Plasmodium reichenowi, and the human malaria par- asite, Plasmodium falciparum, have a similar relation- ship. The implications of these observations are dis-.
T H EJ O U R N A L OF BIOLOGICAL CHEMISTRY Vol. 263,No. 12,Issue of April 25,pp. 5495-5498. 1988 Printed in U.S. A.
Communication
rozoite-derived vaccines be effective, this work will represent a significantstepintheproduction of a multivalent (i.e. multispecies) anti-malarial vaccine. Previous morphologic, biological, and immunologic obserANIMALRESERVOIRSFORHUMAN MALARIA vations indicated that parasites causing human malaria often PARASITES?* have similar forms in nonhuman primates (12, 13).P . brasilianum is established in monkeys in the wild and clearly has (Received for publication, September 21,1987) evolved in such a fashion that thecourse of a blood infection Altaf A. Lal$$, Vidal F. de la CruzS, of this parasite in amonkey is different from that of P. William E. Collinsn, Gary H. Campbelly, malariae. P . malariae infectionsin monkeys and P . brasiPatricia M. Procellv, and Thomas F. McCutchanS lianum infections in man give low transient infections, while From the $Malaria Division, Laboratory of Parasitic in their designated host these parasites produce chronic inDiseases, National Institute of Allergyand Infectious fections; however, striking similarities between the parasites Diseases, National Institutes of Health, do exist. For example, Coatney et al. (13) compared the oocyst Bethesda, Maryland 20892 and the TMalaria Branch, growth rates of P. malariae and P. brasilianum under identical Division of Parasitic Diseases, Center for Infectious Diseases, Centers for Disease Control,Public Health conditions and found that the mean oocyst diameter of the Service, United States Department of Health and Human two parasites in Anopheles freeborni mosquitoes was almost Services, Atlanta, Georgia 30333 identical. Such a close resemblance was not found in any of the other human and simian malaria parasites. Also, Sodeman We describe here the sequence of the circumsporo- et al. (14) showed morphologic similarities in the exoerythrozoite protein gene of the monkey malaria parasite Plas- cytic bodies of the two parasites. Noting the similarity bemodium brasilianum and show that the immunodomi- tween these parasites, Cochrane et al. (15, 16) showed that nant repeat domain is the same as that of the human monoclonal antibodiesagainstthe surface of P. malariae malaria parasite, Plasmodium malariae. The immu- sporozoites also reacted with P. brasilianum sporozoites and nodominant epitope on the surface of sporozoites of a third species of human malaria parasite has, therefore, vice versa. This further supported the idea that there is an been identified. This geneticbased data and the biolog- evolutionary relationship between these two species of Plasical similarities betweenP. brasilianum and P . malar- modium. Clearly, however, monoclonal antibodies to P. falciiae support their putative zoonotic/anthroponotic re- parum sporozoites cross-react with sporozoites from a rodent lationship. We also show that an ape malaria parasite, malaria parasite, Plasmodium berghei, and anti-CS protein Plasmodium reichenowi, and the human malaria par- antibodies derived from the monkey malaria parasite, Plasasite, Plasmodium falciparum,have a similar relation- modium knowlesi, react with P. falciparum sporozoites (17, ship. The implications of these observations are dis- 18).Unexpected cross-reactions of monoclonal antibodies involving apparently dissimilar antigens is common in situacussed with respect to vaccine development. tions where there is high epitope density as in the casewith repeating epitopes (19, 20). Therefore, it was of interest not only to identify the immunodominantregion of the P . brasiOne research approach to developing an antimalarial vac- lianum CS protein butalso to compare the CS protein of both cine is directed toward neutralizing sporozoite the stage of the P. brasilianum and P . malariae directly at the DNA level. Plasmodium life cycle. Sporozoites, the form inoculated by mosquitoes to initiate the human infection, arecovered with MATERIALSANDMETHODS a protein known as the circumsporozoite (CS)’ protein (1, 2). Cloning and Sequencing-P. brasilianum DNA was isolated and Antibodies to the central repeating immunodominant portion restricted in the presence of AluI endonuclease (Bethesda Research of the protein have been implicated in blocking infection in Laboratories). A 1650-bp DNA fragment containing the CS protein vivo (3-6). The genes for the CS protein from two human gene was identified by Southern blot analysis using the oligonucleoparasites, Plasmodium falciparum and Plasmodium uiuax, tide probe (Region II), CCATG(C/A)GTGTAA(G/T)GGAATGGT. have been cloned (7-11). We now report the sequence of the Fragments in 1300-2000-bp range weresize selected, purified, and Plas- ligated into the SmaI site of plasmid pUC13 and used to transform immunodominant region of the monkey malaria parasite Escherichia coli strain HB101. Colonies weretransferred to nitromodium brasilianum and show by DNA hybridization analysis the cellulose and probed with radiolabeled Region I1 oligonucleotide. A that it is the same as the immunodominant sequence of the positive colony (pPbr4) was identified. The insert fragment was human malaria parasite Plasmodium malariae. Should sposubcloned into MI3 bacteriophage vectors, and the nucleotide se-
Circumsporozoite Protein Gene from Plasmodium brasilianum
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in thispaper has been submitted tothe GenBankTM/EMBLData Bankwith accession number($ 503203. § Supported in part by a grant from the Rockefeller Foundation. Present address: Malaria Dept. Biomedical Research Institute, 12111 Parklawn Dr., Rockville, MD 20852. ‘The abbreviations used are: CS, circumsporozoite; bp, base pair(s); kb, kilobase(s); SDS,sodium dodecyl sulfate.
quence was determined using the Sanger dideoxy procedure (21) and a nonrandom deletion subcloning procedure (22). Southern BlotAnalysis-Genomic DNA of P . falciparum and P h modium reichenowi was digested with RsaI, whereas P. brasilianum and P. malariae DNAs were digested with DraI restriction endonuclease. The fragments were electrophoresed through 1.2% agarose, transferred to nitrocellulose paper, and probed with radiolabeled oligonucleotide probes corresponding to Region 11, the repeat region of P. fakiparum AA(C/T)GCAAA(C/T)GCA(T/C)AA(C/T)GCAAA(C/T)CCC, or P . brasiliunum AA(C/T)GCAGCAGGAAA(C/T)GCAGCAGGA (major repeat), or GATGCAGGAAATGAT (minor repeat), and minor repeat probe, where nucleotide substitutions were introduced at position 3 and 7 (GACGCA CGAAATGAT).
5495
CS Protein Gene f i porn P. brasilianum
5496
Hybridization was carried out in4 X SSC, 0.2% polyvinylpyrrolidone, 0.2% Ficoll, 0.1% SDS, and 0.4% bovine serum albumin at 42 "C for 18 h, except in the case of minor repeat oligonucleotideprobes, where hybridization was conducted at 30 "C. Posthybridization washes were carried out for 15 min X 4 in 0.5 X SSC, 0.1% SDS at room temperature followed by a 3-min wash in 6 X SSC, 0.1% SDS at 5°C below the melting temperature of the oligonucleotides.
brasilianum sequence.Comparison of theputativeanchor sequence to otherCS protein anchor sequences shows considerable conservationof this region. For example, 16 of the last 21 aminoacids at the carboxyl terminus of the P. brasilianum and P. vivax genes are thesame. In this portionof the protein, the P. brasitianum CS protein seems to be more similar t o P. vivax, P. knowlesi, and Plasmodium cynomolgi than to P. falciparum. RESULTS ANDDISCUSSION The immunodominant region of P. brasilianum consists of An oligonucleotide probe corresponding toa conserved re- tandem repeats of tetrapeptides. A 12-bp sequence reiterated gion of the CS protein gene (Region 11) was used to clone a 58 times encodes the sequence Asn-Ala-Ala-Gly. There are six copies of the tetrapeptide repeatAsn-Asp-Ala-Gly result1650-bp AluI restriction endonuclease fragment containing the CS protein gene from P. brasilianumDNA. The fragment ing from a second and third base substitution of the codon was sequenced according to the strategyshown in Fig. 1. The GCA (coding for Ala, A) to GAT (coding for Asp, D). Antiserum from tworabbits immunizedwitha peptide nucleotide sequence and thededuced amino acid sequence of conjugate corresponding to the immunodominant region of the fragment are shown in Fig. 2. One end of the fragment the P. brasilianum CS protein showed surface reactivity with begins at Region I, another conserved region of CS proteins, suggesting that the cloned sequence starts about 300 nucleo- the sporozoites of both P. brasilianum and P. malariae by tides after the 5' end of the gene. T h e sequence shown extends immunofluorescence assay (Table I) using both air-dried and up to theDraI site beyond the termination codon. The 1180- glutaraldehyde-fixed sporozoites at equivalent dilutions. This sporozoites of the two species bp coding for an open reading frame show features common indicated that antibodies to the were cross-reactive with respect to thepolyclonal sera raised to other CS protein genes. For example, the two conserved regions, Region I and Region 11, were both found in the P. againstthe P. brasilianum repeat sequence andthatthe immunodominantrepeat of the CS protein from the two species are identical; however, dueto severalexamples of 0 0.5 1.o 1.5 Kb immunological cross-reactivity between different sporozoite I I I I surface proteins (17, 18), we also investigated the DNA seI quence of the CS protein gene of P. malariae. A restriction I c U * fragment of P. malariae DNA that encoded the CS protein gene was identified on a Southern blot using the Region II"* " specific oligonucleotide probe.Using anoligonucleotide probe " whose sequence encodedthe NAAG and NDAG repeat peptide FIG. 1. Strategy for sequencing the CS protein gene from P.brasilianum. The coding sequence is boxed. The conserved areas of the P. brasilianum CS protein, only the band identified Region I (I)and Region I1 (ZZ)are indicated. The hatched areashows with the Region I1 probe hybridized (Fig. 3 A ) . Both repeatthe repeating regionof the gene. The arrows indicate the DNA and Region 11-specific probes hybridized to a DraI fragment
-D
D
t .
sequence data obtained form the various clones.
"_
.-___-__--.___ >I
-- -- - - -
. REGION I A GCT GTT GAA AAT AAA A V E N K ;---BEGIN 4-MER GCA GGA GGA AATGAT G N O A G GGA AATGCA GCA GGT N A A G G GGA GGT LAC GCA GCA G N A A G GCA GGA GGA AAT GCA G N A A G GCA GGT GGA AAT GCA G N A A G GGT AAC GCA GCA GGA G N A A G GGA AAT GAT GCA GGA N D A G G GGA AAT GCA GCA GGA G N A A G GGA AAT GCA GCAGGA G N A A G GGA AATGCA GCA GGA N A A G G GGA AAT GCA GCA GGA G N A A G GGA AAT GCA GCA GGT G N A A G GGA AAT GCA GCA G N A A GCG M A AAT LAG A K N K AAT AATGAT TCG N N D S
TTG AAA CAA CCC L K Q P REPEATS------> AATGAT GCA GGA N O A G AAC GCA GCA GGT N A A G AAT GCA GCA GGA N A A G AAT L C A GCA GGT N A A G AAC GCA GCA GGA N A A G AAT GCA GCA GGA N A A G AAT GCA GCA GGA N A A G AAT GCA GCA GGT N A A G AAT GAT GCA GGA N O A G AAT GCA GCA GGT N A A G AAT GCA GCA GGA N A A G AAC GCA GCA GGA N A A G
GGA AAT GCA GCA G N A A L I T AATAAA GTG D N K V TCT AAT GGT CCA S N G P
CCC GGA GAT GATGAT G 0 O O P
GAC GCA GGA O A G
GCA GCA GGA AAT GCAOCA GGA A A G N A A G GCA GCA GGA AAT GCA GCA GGA L A G N A A G GAT GCA GGA AAT GCA GCA GGT A 0 G N A A G GCA GCA GGT AAC GCA GCA GGA A A G N A A G GCA GGA GCA GCA GGA AATGAT A A G N O A G GCA GCA GGA AAT GCA GCA GGA A A G N A A G GCA GGT GCA GCA GGA AAT GCA A A G N A A G GCA GCA GGT LAC GCA GCA GGA A A G N A A G GCA GCA GGT AAC GCA GCA GGA A A G N A A G GCA GCA GGT AAC GCA GCA GGA A A G N A A G GCA GCA GGT AAC GCA GCA GGA A A G N A A G GCA GCA GGA AAT GCAGCA GGT A A G N A A G EN0 OF REPEATS-------;: GGT AAC GCA GCA GGA AAT GCA GCA GGA G N A A G N A A G GATGCAAAT ACG AATAAA LAG GAC AAC O A N T N K K O N TCT GAA GAACAT I T A AAG AATTATTTAGAA S E E H I K N V L AAT N AAC N AAT N AAC N AAT N AAT N RAT N AAC N AAT N AAC N AAT N AAT N
AAT GAT GAA N O E AAT N AAT N AAC N AAT N AAT N AAT N AAC N AAT N LA1 N AAT N AAT N AAC N
GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A GCA GCA A A
AATGAAAAA N E K CAGGAAGAA Q E E AGT ATT E S I
61 121
181
AAT
N S AGA AGA R R ACT GAA T E GGA A T 1 G I AAATATTAT ATG A T 1 AAA
I AAG K AT1 I GTA V
T E E U S P C S V T C G S GTT GAT GCA AAAAAT AAG AAA CCT GCA GAA TTAGTTTTA Y D A K N K K P L E L Y L TGT TCACTAGATALA TGC TCC AGT ATA 111 AAT GTC GTA C S L O K C S S I F N V V TTAGTTTTA GTC TTA ATA CTC TTT CAC TAAATAAAT AGC L V L V L I L F H CTT TCT TTT TTC A T 1 CATATATATTTATATATATTTTTT ACTAAT GTT TGC ACTTAATTGTATATATATTATATATATTCAATATATAAT T I C CAG TAT T T T AAA
IFAbwith sporozoites
241
301 361 421
481 511
601 661 121 781 841
901 961
CGT AAT AGT A T 1 ACG GAG GAA TGG TCA CCA TGT AGT GTA ACT TGT GGA AGT GGT I T A AGG 1 0 2 1 R GCT A GAA E TTA L TCG TGA TCT
TABLEI Antibody activity of sera, from rabbits immunized with (Asn-Ala-Alaor KLH, against the Gly) X 5-keyhole limpet hemocyanin (KLH) sporozoites or synthetic peptide (Am-Ala-Ala-Gly)X 5
G I R AGT GAC CTT 1081 S O L AGT L A 1 TCG 1141 S H S ATG TAT CTT 1201
TTT 111 TTG 1 2 6 1 1321 1315
FIG. 2. Nucleotide sequence of the CS protein gene from P. brasilianum. The sequence was determined as indicated in the legend to Fig. 1. Region I, Region 11, and boundaries of the repeats are indicated.
Immunogen"
P. mahriae P . brasilinnum, air-dried
AirGDA dried
(NAAG) x ~
(Am-Ala-Ala-Gly) X 5-KLH R29 preimmunization
