Register for Mechanobiology 2016!

Toxoplasma gondii triggers Gi-dependent PI 3-kinase signaling required for inhibition of host cell apoptosis
Leesun Kim, Eric Y. Denkers


Infection with the intracellular parasite Toxoplasma gondii renders cells resistant to multiple pro-apoptotic signals, but underlying mechanisms have not been delineated. The phosphoinositide 3-kinase (PI 3-kinase) pathway and the immediate downstream effector protein kinase B (PKB/Akt) play important roles in cell survival and apoptosis inhibition. Here, we show that Toxoplasma infection of mouse macrophages activates PKB/Akt in vivo and in vitro. In a mixed population of infected and non-infected macrophages, activation is only observed in parasite-infected cells. The PI 3-kinase inhibitors wortmannin and LY294002 block parasite-induced PKB phosphorylation. PKB activation occurs independently of Toll-like receptor adaptor protein MyD88 but uncoupling of Gi-protein-mediated signaling with pertussis toxin prevents PKB phosphorylation. Moreover, in the presence of PI 3-kinase inhibitors or pertussis toxin, not only PKB activation but also ERK1/2 activation during T. gondii infection is defective. Most importantly, the parasite's ability to induce macrophage resistance to pro-apoptotic signaling is prevented by incubation with PI 3-kinase inhibitors. This study demonstrates that T. gondii exploits host Gi-protein-dependent PI 3-kinase signaling to prevent induction of apoptosis in infected macrophages.


The opportunistic protozoan pathogen Toxoplasma gondii is a widespread parasite known for its ability to infect a wide range of host species and cell types (Dubey, 1998; Hill and Dubey, 2002; Luft and Remington, 1992). During infection, highly invasive tachyzoites enter cells using an actin-based motility system coupled with coordinated discharge of apical organelles (Carruthers, 2002; Carruthers and Sibley, 1997; Dobrowolski and Sibley, 1996; Huynh et al., 2003). Inside the host cell, parasites reside within a specialized parasitophorous vacuole that resists endosomal acidification and lysosomal fusion where they display rapid intracellular replication (Sibley, 2003). The ability to resist antimicrobial effector function from within the infected cell and to establish long-term infection despite induction of strong cell-mediated immunity argues that Toxoplasma is a highly effective manipulator of host responses (Denkers et al., 2004).

Several independent lines of evidence indicate that Toxoplasma-infected cells are resistant to inducers of apoptosis (Luder and Gross, 2005; Nash et al., 1998). This phenomenon is associated with upregulation of a subset of anti-apoptosis genes, reduced activation of caspase molecules, and downregulation of poly ADP-ribose polymerase expression (Goebel et al., 2001). Mechanisms involved in parasite-induced suppression of apoptosis are little known. NFκB-dependent activation of anti-apoptosis genes may account for some of these effects (Molestina et al., 2003; Payne et al., 2003), but these results are difficult to reconcile with others that fail to find evidence for NFκB activation during infection (Butcher et al., 2001; Shapira et al., 2002). Therefore, elucidation of parasite-triggered host signaling pathways that mediate resistance to apoptosis remains a high-priority area.

Protein kinase B (PKB)/Akt plays an important role in cell survival and regulation of apoptosis (Scheid and Woodget, 2001; Yang et al., 2004). Activation of PKB is dependent upon signaling by heterodimeric phosphatidylinositol 3-kinase (PI 3-kinase) (Deane and Fruman, 2004; Hawes et al., 1996; Hirsch et al., 2000). The latter uses the lipid mediator phosphatidylinositol (3,4)-bisphosphate [PtdIns(3,4)P2] and phosphatidylinositol (3,4,5)-trisphosphate [PtdIns(3,4,5,)P3] to recruit PKB to the plasma membrane, and further activation occurs through phosphoinositide-dependent kinases (PDKs) (Alessi et al., 1997; Stephens et al., 1998). The PI 3-kinase molecule also contains a protein-kinase domain that stimulates mitogen-activated protein kinase (MAPK) (Bondeva et al., 1998). The anti-apoptotic function of PKB occurs in part through phosphorylative inactivation of the proapototic proteins Bad and caspase-9 (Blume-Jensen et al., 1998; Cardone et al., 1998; Datta et al., 1997; del Peso et al., 1997). In addition, PKB inactivates the Forkhead family transcription factor FKHR1 that regulates apoptosis-inducing genes (Brunet et al., 1999; Kop et al., 1999). PKB has also been implicated in positive regulation of the NFκB signaling pathway that leads to induction of several survival-promoting genes (Arbibe et al., 2000; Kane et al., 1999; Madrid et al., 2001; Patra et al., 2004). Insofar as PKB displays nuclear translocation following activation, the kinase might also be involved in induction of anti-apoptotic genes from within the nucleus (Andjelkovic et al., 1997).

Here, we show that Toxoplasma triggers PKB activation in infected host cells. Activation in macrophages depends upon Gi-protein-coupled signaling, and does not involve TLR adaptor molecule MyD88. In agreement with others, we find that parasite-infected cells are resistant to inducers of apoptosis. However, when PKB activation is inhibited by chemical blockade of the PI 3-kinase pathway, T. gondii no longer prevents induction of apoptosis in infected cells. These results define a major host-biochemical pathway that is exploited by Toxoplasma to prevent induction of apoptosis during infection.


T. gondii activates PKB/Akt upon infection in vivo and in vitro

The serine/threonine protein kinase PKB/Akt mediates many of the downstream events controlled by PI 3-kinase and its activity is regulated by Thr308 and Ser473 phosphorylation. We asked whether in vivo T. gondii infection induces splenic macrophage (MØ) PKB phosphorylation, by using flow cytometry employing phosphorylated-Ser473 PKB antibody (Ab). Tachyzoite-harboring splenic F4/80-positive cells were identified through use of transgenic type-1 strain RH tachyzoites that express yellow fluorescent protein (YFP) (Fig. 1A). When we compared infected and non-infected F4/80-positive macrophages, activation of PKB was only apparent in the infected population (Fig. 1B). We were able to discern peaks of low and high phosphorylated-PKB staining in the infected populations. Back-gating on the FACS plots revealed that these subpopulations corresponded in each case to levels of infection (data not shown).

Fig. 1.

Toxoplasma induces MØ PKB/Akt activation during in vivo and in vitro infection. (A,B) mice were i.p. infected with 103 YFP-expressing RH tachyzoites. Eight days post-infection, spleen cells were isolated and subjected to staining with Ab to MØ marker F4/80 and phosphorylated-PKB (Ser473). (A) YFP expression in F4/80+ cells from infected animals. Inset shows lack of YFP signal in F4/80+ cells from non-infected mice. (B) Expression of phosphorylated-PKB (Ser473) in infected (YFP+, thick line) and non-infected (YFP-, thin line) F4/80+ cells from infected mice described in A. Dotted line, isotype control. (C) Bone-marrow-derived MØs were infected with RH strain tachyzoites. Then, 15 minutes later, cells were collected for flow cytometric analysis. Cells were permeabilized and subjected to Ab staining for the tachyzoite surface protein p30 (SAG-1). The (a) non-infected and (b) infected populations (percentages are indicated) were examined for PKB expression (thick lines) relative to unstimulated cells (dashed lines). Thin lines in a and b show staining of secondary antibody alone on the total MØ population. Infected and non-infected cells gave the same level of background staining (data not shown). These experiments were repeated twice with the same result.

We then prepared bone marrow-derived MØs to determine whether the effects of Toxoplasma on PKB activation were recapitulated during infection in vitro. In this case, cells infected with RH tachyzoites were identified based upon staining with Ab against p30 (SAG-1), a major surface protein of the parasite (Fig. 1C). Whereas non-infected cells showed no increase in PKB phosphorylation (Fig. 1C, panel a), MØs infected with T. gondii displayed increased levels of phospho-PKB (Fig. 1C, panel b). This figure also shows that non-infected macrophages display constitutive PKB phosphorylation that is increased further by infection. Constitutive PKB phosphorylation is also apparent by western blot analysis (see below).

Distinct intracellular signaling cascades lead to activation PKB/Akt and ERK1/2 in T. gondii-infected vs LPS-stimulated cells

Next, we examined activation of PKB and MAPKs in the presence of wortmannin, a potent and specific PI 3-kinase inhibitor (Wymann et al., 1996). Others have reported a partially inhibitory effect of wortmannin on parasite invasion of neutrophils (MacLaren et al., 2004), but we saw no blocking effects at the concentration used in these experiments. There was also no effect of wortmannin on parasite replication in MØs (data not shown).

As shown in Fig. 2, both Toxoplasma and LPS induced rapid phosphorylation of PKB, although activation appeared to be more long-lasting in the case of endotoxin. Parasites and LPS also stimulated phosphorylation of extracellular signal-regulated kinases 1/2 (ERK1/2), the ERK1/2 kinases MEK1/2 (mitogen-activated protein kinase/extracellular signal-regulated kinase kinase1/2), and p38 MAPK. In the presence of wortmannin, Toxoplasma-induced phosphorylation of PKB, ERK1/2 and MEK1/2 was totally eliminated, but activation of p38 MAPK was unaffected. For LPS, PKB activation was abrogated by wortmannin inhibition of PI 3-kinase, but phosphorylation of ERK1/2, MEK1/2 and p38 MAPK was unaffected. Essentially identical results were obtained with LY294002, another PI 3-kinase inhibitor (data not shown). These results show that T. gondii activates PKB and ERK1/2, but not p38 MAPK, in dependence upon PI 3-kinase signaling. Whereas LPS activation of PKB also requires PI 3-kinase signaling, phosphorylation of MAPK family members ERK1/2 and p38 does not depend upon PI 3-kinase.

One of the major pathways for activation of PI 3-kinase is through Gi protein-coupled transmembrane receptors (GiPCR). Therefore, we employed pertussis toxin that uncouples GiPCR to determine whether Gi proteins play a role in PI 3-kinase, ERK1/2 and p38 MAPK activation. In these experiments, pertussis toxin did not affect tachyzoite invasion or replication (data not shown). Similar to blocking of PI 3-kinase, pertussis-toxin-dismantling of GiPCR signaling prevented PKB, ERK1/2 and MEK1/2 activation, but left p38 MAPK activation intact, in infected cells (Fig. 3). During endotoxin stimulation, pertussis-toxin-induced GiPCR inactivation prevented PKB activation but MAPK-signaling modules were unaffected.

Fig. 2.

Requirement for PI 3-kinase signaling in PKB and MAPK activation differs for T. gondii infection and LPS stimulation. MØs were preincubated 2 hours with PI 3-kinase inhibitor wortmannin (WM, 50 ng/ml) or solvent (DMSO) alone (0), then infected with T. gondii (6:1 ratio of parasites to cells; >90% infected) or stimulated with LPS (100 ng/ml). At the indicated time points, cells were collected for western blot analysis for total and phosphorylated forms of each signaling intermediate. This experiment was performed three times with the same result.

Chemokine receptor CCR5 is a seven-transmembrane GiPCR that, according to several studies, is implicated in dendritic cell IL-12 production during T. gondii infection (Aliberti et al., 2000; Aliberti et al., 2003). These reports raised the possibility that the effects reported here are mediated by parasite encounter with CCR5. Nevertheless, MØs generated from CCR5-/- mice responded normally in terms of PKB and MAPK activation during parasite infection (Fig. 4). The minor variation in phosphorylation kinetics is owing to experiment-to-experiment variation. Regardless, the finding shown in Fig. 4 rules out CCR5 involvement in GiPCR activation of PKB and ERK1/2. Similarly, we also examined responses of CCR2-/- MØs and found that they underwent parasite-induced phosphorylation of PKB and ERK1/2, arguing against a role for this chemokine receptor in the responses seen (data not shown).

MyD88, a central adaptor in Toll-like receptor (TLR) signaling, is important in recognition of intact Toxoplasma and Toxoplasma-derived molecules (Scanga et al., 2002; Yarovinsky et al., 2005). Accordingly, we employed MØs deficient in this adaptor protein to determine whether MyD88-dependent signaling cascades interacted with GiPCR signaling to stimulate PKB or ERK1/2 activation. As shown in Fig. 5, absence of MyD88 had no effect on parasite-induced phosphorylation of ERK1/2, MEK1/2 or PKB. By contrast, and as previously recognized (Kawai et al., 1999), absence of MyD88 delayed the kinetics of MEK1/2 and ERK1/2 phosphorylation in response to LPS. Here, we also show a delay in LPS-induced PKB activation in the absence of MyD88 (Fig. 5).

Fig. 3.

Signaling mediated by Gi protein-coupled receptors is required for Toxoplasma-induced PKB and ERK1/2, but not p38 MAPK, activation. MØs were set up as described in Fig. 2, after 2 hours preincubation with pertussis toxin (PTx, 50 ng/ml) or solvent (H2O) alone (0) followed by infection with T. gondii (6:1 ratio of parasites to cells) or stimulation with LPS (100 ng/ml). At the indicated time points, cells were collected for western blot analysis measuring total and phosphorylated form of each signaling intermediate. This experiment was repeated three times with essentially identical results.

In sum (Fig. 6), Toxoplasma infection stimulates pertussis-toxin-sensitive GiPCR signaling leading to activation of PI 3-kinase that can also be blocked by wortmannin and LY294002. In turn, PI 3-kinase stimulates activation of PKB and MEK1/2, the latter of which mediates ERK1/2 phosphorylation. Our data do not show whether PKB is an upstream activator of MEK1/2-ERK1/2 or vice-versa. However, these possibilities are unlikely because it is known from other studies that the PI 3-kinase signaling pathway bifurcates into cascades that result the activation of PKB and MAPK (Bondeva et al., 1998).

Fig. 4.

MAPK and PKB activation induced by T. gondii and LPS are CCR5 independent. MØs from wild-type (WT) and CCR5 knockout (CCR5 KO) mice were incubated with parasites (6:1 ratio of tachyzoites to cells) or LPS (100 ng/ml), and at the indicated time points, samples were subjected to SDS-PAGE followed by western blotting with antibodies against total and phosphorylated forms of PKB, ERK1/2 and p38 MAPK. The experiment was repeated twice with the same result.

Fig. 5.

T. gondii activates ERK1/2 and PKB independently of MyD88, unlike LPS. MØs from wild-type (WT) and MyD88 knockout (KO) mice were incubated with parasites (6:1 ratio of tachyzoites to cells, >90% infected) or LPS (100 ng/ml), and at the indicated time points, samples were collected for western blot analysis measuring total and phosphorylated forms of each signaling intermediate. This experiment was repeated 4 times with the same result.

Fig. 6.

Pathway of T. gondii-induced PKB and ERK1/2 activation in MØs. T. gondii infection activated PI 3-kinase through Gi-protein-coupled receptor signaling. In turn, PI 3-kinase stimulates PKB activation and MEK1/2 phosphorylation leading to ERK1/2 activation. Pertussis toxin, which uncouples Gi-protein signaling, and wortmannin and LY294002, which inhibit PI 3-kinase activation, block PKB/Akt and ERK1/2 activation. PTx, pertussis toxin; WM, wortmannin; Ly, LY294002.

T. gondii exploits the PI 3-kinase signaling pathway to block apoptosis

Several independent lines of evidence indicate that T. gondii infection renders cells resistant to proapoptotic stimuli (Goebel et al., 2001; Luder and Gross, 2005; Nash et al., 1998; Payne et al., 2003). Here, we show that parasite infection renders MØs resistant to staurosporine-induced apoptosis, as determined by TUNEL staining of cells (Fig. 7A-E). Lack of extensive proliferation in these cells at 18 hours may reflect the fact that the host cells were subjected to serum starvation. Fig. 7F shows quantitation of apoptotic cells, revealing a statistically significant inhibitory effect of Toxoplasma on infected MØs. Furthermore, whereas staurosporine induced the cleavage of PARP, a signature event in apoptosis, cells infected with T. gondii failed to undergo cleavage of this DNA repair enzyme (Fig. 8A). Because these effects only occurred in infected cells (Fig. 7) and because extensively washed tachyzoites maintained their ability to block apoptosis (data not shown), this suggests that parasites themselves, rather than carry-over from fibroblast cultures in which the parasites are maintained, are responsible for these effects.

We employed wortmannin and LY294002 to determine the role of PI 3-kinase signaling in parasite-mediated resistance to apoptosis. Blocking PI 3-kinase activity with each of these inhibitors terminated the ability of T. gondii to prevent staurosporine-induced programmed cell death (Fig. 8A). These results were confirmed in an independent colorimetric assay measuring activity of caspase-3, a central regulator of apoptosis (Fig. 8B). Whereas staurosporine induced MØ caspase-3 activation, Toxoplasma infection blocked this effect. As with PARP cleavage, wortmannin inhibition of PI 3-kinase eliminated the ability of the parasite to prevent staurosporine induced apoptosis. The effects of wortmannin were not due to an inhibition of parasite invasion, because infection rate was not affected by PI 3-kinase inhibition (data not shown). Similar effects on caspase-3 activity were seen using LY294002, however, in this case the inhibitor alone increased background levels of apoptosis (data not shown). The combined results demonstrate that Toxoplasma exploits GiPCR signaling to activate PI 3-kinase, rendering infected cells resistant to programmed cell death.

Fig. 7.

T. gondii infection renders cells resistant to apoptosis. (A-E') MØs were serum-starved for 18 hours, infected with Toxoplasma; 2 hours later staurosporine (1 μM) was added to cell cultures. After 18 hours, cells were subjected to TUNEL-FITC (green) and anti-p30 (SAG-1) Alexa Fluor 594 (red) staining, nuclei were stained with DAPI (blue). (A) Non-infected cell showing DAPI-stained nucleus. (A') Same cell as in A, showing TUNEL- and nuclei-staining (green and blue, respectively, appearing as turquoise). (B-E) Cells from the same population. Examples of infected cells, showing parasite-staining (red) and lack of TUNEL-stained nuclei (therefore remaining blue). Arrows point to tachyzoites in an intracellular location. (F) Quantitation of TUNEL-staining-positive cells. Med, cultured in medium alone; STS, cells cultured with staurosporine; I/STS, infected cells subjected to STS treatment; NI/STS, non-infected cells cultured with STS. *P<0.05; 100 cells were counted in each sample. Bars, 5 μm. This experiment was performed three times with the same result.


Many intracellular protozoans disrupt host cell signaling pathways that lead to apoptosis (James and Green, 2004; Luder and Gross, 2005). As such, blocking programmed cell death is emerging as a unifying survival strategy of this class of microbial pathogens. Here, we show that Toxoplasma exploits heterotrimeric Gi-protein-mediated signaling to activate PI 3-kinase, leading to phosphorylation of PKB/Akt and ERK1/2 MAPK, and inhibition of apoptosis. Stimulation of PKB occurred during infection of F4/80-positive splenic macrophages during acute stages of in vivo Toxoplasma infection and also occurred during in vitro infection of bone marrow macrophages.

Fig. 8.

PI 3-kinase inhibition reverses the anti-apoptotic effect of Toxoplasma infection. (A), MØs were serum-starved for 18 hours after which wortmannin (WM), LY294002 (Ly) or solvent alone (DMSO) were added. After 2 hours, cells were infected (6:1 ratio of parasites to cells, >90% infection) or left non-infected in medium (Med). Then, after another 2 hours, staurosporine (STS) was added and 4 hours later cell lysates were prepared for analysis of apoptosis. PARP cleavage, which is indicative of apoptosis, was measured in PARP-specific western blot assays. Apoptotic PARP cleavage generates 89 kDa and 24 kDa fragments (not shown in the blot) from the intact 214 kDa molecule. (B) Cultures were initiated in the presence or absence of WM as described in A. Caspase activity was measured in an ELISA-based assay. Med, cells in medium alone; M+S, cells cultured with staurosporine; Tg, cells infected with T. gondii. Tg+S, cells infected and then triggered with staurosporine. Number above the `Tg+S' group (white bar) indicates percent inhibition relative to the `M+S' group. *P<0.05; ns, not statistically significant. These experiments were performed three times with the same result.

Another apicomplexan with well-known anti-apoptotic effects is Theileria parvum (Heussler et al., 2001). Nevertheless, the effect of this parasite occurs independently of PKB, unlike the results we report here for Toxoplasma. Furthermore, anti-apoptotic effects of Theileria depend upon NFκB signaling (Machado et al., 2000) but we, and others, do not detect activation of this signaling pathway in infected MØs (Butcher et al., 2001; Shapira et al., 2002). In addition, infected lymphocytes undergo apoptosis dependent upon caspase-3 and caspase-9 when Theileria macroschizonts are eliminated from cells by drug treatment (Guergnon et al., 2003). We conclude that Toxoplasma and Theileria employ distinct mechanisms to achieve the same - namely, inhibition of apoptosis,

Although both Toxoplasma as well as LPS activate PKB and ERK1/2 MAPK, the pathways involved are clearly distinct. As reported by others (Fan et al., 2004; Ojaniema et al., 2003), LPS-induced activation of PKB/Akt involves triggering through Gi-proteins and PI 3-kinase. This was determined by blocking-studies with pertussis toxin to inhibit Gi signaling, and wortmannin as well as LY294002 to inactivate PI 3-kinase. By contrast, use of the same inhibitors demonstrated that TLR4 ligand-induced activation of p38 MAPK and ERK1/2 did not require Gi-PI-3-kinase-PKB-mediated signaling. Nevertheless, LPS-mediated signaling to PKB activation involves cross talk between TLR and GiPCR pathways. This is mediated through Gi-protein-mediated association with CD14 that, together with TLR4 and MD2, forms the LPS-receptor complex (Fan et al., 2004; Solomon et al., 1998; Vivarelli et al., 2004).

Using macrophages deficient in MyD88 expression, we found that LPS-induced activation of PKB and MAPK continued to occur in the absence of this TLR adaptor molecule, although with delayed kinetics. Activation of MAPK and NFκB in the absence of MyD88 is known to occur dependent upon Toll-IL1-receptor (TIR)-domain-containing adaptor-inducing IFN-β (TRIF) (Yamamoto et al., 2003; Yamamoto et al., 2004), and it seems likely that MyD88-independent LPS-induced PKB/Akt phosphorylation is also mediated by this adaptor of the TLR signaling pathway.

The effects of Toxoplasma on PKB/Akt and MAPK activation pathways differed. A Gi-PI-3-kinase pathway mediated PKB and ERK1/2 activation, and neither the degree nor the kinetics of phosphorylation were affected by absence of MyD88. In contrast to PKB and ERK1/2, p38 MAPK activation continued to occur when the Gi-PI 3-kinase pathway was dismantled with chemical inhibitors. In this regard, it is interesting that Porphyromonas gingivalis LPS activates PI 3-kinase and PKB/Akt through TLR2, leading to ERK1/2 activation. Nevertheless, this pathway is not required in p38 MAPK or Jun N-terminal kinase 1/2 (JNK1/2) activation triggered by P. gingivalis LPS (Martin et al., 2003). Although similar to the T. gondii-stimulated PI 3-kinase-PKB/Akt-ERK1/2 pathway, lack of an effect of MyD88 and sensitivity to pertussis toxin argue for involvement of Gi-coupled receptors rather than TLR in the PKB/Akt response to T. gondii.

Activation of ERK1/2, like PKB, has been implicated in disrupting pro-apoptotic signaling. This can involve signaling through PI 3-kinase and, at least in some cases, acts by phosphorylation-dependent inactivation of caspase-9 (Allan et al., 2003; von Gise et al., 2001). Although Toxoplasma induced ERK1/2 activation in dependence upon PI 3-kinase, blocking ERK1/2 activity with chemical inhibitors consistently had no influence on the anti-apoptotic effects of the parasite (data not shown). Therefore, PKB is probably the major effector molecule activated by PI 3-kinase that mediates resistance to apoptosis in infected macrophages.

The finding that pertussis toxin prevents Toxoplasma-triggered PKB activation implicates Gi-protein-coupled receptors in recognition of the parasite during infection. This large class of receptors has in common a seven-transmembrane-spanning region and includes receptors required for chemokine recognition. Although CCR5 has been implicated in dendritic cell recognition of a Toxoplasma cyclophilin (Aliberti et al., 2003), we found normal PKB activation by using macrophages from CCR5-/- mice, ruling out a role for this chemokine receptor in the present study. It is nevertheless worthwhile to notice other studies suggesting that CCR5 is permitted through the moving junction that forms when Toxoplasma enters the host cell. After invasion, CCR5 assumes a position co-localizing with the parasitophorous vacuole membrane (Charron and Sibley, 2002). Whereas our results argue that CCR5 is not involved in PKB/Akt activation, it is possible that other host-G-protein-linked seven-transmembrane proteins locate to the nascent parasitophorous vacuole membrane where they might be triggered by parasite-derived molecules.

Whereas the results of this and other studies clearly show that infection with T. gondii induces resistance to apoptosis, the parasite paradoxically displays pro-apoptotic effects in vivo. This is related to the high levels of type-1 cytokines that are produced and that are required to survive Toxoplasma infection (Denkers, 2003; Denkers and Gazzinelli, 1998). For example, Peyer's patch CD4+ T lymphocytes undergo IFN-γ-mediated, Fas-dependent programmed cell death following oral infection (Liesenfeld et al., 1997). During infection with the virulent RH parasite strain that was used in the present studies, high-level proinflammatory cytokine-mediated programmed cell death is induced in parasite-positive lymphoid organs, yet apoptotic areas are distinct from foci of infection (Gavrilescu and Denkers, 2001; Mordue et al., 2001). It seems probable that Toxoplasma renders host cells resistant to pro-apoptotic effects of type-1 cytokines as a strategy to maintain host cell survival and parasite persistence in the presence of an overwhelming pro-inflammatory response.

Materials and Methods

Mice and parasites

Strain C57BL/6 female mice (aged 6-8 weeks) were obtained from Taconic (Germantown, NY). Mice were housed in the animal facility of the College of Veterinary Medicine at Cornell University. RH strain tachyzoites of T. gondii were maintained on human foreskin fibroblast monolayers by biweekly passage in Dulbecco's modified Eagle's medium (DMEM) (Life Technologies, Gaithersburg, MD) supplemented with 1% heat-inactivated fetal calf serum (FCS) (HyClone, Logan, UT), 100 U/ml penicillin (Life Technologies), and 0.1 mg/ml streptomycin (Life Technologies). Transgenic RH parasites expressing yellow fluorescent protein (YFP) were kindly provided by D. Roos (University of Pennsylvania, Philadelphia, PA). YFP-RH parasites were supplemented with 1 mM pyrimethamine (Sigma-Aldrich). Parasite cultures were tested every 4-6 weeks for Mycoplasma contamination using a commercial PCR-ELISA-based kit (Roche Applied Science, Mannheim, Germany).

Reagents and antibodies

Rabbit antibodies recognizing full-length forms of poly-ADP-ribose polymerase (PARP) Ab or phosphorylated forms of p38, ERK1/2, MEK1/2 and PKB (Ser473) were purchased from Cell Signaling Technology (Beverly, MA), rabbit anti-p38 Ab was obtained from Santa Cruz Biotechnology (Santa Cruz, CA), and mouse FITC-conjugated-anti-p30 (SAG-1) Ab was obtained from Argene (North Massapequa, NY). Pertussis toxin, staurosporine and PI 3-kinase inhibitors LY294002 and wortmannin were from Calbiochem (La Jolla, CA). Ultra-pure LPS from Salmonella minnesota R595 was purchased from List Biological Laboratories (Campbell, CA).

Cell culture

Bone-marrow-derived macrophages (MØs) from C57BL/6 (Taconic) or MyD88-/- mice on a partially backcrossed 129/Ola × C57BL/6 background (originally engineered by S. Akira, kindly provided by A. Sher, NIAID) were prepared as described previously (Kim et al., 2004). Some experiments employed MØs from CCR5+/+ (B6129F2/J) and CCR5-/- (B6129P2-Cmkbr5tm1Kuz) mice that were purchased from The Jackson Laboratory (Bar Harbor, ME). The cells generated were >90% F4/80 positive, as determined by flow cytometric analysis. For in vitro infection, RH strain tachyzoites or LPS were added to MØs in plates (Costar, Corning, NY), which were then briefly centrifuged (3 minutes, 200 g) to synchronize contact between cells and parasites. In some experiments, the PI 3-kinase inhibitors wortmannin and LY294002, or pertussis toxin were incubated for 2 hours prior to infection or LPS stimulation. To induce apoptosis, cells were serum-starved for 18 hours, then staurosporine (1 μM, unless indicated otherwise) was added 2 hours after parasite infection. Samples were subsequently collected for biochemical assay.

Flow cytometry

For intracellular phospho-PKB and RH tachyzoite staining, cells were briefly washed with PBS, fixed with 1% paraformaldehyde for 10 minutes at 37°C and permeabilized on ice with 90% ice-cold methanol for 30 minutes. Cells were blocked with 0.5% BSA (Sigma) for 10 minutes at room temperature followed by incubation with rabbit anti-phospho-PKB Ab for 30 minutes at room temperature. After washing, cells were incubated with PE-conjugated donkey anti-rabbit secondary Ab (Jackson ImmunoResearch Laboratories, West Grove, PA) and FITC-conjugated anti-p30 Ab for 30 minutes at room temperature. Cells were washed and subjected to flow cytometry using a FACSCalibur flow cytometer (Becton Dickenson Immunocytometry Systems, San Jose, CA) and Flow Jo software (Tri Star, San Carlos, CA).

In vivo PKB activation

YFP-RH (103) was inoculated intraperitoneally (i.p.) into mice. Eight days after infection, spleens were homogenized, passed through a 70 μm nylon cell strainer (BD Biosciences, San Diego, CA) and red blood cells were removed with lysis buffer (Sigma). Cells were then washed and resuspended in PBS. Fc receptors were blocked with PBS containing 1% BSA (Sigma), 5 μg/ml rat anti-mouse CD16/CD32 (BD Biosciences) and 10% normal mouse serum (Jackson ImmunoResearch Laboratories). Then cells were stained with allophycocyanin-conjugated anti-F4/80 Ab (Caltag Laboratories, Burlingame, CA) for 30 minutes on ice. Intracellular PKB staining was carried out as described above.

Western blotting

Immunoblot analysis was performed as previously described (Kim et al., 2004).

In situ cell death detection by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL)

Following staurosporine treatment to induce apoptosis, cells were fixed in 2% paraformaldehyde at room temperature for 1 hour and permeabilized with 0.1% Triton X-100 (Sigma) and 0.1% sodium citrate (Sigma) for 5 minutes on ice. Cells were then subjected to TUNEL using FITC-conjugated dUTP (Roche Applied Science). Toxoplasma staining was accomplished with a rabbit anti-p30 Ab (BioGenex, San Ramon, CA) followed by Alexa Fluor 594-conjugated goat anti-rabbit Ab (Molecular Probes, Eugene, OR). Coverslips were mounted with ProLong Antifade reagent containing 4′, 6′-diamidino-2-phenylindole (DAPI) for DNA staining (Molecular Probes). Images were collected with a fluorescence microscope (Olympus, Melville, NY).

Caspase 3 colorimetric assay

Caspase-3 activity was detected by the chromophore p-nitroanilide (pNA) after cleavage of the synthetic peptide substrate DEVA-pNA by active caspase-3 using a commercial kit (BioVision, Mountain View, CA). Following staurosporine treatment for 4 hours, cells were lysed and centrifuged to collect cytosolic extracts. Supernatants were incubated with dithiothreitol and DEVA-p-NA in the reaction buffer at 37°C for 3 hours. Caspase-3 activity was measured as optical density at 405 nm.


The statistical significance of the data was determined by unpaired Student's t-test. Values of P<0.05 were considered significant.


We thank B. Butcher for insightful discussion and L. Del Rio for FACS assistance. We gratefully acknowledge S. Akira for permission to use MyD88-/- cells that were generously provided by A. Sher. This work was supported by PHS grant AI50617.

  • Accepted February 14, 2006.


View Abstract