MEN1 GENE MUTATION ANALYSIS OF SPORADIC ADRENOCORTICAL LESIONS
Birgit GÖRTZ1, Jürgen ROTH2, Ernst J.M. SPEEL1, Akiko KRÄHENMANN1, Ronald R. DE KRIJGER3, Xavier MATIAS-GUIU4, Seraina MULETTA-FEURER1, Katrin RÜTMANN1, Parvin SAREMASLANI1, Philipp U. HEITZ1 and Paul KOMMINOTH1,2*
1 Department of Pathology, University of Zurich, Zurich, Switzerland
2 Division of Cell and Molecular Pathology, University of Zurich, Zurich, Switzerland
3Department of Pathology, Erasmus University of Rotterdam, Rotterdam, The Netherlands
4Department of Pathology, Hospital Santa Creu i Sant Pau, Barcelona, Spain
To clarify the role of the MEN1 gene in the tumorigenesis of sporadic adrenocortical tumors, we performed a molecular study on 35 adrenocortical lesions including 6 hyperplasias, 19 adenomas and 10 carcinomas. Loss of heterozygosity (LOH) of the MEN1 gene was assessed by PCR using an intragenic (D11S4946) and 2 flanking microsatellite markers (D11S4936, PYGM) and/or fluorescence in situ hybridization (FISH) with a 40-kb cosmid probe containing the MEN1 gene. The com- plete coding sequence of the MEN1 gene was screened for mutations using non-radioactive, PCR-based single-strand conformation polymorphism (SSCP) analysis and MDE hetero- duplex gel electrophoresis. PCR-LOH and FISH analyses performed in 29 tumors (PCR-LOH in 4, FISH in 17 and both in 8 tumors) revealed allelic deletion of the MEN1 locus in 8 (27.5%) and at 11q13 in 9 (31%) tumors. Furthermore, the frequency of LOH at 11q13 was significantly higher in adreno- cortical carcinomas (60%) than in benign lesions (11%). Muta- tion analysis of tumor samples revealed 9 polymorphisms in 7 tumors (S145S, R171Q, R171Q together with L432L) but no mutations, with the exception of one adrenocortical ad- enoma. The latter tumor contained a somatic E109X stop codon mutation in exon 2 and a 5178-9G-A splice mutation in intron 4, which was also detectable in various non- tumorous tissues and blood indicative of a germ-line muta- tion. The patient, who had no clinical signs or family history of MEN1, later also developed a neuroendocrine carcinoma (atypical carcinoid) of the lung. Our findings indicate that inactivating mutations of the MEN1 tumor-suppressor gene appear not to play a prominent role in the development of sporadic hyperplastic or neoplastic lesions of the adrenal cortex and that the newly reported 5178-9G-+A splice muta- tion in intron 4 might cause a variant of the MEN 1 pheno- type. Int. J. Cancer 80:373-379, 1999.
@ 1999 Wiley-Liss, Inc.
Benign adrenocortical adenomas (ACAs) are a frequent finding in individuals over 50 years of age, with a prevalance of 0.5-2.0%. Most of these tumors are detected incidentally by ultrasound or computer tomography (incidentalomas). In contrast, adrenocortical carcinomas (ACCs) are rare and highly malignant tumors, with an incidence of 1:1.7 million per year and a mortality rate of 70-92% in adults. Histological differentiation of benign from malignant adrenocortical lesions is often difficult by conventional histology, especially in small, well-differentiated tumors. It can be achieved using a combination of clinical and histo-pathological parameters (Schröder et al., 1995) as well as immuno-histochemical assess- ment of tumor proliferation fraction (MiB-1) and p53 expression. Nonetheless, it remains difficult to predict the biological potential of some tumors. Thus, improved understanding of tumor initiation and progression would be useful to better classify and predict the biological behavior of adrenocortical neoplasms.
Currently, our understanding of the molecular pathogenesis of adrenocortical tumors is poor, with the exception of some rare inherited cancer syndromes in which affected patients also develop adrenal lesions and molecular defects have been identified, e.g., Beckwith-Wiedemann, McCune-Albright and Li-Fraumeni syn- dromes. Patients with multiple endocrine neoplasia type 1 (MEN1) syndrome also develop adrenocortical lesions, with a frequency of up to 40% (Komminoth et al., 1998). The vast majority of these lesions are endocrinologically silent and represent macronodular cortical hyperplasia as well as adenomas. Functionally active
adenomas (mostly with hypercortisolism or hyperaldosteronism) or carcinomas are rare. Besides adrenocortical lesions, MEN1 patients may suffer from tumors of parathyroid glands, anterior pituitary, endocrine pancreas and endocrine duodenum. Less frequently, neuroendocrine tumors of the lung, the thymus and the stomach as well as some non-endocrine tumors, e.g., lipomas, angiofibromas and ependymomas, may also be found (Komminoth et al., 1998).
The susceptibility gene for MEN1 has previously been localized on chromosome 11q13 (Larsson et al., 1988) and cloned (Chan- drasekharappa et al., 1997). A tumor-suppressor function for the MENI gene has been suggested based on frequent chromosome 11q13 loss of hetereozygocity (LOH) in neoplasms of affected MEN1 patients (Larsson et al., 1988). Allelic deletions (Jakobovitz et al., 1996) and mutation analysis studies (Debelenko et al., 1997; Heppner et al., 1997; Zhuang et al., 1997) have implicated the MENI gene as a tumor suppressor in a significant fraction of the sporadic counterparts of typical MEN1 neoplasms, including parathyroid tumors, insulinomas, gastrinomas and neuroendocrine (carcinoid) tumors of the lung. However, a systematic examination of the role of the MENI gene in sporadic adrenocortical lesions is missing.
A potential role of the MEN1 gene in adrenocortical tumor formation was suggested by previous studies on 11q13 LOH in MEN1-associated adrenocortical tumors (Beckers et al., 1992; Skogseid et al., 1992), and allelic deletions of the MEN1 gene locus have also been reported in sporadic adrenocortical tumors. For example, Gordon et al. (1996) found LOH at 11q13 in 7 of 33 (21.2%) and Iida et al. (1995) in 5 of 11 (45.4%) informative aldosterone-producing adrenocortical tumors and concluded that one of the genes associated with the development of these tumors may coincide with the MENI locus.
Thus, it was tempting to determine whether the development of sporadic adrenocortical hyperplastic and neoplastic lesions is associated with somatic inactivation of one or both alleles of the MEN1 gene. We therefore examined 35 samples of patients with adrenocortical hyperplasia (n = 6), adenomas (n = 19) and carcinomas (n = 10) for mutations in the MENI gene and allelic deletion at the 11q13 locus.
MATERIAL AND METHODS
Patient samples
Frozen tissues of a total of 35 patients with hyperplastic and neoplastic adrenocortical lesions were obtained from the files of the Department of Pathology of the University of Zurich, Erasmus University (Rotterdam) and the Autonomous University of Barce- lona, Hospital Santa Creu i Sant Pau. Samples included 6 adreno- cortical hyperplasias, 19 adenomas, 9 carcinomas and 1 vertebral metastasis of an ACC (Table I).
*Correspondence to: Division of Cell and Molecular Pathology, Depart- ment of Pathology, University of Zurich, Schmelzbergstrasse 12, CH-8091 Zurich, Switzerland. Fax: (41)1-255- 4551.
E-mail: paul.komminoth@pty.usz.ch
| Adrenocortical lesions | Symptoms |
|---|---|
| Hyperplasia = 6 | Cushing = 5 |
| Inactive = 1 | |
| Adenoma = 19 | Cushing = 8 Conn = 6 |
| Inactive = 5 | |
| Carcinoma = 10 | Cushing = 3 Inactive = 7 |
| Total 35 |
Hematoxylin-eosin-stained sections of formaldehyde-fixed, par- affin-embedded samples of each tumor were evaluated and classi- fied using previously published criteria (Schröder et al., 1995).
The average age of patients with adrenocortical hyperplasia, ACAs and ACCs was 56 (35 to 69), 48 (9 to 84) and 43.8 (0.3 to 72) years, respectively, and the female:male ratio 6:0, 12:7 and 5:5, respectively. The average diameter of ACAs and ACCs was 4.6 and 8.1 cm, respectively.
Of the 19 patients with ACA, 8 suffered from Cushing’s syndrome, 6 suffered from Conn’s syndrome and 5 exhibited functionally inactive neoplasms. Of the 10 patients with ACC, 3 had a history of Cushing’s syndrome and 7 demonstrated function- ally inactive tumors. Five of the 6 patients with adrenal hyperplasia clinically presented with Cushing’s syndrome and 1 patient had no endocrine symptoms (Table I).
DNA extraction
Genomic DNA was isolated from frozen tumors using the D-5000 Puregene DNA Isolation Kit (Gentra Systems, Minneapo- lis, MN). Approximately 2 mm3 of tumor material were homog- enized, and DNA extraction was performed according to the manufacturer’s recommendations. DNA from peripheral blood leukocytes of healthy persons served as negative controls. DNA available from our studies on MEN1 patients carrying MENI point mutations served as positive controls.
Non-isotopic PCR-SSCP and heteroduplex gel electrophoresis
Intronic oligonucleotide primer sets (Table II) were designed to amplify the translated exons 2 to 10 and were synthesized by the phosphoramidite method using a 392 DNA synthesizer (Applied Biosystems, Foster City CA). PCR fragments ranging from 244 to 409 bp in length were amplified from 100 ng tumor or germ-line DNA. PCR amplification of exons 3 to 9 was performed in a total amount of 50 ul reaction mixture containing 0.2 mmol dATP, dTTP, dGTP, dCTP; 50 pmol each sense and antisense primer; 1.5 mM Mg2+; 10 mM Tris-HC1; 50 mM KCI; and 1 U Taq DNA polymerase (AmpliTaq Gold; Perkin-Elmer, Norwalk, CT). For PCR amplification of exons 2 and 10, the amplification mixture contained 1× Expand High Fidelity PCR buffer (Boehringer- Mannheim, Mannheim, Germany); 0.2 mmol dATP, dTTP, dGTP, dCTP; 50 pmol each sense and anti-sense primer; 10% DMSO; and 1.75 U Expand High Fidelity DNA Polymerase (Boehringer- Mannheim).
PCR amplifications were carried out using GeneAmp thin- walled reaction tubes (Perkin-Elmer) and a programmable thermal cycler (DNA thermal cycler 9600, Perkin-Elmer). After initial denaturation at 94°C for 420 sec (AmpliTaq Gold) or 120 sec (Expand High Fidelity), 35 cycles of denaturation for 30 sec at 94℃, annealing for 45 sec at 56° to 67°℃ (Table II) and extension for 45 sec (AmpliTaq Gold) or 60 sec (Expand High Fidelity) at 72℃ was used, followed by a final extension at 72℃ for 300 sec.
For the SSCP analysis, 10 ul of PCR products were diluted 1:1 in stop buffer (95% formamide, 20 mM EDTA, 0.05% xylene cyanol, 0.05% bromophenol blue), heat-denatured at 94℃ for 10 min and quickly chilled in liquid nitrogen before loading onto non- denaturing, 0.8-mm-thick 6% polyacrylamide gels (29:1 acrylamide:
bisacrylamide; BioRad, Glattbrugg, Switzerland) containing 5% glycerol (Komminoth et al., 1996). Electrophoresis was carried out using a sequencing gel electrophoresis apparatus (GIBCO BRL, Zurich, Switzerland) at 8 W for 16 hr at room temperature.
For the heteroduplex method, PCR products of tumors were admixed with those of normal control DNA (1:1), denatured at 94℃ for 10 min and then slowly cooled to room temperature (30 min) to allow heteroduplex formation. Ten microliters of PCR products were diluted 1:1 in stop buffer and directly loaded onto a non-denaturing polyacrylamide-derived gel matrix (MDE Gel Solution; FMC BioProducts, Rockland, ME) containing 15% urea. Electrophoresis was carried out at 500 to 700 V (depending on the size of the PCR products) for 16 hr at room temperature.
After electrophoresis, DNA was visualized by silver staining as described (Komminoth et al., 1996). In brief, gels were fixed for 10 min in 10% ethanol, treated for 3 min with 1% nitric acid, washed twice for 2 min in double-distilled water and then incubated for 20 to 30 min in a 0.2% AgNO3 solution (Merck, Zurich, Switzerland). After a brief wash in double-distilled water, gels were developed in a 0.28 M Na2CO3 (Merck) solution containing 0.02% formalde- hyde, rinsed for 2 min in 10% acetic acid, followed by a wash for 5 min in 50 mM EDTA and two washes for 5 min each in double-distilled water.
Non-isotopic direct sequencing of PCR products
Abnormal bands from PCR-SSCP analysis were excised from additionally prepared SSCP polyacrylamide gels, stained with Sybr Green I nucleic acid (Molecular Probes, Eugene, OR), placed in 100 ul of 1X Tris EDTA buffer (pH 8.0) and incubated for 120 min at 95°℃ to elute DNA. An aliquot (3 to 5 ul) of the supernatant was used as PCR template for 35 further PCR cycles as detailed above to yield PCR products predominantly harboring the mutated appropriate MENI sequence; 40 ul of re-amplified DNA sequences and PCR products showing heteroduplex formation in the electro- phoresis assay were agarose gel-purified using the QIAquick Gel Extraction Kit (Qiagen, Basle, Switzerland), alcohol-precipitated after adding 20 µg of glycogen (Boehringer-Mannheim) and resuspended in 12 ul of 10 mM Tris buffer (pH 8.0). The DNA concentration of purified PCR products was estimated by compar- ing the band intensities of 2 ul sample DNA and the quantified DNA m.w. marker pUCBM21/ Hpa II, Dra I, Hind III (Boehringer Mannheim) in an ethidium bromide-stained agarose gel electropho- resis.
DNA sequences of 30 ng PCR products were determined in sense and anti-sense directions by fluorescence-based dideoxy terminator cycle sequencing using the TaqDyeDeoxyª Terminator Cycle Sequencing kit (Applied Biosystems, Weiterstadt, Germany) followed by gel electrophoresis, data collection and analysis on an automated DNA sequencer (model 373A, Applied Biosystems).
MEN1 allelic deletion analysis
FISH analysis was performed using the fluorescein-labeled cosmid clone c10B11 (40 kb) containing the MENI gene as a probe. Touch preparations from frozen tumor material were used for FISH analysis (Speel et al., 1997). Touch preparations were fixed in 70% ethanol for 1 hr at room temperature (RT); dehyrated in an ethanol series of 70%, 96% and 100%; air-dried; and stored at -80℃ or directly used for FISH. Slides were next treated with 100 ug/ml pepsin (Sigma, St. Louis, MO) in 0.01 N HCI for 20 min at 37℃ and post-fixed for 10 min in 1% formaldehyde in PBS for 10 min at RT (Speel et al., 1997). The cosmid probe was labeled with fluorescein (spectrum green)-dUTP (Vysis, Downers Grove, IL) by nick translation (Boehringer-Mannheim) and precipitated in etha- nol in the presence of 50X herring sperm DNA, 50X yeast tRNA and 50X Cot-1 fraction of human DNA. The DNA pellet was resuspended in the hybridization buffer (50% formamide, 10% dextran sulfate, 2 × SSC, 20 ng/10 ul rhodamine-labeled alpha repetitive DNA specific for chromosome 11) with a final concentra- tion of 250 ng cosmid probe/10 ul. Slides were denatured in 70%
| Primer | Position | Orientation | Sequence | Length (bp) | Temp. (℃) |
|---|---|---|---|---|---|
| M2newF | 2193-2211 | Sense | 5'-CCTTAGCGGACCCTGGGA-3' | 365 | 63 |
| M2newR | 2575-2537 | Anti-sense | 5'-CTTAGCGGACCCTGGGAGGA-3' | ||
| M2NNF | 2436-2455 | Sense | 5'-TCAACCGCGTCATCCCTACC-3' | 405 | 60 |
| M2NR | 2840-2821 | Anti-sense | 5'-CACCTGCCGAACCTCACAAG-3' | ||
| M3NF | 4265-4284 | Sense | 5'-GGACCCTCTTTCATTACCTC-3' | 357 | 57 |
| M3NR | 4631-4612 | Anti-sense | 5'-TCTGTCTTCCCTTCCTATGT-3' | ||
| M4NF | 4639-4658 | Sense | 5'-CTTTTCCTGGCTGTCATTCC-3' | 244 | 62 |
| M4NR | 4900-4883 | Anti-sense | 5'-TCCCACAGCAAGTCAAGTCT-3' | ||
| M5/6NF | 5125-5145 | Sense | 5'-ACCCGTTCTCCTCCCTGTTC-3' | 379 | 65 |
| M5/6NR | 5504-5485 | Anti-sense | 5'-GAGACTGGATGGGCGATACC-3' | ||
| M7NF | 5891-5912 | Sense | 5'-GGTGGGAGTGGAGATGGAGAGG-3' | 340 | 67 |
| M7R | 6233-6212 | Anti-sense | 5'-GGACGAGGGTGGTTGGAAACTG-3' | ||
| M8NF | 6581-6604 | Sense | 5'-AGACCCCTTCAGACCCTACAGAG-3' | 273 | 67 |
| M8R | 6855-6834 | Anti-sense | 5'-CCATCCCTAATCCCGTACATGC-3' | ||
| M9F | 7142-7164 | Sense | 5'-CTGCTAAGGGGTGAGTAAGAGAC-3' | 288 | 63 |
| M9NR | 7430-7411 | Anti-sense | 5'-AAAAGTCTGACAAGCCCGTG-3' | ||
| M10NF | 7546-7565 | Sense | 5'-TTGCTCTCACCTTGCTCTCC-3' | 409 | 56 |
| M10NNR | 7955-7938 | Anti-sense | 5'-CTTGATGGCGCTCGAGTT-3' | ||
| M10NNF | 7896-7915 | Sense | 5'-AAGATGAAGGGCATGAAGGA-3' | 332 | 56 |
| M10NR | 8228-8208 | Anti-sense | 5'-GCTGGAGAAAATCGTGGGTT-3' |
formamide/2 × SSC at 72℃ for 2 min; incubated in a cold (-20℃) ethanol series of 70%, 96% and 100% for 2 min each; and then air-dried. Probes were denatured at 75℃ for 10 min and then incubated for 30 min at 37℃ for pre-annealing. To each slide 10 ul of probe mixture were applied; slides were incubated overnight in a humified chamber at 37℃. Post-hybridization washes were per- formed at 45℃ in 50% formamide/2 × SSC (3 times for 5 min) and 0.1 × SSC (3 times for 5 min). Detection of the hybridized cosmid probe was performed using rabbit anti-fluorescein (DAKO, Glos- trup, Denmark) and swine anti-rabbit Ig fluorescein (each for 30 min at 37℃), followed by washing in PBS/ 0.05% Tween 20 and PBS solution at RT (2 times for 5 min). Slides were mounted in Vectashield (Vector, Burlingame, CA) containing 0.5 µg/ml 4’,6- diamidino-2-phenylindole-antifade (DAPI, Sigma) for nuclear coun- terstaining.
Hybridization signals were scored using a Vysis Quips Genetic Workstation (IG Instrumenten-Gesellschaft, Bern, Switzerland). At least 100 interphases with strong hybridization signals were scored for each tumor. The presence of cells with only one MEN1 cosmid signal at a frequency of more than 30% of the cases was interpreted as LOH. Normal frozen adrenal or connective tissue in the vicinity of tumors served as internal control. These controls exhibited nuclei with one MEN1 signal at a frequency of 3% to 7 %.
LOH analysis
In 12 patients from whom DNA of non-tumorous tissue or blood was available, normal and tumor DNA were screened for LOH using 3 polymorphic markers on 11q13 and visualized by silver staining as described above (Komminoth et al., 1996). The markers included an intragenic microsatellite marker (D11S4946) at the MEN1 5’ region and 2 flanking 11q13 markers, PYGM and D11S4936. A 2-fold difference in relative allele intensity ratios between tumor DNA and normal DNA was scored as LOH.
RESULTS
Representative results are shown in Figures 1 and 2 and the molecular and clinical findings are summarized in Table III.
Of the 12 patients in whom non-tumorous tissue or blood was available for PCR microsatellite-based LOH analysis, 4 (33%) demonstrated LOH at the MENI locus (Fig. 2, Table III) and 2 additionally at the PYGM and 1 at the D11S4936 locus. One further tumor exhibited an isolated LOH at the PYGM locus but not the other investigated loci. In 8/12 tumors, PCR-based LOH results could be compared with those obtained with FISH. In 6 cases,
findings were identical, and in the remaining tumors a loss of one of the MEN1 alleles was observed only in 20% of tumor cells. In 17/35 tumors, only FISH could be performed (due to lack of non-tumorous tissue). A total of 4 tumors demonstrated a loss of the MENI locus on one chromosome in more than 30% of cells, and one of these cases also demonstrated loss of one of the chromo- some 11-specific signals in 71% of cells, indicating loss of major parts of one of the chromosomes 11.
Combined results of PCR-LOH and FISH analyses performed in 29 tumors (PCR-LOH in 4, FISH in 17 and both in 8 tumors) revealed allelic deletion of the MENI locus in 8 (27.5%) and LOH at 11q13 in 9 (31%) tumors.
When comparing the LOH status at 11q13 (including the tumors with isolated LOH at the PYGM locus) and clinical data, LOH was predominantly found in clinically malignant tumors (6/10, 60%) but rarely in benign lesions (1/14, 7.1% adenomas and 1/5, 20% hyperplasias).
PCR-SSCP and MDE gel analyses demonstrated band shifts of exon 2 in 5, exon 3 in 3, exon 5 in 1 and exon 9 in 2 tumors. Sequence analysis of these samples revealed 9 polymorphisms in 7 tumors (S145S in 4 tumors, 11.4%; R171Q in 1 tumor, 2.8%; and R171Q together with L432L in 2 tumors, 5.7%). Interestingly, one patient with a clinically sporadic ACA (A22; Fig. 1a,b) exhibited a somatic E109X stop codon mutation in exon 2 (which was not detectable in non-tumorous tissue; Fig. 1f) and a 5178-9G-A splice mutation in intron 4 (Fig. 1e). This mutation induces a new splice acceptor site with a frameshift resulting in a truncated protein. Two years after removal of the adrenocortical tumor, this patient also developed a neuroendocrine carcinoma (atypical carcinoid) of the dorsobasal right basal lung segment (Fig. 1c,d). Additional molecular analyses of the neuroendocrine lung tumor as well as non-tumorous lymph node and lung tissue, a liver and rectal biopsy and blood exhibited the same 5178-9G-A splice mutation in intron 4 as found in the adrenocortical tumor (Fig. 1e), indicating that the 2 endocrine tumors of this patient were caused by an inheritable germ-line mutation in the MENI gene and that the patient therefore qualifies as a MENI gene carrier. Clinical examination of the patient revealed no familial history of MEN1 or clinical signs of an involvement of parathyroid glands, the endo- crine pancreas or pituitary gland. LOH analysis of the adrenal and bronchial tumor revealed an allelic deletion of the MENI locus in both tumors (Fig. 1g).
No mutations were found in the remaining sporadic adrenocorti- cal lesions tested.
1
a
b
c
d
e
ND
NN
Cd
AA
N1
N2
N3
f
ND
Cd
AA
N1
N2
g
D11S4936
N
Cd D11S4946
AA
CAT AA T CTCT CCT T CAGCT CCT AG A AGCT
AGG AGCOGA AG
CT CG ADAAGGGGG
HLAS exon 2 E109X
CCCCCT TOTCG AG
N
Cd
AA
PYGM
intron 4
5178-9G→A exon 5
reverse
reverse
N
Cd
AA
2
D11S4936
N
T
D11S4946
N
T
PYGM
A55
N
T
| File | Diagnosis | Clinic | Sex | Age (yr) | Polymorphism | Mutation | D11S49361 | D11S49461 | PYGM1 | FISH |
|---|---|---|---|---|---|---|---|---|---|---|
| A 38 | Hyperplasia | Cushing | F | 69 | n | n | n | |||
| A 42 | Hyperplasia | Cushing | F | 69 | n | n | n | |||
| A 43 | Hyperplasia | Cushing | F | 65 | n | n | n | |||
| A 34 | Hyperplasia | Inactive | F | 52 | ||||||
| A 45 | Hyperplasia | Cushing | F | 35 | n | n | n | + | ||
| A 6 | Hyperplasia | Cushing | F | 50 | n | n | n | n | ||
| A 22 | Adenoma | Inactive | M | 43 | E109X, 5178-9G-+ A2 | U | + | (-)3 | ||
| A 49T | Adenoma | Conn | M | 63 | S145S | |||||
| A55T | Adenoma | Inactive | F | / | + | + | + | + ☒ | ||
| A 30 | Adenoma | Inactive | F | 66 | U | n | ||||
| A 35 | Adenoma | Inactive | F | 46 | R171Q, L432L | n | n | n | ||
| A 37 | Adenoma | Cushing | F | 37 | n | n | n | |||
| A 40 | Adenoma | Cushing | F | 84 | R171Q | n | n | n | ||
| A 41 | Adenoma | Cushing | F | 37 | n | n | n | |||
| A 46 | Adenoma | Conn | F | 57 | S145S | n | n | n | ||
| A 47 | Adenoma | Conn | M | 34 | n | n | n | |||
| A 48 | Adenoma | Cushing | M | 9 | n | n | n | n | ||
| A 50 | Adenoma | Conn | F | 42 | S145S | n | n | n | ||
| A 51 | Adenoma | Conn | M | 42 | n | n | n | |||
| A 52 | Adenoma | Conn | M | 61 | n | n | n | |||
| A 53 | Adenoma | Cushing | F | 68 | n | n | n | |||
| A 1 | Adenoma | Cushing | F | 29 | n | n | n | n | ||
| A 3 | Adenoma | Cushing | F | 42 | n | n | n | n | ||
| A 7 | Adenoma | Cushing | M | 45 | n | n | n | n | ||
| A 8 | Adenoma | Inactive | F | 62 | n | n | n | n | ||
| A 12T | Carcinoma | Inactive | F | 44 | U | + | n | |||
| A 15 | Carcinoma | Inactive | M | 39 | U | + | + | (-)3 ☒ | ||
| A 17 | Carcinoma | Inactive | M | 39 | n | |||||
| A 20T | Carcinoma | Cushing | F | 60 | n | |||||
| A 23 | Carcinoma | Inactive | M | 0.3 | R171Q, L432L | + | ||||
| A 33 | Carcinoma | Inactive | F | 33 | ||||||
| A 54T | Carcinoma | Inactive | F | 50 | ||||||
| A 39 | Carcinoma | Cushing | M | 72 | n | n | n | + ☒ | ||
| A 36 | Carcinoma | Inactive | M | 46 | n | n | n | + ☒ | ||
| A 44 | Carcinoma | Cushing | F | 55 | S145S | n | n | n | + |
1D11S4946 intragenic polymorphic marker in 5’ region of the MEN1 gene, PYGM approx. 70 kb centromeric, D11S4936 approx. 100 kb telomeric. n, not tested; U, uninformative .- 2Germ-line mutation .- 3Loss of 1 allele in 20% of cells .— , Retention of constitutional heterozygosity .- +, Loss of heterozygosity.
DISCUSSION
We have assessed the frequency of 11q13 deletions and MEN1 gene mutations in a series of 35 clinically sporadic adrenocortical lesions. Although we could identify one patient with a clinically
FIGURE 1 - Adrenocortical adenoma [(a) macroscopic and (b) microscopic appearance] and neuroendocrine (carcinoid) tumor of the lung [(c) low power and (d) microscopic appearance] of a 43-year-old male patient without family history or clinical signs of MEN1. (e) Molecular analysis revealed band shifts in the SSCP analysis (red arrowheads) and a 5178-9G-A splice mutation in intron 4 (red arrows) in samples of the tumors (Cd, AA) and various normal tissues (N1, N2, N3) indicative of a germ-line mutation (ND, normal control DNA-denatured; NN, non-denatured). (f) A somatic E109X stop codon mutation was found in exon 2 in the DNA of the adrenocortical adenoma (AA) (ND, normal control DNA-denatured; NN, non- denatured). (g) LOH analysis using 11q13 microsatellite markers D11S4936 (telomeric), D11S4946 (intragenic) and PYGM (centro- meric) revealed LOH at the MENI locus in the adrenocortical adenoma (AA) and neuroendocrine (carcinoid) tumor of the lung (Cd) when compared with non-tumorous tissue (N). Results using the D11S4936 marker were uninformative (homozygous alleles), and no LOH at the PYGM locus was found. Scale bar: (a,c) 1 cm, (b,d) 250 um.
FIGURE 2 - Comparison of LOH analysis using microsatellite mark- ers (left) and FISH (right) in a functionally inactive sporadic adrenocor- tical adenoma (A55). Note that LOH is present at all 3 loci (D11S4936, D114946 and PYGM; red arrowheads) in the DNA derived from tumor tissue (T) when compared to non-tumorous tissue (N) and that in the majority of tumor cell nuclei one chromosome 11 (red dots, alpha satellite probe, white arrows) is missing the MENI gene locus (green dots, 40-kb cosmid probe). Scale bar: 5 um.
non-expected variant of MEN1 phenotype, no somatic MEN1 gene mutations could be detected in the remaining sporadic adrenocorti- cal lesions. Thus, our study implicates no obvious function of the MEN1 gene in the development of sporadic adrenocortical tumors.
A role of the MEN1 gene in sporadic adrenocortical tumors was suggested previously by LOH studies on 11q13. However, in these studies, polymorphic markers only flanking the area of the putative MEN1 gene were used and 21% to 45% of the sporadic adrenocor- tical tumors showed LOH at 11q13 (Gordon et al., 1996; Iida et al., 1995). In the present study, we tested for allelic deletions precisely at the location of the MEN1 gene using an intragenic and 2 flanking microsatellite markers as well as FISH. Combined results of PCR-LOH and FISH analyses performed in 29 tumors revealed LOH at the MEN1 locus in 8 (27.5%) and at 11q13 in 9 (31%) tumors. However, no MENI gene mutations were detectable in our series of truly sporadic adrenocortical lesions. There are at least 3 possible reasons why MENI gene mutations were not seen in lesions with loss of one copy of the MEN1 gene. First, our work may in fact under-estimate the prevalence of MENI somatic mutations in sporadic adrenocortical tumors because the sensitivity rate of SSCP for detection of single-base substitutions and small deletions/insertions is estimated to be 80%. However, all tumors were additionally screened with MDE heteroduplex gel electropho- resis to increase the sensitivity of detection and tumors were found to be negative. Furthermore, all positive control samples yielded aberrant band patterns and polymorphisms were easily identified. Second, the tumors may contain a mutation in the promotor region of the MEN1 gene that was not screened. Third, an alternative mechanism, such as hypermethylation of a CpG island, as has been described in other tumor-suppressor genes (Kanai et al., 1997),
may inhibit transcription of the second copy of the MEN1 gene in these tumors. To address this hypothesis, mRNA-expression stud- ies would be required, which were not performed in our study.
Interestingly, one tumor exhibited an isolated LOH at the PYGM locus, which is approximately 70 kb centromeric of the MEN1 gene, indicating that other tumor-suppressor genes on 11q13 might also be involved in the development of adrenocortical lesions. Another interesting finding was the detection of an allelic deletion at 11q13 (without MENI gene mutation) in an adrenocortical hyperplasia. The affected female patient exhibited bilateral (macro- nodular) adrenocortical disease with hypercorticotropism but no family history or clinical signs of multiple endocrine neoplasia. Since only a small fraction of the lesion was available for study, it remains unclear whether the examined sample consisted of an adenomatous nodule with monoclonal growth and allelic loss at 11q13 or whether the whole organ exhibited this genetic alteration. When comparing the LOH status of adrenocortical lesions at 11q13 (including the tumors with isolated LOH at the PYGM locus) and the clinical data, LOH was predominantly found in clinically malignant tumors (6/10, 60%) but rarely in benign lesions (1/14, 7.1% adenomas and 1/5, 20% hyperplasias). This finding may implicate other as yet unidentified tumor-suppressor genes on 11q13 which give rise to more aggressive and potentially metasta- sizing tumors in the formation of adrenocortical lesions. Another possible explanation for the higher incidence of LOH at 11q13 in malignant adrenocortical neoplasms when compared with benign lesions is genetic instability in malignant neoplasms, which may give rise to random loss of chromosomal loci, including 11q13. However, since only one tumor exhibited monosomy and another one more than two copies for the centromere 11 targets in the FISH analysis, the latter explanation appears less likely.
The single patient who demonstrated a somatic E109X stop codon mutation in exon 2 in the tumor DNA from an adrenocortical adenoma additionally developed a neuroendocrine tumor (carci- noid) of the bronchus but exhibited no signs of primary hyperpara- thyroidism or pancreatic or pituitary disease. Despite the lack of additional clinical signs of a MENI phenotype, we consider this patient to harbor a potentially inheritable, special variant of MEN1 since we could demonstrate a 5178-9G-A splice germ-line mutation in intron 4 detectable both in the tumor DNA of the adrenocortical and bronchial neoplasms and in DNA obtained from various non-tumorous tissues (lung, lymph node, liver and rectum) and blood. The 5178-9G-A splice mutation in intron 4 induces a splice acceptor site 7 nucleic acids prior to the splice site in the wild-type DNA and thereby leads to a shift in the reading frame of exon 5, resulting in a trunctated upstream protein. LOH analysis of the adrenal and bronchial tumor revealed an allelic deletion of the MEN1 locus in both tumors. Thus, it appears that somatic inactivation of the other MENI allele was caused by an allelic deletion or, in the case of the adrenal adenoma, by the E109X stop codon mutation in exon 2. We have identified a variant of MEN1 phenotype which is caused by a novel genomic 5178-9G-A splice mutation. Whether or not this special phenotype of MEN1 is entirely attributed to this particular type of mutation remains to be
investigated. In light of our findings, we speculate that the variety of phenotypes caused by germ-line MENI gene mutations is far broader than previously expected and therefore recommend testing of patients with more than one (neuro-)endocrine neoplasm in different organs for MEN1 mutations.
Our study on sporadic adrenocortical lesions did not show mutations of the MENI gene, which is consistent with the theory that other as yet unidentified tumor-suppressor genes/oncogenes or growth factors may contribute to the pathogenesis of these tumors. There are few reports of molecular studies of oncogenes or tumor-suppressor genes in sporadic adrenocortical neoplasms. Although initial reports of Lyons et al. (1990) indicated that oncogenic mutations of G proteins are involved in the formation of adrenocortical tumors, subsequent studies demonstrated that alter- ations of these genes are in fact extremely rare (Yoshimoto et al., 1993) and thus cannot play a major role in the tumorigenesis of adrenocortical neoplasms. Other chromosomal alterations and genetic defects have been described. For example, a breakpoint of 11p13, LOH of alleles on 11p15 and 17p13 as well as over- expression of IGF-II were found in up to 40% of malignant adrenocortical cancers (Gicquel and Le Bouc, 1997), indicating that they could be used as markers for malignancy in these tumors. Furthermore, p53 tumor-suppressor gene mutations have been demonstrated in 20% to 27% of ACCs but rarely in ACAs. However, the results of McNicol et al. (1997) indicate that although p53 appears to be involved in tumor progression, it has no prognostic significance in ACCs. Another gene which has been implicated in the oncogenesis of adrenocortical tumors is the ACTH-receptor gene. Reincke et al. (1997) have demonstrated that LOH of the ACTH-receptor gene is present in a subset of adrenocortical tumors and concluded that this gene might be involved in adrenal tumorigenesis. However, Tsigos et al. (1995), by investigating 17 adenomas, 8 carcinomas and 2 adrenocortical cancer cell lines, concluded that activating mutations of the ACTH-receptor gene do not represent a frequent mechanism of human adrenocortical tumorigenesis. Thus, the role of the ACTH- receptor gene, as well as other occasionally reported genes including the adenomatous polyp coli (APC) gene on chromosome 5q, Carney’s complex gene on chromosome 2p16, the p21 (Waf1/ Cip1) gene on 6p21 and the ZFMI gene on 11q13, needs to be further elucidated.
Considered together, these findings suggest that a complicated interaction of factors and multiple steps is required for adrenocorti- cal tumor formation in most cases.
ACKNOWLEDGEMENTS
We thank Drs. S.C. Chandrasekharappa (NIH, Bethesda, MD), M. Mihatsch and G. Sauter and Ms. M. Kasper (University of Basel, Switzerland) and Ms. C. Eberle (University of Zurich, Switzerland) for providing the c10B11 cosmid probe and frozen tissue samples, respectively; also, Mr. H. Neff and Mr. N. Wey for photographic and computer-assisted reproductions.
REFERENCES
BECKERS, A., ABS, R., WILLEMS, P.J., VAN DER AUWERA, B., KOVACS, K., REZNIK, M. and STEVENAERT, A., Aldosterone-secreting adrenal adenoma as part of multiple endocrine neoplasia type 1 (MEN1): loss of heterozygosity for polymorphic chromosome 11 deoxyribonucleic acid markers, including the MEN1 locus. J. clin. Endocrinol. Metab., 75, 564-570 (1992).
CHANDRASEKHARAPPA, S.C. and 22 OTHERS, Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science, 276, 404-406 (1997).
DEBELENKO, L.V. and 20 OTHERS, Identification of MEN1 gene mutations in sporadic carcinoid tumors of the lung. Hum. mol. Genet., 6, 2285-2290 (1997).
GICQUEL, C. and LE BOUC, Y., Molecular markers for malignancy in adrenocortical tumors. Horm. Res., 47, 269-272 (1997).
GORDON, R., GARTSIDE, M., TUNNY, T. and STOWASSER, M., Different allelic patterns at chromosome 11q13 in paired aldosterone-producing tumours and blood DNA. Clin. exp. Pharmacol. Physiol., 23, 594-596 (1996).
HEPPNER, C. and 20 OTHERS, Somatic mutation of the MEN1 gene in parathyroid tumours. Nature (Genet.), 16, 375-378 (1997).
JAKOBOVITZ, G., NASS, D., DEMARCO, L., BARBOSA, A., SIMONI, F.B., RECHAVI, G. and FRIEDMAN, E., Carcinoid tumors frequently display genetic abnormalities involving chromosome 11. J. clin. Endocrinol. Metab., 81, 3164-3167 (1996).
KANAI, Y., USHIJIMA, S., HUI, A.M., OCHIAI, A., TSUDA, H., SAKAMOTO, M. and HIROHASHI, S., The E-cadherin gene is silenced by CpG methylation in human hepatocellular carcinomas. Int. J. Cancer, 71, 355-359 (1997).
KOMMINOTH, P., HEITZ, P.U. and KLÖPPEL, G., Multiple endocrine disorders. In: L.S. Stefaneanu, H.S. Sasano and K. Kovacs (eds.), Molecular and cellular endocrine pathology, Chapman and Hall, London (1999). (In press).
KOMMINOTH, P., ROTH, J., MULETTA-FEURER, S., SAREMASLANI, P., SEELEN- TAG, W.K.F. and HEITZ, P.U., RET proto-oncogene point mutations in sporadic neuroendocrine tumors. J. clin. Endocrinol. Metab., 81, 2041- 2046 (1996).
LARSSON, C., SKOGSEID, B., OBERG, K., NAKAMURA, Y. and NORDENSKJÖLD, M., Multiple endocrine neoplasia type 1 gene maps to chromosome 11 and is lost in insulinoma. Nature (Lond.), 332, 85-87 (1988).
LYONS, J. and 12 OTHERS, Two G protein oncogenes in human endocrine tumors. Science, 249, 655-659 (1990).
McNICOL, A.M., NOLAN, C.E., STRUTHERS, A.J., FARQUHARSON, M.A., HERMANS, J. and HAAK, H.R., expression of p53 in adrenocortical tumours: clinicopathological correlations. J. Pathol., 181, 146-152 (1997).
REINCKE, M., MORA, P., BEUSCHLEIN, F., ARLT, W., CHROUSOS, G.P. and ALLOLIO, B., Deletion of the adrenocorticotropin receptor gene in human adrenocortical tumors: implications for tumorigenesis. J. clin. Endocrinol. Metab., 82, 3054-3058 (1997).
SCHRÖDER, S., KOMMINOTH, P., PADBERG, B. and HEITZ, P.U., Morphological
typing, evaluation of tumor dignity and prognosis and etiologic classifica- tion of adrenomedullary and adrenocortical neoplasias. Pathologe, 16, 307-314 (1995).
SKOGSEID, B., LARSSON, C., LINDGREN, P.G., KVANTA, E., RASTAD, J., THEODORSSON, E., WIDE, L., WILANDER, E. and OBERG, K., Clinical and genetic features of adrenocortical lesions in multiple endocrine neoplasia type 1. J. clin. Endocrinol. Metab., 75, 76-81 (1992).
SPEEL, E.J.M., RAMAEKERS, F.C.S. and HOPMAN, A.H.N., Sensitive multi- color fluorescence in situ hybridization using catalyzed reporter deposition (CARD) amplification. J. Histochem. Cytochem., 45, 1439-1446 (1997).
TSIGOS, C., ARAI, K., LATRONICO, A.C., DIGEORGE, A.M., RAPAPORT, R. and CHROUSOS, G.P., A novel mutation of the adrenocorticotropin receptor (ACTH-R) gene in a family with the syndrome of isolated glucocorticoid deficiency, but no ACTH-R abnormalities in two families with the triple A syndrome. J. clin. Endocrinol. Metab., 80, 2186-2189 (1995).
YOSHIMOTO, K., IWAHANA, H., FUKUDA, A., SANO, T. and ITAKURA, M., Rare mutations of the Gs alpha subunit gene in human endocrine tumors. Mutation detection by polymerase chain reaction-primer-introduced restric- tion analysis. Cancer, 72, 1386-1393 (1993).
ZHUANG, Z. and 24 OTHERS, Mutations of the MEN1 tumor suppressor gene in pituitary tumors. Cancer Res., 57, 5446-5451 (1997).