Genomic Organization, 5'-Upstream Sequence, and Chromosomal ...

8 downloads 0 Views 2MB Size Report
regulatory elements may account for IA-1 gene expres- sion in different ... insulinoma-associated cDNA, IA-1,' that encodes a protein with zinc finger ... Library-A human liver genomic library in A GEM-11 was purchased from Promega ...
THEJOURNALOF BIOLOGICAL. CHEMISTRY

Vol. 269,No. 19,Issue of May 13, pp. 14170-14174, 1994 Printed in U.S.A.

Genomic Organization, 5’-Upstream Sequence, and Chromosomal IA-1” Localization of an Insulinoma-associated Intronless Gene, (Received for publication, December 8, 1993)

Michael S . Lanf, Qing Li, Jia Lu, William S. Modis, and Abner L. Notkins From the Laboratory of Oral Medicine, NIDR, National Institutesof Health, Bethesda, Maryland 20892 and the $Biological Carcinogenesis Development Program, Program Resources, Inc. IDynCorp, Frederick Cancer Research and Development Center, Frederick, Maryland 21 702-1201

IA-1 is a novelcDNA originally isolated froma human normal tissues (including pancreas, testes, lymph node,brain, insulinoma subtraction library (ISL-153). It encodes a lung, liver, stomach, spleen, thyroid, kidney,and colon). Cliniprotein containing both a zinc finger DNA-binding do- cal studies on a panel of 64 human lung cancer cell lines by IA-1 tran- Northern analysis demonstrated mainand aputativeprohormonedomain. that IA-1mRNA was detected scripts have been found thus far only in tumors of neuin 97% (30/31)of small celllung cancer cell lines and 13% (430) roendocrine origin. Clinical studies have shown that of non-small cell lung cancer cell lines (Lan et al., 1993). The IA-1 is a sensitivemarker for neuroendocrine differen- IA-1-positive non-small cell lung cancer cell lines were shown tiation of human lung tumors. to have neuroendocrine features as evidenced bythe expression IA-1 of one or more known neuroendocrine markers, such as chroIn this study, we cloned and sequenced the entire geneand its 5‘-upstreamregionfrom a humanliver mogranin A and L-dopa decarboxylase. The restricted expresgenomic library.In situ hybridization localized theIA-1 sion of the IA-1 gene in neuroendocrine tumors suggests that genetotheshortarmofhumanchromosome 20. SeIA-1 gene expression may be under the tight controlof certain quence analysis and restriction enzyme mapping IA-1 gene is uninterrupted and appears tissue-specific regulatory elements. showed that the the IA-1 gene from IA-1 is an intronless gene In this study, we cloned and characterized to be intronless. Evidence that in a human liver genomic library. We mapped, by restriction enthat can translate into protein was obtained from that contained the IA-1 gene, vitro translationstudiesthatshowedthatboth IA-1 zymes, a 13-kb genomic fragment cDNA and IA-1 genomic DNA yielded identical protein sequenced the entire gene and its 5”upstream region, and lo20 by fluorescencei n products of approximately 61,000 daltons. Examination calized t h e IA-1gene to chromosome p11.2 (2090 base pairs) revealed sev-situ hybridization. of the 5”upstream region eral tissue-specific regulatory elements, including gluMATERIALSANDMETHODS cokinase upstream promoter elements and a Pit-1 factor binding site. The presence of several different upstream Screening of IA-1 Genomic Clones from Normal Human Liver regulatory elements may account for IA-1 gene expres- Library-A human liver genomic library in A GEM-11 was purchased from Promega (Madison, WI). This library contains 9-23-kb fragments sion in different neuroendocrine tumors.

of normal human genomic DNApartially digested with Sau3AI and two bases filled in by Klenow fragment. The total primary plaque-forming units of this library is 2.0 x lo6,and the final amplified titer is 1.65 x lo9 We have recently reported the sequence of a novel h u m a n plaque-forming units/ml. The library was screened by the standard insulinoma-associated cDNA, IA-1,’ that encodes a protein plaque hybridization method (Sambrooket al., 1989). Briefly,duplicate with zinc finger DNA-binding motifs (Goto et al., 1992). This filters were lifted, baked under vacuum at 80 “C for 1 h, and washed unique cDNA has revealed several interesting features, which with 5 x SSC for 30min. The filter was prehybridized a t 50 “Cwith 40% formamide, 5 x SSC, 10 pg/ml sheared salmon sperm DNA, 6 x Deninclude a putative prohormone domain with several dibasic hardt’s solution (5 Prime + 3 Prime, Inc., Boulder, CO) for 6 hfollowed amino acids, an amidation signal sequence (Pro-Gly-Lys-Arg) by hybridization in the same solution with a randomly primed labeled at the amino terminus, a n d five Cys,-His, zinc finger DNA- (lo6cpdml) cDNAprobe, LA-1-34 (Goto et al., 1992).Hybridization was binding motifsat t h e carboxyl terminus. Expression of t h e IA-1 carried out at 50 “C for 16 h, and the filter was washed with 1x SSC, gene was detected by Northern analysisin five of five h u m a n 0.1% SDS at the same temperature. Positive plaques from duplicate insulinoma tissues, in all of the murine insulinoma cell lines filters were subjected to secondary and tertiary screening. DNA Sequence Analysis and Restriction Enzyme Mapping-A posi(including mouse, rat, and hamster), and in other neuroendo- tive genomic clone, containing a 13-kb insert, was digested with Sac1 as pituitary tumor, pheochromocy- and subcloned into the pBlueScriptI1(sk+)vector (Stratagene, La Jolla, crine tumor cell lines (such toma, medullary thyroid carcinoma, and small cell lung carci- CA). Plasmid DNA was used for double strand DNA sequencing. DNA sequencing was performedusing Sequenase T4 DNApolymerase under noma), but it was not detected in a variety of other tumors (such as melanoma, breast carcinoma, thyroid carcinoma,glio- conditions recommendedby the supplier (U.S. Biochemical Corp.). Inblastoma, pancreatic carcinoma, and choriocarcinoma) or in ternal sense and antisense strand primers were synthesized byBiosynthesis, Inc. (Lewisville,TX). Both strands were sequencedand analyzed using the Geneworks 2.1 software package (IntelliGenetics, * The costs of publication of this article were defrayed in part by the Mountain View, CAI.The current FASTA data base was used for searchpayment of page charges. This article must therefore be hereby marked ing both nucleic acid and protein sequence similarities (Pearson and “aduertisement” in accordance with 18 U.S.C.Section1734solelyto Lipman, 1988). Restriction enzyme mapping of the isolated genomic indicate this fact. The nucleotide sequence($ reported in this paperhas been submitted clone was performed by the method of Whittaker and Southern (1986) to the G’enBankTMIEMBL Data Bank with accession nurnbeds) U07172. with some modifications. Briefly, plasmid DNAfrom the pL3S5.6 or pL3S4.9 clone was dissolved in 100 1.11of 1 x restriction buffer contain$ To whom correspondence should be addressed: Laboratory of Oral ing 10 mM magnesium chloride.Twenty-pl aliquots were W crossMedicine, NIDR, NIH, Bldg. 30, Rm. 121, Bethesda, MD 20892. The abbreviations used are: IA-1, insulinoma-associated cDNA-1; linked for 0, 1, 15, 45, or 60 min using Stratalinker (Stratagene) and bp, base pair(s); kb, kilobase(s);PIPES, 1,4-piperazinediethenesulfonic subjected to appropriate enzyme digestion. The sample was heated at 65 “C for 15 min and run on a 1%agarose gel for Southern analysis. acid.

14170

Intronless

IA-1

0

Gene

14171

10

5 I

I

-2090

+1

13 kb

I dSacl2-IA-1

I

-145

pL3S5.6 pL3s2.0

pL3s4.9

FIG.1. Genomic organization of the LC\-1 gene. The restriction enzyme map of a 13-kb genomic clone containing the IA-1 gene was determined by a combination of sequencing and restriction enzyme mapping with Sad, EcoRI, PstI, HindIII, and BarnHI a s described under "Materials and Methods." The 13-kb fragment was digested with restriction enzyme Sac1 and subcloned into the pBlueScriptI1 vector. Four clones, designated pL3S4.9, pL3SO.8, pL3S2.0, and pL3S5.6, were generated. The rectangular box shows the location of IA-1 cDNA; the shaded area represents thecoding sequence, and the open areas represent the5'- and 3'-untranslatedregions. The 5"upstream sequence (from+1 to -2090 bp) of the IA-1 gene is shown inFig. 6. A clone, dSacl2-IA-I, was prepared asa DNA template for in vitro translation study.No intron was found in the IA-1 gene. EcoRl A

B

C

D

Hindlll E

F

A

B

C

D

E

F

-200

-97.4 - 7 Kb

-6 -5 -4

-68

=3 -2

- 43

FIG.2. Southern analysis.Fifteen pgof genomic DNAisolated from 16 individuals were digested with restriction enzymes EcoRI or HindIII andelectrophoresed on a 0.8% agarose gel. Denatured DNAs were transblotted to Nytran and hybridized with ""P-labeled full-length IA-1 cDNA. The blot was washed under high stringency and exposed a t -70 "C for 3 days. Six samples (A-F) are shown. Hybridization was performeda t 39 "C for 18 h in a solution containing 6 x Denhardt's solution, 10pg/ml sheared salmon spermDNA, 2 x SSC, and loGc p d m l end-labeled probe. Four oligonucleotide probes (TGCCCACTAGTAACTAAC, AGTGGGAAAAGGAVCCG, TGGGGGACTCATCVATC, AAGAAGCCCAAGGCCAT), located a t bothends of pL3S5.6 and pL3S4.9, were end-labeled by T4 polynucleotide kinase (Life Technologies, Inc.) and used to probe partially digestedDNAs. Southern Analysis-Total genomic DNAs were prepared from 16 individuals by standard methods (Sambrook et al., 1989). Fifteen pg of DNA were digested with restriction enzymes EcoRI or HindIII and

FIG.3. In vitro translation. IA-1 cDNA, clone IA-1-18 (lanes 1 and 2 ) and IA-1 genomic DNA, clone dSacl2-IA-1 (lanes 3 and 4 ) were subcloned intothe pBlueScriptI1 vector. Capped mRNAs of sense (lanes 1 and 3 ) and antisense (lanes2 and 4 ) were synthesized in an in vitro transcription system. One pg of each transcript was translated in a rabbit reticulocyte lysate with I"'S1cysteine. Translated products were separated on a 7.5%)SDS-polyacrylamide gel electrophoresis gel, fixed with autoradiography enhancer, and the film was exposed overnight. Thetranslatedproteinismarked by an arrowalongwithprotein markers. electrophoresed on a 0.8% agarose gel. Denatured DNAs were transblotted to Nytran via capillary blotting and hybridized with a "P-labeled cDNA IA-1-18 probe (Goto et al., 1992). Hybridization was performed at 50 "C for 18 h in a solution containing 40%. formamide, 5 x SSC, 10pg/ml sheared salmon sperm DNA, 6 x Denhardt's solution, and loGc p d m l radiolabeled probe. The blot was washedtwice in 0.1x SSC, 0.1% SDS a t 60 "C (high stringency). Autoradiograms wereexposed a t -70 "C for 3 days. RNase Protection Assay-A genomic subclone (dSac3BE), including both IA-1 cDNA and 5"flanking regions, was prepared from BglI digestion and 5"deletion of pL3S2.0. "'P-Labeled antisense riboprobe (ap-

14172

Intronless

IA-1

Gene 1

2

3

nt

121.

102. +

81-

69-

FIG.5. RNase protection assay. A :'ZI"labeled antisense riboprobe was prepared by T7 RNA polymerase transcription from a genomic subclone, dSac3BE. Theriboprobe was purifiedfrom a 6% urea gel, and 1 x 10' cpm riboprobe was mixed with 50 pgof NCI-H69 RNA (lane 11, HeLa RNA (lane 21, or yeasttRNA (lane 3 ) for overnight hybridization. The hybridized mixture was subjected to RNase digestion and analysis on a 8% denaturing sequencing gel. An estimated 100-nucleotide (nt) long fragment (lane 1 ) was protected with small cell lung carcinoma RNA (arrow). Riboprobe size markers were prepared from the same vector. probe (20 ng/p1) was hybridized with metaphase chromosomes overnight at 37 "C in a solutionof 50% formamide and 10% dextran sulfate in 2 x SSC. The slides were washed with 50% formamide in 2 x SSC followed by 2 x SSC alone, both at 40 "C. Detection was carried out usingavidin-fluoresceinisothiocyanate. Chromosome identification was performed following simultaneous staining with Hoechst33258.

Flc:. 4. Chromosomal localization.A , metaphasc ~hr~mosomes filllowing fluorescence in s i t u hybridization with the genomic IA-1 probe. Arrows indicate the signala t 2 0 ~ 1 1 . 2R., simultaneous Hoechst 33258 staining and QFH-bandedchromosomes.

RESULTS AND DISCUSSION

Genomic Structure and Nucleotide Sequence of the IA-1 Gene-In this study, we report the cloning and sequencing of proximately 400 nucleotides)wasprepared by T7 RNA polymerase the IA-1 gene from a human normal liver genomic library. By transcription and subjected 6% to urea gel purification. Theriboprobe (1 screening 2 x lo5recombinants with IA-1-34 cDNA probe, we isolated two identical genomic clones (A 3) containing an apx IO5 cpm) was mixed with 50 pgof HeLa RNA or NCI-H69 small cell lung carcinoma RNA in a hybridization buffer (40 mM PIPES, pH 6.4, proximately 13-kb insert. The A 3 genomic clone was digested 50% formamide). The reaction mixture was heated to 90 "C for 5 min with Sac1 and subcloned into the pBlueScriptI1 (sk+) vector. and slowly cooled down to 45"C for overnight incubation. RNasedigesFour subclones, designated pL3S4.9, pL3SO.8, pL3S2.0, and tion was carried outby adding 300p1 of RNase solution (300 mM NaCI, pL3S5.6, were generated (Fig. 1).The IA-1 gene and the 5'10 mhf Tris-C1, pH 7.4,5 mM EDTA, 2 pg/ml RNase T1,40 pg/ml RNase A) a t room temperature for 10 min. The reaction was stopped by adding upstream flankingregion (2090 bp) were completely sequenced 20 pl of 1% SDS and5 p1 of 10 mg/ml protease K a t 37 "C for 15 min. The from pL3SO.8, pL3S2.0, and the 5"region of pL3S5.6. Our exreaction mixture was extracted with 400 p1 of phenokhloroform twice periments showed that the nucleotide sequence derived from and ethanol precipitated. The protected fragment was analyzed on a n normal human genomic DNA was identical to the IA-1 cDNA 8% denaturing sequencing gel. sequence derived from human insulinoma mRNA (Goto et al., In Vitro IFansZation-An EcoRI fragment (6.5 kb) from genomic IA-1 DNA was subcloned into the pBlueScriptI1 (sk+)vector (Stratagene). 1992). In addition, we have completed the restriction map of the A 3 genomic clone by performing restriction enzyme mapSerialdeletionwas performed to eliminatethe5'-untranslatedsequence. Oneclone (designated dSacl2-IA-1,starting at position -145 bp ping on pL3S5.6 and pL3S4.9. Both sequence analysis and and extending to the EcoRI site in the 3"region of IA-I gene) was chosen restriction enzyme mapping indicated that theIA-1 gene is not for in vitro translation study. Both dSacl2-IA-1 and IA-1-18cDNA, interrupted and, therefore, appears to be an intronless gene containing thecomplete IA-1 open reading frame, were linearized with (Fig. 1). either BarnHIor XhoI restriction enzyme. CappedmRNA (Stratagene) The existence of introns has been reported in most of the or T7 RNA polymerase to produce was synthesized using either T3 either the senseor antisense transcript. One pgof each transcript was genes of eukaryotic organisms. However, some important eukaryotic genes,such as seven transmembrane receptors (Allard addedtoarabbitreticulocyteinvitrotranslationreactionmixture (Promega) in the presence of [?31cysteine (Amersham Corp.) a t 30 "C et al., 1987; Kobilka et al., 1987a, 1987b; Young et al., 1986; for 1 h. Twenty pl of reaction mixture was run on 7.5% SDS-polyacryl- Sunahara et al., 1990; Demchyshyn et al., 1992; Nakayama et amide gel electrophoresis and fixed with autoradiography enhancer al., 1985), human interferon-a and $1 (Nagata et al., 1980; (DuPont NEN). The gel was exposed to the film overnight a t room Lawn et al., 1981), type X collagen (Ninomiya etal., 1986), and temperature. of genes FZuorescence In SituHybridization-The procedure utilizedhas been histone genes (Stein, 1984) are intronless. The number compared with those with introns. reported by Tory et al. (1992).Briefly, biotinylated IA-1 genomic DNA without introns is small as

IA-1 Intronless Gene

14173

UPB

-1990

TTTTGGGGAG TTACTTTTAG ATAGMATATGTTTATMATTTTAGATMACTCATTTTMGATTTCTMC GHP - 1 ACTAGCACTC ACCACAGCAT TTTTMAACT AGGGTGAGAC TTTTCAAMT A T A T T T C A C T T C TTCCCTTTTC AGTGTGTGAG TTMAATATT

-1890

A C M A C A T MM T T A T C A M

-1790

m T G G A C T CTATTTTTAC ATGTTTAATT TCTACGCCAG GGTATTTTCA G A G G G W TC C M A A C M TT A M T A C C T T

-1690

AAMCTCAGCMAATCTGTCTGGTATACM

-1590

GTGGCTTGGC TATTTGTTCC CCAGTGTTGGACGATCAGGG

CCACTACGCA CTCGTGGGGT TCTCTGCCCT GTCCCCCGAA GATTCTGCCT TTTGTGTTTC

-1490

CTCGGGGCTGGGCTGGACAGAGGATGGAGGCAGGCGGCCC

-2090

GAGCTCAAM GAGCAAGGM AT”

UPE TATCTTTCCC TTGATTMCC AACCTGATTA A C C C C T T M TT M T T M G C CT M C M C T G T

CCC-

AP-1 A C T T W

CAGTGGAATT M G C G T M G T

AGCTTTGCAG AAAAATGTTT CCTTCTCTTT C M T A M G T G ACATTTTCCT CACAGCATAT GACATTTTAT

-1390

TTCGCTGTTG TCTCTTCCTG GTGGCACAGG GGAGGCGGCT

AGTCTGGGTC C A M C G G M C GGCAGCGGAG GTGGGGGTGG GGGTGGGGGC GGGGGGAATC T-Ag TGTCCTCTCG GAGGCCTCAG CCTGCCTCAG AGTACAGATT GCCCCCCCTC CCCCCGTCCA

-1290

GCACGCGTCT CTTTTGCGTC CAGATTGGCC GCGACCGGAG

CTCMTAGCAGGAGGTTMT

-1190

TGCGCCAGAC AGGCAGCCCC TTTATTTCGC AGCGCCTTGA TTGGAGCCCT TGATTTAGCA TCTGATGTCA ACCGGCAAAC W T G C C C G CTCGGMTGA T-Ag T-Aa CGAGGCCGAC CGCGCGTTTC GGGCTTCGGC CMCTCGACCCGGATTMCC AAATGCATGA GGCCGCGCGG GGAGGAGAAA GCCMCTTCA CTGGGGCGCA

-1090

TTCCTTCACA M G G T G M G G GGGCACGGCT CCGTGGGGGC

GCCCCGCCM AGATGCCATT CCCMGCCTCCMGAACCCAMAGTTCAGAMTTCACGTT

-990

GAGCCTTGGACCACGCGGAAGCTCCGGGCGGCCTGGGTCG

-890

CCTCTGGGM CCCAGCCCAG CCCGCCCGCG

-790

TCTAGCCCCGTGGGCGCGCGGAGGCTGCGAGCACAAACAT

-690

CTTTTGCTTG GCCTCTTGGG TCCAGCGGCC CATCCGTCCA AGGTCCGGGCGGAGGCGTCC

-590

CACCACGCGGACGCACTGCG

CGCCCCCACC CAATCAGCGC CCTCCGCMC GATCTCCCCC GCASCGCCCC GGCGCGCCCC

-490

CCGGCTCGCG GATTGAACCC

TCCTGACATA TTTGGGGCCA TTCTTCTCCTTTGTTGCTAT

-390

GCGTGCATTG TCGCGCTGAT GGACGGGCCC ATTTGGCGGC TCCGCGCCCC CCGGAGGAGA GACACAMGC CCAGGCACGT GCGCCTCCCC ATAGAGMGC

-290

AGCAGACCGT GMGGGAGGC GGGGCCGGGC GTGTGCCTGG

ACCGGGCGV GCGW GGCGC COGGCOGGGC GACCAGGGGC GCGCGCGGGG GCCCCGCGCC

-190

CTCAGGTACA TCTGCCGCAC CTACCGGGCG ACCCCCGAGT

CCCGGCCCCC TTTTGGCCGC CCCATCGCCC TCCCACCCTG CCAGGCTGAG GAGCTGCGGA

CTCCMTCCC GCGCTTTCGG GAGACTGAGAAGGCCGTGCC CGCCCTCGGC CACTGCCAGA AGGCCGGGCC

M T T C C T T C T CAAACTCGAAAGAAACCTTC CCCTGTCCAC ACTTGGMCC CCGGGGAACC

SP- 1

SP-1

SP-1

SP-1

- 90 13 111

CGCGCTGATT GGCTCCAGGG GAAGCGGGAG

SP-1

GCACMCTTC CCCCTCGGCT

TTTGCTAGCG ACCCGCGGGT MTCCCCGCG CGGGAGGGGG

SP-1

.... ..

?+I

GACGCGCGTG GCGAGAACM TGGCCCCCTC CCCCCGTTM MGGGAGCGG CTGCCGGGCC CGGGGACAGG

CAGGGCGCAGAGCTGGGCCGAGCCGTCGCCGGCGCCACGC AGCGCGGAGG GGGGACCGAG

GGACCCTGCT GCTCTCTCGG ATTCTTGTTT ATTTCCCMA

GAGTCCCGCA GCCGCCGCGC CCGGGCAATG GGCCGGGGGC ACTGAGGGCC GCCGGGGCCG

CCAGTGCCGT GCCCTCGGGC C G C G C G A A C A C C C G C G G CTTCCTGGTG MGCGCAGCA AG

FIG.6. Nucleotide sequenceof the 5‘-endof the human IA-1 gene. Transcription initiation site is assigned to position +l.An asterisk (*) indicates the TATA box-like sequence. Conserved sequences of known regulatory elements, SP-I, T-Ag,A P - I , upstream promotor elements (UPE), and GHF-1 are underlined.

The studies described here show that IA-1, which encodes a zinc finger DNA-binding protein, belongs to this smallgroup of genes that are intronless. Southern analysisof the IA-1 gene (Fig. 2) revealed a single band (6.5 kb) withEcoRI digestion and two bands (3.8 and 0.75 kb) with HindIII digestion. These findings are consistent with the restriction enzyme mapping data shown in Fig. 1 and suggest that the IA-1 is probably a single copy gene. DNAs isolated from a total of 16 individuals (notshown) digested with either EcoRI or HindIII restriction enzymes did not show any evidence of restriction fragment lengthpolymorphisms. In Vitro Dunslation Products from IA-1 cDNA and IA-1 Genomic DNA-To show that both IA-1 cDNAand IA-1 genomic DNA can translate the same proteinproduct without RNA splicing, IA-1 cDNA from position +12 to the poly(A) tail (IA1-18) (Goto et al., 1992) and an IA-1 genomic DNA clone (dSacl2-IA-1) containing 304 bp 5‘-upstream from the ATG start codon were used as DNA templates for synthesizing RNA transcripts. Using the rabbit reticulocyte system for in vitro translation, a prominent proteinproduct with an estimatedM , of 61,000 was produced by both IA-1 cDNA and the intronless IA-1 genomic DNA (Fig. 3). Antisense transcripts revealed no protein product. This demonstrates that the intronless IA-1 gene iscapable of translating into the same protein product as the IA-1 cDNA. Mapping of the ZA-1 Gene to Chromosome 20-The fact that IA-1 is a sensitive markerfor small cell lung cancer (Lan etal., 1993) and hasa restricted tissue distribution(Goto et al., 1992) makes it important to determineits chromosomal localization. Using fluorescence in situ hybridization, a total of 46 metaphase cells were examined. In 42 of these cells, from one to

four chromatids were labeled at 20~11.2(Fig. 4). No significant background was observed a t other chromosomal positions. These results suggest that the IA-1 gene is a single copy gene located on the short arm of chromosome 20. Several hereditary diseases havebeen mappedto the short arm of chromosome 20. Thus far, only one disease, arteriohepatic dysplasia, has been localized specifically t o the p11.2 region (Teebi et al., 1992; Simpson, 1988). Characterization of 5’-Upstream Sequence of the IA-1 GeneThe transcription initiation site of the IA-1 gene was determined by RNase protectionassay using total RNAisolated from both a small cell lung carcinoma cell line (NCI-H69) and HeLa cells. A gel-purified antisense riboprobe extending from 459 to 60 bp upstream of the ATG start site was hybridized with50 pg of NCI-H69 RNA, HeLa cell RNA, or yeast tRNA. As shown in Fig. 5, an estimated 100-nucleotide long fragment (lane1) was protected specifically with small cell lung carcinoma RNA, which strongly expressed IA-1 message, whereas HeLa RNA and tRNA did not protect the probe (lanes 2 and 3 ) . The transcription start site was estimated a t 159 bp upstream of the ATG start codon (designated as +I in Fig. 61, which is 12 bp longer than the IA-1 cDNA reported previously (Goto et al., 1992). At the vicinity of the translation start site, there is a stretch of GC-rich sequence (>80%), which might form secondary structure that could hamper the proper primer extension analysis. In fact, we made several primers but were unable to extend them to the correct transcription initiation site as determined by RNase protection assay. The5”upstream DNA sequence, extending 2090 bp upstream from the transcription start site, was determined by double-stranded sequencing of pL3S2.0 and pL3SO.8 genomic

14174

IA-1 Intronless Gene

subclones (Fig. 6) and compared with known sequences in the lung carcinoma, pheochromocytoma, and medullarythyroid GenBank DNA data base (Pearson and Lipman, 1988). No sig- carcinoma. nificant similarity wasfound except for a number of conserved Acknowledgments-We thank Dr. Y. Goto for screening the genomic regulatoryelements.The classic TATAbox (McKnight and and Drs. P. Zhou, M. F. Young, and R. R. Franks for advice and Kingsbury, 1982) typically located between positions -25 and library critical reading of the manuscript. We gratefully acknowledge the ex-35 bp was not found; however, an AT-rich sequence located cellent editorial help of Eloise Mange. between positions -29 and -34 bp, which might serve as a TATA box, was observed. The so-called CAATT box, which is REFERENCES often found at -80 bp, was absent. Three putative transcription Allard, W. J., Sigal, I. S., and Dixon, A. F. (1987) Nucleic Acids Res. 15, 10604 factor SP-1 binding sites (GC boxes) located between positions Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra, R. J., Rahmsdorf, H. J.,Jonat, C., Herrlich, P., and Karin, M. (1987) Cell 49,729-739 -220 and -250 bp (Kadonaga et al., 1987) and three additional Demchyshyn, L., Sunahara, R. K., Miller, K., Teitler, M., Hoffman, B. J., Kennedy, GC boxes further upstream in reverse orientation also were J. L., Seeman, P., VanTol, H. H. M., and Niznik, H. B. (1992)Proc. Natl. Acad. found. In this context, the promoters of several housekeeping Sei. U. S. A . 89, 5522-5526 genes thatlack the characteristicTATAand CAATT boxes (Mel- Fanning, E., and Knippers, R. (1992) Annu. Rev. Biochern. 6 1 , 5 5 4 5 Y., DeSilva, M. G., Toscani, A., Prabhakar, B. S., Notkins, A. L., and Lan,M. ton et al., 1984; Reynolds et al., 1984; Valerio et al., 1985) are Goto, S. (1992) J. Bid. Chern. 267,15252-15257 known to havea similarly high GC content. Further upstream, Ingraham, H. A., Chen, R., Mangalam, H. J., Elsholtz, H. P., Flynn, S. E., Lin, C. R., Simmons, D. M., Swanson, L., and Rosenfeld, M. G. (1988) Cell 55,519-529 three SV40 T antigen binding sites (Jones and Tjian, 1984)and Jones, K. A,, and Tjian, R. (1984) Cell 36, 155-162 an AP-1 (Lee et al., 1987a, 198713) binding site were found. The Kadonaga, J. T., Carner, K. R., Masiarz, F. R., and Tjian, R. (1987)Cell 51, 1079I n9o SV40 T antigen binding site has been shown t o be both a posiB. K., Matsui, H., Kobilka,T. S.,Yang-Feng, T. L., Francke, U., Caron,M. tive regulator of viral replication and a negative regulator of Kobilka, G.. Lefkowitz., R. J.., and Recan. J. W. (1987a) Science 238.650-656 early gene transcription (Fanning and Knippers, 1992). The Kobilka, B.K., Frielle, T., Dohlman, H. G., Bolanowski, M. A,, Dixon, R. A.F., Keller, P., Caron, M. G., and Lefkowitz, R. J. (1987b) J. B i d . Chern. 262, AF"1 binding site binds a transcription factor, AP-1, and its 7321-7327 activities may be modulated by treatment of cells with the Lan, M. S., Russell, E. K., Lu, J., Johnson,B. E., and Notkins, A. L. (1993) Cancer Res. 53,41694171 (Angel tumor promoter, 12-0-tetradecanoylphorbol-13-acetate Lawn, R. M.,Adelman, J., Franke,A.E., Houck, C.M., Gross, M., Najarian, R., and et al., 1987; Lee et al., 1987a). Goeddel, D. V. (1981) Nucleic Acids Res. 9, 1045-1052 Two P-cell-specific upstream promoter elements, character- Lee, W., Mitchell, P., and Tjian, R. (1987a) Cell 49, 741-752 Lee, W., Haslinger, A,, Karin, M., and Tjian, R. (1987b) Nature 325,368372 istic of the upstream glucokinase promoter, are located at poMcKnight, S. L., and Kingsbury, R. (1982) Science 217, 316-324 sitions -2075 and -1807 bp of IA-l. Gel mobility shift experi- Melton, D. W., Konecki, D. S., Brennand, J., and Caskey, C. T.(1984) Proc. Nutl. ments have shown that a P-cell factor isolated from nuclear Acud. Sci. U. S. A . 81,2147-2151 extracts of murine insulinomacell lines (HITM2.2.2 and P-TC- Nagata, S., Mantei, N., and Weissmann, C. (1980) Nature 287, 401408 Nakayama, N., Miyajima, A,, and Arai, K. (1985) EMBO J. 4,2643-2648 31, but not from the pituitary tumor cell line (AtT-20) or the Ninomiya, Y., Gordon, M., van der Rest, M., Schmid, T., Linsenmayer, T., and Olsen, B. R. (1986) J. Biol. Chem. 261, 5041-5050 fibroblast cell line (NIH-3T31, binds to the upstreampromotor Pearson, W. R., and Lipman, D. J. (1988)Proc. Nutl. Acad. Sci.U. S. A. 86, 2444elements (Shelton et al., 1992). The P-cell factor is the same 2448 factor that binds to similar elements, termed CT boxes, in the Reynolds, G. A., Basu, S. K., Osborne, T. F., Chin, D. J., Gil, G., Brown, M. S., Goldstein, J. L., and Luskey, K. L. (1984) Cell 38, 275-285 insulin promoter. Additionally, at position -1931 bp, there is a Sambrook, J., Fritsch, E. F., and Maniatis, T.(1989) Molecular Cloning: A Laboconserved prolactidgrowth hormone gene recognition element rutoryManual, pp. 9.16-9.19, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY GHF-1 (TATATGCAT),which is the binding site of the pituitarK. D., Franklin, A. J., Khoor, A,, Beechem. J.,and Magnuson, M. A. (1992) y-specific positive transcription factor, Pit-1 (Ingraham et al., Shelton, Mol. Cell. Biol. 12, 45784589 1988). It hasbeen suggested that thebinding of the Pit-1factor Simpson, N. E.(1988) J. Med. Genet. 25, 794-804 to thissequence may confer a characteristic cellular phenotype Stein, G. S., Stein J. L., and Marzluff, W. F. (1984) Histone Genes:Structure, Organization, and Regulation, pp. 397455, John Wiley & Sons, Inc.,New York in the anteriorpituitary. Sunahara, R. K., Niznik, H., Weiner, D., Stormann, T., Brann, M., Kennedy, J., The IA-1 gene is expressed in tumorsof neuroendocrine oriGlernter, J., Rozmahel, R., Yang, Y., Israel, Y., Seeman, P., and O'Dowd, B. (1990) Nature 3 4 7 , 8 0 4 3 gin, including insulinomas and pituitary tumors. The presence Teebi, A. S., Murthy, D. S., Ismail, E. A,, and Redha, A. A. (1992)Am. J . Med. Genet. of several different tissue-specific regulatory elements in the 42, 35-38 upstream region of the IA-1 gene may result in theexpression Tory, K., Latif, F., Modi, W., Schmidt, L., Hui-Wei, M., Li, H., Cobler, P., Orcutt, M. L., Delisio, J., Geil,L., Zbar, B., and Lerman,M. I. (1992) Genornics 13,275-286 of IA-1 in different neuroendocrine cells responding t o different Valerio, D., Duyvesteyn, M. G. C., Dekker, B. M. M., Weeda, G., Berkvens, T. M., tissue-specific transcription factors. Currently, we are attemptvan der Voorn, L., van Ormondt, H., and van der Eb,A. J. (1985) EMBO J. 4, ing to characterize the promoter and other regulatory elements 437443 Whittaker, P. A,, and Southern, E. M. (1986) Gene (Amst.1 41,129-134 of the IA-1 gene as well as tissue-specific transcription factors Young, D., Waitches, G., Birchmeier, C., Fasano, O., and Wigler, M. (1986)Cell 45, 711-719 present in various IA-1-expressing cells, including small cell ~~~

~

Y

I