8/14/2019 274.full.pdf

1/6

COMMENTS

No Evidence of a Role for Mutations or Polymorphisms

of the Follicle-Stimulating Hormone Receptor in OvarianGranulosa Cell Tumors*

PETER J. FULLER, KAREN VERITY, YAN SHEN, PAMELA MAMERS,TOM JOBLING, AND HENRY G. BURGER

Prince Henrys Institute of Medical Research, Monash University, and Department of Obstetrics andGynecology, Monash Medical Center, Clayton, Victoria 3168, Australia

ABSTRACTThe molecular pathogenesis of granulosa cell tumors of the ovary

is not understood, although recent studies have shown that immu-

noreactive inhibin secretion by these tumors may be used as a tumormarker. Granulosacell tumors exhibit many features of normalgran-ulosa cells, including a response to FSH and inhibin secretion. FSHlevels are suppressedin patients withinhibin-secreting granulosacelltumors, suggesting FSH-independent growth of these tumors. Acti-

vating mutations of the FSH receptor might, therefore, be involved intumorigenesis. We sought to identify mutations in the FSH receptorgenes of these tumors using PCR to amplify the exon encoding thetransmembraneand cytoplasmic domains fromthe tumorDNA. Anal-

ysis of the amplicons for single strand conformational polymorphismand direct sequencing confirmed a previouslyreported polymorphismin the C-terminal region of the receptor, but did not identify tumor

specific missense mutations and/or polymorphisms. In addition, ribonucleic acid from 3 granulosa cell tumors was used to confirmexpression of the FSH receptor; expression was unexpectedly alsobserved in several ovarian mucinous cystadenocarcinomas used acontrols. In conclusion, our failure to identify activating mutations othe FSHreceptorin 15 granulosa cell tumorsargues against a role fothe FSH receptor in tumorigenesis and suggests that some subsequent component of this signal transduction pathway may be acti

vated. (J Clin Endocrinol Metab83: 274279, 1998)

OVARIAN SEX cord stromal tumors represent approx-imately 10% of all ovarian tumors. They are generallyclassified as granulosa cell tumors, thecomas, or luteomas;the former is the most common (1). Although the classifica-tion is primarily morphological, rather than functional or

biological, these tumors exhibit many of the features of gran-ulosa cells including FSH binding, a response to FSH, and thesecretion of the peptide hormones inhibin (2 4) and Mulle-rian inhibiting substance (5). Inhibin production by thesetumors appears to suppress plasma FSH levels (3, 4), sug-gesting the FSH-independent growth of these tumors. CyclinD2 is required for granulosa cell proliferation, and its ex-pression is stimulated by FSH. High levels of cyclin D2 mes-senger ribonucleic acid (RNA) have recently been reportedin granulosa cell tumors of the ovary (6), further arguing forconstitutive activation of an FSH-dependent mitogenic sig-naling pathway in these tumors.

Heterotrimeric G proteins couple cell surface receptors for

hormones, including FSH, to intracellular second messengesystems. Activating mutations of both cell surface receptorand G proteins have been reported in various tumors (7, 8)Mutations of the G protein -subunit gene causing inhibition

of the intrinsic guanosine triphosphatase activity anthereby yielding an activating oncogene, gsp, were first described in GH-secreting pituitary adenomas (9). The gsponcogene is found in 40% of somatotroph adenomas as well aless frequently in nonfunctioning pituitary adenomas, corticotroph adenomas, thyroid adenomas, and the McCuneAlbright syndrome (7, 8, 10). In a retrospective study oarchival material from many tumor types, Lyons et al. (10reported the presence of analogous mutations in the Gi-subunit gene (gip2) in 3 of 10 sex cord stromal tumors and 3of 11 adrenocortical tumors. In a recentstudy we were unablto find activating mutations at codon 179 of the Gi-2gene inovarian granulosa cell tumors (11), and Reincke et al. (12were similarly unsuccessful when they examined 18 adre

nocorticoid tumors.The FSH receptor is a member of the seven-transmem

brane domain or G protein-coupled receptor (GPCR) superfamily of cell surface receptors. Activating mutations of thesreceptors have been described in a range of conditions (8)Activating mutations of the TSH receptor have been implicated in the pathogenesis of malignancy in hyperfunctioningthyroid adenomas (13, 14) and well differentiated thyroidcarcinoma (15, 16). Recently, a virally encoded constitutivelyactive GPCR of the chemokine subfamily has been impli

Received April 22, 1997. Revision received August 22, 1997. AcceptedSeptember 28, 1997.

Address all correspondence and requests for reprints to: Dr. Peter J.Fuller, Prince Henrys Institute of Medical Research, P.O. Box 5152,Clayton, Victoria 3168, Australia. E-mail: peter.fuller@med.monash.edu.au.

* This work was supported by a project grant from the Anti-CancerCouncil of Victoria.

Recipient of a Principal Research Fellowship from the NationalHealth and Medical Research Council of Australia.

Supported by a Monash University Graduate Research Scholarship.

0021-972X/98/$03.00/0 Vol. 83, No.Journal of Clinical Endocrinology and Metabolism Printed in U.S.ACopyright 1998 by The Endocrine Society

274

8/14/2019 274.full.pdf

2/6

cated in the pathogenesis of both Kaposis sarcoma and a rare-cell lymphoma (17).

The possibility that activation of the FSH-dependent sig-naling pathways may play a role in the pathogenesis ofgranulosa cell tumors of the ovary, in a manner analogous tothe role of the TSH receptor in the pathogenesis of thyroidtumors, encouraged us to analyze a series of granulosa cell

tumors for activating mutations of the FSH receptor.

Materials and Methods

Isolation of DNA from tissue specimens

Frozen granulosa cell tumor tissue (n 3) or paraffin-embeddedtissue sections (n 12) were studied (11) as part of a much largerMelbourne-wide investigation of seruminhibin levels in ovarian tumors(3, 4). DNA extraction from fresh-frozen tissue was performed as pre-viously described (11). Briefly, approximately 0.5 g tissue was homog-enized in 5 mL extraction buffer (10 mmol/L Tris, 0.1 mol/L ethyl-enediamine tetraacetate, 20 g/mL ribonuclease, and 0.5% SDS),digested with 100 g/mL proteinase K at 37 C for 4 h, extracted twicewith an equal volume of phenol and once with chloroform, precipitatedwith 0.2 vol 10 mol/L ammonium acetate and 2 vol 100% ethanol, andwashed with 70% ethanol. DNA pellets were diluted in distilled waterand stored at 4 C. For isolation of DNA from paraffin-embedded tissue,two or three 10-m sections were cut for each sample, and excessparaffin was removed using a disposable scalpel. Tissue sections werescraped into a 1.5-mL microcentrifuge tube. Paraffin was removed fromthe samples by two 20-min extractions with 1 mL xylene. Xylene wasremoved by twowashes with 1 mL100% ethanol andone wash with 70%ethanol. Tissue sections were vacuum-dried and digested overnight at50 C with 400 ng/L proteinase K in 100200 L 50 mmol/L Tris, pH8.3, as previously described (11). Samples were then boiled for 8 min toinactivate proteinase K, cooled on ice, and spun briefly to pellet outdebris. The supernatant was transferred into a fresh tube for storageat 20 C.

Amplification of genomic DNA

PCR was used to amplify five overlapping amplicons covering exon10 of the FSH receptor gene. The primer sequences are shown in Table1, and the positions of the amplicons are shown schematically in Fig. 1.

The amplicons ranged from 226275 bp in length. The PCR reactionmixture consisted of 100 ng DNA extracted from frozen tissue or 5 Ltissue section lysate, 30 pmol of each primer, 100 mol/L deoxy-NTPs,1.8 mmol/L MgCl

2, 2.5 U Taq polymerase (Boehringer Mannheim,

Mannheim, Germany), and 1 PCR buffer [10 PCR buffer is 100mmol/L Tris (pH 8.5), 500 mmol/L KCl, and 1% Triton X-100] in a totalvolume of 100L. PCR was initiated at 94 C for 5 min, followed by 35cyclesof 94C for 30s, 55C for 40s, and 72C for 60s and a final extensionat 72 C for 8 min. In all PCR experiments, reactions containing no DNAwere included as negative controls.

Direct sequencing

To confirm the above findings, the PCR products were sequenceddirectly using a Sequenase 2.0 kit (U.S. Biochemical Corp., ClevelandOH), [35S]deoxy-ATP, 15 pmol primer, and 500 ng of the PCR productfor each sample.

Single strand conformational polymorphisms (SSCP)

The SSCP analyses were performed using minor modifications of thmethod described originally by Oritaet al.(18). PCR reactions to be usefor the SSCP analysis were performed in the presence of 0.5 C[-33P]deoxy-ATP [2000 Ci/mmol; DuPont (Australia), North SydneyAustralia] in a total volumeof 50L.The products were runon standardenaturing (7 mol/L urea) or nondenaturing 6% polyacrylamide sequencing gels at 28, 20, and 10 C with or without 10% glycerol in the gelConstant temperature was maintained using a Strata Therm Temperature Controller (Stratagene, La Jolla, CA). The conditions of the PCR ardescribed above.

Reverse transcriptase-PCR

RNA was extracted from the three frozen granulosa cell tumors (11)seven mucinous cystadenocarcinomas of the ovary, and normal humancolon using the guanidine thiocyanate/cesium chloride method. Onmicrogram of total RNA was reverse transcribed for 90 min at 42 C ia total volume of 20 L using AMV reverse transcriptase (BoehringeMannheim) with the FSH receptor antisense primer at position 176

(Table 1). Two microliters of this reaction were added to a PCR reactionas described above using the sense and antisense primers for amplicon3 (Table 1). Ten microliters of each product were analyzed on ethidium

bromide-stained 1% agarose gels. Reactions containing either no inpuRNA or no reverse transcriptase were included as negative controls.

Results

The clinical details for 13 of the tumors analyzed havepreviously been reported (11). An additional 2 samples werobtained from paraffin blocks. Preoperative inhibin levelwere not available for these 2 patients. In addition, DNA wa

TABLE 1. Primers used for PCR of exon 10 of the FSH receptor

1 Sense, 5-TCA TGG GGT ACA ACA TCC TCA-3 (1150)Antisense, 5-AAA AGC CAG CAG CAT CAC AGC-3 (1397)

2 Sense, 5-ATC ACA ACT ATG CCA TTG ACT-3 (1360)Antisense, 5-GTA GCT GCT GAT GCC AAA GAT-3 (1587)

3 Sense, 5-CTT TTG CAG CTG CCC TCT TTC-3 (1564)Antisense, 5-TTG GCG ATC CTG GTG TCA CTA-3 (1769)

4 Sense, 5-GCG GAA CCC CAA CAT CGT GTC-3 (1745)Antisense, 5-TGA AGA AAT CTC TGC GAA AGT-3 (1969)

5 Sense, 5-ATT CTG CTG GTT CTG TTT CAC-3 (1899)Antisense, 5-CAC ATT GTG TTT TAG TTT TGG-3 (2153)

The primer positions, numbered according to the published se-quence (36, 37), are shown inparenthesesat the end of each line. Theamplicons are numbered 15.

FIG. 1. Schematic representation of the structure of the FSH receptor. The amplicons (no. 15) that span the region of exon 10 encodingthe transmembrane domains (TMD) and the cytoplasmic tail arindicated. The positions of the inactivating (Ala189Val) (19) and activating (Asp567 Gly) (20) mutations are indicated, as is the positionof the codon 680 polymorphism. Modified from Aittomakiet al. (19

COMMENTS 27

8/14/2019 274.full.pdf

3/6

obtained from normal human tissue for use as a control. AllPCR amplicons obtained were of the expected size, and theiridentities were verified using direct DNA sequencing. Allamplicons were subjected to SSCP analysis under the con-ditions described above. Several putative differences in the

banding patterns were detected, but in none was this con-firmed by direct sequencing. The only consistent difference

in the pattern observed between samples for SSCP analysiswas for the fifth amplicon, where 2 different alleles wereclearly seen (Fig. 2). Direct sequencing revealed this to be apreviously described (19) polymorphism where a G to Atransition results in either a serine or an asparagine residue,respectively, at codon 680(Fig. 3).We further examined DNAfrom 7 mucinous cystadenocarcinomas of the ovary and 22randomly selected normal DNA samples for this polymor-phism (Fig. 2). The distribution of the codon 680 alleles istabulated in Table 2. No clear association of the codon 680allele with the source of the DNA was observed. In theiroriginal description of this polymorphism, Aittomaki et al.(19) found that neither allele was linked to hereditary hy-pergonadotropic ovarian failure, the phenotype being ex-

amined. In addition, direct sequencing was performed onamplicon 4 from 10 of the granulosa cell tumors, given thatthis region encompasses a previously reported activatingmutation (20) of the FSH receptor and corresponds to a hotspot for activating mutations of the TSH receptor (1316)and has very recently been reported to contain an inactivat-

ing mutation in ovarian sex cord tumors (21). As with theSSCP analysis of this amplicon, direct sequencing failed toidentify any nucleotide changes.

In addition, we sought to establish that the FSH receptowas indeed expressed in granulosa cell tumors. RNA waable to be extracted only from the three frozen granulosa celtumors and the seven mucinous cystadenocarcinomas whosDNA was analyzed above. Reverse transcriptase-PCR of thRNA from the granulosa cell tumors clearly confirmed expression of the FSH receptor gene (Fig. 4). Curiously, lowelevels of expression than in the granulosa cell tumors were

noted in most of the mucinous cystadenocarcinomas.

Discussion

Ovarian granulosa cell tumors are relatively rare, but exhibit an interesting phenotype with both unrestricted growthand ovarian peptide hormone hypersecretion. Granuloscells are normally subject to hormonal regulation, particularly by FSH. FSH stimulates granulosa cells to secrete inhibin and Mullerian inhibiting substance, both of which armembers of the transforming growth factor- family o

FIG. 2. SSCP analysis of amplicon 5 of the FSH receptor. Analysis ofDNA from 15 granulosa cell tumors and 7 mucinous cystadenocar-cinomas of the ovary flanked by 2 normal DNA samples. The 33P-labeled PCR products have been run in a denaturing gel containing10% glycerol at 28 C. The normal samples are both heterozygous forthe polymorphism. The positions of the specific conformers for eachallele are shown on the left(upper bands, the asparagine-containingallele) and right (lower bands, the serine-containing allele) of thefigure.

FIG. 3. Direct DNA sequencing of representative tumor samples ether heterozygous or homozygous for the codon 680 polymorphism

TABLE 2. Distribution of codon 680 alleles

GCT MC N

Asn/Asn 4 5Ser/Asn 8 6 9Ser/Ser 3 1 8

GCT, Granulosa cell tumors; MC, mucinous cystadenocarcinomasN, normal; Ser, serine; Asn, asparagine.

276 COMMENTS JCE & M 199Vol 83 No

8/14/2019 274.full.pdf

4/6

dimeric growth factor molecules, as well as stimulates thesecretion of estradiol. As part of the mitogenic response, FSHstimulates increased cyclin D2 expression (6).

Recent clinical reports suggest that hyperstimulation ofthis signal transduction pathway during gonadotropin ther-apy for the treatment of infertility increases the incidence ofgranulosa cell tumors (22, 23); however, the validity of thesestudies is uncertain (24, 25). Granulosa cell tumors also arisein transgenic mice, in which the inhibin -subunit gene has

been deleted. The pathogenesis of these tumors is unclear;however, the mice have markedly elevated levels of FSH andalso activin (26). Up-regulation of the signal transductionpathways initiated by FSH might be postulated to play a rolein the pathogenesis of these tumors. Lyons et al. (10) reportedthe presence of thegip2oncogene in 3 of 10 ovarian sex cordstromal tumors. The relationship of Gi-2to FSH-stimulatedsignal transduction is not clear; indeed, we were unable toconfirm the presence of the gip2oncogene in the granulosacell tumors used in this current study (11).

In addition to mutations of G proteins, mutations of GPCRplay a role in a range of inherited and acquired conditions (78, 17). Inactivating mutations of these receptors have beenextensively reported, including an alanine to valine substitution in the extracellular region of the FSH receptor firsreported by Aittomaki et al. (19) in hereditary hypergonadotropic ovarian failure.

Activating mutations of GPCR were first identified in artificial constructs, but naturally occurring mutations havsubsequently been identified in a range of receptors, including those for chemokines (17), calcium (8), PTH/PTH-relatepeptide (27), and gonadotropins. Activating mutations of thTSH receptor have been identified in both inherited andsporadic conditions. Inherited mutations of the TSH receptocause familial nonautoimmune hyperthyroidism (28)whereas somatic mutations occur in both adenomas (13, 14and well differentiated carcinomas (15, 16) of the thyroidActivating mutations of the LH receptor have been reportedin gonadotropin-independent male-limited precocious pu

berty (29, 30). A heterozygous A to G change at nucleotid1700 in the human FSH receptor, causing an aspartate to

glycine transition in codon 507, has been recently reported ina subject with pituitary-independent spermatogenesis (20)The substituted residue lies in the third intracytoplasmiloop. The majority of activating mutations of GPCR reportedlie in this loop, the sixth transmembrane domain, or, lescommonly, the second transmembrane domain.

In the present study we sought to address the hypothesithat activation of some component of the FSH-stimulatedsignal transduction pathway in granulosa cells may contribute to the pathogenesis of granulosa cell tumors of the ovaryWe examined the FSH receptor not only because it representthe most proximal component of the pathway, but also because of the reports of activating mutations in the threegonadotropin receptors. Our inability to identify any muta

tions in exon 10 of the FSH receptor in the DNA from 15granulosa cell tumors would seem to argue strongly againsa major role for FSH receptor mutations in the pathogenesiof these tumors. It does not exclude the possibility that mutations may occur in a subset of cells or a minority or tumorsand it remains possible, although unlikely, that our exhaustive search using SSCP analysis under a range of conditionhas missed a specific mutation. It is reassuring that a previously reported polymorphism was clearly identified. Limited direct sequencing was also performed; although thishould be 100% sensitive, it may still miss mutations of oneallele where a degree of contamination of the tumor by normal tissue, such as stromal elements, dilutes the abnormaallele.

Subsequent to the submission of this study, Kotlar et al(21) reported a similar study in which they examined smaller region of exon 10 of the FSH receptor gene in 1ovarian sex cord tumors. They amplified the third intracytoplasmic loop together with parts of the fifth and sixthtransmembrane domains. Using direct sequencing of PCRproducts from archival material amplified by nested PCRthey identified a missense mutation at nucleotide 1777 thaconverts codon 591 from phenylaline to serine (F5915) in 69%of the sex cord tumors (21). Functional studies suggest thathis is an inactivating mutation (21), which runs counter to

FIG. 4. Ethidium bromide-stained gels of reverse transcriptase-PCRproducts using the primer pair for the FSH receptor amplicon 3. A,The results from three granulosa cell tumors (GCT). Mol wt markers

(M) and the amplicon from PCR of DNA (D) are also shown. B, Theresults from seven ovarian mucinous cystadenocarcinomas with noreverse transcriptase (T) and no RNA (R) negative controls. ColonicRNA (C) and jejunal RNA (J) negative controls are also shown. TheFSH receptor band is indicated by an arrow.

COMMENTS 27

8/14/2019 274.full.pdf

5/6

their and our initial hypothesis. We have been unable toidentify this mutation in our tumors from the SSCP analysis,

by direct sequencing of both strands of this amplicon from10 of our tumors, or by allele-specific hybridization with theprimers described by Kotlaret al.(21). It is difficult to explainthe discrepancy between our study and that of Kotlar et al.(21); the methodology is fundamentally similar, except, per-

haps, that we did not need to use nested PCR. The maindifference, therefore, lies in the tumor populations studied.Our tumor sample represents essentially consecutive gran-ulosa cell tumors identified in Melbourne as part of a largerstudy of inhibin levels in ovarian tumors (3, 4), whereasKotlar et al. (21) examined archival material accumulated byreferral of specific blocks of tumors over several decades forconfirmation of difficult histological diagnosis. They in-clude a large number of juvenile granulosa cell tumors. Themutation was also observed in a curious subgroup of ovariantumors, small cell carcinomas of the hypercalcemia type. Intheir discussion, Kotlaret al.(21) note that in two additional,presumably relatively unselected, cases, the mutation wasnot observed. An exciting conclusion that might be drawn

from this dichotomy is that distinct subgroups of sex cordtumors might be defined on the basis of the F591S mutationin the FSH receptor gene.

The polymorphism at codon 680 was first recognized assuch by Aittomaki et al. (19). Although the polymorphismwould clearly seem to have no correlation to ovarian cancerand to be fairly evenly distributed, it represents a noncon-servative substitution. This region of the receptor is not wellconserved compared to the TSH or LH receptor, but is rea-sonably well conserved across species (31). Asparagine ispresentat this position in seven other species of FSH receptor(31). We know of no studies that examine the significance ofthis substitution. The cytoplasmic tail is known to be im-portant in G protein coupling (7, 8), and thus changes in this

region might alter the strength or specificity of such inter-actions. A serine at this position also creates a potentialphosphorylation site (32).

FSH binding and stimulation of adenylyl cyclase activityin granulosa cell tumors have previously been reported (33,34), and our demonstration of FSH receptor gene expressionconfirms this finding. The presence of FSH receptor mes-senger RNA, albeit at low levels, in mucinous cystadeno-carcinomas is of interest given that these tumors also secreteinhibin (3).

The lack of activating mutations of the FSH receptor ingranulosa cell tumors in this study, although in contrast tothe situation with the TSH receptor, parallels the findings ofstudies of adrenal tumors in which activating mutations ofthe ACTH receptor were not identified (35). In conclusion, itwould seem that theFSH receptor, like thegip2 oncogene (11)or thegsp oncogene (10), is unlikely to play a major role inthe pathogenesis of granulosa cell tumors of the ovary, al-thoughan inactivatingmutationmay play a role in thepatho-genesis of a subset of sex cord tumors (21). The pathogenesisof ovarian sex cord stromal tumors, therefore, remainslargely unexplained, and perhaps because of their relativerarity, few studies have specifically examined the moleculargenetics of these tumors. The evidence, however, that up-regulation of this signaling pathway occurs remains com-

pelling, and future studies will need to examine moleculedistal to the FSH receptor but proximal to cyclin Dup-regulation.

Acknowledgments

We thank Sue Panckridge and Claudette Thiedeman for preparationof the manuscript. and Simon Chu for performing the allele-specifi

hybridization. We also acknowledge the contribution of our pathologiscolleagues,Drs. Beatrice Susil, Virginia Bilson, Andrew Ostor, and JameScurry, for making available the tissue blocks used.

References

1. Russell P, Bannatyne P. 1989 Surgical pathology of the ovaries. EdinburghChurchill Livingstone.

2. LappohnRE, BurgerHG, Bouma J, BangahM, Krams M,de BruijnHW. 198Inhibin as a marker for granulosa cell tumours. N Engl J Med. 321:790793

3. Healy DL, Burger HG, Mamers P, et al. 1993 Inhibin: a serum marker fomucinous ovarian cancers. N Engl J Med. 329:15391542.

4. Jobling T, Mamers P, Healy DL, MacLachlan V, Burger HG, Day AJ.1994 Aprospective study of inhibin in granulosa cell tumors of the ovary. GynecoOncol. 55:285289.

5. GustafsonML, LeeMM, ScullyRE, et al. 1992 Mullerian inhibiting substancas a marker for ovarian sex-cord tumor. N Engl J Med. 326:466471.

6. Sicinski P, Donaher JL, Geng Y, et al. 1996 Cyclin D2 is an FSH-responsivgene involved in gonadal cell proliferation and oncogenesis. Natur

384:470474.7. Dhanasekaran N, Heasley LE, Johnson GL. 1995 G protein-coupled recepto

systems involved in cell growth and oncogenesis. Endocr Rev. 16:259270.8. Spiegel AM.1996 Genetic basis of endocrine disease: mutations in G protein

andG protein-coupled receptorsin endocrinedisease. J ClinEndocrinol Metab81:24342442.

9. Landis CA, Masters SB, Spada A, Pace AM, Bourne HR, Vallar L. 198GTPase inhibiting mutations activate the -chain of Gs and stimulate adenylycyclase in human pituitary tumours. Nature. 340:692696.

10. Lyons J, Landis CA, Harsh G, et al. 1990 Two G protein oncogenes in humaendocrine tumors. Science. 249:655659.

11. Shen Y, Mamers P, Jobling T, Burger HG, Fuller PJ. 1996 Absence of thpreviously reported G-protein oncogene (gip2) in ovarian granulosa cell tumors. J Clin Endocrinol Metab. 81:41594161.

12. Reincke M, Karl M, Travis W, Chrousos GP. 1993 No evidence for oncogenimutations in guanine nucleotide-binding proteins of human adrenocorticaneoplasms. J Clin Endocrinol Metab.77:14191422.

13. Parma J, Duprez L, Van Sande J, et al. 1993 Somatic mutations in the thyrotropin receptor gene cause hyperfunctioning thyroid adenomas. Nature

365:649651.14. Russo D, Arturi F, Wicker R, et al. 1995 Genetic alterations in thyroid hyperfunctioning adenomas. J Clin Endocrinol Metab. 80:13471351.

15. Russo D, Arturi F, Schlumberger M, et al.1995 Activating mutations of thTSH receptor in differentiated thyroid carcinomas. Oncogene. 11:19071911

16. Spambalg D, Sharifi N, Elisei R, Gross JL, Medeiros-Neto G, Fagin JA. 199Structural studies of the thyrotropin receptor and Gs in human thyroicancers: low prevalence of mutations predicts infrequent involvement in malignant transformation. J Clin Endocrinol Metab. 81:38983901.

17. Murphy PM. 1997 Pirated genes in Kaposis sarcoma. Nature. 385:29629918. Orita M, Suzuki Y, Sekiya T, Hayashi K. 1989 Rapid and sensitive detectio

of point mutations and DNA polymorphisms using the polymerase chaireaction. Genomics. 5:874879.

19. Aittomaki K, Lucena JLD, Pakarinen P, et al. 1995 Mutation in the folliclestimulating hormone receptor gene causes hereditary hypergonadotropovarian failure. Cell 82:959968.

20. Gromoll J, SimoniM, NieschlagE. 1996An activating mutation of thefolliclestimulating hormone receptor autonomously sustains spermatogenesis in hypophysectomized man. J Clin Endocrinol Metab. 81:13671370.

21. Kotlar TJ, Young RH, Albanese C, Crowley Jr WF, Scully RE, Jameson JL

1997A mutationin thefollicle-stimulatinghormonereceptor occurs frequentlin human ovarian sex cord tumors. J Clin Endocrinol Metab. 82:10201026.

22. Willemsen W, Kruitwagen R, Bastiaans B, Hanselaar T, Rolland R. 199Ovarian stimulation and granulosa-cell tumour. Lancet. 341:986988.

23. Rossing MA,DalingJR, Weiss NS,Moore DE,Self SG. 1994 Ovarian tumorin a cohort of infertile women. N Engl J Med. 331:771776.

24. Venn A, Watson L, Lumley J, Giles G, King C, Healy D. 1995 Breast anovarian cancer incidence after infertility and in vitro fertilisation. Lance346:9951000.

25. Rossing MA. 1996Ovariancancer: a riskof infertility treatment? J Br Fertil So1:4650.

26. Matsuk MM, Finegold MJ, Su J-GJ, Hsueh AEW, Bradley A. 1992-Inhibiis a tumour-suppressor gene with gonadal specificity in mice. Natur360:313319.

278 COMMENTS JCE & M 199Vol 83 No

8/14/2019 274.full.pdf

6/6

27. Schipani E, Langman CB, Parfitt AM, et al. 1996 Constitutively activatedreceptors for parathyroid hormone and parathyroid hormone-related peptidein Jansens metaphyseal chondrodysplasia. N Engl J Med. 335:708714.

28. Tonacchera M,Van Sande J,Cetani F, et al. 1996 Functional characteristics ofthree new germline mutations of the thyrotropin receptor gene causing au-tosomal dominant toxic thyroid hyperplasia. J Clin Endocrinol Metab.81:547554.

29. Shenker A, Laue L, Kosugi S, Merendino Jr JJ, Minegishi T, Cutler Jr GB.1993 A constitutively activating mutation of the luteinizing hormone receptorin familial male precocious puberty. Nature. 365:652654.

30. Kremer H, Mariman E, Otten BJ, et al. 1993 Cosegregation of missense mu-tations of the luteinizing hormone receptor gene with familial male-limitedprecocious puberty. Hum Mol Genet. 2:17791783.

31. Richard F, Martinat N, Remy J-J, Salesse R, Combarnous Y. 1997 Cloning,sequencing andin vitrofunctional expression of recombinant donkey follicle-stimulating hormone receptor: a new insight into the binding specificity ofgonadotrophin receptors. J Mol Endocrinol. 18:193202.

32. Gromoll J,Simoni M,NordhoffV, BehreHM, DeGeyterC, NieschlagE. 1996

Functional and clinical consequences of mutations in the FSH receptor. MoCell Endocrinol. 125:177182.

33. Graves PE, Surwit EA, Davis JR, Stouffer RL. 1985 Adenylate cyclase ihuman ovarian cancers: sensitivity to gonadotropins and nonhormonal actvators. Am J Obstet Gynecol. 153:877882.

34. Stouffer RL, Grodin MS, Davis JR, Surwit EA. 1984 Investigation of bindinsites for follicle-stimulating hormone and chorionic gonadotropin in humaovarian cancers. J Clin Endocrinol Metab. 59:441446.

35. Ratronico AC, Reincke M, Mendonca BB, et al.1995 No evidence for onco

genic mutations in the adrenocorticotropin receptor gene in human adrenocortical neoplasms. J Clin Endocrinol Metab. 80:875877.36. MinegishT, Nakamura K,TakakuraY, IbukiY, Igarashi M. 1991Cloning an

sequencing of human FSH receptor cDNA. Biochem Biophys Res Commun175:11251130.

37. Gromoll J, Ried T, Holtgreve-Grez H, Nieschlag E, Gudermann T. 199Localization of thehuman FSHreceptorto chromosome 2 p21 using a genomprobe comprising exon 10. J Mol Endocrinol. 12:265271.

COMMENTS 27