MALIGNANT ADRENAL tumors occur with an incidence

002-972X/00/$03.00/0 Vol. 85,. The Journal of Clinical Endocrinology & Metabolism Printed in U.S.A. Copyright 2000 by The Endocrine Society Complete Sequencing and Messenger Ribonucleic Acid Expression Analysis of the MEN I Gene in Adrenal Cancer* KLAUS-MARTIN SCHULTE, MARTINA MENGEL, MATTHIAS HEINZE, DIETMAR SIMON, SIBYLLE SCHEURING, KARL KÖHRER, A HANS-DIETRICH RÖHER Klinik für Allgemeine Chirurgie und Unfallchirurgie (K.-M.S., M.M., M.H., D.S., H.-D.R.) and Biologisch-Medizinisches Forschungszentrum (S.S., K.K.), Heinrich-Heine-University, 40225 Düsseldorf, Germany ABSTRACT Adrenal cancer is a rare sporadic disease that has also been observed in the context of multiple endocrine neoplasia type I (MEN I). Adrenal lesions occur in up to 40% of MEN I patients. Loss of heterozygosity of the q3 band harboring the menin gene has been reported in more than 50% of patients with adrenal cancer. Despite this high index of suspicion, former screening studies did not reveal mutations of the MEN I gene in 28 patients. We identified loss of heterozygosity of q3 microsatellites in five of five patients (00%). In 40%, heterozygosity was retained in codon 48 of the MEN I gene. Complete direct DNA sequencing data of the entire coding region and adjacent splice sites of the MEN I gene were obtained in 4 patients with sporadic adrenal cancer. In only one of them a heterozygous missense mutation, R76Q (exon 3), was identified. Due to the heterozygous pattern and unknown biological effect of this mutation, it is not clear whether there is a causal relationship with MALIGNANT ADRENAL tumors occur with an incidence of about.7 new cases per million per year (). The genetic basis of either benign or malignant tumors is not completely understood (2 4). Lesions involve mutations of such genes as p53 (5), p2 (6), or the ACTH receptor (7). Some adrenal tumors occur in the context of hereditary tumor syndromes such as the Li-Fraumeni syndrome (8), the Beckwith-Wiedemann syndrome (9 ), or the Carney complex (2). In about 30 40% of patients with multiple endocrine neoplasia type I (MEN I), adenomas or bilateral hyperplasia of the adrenal gland are observed (3 6). Adrenal cancer has been reported in few patients suffering from a MEN I syndrome (3, 7). Loss of heterozygosity (LOH) in adrenal tumors frequently involves the loci 2q, 4p, 3p, 8p (8), 7p (9), p and q (8 2), and 3q (9). With regard to adrenal cancer Received August 27, 999. Revision received October 4, 999. Accepted October 5, 999. Address correspondence and requests for reprints to: Dr. med. Klaus-Martin Schulte, Klinik für Allgemeine Chirurgie und Unfallchirurgie, Medizinische Einrichtungen, Heinrich-Heine-University, Moorenstr. 5, 40225 Düsseldorf, Germany. E-mail: schultekm@ med.uniduesseldorf.de. * Supported by Grant 977204 from the research committee of the Heinrich-Heine-University (Düsseldorf, Germany). K.M.S. is supported by a grant from the German National Research Foundation (Deutsche Forschungsgemeinschaft DFG Schu 270/-2). adrenal cancer. The total mutation frequency in sporadic adrenal cancer is of4 (7%). Menin messenger RNA expression was identified in 4 of 4 patients (00%). Transcriptional inactivation of the menin gene is, hence, unlikely to cause loss of its tumor suppressor function in adrenal cancer. Furthermore, we examined three patients who presented adrenal cancer in the context of sporadic multiglandular endocrine tumor disease previously diagnosed on clinical grounds to be MEN I syndrome. An opal stop codon mutation was identified in codon 26 (exon 2) in the adrenal cancer of one of these patients. Formation of the adrenal cancer in this patient may be rather coincidental because the mutation was present in a heterozygous pattern. There was no mutation of the menin gene in the two other patients. This may mean that formation of adrenal cancer in the context of multiglandular endocrine disease denotes an entity different from MEN I in some patients. (J Clin Endocrinol Metab 85: 44 448, 2000) only, LOH on chromosome q3 is a rather frequent event. In this entity, it was seen in 2 of 2 patients (22), in of (3), 2 of 2 (00%) (2), 4 of 8 (50%) (20), 9 of 9 (47%) (8), of 4 (25%) using a set of q3 microsatellites (23), and 6 of 0 (60%) using the MEN I sequence in a fluorescence in situ hybridization assay (24). In total, it was present in 23 of 44 (52%) examined patients in whom different technical approaches were used. In MEN I patients, the frequency of q3 LOH in either benign or malignant adrenal lesions was only moderate, however (3, 23). In a study of six patients with clinically overt MEN I syndrome and associated adrenal lesions, only the patient with adrenal cancer had LOH of the q3 band (3). The gene responsible for MEN I is located on chromosome q3 and has been characterized as a tumor suppressor gene (25). Inactivating mutations of the MEN I gene occur not only in patients with a typical syndrome, but also in sporadic tumors that otherwise are part of the MEN I disease. Examples are parathyroid adenomas (2%) (26), gastrinomas (33%) (27), bronchial carcinoid tumors (36%) (28), and, rarely, pituitary tumors (29, 30). It is on this background that the MEN I gene has been considered as a candidate gene for the genesis of adrenal neoplasms. Three recent reports described absence of MEN I gene mutation in sporadic adrenocortical neoplasms. These studies used screening techniques for detection of MEN I gene lesions and analyzed a total of 28 patients (8, 23, 24). 44 Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207

442 SCHULTE ET AL. JCE&M 2000 Vol 85 Our study presents complete direct DNA sequencing of the MEN I gene, including adjacent splice sites in 4 patients with sporadic adrenal cancer and 3 patients with sporadic MEN I-like disease. LOH studies were performed in the q3 region harboring the MEN I gene. Messenger RNA (mrna) expression of menin was studied in 4 patients. Subjects Subjects and Methods Tumor tissue was obtained at adrenalectomy (ADX) from the central part of the lesion by an experienced endocrine surgeon, and tissue was immediately frozen at 80 C under tissue tek. Particular care was taken to obtain a part of tumor tissue in which contamination by normal adrenal tissue could be safely excluded already on macroscopical examination. Diagnosis was established by histopathological examination. Malignancy was determined by presence of infiltrative growth including capsule and/or vessels. Clinically, malignancy was defined in presence of metastasis or local recurrence of an infiltrating mass. Patients with hereditary tumor disease were excluded. We included 4 sporadic adrenocortical carcinomas and three patients with adrenal cancer in the setting of sporadic syndrome similar to MEN I disease (patients 5, 4, and 00) (Table ). Tissue or blood was removed after informed consent of the patients. All parts of the study were conducted according to the Declaration of Helsinki principles. Material and methods DNA from blood and tumor was extracted using the Qiagen blood and tissue kits (Qiagen, Hilden, Germany). Tumor tissue was cut to 0- m slices by a microtome. A hematoxylin and eosin-stained slice adjacent to that used for DNA isolation was obtained to confirm the adrenal identity of the tissue sample. LOH analysis We used the polymorphic markers PYGM, DS480, DS987, and DS449. PYGM is located centromeric of the MEN I gene, DS480 is located centromeric to PYGM, and DS449 and DS987 are located telomeric to the MEN I gene (Table 2). PCR conditions and LOH scoring have been formerly published (3). Fluorescence-labeled amplimers were separated by capillary electrophoresis under denaturing conditions with an ABI PRISM 30 Genetic Analyzer A (Perkin- Elmer Biosystems, Branchburg, NJ). GeneScan 350, TAMRA marker, was used as an internal size standard. The allele ratio for the normal sample was calculated by division of the peak area integrals of the two alleles. The LOH ratio was calculated by dividing the LOH ratio of the tumor by that of the corresponding blood control. LOH was defined in presence of an LOH ratio less than 0.4. Mutation analysis Exons 2 0 and the neighboring splice junctions were amplified by PCR using oligonucleotide primers located in the intronic parts of the gene (Table 2). The reaction contained 5 l TaqMasterMix (Qiagen), 00 ng genomic DNA, and m oligonucleotide primers and was cycled at the following conditions: 94 C 5 min, 35 cycles of 94 C 30 sec, 55 60 C 30 sec, 72 C min, and 72 C 5 min. Reaction products were purified using the PCR purification kit or the gel extraction kit (Qiagen) after % agarose gel electrophoresis. Cycle sequencing by Taq DNA-polymerase was performed with M3-oligonucleotides in a 0- l volume containing 20 40 ng PCR products and 0 pmol forward and reverse M3 primer using a dye terminator method with 25 cycles of 96 C 0 sec, 50 C 5 sec, and 60 C 4 min (ABI Prism BigDye Terminator Cycle Sequencing Ready Reaction Kit; Perkin-Elmer Biosystems). Sequencing products were separated by PAGE with an ABI377 DNA sequencer (Perkin-Elmer Biosystems). Chromatograms were analyzed using the Lasergene Navigator software (GATC, Koustauz, Germany) and blasted against the published MEN I genomic sequence (GenBank accession number U93237). In rare cases where sense- and antisense sequencing did not TABLE. Clinical characteristics of patients with adrenal cancer. Tumor size Age Sex Clinical diagnosis Hormone production Surgical treatment 5 7 2 cm 34 M HPT, pituitary adenoma, ne ADX and nephrectomy en bloc MEN I syndrome 0 4 4 cm 67 M Incidentaloma ne ADX and nephrectomy en bloc 4 8 cm 45 F MEN I syndrome, Cushing Cortisol ADX left, subtotal ADX right syndrome, pituitary adenoma, contralat adrenal, hyperplasia 5 0 8 73 M Radiculitis ne ADX 7 8 cm 42 M Hypertension Aldosterone cortisol, renin ADX and nephrectomy en bloc 24 4 0 cm 47 F Cushing syndrome, Cortisol, DHEAS, ADX, transperitoneal hirsutism, hypertension testosterone 29 6 4.5 cm 56 F Cushing syndrome Cortisol Tumor excision, transperitoneal 34 9 8 cm 46 F Cushing syndrome, caval Cortisol aldosterone ADX, transperitoneal vein thrombosis 5 9 5 cm 37 F Met adrenal cancer Cortisol, DHEAS ADX, transperitoneal 57 8 5 cm 5 M Pubertas praecox (at 0 yr) Testosterone ADX, transperitoneal 62 8 4 cm 60 M Gynecomastia Estradiol ADX, transperitoneal 64 8.5 6 cm 28 F Amenorrhea, hirsutism, Cortisol, DHEAS, estradiol ADX, transperitoneal Cushing syndrome 7 9 cm 40 F Back pain ne Tumor resection and splenectomy en bloc 00 Local recurrence 32 F MEN I syndrome, ne Local excision parathyroid adenoma, foll. thyroid adenoma, sec. amenorrhea, father: PTH (2 glands) 0 4.5 4.5 cm 78 F Cushing syndrome, goiter Cortisol ADX 02 7 2 cm 39 F Cushing syndrome, Cortisol Local excision, transperitoneal recurrent adrenal cancer 04 9 8 cm 49 F Liver metastasis ne ADX transperitoneal, liver tumor resection Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207

MEN I GENE IN ADRENAL CANCER 443 TABLE 2. Oligonucleotide primers used for amplification and sequencing Exon Position Sequence (5-3 ) BP T/C 2 (2288-2732) 2285-2302 tgt aaa acg acg gcc agt gcc atg ggg ctg aag gcc 536 57 2782-2762 cag gaa aca gct atg acc gtt ttg aag aag tgg gtc atg 3 (4297-4505) 4096-49 tgt aaa acg acg gcc agt gtt gga cat aga ggg tgt aaa cag 6 63 467-4649 cag gaa aca gct atg acc gtg cct gct tca ggg aat gac ag or 476-495 tgt aaa acg acg gcc agt atc tga ggt tgg gtc aca gg 436 4576-4558 cag gaa aca gct atg acc aaa tgg agt ccc ttg ggt g 4 (476-4844) 4649-467 tgt aaa acg acg gcc agt ctg tca ttc cct gaa gca ggc ac 29 57 4904-488 cag gaa aca gct atg acc ggt ccc aca gca agt caa gtc tgg 5 6 (577-527, 5298-5385) 538-56 tgt aaa acg acg gcc agt cct gtt ccg tgg ctc ata act ctc 38 57 5520-5497 cag gaa aca gct atg acc aca gtt gac aca aaa tga gac tgg 7 (6025-66) 5828-5849 tgt aaa acg acg gcc agt cct cag cca gca gtc ctg tag a 4 68 6203-682 cag gaa aca gct atg acc gaa gaa agg aca ggc tgc agg c 8 (6623-6758) 6577-6600 tgt aaa acg acg gcc agt tgg tga gac ccc ttc aga tcc tac 32 57 6853-6834 cag gaa aca gct atg acc cca atc cct aat ccc gta cat gc 9 (796-7360) 75-774 tgt aaa acg acg gcc agt ggt gag taa gag act gat ctg tgc 3 57 7426-7404 cag gaa aca gct atg acc gtc tga caa gcc cgt ggc tgc tg 0, 5 prime (7578-8060) 7554-7573 tgt aaa acg acg gcc agt acc ttg ctc tcc cca ctg gc 404 58 7922-7903 cag gaa aca gct atg acc cag cag ctc ctt cat gcc ct 0, 3 prime (7578-8060) 778-7737 tgt aaa acg acg gcc agt gcc agc act gga caa ggg cc 339 58 802-8002 cag gaa aca gct atg acc gta gtc act agg ggt gga ca M3 for sequencing Forward tgt aaa acg acg gcc agt 60 cag gaa aca gct atg acc PYGM Forward F-cta gca gag tcc acc tac tg 60 gtc gtc agg tag caa ctg ac DS480 Forward F-ccc tct tgc ctg tgt tga aat 60 ttt gag gta ggc ttc gta ta DS987 Forward F-gac tcc agt ctg ggc aat aaa agc 60 ggt ggc agc atg acc tct aaa g DS449 Forward F-ggt gaa aaa aca cac ttg tct g 60 ggc gac ata gtg aga tcc tgt Menin 4-6 RT-PCR 477-4736 Cy5 gct ggc tgt acc tga aag ga 257 57 5385-5366 gtt gtg gta gag ggt gag tg yield a definite sequence, the exon was reamplified from genomic DNA and resubjected to forward and reverse sequencing. Mutations and polymorphisms were subjected to repeated sequencing from independent tissue samples and independently produced amplimers. RNA isolation and RT-PCR of the menin gene A 0- m microtome slice was used for isolation of RNA using TRIzol (Life Technologies, Inc., Grand Island, NY) by a method modified from Chomczynski and Sacchi (32). RT was done using murine leucemia reverse transcriptase mmulv using a complementary DNA (cdna) synthesis kit according to the instructions of the manufacturer (Apbiotech, Freiburg, Germany). PCR of the cdna was performed using the oligonucleotides menin 4 and menin 6 (Table2), which span introns 4 and 5 and yield a product of 257 bp. The corresponding DNA amplimer is 668 bp long. PCR conditions were identical to those given above, with an annealing temperature of 57 C. The product was visualized by % agarose electrophoresis and ethidium bromide staining. Quality of the cdna was assured by previous amplification of a house-keeping gene ( -actin) and a low copy number mrna (transforming growth factor -type II receptor or edg-2-receptor). Only cdna, which consistently yielded amplimers with these test primers, was used for menin expression analysis. Failure to obtain the mrna-specific band at 257 bp was controlled by RT-PCR repeated two times. Results LOH analysis Pairs of tumor and leukocyte DNA were available in five patients with sporadic adrenocortical carcinoma. Fig. demonstrates the LOH in informative alleles of q3 loci and corresponding sequence information on intragenic polymorphic sites in five adrenal cancers. In all patients with adrenal cancer, LOH could be demonstrated by at least one variable number of tandem repeats-marker. Markers were informative in 75 80% of cases examined. LOH of the q3 region did not affect the MEN I gene in two of five patients with q3 LOH because these patients had retained heterozygosity at position 7264 (patient 64) or position 2722 (patient 7) in the same DNA sample that had been used for LOH determination. In 2 additional adrenal cancers no patient blood was available. In 5 of these (42%) LOH was excluded by heterozygous sequence determination of the intragenic polymorphic sites 2722 (codon 45) and 7264 (codon 423). In seven patients (58%), sequence information did not allow further discrimination of homo- vs. hemizygosity. Mutation analysis Unequivocal sequence determination by direct sequencing of strand and antistrand DNA was achieved for 7 patients with adrenal cancer. All coding exons (2 0) were analyzed. Amplimers typically included the splice junctional positions at the exon/intron borders (Table 2). Analysis of the adjacent intronic sequences of exons 2 0 did not yield any sign of mutations in splice sites. In position 682 (intron 8) we regularly identified the sequence aca ggcca and not acagggcca, as published previously (GenBank U93237). In position 4243 (intron 2) we regularly identified the sequence tggccccctttc and not tggcccc tttc, as previously published (GenBank U93237). There were only few mutations and polymorphisms in the Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207

444 SCHULTE ET AL. JCE&M 2000 Vol 85 Fig.. LOH analysis in five patients with adrenal cancer. entire open reading frame of the MEN I gene in patients with malignant adrenal tumors. Patient 7 had a heterozygous missense mutation R76Q in exon 3 (Fig. 2). Patient 5 had a stop codon mutation in codon 26 by transition of TGG to TGA, forming an OPAL stop codon from tryptophane (Fig. 3). Patients 7 and 64 showed heterozygous transition from AGC to AGT S45S, a previously defined intragenic polymorphism (genomic position 2722). Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207

MEN I GENE IN ADRENAL CANCER 445 Menin mrna detection by RT-PCR High-quality RNA could be obtained from archival tissues of 4 patients with adrenal cancer. Samples with potential contamination by surrounding adrenal tissue were excluded from analysis. Suitability of derived cdnas for gene expression analysis was ascertained by amplification of a 446-bp b-actin fragment and a 63-bp fragment of the transforming growth factor b receptor type II using intron-spanning primers (data not shown). Expression of menin was demonstrated by amplification of a 257-bp fragment including parts of exons 4, 5, and 6. Contaminating DNA yielded a 668-bp amplimer (Fig. 4). The 257-bp menin mrna amplimer was present in 4 of 4 (00%) of adrenal cancer tissues (Table 3). Discussion Fig. 2. Heterozygosity is present in position 4363 (codon 76) in a patient with adrenal cancer. The missense mutation CAG codes for an exchange from arginine to glutamine in the third exon of the menin gene. Forward and reverse sequence are presented. The above lane shows the generic sequence. FIG. 3. Opal stop codon mutation in patient 5. Heterozygosity is present in position 2665 (codon 26). The missense mutation TGA codes for an exchange from tryptophan to an opal stop codon in the second exon of the menin gene. Presented are two reverse and one forward sequence. Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207 In our study, all five patients in whom LOH data were available showed at least one LOH on q3. This finding is well in accordance with published data that demonstrated 52% of LOH on q3 in a total of 44 patients (3, 8, 20 24). In two of our patients, however, retained intragenic heterozygosity in codon 48 rendered unlikely that the LOH included the MEN I gene. This could hint toward a relevant different tumor suppressor gene located in this chromosomal area. The frequency of intragenic heterozygosity in the polymorphic codon 48 was 35%, comparable to data obtained in large groups of MEN I kindreds (33). Hitherto, screening data for mutations are available for a total of 28 patients with adrenal cancer from three independent studies. Dideoxyfingerprinting, which was used in 0

446 SCHULTE ET AL. patients (26), has a high sensitivity to detect mutations in the MEN I gene (25, 26, 34 36). Single-strand conformational polymorphism/single-strand conformational analysis/single-strand conformational variation used for the remaining 8 patients is less sensitive and detects about 60 80% of mutations of the MEN I gene when applied after prior LOH analysis (8, 33). Because MEN I mutations in adrenal cancer may have escaped detection due to the relatively small sample size and the limited sensitivity of single-strand confortable 3. Menin mrna expression in adrenal cancer Patient. Menin 257 LOH by VNTR Intragenically heterozygous 5 0 5 7 24 29 5 57 62 64 7 00 0 04 07 Presence of menin mrna expression compared to presence of an extragenic LOH on chromosome band q3 and presence of retained intragenic heterozygozity in bp positions 2722, 6053, and 7264. Menin 257: presence of the 257 bp menin RT-PCR product in amplimers from tumor tissue. LOH by VNTR: presence of loss of heterozygozity in at least one microsatellite on q3., t done. Intragenically heterozygous: presence of a heterozygous constellation in at least one out of three identified intragenic polymorphic sites: 2722, 6053 and 7264. FIG. 4. Menin nrna expression in adrenal cancer assessed by RT-PCR. Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207 JCE & M 2000 Vol 85 mational polymorphism screening techniques, we here add complete direct sequencing data on 4 additional patients with sporadic isolated adrenal cancer. A single heterozygous missense mutation was observed in exon 3, codon 76, coding for an amino acid exchange from arginin to glutamine in a patient with sporadic adrenal cancer. If the heterozygous amino acid exchange decreases, menin function could not be examined because useful tests for the biological consequences of a mutation are lacking. This 40-yr-old female had neither a family history nor other coincident or metachronic endocrine tumors. mutation was identified in 3 patients yielding a mutation frequency of 7%, if the above heterozygous mutation is taken into consideration and 0% if it is excluded. It has long been discussed whether adrenal cancer truly belongs to the clinical spectrum of the MEN I disease. The presentation of two very rare diseases in one patient argues in favor of a mechanistic relation, but molecular proof has not been given yet. A stop codon mutation in exon 2 was identified in patient 5. However, the mutation occurred in a heterozygous pattern not fulfilling Knudson s theory for the inactivation of tumor suppressor genes (Knudson, 97 #292). In this particular patient, admixture of normal adrenal tissue was unlikely because he presented with a bulky, recurrent, infiltrative, and locally inoperable tumor in our institution and the sample was obtained from this tumor mass. This patient fulfills the criteria for diagnosis of sporadic MEN I syndrome by former operations of a pituitary adenoma and multiglandular hyperparathyroidism. This is the first patient with clinical MEN I disease and adrenal cancer in whom molecular proof of an inactivating menin mutation is reported, but the genesis of this adrenal cancer is prone to be accidental or related to a mechanism only indirectly related to his menin gene defect. mutation was detected in patient 00. In this 32-yr-old

MEN I GENE IN ADRENAL CANCER 447 woman, a single parathyroid adenoma had previously been removed at the age 25 yr. One further parathyroid gland was normal, two were not identified. She also presented a follicular thyroid adenoma and a secondary amenorrhea from age of 6 yr until death. Her father had been operated on a double parathyroid adenoma. This constellation can not be described as typical MEN I disease, but may reflect a different entity of multiglandular endocrine disease. We equally did not detect any mutation in patient 4. This 45- yr-old female presented with Cushing s syndrome caused by bilateral adrenal disease. The patient also suffered from a bicentric prolactinoma. Histopathological examination revealed massive multinodular hyperplasia on the right side and an adrenal cancer of 30 g on the left side. There was no family history. Analysis of mrna expression is a suitable method to monitor transcription of the menin gene (37). In our context, we did not search for the gradual decrease observed by deletion of one allele but only for complete loss of menin message that would be necessary to cause loss of menin function in the absence of any mutation in the coding sequence, as had been confirmed in our samples by prior sequencing. All adrenal cancers maintained expression of menin mrna. Particular care was taken to obtain tumor tissue from the center of the macroscopically identified cancer lesion later histopathologically confirmed. Such samples may be particularly suitable for analysis since by histopathological examination alone it may be virtually impossible to confirm the malignant character of particular areas of adrenal tissue. In presumed absence of contamination by normal adrenal tissue and the well-known monoclonal character of tumorous adrenal lesions (38), the presence of menin mrna in all samples of adrenal cancer examined renders unlikely that transcriptional inactivation of the MEN I gene is responsible for loss of its tumor suppressor function in such tumors. In summary, our data identify for the first time a heterozygous missense mutation of the MEN I gene in a patient with sporadic adrenal cancer. However, the biological consequences of the observed heterozygous mutation and its relation to tumor formation are uncertain. The mutation frequency in our group of 4 sporadic adrenal cancers is 7% by full-length direct DNA sequencing, if this mutation is taken into consideration and 0% if it is excluded. We also present the first case where adrenal cancer occurs in a patient in whom MEN I disease is proven by mutational analysis. The heterozygous pattern of this mutation in the tumor tissue argues in favor of a mechanism of tumorigenesis that is accidental or only indirectly related to the MEN I gene defect. Absence of MEN I gene mutations in adrenal cancers of two additional patients with atypical multiglandular endocrine tumor disease argues in favor of the presence of entities that resemble in the phenotype but not the genotype. Presence of menin mrna expression renders it unlikely that inactivation of menin transcription is responsible for formation of adrenal cancer. The frequent evidence of LOH at q3 in adrenal cancer in absence of MEN I gene defects hints toward a role for a different tumor suppressor gene located in this chromosome band. Acknowledgments We thank Mrs. K. Alemazkour, Mrs. B. Weller, and Mrs. B. Bosilj for excellent technical assistance. References. Lipsett M, Hertz R, Ross G. 963 Clinical and pathophysiologic aspects of adrenocortical adenoma. Am J Med. 35:374 379. 2. Reincke M. 998 Mutations in adrenocortical tumors. Horm Metab Res. 30:447 455. 3. Latronico AC, Chrousos GP. 997 Extensive personal experience: adrenocortical tumors. J Clin Endocrinol Metab. 82:37 324. 4. Bornstein SR, Stratakis CA, Chrousos GP. 999 Adrenocortical tumors: recent advances in basic concepts and clinical management. Ann Intern Med. 30:759 77. 5. Ohgaki H, Kleihues P, Heitz PU. 993 p53 mutations in sporadic adrenocortical tumors. Int J Cancer. 54:408 40. 6. Iida S, Fujii H, Moriwaki K. 997 A somatic mutation of the p2(waf/cip) gene in a human adrenocortical adenoma. Anticancer Res. 7:633 636. 7. Reincke M, Mora P, Beuschlein F, Arlt W, Chrousos GP, Allolio B. 997 Deletion of the adrenocorticotropin receptor gene in human adrenocortical tumors: implications for tumorigenesis. J Clin Endocrinol Metab. 82:3054 3058. 8. Li F, Fraumeni J. 969 Soft-tissue sarcomas, breast cancer, and other neoplasms: A familial syndrome? Ann Int Med. 7:747 752. 9. Beckwith J. 969 Macroglossia, omphalocele, adrenal cytomegaly, gigantism and hyperplastic visceromegalie. Birth Defects. 5:88 96. 0. Clouston WM, Cannell GC, Fryar BG, Searle JW, Martin NI, Mortimer RH. 989 Virilizing adrenal adenoma in an adult with the Beckwith-Wiedemann syndrome: paradoxical response to dexamethasone. Clin Endocrinol (Oxf) 3:467 473.. Kleihues P, Schauble B, zur Hausen A, Esteve J, Ohgaki H. 997 Tumors associated with p53 germline mutations: a synopsis of 9 families. Am J Pathol. 50: 3. 2. Carney JA. 995 Carney complex: the complex of myxomas, spotty pigmentation, endocrine overactivity, and schwannomas. Semin Dermatol. 4:90 98. 3. Skogseid B, Larsson C, Lindgren PG, et al. 992 Clinical and genetic features of adrenocortical lesions in multiple endocrine neoplasia type. J Clin Endocrinol Metab. 75:76 8. 4. Beckers A, Abs R, Willems PJ, et al. 992 Aldosterone-secreting adrenal adenoma as part of multiple endocrine neoplasia type (MEN): loss of heterozygosity for polymorphic chromosome deoxyribonucleic acid markers, including the MEN locus. J Clin Endocrinol Metab. 75:564 570. 5. Skogseid B, Rastad J, Gobl A, et al. 995 Adrenal lesion in multiple endocrine neoplasia type. Surgery. 8:077 082. 6. Burgess JR, Harle RA, Tucker P, et al. 996 Adrenal lesions in a large kindred with multiple endocrine neoplasia type. Arch Surg. 3:699 702. 7. Houdelette P, Chagnon A, Dumotier J, Marthan E. 989 Malignant adrenocortical tumor as a part of Wermer s syndrome. Apropos of a case. J Chir (Paris). 26:385 387. 8. Kjellman M, Roshani L, Teh BT, et al. 999 Genotyping of adrenocortical tumors: very frequent deletions of the MEN locus in q3 and of a -centimorgan region in 2p6. J Clin Endocrinol Metab. 84:730 735. 9. Yano T, Linehan M, Anglard P, et al. 989 Genetic changes in human adrenocortical carcinomas (published erratum appears in J Natl Cancer Inst 989 Aug 6;8:263). J Natl Cancer Inst. 8:58 523. 20. Kjellman M, Kallioniemi OP, Karhu R, et al. 996 Genetic aberrations in adrenocortical tumors detected using comparative genomic hybridization correlate with tumor size and malignancy. Cancer Res. 56:429 4223. 2. Iida A, Blake K, Tunny T, et al. 995 Allelic losses on chromosome band q3 in aldosterone-producing adrenal tumors. Genes Chromosomes Cancer. 2:73 5. 22. Gordon RD, Klemm SA, Tunny TJ, Stowasser M. 994 Genetics of primary aldosteronism. Clin Exp Pharmacol Physiol. 2:95 98. 23. Heppner C, Reincke M, Agarwal SK, et al. 999 MEN gene analysis in sporadic adrenocortical neoplasms. J Clin Endocrinol Metab. 84:26 29. 24. Gortz B, Roth J, Speel EJ, et al. 999 MEN gene mutation analysis of sporadic adrenocortical lesions. Int J Cancer. 80:373 379. 25. Chandrasekharappa SC, Guru SC, Manickam P, et al. 997 Positional cloning of the gene for multiple endocrine neoplasia-type. Science. 276:404 407. 26. Heppner C, Kester MB, Agarwal SK, et al. 997 Somatic mutation of the MEN gene in parathyroid tumours. Nat Genet. 6:375 378. 27. Zhuang Z, Vortmeyer AO, Pack S, et al. 997 Somatic mutations of the MEN tumor suppressor gene in sporadic gastrinomas and insulinomas. Cancer Res. 57:4682 4686. 28. Debelenko LV, Brambilla E, Agarwal SK, et al. 997 Identification of MEN gene mutations in sporadic carcinoid tumors of the lung. Hum Mol Genet. 6:2285 2290. 29. Prezant TR, Levine J, Melmed S. 998 Molecular characterization of the men tumor suppressor gene in sporadic pituitary tumors. J Clin Endocrinol Metab. 83:388 39. Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207

448 SCHULTE ET AL. JCE&M 2000 Vol 85 30. Tanaka C, Kimura T, Yang P, et al. 998 Analysis of loss of heterozygosity on chromosome and infrequent inactivation of the MEN gene in sporadic pituitary adenomas (see comments). J Clin Endocrinol Metab. 83:263 2634. 3. Lubensky IA, Debelenko LV, Zhuang Z, et al. 996 Allelic deletions on chromosome q3 in multiple tumors from individual MEN patients. Cancer Res. 56:5272 5278. 32. Chomczynski P, Sacchi N. 987 A single step method of RNA isolation by acid guadinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 62:56 59. 33. Mutch MG, Dilley WG, Sanjurjo F, et al. 999 Germline mutations in the multiple endocrine neoplasia type gene: evidence for frequent splicing defects (In Process Citation). Hum Mutat. 3:75 85. 34. Sarkar G, Yoon HS, Sommer SS. 992 Dideoxy fingerprinting (dde): a rapid and efficient screen for the presence of mutations. Genomics. 3:44 443. 35. Agarwal SK, Kester MB, Debelenko LV, et al. 997 Germline mutations of the MEN gene in familial multiple endocrine neoplasia type and related states. Hum Mol Genet. 6:69 75. 36. Marx SJ, Agarwal SK, Kester MB, et al. 998 Germline and somatic mutation of the gene for multiple endocrine neoplasia type (MEN). J Intern Med. 243:447 453. 37. Asa SL, Somers K, Ezzat S. 998 The MEN- gene is rarely down-regulated in pituitary adenomas. J Clin Endocrinol Metab. 83:320 322. 38. Beuschlein F, Reincke M, Karl M, et al. 994 Clonal composition of human adrenocortical neoplasms. Cancer Res. 54:4927 4932. Erratum Corrections have been made to two tables in the article Modest hormonal effects of soy isoflavones in postmenopausal women by Alison M. Duncan et al. (The Journal of Clinical Endocrinology & Metabolism 84:3479 3484). Revisions to Tables 2 and 3 were inadvertently omitted from the article. The correct tables appear here. The printer regrets the errors. TABLE 2. Daily energy and macronutrient intake Prestudy a Control b Low-iso b High-iso b Energy (kcal) 772 55.4 799 36.4 755 38.0 783 36.4 Protein (g) 69.0 2.5 c.7.7 d 3.8.8 d 4.4.7 d Carbohydrate (g) 26.6 7.6 c 242.7 5.8 d 232.3 6. d 233. 5.8 d Fat (g) 54.3 3. c 45.9 2. d 45.3 2. d 47.7 2. d Dietary fiber (g) 7.6 0.8 c 3.4 0.6 d 3. 0.7 d 2.9 0.6 d a Lsmean SE, based on one 3-day food record per subject. b Lsmean SE, based on three 3-day food records per subject; these data include contributions from the soy powder. c,d Means in the same row with different superscripts are significantly different (P 0.05). TABLE 3. Plasma hormone concentrations Baseline Control Low-iso High-iso LH (IU/L) 46.2 9. 43.6 0.96 45..00 46.0 0.96 FSH (IU/L) 56.2.3 a 52.6 0.67 b 52.6 0.70 b 54.2 0.67 a,b Estradiol (pmol/l) 35.6 3. a 33.7 2.3 a,b 34.7 2.4 a,b 29.7 2.3 b Estrone (pmol/l) 52.8 2.7 a 46.6.5 a,b 46..5 a,b 44.0.5 b Estrone-sulfate (nmol/l) 3.26 0.5 a 3.03 0.06 a 3.06 0.07 a 2.73 0.06 b Testosterone (nmol/l) 0.94 0.04 a,b 0.85 0.02 a,b 0.90 0.02 a 0.83 0.02 b Androstenedione (nmol/l) 4.78 0.28 4.9 0.4 4.48 0.4 4.2 0.4 DHEA-S (nmol/l) d 2398 232 94 44.8 2037 46.7 857 44.8 2063 (794, 2372) a,b 734 (665, 806) a,c 855 (778, 936) b 708 (640, 779) c SHBG (nmol/l) 36.6 0.90 a 32.3 0.59 b 32.4 0.6 b,c 34. 0.59 c Prolactin ( g/l) 7.30 3.6 6.93 0.39 7.03 0.4 7.8 0.39 Cortisol (nmol/l) 495.0 32.8 432.9 2.3 446.5 2.8 40.4 2.3 Insulin (pmol/l) 32.8 3.4 26.4.44 27..5 25.8.44 Free T 3 (pmol/l) 3.5 0.09 3.39 0.06 3.53 0.06 3.39 0.06 Total T 3 (nmol/l) 0.030 0.0003 0.025 0.0003 0.026 0.0003 0.026 0.0003 Free T 4 (pmol/l) 5.20 0.39 4.95 0.26 4.97 0.26 5.60 0.26 Total T 4 (nmol/l) 93..6 90.9.03 90.7.03 93.6.03 TSH (mu/l) 3.48 0.5 3.25 0.3 3.33 0.4 3.49 0.3 TBG (nmol/l) 605.9 3.6 a 559.3 2.2 b 600.4 2.8 a 554. 2.2 b Values are the lsmean SE. a,b,c Means in the same row with different superscripts are significantly different (P 0.05). d To satisfy the ANOVA assumption of normality, data were log-transformed. Geometric mean (95% confidence interval) are presented below the lsmean SE. Downloaded from https://academic.oup.com/jcem/article-abstract/85//44/2856273 on 28 vember 207