Genotyping of Adrenocortical Tumors: Very Frequent Deletions of the MEN1 Locus in 11q13 and of a 1-Centimorgan Region in 2p16*
MAGNUS KJELLMAN, LEYLA ROSHANI, BIN TEAN TEH, OLLI-PEKKA KALLIONIEMI, ANDERS HÖÖG, STEVEN GRAY, LARS-OVE FARNEBO, MIKAEL HOLST, MARTIN BÄCKDAHL, AND CATHARINA LARSSON
Departments of Surgery (M.K., L .- O.F., M.B.), Molecular Medicine (M.K., L.R., B.T., C.L.), Clinical Neuroscience (S.G.), Woman and Child Health (M.K., M.H.), and Pathology (A.H.) Karolinska Hospital, S-171 76 Stockholm, Sweden; and Laboratory for Cancer Genetics (O .- P.K.), Institute of Medical Technology, University of Tampere and Tampere University Hospital, FIN-33521 Tampere, Finland
ABSTRACT
To identify chromosomal regions that may contain loci for tumor suppressor genes involved in adrenocortical tumor development, a panel of 60 tumors (39 carcinomas and 21 adenomas) were screened for loss of heterozygosity. Although the vast majority of loss of het- erozygosity (LOH) were detected in the carcinomas and involved chromosomes 2, 4, 11, and 18, only few were found in the adenomas. Therefore, 2 loci that harbor the familial cancer syndromes Carney complex in 2p16 and the multiple endocrine neoplasia type 1 gene in 11q13 were further studied in 27 (13 carcinomas and 14 adenomas) of the 60 tumors. Detailed analysis of the 2p16 region mapped a minimal area of overlapping deletions to a 1-centimorgan region, which is separate from the Carney complex locus. LOH for a micro- satellite marker (PYGM), very close to the MEN1 gene, was detected
in all 8 informative carcinomas (100%) and in 2 of 14 adenomas. Of the 27 cases analyzed in detail, 13 cases (11 carcinomas and 2 ade- nomas) showed LOH on chromosome 11 and was therefore selected for MEN1 gene mutation analysis. In 6 cases a common polymorphism (Asp418Asp) was found, but no mutation was detected. In conclusion, our data indicate the existence of tumor suppressor genes at multiple chromosomal locations, whose inactivations are involved in the de- velopment of adrenocortical carcinomas. Loss of genetic material from 2p16 was strongly associated with the malignant phenotype, as it was seen in almost all carcinomas but not in any of the adenomas. LOH in 11q13 also occurred frequently in the carcinomas, but was not associated with a MEN1 mutation, suggesting the involvement of a different tumor suppressor gene on this chromosome. (J Clin Endo- crinol Metab 84: 730-735, 1999)
A DRENOCORTICAL lesions most often occur sporadi- cally, but are also found in conjunction with familial tumor syndromes such as Beckwith-Wiedemann (11p15.5), Li-Fraumeni (TP53 in 17p13), multiple endocrine neoplasia type 1 (MEN1), and Carney complex (CC). MEN1 is an au- tosomal dominant disease characterized by neoplasia of the parathyroid glands, the endocrine pancreas, and the anterior pituitary gland. In addition, the patients may develop ad- renocortical tumors, carcinoids, thyroid nodules, and lipo- mas (1). The involvement of the adrenal gland has been reported in about 40% of MEN1 patients and represent bi- lateral hyperplasia, adenomas and in a few cases carcinomas (2-4). The MEN1 tumor suppressor gene in 11q13 (5) was recently cloned (6, 7) and has since been shown to be so- matically mutated in half of the sporadic parathyroid, pan-
creatic, and pituitary tumors displaying a concomitant loss of one 11q13 allele (1). The CC is a familial multiple neoplasia syndrome inherited in an autosomal dominant manner. Af- fected family members may develop primary pigmented nodular adrenocortical disease, lentiginosis, myxomas, and a variety of endocrine and nonendocrine tumors. The CC locus has been assigned to 2p16, but the gene has not yet been cloned (8).
Studies of genetic alterations in adrenocortical tumors have demonstrated the involvement of multiple chromo- somal regions, which to some extent overlap with the map- ping of the familial forms of these tumors. Thus, both familial and sporadic adrenocortical tumors show loss of heterozy- gosity /allelic imbalance (LOH/AI) for markers in 11p, 11q, and 17p. Comparative genomic hybridization performed on 22 sporadic tumors showed an increased number of sequence copy number aberrations by tumor size in carcinomas, whereas the benign tumors were genetically more stable (9). Moreover, losses were most often found on chromosomes 2, 11, and 17, whereas gains were detected on chromosomes 4 and 5. The results suggest that putative tumor suppressor genes (TSG) and oncogenes are located on these chromo- somes. Other studies have focused on genetic alterations within candidate genes such as TP53, p21, ACTH receptor, CYP-21, and IGF2 (10-16a).
Received August 14, 1998. Received November 9, 1998. Accepted November 12, 1998.
Address all correspondence and requests for reprints to: Dr. Magnus Kjellman, Department of Surgery, Karolinska Hospital P9:03, S-171 76 Stockholm, Sweden. E-mail: kmak@kir.ks.se.
* This work was supported by Grant 2330, 10170, and 102 from the Swedish Medical Research Council, the Cancer Society in Stockholm (Project NO 96:115), the Swedish Cancer Foundation, the Janssen-Cilags Foundation, the Fredrik and Ingrid Thuring’s Foundation, and the Tor- sten and Ragnar Söderberg Memory Foundations. The experiments comply with the current laws of Sweden.
Here we have applied LOH analysis to a panel of adre- nocortical adenomas and carcinomas to identify chromo- somal regions that may be involved in the development and progression of this tumor form. With the aim to get an insight into the importance of the MEN1 gene and CC locus, the 11q13 and 2p16 regions were analyzed in more detail. Fur- thermore, mutation analysis of the MEN1 gene was performed.
Subjects and Methods
Patients and tumor samples
This study includes matched pairs of constitutional and tumor sam- ples from 60 patients with primary sporadic adrenocortical tumors (39 carcinomas and 21 adenomas) operated on between 1973 and 1995. The tumors were classified as malignant or benign based on histopatholog- ical findings (invasion, necrosis, pleomorphism, and number of mitosis), tumor size (median, 7.5; range, 0.5-22.0 cm), and clinical outcome. En- docrine symptoms were present in 38 of the patients; 17 showed signs of Cushing’s disease (11 carcinomas and 6 adenomas), 4 showed viril- ization (carcinomas), 2 showed feminization (carcinomas), and 15 were classified as having Conn’s disease (4 carcinomas and 11 adenomas). Fifty-six of the tumors were screened for LOH/ AI on chromosome 1 to X (the Y chromosome was not analyzed). Four additional carcinomas from patients 2, 6, 11, and 12 were included after the initial screening had been performed and were only analyzed for LOH/AI on chromosomes 2, 4, 11, 14, 17, 18, and 22 (Table 2). Further analysis of 11q13 and 2p16 was performed in 27 tumors (13 carcinomas and 14 adenomas) for which normal and tumor tissue was still available. The MEN1 gene was sub- sequently screened for mutations, using single strand conformation analysis (SSCA), in the 11 carcinomas and in the 2 adenomas that showed chromosome 11 LOH/ AI. Informed consent was obtained from all living patients, and the study was approved by the ethical committee of the Karolinska Hospital.
DNA isolation
Immediately after surgical removal, adrenocortical tissue was frozen in liquid nitrogen and stored at -70 C. High mol wt DNA was extracted from 27 frozen tumor tissues by phenol chloroform extraction and eth- anol precipitation according to standard methods. To ensure represen- tativity of the tumor material, pieces were cut from all specimens for histopathological examination, and the samples that were included in the study contained 70% or more tumor cells. In 33 cases where no frozen tumor tissue was available, tumor DNA was obtained from sections of routinely processed paraffin-embedded tumor tissue as previously de- scribed (17). In short, 5-um sections were deparaffinized with lemonene, and tumor cell areas were dissected and incubated with proteinase K in lysis buffer [500 µg/mL proteinase K, 0.05 M Tris-HCI (pH 7.9), 0.15 M NaCl, 5 mM ethylenediamine tetraacetate, and 1% SDS] at 45 C over- night, followed by phenol-chloroform extraction and ethanol precipi- tation. Constitutional DNA was extracted from peripheral leukocytes or paraffin-embedded normal tissues.
LOH/AI studies
The 36 microsatellite markers used for the initial screening and their chromsomal localizations are detailed in Table 1. Standard PCR reac- tions were carried out on microtiter plates in a final volume of 50 AL containing 40 ng constitutional or tumor DNA, 10 mM Tris base (pH 9), 50 mM KČ1, 1.5 mm MgCl2, 0.1% (vol/vol) Triton X-100, 0.01% (wt/vol) gelatin, 50 pmol of each primer, 1.25 mm of each deoxy (d)-NTP, and 1 U Taq polymerase. Samples were amplified using a hot start, with Taq polymerase added after a 5-min denaturation step at 96 C, followed by 35 two-step cycles with annealing at 55 C for 30 s and denaturation at 94 C for 40 s, and a final elongation step at 72 C for 2 min. To minimize the number of sequencing gels and gel loadings, a procedure developed by Hazan et al. (18) was used. In short, PCR amplifications of different markers from the same genomic DNA sample were pooled and loaded into a single well of a sequencing gel. After migration, the products were transferred onto a nylon filter (Hybond N+) and hybridized with one of
the primers (end labeled with 33P) used in the PCR reaction. The filters were then subjected to autoradiography for 7-10 days.
The following microsatellite markers were used for the additional studies of LOH/AI on chromosomes 2 and 11: D2S319, POMC, D2S391, D2S288, CA5, CA7, D2S123, CA2, and D2S125 on chromosome 2 and D11S909, PYGM, D11S449, and D11S968 on chromosome 11. PCR anal- yses were performed using standard methods. One primer was end labeled with 32P. The thermal cycling conditions were incubation at 94 C for 4 min, followed by 30 step cycles of 94 C for 1 min, 60 C for 1 min, and 72 C for 1 min, and a final extension for 7 min at 72 C. Aliquots of the PCR product were electrophoresed in denaturing 6% polyacryl- amide DNA sequencing gels. Gels were then subjected to autoradiog- raphy for 12-24 h.
The location and order of markers are according to GDB (http:// www.gdb.org). LOH/AI was defined as either a total absence or a reduction by 50% or more of the signal intensity of one of the consti- tutional alleles in the tumor compared to the constitutional DNA that could be detected visually.
SSCA
The nine coding exons of the MEN1 gene were amplified using 15 different fragments of 200-300 bp each, as previously described (7). Fifty nanograms of genomic DNA were amplified by PCR in 50 mM KCI; 10 mM Tris-HCI (pH 9.0); 1.5 mm MgCl2 (Promega Corp., Madison, WI); 0.2 mM dTTP, dCTP, and dGTP; 0.05 mm dATP; and [@-32P]dATP (Amer- sham, Arlington Heights, IL) at 1 mCi/reaction and 2 U Taq DNA polymerase (Promega Corp.) in a final volume of 15 p.L. Thermocycling conditions were 30 cycles of 1 min at 94 C, 1 min at 62 C, followed by one cycle of 5-min extension at 72 C. The PCR products were then electrophoresed in 25% MDE (FMC, Rockland, ME) gels at room tem- perature for 12 h at 6-8 watts. Gels were dried before autoradiography.
Direct DNA sequencing
SSCA-shifted bands were excised from the MDE gel and placed in 50 pL ddH2O at 37 C for 1 h. A 5-pL aliquot of this solution was then amplified in a 50-pL reaction with the following components: 50 mM KCÎ; 10 mM Tris-HCI (pH 9.0); 1.5 mm MgCl2 (Promega Corp.); 0.2 mM dTTP, dCTP, dGTP, and dATP; and 15 U Taq DNA polymerase (Pro- mega Corp.). The purified PCR products were sequenced using cycle sequencing reactions from the ABI PRISM 377 BigDye Primer cycle sequencing kit (Perkin-Elmer, Norwalk, CT) and run on the automated sequencer ABI 377 (PE Applied Biosystems, Foster City, CA).
Results
Genotyping of adrenocortical tumors
In the present study adrenocortical carcinomas and ade- nomas from 60 patients were screened for LOH/AI using microsatellite markers (Table 1). For the acrocentric chro- mosomes, one locus on the q-arm was analyzed, and for most of the nonacrocentric chromosomes, markers on both the p- and q-arms were tested. As several of the DNA samples were isolated from paraffin-embedded tissues, all samples did not amplify for all the markers. Overall LOH/ AI was detected on most chromosomes, and in frequencies that varied between 0-29%. Detailed study of the LOH/AI with regard to the malignant (39 cases) and benign (21 cases) classification re- vealed that the vast majority of losses occurred in the car- cinomas. In the carcinomas the LOH/AI frequency per marker varied from 0-60%, and in the adenomas it varied from 0-14%. Similarly, the individual variation in the num- ber of detectable chromosomal alterations fell within a range of 0-18 for the carcinomas and 0-2 for the adenomas.
Allele losses were detected in 79 of 813 informative in- stances (Table 1). Seventy-four of these occurred in the car- cinomas, and most often involved chromosome arms 2q
| Chromosomal location | Locus | Marker | Total no. of LOH/informative cases (%) | Cancer, no. of LOH/informative cases (%) | Adenoma, no. of LOH/informative cases (%) |
|---|---|---|---|---|---|
| 1pter | D1S243 | AFM214yg7 | 3/23 (13) | 3/9 (33) | 0/14 (0) |
| 1qter | D1S423 | AFM042XE3 | 5/28 (18) | 4/11 (36) | 1/17 (6) |
| 2pter | D2S319 | AFM108XH8 | 2/28 (7) | 2/17 (12) | 0/11 (0) |
| 2qter | D2S125 | AFM112YD4 | 12/41 (29) | 12/24 (50) | 0/17 (0) |
| 3p25-3pter | D3S1307 | AFM238WB12 | 4/27 (15) | 4/14 (29) | 0/13 (0) |
| 3q25.2-3q26.2 | D3S1265 | AFM087YB7 | 1/18 (6) | 0/7 (0) | 1/11 (9) |
| 4p16.3 | D4S227 | C5E | 4/15 (27) | 4/9 (44) | 0/6 (0) |
| 4q35-4qter | D4S426 | AMF238VE3 | 1/26 (4) | 0/13 (0) | 1/13 (8) |
| 5pter | D5S392 | AMF028XB12 | 0/31 (0) | 0/15 (0) | 0/16 (0) |
| 5qter | D5S408 | AMF164XB8 | 0/23 (0) | 0/9 (0) | 0/14 (0) |
| 6p25 | D6S344 | AMF092XB7 | 0/25 (0) | 0/8 (0) | 0/17 (0) |
| 6q25.2-7q27 | D6S305 | AFM242ZG5 | 2/13 (15) | 2/7 (29) | 0/6 (0) |
| 7p15-7pter | D7S531 | AFM254yc9 | 1/19 (5) | 1/11 (9) | 0/8 (0) |
| 7q36-7qter | D7S550 | AMF224XH4 | 1/28 (4) | 1/15 (7) | 0/13 (0) |
| 8p23-8pter | D8S264 | AMF143XD8 | 5/33 (15) | 5/19 (26) | 0/14 (0) |
| 8q24.13-8qter | D8S272 | AMF175XB4 | 1/22 (4) | 1/9 (11) | 0/13 (0) |
| 9p ND | |||||
| 9q34.3 | D9S158 | AMF073YB11 | 0/11 (0) | 0/5 (0) | 0/6 (0) |
| 10pter | D10S249 | AMF207WD12 | 1/26 (4) | 1/16 (6) | 0/10 (0) |
| 10q ND | |||||
| 11p15.5 | D11S922 | AFM217YB10 | 3/11 (27) | 3/5 (60) | 0/6 (0) |
| 11q24.1-11q25 | D11S968 | AFM109XC3 | 9/32 (28) | 9/19 (47) | 0/13 (0) |
| 12p13.2-12pter | D12S352 | AFM303XD9 | 0/21 (0) | 0/10 (0) | 0/11 (0) |
| 12qter | D12S357 | AFM310VD5 | 0/14 (0) | 0/4 (0) | 0/10 (0) |
| 13q34 | D13S285 | AFM309VA9 | 3/35 (9) | 3/16 (19) | 0/19 (0) |
| 14q32.1 | D14S292 | AFM120XG5 | 3/15 (20) | 2/8 (25) | 1/7 (14) |
| 15q26-15qter | D15S120 | AFM164ZC9 | 0/24 (0) | 0/9 (0) | 0/15 (0) |
| 16p13.3 | D16S521 | AFM139WG1 | 0/13 (0) | 0/6 (0) | 0/7 (0) |
| 16q ND | |||||
| 17p13-17pter | D17S926 | AFM207XA11 | 1/17 (6) | 1/9 (11) | 0/8 (0) |
| 17qter | D17S788 | AFM095ZD11 | 1/9 (11) | 1/3 (33) | 0/6 (0) |
| 18p11.22-18pter | D18S59 | AFM178XC3 | 4/17 (24) | 4/7 (57) | 0/10 (0) |
| 18qter | D18S554 | AFM296WD5 | 7/32 (22) | 6/17 (35) | 1/15 (7) |
| 19p ND | |||||
| 19q13.4 | D19S210 | AFM119XH6 | 0/38 (0) | 0/21 (0) | 0/17 (0) |
| 20pter | D20S171 | AFM046XF6 | 0/35 (0) | 0/18 (0) | 0/17 (0) |
| 20q ND | |||||
| 21qter | D21S1259 | AFM326TF1 | 0/6 (0) | 0/1 (0) | 0/5 (0) |
| 22q22p13.3qter | D22S274 | AFM164TH8 | 2/21 (10) | 2/10 (20) | 0/11 (0) |
| Xp22.3 | DXS996 | AFM212XE5 | 3/19 (16) | 3/10 (30) | 0/9 (0) |
| Xq28 | DXS1193 | AFM199WC7 | 0/17 (0) | 0/5 (0) | 0/12 (0) |
ND, not done. Percentages are given in parentheses.
(50%), 4p (44%), 11p (60%), 11q (47%), and 18p (57%; Table 1 and Fig. 1). In addition, 1p, 1q, 3p, 6q, 8p, 14q, 17p, 17q, 18q, 22q, and Xp showed LOH/AI in 20-36% of the informative cases. The five losses identified in the adenomas involved five different chromosome arms, i.e. 1q, 3q, 4q, 14q, and 18q (Table 1). Among the carcinomas, the larger tumors showed an increased number of LOH/AI compared to the smaller tumors. However, the pattern of chromosomes involved did not differ between tumors of different hormonal status.
Twenty-seven tumors (13 carcinomas and 14 adenomas) were further analyzed for LOH/AI within chromosomal re- gions 11q13 and 2p16, where the MEN1 gene and CC locus
are located, respectively. Furthermore, mutation screening of the MEN1 gene was performed in 13 of these cases (11 car- cinomas and 2 adenomas). Table 2 shows the clinical, mor- phological, hormonal, and genetic data for each patient in this subgroup.
LOH/AI in chromosome 11q13 and mutation analysis of the MEN1 gene
The PYGM marker located very close to the MEN1 gene in 11q13 showed LOH/AI in all 8 informative carcinomas (100%) and in 2 of the 13 informative adenomas (13%). How-
D2S125
D2S125
D2S319
D4S227
D4S227
*
CT
CT
CT
CT
CT
D11S922
D11S968
D18S54
D18S54
D11S922
CT
CT
CT
CT
CT
ever, losses were less frequent in the distal parts of chromo- some 11. One of the carcinomas and the 2 adenomas with loss of PYGM (cases 4, 21, and 27) showed retained heterozy- gosity with D11S909 in 11pter (Fig. 2). Similarly, no losses were detected with D11S968 at 11qter in 3 carcinomas and 1 adenoma displaying LOH for PYGM in 11q13 (Fig. 2). SSCA analysis of the 9 coding exons of the MEN1 gene revealed no mutations in the 11 tumors analyzed. In 6 cases, a common polymorphism (GAC to GAT) involving codon 418 in exon 9 was detected. This polymorphism (Asp418Asp) was de- tected in both constitutional and tumor DNA and has been reported to occur in up to 44% of the population (6).
LOH/AI in chromosome 2p16
LOH/AI on chromosome 2 was only detected in the car- cinomas, with increasing frequency for markers within 2p16 (median, 72%; range, 54-88%) compared to D2S319 in 2pter (8%) and D2S125 in 2qter (44%). Twelve of the 13 carcinomas showed LOH/ AI for at least 1 of the 7 microsatellite markers from the 2p16 region typed (Fig. 3). In 7 of these, all infor- mative markers showed loss (tumors 1, 2, 3, 4, 6, 10, and 11). However, in the remaining 5 tumors, LOH/ AI only occurred at some loci, whereas other loci showed retained heterozy- gosity (tumors 7, 8, 9, 12, and 13). Based on the results from tumors 9, 12, and 13, a single minimal region of overlapping deletions could be mapped between D2S391 and D2S288 (Fig. 3). The interval between D2S391 and D2S288 defines a 1-centimorgan (cM) region that is separate from the location of the 6-cM CC region, as well as from the human MSH2, human MSH6, and POMC genes (Fig. 3).
Discussion
In this study we have screened adrenocortical tumors for genetic alterations using microsatellite markers and SSCA analysis. The carcinomas showed a high percentage of LOH/AI at 11q13, but no mutations were detected in the MEN1 gene. LOH at 2p16 was found in almost all of the carcinomas, and a 1-cM target region outside the CC locus was delineated. The approach used in this study have iden- tified some chromosome regions with frequent deletions, indicating that they may harbor TSGs important for the de- velopment and progression of this tumor form. However, this approach does not exclude that other chromosome re- gions may be of importance.
Adrenocortical tumors in MEN1 patients are always seen in conjunction with tumors in the endocrine pancreas (3, 4, 19). LOH at 11q13 in adrenocortical lesions from MEN1 pa- tients has to date only been reported in a rare case of an aldosterone-producing adenoma (20) and in one adrenocor- tical carcinoma (3). In our study of sporadic adrenocortical tumors, two benign tumors, both aldosteronomas, showed LOH at 11q13. This has previously been described in patients with familial hyperaldosteronism type II, in one sporadic aldosteronoma as well as in the MEN1 case mentioned above (20, 21). The involvement of the MEN1 gene in the patho- genesis of sporadic adrenocortical tumors cannot be com- pletely ruled out even though no mutations were found in this study. We employed the commonly used method of SSCA to screen for mutations. For the MEN1 gene this method has a detection rate of approximately 60-80% in family studies and somewhat higher in studies of tumors with loss of one 11q13 allele. Clearly, one allele of the MEN1 gene is frequently lost, and other mechanisms than muta- tions within the coding part of the gene, such as splice mu- tations or hypermethylation of the promoter region, may be of importance. Furthermore, another gene in this chromo- somal region may be involved.
It has not yet been established whether there is a progres- sion of adrenocortical tumors from the benign to the malig- nant phenotype, or if malignant and benign tumors have different origins. It could be speculated, however, that ad- renocortical cell proliferation may be initiated by a mitogen stimulus, which, in turn, facilitates for key genetic alterations to take place. Mutational events in genes that are important to maintain the stability of the genome may then lead to the multiple numerical and structural chromosomal aberrations seen in tumors larger than 4-5 cm, as demonstrated in this study and previously (9). Until mutations in genes have been identified that initiate the development and are involved in the malignant transformation of this tumor form, numerical changes in chromosomes and structural genetic changes at specific chromosome regions that occur in the carcinomas may work as a preoperative tumor marker for malignancy. Fluorescence in situ hybridization and/or PCR methods can be used on fine needle aspiration material obtained preoperatively.
As LOH/AI in 2p16 was found frequently in the carcino- mas but not in the benign tumors, it indicates that the region is important in the malignant progression of adrenocortical tumors. Furthermore, the defined 1-cM interval between
| Clinical data | Morphology | Hormonal data, steroid profile in urine | Genetic data | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Patiend (ID) | Age (yr)/sex | Endocrine symptoms | Clinical outcome | Follow-up (yr) | Size (cm) | Histopathological diagnosis | % Tumor cells | No. of markers with LOH/no. of informative markers | MEN1 mutation | LOH/AI 11q13 | LOH/AI 2p16 | |
| 1(B1A) | 63/F | Virilization | Met, DFD | 0.5 | 20 | Cancer | 66 | Malignant | 60 | No | NI | LOH |
| 2(C1Y) | 39/F | Virilization | No rec, alive | 2.5 | 18 | Cancer | 95 | Malignant | A | No | NI | LOH |
| 3(A1T) | 78/M | Feminization | Met, DFD | 0.3 | 15 | Cancer | 95 | Malignant | 29 | No | NI | LOH |
| 4(A2Q) | 44/M | None | No rec, alive | 7.5 | 15 | Cancer | 95 | Malignant | 58 | No | LOH | LOH |
| 5(B1E) | 30/M | None | Met, DFD | 4.7 | 11 | Cancer | 75 | Undetermined | 13 | No | LOH | Ret |
| 6(C1T) | 63/M | None | No rec, alive | 2.5 | 11 | Cancer | 85 | Malignant | A | No | LOH | LOH |
| 7(A2U) | 54/M | Feminization | Rec, alive | 9.8 | 10 | Cancer | 75 | Malignant | 25 | — | NI | LOH |
| 8(B1D) | 69/F | Virilization | Rec, DFD | 7.0 | 10 | Cancer | 95 | Malignant | 22 | No | NI | LOH |
| 9(A2R) | 47/F | Cushing | Met, DFD | 1.5 | 10 | Cancer | 80 | Malignant | 22 | — | LOH | LOH |
| 10(A2A) | 61/F | None | Rec, DFD | 7.5 | 9 | Cancer | 95 | Malignant | 16 | No | LOH | LOH |
| 11(C1X) | 55/F | Cushing | Rec, DFD | 0.8 | 9 | Cancer | 90 | Malignant | A | No | LOH | LOH |
| 12(C1L) | 33/F | Cushing | No rec, alive | 3.6 | 7,5 | Cancer | 95 | Benign | A | No | LOH | LOH |
| 13(B1F) | 72/F | Cushing | Met, DFD | 0.5 | 7 | Cancer | 70 | Malignant | 14 | No | LOH | LOH |
| 14(A10) | 63/F | None | No rec, alive | 4.1 | 5 | Adenoma | 95 | — | 5 | — | Ret | Ret |
| 15(A2D) | 62/F | Cushing | No rec, alive | 5.3 | 5 | Adenoma | 95 | Benign | 0 | — | Ret | Ret |
| 16(A2B) | 52/M | Cushing | No rec, alive | 5.3 | 4 | Adenoma | 95 | — | 0 | — | Ret | Ret |
| 17(A1A) | 66/F | None | No rec, DFID | 7.1 | 4 | Adenoma | 80 | — | 7 | — | Ret | Ret |
| 18(A1Z) | 63/F | None | No rec, alive | 10.3 | 3,5 | Adenoma | 95 | Benign | 0 | — | Ret | Ret |
| 19(A2A) | 38/F | Cushing | No rec, alive | 5.4 | 2,5 | Adenoma | 95 | Undetermined | 0 | — | Ret | Ret |
| 20(A2C) | 46/F | Conn | No rec, alive | 4.5 | 2.5 | Adenoma | 95 | — | 0 | Ret | Ret | |
| 21(A1L) | 50/F | Conn | No rec, alive | 8.2 | 2 | Adenoma | 95 | — | 4 | No | LOH | Ret |
| 22(A1K) | 47/F | Conn | No rec, alive | 9.8 | 2 | Adenoma | 80 | — | 0 | — | Ret | Ret |
| 23(A2F) | 33/F | Conn | No rec, alive | 4.1 | 1.8 | Adenoma | 95 | — | 3 | — | NI | Ret |
| 24(A1J) | 47/F | Conn | No rec, alive | 10.1 | 1,5 | Adenoma | 90 | — | 0 | — | Ret | Ret |
| 25(A1D) | 38/F | Conn | No rec, alive | 11.0 | 1,2 | Adenoma | 95 | — | 0 | — | Ret | Ret |
| 26(A2H) | 27/F | Conn | No rec, alive | 6.2 | 1 | Adenoma | 95 | Benign | 0 | — | Ret | Ret |
| 27(A2G) | 79/F | Conn | No rec, alive | 4.7 | 0,5 | Adenoma | 90 | — | 2 | No | LOH/Ret | Ret |
F, female; M, male; - , not done; LOH, loss of heterozygosity; Ret, retention; NI, not informative; Met, metastasis before surgery; Rec, recurrence after surgery; DFD, dead from disease; DFID, dead from intercurrent disease; Ø, analyzed for chromosome 2, 4, 6, 11, 14, 17, 18, and 22 only; ID, identification number.
Carcinomas
Adenomas
2 3 4 5 6 8 9 10 11 12 13
21 27
D11S909
IGF2
PYGM
11q13
MENI
S449
O
D11S968
☒
11
LOH/AI
Retained heterozygosity
Uninformative
Carcinomas
1 2 3 4 5 67 8 9 10 11 12 13
D2S319
POMC
2p16
D2S391
1 cM
D2S288
hMSH6
CA5
hMSH2
CA7
D2S123
Carney
CA2
D2S125
2
LOH/AI
Retained heterozygosity
Uninformative
D2S391 and D2S288 separates from the location of the 6-cM CC region as well as from the human MSH2, huamn MSH6, and POMC genes. This would exclude the CC gene as the target for the 2p16 deletions found in this study. However, the CC locus can still be involved in adrenocortical tumors because of earlier published data on Carney tumors indicating that the CC gene may have a dominant, rather than a recessive, tumorigenic function (22). The DNA repair genes located on 2p16 are other candidates, but are not likely to be involved because they are located outside the 1-cM interval, and no frequent finding of microsatellite instability was detected in this study.
In conclusion, gross genetic alterations seen in adrenocor- tical tumors are complex and an increased genetic instability is a feature of the malignant tumors. Further studies on the mechanisms that cause chromosomal instability in this tu- mor form may lead us to genes involved in the malignant transition. Isolation and characterization of the putative TSG in the 2p16 region may provide insights into such maligni- fication of adrenocortical tumors. Furthermore, no mutations in the MEN1 gene was found in this study.
Acknowledgments
The authors wish to thank the Genethon Laboratories (Paris, France) for providing technical equipment as well as Ann Svensson for valuable help with the DNA extraction from frozen tumor tissues.
References
1. Teh B. 1998 Recent advances in multiple endocrine neoplasia type 1. Curr Opin Endocrinol Diabetes. 5:35-39.
2. Ballard H, Frame B, Hartsock R. 1964 Familial multiple endocrine adenoma- peptic ulcer complex. Medicine. 43:494-525.
3. Skogseid B, Larsson C, Lindgren P, et al. 1992 Clinical and genetic features of adrenocortical lesions in multiple endocrine neoplasia type 1. J Clin Endo- crinol Metab. 75:76-81.
4. Burgess J, Harle R, Tucker P, et al. 1996 Adrenal lesions in a large kindred with multiple endocrine neoplasia type 1. Arch Surg. 131:699-702.
5. Larsson C, Skogseid B, Öberg K, Nakamura Y, Nordenskjöld M. 1988 Mul- tiple endocrine neoplasia type 1 gene maps to chromosome 11 and is lost in insulinoma. Nature. 332:85-87.
6. Chandrasekharappa S, Guru S, Manickam P, et al. 1997 Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science. 276:404-407.
7. The European Consortium on MEN. 1 1997 Identification of the multiple endocrine neoplasia type 1 (MEN1) gene. Hum Mol Genet. 6:1177-1183.
8. Stratakis C, Carney J, Lin J, et al. 1996 Carney complex, a familial multiple neoplasia and lentiginosis syndrome. Analysis of 11 kindreds and linkage to the short arm of chromosome 2. J Clin Invest. 97:699-705.
9. Kjellman M, Kallioniemi O-P, Karhu R, et al. 1996 Genetic aberrations in adrenocortical tumors detected using comparative genomic hybridization cor- relate with tumor size and malignancy. Cancer Res. 56:4219-4223.
10. Sameshima Y, Tsunematsu Y, Watanabe S, et al. 1992 Detection of novel germ-line p53 mutations in diverse-cancer-prone families identified by select- ing patients with childhood adrenocortical carcinoma. J Natl Cancer Inst. 84:703-707.
11. Ohgaki H, Kleihues P, Heitz P. 1993 p53 mutations in sporadic adrenocortical tumors. Int J Cancer. 54:408-410.
12. Iida S, Fujii H, Moriwaki K. 1997 A somatic mutation of the p21(Waf1/Cip1) gene in a human adrenocortical adenoma. Anticancer Res. 17:633-636.
13. Reincke M, Mora P, Beuschlein F, Arlt W, Chrousos G, Allolio B. 1997 Deletion of the adrenocorticotropin receptor gene in human adrenocortical tumors: implications for tumorigenesis. J Clin Endocrinol Metab. 82: 3054-3058.
14. Gray S, Kjellman M, Larsson C, Ekström T. 1997 Allelic expression imbalance and preferential splicing of IGF2 in adrenocortical tumors. Biochem Biophys Res Commun. 239:878-883.
15. Beuschlein F, Schultze E, Mora P, et al. 1998 Steroid 21-hydroxylase mutations and 21-hydroxylase messenger ribonuceic acid expression in human adreno- cortical tumors. J Clin Endocrinol Metab. 83:2585-2588.
16. Gicquel C, Raffin-Sanson M, Gaston V, et al. 1997 Structural and functional abnormalities at 11p15 are associated with the malignant phenotype in spo- radic adrenocortical tumors: study on a series of 82 tumors. J Clin Endocrinol Metab. 82:2559-2565.
16a.Kjellman, M., et al. No overrepresentation of congenital adrenal hyperplasia in Swedish patients with adrenocortical tumors. Clin Endocrinol (Oxf). In press.
17. Shibata D. 1994 Extraction of DNA from paraffin-embedded tissue for analysis by polymerase chain reaction: new tricks from an old friend. Hum Pathol. 25:561-563.
18. Hazan J, Dubay C, Pankowiak M, Becuwe N, Weissenbach J. 1992 A genetic linkage map of human chromosome 20 composed entirely of microsatellite markers. Genomics. 12:183-189.
19. Hedeland H, Östberg G, Hökfelt B. 1968 On the prevalence of adrenocortical adenomas in an autopsy material in relation to hypertension and diabetes. Acta Med Scand. 184:211-214.
20. Beckers A, Abs R, Willems P, et al. 1992 Aldosterone-secreting adrenal ad- enoma as part of multiple endocrine neoplasia type 1 (MEN1): loss of het- erozygosity for polymorphic chromosome 11 deoxyribonucleic acid markers, including the MEN1 locus. J Clin Endocrinol Metab. 75:564-570.
21. Iida A, Blake K, Tunny T, et al. 1995 Allelic losses on chromosome band 11q13 in aldosterone-producing adrenal tumors. Gene Chromosome Cancer. 12:73-75.
22. Stratakis CA, Jenkins RB, Pras E, et al. 1996 Cytogenetic and microsatellite alterations in tumors from patients with the syndrome of myxomas, spotty skin pigmentation, and endocrine overactivity. J Clin Endocrinol Metab. 81:3607- 3614.