which is expressed by adrenocortical cells and is the most potent growth factor for endocrine adrenocortical cells, is a relay of ACTH action. More recently, we described that part of this trophic effect might be indirectly mediated by the adrenocortical vasculature via vascular endothelial growth factor (35). A link between AC activation and adrenocortical cell proliferation certainly exists, probably involving such paracrine mediators acting directly or indirectly. Moreover, despite an enlarged tissue, the basal cortisol production by GIPR mouse is not higher than that observed in control mouse. This might indicate a switch toward a proliferative phenotype of the cells expressing the GIPR gene with higher sensitivity to proliferative stimuli in vivo. Thus, the generation of GIPR transgenic tissues should give us the opportunity to further study the signaling pathways involved in proliferation and the cross-talk between pathways in adrenocortical cells.
We previously noted that an important element for success of adrenocortical cell transplantation is the supply of FGF-1 from the cotransplanted FGF-secreting 3T3 cells (23, 24). The 3T3 cells were rendered incapable of further division by mitomycin C treatment. At 36 d after cell transplantation, transplants that were not formed by the inclusion of FGF-secreting cells always contained a necrotic area, although the tissue produced plasma steroid levels in the same range as the 3T3 cell-containing transplants. Tissues formed with or without 3T3 cells had very similar structures outside of the necrotic area, indicating that the 3T3 cells did not substantially change the nature of the transplant tissue (24). Labeling studies indicated that rare 3T3 cells were still present after 36 d and were located within or next to the endothelium that forms within the transplant tissue (Thomas, M., unpublished observations). We cannot, however, rule out the possibility that GIPR-expressing cells could affect the survival and the function of the cotransplanted 3T3 cells through some unknown mechanism.
Programmed cell death inhibition might contribute to the cellular hyperplasia. However, in the present study, the development of a hyperplastic neo-organ derived from GIPR cells cannot be explained by an inhibition of apoptosis because we clearly observed a long-term imbalance of cell proliferation and no change in the rate of cell attrition. A genome-wide cDNA array analysis of gene expression in the GIPR neo-organ compared with the normal tissue derived from control cells might help to understand these mechanisms, although the differences in the respective abundance of the distinct cell types present in these tissues (abundant fibrous tissue in GIPR neo-organs) might render the interpretation of these data quite difficult.
The retroviral transduction of the GIPR gene allowed the stable expression of an ectopic gene in adrenocortical cells after transplantation in mice. Some studies had demonstrated the difficulty to obtain functional GPCRs in heterologous systems, which could be expressed only in cell lines having a minimal level of endogenous expression of the GPCR studied (39, 40). In our system, we had shown that the delivery of GIPR gene in adrenocortical cells was able to trigger normal receptor trafficking to the plasma membrane (as evidenced by immunolabeling, Fig. 3D) and normal receptor coupling to Gs protein leading to AC activation and steroidogenic response. This suggests that specific pharmacological antagonists of this aberrant receptor might be sufficient to treat GPCR-dependent AIMAH and adrenal adenomas. However, such attempts at controlling cortisol secretion have proven clinically unsuccessful to reverse adrenal overgrowth (17, 19), indicating that a number of yet unidentified secondary genetic events may have also occurred in these tumors. In a recent paper, Swords et al. (36) demonstrated functional expression of GIPR in primary adrenocortical cell cultures isolated from patients presenting an ACTH-dependent adrenal hyperplasia with no clinical evidence of food-dependent cortisol secretion. The authors suggested that ACTH stimulation increased GIPR expression, which, in some cases, could lead to GIP-dependent cortisol production and adrenal growth through the activation of GIPR signaling cascade. However, other reports including ours failed to detect the presence of the GIPR message in ACTH-dependent CS and Cushing’s disease by sensitive RT-PCR methods (8, 12). More studies are clearly needed before any conclusion can be drawn on this important issue. Lastly, our xenotransplantation mouse model should make it possible to explore new therapeutic treatments of such adrenocortical masses.
Taken together, our data demonstrate that the ectopic expression of a G protein-coupled receptor gene is a sufficient genetic event to initiate a gain-of-function leading to hyperplasic transformation of adrenocortical cells. Additional studies need to be undertaken to explain the molecular mechanism that drives the ectopic expression of a normal GPCR gene in adrenocortical cells.
Acknowledgments
We thank Dr. T. Usdin for the plasmid pCMV-IRES-rGIPR, Dr. T. Kieffer for the GIPR antibody, C. Guillermet for her assistance in immunostaining, and Dr. P. Faure and M. Martinie for performing the ACTH immunoradiometric assay. We are indebted to Drs. P. J. Hornsby, J. B. Calixto, and P. Rodien for critical reading of the manuscript.
Footnotes
This work was supported by Institut National de la Sante et de la Recherche Medicale, Commissariat a l’Energie Atomique (DSV/DRDC), Fondation de France (Research Grant 2004012837 to M.T.), Association pour la Recherche sur le Cancer (ARC, France, Research Grant 4713 to M.T.), and Programe Hosptalier de Recherche Clinique (Grant AOM 02068) to the COMETE Network. T.L.M. was supported in succession by doctoral grants from the Association pour la Recherche sur le Cancer, the Fondation pour la Recherche Medicale, and the Coordenaao de Aperfeioamento de Pessoal de Nivel Superior, Ministerio de Educaao, Brazil.
First Published Online October 27, 2005
Abbreviations: AC, Adenyl cyclase; AIMAH, ACTH-independent macronodular adrenal hyperplasia; CS, Cushing’s syndrome; CYP17, 17-hydroxylase; FACS, fluorescence-activated cell sorting; FGF, fibroblast growth factor; GIP, gastric inhibitory polypeptide; GIPR, GIP receptor; GPCR, G protein-coupled receptor; 3HSD, --hydroxy-5-steroid dehydrogenase/isomerase-1; MC2R, melanocortin 2 receptor;
Accepted for publication October 14, 2005.
References
Spiegel AM 1998 G proteins, receptors, and disease. Totawa, NJ: Humana Press
Lacroix A, Baldacchino V, Bourdeau I, Hamet P, Tremblay J 2004 Cushing’s syndrome variants secondary to aberrant hormone receptors. Trends Endocrinol Metab 15:375–382
Groussin L, Perlemoine K, Contesse V, Lefebvre H, Tabarin A, Thieblot P, Schlienger JL, Luton JP, Bertagna X, Bertherat J 2002 The ectopic expression of the gastric inhibitory polypeptide receptor is frequent in adrenocorticotropin-independent bilateral macronodular adrenal hyperplasia, but rare in unilateral tumors. J Clin Endocrinol Metab 87:1980–1985
Lu M, Wheeler MB, Leng XH, Boyd 3rd AE 1993 The role of the free cytosolic calcium level in -cell signal transduction by gastric inhibitory polypeptide and glucagon-like peptide I(7–37). Endocrinology 132:94–100
Volz A, Goke R, Lankat-Buttgereit B, Fehmann HC, Bode HP, Goke B 1995 Molecular cloning, functional expression, and signal transduction of the GIP-receptor cloned from a human insulinoma. FEBS Lett 373:23–29
Kubota A, Yamada Y, Yasuda K, Someya Y, Ihara Y, Kagimoto S, Watanabe R, Kuroe A, Ishida H, Seino Y 1997 Gastric inhibitory polypeptide activates MAP kinase through the wortmannin-sensitive and -insensitive pathways. Biochem Biophys Res Commun 235:171–175
Trumper A, Trumper K, Trusheim H, Arnold R, Goke B, Horsch D 2001 Glucose-dependent insulinotropic polypeptide is a growth factor for (INS-1) cells by pleiotropic signaling. Mol Endocrinol 15:1559–1570
Chabre O, Liakos P, Vivier J, Chaffanjon P, Labat-Moleur F, Martinie M, Bottari SP, Bachelot I, Chambaz EM, Defaye G, Feige JJ 1998 Cushing’s syndrome due to a gastric inhibitory polypeptide-dependent adrenal adenoma: insights into hormonal control of adrenocortical tumorigenesis. J Clin Endocrinol Metab 83:3134–3143
Lebrethon MC, Avallet O, Reznik Y, Archambeaud F, Combes J, Usdin TB, Narboni G, Mahoudeau J, Saez JM 1998 Food-dependent Cushing’s syndrome: characterization and functional role of gastric inhibitory polypeptide receptor in the adrenals of three patients. J Clin Endocrinol Metab 83:4514–4519
Luton JP, Bertherat J, Kuhn JM, Bertagna X 1998 [Aberrant expression of the GIP (gastric inhibitory polypeptide) receptor in an adrenal cortical adenoma responsible for a case of food-dependent Cushing’s syndrome]. Bull Acad Natl Med 182:1839–1849; discussion 1849–1850 (French)
N'Diaye N, Hamet P, Tremblay J, Boutin JM, Gaboury L, Lacroix A 1999 Asynchronous development of bilateral nodular adrenal hyperplasia in gastric inhibitory polypeptide-dependent Cushing’s syndrome. J Clin Endocrinol Metab 84:2616–2622
N'Diaye N, Tremblay J, Hamet P, De Herder WW, Lacroix A 1998 Adrenocortical overexpression of gastric inhibitory polypeptide receptor underlies food-dependent Cushing’s syndrome. J Clin Endocrinol Metab 83:2781–2785
Usdin TB, Mezey E, Button DC, Brownstein MJ, Bonner TI 1993 Gastric inhibitory polypeptide receptor, a member of the secretin- vasoactive intestinal peptide receptor family, is widely distributed in peripheral organs and the brain. Endocrinology 133:2861–2870
Wheeler MB, Gelling RW, McIntosh CH, Georgiou J, Brown JC, Pederson RA 1995 Functional expression of the rat pancreatic islet glucose-dependent insulinotropic polypeptide receptor: ligand binding and intracellular signaling properties. Endocrinology 136:4629–4639
Hamet P, Larochelle P, Franks DJ, Cartier P, Bolte E 1987 Cushing syndrome with food-dependent periodic hormonogenesis. Clin Invest Med 10:530–533
Lacroix A, Bolte E, Tremblay J, Dupre J, Poitras P, Fournier H, Garon J, Garrel D, Bayard F, Taillefer R 1992 Gastric inhibitory polypeptide-dependent cortisol hypersecretion: a new cause of Cushing’s syndrome. N Engl J Med 327:974–980
Reznik Y, Allali-Zerah V, Chayvialle JA, Leroyer R, Leymarie P, Travert G, Lebrethon MC, Budi I, Balliere AM, Mahoudeau J 1992 Food-dependent Cushing’s syndrome mediated by aberrant adrenal sensitivity to gastric inhibitory polypeptide. N Engl J Med 327:981–986
Croughs RJ, Zelissen PM, van Vroonhoven TJ, Hofland LJ, N'Diaye N, Lacroix A, de Herder WW 2000 GIP-dependent adrenal Cushing’s syndrome with incomplete suppression of ACTH. Clin Endocrinol (Oxf) 52:235–240
de Herder WW, Hofland LJ, Usdin TB, de Jong FH, Uitterlinden P, van Koetsveld P, Mezey E, Bonner TI, Bonjer HJ, Lamberts SW 1996 Food-dependent Cushing’s syndrome resulting from abundant expression of gastric inhibitory polypeptide receptors in adrenal adenoma cells. J Clin Endocrinol Metab 81:3168–3172
Pralong FP, Gomez F, Guillou L, Mosimann F, Franscella S, Gaillard RC 1999 Food-dependent Cushing’s syndrome: possible involvement of leptin in cortisol hypersecretion. J Clin Endocrinol Metab 84:3817–3822
Tsagarakis S, Tsigos C, Vassiliou V, Tsiotra P, Pratsinis H, Kletsas D, Trivizas P, Nikou A, Mavromatis T, Sotsiou F, Raptis S, Thalassinos N 2001 Food-dependent androgen and cortisol secretion by a gastric inhibitory polypeptide-receptor expressive adrenocortical adenoma leading to hirsutism and subclinical Cushing’s syndrome: in vivo and in vitro studies. J Clin Endocrinol Metab 86:583–589
Antonini SR, N'Diaye N, Baldacchino V, Hamet P, Tremblay J, Lacroix A 2004 Analysis of the putative regulatory region of the gastric inhibitory polypeptide receptor gene in food-dependent Cushing’s syndrome. J Steroid Biochem Mol Biol 91:171–177
Thomas M, Northrup SR, Hornsby PJ 1997 Adrenocortical tissue formed by transplantation of normal clones of bovine adrenocortical cells in scid mice replaces the essential functions of the animals’ adrenal glands. Nat Med 3:978–983
Thomas M, Hornsby PJ 1999 Transplantation of primary bovine adrenocortical cells into scid mice. Mol Cell Endocrinol 153:125–136
Thomas M, Suwa T, Yang L, Zhao L, Hawks CL, Hornsby PJ 2002 Cooperation of hTERT, SV40 T antigen and oncogenic Ras in tumorigenesis: a cell transplantation model using bovine adrenocortical cells. Neoplasia 4:493–500
Duperray A, Chambaz EM 1980 Effect of prostaglandin E1 and ACTH on proliferation and steroidogenic activity of bovine adreno-cortical cells in primary culture. J Steroid Biochem 13:1359–1364
Vander Heiden MG, Thompson CB 1999 Bcl-2 proteins: regulators of apoptosis or of mitochondrial homeostasis Nat Cell Biol 1:E209–E216
Forough R, Xi Z, MacPhee M, Friedman S, Engleka KA, Sayers T, Wiltrout RH, Maciag T 1993 Differential transforming abilities of non-secreted and secreted forms of human fibroblast growth factor-1. J Biol Chem 268:2960–2968
Lewis JT, Dayanandan B, Habener JF, Kieffer TJ 2000 Glucose-dependent insulinotropic polypeptide confers early phase insulin release to oral glucose in rats: demonstration by a receptor antagonist. Endocrinology 141:3710–3716
Negoescu A, Labat-Moleur F, Defaye G, Mezin P, Drouet C, Brambilla E, Chambaz EM, Feige JJ 1995 Contribution of apoptosis to the phenotypic changes of adrenocortical cells in primary culture. Mol Cell Endocrinol 110:175–184
Arnaldi G, Angeli A, Atkinson AB, Bertagna X, Cavagnini F, Chrousos GP, Fava GA, Findling JW, Gaillard RC, Grossman AB, Kola B, Lacroix A, Mancini T, Mantero F, Newell-Price J, Nieman LK, Sonino N, Vance ML, Giustina A, Boscaro M 2003 Diagnosis and complications of Cushing’s syndrome: a consensus statement. J Clin Endocrinol Metab 88:5593–5602
Stenzel-Poore MP, Cameron VA, Vaughan J, Sawchenko PE, Vale W 1992 Development of Cushing’s syndrome in corticotropin-releasing factor transgenic mice. Endocrinology 130:3378–3386
Tronche F, Kellendonk C, Kretz O, Gass P, Anlag K, Orban PC, Bock R, Klein R, Schutz G 1999 Disruption of the glucocorticoid receptor gene in the nervous system results in reduced anxiety. Nat Genet 23:99–103
Yano H, Readhead C, Nakashima M, Ren SG, Melmed S 1998 Pituitary-directed leukemia inhibitory factor transgene causes Cushing’s syndrome: neuro-immune-endocrine modulation of pituitary development. Mol Endocrinol 12:1708–1720
Thomas M, Keramidas M, Monchaux E, Feige JJ 2004 Dual hormonal regulation of endocrine tissue mass and vasculature by adrenocorticotropin in the adrenal cortex. Endocrinology 145:4320–4329
Swords FM, Aylwin S, Perry L, Arola J, Grossman AB, Monson JP, Clark AJ 2005 The aberrant expression of the gastric inhibitory polypeptide (GIP) receptor in adrenal hyperplasia: does chronic adrenocorticotropin exposure stimulate up-regulation of GIP receptors in Cushing’s disease J Clin Endocrinol Metab 90:3009–3016
Dallman MF 1984 Control of adrenocortical growth in vivo. Endocr Res 10:213–242
Hornsby PJ 1984 Regulation of adrenocortical cell proliferation in culture. Endocr Res 10:259–281
Noon LA, Franklin JM, King PJ, Goulding NJ, Hunyady L, Clark AJ 2002 Failed export of the adrenocorticotrophin receptor from the endoplasmic reticulum in non-adrenal cells: evidence in support of a requirement for a specific adrenal accessory factor. J Endocrinol 174:17–25
Rohrer DK, Kobilka BK 1998 G protein-coupled receptors: functional and mechanistic insights through altered gene expression. Physiol Rev 78:35–52(Tnia L. Mazzuco, Olivier Chabre, Nathali)
Commissariat a l’Energie Atomique, Departement Reponse et Dynamique Cellulaires, Laboratoire ANGIO (T.L.M., O.C., J.-J.F., M.T.), 38054 Grenoble, France
Centre Hospitalier Regional Universitaire de Grenoble, Departement de Diabetologie, Urologie, Nephrologie, et Endocrinologie, Service d’Endocrinologie (T.L.M., O.C.)
Departement d’Anatomie et de Cytologie Pathologique, Laboratoire de Pathologie Cellulaire (N.S.), 38043 Grenoble, France
Abstract
Aberrant expression of ectopic G protein-coupled receptors (GPCRs) in adrenal cortex tissue has been observed in several cases of ACTH-independent macronodular adrenal hyperplasias and adenomas associated with Cushing’s syndrome. Although there is clear clinical evidence for the implication of these ectopic receptors in abnormal regulation of cortisol production, whether this aberrant GPCR expression is the cause or the consequence of the development of an adrenal hyperplasia is still an open question. To answer it, we genetically engineered primary bovine adrenocortical cells to have them express the gastric inhibitory polypeptide receptor. After transplantation of these modified cells under the renal capsule of adrenalectomized immunodeficient mice, tissues formed had their functional and histological characteristics analyzed. We observed the formation of an enlarged and hyperproliferative adenomatous adrenocortical tissue that secreted cortisol in a gastric inhibitory polypeptide-dependent manner and induced a mild Cushing’s syndrome with hyperglycemia. Moreover, we show that tumor development was ACTH independent. Thus, a single genetic event, inappropriate expression of a nonmutated GPCR gene, is sufficient to initiate the complete phenotypic alterations that ultimately lead to the formation of a benign adrenocortical tumor.
Introduction
DEFECTS IN THE expression or activity of G protein-coupled receptors (GPCRs) have been identified in a wide variety of endocrine disorders (1). Several natural mutations of GPCRs have been characterized that result in receptor loss-of-function, whereas other mutations constitutively stimulate G protein coupling or alter the specificity of the binding domain, resulting in gain-of-function (1). Aberrant expression of a GPCR in a tissue that normally does not express or poorly expresses it is another mechanism leading to endocrine diseases, which is observed in particular in several variants of Cushing’s syndrome (CS). CS results from lasting exposure to excess glucocorticoids and is either ACTH dependent or independent. Recent studies have clearly established that some cortisol-producing ACTH-independent macronodular adrenal hyperplasia (AIMAH) and adenomas are controlled by the aberrant adrenocortical expression of ectopic or hyperactive eutopic GPCRs (2). In such cases, in vitro studies have demonstrated that cortisol secretion becomes driven by an unusual hormone that activates the adenylyl cyclase (AC)/cAMP signaling cascade normally triggered by the ACTH receptor. During the last decade, the GPCRs for gastric inhibitory polypeptide (GIP), catecholamines, TSH, vasopressin, serotonin, and LH have been shown to be overexpressed in several cases of unusual AIMAH and adenomas (2). However, although there is clear clinical evidence for the implication of these ectopic receptors in the dysregulated cortisol production observed in these pathologies, whether this aberrant GPCR expression is the cause or the consequence of the development of an adrenal benign tumor is still an unanswered question. We chose to address this question in the present work using the GIP receptor as a model GPCR, because it appears to be one of the most commonly encountered receptors in vitro among abnormally expressed GPCR-dependent CS (3).
GIP is a gastrointestinal hormone that is released during meals. GIP acts on pancreatic -cells to stimulate insulin release through binding to a seven-transmembrane domain GPCR. Once bound by GIP, this pancreatic -cell receptor activates AC, increases cAMP production, and ultimately leads to the elevation of intracellular calcium and insulin exocytosis (4, 5). Some studies have indicated that GIP may contribute to -cell growth (6, 7). Moreover, the GIP receptor (GIPR) gene has been found to be expressed at a lower level in brain, intestine, adipose tissue, and heart but not in normal adrenal cortex (8, 9, 10, 11, 12, 13, 14). Ectopic GIPR expression in the adrenal cortex is observed in food-dependent CS associated with benign adrenal tumors (15). The pathophysiology of this syndrome is characterized by a hypercortisolism induced by physiological postprandial increase in circulating GIP concentrations (16, 17). Abnormal GIPR expression has also been described in other cases of benign adrenal tumors (8, 9, 11, 15, 18, 19, 20, 21) but was not observed in a small cohort of malignant adrenal tissues (3). No mutation in GIPR gene coding sequence and promoter could be identified so far (12, 22), and the molecular mechanisms of this ectopic GIPR expression remain unclear.
To investigate the role of ectopic GIPR expression in the development of benign adrenocortical tumors, we used an in vivo model of cell transplantation and tissue reconstruction (23). In this model, primary bovine adrenocortical cells are transplanted under the renal capsule of adrenalectomized immunodeficient mice reconstituting a vascularized and functional adrenocortical tissue, which secretes cortisol and avoids the otherwise lethal effect of adrenalectomy (24). We have shown previously that retroviral introduction of a limited number of transforming genes (hTERT, SV40 large T, and ras val12) in the primary cells before transplantation leads to the formation of an invasive tumor (25). The introduction of single genes or combinations, other than all three, leads to the formation of tissue with various degrees of abnormality, which is neither continuously expanding nor invasive (25). In the present work, we genetically engineered primary bovine adrenocortical cells to have them abnormally express the GIPR and analyzed the functional and proliferative characteristics of the tissue formed after cell transplantation. In this way, we could directly evaluate whether the unique genetic alteration that results in ectopic expression of the GIPR in adrenal cortex is the cause of both benign adrenocortical mass and aberrant cortisol production observed in the food-dependent CS. Our results clearly validate this hypothesis.
Materials and Methods
GIPR vector construction and retrovirus production
The rat GIPR cDNA initially cloned in a modified pCMV vector (13) was amplified by PCR with primers designed for creation of HindIII and ClaI restriction sites at 5' and 3' ends, respectively. The 1.4-kb DNA fragment was subcloned into the pGEM-T Easy cloning vector (Promega, Madison WI); it corresponds to the GIPR cDNA coding sequence 163-1534, which includes part of exon 2 and exons 3–14 (according to sequence number L19660, EMBL database). The sequence-verified GIPR cDNA was transferred from construct rGIPR-pGEM-T Easy to a pLNCX2 retroviral expression vector (Clontech, Palo Alto, CA), downstream of its immediate-early cytomegalovirus promoter. Retroviral vector without insert was used as a negative control. The packaging PT67 cells (Clontech) were transfected with vectors and used for the production of replication-incompetent viruses. After 7 d of 400 μg/ml G418 antibiotic selection, viral supernatant was passed through a 0.45-μm-syringe filter to obtain cell-free viruses for adrenocortical cell infection.
Growth of bovine adrenocortical cells in culture and retroviral transduction
Primary adrenocortical cells were prepared by enzymatic digestion of adrenal glands from 2-yr-old steers, as previously described (26). Primary cell suspensions were stored frozen in liquid nitrogen. Frozen cells were thawed and plated in DMEM/Ham’s F-12 1:1 supplemented with 10% fetal calf serum, 10% horse serum, and 1% (vol/vol) UltroSer G (Biosepra, Villeneuve-la-Garenne, France). Adrenocortical cells were infected for 6 h by adding the filtered medium containing retroviruses at a ratio of 1:1 with fresh culture medium. Infected cells were selected with G418 for 7 d generating either control cells or GIPR cells.
In vitro studies
cAMP production and cortisol secretion by transduced cells were measured after 2 h of incubation with 10 μM forskolin (Sigma, Saint Quentin Fallavier, France), 100 nM ACTH (1–24) (Neosystem, Strasbourg, France), 100 nM GIP (Bachem, Voisins-le-Bretonneux, France), or fresh medium. Cortisol secretion was assayed directly on culture medium by a specific RIA with a cortisol antiserum (Endocrine Sciences, Calabasas, CA). cAMP production was measured from the supernatant of homogenized cells in the presence of 1 mmol/liter IBMX using an ELISA kit (Neogen Corp., Lansing, MI) following the manufacturer’s recommendations.
For cell cycle analysis by fluorescence-activated cell sorting (FACS), cells were grown for 72 h in serum-free medium (basal condition). During the last 24 h, 100 nM GIP or ACTH was added in some plates. A positive control of proliferation was performed by adding 20% serum to the culture medium. Before FACS analysis, cells were incubated with 20 μg/ml propidium iodide and 100 μg/ml RNase A. In each assay, 25,000 propidium iodide-positive cells were analyzed by FACScan using the CellQuest program (Becton Dickinson, Franklin Lakes, NJ). This experiment was repeated three times.
Gene expression analysis was assessed after total RNA extraction from cultured cells (RNeasy Mini Kit; QIAGEN, Courtaboeuf, France) and cDNA synthesis from 1 μg RNA with ImProm-II reverse transcriptase (Promega) using random primers. Semiquantitative PCR was performed with specific primers for human and bovine GIPR (forward primer, 5'-ATCTGCTGGTGGTTGTGAGACG-3'; reverse primer, 5'-ACGGTTCCTACGCCTATTTC-3'), bovine melanocortin 2 receptor (MC2R) (forward primer, 5'-GGGGTTTTGGAGAACCTGATGG-3'; reverse primer, 5'-GGCGACTGGCGATGTAGTGTTA-3'), human MC2R (forward primer, 5'-GACTGTCCTCGTGTGGTTTTGC-3'; reverse primer, 5'-CTTTGGTGTCGGCTACTGTAGTA-3'), bovine 3--hydroxy-5-steroid dehydrogenase/isomerase-1 (3HSD) (forward primer, 5'-GAGACCATCATGAACGTCAA-3'; reverse primer, 5'-GTCCTTCGAGTGATAAAGGT-3'), bovine 17-hydroxylase (CYP17) (forward primer, 5'-GGCCCCATCTATTCCTTTCGT-3'; reverse primer, 5'-CGTCTGTTATTGTTACGACCG-3'), bovine Bax (forward primer, 5'-TGCTTCAGGGTTTCATCCAG-3'; reverse primer, 5'-GGAGAGGATGAAACCCTGTG-3'), bovine Bcl-xL (forward primer, 5'-GGATAGCCCTGCTGTGAATGG-3'; reverse primer, 5'-CACACCAAGACGACCCGAGT-3'), bovine Col-1A2 (forward primer, 5'-CAACCATGCCTCTCAGAACA-3'; reverse primer, 5'-TGTACCTACTCCTTTGACCG-3'), and RP-L27 (forward primer, 5'-GAACATTGATGATGGCACCTC-3'; reverse primer, 5'-GGGGATATCCACAGAGTACC-3') genes. Total RNA extraction from transplanted tissues was also performed to analyze GIPR gene expression by RT-PCR. Amplification products were separated by 2% agarose gel electrophoresis, scanned with a FluorImager (Molecular Dynamics, Sunnyvale, CA), and normalized to RP-L27. An apoptosis index was calculated based on the ratio of expression levels of proapoptotic Bax and antiapoptotic Bcl-xL (27).
Cell transplantation and animal experimentation
Immunodeficient RAG 2 –/– mice originally purchased from CDTA (Orleans, France) were maintained in our animal facility as a breeding colony. All animal studies were approved by the institutional guidelines and the European Community for the Use of Experimental Animals. Under Avertin anesthesia, male and female mice at an age between 8 and 12 wk (22 g body weight) were adrenalectomized, and 2 x 106 control or GIPR adrenocortical cells were transplanted under the renal capsule (23, 24) together with 4 x 105 fibroblast growth factor (FGF)-1-secreting 3T3 cells pretreated by 2 μg/ml mitomycin C (Sigma) to prevent their further proliferation The 3T3 cell line stably expressed FGF-1 fused in frame with a signal peptide from hst/KS3 gene, yielding a highly angiogenic secreted product (28). Postoperative animal care consisted of administration of an analgesic and a mixture of antibiotics in the drinking water for 7 d. After surgery, the mouse body weight was surveyed on a daily basis during the first week and then twice a week until euthanasia.
After 4 wk, tail blood samples were taken at time 0 and 15 min after the injection of ACTH (1–39) (Neosystem, 42 pmol/g body weight) or GIP (50 pmol/g body weight). Plasma cortisol and ACTH were measured by RIA and immunoradiometric assay, respectively. Blood glucose values were determined from whole blood using an automatic glucose monitor (Accu-Chek Active, Roche, France). Blood samples were collected in the afternoon, during the physiological fasting period and nadir of cortisol level. After 8 wk, the animals were killed and the kidneys bearing the adrenocortical transplant were excised for histological analysis.
In some experiments, 1 μg/g·d dexamethasone (Sigma) was delivered to animals by sc osmotic mini-pump (Alzet, Cupertino, CA) implanted at the time of cell transplantation to suppress pituitary ACTH secretion. Mice were killed after 14 d of dexamethasone or excipient (cyclodextrin) treatment. Cardiac blood samples were collected, and transplants were processed for histological analysis.
Histological studies
Paraformaldehyde-fixed tissues were paraffin embedded and sectioned using standard methods. Sections were stained with hematoxylin and eosin. Picro-Sirius Red staining was used to localize collagen. Tissues were immunostained with antibodies to CYP17, 3HSD, GIPR (29), and Ki-67 (Dako A/S, Trappes, France), with biotinylated antimouse or antirabbit secondary antibodies as required, followed by detection using an avidin-biotin-peroxidase complex and diaminobenzidine (Dako A/S). For 3HSD-antibody detection we used the Fast Red chromogen as a substrate. DNA fragmentation associated with apoptosis was detected by nick end labeling of sections using the TdT-FragEL kit (Oncogene Research Products, Boston, MA).
The number of Ki-67-positive nuclei per 100 adrenocortical cell nuclei was used as the proliferation index. Counting was performed manually, using two nonconsecutive tissue sections per tissue sample, selected at random in both groups (control and GIPR tissues); n = 10 sections of each group. Overall comparison was performed with graphical statistical summary and Student’s t test.
Results
GIPR expression in adrenocortical cells triggers cortisol secretion in vitro
We cloned the coding sequence of the rat GIPR cDNA in a retroviral vector and transduced primary cultures of bovine adrenocortical cells to induce stable ectopic expression in these cells (Fig. 1A). The rat GIPR nucleotide sequence displays 85.5% identity with the human GIPR coding sequence, and the rat and human proteins display similar affinities for GIPR (13). We first compared the phenotypes and steroidogenic capacities of the infected cells in vitro. Uninfected primary adrenocortical cells, selected stable infected cells transduced with an empty vector encoding only the drug resistance gene (named control cells hereafter), and selected stable infected cells transduced with a GIPR-expressing vector (named GIPR cells hereafter) all presented a similar morphology (Fig. 1B) and identical growth rates (data not shown) and presented contact inhibition at confluence. We analyzed also the GIPR cells grown in vitro under basal conditions and after either ACTH or GIP addition, as regards their cell cycle distribution. DNA labeling and FACS analysis revealed a similar distribution in the S-G2/M phases of the cell cycle (10.8 ± 1.2, 11.5 ± 0.3, and 10.2 ± 0.3%, respectively). The control cells exhibited a similar S-G2/M distribution as the GIPR cells under the same conditions (data not shown). Under stimulation by a 20% serum-rich culture medium, both control and GIPR cells exhibited a significant increase in the percentage of cells in S-G2/M phases (28.4 ± 1.0%, P < 0.001; 29.7 ± 0.6%, P < 0.001, respectively) compared with basal conditions and after either ACTH or GIP incubation. Thus, the expression of a functional GIPR in adrenocortical cells did not induce morphological or proliferative changes in vitro but triggered cortisol secretion in response to GIP.
As shown by RT-PCR analysis, GIPR gene expression was observed only in the GIPR-infected cell population and in a human adrenocortical adenoma from a patient presenting a food-dependent CS (Fig. 1C). We confirmed the absence of eutopic GIPR gene expression in primary adrenocortical cells. The expression of MC2R (the ACTH receptor) was preserved in all three types of cells, even after the antibiotic selection steps undergone by the infected cells (Fig. 1C). We also observed that genes encoding the steroid-converting enzymes type II 3HSD and steroid CYP17, involved in cortisol biosynthesis, were unaffected by retroviral infection-mediated gene transduction and were equally expressed in all cell types (Fig. 1C).
Cortisol is the main corticosteroid produced by bovine and human adrenocortical cells, whereas mouse cells, which lack the steroid CYP17 enzyme, secrete corticosterone. However, long-term in vitro culture of adrenocortical cells may result in loss of their steroidogenic capacities, despite detectable expression of 3HSD and CYP17 genes (30). We thus ascertained that our infected cells were still hormonally responsive after the 2 wk of culture necessary for the infection and antibiotic selection steps. We measured cAMP production in response to addition of either ACTH or GIP. Upon ACTH (100 nM) stimulation, we observed a strong (7-fold) elevation of cAMP in both infected cell types (control and GIPR cells), whereas upon GIP (100 nM) stimulation, only the GIPR cells responded (16-fold cAMP elevation) (Fig. 1D). We then measured the cortisol secretion in response to ACTH or forskolin (10 μM), a direct activator of AC, which appeared to be preserved in all transduced cells (2-fold over basal, Fig. 1D). In contrast, under GIP treatment, only the GIPR cells showed a 2-fold elevation of cortisol secretion (P < 0.05) (Fig. 1D). Thus, adrenocortical steroid synthesis appears to be correlated with receptor activation and cAMP production in both control and GIPR cells, and ectopic GIPR gene expression clearly confers to adrenocortical cells the new capacity to secrete cortisol in response to GIP.
Transplantation of GIPR cells into mice induces the formation of a functional GIP-responsive transgenic adrenocortical tissue
We then implanted GIPR cells under the renal capsule of adrenalectomized immunodeficient mice, allowed the grafted cells to form a neo-organ, and first analyzed the physiological responses of the grafted mice to hormonal challenges 6 wk after transplantation. In a previous study, we have shown that implantation of unmodified primary bovine adrenocortical cells leads to the formation of a vascularized and cortisol-secreting organ that saves mice from the lethal effect of adrenalectomy (24). To assess the functional status of the tissue formed from the transplanted GIPR cells, we first intended to check the direct effect of food intake on plasma cortisol levels. However, because mice eat several times over a long period of time during darkness, it was technically difficult to monitor cortisol levels as a function of food intake. It also turned out to be impossible to check the effect of food deprivation on the plasma cortisol levels of GIPR mice because, after one night of starving, mice had significantly lost weight, making the conclusions of such an experiment extremely uncertain. We thus decided to measure plasma cortisol levels in response to injection of a physiological postprandial concentration (50 pmol/g ip) of GIP. The mice transplanted with GIPR cells did show elevated cortisol in response to GIP injection (+44%), whereas the mice transplanted with control cells did not (Fig. 2, A and B). In contrast, as expected, injection of ACTH (2 pmol/g ip) resulted in a mild but reproducible elevation (+26%) of plasma cortisol levels in GIPR mice as well as in control mice (Fig. 2B). Because GIPR mice exhibited elevated cortisol levels in response to GIP, we screened some physical and biological features that might occur in a situation of cortisol hypersecretion such as CS. Eight weeks after transplantation, the plasma ACTH levels in GIPR mice (21.4 ± 6.3 pmol/liter) were decreased to 48% of those in age-matched control mice (44.6 ± 23.5 pmol/liter). This difference did not reach the threshold of statistical significance (P > 0.05). This partial ACTH suppression is consistent with a discontinuous inhibition of pituitary corticotrope cells due to fluctuating cortisol levels, as seen in some atypical clinical presentations of CS (mild hypercortisolism, cyclical CS, and subclinical CS) (31). Because these forms of hypercortisolism can alter glucose metabolism, resulting in glucose intolerance or overt diabetes mellitus (31), we assessed the insulin sensitivity of transplanted mice by measuring glycemia during their physiological fasting period. In a series of measurements performed between d 28 and 53 after transplantation, glycemia was always higher in GIPR mice than in control mice, and the mean values were significantly different (P < 0.001; n = 9) (Fig. 2C). We also noticed growth attenuation along with weight loss in mice transplanted with GIPR cells compared with mice transplanted with control cells (–15% on d 53 after surgery). Similar growth retardation is encountered in other mouse models of CS, as observed in glucocorticoid-treated mice (our unpublished observations) and in different transgenic mouse models of CS (32, 33, 34). No remarkable changes in distribution of body fat distribution such as the presence of a buffalo hump or intraabdominal fat deposits were observed at necropsy.
Mice transplanted with GIPR cells develop a hyperplasic adrenocortical tissue
We analyzed the histology and proliferative status of the neo-organs formed in the renal capsule niche. At the time of killing (8 wk after transplantation), tissues formed from transplanted adrenocortical cells were fixed and processed for conventional histology and histochemistry. The pale tissue originating from control cells was spread over the kidney parenchyma and occupied a thin space beneath the kidney capsule (Fig. 3A). In contrast, transplanted GIPR cells gave rise to a voluminous mass that was prominently yellow in its center and white on its periphery. Cross-sections of this tissue were about 15 times larger than those of the control neo-organ, and compression of the renal parenchyma was visible (Fig. 3B). Control tissue presented a uniform structure of regular eosinophilic adrenocortical cells in close contact with the kidney parenchyma on its lower side and the renal capsule on its upper side (Fig. 3B). In contrast, GIPR tissue had an aspect of an encapsulated mass losing the contact between adrenocortical cells and the kidney surface; the tissue showed an irregular architecture with cellular pleiomorphism and nuclear atypia (Fig. 3B). Both clear, lipid-laden (fasciculata-type) cells and eosinophilic lipid-depleted (reticularis-type) cells were dispersed in a stromal reaction, without any sign of necrosis. All 14 GIPR tumors collected up to 53 d after transplantation showed a clear border between the benign tumor and the kidney without any sign of invasion. Comparatively, the histological appearance of a human GIPR adenoma is shown in the right panels (Fig. 3B) to illustrate their irregular cellular distribution consisting in sheets of fasciculata-type cells interspersed with reticularis-type cells. This human adenoma was encapsulated and had some areas composed of tumor stroma, as revealed by Picro-Sirius Red staining (Fig. 3B). A pronounced fibrous stroma was similarly observed in GIPR transplants, where a framework of connective tissue is delineated by red-stained collagen fibers. In marked contrast, only the renal capsule was stained in control tissue (Fig. 3B). The stromal reaction was a result of the in vivo development of GIPR tissue because no difference in the levels of 2 type I collagen (Col-I) gene expression was observable by RT-PCR in GIPR cells in vitro (data not shown).
Preservation of the steroidogenic phenotype of the developing neo-organs was demonstrated by immunohistochemistry using antibodies against the steroid-converting enzymes 3HSD and CYP17 (Fig. 3C). Interestingly, in the GIPR tumor, all cells with adrenocortical histological features appeared to be steroidogenic, forming islets or nodular regions between unstained stromal cells. The expression of the GIPR gene and the detection of the GIPR protein were confirmed in 8-wk-old tissues formed from transplanted GIPR cells by RT-PCR (data not shown) and by immunohistochemistry (Fig. 3D). GIPR gene and protein expressions were not detected in tissues formed from control cells (data not shown).
To further characterize the enlargement of GIPR tissue mass compared with control tissue, we checked whether this hyperplasia resulted from increased proliferation or decreased apoptosis. Regarding proliferation, we used the antibody Ki-67, which binds a nonhistone nuclear protein expressed in all cell cycle phases, except G0/early G1. Ki-67 staining was observed throughout the transplant in 17.7 ± 1.3% of cells in GIPR tissue, a value significantly higher than that observed in control tissue (3.5 ± 1.0%, P < 0.001; Fig. 4, A and B). There was no difference between the apoptotic rates of control and GIPR cells before transplantation, as demonstrated by the proapoptotic index based on expression of BCL2-like 1 gene (Bcl-xL) and BCL2-associated X gene (Bax) (Fig. 4C). The detection of apoptotic cells by terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling showed no significant differences between both types of neo-organs (data not shown), indicating that decreased apoptosis was certainly not the cause for the observed hyperplasia. Thus, cells encoding GIPR gene showed a growth advantage after transplantation in mice, leading to an enlarged hyperproliferative adrenocortical mass.
To investigate whether the growth of the GIPR transplants was ACTH independent, we used a mouse model of synthetic glucocorticoid-induced ACTH suppression. In a previous study (35), we have shown that dexamethasone treatment induces adrenal atrophy as a consequence of ACTH suppression and that this tissue regression does not result from a direct glucocorticoid effect on the adrenal cortex. Adrenalectomized mice were transplanted with control or GIPR cells and chronically perfused with dexamethasone via sc osmotic mini-pumps from the onset of transplantation. Dexamethasone dramatically reduced plasma ACTH concentration in both groups of mice (1.3 ± 0.8 pmol/liter in control mice; 2.4 ± 1.2 pmol/liter in GIPR mice) as soon as 24 h after dexamethasone treatment. When dexamethasone-treated animals transplanted with control adrenocortical cells were killed on d 14 after surgery, an atrophic transplant tissue was present at the site of injection, with reduced cellularity and disorganized architecture compared with the one in untreated control mice (Fig. 5). In contrast, GIPR transplants had developed to a similar large size in the absence as well as in the presence of ACTH (Fig. 5). The ACTH-independent development of GIPR tumors thus suggests that physiological GIP may act as a trophic factor through activation of the adrenal ectopic GIPR.
Discussion
We report here on the generation of a GIP-dependent adrenocortical mass associated with a mild CS through the transplantation of genetically engineered bovine adrenocortical cells into immunodeficient mice. Very interestingly, the endocrine adrenocortical cells giving rise to these tumors were minimally modified by the retroviral introduction into their genome of a single nonmutated gene encoding the GIPR. This clearly demonstrates that in our animal model, the ectopic adrenal expression of this AC-activating receptor (physiologically expressed in the pancreatic -cells but not in adrenocortical cells) is sufficient to induce not only aberrant cortisol secretion but also hyperproliferation and benign cell transformation. These findings give strength to the hypothesis that aberrant expression of GPCRs plays a direct role in the development of these hyperplasic tissues. This is a novel and important contribution to the understanding of food-dependent CS in humans and more generally to all cases of benign adrenal tumors associated with aberrant GPCR expression. The clinical observations collected so far in the description of food-dependent CS, vasopressin-responsive CS, catecholamine-dependent CS, postmenopausal CS, and serotonin-dependent CS have clearly demonstrated the link between these endocrine disorders and the aberrant adrenal expression of GIPR, arginine vasopressin V1-3 receptors, -adrenergic receptors, LH-human chorionic gonadotropin receptor, and serotonin 5-HT4 and 5-HT7 receptors, respectively (2). In particular, in vitro studies using primary cell cultures derived from these tumors have established that the aberrantly expressed GPCRs confer steroidogenic response to the corresponding ligands, suggesting that these receptors directly couple at the cell surface with the AC-G-proteins complex and substitute for the MC2R function. Because cortisol overproduction exerts a feedback control on pituitary ACTH secretion but does not control the circulating levels of the ectopic receptor ligand, steroidogenesis becomes uncontrolled. It is noteworthy that the transplanted animals with GIPR cells had detectable ACTH levels. This lack of suppression of pituitary ACTH had previously been reported in the documented cases of food-dependent CS and might indicate that the discontinuous elevation of plasma GIP levels inducing cortisol secretion does not always result in complete suppression of the hypothalamic-pituitary-adrenal axis (18, 21, 36).
As concerns proliferation, only very few in vitro studies have addressed this question in benign adrenal tumor-derived cells. Interestingly, in our experiment, the percentage of genetically engineered cell population expressing GIPR in the S-G2/M phases was not modified by the addition of either GIP or ACTH to the medium. Our group reported previously that GIP could weakly stimulate MAPK activity and thymidine incorporation in adrenocortical cells from an adenoma of a patient with food-dependent CS (8). Conversely, GIP has been shown to have an inhibitory action on thymidine incorporation in adrenocortical cells isolated from a food-dependent adenoma (9). This type of analysis has unfortunately not been performed in other cases of CS secondary to aberrant GPCR expression. How does stimulation of GIPR in adrenocortical cells lead to cellular hyperplasia and pleiomorphism with stromal reaction, as obtained in GIPR transplants The most evident but quite vague answer to this question is that GIP acts through the same pathways as ACTH. It is long known that in vivo, an excess of endogenous or exogenous ACTH causes adrenal hypertrophy, rapidly followed by hyperplasia (37), whereas in vitro studies using adrenal cells from several species have shown that ACTH behaves as antimitogenic leading to the notion that ACTH is an indirect mitogen in intact animals (38). Although known for several decades, the trophic action of ACTH in vivo is still poorly deciphered at the molecular level because of this paradoxical effect on proliferation. It has been proposed that FGF-2, which is expressed by adrenocortical cells and is the most potent growth factor for endocrine adrenocortical cells, is a relay of ACTH action. More recently, we described that part of this trophic effect might be indirectly mediated by the adrenocortical vasculature via vascular endothelial growth factor (35). A link between AC activation and adrenocortical cell proliferation certainly exists, probably involving such paracrine mediators acting directly or indirectly. Moreover, despite an enlarged tissue, the basal cortisol production by GIPR mouse is not higher than that observed in control mouse. This might indicate a switch toward a proliferative phenotype of the cells expressing the GIPR gene with higher sensitivity to proliferative stimuli in vivo. Thus, the generation of GIPR transgenic tissues should give us the opportunity to further study the signaling pathways involved in proliferation and the cross-talk between pathways in adrenocortical cells.
We previously noted that an important element for success of adrenocortical cell transplantation is the supply of FGF-1 from the cotransplanted FGF-secreting 3T3 cells (23, 24). The 3T3 cells were rendered incapable of further division by mitomycin C treatment. At 36 d after cell transplantation, transplants that were not formed by the inclusion of FGF-secreting cells always contained a necrotic area, although the tissue produced plasma steroid levels in the same range as the 3T3 cell-containing transplants. Tissues formed with or without 3T3 cells had very similar structures outside of the necrotic area, indicating that the 3T3 cells did not substantially change the nature of the transplant tissue (24). Labeling studies indicated that rare 3T3 cells were still present after 36 d and were located within or next to the endothelium that forms within the transplant tissue (Thomas, M., unpublished observations). We cannot, however, rule out the possibility that GIPR-expressing cells could affect the survival and the function of the cotransplanted 3T3 cells through some unknown mechanism.
Programmed cell death inhibition might contribute to the cellular hyperplasia. However, in the present study, the development of a hyperplastic neo-organ derived from GIPR cells cannot be explained by an inhibition of apoptosis because we clearly observed a long-term imbalance of cell proliferation and no change in the rate of cell attrition. A genome-wide cDNA array analysis of gene expression in the GIPR neo-organ compared with the normal tissue derived from control cells might help to understand these mechanisms, although the differences in the respective abundance of the distinct cell types present in these tissues (abundant fibrous tissue in GIPR neo-organs) might render the interpretation of these data quite difficult.
The retroviral transduction of the GIPR gene allowed the stable expression of an ectopic gene in adrenocortical cells after transplantation in mice. Some studies had demonstrated the difficulty to obtain functional GPCRs in heterologous systems, which could be expressed only in cell lines having a minimal level of endogenous expression of the GPCR studied (39, 40). In our system, we had shown that the delivery of GIPR gene in adrenocortical cells was able to trigger normal receptor trafficking to the plasma membrane (as evidenced by immunolabeling, Fig. 3D) and normal receptor coupling to Gs protein leading to AC activation and steroidogenic response. This suggests that specific pharmacological antagonists of this aberrant receptor might be sufficient to treat GPCR-dependent AIMAH and adrenal adenomas. However, such attempts at controlling cortisol secretion have proven clinically unsuccessful to reverse adrenal overgrowth (17, 19), indicating that a number of yet unidentified secondary genetic events may have also occurred in these tumors. In a recent paper, Swords et al. (36) demonstrated functional expression of GIPR in primary adrenocortical cell cultures isolated from patients presenting an ACTH-dependent adrenal hyperplasia with no clinical evidence of food-dependent cortisol secretion. The authors suggested that ACTH stimulation increased GIPR expression, which, in some cases, could lead to GIP-dependent cortisol production and adrenal growth through the activation of GIPR signaling cascade. However, other reports including ours failed to detect the presence of the GIPR message in ACTH-dependent CS and Cushing’s disease by sensitive RT-PCR methods (8, 12). More studies are clearly needed before any conclusion can be drawn on this important issue. Lastly, our xenotransplantation mouse model should make it possible to explore new therapeutic treatments of such adrenocortical masses.
Taken together, our data demonstrate that the ectopic expression of a G protein-coupled receptor gene is a sufficient genetic event to initiate a gain-of-function leading to hyperplasic transformation of adrenocortical cells. Additional studies need to be undertaken to explain the molecular mechanism that drives the ectopic expression of a normal GPCR gene in adrenocortical cells.
Acknowledgments
We thank Dr. T. Usdin for the plasmid pCMV-IRES-rGIPR, Dr. T. Kieffer for the GIPR antibody, C. Guillermet for her assistance in immunostaining, and Dr. P. Faure and M. Martinie for performing the ACTH immunoradiometric assay. We are indebted to Drs. P. J. Hornsby, J. B. Calixto, and P. Rodien for critical reading of the manuscript.
Footnotes
This work was supported by Institut National de la Sante et de la Recherche Medicale, Commissariat a l’Energie Atomique (DSV/DRDC), Fondation de France (Research Grant 2004012837 to M.T.), Association pour la Recherche sur le Cancer (ARC, France, Research Grant 4713 to M.T.), and Programe Hosptalier de Recherche Clinique (Grant AOM 02068) to the COMETE Network. T.L.M. was supported in succession by doctoral grants from the Association pour la Recherche sur le Cancer, the Fondation pour la Recherche Medicale, and the Coordenaao de Aperfeioamento de Pessoal de Nivel Superior, Ministerio de Educaao, Brazil.
First Published Online October 27, 2005
Abbreviations: AC, Adenyl cyclase; AIMAH, ACTH-independent macronodular adrenal hyperplasia; CS, Cushing’s syndrome; CYP17, 17-hydroxylase; FACS, fluorescence-activated cell sorting; FGF, fibroblast growth factor; GIP, gastric inhibitory polypeptide; GIPR, GIP receptor; GPCR, G protein-coupled receptor; 3HSD, --hydroxy-5-steroid dehydrogenase/isomerase-1; MC2R, melanocortin 2 receptor;
Accepted for publication October 14, 2005.
References
Spiegel AM 1998 G proteins, receptors, and disease. Totawa, NJ: Humana Press
Lacroix A, Baldacchino V, Bourdeau I, Hamet P, Tremblay J 2004 Cushing’s syndrome variants secondary to aberrant hormone receptors. Trends Endocrinol Metab 15:375–382
Groussin L, Perlemoine K, Contesse V, Lefebvre H, Tabarin A, Thieblot P, Schlienger JL, Luton JP, Bertagna X, Bertherat J 2002 The ectopic expression of the gastric inhibitory polypeptide receptor is frequent in adrenocorticotropin-independent bilateral macronodular adrenal hyperplasia, but rare in unilateral tumors. J Clin Endocrinol Metab 87:1980–1985
Lu M, Wheeler MB, Leng XH, Boyd 3rd AE 1993 The role of the free cytosolic calcium level in -cell signal transduction by gastric inhibitory polypeptide and glucagon-like peptide I(7–37). Endocrinology 132:94–100
Volz A, Goke R, Lankat-Buttgereit B, Fehmann HC, Bode HP, Goke B 1995 Molecular cloning, functional expression, and signal transduction of the GIP-receptor cloned from a human insulinoma. FEBS Lett 373:23–29
Kubota A, Yamada Y, Yasuda K, Someya Y, Ihara Y, Kagimoto S, Watanabe R, Kuroe A, Ishida H, Seino Y 1997 Gastric inhibitory polypeptide activates MAP kinase through the wortmannin-sensitive and -insensitive pathways. Biochem Biophys Res Commun 235:171–175
Trumper A, Trumper K, Trusheim H, Arnold R, Goke B, Horsch D 2001 Glucose-dependent insulinotropic polypeptide is a growth factor for (INS-1) cells by pleiotropic signaling. Mol Endocrinol 15:1559–1570
Chabre O, Liakos P, Vivier J, Chaffanjon P, Labat-Moleur F, Martinie M, Bottari SP, Bachelot I, Chambaz EM, Defaye G, Feige JJ 1998 Cushing’s syndrome due to a gastric inhibitory polypeptide-dependent adrenal adenoma: insights into hormonal control of adrenocortical tumorigenesis. J Clin Endocrinol Metab 83:3134–3143
Lebrethon MC, Avallet O, Reznik Y, Archambeaud F, Combes J, Usdin TB, Narboni G, Mahoudeau J, Saez JM 1998 Food-dependent Cushing’s syndrome: characterization and functional role of gastric inhibitory polypeptide receptor in the adrenals of three patients. J Clin Endocrinol Metab 83:4514–4519
Luton JP, Bertherat J, Kuhn JM, Bertagna X 1998 [Aberrant expression of the GIP (gastric inhibitory polypeptide) receptor in an adrenal cortical adenoma responsible for a case of food-dependent Cushing’s syndrome]. Bull Acad Natl Med 182:1839–1849; discussion 1849–1850 (French)
N'Diaye N, Hamet P, Tremblay J, Boutin JM, Gaboury L, Lacroix A 1999 Asynchronous development of bilateral nodular adrenal hyperplasia in gastric inhibitory polypeptide-dependent Cushing’s syndrome. J Clin Endocrinol Metab 84:2616–2622
N'Diaye N, Tremblay J, Hamet P, De Herder WW, Lacroix A 1998 Adrenocortical overexpression of gastric inhibitory polypeptide receptor underlies food-dependent Cushing’s syndrome. J Clin Endocrinol Metab 83:2781–2785
Usdin TB, Mezey E, Button DC, Brownstein MJ, Bonner TI 1993 Gastric inhibitory polypeptide receptor, a member of the secretin- vasoactive intestinal peptide receptor family, is widely distributed in peripheral organs and the brain. Endocrinology 133:2861–2870
Wheeler MB, Gelling RW, McIntosh CH, Georgiou J, Brown JC, Pederson RA 1995 Functional expression of the rat pancreatic islet glucose-dependent insulinotropic polypeptide receptor: ligand binding and intracellular signaling properties. Endocrinology 136:4629–4639
Hamet P, Larochelle P, Franks DJ, Cartier P, Bolte E 1987 Cushing syndrome with food-dependent periodic hormonogenesis. Clin Invest Med 10:530–533
Lacroix A, Bolte E, Tremblay J, Dupre J, Poitras P, Fournier H, Garon J, Garrel D, Bayard F, Taillefer R 1992 Gastric inhibitory polypeptide-dependent cortisol hypersecretion: a new cause of Cushing’s syndrome. N Engl J Med 327:974–980
Reznik Y, Allali-Zerah V, Chayvialle JA, Leroyer R, Leymarie P, Travert G, Lebrethon MC, Budi I, Balliere AM, Mahoudeau J 1992 Food-dependent Cushing’s syndrome mediated by aberrant adrenal sensitivity to gastric inhibitory polypeptide. N Engl J Med 327:981–986
Croughs RJ, Zelissen PM, van Vroonhoven TJ, Hofland LJ, N'Diaye N, Lacroix A, de Herder WW 2000 GIP-dependent adrenal Cushing’s syndrome with incomplete suppression of ACTH. Clin Endocrinol (Oxf) 52:235–240
de Herder WW, Hofland LJ, Usdin TB, de Jong FH, Uitterlinden P, van Koetsveld P, Mezey E, Bonner TI, Bonjer HJ, Lamberts SW 1996 Food-dependent Cushing’s syndrome resulting from abundant expression of gastric inhibitory polypeptide receptors in adrenal adenoma cells. J Clin Endocrinol Metab 81:3168–3172
Pralong FP, Gomez F, Guillou L, Mosimann F, Franscella S, Gaillard RC 1999 Food-dependent Cushing’s syndrome: possible involvement of leptin in cortisol hypersecretion. J Clin Endocrinol Metab 84:3817–3822
Tsagarakis S, Tsigos C, Vassiliou V, Tsiotra P, Pratsinis H, Kletsas D, Trivizas P, Nikou A, Mavromatis T, Sotsiou F, Raptis S, Thalassinos N 2001 Food-dependent androgen and cortisol secretion by a gastric inhibitory polypeptide-receptor expressive adrenocortical adenoma leading to hirsutism and subclinical Cushing’s syndrome: in vivo and in vitro studies. J Clin Endocrinol Metab 86:583–589
Antonini SR, N'Diaye N, Baldacchino V, Hamet P, Tremblay J, Lacroix A 2004 Analysis of the putative regulatory region of the gastric inhibitory polypeptide receptor gene in food-dependent Cushing’s syndrome. J Steroid Biochem Mol Biol 91:171–177
Thomas M, Northrup SR, Hornsby PJ 1997 Adrenocortical tissue formed by transplantation of normal clones of bovine adrenocortical cells in scid mice replaces the essential functions of the animals’ adrenal glands. Nat Med 3:978–983
Thomas M, Hornsby PJ 1999 Transplantation of primary bovine adrenocortical cells into scid mice. Mol Cell Endocrinol 153:125–136
Thomas M, Suwa T, Yang L, Zhao L, Hawks CL, Hornsby PJ 2002 Cooperation of hTERT, SV40 T antigen and oncogenic Ras in tumorigenesis: a cell transplantation model using bovine adrenocortical cells. Neoplasia 4:493–500
Duperray A, Chambaz EM 1980 Effect of prostaglandin E1 and ACTH on proliferation and steroidogenic activity of bovine adreno-cortical cells in primary culture. J Steroid Biochem 13:1359–1364
Vander Heiden MG, Thompson CB 1999 Bcl-2 proteins: regulators of apoptosis or of mitochondrial homeostasis Nat Cell Biol 1:E209–E216
Forough R, Xi Z, MacPhee M, Friedman S, Engleka KA, Sayers T, Wiltrout RH, Maciag T 1993 Differential transforming abilities of non-secreted and secreted forms of human fibroblast growth factor-1. J Biol Chem 268:2960–2968
Lewis JT, Dayanandan B, Habener JF, Kieffer TJ 2000 Glucose-dependent insulinotropic polypeptide confers early phase insulin release to oral glucose in rats: demonstration by a receptor antagonist. Endocrinology 141:3710–3716
Negoescu A, Labat-Moleur F, Defaye G, Mezin P, Drouet C, Brambilla E, Chambaz EM, Feige JJ 1995 Contribution of apoptosis to the phenotypic changes of adrenocortical cells in primary culture. Mol Cell Endocrinol 110:175–184
Arnaldi G, Angeli A, Atkinson AB, Bertagna X, Cavagnini F, Chrousos GP, Fava GA, Findling JW, Gaillard RC, Grossman AB, Kola B, Lacroix A, Mancini T, Mantero F, Newell-Price J, Nieman LK, Sonino N, Vance ML, Giustina A, Boscaro M 2003 Diagnosis and complications of Cushing’s syndrome: a consensus statement. J Clin Endocrinol Metab 88:5593–5602
Stenzel-Poore MP, Cameron VA, Vaughan J, Sawchenko PE, Vale W 1992 Development of Cushing’s syndrome in corticotropin-releasing factor transgenic mice. Endocrinology 130:3378–3386
Tronche F, Kellendonk C, Kretz O, Gass P, Anlag K, Orban PC, Bock R, Klein R, Schutz G 1999 Disruption of the glucocorticoid receptor gene in the nervous system results in reduced anxiety. Nat Genet 23:99–103
Yano H, Readhead C, Nakashima M, Ren SG, Melmed S 1998 Pituitary-directed leukemia inhibitory factor transgene causes Cushing’s syndrome: neuro-immune-endocrine modulation of pituitary development. Mol Endocrinol 12:1708–1720
Thomas M, Keramidas M, Monchaux E, Feige JJ 2004 Dual hormonal regulation of endocrine tissue mass and vasculature by adrenocorticotropin in the adrenal cortex. Endocrinology 145:4320–4329
Swords FM, Aylwin S, Perry L, Arola J, Grossman AB, Monson JP, Clark AJ 2005 The aberrant expression of the gastric inhibitory polypeptide (GIP) receptor in adrenal hyperplasia: does chronic adrenocorticotropin exposure stimulate up-regulation of GIP receptors in Cushing’s disease J Clin Endocrinol Metab 90:3009–3016
Dallman MF 1984 Control of adrenocortical growth in vivo. Endocr Res 10:213–242
Hornsby PJ 1984 Regulation of adrenocortical cell proliferation in culture. Endocr Res 10:259–281
Noon LA, Franklin JM, King PJ, Goulding NJ, Hunyady L, Clark AJ 2002 Failed export of the adrenocorticotrophin receptor from the endoplasmic reticulum in non-adrenal cells: evidence in support of a requirement for a specific adrenal accessory factor. J Endocrinol 174:17–25
Rohrer DK, Kobilka BK 1998 G protein-coupled receptors: functional and mechanistic insights through altered gene expression. Physiol Rev 78:35–52(Tnia L. Mazzuco, Olivier Chabre, Nathali)