Immortalisation of human ovarian surface epithelium with telomerase and temperature-senstitive SV40 large T antigen

Available online at www.sciencedirect.com R

Experimental Cell Research 288 (2003) 390 – 402

www.elsevier.com/locate/yexcr

Immortalisation of human ovarian surface epithelium with telomerase and temperature-senstitive SV40 large T antigen Barry R. Davies,a,* Islay A. Steele,a Richard J. Edmondson,a Simon A. Zwolinski,b Gabriele Saretzki,c Thomas von Zglinicki,c and Michael J. O’Hared a

Department of Surgery, University of Newcastle, Newcastle-Upon-Tyne, NE2 4HH, UK Department of Cytogenetics, Institute of Human Genetics, University of Newcastle, Newcastle-Upon-Tyne, NE2 4HH, UK c Department of Gerontology, University of Newcastle, Newcastle-Upon-Tyne, NE2 4HH, UK d Ludwig Institute for Cancer Research-University College London Breast Cancer Laboratory, 67-73 Riding House Street, London W1W 7EJ, UK b

Received 12 November 2002, revised version received 24 March 2003

Abstract Epithelial ovarian cancer is the most common form of gynaecological malignancy. This lethal disease is thought to arise in ovarian surface epithelial (OSE) cells. The biology of these cells is not well understood, due to the limited amount of tissue that can be obtained from a single biopsy and their limited life span in culture. To overcome these problems, we have conditionally immortalised OSE cells with the catalytic subunit of telomerase (hTERT) and a temperature-sensitive form of SV40 Large T antigen (tsT). We have maintained these cells (designated OSE-C2) in culture for more than 100 population doublings after introduction of the immortalising genes. Early passage OSE-C2 cells have a near-tetraploid karyotype and exhibit a dual mesenchymal– epithelial phenotype, with consistent expression of vimentin and variable expression of cytokeratins and type III collagen, and absence of E cadherin expression. OSE-C2 cells proliferate steadily at the permissive temperature of 33°C, but fail to increase in number at the nonpermissive temperature of 39°C. Serum-deprived OSE-C2 cells are stimulated to grow at 33°C by EGF, whereas they are growth inhibited at 33°C by TGF␤ in the presence or the absence of serum. When temperature shifted to the nonpermissive temperature, OSE-C2 cells modulate to a more mesenchymal phenotype, and a proportion of the cells undergo senescence and/or apoptosis. Moreover, at the nonpermissive temperature, the levels of p53 and SV40 Large T antigen diminish, whilst the level of p21 increases, whereas the level of p16 and telomerase activity is unchanged. This experimental system shows that expression of telomerase alone only allows limited proliferative potential of OSE cells; expression of tsT is necessary to maintain these cells in culture for longer periods, perhaps by its ability to inactivate components of the p53/Rb pathway. OSE-C2 cells may be useful in studying the physiology and differentiation of human OSE cells and provide insight into the poorly understood earliest stages of epithelial ovarian cancer. © 2003 Elsevier Science (USA). All rights reserved. Keywords: Ovarian surface epithelium; Immortalisation; SV40 Large T antigen; Telomerase; Proliferation; Senescence; Apoptosis

Introduction Ovarian surface epithelial (OSE) cells are a single layer of simple epithelial cells that line the surface of the ovary. These cells are thought to be the site of origin of epithelial ovarian cancer (EOC), the commonest gynaecological ma-

* Corresponding author. Department of Surgery, The Medical School, University of Newcastle, Newcastle-Upon-Tyne, NE2 4HH, United Kingdom. Fax: ⫹0191-222-8514. E-mail address: [email protected] (B.R. Davies).

lignancy in women [1]. EOC is a highly lethal malignancy because it is grows insidiously and is difficult to detect. Consequently, most of these cancers are highly advanced at presentation, and the earliest stages of malignant transformation of OSE cells are poorly understood. In order to further our understanding of the growth and differentiation properties of OSE cells and how these properties are perturbed in malignancy, we have immortalised human OSE cells from a woman undergoing surgery for nonmalignant disease. Normal human somatic cells, including OSE cells, un-

0014-4827/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0014-4827(03)00218-0

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

dergo a limited number of divisions in vitro before entering an irreversible growth-arrest state defined as senescence [2]. However, overriding normal cell cycle checkpoints with viral oncogenes such as SV40 Large T antigen can result in an extension of growth potential, with most cells finally terminating in “crisis,” in which abortive or abnormal mitosis leads to progressive cell death [3]. In human cells, the progressive shortening of the telomeres with each cell division has been proposed as the “mitotic clock” that regulates the loss of replicative potential [4]. Telomeres are repetitive DNA sequences at the ends of chromosomes which can be maintained by telomerase [5,6]. Telomerase is a multicomponent enzyme which comprises a template RNA plus an essential catalytic protein subunit (hTERT) [7,8]. When adult somatic cells divide, their telomeres shorten by 10 to 200 bp per division [9], because the cells do not contain functional telomerase. The reconstitution of telomerase activity by ectopic expression of hTERT is sufficient for the immortalisation of some human somatic cells, including certain types of fibroblasts [10 –12]. Other cell types, including neonatal keratinocytes, mammary epithelium, mammary fibroblasts, and mammary endothelial cells, appear more sensitive to a telomere length-independent, probably stress-induced arrest. They seem to require specific cell culture conditions or inactivation of the p53/pRB/ p16INK4 pathway as well as functional telomerase [13,14]. Human OSE cells can be cultured from overtly normal ovaries at surgery for nonmalignant gynaecological diseases by gently scraping the ovarian surface with a rubber scraper [15]. The cultures usually proliferate for two to four passages and then senesce. The problem of obtaining sufficient numbers of OSE cells for experiments requiring large cell numbers has been approached by introduction of SV40 Large T antigen and the HPV E6 and E7 genes into human OSE [16,17]. However, such treatment does not result in true immortality but rather an increased but finite growth potential. Moreover, although many characteristics of normal OSE cells are retained by the introduction of these viral oncogenes, their proliferative properties are significantly altered. In order to produce a more representative system to study the biology and neoplastic progression of OSE, a temperature-sensitive, conditionally immortalising form of SV40 Large T antigen (tsT), with two point mutations which make the protein thermolabile at temperatures higher than 39°C, has been introduced into primary cultures of OSE cells [18]. By shifting the temperature of cells conditionally immortalised by this protein to 39°C, the effects of immortalisation are reversed and normal phenotypic characteristics are restored in certain differentiated cell types. In such conditionally immortalised cells, epithelial characteristics are retained or enhanced upon inactivation of tsT at 39°C, but proliferative arrest and subsequent senescence rapidly occur. Moreover, such cells also have a finite growth potential even when cultured at the permissive temperature (33°C). In order to generate truly immortal OSE cells with the

391

potential of retaining normal phenotypic and growth properties, we have immortalised human OSE cells from a woman undergoing surgery for nonmalignant disease, using retroviruses which transduce the catalytic subunit of telomerase (hTERT) [10,14] and a temperature-sensitive form of SV40 Large T antigen (tsT) [19,20]. We have maintained this pool of cells for at least 100 population doublings at the permissive temperature. We now report on the biological properties of these cells.

Materials and methods Establishment and transduction of primary cell cultures A primary culture of OSE cells was established from an ovary of a 48-year-old woman undergoing surgery for menorrahgia by scraping the surface with a sterile cell scraper [15]. Subsequent histological examination of the ovary showed no evidence of disease. Cells were maintained for two passages (approximately 4 population doublings) in routine medium (RPMI/Hams F12 medium supplemented with 10% FBS and antibiotics). The purity of the epithelial phenotype was confirmed by staining with a pan-cytokeratin antiserum (FITC conjugate, clone MF116, Dako). A midconfluence proliferating culture was exposed to filtered (0.4 ␮m) supernatants from the retroviral packaging lines TEFLY-A and PA317 producing high-titre full-length pBabehygro-hTERT [14] and pS/tsA58-U19/8 [20], respectively. Incubation with the retroviral packaging lines was carried out for 18 h with 8 mg/ml of polybrene. After this time, the medium was replaced by routine medium and the cells were grown to confluency. They were then selected with either 0.5 mg/ml of G418 or 25 mg/ml of hygromycin B until all untransduced cells had been killed (⬍14 days) [14]. Transduced cultures were grown at 33°C. Temperature-arrest experiments were carried out at 39°C. Growth assays Cell counting experiments were carried out by plating 1 ⫻ 104 cells/well in 96-well plates and counting sextuplicate wells using the MTT assay described by Mossman with minor modifications [21]. Briefly, 11 ␮l of 5 mg/ml 3-(4,5demethylthiazol-2-yl)-2,5-dipheyl tetrazolium bromide (MTT; Sigma) was added to 100 ␮l of PBS in each assay well and plates were incubated for 6 h; 100 ␮l of dimethyl sufoxide (DMSO) was added to each well and plates were read at 490 nm on a Dynex reader (Dynex). Senescence and apoptosis assays Senescence was assayed using the senescence-associated ␤-galactosidase method [22]. Apoptosis was analysed in two ways; caspase 3 activity was determined using the ApoAlert caspase 3 assay kit (Clontech) according to the

392

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

manufacturer’s instructions, and the proportion of apoptotic cells within a culture was determined by analysing forward/ sideward scatter by flow cytometry; apoptotic cells are smaller and more granular, resulting in lower forward scatter and greater side scatter, respectively [23]. Immunofluorescence microscopy The expression of proteins in cultured cells was studied by immunofluorescence microscopy using antisera to pan cytokeratins (FITC conjugate, clone MF116, Dako), cytokeratin 18 (clone LE61, obtained from Professor E.B. Lane, University of Dundee, UK), cytokeratin 19 (clone RCK108, Dako), collagen type III (clone UNLB, Southern Biotechnology Associates, Birmingham, AL), vimentin (clone J144, FITC conjugate, Serotec), Von Willebrand factor (clone F8/86, Dako), E cadherin (clone HECD-1, R & D Systems), SV40 Large T antigen (clone mAb423 (Dr. P. Yat), and p53 (Ab-6, Oncogene Research Products). Cells were plated in 8-well permanox chamber slides (Nunc) for approximately 48 h before fixing for 6 min in ice-cold methanol. After washing in PBS, fixed cells were incubated with primary antisera, some of which were directly conjugated to fluorescent tags. Samples incubated with nondirectly conjugated antisera were washed with PBS and subsequently incubated with a suitable flourescently conjugated second antibody. Some of the fixed cells were costained with 1.25 ␮g/ml of propidium iodide (Sigma) for 60 min before incubating with primary antiserum. Slides were mounted using aqueous mounting medium (Vector Laboratories, Burlingame, CA) and examined on a Leica DMR fluorescence microscope. Images were captured using a SPOT camera. Western blotting Cellular proteins were extracted by scraping subconfluent cultures into a buffer consisting of 4% (w/v) sodium dodecyl sulphate (SDS), 10% (w/v) sucrose, 100 mM dithiothreitol, 0.125 M Tris–HCl, pH 6.8, and 25 ␮g/ml of aprotinin. Equal amounts of total cellular proteins were separated by electrophoresis through SDS–12.5% polyacrylamide gels. Proteins were electroblotted onto Hybond ECL membrane (Amersham, Bucks, UK) using a mini-trans blot apparatus (Bio-Rad). Filters were incubated with antibody to p53 (Ab-6, Oncogene Research Products), p21 (Ab-1, Oncogene Research Products), or p16 (Pharmingen), and the bound antibody was detected using peroxidaseconjugated mouse immunoglobulins (Dako) and the ECL detection system (Pierce). Measurement of telomerase activity Telomerase activity was assayed via the Telo TTAGGG Telomerase PCR ELISA (Roche). Briefly, cell pellets were

resuspended in 200 ␮l of lysis buffer and centrifuged to remove cell debris. The supernatants were used in the telomeric repeat amplification protocol (TRAP), where telomerase adds telomeric repeats (TTAGGG) to the 3⬘ end of the biotin-labelled synthetic primer. These elongation products were subsequently amplified by PCR and detected by a digoxigenin (DIG)-labelled, telomeric repeat-specific ELISA according to the manufacturer’s instructions. As controls, lysis buffer alone, heat denaturing, and mortal cells were used. Telomere length measurements To prepare high-quality genomic DNA, cells at subconfluency were embedded in 0.65% low-melting agarose plugs at a density of 107 cells/ml before treatment with proteinase K. DNA was completely digested by HinfI (60 U per plug, Boehringer Mannheim, Germany) at 37°C. Plugs were analysed in a 1% agarose gel by pulsed-field gel electrophoresis (Bio-Rad, Hercules, CA). Gels were blotted to Hybond N⫹ membranes (Amersham) and hybridized with the telomeric probe (TTAGGG)4, directly conjugated to alkaline phosphatase (Promega, Madison, WI). A chemiluminescence signal was recorded on film within the linear range and analysed in an imaging densitometer (Fuji LAS100). Flow cytometric measurement of growth fraction Subconfluent cells were grown for 5 days at both the permissive (33°C) and nonpermissive (39°C) temperatures, before labelling with 10 ␮mol/L of BrdUrd for 30 min. After harvesting by trypsinisation, cells were fixed in icecold 70% ethanol, before incubating in 2 mol/L of HCl, 0.5% Triton X-100 for 30 min. Following neutralisation, the cell concentration was adjusted to 1 ⫻ 106 per 100 ␮l, and 100 ␮l of cells was incubated with 10 ␮l of antibody to BrdUrd (clone F7210, Dako). A nonreactive FITC-conjugated monoclonal antibody of the same isotype was used as a negative control. After counterstaining with 40 mg/ml of propridium iodide (PI; Sigma), cells were analysed on a flow cytometer. Tumorigenicty studies Subconfluent cells growing at the permissive temperature were harvested and resuspended in PBS at a concentration of 2 ⫻ 107 cell/ml; 0.2 ml of cell suspension (i.e., 4 ⫻ 106 cells) was injected ip or sc into nude mice. Mice were monitored for 4 months before they were killed humanely, and a postmortem was carried out. Animal experiments were carried out under UK Home Office Project Licence 60/2610 to Dr. B.R. Davies.

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

Karyotype analysis Metaphase spreads were prepared from colchicine-arrested cells grown at the permissive temperature using standard harvesting techniques [24].

Results Growth characteristics of immortalised OSE-C2 cells The growth characteristics of early passage (passage 8) and late passage (passage 28) OSE-C2 cells were studied at 33°C and after temperature shifts to 37 and 39°C. At the permissive temperature of 33°C, early and late passage OSE-C2 cells grew slowly but steadily, with average doubling times of between 100 and 120 h (Fig. 1A). The doubling times at early and late passage did not differ significantly from one another when the cells were maintained at 33°C (Table 1). When the temperature was raised from 33 to 37°C, early and late passage OSE-C2 cells grew at a similar rate to cells maintained at 33°C for the first 48 h, but thereafter the number of viable cells in the population decreased (Fig. 1A). After temperature shift from 33°C to the nonpermissive temperature of 39°C, the number of early passage OSE-C2 cells did not change significantly for 72 h, but after this time their number fell rapidly. The number of viable late passage OSE-C2 cells started declining after 24 h at 39°C (Fig. 1A). In contrast, two primary cultures of human OSE cells grew steadily at 33, 37, and 39°C. The doubling times of the two primary cultures did not differ significantly at the three temperatures analysed (Table 1). We have maintained OSE-C2 cells at 33°C for at least 52 passages (104 population doublings), whereas primary cultures of normal OSE have a very limited life span in culture before they become flattened and fail to proliferate further. We have studied the life span in culture of 39 primary cultures of human OSE cells (Fig. 1B). The median life span of these cultures is 2 (range 1–9). Only 4 primary cultures survived beyond the fourth passage in 75-cm2 petri dishes, when they were split at the same 1:4 ratio we have adopted for maintaining OSE-C2 cells. To gain further insight into the behaviour of OSE-C2 cells at the nonpermissive temperature, the cell cycle profile and the phenotypes of apoptosis and senescence were studied. The proportion of viable cells in S phase after 3 days, as determined by BrdURd flow cytometry, was approximately 16.5% when the cells were maintained at 33°C, but this decreased to 4.4% after incubation for 5 days at 39°C (Fig. 1C). Staining with propidium iodide showed that there was an increase in the sub-G1 peak at the nonpermissive temperature, suggesting that cell death occurs at this temperature (data not shown). Therefore, we determined the relative level of caspase 3 activity and quantified cell death by flow cytometry after temperature shift. After 2, 3, and 5 days at 39°C, the caspase 3 activity increased by a factor of

393

up to 10-fold compared with that in cells maintained at 33°C. However, this was still substantially less than the caspase 3 activity induced by incubation with 2.5 mM staurosporine for 4 h (Fig. 2A). Nonetheless, the proportion of dead cells in the culture increased from approximately 6% at 33°C to approximately 49% after 3 days at 39°C (Fig. 2B). Finally, we determined whether the cells undergo senescence after temperature shift to 39°C by staining early passage cultures for senescence-associated ␤-galactosidase. After 3 days at 39°C, 5.3 ⫹ 0.8% cells were positive, compared with 0.5 ⫹ 0.1% at 33°C, indicating that the higher temperature induces senescence-associated ␤-galactosidase activity by approximately 10-fold. After 7 days at 39°C, this increased further to 18.4 ⫹ 2.7% (Fig. 3). Expression of immortalising and cell cycle-associated proteins by immortalised OSE-C2 cells Nuclear immunoreactivity with antiserum to SV40 Large T antigen was intense when OSE-C2 cells were maintained at 33°C (Fig. 4A), whereas it was weak after 3 days at 39°C (Fig. 4B). Similarly, staining with antiserum to p53 gave clear nuclear immunoreactivity at 33°C (Fig. 4C), but this diminished considerably after 3 days at 39°C (Fig. 4D). Diminished expression of p53 at the nonpermissive temperature was confirmed by Western blotting experiments. After 3 days at 39°C, p53 expression was reduced but still detectable, whereas after 5 days no immunoreactive p53 was detectable (Fig. 5A). We then investigated whether temperature shift to the nonpermissive temperature was accompanied by an induction of the cyclin-dependent kinase inhibitors p21 and p16. p21 expression showed a gradual increase with time at the nonpermissive temperature, whereas p16 expression did not change significantly (Fig. 5B). Expression of telomerase activity in OSE-C2 cells was then studied by ELISA. Telomerase activity in early passage OSE-C2 cells was almost as great as that observed in the 41-M ovarian carcinoma cell line and exceeded the activity observed in the OVCAR3 ovarian carcinoma cell line, whereas virtually no telomerase activity was observed in two primary cultures of human ovarian surface epithelium (Fig. 6A). Telomerase activity was similar after extended passage in vitro (passage 33 versus passage 10) and after temperature shift to 39°C for 5 days (Fig. 6A). Telomerase activity slightly elongated the telomeres of both early and late passage OSE-C2 cells in comparison to those observed in a fresh primary culture of OSE-C2 cells (Fig. 6B). Expression of differentiation markers by immortalised OSE-C2 cells Over 95% of the cells in early passage OSE-C2 cultures stained with antiserum to pan cytokeratins and cytokeratin 19 when cultured at 33°C (Table 1 and Figs. 4e and f). After extensive passaging in vitro, the number of cytokeratinpositive cells at 33°C was reduced, but still remained ⬎50%

394

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

Fig. 1. Growth characteristics of OSE-C2 cells and primary cultures of normal OSE cells at various temperatures. (A) Growth curves at 33, 37, and 39°C for OSE-C2 cells at early and late passage and two independent primary cultures of OSE cells. Cells were plated in 96-well plates and cell number was determined by the MTT assay at 24, 48, 72, and 96 h. Cell number is proportional to the optical density at 490 nm. The data are representative of two independent experiments. (B) Graph showing the maximum passage number of 39 primary cultures of human OSE cells. (C) The S-phase fraction of OSE-C2 cells, determined after 5 days at 33 and 39°C by BrdUrd flow cytometry. The data are representative of three independent experiments.

(Table 2). After 3 days of culture at the nonpermissive temperature, the number of cytokeratin-positive cells was consistently less in both early and late passage cultures (Table 2). Approximately 39% of the cells in early passage OESC-2 cultures stained weakly with antiserum to collagen type III at the permissive temperature, but this increased

dramatically both in intensity and number to nearly 100% when the cells were temperature shifted to 39°C and maintained at this temperature for 3 days (Table 2 and Figs. 4g and h). In late passage cultures, approximately 45% of the cells stained with antiserum to collagen type III; this was only slightly increased when the cells were temperature

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

395

Table 1 Doubling times of OSE-C2 cells and primary cultures of human OSE cells at various temperatures Cells

OSE-C2 passage 8 OSE-C2 passage 28 Primary OSE 1 Primary OSE 2

Doubling time (h) at 33°C

37°C

39°C

106 ⫾ 13 118 ⫾ 11 92 ⫾ 6 111 ⫾ 8

Did not double Did not double 84 ⫾ 3 117 ⫾ 13

Did not double Did not double 95 ⫾ 13 127 ⫾ 9

Doubling times were determined using the MTT assay. The data are the mean ⫾ SD of sextuplicate replicates and are representative of two independent experiments.

Fig. 3. The effect of shift to the nonpermissive temperature on senescence in OSE-C2 passage 24 cells. Senescence was measured using the senescence-associated ␤-galactosidase assay. The mean and SE of five fields at ⫻40 magnification are shown. The data are representative of three independent experiments.

Fig. 2. The effect of shift to the nonpermissive temperature on apoptosis in OSE-C2 passage 24 cells. OSE-C2 cells were grown for the times indicated at 33 or 39°C. (A) Caspase 3 activity was determined by the ApoAlert caspase 3 assay kit, which measures the absorbance at 405 nm due to production of the chromophore p-nitroaniline (pNA) after its cleavage by caspases from labelled caspase-specific substrates (Clontech). The mean and SE of triplicate measurements are shown. The data are representative of two independent experiments. (B) Apoptotic cells were identified in density plots by their lower forward (FSC) and higher sideward (SSC) scatter intensity as measured by flow cytometry. The data are representative of two independent experiments.

396

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

397

serum-starved controls after 7 days. In contrast, EGF and FBS both significantly increased the number of cells, and TGF␤ significantly decreased the number of cells, compared to cells maintained in serum-free medium (Fig. 7A). In the presence of serum, TGF␤ still resulted in a decrease in cell number, whereas addition of EGF did not result in a further increase in cell number compared with serum alone (Fig. 7B). Tumorigenicity and anchorage dependence of immortalised OSEC-2 cells Innocula of 4 ⫻ 106 early and late passage OSE-C2 cells failed to produce tumours at subcutaneous or intraperitoneal sites in any nude mice after a 4-month experimental period. Moreover, these cells failed to produce colonies in soft agar at the permissive and nonpermissive temperatures, whereas two ovarian cancer cell lines, R199neu and 41-M, cloned with an efficiency of ⬎50%, as described previously [25]. Karyotype of immortalised OSEC-2 cells

Fig. 5. Expression of the p53 (A) and p21 and p16 (B) proteins by OSE-C2 cells at the permissive and nonpermissive temperatures. Cell extracts were prepared from OSE-C2 cells grown at 33 or at 39°C for 3, 5, and 7 days. Proteins were separated by SDS–PAGE under denaturing conditions and probed with the relevant antisera. To check equal loading, blots were stripped and reprobed with antiserum to ␣-tubulin.

shifted to 39°C. Both early and late passage OSEC-2 cells stained consistently with antiserum to vimentin and failed consistently to stain with antisera to E cadherin and Von Willibrand factor (Table 2). Effect of growth factors and hormones on immortalised OSEC-2 cells The growth responses of OSE-C2 cells to several polypeptide growth factors and gondotrophic hormones were studied using MTT growth assays. The effects of growth factors and hormones were first analysed in serumfree medium. Concentrations of the growth factors KGF (FGF-7) and HGF, which have been previously shown to stimulate DNA synthesis in ovarian cancer cell lines, and physiological doses of the gonadotrophins FSH and LH did not result in an increase in cell number compared with

OSE-C2 cells have a near-tetraploid karyotype, with a mode chromosome number of 86 (range 82– 87). Seven marker chromosomes were commonly observed (Fig. 8). Apparent monosomy of chromosome 11 is a consistent feature, though marker chromosomes mar1 and mar2 are probably derived from chromosome 11.

Discussion The OSE is a single layer of cells that covers the ovary. Primary cultures of human OSE typically produce fewer than 2 million cells and have a very limited median life span of two passages in culture. In our laboratory, only 10% of such primary cultures survive at least five passages, and we have never succeeded in maintaining a culture beyond the ninth passage. Sequential transduction of the hTERT and U19tsa58 LT (tsLT) genes into a primary culture of human OSE cells has resulted in an immortal culture that grows without either a senescent or crisis phase being observed, when the cells are maintained at 33°C. We have maintained these cells for over 104 population doublings, whereas human OSE cells expressing tsLT alone are limited to between 52 and 71 population doublings [18]. Although reconstitution of telomerase activity alone is sufficient for immortalisation of freshly isolated adult human mammary fibroblasts, microvascular endothelial cells, and myometrial cells

Fig. 4. Immunofluorescence microscopy showing expression of cell cycle-associated proteins and differentiation-associated proteins in OSE-C2 cells. OSE-C2 cells were grown on chamber slides, fixed in ice-cold methanol, and stained with the appropriate antisera. (A) SV40 Large T antigen, 33°C; (B) SV40 Large T antigen, after 5 days at 39°C; (C) p53, 33°C; (D) p53, after 5 days at 39°C; (E) Costaining with prodridium iodide and FITC-cytokeratin 19, 33°C; (F) TRITC-cytokeratin 18, 33°C; (G) Costaining with propridium iodide and FITC-collagen type III, 33°C; (H) Costaining with propridium iodide and FITC-collagen type III, after 3 days at 39°C. Original magnification of all panels, ⫻40. The data are representative of two independent experiments.

398

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

Fig. 6. Telomerase activity and telomere lengths of OSE-C2 cells. (A) Telomerase activity was determined by the Telo TAGGG Telomerase PCR ELISA in two independent primary cultures of human OSE cells, OSE-C2 cells at early and late passage at 33°C or after 5 days at 39°C, and in the ovarian carcinoma cell lines 41M and OVCAR3. The data are representative of three independent experiments. (B) Telomere lengths were determined by digestion with Hinf1 and pulsed-field gel electrophoresis. Lane 1, primary culture of OSE cells; lane 2, early passage OSEC-2 cells (p9); lane 3, late passage OSE-C2 cells (p25).

[10 –12,26], it would appear that these human OSE cells depend on LT antigen for maintenance of growth, at least under the culture conditions employed. In this respect they are similar to freshly isolated human mammary fibroblasts, mammary endothelial cells, and neonatal keratinocytes and primary human airway epithelial cells [14,27]. Although ectopic expression of hTERT resulted in the stabilisation of telomeres, temperature shift to 39°C and inactivation of T antigen resulted in a very significant reduction in growth and DNA synthesis such that the cells could not be passaged

further, along with an induction of senescent and apoptotic phenotypes. It is unlikely that a small subpopulation of cells can grow independently of T antigen because we have been unable to clone any cells maintained at 39°C. Moreover, at this temperature, OSE-C2 cells fail to form colonies when plated at low seeding densities, whereas they readily do so at 33°C (B.R. Davies and I.A. Steele, unpublished observations). Whilst these data strongly suggest that OSE cells cannot be immortalised by telomerase alone, it is not conclusive proof, as it is possible that the requirement for Large

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

399

Table 2 Expression of markers by OSE-C2 cells Antiserum

Pan cytokeratin CK8 (LE61) CK19 Collage Type 3 Vimentin E cadherin Von Willebrand factor

Early passage

Late passage

33°C

39°C, 3 days

33°C

39°C, 3 days

99.3 ⫹ 0.2 42.9 ⫹ 2.0 96.2 ⫹ 0.5 38.9 ⫹ 4.1 100 ⫹ 0.0 0.0 ⫹ 0.0 0.0 ⫹ 0.0

50.3 ⫹ 2.6 28.3 ⫹ 4.4 37.6 ⫹ 1.9 98.2 ⫹ 0.9 100 ⫹ 0.0 0.0 ⫹ 0.0 0.0 ⫹ 0.0

62.7 ⫹ 1.5 ND 56.8 ⫹ 2.2 44.6 ⫹ 2.2 100 ⫹ 0.0 0.0 ⫹ 0.0 0.0 ⫹ 0.0

27.3 ⫹ 4.6 ND 38.1 ⫹ 1.8 49.5 ⫹ 1.7 100 ⫹ 0.0 0.0 ⫹ 0.0 0.0 ⫹ 0.0

Note. Expression of the various markers was determined by immunofluorescence microscopy. The data represent the mean ⫾ SD of at least five fields at ⫻40 magnification and are representative of two independent experiments.

T is a substitution for inadequate culture conditions, such as absence of epithelial–mesenchymal interactions, absence of feeder cells, or suboptimal medium composition. This having been said, the apparent requirement for hTERT alone to immortalise some primary cultures of human cells may also involve abrogation of the intrinsic cell cycle regulating pathways that are inhibited by Large T, independently of the external milieu, or viral immortalising genes. Indeed, additional events, such as inactivation of the p16INK4a or p14ARF genes, have been shown in hTERT immortalised human foreskin keratinocytes and adenoid epithelial cell lines [28]. The behaviour of the cultures at 39°C is heterogeneous, with approximately 50% of the cells undergoing apoptosis within 3 days, whilst approximately 20% of the cells remaining after 7 days show induction of a senescent phenotype. Moreover, there is a significant decrease in the Sphase fraction of the remaining viable cells after 5 days at 39°C. The factors that control the cells’ decision whether to undergo growth arrest, senescence, or apoptosis are not known but may reflect differences in the age or numbers of replications of the cells in the original donor population, the degree of free radical or other stress-associated damage, or the mutational status of other components of cell cycle control. At the molecular level, there is a progressive loss of immunoreactive nuclear p53 and SV40 Large T antigen at the nonpermissive temperature, whereas there is no significant change in telomerase activity. The progressive loss of p53 expression is most probably a consequence of the inactivation of the LT protein; LT is known to sequester and inactivate p53. Presumably, inactivation of LT leads to degradation of the majority of the nuclear p53 protein, but sufficient wild-type p53 protein is retained to activate transcription of the p21 gene, which can contribute to the induction of growth arrest and possibly senescence and apoptosis. However, a proportion of OSE-C2 cells remain viable even after maintenance at the nonpermissive temperature for 2 weeks; when these cells are replated at 33°C they grow and divide with similar kinetics to those observed before temperature shift. Moreover, the replated population retains a mixture of cytokeratin and collagen type-III-posi-

tive cells, so both more epithelial- and mesenchymal-like cells can reenter the cell cycle following a period of extended growth arrest (B.R. Davies and I.A. Steele, unpublished observations). Immortalised OSEC-2 cells exhibit many of the properties characteristic of primary cultures of human OSE and human OSE cell lines with extended life spans, such as those conditionally immortalised with temperature-sensitive Large T antigen alone [18]. For example, they exhibit a poorly committed, dual mesenchymal- epithelial phenotype [29,30]. Early passage cultures express markers of simple epithelium, such as cytokeratins 8, 18, and 19, in the majority of the cells [31], but fail to express markers of complex epithelial differentiation, such as E cadherin; this is consistent with the phenotype of human OSE in vivo [32,33]. After extensive passaging in vitro, the percentage of cells expressing cytokeratin is reduced, whereas the reverse is true of the mesenchymal marker collagen type III. This is consistent with previous observations that OSE cells tend to lose epithelial characteristics and gain mesenchymal characteristics with increasing passage number in vitro [34]. Both early and late passage cultures express vimentin, which is also a characteristic of OSE cells in vivo [29]. However, they fail to stain with antiserum to Von Willebrand factor, showing that they are not endothelial cells. Conditionally immortalised cells often show characteristics of differentiation concurrent with loss of proliferative potential when temperature shifted to the nonpermissive temperature. OSE cells expressing tsLT alone showed induction of cytokeratin and collagen type III when grown at the nonpermissive temperature [18], whereas OSE-C2 cells immortalised with tsLT and hTERT showed induction of collagen type III and a reduction in cytokeratin expression at the nonpermissive temperature. In both experimental systems, shift to the nonpermissive temperature resulted in growth arrest, but in the cultures expressing tsT alone a greater proportion of the cells expressed senescence-associated ␤ galactosidase activity than in the tsLT/hTERT cultures [18]. The reason for this discrepancy is not known, but it may be that our hTERT/tsLT immortalised cells are more firmly committed to the mesenchymal differentiation

400

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

Fig. 7. Effects of growth factors and hormones on passage 26 OSE-C2 cells. (A) OSE-C2 cells were plated in 96-well plates in serum-free medium or serum-free medium containing the growth factors or hormones indicated. After 7 days, cell numbers were determined by the MTT assay. The mean and SE of sextuplet measurements are shown. (B) OSE-C2 cells were plated in 96-well plates in medium containing 10% FBS and the growth factors or hormones indicated. After 7 days, cell numbers were determined by the MTT assay. The mean and SE of sextuplet measurements are shown. The data are representative of three independent experiments.

pathway, whereas the tsLT cells retain the ability to enter both differentiation pathways. Alternatively, it is possible that the more epithelial-like cytokeratin-expressing subpopulation of tsT/hTERT cells is more prone to undergo apoptosis at the nonpermissive temperature. The growth of tsLT/hTERT immortalised OSE-C2 cells can be influenced by at least two defined exogenous growth factors: EGF results in an increase in cell number when the cells are grown in serum-free medium, whereas TGF␤ results in a decrease in cell number after 7 days in serum-free or serum-containing medium. The ability of EGF and TGF␤ to respectively increase and decrease cell number is consistent with the findings of previous studies in primary cultures of OSE cells [35]. This property distinguishes OSE-C2 cells

from mesothelial cells, in which TGF␤ has been reported to induce DNA synthesis [36]. In summary, we have generated a heterogenous culture of immortalised human OSE cells by transduction with retroviruses directing expression of a temperature-sensitive form of the viral oncogene SV40 Large T antigen and the catalytic subunit of human telomerase. These cells will be of value in understanding the fundamental mechanisms of immortalisation of human OSE cells and how the processes of senescence, apoptosis, and differentiation are deregulated in the early stages of malignant transformation of this tissue. Such cells also provide a first step in the generation of “designer ovarian cancers” in which the ability of defined oncogenes and tumour suppressor genes to cooperate with

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402

401

Fig. 8. Karyotype of OSE-C2 cells. A typical karyotype of an OSE-C2 cell with 83 chromosomes, including 7 commonly observed marker chromosomes. Whilst there is only one copy of chromosome 11, there is probably chromosome 11 material present on mar1 and mar2.

telomerase and the p53/Rb pathway to generate ovarian cancer cells may be assayed.

Acknowledgments We thank Dr. Brian Shenton (Department of Surgery, University of Newcastle) for assistance with the flow cytometry. This work was supported in part by a Cancer Research UK William Ross Ph.D. studentship to I.A. Steele.

References [1] N. Auersperg, A.S.T. Wong, K-C. Choi, S.K. Kang, P.C.K. Leung, Ovarian surface epithelium: biology, endocrinology and pathology, Endocr. Rev. 22 (2001) 255–288. [2] V.J. Cristofalo, R.G. Allen, R.J. Pignolo, B.G. Martin, J.C. Beck, Relationship between donor age and the replicative lifespan of human cells in culture: a re-evaluation, Proc. Natl. Acad. Sci. USA 95 (1998) 10614 –10619. [3] J.W. Shay, W.E. Wright, H. Werbin, Defining the molecular mechanisms of human cell immortalisation, Biochim. Biophys. Acta 1072 (1991) 1–7. [4] R.C. Allsopp, H. Vaziri, C. Patterson, S. Goldstein, E.V. Younglai, A.B. Futcher, C.W. Greider, C.B. Harley, Telomere length predicts replicative capacity of human fibroblasts, Proc. Natl. Acad. Sci. USA 89 (1992) 10114 –10118.

[5] D. Hanahan, R.A. Weinberg, The hallmarks of cancer, Cell 7 (2000) 57–70. [6] S. Goldstein, Replicative senescence: the human fibroblast comes of age, Science 249 (1990) 1129 –1133. [7] M. Meyerson, C.M. Counter, E.N. Eaton, L.W. Ellisen, P. Steiner, S.D. Caddle, L. Ziaugra, R.L. Beijersbergen, M.J. Davidoff, Q. Liu, hEST2, the putative human telomerase catalytic subunit gene, is up-regulated in tumour cells and during immortalisation, Cell 90 (1997) 785–795. [8] J. Zhu, J.M. Bishop, E.H. Blackburn, Telomerase extends the lifespan of virus-transformed human cells without net telomere lengthening, Proc. Natl. Acad. Sci. USA 96 (1999) 3723–3728. [9] T. Von Zglinicki, V. Serra, M. Lorenz, G. Saretzki, R. LenzenGrossimlighaus, R. Gessner, A. Risch, E. Steinhagan-Thiessen, Short telomeres in patients with vascular dementia: an indicator of low antioxidative capacity and a possible prognostic factor, Lab. Invest. 80 (2000) 1739 –1747. [10] A.G. Bodnar, M. Ouellette, M. Frolkis, S.E. Holt, C.P. Chiu, G.B. Morin, C.B. Harley, J.W. Shay, S. Lichtsteiner, W.E. Wright, Extension of lifespan by introduction of telomerase into normal cells, Science 279 (1998) 349 –352. [11] J. Yang, E. Chang, A.M. Cherry, C.D. Bangs, Y. Oei, A. Bodnar, A. Bronstein, C.P. Chiu, G.S. Herron, Human endothelial cell life extension by telomerase expression, J. Biol. Chem. 274 (1999) 26141– 26148. [12] M.M. Ouellette, M. Liao, B.S. Herbert, M. Johnson, S.E. Holt, H.S. Liss, J.W. Shay, W.E. Wright, Senescent telomere lengths in fibroblasts immortalised by limiting amounts of telomerase, J. Biol. Chem. 275 (2000) 10072–10076. [13] T. Kiyono, S.A. Foster, J.I. Koop, J.K. Mc Dougall, D.A. Galloway, A.J. Klingelhutz, Both Rb/p16 (INK4a) inactivation and telomerase

402

[14]

[15]

[16]

[17]

[18]

[19] [20]

[21]

[22]

[23]

[24]

[25]

B.R. Davies et al. / Experimental Cell Research 288 (2003) 390 – 402 activity are required to immortalise human epithelial cells, Nature 396 (1998) 84 – 88. M.J. O’Hare, J. Bond, C. Clarke, Y. Takeuchi, A.J. Atherton, C. Berry, J. Moody, A.R.J. Silver, D.C. Davies, A.E. Alsop, A. Munro Neville, P.S. Jat, Conditional immortalisation of freshly isolated human mammary fibroblasts and endothelial cells, Proc. Natl. Acad. Sci. USA 98 (2001) 646 – 651. P.A. Kruk, S.L. Maines-Bandiera, N. Auersperg, A simplified method to culture human ovarian surface epithelium, Lab. Invest. 63 (1990) 132–136. S.L. Maines-Bandiera, P.A. Kruk, N. Auersperg, SV40-transformed human OSE cells escape normal growth controls but retain morphogenetic responses to extracellular matrix, Am. J. Obstet. Gynecol. 167 (1992) 729 –735. S.W. Tsao, S.C. Mok, E.G. Fey, J.A. Fletcher, T.S.K. Wan, E-C. Chew, M.G. Muto, R.C. Knapp, R.S. Berkowitz, Characterisation of human OSE cells immortalised by human papilloma viral oncogenes (HPV-E6/E7 ORFs), Exp. Cell Res. 218 (1995) 499 –507. E.H.L. Leung, P.C.K. Leung, N. Auersperg, Differentiation and growth potential of human OSE cells expressing temperature-sensitive SV40 T antigen, In Vitro Cell. Dev. Biol.-Anim. 37 (2001) 515–521. G. Almazan, R. McKay, An oligodendrocyte precursor cell line from rat optic nerve, Brain Res. 579 (1992) 234 –245. A.C. Stamps, S.C. Davies, J. Burman, M.J. O’Hare, Analysis of proviral integration in human mammary epithelial cell lines immortalised by retroviral infection with a temperature sensitive Large T antigen construct, Int. J. Cancer 57 (1994) 865– 874. T. Massmann, Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays, J. Immunol. Methods 65 (1983) 55– 63. G.P. Dimri, X. Lee, G. Basile, M. Acosta, G. Scott, C. Roskelly, E.E. Medrano, M. Linskens, I. Rubelj, O. Pereira-Smith, M. Peacocke, J. Campisi, A biomarker that identifies senescent human cells in culture and in aging skin in vivo, Proc. Natl. Acad. Sci. USA 92 (1995) 9363–9367. G. Saretzki, A. Lugwig, T. von Zglinicki, I.B. Runnebaum, Ribozyme-mediated telomerase inhibition induces immediate cell loss but not telomere shortening in ovarian cancer cells, Cancer Gene Ther. 8 (2001) 827– 834. D.E. Rooney (Ed.), Human Cytogenetics: Malignancy and Acquired Abnormalities. A Practical Approach, third ed., Oxford Univ. Press, Oxford. B.R. Davies, N. Auersperg, S.D. Worsley, B.A.J. Ponder, Transfection of rat ovarian surface epithelium with neulerb-B2 induces trans-

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

formed phenotypes in vitro and the tumorigenic phenotype in vivo, Am. J. Pathol. 152 (1998) 297–306. J. Condon, S. Yin, B. Mayhew, R.A. Word, W.E. Wright, J.W. Shay, W.E. Rainey, Telomerase immortalisation of human myometrial cells, Biol. Reprod. 67 (2002) 206 –514. A.S. Lundberg, S.H. Randell, S.A. Stewart, B. Elenbaas, K.A. Hartwell, M.W. Brooks, M.D. Fleming, J.C. Olsen, S.W. Miller, R.A. Weinberg, W.C. Hahn, Immortalisation and transformation of primary human airway epithelial cells by gene transfer, Oncogene 21 (2002) 4577– 4586. D.G. Farwell, K.A. Shera, J.I. Koop, G.A. Bonnet, C.P. Matthews, G.W. Reuther, M.D. Coltrera, J.K. McDougall, A.J. Klingelhutz, Genetic and epigenetic changes in human epithelial cells immortalised by telomerase, Am. J. Pathol. 156 (2000) 1537–1547. N. Auersperg, S.L. Maines-Bandiera, H.G. Dyck, P.A. Kruk, Characterisation of cultured human OSE cells: phenotypic plasticity and premalignant changes, Lab. Invest. 71 (1994) 510 –518. P.A. Kruk, V.J. Uitto, J.D. Firth, S. Dedhar, N. Auersperg, Reciprocal interactions between human OSE cells and adjacent extracellular matrix, Exp. Cell. Res. 215 (1994) 97–108. C.C. Van Niekerk, O.C. Boerman, F.C.S. Ramaekers, L.G. Poels, Marker profile of different phases in the transition of normal human ovarian epithelium to ovarian carcinomas, Am. J. Pathol. 138 (1991) 455– 463. B.R. Davies, S.D. Worsley, B.A.J. Ponder, Expression of E cadherin, ␣ catenin and ␤ catenin in normal ovarian surface epithelium and epithelial ovarian cancers, Histopathology 32 (1998) 69 – 80. K. Sundfeldt, Y. Piontkewitz, K. Ivarsson, O. Nilsson, P. Hellberg, M. Brannstrom, P.O. Janson, S. Enerback, L. Hedin, E cadherin expression in human epithelial ovarian cancer and normal ovary, Int. J. Cancer 74 (1997) 275–280. H.G. Dyck, T.C. Hamilton, A.K. Godwin, H.T. Lynch, S. MainesBandiera, N. Auersperg, Autonomy of the epithelial phenotype in human ovarian surface epithelium: changes with neoplastic progression and with a family history of epithelial ovarian cancer, Int. J. Cancer 69 (1996) 429 – 436. A. Burchuck, G. Rodriguez, G. Olt, R. Whitaker, M.P. Boente, B.A. Arrick, D.L. Clarke-Pearson, R.C. Bast, Regulation of growth of normal ovarian epithelial cells and ovarian cancer cell lines by transforming growth factor ␤, Am. J. Obstet. Gynecol. 166 (1994) 676 – 684. E.W. Gabrielson, B.I. Gerwin, C.C. Harris, A.B. Roberts, M.B. Sporn, J.F. Lechner, Stimulation of DNA synthesis in cultured primary human mesothelial cells by specific growth factors, FASEB J. 2 (1998) 2717–2721.