table of contents table of contents

This review focus on basic aspects, from a cell biology perspective, of …

Home » Biology Articles » Protistology » Protozoology » Secretory organelles of pathogenic protozoa » Cell secretion in apicomplexa

Cell secretion in apicomplexa
- Secretory organelles of pathogenic protozoa

Since the first studies on the fine structure of protozoa belonging to the Apicomplexa group, especially studies carried out with T. gondii, Eimeria and Plasmodium, it became clear that the anterior region of the infective forms (trophozoites, merozoites, sporozoites) was highly specialized, forming what is generally known as the apical complex (Fig. 1). In addition to cytoskeleton components, such as the conoid-sub-pellicular microtubules complex, two organelles were initially recognized and designated as micronemes and rhoptries. Later on another structure, which can be found in other regions of the protozoan body, was recognized and designated as dense granules. 

Apicomplexan protozoa present a typical endoplasmic reticulum with associated ribosomes. In T. gondii the ER is distributed throughout the cell and the nuclear envelope itself provides a significant fraction of the ER. Vesicles with a fuzzy coat bud off from the nuclear envelope and the ER. In Plasmodium the ER is less well developed. Several cisternae of the Golgi complex are observed in the anterior region, just above the nucleus of T. gondii. Vesicles with a clathrin-like coat bud off from the trans portion of the Golgi complex. Proteins such as COP I and II, Arf 1 and Sar 1, involved in vesicle formation at the ER-Golgi complex system, have been found in T. gondii (Review in Joiner and Roos 2002, Ngô et al. 2000).


Micronemes are small, cigar-shaped organelles that are restricted to the apical third of the protozoan body. Their number varies according to the species and the developmental stages. In some species are hardly seen while in others are so numerous that correspond to the most abundant organelle found in the cell (Fig. 2). The organelle is surrounded by a typical unit membrane and presents an electron dense matrix due to its large protein content. Indeed the organelle is intensely stained when the protozoa are submitted to the ethanolic phosphotungstic acid technique, which reveals basic proteins (De Souza and Souto-Padrón 1978). At present, we still do not have a clear explanation for this labeling pattern since the known micronemal proteins have isolectric points lower than 7.0. All proteins found in the micronemes are synthesized with an N-terminal signal sequence that mediates their entrance into the secretory pathway by translocation across the endoplasmic reticulum membrane. A mutagenesis analysis of the C-terminal portion of MIC2 showed the presence of two conserved amino acid motifs mediating the target of this protein to the micronemes (Di Cristina et al. 2000). One motif is a tyrosine-based signal and the other one consists of a stretch of acidic residues. 

Many of the micronemal proteins are glycosylated as seen by labeling of the micronemes when thin sections of T. gondii are incubated in the presence of gold-labeled lectins (Carvalho et al. 1991).

It has been shown that when the infective forms of Apicomplexan parasites touch the host cell surface they trigger a process of Ca2+ release and the discharge of the content of the micronemes at the junction between the parasite and the host cell (Carruthers et al. 1999a, Bouchot et al. 1999, Vieira and Moreno 2000) which then mediates parasite attachment (Carruthers et al. 1999a, b, Carruthers and Sibley 1999). This process takes place in a few seconds and the released proteins are not incorporated together with the parasites but instead are capped and released from the posterior end of the protozoan (Carruthers et al. 1999a). During redistribution on the parasite surface, transmembrane MICs are thought to connect external receptors to the submembranous acto-myosin motor that provides the power for parasite motility. Chelating of parasite intracellular Ca2+ inhibited both microneme release and invasion of host cells. What is the origin of the Ca2+ used by the protozoan? There is enough data indicating that the Ca2+ used for protozoan motility, micronemal secretion and cell invasion comes from organelles found in the protozoan. Fluorescence microscopy of protozoa labeled with the calcium indicator fluo-4 showed that cytosolic calcium levels underwent dramatic and rapid fluxes (Lovett and Sibley 2003). Several protozoan organelles, such as the endoplasmic reticulum, the Golgi complex, the mitochondrion and the acidocalcisomes may store calcium (Moreno and Zhong 1996, Review in Arrizabalaga and Boothroyd 2004). It has been shown that there is an intracellular calcium release channel with properties of the inositol 1,4,5-triphosphate/ryanodine receptor superfamily (Lovett et al. 2002).

It has been shown that isoforms of phosphoglucomutase, a cytosolic enzyme, are implicated in Ca2+-mediated signaling events. One isoform, known as parafusin, plays an important role during exocytic activity in ciliated protozoa (Zhao and Satir 1998). A protein called parafusin-related protein has been identified in T. gondii and showed tobe localized to an apical subpopulation of miconemes and to be redistributed during invasion of the host cell by the protozoan (Matthiesen et al. 2001a, b).

Several micronemal proteins contain one or more adhesive motifs found in mammalian proteins such as Epidermal Growth Factor, integrins, thrombospondin and kallikrein. Four of them, all designated as MIC (MIC1-4, from micronemal proteins), have been studied in some detail.

MIC1 has a size of 60 kDa and contains two degenerate repeats similar to an adhesive sequence found in thrombospodin and known as type I repeats (TSP-I), and is able to bind to the host cell surface (Fourmaux et al. 1996). It has been shown that it is a lactose-binding lectin (Lourenço et al. 2001). It was also shown that its N-terminal portion functions as an independent adhesin and promotes association with TgMIC4. A galectin-like domain interacts and stabilizes TgMIC6 (Saouros et al. 2005). Single deletion of MIC1 gene decreased invasion of fibroblasts and slightly reduced virulence of the parasite to mouse (Cérède et al. 2005).

MIC2 has a size of 115 kDa and comprises an adhesive N-terminal integrin-like A domain that has been implicated in binding ICAM 1 (Barragan et al. 2005) and glycosaminoglycans (Harper et al. 2004), which are ubiquitous sulfated proteoglycans found in the extracellular matrix. MIC2 also displays six thrombospodin type-1 repeats that also have the potential to bind glycosaminoglycans (Wan et al. 1997). It has been shown that a second micronemal protein, known as M2AP, facilitates the transport of MIC2 through the secretory pathway (Huynh et al. 2003, Rabenau et al. 2001). These two proteins seem to form stable hexamers consisting of three ab dimmers (Jewett and Sibley 2004).

Following discharge MIC2 is proteoliticallycleaved by proteases with release of its ectodomain from the parasite surface, a process that seems to be involved on parasite invasion (Carruthers et al. 2000, Brossier et al. 2003). Recently, proteins belonging to the rhomboid family of intramembrane-cleaving serine proteases, designated as TgROMs, were detected (Brossier et al. 2005, Dowse et al. 2005).

The binding of MIC2 to the host cell surface may establish connection between the host cell surface receptor and the cytoskeleton machinery of the parasites activating the gliding process necessary for the penetration of the parasite into the host cell.It has been shown that Plasmodium sporozoites, which do not express a homologue of MIC2, known as TRAP, fail to glide or invade host cells (Sultan et al. 1997).

MIC3 has a size of 90 kDa and possessesfive partially overlapping epidermal growth factor (EGF) domains and an NH2 terminal chitin binding-like domain, which probably are also involved in the process of parasite association to the host cell surface (Fourmaux et al. 1996, Soldati et al. 2001). It is a disulfide-linked heterodimer comprised of two 38 kDa isoforms (Achbarou et al. 1991). Single substitution of two critical amino acids in the chitin binding-like domains of this protein abolished its binding to cells and decrease parasite virulence (Cérède et al. 2005).

MIC 4 contains six domains and it binds to host cells (Brecht et al. 2001). In the case of Eimeria tenella MIC4 is a transmembrane protein with a molecular weight of 240 kDa, containing 31 tandemly arranged EGF-like repeats in the extracellular domain. These repeats have calcium binding consensus that seem to be involved in make the molecule to adopt a protease-resistant, rigid structure that could favor its interaction with host cell ligands (Periz et al. 2005).

MIC 6, which is a transmembrane protein, forms trimeric complexes with soluble micronemal proteins such as MIC1 and MIC 4, functioning as an escort protein. Deletion of MIC 6 prevents targeting of these two proteins to the microneme (Reiss et al. 2001). MIC 8 also seems to work as a escort protein to MIC 3 (Meissner et al. 2002).

In Plasmodium falciparum a micronemal protein, known as EBA 175, has been shown to bindto sialic acid (Sim 1995) a molecule, which plays a fundamental role on the process of parasite-erythrocyte interaction.

More recently several other proteins which do not present recognizable adhesive motifs have been identified in the micronemes of T. gondii. TgMIC 5 has homology to the parvulin family of peptidyl prolyl cis-trans isomerases and may assist in the folding of other micronemal proteins (Brydges et al. 2000). Other proteins, including Tg MIC10 entirely devoid of cysteines, were recently identified (Hoff et al. 2001). MIC 11, a 16 kDa protein, was recently identified in several coccidian parasites. During its traffic through the secretory pathway it is proteoliticaly cleaved with removal of an internal propeptide, resulting in a mature form containing a a-chain and a b-chain tethered by a single disulfide bond (Harper et al. 2004).

Members of the genus Plasmodium present several developmental stages, which are able to invade different cells in both vertebrate and invertebrate cells. For instance, merozoites invade vertebrate red blood cells, the ookinete invades epithelial cells of the insect vector while sporozoites invade the salivary gland epithelial cells of the insect and when inoculated into the vertebrate host, traverse several cellular barriers until invade hepatic cells. Recent studies have shown that Plasmodium presents several genes coding for proteins, which present a membrane-attack complex/perforin-like domain. One of these proteins (PLP1/SPECT2) found in sporozoites was shown to be localized in the micronemes (Kaiser et al. 2004) and is necessary for cell traversal (Ishino et al. 2005). Genes encoding proteins with similar properties were identified in the genome of Eimeria and Toxoplasma.


Rhoptries are long, club-shaped organelles connected by thin necks to the extreme apical pole of the parasite (Figs. 1-4). At their basal portion the matrix of the organelle shows a spongeous appearance while the neck region is uniformly electron dense making it difficult the distinction from the micronemes. Their number varies according to the species. Several can be seen in T. gondii whereas only two, often designated as paired organelle, are found in Plasmodium. The organelle is surrounded by a typical unit membrane and is heavily stained with ethanolic phosphotungstic acid (Fig. 1) (De Souza and Souto-Padrón 1978). Cytochemical studies have shown the presence of glycconjugates and lectin-like molecules (Carvalho et al. 1991) and Ca2+ (Pezzela et al. 1997) in the rhoptries. Allproteins found in the rhoptries were synthesized in the endoplasmic reticulum and passed through the Golgi complex. Members of the ROP2 family contain multiple independent targeting signals (Bradley and Boothroyd 2001, Striepen et al. 2001). ROP 2 displays both YXXF and LL motifs. Deletion or alteration of these motifs abolishes protein delivery to the rhoptries, with its accumulation in a post Golgi compartment (Review in Joiner and Roos 2002). There is some evidence that rhoptry proteins after leaving the Golgi complex are first accumulated in an intermediate compartment, a type of immature rhoptry which is a acidic organelle, with a pH of 3.5 to 5.5 whereas the mature organelle has a pH of 5.0 to 7.0 as determined using the DAMP technique (Metsis et al. 1995, Shaw et al. 1998). Based on this fact and the involvement of both secretory and endocytic pathways in the rhoptry formation it has been considered that this organelle is equivalent to a lysosome (Metsis et al. 1995, Ngô et al. 2004). More recently, a novel Na+/H+ exchanger, designated as TgNEH2, was localized in the rhoptry of T. gondii (Karasov et al. 2005). This protein may be involved in rhoptry pH regulation. This organelle presents as a characteristic feature the fact that when its lipids and proteins are exocytosed through the duct the organellar membrane is retained and an empty organelle, which can be easily identified by electron microscopy, remains.

Immunocytochemistry provided evidence that the rhoptry content is not homogeneous. For instance, some proteins are localized in the basal region of the organelle whereas others are located inits apical portion (Review in Blackman and Bannister 2001). 

Secretion of rhoptry proteins takes place immediately after adhesion of the parasites to the host cell surface. In the case of T. gondii kinetic studies showed that release of the proteins is completed in about 1 minute and that the proteins are internalized and will make part of the membrane lining the parasitophorous vacuole (Sam-Yellowe et al. 1988, Saffer et al. 1992, Carruthers and Sibley 1997). Several rhoptry proteins have been identified and characterized. ROP1 has a size of 60 kDa and exhibits an extreme charge asymmetry with a highly acidic N-terminal domain and a basic C-terminal domain suggesting that it participates in protein-protein interactions (Ossorio et al. 1992). It has been suggested (Schwartzman 1986) that ROP1 corresponds to the penetration enhancing factor previously characterized in T. gondii (Lycke et al. 1968) and for some time it was considered as a key molecule for the penetration of host cells by this parasite. However, it was shown that ROP1 null mutants invade host cells normally (Kim and Boothroyd 1993). ROP2, which has a size of 54 kDa, ROP3, ROP4 and ROP8 are antigenically cross-reactive and may have overlapping functions. ROPs 2 and 8 have been characterized in some detail and cDNA sequences have shown that although they are highly homologous with one another they share no significant homology with any other proteins in the databases (Beckers et al. 1994, 1997). Both possess single putative transmembrane segments 75 amino acids from their respective C-termini. OP2 is secreted during the process of penetration of T. gondii into the host cell and is found in association with the membrane lining the parasitophorous vacuole. Its N-terminal domain is exposed on the cytoplasmic face of the parasitophorous vacuole and may be involved in the association of the vacuoles with cytoplasmic organelles of the host cell such as the endoplasmic reticulum and mitochondria (Sinai et al. 1997, Sinai and Joiner 2001). Little is known about the other ROP proteins, which have been identified using monoclonal antibodies (Leriche and Dubremetz 1991). A novel rhoptry protein, designated as BRP1, was recently identified in nascent organelles found during the first division of bradizoites, but not in tachyzoites (Schwarz et al. 2005).

The rhoptries of T. gondii have been isolatedby subcellular fractionation procedures and biochemical analysis showed a lipid to protein ratio of 0.26, thus indicating their richness in proteins. The cholesterol to phospholipid ratio was 1.48. Phosphatidylcholine was the major phospholipid (Foussard et al. 1991). Proteomic analysis using mass spectrometry of the fraction identified 38 novel proteins. At least 11 of them were localized in the rhoptries, as shown by immunofluorescence microscopy. Some are localized in the bulbous basal portion of the organelle while others are restricted to the neck portion, an observation which points to the existence of different domains in the rhoptries. In addition other proteins such as toxofilin, Rab 11, kinases and phosphatases were also found in the rhoptry (Bradley et al. 2006).

The rhoptries isolated from P. falciparum merozoites showed a large number of proteins(Etzion et al. 1991). A large number of rhoptry proteins have been identified in Babesia, Plasmodium and Eimeria (Review in Sam-Yellowe 1999). Many of them present as a special feature the ability to bind to erythrocytes.


The dense granules are spherical organelles distributed throughout the cell rather than localized at the apical complex, with a mean diameter of 0.2mm (Fig. 4). Its matrix is uniformly electron dense due to the high concentration of protein. Kinetic studies have shown that secretion of the dense granule content takes place after parasite invasion and localization within the parasitophorous vacuole persisting for several minutes (Carruthers and Sibley 1997). In contrast to secretion of micronemes and rhoptries, which takes place in the apical region, dense granule secretion occurs at the lateral regions of the protozoan. The secreted proteins associate with the membrane of the parasitophorous vacuole and with the parasite derived intravacuolar membranous network. Proteins are delivered to the dense granule by the bulk flow pathway (Coppens et al. 1999). Proteins from which specific targeting signals for other organelles have been deleted are localized in the dense granules (Striepen et al. 2001, Reiss et al. 2001).

Several proteins have been identified in the dense granules. GRA1 has a size of 22-27 kDa is relatively abundant and remains either soluble within the vacuole or loosely associated to the membranous network. It binds Ca2+ with two EF hand motifs and therefore has been considered to be involved in homeostasis of this ion within the vacuole (Cesbron-Delauw et al. 1989). GRA2 has a size of 28.5 kDa and after secretion is tightly associated with the membranous network (Mercier et al. 1993, 1998a) through two amphiphatic alpha-helical domains (Mercier et al. 1998a, Sibley et al. 1995). The expression of an HA9 epitope-tagged form of GRA2 by stable transformation of T. gondii showed that it is correctly packaged secreted and targeted. Expression of deletion mutants lacking either of two amphiphatic alpha helices resulted in the production and secretion of proteins, which did not associate in a stable way to the membranous network (Mercier et al. 1998b). Based on the observation that a GRA2 null mutant of T. gondii presents an attenuated virulence to mice it has been suggested that this protein plays some role in the virulence of the parasite (Mercier et al. 1998b). GRA3 has a size of 30 kDa and forms multimeric complexes that associate with the membrane lining the vacuole through hydrophobic interactions (Ossorio et al. 1994), although there is no predicted membrane-spanning domain for it. However, it was recently shown that GRA3 is actually an artificial chimera of 2 proteins. One, with a molecular weight of 65 kDa, shares the C-terminus of GRA3 and the other, with a predicted molecular weight of 24 kDa, shares the N-terminal region and is recognized by antibodies previously shown to label the dense granules. The corrected GRA3 has a N-terminal secretory signal sequence and a transmembrane domain, which explains its insertion into the membrane lining the parasitophorous vacuole (Henriquez et al. 2005).

GRA 4, 5, 6, 7 and 8 each have one putative transmembrane segment and it has been suggested that they may constitute the molecular sieve that allows the passage of molecules smaller than 1900 Da across the vacuolar membrane (Schwab et al. 1994). These proteins have a size in the range of 21 to 32 kDa. The level of GRA 7 expression is lower in a less virulent strain of T. gondii. An additional form of GRA 7 with reduced mobility, probably due to some modification of the protein after exocytosis, was detected on the surface of intact host cells (Neudeck et al. 2002). How this protein reaches the host cell surface is not yet clear. More recently a new protein, designated as GRA 9, was identified. It is a 41 kDa protein, which associates with the tubular network found within the PV (Adjogble et al. 2004). In addition to the GRA series the dense granules also contain two closely related isoenzymes of the nucleotide triphosphatase NTPase which are involved in breaking down host-supplied di- and triphosphate nucleotides participating in the purine salvage pathway, for which T. gondii is auxotrophic (Asai et al. 1983, 1995, Sibley et al. 1994) and in the initiation of the releaseof the parasites from the vacuole (Stommel et al. 1997, Silverman et al. 1998) in a process which involves depletion of ATP and increase in Ca2+ concentration (Stommel et al. 1997, Silverman et al. 1998). Two protease inhibitors have also been identified in the dense granules (Morris et al. 2002, Pszenny et al. 2002).


rating: 3.06 from 16 votes | updated on: 29 Apr 2007 | views: 22858 |

Rate article: