To define egress by bradyzoites, we induced the differentiation of tachyzoites to bradyzoites by culture in human foreskin fibroblast (HFF) cells at alkaline pH (8.2) for 7 days. We treated both tachyzoite cultures and in vitro-differentiated cysts with Ca²⁺ ionophore A23187 to trigger egress from parasitophorous vacuoles (PVs) or bradyzoite cysts, as detected by indirect immunofluorescence assay (IFA) or time-lapse video microscopy. We observed that A23187 induced complete egress of tachyzoites from disrupted PVs while only a few bradyzoites were released from cysts that remained largely intact (Figure 1A). This result was also confirmed by time-lapse video microscopy using the ME49 BAG1-mCherry strain either grown as tachyzoites (Figure 1—video 1) or bradyzoites (Figure 1—video 2). We quantified the percentage of tachyzoites or bradyzoites that were released during egress in response to A23187 or the agonist zaprinast, which is a cGMP-specific phosphodiesterase (PDE) inhibitor that activates PKG-mediated Ca²⁺ signaling, leading to egress. In contrast to tachyzoites, we found significantly lower egress rate of bradyzoites in response to A23817 or zaprinast (Figure 1B). To examine the behavior of released parasites, we determined the maximum egress distance that parasites moved away from the original vacuole or cyst following egress. Tachyzoites migrated much further than bradyzoites after induced egress (Figure 1C). Bradyzoites also moved more slowly than tachyzoites (Figure 1D), as shown by quantification of their trajectories from time-lapse video microscopy images. Taken together, these findings indicate that egress by bradyzoites in response to Ca²⁺ ionophore or zaprinast is incomplete and restricted.

Egress by parasites requires Ca²⁺-stimulated microneme secretion. To examine the reason for inefficient egress by bradyzoites, we monitored microneme secretion by quantitative secretion analysis of MIC2 fused with Gaussia luciferase (Gluc). The MIC2-Gluc reporter was randomly integrated into the genome of the BAG1-mCherry strain (Figure 2A). IFA revealed that MIC2-Gluc was expressed and localized to the apical pole in tachyzoites and bradyzoites induced for 7 days at pH 8.2 in vitro (Figure 2B). We also confirmed the expression of MIC2-GLuc and MIC2-associated protein M2AP, which forms a protein complex with MIC2 (Huynh et al., 2003), by western blotting. Although both proteins were readily detectable, they were expressed at lower levels in bradyzoites compared with tachyzoites (Figure 2C). Furthermore, IFA demonstrated that MIC2 and M2AP were co-localized to the apical region in bradyzoites, consistent with being located in micronemes (Figure 2—figure supplement 1A). Next, to enrich highly purified MIC2-GLuc reporter parasites, BAG1-mCherry MIC2-GLuc strain tachyzoites, and bradyzoites liberated from cysts produced by cultivation for 7 days at pH 8.2 in vitro, were sorted by fluorescence-activated cell sorting (FACS) (Figure 2—figure supplement 1B). FACS-sorted tachyzoites and bradyzoites were treated with zaprinast or ionomycin, a Ca²⁺ ionophore that induces release of Ca²⁺ from the ER (Beeler et al., 1979). Bradyzoites secreted much less MIC2-Gluc protein compared to tachyzoites in response to Ca²⁺ agonists, zaprinast and ionomycin, as shown by Gaussia luciferase assays performed on excretory-secretory antigen (ESA) fractions collected following stimulation (Figure 2D). To further investigate the process of microneme secretion by bradyzoites, we randomly integrated an mCherry secretion reporter, based on the signal peptide sequence of ferredoxin-NADP(+)-reductase (FNR-mCherry), into the genome of BAG1-EGFP parasites (Figure 2E). The FNR-mCherry reporter is an improved version of DsRed reporter that is secreted constitutively and released from the PV surrounding tachyzoites following the discharge of PLP1 in response to Ca²⁺ agonists (Kafsack et al., 2009). Then, we monitored the permeabilization of PV membrane surrounding either tachyzoite vacuoles or in vitro-differentiated bradyzoites after stimulation with A23187 based on the diffusion of FNR-mCherry using time-lapse fluorescence video microscopy. Consistent with previous reports (Lourido et al., 2010), we observed that A23187 stimulated fast leakage of FNR-mCherry from the PV surrounding tachyzoites (Figure 2F, top panel, and Figure 2—video 1). However, FNR-mCherry was not released from the cyst after A23187 stimulation (Figure 2F, middle panel, and Figure 2—video 3). As a control to confirm that the FNR-mCherry was indeed secreted into the lumen of the cyst matrix, we treated cysts with trypsin to release bradyzoites. Once the cyst wall was digested, the FNR-mCherry dissipated rapidly, or was digested, confirming that it was present in the matrix of the cyst and not trapped in the parasite (Figure 2F, bottom panel, and Figure 2—video 2). These data were also confirmed by plotting FNR-mCherry fluorescence intensity changes vs. time for tachyzoites vs. intact or trypsin-treated cysts (Figure 2G). These findings demonstrate dampened microneme secretion by bradyzoites, which may explain their incomplete egress.

To investigate Ca²⁺ signaling in bradyzoites, we established a dual fluorescent reporter system containing constitutively expressed GCaMP6f and mCherry under the control of bradyzoite stage-specific promoter BAG1 (Figure 3A). Using this system, both tachyzoites and bradyzoites express the same levels of GCaMP6f, while only bradyzoites express mCherry, allowing specific monitoring of Ca²⁺ signals in both stages. To confirm the differentiation stage of bradyzoites expressing this dual reporter, we monitored BAG1 and SAG1 expression in bradyzoites induced for different times at alkaline pH by IFA. We observed a gradual increase in the percentage of BAG1-positive and SAG1-negative bradyzoites (mature bradyzoites) from 3 days to 7 days after induction. Based on this criterion, ~70% of parasites were mature bradyzoites in cysts that were induced for 7 days (Figure 3—figure supplement 1A and B), and we chose this time point for further studies. Next, we compared the response of BAG1-mCherry GCaMP6f reporter parasites that were grown as tachyzoites to those induced to form bradyzoites by cultivation in HFF cells for 7 days at pH 8.2 in vitro after treatment with Ca²⁺ ionophore A23187. Ionophore treatment induced rapid and high-level increases in GCaMP6f fluorescence in tachyzoites but delayed and much weaker responses in bradyzoites as monitored by time-lapse video microscopy (Figure 3B, Figure 3—video 1, Figure 3—video 2). To rule out an effect of differences in expression level of GCaMP6f during differentiation, we measured fluorescence intensities of GCaMP6f and BAG1-mCherry in different parasites within the same cyst. We observed no correlation between the signals of BAG1-mCherry and the basal expression of GCaMP6f in the absence of ionophore stimulation, indicating that the low responses to ionophore were not due to expression differences in the Ca²⁺-sensitive reporter (Figure 3—figure supplement 1C and D). To determine the effect of bradyzoite development on Ca²⁺ signaling, we treated intracellular tachyzoites vs. bradyzoites induced by cultivation in HFF cells at pH 8.2 in vitro for 4–7 days and quantified time of each tachyzoite vacuole or bradyzoite cyst to reach Ca²⁺ peak level after addition of A23187 ionophore by video microscopy. Increasing time of bradyzoites development was associated with progressively longer times to reach peak fluorescence of GCaMP6f (Figure 3C). Time-lapse recording of GCaMP6f fluorescence intensity ratio changes (F/F₀) showed delayed Ca²⁺ increase and lower fold changes in bradyzoites compared with tachyzoites in response to A23187 stimulation (Figure 3D). Zaprinast also elicited slower Ca²⁺ increases and lower fold changes in bradyzoites compared with tachyzoites even in the presence of extracellular Ca²⁺ (Figure 3E). To better characterize Ca²⁺ responses of bradyzoites, we performed live video imaging using spinning disc confocal microscopy to distinguish individual bradyzoites within in vitro-differentiated cysts and identify motile bradyzoites within cysts by comparing consecutive images (Figure 3F). Motile bradyzoites were also observed to have higher GCaMP6f signals and these typically oscillated over time. In response to Ca²⁺ agonists, intracellular bradyzoites showed reduced percentages of motility compared to tachyzoites (Figure 3G). In summary, Ca²⁺ dynamics are delayed and reduced in bradyzoites in response to Ca²⁺ agonists.

To rule out the possibility that alkaline pH stress used for differentiation resulted in lowered Ca²⁺ signals in bradyzoites, we examined Ca²⁺ signaling in bradyzoites within cysts that formed naturally in differentiated C2C12 myocytes. Differentiated myocytes stained positively for skeletal myosin and facilitated the development of bradyzoites, as shown using the bradyzoite stage-specific protein BAG1 (Figure 4A). We tested Ca²⁺ responses of bradyzoites formed in muscle cells using the dual fluorescent reporter GCaMP6f BAG1-mCherry parasites in response to A23187 or zaprinast by time-lapse video recording. Time-lapse imaging showed slow increase of GCaMP6f fluorescence in response to A23187 in tissue cysts formed in C2C12 myocytes (Figure 4B). Both the rate of increase and the maximum amplitude of the GCaMP6f signal were much lower in bradyzoites differentiated in myocytes compared to tachyzoites cultured in undifferentiated myoblasts (Figure 4C). The time to reach the peak GCaMP6f fluorescence was also delayed in bradyzoites formed in C2C12 myocytes compared with tachyzoites grown in myoblasts (Figure 4D). Bradyzoites cultured in C2C12 myocytes show significantly lower motility in response to A23187 and zaprinast when compared with tachyzoites (Figure 4E).

To further examine Ca²⁺ signaling in bradyzoites, we harvested tissue cysts containing BAG1-mCherry GCaMP6f bradyzoites from the brains of chronically infected CD-1 mice and investigated their responses ex vivo. Video microscopy of ex vivo tissue cysts showed slow increases in GCaMP6f fluorescence in response to A23187 or zaprinast (Figure 4F). The ratio of GCaMP6f fluorescence changes vs. time (F/F₀) from bradyzoites within ex vivo cysts demonstrated lower and slower changes, consistent with lower Ca²⁺ levels, compared with extracellular tachyzoites in response to Ca²⁺ agonists (Figure 4G). In comparing the response of extracellular, ex vivo tissue cysts (Figure 4F and G) to intracellular cysts formed during infection of C2C12 myocytes (Figure 4B and C), it was evident that the extracellular cysts respond somewhat faster, albeit still much slower than tachyzoites. This intermediate level of response was also seen in in vitro-differentiated tissue cyst (produced by cultivation in HFF cells at pH 8.2 for 7 days) that were liberated from HFF cells and tested in vitro (Figure 4—figure supplement 1). Next, we measured the percentage of motile and egressed bradyzoites within ex vivo tissue cyst treated with A23187 and zaprinast. Strikingly, no egressed bradyzoites were observed, although all the bradyzoites within ex vivo cysts became motile after stimulation (Figure 4H, Figure 4—video 1, Figure 4—video 2). Taken together, these findings indicate that bradyzoites formed spontaneously in muscle myocytes and within ex vivo cysts from chronically infected mice display dampened Ca²⁺ dynamics when treated with Ca²⁺ agonists.

The cyst wall surrounding bradyzoites may restrict access to Ca²⁺ agonists and hence dampen signals from GCaMP6f in response to Ca²⁺ agonists in the studies described above. To test this possibility, we monitored GCaMP6f fluorescence changes in extracellular bradyzoites vs. tachyzoites of the BAG1-mCherry GCaMP6f strain by live imaging. Bradyzoites were induced by cultivation in HFF cells at pH 8.2 for 7 days and liberated from cysts by trypsin treatment, followed by washing and resuspension for analysis. We also observed slower increases in GCaMP6f fluorescence intensity in bradyzoites (Figure 5—video 2) compared with tachyzoites (Figure 5—video 1) in response to A23187 (Figure 5A). Quantitative analysis of Ca²⁺ fluorescence changes (F/F₀) after stimulation by A23187 and zaprinast showed slower Ca²⁺ responses in extracellular bradyzoites when compared to tachyzoites (Figure 5B). To confirm that extracellular bradyzoites were viable after liberation from in vitro-cultured cysts by trypsin treatment, we utilized SYTOX Red, which is a DNA dye excluded by intact membranes of viable cells. In contrast to bradyzoites that were formaldehyde-fixed as a positive control, extracellular bradyzoites were not stained by SYTOX after the liberation from in vitro cysts (Figure 5C), indicating that they were still viable after trypsin treatment.

We hypothesized that bradyzoites might have dampened GCaMP6f responses because they fail to release Ca²⁺ from intracellular stores. We tested Ca²⁺ responses of BAG1-mCherry and GCaMP6f -expressing bradyzoites and tachyzoites treated with ionomycin, which releases Ca²⁺ mainly from the ER (Beeler et al., 1979), thapsigargin, which inhibits SERCA-type Ca²⁺-ATPase causing an increase of cytosolic Ca²⁺ due to uncompensated leakage from the ER (Thastrup et al., 1990), and NH₄Cl, an alkalizing reagent that releases Ca²⁺ from acidic stores like acidocalcisomes (Moreno and Zhong, 1996). Both ionomycin and thapsigargin induced delayed and lower amplitude changes in GCaMP6f fluorescence in bradyzoites vs. tachyzoites as shown by plotting fluorescence intensity fold changes (F/F₀) vs. time (Figure 5D), indicative of lower ER-stored Ca²⁺. In contrast, bradyzoites treated with NH₄Cl showed no meaningful change in GCaMP6f fluorescence, suggesting that they lack mobilizable acidic Ca²⁺ (Figure 5D). To rule out the possibility that the Ca²⁺ indicator GCaMP6f is less sensitive in bradyzoites due to some intrinsic defect, we loaded BAG1-mCherry-expressing tachyzoite or bradyzoites with the Ca²⁺-sensitive vital dye Fluo-8 AM and used these cells for imaging. Fluo-8 AM-labeled bradyzoites displayed dampened Ca²⁺ signaling after stimulation by ionomycin, thapsigargin, or NH₄Cl, relative to tachyzoites that responded normally (Figure 5E). Collectively, these findings indicate that bradyzoites are less able to mobilize Ca²⁺ from the ER and acidic stores in response to agonists.

To more precisely compare Ca²⁺ levels in tachyzoites and bradyzoites, we constructed a ratiometric fluorescence reporter by coexpression of GCaMP6f with blue fluorescent protein (BFP) mTagBFP2 linked by a P2A split peptide (Figure 6A, Figure 6—figure supplement 1A, Figure 6—figure supplement 1B). Because both proteins are coexpressed from the same promoter, the mTagBFP2 serves as a control for expression level as mTagBFP2 is nonresponsive to Ca²⁺ levels (Cranfill et al., 2016). Live fluorescence microscopy showed simultaneous expression of GCaMP6f and mTagBFP2 in tachyzoites, and additionally mCherry in bradyzoites (Figure 6B). Equal expression of GCaMP6f (His tag) and mTagBFP2, as well as separation of tachyzoites and bradyzoite populations (detected with SAG1 and BAG1, respectively), was validated by western blotting (Figure 6C). To compare Ca²⁺ basal levels, we quantified the fluorescence intensity ratio F_GCaMP6f/F_mTagBFP2 of intracellular and extracellular tachyzoites and bradyzoites in EC buffer with or without Ca²⁺. We observed significant reductions in the fluorescence intensity ratio of both intracellular and extracellular bradyzoites relative to tachyzoites (Figure 6D), indicative of lower resting Ca²⁺ levels in bradyzoites. We next compared Ca²⁺ dynamics of intracellular tachyzoites and bradyzoites in response to Ca²⁺ agonists ionomycin, NH₄Cl, and thapsigargin. Changes in the fluorescence of GCaMP6f were much slower and of lower amplitude in bradyzoites relative to tachyzoites (Figure 6E). We also observed lower resting Ca²⁺ and peak levels in extracellular bradyzoites compared to tachyzoites (Figure 6F), indicating lower activity or expression of cytoplasmic influx mechanisms like the PM Ca²⁺ entry or ER Ca²⁺ release channels. To understand the molecular basis for the reduced stored Ca²⁺ and responses in bradyzoites, we performed real-time PCR to compare mRNA expression levels of several Ca²⁺ transporters and channels. Included in this list are TgSERCA (Nagamune et al., 2007), which is the molecular target of thapsigargin and transfers Ca²⁺ from the cytosol of parasites to ER, TgA1 (Luo et al., 2004), which plays important roles in the accumulation of Ca²⁺ in the acidocalcisome and other acidic stores, TgTRPPL-2 (Márquez-Nogueras et al., 2021), which is a TRP channel key for Ca²⁺ influx into the cytosol, and additional Ca²⁺-related proteins, such as TgPMCA1, TgA2, and the Ca²⁺/H⁺ exchanger (Nagamune et al., 2008). We observed significant reduction in the relative expression level of TgSERCA, TgA1, TgPMCA1, TgA2, Ca²⁺/H⁺ exchanger, and TgTRPPL-2 in bradyzoites compared to tachyzoites (Figure 6G). Taken together, these findings indicate that bradyzoites have lower levels of stored Ca²⁺, which is associated with the overall downregulation of Ca²⁺-related pumps and channels.

To test whether dampened Ca²⁺ signaling would still be sufficient to drive gliding motility of bradyzoites, we treated BAG1-mCherry GCaMP6f-expressing cysts cultured in vitro with trypsin to liberate bradyzoites (Figure 7A). There were no obvious changes in the Ca²⁺ levels nor motility during trypsin treatment and release of bradyzoites that were monitored for 5 min (Figure 7B, Figure 7—video 1). We then extended the trypsin treatment to 10 min to assure complete digestion of the cysts and to allow bradyzoites to settle before imaging by time-lapse microscopy. Again, we observed that the majority of bradyzoites failed to show enhanced GCaMP fluorescence or initiate gliding motility. However, the small number of bradyzoites that did undergo gliding displayed patterns that were highly reminiscent of tachyzoite motility including circular gliding (Figure 7C, Figure 7—video 2). Similar to previous descriptions of oscillating Ca²⁺ patterns in gliding tachyzoites (Lovett and Sibley, 2003), we observed fluctuations of GCaMP6f fluorescence intensities in individual extracellular bradyzoites that exhibited gliding motility (Figure 7D). To evaluate the potential effect of trypsin treatment on Ca²⁺ levels, we incubated extracellular GCaMP6f-expressing tachyzoites with trypsin and observed no difference in the Ca²⁺ responses induced by ionomycin (Figure 7—figure supplement 1A), ruling out the possibility that the low Ca²⁺ response in bradyzoites was caused by trypsin treatment.

To further characterize the role of Ca²⁺ signaling in bradyzoite motility, we treated cells with the Ca²⁺ chelator BAPTA-AM, the PKG inhibitor compound 1, and the CDPK1 inhibitor 3-MB-PP1 to block Ca²⁺ signaling in bradyzoites. All these inhibitors significantly impaired gliding motility of tachyzoites and bradyzoites (Figure 7E and F), confirming a key role of Ca²⁺ signaling in parasite motility. Bradyzoites displayed shorter gliding distance compared with tachyzoites as determined by measurements of trail lengths detected with SAG1 (tachyzoite) or SRS9 (bradyzoites) (Figure 7F). These two surface markers are both members of the cysteine-rich SRS family that are tethered to the surface membrane by a GPI anchor, thus they represent comparable reporters for each stage. In summary, despite having dampened Ca²⁺ stores and reduced responses to agonist when intracellular, extracellular bradyzoites require Ca²⁺ signaling to activate gliding motility.

Following reactivation of tissue cysts, we hypothesized that bradyzoites must replenish their Ca²⁺ and energy stores to meet the demands of cell to cell transmission. To test this idea, we released bradyzoites using trypsin treatment and then treated extracellular bradyzoites with EC buffer with or without Ca²⁺ (1.8 mM) and with or without glucose (5.6 mM) for different times and stimulated the Ca²⁺ responses using ionomycin. Quantitative analysis of Ca²⁺ fluorescence changes (F/F₀) showed that bradyzoites substantially recovered stored Ca²⁺ in the presence of exogenous Ca²⁺ and glucose for 1 hr compared to 10 min (Figure 8A and B). A more modest recovery was observed in the presence of Ca²⁺ but absence of glucose (Figure 8B). We also monitored the effect of recovery of Ca²⁺ pools on microneme secretion by bradyzoites and found that bradyzoites secreted more MIC2-GLuc after being treated with exogenous Ca²⁺ and glucose for 1 hr compared to 10 min (Figure 8C). Next, we investigated the effect of exogenous Ca²⁺ and glucose on gliding motility by bradyzoites. We used time-lapse video microscopy to determine the percentage of extracellular bradyzoites undergoing twirling, circular and helical motility after incubation in EC buffer ± Ca²⁺ and glucose for 10 min vs. 1 hr. Quantitative analysis showed that bradyzoites underwent all forms of gliding motility and substantially recovered gliding motility after incubation with EC buffer containing both Ca²⁺ and glucose for 1 hr, while very few bradyzoites were able to glide following incubation with exogenous Ca²⁺ or glucose alone (Figure 8D).

We reasoned that exogenous glucose could be utilized by parasites to produce ATP via glycolysis or oxidative phosphorylation to maintain a variety of cellular functions. To investigate the ATP source for supporting gliding motility, we treated exogenous bradyzoites in EC buffer containing Ca²⁺ (1.8 mM) with glucose to support glycolysis vs. the glucose analogue 2-deoxy-D-glucose (2-DOG) to block glycolysis (Figure 8E). Alternatively, similar preparations of bradyzoites were incubated with glutamine to provide substrates for the tricarboxylic acid (TCA) cycle or the ATP synthase inhibitor oligomycin A to inhibit oxidative phosphorylation (Figure 8E). Quantitative analysis of percentage of gliding motility showed either glucose or glutamine significantly increased gliding motility by bradyzoites (Figure 8F), indicating that either carbon source can be used to produce ATP for maintaining gliding motility. Either 2-DOG or oligomycin A blocked gliding motility by bradyzoites even in the presence of exogenous glucose or glutamine (Figure 8F), demonstrating that both oxidative phosphorylation and glycolysis are ATP sources for driving gliding motility by bradyzoites. Interestingly, oligomycin A blocked glucose-dependent gliding (Figure 8F), indicative of an essential role of mitochondrial electron transport chain in ATP production for supporting gliding motility of bradyzoites.

To further investigate the energy status of bradyzoites, we utilized reversed-phase high-performance liquid chromatography (RP-HPLC) to measure the adenosine triphosphate (ATP), adenosine diphosphate (ADP), and adenosine monophosphate (AMP) levels in bradyzoites treated with EC buffer containing both Ca²⁺ (1.8 mM) and glucose (5.6 mM) for different times (Figure 8—figure supplement 1A-C). We observed that after the incubation in EC buffer for 1 hr, bradyzoites had significantly higher ATP, ADP, and AMP levels (Figure 8G), demonstrating enhanced ATP production during incubation. In contrast, ATP and ADP levels did not change significantly in tachyzoites despite the increase in AMP level after incubation with exogenous Ca²⁺ and glucose (Figure 8G). The ATP/ADP ratio and energy charge have been widely used to evaluate cellular energy status, which controls the free-energy change for ATP hydrolysis for different cellular functions (Maldonado and Lemasters, 2014). Bradyzoites incubated with EC buffer for 1 hr displayed significantly increased ATP/ADP ratio and energy charge while no obvious changes were observed in tachyzoites (Figure 8H and I), indicating that bradyzoites rapidly recover their energy status following incubation with glucose. Collectively, exogenous Ca²⁺ and glucose altogether activate bradyzoite gliding motility via restoration of ATP levels and Ca²⁺ stores.

T. gondii tachyzoites were passaged in confluent monolayers of HFFs obtained from the Boothroyd Laboratory at Stanford University. The ME49 ∆hxgprt::Fluc type II strain of T. gondii (Tobin et al., 2012) was used as a parental strain for genetic modification. Tachyzoites were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Life Technologies) pH 7.4, supplemented with 10% fetal bovine serum (FBS), penicillin, and streptomycin (Life Technologies) at 37°C in 5% CO₂. Parasite and host cell lines were confirmed to be negative for mycoplasma using an e-Myco plus kit (Intron Biotechnology). For in vitro induction of bradyzoites, parasites were cultured in alkaline medium in ambient CO₂ as described previously (Wang et al., 2015). In brief, infected HFF monolayers were switched to RPMI 1640 medium (MP Biomedicals) buffered to pH 8.2 with HEPES and supplemented with 5% FBS and cultured at 37°C in ambient CO₂, during which time the alkaline medium was changed every 2 days. For spontaneous induction of bradyzoites, C2C12 muscle myoblast cells (ATCC CRL-1772) were maintained in DMEM supplemented with 20% FBS. C2C12 myoblast differentiation and myotube formation were induced in DMEM containing 2% horse serum (Biochrom) by cultivation at 37°C in 5% CO₂ for 5 days. Tachyzoites were inoculated into the differentiated muscle cells and cultured for another 7 days to induce bradyzoite formation, during which time the induction medium was changed every 2 days. For harvesting bradyzoites, infected monolayers were scraped into intracellular (IC) buffer (142 mM KCl, 5 mM NaCl, 1 mM MgCl₂, 5.6 mM D-glucose, 2 mM EGTA, 25 mM HEPES, pH 7.4) and released from cells by serially passing through 18G, 20G, and 25G needles, followed by centrifugation (150 × g, 4°C) for 10 min. The pellet containing cysts was resuspended in IC buffer. Bradyzoites were liberated from cysts by digestion with 0.25 mg/ml trypsin at room temperature for 5 min, followed by centrifugation (150 × g, 4°C) for 10 min. The supernatant containing liberated bradyzoites was further centrifuged (400 × g, 4°C) for 10 min. The pellet containing purified bradyzoites was resuspended in extracellular (EC) buffer (5 mM KCl, 142 mM NaCl, 1 mM MgCl₂, 5.6 mM D-glucose, 25 mM HEPES, pH 7.4) with (1.8 mM Ca²⁺) or without CaCl₂, as indicated for different assays and in the legends.

Tachyzoites were induced to form bradyzoites at a multiplicity of infection (MOI) of 0.5 by culture at pH 8.2 in RPMI 1640 medium under ambient air (low CO₂) for 7 days followed by scraping into PBS containing 0.1% bovine serum albumin (BSA). Cysts were released from host cells by repeated passage through a 23G needle and collected by centrifugation at 150 × g for 10 min. To purify large-scale bradyzoites used for western blotting, pelleted cysts were resuspended in 1 ml PBS containing 10 µl biotinylated Dolichos biflorus agglutinin (DBA) (Vector Laboratories) and 100 µl Pierce Streptavidin Magnetic Beads (Thermo Fisher) and incubated for 1 hr at 4 °C. The beads and absorbed cysts were collected using a magnetic stand and resuspended in 1 ml PBS containing 0.25 mg/ml trypsin and incubated for 10 min at 16°C. The supernatant containing released bradyzoites was separated from the beads and retained. To remove any residual tachyzoites in the supernatant, 5 µl of mAb DG52 pre-coupled to 100 µl Dynabeads Protein G (Thermo Fisher) was added to the supernatant and incubated for 1 hr at 4 °C. The supernatant was separated from the beads, bradyzoites centrifuged at 600 × g, 4°C for 10 min and kept for different assays. To purify bradyzoites used for investigation of recovery of microneme secretion and high-performance liquid chromatography UV (HPLC-UV) analysis of ATP, ADP, and AMP levels in bradyzoites, PBS was replaced with EC buffer without Ca²⁺ or glucose during the purification by magnetic beads.

A23187, zaprinast, ionomycin, thapsigargin, NH₄Cl, fluorescein isothiocyanate-conjugated DBA, and BAPTA-AM were obtained from Sigma. Fluo-8 AM was obtained from Abcam. SYTOX Red Dead Cell Stain was obtained from Thermo Fisher. The compounds 3-MB-PP1 (Lourido et al., 2010) and compound 1 (Brown et al., 2016) were obtained as described previously. Trypsin and L-glutamine were purchased from MP Biomedicals. Adenosine 5'-triphosphate (ATP) disodium salt, adenosine 5'-diphosphate (ADP) sodium salt, adenosine 5'-monophosphate (AMP) disodium salt, oligomycin A, and 2-DOG were purchased from Sigma. Primary antibodies include mouse mAb DG52 anti-SAG1 (provided by John Boothroyd), mouse mAb 6D10 anti-MIC2 (Carruthers et al., 1999a), rabbit anti-GRA7 (Alaganan et al., 2014), mouse mAb 8.25.8 anti-BAG1 (obtained from Louis Weiss), rabbit anti-BAG1 (obtained from Louis Weiss), rabbit anti-M2AP (obtained from Vernon B. Carruthers), mouse anti-c-myc (mAb 9E10, BioLegend), mouse anti-acetylated Tubulin (mAb 6-11B-1, Sigma), rat anti-mCherry (mAb 16D7, Life Technologies), rabbit-anti SRS9 (obtained from John Boothroyd), rabbit anti-tRFP (Axxora), and mouse anti-6XHis (mAbHIS.H8, Life Technologies). Secondary antibodies for IFAs include goat anti-mouse IgG conjugated to Alexa Fluor-488, goat anti-rabbit IgG conjugated to Alexa Fluor-488, anti-mouse IgG conjugated to Alexa Fluor-568, goat anti-rat IgG conjugated to Alexa Fluor-568, and goat anti-mouse IgG conjugated to Alexa Fluor-594 (Life Technologies). For western blotting, secondary antibodies consisted of goat anti-mouse IgG, goat anti-rabbit IgG, or goat anti-rat IgG conjugated to LI-COR C800 or C680 IR-dyes and detected with an Odyssey Infrared Imaging System (LI-COR Biotechnology).

Freshly harvested parasites were loaded with 500 nM Fluo-8 AM for 10 min at room temperature, followed by centrifugation at 400 × g for 5 min and washing in EC buffer without Ca²⁺. Parasites were resuspended in EC buffer without Ca²⁺ and added directly to glass-bottom culture dishes. After addition of agonists, time-lapse images were recorded and analyzed as described above.

Infected cells were treated with 2 μM A23187 or 500 μM zaprinast for 15 min at 37°C. Following incubation, samples were stained by IFA using antibodies against SAG1 (mouse), GRA7 (rabbit), FITC-conjugated DBA or BAG1 (rabbit), and followed by secondary antibodies conjugated to Alexa Fluor. Samples were examined by fluorescence microscopy, and the percentages of egressed or released parasites per vacuole or cyst was determined at least for 20 vacuoles or cysts per experiment. The maximum egress distance of parasites from vacuole or cysts was measured from scanned tiff images in ImageJ.

ME49 BAG1-mCherry MIC2-GLuc reporter bradyzoites were induced for 7 days at pH 8.2, harvested in IC buffer as described above, and passed through a 3 μm polycarbonate membrane filter. ME49 ∆hxgprt::Fluc tachyzoites, cultured and harvested as indicated above, were used for gating. Approximately 1 × 10⁶ parasites from each sample (ME49 BAG1-mCherry MIC2-GLuc reporter tachyzoites and ME49 BAG1-mCherry MIC2-GLuc reporter bradyzoites) were sorted on Sony SH800S Cell Sorter directly into 500 µl IC buffer followed by centrifugation. Flow cytometry data were processed using FlowJo version 10 (FlowJo, LLC).

FACS-sorted MIC2-GLuc reporter tachyzoites and bradyzoites, or bradyzoites purified by magnetic beads, were suspended with EC buffer and incubated with different agonists at 37°C for 10 min. ESA was collected by centrifugation and mixed with Pierce Gaussia Luciferase Glow Assay Kit reagent (Thermo Scientific), and luminescence was detected using a Cytation 3 Cell Imaging Multimode Imager (BioTek Instruments, Inc). Buffer control values were subtracted from their corresponding sample values to correct for background.

RNA was extracted from ME49 ∆hxgprt::Fluc tachyzoites and bradyzoites induced for 7 days at pH 8.2 using RNeasy Mini Kit (QIAGEN) combined with QIAshredder (QIAGEN) followed by DNA Removal using DNA-free DNA Removal Kit (Thermo Fisher) and subsequent reverse transcription using High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher). Quantitative real-time PCR was performed on Applied Biosystems QuantStudio 3 Real-Time PCR System (Thermo Fisher) using SYBR Green JumpStart Taq ReadyMix (Sigma) with primers shown in Supplementary file 1. Mean fold changes from two independent experiments were calculated from ∆∆ Ct values using actin1 transcript as housekeeping gene, as described previously (Livak and Schmittgen, 2001).

Coverslips were precoated by incubation in 50% FBS diluted in PBS for 1 hr at 37°C followed by rinsing in PBS. Freshly harvested tachyzoites or bradyzoites were resuspended in EC buffer, treated with DMSO (0.1%, v/v), or inhibitors (in 0.1% DMSO, v/v), and then added to precoated glass coverslips and incubated at 37°C for 15 min. Coverslips were fixed in 2.5% formalin in PBS for 10 min, and the surface proteins were detected by IFA as described above using anti-SAG1 and anti-SRS9 antibodies as stage-specific markers for tachyzoites and bradyzoites, respectively. Gliding trails were captured by IFA microscopy as described above, and the frequency of trails was measured from tiff images using ImageJ.

BAG1-mCherry parasites were induced to form bradyzoites by culture at pH 8.2 in RPMI 1640 medium under ambient air (low CO₂) for 7 days followed by scraping into EC buffer without Ca²⁺ or glucose and repeated passage through a 23G needle. Intact but extracellular cysts were pellet by centrifugation at 150 × g for 10 min and resuspended in EC buffer without Ca²⁺ or glucose and incubated for ~2 hr at 4°C. MatTek 25 mm glass-bottom dishes (coverslip dishes) were precoated with 2 ml 50% FBS at 4°C overnight and rinsed twice using PBS prior to use. Purified cysts were added to the precoated coverslip dishes in EC buffer without Ca²⁺ or glucose but containing 0.25 mg/ml trypsin and incubated for 10 min at 16°C. The medium was removed and 2 ml EC buffer ± 1.8 mM Ca²⁺ and/or ±5.6 mM glucose was added and incubated for 10 min or 1 hr at 16°C. Prior to imaging, the coverslip dishes were heated to 37°C using a Heating Unit XL S (Zeiss) attached to the Zeiss AxioObserver Z1 (Carl Zeiss, Inc). Images were collected under bright-field illumination using a ×40 C-Apochromat water immersion objective (N.A. 1.20) and ORCA-ER digital camera (Hamamatsu Photonics) at 1 s intervals for 5 min per field. The percentage of BAG1-mCherry-positive bradyzoites displaying different types of gliding motility was calculated from six movies per sample. Images were imported into NIH ImageJ with a Cell Counter plug-in for quantification of the types of motility based on visual inspection.

Bradyzoites were induced for 7 days at alkaline pH and purified by magnetic beads as described above, followed by resuspension in 1 ml EC buffer containing 1.8 mM Ca²⁺ and 5.6 mM glucose for 10 min or 1 hr at 16°C. As control, tachyzoites were harvested and incubated in EC buffer without Ca²⁺ or glucose for 2 hr at 4°C, followed by treatment with EC buffer containing 1.8 mM Ca²⁺ and 5.6 mM glucose for 10 min or 1 hr 16°C. Following incubation, parasites were pelleted at 600 × g, 4°C for 10 min, and stored at –80°C until analysis.

A previously described method for extraction of ATP, ADP, and AMP (Menegollo et al., 2019) was adapted for use here. In brief, 95 µl of extraction buffer (0.3 M perchloric acid [HClO₄], 1 mM ethylenediaminetetraacetic acid disodium salt [Na₂EDTA], pH 8.0) was used to resuspend cell pellets and incubated for 5 min at room temperature. Extraction was stopped by addition of 17 µl of neutralization buffer (2 M potassium hydroxide) to the samples followed by mixing. Samples were centrifuged at 14,000 × g for 10 min at 4°C, and the supernatant was transferred to a new tube for HPLC analysis. Analysis was performed using an HPLC system consisting of a SPD-20A UV/VIS detector (Shimadzu) equipped with SIL-20A autosampler (Shimadzu), with a Luna Omega Polar C18 column (4.6 mm internal diameter × 150 mm length, 3 μm particle size, 100 Å pore size) and LC-20AD pump (Shimadzu). The protocol was setup as isocratic separation using a mobile phase containing 0.1 M ammonium dihydrogen phosphate (NH₄H₂PO_4, Sigma), pH 6.0, containing 1% methanol with a flow rate of 0.8 ml/min. Injection volume was 30 µl, and peak detection was monitored at 254 nm. A series of standards containing ATP, ADP, and AMP with different concentrations were used to establish retention times and standard calibration curves by calculating peak area. Samples from two independent biological replicates were analyzed using three technical replicates. The retention time and peak areas were used to calculate the corresponding concentration of each nucleotide from each sample according to the standard curve.

Mice were housed in an Association for Assessment and Accreditation of Laboratory Animal Care International-approved facility at Washington University School of Medicine. All animal studies were conducted in accordance with the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals, and protocols were approved by the Institutional Animal Care and Use Committee at the School of Medicine, Washington University in St. Louis.

Eight-week-old female CD-1 mice (Charles River) were infected with 200 ME49 BAG1-mCherry GCaMP6f tachyzoites by intraperitoneal injection. After 30 days of infection, animals were sacrificed, the brain removed and homogenized, and the number of brain cyst was determined by DBA staining and microscopy as previously described (Wang et al., 2015). Eight-week-old female CD-1 mice (Charles River) were infected with five cysts from the brain homogenate by oral gavage. Following a 30-day period, these mice were euthanized, and brain homogenate was collected and added to glass-bottom dishes for live imaging of tissue cysts.

Statistical analyses were performed in Prism (GraphPad). Data that passed normally distribution were analyzed by one-way ANOVA or Student’s t-tests, while data that were not normally distributed, or contain too few samples to validate the distribution, were analyzed by Mann–Whitney or Kruskal–Wallis nonparametric tests. *p<0.05, **p<0.01, ***p<0.001.