We initially designed experiments to track the destination of Ca²⁺ taken up by extracellular tachyzoites from the extracellular milieu into their cytosol. Tachyzoites were loaded with the ratiometric Ca²⁺ indicator Fura-2 (Figure 1A) and incubated in Ringer buffer containing 100 µM EGTA to chelate extracellular Ca²⁺ and prevent further uptake. Under these conditions, we investigated the release of intracellular Ca²⁺ pools by various pharmacological agents.

When TG, an inhibitor of SERCA, was added, it blocked the re-uptake of Ca²⁺ into the ER and unmasked the passive leak of Ca²⁺ from the ER into the cytosol, resulting in a cytosolic Ca²⁺ increase of approximately 160 nM (Figure 1B–C). We compared this response to the effect of ionomycin (IO), a Ca²⁺/H⁺ ionophore which acts on neutral intracellular Ca²⁺ stores, inducing depletion of Ca²⁺, mainly from the ER (Smith et al., 1989). Exposure of T. gondii to 1 µM IO caused a cytosolic increase of approximately 700–1,100 nM Ca²⁺ (Figure 1C). We next tested another SERCA inhibitor, cyclopiazonic acid (CPA), which is structurally unrelated to TG and with a different mode of action (Inesi and Sagara, 1994). CPA induced a smaller increase in cytosolic Ca²⁺ compared to TG, possibly indicating less efficacy toward TgSERCA (Figure 1—figure supplement 1A–B). Exposure to TG before CPA abolished the effect of CPA, whereas exposure to CPA did not prevent the effect of TG. (Figure 1—figure supplement 1B–C). Although TG does a better job at depleting the ER of Ca²⁺ in intact parasites, the resulting increase of cytosolic Ca²⁺ after adding TG is modest compared to the response of IO (Figure 1C). This result could reflect the slow kinetics of Ca²⁺ leak from the ER, allowing other buffering and transport mechanisms to mitigate the phenomenon. Alternatively, it may indicate that the duration after TG treatment was sufficient to complete store depletion. As shown in Figure 1B–C, residual Ca²⁺ remains in the stores after TG treatment, and the TG-induced phenomenon does not return to baseline, suggesting that the leak remains active.

Next, we tested the effect of Ca²⁺ entry in the filling of intracellular stores by measuring the cytosolic Ca²⁺ increases in response to inhibitors after pre-exposing the parasite suspension to extracellular Ca²⁺ (Figure 1D–G, darker traces). Addition of 1.8 mM Ca²⁺ caused a cytosolic increase due to influx through the plasma membrane (Figure 1D, dark blue trace) (Pace et al., 2014). A substantial portion of the entering Ca²⁺ appeared to be rapidly sequestered by the ER, as evidenced by the significantly greater TG-induced response in parasites previously exposed to extracellular Ca²⁺ (compare light and dark blue traces) (Figure 1D). We next tested the lysosomotropic agent glycyl-L-phenylalanine-naphthylamide (GPN), which primarily mobilizes Ca²⁺ from acidic organelles (Haller et al., 1996; Lloyd-Evans et al., 2008; Miranda et al., 2010; Yuan et al., 2021) and observed a similar pattern (Figure 1E). The increase in cytosolic Ca²⁺ following GPN addition was markedly greater in parasites previously exposed to extracellular Ca²⁺ than in those that had not been exposed (Figure 1E, compare light and dark blue traces). We previously proposed that GPN may act on the lysosome-like Plant-Like Vacuolar Compartment (PLVAC), a dynamic acidic organelle involved in calcium storage, and the processing of secretory proteins (Miranda et al., 2010; Stasic et al., 2022). The increase in cytosolic Ca²⁺ in response to the addition of nigericin (which acts on acidic stores) or CCCP (which likely targets mitochondria) was also greater in cells previously loaded with extracellular Ca²⁺ (Figure 1F–G, dark purple and dark orange traces, respectively). These data indicate that the ER, mitochondrion, PLVAC, and other acidic stores release more calcium into the cytosol of T. gondii tachyzoites following exposure to extracellular Ca²⁺, which stimulates its influx through the plasma membrane. The ER displayed high capacity to access a large portion of extracellular Ca²⁺, with TG producing close to ~300–400 nM of Ca²⁺ increase after pre-exposure to Ca²⁺ and only ~150–200 nM Ca²⁺ without Ca²⁺ pre-exposure (Figure 1D).

We next aimed to understand how other compartments are replenished with Ca²⁺, given that the ER appears to be particularly effective at taking up Ca²⁺ from the cytosol. We designed an experiment where parasites were first loaded with Ca²⁺, followed by inhibition of SERCA using TG. This inhibition prevents ER Ca²⁺ uptake, allowing Ca²⁺ to accumulate on the cytosolic side of the ER membrane. Under these conditions, we added agonists such as GPN or nigericin following the addition of TG. As shown in Figure 1H, Ca²⁺ was first added to load intracellular stores, followed by TG to induce ER Ca²⁺ leakage, and then GPN to trigger Ca²⁺ release from acidic stores. Comparison of GPN-induced cytosolic Ca²⁺ signals with and without TG pre-treatment revealed a significantly greater response in the TG condition. A similar enhancement was observed for the nigericin-induced response following TG treatment (Figure 1I).

These results support our hypothesis that extracellular Ca²⁺ is primarily taken up by the ER and subsequently redistributed to other organelles. Importantly, Ca²⁺ was added prior to TG to allow store loading, and TG treatment then permitted ER Ca²⁺ leakage, facilitating Ca²⁺ transfer to other compartments.

However, we considered the possibility that the enhanced responses to GPN or nigericin could be due to increased PM Ca²⁺ influx triggered by elevated cytosolic Ca²⁺. To test this, we repeated the experiments in the absence of extracellular Ca²⁺ (Figure 1J–K). Notably, prior addition of TG resulted in an enhanced cytosolic Ca²⁺ response to both GPN and nigericin. These results further support the notion that Ca²⁺ can be transferred from the ER to other intracellular stores independently of extracellular Ca²⁺ influx.

We also performed an additional experiment in which SERCA was inhibited with TG prior to Ca²⁺ addition. We then quantified the subsequent GPN response in conditions with and without TG preincubation and observed a significant increase in the TG-treated group (Figure 1—figure supplement 2A). This result suggests that, under non-physiological conditions where SERCA is blocked, the PLVAC may take up Ca²⁺ directly from the cytosol. However, this is unlikely to occur under normal conditions, as functional SERCA likely has a higher affinity for Ca²⁺ and would sequester it limiting its availability to other compartments.

In summary, pre-exposure of T. gondii to physiological levels of extracellular Ca²⁺ markedly enhanced the capacity of the ER, mitochondria, and acidic stores to release Ca²⁺ into the cytosol, with the ER and GPN-sensitive stores exhibiting the most pronounced responses.

The previous results highlight the central role of the ER in taking up Ca²⁺ from the cytosol following an influx from the extracellular milieu. We propose that this ER uptake, essential for maintaining Ca²⁺ store levels, is driven by the high Ca²⁺ affinity of TgSERCA. As a key mechanism, SERCA enables the ER to sustain its Ca²⁺ concentration despite the constitutive and passive leakage of Ca²⁺ from the ER into the cytosol (Camello et al., 2002).

To characterize the activity of TgSERCA in situ, we adapted a protocol to directly measure Ca²⁺ uptake by the stores in which TgSERCA localizes (ER and Golgi apparatus) (Calixto et al., 2025). This approach, which has been widely used in mammalian cells to assess Ca²⁺ release from the ER, employs the low-affinity Ca²⁺ indicator Mag-Fluo-4 (Kd ~22 µM) (Rossi and Taylor, 2020). The cytosolic concentration of Ca²⁺ in T. gondii is approximately 70 nM (Moreno and Zhong, 1996), which is well below the detection threshold of Mag-Fluo-4. To facilitate loading into organelles, we incubated parasites for an extended period with higher concentrations of Mag-Fluo-4-AM, promoting its compartmentalization into intracellular stores. Following incubation, parasites were washed and treated with a low concentration of digitonin, which selectively permeabilizes the plasma membrane while preserving the integrity of organellar membranes. Under these conditions, the parasites retained the Ca²⁺ indicator within their organelles (Figure 2A). We next assessed the capacity of these permeabilized parasites to take up Ca²⁺. Since the activity of SERCA depends on MgATP (Figure 2B), we added this substrate in the presence of defined Ca²⁺ concentrations calculated using the MaxChelator program (Bers et al., 1994). Under these conditions (free calcium ranging from 55 to 880 nM and MgATP at concentrations of 25–500 μM), we observed consistent and reproducible Ca²⁺ uptake, as shown in Figure 2B–C. We selected 220 nM Ca²⁺ for our study because this concentration approximates physiological cytosolic fluctuations and supports detectable Ca²⁺ uptake. Additionally, this concentration of Ca²⁺ has been used in previous studies of mammalian SERCA (Rossi and Taylor, 2020). Validation that this activity is mediated by TgSERCA is demonstrated by the addition of TG, which inhibits SERCA allowing Ca²⁺ leakage from the organelle (Figure 2D). Although the Ca²⁺ released after adding TG appears modest, consistent with the slow leak characteristics of ER calcium, the high Kd (22 µM) of the indicator implies that even small decreases in fluorescence signal, represent significant Ca²⁺ efflux. In contrast, IO at 1 µM caused a more pronounced Ca²⁺ release, lowering the Ca²⁺ concentration below the baseline level. This effect is likely due to IO targeting multiple intracellular compartments in addition to the ER, as well as the fundamental difference in mechanisms: IO acts as an ionophore, directly facilitating Ca²⁺ efflux across membranes, whereas TG inhibits SERCA, resulting in Ca²⁺ release through the ER’s natural leak pathway (Figure 2E).

We used the Mag-Fluo-4 assay to directly compare the inhibitory effects of CPA and TG (Figure 2—figure supplement 1). Under the conditions of the Mag-Fluo-4 assay, using digitonin-permeabilized parasites, both inhibitors produced comparable levels of Ca²⁺ efflux suggesting that at the concentrations used both inhibited SERCA and the efflux rate corresponds to the intrinsic ER leak mechanism (Figure 2—figure supplement 1A–C). This finding suggests that CPA may be less effective at inhibiting SERCA in intact parasites, possibly due to its reversibility and partial dissociation over time, allowing residual Ca²⁺ reuptake into the ER and resulting in a smaller cytosolic Ca²⁺ increase compared to TG.

In summary, these results demonstrate that the activity of TgSERCA in T. gondii tachyzoites can be measured in situ using permeabilized parasites loaded with the low-affinity Ca²⁺ indicator Mag-Fluo-4. This activity is MgATP-dependent and both TG and CPA can inhibit TgSERCA activity, leading to leakage of the accumulated Ca²⁺. The larger effect of IO compared to TG is likely due to differences in their mechanisms of action.

To investigate the role of TgSERCA (TGGT1_230420) in the biology of T. gondii, we generated conditional knockout parasites (i△TgSERCA), based on the gene’s predicted essentiality (fitness score –5.44) (Sidik et al., 2016). A tetracycline regulatable element was inserted at the 5’ end of the TgSERCA gene locus to control its expression with anhydrotetracycline (ATc) (Sheiner et al., 2011). In addition, we endogenously tagged TgSERCA with a C-terminal 3xHA epitope and generated clonal lines of both iΔTgSERCA and iΔTgSERCA-3HA (Figure 3A).

With the aim of detecting the protein, we generated a guinea pig polyclonal antibody against the phosphorylation (P) and nucleotide-binding (N) domains of TgSERCA, which was affinity-purified and validated by Western blotting (Figure 3—figure supplement 1A) and IFAs (Figure 3—figure supplement 1B). Colocalization of the anti-TgSERCA with the anti-HA signal was confirmed by IFA. Although the signals from the anti-HA and anti-TgSERCA did not completely overlap, both were lost in the iΔTgSERCA mutant grown in the presence of ATc (Figure 3—figure supplement 1B, +ATc). The partial colocalization may reflect differences in antibody accessibility or that the two antibodies recognize distinct regions of the protein. Both Western blots and IFAs confirmed that TgSERCA expression is tightly regulated by ATc and becomes undetectable after 2.5 days in culture (Figure 3B, Figure 3—figure supplement 1B). Growth of the iΔTgSERCA mutant was severely impaired in the presence of ATc, as assessed by plaque assays (Figure 3C–D). In this assay, parasites undergo successive rounds of invasion, replication, and egress, leading to host cell lysis and the formation of plaques on confluent monolayers. Downregulation of TgSERCA expression led to a marked defect in replication, with parasites failing to progress beyond one or two rounds of division (Figure 3E–F). All parasitophorous vacuoles (PVs) in ATc-treated cultures contained four or fewer parasites (Figure 3F, bar graph on the right). Host cell invasion was also reduced in the iΔTgSERCA mutant (Figure 3G) when cultured with ATc.

Parasite egress was significantly affected by TgSERCA depletion. Ionomycin (IO), which has been known to trigger egress by inducing Ca²⁺ release (Borges-Pereira et al., 2015), and natural egress following pre-incubation with 1 μM compound 1 (Donald et al., 2002), previously shown to synchronize parasite exit (Vella et al., 2021), were both markedly reduced in ATc-treated parasites, underscoring the critical role of ER Ca²⁺ stores in supporting both ionophore-induced and spontaneous egress (Figure 3H, IO, and natural egress). Interestingly, however, egress induced by saponin in the presence of extracellular Ca²⁺ was accelerated in ATc-treated parasites (Figure 3H, saponin egress). This enhancement may result from a more rapid rise in cytosolic Ca²⁺, reaching the egress threshold more quickly due to impaired SERCA activity combined with ongoing Ca²⁺ leak from the ER (Vella et al., 2021). The saponin concentration used selectively permeabilizes the host cell membrane, allowing extracellular Ca²⁺ to enter the parasite cytosol without compromising the integrity of the parasite plasma membrane. This is consistent with previous observations showing that tachyzoites remain motile and exhibit Ca²⁺ oscillations under similar conditions (Borges-Pereira et al., 2015). The resulting rise in cytosolic Ca²⁺ within the parasite stimulates motility and triggers egress. To further examine this phenomenon, we directly compared the timing of egress between untreated and ATc-treated iΔTgSERCA parasites under identical saponin exposure conditions (Figure 3H, Saponin egress).

In summary, our findings demonstrate that TgSERCA is essential for T. gondii replication, invasion, and natural egress. Interestingly, when host cells were selectively permeabilized, parasites with reduced TgSERCA expression displayed accelerated egress, likely due to altered calcium dynamics.

Further characterization of the i△TgSERCA mutant showed a diminished cytosolic Ca²⁺ response to TG (Figure 4A), which was also observed when TG was applied after extracellular Ca²⁺ had been added to fill the stores (Figure 4B). This was most likely due to reduced Ca²⁺ accumulation by the ER in the iΔTgSERCA (+ATc) mutant. Note that the change in Ca²⁺ in Figure 4A (TatiΔku80 cells) is larger than in Figure 1B (RH strain), which we attribute to differences between the two cell lines (RH vs TatiΔku80).

Most importantly, MgATP-driven Ca²⁺ uptake by permeabilized cells measured using Mag-Fluo-4, showed no TgSERCA activity after 48 hr of culture with ATc (Figure 4C). This experiment validated the Mag-Fluo-4 method for assessing SERCA activity. At 24 hr post-culture with ATc, some residual SERCA activity was still detected (Figure 4C).

Interestingly, Ca²⁺ entry measured in Fura-2 loaded iΔTgSERCA parasites (±ATc) was not affected by the downregulation of TgSERCA (Figure 4D). This finding argues against the presence of an ER-based mechanism that regulates Ca²⁺ entry. Moreover, the cytosolic resting Ca²⁺ concentration remained unchanged in the iΔTgSERCA (+ATc) mutant (Figure 4A, B and D–H) highlighting a critical role of the plasma membrane Ca²⁺ pump in maintaining cytosolic Ca²⁺ homeostasis.

The response to Zaprinast was diminished but was still present (Figure 4E) indicating that Zaprinast induces Ca²⁺ release from the ER and from an additional compartment. When Zaprinast was added after Ca²⁺ replenishment, the response remained reduced in the mutant pre-incubated with ATc (Figure 4F). We next tested GPN, which primarily targets acidic stores, and observed a decreased response (Figure 4G). Adding GPN after replenishing the cells with Ca²⁺ resulted in an increased response, as we showed in Figure 1, but this response was also reduced when the mutant was grown with ATc (Figure 4H).

Given that the Ca²⁺ phenotypes were assessed after 24 hr of ATc treatment, when approximately 50% of TgSERCA activity remains, the response to Zaprinast may still reflect ER involvement, and may not provide definitive evidence for the contribution of an additional Ca²⁺ pool. To further investigate this, we conducted an experiment in which TG was added prior to GPN and Zaprinast. In this setting, GPN significantly reduced the Zaprinast-induced response (Figure 1—figure supplement 2B). This result suggests that Zaprinast also targets a non-ER Ca²⁺ store, and that this store is likely the same one affected by GPN.

These results support a functional link between the stores targeted by GPN and the ER. Given that SERCA downregulation impaired ER Ca²⁺ storage without affecting cytosolic Ca²⁺ uptake and cytosolic Ca²⁺ levels, the diminished response to GPN suggests that Ca²⁺ released or leaked from the ER is important for refilling the store targeted by GPN.

In mammalian cells, the high concentration of Ca²⁺ in the ER is important for mitochondrial ATP production (Wenzel et al., 2022). This is because of the close proximity between the ER and mitochondria which allows for the directional flow of Ca²⁺ from the ER to the mitochondria (Gincel et al., 2001; Rapizzi et al., 2002). With the aim of verifying if the T. gondii mitochondria can take up calcium, we introduced a genetic Ca²⁺ indicator in the mitochondrion of T. gondii tachyzoites by attaching the GCaMP6f gene (Chen et al., 2013) to the mitochondrial targeting signal of the T. gondii superoxide dismutase 2 (SOD2) gene (Pino et al., 2007) and isolated stable transgenic clones (RH-SOD2-GCaMP6f) (Vella et al., 2020). Fluorescence microscopy of live cells confirmed GCaMP6f localization to the mitochondria (Figure 5A). Direct Ca²⁺ uptake was observed in digitonin-permeabilized parasites incubated in the presence of increasing concentrations of Ca²⁺ (Figure 5B–C). Although a measurable increase in mitochondrial fluorescence was observed, it required high Ca²⁺ concentrations, indicating that the T. gondii mitochondrion can take up Ca²⁺ but do so with very low affinity. These Ca²⁺ levels were significantly higher than the typical cytosolic Ca²⁺ concentrations found in healthy cells (Figure 5B–C).

We hypothesized that the T. gondii mitochondrion may take up Ca²⁺ through close membrane contacts with the ER, where localized Ca²⁺ concentrations in microdomains could be significantly higher than in the cytosol, a mechanism previously described in mammalian cells (Rizzuto et al., 1998). We next loaded the RH-SOD2-GCaMP6f mutant with Fura-2 to simultaneously monitor cytosolic and mitochondrial Ca²⁺ in intact parasites. Upon addition of extracellular Ca²⁺, an increase in cytosolic Ca²⁺ was observed, however, mitochondrial GCaMP6f fluorescence remained unchanged (Figure 5D–E), suggesting that mitochondria are unable to take up Ca²⁺ at the cytosolic concentrations reached under these conditions. This also validates the proper localization of the indicator, confirming its absence from the cytosol. Addition of TG followed by extracellular Ca²⁺ resulted in a cytosolic Ca²⁺ increase, readily detected in Fura-2-loaded parasites. However, and most importantly, only TG triggered a measurable increase in the mitochondrial GCaMP6f signal, whereas a rise in cytosolic Ca²⁺ induced by extracellular Ca²⁺ addition alone did not (Figure 5G). Our interpretation is that TG-induced ER Ca²⁺ leakage led to local accumulation of Ca²⁺ at the cytosolic face of the ER membrane, creating microdomains of high Ca²⁺ concentration sufficient to trigger mitochondrial uptake. Addition of Ca²⁺ after TG resulted in a greater increase in cytosolic Ca²⁺ (Figure 5F) compared to TG alone. However, even under these conditions, no corresponding increase in mitochondrial GCaMP6f fluorescence was observed. This further confirms that mitochondria are unable to take up cytosolic Ca²⁺ at these low concentrations.

We next introduced the same SOD2-GCaMP6f chimeric gene into the iΔTgSERCA mutant background and isolated a clonal line (iΔTgSERCA-SOD2-GCaMP6f) (Figure 5—figure supplement 1A–B). Fluorescence imaging of live cells confirmed proper localization of the indicator (Figure 5—figure supplement 1A), and fluorescence measurements of intact cells corroborated that adding extracellular Ca²⁺ did not increase GCaMP6f fluorescence, whereas addition of TG to the suspension resulted in a fluorescence increase (Figure 5—figure supplement 1B). We next monitored changes in GCaMP6f fluorescence in the mutant and compared results between parasites grown with and without ATc. In line with prior observations, parasites cultured without ATc showed a consistent and measurable increase in mitochondrial GCaMP6f signal upon TG treatment (Figure 5H, blue trace). In contrast, this response was abolished in parasites cultured with ATc, consistent with reduced TgSERCA expression leading to ER Ca²⁺ depletion (Figure 5H, pink trace).

We next tested additional stimuli and observed a clear increase in mitochondrial GCaMP6f fluorescence in the iΔTgSERCA mutant following the addition of GPN, Zaprinast, or IO (Figure 5I-K, blue traces). In all cases, this fluorescence increase was significantly reduced in the iΔTgSERCA (+ATc) mutant (Figure 5I-K, pink traces). These results suggest a potential direct interaction between the mitochondrion and acidic Ca²⁺ stores, such as the PLVAC and/or Golgi apparatus. The reduced Ca²⁺ content of these compartments, resulting from TgSERCA downregulation, appears to impact mitochondrial Ca²⁺ uptake.

In summary, we demonstrated that the T. gondii mitochondrion is capable of Ca²⁺ uptake via transfer from the ER, a process that becomes apparent upon inhibition of ER Ca²⁺ uptake with TG. This suggests that the high Ca²⁺ concentrations required for mitochondrial uptake are achieved only at membrane contact sites between the ER and mitochondria. In the iΔTgSERCA (+ATc) mutant, impaired ER Ca²⁺ storage due to TgSERCA downregulation compromises mitochondrial Ca²⁺ uptake. Additionally, our data suggest that the T. gondii mitochondrion may also take up Ca²⁺ from acidic stores, such as the PLVAC or Golgi, which appear to rely indirectly on ER Ca²⁺ refilling. When TgSERCA is downregulated, depletion of ER Ca²⁺ likely compromises the Ca²⁺ content of these acidic compartments, and impairing mitochondrial Ca²⁺ uptake from these stores.

We next investigated whether proximity between the ER and other organelles could be detected by IFA and/or electron microscopy (EM). We performed IFAs with ER and mitochondria markers and ER and PLVAC markers (Figure 6). In intracellular parasites, the mitochondrion was observed to surround the ER, forming multiple potential sites of interaction (Figure 6A and Figure 6—video 1). As previously described, the mitochondrion of intracellular T. gondii tachyzoites surrounds the periphery of the cell in a lasso-shape morphology (Ovciarikova et al., 2017). In contrast, in extracellular parasites, the mitochondrion displays a marked morphological change, adopting either a sperm-like or collapsed conformation (Ovciarikova et al., 2017; Figure 6B and Figure 6—video 2). Our hypothesis is that retraction of the mitochondrion allows the ER membranes to expand in extracellular parasites and extend toward the apical domain, where Ca²⁺ is required for micronemes secretion and conoid extrusion.

The PLVAC also formed points of contact with the ER (Figure 6C and Figure 6—video 3). Multiple points of contact were also observed by EM between the ER and the PLVAC, the ER and the apicoplast, and the ER and the mitochondrion (Figure 6D).

Interestingly, these contacts were still present in the iΔTgSERCA (+ATc) mutant (Figure 6—figure supplement 1A–D), as most likely TgSERCA would not be directly involved in the establishment of contacts. We quantified the length of the limiting membrane of the organelle in contact with ER membranes at a distance of less than 30 nm and found that after knockdown of TgSERCA, the contact was not altered. However, Ca²⁺ transported from the ER into the mitochondrion after TG treatment was significantly decreased which means that this phenotype is due to reduced ER Ca²⁺ and not to lack of contacts (Figure 6—figure supplement 1E). A similar result was seen when measuring contacts between the ER and the apicoplast (Figure 6—figure supplement 1F) and between the ER and the PLVAC (Figure 6—figure supplement 1G).

In summary, this data supports the presence of points of contact between the ER and other organelles like the mitochondrion, the PLVAC, and the apicoplast. These contacts likely facilitate the transfer of Ca²⁺ from the ER, the organelle with the highest Ca²⁺ content, to other compartments.

Toxoplasma gondii tachyzoites (RH and TatiΔku80 strain) were maintained in human telomerase reverse transcriptase immortalized foreskin fibroblasts (hTERT) (Farwell et al., 2000) grown in Dulbecco’s modified minimal essential media (DMEM) with 1% FBS. These cells are tested for Mycoplasma contamination regularly and are treated with mycoplasma removal agent. The hTERT cell line (ATCC CRL-3627) is the only human cell line used in this project. It was obtained directly from ATCC and expanded in the absence of any other mammalian cells. After amplification, the cells were cryopreserved in liquid nitrogen. A single vial is thawed and used for approximately six months before being discarded and replaced with a new vial. The hTERT cells were used solely for the growth of Toxoplasma gondii tachyzoites. No experiments were performed using this cell line itself.

A promoter insertion plasmid was generated by cloning three PCR fragments into a modified pCR2.1-TOPO vector using the Gibson Assembly Cloning Kit (NEB #E5510). One fragment corresponding to the TgSERCA flanking region (predicted promoter/5’UTR) was amplified with primers 1 and 2 (Supplementary file 1). The second fragment corresponds to DHFR + T7 S4 (Sheiner et al., 2011) and was amplified with primers 3 and 4. Another fragment corresponds to the 5’ TgSERCA coding sequence beginning with start codon and was amplified with primers 5 and 6. The vector pCR2.1-TOPO was used, which had only one EcoRI site and it was cut with the enzyme NotI to use as vector backbone. The promoter insertion plasmid was transfected into the TatiΔku80 cells and selected with 1 μM pyrimethamine using an ‘ultra-aggressive’ screening method. Briefly, 200 μl of the suspension of transfected parasites was added to 10 ml of medium, and one to three drops (~65 μl per drop) were inoculated into each well of three 24-well plates already filled with medium. The clonal lines created after selection and subcloning were termed iΔTgSERCA.

For in situ tagging, an approximately 2 kb fragment was amplified from the genomic locus (3’ region) of the TgSERCA gene using primers 7 and 8. The fragment was cloned into the pLic-3HA-CAT plasmid (Huynh and Carruthers, 2009) and the construct was linearized with the enzyme NheI for transfection into the iΔTgSERCA mutant. Clonal cell lines were generated after selection with chloramphenicol and subcloning and termed iΔTgSERCA-3HA.

The phosphorylation (P) and nucleotide binding (N) domains of TgSERCA (TGGT1_230420) (nucleotides 1123–2415, amino acid residues 375–805) were cloned into XmaI and HindIII sites of pQE-80L with primers 13 and 14 (Supplementary file 1) to create recombinant protein with a N-terminal 6xHis tag. The resulting plasmid was transformed into Escherichia coli BL21-CodonPlus competent cells and expression was induced by addition of 0.4 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) for 4 hr at 37 °C. Cells were pelleted and resuspended in equilibration/binding buffer (50 mM Na₃PO₄, 300 mM NaCl, 10 mM Imidazole, 8 M Urea, and protease inhibitor cocktail, Sigma, P-8849). The cells were then sonicated for 80 s and centrifuged at 12,000 rpm for 20 min at 4 °C. The supernatant was filtered through a 0.45 μm membrane and the protein was purified using HisPur Ni-NTA Chromatography Cartridge (Thermo Scientific) following instructions from the manufacturer. Proteins that were unbound were washed with 12 ml of wash buffer (50 mM Na₃PO₄, 300 mM NaCl, 40 mM imidazole, and 8 M urea), and the recombinant protein was eluted with 5 ml elution buffer (50 mM Na₃PO₄, 300 mM NaCl, 250 mM imidazole, and 8 M urea). Eluted protein fractions were concentrated and desalted with an Amicon Ultra-0.5 mL centrifugal filter (Millipore Sigma).

Two guinea pigs were each immunized with 0.2 mg of purified recombinant protein mixed with equal volume of Freund’s Complete Adjuvant (Sigma F5581), followed by two boosts of 0.1 mg antigen mixed with equal volume of Freund’s Incomplete Adjuvant (Sigma F5506) for guinea pig 1 and three boosts for guinea pig 2. The resulting antibodies were tested at 1:1,000 in western blot against RH lysates and were developed with Alexa Fluor 488 goat anti-guinea pig (1:1,000). The antibodies were compared with Dr. Sibley’s mouse anti-SERCA antibody (Nagamune et al., 2007a) to confirm size and purity (Figure 3—figure supplement 1A). The anti-TgSERCA antibodies were then affinity-purified. Guinea pigs were handled according to our approved institutional animal care and use committee (IACUC) protocols (A2021 03–005 A5) of the University of Georgia.

T. gondii tachyzoites were loaded with Fura-2 AM as previously described (Vella et al., 2020; Stasic et al., 2021). Freshly released tachyzoites were washed twice with buffer A plus glucose (BAG; 116 mM NaCl, 5.4 mM KCl, 0.8 mM MgSO4, 50 mM HEPES, pH 7.3, and 5.5 mM glucose), by centrifugation (706×g for 10 min) and re-suspended to a final density of 1×10⁹ parasites/ml in loading buffer (BAG plus 1.5% sucrose, and 5 μM Fura-2-AM). The suspension was incubated for 26 min at 26 °C with mild agitation. Subsequently, the parasites were washed twice (2,000×g for 2 min) with BAG to remove extracellular dye, re-suspended to a final density of 1×10⁹ parasites per ml in BAG and kept on ice. This loading protocol is specifically designed to minimize Fura-2 compartmentalization, which is typically indicated by elevated resting Ca²⁺ concentrations. All experiments are conducted within a time frame during which resting Ca²⁺ levels remain stable, typically below or at 100 nM. For fluorescence measurements, 2×10⁷ parasites/mL were placed in a cuvette with 2.5 mL of Ringer’s buffer without calcium (155 mM NaCl, 3 mM KCl, 1 mM MgCl₂, 3 mM NaH₂PO₄, and 10 mM Hepes, and 10 mM glucose). Fluorescence measurements were done in a Hitachi F-7000 or F-4500 fluorescence spectrophotometer using the Fura-2 conditions for excitation (340 and 380 nm) and emission (510 nm). The Fura-2 fluorescence response to Ca²⁺ was calibrated from the ratio of 340/380 nm fluorescence values after subtraction of the background fluorescence of the cells at 340 and 380 nm as previously described (Grynkiewicz et al., 1985). The Ca²⁺ release rate was defined as the change in Ca²⁺ concentration during the initial 20 s after reagent addition. ΔF was calculated as the difference between the highest Ca²⁺ peak and basal Ca²⁺levels, and recovery was defined as the change in Ca²⁺ concentration after the peak was reached, measured over the indicated time intervals.

Tachyzoites freshly egressed and washed as described above were resuspend to a final density of 1 × 10⁹ cells/ml in HBS buffer (135 mM NaCl, 5.9 mM KCl, 1.2 mM MgCl₂, 11.6 mM HEPES pH 7.3, 1.5 mM CaCl₂, 11.5 mM glucose) containing 1 mg/ml BSA, 0.2 mg/ml of pluronic F127 and 20 μM Mag-Fluo4-AM. The suspension was incubated at RT with mild shaking for 1 h, in the dark. Subsequently, parasites were washed two times and centrifuged at 5,000 rpm for 2 min to remove extracellular dye. The pellet was resuspended in 1.8 ml of CLM buffer (20 mM NaCl, 140 mM KCl, 20 mM PIPES, pH 7.0) containing 1 mM EGTA at 1 × 10⁹ cells/ml. Parasites were permeabilized with 44.4 μg/ml digitonin for 6 min, washed twice with CLM containing 1 mM EGTA and cetrifuged at 5,000 rpm for 2 min to remove digitonin, then resuspended to a final density of 1 × 10⁹ tachyzoites/ml and kept on ice. For each test, 50 μl (5 × 10⁷) of parasite suspension was added to 1.95 ml of CLM containing 1 mM EGTA and 0.375 mM CaCl₂ which results in 220 nM free Ca²⁺ as calculated with MaxChelator. Fluorescence was measured with a Hitachi F-7000 or F-4500 fluorescence spectrophotometer (Excitation at 485 nm and emission at 520 nm). Ratio (∆F/F₀/s) was evaluated by measuring the rate of change in fluorescence over 20 s after reagent addition.

The organelle targeting of GCaMP6f was made by overlapping PCR. The N-terminal mitochondrial targeting sequence of the T. gondii SOD2 gene (Pino et al., 2007) was used to target GCaMP6f to the mitochondrion. The GCaMP6f gene for this construct was amplified by primers 9 and 10 (Supplementary file 1). After gel purification of the GCaMP6f and SOD2 sequences, the mitochondria targeting construct was built by overlapping PCR with the purified PCR products as template. This construct was then cloned into the Topo-blunt vector. After the sequence was verified by sequencing, the SOD2-GCaMP6f fragment was removed by BglII and AvrII digestion and cloned into the same restriction sites of the pDT7S4H3 (Sheiner et al., 2011) and pCTH3 (Vella et al., 2020) vectors. The pDT7S4H3-SOD2-GCaMP6f construct was introduced into RH parasites by electroporation. After selection with pyrimethamine, the parasites were sorted by FACS and then subcloned. Clones were selected based on the dynamic range of the response to ionomycin. The pCTH3-SOD2-GCaMP6f was introduced into the iΔTgSERCA mutant by electroporation. After selection with chloramphenicol, the parasites were sorted by FACS and then subcloned. The expression of GCaMP6f was verified by live-cell imaging, and western blots. The clone with the largest dynamic range, as evaluated using Ionomycin, was selected for further experiments. Figure 5—figure supplement 1 shows live fluorescence confirming mitochondrial localization and fluorescence traces showing the response of whole parasites expressing GCaMP6f to the addition of Ca²⁺ and Thapsigargin.

Measurements with permeabilized parasites: T. gondii tachyzoites expressing SOD2-GCaMP6f were collected and washed two times at 5,000 rpm for 2 min with BAG. The parasite pellet was resuspended in 1.8 ml of BAG buffer containing 0.1 mM EGTA at 1x10⁹ cells/ml. Permeabilization with 44.4 μg/ml digitonin for 6 min was done by following the fluorescence of GCaMP6f. Parasites were washed twice with the same buffer and centrifuged at 5,000 rpm for 2 min to remove digitonin, resuspended to a final density of 1×10⁹ parasites/ml in intracellular buffer (140 mM Kgluconate, 10 mM NaCl, 2.7 mM MgSO₄, 200 μM EGTA, 65 μM CaCl₂, 10 mM HEPES, 10 mM Tris, pH 7.3, 1 mM Glucose) and kept on ice. 50 μl (5 x 10⁷) of the parasite suspension was mixed with 1.95 ml intracellular buffer for measurement. Measurements were done in a Hitachi 7000 fluorescence spectrophotometer set at 485 nm excitation and 509 nm emission. The uptake rate (∆F/F₀/s) was evaluated by measuring the % of change in fluorescence per second during the initial 20 s after reagent addition.

For measurements with intact parasites, they were collected, washed, and resuspended in BAG at 1 × 10⁹ cells/ml for testing. 50 μl (5 × 10⁷) of the parasite suspension was mixed with 1.95 ml BAG containing 0.1 mM EGTA for measurement. The ratio (∆F/F₀) was evaluated by measuring the maximum change in fluorescence over 20 s after reagent addition (linear rate).

Red-green invasion assays were performed as originally described (Kafsack et al., 2004), modified (Chasen et al., 2017) and adapted to use td-RFP-expressing parasites. The number of tachyzoites used was 2 × 10⁷, and invasion was for 5 min. Plaque assays were performed as previously described (Roos et al., 1994) with modifications (Liu et al., 2014). 125 tachyzoites were used for infection of confluent six-well plates with hTERT fibroblasts, followed by an incubation time of 10 days prior to fixing and staining with crystal violet.

For egress assays, the monolayers of hTERT cells grown in 35  mm Mattek dishes were infected with 50,000 tdTomato-expressing parasites for 24 or 48 hr. Parasitophorous vacuoles containing 4–8 parasites were observed by microscopy after washing twice with Ringer’s buffer without calcium. Dishes were filled with 1 ml of Ringer’s buffer supplemented with either 100 μM EGTA or 1.8 mM CaCl₂. Images were collected in time-lapse mode with an acquisition rate of 3 s for 15 min. We observed that most of the iΔTgSERCA cells +/-ATc were still able to egress when stimulated with 1 or 0.5 µM ionomycin added 2 min after the start of the recordings with either 100 μM EGTA or 1.8 mM CaCl₂. We next tested lower concentrations of ionomycin (100 nM and 50 nM) in Ringer’s buffer containing 1.8 mM CaCl₂. Egress was also triggered by adding 0.01% Saponin in the presence of 1.8 mM CaCl₂. For egress triggered by ionomycin, the percentage of vacuoles egressed after adding ionomycin during the 15 min of the video (2 min baseline + 13 min after adding ionomycin) was quantified (from 100 vacuoles). For egress triggered by saponin, the time to egress after adding saponin was quantified.

For natural egress, the iΔTgSERCA mutant expressing td-tomato RFP was used to infect confluent hTERT cell monolayers 36 hr before adding ATc and 1 μM compound 1 (pyrrole 4-[2-(4-fluorophenyl)–5-(1-methylpiperidine-4-yl)–1H-pyrrol-3-yl]pyridine) (Cpd1) (Donald et al., 2002) dissolved in ethanol and the culture continued for 24 hr. After treating with Cpd1 for 24 hr, cultures showed intact vacuoles, which differed from the vehicle-treated plates (36 hr cultures plus ATc treatment for 24 hr or 48 hr without ATc), which were fully lysed. Following synchronization, the Cpd1-containing media was removed, and the vacuoles were washed twice with warm media lacking Cpd1. Fresh media without Cpd1 was added, and the plates were transferred to a prewarmed DeltaVision microscope stage set to 37  °C. After 10 min at 37 °C, egress of the full vacuoles was enumerated. We counted each plate for 1 min and evaluated at least 100 vacuoles per experiment. Three independent biological experiments were conducted and summarized.

For replication assays, hTERT cells were grown on 35  mm MatTek dishes. Each dish was infected with 50,000 tdTomato-expressing parasites. 24  hr after the infection, the number of parasites per PV was counted using a fluorescence microscope. For each experiment, at least 100 PVs were counted. Results were the average of three independent experiments (Li et al., 2021).

Tachyzoites were grown on hTERT cells on cover slips for ~24 hr, washed twice with BAG and fixed with 4% formaldehyde for 1 hr, followed by permeabilization with 0.3% Triton X-100 for 20 min, and blocking with 3% bovine serum albumin. IFAs were performed as previously described (Miranda et al., 2010). Fluorescence images were collected with an Olympus IX-71 inverted fluorescence microscope with a Photometrix CoolSnapHQ CCD camera driven by DeltaVision software (Applied Precision, Seattle, WA). Super-resolution microscopy was performed using a Zeiss ELYRA S1 (SR-SIM) system mounted on a high-resolution Axio Observer Z1 inverted microscope. The setup included transmitted light (HAL), UV (HBO), and high-power solid-state laser illumination sources (405/488/561 nm), a 100× oil immersion objective, and an Andor iXon EM-CCD camera. Image acquisition and structured illumination analysis were conducted using ZEN software (Zeiss) with the SIM analysis module. Rat anti-HA antibody (Roche) was used at a 1:25 dilution, and mouse anti-HA antibody (Covance) was used at a 1:200 dilution. Affinity-purified guinea pig anti-TgSERCA antibody was used at a 1:500 dilution.

Western blot analysis was performed as previously described (Liu et al., 2014). Rat anti-HA antibody from Roche was used at a dilution of 1:200. Mouse anti-HA antibody from Covance was used at a dilution of 1:1,000. The guinea pig anti-TgSERCA antibody was used at a dilution of 1:2000. Secondary goat anti-rat or mouse antibody conjugated with HRP was used at 1:5,000. Mouse anti-α-tubulin at a dilution of 1:5,000 was used for loading control.

For ultrastructural observations of intracellular T. gondii by thin-section transmission EM, infected human foreskin fibroblast cells were fixed in 2.5% glutaraldehyde in 0.1 mM sodium cacodylate (EMS) and processed as described (Coppens and Joiner, 2003). Ultrathin sections of infected host cells were stained before examination with a Hitachi 7600 EM under 80 kV. For quantitative measurement of distance between organelles, the closest point between T. gondii’s organelles and ER membrane was measured using ImageJ and was performed on 47 representative electron micrographs at the same magnification for accurate comparison between organelles.

Statistical analyses were performed by Student’s t-test using GraphPad PRISM version 9. Error bars shown represent mean ± SD (standard deviation) of at least three independent biological replicates. Unpaired two-tailed t-test performed in all comparisons.