The role of the reticulocyte-binding-like protein homologues of Plasmodium in erythrocyte sensing and invasion

Authors

  • Karthigayan Gunalan,

    1. Division of Molecular Genetics & Cell Biology, School of Biological Sciences, Nanyang Technological University, Singapore, Singapore
    Search for more papers by this author
  • Xiaohong Gao,

    1. Division of Molecular Genetics & Cell Biology, School of Biological Sciences, Nanyang Technological University, Singapore, Singapore
    Search for more papers by this author
  • Sally Shu Lin Yap,

    1. Division of Molecular Genetics & Cell Biology, School of Biological Sciences, Nanyang Technological University, Singapore, Singapore
    Search for more papers by this author
  • Ximei Huang,

    1. Division of Molecular Genetics & Cell Biology, School of Biological Sciences, Nanyang Technological University, Singapore, Singapore
    Search for more papers by this author
  • Peter R. Preiser

    Corresponding author
    1. Division of Molecular Genetics & Cell Biology, School of Biological Sciences, Nanyang Technological University, Singapore, Singapore
    Search for more papers by this author

Summary

Malaria remains a serious public health problem with significant morbidity and mortality accounting for nearly 20% of all childhood deaths in Africa. The cyclical invasion, cytoadherence and destruction of the host's erythrocyte by the parasite are responsible for the observed disease pathology. The invasive form of the parasite, the merozoite, uses a complex set of interactions between parasite ligands and erythrocyte receptors that leads to the formation of a tight junction and ultimately successful erythrocyte invasion. Understanding the molecular mechanism underlying host cell recognition and invasion is crucial for the development of a targeted intervention strategy. Two parasite protein families termed reticulocyte-binding-like protein homologues (RBL) and the erythrocyte-binding-like (EBL) protein family are conserved in all Plasmodium species and have been shown to play an important role in host cell recognition and invasion. Over the last few years significant new insights have been gained in understanding the function of the RBL family and this review attempts to provide an update with a specific focus on the role of RBL in signal transduction pathways during invasion.

Introduction

Malaria continues to be a serious public health problem with significant morbidity and mortality. Nearly half of the world's population live in malaria endemic areas. It is estimated that 190–311 million cases of malaria occur worldwide of which ∼ 1 million people die, most of them being children under the age of five (WHO and CDC). Apart from the toll on human life, the disease also greatly impacts the affected country's economy, whereby a significant portion of its health budget is used on controlling malaria. Malaria is caused by the obligate intracellular protozoan parasites belonging to the phylum Apicomplexa, genus Plasmodium. Five Plasmodium species are known to infect humans with Plasmodium falciparum being the most virulent causing the majority of death. The life cycle of Plasmodium is a complex process involving an insect vector and a vertebrate host. The parasite has three invasive stages, the sporozoite, merozoite and ookinete that invade the salivary gland and hepatocyte, the erythrocyte and the mosquito midgut respectively. During the erythrocytic part of the life cycle, the merozoite after successful erythrocyte invasion develops through a number of distinct morphological forms – rings, trophozoites and schizonts. Finally the mature schizonts burst to release between 8 and 32 invasive merozoites, which rapidly invade new erythrocytes to continue the cycle. All apicomplexan parasites are characterized by apical organelles (rhoptries, micronemes and dense granules), which play a fundamental role in the invasion process. In Plasmodium, rhoptries are the most prominent large secretory organelle present in pairs at the apical tip of the merozoite and their content is thought to be important during both initial host cell sensing and the establishment of the parasitophorus vacuole. In contrast micronemes are smaller secretory organelles that have one end attached to the rhoptry and their contents contributes to host cell recognition and junction formation (Preiser et al., 2000).

Key cellular events enabling successful invasion of merozoites

Schizont rupture to release invasive merozoites is an explosive event called egress that involves the disruption of both the parasitophorous vacuolar and red blood cell membrane. Live video microscopy indicates that egress occurs within 400 milliseconds (ms). Initially, the erythrocyte membrane curls around to form buckles causing the eversion of the erythrocyte membrane thereby providing the force for the merozoites to be propelled outward (Fig. 1) (Abkarian et al., 2011). The released merozoites then proceed to rapidly invade an appropriate erythrocyte.

Figure 1.

Schematic representation of stages involved during egress and merozoite invasion. Diagrammatic representation of sequence of events that the RBC membrane undergoes during the egress of merozoites from schizonts, i.e. the erythrocyte membrane undergoes a sequence of events during egress, starting with the membrane curling outwards. Following the membrane curling, a circular toroid shape is formed around the opening. The erythrocyte membrane further curls around to form buckles. This step provides the force for the merozoites to propel outward. Further buckling leads to eversion of the membrane. Once the merozoites are released from the schizont, the merozoites randomly attaches to the red blood cells, followed by apical reorientation. Junction is formed between the merozoites and the erythrocyte with the help of RBLs and EBLs. The immunofluorescence images in the schematic diagram indicate the colocalization (yellow) of PfRH (red) and EBA175 or AMA1 (green) during different stages of invasion – (i) reorientation of merozoites with the apical end facing the erythrocytes, (ii) tight junction formation and (iii) penetration of merozoite into the erythrocyte. Once junction formation is initiated the merozoites gradually penetrates and after the merozoite entry into the erythrocyte the erythrocyte membrane reseals again.

Invasion begins with the random attachment of any part of the merozoite with the erythrocyte surface. This molecular factors contributing to the initial low-affinity and reversible attachment is poorly understood although merozoites surface proteins like MSP1 are thought to play a role in this step. After attachment, the merozoite reorients itself such that the apical end faces towards the erythrocyte membrane (Fig. 1), a process thought to be at least in part mediated by the micronemal protein AMA1 (Cowman and Crabb, 2006; Gaur and Chitnis, 2011). Following reorientation, rhoptry and micronemal proteins mediate direct high-affinity merozoite-erythrocyte interactions (Preiser et al., 2000) that ultimately lead to tight junction formation and irreversible commitment of the merozoite to invasion. The tight junction is characterized by an electron dense thickening between the erythrocyte membrane and the merozoite and its molecular make up is not yet fully understood although it is known to include a number of rhoptry neck proteins as well as the micronemal protein AMA1. Some of the rhoptry neck proteins are directly inserted into the erythrocyte membrane while others appear to form the link between these inserted proteins and the merozoite (Gaur and Chitnis, 2011). Members of both the erythrocyte-binding-like (EBL) and reticulocyte-binding-like (RBL) protein family that are located in the micronemes and rhoptries have been shown to be translocated to the junction although they also play a direct role in the recognition and signalling events leading to junction formation. Importantly, the junction is directly associated with the actin-myosin motor that drives merozoite penetration. The movement generated by this parasite derived motor results in the junction moving from the anterior to the posterior end of the penetrating merozoite thereby facilitating parasite entry (reviewed in Gaur and Chitnis, 2011). Subsequently, the link between the merozoite and the erythrocyte breaks and the erythrocyte membrane reseals again once the parasite is successfully inside the erythrocyte (Fig. 1). Although, invasion represents a highly complex molecular event, live video microscopy indicates that merozoite entry into the erythrocyte completes within a minute. The first 11 s is the pre-invasion phase where dramatic erythrocyte deformation occurs after merozoite encounter. This is followed by the classical invasion phase where the merozoite becomes internalized within the erythrocyte. By 36 s the infected erythrocyte undergoes a dramatic dehydration type of morphology called echinocytosis, before recovering back to the normal shape over a period of 5–11 min (Gilson and Crabb, 2009).

The role of the reticulocyte-binding-like proteins in invasion: sensing, binding, signalling

Different human malaria parasite species have different erythrocyte preferences with Plasmodium vivax preferentially invading reticulocytes, Plasmodium malariae invading normocytes, while P. falciparum invades erythrocytes of all ages. Similar differences of erythrocyte preference are seen in other Plasmodium species. Importantly, restriction to a subset of circulating erythrocytes often results in self-limiting parasitaemia thereby directly impacting on parasite virulence. The RBLs represents key players that are directly involved in host cell selection and thereby virulence. Members of RBLs are found in all Plasmodium species and have been shown to be involved in the differential recognition of erythrocyte receptors (Cowman and Crabb, 2006). Here, we focus on recent important findings that provide new insights on the functional role of the RBLs during merozoite invasion.

Plasmodium yoelii RBLs

The first RBL family member was discovered in Plasmodium yoelii and termed Py235 (reviewed in Gruner et al., 2004; Iyer et al., 2007a). It was identified using monoclonal antibodies obtained from mice protected against the virulent YM strain of P. yoelii. These antibodies were used to locate these proteins in the rhoptries of P. yoelii merozoites. The same antibodies also recognized proteins in other rodent Plasmodium species, suggesting that Py235 is conserved across species. The ability of Py235 antibodies to restrict the invasion characteristics of the virulent P. yoelii line to reticulocytes suggested that Py235 plays a role in host cell selection and parasite virulence. The subsequent identification of genes coding for Py235 led to the realization that this protein is a member of a multigene family containing up to 50 members in P. yoelii with each member being characterized by a homologous 500-amino-acid region and conserved cysteine residues. In the reference genome 14 copies of Py235 were found suggesting that the initial copy number represented an overestimation of the PCR- and RFLP-based analysis. Transcriptional studies have identified quantitative differences in the level of Py235 expression as being a main contributor to parasite virulence, with high levels of Py235 transcription being associated with increased severity of disease, although differences in the Py235 repertoire of virulent and avirulent P. yoelii strains could also be a contributing factor. Not all members of Py235 are transcribed equally within a parasite population suggesting that different Py235 provide the parasite with slightly different invasion properties that are subsequently selected for the host for optimal parasite replication (Iyer et al., 2007b). One study showed that disruption of the dominantly transcribed Py235 does indeed impact on parasite virulence (Bapat et al., 2011) suggesting that other members of the protein family are only able to partially compensate for the missing member (Ogun et al., 2011).

Py235 is also involved in a novel form of clonal phenotypic variation in which merozoites originating from a single schizont transcribe distinct Py235 members (Preiser et al., 1999). Merozoites produced from a single schizont would therefore be expected to have differences in invasion potential as well as immunogenicity implicating Py235 in playing an important role in adaptation to the host environment as well as immune evasion. The role of Py235 in host cell selection is further supported by studies showing direct binding of Py235 to a specific erythrocyte receptor (Ogun and Holder, 1996).

Plasmodium vivax RBLs

Plasmodium vivax preferentially invades reticulocytes and this selective recognition was initially shown to be mediated by two proteins called P. vivax reticulocyte-binding proteins (PvRBP1 and 2) that appear to form a complex through non-covalent interactions (reviewed in Iyer et al., 2007a). These proteins are localized at the apical end of the merozoite and are able to specifically bind to reticulocytes through unknown receptors. Sequence comparison showed that PvRBP1 and PvRBP2 have weak but significant homology with the Py235 family members providing the first evidence that RBLs are conserved in human Plasmodium species. The recent completion of the P. vivax genome suggested that there are additional members of PvRBP (PvRBP1a, 1b, PvRBP2a, 2b, 2c, 2d and PvRBP 3) of which PvRBP2d and PvRBP3 appear to be pseudogenes (Kosaisavee et al., 2012). Overall members of RBL show limited diversity between different strains of P. falciparum or P. vivax with the exception of PvRBP2c (earlier named as PvRBP2) which is highly polymorphic suggesting it is under greater selection pressure.

Plasmodium falciparum RHs

Plasmodium falciparum reticulocyte-binding proteins homologues (PfRH) were initially identified using bioinformatics approaches as part of the genome-sequencing project. Currently, six PfRH members have been identified (Fig. 2A), Reticulocyte-binding protein homologue 1 (PfRH1), PfRH2a, PfRH2b, PfRH3, PfRH4 and PfRH5 (Cowman and Crabb, 2006). The gene structure of PfRH members (except PfRH5) comprises two conserved exons (with the first exon coding for signal sequence and second exon coding for rest of the protein), one intron and a transmembrane domain. The 500-amino-acid homology regions initially identified in Py235 and PvRBP showed low sequence homology between the different PfRH members but there were a number of conserved amino acid blocks which clearly identify them as members of this family.

Figure 2.

RBL family in P. falciparum, P. yoelii and P. vivax and their receptor specificity.

A. Sequence alignment of PfRH members showed a homology region within the first 500 amino acids (blue shade). Green bars represent homology region based on alignment of PfRH2a, Py235 and PvRBP2. Nucleotide (ATP/ADP) binding region has been confirmed in PY01365, the dominant member of Py235. Red bars indicate region of each PfRH members to which antibody has been raised against and these antibodies inhibit parasite invasion in vitro. Each member was represented based on the actual size of the protein, as indicated by the scale bar of 250 amino acids (black bar).

B. Receptors for P. falciparum reticulocyte-binding protein and its property. Complement receptor 1 (CR1) and basigin are the erythrocyte receptors for PfRH4 and RH5 respectively. Enzyme sensitivity of RHs to neuraminidase (Nm), trypsin and chymotrypsin treated erythrocytes are also listed. S, sensitive; R, resistance.

PfRH1

PfRH1 is an orthologue of the PvRBP1 located on chromosome 4. PfRH1 is expressed at the apical end of the merozoite and appears to reside in the rhoptries. PfRH1 binds to a sialic acid-containing, trypsin- and chymotrypsin-resistant receptor named Y (Fig. 2B) (Gao et al., 2008). Targeted gene disruption of PfRH1 showed that it is required for sialic acid-dependent invasion and consistent with this, expression levels of PfRH1 have marked differences between parasite strains using different invasion pathways (Cowman and Crabb, 2006). The copy number of PfRH1 can vary between different laboratory strains although this has not been confirmed in patient derived parasite populations (Nair et al., 2010), suggesting that PfRH1 copy number is not a major determinant of level of gene expression and invasion in the natural population. PfRH1 undergoes a series of proteolytic processing event resulting in products of around 120 and 240 kDa before and during invasion (Triglia et al., 2009). Both products are localized at the tight junction and it appears that at least some of the 120 kDa fragment is taken into the ring stage of the parasite.

PfRH2a and PfRH2b

PfRH2a and RH2b were identified by comparative analysis with PvRBP2 (Rayner et al., 2000) and share 8 kb of identical gene sequence at the 5′ end before a divergent and unique 3′ end. They are both located on chromosome 13 and are thought to have arisen by gene duplication. Both proteins are predicted to be more than 350 kDa in size. Both PfRH2a and 2b are located in the neck of the rhoptries of the merozoites (Cowman and Crabb, 2006) and like PfRH1, PfRH2a is located at the junction during invasion (Gunalan et al., 2011). PfRH2a and 2b show different expression levels in different P. falciparum strains and PfRH2b has been implicated in mediating a novel invasion pathway by interacting with a chymotrypsin-sensitive erythrocyte receptor (Cowman and Crabb, 2006). Furthermore, disruption of PfRH2a and PfRH2b shows that both ligands are required for sialic acid independent invasion (Desimone et al., 2009). Recent studies showed that the 140, 270 and 360 kDa fragments of PfRH2a directly bind to erythrocytes each with unique binding specificity (Gunalan et al., 2011) in line with the observation that a 220 kDa proteolytic product of PfRH2a/2b binds to erythrocytes with similar erythrocyte-binding properties (Sahar et al., 2011). In addition, the N-terminal 85 kDa fragment of both PfRH2a and 2b also binds to erythrocyte (Sahar et al., 2011; Triglia et al., 2011) consistent with the observation that different processed products of PfRH2a/b mediate different receptor interactions during merozoite invasion (Gunalan et al., 2011). Functional differences between PfRH2a and 2b appear not only to be mediated via distinct receptor-binding domains located in the ectodomain but also by differences in the cytoplasmic domain, suggesting differential signalling via the binding of unknown cytoplasmic proteins (Dvorin et al., 2010).

In clinical isolates as well as culture adapted parasites expression levels of PfRH1 and PfRH2a/2b varied in a fashion consistent with parasites utilizing different invasion pathways. PfRH2 is naturally immunogenic and anti-PfRH2 antibodies are associated with protection from malaria (Tham et al., 2012) in line with the observation that antibodies against PfRH2a/2b are able to inhibit merozoite invasion in vitro. Moreover, it was shown that PfRH2a, PfRH2b and PfRH4 cooperatively act together for invasion to occur (Desimone et al., 2009) suggesting that merozoite invasion properties are determined by the overall ratio of PfRH expression. The observation that loss of erythrocyte-binding ligand EBA181 ablates the ability of PfRH2a/b protein antibodies to inhibit merozoite invasion also suggests a so far poorly understood complex interplay between these two invasion protein families (Tham et al., 2012).

PfRH3

Based on the presence of mutations within the coding region that lead to reading frameshifts at the 5′ end of the gene of all sequences obtained to date, PfRH3 is thought to be non-translated pseudogene (Cowman and Crabb, 2006). Interestingly, different mutations have occurred in diverse parasite isolates leading to speculations that these mutations may reflect the evolution of the parasite through recent events. The gene is located on chromosome 12 and is transcribed in a stage-specific manner. Whether these ‘sterile’ transcripts still have a functional role during parasite development is not known.

PfRH4

PfRH4 codes for a 220 kDa protein and is located on chromosome 4 and like all the other members of the RH family undergoes processing to a 160 kDa protein (reviewed in Cowman and Crabb, 2006; Gaur and Chitnis, 2011; Tham et al., 2012). PfRH4 is located in the rhoptries and sequence comparisons shows a close similarity to PvRBP-1. Activation of PfRH4 expression changes the invasion properties of the W2mef parasite line from a sialic acid-dependent to a sialic acid-independent one demonstrating that differential activation of PfRH4 alters the ability of the parasite to utilize different erythrocyte receptors. This finding clearly demonstrated that differential expression of PfRH directly impacted on the invasion pathway utilized by the parasite. The erythrocyte-binding ability of PfRH4 was shown to be sialic acid independent and trypsin and chymotrypsin sensitive in line with the identification of complement receptor 1 (CR1) as its erythrocyte receptor. Antibodies raised to the binding domain of PfRH4 are able to block erythrocyte invasion as well as erythrocyte binding of native PfRH4 (Tham et al., 2012).

PfRH5

Unlike all the others, PfRH5 the smallest member of the PfRH family members appears to be essential for parasite invasion making this protein a unique target for vaccine intervention (Crosnier et al., 2011). In line with this PfRH5 has been shown to be expressed in all the parasites lines investigated so far (Tham et al., 2012). PfRH5 is localized in the rhoptries and moves with the tight junction during erythrocyte invasion (Baum et al., 2009). At around 63 kDa in size, PfRH5 is significantly smaller than the other PfRH and the fact that it also lacks a transmembrane and cytoplasmic domain at the C-terminus end would imply some unique associations with other proteins to ensure proper anchoring to the merozoite surface during invasion. Interestingly, a micronemal protein PfRipr (P. falciparum RH5 interacting proteins) was found to interact with PfRH5 (Chen et al., 2011) suggesting that they complex after their release from their respective organelles. PfRH5 is cleaved to a 45 kDa protein which is commonly detected in all the parasite culture supernatants (Baum et al., 2009) although a 28 kDa fragment was also detected (Hayton et al., 2008).

Analysis of a genetic cross have identified sequence differences in PfRH5 as playing an important role in allowing P. falciparum clones to invade and bind to receptors on red blood cells of both humans and monkeys (Hayton et al., 2008). Recently, basigin was identified as the receptor for PfRH5 (Crosnier et al., 2011) and it is now interesting to establish how changes in PfRH5 enable it to recognizes a possible basigin variant on monkey erythrocytes. Importantly, antibodies against the full-length PfRH5 are found to effectively inhibit merozoite invasion making this protein an attractive target for invasion inhibition antibodies (Douglas et al., 2011).

Plasmodium reichenowi and other simian parasites RBLs

Five PfRHs homologues were identified in the closely related chimpanzee parasite Plasmodium reichenowi. Although, the overall sequence homology between P. reichenowi normocyte-binding protein 1 (PrNBP1) and PfRH1 is around 80%, it is a pseudogene due to the presence of 224 premature termination codons in exon 2 of PrNBP1 (Rayner et al., 2004). The P. reichenowi homologue of PfRH2a has a different gene structure and is conserved only at the C-terminal region. Unlike, PfRH3, its homologue in P. reichenowi, PrNBP3 encodes a complete protein and both PfRH2b and PfRH4 are highly conserved. The smallest member of PfRH family, PfRH5, which has no orthologue in any other human, monkey and rodent malarial parasites, is conserved in P. reichenowi (Hayton et al., 2008).

The P. vivax (PvRBP1 and PvRBP2) orthologues are also found in another simian malaria parasite Plasmodium cynomolgi and the recently published genome sequence found multiple RBL genes (eight members) of which some are truncated or pseudogenes (Tachibana et al., 2012). RBLs are also found in Plasmodium knowlesi with the PvRBP1 orthologue in P. knowlesi (PkNBP1) being a pseudogene (Pain et al., 2008). Interestingly, two novel P. knowlesi RBLs termed PkNBPXa and PkNBPXb are expressed at the schizont stage with the expected molecular weight of 324 and 334 kDa respectively. However, antibodies against PkNBPXa and PkNBPXb recognized additional proteolytic fragments of 220 and 140 kDa products respectively. Localization studies showed that PkNBPXa and PkNBPXb are located in microneme and bind strongly to rhesus macaque erythrocytes (Meyer et al., 2009) and it will be interesting to see whether these proteins can also bind human erythrocytes. P. vivax appears to have a homologue of PkNBPXa and homologues of PkNBPXa and PkNBPXb were identified in P. cynomolgi and Plasmodium coatneyi (Meyer et al., 2009). In addition orthologue of PvRBP1 with 84% nucleotide identity has been identified in the simian malaria parasite P. coatneyi while Plasmodium fragile has an intact functional NBP1 gene. However, there is no indication of PvRBP2 in P. knowlesi and P. coatneyi parasites. Taken together, the conservation of RBL genes across different Plasmodium species highlights a conserved function while the retention of different functional members is in line with species-specific differences in the receptors recognized by these proteins.

Erythrocyte-binding domains (EBD) of PfRHs

The erythrocyte-binding properties for a number of members of the RBL have been documented for P. falciparum, P. vivax, P. knowlesi and P. yoelii (Tham et al., 2012). Further insights have been gained in the case of P. falciparum were the exact location of the EBD within different RH has been delineated. For the case of PfRH1 a ∼ 300-amino-acid region located towards the N-terminal region of the protein and characterized by a coiled coil motif which is possibly involved in protein–receptor interaction has been shown to contain the functional EBD (Gao et al., 2008). Based on the phylogenetic comparison of PfRH4 with PvRBP1 of P. vivax a 261 aa region of PfRH4 has been identified as the EBD and been shown to retain the same erythrocyte-binding property as the native protein (Gaur et al., 2007). In the case of PfRH5 a single amino acid change (I204K) has been linked to differences in the binding and invasion property of the parasite (Hayton et al., 2008), this adds another level of complexity to understand the role of RH in host cell recognition as it suggests that relatively few amino acid changes in the binding region can lead to different receptor recognition of the protein. Recently, by comparing the erythrocyte-binding region of PfRH1 and RH4, a 40 kDa region common to both PfRH2a and 2b was determined to be a functional EBD of PfRH2a/2b (Sahar et al., 2011) and this was subsequently further delineated to a 15 kDa region (Triglia et al., 2011). Interestingly, a recent study showed that different proteolytic fragments of PfRH2a possessed different binding properties indicating an EDB at both the C-terminal end and the N-terminus (Gunalan et al., 2011). In addition, using PfRH2a/2b chimeric parasites it was shown that the unique C-terminal cytoplasmic domain of PfRH2a/2b is important for differential invasion pathway utilization (Dvorin et al., 2010). This suggests that different RH may have multiple binding receptor recognition properties that are utilized at different steps of the invasion process and may explain the need for the precise processing events that are observed for all the RHs. In addition to P. falciparum, the EDB of one member of the Py235 family of P. yoelii has been identified as a 194 aa region that shows limited amino acid conservation with the erythrocyte-binding regions of PfRH1 and RH4 (Gruber et al., 2011). Low resolution structure analysis of this EBD indicated striking conservation of the overall shape of the molecule to both the DBL domains of EBL and the CIDR domain of PfEMP1 two known receptor binding regions of P. falciparum (Gruber et al., 2011). This would suggest that despite diverse sequences, receptor binding regions of Plasmodium proteins retain a conserved shape (Gruber et al., 2012). All these observations indicate the RBLs contain at least one EDB located towards the N-terminal part of the protein and that only a relatively small proportion of the overall protein is actually involved in receptor binding (Fig. 2A). While the work on PfRH2a suggests that there may be other EBD in addition to the one located at the N-terminus, the large size of the protein in relation to the binding region suggests that other functions important during the invasion process may be carried out by these proteins.

Role of ATP during invasion

One surprising potential additional role came from the detailed bioinformatics analysis of the RBL that indicated the presence of a putative nucleotide-binding domain (NBD) in all members of the RBL family. This predication was validated using biochemical and biophysical approaches to directly demonstrate the specific binding of ATP and ADP to an approximately 94 kDa region (NBD94) of Py235 (Ramalingam et al., 2008). While there was no evidence for the NBD being involved in ATP hydrolysis, ATP/ADP binding induces conformational changes in a C-terminal hinge-region of NBD94, indicating that NBD94 may act as a potential sensor for ATP/ADP. The adenine nucleotide binding segment in NBD94 is a short amino acids sequences 483FNEIKEKLKHYNFDDFVKEE502 and interestingly this sequences directly interferes with the binding of native Py235 to erythrocytes suggesting that this represents a novel new drug target. Subsequently it was shown that the NBD and the EBD are coupled via a hinge-like structure region suggesting a mechanisms in which binding of ATP leads to a conformational change within the protein that results in the exposure/activation of the binding region of the protein. This idea is supported by the findings that ATP/ADP is able to modulate the direct binding of Py235 to the erythrocyte (reviewed in Gruber et al., 2012).

Identification of a suitable erythrocyte for invasion is key for parasite survival and while specific receptor ligand interactions are key for the identification of a specific cell type these interactions may not be enough to identify the suitable ‘health’ of the target cell. Earlier studies showed that intracellular ATP is an important factor in host cell determination with levels of parasitaemia correlating strongly with intracellular ATP levels (Eaton and Brewer, 1969). Furthermore, intracellular ATP has been shown to be essential for invasion. Within the erythrocyte, intracellular ATP is used to phosphorylate membrane-associated proteins directly impacting on the arrangement of the cytoskeleton and thereby affecting the deformability of erythrocytes. Mechanical stress induces ATP release from erythrocyte and recent studies on purinergic signalling in Plasmodium have provided evidences of the importance of extracellular ATP, which is likely to be released from the erythrocyte during invasion (Levano-Garcia et al., 2010; Cruz et al., 2012). Importantly, it suggests that ATP released from erythrocytes forms a gradient at the erythrocyte membrane that is sensed by the RBLs. This strategy of sensing ATP could ensure the binding domains of RBLs are only exposed to the environment at a time when they are needed thereby providing the parasite with two advantages; protection against invasion blocking antibodies and ensuring the invasion of a healthy erythrocytes.

Role of calcium (Ca2+) signalling during invasion

Invasion is preceded by high-affinity apical attachment to the host cell surface, with rhoptry and microneme proteins mediating many of these interactions. However, the external signals and signalling mechanisms responsible for secretion of these invasion proteins are not fully understood. Calcium is a second messenger in cell signalling and Ca2+ homeostasis is critical to the function of eukaryotic cells controlling processes such as secretion, transcription and cell division (Berridge et al., 2003). Ca2+ is essential for the normal growth, regulation of cell cycle and long-term survival of P. falciparum and intracellular Ca2+ is also important for erythrocyte invasion in P. falciparum (Wasserman et al., 1982). The ability to isolate viable merozoites has made it possible to study signalling events during invasion and it has been shown that intracellular Ca2+ regulates the release of microneme and rhoptry proteins such as PfEBLs and RHs (Singh et al., 2010). Exposure of free merozoites to low K+ ion concentrations, such as seen in the bloodstream, provides the external signal that leads to a rise in cytosolic Ca2+ through a phospholipase C (PLC) pathway. This triggers the translocation of microneme proteins such as EBA175 and AMA1 to the merozoite surface. Subsequently, the interaction of EBA175 with glycophorin A, restores basal cytosolic Ca2+ levels and triggers release of rhoptry proteins such as CLAG3.1 and PfRH2b. Similarly, in Toxoplasma gondi the discharge of microneme proteins is triggered by an increase of intracellular Ca2+ leading to the release of proteins such as MIC and AMA1 (Carruthers et al., 1999). Interestingly, DOC2 a protein playing a role in regulating Ca2+-dependent exocytosis has been implicated in microneme secretion in both T. gondi and P. falciparum (Farrell et al., 2012) providing some mechanistic insights on the Ca2+ induced secretion of AMA1 and EBA175. A recent study shows that purinergic signalling triggers cytosolic Ca2+ rise and MSP1 proteolysis during invasion in Plasmodium berghei and P. yoelii parasites potentially establishing a link between extracellular ATP and Ca2+ signalling (Cruz et al., 2012). This leads to an attractive hypothesis that RBLs as putative nucleotide sensors are involved in regulating the release of cytosolic Ca2+ thereby triggering microneme discharge and junction formation. Further studies in this direction will open up a path for the analysis of signalling pathways involved in regulated secretion of apical organelles during invasion.

Conclusion and future perspectives

Our understanding about the function of RBLs in host pathogen interaction has grown significantly over the last decade. The erythrocyte-binding characteristics of various RBLs has been determined and receptors for a number of RBLs have been elucidated. Expanding this information to fully understand the interactions between all the RBL and their cognate erythrocyte receptors is now feasible. Structural characterization of the receptor–EBD complex along with detailed mapping of invasion blocking antibodies could lead to new strategies for an epitope vaccine that combines crucial peptide sequences of multiple RBL into a single synthetic vaccine. In addition structural information will be of immense help in developing new strategies, like small molecule inhibitors, to target receptor ligand interactions during parasite invasion. The findings that proteolytic processing of RBLs is tightly regulated and can change the erythrocyte-binding ability of the product not only suggests an additional complexity of RBL–receptor interactions that needs to be investigated further but also provides us with another putative target for blood-stage intervention.

The finding that RBLs have a potential role as sensing molecules that could trigger the release of other invasion proteins provides a new avenue to explore. The link of RBL to ATP sensing as well as in Ca2+ signalling not only open up a whole new area of investigation but also provides an opportunity to explore how all these processes are linked together to enable the merozoite to successfully invade a RBC. In particular the conformational changes in RBL induced by ATP binding could be seen as a mechanism to bring the erythrocyte into closer proximity to the merozoite thus aiding the binding of other parasite ligand to their cognate receptors. Alternatively, ATP might provide the trigger that is needed to expose the binding domain of an RBL, thereby enabling binding to the specific receptor. Such a mechanism would provide a novel form of immune evasion. Finally the sensing of ATP by RBLs along with specific receptor binding might trigger the release of intracellular Ca2+ that leads to the release of other invasion proteins. Interestingly, the combination of the dual signal of ATP sensing along with receptor binding provides a safeguard that controls Ca2+ signalling and thereby limits premature secretion of invasion proteins. To resolve all these questions requires novel high speed real-time imaging approaches as interactions between RBL and ATP as well as Ca2+ release are most likely transient and occur only within a relatively short time window during invasion. The fact that it is now feasible to produce viable merozoites is a key breakthrough that will enable the community to move into this direction particularly if it is possible to develop unique inhibitors in the form of monoclonal antibodies or small molecules that target RBL and arrest invasion at a precise step. All these studies will provide a new dimension in understanding the potential role of RBLs in successful invasion of the host erythrocyte by the parasite as well as provide new and novel ways of targeting invasion with new therapeutic approaches.

Acknowledgements

This work was supported by a grant from the Biomedical Research Council (Singapore).

Ancillary