(2002) 21, 7605 – 7608 ª 2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00 www.nature.com/onc

Absence of SDHD in sporadic neuroendocrine tumors and detection of two germline variants in paraganglioma patients

Aurel Perren*,1, Andre´ Barghorn1, Sonja Schmid1, Parvin Saremaslani1,Ju¨ rgen Roth1, Philipp U Heitz1 and Paul Komminoth2

1Department of Pathology, University Hospital, 8091 Zurich, Switzerland; 2Institue of Pathology, 5404 Baden, Switzerland

Allelic loss of the long arm of 11 is frequent forms of these tumors is not well established, in in neuroendocrine tumors (NET) of different organs. contrast to the familial forms which are associated with However, the MEN1 on 11q13 is mutated only in a the MEN1, MEN2 or VHL syndromes. Several studies subset of NET and allelic losses on 11q frequently on different types of sporadic NET have shown extend to the telomere. In this genetic region lies the frequent allelic loss of the long arm of chromosome SDHD which is associated with 11. More specifically, LOH and CGH studies have hereditary paragangliomas (PGL1). We sought to demonstrated a loss of 11q in 26 to 53% of PTA, 28% determine whether SDHD plays a role in the develop- of PCC, 25 to 80% of NET of the lung and ment of sporadic NET. By and deletion gastrointestinal tract and 28 to 69% of EPT (Table analysis of SDHD we were unable to detect any SDHD 1). In addition, the allelic loss has been shown to mutation in 45 NET of the lung, gastrointestinal tract, extend to the telomere of 11q in all types of NET pancreas or parathyroid. However, we found allelic investigated (Table 1). These findings point towards a deletions in 20 to 50% of all tumors but parathyroid tumor suppressor gene located on 11q involved in the adenomas. Furthermore, we found heterozygous germline tumorigenesis of sporadic NET with the MEN1 gene variants in 2/8 paragangliomas. A first case of variant on 11q13 representing a candidate. However, we and c.149 A4G (H50R) was found in a patient with an others could demonstrate that only a subset of EPT, extra-adrenal pheochromocytoma, the other variant c.34 PTA and PCC carries mutations of the MEN1 gene G4A (G12S) in a patient with a paratracheal and that these tumors exhibit a 2 – 3-fold higher paraganglioma, C-cell hyperplasia of the thyroid and frequency of 11q13 LOH when compared to mutation hyperplasia of ACTH-producing cells of the pituitary rates (Table 1). Other studies and our own data gland. Both variants were absent in 93 controls. Our showing isolated chromosomal losses more telomeric of results demonstrate that somatic SDHD mutations are the MEN1 gene also indicate that there might be other rare in sporadic NET. However, LOH alone could lead tumor suppressor involved in the initiation and to a complete loss of function since SDHD is an progression of NET (Eubanks et al., 1994; Chakrabarti imprinted gene. Furthermore, we describe two germline et al., 1998; Gortz et al., 1999; Rigaud et al., 2001). variants possibly causing hereditary paragangliomas. One of the regions of interest which has been narrowed Oncogene (2002) 21, 7605 – 7608. doi:10.1038/sj.onc. down is 11q23 (Rigaud et al., 2001). It harbors the 1205812 tumor suppressor gene succinate dehydrogenase subunit D (SDHD) (Hirawake et al., 1999) which acts as Keywords: SDHD; endocrine tumors; mutation; allelic hydrophobic membrane anchor for the catalytically loss active subunits of cytochrome II. In addition, it participates in electron transport and interacts with quinones (Scheffler, 1998). Indirect evidence is in favor Tumors of the diffuse neuroendocrine system are for a tumor suppressor function of SDHD in endocrine defined by the common expression of neuroendocrine tumors: It is responsible for familial PGA type 1 markers and share certain morphologic features. They (PGL1, OMIM 168000); it is mutated in 10% of encompass neuroendocrine tumors (NET) of the lung apparently sporadic PCC (Gimm et al., 2000) and it and gastrointestinal tract (carcinoids), endocrine seems to be selectively imprinted in some neuroendo- pancreatic tumors (EPT), parathyroid adenomas crine tissues (van der Mey et al., 1989). (PTA), pheochromocytomas (PCC) and paraganglio- The aim of the present study was to investigate a mas (PGA). The genetic background of the sporadic possible role of SDHD inactivation in different types of sporadic NET of different organ locations. We there- fore examined 62 sporadic NET for SDHD mutations by PCR/SSCP and allelic deletions. *Correspondence: A Perren, Department of Pathology, University Somatic SDHD mutations were undetectable in 21 Hospital Zurich, Schmelzbergstr 12, CH-8091 Zurich, Switzerland; E-mail: [email protected] EPT, 14 NET of the gastrointestinal tract and the lung, Received 3 April 2002; revised 28 June 2002; accepted 28 June nine PCC, eight PGA and 10 PTA (Table 2). The only 2002 somatic SDHD mutation reported to date was found in SDHD in sporadic neuroendocrine tumors A Perren et al 7606 1/18 sporadic PCC (5%) (Gimm et al., 2000). recently been described in another pheochromocytoma However, a definite answer about the role of SDHD patient (Gimm et al., 2000). The of our in these tumors cannot be given. The disease phenotype patient (PGA3) with a G12S germline variation was of familial PGL1 caused by SDHD mutations is exceptional, consisting of a paratracheal PGA, C-cell inherited as an autosomal dominant trait with incomplete penetrance when transmitted through fathers whereas no disease phenotype occurs when transmitted maternally. This inheritance pattern is consistent with genomic imprinting of the maternal allele of the SDHD gene (van der Mey et al., 1989). However, allelic expression analysis in fetal brain and lymphomatoid cell lines revealed biallelic expression (Malik et al., 2000). Recently, SDHD was shown to be monoallelically expressed in paragangliomas (Baden- hop et al., 2001). If SDHD is also a monoallelically expressed tumor suppressor in other endocrine tissues, it is predisposed to gene inactivation through a single genetic ‘hit’ by loss of the expressed paternal allele, leading to a cell devoid of SDHD activity. With the exception of PTA, which exhibited neither SDHD mutations nor LOH, the examined neuroendocrine Figure 1 LOH rates of 11q23 in neuroendocrine tumors using tumor types showed allelic losses of the SDHD locus in two microsatellite markers flanking SDHD (D11S900, D11S1347). EPT: endocrine pancreatic tumor; CD: neuroendo- 20 to 57% (Table 2, Figure 1). Thirty-four out of 44 crine tumor of lung and gastrointestinal tract; PCC: pheochromo- (77%) pairs of tumor and germline DNA were cytoma (adrenal); PGA: paraganglioma; PTA: parathyroid informative for at least one of the two markers. adenoma Ongoing work is examining whether there is selective allelic loss of the non imprinted allele and whether Table 2 Material and results SDHD expression is completely lost in these tumors. Although we studied clinically sporadic NET, we Number LOH rate SDHD detected two amino acid variants of the SDHD gene in NET lung 8 3/4 0/8 tumor and germline DNA of two patients, indicating a NET GIT 6 1/3 0/6 heritable tumor predisposition (Figure 2). They were EPT 21 4/14 0/21 PTA 10 0/5 0/10 found in one patient (PGA8) suffering from a para- PCC 9 1/3 0/9 adrenal sympathic PGA (para-adrenal pheochromocy- PGA parasympathic 4 0/2 1/4* toma) and in one patient (PGA3) with a cervical PGA PGA symapthic 4 1/3 1/4* among a total of eight patients with PGA (25%). The NET: neuroendocrine tumor; GIT: gastrointestinal tract; EPT: former patient exhibited an SDHD variant H50R in endocrine pancreatic tumor; PTA: parathyroid adenoma; PCC: exon 2 which has not yet been described while the pheochromocytoma; PGA: paraganglioma; No sporadic SHDH G12S variant in exon 1 found in the other patient has mutations but two SDHD germline variants were detected

Table 1 11q losses distal of the MEN1 locus Tumor typea 11q loss (%) MEN1 mutation (%) Referenceb

NET lung 34/46 (78) Jakobovitz et al., 1996 12/23 (52) Walch et al., 1998b 3/12 (25) 3/12 (25) Gortz et al., 1999b 4/11 (36) Zhao et al., 2000b 13/20 (65) Petzmann et al., 2001b NET GIT 3/9 (33) 2/9 (22) Gortz et al., 1999b 14/27 (52) D’Adda et al., 1999b 1/21 (5) Zhao et al., 2000b EPT 11/39 (28) Eubanks et al., 1994 14/30 (47) 4/30 (13) Gortz et al., 1999b 16/44 (36) Speel et al., 1999b 9/13 (69) Rigaud et al., 2001b PTA 16/61 (26) Friedman et al., 1992 15/43 (35) Iwasaki, 1996 6/25 (24) Farnebo et al., 1997 14/26 (54) 7/26 (27) Farnebo et al., 1999b PCC 8/29 (28) Dannenberg et al., 2000b

aNET: neuroendocrine tumor (carcinoid); GIT: gastrointestinal tract; EPT: endocrine pancreatic tumor; PTA: parathyroid adenoma; PCC: pheochromocytoma. bThese authors showed allelic loss extending to 11q23 or postulated a tumor suppressor distally of 11q13

Oncogene SDHD in sporadic neuroendocrine tumors A Perren et al 7607 2000), Gimm et al. (2000) described the G12S variant in one patient with a pheochromocytoma and in 1/78 control alleles. Our data provide further evidence that this G12S variant is a mutation causing familial PGA and PCC. Both PGA and PCC are tumors that have been underdiagnosed in the past and the complex inheritance pattern due to imprinting as well as the incomplete penetrance makes it even more difficult to clinically recognize familial forms. The fact that the tumor PGA3 showed retention of heterozygosity of both flanking polymorphic markers is intriguing since most of the examined tumors from PGL1 patients showed LOH of the wildtype allele, including the PCC patient with the same germline sequence variant G12S. Apart from technical problems, i.e. admixture of non-neoplastic tissue in the tumor DNA, which we would like to exclude due to careful microdissection, there are at least three possible explanations for this finding. First, there might be only a small allelic loss spanning neither of the two adjacent markers. Second, one allele might be inactivated by epigenetic mechanisms as (Baysal et al., 1999, 2000) or third and most probably, one allele is the imprinted maternal allele (Badenhop et al., 2001). SDHD is unusual among tumor suppressor genes in that it encodes the small subunit of cytochrome b (CybS), a mitochondrial protein. Cytochrome b is part of the mitochondrial complex II (succinate-ubiquinone oxidoreductase) which is involved in the citric acid cycle (Hagerhall, 1997). CybS acts together with CybL as hydrophobic membrane anchor for the catalytically active flavoprotein and iron-sulfur subunits. In addi- tion, it participates in electron transport and Figure 2 DNA was extracted from frozen or paraffin embedded interaction with quinones (Scheffler, 1998). The neoplastic and non-neoplastic tissue (Perren et al., 1998). PCR sequence variant of PGA3 in exon 1 changes Gly to was performed using described conditions (Baysal et al., 2000). a larger neutral and hydrophilic Ser in exon 1. This SSCP Analysis was carried out on non-denaturing polyacrylamide could potentially impair correct localization of the gels at 80 W for 5 h at room temperature followed by silver stain- ing as described (Komminoth et al., 1994): Left lane: normal con- protein as the amino-terminal sequence directs it to trol, denatured. Middle lane: undenatured probe. Right lane: mitochondrial and the appropriate mitochondrial Denatured samples from PGL3 and PGL8 showing additional compartment (Koehler, 2000; Pfanner, 2000). In bands (?). Cycle sequencing reveals H50R variant in exon 2 of PGA8, His is replaced by Arg. A specific function of PGL3 and G12S variant in exon 1 of PGL8 this region is not known but this amino acid change could result in a conformational change of the protein. To date no functional analysis exists that examined the cellular effect of these amino acid variants. hyperplasia of the thyroid and hyperplasia of ACTH In conclusion we demonstrate the lack of somatic producing cells of the pituitary suggestive of a SDHD mutations in NET of the lung, gastrointestinal predisposition to neuroendocrine tumors. History trace, pancreas and parathyroids. However, LOH rates taking revealed these two patients having no family ranging from 20 to 50% except in PGA, do not exclude background of PGA or of other NET. In order to a role of SDHD in the examined tumors, as the gene is elucidate whether these SDHD variants represent potentially imprinted in these tissues. Studies regarding polymorphisms or mutations, we examined blood the SDHD expression in NET will clarify this issue. samples of 93 unrelated control persons (186 control We detected two germline sequence variants in PGA alleles) without a history of endocrine tumors and did which potentially are mutations. As the familial nature not detect any SDHD variant in exons 1 and 2. While of this disease is difficult to recognize clinically, Baysal and collaborators did not find these two screening of PGA and PCC patients for SDHD variants in 200 control (Baysal et al., mutations seems to be mandatory.

Oncogene SDHD in sporadic neuroendocrine tumors A Perren et al 7608 Acknowledgements Supported by Swiss League grant SKL-997-02-2000 and Swiss National Science Foundation grant 31- 618845.00.

References

Badenhop RF, Cherian S, Lord RS, Baysal BE, Taschner PE Hirawake H, Taniwaki M, Tamura A, Amino H, Tomitsuka and Schofield PR. (2001). Genes Chromosomes Cancer, 31, E and Kita K. (1999). Biochim. Biophys Acta, 1412, 295 – 255 – 263. 300. Baysal BE, Ferrell RE, Willett-Brozick JE, Lawrence EC, Iwasaki H. (1996). Int. Surg., 81, 71 – 76. Myssiorek D, Bosch A, van der Mey A, Taschner PE, Jakobovitz G, Nass D, Demarco L, Barbosa A, Simoni FB, Rubinstein WS, Myers EN, Richard III CW, Cornelisse Rechavi G and Friedman E. (1996). J. Clin. Endocrinol. CJ, Devilee P and Devlin B. (2000). Science, 287, 848 – 851. Metab., 81, 3164 – 3167. Baysal BE, van Schothorst EM, Farr JE, Grashof P, Koehler CM. (2000). FEBS Lett., 476, 27 – 31. Myssiorek D, Rubinstein WS, Taschner P, Cornelisse CJ, Komminoth P, Kunz E, Hiort O, Schroder S, Matias-Guiu Devlin B, Devilee P and Richard 3rd CW. (1999). Hum. X, Christiansen G, Roth J and Heitz PU. (1994). Am. J. Genet., 104, 219 – 225. Pathol., 145, 922 – 929. Chakrabarti R, Srivatsan ES, Wood TF, Eubanks PJ, Malik K, Salpekar A, Hancock A, Moorwood K, Jackson S, Ebrahimi SA, Gatti RA, Passaro EJ and Sawicki MP. Charles A and Brown KW. (2000). Cancer Res., 60, 2356 – (1998). Genes Chromosomes Cancer, 22, 130 – 137. 2360. D’Adda T, Keller G, Bordi C and Hofler H. (1999). Lab. Perren A, Roth J, Muletta-Feurer S, Saremaslani P, Speel Invest., 79, 671 – 677. EJ, Heitz PU and Komminoth P. (1998). J. Pathol., 186, Dannenberg H, Speel EJ, Zhao J, Saremaslani P, van Der 363 – 371. Harst E, Roth J, Heitz PU, Bonjer HJ, Dinjens WN, Mooi Petzmann S, Ullmann R, Klemen H, Renner H and Popper WJ, Komminoth P and de Krijger RR. (2000). Am. J. HH. (2001). Hum. Pathol., 32, 333 – 338. Pathol., 157, 353 – 359. Pfanner N. (2000). Curr. Biol., 10, R412 – R415. Eubanks PJ, Sawicki MP, Samara GJ, Gatti R, Nakamura Rigaud G, Missiaglia E, Moore PS, Zamboni G, Falconi M, Y, Tsao D, Johnson C, Hurwitz M, Wan Y-JY and Passaro TalaminiG,PesciA,BaronA,LissandriniD,RindiG, E. (1994). Am. J. Surg., 167, 180 – 185. Grigolato P, Pederzoli P and Scarpa A. (2001). Cancer FarneboF,KytolaS,TehBT,DwightT,WongFK,HoogA, Res., 61, 285 – 292. Elvius M, Wassif WS, Thompson NW, Farnebo LO, Scheffler IE. (1998). Prog. Nucleic Acids. Res. Mol. Biol., 60, Sandelin K and Larsson C. (1999). J. Clin. Endocrinol. 267 – 315. Metab., 84, 3775 – 3780. Speel EJ, Richter J, Moch H, Egenter C, Saremaslani P, Farnebo F, Teh BT, Dotzenrath C, Wassif WS, Svensson A, Rutimann K, Zhao J, Barghorn A, Roth J, Heitz PU and White I, Betz R, Goretzki P, Sandelin K, Farnebo LO and Komminoth P. (1999). Am. J. Pathol., 155, 1787 – 1794. Larsson C. (1997). Hum. Genet., 99, 342 – 349. van der Mey AG, Maaswinkel-Mooy PD, Cornelisse CJ, Friedman E, De Marco L, Gejman PV, Norton JA, Bale AE, Schmidt PH and van de Kamp JJ. (1989). Lancet, 2, 1291 – Aurbach GD, Spiegel AM and Marx SJ. (1992). Cancer 1294. Res., 52, 6804 – 6809. Walch AK, Zitzelsberger HF, Aubele MM, Mattis AE, Gimm O, Armanios M, Dziema H, Neumann HP and Eng C. Bauchinger M, Candidus S, Prauer HW, Werner M and (2000). Cancer Res., 60, 6822 – 6825. Hofler H. (1998). Am. J. Pathol., 153, 1089 – 1098. Gortz B, Roth J, Krahenmann A, de Krijger RR, Muletta- Zhao J, de Krijger RR, Meier D, Speel EJ, Saremaslani P, Feurer S, Rutimann K, Saremaslani P, Speel EJ, Heitz PU Muletta-Feurer S, Matter C, Roth J, Heitz PU and and Komminoth P. (1999). Am. J. Pathol., 154, 429 – 436. Komminoth P. (2000). Am. J. Pathol., 157, 1431 – 1438. Hagerhall C. (1997). Biochim. Biophys. Acta., 1320, 107 – 141.

Oncogene