Detailed Deletion Mapping of Chromosome Arm 3pin Breast Cancers: A 2-cM Region on 3p14.3-21.1and a 5-cM Region on 3p24.3-25.1 CommonlyDeleted in Tumors

Satoshi Matsumoto,1,4 Fujio Kasumi,2 Goi Sakamoto,3 Masahiko Onda,4 Yusuke Nakamura,5 and Mitsuru Emi1*1Department of Molecular Biology, Institute of Gerontology, Nippon Medical School, Kawasaki, Japan2Department of Surgery, Cancer Institute, Tokyo, Japan3Department of Pathology, Cancer Institute, Tokyo, Japan4First Department of Surgery, Nippon Medical School, Tokyo, Japan5Laboratory of Molecular Medicine, Institute of Medical Science, University of Tokyo, Tokyo, Japan

Loss of heterozygosity (LOH) on 3p is frequent in human renal cell carcinomas, lung cancers, and breast cancers. To define theregion(s) on 3p that harbor presumptive tumor suppressor gene(s) for breast cancer, we examined 196 primary breast tumorsfor their patterns of LOH at 22 microsatellite marker loci distributed along this chromosome arm. Allelic loss at one or moreloci was observed in 101 (52%) of these tumors. Detailed deletion mapping identified two distinct commonly deleted regions;one was localized to a 2-cM interval flanked by D3S1547 and D3S1295 at 3p14.3-21.1, and the other to a 5-cM interval flankedby D3S1286 and D3S1585 at 3p24.3-25.1. The FHIT gene lies in the vicinity of the proximal commonly deleted region. Attemptsto correlate LOH on 3p to clinicopathological parameters detected an association with the absence of the progesteronereceptor (P 5 0.0096). The results suggest that inactivation of unidentified tumor suppressor genes on 3p plays a role in themechanism whereby hormone dependency is lost in the course of breast carcinogenesis. Genes Chromosomes Cancer20:268–274, 1997. r 1997 Wiley-Liss, Inc.

INTRODUCTION

Frequent loss of heterozygosity (LOH) on chro-mosome arms 1p, 3p, 11p, 13q, 16q, 17p, and 17qhas been observed in primary breast cancers (Satoet al., 1990, 1991; Takita et al., 1992; Harada et al.,1994; Ito et al., 1995), and suggests that a variety oftumor suppressor genes may influence the develop-ment and/or progression of breast cancer. Cytoge-netic and molecular changes on 3p are key featuresof renal cell carcinomas (RCC) and small cell andnon-small cell lung cancers, where cytogeneticallyvisible deletions of 3p and LOH at loci on 3p havebeen described (Hibi et al., 1992; Druck et al.,1996; van den Berg et al., 1996). Two candidategenes on 3p have already been identified: the VHLgene, which was isolated from band 3p25 followinglinkage analysis of families with von Hippel-Lindau disease, a hereditary renal cell carcinomasyndrome (Latif et al., 1993), was found to harborsomatic mutations in a majority of sporadic renalcell carcinomas examined by Gnarra et al. (1994);and the FHIT gene was recently isolated from aregion at 3p14 that contains the fragile site FRA3Band the break point of the t(3; 8) translocationfound in familial cases of renal cell carcinoma.Abnormalities in RNA transcripts and allelic loss of

FHIT subsequently were reported in lung cancersand some digestive tract cancers (Ohta et al., 1996;Sozzi et al., 1996), as well as in breast cancers(Negrini et al., 1996).

Cytogenetically, complete or partial deletion ofthe short arm of chromosome 3 has been observedin primary cultures of breast cancer cells (Pandis etal., 1993; Lebeau et al., 1995; Teixeira et al., 1996).Others have observed LOH in sporadic breastcancers by means of conventional RFLP analyses(Ali et al., 1989; Sato et al., 1991; Chen et al., 1992,1994). A high-resolution deletion map would beuseful in defining the locations of putative tumorsuppressor genes for breast cancer on 3p. There-fore, we carried out an LOH analysis of 196 primarybreast cancers using 22 highly informative microsat-ellite markers for this region. In addition to con-structing a detailed deletion map of chromosome

Supported by: Ministry of Education, Science, and Culture ofJapan, Ministry of Health and Welfare of Japan, Genome Science:New Frontiers in Bioscience from the Ministry of Education,Science, and Culture of Japan, and Kanagawa Academy of Scienceand Technology Research Grants.

*Correspondence to: Mitsuru Emi, M.D., Ph.D., Department ofMolecular Biology, Institute of Gerontology, Nippon Medical School,1-396 Kosugi-cho, Nakahara-ku, Kawasaki 211, Japan.

Received 27 December 1996; Accepted 13 May 1997

GENES, CHROMOSOMES & CANCER 20:268–274 (1997)

r 1997 Wiley-Liss, Inc.

arm 3p, we looked for correlations between LOHon 3p and certain clinicopathological parameters.

MATERIALS AND METHODS

Samples and DNA Preparation

Carcinoma tissues and corresponding normal tis-sues were obtained from 196 patients with primarybreast cancers at the time of mastectomy at theCancer Institute Hospital. Portions of these tissueswere frozen immediately after surgery and stored at-80°C. Genomic DNA was extracted from frozentissues in a manner described previously (Sato etal., 1990)

LOH Analysis

For LOH analysis by a PCR-based approach, weused 22 microsatellite markers (Gyapay et al., 1994;Gemmill et al., 1995) covering the entire p arm ofchromosome 3. The order of loci based on theconsensus genetic and physical maps of 3p (Gyapayet al., 1994; Gemmill et al., 1995) is shown in Figure1. Each PCR reaction contained 20 ng of genomicDNA, 30 mM of Tris HCl (pH 8.8), 50 mM of KCl,2 mM of MgCl2, 5 mM of 2-mercaptoethanol, 100µM of dNTPs, 1.6 pmol each of [g-32P] ATP-end-labeled primer and non-labeled primer, and 0.25unit of Taq polymerase, in a final volume of 10 µl.Cycle conditions were 94°C for 4 min, then 30cycles of 94°C for 30 sec, 52-62°C for 30 sec, and72°C for 30 sec, with a final extension step of 5 minat 72°C, in a Gene Amp PCR System (PerkinElmer Cetus, CA). After amplification, PCR prod-ucts were diluted twofold with loading buffer andheat-denatured at 94°C for 5 min. A 3.5 µl aliquot ofeach sample was electrophoresed in a 6% polyacryl-amide gel containing 36% formamide and 8 M ureafor 2–4 hr at 2,000 volts. Gels were dried andexposed to X-ray film at room temperature for16–24 hr.

Definition of LOH

Signal intensities of polymorphic alleles werequantified by a Hoefer GS-300 scanning densitom-eter; peak areas corresponding to each signal werecalculated by electric integration using the GS-370electrophoresis data system (Hoefer Scientific In-struments, San Francisco, CA). The signal intensi-ties of alleles of tumor-tissue DNA were comparedto those of normal-tissue DNA. We judged areduction in signal intensity .50% to be loss ofheterozygosity. We distinguished LOH from chro-mosome multiplication by normalizing each signal

to the signal obtained when the same DNA wasanalyzed with markers for loci on other chromo-somes; thus, we compared band intensities ofalleles in normal DNA and tumor DNA detected byPCR of each chromosome 3 marker with intensitiesof bands detected by PCR of two control markers,IL1B on 2q and LDLR on 19q. Chromosome lossor gain is observed only rarely in these regions inbreast cancers by cytogenetic analysis, FISH analy-sis, or RFLP analysis. A preliminary comparativestudy of the microsatellite-PCR method and theFISH method for detection of 1p loss and 1q gain inthe same series of breast cancer specimens con-firmed the reliability of our microsatellite-PCRmethod for the detection of allelic loss and multipli-cation (unpublished observation).

Figure 1. Frequencies of LOH detected at 22 microsatellite markerson chromosome arm 3p. Locations and order of the markers (left) werederived from published linkage and physical information (Gyapay et al.,1994; Gemmill et al., 1995).

269DELETION MAPPING OF Ch 3p IN BREAST CANCER

Clinicopathological Parameters

The panel of parameters studied included: histo-logic type, tumor size and infiltration, and presenceand number of lymph node metastases. Tumorswere classified by pathologists according to thehistologic TNM classification and the histologictyping scheme of the Japanese Breast Cancer Soci-ety into the following types: noninvasive ductal(1a), invasive papillotubular (a1), invasive solidtubular (a2), invasive scirrhous carcinoma (a3), andother specific types (b). This classification is essen-tially the same as the World Health Organizationscheme for typing breast tumors. Estrogen receptor(ER) and progesterone receptor (PgR) status wasmeasured by radioreceptor assay according to astandard dextran-coated charcoal (DCC) method,using [125I]-estradiol as labeled ligand on fresh-frozen tissue (Otsuka Pharmaceutical). All samplescontaining .5 fmol of ER or PgR per mg proteinwere considered receptor- positive. The chi-squaretest and Fisher’s exact test were used for statistical

analysis of the results. One-tailed P values of lessthan 0.05 were considered statistically significant.

RESULTS

LOH was detected in 101 breast tumors (52%)with at least one of 22 polymorphic microsatellitemarkers on 3p. The marker loci and the observedfrequencies of LOH are listed in Figure 1. Thehighest frequency of LOH (45%) was detectedwith marker D3S1295 at 3p14.3-p21.1. Among the101 tumors with LOH, 29 showed LOH at allinformative loci, implying loss of the whole shortarm. The other 72 showed partial or interstitialdeletions which could be further classified intothree groups on the basis of their pattern of LOH.Figure 2A illustrates LOH patterns of 13 tumorsthat exhibited LOH in a small distal portion,including the p24.3-25.1 region. Figure 2B illus-trates the patterns of 35 tumors that exhibited LOHin a small proximal portion, including the p14.3-

Figure 2. Schematic representation of deletion mapping on 3p inbreast cancers. Allelic losses were found in a distal restricted region in13 tumors (A) and at a proximal restricted region in 35 tumors (B).Tumor identification numbers are at the top of each column. Solid and

unfilled circles indicate loss and retention of heterozygosity, respec-tively. Gaps reflect uninformativeness for specific markers. Commonlydeleted regions are shown as rectangles indicated with arrows.

270 MATSUMOTO ET AL.

21.1 region. Both regions were deleted in 24 othertumors.

Figure 3A shows autoradiograms representativeof cases that exhibited partial deletions aroundp24.3-25.1. These and other cases shown in Figure2A were used to localize the distal commonlydeleted region to a 5-cM interval flanked byD3S1286 and D3S1585 at p24.3-25.1. The VHLgene, previously mapped between D3S1259 andD3S1597 at 3p25 (Latif et al., 1993; Gemmill et al.,1995), lies telomeric to the commonly deletedregion in our panel of breast cancers.

Figure 3B shows autoradiograms representativeof cases that exhibited partial deletions aroundp14.3-21.1. Analysis of the LOH patterns in theseand the other cases shown in Figure 2B was used tolocalize the proximal commonly deleted region to a2-cM interval flanked by D3S1547 and D3S1295 at

3p14.3-p21.1. The recently isolated FHIT gene(Ohta et al., 1996) lies in the vicinity of theD3S1547 locus.

LOH at 3p24.3-25.1, 3p14.3-21.1, and on 3p ingeneral was compared with hormone receptor (ERor PgR) status (Table 1). PgR-negative status wasmore frequent in tumors that had lost one allele of3p24.3-25.1 (22 of 56, 39%) than in tumors thatretained both alleles (26 of 111, 23%) (P 5 0.0325).PgR-negative status was also more frequent intumors that had lost one allele of 3p14.3-21.1 (29 of73, 40%) than in tumors that retained both alleles(19 of 94, 20%) (P 5 0.0057). When analyzed as awhole arm, PgR-negative status was also morefrequent in tumors that had lost at least one allelefrom 3p (32 of 85, 38%) than in tumors withoutLOH (16 of 82, 20%) (P 5 0.0096). ER-negativestatus also had a similar tendency toward correla-

Figure 3. Representative autoradiograms that helped determinetwo distinct commonly deleted regions on 3p (A: the distal region, B:the proximal region). A vertical bar at the left of each set of diagrams ofchromosome arm 3p indicates the commonly deleted region. Tumornumbers are shown above each set of lanes; N and T indicate matchedDNA samples isolated from normal and tumor tissues, respectively.Tumor 388 retains heterozygosity at the D3S1286 locus, but shows

LOH at the D3S1560 locus. Tumor 524 shows retention of heterozygos-ity at the D3S1293 and D3S1585 loci, but shows LOH at the D3S1286locus. Tumor 200 retains heterozygosity at the D3S1547 locus, butshows LOH at the D3S1573 locus. Tumor 352 shows retention ofheterozygosity at the D3S1300 and D3S1581 loci, but shows LOH at theD3S1295 locus. Tumor 864 shows retention of heterozygosity at theD3S1312 and D3S1295 loci, but shows LOH at the D3S1547 locus.

271DELETION MAPPING OF Ch 3p IN BREAST CANCER

tion with LOH on 3p, although our data did notreach statistical significance. LOH on 3p had noparticular association with tumor size or infiltration,lymph node metastasis, or histopathological classifi-cation.

DISCUSSION

High-resolution deletion mapping carried out inthe present study using a large number of tumorsand a panel of highly informative microsatellitemarkers enabled definition of two small deletedregions of 3p in breast cancers. One commonlydeleted region was defined in a 5-cM intervalbetween D3S1286 and D3S1585 at 3p 24.3-25.1. Inprevious LOH studies, deletions in the 3p24-26region have been reported in various tumors, includ-ing lung cancer, RCC, oral squamous cell carci-noma, ovarian cancer, cervical cancer, and breastcancer (Chen et al., 1992, 1994; Hibi et al., 1992;Rimessi et al., 1994; Wu et al., 1994; Roz et al.,1996; van den Berg et al., 1996). These findings,together with our results, indicate the presence ofat least one tumor suppressor gene at 3p24-26 andraise the possibility that a single gene in thatlocation may be involved in tumors originating fromdifferent organs. The VHL gene, a tumor suppres-

sor gene at 3p25 that is associated with RCC, is alikely candidate for involvement in breast cancer aswell. However, Gnarra et al. (1994) found no pointmutations of VHL in breast cancers, and the com-monly deleted region at 3p24.3-25.1 which wedefined in breast cancers did not include the VHLgene. Thus, a tumor suppressor gene for breastcancer other than VHL may be present at thislocation.

The commonly deleted region we defined at3p14.3-21.1 in the present study lies within a 2-cMinterval between D3S1547 and D3S1295. Cytoge-netic abnormalities and allelic losses involving the3p12-21 region have been described before inbreast cancers. Teixeira et al. (1996) demonstratedfrequent interstitial 3p deletions, del(3)(p12-13p14-21), in primary cultures of breast tumor cells; Ali etal. (1989) detected LOH at 3p21-25 in 30% ofbreast tumors investigated with RFLP analysis;Bergthorsson et al. (1995) detected LOH at 3p14 in68% of breast cancers in eight breast cancer fami-lies; Sato et al. (1991) detected LOH at 3p13-14.3 in47% of the breast tumors they examined; and Chenet al. (1994) found a common deletion at 3p13-14 inbreast cancers through a combination of RFLPanalysis and FISH. Taking advantage of highly

TABLE 1. LOH on 3p and Clinicopathological Parameters

Whole p-arm At p24.3–25.1 At p14.3–21.1

LOH (no.) Retention (no.) LOH (no.) Retention (no.) LOH (no.) Retention (no.)

Histologic typea

1a 0 5 0 5 0 5a1 16 20 8 28 14 22a2 33 20 23 30 29 24a3 43 34 30 47 35 42

Tumor sizeb

t1 28 25 14 39 23 30t2 61 57 44 74 53 65t3 5 10 4 11 4 11

Lymph node metastasis(2) 46 49 31 64 38 57(1) 48 45 31 64 42 51

Hormone receptorc

ER(2) 41 30 28 43 35 36(1) 44 54 28 70 38 60

PgR(2) 32 16 22 26 29 19(1) 53 66 34 85 44 75

P 5 0.0096 P 5 0.0325 P 5 0.0057

a1a: noninvasive tubular carcinoma, a1: invasive papillotubular carcinoma. a2: invasive solid tubular carcinoma, a3: invasive scirrhous carcinoma.bt1: tumor %2 cm, t2: tumor ,2 cm %5 cm, t3: tumor .5 cm.cER(2) or PgR(2): estrogen receptor or progesterone receptor level below 5 fmol/mg protein. ER(1) or PgR(1): estrogen receptor or progesteronereceptor level above 5 fmol/mg protein.Of the total of 196 tumors we analyzed for 3p LOH, information on histologic type, tumor size, lymph node metastasis, and hormone receptors ER andPgR, respectively, were available in 171, 186, 188, 170, and 169 cases, respectively.

272 MATSUMOTO ET AL.

informative microsatellite markers, we were able todetect several small interstitial deletions at the3p14-21 region in breast tumors. This region encom-passes the homozygous deletion at 3p14.3-21.1 thatBuchhagen et al. (1994) detected in a primarybreast cancer by Southern blot analysis using D3S2as a probe. All these data support the idea that atumor suppressor gene associated with breast can-cer exists at 3p14-21.

Recently, the FHIT gene was isolated from the3p14 region, and abnormalities in RNA transcriptsof this gene were subsequently detected in diges-tive tract cancers and lung cancer (Ohta et al., 1996;Sozzi et al., 1996). Man et al. (1996) detected LOHin 44% of breast cancers at D3S1300, which mapswithin intron 5 of this gene. Panagopoulos et al.(1996) reported deletion of the FHIT gene in someatypical hyperplasias of the breast. Negrini et al.(1996) found that abnormalities in FHIT transcrip-tion was exhibited in 30% (9 of 30) of primary breastcancers. The present study shows that the FHITgene lies in the vicinity of our proximal commonlydeleted region and supports the idea that this geneis a candidate tumor suppressor gene in breastcancer.

Homozygous deletions have been found at 3p21.3in lung cancers as well (Yamakawa et al., 1993; Koket al., 1994). Hibi et al. (1992) showed that 3p21.3was the common deleted region and that it was lostin almost 100% of the cases in an RFLP study oflung cancer. In breast cancer, Eiriksdottir et al.(1995) reported that the highest levels (6 of 13cases) of allelic loss were detected at the markerPH3H2 on 3p21.3. The three markers we used for3p21.3, i.e., D3S1277, D3S1298, and D3S1613,detected LOH in 31/117 (27%), 49/157 (31%), and41/126 (33%) informative cases, respectively (Fig.1). In our series, the most frequent LOH wasobserved for marker D3S1295 at 3p14.3-21.1 (43/96, 45%).

Eiriksdottir et al. (1995) demonstrated that allelicloss on 3p was significantly correlated with lowhormone receptor content, which is in agreementwith our finding of a significant association be-tween LOH at 3p and loss of PgR. We previouslyreported associations of LOH at 17q21 and 17p13.3with the loss of ER and/or PgR (Ito et al., 1995).Associations between specific genetic alterationsand loss of hormone receptors suggests that inactiva-tion of tumor suppressor genes on 3p also plays arole in the mechanism whereby hormone depen-dency is lost in the course of breast carcinogenesis.

ACKNOWLEDGMENTS

The authors thank Drs. Futoshi Akiyama andGoi Sakamoto for pathological examinations, andDrs. Toyomasa Katagiri, Yousuke Harada, Isao Ito,and Kenji Kobayashi for preparing tumor DNAs.

REFERENCES

Ali IU, Lidereau R, Callahan R (1989) Presence of two members ofc-erbA receptor gene family (c-erbA beta and c-erbA2) in smallestregion of somatic homozygosity on chromosome 3p21-p25 inhuman breast carcinoma. J Natl Cancer Inst 81:1815–1820.

Bergthorsson JT, Eiriksdottir G, Barkardottir RB, Egilsson V, ArasonA, Ingvarsson S (1995) Linkage analysis and allelic imbalance inhuman breast cancer kindreds using microsatellite markers fromthe short arm of chromosome 3. Hum Genet 96:437–443.

Buchhagen DL, Qiu L, Etkind P (1994) Homozygous deletion,rearrangement and hypermethylation implicate chromosome re-gion 3p14.3-3p21.3 in sporadic breast-cancer development. Int JCancer 57:473–479.

Chen LC, Kurisu W, Ljung BM, Goldman ES, Moore D 2nd, SmithHS (1992) Heterogeneity for allelic loss in human breast cancer. JNatl Cancer Inst 84:506–510.

Chen LC, Matsumura K, Deng G, Kurisu W, Ljung BM, LermanMI, Waldman FM, Smith HS (1994) Deletion of two separateregions on chromosome 3p in breast cancers. Cancer Res 54:3021–3024.

Druck T, Kastury K, Hadaczek Piotr, Podolski J, Toloczko A,Sikorski A, Ohta M, LaForgia S, Lasota J, McCue P, Lubinski J,Huebner K (1995) Loss of heterozygosity at the familial RCCt(3;8) locus in most clear cell renal carcinomas. Cancer Res55:5348–5353.

Eiriksdottir G, Bergthorsson JT, Sigurdsson H, Gudmundsson J,Skirnisdottir S, Egilsson V, Barkardottir RB, Incvarsson S (1995)Mapping of chromosome 3 alterations in human breast cancerusing microsatellite PCR markers: Correlation with clinical vari-ables. Int J Oncol 6:369–375.

Gemmill RM, Chumakov I, Scott P, Waggoner B, Rigault P, Cypser J,Chen Q, Weissenbach J, Gardiner K, Wang H, Pekarsky Y, Le GallI, Le Paslier D, Guillou S, Li E, Robinson L, Hahner L, Todd S,Cohen D, Drabkin HA (1995) A second-generation YAC contigmap of human chromosome 3. Nature 377:299–319.

Gnarra JR, Tory K, Weng Y, Schmidt L, Wei MH, Latif F, Liu S,Chen F, Duh FM, Lubensky I, Duan DR, Florence C, Pozzatti R,Walther MM, Bander NH, Grossman HB, Brauch H, Pomer S,Brooks JD, Isaacs WB, Lerman MI, Zbar B, Linehan WM (1994)Mutations of the VHL tumor suppressor gene in renal carcinoma.Nat Genet :85–90.

Gyapay G, Morissette J, Vignal A, Dib C, Fizames C, Millasseau P,Marc S, Bernardi G, Lathrop M, Weissenbach J (1994) Genethonhuman genetic linkage map. Nat Genet 7:246–339.

Harada Y, Katagiri T, Ito I, Akiyama F, Sakamoto G, Kasumi F, EmiM, Nakamura Y (1994) Genetic studies of 457 breast cancers:Clinicopathologic parameters compared with genetic alterations.Cancer 74:2281–2286.

Hibi K, Takahashi T, Yamakawa K, Ueda R, Sekido Y, Ariyoshi Y,Suyama M, Takagi H, Nakamura Y, Takahashi T (1992) Threedistinct regions involved in 3p deletion in human lung cancer.Oncogene 7:445–449.

Ito I, Yoshimoto M, Iwase T, Watanabe S, Katagiri T, Harada Y,Kasumi F, Yasuda S, Mitomi T, Emi M, Nakamura Y (1995)Association of genetic alterations on chromosome 17 and loss ofhormone receptors in breast cancer. Br J Cancer :438–441.

Kok K, van den Berg A, Veldhuis PM, van der Veen AY, Franke M,Schoenmakers EF, Hulsbeek MM, von der Hout AH, de Leij L,van de Ven W, Buys CH (1994) A homozygous deletion in a smallcell lung cancer cell line involving a 3p21 region with a markedinstability in yeast artificial chromosomes. Cancer Res 54:4183–4187.

Latif F, Tory K, Gnarra J, Yao M, Duh FM, Orcutt ML, StackhouseT, Kuzmin I, Modi W, Geil L, Schmidt L, Zhou F, Li H, Wei MH,Chen F, Glenn G, Choyke P, Walther MM, Weng Y, Duan DSR,Dean M, Glavac D, Richards FM, Crossey PA, Ferguson-SmithMA, le Paslier D, Chumakov I, Cohen D, Chinault C, Maher ER,Linehan WM, Zbar B, Lerman MI (1993) Identification of the vonHippel-Lindau disease tumor suppressor gene. Science 260:1317–1320.

273DELETION MAPPING OF Ch 3p IN BREAST CANCER

Lebeau J, Gerbault-Seureau M, Lemieux N, Apiou F, Calvo F,Berthon P, Goubin G, Dutrillaux B (1995) Loss of chromosome 3parm differentiating tumorigenic from non-tumorigenic cells de-rived from the same SV40-transformed human mammary epithe-lial cells. Int J Cancer 60:244–248.

Man S, Ellis IO, Sibbering M, Blamey RW, Brook JD (1996) Highlevels of allele loss at the FHIT and ATM genes in non-comedoductal carcinoma in situ and grade I tubular invasive breast cancers.Cancer Res 56:5484–5489.

Negrini M, Monaco C, Vorechovsky I, Ohta M, Druck T, Baffa R,Huebner K, Croce CM (1996) The FHIT gene at 3p14.2 isabnormal in breast carcinomas. Cancer Res 56:3173–3179.

Ohta M, Inoue H, Cotticelli MG, Kastury K, Baffa R, Palazzo J,Siprashvili Z, Mori M, McCue P, Druck T, Croce CM, Huebner K(1996) The FHIT gene, spanning the chromosome 3p14.2 fragilesite and renal carcinoma-associated t(3;8) breakpoint, is abnormalin digestive tract cancers. Cell 84:587–597.

Panagopoulos I, Pandis N, Thelin S, Petersson C, Mertens F, Borg A,Kristoffersson U, Mitelman F, Aman P (1996) The FHIT andPTPRG genes are deleted in benign proliferative breast diseaseassociated with familial breast cancer and cytogenetic rearrenge-ments of chromosome band 3p14. Cancer Res 56:4871–4875.

Pandis N, Jin Y, Limon J, Bardi G, Idvall I, Mandahl N, Mitelman F,Heim S (1993) Interstitial deletion of the short arm of chromosome3 as a primary chromosome abnormality in carcinomas of thebreast. Genes Chromosomes Cancer 6:151–155.

Rimessi P, Gualandi F, Morelli C, Trabanelli C, Wu Q, Possati L,Montesi M, Barrett JC, Barbanti-Brodano G (1994) Transfer ofhuman chromosome 3 to an ovarian carcinoma cell line identifiesthree regions on 3p involved in ovarian cancer. Oncogene 9:3467–3474.

Roz L, Wu CL, Porter S, Scully C, Speight P, Read A, Sloan P,Thakker N (1996) Allelic imbalance on chromosome 3p in oral

dysplastic lesions: An early event in oral carcinogenesis. CancerRes 56:1228–1231.

Sato T, Tanigami A, Yamakawa K, Akiyama F, Kasumi F, SakamotoG, Nakamura Y (1990) Allelotype of breast cancer: Cumulativeallele losses promote tumor progression in primary breast cancer.Cancer Res 50:7184–7189.

Sato T, Akiyama F, Sakamoto G, Kasumi F, Nakamura Y (1991)Accumulation of genetic alterations and progression of primarybreast cancer. Cancer Res 51:5794–5799.

Sozzi G, Veronese ML, Negrini M, Baffa R, Cotticelli MG, Inoue H,Tornielli S, Pilotti S, de Gregorio L, Pastorio U, Pierotti MA, OhtaM, Huebner K, Croce CM (1996) The FHIT gene at 3p14.2 isabnormal in lung cancer. Cell 85:17–26.

Takita K, Sato T, Miyagi M, Watatani M, Akiyama F, Sakamoto G,Kasumi F, Abe R, Nakamura Y (1992) Correlation of loss of alleleson the short arms of chromosomes 11 and 17 with metastasis ofprimary breast cancer to lymph nodes. Cancer Res 52:3914–3917.

Teixeira MR, Pandis N, Bardi G, Andersen JA, Heim S (1996)Karyotypic comparisons of multiple tumorous and macroscopicallynormal surrounding tissue samples from patients with breastcancer. Cancer Res 56:855–859.

van den Berg A, Hulsbeek MMF, de Jong D, Kok K, Veldhuis PMJF,Roche J, Buys CHCM (1996) Major role for a 3p21 region and lackof involvement of the t(3;8) breakpoint region in the developmentof renal cell carcinoma suggested by loss of heterozygosityanalysis. Genes Chromosomes Cancer 15:64–72.

Wu CL, Sloan P, Read AP, Harris R, Thakker N (1994) Deletionmapping on the short arm of chromosome 3 in squamous cellcarcinoma of the oral cavity. Cancer Res 54:6484–6488

Yamakawa K, Takahashi T, Horio Y, Murata Y, Takahashi E, Hibi K,Yokoyama S, Ueda R, Takahashi T, Nakamura Y (1993) Frequenthomozygous deletions in lung cancer cell lines detected by a DNAmarker located at 3p21.3-p22. Oncogene 8:327–330.

274 MATSUMOTO ET AL.