Neoplastic Transformation of Human Epithelial Cells in ... · agar prepared in Roswell Park Memorial Institute Medium 1629 (Grand Island Biological Co., Grand Island, N. Y.), supplemented

[CANCER RESEARCH 41, 5096-5102, December 1981]0008-5472/81/0041-OOOOS02.00

Neoplastic Transformation of Human Epithelial Cells in Vitro afterExposure to Chemical Carcinogens1

George E. Milo,2 Inge Noyes, John Donahoe, and Steven Weisbrode

Department ol Physiological Chemistry and Comprehensive Cancer Center [G. E. M.. I. N.], Department of Veterinary Pathobiology ¡S.W.J, and Department ofVeterinary Preventive Medicine ¡J.D.I. The Ohio State University. Columbus. Ohio 43210

ABSTRACT

Human foreskin epithelial cells were transformed to an anchorage-independent state of growth (in soft agar) and neoplasia (invasion of chick embryonic skin in vitro). Aflatoxin B,, N-methyl-A/'-nitro-/V-nitrosoguanidine, propanesultone, /3-propi-

olactone, or ultraviolet absorbance at 254 nm were used successfully as carcinogens. These foreskin epithelial cells, likehuman foreskin fibroblasts, were readily transformed whentreated in S phase but, unlike the transformed fibroblasts,expression of cellular neoplasia did not require an extendedperiod of time in culture. The invasive features of the transformed human epithelial cells in chick embryonic skin in vitrosimulated squamous cell carcinoma.

INTRODUCTION

Although there are many reports of transformations of humandiploid fibroblast cells in vitro using chemical (1, 6, 10, 12, 25),physical (1, 9, 14, 18, 25, 28), or viral (16) carcinogens, weknow of no detailed reports on the successful reproducibletransformation of human epithelial cells by carcinogens. Theunique feature of our procedure for transformation is basedupon the ability to grow epithelial cells in vitro, free of detectable contaminating fibroblasts, but on the same substrata onwhich the fibroblasts are propagated. There are many reportsin the literature describing how to grow human epithelial cellson substrata that restrict fibroblast growth, such as pig collagen(2) rat tail collagen (7, 8, 22), or in the presence of lethallyirradiated 3T3 mouse cell populations (22, 24). However, noneof these reports discuss the procedures for transforming human epithelial cells. In a previous report (12), we demonstratedthat it was important to treat the fibroblasts in the S phase ofthe cell cycle in order to reproducibly induce a carcinogenicevent. We have recently presented evidence for the transformation of human epithelial cell populations (14) using theconcept of treatment of the cells in S phase. We now reportthat low-passage epithelial cells (11 ) from many different fore

skins can be neoplastically transformed in vitro following carcinogenic treatment.

MATERIALS AND METHODS

Cell Culture. Reproducible in vitro cultivation of normal proliferatinghuman epithelial cells has been difficult to achieve with the presentmethodologies, and the success rate has been low (2, 7). Recentreports (20, 21 ) have described a procedure for reproducibly growing

human newborn foreskin epithelial cells on lethally irradiated fibroblast3T3 feeder layers. We recently reported an alternate method forsuccessful reproducible establishment of proliferating human foreskinepithelial cells in vitro, free of detectable fibroblasts (11, 20), andwithout a requirement for growth on irradiated feeder layers or epidermal growth factor.

Human foreskins were minced into 2-mm segments in CM.3 The

tissue was rinsed 3 times in this medium, and the tissue fragmentswere transferred to 20 ml of CM supplemented with 20% FBS (SterileSystems, Logan, Utah) and containing 0.25% collagenase (115 unitslot no. 4197 CLS per mg; Worthington Biochemical Corp., Freehold,N. J.). Enzymatic dispersion of the tissue was completed at 37° in a

4% COj-enriched air atmosphere for 6 to 8 hr. The cells were recoveredby centrifugation at 150 x g for 7 min at 12°. The cell pellet was

washed twice, and the final pellet was resuspended in CM. The cellsuspension was seeded into a 75-sq cm flask. The cell populations

were allowed to attach for 72 hr and were refed once.On Day 5 following seeding, the fibroblasts were selectively de

tached with trypsin (20). To ensure that these cultures were free offibroblasts, the selective detachment was repeated as necessary, andthe pure epithelial cultures were stained for the presence of keratin(22, 24, 26). These epithelial cultures were refed with CM supplemented with 20% FBS. The epithelial cell cultures in a state of growth(Phase II) described in the papers of Karasek and Su-Chin (8) and Su-

Chin et al. (27) were used for carcinogen treatment.Suspension of Chemical Carcinogens and Potentiating Agent. In

the treatment process of adding either the carcinogen(s) or sensitizingagent to the cells, these chemicals were solvated in a specific manner.Insulin was used for these studies at a concentration of 1.0 unit/ml(Sigma Chemical Co., St. Louis, Mo.; No. I-5500). The stock solutions

were dissolved in water and glacial acetic acid (20:1, v/v), with a finalpH of 4.5 at 100 units/ml, and were made up fresh for each experiment.The chemical carcinogens MNNG, PRS, ß-PL,and AFB, were prepared

as described previously (12, 15). The stock solutions were made up inspectra-grade acetone, and the working solutions were made up in

CM. These solutions were then added immediately to the experimentalmedium over the cells.

U.V. Treatment at 254 nm. When UV was used to transform thecells, an exposure value of 5 J/sq m at a fluence of 1.2 J/sq m/secwas used to treat the cells.

Cytotoxicity Characterization. The procedures used to determinethe toxic concentrations of chemical carcinogens suitable for use inthe fibroblast transformation system (21) were not suitable when applied to epithelial cell cultures. For example, a 50% cytotoxic concentration of PRS (18.0 /¿g/ml)on fibroblasts when applied to epithelialcells killed all the epithelial cells.

The final toxic concentration of chemical carcinogens suitable foruse on the epithelial cell cultures was determined by the toxic sensitivityof the cells to the compounds. The toxic response was measured inthe following manner. The highest concentration of the chemical car-

1This work was supported in part by National Cancer Institute Grant R01 -CA-

25907, EPA R-806638.2 To whom requests for reprints should be addressed, at OSU Comprehensive

Cancer Center. 410 West 12th Avenue. Columbus, Ohio 43210.Received December 22. 1980: accepted September 14, 1981.

3 The abbreviations used are: CM. minimum essential medium-Hanks' bal

anced salt medium containing 25 rriM 4-(2-hydroxethyl)-1-piperazineethanesul-fonic acid buffer, pH 7.2: FBS, fetal bovine serum; MNNG, N-methyl-W'-nitro-fV-nitrosoguanidine: PRS, propanesultone; ß-PL,/ì-propiolactone; AFBi, aflatoxinBi; DM. Dulbecco's modified minimum essential medium; CES, chick embryonic

skin; PDL, population doubling.

5096 CANCER RESEARCH VOL. 41

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Neoplastia Transformation

cinogen or strongest dose of UV at 254 nm that did not lead todetachment of the epithelial colonies from the plastic substratum following treatment was designated as a nontoxic carcinogenic dose usedto treat the cells in the transformation protocol.

Preparation of Epithelial Cell Populations for Passage through SPhase. To reproducibly transform human epithelial cells, we treatedcells from 12-day-old cultures containing 150 colonies/25 sq cm

(250,000 cells) while a large proportion of the cells were in S phase.Due to the low number of randomly proliferating cells in S phase at anytime over the 24-hr cell cycle, it was necessary to increase the number

of cells going through the cell cycle at any one time. We observed thatthere was less than 1 cell/5 x 10s cells in S phase over a 24-hr period

(14) in a randomly proliferating population.When the CM medium was replaced with DM minus arginine and

glutamine, containing 1.0 unit of insulin per ml supplemented with 10%dialyzed FBS (15), the cells left G and entered S. Immediately uponreplacement of CM with DM, [mef/7y/-3H]thymidine (New England Nu

clear, Boston, Mass.; 60 Ci/mmol) at 1 /iCi/ml was added to themedium. Populations were removed from the labeling medium at 1-hrintervals over a 24-hr period. The first population was removed at 30

min. A slight modification was also introduced in order to follow thepopulations through the transformation protocol (described below); DMwas used to replace CM for 30 min and then replaced again with CM.The addition of the 3H on the methyl moiety was the same as described

above.Transformation Protocol Treatment of Cells in S Phase with

Carcinogen. Two hr into S phase, the DM medium was replaced withCM plus 10% FBS and 1.0 unit of insulin per ml, containing eitherMNNG (0.4 fig/ml), PRS (7.5 mg/ml), /?-PL (7.5 /ig/ml), or AFB,, (2.5

jig/ml) (Chart 1). The carcinogen medium was removed 6 hr afterinitiating carcinogen treatment. When UV irradiation at 254 nm wasused to transform the cells, we discontinued the exposure after 4.2sec and replaced the medium as described above for chemical carcinogens (9). When UV treatment was used CM did not contain phenolred (9). At the conclusion of the carcinogen (Chart 1) treatment period,the cells were washed 3 times with CM plus 20% FBS and incubatedfor 2 to 3 weeks until the epithelial colonies grew to a confluent density.The confluent cultures were rinsed once with Dulbecco's phosphate-

buffered saline minus calcium and magnesium and serially passaged1:2, using a solution of 0.25% trypsin and 0.02% Versene.

Growth in Soft Agar. After the treated cultures were serially passaged twice, they were then seeded at a cell density of 50,000/25 sqcm into 2 ml of 0.33% agar prepared in Dulbecco's LoCal medium (Bio

Labs, Northwood, III.) supplemented with 1 rriM sodium pyruvate, 1 xnonessential amino acids, 1 x essential amino acids, 2 mw glutamine,1 X vitamins, 0.2% sodium bicarbonate, 5.0 fig gentamicin per ml, and20% FBS (15). This agar preparation was layered over a base of 2%agar prepared in Roswell Park Memorial Institute Medium 1629 (GrandIsland Biological Co., Grand Island, N. Y.), supplemented in the samemanner as the 0.33% overlayer. The cultures were examined for colonyformation weekly, and they were considered negative when no growthwas seen after 4 weeks. The colonies were removed from soft agarand seeded onto a plastic substratum of a 75-sq cm flask. The cultureswere fed with CM plus 20% FBS (see "Cell Culture").

Rhodanile Blue Stain. Epithelial populations from the epidermiswere stained for cornification as described by Sisskin and Barrett (26).The cells were fixed in 10% buffered formate, and a 1% preparation ofRhodanile blue (Aldrich Chemical Co., Milwaukee, Wis.) was added tothe culture. After 30 min, the cells were rinsed with several washes ofwater until no pink color was apparent. The cultures were then examined under light microscopy.

Assay for Cellular Neoplasia. The carcinogen-treated cell populations at Passage 6 to 7 were seeded onto CES (organ culture) to assayfor cellular neoplasia (19). This procedure was used in lieu of the nudemouse (3). However, tumors were observed in nude mice (13, 15).

The CES organ culture (5, 19) was modified in the following mannerto optimize the sensitivity and frequency of success for a rapid assay

of cellular neoplasia for carcinogen-transformed epithelial cells. Eggs

were incubated for 9 to 10 days in a humidified egg incubator (HumidairIncubator Co., New Madison, Ohio). The embryos were removed fromthe eggs, and the skins were separated from the dorsal part of theembryo and placed on previously prepared agar bases. The agar basecontained 10 parts of 1% agar (Bacto-Agar) in Earle's balanced salt

solution without sodium bicarbonate, 4 parts of FBS, and 4 parts ofchick embryo extract. The chick embryo extract was prepared from10-day-old chick embryos by high-speed mincing and then used in the

agar base. Organ culture grids (60 mesh stainless steel; Falcon Plastics, Oxnard, Calif.) were placed in 35-mm-diameter plastic dishes, and

then enriched agar base was poured over the grids to cover them to adepth of 1 to 2 mm. A 6- to 8-mm-diameter section of skin was layered

on the medium immediately over the wire grid with the dermis side up.A sterile glass ring 2 mm thick and 8 mm in diameter was laid over theskin. Then 105 cells from the treated and untreated cultures were

suspended in 0.025 ml of CM plus 20% FBS and seeded into theserings. Seven 35-mm closed agar dishes were set in a 150-mm-diameterFalcon Petri dish. One 60-mm open Petri plate containing steriledistilled water was placed in each 150-mm plate in order to ensure ahumidified atmosphere. This 150-mm Petri dish was then covered andplaced in a 4% CO2 humidified atmosphere at 37°.On Day 2 following

seeding, the cultures were fed an additional 0.025 ml of CM plus 20%FBS. On Day 4, the skins were removed and fixed in Bouin's solution.

Fixed skins were cut in half and embedded in paraffin with cut edgestoward the face of the block. Five-jum transverse sections were made,

stained with hematoxylin and eosin, and examined by light microscopy.

RESULTS

As in carcinogen treatment of randomly proliferating companion fibroblast cultures (12, 15), we rarely observed a transformation event in carcinogen-treated randomly proliferating

epithelial cells. To successfully and reproducibly transformhuman foreskin epithelial cell populations, it was necessary totreat the cells while they were in S phase (Chart 1, Induction 6hr). Due to the low rate of proliferation of randomly growingepithelial cells compared to the rapidly proliferating (26% ra-diolabeled nuclei; labeling index) low-passage fibroblasts, weselected alternate ways to accelerate the process by whichepithelial cells would traverse from G, to S phase. Continuousradiolabeling of the epithelial or fibroblasts with [3H]thymidine

over a 24-hr period revealed that 83 to 90% of the fibroblast

populations were radiolabeled (12), while 1% of the epithelialcell nuclei were radiolabeled (Fig. 1/0.

To increase the number of cells traversing through S phase,the epithelial cell population was transferred to DM, minusarginine and glutamine, and 10% dialyzed FBS. Radiolabelingof these epithelial cells after 0.5 to 1.0 hr in DM followed byreplacement of DM with CM over a 4.0-hr period revealed that56% of the cells were in S phase (Fig. IB).

Induction of the carcinogenic process in the epithelial cellswas initiated when the cells were in S phase. The carcinogenswere added to the cells in S phase, and the carcinogenicresponse was measured as a function of their ability to expressanchorage-independent growth and invasion of CES. Exposure

of the cells to UV at 254 nm was accomplished during S phasealso. To avoid excessive cytotoxicity because the colonies ofepithelial cells responded to the agents as a unit rather than asindividual cells, as demonstrated by the fibroblast response totoxic concentrations, we determined doses according to theinability of the cells to detach from the substratum when treatedwith the carcinogen. This dose was identified as the carcino-

DECEMBER 1981 5097

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

G. E. Milo et al.

Pf«induefiOfMEM-Arg-Glut

MEDIUM2

hrInduri"^MEM*IO%FBS6 hrPov-TrtotmtntCM

* 10%FBS3

weeksAnchoraqt

IndcptrtdcntGrowth2-3w«tk*orN«oplosro

Evaluation onCES3

doy*

Chart 1. This sequence graphically represents the events as they occur inhuman epithelial cell carcinogenesis as described in the text. There was asingular modification that occurs when the cells were irradiated with UV at 254nm; the cells were exposed for only a few sec to UV (see text). Followingirradiation, the experimental medium was removed, and the posttreatment stageand other stages remained the same in reference to time. MEM. minimumessential medium: Arg. arginine; Glut., glutamina.

genie dose. We suspect that the difference in sensitivity of the2 cell types to the carcinogens could, in part, be due tometabolism or permeability of the carcinogen. For example,foreskin epithelial cells bioxygenate in excess of 83%benzo(a)pyrene to toxic oxygenated metabolites, phenols(17), while transformable fibroblasts bioxygenate only 2%benzo(a)pyrene over a 24-hr period (29). Therefore, the toxicityvalues for the compound(s) on foreskin epithelial cells or fibroblasts were different.

We therefore selected the 20% cytotoxic concentration asdetermined by inhibition of fibroblast proliferation (21) as acarcinogenic dose to be used to treat the epithelial cell populations: AFB,, 2.5 fig/ml; MNNG, 0.4 /iig/ml; PRS, 7.5 jag/ml;ß-PL,7.5 jug/ml. For UV at 254 nm, we exposed the cells to 5J/sq m. If we exceeded this concentration or exposure to thecarcinogen by 25%, the cells died. After the experimentalmedium was removed, these cultures were incubated in CMfor 2 to 3 weeks and then serially passaged twice at a 1:2 splitratio.

The cells were then seeded into soft agar (0.33%) to evaluateanchorage-independent growth (16) or onto CES to evaluate

cellular neoplasia (19). The number of colonies that formed insoft agar following carcinogen treatment was: PRS-trans-formed cells, 58/106 cells; MNNG-transformed cells, 32/10scells; AFB,, 41 /106 cells; ß-PL,23/10s cells; cells transformedby UV irradiation at 254 nm, 100/105 seeded cells (Table 1).

Growth in soft agar (anchorage-independent growth) has been

repeated at least twice with similar results as reported above.Examination of the colonies that were 50 cells/colony orgreater in number in soft agar by Rhodanile blue staining (22,24) revealed that none of these colonies contained cornifiedenvelopes that stained pink. The transformed cells seeded inculture at 250,000 cells/75 sq cm yield 6 to 8 x 106 cells in5 days, while normal cells yield 4 to 6 x 106 cells in 14 days

(11). These transformed populations do not make keratin asdetermined by Rhodanile blue stain (22, 24); when reseededin soft agar, they produced colonies. They exhibit an extendedlife span of 80 to 120 PDL (Table 2), and preliminary evidenceindicates that they are tumorigenic in nude mice (13, 14).Control cultures ceased proliferating at 20 ±3 (S.D.) PDL (11)and produce no tumors in mice. Each of the above coloniesexhibiting anchorage independence was checked for tumorformation on CES.

In all of the specimens observed to date, the following treatedpopulations were invasive out of the number seeded on CES:7 of 9 aflatoxin-treated populations; 2 of 3 PRS-treated populations; 7 of 16 /3-PL-treated populations; 1 of 1 MNNG-treatedpopulations; and 2 of 2 UV-treated populations. In 11 different

control cell populations, we observed only one population that

Table 1Colony formation capability in 0.33% agar of control and carcinogen-treated

epithelial cell populationsAnchorage-independent growth was measured following completion of treat

ment of primary Phase II foreskin populations.

Incidence ofNo. that colony for-

grew in soft mation/105Carcinogen" Dose POL" agarc cells"

AFB,MNNGPRSß-PLUV

at 254 nm'2.5ng/ml0.4

fig/ml7.5(ig/ml7.5

jig/ml5.0J/sq m999922233741325823100

* Each carcinogen was applied to the cells as described in the text. UV

treatment at 254 nm was accomplished as described in the text." Each cell population upon completion of the treatment was serially passaged

1:2 after 2 to 3 weeks following removal of the carcinogen. These cultures wereserially passaged twice and then seeded in soft 0.33% agar over a 2% agar base(see text).

c At the present time, we have successfully transformed 2 to 7 different

epithelial cell populations from different foreskin samples." Fifty thousand cells were seeded into 2 ml of 0.33% agar over a 5-ml 2%

agar base in 25-sq cm wells. These cultures were incubated at 37°in a 4% CÛ2-

enriched air atmosphere. These cultures were refed weekly over a 4-weekperiod. The values represent number of colonies containing 50 cells or more percolony per 105 cells seeded in 0.33% agar for one experiment. Repeat experi

ments gave similar results. None of the untreated cell populations grew in softagar (formed colonies).

" UV at 254 nm. a physical carcinogen, was used to treat epithelial cells at a

fluence of 1.2 J/sq m/sec.

Table 2Growth characteristics of untreated and carcinogen-transformed human

keratinocytes

Culturetype'UntreatedMNNGPRSßPLAFB,UV

at 254 nmCell

density (x10e)4-6(14)"6.5-7.5

(5)7.0-8.0(5)6.0-7.2

(5)7.2-8.0(5)6.0-7.4

(5)Life

spanPDL20

±31111058090120

" All treated and untreated populations were treated at a cell density of

250,000 cells (see text)."Numbers in parentheses, days; the cell numbers were obtained 14 days

after seeding for confluent dense cultures of untreated cells and 5 days fortreated cell populations that exhibited a confluent dense culture; n = 8.

c Mean ±S.D.

invaded the CES. In this one case, the invasion was only intothe first layer of cells of the dermis. All of the treated cellpopulations that would not grow in soft agar did not invade theCES. All untreated epithelial cell populations examined to date(10 different tissues), did not invade the chick skin. Histopath-

ologically, the tumor simulated squamous cell carcinoma (Fig.2).

DISCUSSION

Recently, we were able to report on the demonstrated neo-plastic transformation in vitro of early-passage diploid fibroblastcells from human foreskin by several direct-acting chemical

carcinogens (12). Subsequent to that achievement, we wereable to demonstrate neoplastic transformation of foreskin fibroblasts with physical carcinogens (9, 15). Recently, Greiner efal. (4) have been able to induce transformation of human lungfibroblast (MRC-5), using the same concept. Extensive characterization of the carcinogen induction process in foreskinfibroblasts indicated that 16 to 20 PDL had to elapse followingtreatment before anchorage-independent growth was ex-

5098 CANCER RESEARCH VOL. 41

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Neoplastia Transformation

pressed (15). Following growth in soft agar, the transformedcells would then produce tumors when injected s.c. into nudemice (15). However, evaluation of the tumors in the nude micewas undertaken 6 weeks after injection of the cell suspensionsinto the mice. The same time span was necessary beforeintracranial tumors could be evaluated. We found that therewas a 100% correlation between tumor induction, as measuredby tumor formation by transformed fibroblast populations ofMNNG, PRS, AFB,, ß-PL,transformed fibroblast cells in nudemice, and the development of intracranial tumors and growthinto CES. We therefore investigated the use of anchorage-

independent growth and invasiveness of CES by transformedepithelial cells as a measure of their neoplastic potential. Positive control populations of human nasoseptal and testicularcarcinoma cells formed tumors in all 3 different systems. Thesesame 2 cell populations also grew in soft agar; therefore, therewas again a correlation demonstrated between anchorage-

independent growth and tumorigenicity. The chemical carcinogen-transformed foreskin epithelial populations formed colonies in soft agar. Moreover, we did note that the colonies thatcontained 50 or more cells per colony did not cornify. Apparently, Rheinwald and Beckett (23) found that a squamous cellcarcinoma line that grew in soft agar did not form a cornifiedenvelope, while the majority of the squamous cell carcinomapopulations did cornify. They speculate that colony formationin a semisolid medium is unlikely to be useful as an in vitromarker for neoplastic transformation of keratinocytes. In this invitro system, we have found that transformed populations thatform colonies in soft agar of 50 cells or more do not cornifyand invade CES. We found that all (12-17) of the untreated

fibroblast cell lines or epithelial cell lines would not invade theCES in vitro.

Upon intracranial injection of 5 x 105 carcinogen-trans

formed epithelial cells, we have observed aberrant motor functions, including hyperkinetic activity and progressive paralysisof the mice 4 to 5 weeks following injection (3). We followedexplicitly the procedure described by Greiner ef al. (4); if themice were not sacrificed for histopathology, they routinely diedin 6 to 8 weeks. Histopathological examination of the cranialtissue failed to detect presence of transformed human epithelialcells. Mice (15 of them) receiving injections of 5 x 105 normal

cells did not exhibit any changes in neurological function andlived normal spans of 1 to 1.5 years.

The characteristics of growth of the transformed humanepithelial cells into the CES enabled us to observe the invasivefeatures of the transformed epithelial cells during the formationof localized tumors. The tumors simulated squamous cell carcinomas. The invasive features of these cells on CES implymalignancy but are not to be compared to métastases(Fig. 1).We want to convey the fact that these transformed cells containthe ability to form localized growths (tumors) within 3 days onCES. We wish also to indicate that the program for carcinogen-induced transformation of human epithelial cells follows thesame concept as does the chemical carcinogen-induced transformation of human fibroblast cells (12, 15). However, themovement of a wave of cells (56%) through S phase appearedto be enhanced upon replacement of the growth medium withminimum essential medium minus arginine and glutamine. If wesubstituted DM minus arginine and glutamine, we observed asimilar response. Other stress situations appeared to triggerthis phenomenon, such as either exposure to high temperature

(41 °)tor a few min or mechanical injury (using a sharp edge to

injure the colony by cutting through the colony). It appears asthough nonproliferating human keratinocytes in Gìcell arrestwere stopped at a point close to the G,-S interphase and were

stimulated to proceed into S phase when stressed by thedeletion of arginine and glutamine from the growth medium.Moreover, the program for expression of anchorage-independ

ent growth and cellular neoplasia for transformed epithelialcells is not programmed chronologically the same as for transformed fibroblasts.

These carcinogen-treated epithelial cells exhibit an extendedlife span and growth patterns similar in concept to carcinogen-

transformed fibroblast cells; however, they exhibit altered behavior for expression of anchorage-independent growth andtumor formation in nude mice.4 The temporal relationship be

tween growth in soft agar and tumor formation, CES, is notextended (see Chart 1) compared to transformed fibroblast(12, 13). Persistence of 0.8- to 1.5-cm tumors s.c. in the nude

mouse is shortened to 11 to 14 days, while tumors formed fromcarcinogen-transformed fibroblast populations are not evalu

ated for 4 to 6 weeks following injection of the transformed cellpopulation.

Reproducible transformation of human epithelial cells in vitrorequires that the cells are treated in S phase. The expressionof anchorage-independent growth and cellular neoplasia is

expressed within 7 PDL following treatment. The human foreskin epithelial cells as utilized in these studies provide an invitro model to study human carcinogen-induced transformation

of normal epithelial cells to neoplastic cells, in vitro.

REFERENCES

1. Borek, C. X-ray-induced in vitro neoplastic transformation of human diploidcells. Nature (Lond.). 288. 776-778, 1980.

2. Freeman, A., Igel, H., Herrman, B., and Klinfeld, K. Growth and characterization of human skin epithelial cell cultures. In Vitro (Rockville), 12: 352-362, 1980.

3. Giovanella, B., Nilsson, K., Zech, L., Klein, G., and Stehlin, J. Growth ofdiploid, Epstein-Barr virus carrying human lymphoblastoid cell lines hetero-transplanted into nude mice under immunologically privileged conditions.Int. J. Cancer, 24: 103-113, 1979.

4. Greiner. J.. Evans, C., and DiPaolo, J. Carcinogen induced neoplastictransformation of human MRC-5 cells. Carcinogenesis, 2. 359-362, 1980.

5. Johnson, J., O'Donnell, R.. and Noguchi, P. An organ culture tumorigenicity

assay. TCA Man., 4: 857-859, 1979.6. Kakunaga, T. Neoplastic transformation of human diploid fibroblast cells by

chemical carcinogens. Proc. Nati. Acad. Sei. U. S. A., 75: 1334-1338.1978.

7. Karasek, M. In vitro culture of human skin epithelial cells. J. Invest. Derma-tol.. 47: 533-540. 1966.

8. Karasek, M.. and Su-Chin. L. Keratinization of epidermal cells in cell culture.In M. Seiji and I. Bernstein, (eds.), Biochemistry of Cutaneous EpidermalDifferentiation, pp. 336-358. Baltimore: University Park Press. Inc., 1977.

9. McCloskey. J. A., and Milo, G. In vitro transformation of normal diploidhuman cells by UV and X-rays. In: Abstracts of the Fifth Annual Meeting ofthe American Society for Photobiology, San Juan. Puerto Rico. 1977, p.110.

10. McCormick, J., Silinskas, C. K., and Mäher. V. Transformation of diploidhuman fibroblasts by chemical carcinogens. In: P. Tri and H. Gelboin (eds.).Fundamental Mechanisms and Environmental Effects, pp 491-498 Dordrecht, The Netherlands: D. Reidel Publishing Co.. 1980.

11. Milo. G.. Ackerman, A., and Noyes, I. Growth and ultrastructural characterization of proliferating human keratinocytes in vitro without added extrinsicfactors. In Vitro (Rockville), 16: 20-30. 1980.

12. Milo, G., and DiPaolo, J. A. Neoplastic transformation of human diploid cellsin vitro after chemical carcinogen treatment. Nature (Lond.), 275. 130-132,1978.

13. Milo, G., and DiPaolo, J. Presensitization of human cells with extrinsic

' G. E. Milo and J. A. DiPaolo, unpublished data.

DECEMBER 1981 5099

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

G. E. Milo et al.

signals to induced chemical carcinogenesis. Int. J Cancer, 26: 805-812,

1980.14. Milo, G., Noyes, I., Donahoe, J., and Weisbrode. S. Induction of neoplasia

after in vitro exposure of human epithelial cells to carcinogens. Proc. Am.Assoc. Cancer Res., 22. 117, 1981.

15. Milo. G.. Oldham, J., Zimmerman, R., Hatch, G., and Weisbrode, S. Characterization of human cells transformed by chemical and physical carcinogens, in vitro. In Vitro, (Rockville), T7. 719-729, 1981.

16. Milo, G., Olsen. R., Weisbrode. S., and McCloskey, J. Feline sarcoma virusinduced in vitro progression from premalignant to neoplastic transformationof human diploid cells. In Vitro, (Rockville), Õ6:813-822, 1980.

17. Milo, G., Trewyn. R., Tejwani, R., and Oldham, J. Intertissue variation inbenzo(a) pyrene metabolism by human skin, lung and liver, in vitro. In:advisory group for aerospace research and development. NATO AdvisoryGroup for Aerospace Research and Development Conf. Proc. No. 309,Toxic Hazards in Aviation, B7-1 to B7-9, 1980.

18. Namba, N.. Nishitani. K., and Kimolo, T. Carcinogenesis in tissue culture29: neoplastic transformation of a normal human diploid cell strain, WI-38,with Co-60 gamma rays. Jpn. J. Exp. Med., 48: 303-311, 1978.

19. Noguchi, P.. Johnson, J., O'Donnell, R., and Petriciani, J. Chick embryonic

skin as a rapid organ culture assay for cellular neoplasia. Science (Wash. D.C.). ; 99. 980-983, 1978.

20. Noyes, I., Milo. G., and Cunningham, C. Establishment of proliferating humanepithelial cells in vitro from cell suspensions of neonatal foreskin. TCA Man.,5: 1173-1176, 1980.

21. Oldham, J. W., Allred, L. E., Milo. G. E.. Kindig, O., and Capen, C. C. The

toxicological evaluation of the mycotoxins T-2 and T-2 tetraol using normalhuman fibroblasts in vitro. Toxicol. Appi. Pharmacol., 52: 159-168, 1980.

22. Rheinwald. J. Serial cultivation of normal human epidermal keratinocytes.Methods Biol., 21 A: 230-254. 1980.

23 Rheinwald. J., and Beckett. M. Defective terminal differentiation in cultureas a consistent and selectable character of malignant keratinocytes. Cell,22:629-632, 1980.

24. Rheinwald. J. G., and Green. H. Serial cultivation of strains of humanepidermal keratinocytes: the formation of keratinizing colonies from singlecells. Cell, 6: 331-344, 1975.

25. Silinskas, K. C., Kateley. S. A., Tower, J. E., Mäher.V. M., and McCormick.J. J. Induction of anchorage-independent growth in human fibroblasts bypropane sultone. Cancer Res., 41: 1620-1627. 1981.

26. Sisskind, E. E., and Barrett, J. C.. Inhibition of terminal differentiation ofhamster epidermal cells in culture by the phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate. Cancer Res., 41: 593-603, 1981.

27. Su-Chin, G., Eaton, M., and Karasek, M. Growth characterization of humanepidermal keratinocytes from newborn foreskin in primary and serial cultures. In Vitro (Rockville), 15: 813-822. 1979.

28. Sutherland, B. M., Cimino, J. J., Delihas. N., Shin, A., and Oliver, R.Ultraviolet light-induced transformation of human cells to anchorage-independent growth. Cancer Res., 40: 1934-1939, 1980.

29. Tejwani. R., Newnow, S., and Milo, G. Analysis of intracellular distributionand binding of benzo(a)pyrene in human diploid fibroblast. Cancer Lett., 10:576-585, 1980.

5100 CANCER RESEARCH VOL. 41

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Neoplastia Transformation

Fig. 1. In A, the cell population was treatedwith [mefriy/-3H]thymidine for 24 hr, containedin CM (see text), x 420. The major portion ofthe label, as determined by autoradiography,localized in the nuclei of the cells located onthe periphery of the colonies. In B, the cellpopulation, treated with DM for 0.5 to 1 hr,followed by a change to CM containing[mef/iy/-3H]thymidine, was observed for 3.5 hr.x 420. The total elapsed time shown for thesecultures was 4 hr. The percentage of radiola-

beled nuclei was determined by dividing thetotal number of radiolabeled nuclei by the totalnumber of nuclei in a colony x 100.

1A

'~30Llípí:;^~'5*v

'^?T*>»:>-/\\ ^W

% *K^%%- 111 ^

^ . , >^^^p

DECEMBER 1981 5101

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

»•,>'

w$$m¿í

S 2 QI o §

Pi*" i-•°§'Sto g S»è2.O^Ss

i 2 g co1 m UJ

f«

iii-fiJ2 0) E e'l'I

ossisii a¿2 « cu —(C s s- Oï

Síf*fv;liïSiK §i 2

9 B E ^

S«2!_ ~ 4) *-

E •i O) f̂ B

<" O ^-; ®

•éficg 3o 2S «•?W03Ô 5«gu >

i«!IIrli'S 'S c -2

ini0 - C. ~

sl|s.

¡Sii2 ° 21 °~- S? <uta E!D o ||_ f**£ «< xEaa'w<" — CO LU

Ti lg O0 t ° oSSÒS»S o °" - «SM c g o1 °si

lili5 £^E o 5 oS5E£E P ott§5 S= S .cSi C IN

L ^ "^ =^ tf) v O)O, O) X_ O

S £M E£- uj p

^ > Oí*-:ilji1^"=?lï&SCO r; *- •—

sii«?5i-s•;e c o

1|«|O 0) w O"-Sic

~£^1¡talO Îo O)

li >.£ ra¿a e~ tt\ (11

5102

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1981;41:5096-5102. Cancer Res   George E. Milo, Inge Noyes, John Donahoe, et al.   after Exposure to Chemical Carcinogens

in VitroNeoplastic Transformation of Human Epithelial Cells

  Updated version

  http://cancerres.aacrjournals.org/content/41/12_Part_1/5096

Access the most recent version of this article at:

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  [email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications

on April 20, 2017. © 1981 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from