- Email: [email protected]
Overexpression of p53 Is Not a Feature of Ovarian Granulosa Cell Tumors
GYNECOLOGIC ONCOLOGY ARTICLE NO.
61, 50–53 (1996)
0095
Overexpression of p53 Is Not a Feature of Ovarian Granulosa Cell Tumors FU-SHING LIU, M.D.,* ESTHER SHIH-CHU HO, M.D.,* CHIUNG-RU LAI, M.D.,† JUNG-TA CHEN, M.D.,‡ ROCKY TAI-PING SHIH, M.D.,‡ CHING-HWA YANG, M.D.,* AND CHIEN-MING TSAO, M.D.* Departments of *Obstetrics and Gynecology and ‡Pathology, Taichung Veterans General Hospital, Taichung, Taiwan, Republic of China; and †Department of Pathology, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China Received July 12, 1995
sion of the Wilms’ tumor suppressor gene WT1 was analyzed in granulosa cell tumors and no WT1 mutation was detected [7]. Another tumor suppressor gene, p53, has not yet been studied in granulosa cell tumors, although it plays an important role in the pathogenesis of various human neoplasms including epithelial ovarian cancers [8–11]. In the present study, we used immunohistochemical assay to examine whether p53 overexpression occurred in these specific ovarian neoplasms.
The p53 tumor suppressor gene has been extensively studied in various human tumors including epithelial ovarian cancers. However, little is known about the expression of this gene in ovarian granulosa cell tumors, the most common histologic type of sex cord–stromal tumors. We investigated whether overexpression of the p53 gene product occurs in this specific ovarian tumor. Nineteen patients with ovarian granulosa cell tumors were recruited in this study. Immunohistochemical staining for the p53 protein with monoclonal antibody PAb 1801 was performed in the paraffin-embedded tissue of each case to screen for p53 overexpression. Among the 19 ovarian granulosa cell tumors, there was only one well-differentiated tumor found to have nuclear immunoreactivity in a small fraction of tumor cells. Polymerase chain reaction– single-stranded conformation polymorphism was used to study the tumor showing focal p53 positivity, but no mobility shift was noted from exon 4 through exon 9 of the p53 gene. On the basis of this observation, we propose that alteration of the p53 tumor suppressor gene is not a common finding in ovarian granulosa cell tumors. q 1996 Academic Press, Inc.
MATERIALS AND METHODS
Tissue specimens. Nineteen patients with ovarian granulosa cell tumors were included in this study. Each patient underwent exploratory laparotomy at either Taichung Veterans General Hospital (10 patients) or Taipei Veterans General Hospital (9 patients) between 1987 and 1994. The stage of disease was assigned based on surgical–pathologic findings according to the International Federation of Gynecology and Obstetrics (FIGO) staging system for carcinoma of the ovary. All histopathology slides were reviewed by pathologists at these two hospitals to confirm the diagnosis and the tumor differentiation grading. Tumors were classified as well-differentiated if the histologic patterns were microfollicular, macrofollicular, insular, trabecular, or tubular, and as poorly differentiated if the patterns are watered silk, gyriform, or diffuse [12]. Immunohistochemical staining. The procedures of immunohistochemical staining for p53 were according to the previously published techniques [13]. Briefly, paraffin blocks of tumors from the 19 patients were cut at 4 mm and placed on silanized slides (DAKO, Carpinteria, CA). After deparaffinization, the sections were treated with 3% hydrogen peroxide to eliminate staining due to endogeneous peroxidase. To retrieve masked antigens in formalin-fixed tissues, slides were immersed within citrate buffer (pH 6.0) and heated in a microwave oven (700 W) three times for 5
INTRODUCTION
Granulosa cell tumors represent the ovarian line of differentiation in sex cord–stromal tumor and account for approximately 1.5 to 3% of primary and 6 to 10% of malignant ovarian neoplasms [1]. Clinically, granulosa cell tumors behave as low-grade malignancies, with 86 to 100% longterm survival rates for stage Ia tumors, but the prognosis is significantly worse for larger tumors that have ruptured (stage Ic, 60% long-term survivors) or when there is extraovarian spread (26 to 49% long-term survivors) [1, 2]. Recently various risk factors, such as mitotic rate and DNA ploidy of tumor cells, were evaluated to determine the patient prognosis [3–5]. Histopathologic observation of ovarian granulosa cell tumorigenesis in mice model was also described [6]. Little is known, however, about the molecular pathogenesis of these tumors. In a recent study, the expres50
0090-8258/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
/ 6d0a$$4222
03-11-96 23:50:37
goas
AP: Gyn Onc
51
p53 IN OVARIAN GRANULOSA CELL TUMORS
TABLE 1 Patient Characteristics Case
Age
Histologic type
Differentiation
Stage
1a 2 3a 4 5a,b 6 7 8 9 10 11 12 13 14 15 16 17 18 19
9 57 45 73 75 73 68 38 49 55 66 43 17 75 56 45 55 46 40
Juvenile Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult
Well Well Well Well Well Well Well Well Well Well Well Well Well Poor Poor Poor Poor Poor Poor
Ia Ia IIIa Ia Ia Ia Ia Ia Ic Ic Ia Ia Ia IIIb Ia Ia IIIb Ia Ia
a DNA was extracted from paraffin-embedded tissue and PCR–SSCP was performed. b Nuclear immunostaining of p53 was found in a focal area of tumor cells.
min each time. After being rinsed in two changes of phosphate-buffered saline (PBS, pH 7.6), slides were preincubated in serum blocking solution (10% goat serum) and followed by incubation for 1 hr with primary antibody. Monoclonal antibody PAb 1801 (1:100) (Oncogene Science, Manhasset, NY), which recognizes an epitope located between amino acids 32 and 79, was used to detect accumulated p53 protein. A slide of a paraffin-embedded epithelial ovarian cancer, known to possess p53 mutation, was used as positive control; another section of this tumor blank for primary antibody was used as negative control. Slides were then rinsed in three changes of PBS and incubated with the biotinylated secondary antibody (DAKO LSAB Kit K681). Following an incubation with peroxidase-labeled streptavidin (DAKO LSAB), slides were rinsed again and sections were developed with the enzyme substrate diaminobenzidine. p53 immunostaining was classified as overexpressing only if nuclear staining was present in the majority of tumor cells. Polymerase chain reaction (PCR)–single-stranded conformation polymorphism (SSCP) analysis. There was one tumor where focal immunoreactivity was noted in immunohistochemical study (case 5 in Table 1). To further examine whether p53 mutation was present in this tumor, molecular study was performed. DNAs were extracted from the paraffin-embedded tissues of this tumor and another two tumors (cases 1 and 3 in Table 1). Intron-based oligonucleotide
/ 6d0a$$4222
03-11-96 23:50:37
goas
primers were used to individually amplify exons 4, 5, 6, 7, 8, and 9 of the p53 gene using polymerase chain reaction as described by Kohler et al. [9]. The PCR conditions and the procedures of SSCP were the same as described previously [13]. DNA samples from blood of a healthy man and from tumors with known p53 mutation in appropriate exons were included in the PCR–SSCP analysis, serving as negative and positive controls, respectively. RESULTS
The clinical data and histologic classifications of the 19 ovarian granulosa cell tumors are shown in Table 1. Eighteen cases were adult granulosa cell tumors and only one was juvenile type. As for tumor differentiation, six cases were poorly differentiated, while the others were well differentiated. Overexpression of p53 was not observed in any of the 19 ovarian granulosa cell tumors, but nuclear staining was noted in a small portion of a well-differentiated tumor (Case 5 in Table 1 and Fig. 1A). In contrast, the positive control of the epithelial ovarian cancer known to have p53 mutation revealed homogeneous nuclear staining among the tumor cells (Fig. 1B). For the three tumors subjected to PCR– SSCP analysis, none showed mobility shift on SSCP from exon 4 through exon 9 of the p53 gene. Examples of the SSCP analysis are shown in Fig. 2. DISCUSSION
The p53 gene has been implicated in many inherited and sporadic forms of malignancies in humans. This human gene lies on chromosome 17p and has been examined in a wide variety of primary tumors [14]. Of these, epithelial ovarian tumors have been studied sequentially in detail from advanced-stage [8] and early-stage [9] to benign and borderline epithelial ovarian tumors [10]. Through observations from these studies it was proposed that alterations of p53 were late events in the carcinogenesis of epithelial tumors [9]. Conversely, a recent study reported a significant association between p53 protein accumulation and tumor progression in mucinous borderline tumors. It was therefore controversial that p53 abnormalities might be early events in ovarian carcinoma [15]. In addition, overexpression of p53 was reported to be associated with a poor prognosis in epithelial ovarian cancer [16]. In the present study we attempted to investigate with immunohistochemical assay to determine whether p53 aberration occurred in ovarian granulosa cell tumors, a topic with no information in the literatures. In addition to molecular biology studies, immunohistochemistry is another useful method to investigate p53 abnormality. Some limitations of immunohistochemistry in detecting p53 alterations are noted, e.g., deletion, nonsense mutations, and splice site mutations can result in absence
AP: Gyn Onc
52
LIU ET AL.
FIG. 1. Immunohistochemical staining for p53. (A) Focal area of nuclear staining in a well-differentiated granulosa cell tumor. (B) Overexpression in an epithelial ovarian cancer as positive control.
of p53 protein overexpression [11]. However, a very good correlation was found between the presence of p53 missence mutations and accumulation of the mutated p53 protein [11, 17–19]. Moreover, immunohistochemical study has the advantage of detecting a gene abnormality at the single-cell level while the molecular analysis does not [19]. It is thus a reliable and ideal method to screen p53 alterations [20]. In our study, immunohistochemical staining with PAb 1801 could not detect nuclear accumulation of p53 in most of the studied granulosa cell tumors (18/19) regardless of the histologic type and tumor cell differentiation. The only exception was an early-stage (Ia), well-differentiated tumor in which a small fraction of tumor cells revealed nuclear immunoreactivity. Such heterogenous staining of p53 was also noted in other kinds of tumors and different conclusions on the occurrence of p53 mutation were made [13, 21]. The data presented in the current study suggested that p53 alteration was not a common finding in ovarian granulosa cell tumors. Because p53 mutations have been demonstrated to be the most common genetic change in a wide variety of primary tumors, the absence of p53 abnormality found in most of the studied ovarian granulosa cell tumors may indicate that a different mechanism is involved in the tumor development.
FIG. 2. Single-stranded conformation polymorphism analysis of exons 7 (A) and 8 (B) of the p53 gene. Lanes 1, 2, and 3: DNAs extracted from cases 1, 3, and 5 in Table 1, respectively. N, negative control, DNA extracted from blood of a healthy man; P, positive control, DNA extracted from epithelial ovarian cancers known to have mutations on exon 7 and 8, respectively.
/ 6d0a$$4222
03-11-96 23:50:37
goas
Recently the granulosa cell tumorigenesis was studied with animal models. It was noted that ovarian granulosa cell tumors could be induced in some recombinant inbred strains of mice with administration of dehydroepiandrosterone or testosterone [22]. The mice granulosa cell tumor bears a striking resemblance to human juvenile granulosa cell tumors. It was soon found that the mice tumorigenesis was controlled by a small number of genes. One of these, granulosa cell tumor susceptibility (Gct), was located on mouse chromosome 4 and could be revealed by dehydroepiandrosterone [23]. It is thus of interest to investigate whether human juvenile granulosa cell tumor is also related to an abnormal androgen environment. In another study, the a-subunit of inhibin, a dimeric ovarian glycoprotein hormone, was described as a tumor suppressor gene with gonadal specificity in mice [24]. In this study gonadal stromal tumors ranging from granulosa cell hyperplasia to granulosa–thecal cell tumors developed in the ainhibin-deficient mice within the first few weeks of life. On the other hand, the serum inhibin levels were found to be elevated in patients with ovarian granulosa cell tumors and it was suggested that granulosa cell tumors produced inhibin [25]. This proposal was recently further confirmed by immunohistochemical study in which strong positive a-inhibin was observed in a granulosa cell tumor [26]. The role of ainhibin in the pathogenesis of granulosa cell tumors therefore needs additional investigation. In addition, chromosome abnormality, i.e., trisomy 12, was noted in human ovarian granulosa cell tumors [27, 28]. It was suggested that increased copy number of gene(s) located on chromosome 12 might give rise to the occurrence of granulosa cell tumors [28]. Several potential oncogenes and known oncogenes which have been mapped to chromosome 12 (KRAS2, GLI, INT1, RAP1B, IGF1, and IFNG) were proposed as probable candidates, but expression of
AP: Gyn Onc
p53 IN OVARIAN GRANULOSA CELL TUMORS
these genes has not been studied in ovarian granulosa cell tumors [28]. Further studies focused on the precise nature of these genes may be helpful in understanding the molecular pathogenesis of granulosa cell tumors.
14. 15.
REFERENCES 16. 1. Costa, M. J., DeRose, P. B., Roth, L. M., Brescia, R. J., Zaloudek, C. J., and Cohen, C. Immunohistochemical phenotype of ovarian granulosa cell tumors, Hum. Pathol. 25, 60–66 (1994). 2. Bjorkholm, E., and Silfversward, C. Prognostic factors in granulosa cell tumors, Gynecol. Oncol. 11, 261–274 (1981). 3. Haba, R., Miki, H., Kobayashi, S., and Ohmori, M. Combined analysis of flow cytometry and morphometry of ovarian granulosa cell tumor, Cancer 72, 3258–3262 (1993).
17.
18.
4. Malmstrom, H., Hogberg, T., Risberg, B., and Simonsen, E. Granulosa cell tumors of the ovary: Prognostic factors and outcome, Gynecol. Oncol. 52, 50–55 (1994).
19.
5. Evans, M. P., Webb, M. J., Gaffey, T. A., Katzmann, J. A., Suman, V. J., and Hu, T. C. DNA ploidy of ovarian granulosa cell tumors, Cancer 75, 2295–2298 (1995).
20.
6. Tennent, B. J., Shultz, K. L., Sundberg, J. P., and Beamer, W. G. Ovarian granulosa cell tumorigenesis in SWR-derived F1 hybrid mice: Preneoplastic follicular abnormality and malignant disease progression, Am. J. Obstet. Gynecol. 163, 625–634 (1990).
21.
7. Coppes, M. J., Ye, Y., Rackley, R., Zhao, X. I., Liefers, G. J., Casey, G., and Williams, B. R. G. Analysis of WT1 in granulosa cell and other sex cord-stromal tumors, Cancer Res. 53, 2712–2714 (1993). 8. Marks, J. R., Davidoff, A. M., Kerns, B. J., Humphrey, P. A., Pence, J. C., Dodge, R. K., and Berchuck, A. Overexpression and mutation of p53 in epithelial ovarian cancer, Cancer Res. 51, 2979–2984 (1991). 9. Kohler, M. F., Kerns, B. J., Humphrey, P. A., Marks, J. R., Bast, R. C. Jr., and Berchuck, A. Mutation and overexpression of p53 in early-stage epithelial ovarian cancer, Obstet. Gynecol. 81, 643–650 (1993). 10. Berchuck, A., Kohler, M. F., Hopkins, M. P., Humphrey, P. A., Robboy, S. J., and Rodriguez, G. C. Overexpression of p53 is not a feature of benign and early-stage borderline epithelial ovarian tumors, Gynecol. Oncol. 52, 232–236 (1994). 11. Kupryjanczyk, J., Thor, A. D., Beauchamp, R., Merritt, V., Edgerton, S. M., and Bell, D. A. P53 gene mutations and protein accumulation in human ovarian cancer, Proc. Natl. Acad. Sci. USA 90, 4961–4965 (1993). 12. Young, R. H., and Scully, R. E. Sex cord-stromal, steroid cell, and other ovarian tumors, with endocrine, paraendocrine, and paraneoplastic manifestations, in Blaustein’s Pathology of the female genital tract (R. J. Kurman, Ed.), Springer-Verlag, New York, pp. 783–847 (1994). 13. Liu, F. S., Ho, E. S-C., Chen, J. T., Shih, R. T-P., Yang, C. H., and Shih, A. Overexpression or mutation of the p53 tumor suppressor gene
/ 6d0a$$4222
03-11-96 23:50:37
goas
22.
23.
24.
25.
26.
27.
28.
53
does not occur in malignant ovarian germ cell tumors, Cancer 76, 291– 295 (1995). Levine, A. J., Momand, J., and Finlay, C. A. The p53 tumor suppressor gene, Nature 351, 453–456 (1991). Kupryjanczyk, J., Bell, D. A., Yandell, D. W., Scully, R. E., and Thor, A. D. p53 expression in ovarian borderline tumors and stage I carcinomas, Am. J. Clin. Pathol. 102, 671–676 (1994). Hartmann, L. C., Podratz, K. C., Keeney, G. L., Kamel, N. A., Edmonson, J. H., Grill, J. P., Su, J. Q., Katzmann, J. A., and Roche, P. C. Prognostic significance of p53 immunostaining in epithelial ovarian cancer, J. Clin. Oncol. 12, 64–69 (1994). Esrig, D., Spruck, C. H., III, Nichols, P. W., Chaiwun, B., Steven, K., and Groshen, S. p53 nuclear protein accumulation correlates with mutations in the p53 gene, tumor grade, and stage in bladder cancer, Am. J. Pathol. 143, 1389–1397 (1993). Bartek, J., Bartkova, J., Lukas, J., Staskova, Z., Vojtesek, B., and Lane, D. P. Immunohistochemical analysis of the p53 oncoprotein on paraffin sections using a series of novel monoclonal antibodies, J. Pathol. 169, 27–34 (1993). Inoue, M., Fujita, M., Enomoto, T., Morimoto, H., Monden, T., and Shimano, T. Immunohistochemical analysis of p53 in gynecologic tumors, Am. J. Clin. Pathol. 102, 665–670 (1994). Hurlimann, J., Chaubert, P., and Benhattar, J. p53 gene alterations and p53 protein accumulation in infiltrating ductal breast carcinomas: Correlation between immunohistochemical and molecular biology techniques, Mod. Pathol. 7, 423–428 (1994). Bartkova, J., Bartek, J., Lukas, J., Vojtesek, B., Staskova, Z., Rejthar, A., Kovarik, J., Midgley, C. A., and Lane, D. P. p53 protein alterations in human testicular cancer including pre-invasive intratubular germcell neoplasia, Int. J. Cancer 49, 196–202 (1991). Beamer, W. G., Shultz, K. L., and Tennent, B. J. Induction of ovarian granulosa cell tumors in SWXJ-9 mice with dehydroepiandrosterone, Cancer Res. 48, 2788–2792 (1988). Beamer, W. G., Tennent, B. J., Shultz, K. L., Nadeau, J. H., Shultz, L. D., and Skow, L. C. Gene for ovarian granulosa cell tumor susceptibility, Gct, in SWXJ recombinant inbred strains of mice revealed by dehydroepiendrosterone, Cancer Res. 48, 5092–5095 (1988). Matzuk, M. M., Finegold, M. J., Su, J-G. J., Hsueh, A. J. W., and Bradley, A. a-Inhibin is a tumor-suppressor gene with gonadal specificity in mice, Nature 360, 313–319 (1992). Lappohn, R. E., Burger, H. G., Bouma, J., Bangah, M., Krans, M., and de Bruijin, H. Inhibin as a marker for granulosa cell tumors, N. Engl. J. Med. 321, 790–793 (1989). Gurusinghe, C. J., Healy, D. L., Jobling, T., Mamers, P., and Burger, H. G. Inhibin and activin are demonstrable by immunohistochemistry in ovarian tumor tissue, Gynecol. Oncol. 57, 27–32 (1995). Leung, W. Y., Schwartz, P. E., Ng, H. T., and Yang-Feng, T. L. Trisomy 12 in benign fibroma and granulosa cell tumor of the ovary, Gynecol. Oncol. 38, 28–31 (1990). Fletcher, J. A., Gibas, Z., Donovan, K., Perez-Atayde, A., Genest, D., Morton, C. C., and Lage, J. M. Ovarian granulosa–stromal cell tumors are characterized by trisomy 12, Am. J. Pathol. 138, 515–520 (1991).
AP: Gyn Onc
61, 50–53 (1996)
0095
Overexpression of p53 Is Not a Feature of Ovarian Granulosa Cell Tumors FU-SHING LIU, M.D.,* ESTHER SHIH-CHU HO, M.D.,* CHIUNG-RU LAI, M.D.,† JUNG-TA CHEN, M.D.,‡ ROCKY TAI-PING SHIH, M.D.,‡ CHING-HWA YANG, M.D.,* AND CHIEN-MING TSAO, M.D.* Departments of *Obstetrics and Gynecology and ‡Pathology, Taichung Veterans General Hospital, Taichung, Taiwan, Republic of China; and †Department of Pathology, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China Received July 12, 1995
sion of the Wilms’ tumor suppressor gene WT1 was analyzed in granulosa cell tumors and no WT1 mutation was detected [7]. Another tumor suppressor gene, p53, has not yet been studied in granulosa cell tumors, although it plays an important role in the pathogenesis of various human neoplasms including epithelial ovarian cancers [8–11]. In the present study, we used immunohistochemical assay to examine whether p53 overexpression occurred in these specific ovarian neoplasms.
The p53 tumor suppressor gene has been extensively studied in various human tumors including epithelial ovarian cancers. However, little is known about the expression of this gene in ovarian granulosa cell tumors, the most common histologic type of sex cord–stromal tumors. We investigated whether overexpression of the p53 gene product occurs in this specific ovarian tumor. Nineteen patients with ovarian granulosa cell tumors were recruited in this study. Immunohistochemical staining for the p53 protein with monoclonal antibody PAb 1801 was performed in the paraffin-embedded tissue of each case to screen for p53 overexpression. Among the 19 ovarian granulosa cell tumors, there was only one well-differentiated tumor found to have nuclear immunoreactivity in a small fraction of tumor cells. Polymerase chain reaction– single-stranded conformation polymorphism was used to study the tumor showing focal p53 positivity, but no mobility shift was noted from exon 4 through exon 9 of the p53 gene. On the basis of this observation, we propose that alteration of the p53 tumor suppressor gene is not a common finding in ovarian granulosa cell tumors. q 1996 Academic Press, Inc.
MATERIALS AND METHODS
Tissue specimens. Nineteen patients with ovarian granulosa cell tumors were included in this study. Each patient underwent exploratory laparotomy at either Taichung Veterans General Hospital (10 patients) or Taipei Veterans General Hospital (9 patients) between 1987 and 1994. The stage of disease was assigned based on surgical–pathologic findings according to the International Federation of Gynecology and Obstetrics (FIGO) staging system for carcinoma of the ovary. All histopathology slides were reviewed by pathologists at these two hospitals to confirm the diagnosis and the tumor differentiation grading. Tumors were classified as well-differentiated if the histologic patterns were microfollicular, macrofollicular, insular, trabecular, or tubular, and as poorly differentiated if the patterns are watered silk, gyriform, or diffuse [12]. Immunohistochemical staining. The procedures of immunohistochemical staining for p53 were according to the previously published techniques [13]. Briefly, paraffin blocks of tumors from the 19 patients were cut at 4 mm and placed on silanized slides (DAKO, Carpinteria, CA). After deparaffinization, the sections were treated with 3% hydrogen peroxide to eliminate staining due to endogeneous peroxidase. To retrieve masked antigens in formalin-fixed tissues, slides were immersed within citrate buffer (pH 6.0) and heated in a microwave oven (700 W) three times for 5
INTRODUCTION
Granulosa cell tumors represent the ovarian line of differentiation in sex cord–stromal tumor and account for approximately 1.5 to 3% of primary and 6 to 10% of malignant ovarian neoplasms [1]. Clinically, granulosa cell tumors behave as low-grade malignancies, with 86 to 100% longterm survival rates for stage Ia tumors, but the prognosis is significantly worse for larger tumors that have ruptured (stage Ic, 60% long-term survivors) or when there is extraovarian spread (26 to 49% long-term survivors) [1, 2]. Recently various risk factors, such as mitotic rate and DNA ploidy of tumor cells, were evaluated to determine the patient prognosis [3–5]. Histopathologic observation of ovarian granulosa cell tumorigenesis in mice model was also described [6]. Little is known, however, about the molecular pathogenesis of these tumors. In a recent study, the expres50
0090-8258/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
/ 6d0a$$4222
03-11-96 23:50:37
goas
AP: Gyn Onc
51
p53 IN OVARIAN GRANULOSA CELL TUMORS
TABLE 1 Patient Characteristics Case
Age
Histologic type
Differentiation
Stage
1a 2 3a 4 5a,b 6 7 8 9 10 11 12 13 14 15 16 17 18 19
9 57 45 73 75 73 68 38 49 55 66 43 17 75 56 45 55 46 40
Juvenile Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult Adult
Well Well Well Well Well Well Well Well Well Well Well Well Well Poor Poor Poor Poor Poor Poor
Ia Ia IIIa Ia Ia Ia Ia Ia Ic Ic Ia Ia Ia IIIb Ia Ia IIIb Ia Ia
a DNA was extracted from paraffin-embedded tissue and PCR–SSCP was performed. b Nuclear immunostaining of p53 was found in a focal area of tumor cells.
min each time. After being rinsed in two changes of phosphate-buffered saline (PBS, pH 7.6), slides were preincubated in serum blocking solution (10% goat serum) and followed by incubation for 1 hr with primary antibody. Monoclonal antibody PAb 1801 (1:100) (Oncogene Science, Manhasset, NY), which recognizes an epitope located between amino acids 32 and 79, was used to detect accumulated p53 protein. A slide of a paraffin-embedded epithelial ovarian cancer, known to possess p53 mutation, was used as positive control; another section of this tumor blank for primary antibody was used as negative control. Slides were then rinsed in three changes of PBS and incubated with the biotinylated secondary antibody (DAKO LSAB Kit K681). Following an incubation with peroxidase-labeled streptavidin (DAKO LSAB), slides were rinsed again and sections were developed with the enzyme substrate diaminobenzidine. p53 immunostaining was classified as overexpressing only if nuclear staining was present in the majority of tumor cells. Polymerase chain reaction (PCR)–single-stranded conformation polymorphism (SSCP) analysis. There was one tumor where focal immunoreactivity was noted in immunohistochemical study (case 5 in Table 1). To further examine whether p53 mutation was present in this tumor, molecular study was performed. DNAs were extracted from the paraffin-embedded tissues of this tumor and another two tumors (cases 1 and 3 in Table 1). Intron-based oligonucleotide
/ 6d0a$$4222
03-11-96 23:50:37
goas
primers were used to individually amplify exons 4, 5, 6, 7, 8, and 9 of the p53 gene using polymerase chain reaction as described by Kohler et al. [9]. The PCR conditions and the procedures of SSCP were the same as described previously [13]. DNA samples from blood of a healthy man and from tumors with known p53 mutation in appropriate exons were included in the PCR–SSCP analysis, serving as negative and positive controls, respectively. RESULTS
The clinical data and histologic classifications of the 19 ovarian granulosa cell tumors are shown in Table 1. Eighteen cases were adult granulosa cell tumors and only one was juvenile type. As for tumor differentiation, six cases were poorly differentiated, while the others were well differentiated. Overexpression of p53 was not observed in any of the 19 ovarian granulosa cell tumors, but nuclear staining was noted in a small portion of a well-differentiated tumor (Case 5 in Table 1 and Fig. 1A). In contrast, the positive control of the epithelial ovarian cancer known to have p53 mutation revealed homogeneous nuclear staining among the tumor cells (Fig. 1B). For the three tumors subjected to PCR– SSCP analysis, none showed mobility shift on SSCP from exon 4 through exon 9 of the p53 gene. Examples of the SSCP analysis are shown in Fig. 2. DISCUSSION
The p53 gene has been implicated in many inherited and sporadic forms of malignancies in humans. This human gene lies on chromosome 17p and has been examined in a wide variety of primary tumors [14]. Of these, epithelial ovarian tumors have been studied sequentially in detail from advanced-stage [8] and early-stage [9] to benign and borderline epithelial ovarian tumors [10]. Through observations from these studies it was proposed that alterations of p53 were late events in the carcinogenesis of epithelial tumors [9]. Conversely, a recent study reported a significant association between p53 protein accumulation and tumor progression in mucinous borderline tumors. It was therefore controversial that p53 abnormalities might be early events in ovarian carcinoma [15]. In addition, overexpression of p53 was reported to be associated with a poor prognosis in epithelial ovarian cancer [16]. In the present study we attempted to investigate with immunohistochemical assay to determine whether p53 aberration occurred in ovarian granulosa cell tumors, a topic with no information in the literatures. In addition to molecular biology studies, immunohistochemistry is another useful method to investigate p53 abnormality. Some limitations of immunohistochemistry in detecting p53 alterations are noted, e.g., deletion, nonsense mutations, and splice site mutations can result in absence
AP: Gyn Onc
52
LIU ET AL.
FIG. 1. Immunohistochemical staining for p53. (A) Focal area of nuclear staining in a well-differentiated granulosa cell tumor. (B) Overexpression in an epithelial ovarian cancer as positive control.
of p53 protein overexpression [11]. However, a very good correlation was found between the presence of p53 missence mutations and accumulation of the mutated p53 protein [11, 17–19]. Moreover, immunohistochemical study has the advantage of detecting a gene abnormality at the single-cell level while the molecular analysis does not [19]. It is thus a reliable and ideal method to screen p53 alterations [20]. In our study, immunohistochemical staining with PAb 1801 could not detect nuclear accumulation of p53 in most of the studied granulosa cell tumors (18/19) regardless of the histologic type and tumor cell differentiation. The only exception was an early-stage (Ia), well-differentiated tumor in which a small fraction of tumor cells revealed nuclear immunoreactivity. Such heterogenous staining of p53 was also noted in other kinds of tumors and different conclusions on the occurrence of p53 mutation were made [13, 21]. The data presented in the current study suggested that p53 alteration was not a common finding in ovarian granulosa cell tumors. Because p53 mutations have been demonstrated to be the most common genetic change in a wide variety of primary tumors, the absence of p53 abnormality found in most of the studied ovarian granulosa cell tumors may indicate that a different mechanism is involved in the tumor development.
FIG. 2. Single-stranded conformation polymorphism analysis of exons 7 (A) and 8 (B) of the p53 gene. Lanes 1, 2, and 3: DNAs extracted from cases 1, 3, and 5 in Table 1, respectively. N, negative control, DNA extracted from blood of a healthy man; P, positive control, DNA extracted from epithelial ovarian cancers known to have mutations on exon 7 and 8, respectively.
/ 6d0a$$4222
03-11-96 23:50:37
goas
Recently the granulosa cell tumorigenesis was studied with animal models. It was noted that ovarian granulosa cell tumors could be induced in some recombinant inbred strains of mice with administration of dehydroepiandrosterone or testosterone [22]. The mice granulosa cell tumor bears a striking resemblance to human juvenile granulosa cell tumors. It was soon found that the mice tumorigenesis was controlled by a small number of genes. One of these, granulosa cell tumor susceptibility (Gct), was located on mouse chromosome 4 and could be revealed by dehydroepiandrosterone [23]. It is thus of interest to investigate whether human juvenile granulosa cell tumor is also related to an abnormal androgen environment. In another study, the a-subunit of inhibin, a dimeric ovarian glycoprotein hormone, was described as a tumor suppressor gene with gonadal specificity in mice [24]. In this study gonadal stromal tumors ranging from granulosa cell hyperplasia to granulosa–thecal cell tumors developed in the ainhibin-deficient mice within the first few weeks of life. On the other hand, the serum inhibin levels were found to be elevated in patients with ovarian granulosa cell tumors and it was suggested that granulosa cell tumors produced inhibin [25]. This proposal was recently further confirmed by immunohistochemical study in which strong positive a-inhibin was observed in a granulosa cell tumor [26]. The role of ainhibin in the pathogenesis of granulosa cell tumors therefore needs additional investigation. In addition, chromosome abnormality, i.e., trisomy 12, was noted in human ovarian granulosa cell tumors [27, 28]. It was suggested that increased copy number of gene(s) located on chromosome 12 might give rise to the occurrence of granulosa cell tumors [28]. Several potential oncogenes and known oncogenes which have been mapped to chromosome 12 (KRAS2, GLI, INT1, RAP1B, IGF1, and IFNG) were proposed as probable candidates, but expression of
AP: Gyn Onc
p53 IN OVARIAN GRANULOSA CELL TUMORS
these genes has not been studied in ovarian granulosa cell tumors [28]. Further studies focused on the precise nature of these genes may be helpful in understanding the molecular pathogenesis of granulosa cell tumors.
14. 15.
REFERENCES 16. 1. Costa, M. J., DeRose, P. B., Roth, L. M., Brescia, R. J., Zaloudek, C. J., and Cohen, C. Immunohistochemical phenotype of ovarian granulosa cell tumors, Hum. Pathol. 25, 60–66 (1994). 2. Bjorkholm, E., and Silfversward, C. Prognostic factors in granulosa cell tumors, Gynecol. Oncol. 11, 261–274 (1981). 3. Haba, R., Miki, H., Kobayashi, S., and Ohmori, M. Combined analysis of flow cytometry and morphometry of ovarian granulosa cell tumor, Cancer 72, 3258–3262 (1993).
17.
18.
4. Malmstrom, H., Hogberg, T., Risberg, B., and Simonsen, E. Granulosa cell tumors of the ovary: Prognostic factors and outcome, Gynecol. Oncol. 52, 50–55 (1994).
19.
5. Evans, M. P., Webb, M. J., Gaffey, T. A., Katzmann, J. A., Suman, V. J., and Hu, T. C. DNA ploidy of ovarian granulosa cell tumors, Cancer 75, 2295–2298 (1995).
20.
6. Tennent, B. J., Shultz, K. L., Sundberg, J. P., and Beamer, W. G. Ovarian granulosa cell tumorigenesis in SWR-derived F1 hybrid mice: Preneoplastic follicular abnormality and malignant disease progression, Am. J. Obstet. Gynecol. 163, 625–634 (1990).
21.
7. Coppes, M. J., Ye, Y., Rackley, R., Zhao, X. I., Liefers, G. J., Casey, G., and Williams, B. R. G. Analysis of WT1 in granulosa cell and other sex cord-stromal tumors, Cancer Res. 53, 2712–2714 (1993). 8. Marks, J. R., Davidoff, A. M., Kerns, B. J., Humphrey, P. A., Pence, J. C., Dodge, R. K., and Berchuck, A. Overexpression and mutation of p53 in epithelial ovarian cancer, Cancer Res. 51, 2979–2984 (1991). 9. Kohler, M. F., Kerns, B. J., Humphrey, P. A., Marks, J. R., Bast, R. C. Jr., and Berchuck, A. Mutation and overexpression of p53 in early-stage epithelial ovarian cancer, Obstet. Gynecol. 81, 643–650 (1993). 10. Berchuck, A., Kohler, M. F., Hopkins, M. P., Humphrey, P. A., Robboy, S. J., and Rodriguez, G. C. Overexpression of p53 is not a feature of benign and early-stage borderline epithelial ovarian tumors, Gynecol. Oncol. 52, 232–236 (1994). 11. Kupryjanczyk, J., Thor, A. D., Beauchamp, R., Merritt, V., Edgerton, S. M., and Bell, D. A. P53 gene mutations and protein accumulation in human ovarian cancer, Proc. Natl. Acad. Sci. USA 90, 4961–4965 (1993). 12. Young, R. H., and Scully, R. E. Sex cord-stromal, steroid cell, and other ovarian tumors, with endocrine, paraendocrine, and paraneoplastic manifestations, in Blaustein’s Pathology of the female genital tract (R. J. Kurman, Ed.), Springer-Verlag, New York, pp. 783–847 (1994). 13. Liu, F. S., Ho, E. S-C., Chen, J. T., Shih, R. T-P., Yang, C. H., and Shih, A. Overexpression or mutation of the p53 tumor suppressor gene
/ 6d0a$$4222
03-11-96 23:50:37
goas
22.
23.
24.
25.
26.
27.
28.
53
does not occur in malignant ovarian germ cell tumors, Cancer 76, 291– 295 (1995). Levine, A. J., Momand, J., and Finlay, C. A. The p53 tumor suppressor gene, Nature 351, 453–456 (1991). Kupryjanczyk, J., Bell, D. A., Yandell, D. W., Scully, R. E., and Thor, A. D. p53 expression in ovarian borderline tumors and stage I carcinomas, Am. J. Clin. Pathol. 102, 671–676 (1994). Hartmann, L. C., Podratz, K. C., Keeney, G. L., Kamel, N. A., Edmonson, J. H., Grill, J. P., Su, J. Q., Katzmann, J. A., and Roche, P. C. Prognostic significance of p53 immunostaining in epithelial ovarian cancer, J. Clin. Oncol. 12, 64–69 (1994). Esrig, D., Spruck, C. H., III, Nichols, P. W., Chaiwun, B., Steven, K., and Groshen, S. p53 nuclear protein accumulation correlates with mutations in the p53 gene, tumor grade, and stage in bladder cancer, Am. J. Pathol. 143, 1389–1397 (1993). Bartek, J., Bartkova, J., Lukas, J., Staskova, Z., Vojtesek, B., and Lane, D. P. Immunohistochemical analysis of the p53 oncoprotein on paraffin sections using a series of novel monoclonal antibodies, J. Pathol. 169, 27–34 (1993). Inoue, M., Fujita, M., Enomoto, T., Morimoto, H., Monden, T., and Shimano, T. Immunohistochemical analysis of p53 in gynecologic tumors, Am. J. Clin. Pathol. 102, 665–670 (1994). Hurlimann, J., Chaubert, P., and Benhattar, J. p53 gene alterations and p53 protein accumulation in infiltrating ductal breast carcinomas: Correlation between immunohistochemical and molecular biology techniques, Mod. Pathol. 7, 423–428 (1994). Bartkova, J., Bartek, J., Lukas, J., Vojtesek, B., Staskova, Z., Rejthar, A., Kovarik, J., Midgley, C. A., and Lane, D. P. p53 protein alterations in human testicular cancer including pre-invasive intratubular germcell neoplasia, Int. J. Cancer 49, 196–202 (1991). Beamer, W. G., Shultz, K. L., and Tennent, B. J. Induction of ovarian granulosa cell tumors in SWXJ-9 mice with dehydroepiandrosterone, Cancer Res. 48, 2788–2792 (1988). Beamer, W. G., Tennent, B. J., Shultz, K. L., Nadeau, J. H., Shultz, L. D., and Skow, L. C. Gene for ovarian granulosa cell tumor susceptibility, Gct, in SWXJ recombinant inbred strains of mice revealed by dehydroepiendrosterone, Cancer Res. 48, 5092–5095 (1988). Matzuk, M. M., Finegold, M. J., Su, J-G. J., Hsueh, A. J. W., and Bradley, A. a-Inhibin is a tumor-suppressor gene with gonadal specificity in mice, Nature 360, 313–319 (1992). Lappohn, R. E., Burger, H. G., Bouma, J., Bangah, M., Krans, M., and de Bruijin, H. Inhibin as a marker for granulosa cell tumors, N. Engl. J. Med. 321, 790–793 (1989). Gurusinghe, C. J., Healy, D. L., Jobling, T., Mamers, P., and Burger, H. G. Inhibin and activin are demonstrable by immunohistochemistry in ovarian tumor tissue, Gynecol. Oncol. 57, 27–32 (1995). Leung, W. Y., Schwartz, P. E., Ng, H. T., and Yang-Feng, T. L. Trisomy 12 in benign fibroma and granulosa cell tumor of the ovary, Gynecol. Oncol. 38, 28–31 (1990). Fletcher, J. A., Gibas, Z., Donovan, K., Perez-Atayde, A., Genest, D., Morton, C. C., and Lage, J. M. Ovarian granulosa–stromal cell tumors are characterized by trisomy 12, Am. J. Pathol. 138, 515–520 (1991).
AP: Gyn Onc