Responses of human cervical keratinocytes in vitro to ...

Carcinogenesis vol.6 no.7 pp. 1011-1015, 1985

Responses of human cervical keratinocytes in vitro to tumour promoters and diethylstilboestrol

M.A. Stanley, N.S. Crowcroft, J.P. Quigley and E.K. Parkinson1 Department of Pathology, Tennis Court Road, Cambridge CB2 1QP, and 'The Paterson Laboratories, Christie Hospital and Holt Radium Institute, Wilmslow Road, Manchester M20 9BX, UK

Introduction Diethylstilboestrol (DES)* is a synthetic non-steroidal oestrogen causally associated with the development of clear cell adenocarcinoma of the vagina and cervix in the female progeny of women treated with the drug during pregnancy (1). In addition to this rare lesion of tuboendometrial origin (2) other reproductive abnormalities have been described in the exposed offspring. These include vaginal adenosis, cervical ectropion, squamous metaplasia and gross abnormalities of the vagina and cervix (3). Experimental investigations using the neonatal mouse (4,5) have shown that •Abbreviations: DES, diethylstilboestrol; HCE, human ectocervical cells; PMA, phorbol-12-myris(ate-13-acetate; FBS foetal bovine serum; EGF, epidermal growth factor. © IRL Press Limited, Oxford, England.

Materials and methods Cell culture HCE were prepared as described previously (18) and kept as frozen stocks. All cutlures of normal HCE were grown in the Glasgow modification of Eagle's medium (19) supplemented with 10% foetal bovine serum (FBS, Sera Labs), 0.5 ^g/ml hydrocortisone and choleragen (Sigma) at 10~10 M. Epidermal growth factor (EGF, Collaborative Research) at 10 ng/ml was added 2 4 - 7 2 h after plating. All cultures were grown at 37°C in a humidified atmosphere of 5% COj in air. One cell strain AC 69 was used in all experiments involving normal keratinocytes. This cell strain was derived from a 36-year-old female undergoing hysterectomy for benign uterine disease. The cells were used at second passage and had undergone 27 population doublings in culture. Strain AC 69 exhibits a normal diploid karyotype. All normal HCE were routinely cultivated in the presence of lethally irradiated Swiss 3T3 fibroblast (XR SW 3T3) feeder cells. Swiss 3T3, fibroblasts were grown and irradiated for use as feeder layers as described previously (18). Swiss 3T3 conditioned medium was prepared as described by Rheinwald (17). Line CC22 originated from a squamous-cell carcinoma of the cervix and was derived in his laboratory using the technique described by Melnick (20). The line was a gift from Dr. K. Powell, Department of Medical Microbiology, Leeds University. This line was routinely cultivated in the absence of 3T3 feeder cells and was grown in GMEM supplemented with 10% FBS and 0.5 /ig/ml hydrocortisone. The inoculation of 5 x 10* cells of CC22 into the nude mouse resulted in a progressively enlarging nodule which on histological examination was found

1011

Downloaded from http://carcin.oxfordjournals.org/ at Queen Mary, University of London on October 20, 2015

We have compared the responses of normal human cervical keratinocytes (HCE) to diethylstilboestrol (DES), and the promoting agents, phorbol-12-myristate-13-acetate (PMA) and mezerein using the loss of cloning efficiency as a measure of terminal differentiation in vitro. Dose-response studies showed that normal HCE are growth inhibited by chronic exposure to DES at concentrations >2.5 x 10~ 5 M, to PMA at concentrations >10~ 8 M and mezerein at concentrations > 10 ~9 M. Compared to acetone controls, promoter or DEStreated cells exhibited a 10- to 12-fold increase in cornifledenvelope formation. Normal HCE exhibit a heterogeneous response to PMA in that 85 — 90% of colony-forming cells lose their colony-forming ability after a 24-h exposure to 10 ~6 M PMA. The PMA-resistant subpopulation, PMAR, remains constant and is not reduced even after 96 h chronic exposure to PMA. In contrast, the colony-forming ability of normal HCE is almost totally suppressed after 24 h exposure to 10~ ( M mezerein. After 24 h incubation with 5 x 10~ 5 M DES, 20% of normal HCE are capable of colony formation but this resistant fraction is eliminated after 96 h chronic exposure. Comified-envelope formation was negligible in malignant cervical keratinocytes grown in the presence of DES or promotors and these cells were characterised by a very large PMAR fraction — 85 — 90% of cells retained colony-forming ability after exposure to 10~6 M PMA for 24 h. Furthermore, 90 —100% of malignant cervical keratinocytes retained their colony-forming capacity after exposure to 10~ 6 M mezerein. However, colony-forming ability declined steadily in the presence of 5 x 10~ 5 M DES and after % h only a tiny fraction, 1%, of malignant cervical keratinocytes could form colonies on replating. The mechanisms by which DES inhibits growth and induces cornified-envelope formation in HCE would appear to be distinct from those activated by PMA and mezerein.

the administration of DES during critical periods of genital tract organogenesis results in genital-tract neoplasms which may have a glandular or squamous morphology. These studies clearly indicate that the target tissue for DES is the Mullerian cervicovaginal epithelium but the cellular mechanism whereby DES induces neoplasia is not clear. The drug is negative in bacterial mutagenicity tests (6) but transforms neonatal Syrian hamster cells in vitro (7), an effect which has been ascribed to the induction of aneuploidy and chromosomal rearrangement; such an action of DES has been reported in various systems (8,9). Recent work has implicated DES as a co-carcinogen (10) or as a promoter in two-stage carcinogenesis systems in both rodent (11) and human cells (12). In these systems DES mimicked the effect of the prototype promoting agent phorbol-12-myristate-13-acetate (PMA). At high concentrations, 5 x 10~5 to 10~* M, DES inhibits the growth of human ectocervical cells (HCE) in vitro (13) and induces terminal differentiation in these cultures. Growth inhibition and the induction of terminal differentiation is observed in normal epidermal keratinocytes in vitro after exposure to the promotor PMA (14). However, work with normal epidermal keratinocytes reveals that the cell population is heterogeneous in the response to PMA, in that a small but significant fraction of the population is resistant to PMA-induced terminal differentiation retaining the capacity to form viable cell colonies after replating (15). Malignant or transformed keratinocytes are distinguished by a very much increased fraction of PMA-resistant cells (15) and the acquisition by this fraction of resistance to the secondstage promoter, mezerein (16). These observations prompted us to compare the responses of human cervical keratinocytes to DES and the promoting agents PMA and mezerein using the loss of cloning efficiency, the first detectable step in the sequence of changes leading to terminal differentiation (17), as a measure of terminal differentiation in HCE in vitro.

M.A. Stanley et al. to be an infiltrating keratinising squamous cell carcinoma. The cells of CC22 are keratinocytes and have a cytokeratin profile after 2D gel electrophoresis consistent with an ectocervical origin (Stanley and Quigley, unpublished data). For experiments a frozen ampoule of strain AC 69 was rapidly recovered from frozen stocks and cells were plated at 5 x 10* HCE together with 5 x 105 XR SW 3T3 in 50-mm dishes (Falcon) and grown for 12 days before use. Line CC22 was inoculated at 5 x 105 onto 50 mm culture dishes and grown for 5 days before use. DES and PMA were obtained from the Sigma Chemical Co. and mezerein from LCS Services, Wobum, MA, USA. All compounds were dissolved in acetone before being added to the cultures, in which the final acetone concentration did not exceed 0.1 %.

O

Colony Number % Control 10

To9

Concentration (M)

Fig. 1. The effect of chronic exposure to PMA or DES or mezerein over ; range of concentrations on the colony-forming efficiency of normal HCE relative to acetone controls. The plating efficiency of the controls was 2 9 ± 0.31%. The values represent the mean of two experiments, n = 8. Open circles, DES; closed circles, PMA; triangles, mezerein.

Comified-envelope formation This was carried out after exposure to the test compounds as described by Parkinson et al. (15) using the method of Green (23).

Results The effects of PMA and mezerein on normal HCE growth The effect of mezerein and PMA in the dose range 10~4 to 10~10 M on the growth of normal HCE was examined and the results are shown in Figure 1. HCE are growth inhibited, in terms of colony number, by chronic exposure to PMA at doses ;> 10~ 8 M. PMA doses of lO" 9 M and KT 1 0 M had no effect on HCE growth. Mezerein at doses of 2:10~9 M inhibited HCE growth as measured by colony number. These effects are similar to those observed on epidermal keratinocytes (14) although HCE appear to be less sensitive to both PMA and mezerein growth inhibition. The presence or absence of EGF or cholera toxin had only marginal effects on the growth inhibition of HCE by PMA or mezerein. HCE colony formation is inhibited by DES at concentrations 2 : 2 . 5 x l O ~ 5 M , a n effect which has been described in previous work (13). The effect of PMA, mezerein and DES on the formation ofcornified envelopes in normal and malignant cervical keratinocytes The results displayed in Figure 2 show the effect of a 6-day exposure of normal HCE to PMA or mezerein or DES on comified-envelope formation. Compared to acetone controls, promoter- or DES-treated cells exhibited a 10- to 12-fold increase in cornified envelopes. Comified-envelope formation was negligible in CC22 cells grown in the presence of DES or promoters for 6 days. Colony forming ability following PMA, mezerein or DES treatment The data displayed in Figure 3 show that HCE exhibit a response to a growth-inhibitory dose of PMA comparable to that described for epidermal keratinocytes (15). Thus exposure of HCE to PMA at 10 ~6 M for 24 h results in a loss of colony-forming ability by 85—90% of cells capable of colony formation in acetone-treated control cultures. This resistant population is not reduced even after 96 h incubation in the presence of 10~6 M PMA. The size of the resistant population is very constant under the experimental conditions used. A subpopulation of HCE resis1012

o —O

1

20

I

18 ^ b Cornified Envelopes 14 12 10 8 6 4 2 0

A Control

1O6M 1O6M 5 x 1 0 ^ PMA MEZ DES

Fig. 2. The effect of PMA or DES or mezerein on the formation of cornified envelopes by normal HCE (strain AC 69) after 6 days in the presence of the compound. The data represent the fraction of cornified envelopes present in disaggregated cell suspensions expressed as a percentage of the total cells and are the means of eight replicates ± SEM.

tant to mezerein is not apparent and colony formation is almost totally suppressed after 24 h incubation with 10~6 M mezerein. This effect is not reversed or modified even after 96 h in the presence of the chemical. Decreasing the dose of mezerein to 10 ~7 M did not increase colony-plating efficiency significantly (data not shown). The response to DES by HCE is different to that of PMA or mezerein. After 24 h incubation in the presence of 5 x 10~ ! M DES, only 20% of cells are capable of colony formation compared to controls but this resistant fraction decreases steadily until after 96 h chronic exposure to DES, colony-forming ability is negligible. The response of malignant cervical keratinocytes to growth-


Measurement of cell survival after drug treatment Feeder layers were removed from HCE cultures after 12 days in culture by spraying with 0.01% EDTA (21). The cultures were then washed with calcium and magnesium-free phosphate-buffered saline and 3T3 conditioned medium was added. These cultures were then exposed to acetone at 0.1 % or to the appropriate drug in 0.1 % acetone. After 24, 48, 72 or 96 h some cultures were disaggregated using TEGPED (22) and replated onto 100-mm dishes (Falcon) in the presence of 2 x 10* XR SW 3T3. The numbers of cells inoculated after each treatment were as follows: acetone controls, 104 cells per dish, PMA-, mezerein- or DEStreated cells, 10s cells per dish. Line CC22 was routinely cultivated in the absence of feeder cells but treatment with PMA, mezerein and DES followed the same protocol as strain AC 69. After exposure to PMA, mezerein or acetone, CC22 cells were inoculated at 103 cells per dish in the presence of feeder cells. Cells exposed to DES were inoculated at 10s or 103 per dish in the presence of feeder cells.

100

Responses of human cervical keratinocytes in vitro

120

100 Cloning Efficiency 80 % Control

100 Cloning Efficiency % Control 80

60

60

40

40

20

24

48 Hours -+•

72

96 24

Fig. 3. The effect of exposure to PMA or DES or mezerein for varying time periods on the cloning efficiency of HCE relative to acetone controls. The results are expressed as a percentage of control values and represent the mean of three experiments ± SEM. Colony-forming efficiency of the acetone control was 3.2 ± 0.4%. Symbols: open circles, mezerein; triangles, DES; squares, PMA

100 Cloning Efficiency % Control 80

24

48 Hours • -

72

Fig. 4. The effect of exposure to mixtures of DES of PMA or DES and mezerein or mezerein and PMA for varying time periods on the cloning efficiency of normal HCE relative to acetone controls. The results arc expressed as a percentage of control values and represent the means of eight replicates from two experiments ± SEM. Colony-forming efficiency of the acetone controls was 3.2 ± 0.4%. Symbols: closed squares, PMA; open squares, PMA + mezerein; circles, PMA + DES; triangles, DES + mezerein.

inhibitory levels of PMA, mezerein and DES were examined and the results are shown in Figure 5. In contrast to the effects on normal HCE, CC22 cells exhibited only a 10-15% loss in colony-forming efficiency after exposure to 10~6 PMA. This large resistant fraction was not decreased even after 96 h chronic exposure to PMA. The response of CC22 cells to 10~6 M mezerein was the reverse of that exhibited by normal HCE. After 24 h there was no decrease in colony-forming efficiency com-

48

72

96

Hours * • Fig. 5. The effect of PMA or mezerein or DES on the cloning efficiency of the malignant keratinocyte line CC22 relative to acetone controls The results are expressed as a percentage of control values and represent the mean of 12 replicates from two experiments ± SEM. The colony-forming efficiency of the acetone controls was 14 ± 1.6%. Symbols: open circles, DES; closed circles, PMA; triangles, mezerein.

pared to controls and this response was not modified even after 96 h chronic exposure to mezerein. The response of CC22 cells to 5 x 1 0 - ' M DES differed slightly from that of normal HCE, in that the time-course response to DES exhibited by CC22 was slower than that shown by AC 69. After 24 h exposure to DES, colony-forming efficiency had declined to 50% of the acetone controls. Colony-forming ability declined steadily in the presence of DES and after 96 h only a tiny fraction of cells, ~ 1 %, retained the ability to form viable colonies. However, this fraction of cells remained even after prolonged exposure to DES (at least 14 days), and colonies would form when the drug was removed, an observation comparable to that described for MCF 7 cells (24). No fraction of normal HCE cells survived a 96-h exposure to 5 x 10~5 M DES. The data displayed in Figures 4 and 6 show the results of experiments in which AC 29 cells or CC22 cells were grown in the presence of mixtures of PMA, mezerein and DES. The PMAresistant fraction, (PMAR), is suppressed in normal HCE cells in the presence of either mezerein or DES. In experiments with CC22 cells, survival after exposure to equimolar concentrations of PMA and mezerein was not significantly different from when the cultures were treated with the compounds alone. However, the presence of DES in the medium always resulted in the loss of colony-forming ability. Discussion The results shown in Figures 1 and 2 show that chronic exposure of normal human cervical keratinocytes to high concentrations of PMA or mezerein (10~6 M) or DES (5 x 10~ s M) results in the inhibition of growth and the induction of terminal differentiation in these cells. With the first two compounds the in vitro responses of normal and malignant cervical keratinocytes are similar to those described for normal and transformed human 1013


20

M.A. Stanley et al.

120

100 Cloning Efficiency °*o Control

60

40

20

48 Hours-*-

72

96

Fig. 6. The effect of mixtures of PMA and DES or PMA and mezerein or DES and mezerein on the cloning efficiency of CC22 cells relative to acetone controls after varying exposure times. The results are expressed as a percentage of the control values and represent the mean of 12 replicates from two experiments ± SEM. The cloning efficiency of the acetone controls was 14 ± 1.6%. Symbols: open circles, DES + mezerein; closed circles, DES + PMA; triangles, PMA + mezerein.

and mouse keratinocytes (16). Normal HCE are induced to terminally differentiate after exposure to PMA but a subpopulation of these cells (PMAR) exhibits resistance to the PMA-induced terminal differentiation even at the very high dose of 10~5 M (Figure 1). Mezerein is growth inhibitory to normal HCE and induces terminal differentiation but no subpopulation of mezereinresistant keratinocytes is apparent. However, the malignant cervical line, CC22, consists almost entirely of cells resistant to PMA. Furthermore, this cell line is resistant to^the growthinhibitory effects of mezerein and the cloning efficiency of these cells after exposure to 10~6 M mezerein is not significantly different from the acetone controls. In the two-stage (initiation-promotion) system of carcinogenesis in mouse epidermis, PMA is a potent complete promoter (25) but mezerein is a weak promoter although it can act as an effective second-stage promoter if applied after PMA in a two-stage promotion system (26,27). In epidermis, initiation is essentially irreversible (28,29) suggesting that either the target cell is a stem cell or that the initiation event alters committed cells so that they can self-renew and function as stem cells. Keratinocytes cultured using 3T3 feeder layers can be grafted into nude mice (30) or humans (31) to form a structurally normal epidermis and hence these cultures must contain self-renewing cells capable of epidermal regeneration, i.e., stem cells. When treated with PMA most epidermal keratinocytes lose the capacity to divide and proceed to terminally differentiate and it follows that the initiated cells (and the stem cells) should resist these changes. Indeed such resistant changes have been described in cultures of mouse (16,32) and human (15,16) epidermal keratinocytes (PMAR) and several lines of evidence suggest that the PMAR cells are less committed to terminal differentiation than the other dividing keratinocytes (15,16). In epidermis there is persuasive evidence that tumour initia1014


24

tion results in defective commitment of the keratinocytes towards terminal differentiation (33) and that PMA in part achieves tumour promotion by selecting for cells which have this defect (15,16,33). This would occur because in clones of initiated cells, the less-committed PMAR cells would have an increased tendency to self-renew instead of becoming committed to terminal differentiation and as a consequence PMA-sensitive. The greater fraction of PMAR cells in initiated clones would clearly give them a selective advantage (15,16,33). The first stage of tumour promotion is longer lasting than the second stage (34,35) but is dependent upon the wave of DNA synthesis which follows PMA treatment (35 — 37) indicating that some clonal expansion of the initiated population is important even in the early stages of promotion. The reduced reversibility of the first stage in promotion could be based in part on the observation that whilst the first treatment of the epidermis results in an initial reduction in DNA synthesis and considerable cell loss as a result of differentiation (38), subsequent treatments do not (39). This would result in the survivors of the first treatment (including the initiated cells) having considerable space in which to expand but in the case of subsequent treatments clonal expansion of the initiated population would be restricted largely to the upward direction where space is created as a result of cell loss due to desquamation. In the current study the reduced response to PMA-induced terminal differentiation shown by the malignant cervical line CC22 when compared with normal HCE indicates that similar selective mechanisms might be operational in cervical squamous epithelium. In both epidermal (16) and cervical (Figures 1 and 3) keratinocytes, the PMAR cells, like initiated cells in vivo (40), are sensitive to the second-stage promoter mezerein but become resistant to this compound in cultures derived from benign or malignant (16, Figure 4) lesions where the transformed phenotype is expressed. The inefficiency of mezerein as a complete promoter might therefore be the result of its ability to induce terminal differentiation in a subpopulation of keratinocytes which contains the initiated cells (16,40) but which in contrast resist the effects of PMA (Figure 1). The acquisition of resistance to mezerein by CC22 (Figure 4) and other transformed human and mouse keratinocyte lines (16) could well have relevance to the observation that mezerein is a much better second-stage promoter than a complete promoter in certain strains of mice (26,27). Both PMA and mezerein would be capable of giving a selective advantage to both benign and malignant keratinocytes over their normal counterparts (Figures 1,4) and (15,16), whereas DES which has been identified as a 'promoter' principally from in vitro studies (10,11) would be much less effective (cf. Figures 1 and 4). The oxidative metabolism of DES mediated by prostaglandin synthetase results in the formation of reactive radical species such as phenoxyradical (41,42). The improved ability of initiated cells to scavenge free radicals (S. Yuspa, personal communication) might be responsible for the slower rate of death in line CC22 compared to HCE. The data presented in Figures 3—6 suggest that the mechanisms by which DES inhibits growth and induces cornified-envelope formation are distinct from those activated by PMA or mezerein as they are retained by the malignant cervical keratinocytes. The data also suggest that DES is unlikely to give transformed keratinocytes a great selective advantage over their normal counterparts and as a result is unlikely to operate as a very effective second-stage promoter although its ability to interfere with cell division (7,24,40) and generate free radicals (41,42) might make it a candidate for an effective first-stage promoter.

Responses of human cervical keratinocytes in vitro

Acknowledgements This work was supported by a grant from the Cancer Research Campaign.

References

Received on 11 January 1985; accepted on I April 1985

1015


1. Herbst.A.L., Ulfelder.H. and Poskanzer.D.C. (1971), Adenocarcinoma of the vagina: association of maternal stilboestrol therapy with tumour appearances in young women. New Engl. J. Med., 284, 878-881. 2. Robboy.S.J., Young.R.H., Welch.W.R., Truslow.G.Y., PraU., HerbsJ.A.L. and Scully,R.E. (1984), Atypical vaginal adenosis and cervical ectropion. Association with clear cell adenocarcinoma in diethylstilboestrol-exposed offspring. Cancer, 54, 869-875. 3. Robboy.S.J., Young.R.H. and HerbsJ.A.L. (1981), Female genital tract changes related to prenatal dtethylstilboestrol exposure, in Blaustein.A. (ed.), Pathology of the Female Genital Tract, Springer Verlag, New York, Heidelberg, Berlin, pp. 99-118. 4. McLachlanJ.A., Newbold.R.R. and Bullock,B.C. (1980), Long term effects on the female mouse genital tract associated with prenatal exposure to diethylstilboestrol. Cancer Res., 40, 3988-3999. 5. ForsbergJ.G. and Kalland.T. (1981), Neonatal estrogen treatment and epithelial abnormalities in the cervico-vaginal epithelium of adult mice, Cancer Res., 41, 721-734. 6. Glatt.H.R., Metzler.M. and Oesh.F. (1979), Diethylstilboestrol and 11 derivatives. A mutageniciry study with Salmonella typhimurium, Musat. Res., 67, 113-121. 7. Tsutsui.T., Maizumi.H., McLachlanJ.A. and BarretU-C. (1983), Aneuptoidy induction and cell transformation by diethylstilboestrol: a possible chromosomal mechanism in carcinogenesis, Cancer Res., 43, 3814-3821. 8. Chrisman.C.L. (1974), Aneuploidy in mouse embryos induced by diethylstilboestrol diphosphate. Teratology, 9, 229-232. 9. Ishidate.M. Jr. and Odashima.S. (1977), Chromosome tests with 134 compounds on Chinese hamster cells in vitro: a screening for chemical carcinogens, Mutal. Res., 48, 337-354. 10. Kinsella.A.R. (1982), Elimination of metabolic co-operation and the induction of sister chromatid exchanges are not properties common to all promoting or co-carcinogenic agents, Carcinogenesis, 3, 499-503. 11. LillehaugJ.R. and Djurhuus (1982), Effect of diethylstilboestrol on the transformable mouse embryo fibroblast C3H/10T1/2C18 cells. Tumour promotion, cell growth, DNA synthesis and omithine decarboxylase, Carcinogenesis, 3, 797-799. 12. SiegfriedJ.M., Nelson,K.G., MartmJ.L. and Kaufmann.D.G. (1984), Promotional effect of diethylstilboestrol on human endometrial stromal cells pretreated with a direct acting carcinogen, Carcinogenesis, 5, 641-646. 13. Johnstone.S.R.D., Stanley,M.A. and QuigleyJ.P. (1984), Trie effect of diethylstilboestrol on the in vitro growth of human ectocervical cells, Carcinogenesis, 5, 741-745. 14. Parkinson,E.K. and Emmerson.A. (1982), The effects of tumour promoters on the multiplication and morphology of cultured human epidermal keratinocytes, Carcinogenesis, 3, 525-531. 15 Parkinson,E.K., Grabham.P. and Emmerson.A. (1983), A subpopulation of cultured human keratinocytes which is resistant to the induction of terminal differentiation related changes by phorbol-12-myristate-13-acetate: evidence for an increase in the resistant population following transformation, Carcinogenesis, 4, 857-861. 16. Parkinson,E.K., Pera.M.F., Emmerson.A. and Gormann.P.A. (1984), Differential effects of complete and second stage tumour promoters in normal but not transformed human and mouse keratinocytes, Carcinogenesis, 5, 1071-1077. 17. RheinwakU.G. (1980), Serial cultivation of normal human epidermal keratinocytes, in Harris.C.C, Trump.B.F. and Stoner.G.D. (eds.), Methods in Cell Biology, Vol. 21, Normal Human Tissue and Cell Culture, A. Respiratory, Cardiovascular and Integumentary Systems, Academic Press, New York, pp. 229-254. 18. Stanley,M.A. and Parkinson,E.K. (1979), Growth requirements of human cervical cells in culture, Int. J. Cancer, 24, 407-414. 19. Stoker.M.G.P. and MacPherson.I. (1961), Studies on transformation of hamster cells by polyoma virus in vitro. Virology, 14, 359-370. 20. Melnick,J.L., Lewis,R., Wimberiy.I., Kaufman,R.H. and Adam.E. (1978), Association of cytomegalovirus (CMV) infection with cervical carcinoma: Isolation of CMV from cell cultures derived from cervical biopsy, Intervirology, 10, 115-119. 21. Rheinwald.J.G. and Green,H. (1975), Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinising colonies from single cells, Cell, 6, 331-343.

22. Chang.S.E., Kcen.J.E., Lane.E.B. and Taylor-PapadimitriouJ. (1982), Establishment and characterisation of SV40-transformed human breast epithelial cell lines, Cancer Res., 42, 2040-2053. 23. Green.H. (1977), Terminal differentiation of cultured human epidermal cells, Cell, 11, 405-416. 24. Brandes.L.J. and Hermonat.M.W. (1983), Receptor status and subsequent sensitivity of subclones of MCF-7 human breast carcinoma cells surviving exposure to diethylstilboestrol, Cancer Res., 43, 2381-2385. 25. Scnb'nerJ.D. and Suss.R. (1978), Tumour initiation and promotion, in Richter.G.W. and Epstein.M.A. (eds.). International Review of Experimental Pathology, Vol. 18, Academic Press, New York, pp. 137-198. 26. Furstenberger.G., Berry,D.L., Sorg,B. and Marks.F. (1981), Skin tumor promotion by phorbol esters is a two-stage process, Proc. Nail. Acad. Sci. USA, 78, 7722-7726. 27. Slaga.T.J., Klein-Szanto,AJ.P., Fischer.S.M., Weeks.C.E., Nelson.K. and Major.S. (1980), Studies on mechanism of action of anti-tumor agentspromoting agents: their specificity in two-stage promotion, Proc. Nail. Acad. Sci. USA, 77, 2251-2254. 28. Loehrke,H., Schweizer.J., Dedener.E., Hesse,B., Rosenkranz.G. and Goertier.K. (1983), On the persistence of tumour initiation in two-stage carcinogenesis in mouse skin, Carcinogenesis, 4, 771-775. 29. Van Duuren.B.L., Sivak.A., Katz.C, Seidman.I. and Melchionne.S. (1975), The effect of aging and interval between primary and secondary treatment in two-stage carcinogenesis in mouse skin, Cancer Res., 35, 502-505. 30. Banks-Schlegel.S. and Green,H. (1980), Formation of epidermis by serially cultivated human epidermal cells transplanted as an epithelium to athymic mice. Transplantation, 29, 308-313. 31. O'Connor.N.E., MullikinJ.B., Banks-Schlegel,S., Kehinde.O. and Green,H. (1981), Grafting of bums with cultured epithelium prepared from autologous epidermal cells, Lancet, 1, 75-78. 32. Yuspa.S.H., Ben,T., Hennings,H. and Uchti,U. (1982), Divergent responses in epidermal basal cells exposed to the tumor promoter 12-O-tetradecanoylphorbol-13-acetate, Cancer Res., 42, 2344-2349. 33. Yuspa.S.H. and Morgan,D.L. (1981), Mouse skin cells resistant to terminal differentiation associated with initiation of carcinogenesis, Nature, 293, 72-74. 34. Furstenberger.G., Sorg.B. and Marks,F. (1983), Tumor promotion in skin: evidence for a memory effect, Science (Wash.), 220, 89-91. 35. Slaga.T.J., Fischer.S.M., Nelson,K. and Gleason,G.L. (1980), Studies on the mechanism of skin tumor promotion: evidence for several stages in promotion, Proc. Nail. Acad. Sci. USA, 77, 3659-3663. 36. Kinzel.V., Loehrke.H., Goertlcr.K., Furstenberger.G. and Marks.F. (1984), Suppression of the first stage of phorbol 12-tetradecanate-13-ace«ate effected tumor promotion in mouse skin by non-toxic inhibition of DNA synthesis, Proc. Nail. Acad. Sci. USA, 81, 5858-5862. 37. Furstenberger.G., SchweizerJ. and Marks.F. (1985), Development of phorbol ester responsiveness in neonatal mouse epidermis: correlation between hyperplastic response and sensitivity to first-stage promotion, Carcinogenesis, 6, 289-294. 38. Argyris.T.S. (1980), Epidermal growth following a single application of 12-Otetradecanoyl-phorbol-13-acetate in mice, Am. J. Pathol, 98, 639-648. 39.Raick,A.N., Thumm.K. and Chivers.B.R. (1972), Early effects of 12-Otetradecanoyl-phorbol-13-acetate on the incorporation of tritiated precursors in DNA and the thickness of the interfollicular epidermis and their relation to tumor promotion in mouse skin, Cancer Res., 32, 1562-1568. 40. Argyris.T.S. (1983), Nature of the epidermal hyperplasia produced by mezerein, a weak tumor promoter, in initiated skin of mice, Cancer Res., 43, 1768-1773. 41. Degen.G.H., Eling,T.E. and McLachlanJ. (1982), Oxidative metabolism of diethylstilboestrol by prostaglandin synthetase, Cancer Res., 42, 919-923. 42. Tsutsui.T., Degen.G.H., Schiffmann.D., Wong.A., Maizumi,H., McLachIan.J.A. and BarrettJ.C. (1984), Dependence on exogenous metabolic activation for the induction of unscheduled DNA synthesis in Syrian hamster embryo cells by diethylstilboestrol and related compounds, Cancer Res., 44, 184-189. 43.Sawada,M. and Ishidate.M.,Jr. (1978), Colchicine-like effect of diethylstilboestrol (DES) on mammalian cells in vitro, Mutat. Res., 57, 175-182.


L