DNA AND CELL BIOLOGYVolume 10, Number 6, 1991Mary Ann Liebert, Inc., PublishersPp. 467-474

Human Spermidine Synthase Gene: Structure andChromosomal Localization

SANNA MYÖHÄNEN, LEILA KAUPPINEN, JARMO WAHLFORS, LEENA ALHONEN,and JUHANI JÄNNE

ABSTRACT

The human spermidine synthase (EC 2.5.1.16) gene was isolated from a genomic library constructed withDNA obtained from a human immunoglobulin G (IgG) myeloma cell line. Subsequent sequence analyses re-vealed that the gene comprised of 5,818 nucleotides from the cap site to the last A of the putative polyade-nylation signal with 8 exons and 7 intervening sequences. The 5 -flanking region of the gene was extremelyGC rich, lacking any TATA box but containing CCAAT consensus sequences. No perfect consensus se-

quence for the cAMP-responsive element for the AP-1 binding site was found, yet the gene contained sevenAP-2 binding site consensus sequences. The putative polyadenylation signal was an unusual AATACA in-stead of AATAAA. Polymerase chain reaction analysis with DNA obtained from human x hamster somaticcell hybrids indicated that human spermidine synthase genomic sequences segregate with human chromo-some 1. Transfection of the genomic clone into Chinese hamster ovary cells displaying a low endogenousspermidine synthase activity revealed that the gene was transiently expressed and hence in all likelihood rep-resents a functional gene.

INTRODUCTION

The biosynthesis of the poLYAMiNEs putrescine, sper-midine, and spermine in mammalian cells is accom-

plished by the concerted action of four separate enzymes—ornithine decarboxylase (EC 4.1.1.17), adenosylmethio-

nine decarboxylase (4.1.1.50), spermidine synthase (EC2.5.1.16), and spermine synthase (EC 2.5.1.22). The twodecarboxylases, especially ornithine decarboxylase, havebeen the subjects of extensive investigation due to theirstriking inducibility and short half-lives (Jänne et al.,1978). Much less attention has been paid to the two pro-pylamine transferases: spermidine synthase and sperminesynthase, apparently because these enzymes are stable andonly modestly inducible. The regulation of spermidine andspermine synthases is believed to occur at the level of sub-strate (decarboxylated adenosylmethionine) availability(Jänne et al., 1978). However, there are reports indicatingthat spermidine synthase, but not spermine synthase, is in-duced in response to accelerated proliferation such as liverregeneration (Hannonen et al., 1972), hormone-inducedgrowth of tissues (Oka et al, 1977; Käpyaho et al., 1980),

and lymphocyte activation (Korpela et al., 1981). In fact,under these conditions, the stimulation of spermidine syn-thase activity was fully comparable to that of ornithineand adenosylmethionine decarboxylases.

Recently, we isolated and sequenced cDNA encodinghuman spermidine synthase from a human decidual cDNAlibrary (Wahlfors et al., 1990). By using this cDNA as a

probe we now have isolated the whole human spermidinesynthase gene from a genomic library constructed withDNA from a human myeloma cell line. The gene was sub-sequently sequenced, its chromosomal location assigned,and its functionality confirmed by transfection experi-ments.

MATERIALS AND METHODSMaterials

Restriction and other DNA-modifying enzymes were

purchased from Boehringer Mannheim, Promega, or NewEngland Biolabs. LambdaGEM Xho I arms were fromPromega and M13mpl8 RF DNA from Boehringer Mann-heim. Radioactive isotopes [a-32P]dCTP and [7-35S]dATP

Department of Biochemistry & Biotechnology, University of Kuopio, P.O.B. 1627, SF-70211 Kuopio, Finland.

467

468 MYOHANEN ET AL.

and Hybond N nylon membranes were from Amersham. ASequenase 2.0 sequencing kit was purchased from U.S.Biochemicals and the AutoRead sequencing kit was fromPharmacia. Sequencing gel reagents were purchased fromBioRad or Pharmacia. PCRable DNA system was fromBIOS Corporation. DL-[2-'4C]methionine was from Amer-sham. Chinese hamster ovary cell line CHO-K1 was ob-tained from the American Type Culture Collection. Thecell culture medium F12 was from NordCell and Eagle'sminimum essential medium (MEM) from NordVacc; othermaterials for cell culture were purchased from GIBCO.

Isolation and characterization of a genomic cloneGenomic DNA was isolated from human IgG myeloma

cell line Sultan 20D with ornithine decarboxylase gene am-

plification and modest spermidine synthase activity (Hir-vonen et al, 1989). A genomic library was constructed intoa lambdaGEM12 vector and screened with the nick-trans-lated insert of the cDNA plasmid phSDl (Wahlfors et al.,1990). The insert of a positive clone was subcloned intopUC19 and characterized with Southern blotting and DNAsequencing of double-stranded X template using primerspreviously constructed for characterization of the cDNA.

DNA sequencingThe nucleotide sequence of the human spermidine syn-

thase gene was determined using the shotgun strategy(Bankier and Barrel, 1983). An Eco RI fragment of hg-SPDSY1 was sheared by sonication or by Alu I or Sau 3Adigestions. Fragments were cloned into M13mpl8 andabout 100 individual clones were sequenced using thechain-termination method of Sanger et al. (1977). Se-quencing reactions and analyses of the reaction productswere performed either manually using Sequenase enzymeaccording to the protocol of U.S. Biochemicals or auto-matically using Pharmacia's AutoRead fluorescent se-

quencing kit and A.L.F automated DNA sequencer. dITPwas occasionally used instead of dGTP to resolve compres-sions due to extreme high G/C content of the 5'-flankingregion. Double-stranded sequencing was performed occa-

sionally using pUC19 as a vector and following the instruc-tions of the AutoRead sequencing kit. To fill gaps in thesequence or to obtain sequence from the opposite strand,the reverse end of some M13 clones were sequenced usingfluorescent reverse M13 primer by the method ofMyöhänen and Wahlfors (1991). Sequence data was com-

pared, assembled and analysed with the aid of DNASISsoftware system (Hitachi America, Ltd. and PharmaciaBiotechnology).

Chromosomal localizationTwo oligonucleotide primers complementary to fourth

and fifth introns of human spermidine synthase gene were

synthesized with Applied Biosystems 381A DNA synthe-sizer. Sequences of the primers were 5-CTGGCTCTGG-CCACCTGGTA-3' (4th intron) and 5'-TAAGCATCAGC-ATCCGGCAG-3' (5th intron), giving a PCR product of

248 bp. PCRable DNA from human x hamster somaticcell hybrids was used as the template in polymerase chainreaction according to BIOS Corporations instructions. TaqDNA polymerase (0.25 U/reaction) and appropriate bufferwere from Promega. The amount of both primers was 20pmole/reaction and the titrated optimal magnesium con-

centration was 1.5 mM. After denaturation in 96° C for 3min, 30 cycles in Hybaid thermal cycler were carried out

according to the following program: 95°C for 1 min, 62°Cfor 1 min, and 72°C for 30 sec. Polymerization step in thelast cycle was extended to 5 min at 72°C. The analysis ofthe PCR products was carried out on 1.5% agarose gelstained with ethidium bromide.

Transfection experimentsA wild-type Chinese hamster ovary cell line CHO-K1

was cultured in Nutrient mixture F12/MEM supplementedwith 10% fetal calf serum and 50 mg/liter gentamicin. TheEco RI fragment of spermidine synthase genomic clone inpUC19 plasmid, phgSPD65, was used for the transfectionexperiments. The circular plasmid was transfected into theCHO-K1 cells by the calcium phosphate coprecipitationtechnique in low pH and low C02. The cells (0.5-1 x 106)were incubated overnight on 10-cm diameter plates in thegrowth medium with 10 fig of plasmid DNA using buffersand conditions described by Chen and Okayama (1987).After transfection the cells were grown in the F12/MEMmedium and spermidine synthase activity was measured bythe method of Raina et al. (1983) at 48 and 72 hr after thetransfection. Mock transfection was performed with 10 figof sonicated herring sperm DNA.

RESULTSIsolation and nucleotide sequence of the humanspermidine synthase gene

A human myeloma cell line genomic library constructedinto the replacement vector LambdaGEM-12 was screenedusing the insert of cDNA plasmid phSDl as a probe. Outof 130,000 plaques, one positive X clone was obtained.This clone, designated as hgSPDSYl, had an insert of 10.5kb (Fig. 1) and was potentially a full-length gene accordingto Southern blot analyses. This was further confirmed bydouble-stranded DNA sequencing using oligonucleotidescomplementary to the cDNA sequence as primers. Se-quencing revealed that the clone hySPDSYl contained atleast the whole protein coding region of the spermidinesynthase gene. The 10-kb Eco RI fragment (Fig. 1) wassubcloned into plasmid pUC19 for propagation and desig-nated as phgSPD65. The Eco RI fragment was also usedfor sequence analysis.

The shotgun sequencing approach was used to determinethe nucleotide sequence of human spermidine synthasegene and the flanking regions. Ml3 clones from severalseparate clonings were sequenced to obtain the sequencefrom both strands. Figure 2 gives the nucleotide sequenceof the Pst l-Eco RI fragment containing the whole gene.

The human spermidine synthase gene is 5,818 núcleo-

HUMAN SPERMIDINE SYNTHASE GENE 469

I I

FIG. 1. Structure of human spermidine synthase gene. a. Restriction enzyme map of the insert in the X clone hgSPD-SY1 containing a functional spermidine synthase gene. Restriction sites shown: X, Xho I; E, Eco RI; S, Sac I; H, HindIII; P, Pst I; C, Cla I; B, Bam HI. b. Schematic structure of spermidine synthase gene. Flanking regions and intronsare shown as solid line, while open boxes represent untranslated regions and solid boxes represent coding regions. Exonsare numbered with Roman numbers. The lengths of coding and untranslated regions in the same exon are marked sepa-rately.

tides long from the cap site of the mRNA (determined pre-viously by Wahlfors et al, 1990) to the last A of the puta-tive polyadenylation signal. The gene contains 8 exons and7 introns, with all the junctions conforming to the GT-AGrule. The structure of the gene as well as the lengths ofexons and introns are shown in Fig. 1. All the exonic se-

quences match with the cDNA sequence, except nucleotidenumber 6,456 in the last exon. This nucleotide change isconservative, replacing codon number 297 in the cDNA,GCC, by GCA, both of which encode alanine. The lastexon was predicted to be 679 nucleotides long, containing21 coding nucleotides with the rest being noncoding. Thereal polyadenylation signal could not be determined bycomparison, because the cDNA clone phSDl is not fulllength at the 3' end (Wahlfors et al, 1990). However, theputative signal AATACA was the best alternative in thenear 3'-flanking region. The calculated mRNA size was

equal to the one obtained by Northern analysis (about 1.6kb). Further, a radioactive single-stranded probe comple-mentary to nucleotides 49-336 after the AATACA signaldid not hybridize to any mRNA. However, in Southernblot analysis this probe recognized a 7-kb band of spermi-dine synthase gene.

The two longest introns, 3 and 6, contain 4 and 2 Alu re-

peats, respectively. These particular introns consist almostentirely of /l/«-related sequences and represent more than30% of th gene. The Alu repeats in spermidine synthasegene show the highest similarity to the "old" subclass(Deininger and Slagel, 1988), as did also the human orni-thine decarboxylase gene (Hickok et al, 1990), codinganother polyamine synthesizing enzyme.

A typical CpG-rich island contained about 550 nucleo-tides in the 5'-flanking region, the first exon and the 5' halfof the first intron of spermidine synthase gene. The aver-age G/C percentage of this 1,100-nucleotide region was78%, but the percentage around the transcriptional startsite (from nucleotide 1,200 to nucleotide 1,400) was even

higher, over 83%. No TATA box, but two consensusCCAAT sequences were present, one in the sense strand(nucleotides 1,212-1,216 in Fig. 2) and one in the oppositestrand (nucleotides 1,242-1,246 in Fig. 2), 103 and 73 nu-cleotides away from the transcriptional start site, respec-tively. Both of these sites are also putative binding sites fortranscription factor NF-1 (Mitchell and Tjian, 1989). EightGC boxes (GGGCGGG) in both strands were present inthe promoter region and one was in the first exon of thegene. Among these nine sequences, one appeared to be a

part of full consensus sequence of Spl binding site (GCC-CCGCCC, at 1,275-1,283 in Fig. 2). According to Mitchelland Tjian (1989), no perfect consensus sequence for CREBor AP-1 (z-juri) binding sites conferring cAMP or TPA in-ducibility were present in the gene. However, seven AP-2binding site consensus sequences, also conferring TPA andcAMP inducibility, were found in both orientations (CCC-CAGGC; at nucleotides 2,391, 3,084, 4,794, 5,241, 5,408,5,927, and 6,904 in Fig. 2). Whether these putative bindingsites are relevant for the transcription of human spermi-dine synthase gene in vivo is not known.

Chromosomal localizationPCR was carried out as described in Materials and

Methods and reaction products were analyzed by agarosegel electrophoresis. In addition to human genomic DNAcontrol, a single PCR product of the predicted size was ob-tained with DNA from hybrid cell lines 683, 867, 937, and1,099 as templates (Table 1). No reaction product was ob-tained with hamster genomic DNA or with reagent blank.These results clearly map the human spermidine synthasegene to chromosome 1, despite the incorrect positive signalfrom hybrid cell line 683. According to the manufacturer,this cell line is not supposed to contain chromosome 1.However, our recent studies with genes known to segregatewith chromosome 1 have indicated that this particular cell

470 MYOHANEN ET AL.

10 20 30 40 50 60 70 80 i « i * I » * * i l i i i l i i » ' i ' i i » I

ctgcaggcgcgcactaccatgcccagctaatttttgtatttttagtacagacggggtttcaccatgttggccaggatcgt 80cttgatctcttgacctcgtgatccgcccgcctccgcctccgcctcccacagtgctgggattaccggggtgaaccatcocg 160cccaggcacctagcaattctttagcggtcttggfttacctccccittagaaggagcttaaaagcaagcaaggcacattgt 240tgcctaggcttgaggcttgctctcacccataaacaacgctgttactctgctctggggacgacacaggaaacgttccccac 320ctccaggtggaggctgcaaaacgtgtcaaaaccatccctgacataatgtcaagagtagcttatactagtttcatcttcct 400tccttggcattcgaactgcgtgtggaacaattagcatttaatatttggtaattagcatgcttaatgtgattctgagaagt 480tctttgacattctcataaaaacagcacattcccacccacccttcaaagagcaagacccagtttgtcaagaaaaattgcgt 560gccagtcttttctggtgctgaatatgtatgttctgggcctcttcctggacactgctggtaaatttagaaactcgtttaga 610aaagcacttctctcgtattcaacagcctataggctcatggcgcagaatctaagggaaaatggctaaatccagcttgttaa 720ttcgcgggctgtgatgattctttccaagtaaataaaaaccctcggttcgccccgacgagccacataatctgttcaaatcc 800aacaaggaaccagattttggacgcaaagaaggatacgttctactcgccccgtgcaacaacgtaaaccactgtagccgccc 880gctcccgtgtctccagcccaaaggctgactctccagtccgcacgtcgcagcgctcttgccctccacaccaagcccgagtc 960ccgcagcccctcgaggccctcggtgcctcccaaccccgagaaggaagcgggggccggtggtgcaccgccccggctgcttg 1040gggcggaggaaggacccggaccccttccgccggcccagcccgccccggaacccgacccggccgcccggccccggccgggc 1120cccacgtggcccctggagcgggccgcactaccctgctgccgccgacggacggcgcgccacagccactctgcgccgctctg 1200cgagccggtggccaatgagcgccaggcgaggccgctttgccattggcgagcgcgggctccgcccccgccggcaggccccg 1280ccccgcgcccgggttaggttgcggcgcgggcggcGGGCGGAGCTGGTCCCGTTGTGCTGCGGCGCCGCGCGGCCTGCAGT 1360CCCGGGCCCGCGCCCCGCGCCGCCCGCCCGCCCGCCATGGAGCCCGGCCCCGACGGCCCCGCCGCCTCCGGCCCCGCCGC 1440CATCCGCGAGGGCTGGTTCCGCGAGACCTGCAGCCTGTGGCCCGGCCAGGCCCTGTCGCTGCAGGTGGAGCAGCTGCTCC 1520ACCACCGGCGCTCGCGCTACCAGGACATCCTCGTCTTCCGCAGgtaccgccgctgcccgcaggcgcctgccccctaggct 1600cagcccgggccgcctgctgcccgcctcacgggcctctccacgccgggacccaagcgggctggacctcgtcctgccctggc 1680cccctcgccacccctcacaccgcctccctgggctggggctgggactgcgggctggcctcttgggtgggggagtcggagtc 1760tgcgccccgctccacgtgtcagccctcaggacacgtcagagcccgaggagaccccgggtcccaccccggcctccacccgg 1840cggcccgcctgccgttcctcgccacgtgtcaecategetcetcatccctgggacccctaggcgggatggggagaccctec 1920tcacccagggcggcttggggtacgttttccccaccccagagaacccaggtccccgactgtcactccgcccgcagTAAGAC 2000CTATGGCAACGTGCTGGTGTTGGACGGTGTCATCCAGTGCACGGAGAGAGACGAGTTCTCCTACCAGGAGATGATCGCCA 2080ACCTGCCTCTCTGCAGCCACCCCAACCCGCGAAAGgtaccccagtgtcccctggaacagtgccggacgaggggcggcccc 2160aggtgtgctccgggctcttcccagatgctgcctgcatggttgtcagagaaagtgctagcaaggccaggggcgtcccgcgg 2240aggggtgggggccgacactgacgcggcctcggaatcctagggcagccctggaaggaacttccaggaaaggggacaccggc 2320aegaaagegtttccgagggtagaaaaagatgaggcccgtgggtccgaggggtcagggggtctgcttcaggggcctggggg 2400ctcccagtcctgccagggcccctgccttgactgccccctcctcccagGTGCTGATCATCGGGGGCGGAGATGGAGGTGTC 2480CTGCGGGAGGTGGTGAAGCACCCCTCCGTGGAGTCCGTGGTCCAGTGTGAGATCGACGAGgtgagtgccggcgtagagcc 2560aggtttgagtcctggttctcccagcggccagctgtgccctgaaatggctgcacacccccgagcaaggcaggtagggcctg 2640tttctccatctggaaaacacctggtcggggagggttcagtaggaaaaccagatggcagagggcctggcaggtggtgaggg 2720cacctgcgtggcgagctcttactaaaactgagctgatttttttttttttttttttttgagacagagtttcgctcttgttg 2800cccaggctggagtgtgatggtgcgatctcggctcactgcaacctccacctcctgggctcaagtagttcttctgcctcagc 2880ctccggagtagctgggattacagacatgcgctaccatgcccggctaattttgtatttttagtagagacagggtttctcca 2960tgttggtcaggctggtctcggacctggcgaccacaggtgatccgccagcctcgtcctcccaaagtactgggattacaggc 3040gtgagccaccacgcccagccgactaagctgatttttaatctgagccccaggcaggqccccaagacaqctcaactatttqt 3120acg Liaccccttacactcagtagctgctcactaaaatcatgctacgtgccaggtgttgcccgggtatggggacagtggta 3200gacgacagatcagtccctgccctctaggagctgatgtcgtagttaaaggagacatcagatggccagacgtggtggctcac 3280acctgtaatcccagcactttgggacgccaaggcgggcagatcacctgaggtcaggagttcaagagcagcctggccaacct 3360ggtgaaaccccatttctactaaaaatacaaaaattagccgggcatggtagtgcatgtctgtaaacccagctactagggag 3440gctgaggtggaagaat tacttgagcccgggaggcggaagttgcaatgaaccgagatctcgccactgcactccagcctggg 3520tgacagaggaagaatctgtctcaaaaaaaacaaacaacaaaaatagagacatcaaaqgatqgtctgatgaagqcaaqaca 3600ggggctgggggacaggagaaggcagggttcctgtgaatgcatggggggtggtcagggcaggcctccaggaggtggcgttt 3680gagctgagacctcagtgaaaagcaggtggccgtgtgcagggagggggaggttctcctggccagaggttggaattgcatcc 3760ttctaaaataggaaacaggccaagcgctggtggctcacacctgtaatctcagcactctgggaggctgaggcgggcagatc 3840acaaggtctctactaoaaatacaaQaaaaaaaaaaaatggcccagcttggtggcgtgtgcctgtaatcccaactactcgg 3920gaggctgaggcaggagggtgcagtgagctgagatcgtgccactgcactccagcctgggcagcagagcaagactgtctcga 4000aaaataaataaataaaataqqaaqcqacaaqaaaqccactcaqatqgqqcqatttqqtctqqaaqqaqqgqataaqqatq 4080ggqgagaggagcccaagccgccggcaggagccagctctcaaggagaaatggaggtaccagagttctccccgctttacaca 4160ttataaactgaggttcccaaaaggggccaggtgtggtggctcacagctgtaatcctagcacttttggaggccgaqqtggg 4240aggatcgaggagttcgaggcaacatagggagactctatctctaccaaaaatttaaaaagtagccaagtatgattqaacac 4320acttgtcccagctactcaggaggctgatgggggaggatcacctgagccccggaggccgagggtgtagtgagccatqatcg 4400atgccactgcactccagcctgggccacaaagtgagaccctgtctcaaaaaaaaataataaaaaaaagggaagqqqttqqc 4480caaggeggettgcctgtgaggcacttggagagtcccacgtggctgtgctggctccaggtcccccagccccttggcccaga 4560ctggtccctcccatcccctccagGATGTCATCCAAGTCTCCAAGAAGTTCCTGCCAGGCATGGCCATTGGCTACTCTAGC 4640TCGAAGCTGACCCTACATGTGGGTGACGGTTTTGAGTTCATGAAACAGAATCAGGATGCCTTCGACGTGATCATCACTGA 4720CTCCTCAGACCCCATGGgtaagcagtggatgggccccagggttttctggcagctgcaggtctggaggtcagcctccccca 4800

HUMAN SPERMIDINE SYNTHASE GENE 471

4810 4820 4830 4840 4850 4860 4870 4880' ' ' I ' ' ' I ' ' I ' ' ' ' ' ' I I ' I I ' I l I ' I ' l ! I I I I I I I I I I ' I I- l l I I I ' ' I

ggccttcagagtaaaggatagagcggcctcccaccccccgaactagagctgtacttttcccttctcatttgttacctgcc 4880ctctgaaacatggctcaggacagtaggcaggagccaggcgactgcccagattcacaagctggtgaccaaggagagtgggg 4960atctggcattgggacactgaggaccctgtgtcctcttcagcctcccctctgctctgaagtggtcagcactggagtggggg 5040caggttctagtcttgaacgaaggcctaggttagaggttcctctgctgtggtgccaatgagactcccccaagaatgggatt 5120caggtgtggatcccccacagacctgggttcagatcctggctctggccacctggtagctgtgtgggtgacagtggccctgg 5200aggtcacagccagactgttcagtgtgtctccctctgtcttccccagGCCCCGCCGAAAGTCTCTTCAAGGAGTCCTATTA 5280CCAGCTCATGAAGACAGCCCTCAAGGAAGATGGTGTCCTCTGCTGCCAGGgtgagccacaggcctggagcactggggcgg 5360ggcggggtggggcagggcaggccctgccggatgctgatgcttaggggcccccagGCGAGTGCCAGTGGCTGCACCTGGAC 5440CTCATCAAGGAGATGCGGCAGTTCTGCCAGTCCCTGTTCCCCGTGGTGGCCTATGCCTACTGCACCATCCCCACCTACCC 5520CAGCGGCCAGATCGGCTTCATGCTGTGCAGCAAGAACCCGgtgagatgggggtgtctgggggtgggggttggggggaagg 5600tgggcataaatagagatccctgcccctgccgggcgcggtggctcacacctgtaacccagcactttgggaggctgaggcgg 5680gcagatcacaaggtcaggagatcgagaccatcttggctaacacggcgaaaccccgtctctactaaaaatacaaaaaaatt 5760agccaggcatggcagcgcgcgcctgtagtcccagctgctggggaggctgaggcaggagaatggcgtgaacccgggaggcg 5840gagcttgcagtgagccgagattgcgccactacattccagcctgggtaacagaggaagattccgactcaaaaaaaaaaaaa 5920aggccctccccaggccaggtgcggtgtctcatgcctgtaattccagcattttgggagaccaaggtgggcggatcacttga 6000ggtcaggagtacaagaccagcctgaccaacatggagaaaccccatcactactcaaaatacaaaaaaaaaaaaaattagcc 6080gggcgtggtggcgcgtacctgaggctgaggcaggagaat.cacttgaacctgggagacagaggt'tgcagtgagctgagatg 6160acqccactgcactccagcgtgqcaacagtqagactccgtctcaaaaaaaaaaaaaaaagtgccccccctgatgtqcccct 6240ggcccggtccccagAGCACGAACTTCCAGGAGCCGGTGCAGCCGCTGACACAGCAGCAGGTGGCGCAGATGCAGCTGAAG 6320TACTACAACTCCGACGTGCACCGCGCCGCCTTTGTGCTGCCCGAGTTTGCCCGCAAGgtgggtggcctgcggggctgggt 6400ggtgggacccagggacccagagcgccctcctgactggcctcatgtccctccagGCACTGAATGATGTGAGCTGAGCCCAG 6480GCGCCACCACTGATGCCACCCAGGACCTCGGACCTTGGAGCCTGCGGGGTGCCTCGGCCCCTCCAGCCCCGGGCCGGACC 6560TCCTGCTGGCTCTCGCCCACCAACCAAGTGTTACAAGCCCCAGAATGCTGCCCGGCCTGCCCTGCTGGGCGGACTGTCTG 6640TGTGTCTGTCTCTCTGGCGTTCCACCTCCAAGCCTATACCAGCTGTGTACAGCGCCATCTCTCTGCCTTCTGTTGCCCCT 6720CACTCACCAAACACGTGTATTTATAGCAAAGATTGGAGTCCTGTGTCTCCTGACCTTGGCTGGGCCCAGGCAGGGCCACA 6800TTCACCATTGGGTGCCTCTGGGGTGAGGGTCTGCAGAGGCCTTGCTGGCTGACCCCCAAGTGTCTGCTGCAGGGCTGAGG 6880CTGCAGGCGGGCCATCGTGGATAGCCTGGGGCACAGAGGGTCACCGCAGTCGTCACGTGGGACCCAGAGCTGTCCTGGGA 6960AGCTGACTTAGCTGTCCTTTTACCAAGCCCTTCACAAGGCCACTGGTGACAGCCCCCCAGGGCAGTGGGGTGGGTGAGAT 7040CAGGGTGGGGCTGCCCGGGAGCATTCTCAGAAAAATTGGGGACACTCACAGGTGTAAGTCAGGTCCCATCCAGGTACTCC 7120AGGGCAAATACAggaaggggtggcggggctggttaccttcggcctttttaagcacatcaggagcttaacactggcccagt 7200gactgtgccctgactccacccggcattcagacttgggttcaaattcccaccatgccccgccccctatgtggacaaattga 7280gaaagcaagtgtgggcaccccaccagggactgcgaggaccagggctgtcccctctccaaggtgctgaactcccgccttcc 7360aggacccaacggtggtgggaggacaggaaaggaaccctctttgcatgggcctgagttgccaacccctttccccaccctgg 7440gcaggggctgggctagcggacgcatcagggagggaggccccactcccagccgaggcagccaccttggagccctaactcac 7520ccgggtatgttttctgggacaccagtgtaagggggattcagtttcgccatcaactctggcttcaggccagtcatagccct 7600ccagtctccacctgccccccact 7623

FIG. 2. The nucleotide sequence of human spermidine synthase gene. Exons and introns are specified by capital andlowercase letters, respectively. The first nucleotide in the Pst I recognition sequence (see Fig. 1) is designated as nucleo-tide number 1. The Alu repeats in the 5'-flanking region and in the introns 3 and 6 are underlined.

line contains at least fragments of human chromosome 1(Kolmer et al, 1991).Expression of human spermidine synthase genein CHO cells

To verify that the isolated human spermidine synthaseclone hgSPDSYl was functional, the plasmid phgSPD65was transfected into CHO-K1 cells with low endogenousspermidine synthase activity. Spermidine synthase activityof the cells transfected with phgSPD65 was three timeshigher at 48 hr after the transfection than in the cells trans-fected with herring sperm DNA. When measured at 72 hrafter the transfection, the activity was still over two timeshigher than in the control cells (Table 2).

DISCUSSION

The described human spermidine synthase gene probablyrepresents a functional and actively expressed gene. Thisview is supported by the fact that the gene was transiently

expressed when transferred into Chinese hamster ovarycells (Table 2). Furthermore, the only discrepancy betweenthe cDNA sequence for human spermidine synthase we re-

cently reported (Wahlfors et al, 1990) and the present se-

quence of the coding region was a conservative replace-ment of the alanine codon 297 GCC by GCA, the lattertriplet still coding for alanine. This difference may havebeen resulted from the different source of the cDNA (de-cidual) and genomic (myeloma) libraries used and may justrepresent DNA polymorphism. The latter possibility hasnot been proved because no suitable restriction enzymethat cleaves at this site is available.

Based on its promoter structure, the human spermidinesynthase gene appears to belong to the so-called "house-keeping" genes characterized by an extreme GC richnessand the lack of a TATA box (Dynan, 1986). The transcrip-tion factors, such as ETF, are supposed to bind to GC-richregions of the promoter (Kageyama et al, 1989). If onelooks at the promoter region of human spermidine syn-thase, it is not only GC-rich but contains several 5'-CCCC-3'

MYOHANEN ET AL.

O)EoOEo

ü

"OO.

CMCM

OCvl

C

coo

. . m +tM

+ . o . +

+ . m +

+ + LO

+ ICI +

+ Q + + + + D>++Q . ++Q + +Q8loio

oCD

CM

COCM

OCM

COCM

CM

COCM

CM

^3-CM

COCM

+ + + + Q

it*0*0(>)ONOCMSn<D^OCMO)N7^-NCO(DOlcllOwiMnmoioiD^^ojiio^coooninoiDcoo^Ngn^NMOaittlCIltMLOlONOOCIICOOICIICOOllDOIOOO

CMCO

CM

CMco

CM

01 Tt "D

° o »Si 0) - UÎ

to cE »5 tí

JZ CD—' 0-

HUMAN SPERMIDINE SYNTHASE GENE 473

Table 2. The Transient Expression of HumanSpermidine Synthase Gene in CHO-K1 Cells

Spermidine synthaseTime after activity

Transfected gene transfection (hr) (pmole/min/fig)NonephgSPD65NonephgSPD65

48487272

0.3340.9210.3050.652

The results are means for triplicate cultures.

motifs which are supposed to be the best binding sites forETF (Kageyama et al, 1989).

Another slightly unusual feature of human spermidinesynthase gene was the apparent replacement of the by farmost common polyadenylation signal AATAAA by AAT-ACA. All the experimental evidence, however, supportsthe view that this sequence signals the polyadenylation ofthe mRNA. These include the correct size (1.6 kb) of sper-midine synthase mRNA in Northern blotting and the factthat a probe covering nucleotides 49 to 336 after the puta-tive polyadenylation signal did not recognize any mRNAspecies but did recognize spermidine synthase gene inSouthern blot analysis.

The human spermidine synthase gene appears to segre-gate with chromosome 1 as revealed by our polymerasechain reaction assay with genomic DNA obtained from hu-man x hamster somatic cell hybrids. Chromosome 1showed by far the lowest discordance rate (4%), givingonly one incorrect, according to the information of thesupplier of this panel, signal. However, the particular cellline (683 in Table 1) not supposed to contain human chro-mosome 1 in all likelihood carries at least fragments of it.We recently encountered similar difficulties with the panelin the connection of another chromosomal assignment(Kolmer et al., 1991). Similar discordance was found andwe subsequently analyzed the panel with primers for the(3-subunit gene of thyroid-stimulating hormone known tosegregate with human chromosome 1 (Fugushige et al,1986). Positive signals were obtained from five differentcell lines in Table 1, including the line 683 not supposed tocontain human chromosome 1 or fragments of it as re-vealed by karyotyping. As indicated earlier, these spermi-dine synthase sequences on chromosome 1 apparently rep-resent an actively expressed gene; no signs for the presenceof a pseudogene(s) in human genome were obtained. This,however, does not exclude the presence of processed pseu-dogenes as we used primers targeted to intron sequences toscreen the DNA panel.

Of the genes for enzymes engaged to the biosynthesis ofthe polyamines in man, the actively expressed ornithine de-carboxylase gene is located on the short arm of chromo-some 2 together with a processed pseudogene on chromo-some 7. The genomic sequences for human adenosylmethi-onine decarboxylase segregate with chromosomes X and 6(Radford et al, 1988), thus indicating an evolutionary di-vergence of the genes of these three enzymes involved inpolyamine biosynthesis in human.

ACKNOWLEDGMENTS

We thank Anne Karppinen for synthesizing the oligonu-cleotide primers and Meelis Kolmer for designing Table 1.We also thank Ms Eija Korhonen for preparing labeled de-carboxylated S-adenosylmethiothine and Dr. TerhoEloranta for putting this reagent at our disposal. The skill-ful secretarial assistance of Miss Taru Koponen is grate-fully acknowledged. This work was supported by the Fin-nish Academy of Sciences and the Finnish Foundation forCancer Research. The nucleotide sequence reported in thispaper has been submitted to GenBank with accession num-ber M64231.

REFERENCES

BANKIER, A.T., and BARREL, R.T. (1983). Shotgun DNA se-

quencing. In Techniques in Life Sciences. B5: Nucleic AcidBiochemistry. R.A. Flavell, ed. (Elseiver, Limerick) pp. 1-34.

CHEN, C, and OKAYAMA, H. (1987). High-efficiency trans-formation of mammalian cells by plasmid DNA. Mol. Cell.Biol. 7, 2745-2752.

DEININGER, P.L., and SLAGEL, V.K. (1988). Recently ampli-fied Alu family members share a common parental Alu se-

quence. Mol. Cell. Biol. 8, 4566-4569.DYNAN, W.S. (1986). Promoter for housekeeping genes. Trends

Genet. 2, 196-197.FUGUSHIGE, S., MUROTSU, T., and MATSUBARA, K.

(1986). Chromosomal assignment of human gastrin, thyro-tropin (TSH)-beta subunit and c-erbB-2 by chromosome sort-ing combined with velocity sedimentation and Southern hybrid-ization. Biochem. Biophys. Res. Commun. 134, 477-483.

HANNONEN, P., RAINA, A., and JÄNNE, J. (1972). Poly-amine synthesis in the regenerating rat liver: Stimulation of S-adenosylmethionine decarboxylases and spermidine and sper-mine synthases after partial hepatectomy. Biochim. Biophys.Acta 279, 84-90.

HICKOK, N.J., WAHLFORS, J., CROZAT, A., HALME-KYTÖ, M., ALHONEN, L., JÄNNE, J., and JÄNNE, O.A.(1990). Human ornithine decarboxylase-encoding loci: Nucleo-tide sequence of the expressed gene and characterization of a

pseudogene. Gene 93, 257-263.HIRVONEN, A., ELORANTA, T., HYVÖNEN, T., AL-

HONEN, L., and JÄNNE, J. (1989). Characterization of diflu-oromethylornithine-resistant mouse and human tumour celllines. Biochem. J. 258, 709-713.

KAGEYAMA, R., MERLIND, G.T., and PASTAN, I. (1989).Nuclear factor ETF specifically stimulates transcription frompromoters without a TATA box. J. Biol. Chem. 264, 15508-15514.

KÄPYAHO, K., PÖSÖ, H., and JÄNNE, J. (1980). Role of pro-pylamine transferases in hormone-induced stimulation of poly-amine biosynthesis. Biochem. J. 192, 59-63.

KOLMER, M., ÖRD, T., ALHONEN, L., HYTTINEN, J.,SAARMA, M., VILLEMS, R., and JÄNNE, J. (1991). Assign-ment of human prochymosin pseudogene to chromosome 1.Genomics 10, 496-498.

KORPELA, H., HÖLTTÄ, E., HOVI, T., and JÄNNE, J.(1981). Response of enzymes involved in the metabolism ofpolyamines to phytohaemagglutinin-induced activation of hu-man lymphocytes. Biochem. J. 196, 733-738.

MITCHELL, P.J., and TJIAN, R. (1989). Transcriptional regu-lation in mammalian cells by sequence-specific DNA binding

474 MYOHANEN ET AL.

proteins. Science 245, 371-378.MYÖHÄNEN, S., and WAHLFORS, J. (1991). Reverse fluores-

cent sequencing of large M13 clones. Biotechniques (submittedfor publication).

OKA, T., KANO, K., and PERRY, J.W. (1977). Hormonal regu-lation of spermidine synthase during the development of mouse

mammary epithelium in vitro. Biochem. Biophys. Res. Com-mun. 79, 979-985.

RADFORD, D.M., EDDY, R., HALEY, L., HENRY, W.M.,PEGG, A.E., PASUNEN, A., and SHOWS, T.B. (1988). Genesequences coding for S-adenosylmethionine decarboxylase are

present on human chromosome 6 and the X and are not ampli-fied in colon neoplasia. Cytogen. Cell Genet. 49, 285-288.

RAINA, A., ELORANTA, T., and PAJULA, R.-L. (1983).Rapid assays for putrescine amino-propyltransferase (spermi-dine synthase) and spermidine aminopropyltransferase (sper-mine synthase). Methods Enzymol. 94, 257-260.

SANGER, F., NICKLEN, S., and COULSON, A.R. (1977).DNA sequencing with chain-terminating inhibitors. Proc. Nati.Acad. Sei. USA 74, 5463-5467.

WAHLFORS, J., ALHONEN, L., KAUPPINEN, L., HYVÖ-NEN, T., JÄNNE, J., and ELORANTA, T.O. (1990). Humanspermidine synthase: Cloning and primary structure. DNA CellBiol. 9, 103-110.

Address reprint requests to:Dr. Juhani Jänne

Department of Biochemistry and BiotechnologyUniversity of Kuopio

P.O. Box 1627SF-70211 Kuopio, Finland

Received for publication April 29, 1991.