Jun 14, 2002 - parasite development. Charles ... in sporogonic development, PxSR may have a multi- plicity of .... (Fig. 2B). The atomic structure of the SRCR domain of Mac-2 ..... along with the Naval Medical Research Center (USA),.
Blackwell Science, LtdOxford, UKMMIMolecular Microbiology 0950-382X Blackwell Science, 200245Original ArticlePbSR knockout arrests development of malaria in mosquitoC. Claudianos et al.
Molecular Microbiology (2002) 45(6), 1473–1484
A malaria scavenger receptor-like protein essential for parasite development Charles Claudianos,1,2,3* Johannes T. Dessens,1 Holly E. Trueman,1 Meiji Arai,1,4 Jacqui Mendoza,1 Geoff A. Butcher,1 Tessa Crompton1 and Robert E. Sinden1 1 Department of Biological Sciences, Imperial College of Science Technology and Medicine, London SW7 2AZ, UK. 2 Research School of Biological Sciences, Australian National University, Canberra, ACT 2601, Australia. 3 CSIRO Division of Entomology, Canberra, ACT 2601, Australia. 4 Department of Medical Zoology, Jichi Medical School, Tochigi 329–0498, Japan.
Summary Malaria parasites suffer severe losses in the mosquito as they cross the midgut, haemolymph and salivary gland tissues, in part caused by immune responses of the insect. The parasite compensates for these losses by multiplying during the oocyst stage to form the infectious sporozoites. Upon human infection, malaria parasites are again attenuated by sustained immune attack. Here, we report a single copy gene that is highly conserved amongst Plasmodium species that encodes a secreted protein named PxSR. The predicted protein is composed of a unique combination of metazoan protein domains that have been previously associated with immune recognition/ activation and lipid/protein adhesion interactions at the cell surface, namely: (i) scavenger receptor cysteine rich (SRCR); (ii) pentraxin (PTX); (iii) polycystine-1, lipoxygenase, alpha toxin (LH2/PLAT); (iv) Limulus clotting factor C, Coch-5b2 and Lgl1 (LCCL). In our assessment the PxSR molecule is completely novel in biology and is only found in Apicomplexa parasites. We show that PxSR is expressed in sporozoites of both human and rodent malaria species. Disruption of the PbSR gene in the rodent malaria parasite P. berghei results in parasites that form normal numbers of oocysts, but fail to produce any sporozoites. We suggest that, in addition to a role in sporogonic development, PxSR may have a multiplicity of functions. Accepted 14 June, 2002. *For correspondence. E-mail claudianos@ rsbs.anu.edu.au; Tel. (+61) 2 61255065; Fax (+61) 2 61258294.
© 2002 Blackwell Science Ltd
Introduction The malaria life cycle in the mosquito vector starts with the uptake of gametocytes from parasite-infected blood. After undergoing gametogenesis and fertilization, ookinetes are formed in the mosquito midgut. Ookinetes cross the midgut wall and transform into oocysts under the basal lamina (Sinden, 1999). Each oocyst produces up to 8000 sporozoites. Sporozoites must traverse the mosquito haemolymph and salivary gland tissues before they can enter the vertebrate host via mosquito bite. In the mammalian host, sporozoites are transported from the site of bite via the bloodstream to hepatocytes where they develop into pre-erythrocytic liver schizonts. Each liver schizont produces up to 4000 merozoites, which upon release into the blood stream initiate the erythrocytic cycle, the prime cause of clinical symptoms in malaria patients. As various Plasmodium genome projects near completion it is becoming apparent that parasite gene sequences are being identified or categorized as having bacterial, plant or even vertebrate origins. In some cases the evolutionary origins of whole protein families are being reassessed. The origins of many other sequences remain uncertain; of these unknowns many will inevitably define new gene and protein families. However, the atypical nature of some sequences is suggestive of lateral gene transfer events. In Plasmodium, the enzyme enolase is thought to be the result of lateral gene transfer from vertebrates (Keeling and Palmer, 2001). The emergence of pathogenic strains of bacteria has been associated with the acquisition of numerous virulence factors by lateral gene transfer (Ochman, 2001). Similarly, putative lateral acquisition or mimicry of host immune factors has been observed in helminth and nematode worms (Riffkin et al., 1996; Maizels et al., 2001). Aside from atypical immune factors found in pathogens the origins of immunity is increasingly understood to have occurred in early metazoans (Pancer et al., 1999; Müller, 2001). Recent studies of marine sponges and sea urchins highlight a diversity of proteins that contain scavenger receptor cysteine-rich (SRCR) domains. Notable among these are extracellular proteins associated to sperm and to primitive immune cells known as coelomocytes (Müller, 1997; Pancer, 2000). Many of these molecules are mosaic proteins, the genes of which have complex transcriptional profiles. The
1474 C. Claudianos et al. dynamic expression and diversity of SRCR proteins is thought to be associated with the evolution of antipathogenic defence albeit such multifaceted proteins are expected to have more than one function (Pancer, 2000). In this paper we report and characterize a novel Plasmodium multidomain scavenger receptor-like protein, named PxSR, the structure of which is reminiscent of vertebrate and invertebrate proteins involved in immunity. Here we show that disruption of the PbSR gene in the rodent malaria parasite P. berghei blocks sporozoite development. The biological significance of this observation in the context of its structural features is discussed. Results Identification of PxSR To investigate whether Plasmodium possessed genes potentially involved in immune modulation we searched the P. falciparum genome for SRCR domains, which are frequently found in proteins involved in immunity. A single putative gene (PfSR) containing two tandem SRCR domains was predicted from sequence data of chromosome 14. A putative orthologue was subsequently identified in the rodent malaria species P. berghei by screening an ookinete cDNA library with a heterologous PfSR probe. The mRNA sequence of PbSR (GenBank accession number AY034780) contains an open reading frame encoding a 1304 amino acid protein with a calculated Mr of 148 247. The encoded protein contains a predicted 22 amino acid N-terminal signal peptide, but no identifiable transmembrane or glycosylphosphatidylinositol (GPI) domains, indicating the protein is secreted and soluble. Confident prediction of PySR, PkSR genes and a partial PcSR gene was achieved via homology searches of the P. yoelii, P. knowlesi and P. chabaudi genome projects. A putative orthologue was similarly identified from data made available by Cryptosporidium parvum genome project. More recently, homologous sequences from other Apicomplexa parasite genome projects, including Toxoplasma gondii and Theileria parva have been identified, although these
partial sequences cannot confirm a complete gene. Protein comparisons show that PbSR is 95% identical to PySR, consistent with their close ancestry, while identities with PfSR PkSR and CpSR are 64%, 58% and 33% respectively (Fig. 1A). Structural characterization of PxSR The PxSR proteins are characterized by eight peptide modules corresponding to four different protein families (Bateman et al., 2000), namely SRCR, PTX, LCCL and LH2/PLAT (Figs 1 and 2). Extensive homology searches of bacterial, protist and metazoan genome projects failed to detect any genes with a similar modular composition to PxSR outside the Apicomplexa. In the near complete P. falciparum, P. yoelii and C. parvum genomes the LH2/ PLAT, SRCR and PTX domains appear to be unique to PxSR, while there are two other duplicated sequences that encode a putative LCCL domain (data not shown). Thus, from available sequence data it appears that PxSR is encoded by highly conserved single copy genes that constitute a unique and novel gene family of Apicomplexa parasites. The SRCR domain is a highly conserved protein module of approximately 110 amino acids in length containing six conserved cysteine residues that form three intradomain disulphide bonds (Resnick et al., 1994). The SRCR domain defines an ancient superfamily of proteins that occur in a diverse range of organisms from marine sponges to vertebrates. SRCR proteins have often been used to help delineate the metazoan threshold of life (Müller, 2001). SRCR domains are divided into two subfamilies, group A and B. Both groups have three disulphide bridges in common, but group B SRCRs contain a unique pair of cysteine residues that form an additional disulphide (Resnick et al., 1994). SRCR domains are thought to mediate protein-to-protein and ligand binding interactions. Cell-surface and secreted members of the SRCR protein family (Pfam 00530) include CD5, CD6, Spa (Figs 2 and 3), gp340, AIM, macrophage scavenger receptor type I, CD163, complement factor I and the
Fig. 1. Protein sequence analysis of PxSR. A. CLUSTAL W multiple alignment of PbSR, PySR (chrPy1 453), PfSR (chr14_C14m33), PkSR (Sanger_PKN.0.008677) and CpSR (cparvum_Contig 1948). Identical residues to PbSR are represented by (.) and insertion/deletion gaps by (–). A quantitative consensus of PxSR sequences, interpreted using a 66% identity threshold, is shown below the alignment. Conserved domain homology patterns were identified from PSI-BLAST protein searches (Blosum80 matrix) and SMART and PFAM database searches. Colour-coded domains constitute signal peptide (red); LH2/PLAT (green); LCCL (black); SRCR (blue); PTX (magenta). Cysteine motifs of LCCL and SRCR domains are highlighted in yellow. Grey boxes mark conserved peptide sequences used for polyclonal antibody production in rabbits (Eurogentec). B. Schematic diagram of PxSR proteins showing the modular structure, colour-coded as in (A). C. PDB Brookhaven and the Molecular Modelling DataBase (MMDB) at NCBI, were used to construct homology models based on PDB:1BU8_A (pancreatic lipase); PDB:1JBI (inner ear protein; Coch-5b2); PDB:1BY2 (Mac-2 binding protein; M2BP); and PDB:1SAC_A (serum amyloid protein; SAP). MMDB alignments were interpreted using the RasWin program. Shown are regions of high confidence mapping (>85% of primary sequence aligns to known 3D-structure) that functionally annotate PxSR domains to the colipase binding domain of pancreatic lipase, the interaction site of the SAP homopentamer, confirming the disulphide bridge structure of LCCL and intradomain cysteine pattern of SRCR domains (see also Fig. 3A). © 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
PbSR knockout arrests development of malaria in mosquito 1475
© 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
1476 C. Claudianos et al. Fig. 2. A. Schematic model of PxSR, together with various metazoan proteins that share homologous domains (in black). Other domains are given with accession codes as described in the SMART database and include: EGF, epidermal growth factor-like domain (SM0181); Cht, chitin-binding type 2 domain (SM0494); Try_SPc, trypsin-like serine protease domain (SM0020); LDLa, low-density lipoprotein receptor class A domain (SM0192); CLECT, C-type lectin domain (SM0034); CCP, complement control protein domain (SM0032); and vWA, von Willebrand factor type A domain (SM0327). Signal and transmembrane domains are also shown. B. A list of homologies: pairwise comparison of related proteins/domains with PxSR consensus sequence (Fig. 1A), indicating amino acid identity and similarity scores.
sperm SPERACT egg-peptide receptor of sea urchins (Catterall et al., 1987; Dangott et al., 1989; Freeman, 1994; Droste et al., 1999; Hohenester, et al., 1999; Holmskov et al., 1999; Miyazaki et al., 1999; Gebe et al., 2000). Database searches indicate PxSR most closely resembles SRCR domain-containing proteins involved in development and regulation of the mammalian immune system. The highest homology score in terms of size and identity (32%) was to the region containing SRCR domains 2 and 3 of the mouse lymphoid protein, Spa (Fig. 2B). Spa is a secreted protein comprised of three tandem group B SRCR domains and is closely related to the ectodomain of lymphocyte antigens CD5 and CD6 of vertebrates (Gebe et al., 2000). To date, group B SRCR domains have been uniquely found in proteins implicated in immunity. Further searches using the separated domains of PxSR confirmed SRCR 1 had highest homology to domain 2 of Spa while SRCR 2 had highest homology to domains 6 and 7 of the SpSRCR7 variant 1 protein from the sea urchin, Stongylocentrotus purpuratus (Pancer, 2000) (Fig. 2B). The atomic structure of the SRCR domain of Mac-2 binding protein (M2BP), a tumour-associated antigen and
matrix protein has been resolved (Hohenester et al., 1999). Molecular modelling using the three-dimensional structure of M2BP helped us identify structural features in the corresponding primary protein sequence of PxSR and related SRCR domains (Fig. 1C and Fig. 3). M2BP has been previously used as a structural prototype in a functional study of the CD6 receptor (Bowen et al., 2000). An amino acid sequence alignment of PxSR shown together with closely related group A and B SRCR domains, including those from M2BP and CD6, draws attention to regions of conservation and variation (Fig. 3A). Intuitively, variation most likely underpins functional specificity between related proteins. The multiple alignment of SRCR domains shows regions characterized by extensive variation (replacements, sequence deletion and insertion) occur in the middle and C-terminal regions of these domains. The SRCRs of PxSR contain a conserved eight-cysteine motif (Fig. 1A). However, the position of the two additional cysteines is distinct to those of group B SRCRs (Fig. 3A). The SRCR domains of PxSR may thus represent a novel subfamily within the SRCR superfamily. We propose these be known as ‘Group C’ SRCR domains. Confident alignment and subsequent structural mapping show the unique cysteines of apicom© 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
PbSR knockout arrests development of malaria in mosquito 1477 Fig. 3. Analysis of the SRCR domain. A. Structurally based sequence alignment of selected group A and B domains with SRCRs of PxSR. Sequences were extracted from the following proteins found in GenBank; PxSR consensus (Fig. 1A); human Spa (U82812); mouse Spa (AF018268); sponge MAP_GEOCY (CAC38754); human CD5 (P06127); human CD6 (X60992); Drosophila GRAAL (AJ251803); sea urchin SpSRCR7 (AF228824); sea urchin speract receptor-sp85 (AAA75510) and human M2BP (1BY2). Where appropriate, sequence descriptors include species initials, protein name and the corresponding number of the SRCR domain occurring from the N-terminus of that protein. Conserved residues shaded black and less conserved residues shaded grey represent thresholds of >95% and >40% respectively. Location of the six conserved Cys residues that characterize SRCR are marked by , the two diagnostic group B Cys by and unique Cys residues of PxSR by . Secondary structural elements b (sheets) and a (helix) of M2BP are shown below alignment. B. Unrooted distance neighbour-joining tree showing phylogenetic relationship of PxSR domains in relation to SRCR domains of related vertebrate and invertebrate proteins. Parsimony analysis using the same dataset produced a congruent tree (data not shown). Distance/ Parsimony bootstrap values of >70% are indicated at nodes. Scale bar indicates a distance of 0.1 amino acid substitutions per position in the sequence. C. 3 D cartoon structure of M2BP (PDB; 1BY2) generated using the RasWin program, showing structural elements and linking loop regions. Six conserved Cys residues that characterize SRCR domains are shaded black and numbered according to (A), also shown is the loopout region (grey) that corresponds to the hypervariable b4–b5 region. A solvent-accessible model (not shown) shows the b4–b5 region remains largely exposed at surface of the molecule.
plexan SRCRs (residue 72 and 81; Fig. 3A) occur in a hypervariable region between sheets b4 and b5 of M2BP (Fig. 3C). Intriguingly, domains 2 and 3 of the Spa protein also contain cysteine residues at equivalent positions in the alignment. Mutagenic analyses show ligand binding interactions are critically associated with residues that occur in the hypervariable regions of SRCRs from CD6 (Bowen et al., 2000). The putative loop-out region be© 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
tween sheets b4 and b5 of the SRCR domain (Fig. 3C) may thus be important in specifying the ligand interactions of PxSR proteins. The unique cysteine residues contained in this region may form a new intradomain disulphide, alternately they may remain free to interact with other proteins or other domains within PxSR. Phylogenetic analyses of the 142-residue alignment above (Fig. 3A) show PxSR domains are more closely related to each other
1478 C. Claudianos et al. than to other SRCR domains. The tandem organisation of these domains is most likely the result of a domain duplication event. The SRCRs of PxSR are also shown as the most distant group in phylogeny though more closely related to domain 2 of Spa than to other SRCR from vertebrates or invertebrates (Fig. 3B). LCCL domains are characterized by a four-cysteine residue motif (Trexler et al., 2000). The LCCL module is thought to be important for the structural integrity of multidomain proteins, many of which contain immune factors such as complement proteases with antimicrobial properties. The LCCL domains of Limulus clotting factor C protein and late gestation lung protein are associated with antibody-independent host defence mechanisms, which are believed to recognise cell-surface carbohydrates of invading pathogens (Trexler et al., 2000). The structure of the LCCL module of the human cochlear protein Coch5b2 protein (Fig. 2) has recently been resolved (Liepinsh et al., 2001). PxSR contains four LCCL domains with high structural homology to Coch-5b2. Molecular modelling of the LCCL-4 domain of PxSR on that of Coch-5b2 clearly shows the conserved four-cysteine motif that forms two disulphide bridges (Fig. 1C). PTX domains belong to a family of proteins (Pfam 00354) that are involved in acute immunological responses including agglutination, bacterial capsular swelling, complement activation and phagocytosis. Members typically contain one PTX domain. The individual monomers are organised in a discoid arrangement of five or eight non-covalently bound subunits. Members of this family of serum proteins include the opsonins C-reactive protein (CRP) and serum amyloid protein (SAP) (Fig. 2) (Bharadwaj et al., 2001). Structural similarities also exist with proteins such as Limulus SAP (Shrive et al., 1999) and the predicted Drosophila melanogaster b6 pentraxinlike gene (FlyBase; FBgn0024897). Molecular modelling of the PxSR PTX domain on human SAP (Emsley et al., 1994) indicates these domains have similar structure (Fig. 1C). To our knowledge, PxSR is the first report of a PTX domain within a large multidomain protein. LH2/PLAT domains are found in a variety of membrane or lipid-associated proteins of plants, metazoans and some pathogenic bacteria (Pfam 01477). These domains are implicated in protein–lipid interactions (Bateman and Sandford, 1999). The amino-terminal domain of lipoxygenase enzymes has structural homology to the carboxyterminal domain in bacterial alpha toxin and mammalian pancreatic lipases (Fig. 2). In lipoxygenase this domain binds to colipase, which mediates binding to membranes (van Tilbeurgh et al., 1999). Molecular modelling of the PxSR LH2/PLAT domain on the atomic structure of rat pancreatic lipase (Roussel et al., 1998) again indicates these domains have similar structures, which may reflect similar function (Fig. 1C).
Expression of PxSR We investigated the expression of PbSR by reverse transcription-PCR using total RNA isolated from purified P. berghei asexual blood stages (derived from the nongametocyte producing clone 2.33), gametocytes, ookinetes, day 5 and day 15 oocysts and day 23 sporozoites from infected salivary glands. The results indicate that PbSR transcript is present in all these life stages (Fig. 4A). The highest specific amplification level with respect to the cytoskeletal reference gene tubulin-1 was found in day 15 oocysts that typically contain sporozoites (Sinden, 1981) (see also Fig. 6). The tubulin-1 specific primers (flanking an intron) uniquely amplified a product of a size consistent with that of its mRNA (Fig. 4A), indicating that amplification was derived from cDNA and not contaminating genomic DNA. A limited epitope anti-PxSR immune serum raised against two conserved peptides of PxSR (Fig. 1A) readily stained sporozoites from newly ruptured oocysts and salivary gland sporozoites in both P. berghei and P. falciparum (Fig. 4B and C), while preimmune serum did not stain (data not shown). Interestingly, PxSR staining appeared polarized towards the apical end in many sporozoites examined, indicating that the protein may be targeted to the apical secretory organelles (Fig. 4B). Immunoblot analysis of P. berghei sporozoite-infected salivary glands revealed two bands of approximately 40 and 60 kDa that were absent in uninfected glands (Fig. 4D), supporting sporozoite expression of PbSR and suggesting that the protein may be post-translationally modified. Complex post-translational modification has been reported in other modular SRCR domain-containing proteins like Anopheles gambiae Sp22D (Danielli et al., 2000) (Fig. 2). Alternatively, these bands may constitute degradation products from the full-length PbSR protein. PbSR protein was conclusively detected only in sporozoites, despite PbSR messenger being present in the other life stages studied (Fig. 4). This may indicate that transcription and translation of the PbSR gene are uncoupled, as has been reported for other P. berghei genes such as Pbs21 and Pbs25 (Paton et al., 1993; del Carmen Rodriguez et al., 2000).
PbSR gene disruption To investigate possible functions of PxSR we generated transgenic PbSR-disrupted P. berghei ANKA clone 2.34 parasites by insertion of a modified Toxoplasma gondii dihydrofolate reductase/thymidylate synthase gene (DHFR/TS) cassette into PbSR by double homologous recombination (van Dijk et al., 1995) (Fig. 5A). The DHFR/ TS cassette was inserted between nucleotide positions 880 and 2776 of PbSR, thereby removing 1.9 kb of PbSR central coding sequence. Two independent clonal © 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
PbSR knockout arrests development of malaria in mosquito 1479 Fig. 4. Expression of PxSR. A. Multiplexed reverse transcription-PCR of PbSR (550 bp) from total RNA samples isolated from purified asexual blood stages (derived from a non-gametocyte producing clone), gametocytes, ookinetes, day 5 and day 15 oocysts from mosquito midgets, and day 23 sporozoite infected salivary glands. RNA amounts were normalized for the cytoskeletal tubulin-1 mRNA (Tub-1; 300 bp). B. FITC immunofluorescent-antibody staining of P. berghei day-23 salivary gland sporozoite, arrow indicates apical end. C. Field of P. falciparum day-23 salivary gland sporozoites similarly stained as in (B). Bars equal 10 mm (D) Immunoblot of uninfected (ui) and P. berghei wild-type parasite-infected (i) mosquito salivary glands (glands of 30 mosquitoes per sample, infected sample dissected at 23 days post infection), showing bands at approximately 40 and 60 kDa.
pyrimethamine-resistant parasite populations (named PbSR ko, clones 5 and 6) were assessed by Southern blot analysis of HindIII-digested genomic DNA. A PbSR probe gave rise to a single band in parental (wild-type) parasites and no bands in the PbSR ko parasites (Fig. 5B), demonstrating the successful deletion of the central PbSR coding sequence via insertion of the DHFR/TS cassette. Conversely, a DHFR/TS probe gave rise to a single band in the PbSR ko parasites and no signal in wild-type parasites, confirming successful integration (Fig. 5B). PbSR-disrupted phenotype PbSR ko parasites developed fully in the erythrocytic cycle and formed normal numbers of gametocytes in mice. Upon infection of Anopheles stephensi mosquitoes © 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
PbSR ko parasites developed oocysts in numbers comparable to wild-type parasites at 10-days post infection (Table 1). However, a clear phenotype was observed on closer examination of the oocysts. In sharp contrast to wild-type oocysts, PbSR ko oocysts did not produce any sporozoites (Fig. 6A). Analysis of the nuclear organisation in the PbSR ko oocysts by DNA staining showed similar developmental progression to that in wild-type oocysts (data not shown). However, we found no evidence of cytokinesis and subsequent sporozoite formation (Fig. 6A). Indeed, PbSR ko parasite-infected mosquitoes were unable to infect mice by mosquito bite (data not shown), an observation consistent with the absence of sporozoite development. Both independent PbSR ko clones behaved in the same way, indicating the PbSR ko phenotype is not the result of clonal variation. Thus, PbSR
1480 C. Claudianos et al. gene from the rodent malaria parasite P. berghei. Highly conserved orthologues exist in other Apicomplexa species including the human malaria parasite P. falciparum. We show that PxSR proteins are expressed in the infective sporozoite stage. PbSR-disrupted parasites develop fully during the vertebrate erythrocytic cycle and undergo apparently normal gametogenesis, fertilisation, ookinete development and oocyst transition in the mosquito. However, PbSR-disrupted parasites fail to develop past the oocyst stage. PxSR is only the third malaria sporozoite protein to be functionally characterized via targeted gene disruption, the others being circumsporozoite protein (CS) (Menard et al., 1997) and thrombospondin-related adhesive protein (TRAP) (Sultan et al., 1997).
Fig. 5. Construction and analysis of PbSR disrupted parasites. A. Schematic diagram of the transfection plasmid pPbSR and the integration of the modified Toxoplasma gondii dihydrofolate reductase-thymidylate synthase gene (DHFR/TS) cassette, conferring pyrimethamine resistance, into PbSR (grey) by double homologous recombination. Recombination sequences (dark grey), crossover sites (crossed lines), crossover positions (small arrows), and probes used in Southern blot analyses (thick lines) are shown. B. Southern blot analysis of HindIII-digested genomic DNA from wildtype and PbSR ko parasite clones 5 and 6 with probes indicated in (A).
is an essential molecule for sporozoite development in mosquitoes, an observation consistent with its expression in this life stage (Fig. 4). Oocyst growth curves diverged after 5 days post infection, approximately the time when sporozoites normally start to appear in P. berghei occysts (Sinden, 1981). Interestingly, PbSR ko oocysts reached a significantly (35%) greater size than wild-type oocysts (Fig. 6B).
Fig. 6. PbSR knockout phenotype in mosquito. A. Twenty-day-old oocysts in Anopheles stephensi show the presence of mature sporozoites (s) budding off sporoblast (sb) in wild-type oocysts, and the absence of sporozoite formation in PbSR ko oocysts. B. Oocyst development time course in A. stephensi infected with wildtype (squares) or PbSR ko (circles) parasites. Each time point shows mean oocyst size of 100 oocysts. s.e.m. were negligible.
Discussion We have cloned and functionally characterized the PbSR
Table 1. Effect of PbSR disruption on P. berghei infectivity to A. stephensi. Mean number of oocysts ± SEM from mosquito infected midguts (number of midguts dissected)b Expt. 1 2 3 4
Type of feed (clone)a gct (5) gct (6) ook (5) ook (6)
Wild type 18.0 ± 111 ± 38.7 ± 46.6 ±
2.2 6.8 3.3 3.2
PbSR ko (149) (97) (106) (105)
2.7c
39.2 ± (135) 143 ± 8.6 (98) 43.7 ± 3.1 (120) 41.7 ± 2.8 (106)
% of wild type 217 129 113 89
a. Gametocye (gct) or ookinete (ook) feed, and clone used. b. Each experiment is based on pooled data from three mice (gct) and three membrane feeders (ook). c. Significantly different (P < 0.001) from value for wild type infected control group as calculated by Mann–Whitney U-test. © 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
PbSR knockout arrests development of malaria in mosquito 1481 There are obvious questions concerning the evolutionary origins of PxSR. To our knowledge, SRCR, PTX and LCCL domains have to date only been found in metazoans (Müller, 2001). The identification of PxSR may thus redefine the basal lineage of these protein domain families. Our phylogenetic results concerning the SRCR domain support this view (Fig. 3B). Interestingly, PxSR is a modular protein characterized by a number of domain duplication (SRCR and LCCL). The phylogenetic relationship of these duplications suggests PxSR has evolved in a step-wise manner (unpublished results). However, we found no significant PxSR sequence homologies outside Apicomplexa. Searches of other protist genome projects including, Giardia intestinalis, Entamoeba histolytica, Leishmania major, Trypanosoma spp. and the recently completed genome of the microsporidian Encephalitozoon cuniculi (Katinka et al., 2001) failed to detect PxSR homologues (data not shown). Although there is debate concerning the phylogeny of these protist lineages, Giardia and Entamoeba are thought to predate the evolution of the Apicomplexa (Baldauf and Doolittle, 1997). The absence of related PxSR sequences from other protist lineages foster speculation that PxSR may have arisen through one or more lateral gene transfer events, the sequences of which may have combined and recombined in some ancestral lineage of Apicomplexa. Alternately, the absence of related sequences may reflect a lack of current sequencing data. At best, the ontogeny of this gene remains uncertain. Why do PbSR ko oocysts not support development of viable sporozoites? PxSR encodes a complex multidomain protein and its disruption may have pleiotropic molecular consequences during oocyst development, as has been postulated as an explanation for the similar phenotype observed in CS-disrupted parasites, where sporozoite development in the oocyst is also impaired (Menard et al., 1997). There are, however, notable differences between the CS- and PbSR-disrupted phenotypes. CS-disrupted parasites produced some sporozoites and their oocysts were of comparable size to wild-type oocysts (Menard et al., 1997). These apparently distinct phenotypes suggest that the mechanisms that cause the reduction of sporozoite formation may be different between the CS- and PbSR-disrupted parasites. Recognising the structural features of PxSR, we should also consider that the protein could play a protective role against mosquito immune factors that target the parasite and adversely affect sporozoite formation. The oocyst and the sporozoite are extracellular parasite life stages that are exposed to mosquito immune responses for substantial periods of time. Plasmodium infections are known to elicit the up-regulation of mosquito genes that have putative immune functions. Nitric oxide synthase, antimicrobial peptide and serine protease genes such as ISP13 and © 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
ISPL5 are upregulated in response to bacterial and Plasmodium infections (Dimopoulos et al., 1997; 1998). Sp22D, the orthologue of Drosophila GRAAL, is expressed by haemocytes (the insect equivalent of the macrophage) and is also upregulated by bacterial challenge (Danielli et al., 2000). The modular structure of Sp22D has been proposed to be consistent with an immune recognition or ‘immune sensing’ function analogous to that of the horseshoe crab clotting factor C (Gorman and Paskewitz, 2001). An in vitro culture system of P. berghei oocysts and sporozoites has very recently been reported (Al-Olayan et al., 2002). This system may be useful to study the in vitro development of PbSR ko oocysts in the absence of mosquito immune factors and could shed light on a potential role of PbSR in mosquito immune evasion. Very recently it was reported that CS is directly involved in sporozoite budding and thus has multiple functions that include both morphogenesis and infectivity of the sporozoite (Thathy et al., 2002). A similar scenario could apply to PxSR. Indeed, the presence of PbSR in day 23 salivary gland sporozoites and the high level of PbSR transcript observed in day 15 oocysts that typically contain sporozoites indicate that PbSR synthesis continues after sporozoite formation. As such, the protein is likely to be carried to and released in liver cells by infecting sporozoites as has been reported for CS and TRAP. It is thus conceivable that the action of PbSR, besides allowing sporozoite morphogenesis, is important for sporozoite infectivity in the insect or vertebrate host. One hypothesis is that PxSR acts as an immunomodulatory factor that could scavenge or compete with the ligands of its molecular homologues such as CD163, CD5, CD6, CRP and SAP in the vertebrate, or equivalent molecules such as Sp22D in the mosquito. It is noteworthy that one of the other P. berghei sequences that contain an LCCL domain, CCP2, has recently been reported in GenBank (accession number AF491294). This gene also encodes a multidomain adhesive protein, which in addition to a single LCCL domain contains ricin, discoidin and fibrillar-collagen protein domains not previously reported in protist lineages. We have identified a number of homologous sequences from various apicomlexan genome sequencing projects including P. falciparum (data not shown). In P. falciparum the orthologous CCP2 sequence is situated on the same chromosome as PfSR (chr14). A putative gene duplication of CCP2 is situated approximately 0.8 Mb further along the same chromosome. Homology searches identify the C. parvum protein Cpa135 (GenBank accession number AJ006593.2), previously known as SA35 (Tosini et al., 1999), and a cDNA from T. gondii (GenBank accession number BM131310), as the putative orthologues of Plasmodium CCP2. Tosini et al. (1999) show Cpa135 is also expressed in the late oocyst that contains sporozoites.
1482 C. Claudianos et al. Moreover, Cpa135 is reported to be a secreted protein, which is associated with the apical complex and micronemes. These observations parallel our PxSR data at a number of levels. It is possible the LCCL domains of PxSR and CCP2 may be important in the localization and secretion of these novel mosaic proteins. The absence of sporozoite development in our PbSR ko parasites precludes a direct comparison of infectivity with wild-type sporozoites in mosquito and mouse. Instead of a PbSR null mutant parasite like the one reported here, transgenic parasites encoding altered versions of PbSR (e.g. ‘domain knockouts’) are more likely to give rise to sporozoite formation and will be instrumental in elucidating the roles of PxSR and its individual protein domains in sporozoite development and infectivity. The construction of such parasites is underway. Very recently, related studies by a different laboratory reported expression of PfSR in P. falciparum gametocytes (Delrieu et al., 2002). This suggests that PfSR may have additional functions to those of PbSR in P. berghei. Clearly, these interesting differences warrant further investigation. Experimental procedures Parasite maintenance and purification, RNA extraction, reverse transcription-PCR, Southern blotting, mosquito infections and immunodetection were as described (Butcher et al., 1991; Dessens et al., 1999; 2001).
Polyclonal anti-peptide antibodies Two peptides from the protein sequence were chosen on the basis of sequence conservation and predicted antigenic profile, using the GCG program ANTIGENIC (Fig. 1). Synthesis of 15–25 mg of each peptide, coupling of 5 mg of each peptide to a carrier protein (KLH, BSA, OVA or THY) and immunization of two SPF rabbits using a standard protocol with the mix of the coupled peptides (four injections) over 3 months, was carried out by Oswel/ Eurogentec. Preimmune and final bleed antisera were used at a final concentration of 1:20. Identification of PxSR A lTriplEX2 cDNA library was constructed from total parasite RNA using the cDNA Library Construction Kit (Clontech), Superscript II (Life Sciences) and Gigapack III gold packaging (Stratagene), according to manufacturer's instructions. Escherichia coli strain XL1-Blue was used for all library screens and strain BM25.8 for plasmid conversion. The library was screened using a [a-32P]-dATP random primer labelled PfSR sequence. The 3¢ end of PbSR was amplified using rapid amplification of cDNA ends (SMART™RACE, Clontech) with the PbSR-specific primer (CCCAAATTTGAG CACCCTTGTCATCTTGTT). Cloned cDNAs were analysed by automated sequencing and primer walking.
Transgenic parasite construction A 606-bp sequence was PCR amplified using primers PbSR/ NotI (GCGGCCGCGAGAATCTATAACTGGGTCAG) and PbSR/BamHI (GGATCCAGATATGAACCCTTCATGTAACAT), digested with NotI and BamHI and ligated into a NotI/BamHIdigested pBS-DHFR (Dessens et al., 1999), to give pPbSRNB. A 481-bp fragment was PCR amplified using primers PbSR/KpnI (GGTACCTCTCCTATAAAATAATCAGTTGC) and PbSR/HindIII (AAGCTTAGAATCTATAACTGGGTCAG), digested with HindIII and KpnI and ligated into HindIII/KpnIdigested pPbSR-NB, to give the transfection plasmid pPbSR. Parasite transfection, pyrimethamine selection and dilution cloning were performed as described (Waters et al., 1997).
Computer analyses Pairwise alignments were obtained using the GCG program GAP with gap weight 3.0 and gap length penalty of 0.1 (Devereux et al., 1984). Multiple sequence alignments were created using the CLUSTAL W program with the Blosum62 matrix, gap open penalty of 10 and gap extension penalty of 0.5 and gap separation of 8 (Thompson et al., 1994). Alignment shown in Fig. 3A was built through iterative pairwise comparisons using the BioEdit program (Hall, 1999) with CLUSTAL W, alignment was anchored at residues corresponding to key secondary structural elements as interpreted from M2BP (PDB; 1BY2). Phylogenetic analyses were done using the PAUP* program (Swofford, 2000). Distance trees were created using the neighbour-joining method with standard distances and mean character differences. Parsimony trees were created using simple heuristic search method with fast stepwise addition. Confidence values were obtained via random resampling with 1000 bootstrap replications.
Acknowledgements We wish to thank the scientists and funding agencies comprising the international Malaria Genome Project for making sequence data from the genome of P. falciparum (3D7) public prior to publication of the completed sequence. The Sanger Centre (UK) provided sequence for chromosomes 1, 3–9, and 13, with financial support from the Wellcome Trust. A consortium composed of The Institute for Genome Research, along with the Naval Medical Research Center (USA), sequenced chromosomes 2, 10, 11 and 14, with support from NIAID/NIH, the Burroughs Wellcome Fund, and the Department of Defense. The Stanford Genome Technology Center (USA) sequenced chromosome 12, with support from the Burroughs Wellcome Fund. The Plasmodium Genome Database is a collaborative effort of investigators at the University of Pennsylvania (USA) and Monash University (Melbourne, Australia), supported by the Burroughs Wellcome Fund. Preliminary sequence and/or preliminary annotated sequence data from the Plasmodium yoelii genome was obtained from The Institute for Genomic Research website (http://www.tigr.org). This sequencing program is carried on in collaboration with the Naval Medical Research Center and is supported by the U.S. Department of Defense. Partial shotgun sequencing data of Plasmodium knowlesi was obtained from the Sanger Institute, Pathogen © 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
PbSR knockout arrests development of malaria in mosquito 1483 Sequencing Unit, funded by the Wellcome Trust. Cryptosporidium parvum sequence data was obtained through the respective Genome Sequencing Project at the Virginia Commonwealth University/Tufts University School of Veterinary Medicine. We thank Dr P. Foster for helpful suggestions and E. Khater and C. McKeown for assistance with immunofluorescence and library construction, and Dr A. P. Waters for providing the Tubulin-1 primers. This work was supported by grants from the National Health and Medical Research Council of Australia (997038) (CC), the Wellcome Trust (J.T.D., T.C.), the European Union (H.E.T.) and the Japan Society for the Promotion of Science (M.A.). The sequence reported in this paper has been deposited in the GenBank database (accession no. AY034780).
References Al-Olayan, E.M., Beetsma, A.L., Butcher, G.A., Sinden, R.E., and Hurd, H. (2002) Complete development of mosquito phases of the malaria parasite in vitro. Science 295: 677– 679. Baldauf, S.L., and Doolittle, W.F. (1997) Origin and evolution of the slime molds (Mycetozoa). Proc Natl Acad Sci USA 94: 12007–12012. Bateman, A., and Sandford, R. (1999) The PLAT domain: a new piece in the PKD1 puzzle. Curr Biol 9: 588–590. Bateman, A., Birney, E., Durbin, R., Eddy, S.R., Howe, K.L., and Sonnhammer, E.L. (2000) The Pfam protein families database. Nucl Acids Res 28: 263–266. Bharadwaj, D., Mold, C., Markham, E., and Du Clos, T.W. (2001) Serum Amyloid P component binds to Fc (gamma) receptors and opsonizes particles for phagocytosis. J Immunol 166: 6735–6741. Bowen, M.A., Aruffo, A.A., and Bajorath, J. (2000) Cell surface receptors and their ligands: In vitro analysis of CD6–CD166 interactions. Proteins: Struct Funct Genet 40: 420–428. Butcher, G.A., Carr, R.E., and Fleck, S.L. (1991) The antimalarial activity of mouse tumour necrosis serum is blocked by purines. Ann Trop Med Parasitol 85: 271–273. del Carmen Rodriguez, M., Gerold, P., Dessens, J., Kurtenbach, K., Schwartz, R.T., Sinden, R.E., and Margos, G. (2000) Characterisation and expression of pbs25, a sexual and sporogonic stage specific protein of Plasmodium berghei. Mol Biochem Parasitol 110: 147–159. Catterall, C.F., Lyons, A., Sim, R.B., Day, A.J., and Harris, T.J. (1987) Characterization of primary amino acid sequence of human complement control protein factor I from an analysis of cDNA clones. Biochem J 242: 849–856. Dangott, L.J., Jordan, J.E., Bellet, R.A., and Garbers, D.L. (1989) Cloning of the mRNA for the protein that crosslinks to the egg peptide speract. Proc Natl Acad Sci USA 86: 2128–2132. Danielli, A., Loukeris, T.G., Lagueux, M., Muller, H.M., Richman, A., and Kafatos, F.C. (2000) A modular chitinbinding protease associated with hemocytes and hemolymph in the mosquito Anopheles gambiae. Proc Natl Acad Sci USA 97: 7136–7141. Delrieu, I., Waller, C.C., Mota, M.M., Grainger, M., Langhorne, J., and Holder, A.A. (2002) PSLAP, a protein with multiple adhesive motifs, is expressed in Plasmodium © 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
falciparum gametocytes. Mol Biochem Parasitol 121: 11– 20. Dessens, J.T., Beetsma, A.L., Dimopoulos, G., Wengelnik, K., Crisanti, A., Kafatos, F.C., and Sinden, R.E. (1999) CTRP is essential for mosquito infection by malaria ookinete. EMBO J 18: 6221–6227. Dessens, J.T., Mendoza, J., Claudianos, C., Vinetz, J.M., Khater, E., Hassard, S., et al. (2001) Knockout of the rodent malaria parasite chitinase pbcht1 reduces infectivity to mosquitoes. Infect Immun 69: 4041–4047. Devereux, J., Haeberli, P., and Smithies, O. (1984) A comprehensive set of sequence analysis programs for the vax. Nucl Acids Res 12: 387–395. van Dijk, M.R., Waters, A.P., and Janse, C.J. (1995) Stable transfection of malaria parasite blood stages. Science 268: 1358–1362. Dimopoulos, G., Richman, A., Muller, H.M., and Kafatos, F.C. (1997) Molecular immune responses of the mosquito Anopheles gambiae to bacteria and malaria parasites. Proc Natl Acad Sci USA 94: 11508–11513. Dimopoulos, G., Seeley, D., Wolf, A., and Kafatos, F.C. (1998) Malaria infection of the mosquito Anopheles gambiae activates immune-responsive genes during critical transition stages of the parasite life cycle. EMBO J 17: 6115–6123. Droste, A., Sorg, C., and Hogger, P. (1999) Shedding of CD163, a novel regulatory mechanism for a member of the scavenger receptor cysteine-rich family. Biochem Biophys Res Commun 256: 110–113. Emsley, J., White, H.E., O’Hara, B.P., Oliva, G., Srinivasan, N., Tickle, I.J., et al. (1994) Structure of pentameric human serum amyloid P component. Nature 367: 338–345. Freeman, M.W. (1994) Macrophage scavenger receptors. Curr Opin Lipidol 5: 143–148. Gebe, J.A., Llewellyn, M., Hoggatt, H., and Aruffo, A. (2000) Molecular cloning, genomic organization and cell-binding characteristics of mouse Sp-alpha. Immunology 99: 78– 86. Gorman, M.J., and Paskewitz, S.M. (2001) Serine proteases as mediators of mosquito immune responses. Insect Biochem Mol Biol 31: 257–262. Hall, T.A. (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/ NT. Nucl Acids Symp Series 41, 95–98. Hohenester, E., Sasaki, T., and Timpl, R. (1999) Crystal structure of a scavenger receptor cysteine-rich domain sheds light on an ancient superfamily. Nat Struct Biol 6: 228–232. Holmskov, U., Mollenhauer, J., Madsen, J., Vitved, L., Gronlund, J., Tornoe, I., et al. (1999) Cloning of gp-340, a putative opsonin receptor for lung surfactant protein D. Proc Natl Acad Sci USA 96: 10794–10799. Katinka, M.D., Duprat, S., Cornillot, E., Metenier, G., Thomarat, F., Prensier, G., et al. (2001) Genome sequence and gene compaction of the eukaryote parasite Encephalitozoon cuniculi. Nature 414: 450–453. Keeling, P.J., and Palmer, J.D. (2001) Lateral transfer at the gene and subgenic levels in the evolution of eukaryotic enolase. Proc Natl Acad Sci USA 98: 10745–10750. Liepinsh, E., Trexler, M., Kaikkonen, A., Weigelt, J., Banyai, L., Patthy, L., and Otting, G. (2001) NMR structure of the
1484 C. Claudianos et al. LCCL domain and implications for DFNA9 deafness disorder. EMBO J 20: 5347–5353. Maizels, R.M., Gomez-Escobar, N., Gregory, W.F., Murray, J., and Zang, X. (2001) Immune evasion genes from filarial nematodes. Int J Parasitol 31: 889–898. Menard, R., Sultan, A.A., Cortes, C., Altszuler, R., van Dijk, M.R., Janse, C.J., et al. (1997) Circumsporozoite protein is required for development of malaria sporozoites in mosquitoes. Nature 385: 336–340. Miyazaki, T., Hirokami, Y., Matsuhashi, N., Takatsuka, H., and Naito, M. (1999) Increased susceptibility of thymocytes to apoptosis in mice lacking AIM, a novel murine macrophage-derived soluble factor belonging to the scavenger receptor cysteine-rich domain superfamily. J Exp Med 189: 413–422. Müller, W.E. (1997) Origin of metazoan adhesion molecules and adhesion receptors as described from mRNA analyses in the marine sponge Geodia cydonium: a review. Cell Tissue Res 289: 383–395. Müller, W.E. (2001) Review: How was metazoan threshold crossed? The hypothetical Urmetazoa. Comp Biochem Physiol A Mol Integr Physiol 129: 433–460. Ochman, H. (2001) Lateral and oblique gene transfer. Curr Opin Genet Dev 11: 616–619. Pancer, Z. (2000) Dynamic expression of multiple scavenger receptor cysteine-rich genes in coelomyctes of the purple sea urchin. Proc Natl Acad Sci USA 97: 13156–13161. Pancer, Z., Rast, J.P., and Davidson, E.H. (1999) Origins of immunity: transcription factors and homologues of effector genes of the vertebrate immune system expressed in the sea urchin coelomocytes. Immnuogenetics 49: 773–786. Paton, M.G., Barker, G.C., Matsuoka, H., Ramesar, J., Janse, C.J., Waters, A.P., and Sinden, R.E. (1993) Structure and expression of a post-transcriptionally regulated malaria gene encoding a surface protein from the sexual stages of Plasmodium berghei. Mol Biochem Parasitol 59: 263–275. Resnick, D., Pearson, A., and Krieger, M. (1994) The SRCR superfamily: a family reminiscent of the Ig superfamily. Trends Biochem Sci 19: 5–8. Riffkin, M., Seow, H.F., Jackson, D., Brown, L., and Wood, P. (1996) Defence against the immune barrage: helminth survival strategies. Immunol Cell Biol 74: 564–574.
Roussel, A., Yang, Y., Ferrato, F., Verger, R., Cambillau, C., and Lowe, M. (1998) Structure and activity of rat pancreatic lipase-related protein 2. J Biol Chem 273: 32121–32128. Shrive, A.K., Metcalfe, A.M., Cartwright, J.R., and Greenhough, T.J. (1999) C-reactive protein and SAP-like pentraxin are both present in Limulus polyphemus hemolymph: crystal structure of Limulus SAP. J Mol Biol 290: 997–1008. Sinden, R.E. (1981) Sexual development of malarial parasites in their mosquito vectors. Trans Royal Soc Trop Med Hyg 75: 171–172. Sinden, R.E. (1999) Plasmodium: differentiation in the mosquito Parassitologia 41: 139–148. Sultan, A.A., Thathy, V., Frevert, U., Robson, K.J.H., Crisanti, A., Nussenzweig, V., et al. (1997) TRAP is necessary for gliding motility and infectivity of Plasmodium sporozoites. Cell 90: 511–522. Swofford, D.L. (2000) PAUP* Phylogenetic analysis using parsimiony (*and other methods). Version 4. Sinauer Associates: Sunderland, Massachusetts. Thathy, V., Fujioka, H., Gantt, S., Nussenzweig, R., Nussenzweig, V., and Menard, R. (2002) Levels of circumsporozoite protein in the Plasmodium oocyst determine sporozoite morphology. EMBO J 21: 1586–1596. Thompson, J.D., Higgins, D.G., and Gibson, T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucl Acids Res 22: 4673–4680. van Tilbeurgh, H., Bezzine, S., Cambillau, C., Verger, R., and Carriere, F. (1999) Colipase: structure and interaction with pancreatic lipase. Biochim Biophys Acta 1441: 173– 184. Tosini, F., Caccio, S., Tamburrini, A., La Rosa, G., and Pozio, E. (1999) Identification and characterisation of three antigenic proteins from Cryptosporidium parvum sporozoites using a DNA library expressing poly- histidine tagged peptides. Int J Parasitol 29: 1925–1933. Trexler, M., Banyai, L., and Patthy, L. (2000) The LCCL module. Eur J Biochem 267: 5751–5757. Waters, A.P., Thomas, A.W., Vandijk, M.R., and Janse, C.J. (1997) Transfection of malaria parasites. Methods 13: 134– 147.
© 2002 Blackwell Science Ltd, Molecular Microbiology, 45, 1473–1484
