Receptor-ligand and parasite protein-protein interactions in Plasmodium vivax : Analysing rhoptry neck proteins 2 and 4
2018; Wiley; Volume: 20; Issue: 7 Linguagem: Inglês
10.1111/cmi.12835
ISSN1462-5822
AutoresMaritza Bermúdez, Gabriela Arévalo‐Pinzón, Laura Peralta Rubio, Olivier Chaloin, Sylviane Muller, Hernando Curtidor, Manuel A. Patarroyo,
Tópico(s)Vector-borne infectious diseases
ResumoCellular MicrobiologyVolume 20, Issue 7 e12835 RESEARCH ARTICLEFree Access Receptor–ligand and parasite protein–protein interactions in Plasmodium vivax: Analysing rhoptry neck proteins 2 and 4 Maritza Bermúdez, Maritza Bermúdez Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, ColombiaThese authors are equal contributors.Search for more papers by this authorGabriela Arévalo-Pinzón, Gabriela Arévalo-Pinzón Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, Colombia PhD Programme in Biomedical and Biological Sciences, Universidad del Rosario, Bogotá, ColombiaThese authors are equal contributors.Search for more papers by this authorLaura Rubio, Laura Rubio Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, ColombiaSearch for more papers by this authorOlivier Chaloin, Olivier Chaloin CNRS, Immunopathology and therapeutic chemistry, Institut de Biologie Moléculaire et Cellulaire (IBMC), Strasbourg, FranceSearch for more papers by this authorSylviane Muller, Sylviane Muller CNRS, Immunopathology and therapeutic chemistry, Institut de Biologie Moléculaire et Cellulaire (IBMC), Strasbourg, France CNRS, Biotechnology and cell signaling, University of Strasbourg, France / Laboratory of Excellence Medalis, France University of Strasbourg Institute for Advanced Study (USIAS), Strasbourg, FranceSearch for more papers by this authorHernando Curtidor, Hernando Curtidor Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, Colombia School of Medicine and Health Sciences, Universidad del Rosario, Bogotá, ColombiaSearch for more papers by this authorManuel Alfonso Patarroyo, Corresponding Author Manuel Alfonso Patarroyo mapatarr.fidic@gmail.com orcid.org/0000-0002-4751-2500 Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, Colombia School of Medicine and Health Sciences, Universidad del Rosario, Bogotá, Colombia Correspondence Manuel Alfonso Patarroyo, Fundación Instituto de Inmunología de Colombia (FIDIC), Carrera 50 # 26-20, Bogotá, Colombia. Email: mapatarr.fidic@gmail.comSearch for more papers by this author Maritza Bermúdez, Maritza Bermúdez Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, ColombiaThese authors are equal contributors.Search for more papers by this authorGabriela Arévalo-Pinzón, Gabriela Arévalo-Pinzón Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, Colombia PhD Programme in Biomedical and Biological Sciences, Universidad del Rosario, Bogotá, ColombiaThese authors are equal contributors.Search for more papers by this authorLaura Rubio, Laura Rubio Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, ColombiaSearch for more papers by this authorOlivier Chaloin, Olivier Chaloin CNRS, Immunopathology and therapeutic chemistry, Institut de Biologie Moléculaire et Cellulaire (IBMC), Strasbourg, FranceSearch for more papers by this authorSylviane Muller, Sylviane Muller CNRS, Immunopathology and therapeutic chemistry, Institut de Biologie Moléculaire et Cellulaire (IBMC), Strasbourg, France CNRS, Biotechnology and cell signaling, University of Strasbourg, France / Laboratory of Excellence Medalis, France University of Strasbourg Institute for Advanced Study (USIAS), Strasbourg, FranceSearch for more papers by this authorHernando Curtidor, Hernando Curtidor Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, Colombia School of Medicine and Health Sciences, Universidad del Rosario, Bogotá, ColombiaSearch for more papers by this authorManuel Alfonso Patarroyo, Corresponding Author Manuel Alfonso Patarroyo mapatarr.fidic@gmail.com orcid.org/0000-0002-4751-2500 Fundación Instituto de Inmunología de Colombia (FIDIC), Bogotá, Colombia School of Medicine and Health Sciences, Universidad del Rosario, Bogotá, Colombia Correspondence Manuel Alfonso Patarroyo, Fundación Instituto de Inmunología de Colombia (FIDIC), Carrera 50 # 26-20, Bogotá, Colombia. Email: mapatarr.fidic@gmail.comSearch for more papers by this author First published: 27 February 2018 https://doi.org/10.1111/cmi.12835Citations: 11 AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract Elucidating receptor–ligand and protein–protein interactions represents an attractive alternative for designing effective Plasmodium vivax control methods. This article describes the ability of P. vivax rhoptry neck proteins 2 and 4 (RON2 and RON4) to bind to human reticulocytes. Biochemical and cellular studies have shown that two PvRON2- and PvRON4-derived conserved regions specifically interact with protein receptors on reticulocytes marked by the CD71 surface transferrin receptor. Mapping each protein fragment's binding region led to defining the specific participation of two 20 amino acid-long regions selectively competing for PvRON2 and PvRON4 binding to reticulocytes. Binary interactions between PvRON2 (ligand) and other parasite proteins, such as PvRON4, PvRON5, and apical membrane antigen 1 (AMA1), were evaluated and characterised by surface plasmon resonance. The results revealed that both PvRON2 cysteine-rich regions strongly interact with PvAMA1 Domains II and III (equilibrium constants in the nanomolar range) and at a lower extent with the complete PvAMA1 ectodomain and Domains I and II. These results strongly support that these proteins participate in P. vivax's complex invasion process, thus providing new pertinent targets for blocking P. vivax merozoites' specific entry to their target cells. 1 INTRODUCTION The invasion cycle of parasites from the phylum Apicomplexa (i.e., Toxoplasma gondii or Plasmodium spp) represents one of the most complex pathogen invasion processes that is actively orchestrated by the parasite itself through a series of events involving a diverse set of molecular interactions (Carruthers, 2002; Weiss, Crabb, & Gilson, 2016). Such process requires a sequence of coordinated activities, including recognition, host cell attachment, protein secretion (from micronemes and rhoptries) and motility, supported by an actin-/myosin-based gliding in Apicomplexan parasites (Weiss et al., 2015). One of its most outstanding features concerns the formation of stable, high-avidity merozoite (Mz) apex binding to a host cell, resembling a form of ring moving progressively (moving junction [MJ]) from the apical pole towards the back of the parasite, propelling it into a nascent parasitophorous vacuole, thereby allowing it to survive and replicate within a target cell (Aikawa, Miller, Johnson, & Rabbege, 1978). Studies orientated towards identifying the components forming the circumferential ring have shown that it appears to be mainly formed by a multimeric protein complex, involving the parasite's apical membrane antigen 1 (AMA1) and rhoptry neck (RON) proteins -2, -4, -5, and -8 (the latter only in T. gondii) (Alexander, Mital, Ward, Bradley, & Boothroyd, 2005; Besteiro, Michelin, Poncet, Dubremetz, & Lebrun, 2009; Collins, Withers-Martinez, Hackett, & Blackman, 2009; Straub, Cheng, Sohn, & Bradley, 2009). The RON/AMA1 complex acts independently of host cell receptors as Apicomplexans secrete RON proteins into a host cell, using RON2 as receptor on target cell membrane. RON2 is anchored to host cell surface, and the carboxyl terminal region remains exposed for binding to AMA1, which acts as parasite ligand (Besteiro et al., 2009; Lamarque et al., ; Srinivasan et al., 2011). Regarding the other RONs, it has been described that RON5 and RON8 in T. gondii are involved in stabilising and organising the complex in a host cell (Beck, Chen, Kim, & Bradley, 2014; Straub, Peng, Hajagos, Tyler, & Bradley, 2011). It has been found that conditional knockdown involving a loss of TgRON5 has led to the complete degradation of TgRON2 and mistargeting of TgRON4 (Beck et al., 2014), whereas a loss of TgRON4 has significantly reduced TgRON5 and TgRON2 levels (Guerin et al., 2017). It has also been shown that four host proteins (ALIX-CD2AP-CIN85 and TSG101) are specifically recruited to the MJ through specific binding motifs present on TgRON proteins (Guerin et al., 2017). Structural studies of the complex formed between AMA1 and a RON2 peptide have shown the participation of the AMA1 hydrophobic trough, so that a binding pocket accepts the RON2 critical loop region, having significant shape and complementary charge (Tonkin et al., 2011; Vulliez-Le Normand et al., 2012; Vulliez-Le Normand, Saul, Hoos, Faber, & Bentley, 2017). This interaction's importance in Plasmodium falciparum has been established by using antibodies, peptides, or small molecules directed against the PfAMA1-PfRON2 complex interface, thereby significantly inhibiting invasion (Collins et al., 2009; Lamarque et al., ; Srinivasan et al., 2013). Immunising mice with the AMA1-RON2L complex (but not with individual antigens) has induced qualitatively higher growth inhibitory antibodies capable of protecting mice against experimental challenge with a lethal Plasmodium yoelii strain, highlighting such protein–protein interaction's important role during invasion (Srinivasan et al., 2014). AMA1 was initially identified in Plasmodium knowlesi (Deans et al., 1982) and then found throughout all Apicomplexa, being essential for T. gondii and P. falciparum survival (Mital, Meissner, Soldati, & Ward, 2005; Triglia et al., 2000). Regarding AMA1's function in MJ formation, various roles involved in P. falciparum invasion have been attributed to it, including Mz reorientation on red blood cell (RBC) surface (Mitchell, Thomas, Margos, Dluzewski, & Bannister, 2004), participation in resealing RBC at the end of invasion (Yap et al., 2014), interaction with aldolase that provides extracellular linkage for the actin/myosin motor (Diaz et al., 2016), and interaction with receptors on RBC surface (Kato, Mayer, Singh, Reid, & Miller, 2005; Urquiza et al., 2000). Previous observations showed that tachyzoites (Tz) and sporozoites (Spz) having conditional knockouts for AMA1 were capable of invading a host cell and appeared to form normal MJ, suggesting that AMA1 was acting independently of RONs and that AMA1–RON2 interaction was not essential (Bargieri et al., 2013; Giovannini et al., 2011). However, later studies showed that Tz used paralogues of the generic AMA1–RON2 pair as substitutes, partially compensating for the loss of TgAMA1 (Lamarque et al., 2014). Alternative pathways for the lack of AMA1 and RON2 have not been reported in P. falciparum (Lamarque et al., 2014), unlike that observed in other invasion steps (such as reorientation) using functionally redundant proteins, such as erythrocyte binding antigens (EBAs) and reticulocyte-binding-like protein homologues (RH; Lopaticki et al., 2011). This would suggest that although it might be a conserved mechanism amongst Apicomplexa, differences concerning regulation of MJ components, composition, and interaction regions could partly define selectivity by their respective host cells. P. falciparum RONs were identified by coimmunoprecipitation assays showing AMA1 association with high molecular weight components located in the rhoptry neck (Curtidor, Patino, Arevalo-Pinzon, Patarroyo, & Patarroyo, 2011) homologues of those reported in T. gondii (Alexander, Arastu-Kapur, Dubremetz, & Boothroyd, 2006; Bradley et al., 2005; Cao et al., 2009). Even though there is no clear evidence of PfRON2, PfRON4, and PfRON5 translocation across/onto host cell membrane and RBC cytosol in P. falciparum, previous studies have shown that PfRON2 uses a cysteine-rich region (PfRON2-C) located in the carboxyl terminal extreme for interacting with receptors on a host cell; these are sensitive to treatment with trypsin and neuraminidase (Hossain, Dhawan, & Mohmmed, 2012). However, the cysteine-rich region located in PfRON2's central portion (PfRON2-M) and the PfRON2-C region interact with PfAMA1 (Hossain et al., 2012). PfRON5 contains peptides having high affinity binding to receptors on RBC, which are capable of inhibiting Mz entry to target cells. Such data highlight RON and AMA1 participation in establishing receptor–ligand and protein–protein interactions (Counihan, Kalanon, Coppel, & de Koning-Ward, 2013; Remarque, Faber, Kocken, & Thomas, 2008). Regarding Plasmodium vivax, the geographically most widespread cause of human malaria (Howes et al., 2016), there is little information concerning the proteins participating in binding (ligand–receptor interaction) or protein–protein interactions, partly due to difficulties in culturing P. vivax in vitro thereby impeding genetic approaches for studying proteins' role during the cycle of invasion of reticulocytes. Such limitations have been a barrier to studying the biological, clinical, and immunological characteristics necessary for designing control measures against P. vivax. Using P. vivax strains adapted in non-human primates (Pico de Coana et al., 2003) or infected patients' blood (Bozdech et al., 2008) has enabled comparative approaches with other Plasmodium species (Patarroyo, Calderon, & Moreno-Perez, 2012) or using omic sciences (Moreno-Perez, Degano, Ibarrola, Muro, & Patarroyo, 2014; Venkatesh et al., 2016) to identify a significant amount of proteins expressed in P. vivax schizonts (Sch) and Mz (Patarroyo et al., 2012). PvRON2 (Arevalo-Pinzon, Curtidor, Patino, & Patarroyo, 2011), PvRON4 (Arevalo-Pinzon, Curtidor, Abril, & Patarroyo, 2013), and PvRON5 (Arevalo-Pinzon, Bermudez, Curtidor, & Patarroyo, 2015) have been identified recently; these are homologous to those identified in P. falciparum, which are located in the apical extreme of P. vivax Colombia Guaviare 1 (VCG-1) strain Sch. Different biochemical techniques have been used to report a conserved PvRON5 fragment located towards the carboxyl-terminal extreme interacting with RBC, having a preference for CD71+ cells (Arevalo-Pinzon et al., 2015), suggesting that RONs could be participating in host–parasite interactions, similar to that found in P. falciparum. PvRON2 and PvRON4 ability to interact specifically with RBC was thus evaluated to advance the functional characterisation of P. vivax RONs; this was followed by determining binary interactions between PvRON2, PvRON4, PvRON5, and PvAMA1 by surface plasmon resonance (SPR). The data indicate that the PvRON2 central and PvRON4 carboxyl terminal regions specifically interact with protein receptors on reticulocyte membrane having CD71+CD45− phenotype. It was also shown here that PvRON2 also participates in interaction with other parasite proteins, such as PvAMA1, PvRON4, and PvRON5, highlighting P. vivax RON participation in establishing several protein–protein and host–pathogen interactions. 2 RESULTS AND DISCUSSION 2.1 The PvRON4 carboxyl terminal region interacted with umbilical cord blood RBC Recombinant plasmids were constructed based on evidence regarding two P. falciparum RON2 cysteine-rich regions' binding and interaction activity (Hossain et al., 2012; Lamarque et al., ); they contained the regions encoding the P. vivax central (PvRON2-RI) and carboxyl terminal (PvRON2-RII) regions (Figure 1a). A PvRON4 region located towards the carboxyl terminal extreme was amplified (Figure 1a); it is highly conserved at intraspecies and interspecies level and is under purifying selection, contrary to that reported for the amino terminal region having extensive size polymorphism (Buitrago, Garzon-Ospina, & Patarroyo, 2016). Selected regions were cloned in pRE4 vector in frame with herpes simplex virus (HSV) glycoprotein D signal peptide and transmembrane domain thereby enabling proteins to be expressed and anchored to cell membrane (Cohen et al., 1988). Following transfection and immunofluorescent studies on nonpermeabilised cells, it was found that only PvRON4 was correctly and efficiently expressed on COS-7 cell membrane (Figure 1b). Several modifications to COS-7 cell transfection and expression protocols for PvRON2-RI and PvRON2-RII fragments led to low transfection and expression efficiency compared with that obtained with PvRON4 and PvDBP-RII recombinant fragments. It should be noted that some transient transfections in COS-7 cells may have relatively low protein expression rates, as has been reported previously for other proteins (Berntzen et al., 2005). Figure 1Open in figure viewerPowerPoint PvRON4 binding to RBCs. (a) PvRON2 and PvRON4 schematic representation. Each protein's main characteristics are shown. Signal sequence (dark blue), transmembrane domains (black), coiled-coil domains (grey), and cysteine residues conserved amongst Plasmodium vivax strains (dotted red lines) (Arevalo-Pinzon et al., 2011; Arevalo-Pinzon, Curtidor, Abril, & Patarroyo, 2013). The locations are shown of two PvRON2 fragments (PvRON2-RI and PvRON2-RII) and one PvRON4 fragment, which were cloned in vectors to be expressed in COS-7 cells and Escherichia coli. (b) PvDBPRII and PvRON4 expression and location on COS-7 eukaryote cell surface. Above: Green shows expressed proteins and blue shows DAPI-stained cell nuclei. Transfection efficiency was ~5%, which was used for normalising the Rosetting data. Below: An example of a rosette indicating RBC bound to COS-7 cell surface. (c) Rosette formation assay. Computation of the amount of rosettes counted in 20 fields using UCB RBC from different Duffy phenotypes with COS-7 cells transfected with PvDBPRII or PvRON4. A rosette assay involving normocytes having Fya+Fyb− phenotype was also included (d) Enzyme treatment assays. The amount of rosettes formed between cells transfected with PvDBPRII or PvRON4 and UCB RBC treated with different enzymes is shown. All Rosetting assays were performed three times in triplicate The P. vivax infection has been seen to be limited to Duffy (Fy) positive reticulocytes, also known as the Duffy antigen/receptor for chemokines (DARC). Such preference seems to be mediated by specific receptor–ligand interactions between the DBP protein expressed by the parasite and the DARC receptor on RBC membrane (Miller, Mason, Clyde, & McGinniss, 1976). DBP–DARC interaction was taken as positive control based on the above information where PvDBP cysteine-rich Region II was expressed on COS-7 cell membrane (Figure 1b). As expected, PvDBPRII interacted with DARC whilst having low negative Duffy cell binding capacity, ANOVA-Tukey: F(3, 12) = 24.8, p < .0001 (Figure 1b,c). The PvRON4 C-terminal cysteine-rich region (pRE4-PvRON4 construct) bound to umbilical cord blood (UCB) RBC having different Duffy phenotypes, although their binding activity was lower than that found for PvDBP-RII (Figure 1b,c). PvRON4 binding to Fya+Fyb− had statistically significant differences regarding binding to Fya−Fyb− RBC, ANOVA-Tukey: F(3, 12) = 4.26, p = .020. When PvRON4 binding to RBC (m = 34 ± 2.7) was compared with binding to normocytes (m = 11 ± 1.4), it was observed that the amount of rosettes compared with normocytes became significantly reduced, t test: t(10) = 18.68, p = .001 (Figure 1b,c). This suggested that PvRON4 binding to RBC depends on a more immature RBC population, similar to that reported for other P. vivax antigens, such as PvAMA1 (Arevalo-Pinzon, Bermudez, Hernandez, Curtidor, & Patarroyo, 2017), PvGAMA (Baquero et al., 2017), and EBP2 (Ntumngia et al., 2016), whereas PvDBPRII bound to both RBC and normocytes (Figure 1c). Several reports have shown that different P. falciparum strains differ regarding their ability to invade RBC treated with different enzymes, such as trypsin, chymotrypsin, and neuraminidase, removing different receptors from RBC surface (Duraisingh et al., 2003; DeSimone et al., 2009). Such assays enabled ascertaining that proteins like erythrocyte binding antigen 175 (EBA175) recognised the sialic acid component in glycophorin A (Duraisingh, Maier, Triglia, & Cowman, 2003) and EBA 140 bound to glycophorin C (Gilberger et al., 2003). It has been reported (and also found here) that PvDBP binding to DARC was susceptible to enzymatic treatment with chymotrypsin (Barnwell, Nichols, & Rubinstein, 1989), t test: t(4) = 24.4, p = .001 (Figure 1d). It has also been reported that PvAMA1 binding was susceptible to neuraminidase and chymotrypsin action (Arevalo-Pinzon et al., 2017). Interestingly, it was found here that PvRON4 interacted with RBC membrane protein receptors susceptible to treatment with the three enzymes evaluated, ANOVA-Tukey: F(3, 8) = 59, p = .001 (Figure 1d). Such binding profile did not fit any enzymatic behaviour typical of previously described malaria receptors. Concerning other parasites such as T. gondii, a 17-kDa region in TgRON4 proximal C-terminus was sufficient for binding to host cell β-tubulin carboxyl-terminal region (Takemae et al., 2013). Even though the overall similarity between T. gondii, P. falciparum, and P. vivax RON4 is not very high (Supporting Information), a pattern of conservation could be observed towards these proteins' carboxyl terminal region. This suggested that functional RON4 binding region was located towards this region, even though receptor type differed according to species, as shown by TgRON4 (Takemae et al., 2013) and PvRON4 interaction results (Figure 1). Although it is still not clear whether RON4 is exported towards host cell cytosol, its participation in MJ formation is crucial for target cell invasion by T. gondii and P. falciparum (Giovannini et al., 2011). Recent studies have found that the PvRON4 central region contained a sterease/lipase domain, which could be associated with ester bonds rupture in the phospholipids constituting host cell membrane (Buitrago et al., 2016). This suggested the presence of several functional domains in RON4, where the C-terminal region could interact with RBC membrane whilst the sterease/lipase (central region) domain could then enable RON4 internalisation. Future studies aimed at elucidating such functional domains' coordinated involvement in invasion should be carried out. The mechanism by which proteins such as Tg, Pf, and Pv RON4, RON5, and RON8 are translocated to host cell cytoplasm has not yet been elucidated. It cannot be ruled out that proteins must first interact with proteins on host cell membrane for this to occur, as seen in this PvRON4 study (Figure 1). It has been found that one PvRON5 region and some PfRON5 20 residue-long peptides specifically interact with RBC membrane (Arevalo-Pinzon et al., 2015; Curtidor et al., 2014). 2.2 PvRON4 and PvRON2-RII bound specifically to CD71-labelled reticulocytes Three protein fragments were obtained in recombinant form in Escherichia coli for determining PvRON4–RBC interaction specificity and evaluating PvRON2-RI and PvRON2-RII specific binding capability (Figure 1a); these regions were then purified by affinity chromatography. A single band was found for each fragment by western blot and Coomassie blue staining, coinciding with the expected molecular weights (Figure 2a, left). The three recombinant fragments contained conserved cysteine residues amongst the different P. vivax strains (Arevalo-Pinzon et al., 2011; Arevalo-Pinzon, Curtidor, Abril, & Patarroyo, 2013). Although internal disulphide linkage formation in these fragments is still unknown, previous studies with the RON2sp1 show that disulphide bridge formation between Cys2051-Cys2063 (PvRON2sp1) and Cys2037-Cys2049 (PfRON2sp1) in RON2 carboxy terminal region is important for this peptide's interaction with the AMA-1 hydrophobic groove (Vulliez-Le Normand et al., 2012; Vulliez-Le Normand et al., 2017). Ellman test did not reveal sulfhydryl groups in these fragments, that is, no free thiol groups in any of the three recombinant proteins, indicating disulphide bridges. Thionitrobenzoic acid formation was observed when proteins were incubated with a molar excess of DTT, which is formed only in the presence of free sulfhydryl groups. Figure 2Open in figure viewerPowerPoint PvRON2-RI and PvRON4 proteins interacted specifically with CD71+ reticulocytes. (a) PvRON2 and PvRON4 expression and specific binding to UCB RBC. Left-hand side: protein recognition by western blot (1) with monoclonal antihistidine antibody and Coomassie blue staining (2) after the purification of the three proteins obtained in Escherichia coli. Right-hand side: specific binding assays. All recombinant proteins were radio-labelled with Na125I and made to compete for binding to UCB RBC in the absence (total binding) or presence of the same nonradiolabelled protein (non-specific binding). A specific binding curve was obtained from the two curves. CPM: counts per minute. Data are shown as the mean values (±SD) of at least three independent experiments. (b) PvRON2-RI and PvRON4 binding to CD71+CD45− as measured by flow cytometry. Dot plots show the evaluated recombinant proteins' binding to a CD71+CD45− reticulocyte population. Binding was detected by a PE-conjugated monoclonal antihistidine antibody. The percentage obtained for negative control (cells without recombinant protein) was subtracted from each binding assay. The bar graph represents each protein's average binding to CD71+CD45− cells obtained in three assays performed in triplicate Every recombinant fragment was radiolabelled, quantified, and incubated in a UCB RBC competition assay with high concentrations of the same nonradiolabelled recombinant protein. The results showed that even though the three fragments did bind (total binding), only PvRON4 and PvRON2-RI bound specifically to RBC, because radiolabelled protein binding became reduced (by ~55%) in the presence of nonradiolabelled protein (non-specific binding; Figure 2a, right). When graphing specific binding, it was observed that PvRON4 and PvRON2-RI specific binding activity was 1%-PvRON4 to 2.6%-PvRON2-RI (curve slope) compared with PvRON2-RI 0.22% specific activity (Figure 2a, right). As P. vivax Mrz have a strong preference for reticulocytes expressing transferrin receptor 1 (CD71; becoming sequentially lost as reticulocytes mature), cytometry was used for determining whether PvRON4 and PvRON2-RI binding was restricted to interaction with this type of immature cell (reticulocyte stages I-III). Figure 2b shows that PvRON2-RI had 6.56% binding to CD71+CD45− cells (m = 7.6 ± 0.94%, 95% CI [5.2, 9.9]) and PvRON4 5.60% (m = 5.15 ± 0.55%, 95% CI [3.7, 6.5]), whereas low CD71−CD45− cell interaction (~0.1%) was found (data not shown). PvDBPRII recombinant protein had 10.5% binding (positive control; m = 12.3 ± 1.47%, 95% CI [8.6, 15.95]) whereas PvDBPRIII-IV recombinant protein, expressing Regions III and IV (negative control), had low binding capacity (m = 1.0 ± 0.55%, [−0.3, 2.4]), thereby coinciding with previous studies showing that these regions do not participate in target cell binding (Chitnis & Miller, 1994; Ocampo et al., 2002). Although these proteins interacted preferentially with CD71-labelled reticulocytes, the PvRON4 protein receptor's enzymatic profile did not match the CD71 protein profile, indicating that this is not the PvRON4 receptor. It has been reported very recently that CD71 acts as receptor for the PvRBP2b ligand and that interaction is trypsin- and chymotrypsin-sensitive (Gruszczyk et al., 2018). It has been reported that the PfRON2 fragment involved in interaction with RBC is located in the carboxyl-terminal region (Hossain et al., 2012), different to that found for PvRON2 where, even though both regions bound, only the region located towards the protein's central portion (called PvRON2-RI here) specifically interacted with UCB RBC (Figure 2). Such data, added to that available for P. vivax proteins (e.g., PvMSP1 and PvAMA1, Arevalo-Pinzon et al., 2017, and now PvRON2), have shown that some P. vivax proteins having orthologues in P. falciparum fulfil the same functions but use different functional regions. This has important implications when designing effective control mechanisms against P. vivax malaria and could be partly responsible for the specificity of interaction with respective target cells. It is still not clear why P. vivax has tropism for CD71+ reticulocytes; the interaction defined between DBP and DARC was initially considered essential for parasite binding; however, DARC is expressed on both reticulocytes and erythrocytes. Recent studies have managed to show that DARC epitope recognised by DBP has increased its exposure during early stages of reticulocyte maturation and decreased it during maturation into normocytes (Ovchynnikova et al., 2017). DARC epitope exposure during reticulocyte stages I–III enables greater DBP association with CD71high/TOhigh reticulocytes, but not in mature reticulocytes or erythrocytes, partly explaining P. vivax tropism for reticulocytes (Ovchynnikova et al., 2017). It is worth noting that specific tropism for CD71+ reticulocytes has also been shown for other P. vivax proteins, including those studied here. Receptor remodelling during reticulocyte maturation, added to the fact that proteins have distinct binding domains in each species, could complement each parasite's selectivity for its respective host cell. 2.3 Twenty-residue-long peptides were capable of specifically competing for PvRON4 and PvRON2 binding Competition assays were carried out between peptides covering radiolabelled protein and recombinant protein fragments' amino acid sequences to define the regions responsible for PvRON2 and PvRON4 interaction with CD71+CD45− cells. The results showed that PvRON4 fragment-derived peptides 40305, 40306, 40312, and 40313
Referência(s)