(CANCER RESEARCH 58. 10.14-1041. March 1. 1998|

Human Lung Cancer Antigens Recognized by Autologous Antibodies:

Definition of a Novel cDNA Derived from the Tumor Suppressor

Gene Locus on Chromosome 3p21.3Ali O. Cure, Nasser K. Altorki, Elisabeth Stockent, Matthew J. Scanlan, Lloyd J. Old, and Yao-Tseng Chen1

Ludwig Institute for Cancer Research. Ne»' York Unii at Memorial Sloan-Ketlering Cancer Center ¡A.O. G.. E. S.. M. J. S., L J. O., Y-T. C.]; and Departments of Surgery

¡N.K. A.I and Pathology ¡Y-T.C.I, Cornell University Medical College, New York, New York 10021

ABSTRACT

Serological analysis of a recombinant lung cancer cDNA expressionlibrary with the autologous patient serum led to the isolation of 20 clonesrepresenting 12 different genes: 4 of these were known genes, and theother 8 were previously unknown genes. Of the four known genes, aldolaseA (V)-/,(-/1, previously shown to be overexpressed in lung cancer, was

most frequently isolated. The other three genes were annexin XI, humanHIV Rev-interacting protein Rip-1, and the human homologue of theATP-binding arsA component of the bacterial arsenite transporter, all of

which are known to be widely expressed in human tissues. Among theeight unknown genes, of most interest was NY-LU-12. Cloning of full-

length NY-LU-12 showed that this cDNA was derived from the same gene

as gI6, a partially sequenced gene that mapped to the lung cancer tumorsuppressor gene locus on chromosome 3p21. The reported g!6 sequence,however, was significantly shorter (2433 versus 3591 bp). As a result ofalternate splicing and subsequent frameshift, the reported gl6 protein is603 amino acids shorter than the NY-LU-12 product (1123 residues) at its

COOH terminus and would therefore lack the epitopes recognized by theautologous serum. Analysis of the putative NY-LU-12 protein sequencepredicted that it is a nuclear zinc finger protein with two RNA-binding

domains, and Southern analysis showed that this gene is partially deletedin the lung cancer line NCI-H740 but not in nine other lung cancer lines.Screening of normal and cancer patient sera showed anti-NY-LU-12

seroreactivity in 2 of 21 allogeneic lung cancer patients but not in 24patients with other tumors or in 16 sera from healthy donors. Comparisonof NY-LU-12 cDNA from lu 15 tumor and normal lung tissue by DNAsequencing and/or single-strand conformation polymorphism analysis

showed no evidence of mutation. Considering the high frequency of 3p21alterations in lung cancer and the fact that the tumor suppressor gene orgenes in this locus have not been identified, additional studies on theNY-LU-12 gene and its product are warranted.

INTRODUCTION

The structural definition of human tumor antigens recognized bythe autologous host has been a long-standing challenge in tumor

immunology. The growing list of human tumor antigens recognizedby cytotoxic T cells (1,2) and antibodies (3, 4) provides convincingevidence for immune recognition in the host of origin, as well asattractive targets for vaccine-based approaches to cancer therapy.

Using specific antitumor CTLs as probes, a number of CTL targetshave been cloned on the basis of MHC class I reactivity (1, 2).However, this approach involves cultured cancer cell lines and stableCTL lines from the same patient, two requirements that cannot easilybe met in many tumor types.

Sahin el al. (3) have recently introduced a powerful new methodology for identifying human tumor antigens eliciting humoral immune

Received 11/7/97; accepted 1/16/98.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 in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed, at Department of Pathology,

Cornell University Medical College, 1300 York Avenue. New York. NY 10021. Phone:(212)746-6472; Fax: (212)746-4483; E-mail: ytchen@mail.med.comell.edu.

response. The method has been called SEREX.2 Novel as well as

previously defined tumor antigens have been identified by the SEREXmethod, including MAGE-1 and tyrosinase. both originally identifiedby cloning of the epitopes recognized by CTL (5-7). Thus, antibody

screening of cDNA libraries prepared from frozen human tumors canbe used to identify antigens eliciting a cellular immune response,including CTL, circumventing the need for established cultured autologous cell lines and stable CTL lines.

SEREX has been used to analyze expression cDNA libraries fromvarious malignancies, including melanoma (3), renal cell carcinoma(3), Hodgkin's disease (3), astrocytoma (3), esophageal squamous cell

carcinoma (4), and colon cancer.3 In addition to cancer tissues,

SEREX has also been applied to the screening of allogeneic testicularcDNA libraries (8), aiming specifically at the isolation of cancer-testisantigens, i.e., antigens such as MAGE and NY-ESO-1 (4). which

show restricted tissue expression primarily in cancer and in normaltestis (4). These efforts have led to the identification of a number ofnovel tumor antigens that fall into several categories, including lineage-specific differentiation antigens (3), cancer-testis antigens (4, 8),

and mutated gene products, e.g.. p53. The isolation of mutated p53 bySEREX proved that, using host humoral immune responses as a tool,it is possible to isolate mutated TSG products through SEREX, thusproviding a novel approach to defining TSGs. Here, we have extendedSEREX analysis to lung cancer and characterized one of the isolatedcDNA clones as a gene that maps to the lung cancer TSG locus onchromosome 3p21 (9).

MATERIALS AND METHODS

RNA Extraction and Construction of cDNA Expression Library. TotalRNA was extracted from cultured cell lines and from normal and tumor tissues.A cDNA library was constructed from a case of moderately differentiatedadenocarcinoma of the lung, obtained from the Department of Pathology atThe New York Hospital (New York, NY). The library was constructed in aÀZAPExpress vector using a cDNA library kit (Stratagene, La Jolla, CA).

Immunoscreening of the cDNA Library. The cDNA library was screenedwith autologous patient's serum as described previously (3. 4). Briefly, the

serum was diluted 1:10 and preabsorbed with phage-transfected Escherichia

coli lysate. and a 1:10 dilution of the absorbed serum (final dilution of serum,1:100) was incubated overnight at room temperature with the nitrocellulosemembranes containing the phage plaques. After washing, the filters wereincubated with alkaline phosphatase-conjugated goat antihuman Fc-y second

ary antibodies and the reactive phage plaques were visualized by incubatingwith 5-bromo-4-chloro-3-indolyl-phosphate and nitroblue tetrazolium. Phage-

mid clones encoding human immunoglobulin sequences were subsequentlyeliminated during the secondary screening.

Sequence Analysis of the Reactive Clones. The reactive clones weresubcloned. purified, and in vitro excised to pBK-CMV plasmid forms (Strat-

2 The abbreviations used are: SEREX. serological analysis of recombinant cDNA

expression libraries; TSG. tumor suppressor gene; RT, reverse transcription; RACE, rapidamplification of cDNA ends; SSCP. single-strand conformation polymorphism; EST,expressed sequence tag; ORF, open reading frame; nt, nucleotide(s).

' M. J. Scanlan. Y-T. Chen. B. Williamson, A. O. Cure, E. Stocken, J. D. Gordan, Ö.

Tureci, U. Sahin, M. Pfreundschuh. and L. J. Old. Characterization of human colon cancerantigens recognized by autologous antibodies. Int. J. Cancer, in press. 1998.

1034

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

IMMUNOGENIC LUNG CANCER ANTIGEN FROM CHROMOSOME 3P2I.3

agene). Plasmid DNA was prepared using Wizard Miniprep DNA PurificationSystem (Promega. Madison, WI). The inserted DNA was evaluated by EcoRl-

Xba\ restriction mapping, and clones representing different cDNA inserts weresequenced. The sequencing reactions were performed by DNA Services atCornell University (Ithaca. NY) using ABI PRISM (Perkin-Elmer Corp.)

automated sequencers.RT-PCR. To evaluate the mRNA expression pattern of the cloned cDNA

in normal and malignant tissues, gene-specific oligonucleotide primers forPCR were designed to amplify cDNA segments of 300-400 bp in length, withthe estimated primer melting temperature in the range of 65-70°C. All primers

were commercially synthesized (Operon Technologies, Alameda, CA). RT-

PCRs were performed using 35 amplification cycles in a thermal cycler(Perkin-Elmer) at an annealing temperature of 60°C.

Genomic Southern Blot Analysis. Genomic DNA was extracted from celllines and frozen tumor tissue. Following restriction enzyme digestion, theDNA was separated on a 0.7% agarose gel, blotted onto nitrocellulose filters,and hybridized to a 12P-labeled DNA probe at high stringency (65°C.aqueous

buffer). Washing of the blot was also under high-stringency conditions, with afinal wash in 0.2 x SSC with 0.2% SDS at 65°C.

5' RACE Analysis of mRNA Transcripts. To identify the 5' end of the

mRNA transcripts, we used RACE methodology, with the Marathon cDNAamplification kit (Clontech) and adaptor-ligated testicular cDNA as the sub

strate. The PCR products, after separation by agarose gel electrophoresis. werecloned into the direct PCR cloning vector pGEM-T (Promega).

SSCP Analysis. SSCP analysis was performed to analyze cDNA fromvarious tissues, as described previously (10). Briefly. PCR was performed asdescribed above, using 2 /il of RT product in a final volume of 25 \t\, with 1juCi of ["P]dCTP (-3000 Ci/mmol; New England Nuclear) per reaction.

After the PCR. 1 \u of the reaction was diluted 1:10 in 0.1% SDS-10 mM

EDTA. Five JLI!of the diluted sample were mixed with 5 \¡\of 2X denaturingbuffer (95% formamide. 20 mM EDTA, 0.05% bromphenol blue, and 0.05%xylene cyanol). heat-denatured at 92°Cfor 3 min. and electrophoresed through

a 4.5% polyacrylamide gel with 10% glycerol. As controls, aliquots of thesame samples were diluted with a standard nondenaturing DNA loading dyeand electrophoresed in parallel. The electrophoresis was performed at roomtemperature at a constant power of 8-10 W. The gel was then dried andautoradiography performed for 15-24 h with an intensifying screen.

RESULTS

Identification of Immunoreactive cDNA Clones. A cDNA expression library of 1.42 X IO7 primary clones was prepared from

Lu 15, a specimen of moderately differentiated adenocarcinoma of thelung, and 8 X IO5phage plaques were immunoscreened with absorbed

autologous patient serum at 1:100 dilution. Excluding false-positive

clones encoding immunoglobulin gene fragments. 20 positive cloneswere identified. These clones were purified, and their sequences wereanalyzed. Comparisons of the sequences showed that these clonesrepresented cDNAs from 12 distinct genes, designated NY-LU-Ì-NY-LU-12 (Table 1). A homology search through the GenBank/European

Moleculary Biology Organization databases revealed that 4 of the 12genes corresponded to previously known molecules, and the other 8were unknown genes, with sequence identity limited only to ESTs orto short segments of known genes.

Of the 4 known genes, aldolase A (NY-LU-I) was most frequently

isolated, representing 6 of 20 primary positive clones in the entirescreening. NY-LU-2, represented by two isolates, was the humanhomologue of the ATP-binding arsA component of the bacterial

arsenite transporter, a gene which has been shown to be ubiquitouslyexpressed in various tissues (11). NY-LU-3 encodes annexin XI,which is a M, 56,000 ubiquitously expressed antigen to which autoan-

tibodies have been described in sera from patients with variousautoimmune diseases (12, 13). The last gene in this group, NY-LU-4,codes for the human HIV Rev-interacting protein Rip-1, which has

been shown to be expressed in the monocyte cell line U937 and therhabdomyosarcoma cell line RD, as well as in adherent monocytesand primary lymphocytes (14).

Of the eight unknown genes, six (NY-LU-5, -7, -8, -9. -10, and -//)

shared sequence identity with reported ESTs. possibly representingcDNA products derived from the same genes. These ESTs werederived from various somatic tissues unrelated to lung, e.g., neuron,pregnant uterus, colon, endothelial cells, and so on, suggesting thatthese genes are widely expressed in human tissues (Table 1). Theseclones were not further investigated. Another unknown gene, NY-LU-6, showed no homology to sequences in public databases. Thetissue expression pattern of this gene was evaluated by RT-PCRanalysis using gene-specific primers and a normal tissue RNA panel

consisting of lung, colon, kidney, liver, brain, and testis. Resultsshowed universal expression in these tissues, and this clone was notfurther analyzed.

NY-LU-12 Is on the TSG Locus of Chromosome 3p21. The lastgene in the unknown gene group. NY-LU-12, was represented by the

immunoreactive clone Lu 15/251. This clone, 1081 bp in length,contained an uninterrupted ORF of 952 bp. followed by a 129-bp 3'

untranslated region. No translation initiation codon was identified,indicating that this was a partial cDNA clone.

Table 1 Genes isolated by SEREX analysis of an adenocarcinoma of the lung

Gene designation Gene/sequence identity (accession no.) cDNA Comments

NY-LU-I

NY-W-2

NY-LU-3

NY-LU-4

NY-LU-5NY-LU-6

NY-LU-7

NY-LU-8NY-LU-9

NY-LU-10

NY-LU-Il

NY-LU-12

Aldolase A. N and H type (X06352)

hASNA-1 (U60276)

Annexin XI (L19605)

Rip-1 (U55766)

Unknown (W61291. W92962, and so on)Unknown (EST: none)

Unknown (EST: W23466, AA167732, and so on)

Unknown (EST: Z78323, N39225. and so on)Unknown (EST: W26569. AA036884. and so on)

Lul5/24, 72,83. 158, 219.241

Lu 15/26, 66

Lu 15/64

Lu 15/65

Lu 15/80Lu 15/85

Lul5/135, 217

Lul5/l39Lul5/145

Unknown (M29204; EST: AA057859. AA088404. and so on) Lul5/I54

Unknown (EST: W23466, AA057400. and so on) Lu 15/270

gl6 (U50839) Lul5/251

Human fructose-1,6-diphosphate aldolase A. Expressed in

muscle (M type) hut also in most other tissues (N and Htypes). Levels increased in most lung cancers.

Human homologue of the ATP-binding arsA component of

the bacterial arsenite transporter. Ubiquitously expressed.Homo sapiens Mr 56,000 autoanligen. Antibodies to annexin

XI are found in multiple autoimmune diseases.Ubiquitously expressed.

Human HIV Rev-interacting protein. Expressed in B cells,

monocytes. and rhahdomyoma cells.Expressed ubiquitously (by RT-PCR; data not shown).Sequence contains no ORF. Expressed ubiquitously (by RT-

PCR: data not shown).EST derived from neuron, pregnant uterus, lung cancer,

parathyroid tumors, and so on.EST derived from fetal heart, relina, and so on.EST derived from retina, pregnant uterus, fetal liver/spleen,

and so on.Similar to DNA-binding factor. EST derived from colon,

pancreas, pregnant uterus, fibroblasts. and so on.EST derived from retina, pregnant uterus, fetal heart, fetal

liver/spleen, parathyroid tumors, and so on.Located at the 3p2l TSG locus (see text).

1035

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

IMMUNOGENIC LUNG CANCER ANTIGEN FROM CHROMOSOME 3P2I.3

LUCA15 PELVRNGDEENPLKRGLVAAYSGDSDNEE ELVERLESEEEKLADWKKMACLLCRRQFPNKDAL 662DXS8237E DLPKLASDDRPSPPRGLVAAYSGESD.SEE EQERGGPEREEKLTDWQKLACLLCRRQFPSKEAL 233

NY-LU-12 IKHQQLSDLHKQNLEIHRKIKQSEQELAYLERRERE.GKFKGRGNDRREKLQSFDSPERKRIKYSRETDS..DRKLVDKEDID 1050

LUCA15 VRHQQLSDLHKQNMDIYRRSRLSEQELEALELRERE.MKYRDRAAERREKYGIPEPPEPKRKKQFDAGTV. .NYEQPTKDGID 742DXS8237E IRHQQLSGLHKQNLEIHRRAHLSENELEALEKNDMEQMKYRDRAAERREKYGIPEPPEPKRRKYGGISTASVDFEQPTRDGLG 316

NY-LU-12 TSSKGGCVQQATGiraKGTGLGYGHPGLASSEEAEGRMRGPSVGASGRTSKRQSNETYKDAVRPVMFARYKELD 1123

LUCA15 HSNIGNKMLQAMGWREGSGLGRKCQGITAPIEAQVRLKGAGLGAKGSAYGLSGADSYKDAVRKAMFARFIEME 815DXS8237E SDNIGSRMLÇjAMGWKEGSGLGRKKQGIVTPIEAQTRVRGSGLGARGSSYGVTSTESYKETLHKTMVTRFNEAQ 389

Fig. I. Amino acid homology between NY-LU-12, LUCA15. and DXS8237E. Boldface type, sequences shared by al least two genes. A highly conserved segment, almost identicalin sequences of all three genes, can be identified between NY-LU-12 amino acid positions 945 and 989. A C2H2-type /.inc-finger motif is contained within this region.

A sequence homology search revealed that NY-LU-12 shared up to

30% homology with two other human proteins at its COOH terminus(Fig. 1), LUCA15 (Ref. 15; GenBank accession no. U23946) andDXS8237E (Ref. 16, GenBank accession no. P98175), and to Sl-1.

the rat counterpart of DXS8237E (17). Both LUCA15 and DXS8237Econtain a bipartite nuclear localization signal and a C2H2-type zinc

finger domain in this homologous region. LUCA 15 has previouslybeen mapped to the TSG locus on chromosome 3p21.3 (15), andDXS8237E was mapped to chromosome Xpl 1.23 (16). Of particularinterest, however, was that a short segment (92 bp) at the 5' end ofNY-LU-12 was identical to the 3' terminus of a previously identifiedgene, gl6 (GenBank accession number U50839).4 gl6 was mapped to

chromosome 3p21.3 and was involved in the 3p21 deletion in thesmall cell lung cancer line NCI-H740, and these two features were

shared by the LUCAI5 gene (15).To compare NY-LU-12 with gI6, the full-length NY-LU-Ì2cDNA

sequence was obtained from normal testicular mRNA through acombination of 5' RACE and direct PCR cloning strategies. The

predominant cDNA form, excluding the poly(A) tail, is 3591 bp long.An ORF of 1123 amino acid residues was identified (nt 102-3470),with 101 bp of 5' untranslated and 129 bp of the 3' untranslated

region. The nt and amino acid sequences are shown in Fig. 2.The comparison of NY-LU-12 with the g!6 sequence verified that

these two are identical genes and mapped NY-LU-12 to the TSG locus

on 3p21.3. The reported g¡6sequence, 2433 bp in length, lacks the110 bases at the 5' end. which include the translation initiation codonat nt 102, and also the 980 nt at the 3' end of NY-LU-12. In addition,

a 74-bp DNA segment (nt 1586-1659 of NY-LU-12) was absent in thereported $16 sequence. Oligonucleotide primers flanking this 74-bp

region were designed and used to amplify RNA from one normal lungspecimen, five lung cancer cell lines, and six lung cancer specimens.Two RT-PCR products were seen in every specimen, corresponding to

the sizes of the two cDNA variants. Thus, it was concluded that thisvariation represents an alternate splicing event that occurs in bothnormal and cancerous lung tissues. Of interest, however, was thedifference in the putative translation products resulting from thisadditional 74-bp exon. In the absence of this exon, the ORF ofNY-LU-12 would end in the termination codon at nt 1736. as reported

for gl6, with a total length of 520 amino acid residues (in contrast to1123 residues in the longer transcript). Moreover, this shorter formwould not encode the COOH-terminal portion of the NY-LU-12

protein, the segment responsible for the immunoreactivity of Lul5/251 to the autologous patient serum.

Features of NY-LU-12 and Its Putative Gene Product. Analysisof the NY-LU-12 amino acid sequence showed 20 inexact 6-amino

acid repeats, close to the NH-, terminus, with a D(F/Y)RGR(D/E)consensus sequence (Fig. 2). These repeats were separated by 4- to6-amino acid intervals, which did not show sequence homology

4 F. Latif, F. M. Duh, M. H. Wei. Y. Sekido. J. Minna, and M. Lerman. unpublished

observations.

among each other. This feature in the primary sequence is distinctiveamong known proteins. Hydrophilicity analysis revealed that thisregion, hydrophilic in general, contains 20 hydrophilic peaks followedby hydrophobic turns, each cycle corresponding to one hexamerrepeat (Fig. 3). Although the significance of this characteristic isunclear at present, this segment of the sequence is highly rich inarginine and aspartic acid, a feature shared by RNA-binding proteins(18-21 ), e.g., RNA helicase (20), heteronuclear RNA 3' end cleavage

stimulation factor (21 ), and so on. Analysis of the putative NY-LU-12

protein against the PROSITE and PfamA protein profile databases,indeed, predicted two RNA-binding domains, located downstreamfrom the hexapeptide repeats at amino acid residues 456-530 and656-740 (Fig. 4). In addition, NY-LU-12 was also predicted to be aDNA-binding protein, with two zinc-finger motifs identified (aminoacids 537-565 and 955-985). Consistent with the notion that theNY-LU-12 product is a DNA-binding/RNA-binding protein, a bipartite nuclear localization signal (amino acids 1016-1032) and a four-residue nuclear localization pattern (PRKR, amino acids 604-607)were identified. These domain characteristics of NY-LU-12, whencompared to its homologous genes LUCA15 and SI-I (DXS8237E),

showed a striking similarity. As shown in Fig. 4. all predicted functional domains in NY-LU-12 were also present in LUCA 15 and Sl-1.

Moreover, the order and spatial relationships of these domains in eachprotein are highly conserved, implicating an underlying functionalsignificance.

Possible sites for posttranslational modifications of NY-LU-12were also analyzed. Results showed potential sites for tyrosine sulfa-

tation, amidation, as well as phosphorylation sites for protein kinaseA, protein kinase C. casein kinase II, and tyrosine kinase. A PESTregion, peptide sequences rich in proline (P), glutamic acid (E), serine(S), threonine (T), and aspartic acid, consistently found among unstable proteins with short half-lives, was identified at amino acids 897-928 (Fig. 2), implying that NY-LU-12 was an unstable protein.

Additional cDNA Variants of NY-LU-12. In the process of 5'

RACE cloning of the full-length NY-LU-12, three minor forms of

cDNA products were identified that varied in their transcriptionalinitiation sites and in their exon usage in the 5' segment of this gene.

These variants will be described as transcripts B, C, and D. Fig. 5shows the comparison of these transcripts to the predominant cDNAform (transcript A, see Fig. 2)

Transcript B (Fig. 5. bottom) contains an additional exon of 208 bp,inserted at nt 145 of the NY-LU-12 sequence. The original ORF ofNY-LU-12 is disrupted due to this inserted sequence, and the AUG

initiation codon used by transcript A is, thus, unlikely to be used bythis transcript. A new potential translation initiation site, however, isfound within this new exon and would continue the translation into theORF of transcript A. The final product would be a protein of 1177amino acids, with the 69 residues at the NH2 terminus being differentfrom those of transcript A. Interestingly, this new exon encodes asignal peptide that is not present in the transcript A, and it is possible

1036

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

[MMUNOOENIC LUNG CANCER ANTIGEN FROM CHROMOSOME 3P2I.3

:

34

100

200

261

301

334

501

534

634

66'

901

834

BCI

•¿�M!

934

461

1001

1034

1067

1101

MWGDSRPANRTGPFRGSQEERFAPGWNRDYPPP

PLKSHAQERHSGNFPGRDSLPFDFQGHSGPPFA

ATGTAGAGGAUVEETTCGTCTTCl

S S SGCATTCTTTCAGCTATGGAGCTAGAGACGG

HSFSYGARDG3ATTTCCAGAGCAGAGA1D

F Q S R DTCATCACAGTTGS S Q LACCGCATGGT

P H G3ACTATCGAGGAGGGGAGD Y R G GE3ACTTCAGGGGTAGGGAC

D F R G R D\TACATTCTGIKSMGOC GACCTGGACAT

P G HGATTTCAGGGGGGGAGAID F R G GDGATTTTCGGGATAGAGAA

D F R D R E3GACCACCTAGPP

3ACTATAGGGGTGGAGAT ÕGTACTTCTATC ATTATAGAGGTAGGGAC 3CACCTCATATG \ACTACAGAGACAGGGAT 3CTCACGCTGT1 SACTTCAGAGD Y R G G D D Y R G R E N Y R D R D

D F R G R G D F R G R D D F R G R D

ATCAGATTTG ÃŽATâ„¢. . :GTTC1 JATTTTAGGAATAGAGAI TATCTGATTTC ÃŽACTTTAGAGACAAAGAC 3GAACACAAGTAD F R A R E D F R N R D D F R D K D

3ACTTTAGAGGCCGAGGT TCAGGTACTACTGATCTA 3ACTTTAGGGACAGGGA1 I\CGCCACATTCJ;ATTTCAGAGGTAGACAC:GATCTAGGACTGATCD F R G R G S G T T D I D F R D R D D F R G R H

D F R G R E E F K D R E MPPVDPNILDYIOPST

QDREHSGMNVNRREESTHDHTIERPAFGIQKGE

FEHSETREGETQGVAFEHESPADFQNSQSPVQD

QDKSQLSGREEQSSDAGLFKEEGGLDFLGRQDTD

YRSMEYRDVDHRLPGSQMFGYGOSKSFPEGKTA

RDAQRDLQDQDYRTGPSEEKPSRLIRLSGVPED

ATKEEILNAFRTPDGMPVKNLOLKEYNTI

TGTCTGCGTGGAGTTTTCACTCTTGGAAGATGCCATCGGATGCATGGAGGCCAACCAG AA . A A ,/, : A

v c v E F :> LLEDAIG:MEANQGTLMIQDKEVTLEY

VSSLDFWYCKRCKANIGGHRSSCSFCKNPREVT

EAKQELITYPQPQKTSIPAPLEKQPNQPLRPADK

EPE | P R K R | EEGQESRLGHQKREAERYLPPSRREG

PTFRRDRERESWSGETRQDGESKTIMLKRIYRS

TPPEVIVEVLEPYVRLTTANVRIIKNRTGPMGHT

YGFIDLDSHVEALRVVKILQNLDPPFSIDGKMV

AVNLATGKRRNDSGDHSDHMHYYQGKKYFRDRR

GGGRNSDWSSDTNRQGQQSSSDCYIYDSASGYYY

DPLAGTYYDPNTQQEVYVPQDPGLPEEEEIKEK

KPTSQGKSSSKKEMSKRDGKEKKDRGVTRFOEN

ASEGKAPAEDVFKKPLPPTVKKEESPPPPK V V N P

LIGLLGEYGGDSDYEEEEEEEQTPPPQP R T A Q P

QKREEQTKKENEEDKLTDWNKLACLLCRRQFPN

KEVLIKHQQLSDLHKQNLEIHRKIKQSEQELAYL

ERREREGKFKGRGNDRR

SRETDSDRKLVDKEDIDTSSKGGCVQQATGMRK

GTGLGYGHPGLASSEEAEGRMRGPSVGASGRTSK

RQSNETYRDAVRRVMFARYKELD'"

buu

TOO

h00

Milu

1000

1100

1200

1300

1400

1500

1600

noo

1800

1900

2000

2100

2200

2300

2400

2500

2600

2700

3000

3100

3200

3300

3599

Fig. 2. nt and amino acid sequences of NY-LU-12, derived from (he predominant cDNA form (see textl. Riçhi.ni numbers; left, amino acid numbers. Boldface type, putativetranslation initiation site (ATO): *. termination codon. D (between nt 350 and 1020), 20 inexact hexapeptide repeats with (he D(F/Y)RGR(D/E) consensus (see texll. Shaded regions,74-bp exon, absent in the reported glfi sequence (nt 1586-1659). The amino acid sequences 604-607 and 1016-1032 (D) are nuclear localization signals, and (he amino acid sequences887-929 (underlined) show the PEST sequence.

1037

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

IMMLINOOENIC LUNG CANCER ANTIGEN FROM CHROMOSOME 3P2I.3

70 110 150

Amino acid residueFig. 3. Hydrophilicity plol of NY-LU-12, amino acids 70-300. Numbers. 20 inexact

hexaneptide repeats, revealing cyclic changes in hydrophilicity corresponding to therepeats.

that these two products are localized to different subcellular compartments.

Similar to transcript B, transcripts C and D both contained additional exon(s) that were not present in transcript A. Transcript Ccontained two extra exons in tandem that were 364 bp long, only oneof which (137 bp) was present in transcript D. These extra exon(s).inserted at the same alternate splicing site as transcript B, disruptedthe original ORF. and the only long ORF would initiate at nt 498 ofNY-LU-12 (nt 959 of transcript C and nt 635 of transcript D). Considering the long untranslated regions at the 5' ends, it is doubtful

whether transcripts C and D are, indeed, translated in vivo.Correlating with this variation of NY-LU-12 mRNA. Northern blot

analysis showed several RNA species in normal tissues, rangingapproximately from 3 to 4.4 kb. The intensity of individual bands alsoappear to vary among different tissues, suggesting tissue-specificregulation of NY-LU-12 mRNA at transcriptional and posttranscrip-

tional levels (Fig. 6).Southern Blot Analysis of NY-LU-12 in Normal and Tumor

Tissues. To investigate the status of NY-LU-Ì2in normal and tumor

cells. Southern blot analysis was performed on nine lung cancer celllines (three adenocarcinoma, two squamous cell, three large cellanaplastic. and one small cell). Lu 15 tumor DNA. and a colon cancercell line HT29 (Fig. 7). (HT29 was included due to the finding of anEST identified in GenBank, accession no. AA079461. which appearedto be a fusion sequence between the semaphorin IV gene and NY-LU-12.) Using a 1.1-kbcDNA probe (nt 1095-2140) and WmdIII-digested

DNA. the results showed that one of the two hybridizing bands wasabsent in NCI-H740, the cell line in which i>!6 has been shown to beinterrupted (see above), providing additional evidence that NY-LU-Ì2and i-Io are identical genes. The breakpoint of this deletion, deter

mined by use of primers from different regions, was further defined tobe between nt 1433 and 1777 of NY-LU-12. with the 3' sequences

homozygously deleted. Besides NCI-H740, however, no evidence of

homo/.ygous deletion was seen in other tumor cell line samples or inthe Lu 15 tumor sample. The similar band intensities and identicalsizes of the hybridizing DNA species in all specimens also arguedagainst the possibility of a heterozygous deletion or translocation ofthis gene, at least in the region analyzed. No change was found inHT29. suggesting that the semaphorin IV/NY-LU-Ì2fusion sequence

in the GenBank probably represents a cloning artifact.SSCP and Sequence Analysis of NY-LU-12 in LulS Tumor

DNA. The mapping oíNY-LU-12 to the lung cancer TSC locus raised

the possibility that an altered protein product due to a mutational eventmay be the basis for the autologous immune recognition. This possibility was explored using DNA sequencing and SSCP analysis.

The DNA sequence contained in the immunoreactive clone LulS/251 (nt 2518-3599 of NY-LU-12) was obtained from the normalcounterpart by RT-PCR cloning using autologous normal lung tissue,

and no mutations were found when it was compared to Lu 15/251.RT-PCR SSCP was then used to analyze the entire NY-LU-Ì2gene,

comparing Lu 15 tumor tissue and autologous normal lung tissue. Toencompass the whole sequence, 10 sets of primer pairs were designed,each amplifying a range of 205-603 bp. For products of >400 bp, arestriction enzyme digestion step was added prior to the electrophore-

sis step to further reduce the fragment sizes and increase the assaysensitivity. Results showed no reproducible changes between normaland tumor tissues and. thus, no evidence of mutation in Lu 15 tumorcDNA (data not shown).

Serological Response to NY-LU-12 in Lung Cancer Patients.The frequency of anti-NY-LU-12 response was examined among

normal adult and patient sera using the phage plaque assay identical tothe original immunoscreening procedure. Of 21 absorbed sera fromallogeneic lung cancer patients, one (Lu22) reacted strongly at 1:1000dilution, and another (Lu7) also reacted at 1:1000. but only weakly.Nineteen other lung cancer patient sera were nonreactive, and the serafrom 16 healthy donors, 15 colon cancer patients. 5 breast cancerpatients, 1 renal cancer patient, 1 prostate cancer patient, 1 esophagealcancer patient, and 1 melanoma patient were nonreactive as well.

DISCUSSION

Following the original description by Sahin el al. (3), SEREXmethodology has been applied to several different tumor types (4. 8,

100 200 300 400 500

I h600

H 1—700 900 1000 1100 amino acid it

20 REPEATS NY-LU-12

RÕÕZ LUCA15

R IIZI S1.1Fig. 4. Comparison of the protein domains in NY-LU-12. LUCA15. and Sl-l. showing locations of the RNA-binding domains (R], /ine linger motifs (Z). and bipartite nuclear

localization signals (N). The NH, terminus zinc-finger motif is of the Ran-binding protein type (38), whereas the COOH-terminal one is of the C-.H, type. The 20 repeats al the NH,terminus of NY-LU-12 are not found in LUCA 15 or Sl-l.

1038

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

IMMUNOGENIC LUNG CANCER ANTIGEN FROM CHROMOSOME 3P2I.3

bp 1 100 200 300 400 500 600 700 80090011 ' ¡ i l '1ATGAATG208

bpA227

bp 137bpAlD'137

bp1000reG

Transcript Variant B

120 (120)

T..: A, ...-:T i:/-'.:ìTTT;C;A6ATGGCGTTTCGCCGTGTTGBCCGGGCTGGTCTCGGACTCCTGACCTCaAGTGATCCACCTACCTCG6CCTCCCRAAGTGCTGG<SACTRTAGST .'•',!!•¿�AGATGGCGTTTCGCI » * '*-?

GTGAGCCACCGCACCTGCCATTTGGATTGGCAATCTGCAAGATTTTATTACTTAAATGCAAÕ-AGATnTTCTCATTCATTGTTCTGARGCTTGGAGTTOOAATGAAAAATTTAG

VSHRTCHLDWQSARFYYLNATDVLIHCSEAWSSNEKFSGS

CAAGAAGAAAGGTTTGCTCCCGGGTGG..QEERFAPGW..

Fig. 5. Top. schematic comparison of the different cDNA variants of NY-LU-12. The predominant cDNA form (sequence shown in Fig. 2) was designated as transcript A. Boltlfaretype, putative translation initiation site (ATG1 for each transcript. D, additional exon(s) in transcripts B, C. and D; the sizes of these exons are shown. *, transcripts C and D indicatethe translation termination sites should the translation initiation site of transcript A be used. The transcriptional start site of transcript C is 97 bp 5' to the other variants, suggesting

a possible difference in promoter usage. Bottom, The nt and putative amino acid sequence of transcript B. Shaded re^iitns. additional exon absent in transcript A: D putative signalsequence. Transcripts A and B have identical ORFs following the 3' end of this exon.

22).3 Compilation of these studies indicates that SEREX identified

antigens can be classified into at least five distinct categories.Differentiation antigens, e.g., tyrosinase in melanoma (7), represent

the first category. Two novel antigens recently identified from colon

(O

C0OCo

kb

7.5-

4.4-

2.4-

2.4-

1.35-

II

_0)o13

E 0C

C/D ^

NY-LU-12

Actin

Fig. 6. Northern blot analysis of NY-LU-12 using 2 fig of each normal tissue poly( A )+RNA. Results showed ubiquitous expression of NY-LU-12. The mRNA size is heterogeneous within individual specimens and also among the five tissues tested. Longertranscripts are seen in liver and kidney. The same blot was hybridized to a ß-actinprobe

as control.

cancer libraries, shown to have mRNA expression restricted to thegastrointestinal tract, may also belong to this category. The secondcategory are cancer-testis antigens (4), i.e., antigens with high-levelmRNA expression restricted to cancer and normal testis. SEREX-

defined antigens in this category include the SSX family (3, 8),NY-ESO-1 (4), and SCP-1.5 None of the isolates from our lung cancer

series belongs to these two categories.The third category includes overexpressed gene products that are

otherwise broadly expressed in normal tissues. An example of this isaldolase A. Aldolase A (fructose-1.6-diphosphate aldolase), whichcatalyzes the reversible aldol cleavage of fructose-1,6-diphosphate toglyceraldehyde-3-phosphate and dihydroxyacetone phosphate, has

been shown to be overexpressed in tumors, and levels of this enzymeare known to be increased up to 10 times in lung cancer and othermalignancies (23-25). Additional examples of overexpressed antigens

identified by SEREX are the carbonic anhydrase-like molecule iso

lated from renal cancer, shown to be increased up to 20 times in renalcancer (3). and eukaryotic translation initiation factor-4-y, a gene

previously shown to be amplified in squamous cell carcinoma of thelung (22).

The fourth category includes antigens that have been previouslyknown to be associated with autoimmune state or disorders. Here,annexin XI. known to be reactive to patient sera in various autoimmune diseases, is an example of such antigens. The immune recognition of these antigens probably stems from the high immunogenicityof these molecules, and host immune responses are elicited followingtumor growth and/or necrosis. Such antigens have been identified inalmost all SEREX studies, including our studies on esophageal cancer(4) and colon cancer.

"*0. Tureci, U. Sahin. C. Zwick. M. Koslowski, G. Seitz. and M. Pfreundschuh.

Identification of a meiosis-specific protein as a new member of (he cancer/testis antigensuperfamily. submitted for publication.

1039

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

IMMUNOOENIC LUNG CANCER ANTIGEN FROM CHROMOSOME 3P2I.3

Fig. 7. Southern blot analysis of NY-LU-12

with DNA from eight lung cancer cell lines, thecolon cancer cell line HT29, and the Lu 15 lungtumor specimen (15 ¡igof W/ndlll-digested DNAper lane). NCI-H740 lacked one of the two hybrid

izing DNA bands, consistent with a partial deletion. No evidence of a genetic alteration was seenin other cell lines or in the lung tumor specimenLul5. SK-LC-5, not included in this figure,showed an identical pattern to the other SK-LC cell

lines.

Kb

9.4 -

6.6 -

4.4-

co co r*. r-

Ó Ó Ó Ó

CO CO CO CO CO

yC/D

CMCO|COzCO1"(5

OO)CMIIDiD

o^

u

The fifth category includes mutated gene products. The first example of this was found in a colon cancer library from which mutatedp53 was isolated. The finding that such an altered gene product can beisolated by SEREX reveals a previously unrecognized potential forSEREX, i.e., as a way to identify mutated gene products in cancer,such as oncogene or TSG products. NY-LU-12, defined in this study,is a clear candidate for inclusion in this category.

It has long been recognized that all small cell lung cancers and mostnon-small cell lung cancers have genetic alterations involving 3p21[see Kok et al. (9) for a recent review]. Several groups have documented this finding by showing loss of heterozygosity of specific3p21 markers (9, 26, 27). When microcell hybrids from humanchromosome 3p21 were introduced into the mouse fibrosarcoma lineA9 and injected into nude mice, they were observed to decrease tumorgrowth in vivo (28). Among the hybrids introduced into the cellscapable of causing tumor suppression, the one with the smallestchromosome 3 fragment corresponded to an approximately 2-MbDNA fragment, and thus, it was concluded that this chromosomalregion harbored tumor suppressor activity, hence, the designation TSGlocus (9, 28). Subsequently, four of five markers derived from this2-Mb stretch were found to be homozygously deleted in the NCI-H740 lung cancer cell line, correlating this deletion with tumorsuppressor activity (29). By comparing the 3p deletion breakpoints ofNCI-H740 and GLC20 lung cancer cell lines, the minimal commonregion that was deleted was narrowed to a 370-kb stretch at 3p21.3(15, 30, 31). The telomeric end of this deletion in the NCI-H740 linewas mapped to gene 16 (15), presumably the same gene as g16(GenBank accession no. U50839).4 Sequence analysis showed NY-

LU-12 to be identical to gl6, and full-length NY-LU-12 cDNA sequence was obtained, including coding regions absent in gl6. Ourdata from PCR and Southern analysis confirmed that the 5' region ofNY-LU-12/gl6 is intact in NCI-H740, whereas the 3' end is deleted.

The telomeric breakpoint of 3p21 in NCI-H740 was within a 350-bpcDNA segment of NY-LU-12. By developing a 450-kb PI contig,Todd et al. (32) identified an 80-kb PI clone (P294) from this regionwhich showed tumor suppressor activity in transfection studies. Although the 5' end of the NY-LU-12 seemed to be telomeric to thisclone, whether the 3' end of this gene is within this gene segment is

unclear.Well over 50 genes have been assigned to 3p21. Although several

of them have been analyzed as possible candidates for TSGs, including ACYI (amino acid acylase), UBE1L (ubiquitin-activating enzymeEl), CTNNB1(ß-cateningene), SEMA-IV (one of semaphorin genes),and so on, thus far, none of them has fulfilled the criteria for TSGs(9).Intriguingly, LUCA15, a gene located immediately centromeric togene 16 (15), encodes a protein which shares several functional

domains with NY-LU-12 and up to 30% homology at their COOHtermini (Fig. 1), indicating an evolutionary and likely functionallinkage between these two genes. The chromosomal location of NY-LU-12 to 3p21 suggested that it might be a TSG that was mutated inLul5 tumor cells, particularly in light of the following findings: (a) itis located within the 370-kb minimal-deletion segment of 3p21.3 andis partially deleted in NCI-H740, (b) its protein elicited a high-literIgG response in the Lu 15 patient and in two additional allogeneic lungcancer patients, and (c) it is predicted to code for a nuclear zinc-fingerprotein with RNA-binding domains. Despite vigorous comparisonbetween the Lu15 tumor and normal NY-LU-12 cDNA, however, wewere not able to demonstrate any evidence of mutation, deletion, orother genetic changes in NY-LU-12. Examination of a panel of lungcancer cell lines by Southern blot analysis also showed no evidence ofdetectable deletion or translocation, with the exception of NCI-H740.

Although no genetic change involving NY-LU-12has been found inthe Lul5 lung cancer, our analysis could have missed genetic alterations in this gene for at least several reasons.

(a) The immunoreactive NY-LU-12 cDNA clone may have represented a wild-type sequence derived from nonneoplastic cells in thetumor specimen or from the unaltered alíelein the tumor cells.Wild-type p53 antigen, for example, has been shown to react with thepatients' antibodies generated against mutated p53 (33, 34).

(b) SSCP analysis, the technique that we used to detect NY-LU-12mutations in the Lu15 tumor, has a sensitivity in the range of 80-90%(10) and false-negative results do occur, particularly when a mixedpopulation of cells are analyzed. For instance, the change(s) in theLulS tumor could have been masked by nonneoplastic cells and/or bythe unaltered alíelein the tumor cells.

(c) NY-LU-12deletions in other lung cancer cell lines may not have

been detected because of the choice of the probe and/or the restrictionenzyme.

(d) Alterations in the TSG may be at the genomic DNA level andnot reflected at the cDNA transcripts that we have analyzed, e.g., p53mutations in lung cancer have been reported in the intronic sequences(35, 36).

(e) The possibility of an aberrant cDNA product, e.g., a fusionproduct between NY-LU-12and other genes (such as the semaphorinl\/NY-LU-12 fusion cDNA reported in HT29), has not been entirelyruled out.

Given these reasons, other experimental approaches have to betaken to exclude mutations in NY-LU-12and to investigate its role asa TSG. Such studies would include mutational analysis using othertechniques (37), additional cloning and sequence analysis at thegenomic and cDNA levels, transfection studies to define the effect ofNY-LU-12 on tumorigenesis in vivo, knockout studies of the mouse

1040

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

IMMUNOGENIC LUNG CANCER ANTIGEN FROM CHROMOSOME 3P21.3

NY-LU-12 counterpart, and so on. Considering the high frequency of

3p21 changes in lung cancer and the fact that TSG(s) at this locus haveeluded extensive analysis of this region, experiments to further characterize the NY-LU-12 gene and its product need to be vigorously

pursued.

REFERENCES

1. Boon, T., and van der Bruggen. P. Human tumor antigens recognized by T lymphocytes. J. Exp. Med., 183: 725-729, 1996.

2. Rosenberg, S. A. Development of cancer immunothérapiesbased on identification ofthe genes encoding cancer regression antigens. J. Nail. Cancer Inst. (Bethesda). 88:1635-1644, 1996.

3. Çahin,U., Tiireci, O., Schmitt, H.. Cochlovius, B., Johannes. T., Schmits. R.. Stenner.F., Luo, G., Schobert. I., and Pfreundschuh. M. Human neoplasms elicit multiplespecific immune responses in the autologous host. Proc. Nati. Acad. Sci. USA, 92:11810-11813, 1995.

4. Chen, Y. T., Scanlan. M. J., Sahin, U., Türeci,0.. Cure, A. O., Tsang, S., Williamson,B., Stocken, E., Pfreundschuh, M., and Old. L. J. A testicular antigen aberrantlyexpressed in human cancers detected by autologous antibody screening. Proc. Nati.Acad. Sci. USA, 94: 1914-1918, 1997.

5. Robbins, P. F.. el-Gamil, M., Kawakami. Y.. Stevens, E., Yannelli. J. R., andRosenberg, S. A. Recognition of tyrosinase by tumor-infiltrating lymphocytes from apatient responding to immunotherapy. Cancer Res., 54: 3124-3126, 1994.

6. van der Bruggen. P., Traversari, C, Chômez, P., Lurquin, C., De Plaen, E., Van denEynde, B., Knuth, A., and Boon. T. A gene encoding an antigen recognized bycytolytic T lymphocytes on a human melanoma. Science (Washington DC), 254:1643-1647, 1991.

7. Brichard, V., Van Pel. A.. Wolfel, T., Wolfel, C., De Plaen, E., Lethe, B., Coulie, P.,and Boon, T. The tyrosinase gene codes for an antigen recognized by autologouscytolytic T lymphocytes on HLA-A2 melanomas. J. Exp. Med., 178: 489-495, 1993.

8. Cure, A. O., Türeci,O., Sahin, U.. Tsang. S.. Scanlan, M. J., Jager, E., Knuth, A.,Pfreundschuh, M., Old, L. J., and Chen, Y. T. SSX: a multigene family with severalmembers actively transcribed in normal testis and cancer. Int. J. Cancer, 72: 965-971,1997.

9. Kok, K., Naylor, S. L., and Buys, C. H. Deletions of the short arm of chromosome 3in solid tumors and the search for suppressor genes. Adv. Cancer Res.. 71: 27-92,1997.

10. Dracopoli. C. D., Haines, J. L., Korf, B. R., Moir, D. T., Morton, C. C., Seidman,J. G., and Smith. D. R. (eds.). Current Protocols in Human Genetics. Chapter 7.4.New York: John Wiley & Sons. Inc., 1997.

11. Kurdi-Haidar, B., Aebi, S., Heath, D., Enns, R. E., Naredi, P., Horn, D. K., andHowell, S. B. Isolation of the ATP-binding human homolog of the arsA componentof the bacterial arsenite transporter. Genomics, 36: 486-491. 1996.

12. Misaki, Y., Pruijn. G. J., van der Kemp, A. W.. and van Venrooij, W. J. The 56Kautoantigen is identical to human annexin XI. J. Biol. Chem.. 269: 4240-4246, 1994.

13. Misaki, Y.. Van Venrooij, W. J., and Pruijn, G. J. Prevalence and characteristics ofanti-56K/annexin XI autoantibodies in systemic autoimmune diseases. J. Rheumatol.,22: 97-102, 1995.

14. Refaeli, Y., Levy, D. N., and Weiner, D. B. The glucocorticoid receptor type IIcomplex is a target of the HIV-1 vpr gene product. Proc. Nati. Acad. Sci. USA, 92:3621-3625, 1995.

15. Wei, M. H., Latif, F., Bader. S., Kashuba, V., Chen. J. Y., Duh, F. M., Sekido, Y.,Lee, C. C., Geil, L., Kuzmin, 1., Zabarovsky, E., Klein, G., Zbar, B., Minna, J. D.. andLerman. M. I. Construction of a 600-kilobase cosmid clone contig and generation ofa transcriptional map surrounding the lung cancer tumor suppressor gene (TSG) locuson human chromosome 3p21.3: progress toward the isolation of a lung cancer TSG.Cancer Res.. 56: 1487-1492. 1996.

16. Coleman, M. P., Ambrose, H. J., Carrel, L., Nemeth, A. H., Willard, H. F.. andDavies, K. E. A novel gene, DXS8237E, lies within 20 kb upstream of UBE1 inXpll.23 and has a different X inactivation status. Genomics, 31: 135-138, 1996.

17. Inoue, A., Takahashi, K. P., Kimura, M., Watanabe, T., and Morisawa, S. Molecularcloning of a RNA binding protein. Sl-1. Nucleic Acids Res., 24: 2990-2997. 1996.

18. Witte, M. M., and Scott, R. E. The proliferation potential protein-related (P2P-R1

gene with domains encoding heterogeneous nuclear ribonucleoprotein association andRbl binding shows repressed expression during terminal differentiation. Proc. Nati.Acad. Sci. USA, 94: 1212-1217, 1997.

19. Wilson, R., Ainscough, R., Anderson, K., Baynes, C., Berks. M., Bonfield. J., Burton.J., Connell, M., Copsey. T., Cooper, J., et al. 2.2 Mb of contiguous nucleotidesequence from chromosome III of C. elegans. Nature (Lond.). 368: 32-38, 1994.

20. Seraphin. B.. Simon. M., Boulet. A., and Faye. G. Mitochondria! splicing requires aprotein from a novel helicase family. Nature (Lond.), 337: 84-87, 1989.

21. Takagaki. Y., MacDonald, C. C., Shenk, T., and Manley. J. L. The human 64-kDapolyadenylylation factor contains a ribonucleoprotein-type RNA binding domain andunusual auxiliary motifs. Proc. Nati. Acad. Sci. USA, 89: 1403-1407, 1992.

22. Brass, N.. Heckel, D.. §ahin.U., Pfreundschuh, M.. Sybrecht, G. W., and Méese,E.Translation initiation factor eIF-4-y is encoded by an amplified gene and induces an

immune response in squamous cell lung carcinoma. Hum. Mol. Genet.. 6: 33-39,

1997.23. Schapira, F. Isozymes and cancer. Adv. Cancer Res., 18: 77-153, 1973.24. Ojika, T.. Imaizumi. M.. Abe. T., and Kalo, K. Immunochemical and immunohisto-

chemical studies on three aldolase isozymes in human lung cancer. Cancer (Phila.l.67: 2153-2158, 1991.

25. Asaka, M., Kimura. T., Meguro. T., Kalo, M.. Kudo, M., Miyazaki. T.. and Alpert, E.Alteration of aldolase isozymes in serum and tissues of patients with cancer and otherdiseases. J Clin. Lab. Anal., 8: 144-148, 1994.

26. Naylor. S. L.. Johnson. B. E.. Minna. J. D., and Sakaguchi, A. Y. Loss of heterozy-gosity of chromosome 3p markers in small-cell lung cancer. Nature (Lond.), _?29:451-454, 1987.

27. Yokota. J., Wada, M., Shimosato, Y., Terada, M., and Sugimura, T. Loss of het-erozygosity on chromosomes 3. 13, and 17 in small-cell carcinoma and on chromosome 3 in adenocarcinoma of the lung. Proc. Nail. Acad. Sci. USA, 84: 9252-9256,1987.

28. Killary. A. M.. Wolf. M. E.. Giambemardi, T. A., and Naylor, S. L. Definition of atumor suppressor locus within human chromosome 3p2l-p22. Proc. Nati. Acad. Sci.USA. 89: 10877-10881, 1992.

29. Daly. M. C., Xiang, R. H.. Buchhagen. D.. Hensel. C. H., Garcia. D. K.. Killary.A. M., Minna, J. D., and Naylor, S. L. A homozygous deletion on chromosome 3 ina small cell lung cancer cell line correlates with a region of tumor suppressor activity.Oncogene, 8: 1721-1729, 1993.

30. Kok, K., van den Berg, A., Veldhuis, P. M., van der Veen, A. Y., Franke, M.,Schoenmakers. E. F., Hulsbeek, M. M., van der Houl. A. H., de Leij, L., van de Ven,W., and Buys. C. H. A homozygous deletion in a small cell lung cancer cell lineinvolving a 3p2I region with a marked instability in yeast artificial chromosomes.Cancer Res., 54: 4183-4187, 1994.

31. Roche, J., Boldog, F., Robinson, M., Robinson, L., Varella-Garcia, M.. Swanlon. M..

Waggoner. B.. Fishel, R.. Franklin, W., Gemmili. R.. and Drabkin. H. Distinct 3p21.3deletions in lung cancer and identification of a new human semaphorin. Oncogene,12: 1289-1297, 1996.

32. Todd, M. C., Xiang, R. H., Garcia, D. K., Kerbacher, K. E., Moore, S. L., Hensel.C. H., Liu, P., Siciliano, M. J., Kok, K., van den Berg. A.. Veldhuis, P., Buys, C. H.,Killary, A. M., and Naylor. S. L. An 80-kb PI clone from chromosome 3p21.3suppresses tumor growth in \'i\'tt. Oncogene. 13: 2387-2396. 1996.

33. Baker, S. J., Preisinger. A. C.. Jessup, J. M., Paraskeva, C., Markowitz. S.. Willson.J. K., Hamilton, S.. and Vogelstein, B. p53 gene mutations occur in combination with17p allelic deletions as late events in colorectal lumorigenesis. Cancer Res., 50:7717-7722, 1990.

34. Bennett, W. P., Hollstein, M. C., He, A., Zhu, S. M., Resau. J. H., Trump, B. F.,Metcalf, R. A.. Welsh. J. A.. Midgley, C., Lane. D. P.. and Harris, C. C. Archivalanalysis of p53 genetic and protein alterations in Chinese esophageal cancer. Oncogene, 6: 1779-1784. 1991.

35. Takahashi, T.. D'Amico. D.. Chiba. !.. Buchhagen. D. L., and Minna. J. D. Identifi

cation of intronic point mutations as an alternative mechanism for p53 inactivation inlung cancer. J. Clin. Invest., 86: 363-369, 1990.

36. Bodner, S. M.. Minna. J. D.. Jensen, S. M., D'Amico. D., Carbone. D.. Mitsudomi.

T., Fedorko, J., Buchhagen, D. L., Nau, M. M.. Gazdar, A. F., and Linnoila, R. I.Expression of mutant p53 proteins in lung cancer correlates with the class of p53 genemutation. Oncogene. 7: 743-749, 1992.

37. Mashal, R. D., and Sklar, J. Practical methods of mutation detection. Curr. Opin.Genet. Dev.. 6: 275-280, 1996.

38. Yokoyama. N.. Hayashi, N., Seki. T.. Panie, N.. Ohba. T.. Nishii, K.. Kuma, K.,Hayashida. T.. Miyata, T., Aebi, U., Fukui, M., and Nishimoto, T. A giant nucleoporeprotein that binds Ran/TC4. Nature (Lond.), 376: 184-188, 1995.

1041

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

1998;58:1034-1041. Cancer Res   Ali O. Güre, Nasser K. Altorki, Elisabeth Stockert, et al.   Suppressor Gene Locus on Chromosome 3p21.3Antibodies: Definition of a Novel cDNA Derived from the Tumor Human Lung Cancer Antigens Recognized by Autologous

  Updated version

  http://cancerres.aacrjournals.org/content/58/5/1034

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/58/5/1034To request permission to re-use all or part of this article, use this link

on March 31, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from