Using a Genetically Encoded Sensor to Identify Inhibitors of Toxoplasma gondii Ca2+ Signaling

Abstract

The life cycles of apicomplexan parasites progress in accordance with fluxes in cytosolic Ca(2+) Such fluxes are necessary for events like motility and egress from host cells. We used genetically encoded Ca(2+) indicators (GCaMPs) to develop a cell-based phenotypic screen for compounds that modulate Ca(2+) signaling in the model apicomplexan Toxoplasma gondii In doing so, we took advantage of the phosphodiesterase inhibitor zaprinast, which we show acts in part through cGMP-dependent protein kinase (protein kinase G; PKG) to raise levels of cytosolic Ca(2+) We define the pool of Ca(2+) regulated by PKG to be a neutral store distinct from the endoplasmic reticulum. Screening a library of 823 ATP mimetics, we identify both inhibitors and enhancers of Ca(2+) signaling. Two such compounds constitute novel PKG inhibitors and prevent zaprinast from increasing cytosolic Ca(2+) The enhancers identified are capable of releasing intracellular Ca(2+) stores independently of zaprinast or PKG. One of these enhancers blocks parasite egress and invasion and shows strong antiparasitic activity against T. gondii The same compound inhibits invasion of the most lethal malaria parasite, Plasmodium falciparum Inhibition of Ca(2+)-related phenotypes in these two apicomplexan parasites suggests that depletion of intracellular Ca(2+) stores by the enhancer may be an effective antiparasitic strategy. These results establish a powerful new strategy for identifying compounds that modulate the essential parasite signaling pathways regulated by Ca(2+), underscoring the importance of these pathways and the therapeutic potential of their inhibition.

Keywords: calcium; calcium intracellular release; drug screening; parasitology; protein kinase G (PKG); signal transduction.

FIGURE 1.

Zaprinast raises cytosolic Ca²⁺ through the activation of PKG. A, video microscopy of intracellular parasites expressing both GCaMP5 and constitutively secreted DsRed, following the addition of zaprinast at 0 s. B, GCaMP5 fluorescence in the region of the parasitophorous vacuole (green) or DsRed fluorescence in adjacent areas of the infected host cell (red) following the addition of zaprinast. Results shown are mean ± S.E. for four experiments. Kymographs for GCaMP5 fluorescence in all four experiments are shown to indicate times of peak fluorescence (white asterisks) relative to the initiation of egress from the region (white vertical lines). C, intracellular Ca²⁺ concentrations, monitored over time, for wild-type parasites loaded with Fura-2/AM, suspended in buffer containing extracellular (1.8 mm) or basal (100 nm) free Ca²⁺, and stimulated with 100 μm zaprinast at 400 s. D, similar measurements were performed for PKG-M and PKG-T parasites loaded with Fura-2/AM and suspended in basal Ca²⁺. Cpd1 or vehicle was added at 100 s, and zaprinast was added at 400 s, as indicated. E–G, intracellular calcium concentrations from parasites loaded with Fura-2/AM and then treated with zaprinast, ionomycin, GPN, or thapsigargin at 100 s (1) and 400 s (2), as indicated. Traces are representative of three independent experiments. Bar graphs report the change in cytosolic calcium over 20 s following the addition of each drug. Results shown are mean ± S.E. (error bars). Z, zaprinast; I, ionomycin; G, GPN; T, thapsigargin; *, p < 0.05; one-tailed t test.

FIGURE 2.

Compound screen identifies modulators of zaprinast-induced Ca²⁺ signaling. A, GCaMP5-expressing parasites pretreated with Cpd2 or a vehicle control were stimulated with zaprinast or a vehicle control, and fluorescence was measured to determine a Z′ score for a zaprinast-based screen for Ca²⁺ modulators. B, GCaMP5-expressing parasites were pretreated with 823 compounds from the PKIS libraries. Fluorescence was measured after zaprinast stimulation. Results are -fold change from zaprinast alone after background subtraction. Compounds that fluoresced independently are indicated (blue) along with selected enhancers (Enh; green) and inhibitors (Inh; red). The mean ± S.D. (error bars) of two experiments is shown, and dashed lines indicate two S.D. values above and below the mean of all compounds. C, structures of Enh1, Enh2, Inh1, and Inh2 as well as the known PKG inhibitors: Cpd1 and Cpd2. D, cumulative frequencies of screen values for all compounds (black), Inh1 and its analogs (red), or Enh1 and its analogs (green).

FIGURE 3.

PKG inhibitors display distinct modes of inhibition. A, activity of recombinant PKG in the presence of increasing concentrations of Inh1, Inh2, or the PKG inhibitor Cpd2. B, binding of PP derivatives like Inh1 to protein kinases orients the C2 position toward the gatekeeper residue (orange). Left, the PP scaffold's orientation for a disubstituted PP in complex with CDK2. Middle, the structure of Inh1 is highlighted to indicate the PP scaffold (red) and the C2 fluorophenyl group (green). Two other similar compounds from other kinase structures are superimposed over the CDK2 structure in complex with its inhibitor, all showing a similar positioning of the PP scaffold. Right, human p38 MAPK with a trisubstituted monocyclic heterocycle oriented similarly as the PP scaffold and extending a fluorophenyl group in the direction of the gatekeeper. C, oxindole derivatives bind to protein kinases in a manner that orients their C2 and C5 positions away from the gatekeeper. Left, oxindole scaffold of a derivative in complex with NEK2. Right, the oxindole scaffold of Inh2. Two other similar compounds from other kinase structures have been superimposed on the structure of NEK2 with its inhibitor. D, zaprinast-induced egress of parasites carrying Cpd2-sensitive (PKG-T) or resistant (PKG-M) alleles of PKG, following pretreatment with Inh1, Inh2, or Cpd2. Results shown are mean ± S.E. (error bars) for n = 3 independent experiments.

FIGURE 4.

Enh1 and Enh2 mobilize intracellular Ca²⁺ stores. A, GCaMP6-expressing parasites suspended in buffer supplemented with either extracellular (1 mm) or basal (100 nm) Ca²⁺ concentrations, with or without 1% FBS, and treated with 10 μm Enh1 or Enh2 at time 0. B, GCaMP6-expressing parasites treated with 1 μm Cpd1 or vehicle at 100 s and then with 10 μm Enh1 or 100 μm zaprinast at 400 or 750 s, respectively. A gap indicates incubation on ice before the addition of zaprinast, in order to capture the peak of the response. Measurements represent fluorescence after the subtraction of background obtained from samples treated with Cpd1 or vehicle, as indicated. Results shown are mean ± S.E. (error bars) for n = 3 independent experiments. C and D, intensity of R-GECO-expressing HeLa cells treated with Enh1, zaprinast, or A23187 over 10 min (C) or acquired at a faster rate for 1 min (D).

FIGURE 5.

Enhancers of Ca²⁺ mobilization show antiparasitic activity. A, plaque formation in the presence of the indicated concentrations of Enh1 or zaprinast. The drug concentrations indicated did not affect host cell survival. B, dose-dependent effect of Enh1 and zaprinast on parasite viability, assayed by monolayer disruption, 3 days postinfection, at the indicated drug concentrations. Results shown are mean ± S.E. (error bars) for n = 3 independent experiments. C, host cell lysis following 1 h of infection in the presence of varying zaprinast concentrations.

FIGURE 6.

Enh1 elicits asynchronous cytosolic Ca²⁺ fluxes and blocks zaprinast-induced egress. A, video microscopy of GCaMP6f-expressing parasites treated with Enh1 or zaprinast. Time after the addition of the compound is indicated. Different times were used to capture the fast and slow responses of zaprinast and Enh1, respectively. B, kymographs illustrate average fluorescence intensities of individual parasites, per row, during the course of the treatment indicated. Black indicates that parasites egressed from vacuoles. C, change in fluorescence of the parasites illustrated in B over the 40 s following the addition of zaprinast. Measurements from each biological replicate are colored separately. Mean change for each group is indicated with a horizontal line. ****, p < 0.0001, two-tailed t test.

FIGURE 7.

Enh1 blocks egress Ca²⁺-related phenotypes in T. gondii and P. falciparum. A, egress of intracellular parasites treated with zaprinast, Enh1, or a vehicle control. The number of intact vacuoles was monitored by live microscopy over 30 min. Representative images before and after treatment are shown. B, dose-dependent inhibition of zaprinast-induced egress following pretreatment with Enh1 or vehicle. C, Enh1 inhibition of egress induced by either zaprinast or A23187. Results shown are mean ± S.E. (error bars) for n = 3 independent experiments. ***, p < 0.001; **, p < 0.01. D–E, schizonts were released from Cpd2 arrest immediately preceding the addition of Enh1. D, after 1–2 h, the remaining schizonts (mean ± S.D. for three technical replicates) were counted and normalized to their initial abundance (3.6 and 5.4% in each experiment, respectively). E, Enh1 blocks invasion of erythrocytes by P. falciparum, measured 1–2 h following release from Cpd2, as assessed from the ring stage parasitemia. Mean invasion ± S.E. is expressed as a percentage of invasion without drug (9.5 and 8.1% in each experiment, respectively). Background was assessed using heparin as a specific blocker of invasion and was comparable with the signal observed with saturating concentrations of Enh1. F, dose-dependent inhibition of T. gondii invasion following 10-min pretreatment of parasites before invasion. Results shown are mean ± S.E. for n = 3 independent experiments.