UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Serial cultivation of normal human ovarian surface epithelium in tissue culture : a potential model… Siemens, Craig Henry 1986

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1986_A6_7 S53.pdf [ 16.36MB ]
Metadata
JSON: 831-1.0096807.json
JSON-LD: 831-1.0096807-ld.json
RDF/XML (Pretty): 831-1.0096807-rdf.xml
RDF/JSON: 831-1.0096807-rdf.json
Turtle: 831-1.0096807-turtle.txt
N-Triples: 831-1.0096807-rdf-ntriples.txt
Original Record: 831-1.0096807-source.json
Full Text
831-1.0096807-fulltext.txt
Citation
831-1.0096807.ris

Full Text

SERIAL CULTIVATION OF NORMAL HUMAN OVARIAN SURFACE EPITHELIUM IN TISSUE CULTURE: A POTENTIAL MODEL FOR OVARIAN CARCINOGENESIS By CRAIG HENRY SIEMENS B.Sc, The University of Victoria, 1981 A THESIS SUBMITTED. IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of Anatomy We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1986 (Q) Craig H. Siemens, 1986. In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree a t the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department of A r° fiTo  The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date ABSTRACT Most human ovarian cancers are thought to arise i n the surface epithelium of the ovary. However, few models for the study of this tissue exist and no methods for i t s propagation i n culture are available. This report addresses two problems encountered previously in the culture of normal human ovarian surface epithelium (OSE): (i) limited growth; and ( i i ) d i f f i c u l t i e s i n c e l l identification because of the pleomorphism of cultured human OSE c e l l s . The poor growth potential exhibited by human OSE i n culture was greatly improved by the development of a method that permits, for the f i r s t time, the seri a l cultivation and clonal growth of human OSE i n culture. To achieve this, OSE cells were cultured from premenopausal ovarian tissue obtained at surgery. Growth of OSE cells was compared in several culture media: Medium 199 (199), MCDB 202 (202), Waymouth's 752/1 (WM), and 199/202 ( l : l ) , with 5, 15, or 25$ fetal bovine serum (FBS) and with/without epidermal growth factor (EGF, 20 ng/ml) and hydrocortisone (HC, 0.4 ug/ml). Proliferation was greater in 199/202 ( l : l ) with 15$ FBS, EGF, and HC, than i n the same medium with 5% FBS or Waymouth's medium with any supplement. At clonal densities (100 cells/60 mm culture dish) with EGF/HC, seeding efficiencies in most media ranged from 30 to 50$. Cloning efficiencies differed significantly between media and ranged from 1.0% to 13%. In the medium optimal for maximal proliferation, 15$ FBS/199/202 (1:l)/EGF/HC, OSE cells were subcultured up to 10 times and underwent 20 to 25 population doublings. The population doubling time during the log phase of growth was approximately 48 hours. Seeding - i i i -efficiencies of up to 53% and cloning efficiencies of up to 13% were obtained. The r a d i i of outgrowths were significantly greater than in the other media. Early passage OSE cells reversibly modulated from an epithelial morphology, i n medium lacking EGF/HC, to an elongate morphology in medium containing EGF/HC. The cytokeratin filaments of OSE modulated from a strongly-stained, filamentous pattern i n the absence of EGF/HC, to a more diffuse, lightly-stained pattern with EGF/HC. In contrast to OSE cells grown in 15FBS/199/202/EGF/HC, OSE cells grown in WM with 5 or 15% FBS were near-stationary with a flattened, epithelioid morphology. Thus, culture conditions have been defined for human OSE that w i l l allow the application of short-term carcinogen assays that require either rapid or absent proliferative activity. As a model to examine the pleomorphism exhibited by human OSE i n primary culture, a continuous c e l l line of rat ovarian surface epithelium (ROSE 239) was cultured and cloned in WM with 10% FBS. The use of primary cultures of human OSE for this purpose was impractical due to the high degree of v a r i a b i l i t y i n explants and OSE outgrowths in primary culture. In clonal culture, ROSE 239 and two clones derived from i t (R0SE-B2, and R0SE-C4) formed colonies with three morphological forms: compact, flattened and atypical. The morphologies exhibited by these colonies were similar to those previously observed in human OSE cells in primary culture (Auersperg et a l , 1984), and appeared to modulate between the different morphological forms. R0SE-C4 grew as a compact, cobble-stone-like monolayer in sparse and confluent cultures. R0SE-B2 displayed a more fusiform growth pattern i n sparse culture but formed a cobble-stone-like monolayer when confluent. Confluent cultures of R0SE-C4 - iv -remained contact-inhibited whether fed continuously or starved for up to 2 months. Confluent R0SE-B2 cultures continued to proliferate when fed continuously and formed multilayered ridges,but starving resulted i n c e l l death which began within 2 weeks. Both R0SE-C4 and R0SE-B2 were variably positive for cytokeratin filaments. Thus, R0SE-C4 cells appear to behave like normal OSE i n vivo, while R0SE-B2 cells appear to have lost their responsiveness to the mechanisms that control the growth of normal OSE c e l l s . This culture model could perhaps be used to provide clues about the mechanisms controlling OSE c e l l growth, the changes that occur i n OSE during neoplasia, and the pleomorphic nature of OSE-derived tumours. ACKNOWLEDGEMENTS I would like to thank Dr. Nelly Auersperg for her advice, guidance and support throughout this research. I would like to thank Dr. C.E. Slonecker of the Department of Anatomy for the f a c i l i t i e s to carry out this research. I wish to express my appreciation to Drs. G.W. Korn, L.S. Deane, H.J. Pendleton, and I. Dunnett, and the operating room staff at Vancouver General, Shaughnessy, and University Health Sciences hospitals, for their cooperation in supplying ovarian biopsy specimens. To Mrs. J. Black, Mrs. C. Burr, and Ms. P. Kruk I extend thanks for expert technical help. A special thanks goes to everyone else who tolerated me through the course of this research. This research was supported by grants from the National Cancer Institute of Canada to Dr. N. Auersperg and a University of B.C., Faculty of Medicine Summer Studentship. - v i -TABLE OF CONTENTS ABSTRACT i i ACKNOWLEDGMENTS v TABLE OF CONTENTS v i LIST OF FIGURES ix LIST OF TABLES x i ABBREVIATIONS x i i INTRODUCTION 1 (A) OVERVIEW 1 (1) Development of Culture Conditions for Human OSE 2 (2) OSE Pleomorphism 3 (B) HUMAN OVARIAN SURFACE EPITHELIUM 4 (1) OVARIAN SURFACE EPITHELIUM IN VIVO 4 (a) Embryology 4 (b) Adult Morphology and Function 6 (2) HUMAN OVARIAN CANCER 7 (3) MODELS OF HUMAN OVARIAN CANCER 12 (a) Spontaneous Tumours 12 (b) Animal Models 12 (c) Culture Models 13 (4) TISSUE CULTURE OF EPITHELIA 15 (5) SHORT-TERM CARCINOGEN ASSAYS 17 (6) RATIONALE 21 (C) RAT OVARIAN SURFACE EPITHELIUM 24 (1) BACKGROUND 24 (2) RATIONALE 26 - v i i -METHODS AND MATERIALS < 2 8 (A) HUMAN OVARIAN SURFACE EPITHELIUM .....28 (1) TISSUES 28 (2) TISSUE CULTURE 28 (a) Tissue Preparation 28 (b) Media and Culture Conditions 32 (c) Evaluation of Culture Conditions 34 (i) OSE Outgrowth 34 ( i i ) Growth of OSE in Mass Culture 35 ( i i i ) Growth Curves 35 (iv) Clonal Growth 36 (3) HUMAN PERITONEAL MESOTHELIUM 36 (4) IMMUNOFLUORESCENCE MICROSCOPY 37 (5) THYMIDINE INCORPORATION ASSAYS 38 (6) EFFECT OF 17S-ESTRADIOL ON DNA REPLICATION 39 (7) PHOTOGRAPHY 39 (8) STATISTICS 40 (B) RAT OVARIAN SURFACE EPITHELIUM 40 RESULTS 42 (A) HUMAN OVARIAN SURFACE EPITHELIUM 42 (l) EFFECT OF CULTURE CONDITIONS ON OSE GROWTH 42 (a) OSE Outgrowth in Primary Cultures 42 (b) Growth i n Mass Culture 51 (c) Growth Curves 56 (d) Clonal Growth 67 - v i i i -(2) EFFECT OF EGF AND HYDROCORTISONE ON OSE MORPHOLOGY 80 (a) Primary Cultures 80 (b) After Subculture 85 (3) EFFECT OF EGF AND HC ON KERATIN FILAMENT ORGANIZATION..94 (4) EFFECT OF EGF AND HC ON CULTURE LIFESPAN 99 (5) EFFECT OF 17B-ESTRADI0L ON OSE PROLIFERATION 100 (6) COMPARISON WITH HUMAN PERITONEAL MESOTHELIUM 100 (B) RAT OVARIAN SURFACE EPITHELIUM 103 (1) ROSE 239 CLONES 103 (2) MORPHOLOGY OF CLONES OF R0SE-C4 and R0SE-B2 I l l DISCUSSION 123 (A) HUMAN OVARIAN SURFACE EPITHELIUM 123 (1) EFFECT OF CULTURE MEDIA ON OSE CELL GROWTH 123 (2) EFFECT OF EGF AND HC ON OSE CELL GROWTH 127 (3) SERUM SUPPLEMENTATION 131 (4) OSE MORPHOLOGY IN CULTURE 132 (5) APPLICATIONS OF HUMAN OSE IN CULTURE 137 (a) Short-term Carcinogen Assays 137 (b) Other Applications 139 (B) RAT OVARIAN SURFACE EPITHELIUM 140 SUMMARY 144 BIBLIOGRAPHY 146 APPENDIX I. SOURCES OF MATERIALS 162 APPENDIX II. IMMUNOFLUORESCENCE STAINING OF KERATIN FILAMENTS...164 APPENDIX III. HOECHST DNA ASSAY 166 APPENDIX IV TRITIATED THYMIDINE INCORPORATION INTO DNA 168 APPENDIX V DEXTRAN CHARCOAL TREATED SERUM 170 - ix -LIST OP FIGURES Figure 1. Culture method for human ovarian surface epithelium 30 Figure 2. Outgrowth of human ovarian surface epithelium in primary culture 45 Figure 3« Outgrowth of human ovarian surface epithelium in primary culture (Case I) 47 Figure 4- Relative outgrowth of human ovarian surface epithelium in primary culture 53 Figure 5. Growth of human ovarian surface epithelium in mass culture 55 Figure 6. Growth curve for human ovarian surface epithelium: Passage 3 59 Figure 7- Growth curve for human ovarian surface epithelium: Passage 5 61 Figure 8. Growth curve for human ovarian surface epithelium: Passage 7 63 Figure 9. Morphology of human OSE cells in growth curve experiments..66 Figure 10. Effect of culture media on relative seeding efficiency of human OSE 69 Figure 11. Effect of culture media on seeding efficiency of human OSE 71 Figure 12. Effect of culture media on relative cloning efficiency of human OSE 74 Figure 13. Effect of culture media on cloning efficiency of human OSE 76 Figure 14. Morphology of clonal colonies of human ovarian surface epithelium. 79 Figure 15. Morphology of human OSE in primary culture 82 Figure 16. Effect of EGF and HC on the morphology of human OSE in primary culture 84 Figure 17. Effect of EGF and HC on the morphology of early passage human OSE cells 87 Figure 18. Effect of EGF and HC on the morphology of early passage human OSE cells 89 Figure 19. Morphology of human ovarian surface epithelium in Waymouth's medium 91 Figure 20. Effect of EGF and HC on the morphology of early passage human peritoneal meosthelial cells (Strain LP-9) 93 Figure 21. Further observations on human OSE morphology in culture....96 Figure 22. Effect of EGF and HC on keratin filament expression in human OSE cells 98 Figure 23. Effect of estradiol (E 2), EGF and HC on (^H)TdR incorporation i n human OSE cells 102 Figure 24. Morphological forms exhibited by clones of ROSE 239 105 Figure 25. Morphology of R0SE-C4 and R0SE-B2 in sparse and confluent cultures Figure 26. Confluent cultures of R0SE-C4 maintained without addition of culture medium 110 - X -Figure 27. Confluent cultures of R0SE-B2 maintained without addition of culture medium 113 Figure 28. Confluent cultures of R0SE-C4 cells fed biweekly 115 Figure 29- Confluent cultures of R0SE-B2 cells fed biweekly 117 Figure 30. Growth patterns of clones of R0SE-C4 and R0SE-B2 119 Figure 31. Morphology of a growing ROSE 239 clone. 122 LIST OF TABLES TABLE I. Culture media and serum concentrations tested 33 TABLE II. Growth of OSE in mass culture: Visual rankings of c e l l growth on different media ..43 TABLE III. Effect of culture conditions on OSE growth i n primary-culture 49 TABLE IV. Effect of culture conditions on OSE growth pattern i n primary culture (Case I) .....57 TABLE V. Cloning efficiency and size of colonies of OSE in clonal culture 77 - x i i -ABBREVIATIONS OSE - human ovarian surface epithelium (or epithelial). ROSE - rat ovarian surface epithelium (or epithelial). ROSE 239 - continuous c e l l line of rat ovarian surface epithelium. R0SE-B2 - clone of ROSE 239-R0SE-C4 - clone of ROSE 239-WM - Waymouth's 752/1 Medium. 199 - Medium 199 . 202 - MCDB 202 Medium. 199/202 - 1:1 mixture of 199 and 202. FBS - Fetal bovine serum. CS - Calf serum. EGF - Epidermal growth factor. HC - Hydrocortisone. EH - Epidermal growth factor/hydrocortisone PDGF - Platelet-derived growth factor. E 2 - 17e-estradiol [l,3 , 5(lO)-estratrien - 3 , 176-diol]. HSD - Hydroxysteroid dehydrogenase dH20 - Dist i l l e d water. ddH20 - Double-distilled water. SCE - Sister chromatid exchange. (^H)TdR - Tritiated thymidine. NRS - Normal rabbit serum Ab - antibody 1 ° - primary 2° - secondary INTRODUCTION (A) OVERVIEW The human ovary i s a small but highly complex organ which undergoes cyc l i c a l structural and biochemical changes related to f o l l i c u l a r maturation, ovulation and steroidogenesis. Investing the ovary and intimately involved in i t s function and embryology, i s a simple epithelium, the ovarian surface epithelium (ovarian mesothelium, "germinal" epithelium). Although i t s contribution to the total ovarian c e l l mass i s small, the ovarian surface epithelium (OSE) i n humans is thought to give rise to over 80$ of a l l ovarian cancers. These "common epithelial tumours" of the ovary are responsible for considerable morbidity and mortality i n women in the industrialized countries of the world, where ovarian cancer i s now ranked f i f t h as a fat a l form of cancer. Patients with ovarian cancer have a poor prognosis and low survival rate, due mainly to a lack of suitable methods for detection of the cancer when i t i s s t i l l curable by surgery. Despite the prevalence of ovarian cancer, knowledge concerning the pathogenesis and etiology of ovarian cancer i s fragmentary and inconclusive. However, there i s considerable evidence implicating environmental agents in the genesis of ovarian neoplasms, although no specific carcinogenic agents have been identified. In spite of i t s c l i n i c a l importance, no useful experimental models for the study of human OSE exist. As a result, the biology of OSE i s poorly understood and evidence for i t s role i n carcinogenesis i s based almost entirely on histological examination of c l i n i c a l specimens. The lack of a useful model of human OSE has recently prompted our laboratory to develop procedures to isolate, characterize, and cultivate human OSE i n tissue culture. This report addresses two problems encountered previously in the culture of human OSE: (l) - 2 -the limited growth and culture lifespan of human OSE; and (2) the pleomorphism exhibited by human OSE in primary culture. (l) Development of Culture Conditions for Human OSE Although animal models are useful for studies of some types of cancer, the relevance of i n vivo animal models for the study of ovarian cancer i s questionable. In animals, ovarian tumours rarely arise i n the surface epithelium and definitive evidence for carcinogenesis i n the OSE in these models i s lacking. Recently, several methods have been reported for the culture of the OSE of the rat and rabbit, but again, because of species differences, their usefulness as models of human ovarian cancer i s limited. Thus, experimental approaches to the study of human OSE-derived cancers have been hampered by the lack of a suitable culture model - namely, normal human OSE in culture. Prior to the work done in our laboratory, attempts to culture human OSE have been unsuccessful (Wu et a l . , 1982), presumably because of d i f f i c u l t i e s i n separating the OSE cells from other, predominant ovarian c e l l types, and because the cellular growth requirements of normal human OSE were unknown. In a preliminary report (Auersperg et a l . , 1984), a method to obtain relatively pure primary cultures of human OSE was described. With this method primary cultures of OSE have been established, and OSE can be distinguished from other similar c e l l types in culture. However, under the conditions used OSE growth was limited and unpredictable. The cells could be maintained in culture for only a few weeks, and could be subcultured only once or twice before senescence. Consequently, the main objective of the present research was to define culture conditions in which normal human OSE cells could be grown reproducibly and i n sufficient numbers to permit investigation into ovarian carcinogenesis and biology. With culture conditions that allow serial - 3 -cultivation of human OSE in vitro, this model system could be invaluable i n the study of ovarian biology and carcinogenesis. The literature describing the types, classification, diagnosis, treatment, biology, etiology, and epidemiology of ovarian cancer i s immense. Since much of this information is only indirectly related to this research, the following review largely ignores pathological and c l i n i c a l aspects of ovarian cancer. Instead, emphasis is placed on aspects of ovarian biology and tissue culture methodology relevant to the development of an i n vitro model system for the study of ovarian carcinogenesis. Also discussed are: the current views on ovarian cancer etiology; the i n vivo and i n vitro models presently available for the study of ovarian cancer; and the use of cultured mammalian cells in short-term i n vitro tests for potential carcinogens. (2) OSE Pleomorphism In primary culture, outgrowths of human OSE generally exhibit one of three morphological forms: (i) a cobble-stone arrangement of polygonal epithelial cells (compact); ( i i ) flattened epithelial c e l l s ; and ( i i i ) elongate, irregularly-shaped epithelial cells (atypical) (Auersperg et a l . , 1984). In vivo, the OSE also exhibits a range of morphological forms under both normal and pathological conditions. It is not clear whether these morphological differences are a response to prevailing physiological conditions (eg. hormones, growth factors, CO2 or 0^ tension, mechanical stress) or perhaps an expression of inherent characteristics of the multipotent epithelium from which i t arises. The morphological forms exhibited by the OSE cells may represent terminal differentiation of OSE or alternatively, OSE cells may have the capacity to modulate between different morphological forms. These morphological aspects of OSE biology are poorly understood due to the absence of suitable animal or culture models to examine them. - 4 -In t h i s report, a continuous l i n e of r a t ovarian surface e p i t h e l i a l c e l l s (ROSE-239) i s used as a model to examine the differences observed i n human OSE morphology. Ideally, a culture model f o r the study of OSE pleomorphism should u t i l i z e human OSE. Unfortunately, t h i s i s impractical because of the high degree of v a r i a b i l i t y i n the composition of ovarian explants and i n the extent of OSE outgrowths i n primary cul t u r e . ROSE-239 c e l l s appear to be a u s e f u l model f o r the study of OSE pleomorphism and the i n t e r r e l a t i o n s h i p s between d i f f e r e n t OSE c e l l forms. (B) HUMAN OVARIAN SURFACE EPITHELIUM (1) OVARIAN SURFACE EPITHELIUM IN VIVO (a) Embryology The ovarian surface epithelium i s of mesodermal o r i g i n , derived from the mesothelial l i n i n g of the intraembryonic coelom. During the 5th week of gestation, gonadal development begins. The coelomic epithelium of the urogenital ridge p r o l i f e r a t e s and d i f f e r e n t i a t e s to form the gonadal (genital) ridge (Moore, 1977). Just l a t e r a l to each urogenital ridge, the coelomic epithelium invaginates to form the Mullerian (paramesonephric) ducts from which the mucosae of the f a l l o p i a n tubes, endometrium and endocervix are derived (Hart, 1981). This anatomical r e l a t i o n s h i p i s c l i n i c a l l y important i n adult l i f e , because the histogenesis of most common e p i t h e l i a l ovarian tumors i s linked to the embryonic coelomic epithelium; metaplasia of OSE c e l l s frequently produces c e l l s which resemble the e p i t h e l i a of Mullerian duct derivatives (Parmley and Woodruff, 1974). Shortly a f t e r gonadal ridge formation, the primary sex cords develop as an ingrowth of the coelomic epithelium. At 7 weeks, i n the absence of Y-chromosomes, the primordial gonad - 5 -slowly begins to develop into an ovary. During this period, the OSE consists of a single, columnar layer of cells (Gondos, 1975)* The primary sex cords degenerate and at 12 to 16 weeks, the OSE proliferates to produce the secondary (cortical) sex cords cords which extend into the ovarian cortex (Moore, 1977; Motta and Makabe, 1982). As the secondary sex cords develop, the primordial germ c e l l s , which have migrated to the underlying mesenchyme, are incorporated into them. The cords break up into isolated c e l l clusters (primordial f o l l i c l e s ) consisting of an oogonium surrounded by f o l l i c u l a r cells derived from the OSE. Thus, the granulosa c e l l s of the adult ovary appear to be derived from the OSE of the developing ovary (Moore, 1977; Motta and Makabe, 1982). During the 4th and 5th months of fetal development, the OSE undergoes intense mitotic activity forming a multilayered epithelium (Anderson et a l . , 1976; Moore, 1977; Parmley and Woodruff, 1974). During this period, the OSE undergoes strat i f i c a t i o n , proliferation and accompanying alterations i n nuclear and cytoplasmic morphology which resemble changes seen in OSE neoplasms in adults (Gondos, 1975)* At 22 to 24 weeks, the OSE is generally reduced to a single layer with occasional multilayered areas. At birth, the OSE i s a monolayered epithelium resting on the tunica albuginea, and i t i s occasionally invaginated into the underlying cortex as crypts or cords of cells (Motta and Van Blerkom, 1980; Nicosia, 1983) and epithelial inclusion cysts may be present i n the ovarian cortex (Blaustein, 1981). The OSE i s continuous with the mesothelial lining of the peritoneal cavity at the mesovarium. However, i t i s functionally and morphologically somewhat different from other mesothelial linings (Andrews and Porter, 1973; Blaustein and Lee, 1979). - 6 -(b) Adult Morphology and Function In the adult, the OSE undergoes c y c l i c changes i n morphology and p r o l i f e r a t i v e rate r e l a t e d to processes occurring i n the ovarian cortex during the menstrual cycle (Van Blerkom and Motta, 1979)• The most notable changes occur during the extensive p r o l i f e r a t i o n and migration of the OSE during the r e p a i r of the ovarian surface a f t e r ovulation. These changes may be mediated by hormones, as i n the tissues of the embryologically-related Mullerian duct system (Heap and I l l i n g t o n , 1977; Papadaki and Beilby, 1971). Changes i n the OSE are also evident with pregnancy, menopause and age (Balboni, 1983; N i c o s i a , 1983; Papadaki and Beilby, 1971; Van Blerkom and Motta, 1979). OSE c e l l s are squamous to columnar depending on t h e i r p o s i t i o n r e l a t i v e to f o l l i c l e s , corpora lutea, and crypts (Papadaki and Beilby, 1971; Motta and Van Blerkom, 1980). Their a p i c a l surface i s covered with numerous m i c r o v i l l i and the occasional c i l i u m (Anderson et a l . , 1976; Blaustein and Lee, 1979). The c e l l s l i e on a basal lamina and are joined at t h e i r l a t e r a l borders by desmosomes, f o c a l t i g h t junctions, gap junctions, and i n t e r c e l l u l a r i n t e r d i g i t a t i o n s (Anderson et a l . , 1976; Ferenczy and Richart, 1974; Nicosia, 1983). U l t r a s t r u c t u r a l studies of OSE i n d i c a t e that these c e l l s are involved i n both secretory (rough endoplasmic reticulum, Golgi complexes, secretory v e s i c l e s , and l i p i d i n clusions) and endocytotic (endocytotic p i t s and v e s i c l e s , lysosomes, m u l t i v e s i c u l a r bodies) a c t i v i t i e s (Anderson et a l . , 1976; N i c o s i a , 1983). The OSE may produce p r o t e o l y t i c enzymes involved i n the d i s s o l u t i o n of the f o l l i c u l a r apex at ovulation (Cajander and Bjersing, 1977) and has been shown, histochemically, to be p o s i t i v e f o r e p i t h e l i a l mucin, glycosaminoglycans (Blaustein and Lee, 1979), glycoproteins (Guraya, 1980) and 17t*-hydroxysteroid dehydrogenase (HSD) (Auersperg et a l . , 1983, 1984a; Blaustein and Lee, 1979). That the OSE i s p o s i t i v e f o r 17S-HSD but lacks A5.33-HSD (the enzyme responsible for producing the substrates of 17&-HSD; i.e. testosterone and estradiol) suggests a possible role i n the processing of steroids produced i n the f o l l i c l e s and corpora lutea of the ovary, rather than a steroid synthetic function (Adams, 1981; B a i l l i e et a l . , 1966). Or, since OSE possesses estrogen receptors (Adams and Auersperg, 1981b), 17B-HSD may be required for processing internalized estrogen. The presence of smooth-surfaced and coated endocytotic pits i n the plasma membrane, and large, apical cytoplasmic extensions in regions of intense metabolic activity (e.g. over newly-formed corpora lutea) are indicative of an endocytotic function i n the OSE (Papadiki and Beilby, 1971; Van Blerkom and Motta, 1979). In addition, tracers (colloidal gold and fe r r i t i n ) are taken up from the peritoneal cavity into endocytotic vesicles and may be transported across the epithelium to the underlying stroma (Anderson et a l . , 1976; Donaldson, 1976). Human OSE cells in vivo contain several types of intermediate filaments, including cytokeratin, vimentin, and desmoplakins (associated with desmosomes) (Czernobils'ky et a l . , 1985). Several subclasses of cytokeratin filaments have been identified i n the OSE (subclasses 7, 8, 18, and 19) by gel electrophoresis. These subclasses are also expressed i n human peritoneal cells (Connell and Rheinwald, 1983), and ovarian mesotheliomas and carcinomas (Czernobilsky et a l . , 1985). (2) HUMAN OVARIAN CANCER The human ovary i s the site of some 40 types of cancer, varying widely i n virulence, and c e l l type of origin (Janovski and Panamandhan, 1973; Scully, 1977). Over the past 30 years, the mortality due to ovarian cancer has increased considerably (Greene et a l . , 1984). Presently, ovarian cancer i s - 8 -the f i f t h or sixth most frequent malignancy occurring i n American women (Heintz et a l . , 1985; Richardson et a l . , 1985a). Of the numerous histologically-distinct neoplasms that arise i n the ovary, those derived from the OSE account for 80 to 90$ of a l l ovarian cancers, including those that contribute most to cancer mortality (Feniglio, 1980; Janovski et a l . , 1973; Scully, 1977). In British Columbia i n 1981, common epithelial cancers resulted in an annual mortality rate of 8.5/100,000 women (R.P. Gallagher, pers.comm.). Although ovarian carcinomas occur less frequently than do carcinomas of the cervix or endometrium, they are responsible for more deaths than any other group of malignancies of the female genital tract (Hart, 1981; Richardson et a l . , 1985a). Patients with ovarian cancer have a poor 5-year survival rate (20-47$) (Janovski et a l . , 1973; Morrow, 1981), due mainly to the fact that about 65$ of ovarian cancers are inoperable when f i r s t diagnosed (McGowan, 1978; Morrow, 1981). Failure of diagnosis of ovarian cancer i n curable stages i s d i f f i c u l t because: (i) the ovaries are relatively inaccessible for accurate evaluation; ( i i ) ovarian carcinomas, like other visceral malignancies, do not display early symptoms; and ( i i i ) screening procedures, such as specific immunologic or biochemical markers in the blood or urine, have yet to be developed (Lingeman, 1983; Morrow, 1981; Van Nagell, 1983). As a result, at i n i t i a l surgery the majority of cases are found to have spread beyond the ovaries (Morrow, 1981). Further decreases i n survival can be attributed to the fact that most ovarian cancers respond poorly to chemotherapy and radiotherapy (Morrow, 1981; Scully, 1970; Richardson et a l . , 1985b). Neoplasms thought to be derived from the OSE are classified as 'common epithelial tumours' of the ovary i n the World Health Organization's classification of ovarian tumours (Scully, 1977). This group of neoplasms - 9 -includes serous, mucinous, endometrioid, clear c e l l and Brenners tumors, solid adenocarcinomas, carcinomas and mixed Mullerian tumors (Fenoglio, 1980; Scully, 1977). These tumours are notorious for their wide variation i n the degree of histologic differentiation that can be found i n different areas of the same neoplasm (Hart, 1981). The other malignant ovarian tumours are mainly of the germ c e l l or gonadal stromal (sex cord-stromal) type (Fox and Langley, 1981; Richarson et a l . , 1985a). Malignant neoplasms of the ovary occur at a l l ages from infancy to old age. Nonepithelial neoplasms account for most of the neoplasms in younger women, while adenocarcinomas predominate in women 25 years or older i n America and northern Europe (Lingeman, 1983). After age 40, the incidence of epithelial tumours rises rapidly u n t i l 80 years when a plateau occurs . The average age of diagnosis i s 50 to 55 years of age (Hart, 1981; Morrow, 1981). Knowledge concerning the etiology and pathogenesis of human epithelium-derived ovarian cancer i s fragmentary and inconclusive (Tunca et a l . , 1985; Woodruff, 1979). Epidemiologic studies indicate that ovarian cancer i s associated with low parity but not specifically with i n f e r t i l i t y (Richarson et a l . , 1985a), and some studies suggest a preventative effect of oral contraceptives (Casagrande et a l . , 1979). In general, i t appears that ovarian cancer increases with "ovulatory age" - the total time in a woman's l i f e during which her ovarian cycle i s not suppressed by pregnancy, lactation, or the use of oral contraceptives (Casagrande et a l . , 1979; Joly et a l . , 1974; Risen, 1983). Other observations that lend support to this notion are: (i) epithelial cancers are seen with regularity only after puberty; ( i i ) patients with gonadal dysgenesis show a low frequency of epithelial tumors; ( i i i ) there i s an increased incidence of epithelial tumours in nuns; and (iv) mammals with estrous cycles display a decreased incidence of epithelial tumours , while - 10 -animals with frequent ovulations, such as egg-laying domestic fowl, display an increased incidence of adenocarcinomas (Hart, 1981; Lingeman, 1983). Frequent ovulation i s thought to increase the frequency of formation of OSE-lined cysts which have been shown to be associated with many ovarian cancers (Hart, 1981; Sc u l l y , 1977; Ra d i s a v l j e v i c , 1977). Inclusion cysts may represent receptacles i n which carcinogens are maintained i n close approximation with OSE f o r prolonged periods of time (Parmley and Woodruff, 1974). Cramer et a l (1983a) and Cramer and Welch (1983) have suggested that estrogen or estrogen-precursors and gonadotropins may be involved i n d i f f e r e n t i a t i o n , p r o l i f e r a t i o n and eventual malignant transformation of the entrapped OSE. A l l ovarian neoplasms are rare i n the non-industrialized countries of the world, but very frequent ( e s p e c i a l l y neoplasms of e p i t h e l i a l o r i g i n ) i n highly i n d u s t r i a l i z e d countries (Cramer et a l . , 1982; Lingeman, 1983). Migrants from low incidence countries to high incidence countries u s u a l l y develop the incidence rate of the new host country within a generation or two (Lingeman, 1983). A long l i s t of environmental agents have been implicated i n human ovarian carcinogensis i n c l u d i n g : p o l y c y c l i c aromatic hydrocarbons, i n d u s t r i a l p o l l u t a n t s , and smoking (Mattison and Thorgeirsson, 1978), asbestos (Newhouse et a l 1972), t a l c (Cramer et a l . , 1982; Henderson et a l . , 1979; Longo and Young, 1979), coffee (Trichanpoulos et a l . , 1981), a n t i r u s t o i l (Jarvholm, 1981), i n f e c t i o u s agents (mumps vir u s ) (Cramer et a l . , 1983), and die t (Lingeman, 1983). In addition, some studies point to a possible f a m i l i a l and genetic ( C h r i s t i a n , 1971; Dozois et a l . , 1971; Fraumeni, 1975; Kimbrough, 1929; Liber, 1950; Lynch et a l . , 1981), and r a c i a l and geographic (Lingeman, 1974; Lingeman, 1983) f a c t o r s . Mesothelial c e l l s of the p e l v i c and ovarian surfaces are unique i n females i n that environmental substances applied to the perineal region may reach them - 11 -by upward passage through the interconnected lumens of the genital tract organs. Agents such radio-opaque dyes, creams, suppositories, carbon particles (Egli and Newton, 1961) and radioactive materials (Venter and Iturralde, 1979) may reach the ovarian surface i n a manner analagous to the migration of spermatozoa or gonococci (Tunca et a l . , 1985)• As well, endogenous fluids and debris released from the endometrium and fallopian tubes may reach these epithelia (Hart, 1981). Cramer and Welch (1983) attempted to combine these different etiologic factors into one theory. In this theory, the f i r s t step toward malignant growth i s the formation of inclusion cysts by entrapment of surface epithelium i n the ovarian stroma. The second step includes differentiation, proliferation and sometimes malignant transformation of the epithelium lining the inclusion cysts, and occurs as a consequence of either direct stimulation by gonadotropins or indirect stimulation via steroidogenesis induced by gonadotropins. However, the f i n a l stimulus to malignant transformation remains unknown. In this model, ovarian cancer risk i s determined by a l l factors that cause: (i) inclusion cyst formation; ( i i ) high estrogen from extraglandular sources; and/or ( i i i ) high gonadotropin levels. Consequently, factors that prevent one or more of those stimuli may protect against this disease. Conditions which predispose to inclusion-cyst formation are not f u l l y understood. Ovulation has been proposed as the major mechanism by which inclusion-cysts are aquired during adult l i f e (Radisavljevic, 1977). However, inclusion cysts have also been observed in ovaries prior to menarche (Blaustein, 1981). The direct action of foreign bodies such as talc on the OSE may also enhance inclusion-cyst formation (Cramer and Welch, 1983)* - 12 -(3) MODELS OF HUMAN OVARIAN CANCER There i s no direct evidence that human ovarian carcinomas arise i n the OSE. OSE has been implicated as the source of most ovarian cancers primarily as a result of frequent histopathological demonstration of premalignant and early malignant changes in the OSE of human ovaries, sometimes in direct continuation with f u l l y malignant tumours. As well, the identification of a spectrum of pathological changes, from benign to malignant, i n conjunction with epithelium-lined inclusion cysts, provides further support for this contention (Motta et a l . , 1980; Scully, 1977; Tunca et a l . , 1985). In the study of OSE-derived malignancies, three types of biological models have been uti l i z e d : (i) spontaneously occurring epithelial tumours in animals; ( i i ) in vivo experimental models; and ( i i i ) i n vitro models. (a) Spontaneous Tumours Spontaneously occuring tumours in animal species have contributed l i t t l e to the understanding of OSE-derived malignancies. In species other than man, ovarian tumours generally arise in granulosa, theca, stromal or germ cells and rarely in the OSE (Cramer and Welch, 1983; Hamilton et a l . , 1984; Tunca et a l . , 1985). The exceptions are, OSE-derived tubular adenomas in mice with genetic deficiency of germ cells (Murphy, 1980), adenocarcinomas i n aging hens and adenomas and carcinomas in dogs (Cotchin, 1977). (b) Animal Models Many attempts have been made to produce ovarian neoplasms in experimental animals (Woodruff, 1975; Woodruff, 1976). These usually resulted in the generation of uncommon stromal or germ c e l l tumours and not the prevalent adenocarcinomas of women (Murphy, 1980; Woodruff, 1975). Ovarian epithelial - 13 -tumours have been induced i n rodents by i r r a d i a t i o n ( M i l l e r , 1972), by implantation of chemically impregnated sutures i n t o ovarian t i s s u e of rat s (Jacobs et a l . , 1984; Kato et a l . , 1973; Tunca et a l . , 1985; Sekiya et a l . , 1979), i n t r a p e r i t o n e a l i n j e c t i o n of asbestos (Graham and Graham, 1967), transplantation of embryos to extrauterine s i t e s i n mice (Damanov and S o l t e r , 1976), and by prenatal exposure to urethane (Nomura, 1973). Also described are p a p i l l a r y changes i n the OSE of rats i n response to t a l c i n j e c t i o n into the ovarian bursa (Hamilton et a l . , 1984b). The mechanism of a c t i o n i n these models i s incompletely understood. Experimental techniques that cause high gonadotropin production are associated with ovarian tumors i n animals. However, these tumours generally o r i g i n a t e from the stromal c e l l s of the ovary and do not duplicate the complex e p i t h e l i a l patterns observed i n human ovarian tumours (Cramer and Welch, 1985). (c) Culture Models In v i t r o models f o r the study of e p i t h e l i a l ovarian malignancies include the culture of OSE derived from both normal and malignant t i s s u e s . In several recent reports, c e l l l i n e s from human carcinomas have been established and t h e i r i n v i t r o properties characterized (Buick et a l . , 1985; Hamilton et a l . , 1984a; Woods et a l . , 1979)* Such models are us e f u l f o r describing the b i o l o g i c a l properties of ovarian cancers, and f o r assessing the u t i l i t y of possible chemotherapeutic drugs, but contribute l i t t l e to our understanding of the genesis of ovarian tumours. Only a handful of studies over the past 30 years have dealt with the culture of normal ovarian surface epithelium. These studies have u t i l i z e d explants of whole ovaries or ovarian fragments obtained from mice (Martinovich, 1937, 1984; Long, 1940; Nic o s i a et a l . , 1984), rabbits (Champy, - 14 -1926; Nicosia et a l . , 1984), chicken embryos (Pano and Garofolini, 1927; Nicosia et a l . , 1984), human fetus (Gallard, 1953; Nicosia et a l . , 1984) and canary embryos (Wolff and Haffen, 1952; Nicosia et a l . , 1984). In these studies, OSE growth was limited and OSE cells were poorly characterized. With the increasing c l i n i c a l importance of ovarian cancer, and with recent advances i n c e l l isolation and culture techniques, the OSE of several mammalian species has been grown successfully and characterized in culture. In 1980, two groups independently published procedures to culture the OSE of the rat (Hamilton et a l . , 1980; Adams and Auersperg, 1980). Rat OSE cells i n culture grow rapidly, can be readily passaged (Hamilton et a l . , 1980; Adams and Auersperg, 1981a), and often form continuous c e l l lines (Adams and Auersperg, 1981a). This model has provided evidence that the OSE cells have estrogen receptors and HSD activity (Adams, 1981; Adams and Auersperg, 1981a, 1981b and 1983; Hamilton et a l . , 1982;), and that they are susceptible to transformation by an oncogenic retrovirus (Adams and Auersperg, 1980 and 1981a). In the past three years, rabbit OSE has been successfully isolated and characterized (Nicosia and Johnson, 1984a; Nicosia et a l . , 1983; 1984, and 1985). Rabbit OSE i n culture can be passaged up to 6 times, undergoes 24 cell-population doublings during 6 weeks and has a population doubling time of 30 - 36 hours (Nicosia et a l . , 1985). A possible role of asbestos in the genesis of ovarian neoplasms has been postulated using this model (Nicosia and Johnson, 1984b). In spite of this apparent success, the fact remains that epithelial tumours rarely arise in these species, so their usefulness as models of human ovarian cancer may be limited. The f i r s t and only reports of the growth of human OSE in primary culture have recently come from our laboratory (Auersperg et a l . , 1983, 1984a, and - 15 -1984b; Siemens and Auersperg, 1986). In these studies, OSE c e l l s were grown from explants of premenopausal ovarian surface i n p l a s t i c culture dishes using Waymouth's 752/1 medium with 25$ f e t a l bovine serum. OSE c e l l s were distinguished from other morphologically-similar c e l l types (mesothelial and endothelial c e l l s ) , and from other ovarian c e l l s by antibodies to k e r a t i n filaments, histochemistry f o r 173-HSD a c t i v i t y , and by morphologic and u l t r a s t r u c t u r a l c r i t e r i a . Growth of human OSE under the above conditions i s unpredictable and l i m i t e d to primary cultures or one c e l l passage. Growth of OSE c e l l s on a v a r i e t y of substrata (glass, f i b r o n e c t i n , g e l a t i n , collagen) does not s i g n i f i c a n t l y improve the rate of OSE p r o l i f e r a t i o n (Auersperg et a l . , 1984; N. Auersperg, pers. comm.). The above studies of r a t , r a b b i t , and human OSE i n cult u r e , have described the p r o l i f e r a t i v e tendencies of these c e l l s i n v i t r o . Also, OSE c e l l s are known to p r o l i f e r a t e i n vivo during the reparative processes that follow ovulation. These observations i n d i c a t e that through the use of appropriate culture strategy, i t should be possible to su c c e s s f u l l y propagate human OSE i n long term c u l t u r e . (4) TISSUE CULTURE OF EPITHELIA Growth i n culture of normal human e p i t h e l i a l c e l l s has t r a d i t i o n a l l y been very d i f f i c u l t , due both to f i b r o b l a s t i c overgrowth and to the inadequacy of av a i l a b l e culture media. According to Tsau et a l . (1982), d i f f i c u l t i e s i n the culture of e p i t h e l i a l i e i n : ( i ) f a i l u r e to appreciate the extent to which c e l l u l a r growth requirements d i f f e r between species, and from one c e l l type to another; and ( i i ) f a i l u r e to appreciate quantitative rather than q u a l i t a t i v e differences i n growth requirements. Recently, with advances i n t i s s u e culture technology, the c e l l u l a r growth requirements f o r a v a r i e t y of human e p i t h e l i a - 16 -been defined. These include: epidermal keratinocytes (Tsau et a l . , 1982); melanocytes (Eisinger and Marko, 1982); bronchial epithelium (Lechner et a l . , 1982; Stoner et a l . , 1980); transitional epithelium of the urinary bladder (Kirk et a l . , 1985); prostatic epithelium (Kirk et a l . , 1985), endothelium (Maciag et a l . , 1981; Thornton et a l . , 1983); mammary epithelium (Hammond et a l . , 1984; Stampfer et a l . , 1980); colonic epithelium (Siddiqui and Chopra, 1986); endometrial epithelium (Kleinman et a l . , 1983; Kaufman et a l . , 1983); and hepatocytes (Butterworth et a l . , 1983). The conditions used in the culture of these epithelia vary greatly, presumably due to differences i n the nutrient requirements of differentiated epithelia. According to Ham (1974, 1984), a culture medium should ideally be designed specifically to meet the needs of the cells to be grown in i t . It should be capable of satisfying the nutrient requirements of isolated single cells (clonal growth) as well as relatively large c e l l populations, without preliminary conditioning or detoxification. This method of medium design not only improves c e l l growth, but i t may select against undesirable c e l l types, such as fibroblasts (Tsau et a l . , 1982). Ham (1974) suggests that i t may be desirable to have two media; one that provides only those factors needed for c e l l multiplication and a second that provides additional factors that might be needed for optimum expression of differentiated properties. Fetal calf serum, calf serum, or horse serum i s often added to culture media to compensate for inadequacies in the culture conditions such as suboptimal concentrations of nutrients, lack of appropriate growth factors, or inadequate substratum. However, the tremendous biochemical complexity of serum results i n many undefined and perhaps toxic components i n the culture system. As well, large serum supplements in some media enhance the growth of unwanted fibroblasts (Tsau et a l . , 1982). By modification of the culture - 17 -medium, i t i s possible to l i m i t the concentration of serum or, a l t e r n a t i v e l y , to use small amounts of dialyzed serum, r e s u l t i n g i n a medium with fewer undefined components (McKeehan and Ham, 1977; McKeehan et a l . , 1977). McKeehan et a l . (1977) and Ham (1981) suggest that i n d e f i n i n g a culture medium f o r a s p e c i f i c c e l l type, one should systematically analyze the e f f e c t s on c l o n a l growth of l i m i t i n g amounts of serum protein while making modifications to the synthetic portion of the medium. In. t h i s system, each nutri e n t added to the synthetic medium i s tested at various concentrations and every i n d i v i d u a l change that improves growth even s l i g h t l y (at l i m i t i n g concentrations of serum) i s incorporated i n t o future experiments. The concentration of the serum i s again made l i m i t i n g , and the process i s repeated. The only requirement i s that i n i t i a l l y i t must be possible to obtain a s u f f i c i e n t l e v e l of m u l t i p l i c a t i o n to perform the needed assays (Tsau et a l . , 1982). Unfortunately, the procedures involved i n t h i s system of designing a chemically defined medium are extremely time consuming, and therefore beyond the l i m i t s of t h i s research. However, culture conditions have been defined f o r human pe r i t o n e a l mesothelial c e l l s (Connell and Rheinwald, 1983; McKeehan and Ham, 1977; McKeehan et a l . , 1977), an epithelium which i s c l o s e l y r e l a t e d anatomically and embryologically to human OSE. These conditions were therefore chosen as a baseline to define a medium f o r the propagation of human OSE i n c u l t u r e . (5) SHORT TERM CARCINOGEN ASSAYS Short term, i n v i t r o t e s t s f o r carcinogen i d e n t i f i c a t i o n are an i n c r e a s i n g l y valuable means of rapid detection of the carcinogenic p o t e n t i a l substances, and f o r studying the mechanism of carcinogen a c t i o n . These tests are diverse, employing v i r a l systems ( E l i s p u r u , 1981), microbial systems - 18 -(Haroun and Ames, 1981), and cultured mammalian cells (Siegfied and Nesnow, 1984; Williams, 1981). However, a common feature of a l l of these tests i s that they, directly or indirectly, assess the DNA or chromosome-damaging capablity of potential mutagens or carcinogens. The general concept is that i f " DNA damage of any kind occurs, there i s a possibility that mutation may ensue. Therefore any agent that damages DNA i s a potential mutagen. Because mutagenesis and carcinogenesis are closely correlated, and because most, i f not a l l mutagens are also carcinogens, i t can be assumed that any agent that causes DNA damage i s also capable of inducing the process that eventually causes human cancer (Painter, 1981). Of the many types of short-term tests available, the use of mammalian cells in culture appears the most relevant. Studies with intact animals are very important i n analyzing many aspects of carcinogensis, but the use of cells i n culture permits a more direct experimental manipulation and quantitation of the individual steps and can yield information on their nature (Butterworth, 1985). In the past, short-term carcinogen assays employed mainly embryo cultures or fibroblastic c e l l lines. Although such c e l l types are easy to maintain i n culture, the fibroblastic character of these cultures represents a serious limitation; over 85% of a l l human cancers arise from epithelial rather than fibroblastic tissues (Weinstein et a l . , 1976). In recent years, attention has been directed toward developing culture conditions for the growth of a variety of normal epithelial c e l l types derived from both rodent and human tissues. Although rodent epithelia are generally easier to maintain i n culture, their usefulness as models for human epithelia has i t s limitations due to species diferences, particularly i n the study of carcinogenesis. For example, human and rat epithelia (mammary, endometrium) have recently been shown to differ in their capacity for carcinogen activation - 19 -(Moore et a l . , 1986) and DNA adduct formation (Kulkarni and Kaufman, 1986). Fortunately, with recent advances i n the f i e l d of tissue culture, a variety of normal human epithelia can now be maintained i n culture. Some of these epithelia are now being used, to a limited extent, i n the testing of chemical carcinogens and i n the study of the mechanisms underlying human carcinogenesis (Greenebaum et a l . , 1983; Eisinger et a l . , 1983; Kulkarni and Kaufman, 1986; Moore et a l . , 1986; Pruess-Schwartz et a l . , 1986; Siddiqui and Chopra, 1986; Siegfried and Nesnow, 1984; Willey et a l . , 1984). The use of human epithelial c e l l cultures such as these, rather than the usual rodent culture systems, is becoming an increasingly more valuable approach to mutagenicity and carcinogenicity testing (McCormick and Maher, 1985). Short-term tests for carcinogens can be divided into three major groups: (i ) those that measure DNA repair synthesis (unsheduled DNA synthesis) (Stich et a l . , 1981; Williams, 1985); ( i i ) those that measure inhibition of DNA synthesis (Painter, 1981); and ( i i i ) those that assess chromosomal damage such as chromosomal aberrations, DNA fragmentation, sister chromatid exchange (SCE) (Wolff, 1981; Wolff and Perry, 1974), or micronucleus production (Heddle et a l . , 1978; Schmid, 1975). Tests that measure DNA repair synthesis require c e l l cultures that are stationary or dividing very slowly so that the cells undergoing DNA repair synthesis (after exposure to a carcinogen) are not obscured by cells undergoing regular DNA replication (Williams, 1976, 1978, 1985; Stich et a l . , 1981). Cells such as rat l i v e r hepatocytes (Williams, 1985) and colon epithelial cells (Siddiqui and Chopra, 1976) are routinely used for such assays. Cell division i n replicating cultures can be limited by arginine-deprivation, thus extending the range of c e l l types to which this assay can be applied (Stich and San, 1970). This technique involves (i) the exposure of nondividing cells to graded concentations of a potential - 20 -carcinogen; ( i i ) the simultaneous or subsequent exposure of these cells to t r i t i a t e d thymidine ( H)TdR; ( i i i ) quantification of DNA repair synthesis by autoradiography or s c i n t i l l a t i o n counting (Stich et a l . , 198l). Tests measuring the inhibition of DNA synthesis require rapidly dividing c e l l populations and measure the differences i n the rate of DNA synthesis (by ( H)-TdR uptake) after exposure to a carcinogen. Exposure of DNA to a mutagen causes lesions that inhibit the overall rate of DNA synthesis as a function of time, u n t i l the repair processes intervene (Painter, 1981) Tests that assess chromosomal damage in cultured cells require a rapidly dividing population of cells (Wolff, 1981). The general test procedure is as follows: (i) rapidly dividing cells are exposed to a graded series of a carcinogen; ( i i ) chemical analogues of thymidine (in the case of SCE assay) are added to produce different sister chromatids; ( i i i ) c e l l division i s arrested at the second mitosis after treatment (for SCE) or between the third and fourth mitoses (micronucleus assay); (iv) the cells are fixed and stained with Giemsa stain; and (5) the cultures are assessed visually for chromosomal damage, SCE, or the presence of micronuclei in the cytoplasm. A c e l l cycle time between 24 and 48 hours i s considered optimal for this type of assay (R.H.C. San, pers. comm.). Because of the low level of carcinogen-activating capability of some cells i n long term culture, primary cultures or freshly-isolated cultures are commonly used in short-term carcinogen assays (Williams, 1976; Stich et a l . , 1981). Early passage cells generally retain their i n vivo characteristics (such as enzyme systems) making them particularly useful in detecting tissue-specific carcinogens or procarcinogens requiring enzymatic activation (Williams, 1978). In systems where carcinogen-activating enzymes may be lacking, exogenous enzymes, in the form of rat l i v e r homogenate, are often added to ensure detection of metabolism-requiring agents (R.H.C. San, pers. comm.). - 21 -(6) RATIONALE The objective of this research was to define culture conditions that permit ser i a l cultivation of normal human OSE i n culture. In developing a culture medium for the growth of any epithelial' c e l l type for use in the study of carcinogenesis, i t i s important to reproduce, as closely as possible, the proliferative characteristics of the cells i n vivo. Recent evidence suggests that an increase i n the proliferative activity of epithelial cells (rat nasal and prostatic epithelium) increases the susceptiblility of the epithelium to carcinogen-induced tumorigenesis (Bosland et a l . , 1986; P a t i l et a l . , 1986). OSE i n vivo (like endometrial epithelium) i s unusual in that i t regularly "cycles" between two proliferative states: (i) stationary or slowly dividing such as over the general ovarian surface; and ( i i ) rapidly dividing, such as in the repair of the ovarian surface after ovulation. The ovulatory defect i s usually covered by OSE in 3 to 5 days (Harrison and Weir, 1977; Papadaki and Beilby, 1971). In most other simple or str a t i f i e d human epithelia in vivo, rapid c e l l proliferation occurs only i n response to wounding (Rheinwald and Green, 1975). By culturing OSE under conditions that allow both rapid and slow proliferation, the proliferative states in which the OSE i s exposed to carcinogens i n vivo are reproduced. In addition, short-term tests for carcinogens are available which use either stationary or rapidly-dividing cultures. For the serial cultivation and rapid growth of OSE i n culture, a suitable synthetic medium was required. Prior to the present research, the only synthetic culture medium used in the growth of human OSE was Waymouth's 752/1 medium (Auersperg et a l . , 1984), which was chosen primarily so the results could be compared to previous and related studies (Adams and Auersperg, 1981a). Therefore, i t was unlikely that this medium was optimal medium for the growth of human OSE. - 22 -It was mentioned earlier that a medium should ideally be designed specifically for the c e l l type to be grown in i t . This can be accomplished by systematically analyzing the growth-promoting effects of different concentrations of each component of the medium. Unfortunately, due to the complexity of most synthetic culture media (70 to 80 different components), this system of designing a chemically defined medium was beyond the time constraints of this research. As an alternative, I elected to assess the potential of several available culture media for their a b i l i t y to stimulate the growth of OSE in culture. One of these media has been shown to stimulate the growth of normal human peritoneal mesothelial cells i n culture (Connell and Rheinwald, 1983; McKeehan et a l . , 1977; McKeehan and Ham, 1977). Although OSE and peritoneal mesothelium diff e r in morphology and function i n vivo (Blaustein and Lee, 1979), they share: (i) a common embryological origin; ( i i ) a common apical extracellular environment; and ( i i i ) some cellular functions (selective barrier, transcytotic a c t i v i t y ) . In view of these similarities, i t was thought that the culture conditions used successfully for peritoneal mesothelium in culture might likewise stimulate the growth of OSE and perhaps prolong i t s i n vitro lifespan. The medium used for mesothelial cells consists of a 1:1 mixture of two synthetic culture media (MCDB 202 and M199), epidermal growth factor (EGF), hydrocortisone (HC), and 15% fetal bovine serum (FBS). In this medium, mesothelial cells grew rapidly and underwent 40 to 50 population doublings before senescence. Both EGF and HC were required for optimal growth of human mesothelial cells (Connell and Rheinwald, 1983). In the present study, the growth of human OSE i n this medium was compared to growth i n the previously used culture medium (25$ FBS/WM) and to the synthetic media, M199 and MCDB 202 individually. Because requirements for c e l l growth could vary with the presence or absence of explants, and with c e l l density (Ham, 1974), c e l l - 23 -growth was examined under a variety of conditions. Maintainance of i n vivo characteristics was monitored by phase microscopy of l i v e cultures, bright-field microscopy of stained cultures, and keratin immunofluorescence. The growth promoting effects of EGF and HC on epithelia other than mesothelium are well documented (Gospodarowicz et a l . , 1977; Gospodarowicz and Bialecki, 1979; Kidwell et a l . , 1982; Cristofalo and Rosner, 1981). Because human OSE and peritoneal mesothelial cells d i f f e r considerably in their neoplastic potential, i t was of particular interest to determine i f these epithelia also differed i n their response to these mitogens. Recent reports indicate that steroid hormones modulate the effects of growth factors i n a variety of c e l l types (Chabot et a l . , 1985; Sadiq and Devaskar, 1984). In Xenopus hepatocytes, 178-estradiol (Eg) has been shown to inhibit the mitogenic effect of EGF (Wolffe et a l . , 1985). These observations, the presence of estrogen receptors (Adams and Auersperg, 1981b), and EGF receptors in rat OSE (Chabot et a l . , 1986), and the presence of EGF and Eg in the ovary i n vivo, prompted investigation into possible interactive effects of these agents in OSE. The effect of a l l possible combinations of EGF (20 ng/ml), HC (0.4 yg/ml), and three concentrations of Eg on ( H)TdR incorporation in human OSE cultures was examined. Since i t i s advantageous to eliminate as much serum from the culture medium as possible (ie. undefined components), the effects of decreasing the serum supplement of the medium below that previously used, 25% FBS, was also examined. The av a i l a b i l i t y of a model system for human OSE has numerous important applications. Since no relevant models for ovarian carcinogenesis exist at present, the use of human OSE in culture as a model for ovarian carcinogensis - 24 -may allow the identification of specific ovarian carcinogens, and perhaps provide clues regarding the susceptiblity of this tissue to neoplasia. The apparent capability of OSE for both rapid and slow proliferation rates may qualify primary or secondary cultures of this tissue as suitable for several types of short-term carcinogen assays. OSE i n culture could potentially be important i n the identification of badly needed immunological markers for the early detection of OSE-derived cancers. And f i n a l l y , this culture system w i l l undoubtably yield valuable information on the biology of OSE. (C) RAT OVARIAN SURFACE EPITHELIUM (1) BACKGROUND One of the most striking features of mammalian ovarian surface epithelium i s i t s pleomorphism under both normal and pathological conditions i n vivo, and in tissue culture. On the surface of the normal ovary, the OSE ranges from squamous (on the distended surface overlying the Graafian f o l l i c l e ) , to cuboidal (over the general ovarian surface), to columnar in association with surface crypts and inclusion cysts (Van Blerkom and Motta, 1979; Motta and Van Blerkom, 1980). OSE cells on the post-ovulatory defect assume a very flattened morphology as they migrate over the connective tissue matrix (Motta and Van Blerkom, 1980). In ovarian inclusion cysts, the OSE may exhibit a variety of morphological forms under "normal" conditions. Here, the OSE ranges from a f l a t to a cuboidal epithelium, to epithelia with morphological and histochemical characteristics of the Mullerian duct derivatives: endometrial epithelium, and endosalpingeal epithelium (Blaustein et a l . , 1982; Kerner et a l . , 1981). Morphological changes i n the OSE are also observed in many types of ovarian malignancies. 'Common epithelial tumours' of the ovary often exhibit - 25 -metaplasia to a c e l l type resembling the e p i t h e l i a of the ' Mullerian duct d e r i v a t i v e s , namely serous, mucinous and endometrioid which have h i s t o l o g i c a l and secretory c h a r a c t e r i s t i c s of tubal, endocervical, and endometrial e p i t h e l i a r e s p e c t i v e l y (Parmley and Woodruff, 1974). There i s also some evidence that the stromal component of ovarian cancers, generally thought to be derived from the ovarian stroma (or i n metastases, from stroma at other s i t e s ) , may be derived from the OSE (Parmley and Woodruff, 1974; Adams and Auersperg, 1981a, 1985). In tissue culture, the OSE exhibits a range of morphological forms. A r a t OSE c e l l l i n e , ROSE 199, forms p a p i l l a r y structures i n culture that resemble serous p a p i l l a r y cystadenomas seen i n humans (Adams and Auersperg, 1985). Within these structures, ROSE 199 c e l l s maintain an e p i t h e l i a l morphology but express the stromal c h a r a c t e r i s t i c of abundant collagen production. Rabbit OSE c e l l s i n culture also d i s p l a y a range of morphologies i n c l u d i n g the formation of v i l l o u s processes, hemicysts, multilayered configurations, and e p i t h e l i i a l invaginations (Nicosia et a l . , 1984). In preliminary studies from our laboratory (Auersperg et a l . , 1983, 1984a), outgrowths of human OSE i n primary culture assumed several forms: ( i ) compact, monolayered e p i t h e l i a l c e l l s ; ( i i ) flattened e p i t h e l i a l c e l l s ; and ( i i i ) a t y p i c a l e p i t h e l i a l c e l l s . In addition, OSE c e l l s with e p i t h e l i a l morphology tended to become more a t y p i c a l with time i n c u l t u r e . A l l morphological types of OSE colonies possessed numerous m i c r o v i l l i and were p o s i t i v e f o r k e r a t i n by immunofluorescence. In some cultures, OSE c e l l s retracted from the substratum and formed m u l t i c e l l u l a r , p a p i l l a r y aggregates. In other cultures, the c e l l s became loos e l y adherent and spindly, or formed concentric arrays at the periphery of colonies, resembling cultures of squamous k e r a t i n i z i n g e p i t h e l i a (Kulesz-Martin et a l . , 1980). - 26 -Recently, two methods were reported that permit the culture of the ovarian surface epithelium of the rat (Adams and Auersperg, 1980, 1981a; Hamilton et a l . , 1981). Rat OSE (ROSE) cells i n explant culture reproducibly grow as cobblestone-like outgrowths of epithelial c e l l s , when cultured i n Waymouth's medium with 25% FBS (25FBS/WM). Most of these cultures could be passaged only a few times before senescence, but some proceeded to form continuous c e l l lines (Adams and Auersperg, 1981a, 1985). ROSE cells in culture have provided some direct evidence that the OSE i s the site of origin of 'common epithelial tumours' o f the ovary. Kirsten murine sarcoma virus-transformed ROSE cells produced highly malignant tumours in rats, resembling ovarian endometrioid stromal sarcomas in humans (Adams and Auersperg, 1980). ROSE cells have also been shown to share characteristics of the Mullerian duct epithelia: they possess estrogen-receptors (Adams and Auersperg, 1983; Hamilton et a l . , 1982) and 17(i-HSD activity (Adams and Auersperg, 1981b). (2) RATIONALE In this study, a continuous line of rat OSE cells (ROSE-239), which was established from explant cultures, was used as a model to examine the morphological forms exhibited by human OSE i n culture. Human OSE cells are impractical for this purpose, for reasons described i n Section A2. Preliminary results with ROSE 239 cultures revealed that when seeded at clonal density, the resultant clones displayed one of three morphological forms (compact epithelial, flattened epithelial, and atypical), which bore striking resemblance to the morphological forms exhibited by human OSE i n primary culture (Auersperg et a l . , 1984). It was thought that ROSE 239 cells may therefore represent a useful model for the study of OSE pleomorphism and the interrelationships between different OSE c e l l forms. To examine the - 27 -relationship between the morphological phenotypes observed in ROSE ' 239 cultures, attempts were made to clone colonies of each of the observed morphologies. These clones were then reseeded at clonal density and the morphology of the resultant clones analyzed. Further, the morphologies of clones were followed from single-cell to large colony, to examine whether the observed morphologies were stable or i f the cells were capable of modulation between different morphological forms. Clonal lines of ROSE derived from colonies of different morphology were also examined i n long term culture, to determine i f characteristics of their growth or morphology varied with time or culture conditions. - 28 -METHODS AND MATERIALS (A) HUMAN OVARIAN SURFACE EPITHELIUM (1) TISSUES Biopsy specimens of normal ovary were obtained at surgery from 5 premenopausal women undergoing surgery for nonmalignant gynecological 'disorders (ages 30 to 37 years). The types of operations from which samples were obtained include (i) total hysterectomy; ( i i ) myomectomy and bilateral oophorectomy for fibroids; ( i i i ) hysterectomy for pelvic inflammatory disease. The specimens were collected under sterile conditions and transported to the laboratory i n Waymouth's 752/1 medium (WM) with 25$ fetal bovine serum (FBS) at ambient temperature. (2) TISSUE CULTURE (a) Tissue Preparation The ovarian sample was processed for culture within 30 minutes of obtaining the specimen, as described by Auersperg et a l . (1984). More specifically, the tissue was transferred t-o a glass dissecting dish containing WM, then, using a dissecting microscope, extraneous stroma was trimmed from beneath the ovarian surface, leaving a sheet of tissue 1 to 2 mm in thickness (Figure l ) . To minimize disruption of the OSE while removing the stroma, the tissue was held i n place with forceps (by the stroma) then cut with a sharp scalpel by direct pressure rather than a "sawing motion". The tissue was then transferred to another dish containing fresh culture medium. The sheet of ovarian surface was cut into strips, then into cubes ( l to 2 mm ) using the same dissection technique. The cubes, or explants, were placed with the ovarian surface facing up into 35 mm culture dishes in small drops of culture medium. - 29 -Figure 1. Culture Method For Human Ovarian Surface Epithelium Diagram to demonstrate the culturing method for human ovarian surface epithelial cells (OSE). a. Biopsy specimens are obtained at surgery and transported to the laboratory i n st e r i l e culture medium. b. Excess stroma i s trimmed from beneath the ovarian surface. c. Ovarian surface i s cut with a scalpel into 1 - 2 mm^  explants. d. Explants are placed epithelium-side-up i n a plastic culture dish and weighed down with a coverslip and cloning cylinder. OSE outgrowths extend out onto the dish or coverslip. e. Explants are removed to another culture dish when the OSE outgrowths approach one another. f. OSE outgrowths are passaged when subconfluent, cultured for several passages, and used i n experiments to assess the growth-promoting effects of different media. - 30 -F i g u r e 1. C u l t u r e Method f o r Human Ovarian Surface E p i t h e l i u m a. b. c. d . e. f . BIOPSY SPECIMENS FROM 5 HUMAN PREMENOPAUSAL OVARIES EXPLANTS REMOVED EPITHELIAL OUTGROWTHS SUBCULTURED CLONAL GROWTH GROWTH CURVES . MASS CULTURE KERATIN IMMUNOFLUORESCENCE - 31 -Five to 7 explants were placed i n each dish and the explants were allowed to adhere to the dish for 6 minutes. A round polystyrene or Thermanox coverslip was placed on top of the explants and weighed down with a glass cloning cylinder, and 2 ml of culture medium was then carefully added to the dish. The explants were arranged near the periphery of the coverslip to allow adequate exchange of nutrients and gases. In addition to explant cultures, the culture medium i n the f i n a l dissecting dish was collected, spun at 1000 RPM (180 g) in a c l i n i c a l centrifuge, and the cells were plated into a single culture dish. The cultures were incubated at 37° C i n a humidified incubator with 5% C0^:S5% air, and l e f t undisturbed for 3 to 4 days after explantation. In total, 5 ovarian biopsy specimens (approximately 235 explants) were cultured i n this manner. Cell outgrowths extended out onto culture dishes and coverslips. When OSE outgrowths began to approach each other, the coverslips were removed and the explants were transferred to new culture dishes. The explants were again covered with a coverslip and cloning cylinder, and immersed in 2 ml of medium to obtain additional outgrowths. The outgrowths were identified as epithelial or fibroblastic by morphological c r i t e r i a . Outgrowths of OSE cells i n culture frequently form a confluent, pavement-like monolayer, while fibroblastic outgrowths appear as an array of aligned cells radiating from the explant. Generally, only cultures i n which a l l explants produced typical "compact" (see Introduction A-2) epithelial outgrowths were used for further experimentation. However, i n some cases, minor contaminating fibroblastic outgrowths were removed from the dish by (i) scraping the cells (circled on the bottom of the dish with a f e l t pen) with a rubber policeman; or ( i i ) cutting the outgrowths from the coverslip with s t e r i l e scissors. Dishes badly contaminated with fibroblasts were not used. To subculture, subconfluent OSE cells were dissociated with 0.125$ trypsin and 0.02$ EGTA i n calcium/magnesium-- 32 -free Hank's "balanced salt solution (trypsin/EGTA), and spun at 1000 RPM i n a c l i n i c a l centrifuge for 4-5 minutes. The cells were reseeded at a s p l i t ratio of about 1:3« OSE cells were stored frozen under liquid nitrogen i n WM with 25$ FBS and 8% dimethylsulf oxide (DMSO). Cell counts were made using a hemocytometer or an Artek 980 c e l l counter. (b) Media and Culture Conditions Several types of synthetic media were used in the culture of OSE cells (Table I ) : (i) Waymouth's Medium 752/1 (WM); ( i i ) Medium 199 with Earle's salts (199); ( i i i ) MCLB 202 (202) (McKeehan and Ham, 1977; McKeehan et a l . , 1977); and (iv) 199/MCDB 202 (mixed 1:1) (199/202). The culture media were supplemented with either 5%, 15$, or 25$ fetal bovine serum (FBS) and contained 100 IU p e n i c i l l i n G/ml, and 100 ug streptomycin/ml. The FBS used i n this study (Hyclone lot # 110451) i s partially characterized so the concentrations of many of i t s components are known. Epidermal growth factor (20 ng/ml) (EGF) and hydrocortisone (4.0 ug/ml) (HC) were added to some culture media. OSE cells were passaged in 199/202 with 15$ FBS, EGF (E), and HC (H) (15FBS/199/202/EH) to obtain enough cells f o r . medium-testing experiments. When sufficient OSE ce l l s were obtained, the cultures were changed to medium lacking EGF and HC for several days, then passaged to the culture conditions to be tested for their effects on OSE c e l l proliferation. This was required to eliminate the mitogenic and morphological effects of EGF and HC on OSE c e l l s . In routine cultures, the medium was changed at 3 to 4 day intervals or more often in crowded cultures. In some cultures c e l l identity was questionable, therefore at subculture some cells were seeded on glass coverslips and were examined for keratin by indirect immunofluorescence. Synthetic culture media used i n these experiments were prepared, and FBS - 33 -TABLE I: CULTURE MEDIA AND SERUM CONCENTRATIONS TESTED 25$ PBS, WAYMOUTH MEDIUM 25$ FBS, WAYMOUTH MEDIUM, EGF, HYDROCORTISONE 5$ AND 15$ FBS, WAYMOUTH MEDIUM, EGF, HYDROCORTISONE 5$ AND 15$ FBS, MEDIUM 199, EGF, HYDROCORTISONE 5$ AND 15$ FBS, MCDB 202, EGF, HYDROCORTISONE 5$ AND 15$ FBS, MEDIUM 199/MCDB 202 ( l : l ) , EGF, HYDROCORTISONE EGF: 20 ng/ml HYDROCORTISONE: 4 ug/ml - 34 -added, less than 2 weeks before use. For cloning experiments, fresh medium (less than 1 week old) was used. MCDB 202 medium was prepared by the method of McKeehan and Ham (1977), and McKeehan et a l . (1977). That i s , the medium was prepared and stored as a series of stock solutions (Appendix I) which were combined to form the complete medium. For addition to cultures, lyophylized EGF was reconstituted to 100 ug/ml in s t e r i l e double-distilled HgO (ddH^O), aliquoted, and stored at -70°C. Prior to use, aliquots were thawed, diluted to 10 ug/ml i n ddHgO, and frozen at -20°C for 1 to 2 weeks u n t i l used. Lyophylized HC was reconstituted to 1 mg/ml in 70$ ethanol, aliquoted and stored at -70°C. Prior to use, aliquots were diluted to working concentrations (200 ug/ml) and stored at -20°C for 1 to 2 weeks un t i l used. Ethanol concentration was kept below 0.15 % i n a l l cultures. EGF and HC were replenished at medium change or by separate addition at 3 day intervals in a l l cultures. (c) Evaluation of Culture Conditions A variety of approaches were used to analyze the growth of OSE c e l l s . These included: (i) measuring the extent of OSE outgrowth from explants i n primary culture; and, after subculture, ( i i ) visual inspection of growth in mass culture; ( i i i ) growth curves; (iv) seeding efficiency; and (v) cloning efficiency. (i) OSE Outgrowth. In 3 of the 5 biopsies obtained, the extent of outgrowth was assessed by measuring the maximum radius of outgrowth of OSE from explants at 10 to 12 days after explantation into the various culture media. In addition, the percentage of explants producing outgrowths after 4 or 5 days in culture was recorded. Between 5 and 15 explants were examined for each medium. - 35 -( i i ) Growth of OSE i n mass culture. Growth of OSE i n mass culture (4 experiments; 2 cases; passages 7 to 10) was used as a preliminary method for screening the various culture media, the FBS concentrations, and the effectiveness of EGF and HC addition. In an early experiment (Experiment 2), 14 different culture conditions were tested, including 4 culture media, 3 FBS concentrations, and the presence or absence of EGF and HC (Table IV). This experiment was used as a guideline for later experiments in which only the media yielding the best results were tested. OSE c e l l s , passaged in 15FBS/199/202/EH then transferred to 25FBS/WM as described i n Section A-2-b, 2 • were plated i n 2 cm culture wells (4 wells for each medium) at 1500 2 cells/cm . After trypsinization and centrifugation, the cells were suspended in WM, then added in 0.1 ml of WM to 0.4 ml of the appropriate medium. The cells were allowed to proliferate for 10 days, with a medium change every 3 to 4 days. The wells were then fixed i n 10$ formol saline, stained for 1 minute in 0.025$ crystal violet, and allowed to dry. Major differences i n c e l l mass were assessed visually. ( i i i ) Growth Curves. For OSE growth curves (3 ser i a l passages of 1 case), 2 2 13,000 OSE cells/cm (passages 3 to 5) were plated i n 2 cm wells i n 0.5 ml of 25$ FBS/WM (to avoid differences' i n plating density). After 24 hours, the wells were scanned to eliminate wells of obviously different density, then changed to one of three media: Waymouth's medium with 5$ FBS (5FBS/WM); Waymouth's medium with 15$ FBS (15FBS/WM); Waymouth's medium with 5$ calf serum (5CS/WM); or 15FBS/199/202/EH. Three replicate wells were prepared for each medium. The culture medium was changed every 3 days i n i t i a l l y , then as required i n wells that became crowded. At 2 to 4 day intervals, the culture medium was removed and the cells were frozen i n culture wells i n 0.4 ml of 0.25$ TETPCK lyophylized trypsin dissolved i n phosphate-saline (0.05M - 36 -Na2HP04; 2M NaCl; 0.002M EDTA) buffer, pH 7.4- When the growth curve was completed, the wells were thawed and the DNA content quantified using a technique based on the enhancement of fluorescence seen when the fluorochrome bisbenzimidazol (Hoechst 33258) binds to DNA (Labarca and Paigan, 1980) (Appendix III). Briefly, the thawed wells were diluted to 3 ml i n PBS and 50 ul or 150 ul (depending on c e l l density) of Hoechst dye (20 ug dye/ml HgO) was added. The percent transmission at 458 nm was then read using an American Instruments Company spectrophotofluorometer. DNA standards (20 ng/ml to 3 ug/ml) were prepared for each expermiment from a stock DNA solution (100 ug calf thymus DNA/ml PBS) stored i n aliquots frozen at -20°C. Approximate c e l l numbers were calculated on the basis of 7 pg of DNA per c e l l (Altman and Katz, 1976) (iv) Clonal Growth. For clonal growth of OSE (4 experiments; 3 cases), a single-cell suspension of OSE was prepared in trypsin/EGTA by gentle pipetting. The suspended cells were assessed as primarily single and counted with a hemocytometer, centrifuged, and seeded at a density of 100 cells/60mm culture dish in 5 ml of the medium to be tested. Three replicate dishes were prepared for each medium. The cells were maintained for 14 days without medium change, however, EGF and HC were added every 3 days. The cells were then fixed and stained as i n Section A-2-c-ii. Seeding efficiency (number of single cells attached + number of colonies) and cloning efficiency (number of.colonies > 2 cells) were microscopically quantified. (3) HUMAN PERITONEAL MESOTHELIAL CELL CULTURE Cultures of human peritoneal mesothelial cells (Strain LP-9) (Connell and Rheinwald, 1983) were obtained from Dr. J. G. Rheinwald (Dana-Farbar Cancer Institute, Boston, MA). Early passage LP-9 cells are available from the - 37 -Aging Cell Repository of the Institute for Medical Research (Camden, NJ. -Repository # AG7086) (LaRocca and Rheinwald, 1985)• LP-9 cultures were used to ensure that the results obtained with these c e l l s by Connell and Rheinwald (1983) could be reproduced i n our laboratory and compared to OSE cultuures. LP-9 cells were cultured i n 15PBS/199/202 medium with or without EGF and HC and handled similarly to OSE. .(4) IMMUNOFLUORESCENCE MICROSCOPY Immunofluorescence microscopy for keratin was used to identify OSE in culture and to observe changes in keratin filament pattern and abundance in media with and without EGF and HC. For immunofluorescence, OSE were grown for several days i n medium without EGF and HC, then dissociated with trypsin/EGTA and plated in drops on glass coverslips in 35 mm culture dishes. The cells were allowed to settle for several minutes before adding the culture media (25FBS WM, 15PBS/199/202, or 15FBS/199/202/EH). After several days, well before the cells became confluent, excess medium was drained from the coverslips and the cells were fixed by immersing the coverslip directly into a Columbia jar containing cold (-20°C) 100$ methanol. The cells were then processed using a method similar to that of O'Guin et a l . (1985) (Appendix II) . Briefly, the cells were fixed i n cold methanol-acetone ( l : l ) for 5 minutes, allowed to dry, and rehydrated i n 0.02 M phosphate buffer (PBS). Next, the cells were incubated in a humidified chamber for 30 minutes at 37 C i n 50 ul of primary antibody (rabbit antiserum prepared against human epidermal (callus) keratins (l° Ab) - g i f t from T.T. Sun), rinsed i n PBS for 1 hour, and incubated for 30 minutes at 37°C in secondary antibody (2°Ab) (FITC-conjugated goat anti-rabbit IgG, H and L chain specific). Negative control preparations which were run concurrently included: (i) OSE cells stained with normal rabbit serum (NRS) i n place of the l°Ab then with - 38 -2°Ab; ( i i ) OSE ce l l s stained with 2°Ab only; and ( i i i ) rat lung fibroblasts stained with both 1° and 2°Ab's. A positive control of rat ovarian surface epithelium (ROSE 239, see below) was also included i n each experiment. The coverslips were then rinsed i n PBS overnight, mounted on chromic acid-cleaned glass microscope slides i n a drop of 16.7$ Gelvatol (pH 7.8) (Rodrigues and Dienhardt, I960), and the coverslips were sealed with clear nailpolish. The mounted slides were viewed and photographed on a Zeiss Photomicroscope II (Appendix II). The same indirect immunofluorescence technique was used to observe differences i n keratin filament pattern in R0SE-C4 and R0SE-B2 clones. (5) THYMIDINE INCORPORATION ASSAYS 3 3 Incorporation of ( H)thymidine (( H)TdR) into DNA was used i n some OSE cultures as a measure of c e l l proliferation (Appendix IV). OSE cells were plated i n 0.1 ml of medium at 3000 cells/6 mm well in 96-well plates. The cells were plated i n 25FBS/WM to avoid differences i n seeding efficiency that might arise from using different media. Approximately 8 hours after plating, the medium was changed to the test medium (4 wells for each medium) and the ce l l s were grown to a subconfluent state. After approximately 3 days, (3 )TdR (lmCi/ml i n sterile water, 2 Ci/mmole) was added i n 0.1 ml of WM (at 1 LiCi/ml) to the medium i n each well, and the wells were incubated for 3 hours. The medium was then replaced with 0.2 ml of 0.25$ trypsin (TRTPCK lyophilized)/0.02$ EDTA/0.8$ Triton X-100 and incubated at 37° C for 30 minutes. The lysate was harvested onto glass microfiber f i l t e r s using a microharvester. The f i l t e r discs were washed i n d i s t i l l e d H^ O, then 70$ ethanol, dryed for 30 minutes, and counted i n Omnifluor liquid s c i n t i l l a t i o n f l u i d using a Phillips PW40 sc i n t i l l a t i o n counter. This procedure was repeated i n three experiments. - 39 -(6) EFFECT OF 173-ESTRADIOL ON DNA REPLICATION To determine the effect of 17B-estradiol (1,3,5(l0)-estratrien-3,17B-diol) (E 2) on c e l l proliferation, OSE cells were plated into 96-well plates in 199 medium with 15$ dextran charcoal-treated FBS (DCT-FBS) (Appendix V). Approximately 8 hours after plating, the plating medium was replaced with 15DCT-FBS/199 containing combinations of EGF (20 ng/ml), HC (0.4 ug/ml), and —10 —8 —6 3 concentrations of (10 M, 10 M, and 10 M f i n a l concentrations) i n 11 possible combinations (Figure 23). Quadruplicate wells were used for each treatment. A stock solution of was prepared by reconstituting E^ i n 100$ ethanol and storing i t at -20°C. The stock was diluted to working solutions in 70$ ethanol just prior to use. Ethanol concentration was about 0.3$ i n a l l cultures, including control wells. The cells were grown to subconfluence (approximately 3 days) and assayed for HTdR incorporation as described above. (7) PHOTOGRAPHY Photomicrographs of c e l l cultures were obtained using a Wild M40 inverted photomicroscope and Kodak 2415 Technical Pan-black and white film. The film was developed i n Kodak HC-110 developer (dilution D) for 6 minutes and printed on Ilford Ilfospeed photographic paper. Fluorescence photomicrographs were taken on a Zeiss Photomicroscope II using Kodak Tri-X Pan (400 ASA taken with a camera setting of 800 ASA) black and white or Kodak Ektachrome (1600 ASA taken with a camera setting of 800 ASA) colour (daylight) slide film. Tri-X Pan film was developed i n Kodak Microdol developer (Appendix IV). - 40 -(8) STATISTICS St a t i s t i c a l analysis of the data presented here was performed on an Apple l i e computer, using a Stats Plus (Madigan and Lawrence, 1983) s t a t i s t i c a l program. The data were analyzed using Student's T-test, where 2 groups of data were to be compared, and by one-way ANOVA followed by Tukey's test, where more than two groups were compared. When the data from more than one experiment were pooled, the data for each experiment were normalized to 25FBS/WM (Data Medium X/Data 25FBS/WM X 100$), to eliminate differences between individual experiments. (B) RAT OVARIAN SURFACE EPITHELIUM A continuous c e l l line derived originally from primary cultures of rat ovarian surface epithelium (ROSE 239) (Adams and Auersperg, 1981a), was used as a model to investigate OSE pleomorphism. ROSE 239 cells were routinely cultured i n 10FBS/WM i n plastic culture vessels, and maintained i n humididfied 5% COg/air as described for OSE c e l l s . To separate clones of different morphologies, ROSE cells (passage 26 to 31) were suspended i n trypsin/EGTA, counted and assessed as single by hemocytometer, then seeded at clonal density (100 to 500 cells/60mm culture dish). To ensure that the colonies were of clonal origin, attached single c e l l s were identified microscopically and marked, within 12 hours of plating. At 12 to 14 days, colonies of different morphologies were cloned by: (i) isolating the colony with a glass cloning cylinder dipped i n sterile silicone grease; ( i i ) f i l l i n g the cloning cylinder with 1 or 2 drops of trypsin/EGTA; ( i i i ) washing the cells within the cloning cylinder from the plastic with a fine-tipped pipette; and (iv) transferring the cells (without centrifugation) - 41 -2 to a 2 cm culture well containing 0.4 ml of culture medium. Two single-cell-clones (R0SE-C4 and R0SE-B2) were selected for their morphology, were grown to confluence, then reseeded at clonal density (500 cells/60mm culture dish; 3 replicate dishes of each) within 5 passages to establish the st a b i l i t y of the observed morphologies. The dishes were l e f t undisturbed for 7 or 14 days, then fixed in 10$ formol saline and stained with crystal violet. The clones of each observed morphology were counted. In several instances, clones were followed for up to 3 weeks by photomicrographs taken every 4 to 5 days. To assess the response of R0SE-C4 and R0SE-B2 to crowding, several cultures of these clones were maintained i n a confluent state for up to 10 weeks. Duplicate flasks of each clone were either "starved" by changing the medium only once every 3 to 6 weeks, or supplied with new medium every 3 to 4 days. - 42 -RESULTS (A) HUMAN OVARIAN SURFACE EPITHELIUM (1) EFFECT OF CULTURE CONDITIONS ON OSE GROWTH (a) OSE Outgrowth In Primary Cultures The extent of OSE outgrowth under different culture conditions was examined i n 3 of the 5 biopsy specimens (cases) obtained. Further testing of different culture media on primary cultures was impractical due to the limited number of ovarian biopsies available and the small size of the biopsies obtained. The remaining 2 biopsies were cultured i n a medium (15FBS/199/-202/EH) that, i n i n i t i a l tests, was found to provide the greatest OSE outgrowth and to delay c e l l senescence. As a result, i t was possible to adequately assess the effects of only two media, 25FBS/WM and 15FBS/199/-202/EH, on OSE c e l l growth in primary culture. The results obtained on the effect of the various culture conditions on OSE outgrowth are therefore best described on the basis of individual cases. CASE I: In Case I, OSE outgrowth was assessed in 15FBS/WM, 15FBS/WM/EH, and 25FBS/WM from 4 to 10 days after explantation (Figure 2; Table II). The number of explants producing outgrowth after 4 days i n culture was highest i n 25FBS/WM (67$), lowest i n 15FBS/WM (47$), and intermediate i n 15FBS/WM/EH (57$). The average maximum radius of outgrowth increased at a similar rate i n a l l three media, from day 4 to day 9 (Figure 3)« No s t a t i s t i c a l l y significant difference i n OSE outgrowth was evident i n the three media tested after 9 days i n culture. This experiment provided preliminary evidence that the addition of EGF and HC to the culture medium promotes a more fusiform morphology i n OSE outgrowth in primary cultures. However, the growth promoting effects of EGF and HC appear to be less effective i n WM (as evidenced by the similarity i n - 43 -TABLE II: EFFECT OF CULTURE CONDITIONS ON OSE GROWTH IN PRIMARY CULTURE PERCENT (NUMBER) OF MAXIMUM RADIUS EXPLANTS WITH OSE OF OUTGROWTH CASE MEDIUM OUTGROWTHS (mm _+ SEM)) I DAY 4 DAY 9 15FBS/WM 48 (10) 1.5 (0.24) 15FBS/WM/EH 57 (8) 1.8 (0.37) 25FBS/WM 67 (10) 1.9 (0.25) II DAY 5 DAY 12 25FBS/WM 0.0 0.79 (0.25)** 25FBS/WM/EH 50.0 (2) 1.05 (0.25) 15FBS/WM/EH 92.3 (12) 1.91 (0.14) 15FBS/199/EH 50.0 (7) 1.21 (0.15) 15FBS/202/EH 33.3 (3) 1.09 (0.15) 15FBS/199/202/EH 100 (9) 1.93 (0.19)** IV DAY 7 DAY 11 25FBS/WM 75.0 (6) 1.49 (0.31)* 15FBS/199/202/EH 75.0 (6) 2.03 (0.20)* V DAY 8 DAY 11 15FBS/199/202/EH 58.5 (7) 1.65 (0.42) * P< 0.05, n = 8 ** P< 0.01, n = 4 - 13 - 44 -Figure 2. Outgrowth of Human Ovarian Surface Epithelium i n Primary-Culture Histogram showing average r a d i i of outgrowth (mm) of human ovarian surface epithelium (OSE) from explants grown under different culture conditions. Bars represent mean radius of outgrowth i n each medium +_ SEM for 4 different biopsy specimens. n = number of explants/medium = 4-15« The average radius of OSE outgrowth was significantly greater in 15FBS/199/202/EH than i n 25FBS/WM i n Case II and in Case IV (P<0.05). In Case II, outgrowth of OSE i n 15FBS/WM/EH was also greater than 25FBS/WM (p<0.05). The two bars at the bottom of the graph show outgrowth of OSE from previously cultured explants two days after transfer to a new culture dish. Figure 2 . Outgrowth of Human Ovarian Surface E p i t h e l i u m ' i n Primary C u l t u r e Case I Case l l CaseLYf Explant transfer %FBS Medium Supplement 25 WM 15 WM 15 WM EGF/HC 25 WM — 25 WM EGF/HC 15 WM EGF/HC 15 199 EGF/HC 15 202 EGF/HC 15 199/202 EGF/HC 25 WM — 15 199/202 EGF/HC 15 199/202 EGF/HC 25 WM — : 15 199/202 EGF/HC Average outgrowth (mm) 0 1.0 2.0 i 1 1 1 1 1 1 1 1 1 1 1 r 3 — • ' l 1 3-i——i j — i — i 1 1 ' Q — • > I ' ± = 3 • I I I I I I I I I I I I L - 46 -Figure 3« Outgrowth of Human Ovarian Surface Epithelium i n Primary-Culture: (Case I) Graph showing average radius of OSE outgrowth at 2 day intervals from explants of Case I grown i n 25FBS/WM ( o O ), 15FBS/WM/EH ( O O ) , and 15FBS/WM ( • • ). Points represent mean radius of outgrowth + SEM for 1 experiment. - 47 -F i g u r e 3 . O u t g r o w t h o f Human O v a r i a n S u r f a c e E p i t h e l i u m i n P r i m a r y C u l t u r e ( C a s e I) 0 2 4 6 8 10 DAYS - 48 -OSE outgrowths i n 15FBS/WM and 15FBS/WM/EH) than i n other media tested i n later experiments. The growth pattern of OSE outgrowths varied with time in culture and with the addition of EGF and HC to the culture medium (Table III). In a l l three media, outgrowths were i n i t i a l l y a mixture of epithelial, atypical and mixed epithelial/atypical growth patterns. By 10 days after explantation, no purely atypical outgrowths remained i n the 15FBS/WM and 25FBS/TO cultures; the outgrowths were either epithelial or mixed epithelial/atypical. In contrast, after 10 days a l l outgrowths in 15FBS/WM/EH were either atypical or mixed epithelial/atypical, with no purely epithelial outgrowths. CASE II: Explants of Case II ovarian surface were cultured under 6 different culture conditions (25FBS/WM, 25FBS/WM/EH, 15FBS/WM/EH, 15FBS/202/EH, 15FBS/199, and 15FBS/199/202/EH) with 1 to 3 culture dishes (4-5 explants/culture dish) for each medium. The percent of explants producing OSE outgrowths at day 5 was highest in 15FBS/199/202/EH (100$) and 15FBS/WM/EH (92.3$) (Figure 2; Table II). In 25FBS/WM there were no outgrowths at Day 5, while in the other media OSE outgrowths occurred i n 33$ to 50$ of explants. The average maximum radius of outgrowth at day 12 was greater i n 15FBS/199/202/EH (1.93 mm; p<0.0l) and 15FBS/WM/EH (1.91 mm; p<0.05) than i n 25FBS/WM (0.79 mm). CASE III: Explants of ovarian surface from Case III were cultured in the following media: 25FBS/WM, 15FBS/WM/EH, 15FBS/199/EH, 15FBS/202/EH (one dish of each) and 15FBS/199/202/EH (5 dishes). Unfortunately, no explants produced typical epithelial OSE outgrowths. The cel l s had a fusiform morphology and grew i n a pattern typical of fibroblastic outgrowths: the cells were arranged - 49 -TABLE III: EFFECT OF CULTURE CONDITIONS ON OSE GROWTH PATTERN IN PRIMARY CULTURE (CASE I) GROWTH PATTERN MEDIUM DAY EPITHELIAL ATYPICAL MIXED 15FBS/WM 4 5 6 1 10 5 0 5 15FBS/WM/EH 4 6 3 4 10 0 5 5 25FBS/WM 4 2 4 1 10 3 0 3 - 50 -in a parallel array radiating out from the explants. In addition, most of the dishes became infected with yeast, indicating non-sterile handling procedures somewhere between surgery and explantation. Therefore, this case i s not considered i n the remainder of this report. CASE IV: Explants of ovarian surface from Case IV were cultured i n two media only, 25FBS/WM and 15FBS/199/202/EH, with two dishes of each medium. By 7 days after explantation, 75$ of explants i n both media had OSE outgrowths (Figure 2; Table II). At Day 11, the average maximum radius of OSE outgrowth was significantly greater (p<0.05) i n 15FBS/199/202/EH (2.03 mm) than i n 25FBS/WM (1.49 mm). CASE V: Explants of Case V were cultured i n 15FBS/199/202/EH only (Figure 2; Table II). Most OSE outgrowths grew slowly and some explants never produced outgrowths. By 8 days after explantation, 58$ of explants had outgrowths. At 11 days after explantation, 81.8$ of explants had outgrowths averaging 1.65 mm (SEM _+ 1.26). In the 5 cases discussed here, there are two media for which the data from different cases can be combined to further assess OSE outgrowth; 25FBS/WM (Cases I, II and IV) and 15FBS/199/202/EH (Cases II, IV and V). When the maximum rad i i of outgrowth i n each medium for a l l experiments are pooled without normalizing the data to 25FBS/WM, the outgrowths were significantly more extensive i n 15FBS/199/202/EH than i n 25FBS/WM (p<0.05). This indicates an enhanced probability of greater outgrowth i n 15FBS/199/202/EH for any explant cultured i n this medium, regardless of i t s source. The pooled average OSE outgrowth i n 15FBS/199/202/EH was 1.87 mm (SEM _+ O.ll) while i n 25FBS/WM outgrowths averaged 1.39 mm (SEM +_ 0.32). When the outgrowth data from a l l - 51 -experiments are normalized to 25FBS/WM and then pooled, no significant differences i n the relative radius of outgrowth was evident between media, probably due to the high variability i n OSE outgrowth even when cultured i n the same medium (Figure 4). Overall, 70 of 115 explants i n the cases described above produced OSE outgrowths within 4 to 8 days. However, the number of explants producing OSE outgrowths and the extent of outgrowths i n some media, was highly variable. For example, OSE outgrowth at 9 to 12 days after explantation into 25FBS/WM ranged from 0.79 mm i n Case II to 1.49 mm i n Case IV. OSE outgrowth appeared to be less variable in 15FBS/199/202/EH, ranging from 1.65 mm in Case V to 2.03 mm i n Case IV. A dramatic improvement i n OSE c e l l yield was achieved by transferring explants to new culture dishes after OSE outgrowth occurred. This method has been used by others for the growth of human bronchial epithelial explants i n culture (Lechner et a l . , 1982). In one case (Case II), the explants continued to produce primarily OSE outgrowths for at least 7 transfers. In other cases (Cases IV and V), OSE outgrowths were produced for 2 or 3 explant transfers after which the outgrowths were primarily fibroblastic on 25FBS/WM or 15FBS/199/202. OSE outgrowths occurred much more quickly from transferred explants (within 2 days after transfer), presumably because the surface of these organoids had become covered with a continuous sheet of OSE. No s t a t i s t i c a l difference was observed i n the extent of outgrowth (at 2 days after transfer) from explants transplanted into 25FBS/WM and 15FBS/199/202/EH (Figure 2). (b) Growth In Mass Culture Growth of OSE in mass culture was assessed i n 4 experiments and included ce l l s of passage 7 to 10 and Cases I and II. Figure 5 shows an early - 52 -Figure 4. Relative Outgrowth of Human Ovarian Surface Epithelium i n Primary Culture Histogram showing relative outgrowth of OSE i n primary culture. Radius of outgrowth data for each medium i n each experiment has been normalized to the average radius of outgrowth i n 25FBS/WM in the same experiment for comparative purposes. Bars represent mean relative outgrowth _+ SEM for 1 - 3 experiments. n = number of outgrowths assessed = 4 - 3 6 . There was no s t a t i s t i c a l difference i n relative OSE outgrowth i n the different media. Figure 4. R e l a t i v e Outgrowth of Human Ovarian Surface E p i t h e l i u m i n Primary C u l t u r e %FBS Medium 25 25 15 15 15 15 15 WM WM WM WM 199 202 199/202 Relative outgrowth (normal ized to 25% F B S / W M ) Supplement 0 1.0 2.0 3.0 I 1 1 1 1 1 1 1 1 1 1 1 1 1 r -EGF/HC EGF/HC EGF/HC EGF/HC EGF/HC T = \ 1 3 1 I L J I I I i J I I L - 54 -Figure 5- Growth of Human Ovarian Surface Epithelium i n Mass Culture Photograph of wells from a representative experiment showing human OSE (passage 8) seeded at 1500 cells/cm^, grown under a variety of different culture conditions for 10 days, then fixed i n 10$ formol saline and stained i n 0.025$ crystal violet. Note that OSE c e l l growth i s greatest i n 5FBS and 15FBS/199/202/EH, lowest in WM with any supplement, and intermediate i n the other media. OSE c e l l growth is lower in a l l media lacking EGF and HC. - 56 -experiment (Case I) i n which OSE cells were grown i n 15 different culture media. OSE growth was greatest i n 5FBS and 15FBS/199/202/EH, somewhat lower in 5PBS and 15FBS/202/EH, and 15FBS/199/EH, and only marginal i n the remaining media. The results from this experiment and three similar experiments are summarized i n Table IV. In a l l experiments, OSE c e l l mass was greater (by visual assessment) i n 15FBS/199/202/EH than in the other media tested. In this medium, OSE cells underwent 20 to 25 population doublings based on 1:3 s p l i t ratio. 5FBS/199/202/EH was rated second best in the two experiments i n which i t was tested. Generally, 5FBS and 15FBS/199/EH, and 5FBS and 15FBS/202/EH promoted OSE growth, but not to the same extent as either 5EBS and 15FBS/199/202/EH media. A significant decrease i n OSE c e l l mass was observed in 199, 202, and 199/202 cultures when EGF and HC were ommitted from the culture medium. Cultures grown in WM invariably produced the lowest OSE c e l l mass, despite the presence of FBS concentrations up to 25%, and the addition of EGF and HC. No differences i n OSE growth related to the source of the OSE or to c e l l passage were evident. Microscopic observation of fixed cultures after the 10 day culture period indicated the presence of cells with an epitheloid morphology in media lacking EGF. and HC, and a more fusiform morphology i n media containing EGF and HC. (c) Growth Curves OSE growth was monitored over 12-14 days using a fluorometric DNA assay. Once again, 15FBS/199/202/EH was superior for rapid growth of OSE c e l l s . Figures 6 to 8 show three experiments (Case I) where OSE was cultured i n 5FBS/WM, 15FBS/WM, and 15FBS/199/202/EH. In 15FBS/199/202/EH, OSE reached the logarithmic phase of the growth curve within 4 days, and had an average population doubling time of 48 hours (range 24 - 72 hours) during the log TABLE IV: GROWTH OF OSE IN MASS CULTURE: VISUAL RANKINGS* OF CELL GROWTH ON DIFFERENT MEDIA EXPERIMENT A EXPERIMENT B EXPERIMENT C EXPERIMENT D CASE I, p. 7 CASE I p. 8 CASE I, p. 10 CASE I I , p.2, t . 3 MEDIUM VISUAL RANKING VISUAL RANKING VISUAL RANKING VISUAL RANKING 5FBS/WM - - - -5FBS/WM/EH 4 6 8 -15FBS/WM - 6 - -15FBS/WM/EH 3 6 8 3 25FBS/WM 3 6 7 4 25FBS/WM/EH 3 5 8 -5FBS/199 2 - - -5FBS/199/EH 1 5 6 -15FBS/199 - 6 - -15FBS/199/EH - 3 3 2 5FBS/202 - - - -5FBS/202/EH - 3 5 -15FBS/202 - 4 - -15FBS/202/EH - 2 4 2 5FBS/199/202 - - - -5FBS/199/202/EH - 2 2 -15FBS/199/202 - 5 - -15FBS/199/202/EH - 1 1 1 * Wells fixed and stained a f t e r 10 days i n culture and ranked v i s u a l l y f o r c e l l mass present ( l = highest ranking; i . e . greatest c e l l mass present). -- Not done; p = c e l l passages; t = number of times explant transferred to a new dis h . - 58 -Figure 6. Growth Curves f o r Human Ovarian Surface Epithelium: Passage 3 Growth curves f o r human OSE (Case IV, passage 3) grown i n several media. C e l l growth was assessed by a spectrofluorometric technique using a fluorescent dye (Hoechst) that binds to DNA. Points represent mean _+ SEM f o r 3 r e p l i c a t e wells. This technique y i e l d s an approximation of c e l l number based on 7 pg DNA/cell. Note that c e l l growth i s most rapid i n 15FBS/199/202/EH and minimal i n 5FBS and 15FBS/WM. F i g u r e 6 . G r o w t h C u r v e s f o r Human O v a r i a n S u r f a c e E p i t h e l i u m : P a s s a g e 3 . 0 2 4 6 8 10 12 14 Days in culture - 60 -Figure 7. Growth Curves for Human Ovarian Surface Epithelium: Passage 5 Growth curves for human OSE (Case IV, passage 5) grown i n several media. Cell growth was assessed by a spectrofluorometric technique using a fluorescent dye (Hoechst) that binds to DNA. Points represent mean _+ SEM for 3 replicate wells. This technique yields an approximation of c e l l number based on 7 pg DNA/cell. Note that c e l l growth i s most rapid i n 15FBS/199/202/EH, and minimal i n 5 and 15FBS/WM. F i g u r e 7. G r o w t h C u r v e s f o r Human O v a r i a n S u r f a c e E p i t h e l i u m : P a s s a g e 5 . 0 2 4 6 8 10 12 14 D a y s in culture - 62 -Figure 8. Growth Curves for Human Ovarian Surface Epithelium: Passage 7 Growth curves for human OSE (Case IV, passage 7) grown i n several media. Growth was assessed by a spectrofluorometric technique using a fluorescent dye (Hoechst) that binds to DNA. Points represent mean +_ SEM for 3 replicate wells. This technique yields an approximation of c e l l number based on 7 pg DNA/cell. Note that c e l l growth i s most rapid i n 15FBS/199/202/EH and minimal i n 5 and 15FBS/VM. Although DNA was not measured on day 12 for 15FBS/199/202/EH, c e l l growth appears to have diminished i n this passage as compared to passage 3 (Figure 6). F i g u r e 8 . G r o w t h C u r v e s f o r Human O v a r i a n S u r f a c e E p i t h e l i u m : P a s s a g e 7 . Days in culture - 64 -phase of growth. Saturation density was not achieved i n 15FBS/199/202/EH even a f t e r 12 days i n cult u r e . A f t e r 1 2 days i n cu l t u r e , OSE c e l l s "became very crowded, and began to p i l e up and p u l l back into clumps (Figure 9a). This behaviour may have provided a d d i t i o n a l surface f o r OSE c e l l growth and may explain why saturation density was not achieved. A f t e r 12 days i n culture, the OSE c e l l number i n 15FBS/199/202/EH ranged from 4-3X105 to 5.4X105 (from two experiments which ran to 12 days), based on an approximation of 7 pg DNA/cell (Altman and Katz, 1976). The growth p o t e n t i a l of t h i s s t r a i n of OSE i n 15FBS/199/202/EH appeared to decrease from passage 3 (Figure 6) to passage 5 (Figure 8), while the curves f o r 15WM and 5WM remained unchanged between passages. A l l cultures were passaged i n 15FBS/199/202/EH between experiments. The growth c h a r a c t e r i s t i c s of OSE c e l l s grown i n 5FBS/WM and 15FBS/WM were very d i f f e r e n t than i n 15FBS/199/202/EH. In these media, OSE c e l l s never entered a log phase of growth. Instead, the c e l l number remained constant or increased only s l i g h t l y over the 12 to 14 day period. The average c e l l number at Day 12 f o r the 3 experiments was s l i g h t l y higher i n 15FBS/WM (5-6X10 4 +_ 9-4X105 SEM) than i n 5FBS/WM (3.0X10 4 _+ 3.1X10-5 SEM). However, the average c e l l number at Day 12 was s i g n i f i c a n t l y lower than i n 15FBS/199/202/EH (p<O.Ol). OSE cultured i n 5CS/WM began to f l o a t o f f into culture medium s h o r t l y a f t e r p l a t i n g and only a few c e l l s remained a f t e r 12 days i n culture (Figure 9e). Therefore, these cultures were not analyzed f o r DNA content. In 5FBS/WM and 15FBS/WM, OSE c e l l s maintained a fla t t e n e d , e p i t h e l i a l morphology over the 14 day period of the growth (Figure 9c and 9d). In 15FBS/199/202/EH, OSE c e l l s had a more fusiform morphology c h a r a c t e r i s t i c of c e l l s grown i n the presence of EGF and HC (Figure 9a). - 65 -Figure 9« Morphology of Human OSE Cells i n Growth Curve Experiments Phase photomicrographs of l i v e cultures of human OSE cells (Case IV, Passage 5) after 10 days i n culture i n growth curve experiments. OSE cells were grown i n : a and b. 15FBS/199/202/EH (a. xlJO and b. x50); c. 15FBS/WM (xl30) ; d. 5FBS/WM (xl30); and e. 5CS/WM (xl30). Note that OSE cells i n 15FBS/199/202/EH are fusiform and pulling back into clumps or ridges, while those i n 5FBS and 15FBS/WM display a flattened, epithelioid morphology. The majority of cel l s i n 5CS/WM died and floated off by Day 10. - 67 -(d) Clonal Growth Clonal growth was assessed in four experiments and included cells from I, II, and IV and of passages 2 to 8. The data for individual seeding efficiency experiments and the pooled data are shown i n Figures 10 and 11. Overall, seeding efficiency in most media was between 30$ and 50%. The pooled data (where seeding efficiency i s normalized to 25FBS/WM) indicate that the relative seeding efficiency ranged from 111$ to 180$ of 25FBS/WM i n a l l media (except 5FBS/WM) (Figure 10). Seeding efficiency was significantly lower i n 5FBS/WM (p<0.01 - 0.05) than i n a l l the other media, which were not st a t i s t i c a l l y different. Analysis of individual experiments revealed similar seeding efficiency results, with some variability between cases (Figure l l ) . In Case I, as i n the pooled data, a l l media had a higher seeding efficiency than 5FBS/WM (p<0.0l). In Case II, seeding efficiency was significantly higher in 5FBS and 15FBS/199/202/EH (p<0.01 - 0.05) than i n a l l the other media, which were not s t a t i s i c a l l y different. In Case IV, seeding efficiency was higher in both 5 FBS and 15FBS/199/202/EH than i n 5FBS/WM (p<0.05 and p<0.01 respectively). In addition, seeding efficiency was significantly greater i n 15FBS/202/EH than in 15FBS/WM/EH, 25FBS/WM and 5FBS/199/EH. The remainder were s t a t i s t i c a l l y indiscernible. In a l l experiments, seeding efficiency was not significantly decreased when the FBS concentration was lowered to 5$ (except i n WM Case I) . Therefore, FBS concentration does not appear to be an important factor in OSE seeding i n the media tested. Seeding efficiency was variable between different experiments, however, there was no consistent relationship between c e l l passage and seeding efficiency. Seeding efficiency appears to be less variable in 5FBS and 15FBS/199/202/EH than i n the other media tested. The cloning efficiencies were similarly analyzed as individual and - 68 -Figure 10. E f f e c t of Culture Media on Relative Seeding E f f i c i e n c y of Human OSE Histogram showing r e l a t i v e seeding e f f i c i e n c y of ea r l y passage human OSE. Seeding e f f i c i e n c y data f o r each experiment has been normalized to the average seeding e f f i c i e n c y of 25FBS/WM f o r the same experiment f o r comparative purposes. Bars represent mean r e l a t i v e seeding e f f i c i e n c y + SEM f o r 2 experiments ( f o r those media containing 5% FBS) or 3 experiments f o r the other media. Note that the r e l a t i v e seeding e f f i c i e n c y i s s i m i l a r i n a l l media except 5FBS/WM/EH i n which the r e l a t i v e seeding e f f i c i e n c y i s s i g n i f i c a n t l y lower (p<0.01 - 0.05). F i g u r e 10. E f f e c t of Culture Media on R e l a t i v e Seeding E f f i c i e n c y o f Human OSE Relative s eed ing efficiency ( n o r m a l i z e d to 25% F B S / W M ) % F B S M e d i u m S u p p l e m e n t 25 WM — 15 WM EGF/HC 5 W M EGF/HC 15 199 EGF/HC 5 199 EGF/HC 15 202 EGF/HC 5 202 EGF/HC 15 199/202 EGF/HC 5 199/202 EGF/HC 0 0 . 2 0 . 4 0 . 6 0 . 8 1.0 1.2 1.4 1.6 1.8 2 . 0 2 . 2 i I I I I I I I I [ i 1 3 — I 1 I I I I I I I ' - 70 -Figure 11. Effect of Culture Media on Seeding Efficiency of Human OSE Histogram showing individual seeding efficiency experiments of early passage human OSE cells plated at clonal density (100 cells/dish) and l e f t for 14 days. Bars represent mean seeding efficiency _+ SEM for each experiment with 3 replicate dishes/experiment. Roman numerals represent case numbers. The c e l l passage at the time of experimentation are: Case I - Passage 8; Case II - Passage 2 + 3 explant transfers; and Case IV - Passage 2. Note that seeding efficiency varies between cases and that seeding efficiency i s generally higher i n the 199 and 202 containing media, and lower in the WM containing media. Figure 11. E f f e c t of Culture Media on Seeding E f f i c i e n c y o f Human OSE %FBS 25 25 15 5 15 15 15 M e d turn WM WM WM WM 199 199 202 202 199/202 199/202 S e e d i n g efficiency ( %) S u p p l e m e n t 0 10 2 0 3 0 4 0 5 0 6 0 / 1 0 0 EGF/HC I I — 11 IV I EGF/HC 1 1 IV EGF/HC ,v EGF/HC EGF/HC EGF/HC EGF/HC I EGF/HC II IV I IV I IV I IV I EGF/HC ' I IV I IV 3^ I—r—• 3 - « 3 3 ^ 3* 3 ' J = r — " 3 -3 1 i - 72 -pooled data (Figures 12 and 13). Pooled data (normalized to 25FBS/WM) indicate that the relative cloning efficiency i s greater i n 15FBS/199/202/EH (p<O.Ol), than i n 5FBS/WM/EH, 15FBS/WM/EH, 5FBS/199/EH, and 5FBS/202/EH (Figure 12). However, cloning efficiencies i n 5FBS and 15FBS/199/202/EH, 15FBS/199/EH and 15FBS/202/EH, were not s t a t i s t i c a l l y different. In a l l media, lowering the FBS concentration to 5% resulted in a decrease (albeit not st a t i s i c a l l y significant by ANOVA) i n cloning efficiency and WM was consistently inferior to other synthetic media. In individual cases, cloning efficiency was s t a t i s t i c a l l y similar for a l l media i n Cases I and II. In Case IV, cloning efficiency was higher i n 15FBS/199/202/EH (p<0.01 - 0.05) than in a l l other media except 15FBS/202/EH (Figure 13). Although not st a t i s t i c a l l y significant, the pooled average cloning efficiency i n each case was almost always highest i n 15FBS/199/202/EH and lowest i n WM supplemented with 5%, 15%, or 25% FBS, even when EGF and HC were added (Table V). As observed for seeding efficiency, there were no obvious differences i n cloning efficiency due to donor or c e l l passage. Table V shows the distribution of the sizes of clones observed i n each medium. In 5%, 15%, and 25% FBS/WM, very few colonies with more than 5 cells formed. Likewise, i n the other media supplemented with 5% FBS (5FBS/199/EH and 5FBS/202/EH) there were very few colonies with more than 5 cells were formed. Only i n 5FBS/199/202/EH and i n 199, 202, and 199/202 media supplemented with 15% FBS, EGF and HC, did an appreciable number of larger colonies form. More colonies greater than 10 cells (often 50 to 250 cells) formed i n 15FBS/199/202/EH than i n any other medium (p<0.01 using normalized data), except 15FBS/199EH. OSE clones were typically composed of dispersed OSE ce l l s , that appeared to have migrated away from one another (Figure 14). OSE cells did not form - 73 -Figure 12. Effect of Culture Media on Relative Cloning Efficiency of Human OSE Histogram showing the relative cloning efficiency of early passage human OSE c e l l s . Cloning efficiency data for the different media i n individual experiments has been normalized to the cloning efficiency data for 25FBS/WM in the same experiment for comparative purposes. Bars represent mean relative cloning efficiency _+ SEM for 2 experiments (those media containing 5$ FBS) or 3 experiments for the other media. Note that relative cloning efficiency i s significantly greater i n 15FBS/199/202/EH (p<O.Ol), than i n 5FBS/WM/EH, 5FBS/199/EH, and 5FBS/202/EH. 5FBS and 15FBS/199/202/EH, 15FBS/199/EH, and 15FBS/202/EH are s t a t i s t i c a l l y the same. Figure 12. E f f e c t of Culture Media on R e l a t i v e Cloning E f f i c i e n c y of Human OSE Relative c lon ing eff ic iency (normal i zed to 25% F B S / W M ) 0 1.0 2 . 0 3 . 0 4.0 5 . 0 6 . 0 7 . 0 8 . 0 9 . 0 1 0 . 0 % F B S M e d i u m S u p p l e m e n t 25 W M — 15 W M E G F / H C 5 W M E G F / H C 15 199 E G F / H C 5 199 E G F / H C 15 2 0 2 E G F / H C 5 2 0 2 E G F / H C 15 1 9 9 / 2 0 2 E G F / H C 5 1 9 9 / 2 0 2 E G F / H C J I I I I L - 75 -Figure 13• Effect of Culture Media on Cloning Efficiency of Human OSE Histogram showing individual cloning efficiency experiments of early passage human OSE cells plated at 100 cells/culture dish and l e f t for 14 days. Bars represent mean cloning efficiency _+ SEM for each experiment. Roman numerals represent case numbers. The c e l l passages at the time of experimentation are the same as for Figure 11. Note that in Case IV cloning efficiency was higher i n 15FBS/199/202/EH than i n any of the other media for that case (p<0.01 - 0.05) except 15FBS/202/EH. - 76 -Figure 13. E f f e c t of C u l t u r e Media on Cloning E f f i c i e n c y of Human OSE %FBS Med lum 2 5 2 5 15 5 15 15 15 W M W M W M W M 199 199 2 0 2 2 0 2 1 9 9 / 2 0 2 1 9 9 / 2 0 2 C l o n i n g efficiency (%) Supplement 0 5 .0 10 15 I—i i I I i I l I l i l I I I l l 1 0 0 E G F / H C , — 11 IV E G F / H C n L t h iv 0T E G F / H C * \ E G F / H C j', IV E G F / H C E G F / H C II IV 3-< E G F / H C | V E G F / H C 11 IV E G F / H C , v 3 1 ' • i ' ' i i i ' i i ' - 7 7 -TABLE V: CLONING EFFICIENCY AND SIZE OF COLONIES OF OSE IN CLONAL CULTURE AVERAGE CLONING EFFICIENCY* (PERCENT) MEDIUM 2 - 5 CELLS/ 6 - 1 0 CELLS/ > 10 CELLS/ TOTAL COLONY COLONY COLONY 25FBS/WM 0.89 (3-1) 0 0.11 (0.11) 1.0 (0 . 3 7 ) 15FBS/WM/EH 0.89 (3.1) 0.11 (0.11) 0 1.0 (0.33) 5FBS/WM/EH 0 0 0 0 5FBS/199/EH 0.67 (0.49) 0 0 0.67 (0.49) 15FBS/199/EH 2.40 (0.87) 1.00 (0.33) 1.2 (0.57) 4.60 (2.00) 5FBS/202/EH 1.50 (0.57) 0.22 (0.33) 0.11 (0.11) 1.83 (0.85) 15FBS/202/EH 1.67 (0.43) 1.33 (0 . 3 3 ) 0.78 (0.37) 3.78 (0.70) 5FBS/199/202/EH 2.20 (0.69) 0.83 (0.54) 0.33 (0.21) 3-36 (0.65) 15FBS/199/202/EH 3.78 (0.97) 0.78 (0.28) 2 . 3 0 (0.43) 6.86 (1.33) *Numbers represent mean _+ SEM of: ( i ) 2 experiments ( t r i p l i c a t e dishes f o r each expt.) f o r 25FBS/WM and media c o n t a i n i n g 5% FBS; and ( i i ) 3 experiments f o r other media. - 78 -Figure 14• Morphology of Clonal Colonies of Human Ovarian Surface Epithelium Light photomicrographs of a single clonal colony of human OSE cells after 14 days i n culture i n 15FBS/199/202/EH (a. x50; b. xlJO). Cells were fixed in 10$ formol saline and stained in 0.025$ crystal violet. Note that the cells do not form a cohesive clone but rather appear to migrate away from one another. Many of the single cells have a flattened, fan-shaped morphology and i n several regions the cel l s have joined to form a small epithelial sheet. - 80 -compact colonies as seen i n other epithelial c e l l types and rat ovarian surface epithelium (ROSE 239) i n culture. Individual OSE c e l l s i n clones had a flattened, fan-shaped morphology with a large, ruffled leading lamellipodium extending out from the c e l l body. They generally lacked a long retraction fiber characteristic of fibroblasts. This morphology i s typical of migrating human urothelial cells (Kirk et a l . , 1985). In some regions of the clones, OSE cells become closely associated to form confluent groups of c e l l s . (2) EFFECT OF EGF AND HC ON OSE MORPHOLOGY (a) Primary Cultures When primary cultures of OSE were initiated i n medium containing EGF and HC, the morphology of the OSE outgrowth was variable. Some explants produced outgrowths that i n i t i a l l y assumed a typical cobble-stone pattern and then, after a few days i n culture, formed very dense, parallel arrays of cells radiating out from the explant; a morphology similar to that seen in fibroblasts i n culture (Figure 15a). In contrast, some explants (often i n the same culture dish as those described above) produce outgrowths with a confluent, cobblestone pattern that was maintained despite the presence of EGF and HC (Figure 15b). Other explants formed outgrowths of irregularly-shaped ce l l s (Figure 15c) or a mixture of irregular and flattened cells (Figure 15d). A similar v a r i a b i l i t y i n human OSE morphology in primary culture was observed by Auersperg et a l (1984a). On several occasions, primary cultures of very crowded, fusiform OSE cells (Figure 16a), cultured in 15FBS/199/202/EH, were subcultured to medium without EGF and HC. Within several days of subculture, the cells had undergone a drastic morphological change to a flattened, epithelial form (Figure 16b). This exemplifies the u t i l i t y of this culture system i n distinguishing between the morphologically similar fusiform OSE cells and fibroblasts. - 81 -Figure 15• Morphology of Human OSE i n Primary Culture Phase photomicrographs of liv e cultures of human OSE i n primary culture. A l l cultures were grown i n 15FBS/199/202/EH, but display quite different morphologies, a. Very dense, fusiform OSE (?) cells resembling fibroblastic outgrowths (xlJO). b. Confluent sheet of OSE forming a cobblestone pattern typical of other epithelial c e l l types (xl30). c. Irregularly-shaped or atypical OSE cells (xl30). d. Mixed atypical and flattened OSE cells (xl30). - 82 -Figure 15. - 83 -Figure 16. Effect of EGF and HC on the Morphology of Human OSE i n Primary Culture Phase photomicrographs of liv e cultures of human OSE. a. A dense, fibroblastic outgrowth of OSE cells i n primary culture grown i n 15FBS/199/202/EH (x 130). b. The same culture exhibiting an epithelial morphology several days after subculture into the same medium but lacking EGF and HC (xl30). - 84 -Figure 16. - 85 -(b) After Subculture The addition of EGF (20 ng/ml) and HC (0.4 ug/ml) to culture media altered the morphology of OSE. Subcultured OSE cells grown on medium without EGF and HC displayed a flattened, epithelioid morphology. When crowded, these cells formed a loosely arranged sheet of flattened OSE cells (Figure 17a and 17b). The cells were similar in size and shape and form close contacts with neighbouring c e l l s . When EGF and HC were added to the culture medium, the cells reached confluence more rapidly than cultures lacking EGF and HC. In addition, the majority of the cells assumed an elongate or fusiform morphology, often with long cytoplasmic extensions (Figure 17b). Many of the cells had highly ruffled edges and cytoplasmic extensions. In crowded cultures, OSE cells exposed to EGF and HC overlappped and formed parallel arrays of cells (Figure 17b). When OSE cultures containing EGF and HC (Figure 18a) were changed (or subcultured) to medium without EGF and HC, OSE ce l l s became noticeably f l a t t e r within 24 hours and assumed a more epithelial form indistinguishable from cells not exposed to EGF and HC (Figure 18d). OSE cells grown on 25FBS/WM were much fl a t t e r than those grown on 15FBS/199/202 (Figure 19). This modulation of morphology i s comparable to that displayed by human peritoneal mesothelial cells cultured under similar conditions (Connell and Rheinwald, 1983). Figures 20a and 20b show peritoneal mesothelial cells (Strain LP-9) grown on 15FBS/199/202 i n the absence of EGF and HC. In this medium the cells displayed a flattened morphology resembling OSE i n the same medium. Figures 20c and 20d show mesothelial c e l l s , under sparse and dense conditions respectively, cultured i n the presence of EGF and HC. Under these conditions the ce l l s were more fusiform and formed dense parallel arrays when crowded. These observations indicate that the culture system used i n this - 86 -Figure 17. E f f e c t of EGF and HC on the Morphology of E a r l y Passage Human OSE C e l l s Phase photomicrographs of l i v e cultures of human OSE c e l l s grown i n 15FBS/199/202 medium: a. without EGF/HC (xl30); and b. with EGF/HC (xl30). Note that the c e l l s not exposed to EGF/HC e x h i b i t a fla t t e n e d , e p i t h e l i o i d morphology and form a l o o s e l y associated c e l l sheet, while the majority of the c e l l s exposed to EGF/HC e x h i b i t an elongate or fusiform morphology. - 88 -Figure 18. Effect of EGF and HC on the Morphology of Early Passage Human OSE Cells Phase photomicrographs of live cultures of early passage human OSE cells grown i n 15FBS/199/202. a. Cells exposed to EGF/HC (xl30). b. The same culture as i n a., 24 hours after changing to medium lacking EGF/HC (xl30). c. OSE cells maintained u n t i l confluent i n medium containing EGF/HC (xl30). d. OSE ce l l s not exposed to EGF/HC. Note that OSE cells modulate from a fusiform morphology i n medium containing EGF/HC (a and d), to a more epithelial form in medium lacking EGF/HC (b). OSE cells grown i n 25FBS/WM and not exposed to EGF/HC remain flattened and epithelioid (d). - 90 -Figure 19- Morphology of Human Ovarian Surface Epithelium i n Waymouth's Medium Phase photomicrographs of liv e cultures of early passage human OSE grown in 25FBS/WM (xlJO). Note that cells grown i n this medium are noticably fl a t t e r than those grown i n 15FBS/199/202 (Figure 18). - 92 -Figure 20. Effect of EGF and HC on the Morphology of Early Passage Human Peritoneal Mesothelial Cells (Strain LP-9) Phase photomicrographs of liv e cultures of early passage normal human peritoneal mesothelial cells (Strain LP-9) grown i n 15FBS/199/202. a. Sparse culture grown i n the absence of EGF/HC (xllO). b. Subconfluent culture grown i n the absence of EGF/HC (xllO). c. Sparse culture grown i n the presence of EGF/HC (xllO). d. Dense culture grown i n the presence of EGF/HC (xllO). Note that human meosthelial c e l l s i n culture are morphologically very similar to human OSE cells (Figure 18), and undergo a similar modulation of morphology i n response to EGF and HC. a - 94 -study was comparable to that used by Connell and Rheinwald, (1983) despite the different source of MCDB 202 medium. In one culture, OSE cells cultured for 2 weeks i n 15FBS/199/202/EH, retracted from the substratum and formed multicellular aggregates which extended up into the culture medium (Figure 21b). The aggregates were often associated with large, multinucleated cells with granular cytoplasm. This phenomenon has been observed previously by Auersperg et a l . (1984) i n primary-cultures of OSE ce l l s cultured i n 25FBS/WM. (3) EFFECT OF EGF AND HC ON KERATIN FILAMENT ORGANIZATION IN OSE OSE cells cultured in medium (15FBS/199/202) lacking EGF and HC had a filamentous arrangement of cytokeratin filaments distributed throughout the cytoplasm of approximately 80 - 90$ of OSE cells (Figure 22a). In the remaining cells cytokeratin staining was more diffuse and similar in intensity to background staining (OSE cells stained with NRS and 2°Ab). The keratin filament distribution of OSE cells subcultured to medium containing EGF and HC changed markedly within 48 hours of subculture. Only about 20 - 30$ of OSE cel l s displayed the typical filamentous arrangement of keratin filaments (Figure 22b). The majority of cells were diffusely stained and staining was again similar in intensity to background staining. Negative OSE controls and rat lung fibroblasts were stained nonspecifically i n a diffuse pattern (Appendix II) . Thus, OSE cells i n culture appear to modulate their keratin filament composition from an intensely staining, filamentous form in more slowly dividing cells grown i n the absence of EGF and HC, to a more diffuse form i n more rapidly growing cells cultured i n the presence of EGF and HC. A similar modulation of keratin filament composition has been described for human peritoneal mesothelial cells i n culture (Connell and Rheinwald, 1983). - 95 -Figure 2 1 . Further Observations on Human OSE Morphology i n Culture Photomicrographs of l i v e cultures of human OSE. a. OSE (?) c e l l s (Passage 8) grown i n i t i a l l y i n 15FBS/199/202/EH, then transferred to 25FBS/WM f o r several passages (xl30). Note the fusiform morphology of these c e l l s despite the absence of EGF/HC. b. OSE c e l l s cultured i n 15FBS/199/202/EH (xl30). C e l l s have retracted from the substratum to form m u l t i c e l l u l a r aggregates which extend up int o the culture medium, c. Culture of aging OSE c e l l s (xl50). -97 -Figure 22. Effect of EGF and HC on Keratin Filament Expression i n Human OSE Cells Photomicrographs of representative areas of early passage human OSE cells cultured i n 15FBS/199/202 with and without EGF/HC, then stained by immunofluorescence for cytokeratin filaments (xl60). a. OSE cells grown in medium without EGF/HC and stained with antikeratin 1° Ab and FITC-conjugated 2° Ab. b. OSE cells grown i n medium with EGF/HC and stained the same as in a. c. OSE cells grown in medium without EGF/HC and stained with normal rabbit serum and 2° Ab. Note that OSE cells grown i n medium without EGF/HC have distinct cytokeratin filaments throughout the cytoplasm, while the majority of cells grown i n medium with EGF/HC have greatly diminished, diffusely stained keratin filaments (arrows). A positive c e l l i s included i n b to ill u s t r a t e that most, but not a l l , cells have diminished keratin. Photographs a and b were printed identically to the negative control culture (c), which.was printed to eliminate diffuse background staining. - 99 -Negative OSE controls stained only with second antibody were negative. Immunofluorescence photomicrographs were taken at the same time with the same exposure times, and printed under identical conditions as OSE control (NRS/2°Ab) photomicrographs (Figure 22c). (4) EFFECT OF EGF AND HC ON OSE CULTURE LIFESPAN Prior to this research, OSE could be passaged only once or twice before the cells underwent senescence and death (Auersperg et a l , 1984). In this study OSE cells were subcultured up to 10 times i n 15FBS/199/202/EH, the medium now routinely used to passage OSE c e l l s . The cells underwent 20 to 25 population doublings before senescence or modulation to a fusiform morphology which could not be reverted to epithelial by removing EGF and HC from the culture medium. Availability of OSE cells did not allow the passaging of cultures for the sole purpose of defining c e l l lifespan, therefore population doublings are inferred from the available data (ie: calculations from c e l l dilutions). Figure 21c shows a senescent culture of OSE cells which have ceased to divide. These cells possessed long cytoplasmic extensions, stress fibres and detached easily from the plastic when pipetted gently. In several cases OSE c e l l s , ( i n i t i a l l y epithelial i n morphology) were repeatedly subcultured i n 15FBS/199/202/EH (8 to 10 passages). In early passage, these cel l s grew slowly and exhibited a flattened epithelioid morphology when transferred to medium lacking EGF and HC. However, after 8 - 1 0 passages when these cells were transferred to media lacking EGF and HC (25FBS/WM and 15FBS/199/202) they continued to grow rapidly and maintained a fusiform morphology (Figure 21a). In one experiment, these cells were negative for keratin filaments by immunofluorescence. - 100 -(5) EFFECT OF 17 B-ESTRADIOL ON OSE PROLIFERATION. The i n t e r a c t i v e e f f e c t s of EGF, HC, and e s t r a d i o l ( E 2 ) on OSE p r o l i f e r a t i o n were examined i n one experiment by ( H)TdR incorporation i n t o OSE DNA (Figure 23). When no E 2 was added to the culture medium, EGF alone has a stimulatory e f f e c t on ( H)TdR incorporation; 38$ above con t r o l cultures ( i . e . c o n t r o l = background cultures with no medium a d d i t i v e s ) . When HC alone was added, ( H)TdR incorporation increased to 107$ above con t r o l c u l t u r e s . When EGF and HC were added together, t h e i r e f f e c t s were more than 10 a d d i t i v e (215$ above c o n t r o l ) . When E 2 was added at 10 M, ( H)TdR incorporation was stimulated to 3 4 $ above c o n t r o l c u l t u r e s , while at other 3 E^ concentrations ( H)TdR incorporation was the same as c o n t r o l s . I n t e r e s t i n g l y , the stimulatory e f f e c t of EGF observed i n cultures with no —10 —6 ad d i t i v e s , appeared to be i n h i b i t e d by E^, at 10~ M and 10 M E^. Also, the a d d i t i v e e f f e c t of EGF and HC was diminished by E^ added at —8 —10 10" M and 10" M. The stimulatory e f f e c t of HC observed i n media without —6 —8 E 2 was enhanced by the presence of 10 M (80$ above control) and 10 M (15$ above control) E 2 > and diminished at 10 _ 1^M (20$ below c o n t r o l ) . Thus, i t appears that the stimulatory e f f e c t of EGF may be i n h i b i t e d by E 2 1 0 ft (at 10" and 10" M E„), while the stimulatory e f f e c t of HC may be —8 —10 enhanced by E 2 (at 10 and 10 M E 2 ) . Further experimentation i s necessary to v a l i d a t e t h i s . (7) HUMAN PERITONEAL MESOTHELIUM The morphology and growth pattern of human per i t o n e a l mesothelial c e l l s ( S t r a i n L P - 9 ) , cultured f o r comparison to OSE, were comparable to LP - 9 cultures previously reported (Connell and Rheinwald, 1 9 8 3 ) . In 15FBS/199/202/EH, LP - 9 c e l l s grew r a p i d l y and adopted a fusiform morphology with r u f f l e d c e l l edges and cytoplasmic extensions (Figure 20). In crowded - 101 -Figure 23. Effect of Estradiol ( E 2 ) , EGF and HC on (3H)TdR Incorporation i n Human OSE Histogram showing the effect of E 2, EGF and HC on the incorporation of (3H)TdR i n human OSE cells i n culture. Bars represent mean _+ SEM for 1 experiment only (n = 4 wells/treatment). DPM = disintegrations per minute. Note that EGF appears to be stimulatory i n the absence of E 2; HC alone or with EGF and/or E 2 appears to be stimulatory; and that E 2 appears to inhibit the stimulatory effect of EGF. Figure 23. E f f e c t of E s t r a d i o l (E 2)» Epidermal Growth F a c t o r (EGF), and 3 Hydrocortisone (HC) on ( H)TdR I n c o r p o r a t i o n D P M E2 cone. Supplement 0 100 200 300 400 500 600 700 i I I i i i < • E G F H - f — l H C H - i E G F / H C ' 1 ' 10-1 0M — E G F H C E G F / H C 1—1—i 10" 8 M — i — i E G F H C E G F / H C t-4-H — H 10"6M E G F H C 1 1 , E G F / H C I I i 1 1 i 1 - 103 -cultures on this medium, the cells assumed a very dense, aligned growth pattern. In 15FBS/199/202, LP-9 cells grew more slowly. The cells formed a loosely associated monolayer of irregularly-shaped, flattened c e l l s . Fusiform LP-9 c e l l s grown on 15FBS/199/202/EH assumed a more flattened, epithelial morphology when subcultured of changed to medium without EGF and HC. At the light microscope level, LP-9 cultures were indistinguishable from OSE c e l l s . These results indicate that the culture conditions used in this study for the growth of OSE are comparable to those used by Connell and Rheinwald (1983), despite the different source of the media and other culture materials. In addition, they are i n keeping with the close anatomical and developmental relationship between human OSE and peritoneal mesothelial c e l l s . (B) RAT OVARIAN SURFACE EPITHELIUM (ROSE 239) (1) ROSE 239 CLONES ROSE 239 cells seeded at clonal density (100 to 500 c e l l s / 60 mm culture dish) had an average cloning efficiency of 43*6$ (range: 41$ to 50$) i n one experiment. The clones exhibited one of three distinct growth patterns: compact, cobblestone-like epithelial c e l l s ; irregular (atypical) epithelial c e l l s ; and large, flattened epithelial cells (Figure 24). These growth patterns were strikingly similar to those observed in primary cultures of human ovarian surface epithelial cells (Auersperg et a l , 1984). ROSE 239 colonies with a "compact" growth pattern were composed of cells of a similar size and shape. Adjacent cells were closely apposed and formed a characteristic cobblestone pattern (Figure 24a). Colonies edges were typically smooth in outline. "Atypical" colonies were composed of irregularly shaped cells which generally did not form a confluent sheet. The colonies - 104 -Figure 24. Morphological Forms Exhibited by Clones of ROSE 239 Photomicrographs of clonal colonies of ROSE 239 ce l l s , grown for 14 days i n 10FBS/WM then fixed i n 10$ formol saline and stained i n 0.025$ crystal violet. ROSE 239 clones exhibit one of three morphological forms. a. Colonies with an epithelial, cobblestone growth pattern (xl60). b. Colonies with an irregular or atypical growth pattern (xl60). c. Colonies with a very flattened morphology (xl60; inset x385). - 105 - 106 -varied i n shape and had irregular edges (Figure 24b). "Flat" colonies were generally small, rarely exceeded 12 c e l l s , and were composed of very large, flattened cells (Figure 24c). When stained with crystal violet these cells exhibited stress fibers and poorly stained cytoplasm suggestive of nondividing, aging c e l l s . Several colonies (derived from single cells) exhibiting atypical and compact morphologies were isolated using a cloning cylinder and propagated. By this method, a clone with the atypical growth pattern (R0SE-B2) and a clone with the compact growth pattern (R0SE-C4) • were established. These clones retained their respective morphologies for at least 5 passages. Several attempts were made to clone f l a t colonies, but these were unsuccessful. This may be a result of the small c e l l number or perhaps a reflection of their state of aging or differentiation. R0SE-C4 ce l l s , when seeded sparsely, formed mainly compact colonies (Figure 25a) similar to the colony from which i t arose. Sparse R0SE-B2 cells displayed an irregular growth pattern (Figure 25c). When both of these c e l l lines reached confluence, they formed cobblestone-like monolayers indistinguishable from one another (Figure 25b and 25d).. Confluent R0SE-B2 cells appeared to have a smaller surface area (more columnar ?) than confluent R0SE-C4 c e l l s . The two clones remained morphologically similar for up to 3 weeks, whether the medium was changed at 3 to 4 day intervals or not changed at a l l (starved cultures). In a preliminary experiment, the majority of both R0SE-B2 and R0SE-C4 cells have been shown to express cytokeratin filaments by immunofluorescence (P. Kruk, pers. comm.). When R0SE-C4 and R0SE-B2 lines were grown to confluence and then starved for longer periods of time, differences in the growth characteristics of the two lines became apparent. R0SE-C4 cells maintained a confluent monolayer for almost 2 months without medium change (Figure 26). The cells remained - 107 -Figure 25. Morphology of R0SE-C4 and R0SE-B2 i n Sparse and Confluent Cultures Phase photomicrographs of liv e cultures of the clonal lines R0SE-C4 and R0SE-B2. a. R0SE-C4 i n sparse culture, b. Confluent culture of R0SE-C4-c. R0SE-B2 i n sparse culture. d. Confluent culture of R0SE-B2. Note that R0SE-C4 cells have a compact, epithelial morphology i n both sparse and confluent cultures, while R0SE-B2 cells are more irregularly shaped i n sparse culture but modulate to a compact morphology i n confluent culture. - 108 -F i g u r e 25. - 109 -Figure 26. Confluent Cultures of R0SE-C4 Maintained Without Addition of Culture Medium Phase photomicrographs of liv e cultures of R0SE-C4 which have had no culture medium added for a period of: a. 2 weeks; b. 5 weeks; and c. 7 weeks (xl30). Note that the cells remain as a confluent monolayer despite nutritional starvation. F i g u r e 26. - 110 -- I l l -viable and could be readily subcultured after this starvation period. However, i n starved R0SE-B2 cultures the medium became more acidic after 1 to 2 weeks and the cells began to shed into the culture medium leaving small holes i n the monolayer (Figure 27a). Shed cells did not appear to be viable because they did not attach or divide i f replated i n fresh medium (on the basis of 1 experiment). After 5 weeks most of the R0SE-B2 cells were in the culture medium and only strips of cells remained attached to the culture flask (Figure 27b). When fresh medium was supplied to cultures starved for 8 weeks, a confluent monolayer reformed within a few days (Figure 27c). When confluent cultures of R0SE-C4 and R0SE-B2 ce l l s were given fresh culture medium at 3 to 4 day intervals, differences between the two lines were again evident. R0SE-C4 cells again remained unchanged as a confluent monolayer for up to 2 months (Figure 28). R0SE-B2 ce l l s , however, began to form multilayered ridges of cells after 4 to 5 weeks at confluence (Figure 29a). Monolayered areas of cobblestone-like cells were present between the ridges. After 2 months, the ridges became progressively larger, and more dense but the background of monolayered cells remained (Figure 29b and 29c). This growth pattern i s reminiscent of another rat ovarian surface epithelial c e l l line (ROSE-199) grown under similar conditions (Adams and Auersperg, 1985) which histologically and ultrastructurally resembled papillary cystadenomas seen i n the human ovary. (2) MORPHOLOGY OF CLONES OF R0SE-C4 AND R0SE-B2 When cultures of the R0SE-C4 and R0SE-B2 clones were reseeded at clonal density (within 5 passages of being established) and allowed to grow for 7 days, the growth patterns of the majority of the resulting clones resembled those of the parent clone (Figure 30). The pooled results of 3 experiments - 112 -Figure 27.. Confluent Cultures of R0SE-B2 Maintained Without Addition of Culture Medium Phase photomicrographs of liv e cultures of R0SE-B2 cells which have had no medium added for a period of: a. 2 weeks; and b. 5 weeks, c. Culture of R0SE-B2 3 days after a medium change, prior to which the culture was starved for 8 weeks (xlJO). Note that R0SE-B2 cells begin to die off after several weeks of starvation but quickly reform a confluent monolayer when fresh culture medium is added to the starved culture. - 114 -Figure 28. Confluent Cultures of E0SE-C4 Cells Fed Biweekly-Phase photomicrographs of liv e confluent cultures of R0SE-C4 cells receiving fresh culture medium biweekly. a. 5 weeks. b. 7 weeks (xlJO). Note that R0SE-C4 cells maintain a confluent monolayer despite addition of new culture medium. - 115 -Figure 28. - 116 -Figure 29- Confluent Cultures of R0SE-B2 Cells Fed Biweekly-Phase photomicrographs of live confluent cultures of R0SE-B2 cells receiving fresh culture medium biweekly, a. 5 weeks (xl^O). b. 7 weeks (xl30). c. 7 weeks (x320). Note that after 5 weeks i n culture, the ce l l s begin to retract from the culture dish to form multilayered ridges of very dense c e l l s . Between the ridges, the cells exhibit a compact, epithelial morphology similar to confluent R0SE-C4 cultures (Figure 28). F i g u r e 29. - 117 -- 118 -Figure 30. Growth Patterns of Clones of R0SE-C4 and R0SE-B2 Histogram showing the growth patterns exhibited when R0SE-C4 and R0SE-B2 7 days a f t e r r e p l a t i n g the c e l l s at c l o n a l density (100 c e l l s / c u l t u r e d i s h ) . Bars represent the mean number of colonies (expressed as % of t o t a l colonies/dish) +_ SEM f o r 3 experiments with 3 replicates/experiment. Note that clones of the c e l l l i n e e x h i b i t i n g a compact growth pattern (R0SE-C4) were p r i m a r i l y of the compact form (p<0.0l), with some f l a t and a t y p i c a l colonies also present. Clones of the a t y p i c a l c e l l l i n e (R0SE-B2) were mainly of the a t y p i c a l form (p<0.0l), with some f l a t and compact colonies also present. - 119 -Figure 30. Growth Patterns of Clones of R0SE-C4 and R0SE-B2 Average number of colonies (y. total) 0 2 0 4 0 6 0 8 0 1 0 0 r S I N G L E F L A T C O M P A C T A T Y P I C A L A T Y P I C A L / F L A T O T H E R H R O S E - C 4 S I N G L E F L A T C O M P A C T A T Y P I C A L A T Y P I C A L / F L A T O T H E R ROSE-B2 3^ L - 120 -revealed that clones derived from the R0SE-B2 parent clone had primarily (p<0.00l) atypical growth patterns (66.2$ of total colonies). However, some colonies with f l a t (15.4$) and compact (8.2$) morphologies also formed. Similarly, clones derived from the R0SE-C4 parent clone produced mainly (p<0.00l) colonies with compact growth patterns (54.0$). Again, some atypical (16.9$) and some f l a t (l5«4$) colonies also formed. Thus, these c e l l lines appeared to modulate from one morphology to another during the course of only a few c e l l divisions. Clones derived from R0SE-B2, cultured for 14 days, became very dense in the center and the c e l l s began to pile up and form ridges, much like those observed i n R0SE-B2 cultures that were repeatedly changed. R0SE-C4 clones grown for 14 days did not exhibit this growth pattern. When clones of the parent c e l l line - ROSE 239 - were monitored from a single c e l l to large colony, changes in c e l l morphology occurred over time. Figure 31 shows a single c e l l that divided to form a small colony of large flattened c e l l s . The flattened cells proliferated to form a colony with a compact growth pattern which later assumed a more atypical morphology. Other clones grew into atypical colonies which later assumed a more compact morphology. In general, flattened cells appeared early in the growth of most clones regardless of their eventual growth pattern. -However, i n clones grown for longer periods (14 days) there an increased number of flattened cells appeared within compact and atypical colonies. Apparently, the progeny of the ce l l s within a l l clone types, compact, atypical, and f l a t , were able to modulate into each of the other morphologies. However, since many colonies of flattened cells failed to expand beyond 12 c e l l s , the assumption of the f l a t phenotype, in such cases, appeared to be irreversible, and perhaps represented senescent c e l l s . - 121 -Figure 31. Morphology of a Growing ROSE 239 Clone Phase photomicrographs of a single ROSE-239 clone taken as the colony grew from 2 cells (a) to large colony (d) (x80). Note that the morphology of the cells of clone change from large, flattened cells (b) to a compact, cobblestone epithelium (c) to a mixture of fusiform and compact cells (d). Other clones were observed to change from f l a t to atypical to compact (not shown). - 122 -Figure 31. - 123 -DISCUSSION (A) HUMAN OVARIAN SURFACE EPITHELIUM (l) EFFECT OF CULTURE MEDIA ON OSE CELL GROWTH Historically, the establishment of replicative cultures of normal human epithelial cells has proven to been a d i f f i c u l t task. Selecting an optimal nutrient formulation (Hammond et a l . , 1984; Tsau et a l . , 1982); adjusting the Ca + + concentration (Lechner et a l . , 1982; Tsau et a l . , 1982); lowering the pH of the medium (Peehl and Ham, 1980a); coating culture dishes with a variety of extracellular matrix components (Lechner et a l . , 1981); supplementing with specific growth factors, hormones and conditioned medium (Connell and Rheinwald, 1983; Kirk et a l . , 1985; Stampfer et a l . , 1980; Tsau et a l . , 1982); and co-culturing with feeder cells (Rheinwald and Green, 1975; Taylor-Papadimitriou et a l . , 1977), are examples of techniques used to maintain a population of growing epithelial c e l l s . In this report, a synthetic culture medium, supplemented with FBS and selected* mitogens, has been selected that permits serial cultivation of normal human OSE cells i n culture for the f i r s t time. This culture medium not only stimulates OSE c e l l growth but also delays senescence of OSE in culture. It consists of a mixture of two synthetic culture media, Medium 199 and MCDB 202 ( l : l ) , 15$ FBS, 20 ng/ml EGF and 0.4 vg/ml HC. This medium has previously been used successfully for the culture of human peritoneal mesothelial cells in culture (Connell and Rheinwald, 1983). In this medium OSE ce l l s : (i) can be subcultured 8 to 10 times; ( i i ) undergo approximately 20 - 25 population doublings before senescence; ( i i i ) have an average population doubling time of 48 hours; (iv) have a seeding efficiency of up to 53$ (when seeded at clonal density); and (v) have a cloning efficiency of up to 13$. Prior to this - 124 -research, i t was possible to maintain OSE cells i n • culture for only a few weeks before c e l l division ceased and the cells could not be successfully subcultured (Auersperg et a l . , 1984a). One of the prerequisites for defining a medium for improved c e l l growth i s to obtain a sufficient level of replication to perform the needed assays. Fortunately, i t became apparent very early i n this study that the 199/202 medium, EGF, and HC greatly improved OSE c e l l proliferation and culture lifespan, so these conditions could be used i n stock cultures to obtain enough OSE for experimentation. The growth of OSE achieved i n this medium compares favourably, by the parameters examined, with the growth of other human epithelia and the OSE of other species, i n culture (Adams and Auersperg, 1981a; Connell and Rheinwald, 1983; Lechner et a l . , 1982; Maciag et a l . , 1981; Rheinwald and Green, 1980). The loss of OSE growth capacity upon continued c e l l passage could be due to an inherent limitation of c e l l division, or to the absence of necessary nutritional factors, appropriate substratum, or c e l l - c e l l interactions, or a l l of these. Because of the complexity of the synthetic culture media, and the nature of experiments used i n this study, i t i s d i f f i c u l t to determine which component(s) of 15FBS/199/202/EH are responsible for the success of this medium in the growth of OSE. In mass culture, c e l l growth in the absence of EGF and HC was significantly greater in 5FBS and 15FBS/199/202 than in the other media with the same FBS supplement. This suggests that the 199/202 synthetic medium contains more of the nutrients (in appropriate quantities) required for OSE growth, than any of the other media. The 202 portion of this medium was i n i t i a l l y designed for the growth of human diploid fibroblasts and later modified for chicken embryo fibroblasts (McKeehan and Ham, 1977; McKeehan et a l . , 1977), but has recently been used alone, or in combination with other media, for the culture of several human epithelia (Connell and - 125 -Rheinwald, 1983; Hammond et a l . , 1984). Qualitatively, 202 contains : (i) trace elements (Cu, Mn, Si, Sn, Mo, V, Se, Ni, and Zn); ( i i ) pyruvate; and ( i i i ) Hepes buffer, while the other media do not. Quantitatively, this medium has a higher C a + + (2mM) content than WM (0.8mM) and M199 (l.lmM), a lower glucose (8mM) content than WM(29mM), and a higher glucose content than 199 (5.7mM). Ca has been shown to be important i n selective growth of specific c e l l types and i n modulation of cellular differentiation i n culture (Ham, 1984). There are numerous other quantitative differences i n other components such as amino acids and vitamins. OSE growth in media containing 199 or 202 alone (with FBS and mitogens), was not nearly as successful as 199/202 mixed 1:1, suggesting that quantitiative, not qualitative, differences i n the constituents of the medium are responsible for i t s success. This has been shown to be true for other human epithelial c e l l types (Ham and McKeehan 1978; Hammond et a l . , 1984; Tsau et a l . , 1982). Minor changes in pH have been shown to affect the growth of some c e l l types i n culture (McKeehan et a l . , 1977). Unlike conventional carbon dioxide-bicarbonate buffering systems, Hepes buffer minimizes large changes i n pH that occur during the accumulation of cellular metabolic products, and when cultures are removed from the incubator (Ham and McKeehan, 1978). Its presence in 202 may have been partially responsible for the success of this medium for OSE growth. OSE c e l l growth was poor i n Waymouth's 752/1 medium, even when supplemented with 25% FBS, EGF and HC. This i s perhaps not surprising considering that Waymouth's medium is a comparatively simple medium, developed i n i t i a l l y for the growth of mouse NCTC cells (Waymouth, 1959). It contains no trace elements or pyruvate, and has a conventional carbon dioxide-bicarbonate buffering system, a high glucose (29mM) and low Ca + + (0.8mM) content. - 126 -Because 15EBS/199/202/EH was chosen essentially by " t r i a l and error", i t probably does not yet represent an optimal medium for OSE culture. However, in i t s present form, i t supports a level of OSE proliferation sufficient for numerous applications, and provides a basis for further investigation into the specific growth requirements of OSE i n culture. Additional nutritional factors that should be considered i f this medium is to be optimized further are: (i) hormones present i n the ovary, such as testosterone and 5<*-dihydrotestosterone, which have been shown to stimulate the growth of rat OSE i n culture (Hamilton et a l . , 1980), and progesterone, follicle-stimulating hormone, and luteinizing hormone; ( i i ) PDGF, which i s released during blood clotting and therefore probably present at ovulation; ( i i i ) insulin, which stimulates the growth of other human epithelia in culture (Ham, 1984; Stampfer et a l . , 1980); and (iv) transferrin, which is mitogenic to other human epithelia i n culture (Hammond et a l . , 1984; Lechner et a l . , 1982). The relatively low average colony-forming efficiency, and the small size and dispersed growth pattern of OSE clones grown at clonal density, may be related to the fact that conditions that support the growth of large c e l l populations are not necessarily optimal for clonal cultures (McKeehan and Ham, 1977). According to Ham (1981), i n moderately crowded cultures where the volume of medium per c e l l i s not too large, complex interactions occur between the cells and the medium, which are collectively known as "conditioning" of the medium, a process often essential for cellular multiplication. The two main features of conditioning are detoxification, and release into the medium of a set of metabolic intermediates that are synthesized by the c e l l s , and accumulate in the conditioned medium. Under clonal conditions, the volume of medium per c e l l i s so large that essentially no conditioning can be accomplished. Cellular proliferation can occur under these conditions only i f - 127 -the medium as formulated i s free from toxicity and satisfies the growth requirements of the c e l l s . Included i n the "nutrients" that must be provided by the medium are carbon dioxide, pyruvate (or other 2-oxocarboxylic acids), and nonessential amino acids that tend to be lost into the surrounding medium. Since a sufficient level of OSE replication has now been achieved i n mass culture, i t may now be advantageous to systematically analyze the growth promoting effects of different medium components in clonal cultures, in a manner similar to that recommended by McKeehan and Ham (1977) and Ham (1984). The seeding efficiency of OSE i n clonal culture (100 cells/dish) was less dependent on the type of medium or the quantity of the serum added to the medium than was colony formation. One exception was WM, in which seeding efficiency was significantly decreased with lower serum supplements. Apparently, the requirements for attachment of OSE cells to plastic are far less stringent than for c e l l proliferation in clonal culture. In primary culture, the number of explants producing OSE outgrowths and the extent of OSE outgrowths was variable in a l l media, including 15FBS/199/-202/EH. In one case, no explants produced OSE outgrowths. Such variation i s perhaps not surprising considering the large number of variables potentially introduced by (i) the donor's age and stage of menstrual cycle; ( i i ) random sampling from the anatomically and morphologically diverse ovarian surface; and ( i i i ) the c l i n i c a l disorder of the patient. (2) EFFECT OF EGF AND HC ON OSE CELL GROWTH The most dramatic improvement in OSE c e l l growth and culture lifespan was achieved by the addition of EGF and HC to the culture media. These factors are mitogenic to other types of human epithelia i n culture including: epidermal keratinocytes (Rheinwald and Green, 1977), peritoneal mesothelial - 128 -c e l l s (Connell and Rheinwald, 1983), p l e u r a l mesothelial c e l l s (Lechner et a l . , 1985), bronchial epithelium ( S i e g f r i e d and Nesnow, 1984), corneal endothelium (Gospodarowizc, 1977), mammary epithelium (Hammond et a l . , 1984), and others (Gospodarowicz et a l . , 1978; C r i s t o f a l o and Rosner, 1981). Although only one concentration of EGF (20 ng/ml) and HC (0.4 ug/ml) were used i n t h i s study, these concentrations were comparable to those used i n other human e p i t h e l i a l culture systems: EGF, 5 - 3 0 ng/ml and HC, 0.4 - 0.5 ug/ml (Connell and Rheinwald, 1983; Hammond et a l . , 1984; Rheinwald and Green, 1977; Tsau et a l . , 1982; S i e g f r i e d and Nesnow, 1984). Since the mitogenic response to these factors i n d i f f e r e n t c e l l types v a r i e s with the concentration of the f a c t o r ( i . e . high concentrations may be i n h i b i t o r y ) (Rheinwald and Green, 1977), the e f f e c t of a range of concentrations on OSE c e l l growth should be examined. Although the i n d i v i d u a l mitogenic e f f e c t s of EGF and HC were not examined i n c l o n a l or mass cultures, a preliminary assessment the i n d i v i d u a l e f f e c t s of these mitogens can be made from the experiment on the e f f e c t of on ( H)TdR incorporation. I n d i v i d u a l l y , both EGF and HC enhance ( H)TdR incorporation considerably, but when added together the stimulatory e f f e c t i s more than a d d i t i v e . S i n g l e t a r y et a l . (1985) noted a s i m i l a r stimulation of c e l l growth when EGF and HC were added i n combination rather than separately to human tumour c e l l s i n c u l t u r e . The stimulatory e f f e c t of glucocorticoids appears to be mediated through the regulation of the c e l l ' s responsiveness to a second hormone or growth f a c t o r ( C r i s t o f a l o and Rosner, 1981). Baker et a l . (1978) showed that g l u c o c o r t i c o i d s , i n the presence of EGF, stimulate c e l l d i v i s i o n and suggested the enhancement of binding of EGF to surface receptors as a mechanism. A l t e r n a t i v e l y , there i s evidence from studies using r a t mammary epithelium that EGF and HC may act v i a t h e i r a b i l i t y to enhance the production of - 129 -basement-membrane collagen substratum (Kidwell et a l . , 1982). HC may act by blocking collagen turnover, while EGF may act by enhancing collagen synthesis. However, the mechanism underlying the stimulatory effects of glucocorticoids i s poorly understood. The observed requirement of OSE cells for EGF and HC even i n the presence of large amounts of serum i s common to other human epithelial c e l l types i n culture (Rheinwald and Green, 1975; Connell and Rheinwald, 1983). This suggests that EGF i s not present i n an appreciable amount in the serum used i n these experiments, and that other factors present i n the serum can not replace EGF as a requirement for OSE growth. The recent report by Oka and Orth (1983) that EGF is not present in large quantities i n human plasma, but i s released from platelets during the clotting process suggests that EGF i s only required by many cells in vivo for growth during wound repair. In the ovary, EGF (and perhaps platelet-derived growth factor) are therefore l i k e l y present i n increased amounts due to the hemorrhaging of small vessels that occurs when the f o l l i c l e ruptures (Gillet et a l . , 1980). Localized stimulation of OSE proliferation by these factors could result in the observed rapid repair of the ovulatory defect. In the epidermis of the skin, wound healing has similarly been shown to be enhanced by EGF, through stimulation of epidermal keratinocyte proliferation (Buckley et a l . , 1985). In the ovary, the change i n the substratum underlying the OSE from basal lamina to a fibrin-clot substratum at ovulation, may also influence OSE organization and proliferation. Fibrin clot has been shown to alter the growth pattern of vascular endothelial and thyroid f o l l i c u l a r cells i n culture (Garbi et a l . , 1984; Kadish et a l . , 1979). In some c e l l types i t has been shown that EGF may act as a substitute for other factors that normally stimulate growth in culture (LaRocca and Rheinwald, 1982). For example, thrombin has been reported to largely - 130 -compensate for the absence of EGF i n a serum-free medium for hamster fibroblasts (Cherington and Pardee, 1980). Similarly, i n OSE cultures EGF and HC may i n part substitute for other unidentified factors that normally stimulate OSE c e l l growth in vivo. Recent reports indicate that EGF may have a physiological role in the control of ovarian function. EGF inhibits the gonadotropin induction of LH receptors (Mondschein and Schomberg, 1981; Symanski and Schomberg), inhibits estrogen production (Hseuh et a l . , 1981), inhibits ovarian theca i n t e r s t i t i a l cytodifferentiation (Erickson and Case, 1983), induces the maturation of rat follicle-enclosed oocytes (Dekel and Sherizey, 1985), and stimulates progestin biosynthesis (Knecht and Catt, 1983). In addition, EGF receptors have been localized i n the theca interna cells of atretic f o l l i c l e s and i n luteal cells of the corpus luteum of the rat (Chabot et a l . , 1985). Also, EGF receptors appeared to be present i n the OSE in autoradiographs in a paper by Chabot et a l . (1985), although i t is not mentioned in the text. In the present study, EGF stimulated c e l l division of human OSE i n culture, but the mitogenic effect of EGF appeared to be inhibited by 17S-estradiol (Eg). This suggests a possible hormonal role of EGF i n the OSE. Eg has also been shown to inhibit the mitogenic effect of EGF in Xenopus hepatocytes in culture, apparently by a reduction in the number or a f f i n i t y of c e l l surface EGF-receptors (Wolffe et a l . , 1985). Steroid hormones are known to affect the level of EGF receptors i n other c e l l types and in other tissues (Chabot et a l . , 1985; Sadiq and Devaskar, 1984). In addition to i t s mitogenic effects, EGF inhibits the gonadotropin stimulation of estrogen synthesis in primary cultures of mouse granulosa cells (Hsueh et a l . , 1981). In the ovary, Eg i s produced prior to ovulation by the cells of the maturing f o l l i c l e and after ovulation by the corpus luteum. In the f i r s t part of the ovarian cycle, the circulating levels - 131 -of Eg increase toward ovulation then decreases to a low level during and shortly after ovulation (Guraya, 1985; Greenwald, 1980). Possibly, increased levels of EGF at ovulation, by release from platelets, may simultaneously inhibit Eg production by granulosa cells and stimulate the proliferation of OSE c e l l s , while Eg levels are low. In vivo the OSE cells surrounding the base of the f o l l i c u l a r defect and overlying the recently established corpus luteum have apical cytoplasmic extensions (Motta and Van Blerkom, 1980). This phenomenon has been shown, in other c e l l types in vitro, to be involved in the enhanced endocytotic activity, and endocytosis and ruffling can be stimulated by various growth factors,, including EGF (Myrdal and Auersperg, 1986; Wiley and Cunningham, 1982). (3) SERUM SUPPLEMENTATION According to McKeehan et a l (1977), the apparent requirement of normal cells for high concentrations of serum protein i s due to the use of inadequate culture media. That i s , large amounts of serum protein enable cells to grow in media which are not well suited to their nutritional needs. When the medium i s nutritionally complete and qualitatively adjusted to the needs of the c e l l s , the actual amount of serum protein required for c e l l growth i s quite small. The amount of serum used in the present culture system, 15% FBS, i s high but comparable to that used for other human epithelia i n culture, particularly before a serum-free or serum-reduced medium had been defined (Connell and Rheinwald, 1983; Hamilton et a l . , 1980; LaRocca and Rheinwald, 1985; Maciag et a l . , 1981). The serum supplement used here i s considerably lower than that previously used for OSE cells (25$ FBS). In addition, the results for OSE growth mass culture suggest that 5% FBS may provide an acceptable level of OSE proliferation, at least when the cells are under - 132 -relatively crowded conditions. However, at clonal density lowering the FBS supplement from 15$ to 5$ significantly reduces the clonal growth of OSE. This suggests that some nutrients important to OSE growth i n clonal culture are either lacking or present in suboptimal quantities. In retrospect, the presence of small amounts of E^ (10 pg/ral), f o l l i c l e stimulating hormone (36.02 ng/ml), testosterone ( l l ng/dL), progesterone (10 ng/dL), and perhaps other components in the serum used in this study (HyClone lot #110451), may have modulated the mitogenic effect of EGF and HC. This could be tested by using dextran/charcoal-stripped serum. The presence of serum i n the medium w i l l not affect the potential of OSE cultures for use in carcinogen assays, but for other applications a reduction i n the serum supplement in the medium may be advantageous. (4) OSE MORPHOLOGY IN CULTURE Embryologically, the ovarian surface epithelium, peritoneal mesothelium, and the epithelia of the Mullerian duct derivatives (endometrial, endosalp-ingeal, and endocervical epithelia) arise from the mesothelial lining of the intraembryonic coelom. That they are a l l epithelia of mesodermal origin may, i n part, explain the unusual properties of OSE (and peritoneal mesothelium) in culture. In the absence of EGF and HC OSE cells exhibited an epithelioid morphology and expressed high levels of cytokeratin filaments arranged i n a filamentous pattern throughout the cytoplasm. When EGF and HC were added to the culture medium, OSE cells assumed a fibroblastoid morphology, and the keratin composition of the majority of the cells changed to a l i g h t l y stained, diffusely distributed pattern. Prior to this research, the reversible modulation of cytokeratin expression was thought to be unique to normal human peritoneal mesothelial cells i n culture (Connell and Rheinwald, 1983). In - 133 -medium containing EGF and HC, human mesothelial cells reportedly grew rapidly, adopted a fibroblastic shape, and showed decreased keratin and increased vimentin synthesis. In the absence of EGF and HC, the cells grew more slowly, had an epithelial morphology, and a high keratin, low vimentin content. The observed filament changes were reversible. Klymkowsky (1982) has shown that intracellular injection of monoclonal antibodies against vimentin or keratin filaments disrupts the organization of both filament types i n cultured PtKg epithelial c e l l s , suggesting that these filaments networks are functionally interrelated. Rat OSE cells i n culture express vimentin filaments (A. Hornby, pers. comm.) and human OSE i n vivo express vimentin, desmoplakins, and four subclasses of cytokeratins (subclasses 7, 8, 18, and 19) (Czernobilsky et a l . , 1985). Coexpression of keratin and vimentin i n normal epithelia i n vivo i s apparently a rare phenomenon, however i t has been demonstrated i n some epithelial c e l l cultures (Czernobilsky et a l . , 1985). Normal human peritoneal mesothelial cells (Strain LP-9) in culture express the same subclasses of cytokeratins as OSE i n vivo (Connell and Rheinwald, 1983; Czernobilsky et a l . , 1985) and are morphologically indistinguishable at the light microscopic level from human OSE when cultured i n the same medium. OSE cells in seri a l cultures lack the cuboidal morphology typical of OSE i n vivo, and resemble peritoneal mesothelial c e l l s , suggesting modulation to a epithelium of less differentiated form. It has been noted previously that human OSE cells in primary culture exhibit a number of morphological forms that are not attributable to known differences i n culture conditions (Auersperg et a l . , 1984). The response of primary cultures of OSE to EGF and HC appears to be variable, and not a predictable change from epithelial to fusiform as i n secondary cultures. In the presence of EGF/HC, OSE cells i n some outgrowths have an irregular, - 134 -elongate morphology, but do not align i n parallel -arrays like fibroblast outgrowths. In other OSE outgrowths, the cells maintain a typical cobblestone growth pattern i n the presence of EGF/HC. Both of these growth forms have previously been shown to contain cytokeratin filaments by immunofluorescence (Auersperg et a l . , 1984a). These observations and the fact that cobblestone OSE cells often became fusiform within 48 hours after removing the explant suggest that the explant i s influencing OSE morphology, perhaps through hormone, growth factor, or extracellular matrix production by the cells within the explant, or by some mechanical means, or perhaps by information passed via gap junctions from the cells on the explant. Although poorly understood, the a b i l i t y of OSE to maintain an epithelial morphology in the presence of EGF/HC may be important i n understanding the genesis of OSE-derived tumours. For example, i f the extracellular matrix underlying OSE in vivo i s necessary to maintain the cuboidal morphology in the presence of EGF, alteration of cellular receptors for the ECM may influence the response of the OSE to growth factors. Unregulated c e l l proliferation could possibly ensue. Changes in cellular morphology in response to culture conditions are not a property unique to human OSE. Others have shown that c e l l shape can be modulated by the adhesiveness of the substratum (Folkman and Moscona, 1978), ++ by addition of EGF to the culture medium, and by altering the Ca content of the culture medium (Marceau and Swierenga, 1985). In culture, c e l l shape appears to play a regulatory role in c e l l growth. Folkman and Moscona (1978) showed that by plating cells on culture dishes of different adhesiveness (precoated with graded concentrations of the polymer poly(2-hydroxyethyl-methacrylate), c e l l spreading could be restricted to provide a graded series of c e l l shapes. Using this method i t was shown that there i s a direct correlation between the i n i t i a t i o n of DNA synthesis and the amount of c e l l - 135 -spreading. The mechanisms underlying the relationship between changes in c e l l shape and DNA synthesis are poorly understood, but there i s increasing evidence that cytoskeletal components are involved i n the i n i t i a t i o n of DNA synthesis (Otto, 1982; Marceau and Swierenga, 1985). For example, when EGF or C a + + are added to cells growth-arrested through reduced C a + + levels, the cells undergo a modulation in c e l l shape, prior to the i n i t i a t i o n of DNA synthesis, which i s associated with marked changes in the organization of the cytoskeletal components (Marceau and Swierenga, 1985). The cytoskelelton i s a complex structure composed of microtubules, microfilaments, and intermediate filaments. The way in which each of these elements participates i n the events that lead to i n i t i a t i o n of DNA synthesis and mitosis, as well as the molecular basis underlying the i n i t i a t i o n process, i s not well understood. However, the phosphorylation of cytoskeletal elements i s thought to be involved (Marceau and Swierenga, 1985). EGF i s a small peptide growth factor that binds to c e l l surface receptors. The main feature underlying the biological action of EGF i s i t s capacity to activate protein kinases, including tyrosine specific phosphorylation of the EGF-receptor, i t s e l f a protein kinase (Hunter and Cooper, 1985). Activation of protein kinase activity leads to a cascade of intracellular events which either directly, through the protein kinase activity of the EGF receptor, or indirectly through the activation of a number of cAMP-dependent protein kinases, appears to result in the phosphorylation of such cytoskeletal elements as microtubules or intermediate filaments (Marceau and Swierenga, 1985). Recently, Aoyagi et a l (1985) have shown that EGF stimulates tyrosine phosphorylation of cytokeratins i n pig epidermal keratinocytes. Phosphorylation of cytoskeletal elements may invoke permanent changes in cytoskeletal structures resulting in release of the c e l l from - 136 -adhesion plaques, a reduction i n c e l l - t o - c e l l communication, and separation of the centrioles (Marceau and Swierenga, 1985). A l l of these constitute early-events related to c e l l commitment to division, and may allow the c e l l to progress through the DNA prereplicative period. Thus, i t appears that the cytoskeleton may have a direct role in the the transmission of proliferative signals from external receptor sites to the nucleus. In addition to the effect of c e l l shape on DNA synthesis, c e l l shape is also thought to be important in the regulation of gene expression. Ben-Ze'ev (1985a) showed that changes in c e l l shape and c e l l - c e l l interactions e l i c i t regulatory responses of cytoskeletal and other genes. Cytoskeletal elements are thought to regulate gene expression. According to Ben Ze'ev (1985b), extensive c e l l - c e l l interaction may induce the synthesis of cytokeratins, perhaps mediated by the formation of desmosome-type intercellular junctions in which keratin filaments often terminate. This i s in keeping with the enhanced expression of cytokeratins observed in slowly-dividing mesothelial and OSE c e l l s , and the decreased expression of cytokeratins in rapidly dividing cells i n culture. Whether modulation of OSE c e l l shape and cytokeratin expression are equally important in vivo i s not known. However, rapidly-dividing OSE modulates to a more flattened morphology in the repair the ovulatory defect (Motta and Van Blerkom, 1980), suggesting perhaps that exogenous factors such as EGF may play a similar role in vivo. In this study, some apparently pure OSE cultures did not undergo senescence after repeated passage (up to 10 passages then stored by freezing) i n the presence of EGF and HC, but instead appeared to lose their responsiveness to the mitogens. When EGF and HC were removed, the cells continued to grow rapidly, did not revert back to a more epithelial growth - 137 -pattern, and were negative for keratin filaments, by immunofluorescence. Although possible overgrowth by fibroblasts can not be eliminated, i t i s possible that these OSE cells were permanently "transformed" to a fast-growing, fusiform growth pattern, perhaps comparable to the stromal component of some OSE-derived tumours. However, further characterization of these cultures i s required to determine i f these cells are i n fact, fusiform OSE cells with a low cytokeratin content, or fibroblasts. (6) APPLICATIONS OF HUMAN OSE IN CULTURE (a) Short-term Carcinogen Assays Human OSE i n vivo regularly "cycles" between two proliferative states; (i) stationary or slowly-dividing, as over the general ovarian surface; and ( i i ) rapidly dividing, as in repair of the ovarian surface after ovulation. Similarly, human OSE cells i n culture can be maintained in either a stationary or slowly-dividing condition, in WM with low FBS supplementation, or as rapidly proliferating c e l l s , i n 15FBS/199/202/EH (Figure 12). Therefore, not only do the conditions defined in this study for the culture of OSE mimic the proliferative states of OSE in vivo, but they may also permit the application of short-term carcinogen assays that examine carcinogenic activity under conditions similar to those present in the human ovary. Rapidly-dividing OSE ce l l s , with a population doubling time of about 48 hours, should be suitable for several carcinogen assay tests that require an increased level of c e l l division. For example, tests assessing sister chromatid exchange, micronucleus production, or chromosomal aberrations after exposure to a carcinogen should be applicable (Heddle et a l . , 1978; Wolff, 1981). Stationary OSE c e l l s , cultured in WM are appropriate for carcinogen assay tests that assess DNA-repair synthesis after carcinogen exposure (Williams, - 138 -1985). If the level of replication i n these slowly dividing cultures i s too high, DNA replicative synthesis can be arrested by transferring the cultures to an arginine-deficient medium for several days before assaying (Stich and San, 1970). Some chemical carcinogens damage DNA directly, while other compounds require metabolic activation into intermediates before becoming active mutagens. Often c e l l s i n culture lose their capacity to activate certain compounds. Therefore, to ensure the detection of metabolism-requiring agents, test samples should be assayed i n the presence of an exogenous rat l i v e r microsome preparation (Ames et a l . , 1975; R.H.C. San, pers. comm.). The avai l a b i l i t y of human OSE i n tissue culture should prove to be a valuable tool in the testing of presumptive carcinogens thought to be specific for the OSE. Any tests of potential ovarian carcinogens should include talc and asbestos, which i s often found i n cosmetic and pharmaceutical talc preparations (Hildrick-Smith, 1976; Rohl et a l . , 1976; Paolletti et a l . , 1985). These agents are found i n association with diaphrams, condoms, tampons, sanitary napkins, perineal dusting powders and sprays, and in related industries. Talc has been identified i n association with ovarian tumours (Henderson et a l . , "1971; Henderson and Griffiths, 1975; Mostafa et a l . , 1985). Asbestos and talc have been implicated i n ovarian cancer by epidemiological studies (Newhouse et a l . , 1972; Cramer et a l . , 1982). Recently, asbestos has been shown to be cytotoxic and mutagenic to human pleural mesothelial cells in culture (Lechner et a l . , 1985). Likewise, the wide variety of other active agents that could potentially contact the ovarian surface epithelium via the vagina and fallopian tubes, such as components of spermacidal foams or other chemicals present i n feminine hygiene products, should be tested for mutagenic activity. Human OSE in culture could prove to be valuable i n obtaining new information on ovarian carcinogensis, and perhaps - 139 -add to our understanding of the pleomorphic nature of 'common epithelial tumours' of the ovary, and the neoplastic potential of the ovarian surface epithelium. (b) Other Applications Aside for i t s potential for use in the study of ovarian carcinogensis, human OSE i n culture may have other important applications. In recent years, considerable emphasis has been placed on establishing immunological or other biochemical markers for early detection of "common epithelial tumours" of the ovary. Carcinoembryonic antigen (CEA), human chorionic gonadotropin, antigen CA 125, monoclonal antibodies to keratins and a variety of other tumour associated antigens (Blaustein et a l . , 1982; Cooper et a l . , 1985; Heinonen et a l . , 1985; Van Nagell, 1983) have been evaluated as markers for ovarian tumours, but to date no truly tumour-specific markers have been detected. The ava i l a b i l i t y of normal human OSE i n culture may aid i n the identification of biologic markers specific to OSE that are maintained in malignant transformation. The capacity to extend the growth potential of OSE in culture could also greatly f a c i l i t a t e investigations into the biology of the OSE in vivo, which, because of the relatively small amount of OSE present in the human ovary and differences between human OSE and that of other mammals, i s s t i l l poorly understood. OSE i n culture, where conditions • are defined and can be manipulated, may yield new information on the hormone responsiveness and differentiated characteristics of this tissue. - 140 -(B) RAT OVARIAN SURFACE EPITHELIUM The capacity of ROSE-239 and i t s clones to display a range of morphological forms i n culture, i s reminiscent of the pleomorphism exhibited by human OSE i n vivo (Motta and Van Blerkom, 1980; Blaustein et a l . , 1982), and i n "common e p i t h e l i a l tumours" of the ovary (Parmley and Woodruff, 1974). However, the most i n t e r e s t i n g observation was that ROSE 239 and s i n g l e - c e l l -clones of ROSE 239 (R0SE-C4 and R0SE-B2) exhibited three morphological forms (compact, a t y p i c a l and f l a t ) that were s t r i k i n g l y s i m i l a r to outgrowths of human OSE i n primary culture (Auersperg et a l . , 1984a). The d i f f e r e n t growth patterns i n i t i a l l y exhibited by uncloned ROSE c e l l s i n culture suggested e i t h e r that contaminating c e l l types were present or that ROSE c e l l s i n cult u r e were capable of manifesting d i f f e r e n t phenotypes. The a v a i l a b l e evidence supports the l a t t e r hypothesis. F i r s t , when the c l o n a l l y - d e r i v e d l i n e s R0SE-C4 and R0SE-B2 were seeded at c l o n a l density, the resultant colonies had p r i m a r i l y the morphology of the parent clone ( i e : e i t h e r compact or a t y p i c a l ) , but a small proportion of the other morphologies were always present. Second, growing clones changed i n morphology as they proceeded from s i n g l e - c e l l to large colony. And t h i r d , both R0SE-B2 and R0SE-C4 c l o n a l l i n e s were p o s i t i v e f o r k e r a t i n filaments and grew as an cobblestone monolayer when confluent, excluding the p o s s i b i l i t y of contamination by other ovarian c e l l types. The observed pleomorphism i n ROSE c e l l s i n culture suggests e i t h e r ( i ) that subpopulations of c e l l s e x i s t within the mammalian OSE that exhibit d i f f e r e n t but stable morphological phenotypes; or ( i i ) that OSE c e l l s are capable of modulating between d i f f e r e n t morphological forms. Observations of growing clones provided evidence f o r the l a t t e r . Clonal colonies derived from R0SE-C4 and R0SE-B2 existed as small colonies of flattened c e l l s e arly i n - 141 -their "lifespan", then later formed compact or atypical colonies. Eventual compact colonies had an atypical stage in their history, and vice versa. Many of the flattened colonies did not progress to form larger colonies, suggesting that the flattened form, in these cases, represented a loss of replicative potential. Because a l l colonies were exposed to the same culture conditions, environmental factors were probably not responsible for the observed differences i n morphology. Perhaps this apparent multipotentiality of ROSE cells i s an inherent characteristic derived from the multipotential coelomic epithelium from which they arose. Presumably, the different morphological forms observed in ROSE cultures perform different functional roles i n vivo. The morphology of R0SE-C4 cell s , a compact, cobblestone monolayer, is very similar i n appearance to the OSE present over much of the ovarian surface i n most mammalian ovaries, including the human (Van Blerkom and Motta, 1979; Motta and Van Blerkom, 1980). However, the relationship of cells exhibiting the more fusiform or "atypical" morphology, to cells found in OSE in vivo i s not clear. Several studies of OSE i n vivo suggest that OSE cells differ i n their characteristics from c e l l to c e l l . Czernobilsky et a l . (1985) noted differences in cytokeratin and vimentin staining between human OSE cells i n vivo. Blaustein and Lee (1979) observed some variablity i n staining for epithelial mucin and 17S-HSD in human OSE. And, as discussed previously, the morphology of the OSE varies markedly over the surface of the ovary, in association with crypts and cysts, and with the stage of the menstrual cycle. Possibly, the morphological forms exhibited by OSE c e l l s perform different functions which are important at different stages in the ovarian cycle or at a particular position on the ovarian surface. - 142 -Multipotential behaviour of normal epithelial c e l l s i n culture i s not unique to OSE. Cloned c e l l lines of rat mammary epithelium exhibits two morphological forms; cuboidal and elongate (Ormerod and Rudland, 1985). In this model system, cuboidal cells appeared to modulate irreversibly to elongate c e l l s . Bovine vascular endothelial cells i n primary culture generally grow as a single layer of polygonal c e l l s , however elongate or "sprouting" cells appear i n crowded cultures (Cotta-Pereira et a l . , 1980). The "sprouting" cells had properties resembling endothelial cells i n regenerating or recently formed vessels. McAuslan et a l . (1982) demonstrated that c e l l lines of bovine vascular endothelium could undergo reversible changes i n morphologic phenotype, i n response to factors present i n the culture medium. Brown et a l . (1985) found that cloned rat mesothelioma cells modulated from an i n i t i a l epithelial morphology to a mixed epithelial-fibroblastic culture after passaging 20 times i n culture, indicating the multipotential capability of mesothelioma c e l l s . The clonal lines R0SE-C4 and R0SE-B2 in mass culture had very consistent properties that persisted for numerous passages, despite their capacity for modulation to other morphological forms. Confluent R0SE-C4 cultures remained contact-inhibited under starvation conditions and when continuously fed. In contrast, confluent R0SE-B2 cultures, although indistinguishable i n i t i a l l y from confluent R0SE-C4 cultures, appeared to lose their responsiveness to culture conditions; the cells continued to grew rapidly under starvation conditions and when continuously fed. This resulted i n c e l l death and multilayered cultures, respectively. These properties of R0SE-B2 are not unlike those of many malignant c e l l types, in vivo and i n culture. It has yet to be determined whether R0SE-B2 cells are tumorigenic, or i f they exhibit autonomous growth factor (e.g. a-TGF) or extracellular matrix production, - 143 -which may be responsible for their altered growth pattern i n crowded cultures. The fusiform growth pattern and decreased responsiveness to culture conditions of R0SE-B2 suggests modulation to a morphology similar to the stromal component of many OSE-derived tumours (Parmley and Woodruff, 1974)• Further investigation into the factors regulating growth and morphological modulation of R0SE-B2 and R0SE-C4 may provide insight into early disturbances in the regulation of growth and differention of "common epithelial tumours" of the ovary. - 144 -SUMMARY Most human ovarian cancers are thought to arise i n the surface epithelium of the ovary. However, few models for the study of this tissue exist and no methods for i t s propagation i n culture are available. In this report: (i) culture conditions have been defined for the seria l cultivation and clonal growth of normal human ovarian surface epithelium (OSE); and ( i i ) a rat OSE c e l l line (ROSE 239) i s used as a model to examine pleomorphism in human OSE. Growth of human OSE, obtained from explant cultures of premenopausal ovarian tissue, was compared i n several culture media: Medium 199 (199), MCDB 202 (202), Waymouth's 752/1 medium (WM) and 199/202 ( l : l ) with 5, 15, or 25$ fetal bovine serum (PBS), and with or without epidermal growth factor (EGF, 20 ng/ml) and hydrocortisone (HC, 0.4 ug/ml). Extent of OSE outgrowth, plating efficiency, cloning efficiency, and growth curves were used to measure the influence of different culture media on OSE growth. ROSE 239 cells were cultured and cloned in WM with 10$ FBS.. Of the media examined for the culture of human OSE, the medium best suited for OSE growth contains: (i) 199/202 mixed 1:1; ( i i ) 15$ FBS; ( i i i ) 20 ng/ml EGF; (iv) 0.40 ug/ml HC. In this medium, human OSE: (i) underwent 20 - 25 population doublings before senescence; ( i i ) had a population doubling time of about 48 hours in the log phase of growth; ( i i i ) - could be subcultured 8 to 10 times; (iv) had a seeding efficiency of up to 53$; and (v) had a cloning efficiency of up to 13$. This medium was superior to a l l of the other media - 145 -examined under the conditions tested. Human OSE cell's i n culture reversibly modulated their morphology and exhibited altered cytokeratin filament expression i n response to EGF and HC, i n a manner similar to human peritoneal mesothelium in culture. There was also some evidence that 17B-estradiol inhibited the mitogenic effect of EGF on OSE c e l l s . ROSE 239 cells and two clones derived from this line (R0SE-C4 and R0SE-B2) exhibited three morphological forms in clonal culture (compact, flattened and atypical) that resembled the morphological forms of human OSE i n primary culture. The compact and atypical forms of OSE appeared to modulate between different morphologies. In mass culture, R0SE-C4 cells grew as a compact, cobblestone-like epithelium, while R0SE-B2 cells were fusiform (atypical) when sparse and similar to R0SE-C4 when confluent. Confluent cultures of R0SE-C4 remained contact-inhibited whether fed continuously or starved for up to 2 months. However, confluent R0SE-B2 cultures continued to proliferate when fed continuously forming multilayered ridges, but starving resulted i n c e l l death within 2 weeks. Human OSE in culture could potentially be valuable in the study of ovarian biology and carcinogenesis. Here, culture conditions have been defined that permit serial cultivation of human OSE for the f i r s t time in culture. This may allow the application of short-term i n vitro carcinogen assays that require cultures of either rapidly dividing or stationary c e l l s . The behaviour of ROSE 239 cells i n culture suggests that OSE cells are capable of exhibiting several different morphological forms and that modulation between forms may be possible. ROSE 239 cells may represent a useful model to examine further the apparent mulipotentiality of normal and neoplastic OSE in vivo. - 146 -BIBLIOGRAPHY 1. Adams, A.T. 1981. The Rat Ovarian Surface Epithelium In Vitro. Ph.D. Thesis, Department of Zoology, University of British Columbia. 2. Adams, A.T. and N. Auersperg. 1980. Transformation of rat ovarian surface epithelial (ROSE) cells by Kirsten murine sarcoma virus (Ki-MSV) in vi t r o . Proceedings of the American Association for Cancer Research 21: 92. 3. Adams, A.T. and N. Auersperg. 1981a. Transformation of cultured rat ovarian surface epithelium by Kirstin murine sarcoma virus. Cancer Research_41: 2063-2072. 4. Adams, A.T. and N. Auersperg. 1981b. Autoradiographic investigation of estrogen receptors i n cultured rat ovarian surface epithelial c e l l s . Journal of Cell Biology _9_1: 209a. 5. Adams, A.T. and N. Auersperg. 1983* Autoradiographic investigation of estrogen binding in cultured rat ovarian surface epithelial c e l l s . Journal of Histochemistry and Cytochemistry 31? 1321-1325. 6. Adams, A.T. and N. Auersperg. 1985. A c e l l line, ROSE 199, derived from normal rat ovarian surface epithelium. Experimental Cell Biology 53• 181-188. 7. Altman, P.L., and D.D. Katz, eds. 1976. Cell Biology. Federation of American Societies for Experimental Biology, Bethesda. pp. 367-368. 8. Ames, B.N., J. McCann, and E. Yamaski. 1975. Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome mutagenicity test. Mutation Research _3_1: 347-364. 9. Anderson, E., G. Lee, R. Letourneau, D.F. Albertini, and S.M. Mueller. 1976. Cytological observations of the ovarian epithelium in mammals during the reproductive cycle. Journal of Morphology 150: 135-164. 10. Andrews, P.M. and K.P. Porter. 1973« The ultrastructural morphology and possible functional significance of mesothelial m i c r o v i l l i . Anatomical Record 177: 409-426. 11. Aoyagi, T., H. Suya, K. Umeda, N. Kato, 0. Nemoto, H. Kobayashi, and Y. Miura. 1985* Epidermal growth factor stimulates tyrosine phosphor-ylation of pig epidermal fibrous keratin. The Journal of Investigative Dermatology 84: 118-121. 12. Auersperg, N., C.H. Siemens, and S.E. Myrdal. 1983. Culture and character-ization of human ovarian surface epithelial c e l l s . Journal of Cell Biology 97: 351a. - 147 -13. Auersperg, N.f CH. Siemens, and S.E. Myrdal. 1984a. Human ovarian surface epithelium i n primary culture. In Vitro 20: 743-755. 14. Auersperg N.f CH. Siemens, and S.E. Myrdal. 1984b. Growth patterns of human ovarian surface epithelium i n culture. Canadian Federation of Biological Societies 27: 49-15- B a i l l i e , A.H., M.M. Ferguson, and D. McHart. 1966. Developments In Steroid Histochemistry. Academic Press, New York. 16. Baker, J.B., G.S. Barsh, D.H. Carney, and D. Cunningham. 1978. Dexamethasone modulates binding and action of epidermal growth factor , i n serum-free medium. Proceedings of the National Academy of Science, U.S.A. 75.: 1882-1886. 17. Balboni, G.C 1983. Structural changes: Ovulation and luteal phase. In: Comprehensive Endocrinology: The Ovary. CB. Serra, ed. Raven Press, New York. pp. 123-142. 18. Ben-Ze'ev, A. 1983- Cell configuration-related control of vimentin biosynthesis and phosphorylation i n cultured mammalian c e l l s . The Journal of Cell Biology .97: 858-865-19. Ben-Ze'ev, A. 1985a. Cell shape, the complex cellular networks, and gene expression: Cytoskeletal protein genes as a model system. Cell and Muscle Motility 16.: 23-53. 20. Ben-Ze'ev, A. 1985b. C e l l - c e l l interaction and c e l l configuration related control of cytokeratins and vimentin expression i n epithelial cells and fibroblasts. _In: Intermediate Filaments: Annals of the New York Academy of Science. Volume 455. E. Wang, D. Fischman, R.K.H. Liem, and T.-T. Sun, eds. New York Academy of Science, New York. pp. 597-613-21. Blaustein, A. 1981. Surface cells and inclusion cysts i n f e t a l ovaries. Gynecologic Oncology .12: 222-233. 22. Blaustein, A., A. Kaganowicz, and J. Wells. 1982. Tumor markers i n inclusion cysts of the ovary. Cancer .49: 722-726. 23. Blaustein, A. and H. Lee. 1979- Surface cells of the ovary and pelvic peritoneum: A histochemical and ultrastructural comparison. Gynecologic Oncology 8_: 34-43-24- Bosland, M.C, T.J. Dirksen, and M.K. Prinsen. 1986. The role of c e l l proliferation in prostatic carcinogensis. Proceedings of the American Association for Cancer Research 27_: 92. 25- Brown, D.G., N.F. Johnson, and M.M.F. Wagner. 1985. Multipotential behaviour of cloned rat mesothelioma cells with epithelial phenotype. British Journal of Cancer 51} 245-252. 26. Buckley, A., J.M. Davidson, CD. Kamerath, T.B. Wolt, and S.C. Woodward. 1985. Sustained release of epidermal growth factor accelerates wound repair. Proceedings of the National Academy of Science, U.S.A. 82: 7340-7344. - 148 -27- Buick, R.N., R. Pullano, and J.M. Trent. 1985. Comparative properties of f i v e human ovarian adenocarcinoma c e l l l i n e s . Cancer Research. _45: 3668-3676. 28. Butterworth, B.E. 1985. Measurement of chemically-induced DNA r e p a i r i n rodent and human c e l l s . In: New Approaches i n T o x i c i t y Testing and Their A p p l i c a t i o n i n Human Risk Assessment. A.P. L i , ed. Raven Press, New York. pp. 33-40 29. Butterworth, B.E., L.L. Earle, S. Strom, R. J i r t l e , and G. Michalopoulos. 1983. Measurement of chemically induced DNA r e p a i r i n human hepatocytes. Proceedings of the American Ass o c i a t i o n f o r Cancer Research 2£: 69 • 30. Cajander, S. and L. Bjersing. 1977. Ovarian histochemistry. _In: The Ovary I: General Aspects. P.L. Zuckerman and B.J. Weir, eds. Academic Press, New York. pp. 303-391. 31. Casagrande, J.T., M.C. Pike, R.K. Ross, E.W. Louie, S. Roy, and B.E. Henderson. 1979• "Incessant ovulation" and ovarian cancer. The Lancet _2: 170-173-32. Chabot, J.-G., R. St-Arnaud, P. Walker, and G. P e l l e t i e r . 1986. D i s t r i b u t i o n of epidermal growth f a c t o r receptors i n the r a t ovary. Molecular and C e l l u l a r Endocrinology _44. 99-109-33« Champy, C. 1926. Sur l e s cultures d'epithelium germanatif i n v i t r o . Societe B i o l . Compt. Rend. _9_4: 1082-1084-34. Cherington, P.V. and A.B. Pardee. 1980. Sy n e r g i s t i c e f f e c t s of epidermal growth f a c t o r and thrombin on the growth stimulation of d i p l o i d chines hamster f i b r o b l a s t s . Journal of C e l l u l a r Physiology 103: 25-32. 35« C h r i s t i a n , CD. 1971. Ovarian tumors - an extension of the Peutz-Jeghers syndrome. American Journal of Obstetrics and Gynecology 111: 529-534-36. Connell, N.D., and J.G. Rheinwald. 1983- Regulation of the cytoskeleton i n mesothelial c e l l s : Reversible los s of k e r a t i n and increase i n vimentin during rapid growth i n cult u r e . C e l l _34: 245-253-37 Cooper, D-, A. Schermer, and T.-T. Sun. 1985. C l a s s i f i c t i o n of human e p i t h e l i a and t h e i r neoplasms using monoclonal antibodies to keratins: Strategies, Applications, and Limi t a t i o n s . Laboratory Investigation _52: 243-256. 38. Cotchin, E. 1977. Spontaneous tumours of the uterus and ovaries i n animals. Pathology of the Female Genital Tract. A. Blaustein, ed. Springer Verlag, New York. pp. 822-861. 39. Cotta-Pereira, G., H. Sage, P. Bornstein, R. Ross, and S. Schwartz. 1980. Studies of morphologically a t y p i c a l ("sprouting") cultures of bovine a o r t i c e n d o t h e l i a l c e l l s . Growth c h a r a c t e r i s t i c s and connective tissue p r o t e i n synthesis. Journal of C e l l u l a r Physiology 102: 183-191. - 1 4 9 -40. Cramer, D.W., W.E. Welch, E.S. Sc u l l y , and C A . Wojciechowski. 1982. Ovarian cancer and t a l c . Cancer 50; 372-376. 41. Cramer, D.W. and W.E. Welch. 1983* Determinants of ovarian cancer r i s k . I I . Inferences regarding pathogenesis. Journal of the National Cancer I n s t i t u t e 71: 717-721. 42 Cramer, D.W., C B . Hutchison, W.E. Welch, E.E. S c u l l y , and K.J. Eyan. 1983a. Determinants of ovarian cancer r i s k . I. Eeproductive and family h i s t o r y . Journal of the National Cancer I n s t i t u t e _71: 711-716. 43. Cramer, D.W., W.R. Welch, S. Ca s s e l l s , and E.E. S c u l l y . 1983b. Mumps, menarche, menopause, and ovarian cancer. American Journal of Obstetrics and Gynecology 147; 1-6. 44. C r i s t o f a l o , V.J. and B.A. Eosner. 1981. Glucoc o r t i c o i d modulation of c e l l p r o l i f e r a t i o n . In: Handbook of Experimental Pharmacology. Volume _57. E. Baserga, ed. Springer-Verlag, New York. pp. 208-227. 45. Czernobilsky, B., E. Moll, E. Levy, and W.W. Franke. 1985. Co-expression of cytokeratin and vimentin filaments i n mesothelial, granulosa and rete o v a r i i c e l l s of the human ovary. European Journal of C e l l Biology 37: 175-190. 46. Damanov, I. and D. So l t e r . 1976. Animal model: Embryo-derived teratomas and teratocarcinomas i n mice. American Journal of Pathology 83.: 241-244-47. Dekel, N., and I. Sh e r i z l y . 1985. Epidermal growth fa c t o r induces maturation of f o l l i c l e - e n c l o s e d oocytes. Endocrinology 116: 406-409. 48. Donaldson, E.E. 1976. The Ult r a s t r u c t u r e of the Eat Ovarian Germinal Epithelium and i t s P e r m i a b i l i t y to Electron Demonstrable Trace  Molecules. M.Sc. Thesis, Department of Anatomy, Univ e r s i t y of B r i t i s h Columbia. 49. Dozois, E.E., E.D. Kempers, D.C Dahlin, and L.G. Bartholomew. 1971. Ovarian tumours associated with Peutz-jeghers syndrome. Annals of Surgery 172: 233-290. 50. E g l i , G.E., and M. Newton. 1961. The transport of carbon p a r t i c l e s i n the human reproductive t r a c t . F e r t i l i t y and S t e r i l i t y 12: 151-155. 52. E i s i n g e r , M. and 0. Marko. 1982. Se l e c t i v e p r o l i f e r a t i o n of normal human melanocytes i n v i t r o i n the presence of phorbol ester and cholera toxin. Proceedings of the National Academy of Sciences, USA 7 9 : 2018-2022. 53. E i s i n g e r , M., 0. Marko, and I.B. Weinstein. 1983- Stimulation of growth of human melanocytes by tumor promoters. Carcinogenesis 4_: 779-781. 54. E l i s p u r u , E.K. 1981. A biochemical phage induction assay f o r carcinogens. In: Short-term Tests For Chemical Carcinogens. H.F. S t i c h and R.H.C. San, eds. Springer-Verlag, New York. pp. 1-11. - 150 -55« Erickson, G.F., and E. Case. 1983. Epidermal growth f a c t o r antagonizes ovarian t h e c a - i n t e r s t i t i a l c y t o d i f f e r e n t i a t i o n . Molecular and C e l l u l a r Endocrinology _3_1: 71-76. 56. Fano, G., and L. G a r o f o l i n i . 1927. Culture i n v i t r o of germinal epithelium from chick embryo. Journal of Physiology, Proceedings 63'. v i - v i i . 57. F a t h a l l a , M.F. 1972. Factors i n the causation and incidence of ovarian cancer. Obstetrics and Gynecologocal Surveys 27; 751. 58. F e n i g l i o , CM. 1980. Overview a r t i c l e : U l t r a s t r u c t u r a l features of the common e p i t h e l i a l tumors of the ovary. U l t r a s t r u c t u r a l Pathology _1: 419-444. 59- Ferenczy, A. and R.M. Richart. 1974. The Female Reproductive System: Dynamics of Scanning and Transmission Electron Microscopy. John Wiley and Sons, New York. 60. Folkman, J. and A. Moscona. 1978. Role of c e l l shape i n growth c o n t r o l . Nature 273: 345-349-61. Fox, H. and F.A. Langley. 1981. Malignant gonadal stromal tumours of ovary. In: Gynecologic Oncology: Fundamentals and C l i n i c a l P r a c t i c e . Volume 2. M. Coppleson, ed. C h u r c h i l l Livingston, New York. pp.697-709. 62. Fraumeni, J.F., J r . , G.W. Grundy, E.T. Creagan, and R.B. Everson. 1975. Six f a m i l i e s prone to ovarian cancer. Cancer _36_: 364-369. 63. Gallard, P.J. 1953. Growth and d i f f e r e n t i a t i o n of explanted t i s s u e s . International Review of Cytology 2: 331-401. 64. Garbi, C , L. Nistch, and S.H. Wollman. 1984. Embedding i n collagen g e l s t a b i l i z e s the p o l a r i t y of e p i t h e l i a l c e l l s i n thyroid f o l l i c l e s i n suspension. Experimental C e l l Research 151: 458-465-65- G i l l e t , J.Y., R. M a i l l e t , and C. Gautier. 1980. Blood supply and lymph supply of the ovary. In: Biology of the Ovary. P.M. Motta and E.S.E. Makabe, eds. Martinus N i j h o f f Publishers, Boston, pp. 86-98. 66. Gondos, B. 1975. Surface epithelium of the developing ovary. American Journal of Pathology 81: 303-320. 67. Gospodarowicz, D., A.L. Mescher, and CR. Bird w e l l . 1978. Control of c e l l u l a r p r o l i f e r a t i o n by the f i b r o b l a s t and epidermal growth f a c t o r s . National Cancer I n s t i t u t e Monograph 48: 109-130. 68. Gospodarowicz, D-, A.L. Mescher, CR. Birdwell. 1977. Stimulation of corneal endothelial c e l l p r o l i f e r a t i o n i n v i t r o by FGF and EGF. Experimental Eye Research _2_5_: 75-89. 69. Gospodarowicz, D., and H. B i a l e c k i . 1979. F i b r o b l a s t and epidermal growth factors are mitogenic agents f o r cultured granulosa c e l l s of rodent, porcine and human o r i g i n . Endocrinology 104s 757-764. - 151 -70. Graham, J. and R. Graham. 1967- Ovarian cancer and asbestos. Environmental Research _1: 115-128. 71. Greene, M.H., J.W. Clark, D.W. Blayney. 1984. The epidemiology of ovarian cancer. Seminars i n Oncology _11: 209-226. 72. Greenebaum, E., M. Nicolaides, M. Ei s i n g e r , R.H. Vogel, and I.B. Weinstein. 1983- Binding of phorbol dibutyrate and epidermal growth fac t o r to cultured human epidermal c e l l s . Journal of the National Cancer I n s t i t u t e J O : 435-441. 73. Greenwald, G.S. 1980. In-vivo and i n - v i t r o steroidogenesis by the human ovary during the menstrual c y c l e . In: Biology of the Ovary. P.M. Motta and E.S.E. Makabe, eds. Martinus N i j h o f f , Boston, pp. 244-253-74. Guraya, S.S. 1980. Histochemistry of the ovary. In: The Biology of the Ovary. P.M. Motta and E.S.E. Hafez, eds. Martimus N i j h o f f Publishers, Boston, pp. 33-51-75- Guraya, S.S. 1985. Biology of Ovarian F o l l i c l e s i n Mammals. Springer-Verlag, New York. 76. Ham, R.G. 1974. N u t r i t i o n a l requirements of primary c u l t u r e s : A neglected problem of modern biology. In V i t r o 10: 119-129-77. Ham, R.G. 1981. Su r v i v a l and growth requirements of nontransformed c e l l s . In: Handbook of Experimental Pharmacology. Volume 57- R- Baserga, ed. Springer-Verlag, New York. pp. 13-88. 78. Ham, R.G. 1984. S e l e c t i v e media. In: C e l l Separation: Methods and Selected Applications 3} 209-236. Academic Press, New York. T.G. Pretlow and T.P. Pretlow, eds. 79- Ham, R.G. and W.L. McKeehan. 1978. N u t r i t i o n a l requirements f o r c l o n a l growth of nontransformed c e l l s . In: N u t r i t i o n a l Requirements of Cultured C e l l s . H. Katsuta, ed. U n i v e r s i t y Park Press, Baltimore, pp. 63-111-80. Hamilton, T.C., P. Davies, and K. G r i f f i t h s . 1982. Estrogen r e c e p t o r - l i k e binding i n the surface germinal epithelium of the r a t ovary. Journal of Endocrinology 95: 377-385. 81. Hamilton, T.C., R.C. Young, and R.F. Ozols. 1984a. Experimental model systems of ovarian cancer: Applications to the design and evaluation of new treatment approaches. Seminars In Oncology 11: 285-298. 82. Hamilton, T.C., H. Fox, C H . Buckley, W.J. Henderson, and K. G r i f f i t h s . 1984b. E f f e c t s of t a l c on the rat ovary. B r i t i s h Journal of Experimental Pathology 65: 101-106. 83. Hamilton, T.C, W.J. Henderson, and C. Eaton. 1980. I s o l a t i o n and growth of the rat ovarian germinal epithelium. In: Tissue Culture i n Medical Research. I I . R.J. Richards and K.T. Rajan, eds. Pergamon Press, New York. pp. 237-244. - 152 -84. Hammond, S.L., R.G. Ham, and M.R. Stampfer. 1984. Serum-free growth of human mammary e p i t h e l i a l c e l l s : Rapid c l o n a l growth i n defined medium and extended s e r i a l passage with p i t u i t a r y e x t r a c t . Proceedings of the National Academy of Sciences, USA 81: 5435-5439. 85. Haroun, L. and B.N. Ames. 1981. The Salmonella mutagenicity t e s t : An overview. _In: Short-term Tests For Chemical Carcinogens. H.F. S t i c h and R.H.C. San, eds. Springer-Verlag, New York. pp. 108-119-86. Harrison, R.J. and B.J. Weir. 1977. Structure of the mammalian ovary. In: The Ovary. I. General Aspects. P.L. Zuckerman and B.J. Weir, eds. Academic Press, New York. pp. 113-217. 87. Hart, W.R. 1981. Pathology of malignant and borderline e p i t h e l i a l tumours of the ovary. jCn: Gynecologic Oncology: Fundamental P r i n c i p l e s and C l i n i c a l P r a c t i c e . Volume 2. M. Coppleson, ed. C h u r c h i l l Livingston, New York. pp.633-654. 88. Heap, R.B., and D.V. I l l i n g t o n . 1977. The mechanisms of a c t i o n of estrogens and progesterone. _In: The Ovary I I I : Regulation of Oogenesis and Steroidogenesis. S. Zuckerman and B.J. Weir, eds. Academic Press, New York. pp. 59-150. 89. Heddle, J.A., B.C. Lue, E.F. Saunders, and R.D. Benz. 1978. S e n s i t i v i t y to f i v e mutagens i n Fanconi's anemia as measured by the micronucleus method. Cancer Research 38: 2983-2988. 90. Heddle, J.A., and M.F. Salamone. 1981. The micronucleus t e s t : In vivo . In: Short-term Tests f o r Chemical Carcinogens. H.F. S t i c h and R.H.C. San, eds. Springer-Verlag, New York. pp. 243-249. 91. Heinonen, P.K., K. T o n t t i , T. Koivula, and P. Pystynen. 1985-Tumour-associated antigen CA 125 i n patients with ovarian cancer. B r i t i s h Journal of Obstetrics and Gynecology _9_2: 528-531. 92. Heintz, A.P., N.F. Hacker, and L.D. Lagasse. 1985. Epidemiology and etiology of ovarian cancer: A review. Obstetrics and Gynecology 66: 127-135. 93. Henderson, W.J., and K. G r i f f i t h s . 1975. I d e n t i f i c a t i o n of p a r t i c l e s i n ovarian t i s s u e . In: Gynaecologic Malignancy: C l i n i c a l and Experimental Studies. M.G. Brush and R.W. Taylor, eds. B a i l l i e r e T i n d a l l , London, pp. 227-240. 94. Henderson, W.J., T.C. Hamilton, and K. G r i f f i t h s . 1979- Talc i n normal and malignant ovarian t i s s u e . The Lancet 1_: 499« 95. Henderson, W.J., C.A.F. J o s l i n , and A.C. Turnbull. 1971. Talc and carcinoma of the ovary and cervix. The Journal of Obstetrics and Gynaecology of the B r i t i s h Commonwealth _78: 266-272. 96. Hildrick-Smith, G.Y. 1976. The biology of t a l c . B r i t i s h Journal of I n d u s t r i a l Medicine 33: 217-229-- 153 -97. Hseuh, A.J.W., T.H. Welsh, and P.B.C. Jones. 1981. I n h i b i t i o n of ovarian and t e s t i c u l a r steroidogensis by epidermal growth f a c t o r . Endocrinology 108; 2002-2004. 98. Hunter, T. and J.A. Cooper. 1985. Protein-tyrosine kinases. Annual Review of Biochemistry 54_: 897-930. 99. Jacobs, A.J., G.L. C u r t i s , J.R. Newland, R.B. Wilson, and W.L. Ryan. 1984. Chemical induction of ovarian e p i t h e l i a l carcinoma i n mice. Gynecologic Oncology _18: 177-180. 100. Janovski, N.A., and T.L. Paramandhan. 1973. Ovarian Tumours. W.B. Saunders, Phil a d e l p h i a . 101. Jarvholm, B., and B. Lavenius. 1981. A cohort study on cancer among workers exposed to an a n t i r u s t o i l . Scandanavian Journal of Work and Environmental Health 7: 179-184. 102. Joly, D.J., A.M. L i l i e f i e l d , E.L. Diamond, and I.D.J. Bross. 1974. An epidemiologic study of the r e l a t i o n s h i p of reproductive experience to cancer of the ovary. American Journal of Epidemiology 99: 190-209-103. Kadish. J.L., B u t t e r f i e l d , and J . Folkman. 1979. The e f f e c t of f i b r i n on cultured vascular endothelial c e l l s . Tissue and C e l l Research 11: 99-108. 104. Kato, T., M. Ya k u s h i j i , A. Tsunawaki, and K. Ide. 1973« A study of experimental ovarian tumors i n rats by the chemical carcinogen, 20-methylcholanthrene. Kurume Medical Journal_20: 159-167. 105. Kaufman, D.G., J.M. S i e g f r i e d , B.H. Dorman, K.G. Nelson, and L.A. Walton. 1983. Carcinogen-induced changes i n cultured human endometrial c e l l s . In: Human Carcinogensis. C C . Harris and H.N. Autrup, eds. Academic Press, New York. pp. 469-508. 106. Kerner, H., E. Gaton, and B. Czernobilsky. 1981. Unusual ovarian, tubal and p e l v i c mesothelial i n c l u s i o n s i n patients with endometriosis. Histopathology 5: 277-283. 107. Kidwell, W.R., D.S. Salomon, L.A. L i o t t a , J.A. Zweibel, and M. Bano. 1982. E f f e c t s of growth factors on mammary e p i t h e l i a l c e l l p r o l i f e r a t i o n and basement membrane synthesis. In: Growth of C e l l s i n Hormonally Defined Media. Cold Spring Harbor Conference 9_: 807-818. 108. Kimbrough, R.A. 1929. Coincident carcinoma of ovary i n twins. American Journal of Obstetrics and Gynecology 18: 148-149-109- Kirk, D-, S. Kagawa, S.K. Narayn, and Y. Ohnuki. 1985- Novel antigen system f o r tracking e p i t h e l i a l c e l l migration i n serum-free c u l t u r e . Experimental C e l l Research 160: 221-229. 110. Kleinman, D., Y. Sharon, I. Sarov, and V. I n s l e r . 1983- Human endometrium i n c e l l c ulture: A new method f o r c u l t u r i n g human endometrium as separate e p i t h e l i a l and stromal components. Archives of Gynecology 234: 103-112. - 154 -111. Klymkowsky, M.W. 1982. Vimentin and k e r a t i n intermediate filament systems i n cultured PtK 2 e p i t h e l i a l c e l l s are i n t e r r e l a t e d . The EMBO Journal 1_: 161-165. 112. Knecht, M. and K.J. Catt. 1983- Modulation of cAMP-mediated d i f f e r e n t i a t i o n i n ovarian granulosa c e l l by epidermal growth f a c t o r and' p l a t e l e t - d e r i v e d growth f a c t o r . Journal of B i o l o g i c a l Chemistry 258: 2789-2794. 113. K r y s t a l , G. 1983. A simple microassay forerythropoetin based on ^H-thymidine incorporation into spleen c e l l s from phenylhydrazine treated mice. Experimental Hematology 11: 649-660. 114. Kulesz-Martin, M.F., B. Koeler, H. Hennings, and S.H. Yuspa. 1980. Quantitative assay f o r carcinogen a l t e r e d d i f f e r e n t i a t i o n i n mouse epidermal c e l l s . Carcinogenesis _1: 995-1006. 115. Kulkarni, M.S. and D.G. Kaufman. 1986. Species di f f e r e n c e s i n the formation of benzo(a)pyrene DNA adducts i n rodent and human endometrium i n organ c u l t u r e . American Association f o r Cancer Research, Proceedings _27: 83* 116. Labarca, C. and K. Paigen. 1980. A simple, rapid, and s e n s i t i v e DNA assay procedure. A n a l y t i c a l Biochemistry 102: 344-352. 117. LaRocca, P.J. and J.G. Rheinwald. 1985- Anchorage-independent growth of normal human mesothelial c e l l s : A s e n s i t i v e bioassay f o r EGF which d i s c l o s e s the absence of t h i s f a c t o r i n f e t a l c a l f serum. In V i t r o C e l l u l a r and Developmental Biology 21: 67-72. 118. Lechner, J.F., A. Haugen, H. Autrup, I.A. McClendon, B.F. Trump, C.C. H a r r i s . 1981. Clonal growth of e p i t h e l i a l c e l l s from normal adult human bronchus. Cancer Research _41: 2294-2304. 119. Lechner, J.F., A. Haugen, I.A. McClendon, and E.W. P e t t i s . 1982. Clonal growth of normal adult human bronchial e p i t h e l i a l c e l l s i n a serum-free medium. In V i t r o 18: 633-642. 120. Lechner, J.F., T. Tokiwa, M. LaVeck, W. F. Benedict, S.B. Banks-Schlegel, H. Yeager, A. Banerjee, and C.C. H a r r i s . 1985* Asbestos-associated chromosomal changes i n human mesothelial c e l l s . Proceedings of the National Academy of Sciences 82_: 3884-3888. 121. Liber, A.F. 1950. Ovarian cancer i n mother and f i v e daughters. Archives of Pathology 49_: 280-290. 122. Lingeman, C.H. 1974- Etio l o g y of cancer of the human ovary: A review. Journal of the National Cancer I n s t i t u t e 53_: 1673-1618. 123. Lingeman, C.H. 1983- Environmental factors i n the etio l o g y of carcinoma of the human ovary: A review. American Journal of I n d u s t r i a l Medicine l i 365-379. 124. Long, H.J. 1940. Growth i n v i t r o of ovarian germinal epithelium. In: Contributions to Embryology 28: 91-96. - 155 -125. Longo, D.L., and R.C. Young. 1979. Cosmetic t a l c and ovarian cancer. The Lancet 2: 349-351. 126. Lynch, H.T., V. Albano, L. Black, J.F. Lynch, J . Recabaren, and R. Pierson. 1981. F a m i l i a l excess of cancer of the ovary and other anatomic s i t e s . Journal of the American Medical Association 245: 261-264. 127. Maciag, T., G.A. Hoover, M.B. Stemerman, and R. Weinstein. 1981. S e r i a l propagation of human endothelial c e l l s i n v i t r o . The Journal of C e l l Biology 91: 420-426. 128. Madigan, S. and V. Lawrence. 1983. Stats Plus: A General S t a t i s t i c s Package For the Apple I I . Human Systems Dynamics, Northridge, C a l i f . 129. Marceau, N., H. Baribault, and I. Leroux-Nicollet. 1984* Dexamethasone can modulate the synthesis and organization of cytokeratins i n cultured d i f f e r e n t i a t i n g rat hepatocytes. Canadian Journal of Biochemistry and C e l l Biology 63j 448-457-130. Marceau, N. and S.H.H. Swierenga. 1985. Cytoskeletal events during calcium- or EGF-induced i n i t i a t i o n of DNA synthesis i n cultured c e l l s : Role of protein phosphorylation and clues i n the transformation process. C e l l and Muscle M o t i l i t y 6_: 97-140. 131. Martinovich, P.N. 1937. Development i n v i t r o of the mammalian gonads. Nature 139: 413. 132. Mattison, D.R., and Thorgeirrsson. 1978. Smoking and i n d u s t r i a l p o l l u t i o n , and t h e i r e f f e c t s on menopause and ovarian cancer. The Lancet _1: 187-188. 133. McAuslan, B.R., G.N. Hannan, and W. R e i l l y . 1982. Signals causing change i n morphological phenotype, growth mode, and gene expression of vascular endothelial c e l l s . Journal of C e l l u l a r Physiology 112: 96-106. 134. McCormick, J . J . and V.M. Maher. 1985- Use of human c e l l s i n mutagenicity and carcinogenicity determination. _In: New Approaches i n T o x i c i t y Testing and Their A p p l i c a t i o n In Human Risk Assessment. A.P. L i , ed. Raven Press, New York. pp. 17-32. 135. McGowan, L.M.. 1978. Ovarian cancer. In: Gynecologic Oncology. L.M. McGowan, ed. Appleton-Century-Crofts, New York. pp. 283-331. 136. McKeehan, V.L. and R.G. Ham. 1977. Methods f o r reducing the serum requirement f o r growth i n v i t r o of nontrnasformed d i p l o i d f i b r o b l a s t s . J o i n t WHO/IABS symposium on the standardization of c e l l substrates f o r the production of vi r u s vaccines. Development of B i o l o g i c a l Standards 37. F.T. Perkins, and R.H. Regamy, eds. S. Karger, Basel, pp. 97-108. 137. McKeehan, W.L., K.A. McKeehan, S.L. Hammond, and R.C. Ham. 1977. Improved medium f o r c l o n a l growth of human d i p l o i d f i b r o b l a s t s at low concentrations of serum pro t e i n . In V i t r o 13: 399-416. - 156 -138. M i l l e r , R.W. 1972. Radiation induced cancer. Journal of the National Cancer I n s t i t u t e 49_: 1221-1227. 139. Mondshein, J.S. and D.W. Schomberg. 1981. Growth factors modulate gonadotropin receptor induction i n granulosa c e l l c u l t u r e s . Science 211: 1179-1180. 140. Moore, K.L. 1977. The Developing Human. W.B. Saunders, Toronto. 141. Moore, C.J., W.A. Tricomi, S.R. Eldridge, K. Kamiya, and M.N. Gould. 1986. Q u a l i t a t i v e s p e c i e s - s p e c i f i c differences DMBA and B(A)P a c t i v a t i o n i n human and r a t mammary e p i t h e l i a l c e l l s . Proceedings of American Association f o r Cancer Research _27: 88. 142. Morrow, C P . 1981. Malignant and borderline e p i t h e l i a l tumours of ovary: C l i n i c a l features, staging, diagnosis, intraoperative assessment and review of management. In: Gynecologic Oncology: Fundamental P r i n c i p l e s and C l i n i c a l P r a c t i c e . Volume 2. M. Coppleson, ed. C h u r c h i l l Livingston, New York, pp.-655-686. 143. Mostafa, S.A.M., C B . Bargeron, R.W. Flower, N.B. Rosenshein, T.H. Parmley, and J.D. Woodruff. 1985. Foreign body granulomas i n normal ovaries. Obsetrics and Gynecology 66: 701-702. 144. Motta, P.M., J . Van Blerkom, and S. Makabe. 1980. Changes i n the surface morphology of the ovarian germinal epithelium during the reproductive cycle and i n some pathological conditions. Journal of Submicroscopical Cytology JL2: 407-425-145. Motta, P.M. and J . Van Blerkom. 1980. Scanning electron microscopy of the mammalian ovary. In: Biology of the Ovary. P.M. Motta and E.S.E. Hafez, eds. Martinu N i j h o f f Publishers, Boston, pp. 162-175-146. Motta, P.M. and S. Makabe. 1982. Development of the ovarian surface and associated germ c e l l s i n the human fetus. C e l l and Tissue Research 226: 493-510. 147- Murphy, E.D. 1980. Major experimental models: Histogenesis and e v a l -uation. In: Biology of Ovarian Neoplasia. I I . E.D. Murphy and W.C Beamer, eds. UICC Technical Report Series, Volume 50. pp. 66-73-148. Myrdal, S.E. and N. Auersperg. 1986. An agent or agents produced by virus-transformed c e l l s cause unregulated r u f f l i n g i n untransformed c e l l s . The Journal of C e l l Biology 102: 1224-1229-149- Newhouse, M.L., G. Berry, J . C Wagner, and M.E. Turok. 1972. A study of the m o r tality of female asbestos workers. B r i t i s h Journal of I n d u s t r i a l Medicine 29: 134-141. 150. Nicosi a , S.V. 1983. Morphological changes or the human ovary throughout l i f e . In: The Ovary. C B . Serra, ed. Raven Press, New York. pp. 57-81. - 157 -151. Nico s i a , S.V. and J.H. Johnson. 1984a. Surface morphology of ovarian mesothelium (surface epithelium) and of other p e l v i c and extrapelvic mesothelial s i t e s i n the r a b b i t . International Journal of Gynecological Pathology 2= 249-260. 152. Nicosia, S.V. and J.H. Johnson. 1984b. E f f e c t s of two i n t e r n a t i o n a l union against cancer standard reference samples on cultured r a b b i t ovarian surface e p i t h e l i a l c e l l s . Scanning E l e c t r o n Microscopy 3} 1329-1336. 153- Nicosia S.V., J.H. Johnson, and E.J. S t r e i b e l . 1984. I s o l a t i o n and ch a r a c t e r i z a t i o n of rabbit ovarian mesothelium (surface epithelium). International Journal of Gynecological Pathology 3_: 348-360. 154. Nicosia, S.V., J.H. Johnson, E.J. S t r e i b e l . 1985. Growth c h a r a c t e r i s t i c s of r a b b i t ovarian mesothelial (surface e p i t h e l i a l ) c e l l s . International Journal of Gynecological Pathology _4: 58-74. 155. Nomura, T. 1973« Carcinogensis by urethane v i a mother's milk and i t s enhancement of transplacental carcinogenesis i n mice. Cancer Research 33_: 1677-1683. 156. O'Guin, W.M., A. Schermer,and T.-T. Sun. 1985. Immumofluorescence s t a i n i n g of k e r a t i n filaments i n cultured e p i t h e l i a l c e l l s . Journal of Tissue Culture Methods .9: 123-128. 157. Oka, Y., and D.N. Orth. 1983. Human plasma epidermal growth factor/B—urogastrone i s associated with blood p l a t e l e t s . Journal of C l i n i c a l I n v e s t i g a t i o n 72: 249-259. 158. Ormerod, E.J. and P. Rudland. 1985. I s o l a t i o n and d i f f e r e n t i a t i o n of cloned e p i t h e l i a l c e l l l i n e s from normal r a t mammary glands. In V i t r o C e l l u l a r and Developmental Biology _21: 143-153. 159. Otto, A.M. 1982. Microtubules and the regulation of DNA synthesis i n f i b r o b l a s t i c c e l l s . C e l l Biology International Reports _6: 1-18. 160. Painter, R.A. 1981. DNA synthesis i n h i b i t i o n i n mammalian c e l l s as a tes t f o r mutagenic carcinogens. _In: Short-term Tests For Chemical Carcinogens. H.F. S t i c h and R.H.C. San, eds. Springer-Verlag, New York, pp. 59-64. 161. P a o l e t t i , L., S. Caiazza, G. D o n e l l i , and F. P o c c h i a r i . 1984- Evaluation by e l e c t r o n microscopy techniques of asbestos contamination i n i n d u s t r i a l , cosmetic, and pharmaceutical t a l c s . Regulatory Toxicology and Pharmacology 4_: 222-235. 162. Parmley, T.H., and J.D. Woodruff. 1974. The ovarian mesothelioma. American Journal of Obstetrics and Gynecology 120: 234-241. 163. Papadaki, L., and J.O.W. Beilby. 1971. The fi n e structure of the surface epithelium of the ovary. Journal of C e l l Science 8_: 445-465. - 158 -164. P a t i l , G., F.J. Burns, A. Sellakumar, and C A . Snyder. 1986. C e l l p r o l i f e r a t i o n k i n e t i c s i n rat nasal epithelium a f t e r i n h a l a t i o n of d i r e c t - a c t i n g a l k y l a t i n g carcinogens. American Ass o c i a t i o n f o r Cancer Research, Proceedings _27: 74. 165. Peehl, D.M. 1985. An improved culture system f o r the culture of adult human p r o s t a t i c e p i t h e l i a l c e l l s . In V i t r o £1 (Part I I ) : 65a. 166. Peehl, D.M., and R.G. Ham. 1980a. Growth and d i f f e r e n t i a t i o n of human keratinocytes without a feeder layer or conditioned medium. In V i t r o 16: 516-525-167. Peehl, D.M., and R.G. Ham. 1980b. Clonal growth of human keratinocytes with small amounts of dialyzed serum. In V i t r o 16: 526-538. 168. Ponec, M. 1984* E f f e c t s of glucocorticoids on cultured s k i n f i b r o b l a s t s and keratinocytes. International Journal of Dermatology 23: 11-24. 169. Pruess-Schwartz, D., W.M. Baird, A. Nikbakht, B.A. Merrick, and J.K. S e l k i r k . 1986. Benzo(a)pyrene-DNA assuct formation i n normal human mammary e p i t h e l i a l c e l l cultures and the human mammary carcinoma T 47D c e l l l i n e . American Association f o r Cancer Research, Proceedings 27: 87. 170. R a d i s a v l j e v i c , S.V. 1977. The pathogenesis of ovarian cysts and cystomas. Obstetrics and Gynecology 49: 424-429-171. Rheinwald, J.G., and H. Green. 1975. S e r i a l c u l t i v a t i o n of human epidermal keratinocytes: The formation of k e r a t i n i z i n g colonies from s i n g l e c e l l s . C e l l _6: 331-344. 172. Rheinwald J.G., and H. Green. 1977. Epidermal growth f a c t o r and the m u l t i p l i c a t i o n of human epidermal keratinocytes. Nature 265: 421-424. 173- Richardson, G.S., R.E. Sc u l l y , N. Najamosama, and J.H. Nelson. 1985a. Common e p i t h e l i a l cancer of the ovary (Part I ) . The New England Journal of Medicine 312: 415- 424. 174. Richardson, G.S., R.E. Sc u l l y , N. Najamosame, and J.H. Nelson. 1985b. Common e p i t h e l i a l cancer of the ovary (Part I I ) . The New England Journal of Medicine 312: 474-483. 175- Risch, H.A., N.S. Weiss, J.L. Lyon, J.R. Daling, and J.M. L i f f . 1983. Events of reproductive l i f e and the incidence of e p i t h e l i a l ovarian cancer. American Journal of Epidemiology 117: 128-139-176. Rodrigues, J . and F. Dienhardt. I960. Preparation of a semipermanent mounting medium f o r fluorescent antibody studies. Virology_12: 316-317-177- Rohl, A.N., A.M. Langer, I . J . S e l i k o f f , A. T o r d i n i , and R. Kli m e n t i d i s . 1976. Consumer talcums and powders: Mineral and chemical c h a r a c t e r i z a t i o n . Journal of Toxicolgy and Environmental Health _2: 255-284-- 159 -178. Romijn, J.C., C F . Verkoelen, and F.H. Schroeder. 1984. Determination o f the growth rate of human p r o s t a t i c c e l l s i n primary culture by a morphometric technique. C e l l Biology International Reports 8_: 363-371• 179* Sadiq, H.F., and U.P. Devaskar. 1984. Glucocoticoids increase pulmonary epidermal growth f a c t o r receptors i n female and male r a b b i t . Biochemical and Biophysical Research Communications 119: 408-414. 180. Schmid, W. 1975* The micronucleus t e s t . Mutation Research 31} 9-15. 181. S c u l l y , R.E. 1977. Ovarian Tumors: A Review. American Journal of Pathology 87: 686-720. 182. Sekiya, S., N. Endoh, Y. Kikuchi, T. Katoh, A. Matsuura, H. Iwasawa, B. Takeda, and H. Takamizama. 1979* In vivo and i n v i t r o studies of experimental ovarian adenocarcinoma i n r a t s . Cancer Research 39: 1108-1112. 183. S i d d i q u i , K.M. and D.P. Chopra. 1986. Unscheduled DNA synthesis i n colon e p i t h e l i a l c e l l s exposed i n v i t r o to carcinogens or b i l e a c i d s . American Association f o r Cancer Research, Proceedings _27: 102. 184. S i e g f r i e d , J.M., and S. Nesnow. 1984. C y t o t o x i c i t y of chemical carcinogens towards human bronchial e p i t h e l i a l c e l l s evaluated i n a cl o n a l assay. Carcinogensis 5} 1317-1322. 185. Siemens, C H . and N. Auersperg. 1986. Cultured human ovarian surface epithelium (OSE): A p o t e n t i a l carcinogen assay system. Proceedings of the Ammerican Association f o r Cancer Research _27: 27. 186. Sing l e t a r y , S.E., B. Tomasovic, C Sp i t s e r , S.L. tucker, V. Hug, and B. Drewinko. 1985. E f f e c t s and i n t e r a c t i o n s of epidermal growth f a c t o r , i n s u l i n , hydrocortisone, and e s t r a d i o l on the cloning of human tumor c e l l s . International Journal of C e l l Cloning 3} 407-414. ^ 187. Stampfer, M., R.C. Hallowes, and A.J. Hackett. 1980. Growth of normal human mammary c e l l s i n cu l t u r e . In V i t r o 16: 415-425. 188. S t i c h , H.F., and R.H.C San. 1970. DNA r e p a i r and chromatid anomalies i n mammalian c e l l s exposed to 4-nitroquinoline 1-oxide. Mutation Research 10: 389-404. 189. S t i c h , H.F., R.H.C. San, and H.J. Freeman. 1981. DNA Repair Synthesis (UDS) as an i n v i t r o and i n vivo bioassay to detect prcarcinogens, ultimate carcinogens, and organotropic carcinogens. In: Short-term Tests f o r Chemical Carcinogens. H.F. S t i c h and R.H.C. San, eds. Springer-Verlag, New York. pp. 65-82. 190. Stoner, G.D., C C . Harris, C A . Myers, B.F. Trump, and R.D. Connor. 1980. Putrescine stimulates growth of human bronchial e p i t h e l i a l c e l l s i n primary cu l t u r e . In V i t r o 16: 399-406. - 160 -191. Symanski, J.A. and D.W. Schomberg. 1980. A novel r o l e f o r growth f a c t o r s i n granulosa c e l l c u l t u r e : I n h i b i t i o n of FSH-dependent LH receptor induction. Endocrine Society 62nd Meeting, Abstract 607. 192. Taylor-Papadimitriou, J., M. Shearer, and M.G.P. Stoker. 1977. Growth requirements f o r human mammary e p i t h e l i a l c e l l s i n c u l t u r e . I n t e r n a t i o n a l Journal of Cancer _20: 903-908. 193. Thornton, S.C, S.N. Mueller, and E.M. Levine. 1983* Human e n d o t h e l i a l c e l l s : Use of heparin i n cloning and long-term s e r i a l c u l t i v a t i o n . Science 222: 623-625. 194. Trichanpoulos, D., M. Papapostolou, and A. Polychronopoulou. 1981. Coffee and ovarian cancer. International Journal of Cancer 28: 691-693. 195. Tsau, M.C, B.J. Walthall, and R.G. Ham. 1982. Clonal growth of normal human keratinocytes i n a defined medium. Journal of C e l l u l a r Physiology 110: 219-229. 196. Tunca, J . C , E. Erturk, E. Erturk, and G.T. Bryan. 1985- Chemical induction of ovarian tumors i n r a t s . Gynecologic Oncology 21: 54-64-197- Van Nagell, J.R. 1983- Tumor markers i n gynecologic malignancies. In: Gynecologic Oncology. C.T. G r i f f i t h s and A.P. F u l l e r , eds. Martimus N i j h o f f Publishers, Boston, pp. 63-79. 198. Van Blerkom, J . and P.M. Motta. 1979- In: The C e l l u l a r Basis of Mammalian Reproduction. Urban and Scharzenberg, Baltimore. 199. Venter, P.F. and M. I t u r r a l d e . Migration of a p a r t i c u l a t e r a d i o a c t i v e t r a c e r from the vagina to the p e r i t o n e a l c a v i t i e s and ovaries. South A f r i c a n Journal of Medicine 55: 917-919-200. Waymouth, C. 1959- Rapid p r o l i f e r a t i o n of sublines of NCTC Clone 929 ( S t r a i n L) mouse c e l l s i n a simple chemically defined medium (MB 752/1). Journal of the National Cancer Institute_22: 1003-1017-201. Weinstein, I.B., M. Wigler, and U. Stadler. 1976. Analysis of the mechanism of chemical carcinogensis i n e p i t h e l i a l c e l l c u l t u r e s . In: Screening Tests i n Chemical Carcinogenesis. R. Montesano, H. Bartsch, and L. Tomatis, eds. International Agency For Cancer Research, Lyon, pp. 355-381. 202. Wiley, H.S. and D.D. Cunningham. 1982. Epidermal growth f a c t o r stimulates f l u i d phase endocytosis i n human f i b r o b l a s t s through a s i g n a l generated at the c e l l surface. Journal of C e l l u l a r Biochemistry _19: 383-394. 203. Willey, J . C , A.J. Saladino, C. Ozanne, J.F. Lechner, and C.C H a r r i s . 1984. Acute e f f e c t s of 12-0-tetradecanoylphorbol-13-acetate, t e l e c i d i n B or 2,3,7,8,-tetrachlorodibenzo-p-dioxin on cultured normal human bronchial e p i t h e l i a l c e l l s . Carcinogenesis 5_: 209-215. 204. Williams, CM. 1976. Further improvements i n the hepatocyte primary cu l t u r e DNA r e p a i r t e s t f o r carcinogens. Cancer L e t t e r s _4: 69-75. - 161 -205- Williams, G.M. 1978. The use of l i v e r e p i t h e l i a l cultures f o r the study of chemical carcinogensis. American Journal of Pathology 85: 739-753. 206. Williams, G.M. 1981. L i v e r culture i n d i c a t o r s f o r detection of chemical carcinogens. In; Short-term t e s t s f o r chemical carcinogens. H.F. S t i c h and R.H.C. San, eds. Springer-Verlag, New York. 277-289. 207. Williams, G.M. 1985. I d e n t i f i c a t i o n of genotoxic and epigenetic carcinogens i n l i v e r culture systems. Regulatory Toxicology and Pharmacology 5_: 132-144. 208. Wolff, S. 1981. The s i s t e r chromatid exchange t e s t . _In: Short-term Tests f o r Chemical Carcinogens. H.F. S t i c h and R.H.C. San, eds. Springer-Verlag, New York. pp. 236-242 209. Wolff, E. and K. Haffen. 1952. Sur l ' i n t e r s e x u a l i t e experimental des gonades embryonaires due canard c u l t i v e s i n v i t r o . Archives d'Anatomie Microsopique et de Morphologie Experimentale (Paris) _4_1: 185-207. 210. Wolff, S. and P. Perry. 1974. D i f f e r e n t i a l Geimsa s t a i n i n g of s i s t e r chromatids and the study of s i s t e r chromatid exchanges without autoradiography. Chromosoma 48: 341-353. 211. Wolffe, A.P., R.I. Bersimbaev, and J.R. Tata. 1985. I n h i b i t i o n by e s t r a d i o l of binding and mitogenic e f f e c t of epidermal growth f a c t o r i n primary cultures of Xenopus hepatocytes. Molecular and C e l l u l a r Endocrinology 40: 167-173-212. Woodruff, J.D. 1975- Experimental production of ovarian neoplasia. In: Diagnosis and Treatment of Ovarian Neoplastic A l t e r a t i o n s . H. de Wa t t e v i l l e and P.R.J. Burch, eds. Excerpta Medica, Amsterdam, pp. 53-55. 213. Woodruff, J.D. 1976. History of ovarian neoplasia: Facts or fancy. Obstetrics and Gynecology Annual _5_: 331-344. 214. Woodruff, J.D. 1979. The pathogensis of ovarian neoplasia. Johns Hopkins Medical Journal 144: 117-123. 215. Woods, L.K., R.T. Morgan, L.A. Quinn, G.E. Moore, T.U. Semple, and K.E. Stedman. 1979. Comparison of four new c e l l l i n e s from patients with adenocarcinoma of the ovary. Cancer Research _39: 4449-4459. 216. Wu, Y.-J, L.M. Parker, N.E. Binder, M.A. Beckett, J.H. Sinard, C.T. G r i f f i t h s , and J.G. Rheinwald. 1982. The mesothelial keratins: A new family of c y t o s k e l e t a l proteins i d e n t i f i e d i n cultured mesothelial c e l l s and nonkeratinizing e p i t h e l i a . C e l l 31: 693-703. - 162 -APPENDIX I - SOURCES OF MATERIALS (1) TISSUE CULTURE Way-mouths' 752/1 Medium - Sigma Chemical Co., St. Louis, MO. Medium 199 ( E a r l e s ' s a l t s ) - Sigma Chemical Co., St. Louis, MO. MCDB 202 Medium (prepared as i n McKeehan and Ham, 1977) - Terry Fox Laboratory, B.C. Cancer Research Center, Vancouver, B.C. Hanks' Balanced S a l t Solution ( Ca + +/Mg + +-free) - Gibco Laboratories, Grand Island, N.Y. F e t a l Bovine Serum (Hyclone l o t #-110451) - Hyclone Laboratories Inc., Logan, Utah. F e t a l C a l f Serum - Gibco Laboratories, Grand Island, NY. Trypsin (250:1) - Gibco Laboratories, Grand Island, NY. EGTA [e t h y l e n e b i s ( o x y e t h l e n e n i t r i l o ) J t e t r a a c e t i c a c i d - J.T. Baker Chemical Co., P h i l l i p s b u r g , NJ. P e n i c i l l i n / S t r e p t o m y c i n - Gibco Laboratories, Grand Island, NY. Epidermal Growth Factor (EGF) - Collaborative Research, Lexington, MA. Hydrocortisone - Sigma Chemical Co., St. Louis, MO. 176-Estradiol [ l , 3 , 5 , ( l 0 ) - e s t r a t r i e n - 3 , 173-diol] - S t e r a l o i d s Inc., Wilton, NH. Dextran - Sigma Chemical Co, St. Louis, MO. N o r i t activated charcoal - F i s h e r S c i e n t i f i c Co., Vancouver, B.C. P l a s t i c culture vessels - Corning Glass Works, Cornig, NY. . P l a s t i c culture wells - Nunc, Denmark. Polystyrene c o v e r s l i p s - Lux S c i e n t i f i c Corporation, Thousand Oaks, C a l i f . Thermanox coverslips - Lux S c i e n t i f i c Corporation, Thousand Oaks, CA. Glass c o v e r s l i p s (22 X 22 mm) - Canlab, Vancouver, B.C. Artek Model 980 C e l l Counter - Artek Systems Corp., Farmingdale, NY. Urisystem disposable s l i d e s - F i s h e r S c i e n t i f i c Ltd., Vancouver, B.C. Gelman Acrodisc low protein binding disposable f i l t e r s (0.20 um) -Gelman, Ann Arbor, MI. (2) Immunofluorescence Rabbit a n t i - k e r a t i n polyclonal antibody - g i f t from T.-T. Sun. Goat a n t i - r a b b i t IgG (heavy and l i g h t chain s p e c i f i c ) , rhodamine-conjugated - C a p p e l - S c i e n t i f i c , D i v i s i o n of Cooper Biomedical, Malvern, PA. Gelvatol - Monsanto, S p r i n g f i e l d , MA. (3) Hoechst DNA Assay TRTPCK t r y p s i n (lyophilyzed) - Cooper Biomedical, Malvern, PA. Hoechst (bisbenzimidazol) - Sigma Chemical Co., St. Louis, MO. Calf-thymus DNA - Sigma Chemical Co., St. Louis, MO - 163 -APPENDIX I - SOURCES OF MATERIALS (CON'T) (4) Tritiated-thymidine Incorporation Glass micro-fiber f i l t e r s - Whatman, C l i f t o n , NJ. T r i t o n X-lOO - Sigma Chemical Co., St. Louis, MO. Omnifluor S c i n t i l l a t i o n F l u i d - New England Nuclear, Boston, MA. TRTPCK-trypsin - as above. (5) Photography Kodak 2415 Fan Technical black and white f i l m Kodak Tri-X Pan black and white (ASA 400) Kodak Ektachrome color s l i d e f i l m (ASA 400, daylight) Kodak HC110 developer Kodak Microdol developer A l l of above from Eastman Kodak Co., Rochester, NY. - 1 6 4 -APPENDIX I I - IMMUNOPLUORESCENCE STAINING OF KERATIN FILAMENTS (Afte r O'Guin et a l . , 1985) Solutions a. Phosphate-buffered s a l i n e (PBS) To 750 ml d i s t i l l e d H 20 (dH 20) add, i n order, 8.0 g NaCl; 2 g KC1; 1.15 g Na 2HP0 4; 0.2 g KH 2P0 4; 0.2 g KH 2P0 4. Add s u f f i c i e n t volume dH 20 to make 1.0 l i t e r . pH 7«3« b. Gelvatol (Rodrigues and Dienhardt, I960). Add 20 g Gelvatol to 80 ml PBS. S t i r overnight. Add 40 ml g l y c e r o l , and s t i r overnight. Decant supernatant, check pH (7.8), and store i n an a i r t i g h t b o t t l e at 4°C. Preparation of c u l t u r e s . a. Plate c e l l s on s t e r i l e , 22 X 22 mm glass c o v e r s l i p s i n 35 mm culture dishes. P r i o r to use, c o v e r s l i p s should be washed f o r 15 min. i n hot dH 20 containing 7X-detergent, rinsed f o r 30 min. i n running tap H 20, rinsed 3X i n dH 20, and autoclaved, separated by gauze, i n a glass p e t r i d i s h . b. When the cultures have grown to the desired density, remove the culture medium, and immerse s l i p d i r e c t l y i n t o -20°C methanol-acetone ( l : l ) i n a Columbia s t a i n i n g j a r ; should be prepared i n advance and stored at -20°C) f o r 5 min. to f i x and permeablize the c u l t u r e s . Allow cover s l i p s to a i r - d r y (usually 2 to 3 min.) S l i p s can be stored f o r several weeks i f necessary i n -20°C methanol, then transferred to methanol-acetone at the time of s t a i n i n g . c. Rehydrate the specimens i n PBS f o r 5 min. In d i r e c t immunofluorescence s t a i n i n g . a. Remove the c o v e r s l i p s from PBS and drain by touching the edge of the c o v e r s l i p to b l o t t i n g paper. Do not allow specimens to dry out at any time. b. Apply approximately 50 ul of a properly d i l u t e d s o l u t i o n of a n t i k e r a t i n antibody to the surface of each c o v e r s l i p . Place c o v e r l i p s i n a container containing moistened foam rubber or f i l t e r paper to provide humidity. c. Close humidified chamber and incubate at 37°C f o r 30 min. d. Rinse well with several changes of PBS and store i n Columbia j a r containing PBS f o r 5 min. - 165 -APPENDIX II - IMMUNOFLUORESCENCE STAINING OF KERATIN FILAMENTS (CON'T) e. Apply 50ul of properly d i l u t e d f l u o r e s c e n t l y tagged secondary antibody; goat a n t i - r a b b i t IgG. f . Incubate as i n c. g. Rinse as i n d. h. Controls. ( i ) P o s i t i v e - c e l l s known to be p o s i t i v e f o r k e r a t i n stained with primary and secondary antibodies (eg. 1° cultures of ROSE c e l l s ) . ( i i ) Negative - rat lung or human f i b r o b l a s t s stained with primary and secondary antibodies. - OSE c e l l s stained with normal rabbit serum i n place of primary antibody then secondary antibody. - OSE c e l l s stained with secondary antibody only. (4) Mounting. a. Place small drop of Gelvatol s o l u t i o n onto a glass microscope s l i d e . P r i o r to use, microscope s l i d e s should be cleaned by immersion i n chromic a c i d f o r 15 to 30, rinsed i n running tap H2O f o r 30 min., rinsed 3X i n dH^O. b. Wipe PBS from surface of c o v e r s l i p without c e l l s , and i n v e r t c o v e r s l i p onto the Gelvatol. c. Allow to dry f o r several hours then s e a l edges of c o v e r s l i p with c l e a r n a i l p o l i s h . d. View by epifluorescence microscopy. (5) Photography a. Black and white p r i n t s . Kodak Tri-X Pan f i l m (ASA 400) exposed f o r 15 to 45 seconds at a camera ASA s e t t i n g of 1600. Develop i n Microdol according to manufacturer's s p e c i f i c a t i o n s . P r i n t on Ilfospeed paper. b. Color s l i d e s Kodak Ektachrome c o l o r s l i d e f i l m ASA 400 exposed f o r 1 to 3 min at a camera ASA s e t t i n g of 800. Develop by "push processing" (ASA 800). - 166 -APPENDIX I I I - HOECHST DMA ASSAY (A f t e r Labarca and Paigan, 1980) (1) Solutions a. Hoechst Dye #33258 (bisbenzimide) (20 u/ml) Slowly disso l v e 20 ug Hoescht dye/ml dH 20. Store frozen at -20° C i n 5 ml aliquotes b. Phosphate-buffered s a l i n e (PBS). To 750 ml dH 20 add 7.1 g Na 2HP0 4; 116.88 g NaCl; and 0.84 g EDTA . Add s u f f i c i e n t volume of dH 20 to make 1.0 1. Adjust pH to 7.4. c. TRTPCK,trypsin (0.25$). Dissolve 100 mg TRTPCK lyophilyzed t r y p s i n i n 40 ml PBS. Store at -20°C. d. DNA standard stock (100 ug/ml). Dissolve completely 1 mg calf-thymus DNA i n 10 ml PBS. Store at -20°C i n a l i q u o t s . Do not refreeze. (2) Preparation of DNA standards. a. Thaw 100 ug DNA/ml stock and d i l u t e to 10 ug DNA/ml with PBS. Prepare the following d i l u t i o n s with 10 ug DNA/ml stock and PBS: 1 ug/ml, 2 ug/ml, 3 ug/ml, and 4 ug/ml. b. D i l u t e 10 ug DNA/ml stock to 1 ug DNA/ml and prepare the following d i l u t i o n s : 100 ng/ml, 200 ng/ml, 400 ng/ml, 600 ng/ml, and 800 ng/ml. c. Dilute 1 ug DNA/ml stock to 0.1 ug DNA/ml an prepare the following d i l u t i o n s : 10 ng/ml, 20 ng/ml, 40 ng/ml, 60 ng/ml, and 80 ng/ml. DNA standards required w i l l depend on the number of c e l l s i n cultures to be examined. Standards are stable f o r at l e a s t 16 hours i f stored at -20°C. (3) Add 150 ul Hoechst dye stock (20 u/ml) to 3 ml of each of 2a, 2b, and 2c. Mix well and read % transmission on spectrophotofluorometer. E x c i t a t i o n wavelength - 354 nm; Emission wavelength - 458 nm. (4) Preparation of c e l l samples a. Grow c e l l s i n 2 cm^ wells. At appropriate density, add 0.5 ml TRTPCK t r y p s i n to each well, cover with Parafilm and store at -20°C. b. When convenient, thaw cultures, allow t r y p s i n to act f o r 15 - 30 min. Mix well by p i p e t t i n g , t r a n s f e r to test tube, and add PBS to make 3 ml. Add 150 ul Hoechst stock (20 mg/ml). Mix well and read fluorescence. c. The number of c e l l s present can be calculated on the basis of approximately 7 pg DNA (Altman and Katz, 1976). - 167 -APPENDIX IV - TRITIATED THYMIDINE [(^H)TdR] INCORPORATION INTO DNA (After K r y s t a l , 1983) (1) C e l l culture Seed c e l l s i n 96 well m i c r o t i t r e plate i n 0.1 ml of medium. Allow c e l l s to grow to subconfluence. (2) (3H)TdR Addition a. Add (^H)TdR, 1 uCi/well by d i l u t i n g stock (lmCi/ml) 10 times to obtain 0.10 mCi/ml and add 10 ul of t h i s to each w e l l . b. Incubate f o r 3 hours at 37°C. c. Remove culture medium, and add 0.2 ml of 0.25$ trypsin/0.02$ EDTA/0.8$ Triton-X 100 i n Hank's Balanced S a l t Solution ( C a + + / M g + + - f r e e ) . Incubate at 37°C u n t i l c e l l s l y s e (approximately 15 minutes). (3) C e l l Harvesting a. Place glass micro-fiber f i l t e r into c e l l harvestor (rough-side up). b. Aspirate 12 wells with trypsin/EDTA only f o r background counts. c. Wash 20X with dH 20 and then with 10 to 20 ml of MeOH. d. Transfer f i l t e r s to s c i n t i l l a t i o n v i a l s , allow to dry i n fume hood f o r 1 hour, add 10 ml Omnifluor s c i n t i l l a t i o n f l u i d , and count. e. Proceed with sample wells as i n b to d. (4) Controls a. C e l l c o n t r o l - Aspirate several wells with c e l l s only and no (^H)TdR. b. (^H)TdR c o n t r o l - Set up several wells that contain medium only and no c e l l s , add (^H)TdR, then treat the same as wells containing c e l l s . T o t a l CPM - Add 0.01 ml of (?H)TdR s o l u t i o n to small piece of clean glass micro-fibre f i l t e r , t ransfer to s c i n t i l l a t i o n v i a l , and l e t dry f o r 1 hour before adding s c i n t i l l a t i o n f l u i d . - 168 -APPENDIX V - DEXTRAN CHARCOAL TREATED SERUM a. Add 10 mg Norit activated charcoal and 1 mg Dextran, per ml of serum to be treated to the serum. b. Shake i n water bath or incubator at 37°C f o r 2 hours. c. Spin at 4000 RPM (3000 g) f o r 30 min. at 4°C. d. F i l t e r through 0.22 ym low-protein-binding Gelman f i l t e r . e. Add to culture medium. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0096807/manifest

Comment

Related Items