| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Unité dEndocrinologie-Gynécologie Pédiatrique (N.K., A.E., F.A., S.L., C.S., F.P.), Service de Pédiatrie 1, Hôpital Arnaud-de-Villeneuve, and Service dHormonologie du Développement et de la Reproduction, Hôpital Lapeyronie, Centre Hospitalier Universitaire (CHU) Montpellier, 34295 Montpellier, France; Institut National de la Santé et de la Recherche Médicale U540 (N.K., S.L., C.S.), Hormones et Cancers, 34000 Montpellier, France; Service dOncologie Pédiatrique (C.Pa.), Institut Gustave Roussy, 94805 Villejuif, France; Service dAnatomo-pathologie (P.D.), Institut Gustave Roussy, 94805 Villejuif, France; Service dEndocrinologie Pédiatrique (C.Pi.), Hôpital des Enfants, CHU Toulouse, 31059 Toulouse, France; Service dEndocrinologie Pédiatrique (E.T.), Hôpital Necker, Assistance PubliqueHôpitaux de Paris (APHP), 75743 Paris, France; Service dEndocrinologie Pédiatrique (R.B.), Hôpital Kremlin-Bicêtre, APHP, 94270 Paris, France; Service dEndocrinologie Pédiatrique (C.L.), Hôpital Charles Nicolle, CHU Rouen, 76031 Rouen, France; Service de Pédiatrie (D.P.), CHU Grenoble, 38043 Grenoble, France; Service dEndocrinologie (A.-M.G.), Hôpital Caremeau, CHU Nîmes, 30029 Nîmes, France; and Service dAnatomopathologie (P.B.), Hôpital Lapeyronie, CHU Montpellier, 34295 Montpellier, France
Address all correspondence and requests for reprints to: Prof. Charles Sultan, Unité dEndocrinologie-Gynécologie Pédiatriques, Service de Pédiatrie 1, Hôpital Arnaud de Villeneuve, Centre Hospitalier Universitaire Montpellier, 34295 Montpellier, France. E-mail: c-sultan{at}chu-montpellier.fr.
| Abstract |
|---|
|
|
|---|
s in ovarian granulosa cell tumors (OGCTs). Although the precise role of this mutation in the transformation of ovarian cells into malignant cells remains debatable, it has been demonstrated in other tissues that the rate of cell proliferation and invasiveness can be influenced by the gsp oncogene.
Objective: The objective of this study was to determine whether activating mutations of G
s or G
i are present in juvenile OGCTs and, if so, whether these mutations are significant prognostic factors.
Design and Setting: This was a multicentric nationwide study.
Patients and Methods: Thirty children with juvenile OGCT were included from the malignant germinal tumor protocol of the French Society for Childhood Cancer. Genetic studies of the tumoral DNA used nested PCR, laser microdissection, and direct sequencing.
Results: G
s-activating mutations in hot spot position 201 were found in nine patients (30%). Laser microdissection confirmed that mutations R201C and R201H were exclusively localized in the tumoral granulosa cells and were absent in the ovarian stroma. Patients with a hyperactivated G
s exhibited a significantly more advanced tumor (P < 0.05) because seven of them (77.7%) were staged as Ic or had had a recurrence. G
i did not exhibit any mutation.
Conclusions: Activating mutations of G
s are present in 30% of juvenile OGCTs. The gsp oncogene, which is known to be implicated in cell proliferation and tumoral invasiveness, can be considered as a new prognostic factor of these tumors.
| Introduction |
|---|
|
|
|---|
The pathophysiology of OGCT is still unclear. The usual risk factors for epithelial cancers, such as early menarche and association with the familial breast cancer genes BRCA1 and BRCA2, do not seem to influence the occurrence of OGCT. On the other hand, excessive treatment by gonadotropins in the context of infertility has been discussed as a risk factor of adult OGCT (5). Trisomy of chromosomes 12 and 14 and monosomy 22 have been described as nonrandom and nonobligatory events in these tumors (6, 7). In the juvenile form, which mainly occurs at 89 yr and between 13 and 17 yr (8), insight into the molecular mechanisms and reliable prognostic factors are lacking. Although their classification was originally morphological, OGCTs exhibit many of the features of normal granulosa cells: expression of the FSH receptor gene (9), FSH binding (10), and estrogen and inhibin synthesis (11, 12). The gene expression profile of OGCT is consistent with an activation of FSH signaling (13). Although FSH production is blunted by the inhibin production of the tumor (14), an autonomous activation of the post-FSH-receptor transduction signal has been hypothesized (15). FSH signaling involves the coupling of heterotrimeric G proteins to activate intracellular second messenger systems, mainly the cAMP-mediated protein kinase. Conflicting data regarding the presence of a gsp oncogene in OGCT have been reported (16, 17). Although the precise role of this mutation in the transformation of ovarian cells into malignant cells remains debatable, it has been clearly demonstrated in other tissues that the rate of cell proliferation and invasiveness can be influenced by constitutive activation of G
s (18).
The aim of this multicentric nationwide study was to determine whether activating mutations of G
s or inhibiting mutations of G
i are present in juvenile OGCTs and, if so, whether these mutations are significant prognostic factors.
| Patients and Methods |
|---|
|
|
|---|
Between 1990 and 2004, 30 patients with juvenile granulosa cell tumor were reported from the malignant germinal tumor protocol of the French Society for Childhood Cancer and from eight pediatric endocrinology centers. All 30 girls had histologically proven juvenile granulosa tumors, and 20 of the girls were prepubertal. The patients had been referred to pediatric centers either for treatment of a diagnosed abdominal mass or for investigation of precocious pseudopuberty. The clinical data were reviewed for all patients, and the tumors were staged retrospectively from the histological and surgical reports using the International Federation of Gynecology and Obstetrics classification (19). The conditions necessary to assume that resection had been complete were removal of the intact tumor in one piece and margins free of tumor cells on microscopic examination. Spontaneous preoperative rupture and malignant ascites, as well as accidental peritoneal spreading (rupture of the tumor capsule during surgery), were classified stage Ic.
The childrens pubertal stages were determined according to Tanner (20). Linear growth was monitored with reference to the Centre International de lEnfance curves. In cases of tumors revealed by precocious pseudopuberty with signs of estrogenic activity (breast development, growth acceleration, vaginal bleeding), plasma estradiol was measured. Ten patients with adult-type OGCT were also included. This work was approved by the ethics committee of the referring university hospital. Informed consent was obtained from each patients parents.
DNA analysis
Genomic DNA was isolated from paraffin-embedded tumors by standard procedures. Three 7-µm sections were cut from each block and mounted on glass slides. The middle section was stained with hematoxylin-eosin and reviewed by the pathologist to confirm diagnosis and to demarcate areas of tumor and nontumor in each case. The excess paraffin was removed, and the tissue was scraped into sterile 1.5-ml tubes. Isolation of genomic DNA from these tumor tissues was performed according to previously described procedures (21). PCR was used to amplify DNA fragments including codons 201 and 227 of G
s, based on the nested PCR method (22, 23). A modified primer was used to obtain a PCR product from normal DNA, which could then be digested by a restriction enzyme (EagI). In contrast, the PCR product obtained from mutated DNA was resistant to this enzyme. Successive PCR and enzyme digestion steps, however, resulted in a selective enrichment of the mutated allele. PCR and digestion were carried out in three successive steps. The primers for amplification are summarized in Table 1
. To ensure the absence of contamination due to the nested PCR, normal granulosa cells provided by Prof. Y. Menezo (Lyon, France) and DNA-free samples were run simultaneously.
|
Exons 5 and 6 of the G
i2 gene were amplified using a pair of oligonucleotides as indicated in Table 2
. The cycling protocol was identical for both exons and consisted of an initial denaturation step at 93 C (2 min), followed by 39 cycles of annealing at 93 C (30 sec), extension at 58 C (30 sec), and denaturation at 72 C (30 sec), with a final extension step of 6 min at 72 C.
|
Direct sequencing
DNA sequencing was performed on all PCR products with the antisense primer. The ABI310 PCR system (Applied Biosystems, Foster City, CA) was used. Sequencing reactions were repeated twice with at least two different PCR products.
Laser microdissection
To confirm the presence of a mutation in the granulosa cells, laser microdissection of slides was performed using a PixCell II LCM System (Arcturus Engineering, Inc., Mountain View, CA). As previously described (24), the film with attached specimen was mounted in reverse (film side up) on a coverslip. The specimens were then subjected to a laser beam, and the desired cells were catapulted by laser pressure onto mineral oil-coated caps of a PCR tube. The transfer of the microdissected cells onto the cap was confirmed under microscope. The laser beam diameter was 15 µm, and 25003000 impulsions were performed (30 milliwatts, 15 msec). The genetics of the tumoral cells and the adjacent ovarian stroma were thus studied separately.
Statistical analysis
Statistical analysis was performed with the SPSS 11.1 program.
2 tests for qualitative data and Students t tests for quantitative data were used. Significance was accepted at P < 0.05.
| Results |
|---|
|
|
|---|
s in hot spot position 201 in nine of the 30 patients DNAs (30%). In eight cases, a mutation from C to T in the first position at codon 201 of exon 8 of the G
s gene resulted in the encoded amino acid changing from arginine (CGT) to cysteine (TGT). In one case, a transition from G to A in the second position of the same codon resulted in a mutation from arginine (CGT) to histidine (CAT). No mutation in position 227 was found. Moreover, laser microdissection confirmed that mutations R201C and R201H of G
s were exclusively localized in the tumoral granulosa cells and were absent in the ovarian stroma (Fig. 1
i did not exhibit any mutation. Sequences of normal granulosa cell cultures and from adult OGCTs did not exhibit any mutation.
|
s mutation status. When compared with the patients with normal G
s, the clinical symptoms and age of occurrence were not significantly different. We did not observe more severe hyperestrogenic symptoms in the cases of G
s mutation. Nevertheless, the oncologic stages were significantly different according to the gsp oncogene status. Patients with a hyperactivated G
s exhibited a significantly more advanced tumor (P < 0.05) (Table 3
s mutation had a tumor strictly limited to the ovary with neither extracapsular extension nor recurrence (76.2%).
|
| Discussion |
|---|
|
|
|---|
Because it represents the first step in FSH signal transduction, the FSH receptor is suspected to play a role in the oncogenesis. This has not been confirmed, however, because no activating mutation of the FSH receptor gene has ever been reported (9, 17, 26). The second step in FSH signal transduction includes the G proteins. Heterotrimeric G proteins couple the cell surface receptors for hormones and thus activate intracellular second messenger systems. Inhibition of the intrinsic guanine triphosphatase activity of the G protein
-subunit results in retention of the active, GTP-bound state and in ligand-independent activation of the signaling pathway. Such activating mutations of the G
s gene or inhibiting mutations of the G
i gene can theoretically occur. For the gip oncogene, mutations have been reported in codon 179 of the G
i subunit gene in adrenocortical and ovarian sex cord stromal tumors (27) and in codon 205 in nonfunctioning pituitary tumors (28). The present study indicates that somatic mutations of G
i are not implicated in granulosa cell tumors. Our data are in accordance with those of Ligtenberg et al. (17) and Shen et al. (29).
The gsp mutation (R201C) was reported in sex cord stromal tumors (16). However, the implication of G
s-activating mutations as a primary mechanism of oncogenic transformation of granulosa cells has not been confirmed (17, 30). The studies that drew the conclusion that FSH receptor-coupled G proteins do not play a major role mainly explored adult tumors and did not use a highly sensitive nested PCR method. In the present study, we chose to study the juvenile type of OGCT. In this particular and rare form of granulosa cell tumor, endocrine signs like precocious pseudopuberty (precocious thelarche and/or menarche) are the predominant clinical symptoms in children. Juvenile OGCTs are strong estrogen-secreting tumors. In contrast, in the adult form, OGCTs are most often diagnosed because of an abdominal mass, and the endocrine symptoms are minor and less frequent. It thus appeared logical to look for a hyperactivation of the FSH signaling pathway in children. We report here that G
s-activating mutations were identified in 30% of the juvenile OGCT cases. This may even be an underestimation because only tissue from a selected block was studied. As expected, we found no G
s mutations in the adult type of OGCT in this study (17).
Although this percentage is significant, we cannot conclude that the gsp oncogene is a predominant mechanism of cell transformation. Further studies are needed to determine whether a gsp mutation alone is sufficient to induce human juvenile OGCT. Whatever its potential role in the origin of the tumor, however, the gsp oncogene is a prognostic factor. In our series, more than half of the patients with an extraovarian tumor exhibited a gsp-positive status. On the other hand, all but two tumors staged as Ia were gsp negative. These findings are consistent with the known genomic effects of G
s protein. The oncogenic activity of the gsp oncogene is dependent on the cell type. G
s stimulates proliferation only in cells that are positively responsive to cAMP-protein kinase A for their cell growth. Granulosa cells, which are highly specialized endocrine cells, use this signaling pathway. The rate of cell proliferation in an OGCT can thus be influenced by the presence of a gsp oncogene.
Besides this outcome on cell survival and proliferation, other effects of an activated G
s protein could explain why the gsp oncogene appears to be a new prognostic factor for OGCT. Invasion capacity and cell motility are increased in endocrine-dependent tumors transfected with an activated G
s (18, 31). The control of cell invasion by the gsp oncogene is not only protein kinase A-dependent, but it involves other cAMP-dependent mechanisms, e.g. the guanine-nucleotide exchange factor Epacs/cAMP-GEFs pathway, leading to the activation of Rap1, a small guanosine triphosphatase involved in activating the integrins that play a key role in cellular invasion (32, 33, 34, 35). Hyperactivated G
s increases tumorigenicity and the metastasis-forming ability of tumoral cells. This may explain why gsp-positive OGCTs more frequently exhibit extracapsular extension.
From a clinical point of view, this molecular study addresses three key issues. First, prognostication in juvenile OGCT has proved very difficult. No association between tumor size and histology (nuclear atypia, mitotic count, or cytogenetic studies) and outcome has been undisputedly demonstrated (36, 37). The outcome is good in most cases, but the risk of a fatal outcome is not negligible (38). The survival of patients with stage Ia tumors is 8398%, the overall survival in the whole group being 7892%. However, those tumors that have spread within the abdomen or recurred after initial therapy have a much poorer prognosis (8). We show here that the gsp oncogene appears to be a reliable prognostic factor of juvenile OGCT because it is significantly correlated with stage Ic, which, as in other types of ovarian tumor, heightens the risk of local relapse. Although the measurement of serum Mullerian-inhibiting substance and inhibin levels has proved to be a powerful and clinically useful predictor of impending relapse (39, 40), it is still not possible to identify the patients who will relapse in the future. The frequency and length of follow-up especially have not been established for children with OGCT. Yet, long-term recurrence has been described in adulthood (41, 42), and even patients with stage I disease appear to be at higher risk in some series (37). Because it has been correlated with cell motility and tumor rupture, the gsp oncogene may be a useful tool to rationalize the follow-up of children with OGCT, with its detection indicating the need for lifelong surveillance.
Second, the gsp oncogene may be implicated not only in granulosa cell tumors but more generally in all tumors arising from FSH-sensitive cells. As compared with the ovary, in which granulosa cell activity is FSH-dependent, FSH in the testis regulates Sertoli cell activity. It is conceivable that a constitutionally hyperactivated G
s could also induce hyperplasia of Sertoli cells. Indeed, we have reported that Sertoli cell adenoma can be linked to an activating mutation of the G
s gene (43).
The third issue addressed in this work is the potential relationship between OGCT and McCune-Albright syndrome (MAS). Although the initial definition of MAS was clinical, the finding that the G
s mutation is a major pathophysiological mechanism has had an impact, and this mutation is gradually becoming a new diagnostic element. Yet, the G
s mutation has been reported in several pathologies, without signs of associated MAS, including thyroid or somatotropic adenomas. Molecular analysis has thus posed the question of a relationship between OGCT and MAS. In this study, no patient with OGCT presented with café au lait spots or bone dysplasia. Conversely, none of our 113 patients with MAS has developed a granulosa tumor to date (23). The question thus remains open.
Overall, the data from this nationwide study indicate that the gsp oncogene may be a significant prognostic factor in juvenile OGCT. However, whether gsp alone has the capacity to induce the malignant transformation of granulosa cells and the specific events associated with G
s-mediated tumor progression remains to be investigated.
| Acknowledgments |
|---|
| Footnotes |
|---|
Present address for S.L.: Laboratoire de Biochimie, Centre Hospitalier Universitaire Caremeau, Place du Professor Robert Debré, 30029 Nimes, France.
Disclosure statement: All authors have nothing to declare.
First Published Online February 28, 2006
Abbreviations: MAS, McCune-Albright syndrome; OGCT, ovarian granulosa cell tumor.
Received December 13, 2005.
Accepted February 22, 2006.
| References |
|---|
|
|
|---|
QL)-mediated signaling increases invasiveness and tumorigenicity of PC-3M prostate cancer cells. Oncogene 18:33763382[CrossRef][Medline]
s from patients with fibrous dysplasia of bone. Bone 21:201206[Medline]
mutations: analysis of 113 patients with signs of McCune-Albright syndrome: a European Collaborative Study. J Clin Endocrinol Metab 89:21072113
-gene in nonfunctioning pituitary tumors. J Clin Endocrinol Metab 77:765769[Abstract]
activating mutations. J Pediatr Endocrinol Metab 15 (Suppl 3):891895
(olf) subunit promotes cellular invasion, survival, and neuroendocrine differentiation in digestive and urogenital epithelial cells. Oncogene 21:40204031[CrossRef][Medline]
6ß4 integrin-dependent activation of phosphoinositide 3-OH kinase and promotion of invasion. Mol Cell Biol 21:50825093
gene mutation: an unusual expression of McCune-Albright syndrome in a prepubertal boy. J Clin Endocrinol Metab 86:17781781This article has been cited by other articles:
![]() |
C. Aydin, N. Aytan, M. J. Mahon, H. A. W. Tawfeek, N. W. Kowall, A. Dedeoglu, and M. Bastepe Extralarge XL{alpha}s (XXL{alpha}s), a Variant of Stimulatory G Protein {alpha}-Subunit (Gs{alpha}), Is a Distinct, Membrane-Anchored GNAS Product that Can Mimic Gs{alpha} Endocrinology, August 1, 2009; 150(8): 3567 - 3575. [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 |