DNA replication error is frequent in ovarian granulosa cell tumors

Cancer Genetics and Cytogenetics 122 (2000) 55–58

Short communication

DNA replication error is frequent in ovarian granulosa cell tumors Mitsuaki Suzukia,*, Michitaka Ohwadaa, Yasushi Sagaa, Kazunori Ochiaib, Ikuo Satoa a

Department of Obstetrics and Gynecology, Jichi Medical School, 3311 Yakushiji, Minamikawachi, Kawachi, Tochigi 329-0498, Japan b Jikei University School of Medicine, Tokyo, Japan Received 12 February 2000; received in revised form 12 April 2000; accepted 14 April 2000

Abstract

DNA replication errors (RER) have been detected in epithelial ovarian cancers, as well as in other human tumor types. These observations suggest that this genetic defect is present in ovarian granulosa cell tumors, and that a DNA mismatch repair deficiency may be involved in their development and/or progression. We therefore assayed tissue samples from 29 patients with granulosa cell tumors for RER, using polymerase chain reaction (PCR) and 5 microsatellite markers. The RER were observed at greater than or equal to 1 loci in 15 (58%) of 26 informative cases. The incidence of RER was unrelated to the patient’s age or the histologic subtype or clinical stage of the tumors. The RER, however, were observed in 57% (8/14) of the informative patients with stage IA disease. These findings suggest that a DNA mismatch repair deficiency may contribute to the pathogenesis of ovarian granulosa cell tumors, and that this deficiency may be an early event in their development and/or progression. © 2000 Elsevier Science Inc. All rights reserved.

1. Introduction

2. Materials and methods

The ovarian line of differentiation in sex cord-stromal tumors is represented by ovarian granulosa cell tumors. These tumors account for approximately 1.5–3% of all ovarian tumors and 6–10% of malignant ovarian neoplasms [1–3]. Although little is known about the molecular pathogenesis of these neoplasms, recent observations, showing an association between a DNA mismatch repair deficiency and the development of hereditary nonpolyposis colorectal cancer (HNPCC) [4–6] and epithelial ovarian cancer [7–11], suggested to us that this genetic alteration may also occur in ovarian granulosa cell tumors. Mismatch repair deficiencies cause alterations in simple repeat sequences, which are most easily observed as changes in the lengths of microsatellite sequences. These changes are revealed when tumor DNA is compared with normal DNA from the same individual. Using polymerase chain reaction (PCR) and microsatellite markers, mismatch repair deficiencies can be identified as replication errors (RER). We assayed the incidence of RER in patients with granulosa cell tumors. We also evaluated the involvement of a DNA mismatch repair deficiency in the development and progression of these tumors.

2.1. Patients

* Corresponding author. Tel.: ⫹81-285-58-7376; fax: ⫹81-285-44-8505.

The study population consisted of 29 Japanese women who had surgery for ovarian granulosa cell tumors between 1974 and 1997 at the Department of Obstetrics and Gynecology, Jichi Medical School, and the Jikei University School of Medicine. Their median age was 44 years (range 6–75). Laparotomy had been performed on each patient as part of the treatment. Surgical staging was determined according to the classification of the International Federation of Gynecology and Obsetrics (FIGO), and was as follows: stage IA disease (16), stage IB (1), stage IC (9), and stage IIIC (3). Histologically, the tumors were of the adult type in 25 patients, and of the juvenile type in 4 patients. 2.2. Tissue preparation and DNA extraction Formalin-fixed, paraffin-embedded tissue from each resected ovary was cut into 10-␮m sections that were mounted on glass slides for counterstaining with hematoxylin. Using an 18-G disposable needle under a stereoscopic microscope, three or more samples of both abnormal and normal tissue were collected from each patient. DNA was extracted from tissue samples according to the method of Wright et al. [12]. Briefly, each sample was digested overnight at 60⬚C in 500 ␮l of proteinase K (20 mg/mL in 50 mM Tris, pH 8.5, 0.5% Tween 20, 1 mM EDTA). After heating

0165-4608/00/$ – see front matter © 2000 Elsevier Science Inc. All rights reserved. PII: S0165-4608(00)00 2 6 9 - 7

56

M. Suzuki et al. / Cancer Genetics and Cytogenetics 122 (2000) 55–58

Table 1 Loci, map positions, and primer sequences Locus

Chromosomal position

Sequence (5⬘ → 3⬘)

TP53

17p

ACTGCCACTCCTTGCCCCATTC AGGGATACTATTCAGCCCGAGGTG AAACAGGATGCCTGCCTTTA GGACTTTCCACCTATGGGAC ATACTCTGGACCCAGATTGATTAC TAATTCCCAAATGGTTTAGGGGAG CCCCAAGGCTGCACTT AGCTGAGACTACAGGCATTTG CAGAAAATTCTCTCTGGCTA CTCATGTTCCTGGCAAGATT

D2S123

2p

D3S1029

3p

D3S1611

3p

D18S34

18q

at 100⬚C for 10 minutes to inactivate the enzyme and cooling on ice, the samples were centrifuged at 12,000 rpm for 5 minutes to remove any debris. The DNA was refined by extraction with phenol/chloroform. 2.3. DNA RERs Five sets of microsatellite primers (TP53 [13], D2S123 [14], D3S1029 [15], D3S1611 [14], and D18S34 [16]) (Table 1), were utilized for PCR amplification, using the method of Senba et al. [17]. Amplification was for 30 cycles of denaturation (1 minute at 94⬚C), annealing (1 minute at 55⬚C), and extension (2 minutes at 72⬚C). The PCR products were diluted with formamide dye solution (Amersham Life Science, Inc., Cleveland, OH, USA), heated for 3 minutes at 90⬚C, electrophoresed on 8% polyacrylamide gels containing 7 M urea, and transferred to nylon membranes. The membranes were incubated with Fab fragments of antidigoxigenin antibody (Boehringer Mannheim Biochemica, Mannheim, Germany) conjugated to alkaline phosphatase. Following the addition of CSPD solution (Boehringer Mannheim), the membranes were exposed to x-ray film for about 2 hours. Samples in which PCR products were detected at four or more of the five microsatellite loci were considered to be informative. An RER was denoted by the occurrence of band shifts or extra bands in tumor samples compared to matched normal tissue (Fig. 1); any tumor containing a band shift or extra band at one or more microsatellite locus was considered positive for RER.

positive patients per locus did not differ significantly (Table 2). Of the 29 granulosa cell tumor patients included in the study, 26 (90%) were informative (i.e., had PCR products at 4 or 5 loci) and 15 (58%) of these were RER-positive at 1 or more locus. When we analyzed the relationship between RER and the clinicopathological characteristics of the granulosa cell tumor patients, we found that RER-positivity was not affected by patient age, or histologic subtype or FIGO clinical stage of the tumor (Table 3). With reference to survival rate, 2 of the 3 patients with stage IIIC disease died within 15 months, while the third, who was alive after 24 months, was not followed afterward. In contrast, all of the 23 patients with stage IA to stage IC disease are still alive (range 11–265 months, median 58 months); in this group, the survival rate did not differ between the RER-positive (n ⫽ 14) and the RER-negative (n ⫽ 9) patients. 4. Discussion Several genetic alterations common to many types of human tumors have been investigated in ovarian granulosa cell tumors. Although several investigators have reported

2.4. Survival and statistical analysis Using the method of Kaplan and Meier [18], survival was calculated from the time of initial surgery, and differences in survival rate were analyzed by the generalized Wilcoxon and log-rank tests. Differences in RER-positivity between groups were calculated using the Chi-square test. All tests were two-tailed. A level of P ⬍ 0.05 was considered statistically significant. 3. Results While extra bands or band shifts were detected at each of the five microsatellite loci examined, the number of RER-

Fig. 1. The RER in granulosa cell tumors. The presence of extra bands in the tumor sample (T) and thin absence in matched normal tissue of the same patient (N), indicates the presence of an RER.

M. Suzuki et al. / Cancer Genetics and Cytogenetics 122 (2000) 55–58 Table 2 RERs at microsatellite loci in patients with granulosa cell tumors No. of tumors examined No. of informativea tumors (%) No. of tumors with alterations at TP53 D2S123 D3S1029 D3S1611 D18S34 Total no. of RER-positive tumors (%)

29 26 (90) 3 2 5 6 4 15 (58)

a Informative was defined as samples in which PCR products were detected at four or more of the five microsatellite loci.

the overexpression of p53 protein in granulosa cell tumors, mutations of this gene were very rare [19, 20]. The finding of a mutation in exon 9 of the Wilms’ tumor suppressor gene (WT1) in a patient with a juvenile granulosa cell tumor suggested that there may be an association between WT1 mutations and this type of tumor [21]. When Coppes et al. [22] assayed 11 granulosa cell tumors for WT1 mutations; however, mutation at this locus was not detected in this type of tumor. Mutations in the gip 2 oncogene have also been assayed in ovarian granulosa cell tumors. Examination of 10 sex cord-stromal tumors, including 7 granulosa cell tumors, showed that 3 had mutations in the gip 2 oncogene [23]. In contrast, a subsequent study failed to detect gip 2 oncogene mutations in 13 granulosa cell tumors [24]. Therefore, no genetic alterations closely related to the development and/or progression of granulosa cell tumors have been detected. We found that 58% of informative patients with granulosa cell tumors were RER-positive, a rate considerably higher than that in patients with epithelial ovarian cancer (6.4–17%) [7–11]. This observation suggests that a DNA mismatch repair deficiency may be involved in the molecular pathogenesis of ovarian granulosa cell tumors. While we detected no significant differences in clinicopathological characteristics between RER-positive and RER-negative patients, we found that 57% of the RER-positive patients

Table 3 RERs in granulosa cell tumor patients classified by clinicopathological characteristics Characteristic Age (years) ⬍50 ⭓50 FIGO stage IA IB IC IIIC Histologic subtype Adult type Juvenile type NS, not significant.

n

No. of patients with RER at ⭓1 locus (%)

P-value

14 12

7 (50) 8 (67)

NS

14 1 8 3

8 (57) 1 (100) 5 (63) 1 (33)

NS

23 3

13 (57) 2 (67)

NS

57

had stage IA disease. Our findings therefore suggest that a DNA mismatch repair deficiency is an early event in the development and/or progression of granulosa cell tumors. Acknowledgments We are grateful to Mrs. E. Kurokawa, Mrs. M. Ohashi, and Ms. Y. Murata for technical assistance. References [1] Young RH, Scully RE. Ovarian sex cord-stromal tumors. Recent progress. Int J Gynecol Pathol 1982;1:101–23. [2] Young RH, Scully RE. Ovarian sex cord-stromal tumors. Problems in differential diagnosis. Pathol Annu 1988; 23:237–96. [3] Costa MJ, DeRose PB, Roth LM, Brescia RJ, Zaloudek CJ, Cohen C. Immunohistochemical phenotype of ovarian granulosa cell tumors. Absence of epithelial membrane antigen has diagnostic value. Hum Pathol 1994;25:60–66. [4] Aaltonen LA, Peltomäki P, Leach FS, Sistonen P, Pylkkänen L, Mecklin J-P, Järvinen H, Powell SM, Jen J, Hamilton SR, Petersen GM, Kinzler KW, Vogelstein B, de la Chapelle A. Clues to the pathogenesis of familial colorectal cancer. Science 1993;260:812–6. [5] Aaltonen LA, Peltomäki P, Mecklin J-P, Järvinen H, Jass JR, Green JS, Lynch HT, Watson P, Tallqvist G, Juhola M, Sistonen P, Hamilton SR, Kinzler KW, Vogelstein B, de la Chapelle A. Replication errors in benign and malignant tumors from hereditary nonpolyposis colorectal cancer patients. Cancer Res 1994;54:1645–8. [6] Jacoby RF, Marshall DJ, Kailas S, Schlack S, Harms B, Love R. Genetic instability associated with adenoma to carcinoma progression in hereditary nonpolyposis colon cancer. Gastroenterology 1995;109: 73–82. [7] Osborne RJ, Leech V. Polymerase chain reaction allelotyping of human ovarian cancer. Br J Cancer 1994;69:429–38. [8] Fujita M, Enomoto T, Yoshino K, Nomura T, Buzard GS, Inoue M, Okudaira Y. Microsatellite instability and alterations in the hMSH2 gene in human ovarian cancer. Int J Cancer 1995;64:361–6. [9] King BL, Carcangiu M-L, Carter D, Kiechle M, Pfisterer J, Pfleiderer A, Kacinski BM. Microsatellite instability in ovarian neoplasms. Br J Cancer 1995;72:376–82. [10] Tangir J, Loughridge NS, Berkowitz RS, Muto MG, Bell DA, Welch WR, Mok SC. Frequent microsatellite instability in epithelial borderline ovarian tumors. Cancer Res 1996;56:2501–5. [11] Arzimanoglou II, Lallas T, Osborne M, Barber H, Gilbert F. Microsatellite instability differences between familial and sporadic ovarian cancers. Carcinogenesis 1996;17:1799–804. [12] Wright DK, Manos MM (1990): Sample preparation from paraffinembedded tissues. In: Innis MA, Gelfand DH, Shinsky JJ, White TJ, editors. PCR Protocols. A Guide to Methods and Applications. New York: Academic Press, 1990. pp. 153–8. [13] Jones MH, Nakamura Y. Detection of loss of heterozygosity at the human TP53 locus using a dinucleotide repeat polymorphism. Genes Chromosom Cancer 1992;5:89–90. [14] Gyapay G, Morissette J, Vignal A, Dib C, Fizames C, Millasseau P, Marc S, Bemardi G, Lathrop M, Weissenbach J. The 1993–94 Généthon human genetic linkage map. Nature Genet 1994;7:246–339. [15] Jones MH, Yamakawa K, Nakamura Y. Isolation and characterization of 19 dinucleotide repeat polymorphisms on chromosome 3P. Hum Mol Genet 1992;1:131–3. [16] Caduff RF, Johnston CM, Svoboda-Newman SM, Poy EL, Merajver SD, Frank TS. Clinical and pathological significance of microsatellite instability in sporadic endometrial carcinoma. Am J Pathol 1996;148: 1671–8. [17] Senba S, Konishi F, Okamoto T, Kashiwagi H, Kanazawa K, Miyaki

58

[18] [19]

[20]

[21]

M. Suzuki et al. / Cancer Genetics and Cytogenetics 122 (2000) 55–58 M, Konishi M, Tsukamoto T. Clinicopathologic and genetic features of non-familial colorectal carcinomas with a DNA replication error. Cancer 1998;82:279–85. Kaplan EL, Meier P. Nonparametric estimation for incomplete observations. J Am Stat Assoc 1958;53:457–81. Liu F-S, Ho ES-C, Lai C-R, Chen J-T, Shih RT-P, Yang C-H, Tsao C-M. Overexpression of p534 is not a feature of ovarian granulosa cell tumors. Gynecol Oncol 1996;61:50–3. Kappes S, Milde-Langosch K, Kressin P, Passlack B, DockhornDworniczak B, Rohlke P, Loning T. p53 mutations in ovarian tumors, detected by temperature-gradient gel electrophoresis, direct sequencing and immunohistochemistry. Int J Cancer 1995;64:52–9. Pelletier J, Bruening W, Kashtan CE, Mauer SM, Manivel JC, Strie-

gel JE, Houghton DC, Junien C, Habib R, Fouser L, Fine RN, Silverman BL, Haber DA, Housman D. Germline mutations in the Wilms’ tumor suppressor gene are associated with abnormal urogenital development in Denys-Drash syndrome. Cell 1991;67:437–47. [22] Coppes MJ, Ye Y, Rackley R, Zhao X-I, Liefers GJ, Casey G, Williams BRG. Analysis of WT1 in granulosa cell and other sex cordstromal tumors. Cancer Res 1993;53:2712–4. [23] Lyons J, Landis CA, Harsh G, Vallar L, Grunewald K, Feichtinger H, Duh Q-Y, Clark OH, Kawasaki E, Bourne HR, McCormick F. Two G protein oncogenes in human endocrine tumors. Science 1990;249:655–9. [24] Shen Y, Mamers P, Jobling T, Burger HG, Fuller PJ. Absence of the previously reported G protein oncogene (gip2) in ovarian granulosa cell tumors. J Clin Endocrinol Metab 1996;81:4159–61.