Tumor and Lymphoid Cell Lines from a Patient with Carcinoma of … · 201, COLO 205, and COLO 206 were derived from an ascites fluid specimen. COLO 197 had a normal pattern of estab

[CANCER RESEARCH 38, 1345-1355, May 1978]

Tumor and Lymphoid Cell Lines from a Patient with Carcinoma ofthe Colon for a Cytotoxicity Model1

Toni U. Semple, Laurie A. Quinn, Linda K. Woods, and George E. Moore

Department of Surgery, Denver General Hospital, Denver, Colorado 80204

ABSTRACT

Three tumor cell lines (COLO 201, COLO 205, and COLO206) have been established from ascites fluid obtainedfrom a male patient with adenocarcinoma of the colon. Inaddition to the tumor lines, two lymphoid lines (COLO 197and COLO 200) have been established from the same patient, with one line from the original biopsy and one fromperipheral blood.

Characterization of the tumor cell lines revealed fourcell types that differ from most colon cell lines reported byothers. Chromosome markers were identical in COLO 201and COLO 205. A long-arm isochromosome 5 observed inCOLO 201 and COLO 205 was absent in COLO 206.Statistical analysis of autosomal polysomy revealed thatthese cell lines were stable and indicated that there maybe a cytogenetic basis for the three predominant types ofcell morphology.

The lymphoid cell line derived from the peripheral bloodhad a normal male karyotype. The lymphoid cell linederived from a biopsy specimen had a mode of 46 and adeleted chromosome 7 marker. Both lymphoid cell lineshad B-cell characteristics.

These autochthonous cell lines have been used forimmunological studies in cytotoxicity assays and immu-

noglobulin characterization.

INTRODUCTION

An in vitro autochthonous human model for studies ofinteractions between tumor cells and immune cells is avaluable asset in studies of cancers. We have recentlyreported such models derived from several types of cancer(16, 18). Here, we describe 3 colon carcinoma cell lines(COLO2 201, COLO 205, and COLO 206) and 2 B-lymphoid

cell lines (COLO 197 and COLO 200) from the same patient.Colon cancer is second only to skin cancer in prevalence

in this country and as such has been the subject of muchresearch. The recognition that colon cancer cells elaborateCEA has provided a basic means for identification of thiscell type. In addition to CEA production, studies of cellmorphology, cytogenetics, and surface antigens have contributed to the general knowledge of colon carcinoma.

' Supported by Grant CA15018, awarded by the National Cancer Insti

tute, Department of Health, Education and Welfare, and by the Mary B. andL. H. Marshall Fund.

2 The abbreviations used are: COLO, prefix for cell lines established by

our laboratory in Colorado; CEA, carcinoembryonic antigen; RPMI, series ofmedia and cell lines developed at Roswell Park Memorial Institute; GEM,series of media and cell lines developed at Denver General Hospital; PCS,fetal calf serum; CTC, cultured tumor cell; CL, cultured lymphocyte; SCL,sensitized cultured lymphocyte; PPLO, pleuropneumonia-like organisms.

Received October 25, 1977; accepted February 10, 1978.

With the availability of autochthonous B-lymphoid celllines, the interactions between tumor cell targets andB-lymphoid effectors can be studied for a basic understanding of tumor immunology. The intent of this report isto provide a description of the cell lines and preliminarycytotoxicity data on the interactions of autochthonous lymphocytes and colon carcinoma cells.

CASE HISTORY

A 70-year-old white male presented on June 6, 1975, withcarcinoma of the colon; a peripheral blood sample wastaken for cell culture. On June 11, 1975, a right colectomywas performed, and tissue specimens from the right colonwere submitted to pathological examination and tissueculture. The pathology report indicated that the patient hadadenocarcinoma of the colon. The anaplastic carcinomacells invaded through the muscular wall and formed glands,clumps, and sheets of cells. MÃ©tastaseswere present in theliver.

Postoperatively, the patient was started on 5-fluorouraciltherapy and was discharged on June 19, 1975.

On July 23,1975, the patient was readmitted with discomfort from ascitic fluid. Five liters of fluid were taken on July25, 1975, and were submitted for tissue culture. A secondtap of 2.5 liters of fluid was made on July 31, 1975, and thefluid was submitted for tissue culture.

The patient's status deteriorated rapidly, and he died on

August 1, 1975.

MATERIALS AND METHODS

Methods for establishing tumor and lymphoid cell lineshave been previously described by Morgan ef a/. (16).Briefly, the tumor tissue was minced and cultured in RPMI1640 and GEM 1717 (15) supplemented with 10 or 20% heat-inactivated PCS (Reheis Chemical Co., Kankakee, III.), penicillin (100 ID/ml), and streptomycin (50 Â¿Â¿g/ml).The tissuesuspensions were pipeted into loosely capped 4-oz flintglass bottles and incubated at 37Â°in humidified air with

10% COÂ».Each week 1 to 5 ml of fresh medium were addedto the cultures until growth was stabilized. Tumor cellcultures were considered to be established when there wasa single cell type present in the culture, when there wassemiconfluent growth, and when there were several successful subcultures.

The lymphoid cultures were established according tomethods described earlier (14). In short, the buffy coatobtained from 30 ml of anticoagulated blood that had beenallowed to settle was placed in 4-oz glass bottles, mediumwas added, and the cultures were incubated. Lymphoidlines were considered established when clumps of cells

MAY 1978 1345

Association for Cancer Research. by guest on September 5, 2020. Copyright 1978 Americanhttps://bloodcancerdiscov.aacrjournals.orgDownloaded from

https://bloodcancerdiscov.aacrjournals.org

7. U. Semple et al.

began to form in the medium and the culture was successfully subcultured. The individually established cultures werethen assigned a cell line number.

Heterologous transplantation of the cultured tumor cellswas performed in 5 female C3H/HeJ mice irradiated with600 R. Each mouse was given a s.c. injection of 106 viable

tumor cells in 0.1 ml of 0.09% NaCI solution. When palpablenodules were observed, they were surgically removed andprepared for histological studies.

Coverslip preparations of the tumor cell lines were usedfor phase microscopy morphology studies. Giemsa-stained

smears of the lymphoid cell lines were observed with lightmicroscopy. Mucicarmine stains were done according tothe method of Mallory (10). Electron microscopy methodswere used as previously described (13).

Chromosome preparations were from 2-day-old subcultures of the cultured tumor cells incubated for 2 hr at 37Â°

with 0.1 /LilColcemid (Ciba Pharmaceuticals, Summit, N. J.)per 10 ml culture medium. Cell suspensions were centri-

fuged, resuspended in 0.075 M KCI for 12 min, centrifuged,resuspended in fixative (methanohacetic acid, 3:1) for 10min, centrifuged, and then resuspended in fresh fixative foran additional 10 min. Slides were made according to thehot-plate method of Lubs ef al. (9). Giemsa-banded chro

mosomes were obtained by dipping slides in a trypsin(1:250; Difco Laboratories, Detroit, Mich.)-buffered 0.9%

NaCI solution (pH 6.8) for 30 to 90 sec and by then stainingin a buffered (pH 6.8) 2% Giemsa solution (Harleco, Philadelphia, Pa.) for 3 min. Constitutive heterochromatin banding was obtained by a modification of the method ofMcKenzie and Lubs (12). Slides that were used previouslyfor Giemsa-banded preparations were incubated at 60Â°for

5 to 20 min and were stained in a buffered (pH 6.8) 5%Giemsa solution.

The degree of polysomy for each autosome of the tumorcell lines was studied by totaling the number of copies ofeach autosome in 6 to 9 metaphases/cell line. The totalswere ranked for each cell line and were analyzed by subtraction of the mean rank from the rank total and divisionby the S.D. Values greater than 2.08 were significant at the5% level.

CEA analyses were performed by Associated Laboratories, Inc., Wichita, Kans., on spent medium from culturesthat had not been fed for 10 days.

Surface membrane immunoglobulin studies of the celllines were performed as follows: (a) 1 x 106 cells wereharvested and washed once with phosphate-buffered saline

[NaCI (8.0 g/liter):KCI (0.2 g/liter):NaHPO, (1.15 g/liter):KH2PO4 (0.2 g/liter)]; (b) the resultant pellet wasstained with 100 /J of diluted fluorescein isothiocyanate-conjugated goat anti-human antibody (Hyland Laboratories,

Los Angeles, Calif.) for 30 min at room temperature; (c) thestained cells were washed; (d) the cell pellet was mixed in100 /Â¿Iof glycerin:phosphate-buffered saline (3:1); and (e) 1drop of this suspension was placed on alcohol-cleanedmicroscope slides and covered with a 22-sq mm No. 1

coverslip for observation with a Zeiss Photomicroscope IIIequipped with Ã©pi-illumination, a No. 510 exciter filter, and

a No. 530 barrier filter.T- and B-cell rosettes were studied by the E-rosette

methods of Jondal ef al. (7) and Pellegrino ef a/. (17).

Cytotoxicity assays were performed by seeding 2.5 x 10*

CTC suspended in 250 /^l of GEM 1717 medium supplemented with 10% PCS and 1.5x glucose into the wells ofthe TEST II plate (Falcon Plastics, Los Angeles, Calif.). Theplates were incubated at 37Â°with 10% CO; in humidified air

for 8 to 24 hr to allow the CTC to attach to the bottoms ofthe wells. The medium was then aspirated from the wellswith a beveled capillary pipet, and 5 x 105 CL or SCL (20:1)

suspended in 250 /Â¿\of the same medium were added toappropriate target-containing wells. At intervals, succeed

ing rows of CTC and CL together, plus rows of control CTC,were time stopped by aspirating the medium from the wellsand by fixing the adherent cells with absolute methanol for3 min. Subsequently, the cells were stained with 10%Giemsa stain for 10 min, washed in cold tap water, and airdried. The wells were evaluated by 2 or more observers.Values of 0, 2, 4, 6, 8, or 10 were assigned to each wellrepresenting 0, 20, 40, 60, 80, and 100% kill of the CTC byCL or the SCL. SCL were generated by cocultivation of theCL with the CTC for a period of 72 hr. The SCL were thenharvested, counted, suspended in half-fresh, half-sensitized

medium, and added to new target cells for measurement oftarget destruction.

The cell lines were tested for the presence of PPLOaccording to the bisbenzamide fluorescent staining methodof Chen (1).

RESULTS

The history of the cell lines is outlined in Table 1. COLO201, COLO 205, and COLO 206 were derived from an ascitesfluid specimen. COLO 197 had a normal pattern of establishment for lymphoid cells derived from peripheral bloodspecimens (14). COLO 200, a lymphoid line, was established in a primary culture from a tumor biopsy specimen.Lymphoid cells began to divide rapidly during the secondmonth of culture. Neither tumor cells nor fibroblasts wereobserved by the end of the second month. All of these lineshave been maintained in GEM 1717 and RPMI 1640 mediasupplemented with 10 or 20% FCS.

Photographs of the original tumor specimen were obtained for comparison with heterologous transplant tumors(Fig. 1). COLO 201 produced tumor masses in 6 of 7 mice in10 days. The seventh mouse succumbed to infection on Day8. The data were unobtainable for that animal. Evidence ofdifferentiated tumor masses was observed in paraffin-

embedded sections obtained from the other 6 mice (Fig. 2).The 3 mice given injections of COLO 205 cells had solidtumor masses at 10 days. Histological examination confirmed evidence of tumor cell growth. Palpable noduleswere also observed in 4 of 5 mice given injections of COLO206 cells; histological study revealed semidifferentiatedepithelial tumors.

Phase-contrast micrographs revealed 4 cell types in the

colon tumor cell lines: small round cells, bipolar cells,epithelial-like polygonal cells, and signet ring-like cells. The

cell types were evident in different proportions in the 3tumor cell lines. Bipolar cells predominated in COLO 201(Fig. 3), although scattered islands of epithelial colonieswith some signet ring cells and small round cells were alsoobserved.

1346 CANCER RESEARCH VOL. 38



Autochthonous Colon Carcinoma Cell Lines

Table 1History of cell lines established from Patient 54-62-84

CelllineCOLO

197COLO

200COLO

201COLO

205COLO

206Tissue

oforiginPeripheral

bloodPrimary

tumortissueAscites

fluidAscites

fluidAscites

fluidDate

initiated6/6/756/11/757/31/757/31/757/31/75Date

established8/22/758/29/759/15/759/25/759/25/75CelltypeLymphoid

suspensionLymphoid

suspensionTumorTumorTumorTherapy

prior tocultureNoneNone5-FUra5-FUra5-FUra

Epithelial-like colonies composed of polygonal cells andsignet ring-like cells predominated in COLO 205 (Fig. 4).

COLO 206 (Fig. 5) was largely comprised of small roundcells with scattered bipolar cells and occasional epithelialcolonies. The epithelial-like colonies and bipolar cell types

were attached to the glass more firmly than were the smallround cells. In all of the lines, some cells proliferated asfree-floating cells. All 3 lines could be subcultured bypouring 10 ml of suspended cells into a new flask. Muci-

carmine stains indicated that all of the cell lines had somemucin-producing cells.

Electron micrographs of COLO 201 revealed polygonalcells with numerous microvilli, which resembled the brushborder of intestinal epithelial cells. The microvilli of adjacent cells interdigitated (Fig. 6). Desmosomes were observed at some cell junctions. A glycocalyx (fuzzy coat) waspresent on the surface of some microvilli. The nuclei hadmore than 1 nucleolus, and often 1 nucleolus was associated with the nuclear membrane as reported by Drewinkoef al. (3). The cytoplasm contained rough endoplasmicreticulum, abundant ribosomes, numerous mitochondria,bundles of parallel fibrils, vacuoles, and microvesicles inthe cytoplasm and near the membrane, and electron-dense

granules. No virus particles were seen.The cell population in COLO 197 was comprised of

immature lymphocytes of various sizes. Electron micrographs of cells from COLO 200 were typical of B-lymphoidcell lines (Fig. 7) (14). Phase-contrast microscopy andGiemsa-stained smears revealed some small cells with largenuclei and hairy, "rotor-like" projections from the surface

membranes.Cytogenetic analysis of COLO 201 was made on the basis

of 71 metaphases, of which 12 were photographed and 9were karyotyped. COLO 201 was bimodal with modes of 72and 75 (range, 72 to 77). COLO 205 analysis was made on46 metaphases, of which 8 were photographed and 7 werekaryotyped. The mode of COLO 205 was 75 (range, 71 to77). Analysis of COLO 206 was made on 56 metaphases, ofwhich 16 were photographed and 6 were karyotyped. Themode of COLO 206 was 75 (range, 71 to 78).

The marker chromosomes observed in COLO 201 andCOLO 205 were the same and are depicted in Fig. 9a. Themarker chromosomes observed were as follows: (a) M,:t(1?;?;3) (1q2?-Â»1q1?::?-Â»cen-Â»?::3q11-Â»3qter). The first

portion of this marker has a heterochromatic band ofunknown origin (Fig. 9Â£>).Since the heterochromatic bandstains darkly with both G- and C-banding, it is likely to be

chromosome 1, 16, or Y in origin. The length of thissegment makes chromosome 1 the most likely derivativechromosome. It is difficult to describe with accuracy thechromosome segment before and after the centromere. The3q origin of the terminal segment of this chromosomemarker was clear, (b) M,: t(?;7:?) (?::7pter-Â»7q1 ?::?). Both

unidentified terminal portions of this marker were difficultto resolve. Possibly, segments of 13q and/or 2q are involved, (c) M:1: i(5) (qter-*cen-Â»qter). (d) M,, unidentified

metacentric marker; M., is similar in size and bandingpattern to M4 of COLO 206. It may also be a 14p+. (e) M.,,del(10) (qter-Â»pl?:). (/) M,;, t(19;?) (19::?). (g) M-, unidentified E-sized marker.

The marker chromosomes observed in COLO 206 (Fig.10) were as follows: (a) M,, same as M, of COLO 201 andCOLO 205; (b)M,, del(3) (qter-Â»p14:); (c) M:l, same as M, ofCOLO 201 and COLO 205; (d) M4, t(8;?) (8qter-*8p2?::?);

(e) M5, same as M5 of COLO 201 and COLO 205; (f) M,,, sameas M,. of COLO 201 and COLO 205; (g) M7, same as M7 ofCOLO 201 and COLO 205; (h) MK (marker observed only inthis particular metaphase), Â¡(9)(qter-Â»cen-Â»qter); (/') M9

(marker observed only in this particular metaphase), unidentified E-sized chromosome.

Data for statistical analysis of the frequency of eachnormal autosome in COLO 201, COLO 205, and COLO 206are presented in Table 2. COLO 201 and COLO 206 had asignificant polysemy of chromosome 20. COLO 205 had asignificant polysomy of chromosome 9.

Cytogenetic analysis of COLO 197 (Fig. 11a) was made on51 metaphases, of which 16 were photographed and 9 werekaryotyped. The mode of COLO 197 was 46, and thekaryotype was consistent with that of a normal male.

The cytogenetic analysis of COLO 200 (Fig. 110) wasmade on 54 metaphases, of which 13 were photographedand 10 were karyotyped. The mode of COLO 200 was 46 buta 7p- marker chromosome was present.

CEA was not detected in spent medium from the lymphoidcell lines COLO 197 and COLO 200. However, spent mediataken from the tumor cell cultures, fed 10 days previously,revealed that COLO 201 produced 23.1 ng CEA per 106cells, COLO 205 produced 105.7 ng CEA per 106 cells, andCOLO 206 produced 11.8 ng CEA per 106 cells.

Both lymphoid cell lines produced the same classes ofÂ¡mmunoglobulin. However, compared to COLO 197, amarkedly greater number of COLO 200 cells producedimmunoglobulin. Results of the immunofluorescent studiesare presented in Table 3.

MAY 1978 1347



T. U. Semple et al.

Table 2Number of normal chromosomes per metaphase

Normalchromo

some12345678910111213141C1916171819202122Total234412334131225322422no.ofeach

normal chromosomeinCOLO

201(9metaphases)33114232513114322533333332232422224â€¢3233522322232314231333223621333332334131243222623251311224132223233732322032323131213312411323331325332235333311332222324232141224622a

metaphaseCOLO

205(6metaphases)322221324241322012622332322324232243323334332222326223323321533332323222222242222532331332316213224222123332322236123142232524223322422222132121513COLO

206(7metaphases)23234242222334431234233333221323232.331242233234242222222354242333234331432233223154233335242522232321253333335242422313.1322723COLO

201RT"2526172120162619371426131727OQdo29211727451819RT

MR

S.D.0.170.29-0.80-0.31-0.43-0.920.29-0.551.62-1.160.29-1.28-0.800.411

QÂ£1.OD0.65-0.31-0.800.412.59''0.68-0.55COLO

205RT1817111613121612281114141318ooÂ¿Â¿16121212211518RT

MR

S.D.0.600.36-1.070.12-0.60-0.830.12-0.832.98"-1.07-0.36-0.36-0.600.60ICC.OO0.12-0.83-0.83-0.831.31-0.120.60COLO

206RT1920182127152512221515161619OAÂ¿D18191412331818RT

MR

S.D.0.000.20-0.200.391.56-0.781.17-1.370.59-0.78-0.78-0.59-0.590.001

QTI.J/-0.200.00-0.98-1.372.73''-0.200.20

" RT, rank total; MR, mean rank.b Values greater than 2.08 are significant at the 5% level.

Table 3Results of immunofluorescence studies

% of fluorescingcellsImmunoglobulin

classigA

igGigMIgDigEC'3

C'4COLO

197338

00

2851COLO

2002588

00

7525

0

Neither COLO 197 nor COLO 200 resetted spontaneouslywith sheep RBC. However, 95% of the cells from COLO 197and 97% of the cells from COLO 200 rosetted spontaneouslywith RBC from Macaca artoides, thus confirming their B-cell origin.

Cytotoxicity assays that compare the tumor cell destruction capabilities of COLO 197 and COLO 200 showed thatCOLO 197 destroyed 100% of the targets from COLO 201 in140 hr of cocultivation, 100% of the targets from COLO 205in 119 hr of cocultivation, and 100% of the targets fromCOLO 206 in 140 hr of cocultivation. In contrast, the sameratio of COLO 200 cells destroyed 100% of the tumor targetsfrom COLO 201 in 70 hr, 100% of the tumor targets fromCOLO 205 in 74 hr, and 100% of the tumor targets fromCOLO 206 in 72 hr of cocultivation. Electron microscopyshowed numerous sites of attachments between the lymph-oid cell membrane and the tumor cell microvilli of COLO201 cocultivated with COLO 200 (Fig. 8).

All 3 tumor cell lines and the lymphoid cell line COLO 197had no evidence of PPLO contamination. PPLO were observed in COLO 200.

DISCUSSION

This report concerns 3 human colon carcinoma cell lines(COLO 201, COLO 205, and COLO 206) and 2 lymphoid celllines (COLO 197 and COLO 200) derived from the samepatient. Cell lines derived from the same biopsy specimenwere given separate identification numbers because theywere initiated separately, were maintained under differentconditions, and had varying characteristics. Heterologoustransplants, cell morphology, cell products, and cytoge-netic evidence revealed that COLO 201, COLO 205, andCOLO 206 were malignant cell lines. Morphological, immu-nological, and cytogenetic studies revealed that COLO 197and COLO 200 were B-lymphoid cell lines.

The colon carcinoma cell lines did not exhibit the slowgrowth and the formation of tight epithelial colonies thatwere characteristic of most previously described coloncarcinoma cell lines (8, 20, 21). Our tumor lines mostclosely resembled the neoplastic colon cell lines SW 480and SW 620 described by Leibovitz ef al. (8). SW 480 was acell line derived from a primary tumor specimen, whereasSW 620 was derived from a metastatic lesion from the samepatient. Our cell lines were initiated from an ascites fluidspecimen. Leibovitz ef al. (8) described SW 480 and SW 620as adenocarcinoma cell lines that grew as a mixture of smallislands of epithelial cells and individual bipolar cells. COLO205 was most similar to the description of SW 480; COLO206 was somewhat similar to SW 620, except that fewerbipolar cells were evident than were depicted in SW 620.COLO 201 had many more bipolar cells than were evident inSW 620 and also had some loosely attached epithelialcolonies like those of COLO 205 and SW 480.

The cells of COLO 201 contained distinct intestinal epithelial phenotypic traits; microvesicles and vacuoles, often





associated with CEA-producing colon carcinoma cells,were observed in the cytoplasm and near the microvilli (5).It will be of interest to study the â€¢¿�iltramorphologyof the

other 2 cell lines (COLO 205 and COLO 206) and to ascertainwhether the morphological differences observed with phasemicroscopy are also evident with electron microscopicdata. The different cell types in these cell lines probablyrepresent different levels of differentiation.

Cytogenetic analysis of our colon cell lines revealed thatthe modes ranged from 71 to 78. Most colon lines reportedby others have modes near diploid (3, 8, 20, 21), althoughchromosome modes from 50 to 78 have also been reportedfor a few colon carcinoma cell lines (8) and hundreds tothousands of chromosomes have been observed in hugepolyploid cells from fresh colon carcinoma tissue preparations (19).

The 3 tumor cell lines reported here did not have identicalkaryotypes. COLO 206 differed from COLO 205 by theabsence of a statistically significant number of No. 9 chromosomes, and COLO 206 differed from COLO 201 andCOLO 205 by the absence of an isochromosome 5 long-armmarker. COLO 201, unlike COLO 205, had a statisticallysignificant number of No. 20 chromosomes and lacked asignificant number of No. 9 chromosomes. Although minorkaryotypic differences may be related to morphologicaland/or biochemical differences in these cell lines, theabsence of wide variability in the copies of autosomes andmarkers indicated that these cell lines had stable cellpopulations.

The tumor cell line with the slowest growth rate, COLO205, produced the largest quantity of CEA. COLO 205 alsohad the most epithelium-like morphology of the 3 lines.COLO 201 produced markedly less CEA than COLO 205; itsgrowth rate was more rapid and the cells were less differentiated. COLO 206 grew more rapidly than did either COLO201 or COLO 205 and consisted almost exclusively of smallround cells. COLO 206 produced less CEA than did eitherCOLO 201 or COLO 205. Further study would be necessaryto determine whether the levels of cell differentiation,growth rate and/or cytogenetic variability of the cell lines,and CEA production by these cell lines are stable characteristics.

Whether the minor differences among these cell linesarose in vivo or in vitro is not easy to determine. Heterogeneity of tumor cell populations in vivo has been described(4), as have cytogenetic, morphological, and biochemicalinstability in vitro (2, 6, 11). As mentioned previously, wemaintained the tumor cell lines derived from 1 specimen asseparate entities. Cell cultures initiated from the samebiopsy in different types of media or handled in slightlydifferent manners may have different predominant cellpopulations and very different characteristics. The tumorcell lines in this report exhibit differences in morphology,cytogenetics, and CEA production.

Giemsa-stained smears and phase microscopy of COLO197, derived from a peripheral blood specimen, consistedof cells with lymphoblastoid morphology and B-cell characteristics, as is true for all known lymphoid cell linesderived from normal cells. Cytogenetic analysis confirmedthat the origin of COLO 197 was a normal cell population.The majority of COLO 197 cells formed hairy, rotor-like

projections on the surface membrane after exposure to thetumor cells in vitro. In contrast, COLO 200 cells had numerous hairy, rotor-like projections on the surface membranein static culture. Possibly, COLO 200 was sensitized totumor cells in vivo. Cytogenetic analysis revealed thatCOLO 200 has a mode of 46 chromosomes and markerchromosome with a consistent 7p- deletion. This markerwas not observed in the tumor cell lines or in COLO 197.

The failure of COLO 197 and COLO 200 to rosette spontaneously with sheep RBC and their proclivity for rosettingwith monkey RBC, in addition to the presence of surfacemembrane Â¡mmunoglobulin, were evidence of their B-cellnature.

The autochthonous system described herein allowed repetitive cytotoxicity studies of the immune reaction betweenB-lymphoid and tumor cells. Total tumor cell destructionwas obtained at low (20:1) lymphocyte:tumor cell ratioswith both COLO 197 and COLO 200. COLO 200, derivedfrom the original biopsy, killed 100% of the tumor celltargets in one-half of the time it took COLO 197, derivedfrom the peripheral blood, to kill 100%. even though growthkinetics of the 2 lines were similar. One hypothesis thatmight explain different killing rates of the 2 lines is thatCOLO 200 consists of a population or populations of cellsnow propagating in vitro that were "presensitized" or acti

vated in vivo against the tumor cells. They may haveretained a "memory" for tumor cell antigenicity for the 2

years for which they have been in culture, thereby explaining their enhanced tumor cell-destructive capabilities. Additionally, COLO 200 had a far greater percentage of cellsproducing Â¡mmunoglobulin than did COLO 197. Whetherthese 2 findings are directly related remains to be proven.An additional consideration must be given to the positivePPLO test on COLO 200 cells. Presence of such a microorganism may increase the immunological reactivity of thelymphoid cells.

These autochthonous cell lines provide the investigatorwith a system for careful analysis of the immunobiology andantigenicity of human adenocarcinoma of the colon withoutthe complications of histocompatibility differences. Thecytotoxic ability of the B-cell may be studied in vitro, andcellular products such as CEA and/or lymphokines may bemonitored for long periods of time with varying cultureconditions. It would be nice to have parallel T-lymphoid celllines for use in comparative assays of T- and B-cells,separately and mixed together, but we still cannot maintainnormal T-cells in vitro for more than a few weeks.

ACKNOWLEDGMENTS

We gratefully acknowledge V. Buric for electron microscopy. G Knappfor the statistical analysis, and W N. Palmer for heterologous transplantationof the cultured cells.

REFERENCES

1 Chen. T. R. Microscopic Demonstration of Mycoplasma Contaminationin Cell Cultures and Cell Culture Media. Tissue Culture AssociationManual 1: 229-232. 1975.

2. Chen, T. R., and Shaw. M. W. Studies of a Cell Line Derived from aHuman Malignant Melanoma. In Vitro, W: 216-223. 1974.

3. Drewinko. B.. Romsdahl. M. M., Yang, L. Y., Ahearn, M. J.. and Trujillo.J. M. Establishment of a Human Carcinoembryonic Antigen-ProducingColon Adenocarcinoma Cell Line. Cancer Res.. 36: 467-475. 1976

4. Fidler. I J., and Kripke. M. L Metastasis Results from Preexisting

MAY 1978 1349



T. U. Semple et al.

Variant Cells within a Malignant Tumor. Science, 197: 893-895, 1977. 14.5. Gold, P., Krupey, J., and Ansari. H. Position of the Carcinoembryonic

Antigen of the Human Digestive System in Ultrastructure of Tumor Cell 15.Surface. J. Nati. Cancer Inst., 45: 219-222. 1970.

6 Jacobs, B . and Uphoff, D. Immunologie Modification: A Basic SurvivalMechanism. Science, 785. 582-586, 1974. 16.

7. Jondal. M., Holm, G.. and Wigzell. H Surface Markers on Human T andB Lymphocytes. J. Exptl. Med., 136: 207-215. 1972.

8. Leibovitz, A., Stinson, J., McCombs. W., McCoy. C., Mazur, K., and 17.Mabry, N. Classification of Human Colorectal Adenocarcinoma CellLines. Cancer Res.. 36. 4562-4569. 1976

9. Lubs, H. A., McKenzie, W. H., Patii, S. R., and Merrick, S New Staining 18.Methods for Chromosomes. Methods Cell Biol. 6: 346-380. 1973.

10. Mallory. F. B. Pathological Technique, p. 130. New York: Hafner Publishing Co.. 1961. 19.

11. McCombs. W. B.. Leibovitz. A.. McCoy. C. E.. Stinson, J. C.. and Berlin,J. D. Morphologic and Immunologie Studies of a Human Colon Tumor 20.Cell Line (SW-48). Cancer, 38. 2316-2327, 1976.

12. McKenzie, W. H., and Lubs. H. A. An Analysis of Technical Variables inthe Production of C Bands. Chromosoma. 41: 175-182, 1973. 21.

13. Mitchen, J. R., Moore. G. E.. Gerner, R. E., and Woods. L. K. Interactionof Human Melanoma Cell Lines with Autochthonous Lymphoid Cells.Yale J. Biol. Med., 46: 669-680. 1973.

Moore. G E.. Gerner, R. E., and Franklin, H. A. Culture of NormalHuman Leukocytes. J. Am. Med. Assoc.. )99: 519-524, 1967.Moore. G E., and Woods, L. K. Culture Media for Human Cells â€”¿�RPMI1603, RPMI 1634. RPMI 1640 and GEM 1717. Tissue Culture AssociationManual, 3: 503-509, 1977.Morgan, R. T., Quinn. L. A.. Woods. L. K., and Moore. G. E MarkerProperties of Tumor and Lymphoid Cell Lines Derived from a Patientwith Squamous Cell Carcinoma. Cancer Res.. 37 2030-2035. 1977.Pellegrino. M. A.. Ferrone, S.. and Theofilopoulos. A. N. RosetteFormation of Human Lymphoid Cells with Monkey Red Blood Cells. J.Immunol.. 115: 1065-1071, 1975.Quinn, L. A., Woods, L. K., Merrick. S. B.. Arabasz. N. M., and Moore.G. E. Cytogenetic Analysis of Twelve Human Malignant Melanoma CellLines. J. Nati. Cancer Inst. 59: 301-307. 1977.Sandberg, A. A.. Ishihara. T., Moore, G. E., and Pickren, J. W. UnusuallyHigh Polyploidy in a Human Cancer. Cancer, 16: 1246-1254. 1963.

Tom. B., Rutzky. L. P., Jakstys. M. M., Oyasu. R., Kaye, C. I., and Kahan,B. D. Human Colonie Adenocarcinoma Cells. I. Establishment andDescription of a New Line. In Vitro. 12: 180-191, 1976.Tompkins. W. A. F., Watrach, A. M.. Schmale. J. D., Schultz, R M.. andHarris, J. A. Cultural and Antigenic Properties of Newly Established CellStains Derived from Adenocarcinomas o' the Human Colon and Rectum.J. Nati. Cancer Inst. 52: 1101-1106, 197*.

Fig. 1. Histology section from original tumor specimen. H & E, x 10.

Fig. 2. Example of tumor mass grown in heterologous host. Section is from COLO 201. H & E, x 10.Fig. 3. Phase-contrast photomicrograph of COLO 201. Note predominance of bipolar cells, x 16.

Fig. 4. Phase-contrast photomicrograph of COLO 205. Note predominance of polygonal, epithelial-like morphology, x 16.

Fig. 5. Phase-contrast photomicrograph of COLO 206 showing small, rounded cells x 16.

Fig. 6. Electron micrograph of COLO 201 revealed polygonal epithelial-like morphology with interdigitating brush border x 8900.

Fig. 7. Electron micrograph of lymphoblast cell line COLO 200 revealed numerous free ribosomes and regular nuclei, x 8900.

Fig. 8. Electron micrograph of COLO 200 versus COLO 201. Note attachment of lymphoid cell to the tumor cell microvilli. x 15,000.Fig. 9. a, Giemsa-banded karyotype representative of COLO 201 and COLO 205; o, Giemsa-banded reference marker M and C-banded M, indicating

heterochromatic band, x 100.Fig. 10. a, Giemsa-banded karyotype of COLO 206; b, additional example of Giemsa-banded markers from COLO 206. x 100.

Fig. 11. Giemsa-banded karyotype of (a) COLO 197 and (D) COLO 200 x 100.




r Â«â€¢â€¢^a^ÃT*w

^t

ff- 'v v S*" >Â».V^*.>'"â€¢


â€¢¿�â€¢â€¢â€¢â€¢I

^3Ã•:: Ã•I

i \ V 4

MAY 1978 1351



T. U. Semple et al.

vâ€¢¿�

â€¢¿�irÂ«Â»'â€¢¿�-3R

*- ' * Â»O

'^â€¢r





COLO 201 & COLO 205

KÃ¬R W if

10 11 12

>M Â»Â«EÂ«| rillt Si13 14 15 16 17 18

iti mt Â«Â»Â»*â€¢19 20 21 22 XX

ÃŽÃŽJÃ•C f I lÃ Â»Â»MO ^"3 4 5 fi 7 2^ d

rM1 M1

G- C- G- C-

Banding Banding

9b

MAY 1978 1353



T. U. Semple et al.

COLO 206 10a

*â€¢*w*m mW

s

Â¥ - *. Ã¯**

10 11 12

Â«Mil fca a* â€¢¿�*i3 14

19

15 16 17 18

20 21 22 X Y

*% Ã±

I sii U t Â«IÂ«M,

10b

II H3 M1 7 M, 2 13 8 M, 10 MR 19 M,





COLO 197

If li C< II If123 45

it Â¡i II 13 II Ã®Ã¯6 7 8 9 10 11 12

â€¢¿�Â»II M il i*13 14 15 16 17 18

Ulti â€¢¿�* Â«*19 20 21 22

IÂ» 11aCOLO 200

m A A

I ? Il il I! HÂ«i Â¿i i Ã¡

10 11 12

tÃ© ii ii 91 U li13 14 15 16 17 18

if *â€¢

I19 20 21 22

X Y 11b

MAY 1978 1355



Documents

Tumor and Lymphoid Cell Lines from a Patient with Carcinoma of … · 201, COLO 205, and COLO 206 were derived from an ascites fluid specimen. COLO 197 had a normal pattern of estab