Isolation and Characterization of Mouse Ovarian Surface Epithelial Cell Lines

Isolation and Characterization of Mouse Ovarian Surface Epithelial Cell Lines

PATHOLOGY RESEARCH AND PRACTICE © Gustav Fischer Verlag Isolation and Characterization of Mouse Ovarian Surface Epithelial Cell Lines Michiko Kido 1 ...

7MB Sizes 0 Downloads 103 Views

PATHOLOGY RESEARCH AND PRACTICE © Gustav Fischer Verlag

Isolation and Characterization of Mouse Ovarian Surface Epithelial Cell Lines Michiko Kido 1 and Masabumi Shibuya 2 lDepartment of Obstetrics and Gynecology, Tokyo University Branch Hospital, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo, Japan 2Department of Genetics, Institute of Medical Science, University of Tokyo, Shirokanedai, Minato-ku, Tokyo, Japan

Introduction

ovarian malignancies was only 650 in 1960, but has exceeded 3500 in 1992 in Japan [14]. The ovary is made up of various kinds of cells, including germ cells and hormone-producing cells. Several types of tumors arise from this organ. However, two thirds of ovarian tumors, and over 80% of malignancies, are derived from ovarian surface epithelium (OSE), which arise as a single layer of cells covering the ovary [28J. It is still unclear as to why OSE can produce histologically various types of tumors and why many ovarian tumors arise from OSE. It is, therefore, important to know the growth characteristics and mechanism of malignant transformation of OSE to clarify the histogenesis of epithelial ovarian tumors. OSE cells do not have specific conspicuous characteristics, and the lack of any methodological approach for analysis has made its characterization difficult. Further, OSE cells have no specific markers by which they could be distinguished from other cells. Since the 1980's several investigators have developed a culture system for rat, rabbit and human OSE's [2, 3, 25J, but only one old, unconfirmed report for the culture of mouse OSE exists [2IJ. Genetic background of definite strains of mice is clearer than that of humans and other experimental animals. Recently, gene targeting technology has advanced and animals deleted of an important gene can be used to study the exact role of the gene. Most of the animals now available for this study are mice, thus, a variety of gene-deficient mice are now

Ovarian cancer is the leading cause of death of gynecological malignancies in Western countries [8]. In Asia including Japan, it is still less frequent than in the United States or European countries. But as life style has become more Westernized, the disease has been conspicuously increasing. The number of deaths due to

Address for correspondence: Michiko Kido, M.D., Ph.D., Department of Obstetrics and Gynecology, Tokyo University Branch Hospital, Faculty of Medicine, University of Tokyo, 3-28-6, Mejirodai, Bunkyo-ku, Tokyo, 112-0015, Japan. Tel.: +81-3-3943-1151, Fax: +81-3-3943-0564, E-mail: [email protected]

Summary The incidence of ovarian malignancies has significantly increased in the past decades in many countries, however, an appropriate animal model system enabling the study of ovarian cancer such as stable mouse ovarian epithelial cell lines has not yet been developed. Here we report the establishment of cell lines derived from mouse ovarian surface epithelium (MOSE) by two procedures - one, through the introduction of SV40 large T antigen DNA into C3H/He MOSE (T-Ag-MOSE) and another through spontaneous immortalization of cells from p53-deficient MOSE (p53-def-MOSE). p53-defMOSE cell line did not show any transformed phenotype either in vitro culture system nor in vivo tumorigenicity assay, whereas T-Ag-MOSE formed tumors in nude mice. Tumors formed by the injection of T-AgMOSE were undifferentiated malignancies associated with heterologous mesothelial tissues such as those of the osteoid phenotype. The established MOSE cell lines are useful in the molecular analysis of the multistep carcinogenesis of ovarian tissues in humans. Key words: Ovarian surface epithelium - SV40 T antigen - p53-deficient mouse - Transformation

Pathol. Res. Pract. 194: 725-730 (1998)

0344-0338/98/0194-0725$5.00/0

726 . M. Kido and M. Shibuya available for studying the process of carcinogenesis. It is therefore advantageous to develop a culture system for mouse OSE to study the growth characteristics of OSE. We have isolated mouse OSE (MOSE) and established cell lines by two kinds of immortalizing methods. One, by transfection of SV40 large T antigen DNA (T-Ag-MOSE), and another by using the strain of mouse homozygously deleted of p53 gene (p53-defMOSE). These cell lines were then characterized.

Material and Methods

Effect ofgrowth factors or hormones on cell growth

Exactly 2 ml of a single cell suspension (5xl()4 cells/ml) was put into culture dishes 35 mm in diameter. After incubation for certain periods, three dishes were arbitrarily selected and the living cell number per dish was counted by blood cell counter. Average and standard deviation of triplicate dishes were calculated and plotted on a semilogarithmic graph. The concentration of FBS in the control dishes was 0.05-0.5%. Growth factors or hormones were added to the control medium at a final concentration of -EGF: 10 ng/ml, aFGF: 10 ng/ml, bFGF: 10 ng/ml, VEGF: 10 ng/mI, insulin: 10 J.lg/ ml, hydrocortisone: 500 ng/mI, estradiol-17~: 10-7 M, progesterone: 10-8 M, and androstenedione: 10-8 M. To examine the effect of estradiol, we used the medium which did not include phenol red to exclude the estrogen-like effect of this substance [6J.

Primary culture and immortalization ofmouse ovarian suiface epithelial cells

Ovaries were removed from adult female C3H1He mice (8 weeks of age) under aseptic conditions, and transferred to a culture dish containing the high-glucose version of Dulbecco's modified Eagle medium (GIBCO laboratories) supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/ml) and streptomycin (100 J.lg/ml) (complete growth medium). After fallopian tubes and connective tissue surrounding the ovary were carefully removed, ovaries were washed three times by phosphate-buffered saline (PBS) and treated as a whole with 0.025% trypsin for 60 minutes. Then the ovaries were removed and the above mentioned complete growth medium was added to the reaction solution to dilute the trypsin. The cell suspension was then placed in tissue culture dish (LUX) and cultured under conditions of 5% CO2 in 37°C humidified air. The cultured cells did not grow well after several passages, and to obtain immortalized cells, the primary culture cells in the dish were transfected with elongation factor promoterdriven SV40 large T antigen DNA containing plasmid pEF321-T [l8J (kindly provided by Drs. Yamaguchi and Sugano, Institute of Medical Science, Univ. of Tokyo) by the polybrene DMSO method [17]. Colonies of the OSE showing cobblestone appearance were isolated using cloning cylinders. Then the epithelial cell suspension was diluted and placed in 96-well cell culture plates. The wells containing only one cell were marked and a monoclonal cell line (T-Ag-MOSE) were obtained from these wells. T-Ag-MOSE cells have been passaged up to 50 times (I: 10 splits).

Establishment ofa cell line ofovarian suiface epithelial cells ofp53-deficient mouse

A mouse that was homozygously deleted of the p53 gene [31] was kindly given by Dr. Shinichi Aizawa (Kumamoto

Univ., Japan). Primary cultures of ovarian surface epithelial cells of this mouse were obtained with the same method as that of the C3H1He mice mentioned above. A monoclonal cell line (p53-def-MOSE) was isolated and grown from a single cell-containing well. The cells of p53-def-MOSE did not undergo a reduction of growth without being introduced pEF321-T and have been passaged up to 50 times (I: 10 splits).

Test ofcapacity for anchorage independent growth

Cell suspension was centrifuged and the precipitate was resuspended in 0.5% agarose (SEAKEM, low mealting) in complete growth medium at a concentration of lxl04 cells/ml. Three milliliters of the single cell suspension was layered over a solidified bottom layer of 0.8% agarose in medium in a 60 mm culture dish. After the upper layer was solidified, the dishes were incubated at 37°C in 5% CO2 for 3 weeks. Colonies larger than 16 cells in the cell-containing layer were counted. Cloning efficiency was calculated as follows: (the number of colonies)/(the number of cells plated)x100 (%).

Test for tumorigenicity

Three to seven female Balb/c nu/nu mice (six or seven weeks old) were injected with Ix107 cells suspended in medium either subcutaneously or intraperitoneally and monitored for 6 months. Tumors were resected when their diameter exceeded I cm or when the mice died, and examined histologically. Tumor samples were fixed with formalin, embedded, sliced and stained with Hematoxylin and Eosin.

Results Morphology of T-Ag-MOSE and p53-def-MOSE cells

Primary ovarian surface epithelial cells were cultured and used for the establishment of T-Antigen-carrying MOSE (T-Ag-MOSE) or p53-deficient MOSE (p53def-MOSE) (see Material and Methods). Cells ofT-AgMOSE or p53-def-MOSE resembled that of a cobblestone-like epithelial monolayer (Fig. la, b). The cells showed contact inhibition on confluency. On immunohistochemical examination, these cells were laminin positive, strongly suggesting that these cells were derived from epithelial cells (data not shown). Expression of SV40 T-Ag in T-Ag-MOSE and absence of expression of p53 in p53-def-MOSE were both confirmed by Western blotting.

Mouse Ovarian Epithelial Cell Lines . 727

• control, FBS 0.5%

~_+---r--f-o"f cEGF -insulin ~~~ ohydrocortisone

103,-'---+---+----.---.------<~--t

o

6

4

3

2

days

Fig. 2. Growth curve of early passage (at passage 5) p53-defMOSE. Cell growth was promoted by addition of 10 ng/ml of EGF. Similarly, 500 ng/ml of hydrocortisone or 10 Jlg/ml of insulin promoted cell growth.

Fig. 1. Morphology of MOSE. a: T-Ag-MOSE. b: p53-defMOSE. Both cells grew resembling a cobblestone-like epithelial monolayer and showed contact inhibition.

Effect of growth factors and hormones on the growth of the T-Ag-MOSE and p53-def-MOSE

One of the characteristics of OSE is that these cells are growth-stimulated by hydrocortisone [9, 16, 29 J, which suppresses the proliferation of most of the cells such as fibroblasts and epithelial cells.

Growth curve of early passage cells (at passage 5) of p53-def-MOSE are shown in Fig. 2. Cell growth was promoted by the addition of EGF, hydrocortisone, and insulin at the concentration described in Material and Methods. Addition of estradiol 17-~, aFGF, bFGF, VEGF, progesterone and androstenedione did not have any significant effect on growth (data not shown). Growth characteristics of early passage cells (at passage 5) of T-Ag-MOSE were similar to those of the above mentioned p53-def-MOSE. Capacity for anchorage-independent growth of T-Ag-OSE and p53-def-OSE

3Y1 cells [l9J and ras-transformed 3YI were used as a negative and positive control, respectively, for soft agar cloning assay. 3Yl gave essentially no colonies, whereas 3Yl, K-ras- showed many (several hundreds per well) colonies in the assay. Cloning efficiency of these cells, T-Ag-MOSE and p53-def-MOSE, are shown in Fig. 3. Each of T-Ag-MOSE cells formed

cloning efficiency of MOSE

3Y10 3Y1, K-ras

15.64

T-Ag-MOSE

Fig. 3. Cloning efficiency in soft agar of MOSE. The T-Ag-MOSE cells formed small colonies with only 50% of the efficiency of 3YI, K-ras used as the positive control.

p53-def-MOSE

a a

2

4

6

8

10

12

14

16

(0/0)

728 . M. Kido and M. Shibuya Table 1. Tumorigenicity of MOSE cell lines in nude mice Injected cell line

T-Ag-MOSE p53-def-MOSE

a

Route of injection subcutaneous

intraperitoneal

3/8 0/6

11112 0/6

MOSE: mouse ovarian surface epithelium

Capacity for tumorigenicity

First we preliminarily injected T-Ag-MOSE subcutaneously to one nude mouse, and intraperitoneally to another two mice. Tumors were first detected after about two months in the two mice which were injected intraperitoneally. Then tumorigenicity of T-Ag-MOSE and p53-def-MOSE were further examined. The results are summarized in Table 1. T-Ag-MOSE cells were tumorigenic at high efficiency, but as for p53-def-MOSE, no tumors were formed even after six months of the injection.

b

Analysis of tumors formed in mice injected with T-Ag-MOSE cells

c

Fig. 4. Histopathology of tumors in nude mice injected MOSE. a: T-Ag-MOSE-injected tumor (x200). The tumor cells were arranged irregularly and it was difficult to identify the origin of the tumor cells. b: Myxomatous part detected in T-Ag-MOSE-injected tumor (x40). Tumor cells destructively invaded across the basement membrane into the skin or muscle tissue. c: Osteoids detected in T-Ag-MOSE-injected tumor (x 100).

The tumors derived from T-Ag-MOSE were white or yellow-white, smooth, hard and slightly adhesive with surrounding tissues. Bloody ascites were detected in three mice that were intraperitoneally injected. In two mice, intraperitoneal tumors were formed under the diaphragm, which invaded into the pleural cavity. Distant metastases were not detected macroscopically in any mice. Microscopic examination of these tumors was as follows (Fig. 4a). Tumor cells destructively invaded across the basement membrane into the skin or muscle tissue. The tumor cells had abundant cytoplasm, and oval or polygonal, chromatin-rich, atypical nuclei. The cells were arranged irregularly, and it was difficult to identify the origin of the tumor cells. In some small parts of the tumors, myxomatous (Fig. 4b) or epithelial-like arrangements were seen. In several tumors, differentiation to mesothelial tissues was detected. Osteoid tissues were seen in four tumors (Fig. 4c). Angiosarcomatous atypical vessels were detected in one tumor.

Discussion small colonies at nearly half the efficiency when compared to that of the positive control. On the other hand, p53-def-MOSE did not form any colonies.

Surface epithelial cells of the human ovary are repeatedly disarranged during ovulation, and the genetic errors can accumulate by cell division necessary for the repairment [22J. This 'incessant ovulation hypothesis' in

Mouse Ovarian Epithelial Cell Lines . 729 ovarian tumorigenesis has been supported by epidemiological and experimental data [12,28]. The pathway toward tumor formation is still unclear and it is necessary to clarify each step. Many investigators have reported genetic changes in human ovarian tumor samples [4,10, 24, 30, 32], such as activation of H-ras, K-ras, c-erbB2 and c-myc as well as inactivation of p53. By using polymorphic DNA markers, several loci having loss of heterozygosity have been detected which could be possible sites of tumor suppressor genes [7, 27]. But every human tumor sample has a different genetic background and it is difficult to study the specific role of each alteration in oncogenes or tumor suppressor genes. Several investigators have reported the development of various culture systems for the OSE of rats, human, or rabbits since 1980 [2, 9, 11,23,25]. By spontaneous immortalization or using DNA tumor virus antigen, some cell lines of OSE have been established [1, 11, 23]. Since the genetic background of mice has been extensively studied and various strains of mice are useful in the analysis of multi-step carcinogenesis, we have tried to culture OSE of mice, and established cell lines involving two different methods of immortalization. One was by transfection of the SV40 large T antigen DNA, similar to the method for the immortalization of human OSE [23]. Another was by using the OSE of mice homozygously knocked out of the p53 gene. Fibroblasts or epithelial cells of the mouse of this strain have been reported to immortalize spontaneously [31]. Characterization of the obtained MOSE cells is difficult because they lack specific cell markers. But MOSE lines we obtained showed a typical cobblestone-like epithelial arrangement, were laminin positive, and had a specific growth pattern. Several reports described that the growth of OSE of human or rat was promoted by the addition of EGF or hydrocortisone [9, 13, 29]. Insulin also had a growth promoting effect on human fetal OSE, while estrogen, progesterone, follicle stimulating hormone and Mullerian inhibiting substance had no effect on growth [9]. By immunohistochemistry, EGFR was detected in human OSE [51], and the glucocorticoid receptor was detected in rat OSE [13]. In general, growth of many cells such as fibroblasts are inhibited by hydrocortisone, while only certain kinds of cells including OSE exhibit a growth promoting effect. The extent of transformation can be tested by fociforming in monolayer culture, anchorage-independent growth in soft agar, and tumorigenicity in nude mice. TAg-MOSE retained contact inhibition, made fewer colonies in soft agar than positive control, but made malignant tumors in nude mice in high efficiency. So TAg-MOSE was transformed by the introduction of the SV40 T antigen. On the other hand, p53-def-MOSE did not show any positive results when tested for transformation. In T-Ag-MOSE cells, two pathways, Rb and p53 are thought to be inactivated by the T-Ag, whereas

p53-def-MOSE cells carry only the p53-deficiency. Therefore, at least in the case of mice, inactivation of p53 alone is not sufficient for the development of ovarian carcinoma. Tumor cells from T-Ag-MOSE injected mice showed some differences from their original T-Ag-MOSE cells. If the tumors formed in nude mice had a histological resemblance to that of human ovarian tumors, it could be a useful in vitro model. Godwin et al. showed that repeated subculture of rat OSE caused malignant transformation and formed a nude mouse tumor that closely resembled the human serous adenocarcinoma [11]. Hoffman et al. reported that rat OSE which spontaneously immortalized were themselves non-tumorigenic, but introduction of SV40 T antigen to them caused tumor formation, that histologically resembled the human poorly differentiated serous papillary adenocarcinoma or poorly differentiated carcinoma [15]. In the same report, introduction of the SV40 T antigen into the early passage of rat OSE caused poorly differentiated sarcoma in nude mice, which was similar to T-AgMOSE-injected tumors. The phenotypic difference may be due to the extent of transformation of the cells introduced of SV40 T antigen. Some of the T-Ag-MOSE-injected tumors showed differentiation to mesothelial tissues such as osteoid tissues. Since coelomic epithelium has the potential to differentiate into various mesothelial tissues, the OSE derived from it could have a similar potential. In Lauchlan's reports (1968), it has been proposed that the female genital tract, ovaries and peritoneal cavity as a whole constitute a secondary Mullerian system [20]. In our results, heterologous mesenchymal tissues in T-Ag-MOSE-injected tumors originated from OSE. Human ovarian sarcomas are relatively rare but several cases of tumors containing heterologous mesenchymal tissues have been reported [26]. The origin of the tissues was not clear, but it could have been derived or originated from OSE considering our results. T-AgMOSE and p53-def-MOSE can be used to assay the transforming activity of other known oncogenes. It is possible to study the system of signal transduction of OSE using these cells. Further, by transfection of the human ovarian tumor DNA fragment, it is possible to detect transforming activity, as an unknown gene important for ovarian tumorigenesis could be isolated. Thus T-Ag-MOSE and p53-def-MOSE are useful systems to study the role of various genes in ovarian tumor development.

Acknowledgements. We thank Dr. Shinichi Aizawa for the kind supply of the p53 deficient mouse, Dr. Atsuhiko Sakamoto, Dr. Kuniko Iihara, and Dr. Kouichi Suzuki for their help in histological analysis, and Dr. Takashi Kawana and Dr. Lata Seetharam for reviewing the manuscript.

730 . M. Kido and M. Shibuya This work was supported by a Grant-in-Aid (09771258) from the Ministry of Education, Science and Culture of Japan. 16.

References 17. I. Adams AT, Auersperg N (1985) A cell line, ROSE 199, derived from normal rat ovarian surface epithelium. Exp Cell Bioi 53: 181-188 2. Auersperg N, Siemens CH, Myrdal SE (1983) Culture and characterization of human ovarian surface epithelial cells. J Cell Bioi 97: 351 a 3. Auersperg N, Siemens CH (1984) Human ovarian surface epithelium in primary culture. In Vitro 20: 743-755 4. Berchuck A, Kamel A, Whitaker R (1990) Overexpression of Her-2/neu is associated with poor survival in advanced epithelial ovarian cancer. Cancer Res 50: 4087-4091 5. Berchuck A, Rodriguez GC, Kamel A, Dodge RK, Soper JT, Clarke-Perason DL, Bast Jr. RC (1991) Epidermal growth factor receptor expression in normal ovarian epithelium and ovarian cancer: 1. Correlation of receptor expression with prognostic factors in patients with ovarian cancer. Am J Obstet Gynecol164: 669-674 6. Berthois Y, Katzenellenbogen JA, Katzenellenbogen BS (1986) Phenol red in tissue culture media is a weak estrogen: implications concerning the study of estrogen-responsive cells in culture. Proc Natl Acad Sci USA 83: 2496-2500 7. Cliby W, Ritland S, Hartmann L, Dodson M, Halling KC, Keeney G, Podratz KC, Jenkins RB (1993) Human epithelial ovarian cancer allelotype. Cancer Res 53: 2393-2398 8. Disaia PJ, Creasman WT (1993) Clinical Gynecologic Oncology. fourth edition: 333-340. Mosby-Year Book, St. Louis 9. Dubeau L, Velicescu M, Sherrod AE, Schreiber G, Holt G (1990) Culture of human fetal ovarian epithelium in a chemically-defined, serum-free medium: a model of ovarian carcinogenesis. Anticancer Res 10: 1233-1240 10. Feig L, Bast R, Knapp R (1984) Somatic activation of rasK gene in a human ovarian carcinoma. Science 223: 698-700 11. Godwin AK, Testa JR, Handel LM, Liu Z, Vanderveer LA, Tracey PA, Hamilton TC (1992) Spontaneous transformation of rat ovarian surface epithelial cells: association with cytogenetic changes and implications of repeated ovulation in the etiology of ovarian cancer. J Natl CancerInst84:592-601 12. Greene M, Clark J, Blayney D (1984) The epidemiology of ovarian cancer. Semin Oncolll: 209-226 13. Hamilton TC, Davis P, Griffiths K (1983) Steroid-hormone receptor status of the normal and neoplastic ovarian surface germinal epithelium. In: Greenwald GS, Terranova PF (Eds) Factors regulating ovarian function, pp 81-85. Raven Pr., New York 14. Health and Welfare Statistics Association (1994) Kousei no shihyou: kokumin eisei no doukou, Vol 41: 421-422. Health and Welfare Statistics Association, Tokyo 15. Hoffman AG, Burghardt RC, Tilley R, Auersperg N (1993) An in vitro model of ovarian epithelial carcino-

18.

19.

20. 21. 22. 23.

24.

25.

26. 27.

28.

29.

30.

31.

32.

genesis: changes in cell-cell communication and adhesion occurring during neoplastic progression. Int J Cancer 54:828-838 Kano H, Matsui Y, Makino Y, Okudaira Y (1990) Primary culture and characterization of human ovarian surface epithelium. Acta Obst Gynaec Jpn 42: 45-52 Kawai S, Nishizawa M (1984) New procedure for DNA transfection with polycation and dimethyl sulfoxide. Mol Cell Bioi 4: 1172-1174 Kim DW, Uetsuki T, Kaziro Y, Yamaguchi N, Sugano S (1990) Use of the human elongation factor la promoter as a versatile and efficient expression system. Gene 91: 217-223 Kimura G, Itagaki A, Summers J (1975) Rat cell line 3y1 and its virogenic polyoma- and sv40-transformed derivatives. Int J Cancer 15: 694-706 Lauchlan SC (1968) Conceptual unity of the mullerian group tumors. Cancer 22: 601-610 Long JH (1940) Growth in vitro of ovarian germinal epithelium. Contrib Embryol172: 89-97 Lowry S, Russell H, Hickey I (1991) Incessant ovulation and ovarian cancer. Lancet 338: 1544-1545 Maines-Bandiera SL, Kruk PA, Auersperg N (1992) Simian virus40-transformed human ovarian surface epithelial cells escape normal growth controls but retain morphogenetic responses to extracellular matrix. Am J Obstet Gynecol167: 729-735 Marks JR, Davidoff AM, Kerns BK (1991) Overexpression and mutation of p53 in epithelial ovarian cancer. Cancer Res 51: 2979-2984 Nicosia SV, Johnson JH, Streibel EJ (1984) Isolation and characterization of rabbit ovarian mesothelium (surface epithelium). Int J Gynecol Pathol3: 348-360 Russell P, Bannatyne P (1989) Surgical pathology of the ovaries, pp 468-473. Churchill Livingstone, New York Sato T, Saito H, Morita R, Koi S, Lee JH, Nakamura Y (1991) Allelotype of human ovarian cancer. Cancer Res 51: 5118-5122 Scully RE (1979) Tumors of the ovary and maldeveloped gonads, pp 53-54. Armed Forces Institute of Pathology, Washington, D.C. Siemens CH, Auersperg N (1988) Serial propagation of human ovarian surface epithelium in tissue culture. J Cell Physiol134: 347-356 Smith DM, Groff DE, Pokul RK, Bear JL, Delgado G (1989) Determination of cellular oncogene rearrangement or amplification in ovarian adenocarcinomas. Am J Obstet Gynecol161: 911-915 Tsukada T, Tomooka Y, Takai S, Ueda Y, Nishikawa S, Yagi T, Tokunaga T, Takeda N, Suda Y, Abe S, Matsuo I, Ikawa Y, Aizawa S (1993) Enhanced proliferative potential in culture of cells from p53-deficient mice. Oncogene 8:3313-3322 Zhou OJ, Gonzalez-Cadavid N, Ahuja H, Battifora H, Moore GE, Cline MK (1988) A unique pattern of protooncogene abnormalities in ovarian adenocarcinoma. Cancer 62: 1573-1576

Received: April 3, 1998 Accepted in revised form: June 9, 1998