[CANCER RESEARCH 45, 3668-3676, August 1985]

Comparative Properties of Five Human Ovarian Adenocarcinoma Cell Lines1

R. N. Buick,2 R. Rullano, and J. M. Trent3

Ontario Cancer Institute and Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada [R. N. B., R. P.], and Cancer Center, University ofArizona, Tucson, Arizona 85721 [J. M. T.J

ABSTRACT

We describe the derivation of three human ovarian carcinomacell lines and the comparison of their properties with two previously described cell lines of like histology (SKOV-3 and CAOV-3). Two of the new lines (HOC-1 and HOC-7) were derived fromseparate ascites tumors (at 9-month intervals) of a patient withwell-differentiated serous adenocarcinoma of the ovary. The thirdnew line, HEY, was derived from a human ovarian cancer xeno-graft (HX-62) originally grown from a peritoneal deposit of apatient with moderately differentiated papillary cystadenocarci-

noma of the ovary. The cell lines demonstrated differential abilityto grow in semisolid culture and as xenografts in immunologicallydeprived CBA/CJ mice. Dose-response curves were generatedfor clonogenic cell survival of cells exposed to common chemo-

therapeutic agents; one of the lines (HEY) shows a degree ofresistance to the alkylating agent c/s-diamminedichloroplati-num(ll) (c;s-platinum). Common karyological features included

structural abnormalities of chromosomes 3 and 11. Heterogeneity of expression of ovarian tumor-associated antigens was

documented.

INTRODUCTION

A limited number of established cell lines from human ovariancarcinoma cells have been described previously (8, 10, 12, 22,29). To varying degrees, these cell lines have been shown toretain tissue-specific characteristics enabling their use in studiesof tumor-associated antigenicity (18), hormone production anddependency (14), and growth control (28).

A number of recent studies have shown that primary ovariantumor cells are also able to be grown in short-term semisolid

cultures and that assessments can be made of the properties ofclonogenic tumor cells (6, 7, 11). Measurement of fractionalsurvival of such cells to in vitro exposure to chemotherapeuticdrugs has been shown to have a degree of predictive value interms of patient response (1, 19). Such studies suffer in aquantitative sense from the inability to obtain reproducible tumorcell aliquots to allow analysis of dose-response relationships and

the relationship of in vitro clonogenic cell survival to in vivoresponse (20).

We have therefore studied the properties of 5 establishedovarian adenocarcinoma cell lines (2 previously described) toobtain the quantitative information necessary to use these linesas tissue-specific, reproducible sources of cells for analysis of

ovarian tumor biology and of the effects of radiation and chemotherapeutic drugs on human ovarian tumor cells growing inculture and as xenografted tumors in immunologically deprivedmice.

1Supported by grants from the National Cancer Institute of Canada (R. N. B.)

and by Department of Health and Human Services Grants CA 29476 and CA17094from the National Cancer Institute (J. M. T.).

2To whom requests for reprints should be addressed.3Scholar of the Leukemia Society of America.

Received 6/25/84; revised 4/2/85; accepted 4/17/85.

MATERIALS AND METHODS

Patient Material

Case A. Cell lines HOC-1 and HOC-7 were initiated from separate

malignant ascites (April 29, 1980 and December 16, 1980) of a patientwith a serous well-differentiated Stage III adenocarcinoma of the ovary.

Her treatment history was as follows. Diagnosis was at exploratorylaparotomy 2 years prior to first ascites sample. For 7 months prior tofirst paracentesis, she had no treatment; for the 9 months betweeninititiation of HOC-1 and HOC-7, she had paracentesis at approximately

monthly intervals. The characteristics of the cells obtained by paracentesis are described elsewhere (17).

Case B. Cell line HEY was initiated from a disaggregated xenograftedovarian tumor. The derivation of the xenografted tumor (HX-62) (obtained

from Dr. A. Jones, Institute for Cancer Research, Button, England) hasbeen described (21). The initial biopsy was a peritoneal deposit of amoderately differentiated papillary cystadenocarcinoma of the ovary. Thecell line was initiated after 3 passages of the xenograft in immunologicallydeprived mice.

Initiation and Maintenance of Cell Lines HOC-1, HOC-7, and HEY

Case A. HOC-1 and HOC-7 were initiated from separate ascitessamples in identical fashion. Ascites cells were harvested by centrifuga-tion at 600 x g for 10 min and resuspended in «-MEM4medium plus

PCS (10%, v/v). Mononuclear cells were prepared by FicolhHypaque,and the washed cells were frozen in liquid nitrogen in dimethyl sulfoxide(10%, v/v) and PCS (20% v/v). Thawed cells were subsequently seededin MEM plus 10% PCS at a concentration of approximately 106 cells/

flask, and culture was initiated. Passaging by trypsinization (0.025%)has continued for approximately 24 months on a weekly basis (1:10 in«-MEM:10% PCS), and cells at various passages have been maintained

frozen.Case B. An ovarian tumor xenograft HX-62 (21), passage 3, was

grown in the flank of an immunologically deprived CBA/CJ mouse (15).Tumor pieces were aseptically removed and washed in «-MEM:10%

PCS. They were minced with crossed scalpels, forced through a finecopper screen, and passed through needles of decreasing size to 23gauge. The resulting cell suspension was seeded and passaged asdescribed for Case A. The line has been maintained in this manner andby freezing at various passages for 18 months.

Maintenance of Established Cell Lines

Cell lines CAOV-3 and SKOV-3 were obtained from the American

Type Culture Collection, Rockville, MD. No banded karyotypes havebeen published previously on these lines. The cell lines were obtained atpassages 30 and 19, respectively. Both have been maintained by passaging continuously on a weekly basis (SKOV-3,1:8 in «-MEMcontaining15% PCS; CAOV-3,1:4 in «-MEMcontaining 10% PCS) and by cryopres-

ervation.

Measurement of Clonogenicity

Assessments of clonogenic cell frequency in plastic-adherent cultureswere performed by seeding an appropriate cell number in triplicate T-25flasks in «-MEMcontaining 10% PCS, staining with méthylèneblue after

1 to 2 weeks of culture, and counting colony number (>50 cells) under

4The abbreviations used are: MEM, minimal essential medium; FCS, fetal calf

serum; CEA, carcinoembryonic antigen; PBS. phosphate-buffered saline; cis-plati-num, c/s-diamminedichloroplatinum(ll).

CANCER RESEARCH VOL. 45 AUGUST 1985

3668

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HUMAN OVARIAN CANCER CELL LINES

low-power inverted microscopy.Semisolid cultures were initiated in similar fashion in 35-mm tissue

culture dishes in 2-layer systems of agar (0.3%, w/v) on agar (0.5%, v/

v) or methylcellulose (0.8%, v/v) on agar (0.5%, v/v). Colony counts wereperformed at an appropriate time of incubation determined by serialobservation, using a criterion of 3>5Qcells as designation of a clonal unit.

Growth as Xenografts

Cell lines (2 x 106 cells) were injected into the right thigh muscle of

¡mmunologically deprived CBA/CJ mice (15). Following transplantation,all animals were observed weekly until their tumors grew to a meandiameter of 1.0 cm or for at least 4 months in the absence of tumors.Excised tumors were fixed in formalin, and thin paraffin sections werecut and stained with hematoxylin and eosin for histological analysis.

Cytofluorometric Analysis of Antigen Presentation

The antibodies used detected NB/70K (4), OC-125 (3), Ca antigen (2),CEA (4), and Ba-2.5 A rabbit antiserum directed against NB/70K and the

murine monoclonal antibodies detecting OC-125 and Ba-2 were gener

ously provided by Dr. G. I. Urbach, Wellesley Hospital, Toronto, Ontario,Canada; Dr. R. Bast, Sydney Farber Cancer Center, Boston, MA; andDr. T. Lebien, University of Minnesota, Minneapolis, MN, respectively.

Murine monoclonal antibodies directed against the antigens Ca andCEA were purchased from Wellcome Diagnostics, England, and Hybri-tech, La Jolla, CA, respectively. Either fluorescent goat anti-rabbit orgoat anti-mouse IgG (Cappel) was used as second antibody as appro

priate.Quantitation of the levels of ovarian tumor-associated surface antigens

on cells was performed on a Coulter Epics flow cytofluorimeter. Cells(1.5 x 106) suspended in PBS were reacted with the first antibody for30 min at 4°C.The cells were then washed twice and resuspended in

PBS containing an appropriate fluorescein-conjugated second antibody.After 30 min incubation at 4°C,cells were washed 3 times in PBS and

analyzed at 488 nm. Antibodies were always used at saturating concentrations.

Karyotyping

Harvesting, slide preparation, and chromosome banding analysis ofall cell lines were performed as described previously (26).

Electron Microscopy

Cell layers were fixed in situ with 3.5% glutaraldehyde in cacodylatebuffer, postfixed for 1 h in 1% buffered osmium tetroxide, and stainedfor 30 min with 20% aqueous uranyl acetate. Following dehydration inethanol and embedding in Epon, portions of the cell layer were removedfor sectioning using an inverted capsule method. Thin sections were cutparallel to the monolayer and stained with uranyl acetate and lead citratebefore being examined in an Elmiskop I electron microscope.

RESULTS

Morphological Characteristics of Cell Lines. All 5 lines growas monolayer culture on plastic. Morphology is epithelial-like,

with the cells forming a whorling pattern of growth. An exampleof typical morphology of the line HOC-7 is shown in Fig. 1A

Evidence for the epithelial nature of the cells was obtained byelectron microscopy. All cell lines demonstrated tight junctionsand surface microvilli; Fig. 16 shows typical cell junctions fromthe cell line HOC-7.

Cell Growth Characteristics. The doubling times for plastic-

adherent growth of the lines under the growth conditions described in "Materials and Methods" are: HOC-1, 36 h; HOC-7,

16.4 h; HEY, 30 h; SKOV-3, 28.8 h; CAOV-3, 78 h. All cell lines

will form recognizable colonies when grown in low cell densityon plastic substrate (Fig. ~\A).Colony growth efficiency is linearly

related to plated cell number for all lines. The plating efficienciesand gross colony descriptions are given in detail in Table 1A. Inaddition, clonal growth was initiated in culture rendered semisolidby agar or methylcellulose (Table 1B). In all cases of positiveclonogenicity, frequency of colony growth could be linearly related to the number of cells plated. Gross colony morphologywas similar for all positive lines,, tightly packed spherical groupsof cells allowing easy quantitation.

Aliquots (2 x 106) of cells from all lines were also injected into

the flanks of immunologically deprived CBA/CJ mice, and subsequent tumor development was monitored. The HEY andSKOV-3 cell lines produced continuously growing tumors in

100% of mice. Tumors reached 1 cm diameter after approximately 5 weeks. No tumors developed within 3 months afterinjection of HOC-1, HOC-7, or CaOV-3. Resulting tumors fromHEY and SKOV-3 were assessed pathologically and had a

history consistent with that of adenocarcinoma. A representativeexample of a tumor formed from the HEY cell line is shown inhematoxylin-eosin-stained section in Fig. 1C.

Presentation of Tumor-associated Antigens. Expression ofantigens NB/70K, OC-125, Ca-1, CEA, and Ba-2 were deter

mined by flow cytofluorometry on cells harvested by gentletrypsinization and exposed to appropriate antibody combinations. The result of a typical experiment with the expression ofCEA, OC-125, and Ba-2 on cell line HOC-7 is shown in Chart 1.

Expression of antigens was classified as positive (strong toweak) or negative, and the accumulated results are presented inTable 2. Trypsinized cells resuspended in a-MEM:10% FCS and

incubated in roller tubes for 24 h showed the same patterns ofantibody recognition. Parallel assessment of antigen presentation by fluorescence microscopy yielded qualitatively similar data(results not shown).

Karyology. A comparison of numerical and structural chromosome alterations between all 5 cell lines is presented in Table3, with documentation of chromosome alterations presented inFigs. 2 to 5. Results from these cell lines are in good generalagreement with chromosome alterations commonly observed indirect specimens from human ovarian adenocarcinomas (Ref.25; for review, see Ref. 23). Those chromosomes most consistently involved in either structural or numeric alterations werechromosomes 3, 6, and 11. Structural alterations of the shortarm of chromosome 3 involving bands p12-13 were found in 4

of 5 cell lines. One line (CAOV) failed to demonstrate an alterationof 3p but did display a simple deletion of the long arm

Table 1

Assessment of clonogenic growth of ovarian cancer cell lines

A. Growth on plastic substrateB. Growth in semisolid

medium

Plating efficiency(%)Cell

UneHOC-1

HOC-7HEYSKOV-3

CAOV-3Plating

efficiency(%)11.5

±1.0a

20.5 ±3.017.0±3.017.5

±2.06.1 ±1.5Gross

colonymorphologyTightly

packedTightly packedLoose aggrega

tionTightly packedTightly packedAgarNo

growth10.0 ±2.028.5 ±5.015.0±

1.13.0 ±0.1Methyl-

celluloseNo

growth0.5 ±0.1

23.5 ±6.538.0

±9.03.0 ±0.1

51.Royston, personal communication. 8 Mean ±SEof triplicate measurements.

CANCER RESEARCH VOL. 45 AUGUST 1985

3669

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HUMAN OVARIAN CANCER CELL LINES

»

Control

CEAOC-125

Table 2

Antigen presentation by ovarian cancer cell lines

Tumor-associated antigen

Cell line NB/70K OC-125 Ca-1 CEA Ba-2

HOC-1HOC-7HEYSKOV-3CAOV-3+"+ Not tested Not tested Nottested++ — —++—

— —++__ _ __±+ ++ ++ ++

O 20 60 100 140 180 220 258Fluorescencetchonnel number)

Chart 1. Flow cytometric analysis of binding of antibodies to CEA, OC-125, and Ba-2 to the HOC-7 cell line. Control represents HOC-7 cells exposed to nonspecific

mouse IgG.

displayed an apparently bimodal distribution around 49 and 94chromosomes (with a wide variation in chromosome numberobserved between these "modes"). The remaining tumors dis

played a mode near-triploid for SKOV and CAOV and pseudo-diploid for HOC-1 and HOC-7.

Drug Sensitivity. We have assessed the drug sensitivity ofthe lines HOC-7, HEY, SKOV-3, and CAOV-3 to 3 agents commonly applied to the chemotherapy of ovarian cancer, Adriamy-cin, vinblastine, and c/s-platinum. Response of the lines to the

drugs was assessed by measurement of fractional survival ofclonogenic cells (plastic adherent) in response to continuouscontact with the drugs during growth of the cells. An example ofthe dose-response curves obtained is shown in Chart 2 for HEYand SKOV-3 cells exposed to c/s-platinum. The combined data

are summarized in Table 4 as the doses of drug required toreduce survival to 10% of control.

DISCUSSION

We describe the establishment of 3 new ovarian carcinomacell lines. Two of these lines (HOC-1 and HOC-7) were estab

lished from the ascites tumor of an individual patient at differentstages of her disease. This patient has been the subject ofprevious reports concerning clonal evolution and disease progression (17, 26). The third line (HEY) was established from ahuman ovarian tumor which had been passaged previously as axenograft in immunologically deprived mice (21).

All 3 lines display epithelial morphology under light and electronmicroscopy (Fig. 1). Growth on plastic displayed similar largecuboidal cells forming a whoriing cobblestone appearance without obvious fibroblastic contaminants. When plated at low cellnumbers for colony formation, the cell line HEY was noted toform clonal units which were much more loosely packed thanthe other cell lines (Table 1). This would make quantitation ofclonogenicity more difficult at high cell inocula.

The proliferative features of the cell lines are characteristic. Tosummarize the data of Table 1 and xenografting experiments,

* ++, strongly positive; +, positive; ±,weakly positive; -, negative.

[del(3Xp13:)]. Consistent with earlier reports of ovarian cancers(16, 23, 27), alteration of the long arm of chromosome 6 was acommon finding in these cell lines. In all lines except HOC-1 andHOC-7, a simple deletion involving band regions 6q21-23 was

observed to be a clonal feature, present in approximately 100%of all cells examined. Finally, structural alterations of chromosome 11 were found in 4 of 5 tumor lines, with the mostfrequently altered band region being 11q13-21. In 3 cases, a

simple deletion of 11q was observed; while in a single case(HOC-7) (Fig. 5), a translocation involving 11q13 was observed.

In addition to these identifiable chromosome alterations, all tumorlines with the exception of HOC-1 and HOC-7 contained numer

ous unidentified marker chromosomes (umars) (Figs. 2 to 4).Unfortunately, karyotypic information on the primary tumor

specimens from which these lines were derived was availableonly for the HOC-1 and HOC-7 cell lines. In these lines, results

comparing the in vivo karyotype with the karyotype of cellsgrown in vitro in monolayer culture revealed conservation ofmost structural and numeric clonal alterations. Finally, the HOC-7 cell line demonstrated significant karyotypic "evolution" (as

defined as increasing karyotypic progression) based upon resultsof serial harvests over a 1-year period in vitro (26).

The HEY cell line demonstrated a highly aneuploid karyotypewith a high percentage of polyploid cells (>50%) and numerousclonal marker chromosomes. The chromosome number of the75 cells examined varied within a range of 43 to 100 and

CANCER RESEARCH VOL. 45 AUGUST 1985

3670

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HUMAN OVARIAN CANCER CELL LINES

Table 3

Cytogenetic features ol 5 established ovarian carcinoma cell lines

Cell line BandingNo. of cells

counted Range Mode Clonal structural abnormalities

SKOV

CAOV

HEY

HOC-1

HOC-7

G-,Q-

G-,Q-,C-

G-

25 32-88 74

25 43-4N 69

t(1;13X1p22;13q34)del(3Xp13:)del(6Xq23:)del(11Xq21:)del(12Xq22:)±15-19 umars'

del(1Xp22:)del(3Xq21:)del(6Xq23:)t(7;?Xp22;?)

75 45-113 NO

G-,Q-

G-,Q-

100

100

34-4N

42-4N

49

50

del(XXq26:)

del(3Xp12:)t(5:?X5p15;?)del(6Xq21-23:)i(8q)del(11Xq21:)+3-5 umars

inv(3Xp1 3q23)

inv(3Xp1 3q23)t(11;13Xq13;p13)t(12;19Xq24;q13)

8 umars, unidentified marker chromosome; ND, not determined.

IOO

O.OOI OOl O.I I.OCis-platinum( ¿ig/ml)

Chart 2. Dose-response curves for cytotoxicity of c/s-platinum to ovarian cancercell lines; drug was included in the culture medium at the concentrations indicated.O, HEY; •,SKOV-3. Clonogenic cells on plastic were assessed, and fractionalsurvival was calculated. Points, means for triplicate flasks; bars, SE.

Table 4

Drug sensitivity of 4 ovarian cancer cell lines

D,o(ng/ml)

CelllineHOC-7

HEYSKOV-3CAOV-3Adriamycin4

x 10-34 x 10-'8x10-*

2.5 x 10-3Vinblastine2x

1Q-44x 10-56x IO'53x IO"6c/s-Platinum5X10-2

1.07.5 x 1Q-2

<2.5x 10-'

HOC-1 is Clonogenic on plastic but will not grow in semisolidmedium or in immunologically deprived mice. HOC-7, however,

will grow as colonies on plastic and in semisolid medium but also

is not tumorigenic in the xenogeneic host. HEY grows under all3 circumstances. Of the previously established cell lines, SKOV-3 likewise grows under all 3 circumstances, while CAOV-3 fails

to grow in semisolid medium but is tumorigenic in immunologically deprived mice (9). Appropriate selection of cell line cantherefore be made to study therapeutic effects in culture and invivo.

Analysis of the cell surface display of ovarian cancer-associated antigens revealed an individual pattern of expression. HOC-1 and HOC-7 were NB/70K, OC-125, and Ba-2 positive. HEYwas negative for all antigens except Ba-2. Of the 2 establishedlines, SKOV-3 was negative for all markers, and CaoV-3 was

positive for all antigens except CEA. The tumors of origin of thelines HOC-1, HOC-7, and HEY contained some cells positive forall 5 tumor-associated antigens (4).6However, it is also clear thatmany of the original tumor cells were "null" with respect to these

markers. It is possible, therefore, that the lack of expression ofsome of the ovarian tumor-associated antigens in these cell linesreflects the origin of the lines from non-antigen-expressing tumor

cells. Alternatively, the expression of these antigens may berelated to a process of cell differentiation rather than to thetransformed state (4). Nevertheless, the antigen expression pattern clearly is a characteristic of the individual line; as such, thesecells may prove valuable in assessing the biological function ofthese tumor-associated antigens.

Considerable heterogeneity is seen in the ability of the celllines to form colonies in semisolid or viscous cultures. HOC-7

demonstrates a high plating efficiency (10%) in agar but growspoorly in methylcellulose (0.5% plating efficiency). SKOV-3, how

ever, shows higher plating efficiency in methylcellulose. Suchindividual preferences for semisolid support have also been notedamong directly cultured ovarian carcinoma ascites samples (5).There is no concordance between features of tissue culture

•Unpublishedobservations.

CANCER RESEARCH VOL. 45 AUGUST 1985

3671

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HUMAN OVARIAN CANCER CELL LINES

growth and tumorigenicity in immunologically deprived mice.HOC-1 and HOC-7 were derived from biopsies from an individ

ual patient at different stages of disease progression. The onlycharacteristic of the cell lines which point to a clear divergenceof properties is that of growth in semisolid medium. HOC-1 isunable to proliferate in agar or methylcellulose while HOC-7

grows well in agar and to a lesser degree in methylcellulose. Allother measured characteristics are indistinguishable, includingtheir inability to form tumors in immunologically deprived mice.

Banding analysis of all 5 ovarian tumor lines revealed ananeuploid karyotype with numerous numeric and structural abnormalities. Chromosomes 3, 6, and 11 appeared to representthe most frequently altered chromosomes in these lines, a findingconsistent with earlier reports (16, 23, 24, 27). Banding analysisof 3 cell lines (SKOV, CAOV, and HEY) demonstrated a simpledeletion of chromosome 6q21-23;, a region of the long arm of

chromosome 6 which has been suggested by several investigators to represent a nonrandom site of chromosome change inovarian neoplasia (16, 23, 27). It has been reported recently thattranslocations or deletions involving this same region of chromosome 6 have also been shown to occur nonrandomly in humanmalignant melanoma (25). The molecular basis which may explainthe similarity in chromosome change between melanoma andovarian cancer may have been identified recently when Harperef al. (13) mapped the c-myb oncogene to chromsome 6q22-

24. Currently, it is unknown whether the alterations of chromosome 6 in melanoma or ovarian cancer are directly related to thelocation of the c-myb oncogene. Investigations are under way inour laboratories to elucidate whether the c-myb oncogene is

located at or near the breakpoint of 6q demonstrated in theseovarian cell lines (or whether it is translocated elsewhere withinthe genome). However, preliminary experiments indicate that thec-myb gene is not expressed at abnormally high levels in theHEY cell line.6

The range of drug sensitivity seen in 4 lines to Adriamycin,vinblastine, and c/s-platinum is shown in Table 4. In general,

Adriamycin and vinblastine sensitivity falls within a narrow range,but considerably more variability is seen in the sensitivity to c/s-

platinum. Of particular interest is the apparent resistance of theHEY cell line to c/s-platinum. This line may therefore prove to bevaluable in assessing mechanisms of c/s-platinum resistance in

culture and in vivo since it also can be grown as a xenograftedtumor.

In summary, we believe that it will be possible, on the basis ofthese data, to select a model ovarian carcinoma cell line with thecombination of characteristics necessary to answer many questions relating to ovarian tumor biology and therapeutic effects.

ACKNOWLEDGMENTS

We gratefully acknowledge the assistance of Dr. S. Fine in providing the tumormaterial used to derive cell lines HOC-1 and HOC-7 and Dr. P. Selby for initiatingthe xenografted tumor HX-62 (the source of HEY cell line) in our laboratory. Ourthanks also to Dr. I. Tannock for providing the facilities of his laboratory in thexenografting studies, to the late Dr. A. Howatson for the electron microscopicstudies, and to F. Thompson, J. Coxhead, and K. Massey for excellent technicalassistance in the karyotyping studies.

REFERENCES1. Alberts, D. S., Chen, H. S. G., Soehnlen, B., Salmon, S. E., Surwit, E. A.,

Young, L., and Moon, T. E. In vitro clonogenic assay for predicting response

of ovarian cancer to chemotherapy. Lancet, 2. 340-342, 1980.2. Ashall, F., Bramwell, M. E., and Harris, H. A new marker for human cancer

cells. I. The Ca antigen and the Gal antibody. Lancet, 2: 1-6, 1982.3. Bast, R. C., Jr., Feeney, M., Lazarus, H., Nadler, L. M., Colvin, R. B., and

Knapp, R. C. Reactivity of a monoclonal antibody with human ovarian carcinoma. J. Clin. Invest. 68:1331-1337, 1981.

4. Bizzari, J. P., Mackillop, W. J., and Buick, R. N. Cellularspecificityof NB/70K,a putative human ovarian tumor antigen. Cancer Res., 43: 864-867, 1983.

5. Buick, R. N., and Fry, S. E. A comparison of human tumor cell clonogenicityin methylcelluloseand agar culture. Br. J. Cancer42: 933-936,1980.

6. Buick, R. N., Fry, S. E., and Salmon, S. E. Effect of host cell interactions onclonogenic carcinoma cells in human malignant effusions. Br. J. Cancer 41:695-704,1980.

7. Buick, R. N., and Mackillop, W. J. Measurementof self-renewal in culture ofclonogenic cells from human ovarian carcinoma. Br. J. Cancer 44: 349-355,1981.

8. DiSaia,P. J., Morrow, M., Kanagus,J., Piechal,W., and Townsend, D. E. Twonew tissue culture lines from ovarian cancer. Gynecol. Oncol. 3, 215-219,1975.

9. Fogh, J., Fogh, J. M., and Orfeo, T. One hundred and twenty-seven culturedhuman tumor cell lines producing tumors in nude mice. J. Nati. Cancer Inst,59: 221-226,1977.

10. Fogh,J., and Trempe, G. New humantumor cell lines.In: J. Fogh(ed.),HumanTumor Cells in Vitro, pp. 115-159. New York: PlenumPublishingCorp., 1975.

11. Hamburger, A. W., Salmon, S. E., Kim, M. D., Trent, J. M., Soehnlen,B. J.,Alberts, D. S., and Schmidt, H. J. Direct cloning of human ovarian carcinomacells in agar. Cancer Res., 38: 3438-3444,1978.

12. Hamilton,T. C., Young, R. C., McCoy, W. M., Grotzinger, K. R., Green,J. A.,Chu, E. W., Whang-Peng,J., Rogan, A. M., Green, W. R., and Ozols, R. F.Characterizationof a human ovarian carcinoma cell line (NIH: OVCAR-3)withandrogenand estrogen receptors. Cancer Res., 43: 5379-5389,1983.

13. Harper, M. E., Franchira,G., Love, J., Simon, M. I., Gallo, R. C., and Wong-Staal, F. Chromosomalsublocalizationof human c-myb and c-fes cellularonegenes. Nature (Lond.),304: 169-171, 1983.

14. Kanabus, J., Braunstein, G. D., Emry, P. K., DiSaia, P. J., and Wade, M. E.Kinetics of growth and ectopie production of human chorionic gonadotropinby an ovarian cystadenocarcinomacell line maintainedin vitro. Cancer Res.,38:765-770,1978.

15. Kovnat, A., Armitage, M., and Tannock, I. Xenograftsof humanbladdercancerin immune-deprivedmice. Cancer Res., 42: 3696-3703,1982.

16. Kusyk, C. J., Edwards, C. L., Arrighi, F. E., and Romsdahl, M. M. Improvedmethod for cytogenetic analysis of solid tumors. J. Nati. Cancer Inst., 65:1199-1203, 1979.

17. Mackillop, W. J., Trent, J. M., Stewart, S. S., and Buick, R. N. Tumorprogression studied by analysis of cellular features of serial ascitic ovariancarcinoma tumors. Cancer Res., 43: 874-878, 1983.

18. Palillo, R. A., Story, M. T., and Ruckert, C. F. Expression of cell-mediatedimmunity and blocking factor using a new line of ovarian cancer cells in vitro.Cancer Res., 39: 1185-1191, 1979.

19. Salmon, S. E., Hamburger, A. W., Soehnlen, B., Durie, B. G. M., Alberts, D.S., and Moon, T. E. Quantitation of differentialsensitivityof humantumor stemcells to anti-cancerdrugs. N. Engl. J. Med., 298: 1321-1327, 1978.

20. Selby, P., Buick, R. N., and Tannock, I. A critical appraisal of the "humantumor stem-cell assay." N. Engl. J. Med., 308: 129-134,1983.

21. Selby, P. J., Thomas, J. M., Monaghan, P., Sloane, J., and Peckham, M. J.Human tumor xenografts established and serially transplanted in mice immunologically deprived by thymectomy, cytosine arabinoside and whole-bodyirradiation. Br. J. Cancer,41: 52-61, 1980.

22. Sinna, G. A., Beckman, G., Lundgren, E., Nordenson, I., and Roos., G.Characterization of two new human ovarian carcinoma cell lines. Gynecol.Oncol. 7: 267-280, 1979.

23. Trent, J. M. Prevalenceand clinical significanceof cytogenic abnormalities inhuman neoplasms.In: D. S. Alberts and E. A. Surwit (eds.), Ovarian Cancer,pp. 1-22. Boston: Martinus Nijhoff Publishers, 1985.

24. Trent, J. M., Rosenfeld, S. B., and Meyskens, F. L. Chromosome 6q involvement in human malignant melanoma.Cancer Genet. Cytogenet., 9: 177-180,1983.

25. Trent, Ü.M.,andSalmon,S. E. Karyotypicanalysisof humanovariancarcinomacells cloned in short-term agar culture. Cancer Genet. Cytogenet. 3: 279-291,1981.

26. Trent, J. M., Thompson, F. H., and Buick, R. N. Generationof clonal variantsin a human ovarian carcinoma studied by chromosome banding analysis.Cancer Genet. Cytogenet. 14: 153-161, 1984.

27. Wake, N., Hreschyshyn,M., Piver,S., Matsui, S., and Sandberg,A. A. Specificcytogenetic changes in ovarian cancer involving chromosomes 6 and 14.Cancer Res.. 40: 4512-4518, 1980.

28. Welander, C. E., Natale, R. B., and Lewis, J. L., Jr. In vitro stimulation ofhuman ovarian cancer cells by xenogeneic peritoneal macrophages. J. Nati.Cancer Inst. 69: 1039-1047, 1982.

29. Woods, L. K., Morgan, R. T., Quinn, L. A., Moore, G. E., Semple,T. U., andStedman, K. E. Comparison of four new cell lines from patients with adeno-carcinoma of the ovary. Cancer Res., 39: 4449-4459,1979.

CANCER RESEARCH VOL. 45 AUGUST 1985

3672

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

•t ÎB

Fig. 1. ^, representative Wright's-stained colony of HOC-7 cells adherent to the culture flask. x^OO.B, high-power electron micrograph of HOC-7 cell line in monolayer

culture. Thick arrows, macula adherens; thin arrow, puncta (zonula) adherens. x 3200. C, thin section of tumor formed by HEY cell line in immunologically deprived mice.H&E.x 350.

CANCER RESEARCH VOL. 45 AUGUST 1985

3673

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HUMAN OVARIAN CANCER CELL LINES

.

cifi- sic, ,s e«6«*»»•••*

» *»»,9 20

C S««t »litie te

21 22 X

*ce*.

fa*»-

•1 2 3

444 iVi,4 5

I iti! ti Si II I8s«8S 6 7 8 9 10 li \ \ V2

Oli 55 ila a13 14 15 16 17 18

21 22 4 Xti SUBII

19 20

H BUU 56 )! IIS OD •a» * •

Fig.2. Q-banded karyotype of SKOV cell line. Identifiableclonal structural markers (arrows)are described in more detail in Table 3. Deletion 10p chromosome seenin this karyotype was not a clonal change.

Fig.3. Q-banded metaphase karyotype of the CAOV cell line.Arrows, identifiableclonal structural markers. A detailed description of these markers is given in Table3.

CANCER RESEARCH VOL. 45 AUGUST 1985

3674

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HUMAN OVARIAN CANCER CELL LINES

)() JiffIHIt. fum

13

m3

<ÌHlK<ìl)«f7 8 " 99 in 10

Mil Mil14 15 16

11 12

»I i*17 18

IK fill*19 20

MM21

II22

umar c mg

X/Y

X

Fig. 4. G-banded karyotype of the HEY cell line. Arrow, del(6Xq21-23:) chromosome. Other clonal structural markers observed in this karyotype include: m, =t(5;?X5p15;?); m2 = i(8q), m3 = del(3Xp12:);m4 = a putative del(9Xp13:)chromosome; and a number of unidentified marker chromosomes (umar). Inset, additionalexamples of clonal markers from another cell, including a del(11Xp21:)chromosome which was a clonal marker that was not present in the karyotype. Clonal markersmsand m„could not be positively identified.

CANCER RESEARCH VOL. 45 AUGUST 1985

3675

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HUMAN OVARIAN CANCER CELL LINES

B •

\* •

t (II 131

/

del|l2q|

t!l2;l9)

Fig. 5. A, G-bandedpartial metaphasespread from the HOC-1 cell line showing the only clonal structural abnormalitydetected, a pericentric inversionof chromosome3. A G-banded partial metaphase from a late passage of the HOC-7 cell line in B displays the same inv(3)marker in addition to other clonal structural rearrangementswhich includes a t(11;13) chromosome. A nucleolar organizing region (NOR)-bandedpartial spread of late-passage HOC-7 in C demonstrates that the t(11;13) markerhas retained the "stalk" region of chromosome 13, thereby serving to confirm the proposed breakpoints at 11q13 and 13p13. Open arrows in C, the NOR bands onanother D-group chromosome and 2 G-group chromosomes which are in satellite association. Further description of clonal markers is given in Table 3.

CANCER RESEARCH VOL. 45 AUGUST 1985

3676

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1985;45:3668-3676. Cancer Res   R. N. Buick, R. Pullano and J. M. Trent  Adenocarcinoma Cell LinesComparative Properties of Five Human Ovarian

  Updated version

  http://cancerres.aacrjournals.org/content/45/8/3668

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

  .pubs@aacr.orgDepartment at

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

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/45/8/3668To request permission to re-use all or part of this article, use this link

on June 26, 2020. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from