点击显示 收起
Unité de Biologie des Interactions Hte-Parasite, URA , Institut Pasteur, Paris
Seattle Biomedical Research Institute
Department of Pathobiology, University of Washington, Seattle
Laboratory of Molecular and Vector Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland
Malaria in pregnancy is a serious complication associated with parasitized erythrocyte (PE) sequestration in the placenta. Recent work suggests that var genes could play an important role in PE binding to chondroitin sulfate A (CSA), a primary placental adherence receptor. Here, we confirm that var2CSA is transcriptionally up-regulated in CSA-binding parasites and define CSA-binding domains in var2CSA. The identification of multiple binding domains in var2CSA strengthens the evidence for their involvement in malaria during pregnancy and may have applications for the development of a vaccine against malaria in pregnancy.
Plasmodium falciparum infection during pregnancy is associated with parasitized erythrocyte (PE) sequestration in the placenta and contributes to low birth weight and neonatal mortality [1]. Placental isolates are functionally distinct, because they do not bind to CD36 but instead bind to chondroitin sulfate A (CSA) [2] and hyaluronic acid [3]. During pregnancy, antibodies to the PE surface develop that are broadly strain transcendent and have been suggested to play a role in protective immunity [4]. These observations suggest it may be possible to vaccinate against malaria during pregnancy, if the parasite adhesion ligands can be defined.
P. falciparuminfected erythrocytes employ a variable family of erythrocyte surface adhesion ligands called "var" genes, which encode P. falciparum erythrocyte membrane protein1 (PfEMP-1) [5], to sequester in different microvasculature sites. PfEMP-1 proteins have multiple adhesion domains, called "Duffy bindinglike" (DBL) domains and cysteine-rich interdomain regions (CIDRs), which determine PE binding specificity [5]. Of importance to placental sequestration, CD36 and CSA are mutually exclusive PE adhesion phenotypes [2, 6] that are functionally incompatible in the same PfEMP-1 protein [7]. Mechanistically, this may have arisen because CD36 binding and nonbinding var genes localize to distinct chromosomal regions and are transcribed in opposite orientations [810]. This genetic organization may cause them to recombine separately and evolve different structures and functions under distinct selective pressures. Since placental isolates do not bind to CD36 [2, 3], the maternal placenta selects for nonCD36-binding PfEMP-1 proteins that ensure localization in placenta and not on vascular endothelium.
Within the group of PfEMP-1 proteins predicted not to bind to CD36, a small subset, including var1CSA, have DBL- domains that bind to CSA [7, 11, 12]. Another gene within this group, var2CSA, does not contain DBL- domains but was recently found to be up-regulated in PEs that bind to CSA [13]. var2CSA is considered to be an additional candidate for CSA binding and placental adhesion, because of its expression profile and lack of a typical CD36-binding domain. Of interest, var2CSA contains sequence homology to the minimal CSA-binding region from var1CSA [12] and is unusually conserved between parasite strains [10, 13], which are features that might be expected of a placental adhesion ligand. The aim of the present study was to determine whether var2CSA contains CSA-binding domains.
Materials and methods.
For var2CSA Northern blot analysis, total RNA was prepared from ring- and trophozoite-stage cultures 10 h and 24 h after invasion, respectively. RNA preparation, electrophoresis, membrane transfer (Hybond N+; Amersham), and hybridization with radiolabeled A4 var2CSA DBL3 probe (36604623 bp) or a probe based on the var semiconserved exon 2 (var T11.1 gene; GenBank accession no. U67959; 79309147 bp) generated using the Megaprime Labeling Kit (Amersham) were performed as described elsewhere [14]. Membranes were hybridized at high-stringency conditions overnight at 60°C and were washed twice with 0.2× saline sodium citrate and 0.1% SDS for 30 min at 60°C.
For the determination of surface expression of various domains in CHO-745 cells, constructs were amplified from genomic DNA by polymerase chain reaction and were cloned into the pSR5-12CA5 vector (Affymax Research Institute), as described elsewhere [12]. The following domains from 3D7 PFL0030c var2CSA (GenBank accession no. NP_701371) and from A4 var2CSA (GenBank accession no. AAQ73926) were used (amino acid boundaries of each clone are given in parentheses): 3D7 DBL1 (aa 57382), 3D7 DBL2 (aa 542853), 3D7 DBL3 (aa 12131523), 3D7 DBL3b (aa 12131571), 3D7 DBL4 (aa 15761883), 3D7 DBL5 (aa 20012272), 3D7 DBL6 (aa 23182589), A4 DBL1 (aa 52383), A4 DBL2 (aa 543858), A4 DBL3 (aa 12201541), and A4 DBL3b (aa 12201580). Chinese hamster ovary PgsA-745 (CHO-745) cells deficient in glycosaminoglycans (American Type Culture Collection) were transfected and selected by single-cell cloning by use of a fluorescence-activated cell sorter (Becton Dickinson), as described elsewhere [12].
Binding assays with bovine trachea CSA (Sigma) or shark cartilage chondroitin sulfate C (CSC) (Sigma or Seikagaku) linked to biotin ("biot-CSA" and "biot-CSC," respectively) were performed as described elsewhere [11, 12]. In brief, 50 g of biot-CSA or biot-CSC was incubated for 1 h with stably transfected CHO-745 clones grown on coverslips. Binding was visualized with Dynabeads (Dynal) coated with anti-biotin monoclonal antibody (Jackson Immunoresearch Labs). For inhibition assays, the cells were incubated for 1 h with 200 g/mL CSA or CSC (Sigma) before addition of biot-CSA.
Results and discussion.
To determine the role of var2CSA in CSA adhesion, Northern blot analysis was performed on a parasite line in which var1CSA had been genetically disrupted (C1 mutant) [15]. C1 parasites bind to CD36 [15] and predominantly express an 89-kb var product (figure 1). Upon CSA selection, the C1-CSA parasite line switches to the larger, var2CSA transcript (figure 1), confirming previous results showing that var2CSA is transcriptionally up-regulated in different parasite strains selected to bind to CSA [13].
3D7 var2CSA contains 3 DBL-Xtype and 3 DBL-type domains [13]. For determination of the CSA-binding domain(s) from var2CSA, recombinant proteins corresponding to the different 3D7 var2CSA individual domains were expressed on the surface of CHO-745 cells and tested for binding. 3D7 var2CSA DBL2-X and DBL6- bound to biot-CSA (table 1). The binding was specific to CSA, since the adhesion of biot-CSA was inhibited by soluble CSA but not by soluble CSC (table 1). We also tested the DBL2-X domain of A4 var2CSA. Like 3D7, it bound to biot-CSA (table 1). None of the domains we tested bound to biot-CSC (data not shown).
Interestingly, the 3D7 var2CSA DBL3-X domain tested negative for CSA binding, although this domain had the closest sequence similarity to the minimal CSA-binding region from the FCR3 var1CSA DBL3- domain [12]. This same domain still did not bind to biot-CSA when expressed as a slightly longer domain (3D7 DBL3b-X; table 1). Sequence comparison of var2CSA from the A4, MC, 3D7, and Dd2 strains [10] showed that DBL3-X domains were highly similar (80% homology) but contained a 12-residue deletion in the 3D7 DBL3-X domain (figure 2A). This modification eliminated, among other things, a cysteine residue (figure 2A and 2B) present in the var1CSA DBL3- minimal binding domain and might explain the inability of the 3D7 DBL3-X domain to bind to CSA.
To examine this hypothesis, the DBL3-X and DBL3b-X domains from A4 var2CSA were expressed on CHO-745 cells and tested for binding (table 1). The A4 DBL3-X and DBL3b-X domains bound specifically to biot-CSA (table 1) but not to biot-CSC (data not shown). Considered with the previous mapping of the minimal binding domain in var1CSA [12], this var2CSA binding comparison additionally highlights the potential importance of the C-terminal region in CSA-binding DBL domains. Proof that the C-terminal region defines a general CSA-binding pocket must await further studies to confirm the minimal domain(s) of DBL2-X, DBL3-X, and DBL6- responsible for the CSA-binding phenotype.
Since the initial report that maternal antibodies recognize placental isolates from geographically dispersed regions and block infected erythrocyte binding to CSA [4], there has been intense interest in defining the parasite adhesion ligands, since these may form the basis of a vaccine against malaria during pregnancy. A leading candidate has been the parasite variant surface antigens responsible for cytoadhesive activities, but questions have arisen as to whether these are too divergent between parasite strains to explain the cross-reactive antibody response. Unlike the majority of the var gene family, var2CSA sequences are unusually conserved between parasite strains [10, 13]. In the present study, we demonstrate that 2 slightly distinct var2CSA sequences, which are transcriptionally up-regulated in different CSA-adherent parasite strains, contain multiple CSA-binding domains. This is the first evidence, to our knowledge, that var2CSA sequences bind to CSA, and it raises the possibility that multivalency is important for infected erythrocyte sequestration in the placenta. In the present study and others [12], several different DBL domains, including DBL-, -, and -X types, have now been shown to specifically bind to CSA (figure 2B). Binding sequences are highly diverse, although a consensus sequence based on the minimal binding region from FCR3 var1CSA DBL3- and other CSA-binding domains (figure 2B) would be defined as follows (using the single-letter code, with "X" meaning any amino acid and conserved cysteine residues in bold font): (hydrophobic)XEWX(E/D)X(F/Y)(C1)X2RX6(aliphatic)X3C2(variable length with 1 or 3 cysteines)C4X3C5X2YX2(aromatic)(aliphatic)(variable length)(aromatic)X6/7(F/Y)X8. These consensus residues can also be detected in DBL- domains that do not bind to CSA [12], suggesting that they are necessary but not sufficient for binding. Evidence for this will require studies of the effects of mutations of critical residues on binding to CSA.
In conclusion, the var2CSA is a strain-transcendent member of the parasite variant antigen family, which is transcriptionally up-regulated in infected erythrocytes selected to bind to CSA. Identification of CSA-binding domains in var2CSA strengthens the evidence for their involvement in malaria during pregnancy and is important for future design(s) of a broad-spectrum vaccine against malaria during pregnancy.
Acknowledgments
References
1. Brabin BJ, Romagosa C, Abdelgalil S, et al. The sick placentathe role of malaria. Placenta 2004; 25:35978. First citation in article
2. Fried M, Duffy PE. Adherence of Plasmodium falciparum to chondroitin sulfate A in the human placenta. Science 1996; 272:15024. First citation in article
3. Beeson JG, Rogerson SJ, Cooke BM, et al. Adhesion of Plasmodium falciparum-infected erythrocytes to hyaluronic acid in placental malaria. Nat Med 2000; 6:8690. First citation in article
4. Fried M, Nosten F, Brockman A, Brabin BJ, Duffy PE. Maternal antibodies block malaria. Nature 1998; 395:8512. First citation in article
5. Miller LH, Baruch DI, Marsh K, Doumbo OK. The pathogenic basis of malaria. Nature 2002; 415:6739. First citation in article
6. Rogerson SJ, Chaiyaroj SC, Ng K, Reeder JC, Brown GV. Chondroitin sulfate A is a cell surface receptor for Plasmodium falciparum-infected erythrocytes. J Exp Med 1995; 182:1520. First citation in article
7. Gamain B, Gratepanche S, Miller LH, Baruch DI. Molecular basis for the dichotomy in Plasmodium falciparum adhesion to CD36 and chondroitin sulfate A. Proc Natl Acad Sci USA 2002; 99:100204. First citation in article
8. Robinson BA, Welch TL, Smith JD. Widespread functional specialization of Plasmodium falciparum erythrocyte membrane protein 1 family members to bind CD36 analysed across a parasite genome. Mol Microbiol 2003; 47:126578. First citation in article
9. Lavstsen T, Salanti A, Jensen AT, Arnot DE, Theander TG. Sub-grouping of Plasmodium falciparum 3D7 var genes based on sequence analysis of coding and non-coding regions. Malar J 2003; 2:27. First citation in article
10. Kraemer SM, Smith JD. Evidence for the importance of genetic structuring to the structural and functional specialization of the Plasmodium falciparum var gene family. Mol Microbiol 2003; 50:152738. First citation in article
11. Buffet PA, Gamain B, Scheidig C, et al. Plasmodium falciparum domain mediating adhesion to chondroitin sulfate A: a receptor for human placental infection. Proc Natl Acad Sci USA 1999; 96:127438. First citation in article
12. Gamain B, Smith JD, Avril M, et al. Identification of a 67-amino-acid region of the Plasmodium falciparum variant surface antigen that binds chondroitin sulphate A and elicits antibodies reactive with the surface of placental isolates. Mol Microbiol 2004; 53:44555. First citation in article
13. Salanti A, Staalsoe T, Lavstsen T, et al. Selective upregulation of a single distinctly structured var gene in chondroitin sulphate A-adhering Plasmodium falciparum involved in pregnancy-associated malaria. Mol Microbiol 2003; 49:17991. First citation in article
14. Kyes S, Pinches R, Newbold C. A simple RNA analysis method shows var and rif multigene family expression patterns in Plasmodium falciparum. Mol Biochem Parasitol 2000; 105:3115. First citation in article
15. Andrews KT, Pirrit LA, Przyborski JM, et al. Recovery of adhesion to chondroitin-4-sulphate in Plasmodium falciparum varCSA disruption mutants by antigenically similar PfEMP1 variants. Mol Microbiol 2003; 49:65569. First citation in article