cell lines from human colon carcinoma with unusual cell ... · colo 320 was examined at 1.5, 3.0,...

[CANCER RESEARCH 39, 4914-4924. December 1979]0008-5472/79/0039-0000$02.00

Cell Lines from Human Colon Carcinoma with Unusual Cell Products,Double Minutes, and Homogeneously Staining Regions1

Laurie A. Quinn, George E. Moore,2 Robert T. Morgan, and Linda K. Woods

Department of Surgery, Division of Surgical Oncology, Denver General Hospital, Denver, Colorado 80204

ABSTRACT

Two human colon carcinoma cell lines derived from the sametumor specimen were characterized. The cell lines, COLO 320and COLO 321 , have amine precursor uptake and decarboxylation cell properties, such as ectopic production of norepinephrine, epinephrine, serotonin, adrenocorticotropic hormone, and parathyroid hormone. The cells were morphologically different from most colon cell lines. Double minutes (DM)were initially present in nearly 100% of the metaphases. In afew subcultures of COLO 320, DM have persisted for 1.5years. However, in COLO 321 and some subcultures of COLO320, a loss of DM was observed and new marker chromosomeswith homogeneously staining regions were observed. Theseunusual cell lines should be valuable for studies of apudomasof the colon and the cytogenetic phenomena of DM and hornogeneously staining regions.

INTRODUCTION

Cultured malignant cell lines provide a valuable resourcematerial for cancer research. Large numbers of cell lines fromdifferent tumor types are required to reflect the diversity of celltypeandcell responseof tumorsobservedby a clinician.Sequential biochemical, physiological, and immunologicalanalyses made possible by the accessibility of the culturedcells can lead to improved and new methods for accuratediagnosis, define new subclassifications of tumors, aid in assessrnent of prognosis, and suggest new approaches for developrnent of anticancer agents.

Cell lines derived from human colon carcinomas have beenreported (9, 20, 26, 38, 42, 43, 46). Ultrastructural studiesreveal brush borders and numerous cell junctions (9, 20, 38,46). Synthesis of CEA3 is the most frequently reported biochemical marker of malignant colonic cell lines. In this report,we describe 2 colon carcinoma cell lines, derived from thesame tumor specimen, with characteristics that differ significantly from previously reported colon carcinoma lines. Thepossible origin of these unique cell lines, unusual cell products,and 2 cytogenetic aberrations, DM and HSR's are the reasonsfor this report.

1 This investigation was supported by Grant i -ROi -CAl 501 8, awarded by the

National Cancer Institute, Department of Health, Education and Welfare, and bythe Mary B. and L. H. Marshall Fund.

2 To whom requests for reprints should be addressed.

3 The abbreviations used are: CEA, carcinoembryonic antigen; DM, double

minutes; HSR, homogeneously staining region; FBS, fetal bovine serum heatinactivated at 56Â°for 30 mm; COLO, prefix for cell lines established by ourlaboratory in Colorado; ACTH, adrenocorticotropic hormone; PTH, parathyroidhormone; ESD, esterase D; APUD, amine precursor uptake and decarboxylate;DHFR, dihydrofolate reductase.

Received May i4, 1979; accepted August 27, i979.

A 55-year-old Caucasian female was diagnosed as having acarcinoma of the sigrnoid colon after a brief interval of pain anddysuria. Preoperative levels of electrolytes, minerals, and enzymes were normal. The CEA level of I 5 mg/mI in a nonsmokerwas consistent with an extensive tumor. On May 11, 1977, aninvasive carcinoma of the colon was resected in continuity withthe dome of the urinary bladder. Surgical specimens wereobtained for pathology studies and tissue culture.

A majority of the histology sections revealed a moderatelyundifferentiated adenocarcinoma of the colon with atypicalpseudo-gland formation in the muscle and serosal layers (Fig.1). The cytological details of the invasive glands and clustersof cells were not unusual, except for one area in which thetissue spaces between the malignant glands were completelyfilled with a second kind of epithelial-like cells that wereâ€˜â€˜suggestiveof a poorly differentiated carcinoid. â€˜â€No rosette,cord, or duct-like structures were observed. The cells werelarge; immaturity was emphasized by the large globular nucleusand large nucleolus. Mitotic figures were infrequent.

Clinical episodes that might be related to our subsequentfindings included a severe hyponatremia in the postoperativeperiod, labile blood pressure, hypercalcemia (15.4 mg-i 00 ml,and worsening liver function tests. Chemotherapy was initiatedpostoperatively, but the patient expired 4 months later.

MATERIALS AND METHODS

Fragments of the colon carcinoma were minced, placed inglass flasks, covered with a thin layer of GEM 17i 7 (developedat Denver General Hospital) or Roswell Park Memorial InstituteMedium 1640 supplemented with 10% heat-inactivated FBSand incubated at 37Â°.After 24 hr, the cultures were tightlycapped and placed in a dry incubator at 37Â°.Every 2 to 4days, 1 to 5 ml of fresh medium were added to the cell cultures.At no time was the medium completely replaced with freshmedium. Subcultures were made by shaking the flask andpouring a portion of the medium and suspended cells into anew culture flask. The original culture flasks were retained.When successful subcultures solely comprised of tumor cellswere obtained, the cultures were considered established. Cellcultures were periodically tested for mycoplasmas by the bisbenzarnide fluorescent-staining method of Chen (7).

Heterologous transplantation of the cultured COLO 320 cellswere performed, as previously described (38).

Cells grown on coverslips were used for phase and otherlight microscopy studies. HCI-toluidine blue and silver stainswere performed according to the method of Grirnelius (13).Samples of cells were prepared for electron microscopy asdescribed by Mallory (25).

Cell products were assayed in 7-day-old spent-culture mediurn. Parallel assays were performed on fresh medium supple

4914 CANCER RESEARCH VOL. 39

CASE REPORT

Research. on October 20, 2020. © 1979 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

Colon Carcinoma Cell Lines with DM and HSR s

Fig. 1. Biopsy from the primary colon carcinoma. H & E, x 130.

merited with FBS. Assays for ectopie production of calcitoninand ACTH were performed by Nichols Institute (San Pedro,Calif.). PTH, a-fetoprotein, and alkaline phosphatase levelswere determined by Associated Laboratories (Wichita, Kansas). Analyses of catecholamines were performed by Laboratory Procedures (Woodland Hills, Calif.). Serotonin synthesiswas assayed by Bioscience Laboratories (Van Nuys, Calif.).

The cell lines were tested for CEA production by a radioimmune assay kit (Roche Diagnostics, Nutley, N. J.) in April 1978and January 1979. Seven-day-old spent medium supplemented with 10% FBS from a culture with 107 viable cells and

fresh culture medium supplemented with 10% FBS were usedfor the analysis of CEA levels. COLO 320 was also tested forCEA by Nichols Institute.

Isozyme phenotypes were analyzed in the laboratory of Dr.J. Fogh (Sloan-Kettering Institute). Starch-gel electrophoresis

(14) of tumor cell cytosols was performed for the followingisozymes: "red cell" acid phosphatase, adenosine deaminase,

first and third loci of phosphoglucomutase, ESD, and mito-chondrial glutamate-oxaloacetate. The cytosol from 107 tumor

cells lysed by freeze-thaw in 0.5 ml 0.9% NaCI solution wasused to examine glucose 6-phosphate isozymes by cellulose-

acetate electrophoresis, as described previously (38).Standard cytogenetic methods were used to obtain chro

mosome preparations. G-, Q-, and C-banding were obtained

by the methods of Lavappa (19), Lin era/. (23), and McKenzieand Lubs (27). COLO 320 was examined at 1.5, 3.0, 11.0,16.0, and 20.0 months after initiation of the culture. Some ofthe cytogenetic preparations at 3, 11, and 16 months weremade from the same subculture. COLO 321 was examined 1.5,11, and 20 months postinitiation. Twenty to 75 metaphaseswere examined, and a minimum of 10 karyotypes were madefrom each chromosome harvest. The Hoechst-bromodeoxyuri-

dine methods of Latt (18) and Goto et al. (10) were used tostudy the replicative pattern of the chromosomes and DM.

RESULTS

The primary cultures began to rapidly metabolize the culturemedium within the first 24 hr postinitiation and were subcul-

tured by pouring half of the supernatant into new flasks. Thecultures initiated with Roswell Park Memorial Institute Medium1640 supplemented with 10 and 15% FBS continued to proliferate. It was possible to split the cultures 1:2 every 2 or 3 days.Fibroblasts were not apparent in the cultures. Forty-five days

postinitiation, 2 of the original primary cultures and the subcultures derived from them were designated COLO 320 andCOLO 321. Mycoplasmas were observed in some subculturesafter approximately 1 year.

Heterologous Transplants. Tumor nodules were apparent 8days postinjection of 8 x 106 cells (COLO 320) in all 3 of the

irradiated C3H/HeJ mice used as heterologous hosts. Examination of histology sections of the tumor nodules revealedpoorly differentiated tumors with large areas of necrosis.

Morphology. The cells of COLO 320 and COLO 321 proliferated both attached to the culture flask and in suspension.Phase microscopy revealed small flat clumps and isolated smallround cells (Fig. 2,4). Occasionally, elongated or multinucle-

ated cells were observed.Coverslip preparations were stained with acidified toluidine

blue or Grimelius' silver nitrate. These stains were used to

ascertain whether secretory granules, often present in endocrine and paraendocrine cells, could be observed. Neither stainrevealed the presence of secretory granules.

The original tumor specimen was not examined by electronmicrography at the time of surgical excision. A tissue blockwas later deparaffinated, refixed, and stained for electronmicrosropy. The overall quality of the preparation was poor.Few organelles were observed in this preparation. One cellappeared to have 10 to 15 secretory granules. The nuclearmaterial had numerous crystalloid structures that probablyresulted from the preparation of the specimen for electronmicroscopy.

Electron microscopic studies of COLO 320 and COLO 321were similar. The cells were spheroid. Microvilli were absent.Occasional desmosomes were observed. The presence of multiple convoluted Golgi apparatuses reflected a possible secretory activity. The dilated endoplasmic reticulum was moderatelyrough. Single ribosomes and polysomes were numerous. Cy-

toplasmic fibrils were observed in the perinuclear region nearvesicles and near the cytoplasmic membrane. Large lipid vacuoles, a few clear vacuoles, and rare dense osmiophilic vacuoles, similar in size to seer .tory granules observed in endocrinecells, were present. The dense osmiophilic vacuoles were morefrequent in COLO 321 than in COLO 320. A representativemicrograph of COLO 321 is depicted in Fig. 26.

Cell Products. Levels of hormones and polypeptides testedin spent medium from COLO 320 are summarized in Table 1.PTH, ACTH, serotonin, norepinephrine, and epinephrine werepresent in quantities above control levels. Alkaline phosphatase, calcitonin, and dopamine were not detectable. Â«-Fetopro-

tein and CEA also were not present in detectable quantities.Isozymes. Results of the isozyme analysis for COLO 320

were as follows: red cell acid phosphatase type BA, adenosinedeaminase type 2, first locus of phosphoglucomutase type 1,third locus of phosphoglucomutase type 2, ESD type 1, mito-chondrial glutamate-oxaloacetate transaminase type 1, andglucose-6-phosphate dehydrogenase type B. It should benoted that the ESD isozyme was present in much greaterquantity than were the other isozymes in COLO 320 and 4other cell lines tested at the same time.

DECEMBER 1979 4915



CeII productSero

Norepi EpiPTHtoninSpentme Calci a-Feto nephrinenephrlneDopACTH(pg/(ng/diumtoninCEAprotein(pg/mI)(pg/mI)amine(pg/mI)ml)ml)COLO

320NegativeNegativeNegative4852Negative29.3377090RPMI1640 + 10%NegativeNegativeNegative1 46NegativeNegativei24045FBSmedium

L. A. Quinn et al.

Table iCell product assays (COLO 320)

@l _@ __@@ - I A

Cytogenetics. The earliest cytogenetic preparations fromCOLO 320 and COLO 321 revealed numerous DM in nearly100% of the metaphases. The DM were darkly stained withGiemsa (Fig. 2C), lightly stained with trypsin G-banding (Fig.2D), and pale with quinacrine dihydrochloride stain. Centromeric heterochromatin, as revealed by C-banding was absentin DM(Fig. 2E). Hoechst-bromodioxyuridinedifferentialstaining confirmed that the DM replicated in vitro (Fig. 2F). Thenumber of DM per metaphase varied from 2 to several hundredin both COLO 320 and COLO 321 (Chart 1).

The chromosome mode of COLO 320 was 51 . Chromosomebanding was performed on COLO 320; a complete metaphaseand markers from 4 additional metaphases are represented inFig. 3. Our analysis of the chromosome markers was as follows:4

Ml : del(i ) (qterâ€”sp3i:)M2: t(i ;?) (1qterâ€”+ip36::?)M3: t(i ;5)(i pterâ€”.i p23 or 1 p1 3::Spi 3 or 5pi 5â€”*Sqter)

M4: B-like marker with dark bands near the centromereand lower one-third of the long arm.

M5: t(2?;i 1?;i 4) (14qterâ€”+cen.-+2?q3?::i1?qi 3â€”+11?qter)

M6: t(2?;?;i 4) (14qterâ€”+14q1 ?::2?pI?â€”s2q3?::?)M7: t(2;?) (2qterâ€”*2p2?::?)M8: t(4;7?) (4pterâ€”s4q3?::7?qi?â€”*7?qter)M9: t(5;?) (Sqterâ€”sspi ?::?)MiO: del(3)(pterâ€”+qi32: or qi33:)Mi 1: t(3;9) (3qterâ€”+3q2i::9p2?â€”+9qter)Mi 2: t(9;i 5) (9qterâ€”scen--*iSqter)M13: t(13;i?)(i3qterâ€”.i3q1i::i?q2i--@i?p2i:)M14: dic(13;i3) (i3qterâ€”*i3pl?::i3pl?-â€”*i3qter). One

centromeric constriction was more pronouncedthan the other.

Ml 5: Small metacentric; possible 11p to 11qi 4 withsome additional light-staining material.

Mi 6: E-size submetacentric; possible a deleted i 2q.Mi 7: t(i 9;?) (19::?)Mi8: t(22;?)(22pterâ€”*22q13::?)Mi 9: del(22) (pterâ€”@q11:)M20: Small isochromosome

Variable sizes of small ring chromosomes were also observedin 10 to 15% of the metaphases. When ring chromosomeswere present, the number of rings per metaphase varied from

4 Nomenclature for chromosomes is according to Rets. 3i and 32.

1_@ - - -- - B

30,

C

â€˜â€¢@-r@-â€”r'@-r@-iâ€” -k-â€˜@@@ --r-,-

L@

JJ'â€¢t'@fâ€¢@'iâ€•A@J0@ -at

Numbov of DMs p.r m.tWM$â€¢

Chart 1. Numbers of DMs per metaphase. A, 1.5 months postinitiation ofCOLO 320; B, 3.0 months postinitiation of COLO 320; C, 11 months postinltlationof COLO 320; D, 16 months postinitiation of COLO 320, and first observation ofa HSRmarker in one subculture; E, 16 months postinitlatlon of another subculturewith very few metaphases with HSR markers.

1 to 7. The complement of normal chromosomes per metaphase varied from 38 to more than 100. Chromosome 5 wasobserved least frequently. Multiple copies of chromosome 20were common.

A startling change was observed in the cytogenetic analysisof one subculture after 16 months of continuous culture. DMwere absent in 49% of the metaphases, and 23% of themetaphases had only i DM (Chart 1). A new marker chromosome with a long HSR was present in 94% of the metaphases.A marker with a HSR was present in all cells with low (1 to 7)numbers of DM. The HSR was observed distal to the Xq21region, and an additional dark band was located at the terminalend of the HSR (Figs 4 and 6A). The variant HSR marker hada shorter HSR in the long arm of the X, an additional darkGiemsa band after Xp21 band, and additional HSR. One exceptional metaphase had 37 DM and a HSR marker. Multiplecopies of the HSR marker were not observed. The few cellswithout HSR markers had 15 to hundreds of DM per cell.

Analysis of the markers present revealed that the cell population with HSR marker chromosomes was more homo9eneous than prior analyses showed. MS to M7 of COLO 320-DM

[email protected]..@ â€”I I

20-

DM

4916 CANCERRESEARCHVOL. 39



Colon Carcinoma Ce!! Lines with DM and HSR â€˜s

were seldom observed. The chromosome mode shifted upwardby 2, but we were unable to detect the cause of the altered

mode. One complete karyotype and examples of markers from3 additional metaphases are depicted in Fig. 5. Our analysis ofthe markers was as follows:4

Same as Mi of COLO 320-DMSame as M2 of COLO 320-DMSame as M3 of COLO 320-DMt(2;?) (2qterâ€”+2p1?::?)Same as M10 of COLO 320-DMt(4;?)(4qterâ€”+4p1?::?)Same as Mi 1 of COLO 320-DMt(9;1 2?) (9qterâ€”.9p2?::i2?ql 3â€”*12qter)Same as Mi 2 of COLO 320-DMt(1 1;?) (1 i qterâ€”+i1p1?::?)t(1 3;20) (1 3qterâ€”@i3qi 1::2Oqi ?â€”*20pteror

dic(1 3;20) (i 3qterâ€”@13pi ?::2Oqi ?â€”*20pter)Same as Mi 3 of COLO 320-DMSame as M14 of COLO 320-DMt(i 4;?) (14qterâ€”+cen--+?)t(i 5;?) (15qterâ€”+cen--*?)Same as Mi 7 of COLO 320-DMSame as M16 of COLO 320-DMSmall metacentricVery small unidentifiable piece of chromatin(X;HSR:?) (Xpterâ€”*Xq2i ?::HSR::?)(HSR;?;X;HSR;?) (HSRâ€”@?::Xp22'?--@Xq21 ?::HSR:

After the HSR marker was observed in one subculture, othersubcultures of COLO 320 were examined. One subculturecontinued to have DM and marker complements very similar tothe earliest cytogenetic preparation. Another subculture hadHSR markers in 65% of the metaphases. The remaining cellshad variable numbers of DM. The HSR markers were morphologically similar to those previously seen in a COLO 320-HSRsubculture.

All of the subcultures of COLO 320, examined at 20 months,appeared to have approximately the same cytogenetic patternsas did the subcultures analyzed at 16 months, i.e. , culturesthat previously had HSR markers continued to have thesemarkers; cultures that had DM in the majority of the metaphasescontinued to have the same percentage of metaphases withDM. The earliest cytogenetic harvests were reexamined; noHSR markers were observed.

COLO 321 was not studied as extensively as was COLO320. The chromosome mode was 48. Fewer copies of chromosomes10, 15, 16, 2i , and 22 wereobservedin COLO321than in COLO 320. Present in 98% of the metaphases were 2copies of a long acrocentric chromosome with the stainingpropertiesof a HSR(Fig.6B).Thepresenceof thesemarkerswas first brought to our attention by Dr. Frances Arrighi. Oneto 3 DM were observed in many of the cells. In most cases, 2DM were larger than the average size DM observed in theearliest preparations. The HSR's were probably associatedwith a D-group chromosome. A dark Giemsa band was evidentnear the centromere. The remainder of the chromosome(s)stained more uniformly than did normal human chromosomes,but small very minor bands could be observed in some metaphases. A darkly staining band was observed near the distal

end of the chromosome arm in some of these markers. Onekaryotype and markers from 4 additional metaphases are depicted in Fig. S. Our analysis of the chromosome markers wasas follows:4

Ml : Same as Mi of COLO 320-DM and Mi of COLO320-HSR

M2: Same as M2 of COLO 320-DM and M2 of COLO320-HSR

M3: Same as M3 of COLO 320-DM and M3 of COLO320-HSR

M4: Same as Mi 0 of COLO 320-DM and MS of COLO320-HSR

M5: Same as Mi 1 of COLO 320-DM and M7 of COLO320-HSR

M6: Same as Ml 2 of COLO 320-DM and M9 of COLO320-HSR

M7: t(i 0;?) (10qterâ€”@iOpi ?::?)M8: t(X;i 3) (Xqter-+Xp22::1 3qi 1â€”*i3qter)M9: dic(i 3;20) (1 3qterâ€”@i3pi ?::2Oqi ?::2Opter) prob

ably the same as Mi 1 of COLO 320-HSRMl 0: t(i 4;?) (i 4qterâ€”+cen-.+?)same as Mi 4 of COLO

320-HSRE-size submetacentricE-size metacentricE-size submetacentric; probably same derivation as

M16 of COLO 320-DM and Mi7 of COLO 320-HSR

M14: Small isochromosome

DISCUSSION

APUD cells were initially characterized by a capacity toproduce catecholamines or indolamines after uptake and decarboxylation of amine precursors (34, 35). Cell products suchas cholinesterases, nonspecific esterases, and a-glycerylphosphate dehydrogenase as well as secretion of a variety ofamines and polypeptides are now considered more consistentproperties of an APUD cell (41). Pearse theorized that theAPUD cells were probably of neurogenic origin (33).

Biochemical studies of COLO 320 suggest that the line is ofneuroendocrine origin rather than of colonic adenoepithelialorigin.Endocrinecellsbetweenadenoepithelialcellshavebeenreported (5, 6, 8). Malignant transformations are rare. Histological and ultrastructural studies reveal numerous secretorygranules in the cytoplasm of colonic endocrine cells. Thenumber, size, and staining characteristics of the granules arevariable (5, 8). Information on the hormones released by thecolon endocrine cells is minimal (44). However, serotonin,enteroglucagon, Substance P, vasoactive intestinal peptide,somatostatin, and/or a number of less defined amines orpeptides have been detected (5, 8, 11, 34, 35).

The pathology of the tumor specimen indicated that 2 celltypes were present. One cell type was typical of an adenocarcinoma. The second cell type was very undifferentiated andresembled a carcinoid. Unfortunately, we were unable to obtaingood quality ultrastructural studies on the original tumor specimen, which may have clarified the patient's diagnosis. Theonly clinical symptom suggestive of an apudoma was hypercalcemia. Possibly, the patient had 2 types of cancer, ormalignant cells of an original clone mutated and resulted in thedevelopment of new clonal types.

Mi:M2:M3:M4:M5:M6:M7:M8:M9:M10:Mu:

Ml 2:Mi3:Mi4:M15:Mi6:Mi 7:Mi8:M19:M20:M2i:

Mu:M12:Mi3:

DECEMBER 1979 4917



L. A. Quinn et a!.

COLO 320 and COLO 32i differ considerably from previously reported colon carcinoma cell lines (9, 20, 26, 38, 43,46). In most cases, colon carcinoma cell lines grow slowly,adhere tightly to the culture flask, form dense cell colonies,have cell membranes with interdigitating brush borders, andproduce variable quantities of CEA. In contrast, COLO 320 andCOLO 321 grow very rapidly, high-density epithelial coloniesare not apparent, and the cells are easily loosened from theculture flask. Ultrastructure studies revealed relatively uniform,round cells with few microvilli or desmosomes. Repeated assays of COLO 320 and COLO 32i were negative for CEA.

Assays for endocrine-like products indicated that COLO 320secreted serotonin, PTH, ACTH, catecholamines, and elevatedlevels of ESD isozyme, which are characteristics common toAPUD cells. None of the previously reported colon carcinomacell lines are known to produce these cell products. However,the question remains as to whether all such endocrine-secreting cancers arise from precursor APUD cells, or whether transformed intestinal cells may become so undifferentiated thatthey develop the capacity to produce polypeptides with endocrine function, just as a variety of tumor types produce CEA.

Cytogenetic analyses of COLO 320 and COLO 32i revealeda number of marker chromosomes derived from structuralrearrangements of normal chromosomes. There have beenvery few banded cytogenetic characterizations of colon carcinomas (i 6, 38, 39, 42). Therefore, substantative conclusionsconcerning whether there are nonrandom patterns of chromosome involvement in markers are not possible at this time.Furthermore, these unusual cell lines may not have the samechromosome aberrations that might be found in adenocarcinomas of the colon.

The most striking cytogenetic feature of the cell lines wasthe presence of numerous DM. Tumors of neurogenic originare most frequently reported to have DM (22, 24). It is temptingto suggest that the presence of DM in COLO 320 and COLO32i is further evidence of the neuroendocrine origin of thecells. However, DM have been observed in a number of otherhuman tumor types (3, 12, i 5, 28, 29, 36, 40, 45). In a surveyof 35 of our malignant tumor cell lines, at least one metaphaseof 75 metaphases examined had DM. The number and frequency of DM per cell varied considerably among the differentcell lines. Only one additional cell line, a carcinoma of therectum with atypical morphology, had large numbers of DMsimilar to COLO 320 and COLO 321 . The significance of DMin malignant cells has not yet been elucidated.

Although we have no direct evidence, DM were probablypresent in the original tumor. DM have been observed in freshtumor specimens and cultured tumor lines of a variety oftumors. Studies by Levan et a!. (22), Barker and Hsu (3), andourselves have demonstrated that DM replicate synchronouslywith other chromosomes and, therefore, have the potential topersist in vitro. Our studies also indicate that in some subcultures, the DM have persisted in nearly the same frequency forapproximately i .5 years.

Recently, there has been speculation that DM are related tothe de novo apeparance of unusual chromosomes or markerswith long HSR's (2, 2i ). Levan et a!. (2i ) described a murinesarcoma line, SEWA1A, that had DM when passaged in vivo.The DM were lost after 3.5 months of in vitro cultivation, andnew marker chromosomes appeared with the same stainingproperties as those of DM. When the in vitro line was again

passaged in animals, the DM reappeared in the same frequencyas in the original. tumor. Balaban-Malenbaum and Gilbert (2)described a human neuroblastoma cell line that had 2 cellpopulations. In one cell population, DM were observed. In thesecond cell population, no DM were evident, but marker chromosomes with HSR's were present. Since there were markerchromosomes common to both cell populations, it was probable that both cell populations were derived from a commonprecursor.

We observed a similar phenomenon in some subcultures ofCOLO 320 and COLO 321 . In both cases, DM were observedin metaphases from the earliest cytogenetic analyses. A number of subcultures lost DM and gained HSR markers after 1 to1.5 years. Although occasional metaphases were observedthat had a few DM in addition to the HSR markers, the majorityof the metaphases had either DM or HSR markers. COLO 321had a few DM that were observed in addition to the HSRmarkers. Whether these DM are different from the numerousDM observed in early harvests cannot be determined at thistime. In any case, the number of DM was drastically reducedin 99% of the cells with HSR markers.

The basis for the sudden appearance of a new marker witha HSR and concurrent disappearance of DM is not clear.Alterations of in vitro conditions, such as addition of antibioticsand variations in percentages of FBS supplement, have notresulted in the formation of HSR markers in cultures with DMor gain of DM in cultures that had cell populations with HSRmarkers. It is conceivable that tumor cells with HSR markersand without DM were present in a very small percentage ofcells of the original tumor. kovacs (17) reported HSR's in directpreparations in 5 of 14 malignant specimens.

Levan et a!. (21) suggested that cells with HSR's may havea selective advantage in vitro over cells that have DM. Certainly,one subculture of COLO 320 very rapidly became dominatedby cells with a HSR marker. However, other subcultures derived from the flask that later had HSR markers in nearly all ofthe cells and parallel subcultures have had concomitant cellpopulations with either HSR markers or DM for approximately6 months.

Analysis of the markers present in the subculture dominatedby cells with a HSR revealed less variety in the complementsof markers than was found in the early preparations. Themarkers were not sufficiently uniform to conclude that all cellswith a HSR were derived from a single cell that overgrew theother cells with DM. In vitro instability, selection processes, orchromosome harvest technique-induced random losses ofchromosomes from metaphases could account for some of thevariability in marker complements.

The most recent theory for the origin of DM and HSR'ssuggests that gene amplification is responsible for the development of these aberrations. Biedler and Spengler (4) correlated the appearance of HSR's in methotrexate-resistant cellswith manifold increases in the levels of DHFR. Others havedemonstrated that induced methotrexate resistance in mousesarcoma cells results from increased numbers of copies of theDHFR genes (1, 37). !n situ hybridization studies have confirmed that DHFR genes are present in the HSR's of methotrexate-resistant cells (30). No specific cell products have yetbeen associated with HSR's or DM occurring in human tumorsor human tumor lines. If gene amplification is the origin of bothDM and HSR's, these abnormalities may have biological sig




Colon Carcinoma Cell Lines with DM and HSR â€˜s

the double minutes in a mouse tumor. Cyto9enet. Cell Genet., 20: 12â€”23,1978.

22. Levan, A., Levan, G., and Mitelman, F. Chromosomesand cancer. Hereditas,86: 15-30, 1977.

23. Un, C. C.. van de Sande, H., Smink, W. K., and Newton, D. R. Quinacrinefluorescence and 0-banding patterns of human chromosomes. i . Effects ofvarying factors. Can. J. Genet. Cytol., i 7: 81â€”92,1975.

24. Lubs, H. A., Jr., Salmon, J. H., and Flanigan, S. Studies of a glial tumor withmultiple minute chromosomes. Cancer (Phila.). i 9: 591-599, 1966.

25. Mallory, F. B. Pathological Technique, p. 130. New York: Hafner Press,1961.

26. McCombs, W. B.. Leibovitz, A., McCoy, C. E., Stinson, J. C., and Berlin, J.D. Morphologic and immunologic studies of a human colon tumor cell line(SW 48). Cancer (Phila.), 38: 2316â€”2327,1976.

27. McKenzie, W. H., and Lubs, H. A., Jr. An analysis of the technical variablesIntheproductionof C-bands.Chromosoma(Berl.),4i : 175â€”182, 1973.

28. Miles, C. P. Chromosome analysis of solid tumors. I. Twenty-eight nonepithelial tumors. Cancer (Phila.), 20: 1253â€”1273, 1967.

29. Moore, G. E., Woods, L. K., Quinn, L. A., Morgan, R. T., and Semple, T. U.Characterization of cell lines from four undifferentiated human malignancies.Cancer (Phila.), in press, 1979.

30. Nunberg, J. H., Kaufman, R. J., Schimke, R. T., Urlaub, G., and Chasm, L.A. Amplified dihydrofolate reductase genes are localized to a homogeneouslystainingregionof a singlechromosomein a methotrexate-reslstantChinese hamster overy cell line. Proc. NatI. Acad. Sci., U. S. A., 75: 5535â€”5556, 1978.

31 . Paris Conference (1971). Standardization In human cytogenetics. In: D.Bergsma (ser. ed), Birth Defects Original Article Series, Vol. 8, No. 7. WhitePlains, N. Y.: National Foundation-March of Dimes, i 972.

32. Paris Conference (i 971). Supplement (1975). Standardization in humancytogenetics.In:D.Bergsma(ser.ed), BirthDefects,OriginalArticleSeries,Vol. 11, No. 9. White Plains, N. Y.: National Foundation-March of Dimes,1975.

33. Pearse, A. G. E. The cytochemistry and ultrastructure of polypeptide hormone-producing cells of the APUD series and the embryologic, physiologicand pathologic implications of the concept. J. Histochem. Cytochem., i 7:303-313, 1969.

34. Pearse, A. G. E., and Polak, J. M. Endocrine tumors of neural crest origin:neurolophomas, apudomas and the APUD concept. Med. Biol. (Helsinki),52: 3â€”18,1974.

35. Pearse, A. G. E., and Welbourn, R. B. The apudomas. Br. J. Hosp. Med.,iO:6i7â€”624, 1973.

36. Pierre, A. V., Hoagland, K. C., and Linman, J. W. Microchromosomes inhuman preleukemia and leukemia. Cancer (Phila.), 27: 160â€”i75, 1971.

37. Schimke, R.T., Kaufman, A. J., Alt, F. W., and Kellems, R. Geneamplificationand drug resistance in cultured murine cells. Science, 202: 1051â€”1055,1979.

38. Semple, T. U.. Quinn, L. A., Woods, L. K., and Moore, G. E. Tumor andlymphoid cell lines from a patient with carcinoma of the colon for a cytotoxicity model. Cancer Res.. 38: 1345â€”1355, 1978.

39. Sonta, S., and Sandberg, A. A. Chromosomes and causation of humancancer and leukemia, XXX. Banding studies of primary Intestinal tumors.Cancer(Phila.), 4i: i64â€”i73, 1978.

40. Spriggs, A. I., and Bodington, M. M. Chromosomes of human cancer cells.Br. Med. J., 2: 1431-1435, 1962.

4i . Tlschler, A. 5. Small cell carcinoma of the lung: cellular origin and relationship to other neoplasms. Semin. Oncol., 5: 244â€”252,1978.

42. Tom, B., Rutzky, L. P., Jakstys, M. M., Oyasu, R., Kaye, C. T., and Kahan,B. D. Humancolonic adenocarcinoma cells. 1. Establishmentand descriptionofnewline.lnVitro, i2: 180â€”19i,i976.

43. Tompkins, W. A. F., Watrach, A. M., Schmale, J. 0., Schultz, R. M., andHarris, J. A. Cultured and antigenic properties of newly established cellstrains derived from adenocarcinoma of the human colon and rectum. J.NatI. Cancer Inst., 52: 1i0iâ€”1106, 1974.

44. Welbourn, A. B. Apudomas of the gut. Am. J. Surg., i33: 13â€”22,1978.45. White, L., and Cox, D. Chromosomechanges in a rhabdomyosarcoma during

recurrence and in cell culture. Br. J. Cancer, 21: 684â€”696,1967.46. Yaniv, A., Altboum, Z., Gazit, A., Block-Shtacher, N., and Elyan, E. Estab

lishment and characterization of a cell line derived from a human colonadenocarcinoma (HuCCL-i 4). Exp. Cell Biol., 46: 220â€”230,1978.

4919

nificance for tumor cell development and survival against thedefenses of the host and cancer therapies.

COLO 320 and COLO 321 should provide an accessiblesource of cells for studies of the biochemical properties ofunusual colon carcinoma and the cytogenetic phenomenon ofDM and HSR's.

ACKNOWLEDGMENTS

The authors thank V. Buric for electron microscopy.

REFERENCES

1. Alt, F. W., Kellems, R. E., Bertino, J. R., and Schimke, R. T. SelectivemultIplIcationof dlhydrofolatereductasegenesin methotrexate-resistantvariantsof culturedmurinecells.J. Biol.Chem.,253: 1357â€”i370, i 978.

2. Balaban-Malenbaum, G., and Gilbert, F. Double minute chromosomes andhomogeneouslystainingregionsinchromosomesofahumanneuroblastomacellline.Sclence, 198:739â€”741,1977.

3. Barker, P. E., and Hsu, T. C. Double minutes in human carcinoma cell lines,with specIal reference to breast tumors. J. NatI. Cancer Inst., 62: 257â€”262,1979.

4. Bledler, J. L, and SpongIer, B. A. Metaphase chromosome anomaly: assoclatlon wIth drug resistance and cell specific products. Science, i 9 i : 185â€”187, 1976.

5. Buffa, R., Capella, C., Fontana, P., Usellini, L., and Solcia, E. Types ofendocrInecellsInhumancolonandrectum.CellTissueRes.,i 92:227â€”240, 1978.

6. Calvert, R. Sequential differentiation of intestinal endocrine cells in the fetalmouse.CellTissueRes.192:267â€”276,1978.

7. Chen, T. R. Microscopic demonstration of mycoplasma contamination in cellcultures and cell culture media. Tissue Culture Association Manual, i : 229â€”232, 1975.

8. ChrIstina, M. L., Lehey, T., Zeitun, P., and Dufougeray, F. Fine structuralclassification and comparative distribution of endocrine cells in normal andhuman large Intestine. Gastroenterology, 75: 20â€”28,1978.

9. Drewinko, B.. Romsdahl, M. M., Yang, L. Y., Ahearn, M. J., and Trujillo, J.M. Establishment of a human carcinoembryonic antigen-producing colon

adenocarcinoma cell line. Cancer Res., 36: 467â€”475,1976.10. Goto, K., Akematsu, T., Shemazu, H., and Sugiyama, T. Simple differential

Glemea stainIng of sister chromatlds after treatment with photosensitivedyes and exposure to light and the mechanism of staining. Chromosoma(Bait), 53: 233â€”250,1975.

11. Gould.V.,andChejfee,G.Neuroendocrinecarcinomasof thecolon.Am.J.Surg.Pathol.,2: 31-38. 1978.

I 2. Granberg, I., and Mark. J. The chromosomal aberration of double minutesin a human embryonic rhabdomyosarcoma. Acta Cytol., 15: 42-45, 1971.

13. Grlmellus, L. A slIver nitrate stain of a3cells in human pancreatic islets. ActaSoc.Med.Upsal.,73:243-270,1968.

14. HarrIs, H., and Hopkins, D. A. Handbook of Enzyme Electrophoresis inHuman Genetics. Amsterdam: Elsevier-North Holland Publishing Co., 1976.

15. Jones, G. W., SlmkovIc, D., Bledler, J. L., and Southam, C. M. Humananaplastlcthyroidcarcinomaintissueculture.Proc.Soc.Exp.Biol.Med.,126: 426â€”428,1967.

16. Kovace, G. Abnormalities of chromosome no. 1 In human solid malignanttumours.Int.J. Cancer,21: 688â€”694.1978.

17. Kovacs, G. Homogeneously staining regions on marker chromosomes inmalignancy. Int. J. Cancer. 23: 299-301 , 1979.

18. Latt, S. LocalIzation of sister chromatid exchanges In human chromosomes.Proc. Neff. Aced. 5cl. U. 5. A., 71: 3162-3i66, 1974.

19. Lavappa, K. S. Trypsin-Glemsa banding procedure for chromosome preparationsfromculturedmammaliancells.TissueCultureAssociationManual,4: 761â€”764, 1978.

20. LeIbOVItZ,A., Stinson, J., McCombs, W., McCoy, C., Mazur, L., and Mabry,N. ClassIfIcatIon of human colorectal adenocarcinoma cell lines. CancerRae.,36:4562-4569,1976.

21. Levan, A., Levan, G., and Mandahl, N. A new chromosome type replacing

DECEMBER1979



t@.

@ .

t.ql@

2B.@

S

@#,.w @w'

a@

@ @Ij1Iwrt.

@ ;@s

Fig. 2. A, phase photomicrograph of COLO 320. x 200; B, Electron micrograph of COLO 321 . Note the absence of a brush border and presence of a fewosmiophilic granules. x 6600; C, Giemsa-stained metaphase of COLO 320 with DM of variable sizes; D, trypsin-treated metaphase of COLO 320. DM stain lightly;E, C-banded metaphase of COLO 320. Note absence of C-bands in DM; F, Hoechst-bromodeoxyuridine differential staining of COLO 320 that indicates DM andchromosomes have replicated.

-@

2C

.ft@...,-$

0

.

2E 2F

4920

@@.1@

C

I .â€˜@2@

CS




COLO 320 - DMs

ÃŽÃŒ(C ÃŒ4i?J ?t Ã®> il23456 7 8 9 10 11 12 X 13 15

t. Â«a Â»; r j . . Â« * Â»Â« â€¢Â«.â€¢ *

16 17 18 19 20 21 22

*â€¢â€¢â€¢

Ml M2 M3 M4 M5 M6 M9 M10 M11 M12 M15M16 M17 M18M19 M20 DM

B

"2 M3 M7 M10 M11 M12 M13 M14 M16 DM

*â€¢ tÃ f â€¢*3 Ã•

n â€žM1 M2 M3 M10 M11 M12 M14 M16 M18 DM

? if tifitifi. .M5 M7

i I ii I

M1 M2 M3 M4 M5 M7 M8 M9 M10 M11 M12 M13 M14 M15M16 DM

M1 M2 M3 M4 M5 M7 M9 M10 M11 M12 M14

3

Fig. 3. A, complete karyotype of COLO 320 11.0 months postinitiation; B to E, additional examples of COLO 320 marker chromosomes from 4 differentmetaphases. All metaphases had numerous DM.

DECEMBER 1979 4921



L. A. Quinn et al.

COLO 320 -HSR

@@ ! 161@ â€˜8@ 10@ 12

@ S@NNSal@ a. 01 â€¢. e4@ 14 16 17 18 19 120i 21 22

1 N2 M3 M5 M6 M7 M8 M9 M11 M12 M13 M14 M15 M@M19 M20 X

B@ @. $ I i I I@ I@ C@@ a@ I

M1 M2 M3 M4 M5 M6 M7 M8 M9 M12 M13 M14 M15 M16 M17 M20 x

C@@ ;@ I@ .1'

M1 M2 M3 M5 M7 M8 M9 M10 M12 M13 M14 M15 M19 M21 X

D@I t g@@@ I

M1 M2 M3 M5 M7 M8 Mg M12 M13 M14 M15

4

Fig. 4. A, complete karyotype of COLO 320 16 months postinitiation. Note loss of DM and gain of a marker with a HSR (M20); B to D, additional examples ofmarkers from 3 different metaphases of COLO 320. Note variations in the markers with H5R's (M20 and M2i).

4922 CANCERRESEARCHVOL. 39




COLO 321-HSR

iÃi* iu Ãiisi ;* i * aÂ«* .-., **23456 7 8 9 10 11 12 13 16 17 18 19 21 22

M i ! M 5 i t k Â»Â« -M2 M3 M4 MS Mg M7 M8 M9 M10 M^ M12 M14 M15

Bf { ) I f f Ã® j , ,M! M2 M3 M4 M5 M6 M7 M9 M,, M12 M15 DM

lili!) Ili * lÃ¬M1 M2 M3 M4 M5 /I6 M8 Mg M10 <*1 M15

\ l } 5 : (lÂ¡i Ã¼iMI M2 M3 M4 M5 M6 M8 M9 Â«10 M15 DM

I I ; * i I Â§ * iM2 M3 M4 M5 M6 M8 M10 M13 M14 M15 DM

5

Fig. 5. A, complete karyotype of COLO 321 16 months postinitiation; ÃŸto E, additional examples of markers from 4 different metaphases. Note HSR markers(M15).

DECEMBER 1979 4923



L. A. Quinn et al.

v <* Â¡â€¢â€¢

6A

-Ã¯Ã¯

Fig. 6. A, metaphase from COLO 320 with HSR marker; ÃŸ.metaphase from COLO 321 with HSR markers. Arrow, markers with a HSR.




1979;39:4914-4924. Cancer Res Laurie A. Quinn, George E. Moore, Robert T. Morgan, et al. RegionsProducts, Double Minutes, and Homogeneously Staining Cell Lines from Human Colon Carcinoma with Unusual Cell

Updated version

http://cancerres.aacrjournals.org/content/39/12/4914

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

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

.http://cancerres.aacrjournals.org/content/39/12/4914To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



cell lines from human colon carcinoma with unusual cell ... · colo 320 was examined at 1.5, 3.0,...

Documents