UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Immortalization of human ovarian surface epithelium with a temperature sensitive immortalization agent Leung, Earnest Hin-Lung 1997

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

Item Metadata


831-ubc_1997-0256.pdf [ 6.91MB ]
JSON: 831-1.0087702.json
JSON-LD: 831-1.0087702-ld.json
RDF/XML (Pretty): 831-1.0087702-rdf.xml
RDF/JSON: 831-1.0087702-rdf.json
Turtle: 831-1.0087702-turtle.txt
N-Triples: 831-1.0087702-rdf-ntriples.txt
Original Record: 831-1.0087702-source.json
Full Text

Full Text

IMMORTALIZATION OF HUMAN OVARIAN SURFACE EPITHELIUM WITH A TEMPERATURE SENSITIVE IMMORTALIZATION AGENT by Earnest Hin-Lung Leung B.Sc, The University of British Columbia, 1994 A THESIS SUBMITTED LN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Reproductive and Developmental Sciences Program Department of Obstetrics and Gynaecology University of British Columbia We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA 1997 © Earnest Hin-Lung Leung, 1997  In  presenting  degree freely  at  this  the  available  copying  of  department publication  thesis  in  partial  fulfilment  of  the  University  of  British  Columbia,  I  agree  for  this or of  reference  thesis by  this  for  his thesis  and  study.  scholarly  or for  her  1 further  purposes  gain  that  agree  may  be  It  is  representatives.  financial  requirements  shall  not  that  the  Library  permission  granted  by  understood be  for  allowed  an  advanced  shall for  the that  without  make  extensive  head  of  my  copying  or  my  written  permission.  Department The  Vancouver,  Date  DE-6  of  University  (2/88)  3$  of  British  Canada  Qui  ? <r  Columbia  %«W*o  it  Qufywn)  Abstract  Although ovarian cancer is the leading cause of death from gynecological malignancies among North American women, the etiology and early developments in the progression of this disease are still among the least understood and characterized of all human malignancies. Understanding early events in the development of ovarian carcinogenesis is important.  Over 90% of ovarian  carcinomas are thought to arise from the ovarian surface epithelium (OSE) (Young et al., 1989). Previous attempts to study hOSE were hampered by lack of specimen availability and the limited lifespan of OSE  in vitro.  Immortalization  with wild type SV40 large T antigen (Tag) caused certain characteristics of hOSE to be lost: For example, CA125 is lost when hOSE are immortalized with wild type Tag. In the present study, immortalization of hOSE with a temperature sensitive Tag protein was attempted as a possible solution to the loss of these phenotypic characteristics. Four cases of hOSE were infected with a tsTag A 209 construct, and 28 monoclonal and 4 polyclonal lines were isolated. Two monoclonal and two polyclonal lines were observed for morphology and characterized by proliferation assays, indicators of immortalization (Tag and p53), differentiation markers (keratin, collagen and CA125) and senescence (SA j3-galactosidase) at the permissive (34°C) and nonpermissive (39°C) temperatures. All clones were  ii  composed of small, compact epithelial cells at the permissive temperature, and the cells became larger and flatter at the nonpermissive temperature. tsIOSE clones underwent 52-71 population doublings (pd) post infection at the permissive temperature and 2 -4 pd post infection at the nonpermissive temperature. Two of four cases increased expression of keratin and collagen when shifting from the permissive to the nonpermissive temperature of tsTag. The other two keratin and collagen values remained consistent through the temperature shifts. All cases expressed the epithelial marker at the nonpermissive temperature, as well as mamtaining a low level of CA125 that did not modulate with temperature in any case. No clones expressed SA P-galactosidase at the permissive temperature and all the clones expressed SA P-galactosidase at the nonpermissive temperature. The immortalization, differentiation, and senescence characteristics changed asynchronously after a temperature shift from the permissive to the nonpermissive temperature. Tag was lost by 24h, yet it took p53 at least 2 weeks to return to baseline levels. Senescence associated P-galactosidase began to return in 6 h, and was completely positive after 24h whereas senescence associated morphology did not return until after 24h. Keratin and collagen HI levels that were lost or affected by tsTag began to change after 24h. CA125 levels did not seem to be affected by temperature shifts.  in  One of four cases survived crisis and continued to replicate up to 150 population doublings past the expected lifespan of hOSE. This post crisis line maintained keratin at both temperatures despite the fact that its parent line was one that modulated keratin levels according to temperature shifts. This line has not formed subcutaneous or intraperitoneal tumors after four months in scid mice. Immortalization of hOSE with tsTag is superior to immortalization with wild type Tag because the system is capable of producing a large number of cells that can be phenotypically similar to hOSE. tsIOSE with inactivated Tag maintain the differentiated characteristics of hOSE better than cells immortalized with wt Tag, and thus are more like normal hOSE than IOSE. These cells are a potential source of cells on which large scale studies can be performed. hOSE was induced to form a continuous line, which is now available for further study.  IV  Table of Contents Abstract List of Figures List of Tables Key words and abbreviations Acknowledgments  ii vii viii ix xi  I. Introduction I. A. Human Ovarian Surface Epithelium (hOSE) I. B. Ovarian Cancer I. C. Problem and Approach I. D. SV40 Large T I D . 1. Introduction I. D. 2. Regions of SV40 Tag related to transforming activity I. D. 3. Cellular targets for the Tag I. E. Temperature Sensitivity I. F. Adenoviral constructs I. G. Infection and Cloning I. H. Characteristics of immortalized cells I. H. 1. Tagandp53 : I. H. 2. Growth potential and Senescence Associated /3-galactosidase I. H. 3, Differentiation I. H. 3. a. Keratin I. H. 3. b. Collagen I. H. 3. c. CA125 1.1. Post Crisis Lines I. J. Tumorigenicity IT. Objectives HI. Materials and Methods ITI. A. Culturing m. A. 1. Culture Medium.... III. A. 2. Primary cultures HI. A. 3. Decontamination HI. A. 4. Passaging III. A. 5. Freezing ITI. A. 6. Thawing III. B. Infection and Cloning HI. B. 1. Production of viral stock m. B. 2. Infection of hOSE HI. B. 3. Isolation of clones III. C. Analysis of tsIOSE HI. C. 1. Clones selected HI. C. 2. Proliferative capacity  V  1 1 3 5 :.7 7 8 11 14 15 16 18 18 19 20 20 22 22 23 24 27 29 ,.29 29 29 30 31 31 32 32 32 33 34 34 34 35  III. C: 3. Microscopy... HI. C. 4. Irnmunofluorescence m. C. 4. a. Tag.... m. C. 4. b. Keratin III. C. 4. c. Collagen III. C. 5. Immunohistochemical Staining i n . C. 5. a. CA125.. III. C. 5. b. Senescence Associated /3-galactosidase HI. C. 6.H-thymidine Incorporation m. C. 7. Southern Blot III. D. Post Crisis Lines n i . E. Tumorigenicity n i . F. Statistical Analysis IV. Results IV. A. Infection and Cloning IV. A. 1. Proliferation of stock IV. A. 2. Titration of virus IV. B. Characteristics of Immortalized cells IV. B. 1. Tag IV. B. 2. p53 IV. B. 3. Proliferative capacity IV. C. Phenotypic characteristics of tsIOSE IV. C. 1. Morphological changes IV. C. 2. Differentiation Markers IV. C. 3. Senescence Associated /J-gailactosidase IV. D. 37°C Cultures TV. E. Post Crisis Line V. Discussion.... VI. References  : .....2 2 2 2 1  3  vi  L L L  •••••  L  ,  L 1 1 1  A A ... L  A  :  f f i (. t , t t 8!  L i s t of Figures  Figure 1. The regions of the S V40 Tag and the binding sites for various cellular proteins that the Tag interacts with 9 Figure 2. Tag staining was performed on clones at (a) the permissive temperature (34°C) and (b) after 3 days at the nonpermissive temperature (39°C) 48 Figure 3. The presence of Tag transcript was tested by Southern blot in all four clones at the permissive and nonpermissive temperature 49 Figure 4. p53 immunofluorescent staining is very prominent in tsIOSE at 34°C (a), but is greatly reduced after 3 days at 39°C (b) 50 Figure 5. Morphological differences of tsIOSE at 34°C (a) and 39°C (b) and their differences over time 53 Figure 6. Keratin immunofluorescent staining in tsIOSE at 34°C (a) and 39°C (b) 57 Figure 7. Collagen immunofluorescent staining in tsIOSE at 34°C (a) and 39°C (b) 58 Figure 8. CA125 immunohistochemical staining in IOSE and tsIOSE at 34°C and 39°C.. 59  vu  List of Tables Table 1 Main characteristics and differentiation markers of cell lines cultured at the permissive and nonpermissive temperatures for tsTag 54 Table 2. Percent of cells stained positive for SA P-galactosidase at the permissive and nonpermissive temperatures for tsTag 62 Table 3. Percent of cells staining for differentiation and senescence markers at 37°C 63 Table 4. Percent of cells expressing differentiation and senescence markers by the post crisis line PC 160c 66  •ar-'  viii  Key Words and Abbreviations BSA  bovine serum albumin  DMSO  dimethyl sulfoxide  FBS  fetal bovine serum  FITC  fluorescein isothiocyanate  HBSS  Hanks' balanced salt solution  hOSE  human ovarian surface epithelium  HRP  horse radish peroxidase  IOSE  immortalized ovarian surface epithelium  NCS  neonatal calf serum  nonpermissive temperature the temperature at which a temperature sensitive protein is not active, (in this case, it is 39°C)  PBS  phosphate buffered saline  pds  population doublings  permissive temperature  the temperature at which a temperature sensitive  protein is active, (in this case, it is 34°C)  ix  SV40  Simian virus 40  Tag  large T antigen  ts  temperature sensitive  tsIOSE  temperature  sensitive  immortalized  epithelium tsTag  temperature sensitive T antigen  X  human  ovarian  surface  Acknowledgments  This thesis is dedicated to my parents, John and Rosaline, and my brother Constant for their support and love during the time I spent working on this thesis. I would like to thank my supervisor Dr. Nelly Auersperg for her constant moral support and guidance during my residence in her lab, my committee chair Dr. Peter Leung, who gave me a chance as a first year undergraduate to be exposed to real research and develop an interest and feel for it, and my committee members Dr. Shirley Gillam and Dr. Calvin Roskelley for their comments and suggestions in the development of the experiments for and the completion of this thesis. I would also like to thank the BC Foundation for Non-Animal Research for their financial support over the two years of my graduate program.  XI  I. Introduction I. A. Human Ovarian Surface Epithelium (hOSE)  hOSE in vivo is composed of cuboidal to flat cells that cover the ovary and is separated from the connective tissue cortex by a thin layer of tunica albuginea, inside which collagen fibres run along the surface of the ovary. hOSE has certain prototypical epithelial characteristics  in vivo.  The cells have a domed/rounded  apical surface covered with microvilli, and a basal lamina of varying thickness. There are desmosomes and tight and gap junctions connecting the cells. There are large nuclei, cytoplasmic free ribosomes, Golgi bodies, and rough and smooth endoplasmic reticulum dispersed in the cells. Cuboidal epithelium is also known as type A epithelium and flat or squamous epithelium is known as type B epithelium. Gillet et al. (1991) have suggested that type A is the original shape, and type B arises during post ovulatory repair of the hOSE. In general, mesodermally derived epithelia are quite differentiated. hOSE, however, has other characteristics in addition to these epithelial markers that suggest that hOSE is not as firmly differentiated as many other epithelia. An example of this is the presence of both keratin and vimentin. Furthermore, during postovulatory repair, hOSE changes to a morefibroblastic/connectivetissuelike form (Hafez and Makabe, 1982). This notion is supported when one looks at other experimental evidence. It is known that cells, in general, respond to explanation  1  into culture as they would to wounding. As a result, changes in gene expression occur that would simulate regenerative responses. hOSE responds to the culture environment by modulation from an epithelial to a more mesenchymal phenotype as indicated by anterior-posterior polarity, reduced intercellular cohesion, gel contraction, secretion of collagen types I and fn, and loss of the epithelial markers t  keratin and desmoplakin (Kruk and Auersperg, 1992; Auersperg et al., 1994). hOSE transports materials to and from the peritoneal cavity (Owens et al., 1992) and is involved in ovulatory rupture and repair during the reproductive cycle. It may also contribute to ovarian shrinkage that occurs with age. It has been implicated in the formation of OSE lined clefts, or invaginations and inclusion cysts, pockets within the ovary that are lined with hOSE (Kruk and Auersperg, 1992) and is suspected to be the origin of 85-90% of all ovarian cancers (Nicosia and Nicosia, 1988; Yancik, 1993). Developmentally, the hOSE is a form of modified pelvic mesothelium that arisesfromthe coelomic epithelium in the pregonadal area. This region is near the area that invaginated during development to give rise to the Mullerian ducts, which demonstrates a developmental relationship between hOSE, the epithelia of the oviduct, uterus and upper vaginal cells. This is important when considering the following: Neoplasms of hOSE tend to acquire more specialized and complex phenotypic characteristics of oviductal, endometrial or endocervical epithelium as  2  they progress to malignancy. This is an unusual observation, since most tumors become less, not more, differentiated as they progress (Clement, 1987.) I. B. Ovarian Cancer  Ovarian cancer is the fourth most common cause of death due to cancers among North American women, and is the leading cause of death among gynecological malignancies. One in seventy women (1.4%) will develop ovarian cancer, and most of these cancers will occur between the ages of 40 to 80 years, the average age being 59 years (Lynch et al., 1993). Despite the serious nature of this disease, the etiology and early events in ovarian carcinogenesis are among the least understood of all major human malignancies. The five year survival rate once diagnosed is poor (37% overall) (Hankinson et al. , 1995) and can be attributed largely to two facts: First, ovarian cancers are generally inoperable when they are first diagnosed (Gloeckler, 1993). Second, these cancers respond poorly to therapy (Merino, 1993). Although screening tests are available for patient follow-up and detection of advanced cases, no reliable means for early detection exist as genetic screens can detect less than 1% of all cases (Piver et al., 1993) and penetrance of these cases is not 100%. Women with at least two first degree relatives diagnosed with ovarian cancer are said to have a family history of ovarian cancer. The hereditary component of ovarian cancer can be divided into three groups: (1) breast-ovarian  3  cancer syndrome, (2) site specific ovarian cancer, and (3) Lynch syndrome II. Breast-ovarian syndrome involves families with a high incidence of breast and ovarian cancer. Site specific ovarian cancer deals with only clustering of ovarian o  cancer. Lynch syndrome II involves nonpolyposis colorectal cancer, endometrial cancer and ovarian carcinomas (Lynch et al., 1993). Development of ovarian cancer can also depend on nonhereditary factors, two of which seem to be readily identifiable. Use of oral contraceptives and multiparity seem to reduce the overall risk of "acquiring" ovarian cancer, even in women with family histories (reviewed in Piver et al., 1991).  It is these  correlations that have prompted theories that the continuous monthly ovulation is a factor in ovarian carcinogenesis (Bathalla, 1971).  The continuous ovulation  suggests a greater potential for cells to form inclusion cysts, and an increase in number of DNA mutations due to an increased number of divisions (Codwin et al., 1977) that results from the mitogenic response of hOSE to factors that are secreted to promote wound healing. Human ovarian surface epithelium (hOSE) is suspected to be the source of over 85% of ovarian carcinomas (Nicosia, 1986; Nicosia and Nicosia, 1988; Yancik, 1993).  This estimate is based mainly on histopathological changes  observed in hOSE during neoplastic progression of ovarian tumors. Chromosomal changes in histologically normal OSE adjacent to carcinomas (Zheng et al., 1993) have been observed, suggesting genetic changes in these cells. hOSE is a simple 4  epithelium, but as these neoplasms progress to malignany, they often acquire the more specialized and complex phenotyopic characteristics  of oviductal,  endometrial or endocervical epithelium. A good example of this is the change in expression of E-cadherin, which is normally reduced during neoplastic progression.  In hOSE, the normal level of E-cadherin is low, but as the  malignancy progresses, the levels of E-cadherin increase (Auersperg et al., 1995; Veatch et al., 1994). Cancer is thought to develop over time. One of the first stages is suspected to be immortalization of cells by mechanisms such as carcinogen-stimulated spontaneous mutation, or viral induction. This alone is not enough, but perhaps is the first step of a tumor progressing to malignancy.  Other steps, including  activation of oncogenes, are necessary before a malignant cell type can form (reviewed in DiPaolo et al., 1993). These steps seem necessary since neither normal hOSE nor hOSE infected with a viral immortalization agent immediately cause malignant tumors in immunocompromised mice. I. C. Problem and Approach  Studies of hOSE have been hampered by the small number of cells that can be obtained per specimen, and the relatively short lifespan of hOSE  in vitro.  These problems make it difficult for studies requiring many cells. This problem can be solved by introducing an immortalization agent into the cells to increase the  5  proliferative capacity of hOSE. This has been done with a wild type Tag, but it was found that certain characteristics such as CA125 were lost after the introduction of Tag into the cells . This problem can be solved by inactivating the Tag once proliferation of hOSE for a large number of cells is complete. Inactivation of Tag might reverse the effects of the Tag induced changes that occurred when using wild type Tag. This reversion of phenotypic characteristics would hopefully produce a large number of cells that would phenotypically be more be like hOSE without Tag than those hOSE produced by using a wild type Tag. After a literature review summarized in the next few sections, the approach to solving the problems described above was chosen. Although other methods of inducing immortality were available, the Tag system was chosen since it had proven to work in the hOSE system.  Different methods had been used to  introduce genes into human cells, and the most efficient method appeared to be by infecting the cells using adenovirus constructs containing the foreign gene. The temperature sensitive protein system was chosen since inducible systems such as the dexamethasone-induced Tag production method would introduce additional chemicals in the medium which may affect hOSE physiology in other ways.  6  I. D. SV40 Large T I. D. 1. Introduction Throughout this proposal, two words (irnmortalization and transformation) that are similar will be used, but they will have distinctly different meanings. There are controversies in the literature as to how and when these words are to be used. To avoid confusion, they will be defined here as they are used for the remainder of this paper. Immortalization is used to mean the extension of a cell's proliferative capacity over that of the normal cell by using an agent that interferes with cell cycle regulation. Transformation is used to mean that the cell has changed to a form that is tumorigenic when transplanted in a syngenic host. Infection of many rodent cells with SV40 DNA will result in full phenotypic transformation of these cells (Manfredi and Prives, 1994). If these are injected into syngenic hosts, they are highly tumorigenic. In contrast to the rodent cells, human cells infected with SV40 DNA results in immortalization. If injected into the syngenic hosts, they will not be tumorigenic. The initial sequences of the SV40 genome encodes two tumor antigens, Tag and small t, which are responsible for the tumorigenic activity (Fanning and Nippers, 1992). Tag is required for immortalization. Vectors which express only Tag show oncogenic properties which demonstrate that Tag is necessary and sufficient to transform established and primary cells (Cherington et al., 1988).  7  Small t itself does not have transfonning activities, but can enhance the activity of limited levels of Tag (Montano et al., 1990). The Tag is functionally complex. It is a nuclear phosphoprotein composed of 708 amino acids. It is the only protein necessary for SV40 viral DNA replication (Montano et al., 1990). There are many biochemical activities related to replication, but none of them are directly related to transformation or immortalization. Deletions and missense mutations of the Tag show that functions can be genetically linked to different sections of the amino acid chain (Rutila et al., 1986). I. D. 2. Regions of SV40 Tag related totiansformingactivity There are four regions directly and two regions indirectly related to the ttansforming ability of the Tag. Three regions have been identified as domains important in transformation, regions A, B and D. Each targets different cellular pathways.  8  106 112 • 20 123 IM PPPPP  «• p  IM 199 1(7 209  CD300 • 1  locaaiaiion pRB signal  DNA J *  ^  0  /  p  |«  [/> | mutations  933 STI 899 828 ••If a f ij.i.t  nudaar  PP a?7 an ..  tt 101119 128 112  IjptotphoiyUtna  •77 a n 7oi  A /„„/,^t.dJ—  hy*o(>hot>«:  zinc ffngsr  708  ragkm  DNA . POl  *107\  heal shock p73  DHAbnZng  ATPase  Figure 1. The regions of the SV40 Tag and the binding sites for various cellular proteins that the Tag interacts with. The numbers above the figure represent the amino acid residues of SV40 Tag, out of 708 total amino acids. The darkened regions in the horizontal bar designates the regions identified to be important in the function of Tag. The shapes under the bar show what cellular proteins Tag has been shown to interact with, (taken from Manfredi and Prives, 1994)  9  Region A, amino acids 17 - 70, affects transformation (Marsilio et al., 1991; Pipas et al., 1983). A protein defective in this region binds p53, pRb and pi07 (Montano et al., 1990), and but does not transform. Addition of small t complements this defect, which suggests that small t and Tag share a transformation region (domain) or that small t has an independent function that works similarly to Tag. The amino terminal is important for transformation, although no actual function has been assigned to it. Potential mechanisms for its activity or role in transformation are the binding to p300 (a cell cycle regulator responsible for preventing unchecked growth),  the involvement in the  transformation activity of Tag, and the binding of the heat shock protein p73 to the A region of Tag (Sawai and Butal, 1989). Region B, amino acids 101 - 118, also affects transformation. This region binds pRb and pl07, two cellular proteins. This region along with region A (aa 1 120) has been shown to be sufficient to cause immortahzation of cells in primary cultures (Colby and Shenk, 1982) and expression of Tag in established cells (Clayton et al., 1982). Region C, amino acids 126 - 132, is responsible for nuclear localization. Without this area, foci can not be induced in cells in primary culture, and results in no immortalization. This suggests that nuclear localization is important for immortalization. Established cells, cells that have been previously immortalized,  10  can be transformed even without the formation of foci and without nuclear localization (Kalderon et al., 1984). Region D, amino acids 351 -450 and 533 - 626, binds to p53 (Montenarh et al., 1985). Polymerase a primase has also been shown to compete for binding to Tag, suggesting that two proteins have overlapping binding sites on the Tag (Gagnon and Lane, 1987; Gagnon and Lane, 1990). A zinc finger region, amino acids 302 - 318, is important for the conformation of the T antigen. Without this region, reduced ability for viral DNA replication is seen. It suggests that the overall conformation breaks down if the zinc finger is removed (Schneider and Fanning, 1988). A hydrophobic region, amino acids 571 - 589, is responsible for the protein's overall stability. If this region is mutated, the metabolic half life is much shorter, but the protein is still active. I. D. 3. Cellular targets for the Tag There are two known main cellular targets for SV40 Tag, the tumor suppresser gene products p53 and pRb, as well as other potential targets. They all have different functions in the cell, and if their functions are disrupted, transformation may occur. Four main cellular targets have been suggested, although there may be many more. p53, pRB, pl07 and p300 all form complexes with SV40 Tag. Three of them, p53, pRb and pi07 seem to be inactivated by 11  either bmding to Tag or Tag-induced phosphorylation which results in altered cell growth and regulation (Schneider and Fanning, 1988).  p300 seems to be  inactivated by binding to Tag and is not controlled by phosphorylation. p53 is a tumor suppresser gene product that stops growth in DNA damaged cells and regulates apoptosis, or programmed cell death. It is a nuclear phosphoprotein that binds to DNA in a sequence specific manner and activates transcription (Lane and Benchimol, 1990; Marshall, 1991). It was first studied because rearrangements, deletions and missense mutations of the p53 gene were found to be the most common genetic alterations found in human cancers (Holstein et al., 1991). SV40 binds and either forms a stable complex with the p53 molecule or phosphorylates the p53 molecule rendering it inactive. Some immortalized cells seem to have lost the ability to regulate or tolerate overexpression of p53. Tag can induce cells to tolerate high expression of p53 by inactivating the negative growth regulating activity of p53 (Diller et al., 1990; Johnson et al., 1991; Shaw et al., 1992). Inactivation of p53 results in cells producing more p53 to compensate for loss of regulatory function. This causes a buildup of p53 in the cells expressing Tag. Cells infected with other agents may not be able to inactivate p53, and its overexpression can then lead to apoptosis and death of the cell (Yonish-Rouach et al., 1991; Shaw et al., 1992). SV40 Tag inhibits p53 from binding to DNA, which allows binding of DNA polymerase a to bind to SV40 Tag and results in the autostimulation of the production of Tag. 12  pRb (MW 110,000) is a nuclear phosphoprotein that is phosphorylated and dephosphorylated by the cell cycle dependent kinases p34 dc2 & p33cdk2 pRb c  m  binds to a nuclear anchor depending on its phosphorylation state, while Tag binds only to pRb in the under (hypo) phosphorylated state (Ludlow et al., 1989). Other DNA viruses (including human papillomavirus E7) target hypophosphorylated pRb, and it has been suggested that the hypophosphorylated form is active in the GI phase, exerting a negative growth effect on the cells when necessary. Phosphorylation allows progression into the S phase. Recent studies show that SV40 Tag stimulates the phosphorylation of pRb in quiescent cells (ones that are not growing) (Hu et al., 1992). This evidence suggests that SV40 Tag inhibits pRb function by direct binding and by phosphorylation (Hu et al, 1992). Adenovirus Ela and human papilloma virus E7 also bind to pi07, suggesting that they have functions in cell cycle regulation. pl07 is usually found in both phosphorylated and nonphosphorylated forms in cells. Cyclin E-p33 dk2 c  + pi07 complexes are normally present in GI phase of the cell cycle, and cyclin A-p33  cdc2  + io7 complexes in S phase (Ewen et al., 1992). Both these p  complexes help regulate cell cycles, and SV40 Tag seems to be able to disrupt both these complexes, resulting in transformation and immortalization. No references directly demonstrating that pl07 is a tumor suppressor have been found, although evidence correlating mutations in pi07 and tumors have been found, suggesting that p 107 is a likely candidate to be a tumor suppressor. 13  p300 shares some sequence homology with both pRb and pl07 (monoclonal antibodies cross react between the three proteins) (Hu et al., 1991), and also demonstrates a role in the regulation of cellular DNA synthesis (Howe et al., 1990). Oligonucleotide enrichment methods showed that p300 binds to the sequence GGGAGTG, a sequence that is found in many cellular enhancers and promoters. p300 and TATA binding protein co-immunoprecipitate, and although it is possible that the two proteins interact with each other, it is more likely that they share common or close binding sequences on the transcript (Abraham et al., 1993). I. E . Temperature Sensitivity  Proteins are polypeptides of amino acids that form 3-D structures as the protein is translated from an RNA sequence. Each protein has a specific stability and half life at any given temperature. Mutations in the amino acid sequence that determines the proteins' 3-D structure may result in either a nonfunctional protein, a very stable protein, or an unstable protein. Unstable proteins can be divided into either proteins that are unstable and break down more rapidly than the unmutated protein, or proteins that are not as stable at the normal temperature and are more sensitive to temperature shifts. The temperature at which the protein is active is called the permissive temperature, and the temperature at which it is inactive is called the nonpermissive temperature.  14  Sensitivity of a protein to temperature shifts introduces a way to determine the effects of active and inactive proteins in cells. A nascent polypeptide emerging to form a protein forms a stable protein at the permissive temperature and may form an active or inactive protein at the nonpermissive temperature. The protein at the nonpermissive temperature becomes inactive over time if the nascent protein is active. Cells with the temperature sensitive protein can be grown at the permissive temperature to examine the effects of the active protein in the cell and the protein inactivated by a temperature shift. This would then allow one to examine the effects of the protein not being active in the cell, and compare the differences. I. F. Adenoviral constructs  Adenoviruses were first cultured and reported as viral agents in 1953 (Rowe et al., 1953) and recognized as a transmissible agent that caused degeneration of various epithelial cells. They are a non-enveloped, regular icosahedrons (20 triangular faces and 12 vertices) which have a diameter of about 65 - 80 nm (Home et al., 1959.) Fibres (structures projecting from the surface of each vertice) detennine the serotype or subclass of the adenovirus, and are thought to be the component of the virus that attaches to the cells to allow them to be endocytosed. Adenovirus contains 13% DNA and 87% protein, making it stable in solvents like ether and ethanol (Green and Pina, 1963), but can be lysed by 5M urea (Maizel et al., 1968), 10% pyridine (Prage et al., 1970), acetone (Laver et al., 1967), or multiple freeze-thaw cycles (Maizel et al., 1968). 15  Adenoviral replication is temporally regulated, meaning that early genes in the sequence of genome must be produced before the later genes can be produced (Chow et al., 1980). This is useful in producing adenoviral DNA constructs that can be introduced into cells by infection (with the DNA of interest incorporated within the adenoviral coat), obviating the need to use calcium phosphate, lipids or electroporation to introduce DNA into cells receptive to adenoviral infection. The constructs have the early genome of the adenovirus fused to the DNA of interest, and can be replicated by infecting 293 cells (Graham et al., 1977), a human embryonic kidney cell line that contains the late adenovirus genes. Early and late genes combine to allow all the proteins required for adenoviral replication. The DNA construct is packaged in adenoviral coat, and viral particles can be purified by centrifuging out cell debris (Green and Pina, 1963). I. G. Infection and Cloning  The introduction of DNA into mammalian cells have been accomplished by several methods.  Classic methods like calcium phosphate, lipofectin,  hpofectamine, have been used for years (Feigner et al., 1987). Calcium phosphate is a method that uses high phosphate concentrations to introduce DNA into mammalian cells.  Lipofectin, a monocationic lipid, and hpofectamine, a  polycationic lipid, coat DNA, and help carry the DNA across the cell membrane, and into the nucleus, where it may be translated and transcribed. Infection with engineered viral constructs have also been used as a technique to introduce foreign 16  DNA into human cells. Adenoviral infection of hOSE is much more efficient when compared to methods using lipids to introduce genes into hOSE (Leung et al., in press). Infection of cells involves exposing a culture of cells to a virus that can be taken up by cells through the cell membrane. Exposure is limited to one hour for two major reasons: a) too much virus can be lethal to the cells, and b) the infection is carried out in serum free medium, and most cells do not survive well for prolonged periods without serum. Cloning of normal human cells with limited lifespans is a simple procedure. No selection in the classical sense of using G418 or some other selection product is required since the ability of hOSE to continue to proliferate is very limited compared to that of immortahzed hOSE. Infected cells are passaged, reseeded sparsely, and allowed to grow. Colonies of cells that have acquired the ability to grow under cloning conditions will appear. Uninfected cells will grow slowly, but can be differentiated from immortalized cells by morphological differences. Such morphological differences exhibited by the infected cells include smaller and more compact shape, more uniform size, lack of contact inhibition of growth and movement and large numbers of replicating cells. Since the cells are seeded very sparsely, the uninfected cells replicate many times, and senescence before confluence.  17  Over several weeks, infected cells (in this study, cells containing the temperature sensitive immortalization agent) form foci (presumably clones) in various areas of the dish. When the colonies reach approximately 3 mm in diameter, they are scraped off and put first into wells, then into dishes to passage and/or freeze in preparation for characterization. I. H . Characteristics of immortalized cells  I. H. 1. Tag andp53 The Tag is the component of the SV40 genome that causes the prolonged lifespan of immortalized cells. It is important to ascertain that Tag is present in the nuclei of the suspected clones in order to show that this is the component introduced into the host or recipient cell.  This can be shown by  immunofluorescence or Southern blotting. Immunofluorescence is a method that can show if an antigenic protein is present in the cell or not. This method requires that the protein maintain the 3 D configuration of the epitope, and may not work if the protein is inactivated or structurally altered. Southern blotting will detect the presence or absence of a segment of DNA in the genome of the cells. This method can show the relative amount of DNA present in the cells. Both methods provide useful information about the state of Tag in the cells of interest. Tag binds to p53 so that it can not act in the cell, and the cell therefore continues to produce it and it accumulates in the nuclei in increased amounts. Immunofluorescent staining for  18  p53 therefore determines if Tag is still biologically active in disrupting the cell cycle regulatory mechanism. I. H. 2. Growth potential and Senescence Associated p-galactosidase Tag that is expressed in cells causes the cells to lose their growth regulatory capabilities and seem to be immortalized. However, in contrast to rodent cells, human cells rarely become truly immortal and usually acquire a prolonged but finite replicative potential. Normal somatic cells invariably enter a state of irreversibly arrested growth after afinitenumber of divisions, a process termed replicative senescence. It is thought to be a tumor suppressive mechanism and underlying cause of aging. Immortality, or escape from senescence, is an important step in malignant transformation, but the role of replicative senescence in aging is controversial. Studies on cells cultured from donors of different ages, genetic backgrounds or species suggest that replicative senescence occurs  in vivo, and  that organismic  lifespan and cellular replicative lifespan are under overlapping genetic control. Currently, the methods to measure the potential for cellular growth include growth curves, mitogenic response and ^H mymidine uptake. These methods are neither rapid, easy to control, nor useful for in vivo experiments, but are none the less the accepted methods for quantifying age of cells. The reactions that these methods invoke are not specific to senescent cells, so they can not be distinguished  19  quiescent or terminally differentiated cells in tissues. Evidence that senescent cells exist and accumulate with age  in vivo  has been lacking until recently (Dimri  et al., 1995). Upon replicative senescence in culture, several human cell types mcluding fibroblast, ovarian surface epithelium, hair follicles, and skin have been shown to express a P-galactosidase, histochemically detectable at pH 6, which was not expressed in either presenescent fibroblasts and keratinocytes, or quiescent fibroblasts and terminally differentiated keratinocytes (Dimri et al., 1995). It was also absent from immortal cells, but was induced by genetic manipulations that reversed immortality. This method was used in this study to detect the effect of temperature sensitive modulations on replicative potential. I. H. 3. Differentiation I. H. 3. a. Keratin Keratins are a group of water insoluble proteins (40 - 70 kDa) that form cytoplasmic intermediate filaments in many epithelial cells. There are 24 (hstinct types of keratin in human epithelia, and at least 8 more keratins specific to hair and nails, aka hard keratins. These can be divided into at least 7 major classes according to their immunologic reactivity and size. Keratins found in human cells are of an a-type, as opposed to a p-kerarin found in bird feathers. P-cytokeratins are distinguished by differences in structure and function, but are not discussed 20  since they are not relevant to human cell types, a-keratins are classified by acidity (type I) or neutrality/basicity (type LT), function and location. They are very insoluble proteins, a characteristic of proteins with many sulfur-containing amino acids. Keratins 7, 8, 18, and 19 are found in hOSE (van Niekerk et al., 1991; Auersperg et al., 1994; Cooper et al., 1985). These are important to consider because keratin is found in almost all epithelial cells in vivo and in vitro, but not in non-epithelial cell types like fibroblast, muscle, or neurons (Sun and Green, 1978; Sun et al., 1979.; Franke et al., 1978). Cells that stain positive for keratin demonstrate their epithelial nature which can help differentiate these cells from underlying stromal and fibroblastic tissue. Two factors were used to characterize the clones using this parameter. It has been shown to change as cells are passaged in culture. Such changes should be maintained in a cell line produced with an immortalization agent that has been successful in mamtaining the characteristics of the original cell type. It is also examined as a parameter because the antibodies used help distinguish hOSE from other cells of the ovary. Specimens from surgical cases may not be pure, and may have contaminants from other cell types in the peritoneal cavity for example connective and stromal tissue.  21  I. H. 3. b. Collagen Collagen is an extracellular protein that is normally found in the connective tissues. Collagen i n was chosen as a marker since it is differentiates hOSE from the underlying stroma  in vivo.  Epithelial-mesenchymal conversion was a  characteristic of hOSE, and if hOSE responded to culture conditions as if they were wounded, the cells would likely produce collagen, a mesenchymal marker. I. H. 3. c. CA125 CA125 is an antigenic determinant with protein and carbohydrate components that is located on the surface of ovarian carcinoma cells. It is secreted into blood circulation in >90% of patients with ovarian cancer (Einhorn et al., 1990). The antigenic portion of the protein is 2000-3000 kDa and the mucin like glycoprotein is >200kDa. It is normally found in normal mucosae derived from the Mullerian duct (such as endometrium and fallopian tubes), lung, and mesothelium. It is present during endometriosis and pregnancy. It is present in the amnion cells, so it is speculated that CA125 has a function in development. It is useful as a marker to track the efficacy of anticancer therapy, but its presence in normaltissuesmakes this difficult to be specific to cancer cells. It is currently used to monitor cancer relapse in patients with initially high levels of CA125, since antigens leak to blood and increase in the circulation if the tumor begins to grow again. Return to a negative value does not guarantee a complete response to  22  therapy since patients who undergo restaging laparotomies have been found to have persistent cancer in 50% of the cases with negative preoperative CA125 levels (Berek et al., 1986). Nonetheless, it is the best marker for ovarian cancer so far, even if it is not specific enough. Einhorn (1990) showed that 175 of 5550 women had elevated CAI25 serum levels, and 6 cases of ovarian cancer were detected. The positive predictive value for the study was 3.4%, and 4 of the 6 women had stage ni cancers at detection. A study by Jacobs (1988) of 1010 postmenopausal women had a positive predictive value of 0.04 % with a false positive rate of 3%. CA125 was chosen as a marker for several reasons. IOSE did not retain CA125 when immortalized with a wild type Tag, even though the culture from which it was derived had CA125 positive cells. tsTag is known to retain some of the cell specific phenotypic markers, and if CA125 is maintained with this immortalization agent, it this immortalized cell may be more like normal OSE cells. CA125 status is thought to be different between specimens from women with and without family histories of ovarian cancer. 1.1. Post Crisis Lines Crisis is a stage that all human cells in  vitro,  normal or immortalized, go  through. Normal cells as they senesce undergo crisis. At such a stage, a few cells may stay attached to the culture dish and become quiescent. Human cells have  23  never spontaneously immortalized or survived crisis, so normal cultures reaching senescence as defined by Schaeffer's criteria are generally discarded. Cells that have survived crisis and propagated 150 pd more than the average normal cell (of the same lineage) can be referred to as continuous line (Schaeffer, 1984). Such a line is generally considered to be truly immortalized, and may continue to proliferate for many more passages. It is also possible, however, that more mutations that accumulate due to unchecked replication of DNA and division of cells will cause the cells to die suddenly. The establishment of post crisis lines was attempted in this study to see if a continuous line could be produced for further study. tsTag is an immortalization agent which has been successful in producing continuous lines from cell types that are traditionally difficult to immortalize, and from cells that are not isolated from tumors. I. J. Tumorigenicity  Several studies using animal OSE have been performed to try and understand ovarian carcinogenesis. Studies such as the effects of corpus luteum extracts (Setrakin et al., 1985) and OSE proliferation from reflex ovulators (rabbits) (Osterholtzer et al., 1985), but the results are not comparable to human in vivo  or in vitro situations since the majority of human ovarian cancers originate in  the OSE, whereas in other species except for the hen, ovarian cancers originates in  24  either the stroma, the granulosa or corpus luteum. Although the reproductive system of the hen is a good model for ovarian endocrinology and OSE-derived carcinogenesis, it differs biologically from that of the human, (the most obvious difference being that hens incubate their developing young externally while humans incubate their young internally), and any correlations made to this animal model may be suspect. It has been speculated though, that the repeated ovulation mat hens undergo may be the cause of the ovarian cancers originating in the OSE, which correlates with one of the theories of the causes of ovarian cancers in humans. The common method used to study tumorigenicity of cells is injection of samples into immunocompromised mice. In the past, nude and scid mice have been used to test this property. Nude mice are athymic, immunodeficient mice which lack a functional T-cell system. They have an inherited defect of the epithelial cells in the skin, and in the lining of the third and fourth pharyngeal pouches, causing thymic hypoplasia. A basic thymus is present, but T-cell maturation can not occur, resulting in failure of cell-mediated immune reactions such as allograft rejection, delayed type hypersensitivity, and antibody responses to T cell dependent protein antigens. Normal natural killer (NK) cell development and macrophage stimulation with y-interferon occurs in these mice, which seems to be responsible for the resistance of nude mice to microbes and tumorigenic cells that do not replicate rapidly. Scid (Severely combined immunodeficient) mice  25  lack B-, T-, and NK cells due to an early developmental bone marrow malformation, and are therefore susceptible to microbial assault and less likely than nude mice to resist formation of tumors by slow growing cells. In the present study, cells were injected subcutaneously to attempt to cause tumors under the skin of scid mice. IP injections were performed to see what effects the immortalized, post crisis OSE had in the peritoneal cavity, the natural niche for ovarian surface epithelium cells.  26  n. Objectives  The principal objective of this thesis project was to establish cell lines that could express the phenotypic characteristics of normal hOSE under experimental conditions. The cell lines should have a high proliferative capacity and should either maintain the differentiated characteristics of hOSE through the immortalization, or revert to the hOSE characteristics upon inactivation of the immortalization agent.  The mechanism chosen for this inactivation is the  temperature sensitive Tag protein system. Ultimately, the system should provide a large number of cells that phenotypically resemble hOSE under proper culture conditions. Using these cells, the following hypotheses were addressed: 1. that the immortalized clones have an extended lifespan, and their proliferative capacity is increased by the introduction of tsTag into the genome. This hypothesis arises from the literature studies that tsTag is an effective immortalization agent in some human cell types. The confirmation of this hypothesis would prove that tsTag is effective in hOSE and that tsTag immortalized cells are a potential source of hOSE to perform large scale studies on. 2. that certain phenotypic characteristics are changed and others maintained by the immortalization procedure which can be changed by temperature shifts. This hypothesis arises from studies showing that keratin and collagen  27  expression are maintained and CAI25 is lost in hOSE that have been immortalized with wild type Tag and that characteristics induced by tsTag in hepatocytes (Harris et al., 1995) and trophoblast cells (Chou, 1989; Lei et al., 1992) can be reversed by temperature shifts.  The confirmation of this  hypothesis would suggest that the cells with the temperature-inactivated Tag are more like normal hOSE than hOSE with wild type Tag. 3. that a continuous line can be produced using this method and immortalization agent. This hypothesis is based on placental and hepatocyte continuous lines that have been produced using this immortalization agent. The confirmation of this hypothesis would prove that hOSE can be artificially induced to form continuous lines and introduce the first continuous hOSE lines for further studies. 4. that the characteristics of continuous hOSE lines may not be the same as immortalized hOSE. This hypothesis is based on the fact that continuous lines are more like cancer lines than regularly immortalized lines in that they continue to proliferate without an endpoint.  The confirmation of this  hypothesis would show that clones formed by tsTag are able to form truly immortalized (continuous) clonal lines.  28  HI. Materials and Methods DI. A. Culturing  HI. A. 1. Culture Medium Unless otherwise stated, the following conditions were used to culture cells for this project.  hOSE in primary culture were cultured in M199:MCDB  105(Sigma, St. Louis, MO.), 15% FBS, and 50fig gentamicin/ml (Canada Life Technologies, Burlington, Ont.) in a humid, 5% C0 /air, 37°C incubator. hOSE 2  infected with tsTag were cultured in M199:MCDB 105and 5% NCS (Canada Life Technologies, Burlington, Ont.) at 34°C for isolation and proliferation of immortalized cells, but in M199:MCDB 105 and 15% FBS in a humid, 5% C0 , 2  34°C, 37°C or 39°C incubator, depending on the temperature required for the comparison of tsTag induced effects. III. A. 2. Primary cultures The use of human tissues for research was been approved by UBC Screening Committee for Research on Human Subjects. Primary cultures were obtained from three sources, two in Vancouver, one source from out of town. Biopsy specimens were obtained from appropriate abdominal surgeries and scraped specimens were obtainedfromlaporoscopies at Vancouver Hospital, from women with no indications of cancer. Sterile specimens were obtained directly  29  from the surgical suite and immediately put into CO2 independent medium (Canada Life Technologies, Burlington, Ont.) for transport to BC's Women's Hospital. The rest of the biological work was done in a biological safety cabinet. Surface blood was washed off from biopsy specimens by rinsing two times in media. The external surface of the specimen was scraped with a rubber scraper, and epithelium dislodged into medium in 35 mm tissue culture dishes (VWR Scientific, Edmonton, AB). Scrapings are collected by centrifuging 1000 rpm for 5 min in a clinical centrifuge and resuspended in supplemented M199:105. The culture dishes were left undisturbed for a week to allow the epithelium to settle and stick to the dishes. When the dishes were 80% confluent, the cultures were infected with the adenovirus tsSV40 Tag as described below. i n . A. 3. Decontamination Cells between passages 0 and 7 that were rnildly contaminated with bacteria were treated with 50 |ig gentamicin/ml for a week. Cells contaminated with molds were treated with fungizone (40 M-g/rnL) (Canada Life Technologies, Burlington, Ont.) for a week. If no improvement to the culture was evident after two days, the culture was discarded. If after a week, the culture seemed healthy, the treatment was stopped, and continued to grow for experiments. If after a week, the culture was still contaminated or contamination recurred, the culture was discarded.  30  HI. A. 4. Passaging Cells were passaged using 0.06% trypsin (Worthington Biochemicals Corporations, Freehold, NJ), 0.01% EDTA (VWR Scientific, Edmonton, AB) in Ca , Mg ++  ++  free Hanks' Balanced Salt Solution (HBSS) (Canada Life  Technologies, Burlington, Ont.). Culture medium was removed, trypsin/EDTA added and the cells were observed until they rounded up. Cells were flushed off with a curved pipette, and trypsin neutralized with an equal volume of medium containing 5% serum. Cells were pelletted with a clinical centrifuge at 1000 rpm for 5 min. Supernatant was removed, and the cell pellet resuspended in an appropriate volume of medium for subculture or cryopreservant for freezing. III. A. 5. Freezing All cells werefrozenusing 10% DMSO (VWR Scientific, Edmonton, AB), 40% FBS in M199:MCDB 105 as a cryopreservant.  Cells were tiypsinized,  centrifuged and resuspended in 1.2 ml cryopreservant solution per vial. Approximately 10 cells were frozen per vial. 6  The freezing protocol was  according to instructions accompanying the Taylor-Wharton Handi-Freeze freezing tray (Essex Medical Products, Edmonton, AB). Briefly, it was a decrease of 1°C per min until the solution reached -20°C. Vials were then quickly cooled to liquid nitrogen temperature (-120°C) and then stored in a liquid nitrogen dewar until needed.  31  ill. A. 6. Thawing Cells were thawed by removing cryovials from storage in liquid nitrogen, and plunging them into a 37°C water bath. They were brought to 37°C within 1 2 min and the cryopreservant solution diluted 1:1 with culture medium. The cell mixture was pelletted in a clinical centrifuge at 1000 rpm for 5 min, supernatant removed, cells resuspended in culture medium and plated into culture dishes. Cells to be examined were cultured on glass coverslips until confluent. III. B. Infection and Cloning  HI. B. 1. Production of viral stock The temperature sensitive, origin deficient virus was obtained through Dr. Janice Chou, NIH (Chou and Martin, 1974). It is a chimeric virus that contains the late genes coding for late replication products of the adenovirus replication cycle, plus a temperature sensitive SV40 large T antigen. The mutation that causes the temperature sensitivity is a substitution at position 209, hence its designation A o92  The 293 cells (Graham et al., 1977) were obtained from Dr. Eva Thomas, BCCH Department of Pathology, and cultured in Dulbeco's modified minimum essential medium (DMEM) (Canada Life Technologies, Burlington, Ont) with 4% FBS at 37°C, 5% C 0 in a humidified incubator. Medium was changed when necessary, 2  and cells grown to confluence before freezing for storage or infection for viral production. 32  For viral production, 293 cells were rinsed 3 x 1 min with serum free medium, and covered with virus estimated at 10 particles in 1 mL serum free 5  DMEM and incubated for 1 h at 34°C. The cultures were then rinsed 3 x 1 min with serum free medium, and cells incubated with growth medium. When a cytopathic effect was evident in about 80% of the cells approximately 4 days later, the cell debris was centrifuged out at 1000 rpm in a clinical centrifuge for 10 minutes and supernatant harvested. Vims-containing supernatant was aliquoted and frozen at -70°C. HI. B. 2. Infection of hOSE Four cultures of hOSE were taken at confluence in primary passage, and infected with the virus. Three cases were specimens from women with no family histories of ovarian cancer, and one case was from a woman with a family history of ovarian cancer. The woman with a family history of ovarian cancer had a sister who died of breast cancer, an aunt, grandaunt, maternal grandmother, mother and another sister who had been diagnosed with ovarian cancer. The woman had undergone a prophylactic oophorectomy from which we obtained a scraping. Cultures in 35 mm culture dishes were rinsed three times with serum free medium, and 1 ml virus (Stock solution diluted 10" in serumfreemedium) was applied to 9  the cells for 1 h, with gentle rocking every 15 min. Virus solution was rinsed off three times with serumfreemedium and the cells incubated in complete medium  33  for 48 hours at 34°C. Then the cells were split 1:8 and incubated without disturbing for two weeks. Medium was then changed as needed, and the cultures observed for colonies. 111. B. 3. Isolation of clones Colonies of infected cells appeared as foci of rapidly growing cells on a background of flat, senescing uninfected hOSE consistently after 3 weeks. When the colonies reached a diameter of about 3 mm, they were scraped off using a rubber scraper, and trypsinized to break up the clumps. Cells from individual colonies (clonal lines) or from pooled colonies (polyclonal lines) were seeded into one well of a four well plate (1.5 cm ), and allowed to proliferate at 34°C. The 2  cells were passaged for 2 x at 1:3 splits to allow the majority of non-infected cells which may have been scraped off with the colony to senesce. The cells were then either frozen for stock, or passaged onto coverslips for experiments. HI. C. Analysis of tsIOSE HI. C. 1. Clones selected Several clones were selected for this study. 160a, a nonfamilial monoclonal line with atypical morphology was chosen to represent the atypical hOSE that one finds in normal hOSE cultures.  160c, a nonfamilial monoclonal line with  epithelial morphology, was chosen to represent the epithelial population that one  34  finds in normal hOSE cultures. 166h, a nonfamilial polyclonal line, was chosen because polyclonal lines are more likely to represent a culture of hOSE as a whole. 178x, a familial polyclonal line, was chosen to investigate the possible differences between immortal familial an nonfamilial cases of immortalized clones. LTI. C. 2. Proliferative capacity Clones in early passage (p 4-8) were seeded in four wells (2 cm^) at 1.5 x 104 cells per well and incubated. Cells were dissociated and counted when the cells were confluent, and values for the four wells averaged. The cells were pooled and 1.5 x 10^ cells reseeded into four wells as described above, and the cycle continued. The experiment was terminated when the cells senesced (do not reach confluence after thirty days) as defined by Schaeffer (1984). Cases of unimmortahzed hOSE were treated in the same manner as the clones to compare growth rates. The time between plating and the next counting was measured in days, and cell doublings were calculated by the formula in Schaeffer (1984) \ Population doubling time was calculated by dividing number of cell doublings by days between counting and expressed as mean ± sd.  Number population doublings=Log|Q(N/NQ) x 3.33 where N=number of cells attimeof counting, and Nn^number of cells seeded.  35  A cyclic GMP inhibitor, H89 (Fujii et al., 1995), had been recently shown to block senescence of a tsTag immortalized human dermal fibroblast post temperature shift to the nonpermissive temperature. This compound was tested on tsIOSE at 39°C to try and extend the proliferative capacity of tsIOSE at the nonpermissive temperature. Cells were plated at 34°C and grown for 3 days. Cells were then treated with H89 diluted in M199:MCDB 105 at 10" , 10" , 10" , 4  5  6  10", and 10" M and shifted to the nonpermissive temperature. Cells were maintained until senescence at the nonpermissive temperature. This experiment was performed 3 times in triplicate dishes each time. UI. C. 3. Microscopy Live cultures were examined with a Leitz inverted phase microscope (Diavert Microscope), and photographed with a Leitz microscope camera (Wild PhotoAutomat MPS45) using Kodak 2415 film. Fixed and stained cultures were examined with an Axiophot photomicroscope (Carl Zeiss, Oberkochen, Germany) and photographed using Kodak T-MAX 400 film (pushed to 800) (Kodak, Rochester, NY) with a 40X neofluor (Carl Zeiss, Oberkochen, Germany) objective. 1000 cells per experiment were counted and positive cells were expressed as a percent of total.  36  HI. C. 4. Irnmunofluorescence Cells on coverslips were first rinsed in serum free medium for 5 min, fixed with cold (-20°C) methanol (VWR Scientific, Edmonton, AB), and stored at -20°C for at most 1 week before staining.  On the day of staining, cells were  permeabilized for 5 minutes in -20°C methanokacetone (1:1). Coverslips were rinsed 3 x 5 min in PBS and incubated 20 min in 5% goat serum (Canada Life Technologies, Burlington, Ont) in PBS/1% BSA (Sigma, St. Louis, MO.) to block non-specific staining by the secondary antibody. Experimental cultures were blotted and the primary antibody added and then incubated for 1 h at room temperature, unless otherwise noted. Control cultures did not receive primary antibody and served as a specificity control for the second antibody. All cultures were then rinsed 3 x 5 min in PBS and incubated for 1 h. at room temperature with the secondary antibody to which was attached the immunofluorescent label. They were then rinsed 3 x 5 min. in PBS and mounted onto slides with Gelvatol pH 6.5 (O'Guin et al., 1985). The positive control for Tag, p53, keratin and collagen staining was IOSE-80, aka IVAN (Maines-Bandiera et al., 1992) IE. C. 4. a. Tag The primary antibody for the Tag was aT Ab2 @ 5^ig/ml (original cone, dilution) (mouse Mab, Oncogene Science) and secondary antibody was a goat mouse IgG-FITC conjugate (Jackson Labs at 1:100). Primary antibody for p53  37  was ccp53 @ 5[ig/ml (mouse Mab, Oncogene Science) and secondary antibody is a goat mouse IgG-FITC conjugate (Jackson Labs at 1:40). m. C. 4. b. Keratin The primary antibody, a mixture of mouse a-AEl (1:8 final dilution) and mouse ct-AE3 (1:4 final dilution), were monoclonal antibodies against the acidic and basic keratins found in epithelial cells obtained from Dr. TT Sun (1978). Secondary antibody was a goat cc-mouse antibody conjugated to Texas red (Jackson ImmunoResearch Labs, Mississagua, Ont), a fluorescent label used at 1:160 final dilution. UI. C. 4. c. Collagen Primary antibody was a rabbit a Collagen HI (Southern Biotechnology Associates Inc., Birmingham, AB) used at 1:40 final dilution and second antibody was a goat anti a rabbit antibody conjugated to FITC (Jackson ImmunoResearch Labs, Mississagua, Ont), a fluorescent label used at 1:300 final dilution. HI. C. 5. lmmunohistochemical Staining HI. C. 5. a. CA125 Cells on coverslips were first rinsed in serumfreemedium for 5 min, fixed at room temperature with 10% formaldehyde (VWR Scientific, Edmonton, AB)/PBS, rinsed 3timesand stored in PBS at 4°C for at most 2 weeks before staining. On the day of staining, coverslips were incubated 20 min. in normal goat  38  serum (Canada Life Technologies, Burlington, Ont) to block non-specific staining. Experimental cultures were blotted and the primary antibody added and incubated for 1 hr. at room temperature. Duplicate cultures did not receive primary antibody and served as a specificity control for the second antibody. Coverslips were rinsed 3 x 5 min in PBS, a biotinylated second antibody applied to all cover slips, and incubated for an hour at room temperature. They were again rinsed 3 x 5 min. in PBS and incubated for 30 min at room temperature with streptavidin-horseradish peroxidase conjugate. Color was developed in a TRIS-Saline DAB solution until color in the positive control showed a definite positive. Color development was stopped by washing 2x5 min in distilled water. Coverslips were dried by passing through 75% and absolute ethanol (VWR Scientific, Edmonton, AB), then through 2 passes of xylene (VWR Scientific, Edmonton, AB). Cover slips were mounted on slides with Entellan (VWR Scientific, Edmonton, AB). Primary antibody was Mab OC125 (1:2000), obtained from Dr. R. Bast (1983). HRP-labeled goat anti-mouse IgG (Biorad, Mississaugua, Ont) was used at 1:200 final dilution as the second antibody. Color was developed in 0.025% chammobenzidine (Canada Life Technologies, Burlington, Ont)/0.01% hydrogen peroxide (VWR Scientific, Edmonton, AB)and 0.04% nickel chloride (VWR Scientific, Edmonton, AB). The ovarian carcinoma line OVCAR-3 was used as a positive control, and development time was adjusted for intensity for this reaction.  39  LTJ. C. 5. b. Senescence Associated P-galactosidase Cells were grown in culture dishes until 50% confluence, and serum starved for 48 h to make them quiescent. Cells were then fed with medium containing 40 LtCi  H-thymidine in 2 mL complete medium and incubated for 2 days. Cells were  rinsed in serum free medium for 5 min, fixed with 2% formaldehyde and 0.2% gluteraldehyde for 20 minutes, and stained for pH 6 P-galactosidase activity (Dimri et al., 1995) with 1 mg X-gal in 40mM citric acid/sodium phosphate, pH 6.0/5mM potassium ferrocyanide/5mM potassium ferricyanide/150mM NaCl/2mM MgCi2 overnight at 37°C. Dishes were washed with PBS and dried with methanol in preparation for autoradiography. The experiment was performed 3 times in duplicates each time. HI. C. 6. H-myrnidine Incorporation 3  Dishes were processed for autoradiography as described by Seshadri and Campisi (1990). Dried dishes were coated with a liquid emulsion under darkroom conditions. They were then placed in a lighttightbox with desiccant and stored at 4°C for four days. Before developing, the dishes were warmed up to room temperature to prevent condensation from forming on the dishes and mating development. Dishes were then developed with D19 (Kodak, Rochester, NY) for five minutes, washed with tap water for 30 seconds, fixed with Kodak fixer (Kodak, Rochester, NY) for 5 minutes, and rinsed again in water for five minutes  40  under darkroom conditions. The dishes were then air dried and examined by tight microscopy. HI. C. 7. Southern Blot 1 ug of the PX-8 plasmid (Fromm and Berg, 1987) were digested for 1 h with Bam HI in 50 mM Tris-HCL 1.0 mM MgC12, and 50 mM NaCl, and for a second hour in Xho I in 50 mM Tris HCl, 10 mM MgC12, and 100 mM NaCl at 37°C in a total of 50 ul. 10 ul loading buffer was added to the sample, and electrophoresed on an agarose gel for 12h (including molecular weight markers). The proper band was cut out using a scalpel, and probe purified by using a Wizard™ Plus Miniprep kit (Promega, Madison, WI). The probe was then labeled with alpha- P using a Nick Translation Reagent system (#18160-010, Canada Life Technologies, Burlington, Ont). lug of DNA per sample was digested to completion using Xba I and Bam HI , and electrophoresed on 5% polyacrylamide gels and stained with 0.5 ug.mL ethidium bromide solution. Gels were photographed under long UV light, then treated with 500 mL of 0.2N HCl (10 min), rinsed in water (4x1 min), 500 mL denaturation solution of 1.5 M NaCl/0.5 M NaOH (2 x 15 min), and 500 mL neutralization solution of 1.5 M NaCl/0.5M TrisCl, pH 7.0 (1 x 30 min). Gels were transferred on top of the Whatman Wick, and any air bubbles removed. Nylon membrane cut to the same dimensions of the gel was wetted with  41  20X SSC (3M NaCl, 0,3 M, Na3citrate 2H 0, pH 7.0) and placed on the gel. A 2  dry Whatman 3MM paper and 10 cm of paper towels were placed on top of the membrane. A glass plate and 0.4kg weight was placed on top of the paper towels and left overnight at room temperature. The system was dismantled the next day to allow for capillary transfer of the DNA from the gel onto the membrane. Nylon membranes were dried and irradiated by long wave UV light for 3 minutes each side to crosslink the DNA to the membrane.  Hybridization was performed  overnight with the labeled probe, washed, and subjected to overnight autoradiography at -70°C.  III. D. Post Crisis Lines  Cells reaching crisis at 34°C were maintained in that incubator for several months to test for the possibility of true immortalization by tsTag. Medium was changed once every three weeks to maintain L-glutamine levels, and cells observed at that time. When colonies appeared, surviving cells were cultured as before for four passages to produce stock for further experiments, and a population of the cells was maintained to see if the cells would fulfill the criteria of continuous lines. This included passaging cells until 200 pd over their normal lifespan.  42  III. E . Tumorigenicity  Eight scid mice were obtained from the animal unit at Terry Fox Laboratories (TFL).  The use of these animals was approved by the UBC  Screening Committee for use of Animals in Research. Cells post crisis were 6  7  mjected at 10 and 10 cells/mouse either subcutaneously or interperitoneally at two mice per treatment. Injections were performed at TFL, Vancouver, B.C. by Cindy Miller. The mice were housed at the animal unit at the BC Research Institute for Child and Family Health (BCRICFH), and cages, food and water autoclaved and changed weekly by animal unit technicians supervised by Trish Pomeroy, BCRICFH. This experiment was started in September, 1996. I am observing the mice for signs of tumorigenesis. Sites of subcutaneous injections are examined every two days for signs of tumor growth. Intraperitoneally injected mice are observed for abdominal distention, twitches, and matted hair, signs of internal problems. III. F. Statistical Analysis The results of three or more independent experiments were expressed as mean and standard error. The results of two experiments are listed side by side for comparison of range and error.  Statistical significance between means was  determined by a student t test.  43  IV. Results IV. A. Infection and Cloning  IV. A. 1. Proliferation of stock Some 293 cells were stored frozen for future use and the rest was used to propagate the virus. After infection of 293 with stock virus, cytopathic effects were evident after four to five days of incubation as cells progressively became larger and rounder before lysing. Cell death was evident when debris was found floating in the media. At harvesting, cell debris was centrifuged out with a clinical centrifuge. Since virus cannot be centrifuged out with a low speed centrifuge, this was an effective means of removing a major portion of the contaminants in the solution. The virus remaining in the supernatant was aliquoted out and frozen. IV. A. 2. Titration of virus A limiting dilution titration was performed on the virus solution with -2  -12  dilutions ranging from 10 to 10  using 293 cells as the target. The cells were  incubated for four days, to mimic the conditions at which the virus had been -2  -6  propagated. Virus, killed virtually all the cells with dilutions 10 to 10 . Some -7  -9  but not all cells died when cells was infected at viral dilutions of 10 to 10 . No cells died when virus was added at dilutions of 10 to 10 10  44  The results suggested that 10" to 10" dilutions would be a useful 7  9  concentration to infect cultures to ensure that some cells would be infected, but not to a very high degree such that several viral integrations would be possible. When the experiments were performed, the concern was that more than one copy of the virus would enter the cells, and multiple infections may cause different results than single infections. After thistitration,a dilution of virus harvest solution to 10"  9  was chosen since some but not all of the cells showed cytopathic effect. This suggested that a high proportion of cells were infected with just one virus particle. A lower dilution killed all the cells in the dish. Enough virus has been frozen and stored to infect 50 more primary cultures. The highest dilution that one could use to infect the cells effectively is 10" . 9  When infected cells formed foci or colonies among sparsely populated uninfected (and therefore unimmortalized) cells, they were scraped and trypsinized, then replated into new wells and subsequently into new dishes. The cells were frozen until needed for experiments. I have isolated 28 monoclonal and 4 polyclonal cell lines. The 28 monoclonal lines were each obtained from a distinct colony in the culture dish. Since the cells were split sparsely for cloning, each foci formed was assumed to have arisen from one cell, hence the monoclonal designation. The polyclonal lines were derived from a mixture of small clones that had fused, or were left over after the >3mm clones were removed.  45  The number of clones obtained are in accordance with generally accepted values published. Infection for the purpose of permanently introducing a gene into human cells is generally low, less than 1% , but is higher than transfection values for the purpose of permanently introducing a gene into human cells (Doren and Gluzman, 1984). IV. B. Characteristics of Immortalized cells  IV. B. 1. Tag A positive Tag result in the nucleus of the cell suggests that the cell is producing and expressing Tag. This suggests that the cells from the clone could be immortalized. All 4 clones had the Tag present in the nucleus at the permissive temperature at the same intensity as the positive control, IOSE-80. None of the four clones tested positive for the tsTag at the nonpermissive temperature, although the intensity of the positive control was the same as that cultured at the permissive temperature. An example of positive Tag staining is shown in figure 2. The presence of Tag gene was tested for by Southern blot in all four clones at the permissive and nonpermissive temperature. It was present in all clones at both temperatures at the same size or location in the autoradiograph (see figure 3). TV. B. 2. p53 Cells were stained for p53 to see if the Tag was biologically active at the permissive and nonpermissive temperatures. At the permissive temperature, the 46  intensity of p53 was as high as that of the positive control, IOSE-80. At the nonpermissive temperature, the intensity of p53 was lower after two days. After one week, the levels of p53 had decreased. After two weeks, p53 levels had diminished greatly, but were still higher than the levels shown by the negative control. An example of positive p53 staining is shown in figure 4.  47  Figure 2. Tag staining was performed on clones at (a) the permissive temperature (34°C) and (b) after 3 days at the nonpermissive temperature (39°C). Tag was stained by the indirect immunofluorescent technique using ms a Tag (l°Ab) and gt a ms FITC (2°Ab). Tag was present at 34°C, but absent at 39°C in all cases. Representative pictures are shown, magnification at lOOx.  48  C  160a 34C39C  160c 34C39°C  166h 34C 39C  178x 34C 39C  Figure 3. The presence of Tag transcript was tested by Southern blot in all four clones at the permissive and nonpermissive temperature. The probe for Tag was made from PX-8 plasmid and labeled with ct- P, and used to detect the presence of the Tag transcript in all clones at both temperatures. The lanes in order from left to right are: control (IOSE-80); 160a, 34°C, 39°C; 160c, 34°C, 39°C; 166h, 34°C, 39°C; 178x, 34°C, 39°C. The bands appeared to be the same size based on their location in the autoradiograph. 32  49  Figure 4. p53 imrnunofluorescent staining is very prominent in tsIOSE at 34°C (a), but is greatly reduced after 3 days at 39°C (b). p53 was stained by the indirect immunofluorescent technique using mouse a p53 (l°Ab) and goat a mouse FITC (2°Ab). High levels of p53 were present at 34°C, greatly reduced after 3 days at 39°C in all cases. Representative pictures are shown at magnification 400x.  50  IV. B. 3. Proliferative capacity Clones had the potential to undergo 52 to 71 population doublings (12 - 15 passages in this system) post infection before reaching senescence as defined by Schaeffer (1984) at the permissive temperature, but did not undergo more than two population doublings at the nonpermissive temperature. Clones in early passage (4 - 8) were followed for several weeks at the permissive and nonpermissive temperature.  Clones at the permissive temperature continued to grow at a  logarithmic rate. Clones at the nonpermissive temperature doubled one to two times before senescing after four weeks. If clones at high passage (9 - 13) were shifted to 39°C, they senesced within 24h (ie. before any population doublings occurred.) H89 did not extend the lifespan of the cultures shifted to the nonpermissive temperature at any of the concentrations tested. IV. C. Phenotypic characteristics of tsIOSE IV. C. 1. Morphological changes Detailed culture records and some pictures had been taken as cells were grown for experiments. Cells were counted at confluence 3 times and results averaged to deteimine how many cells could fit into the same surface areas. 35 mm culture dishes contained approximately 280,000 normal passage 1 hOSE and 75,000 normal passage 4 hOSE, and approximately 410,000 tsIOSE. Therefore, cells at the permissive temperature expressed a saturation density higher than  51  normal hOSE. The morphology of most of the clones remained similar to normal hOSE in low passage during comparable states of confluence. Two clones were compact epithelial in morphology when they were near confluence. The other two lines were more atypical throughout their life like some hOSE, but me line derived from a woman with a family history of ovarian cancer had whorly patches among the cells that looked faintly like fingerprint. There was little difference in cell morphology at the permissive temperature until the last two passages before senescence at passage 12 - 15. Then, cells slowly began dying and cells became larger and more rounded. Cells at the nonpermissive temperature acted differently depending on passage. Cells at low passage (p 6 - 9), once moved to the nonpermissive temperature grew very slowly if at all between one and three days. By four days to a week, the cells became big, flat and rounded. By the end of two weeks, many of the cells had died and floated off the dish. For cells in high passage (p 12 - 14), once moved to the nonpermissive temperature, the same change in morphology occurred, but over a shorter period of two to three days.  52  Figure 5. Morphological differences of tsIOSE at 34°C (a) and 39°C (b) and their differences over time. tsIOSE cells at the permissive temperature are small, compact and epithelial. After 3 days at the nonpermissive temperature, the cells are large, flat and still epithelial. Representative pictures are shown at lOOx.  53  Cell line/Clone Clonality Family history of Ovarian Cancer Morphology % Keratin 34°C % Keratin 39°C % Collagen 34°C % Collagen 39°C % CA125 34°C % CA125 39°C  160a  160c  166h  178x  clonal No  clonal No  polyclonal No  polyclonal Yes  atypical 96.3+0.3  epithelial 0.2 ±0.2*  epithelial 34.3 ± 1.5*  atypical 97.0 ±1.5  96.0 ± 0.6  92.3 ± 1.8*  85.0 ± 1.5*  99.7 ±0.3  42.7 ±3.5  24.0 ± 1.5*  33.7 ±4.1*  24.0 ± 1.7  39.7 ±3.2  48.0 ±3.8*  60.3 ±2.3*  22.7 ± 0.9  1.0 ±0.1  1.9 ±0.3  0.8 ±0.2  1.1 ±0.2 .  1.4 ±0.2  1.8 ±0.2  0.5 ±0.1  1.7 + 0.2  Table 1 Main characteristics and differentiation markers of cell lines cultured at the permissive and nonpermissive temperatures for tsTag. Values are from three experiments with cells in passages 8-11 and based on counts of 1000 cells per experiment (mean + se). * indicates that the difference between the temperatures is statistically significant. (p<0.05).  54  IV. C. 2. Differentiation Markers Keratin was examined in 3 clones from non-familial cases (2 monoclonal and 1 polyclonal line) and in 1 clone from a female with family history of ovarian cancer. The clones did not all react similarly to temperatures shifts. No lines were keratin negative at both temperatures. Two of four clones changed from a low keratin population to a high keratin population while the other two clones did not change. Interestingly, as shown in table 1, one of two clonal lines shifted from completely negative at the permissive temperature to almost completely positive at the nonpermissive temperature. All clones expressed the epithelial characteristic keratin at the nonpermissive temperature. A brief time course was performed in duplicate with this line to see at what time or how rapidly keratin appeared in the cell. Keratin appeared within 24h at 33%, at 66% by 48 h, and near 100% at 72 h. The other clonal line maintained a high level of keratin at both the permissive and nonpermissive temperature. The polyclonal line from the nonfamilial case had 34% keratin at the permissive temperature, which increased to 86% at the nonpermissive temperature. The magnitude in shift is not as great as that of the monoclonal lines, but may be a result of mixing lines that modulate expression due to temperature shifts and those that do not. The polyclonal line from the familial case was highly positive at the permissive temperature and the nonpermissive temperature.  55  Collagen lTi was examined in 3 clones from non-familial cases (2 monoclonal and 1 polyclonal line) and in 1 clone from a female with family history of ovarian cancer. The clones did not all react similarly to each other upon temperatures shifts. No lines were collagen negative at either temperature. With a shift from 34°C to 39°C, two of four clones changed from a low collagen population to a higher collagen population while the other two clones did not change (table 2). One monoclonal line, the one that had shifted from negative keratin to high keratin, modulated from 25% at the permissive temperature to 47% at the nonpermissive temperature. The other clonal line maintained 36 - 42% collagen level at both the permissive and nonpermissive temperature.  The  polyclonal line from the nonfamilial case had 33% keratin at the permissive temperature, which increased to 61% at the nonpermissive temperature. The magnitude in shift is as great as that of the monoclonal lines. The polyclonal line from the familial case showed similar, relatively low collagen levels at the permissive temperature and the nonpermissive temperature. All of the clones maintained low and relatively consistent values of CA125 through temperature shifts and were consistently expressed.  56  Figure 6. Keratin immunofluorescent staining in tsIOSE at 34°C (a) and 39°C (b). Keratin was stained by indirect immunofluorescent technique using mouse a AE1 and mouse a AE3 (PAb) and goat a mouse TR. Keratin was present at 34°C in most cases, and at 39°C in all cases. Representative pictures are shown at 400x.  57  Figure 7. Collagen mimunofluorescent staiiiing in tsIOSE at 34°C (a) and 39°C (b). Collagen was stained by indirect immunofluorescent technique using rabbit a collagen III (PAb) and goat a rabbit FITC (2°Ab). Collagen was present at 34°C and at 39°C in all cases. Representative pictures are shown at 400x.  58  Figure 8. CA125 irnmunoMstochemical staining in IOSE and tsIOSE at 34°C and 39°C. CA125 was stained using OC125 (mouse a CA125, l°Ab), goat a mouse-FLRP (2°Ab), and 0.025% DAB, 0.01% H202, 0.04% NiC12 for color development. Cells retained CA125 at both 34°C and 39°C. Representative pictures are shown at lOOx.  59  IV. C. 3. Senescence Associated (3-galactosidase SA p-gal expression was absent at the temperature at which Tag is active at all passages tested (8 - 12). Upon a shift in temperature to 39°C, SA P-gal begins to be expressed rapidly. At 6h, a small amount of SA p-gal was present, but levels increase dramatically by 12 h. At 24h, virtually all the cells expressed the senescence marker SA p-gal (Table 2). IV. D. 37°C Cultures  The purpose of testing the intermediate temperature was twofold: (1) to see if cells would survive longer than at the nonpermissive temperature, yet maintain characteristics demonstrated at 39°C.  This would provide more time for  experiments at the nonpermissive (or less permissive) temperature. (2) to see if one could avoid using a third incubator. The change in morphology from small, compact cells to large flat senescent cells was consistent with the change that appeared in the cultures at 39°C, although thetimeto cell death was longer by a week than at 39°C. Cells did not grow more than 4 population doublings at this temperature as compared to dying before population measurements could be taken. H89 did not visibly extend the proliferative potential of the tsIOSE at this temperature either.  The clones generally acted as if they were at the  nonpermissive temperature and expressed those characteristics (table 3). The only  60  difference was that in line 166h, keratin expression resembled 34°C, while it increased significantly at 39°C (Table 1).  61  Cell line/Clone % SA p-gal 34°C % SA p-gal 39°C 6h post shift % SA p-gal 39°C 12h post shift % SA p-gal 39°C 24h post shift  160a  160c  166h  178x  0.0 + 0.0  0.0 + 0.0  0.0 ±0.0  0.0 ± 0.0  7.0 ± 1.5  11.3 ±2.9  3.0 ± 1.0  7.3 ± 2.9  68.3 ±4.1  84.0 ±5.5  86.3 ±3.8  75.3 ±5.6  99.7 ± 0.3  100.0 ± 0.0  100.0 ±0.0  96.3 ±2.0  Table 2. Percent of cells stained positive for SA P-galactosidase at the permissive and nonpermissive temperatures for tsTag. Values are from 3 experiments with cells in passages 8 - 12, and based on counts of 1000 cells per experiment (mean ± se). The percent SA P-galactosidase increases from 6h to 24h post sMfting the cells from 34°C to 39°C. The increase seems initially slow at first, but increases rapidly after 6h.  62  Keratin Collagen CA125 SA 3-gal 24h post shift  160a  160c  166h  87, 92 25,32 0.9, 1.0 100, 100  178h  85, 97 49, 56 1.3, 2.0 100, 100  37,44 57, 62 0.6, 1.0 100, 100  100, 100 17, 26 1.1, 1.5 100, 100  Table 3. Percent of cells staining for (lifferentiation and senescence markers at 37°C. Values are from 2 experiments with cells in passages 5 - 9 , and based on 1000 cells per experiment. The two values are shown to indicate the range over which the values fell. Cells at 37°C were generally similar to the 39°C phenotypic characteristics.  63  IV. E. Post Crisis Line  One line (160c) survived crisis at 34°C and resumed proliferation after three months of quiescence. The line was passaged until it had undergone a total of 200 population doublings post infection. At that time, the experiment to quantitate the lifespan of the post crisis line was terminated since it fulfilled the definitional requirements of a continuous line (Schaeffer, 1984). The cell line was named PC 160c for convenience. Several vials of the post crisis line were frozen and stored for future use.  The cell line expressed all the immortalization,  differentiation and senescence markers. Keratin, collagen and CA125 levels did not change upon shift to the nonpermissive temperature for 72h, and were consistently around 74%, 29% and 1.4% respectively. SA P-gal expression increased from 0 % in the immortalized state to 100 % in the nonimmortalized state. The line was now presumably truly immortal at the permissive temperature, and the growth pattern resembled transformed lines.  The cells were small,  compact, and were not contact inhibited. They pulled together and formed clusters of cells when they neared confluence. The cells were tested for tumorigenicity. Eight scid mice were injected with 10 and 10 cells that had passed crisis on 13 6  7  September 1996, and to date, no subcutaneous tumors have arisen. Animals injected intraperitoneally still look normal and healthy. I will continue to observe  64  the mice for 6 months, and if no tumors have arisen by that time, the experiment will be terminated. Should tumors arise, they then will be 1) transplanted to new mice, 2) cultured, 3) fixed for histologic examination.  65  34°C 39°C  Keratin  Collagen  74.7 ±3.0 73.3 ±4.7  26.7 ±3.2 29.2 ±2.5  CA125 1.1 ±0.1  1.2 ±0.1  SA p-gal  0.0 ±0.0 100.0 ±0.0  Table 4. Percent of cells expressing differentiation and senescence markers by the post crisis line PC 160c. Values are from 3 experiments with cells in passages 16-21 and based on counts of 1000 cells per experiment. Differentiation markers collagen and CA125, and senescence marker SA P-gal were similar to the parent line, 160c. Keratin, the other differentiation marker tested, was higher than the parent line at 34°C, and the same as the parent line at 39°C.  66  V . Discussion  Four cases of hOSE were successfully infected with a tsTag A209 construct which produced clonal cell lines that had extended proliferative capacities between 52 - 71 population doublings (pds) over the normal lifespan of hOSE, which is between 15 - 23 pds. All clones expressed SA P-galactosidase only after shifting to the nonpermissive temperature. Two of four clones modulatedfromlow to high expression of keratin and collagen, suggesting that active Tag does affect the expression in some hOSE, but the mechanism by which this happens is not known. The tsTag system maintained differentiated characteristics keratin, collagen and CA125 at levels similar to normal hOSE. One of four cases survived crisis an continued to replicate up to 150 population doublings past the expected lifespan of hOSE. This post crisis line maintained keratin at both temperatures despite the fact that its parent line was one that modulated keratin levels according to temperature shifts. This line has not formed subcutaneous or intraperitoneal tumors after four months in scid mice. Several cases of hOSE were obtained at surgery through laparoscopy with informed consent from women with no indications of gynecological cancer. Laparoscopy specimens generally contained hOSE scrapings and in some cases blood.  Laparoscopy specimens are cleaner than biopsy specimens when  processing, and were preferable since one could be more sure of infecting only  67  hOSE. This avoids the possibility of infecting and immortalizing the underlying stroma or fibroblasts of the ovaries. Biopsy specimens contain red blood cells, and sometimes fibroblast and stroma as well as hOSE, all cells that may be dislodged by manipulation. If we used these types of specimens, they would have grown more like the in vivo situation since growth factors and stimuli supplied by the stroma and fibroblasts may be present in the transport medium and initial stages of in vitro growth. Viral stock and cell lines necessary to produce virus had to be proliferated and virustitratedbefore infection of hOSE. The protocol for producing the virus quoted an expected harvest, and it was not necessary to perform a plaque assay. However, the virus has not been used on hOSE before, so I performed a TCLD  50  (Tissue Culture Dose Infected 50%). This value expresses the dilution at which 50% of the cells will be infected and die. This value for my preparation was 10" . 9  This suggests that many virus particles were present in my preparation, and although I do not have a precise value, I do know that if I infect hOSE at 10" , 9  some but not all of the culture will be infected with the virus, which could suggest that only one copy of the virus would enter each cell. This is important since the effect of having multiple copies of the virus active in human cells, (ie. having multiple doses of Tag) has not been characterized. Single integrations of virus, especially of adenovirus-SV40 tsTag, have been better characterized in other cell types, for example human placental cells (Lei et al., 1992), urothelial cells  68  (Masters and Petzoldt, 1993), fibroblasts (Ferber et al., 1993), endometrial stromal cells (Reinhart et al., 1993) and adult CNS cells (Eves et al., 1994). Interestingly, the success of viral integration into cells in vitro is not the same as the in vivo situation.  Gene therapy using adenovirus vectors to  permanently introduce genes into human cells in vivo has been studied and attempted (reviewed in Hess,. 1996), but has been unsuccessful. The problem is that replication defective adenovirus-construct generally maintains episomes in the nucleus of the cells that it infects, and does not integrate into the genome (Leber et al., 1996). This allows the infected cells to produce the gene products encoded on the construct for a limited period of time, but as cells proliferate, the episomal adenovirus construct is diluted and/or is lost. In vitro, however, a hot spot for viral integration using replication defective adenovirus with tsTag's is at 1 p 36 of human chromosomes (Romani et al., 1990; Romani et al., 1993). It is not known how the components of the Tag assist in the integration of the viral particle, but it is speculated that it must be involved since adenovirus constructs without the Tag do not integrate into the genome of its host. Several factors obviated the need for formal selection with antibiotics such as G418 with hOSE. hOSE do not grow well in sparse situations whereas immortal cells tend to grow in sparse situations, even if more slowly than under less sparse conditions.  Growth in lower serum is another characteristic of  immortalized cells. hOSE need higher concentrations of serum to grow, whereas 69  tsIOSE do not, and this is a selection in itself. Another selection method would only complicate the procedure. G418, like any other toxin, may also have an effect on the immortal cells that has not been documented. The consistent three weeks that elapsed post infection before clones appeared was striking. The tsTag construct is first located in episomes at the edge of the nucleus, as with any newly introduced DNA construct. Expression of adenovirally introduced DNA is quite high, as demonstrated by expression of the lac Z gene product in virtually 100% of the population infected with the adenovirus-lac Z construct within 72 h (Leung et al., in press). Therefore, the tsTag must have been expressed since the infection. If the cells had been dividing right from the start of incubation, one would have seen foci form much faster than at 3 weeks. One possible explanation for the lag between expression of tsTag and immortal phenotype could be that the cells were sparse and were not very metabolically active. The tsTag took some time to accumulate, and its effects could not be seen right away since there was not enough enzyme for the system to be active. That is, the tsTag enzyme could not inactivate p53, pRb, and other cell cycle regulators faster than the cell could replace them. When enough tsTag accumulated, the reverse would be true, and the cell cycle regulators would be inactivated faster than the cell could produce them, therefore the cell became immortal. Not all cells would continue on this pathway, however, since the  70  episomal DNA would be lost. Only the cells with integrated adenoviral constructs would continue to proliferate. Irnmortalized cells tend to have a distinct morphology and growth characteristics when compared to nonimmortalized cells. This feature is also used in determining which cell colonies are to be considered clones and which are not. Immortal clones are small, compact and are not contact inhibited, and also grow rapidly in low serum. Nonimmortal cells growing in patches are epithelial, sometimes compact, but contact inhibited and grow slowly in low serum. After observing the cells for a few days in culture gives one an indication of which type of colonies one is observing. hOSE only grow a limited number of population doublings, and after 4 passages, most of the nonimmortalized cells are expected to be senescent. These characteristics were similar to those found when isolating uroepithelial cells (Masters and Petzoldt, 1993) or endometrial stromal cells (Reinhart et al., 1993). The initial investigations involved splitting cells from 34°C and incubating at 34°C and 39°C. Early passages demonstrated certain characteristics. At 34°C, they continued to grow rapidly and to confluence while maintaining a small, compact, epithelial phenotype. At 39°C, early passages of cells slow down almost immediately and lost their small, compact, epithelial phenotype over two weeks. By the end of two weeks, the majority of cells had died and lifted off the plate.  71  Those that remained were vacuolated, very thin and flat, and some were multinucleate. This is expected since cells losing their immortal phenotype will revert back to their pre-immortalization state, including their morphology and often their function. Quantitative assessment is difficult to perform without cell dimension measuring equipment in the lab, but qualitative assessments from pictures show the result. This change in morphology happened over two weeks, whereas the same general pattern follows in normal hOSE in culturefromprimary culture to senescence. Interestingly, if cells are passaged at 34°C until passage 9 or 10, then shifted 39°C, a similar loss of the compact epithelial phenotype occurs, but over a shortertimeperiod of 1 to 2 days. These results suggest that some form of counting mechanism is present in the cells, and is active while the cell is in the immortal state. tsIOSE mimics the old and senescing hOSE in culture at the nonpermissive temperature.  If the  immortalization agent is gone, cells revert to the state at which they should be had they not been immortalized, suggesting that the mechanism for counting and measuring how many population doublings the cells have undergone continues to work while the cells are immortalized, even though the counter can not do anything to stop the growth. After the immortalization is inactivated, the counter still has counted all the population doublings that the cell had undergone while in its immortal state. The cell may recognize this and express characteristics that would/should be presented at that physiological age of the cell. Several potential  72  mechanisms for determining how long a cell lives have been studied. It was first discovered that certain profiles of triton-extracted enzymes suggested that a population of cells was senescing. Individual cells could be isolated by flow sorting and the enzyme activity determined by ultramicrochemistry (Jongkind and Verkerk, 1989). Cellular aging was linked to cellular dysfunctions, but the mechanism was not clear. Cellular aging was also seen in cells treated with micrombule-mteracting drugs. Microtubules are ubiquitous cellular components involved in control of cell structure and functions, and it seemed logical for them to be affected as cells age (Reviewed in Raes, 1991). Mutations in mitochondrial DNA were found in cells from patients with aging and degenerative diseases such as Alzheimer's disease and Werner syndrome as well as in elderly patients (reviewed in Ozawa, 1995 and Grossman, 1995). Mitochondrial DNA that encodes for protein subunits necessary for the maintenance of mitochondrial ATP synthesis acquires mutations at a rate faster than nuclear DNA, due to oxygen damage (Wallace, 1992). GTP-binding proteins were found to befrequentlyinvolved in determining the longevity of yeast Saccharomyces cerevisiae (Jazwinski, 1993), and cGMP inhibitors were found to extend the lifespan of certain cells (Fujii et al., 1995). Telomere length seems to correlate with the expected lifespan of a cell: Short telomeres suggest little lifespan left, and long telomeres suggest long lifespan left (Vazin et al., 1993; Hastie et al., 1990). Immortal cells seem also to 73  have gained the activity of telomerase, and the ability to synthesize telomeric DNA onto chromosomal ends using a segment of its tRNA as a template (Greider and Blackburn, 1989; Yu et al., 1990). This is obviously not what happens in these cells since it has been found that telomerase activity is not increased in SV40 immortalized lines (Kim et al., 1994). Immortality may also be due to expression of recessive changes in normal growth-regulatory genes of the cell (Smith and Pereira-Smith, 1988). Immortal cells have been assigned to four different complementation groups by cell fusion analysis, each group corresponding to four genes or processes involved in senescence (Smith and Pereira-Smith, 1989a). Alterations in gene expression accompanying senescence cause cells to express proteins that first inhibit DNA synthesis, second induce new cell surface epitopes and third induce changes in the extracellular matrix. These changes would likely include the production of SA Pgalactosidase, a marker that is produced by senescent cells (Smith and PereiraSmith, 1989b). Although no function has been assigned to this enzyme, it does not mean that it does not exist. Interestingly, it was found that all human SV40 transformed lines studied assigned to the same group, suggesting that this virus immortalizes various human cells by the same mechanism (Smith and PereiraSmith, 1988). In this system, the most likely explanation of the rapid aging that occurs post shift to 39°C is the re-expression of the dominant senescence genes as 74  described by Smith's complementation groups. Tag could interfere with the expression of the genes that allow cells to senesce, but when it is inactivated, tsTag can no longer function in that respect, so the dominant senescence genes take over, and the cells senesce. Mitochondrial mutations and microtubule alterations are possible since the cells are growing rapidly and they may accumulate mutations, but cells at both temperatures need properly working mitochondria and microtubules, so it is not likely that Tag interferes with either of these mechanisms. It is also not likely to be a cGMP related mechanism since experiments with H89, a cGMP inhibitor, did not do anything to extend the lifespan of tsIOSE at the nonpermissive temperature. Tag phosphorylates and interferes with cell cycle regulators in the cells, that are free in the nucleus. Unpublished preliminary studies suggest that SA P-gal measurement is a method that has the potential to help estimate the proliferative potential in OSE and is a good marker to differentiate senescent and proliferating tsIOSE. The absence of SA p-gal in cells that are expressing the inmiortalizing agent is expected (Dimri et al., 1995) since cancerous cells do not express SA P-gal and proliferating immortalized cells only express SA p-gal just prior to senescing. Expression of SA P-gal in cells where Tag is inactivated was also expected since if the cells are no longer immortal, they should express the senescence marker that is present in normal aging hOSE.  75  SA p-gal is a characteristic that consistently and radically changes once the tsTag is inactivated by a temperature shift, and can be seen within 6 hours of that shift. At the permissive temperature, the Tag is stable and acts to allow cell to continue to proliferate. At this time, no SA P-gal can be detected, and this is like normally immortalized and cancerous cells. Expression of SA P-gal is either verylow or nonexistent. Once the tsTag is inactivated by a temperature shift, SA P-gal begins to express almost instantaneously at very high levels which compares to senescent and dying cells. Morphologically, the cells begin to change shape from their immortalized phenotype (small, compact and epithelial) to a phenotype typical of old and dying cells (large, flat and epithehal). This change is significant because it is so consistent and the changes are so striking. The concomitant expression of a morphological change which shows senescence and a senescence marker suggests that once tsTag is inactivated, something that measured senescence is activated or allowed to express, and this marks a point after which cells can no longer be reverted back to men immortalized state. Tag was not detected at 39°C by immunofluorescence, suggesting that the epitope that the antigen recognizes is affected by the conformational change that occurs when the tsTag protein is shiftedfromthe permissive to the nonpermissive temperature (Wei et al., 1994). Southern blot showed the transcript present at both temperatures, suggesting that the protein was unable to work. The amount of DNA integrated was similar in all the cases based on the density of the 76  autoradiograph band and showed no difference between 34 and 39°C for each line tested. This suggests that integration by the adenoviral mechanism is consistent and stable. This compares to published results where a hot spot for adenoviral type 5 integration has been isolated to chromosome lp36 in an Alu rich sequence (Romani et al., 1990; Romani et al., 1993). If this is consistent, it may explain why adenovirus mediated introduction of genes into hOSE is consistent and reproducible. That is not to say that integration does not happen elsewhere, but only suggests that a likely spot for integration is.at chromosome 1. Fluorescence in situ hybridization (FISH) can be performed to conclusively prove that the integration happened in one location or another. Introduction of chromosome 1 into other cancer cells (HepG2, CMV MJ cells) has prevented the immortal phenotype from expressing, allowing the senescence of cells (Hensler et al., 1994). This theory correlates with Smith's complementation group theory that an expressed dominant gene could cause senescence if the gene is expressed in an immortal cell. One way to test if the tsTag is active in the cell is to measure the cell components that tsTag is known to affect. p53 is a cell cycle regulator that is inactivated by Tag. If p53 is not active in the cell, the cell will produce more, resulting in high levels of p53. p53 levels in tsIOSE were examined at the permissive and nonpermissive temperatures. The ts system produces the expected results of p53 levels.  When tsTag is active, large amounts of p53 were  77  accumulated. When the active tsTag was lost, p53 levels slowly diminished. The half life of p53 in uninfected cells is on the order of 20 minutes (Blagosklonny, 1994), so even a high level of p53 should be dissipated quite rapidly. In the presence of Tag, p53 is more heavily phosphorylated than in the absence of Tag (Sturzbecher et al., 1987; Scheidtmann and Haber, 1990; Patschinsky et al., 1992), and is stabilized post-translationally (Reich et al., 1983), such that its half-life is on the order of several hours (Reihsaus et al., 1990). This suggests that the Tagphosphorylated p53 is more stable than cellular-phosphorylated p53, and the p53 levels that are high post temperature shift to 39°C are due to the slow breakdown of Tag-phosphorylated p53. Two of four lines tested showed a modulation from low to high keratin and collagen upon shifting temperatures. The shift from less to more differentiated phenotypic characteristics due to the inactivation of an immortalizing agent is not unexpected. High keratin in all four lines at 39°C is a strong indicator of epithelial characteristics, and indicates that the cells are expressing epithelial differentiation at the nonpermissive temperature. Although the collagen levels are within the normal range expressed by hOSE in vitro, a shift to a higher percentage of collagen positive cells shows a shift to more mesenchymal properties. It shows that the Tag inhibits or disturbs a certain pathway in hOSE, and upon its inactivation, the inhibition is lost. Without further study here, it is difficult to say whether the mechanism is either a binding phenomenon, or signal transduction  78  interference. What can be said is that it is not a permanent change in the genome since it is an effect that is easily and consistently reversible. A result that was not expected was the higher collagen that these slutting cells acquire. One theory that has been presented in the past is that cells maintain the physiological state at which they were immortalized. If the cells were expressing high collagen, perhaps due to their response to culture (as if wounded) or due to epithelial-mesenchymal conversion, then they acquire this trait when Tag is inactivated. This loss of a phenotypic characteristic due to the immortalization and the reappearance of that phenotypic characteristic due to a change in temperature that inactivates the immortalization agent suggests that a pathway may be inactivated by the tsTag. Other cell types that have been infected with this or a similar virus have also shown changes in phenotypic characteristics that shift with temperature. Some of these include neuronal differentiation of human medullary raphe (Whittemore and White, 1993) and human lens epithelium (Andley et al., 1994). Two of the cell lines maintained high keratin, low collagen and all cell lines maintained low CA125 without modulation due to temperature slnfting. The maintenance of certain differentiated characteristics is similar to immortalization of hOSE with wild type Tag (Maines-Bandiera et al., 1992). CA125 is lost when inmiortalizing with wild type Tag, but is maintained using this immortalization vector. All four of the clones tested maintained a low level of CA125 at both the permissive and nonpermissive temperature. This is significant since it shows a  79  maintenance of a highly differentiated phenotype from hOSE that is lost in transfection with wild type Tag. It is exciting to see a difference between the characteristics maintained by different forms of the same immortalization agent. This suggests that different mechanisms must be involved in each of those mechanisms and that the tsTag is more likely to preserve differentiated phenotypic characteristics. Adenovirus - mediated immortalization has been shown in the past to be a successful means of immortalizing human cells, and has been accomplished in many cell types including lens (Andley et al., 1994), neurons (Whittemore and White, 1993), and stromal cells (Tsao et al., 1994). The one interesting factor among all these is that many of the cell types' differentiated characteristics were maintained through immortalization. This is in comparison to both tsIOSE and other immortalized cell types that lost their differentiated functions at the permissive temperature and regained it at the nonpermissive temperature. All four cell lines, maintained at 34°C, entered crisis at a pd (110 - 130) where the number of population doublings far exceeded the normal lifespan of normal hOSE (20 pd). One of these cell lines when left in the incubator survived crisis and continued to proliferate after a quiescent period of 4 months. Many cells immortalized with Tag do not survive crisis, but a small proportion do (Lane and Benchimol, 1990). Once cells propagate a total of 150 population doublings more than their nonimmortalized counterpart and have passed crisis, they are 80  termed continuous lines (Schaeffer, 1984). This post crisis line has fulfilled the condition, and has been designated PC 160. Other human cell types that have attained this status include trophoblast (Chou, 1976) and endometrial stromal cells (Reinhart et al., 1993). Line PC 160 did not however modulate the characteristics tested, and maintains at least one hOSE specific phenotype of CA125. This would permit a wide range of experiments to be performed with hOSE since one of the limiting factors or experiments with hOSE is the small number of cells we get per specimen. A large number of cells is required for DNA, RNA and protein extractions, optimization of experimental parameters for transfections etc. This cell type would provide a consistent source of cells that are phenotypically similar to hOSE, and is better than pre-crisis lines since the characteristics are similar to hOSE through all the temperatures. In addition, the cell line will presumably not die out since it is truly immortal. One drawback is that the longer one keeps the cells, the more mutations may accumulate in the cells, and they may eventually deviate from normal hOSE phenotypic characteristics. Some human cells that survive crisis become tumorigenic (Lane and Benchimol, 1990). It is believed that the mechanism that causes cells to survive crisis also plays a part in tumorigenicity. This next step was tested in scid mice both subcutaneously and intraperitoneally. The lack of tumor growth at this point (5 months) suggests that the cells are not highly tumorigenic.  If highly  tumorigenic, the cells would have proliferated sufficiently before senescing to  81  causeftrmors.However, this does not mean that tumors will not eventually form. The tumors may just take longer to begin to proliferate. It is not likely that tumor formation in the peritoneal cavity will occur. The temperature sensitive nature of the cells will cause the cells to die rapidly in the internal cavity of the mice which have a body temperature of 39°C. Subcutaneous tumors may form since the surface temperature of the mouse is lower, and closer to the room temperature at which they are housed. This development is slow, if tumors are growing. There were no differences between the familial clone (178x) and the other clones in growth rates, expression of SA-(3-galactosidase or CA125 levels. Morphologically, the 178x at the permissive temperature were more atypical and fibroblastic like, and also had several whorly patches that were much more atypical than any of the other clones isolated. Although phenotypic characteristics of cells derived from specimens from patients with familial histories of ovarian cancer do seem to differ when them comparing normal cells (Dyck et al., 1996), the differences between the polyclonal line and other lines seem small. The levels of keratin and collagen did not modulate with temperatures. This is different from the other polyclonal line since that clone modulated both keratin and collagen to some degree. The immortalization, differentiation, and senescence  characteristics  changed asynchronously after a temperature shift from the permissive to the  82  nonpermissive temperature. Tag was lost by 24h, yet it took p53 at least 2 weeks to return to baseline levels. Senescence associated P-galactosidase began to return in 6 h, and was completely positive after 24h whereas senescence associated morphology did not return until after 24h. Keratin and collagen HI levels that were lost or affected by tsTag began to change after 24h. CA125 levels did not seem to be affected by temperature shifts. The different times that it took for the tested characteristics suggest several things. Senescence associated P-gal, keratin, and collagen are affected by the expression of Tag, and begin to modulate within several hours of Tag being inactivated, and are independent of p53 expression. CA125 levels are not affected by either p53 or Tag expression. P53 levels are high when Tag is active, but begin to decrease when Tag is inactivated. In summary, some but not all characteristics of hOSE are maintained at the permissive temperature, and some of the clones have the ability to change from low to high expression of some of these characteristics by inactivation of tsTag. It is possible to obtain a large number of cellsfromone case by irnmortalizing it with tsTag, but the results may not be uniform since the clones differ and may mutate during replication. The system provides maintenance of a hOSE characteristic like CA125 that is lost with other methods of immortalization. Immortalization with tsTag A209 is ultimately superior to immortalization of hOSE with wild type Tag, and is a good method to produce a large number of immortalized hOSE for large scale production.  83  VI. References  Abraham SE, Lobo S, Yaciuk P, Wang HH and E Moran (1993) p300 and p300 associated proteins are components of TATA binding protein complexes. Oncogene 8:1639-1647. Adams AT and N Auersperg (1980) Transformation of rat ovarian surface epithelial (ROSE) cells by Kirsten murine sarcoma virus (Ki-MSV) in vitro. Proc Am Assn Cancer Res.21,92 - 99. Adams AT and Auersperg N (1981) Transformation of cultured rat ovarian surface epithelial cells by Kirsten murine sarcoma virus. Can. Res. 41, 2063-2072. Andley UP, Rhim JS, Chylack LT Jr, and TP Fleniing(1994) Propagation and immortalization of human lens epithelial cells in culture. Investigative Ophthalmology and visual Science. 35: 3094 - 3102. Auersperg N, Maines-Bandiera S, Booth JH, Lynch HT, Godwin AK and TC Hamilton (1995) Expression of two mucin antigens in cultured human ovarian surface epithehum: influence of a family history of ovarian cancer. Am J Obs Gyn 173, 558-65. Auersperg N, Maines-Bandiera S, and PA Kruk In: Ovarian Cancer 3, Biology, Diagnosis and Management, F. Sharp and WP Mason, eds., Chapman and Hall, pp.97-108, 1994. Auersperg N, Maines-Bandiera SL, Dyck HG, and PA Kruk (1994) Characterization of cultured human ovarian surface epithelial cells: phenotypic plasticity and premalignant changes. Lab. Invest. 71:510-518. Baker SJ, Maarkowitz S, Fearon ER, Willson JK and B Volgstein (1990) Suppression of human colorectal carcinoma cell growth by wild type p53. Science 249: 912 - 915. Bast RC, Clug TL, and E St. John (1983). A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian carcinoma. N Engl J Med 309:169- 171.  Berek JS, Knapp RC, Malkasian GD, Lavin PT, Whitney C, Niloff JM, and RC Bast Jr. (1986) CA125 serum levels correlated with a second look operations among ovarian cancer patients. Obstet Gynecol 67:685-689. Cherrington V, Brown M, Paucha E, Louis JS, Speigelman BM and TM Roberts. (1988) Mol Cell Biol 8: 1380 - 1384. Chou JY and JM Martin (1974) Complementation analysis of SV40 mutants. J Vir 13(5): 1101-1109. Chou JY (1978) Human placental cells transformed by tsA mutants of simian virus 40: A model system for the study of placental functions. Proc Natl Acad Sci USA 75:1409-1413. Chou JY (1989) Differentiated mammalian cell lines immortalized by temperature sensitive tumor viruses. Mol Endo 3:1511-1514. Chow LT, Roberts JM, Lewis JB, and TR Broker (1977) A map of cytoplasmic RNA transcripts from lytic adenovirus type 2, determined by electron microscopy of RNA:DNA hybrids. Cell, 11:819-836. Clayton CE, Murphy D, Lovett M and PWJ Rigby (1982) A fragment of the SV40 large T antigen gene transforms. Nature 299, 59-61. Clement P.B. (1987) Histology of the ovary. Am J Surg Path 11, 277-303. Colby WW and T Shenk (1982) Fragments of the simian virus 40 transforming genes facilitate transformation of rat embryo cells. Proc Natl Acad Sci USA 79, 189 - 193. Dales S (1962) An electron microscope study of the early association between the mammalian viruses and their hosts. J Cell Biol 13: 303-322. Diller L, Kassel J, Nelson CE, Gryka MA, Litwak G, Gebhardt M, Bressac B, Ozturk M, Baker SJ, Vogelstein B and S Friend (1990) p53 functions as a cell cycle control protein in osteosarcomas. Mol Cell Biol 10: 5772-5821. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, Medrano EE, Linskens M, Rubel I, Pereira-Smith O, Peacocke M and J Campisi (1995) A  €5"  biomarker that identifies senescent human cells in culture and in aging skin in vivo. Proc Natl Acad Sci USA. 92;9363 - 9367. DiPaolo JA, Popescu NC, Alvarez L and CD Woodworm (1993) Cellular and molecular alterations in human epithelial cells transformed by recombinant human papilloma virus DNA. Crit Rev Oncog 4(4):337-360. Dyck HG, Hamilton TC, Godwin AK, Lynch HT, Maines-Bandiera S, and N Auersperg (1996) Autonomy of the epithelial phenotype in human ovarian surface epitheUum changes with neoplastic progression and with a family history of ovarian cancer. Int J Can 69,429-436. Dyck HG. and N Auersperg (1994) The expression of mesenchymal characteristics by human ovarian surface epithelial cells diminishes with neoplastic progression. Proc. AACR 85 annual meeting. 35:162. th  Einhorn N, Sjoval K, and G Schoenfeld (1990) Early detection of ovarian cancer using the CA125 radioimmunoassay. Am Soc Clin Oncol; 9:157 - 161. Elliott WM. and N Auersperg (1993) Growth of normal human ovarian surface epithelial cells in reduced-serum and serum-lree media. In vitro Cell Dev Biol 29A(1):9-18. Eves EM, Kwon J, Downen M, Tucker MS, Wainer BH, and MR Rosner (1994) Conditional immortalization of neuronal cells from postmitotic cultures and adult CNS. Brain Research. 656(2):396-404. Ewen ME, Faha B, Harlowe E and DM Livingston (1992) Interaction of pi07 with cyclin A independent of complex formation with viral oncoproteins. Science 255:85-87. Fanning E and R Knippers. (1992) Structure and function of simian virus 40 large tumor antigen. Annu. Rev. Biochem 61: 55 - 85. Ferber A, Chang C, Sell C, Ptasznik A, Cristofalo VJ, Hubbard K, Ozer HL, Adamo M, Roberts CT Jr, and D LeRoith (1993) Failure of senescent human fibroblasts to express the insulin-like growth factor-1 gene. Journal of Biological Chemistry. 268(24): 17883-81788. Feuerbach FJ. Crystal RG. (1996) Progress in human gene therapy. [Review] Kidney International. 49(6): 1791-1794  84  Fromm M and P Berg. (1982) Deletion mapping of DNA regions required for SV40 early region promoter function in vivo. J Mol App Gen 1: 457-481. Fujii M, Ide T, Wadhwa R, Tahara H, Kaul SC, Mitsui Y, Ogata T„ Oishi M, Ayusawa D (1995) Inhibitors of cGMP-dependent protein kinase block senescence induced by inactivation of T antigen in SV40-transformed immortal human fibroblasts. Oncogene. 11:627-634. Gannon JV and Lane DP (1987) p53 and DNA polymerase alpha compete for binding to SV40 large T antigen. Nature 329:456-458. Gannon JV and Lane DP (1990) Interactions between SV40 large T antigen and DNA polymerase a. New Biologist:82-93. Gloeckler L.A. (1993) Ovarian Cancer. Cancer 71:524. Graham FL. Smiley J. Russell WC. Nairn R. (1977) Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J Gen Vir 36:59-74. Grossman LI. (1995) Mitochondrial mutations and human disease. [Review] Environmental and Molecular Mutagenesis. 25 Suppl 26:30 - 37. Hafez ES and S Makabe (1982) Scanning electron microscopy of nonfunctional human ovaries J Reprod Med 27: 271-274. Hamilton TC, Henderson WJ, Eaton C In: Tissue Culture in Medical Res. (II). Proceedings of 2 Intemad. Symp. Richards, R.J., and Rajan, K.T., eds, Pergamon Press, New York, 1980. nd  Hankinson SE, Colditz GA, Hunter DJ, Willett WC, Stampfer MJ, Rosner B, Hennekens CH and FE Speizer. (1995) A prospective study of reproductive factors and risk of epithelial ovarian cancer. Cancer 76: 284-290. Hargest R. Williamson R. (1996) Prophylactic gene therapy for cancer. [Review] Gene Therapy. 3(2):97-102. Hayflick L (1965) Cellular Ageing. Exp. Cell Res. 37, 61636. Hensler PJ, Annab LA, Barrett JC and Pereira-Smith OM. A gene involved in control of human cellular senescence on human chromosome lq. Molecular and Cellular Biology. 14: 2291-7, 1994r. £7  Hess P. (1996) Gene therapy: a brief review. [Review] Clinics in Laboratory Medicine. 16(1): 197-211. Hoffman AG. BurghardtRC. TilleyR. Auersperg N. (1993) An in vitro model of ovarian epithelial carcinogenesis: changes in cell-cell communication and adhesion occurring during neoplastic progression. Int J Can. 54(5):828-38. Hollstein M, Sidransky D, Vogelstein B and CC Harris (1991) p53 mutations in human cancers [Review]. Science 253: 49-53. Howe JA, Mymryk JS, Egan C, Branton PE and ST Bayley (1990) Retinoblastoma growth suppresser and a 300 kDa protein appear to regulate cellular DNA synthesis. Proc Natl Acad Sci 87, 5883-5887. Hu Q, Lees JA, Buchkovich KJ and E Harlow (1992) Mol Cell Biol 12: 971 - 980. Jazwinski SM. (1993) The genetics of aging in the yeast Saccharomyces cerevisiae. [Review] Genetica. 91:35-51. Johnson P, Gray D, Mowat M and S Benchimol (1991) Expression of wild type p53 is not compatible with continued growth of p53 negative tumor cells. Mol Cell Biol 11, 1-11. Jongkind JF. Verkerk A. (1984) Cell sorting and microchemistry of cultured human fibroblasts: applications in genetics and aging research. Cytometry. 5:1827. Kalderon D, Richardson WD, Markham AF and AE Smith (1984) Sequence requirements for nuclear localization of simian virus 40 large T antigen. Nature, 311:33-38. Kim NW, Piatyszek MA, Prowse KR, Harley CB, West WE, Weinrich SL, Shay JW (1994) Specific Association of human telomerase activity with immortal cells and cancer Science 266:2011 - 2015. Kruk PA Maines-Bandiera SL Auersperg N (1990) A simplified method to culture human ovarian surface epithelium. Laboratory Investigation. 63:132-136. Kruk PA, and Auersperg N (1992) Human ovarian epithelial cells are capable of physically restmcturing extracellular matrix. Am J Obstet. Cynecol. 167:14371443. 88  Kruk PA. and Auersperg N. (1991) Percoll centrifugation eliminates mold contaminants from cell cultures. In vitro Cellular and Developmental Biology. 27A:273-6. Lane DP and S Benchimol (1990) p53: Oncogene or anti-oncogene? Genes Dev 4: 1-8. Laver WG, Suriano JR, Green M. (1967) Adenovirus proteins. LT. N-terrninal amino acid analysis. J Virol 1:723 - 728. Leung EH, Chen GTC, Davies BR, Roskelley CD, Leung PCK and N Auersperg (1997) Efficient introduction of genes into human ovarian surface epithelium, In vitro Cellular and Developmental Biology, in press. Lei KJ. Gluzman Y. Pan CJ. Chou JY. (1992) ImmortaUzation of virus-free human placental cells that express tissue-specific functions. Molecular Endocrinology. 6(5):703-12. Ludlow JW, DeCaprio JA, Huang CM, Lee WH, Paucha E and DM Livingston (1989) SV40 large T antigen binds preferentially to an underphosphorylated member of the retinoblastoma susceptibility gene product family. Cell 56: 57-65. Lynch HT. Watson P. Conway TA. Lynch JF. Slominski-Caster SM. Narod SA. Feunteun J. Lenoir G. (1993) DNA screening for breast/ovarian cancer susceptibility based on linked markers. A family study. Arch Int Med 153, 197987. Maines-Bandiera SL, Kruk PA, and Auersperg N (1992) Simian virus 40transformed human ovarian surface epithelial cells escape normal growth controls but retain morphogenic responses to extracellular matrix. Am J Obstet Gyn, 167:729-735. Maizel JV, White DO, Scharff MD (1968) The polypeptides of adenovirus. I. Evidence for multiple protein compartments in the virion and a comparison of types 2, 7 and 12. Virology 36; 115 - 125. Manfredi JJ and C Prives. (1994) The transforming activity of simian virus 40 Tumor antigen. Bioc Biophys Acta 1198: 65 - 83. Marshall CJ (1991) Tumor suppresser genes [Review]Cell 64: 313-326. 89  Marsillio E, Cheng SH, SchafFhausen B, Paucha E and DM Livingston (1991) The T/T common region of simian virus 40 large T antigen contains a distinct transformation governing sequence. J Virol 52: 884 - 891. Masters JR Petzoldt JL (1993) In vitro studies on the pathogenesis of bladder cancer. Verhandlungen der Deutschen Gesellschaft fur Pathologie. 77:157-160. Merino M.J. and Jaffe G. (1993) Age contrast in ovarian pathology. [Review]. Cancer 71, 537-544. Montano X, Millikan R, Newsome JM, Newsome DA, Ludlow K, Arthur AK, Fanning E, Bikel I and DM Livingston. (1990) Simian virus 40 small tumor antigen and an amino terminal domain of large tumor antigen share a common transfonning function. Proc Natl Acad Sci, USA. 87: 7448 - 7452. Montenarh M, Kohler M, Aggeler G and R Henning (1985) Structural prerequisites of simian virus 40 large T antigen for the maintenance of cell transformation. EMBO J. 4, 2941 - 2947. Myrdal SE. and Auersperg N.(1986) An agent or agents produced by virustransformed cells cause unregulated ruffling in untransformed cells. J Cell Biol 102(4): 1224-9. Nicosia SV. In: Aging, Reproduction and the Climacteric. Mastrojanni, L-, Jr., and Paulsen, CA. a's., Plenum Press, New York, pp. 179-199, 1986. Nicosia, S.V., Nicosia, R.F. Path, of Human Neoplasms. HA Azar, Ed. Raven Press, New York, pp.435486; 1988. Nicosia, SV. (1986) In: Aging, Reproduction and the Climacteric. Mastrojanni L Jr., and Paulsen, C A . a's., Plenum Press, New York, pp. 179-199. Owens OJ and Leake RE (1992) Growth factor content in normal and benign ovarian tumours. European J. Obs., Gyn., and Repro Biol 47: 223-228. Ozawa T. (1995) Mitochondrial DNA mutations associated with aging and degenerative diseases. [Review] Experimental Gerontology. 30:269-90. Patschinsky T, Knippschild U and Deppert W. (1992) Species-specific phosphorylation of mouse and rat p53 in simian virus 40-transformed cells. J Virol 66,3846-3859.  90  Pipas JM, Penden KWC, and D Nathans (1983) Mutational analysis of simian virus 40 large T antigen: isolation and characterization with deletions in the T antigen. Mol Cell Biol 3: 203-213. Piver MS, Baker TR, Jishi MF, Sandecki AM, Tsukada Y, Natarajan N, Mettlin CJ and CA Blake. (1993) Familial ovarian cancer. A report of 658 families from the Gilna Radner Familial Ovarian Cancer Registry 1981 - 1991. Cancer 71:582. Radna RL, Caton Y, Jha KK, Kaplan P, Li G, Traganos F, Ozer HL (1989) Growth of immortal simian virus 40 tsA-transformed human fibroblast is temperature dependent. Mol Cell Biol 8:3093-3096. Raes M. (1991) Involvement of microtubules in modifications associated with cellular aging. [Review] Mutation Research. 256:149-68. Reich NC, Oren M, Levine AJ (1983) Two distinct mechanisms regulate the levels of a cellular tumor antigen, p53 .Mol Cell Biol 3, 2143 - 2150. Reihsaus E, Kohler M, Kraiss S, Oren M, Montenarh M (1990) Regulation of the level of the oncoprotein p53 in non-transformed and transformed cells. Oncogene 5, 137 - 145. Rinehart CA, Laundon CH, Mayben JP, Lyn-Cook BD, Kaufman DG. Conditional immortalization of human endometrial stromal cells with a temperature-sensitive simian virus 40. Carcinogenesis. 14(5):993-9, 1993. Romani M, Casciano I, Querzola F, De Ambrosis A and Siniscalco M. Analysis of a viral integration event in a CG-rich region at the lp36 human chromosomal site. Gene. 135:153-60,1993. Romani M, De Ambrosis A, AlhadeffB, Purrello M, Gluzman Y, Siniscalco M, Preferential integration of the Ad5/SV40 hybrid virus at the highly recombinogenic human chromosomal site lp36. Gene. 95:231-41, 1990. Rutila JE, Imperiale MJ, and WW Brockman. (1986) Replication and transformation functions of in vitro generated simian virus large T antigen mutants. Mol Cell Biol 58: 526 - 535. Sawai ET and JS Butel (1989) Association of a cellular heat shock protein with simian virus 40 large T antigen in transformed cells. J Virology 63: 3961 - 3973.  91  Schaeffer WI. Usage of Vertebrate, Invertebrate and Plant, Tissue and Organ Culture Terminology. (1984) In vitro 20:19-24. Scheidtmann KH, and Haber A (1990) Simian virus 40 large T antigen induces or activates a protein kinase which phosphorylates the transformation-associated protein p53. J Virol 64, 672 - 679. Schneider J and E Fanning (1988) Mutations in the phosphorylation sites of simian virus 40 (SV40) T antigen. J Virology 62, 1598 - 1605. Seshadri T and Campisi J (1990) Repression of c-fos transcription and an altered genetic program in senescent human fibroblasts. Science 247, 205-209. Shaw P, Bovey R, Tardy S, Sahli R, Sordat B and J Costa (1992) Induction of apoptosis by wild type p53 in a human colon tumor derived cell line. PNAS 89: 4495.4499.  Siemens CH. and Auersperg N. (1988) Serial propagation of human ovarian surface epithelium intissueculture. J of Cell Phys 134(3):347-56. Sturzbecher ITW, Montenarh M, Henning R (1987) Enhanced protein phosphorylation in SV40-transformed and -infected cells. Virology 160, 445 - 455 Sun TT, Green H. Immunofluorescent staining of keratin fibres in cultured cells. Cell 14:469 - 476, 1978. Sun TT, Shih C, Green H. Keratin cytoskeletons in epithelial cells of internal organs. Proc Natl Acad Sci USA 76:2813 - 2817, 1979. Van Niekerk CC, Ramaekers FCS, Hanselaar AG, Aideweireldt M, Poets, L. G. Changes in expression of differentiation markers between normal ovarian cells and derived tumors. Am. J. Pathol. 142:157-177, 1993. Veatch A.L., Carson L.F. and Ramakrishnan S. (1994) Differential expression of the cell-cell adhesion molecule E-cadherin in ascites and solid human ovarian tumor cells. International J. Cancer 58, 393-399. Wallace DC. (1992) Mitochondrial genetics: a paradigm for aging and degenerative diseases?. Science. 256:628-32, Wei JF. Wei FS. Samulski RJ. Barranger JA.(1994) Expression of the human glucocerebrosidase and arylsulfatase A genes in murine and patient primary 92  fibroblasts transduced by an adeno-associated virus vector. Gene Therapy. l(4):261-8. Whittemore SR, White LA. (1993) Target regulation of neuronal differentiation in a temperature sensitive cell line derived from medullary raphe. Brain Research 615:27-40. Yonish-Rouach E, Resnitzky D, Lotem J, Sachs L, Kimchi A and M Oren (1991) Wild type p53 induces apoptosis of myeloid leukaemic cells that is inhibited by interleukin y. Nature 353: 345-347. Zheng J. Wan M. Zweizig S. Velicescu M. Yu MC. Dubeau L. (1993) Histologically benign or low-grade malignant tumors adjacent to high-grade ovarian carcinomas contain molecular characteristics of high-grade carcinomas. Can Res 53, 4138-42.  93  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items