Identification of the export signal and prediction of exported proteins

A major advance in our understanding of the protein trafficking mechanisms of malaria parasites was the identification of an export signal sequence, and the use of this to predict which proteins the parasite exports (Hiller et al. Reference Hiller, Bhattacharjee, van Ooij, Liolios, Harrison, Lopez-Estrano and Haldar2004; Marti et al. Reference Marti, Good, Rug, Knuepfer and Cowman2004). These studies revealed a consensus conserved amino acid motif downstream of the N-terminal sequences of many exported proteins, which was termed the Plasmodium export element (PEXEL) or the vacuole transport signal (VTS) (Hiller et al. Reference Hiller, Bhattacharjee, van Ooij, Liolios, Harrison, Lopez-Estrano and Haldar2004; Marti et al. Reference Marti, Good, Rug, Knuepfer and Cowman2004), which is conserved structurally and functionally across the Plasmodium species infecting humans and birds (Marti et al. Reference Marti, Good, Rug, Knuepfer and Cowman2004). The PEXEL/VTS motif generally lies within a 20–60 amino-acid stretch downstream of the N-terminal sequence of many exported proteins (Hiller et al. Reference Hiller, Bhattacharjee, van Ooij, Liolios, Harrison, Lopez-Estrano and Haldar2004; Marti et al. Reference Marti, Good, Rug, Knuepfer and Cowman2004; Sargeant et al. Reference Sargeant, Marti, Caler, Carlton, Simpson, Speed and Cowman2006) (Fig. 2). It consists of five amino acids RxLxE/Q/G, of which arginine (R) and lysine (L) are essential for recognition whereas the fifth amino acid glutamate (E) is not essential and can be replaced with glutamine (Q) (frequently) or glycine (G) (rarely) (Hiller et al. Reference Hiller, Bhattacharjee, van Ooij, Liolios, Harrison, Lopez-Estrano and Haldar2004; Marti et al. Reference Marti, Good, Rug, Knuepfer and Cowman2004) (x could be any amino acid). Further, the PEXEL processed proteins get acetylated at the new N-terminus, which appears to be crucial for recognition and transport of exported proteins to red cells (Chang et al. Reference Chang, Falick, Carlton, Sedat, DeRisi and Marletta2008; Boddey et al. Reference Boddey, Moritz, Simpson and Cowman2009). A recent study has identified that many exported proteins can have a relaxed PEXEL motif (RxLxxE), which is functional and processed by the same mechanism used for canonical PEXEL (Boddey et al. Reference Boddey, Carvalho, Hodder, Sargeant, Sleebs, Marapana, Lopaticki, Nebl and Cowman2013). Based on the updated information of the relaxed PEXEL motif, ExportPred version 1.0 (Sargeant et al. Reference Sargeant, Marti, Caler, Carlton, Simpson, Speed and Cowman2006) has been re-designed (version 2.0), which identified 73 additional exported proteins from the falciparum genome (Boddey et al. Reference Boddey, Carvalho, Hodder, Sargeant, Sleebs, Marapana, Lopaticki, Nebl and Cowman2013).

Fig. 2.

Diagrammatic representation of P. falciparum PEXEL and PNEP protein structure. The figure represents a typical structure of an exported protein. The arrow indicates ER resident enzyme Plasmepsin-V that cleaves the PEXEL motif. Scenarios of export of PEXEL-negative exported proteins (PNEPs) are also presented. The bar above the colour indicates hydrophobic regions potentially involved in export and the figure is a modified version of Spielmann and Gilberger (Reference Spielmann and Gilberger2010).

The PEXEL/VTS motif does not appear to be essential for a protein to enter the export pathway, however, as several exported proteins do not possess such a motif. The exported proteins targeted to the host-cell cytoplasm that lack a well-defined PEXEL/VTS motif are denoted as PEXEL negative export proteins (PNEPs). These PNEPs include skeleton binding protein-1 (SBP-1), membrane-associated histidine-rich protein-1 (MAHRP-1), and ring export protein-1 and -2 (REX-1 and REX-2) (Blisnick et al. Reference Blisnick, Morales Betoulle, Barale, Uzureau, Berry, Desroses, Fujioka, Mattei and Braun Breton2000; Spycher et al. Reference Spycher, Klonis, Spielmann, Kump, Steiger, Tilley and Beck2003; Hawthorne et al. Reference Hawthorne, Trenholme, Skinner-Adams, Spielmann, Fischer, Dixon, Ortega, Anderson, Kemp and Gardiner2004; Spielmann et al. Reference Spielmann, Hawthorne, Dixon, Hannemann, Klotz, Kemp, Klonis, Tilley, Trenholme and Gardiner2006; Haase et al. Reference Haase, Herrmann, Gruring, Heiber, Jansen, Langer, Treeck, Cabrera, Bruns, Struck, Kono, Engelberg, Ruch, Stunnenberg, Gilberger and Spielmann2009; Saridaki et al. Reference Saridaki, Frohlich, Braun-Breton and Lanzer2009), which are all resident proteins of MCs. The PNEPs (REX-2, SBP-1 and MAHRP-1) contain a conserved transmembrane domain but lack a signal sequence, whereas REX-1 has a recessed signal sequence (Dixon et al. Reference Dixon, Hawthorne, Spielmann, Anderson, Trenholme and Gardiner2008). These PNEPs have ER intermediates, suggesting export via a classical secretory pathway (Spycher et al. Reference Spycher, Rug, Klonis, Ferguson, Cowman, Beck and Tilley2006; Dixon et al. Reference Dixon, Hawthorne, Spielmann, Anderson, Trenholme and Gardiner2008; Saridaki et al. Reference Saridaki, Sanchez, Pfahler and Lanzer2008; Haase et al. Reference Haase, Herrmann, Gruring, Heiber, Jansen, Langer, Treeck, Cabrera, Bruns, Struck, Kono, Engelberg, Ruch, Stunnenberg, Gilberger and Spielmann2009). It is very likely that processing of a signal peptide could also generate a similar N-terminus of PNEPs as generated by Plasmepsin-V; therefore, it has been proposed (Spielmann and Gilberger, Reference Spielmann and Gilberger2010) that following processing by a signal peptidase, the trafficking of PNEPs may converge with PEXEL proteins at the translocon. The functional evaluation of Plasmodium export signals in Plasmodium berghei suggests that there may be multiple pathways of protein export for PEXEL and PNEP proteins in non-falciparum malaria parasites (Sijwali and Rosenthal, Reference Sijwali and Rosenthal2010). Based on the quantity of PEXEL/VTS positive proteins, it is apparent that the majority of proteins exported by P. falciparum are PEXEL/VTS positive, whereas in non-falciparum Plasmodium species, it is the PEXEL/VTS-independent proteins that make up the majority of the exportome (Table 1). The most well-characterized example of PEXEL/VTS-independent export in a non-P. falciparum malaria parasite is the export of the variant surface proteins of P. vivax, Plasmodium cynomolgi, Plasmodium knowlesi and Plasmodium yoelii/P. berghei/Plasmodium chabaudi (Table 1). Only a few members of the above variant gene families (Table 1) possess a canonical PEXEL/VTS motif near their signal sequences. Plasmodium vivax VIR proteins provide direct evidence for their export at the surface of infected red blood cells (iRBC; Bernabeu et al. Reference Bernabeu, Lopez, Ferrer, Martin-Jaular, Razaname, Corradin, Maier, Del Portillo and Fernandez-Becerra2012; Lopez et al. Reference Lopez, Bernabeu, Fernandez-Becerra and del Portillo2013). Indirect evidence for export of VIR proteins can be inferred from the structural similarity of VIR subfamilies A and D (Merino et al. Reference Merino, Fernandez-Becerra, Durham, Ferreira, Tumilasci, d'Arc-Neves, da Silva-Nunes, Ferreira, Wickramarachchi, Udagama-Randeniya, Handunnetti and Del Portillo2006) with the exported proteins of P. falciparum SURFIN and two transmembrane (Pf2TM) proteins respectively that lack a canonical PEXEL (Sam-Yellowe et al. Reference Sam-Yellowe, Florens, Johnson, Wang, Drazba, Le Roch, Zhou, Batalov, Carucci, Winzeler and Yates2004; Winter et al. Reference Winter, Kawai, Haeggstrom, Kaneko, von Euler, Kawazu, Palm, Fernandez and Wahlgren2005; Alexandre et al. Reference Alexandre, Yahata, Kawai, Torii and Kaneko2011). This suggests that the non-falciparum Plasmodium species may have different mechanisms for the export of proteins necessary for host-cell remodelling and virulence, and corroborates a hypothesis that canonical PEXEL/VTS exported proteins have undergone lineage-specific expansion in the falciparum parasite (Pick et al. Reference Pick, Ebersberger, Spielmann, Bruchhaus and Burmester2011). Further, it is unclear whether the greater abundance of canonical PEXEL/VTS proteins in P. falciparum is due to the fact that P. falciparum requires a high degree of host-cell remodelling compared with non-falciparum malaria parasite species, and so exports more proteins, or to the fact that protein export in the non-falciparum species is mainly mediated through PEXEL/VTS-independent pathways (Pick et al. Reference Pick, Ebersberger, Spielmann, Bruchhaus and Burmester2011).

Table 1.

Canonical and non-canonical PEXEL motifs in exported proteins from Plasmodium species

Data obtained from the Plasmodium genome database (www.plasmodb.org). Canonical and non-canonical PEXEL motifs in the most likely exported proteins were estimated using the sequence pattern search tool (http://www-archbac.u-psud.fr/genomics/patternSearch.html). Number indicates number of paralogs in a gene family. Values in parentheses indicate the actual number of gene family members having a defined canonical or non-canonical PEXEL motif.

^a Canonical PEXEL cleaved by Plasmepsin-V.

^b Non-canonical PEXEL, which is also processed by Plasmepsin-V.

Recognition and processing of the export signal

Canonical PEXEL/VTS-mediated protein export requires the recognition and processing of the export signal in the ER, the first step of entry of exported proteins into the trafficking/secretory pathway. The canonical PEXEL/VTS motif was initially thought to mediate the trafficking of exported protein across the PV; however, it was later identified as a cleavage site for a parasite protease within the ER (Chang et al. Reference Chang, Falick, Carlton, Sedat, DeRisi and Marletta2008; Boddey et al. Reference Boddey, Moritz, Simpson and Cowman2009). The cleavage of the PEXEL/VTS motif in exported proteins directs access to the host-cell cytosol (Chang et al. Reference Chang, Falick, Carlton, Sedat, DeRisi and Marletta2008; Boddey et al. Reference Boddey, Moritz, Simpson and Cowman2009). The PEXEL/VTS cleaving protease has been identified as Plasmepsin-V, an aspartic protease localized in the lumen of ER (Klemba and Goldberg, Reference Klemba and Goldberg2005), by two independent studies conducted with P. falciparum (Boddey et al. Reference Boddey, Hodder, Gunther, Gilson, Patsiouras, Kapp, Pearce, de Koning-Ward, Simpson, Crabb and Cowman2010; Russo et al. Reference Russo, Babbitt, Muralidharan, Butler, Oksman and Goldberg2010). Plasmepsin-V recognizes the conserved PEXEL/VTS sequence (RxLxE/Q/D) and specifically cleaves between the lysine (L) and glutamate (E) residues (Fig. 2). The cleaved fragment carries an xE/Q/D at the N-terminal of the processed protein that is further N-acetylated (Chang et al. Reference Chang, Falick, Carlton, Sedat, DeRisi and Marletta2008; Boddey et al. Reference Boddey, Moritz, Simpson and Cowman2009) and then recruited into the secretory pathway (Chang et al. Reference Chang, Falick, Carlton, Sedat, DeRisi and Marletta2008). The conservation of the xE/D/Q residue in a processed protein seems essential for export but not for processing (Boddey et al. Reference Boddey, Moritz, Simpson and Cowman2009). Plasmepsin-V can process canonical PEXEL and relaxed PEXEL as well, such as RxLxxE (Boddey et al. Reference Boddey, Carvalho, Hodder, Sargeant, Sleebs, Marapana, Lopaticki, Nebl and Cowman2013), indicating that this export signal cleaver is capable of processing a larger number of exported proteins (>400). However, genome analysis of Plasmodium species reveals several fixed patterns (relaxed PEXEL) in the N-terminus of various exported proteins (Table 1). Assuming that these non-canonical PEXEL could be processed by a parasite, as these are exported proteins, we can expect that the parasite might have additional ER resident proteases other than the well-known canonical PEXEL cleaver Plasmepsin-V.

Physical nature of exported proteins in the PV

Shared physical characteristics of exported proteins that traverse the parasite membrane (PM), PV and PVM may offer clues to aid the elucidation of the export/secretory pathway. It has been shown that exported proteins traverse the PV in an unfolded form (Gehde et al. Reference Gehde, Hinrichs, Montilla, Charpian, Lingelbach and Przyborski2009) (Fig. 3). The study was performed with a transgenic parasite expressing a chimeric protein in which a functional export signal is fused to green fluorescent protein (GFP) and to murine dihydrofolate reductase (mDHFR). In the presence of WR99210, all chimeric GFP–mDHFR fusion proteins were found to be accumulated in the PV; on the removal of WR99210, successful entry of the chimeric protein was observed in the host-cell cytoplasm. These unfolded proteins must then be refolded in the iRBC and then trafficked to their final destination (Gehde et al. Reference Gehde, Hinrichs, Montilla, Charpian, Lingelbach and Przyborski2009). This study confirms that exported proteins traverse the PV and PVM in an unfolded form and further underlines the putative role of heat shock proteins and molecular chaperones in unfolding exported proteins for successful translocation and refolding in the iRBC. Proteome analysis of P. falciparum revealed that the PV and iRBC are enriched with a number of heat shock proteins (Florens et al. Reference Florens, Washburn, Raine, Anthony, Grainger, Haynes, Moch, Muster, Sacci, Tabb, Witney, Wolters, Wu, Gardner, Holder, Sinden, Yates and Carucci2002, Reference Florens, Liu, Wang, Yang, Schwartz, Peglar, Carucci, Yates and Wub2004; Nyalwidhe and Lingelbach, Reference Nyalwidhe and Lingelbach2006; Acharya et al. Reference Acharya, Kumar and Tatu2007). Genome sequence analysis has revealed a large repertoire of exported proteins in P. falciparum that are either heat shock proteins (Hiller et al. Reference Hiller, Bhattacharjee, van Ooij, Liolios, Harrison, Lopez-Estrano and Haldar2004; Marti et al. Reference Marti, Good, Rug, Knuepfer and Cowman2004; Nyalwidhe and Lingelbach, Reference Nyalwidhe and Lingelbach2006; Shonhai et al. Reference Shonhai, Boshoff and Blatch2007) or contain the DNA-J domain (which encodes for a molecular chaperone), for example, the seven members of the PHIST-b protein and Hsp40 families (Sargeant et al. Reference Sargeant, Marti, Caler, Carlton, Simpson, Speed and Cowman2006; Acharya et al. Reference Acharya, Kumar and Tatu2007, Reference Acharya, Chaubey, Grover and Tatu2012), and ring-infected erythrocyte surface antigen (RESA) protein (Favaloro et al. Reference Favaloro, Coppel, Corcoran, Foote, Brown, Anders and Kemp1986). These proteins emphasize the importance of chaperone molecules in malaria parasite protein export pathways, and offer a relatively unexplored avenue for exportome research.

Fig. 3.

Schematic representation of an exported protein in the PV in a P. falciparum-infected erythrocyte. Exported proteins enter PV, get unfolded and cross PVM and enter erythrocyte cytoplasm with the help of translocons. In erythrocyte cytoplasm, exported proteins first get refolded with the help of heat shock protein and then reach subcellular locations in erythrocytes. PVM, parasitophorous vacuole membrane; PM, parasite membrane. This image showing translocation of exported protein from PVM is drawn on the basis of work by Gehde et al. (Reference Gehde, Hinrichs, Montilla, Charpian, Lingelbach and Przyborski2009).

Identification of translocon machinery in the PVM

Channel-mediated delivery is a translocon. As channel-mediated protein trafficking necessitates machinery within the PVM to facilitate protein movement through the membrane, so there must exist translocon(s) embedded within the PVM (Schatz and Dobberstein, Reference Schatz and Dobberstein1996; Schnell and Hebert, Reference Schnell and Hebert2003). Such a translocon has recently been described in P. falciparum (de Koning-Ward et al. Reference de Koning-Ward, Gilson, Boddey, Rug, Smith, Papenfuss, Sanders, Lundie, Maier, Cowman and Crabb2009). It appears that the P. falciparum translocon machinery consists of a complex of parasite-encoded proteins embedded within the PVM, termed the Plasmodium translocon of export protein (PTEX). The PTEX complex is an ATP-powered machine and consists of two PTEX proteins (PTEX150 and PTEX88), a heat shock protein (Hsp101), thioredoxin protein-2 (TRX2) (Matthews et al. Reference Matthews, Kalanon, Chisholm, Sturm, Goodman, Dixon, Sanders, Nebl, Fraser, Haase, McFadden, Gilson, Crabb and de Koning-Ward2013) and parasite export protein-2 (EXP2) (de Koning-Ward et al. Reference de Koning-Ward, Gilson, Boddey, Rug, Smith, Papenfuss, Sanders, Lundie, Maier, Cowman and Crabb2009; Bullen et al. Reference Bullen, Charnaud, Kalanon, Riglar, Dekiwadia, Kangwanrangsan, Torii, Tsuboi, Baum, Ralph, Cowman, de Koning-Ward, Crabb and Gilson2012; Riglar et al. Reference Riglar, Rogers, Hanssen, Turnbull, Bullen, Charnaud, Przyborski, Gilson, Whitchurch, Crabb, Baum and Cowman2013). A recent study has shown that a homodimer of EXP2 oligomerizes, forms a pore-like shape and attaches to the remainder of the PTEX complex in PVM (Bullen et al. Reference Bullen, Charnaud, Kalanon, Riglar, Dekiwadia, Kangwanrangsan, Torii, Tsuboi, Baum, Ralph, Cowman, de Koning-Ward, Crabb and Gilson2012) (Fig. 4). Since P. falciparum exports hundreds of highly diverse remodelling and virulence-associated proteins, it is proposed that the parasite may utilize structurally related PVM translocons that may have different accessory proteins or exported proteins may use different cargo. Thus, functional dissection of translocon complexes and modelling of translocon machineries would provide further insights into protein trafficking in P. falciparum.

Fig. 4.

Molecular composition and structure of the translocon in the PVM of the P. falciparum-infected erythrocyte. PTEX150, Plasmodium translocon of exported protein-150; PTEX88, Plasmodium translocon of exported protein-88; TRX-2, thioredoxin-2; HS101, heat shock protein-101; EXP2, exported protein-2. This figure is a modified version of work by de Koning-Ward et al. (Reference de Koning-Ward, Gilson, Boddey, Rug, Smith, Papenfuss, Sanders, Lundie, Maier, Cowman and Crabb2009) and Bullen et al. (Reference Bullen, Charnaud, Kalanon, Riglar, Dekiwadia, Kangwanrangsan, Torii, Tsuboi, Baum, Ralph, Cowman, de Koning-Ward, Crabb and Gilson2012).

Trafficking of exported proteins within the erythrocyte cytoplasm

The fate of exported proteins and the mechanisms behind their journey following their exit from the PVM remains relatively poorly understood. A number of studies have suggested an important role of MCs in the sorting and further trafficking of exported proteins to the surface of infected erythrocytes (Lanzer et al. Reference Lanzer, Wickert, Krohne, Vincensini and Braun Breton2006; Wickert and Krohne, Reference Wickert and Krohne2007; Bhattacharjee et al. Reference Bhattacharjee, van Ooij, Balu, Adams and Haldar2008; Tilley et al. Reference Tilley, Sougrat, Lithgow and Hanssen2008). MCs are parasite-induced, flattened membranous structures scattered throughout the cytoplasm of infected erythrocytes (Spycher et al. Reference Spycher, Rug, Klonis, Ferguson, Cowman, Beck and Tilley2006). However, it remains to be elucidated how the sorting of proteins to the RBC membrane and the surface of the iRBC occurs following protein traversal of the PVM. The resident proteins of MCs have been shown to play an important role in the sorting and trafficking of PfEMP-1 from MCs to the erythrocyte surface. These resident proteins include Pf332, PfEMP-3, PfSBP-1 and MAHRP-I (Waterkeyn et al. Reference Waterkeyn, Wickham, Davern, Cooke, Coppel, Reeder, Culvenor, Waller and Cowman2000; Spycher et al. Reference Spycher, Klonis, Spielmann, Kump, Steiger, Tilley and Beck2003; Maier et al. Reference Maier, Rug, O'Neill, Beeson, Marti, Reeder and Cowman2007; Hodder et al. Reference Hodder, Maier, Rug, Brown, Hommel, Pantic, Puig-de-Morales-Marinkovic, Smith, Triglia, Beeson and Cowman2009). Previously, it was believed that MCs are responsible for the display of exported proteins on the surface of infected erythrocytes by a mechanism in which MCs ultimately merge with the host-cell membrane (Trelka et al. Reference Trelka, Schneider, Reeder and Taraschi2000). Later studies described the role of MCs as a secretory organelle that accumulates parasite proteins and delivers them to the surface of the erythrocyte (Bhattacharjee et al. Reference Bhattacharjee, van Ooij, Balu, Adams and Haldar2008). Further, a large-scale gene knockout study confirmed their role in the trafficking of PfEMP-1 to the surface of the infected erythrocyte (Maier et al. Reference Maier, Rug, O'Neill, Brown, Chakravorty, Szestak, Chesson, Wu, Hughes, Coppel, Newbold, Beeson, Craig, Crabb and Cowman2008). The export of PfEMP-1, one of the most well-characterized virulence-associated proteins, to the surface of the infected erythrocyte has been proposed to be completed in three phases, early, middle and late, wherein each phase involves a number of other parasite-encoded proteins assisting the trafficking through protein–protein interactions (Maier et al. Reference Maier, Rug, O'Neill, Brown, Chakravorty, Szestak, Chesson, Wu, Hughes, Coppel, Newbold, Beeson, Craig, Crabb and Cowman2008). Other parasite-encoded proteins involved in PfEMP-1 trafficking to the surface of infected erythrocytes have been reviewed by Sam-Yellowe (Reference Sam-Yellowe2009). Further investigations are required in order to fully understand the mechanisms behind the transport of proteins beyond the PVM to their final destination in the host-cell cytoplasm and on to the RBC surface.