Molecular & Biochemical Parasitology 134 (2004) 225–232

PTRAMP; a conservedPlasmodiumthrombospondin-relatedapical merozoite protein

Joanne Thompsona,∗, Rachel E. Cookeb, Sally Moorea, Laura F. Andersona,Chris J. Janseb, Andrew P. Watersb

a Institute of Cell, Animal and Population Biology, Ashworth Laboratories, The King’s Buildings, Edinburgh EH9 3JT, UKb Leiden University Medical Centre, Postbus 9600, 2300 RC, Leiden, The Netherlands

Received 25 November 2003; received in revised form 10 December 2003; accepted 11 December 2003

Abstract

A gene encoding a 352 amino acid protein with a putative signal sequence, transmembrane domain and thrombospondin structural homol-ogy repeat was identified in the genome of the human malaria parasite,Plasmodium falciparumand the rodent malaria parasite,Plasmodiumberghei. The protein localises in the apical organelles ofP. falciparumandP. bergheimerozoites within intraerythrocytic schizonts andhas, therefore, been termed thePlasmodiumthrombospondin-related apical merozoite protein (PTRAMP). PTRAMP co-localises withthe Apical Merozoite Antigen-1 (AMA-1) in developing micronemes and subsequently relocates onto the merozoite surface. Althoughthe gene appears to be specific to thePlasmodiumgenus, orthologues are present in the genomes of all malaria parasite species examinedsuggesting a conserved function in host-cell invasion. PTRAMP, therefore, has all the features to merit further evaluation as a malariavaccine candidate.© 2003 Elsevier B.V. All rights reserved.

Keywords:PTRAMP;Plasmodium; TSR; Merozoite; Microneme

1. Introduction

Malaria, caused by the protozoan parasite,Plasmodiumfalciparum, is an increasing health burden throughout thedeveloping world and urgently requires the development ofnew control strategies. The release of genome sequences fora number ofPlasmodiumgenomes has provided an unprece-dented opportunity to identify new proteins that are likelyto play vital roles in the parasite’s development.Plasmod-ium recognises and invades a variety of host cell types tocomplete its complex life-cycle. Although the zoites appeardifferent in their morphology and binding affinities, theyall possess a complex of apical organelles, comprising mi-cronemes, rhoptries and dense granules, from which the pro-

Abbreviations:AMA-1, Apical Merozoite Antigen-1; CSP, Circum-sporozite Protein; RAP1, Rhoptry Associated Protein; TSR, thrombospon-din structural homology repeat; PTRAMP,P. falciparumthrombospondin-related apical merozoite protein; TRAP, thrombospondin-related adhesiveprotein

∗ Corresponding author. Tel.:+44-131-650-8662;fax: +44-131-650-6564.

E-mail address:Joanne.Thompson@ed.ac.uk (J. Thompson).

teins mediating motility and host cell invasion are released(reviewed in[1]). The search for vaccine targets has largelyfocused on this important class of proteins.

A number of microneme proteins that are important oressential for invasion ofApicomplexanparasites contain athrombospondin structural homology repeat (TSR) domain;an evolutionarily-ancient motif found in the extracellularor secreted regions of a number of matrix-based cellularsignalling proteins, including thrombospondin, and com-ponents of the complement pathway (Properdin and Com-plement Factors C6–C9). TSRs mediate a wide variety ofadhesive interactions during cellular development, attach-ment and migration and tissue remodelling through bindingto cell and matrix ligands including Fibronectin, Plasmino-gen, matrix metalloproteases, Heparan Sulphate, CD36and other proteoglycans (reviewed in[2–4]). Tan et al.[5]have recently resolved the three-dimensional structure ofthe TSRs of Thrombospondin-1 and have demonstratedthat each forms three anti-parallel strands held togetherby stacked layers of conserved tryptophan and arginineresidues between cysteine disulphide bonds. Using this as atemplate to compare the likely structure of the TSRs recog-nisable in the genomes of diverse organisms from humans

0166-6851/$ – see front matter © 2003 Elsevier B.V. All rights reserved.doi:10.1016/j.molbiopara.2003.12.003

226 J. Thompson et al. / Molecular & Biochemical Parasitology 134 (2004) 225–232

to Drosophila, Caenorhabditis elegansand Plasmodium,they have proposed that TSRs are divided into two majorgroups[5] exemplified by the TSRs of Thrombospondin-1(Group 1) and F-spondin (Group 2). The groups are dis-tinguished structurally and potentially functionally by thestacking of the strands; in Group 1 TSRs, the second andthird strands are linked by a disulphide bond whilst strand2 and the NH2-terminus are linked by two hydrogen bonds;in Group 2 TSRs, a disulphide bond directly links strand 3to the NH2-terminus.

In Plasmodium, TSRs have been identified and mostextensively characterised in the sporozoite protein,thrombospondin-related adhesive protein (TRAP); a con-served microneme protein that is essential for sporozoitemotility and invasion of the mosquito salivary gland andhost hepatic cells[6,7]. TRAP is a Type 1 integral mem-brane protein that is discharged onto the sporozoite surfaceand binds via its Group 2 TSR to sulphated glycosamino-glycans on the host cell surface[8,9]. Subsequent sporozoitegliding motility is thought to be mediated by the translo-cation of TRAP back along the sporozoite surface throughthe interaction of its short, acidic, cytoplasmic tail withan actin/MyoA cytoskeletal motor complex[10]. Group 2TSRs are also present in CTRP, an ookinete protein thatis essential for invasion of the mosquito midgut[11–13]and in thePlasmodiumCircumsporozite Protein (CSP), theprincipal sporozoite surface protein that is involved both insporozoite gliding motility and targeting and in sporozoitedevelopment within the oocyst[14–18]. In addition, a recentsearch of the releasedPlasmodiumgenome has revealed afurther gene, termedPlasmodiumsporozoite protein withan altered TSR domain (PfSPATR), that is reported to beexpressed both on the sporozoite surface and around themerozoite rhoptries in blood-stage malaria parasites[19].

Given the evident importance of proteins with TSRs in thePlasmodiumlifecycle, we have used this motif to monitor theemergingPlasmodiumgenome databases and to identify twofurther PlasmodiumTSR-containing genes. We report thatone of these, termedPlasmodiumthrombospondin-relatedapical merozoite protein, PTRAMP, encodes a TSR that,uniquely to date, most closely resembles the Group 1 sub-set of TSRs characterised in the mammalian proteins throm-bospondin, Properdin and Brain Angiogenesis Inhibitor[5].Like other knownPlasmodiumproteins that contain conven-tional TSRs, PTRAMP is a microneme protein with a mem-brane anchor that relocates to the zoite surface. We show,however, that PTRAMP has a distinctive pattern of expres-sion during the blood-stages of the parasite’s development.

2. Materials and methods

2.1. PTRAMP identification

TSR sequences from TSP-1, TRAP and CSP were usedas TblastN queries to search sequences released by thePlas-

modiumGenome Consortium curated in PlasmoDb ([20];http://plasmodb.org/) and the Sanger Centre parasite genomesites (http://www.sanger.ac.uk/Projects/).

2.2. Parasite strains and library screening

Plasmodium falciparumstrains used were IT0/A4[21]and the reference strain 3D7.Plasmodium bergheistrainsused were ANKA 2.34 and 2.33[22]. The P. bergheige-nomic ANKA 2.34 strain bacteriophage library was a kindgift from M. Ponzi, Istituto Superiore di Sanita, Roma, Italy.The library was screened with a probe corresponding to nu-cleotides−201 to 170 ofpbtrampamplified fromP. bergheiDNA using primers CTTATATGTGAGTGTGTTGC andATTTATGAAACATTCTGGCTC. Two overlapping clonescontaining the entire open reading frame ofpbtrampwereisolated and sequenced.

2.3. Northern blot hybridisation

Plasmodium falciparumstrain IT0/A4 parasites weretightly synchronised by percol gradient separation to isolateschizont stages followed by sorbitol lysis 3–4 h later oncereinvasion had occurred[23]. This stage was taken as timepoint 1. RNA was prepared by trizol extraction using themethods described by Kyes et al.[24]. RNA from∼1.5×107

parasites was resuspended in formamide, separated on a1% gel and transferred to Hybond N+ (Amersham, UK) bythe method of Kyes et al.[24]. The integrity of the RNAon the gel and blot was confirmed by ethidium bromidestaining.

A 1.07 kb fragment of nucleotides−8 to 2073 ofpftrampwas amplified from schizont cDNA using PCR primersTTTTAAAGATGATTGATGTGTTG and TTATGTGAAG-GCCCCTAGTCG and cloned into the TOPO 2.1 vector(Invitrogen) for use as a template in in vitro transcription ofan anti-sense RNA probe. Five hundred nanograms of DNAwas used in the transcription reaction in the presence of T7RNA polymerase and32P-CTP (Amersham, UK) followingthe protocol for the SP6/T7 Transcription kit from Roche.

2.4. PTRAMP antisera production and immunodetection

PTRAMP antiserum was raised in rats by Eurogentec(Bl) using KLH-conjugated conserved PTRAMP pep-tides: CFKGDLKESRP and (C)EKELYENVAGRYMYD.Sera was affinity-purified against peptides conjugated tosulpha-link columns (Pierce, UK) using the manufacturer’sprocedure.

Protein extracts were prepared from percol-purified andsynchronisedP. falciparum IT0/A4 or P. berghei youngtrophozoites and schizonts[25]. Protein extracts were re-solved on 10% bis–tris polyacrylamide gels (Invitrogen) inthe presence on SDS and transferred to invitrolon PVDFmembranes (Invitrogen) according to the manufacturer’sprotocols. Blots were probed with anti-PTRAMP antisera

J. Thompson et al. / Molecular & Biochemical Parasitology 134 (2004) 225–232 227

at a dilution of 1:1000 and developed with a HorseradishPeroxide-conjugated anti-rat antibody (Sigma) and ECLreagents (Amersham, UK) according to the manufacturer’sprotocol.

For subcellular immunolocalisation; thin smears of syn-chronised IT0/A4 and 3D7P. falciparumor in vitro cul-tured P. bergheiparasites were air-dried, fixed with 2%paraformaldehyde/phosphate buffered saline (PBS) for2 min and 90% methanol/10% acetone for 1 min at 4◦Cand air dried. Smears were blocked with 5% normal goatserum in PBS/0.1% Tween 20 (PBS-T) and sequentiallyprobed with a 1:50 dilution of affinity-purified PTRAMP, a1:500 dilution of AL488-conjugated goat anti-rat antiserum(Molecular Probes) and with anti-RAP1[26] that was di-rectly conjugated with Texas Red (FluoReporter® TexasRed®-X protein labelling kit; Molecular Probes), withPBS-T washes between each step. PTRAMP/AMA-1 IFAswere carried out using purified Apical Merozoite Antigen-1(AMA-1) rat monoclonal antisera 4G2 (a kind gift fromDr. M Blackman, NIMR, London) that was biotinylatedand detected with FITC-Streptavidin (Vector Laboratories).Nuclei were stained with 4′,6′-diamidino-2-phenylindole(DAPI).

3. Results

3.1. Identification of a highly-conserved predicted-surfaceprotein containing a TSR motif in Plasmodium ssp.

The TSR domain is an∼60 amino acid region contain-ing ≥12 highly conserved and characteristically spacedresidues, including six cysteines (although the TSRs ofsome complement factors andPlasmodiumproteins havefewer), two to three tryptophans and two arginines. Re-leased sequences from theP. falciparumgenome sequenc-ing project were examined for genes encoding motifs thatconform to this consensus using as queries the TSR motifsof a variety of TSR-containing proteins, including PfTRAPand CSP. Only two previously uncharacterised genes wereidentified; MAL8P1.45 (Chromosome 8) and PfTRAMP(PFL0870w; Chromosome 12). A GSS fragment corre-sponding to the 3′ region of theP. bergheiTRAMP genewas identified (GSS AZ524971;[27]) and used to iso-late clones containingpbtrampfrom a P. bergheigenomiclibrary. pbtramp is a single copy gene located on Chro-mosome 13/14 of the rodent malaria parasitesP. berghei,Plasmodium yoelii, Plasmodium vinckeiand Plasmodiumchabaudi(R. Cooke and C.J. Janse, unpublished observa-tion).

The deduced protein sequences of PfTRAMP and Pb-TRAMP are shown in alignment inFig. 1A. All PTRAMPsare predicted to contain a signal peptide (SignalP, score=3/4), a single transmembrane (TM) domain and a short,somewhat acidic cytoplasmic region, and are, therefore,most likely Type 1 integral membrane proteins. A PFAM

TSR domain is detected with high significance (E-value=7.6e−07) close to the predicted TM domain. PTRAMPsshow ∼45% identity throughout with the highest levelof conservation in the TSR (Id. = 74%) and theCOOH-terminal tail (Id. = 70%). The level of identitybetween the related but speciated rodent malaria parasites,P. yoelii, P. bergheiandPlasmodium chabaudiand primatemalaria parasites,Plasmodium knowlesiand Plasmodiumvivax is notably high throughout (94 and 87%, respectively)indicating that PTRAMPs are likely to have low polymor-phism. No direct homologues of PTRAMP can be detectedin GenBank or within any of the availableApicomplexangenome sequence releases.

3.2. The TSR of PTRAMP fits the consensus for Group 1TSRs

TSR domains are divided into two sub-groups (Groups1 and 2) based on the arrangement of conserved cysteineand aromatic residues that mediate folding of the three ma-jor strands of the structure[5]. The TSR of PfTRAMP isshown in alignment with representatives of otherP. falci-parum and mammalian Groups 1 and 2 TSRs inFig. 1B.The TSR domains of previously describedPlasmodiumproteins fit the Group 2 pattern. In contrast, the TSR do-main of PfTRAMP aligns most closely with the predictedstructure of Group 1 TSR motifs in mammalian proteins,including Brain Angiogenesis Factor 3 (BAI3) and Prop-erdin. PTRAMP and these mammalian TSR domains sharea similar and characteristic arrangement of the six cysteineresidues and possess three highly conserved motifs that areparticularly significant in structure and binding functions.The first, an amino-terminal W**WG*W motif (where *represents any residue), important for strand 1→ 2 link-age and the presentation of a positively charged grooveon the binding face of the domain[5], is well conservedin PTRAMP although the first tryptophan is substitutedby a pair of alternate aromatic amino acids; phenyalanineand tyrosine. In PTRAMP, the second motif, CS/T*TC,shows an unusual threonine→ aspartic acid substitution.Although the functional significance of this substitution isunclear, there are precedents in the TSRs of the mammalianprotein, Semaphorin 5A, and in the thrombospondin-relatedprotein-1 ofToxoplasma gondii(TgTRP-1). A third motifI/R/Q*R*R, thought to intercalate between the conservedtryptophans at the binding face and to form salt bridgesstabilising the second and third strand fold, is present inPTRAMP.

3.3. PTRAMP is expressed in blood-stage parasites

Northern blot analysis of tightly synchronisedP. fal-ciparum blood-stage parasites show thatpftramp is tran-scribed as a single 2.3 kb message in the late schizont stages(Fig. 2A). The microarray data of Winzler and deRisi areconsistent with maximalpftramp expression in the late

228 J. Thompson et al. / Molecular & Biochemical Parasitology 134 (2004) 225–232

Fig. 1. (A) PTRAMP is highly conserved inPlasmodium. The deduced protein sequences of PTRAMP fromP. falciparumand P. bergheiare shown inalignment. Identical and similar residues are highlighted in black and grey. Predicted signal peptide and transmembrane domains are shown in yellow. TheTSRs are boxed in red. Peptides selected for immunisation are highlighted in grey. (B) The TSR of PfTRAMP shows closest similarity to mammalianGroup 1 TSRs. (1) Alignment of the TSR motif of PfTRAMP with representatives of otherP. falciparumTSRs and the most closely related TSRs of anumber of human proteins. Residues are coloured according to their properties by the ClustalW scheme and those that are particularly significant to TSRstructure are highlighted. Predicted disulphide bridges of Group 1 (above) and Group 2 TSRs (below) are shown in green. Conserved tryptophans andarginines (W and R layers) that mediate binding between layers 1 and 2 in Group 1 TSRs are indicated in red. TSP1, thrombospondin; Prop, Properdin;ADAMTS-4, a disintegrin and metalloproteinase with thrombospondin motifs-4; SM5A, Semaphorin 5A; BAI3, Brain Angiogenesis Inhibitor-1.

Fig. 2. (A) PTRAMP is expressed as a single 2.3 kb transcript in mature blood-stageP. falciparumparasites. Lanes 1–8 show RNA taken from 1.5× 107

synchronisedP. falciparumparasites at 6-h time points after merozoite reinvasion throughout the 46 h life-cycle. PTRAMP expression is detectable fromthe onset of parasite replication at 30 h (lane 5) and 36 h (lane 6) of development and is most abundant in mature and segmenting schizonts at 42 h (lane7) and 46 h (lane 8) after invasion, respectively. Marker sizes are shown in kilobase. (B) PTRAMP is expressed as a 48 kDa protein inP. falciparumand P. bergheischizonts. PTRAMP-antisera was used to detect PTRAMP expression in young trophozoites (lanes 1 and 3) or schizont (lanes 2 and 4)extracts of 1× 106 P. falciparum(lanes 1 and 2) orP. berghei(lanes 3 and 4) synchronised parasites. MWs are shown in kilodaltons.

J. Thompson et al. / Molecular & Biochemical Parasitology 134 (2004) 225–232 229

schizont and merozoite stages ofP. falciparum lines HB3and 3D7[28,29]. pbtramp was also detected as a 2.4 kbtranscript in schizont stage RNA isolated fromP. bergheistrains ANKA HP and 233 that are gametocyte-producersand non-producers, respectively (data not shown).

To detect PTRAMP protein expression we raised an-tisera against conserved peptides located within the TSR

Fig. 3. PTRAMP is a merozoite surface protein that originates in the micronemes. Immunofluorescence assays were performed on fixed smears of mixedstage parasites using affinity purified PTRAMP antisera in conjunction with sera detecting the microneme protein AMA-1 or the rhoptry protein RAP1.Black and white images show the immunofluorescence pattern detected with the designated antisera or of the parasite nuclei visualised with DAPI. (Panel1a) PTRAMP (green fluorescence) is detected in the apical organelles (arrowed) ofP. falciparum IT0/A4 early intraerythrocytic schizonts (s) and atthe merozoite surface in mature segmenting schizonts (ss). Earlier stage parasites, including rings (r) do not express PTRAMP. (Panel 1b) PTRAMP isexpressed at the surface ofP. bergheimerozoites. (Panel 2a) PTRAMP (red fluorescence) co-localises with AMA-1 (green fluorescence) in the developingmicronemes (arrowed) of 3D7P. falciparumintraerythrocytic schizonts that are still undergoing DNA replication. (Panel 2b) PTRAMP is detected at thesurface of released merozoites (m) following schizont rupture whilst AMA-1 localises within the micronemes. (Panel 3a) PTRAMP (red fluorescence)shows a distinct localisation pattern from RAP1 (green fluorescence) in 3D7P. falciparum intraerythrocytic schizonts. (Panel 3b) PTRAMP is detectedat the surface of released merozoites (m) following schizont rupture whilst RAP1 localises within the rhoptries.

domain and at the COOH-terminal region. Affinity-purifiedanti-PTRAMP antisera specifically detect a protein ofMr48 kDa in whole extracts of synchronisedP. falciparumandP. bergheischizonts separated under denaturing conditions(Fig. 2B). This result accords with the identification of aPTRAMP peptide present in the Triton-soluble fractionof a P. falciparum 3D7 merozoite preparation[30]. We

230 J. Thompson et al. / Molecular & Biochemical Parasitology 134 (2004) 225–232

found no evidence for processing of PTRAMP using thisantibody.

3.4. PTRAMP is a microneme protein that relocatesto the merozoite surface

The localisation pattern of PTRAMP inP. falciparumand P. bergheischizonts and merozoites was determinedby immunoflorescence assay using PTRAMP antisera, incombination with antisera against the Rhoptry AssociatedProtein, RAP1, and AMA-1, that resides in the micronemesof intraerythrocytic schizonts before relocating to the sur-face of released merozoites via the rhoptry neck[31,32].In early schizonts, PTRAMP co-localises with AMA-1 inapical organelles of fixed and permeabilisedP. falciparummerozoites but shows a distinct localisation pattern to RAP1(Fig. 3). In segmenting schizonts, PTRAMP appears at themerozoite surface in a pattern that is distinct from bothAMA-1 and RAP1. PTRAMP is detectable at the surfaceof both P. falciparumand P. bergheireleased merozoites.Together these results show that PTRAMP is a merozoiteprotein that originates in the micronemes but relocatesto the merozoite surface before erythrocyte invasion isinitiated.

4. Discussion

We have identified a conserved TSR-motif protein inPlasmodiumand have shown that it is released from api-cal organelles onto the surface of merozoites. In earlyintraerythrocytic schizonts, PTRAMP co-localises withAMA-1, a Type 1 integral membrane protein that is aleading vaccine candidate. Unlike AMA-1 or other de-scribed merozoite apical organellar proteins, however,our data indicate that PTRAMP is released onto the sur-face of the merozoite before schizont rupture, raisingthe possibility that it may have an analogous role to thesporozoite TSR protein, CSP, in determination of zoitemorphology [18]. Overall, PTRAMP has both a locali-sation pattern and the structural features to suggest thatit will play an important role in merozoite invasion andblood-stage development. Indeed, attempts to deletepb-tramp by homologous replacement using well establishedgenetic transformation methods were unsuccessful (R.Cooke, J. Thompson and C.J. Janse, unpublished results),providing preliminary evidence that PbTRAMP is essen-tial for asexual development ofP. berghei. Plasmodiumfalciparum is able to utilise a number of alternate path-ways for merozoite invasion[33], so it will be of greatinterest to establish whetherpftramp is also an essentialgene.

PTRAMP is present and highly conserved in all thePlas-modiumspecies for which genome information is availablebut appears to be specific to thegenus. The presence of aTSR places it within the family ofPlasmodiumTSR-domain

proteins that, without exception, have been shown to playcentral roles in host-cell invasion. All previously describedApicomplexanTSRs fall into the classification of Group 2TSRs. The altered thrombospondin motif of PfSPATR lackstwo of the highly conserved cysteine residues that are im-portant in the formation of the overall TSR fold and may,thus, have a divergent structure. In striking contrast, the TSRmotif of PTRAMP contains six cysteine residues arrangedin the conserved pattern of Group 1 TSRs, together withother structurally important Group 1 motifs. It is tempting tospeculate, therefore, that the TSR in PTRAMP has uniquelyevolved towards a Group 1 structure to allow merozoitesto interact with ligands encountered in the blood-stages ofparasite development that are distinct from those encoun-tered or utilised by the TSRs of ookinete or sporozoiteproteins.

In addition to a TSR, two further structural featuresthat PTRAMP shares with members of theApicomplexanTRAP family and CTRP are a TM domain and a short,acidic and highly charged cytoplasmic tail. It is thus no-table that a conserved Y342ENV sequence close to theCOOH-terminal of PTRAMP is analogous to the Y**�

motif (�: hydrophobic amino acid) that targets eukary-otic proteins to various sub-cellular compartments andthe plasma membrane and is essential for the localisationof TRAP to the sporozoite micronemes[34]. In addi-tion, Jewett and Sibley[35] have recently demonstratedthat the cytoplasmic region of theT. gondii TRAP homo-logue plays a key role in sporozoite motility by forminga complex with the Myosin A domain interacting proteinMTIP [10] and the actin binding protein aldolase, therebyproviding a means of linking parasite-substrate bindingwith the locomotive capacity of the parasite actin–myosinmotor. Merozoites are not known to require the glidingmotility of other invasive stages but must, nevertheless,move and re-orientate over the Red Cell surface beforeinvasion [36]. Gene expression analyses provide evidencethat, apart from TRAP, all members of this translocationcomplex are abundantly expressed in merozoites[28–30].Since TSR-family members are critical to the processof host cell invasion by ookinetes and sporozoites, per-haps PTRAMP plays an analogous role in merozoites,acting as the TSR-family protein driving parasite locomo-tion.

An almost overwhelming number of new potentialanti-malarial vaccine targets have been identified as a re-sult of Plasmodiumgenome sequencing and proteomicanalysis. Given the limited resources available, it is cru-cial that candidates are selected carefully for furtherstudy on the basis of their likely essential role in theparasite’s lifecycle and their potential for exposure toa host immune response. The preliminary characterisa-tion of the localisation and predicted structural featuresof PTRAMP presented here show that this protein fitsthese criteria well and, therefore, justifies further evalua-tion.

J. Thompson et al. / Molecular & Biochemical Parasitology 134 (2004) 225–232 231

Acknowledgements

We would like to thank Dr. Mike Blackman (NIMR,London) for his very valuable comments and for provid-ing the conjugated anti-AMA-1 antibody, and Dr. JanaMcBride (ICAPB, Edinburgh) for anti-RAP1 antibodies.We would like to thank Jai Ramesar and Auke van Wicherenfor excellent technical assistance. Sequence data forP.falciparum Chromosome 12 was obtained from the Stan-ford DNA Sequencing and Technology Center website athttp://www-sequence.stanford.edu/group/malaria. Sequenc-ing of P. falciparum Chromosome 12 was accomplishedas part of the Malaria Genome Project with support bythe Burroughs Wellcome Fund. Financial support for thePlasmodiumGenome Consortium was provided by the Bur-roughs Wellcome Fund, the Wellcome Trust, the NationalInstitutes of Health (NIAID) and the U.S. Department ofDefense, Military Infectious Diseases Research Program.This work was funded by the Medical Research Coun-cil of Great Britain and the European Community, grantERB3514PL972944.

References

[1] Blackman MJ, Bannister LH. Apical organelles ofApicomplexa:biology and isolation by subcellular fractionation. Mol BiochemParasitol 2001;117:11–25.

[2] Chen H, Herndon ME, Lawler J. The cell biology of thrombospondin-1. Matrix Biol 2000;19:597–614.

[3] Adams JC, Tucker RP. The thrombospondin type 1 repeat (TSR)superfamily: diverse proteins with related roles in neuronal develop-ment. Dev Dyn 2000;218:280–99.

[4] Silverstein RL. The face of TSR revealed: an extracellular signalingdomain is exposed. J Cell Biol 2002;159:203–6.

[5] Tan K, Duquette M, Liu JH, Dong Y, Zhang R, Joachimiak A, et al.Crystal structure of the TSP-1 type 1 repeats: a novel layered foldand its biological implication. J Cell Biol 2002;159:373–82.

[6] Sultan AA, Thathy V, Frevert U, Robson KJ, Crisanti A, NussenzweigV, et al. TRAP is necessary for gliding motility and infectivity ofplasmodium sporozoites. Cell 1997;90:511–22.

[7] Wengelnik K, Spaccapelo R, Naitza S, Robson KJ, Janse CJ, BistoniF, et al. The A-domain and the thrombospondin-related motif ofPlasmodium falciparumTRAP are implicated in the invasion processof mosquito salivary glands. EMBO J 1999;18:5195–204.

[8] Muller HM, Reckmann I, Hollingdale MR, Bujard H, Robson KJ,Crisanti A. Thrombospondin related anonymous protein (TRAP) ofPlasmodium falciparumbinds specifically to sulfated glycoconjugatesand to HepG2 hepatoma cells suggesting a role for this molecule insporozoite invasion of hepatocytes. EMBO J 1993;12:2881–9.

[9] Pradel G, Garapaty S, Frevert U. Proteoglycans mediate malariasporozoite targeting to the liver. Mol Microbiol 2002;45:637–51.

[10] Bergman LW, Kaiser K, Fujioka H, Coppens I, Daly TM, Fox S,et al. Myosin A tail domain interacting protein (MTIP) localizes tothe inner membrane complex ofPlasmodiumsporozoites. J Cell Sci2003;116:39–49.

[11] Trottein F, Triglia T, Cowman AF. Molecular cloning of a gene fromPlasmodium falciparumthat codes for a protein sharing motifs foundin adhesive molecules from mammals and plasmodia. Mol BiochemParasitol 1995;74:129–41.

[12] Dessens JT, Beetsma AL, Dimopoulos G, Wengelnik K, CrisantiA, Kafatos FC, et al. CTRP is essential for mosquito infection bymalaria ookinetes. EMBO J 1999;18:6221–7.

[13] Templeton TJ, Kaslow DC, Fidock DA. Developmental arrest of thehuman malaria parasitePlasmodium falciparumwithin the mosquitomidgut via CTRP gene disruption. Mol Microbiol 2000;36:1–9.

[14] Dame JB, Williams JL, McCutchan T, et al. Structure of the geneencoding the immunodominant surface antigen on the sporozoite ofthe human malaria parasitePlasmodium falciparum. Science 1984;225:593–9.

[15] Menard R, Sultan AA, Cortes C, Altszuler R, van Dijk MR, JanseCJ, et al. Circumsporozoite protein is required for development ofmalaria sporozoites in mosquitoes. Nature 1997;385:336–40.

[16] Stewart MJ, Vanderberg JP. Malaria sporozoites release circumsporo-zoite protein from their apical end and translocate it along theirsurface. J Protozool 1991;38:411–21.

[17] Tewari R, Spaccapelo R, Bistoni F, Holder AA, Crisanti A. Functionof region I and II adhesive motifs ofPlasmodium falciparumcir-cumsporozoite protein in sporozoite motility and infectivity. J BiolChem 2002;277:47613–8.

[18] Thathy V, Fujioka H, Gantt S, Nussenzweig R, Nussenzweig V,Menard R. Levels of circumsporozoite protein in thePlasmodiumoocyst determine sporozoite morphology. EMBO J 2002;21:1586–96.

[19] Chattopadhyay R, Rathore D, Fujioka H, Kumar S, De La Vega P,Haynes D, et al. PfSPATR, aPlasmodium falciparumprotein con-taining an altered thrombospondin type I repeat domain is expressedat several stages of the parasite life cycle and is the target of in-hibitory antibodies. J Biol Chem 2003;278:25977–81.

[20] Kissinger JC, et al. ThePlasmodiumgenome database. Nature2002;419:490–2.

[21] Roberts DJ, Craig AG, Berendt AR, Pinches R, Nash G, Marsh K,et al. Rapid switching to multiple antigenic and adhesive phenotypesin malaria. Nature 1992;357:689–92.

[22] Janse CJ, Ramesar J, van den Berg FM, Mons B.Plasmodiumberghei: in vivo generation and selection of karyotype mutants andnon-gametocyte producer mutants. Exp Parasitol 1992;74:1–10.

[23] Lambros C, Vanderberg JP. Synchronization ofPlasmodium falci-parum erythrocytic stages in culture. J Parasitol 1979;65:418–20.

[24] Kyes S, Pinches R, Newbold C. A simple RNA analysis methodshows var and rif multigene family expression patterns inPlasmodiumfalciparum. Mol Biochem Parasitol 2000;105:311–5.

[25] Mons B, Janse CJ, Boorsma EG, Van der Kaay HJ. Synchronized ery-throcytic schizogony and gametocytogenesis ofPlasmodium bergheiin vivo and in vitro. Parasitology 1985;91:423–30.

[26] Fonjungo PN, Stuber D, McBride JS. Antigenicity of recombinantproteins derived from rhoptry-associated protein 1 ofPlasmodiumfalciparum. Infect Immun 1998;66:1037–44.

[27] Carlton JM, Dame JB. ThePlasmodium vivaxand P. bergheigenesequence tag projects. Parasitol Today 2000;16:409.

[28] Bozdech Z, Llinas M, Pulliam BL, Wong ED, Zhu J, DeRisi JL.The transcriptome of the intraerythrocytic developmental cycle ofPlasmodium falciparum. PLoS Biol E5 Epub 2003 Aug 18.

[29] Le Roch KG, Zhou Y, Blair PL, et al. Discovery of gene functionby expression profiling of the malaria parasite life cycle. Science2003;301:503–8.

[30] Florens L, P M, Raine JD, et al. A proteomic view of thePlasmodiumfalciparum life cycle. Nature 2002;419:520–6.

[31] Healer J, Crawford S, Ralph S, McFadden G, Cowman AF. In-dependent translocation of two micronemal proteins in developingPlasmodium falciparummerozoites. Infect Immun 2002;70:5751–8.

[32] Bannister LH, Hopkins JM, Dluzewski AR, Margos G, WilliamsIT, Blackman MJ, et al.Plasmodium falciparumapical membraneantigen 1 (PfAMA-1) is translocated within micronemes along sub-pellicular microtubules during merozoite development. J Cell Sci2003;116:3825–34.

[33] Duraisingh MT, Maier AG, Triglia T, Cowman AF. Erythrocyte-binding antigen 175 mediates invasion inPlasmodium falciparum

232 J. Thompson et al. / Molecular & Biochemical Parasitology 134 (2004) 225–232

utilizing sialic acid-dependent and -independent pathways. Proc NatlAcad Sci USA 2003;100:4796–801.

[34] Bhanot P, Frevert U, Nussenzweig V, Persson C. Defective sort-ing of the thrombospondin-related anonymous protein (TRAP) in-hibits Plasmodium infectivity. Mol Biochem Parasitol 2003;126:263–73.

[35] Jewett TJ, Sibley LD. Aldolase forms a bridge between cell surfaceadhesins and the actin cytoskeleton in apicomplexan parasites. MolCell 2003;11:885–94.

[36] Pinder J, Fowler R, Bannister L, Dluzewski A, Mitchell GH. Motilesystems in malaria merozoites: how is the red blood cell invaded?Parasitol Today 2000;16:240–5.