(CANCER RESEARCH 49. 6098-6107, November 1, 1989]

Characterization of Human Laryngeal Primary and Metastatic Squamous CellCarcinoma Cell Lines UM-SCC-17A and UM-SCC-17B1

Thomas E. Carey,2 Daniel L. Van Dyke, Maria J. Worsham, Carol R. Bradford,3 V. Ramesh Babu,

Donald R. Schwartz, Sandra Hsu, and Shan R. BakerThe Cancer Research Laboratory, Departments ofOtolaryngology/Head and Neck Surgery [T. E. C., C. R. B., D. R. S., S. H., S. R. B.J and Microbiology/Immunology[T. E. C.], The University of Michigan, Ann Arbor, Michigan 48109-0506; and The Cytogenetics Laboratory [D. L. V. £>.,M. J. W., V. R. B.J, Henry Ford Hospital,Detroit, Michigan 48202

ABSTRACT

The squamous cell carcinoma (SCC) cell lines UM-SCC-17A and-17B were derived, respectively, from the primary laryngeal cancer anda mctastatic neck tumor of a patient who failed to respond to radiationtherapy but achieved long-term remission after surgery. The karyotypesof both cell lines and a subline of 17A were pseudodiploid and stable inmultiple in vitro passages. Several karyotypic abnormalities were common to all three cell lines and therefore represent mutations present inthe tumor before the divergence of the metastatic and subline populationswhereas those rearrangements observed only in one cell are more likelyto be secondary. The shared mutations include: duplication of the shortand proximal long arm of chromosome 2, isochromosome 3q, duplication7, inversion 8, duplication of the distal long arm of 18, and monosomy21 or ring 21. Each line had different rearrangements involving chromosome 7 that resulted in three copies of most of the short arm beingpresent in both cell lines, except for high passages of 17B, in which onestructurally normal 7 was replaced by a dicentric isochromosome,dic(7Xqll.22), resulting in four copies of 7p. The dic(7) may representan in vitro mutation. An isochromosome 13q was noted in both thestemline and subline of UM-SCC-17A but not in UM-SCC-17B. Adel(llp) and an iso(21q) were present only in the 17A subline. The celllines expressed the membrane antigen phenotype characteristic of squamous cancers although the UM-SCC-17A subline differed with respectto three markers. Of these, the A9 and blood group antigen changes arethought to be associated with progression. The subline, which carried thedel(llXpl3-pl5.1), also failed to express the E7 antigen mapped to theband Ilpl3. It is possible that the two apparently normal 11s in thissubline carry a point mutation or microscopically undetected deletioninvolving the E7 antigen locus.

INTRODUCTION

Consistent chromosome rearrangements have been identifiedin many animal (1) and human cancers (2). Examples includet(9;22) in chronic myelogenous leukemia (3) and t(8;14), t(2;8),and t(8;22) in Burkitt's lymphoma (4). Similarly in solid tumors

of childhood chromosome 13 (5, 6) and 11 (7, 8) deletions areassociated with retinoblastoma and Wilms' tumor, respectively.

The breakpoints for specific chromosome rearrangements areoften located at or near loci for protooncogenes (9, 10) whichis consistent with the hypothesis that such lesions may alter theregulation of the oncogene products and promote unrestrainedcell growth.

Cytogenetic analysis of solid tumors has progressed slowlybecause the changes are often complex and difficult to analyze,and because it has been difficult to obtain good metaphasepreparations. Recent advances in culture techniques have made

Received 2/3/89; revised 6/7/89, 8/1/89; accepted 8/7/89.The costs of publication of this article were defrayed in part by the payment

of page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1Supported by USPHS Grants CA28564 and CA35929 from the National

Cancer Institute.1Recipient of USPHS Research Career Development Award CA 00621. To

whom requests for reprints should be addressed, at the Cancer Research Laboratory. The University of Michigan, 6020 KHRI, Box 0506, 1301 East AnnStreet, Ann Arbor, Ml 48109-0506.

3Supported by USPHS Training Grant T32-NS07293.

study of solid tumors more feasible. Consistent deletions of 3pin small cell carcinoma (11) and renal cell carcinoma (12), lossof alíeleson chromosome 5p in colon carcinoma (13), andnonrandom rearrangements involving chromosome 1 in neuroblastoma have now been described (14). To begin to addressthe chromosome rearrangements important in SCC,4 we chose

to study a pair of near diploid squamous carcinoma cell linesderived from a primary and metastatic tumor of the larynx ina patient who has remained free of disease for more than 6years after surgery. Unique contributions in this study includethe identification of the original rearrangements in a humantumor by analysis of primary and secondary tumor cell lines;characterization of steps in tumor progression by analysis of asubline and a secondary rearrangement of an inv(8); and demonstration of an immunologic-cytogenetic relationship betweena chromosome 11 rearrangement and expression of the E7 cellsurface antigen in a laryngeal cancer cell line.

MATERIALS AND METHODS

Clinical History of the Cell Line Donor. The patient was a 48-year-old female who had smoked one and one-half packs of cigarettes/dayfor approximately 30 years. In June of 1981 she presented with a sorethroat, hoarseness, and hemoptysis. Clinical examination and biopsyof a small lesion on the left vocal cord revealed T, squamous cellcarcinoma with no evidence of metastatic disease. The patient receivedfull course radiation therapy to the larynx from June through September 1981. The tumor persisted and spread to the soft tissue of the neckin spite of the treatments and in October 1981 a total laryngectomyand radical neck dissection were performed. She has remained in goodhealth to the present time. Her blood type is group O. Also of interestis the observation that this patient had autologous antibody in herserum that binds specifically to her cultured tumor cells but not to hernormal fibroblasts (15, 16). Her family history included a paternal greataunt who died of an unspecified type of cancer, a paternal aunt whodied of breast cancer, and another paternal aunt who died of bladdercancer.

Cell Lines. Tissues were washed, minced, and cultured as describedpreviously (15). Cell lines were established from the primary larynxcancer specimen (UM-SCC-17A) and the neck metastasis (UM-SCC-17B) (15). A subline of the UM-SCC-17A line detected by cytogeneticanalysis is described below. Normal fibroblasts (UM-SCC-17-FB) werecultured from a biopsy of normal forearm skin. A lymphoblastoid cellline, UM-Lyebv-17 was established from peripheral blood lymphocytesby transformation with Epstein-Barr virus (17, 18). Cells were frozenunder liquid nitrogen at multiple early passages in complete minimalessential medium containing 10% v/v dimethyl sulfoxide.

Chromosome Analysis. UM-SCC-17A and -17B cells in log phasewere harvested with 0.1% trypsin and 0.2% EDTA and treated withhypotonie 0.075 M KCl and 3:1 absolute methanohglacial acetic acid.Slides were prepared and G-banded using trypsin (Difco, 1:250) andGiemsa (Harleco). Selected slides were Q-banded, C-banded, or stainedby the Ag-NOR technique. Cells for RBG banding were treated inculture for 6 h with 30 Mg/ml bromodeoxyuridine, slides were prepared,stained in 5% 33258 Hoechst stain, mounted in Mcllvaine's buffer,

4 The abbreviation used is: SCC. squamous cell carcinoma.

6098

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-17B CELL LINES

exposed to UV light for l h and counterstained with Giemsa. Normallymphocytes and lymphoblastoid cells were harvested and G-banded bystandard techniques.

The most consistent karyotype comprises the consensus karyotypeof each cell line. Changes present only in individual cells, representingrandom loss, gain, or rearrangement can be ignored and minor sidelinescan be identified. The number of chromosomes in the consensus karyotype may differ from the modal number of chromosomes because ofrandom chromosome losses and gains.

Serological Reagents and Assays. Monoclonal antibodies UM-E7,UM-G10, and UM-A9 were developed in our laboratory (15). The UM-E7 antibody defines a cell surface molecule controlled by a locusmapped to chromosome band 11pi 3 (19, 20). Loss of expression of E7is hypothesized to define a deletion at this locus.5 The UM-G10

antibody defines the H type 2 blood group precursor, corresponding toblood group O (21). The UM-A9 antibody defines an epithelial germinal

cell/basement membrane antigen that is expressed to a differing extentby squamous carcinomas in a manner that corresponds to clinicalbehavior (22, 23). The EGFR1 monoclonal antibody (Amersham Corp,Cedarbrook, IL) defines the extracellular domain of the epidermalgrowth factor receptor (24). The lot used in these experiments showsno binding to normal fibroblasts at a dilution of 1/100. For comparativepurposes, the 50% positive endpoint with the MDA-468 EGFR-ampli-fied breast cancer cell line (25) is a dilution of 1/12,800 of this samereagent (not shown). Monoclonal antibody UM-ALL was raised to theUM-SCC-17A line and binds to a cell surface antigen expressed by asubset of SCC lines including UM-SCC-17A and -17B, but not bynormal fibroblasts or a variety of nonsquamous cancer lines. Monoclonal antibodies UM-G10, UM-E7, UM-ALL, and EGFR1 were testedusing the anti-immunoglobulin hemadsorption assay (21). Rabbit anti-serum to human ft-microglobulin (Accurate Chemical and ScientificCorp., Westbury, NY) was used to measure expression of class 1histocompatibility antigens. Human sera containing autoimmune antibodies to squamous cell antigens (15) were obtained from patients withactive pemphigus vulgaris and huilons pemphigoid and were providedby Dr. L. Diaz of the Johns Hopkins University. Rabbit and humansera and monoclonal antibody UM-A9 were tested using the protein Ahemadsorption assay (22). Each data point represents the mean value(±SEM)calculated from multiple assays with each serological reagenton several passages of each cell line.

RESULTS

Cell Lines. The UM-SCC-17 cell lines have an unusual morphology for squamous carcinoma as they grow as tightly packedcolonies (Fig. 1).

Cell Surface Antigen Phenotype. UM-SCC-17A and 17B aswell as the UM-SCC-17A subline all express the cell surfaceantigen phenotype typical of cultured squamous epithelial cells(15, 26, 27) (Fig. 2). This phenotype includes class 1 HLAantigens, pemphigus antigen, pemphigoid antigen, the H type2 blood group antigen defined by UM-G10, the epithelial

germinal cell/basement membrane/squamous carcinoma antigen defined by UM-A9, as well as those defined by UM-ALLand -E7. The UM-SCC-17 tumor cell lines also express theEGF receptor as determined by reactivity with the EGFR1antibody. However, in contrast to many other SCC cell lines,these tumor cells express only minimal levels of this marker.

Of the antigens tested, only /32-microglobulin and E7, wereexpressed by fibroblasts. The fibroblasts had significantly higher02-microglobuIin expression than the UM-SCC-17 lines. Malignant keratinocytes often have high class 1 antigen expressionbut, with respect to this marker, the UM-SCC-17 lines aresimilar to normal keratinocytes (not shown) which express lessclass 1 antigen than fibroblasts . The UM-SCC-17-FB fibroblasts expressed E7 antigen more strongly than any of the

B

5C. R. Bradford et al., unpublished results.

Fig. 1. Inverted phase contrast photomicrographs showing the appearance ofthe UM-SCC-17A stemline. the 17A subline and the 17B line in in vitro cultures(x 100). A, UM-SCC-17A stemline passage 24, Day 4 after passage; B, UM-SCC-17A subline passage 100, Day 4 after passage; C, UM-SCC-17B passage 21. Day9 after passage.

autologous tumor lines. The expression by the fibroblasts wasat a level similar to that of normal keratinocytes and fibroblastsfrom other donors. This will be discussed further below.

As shown in Fig. 2, when the membrane antigen expressionof the cell lines was compared, the 17A subline was distinguished from the 17A stemline and 17B by the level of expression of several markers. The 17A subline expressed significantlylower levels of the E7 antigen, the H type 2 blood group antigenand of the antigen defined by UM-ALL than either of the othertumor cell lines. In contrast, the expression of the A9 antigenwas significantly higher on the 17A subline than on either the17A stemline or 17B. This is of interest because A9 expressionis generally stronger on cell lines from recurrent and metastatictumors than on those from primary tumors (22) and becausestrong expression of A9 antigen as well as loss of blood groupantigen expression in tumor biopsies of head and neck canceris significantly associated with early recurrence (23). Thus, therelatively strong expression of A9 and weak expression of theblood group antigen (G10), relative to the other tumor cells

6099

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-I7B CELL LINES

o1C

sILI

coo

200 800 3200

RECIPROCAL DILUTION OF ANTIBODY

Fig. 2. Membrane antigen expression by UM-SCC-17A, the UM-SCC-17Asubline, UM-SCC-17B. and UM-SCC-17-FB normal skin fibroblasts. Resultsshown are the mean values plus or minus the standard error of the mean. A,rabbit antiserum to human ^2-microglobulin; />',serum from a patient with active

pemphigus vulgaris: C, serum from a patient with active bullous pemphigoid; D,monoclonal antibody UM-ALL raised to the UM-SCC-17A line; E, monoclonalantibody UM-G10 (defines the H type 2 blood group antigen corresponding tothe group O blood type of the UM-SCC-17 donor); F, monoclonal antibody UM-A9; G, monoclonal antibody EGFR1; H, monoclonal antibody UM-E7. •¿�,UM-SCC-17A; O, UM-SCC-17A subline; •¿�,UM-SCC-17B, D, UM-SCC17-FB.

from this patient, are suggestive that the 17A subline mayrepresent a population that was in transition to a biologicallymore aggressive state than either the UM-SCC-17A stemlineor the UM-SCC-17B metastatic cell line.

The UM-SCC-17A subline did not express detectable levelsof E7. This antigen has been mapped to a chromosome locusat band 11pi3 (19, 20) and as described below, the sublinecontains a deletion of chromosome 11 [del(ll)(pl3-pl5.5)] thatcorresponds to the E7 locus. This is consistent with other SCClines where reduced expression of E7 is associated with rearrangements affecting the Ilpl3-15 region of at least onehomologue.5 E7 antigen expression by the UM-SCC-17A stem-

line and the 17B line was also lower than that on the normalfibroblasts but no deletion or rearrangement of either chromosome 11 was visible. However, a submicroscopic deletion affecting the E7 locus on one of the 11s in these lines cannot beruled out.

Cytogenetic Results. The karyotype of the donor's phytohe-

magglutinin-stimulated peripheral blood lymphocytes and ofthe lymphoblastoid cell line was 46,XX.

Tumor cells were karyotyped in multiple passages to assessthe heterogeneity of the tumor cell populations and the stabilityof the chromosome rearrangements during long term in vitrogrowth. In cultures established from the primary tumor sitedesignated UM-SCC-17A we identified two populations withmodest karyotypic differences. One, referred to as UM-SCC-17A stemline, was present in multiple stored samples andrepresented by passages 8, 23, 28, and 37. The other referredto as UM-SCC-17A subline was discovered in cells carried inculture and used in experiments from 1983 to 1985 before beingkaryotyped and frozen at passage 65. Although both populations were derived from the same specimen, the two populationsdiffer in their in vitro lineage in that passage 65 was not a laterpassage of the stemline represented by passages 8, 23, 28, and37. Passages 8, 23, 28, and 37 were taken from separate frozenearly stocks.

We analyzed a total of 42 G-banded metaphases from passages 8, 23, 28, and 37 of UM-SCC-17 A, 16 G-banded metaphases from passage 65 of UM-SCC-17A, and 25 G-bandedmetaphases from passages 13, 17, and 52 of UM-SCC-17B. Allof the cultures were near diploid with most cells having 47 to49 chromosomes. Most of the variation in chromosome countswithin each cell lineage appeared to result from random lossesof individual chromosomes. The distribution of chromosomecounts in each passage is shown in Table 1.

UM-SCC-17A Stemline. The consensus karyotype of theUM-SCC-17A stemline was the same for passages 8, 23, 28,and 37 (Table 2). A cell from UM-SCC-17A passage 23 isshown in Fig. 3. The consensus karyotype was interpreted as49,dup(X)(q21.1q22.1), t(X;7)(p22.3;pll.2), +der(2)t(2;18)(q22;q22), +i(3)(qter-cen-qter), t(7;10)(pll;pll), +7, -9,+der(9)t(9;15)(q34;q22), -13, +i(13)(qter-cen-qter), r(?21).The dup(X) had an extra G-dark band below band Xq21,probably representing a duplication of band q21.3, but possiblyrepresenting an insertion of non-X material. This X was late-replicating (genetically inactivated) and was missing from somecells, including the cell shown in Fig. 3 (examples of this X areshown in Fig. 4). It was found in 15 of 31 cells in passages 8,23, and 28 but was not seen in passage 37. The t(X;7) and thet(7;10) appear to represent balanced translocations.

The der(2) consists of a duplication of 2pter-q22 with anabnormal band distal to q22 that may be from the long arm of18. Thus this chromosome has been tentatively identified ast(2;18)(q22;q22). Three copies of chromosome 13 were present(Fig. 3). Since Q-banding did not differ for the two chromosome13 copies in normal cells, it was not possible to determine bybanding whether the i(13q) represents a true isochromosomeor a Robertsonian translocation between homologues.

The ring chromosome was C-band positive. Since the onlycentromere region lost from the karyotype was that of chromosome 21 we tentatively identified the ring as a ring 21. Thering chromosome was missing from 13 of 42 cells (includingthe cell shown in Fig. 3), one copy was present in 19 cells, and10 cells had two to four copies of the ring (Fig. 4). No dicentricchromosomes and no abnormal Ag-NOR segments were observed.

UM-SCC-17A Subline. The karyotype of UM-SCC-17A sub-line, passage 65 was interpreted as 49,dup(X)(q21.1q22.1),t(X;7)(p22.3;pll.2), +der(2)t(2;18)(q22;q22), +i(3)(qter-cen-qter), t(7;10)(pll;pll), +7, inv(8)(pl2q24.1), -9, +der(9)t-(9;15)(q34;q22), +del(ll)(pl3-pl5.5), -13, +i(13)(qter-cen-qter), -18, -21, -21, +i(21)(qter-cen-qter), +r(?21). This sub-

line differs from the 17A stemline in that it has an inv(8)replacing one normal 8, a deleted lip in addition to two

6100

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-17B CELL LINES

Table 1 Chromosome count distribution

Number of cells containing indicated number of chromosomes

Chromosomecount 37 40 41 42 43 44 45 46 47 48 49 50 51 52 88

UM-SCC-17Astemlinep8° 3241

p23 13151p28 1 2511p37 116111

UM-SCC-17A subline

p65 1121551

UM-SCC-17B

pl3 112231 1pl7 11 1 131p52 1 5°Passage number tested.

Table 2 Comparison of the consensus karyotypes of UM-SCC-17A, I7A subline, and 17B, showing number of each normal and rearranged chromosome per cell

Chromosome12der(2)t(2;18)(q22;q22)3iso

(3qter-cen-qter)4der(4)*5inv(5)(pl3q35)6der(6)c7der(7)t(X;7)(p22.3;pll.2)der(7)t(7;10)(pll;pll)dup(7)(q21q22)dic(7)(pter-ql

1.22::ql1.22-pter)8inv(8)(pl2q24.1)der(8)c9der(9)t(9;15)(q34;q22)der(9)*10der(10)t(7;10)(pll;pll)inv(10)(pllq21)11del(ll)(pl3pl5.5)1213iso(l

3)(qter-cen-qter)1415dup(

15)(qter-p 12::q 12-q24::p 12-pter)161718inv(18)(qll.2q23)192021ring(?21)iso(2

1)(qter-cen-qter)22Xder(X)t(X;7)(p22.3;pll.2)dup(X)(q21.1q22.1)UM-SCC-1

7A(stemline)

(p8, p23, p28,p37)47°221212020201110020011011020211220222022110201o-rUM-SCC-17A(subline)

(p65)48-49"22121202020111001*1«011011021«211220221«0220*11*2011*UM-SCC-17D(pl3,

pl7,p52)47"22121I"I*1*1*1»1*1

or0"*0*0*1*,*1*0«i*i0«1*10«I»20*22*0*21*1«22I»1«221*i0*21»0'0*

" Modal number.* UM-SCC-17B had a complex translocation-4,-7,+der(4),+der(9)t(4;7)(4;9)(7;9)(4qter-4pl 5.2::7pl3-7pter;4pter-4pl5.2::9q32-9pter;?7qter-7pl3::9q32-9qter).

The der(7) was not present (see text).c UM-SCC-17A passage 65 had a pericentric inversion inv(8)(pl2q24.1). UM-SCC-17B had a balanced translocation between this chromosome and a chromosome

6 (see text). (Note added in proof: UM-SCC-17A subline has been tested in passage 96 and has the same karyotype as p65).''UM-SCC-HB had one normal 7 and a dup(7) in passages 13 and 17. In passage 52, all six cells analyzed had the dup(7). One passage 52 cell also had a normal

7, but in the other five cells the normal 7 was replaced by a dicentric isochromosome (7p). (Note added in proof: UM-SCC-17B cells from higher passages (p90,p96)derived from the passage 52 cells analyzed here, all have the dic(7p) and have karyotypes otherwise identical to the p52 cells.)

' UM-SCC-17B had monosomy X. UM-SCC-17A and 17A subline had no structurally normal X, but had a balanced t(X;7). 17A subline had a dup(X) in all cells

but the dup(X) was present in only 50% of the cells from passages 8, 23, and 28 and was absent from passage 37., difference from the UM-SCC-17A stemline consensus karyotype.

6101

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-17B CELL LINES

Fig. 3. Karyotype of an early metaphasecell from UM-SCC-17A, passage 23, representing the stemline. The karyotype is given inthe text. Structurally normal chromosomepairs include numbers 1,4, 5, 6, 8, 11, 12, 14-20, and 22. The der(2)t(2;18)(q22;q22) is tothe right of the normal 2s. The isochromosome3q(qter-cen-qter) is to the right of the normal3s. In this cell, chromosomes 4 and 5 havechromatid breaks but these were not consistentin all cells. For chromosome 7, the normal 7is at the left, the der(7)t(7;10)(pl l;pl 1) isin the middle, and at the right is theder(7)t(X;7)(p22.3;pll.2). The 9 at the righthas the segment ISq22-qter attached to itslong arm. The 10 at the right is the der-(10)t(7;10)(pl l;pl 1). One 13 is replaced by ani(13)(qter-cen-qter).Thereismonosomy21.TheX chromosome is a der(X)t(X;7)(p22.3;p 11.2).In this cell the ring chromosome was not present.

iff* li

UM-SCC-17A

Fig. 4. Partial karyotypes from UM-SCC-17A subline, passage 65. In most instances twopartial karyotypes are shown. Left, normalhomologues in a-i; a, normal 2s andder(2)t(2;18)(q22;q22); b, normal 3s and iso-chromosome 3(qter-cen-qter); c, normal andderived 7s with der(7)t(7;10)(pl l;pll) in themiddle and der(7)t(X;7)(p22.3;pll.2) at right;d, normal 8 and inv(8)(pl2q24.1); e. normal 9and der(9)t(9;15)(q34;q22);/, normal 10 andder(10)t(7;10)(pll;pll); g, normal 11s anddel(l I)(pl3pl5.5); A, normal 13 and isochro-mosome 13q; i, normal 18 from two cells;/ring of uncertain origin, which might representa ring 21 (see text); k, isochromosome 21 ort(21;21) from two cells; /, der(X)t(X;7)-(p22.3;p 11.2) and dup(X)(q21.2q22) from fourcells.

apparently normal 1Is, loss of one 18, and an isochromosome21 replacing the normal 21. The dup(X) that was found in onlyhalf of the stemline cells examined was present in all of thesubline cells examined. Partial karyotypes showing examplesof each rearranged chromosome along with the representativenormal copies are given in Fig. 4. To confirm the stability ofthe UM-SCC-17A subline, cells were carried for an additional30 passages and retested. The karyotype in passage 96 wasidentical to the cells tested in passage 65.

UM-SCC-17B. The karyotypes of UM-SCC-17B passages13, 17, and 52 (Figs. 5 and 6) was interpreted as 47, X, —¿�X,+der(2)t(2;18)(q22;q22), +i(3)(qter-cen-qter), -4, -7, -9, +der(4), + der(9)t(4;7)(4;9)(?7;9)(4qter-4pl5.2::7pl3-7pter; 4pter-4pl5.2::9q32-9pter;;qm7qter-7pl5::9q32-9qter),inv(5) (pl3q35), dup(7)(q21q22), t(6;inv(8))(8qter-8q24.1::6p21.16-qter;6pter-6p21.1 ::8p 12-8q24.1 ::8p21 -Spter), inv( 10)(p 11q21 ),dup(15)(qter-pl2::qll-q24::pl2-pter), inv(18)(ql 1.2q23),r(?21). Thus, UM-SCC-17B differed in several ways from UM-SCC-17A. UM-SCC-17B had one genetically active, structurally normal X and lacked both of the rearranged X configurations t(X;7) and dup(X) that were in 17A. In 17B three newinversions appear: inv(5)(pl3q35), inv(10)(pllq21), andinv(18)(ql 1.2q23), each replacing a structurally normal copy.With respect to chromosome 7, both the t(X;7) and the t(7;10)

I —¿�I -k-kfru-UM-SCC-I7A

found in 17A were not present in 17B. Instead chromosome 7was represented one normal 7 (as in 17A), a dup(7)(q21q22)and as part of the der(4) from a complex translocation involvingchromosomes 4, 7, and 9. The derived 4 is comprised of asegment of 7p attached near the end of 4p (Figs. 5 and 6). Thederived 9 consists of most chromosome 9 with the segment ofchromosome 4p attached near the end of the long arm ofchromosome 9 (Figs. 5 and 6). Presumably the missing longarm of 9 was translocated to the short arm of 7 to make up apostulated der(7). This was apparently lost soon after the complex translocation occurred, because the segment missing fromthe long arm of chromosome 9 (9q32-qter) and most of thechromosome 7 (7qter-7pl3) which donated the segment of 7pto the derived 4 were never observed. The inversion 8 detectedin the UM-SCC-17A subline was involved in a translocationwith chromosome 6 in 17B (Figs. 6e and 7). The identificationof the inv(8) in 17A subline helped us to identify the t(6;inv(8))in 17B and provided evidence that the inv(8) was present priorto the separation of the metastatic population from which the17B cell line was derived.

UM-SCC-17B had two apparently normal 13s whereas 17Ahad one 13 and an i(13q). Unlike UM-SCC-17A which had twostructurally normal copies of 15 and a der(9)t(9;15), the 17Bline had one normal chromosome 15 and a dup(15). We iden-

6102

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-17B CELL LINES

Fig. 5. Representative karyotype of a cellfrom UM-SCC-17B, passage 17. Structurallynormal chromosome pairs include numbers 1,11-14, 16, 17, 19, 20, and 22. The der(2) is tothe right of the normal 2s. The isochromosome3q is to the right of the normal 3s. The der(4)consisting of 4qter-4pl5.2::7pl3-7pter; inv-(5)(pl3q35); der(6) from the t(6;inv (8)); dup7(q21q22); der(8)t(6;inv(8)) see text; der-(9)consisting of 4pter-4p 15.2::9q32-9pter; inv-(10)(pllq21);andinv(18)(qll.2q23)aretotheright of their normal homologues. The dup( 15)-(qter-pl2::qll-q24::pl2-pter) is to the left ofthe normal 15. There is monosomy X and 21.The r(?21) is not present.

>(IIli If

«C{erf

UM-SCC-I7B*«>

Fig. 6. Partial karyotypes from cells fromUM-SCC-17B, passage 17. In most instancestwo partial karyotypes are shown. See legendto Fig. 5 for detailed description of the rearranged chromosomes, a, normal 2s and derived2; b, normal 3s and iso (3q); <•.normal 4 andderived 4; d, normal 5 and inverted 5; e. normal6 and 8, derived 6 and 8;_£normal 7 and dup7; g, dup 7 and isochromosome 7p from passage 52; h, normal 9 and der(9); i, normal 10and inv(10);y, normal 15 and dup(15); k, normal 18 and inverted 18; /, ring chromosomefrom two cells; m, normal X from two cells.

¡i ÕÕ

UM-SCC-I7B

Fig. 7. Ideogram depicting chromosome 8 and6 rearrangements in squamous cell carcinoma cellline UM-SCC-17. UM-SCC-17A stemline had apair of normal chromosome 6s and 8s. UM-SCC-17A subline had a pair of normal 6s, but onechromosome 8 had a pericentric inversion, shownhere with breakpoints at 8pl2 and pq241. UM-SCC-17B had one normal 8 and one normal 6 anda balanced translocation between the inversion 8and one chromosome 6, with a breakpoint in theshort arm of the 8 at or near the original inversionbreakpoint, and a breakpoint on chromosome 6 atband 6p21.1, as shown diagrammatically. Thus theinv(8) observed in 17A subline is most likely anintermediate step in the formation of the apparently complex translocation observed in 17B.

8 inv (8) 6 der (6) der (8)

tified the extra material attached to the short arm, as aninsertion of 15qll-15q24, because it was Ag-NOR negativeindicating that it had a deletion of 15p. The chromosome 15rearrangements differed in 17A and 17B; 17A contained a15q22 breakpoint whereas 17B had a 15q24 breakpoint.

Passage 52 of UM-SCC-17B had a similar karyotype to that

of the earlier passages except that the normal 7 was present inonly one of six metaphase cells, being replaced by a dicentric7p(ql 1.22) in the other five cells. This was a stable change sinceit was also present in all cells examined in passages 90 and 96of this same lineage which were otherwise identical to the cellsin lower passages.

6103

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-17B CELL LINES

Net Result of Chromosome Rearrangements. Comparison of is useful for comparison to other tumors. A summary of thethe consensus karyotypes of the two sublines of 17A and the17B line is presented in Table 2. The net result of the chromosome rearrangements in UM-SCC-17A was duplicationXq21.2-Xq22.1, duplication 2pter-2q22, triplication 3q, tri-somy 7, duplication 13ql 1-qter, and duplication 15q24-qter. Inthe siculi ine of UM-SCC-17A there was probable duplication18q22-qter and monosomy 21. In the 17A subline there wasduplication llpl 3-1 Iqter and monosomy 18 or deletion ISpter-q22, but chromosome 21 had been partially returned to balanceby isochromosome formation. The net result in UM-SCC-17Bwas duplication 2pter-2q22.1, triplication 3q, duplication 7pl3-7pter and 7q21-q22, deletion 9q32-9qter, monosomy 21, monosomy X, probable duplication 15ql l-q24 and probable duplication 18q22-qter. In passage 52 (as well as passages 90 and96) of UM-SCC-17B there was triplication 7p, and deletion7qll-7q21 and 7q22-7qter.

Postulated Sequence of Chromosomal Events in the Evolutionof UM-SCC-17A and UM-SCC-17B. On the basis of the chromosome changes we observed in the three tumor lines frompatient UM-SCC-17, a probable sequence of changes has beendeduced. This sequence is outlined in Fig. 8. Trisomy 7 andthree rearrangements (+der(2), +i(3q), and monosomy 21 ±ring ?21) were present in all three cell lines and thus have beenassigned as "early" changes. Early changes are defined here as

those occurring before the separation of the metastatic population. One other rearrangement, inv(8), was found in the UM-SCC-17A subline. When we recognized that the karyotype ofUM-SCC-17B included a translocation between chromosome6 and the inv(8), it appeared that the 17A subline might represent an intermediate step in the progression to the metastatic17B population. However since the 17A subline shared multipleother rearrangements with 17A but not with 17B this lineagecould not be reasonably substantiated. For the inv(8) to bepresent in both 17A subline and 17B, either (a) the samerearrangement arose independently in both lines, or (¿>)theinv(8) was an "early" rearrangement that was subsequently lost

from the 17A stemline with a compensating nondisjunctioninvolving the structurally normal homologues. We favor thelatter and plan to test this possibility by analysis of DNA fromthe 17A stemline for loss of restriction fragment polymorphisms for chromosome 8.

Breakpoints and Possible Regions of Homozygosity. Thepoints at which changes in DNA organization are most likelyto affect gene expression are the chromosome breakpoints. Ifnonrandom chromosome changes do occur in squamous carcinomas, as we expect they do, then a catalogue of the breakpoints

Early changéeat priaory lut»ir site,fu ii ir to rl(i(.il separation of 17A t n m 17B

*der(2|* i(3q)

17*ion at primary site

t(X;7)t(7;10)t(9;15)

/ \at

Progression to aetastatiotuav cell ]

inv(10)<M>(15>inv(18]

t(4;7)(4;9)(7;9)im(5|t(6;inv(8)>

/ \17»subìine(passage 65)

-iro(S) -7, +i(7p) Mosaic

Fig. 8. Postulated sequence of chromosome changes that occurred during thedevelopment and progression of the tumor and cell line UM-SCC-17.

breakpoints found in UM-SCC-17 is shown in Fig. 9. Theseare categorized as either early or secondary rearrangements inTable 3.

Loss of heterozygosity is present or suspect in several cases.We did not find a second copy of 21 in any of the UM-SCC-17cell lines [except perhaps for those cells containing the ring(?r21)]. Similarly, only one intact copy of 18 and a segment of18 (q22-qter) translocated to the der(2) was found in the 17Asubline. Thus the tumor cells are likely to be hemizygous formost alíeleson these chromosomes.

With the information we have now we cannot determine theorigin of the duplicated portions of chromosomes 2 and 3. Onepossibility is that in each case one homologue was first duplicated and then rearranged in which case there should be duplication of some alíelesbut no losses. Alternatively, the first eventmay have been rearrangement that was followed by nondisjunction of the normal homologue in which case these tumor cellsmay have lost heterozygosity for 2q22-2qter and 3pter-3cen-the regions that are not represented in the duplicated segments.Other suspected regions of loss of heterozygosity include chromosome 8 in the 17A stemline, 13q in 17B, and Ilpter-llpl3in the 17A subline (Fig. 9).

DISCUSSION

One of the great challenges for the cancer researcher is toidentify the genetic, cytogenetic, and cytological changes thatare important in the cause and progression of neoplasia. Because the karyotype abnormalities in solid cancers are frequently complex, one approach is to study the karyotype oftumors in very early stages of development. As with other solidtumors, however, the majority of squamous cell cancers areusually quite advanced by the time they are detected, therebyreducing the opportunities for detecting the initial cytogeneticabnormalities. An alternative is to examine cell lines establishedfrom primary and secondary (recurrent or metastatic) tumorsin a series of patients as we did in this report. By analysis ofcell populations cultured from the primary tumor (UM-SCC-17A and UM-SCC-17A subline) and the metastatic tumor(UM-SCC-17B) we were able to distinguish, with reasonableassurance, the "early" chromosome rearrangements from those

associated with tumor progression. Any cytogenetic changeswhich are common to both the 17A and 17B cell lines musthave been present prior to the time of emergence of the metastatic line (Fig. 8). By the same line of reasoning chromosomeabnormalities that are not shared by the primary and metastatictumor cell lines were either lost with tumor progression or aroseseparately subsequent to the development of the metastatic cellpopulation. Thus, from the many chromosome changes observed in UM-SCC-17 A, only five (dup(2pter-2q22), iso(3q),trisomy 7, inv8(pl2q24.1), and —¿�21)emerge as early in vivomutations in this patient's tumor. The rationale for assignment

as early mutations is that it is extremely unlikely that the samerearrangements observed in both 17A and 17B arose independently in different cell lineages either in vivo or in vitro.

Mutations that are important in tumor metastasis may beamong the chromosome abnormalities found in 17B only.Those found only in 17A may be unimportant in metastaticdisease, but may play a role in tumor progression and invasive-ness at the primary site. Five additional chromosome rearrangements, but probably no numerical changes, were associated with progression at the primary site, whereas monosomyX and seven chromosome rearrangements were associated with

6104

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-I7B CELL LINES

Fig. 9. Breakpoints in rearrangements andpostulated regions of homozygosity in UM-SCC-17. O, breakpoints in UM-SCC-17A; D,breakpoints in UM-SCC-17B; *, a postulatedprimary change; and vertical wavy lines, regions of possible homozygosity.

On

!*a

17 18

progression in the metastatic tumor cell population (Fig. 8).Unfortunately this patient received radiation treatments priorto surgery (and the development of the cell lines), thus radiationdamage may account for some of the differences between 17Aand 17B (e.g. the complex translocation involving chromosomes 4, 7, and 9, or the chromosome 7 duplication) making itdifficult to identify with certainty those changes most importantin tumor progression.

Relatively few squamous carcinomas or cell lines from SCCof the head and neck region have been fully characterizedcytogenetically. Thus it is not yet possible to identify consistentchromosome abnormalities in this cancer type. However,among those that have been studied (28-35) translocations,

rearrangements, breakpoints, or numerical differences affectingchromosome segments 4, 7p, 9q, lip, 13, 18, and X that aresimilar to those in UM-SCC-17 have been reported. Further

study of many more tumors will be necessary before it can bedetermined whether these are the sites of nonrandom changesin SCC.

Abnormalities of chromosome 11, especially loss of alíelesor copies have been reported in a number of cancer types (36).The del(l I)(pl3pl5) in the 17A subline may be representativeof a nonrandom chromosome abnormality in squamous cellcancer. A similar del(ll) was present in line 183 (33) and inthe squamous carcinoma of the prostate cell line (UM-SCP-1)

reported by Grossman et al. (27). Recently we found that fourvulvar SCC lines (UM-SCV-1, 2, 3, 5), five head and neck lines(UM-SCC-8, 11A, 22B, 69, HFH-SCC-XP1), and one SCC ofthe bladder also have 1Ip deletions or breakpoints.5 The locus

controlling the E7 antigen was previously mapped to bandIlpl3(19,20) using human hamster-hybrid cells that contained

either a normal or del 11. On this basis we have begun to6105

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATIONOF UM-SCC-I7A AND UM-SCC-17BCELL LINES

Table 3 Chromosome breakpoints in UM-SCC-17A and 17B cell lines organized as early and secondary events

Affectedchromosome23456789101113151821XChromosomerearrangementder(2)t(2;18)(q22;q22)iso(3q)(qter-cen-qter)der(4)inv(5)t(6;inv(8))t(X;7)(p22.3;pll.2)t(7;IO)(pll;pll)der(9)t(4;7)(4;9)(?7;9)dic(7)(qll.23)dup(7)pericentric

inv(8)t(6;inv(8»der(9)t(9;15)(q34;q22)der(9)t(4;7)(4;9)(?7;9)t(7;10)(pll;pll)inv(10)(pll;q21)del(ll)(pl3-pl5.5)iso(

13)(qter-cen-qter)der(9)t(9;15)(q34;q22)dup(

15)(qter-p 12::ql I-q22::pl2-pter)der(2)t(2;18)(q22;q22)inv(18)r?21iso2

1(qter-cen-qter)t(X;7)(p22.3;pll.2)dup(X)(q21.2q22.1)Cell

lineallall17B17B17B17A17A17B17B(P52)17B17A(sub)17B17A17B17A17B17A(sub)17A17A17Ball17Ball17A(sub)17A17ABreakpointsEarlySecondary2q223cen4pl5.25ql3,

5q356p21.17pll.27pll7pl3,

7pl57qll.237q21,7q228pl2,

8q24.18pl29q349q32lOplllOpll,

10q21Ilpl313cen15q2215pl2,15qll,15q2418q2218qll.2,

18q232Icen2

IcenXp22.3Xp21.2,

Xq22.1

evaluate the relationship between E7 antigen expression andlip deletions. Most SCC lines that are poor E7 antigen ex-pressors contain an 11p breakpoint or deletion whereas amongthe karyotyped SCC lines that express E7 well, few have liprearrangements.5 How the level of E7 expression is related to

the lip abnormalities is not known. Since the 17A stemlineand the 17B line expressed lower levels of E7 antigen than didautologous fibroblasts, we expected to find a deletion or breakpoint of lip. However, both lines have apparently normalchromosome 11 pairs. The 17A subline (which contains twoapparently normal 1Is and a deletion 1Ip) has nearly undetect-able E7 expression. We think that one of the 1Is in the UM-SCC-17 stemline and metastasis cultures may have a pointmutation or submicroscopic deletion of lip that reduces thegene dose for the E7 antigen in these cultures. In the 17Asubline, further reduction in E7 expression might be explainedby the creation of the del(ll)(pl3pl5.5) from the normal 11and duplication (by mitotic nondisjunction) of the 11 with thepostulated undetected 11pi3 deletion or point mutation. Oneway to test this hypothesis will be to search for deletions usingprobes for the Ilpl3-llpl5 region using in situ fluorescencehybridization technology which allows for the detection ofspecific DNA sequences not only on metaphase chromosomes,but on interphase nuclei as well (37, 38). Alternatively it maybe possible to segregate the three chromosomes in somatic cellhybrids and use restriction fragment length polymorphismanalysis to search for deletions using complementary DNAprobes for the Ilpl3 to 11pi 5 region. Another approach willbe to isolate the E7 gene segment from normal DNA andexamine the tumors with this karyotypic feature for abnormalities or loss of the gene. These studies are now underway.

Critical to karyotype analysis in tumor cell lines is the question of how well the lines represent the original tumor; i.e.,their stability in vitro. One way to determine this is to comparekaryotypes from direct tumor preparations or primary cultureswith those of the established cell line, as we did for the UM-EC-1 endometrial cancer cell line (18). In UM-EC-1, we found

that the karyotypes of the primary culture and later passageswere the same. With UM-SCC-17 we were not in a position toexamine the karyotype of the direct tumor preparations or

primary cultures. Therefore, we examined multiple passages.Identical karyotypes were observed in four different passages ofthe UM-SCC-17 A stemline, in passages 65 and 96 of the sublineand in three different passages from UM-SCC-17B demonstrating that these lines are stable in vitro. One plausible in vitromutation was the dicentric isochromosome 7p which was firstobserved in passage 52 of UM-SCC-17B (Fig. 6g). Alternativelythe dic(7p) could reflect a minor subline of 17B that was presentbut not recognized in the earlier passages. We cannot excludeeither possibility but as the majority of the cells examined inthis and later passages contained the dic(7p) and as the dic(7p)was never observed in earlier passages, it seems likely to be anin vitro development.

Heterogeneity within the tumor cell lines that we have characterized so far has been minimal (18, 28). The differences weobserved between the 17A sister cell lines illustrate that therewere at least two viable cytogenetically heterogeneous cell populations within the sample of primary tumor tissue we cultured.The 17A subline may represent a population within the primarycancer that was undergoing tumor progression since it hashigher expression of the A9 antigen and reduced blood groupantigen expression which are correlated with early recurrencein vivo (23). Now that the two sister lines have been distinguished by karyotype and antigen phenotype, we are beginningto examine them for differences in growth rate in vitro andtumor production in nude mice to test this hypothesis. This isof special interest because among the cancer cell lines we haveestablished, the UM-SCC-17A and -17B lines are in a categorywe call "low malignant" cell lines. Cell lines in this category

are poorly tumorigenic in nude mice, fail to grow in serum freemedium, and come from donors who exhibit long term survivalafter treatment. Thus identification within a "low malignantcell line" of a subpopulation of cells with characteristics usually

associated with biologically aggressive behavior in vitro and invivo provides an excellent model for testing how well thesecharacteristics do predict behavior.

REFERENCES

1. Klein, G. The role of gene dosage and genetic transpositions in carcinogen-esis. Nature (Lond.), 294: 313-318, 1981.

6106

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

CHARACTERIZATION OF UM-SCC-17A AND UM-SCC-17B CELL LINES

2. Rowley, J. D. Biological implications of consistent chromosome rearrangements in leukemia and lymphoma. Cancer Res., 44:3159-3168, 1984.

3. Nowell, P. C., and Hungerford, D. A. A minute chromosome in humangranulocytic leukemia. Science (Wash. DC), 132: 1497, 1960.

4. Manolov, G., and Manolova, Y. Marker band in one chromosome 14 fromBurkitt lymphomas. Nature (Lond.), 237: 33-34, 1972.

5. Knudson, A. G.. Jr., Meadows, A. T., Nichols, W. W.. and Hill, R. Chromosomal deletion and retinoblastoma. New. Eng. J. Med., 295: 1120-1123,1976.

6. Cavenee, W. K., Dryja, T. P., Phillips, R. A., Benedict, W. F., Godbout, R.,Gallic, B. L., Murphree, A. L., Strong, L. C., and White, R. L. Expressionof recessive alíelesby chromosomal mechanisms in retinoblastoma. Nature(Lond.), 305: 779-784, 1983.

7. Riccardi, V. M., Sujansky, E., Smith, A. C., and Francke, U. Chromosomalimbalance in the Aniridia-Wilms' tumor association: 1Ip interstitial deletion.Pediatrics, 61: 604-610, 1978.

8. Kondo, K., Chilcote, R. R., Maurer, H. S., and Rowley, J. D. Chromosomeabnormalities in tumor cells from patients with sporadic Wilms' tumor.Cancer Res., 44: 5376-5381, 1984.

9. Rowley, J. D. Human oncogene locations and chromosome aberrations.Nature (Lond.), 301: 290-291, 1983.

10. Heim, S., and Mitelman, F. Nineteen of 26 cellular oncogenes preciselylocalized in the human genome map to one of the 83 bands involved inprimary cancer-specific rearrangements. Hum. Genet., 75: 70-72, 1987.

11. Whang-Peng, J., Bunn, P. A., Jr., Kao-Shan, C. S., Lee, E. C, Carney, D.N., Gazdar, A., and Minna, J. D. A nonrandom chromosomal abnormality,del3p(14-23) in human small cell lung cancer (SCLC). Cancer Genet. Cyto-genet., 6: 119-134, 1982.

12. Teyssier, J. R., Henry, I., Dozier, C., Ferre, D., Adent, J. J., and Pluot, M.Recurrent deletion of the short arm of chromosome 3 in human renal cellcarcinoma: shift of the c-ra/1 locus. J. Nati. Cancer Inst., 77: 1187-1195,1986.

13. Solomon, E., Voss, R., Hall, V., Bodmer, W. F., Jass, J. R., Jeffreys, A. J.,Lucibello, F. C., Patel, I., and Rider, S. H. Chromosome 5 alíeleloss inhuman colorectal carcinomas. Nature (Lond.), 328: 616-619, 1987.

14. Brodeur, G. M., Sekhon, G. S., and Goldstein, M. N. Chromosomal aberrations in human neuroblastomas. Cancer (Phila.), 40: 2256-2263, 1977.

15. Carey, T. E., Kimmel, K. A., Schwartz, D. R., Richter, D. E., Baker, S. R.,and Krause, C. J. Antibodies to human squamous cell carcinoma. Otolaryn-gol. Head Neck Surg., 91: 482-491, 1983.

16. Vlock, D. R., Scalise, D., Schwartz, D. R., Richter, D. E., Krause, C. J.,Baker, S. R., and Carey, T. E. Incidence of serum antibody reactivity toautologous head and neck cancer cell lines and augmentation of antibodyreactivity following acid dissociation and ultrafiltration. Cancer Res., 49:1361-1365,1989.

17. Miller, G., and Lipman, M. Release of infectious Epstein-Barr virus bytransformed marmoset leukocytes. Proc. Nati. Acad. Sci. USA, 70:190-194,1973.

18. Grenman, S. E., Van Dyke, D. L., Worsham, M. J., del Rosario, F., Roberts,J. A., McClatchey, K. D., Schwartz, D. R., Babu, V. R., and Carey, T. E.UM-EC-1, a new hypodiploid human cancer cell line derived from a poorlydifferentiated endometrial cancer. Cancer Res., 48: 1864-1873, 1988.

19. Jones, C., Kimmel, K. A., Carey, T. E., Miller, Y. E., Lehman, D. W., andMacKenzie, D. Further studies on a hybrid cell-surface antigen associatedwith human chromosome 11 using a monoclonal antibody. Somat. CellGenet., 9: 489-496, 1983.

20. Fisher, J. H., Miller, Y. E., Sparkes, R. S., Bateman, J. B., Kimmel, K. A.,Carey, T. E., Rodell, T., Shoemaker, S. A., and Scoggin, C. H. Wilms' tumor

aniridia association: segregation of the affected chromosome in somatic cellhybrids, identification of cell surface antigen associated with deleted area,and regional mapping of c-Ha-ras-1 oncogene, insulin gene, and beta globingene. Somat. Cell Mol. Genet., 10: 445-464, 1984.

21. Kimmel, K. A., Carey, T. E., Judd, W. J., and McClatchey, K. D. Monoclonalantibody (G10) to a common antigen of human squamous cell carcinoma:binding of the antibody to the H type 2 blood group determinant. J. Nati.Cancer Inst., 76: 9-19, 1986.

22. Kimmel, K. A., and Carey, T. E. Altered expression in squamous carcinomacells of an orientation restricted epithelial antigen detected by monoclonalantibody A9. Cancer Res., 46: 3614-3623, 1986.

23. Carey, T. E., Wolf, G. T., Hsu, S., Poore, J., Peterson, K., and McClatchey,K. D. Expression of A9 antigen and loss of blood group antigens as determinants of survival in patients with head and neck squamous carcinoma.Otolaryngol. Head Neck Surg., 96: 221-230, 1987.

24. Waterfield, M. D., Mayes, E. L., Stroobant. P., Bennet, P. L., Young, S.,Goodfellow, P. N., Banting, G. S., and Ozanne, B. A monoclonal antibodyto the human epidermal growth factor receptor. J. Cell Biochem., 20: 149-161, 1982.

25. Filmus, J., Pollak, M. N., Cailleau, R., and Buick, R. N. MDA-468, a humanbreast cancer cell line with a high number of epidermal growth factor (EGF)receptors, has an amplified EGF receptor gene and is growth inhibited byEGF. Biochem. Biophys. Res. Commun., 128: 898-905, 1985.

26. Carey, T. E. Establishment of Epidermoid Carcinoma Cell Lines. In: R. E.Wittes (ed.), Head and Neck Cancer, pp. 287-314, London: Wiley & Sons,1985.

27. Grossman, H. B., Wedemeyer, G., Ren, L., and Carey, T. E. UM-SCP-1, anew human cell line derived from a prostatic squamous cell carcinoma.Cancer Res., 44:4111-4117, 1984.

28. Jin, Y. S., Mandahl, N., Heim, S., Biorklund, A., Wennerberg, J., andMitelman, F. Unique karyotypic abnormalities in a squamous cell carcinomaof the larynx. Cancer Genet. Cytogenet., 30: 177-179, 1988.

29. Jin, Y. S., Mandahl, N., Heim, S., Biorklund, A., Wennerberg, J., andMitelman, F. t(6;7)(q23;p22) as the sole chromosomal anomaly in a vocalcord carcinoma. Cancer Genet. Cytogenet., 32: 305-307, 1988.

30. Moore, G. E., Merrick, S. B., Woods, L. K., and Arabasz, N. M. A humansquamous cell carcinoma cell line. Cancer Res., 35: 2684-2688, 1975.

31. Morgan, R. T., Quinn, L. A., Woods, L. K., and Moore, G. E. Markerproperties of tumor and lymphoid cell lines derived from a patient withsquamous cell carcinoma. Cancer Res., 37: 2030-2035, 1977.

32. Boukamp, P., Tilgen, W., Dzarglieva, R. T., Breitkreutz, D., Haag, D., Riehl,R., Bohnet, A., and Fusemig, N. E. Phenotypic and genotypic characteristicsof a cell line from a squamous cell carcinoma of human skin. J. Nati. CancerInst., «A:415-427, 1982.

33. Sacks, P. G., Parnés,S. M., Gallick, G. E., Mansouri, Z., Lichtner, R.,Satya-Prakash, K. L., Pathak, S., and Parsons, D. F. Establishment andcharacterization of two new squamous cell carcinoma cell lines derived fromtumors of the head and neck. Cancer Res., 48: 2858-2866, 1988.

34. Easty, D. M., Easty, G. C., Carter, R. L., Monoghan, P., Pittman, M. R.,and James, T. Five human tumor cell lines derived from a primary squamouscarcinoma of the tongue, two subsequent local recurrences and two nodalmétastases.Br. J. Cancer, 44: 363-370, 1981.

35. Van Dyke, D. L., Carey, T. E., Grenman, S. E., Worsham, M. J., Bradford,C. R., Grenman, R., Babu, V. R., and Schwartz, D. R. Identification ofchromosome changes in squamous cell carcinoma using established cell lines.In: W. E. Fee, Jr., M. E. Johns, E. W. Strong, and P. H. Ward, (eds.), Headand Neck Cancer, Vol 2. Proc. Second Int. Conf. Toronto: B. C. Decker Inc.,in press, 1989.

36. Atkins, N. B., and Baker, M. C. Deficiency of all or part of chromosome 11in several types of cancer: significance of a reduction in the number of normalchromosomes 11. Cytogenet. Cell Genet., 47: 106-107, 1988.

37. Pinkel, D., Straume, T., and Gray, J. W. Cytogenetic analysis using quantitative, high-sensitivity, fluorescence hybridization. Proc. Nati. Acad. Sci.USA, 82: 2934-2938, 1986.

38. Lawrence, J. B., Villnave, C. A., and Singer, R. H. Sensitive, high-resolutionchromatin and chromosome mapping in situ: presence and orientation of twoclosely integrated copies of EBV in a lymphoma line. Cell, 52: 51-61, 1988.

6107

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

1989;49:6098-6107. Cancer Res   Thomas E. Carey, Daniel L. Van Dyke, Maria J. Worsham, et al.   UM-SCC-17BSquamous Cell Carcinoma Cell Lines UM-SCC-17A and Characterization of Human Laryngeal Primary and Metastatic

  Updated version

  http://cancerres.aacrjournals.org/content/49/21/6098

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/49/21/6098To request permission to re-use all or part of this article, use this link

on March 21, 2021. © 1989 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from