[CANCER RESEARCH 49. 6966-6971. December 15. 1989]

Frequent Alteration of the DF3 Tumor-associated Antigen Gene in Primary HumanBreast Carcinomas'

Giorgio R. Merlo, Javed Siddiqui, Craig S. Cropp, Daniel S. Liscia,2 Rosette Lidereau, Robert Callahan, andDonald W. Kufe3

Laboratory of Tumor Immunology and Biology, NCI, NIH, Bethesda, Maryland 20892 (G. R. M., C. S. C., D. S. L., R. C.J; Laboratory of Clinical Pharmacology,Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts 02115 [J. S., D. W. K.J; Centre Rene Huguenin. 5 rue Gaston Latouche,92211 St. Cloud, France ¡R.L.J

ABSTRACT

The gene for the human DF3 breast carcinoma-associated antigencontains a conserved (G+C)-rich 60-base pair tandem repeat and mapsto chromosome lq21-24. In the present study we isolated and characterized 1220 base pairs of nonrepetitive adjacent sequences. Multiple alíeleswere identified by fragment size. Signal intensity of hybrids with thetandem and unique sequence probes indicated that allelic variation is dueto different numbers of repeats. Probes for both the tandem and theunique sequences were used to study the DF3 locus in human breasttumor DNAs. Seventy of 110 breast tumor DNAs were informative atthe DF3 locus. Of these, 20 (29%) showed a loss of heterozygosity, whileeight (11%) had an increased copy number of one alíele.In some cases,the loss of heterozygosity or increased copy number did not extend toother markers on chromosome Iq or Ip. These data indicate that thechromosomal region around the DF3 locus is affected by mutations athigh frequency.

INTRODUCTION

The importance of genetic alterations in neoplastic transformation is increasingly well documented. Mutations as diverseas gene amplifications ( 1), translocations and deletions of chromosomal segments (2), often affecting cellular protooncogenes(3, 4), have been reported in a wide variety of human tumors.However, the precise role these mutations play in multisteptransformation is still unclear.

Molecular, biochemical, and hormonal factors are thought tocontribute to the development of human breast carcinoma(reviewed in Refs. 5 and 6). Genetic analysis of human breasttumors reveals a number of mutations which occur frequentlyand are associated with relevant clinicopathological parametersof the disease (reviewed in Refs. 7 and 8). Amplification of thec-myc (9, 10), int-2 (11), and c-erbB-2 (12, 13) protooncogeneshave been reported in subsets of human breast tumors, pointingto the importance of dominantly acting cellular protooncogenesin breast carcinogenesis.

LOH4 at multiple loci have been observed on chromosomes

lip (14) and 13 (15) in primary human breast tumors. Deletions, such as those involving chromosome 13ql4 in humanretinoblastoma (16, 17), are thought to unmask recessive mutations affecting tumor-suppressor genes. Structural rearrangements of the retinoblastoma susceptibility gene in primary human breast tumors have also been reported (18).

Received 5/24/89; revised 9/18/89; accepted 9/21/89.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.

1This investigation was supported in part by USPHS Grant CA-38869 fromthe National Cancer Institute, DHHS. and by a Burroughs Wellcome Award inClinical Pharmacology (D. W. K.).

2 Present address: Servizio Anatomia Pathologica. USL-1, Via Cavour 31.

10132 Torino, Italy.'To whom requests for reprints should be addressed, at Dana-Farber Cancer

Institute. Division of Medicine. 44 Binney Street. Boston, MA 02115.4The abbreviations used are: LOH, loss of hetero/.ygosity: RFLP. restriction

fragment length polymorphism; VNTR, variable number of tandem repeats; NT,nucleotide: MAb, monoclonal antibody; PUM, polymorphic peanut-lectin-bind-ing urinary mucin; cDNA. complementary DNA.

The DF3 human breast carcinoma-associated antigen is ahigh molecular weight glycoprotein, first identified by a MAbprepared against a membrane-enriched fraction of a humanbreast tumor (19). The DF3 antigen is present on the apicalborder of normal secretory mammary epithelium, at elevatedlevels in the cytosol of less differentiated malignant cells (19),and in the plasma of breast cancer patients (20). The electro-phoretic mobility patterns of the DF3 antigen differ amongindividuals (21). Family studies demonstrated that this heterogeneity is determined by codominant expression of a multial-lelic system at a single locus (21). Using a cDNA coding forthe DF3 antigen, we demonstrated that the polymorphic expression is related to the size of the DF3 alíeles(22). DNA sequenceanalysis revealed a highly conserved (G+C)-rich 60 base pairVNTR (22). Similar findings had been previously reported forthe PUM gene (23, 24). The PUM gene contains a 60-base pairVNTR and nucleotide sequence analysis has indicated that thePUM and DF3 tandem repeats are identical (22, 25). The DF3/PUM locus has been mapped to human chromosome lq21-24(26).

With the availability of a DF3 cDNA clone we sought toextend the sequence analysis to the regions flanking the VNTR,and to investigate the integrity of the DF3 locus in panels ofprimary human breast tumors.

MATERIALS AND METHODS

Clone Isolation and DNA Sequence Analysis. An oligo(dT)-primedcDNA library in XgtlO was prepared from MCF-7 human breastcarcinoma cells and screened with 12P-labeled pDF9.3 probe (22).

Positive clones were isolated from the unamplified library and furtherpurified to homogeneity by repeated plaque hybridization according topublished procedures (27). The cDNA fragments from purified cloneswere cleaved with EcoR\ and subcloned into pGEM-3 (Promega Biotech, Madison, WI). The largest cDNA clone, pDFl.8, was digestedwith Pstl or Kpnl (Fig. \A). The resulting fragments were subclonedinto the respective restriction sites of M13mp 18 or mp 19 and sequencedby the dideoxy chain termination method (28, 29).

Tumor Specimens. Breast tumor biopsies were collected from patientsat the Centre Rene Huguenin, St. Cloud, France. Samples were frozenin liquid nitrogen immediately after surgery and kept at -70°C until

DNA extraction. The patients had not undergone treatment prior tosurgery and complete clinicopathological records were available formost of them. WBCs were collected at the time of surgery and used assources of constitutional DNA for comparison.

DNA Extraction and Southern Hybridization. Genomic DNAs wereextracted (30) and 10 ^g were digested with the restriction enzyme ofchoice. The resulting fragments were fractionated in 0.8% agarose gelsand transferred to Genatran 45 (Fiasco, Woburn, MA) nylon membranes. The membranes were baked at 80°Cfor 2-3 h, prehybridizedand hybridized with "P-labeled DNA probes by standard procedures.Probes were labeled with 12P-dCTP to a specific activity of 1-2 x IO9

cpm//ig using the random primers system (31). After hybridization,filters were washed under stringent conditions (15 mM sodium chloride,1.5 mM sodium citrate, pH 7, 65°C,30 min), exposed for 1-4 days to

X-ray films and the results quantitated by scanning densitometry.

6966

Research. on January 28, 2020. © 1989 American Association for Cancercancerres.aacrjournals.org Downloaded from

ALTERATION OF DFJ TUMOR-ASSOCIATED ANTIGEN GENE

DNA Probes. The 309-base pair EcoRl-EcoR\ DF9.3 cDNA clone,which spans the VNTR, was used for hybridization analysis on BamHl-or /tol-digested DNAs. A 472-base pair Pst\-Pst\ fragment from thepDFl.8 cDNA clone, designated pDFl.8 (Pst), was used for hybridization on ÄamHI-digested DNAs only. Probes for human antithrombinIII (pAT3; 32) and human renin (pHRnESl.9; 33) genes, which identifyDNA sequences located on chromosome lq23-25 (34) and on chromosome Iq42 (35), respectively, were obtained from the AmericanType Culture Collection. L-myc (Ip32; 36) and DlS57/pYNZ2 (Ip;37) markers were used to identify EcoRl and Pst\ RFLPs, respectively.The amount of DNA transferred on li,:ni\\\ blots was standardized byhybridization with probes for p53 (chromosome 17; 38) and/or hcl-i(llq: 39). Pstl blots were standardized with a parathyroid hormonegene, PTH-LF (lip: 14, 40). Purified insert DNAs were used for thehybridization experiments.

RESULTS

Isolation and Sequencing of pDFl.8 cDNA Clone. We previously reported the nucleotide sequence of the pDF9.3 cDNAclone which was isolated by screening a Xgtl 1 expression librarywith MAb DF3 (22). This 309-base pair cDNA consisted entirely of highly conserved (G+C)-rich 60-base pair tandemrepeats. In the present study, the pDF9.3 fragment was used toscreen a XgtlO cDNA library from the MCF-7 breast carcinomacell line. The largest clone, pDFl.8 (1831 base pairs), consistedof 10 60-base pair repeats and 1220 base pairs of nonrepetitivesequence, which were 3' to the repeat. A simplified restriction

map and the complete nucleotide sequence of the pDFl.8 cloneare shown in Fig. 1. A Pstl fragment, DFl.SíPíí),Nt 947 to1446, which contained only nonrepetitive sequences was usedas a probe to analyze AamHI-digested breast tumor DNAs. ThepDF9.3 fragment (22) was also used to hybridize BamHl orPstl blots. Based on fragment size, multiple alíelesranging from8.4 to 14 kilobases and from 3.1 to 8 kilobases were identifiedon BamHl- and /Vl-digested DNAs, respectively.

Bamlll and Pstl RFLPs are Identical. pDF1.8(/)5i) and

pDF9.3 fragments recognize a BamHl RFLP due to the presence of the 60 bp DF3 VNTR within BamHl sites (Fig. 2, Aand C). Pstl-digested DNAs also displayed a RFLP whenhybridized with pDF9.3 (Fig. 2A), but not with DF1.8(Psf)-This result is consistent with the restriction map (Fig. IA). Wehave analyzed the segregation pattern of DF3 alíelesin members of several families. The results of one of these analyses andthe corresponding family tree are shown in Fig. 2, A and B.Smaller-sized fragments were obtained with Pstl (Fig. 2A, bottom panel) compared to BamHl (Fig. 2A, top panel). This occurspresumably because Pstl cuts closer to the VNTR than BamHl.However, the genotypes determined with the two probes co-segregated in each of the families. Using pDF9.3 as the probe,the higher molecular weight fragments tended to display astronger signal, compared to lower molecular weight fragments.This finding is consistent with the VNTR nature of the pDF9.3fragment.

When BamHl digest membranes were hybridized with thepDF9.3 probe and subsequently rehybridized with pDFl.8 fragment, the allelic status and the pattern of bands observed wereidentical (Fig. 2C). This observation indicates that these twoprobes recognize the same human genomic fragment and further confirms the fact that they are part of the same gene.

Genetic Alterations in Breast Tumor DNAs. Human WBCDNA from 48 unrelated donors was hybridized with pDF9.3and pDF1.8(fs/) probes. Twenty-seven individuals (56%) wereheterozygous at the locus. Hybridization with pDFl.&(Pst)showed that the signal intensity was independent of the hag

s' [TGTCACCTCGGCCCCGGAGAOCAGGCCGGCCCCOOGCTCCACCGCCCCCCCAGCCCACCG) ,t

ValThrSerAlaProGluSerArgProAlaProGlySerThrAlaProProAlaHisArq

J 3'

CCTCAGGCTCTGCATCAGOCTCAGCTTCTACTCTGTOCACACAACOOCACn ProGlnAlaLeuHisGlnAlaGlnLeuLeuLeuCysAlaHisAsnGlyTh

CTCTGCCAGGGCTACCACAACCCGGCTACCACAACCCCAGCCAGCAAGAGCACTCCATTCFSerAlaArgAlaThrThrThrArgLeuProGlnProGlnProAlaArgAlaLeuHisSe

TCAATTCCCAGCCACCACTCTGATACTCCTACCACCCTGCCAGCCATACGACCAAGACTGrGlnPheProAlaThrThrLeuIleLeuLeuProProCysGlnProTyrAspGlnAspSlpp

ATGCCAGTAGCACTCACCATAGCACGGTACC TCTCACCTCCTCCAATCACAGCACTT

CTTCCCAGTTGTCTACTGGGGTCTCTTTCTTTTTCCTGTCTTTTCACATTTCAAACCTCCP»H

ACTTTAATTCCTCTCTOGAAGATCCCAGCACCGACTACTACCAAGAGCTGCAGAGAGACA

TTTCTGAAATGTTTTTGCAGATTTATAAACAAGGGGGTTTTCTGGGCCTCTCCAATATTAKjnl

AGTTCAGOCCAAGGATCTGTGGTGGTACAATTGACTCTGGCCTTCCGAGAAG'GTACCA'TC

AATOTCCACGACaGTAGGAGGAACCAACAOTTCAATCAGTATAAAACGGAAGCAOCCTCT

CGATATAACCTGACGATCCCAGACGTCAGCaTGAGTGATaTGCCATTTCCTTCTCTGCCC

AGTCTGGGGCTGGGGTGCCAGGCTGGGGCATCGCGCTGCTGGTGCTGGTCTGTGTTCTGG

TTGCGCTGGCCATTGTCTATCTCATTGCCTTGGCTGTCTGTCAGTGCCGCCGAAAGAACT

ACGGGCAGCTOGACATCTTTCCAGCCCGGGATACCTACCATCCATTAGGCGAGTACCCCA

CCTACCACACCCATGGGCGCTATGTGCCCCTAGCAGTACCGATCGTAGCCCCTATAGAAG

GCCACTTCTGCCAACTTGTAGGGGCACGTCGCCCGCTGAGCTGAGTGGCCAGCCAGTGCC

ATTCCACTCCACTCAGGTTCTTCAGGCCAGAGCCCCTGCACCCTGTTTGGGCTGGTGAGC

TGGGAGTTCAGGTGGGCTGCTCACAGCCTCCTTCAGAGGCCCCACCAATTTCTCGGACAC

TTCTCAGTGTGTGGAAGCTCATGTGGGCCCCTGAGGGCTCATGCCTGGGAAGTGTTGTGG

TGGGGGCTCCCAGGAGGACTGGCCCAGAGAGCCCTGAGATAGCGGGGATCCTGAACTGGA 1800

CTGAATAAAACGTGGTCTCCGGAATTGCGG 3'

B.Fig. 1. A, restriction map of the pDFl.8 cDNA clone. The hatched box (Nt

1-610) represents the 10 60-base pair repeats. The Psll fragment from Nt 947 toNt 1446 was designated pDFLSIftf) and includes only unique sequences. TheEcoRl restriction sites at the termini of the fragment were added during thesubcloning procedure. B. complete 5'-3' nucleotide sequence of PDFl.8 clone.The sequence of the 60-base pair repeat is shown on the first line. Nine identicalrepeats were found, the 10th repeat showed three Nt changes. The region fromNt 61 1 to 1831 represents unique sequence. The deduced amino acid sequencefor this part of the DF3 gene, derived from the nucleotide sequence, is reportedunderneath.

ment size. In contrast, hybridization with pDF9.3 showed signalintensity related to fragment size, thus confirming our earlierobservation (data not shown). The pDFLSiPsf) probe wastherefore used to analyze SawHI-digested tumor DNAs foralterations at the DF3 locus when matching WBC DNA wasnot available. Differential signal intensities, determined byscanning densitometry, were observed in several tumor samples.We considered a two-fold (or more) difference in the signalintensity significant for an alteration at the DF3 locus. Comparison of the hybridization signal obtained using other probesprovided a reference for interpretation of gene copy numberand DNA transfer to the filter in informative patients.

Ninety-seven primary human breast tumor DNAs were analyzed by BamHl only, 12 by Pstl only, and 10 by both BamHland Pstl digestions (Table 1). Seventy patients (59%) wereconstitutionally heterozygous (informative) for the DF3 RFLP.In tumor DNAs from 20 (29%) of these patients one DF3 alíelewas lost. LOH can result from mitotic nondisjunction, mitoticrecombination, deletion, or gene conversion (41 ). To investigatethe possible causes for LOH at the DF3 locus, we examinedother loci on chromosome 1 using RFLP analysis. The poly-

6967

Research. on January 28, 2020. © 1989 American Association for Cancercancerres.aacrjournals.org Downloaded from

ALTERATION OF DF3 TUMOR-ASSOCIATED ANTIGEN GENE

1 4 5 6 10 7 8

Fig. 2. /(, comparison of the hybridizationresults obtained using the pDF9.3 probe onBamHl- and ftrl-digested cellular DNAs froma family. The numbers correspond to specificfamily members, whose pedigree is reported inB. Top. hybridization of ßamHI-digestedDNAs with the pDF9.3 probe; bottom, resultof hybridization on ftrl-digested DNAs for thesame individuals indicated on top. using thepDF9.3 probe. The migration of the molecularweight marker for both panels is reported onthe right. B, family pedigree, with specificmembers indicated by numbers; £',demonstra

tion of RFLP obtained by rehybridizing thesame BamHl digest membrane with pDF9.3(fop) and pDFl .8(ftr) (bottom) probes. Lanes,10/ig of breast tumor DNAs from II unrelatedindividuals. The pattern and the size of thebands observed are identical between the twoprobes, indicating that the two probes hybridize to the same gcnomic fragment.

».

OF 9.3

DF1.8

—¿�9.4

—¿�6.5

—¿�4.3

LJ3Ó6O7~WW4

57S(j

i8ni i9

10

B

Table 1 Number of patients analyzed for each marker

IPTotal

analyzed"Informative (cí)*Alterations (%)cD1S57pYNZ234

29 (85)8(27)L-myc104

56 (54)13(26)DF311970(59)28 (40)lqAT335

15(42)3 (20)HRnES1.954

25 (46)8 (32)

a Includes patients analyzed with cither BamY\\ or Pstl or both.* Parentheses, percentage of informative patients calculated on the total num

ber of patients analyzed.' Parentheses, is the percentage of informative patients showing a numerical

alteration (either LOH or increased copy number) calculated on the total numberof informative patients.

morphic loci tested were L-myc (Ip32), the anonymous DNAsegment DlS57/pYNZ2 (Ip), pAT3 (Iq23), and pHRnES1.9(Iq42). Table 1 summarizes the frequency of LOH (or otheralterations) detected with DNA probes corresponding to theseloci. Of the 20 tumor DNAs containing LOH at the DF3 locus,14 were from patients informative at one or more other markers

on chromosome 1 (Table 2). In tumor DNAs 304 (Fig. 3), 77,112, and 167 (Table 2) LOH was detected at loci on both armsof chromosome 1. LOH in these tumors probably reflectschromosome loss with (tumor DNAs 112 and 167) or without(77 and 304) reduplication of the remaining chromosome. Intumor DNAs 223 and 289 (Fig. 3), 39, 79, 173, 180, 228, and291 (Table 2) LOH was restricted to loci on chromosome lq.LOH in these cases is probably the result of deletion of portionsof the long arm of one chromosome homologue. The patientscorresponding to tumors 70 and 316 were uninformative at theL-myc and D1S57 loci (Table 2). However in tumor DNA 70,constitutional heterozygosity was maintained at the AT3 locus.This suggests that the breakpoint at one end of the deletion islocated between these two loci.

In eight tumor DNAs, constitutional heterozygosity of DF3was maintained with one alíelehaving an increased copy number. Six of these tumors were from patients that were informative at other chromosome 1 loci. In tumor DNA 103 (Table

6968

Research. on January 28, 2020. © 1989 American Association for Cancercancerres.aacrjournals.org Downloaded from

ALTERATION OF DF.l TUMOR-ASSOCIATED ANTIGEN GENE

Table 2 Tumor DNAs containing an alteration at the DF3 locus from patients that were informative at other chromosome I lociTumor DNAs containing cither a LOH or an increased copy number were from patients that were informative at other loci on chromosome I. The pYNJ¿2and

L-myc probes detect sequences on chromosome Ip, whereas pDF9.3. pDFLSfAr). pAT3. and pHRnESI.9 detect sequences at Iq. A dot (.) indicates the tumor DNAwas not tested. A dash (-) indicates the patient was uninformative and the tumor DNA had a normal copy number, a and b. large and small polymorphic restrictionfragments. The presence of only a or b indicates LOH in the tumor DNA and the alíelewhich was detected. [AMP], gene amplification.

IP lq

Tumor DNADIS57pYNZ2

Locus (probe) DF3

L-myc DF9.3 AT3 pAT-3 HRnES1.9

39707779

112167173180223228289291304316103185200237299309

a/ba/ba/ba/ba

a/ba/ba/ba/b

a/b

aa/ba/bbbba/ba/b

a/bba/b

a/h

a[AMPl/ba/bbaa/ba/bb

ahhba/bbaaah

ba/bbaa/ba[AMP]/ba/bbaa/b

a/b

a/b[AMP]a/b

x

304T L

—¿�pYNZ2

289T L

-223

T L

«.

fpDF3dIMb

\pAT3

237T L

200T L

-10.5

- 5.5- 5.0

A. B.Fig. 3. Genetic analysis of the DF3 locus in primary human breast tumor DNAs. Each of the tumor (T) and matching lymphocyte (/.) DNAs »eredigested with

Pstl or /.'K/Õ/III and fractionated by agarose gel electrophoresis. All the samples in this figure represent rehybridi/ation of the same Pslt digest membrane, exceptsample 223 for pDF3 which is a BamHl digest. Probes pYN/2 and pDF3 showed polymorphic bands with either P\t\- or AuniHI-digested DNAs. Each probe detectsmultiple alíelesdue to the VTR sequences within the restriction sites. Probe pAT3 was polymorphic only with Pst\ digested DNAs. It delects a two-allele RFLPconsisting of 10.5-kilobasc and 5.5 -I-5.0-kilobase fragments. In A. samples 304. 289. and 223 show deletion mutation identified by the probes. In lì.tumor samples237 and 200 represent examples of duplication and amplification, respectively, at the DF3 locus.

6969

Research. on January 28, 2020. © 1989 American Association for Cancercancerres.aacrjournals.org Downloaded from

ALTERATION OF DF3 TUMOR-ASSOCIATED ANTIGEN GENE

2) trisomy of chromosome 1 has occurred, since one alíeleeachof L-myc and DF3 was duplicated. A twofold amplification of

one DF3 alíelewas detected in tumor DNA 237 (Fig. 3), 185,299, and 309 (Table 2). Similarly, one DF3 alíelewas amplified5-fold in tumor DNA 200 (Fig. 3). Hybridization with eitherL-myc or pYNZ2 and pAT3 showed a normal copy number for

tumor 237 (Fig. 3) and 309 (Table 2). In tumor 299 (Table 2)the amplification unit includes the AT3, but not the renin locus.

DISCUSSION

The pDF9.3 probe recognized BamHl and Pstl RFLPs, as aresult of the VNTR. Indeed, signal intensity with pDF9.3 wasdirectly related to fragment size. In contrast, use of thepDFl.S(Pst) cDNA probe, which represents nonrepetitive sequences 3' to the VNTR, revealed signal intensity independent

on fragment size, while still recognizing the same RFLP. Thesenonrepetitive sequences 3' to the VNTR of the DF3 gene are

completely different from those reported for the PUM gene(25). While the basis for this discrepancy remains unclear, thefinding that the pDF\.8(Pst) probe recognizes the same gen-omic fragments as the VNTR probe confirms that they areindeed part of the same gene. The present results also demonstrate some degeneracy of the 60-base pair VNTR sequencewith three nucleotides differing in the pDFl.8 clone from thatfound using other libraries prepared from MCF-7 cells (22, 25).Thus the pDFl.8 clone may have been derived from the otheral lui¡ccopy of this gene in MCF-7 cells or from nonoverlappingrepeats in the same alíele.Using the pDF9.3 and pDF1.8(ftf)probes, we demonstrate that the DF3 locus identifies a regionwith a high frequency of numerical alterations in primaryhuman breast tumors. These alterations include chromosomeloss (with and without reduplication of the remaining chromosome), probable deletion of sequences on one homologue, andgene amplification.

Previous cytogenetic studies demonstrated that chromosome1 is frequently altered in many human malignancies (42). Inhuman breast tumors, the distal portion of chromosome Iq isfrequently over represented as result of trisomy or structuralalterations such as translocations (reviewed in Ref. 43). However, the translocation breakpoints as well as the other chromosomal partner in the translocation are variable (43). Improved cytogenetic techniques have recently been described forprimary breast tumor cells which may facilitate the detectionof minute chromosomal alterations that would otherwise remain unrevealed (44). Our study reports that deletions may alsocontribute to the LOH on chromosome Iq in breast tumors.Based on the ubiquitous nature of chromosome Iq alterationsin many solid tumors, it has been suggested (43) that theyrepresent events associated with tumor progression rather thanwith the initiation of tumor development.

Using the same panel of tumors, we previously reportedamplification of the int-2 (chromosome Ilql3), c-myc (chromosome 8q), and c-erbB-2 (chromosome 17q) protooncogenesin 16, 32, and 10% of the tumor DNAs, respectively (11, 10,13). In addition, LOH at multiple loci on chromosome 1Ip hasbeen detected in 20% of the breast tumor DNAs (14). Thedetection of multiple mutations in this panel of DNAs has notbeen limited to a specific subset of the breast tumors. Indeed,the frequency of alterations involving other chromosomes issimilar in both tumors with the LOH on chromosome Iq andin those without this mutation. The presence of multiple mutations in the same subset of tumors could reflect the accumulation of genetic damage during malignant progression. In

contrast, detection of the LOH on chromosome Iq and othermutations at high frequency in different tumors may reflectspecific events that contribute to progression in breast carci-nogenesis.

LOH at the Rb locus in retinoblastomas is hypothesized tounmask a recessive mutation which inactivates a suppressorgene. It is possible the LOH we observed on chromosome Iqmay represent an analogous phenomenon. Reexpression oftumorigenicity upon loss of chromosome 1 in cell hybridsbetween human fibrosarcoma and normal fibroblasts supportsour analogy (45). Similar observations have occurred with hybrids of transformed baby hamster kidney cells and humanfibroblasts (46). A separate panel of breast tumors will be usedto more precisely define the region of chromosome Iq affectedby this LOH.

ACKNOWLEDGMENTS

We thank Drs. J. Minna, O. McBridge (National Cancer Institute,Bethesda), and C. Croce (Wistar Institute, Philadelphia) for kindlyproviding some of the probes used in this study. The pYNZ2 markerwas kindly provided by Dr. Y. Nakamura (Howard Hughes MedicalInstitute, Salt Lake, City, UT).

REFERENCES

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

Alitalo. K., and Schwab. M. Oncogene amplification in tumor cells. Adv.Cancer Res.. 47: 235-281, 1986.Ponder, B. Gene losses in human tumors. Nature (Lond.), 335: 400-402.1988.Land, II.. Parada, L. F., and Weinberg, R. Cellular oncogenes and multistepcarcinogenesis. Science, 222: 771-778, 1983.Bishop, J. M. The molecular genetics of cancer. Science (Wash. DC), 235:305-311, 1987.Harris, J. R.. Hellman. S., Canellos. G. P.. and Fisher. B. Cancer of thebreast. In: V. T. Devita. S. Hellman. and S. A. Rosenberg (eds.). CancerPrinciples and Practice of Oncology, pp. 1119-1178. Philadelphia: Lippin-cott Co., 1985.McGuire. W. L. Prognostic factors in primary breast cancer. Cancer Surv.,5:527-536, 1986.Ali, I. I '.. and Callahan. R. Prognostic significance of genetic alterations in

human breast cancer. In: J. Cossman (ed.). Molecular Genetics and theDiagnosis of Cancer. New York: Elsevier Pubi. Co.. 1988.Ali. I. U.. Lidereau, R., and Callahan. R. Heterogeneity of genetic alterationsin human breast carcinoma. In: M. Lippman. and R. Dikson (eds.). BreastCancer, Cellular & Molecular Biology, pp. 25-48. Boston: Martinus NijhoffPubi.. 1988.Kozbar, D. H., and Croce. C. Amplification of the c-myc oncogene in one offive human breast carcinoma cell lines. Cancer Res., 44:438-441, 1984.Escot, C.. Theillet, C.. Lidereau, R.. Spyratos. F.. Chámpeme. M.. Gest. J.,and Callahan, R. Genetic alteration of the c-myc protooncogene (MYC) inhuman primary breast cancer. Proc. Nati. Acad. Sci. USA, 83: 4824-4838,1986.Lidereau. R.. Callahan. R.. Dickson. C., Peters. G.. Escot. C., and Ali, I.Amplification of the int-2 gene in priman human breast tumors. OncogeneRes.. 2:285-291. 1988.Slamon. D.. Clark. G.. Wong, S., Levin, W.. Ullrich. A., and McGuire, W.Human breast cancer: correlation of relapse and survival with amplificationof HER-2/neu oncogene. Science (Wash. DC). 235: 177-182. 1987.Ali, I. U., Campbell, G., Lidereau, R.. and Callahan, R. Lack of evidence forthe prognostic significance of c-fr/>-B2 amplification in human breast carcinoma. Oncogene Res., 3: 139-146. 1988.Ali. !.. Lidereau, R.. Theillet, C.. and Callahan. R. Reduction of homozygosityof genes on chromosome 11 in human breast neoplasia. Science (Wash. DC).238: 185-188. 1987.Lundberg. C., Scoog, L., Cavenee. W. R.. and Nordenskjoeld. M. Loss ofheterozygosity in human ductal breast tumors indicate a recessive mutationin chromosome 13. Proc. Nati. Acad. Sci. USA, 84: 2372-2376, 1987.Dryja, T., Rapaport. J., Joyce. J.. and Petersen, R. Molecular detection ofdeletions involving band 14 of chromosome 13 in retinoblastomas. Proc.Nati. Acad. Sci. USA. 83: 7391-7394, 1986.Lee, W-H., Bookstein. R., Hong. F.. Young. L-J., Shew. J-Y.. and Lee, Y.-P. Human retinoblastoma susceptibility gene: cloning, identification andsequence. Science (Wash. DC). 235: 1394-1399. 1987.T'ang, A., Varley, J. M., Chakraborty. S.. Linn Murphree. A., and Fung, Y-

K. T. Structural rearrangement of the retinoblastoma gene in human breastcarcinoma. Science (Wash. DC). 242: 263-266. 1988.

6970

Research. on January 28, 2020. © 1989 American Association for Cancercancerres.aacrjournals.org Downloaded from

ALTERATION OF DF3 TUMOR-ASSOCIATED ANTIGEN GENE

19. Kufe. D., Inghirami, G., Abe, G., Hayes, D.. Jusli-Wheeler, H., and Scholm,J. Differential reactivity of a novel monoclonal antibody (DF3) with humanmalignant versus benign breast tumors. Hybridoma, 3: 223-232, 1984.

20. Hayes, D. F., Sckine, H., Ohno, T., Abe, M., Keefe, K., and Kufe, D. Use ofmurine monoclonal antibody for detection of circulating plasma DF3 antigens levels in breast cancer patients. J. Clin. Invest.. 75: 1671-1678. 1985.

21. Hayes, D., Sekine, H., Marcus, D., Alper. C., and Kufe, D. Geneticallydetermined polymorphism of the circulating human breast cancer-associatedDF3 antigen. Blood, 77:436-440, 1988.

22. Siddiqui, J., Abe. M.. Hayes, D.. Shani, E.. Yunis, E., and Kufe, D. Isolationand sequencing of a cDNA coding for the human DF3 breast carcinoma-associated antigen. Proc. Nati. Acad. Sci. USA, 85: 2320-2323. 1988.

23. Swallow, D., Gendler, S., Griffiths, B.. Corney. G.. Taylor-Papadimitriou.J.. and Bramwell. M. The human tumor-associated epithelial mucins arecoded by an expressed hypervariable gene locus PUM. Nature (Lond.). 328:82-84, 1987.

24. Gendler, S., Burchell, J., Duhig, T., Lamport, D.. White, R.. Parker. M.,and Taylor-Papadimitriou, J. Cloning of partial cDNA encoding differentiation and tumor-associated mucin glycoproteins expressed by human mammary epithelium. Proc. Nati. Acad. Sci. USA, 84:6060-6064, 1987.

25. Gendler, S., Taylor-Papadimitriou, J.. Duhig. T., Rothbard. J.. and Burchell.J. A highly immunogenic region of a human polymorphic epithelial mucinexpressed by carcinomas is made up of tandem repeats. J. Biol. Chem.. 263:12820-12823. 1988.

26. Swallow. D., Gendler. S., Griffiths. B., Kearney, A., Povey. S., Sheer, D.,Palmer, R. W., and Taylor-Papadimitriou, J. The hypervariable gene locusPUM, which codes for the tumor-associated epithelial mucins, is located onchromosome 1 within the region Iq 21-24. Ann. Hum. Genet.. 51: 289-294,1987.

27. Huynh. T. V., Young, R. A., and Dawis, R. W. In: D. Glover (ed.), DNACloning Techniques: A Practical Approach. Oxford. England: IRL Press.1985.

28. Sanger, F., Nicklen, S., and Coulson, A. DNA sequencing with chain-terminating inhibitors. Proc. Nati. Acad. Sci. USA, 74: 5463-5467, 1977.

29. Tabor. S., and Richardson, C. C. DNA sequence analysis with a modified T7DNA polymerasc. Proc. Nati. Acad. Sci. USA, «4:4767-4771, 1987.

30. Gross-Bellard, M., Oudet, P., and Chambón. P. Isolation of high molecularweight DNA from mammalian cells. Eur. J. Biochem., 36: 32-38. 1973.

31. Feinberg, A. P., and Vogelstein. B. A technique for radiolabcling DNArestriction endonueleasc fragments to high specific activity. Anal. Biochem..m-6-13. 1983.

32. Prochownik. E., Markham. A., and Orkin, S. Isolation of a cDNA clone foranti-thrombin III. J. Biol. Chem.. 25«:8389-8394, 1983.

33. Hobart, P.. Fogliano, M., O'Connor. B.. Schaefer. I., and Chirgwin. J.

Human renin gene: structure and sequence analysis. Proc. Nati. Acad. Sci.USA, «/:5026-5030. 1984.

34. Bock. S. C., Harris. J. F.. Balazs, I., and Trent. J. M. Assignment of thehuman anti-thrombin III structural gene to chromosome Iq 23-25. Cytoge-net. Cell Genet.. 39: 67-69, 1985.

35. Nakai, H., Inoue, S., Miyazaki, H., Murakami. K.. and Tada, K. Humanrenin gene assigned to chromosome band Iq 42 by in situ hybridization.Cytogenet. Cell Genet., 47: 90-91, 1988.

36. Farndon, P. A., and Simmons, J. LOD scores for markers on chromosome1. Cytogenet. Cell Genet. HGM., 9: 612, 1987.

37. O'Connel, P., Leppert, M., Nakamura. Y., Lathrop. G. M., Cartwright, P.,

Lalouel, J. M., and White, R. A primary genetic linkage map of chromosome1. Cytogenet. Cell Genet. HGM., 9: 672, 1987.

38. McBride, O. W., Merry, D., and Givol, D. The gene for human p53 cellulartumor antigen is located on chromosome 17 short arm (17pl3). Proc. Nati.Acad. Sci. USA, «.*:130-134, 1985.

39. Tsujimoto, Y., Jaffe, E., Cossman, J., Gorham, J., Nowell, P. C., and Croce,C. Clustering of breakpoints on chromosome 11 in human B-cell neoplasmswith the (tl 1;14) chromosomal translocation. Nature (Lond.). 315: 340-343.1985.

40. Antonarakis, S. E., Phillips, J. A.. Mallonce, R. L.. Kazazian, H. H.. Fearon.E. R., Waber, P. C.. Kronenberg, H. K., Ullrich, A., and Meyers, D. A. Beta-globin locus is linked to the parathyroid hormone (PTH) locus and liesbetween the insulin and PTH loci in man. Proc. Nati. Acad. Sci. USA, 80:6615-6619, 1983.

41. 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.

42. Atkin, N. Chromosome 1 aberrations in cancer. Cancer Genet. Cytogenet.,21: 279-285. 1986.

43. Trent, J. M. Cytogenetic and molecular biologic alterations in human breastcancer: a review. Breast Cancer Treat., 5: 221-229, 1985.

44. Devilee. P.. Thierry, R. F.. Kievits, T.. Kolluri. R., Hopman. A. H. N.,Willard, H. F.. Pearson. P. L., and Cornelisse, C. J. Detection of chromosomeaneuploidy in interface nuclei from human primary breast tumors usingchromosome specific repetitive DNA probes. Cancer Res., 4«:5825-5830,1988.

45. Benedict, W. F.. Weissman, B. E., Mark, C., and Stanbridge, E. J. Tumori-genicity of human HT1080 fibrosarcoma x normal fibroblast hybrids: chromosome dosage dependency. Cancer Res., 44: 3471-3479. 1984.

46. Stoler. A., and Bouck. N. Identification of a single chromosome in the normalhuman genome essential for suppression of hamster cell transformation.Proc. Nati. Acad. Sci. USA. 82: 570-574. 1985.

6971

Research. on January 28, 2020. © 1989 American Association for Cancercancerres.aacrjournals.org Downloaded from

1989;49:6966-6971. Cancer Res   Giorgio R. Merlo, Javed Siddiqui, Craig S. Cropp, et al.   in Primary Human Breast Carcinomas

Tumor-associated Antigen GeneDF3Frequent Alteration of the

  Updated version

  http://cancerres.aacrjournals.org/content/49/24_Part_1/6966

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/49/24_Part_1/6966To request permission to re-use all or part of this article, use this link

Research. on January 28, 2020. © 1989 American Association for Cancercancerres.aacrjournals.org Downloaded from