GENOMICS 43, 349–358 (1997)ARTICLE NO. GE974819

A 2-Mb YAC Contig and Physical Map Covering the Chromosome8q12 Breakpoint Cluster Region in Pleomorphic

Adenomas of the Salivary Glands

Koen Kas,*,1 Eva Roijer,†,1 Marianne Voz,* Eva Meyen,*Goran Stenman,† and Wim J. M. Van de Ven*,2

*Laboratory for Molecular Oncology, Center for Human Genetics, University of Leuven, and Flanders Interuniversity Institutefor Biotechnology, Herestraat 49, B-3000 Leuven, Belgium; and †Laboratory of Cancer Genetics,

Department of Pathology, Goteborg University, Sahlgrenska Hospital,S-413 45 Goteborg, Sweden

Received February 27, 1997; accepted May 20, 1997

glands. It is the most common type of salivary glandPleomorphic adenomas are benign epithelial tumors tumor and accounts for almost 50% of all neoplasms in

originating from the major and minor salivary glands. these organs (Waldron, 1991). Microscopically, pleo-Extensive cytogenetic studies have demonstrated that morphic adenomas show a marked histological diver-they frequently show chromosome abnormalities in- sity with epithelial, myoepithelial, and mesenchymalvolving chromosome 8, with consistent breakpoints at components in a variety of patterns. More than half of8q12. In previous studies, we have shown that these

the tumors reveal recurrent chromosome abnormali-breakpoints are located in a 9-cM interval between MOS/ties. Rearrangements of chromosome 8 are most fre-D8S285 and D8S260. Here, we describe directional chro-quent (over 50%), usually involving band 8q12, withmosome walking studies starting from D8S260 as wellthe most common aberration being a t(3;8)(p21;q12).as D8S285. Using the CEPH and ICRF YAC libraries,Other recurrent translocations involve the chromo-these studies resulted in the construction of two non-

overlapping YAC contigs of about 2 and 5 Mb, respec- some region 12q13–q15 (Sandros et al., 1990; Buller-tively. Initial fluorescence in situ hybridization (FISH) diek et al., 1993). With respect to the latter cases, theanalysis suggested that the majority of 8q12 breakpoints high-mobility group protein gene HMGIC was recentlyclustered within the 2-Mb contig, which was mapped to identified as the gene consistently rearranged by thethe centromeric part of chromosome band 8q12. This chromosome 12q13–q15 aberrations (Schoenmakers etcontig has at least double coverage and consists of 34 al., 1995).overlapping YAC clones. The localization of the YACs In an effort to clone the gene(s) affected by the chro-was confirmed by FISH analysis. On the basis of map-

mosome 8q12 aberrations molecularly, we chose direc-ping data of landmarks with an average spacing of 65tional chromosome walking as a structural approachkb as well as restriction enzyme analysis, a long-rangeto define the genomic region encompassing the break-physical map was established for the chromosome re-points. In previous physical mapping studies, we local-gion spanned by the 2-Mb contig. The relative positionsized the 8q12 breakpoints to a 9-cM interval betweenof various known genes and expressed sequence tags

within this contig were also determined. Subsequent D8S285 and D8S260. Therefore, these two polymorphicFISH analyses of pleomorphic adenomas using YACs as markers were used as starting points for chromosomewell as cosmids revealed that all but two of the 8q12 walking.breakpoints in the primary tumors tested mapped The present study describes the construction of twowithin a 300-kb interval between the MOS proto-onco- nonoverlapping YAC contigs covering about 75% of hu-gene and STS EM156. The target gene affected by the man chromosome band 8q12, which spans approxi-chromosome aberrations mapping within this interval

mately 9 Mb of genomic DNA. On the basis of STS-was recently shown to be the PLAG1 gene, which en-content mapping and restriction enzyme analysis, wecodes a novel zinc finger protein. q 1997 Academic Presshave established long-range physical maps of these re-gions. The relative positions of various known and

INTRODUCTION novel genes (CYP7, FAN, PBP1, RAB2, PENK, MOS,The pleomorphic adenoma is a benign epithelial tu- LYN ) as well as expressed sequence tags within these

mor originating from the major and minor salivary regions were also determined. Furthermore, fluores-cence in situ hybridization (FISH) analysis was per-

1 Both authors contributed equally to this work. formed to determine the relative position of the 8q122 To whom correspondence should be addressed. Telephone: 32-16-breakpoints within these contigs. The majority of the345987. Fax: 32-16-346073. E-mail: wim.vandeven@med.kuleuven.

ac.be. breakpoints were shown to cluster within a relatively

3490888-7543/97 $25.00

Copyright q 1997 by Academic PressAll rights of reproduction in any form reserved.

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

KAS ET AL.350

TABLE 1

Pleomorphic Adenomas with Chromosome 8q12 Rearrangements

Case No. Chromsome 8q12 rearrangements Other abnormalities present

CG580 t(8;15)(q12;q14) NoCG588 t(3;8)(p21;q12) NoCG644 t(3;8)(p21;q12) NoCG650 t(5;8)(p13;q12) YesCG666 der(8;11)t(8;11)(p23;q13)inv(8)(p22q12) YesCG682 der(3)del(3)(p23-24)del(3)(p21.3p14.1), der(8)dir ins(8;3)(q12;p21.3p14.1) NoCG714 ins(8;7)(q11-12;p21p14)a YesCG752 t(3;8)(p21;q12) NoCG752 t(3;8)(p21;q12) NoCG787 der(8)del(8)(p11)t(8;13)(q12;q13)del(13)(q14), der(13)t(8;13)(q12;q13) YesT9587 t(3;8)(p21;q12) No

a The interpretation of the karyotype based on FISH is as follows: der(7)ins(7;8)(p22;q11.1q11.2) (wcp 7/, wcp 8-, Y943G4/),der(8)ins(8;7)(q11.2;p21p14) (wcp7/, wcp8/, Y943G4/).

out as previously described (Roijer et al., 1996). Alpha-satellitesmall DNA interval in the centromeric contig. Theseprobes D8Z2 were obtained from Oncor. Slides were examined in aphysical mapping studies proved to be instrumental inZeiss Axiophot epifluorescence microscope using the appropriate fil-the discovery of the PLAG1 gene as the target gene ter combinations. Fluorescence signals were digitalized, enhanced,

affected by the chromosome 8q12 aberrations in pleo- and analyzed using the ProbeMaster FISH image analysis system(Perceptive Scientific Instruments, Houston, TX).morphic adenomas (Kas et al., 1997).

Pulsed-field gel electrophoresis and Southern blot analysis.Pulsed-field gel electrophoresis and Southern blot analysis were per-MATERIALS AND METHODSformed as described by Schoenmakers et al. (1994). Agarose plugscontaining high-molecular-weight yeast / YAC DNA (equivalent to

Tumor material and cytogenetic analysis. Primary pleomorphic 1 1 109 cells ml01) were prepared as described (Schoenmakers et al.,adenomas were obtained from patients at the time of surgery. Pri- 1994). Plugs were thoroughly dialyzed against four changes of 25 mlmary cultures and chromosome preparations were established and T10E1 (pH 8.0) followed by two changes of 0.5 ml 11 restriction bufferanalyzed as described previously (Roijer et al., 1996). The chromo- before they were subjected to either pulsed-field restriction enzymesome 8 rearrangements of the primary tumors used in this study are mapping or YAC-end rescue.listed in Table 1.

PCR analysis. PCR amplification was carried out using a Phar-YAC library screening. YAC clones were isolated from the CEPH macia LKB-Gene ATAQ Controller or a Perkin–Elmer 9600 PCR

mark 1 (Albertsen et al., 1990) and mark 3 (Chumakov et al., 1992) apparatus in final volumes of 50 and 25 ml, respectively, containingYAC libraries, made available to us by the Centre d’Etude du Poly- 10 mM Tris–HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 0.01% gela-morphisme Humain (CEPH). YACs were isolated using a combina- tine, 0.2 mM dNTPs, 20 pmol of each amplimer, 1.25 units of Ampli-tion of PCR-based screening and colony hybridization analysis Taq (Perkin–Elmer Cetus), and 100 ng (for superpools) or 20 ng(Green and Olson, 1990; Schoenmakers et al., 1994). Contaminating (for pools) of DNA. After initial denaturation for 5 min at 947C, 30Candida parapsylosis, which was sometimes encountered, was eradi- amplification cycles were performed, each consisting of denaturationcated by adding terbinafin to the growth medium (final concentration for 1 min at 947C, annealing for 1 min at the appropriate temperatureof 25 mg/ml). The isolated YAC clones were characterized by STS-

(see Table 3), and extension for 1 min at 727C, and with a finalcontent mapping, contour-clamped homogeneous electric fieldextension at 727C for 10 min. Results were evaluated by analysis of(CHEF) electrophoresis (Chu et al., 1986), restriction mapping, hy-10 ml of the reaction product on 2% agarose gels.bridization, and FISH analysis.

Generation of STSs from YAC inserts and insert ends. YAC-endCosmid library screening. Cosmid clones were isolated from anrescue was performed using a vectorette-PCR procedure in combina-arrayed human chromosome 8-specific cosmid library (Wood et al.,tion with direct solid-phase DNA sequencing, as described (Geurts1992) obtained from Los Alamos National Laboratory (LANL). Cos-et al., 1994). STSs specific for YAC inserts were generated from inter-mid DNA was extracted using standard techniques involving purifi-Alu PCR products, isolated using published oligonucleotides TC65cation over Qiagen tips (Qiagen).or 517 (Nelson et al., 1989) to which SalI-tails were added to facilitate

DNA preparation. Extraction of genomic DNA was performed as cloning. After sequence analysis, primer pairs were developed usingdescribed previously (Schoenmakers et al., 1994). DNA from YACs, the OLIGO computer algorithm (Rychlik and Rhoads, 1989). Theycosmids, PCR products, and oligonucleotides was labeled using a were tested on human genomic DNA, according to procedures de-variety of techniques. YAC DNA (100 ng) was amplified by inter-Alu scribed above.PCR (P1: CTGCACTCCAGCCTGGG; P2: TCCCAAAGTGCTGGG-

Nucleotide sequence analysis and oligonucleotides. Nucleotide se-ATTACAG). After initial denaturation for 5 min at 947C, 30 amplifi-quences were determined according to the dideoxy chain terminationcation cycles were performed, each consisting of denaturation for 1method using the T7 polymerase sequencing kit of Pharmacia/LKBmin at 947C, annealing for 30 sec at 377C, and extension for 6 minor the dsDNA Cycle Sequencing System (GIBCO/BRL). Sequencingat 727C, and with a final extension at 727C for 10 min. Amplifiedresults were analyzed using an ALF DNA sequencer (PharmaciaDNA was purified with QIAQuick PCR Purification kit (Qiagen).Biotech) on standard 30-cm, 6% HydrolinkR, Long Range gels (ATFor filter hybridizations, probes were radiolabeled with [a-32P]dCTPBiochem). Sequence analysis utilized Lasergene (DNASTAR) andusing random hexamers (Feinberg and Vogelstein, 1984). In the caseBLAST and BEAUTY searches (NCBI; Altschul et al., 1990). Allof PCR products smaller than 200 bp, a similar protocol was applied,oligonucleotides were purchased from Pharmacia Biotech.but specific oligonucleotides were used to prime the labeling reac-

tions. Oligonucleotides were labelled using [g-32P]ATP. 5*-RACE analysis. For 5*-RACE experiments, the MarathoncDNA Amplification kit (Clontech) was used according the manufac-FISH analysis. Slides for FISH were prepared from cells stored

in fixative at 0207C. Hybridization and probe detection were carried turer’s instructions with minor modifications. The 5*-untranslated

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

2-Mb YAC MAP COVERING THE 8q12 BREAKPOINT CLUSTER REGION 351

end of the normal PLAG1 transcript as well as the chimeric tran- for specific amplification of DNA, establishing STSsscripts were isolated by 5*-RACE. First-strand placenta or adenoma (Table 3). Their location to 8q12–qter was determinedcDNA, respectively, was synthesized from 5 mg total RNA using the by CASH (chromosome assignment using somatic cellMV2 primer (5*-CTG CAC TTG ACC CAC CCC TTG GAT-3 * ) located

hybrids) as well as by FISH after corresponding YACin exon 5 of PLAG1. The ds cDNA was ligated to the adaptor andamplified using the anchor primer AP1 and the MV5 primer (5*-CAG and/or cosmid clones were isolated. With the sequen-GAG AAT GAG TAG CCA TGT GC-3 * ) also located in exon 5. A tially selected and evaluated primer sets, screening ofsecond round of PCR was performed using the anchor primer and the YAC and cosmid libraries was performed to isolatethe MV6 primer (5*-TGC ACT TGT AGG GCC TCT CTC CTG-3* )

the building blocks for contig assembly. Hence, we es-located in exon 4. The final PCR products were purified from agarosetablished an 8q12 centromeric YAC contig consistinggels and cloned into the pCRII vector (Invitrogen).of 34 overlapping YAC clones, covering approximately2 Mb of genomic DNA (Fig. 1). The inserts of the 34RESULTSYACs are between 160 and 1660 kb long. Characteris-tics of the YACs that were used to build this contig areAccording to radiation hybrid mapping data, bandgiven in Table 2. The YAC contig was constructed in8q12 is flanked proximally by the markers D8S165/parallel with contigs WC8.7 and WC-157 from theD8S285 and distally by D8S260 (Sapru et al., 1994).Whitehead Institute/MIT Center for Genome ResearchThe estimated genetic distance between these markersavailable via http://www-genome.wi.mit.edu/cgi-bin/is 9 cM (Gyapay et al., 1994). Using primer sets forcontig/lookup_contig.these microsatellite markers, we selected sets of corre-

sponding YAC clones. These YACs were used for ini-Physical Mapping of Polymorphic Markers, ESTs,tiating the walk toward the 8q12 translocation break-

and Genes in the Centromeric Contigpoints, meanwhile generating a long-range contigbased upon STS-content mapping. Hence we started All markers and genes that were or became publiclyfrom D8S285 to construct a 2-Mb YAC contig covering available during our work were STS-content mapped inthe centromeric part of band 8q12 and from D8S260 to our contig, and those found positive were subsequentlygenerate a 5-Mb contig covering the telomeric part of sublocalized by (primer) hybridization to YAC South-8q12 (for details, see below). Preliminary evaluation of ern blots. The markers that were found to reside withinthe YAC contigs by fluorescence in situ hybridization the centromeric contig were D8S1069 (WI-536),experiments revealed that the majority of the break- AFMB055WG9, D8S125, D8S108, D8S1661 (WI-5459),points were localized within the 2-Mb YAC contig. In D8S166, D8S96, D8S1816, D8S285, D8S1516 (WI-two primary adenomas, the breakpoints mapped to a 3862), D8S1828, and D8S165. Furthermore, in addition1-Mb DNA interval within this contig (Roijer et al., to the MOS proto-oncogene (Testa et al., 1988; Sandros1996). Hence, research was focused especially on the et al., 1990), two other genes could be positioned in2-Mb YAC contig in the centromeric part of 8q12. The our centromeric contig, namely the YES-related proto-two YAC contigs are not overlapping since they could oncogene LYN (Corey and Shapiro, 1994) and the pre-not be linked to each other. proenkephalin gene PENK (Tomfohrde et al., 1992).

Finally, our contig contains three expressed sequenceAssembly of a YAC Contig Starting from D8S285 tags, namely WI-7754 (D8S1930), corresponding to

LYN, IB2045, not resembling any currently knownIn previous studies (Roijer et al., 1996), we have de- gene, and EST D59273. The latter one was encounteredscribed the isolation of two overlapping YACs containing performing BLAST searches with the newly obtainedD8S285, namely 935E9 and 166F4. These YACs, to- STSs. The right end of YAC 143D5 (CH283) revealedgether with a MOS containing YAC (900g10157), were identity with this EST and led to the discovery ofused as a starting point for a chromosome walking pro- PLAG1, the gene affected in pleomorphic adenomasject to define the 8q12 centromeric region. Initially, chro- with t(3;8)(p21;q12) (Kas et al., 1997). The genes encod-mosome walking was performed bidirectionally. This al- ing the k opioid receptor (OPRK1) (Yasuda et al., 1994)lowed us to hook up our 8q12 centromeric contig to an and interleukin-7 (IL7) (Brunton and Lupton, 1990)existing chromosome 8q11 YAC contig (http://www- are not contained in our YAC contigs and map centro-genome.wi.mit.edu/cgi-bin/contig/lookup_contig). Both meric and telomeric, respectively.contigs have the marker D8S593 in common. In the sub-sequent PCR-based screening of the human CEPH YAC Restriction Map and Putative CpG Islands:libraries with D8S108 and D8S166, we isolated Construction of a Rare-Cutter Physical Map of themegaYAC 946B7, which together with megaYAC 935E9 2-Mb YAC Contig in the Centromeric Part of 8q12formed the backbone of our centromeric contig.

In the bidirectional and subsequent unidirectional Southern blots of total yeast and YAC DNA, digestedto completion with the rare-cutter enzymes BssHII,chromosome walking steps, the following general pro-

cedures were used. First, rescuing and sequencing the KpnI, MluI, NotI, PvuI, SalI, and SfiI and separatedon CHEF gels, were hybridized sequentially with (i)ends of YAC clones resulted in DNA markers character-

izing their left and right sides (Table 2). Based on se- the STS used for the initial screening of the YAC inquestion, (ii) pYAC4 right arm sequences, (iii) pYAC4quence data of the ends of the YAC inserts, as well as

of inter-Alu PCR products, primer sets were developed left arm sequences, and (iv) a human Alu-repeat probe

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

KAS ET AL.352

TABLE 2

Analysis of 26 YAC Clones in the Centromeric Contig, Used for YAC End Rescue

CEPH code Size (kb) Landmark left Accession No. Landmark right Accession No. Chimeric

518G7 ND CH131 NA ND NA Y (left)518G12 ND CH132 U84974 ND NA ND770B2 1410 CH281 U84975 CH286 U84989 ND878F2 800 CH282 U84976 CH287 U84990 ND255H10 390 CH273 U84977 ND NA ND531D4 430 CH274 U84978 CH277 U84991 ND15H11 430 ND NA ND NA ND291D12 410 CH29 NA CH34 U84992 Y (left)47E9 370 ND NA ND NA ND26F6 460 ND NA ND NA ND916F12 1230 CH122 U84979 CH128 U84993 ND388D6 430 CH69 U84980 CH100 U84994 ND297E11 350 CH280 U78583 CH285 U84995 ND253H7 160 CH279 U84981 CH284 U84996 ND143D5 260 CH278 U84982 CH283 U84997 ND391H2 620 ND NA ND NA Y392A6 600 ND NA ND NA Y84E12 500 CH289 U84983 CH290 U84998 ND166F4 700 CH67 U84984 CH73 U84999 ND164H5 375 END2 U78594 END6 U85000 ND206F11 ND END3 U84985 END7 NA Y (right)253C7 620 END4 U84986 END8 U85001 ND295A4 ND CH68 NA CH74 U85002 Y (left)917E11 1380 END1 U84987 END5 U85003 ND946B7 1370 CH123 U84988 CH129 U78582 ND935E9 1660 CH33 U78584 CH37 U85004 ND

Note. YAC clones were isolated from CEPH YAC libraries as described under Materials and Methods. ND, not detected by methods used(FISH, YAC end rescue, restriction enzyme mapping). GenBank accession numbers are given. ND, not determined. NA, not applicable(accession numbers were not obtained for chimeric YAC ends).

(BLUR-8). The long-range restriction map that was ob- et al., 1992) using markers contained within YAC166F4. Cosmid clones containing MOS, EM156, andtained in this way was completed by probing with PCR-

isolated STSs/YAC end probes. Restriction maps of in- CH129 were isolated and mapped by FISH (Fig. 1). Thebreakpoints in all nine cases were localized betweendividual YAC clones were aligned, and a consensus re-

striction map was established. The region was searched cosmids CEM1 (MOS ) and CEM23 (EM156) (Figs. 2Band 2C). In adenoma CG666, for example, CEM1 wasfor CpG islands on the basis of the colocalization of

sites for rare-cutter enzymes (Fig. 1). transposed to 8p due to a pericentric inversion of thedicentric marker chromosome (Fig. 2B), while CEM23remained in its normal position at 8q12 (Fig. 2C). InFISH Mapping of 8q12 Breakpoints in Pleomorphicadenoma CG682, which has an ins(8;3)(q12;p21.3p14.1),AdenomasCEM23 spanned the insertion breakpoint (Roijer et al.,

In previous studies, we have mapped the t(3;8)- 1997). This cosmid contig was used in studies leading(p21;q12) breakpoint within a 1-Mb region flanked by to the identification of the Pleomorphic Adenoma Gene,MOS proximally and by D8S166 distally. One YAC PLAG1, the target gene on chromosome 8q12 consis-within this region (166F4) was shown to span the t(3;8) tently affected by the 3;8 rearrangements (Kas et al.,breakpoint in two primary pleomorphic adenomas 1997). In an adenoma with variant translocations, i.e.,(Roijer et al., 1996). We have now extended these previ- CG580, which has a variant t(8;15)(q12;q14), 5*-RACEous observations by analysis of additional tumors in- analysis revealed 38 ectopic nucleotides in front of exoncluding cases with variant translocations (Table 1). 3 of PLAG1 (Kas et al., 1997). This sequence mostNine of eleven primary adenomas with 8q12 abnormal- closely resembled (95% identity) an EST (W17221) thatities had breakpoints mapping within YAC 166F4 (Fig. overlapped with the gene for a-tropomyosin (TPM1)2A). Tumors with 8q12 rearrangements other than the but has the reversed transcriptional orientation. Sinceclassical t(3;8) also had breakpoints within the same TPM1 had been mapped to human chromosome bandYAC, demonstrating that this is the major breakpoint 15q22 by FISH (Eyre et al., 1995) and since the variantcluster region in pleomorphic adenomas with 8q12 ab- translocation involved chromosome band 15q14, we in-normalities. vestigated whether the TPM1 gene was affected by this

To fine map this region, we isolated cosmids from an translocation. Using a TPM1-specific primer set (Eyrearrayed human chromosome 8-specific cosmid library et al., 1995), we isolated two YACs containing this gene.

Both YACs 751C10 and 821H6 were found by FISH to(obtained from Los Alamos National Laboratory; Wood

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

2-Mb YAC MAP COVERING THE 8q12 BREAKPOINT CLUSTER REGION 353

TABLE 3 point was found to reside within the XRCC7 containingYAC 943G4 (Fig. 2D). In the latter case, which had,8q12 STS Primer Sequences, Annealingjudging from cytogenetics, an ins(8;7)(q11-12;p21p14),Temperatures and Expected PCR Product SizesFISH analysis revealed signals on both the der(7) and

Amplimers of newly isolated STSs in the the der(8) chromosomes, demonstrating that this is notcentromeric contig a simple 8;7 insertion but a complex rearrangement

Product with at least two breakpoints on 8q, one of which mapsSTS name Nucleotide sequence 5*–3* size (bp) Ta (7C)within YAC 943G4 and the other proximal to this YAC.5*-RACE and RT-PCR analysis did not reveal any ab-CH33 CCTTTGGCTGGGGTTTATA

GGCCTATGAAGCAAGAGAG 168 60 normalities in PLAG1, suggesting that it is not in-CH34 GTACCCAGAAGGCAAGTAA volved in these cases.

GTGAAAAGGCAGAAATTAG 145 60CH37 TTGCATGAGAATGGAAATG

GGCGTTACTTTCCTTTTGT 176 58 Assembly of a YAC Contig between D8S260 andCH69 AGTGCTTACAATAGGGTGAG D8S507 and Gene Assignments

CCATCCAGAAAGACCATAAT 336 60CH122 TTTGTCTTTGATTTTTATGG Two megaYACs corresponding to D8S260 were used

TGACCAACATACTGCCTAGT 250 57 to start the chromosome walk from the telomeric side ofCH129 CTGAATCCCAGAACAATATAband 8q12. The YAC-insert ends from these two clonesAGGGTAAGTATGTCCTTTAA 110 60

CH273 ATAATGTTGAGACTTTGAGA (783H12 and 814A6) were isolated and used for STS-AAATGTTTATCCTAATTGTA 157 58 content mapping. The most telomeric STS (CH31) is

CH274 CAGGTGAGTGGATGGTGTAA present in two YACs also containing D8S510, andCAAGGGGAGACCAAATCATC 241 60

hence links the 8q12 YAC contig to an 8q13-specificCH277 AATGGCTATGAGGTTGTTTTYAC contig (Doerflinger et al., 1995). When we startedCACATCCTTTCATTTTAGCA 122 58

CH280 GGGCTGATGTTCCATTAACT our physical mapping effort, only one polymorphicGCTTCAACACCAAAAATGCT 163 58 marker, D8S507, was described between D8S285 and

EM76 CTGGGAAGAGATCAAATTC D8S260 (Gyapay et al., 1994). Initially, we were unableTAAAGAGACAGCACCACAAA 220 60

to isolate YACs containing D8S507. Instead, we usedEM156 AGTAGCAGCAGCAACAGTCAthe corresponding amplimers to isolate a cosmid con-TGCGCTATTCAGAGAAGATG 160 60

EM216 CAGTCAGTTCCAGAGGTCATTT taining D8S507 (cosmid CEM3). An insert-end of thisTAGGGAGGGCTTTAATAGTGTT 255 55 cosmid clone (EM73) was then used to isolate the corre-

END1 GCTCACTTCACTCCTACCC sponding megaYACs (872E1, 882D9, and 957C4). It isCAACCAACCACTAAAAACG 161 60unclear why these YACs initially were not recognizedEND2 GTGATTTTACAGCATTTT

TGTAATTTTCAACCAGAAG 91 50 upon screening of the YAC librarary. This might beTC65.2 TACAAACCGGGAGAAAACAG due either to a less efficient PCR amplification of

TTACAGCATTTCCGATTTTG 232 58 D8S507 than of EM73 or to an underrepresentation ofCH31 TCACAGATAATACAGGAT the D8S507 YAC-containing yeast cells in the primaryGATCACTGATGATACTAGG 220 58

pools of the specific library we screened. The availabil-CH32 ATTTGCCTCAGTGTTGCAGATTTTCAGGAGGTCAGGGA 245 62 ity of two new polymorphic markers, D8S1505 and

CH35 CAAAATGACTTATGCTGAA D8S1515 (WI-6879, part of the FAN gene), allowed usTCTATACAGGGCATTGTGA 165 60 to link the two separate megaYAC contigs around

EM73 AAAGCAAGACCCTGTAAAGCD8S260 and D8S507. The composite contig consists ofCTTGGGCTCTATTTTGTGAA 351 6023 CEPH megaYACs and covers approximately 5 Mbof genomic DNA (Fig. 3). The polymorphic markers thatwere tested are publicly available and were found tomap to the der(8) in the t(8;15)(q12;q14) in agreement

with the reported position of TPM1 (Eyre et al., 1995) reside in the telomeric YAC contig: D8S1151 (WI-1155), D8S260, D8S1075 (WI-943), D8S1505, D8S1515and excluding it from involvement in the variant trans-

location (data not shown). Of course, we cannot rule (WI-6879), D8S1723, D8S507, and D8S1957 (WI-9507).The YAC contig was constructed in parallel with con-out the possibility that submicroscopic rearrangements

affect TPM1. On the other hand, the 38 nucleotides tigs WC8.8 and WC-727 from the Whitehead Institute/MIT Center for Genome Research available via http://could be derived from a gene at chromosome band

15q14 showing strong homology with TPM1. All our www-genome.wi.mit.edu/cgi-bin/contig/lookup_contig.Four genes could be mapped in this contig withefforts to clone the corresponding gene failed so far.

Two of the eleven adenomas tested, CG650 and primer sets that we developed from publicly availablesequence data or by updating existing STS/EST dataCG714, had breakpoints clearly proximal to MOS. To

map these breakpoints further, we isolated two addi- using computer-assisted library screening. These genesinclude CYP7, which encodes cholesterol 7a-hydroxy-tional sets of YACs corresponding to CH37, D8S593

and the recently identified XRCC7 gene (Blunt et al., lase (previously assigned to 8q11–q12 using bothmouse–human somatic cell hybrids and FISH; Cohen1995), respectively. In CG650, the breakpoint was

found to reside between ICRF YAC 900g10157 and et al., 1992), RAB-2, which encodes a RAS-related pro-tein (Tachibana et al., 1988), FAN, which encodes aYAC 898G12 (Figs. 1 and 2E), and in CG714 the break-

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

KAS ET AL.354

FIG. 1. YAC contig and STS marker map of the proximal part of chromosomal band 8q12. The upper line indicates the chromosome,with the telomere to the left and the centromere to the right. DNA markers and genes are ordered at the top. Genes are underlined, andpreviously known markers are indicated in boldface. YAC clones are represented as horizontal lines together with their addresses, and theones used for YAC end rescue are described in Table 2. YACs 884B9, 762G8, and 766C7 are not considered part of the contig, but theyconnect with the 8q11 contig. STS locations are indicated by dashed lines. The length of the solid lines indicate their approximate size.YACs 391H2 and 392A6 most likely contain chimeric inserts, and therefore only the parts corresponding to contig markers are shown. YACend clones used as STSs are indicated by arrows. Below the contig the consensus restriction maps for BssHII (B), KpnI (K), MluI (M), NotI(N), PvuI (P), SalI (S) and SfiI (Sf) are indicated. Arrowheads indicate putative CpG islands, defined as the colocalization of sites for (a)K, M, N; (b) K, M, P, S, Sf; (c) K, M, P, Sf; (d) K, N, S; (e) K, N, S; (f) B, N, S; (g) B, K, N, Sf. The DNA region containing the PLAG1 geneis indicated by a shaded box. The right end of YAC 143D5 (CH283) is part of the PLAG1 cDNA. The relative positions of the chromosome8 breakpoints in seven primary adenomas are indicated with asterisks on a schematic drawing of the PLAG1 gene. Shaded areas withinthe boxes represent the coding region of PLAG1 (Kas et al., 1997).

novel WD-repeat protein that couples the p55 TNF- end rescue and/or restriction enzyme mapping and/orFISH) was slightly lower (23%) than the 30–40% ex-receptor to neutral sphingomyelinase (Adam-Klages et

al., 1996), and PBP1, which encodes a novel scaffold pected (Van Ommen, 1993). This is the most detailedavailable physical map of the 8q12 region. The mapprotein (GenBank Accession No. U83463).was constructed in parallel to the generation of a low-resolution map of the human genome and to the physi-DISCUSSIONcal mapping effort of the Whitehead Institute/MIT Cen-

In this report, we present a detailed physical map of ter for Genome Research. It was compared to dataa 2-Mb DNA region representing the centromeric part available electronically via the Internet. The order ofof chromosome band 8q12. The map, which was vali- the markers on the physical map is in agreement withdated by FISH analysis, is covered by the DNA inserts the 1996 Genethon human genetic linkage map (Dib etof 34 overlapping YAC clones. It is tagged by 31 mark- al., 1996). In total, 31 markers/genes were ordered oners with an average spacing of 65 kb. The frequency of the contig as follows: D8S1069–CH274–TC65.2–

AFMB055WG9 – CH277 – CH34 – D8S125 – D8S108 –chimaerism in the set of YACs studied in detail (YAC

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

2-Mb YAC MAP COVERING THE 8q12 BREAKPOINT CLUSTER REGION 355

FIG. 2. FISH mapping of 8q12 translocation breakpoints in pleomorphic adenomas. (A) YAC 166F4 spans the t(3;8)(p21;q12) breakpointin adenoma CG644. Hybridization signals (green) are found on both the der(8) and the der(3), as well as on the normal chromosome 8(red signals represent alpha satellite probes for chromosomes 3 and 8). (B and C) Mapping of the 8q12 breakpoint between cosmids CEM1and CEM23 in adenoma CG666. Cohybridization of cosmids CEM1 (B) and CEM23 (C) (green signals) with alpha satellite probes forchromosomes 8 and 11 (red signals) reveal a breakpoint between the two cosmids. Due to a pericentric inversion of the der(8;11), withbreakpoints at 8q12 and 8p22, CEM1 (B) is transposed to 8p. The signal from cosmid CEM23 remains in its normal position at 8q12 (C).The signal from the normal chromosome 8 is missing in this metaphase. (D) Identification of a more proximal 8q breakpoint in an adenoma(CG714) with an ins(8;7)(q11–q12;p21p14). Cohybridization of YAC 943G4 (green signal) with alpha satellite probes for chromosomes 7and 8 (red signals) results in hybridization signals on the normal chromosome 8, the der(8), and the der(7), indicating that YAC 943G4spans the insertion breakpoint in this case. (E) Mapping of the 8q12 breakpoint in an adenoma (CG650) with a t(5;8). Cohybridization ofYAC 900g10157 (green signal) with alpha satellite probes for chromosomes 5 and 8 (this probe also recognizes chromosomes 1 and 19; redsignals) revealed a breakpoint proximal to YAC 900g10157 since this YAC was translocated to the der(5). Chromosomes are counterstainedin blue with DAPI.

AID GENO 4819 / 6r3e$$4819 07-01-97 06:45:32 gnmal

KAS ET AL.356

FIG. 3. YAC contig and STS marker map of the distal part of chromosome band 8q12. The upper line indicates the chromosome, withthe telomere to the left and the centromere to the right. DNA markers and genes are ordered at the top. Genes are underlined, andpreviously known markers are indicated in boldface. YAC clones and cosmid CEM3 are represented as lines together with their names.YAC end clones used as STSs are represented by arrows.

EM76 – D8S1661 – CH122 – D8S96 – D8S166 – CH69 – 1) affected in adenomas with t(3;8)(p21;q12) transloca-tions (Kas et al., 1997). The translocation results inCH33 – D8S1816 – PENK – CH280 – CH129 – EM156 –

END2 – CH283 – D8S285 – MOS – D8S1516 – LYN – promoter swapping between PLAG1 and the constitu-tively expressed gene for b-catenin (CTNNB1), a pro-D8S1828–IB2045–D8S165–CH37–W1086. A restriction

map generated with the enzymes BssHII, KpnI, MluI, tein interface functioning in the WG/WNT signalingpathway, and specification of cell fate during em-NotI, SalI, SfiI, and SwaI revealed at least seven puta-

tive CpG islands. Three of these islands correspond to bryogenesis. Fusions occur in the 5*-noncoding regionsof both genes, exchanging regulatory control elementspreviously known genes (PENK, MOS, and LYN ). Is-

land ‘‘d’’ in Fig. 1 corresponds to the recently identified while preserving the coding sequences. Due to thet(3;8)(p21;q12), PLAG1 is activated and expression lev-gene PLAG1 (Kas et al., 1997). The others might repre-

sent the 5* end of previously unidentified genes. Our els of CTNNB1 are reduced. Activation of PLAG1 wasalso observed in an adenoma (CG580) with a variant2-Mb map integrates cytogenetic, genetic, and physical

data from chromosome band 8q12. The telomeric YAC translocation t(8;15)(q12;q14). PLAG1 activation dueto promoter swapping seems therefore to be a crucialcontig, which was established between D8S260 and

D8S507, consists of 23 CEPH megaYACs and covers event in salivary gland tumorigenesis.Two of the eleven tumors in the present series hadapproximately 5 Mb. Both contigs represent about 75%

of human chromosome band 8q12, which extends over breakpoints outside the 300-kb breakpoint region. ByFISH we could demonstrate that the breakpoints inapproximately 9 Mb of DNA.

The 8q12 band seems to be rather gene-poor; so far both cases were clearly proximal to MOS. In one casethe breakpoint was mapped between MOS and YAConly 8 genes have been shown to reside in the two

contigs (RAB2, FAN, PBP, CYP7, PENK, MOS, LYN, 898G12, and in the other case it was found to residewithin YAC 943G4. The latter YAC maps at least 2and PLAG1). This is consistent with the view that

Giemsa-positive bands are chromosome regions rather Mb centromeric to MOS. Since the two breakpoints areclearly different and also separate from those locatedlow in G/C content and poor in genes (Craig and Bick-

more, 1993). Chromosome band 8q12 is partially con- in the 300-kb breakpoint region, it is unlikely that theyaffect the same gene. The possibility remains that theseserved in synteny with mouse chromosome 4 (Mock et

al., 1996). Comparative gene mapping, however, did cases have additional rearrangements that may haveescaped detection with the DNA probes used and thatnot reveal any mouse gene or marker not present in

our physical map. they in fact all have rearrangements of the same genein the breakpoint region. Preliminary data of 5*-RACEFISH analyses using YACs and cosmids revealed

that the majority of breakpoints in pleomorphic adeno- and RT-PCR analyses, however, indicate that PLAG1is not involved in these cases.mas clustered within a 300-kb region, both in cases

with the recurrent t(3;8)(p21;q12) and in those car- Recently, we have obtained molecular evidence foran additional breakpoint region distal to our telomericrying 8q12 translocations and having other transloca-

tion partners. A cosmid from this region was also contig (Roijer et al., in preparation). Based on FISHusing YACs derived from the 8q13–q21 region, we haveshown to span the breakpoint in an adenoma with an

ins(8;3) (Roijer et al., 1997). This region was therefore estimated that these breakpoints are about 15 to 20cM telomeric to the 8q12 breakpoint cluster region. Cy-likely to harbor the putative pleomorphic adenoma

gene. BLAST searches of the STSs mapping within the togenetically, these breakpoints are very difficult todistinguish from the classical 8q12 breakpoints. This300-kb region revealed sequence identity between

CH283 and a publicly available EST (D59273). This led is also true for the breakpoints located centromeric toMOS. Collectively, our findings thus suggest that thereto the discovery of a novel, developmentally regulated

zinc-finger gene, PLAG1 (Pleomorphic Adenoma Gene are several genes in the proximal part of 8q that are

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

2-Mb YAC MAP COVERING THE 8q12 BREAKPOINT CLUSTER REGION 357

Chu, G., Vollrath, D., and Davis, R. W. (1986). Separation of largeaffected by chromosome rearrangements in pleomor-DNA molecules by contour-clamped homogeneous electric fields.phic adenomas.Science 234: 1582–1585.In addition to pleomorphic adenomas there are a few

Chumakov, I., Rigault, P., Guillou, S., Ougen, P., Billaut, A., Guas-other types of solid tumors with structural rearrange- coni, G., Gervy, P., LeGall, I., Soularue, P., Grinas, L., Bougueleret,ments involving 8q11–q13, namely lipoblastomas (ref. L., Bellanne-Chantelot, C., Lacroix, B., Barillot, E., Gesnouin, P.,

Pook, S., Vaysseix, G., Frelat, G., Schmitz, A., Sambucy, J. L.,in Dal Cin et al., 1994; Sawyer et al., 1994), rhabdomyo-Bosch, A., Estivill, X., Weissenbach, J., Vignal, A., Riethman, H.,sarcomas (Mitelman, 1994), and renal cell carcinomasCox, D., Patterson, D., Gardiner, K., Hattori, M., Sakaki, Y., Ichi-(Elfving, 1996). It should also be mentioned that pleo-kawa, H., Ohki, M., Le Paslier, D., Heilig, R., Antonarakis, S., and

morphic adenomas of the lacrimal glands show recur- Cohen, D. (1992). Continuum of overlapping clones spanning therent rearrangements of 8q12 (Hrynchak et al., 1994), entire human chromosome 21q. Nature 359: 380–387.including a t(3;8)(p21;q12). Because of the well-known Cohen, D., Chumakov, I., and Weissenbach, J. (1993). A first-genera-

tion physical map of the human genome. Nature 366: 698–701.histopathological and clinical similarities betweenCohen, J. C. C., Cali, J. J., Jelinek, D. F., Mehrabian, M., Sparkes,pleomorphic adenomas originating from the salivary

R. S., Lusis, A. J., Russell, D. W., and Hobbs, H. H. (1992). Cloningand lacrimal glands, it is likely that the 8q12 abnor-of the human cholesterol 7a-hydroxylase gene (CYP7) and localiza-malities in the lacrimal gland tumors also affect tion to chromosome 8q11–q12. Genomics 14: 153–161.

PLAG1. As for the other tumor types mentioned, itCorey, S. J., and Shapiro, D. N. (1994). Localization of the human

remains to be shown whether the rearrangements in gene for Src-related protein tyrosine kinase LYN to chromosomethese cases also affect this gene. Preliminary FISH 8q11–12: A lymphoid signaling cluster? Leukemia 8: 1914–1917.analysis of a lipoblastoma with an 8q12 translocation Craig, J. M., and Bickmore, W. A. (1993). Chromosome bands—Fla-

vours to savour. BioEssays 15: 349–354.breakpoint has shown that the breakpoint, as in pleo-Dal Cin, P., Sciot, R., De Wever, I., Van Damme, B., and Van Denmorphic adenomas, is distal to MOS (unpublished ob-

Berghe, H. (1994). New discrimative chromosomal marker in adi-servations). Unfortunately no markers telomeric topose tissue tumors. The chromosome 8q11–q13 region in lipoblas-MOS could be tested. Additional FISH studies are now toma. Cancer Genet. Cytogenet. 78: 232–235.

in progress to resolve these critical questions. Dib, C., Faure, S., Fizames, C., Samson, D., Drouot, N., Vignal, A.,Millasseau, P., Marc, S., Hazan, J., Seboun, E., Lathrop, M., Gya-pay, G., Morissette, J., and Weissenbach, J. (1996). A comprehen-ACKNOWLEDGMENTSsive genetic map of the human genome based on 5264 microsatel-lites. Nature 380: 152–154.The authors thank C. Huysmans, J. Geurts, and M. M. Behrendt

for their contributions. We thank Nils Mandahl for providing tumor Doerflinger, N., Linder, C., Ouahchi, K., Gyapay, G., Weissenbach,material from a lipoblastoma. This work was supported in part by J., Le Paslier, D., Rigault, P., Belal, S., Ben Hamida, C., Hentati,the EU through the Biomed 1 program Molecular Cytogenetics of F., Ben Hamida, M., Pandolfo, M., DiDonato, S., Sokol, R., Kayden,Solid Tumours, the Geconcerteerde Onderzoekacties 1997-2001, the H., Landrieu, P., Durr, A., Brice, A., Goutieres, F., Kohlschutter,Fonds voor Wetenschappelijk Onderzoek Vlaanderen (FWO), the A., Sabouraud, P., Benomar, A., Yahyaoui, M., Mandel, J.-L., andASLK-programma voor Kankeronderzoek, the Swedish Cancer Soci- Koenig, M. (1995). Ataxia with Vitamin E deficiency: Refinementety, and the IngaBritt and Arne Lundbergs Research Foundation. of genetic localization and analysis of linkage disequilibrium byK. Kas is a postdoc of the FWO. M. Voz has a stipend from the FWO using new markers in 14 families. Am. J. Hum. Genet. 56: 1116–program Kom op tegen Kanker. 1124.

Elfving, P. (1996). Cytogenetic studies of normal kidney tissue andrenal cell carcinoma. Thesis, Lund University, Lund.REFERENCES

Eyre, H., Akkari, P. A., Wilton, S. D., Callen, D. C., Baker, E., andLaing, N. G. (1995). Assignment of the human skeletal muscle a-Adam-Klages, S., Adam, D., Wiegmann, K., Struve, S., Kolanus, W.,tropomyosin gene (TPM1) to band 15q22 by fluorescence in situSchneider-Mergener, J., and Kronke, M. (1996). FAN, a novel WD-hybridization. Cytogenet. Cell Genet. 69: 15–17.repeat protein, couples the p55 TNF-receptor to neutral sphingo-

Feinberg, A. P., and Vogelstein, B. (1984). A technique for radiolabel-myelinase. Cell 86: 937–947.ing DNA restriction endonuclease fragments to high specific activ-Albertsen, H. M., Abderrahim, H., Cann, H. M., Dausset, J., Le Pas-ity. Anal. Biochem. 132: 6–13.lier, D., and Cohen, D. (1990). Construction and characterization

Geurts, J., Schoenmakers, H. F. P. M., Mols, R., and Van de Ven,of a yeast artificial chromosome library containing seven haploidW. J. M. (1994). Improved procedure for rapid isolation and se-human genome equivalents. Proc. Natl. Acad. Sci. USA 87: 4256–quencing of DNA insert termini in yeast artificial chromosomes.4260.Methods Mol. Cell. Biol. 4: 257–265.Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J.

Green, E. D., and Olson, M. V. (1990). Systematic screening of yeast(1990). Basic local alignment search tool. J. Mol. Biol. 215: 403–artificial-chromosome libraries by use of the polymerase chain re-410.action. Proc. Natl. Acad. Sci. USA 87: 1213–1217.Blunt, T., Finnie, N. J., Taccioli, G. E., Smith, G. C. MG. C., De-

mengeot, J., Gottlieb, T. M., Mizuta, R., Varghese, A. J., Alt, F. W., Gyapay, G., Morissette, J., Vignal, A., Dib, C., Fizames, C., Millas-seau, P., Marc, S., Bernardi, G., Lathrop, M., and Weissenbach, J.Jeggo, P. A., and Jackson, S. P. (1995). Defective DNA-dependent

protein kinase activity is linked to V(D)J recombination and DNA (1994). The 1993–94 Genethon human genetic linkage map. Na-ture Genet. 7: 246–339.repair defects associated with the murine scid mutation. Cell 80:

813–823. Hrynchak, M., White, V., Berean, K., and Horsman, D. (1994). Cyto-genetic findings in seven lacrimal gland neoplasms. Cancer Genet.Bullerdiek, J., Wobst, G., Meyer-Bolte, K., Chilla, R., Haubrich, J.,

Thode, B., and Bartnitzke, S. (1993). Cytogenetic subtyping of 220 Cytogenet. 75: 133–138.salivary gland pleomorphic adenomas: Correlation to occurrence, Kas, K., Voz, M., Roijer, E., Astrom, A.-K., Meyen, E., Stenman, G.,histological subtype, and in vitro cellular behavior. Cancer Genet. and Van de Ven, W. J. M. (1997). Promotor swapping between theCytogenet. 65: 27–31. genes for a novel zinc finger protein and b-catenin in pleomorphic

adenomas with t(3;8)(p21;q12) translocations. Nature Genet. 15:Brunton, L. L., and Lupton, S. D. (1990). An STS in the human IL7gene located at 8q12–13. Nucleic Acids Res. 18: 1315. 170–174.

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal

KAS ET AL.358

Mitelman, F. (1994). ‘‘Catalog of Chromosome Aberrations in Can- Bartnitzke, S., Bullerdiek, J., Dal Cin, P., De Jong, P. J., Van denBerghe, H., and Van de Ven, W. J. M. (1994). Physical mapping ofcer,’’ 5th ed., Wiley-Liss, New York.chromosome 12q breakpoints in lipoma, pleomorphic salivaryMock, B. A., Stoye, J., Spence, J., Jackson, I., Eppig, J. T., Fiedorek,gland adenoma, uterine leiomyoma, and myxoid liposarcoma. Ge-F. T., Jr., and Neumann, P. E. (1996). Encyclopedia of the mousenomics 20: 210–222.genome V, Mouse chromosome 4. Mamm. Genome 6: S73–S96.

Schoenmakers, E. F. P. M., Wanschura, S., Mols, R., Bullerdiek, J.,Nelson, D. L., Ledbetter, S. A., Corbo, L., Victoria, M. F., Ramirez-Van den Berghe, H., and Van de Ven, W. J. M. (1995). RecurrentSolis, R., Webster, T. D., Ledbetter, D. H., and Caskey, C. T. (1989).rearrangements in the high mobility group protein gene, HMGI-Alu polymerase chain reaction: A method for rapid isolation ofC, in benign mesenchymal tumours. Nature Genet. 10: 436–444.human-specific sequences from complex DNA sources. Proc. Natl.

Seifert, G., Miehlke, A., Haubrich, J., and Chilla, R. (1986). DiseasesAcad. Sci. USA 86: 6686–6690.of the salivary glands. In ‘‘Pathology Diagnosis. Treatment. FacialRoijer, E., Kas, K., Klawitz, I., Bullerdiek, J., Van de Ven, W., andNerve Surgery,’’ pp. 182–194, Thieme, Stuttgart/New York.Stenman, G. (1996). Identification of a Yeast Artificial Chromo-[Translated by P. M. Stell]some spanning the 8q12 translocation breakpoint in pleomorphic

Tachibana, K., Umezawa, A., Kato, S., and Takano, T. (1988). Nucleo-adenomas with t(3;8)(p21;q12). Genes Chromosomes Cancer 17:tide sequence of a new YPT1-related human cDNA which belongs166–171.to the ras gene superfamily. Nucleic Acids Res. 16: 10368.Roijer, E., Kas, K., Van de Ven, W., and Stenman, G. (1997). Mapping

Testa, J. R., Parsa, N. Z., Le Beau, M. M., and Vande Woude, G. F.of the 8q12 translocation breakpoint to a 40-kb region in a pleomor-(1988). Localization of the proto-oncogene MOS to 8q11–q12 by inphic adenoma with an ins(8;3)(q12;p21.3p14.1). Cytogenet. Cell.situ chromosomal hybridization. Genomics 3: 44–47.Genet. 76: 23–26.

Tomfohrde, J., Wood, S., Schertzer, M., Wagner, M. J., Wells, D. E.,Rychlik, W., and Rhoads, R. E. (1989). A computer program for choos-Parrish, J., Sadler, L. A., Blanton, S. H., Daiger, S. P., Wang, Z.,ing optimal oligonucleotides for filter hybridization, sequencingWilkie, P. J., and Weber, J. L. (1992). Human chromosome 8 link-and in vitro amplification of DNA. Nucleic Acids Res. 17: 8543–age map based on short tandem repeat polymorphisms: Effect of8551.genotyping errors. Genomics 14: 144–152.

Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). ‘‘MolecularVan Ommen, G. J. (1993). First report of the HUGO YAC committee.Cloning: A Laboratory Manual,’’ Cold Spring Harbor Laboratory

In ‘‘Genome Priority Reports’’ (A. J. Cuticchia, P. L. Parson, andPress, Cold Spring Harbor, NY.H. P. Klinger, Eds.), Vol. 1, pp. 885–888, Karger, Basel.

Sandros, J., Stenman, G., and Mark, J. (1990). Cytogenetic and mo- Waldron, C. A. (1991). Mixed tumor (Pleomorphic adenoma) and my-lecular observations in human and experimental salivary gland oepithelioma. In ‘‘Surgical Pathology of the Salivary Glands’’tumors. Cancer Genet. Cytogenet. 44: 153–167. (G. L. Ellis, P. L. Auclair, and D. R. Gnepp, Eds.), pp. 165–186,

Sapru, M., Gu, J., Gu, X., Smith, D., Yu, C.-E., Wells, D., and Wagner, Saunders, Philadelphia.M. (1994). A panel of radiation hybrids for chromosome 8. Geno- Wood, S., Schertzer, M., Drabkin, H., Patterson, D., Longmire, J. L.,mics 21: 208–216. and Deaven, L. L. (1992). Characterization of a human chromo-

Sawyer, J. R., Parsons, E. A., Crowson, M. L., Smith, S., Erickson, some 8 cosmid library constructed from flow-sorted chromosomes.S., and Bell, J. M. (1994). Potential diagnostic implications of Cytogenet. Cell Genet. 59: 243–247.breakpoints in the long arm of chromosome 8 in lipoblastoma. Yasuda, K., Espinosa, R., III, Takeda, J., Le Beau, M. M., and Bell,Cancer Genet. Cytogenet. 76: 39–42. G. I. (1994). Localization of the kappa opioid receptor gene to hu-

man chromosome band 8q11.2. Genomics 19: 596–597.Schoenmakers, E. F. P. M., Kools, P. F. J., Mols, R., Kazmierczak, B.,

AID GENO 4819 / 6r3e$$$101 07-01-97 06:45:32 gnmal