A Patatin-Like Protein Protects Toxoplasma gondii from Degradation in a Nitric Oxide-Dependent Manner

Abstract

Toxoplasma gondii is an obligate intracellular parasite that uses immune cells to disseminate throughout its host. T. gondii can persist and even slowly replicate in activated host macrophages by reducing the antimicrobial effects of molecules such as nitric oxide (NO). A T. gondii patatin-like protein called TgPL1 was previously shown to be important for survival in activated macrophages. Here we show that a T. gondii mutant with a deletion of the TgPL1 gene (ΔTgPL1) is degraded in activated macrophages. This degradation phenotype is abolished by the removal of NO by the use of an inducible NO synthase (iNOS) inhibitor or iNOS-deficient macrophages. The exogenous addition of NO to macrophages results in reduced parasite growth but not the degradation of ΔTgPL1 parasites. These results suggest that NO is necessary but not sufficient for the degradation of ΔTgPL1 parasites in activated macrophages. While some patatin-like proteins have phospholipase A₂ (PLA₂) activity, recombinant TgPL1 purified from Escherichia coli does not have phospholipase activity. This result was not surprising, as TgPL1 contains a G-to-S change at the predicted catalytic serine residue. An epitope-tagged version of TgPL1 partially colocalized with a dense granule protein in the parasitophorous vacuole space. These results may suggest that TgPL1 moves to the parasitophorous vacuole to protect parasites from nitric oxide by an undetermined mechanism.

INTRODUCTION

Toxoplasma gondii is a protozoan parasite that is able to infect nucleated cells in warm-blooded hosts, including humans (12). The parasite uses an active invasion process to enter host cells and create a parasitophorous vacuole (PV), which is necessary for parasite survival and replication (3). The PV is comprised of host cell lipids but excludes host cell proteins, making it distinct from the endocytic network. Entry by active invasion protects T. gondii from lysosomal killing; however, if the parasite enters through the endocytic network, e.g., by phagocytosis, it is killed by standard host mechanisms (21). Through active invasion, T. gondii can use host immune cells to disseminate to peripheral tissues, including the brain.

During the initial phases of Toxoplasma infection, macrophages are one of the first immune cell types to arrive at the infection site. These cells have an array of antimicrobial defenses, such as the production of reactive oxygen and nitrogen species as well as p47 GTPases, all of which help control parasite growth (20). In turn, the parasite has developed the ability to alter these immune cell responses (7). For example, T. gondii has been shown to reduce the amount of nitric oxide (NO) produced by macrophages in vitro by downregulating inducible nitric oxide synthase (iNOS) transcript and protein levels in infected cells (18).

The T. gondii TgPL1 gene has been implicated in the parasite's ability to reduce NO levels and survive in macrophages in vitro (22). TgPL1 encodes a gene with homology to patatin-like phospholipases (PLPs) (22). Members of this gene family, named for a stress response protein in potatoes, are known to have phospholipase A₂ (PLA₂) activity (24). Some PLPs are induced under stress conditions, like Arabidopsis homologs, which use the fatty acids produced as cell signaling intermediates (15). Others are known virulence factors, like ExoU from Pseudomonas aeruginosa, which is secreted into host cells and lyses them by using its PLA₂ activity (23). Residues conserved among members of this gene family include the oxyanion hole (G-G-X-R/K); the hydrolase motif (G-X-S-X-G), which contains the catalytic serine; and a third motif containing the catalytic aspartate (D-X-G/A) (8, 13). All of these residues are conserved in TgPL1, except for the catalytic serine (22).

The TgPL1 insertion mutant is degraded in activated macrophages despite its ability to invade host cells and establish a nonfusogenic parasitophorous vacuole. This mutant does not exhibit defects in acute or chronic infection in the CBA/J mouse model. Because small amounts of TgPL1 transcript are present in the insertional mutant (21), for this study, we generated a TgPL1 deletion mutant (ΔTgPL1) to determine the mechanism that TgPL1 uses to protect parasites within activated macrophages.

MATERIALS AND METHODS

Plasmid construction.

A construct to replace the open reading frame (ORF) of hypoxanthine-xanthine-guanosine phosphoribosyl (HPT) with firefly luciferase (ΔHPT::FLUC) was generated. The 5.3 kb upstream of HPT (5′ fragment) was amplified by using Phusion high-fidelity DNA polymerase (NEB) and primers 5′-CGGTCGACAGTCTCCTGAAGT-3′, which adds a 5′ SalI site, and 5′-CTCGAGAAGCGGACGCAGAAGCGA-3′, which adds a 3′ XhoI site. The 5.4 kb downstream of HPT (3′ fragment) was amplified by using Phusion polymerase and primers 5′-GCGGCCGCAGTACGAGAAGTGGTGCCGGT-3′, which adds a 5′ NotI site, and 5′-CAACCGAAGTTGGTGTTAGTGACTGA-3′. PCR products were cloned by using pCR-TOPO (Invitrogen). The 5′ fragment was cut with XhoI and SalI and subcloned into pTUB-FLUC (14) at the XhoI site. The 3′ fragment was then subcloned into the NotI and SacII sites, creating pΔHPT::FLUC.

A TgPL1 knockout construct was generated such that 80 bp upstream of the predicted translation start site to 10 bp downstream of the predicted poly(A) addition site was replaced with the chloramphenicol acetyltransferase (CAT) gene. Four kilobases upstream of TgPL1 (5′ fragment) was amplified by using Phusion polymerase and primers 5′-GATCACTAGTGCGTCGTATCTGCATGGAGGG-3′, which adds a 5′ SpeI site, and 5′-GATCGTTAACACACCTGAGCGTCTGTTGCC-3′, which adds a 3′ HpaI site. The 3.8 kb downstream of the stop codon (3′ fragment) was amplified by using Phusion polymerase and primers 5′-GATCGGTACCTAAGCAGTCGCCACTCCAAGAG-3′, which adds a 5′ KpnI site, and 5′-GATCCGATCGTGGCTATCATGTCGCGCTGG-3′, which adds a 3′ PvuI site. PCR products were cloned by using pCR-TOPO. The 5′ fragment was subcloned with SpeI and HpaI into pBC-CAT/HPT, which contains the cat gene from pT/230 (25) and the HPT gene from pminiHXGPRT-I (9). The 3′ fragment was subcloned into the resulting plasmid with KpnI and PvuI, creating pTgPL1-KO.

An internal hemagglutinin (HA) epitope tag was added into TgPL1 by amplifying the endogenous promoter and 5′ portion of the ORF by using Phusion polymerase and primers 5′-CAGCAGAAACGCAGATTATG-3′ and 5′-GGGGACCTCCGCGTAGTCTGGGACGTCGTATGGGTATGCGAGTTCCTCTTTGCCGTC-3′, which adds an in-frame hemagglutinin (HA) tag. This fragment was cloned by using pCR-TOPO. A TgPL1 genomic fragment was rescued from the 89B7 insertional mutant as previously described (22). This fragment was not amplified by PCR because of an AT-rich region in the 3′ untranslated region (UTR) that was consistently truncated by this method. The rescue plasmid was digested with SpeI, blunted, and then digested with SacI, and a 3.2-kb fragment corresponding to the 3′ end of the TgPL1 locus was gel purified. The vector containing the 5′ end of the locus was digested with KpnI, blunted, and then digested with SacI to accommodate the ligation of the fragment from the rescue plasmid, creating pTgPL1-SacHA.

T. gondii strains, cell culture, and transfections.

The type II parasite strain Prugniaud (Pru) or ME49 was used in all experiments. Parasites were passaged in confluent human foreskin fibroblasts (HFFs) as previously described (21). Bone marrow cells (BMCs) were isolated from C57BL/6 mice (National Cancer Institute, Charles River Laboratories, Frederick, MD) and developed into bone marrow-derived macrophages (BMMs) in medium containing 20% L929 cell conditioned medium as previously described (21). Subsequent assays were performed with Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS) supplemented with 2 mM glutamate, 0.5% penicillin-streptomycin, and 50 mg/ml gentamicin.

To generate a luciferase-expressing wild-type (WT) T. gondii strain, Pru or ME49 type II parasites were transfected with 50 μg of pΔHPT::FLUC linearized at the SacII site. Transfected parasites were selected with 340 μg/ml 6-thioxanthine for 14 days and then cloned by limiting dilution. Homologous recombination was confirmed by Southern analysis and luciferase activity, resulting in PruΔHPT::Luc and ME49ΔHPT::Luc.

To generate a deletion of the TgPL1 gene, PruΔHPT::Luc was transfected with 50 μg of linearized pTgPL1-KO. Transfections underwent positive selection in 20 μM chloramphenicol until no growth was observed for the mock transfection, followed by negative selection with 340 μg/ml 6-thioxanthine for 9 days to select for homologous recombination. The ΔTgPL1 mutant was cloned by limiting dilution and confirmed by Southern blotting. To complement the ΔTgPL1 mutant, two independent electroporations with 50 μg of pEndoTgPL1 were selected and cloned as previously described (22).

T. gondii bradyzoite development was achieved by incubating parasites in RPMI 1640 without bicarbonate, 1% FBS, 1% penicillin-streptomycin, and 42 mM HEPES (pH 8.0).

In vitro growth assay.

Host cells were passaged and seeded onto coverslips as previously described (26) and infected with 1 × 10⁵ parasites. After 3 h, BMMs were left unstimulated, classically activated with 100 ng/ml lipopolysaccharide (LPS) and 100 U/ml gamma interferon (IFN-γ), or activated with the addition of 10 mM aminoguanidine (an inducible nitric oxide synthase inhibitor). Griess reactions were performed as previously described, to confirm NO production in classically activated samples and the suppression of NO production in classically activated samples containing aminoguanidine (22). Additionally, BMMs were treated with medium containing 50, 100, or 200 μM the nitric oxide donor diethylenetriamine (DETA) nonoate (Cayman Chemicals) plus HEPES or HEPES alone. After 24 h, coverslips were fixed, permeabilized, and stained with chronic mouse serum as previously described (22). Parasite growth was assessed by counting 100 vacuoles per coverslip and categorizing each vacuole as containing degraded parasites, 1 parasite, 2 parasites, 4 parasites, or 8 or more parasites. Samples were blinded to eliminate bias.

Immunofluorescence staining.

BMMs were seeded onto coverslips and infected with T. gondii expressing an HA-tagged version of TgPL1. Parasites were allowed to invade for 3 h. BMMs were classically activated by changing the medium containing 100 U/ml IFN-γ and 100 mg/ml LPS. Samples were fixed after the indicated time for 20 min in 3% formaldehyde. Excess formaldehyde was quenched with 0.1 M glycine for 3 min. Samples were blocked and permeabilized for 1 h at room temperature or overnight at 4°C. Samples were stained with mouse anti-hemagglutinin antibody (Covance) and detected with Alexafluor-488-conjugated donkey anti-mouse secondary antibody and rhodamine-conjugated Dolichos biflorus agglutinin. Coverslips were mounted in Vectashield mounting medium with 4′,6-diamidino-2-phenylindole (DAPI) (Vector Labs). Serial image stacks (0.2-μm Z increment) were taken at a ×100 magnification (PlanApo oil immersion, 1.4 numerical aperture [NA]) using a motorized Zeiss Axioplan III instrument equipped with a rear-mounted excitation filter wheel, a triple-pass (DAPI-fluorescein isothiocyanate [FITC]-Texas Red) emission cube, differential interference contrast optics, and a Hamamatsu Orca-AG charge-coupled-device (CCD) camera operated by OpenLabs 4.0 software (Improvision, Lexington, MA). Fluorescence images were deconvolved by a constrained iterative algorithm, pseudocolored, and merged by using the Volocity software package (Perkin-Elmer).

Induction and purification of TgPL1 from E. coli.

The TgPL1 ORF was amplified by using Accuprime Pfx polymerase (Invitrogen) from cDNA using primers 5′-GATCAGATGTACACACGCTCCAGTGCAAC 3′, which adds a 5′ PciI site, and 5′-GCGGCCGCAGACTCTTCAGACTTTGCCTCTTCG-3′, which adds a 3′ NotI site. This product was cloned into pCR-TOPO (Invitrogen) and subcloned into pET28a at NcoI (which has compatible ends with PciI) and NotI to create pTgPL1-His. This plasmid was transformed into Rosetta(DE3)pLysS competent E. coli cells (Novagen). Protein production was induced for 3 h with 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) in 100-ml culture flasks started from a culture grown overnight and grown to an optical density at 600 nm (OD₆₀₀) of 0.5 to 0.7. Protein was purified from pelleted bacteria in denaturing buffer using nickel-nitrilotriacetic acid (NTA) resin (Qiagen). The protein was then dialyzed into 1× PBS for subsequent assays.

Phospholipase A₂ activity assay.

Phospholipase A₂ activity assays were performed as previously described (4). Briefly, 50 μl of partially purified TgPL1-6×His from E. coli was added to 50 μl 2× PLA₂ buffer (100 mM Tris-HCl [pH 7.5], 200 mM NaCl, 2 mM EDTA, 4 μM fluorescent phospholipid substrate, 0.2% bovine serum albumin [BSA], 12 mM CaCl₂) and incubated at 37°C for 1 h. Fluorescence was measured at excitation and emission wavelengths of 345 and 398 nm, respectively.

Mouse infection.

For intraperitoneal infections, tachyzoites were grown to near lysis in HFFs, syringe lysed, and enumerated on a hemocytometer. A total of 2 × 10⁴ parasites were injected into the peritoneal cavity of 10- to 12-week old female C57BL/6 mice. Infection was allowed to progress until severe neurological symptoms were evident, at which time the mice were sacrificed. Mice that survived to the chronic stage of infection were sacrificed at 25 days postinfection, and their brains were harvested to assess cyst burden as previously described (22). For oral infections, approximately 10⁶ in vitro-derived bradyzoite cysts were fed to female C57BL/6 mice by oral gavage. Mice were sacrificed as indicated above.

Comparative fitness assay.

Growth competition experiments were performed similarly to previously described assays (11). For tissue culture, parasites of the ΔTgPL1 mutant and the ΔTgPL1 mutant complemented with a functional copy of the TgPL1 gene were mixed 1:1, and 10⁶ parasites were then inoculated into HFF host cells and serially passaged without drugs every 4 days for 20 days. At every passage, parasites were removed for plaque assays with or without 1 μM pyrimethamine, as the comp parasites contain the mutant DHFR gene, which confers resistance to pyrimethamine. For in vivo competition studies, C57BL/6 female mice were injected intraperitoneally (i.p.) with 2 × 10⁴ parasites of a 1:1 mix of the ΔTgPL1 mutant and comp. Parasites were harvested by peritoneal lavage every 4 days for serial passage into another mouse and plaque assays with and without pyrimethamine selection. Relative fitness was assessed by the slope of the line, represented by the change over time of the percentage of ΔTgPL1 parasites in the total population of plaques.

RESULTS

The TgPL1 gene was replaced with the CAT gene by homologous recombination.

Four kilobases of upstream and 3.8 kb of downstream TgPL1 genomic sequence were amplified and cloned into a vector with CAT as a positive selectable marker and HPT as a negative selectable marker (Fig. 1A). PruΔHPT::Luc parasites were transfected with plasmid pTgPL1-KO and selected for resistance to both chloramphenicol and 6-thioxanthine. The parental strain PruΔHPT::Luc was used as the wild type (WT) for all experiments. Clones containing the potential TgPL1 deletion were analyzed by Southern analysis using a probe of a 500-bp fragment of the TgPL1 5′ UTR. For the XbaI-digested samples, the expected sizes were 4.8 and 3.7 kb for the WT and TgPL1 knockout clones, respectively. For the ApaLI-digested samples, the expected sizes were 1.5 and 2.2 kb for the WT and the TgPL1 knockout, respectively. ΔTgPL1 exhibited the expected banding pattern (Fig. 1B).

Fig 1.

Replacement of the TgPL1 coding region with CAT. (A) Diagram of the ΔTgPL1 strategy. White bars indicate 5′- and 3′-flanking regions included in the knockout construct (4 kb and 3.8 kb, respectively). The black box above the diagram indicates the fragment of the TgPL1 5′ UTR used as a Southern probe. The black lines indicate non-protein-coding regions of TgPL1 (introns and UTRs). The light gray boxes indicate the two exons of TgPL1 and the coding region of CAT. X indicates XbaI sites, and A indicates ApaLI sites. (B) DNAs from wild-type (WT) and potential knockout parasites were digested with ApaLI or XbaI and analyzed by Southern analysis. The band sizes were as expected for the WT and the ΔTgPL1 mutant when digested with ApaLI (1.5 and 2.2 kb, respectively) and XbaI (3.7 and 4.8 kb, respectively).

The ΔTgPL1 mutant has a growth defect in classically activated but not naïve macrophages.

To determine if the ΔTgPL1 mutant exhibits a growth phenotype in macrophages similar to that of the TgPL1 insertional mutant (21), growth was assessed in naïve and classically activated macrophages. BMMs were infected with WT and ΔTgPL1 parasites, and growth was assessed by an immunofluorescence assay (IFA) at 24 h postinfection. Vacuoles were stained by using serum from chronically infected mice, and growth was evaluated by tallying the number of parasites per vacuole. ΔTgPL1 parasites grew as well WT parasites in naïve macrophages, with over 50% of the vacuoles assessed containing replicating parasites (Fig. 2A). In contrast, ΔTgPL1 parasites exhibited a growth defect in classically activated macrophages, characterized by nearly 30 to 40% of the vacuoles assessed containing degraded parasites, whereas wild-type parasites grew to 4 parasites per vacuole (Fig. 2B). To confirm that the deletion of TgPL1 is responsible for the above-described phenotype, complementation was performed by using the TgPL1 genomic region, which includes the endogenous promoters and UTRs (22). The introduction of the TgPL1 genomic region into the ΔTgPL1 mutant allows the parasites to replicate similarly to WT parasites without the high levels of degradation seen with ΔTgPL1 parasites (Fig. 2B).

Fig 2.

ΔTgPL1 has a NO-dependent growth defect in activated but not naïve bone marrow-derived macrophages. (A to C) Wild-type (WT) parasites, ΔTgPL1 parasites, and ΔTgPL1 parasites complemented with TgPL1 (comp) were grown in bone marrow-derived murine macrophages with no treatment (naïve) (A), activation by IFN-γ and LPS (B), and activation by IFN-γ and LPS plus the NO inhibitor aminoguanidine (C). Parasites were fixed after 24 h and detected by an IFA using serum from chronically infected mice. Vacuoles were first assessed for whether the parasites in each vacuole were intact or degraded. If intact, the number of parasites per vacuole was recorded. (D) Similar assays were performed with naïve or activated (Act) iNOS^−/− macrophages. The Student t test P value of <0.01 is indicated by an asterisk.

The ΔTgPL1 degradation phenotype is abolished when NO is absent.

To investigate whether NO is necessary for the ΔTgPL1 phenotype in classically activated macrophages, growth assays were performed under conditions that reduced or eliminated NO. To reduce the level of NO, the iNOS inhibitor aminoguanidine was added to macrophages concurrent with LPS and IFN-γ activation and in parallel with the experiments shown in Fig. 2A and B. No parasite growth defect was seen for the ΔTgPL1 mutant when macrophages were activated in the presence of aminoguanidine (Fig. 2C). To fully eliminate the presence of NO, BMMs were isolated from iNOS^−/− mice. Similar to the results seen with the iNOS inhibitor aminoguanidine, ΔTgPL1 did not display a phenotype in iNOS^−/− BMMs (Fig. 2D). Whether grown in macrophages with the inhibitor aminoguanidine or in naïve or activated BMMs from iNOS^−/− mice, ΔTgPL1 grew similarly to WT parasites, with the majority of vacuoles containing 4 to 8 parasites (Fig. 2).

ΔTgPL1 does not exhibit a growth defect in BMMs with exogenous nitric oxide.

To determine whether NO is sufficient to cause the ΔTgPL1 mutant to have a replication defect, WT and ΔTgPL1 parasites were grown in naïve macrophages with the NO donor DETA NONOate. WT and ΔTgPL1 parasites were allowed to invade the BMMs before the addition of 50, 100, or 200 μM the NO donor DETA NONOate. WT and ΔTgPL1 parasites both showed a reduction in replication with increasing amounts of the NO donor (Fig. 3). In samples treated with 200 μM DETA NONOate, less than 20% of the vacuoles had more than one parasite (Fig. 3D). However, the ΔTgPL1 mutant did not exhibit a replication defect relative to the WT and was not degraded, showing that exogenously added NO is not sufficient to cause the phenotype of the ΔTgPL1 mutant.

Fig 3.

Addition of an NO donor is not sufficient for the degradation phenotype. Bone marrow-derived macrophages were infected with wild-type (WT) and ΔTgPL1 parasites and treated with 0 μM (A), 50 μM (B), 100 μM (C), or 200 μM (D) the NO donor DETA NONOate. Parasite growth was assessed at 24 h postinfection as described in the legend of Fig. 1.

Peroxynitrite does not inhibit the growth of WT or ΔTgPL1 parasites.

Reactive oxygen intermediates, such as superoxide, are also produced in response to stimulation with IFN-γ. Superoxide can react with NO to form peroxynitrite, which is a powerful oxidizing agent. To determine if peroxynitrite is able to mediate the degradation of ΔTgPL1 parasites in BMMs, Infected BMMs were treated with 0, 50, 100, or 200 μM peroxynitrite and incubated for 24 h, at which point parasite growth was assessed as described above. The 200 μM concentration of peroxynitrite caused the host cells to lift off the coverslip (data not shown). All other concentrations were ineffective at inhibiting parasite growth, indicating that peroxynitrite alone does not cause the degradation of ΔTgPL1 parasites (see Fig. S2 in the supplemental material).

The ΔTgPL1 mutant does not exhibit a virulence defect in C57BL/6 mice.

To examine whether the ΔTgPL1 mutant has a defect in causing disease in mice, 10- to 12-week-old C57BL/6 mice were infected with 2 × 10⁴ or 5 × 10⁴ tachyzoites. Both WT- and ΔTgPL1-infected mice showed about 15% mortality at a dose of 2 × 10⁴ tachyzoites and 100% mortality at a dose of 5 × 10⁴ tachyzoites in C57BL/6 mice (see Fig. S3B and S3C in the supplemental material). To determine if the deletion of TgPL1 affects levels of cyst formation, cyst burdens from the surviving mice infected with 2 × 10⁴ tachyzoites were assessed at 25 days postinfection. WT- and ΔTgPL1-infected mice both had approximately 60,000 cysts per brain (see Fig. S3D in the supplemental material). The virulence of ΔTgPL1 parasites was also examined by the oral inoculation of in vitro-developed bradyzoite cysts, but no significant differences were seen (data not shown). Finally, we examined whether the ΔTgPL1 mutant had a growth defect relative to parasites that had a functional copy of the TgPL1 gene (comp) in a comparative fitness assay. Comp parasites were used in favor of WT parasites because they can be easily differentiated from ΔTgPL1 parasites by the presence of the dihydrofolate reductase (DHFR) marker. The ratio of ΔTgPL1 to comp parasites did not significantly change in tissue culture or in mice over the 20-day experiment (see Fig. S3A in the supplemental material). Taken together, these results show that the TgPL1 gene does not play a significant role in the virulence of T. gondii during acute infection.

Partially purified TgPL1 does not have PLA₂ activity.

TgPL1 belongs to the patatin family of proteins that typically have PLA₂ activity; however, TgPL1 has an S-to-G mutation at the predicted catalytic serine. E. coli-derived TgPL1 with a C-terminal histidine tag does not exhibit PLA₂ activity when incubated with a phosphocholine that has pyrene conjugated to the fatty acid in the sn-2 position (see Fig. S4 in the supplemental material). Surprisingly, even when the catalytic serine is restored by site-directed mutagenesis, the purified mutant TgPL1 protein does not have PLA₂ activity. This lack of PLA₂ activity could be due to a misfolding of recombinant TgPL1 in the E. coli system. Future studies will assess the enzymatic activity of TgPL1 in T. gondii.

The TgPL1 protein localizes to the PV space in macrophages.

To aid in determining the function of TgPL1 in resistance to NO in activated macrophages, a hemagglutinin (HA)-tagged version of TgPL1 (TgPL1-HA) was used to aid the localization of TgPL1. Macrophages were infected and activated as described above for the degradation assay. Naïve and activated samples were fixed, stained, and analyzed by immunofluorescence microscopy. In both activated and naïve samples, TgPL1-HA colocalized with the parasite dense-granule marker GRA4 at day 1 postinfection (Fig. 4). The same localization was seen for parasites grown in activated macrophages for extended periods of time (5 days postinfection). These results suggest that TgPL1 may function in the parasitophorous vacuole to protect parasites from nitric oxide produced by these immune cells.

Fig 4.

TgPL1 localization change in activated macrophages. An HA-tagged version of TgPL1 was used to assess its localization in naïve macrophages at 1 day postinfection (Day 1) and activated macrophages at either 1 day (Day 1-Act) or 5 days (Day 5-Act) postinfection. TgPL1 partially colocalizes with the dense-granule protein GRA4, indicating that TgPL1 is outside the parasite in the parasitophorous vacuole space. The white scale bar is 5 μm. The scales for all panels are identical. DIC, differential interference contrast.

DISCUSSION

Although the production of NO is important for the host to control pathogen replication in many systems, little is known about genes that allow pathogens to evade this host response (5). TgPL1 was previously identified as a parasite factor involved in the ability of T. gondii to survive in activated macrophages (21). We have determined that NO is necessary but not sufficient for the ΔTgPL1 growth defect. Finally, we have seen that TgPL1 localizes to the parasitophorous vacuole in naïve and classically activated macrophages. These data provide a first step toward an understanding of the mechanism of how TgPL1 protects T. gondii against degradation in activated macrophages.

The inability of exogenous NO to degrade ΔTgPL1 mutant parasites may indicate that NO is acting synergistically with other IFN-γ-inducible factors in macrophages to cause this phenotype. For example, immunity-related GTPases are also induced by IFN-γ and have been associated with the disruption of the T. gondii vacuole in macrophages and astrocyte cells (2, 17, 19). These could weaken the vacuole sufficiently to allow the NO to degrade the parasite. This possibility is unlikely, because immunity-related GTPases (IRG) strip the parasitophorous vacuole from the parasite, leaving free parasites in the host cell cytoplasm to be removed by autophagy (17), whereas TgPL1-deficient parasites tend to become degraded before the PV is disrupted (22).

Alternatively, exogenous NO could have different effects than those of NO produced by the cell in which the parasite is growing. The concentration of nitrite in the medium as measured by the Griess reaction might not reflect the intracellular concentration of NO to which the parasite is exposed in the macrophage. This idea is congruent with the presence of both cytosolic and membrane-bound forms of iNOS, which could be used as a way to deliver NO to pathogen-containing vacuoles while reducing the NO exposure of the rest of the cell (28). In this case, dissecting the role of endogenously produced NO versus other IFN-γ-inducible genes is difficult, as there are currently no tools available to induce iNOS in the absence of other IFN-γ genes in macrophages.

While reactive oxygen species have been shown to be unnecessary for the control of parasite replication in activated macrophages in mice, the role of peroxynitrite in this process had not been directly assessed (6). We have eliminated the possibility that degradation occurs because NO combines with reactive oxygen intermediates to form peroxynitrite. This highly reactive compound contributes to the killing of the fungal pathogen Candida albicans in NO-producing macrophages (27); however, this compound is less effective than NO at killing parasitic organisms such as Leishmania major and Giardia lamblia (1, 10). Thus, it is not surprising that it is ineffective against the T. gondii parasite.

The initial mouse studies done with CBA/J mice infected with the TgPL1 insertional mutant showed that there was no virulence defect. We reassessed the contribution of TgPL1 to virulence using a complete deletion in C57BL/6 mice that have high Th-1 polarization, which is important for the production of IFN-γ and, hence, NO. The results of these studies showed that TgPL1 is not essential to the parasite's ability to cause disease during the acute stage of infection. It is also expendable for cyst formation in an oral infection model. This result may indicate that survival in activated macrophages is not important for these processes or that the ΔTgPL1 mutant parasite compensates for its defect by using alternative pathways to grow and reach distal sites of the host. Replication and dissemination could occur in peritoneal cells and naïve macrophages, respectively, where the ΔTgPL1 mutant has no growth defect, or in dendritic cells, where growth of ΔTgPL1 has not been assessed.

Unlike other patatin-like phospholipases associated with pathogens, TgPL1 seems to be acting more like the plant stress response proteins rather than cytolytic PLA₂ enzymes, such as ExoU from Pseudomonas aeruginosa (23). In response to environmental stresses, plant patatins change their location from within low-pH vesicles to the cytoplasm (16). The localization of TgPL1 to the parasitophorous vacuole in macrophages places the protein directly at the interface of the host-parasite interaction, even before the onslaught of NO caused by classical activation. Future studies will investigate the levels and timing of TgPL1 that are necessary for NO protection. A constitutively expressed version of TgPL1 fused with dense granules to localize TgPL1 into parasitophorous vacuoles even without stress will allow the necessary levels and functional domains of TgPL1 to be dissected. Determining the functional domains of TgPL1 will shed light on how T. gondii evades the antimicrobial effects of NO.