Cross-Reactive Surface Epitopes on Chondroitin Sulfate A-Adherent ...

INFECTION AND IMMUNITY, May 2005, p. 2848–2856 0019-9567/05/$08.00⫹0 doi:10.1128/IAI.73.5.2848–2856.2005 Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Vol. 73, No. 5

Cross-Reactive Surface Epitopes on Chondroitin Sulfate A-Adherent Plasmodium falciparum-Infected Erythrocytes Are Associated with Transcription of var2csa Salenna R. Elliott,* Michael F. Duffy, Timothy J. Byrne, James G. Beeson,† Emily J. Mann, Danny W. Wilson,† Stephen J. Rogerson, and Graham V. Brown Department of Medicine, University of Melbourne, Royal Melbourne Hospital, Victoria, Australia Received 18 November 2004/Returned for modification 30 December 2004/Accepted 7 January 2005

Malaria in pregnancy is associated with placental accumulation of Plasmodium falciparum-infected erythrocytes (IE) that adhere to chondroitin sulfate A (CSA). Adhesion is mediated by P. falciparum erythrocyte membrane protein 1 (PfEMP1), a variant parasite protein expressed on the surface of IE and encoded by var genes. Rabbit antiserum was generated against the CSA-adherent P. falciparum line CS2, in which the dominant var transcribed is var2csa, a relatively conserved var gene that has been associated with CSA adhesion. Anti-CS2 recognized genetically distinct CSA-adherent P. falciparum lines but not CD36-adherent parent lines. Reactivity with anti-CS2 correlated with the level of adhesion to CSA. Fluorescence-activated cell sorting according to binding of anti-CS2 showed reactivity was associated with CSA adhesion and transcription of var2csa. These data are consistent with the hypothesis that var2csa encodes a PfEMP1 expressed on the surface of IE, which mediates adhesion to CSA and is relatively conserved between genetically distinct strains of P. falciparum.

pressed by placental IE is probably restricted. Sera from malaria-exposed multigravid women can inhibit adhesion of placental IE to CSA even when IE and sera are obtained from geographically distinct locations, suggesting conservation of CSA binding domains (28). Several var genes purported to encode CSA-binding domains have been described, including varCS2 (43) and FCR3varCSA (12, 50). Binding to CSA has been localized to DBL␥ domains of these genes (12, 18, 32, 38, 43, 44). However, recent studies of FCR3varCSA failed to show a correlation between transcription and the CSA adhesion phenotype or specific transcription in placental-type parasites (21, 27, 35, 47). FCR3varCSA (12, 50) and varCS2 (43) were originally identified in CSA-adherent parasite lines using reverse transcription PCR (RT-PCR). Since single trophozoites can transcribe multiple var genes simultaneously (21), this approach might detect genes that do not encode the expressed PfEMP1 protein. A recent study using quantitative RT-PCR showed that selection of laboratory lines for CSA adhesion was associated with increased transcription of the var gene var2csa (49). The var2csa gene was transcribed at higher levels in placental isolates than in isolates from children. Unlike many var genes, var2csa shows a high level of homology in genetically distinct isolates (33, 49). Recent mass spectrometric analysis failed to identify the var2csa protein in IE membrane extracts of CSAselected or placental isolates (29). However, antisera generated against domains of 3D7 var2csa recognize CSA-selected IE, suggesting that the PfEMP1 encoded by var2csa is expressed on the infected erythrocyte surface (48). In this study, we examined genetically distinct CSA-adherent P. falciparum lines, looking for an association between surface antigenic phenotype and var2csa transcription. Using rabbit antiserum generated against trophozoites of the CSA-adherent

During pregnancy, women are more susceptible to the consequences of Plasmodium falciparum malaria infection. Malaria in pregnancy is associated with increased risk of maternal anemia and fetal intrauterine growth retardation (11, 54) and is characterized by accumulation of P. falciparum-infected erythrocytes (IE) in the placenta (59). Adhesion of IE to multiple host receptors is mediated by P. falciparum erythrocyte membrane protein 1 (PfEMP1) (2, 12, 15, 16, 43; for a review see reference 40), a variant parasite protein expressed on the IE surface and encoded by the var multigene family (3, 51, 55). Transcriptional switching between var genes is associated with changes in parasite antigenic and adhesion phenotypes (51, 55). IE from nonpregnant individuals commonly bind to the endothelial receptors CD36 and intercellular adhesion molecule 1 (ICAM-1) (5, 26, 42). Placental IE have a unique phenotype, adhering to chondroitin sulfate A (CSA) and hyaluronic acid (HA), which are expressed on syncytiotrophoblasts (1, 5, 8, 26). Partial immunity to pregnancy-associated malaria develops after several pregnancies. This is associated with acquisition of antibodies to variant parasite antigens expressed on the IE surface, which inhibit adhesion of placental IE to CSA (7, 28, 39, 45, 53). PfEMP1 is a potential vaccine candidate for pregnancy-associated malaria. There are approximately 60 var genes per parasite genome (30), and since var genes are highly polymorphic (55), the diversity of PfEMP1 molecules is likely to be large. However, the range of PfEMP1 molecules ex-

* Corresponding author. Mailing address: Department of Medicine, University of Melbourne, Royal Melbourne Hospital, Victoria 3050 Australia. Phone: 613 8344 3267. Fax: 613 9347 1863. E-mail: salenna@ unimelb.edu.au. † Present address: The Walter and Eliza Hall Institute of Medical Research, Melbourne, Victoria, Australia. 2848

SURFACE ANTIGENS AND var2csa IN MALARIA

VOL. 73, 2005

FIG. 1. Summary of P. falciparum lines used in this study. Three genetically distinct laboratory lines, FAF-EA8 (derived by selection of ItG on human umbilical vein endothelial cells [HUVEC]), 3D7, and HM (from a traveler infected in Southeast Asia), were used. (A) FAFEA8 was selected for adhesion to CSA to generate CS2, selected on HA to generate E8B-HA, and selected on ICAM-1 to generate E8BICAM. Rabbit antiserum was generated against the CSA-adherent line CS2 (boldface box). (B) 3D7 was selected on CSA to generate 3C, which in turn was selected on ICAM-1 to produce 3CI and then reselected on CSA to derive 3CIC. Direct selection of 3D7 on ICAM-1 produced 3D7-ICAM. (C) The patient isolate HM was selected on CSA to generate HCS3. The adhesion phenotype of each line is indicated in italics.

line CS2, in which the dominant var transcript is var2csa (22), we linked the presence of conserved or cross-reactive epitopes on the surface of CSA-adherent IE with transcription of var2csa. MATERIALS AND METHODS Parasite culture. Parasites were cultured as previously described (4, 57). Parasite lines were regularly tested for Mycoplasma contamination and genotyped (MSP1 and MSP2 alleles). Gelatin enrichment of knob-expressing parasites (31) and sorbitol synchronization (37) were performed every 1 to 2 weeks. Parasite lines. Three genetically different P. falciparum lines were used (Fig. 1): ItG (shown to be identical to FCR3 [21, 46, 58]), 3D7, and HM (isolated from a traveler infected in Southeast Asia). FAF-EA8 (E8B) (Fig. 1A) was from a clone of ItG2F6 (FAF6) (9, 14) and bound predominantly to CD36. CS2 was obtained by repeated selection of FAF-EA8 on Chinese hamster ovary cells and purified CSA (17) (Fig. 1A). CS2 also binds HA (8). E8B-HA was derived from FAF-EA8 by selecting for adhesion to HA and also binds CSA (8) (Fig. 1A). E8B-ICAM was selected from FAF-EA8 by panning on ICAM-1 (Fig. 1A). 3D7

2849

lines were selected for adhesion to CSA and ICAM-1 as described previously (41) (Fig. 1B). 3D7 lines selected for adhesion to CSA do not bind HA (4). 3D7-ICAM was obtained by repeated selection of 3D7 on purified ICAM-1 (Fig. 1B). HM was selected on purified CSA to generate HCS3 (6) (Fig. 1C), which binds HA as well (4). Characterization of adhesion phenotype. Adhesion assays were performed as described previously (5). Receptors were diluted in phosphate-buffered saline (PBS) and coated in triplicate spots on the bases of 25-ml plastic petri dishes (Falcon 1058; Becton Dickinson) by overnight incubation at 4°C (CD36 purified from platelets [a gift from M. Berndt, Baker Institute, Australia]), soluble ICAM-1 [from transfected Chinese hamster ovary cells], CSA [from bovine trachea {Sigma}; 10 ␮g/ml), and HA [from bovine vitreous humor {Sigma}; 100 ␮g/ml). After blocking for 30 min with PBS/1% bovine serum albumin, plates were washed twice in RPMI-HEPES, and trophozoites (3 to 12% parasitemia) were resuspended at 7% hematocrit in adhesion medium (RPMI-HEPES supplemented with 50 ␮g/ml hypoxanthine, 2.5 ␮g/ml gentamicin, and 10% pooled human serum). Thirty-five microliters of this suspension was added per receptor spot and incubated at 37°C for 45 min. Four to six washes with RPMI-HEPES were used to remove unbound cells. Bound cells were fixed by incubation for at least 2 h in 2% glutaraldehyde in PBS and then stained with Giemsa stain. The IE bound in eight fields on each receptor spot were counted, and the mean and standard error of the mean for these 24 counts are given below. Parasite selection for adhesion. Parasite selection for adhesion was performed as described previously (41). Receptors were diluted in PBS (CD36, ICAM-1, CSA, and HA as described above), filter sterilized, and added to a sterile 25-ml petri dish (Falcon 1058; Becton Dickinson) either as 10-␮l spots covering the base of the dish (CD36, ICAM-1) or as 50 ml of diluted receptor (CSA, HA). After overnight incubation, the plate was blocked for 30 min using culture medium containing 10% human serum and then washed twice with RPMIHEPES. Seventy to 100 ml of cultured parasites (5 to 10% mid-trophozoites) was Percoll purified, resuspended in 50 ml of sterile adhesion medium (as described above), and added to the receptor-coated dish. After 45 to 60 min of incubation at 37°C in 1% O2, 4% CO2, 95% N2, unbound cells were removed by four to six washes in RPMI-HEPES until no unbound IE were observed. Directly pipetting 10 ml of culture medium onto bound IE allowed the IE to be removed and transferred to a petri dish for culture with fresh red blood cells. After several selections to obtain the initial phenotype, repeated reselection was required for most parasite lines after 3 to 4 weeks in continuous culture to maintain optimum adhesion. Generation of anti-CS2 rabbit antiserum (R1945). A rabbit was immunized with synchronized whole CS2 trophozoites as described previously (43) (approved by the Animal Ethics Committee, The Walter and Eliza Hall Institute of Medical Research). Rabbit antiserum was heat inactivated (45 min at 56°C) and preabsorbed against normal human erythrocytes for 1 h at room temperature. Malaria-exposed human serum. Serum from malaria-exposed donors (from Malawi or Papua New Guinea) or serum from unexposed Australian controls was heat inactivated (45 min at 56°C). Ethical clearance was obtained from the Human Research Ethics Committee, Royal Melbourne Hospital, the College of Medicine Research Ethics Committee, Malawi, and the Medical Research Advisory Committee, Papua New Guinea. Characterization of antigenic phenotype by flow cytometry. Trophozoites were incubated with anti-CS2 rabbit antiserum or malaria-exposed human serum as described previously (7). Anti-CS2 (1/20) was detected with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit immunoglobulin (Ig) (1/20) (Silenus) or with goat anti-rabbit IgG (1/50) (Jackson ImmunoResearch) and FITCconjugated anti-goat IgG (1/20) (Caltag). Human serum was detected by sequential incubation with goat anti-human IgG (Caltag) (1/50) and FITCconjugated anti-goat IgG (1/20) (Caltag or Chemicon). Quadrants were positioned based on staining with serum from unimmunized rabbits or malaria-naive human serum performed in parallel for each experiment. The percentage of IE specifically recognized by anti-CS2 or human serum was calculated by subtracting the percentage of uninfected erythrocytes that stained positive. Cell sorting. E8B-HA IE were enriched by gelatin sedimentation. After two washes in RPMI-HEPES and three washes in sterile PBS/1% fetal calf serum, staining with anti-CS2 was performed as described previously, with adjustment for the greater volume of cells (6 ml of cells at 0.1% hematocrit). All reagents were filter sterilized before use, and the FITC-conjugated anti-rabbit Ig (Silenus) was dialyzed against PBS overnight to remove sodium azide. Staining with ethidium bromide was omitted. Parasitized erythrocytes were sorted at 4°C, collected into culture medium with 10% human serum, and returned to culture. The phenotypes of sorted populations were characterized after 2 weeks of continuous culture (to generate sufficient numbers of IE).

2850

ELLIOTT ET AL.

Agglutination assay. Agglutination assays were performed as described previously (13). A parasite culture (5 to 10% trophozoites and 3% hematocrit) was spun down and resuspended in an equivalent volume of PBS with ethidium bromide (10 ␮g/ml). IE were incubated with serum at a 1/10 dilution on a rotating wheel at room temperature for 45 min. After gentle resuspension, aliquots were transferred to slides and examined by fluorescence microscopy. Samples were considered positive if more than three agglutinates consisting of more than five IE were observed in the absence of agglutinates of uninfected erythrocytes. Adhesion inhibition. The ability of anti-CS2 to inhibit adhesion of CS2 trophozoites to CSA was determined as described previously (7). CS2 trophozoites were washed in RPMI-HEPES, resuspended at 0.5% hematocrit in RPMIHEPES with anti-CS2 or normal rabbit serum (1/10 dilution), and incubated for 30 min at room temperature. Thirty microliters of the parasite suspension was then added to CSA spots for 20 min at 37°C. Otherwise the assay was performed as described above for the adhesion assays. Nucleic acid isolation and manipulation. Standard phenol-chloroform extraction techniques were used to prepare CS2 genomic DNA (gDNA). RNA was extracted using Trizol reagent (Sigma) as previously described (52). Precipitated RNA was dissolved in either formamide (for Northern blots) or water (for reverse transcription reactions). Northern blots. Northern blotting was performed as previously described (34, 56) using 0.9% agarose gels and random prime [␣-32P]dATP-labeled probes. Northern blots probed with the var exon 2 sequence were washed at low stringency with 2⫻ SSC and 0.1% sodium dodecyl sulfate at 55°C (1⫻ SSC is 0.15 M NaCl plus 0.015 M sodium citrate). Northern blots probed with the specific var2csa probe were washed at a higher stringency with 0.5⫻ SSC and 0.1% sodium dodecyl sulfate at 60°C. Quantitative RT-PCR (Q-RT-PCR). Contaminating gDNA in RNA from each parasite line was removed by repeated digestion with DNase I (DNA-free; Ambion). Ten to 20 ␮g of DNase I-digested RNA from each parasite line was made to 0.3 mg/ml in a solution containing 0.8 mM each of dATP, dCTP, dGTP, and dTTP and 20 ␮g/ml random hexamers. The solution was incubated at 65°C for 5 min and then chilled on ice, made to 20 ␮l containing 50 mM Tris-Cl (pH 8.3), 75 mM KCl, 3 mM MgCl2, 10 mM dithiothreitol, and 40 U of Rnasin (Promega), and incubated first at 25°C for 10 min and then at 42°C for 2 min. Two hundred units of SuperScript II RNase H reverse transcriptase was then added, and the solution was incubated first at 42°C for 50 min and then at 70°C for 15 min. The cDNA was diluted and stored as single-use aliquots at ⫺80°C. Forward and reverse primers to amplify var2csa (c10) (49) were used at a concentration of 2 ␮M. 18S rRNA primers (10) were used at concentrations of 0.7 ␮M (forward) and 6 ␮M (reverse). Each 10-␮l PCR mixture contained 5 ␮l of SYBR Green PCR master mixture (PE Biosystems) and was amplified using the ABI Prism 7900HT sequence detection system. PCR mixtures were incubated at 95°C for 10 min and then subjected to 40 cycles of 95°C for 15 s and 60°C for 1 min. A final incubation of 95°C for 2 min was followed by a dissociation step at 60°C for 2 min with a 2% ramp rate to 95°C for 2 min. The specificity of each primer pair was determined by dissociation curve analysis. DNA from var2csa was cloned into the pGEM-T EASY plasmid (Promega), and the concentration of the recombinant plasmid DNA was accurately determined using a fluorescent DNA quantitation kit (Bio-Rad). The plasmid was thawed and diluted 10-fold five times to generate a six-point standard curve, and 2 ␮l of each dilution was used per Q-RT-PCR mixture, enabling quantitation of 1 ⫻ 106 to 10 cDNA copies. CS2 gDNA at 50 ␮g ml⫺1 was also stored as single-use aliquots at ⫺80°C and diluted 10-fold five times immediately prior to assaying to generate a six-point standard curve. Two microliters of each dilution was used per Q-RT-PCR mixture to generate a gDNA standard curve that spanned 50 ␮g ml⫺1 to 0.5 ng ml⫺1. This standard curve was used to quantitate the 18S rRNA levels that were used to normalize the var2csa data such that cDNA from equivalent numbers of cells were compared for all parasite lines. Absolute quantitation was performed by using the PE Biosystems instructions by interpolation of threshold cycle values for unknown samples onto standard curves of quantity plotted against threshold cycle.

RESULTS Specificity of anti-CS2 rabbit antiserum. When flow cytometry was used, the majority of CS2 IE were recognized by anti-CS2 (mean ⫾ standard deviation, 90.8% ⫾ 5.7% [n ⫽ 10 experiments]) (Fig. 2A). There was minimal binding of serum from an unimmunized rabbit (2.5% ⫾ 0.9% [n ⫽ 10]). Most IE

INFECT. IMMUN.

FIG. 2. Rabbit antiserum generated against CS2 trophozoites (anti-CS2) was screened against the CSA-adherent ItG lines CS2 (A) and E8B-HA (B) and the CD36-adherent parent line FAF-EA8 (C) by flow cytometry. Background staining of CS2 with normal rabbit serum (NRS) from an unimmunized rabbit is also shown (A). Binding of malaria-exposed human serum (HS) to FAF-EA8 IE (C) demonstrated the presence of variant surface antigens. The percentage of trophozoites bound by anti-CS2 is indicated in the upper right quadrant of each dot plot. The bar graph in each panel shows the adhesion phenotype for each line (mean and standard error of the mean). The percentage of parasitemia (trophozoites) used in each adhesion assay is shown on each graph. Representative results from individual experiments are shown.

of another ItG CSA-adherent line, E8B-HA (Fig. 1), also bound to anti-CS2 (83.3% ⫾ 5.1% [n ⫽ 5]) (Fig. 2B), whereas few IE from the CD36-adherent parent line, FAF-EA8, were recognized (6.5% ⫾ 3.4% [n ⫽ 8]) despite expressing variant surface antigens (demonstrated by the binding of malaria-exposed human serum) (Fig. 2C). To exclude nonspecific binding of rabbit IgG to PfEMP1,

VOL. 73, 2005


2851

FIG. 3. Anti-CS2 was screened against the 3D7-derived CSA-adherent lines 3C (A) and 3CIC (B) and the CD36-adherent 3D7 line (C) by flow cytometry. Binding of malaria-exposed human serum (HS) to 3D7 IE (C) indicated the presence of variant surface antigens. The percentage of trophozoites bound by anti-CS2 is indicated in the upper right quadrant of each dot plot. The bar graph in each panel shows the adhesion phenotype for each line (mean and standard error of the mean). The percentage of parasitemia (trophozoites) used in each adhesion assay is shown on each graph. Representative results from individual experiments are shown.

CS2 trophozoites were incubated with purified rabbit IgG (10 mg/ml) in place of serum. The level of binding of purified rabbit IgG to CS2 trophozoites was similar to the level of binding of serum from unimmunized rabbits (results not shown). Antiserum against the conserved cytoplasmic acidic terminal sequence domain of PfEMP1 showed levels of binding to CS2 trophozoites similar to those of normal rabbit serum, confirming that only surface antigens were being detected by anti-CS2 (results not shown). A three-step staining protocol using anti-rabbit IgG secondary antibody gave results similar to the results of the two-step protocol with anti-rabbit Ig described previously (results not shown). Anti-CS2 agglutinated CS2 trophozoites but not the CD36binding parent line, FAF-EA8, and inhibited adhesion of CS2 to CSA (results not shown). Binding of anti-CS2 to genetically different CSA-adherent parasite lines. Anti-CS2 bound to two CSA-adherent 3D7derived parasite lines, 3C and 3CIC (Fig. 1), with a high proportion of IE recognized (for 3C, 83.7% ⫾ 10.6% [n ⫽ 5]; for 3CIC, 64.5% ⫾ 10.2% [n ⫽ 7]) (Fig. 3A and B). Few 3D7 IE,

FIG. 4. Anti-CS2 was screened against the CSA-selected line HCS3 (A) and the Southeast Asian isolate from which it was derived, HM (B), by flow cytometry. Binding of malaria-exposed human serum (HS) to HM IE (B) confirmed the presence of variant surface antigens. The percentage of trophozoites bound by anti-CS2 is indicated in the upper right quadrant of each dot plot. The bar graphs in panels A and B show the adhesion phenotype for each line (mean and standard error of the mean). The percentage of trophozoites used in each adhesion assay is shown on each graph. Representative results from individual experiments are shown. (C) Northern blots of HM and HCS3, obtained using probes against the conserved var exon 2 (which should detect all var genes) and var2csa.

which adhered to CD36, were recognized by anti-CS2 (5.6% ⫾ 3.1% [n ⫽ 6]) (Fig. 3C). Expression of variant surface antigens by 3D7 was confirmed by showing binding of malaria-exposed human serum (Fig. 3C). A third line, HCS3, derived by selection of the Southeast Asian isolate HM on CSA (Fig. 1), also had a high proportion of IE that bound to anti-CS2 (91.2% ⫾ 5.6% [n ⫽ 4]) (Fig. 4A). Anti-CS2 did not recognize the majority of trophozoites from the parent line, HM, although there was low-level binding to a subpopulation (27.1% ⫾ 4.6% [n ⫽ 3]) (Fig. 4B). In

2852

ELLIOTT ET AL.

INFECT. IMMUN.

FIG. 5. E8B-HA IE with mixed adhesion phenotypes after several weeks in culture (A) were reselected on CD36 (B), CSA (C), or HA (D), and their recognition by anti-CS2 was determined. The percentage of trophozoites bound by anti-CS2 is indicated in the upper right quadrant of each dot plot. The bar graph in each panel shows the adhesion phenotype for each line (mean and standard error of the mean). The percentage of trophozoites used in adhesion assays is shown on each graph. Representative results from individual experiments are shown.

parallel assays, the mean fluorescence intensity for anti-CS2 binding to this subpopulation of HM IE was significantly lower than that for binding to HCS3 IE (75.3 ⫾ 21.4 for HM, compared with 383.1 ⫾ 100.3 for HCS3 [n ⫽ 3; P ⬍ 0.05]), suggesting that fewer epitopes on the surface of HM IE were recognized by anti-CS2. A subpopulation of HM (52.9% ⫾ 5.4% [n ⫽ 2]) showed high-level binding of malaria-exposed human serum (mean fluorescence intensity, 590 ⫾ 12.8) (Fig. 4B). In a recent study, the dominant var transcripts in CSAadherent lines derived from ItG and 3D7 hybridized with a var2csa-specific probe on Northern blots (22). In the present study, Northern blotting of HCS3 probed with a conserved var exon 2 sequence revealed two dominant var transcripts, which also hybridized to the var2csa probe (Fig. 4C). In contrast, the dominant var transcripts that hybridized to the conserved var exon 2 probe in the parent line, HM, did not hybridize to the var2csa probe. Binding of anti-CS2 was related to CSA adhesion. Most parasite lines selected for adhesion lose their phenotype in continuous culture and require reselection. Reduced adhesion to CSA in selected lines derived from ItG, 3D7, and HM was associated with a decrease in the percentage of IE recognized by anti-CS2, which increased if parasite lines were reselected on the appropriate receptor (results not shown). After a period of continuous culture, E8B-HA showed reduced adhesion to CSA and HA, increased adhesion to CD36, and decreased recognition by anti-CS2 compared with the phenotype immediately after reselection on HA. This E8B-HA was reselected on CD36, CSA, or HA (Fig. 5). Reselection on either CSA or HA was associated with restoration of adhesion to both CSA and HA and enrichment of IE recognized by anti-CS2. There was no significant change in anti-CS2 binding associated with reselection on CD36. To further establish that recognition by anti-CS2 was linked to CSA adhesion, fluorescence-activated cell sorting according to recognition by anti-CS2 was performed on the same E8B-HA IE with mixed adhesion phenotypes (Fig. 6A), and the sorted populations were characterized. The anti-CS2-positive population showed a dominant CSA and HA adherence

FIG. 6. E8B-HA IE with mixed phenotypes after several weeks in culture (A) were cell sorted according to recognition by anti-CS2. Anti-CS2-positive IE (B) and anti-CS2-negative IE (C) were grown for approximately 2 weeks, and the adhesion phenotypes were determined. The percentage of trophozoites bound by anti-CS2 is indicated in the upper right quadrant of each dot plot. Binding of malariaexposed human serum (HS) to anti-CS2 negative IE (C) demonstrated the presence of variant surface antigens. The bar graph in each panel shows the adhesion phenotype for each line (mean and standard error of the mean). The percentage of trophozoites used in each adhesion assay is shown on each graph. Representative results from individual experiments are shown.

VOL. 73, 2005


2853

scribed above), low levels of the var2csa transcript were detected by Northern blotting (Fig. 7A) and Q-RT-PCR (Fig. 7B). Following reselection on CSA or HA, which increased the proportions of IE recognized by anti-CS2, var2csa was the dominant var transcript that hybridized with the conserved var exon 2 probe on Northern blots (Fig. 7A), and increased transcription of var2csa was detected by Q-RT-PCR (Fig. 7B). In IE reselected on CD36, which did not enrich for IE recognized by anti-CS2, transcription of var2csa was minimal (Fig. 7B). Cell-sorted E8B-HA IE recognized by anti-CS2 showed increased transcription of var2csa by Q-RT-PCR. However, on Northern blots, in addition to var2csa, two other var transcripts were detected. Unlike var2csa, these var genes were not detected on Northern blots of the E8B-HA lines reselected on CSA or HA (Fig. 7A). var2csa transcription was decreased in anti-CS2-negative IE compared with the parent E8B-HA. Taken together, these observations suggested that the major antigenic determinant on the trophozoite surface recognized by anti-CS2 rabbit antiserum was associated with adhesion to CSA and might be the protein product of var2csa. Antiserum reactive with CSA-binding isolates may crossreact with surface antigens on IE having a different adhesion phenotype. Unexpectedly, two 3D7 lines selected for ICAM-1 adhesion, 3CI and 3D7-ICAM (Fig. 1), also showed high-level recognition by anti-CS2 (for 3CI, 64.8% ⫾ 14.1% [n ⫽ 5]; for 3D7-ICAM, 74.8% ⫾ 10.1% [n ⫽ 4]) without detectable adhesion to CSA (Fig. 8A and B). The proportion of IE recognized by anti-CS2 correlated with the level of adhesion to ICAM-1 (results not shown). The dominant var that hybridized with the var exon 2 probe in 3D7-ICAM and 3CI did not hybridize with the var2csa probe (Fig. 8C). A genetically distinct ICAM-1-adherent line, E8B-ICAM, was not recognized by anti-CS2 (13.4% ⫾ 5.1% [n ⫽ 2]) (Fig. 8D). DISCUSSION

FIG. 7. (A) Northern blot of ring-stage RNA from E8B-HA sorted lines (anti-CS2 pos, anti-CS2 neg) and selected lines (sel CSA, sel CD36, sel HA [see Fig. 5 and 6]) using a var exon 2 probe (which should detect all var genes) and a var2csa-specific probe. The dominant var2csa transcript in anti-CS2-positive, CSA-selected (sel CSA), and HA-selected (sel HA) lines is indicated (arrow). (B) Quantitative RTPCR using var2csa-specific primers with E8B-HA sorted and selected lines (means and standard deviations).

phenotype (Fig. 6B). Adhesion to CD36 was also detected. In contrast, IE that were not recognized by anti-CS2 adhered predominantly to CD36, with minimal CSA or HA adhesion (Fig. 6C). Binding of malaria-exposed human serum to antiCS2-negative IE confirmed that their lack of recognition by anti-CS2 was not related to an absence of variant surface antigens. Binding of anti-CS2 was associated with transcription of var2csa. var2csa is the dominant var transcript detected on Northern blots in the ItG lines CS2 and E8B-HA (22). After continuous culture and before reselection of E8B-HA (as de-

Cross-reactive epitopes were detected on the surface of erythrocytes infected with genetically distinct CSA-adherent P. falciparum by rabbit antiserum generated against the CSAadherent P. falciparum line CS2. Similar to placental isolates, CS2 is recognized by sera from pregnant women in a paritydependent manner and is rarely recognized by sera from adult men (7). Recent work has shown that var2csa is the dominant var gene transcribed in CS2 (22); therefore, this antiserum should detect the PfEMP1 product of var2csa if it is expressed on the IE surface. We and others have shown that infected erythrocytes of CSA-adherent P. falciparum lines derived from ItG, 3D7, and the patient isolate HM are recognized by serum antibodies from pregnant women in a parity-dependent manner (7, 48; Beeson et al., unpublished data). These observations suggest that infected erythrocytes of these CSA-selected lines express antigens similar to those expressed during pregnancy-associated malaria. In the current study, anti-CS2 recognized CSAadherent IE from all three P. falciparum lines, ItG, 3D7, and HM. In contrast, anti-CS2 did not bind at high levels to CD36adherent parent lines. We established that the parent lines expressed variant surface antigens that were not recognized by anti-CS2 by demonstrating binding of malaria-exposed human serum and adhesion to CD36. Binding of anti-CS2 was asso-

2854

ELLIOTT ET AL.

FIG. 8. Anti-CS2 was screened against the 3D7 lines selected for adhesion to ICAM-1, 3CI (A) and 3D7-ICAM (B). The percentage of trophozoites bound by anti-CS2 is indicated in the upper right quadrant of each dot plot. The bar graphs in panels A, B, and D show the adhesion phenotype for each line (mean and standard error of the mean). The percentage of trophozoites used in each adhesion assay is shown on each graph. Representative results from individual experiments are shown. (C) Northern blot of ring-stage RNA from 3D7ICAM, 3CI, and HCS3 (included as a positive control) using a var exon 2 probe (which should detect all var genes) and a var2csa-specific probe. (D) Anti-CS2 screened against the genetically distinct ICAM1-adherent line E8B-ICAM.

ciated with adhesion to CSA. Reselection of the ItG line E8B-HA on either CSA or HA confirmed that this process selected for IE recognized by anti-CS2. CSA-adherent IE have been shown to bind nonimmune human IgM (20) and IgG (25) by a mechanism that is not antigen

INFECT. IMMUN.

specific. Immunoglobulins in nonimmune rabbit serum may also bind P. falciparum IE (19). We excluded nonspecific binding of rabbit IgM or IgG as an explanation for cross-reactivity of anti-CS2 with different CSA-adherent P. falciparum lines by showing that serum from unimmunized rabbits and purified rabbit IgG gave staining only marginally greater than the staining in the absence of serum. Comparable levels of staining with anti-Ig and anti-IgG secondary antibodies established that binding of nonspecific IgM did not contribute significantly to reactivity. Similar levels of staining with antiserum against the cytoplasmic acidic terminal sequence domain of PfEMP1 confirmed that anti-CS2 was binding to surface antigens and not to cytoplasmic proteins. Binding of anti-CS2 was associated with transcription of var2csa in CSA-adherent lines. Recently, var2csa has been shown to be the dominant var transcript in CS2, E8B-HA, 3C, and 3CIC (22), and in the current study it was also found to be the dominant var gene transcribed in HCS3. All these lines were recognized by anti-CS2. This association was confirmed by cell sorting IE of the ItG line E8B-HA according to recognition by anti-CS2 and showing that var2csa was transcribed at high levels in anti-CS2-positive IE but not in anti-CS2-negative IE. Interestingly, anti-CS2-positive IE transcribed two var genes in addition to var2csa. Because sorted IE had to be cultured for several weeks before phenotyping, a subpopulation of anti-CS2-positive IE may have switched var gene expression. This could account for the anti-CS2-negative and CD36-adherent IE detected in the anti-CS2-positive population. Alternatively, PfEMP1 proteins encoded by these var genes might be antigenically cross-reactive with the var2csa PfEMP1, resulting in binding of anti-CS2. A level of cross-reactivity between surface antigens expressed by CSA-adherent and non-CSA-adherent IE was evident. The CD36- and ICAM-1-adherent line HM contained a subpopulation that was recognized by anti-CS2, but the level of antibody binding (indicated by mean fluorescence intensity) was lower than the level of binding to CSA-selected HCS3 parasites. Anti-CS2 also bound to IE of the 3D7 ICAM-1adherent lines 3CI and 3D7-ICAM but not to the genetically distinct line E8B-ICAM. Binding by anti-CS2 correlated with the level of adhesion to ICAM-1 in the 3D7 lines, suggesting that it was domains of PfEMP1 that were recognized. Finally, CD36-selected E8B-HA showed low-level reactivity with antiCS2 in the absence of adhesion to CSA or transcription of var2csa. We suggest that our findings are explained by crossreactivity between domains of PfEMP1 separate from the functional domains mediating parasite adhesion, although the possibility of binding to other variant proteins on the infected erythrocyte surface, such as rifins (24, 36), cannot be excluded. In summary, using an antiserum against a P. falciparum line in which the dominant var gene transcribed is var2csa, we detected cross-reactive epitopes on the surface of genetically distinct CSA-adherent malaria parasite lines, the presence of which was associated with increased transcription of var2csa. These findings are consistent with the recently published observations of Salanti et al. (48). Both studies suggest that the 3D7 and ItG lines selected for adhesion to CSA express the protein product of var2csa, which is antigenically cross-reactive between these lines. Additionally, we showed that a third CSAadherent line, HCS3, transcribes var2csa and expresses variant


VOL. 73, 2005

surface antigens that are cross-reactive with those of CSAadherent 3D7 and ItG lines. Our observation that cell sorting infected erythrocytes with anti-CS2 reactivity selected for CSA-adherent IE that transcribed var2csa confirmed the close link between transcription of var2csa and the antigenic surface phenotype associated with CSA adhesion. Given that women in malaria-endemic areas are susceptible to pregnancy-associated malaria, despite preexisting antimalarial immunity, a dedicated vaccine is required for this condition. Antibodies specific for variant surface antigens expressed by placental isolates are associated with protection against the consequences of pregnancy-associated malaria, namely, low birth weight and maternal anemia (53), and probably function by inhibiting parasite adhesion to CSA (23, 28). Ideally, a vaccine for pregnancy-associated malaria would induce antibodies specific for domains of surface proteins that mediate CSA adhesion. Together with recent observations (48), our data suggest that the PfEMP1 encoded by var2csa is expressed on the surface of CSA-adherent malaria-infected erythrocytes and is relatively conserved between genetically distinct P. falciparum lines, which is consistent with the high level of homology reported in the var2csa gene between isolates (33, 49). Recent data (48) indicate that antibodies specific for VAR2CSA are associated with a reduced risk of low birth weight. If VAR2CSA is the dominant protein responsible for CSA adhesion by placental parasites and is highly conserved, it might be a suitable vaccine candidate for malaria in pregnancy. Further studies using fresh placental isolates should be performed to establish the prevalence of VAR2CSA expression and its level of conservation in a natural setting. ACKNOWLEDGMENTS This work was supported by funding from the National Health and Medical Research Council of Australia (grant 215201). We thank Aphrodite Caragounis for technical assistance. REFERENCES 1. Achur, R. N., M. Valiyaveettil, A. Alkhalil, C. F. Ockenhouse, and D. C. Gowda. 2000. Characterization of proteoglycans of human placenta and identification of unique chondroitin sulfate proteoglycans of the intervillous spaces that mediate the adherence of Plasmodium falciparum-infected erythrocytes to the placenta. J. Biol. Chem. 275:40344–40356. 2. Baruch, D. I., J. A. Gormely, C. Ma, R. J. Howard, and B. L. Pasloske. 1996. Plasmodium falciparum erythrocyte membrane protein 1 is a parasitized erythrocyte receptor for adherence to CD36, thrombospondin, and intercellular adhesion molecule 1. Proc. Natl. Acad. Sci. USA 93:3497–3502. 3. Baruch, D. I., B. L. Pasloske, H. B. Singh, X. Bi, X. C. Ma, M. Feldman, T. F. Taraschi, and R. J. Howard. 1995. Cloning the P. falciparum gene encoding PfEMP1, a malarial variant antigen and adherence receptor on the surface of parasitized human erythrocytes. Cell 82:77–87. 4. Beeson, J. G., and G. V. Brown. 2004. Plasmodium falciparum-infected erythrocytes demonstrate dual specificity for adhesion to hyaluronic acid and chondroitin sulfate A and have distinct adhesive properties. J. Infect. Dis. 189:169–179. 5. Beeson, J. G., G. V. Brown, M. E. Molyneux, C. Mhango, F. Dzinjalamala, and S. J. Rogerson. 1999. Plasmodium falciparum isolates from infected pregnant women and children are associated with distinct adhesive and antigenic properties. J. Infect. Dis. 180:464–472. 6. Beeson, J. G., W. Chai, S. J. Rogerson, A. M. Lawson, and G. V. Brown. 1998. Inhibition of binding of malaria-infected erythrocytes by a tetradecasaccharide fraction from chondroitin sulfate A. Infect. Immun. 66:3397–3402. 7. Beeson, J. G., E. J. Mann, S. R. Elliott, V. M. Lema, E. Tadesse, M. E. Molyneux, G. V. Brown, and S. J. Rogerson. 2004. Antibodies to variant surface antigens of Plasmodium falciparum-infected erythrocytes and adhesion inhibitory antibodies are associated with placental malaria and have overlapping and distinct targets. J. Infect. Dis. 189:540–551. 8. Beeson, J. G., S. J. Rogerson, B. M. Cooke, J. C. Reeder, W. Chai, A. M. Lawson, M. E. Molyneux, and G. V. Brown. 2000. Adhesion of Plasmodium falciparum-infected erythrocytes to hyaluronic acid in placental malaria. Nat. Med. 6:86–90.

2855

9. Biggs, B. A., R. F. Anders, H. E. Dillon, K. M. Davern, M. Martin, C. Petersen, and G. V. Brown. 1992. Adherence of infected erythrocytes to venular endothelium selects for antigenic variants of Plasmodium falciparum. J. Immunol. 149:2047–2054. 10. Blair, P. L., A. Witney, J. D. Haynes, J. K. Moch, D. J. Carucci, and J. H. Adams. 2002. Transcripts of developmentally regulated Plasmodium falciparum genes quantified by real-time RT-PCR. Nucleic Acids Res. 30:2224– 2231. 11. Brabin, B. J. 1983. An analysis of malaria in pregnancy in Africa. Bull. W. H. O. 61:1005–1016. 12. Buffet, P. A., B. Gamain, C. Scheidig, D. Baruch, J. D. Smith, R. HernandezRivas, B. Pouvelle, S. Oishi, N. Fujii, T. Fusai, D. Parzy, L. H. Miller, J. Gysin, and A. Scherf. 1999. Plasmodium falciparum domain mediating adhesion to chondroitin sulfate A: a receptor for human placental infection. Proc. Natl. Acad. Sci. USA 96:12743–12748. 13. Bull, P. C., B. S. Lowe, M. Kortok, C. S. Molyneux, C. I. Newbold, and K. Marsh. 1998. Parasite antigens on the infected red cell surface are targets for naturally acquired immunity to malaria. Nat. Med. 4:358–360. 14. Chaiyaroj, S. C., R. L. Coppel, S. Novakovic, and G. V. Brown. 1994. Multiple ligands for cytoadherence can be present simultaneously on the surface of Plasmodium falciparum-infected erythrocytes. Proc. Natl. Acad. Sci. USA 91:10805–10808. 15. Chen, Q., A. Barragan, V. Fernandez, A. Sundstrom, M. Schlichtherle, A. Sahlen, J. Carlson, S. Datta, and M. Wahlgren. 1998. Identification of Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) as the rosetting ligand of the malaria parasite P. falciparum. J. Exp. Med. 187:15– 23. 16. Chen, Q., A. Heddini, A. Barragan, V. Fernandez, S. F. Pearce, and M. Wahlgren. 2000. The semiconserved head structure of Plasmodium falciparum erythrocyte membrane protein 1 mediates binding to multiple independent host receptors. J. Exp. Med. 192:1–10. 17. Cooke, B. M., S. J. Rogerson, G. V. Brown, and R. L. Coppel. 1996. Adhesion of malaria-infected red blood cells to chondroitin sulfate A under flow conditions. Blood 88:4040–4044. 18. Costa, F. T., T. Fusai, D. Parzy, Y. Sterkers, M. Torrentino, J. B. Douki, B. Traore, S. Petres, A. Scherf, and J. Gysin. 2003. Immunization with recombinant duffy binding-like-gamma3 induces pan-reactive and adhesion-blocking antibodies against placental chondroitin sulfate A-binding Plasmodium falciparum parasites. J. Infect. Dis. 188:153–164. 19. Crandall, I., N. Guthrie, and I. W. Sherman. 1996. Plasmodium falciparum: naturally occurring rabbit immunoglobulins recognize human band 3 peptide motifs and malaria-infected red cells. Exp. Parasitol. 82:45–53. 20. Creasey, A. M., T. Staalsoe, A. Raza, D. E. Arnot, and J. A. Rowe. 2003. Nonspecific immunoglobulin M binding and chondroitin sulfate A binding are linked phenotypes of Plasmodium falciparum isolates implicated in malaria during pregnancy. Infect. Immun. 71:4767–4771. 21. Duffy, M. F., G. V. Brown, W. Basuki, E. O. Krejany, R. Noviyanti, A. F. Cowman, and J. C. Reeder. 2002. Transcription of multiple var genes by individual, trophozoite-stage Plasmodium falciparum cells expressing a chondroitin sulphate A binding phenotype. Mol. Microbiol. 43:1285–1293. 22. Duffy, M. F., T. J. Byrne, S. R. Elliott, D. W. Wilson, S. J. Rogerson, J. G. Beeson, R. Noviyanti, and G. V. Brown. Broad analysis reveals a consistent pattern of var gene transcription in Plasmodium falciparum repeatedly selected for a defined adhesion phenotype. Mol. Microbiol., in press. 23. Duffy, P. E., and M. Fried. 2003. Antibodies that inhibit Plasmodium falciparum adhesion to chondroitin sulfate A are associated with increased birth weight and the gestational age of newborns. Infect. Immun. 71:6620–6623. 24. Fernandez, V., M. Hommel, Q. Chen, P. Hagblom, and M. Wahlgren. 1999. Small, clonally variant antigens expressed on the surface of the Plasmodium falciparum-infected erythrocyte are encoded by the rif gene family and are the target of human immune responses. J. Exp. Med. 190:1393–1404. 25. Flick, K., C. Scholander, Q. Chen, V. Fernandez, B. Pouvelle, J. Gysin, and M. Wahlgren. 2001. Role of nonimmune IgG bound to PfEMP1 in placental malaria. Science 293:2098–2100. 26. Fried, M., and P. E. Duffy. 1996. Adherence of Plasmodium falciparum to chondroitin sulfate A in the human placenta. Science 272:1502–1504. 27. Fried, M., and P. E. Duffy. 2002. Two DBLgamma subtypes are commonly expressed by placental isolates of Plasmodium falciparum. Mol. Biochem. Parasitol. 122:201–210. 28. Fried, M., F. Nosten, A. Brockman, B. J. Brabin, and P. E. Duffy. 1998. Maternal antibodies block malaria. Nature 395:851–852. 29. Fried, M., J. P. Wendler, T. K. Mutabingwa, and P. E. Duffy. 2004. Mass spectrometric analysis of Plasmodium falciparum erythrocyte membrane protein-1 variants expressed by placental malaria parasites. Proteomics 4:1086– 1093. 30. Gardner, M. J., N. Hall, E. Fung, O. White, M. Berriman, R. W. Hyman, J. M. Carlton, A. Pain, K. E. Nelson, S. Bowman, I. T. Paulsen, K. James, J. A. Eisen, K. Rutherford, S. L. Salzberg, A. Craig, S. Kyes, M. S. Chan, V. Nene, S. J. Shallom, B. Suh, J. Peterson, S. Angiuoli, M. Pertea, J. Allen, J. Selengut, D. Haft, M. W. Mather, A. B. Vaidya, D. M. Martin, A. H. Fairlamb, M. J. Fraunholz, D. S. Roos, S. A. Ralph, G. I. McFadden, L. M. Cummings, G. M. Subramanian, C. Mungall, J. C. Venter, D. J. Carucci,

2856

31. 32.

33. 34. 35.

36.

37. 38.

39. 40. 41.

42.

43.

44.

ELLIOTT ET AL.

S. L. Hoffman, C. Newbold, R. W. Davis, C. M. Fraser, and B. Barrell. 2002. Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 419:498–511. Goodyer, I. D., J. Johnson, R. Eisenthal, and D. J. Hayes. 1994. Purification of mature-stage Plasmodium falciparum by gelatine flotation. Ann. Trop. Med. Parasitol. 88:209–211. Khattab, A., J. Kun, P. Deloron, P. G. Kremsner, and M. Q. Klinkert. 2001. Variants of Plasmodium falciparum erythrocyte membrane protein 1 expressed by different placental parasites are closely related and adhere to chondroitin sulfate A. J. Infect. Dis. 183:1165–1169. Kraemer, S. M., and J. D. Smith. 2003. Evidence for the importance of genetic structuring to the structural and functional specialization of the Plasmodium falciparum var gene family. Mol. Microbiol. 50:1527–1538. Kyes, S., R. Pinches, and C. I. Newbold. 2000. A simple RNA analysis method shows var and rif multigene family expression patterns in Plasmodium falciparum. Mol. Biochem. Parasitol. 105:311–315. Kyes, S. A., Z. Christodoulou, A. Raza, P. Horrocks, R. Pinches, J. A. Rowe, and C. I. Newbold. 2003. A well-conserved Plasmodium falciparum var gene shows an unusual stage-specific transcript pattern. Mol. Microbiol. 48:1339– 1348. Kyes, S. A., J. A. Rowe, N. Kriek, and C. I. Newbold. 1999. Rifins: a second family of clonally variant proteins expressed on the surface of red cells infected with Plasmodium falciparum. Proc. Natl. Acad. Sci. USA 96:9333– 9338. Lambros, C., and J. P. Vanderberg. 1979. Synchronization of Plasmodium falciparum erythrocytic stages in culture. J. Parasitol. 65:418–420. Lekana Douki, J. B., B. Traore, F. T. Costa, T. Fusai, B. Pouvelle, Y. Sterkers, A. Scherf, and J. Gysin. 2002. Sequestration of Plasmodium falciparum-infected erythrocytes to chondroitin sulfate A, a receptor for maternal malaria: monoclonal antibodies against the native parasite ligand reveal pan-reactive epitopes in placental isolates. Blood 100:1478–1483. Maubert, B., N. Fievet, G. Tami, M. Cot, C. Boudin, and P. Deloron. 1999. Development of antibodies against chondroitin sulfate A-adherent Plasmodium falciparum in pregnant women. Infect. Immun. 67:5367–5371. Newbold, C. I., A. G. Craig, S. Kyes, A. R. Berendt, R. W. Snow, N. Peshu, and K. Marsh. 1997. PfEMP1, polymorphism and pathogenesis. Ann. Trop. Med. Parasitol. 91:551–557. Noviyanti, R., G. V. Brown, M. E. Wickham, M. F. Duffy, A. F. Cowman, and J. C. Reeder. 2001. Multiple var gene transcripts are expressed in Plasmodium falciparum infected erythrocytes selected for adhesion. Mol. Biochem. Parasitol. 114:227–237. Ockenhouse, C. F., M. Ho, N. N. Tandon, G. A. Van Seventer, S. Shaw, N. J. White, G. A. Jamieson, J. D. Chulay, and H. K. Webster. 1991. Molecular basis of sequestration in severe and uncomplicated Plasmodium falciparum malaria: differential adhesion of infected erythrocytes to CD36 and ICAM-1. J. Infect. Dis. 164:163–169. Reeder, J. C., A. F. Cowman, K. M. Davern, J. G. Beeson, J. K. Thompson, S. J. Rogerson, and G. V. Brown. 1999. The adhesion of Plasmodium falciparum-infected erythrocytes to chondroitin sulfate A is mediated by P. falciparum erythrocyte membrane protein 1. Proc. Natl. Acad. Sci. USA 96: 5198–5202. Reeder, J. C., A. N. Hodder, J. G. Beeson, and G. V. Brown. 2000. Identification of glycosaminoglycan binding domains in Plasmodium falciparum erythrocyte membrane protein 1 of a chondroitin sulfate A-adherent parasite. Infect. Immun. 68:3923–3926.

Editor: W. A. Petri, Jr.

INFECT. IMMUN. 45. Ricke, C. H., T. Staalsoe, K. Koram, B. D. Akanmori, E. M. Riley, T. G. Theander, and L. Hviid. 2000. Plasma antibodies from malaria-exposed pregnant women recognize variant surface antigens on Plasmodium falciparum-infected erythrocytes in a parity-dependent manner and block parasite adhesion to chondroitin sulfate A. J. Immunol. 165:3309–3316. 46. Robson, K. J., J. R. Hall, L. C. Davies, A. Crisanti, A. V. Hill, and T. E. Wellems. 1990. Polymorphism of the TRAP gene of Plasmodium falciparum. Proc. R. Soc. Lond. B Biol. Sci. 242:205–216. 47. Rowe, J. A., S. A. Kyes, S. J. Rogerson, H. A. Babiker, and A. Raza. 2002. Identification of a conserved Plasmodium falciparum var gene implicated in malaria in pregnancy. J. Infect. Dis. 185:1207–1211. 48. Salanti, A., M. Dahlback, L. Turner, M. A. Nielsen, L. Barfod, P. Magistrado, A. T. Jensen, T. Lavstsen, M. F. Ofori, K. Marsh, L. Hviid, and T. G. Theander. 2004. Evidence for the involvement of VAR2CSA in pregnancyassociated malaria. J. Exp. Med. 200:1197–1203. 49. Salanti, A., T. Staalsoe, T. Lavstsen, A. T. Jensen, M. P. Sowa, D. E. Arnot, L. Hviid, and T. G. Theander. 2003. Selective upregulation of a single distinctly structured var gene in chondroitin sulphate A-adhering Plasmodium falciparum involved in pregnancy-associated malaria. Mol. Microbiol. 49:179–191. 50. Scherf, A., R. Hernandez-Rivas, P. Buffet, E. Bottius, C. Benatar, B. Pouvelle, J. Gysin, and M. Lanzer. 1998. Antigenic variation in malaria: in situ switching, relaxed and mutually exclusive transcription of var genes during intra-erythrocytic development in Plasmodium falciparum. EMBO J. 17: 5418–5426. 51. Smith, J. D., C. E. Chitnis, A. G. Craig, D. J. Roberts, D. E. Hudson-Taylor, D. S. Peterson, R. Pinches, C. I. Newbold, and L. H. Miller. 1995. Switches in expression of Plasmodium falciparum var genes correlate with changes in antigenic and cytoadherent phenotypes of infected erythrocytes. Cell 82:101– 110. 52. Smith, J. D., S. Kyes, A. G. Craig, T. Fagan, D. Hudson-Taylor, L. H. Miller, D. I. Baruch, and C. I. Newbold. 1998. Analysis of adhesive domains from the A4VAR Plasmodium falciparum erythrocyte membrane protein-1 identifies a CD36 binding domain. Mol. Biochem. Parasitol. 97:133–148. 53. Staalsoe, T., C. E. Shulman, J. N. Bulmer, K. Kawuondo, K. Marsh, and L. Hviid. 2004. Variant surface antigen-specific IgG and protection against clinical consequences of pregnancy-associated Plasmodium falciparum malaria. Lancet 363:283–289. 54. Steketee, R. W., B. L. Nahlen, M. E. Parise, and C. Menendez. 2001. The burden of malaria in pregnancy in malaria-endemic areas. Am. J. Trop. Med. Hyg. 64:28–35. 55. Su, X. Z., V. M. Heatwole, S. P. Wertheimer, F. Guinet, J. A. Herrfeldt, D. S. Peterson, J. A. Ravetch, and T. E. Wellems. 1995. The large diverse gene family var encodes proteins involved in cytoadherence and antigenic variation of Plasmodium falciparum-infected erythrocytes. Cell 82:89–100. 56. Taylor, H. M., S. A. Kyes, D. Harris, N. Kriek, and C. I. Newbold. 2000. A study of var gene transcription in vitro using universal var gene primers. Mol. Biochem. Parasitol. 105:13–23. 57. Trager, W., and J. B. Jensen. 1976. Human malaria parasites in continuous culture. Science 193:673–675. 58. Triglia, T., S. J. Foote, D. J. Kemp, and A. F. Cowman. 1991. Amplification of the multidrug resistance gene pfmdr1 in Plasmodium falciparum has arisen as multiple independent events. Mol. Cell. Biol. 11:5244–5250. 59. Walter, P. R., Y. Garin, and P. Blot. 1982. Placental pathologic changes in malaria. A histologic and ultrastructural study. Am. J. Pathol. 109:330–342.