Regulation of calcium entry by cyclic GMP signaling in Toxoplasma gondii

Abstract

Ca²⁺ signaling impacts almost every aspect of cellular life. Ca²⁺ signals are generated through the opening of ion channels that permit the flow of Ca²⁺ down an electrochemical gradient. Cytosolic Ca²⁺ fluctuations can be generated through Ca²⁺ entry from the extracellular milieu or release from intracellular stores. In Toxoplasma gondii, Ca²⁺ ions play critical roles in several essential functions for the parasite, like invasion of host cells, motility, and egress. Plasma membrane Ca²⁺ entry in T. gondii was previously shown to be activated by cytosolic calcium and inhibited by the voltage-operated Ca²⁺ channel blocker nifedipine. However, Ca²⁺ entry in T. gondii did not show the classical characteristics of store regulation. In this work, we characterized the mechanism by which cytosolic Ca²⁺ regulates plasma membrane Ca²⁺ entry in extracellular T. gondii tachyzoites loaded with the Ca²⁺ indicator Fura-2. We compared the inhibition by nifedipine with the effect of the broad spectrum TRP channel inhibitor, anthranilic acid or ACA, and we find that both inhibitors act on different Ca²⁺ entry activities. We demonstrate, using pharmacological and genetic tools, that an intracellular signaling pathway engaging cyclic GMP, protein kinase G, Ca²⁺, and the phosphatidyl inositol phospholipase C affects Ca²⁺ entry and we present a model for crosstalk between cyclic GMP and cytosolic Ca²⁺ for the activation of T. gondii’s lytic cycle traits.

Toxoplasmagondii is an obligate intracellular parasite that infects approximately one third of the world’s population (1). T. gondii causes disease by engaging in multiple rounds of a lytic cycle, which consists of invasion of host cells, replication inside a parasitophorous vacuole (PV), egress resulting in host cell lysis, and invasion of a new host cell (2, 3). Several key steps of the lytic cycle of T. gondii which are motility, attachment, invasion, and egress are regulated by fluctuations in its cytosolic Ca²⁺ concentration ([Ca²⁺]_cyt) (4, 5).

Ca²⁺ signaling plays important roles in the regulation of many cellular functions (6). However, the concentration of cytosolic Ca²⁺ is highly regulated because sustained high cytosolic Ca²⁺ is toxic and may result in cell death. A variety of Ca²⁺ pumps, channels, and transporters located at the plasma membrane (PM) and intracellular organelles (endoplasmic reticulum (ER), acidic stores, and mitochondria) are involved in regulating [Ca²⁺]_cyt (7).

In T. gondii, both Ca²⁺ entry and release from intracellular stores like the ER may initiate a cascade of signaling pathways that promote progression through the biological steps of the parasite’s lytic cycle. Motile parasites loaded with fluorescent Ca²⁺ indicators or expressing genetically encoded calcium indicators showed Ca²⁺ oscillations (8, 9). Previous studies have shown that a rise in [Ca²⁺]_cyt activates the motility machinery leading to egress. Blocking these cytosolic Ca²⁺ oscillations with intracellular BAPTA (membrane permeable cytosolic Ca²⁺ chelator) blocks motility, conoid extrusion (apical tip of the parasite necessary for attachment), invasion, and egress from the host cell (4).

Extracellular Ca²⁺ entry was demonstrated to be key in these processes in both extracellular (10) and intracellular replicating tachyzoites (11). In previous work, we showed that Ca²⁺ entry can be inhibited ∼80% by voltage-operated Ca²⁺ channel blockers like nifedipine (10) and ∼50% by the wide-spectrum transient receptor potential (TRP) channel inhibitor anthranilic acid (ACA) (12). Cytosolic Ca²⁺ itself enhanced Ca²⁺ entry by a mechanism that we termed Ca²⁺-activated Ca²⁺ entry, and recent work from our lab demonstrated the participation of a TRP-like channel named TgTRPPL-2 (12). However, the experimental evidence did not support that this calcium entry mechanism was regulated by store depletion as seen in mammalian cells by store operated calcium entry channels (13, 14). Furthermore, experiments testing surrogate ions like Mn²⁺ (10) and the absence of the components of the store operated calcium entry pathway, STIM and ORAI, in the T. gondii genome supported this notion (15).

Several recent studies demonstrated that the synthesis of cyclic GMP (cGMP) was fundamental for the control of essential parasite processes (16). cGMP activates the enzyme protein kinase G (PKG) which was proposed to regulate a phosphoinositide phospholipase C (PI-PLC) that produces inositol-1,4,5-trisphosphate (IP₃) which would act on an unidentified channel in the ER allowing the release of Ca²⁺ into the cytosol (17). However, the role of Ca²⁺ itself was not considered in this hypothetical signaling cascade. Ca²⁺ is a known modulator of Ca²⁺ channels (18) and it is known that the activation of PI-PLC depends on the local [Ca²⁺]_cyt (19, 20). In this work, we characterized the role of intracellular signaling in Ca²⁺ entry at the PM and we present a model for crosstalk between cGMP and cytosolic Ca²⁺ for the activation of T. gondii’s lytic cycle traits.

Results

Ca²⁺ entry at the PM

We assessed Ca²⁺ entry (Fig. 1A) in Fura-2–loaded extracellular tachyzoites by adding Ca²⁺ to T. gondii tachyzoites in suspension in a low Ca²⁺ buffer (EGTA buffer) (100 μM EGTA, ∼30 nM free [Ca²⁺]_ext). Fig. S1A shows the loading protocol used, described in detail in the experimental procedures section, and Figure 1A shows a control trace highlighting what we defined as “Basal Ca²⁺ entry” (pink arrow). An increase in fluorescence indicates the increase of cytosolic Ca²⁺ resulting from influx from the extracellular milieu (Fig. 1A, basal Ca²⁺ entry). The rate of increase (Δ[Ca²⁺]/sec) increased with the concentration of extracellular Ca²⁺ (Fig. S1B). This result indicated that extracellular tachyzoites express Ca²⁺ channels at the PM that are either activated by extracellular Ca²⁺ or the electrochemical gradient allows the flow of Ca²⁺ in through an open channel. The peak of cytosolic Ca²⁺ after the initial increase is followed by a recovery phase (due to uptake by other stores or extrusion) and a stabilization of the cytosolic concentration within nM range (Fig. 1A, recovery). This stable concentration is likely the result of an equilibrium between Ca²⁺ entry through the PM channels and the compensatory action of the PM and ER Ca²⁺ ATPases that actively remove Ca²⁺ from the cytosol.

Figure 1.

Calcium entry through the plasma membrane of extracellular Toxoplasma gondii tachyzoites.A, Fura-2–loaded T. gondii tachyzoites at a concentration of 2 × 10⁷/ml are suspended in a buffer containing 100 μM EGTA (EGTA buffer). The resting Ca²⁺ represents the normal cytosolic concentration of Ca²⁺ [Ca²⁺]_cyt. Addition of extracellular 1.8 mM Ca²⁺ stimulates Ca²⁺ entry as observed in the increase in cytosolic Ca²⁺ (basal Ca²⁺ entry). This increase is followed by a recovery. B, addition of extracellular Ca²⁺ (1.8 mM Ca²⁺) to parasites suspended in EGTA buffer (light blue trace) or high Ca²⁺ buffer (1.8 mM Ca²⁺, gray trace). t test: ∗∗p = 0.0018. C, 1.8 mM Ca²⁺ was added at various times after adding 100 μM EGTA: 200, 300, 400, 500 s. Individual dots in the bar graphs represent biological replicates. Data was analyzed using One-Way ANOVA ∗∗p = 0.0082, ∗∗∗p = 0.0005, ∗∗∗∗p = 0.0001. D, the SERCA Ca²⁺-ATPase inhibitor, thapsigargin (TG, 1 μM) was added at 100 s followed by the addition of 1.8 mM Ca²⁺ at 400 s. Enhanced Ca²⁺ entry is highlighted. E, TG was added in EGTA buffer (blue trace) or in high Ca²⁺ buffer (gray trace). The bar graph shows the rate of [Ca²⁺]_cyt increase. t test, ∗p = 0.0133. F, tachyzoites were in suspension in a buffer with 1.8 mM Ca²⁺. 1.8 mM BAPTA (free acid) was added at 100 s to the experiment represented by the dark pink trace. TG at 1 μM was added to both experiments. Individual dots represent the rate of Ca²⁺ increase for three biological replicates. Data was analyzed using Student’s t test ∗p = 0.0188. G, parasites in suspension in a buffer without EGTA, so the concentration of Ca²⁺ would be the contaminating one, ∼10 μM. Hundred micromolar of EGTA was added at 100 s and 1.8 mM BAPTA, free acid, right after adding 1.8 mM Ca²⁺. The bar graph shows the slope values after adding BAPTA from three independent experiments. t test: ∗p = 0.0271. H, parasites in suspension in EGTA buffer and TG was added at 100 s. At 400 s, 1.8 mM Ca²⁺ was added to both experiments and 1.8 mM BAPTA immediately after Ca²⁺ for the experiment represented by the pink trace. t test ∗∗∗p = 0.003. I, 1.8 mM Ca²⁺ was added to parasites pre-incubated with the PMCA inhibitor Vanadate (beige trace) compared to the same experiment without Vanadate (blue trace). Parasites were suspended in EGTA buffer. t test ∗p = 0.0271. All statistical analysis was done with data from three independent biological experiments. SERCA, Sarco/ER Ca2+ ATPase.

Ca²⁺ entry and pumping into intracellular stores is essential for keeping stores replenished. To examine the relationship between the ER, the largest intracellular store in cells (21), and Ca²⁺ entry, we manipulated the extracellular Ca²⁺ concentration. We found that the rate of Ca²⁺ entry after adding extracellular Ca²⁺ was lower when parasites were pre-incubated with 1.8 mM Ca²⁺ (Fig. 1B, gray trace) and was faster if they were pre-incubated with EGTA (100 μM) (Fig. 1B, light blue trace). We next analyzed the rate of Ca²⁺ entry in parasites pre-exposed to EGTA buffer by various lengths of time. The longer the parasites stayed in the EGTA buffer, the faster was the rate of influx, Δ[Ca²⁺]_i/sec, upon addition of extracellular Ca²⁺ (Fig. 1C). This could be the result of stores becoming depleted with the longer incubation in EGTA buffer, which could trigger a higher rate of PM Ca²⁺ entry.

In T. gondii, Ca²⁺ uptake into the ER is actively mobilized by the Sarco/ER Ca²⁺ ATPase (SERCA) homolog (TgSERCA) which, as in other eukaryotes, is sensitive to the SERCA inhibitor, thapsigargin (TG) (22, 23). Inhibition of SERCA by TG resulted in an increase of cytosolic Ca²⁺ due to the ER constitutive efflux/leakage pathway (Fig. 1D, TG-triggered Ca²⁺ increase). We observed that addition of extracellular Ca²⁺ to suspensions previously exposed to TG resulted in enhanced Ca²⁺ entry, which could be due to a store filling requirement or regulation by cytosolic Ca²⁺. The cytosolic increase is followed by a recovery phase (Fig. 1D). To assess the contribution of the TG-triggered cytosolic Ca²⁺ increase to the stimulation of calcium entry, we quantified Ca²⁺ entry after the addition of various concentrations of TG. The rate of Ca²⁺ entry, (Δ[Ca²⁺]_i/sec) was higher when adding Ca²⁺ after 1 μM TG (Fig. S1C). The rate of TG-triggered cytosolic Ca²⁺ (Δ[Ca²⁺]/sec) showed correlation with the concentration of TG added (Fig. S1D). However, the cytosolic Ca²⁺ increase (Δ[Ca²⁺]) only showed significance between 0.5 and 2 μM TG (Fig. S1E). In a similar experiment but in the presence of extracellular Ca²⁺, the rate and amplitude of cytosolic Ca²⁺ increase triggered by TG was dramatically increased (Fig. S1, F–H). However, the rate of Ca²⁺ entry was significantly lower with higher concentrations of TG (Fig. S1F, bar graph). This slower response could be because of the stores being fully replenished or could be an off-target effect of the higher concentration of TG (24). More experiments are needed to clarify the link between the filling state of the stores and its communication with the PM entry mechanism.

The cytosolic Ca²⁺ increase triggered by the addition of TG was significantly higher in cells preincubated in 1.8 mM extracellular Ca²⁺ compared to cells suspended in an EGTA buffer (∼30 nM extracellular Ca²⁺ calculated with maxchelator) (Fig. 1E, compare gray and blue traces). This could be because of increased Ca²⁺ entry or because of higher ER Ca²⁺ stored or a combination of both. To highlight Ca²⁺ entry, we added 1.8 mM BAPTA (free acid) prior to TG to block extracellular Ca²⁺, which resulted in a reduction of the TG-triggered cytosolic increase (Fig. 1F, red trace), indicating that the difference between plus and minus extracellular BAPTA is due in part to Ca²⁺ entry.

The concentration of cytosolic Ca²⁺ is the result of Ca²⁺ influx through the PM and the pumping of Ca²⁺ out by PMCA or into the ER by SERCA. To underline the PMCA’s role in regulating cytosolic Ca²⁺, we blocked the flow of Ca²⁺ entry by adding BAPTA a few seconds after adding extracellular Ca²⁺. We saw a sharp decrease in cytosolic Ca²⁺ after the addition of BAPTA followed by steady state of cytosolic calcium (Fig. 1G). We attributed this effect to the action of the PMCA as BAPTA prevented extracellular Ca²⁺ to re-enter the cytosol, which was masking the pumping activity. This sharp decrease was also seen when adding BAPTA after Ca²⁺ in the presence of TG (Fig. 1H) to block the contribution of the ER. Further demonstration for the role of the PMCA was shown by pre-incubating parasites with vanadate, an inhibitor of P-type ATPases (Fig. 1I), which resulted in a higher rate of Ca²⁺ entry demonstrating the activity of the enzyme in maintaining cytosolic Ca²⁺ at physiological levels. The Δ[Ca²⁺] between the peak and the baseline reached with BAPTA represent the amount of Ca²⁺ that is pumped out by the action of the PMCA, which is approximately 35% and 45% of the cytosolic Ca²⁺ rise after adding Ca²⁺ with and without TG, respectively (Fig. 1, G and H). Ca²⁺ was extruded from the cytosol at a rate similar to the rate seen in cells pretreated with TG, supporting a more important role for the PM Ca²⁺ pump over SERCA in maintaining cytosolic Ca²⁺. This result supports the housekeeping role of PMCA as described for other PMCAs (25, 26).

In summary, these data showed the presence of Ca²⁺ channels at the PM that permeates Ca²⁺ and that are compensated by the action of a PM Ca²⁺ pump to maintain cytosolic Ca²⁺ homeostasis (27, 28). In addition, depletion of intracellular Ca²⁺ by either incubating the cells with extracellular EGTA or BAPTA or by inhibition of SERCA stimulated Ca²⁺ influx (enhanced Ca²⁺ entry).

Ca²⁺ entry pathways with distinct Ca²⁺ affinities and blocker’s sensitivities

With the aim of further characterizing the channel(s) responsible for PM Ca²⁺ entry, Fura-2–loaded tachyzoites in suspension in EGTA buffer were exposed to varying concentrations of extracellular Ca²⁺ (Figs. 2A and S2A). We measured the peak cytosolic Ca²⁺ and found that it increased with the concentration of extracellular Ca²⁺ added to the suspension in a biphasic shape (Fig. 2A). This pattern indicated the presence of more than one type of Ca²⁺ channel for influx with at least two different affinities for Ca²⁺ (Fig. 2A). We next examined the effect of known mammalian Ca²⁺ channel blockers like nifedipine, cilnidipine, verapamil (29), or ACA (30), a wide spectrum TRP channel inhibitor, on the ΔCa²⁺ following the addition of extracellular Ca²⁺. Inhibition by 100 μM verapamil, a phenylalkylamine, was ∼40% and no inhibition was observed with diltiazem, a benzothiazepine type voltage-dependent calcium channel (VDCC) antagonist (Table 1). Dihydropyridines, known blockers of L-type VDCC like nifedipine and cilnidipine, were highly effective and blocked Ca²⁺ influx at ∼80% and ∼75%, respectively (Table 1). Interestingly, ACA inhibited Ca²⁺ influx by approximately 50% (Table 1). Since dihydropyridines and ACA block different types of channels, we next tested these inhibitors at varying concentrations of extracellular Ca²⁺. At 250 μM extracellular Ca²⁺, nifedipine blocked entry at almost ∼100% (Fig. 2, B and C, Table 2), while at higher concentrations of extracellular Ca²⁺ (1.8 mM), a residual ∼20% activity of Ca²⁺ influx was detected (Table 2). This result indicated that at high concentrations of extracellular Ca²⁺, the partial inhibition is due to the activity of more than one Ca²⁺ channel while at lower extracellular Ca²⁺, the nifedipine-sensitive channel appeared to be the main active channel. On the other hand, ACA inhibited Ca²⁺ entry by ∼50% at all tested extracellular [Ca²⁺] (Fig. 2, D and E). The combination of both nifedipine and ACA resulted in an additive inhibition of Ca²⁺ entry, as most likely they block different channels (Fig. 2F). Taken together these results supported the presence of more than one type of Ca²⁺ channel with different affinities for Ca²⁺ and inhibited by different types of blockers.

Figure 2.

Two types of Ca²⁺channels are active at the PM in Toxoplasma gondii tachyzoites.A, cytosolic concentration of Ca²⁺ obtained after adding the indicated extracellular Ca²⁺. Values were taken from similar graphs as the one presented in S2A. Δ[Ca²⁺]_i for the quantifications was calculated as Δ[Ca²⁺] between the peak and basal Ca²⁺. B, 250 μM Ca²⁺ (gray trace) was added to a suspension of Fura-2–loaded tachyzoites in EGTA buffer compared with an identical experiment with parasites pre-incubated with 10 μM nifedipine (red trace). C, quantification of Δ[Ca²⁺]_i in control and nifedipine-treated parasites at various extracellular Ca²⁺ concentrations. Each dot represents an independent biological replicate. Data was analyzed using 2-way ANOVA, ∗∗p = 0.0047, ∗∗∗p = 0.0001, ∗∗∗∗p < 0.0001. D, Ca²⁺ entry stimulated with 250 μM extracellular Ca²⁺ (gray trace) compared with an identical experiment of cells pre-incubated with 1 μM anthranilic acid (ACA) (blue trace). E, quantification of Δ[Ca²⁺]_i in control and ACA-treated parasites after adding different concentrations of extracellular Ca²⁺. Each dot represents an independent biological replicate. Data was analyzed using 2-way ANOVA analysis ns p > 0.1, ∗∗∗p = 0.0001, and ∗∗∗∗p < 0.0001. F, Fura-2–loaded T. gondii tachyzoites in suspension. 1.8 mM Ca²⁺ was added were indicated. Ten micromolars of Nifedipine or/and 1 μM ACA were added at 50 s or/and 150 s, respectively. Each dot represents an independent biological replicate. Data was analyzed using one-way ANOVA analysis ∗∗p = 0.0012, ∗∗p = 0.0083, and ∗p = 0.028. G and H, parasites in suspension in EGTA buffer and pre-loaded with BAPTA-AM (iBAPTA). 1.8 mM extracellular Ca²⁺ was added at 400 s. The red trace in (G) represents an experiment with parasites pre-incubated with Nifedipine. The blue trace in (H) represents an experiment with parasites pre-incubated with ACA. I, quantification and statistical analysis of the Δ[Ca²⁺]_i after adding extracellular Ca²⁺. Individual dots represent biological replicates. Data was analyzed using one-way ANOVA analysis ∗∗∗p = 0.0006, ∗∗∗∗p < 0.0001. J, quantification and statistical analysis of Δ[Ca²⁺]_i after adding extracellular Ca²⁺, in the presence of inhibitors after chelating cytosolic Ca²⁺ with BAPTA. Each dot represents an independent biological replicate. Data was analyzed using one-way ANOVA analysis ∗p = 0.04, ∗∗∗∗p < 0.0001. A, ACA; C, control; iBAPTA, intracellular BAPTA; N, Nifedipine; PM, plasma membrane.

Table 1.

Percentage of inhibition of Ca²⁺ entry

Compounds	% Inhibitiona
40 μM Cilnidipine	78.8 ± 4.6
10 μM Nifedipine	84.7 ± 5.7
100 μM Verapamil	42.7 ± 2.3
1 μM ACA	50.3 ± 10.7

± Standard error was calculated with four independent experiments.

^aControl (0%) was established as Ca²⁺ entry as shown in Figure 1A without inhibitor.

Table 2.

Nifedipine inhibition of Ca²⁺ entry at various extracellular [Ca²⁺]

[Ca2+]E (mM)	% Inhibition by NIFa
0.1	98.6 ± 0.44
0.25	96.5 ± 0.83
0.5	97.2 ± 0.95
1	86.0 ± 4.59
1.8	64.8 ± 4.95

^aNIF = Nifedipine.

We previously showed that cytosolic Ca²⁺ itself could modulate/activate one of the Ca²⁺ channels at the PM (12). With the aim to highlight this activity of Ca²⁺ entry at the PM, we pre-loaded T. gondii tachyzoites with BAPTA-AM to chelate cytosolic Ca²⁺ (intracellular BAPTA). Under these conditions, the “basal Ca²⁺ entry” was significantly reduced as shown after adding extracellular Ca²⁺ (Fig. 2, G and H). Interestingly, under these conditions, nifedipine blocked Ca²⁺ entry by 100% (Fig. 2G) while ACA inhibited the usual 50% (Fig. 2, H–J). This result showed that chelating intracellular Ca²⁺ inhibited Ca²⁺ entry and this inhibition was complete when combined with nifedipine in support for the presence of at least two Ca²⁺ entry pathways (one sensitive to nifedipine and one modulated by cytosolic Ca²⁺). We believe that the residual activity observed in the presence of ACA or intracellular BAPTA is due most likely to Ca²⁺ influx through the nifedipine-sensitive channel (Fig. 2, H–J).

We also studied constitutive Ca²⁺ influx which was observed in extracellular tachyzoites when exposed to 1.8 mM extracellular Ca²⁺. This is evidenced by a steady increase of cytosolic Ca²⁺ over time, which is not observed when the parasites are suspended in EGTA buffer (Fig. S2B). ACA inhibited this activity almost 100% while nifedipine showed only a minor effect (Fig. S2B). This result showed that at least one of the PM channels was leaky, as it allowed constitutive Ca²⁺ influx, and it was sensitive to ACA but not to nifedipine.

Overall Ca²⁺ influx in extracellular tachyzoites of T. gondii is occurring through at least two different types of channels; one channel sensitive to nifedipine with higher affinity for Ca²⁺ (VDCC-like channel) and a channel with lower affinity for Ca²⁺, sensitive to ACA and modulated by cytosolic Ca²⁺ (TRP-like channel).

The role of cGMP signaling in Ca²⁺ entry

With the aim to further characterize Ca²⁺ entry activated by cytosolic Ca²⁺, we tested zaprinast, a cGMP phosphodiesterase inhibitor (Fig. 3A), known to increase cytosolic Ca²⁺ (31). We observed that Ca²⁺ entry was significantly increased after stimulation with zaprinast (Fig. 3B, violet trace) compared to the control rate of entry (Fig. 3B, blue trace, no previous additions). As the cytosolic Ca²⁺ decreases after the peak ensuing from the addition of zaprinast (recovery phase) (Fig. S3A), we added Ca²⁺ at different times after the peak and found that the magnitude of Ca²⁺ entry decreased as the cytosolic Ca²⁺ decreased (Fig. S3A). Adding zaprinast in the presence of extracellular Ca²⁺ resulted in a cytosolic increase from a combination of release from stores and Ca²⁺ entry through the PM (Fig. 3C, brown trace). We demonstrated this by adding extracellular BAPTA (free acid) right before adding zaprinast which would not allow for Ca²⁺ to enter (Fig. 3C, green trace), resulting in a diminished cytosolic Ca²⁺ peak indicating that the difference between both conditions (with and without BAPTA) is due to Ca²⁺ entry (Fig. 3C, green trace). This difference is not found in the control experiments done in the absence of extracellular Ca²⁺ (Fig. S3B).

Figure 3.

Calcium entry in Toxoplasma gondii tachyzoites and the effect of zaprinast.A, diagram illustrating Ca²⁺ entry and the potential participation of the cGMP signaling pathway stimulated by zaprinast. B, extracellular tachyzoites in suspension loaded with Fura-2 treated with zaprinast (Zap, 100 μM) at 100 s and extracellular Ca²⁺ (1.8 mM) at 400 s. Bar graph shows quantification of Δ[Ca²⁺]_i after adding extracellular Ca²⁺ from more than five independent experiments. Each dot represents an independent biological replicate. Data was analyzed using Student’s t test, ∗∗p = 0.0039. C, tachyzoites in buffer with 1.8 mM extracellular Ca²⁺. Zaprinast (100 μM) was added at the indicated time by itself (brown trace) or preceded by the addition of 1.8 mM BAPTA (green trace). Bar graph shows the rate of cytosolic Ca²⁺ increase in response to Zap. Individual dots represent biological replicates. Data was analyzed using Student’s t test ∗p = 0.0354. D, tachyzoites in EGTA buffer were first treated with 100 μM zaprinast (100 s) followed by the addition of 1.8 mM Ca²⁺ at 400 s (control, purple trace). An identical experiment was done with parasites pre-incubated with 10 μM Nifedipine (orange trace). Bar graph shows quantification of Δ[Ca²⁺]_i after adding extracellular Ca²⁺. Each dot represents an independent biological replicate. Data was analyzed using Student’s t test ∗∗∗p = 0.0003. E, similar experiment to the one in (D) but replacing Nifedipine with 1 μM ACA. Bar graph shows quantification of Δ[Ca²⁺]_i after adding extracellular Ca²⁺. Each dot represents an independent biological replicate. Data was analyzed using Student’s t test ∗p = 0.02. F, addition of 100 μM zaprinast to parasites in suspension in a buffer containing 1.8 mM extracellular Ca²⁺ (purple trace) and the same experiment with parasites pre-incubated with Nifedipine (10 μM) (orange trace). Bar graph shows quantification of Δ[Ca²⁺]_i after adding zaprinast. Individual dots represent biological replicates. Data was analyzed using Student’s t test ns p = 0.8. G, similar experiment to the one in (F) but with 1 μM ACA. Each dot represents an independent biological replicate. Data was analyzed using Student’s t test ∗p = 0.012. CGMP, cyclic GMP.

We next tested nifedipine which inhibited Ca²⁺ entry by ∼80% (Fig. 3D, orange trace, Ca²⁺ was added 300 s after zaprinast). Note that nifedipine did not inhibit the zaprinast stimulation of calcium release from intracellular stores, as it was acting only on the PM channel. Addition of ACA inhibited Ca²⁺ entry by ∼50% (Fig. 3E) and did not impact the zaprinast-stimulated calcium release. Interestingly, when testing nifedipine in the presence of extracellular Ca²⁺, a condition in which the stores are filled, Ca²⁺ entry stimulated by zaprinast was insensitive to nifedipine (Fig. 3F) while ACA inhibited ∼70% of the entry (Fig. 3G). This result indicated that the PM Ca²⁺ entry channel activated by the signaling pathway triggered by zaprinast is nifedipine-insensitive but can still be inhibited by ACA. Note that the inhibition of Ca²⁺ entry observed in Figure 3, D and E is measured after recovery of cytosolic Ca²⁺.

To better understand the interplay between cGMP and Ca²⁺ entry, we added a permeable version of cGMP to a suspension of Fura-2–loaded parasites in EGTA buffer, which resulted in an increase in cytosolic Ca²⁺ most likely from intracellular stores. Addition of extracellular Ca²⁺ after cGMP resulted in enhanced Ca²⁺ entry (Fig. 4A, green trace). cGMP activates the enzyme PKG and to ensure that the effect of cGMP was through PKG, we pre-incubated the parasite suspensions with compound 1 (Cpd1), a specific inhibitor of PKG (32). The cytosolic-enhanced Ca²⁺ entry stimulated by cGMP was suppressed by Cpd1, supporting a role for PKG (Fig. 4A, gray trace). The basal Ca²⁺ entry was slightly affected by Cpd1 (Fig. 4, B and C, compare red and blue traces and bars after adding Ca²⁺). However, the enhanced Ca²⁺ entry was fully inhibited (Fig. 4C, compare green and gray bars).

Figure 4.

Ca²⁺entry and cGMP.A, Fura-2–loaded extracellular tachyzoites in suspension were used. Two hundred micromolars of cGMP was added to the suspension in EGTA buffer at 100 s (green and gray traces). Extracellular Ca²⁺ (1.8 mM) was added at 400 s. The gray trace shows a similar experiment with parasites pre-incubated with compound 1 (1 μM). The blue trace represents tachyzoites to which extracellular Ca²⁺ was added at 400 s without extra additions (basal Ca²⁺ entry). B, compound 1 (Cpd1) (1 μM) was added (maroon trace) at 100 s and extracellular Ca²⁺ (1.8 mM) was added at 400 s. C, quantification and statistical analyses of the change in [Ca²⁺]_i measured after addition of extracellular Ca²⁺ without additions (blue bar) or after adding 200 μM cGMP (green bar) or 1 μM Cpd1 (maroon bar) or both (dark Gray bar). Bar graphs represent the statistical analysis from three independent biological replicates using one-way ANOVA analysis; ∗∗∗p = 0.001, ∗p = 0.0343. Individual dots represent biological replicates. D and E, extracellular tachyzoites of the PKG-T mutant (see experimental details) loaded with Fura-2. One micromolar of Cpd1 was added at 50 s (maroon trace), 1.8 mM Ca²⁺ at 400 s, and 200 μM cGMP at 250 s (green trace in E). Bar graphs show the quantification and statistical analysis of Δ[Ca²⁺]_i after adding extracellular Ca²⁺. Individual dots represent biological replicates. Data was analyzed with Student’s t test; ∗p < 0.05. F and G, similar experiments and conditions to the ones used in (D) and (E) but using the PKG-M strain, which expressed PKG (T⁷⁶¹M) resistant to Cpd 1. Individual dots represent biological replicates. Data was analyzed using Student’s t test ns p > 0.05. CGMP, cyclic GMP; PKG, protein kinase G.

To further examine the role of PKG in Ca²⁺ entry, we tested a PKG mutant that is insensitive to Cpd1 (PKGM) and its counterpart PKGT (sensitive to Cpd1). PKGM and PKGT are genetically modified strains in which the PKG gene was replaced by a Ty-tagged allele harboring either the WT gatekeeper (PKGT) or a T⁷⁶¹M mutation (PKGM) (31). Change of the PKGT⁷⁶¹M residue at the base of the ATP-binding pocket made PKG refractory to Cpd1 inhibition. The PKGM mutant was insensitive to growth inhibition by Cpd1 while the PKGT was sensitive (Fig. S4A). Consistent with the growth result, the PKGT mutant showed reduced basal Ca²⁺ entry by 60% in the presence of Cpd1 (Fig. 4D, maroon trace and bar), and enhanced Ca²⁺ entry induced by cGMP was inhibited by ∼75% (Fig. 4E, maroon trace and bar). Cpd1 did not inhibit basal Ca²⁺ entry in the PKGM mutant (Fig. 4F) or the cGMP mediated Ca²⁺ release followed by enhanced Ca²⁺ entry (Fig. 4G). In summary, we showed that the mechanism of Ca²⁺ entry enhanced by cytosolic Ca²⁺ is linked to a signaling pathway in which PKG forms part of.

The role of the PI-PLC in Ca²⁺ entry

It has been proposed that the role of PKG in Ca²⁺ signaling could be through its regulation of PI-PLC (33) (Fig. 5A). To test the potential role of PI-PLC in Ca²⁺ entry, we used the inhibitor U73122 and its inactive analog U73343 (34, 35) (Fig. S4B, inhibitor effect on growth). Pre-incubation with the inhibitor blocked the effect of cGMP on intracellular Ca²⁺ and significantly decreased Ca²⁺ entry after the addition of extracellular Ca²⁺ (Fig. 5B, brown trace). Pre-incubation with the inactive analog, however, resulted in Ca²⁺ responses indistinguishable from the control (Fig. 5B, gray trace). Interestingly, parasites incubated with U73122 and nifedipine, blocking both potential routes of Ca²⁺ influx, showed 100% inhibition of Ca²⁺ entry (Fig. 5C, orange trace) further substantiating the presence of at least two pathways responsible for Ca²⁺ entry.

Figure 5.

The role of PI-PLC in Ca²⁺entry.A, diagram illustrating a model of the role of cGMP, PKG, and PI-PLC in Ca²⁺ entry. B, Fura-2–loaded tachyzoites in EGTA buffer (100 μM EGTA). One micromolar of U73122 (brown trace) or U73343 (green trace) were added at 50 s. Two hundred micromolars of cGMP and 1.8 mM extracellular Ca²⁺ were added where indicated. Bar graphs show the statistical analysis from three independent biological replicates using one-way ANOVA, ns p = 0.08, ∗∗p < 0.002. Individual dots represent biological replicates. C, the experimental set-up was similar to the one presented in (B) but parasites were pre-incubated with 10 μM Nifedipine (orange trace) and U73122 was added at 50 s. Statistical analysis of three independent experiments using t test ∗∗p = 0.003. For experiments shown in (D–K), extracellular tachyzoites of the iΔPIPLC mutant (± ATc for 48) were Fura-2–loaded. Tachyzoites were in suspension in EGTA buffer and 1.8 mM Ca²⁺ was added at 400 s. Pink traces represent the results with the mutant pre-incubated with ATc and green traces represent the same mutant without ATc. D, the basal Ca²⁺ entry was measured by adding 1.8 mM Ca²⁺ at 400 s. E, at 100 s, 1 μM thapsigargin (TG) was added and 1.8 mM Ca²⁺ at 400 s. F, quantification of the Δ[Ca²⁺]_i from parts (D) and (E). ANOVA analysis ∗∗p = 0.001, ∗∗p = 0.005, and ns p > 0.05. G, similar experiment to the one in (E) but with 100 μM zaprinast added at 100 s. Data was analyzed using one-way ANOVA, ∗∗∗p = 0.0004, ∗∗p = 0.005, and ns p > 0.05. Each dot represents an independent biological replicate. H, similar experiment to the one in (G), but 1 μM U73122 was added at 100 s, followed by 1.8 mM Ca²⁺ at 400 s. Individual dots represent biological replicates. One-way ANOVA, ∗∗p = 0.003 and ns p > 0.05. I, identical set-up to (H), but the inactive analog U73343 was added at 100 s. Data from three biological experiments was analyzed using one-way ANOVA ns p > 0.05. For (G), (H), and (I), the one-way ANOVA analysis includes the Ca²⁺ basal control data from (F). J, similar experiment to the one shown in (D) with tachyzoites (+ATc) pre-incubated with Nifedipine, 10 μM (gray trace). Individual dots represent biological replicates. Data was analyzed using Student’s t test ∗p = 0.014. K, similar experimental set-up with 200 μM cGMP added at 100 s and 1.8 mM Ca²⁺ at 400 s. The dotted pink trace represents the result with the iΔPIPLC mutant (+ATc for 48 h) pre-incubated with 10 μM Nif. Individual dots represent biological replicates. Data was analyzed using One-way ANOVA, ∗∗p = 0.009, ∗∗p = 0.002, ∗p = 0.01 and ns p > 0.05. Bar graphs for (B–K) shows the statistical analysis of a minimum of three independent experiments for which the Δ[Ca²⁺]_i was measured. ATc, anhydrotetracycline; cGMP, cyclic GMP; PI-PLC, phosphoinositide phospholipase C; PKG, protein kinase G.

To further characterize the role of PI-PLC in Ca²⁺ entry, we tested the conditional mutant for PI-PLC, iΔPIPLC, in which the expression of the PI-PLC gene is regulated by anhydrotetracycline (ATc) (36). We cultured the mutant in the presence of ATc (+ATc) and compared Ca²⁺ entry with the same mutant grown without ATc (-ATc). The basal Ca²⁺ entry without any stimulation was comparable between +ATc and -ATc parasites (Fig. 5D). We next looked at the enhanced Ca²⁺ entry by adding TG followed by extracellular Ca²⁺. Downregulation of PI-PLC almost entirely abolished the enhanced Ca²⁺ entry and the activity observed was similar to the basal Ca²⁺ entry (Fig. 5E). Quantifications are shown in part F. Most interestingly, the zaprinast-stimulated cytosolic Ca²⁺ influx was highly reduced in the iΔPIPLC (+ATc) mutant and resulted in diminished Ca²⁺ stimulated Ca²⁺ entry (Fig. 5G). This result supports the involvement of PI-PLC in the pathway stimulated by zaprinast on intracellular stores and places PI-PLC as an important player in the Ca²⁺-activated Ca²⁺ entry (or enhanced Ca²⁺ entry).

We tested the effect of U73122 on the iΔPIPLC (+ATc) mutant and showed that they were insensitive as no effect on Ca²⁺ entry was observed upon extracellular Ca²⁺ addition (Fig. 5H, pink trace, compare with the pink trace in part D). The inactive analog, U73343, however, produced a similar response on the iΔPIPLC mutant cultured with or without ATc (Fig. 5I). The basal Ca²⁺ entry of the iΔPIPLC (+ATc) mutant was sensitive to nifedipine at the same level as the parental line indicating that basal Ca²⁺ entry is the result of at least two activities (Fig. 5J, gray trace). Enhanced Ca²⁺ entry stimulated by cGMP was also significantly diminished in the ΔPIPLC (+ATc) mutant (Fig. 5K, pink trace) and if pre-incubated with nifedipine resulted in ∼100% inhibition because both pathways of Ca²⁺ influx were blocked (Fig. 5K, dotted pink trace). These results supported the participation of PI-PLC in a Ca²⁺ entry pathway modulated by cytosolic Ca²⁺ through PKG. The enhanced Ca²⁺ entry would be the combination of the basal Ca²⁺ entry and the Ca²⁺ influx induced by PKG and PI-PLC.

In summary, our results showed that Ca²⁺ entry through an ACA sensitive channel is modulated by Ca²⁺ through a signaling pathway that involves cGMP, PKG, and PIPLC.

Ca²⁺ entry and the lytic cycle

Intracellular tachyzoites replicate inside a porous vacuole which is in equilibrium with the host cytosol where the concentration of Ca²⁺ is ∼100 nM. Under these conditions, parasites can take up Ca²⁺, during physiological host Ca²⁺ fluctuations, which are also impacted by extracellular Ca²⁺ (11). We investigated if the extracellular Ca²⁺ in the culture would influence parasite growth. We measured T. gondii growth using plaque assays and varied the concentration of extracellular Ca²⁺. Parasites engage in repetitive cycles of invasion, replication, and egress causing host cell lysis, and formation of plaques is observed as white spots after staining with crystal violet. Plaque sizes were measured in cultures with 1.8, 0.5, and 0.25 mM extracellular Ca²⁺ after 8 days of growth and, interestingly, we found that plaques were smaller at lower concentrations of Ca²⁺ (Fig. 6A). This result indicated that intracellular tachyzoite growth is impacted by the extracellular Ca²⁺ conditions. Noninfected host cells grew normally under similar conditions although they did not grow in the absence of Ca²⁺ (Fig. S5A and not shown).

Figure 6.

Extracellular calcium and Toxoplasma gondii growth and invasion.A, plaque assays of T. gondii RH tachyzoites cultured in regular growth media without Ca²⁺ and supplemented with Ca²⁺ 1.8 mM, 0.5 mM, or 0.25 mM Ca²⁺. Confluent fibroblast cells were infected with 200 tachyzoites for 8 days. The bar graph shows the quantification of the plaque areas from a minimum of four independent experiments. Individual dots represent biological replicates. Data was analyzed using one-way ANOVA, ∗∗∗∗p < 0.0001. B, plaque assays of T. gondii RH tachyzoites grown in regular media in the presence of the indicated inhibitors. Cilnidipine, 40 μM; compound 1 (Cpd1), 1 μM; anthranilic acid (ACA), 1 μM. The bar graph shows the quantification of the plaque areas from a minimum of four independent experiments. Statistical analysis was done using one-way ANOVA, ∗∗∗∗p < 0.0001. Individual dots represent biological replicates. C, invasion and attachment experiments using the red/green assay of tachyzoites. Confluent fibroblast monolayers were infected with 2 × 10⁷ tachyzoites. Inhibitors were added 5 min prior to the invasion step. Parasites were stained with RabbitαSAG1 at 1:1000 dilution and MouseαSAG1 at 1:200 dilution. Secondary antibodies were goat-αRabbit and Alexa Flour 546 and Alexa Flour 488 goat-αMouse, both at 1:1000 dilution. C, control; Nif, nifedipine, 10 μM; ACA, anthranilic acid, 1 μM; Zap, zaprinast, 100 μM; U22, U73122 1 μM; U33, U73233, 1 μM. 100% is the sum of parasites that have either attached or invaded the host cell in 10 fields of view. Individual dots represent biological replicates. Data was analyzed using one-way ANOVA ∗p < 0.05, ∗∗p < 0.001, and ∗∗∗∗p < 0.0001.

We next investigated growth inhibition by nifedipine, cilnidipine, ACA, and Cpd1 at the concentrations we observed inhibited Ca²⁺ influx, 10 μM, 40 μM, 1 μM, and 1 μM respectively. Nifedipine did not inhibit growth, most likely due to the poor stability of the compound in cultures (Fig. 6B). Nifedipine solutions are extremely unstable and photosensitive and rapidly degrades at 25 °C, and the concentration declines to about 90% within 6 h (Sigma product information). Plaques formed in the presence of ACA were significantly smaller, indicating that Ca²⁺ influx mediated by the TRP-like channel is important for parasite growth. No plaques formed when incubated with Cpd1, a stronger inhibition than the one observed with ACA due to other essential roles of PKG. Ca²⁺ influx through a channel with pharmacological characteristic of a VDCC is important as shown by the lack of plaques in the presence of cilnidipine. Noninfected host cells controls showed no toxicity when grown in the presence of the inhibitors at similar concentrations (Fig. S5B). However, U73122 showed a minor effect in the integrity of the host cell monolayer (Fig. S5B).

We next determined the EC₅₀ (concentration of drug needed to inhibit growth by 50%) of cilnidipine, ACA, and other inhibitors (Table 3). Cilnidipine EC₅₀ was 10 times lower than the effective concentration for inhibiting Ca²⁺ influx, which suggests an additional target. Verapamil showed a modest effect with an EC₅₀ of ∼31 μM. ACA was highly effective with an EC₅₀ of 1.6 μM. The EC₅₀ for the PI-PLC inhibitor U723122 was 0.35 μM and the inactive analog, U73343 was less effective with an EC₅₀ of 2.4 μM. This result was consistent with the specific PI-PLC inhibition properties of both compounds (Table 3). We also tested invasion and attachment and observed inhibition by nifedipine, but no effect was observed with cilnidipine (Fig. 6C). Reduced invasion and increased attachment were also observed in the presence of zaprinast or U73122 (Fig. 6C). The U73122 showed a stronger inhibition of invasion than its inactive analog U73343 (Fig. 6C), consistent with the growth inhibition.

Table 3.

Inhibition of growth by calcium-related inhibitors and channels blockers (EC₅₀)

Compounds	EC50 (μM)a
Cilnidipine	3.5 ± 1.7
Nifedipine	N/A
Verapamil	30.7 ± 6.9
ACA	1.6 ± 0.3
Compound 1	0.11 ± 0.04
Zaprinast	200b
U73122	0.35 ± 0.1
U73343	2.4 ± 0.8

(N/A): Not possible to determine due to instability of the compound.

^aThe EC₅₀ was calculated from three independent growth experiments.

^bSidik et al, 2016.

To further characterize Ca²⁺ influx during egress, we used pharmacological inhibition of the potential Ca²⁺ entry channels and measured time to egress and quantified fluorescence fluctuations of tachyzoites expressing the genetically encoded calcium indicator, GCaMP6f (Fig. S7A). The experimental strategy was to permeabilize the host cell with a very low concentration of saponin (0.01%), added 50 sec after the start of the recording, which will expose the PVs to either high (1.8 mM) or low (∼30 nM) extracellular Ca²⁺. Under these conditions, egress is stimulated as we previously demonstrated (11). Previously, we showed that two peaks of cytosolic Ca²⁺-preceded egress: the first peak originating from intracellular stores followed by a second peak associated with Ca²⁺ influx. This Ca²⁺ influx was sensitive to nifedipine (11). Interestingly, at low extracellular Ca²⁺, ACA significantly delayed parasite egress by affecting the first peak (Fig. 7A, EGTA), while at high extracellular Ca²⁺, there was no difference in Ca²⁺ fluctuations or time of parasites to egress (Fig. 7B, high Ca²⁺). We showed previously that ACA targets the TgTRPPL-2 channel, which localizes to the PM and the ER of T. gondii (12). Deletion of TgTRPPL-2 decreased both ER calcium efflux and Ca²⁺ entry. Since ACA targets both ER- and PM-localized TgTRPPL-2, it would explain the delayed egress observed in the EGTA buffer. When stimulating with zaprinast, ACA did not affect the time of egress at low or high extracellular Ca²⁺ (Fig. S7, A and B). This could be because zaprinast alone can sustain the Ca²⁺ threshold needed to stimulate egress. However, at high extracellular Ca²⁺, ACA produced a modest shift of the first peak of cytosolic Ca²⁺ oscillation which was not statistically different (Fig. S7B).

Figure 7.

Calcium entry and egress.A, control experiment in EGTA buffer (100 μM EGTA). 0.01% saponin was added at 50 s. Fluorescence of single parasites was analyzed with FIJI. F/F₀ represents the normalized fluorescence of single parasites to the resting fluorescence prior to adding saponin. The bar graph shows the statistical analysis of the time to egress (of the leading parasite) from a minimum of three independent experiments. t test analysis, ∗∗p = 0.004. Scale bars represent 5 μm. B, similar experimental set-up to the one in (A) but the buffer contains 1.8 mM Ca²⁺. Bar graph shows quantification of time of egress for the leading parasite of the vacuole. t test analysis ns p = 0.006. Scale bars represent 5 μm. C, GCaMP6f-expressing parasites immobilized in 10% FBS were preincubated with ACA, cilnidipine, or nifedipine. After 1 min of video recording, 1.8 mM Ca²⁺ was added. Violin plot of relative distance shows parasites distribution before and after Ca²⁺ addition. t test ∗∗∗∗p < 0.0004. D and E, relative distance travel by the tachyzoites after 100 μM zaprinast addition in low (D) and high (E) Ca²⁺ buffer. Plot shows data from three independent biological experiments t test ∗∗∗∗p < 0.0001, ∗∗p < 0.005 and t test ∗∗p < 0.005, ∗p < 0.05, respectively.

Finally, we examined the effect of Ca²⁺ inhibitors on parasite motility. We evaluated the relative distance traveled by parasites before and after addition of 1.8 mM Ca²⁺ (Fig. S6B). Using a cell-tracking algorithm (37) and GCaMP6f-expressing parasites, we were able to confirm that extracellular Ca²⁺ enhances motility and this stimulation was inhibited by nifedipine, cilnidipine, or ACA evaluated as the distance traveled by the parasite (Fig. 7C). Next, we tested parasite motility stimulated by zaprinast. We monitored distance traveled in low (Fig. 7D) and high (Fig. 7E) extracellular Ca²⁺ in the presence or absence of the inhibitors: nifedipine, ACA, or Cpd1. This experiment showed that inhibition of PKG by Cpd1 also diminished parasite motility at low and high extracellular Ca²⁺. Interestingly, motility was inhibited by ACA at low extracellular Ca²⁺ but not at high extracellular Ca²⁺. At low extracellular Ca²⁺, motility is likely triggered by intracellular store release and the effect of ACA at low Ca²⁺ could be the result of its inhibition of the ER Ca²⁺ leak (TgTRPPL-2). As the concentration of extracellular Ca²⁺ is low, the low affinity channel is likely nonfunctional so ACA would be irrelevant. When the extracellular Ca²⁺ is high, ACA has no effect as most likely Ca²⁺ is still able to enter the parasite through the high affinity channel.

Discussion

An increase of cytosolic Ca²⁺ in cells is the result of influx from the extracellular milieu through the PM and/or release from intracellular stores mainly from the ER. The resting cytosolic Ca²⁺ is highly regulated, and PM Ca²⁺ ATPases (PMCA) play an essential role by pumping excess cytosolic Ca²⁺ out of the cell (25). The ER Ca²⁺ reuptake is performed by the SERCA Ca²⁺ pump that controls the ER luminal [Ca²⁺]. A poorly defined Ca²⁺ leak in the ER protects the organelle from Ca²⁺ overload. However, this constitutive Ca²⁺ leak combined with other mechanisms at the PM like the PMCA would result in a continuous loss of stored Ca²⁺ from the organelles if a mechanism of entry at the PM would not be present (38).

In T. gondii, both intracellular (11) and extracellular parasites (10) use Ca²⁺ influx for replenishing intracellular stores or for enhancing invasion and motility traits, respectively. The increase of cytosolic Ca²⁺ resulting from both extracellular and/or intracellular influx can contribute to the activation of downstream signaling pathways that promote progression through the parasite’s lytic cycle. Ca²⁺ entry from the extracellular milieu would be the only logical mechanism by which intracellular and extracellular parasites can replenish their intracellular stores which are essential for signaling and lytic cycle progress. However, the mechanisms of Ca²⁺ entry are poorly understood, and few molecular players have been identified.

In this work, we focused on Ca²⁺ entry in extracellular tachyzoites, and we distinguish the basal entry that is observed right after adding extracellular Ca²⁺ (1.8 mM Ca²⁺) to parasites previously suspended in a low extracellular buffer (100 μM EGTA, ∼30 nM free Ca²⁺) from the enhanced entry observed after an increase of cytosolic Ca²⁺ triggered by TG, zaprinast, or cGMP (Ca²⁺ activated Ca²⁺ entry or CACE). This enhanced entry could also be the result of store regulation although previous results did not support the presence of this mechanism (10). We showed that more than one type of PM channel is most likely functional in extracellular tachyzoites of T. gondii with different affinities for Ca²⁺ and different pharmacological profiles. Our hypothesis is that these channels will become functional or activated by specific triggers that would allow them to open at specific points during the lytic cycle of the parasite. Ca²⁺ entry can be inhibited by VDCC blockers like nifedipine and cilnidipine and these inhibitors block Ca²⁺ entry by 80 to 85% at high extracellular Ca²⁺ (>1.5 mM Ca²⁺) and ∼100% at lower extracellular Ca²⁺ (<250 μM Ca²⁺). This inhibition pattern stressed the functioning of a second channel at high extracellular Ca²⁺ concentrations.

In this work, we described the regulation of a PM Ca²⁺ entry activity by Ca²⁺ itself through a signaling pathway that involves cGMP, PKG, and PI-PLC. This activity was sensitive to the broad spectrum TRP channel blocker ACA, supporting that the target of this drug is the previously described TgTRPPL-2 (12). TgTRPPL-2 localized to the PM and the ER, was found to conduct Ca²⁺ and was inhibited by ACA and benzamil. TgTRPPL-2 mutants showed reduced Ca²⁺ influx, a phenotype like the one resulting from ACA inhibition.

The processes of invasion and egress are highly regulated, dynamic, and essential for the propagation of the T. gondii infection. Each step of the parasite lytic cycle is precise, fast, and efficient with distinct and interrelated molecular processes occurring in a coordinated manner and each step is preceded by an increase of cytosolic Ca²⁺ (4, 5). The extracellular motile tachyzoite is surrounded by the high Ca²⁺ concentration of the extracellular milieu (>1.5 mM) and we previously demonstrated Ca²⁺ influx under these conditions, which occurred in a regulated fashion (10). Keeping intracellular Ca²⁺ stores (like the ER) replenished in replicating intracellular T. gondii is essential for the continuation of the parasite’s lytic cycle as exit from the host cell is preceded by a rapid required increase of the parasite’s cytosolic Ca²⁺ (9). Replicating tachyzoites (the fast-growing form of T. gondii) are sequestered in a porous PV that is in equilibrium with the innately low Ca²⁺ (<100 nM) environment of the host’s cytosol. This fluctuates during natural Ca²⁺ signaling events and could reach low μM levels (39) globally or higher levels at membrane contact sites (39). We showed that these transient increases in host cytosolic Ca²⁺ were followed invariably by Ca²⁺ increase in the PV followed by Ca²⁺ entry into the parasite cytosol (11). This previous work demonstrated the functionality of PM-mediated Ca²⁺ entry in intracellular replicating tachyzoites. Host cytosolic Ca²⁺ oscillations were followed by Ca²⁺ entry into the parasite cytosol triggering oscillations where the Ca²⁺ drop would be due (in part) to pumping of Ca²⁺ by the SERCA-Ca²⁺ ATPase into the ER. This Ca²⁺ stockpile in the parasite ER is essential for egress as we showed that a threshold for cytosolic Ca²⁺ had to be attained for successful egress as cytosolic increases that do not reach the threshold resulted in oscillations without egress (11). Interestingly, we found that growing parasites in lower extracellular Ca²⁺ resulted in smaller plaques supporting the relevance of host cytosolic fluctuations for replenishment of tachyzoite intracellular Ca²⁺ stores. We also showed that inhibiting the signaling pathway that leads to enhanced Ca²⁺ entry inhibited T. gondii growth.

We hypothesize that PM Ca²⁺ channels functionally active in intracellular parasites are likely different than the ones that become active in extracellular parasites. The PM Ca²⁺ channels functioning at relatively low levels of extracellular Ca²⁺ (high μM levels) (nifedipine sensitive) supported Ca²⁺ entry in intracellular tachyzoites during normal host calcium fluxes. Ca²⁺ would be captured in the parasite ER to be released through an unknown ER channel. An ER Ca²⁺ leak would protect the ER from Ca²⁺ overload and contribute to the filling of other stores. Upon egress, tachyzoites are in contact with the high Ca²⁺ concentration of the extracellular milieu (>1.5 mM) and PM Ca²⁺ entry in this environment could be through a different type of channel, like the TRP-like channel that our lab characterized (Fig. 8). This PM Ca²⁺ entry in extracellular tachyzoites would maintain parasite motility until they invade a new host cell and begin the replication process anew.

Figure 8.

Proposed model for the regulation of PM Ca²⁺entry of tachyzoites of Toxoplasma gondii. cGMP activates PKG which may activate PI-PLC through phosphorylation. Ca²⁺ is also essential for PIPLC activity which results in the hydrolysis of phosphatidylinositol 4,5-bisphosphate to form inositol trisphosphate or IP₃ which will open an unknown Ca²⁺ release channel at the ER-releasing Ca²⁺ into the cytosol. This Ca²⁺ increase stimulates Ca²⁺ entry through a TRP-like channel. The inhibition by nifedipine and ACA of different channels is shown. CGMP, cyclic GMP; ER, endoplasmic reticulum; IP3, inositol-1,4,5-trisphosphate; PI-PLC, phosphoinositide phospholipase C; PKG, protein kinase G; PM, plasma membrane; TRPPL-2, transient receptor potential channel.

The cyclic nucleotide cGMP is an essential signaling molecule in apicomplexan parasites (32, 40, 41, 42, 43, 44, 45) and the enzyme that synthesizes it, guanylyl cyclase (TgGC), is essential (16). TgGC is a large, 22 transmembrane span protein with a predicted mass of 477 kDa that accumulates at the apical cap region of the PM of T. gondii. Conditional knockdown approaches demonstrated that TgGC is essential for tachyzoite cell-to-cell transmission by regulating egress, migration, and invasion (16, 17, 46). The essentiality of TgGC was also established in an in vivo model of toxoplasmosis, where parasites lacking TgGC were incapable of producing lethal infection in highly sensitive C57Bl/6 mice due to defects in proliferation, dissemination, and persistence (16).

Increase of cGMP levels induces PKG activation which mobilizes intracellular Ca²⁺ presumable through the activation of PI-PLC that hydrolyzes PIP₂ and produces IP₃ and DAG (47). IP₃ would open an unidentified channel in the ER to release Ca²⁺ into the cytosol. Previous PKG-dependent phosphoproteome from Plasmodium falciparum (48) and the most recent phosphoproteome from T. gondii position PI-PLC as a possible phosphorylation substrate of PKG (49). In an additional study with ookinetes of Plasmodium berghei, inhibition of PKG did not result in PI(4,5)P₂ accumulation as could be expected if PKG is activating PI-PLC (50). In this study, it was observed that the phosphoinositide-metabolizing enzymes were less phosphorylated when PKG was inhibited. In addition, lipidomic analysis of total extracts revealed that zaprinast triggered PIP, PIP₂, and PIP₃ depletion (50), indicating PI-PLC activation. More studies are necessary to establish that PI-PLC is a direct target of PKG.

Our present work showed that cGMP is involved in the Ca²⁺ regulation of Ca²⁺ entry as we showed the participation of PKG and PI-PLC in a cascade of molecular events that culminates in the stimulation of Ca²⁺ influx. Increase of cytosolic levels of cGMP induced by the inhibition of PDEs by zaprinast or by the addition of a permeable cGMP resulted in enhanced Ca²⁺ entry that was diminished when PKG or PI-PLC were pharmacologically or genetically ablated. Cytosolic Ca²⁺ increase stimulated by cGMP was significantly reduced upon PI-PLC inhibition, indicating that PKG acts upstream to PI-PLC and could be required for PI-PLC activation. However, PI-PLC requires Ca²⁺ for activity (47), which adds complexity to the proposed signaling cascade: cGMP---PKG---PI-PLC---Ca²⁺ release from intracellular stores; stimulation of Ca²⁺ entry. Is cGMP acting on a Ca²⁺ influx channel (ER or PM) directly and is this initial Ca²⁺ influx contributing to the stimulation of PI-PLC and enhanced Ca²⁺ entry? Addition of cGMP resulted in cytosolic Ca²⁺ increase which could be a combination of its effect in PI-PLC stimulation plus its direct effect in a Ca²⁺ channel. This could explain the residual activity observed in cytosolic Ca²⁺ after adding cGMP to cells previously incubated with a PI-PLC inhibitor or cells in which PI-PLC was downregulated.

In summary, a cytosolic Ca²⁺ increase is important not only for activating Ca²⁺ entry but also is essential for activating PI-PLC as this enzyme has very little activity without Ca²⁺ (47). Most likely, a crosstalk between the cytosolic Ca²⁺ influx mechanisms and release from intracellular stores intersects at the PI-PLC. Interestingly, ∼100% inhibition of Ca²⁺ entry was achieved at high extracellular Ca²⁺ concentrations with nifedipine combined with inhibitors of the PI-PLC. Ca²⁺ entry should be considered as an essential component of the signaling cascades that precede the stimulation of T. gondii biological features.

Experimental procedures

Chemical and reagents

All reagents used were of high analytical grade. Calcium reagents like Fura-2-AM, BAPTA-AM, BAPTA, etc were from AAT Bioquest.

Cell culture and strains

All parasite strains were maintained in human telomerase reverse transcriptase immortalized foreskin fibroblasts (hTERT) (51) grown in Dulbecco’s modified minimal essential media with 1% FBS and streptomycin-penicillin (1:100) (Corning) and regularly tested and treated for Mycoplasma.

Cytosolic calcium measurements with FURA-2

T. gondii tachyzoites were loaded with Fura-2-AM as previously described (22, 52). Briefly, 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 resuspended to a final density of 1 × l0⁹ 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 (2000 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 in ice. Parasites are viable for a few hours under these conditions. 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 dextrose). Fluorescence measurements were done in a Hitachi F-7000 fluorescence spectrofluorometer 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 described previously (53). The rate of Ca²⁺ increase following the addition of Ca²⁺ or inhibitors or agonists was defined as the change in Ca²⁺ concentration during the initial 20 s after the addition of the reagent. The Δ[Ca²⁺]_i was calculated by the difference between the highest peak and basal [Ca²⁺]_i and recovery was defined as the change of Ca²⁺ concentration after the calcium peak was reached for the subsequent 20 s.

For the experiments with BAPTA-AM, T. gondii tachyzoites were incubated in loading buffer with 100 μM BAPTA-AM for 20 min at room temperature in a rotary mixer, followed by washing 2 times with ringer buffer.

The concentrations of agonists used were as follows: TG, 1 μM; zaprinast, 100 μM; 4-[2-(4-fluorophenyl)-5-(1-methylpiperidine-4-yl)-1H-pyrrol-3-yl]pyridine (Compound 1), 1 μM; 1-[6-[{17β-3-methoxyestra-1,3,5(10)-trien-I7-yl}amino]hexyl]-1H-pyrrole-2,5-dione (U-73122) and 1-[6-[{17β-3-methoxyestra-1,3,5(10)-trien-I7-yl}amino]hexyl]- 1H-pyrrole-2,5 pyrrolidine-dione (U73343), 1 μM; nifedipine, 10 μM, ACA, 1 μM; vanadate, 100 μM; BAPTA free acid, 1.8 mM; and the permeable form of guanosine 3′,5′ -cyclic monophosphate 8-bromo- sodium salt (cGMP), 200 μM.

Plaque assays

Plaque assays were performed as previously described (54). Two hundred freshly egressed tachyzoites were used to infect confluent monolayers of hTERT fibroblasts followed by 8 days of growth. Monolayers were fixed and stained with crystal violet and plaque sizes analyzed with FIJI (55) by measuring the area of 15 plaques per biological replicate. At least three biological experiments were done for all conditions tested. Inhibitory compounds were added the first day of the experiment.

Invasion assays

Red-green assays were performed as described (56) with few modifications. Monolayers were infected with 2 × 10⁷ freshly lysed parasites for 20 min on ice and then rapidly transferred to 37 °C for 5 min to stimulate invasion. For experiments including inhibitors, these are added 5 min prior to the invasion step. At the end of the invasion step, parasites were fixed with 3% paraformaldehyde. External (attached) parasites were stained with RabbitαSAG1 (gift from John Boothroyd, Stanford University) at a dilution of 1:1000 followed with 1% Triton X-100 permeabilization. Next, labeling of internal parasites (invaded) was done using a MouseαSAG1 dilution 1:200 (MBS312777, MyBioSource.com). Secondary antibodies were goat-αRabbit and Alexa Flour 546 and Alexa Flour 488 goat-αMouse, both used at 1:1000 dilutions. Images were taken with an Olympus IX-71 inverted fluorescence microscope with a Photometric CoolSnap_HQ CCD camera driven by DeltaVision software (Applied Precision). Data were compiled from three independent experiments and ten fields of view at 1000 total magnifications were counted for each experiment. Fields were randomly selected, and parasites counts were made in a blind fashion. Red or green parasites were counted and percentages of each were calculated by dividing for total parasites.

Egress

GCaMP6f-expressing parasites were used to measure cytosolic calcium fluctuations driving parasite egress as described previously (9). Briefly, glass bottom 35 mm dishes (MatTek Corporation) were used to plate 2 × 10⁶ host cells (hTERT) in Dulbecco’s modified Eagle’s medium supplemented with 10% FBS. After 24 h of growth, host cells were infected with 1 × 10⁶ GCaMP6f-expressing tachyzoites. Twenty-four hours after infection, dishes were washed twice with Ringer buffer plus calcium (155 mM NaCl, 3 mM KCl, 1 mM MgCl2, 3 mM NaH₂PO4, 10 mM Hepes, pH 7.3, glucose 10 mM, and 2 mM CaCl₂ or 100 μM EGTA) and PVs containing two to eight parasites were observed with an inverted fluorescence microscope (Olympus IX-71). The microscope incubation chamber was set at 37 °C, and live cell imaging was recorded with a CoolSnapHQ CCD camera driven by DeltaVision software (Applied Precision). Images were acquired in time-lapse mode with an acquisition rate of 1 to 2 s during 12 min.

Ringer buffer was used as an extracellular buffer throughout all video recording. CaCl₂ was omitted for experiments done in the absence of extracellular Ca²⁺, and the media were supplemented with 100 μM EGTA. ACA (1 μM) was added in Ringer buffer and preincubated for 5 min before imaging. Egress was stimulated with 0.01% saponin or 100 μM zaprinast at 50 sec after the initial video recording.

For analysis of the data, mTrackJ, a plugin freely available for Fiji was used to quantify videos (55) and https://imagescience.org/meijering/software/mtrackj/. We compared the Ca²⁺ response (fluorescence tracings of GCaMP6f parasites) and rate of egress post addition of pharmacological drugs.

Motility

We evaluated the role of Ca²⁺ entry and its impact on motility by exposing parasites to Ca²⁺ in the presence of inhibitors, as described previously (9). Briefly, freshly lysed parasites were collected, purified, and plated on 35-mm bottom glass dish pretreated with 10% FBS. Cells were imaged using a Zeiss LSM 710 confocal microscope set at 37 °C. Then parasites were tracked using a custom-made in-house algorithm (37, 52). The fluorescence of individual parasites (relative cytosolic Ca²⁺) was compared to the motility of parasites after pharmacological or Ca²⁺ stimulation.

EC₅₀ calculation

In vitro growth assays were carried out as described previously using T. gondii tachyzoites expressing red fluorescent protein with the modifications previously described (57). Tachyzoites expressing red fluorescent protein were maintained in human fibroblasts (hTert cells) as described. For drug testing, hTert cells were cultured in 96-well plates 24 h before infection, and 6 × 10⁴ cells were seeded in each well. On the day of the experiment, freshly lysed parasites were purified by passing through a 27-gauge needle, followed by filtration through a 3-μm-pore-size membrane. Each well was seeded with 10⁴ tachyzoites, and fluorescence values were followed for 5 days. The EC₅₀ was calculated on day 5. Plates were read with covered lids, and both excitation (544 nm) and emission (590 nm) were read from the bottom. A Synergy H1 hybrid plate reader from BioTek was used to read fluorescence.

Statistical analysis

All statistical analyses were performed using GraphPad Prism. Unless otherwise noted, all error bars are presented as the SEM and from a minimum of three independent trials. Differences were considered significant if p values were < 0.05. ANOVA was used to compare larger than two sets of data. Comparison between two sets were done using t test.

Data availability

All the data is available in the main text or the supporting information.

Supporting information

This article contains supporting information.

Conflict of interest

The authors declare that they have no conflicts of interest with the contents of this article.

Reviewed by members of the JBC Editorial Board. Edited by Ronald Wek