Intracellular Calcium Channels in Protozoa

Abstract

Ca²⁺-signaling pathways and intracellular Ca²⁺ channels are present in protozoa. Ancient origin of inositol 1,4,5-trisphosphate receptors (IP₃Rs) and other intracellular channels predates the divergence of animals and fungi as evidenced by their presence in the choanoflagellate Monosiga brevicollis, the closest known relative to metazoans. The first protozoan IP₃R cloned, from the ciliate Paramecium, displays strong sequence similarity to the rat type 3 IP₃R. This ciliate has a large number of IP₃- and ryanodine(Ry)-like receptors in 6 subfamilies suggesting the evolutionary adaptation to local requirements for an expanding diversification of vesicle trafficking. IP₃Rs have also been functionally characterized in trypanosomatids, where they are essential for growth, differentiation, and establishment of infection. The presence of the mitochondrial calcium uniporter (MCU) in a number of protozoa indicates that mitochondrial regulation of Ca²⁺ signaling is also an early appearance in evolution, and contributed to the discovery of the molecular nature of this channel in mammalian cells. There is only sequence evidence for the occurrence of two-pore channels (TPCs), transient receptor potential Ca²⁺ channels (TRPCs) and intracellular mechanosensitive Ca²⁺-channels in Paramecium and in parasitic protozoa.

1. Introduction

Calcium ion (Ca²⁺) controls a variety of cellular functions in protozoa. As occurs with mammalian cells, the cytosolic Ca²⁺ concentration [Ca²⁺]_i of protozoa is maintained at very low levels (of the order of 10⁻⁷ M). The cytosolic Ca²⁺ level is responsible for the regulation of Ca²⁺-dependent and Ca²⁺-controlled proteins. Although the total calcium inside protozoan cells is much higher than 10⁻⁷ M, the bulk of this calcium is either bound to proteins, polyphosphate, membranes or other cellular constituents, or is sequestered inside intracellular organelles through the activity of pumps, channels, and exchangers, and released when needed by a variety or intracellular Ca²⁺ channels.

Recent genomic studies (King et al., 2008) have revealed that many ion channels including Ca²⁺ channels previously though to be restricted to animals, can be traced back to one of the unicellular ancestor of animals, Monosiga brevicollis, a choanoflagellate protozoan belonging the supergroup Opisthokonta, which also includes animals, and fungi. Genes encoding homologues to various types of plasma membrane Ca²⁺ channels are present: store-operated channel (Orai) and the endoplasmic reticulum sensor protein stromal interaction molecule (Stim); voltage-operated channel (similar to dihydropyridine-sensitive L-type Ca²⁺ channel); ligand-operated channels (nicotinic acetylcholine receptor and P2X purinergic receptor); transient receptor potential (TRP) channels; and second messenger-operated channel (cyclic nucleotide-gated channel) (Cai, 2008). This protozoan appears to possess all 5 modes of regulated Ca²⁺ entry across the plasma membrane identified in animals (Parekh and Putney, 2005), although their physiological validation is needed (Cai, 2008). Monosiga brevicollis has also 4 homologues of the inositol 1,4,5-trisphosphate receptor (IP₃R), and a homologue to the mitochondrial calcium uniporter (XP_001749044), but no homologues to ryanodine receptors (RyR) (Cai, 2008). However, no functional studies have been reported with any of these channels.

Evidently the evolution of eukaryotic cells is characterized by increasing genomic information that allows for increasing complexity of intracellular structure, dynamics and signaling mechanisms. Target-oriented vesicle trafficking requires not only an inventory of membrane-specific proteins, such as SNAREs (Malsam et al., 2008) and small GTPases (Zerial and McBride, 2001), but also provisions for Ca²⁺ signaling in a very local area where membranes have to interact (Neher, 1998). Ca²⁺ may come from the external medium or be locally released from stores via Ca²⁺-release channels (CRC) so that Ca²⁺ can locally drive docking, priming and eventual fusion of membranes (Rizo et al., 2006). Cell contraction is another example. Ca²⁺ is most appropriate for such functions because of its specific, reversible binding to Ca²⁺-binding proteins, CaBP, which in the end transmit the signal by a conformational change in effector protein molecules (Klee et al., 1980; Rizo et al., 2006). On the one hand global regulation of intracellular Ca²⁺ concentration, [Ca²⁺]_i, is mandatory to avoid the overall toxic effect of Ca²⁺ (Case et al., 2007). On the other hand, local [Ca²⁺]_i regulation also has to account for diffusional spread by a square function, whereas most molecular effects of Ca²⁺ depend on a higher power-function of [Ca²⁺]_i (Neher, 1998). Binding to CaBPs, sequestration into organelles and extrusion from the cell antagonize the occurrence of too high and diffuse [Ca²⁺]_i values after stimulation. Remarkably, the phenomena described in this review, as well as the CRC types mentioned, are all found already in protozoa. Nevertheless, with these cells stringent analyses of Ca²⁺ signaling and the subsiding intracellular CRCs have remained elusive until quite recently.

The protozoan organisms whose Ca²⁺ signaling and subsiding CRCs are currently investigated in our labs include ciliates (Paramecium), their close relatives, Apicomplexa (including pathogenic species of Plasmodium [malaria causing agent] and Toxoplasma) as well as some pathogenic flagellates (trypanosomatids). With these organisms, CRCs have been characterized at a molecular level, in conjunction with functional studies. There is a wide gap between evolutionary levels: ciliates close to recent forms have emerged ~800 to 850 million years ago, non-parasitic Apicomplexa ~500 million years (Douzery et al., 2004) and mammalian apicomplexan parasites ~13 million years ago (Ricklefs and Outlaw, 2010). There is also some information available on the Ca²⁺ dynamics in social amoeba of the myxomycete Dictyostelium, which clearly possesses Ca²⁺ signaling pathways (Allan and Fisher, 2009), but information about CRCs in these cells is scant.

A Paramecium tetraurelia cell is up to ~100 μm in size and exhibits distinct intracellular vesicle trafficking pathways (Allen and Fok, 2000), essentially including all those known from metazoan cells. The pathogenic forms discussed are ~10 times smaller, but also contain specific vesicle-trafficking pathways, such as endocytosis vesicles and organelles for intracellular digestion (trypanosomatids, Apicomplexa). Apicomplexa also possess secretory organelles for exocytosis. Due to their small size and their complicated lifestyle the parasites are much more difficult to study than their free-living relatives. Using fluorescent dyes in both ciliates and Apicomplexa, a considerable Ca²⁺ signal could be recorded during exocytosis of secretory organelles, such as trichocysts (Klauke and Plattner, 1997) and during motility (Lovett and Sibley, 2003), respectively.

Values for steady state [Ca²⁺]_i in widely different cells, from protozoa to mammals, are of the order of 50 to 100 nM at rest and stimulation generally causes an increase by a factor of 10 to 100 (Bootman and Berridge, 1995). This frame also applies to ciliates (Klauke and Plattner, 1997) and to parasitic protozoa (Vieira and Moreno, 2000; Moreno et al., 1994). [Ca²⁺]_i determined in Paramecium under steady state conditions yields values between 60 and 100 nM. It has to be stressed that measurements performed with fluorescent dyes, even when calibrated, systematically underestimate the real local [Ca²⁺]_i increase during activation because of its considerable local restriction. More realistic local, functionally relevant values are obtained by probing the threshold inhibitory effect of Ca²⁺ chelators with appropriate binding properties (Neher, 1995). For instance, during exocytosis stimulation [Ca²⁺]_i in the cell cortex peaked at ~400 nM with fluorescent dyes measurements, whereas chelator application during stimulation indicated the increase in [Ca²⁺]_i to the micromolar range (Klauke and Plattner, 1997).

2. Calcium stores

The paradigm of a Ca²⁺ store in all eukaryotic cells is the endoplasmic reticulum (ER), together with the sarcoplasmic reticulum (SR) in muscle cells (Berridge et al., 2000; Berridge et al., 2003; Clapham, 2007; Cai, 2008). Since Ca²⁺ is stored in many more organelles such stores and their CRCs deserve special attention also in protozoa, including ciliates and parasitic protozoa (Plattner et al., 2012).

Subsequent to stimulation Ca²⁺ is sequestered into different organelles and then may be available later on for release via CRCs in a constitutive manner or in the context of signaling processes. Ca²⁺ can, thus, regulate exocytosis, endocytosis, phagocytosis, fusion of endosomes of different stages with phagosomes, phagosome formation, membrane recycling, phago-lysosome fusion etc. (Hay, 2007; Zampese and Pizzo, 2012). In mammalian cells, many of these organelles, specifically early endosomes (Luzio et al., 2010) and lysosomes (Christensen et al., 2002), are known to store Ca²⁺ (Hay, 2007; Sherwood et al., 2007) and the membranes of many of them contain CRCs (Zampese and Pizzo, 2012). The main types of CRCs found in metazoans up to mammalian cells are IP₃R (Taylor et al., 2004; Bezprozvanny, 2005), RyR (Hamilton, 2005; Mackrill, 2012), transient receptor potential Ca²⁺ channels, TRPC (Patel and Docampo, 2009), and two pore channels, TPC, occurring mainly in acidic compartments (Galione et al., 2009; Galione et al., 2010; Patel and Docampo, 2010). All these channel types also occur in protozoa.

Work with P. tetraurelia was started with Paramecium database (DB) analysis and further evaluation by expression, localization and functional studies. Thus, a plethora of CRCs related to RyRs and to IP₃Rs, or to both, were identified (Ladenburger et al., 2006; Ladenburger et al., 2009; Ladenburger and Plattner, 2011). The different CRC types are scattered over the many sites of specific membrane interactions. The functions of IP₃Rs (Ladenburger et al., 2006) and of RyRs (more safely to be addressed as RyR-like proteins, RyR-LP) (Ladenburger et al., 2009) were investigated in more detail. Further on the analysis concentrated on cortical stores (alveolar sacs) and dense core-secretory organelle (trichocyst) exocytosis as well as on the contractile vacuole complex that serves for osmoregulation and maintaining the internal ionic balance, particularly of Ca²⁺.

In Apicomplexa secretory organelles include rhoptries, micronemes, and dense granules whose exocytosis is mandatory for attachment to a host cell, invasion and establishment of the parasitophorus vacuole. Their secretion requires Ca²⁺ signals, presumably based on IP₃ (Lovett et al., 2002; Lovett and Sibley, 2003) or cADP ribose (cADPR) (Chini et al., 2005; Nagamune et al., 2008) signaling. Database search by different groups did not allow for the identification of either IP₃Rs or of RyRs (or RyR-LPs) in Apicomplexa (Nagamune and Sibley, 2006; Plattner et al., 2012) although intracellular application of IP₃ has facilitated host cell infection (Lovett et al., 2002). In trypanosomatids IP₃Rs have recently been identified on a molecular level, and probed functionally, in two species, Trypanosoma brucei (Huang et al., 2013) and T. cruzi (Hashimoto et al., 2013). In Paramecium, the dense core-secretory organelles called trichocysts can explosively be released by exocytosis within fractions of a second, thus making this system amenable to sub-second analysis (Plattner and Hentschel, 2006). The reaction serves for warding off predators very efficiently (Harumoto and Miyake, 1991).

In summary, CRCs must have evolved early in evolution, i.e. already at the level of protozoa. These CRCs include not only IP₃Rs and RyR-LPs (Plattner and Verkhratsky, 2013) but also TRPCs and TPCs (Patel and Docampo, 2010; Plattner et al., 2012) as well as the mitochondrial calcium uniporter (Docampo and Lukes, 2012) that will be discussed later. On a speculative basis one may envisage also intracellular mechano-sensitive Ca²⁺ channels as part of an ancient CRC inventory. This expectation is based on the finding of stomatin in the membranes of the contractile vacuole complex and of food vacuoles (Reuter et al., 2013), considering that generally the scaffolding protein stomatin is structurally and functionally associated with mechano-sensitive Ca²⁺-channels in metazoans (Lapatsina et al., 2012). By sensing the internal tension in these organelles such channels may initiate a Ca²⁺ signal for the release of contractile vacuole contents by exocytosis or for fusion processes along the food vacuole pathway (Reuter et al., 2013).

Scrutiny at a molecular level, including domain analysis, intracellular localization and functional analysis, including also gene silencing, will provide us with important new insight into Ca²⁺ signaling mechanisms not only in free-living, but also in pathogenic protozoan species. Also the precise function of many of the CRCs recently identified remains to be elucidated.

3. Intracellular calcium channels in ciliates

3.1 IP₃R- and RyR-type CRCs in Paramecium – identification and localization

Only quite recently could such channels be identified, based on genomic sequences indicative of characteristic domain structures. Six subfamilies of putative CRCs, PtCRC-I to PtCRC-VI, which all encompass several paralogs, were found and cloned. These are identified by addition of arabic numbers, e.g. PtCRC-IV-2 (some with further subforms, such as PtCRC-III-1a and -1b). For detailed terminology, see (Ladenburger and Plattner, 2011). In part they show characteristics of IP₃Rs and in part of RyRs and they were the first of these types unambiguously identified in protozoa at a molecular level. Some other CRCs of this family show one or the other, but not all of the characteristic features of either IP₃Rs or RyRs. Domains under consideration include the IP₃-binding domain (IP₃-BD), the RyR/IP₃R-homology (RIH) domain, the pore domain with its transmembrane domains and the selectivity filter as well as regions with variable similarity to mammalian IP₃Rs or RyRs. It was thus identified an unexpected total of 34 IP₃R- and RyR-like channels in the P. tetraurelia cell (Ladenburger and Plattner, 2011). Generally only a selected paralog of one subfamily has been analyzed in more detail. This high number of PtCRCs clearly is the result of several whole genome duplications (Aury et al., 2006) as one can derive from the high similarity of many – though not all - paralogs contained in each subfamily. Remarkably all sequences are expressed, except one from subfamily VI, which may be on the way to pseudogene formation (Ladenburger and Plattner, 2011).

These attempts of identification were complemented by localization of PtCRCs at the light and electron microscope level and by functional analyses: IP₃ binding and activation of injected caged IP₃ on the one hand (Ladenburger et al., 2006) and on the other hand by activation by RyR agonists (Ladenburger et al., 2009) whose secretagogue effect had been probed before (Klauke and Plattner, 1998; Klauke et al., 2000; Plattner and Klauke, 2001). Also discussed in more detail below are fluorescent imaging experiments.

The members of the 6 subfamilies are all distinctly placed in the cell (Fig. 1) and, according to gene silencing experiments, they can account for the regulation of widely different functions in different regions of the highly complex Paramecium cell. In detail, subfamily I channels (in our designation PtCRC-I) are associated with the ER (Ladenburger and Plattner, 2011). PtCRC-II/IP₃Rs are restricted to the contractile vacuole complex (serving in fresh water organisms for the expulsion of water and of some ions, including an excess of Ca²⁺) and, therefore, may fine-tune Ca²⁺ homeostasis by partial reflux of Ca²⁺ (Ladenburger et al., 2006). Spontaneous Ca²⁺ puffs are seen along the tubular extensions of this organelle, indicating constitutively active IP₃R type channels - a phenomenon reported later on also for a chicken lymphocyte cell line (Cardenas et al., 2010) and mammalian atrial myocytes (Horn et al., 2013). PtCRC-III molecules are associated with recycling vesicles engaged in phagosome formation (Ladenburger and Plattner, 2011). A sequence indicating an IP₃-BD occurs in the members of these three subfamilies, I to III, but has been experimentally verified only with PtCRC-II type channels. PtCRC-IV channels display structural and functional characteristics of RyRs (Ladenburger et al., 2009). They are localized to the established subplasmalemmal Ca²⁺-stores, the alveolar sacs. Note that in apicomplexan parasites, the structural equivalent is the “inner membrane complex”, whose relevance for Ca²⁺ signaling is not known (Plattner et al., 2012). Silencing reduces stimulated exocytosis in response to RyR agonists. The remaining PtCRC subfamily types, PtCRC-V and PtCRC-VI, have a more complex distribution (Ladenburger and Plattner, 2011), as shown in Fig. 1.

Fig. 1.

Distribution of CRCs in the P. tetraurelia cell (PtCRC), subfamilies I to VI (and subtypes analyzed in some detail given in arabic numbers), as localized by immuno-fluorescence. Additonal sites of PtCRC localization are not included in the scheme. These are PtCRC-V-4, occurring along the oral cavity (together with PtCRC-VI-3) and along the cleavage furrow; remarkably PtCRC-V-4 also occurs in the parasomal sacs (clathrin coated pits at the cilairy basis) membranes. Beyond structures identified below the scheme, the following abbreviations are used: in the contractile vacuole (cv) system, a marks ampullae (with extending radial/collecting canals), ds = decorated spongiome and ss = smooth spongiome. Other structures are: ac = acidosomes (vesicles of endosomal origin contributing to phagosome/food vacuole formation), cf = cytopharyngeal fibers; ci = cilia; cp = cytoproct, dv = discoidal vesicles, rv = recycling vesicles, mac = macronucleus, oc = oral cavity, pof = postoral fibers, ps = parasomal sacs (clathrin-coated pits), tr = trichocysts and their “ghosts” (gh) occurring after contents release. Compiled according to (Ladenburger and Plattner, 2011).

Altogether, immunolocalization studies with antibodies against subfamily PtCRC members revealed widely different, but distinct localization and functional engagement of the different Ca²⁺-release channel types also turned out to be different. Some PtCRCs occur in different organelles and some organelles possess different CRC types, as summarized in Fig. 1, with some selected examples presented in Figs. 2 and 3. In alveolar sacs PtCRC-IV molecules are scattered over the entire peripheral part where they face the cell membrane (Ladenburger et al., 2009), thus allowing spilling of Ca²⁺ over exocytosis sites. An additional type of CRCs is found in this organelle, i.e. PtCRC-V-4, with characteristics of an IP₃R; it is positioned laterally, i.e. where adjacent sacs approach each other (Ladenburger and Plattner, 2011). Other PtCRCs are generally associated with vesicles participating in trafficking, as known from in vivo and ultrastructural analyses, as well as from the topology of specific SNARE proteins (Plattner, 2010) that mediate specific membrane interactions.

Fig. 2.

Examples of immuno-localization of different PtCRCs. (A) PtCRC-II (IP₃R) is localized to the star-shaped contractile vacuole complex. (B) PtCRC-III-4 is localized to the phagocytic pathway, including recycling vesicles. Here, phagosomes (food vacuoles) show up in dark due to the addition of Indian ink in the medium. For better identification of structures labeled with antibodies against PtCRC-III-4 (green) red antibodies have been applied to stain microtubules. Here, PtCRC-III-4 traffic with recycling vesicles along microtubules that are associated with the postoral fibers (pof) which originate from the oral apparatus (oa). By contrast, the contractile vacuole complex (cvc) is stained only for tubulin, but not for PtCRC-III-4. Also indicated is the number of z-stacks used for a pile-up image in (B). See Fig. 1 for further details. (C) Immunolocalization of PtCRC-IV-1 (RyR-LP) by green fluorescence. Red staining comes from labeling of protein disulfide isomerase (PDI), an ER-specific marker. Note the localization of PtCRC-IV-1 preferably to the alveolar sacs (Ca²⁺-stores, represented by the green patches in the cell cortex) and less to the ER. Abbreviations are: oc = oral cavity, cp = cytoproct. (D) PtCRC-V-4 localizes to micronuclei (mic), i.e. their envelopes, as well as to cell surface components (see Fig. 1 for details). From (Ladenburger et al., 2009). Scale bars = 10 μm.

Fig. 3.

(A) Molecular modeling of the inositol 1,4,5-trisphosphate-BD of PtCRC-II (designated PtIP₃R_N) in comparison to type 1 IP₃R of the mouse (MmIP₃R1). The IP₃-BD of the Paramecium molecule has been modeled by comparison with that of MmIP₃R1. Numbers in parenthesis indicate the amino acid sequence included in the modeling procedure. Note considerable coincidence of the motifs in both IP₃Rs, with some additional loops (of yet unexplained significance) in the IP₃-BD of Paramecium. From (Ladenburger et al., 2006). (B) Exocytosis performance (percent of cells performing exocytosis [ordinate] to a different extent indicated in the abscissa) and the effect of PtCRC-IV-1 (RyR-LP) silencing. Two types of control cells have been evaluated: (i) Cells mock-silenced with the vector (pPD) containing only a GFP sequence served as a negative control; (ii) cells silenced in the exocytosis-relevant ND7 gene served as a positive control. The relevant experiment was carried out with a vector containing a sequence appropriate to silence PtCRC-IV. Note considerable depression of AED-stimulated exocytosis after ND7 and PtCRC-IV silencing, respectively. From (Ladenburger et al., 2009). (C) Effect of silencing of PtCRC-IV-1 (RyR-LP) on Ca²⁺-signaling. Cells were exceptionally contained in a medium with [Ca²⁺]_o reduced to a calculated value of ~30 nM, i.e. slightly below internal resting level ([Ca²⁺]_i ~ 50 nM, thus excluding signals from Ca²⁺-influx. The Ca²⁺ signal generated during exocytosis stimulation was evaluated by ratio imaging (f/f_o). Black lines: non-silenced cells, red lines: silenced cells. Exocytosis was stimulated by aminoethyldextrane (AED), 4-chloro-m-cresol (4CmC) and caffeine, respectively. For further details, see text. From (Ladenburger et al., 2009).

3.2 Functional aspects of IP₃R- and RyR-type CRCs in Paramecium

Only PtCRC-II (IP₃Rs) and PtCRC-IV (RyR-LP) channels have been analyzed in some more detail by post-transcriptional gene silencing, i.e. by applying the method described by (Galvani and Sperling, 2002). Genes diverging by >15 percent can thus be differentially silenced at a post-transcriptional level. This means that not all paralogs of a subfamily can be safely addressed and a similar uncertainty holds for antibodies used for immunolocalization.

3.2.1 IP₃ receptors

The PtCRC-II/IP₃R localized selectively to the contractile vacuole system, from the contractile vacuole bladder, over emanating radial (connecting) arms to the entire “smooth spongiome”, a network of anastomosing membrane-bounded tubules. Here, a H⁺ gradient generated by a vacuolar H⁺-ATPase pump in the adjacent “decorated spongiome” drives chemiosmotically the uptake of Ca²⁺ and water (Fok et al., 1995; Wassmer et al., 2009) for collection by the smooth spongiome, delivery to the vacuole bladder and periodic extrusion by exocytosis (Stock et al., 2002). What may the function of the organelle-specific CRCs then be? In fact, using a Ca²⁺-sensitive fluorescent dye, irregular, spontaneous Ca²⁺ puffs along the radial arms/smooth spongiome part of the organelle were observed (Ladenburger et al., 2006). Ca²⁺ signals have been altered by UV-activation of caged Ca²⁺ and Ca²⁺-dependent biosynthetic pathways (organelle biogenesis) have been inhibited by silencing the PtCRC-II/IP₃R; this, together with the unambiguous identification of an IP₃-BD (with characteristics of a low affinity type of mammalian IP₃R), shown in Fig. 3A, led the authors (Ladenburger et al., 2006) to conclude that these channels allow for some Ca²⁺ reflux from the organelle. However, the function may be dual. First, they may serve for fine-tuning of Ca²⁺ secretion (Ladenburger et al., 2006) similar to systemic functions in kidney. Second, they may provide Ca²⁺ for local restructuring of the spongiome by reversible fusion and fission processes by the numerous organelle-specific SNARE proteins (Schönemann et al., 2013). This may enable the adjustment of organelle structure and function to the actual physiological requirements (Plattner, 2013).

3.2.2 Ryanodine receptor-like proteins

The PtCRC-IV/RyR-LP localized to the alveolar sacs (Ladenburger et al., 2009) that had been previously identified (Stelly et al., 1991) and further characterized (Lange et al., 1995; Hardt and Plattner, 2000) as cortical Ca²⁺-stores. Alveolar sacs are tightly attached to the cell membrane. By immunogold EM analysis these channels have been localized to the outer part of alveolar sacs facing the cell membrane (Ladenburger et al., 2009). Aspects of functional characterization are contained in Fig. 3B, C. These were identified by activation with the secretagogue aminoethyldextran (AED), caffeine, or with the ryanodine substitute 4-chloro-meta-cresol (4CmC) in conjunction with Ca²⁺ imaging, paralleled by gene silencing (Fig. 3C). Since gene silencing inhibited secretagogue-induced Ca²⁺ signals (Fig. 3C), as well as trichocyst exocytosis (Fig. 3B), these channels were concluded to transport Ca²⁺. This function has been identified as the first step of signal transduction, which induces, as a second step, a superimposed Ca²⁺-influx from the medium (store-operated Ca²⁺-entry, SOCE). The activity of both components in concert had been envisioned previously by whole cell-patch electrophysiology (Erxleben and Plattner, 1994; Erxleben et al., 1997), by fluorochrome analysis under selective conditions (Klauke et al., 2000) and by elemental analysis at the EM level using energy-dispersive x-ray microanalysis (Hardt and Plattner, 2000). The situation clearly recalls that of the junction between the SR and the plasma membrane in striated muscle, although SOCE is much more widely distributed.

The simultaneous occurrence of PtCRC-IV- channels (RyR type) and additionally of PtCRC-V-4 (IP₃R type) in alveolar sacs is also not without precedent. It also occurs in the ER of rat sensory neurons (Solovyova and Verkhratsky, 2003) and in mammalian skeletal muscle RyRs and IP₃Rs cooperate to activate Ca²⁺-signaling via the SR (Tjondrokoesoemo et al., 2013). A similar cooperativity has been detected in atrial cells of the heart (Horn et al., 2013).

There are several unexplored properties of PtCRCs including the occurrence of mixed features in some subfamilies. For instance, the molecular size of PtCRCs of the RyR type (e.g. PtCRC-IV) is unusually small. According to their amino acid sequence all PtCRCs are around 300 kDa in size (Ladenburger and Plattner, 2011) and thus resemble IP₃Rs in metazoans. Functional implications of this peculiarity remain to be explored. This may be one of the criteria indicating the occurrence of a common ancestral form of IP₃Rs and RyR-LPs in protozoa (Plattner and Verkhratsky, 2013), whereas in metazoa RyRs are much larger than IP₃Rs (Taylor et al., 2009). Also unusual is the occurrence of two large loops in the PtCRC-II-1 (IP₃R) molecule (Ladenburger et al., 2006) (Fig. 3A).

Evaluation by very recently developed data based algorithms for the determination of transmembrane domains as specified (Ladenburger and Plattner, 2011), has indicated that PtCRCs, including RyR-LPs, possess 6 transmembrane domains (TMD) – previously a matter of debate. Thus, PtRyR-LPs differ from the most widely maintained assumption of only four TMDs. A more recent computational evaluation of the mammalian RyR came to the same conclusion (Ramachandran et al., 2013). Domains for tetramerization are available also in the PtCRC molecules (Ladenburger and Plattner, 2011).

From their occurrence in protozoa, together with the occurrence of mixed type molecules with overlapping characteristics of IP₃Rs and RyRs, we assume that some of the PtCRCs may represent ancestral CRC types, close to such primeval forms. Also the amino acids composing the selectivity filter, i.e. Gly-Ile-Gly-Asp, are identical in both types of PtCRCs (Ladenburger and Plattner, 2011). This also occurs in other lower eukaryotes (Plattner and Verkhratsky, 2013) and, thus, is in contrast to the sequence Gly-Val-Gly-Asp in the IP₃R of metazoan cells (Boehning, 2010). In Paramecium, the latter sequence occurs in PtCRC-I-1a, -1b, and -1c, all with and IP₃-BD (Ladenburger and Plattner, 2011), though its IP₃-binding capacity has not been probed experimentally as yet. Otherwise the diversification of these channels in Paramecium during evolution is quite similar to the diversification of other molecules pertinent to vesicle trafficking, such as SNAREs (Plattner, 2010). This also includes more or less diversification as well as partial elimination by pseudogenization.

3.3 Pharmacology of IP₃R- and RyR-type CRCs and effect of injected cADPR and NAADP on cell function in Paramecium

Although the mechanism of AED as a secretagogue (Plattner and Hentschel, 2006) is not known in detail, some exogenous polyamines are known to induce a SOCE mechanism and exocytosis in various metazoan cells (Williams, 1997; Gamberucci et al., 1998; Plattner and Klauke, 2001; Plattner and Hentschel, 2006). By contrast, caffeine and 4CmC are established activators of RyRs up to mammalian cells (Cheek and Barry, 1993; Westerblad et al., 1998). The amino acids necessary for 4CmC binding (Fessenden et al., 2006) are found in the PtCRC-IV molecule (Ladenburger et al., 2009). Assays for Ca²⁺-dependent ryanodine binding to isolated alveolar sacs gave no results. This is not surprising considering the largely aberrant pharmacology of ciliates (Plattner et al., 2009) and the absence of any evolutionary pressure for this plant toxin. Also the inhibitory effect of Li⁺ on phenomena related to IP₃ in Paramecium (Ladenburger et al., 2006) are no stringent argument considering its pleiotropic effects, whereas the usefulness of compound U73122 as an inhibitor of phosphoinositide-specific phospholipase C in ciliates (Leondaritis et al., 2011) has not yet been known at the time of those analyses.

It is well known that Ca²⁺ effects are strictly locally confined and rapidly counteracted by different mechanisms (see “Introduction”). This explains two observations: microinjection of an excess of Ca²⁺ into a Paramecium cell does not cause any exocytosis (Klauke and Plattner, 1997). Similarly a diffuse Ca²⁺-influx, as achieved by sudden increase of [Ca²⁺]_e, does not result in exocytosis (Erxleben et al., 1997). Also injection of a likely activator of RyRs in mammalian cells (Zalk et al., 2007), cADPR, causes no trichocyst exocytosis, as is the case with injected NAADP, the putative activator of TPCs (Galione et al., 2009; Galione et al., 2010). The localization of such channels is not known for Paramecium. Since either compound changes contractile vacuole pulsations (Plattner et al., 2012), organelles regulating Ca²⁺ homeostasis, one may assume the occurrence of the respective target molecules. However, the proper target may not be reached by the injected compounds, as described above for Ca²⁺, and the effects achieved with the contractile vacuole may be due to Ca²⁺ activation from remote organelles. With Paramecium homogenates a K_d ~3.5 nM for NAADP binding was determined (Plattner et al., 2012). Although a variety of acidic organelles, which could harbor TPCs, are known from Paramecium (Wassmer et al., 2009) acidocalcisomes have not yet been identified. Also unknown is whether vacuoles containing Ca/Mg-phosphate crystals (Grover et al., 1997) would be dynamic Ca²⁺-stores or just waste disposal containers.

4. Intracellular calcium channels in parasitic protozoa

4.1 IP₃R in trypanosomatids

Ca²⁺ homeostasis in trypanosomes differs significantly from that in mammalian cells. There are no orthologs for receptor-operated or store-operated Ca²⁺ channels although the parasites possess orthologs for a putative voltage-sensitive Ca²⁺ channel that, in the case of T. brucei, localizes to the flagellum (Oberholzer et al., 2011). No orthologs for Na⁺/Ca²⁺ exchangers are present in trypanosomes. In contrast to the plasma membrane Ca²⁺-ATPase (PMCA) of higher eukaryotes, trypanosome PMCA apparently lacks a calmodulin-BD and is also localized intracellularly (Lu et al., 1998; Luo et al., 2004). The sarcoplasmic-endoplasmic reticulum Ca²⁺-ATPase (SERCA) of trypanosomatids is insensitive to inhibitors of mammalian SERCA ATPases, such as thapsigargin (Docampo et al., 1993; Vercesi et al., 1993; Furuya et al., 2001). In addition, trypanosomes possess an important acidic calcium store, the acidocalcisomes, which is rich in polyphosphate (Docampo et al., 2005).

Recent results (Huang et al., 2013) have indicated that the IP₃R of T. brucei is localized to acidocalcisomes rather than to the ER (Fig. 4A). The demonstration of the localization of the IP₃R in acidocalcisomes was obtained by tagging the C-terminus of TbIP₃R of procyclic trypomastigotes with an hemagglutinin tag using homologous recombination with the endogenous gene locus (Huang et al., 2013). The TbIP₃R partially co-localized with antibodies against T. brucei acidocalcisome marker vacuolar proton pyrophosphatase (V-H⁺-PPase, or TbVP1) (Fig. 4A). An additional punctate staining of TbVP1 that did not co-localize with TbIP₃R was detected and could correspond to trafficking vesicles. It was described before that adaptor protein-3 (AP-3) complex is involved in sorting proteins, like TbVP1, to acidocalcisomes from the Golgi or from endosomes in both L. major (Besteiro et al., 2008) and T. brucei (Huang et al., 2013). No co-localization with TbBiP, an ER marker (Bangs et al., 1993) with a clear reticular labeling (Fig. 4B), was detected, thus ruling out ER localization of the TbIP₃R. The acidocalcisome localization was confirmed using specific antibodies against TbIP₃R (unpublished results).

Fig. 4.

Localization of TbIP₃R in procyclic trypomastigotes. (A) TbIP₃R partially co-localizes with TbVP1 in acidocalcisomes (Pearson’s correlation coefficient of 0.874), as shown by immunofluorescence microscopy analysis. The merge images show the co-localization in yellow. (B) Lack of co-localization of TbIP₃R with TbBiP in the endoplasmic reticulum (Pearson’s correlation coefficient of 0.156). Scale bars, 10 μm. DIC, differential interference contrast. From (Huang et al., 2013).

Proteomic analysis of contractile vacuole complex (Ulrich et al., 2011) and acidocalcisome fractions (unpublished) of T. cruzi provided evidence of the presence of the TcIP₃R ortholog in these organelles. These results coincided with the punctate and vacuolar localization reported for TcIP₃R by other authors (Hashimoto et al., 2013). These authors suggested an ER localization of TcIP₃R although no clear co-localization with TbBiP antibodies was observed.

The acidocalcisome localization of TbIP₃R led to test for Ca²⁺ release by IP₃ in permeabilized cells under conditions of optimal acidocalcisome function, i.e., in the presence of pyrophosphate (PPi). Addition of PP_i is necessary to acidify acidocalcisomes by the action of TbVP1. This acidification allows Ca²⁺ uptake by the acidocalcisome Ca²⁺-ATPase, which is Ca²⁺/H⁺ countertransporting (it transports Ca²⁺ in exchange for H⁺). Under such conditions, IP₃ addition results in significant Ca²⁺ release (Huang et al., 2013). Similar experiments were done with isolated acidocalcisomes resulting in significant Ca²⁺ release by IP₃ (Huang et al., 2013). Previous attempts to show Ca²⁺ release by IP₃ had been unsuccessful (Moreno et al., 1992a; Moreno et al., 1992b), which was difficult to explain considering that the parasites do have a phosphoinositide phospholipase C (PI-PLC) (Furuya et al., 2000; Okura et al., 2005; Martins et al., 2010), the enzyme that cleaves phosphatidylinositol 4,5-bisphosphate to generate diacylglycerol and IP₃, and that IP₃ was detected in both T. cruzi (Moreno et al., 1992b) and T. brucei (Moreno et al., 1992a). When those experiments were done (1991–2) the presence of acidocalcisomes (Vercesi et al., 1994; Docampo et al., 1995) and an acidocalcisome vacuolar H⁺-PPase in trypanosomes were not known, as they were discovered much latter (Scott et al., 1998; Rodrigues et al., 1999). The reason for the lack of Ca²⁺ release in those experiments was that permeabilization results in dilution of substrates (ATP, PP_i) and alkalinization of acidocalcisomes, as a result of lack of function of the proton pumps in the absence of ATP and PP_i.

The gene coding for TbIP₃R (Tb927.8.2770) shares 41% amino acid identity with T. cruzi IP₃R (TcCLB.509461.90), and orthologs are also present in several Leishmania spp. (Prole and Taylor, 2011). Structural analysis (ELM and TMHMM servers) predicted 5 transmembrane domains in the C-terminal region of these receptors. The ORFs of T. brucei and T. cruzi IP₃Rs predict 3099, and 3011 amino acid proteins, with apparent molecular weights of 343, and 337 kDa, respectively. Trypanosome IP₃Rs possess a series of conserved domains including putative suppressor domain-like (SD), ryanodine receptor IP₃R homology (RIH), and RIH-associated (RIAD) domains (Prole and Taylor, 2011). A motif for a Ca²⁺-specific selectivity filter (GVGD) (Boehning et al., 2001; Boehning, 2010) is present in the putative intraluminal loop between transmembrane domains at the C-terminal region (Huang et al., 2013). This sequence resembles that of IP₃Rs in higher eukaryotes (Boehning, 2010) and also occurs in PtCRC-I type channels (with a putative IP₃-BD), opposite to PtCRC-II channels (with an established IP₃-BD (Ladenburger and Plattner, 2011; Plattner and Verkhratsky, 2013). Of the ten residues that have been proposed to form a basic pocket that binds IP₃ (Yoshikawa et al., 1996; Bosanac et al., 2002), 4 are conserved in TbIP₃R. Other features of trypanosomatid IP₃Rs have been described before (Prole and Taylor, 2011).

In addition to the studies on permeabilized trypanosomes (Huang et al., 2013) functional analyses of TbIP₃R (Huang et al., 2013) and TcIP₃R (Hashimoto et al., 2013) were also done by stable transfection of the respective genes in a chicken B lymphocyte cell line (DT40, R23-11) in which the genes for all three vertebrate IP₃Rs have been stably ablated (DT40-3KO) (Miyakawa et al., 1999). Both TbIP₃R and TcIP₃R1 localized to the ER of DT40-KO cells, and Ca²⁺ release by IP₃ was investigated using permeabilized cells (Huang et al., 2013), microsomal vesicles, or intact cells stimulated by anti-B cell receptor monoclonal antibodies (Hashimoto et al., 2013). Microsomal vesicles from DT40-KO cells expressing TcIP₃R also exhibited IP₃ binding activity (Hashimoto et al., 2013). TcIP₃R was also expressed in HeLa cells, where it localized to the ER, and these permeabilized cells also exhibited Ca²⁺ release in the presence of IP₃ (Hashimoto et al., 2013). TbIP₃R was found to be considerably less sensitive to IP₃ than the rat IP₃R1 (RnIP₃R1) transfected in DT40-3KO cells (Huang et al., 2013). Ca²⁺ release by IP₃ was also investigated in live T. brucei procyclic trypomastigotes loaded with Fluo 4-AM with caged IP₃ (Huang et al., 2013). In cells loaded with caged IP₃ there were rapid Ca²⁺ increases after UV flashes to release free IP₃ (Fig. 5), and these increases were considerably reduced when using trypanosomes in which the expression of TbIP₃R was downregulated by RNAi (Huang et al., 2013).

Fig. 5.

Caged IP₃-dependent Ca²⁺ release in T. brucei. (A–B) Representative traces of Ca²⁺ responses to UV flash in control cells in the absence (A) and presence (B) of caged IP₃, respectively (first flash 3 pulses, second flash 6 pulses). Nigericin (Nig; 5 μM) and Ionomycin (IO; 5μM) were added where indicated. From (Huang et al., 2013).

To study the importance of TbIP₃R and TcIP₃R in the biology of trypanosomes, several strategies were used. Knockdown of TbIP₃R expression was done by induction of RNAi and resulted in growth defects in both bloodstream and procyclic trypomastigotes (Huang et al., 2013). Knockdown of the expression of TbIP₃R in procyclic forms by RNAi reduced the ability of IP₃ to release Ca²⁺ from permeabilized cells and reduced the virulence of bloodstream forms in vivo (Huang et al., 2013). Knockdown of TcIP₃R was done by single-knockout in epimastigotes of the Tulahuen strain. Attempts to obtain null mutants in this or in the Esmeraldo strain were unsuccessful, suggesting the essentiality of this gene (Hashimoto et al., 2013). TcIP₃R knockdown resulted in deficient growth of epimastigotes, deficient metacyclogenesis (transformation of epimastigotes into metacyclic trypomastigotes), deficient host cell invasion by trypomastigotes associated with reduced Ca²⁺ release upon their attachment to the host cells, deficient replication of amastigotes, increased transformation of amastigotes into trypomastigotes, and defects in virulence in vivo (Hashimoto et al., 2013). Overexpression of TcIP₃R also resulted in deficient growth of epimastigotes and amastigotes, and deficient metacyclogenesis, suggesting that an appropriate level of this receptor is necessary for these processes (Hashimoto et al., 2013). In contrast, overexpression of TcIP₃R resulted in increased host cell invasion by trypomastigotes associated with increased Ca²⁺ release upon their attachment to host cells, and decreased transformation of amastigotes into trypomastigotes, with no changes in virulence in vivo except for an early appearance of parasitemia (Hashimoto et al., 2013).

In conclusion, these works (Hashimoto et al., 2013; Huang et al., 2013) clearly established the presence of a functional IP₃ receptor in T. brucei and T. cruzi, and together with previous reports (Docampo and Pignataro, 1991; Moreno et al., 1992a; Moreno et al., 1992b; Furuya et al., 2000; Okura et al., 2005; de Paulo Martins et al., 2010; Martins et al., 2010), the function of a complete IP₃/diacylglycerol pathway in trypanosomes.

Although acidocalcisomes were initially described almost 20 years ago in T. brucei (Vercesi et al., 1994), the mechanism for Ca²⁺ release from these organelles was unknown until now. The localization of the IP₃ receptor in acidocalcisomes (Huang et al., 2013) provides the long sought mechanism for Ca²⁺ release from these organelles.

4.2 Intracellular Ca²⁺ channels in Toxoplasma gondii

Measurements of intracellular Ca²⁺ levels have been done in T. gondii extracellular tachyzoites using the Ca²⁺ dye Fura 2-AM (Fura 2/acetomethoxy) and values of 60–100 nM were obtained (Moreno and Zhong, 1996). The ER and acidocalcisomes were identified as the largest Ca²⁺ stores. A SERCA-type Ca²⁺-ATPase, which is present in the ER (Nagamune et al., 2007) and inhibited by thapsigargin (Moreno and Zhong, 1996), is the main Ca²⁺ uptake mechanism in this compartment. In T. gondii, the presence of intracellular Ca²⁺ stores responsive to IP₃, ryanodine (Lovett et al., 2002) and cADPR (Chini et al., 2005; Nagamune et al., 2008), have been described but there is no genetic evidence for the presence of IP₃ or ryanodine gated channels (Nagamune and Sibley, 2006; Plattner and Verkhratsky, 2013). This is despite the evidence for the presence of enzymes involved in the generation of some of these second messengers such as a phosphoinositide phospholipase C (Fang et al., 2006) and cADPR cyclase and hydrolase activities (Chini et al., 2005). Acidocalcisomes of T. gondii possess a PMCA-type Ca²⁺-ATPase (TgA1) for Ca²⁺ uptake (Luo et al., 2001; Rohloff et al., 2011), but their mechanism of Ca²⁺ release is unknown. Evidence for a Ca²⁺/H⁺ exchanger in these acidocalcisomes has also been reported (Rohloff et al. 2011). The recently described plant-like vacuole (PLV) (Miranda et al., 2010) was also found to be rich in Ca²⁺ and also possesses the PMCA-type Ca²⁺-ATPase (TgA1) for Ca²⁺ uptake. A Ca²⁺/H⁺ exchanger is also present in the PLV (Miranda et al., 2010), but the mechanism of Ca²⁺ release is also unknown. There is no genetic evidence for the presence of a mitochondrial calcium uniporter (MCU) in any Apicomplexan parasite (Bick et al., 2012) and the role of the mitochondria in Ca²⁺ regulation is not clear. A Ca²⁺/H⁺ antiporter is apparently present in the mitochondria of T. gondii (Guttery et al., 2013). Ca²⁺ entry into tachyzoites is regulated (Pace et al., 2013). Other unexplored sources of Ca²⁺ in T. gondii are Golgi complex, apicoplast, inner membrane complex (IMC), and secretory organelles.

T. gondii is unique among parasitic protozoa in possessing an ortholog to two pore channels (TPCs), for which there are no orthologs in other Apicomplexans (Prole and Taylor, 2011). This TPC has substantial similarity to mammalian TPCs in the pore region responsible for ion conduction, suggesting that it may act as Ca²⁺-permeable channel (Prole and Taylor, 2011). However, the localization and function of this channel have not been reported. Mammalian TPCs can be Ca²⁺ channels gated by NAADP, and localized in lysosome-like compartments (Brailoiu et al., 2009; Calcraft et al., 2009; Zong et al., 2009). Their function as Ca²⁺ channels and their stimulation by NAADP have been disputed (Wang et al., 2012; Cang et al., 2013), although the reason for this discrepancy is apparently that tagging the N-terminal region of the channel abolishes its sensitivity to NAADP (Churamani et al., 2013). Fig. 6 shows a schematic representation of Ca²⁺ distribution in tachyzoites.

Fig. 6.

Schematic representation of the distribution of Ca²⁺ in a T. gondii tachyzoite. Ca²⁺ entry is probably through Ca²⁺ channels (a). Once inside the cells, Ca²⁺ can be translocated back to the extracellular environment, primarily by the action of the PMCA (b). In addition, Ca²⁺ will interact with binding proteins (CBPs) or become sequestered by the ER by the action of the SERCA-Ca²⁺-ATPase (c), passively sequestered by the mitochondrion (d), or sequestered by the acidocalcisome (e) or the PLV (f) by the action of a Ca²⁺-ATPase (TgA1). Ca²⁺ appears to diffuse freely into the nucleus. Calcium could also be released into the cytoplasm from the internal stores, such as the ER, through an uncharacterized channel, which appears to respond to ryanodine and caffeine (g). It may also be released from the PLV and the acidocalcisome (AC) through a Ca²⁺/H⁺ exchanger (h). Acidic compartments such as the acidocalcisome and the PLV contain enzymes involved in their acidification e.g. the H⁺- ATPase (i) and the vacuolar-H⁺-pyrophosphatase (j). Mitochondrial Ca²⁺ release is through a Ca²⁺/H⁺ exchanger (k).

4.3 Mitochondrial calcium uniporter

Mammalian mitochondria have been shown to have a central role in intracellular Ca²⁺ homeostasis, and it is well established that intramitochondrial Ca²⁺ concentration can rapidly reach tens or hundreds micromolar values upon cytosolic Ca²⁺ rises of a few micromolar (Rizzuto et al., 1993; Montero et al., 2000). This is because mitochondria are exposed to microdomains of high Ca²⁺ concentration in proximity to sites of Ca²⁺ release at the ER, or to Ca²⁺ channels at the plasma membrane (Rizzuto et al., 1993; Rizzuto et al., 1998; Csordas et al., 1999; Montero et al., 2000; Csordas et al., 2010; Giacomello et al., 2010). This Ca²⁺ uptake is important for shaping the amplitude and spatio-temporal patterns of cytosolic Ca²⁺ increases (Boitier et al., 1999; Hajnoczky et al., 1999; Tinel et al., 1999) and for regulating the activity of three intramitochondrial dehydrogenases that result in increased ATP generation (Denton and McCormack, 1990; McCormack et al., 1990; Hajnoczky et al., 1995; Jouaville et al., 1999; Voronina et al., 2010), as well as in stimulating the activity of the mitochondrial ATP synthase (Balaban, 2009). Ca²⁺ also regulates mitochondrial transporters in the inner membrane (Lasorsa et al., 2003; Satrustegui et al., 2007). Excessive Ca²⁺ uptake, however, favors the formation of the “permeability transition pore” leading to the release of pro-apoptotic factors in the cytosol and cell death (Kroemer et al., 2007).

The ability of mitochondria to take up Ca²⁺ was discovered more than 50 years ago (De Luca and Engstrom, 1961; Vasington and Murphy, 1962) and the channel biophysical properties were well characterized in a patch clamp study of mitoplasts (mitochondria devoid of the outer mitochondrial membrane) (Kirichok et al., 2004). However, the molecular nature of the uniporter remained unknown for decades.

For many years after the discovery of the MCU in mammalian mitochondria, it was thought that less complex life forms such as plants, insects and other invertebrates, or unicellular organisms such as yeast, lacked a specific mitochondrial calcium uptake pathway (Carafoli and Lehninger, 1971). However, in 1989 it was reported (Docampo and Vercesi, 1989a, b) that epimastigotes of T. cruzi possesses a MCU with characteristics similar to those described in mammalian mitochondria: electrogenic transport, sensitivity to ruthenium red, and low affinity for the cation. The evidence of the presence of a MCU in trypanosomes but its absence in yeast was the key to the discovery of the molecular identity of MCU. An elegant study by Perocchi and coworkers (Perocchi et al., 2010) first identified a gene called mitochondrial calcium uptake 1 or MICU1 as encoding a potential regulator of the uniporter. The study was based on the observation that the Ca²⁺ uniporter was detected in T. cruzi (Docampo and Vercesi, 1989a, b) and Leishmania donovani (Vercesi and Docampo, 1992) yet not measurable in the yeast Saccharomyces cerevisiae (Balcavage et al., 1973). From a library of 1,000 mouse mitochondrial proteins, 18 candidate genes were identified that have homologues in vertebrates and trypanosomes but not in yeast (Perocchi et al., 2010). Using short hairpin (sh)RNA silencing of 13 selected genes in a commercially available HeLa cell line that stably expresses a mitochondria-targeted aequorin (mt-AEQ) as a reporter of Ca²⁺ uptake, they identified MICU1 as an important component in Ca²⁺ uptake. Based on the finding of Perocchi et al. (Perocchi et al., 2010), De Stefani et al. (De Stefani et al., 2011) and Baughman et al. (Baughman et al., 2011) used a similar strategy of comparing between mitochondrial genomes of trypanosomes and yeast and performing RNAi experiments of the identified genes and found a gene encoding a protein with all the characteristics of the mitochondrial calcium uniporter (MCU).

The MCU was also found in other trypanosomatids including T. brucei (Docampo and Lukes, 2012). The finding of a MCU uniporter in the bloodstream stage of T. brucei (Vercesi et al., 1992) was surprising because these stages lack a respiratory chain. However, mitochondrial Ca²⁺ uptake can also be energized by ATP in the absence of respiration, in which case it is inhibited by oligomycin, and not by inhibitors of the respiratory chain (Lehninger et al., 1963). This phenomenon also occurs in bloodstream trypomastigotes: the mitochondrial membrane potential is dependent on hydrolysis of ATP by the ATP synthase which acts as an ATPase (Nolan and Voorheis, 1992; Vercesi et al., 1992; Schnaufer et al., 2005; Brown et al., 2006) allowing for Ca²⁺ to still be electrophoretically transported by the MCU (Vercesi et al., 1992). Ca²⁺ uptake by bloodstream forms of T. brucei has three characteristics: 1) it occurs until the ambient free Ca²⁺ concentration is lowered to 0.6–0.7 μM; 2) it is inhibited by oligomycin; and 3) it is associated with the depolarization of the inner membrane energized by ATP. These results indicate that Ca²⁺ uptake is mediated by the ATPase-dependent energization of the inner mitochondrial membrane (Vercesi et al., 1992).

Although present in trypanosomatids and other protozoa, such as Tetrahymena thermophila, and Naegleria gruberi, the MCU is absent in Apicomplexan parasites such as T. gondii or malaria parasites (Bick et al., 2012).

The roles of mitochondrial Ca²⁺ in trypanosomes are apparently more limited than in mammalian cells. None of the dehydrogenases stimulated by Ca²⁺ in vertebrates have been studied in detail in trypanosomatids and there is no evidence of their stimulation by Ca²⁺ (Docampo and Lukes, 2012). Experiments using aequorin targeted to the mitochondria of T. brucei procyclic trypomastigotes revealed that intramitochondrial Ca²⁺ concentrations can reach values much higher than cytosolic Ca²⁺ rises when Ca²⁺ influx through the plasma membrane or Ca²⁺ release from acidocalcisomes are stimulated (Xiong et al., 1997). These results suggest a very close proximity of these organelles and the presence of microdomains of high Ca²⁺ concentration in the vicinity of the plasma membrane and acidocalcisomes (Xiong et al., 1997) and are in agreement with the presence of an IP₃R in acidocalcisomes (Huang et al., 2013). Because the ER-type Ca²⁺-ATPase (SERCA) of T. brucei is insensitive to thapsigargin, a microdomain of high Ca²⁺ concentration between the ER and the mitochondria could not be established in those studies (Xiong et al., 1997). However, these results suggest that one of the main functions of the MCU in procyclic trypomastigotes would be to shape the amplitude and spatio-temporal patterns of cytosolic Ca²⁺ increases. No similar studies have been done with bloodstream trypomastigotes. Fig. 7 shows a scheme of the close contact that would exist between acidocalcisomes and the unique mitochondrion of trypanosomes.

Fig. 7.

Scheme of the potential contact between acidocalcisomes and mitochondrion in trypanosomes. The scheme depicts the molecules mediating Ca²⁺ influx (MCU, MICU1) and efflux (Ca²⁺/H⁺ exchanger, CAX) across the inner mitochondrial membrane (IMM) at an area of acidocalcisome (ACCSOME)-mitochondrial association. The shades of gray represent the [Ca²⁺]: dark gray: > 500 μM; white, 100 nM; PMCA, plasma membrane-type Ca²⁺-ATPase; OMM, outer mitochondrial membrane; VDAC, voltage-dependent anion-selective channel; IP₃R, inositol 1,4,5-trisphosphate receptor.

Mitochondrial Ca²⁺ could also be a contributor to programmed cell death, or apoptosis-like death, in trypanosomatids. Trypanosomatids lack some of the key regulatory or effector molecules involved in apoptosis in mammalian cells, such as the tumor necrosis factor (TNF)-related family of receptors, Bcl-2 family members, and caspases (Ridgley et al., 1999; Smirlis et al., 2010; Kaczanowski et al., 2011). Mitochondrial Ca²⁺ overload with changes in mitochondrial membrane potential, reactive oxygen species (ROS) generation and release of cytochrome c have been observed upon different triggers of cell death in some trypanosomatids (Smirlis and Soteriadou, 2011). In T. brucei procyclic trypomastigotes, the production of ROS impairs mitochondrial Ca²⁺ transport, leading to its accumulation in the nucleus, and causing cell death (Ridgley et al., 1999).

In summary, mitochondrial Ca²⁺ uptake in trypanosomes appears to have a role in shaping the amplitude of cytosolic Ca²⁺ increases after influx through the plasma membrane or release from acidocalcisomes, and in apoptosis-like death, but it is not known whether it has a role in the regulation of ATP production.

4.4 Pharmacology of CRC s in parasitic protozoa

TbIP₃R does not respond to 10 μM NAADP or 1 μM cADPR when expressed in DT40-3KO cells (Huang et al., 2013). cADPR was also ineffective in producting Ca²⁺ release in permeabilized HeLa cells transfected with TcIP₃R (Hashimoto et al., 2013). In contrast, the IP₃ agonist adenophostin A (0.5 μM) showed a Ca²⁺-release that was comparable to that of 10 μM IP₃ in the TbIP₃R-expressing DT40-3KO cells or in permeabilized T. brucei procyclic forms (Huang et al., 2013).

There is pharmacological evidence for the presence of channels responsive to IP₃ in T. gondii and malaria parasites (Passos and Garcia, 1998; Lovett and Sibley, 2003; Alves et al., 2011) although there are no gene orthologs to the mammalian IP₃R or RyR in any of the Apicomplexan genomes (Nagamune and Sibley, 2006; Plattner et al., 2012). There is an ER Ca²⁺ pool, sensitive to thapsigargin, and the Ca²⁺-ATPase involved in pumping Ca²⁺ from the cytosol into the ER has been characterized (Nagamune et al., 2007).

4.5 Calcium signaling and function in parasitic protozoa

Although Ca²⁺ signaling appears to be important for several functions in T. cruzi such as host cell invasion (Moreno et al., 1994), multiplication and differentiation (Lammel et al., 1996), osmoregulation (Rohloff et al., 2003), and programmed cell death (Irigoin et al., 2009), there is much less information on the role of Ca²⁺ in T. brucei. Based on the use of Ca²⁺ ionophores, roles for Ca²⁺ in the release of the bloodstream stage surface coat (Bowles and Voorheis, 1982), and in the maintenance of the cytoskeleton (Selzer et al., 1991) have been proposed. Changes in cytosolic Ca²⁺ levels have also been reported during differentiation from bloodstream to procyclic stages of T. brucei (Stojdl and Clarke, 1996). Results obtained with TbIP₃R (Huang et al., 2013) and TcIP₃R (Hashimoto et al., 2013) knockdowns indicate that Ca²⁺ signaling through the trypanosome IP₃Rs has roles in growth in vitro and in vivo, as well as in cell differentiation. The localization of the TbIP₃R in acidocalcisomes (Huang et al., 2013) also supports a role for acidocalcisomes in Ca²⁺ signaling.

Several studies have looked into the role of Ca²⁺ signaling during the lytic cycle of T. gondii. Using Ca²⁺ ionophores, Ca²⁺-chelating agents and ethanol, a link between Ca²⁺ and conoid extrusion was demonstrated (Mondragon et al., 1994; Mondragon and Frixione, 1996; Del Carmen et al., 2009), although no direct parallel Ca²⁺ measurements were reported. The effect of Ca²⁺ on gliding motility was studied on trypsin-permeabilized tachyzoites (Mondragon and Frixione, 1996) or by measuring Ca²⁺ oscillations in extracellular tachyzoites loaded with the Ca²⁺ dye Fluo 4-AM and studying their correlation with gliding motility (Lovett and Sibley, 2003; Wetzel et al., 2004). The role of Ca²⁺ in microneme secretion was studied in extracellular tachyzoites following the effects of Ca²⁺ ionophores or chelators on protein secretion as evaluated by western blot analyses (Carruthers et al., 1999). A role for Ca²⁺ signaling in invasion was postulated on the basis of the analysis of changes in Ca²⁺ occurring in parasites loaded with Fura 2-AM upon their attachment to host cells (Vieira and Moreno, 2000) or by detecting the cessation of Ca²⁺ oscillations in extracellular tachyzoites loaded with Fluo 4-AM (Lovett and Sibley, 2003). These and more recent studies have provided indirect evidence that Ca²⁺ signaling is part of the pathways that result in the stimulation of conoid extrusion, gliding motility, microneme secretion, and invasion. These traits, which are important steps of the lytic cycle of the parasite are enhanced by extracellular Ca²⁺ (Pace et al., 2013). Ca²⁺ signaling was also proposed to be part of the pathway leading to parasite egress from the host cells based on the effect of Ca²⁺ ionophores, which stimulate egress, although it was never demonstrated that [Ca²⁺]_i increases in tachyzoites before egress (Endo et al., 1982; Garrison et al., 2012). In summary, most of the studies on Ca²⁺ signaling were done with extracellular tachyzoites (conoid extrusion, gliding motility, microneme secretion, invasion) measuring indirectly the involvement of Ca²⁺ (using Ca²⁺ chelators or ionophores) or detecting Ca²⁺ changes with fluorescent dyes independently of the phenomena examined.

5. Conclusions and Perspectives

On the one hand there are many similarities in Ca²⁺ signaling in the protozoa discussed here, but on the other hand there also occur considerable differences. Differences concern the drug sensitivity of the SERCA, the absence or the occurrence of a calmodulin-BD in the PMCA (typical for higher eukaryotes) and the occasional occurrence of this pump in Ca²⁺-storage organelles. Interestingly, a SOCE mechanism has been verified with some protozoa analyzed so far, based in part on the activation of RyR-LPs (Ladenburger et al., 2009). There is, however, no evidence of the presence of Orai or Stim orthologs in either Paramecium (Ladenburger and Plattner, 2011) or parasitic protozoa (EuPathDB.com). Acidocalcisomes are present is some (trypanosomatids, Apicomplexan) but not in other parasitic protozoa (like Giardia, Trichomonas, Entamoeba) (Docampo et al., 2005) while it is not yet known if they are present in Paramecium (Plattner et al., 2012) IP₃Rs orthologs are present in trypanosomatids (Huang et al., 2013; Hashimoto et al., 2013) and Paramecium (Plattner and Verkhratsky, 2013) but unknown in Apicomplexa.

The range of CRC molecules, type IP₃Rs/RyR in Paramecium and of IP₃Rs in Monosiga brevicollis, and trypanosomatids indicates that protozoa already have evolved this Ca²⁺-signaling pathway. Except in ciliates, only IP₃Rs have been unambiguously identified; none have been identified in Apicomplexa – in striking contrast to their response to IP₃ (which, thus, still awaits elucidation). In ciliates some IP₃Rs and RyR-LPs display mixed features of the two. Concomitantly, in Paramecium some of the CRCs can be clearly attributed to either IP₃Rs and to RyR-LPs, all CRCs are differentially localized and – as far as analyzed - can be differentially activated. In Paramecium and other protozoa the selectivity filter in both types of CRCs, IP₃R- and RyR-like channels (or their putative homologues) is identical for most paralogues (Landenburger and Plattner, 2011; Plattner and Verkhratsky, 2013). By contrast the parasitic flagellates have the same amino acid sequence in the selectivity filter of their IP₃R (Huang et al., 2013) as occurring in metazoa (Boehning, 2010). Evidently selective enrichment of Ca²⁺ in the stores can suffice to guarantee selective release of Ca²⁺ as a second messenger. A RyR-type CRC seems to be absent from Apicomplexa; none has been identified in choanoflagellates and in the parasitic flagellates discussed here, whereas their occurrence in other protists has to be left open for the time being (Ladenburger and Plattner, 2011; Plattner and Verkhratsky, 2013). Also the number of transmembrane domains in RyR-LPs, five or six being proposed depending on the database analysis method applied, requires experimental verification.

The presence of the mitochondrial calcium uniporter in a number of protozoa indicates that mitochondrial regulation of Ca²⁺ signaling is also an early appearance in evolution. There is only circumstantial or sequence evidence for the occurrence of TPCs, TRPCs and of intracellular mechano-sensitive Ca²⁺-channels in Paramecium and in parasitic protozoa. This remains a vast field for further explorations.