| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Department of Endocrinology, Metabolism, and Cancer, Institut National de la Santé et de la Recherche Médicale Unit 567, Centre National de la Recherche Scientifique Unit Mixté de Recherche 8104, Institut Cochin (S.G., F.T., L.G., R.L., B.R., X.B., J.B.), Université Paris Descartes (S.G., F.T., L.G., R.L., B.D., X.B., J.B.), Department of Endocrinology, Center for Rare Adrenal Diseases (L.G., R.L., X.B., J.B.), Department of Pathology (F.T., P.L., A.A.) and Department of Digestive and Endocrine Surgery (S.G., B.D.), Assistance Publique Hôpitaux de Paris, Hôpital Cochin, 75014 Paris, France; and COMETE (COrtico-MEdullosurrénale Tumeurs Endocrines)-INCa (Institut National du Cancer)-Rare Adrenal Cancer Network (F.T., L.G., R.L., B.D., X.B., J.B.), 75014 Paris, France
Address all correspondence and requests for reprints to: Professor Jérôme Bertherat, Service des Maladies Endocriniennes et Métaboliques, Hôpital Cochin, 27, rue du Faubourg Saint-Jacques, 75014 Paris, France. E-mail: jerome.bertherat{at}cch.aphp.fr.
| Abstract |
|---|
|
|
|---|
Objective: The objective of the investigation was to study Wnt/β-catenin dysregulation in adrenocortical tumors (ACTs) with cAMP pathway genetic alteration and search for secondary CTNNB1 somatic mutations in heterogeneous tumors.
Patients and methods: Nine PPNADs, including five with macronodules, three ACAs with PRKAR1A somatic mutations, and one heterogeneous tumor with ACC developed within an ACA, were studied by immunohistochemistry and DNA sequencing.
Results: β-Catenin accumulation was observed in all PPNADs, ACAs with PRKAR1A mutations, and the ACC component of the heterogeneous tumor. CTNNB1 somatic activating mutations were found in the macronodule of two of the five macronodular PPNADs, in one ACA with a PRKAR1A somatic mutation, and in the malignant part of the heterogeneous ACT.
Conclusions: The Wnt/β-catenin pathway is activated in PPNADs and ACAs with PRKAR1A mutations, suggesting a cross talk between the cAMP and Wnt/β-catenin pathways in ACT development. In addition, the occurrence as an additional hit of a CTNNB1 somatic mutation is associated with larger or more aggressive ACTs. This underlines the importance of the Wnt/β-catenin pathway in adrenal cortex tumorigenesis and the importance of genetic accumulation in the progression of ACTs.
| Introduction |
|---|
|
|
|---|
The cAMP pathway is involved in primary pigmented nodular adrenocortical disease (PPNAD) that can be observed in Carney complex, and rare ACAs through mutation of the regulatory subunit R1A of the protein kinase A (PRKAR1A) gene, leading to stimulation of protein kinase A (PKA) activity (9).
This study was undertaken with the hypothesis that Wnt/β-catenin signaling pathway activation may be involved in ACTs with cAMP/PKA pathway alteration. In parallel, we searched for somatic CTNNB1 mutations as a secondary event in ACT development.
| Patients and Methods |
|---|
|
|
|---|
Nine patients (seven women; aged 11–54 yr; urinary cortisol 76–934 µg/d) with PPNAD and three female patients (aged 42–81 yr, urinary cortisol 202–322 µg/d) presenting an ACA with a PRKAR1A somatic mutation were studied. Indication for surgery was Cushing syndrome. Additionally, one male patient with an ACC identified as having a heterogeneous macroscopic tumor was also included. As control, 26 subjects with ACA without PRKAR1A mutation, including seven with CTNNB1 mutation, previously reported (7), were used. The hormonal investigations and tumor collection were performed as previously reported (10, 11). Informed signed consent was obtained from all the patients, and the study was approved by our institutional review board.
Microscopy and immunohistochemical staining
Sections, diagnosis, scoring, and immunohistochemistry for β-catenin were performed as previously described (7, 12). Immunohistochemical labeling was evaluated for the presence of membranous, cytoplasmic, and nuclear staining. Cytoplasmic accumulation was considered diffuse if more than 80% of the cells were positive and nuclear accumulation focal if less than 5% of the nuclei were positive.
Nucleic acid extraction and mutation analysis of PRKAR1A and CTNNB1
DNA was extracted from peripheral blood leukocytes and tumor as previously reported (9). For mutation analysis, the 12 exons and the flanking intronic sequences of PRKAR1A, and exon 3 and the flanking intronic sequences of CTNNB1 were amplified by PCR (10, 13). In case of CTNNB1 mutation in PPNAD, DNA from micronodules of the contralateral gland, obtained by macroscopic dissection, was also sequenced.
Protein extraction and analysis
Protein extraction and Western blot were performed as previously described (14). Blots were treated with primary β-catenin antibody (BD Biosciences, San Diego, CA) at a 1:2000 dilution for a night at 4 C.
| Results |
|---|
|
|
|---|
Clinical, hormonal, and radiological investigations of patients with ACTs are summarized in Table 1
. Patient 13 had an incidentally found 5-cm heterogeneous ACT: one component was an ACC with a Weiss score of 6, developed within an ACA with a Weiss score of 1.
|
In eight of the eight PPNAD available, cytoplasmic diffuse and nuclear focal β-catenin accumulation was observed in all the nodules, whatever their size, and in internodular areas (Table 1
and Fig. 1
). No accumulation was observed in the cells of the adrenal medulla. Cytoplasmic diffuse and nuclear focal accumulation was observed in macronodules and micronodules, whatever their CTNNB1 mutational status, and internodular area. Intensity of the staining was everywhere equivalent except for tumor 4, in which the staining was slightly stronger in the mutated nodule than in inter-nodular area.
|
For tumor 13 with a benign and malignant component, cytoplasmic diffuse and nuclear focal (4%) β-catenin accumulation was observed in the malignant component (86%), whereas only focal cytoplasmic staining was seen in the benign component (9%) (Fig. 1
).
Genetic analysis
The results of PRKARIA gene and CTNNB1 sequencing are summarized in Table 1
.
Seven of the nine patients with PPNAD presented a germline PRKAR1A mutation. A CTNNB1 somatic activating mutation was observed in two PPNADs, only in macronodules, defined as nodule larger than 10 mm. No CTNNB1 mutation has been found in micronodules or in the nodules of the controlateral adrenals of the two mutated macronodules. In the five PPNADs with macronodule, two have CTNNB1 somatic mutations, and two of the remaining present with unique genetic alterations: a somatic PRKAR1A mutation in one case (7) and a germline PRKAR1A mutation that give rise to a truncated expressed PRKAR1A mutant protein in the other case. This last mutant protein was previously reported as associated with a more aggressive form of Carney complex (9).
Among the three sporadic ACAs with somatic PRKAR1A mutation, the larger one presented a CTNNB1 somatic activating mutation.
In the ACC (no. 13), sequencing of the exon 3 of the tumoral DNA revealed a not previously reported heterozygous p.Tyr30X CTNNB1 mutation in the malignant component and in the metastasis but not in the benign component (details in Fig. 1
). No TP53 somatic mutation was found in these ACCs, ACAs, or metastasis (data not shown).
β-catenin Western blot analysis
To detect a shortened β-catenin mutant protein, a Western blot analysis of cellular extract was made using the malignant component, the apparently benign ACA part of the heterogeneous tumor, and one of its liver metastasis. A shorter β-catenin mutant protein, weighing around 75 kDA, was detected in the ACC part and the metastasis but not in the ACA component (Fig. 1
). Because β-catenin was accumulated in the tumor and β-catenin antibody used reacts against the C-terminal portion of the protein, we hypothesize that this nonsense mutation leads to an alternative Kozak sequence (alternative ATG), coding for a shortened protein. The first alternative Kozak sequences after the stop mutation and without frameshift are localized in exon 4 after the phosphorylation sites, necessary to its degradation.
| Discussion |
|---|
|
|
|---|
It has been shown that PKA phosphorylates Ser 9 of glycogen synthase kinase-3β (15) and inhibits its activity in the Axin complex, leading to the stabilization of β-catenin. This mechanism could contribute to the accumulation of β-catenin in PPNADs. Another possible cross talk between the cAMP and the Wnt/ β-catenin signaling pathway could be due to a direct phosphorylation of β-catenin on Ser 675 by PKA, inhibiting the degradation of β-catenin, or promoting its binding to its transcriptional coactivator, cAMP response element-binding protein-binding protein (16, 17). It is likely that mutations of PRKAR1A would have the same consequences in PPNADs and ACAs. This cross talk between cAMP and Wnt/β-catenin pathways is supported by real-time PCR experiments in the adrenocortical H295R cells showing that 6 h of forskolin treatment stimulates by 2-fold the cDNA levels of AXIN2, one of the main targets of Wnt/β-catenin (data not shown).
Wnt/β-catenin signaling activation by somatic CTNNB1 mutation was observed in two of the nine PPNADs studied. Interestingly, in both cases these mutations were detected only in macronodules. Considering that a macronodule was present in five of the nine PPNAD studied, this suggests that this molecular alteration might be linked to a more aggressive phenotype. We have previously reported for one of these macronodular PPNAD (case 1) without CTNNB1 mutation, the occurrence of a somatic PRKAR1A mutation (13). This suggests that progressive accumulation of genetic alterations would favor the development of macronodules in PPNADs, a type of ACT with a very low growth potential. Interestingly, for one of the other PPNAD with macronodule and no CTNNB1 mutation, a germline PRKAR1A mutation leads to an expressed truncated protein that has been suggested to play a dominant-negative effect responsible for a more aggressive phenotype (9). However, regulation of the Wnt/β-catenin signaling pathway could involve several others mechanisms because one tumor harbors β-catenin accumulation without PRKAR1A or CTNNB1 mutation.
Interestingly, the ACA studied here harboring both a PRKAR1A and CTNNB1 somatic mutation had no histological or molecular signs of malignancy (no 17p13 allelic losses) and did not recur during follow-up (data not shown). This suggests that the additive effects of PRKAR1A and CTNNB1 mutations keep the ACT in such a differentiated state that it prevents malignant progression.
ACC development might result from the progressive appearance of multiple genetic and epigenetic alterations; however, no clear single gene mutation was reported that could take part in this process. Interestingly, in case 13, the malignant part of the tumor harbors a CTNNB1 mutation that is absent in the benign part. This ACC had a poor outcome despite its small size, leading to metastases. No TP53 mutation could explain this poor outcome (18). It is therefore tempting to speculate that a secondary genetic defect occurring in the malignant part of this ACTs could explain this transformation: CTNNB1 somatic mutation could be the culprit. However, it is most likely that multiple associated alterations could explain this phenotype.
The observation that various types of ACT are associated with CTNNB1 somatic mutations demonstrates that a single event is not sufficient to drive a specific phenotype, as previously reported by Kjellman et al. (19). However, as in the tumors studied here, CTNNB1 mutations were restricted to the portion that was the more aggressive and could lead to a more aggressive tumorigenesis. In this case CTNNB1 mutations would act as a promoting factor rather than an initiating one. Previous germline mutation leading to tumor development, with the occurrence of CTNNB1 activating mutation as a second event in some tumor, has already been described with G6Pase gene mutation in glycogenosis type 1 with liver polyadenomatosis (20). Interestingly, liver adenomas with CTNNB1 activating mutation have also a more aggressive tumorigenesis, i.e. higher risk of malignant transformation.
Conclusion
In conclusion, the Wnt/β-catenin signaling pathway is activated in PPNAD and ACA presenting PRKAR1A mutations. This suggests a cross talk between the cAMP and the Wnt/ β-catenin signaling pathways. Nevertheless, further studies are necessary to determine the mechanisms of this cross talk. The occurrence of CTNNB1 somatic mutation as a second event in ACT is associated with larger and/or more aggressive tumors. To our knowledge, this is the first clear example of multiple genetic alterations in the progression of adrenocortical tumors.
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Summary: S.G. received grant support from Académie Nationale de Médecine (2006), the Association Benoît, Malassagne (2006), and the Association pour la Recherche contre le Cancer (2007). F.T., J.B., L.G., R.L., B.R., P.L., A.A., B.D., and X.B., have nothing to declare.
First Published Online July 22, 2008
1 S.G. and F.T. contributed equally to this work. ![]()
Abbreviations: ACA, Adrenocortical adenoma; ACC, adrenocortical cancer; ACT, adrenocortical tumor; PKA, protein kinase A; PPNAD, primary pigmented nodular adrenocortical disease.
Received March 19, 2008.
Accepted July 11, 2008.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Libe, F. Tissier, M. Bienvenu, L. Groussin, S. Joannidis, C. Hignette, B. Dousset, P. Legmann, M. Faraggi, B. Richard, et al. Adrenocortical Tumor with Two Distinct Elements Revealed by Combined 18F-Fluorodeoxyglucose Positron Emission Tomography and 131I Nor-Cholesterol Scintigraphy J. Clin. Endocrinol. Metab., October 1, 2009; 94(10): 3631 - 3632. [Full Text] [PDF] |
||||
![]() |
S. Schinner, H. S Willenberg, M. Schott, and W. A Scherbaum Pathophysiological aspects of Wnt-signaling in endocrine disease Eur. J. Endocrinol., May 1, 2009; 160(5): 731 - 737. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. V. Shah, A. Muralidharan, M. Gokulgandhi, K. Soan, and S. Thomas Cadherin Switching and Activation of {beta}-Catenin Signaling Underlie Proinvasive Actions of Calcitonin-Calcitonin Receptor Axis in Prostate Cancer J. Biol. Chem., January 9, 2009; 284(2): 1018 - 1030. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |