Genomic structure of human polyoma virus JC: nucleotide sequence of the region containing replication origin and small-T-antigen gene.

(1)

0022-538X/83/010073-07$02.00/0

Genomic

Structure

of

Human

Polyoma Virus JC: Nucleotide

Sequence

of the Region

Containing Replication Origin and

Small-T-Antigen

Gene

TATSUOMIYAMURA,'HIROYUKIJIKUYA,2EIICHI SOEDA,2ANDKUNITOYOSHIIKEl*

DepartmentofEnteroviruses, National Institute of Health, Tokyo141,1and Departmentof Molecular

Genetics,National InstituteofGenetics, Misima

41J2

Japan

Received 5 August 1982/Accepted 23 September 1982

The nucleotide sequence of the region of human polyoma virus JC DNA

between 0.5 and0.7mapunitsfromaunique EcoRI cleavage sitewasdetermined and compared with thoseof the corresponding regions of another human polyoma virus, BK, andsimian virus 40DNAs. Within this region consisting of 945 base pairs, we located the origin ofDNA replication near 0.7 mapunits, the entire coding region for small T antigen, and the splice junctions for large-T-antigen mRNA. Thededuced amino acid sequences for smallTantigen and the partof largeTantigen markedly resembled those of polyoma virus BK and simian virus 40.The results stronglysuggestthatpolyomavirus JC has thesameorganization ofearly genome aspolyomavirus BKand simian virus40 on the physicalmap,

with theEcoRI site as areference point.

Human polyoma virus JC (JCV) has been isolated from brain tissues ofpatients with

pro-gressive multifocal leukoencephalopathy,arare

demyelinating disease, forwhich JCV is proba-bly responsible as an etiological agent (9, 19). Serological surveys have revealed thatJCV in-fection iscommonamonghumanpopulations(2, 18). Like other polyoma viruses, JCV has

tu-morigenic andtransforming capacities(7, 20, 33, 35).Therefore, studies of JCVareboth etiologi-callyandbiologically important.

Biological and biochemical studies on JCV, however, have been greatly hampered because JCVcangrowefficiently onlyinprimaryhuman fetalgliacell cultures rich in spongioblasts (17) andbecause viral DNA from the infected cells is usually heterogeneous in size (12, 16). Although JCV has been successfully adaptedtogrowth in

more readily available cells, such as human amnion cells (31) andhumanembryonic kidney (HEK) cells (14), the viral DNA thus obtained is heterogeneous. A JCV mutant growing well in HEKcellshasbeen foundtocontaintwoclasses

of defective DNAs, whichprobably complement

each otherforviralgrowth (38).

Molecularcloning of JCVDNAinabacterial

vector(10)from DNAobtainedby alow

multi-plicity of infection of primary human fetal glia

cells (12) hasnowmade detailed analyses ofthe

JCVgenome possible. Inthepresent study, we

determined the nucleotide sequence of the

re-gion near 0.7 map units fromEcoRI site in an

attempt to correlate the physical map of JCV

DNA withthe genomic organization. From the

similarity to simian virus 40(SV40) and human polyoma virus BK (BKV) DNAs, we could locate thereplication origin and the entire cod-ing region for the smallTantigen. Alsofound in

thisregionwerethesignal sequencesconserved

nearthe splicing sites of eucaryoticmRNA.

MATERIALSANDMETHODS

JCVDNA. Arestriction endonucleasecleavagemap of JCV(Mad-1)DNA(10, 12,22)is shown inFig.1. A recombinantplasmid,pJC(1-4), which hasaJCV DNA insertatthe BamHI siteofplasmid pBR322(10),was

used. DNA was extracted andpurified as described previously (38).

Enzymes. Restriction endonucleases BamHI,

HaeIII, HindIll, HpaI, PstI, andPvuII were

gener-ously provided by Takara Shuzo Co., Ltd., Kyoto,

Japan, andwereusedasrecommendedbythe

manu-facturer. Bacterial alkalinephosphatase and T4 poly-nucleotide kinase were purchased from Boehringei MannheimBiochemicals, Indianapolis,Ind.

DNAsequencing.Nucleotide sequenceswere deter-minedby the method ofMaxamand Gilbert(13) witha

strategyillustrated inFig. 1. pJC(1-4) DNAdigested

withHindIII,BamHI, HpaI,andPstIwassubjectedto

preparative electrophoresis in 5.5% polyacrylamide

gels. Afterelution, HindIII-B and the shorter

HpaI-cleavedHindIII-Awerefurther cleaved each intotwo

subfragments with HaeIII and PvuII, respectively.

Double-stranded fragments isolated from gels were

labeled at their 5' ends with [y-32PIATP (specific

activity, 2,000 Ci/mmol, Amersham International)

andT4polynucleotidekinase. Thelabeled DNA was

denatured by heatandseparated into eachstrand on

73

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

74 MIYAMURA ET AL.

PvujU

PstI

PvuII .-... .I...

HindIll-A

I

HindIll-C Hindi

^_-

-~---*w *- -0 I

lPstI

FIG. 1. Restriction endonuclease cleavage map of JCV(Mad-1) DNA and schematic representation of the strategy forsequencing. Thecleavagesitesarefrompublisheddata(10,12, 22).Thestraightarrowsindicate the direction andapproximateextentof sequence determination from the indicated restriction enzyme sites.

5% polyacrylamide-0.1% bisacrylamide gels. Frag-mentslabeled atoneendweresubmitted tothe four setsof partialdegradation reaction of A > C, G, C, andC+T.Thedegradation productswererunon

0.5-mm-thick 12%polyacrylamide-0.62% bisacrylamide-7

Mureagels.Autoradiographywasdoneat-70°Cwith

anintensifyingscreen (KodakX-Omatic).

RESULTS

LikeSV40 and BKVDNAs, JCV DNA hasa

unique EcoRI site that can be used as a

refer-ence point (Fig. 1). Cross-hybridization under

lessstringent conditions between restriction

en-donuclease cleavage fragments of these three

polyoma viruses has shown that theirphysical

maps could be colinearly aligned by using the

EcoRIsiteasthe0mapposition (11). To locate

the replication origin precisely on the physical

mapofJCVDNA,wedetermined the nucleotide

sequenceof theareapresumably containing the

replication origin and coding for 5' end of the

early mRNAs and compared it with the

corre-spondingareasof BKVandSV40 DNAs, whose

entire sequences and relation to the genetic

maps have been established (5, 21, 24, 37).

Figure 2 shows the nucleotide sequence of 945

basepairscovering theareafrom0.5to0.7map

units.

Theorigin of JCV DNA replicationwas

locat-edneartheHindIII-A/C junction (Fig. 1). Since

the origin of SV40 DNA replication has been

confinedwithinanarrowregion around the BglI

cleavage site(8, 25), the similar nucleotide

se-quence to that region was searched for in the

JCV DNA sequence. The replication origin of

JCVDNA wasassignedtotheregionof

nucleo-tides 36 to 121 (Fig. 2), which shows extensive

homology with the replication origin of both

SV40 (30) and BKV DNA(3). Thestructuresof replication origins are compared among the

threepolyoma virusesin Fig. 3.

The nucleotide sequence of the replication

origincanbe divided into three parts with differ-ent structuralfeatures, ATstretch (nucleotides from 36to50; boxedinFig. 2),twopairsof the

inverted repeat sequences (nucleotides 54 to

102), and a palindromic sequence with mirror image form (nucleotides 105 to 121; boxed by

dotted line in Fig. 2 and 3). Goldberg-Hogness

box structure (24)was detectedattherightend

of the AT stretch (nucleotides 43to50in Fig. 2

and 3). The regionbetween ATstretchand the

palindrome was rich in GC, with intrastrand

symmetrywhich could formtwopairsof double

hairpin loops, asindicatedbythearrowsinFig.

3.

The coding region for small T antigen was

foundtoinitiate from ATGatnucleotides182to

184andterminateatTAAatnucleotides 698to

700 (Fig. 2). This ATG was the first initiation codonin thelongest openreadingframe

down-streamfromthereplicationorigin. Thetopology

oftheinitiation codonissimilartothatof BKV and SV40 (4, 34). Consequently, JC small-T-J.VIROL.

http://jvi.asm.org/

[image:2.489.53.444.68.320.2]

(3)

5' ATGAGCTCAT TACTCGAGTA

CTTACTCAGA GAATGAGTCT

80 GCTTGGCTGG CAGCCATCCC TTCC TTTTT TTTATATATA CAGGAGGCCG AGGCCGCCTC CGCCTCCAAG CGAACCGACC GTCGGTAGGG AAGGGAAAAA AAATATATAT GTCCTCCGGC TCCGGCGGAG GCGGAGGTTC

HindIII

AGTAGTAAGG

GCG$g.GAGGC

TTTTTAGGAG G3CAGGGAAA TTCCCTTGTT TTTCCCTTTT TTGCAGTAAT

TCATCATTCC CGCACCTCCG AAAAATCCTC CGGTCCCTTT AAGGGAACAA AAAGGGAAAA AACGTCATTA

* ...** @

-240

TTTTTGCTGC AAAAAGCTAA

PEACAAA

GTGCTGAATA GGGAGGAATC CATGGAGCTT ATGGATTTAT TAGGCCTTGA AAAAACGACG TTTTTCGATT TTACCTGTTT CACGACTTAT CCCTCCTTAG GTACCTCGAA TACCTAAATA ATCCGGAACT 320 TAGGTCTGCA TGGGGGAACA TTCCTGTCAT GAGAAAAGCT TATCTGAAAA AATGCAAAGA ACTCCACCCT GATAAAGGTG ATCCAGACGT ACCCCCTTGT AAGGACAGTA CTCTTTTCGA ATAGACTTTT TTACGTTTCT TGAGGTGGGA CTATTTCCAC

HindIII

400 GGGACGAAGA CAAGATGAAG AGAATGAATT TTTTATATAA AAAAATGGAA CAAGGTGTAA AAGTTGCTCA TCAGCCTGAT CCCTGCTTCT GTTCTACTTC TCTTACTTAA AAAATATATT TTTTTACCTT GTTCCACATT TTCAACGAGT AGTCGGACTA

1 480

TTTGGTACAT GGAATAGTTC AGAGGTTGGT TGTGATTTTC CTCCTAATTC TGATACCCTT TATTGCAAGG AATGGCCTAA

AAACCATGTA CCTTATCAAG TCTCCAACCA ACACTAAAAG GAGGATTAAG ACTATGGGAA ATAACGTTCC TTACCGGATT

560 CTGTGCCACT AATCCTTCAG TGCATTGCCC CTGTTTAATG TGCATGCTAA AATTAAGGCA TAGAAACAGA AAATTTTTAA GACACGGTGA TTAGGAAGTC ACGTAACGGG GACAAATTAC ACGTACGATT TTAATTCCGT ATCTTTGTCT TTTAAAAATT 640 GAAGCAGCCC ACTTGTGTGG ATAGATTGCT ATTGCTTTGA TTGCTTCAGA CAATGGTTTG GGTGTGACTT AACCCAAGAA

CTTCGTCGGG TGAACACACC TATCTAACGA TAACGAAACT AACGAAGTCT GTTACCAAAC CCACACTGAA TTGGGTTCTT 720 GCTCTTCATT GCTGGGAGAA AGTTCTTGGA GACACCCCCT ACAGGGATCT

AAAGCTTE2

GGTAAACCAC TATTTTTCTT

CGAGAAGTAA CGACCCTCTT TCAAGAACCT CTGTGGGGGA TGTCCCTAGA TTTCGAAATT CCATTTGGTG ATAAAAAGAA HindIII

1 800

TTGCAGGGCA TTTTGTTTTT TACAATAAAC TTAATTGTTT TTTTTTAGGT GCCAACCTAT GGAACAGATG AATGGGAATC

AACGTCCCGT AAAACAAAAA ATGTTATTTG AATTAACAAA AAAAAATCCA CGGTTGGATA CCTTGTCTAC TTACCCTTAG 880

CTGGTGGAAT ACATTTAATG AGAAGTGGGA TGAAGACCTG TTTTGCCATG AAGAAATGTT TGCCAGTGAT GATGAAAACA GACCACCTTA TGTAAATTAC TCTTCACCCT ACTTCTGGAC AAAACGGTAC TTCTTTACAA ACGGTCACTA CTACTTTTGT

CAGGATCCCA ACACTCTACC CCACCTAAAA AGAAAAAAAA GGTAGAAGAC CCTAAAGACT TTCCT

GTCCTAGGGT TGTGAGATGG GGTGGATTTT TCTTTTTTTT CCATCTTCTG GGATTTCTGA AAGGA 5' BamHI

FIG. 2. Nucleotidesequenceof JCVDNA:theregioncontainingreplicationoriginandsmall-T-antigengene.

Numbering isgiven fromneartheHindIII-A/Cjunction(nucleotide 1)to neartheBamHI site(nucleotide 945).We

identified the orientation ofthestrands fromthesimilaritytoBKV andSV40DNAs. The upperstrand,readfrom lefttoright, is believedtohavethesamepolarityasearlymRNA. The AT stretch within thereplication originis

indicatedin thesolid line box(nucleotides36to50). The dotted line box indicates thepalindrome (nucleotides

105to121). ATGatnucleotides182to 184 isconsidered tobe theinitiation codon forTantigen mRNA, and TAA at nucleotides 698 to 700 is considered to be the termination codon. The arrows refer to the splice

junctions forlarge-T-antigenmRNA. Thearrowbetweennucleotides424and 425isfor the donorsite,and the

other, between nucleotides768and769,is forthe acceptorsite(see Fig.6).

http://jvi.asm.org/

(4)

76 MIYAMURA ET AL.

CT

-- TTTTTTT - A TA-

T

A TA

-CAG

GAGGCCGAGGCCGCCTC

CT -- TTTTTTT T A TAA

T

A TA TAA- GAGGCCGAGGCCGCCTC

CT

AA

TTTTTTTT

T A ---- TA TGCA

GAGGCCGAGGCCGCCTC

80

C4GCCTC CAAG T GCCTC CACC

G

GCCTC

TGAG

BglI

CTI TACT m

AGIAAGTAGT|

A

:...

S@*:

_130

7AGGGCGT:

GGAGGCTTTTT AG GAGGfC - AG GGAAA

A|GGGTGT'

GGAGGCTTTTT

CT

GAGG;ZC

T AG CAAAA

A-!GGAGGCTTTTT

TG

GAGGCC

T

AG-* d .... . ...s@*@ *@ @

184 TTCCCTTGTTTTT-CCCT - T TTTTGCA GTAATTTTTTGCTGCAAAAA GC T AA A ATG 3'

CTATTTGGGGAAATCCCTATTC TTTTGCA --- --ATTTTT GC A AA A ATG 3'

---GC TTTTGCA AAAAGC-TTT GC A AA G ATGL 3'

FIG. 3. Comparison of the sequences around thereplicationoriginsofJCV, BKV,andSV40DNAs. Boxes indicate the sequence homology among the three DNAs in this region. Nucleotides have been aligned to

maximize homology. Arrows indicate the areas with intrastrand symmetry which could formhairpin loops.The 17-nucleotide palindromic structures areboxedbydotted lines. TheBglIrestriction site(GCCN5GGC) isfor SV40 DNA. Numbering for JCV DNA is fromFig. 2.InBKVDNA, the first C and the lastGcorrespondto

nucleotides 140 and 5151, respectively, from Seifetal. (24). InSV40 DNA, the first C and the last G are at

nucleotides 5193 and 5079,respectively, from Reddyetal.(21).

antigengene cancode for 172 amino acids, like those of BKV and SV40encoding 172 and 174 aminoacids,respectively. Furthermore, the de-ducedaminoacidsequencesof smallTantigens showed extensive homology among the three polyoma viruses (Fig. 4).

Fromthehomologywith thesignalsequences

proposed forsplicing oflarge-T-antigen mRNA in SV40, BKV, and mousepolyoma virus (23),

we identified the sequence 5'TCAGAG/ GTTGGT3' atnucleotides 419to430as adonor

site and the sequence 5'TTTTTTTAG/

GTGCCA3' at nucleotides 760 to 774 as an

acceptor site oflarge-T-antigen mRNA of JCV

(Fig.5). In the nucleotidesequencedownstream from theacceptorsiteofJCV,onereading frame

was left open and the other two were closed.

The amino acid sequence predicted from the

openreading framedownstreamfromthe accep-torsitewassimilartothosefromBKVandSV40

T antigen genes in the corresponding regions (Fig. 6).

DISCUSSION

The complete nucleotide sequences and the

genomic organization for BKV, SV40, and

mouse polyoma virus have been established,

andseveralrecognitionsequencesforregulatory

functions have been mapped on the nucleotide

sequences(5,21, 24,28, 37). From the

homolo-gydeterminedbyDNA-DNAhybridization, the

cleavage map of JCV DNA has been aligned, with reference to EcoRI site, to those ofBKV

andSV40DNAs (11), but little is known about the genetic map and genomic organization of JCV. In the present study, we attempted to

locate the origin of DNA replication and the proximalpartof theearlyregionof JCVgenome

by comparingJCV sequences tothose of BKV and SV40 to correlate the physical map to the genomicorganization. Apparently,thesimilarity of thesequencesforreplication originandof the aminoacid sequencesforputative small T

anti-gen between JCV and other primate polyoma virusesproves thatJCV hasearlygenome

orga-nization similartothatof BKV andSV40onthe physicalmap of DNA with the conservedsingle EcoRIsite in each genome as areferencepoint. Since the structural features of the region around the HindIII-A/Cjunction of JCV DNA (at nucleotides 36 to121)were strikinglysimilar

to those ofthe replication origins ofBKV and

SV40 (3, 30), this area was regarded as the

replicationorigin of JCVDNA(Fig. 3). Thearea

iscomposed ofalongATstretch, twoinverted

repeatsequences,andapalindrome. At theright

end ofthe AT stretch, the three primate

poly-omaviruses had sequencessimilartothe Gold-berg-Hogness boxsequenceof 5'TATAAATA3' (24). By analogy to SV40 (1), sequence

5'TATATATA3'of JCV(nucleotides43to50) is probably related to the early promoter region.

TheAT-rich sequencewasfollowedbythe

GC-rich region which could form double hairpin loops (Fig. 3). Although it is unclear whether

this structure is actually formed in vivo, it is

JCv 51

BKV 5'

SV40 5'

CTITTCT

CITC

AAGTAGT Al CT ATTC C AG AAGTAGT G

J. VIROL.

http://jvi.asm.org/

[image:4.489.50.446.54.261.2]

(5)

JCV Met Asp Lys Val Leu Asn Arg Glu Glu Ser Met Glu Leu Met Asp Leu Leu Gly Leu Asp Arg Ser Ala Trp Gly

BKV Met Asp Lys Val Leu Asn Arg Glu Glu Ser Met Glu Leu Met Asp Leu Leu Gly Leu Glu Arg Ala Ala Trp Gly

SV40 Met Asp Lys Val Leu Asn Arg Glu Glu Ser Leu Gln Leu Met Asp Leu Leu Gly Leu Glu Arg Ser Ala Trp Gly

Asn Ile Pro Val Met Arg Lys Ala

Asn Leu Pro Leu Met Arg Lys Ala

Asn Ile Pro Leu Met Arg Lys Ala

Tyr Leu Lys Lys Cys Lys Glu Leu His Pro Asp Lys Gly Gly Asp Glu Asp Tyr Leu Arg Lys Cys Lys Glu Phe His Pro Asp Lys Gly Gly Asp Glu Asp

Tyr Leu Lys[ Cys Lys Glu Phe His Pro Asp Lys Gly Gly Asp Glu Glu Lys Met Lys Arg Met Asn Phe Leu Tyr Lys Lys Met Glu Gln Gly Val Lys Val Ala His Gln Pro Asp Phe Gly Lys Met Lys Arg Met Asn Thr Leu Tyr Lys Lys Met Glu Gln Asp Val Lys Val Ala His Gln Pro Asp Phe Gly Lys Met Lys Lys Met Asn Thr Leu Tyr Lys Lys Met Glu Asp Gly Val Lys TyrAla His Gln Pro Asp Phe Gly Thr - Trp Asn Ser Ser Glu Val Gly Cys - Asp - Phe Pro Pro Asn Ser Asp Thr Leu Tyr Cys Lys|Glu

Thr - Trp Ser Ser Ser Glu Val - Cys Ala Asp - Phe Pro Leu Cys Pro Asp Thr Leu Tyr Cys Lys Glu

Gly Phe Trp Asp Ala Thr Glu Val - Phe Ala Ser Ser Leu Asn Pro Gly Val AsAla MetjTyr Cys LysjGln

Trp Pro Asn Cys Ala Thr Asn ProS Val His sPro Cys Leu Met Cys Met Leu Lys Leu Arg His Arg Asn

Trp Pro Ile Cys Ser Lys Lys Pro Ser Val His Cys Pro Cys Met Leu Cys Gln Leu Arg Leu Arg His Leu Asn

Trp Pro Glu C Ala Lys Lys Met S Ala Asn C Ile&s Leu Leu C Leu L Arg Met Lys His Glu Asn

Arg Lys Phe Leu Arg Ser Ser Pro Leu Val Trp Ile Asp Cys Tyr Cys Phe Asp Cys Phe Arg Gln Trp Phe Gly Arg Lys Phe Leu Arg Lys Glu Pro Leu Val Trp Ile Asp Cys Tyr Cys Ile Asp Cys Phe Thr Gln Trp Phe Gly

Arg Lys Leu Tyr A Lys Asp Pro Leu Val Trp Val Asp Cys Tyr Cys Phe Asp Cys Phe Arg Met|Trp Phe Gly

Cys Asp Leu Thr Gln Glu Ala Leu His Cys Trp Glu Lys Val Leu GlAsp Thr Pro Tyr Arg Asp Leu Lys Leu Leu Asp Leu Thr Glu Glu Thr ILeul Gin Trp ITrpI Val Gin Ile Ile Gly Glu Thr Pro Phe Arg Asp Leu Lys Leu Leu Asp Leu Cys Glu Gly ThrL Leu Leu Trp Cys Asp Ile Ile Gln Thr Thr Tyr Arg Asp Leu Lys Leu

FIG. 4. Amino acidsequencesofJCV, BKV,andSV40small Tantigens.Thesequencewasdeducedfrom

the nucleotide sequence.Thesequencesfor BKV andSV40arefrompublisheddata(4, 34, 36). Gapswereleft

wherenecessarytomaximizehomology.The boxes indicate thehomologousamino acidsamongJCV, BKV,and

SV40.Thearrowbetweenglutamineand valine residues divides the amino terminalregion (commontolargeT

antigens)from thecarboxylterminalregionof theseproteins.

possiblethat thehigherstructuresof DNAnear

thereplicationorigin affecttheexpressionof the viralgenome. In the firsthairpin loop indicated

by an arrow in Fig. 3, 22 out of23 base pairs

wereidentical toeach otheramongJCV,BKV, andSV40. Thecomposition ofthe second

hair-pin structure, on the otherhand, was less

ho-mologous, but exactly the same sequence of

5'AAGTAGTAAGGG3' was found. Another

highly homologous sequence among the three polyomavirus DNAswasfound in 17-base pair palindromes. Of17nucleotides, 15 were identi-cal. Althoughthegeneralfeatures ofreplication originweresharedby mouse polyomavirus (6, 29), the strong homology of nucleotide

se-quences was detected only among the three

primate polyomaviruses.

There was a stretchof60nucleotidesbetween

the palindrome and the initiation codon for T

antigens (Fig. 3). Although thehomology in this noncoding regionamongthe threeprimate

poly-omaviruseswasweaker than thatinthe replica-tion origin, they had repeats of5'TnGC3' and

5'GCAn3',which could formhairpin loops (indi-cated by arrows in Fig. 3), in this region. The

structure ofthe corresponding region inmouse

polyomavirusDNA(27) is distinct from those of the primate polyoma viruses.

It seems reasonable to consider the ATG

codon at nucleotides 182to 184 tobean initia-tion signal for translation of T antigen messen-gers for the following reasons: (i) this was the firstinitiationcodon in the only longopen read-ing frame in this region; (ii) its topology was

almost the same as thatof its BKV and SV40

counterparts; (iii) the deduced amino acid

se-quencefor JCV smallTantigen resembled those

of BKV and SV40 (Fig. 4).

Thus, the small-T-antigen gene of JCV can

code for 172 amino acids (Fig. 4). In the

N-terminal regions, where the small T antigen sharesits amino acids with the largeTantigen,

JCV 5' TCAGA TTGGT...TTTTTTTA TGCCA 3' BKV 5' TCAGA TTTGT...TTTTTATAGGTGCCA 3'

SV40 5' ACTGAG TATTT...GTATTTTAG TTCCA 3'

Py 5' TTCCA TAAGA...TTCTTACAG GCTCT 3'

FIG. 5. Splicing signal sequences for large-T-anti-gen mRNA ofpolyoma viruses. The intervening se-quences of JCV, BKV, SV40, and mouse polyoma (Py) virus are 344, 344, 346, and 386 nucleotides long, respectively. The sequences for BKV, SV40, and

mousepolyoma virus are frompublished data (5, 21, 24, 28).

http://jvi.asm.org/

[image:5.489.47.443.61.335.2]

(6)

78 MIYAMURA ET AL.

JCV Val Pro BKV Val Pro

SV40 Il1e Pro

Thr PheAsnGliu Lys

Ser Phe AsnGlu Lys'

A1lA Php Ac,n rlij-l_

I--I I'-u I J'U"AJ-IU U

Asp Asp Len Cy: His Gilu 'la er spAsp Asn rGly AspGiu Asp Leu Cys His Glu Asp Met

'Ala

Ser AspGiuuG1u Ala ThrAla Asp

Asn Leu Cys Ser Glu Glu Met Pro Ser AspAspGlu Ala Ala Asp

Ser Gln His SerThr Pro Pro Lys Lys Lys Lys LysVal Glu Asp Pro Lys Asp Pro...

Ser Gln His SerThr Pro Pro Lys Lys Lys Arg Lys Val Glu Asp Pro Lys Asp Pro ... Ser GlnHis Ser Thr Pro Pro Lys Lys Lys Arg Lys Val GluAsp Pro Lys Asp Pro ...

FIG. 6. Amino acid sequences encoded by the areas just downstream from the splice junction for large-T-antigen mRNA in JCV, BKV, and SV40. Homologous amino acids are shown by boxes. Aromatic amino acids (e.g., tyrosine, tryptophan, and phenylalanine) are shown by shading. Sequences forBKVandSV40 are from published data (5, 21, 24).

JCV and BKV shared 72 of 81 amino acids

(89%). The homology between JCV and SV40

was83%(67of81), and that between BKV and

SV40wasalso83%. Amino acids shared by the threeviruses in theproximal half of the smallT

antigen were as many as 63 (78%). The restof thesmall-T-antigen coding region seemedto be diverged somewhat more from BKV and SV40 than the former region shared by the large T

antigen. In this region, 63 of 91 amino acids (69%)werecommon toJCV and BKV, 49(54%)

to JCV and SV40, and 54 (59%) to BKV and SV40. The three viruses shared only 43 of 91 amino acids(47%) but hada commonstretchof five amino acids, Arg-Asp-Leu-Lys-Leu, fol-lowed by stop codon TAA. These data are

consistent with the earlier reports showing the relatedness ofT antigens of the three primate polyoma viruses by immunological

cross-reac-tivity (32, 35) or 35S methionine-labeled tryptic peptide patterns (26).

The signal sequences for splicing of mRNA are conserved among eucaryotic species, and

the consensus sequences have been proposed

(15). The latterhalf of the small-T-antigengene

is splicedout inlarge-T-antigen mRNAs of

pa-povaviruses. Thesplicing signals for JCV large-T-antigenmRNA agreewellwith those of

poly-oma viruses (23) (Fig. 5) and the consensus

sequences (15). The amino acid sequence

en-coded downstreamfromthesplicing site

resem-bled those of BKV and SV40 (Fig. 6). Like

BKV, SV40, and mouse polyoma virus (24),

JCV had a stretch of aromatic amino acids

sequence in theregion justdownstreamfromthe

splicing signal.

It has not been determined yet whetherJCV

has a virus-coded middle Tantigen like that of

mouse polyoma virus. Downstream from the

splice junction (acceptor site), mouse polyoma virus DNA has two long open reading frames,

which are used for encoding large and middle T

antigens, respectively (28).Inthecorresponding region of JCV DNA,thetwoframesnotusedfor encoding largeTantigenwereshort(Fig. 2),and

the amino acid sequences deduced from these

twoframes (data not shown) hadno homology with that of mouse polyoma virus middle T

antigen (28). The topology ofcodingframes of JCV in this arearesembled those of SV40 and BKV. Weconclude,therefore,thatJCVcannot

code for the protein like mouse polyoma virus middleTantigen.

The present study showed that, in the region

containing the replication origin and small-T-antigengene, the structureof JCVDNA

resem-bles those of the other polyoma viruses. The homology in the nucleotide or amino acid

se-quence washigherbetween theprimate polyoma

viruses than between JCV andmouse polyoma virus (28) andwasthehighest between JCV and BKV. Despite the similarity between JCV and BKVinthisregion, thetwoviruses have

differ-ent pathogenicity in humans and different host

rangesfor cultured cells. Thesequences

respon-sible for the biological difference between the

twoviruses remainto be studied.

A host range mutant ofJCV, JC-HEK, can

grow well in HEK cell cultures (14), whereas wild-type JCV (Mad-1)cangrowefficiently only in primary human fetal glia cell cultures (17). Interestingly,mutantJC-HEK hasnowthesame

hostrange asBKV. Itispossible that

compari-son of the nucleotide sequence between JCV

andJC-HEK, and between JCV andBKV, will

reveal the structure that determines the host

rangeofthese viruses. Astudy ofthe structure

of JC-HEKDNA isnow in progress.

ACKNOWLEDGMENTS

This workwassupportedin partbygrants-in-aidforcancer researchfrom theMinistry ofEducation, Science,and Cul-ture,Japan.

We thank Peter M. Howley, National Cancer Institute, Bethesda, Md., for his generousgiftofcloned JCV DNA.

LITERATURECITED

1. Benoist, C., and P. Chambon. 1981. In vivo sequence requirementsof theSV40earlypromoterregion.Nature (London) 290:304-310.

J.VIROL.

I'

http://jvi.asm.org/

[image:6.489.94.391.72.163.2]

(7)

2. Brown, P., T. Tsai, and D. C.Gajdusek.1975. Seroepide-miology of human papovaviruses. Am. J. Epidemiol. 102:331-340.

3. Dhar, R., C.-J. Lai, and G. Khoury. 1978. Nucleotide sequence of the DNAreplication origin for human papo-vavirus BKV: sequence and structural homology with SV40. Cell 13:345-358.

4. Dhar, R., I.Self,and G. Khoury. 1979. Nucleotide se-quenceof the BK virus DNA segment encoding smallt antigen. Proc. Natl. Acad. Sci. U.S.A. 76:565-569. 5. Fiers, W., R. Contreras, G. Haegeman, R. Rogiers, A. Van

deVoorde, H. Van Heuverswyn, J. VanHerreweghe, G. Volckaert,and M. Ysebaert. 1978. Complete nucleotide sequenceofSV40 DNA. Nature (London) 273:113-120. 6. Friedmann, T., P. LaPorta, and A.Esty.1978. Nucleotide

sequencestudies of polyoma DNA. TheHpaII3/5 junc-tion to theHpaII4/HaeIII18junction,encodingtheorigin of DNAreplicationand the 5' end of theearly region. J. Biol.Chem. 253:6561-6567.

7. Frisque, R. J., D. B. Riflin, and D. L. Walker. 1980. Transformation of primary hamster brain cells with JC virus andits DNA. J. Virol. 35:265-269.

8. Hay, R. T., and M. L. DePamphilis. 1982. Initiation of SV40 DNAreplicationin vivo:location andstructureof 5' ends of DNAsynthesizedin the oriregion. Cell 28:767-779.

9. Howley, P. M. 1980.Molecularbiology of SV40and the humanpolyomaviruses BKand JC, p. 489-550. In G. Klein(ed.), Viral oncology. Raven Press, New York. 10. Howley, P. M., F.Rentier-Delrue, C. A.Heilman,M.-F.

Law, K. Chowdhury, M. A. Israel, and K. K. Takemoto. 1980. Cloned human polyomavirus JC DNA can trans-formhumanamnioncells. J. Virol. 36:878-882. 11. Law, M.-F., J. D. Martin, K. K. Takemoto, and P. M.

Howley. 1979. The colinear alignment of genomes of papovaviruses, JC, BK, and SV40. Virology 96:576-587. 12. Martin, J. D., R. J. Frisque, B. L. Padgett, and D.L. Walker.1979. Restriction endonucleasecleavage map of the DNA of JC virus. J. Virol.29:846-855.

13. Maxam, A. M., and W. Gilbert. 1980. Sequencing end-labeled DNA with base-specific chemical cleavages. MethodsEnzymol.65:499-560.

14. Miyamura,T., K.Yoshilke,and K. K. Takemoto. 1980. Characterization of JCpapovavirus adaptedtogrowthin humanembryonickidneycells. J. Virol. 35:498-504. 15. Mount, S. M. 1982. A catalogue ofsplice junction

se-quences.Nucleic Acids Res. 10:459-472.

16. Osborn, J. E., S. M. Robertson, B. L. Padgett, G. M. ZuRhein, D. L.Walker,andB.Weisblum.1974. Compari-sonof JC and BK humanpapovaviruseswithsimian virus 40:restriction endonucleasedigestionandgel electropho-resis of resultantfragments.J. Virol. 13:614-622. 17. Padgett, B. L., C. M. Rogers, and D.L.Walker.1977.JC

virus,ahumanpolyomavirus associated with progressive multifocal leukoencephalopathy: additional biological characteristics and antigenic relationships. Infect. Im-mun. 15:656-662.

18. Padgett, B.L., and D. L. Walker. 1973. Prevalence of antibodiesin human seraagainst JC virus, an isolate from acaseofprogressivemultifocalleukoencephalopathy.J. Infect.Dis. 127:467-470.

19. Padgett, B. L., and D. L. Walker. 1976. New human papovaviruses. Prog. Med. Virol. 22:1-35.

20. Padgett,B.L.,D. L.Walker,G.M.ZuRhein,andJ.N. Varakis.1977.Differentialneurooncogenicityof strainsof JC virus, a human polyoma virus, in newborn Syrian

hamsters. Cancer Res. 37:718-720.

21. Reddy, V. B., B.Thlimmappaya, R. Dhar, K. N. Subra-mnanan, B. S. Zain, J. Pan, P. K. Ghosh, M. L. Celma, and S. M. Weissman. 1978.The genome of simian virus 40. Science 200:494-502.

22.Rentier-Delrue, F., A. Lubnlecki, and P. M. Howley. 1981. Analysis of JC virus DNA purifieddirectly from human progressive multifocal leukoencephalopathy brains. J. Virol. 38:761-769.

23. Seif, I., G. Khoury, and R. Dhar. 1979. BKV splice sequences based on analysis of preferred donor and acceptorsites. Nucleic Acids Res. 6:3387-3398. 24. Self,I., G. Khoury, and R. Dhar. 1979. The genomeof

humanpapovavirus BKV. Cell 18:963-977.

25. Shortle, D. R., R. F.Margolskee,and D. Nathans.1979. Mutational analysis of the simian virus 40 replicon: pseu-dorevertants of mutants with a defective replication ori-gin. Proc. Nati. Acad. Sci. U.S.A. 76:6128-6131. 26. Shnmons, D. T., and M. A. Martin. 1978. Common

methionine-tryptic peptides near the amino-terminal end ofprimate papovavirustumorantigens.Proc. Natl.Acad. Sci.U.S.A. 75:1131-1135.

27. Soeda, E., J. R. Arrand, N. Smolar, and B. E. Griffin. 1979. Sequence from early region of polyoma virus DNA containing viral replication origin and encoding small, middle and (part of) large T antigens.Cell17:357-370. 28. Soeda, E., J. R. Arrand, N. Smolar, J. E. Walsh, and B. E.

Griffin. 1980.Coding potential andregulatory signalsof thepolyoma virus genome. Nature(London)283:445-453. 29. Soeda, E., G.Kimura,andK.Miura.1978.Similarityof

nucleotide sequences around theoriginof DNA replica-tion in mousepolyoma virus and simian virus 40. Proc. Natl. Acad. Sci. U.S.A. 75:162-166.

30. Subrama ,K.N., R. Dhar, and S.M.Weissman.1977. Nucleotide sequence of afragment of SV40 DNA that contains theoriginof DNAreplicationandspecifiesthe5' ends of "early" and "late" viral RNA. III. Construction ofthe totalsequenceofEcoRII-GfragmentofSV40 DNA. J. Biol.Chem. 252:355-367.

31. Takemoto, K. K., P. M. Howley, and T. Miyamura. 1979. JC humanpapovavirusreplicationin human amnion cells. J. Virol. 30:384-389.

32. Takemoto,K.K., and M.A.Martin.1976. Transformation of hamster kidney cells by BK papovavirus DNA. J. Virol.17:247-253.

33. Varakis, J., G. M. ZuRhein, B. L. Padgett, and D.L. Walker.1978. Induction ofperipheralneuroblastomas in Syrianhamstersafterinjectionas neonateswithJCvirus, ahumanpolyomavirus.Cancer Res. 38:1718-1722. 34. Volckaert, G., A. VandeVoorde, andW. Fkers. 1978.

Nucleotide sequenceof the simian virus 40 small-t gene. Proc. Natl. Acad. Sci. U.S.A.75:2160-2164.

35. Walker, D. L., B. L. Padgett, G.M. ZuRhein, A. E. Albert,andR.F.Marsh.1973. Humanpapovavirus (JC): inductionofbrain tumors in hamsters. Science 181:674-676.

36. Yang, R. C. A., and R. Wu. 1979. BK virus DNA sequencecodingfor theamino-terminus of theT-antigen. Virology92:340-352.

37. Yang, R. C. A., and R. Wu. 1979. BK virus DNA: complete nucleotide sequenceofa humanpapovavirus. Science206:456-462.

38. Yoshlke, K., T. Mlyamura, H. W. Chan, and K. K. Takemoto.1982. Twodefective DNAs of human polyoma-virusJC adapted to growth in humanembryonic kidney cells. J. Virol.42:395-401.

http://jvi.asm.org/