Intracellular calcium stores in Toxoplasma gondii govern invasion of host cells

The protozoan parasite Toxoplasma gondii is responsible for widespread infections in animals and humans, and is the cause of toxoplasmosis. One of the distinguishing features of the phylum Apicomplexa to which T. gondii belongs is the presence of apical secretory organelles. Discharge of apical organelles occurs during invasion(Carruthers and Sibley, 1997),and their contents mediate attachment to host cells and formation of the specialized parasitophorous vacuole occupied by the parasite(Carruthers, 2002). The first of the specialized secretory organelles to be discharged are micronemes, which contain a variety of adhesive and transmembrane proteins(Soldati et al., 2001; Tomley and Soldati, 2001). The proposed function of microneme proteins is to bridge parasite and host cell membranes during invasion. Consistent with this, several micronemal proteins have been shown to bind to host cells or cell surface molecules including MIC1(Fourmaux et al., 1996), MIC2(Carruthers et al., 2000; Brossier et al., 2003), MIC3(Garcia-Réguet et al.,2000; Cérede et al.,2002), AMA-1 (Donahue et al.,2000; Hehl et al.,2000) and MIC4 (Brecht et al.,2000). The diversity of cell adhesive motifs among various micronemal proteins may reflect the wide rage of host cell types that are recognized by apicomplexan parasites.

Artificially increasing intracellular calcium in the parasite through the use of calcium ionophores A23187 or ionomycin triggers parasite microneme secretion in the absence of host cells(Carruthers and Sibley, 1999). Conversely, chelation of intracellular calcium in the parasite blocks microneme secretion and invasion(Carruthers and Sibley, 1999). Studies showing that EGTA blocks invasion by T. gondii suggest that extracellular calcium might also play a role in the host-parasite interaction(Pezzella et al., 1997). Fluorescent imaging studies have revealed that parasites in association with host cells show elevated levels of cytoplasmic calcium(Vieira and Moreno, 2000). These studies have led to a model suggesting that contact with host cells triggers a rise in cytoplasmic calcium in the parasite, thus triggering exocytosis of adhesins that are needed for strengthening attachment and promoting cell entry.

T. gondii possess several stores of calcium, including the acidocalcisomes, mitochondria, and the endoplasmic reticulum(Moreno and Zhong, 1996). Whether all or some of these stores are released during invasion is unknown. Recent studies have identified IP3 as a second messenger that mediates increases in intracellular calcium in T. gondii following artificial stimuli (Lovett et al., 2002). T. gondii also responds to caffeine and ryanodine, suggesting apicomplexans also contain a ryanodine-like response channel for regulating release of intracellular calcium (Lovett et al., 2002).

Although previous studies have indicated an important role for calcium during parasite invasion of host cells, the source of calcium and whether it acts within the parasite or the host cell remain unresolved. To resolve this important issue, we examined the contribution of intracellular and extracellular calcium sources during parasite motility and invasion into host cells. Additionally, we monitored dynamic calcium changes in parasites and host cells during invasion using the calcium sensitive indicator fluo-4.

T. gondii strain RH (ATCC# 50838) was propagated as tachyzoites in human foreskin fibroblast cells as previously described(Lovett et al., 2002). Freshly harvested parasites were purified by passage through a 23-gauge needle and filtration through three micron Nuclepore membranes (Whatman, Clifton NJ). All cultures tested free of Mycoplasma using the GenProbe™mycoplasma detection system (GenProbe, San Diego, CA).

EGTA (Sigma, St Louis, MO) stock solutions (2×) were prepared in low calcium Ringer's (155 mM NaCl, 3 mM KCl, 5 mM MgCl₂, 3 mM NaH₂PO₄, 10 mM HEPES, 10 mM glucose, 0.1 mM EGTA) and adjusted to pH 6.0 or 7.2. BAPTA-AM [1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid, sodium salt] and BAPTA were purchased from Calbiochem (San Diego, CA),and fluo-4 AM was purchased from Molecular Probes (Eugene, OR). All other reagents were analytical grade. Stock solutions of BAPTA were prepared in distilled water, and BAPTA-AM solutions were made in DMSO.

SDS polyacrylamide gel electrophoresis and western blotting SDS-PAGE was performed in 7% minigels under reducing conditions and transferred to nitrocellulose as previously described(Carruthers and Sibley, 1999). Western blotting was performed with mouse anti-TgMIC2 monoclonal antibody 6D10(ascites, 1:10,000) (Carruthers and Sibley, 1999) and rabbit polyclonal anti-TgACT1 actin antibody(1:10,000) (Dobrowolski et al.,1997). Blots were detected using enzyme conjugated secondary antibodies (Jackson ImmunoResearch Labs, West Grove, PA) combined with SuperSignal ECL (Pierce, Rockford, IL).

Glass coverslips (13 mm round) were incubated overnight with complete media(Dulbecco's modified Eagle's medium, 10 mM HEPES, 44 μM sodium bicarbonate,10% fetal bovine serum, 2 mM glutamine, 20 μg/ml gentamicin). Parasites were resuspended in low calcium Ringer's. BAPTA-AM was loaded into host cells or parasites for 10 minutes at 18°C, while BAPTA or EGTA were added to parasites immediately prior to use to minimize leaching of intracellular calcium stores. Parasites were added to coverslips previously washed with low calcium Ringer's and incubated for 15 minutes in a 37°C water bath. After removal from the water bath, coverslips were washed twice with warm low calcium Ringer's and fixed with 2.5% formalin for 20 minutes at 4°C. Coverslips were stained with Alexa Fluor® 488 (Molecular Probes)conjugated monoclonal antibody DG52 against SAG1 (1:1000) and observed using a Zeiss Axioskop (Carl Zeiss) equipped with epifluorescence and phase contrast optics. Images were captured using an ORCA-ER digital cooled CCD camera(Hamamatsu Photonics K.K., Hamamatsu City, Japan) at 63× magnification controlled by Openlab v3.0.8 imaging software (Improvision, Lexington,MA).

Parasites were resuspended at 10⁸/ml in low calcium Ringer's. BAPTA-AM was loaded for 10 minutes at 18°C, while BAPTA or EGTA were added immediately prior to stimulation to minimize leaching of intracellular Ca²⁺ stores. Secretion was stimulated by addition of ethanol to 1%and incubation at 37°C for 2 minutes(Carruthers et al., 1999b). An unstimulated sample, that was not raised to 37°C and to which ethanol was not added, served as a negative control for background. Parasite supernatants were separated from pellets by centrifugation at 4°C and run on 7%SDS-PAGE gels. Each gel contained standard dilutions of a parasite cell lysate corresponding to percentages of the total number of parasites used. Inadvertent lysis of parasites was monitored by the release of constitutively expressed actin, which is 98% globular in T. gondii(Wetzel et al., 2003).

Glass-bottom 10 mm microwells (MatTek Corp, Ashland, MA) were seeded with host cells to a final density of 60 to 80% confluency. Parasites were resuspended at 10⁸ per ml in low calcium Ringer's. BAPTA-AM was loaded into monolayers of host cells or parasites for 10 minutes at room temperature, while BAPTA or EGTA were added to parasites immediately prior placement on coverslips to minimize leaching of intracellular calcium stores. Microwells were washed with low calcium Ringer's before the addition of parasites and incubated for 15 minutes in a 37°C water bath. After removal from the water bath, coverslips were washed twice with warm low calcium Ringer's and fixed with 2.5% formalin for 20 minutes at 4°C. Coverslips were blocked for 30 minutes with PBS containing 5% (v/v) fetal bovine serum,and 5% (v/v) normal goat serum. Coverslips were sequentially stained with Alexa Fluor® 594 (Molecular Probes) conjugated monoclonal antibody DG52 against SAG1 (1:1000) to visualize extracellular parasites, permeabilized with 0.01% saponin, then stained with Alexa Fluor® 488 conjugated monoclonal antibody DG52 against SAG1 (1:1000)(Håkansson et al., 1999)to visualize intracellular parasites. Coverslips were observed with a Zeiss Axioskop microscope and a 63× objective. Results of three independent experiments were compiled. For each experimental condition, five fields with approximately 20 parasites each were scored for the number of parasites intracellular, extracellular but touching a fibroblast, or extracellular but not in contact with host cells.

Freshly egressed parasites were loaded with 300 nM Fluo-4 for 5 minutes at 37°C, centrifuged at 400 g for 5 minutes at room temperature, and resuspended in 37°C Ringer's plus calcium (155 mM NaCl, 3 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 3 mM NaH₂PO₄, 10 mM HEPES, 10 mM glucose). Sub-confluent monolayers of host cells were loaded by resuspending in Ringer's plus calcium containing 2 μM fluo-4 and incubating at 37°C 15 minutes. After incubation, parasites or cells were washed and resuspended in Ringer's plus calcium. Parasites were added to the culture dishes containing host cells dishes and observed on a Zeiss Axiovert equipped with phase contrast,epifluorescence microcopy and a temperature-controlled stage (Medical Systems,Greenvale, NY) held at 37°C. Parasite motility was observed within minutes after placing the dish on the heated stage and images were recorded over a period of up to 15 minutes. Time-lapse images were collected at two frames per second under low-light illumination using an ORCA-ER digital cooled camera at 63× magnification and 640×480 pixels controlled by Openlab v3.0.8 imaging software (Improvision, Lexington, MA). Phase and fluorescent images were cropped, merged, and saved as Quicktime movies (v5.0).

To determine whether depletion of extracellular calcium affected parasite invasion into host cells, we used BAPTA or EGTA to chelate residual calcium in Ringer's solution that lacked any direct source of calcium (low calcium Ringer's). Invasion assays were performed by incubating parasites with host cells grown on glass coverslips. Parasites were scored as attached to host cells versus intracellular on the basis of differential staining with antibodies before and after addition of detergent, respectively (see Materials and Methods). In order to examine the role of extracellular calcium, parasite attachment and invasion were evaluated in media containing 1 mM BAPTA, a diluent control, or Ringer's containing 5 mM CaCl₂. Parasite attachment and invasion of host cells was similar under all three conditions(Fig. 1).

Previous reports indicate that chelation of calcium with high levels of EGTA decreased parasite invasion (Pezzella et al., 1997). However, our initial attempts to replicate this finding indicated that dilution of a 20× stock solution of EGTA (pH 8.0)into complete culture medium resulted in acidification. Nitrilo nitrogens in EGTA bind protons with pK_as of 8.96 and 9.58(Martell and Smith, 1974),leading to pH-sensitive affinity for calcium. Consequently, the dilution of EGTA stock solutions into calcium-containing medium releases 2 mol H⁺ for each mol Ca²⁺ bound, thus resulting in acidification. Previous studies have indicated that acidic pH is detrimental to parasite invasion (Sibley et al.,1985). To determine the relative contribution of acidification versus calcium chelation, we made 5 mM EGTA solutions in low calcium Ringer's and adjusted them to pH 6.0 or pH 7.2. These solutions were used for attachment and invasion assays shown in Fig. 1. Incubation of parasites in the presence of 5 mM EGTA at a medium pH of 7.2 did not affect attachment or invasion into host cells. In contrast, parasite invasion in the presence of 5 mM EGTA at a medium pH of 6.0 was substantially inhibited, although attachment appeared unchanged (Fig. 1). Collectively, these results indicate that although decreased pH reduces invasion, chelation of extracellular calcium has no effect.

We also examined the effect of chelating intracellular calcium on the attachment and entry of parasites into host cells. Loading of the cell-permeant chelator BAPTA-AM into host cells prior to invasion had no effect when compared to a diluent control(Fig. 1). Thus, host cell calcium does not appear necessary for invasion by T. gondii. Alternatively, preloading parasites with BAPTA-AM prior to invasion inhibited both attachment and invasion of host cells(Fig. 1)(Carruthers et al., 1999a).

Parasite invasion into host cells is an active process that depends on parasite motility (Dobrowolski and Sibley,1996). Parasites exhibit contact-dependent gliding motility that results in formation of membrane trails on the substratum(Håkansson et al.,1999). We examined whether the presence or absence of extracellular calcium affected motility as monitored by trail formation. Parasites were treated with BAPTA-AM, BAPTA, or EGTA before addition to coverslips where they were allowed to glide for 15 minutes. After washing and fixation, trails were visualized by staining of the parasite surface protein SAG1. BAPTA-AM pretreatment inhibited trail formation markedly versus a diluent control (Fig. 2). In contrast, BAPTA treatment had no apparent effect, and a similar density of trails was observed as in the diluent treated control(Fig. 2B,C). Addition of 5 mM calcium did not alter motility compared to the diluent control(Fig. 2D), yet trails appeared discontinuous and punctuated upon close examination. Interestingly, treatment of parasites with EGTA in either pH 6.0 or pH 7.2 media also did not affect trail formation (Fig. 2E,F).

To determine the source of calcium important for influencing microneme secretion, we treated parasites with 50 μM BAPTA-AM to chelate intracellular calcium, or used 1 mM BAPTA or 5 mM EGTA to chelate extracellular calcium. Parasites were then placed at 37°C to stimulate secretion. Cells were removed by centrifugation and secretion was evaluated by detecting MIC2 in the supernatants by western blotting(Fig. 3). MIC2 is a convenient marker for microneme secretion because the secreted 95-100 kDa form (sMIC2) is released into the supernatant, whereas the cell-associated 115 kDa form(cMIC2) remains in the parasite pellet(Carruthers et al., 2000). Constitutive expression of actin was used to control for inadvertent lysis of the parasites during the experiment and was typically between 1% and 5%, as indicated by comparison with dilutions of parasite cell standards (cell stds)(Fig. 3).

Secretion of MIC2 was inhibited by pre-loading of parasites with BAPTA-AM in comparison with a DMSO diluent control, indicating an increase in intracellular calcium is necessary for secretion(Fig. 3). Treatment with non-permeant BAPTA did not affect MIC2 secretion, indicating that extracellular calcium is not required. Interestingly, addition of excess (5 mM) calcium did not provoke additional secretion, providing evidence that extracellular calcium does not increase the efficiency of secretion(Fig. 3). Secretion of MIC2 after treatment with EGTA at pH 7.2 did not compromise in the amount of protein released (Fig. 3). In contrast, treatment with EGTA at pH 6.0 led to a reduction in the amount of MIC2 secreted when compared with the control(Fig. 3). Addition of 1%ethanol to stimulate secretion (Carruthers et al., 1999b) was used as a positive control, whereas unstimulated parasites that were not placed at 37°C were used as a negative control (Fig. 3). Collectively these results show that intracellular calcium in the parasite is essential for microneme secretion, whereas extracellular calcium plays little role in this process.

Although our studies using BAPTA-AM indicated that host cell calcium was not required for parasite invasion, we were interested in determining whether host cell calcium changed during the process of parasite invasion. We monitored host cell calcium during parasite invasion using the indicator fluo-4, which is a qualitative indicator of intracellular calcium in the range of 0.1 to 1 μM (Gee et al.,2000). Monolayers of human fibroblasts were loaded with 2 μM fluo-4 AM and the excess dye washed away, leaving fluo-4 localized throughout the cytoplasm. Images were taken at ∼0.5 second intervals, alternating between phase contrast and fluorescence modes. Unstimulated fibroblasts showed calcium increases lasting approximately six seconds, during which a wave of increased fluorescence spread across the cell (Movie 1, available at jcs.biologists.org/supplemental). The magnitude of these changes was dramatic, with increased signals of fluo-4 being up to 250% higher than background levels (data not shown, n=3 examples monitored over a 15-30 second period). These events were rare,occurring about once every 10 minutes (data not shown). Fibroblast calcium fluxes were likely to be associated with normal signaling processes, as the cells appeared healthy both before and after transients occurred.

To examine changes that occur during invasion, parasites were added to fluo-4-loaded fibroblasts, and phase and fluorescent images were acquired at 0.5 second intervals between each set of images. Host cell calcium levels remained relatively constant during invasion as judged by the fact that only small, localized increases in fluorescence were observed(Fig. 4B,C; Movie 2, available at jcs.biologists.org/supplemental). The magnitude of changes in the fluo-4 signal in host cells was ≤20%different from background (data not shown, n=3 separate examples monitored for 15-30 seconds), in contrast to the rather dramatic natural changes in host cell calcium that occasionally occurred in the absence of invasion (see Fig. 4A). These observations indicate that parasite invasion occurs without an active participation or response of calcium from the host cell.

Previous studies have suggested that increases in parasite intracellular calcium occur during contact with the host cell(Vieira and Moreno, 2000);however, the kinetics of calcium increase were not examined. We predicted that parasites entering cells would show increases in calcium during attachment and invasion. We used fluo-4 AM loaded tachyzoites to visualize qualitative changes in parasite calcium during gliding motility and host cell invasion.

Gliding was associated with brightly fluorescent parasites that underwent periodic cycles of increase followed by dampening of the signal(Fig. 5A; Movie 3, available at jcs.biologists.org/supplemental). Calcium fluxes correlated with parasites that were actively motile and often preceded a burst of motility; whereas dimly stained cells remained immotile. Kinetic changes in fluo-4 fluorescence were plotted over time by taking the average pixel intensity of the parasite from successive frames of the video recording (Fig. 6A). The cycles of increased fluorescence were monitored from independent recordings to determine the average time between successive cycles and the relative length of the cycles (Table 1). Typically, individual cycles lasted ∼30 seconds and were closely spaced in sequence. The initial cycle was the brightest with successive cycles becoming dimmer. The drop in intensity correlated with the gradual loss of motility. Irrespective of whether they are loaded with fluo-4, parasites that are examined by phase contrast or epifluorescence illumination typically maintain motility for only one to two minutes. It is uncertain whether this decrease is because of photodamage or loss of essential factors such as energy stores that are necessary to maintain activity.

We also followed the calcium response of fluo-4-labeled parasites while they invaded fibroblasts. Surprisingly, the bright fluorescence exhibited by motile parasites rapidly diminished during the initial steps of invasion(Fig. 5B,C; Movie 4, available at jcs.biologists.org/supplemental). The drop in signal intensity following invasion occurred more rapidly than had been observed with fluxes occurring during gliding motility(Table 1 and Fig. 6B). The decrease in signal was not due to photobleaching, as other brightly labeled parasites in the same field did not display similar decreases in intensity (data not shown). Intracellular parasites remained dim even when analyzed for extended periods of time (up to five minutes, data not shown).

Calcium is a widespread intracellular second messenger that is known to play a critical role in the invasion of mammalian cells by a number of microorganisms. Several intracellular pathogens modify host cell calcium signaling during invasion. Examples include Trypanosoma cruzimodulation of fibroblast signaling(Rodriguez et al., 1995),activation of calcium-dependent interleukin-8 release from endothelial cells by Salmonella typhimurium (Geritz et al., 2000), and inhibition of macrophage calcium signaling by Mycobacterium tuberculosis (Malik et al., 2000). These examples from a wide variety of pathogens led us to consider whether changes in host cell calcium contribute to invasion of T. gondii.

We addressed the role of host cell calcium by using the intracellular chelator BAPTA-AM. Chelation of host cell calcium using BAPTA-AM had no effect on parasite attachment or invasion. The concentration of BAPTA-AM used should have effectively chelated intracellular calcium in the fibroblasts based on other studies using lower concentrations of this compound on fibroblasts(Scheidegger et al., 1999; Shahrestanifar et al., 1999),and the observation that the cell monolayer began to detach from the substrate if loading times were prolonged (data not shown). Thus, unlike the entry of T. cruzi into mammalian cells(Rodriguez et al., 1995),changes in host cell calcium do not appear to be required for efficient entry of T. gondii.

Consistent with the inhibitor studies, direct monitoring of intracellular calcium in host cells indicated that only weak local calcium transients occur during parasite invasion. Such calcium transients may be associated with membrane perturbation by the parasite that was previously observed as a spike in capacitance of patch-clamped host cells(Suss-Toby et al., 1996). Alternatively, it could indicate activation of local host signaling pathways required for membrane repair. In either case, the very low response observed in the host cell parallels our finding that host cell calcium was not required for parasite invasion. These findings support previous observations that indicate a relatively passive role of the host cell during parasite invasion. For example, unlike phagocytosis or entry of many bacterial pathogens,invasion of mammalian cells by T. gondii does not evoke changes in the host cell cytoskeleton, tyrosine phosphorylation or membrane ruffling(Morisaki et al., 1995; Håkansson et al.,1999).

In contrast to the lack of an essential role for host cell calcium,chelation of intracellular calcium in the parasite using BAPTA-AM treatment prevented microneme secretion and dramatically reduced both motility and invasion into host cells. Since levels of calcium in the parasite appeared critical to the process of invasion, we sought to identify whether extracellular calcium was necessary for parasite invasion into host cells. A previous report that chelation of extracellular calcium by EGTA inhibited T. gondii invasion hypothesized that a flux of calcium from outside to inside the parasite was required during invasion(Pezzella et al., 1997). However, as we show here, this effect was likely to be as a result of the acidifying affects of EGTA on the medium. When EGTA containing medium was adjusted to pH 7.2, or when the non-acidifying chelator BAPTA was used, no effect of chelating extracellular calcium was observed. Furthermore, adding excess calcium had no influence on parasite invasion into host cells. We therefore conclude that extracellular calcium is not required for invasion of cells by T. gondii. These results indicate that a calcium-induced-calcium-entry pathway, where initial increases in cytosolic calcium result in influx of extracellular calcium from outside the cell(Berridge et al., 2000a; Berridge et al., 2000b), is not essential for parasite invasion of host cells.

To explore the role of calcium on T. gondii secretion and motility, we used a similar approach to look at the role of extracellular versus intracellular calcium sources in mediating these events. Chelation of intracellular calcium prevented microneme release, consistent with a previous report (Carruthers and Sibley,1999), whereas chelation of extracellular calcium had no effect. Likewise, although chelation of intracellular calcium with BAPTA-AM prevented motility, chelation of extracellular calcium had little effect on motility. Collectively, these studies indicate that intracellular calcium in the parasite controls microneme secretion and motility.

Monitoring of intracellular calcium in the parasite revealed oscillating fluxes in motile parasites, either immediately preceding or during gliding motility. Previous models suggested that calcium increases would correlate with contact with the host cells immediately prior to invasion. Instead, our observations indicate that oscillating increases in parasite intracellular calcium occur during gliding, regardless of contact with host cells. These changes in calcium may reflect an important role for this second messenger in regulating motility and/or release of adhesins that are necessary for attachment to the substratum. Calcium transients that were observed during motility halted abruptly and the fluorescent calcium signal dramatically decreased during invasion of the parasite into host cells. This finding was unexpected, since previous studies have shown increased cytoplasmic calcium in parasites that were associated with host cells(Vieira and Moreno, 2000). These previous findings may have reflected parasites moving while on top of host cells, which could explain their increased calcium levels. Collectively our findings suggest a more complex model wherein increases in intracellular calcium are needed for motility, while cell entry does not require continued elevation of intracellular calcium in the parasite. Furthermore, the rapid shut off of intracellular calcium in invading parasites may be important for downregulating cellular processes, such as microneme secretion and motility,once invasion is complete.

Despite the central importance of calcium in parasite secretion and motility, relatively little is known about the intracellular pools of calcium or how the release of calcium is regulated. The oscillating patterns of calcium in gliding parasites suggest that intracellular release channels and re-uptake mechanisms are highly active in the parasite. In most cells, the major pool of mobilizable calcium is the endoplasmic reticulum (ER) where release is mediated by IP₃ or ryanodine-type channels(Berridge et al., 2000a; Berridge et al., 2000b). T. gondii responds to agonists/antagonists of both IP₃-type release channels and ryanodine-type channels, suggesting both pathways could operate in the parasite (Lovett et al.,2002). Typically, ER stores are refilled by calcium transporters known as SERCA-type ATPases, which can be inhibited by thapsigargin(Thastrup et al., 1989). Protozoan parasites, including T. gondii, are sensitive to this inhibitor, and yet they also have thapsigargin-insensitive calcium pools(Carruthers et al., 1999). SERCA-type ATPases have not been described in T. gondii, although several putative orthologs are present in the genome databases for T. gondii(http://ToxoDB.org/ToxoDB.shtml). T. gondii also contains plasma membrane type Ca-ATPases, one of which has recently been shown to localize both to the plasma membrane and a unique organelle called the acidocalcisome (Luo et al., 2001). These acidic, calcium-rich organelles appear to play an important role in polyphosphate metabolism and serve as a storage site for calcium in protozoan cells (Docampo and Moreno, 2001; Rodrigues et al., 2002).

Our investigations demonstrate that intracellular calcium plays a crucial role in controlling protein secretion and motility in the parasite T. gondii. The complete reliance of the parasite on its own intracellular calcium stores for motility presents a potential target for intervention. Further studies might reveal sufficient differences between parasite and host calcium homeostasis mechanisms that can be used to disrupt calcium-regulated secretion and/or motility, and thereby prevent infection.

This work was partially supported by a grant to L.D.S. from the NIH (#34036). J.L.L. was partially supported by an Institutional Training Grant to Washington University (AI017172-19). L.D.S. is the recipient of a Scholar Award in Molecular Parasitology from the Burroughs Wellcome Fund. We thank Steve Beverley, Dan Goldberg, Andy Pekosz, Paul Schlessinger, and Tom Steinberg for advice and critical input, and Jeff Diffenderfer for expert technical assistance.