Inactivation of the p16 Tumor Suppressor Gene in Adrenocortical ...

0021-972X/99/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 1999 by The Endocrine Society

Vol. 84, No. 8 Printed in U.S.A.

Inactivation of the p16 Tumor Suppressor Gene in Adrenocortical Tumors* CATIA PILON†, MATTEO PISTORELLO†, ALESSANDRO MOSCON, GIUSEPPE ALTAVILLA, UBERTO PAGOTTO, MARCO BOSCARO, AND FRANCESCO FALLO Department of Medical and Surgical Sciences, Division of Endocrinology (C.P., M.P., A.M., U.P., M.B., F.F.), and the Department of Pathology (G.A.), University of Padova, 35128 Padova, Italy ABSTRACT The mechanisms of adrenocortical tumorigenesis are still unknown. Evidence that the majority of adrenocortical tumors are monoclonal in origin suggests that a progressive accumulation of genetic aberrations, due to activation of protooncogenes and/or inactivation of tumor suppressor genes, leads to abnormal cell proliferation through a multistep process. Inactivation of the p16 tumor suppressor gene (p16INK4A), which encodes the cell cycle protein p16, was investigated in a series of 14 adrenocortical tumors. Using 11 polymorphic microsatellite markers spanning the short arm of chromosome 9, we demonstrated that three of seven adrenocortical carcinomas and one

A

DRENAL tumors include hormonally active adenomas or carcinomas producing specific endocrine syndromes and hormonally silent benign or malignant masses, which are often discovered incidentally (incidentalomas) by imaging studies performed for extraadrenal complaints (1). Distinction of benign from malignant forms based on histological findings is difficult, and clinical criteria are used to predict their biological behavior (2, 3). The mechanisms of adrenocortical tumorigenesis are still unknown (4). The evidence that the majority of these tumors are monoclonal in origin (5, 6) suggests that a progressive accumulation of genetic aberrations leads to abnormal cell proliferation through a multistep process (7). Activation of protooncogenes, i.e. overexpression of insulin-like growth factor II (8), and/or inactivation of tumor-suppressor genes, i.e. p53 deletions (9 –11), are involved. Allelic loss on chromosome 9p21 is known to occur in different tumor types (12–16). The p16 tumor-suppressor gene (p16INK4A or CDKN2A/MTS1) is located within the chromosome region 9p21 and encodes p16, an inhibitor of the cyclin-dependent kinases (CDKs) 4 and 6. Inhibition of CDKs, in turn, prevents phosphorylation of retinoblastoma protein and blocks cell cycle progression from G1 to S phase (17). Alterations of p16INK4A lead to deregulation of cell proliferation and tumorigenesis (18 –20). To assess the role of p16INK4A in tumorigenesis, we exReceived February 3, 1999. Revision received April 1, 1999. Accepted April 15, 1999. Address all correspondence and requests for reprints to: Francesco Fallo, M.D., Department of Medical and Surgical Sciences, Division of Endocrinology, Via Ospedale 105, 35128 Padova, Italy. * This work was supported by Grant 667/01/96 from Regione Veneto, Ricerca Sanitaria Finalizzata (Venezia, Italy). † Joint first authors.

of seven adrenocortical adenomas had loss of heterozygosity (LOH) within chromosome 9p21, the region containing p16INK4A. Immunohistochemistry showed the absence of p16 nuclear staining in all adrenocortical tumors with LOH within 9p21, and positive staining in all remaining tumors without LOH. In conclusion, LOH within 9p21 associated with lack of p16 expression occurs in a considerable proportion of adrenocortical malignant tumors, but is rare in adenomas. Inactivation of p16INK4A may contribute to the deregulation of cell proliferation in this neoplastic disease. (J Clin Endocrinol Metab 84: 2776 –2779, 1999)

amined in a group of 14 adrenocortical tumors loss of heterozygosity (LOH) by using 11 polymorphic microsatellite markers spanning the short arm of chromosome 9, and p16 expression by immunohistochemistry. Subjects and Methods Patients and tumors Fourteen patients (10 women and 4 men; median age, 49 yr, range, 31– 68 yr) with adrenal tumors were referred to the Division of Endocrinology of the University of Padova from 1995–1998. Patients underwent clinical, radiological, and hormonal evaluation. They were considered to have functioning tumors on the basis of clinical picture and hormone levels. Seven patients had non ACTH-dependent Cushing’s syndrome, 2 patients had primary aldosteronism, and 5 patients had adrenal masses with no evidence of hormone dysfunction. Diagnoses were based on standard criteria (21, 22), and hormone assays were performed as previously reported (23). In 3 of the patients with nonfunctioning tumors, adrenal masses were detected incidentally and were apparently benign on the basis of radiological (size, ,4 cm), endocrine, and scintigraphic evaluation (23). The criterion for surgery was the willingness of the patients. All 14 patients underwent adrenalectomy, and at histology 7 tumors were classified as adrenocortical adenomas and 7 as adrenocortical carcinomas. Diagnosis of malignancy was given in accordance with the criteria reported by Weiss et al. (24). Characterization of tumors is shown in Table 1. Staging of the disease in patients with adrenal carcinomas was performed according to the Surveillance, Epidemiology, and End Results classification (25). Among these patients, 3 were at stage II, 2 were at stage III, and 2 were at stage IV. Venous blood samples were obtained from the patients, and portions of the adrenal tissues collected at surgery were snap-frozen and stored at 280 C until assayed. To avoid contamination with surrounding tissue, nonneoplastic area were removed macroscopically, and only a central part of each tissue specimen was used. All patients gave informed consent.

LOH analysis High mol wt genomic DNA was extracted from leukocytes and tumor specimens by standard methods (26). After quantification of the DNA

2776

p16 AND ADRENOCORTICAL TUMORS TABLE 1. Characterization of adrenocortical tumors studied Tumor type

Cortisol-producing adenomas Aldosterone-producing adenomas Nonfunctioning adenomas (incidentalomas) Cortisol-producing carcinomas Nonfunctioning carcinomas Total

No.

2 2 3 5 2 14

content by photometric measurement, an undigested aliquot of the samples was electrophoresed on a 0.6% agarose gel to exclude DNA degradation. Each patient’s matched pair of control leukocyte and tumor DNA was PCR amplified using oligonucleotide primers flanking 11 DNA polymorphic simple sequence repeat loci on chromosome 9p (D9S199, D9S157, D9S162, IFNA, D9S126, D9S1749, D9S1748, D9S171, D9S161, D9S165, D9S178). Primer sequences were obtained from the Genome database. PCRs were carried out in 25-mL reaction volumes with 1.5 mmol/L MgCl2; 200 mmol/L each of deoxy (d)-ATP, dGTP, dTTP, and dCTP; 2 pmol of each primer; template DNA; and 1 U Taq DNA polymerase. PCR was carried out for 25 cycles in a Progene Techne PCR apparatus (Cambridge, UK). Each cycle consisted of denaturation at 94 C for 30 s, annealing at 55– 60 C for 30 s, and extension at 72 C for 45 s. PCR products were run adjacently, separated on 10% nondenaturing polyacrylamide gels, and visualized by silver staining. Allele loss was identified by a reduction in band intensity of greater than 80% or the absence of one of the expected PCR products in amplified DNA.

Immunohistochemistry To test immunohistochemical expression of p16, either frozen or paraffin-embedded tissues were used. Fresh-frozen tumor specimens stored at 280 C were embedded in OCT (Tissue-Tek, Miles Laboratories, Elkhart, IN). Cryostat sections were cut at 4 – 6 mm, mounted on lysinecoated glass slides, and fixed for 10 min at 4 C in 3.7% formaldehyde. This was followed by a 5-min rinse in 0.01 mol/L phosphate-bufferedsaline (PBS) at pH 7.3. A peroxidase-antiperoxidase technique modified from that described by Stenberger (27) was used for frozen section immunohistochemistry. Briefly, slides were rinsed for 10 min in 0.05 mol/L PBS at pH 7.4, followed by a 10-min incubation with 6% rabbit serum to decrease nonspecific Ig binding. The cryostat sections were incubated in a humid chamber at 4 C overnight with the primary monoclonal antibody anti-p16 (F-12, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) in a 1:250 dilution or with PBS as a negative control. After a PBS wash, the Elite avidin-biotin-peroxidase complex (ABC) kit (Vector Laboratories, Inc. Burlingame, CA) was used for subsequent steps according to the manufacturer’s instructions. Chromogenic development was accomplished using 3,39-diaminobenzidine tetrahydrochloride (Sigma Chemical Co., St. Louis, MO) at 0.375 mg/dL with 0.03% hydrogen peroxide. Slides were counterstained with hematoxylin and coverslipped. The ABC method of Hsu et al. (28) was used to demonstrate p16 immunoreactivity in paraffin-embedded tumors specimens obtained from the same tissues. The tissues had been routinely fixed in 10% buffered formalin and were paraffin embedded according to standard surgical pathology laboratory practice. Sections were deparaffinized. Endogenous peroxidase was blocked with 0.3% hydrogen peroxide for 30 min. Tissues were incubated overnight at 4 C with the primary antibody anti-p16 (F-12, Santa Cruz Biotechnology, Inc.) diluted 1:250. The Elite ABC kit (Vector Laboratories, Inc. Burlingame, CA) was used for subsequent steps as described above. Slides were counterstained with hematoxylin, dehydrated, and coverslipped. Slides from either frozen or paraffin-embedded tissues were scored using standard light microscopy; only nuclear staining was considered positive reactivity. The extent of staining was expressed as a visual estimate of the percentage of cells staining. Five hundred cells were examined randomly by the same observer from at least 10 fields.

2777

adrenal tumors from 4 of 14 patients (28.5%) exhibited LOH at 1 or more of tested loci. Figure 1 summarizes allelic losses observed in adrenocortical adenomas and carcinomas. Three of the 4 tumors with LOH were cortisol-producing carcinomas and had allelic loss at 4 or more of tested loci. In 2 cases (K1 and K6) allelic losses were located between IFNA and D9S171, where p16INK4A has been precisely mapped. One of the deleted carcinomas (K7) showed LOH on chromosome 9 outside of this region, but was noninformative for 2 polymorphic loci within it (D9S126 and D9S171). The fourth deleted adrenal tumor (A3) was a nonfunctioning adenoma incidentally discovered, exhibiting a single allelic loss (D9S126). Figure 2 shows deletion pattern data for representative adrenocortical tumors. The allelic analysis of chromosome 9p by the 11 microsatellite DNA markers in a representative case of adrenocortical carcinoma (K6), showing an apparent retention of heterozygosity in a region of LOH, is presented in Fig. 3. Immunohistochemistry

All 4 tumors with evidence of LOH showed a nearly complete, i.e. 10% or less of positive cell nuclei, or complete absence of p16 immunostaining (Fig. 4a). The remaining 10 cases without LOH exhibited a high degree, i.e. 70% or more of positive cell nuclei, of p16 immunohistochemical expression. Normal adrenal tissue adjacent to the negative adenoma was positively stained (Fig. 4b). Discussion

The results of this study indicate that three of seven adrenocortical carcinomas (42.8%) had allelic losses on chromosome 9p21. The prevalence of LOH was similar to that found in other neoplasms (12–14), confirming that allelic losses in this region are frequently associated with malignant tumor phenotype. Besides p16INK4A, the p15 tumor suppressor gene (p15INK4B or CDKN2B/MTS2) has been mapped to chromosome 9p21 (20, 29). This gene encodes p15, another member of CDKs inhibitors that may contribute to the transforming growth factor-b-mediated cell cycle arrest (30). De-

Results LOH analysis

Control leukocyte DNA from all 14 patients was informative (heterozygous) at 7 or more of the markers. Overall,

FIG. 1. Pattern of LOH of chromosome 9p in 14 adrenocortical tumors (K1–K7, adrenocortical carcinomas, n 5 7; A1–A7, adrenocortical adenomas, n 5 7). M, Retention of heterozygosity; f, LOH; NI, noninformative.

2778

PILON ET AL.

JCE & M • 1999 Vol 84 • No 8

FIG. 2. LOH analysis of chromosome 9p in six representative adrenocortical tumors. The microsatellite markers are amplified from blood DNA (lanes B) and tumor (lanes T) DNA. a, The study of the D9S126 microsatellite shows retention of heterozygosity in tumor A1 and LOH in tumor A3 and is not informative in case A2. b, The study of the D9S162 microsatellite shows LOH in tumor K1 and retention of heterozygosity in tumor K3 and is noninformative in case K2. A, Adrenocortical adenomas; K, adrenocortical carcinomas.

FIG. 4. a, Adrenocortical carcinoma showing a nearly complete negative p16 nuclear immunostaining (frozen tissue; magnification, 3250); b, adrenocortical adenoma with unreactive nuclei and adjacent normal tissue intensely positive (paraffin-embedded tissue; magnification, 325).

FIG. 3. Allelic analysis of chromosome 9p by the 11 microsatellite DNA markers in an adrenocortical carcinoma showing an apparent retention of heterozygosity (IFNA) in a region of LOH. B, Blood DNA; T, tumor DNA.

letions of 9p21, inactivating both genes, could therefore affect two major cell proliferation control pathways. A common area of retention of heterozygosity between two areas of deletion has been identified in the deletion map of the three adrenocortical carcinomas showing LOH. As suggested by

Cairns et al. (14) for other tumor types, the presence of one or more closely spaced microsatellite markers with apparent retention of heterozygosity, when flanked by markers showing clear LOH, could indicate a homozygous deletion in the region containing p16INK4. The apparent retention could, in fact, depend on low level of amplification of normal alleles (nontumoral) derived from small amounts of nonneoplastic tissue within the tumor. The same event may also be possible for p15INK4, although a more fine mapping is needed. The total lack of p16 nuclear staining at immunohistochemistry in the tumors with LOH is in accordance with the presence of a homozygous deletion of p16INK4. This finding as well as the positive immunoreactivity for p16 in all remaining tumors without LOH confirm the correspondence between histochemical and genetic analysis found in other types of human cancer (31). Lack of p16 staining in the deleted tumors indicates that p16INK4A product is lost, and this is consistent with its inactivation in the tumorigenesis process. Poor staining due to artifacts induced by tissue fixation in paraffin-embedded specimens, leading to ambig-

p16 AND ADRENOCORTICAL TUMORS

uous staining results, has been reported (32, 33). As we also used fresh-frozen tissues, this possibility seems remote. Among the patients with adrenocortical carcinoma, two eventually died from cancer. Both patients had stage IV of disease at diagnosis. One of these two patients had both LOH and lack of p16 immunohistochemical expression. The other two patients with LOH and negative p16 staining had stage II disease at diagnosis and did not show progression after 1 and 3 yr of follow-up, respectively. Therefore, in this small series of patients with adrenocortical carcinomas genetic alterations do not seem to have a prognostic value. At variance with adrenocortical carcinomas, only one of seven adenomas showed LOH on 9p21, which was restricted to a single microsatellite loss (D9S126). In this case, the adjacent microsatellites were noninformative, and it was not possible to better define a larger area of chromosome loss or apparent retention. The single detected loss was associated with negative p16 tissue staining, suggesting that it was sufficient to prevent p16 formation. The single loss, however, could indicate a small deletion not involving p15INK4B, that is able to disrupt only the p16INK4A-regulated cellular mechanism. Interestingly, the adenoma was an incidentally detected adrenal mass without clinical and histological indications of malignancy. The probability for these apparently benign adrenal masses to evolve toward malignancy is under investigation (34). In this respect, we do not know whether finding LOH on chromosome 9 and lack of p16 expression may represent a higher risk of malignant transformation of these tumors. In conclusion, LOH within 9p21 associated with absence of p16 immunohistochemical expression occurs in a considerable proportion of adrenocortical malignant tumors, but is rare in adenomas. Inactivation of p16INK4A may contribute to the deregulation of cell proliferation in this neoplastic disease. A limitation of our study could be the small number of tumors examined. Further investigations regarding the correlation between p16 tissue expression and p16INK4A aberrations and their possible prognostic impact are required. Acknowledgments We are grateful to C. Lanza and A. Dubrovich for the immunohistochemical reactions, and to R. Leorin for photographs.

References 1. Latronico AC, Chrousos GP. 1997 Adrenocortical tumors. J Clin Endocrinol Metab. 82:1317–1324. 2. Flack MR, Chrousos GP. 1996 Neoplasms of the adrenal cortex. In: Holland R, ed. Cancer medicine, 4th ed. New York: Lea and Fibinger; 1563–1570. 3. Barzon L, Fallo F, Sonino N, Daniele O, Boscaro M. 1997 Adrenocortical carcinoma: experience in 45 cases. Oncology. 54:490 – 496. 4. Gicquel C, Bertagna X, Le Bouc Y. 1995 Recent advances in the pathogenesis of adrenocortical tumors. Eur J Endocrinol. 133:133–144. 5. Beuschlein F, Reincke M, Karl M, et al. 1994 Clonal composition of human adrenocortical neoplasm. Cancer Res. 54:4927– 4932. 6. Gicquel C, Leblond-Francillard M, Bertagna X, et al. 1994 Clonal analysis of human adrenocortical carcinomas and secreting adenomas. Clin Endocrinol (Oxf). 40:465– 477.

2779

7. Reincke M. 1998 Mutations in adrenocortical tumors. Horm Metab Res. 30:447– 455. 8. Gicquel C, Xavier B, Schneid H, et al. 1994 Rearrangements at the 11p15 locus and overexpression of insulin-like growth factor-II gene in sporadic adrenocortical tumors. J Clin Endocrinol Metab. 78:1444 –1453. 9. Lin SR, Lee YJ, Tsai JH. 1994 Mutations of the p53 gene in human functional adrenal neoplasms. J Clin Endocrinol Metab. 78:483– 491. 10. Reincke M, Karl M, Travis WH, et al. 1994 p53 mutations in human adrenocortical neoplasms: immunohistochemical and molecular studies. J Clin Endocrinol Metab. 78:790 –794. 11. Perrett CW, Pistorello M, Boscaro M, Fallo F, Clayton RN. 1996 Molecular genetic studies on adrenal tumors. In: New MI, ed. Where phenotype does not match genotype. Frontiers in endocrinology. Rome: Ares-Serono Symposia; vol 16:139 –143. 12. Kishimoto Y, Sugio K, Hung JY, et al. 1995 Allele-specific loss in chromosome 9p loci in preneoplastic lesions accompanying non-small cell lung cancers. J Natl Cancer Inst. 87:1224 –1229. 13. van der Riet P, Nawroz H, Corio R, et al. 1994 Frequent loss on chromosome 9p21–22 early in head and neck cancer progression. Cancer Res. 54:1156 –1158. 14. Cairns P, Polascik TJ, Eby Y, et al. 1995 Frequency of deletion of p16/CDKN2 in primary human tumours. Nat Genet. 11:210 –212. 15. Farrell WE, Simpson DJ, Bicknell JE, Talbot AJ, Bates AS, Clayton RN. 1997 Chromosome 9p deletions in invasive and noninvasive nonfunctional pituitary adenomas: the deleted region involves markers outside of the MTS1 and MTS2 genes. Cancer Res. 57:2703–2709. 16. Tahara H, Smith AP, Gaz RD, Arnold A. 1996 Loss of chromosome arm 9p DNA and analysis of the p16 and p15 cyclin-dependent kinase inhibitor genes in human parathyroid adenomas. J Clin Endocrinol Metab. 81:3663–3667. 17. Strauss M, Lukas J, Bartek J. 1995 Unrestricted cell cycling and cancer. Nat Med. 1:1245–1246. 18. Kamb A, Gruis NA, Weaver-Feldhaus J, et al. 1994 A cell cycle regulator potentially involved in genesis of many tumor types. Science. 264:436 – 440. 19. Hunter T, Pines J. 1994 Cyclin and cancer II: cyclin D and cdk inhibitors come of age. Cell. 79:573–582. 20. Kamb A. 1995 Cell-cycle regulators and cancer. Trends Genet. 11:136 –140. 21. Kaye TB, Crapo L. 1990 The Cushing’s syndrome: an update on diagnostic tests. Ann Interrn Med. 112:434 – 444. 22. Litchfield WR, Dluhy RG. 1995 Primary aldosteronism. Endocrinol Metab Clin North Am. 24:593– 612. 23. Barzon L, Scaroni C, Sonino N, et al. 1998 Incidentally discovered adrenal tumors: endocrine and scintigrafic correlates. J Clin Endocrinol Metab. 83:55– 62. 24. Weiss LM, Medeiros J, Vickery AL. 1989 Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol. 13:202–206. 25. NIH. 1977 Summary staging guide for cancer surveillance. Epidemiology and end result reporting program. Bethesda: NIH; 12–14. 26. Sambrook J, Fritsch EF, Maniatis T. 1989 Molecular cloning: a laboratory manual, 2nd Ed. Cold Spring Harbor: Cold Spring Harbor Laboratory; 9.16 –9.19. 27. Stenberger LA. 1986 Immunocytochemistry, 3rd Ed. New York: Wiley and Sons. 28. Hsu S-M, Raine L, Fanger H. 1981 Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem. 29:577–580. 29. Stone S, Dayananth P, Jiang P, et al. 1995 Genomic structure, expression and mutational analysis of the P15 (MTS2) gene. Oncogene. 11:987–991. 30. Hannon GJ, Beach D. 1994 p15INK4B is a potential effector of TGF-b-induced cell cycle arrest. Science. 371:257–261. 31. Reed AL, Califano J, Cairns P, et al. 1996 High frequency of p16 (CDNK2/ MTS-1/INK4A) inactivation in head and neck squamous cell carcinoma. Cancer Res. 56:3630 –3633. 32. Kratzke RA, Otterson GA, Lincoln CE, et al. 1995 Immunohistochemical analysis of the p16INK4A cyclin-dependent kinase inhibitor in malignant mesothelioma. J Natl Cancer Inst. 87:1870 –1875. 33. Geradts J, Kratzke RA, Niehanx GA, Lincoln CE. 1995 Immunohistochemical detection of the cyclin-dependent kinase inhibitor 2/multiple tumor suppressor gene 1 (cdkn2/mts1) product p16INK4 in archival human solid tumors: correlation with retinoblastoma protein expression. Cancer Res. 55:6006 – 6011. 34. Barzon L, Scaroni C, Sonino N, Fallo F, Paoletta A, Boscaro M. 1999 Risk factors and long-term follow-up of adrenal incidentalomas. J Clin Endocrinol Metab. 84:520 –526.