key: cord-0003884-ss4b2ko9 authors: Prajapati, Surendra Kumar; Borlon, Céline; Rovira-Vallbona, Eduard; Gruszczyk, Jakub; Menant, Sebastien; Tham, Wai-Hong; Kattenberg, Johanna Helena; Villasis, Elizabeth; De Meulenaere, Katlijn; Gamboa, Dionicia; Vinetz, Joseph; Fujita, Ricardo; Xuan, Xa Nguyen; Urbano Ferreira, Marcelo; Niño, Carlos H.; Patarroyo, Manuel A.; Spanakos, Gregory; Kestens, Luc; Abbeele, Jan Van Den; Rosanas-Urgell, Anna title: Complement Receptor 1 availability on red blood cell surface modulates Plasmodium vivax invasion of human reticulocytes date: 2019-06-20 journal: Sci Rep DOI: 10.1038/s41598-019-45228-6 sha: c6b7bfae68c5c25203f76616c028668feea39e8e doc_id: 3884 cord_uid: ss4b2ko9 Plasmodium vivax parasites preferentially invade reticulocyte cells in a multistep process that is still poorly understood. In this study, we used ex vivo invasion assays and population genetic analyses to investigate the involvement of complement receptor 1 (CR1) in P. vivax invasion. First, we observed that P. vivax invasion of reticulocytes was consistently reduced when CR1 surface expression was reduced through enzymatic cleavage, in the presence of naturally low-CR1-expressing cells compared with high-CR1-expressing cells, and with the addition of soluble CR1, a known inhibitor of P. falciparum invasion. Immuno-precipitation experiments with P. vivax Reticulocyte Binding Proteins showed no evidence of complex formation. In addition, analysis of CR1 genetic data for worldwide human populations with different exposure to malaria parasites show significantly higher frequency of CR1 alleles associated with low receptor expression on the surface of RBCs and higher linkage disequilibrium in human populations exposed to P. vivax malaria compared with unexposed populations. These results are consistent with a positive selection of low-CR1-expressing alleles in vivax-endemic areas. Collectively, our findings demonstrate that CR1 availability on the surface of RBCs modulates P. vivax invasion. The identification of new molecular interactions is crucial to guiding the rational development of new therapeutic interventions against vivax malaria. cleavage of red blood cell receptors has been extensively used to investigate invasion pathways in P. falciparum parasites [28] [29] [30] . To investigate the profile of P. vivax invasion, high-CR1-expressing (H-CR1) reticulocyte-enriched RBCs (reRBCs) ( Figure S1 ) were treated with neuraminidase (removes sialic acid residues), trypsin (cleaves several unknown receptors in addition to CR1) and chymotrypsin (cleaves several unknown receptors in addition to CR1 and DARC). Efficient cleavage of CR1 was confirmed by flow cytometry following antibody labeling with anti-CR1 antibody (Fig. 1a) . The capacity of the treated cells for invasion was confirmed with the P. falciparum 3D7 strain (Fig. 1b) , which results are in agreement with previous findings [31] [32] [33] . P. vivax invasion was reduced by 18.9% (SD 28.74%, N = 5 isolates, signed rank p = 0.13) with neuraminidase treatment, 45.69% (SD 18.26%, N = 5 isolates, signed rank p = 0.043, range 33-78%) with trypsin treatment, and 96% (SD 7%, N = 5 isolates, signed rank p = 0.043) with chymotrypsin treatment (Fig. 1c , Table S1 ). The results demonstrate that P. vivax invasion is sensitive to treatment with chymotrypsin and trypsin. The latest reduced P. vivax invasion efficiency by half in the presence of an intact DARC, suggesting that one or more trypsin-sensitive receptors (e.g., CR1) are involved in the process of P. vivax invasion. On the other hand, although P. vivax invasion was not significantly affected by neuraminidase treatment, a potential interacting role of sialic acid receptors cannot be excluded and deserve further exploration in future assays. Expression of CR1 on the surface of RBCs varies among individuals and this variation is correlated with exon 22 and intron 27 SNPs in non-African populations 19, 22 . To test whether CR1 availability on the surface of reticulocytes affects the efficiency of P. vivax invasion, we characterized the CR1 exon 22 genotype and CR1 expression levels on the surface of reRBC samples (Fig. 2 , Table 1 ). We identified seven homozygote (HH) and six heterozygote (HL) reRBC samples for the exon 22 SNP (rs2274567) (Fig. 2a, www.nature.com/scientificreports www.nature.com/scientificreports/ invasion we compared the invasion rate between reRBC samples expressing high levels of CR1 (High: n = 3; mean MFI: 41.47 SD 10.55) and the sample expressing the lowest levels (Low: n = 1, MFI = 19.69). The level of CR1 expression on L-CR1 reRBC sample was significantly lower than H-CR1 reRBC samples (p = 0.029), while reticulocyte enrichment levels of the L-CR1 reRBC sample (HCR14: 25%) and the H-CR1 samples (HCR40/HCR43/ HCR47: 28%/36%/41%) were not significantly different (p = >0.14). In addition, all reRBC expressed similar levels of CD71 on the surface of reRBCs (HCR14/HCR40/HCR43/HCR47: CD71MFI = 3966/3379/3725/3366) and are Duffy a + b+. Collectively, these results suggest no difference in the expression levels of either Duffy receptor or CD71 between low and high CR1 compared reRBCs samples. In agreement with previous observations 15, 16 , P. falciparum invasion of neuraminidase-treated and -untreated low-CR1-expressing (L-CR1) reRBCs (MFI: 19.69) was 72% and 28% lower than that of H-CR1 reRBCs (mean MFI: 41.47, SD 10.55) (p < 0.05) (Fig. 2d ) demonstrating the validity of our experimental system. P. vivax invasion of a single L-CR1 reRBCs sample was performed using N = 6 isolates, and was 40.8% (SD 11.2%, signed rank p < 0.027) lower than that of H-CR1 reRBCs samples (Fig. 2d ). While these results are consistent with a role for CR1 in mediating P. vivax invasion, additional L-CR1 samples would be necessary to demonstrate whether the impaired invasion is attributable to CR1 levels. To further test the role of CR1 in P. vivax invasion, we performed invasion assays in the presence of soluble recombinant CR1 protein (sCR1). A previous study had demonstrated that in laboratory conditions, incubation of neuraminidase-treated RBCs with sCR1 (50 μg/ml) blocked P. falciparum invasion 15, 16 . In our case, incubation of neuraminidase-treated H-CR1 reRBCs (Table 1) with sCR1 reduced P. falciparum 3D7 invasion by 90%, confirming previous results (Fig. 3 ). In the case of P. vivax, incubation of H-CR1 reRBCs with sCR1 (50 μg/ml) significantly reduced invasion by 42.5% (SD 14.6%, N = 9 isolates, signed rank, p = 0.007) (Fig. 3) . These results indicate that sCR1 directly binds to an unrecognized P. vivax ligand and that this interaction reduces binding between P. vivax parasites and native CR1 on reticulocytes. The observed reduction in invasion capacity is comparable to that observed for L-CR1 reRBCs compared with H-CR1-reRBCs. www.nature.com/scientificreports www.nature.com/scientificreports/ Lack of binding between sCR1 and recombinant P. vivax PvRBP parasite ligands. PvRBPs are homologues of the PfRh family in P. falciparum 34 . It is known that PfRh4 binds to CR1 to mediate glycophorin independent parasite invasion 16 . As our previous results indicate that CR1 has a role in P. vivax invasion, we questioned if a member of the PvRBP family might be the corresponding P. vivax parasite ligand for CR1. To test this hypothesis, we performed immuno-precipitation experiments using sCR1, PfRH4 (88 kDa) and five recombinant PvRBP fragments, PvRBP1a, PvRBP1b, PvRBP2a, PvRBP2b, and PvRBP2c, which migrate at 118, 133, 114, 152 and 94 kDa respectively under reducing conditions (Fig. 4) . While anti-CR1 monoclonal HB8592 successfully immuno-precipitated sCR1, we saw no evidence of complex formation with any of the PvRBP fragments tested, www.nature.com/scientificreports www.nature.com/scientificreports/ suggesting that PvRBPs are not direct ligands for CR1. On the other hand, as expected, we observed that anti-CR1 monoclonal HB8592 successfully immuno-precipitated sCR1 in complex with PfRh4. Biased distribution of CR1 L allele frequency in areas with P. vivax transmission. We observed that reduced CR1 availability on reRBCs decreased the efficiency of P. vivax invasion, even in the presence of an intact DARC. We therefore hypothesized that CR1 genotypes associated with reduced CR1 expression on the surface of reticulocytes may confer protection against P. vivax infection, possibly reflecting positive selection on L alleles in areas exposed to or with a history of exposure to vivax malaria. Positive selection on one allele can be detected by an unusually fast increase in allele frequency in one population relative to another 35 and extended LD surrounding the allele of interest 36 . To explore our hypothesis, we investigated the frequency of CR1 alleles associated with the L-CR1 phenotype through a meta-analysis of peer-reviewed articles and publicly available data. We first confirmed a LD R 2 ≥ 0.8 for the CR1 exon 22 (rs2274567) and intron 27 (rs11118133) alleles in populations sampled within the 1000 Genomes Project (Table S2) 23 . Using data mining of publicly available databases we identified a SNP in intron 26 (rs11118131), which was in LD (R 2 ≥ 0.8) with exon 22 in non-African populations (Table 2 ). LD between exon 22 and intron 26 was further confirmed by genotyping samples from Vietnam, Thailand, Greece, Belgium, Peru, and Brazil. We included intron 26 in the analysis of CR1 allele frequencies to target additional genotyping data from populations in which exon 22 and intron 27 data were not available. We then analyzed the worldwide distribution of CR1 alleles for intron 27 (rs11118133), exon 22 (rs2274567), and intron 26 (rs11118131), hereafter referred to as L-CR1 alleles. In total, data mining and in-house genotyping retrieved L-CR1 allele data for 34,625 individuals from 177 locations across 61 countries (Fig. 5a, Table S3 ). We first compared mean L-CR1 allele frequency in malaria-free areas (defined as areas with no documented history of malaria 37, 38 ), namely, Orkney Island, UK; Xinjiang province, China; and Siberia and Vologda, Russia (L-CR1 = 0.17, range 0.13-0.22), with mean frequencies reported for the other populations in the analysis (Table S3) . We observed a significantly increased frequency of L-CR1 in areas with P. vivax exposure (L-CR1 ≥ 0.29, p < 0.048) but not in sub-Saharan populations (Fig. 5b) , where there is no stable transmission of P. vivax parasites and P. falciparum is the only parasite responsible for overall transmission of malaria disease ( > 99.5%). Further, this significant increase in L-CR1 frequencies ( Figure S2 ) overlapped with the reported transmission intensities of P. vivax in global populations 39 . A significant increase in L-CR1 allele frequencies was still observed when we compared the mean frequency for African populations (Fig. 5a ,b and Figure S2 ) (mean L-CR1 = 0.21, range 0.1-0.28) with that of populations exposed to P. vivax (alone or with P. falciparum) (L-CR1 ≥ 0.33, p < 0.040). These results indicate positive selection on L-CR1 alleles in vivax-endemic areas. LD decay at CR1 locus. Positive selection on CR1 exon 22 was further assessed through LD analysis using data from the 1000 Genomes Project Phase 3. We estimated the number of proxy SNPs (SNPs under LD with exon 22) and the physical distance under LD, which indicates non-random association of alleles at multiple loci. We observed a significantly higher number of SNPs with R 2 ≥ 0.8 in populations exposed to P. vivax malaria, i.e., populations in East Asia (n = 65-83), South Asia (n = 77-79), Europe (n = 56-74), and South and Central America (n = 48-56), compared with African populations not exposed to P. vivax malaria (n = 7-8, p < 0.001) (Fig. 6a ). In addition, LD decay (resulting from genetic recombination) around the exon 22 SNP was significantly faster in populations from Africa (~18 kb, −4,400 bp to 13,500 bp) than in those from vivax-exposed areas of America, Europe, and Asia (~130 kb, −75,000 bp to 54,000 bp) (p ≤ 0.001) (Fig. 6b) . Controls for population stratification is shown in Figure S3 and supplementary information. P. vivax reticulocyte invasion is an essential process characterized by a cascade of events involving specific host-parasite interactions that are poorly understood. Our results, from ex vivo invasion assays and population genetic analyses, indicate that CR1 availability on the reticulocyte surface modulates P. vivax invasion even in the presence of intact DARC. The difficulties associated with investigating P. vivax invasion mechanisms due to the inability to continuously culture the parasites in vitro and limitations of ex vivo invasion assays have been extensively discussed elsewhere 4, 40 . In brief, several limitations have to be taken into account when performing P. vivax invasion assays: (1) the low parasite densities (0.5%-0.01%) in most field/clinical isolates of P. vivax, while >0.1% parasite density is needed to ensure a successful invasion, (2) high proportion of rings (≥80%) are necessary for a successful invasion, (3) ex vivo adaptation is extremely variable between parasite isolates and this may be partially explained by inherent characteristics of the parasite 41 , (4) several samples produce a high proportion of gametocyte during the maturation process and are thus discarded, and (5) cryopreserved parasites show lower invasion efficiency than fresh isolates 42 . The limitations in the number of P. vivax isolates with the correct criteria to perform ex-vivo invasion assays and the number of samples that maturate successfully and present an optimal invasion rate (untreated control well), together with the need to reduce assay variability greatly limit the number of conditions that can be tested with one parasite isolate. Nevertheless, the recent increase in the number of publications using P. vivax ex vivo assays 43, 44 demonstrates the usefulness of this tool to investigate molecular mechanisms related to parasite invasion. In our experiments, we used cryopreserved P. vivax isolates and reticulocyte-enriched samples from hemochromatosis donors. In this regard, CD71 availability on reRBCs was confirmed by flow cytometry and by demonstrating the ability of P. vivax isolates to invade heterogeneous CD71 +ve populations (with variable levels of CD71) 44 . In addition, the invasion rates observed are comparable to rates previously reported for different reticulocyte sou rces 5, 40, 42, [45] [46] [47] . Finally, we observed consistently significant invasion inhibition in different experiments using both natural CR1 expression variations of the reRBCs and biochemical approaches to modify CR1 presence at the reRBC surface. In this regard, our findings of a decrease in P. vivax invasion in trypsin-treated cells contradicts previous www.nature.com/scientificreports www.nature.com/scientificreports/ observations 48, 49 , although a trypsin-sensitive in vitro binding between TfR1-PvRBP2b has been demonstrated to be critical for the recognition of reticulocytes during P. vivax invasion 43 . Discrepancies with previous studies can be at least partially explained by differences in the experimental conditions. In Barnwell et al. 48 the source and enriched proportion of used reticulocytes is unknown, the authors use a single P. vivax Belem strain maintained in squirrel monkey, invasion is monitored after 8-10 hours of incubation, while information on invasion rates is lacking. In comparison with Malleret et al., who used 0.5 mg of trypsin for the digestion treatment 49 , here we used a higher concentration that proved (by flow cytometer) to remove CR1 from the surface of reticulocytes. Overall, although demonstration of invasion inhibition by anti-CR1 antibody may have strengthened our findings further, our results provide strong support for the involvement of CR1 in the process of P. vivax invasion. This role is further supported by our observation of a significant increase in the frequency of L alleles and strong LD around the exon 22 SNP in populations where P. vivax has historically been the major malaria species compared with populations in areas without vivax exposure, such as the African continent 39 . CR1 alleles associated with a low CR1-expression phenotype may exert a beneficial effect in populations exposed to P. vivax by reducing the efficiency of invasion, and thus decreasing the risk of infection and disease. Allele frequencies and LD measures correlated well for populations with recent exposure to the P. vivax parasite (such as Mediterranean, Asian, American, and Pacific populations), while the effect on allele frequencies is faster diluted in populations with past exposure and migration movements (non-Mediterranean European population) 50 . Population differentiation and extended LD in the CR1 genomic region has already been reported for the Sardinian population, which had a long history of endemic malaria until shortly after World War II, indicating positive natural selection of CR1 in this population 51 . The authors proposed P. falciparum as the selective force on the CR1 locus. Indeed, many studies have reported an important role for CR1 in malaria infection and pathogenesis 52 , although investigation of the association between CR1 and P. falciparum susceptibility has depicted a Table 2 . Linkage disequilibrium between CR1 exon 22 (rs2274567) and intron 26 (rs11118131) in global populations. SNP data was retrieved from the 1000 Genomes Project database or in-house genotyping (*). Linkage disequilibrium was analyzed using the LDlink software tool and with CubeX software tool for in-house genotyping data. www.nature.com/scientificreports www.nature.com/scientificreports/ complex scenario with discrepant results 17, 19, 21, 27, 52 . Contradictory results may be explained by multiple factors, including differences in the genetic background of populations and pleiotropic effects of CR1 on different diseases [53] [54] [55] [56] [57] , the promiscuous use of the CR1 receptor by P. falciparum parasites during invasion 15, 16 and rosetting 17 , the complexity of factors influencing CR1 levels in African populations 58 , and (as results from this study indicate) selective pressure exerted by P. vivax malaria. Other studies that have previously demonstrated a role for P. vivax infections in shaping the human genome include studies of Duffy determinants in African individuals and South East Asian ovalocytosis in Papua New Guinea 7,59 . www.nature.com/scientificreports www.nature.com/scientificreports/ In addition to trypsin-sensitive P. vivax invasion, we observed a modest (albeit non-significant) reduction in invasion efficiency in neuraminidase-treated cells. A potential role of sialylated glycoproteins in P. vivax reticulocyte invasion has not been investigated but deserves further attention. Finally, we were unable to demonstrate a direct interaction between CR1 and five members of PvRBPs which were thought to play essential roles during reticulocyte invasion 12 We have shown for the first time that CR1 availability on the surface of reticulocytes modulates P. vivax invasion. We have presented evidence of a recent positive selection of L allele expression as a beneficial trait in populations exposed to P. vivax parasites. Future studies should aim to identify the parasite ligand to CR1 and the possible role of this interaction in Duffy-negative reticulocyte invasion. The identification of new molecular interactions and alternative invasion pathways is crucial for guiding the rational development of therapeutic interventions, i.e., vaccines and drugs, targeting the reduction and prevention of reticulocyte invasion. Ethics statement. Ethical P. vivax isolates. P. vivax isolates were collected from patients with acute P. vivax infection attending clinics in the Shoklo Malaria Research Unit (SMRU) in Mae Sot, Thailand and from communities close to Iquitos city in the region of Loreto, Peru. A 5-ml sample of blood was collected by venipuncture in lithium-heparinized tubes from patients with P. vivax malaria mono-infection with a parasite density >1/1000 red blood cells. Samples with ≥80% parasites at the ring stage were platelet-and leukocyte-depleted using a CF11 column as previously described 67 . Purified parasites were frozen in glycerolyte 42 and stored in liquid nitrogen. The samples were then shipped to ITM, Belgium where the experiments were performed. Table S1 shows a summary of the P. vivax isolates used in each experiment and the parasite density at the time of collection. Reticulocyte-enriched red blood cells from hemochromatosis patients. Reticulocytes were enriched from 450 ml of peripheral blood collected from 13 hemochromatosis patients undergoing therapeutic phlebotomy at the ZNA Sint-Erasmus Hospital in Antwerp, Belgium. Blood samples were collected in SEPACELL bags (Fresenius Kabi) and processed within 48 hours following previously described protocols 68 . In brief, we pass blood through the leucocyte depleting filter (Fresenius Kabi) connected with SEPACELL blood bag. The Duffy phenotype (Fy) was determined by standard serologic methods (DiaMed-ID Micro Typing Systems, DiaMed) and blood grouping was done using a standard ABO antisera kit (Diamed). After depleting platelets and leukocytes, reticulocytes were concentrated by centrifugation (15 min at 400 g without applying a brake) through 70% Percoll 42 with minor modifications to Percoll concentration, which was adjusted in each sample to obtain a higher reticulocyte yield ( Table 1 ). The proportion of reticulocytes was calculated from the thin smear stained with New Methylene Blue (Sigma) under light microscopy. Samples with a proportion of reticulocytes higher than 25% were used for P. vivax invasion tests. Reticulocyte freezing and thawing was performed as previously described 42 . Invasion inhibition assays. P. vivax maturation and invasion assays were performed as previously described 42, 47 . Briefly, thawed P. vivax parasites were cultured to the schizont stage. Mature schizonts were concentrated after centrifugation (15 minutes at 1200 g) through 45% Percoll solution, which yields 90-98% enrichment of schizonts. The concentrated mature schizonts were mixed with the reRBC samples in a 1:6 (1 µl schizont and 6 µl reRBC) ratio, and cultivated in McCoy medium (Invitrogen) supplemented with 25% human serum and 0.5% glucose for 25 h. Invasion inhibition assays were set up in a 96-well culture plate with a final volume of 150 µl per well. Parasite cultures were monitored by microscopic examination of the Giemsa-stained thin film. Invasion rate was defined as the percentage of P. vivax ring-stage invaded RBCs per 9,000 RBCs 24 hours post-invasion. Invasion inhibition was measured in paired reRBC samples (treated vs. untreated control well) using the same parasite isolate. An invasion assay was considered valid when parasitemia of the untreated control well was ≥0.5% with one exception (0.42%). Erythrocyte invasion with 3D7 P. falciparum parasites were used as a positive control for reRBC invasion. The used sCR1 is commercially available (Cat No 5748-CD-050, R&D SYSTEMS, USA) and supplied as lyophilized power in sterile condition without a hazard preservative and constituted in sterile PBS (400 µg/ml) as per manufacturer's instruction. We performed invasion inhibition with sCR1 (50 µg/ml concentration) as described in previous P. falciparum experiments 15, 16 . Bovine serum albumin (BSA: 50 µg/ml) and PBS (12.5%) controls were used to assess the specificity of the assay. Non-hazardous effect of the non-dialyzed sCR1 protein was tested in P. falciparum 3D7 invasion assays compared to the dialyzed sCR1 protein. Paired comparisons of invasion rates between treatments were analyzed using the non-parametric Wilcoxon signed rank test in STATA v13. Enzymatic treatment of red blood cells. Enzymatic treatment of reticulocyte-enriched samples was performed by incubating 10 µl of the cell suspension with 1 mg/ml trypsin (from bovine pancreas, Sigma), 1 mg/ ml chymotrypsin (from bovine pancreas, Sigma) and 0.5 U/ml neuraminidase (from Vibrio cholerae, Sigma) at 37 °C for 1 hour 69 . CR1 enzymatic cleavage efficiency was tested by FACSCalibur 4-color flow cytometer (BD Biosciences) 15 and neuraminidase treatment was assessed by performing an agglutination test using lectin from the peanut Arachis hypogaea, to detect T-antigen which is exposed after sialic acid cleavage 69 . Flow cytometry. Expression levels of CR1 and CD71 were quantified with a FACSCalibur 4-color flow cytometer. Briefly, 200,000 cells were incubated with 5 µl of monoclonal mouse anti-human CR1-PE [E11 mAb] (Biolegend), 20 µl of monoclonal mouse anti-human CD71-APC (BD Biosciences), and 20 µl of monoclonal mouse anti-human DARC/Duffy antigen-FITC (LSBio) for one hour at room temperature. IgG1-PE, IgG1-APC and IgG-FITC conjugated antibodies were used as negative controls. Protein levels are expressed as MFI ( Figure S1 ). Protein expression and purification. DNA sequences encoding PvRBP1a (amino acids 160-1170), PvRBP1b (amino acids 140-1275), PvRBP2a (amino acids 160-1135), PvRBP2b (amino acids 161-1454), and PvRBP2c (amino acid 501-1300) were codon-optimized from the Sal-I protein sequences for Escherichia coli expression and inserted into a pPROEX HTB expression vector (Life Technologies). Protein expression and purification was performed as described earlier 70 . Immuno-precipitation assays. sCR1, recombinant PvRBPs and PfRH4 (88 kDa) 71 were incubated at 0.05 mg/ml in a reaction volume of 50-100 µl for 1 hour at room temperature. Anti-CR1 monoclonal antibody HB8592 (ATCC) was added to the mixture at 0.02 mg/ml for 1 hour, followed by additional incubation with 10 µl of packed Protein G Sepharose beads to capture the anti-CR1 mAb. The beads were washed three times with 200 µl PBS and the proteins were eluted with equivalent volumes of 2X reducing sample buffer and boiled for three www.nature.com/scientificreports www.nature.com/scientificreports/ minutes before separating on SDS/PAGE gels. Immuno-precipitation eluates were fractionated on SDS-PAGE and visualized using SimplyBlue SafeStain (Life Technologies), according to the manufacturer's protocol. CR1 SNP genotyping. Genotyping of 582 blood samples (Supplemental data) was performed following previously described protocols 23 . Primer sequences, amplification conditions, and restriction enzymes are detailed in Table S4 . For some of the reRBC samples an aliquot for DNA extraction was not available and therefore, we extracted RNA and genotyping PCR was performed on cDNA. Allele frequency distribution and linkage disequilibrium (LD) analysis. LD was assessed using data from 1000 Genomes Project Phase 3, genotyping data retrieved from published studies 23, 51, 57, 72 ) and in-house genotyped. R 2 and D' were measured using LDlink web tool (http://analysistools.nci.nih.gov/LDlink/) 73 , Ensembl database (http://www.ensembl.org/index.html) for biallelic SNPs 74 and CubeX (http://www.oege.org/software/ cubex/) 75 for genotype data. R 2 and D' values ≥0.8 were considered indicative of significant correlation and LD between both SNPs. CR1 SNP data was extracted from publicly available databases, i.e., the 1000 Genomes Project Phase 3 76 , ALFRED (Allele Frequency Database, U.S. National Sciences Foundation) 77 , and SGVP (Singapore Genome Variation Project) 78 , and from 39 international peer-reviewed publications (Table S3 ). Comparisons were performed using chi-squared tests. P values < 0.05 were considered significant. Global Epidemiology of Plasmodium vivax Overlaying Molecular and Temporal Aspects of Malaria Parasite Invasion Red cell receptors as access points for malaria infection Hematopoietic stem/progenitor cell sources to generate reticulocytes for Plasmodium vivax culture A receptor for the malarial parasite Plasmodium vivax: the erythrocyte chemokine receptor The resistance factor to Plasmodium vivax in blacks. The Duffy-bloodgroup genotype, FyFy Red blood cell polymorphism and susceptibility to Plasmodium vivax Interaction of Plasmodium vivax Tryptophan-rich Antigen PvTRAg38 with Band 3 on Human Erythrocyte Surface Facilitates Parasite Growth Receptor specific binding regions of Plasmodium vivax tryptophan rich antigens and parasite growth inhibition activity of PvTRAg35.2 Host-parasite interaction: multiple sites in the Plasmodium vivax tryptophanrich antigen PvTRAg38 interact with the erythrocyte receptor band 3 Identification of a reticulocyte-specific binding domain of Plasmodium vivax reticulocyte-binding protein 1 that is homologous to the PfRh4 erythrocyte-binding domain Significant biochemical, biophysical and metabolic diversity in circulating human cord blood reticulocytes The complement system and adaptive immunity Complement receptor 1 is a sialic acid-independent erythrocyte receptor of Plasmodium falciparum Complement receptor 1 is the host erythrocyte receptor for Plasmodium falciparum PfRh4 invasion ligand falciparum rosetting mediated by a parasite-variant erythrocyte membrane protein and complement-receptor 1 Identification of a restriction fragment length polymorphism by a CR1 cDNA that correlates with the number of CR1 on erythrocytes A human complement receptor 1 polymorphism that reduces Plasmodium falciparum rosetting confers protection against severe malaria Analysis of complement receptor type 1 (CR1) expression on erythrocytes and of CR1 allelic markers in Caucasian and African American populations CR1 density polymorphism on erythrocytes of falciparum malaria patients in Thailand Erythrocyte CR1 expression level does not correlate with a HindIII restriction fragment length polymorphism in Africans; implications for studies on malaria susceptibility Quantitative alleles of CR1: coding sequence analysis and comparison of haplotypes in two ethnic groups The genome of the simian and human malaria parasite Plasmodium knowlesi High CR1 level and related polymorphic variants are associated with cerebral malaria in eastern-India. Infection, genetics and evolution: journal of molecular epidemiology and evolutionary genetics in infectious diseases 11 CR1 levels and gene polymorphisms exhibit differential association with falciparum malaria in regions of varying disease endemicity A functional single-nucleotide polymorphism in the CR1 promoter region contributes to protection against cerebral malaria Phenotypic variation of Plasmodium falciparum merozoite proteins directs receptor targeting for invasion of human erythrocytes Evidence for a switching mechanism in the invasion of erythrocytes by Plasmodium falciparum Targeted disruption of an erythrocyte binding antigen in Plasmodium falciparum is associated with a switch toward a sialic acid-independent pathway of invasion Erythrocyte-binding antigen 175 mediates invasion in Plasmodium falciparum utilizing sialic acid-dependent and -independent pathways Ability of Plasmodium falciparum to invade Southeast Asian ovalocytes varies between parasite lines An adaptable two-color flow cytometric assay to quantitate the invasion of erythrocytes by Plasmodium falciparum parasites Identification of proteins from Plasmodium falciparum that are homologous to reticulocyte binding proteins in Plasmodium vivax Sequencing of 50 human exomes reveals adaptation to high altitude Natural selection and infectious disease in human populations The present status of malaria in the world Geography of Malaria: A Medico-Geographic Profile of an Ancient Disease A long neglected world malaria map: Plasmodium vivax endemicity in 2010 Improvement of culture conditions for long-term in vitro culture of Plasmodium vivax Sample related factors affecting short-term culture of erythrocytic Plasmodium vivax in vitro Cryopreserved Plasmodium vivax and cord blood reticulocytes can be used for invasion and short term culture Transferrin receptor 1 is a reticulocyte-specific receptor for Plasmodium vivax Plasmodium vivax: restricted tropism and rapid remodeling of CD71-positive reticulocytes Reticulocytes from cryopreserved erythroblasts support Plasmodium vivax infection in vitro Short-term in vitro culture of field isolates of Plasmodium vivax using umbilical cord blood A reliable ex vivo invasion assay of human reticulocytes by Plasmodium vivax In vitro evaluation of the role of the Duffy blood group in erythrocyte invasion by Plasmodium vivax The unhealthy attraction of Plasmodium vivax to reticulocytes expressing transferrin receptor 1 (CD71) Gene flow and the limits to natural selection Evidence for malaria selection of a CR1 haplotype in Sardinia Complement receptor 1 and malaria Polymorphisms of complement receptor 1 and interleukin-10 genes and systemic lupus erythematosus: a meta-analysis C3b receptor (CR1) genomic polymorphism in rheumatoid arthritis. Low receptor levels on erythrocytes are an acquired phenomenon Acquired but reversible loss of erythrocyte complement receptor 1 (CR1, CD35) and its longitudinal alteration in patients with severe acute respiratory syndrome Complement receptor 1 gene variants are associated with erythrocyte sedimentation rate Complement receptor 1 gene polymorphisms are associated with idiopathic pulmonary fibrosis Red blood cell complement receptor one level varies with Knops blood group, alpha(+)thalassaemia and age among Kenyan children Reduced risk of Plasmodium vivax malaria in Papua New Guinean children with Southeast Asian ovalocytosis in two cohorts and a case-control study Structurally conserved erythrocyte-binding domain in Plasmodium provides a versatile scaffold for alternate receptor engagement A reticulocyte-binding protein complex of Plasmodium vivax merozoites Plasmodium vivax MSP-1 peptides have high specific binding activity to human reticulocytes Identification, characterization and antigenicity of the Plasmodium vivax rhoptry neck protein 1 (PvRON1) a new Plasmodium vivax rhoptry neck antigen Annotation and characterization of the Plasmodium vivax rhoptry neck protein 4 (PvRON4) The Plasmodium vivax rhoptry neck protein 5 is expressed in the apical pole of Plasmodium vivax VCG-1 strain schizonts and binds to human reticulocytes Effective and cheap removal of leukocytes and platelets from Plasmodium vivax infected blood Continuous in vitro propagation of the malaria parasite Plasmodium vivax Invasion profiles of Brazilian field isolates of Plasmodium falciparum: phenotypic and genotypic analyses Gene Models, Expression Repertoire, and Immune Response of Plasmodium vivax Reticulocyte Binding Proteins Antibodies to reticulocyte binding protein-like homologue 4 inhibit invasion of Plasmodium falciparum into human erythrocytes Quantitative polymorphism of complement receptor type 1 (CR1) in patients undergoing haemodialysis LDlink: a web-based application for exploring population-specific haplotype structure and linking correlated alleles of possible functional variants Cubic exact solutions for the estimation of pairwise haplotype frequencies: implications for linkage disequilibrium analyses and a web tool 'CubeX' An integrated map of genetic variation from 1,092 human genomes ALFRED: an allele frequency resource for research and teaching Singapore Genome Variation Project: a haplotype map of three Southeast Asian populations The neglected burden of Plasmodium vivax malaria This work was supported by the ITM's Secondary Research Funding (SOFI), Antwerp, Belgium. We thank the patients and clinical staff at UPCH (Iquitos Peru) and SMRU (Thailand) for their cooperation and contribution to the collection of P. vivax isolates for the ex vivo invasion assays. We are also grateful to L. Boel and A. Ceulemans and Odin Goovaerts for their support in the FACS experiments. This research was made possible thanks to the financial support of the Flemish Government (SOFI) to ARU and the FWO for the short travel grant program [K207216N to S.K.P.] and FWO-SB PhD fellowship to K.D.M. Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-45228-6.Competing Interests: The authors declare no competing interests.Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.