| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Molecular and Population Genetics Laboratory (L.G.C.-C., N.A.A., P.J.P., A.M.J., E.B., N.W., I.P.M.T.) and Histopathology Unit and In Situ Hybridisation Service (R.P.), London Research Institute, Cancer Research UK, London WC2A 3PX, United Kingdom; Department of Pathology and Microbiology, Division of Histopathology, Bristol Royal Infirmary (M.P.), Bristol BS2 8HW, United Kingdom; Department of Pathology, University College Hospital (A.F.), London WC1E 6JJ, United Kingdom; Department of Pathology, Kings College Hospital (S.P.), London SE5 9RS, United Kingdom; Department of Histopathology, Nottingham City Hospital (I.E.), Nottingham NG5 1PB, United Kingdom; Department of Histopathology, Division of Investigative Science, Imperial College (M.A.E.-B.), London WI2 0NN, United Kingdom; and Department of Pathology, St. Bartholomews Hospital (D.M.B.), London EC1A 7BE, United Kingdom
Address all correspondence and requests for reprints to: Dr. Luis G. Carvajal-Carmona, Ph.D., Laboratory of Molecular and Population Genetics, London Research Institute, Cancer Research UK, London WC2A 3PX, United Kingdom. E-mail: luis.carvajal{at}cancer.org.uk.
| Abstract |
|---|
|
|
|---|
Objective: Our objective was to investigate the role of FH mutations in predisposition to LCTs.
Design: We tested for pathogenic effects of the N64T mutation and screened an additional 29 unselected adult LCTs for FH alterations. We also tested these LCTs for mutations in two genes, the LH/choriogonadotropin receptor (LHCGR) and the guanine nucleotide-binding protein
(GNAS) that had been implicated in LCT tumorigenesis.
Results: No mutations were found in GNAS, and one tumor had a LHCGR somatic substitution. In addition to the HLRCC case with the N64T germline FH mutation, we identified one other LCT with a previously unreported FH mutation (M411I). Both LCTs from these patients showed loss of the wild-type FH allele. Immunohistochemical and in situ hybridization analyses demonstrated activation of the hypoxia/angiogenesis pathway not only in the tumors belonging to the FH mutation carriers but also in several other mutation-negative LCTs.
Conclusions: Our study shows that some LCTs are caused by FH mutations and represents one of the first reports of germline mutations in any type of adult testicular tumor.
| Introduction |
|---|
|
|
|---|
The molecular basis of LCTs is poorly understood. Recent studies have shown that childhood LCTs can be caused by a somatic substitution (D578H) in a mutational hotspot of the LH/choriogonadotropin receptor (LHCGR) gene (6, 7, 8). LCTs with this mutation generally result in a phenotype characterized by isosexual pseudoprecocity. LCTs have also been observed infrequently in individuals of families with the Carney complex, an autosomal dominant syndrome caused by germline mutations in the PRKAR1A gene (9). Mutations in PRKAR1A have, however, been reported in Carney complex patients with large-cell calcifying Sertoli cell tumors but not in any patient with an LCT.
Most LCTs occur in adulthood, and their molecular etiology is less understood than that of childhood LCTs (10). To date, only two studies have investigated molecular changes in adult LCTs. Fragoso et al. (11) found one adult LCT with a somatic substitution in a mutational hotspot (R201C) of the guanine nucleotide-binding protein
gene (GNAS, also known as gsp); this gene was evaluated in LCTs because of previously reported mutations in hyperplastic gonadal tissue from McCune-Albright syndrome patients. In another study, Giacaglia et al. (12) screened the D578 LHCGR hotspot in two adult LCTs but failed to identify any mutations. Screening of these mutational sites has not been carried out in a large and representative sample of adult LCTs, and thus, their importance in LCT pathogenesis remains unknown. Although it is possible that some adult LCTs result from somatic mutations and from nongenetic predisposing factors such as cryptorchidism and testicular atrophy, reports on familial clustering, co-occurrence with other types of testicular neoplasms, and their presence in disorders of the endocrine system hint at the existence of uncharacterized genetic factors in at least some LCT cases (2, 13, 14, 15).
We have previously shown that mutations in fumarate hydratase (FH) predispose to hereditary leiomyomatosis and renal cell cancer (HLRCC, MIM605839) (16), a syndrome of uterine fibroids, skin leiomyomas, and renal cancers of type II papillary or collecting duct morphology. FH mutations appear to cause tumor growth through activation of the hypoxia pathway (17). We recently identified an LCT in an adult member of an HLRCC family. This man, an FH N64T mutation carrier, previously developed multiple skin leiomyomata and at the age of 55 yr presented with a benign LCT in the left testicle. This observation of a novel tumor type in an FH mutation carrier raised the possibility of a broader phenotypic spectrum in the HLRCC syndrome.
| Materials and Methods |
|---|
|
|
|---|
Owing to the rarity of adult LCTs, archival specimens of tumor and normal material were provided by six hospitals from the United Kingdom. All available cases were ascertained from the histopathology archives at each center. We obtained paraffin-embedded tumor material belonging to the FH mutation carrier from an HLRCC family and from 29 adult LCT cases where there was no reported family history of leiomyomas. Two genitourinary pathologists (M.A.E.-B. and D.M.B.) verified the diagnosis of all tumor samples and highlighted the tumor and normal areas of the sample to be investigated in the molecular study. Of the 30 tumors, none had evidence of calcification or presented bilaterally, five were LCTs with Sertoli cell elements, and 25 had pure LCT histology. Of the 25 pure LCTs, only two had features of malignancy. The median age of tumor presentation was 48 yr (range, 2479 yr). Our research had full research ethics committee approval.
Mutation screening and loss of heterozygosity
DNA isolation from tumors and matched normal testicular tissue was carried out using standard procedures after manual microdissection. We estimated that all tumor samples contained at least 80% neoplastic cells. The full coding sequence of the FH gene was screened by direct sequencing in forward and reverse orientation and loss of heterozygosity (LOH) analysis was carried out using three microsatellites, including two within FH (experimental details available on request). To genotype the LHCGR D578H mutation, we used methods reported by Liu et al. (7). We genotyped the GNAS R201 hotspot using direct sequencing and the primers reported by Fragoso et al. (11).
Immunohistochemistry (IHC) and in situ hybridization (ISH)
CD34 and the hypoxia-inducible factor 1
(HIF-1
) IHC and vascular endothelial growth factor (VEGF) mRNA ISH analyses and scoring were undertaken as reported by Pollard et al. (18). Two independent observers, who were blinded to the tumors mutational status (M.A.E.-B. and R.P.), scored all IHC and ISH experiments. We scored CD34 by counting the number of microvessels seen in 10 random high-power fields (HPF). HIF-1
IHC was scored on the percentage of positively stained nuclei as weak (<25%), moderate (2550%), strong (5075%), and very strong (75100%). VEGF ISH was examined using a dark-field microscope, and signal was scored as relatively weak, moderate, or strong by three independent observers (L.G.C.-C., M.A.E.-B., and R.P.) relative to the staining of the Leydig cells in the normal surrounding testicular tissue. The region of the tumor with the highest expression level was scored.
| Results |
|---|
|
|
|---|
Sequence analysis revealed the germline A to C mutation at cDNA position 352 (N64T) in the familial case to be homozygous/hemizygous in the LCT, suggesting loss of the wild-type FH allele (Fig. 1A
). The mutation was observed in the heterozygous state in the surrounding normal testicular tissue and in DNA isolated from peripheral blood. LOH in the LCT was verified using microsatellites linked to FH (data not shown).
|
LHCGR and GNAS mutations
The screening of the LHCGR mutation hotspot revealed one sporadic LCT, from a 65-yr-old man, that had acquired the D578H somatic substitution. This finding differs greatly from previous reports in children, where a much larger proportion (typically 70100%) of LCTs have this mutation (7, 8, 12). Our results suggest that the D578H mutation occurs in a small fraction of noninfantile LCTs. The screening of the GNAS R201 hotspot did not reveal any mutations. Our results do not confirm the findings of Fragoso et al. (11), where one of two LCTs examined had mutations at this site. To our knowledge, our study represents the largest systematic screening of these two mutational hotspots in adult LCTs and suggests that they are mutated only in a minority of cases.
Activation of hypoxia/angiogenesis pathway in LCTs
IHC and ISH analyses revealed relatively high expression of HIF-1
and VEGF in many of the LCTs. The familial (N64T) LCT was the only sample with a strong level of VEGF expression in the study (see Fig. 2
), and this tumor also showed very strong HIF-1
expression. Around 25% of the remaining LCTs showed moderate expression of VEGF, and of these, over 85% also showed moderate to strong HIF-1
expression (data not shown). The M411I FH mutant LCT had the highest expression of HIF-1
among all the samples examined, showing strong nuclear staining in nearly 100% of the tumor cells (although, because of lack of sufficient tumor material, we were not able to perform VEGF analysis in this case). The LCT with the somatic LHCGR mutation had moderate levels of VEGF but weak HIF-1
expression. Of the approximately 75% of tumors with weak VEGF expression, over 85% also showed moderate to strong HIF-1
expression (data not shown). Interestingly, we observed some well-demarcated intratumor heterogeneity in VEGF and HIF-1
expression (for example, Fig. 2
); regions of stronger expression generally had a higher cell density, possibly indicative of tumor progression.
|
expression and vessel density did not significantly co-vary (details not shown), suggesting that the hypoxia pathway was not simply being induced in some tumors by physiologically low oxygen levels resulting from suboptimal blood supply. | Discussion |
|---|
|
|
|---|
and VEGF in uterine and renal tumors from the HLRCC syndrome (18), consistent with similar findings in LCTs from FH mutation carriers. It appears, however, that activation of the hypoxia pathway also occurs in the development of at least some LCTs without FH mutations, suggesting an overlapping pathogenic mechanism. The presence of skin leiomyomas, which are often subtle, should be sought in all adult LCT patients. FH mutation screening should then be undertaken in such patients. Correspondingly, males in HLRCC families should be warned of the small absolute, but high relative, risk of LCTs and should be advised to undergo regular testicular (self-)examination. Mutations in the FH gene are one of the few known inherited causes of testicular tumors.
| Acknowledgments |
|---|
| Footnotes |
|---|
First Published Online June 6, 2006
Abbreviations: FH, Fumarate hydratase; HIF-1
, hypoxia-inducible factor 1
; HLRCC, hereditary leiomyomatosis and renal cell cancer; HPF, high-power field; IHC, immunohistochemistry; ISH, in situ hybridization; LCT, Leydig cell tumor; LHCGR, LH/choriogonadotropin receptor; LOH, loss of heterozygosity; VEGF, vascular endothelial growth factor.
Received January 26, 2006.
Accepted May 25, 2006.
| References |
|---|
|
|
|---|
regulatory subunit in patients with the Carney complex. Nat Genet 26:8992[CrossRef][Medline]
in tumours which result from germline FH and SDH mutations. Hum Mol Genet 14:22312239This article has been cited by other articles:
![]() |
J. A. Mayr, D. Meierhofer, F. Zimmermann, R. Feichtinger, C. Kogler, M. Ratschek, N. Schmeller, W. Sperl, and B. Kofler Loss of Complex I due to Mitochondrial DNA Mutations in Renal Oncocytoma Clin. Cancer Res., April 15, 2008; 14(8): 2270 - 2275. [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 |