T. gondii strains ME49 and Prugniaud were maintained in Vero cells in RPMI (Nacalai Tesque) supplemented with 2% heat-inactivated fetal bovine serum (FBS; JRH Bioscience), 100 U/mL penicillin, and 0.1 mg/mL streptomycin (Nacalai Tesque), as previously described (Ma et al., 2014). HFFs were maintained in RPMI (Nacalai Tesque) supplemented with 2% heat-inactivated FBS (JRH Bioscience), 100 U/mL penicillin, and 0.1 mg/mL streptomycin (Nacalai Tesque). IMR-32 cells were maintained in MEM (Nacalai Tesque) containing 10% heat-inactivated FBS, 1% non-essential amino acids (Nacalai Tesque), 100 U/mL penicillin, and 0.1 mg/mL streptomycin. A172 cells were maintained in DMEM (Nacalai Tesque) containing 10% heat-inactivated FBS, 100 U/mL penicillin, and 0.1 mg/mL streptomycin. U251-MG cells were maintained in EMEM (Nacalai Tesque) containing 10% heat-inactivated FBS, 100 U/mL penicillin, and 0.1 mg/mL streptomycin.

To prepare iPSC-derived glutamatergic neurons, ioGlutamatergic neurons were obtained from Abcam (ab259259). ioGlutamatergic neurons (5.7 × 10⁴) were plated in a 24-well plate containing PDL-Geltrex-coated glass coverslips. On days 0–4, the neurons were cultured in complete glutamatergic neuron medium (CGNM) containing 1 μg/mL doxycycline (Sigma–Aldrich). The CGNM comprised 200 mL of Neurobasal medium (Thermo Fisher), 2 mL of GlutaMAX (100×) (Thermo Fisher), 25 μM 2-mercaptoethanol (Thermo Fisher), 4 mL of B27 (Thermo Fisher), 10 ng/mL NT3 (R&D), and 5 ng/mL BDNF (R&D). On days 5–14 days, the neurons were cultured in CGNM without doxycycline. The medium was changed every 48 h. After 12 days in culture, differentiation of iPSC-derived glutamatergic neurons was confirmed by morphology and gene expression of neuron or glutamatergic neuron-specific markers by use of microscopy and an immunofluorescence assay, respectively.

Antibodies against TUBB3 (66375-1-lg), KGA (20170-1-AP), and VGLUT1 (55491-1-AP) were obtained from Proteintech. Salmon E monoclonal antibody for CST1 staining has been described previously (Tomita et al., 2013). The anti-GAP45 antibody was kindly gifted by Dr. Dominique Soldati-Favre (University of Geneva, Switzerland). Recombinant human IFN-γ (300-02) was obtained from Peprotech. Aminoguanidine hydrochloride (396494) was obtained from Sigma–Aldrich. 1-Methyl-DL-tryptophan (sc-224746) and KGA siRNA (sc-105592) were obtained from Santa Cruz Biotechnology, Inc. CB-839 (Cay-22038) was obtained from Cayman Chemical.

Bradyzoite differentiation was confirmed by the gene expression pattern of SAG1 and BAG1, and by staining of the cyst wall protein CST1. HFFs, IMR-32, A172, U-251 MG cells, or iPSC-derived glutamatergic neurons were cultured and infected with T. gondii [multiplicity of infection (MOI) = 0.5]. For alkaline induction, the culture medium was changed to non-induction medium (pH 7.2) or induction medium (pH 8.2) at 2 h post-infection. For IFN-γ induction, the cells were stimulated with or without IFN-γ (10 ng/ml) at 2 h post-infection. Infected host cells were incubated for 24–72 h under low CO₂ conditions. The cyst wall was stained by using an anti-CST1 antibody, and CST1-positive vacuoles were defined as bradyzoite differentiated parasitophorous vacuoles. Quantitative measurements of the cyst wall-positive vacuole rates were performed by counting at least 100 vacuoles per sample.

Total RNA from cells or parasites was extracted by using an RNA basic kit (Nippon genetics), and cDNA was synthesized by using Verso Reverse transcription (Thermo Fisher). Quantitative RT-PCR was performed with a Thermal Cycler Dice Real Time PCR System (Takara) using the Go-Taq Real-Time PCR system (Promega). The values were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) for human cells or Tubulin or Actin for T. gondii in each sample. The primer sequences are listed in S1 Table .

IMR-32 cells or A172 cells were plated in a 6-well plate, and then infected or non-infected cells were incubated for 48 h. The culture supernatants were then collected and filtrated through a 0.4-μm filter membrane before being analyzed. The concentrations of 11 major ions ( F − ,   C l − ,   B r − ,   N O 2 − ,   P O 4 3 − ,   S O 4 2 − , N A + ,   N H 4 + ,   K + ,   M g 2 + ,   C a 2 + ) were measured by using the HIC-20A SUPER ion chromatography system (Shimazu).

The kynurenine concentration in the culture medium was measured by using the Ehrlich reagent method (Braun et al., 2005). Briefly, 70 μL of culture supernatant was mixed with 35 μL of 30% trichloroacetic acid, and centrifuged at 8,000 × g for 5 min. Then, 75 μL of the supernatant was added to an equal volume of Ehrlich reagent (0.8% p-dimethylaminobenzaldehyde in acetic acid) in a 96-well plate, and the absorbance was read at 490 nm. The values were determined by using a standard curve with defined concentrations of kynurenine (Sigma–Aldrich).

IMR-32 cells or iPSC-derived glutamatergic neurons were plated in a 96-well plate, and then infected and non-infected cells were incubated for 24 or 48 h. The concentration of extracellular and intracellular glutamine was measured by using the Glutamine/Glutamate-Glo™ Assay (Promega) and a GloMax Navigator Microplate Luminometer (Promega) according to the manufacturer’s instructions.

HFFs, IMR-32, A172, and U-251 MG cells, or iPSC-derived glutamatergic neurons were cultured on glass coverslips and infected with T. gondii (MOI = 0.5) for the indicated time. The cells were then fixed in PBS containing 4% paraformaldehyde for 15 min at room temperature. Cells were permeabilized with PBS containing 0.1% Triton X-100 for 5 min, and then blocked with 2% FBS in PBS for 1 h at room temperature. Next, the cells were incubated with the indicated primary antibodies for 1 h at 37°C, followed by incubation with Alexa 488- or Alexa 594-conjugated secondary antibodies (Molecular Probes) and DAPI for 1 h at 37°C in the dark. Finally, coverslips were mounted onto glass slides with ProLong Glass Antifade Mountant (Invitrogen) and analyzed by using a BZ-X810 All-in-one Fluorescence Microscope (Keyence).

IMR-32 cells were pre-treated with Nω-Propyl-L-Arginine hydrochloride (2 μM) or 1-methyl-DL-tryptophan (1 mM) for 3 h, and then infected with the parasite. The culture medium was changed 2 h post-infection, and fresh medium containing 50 ng/mL IFN-γ and Nω-Propyl-L-Arginine hydrochloride (2 μM) or 1-Methyl-DL-tryptophan (1 mM) was added for 24–72 h.

HFF cells were pre-treated with CB-839 (2 μM) for 3 h, and then infected with the parasite. The culture medium was changed 2 h post-infection, and fresh medium containing 50 ng/mL IFN-γ with or without CB-839 (2 μM) was added for 5 days. Cells were fixed with 4% paraformaldehyde for 30 min at room temperature, then washed with PBS and stained with 0.1% crystal violet (CV) for 10 min. Images were analyzed by using Image J.

Cells were lysed in a lysis buffer (0.5% Nonidet P-40, 150 mM NaCl, and 20 mM Tris-HCl, pH 7.5) containing a protease inhibitor cocktail (Roche). The cell lysates were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes (Immobilon-P, Millipore) and subjected to Western blot analyses as described previously (Bando et al., 2019).

ioGlutamatergic neurons (5.7 × 10⁴) were plated in a 24-well plate and developed to iPSC-derived glutamatergic neurons. Ten days after development, KGA siRNA was transfected by using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. At 24 h post-lipofection, the wells were washed and then incubated for an additional 24 h. KGA siRNA-transfected, iPSC-derived glutamatergic neurons were infected with T. gondii (MOI = 0.5) for 2 h, and then stimulated with or without IFN-γ (10 ng/ml). Infected host cells were incubated for 24–72 h under low CO₂ conditions.

All statistical analyses were performed by using Excel (Microsoft) or Prism 8 (GraphPad). All experimental points and n values represent an average of three biological replicates (three independent experiments). The statistical significance of differences in mean values was analyzed by using an unpaired two-tailed Student’s t-test. p < 0.05 was considered statistically significant.

The physiological conditions that induce T. gondii stage conversion in human brain cells are unclear. Therefore, we tested the effect of IFN-γ on the expression of the bradyzoite-specific gene BAG1 or the tachyzoite-specific gene SAG1 by using the T. gondii Type II ME49 strain in the following human brain cell lines: astrocytoma cell line (A172), glioblastoma cell line (U-251 MG), and neuroblastoma cell line (IMR-32) ( Figures 1A, B ). Although alkaline stress (pH 8.2) induced BAG1 gene expression in all cell lines tested ( Figure S1A ), IFN-γ-dependent upregulation of BAG1 gene expression was observed in IMR-32 cells, but not in A172 or U-251 MG cells ( Figure 1A ). In addition, we found that SAG1 gene expression was downregulated in IMR-32 cells ( Figure 1B ). T. gondii cyst wall CST1 formation around bradyzoites in IMR-32 cells was found only in IFN-γ-stimulated cells ( Figures 1C, D ). These phenomena were also observed with the Type II T. gondii Prugniaud strain ( Figures S1B, C ). These results indicate that Type II T. gondii differentiate into bradyzoites and form a cyst wall in response to IFN-γ stimulation in human neuroblastoma cells.

IFN-γ-induced bradyzoite differentiation in murine BMDMs depends on NO production (Bohne et al., 1994). In our study, an IFN-γ-dependent increase in NO production was detected in IMR-32 cells, but not in A172 or U251-MG cells ( Figure 2A ). Humans have three isoforms of nitric oxide synthase (NOS): inducible NOS (iNOS), epithelial NOS (eNOS), and neuronal NOS (nNOS) (Umar and van der Laarse, 2010). We therefore tested which NOS isoform is important for NO production in IFN-γ-stimulated IMR-32 cells ( Figure 2B ). We detected nNOS gene expression in IMR-32 cells but not in A172 cells in an IFN-γ-dependent manner ( Figure 2B and Figure S2A ). Then, we tested the effect of IFN-γ-dependent NO production on bradyzoite differentiation in IMR-32 cells by using Nω-propyl-L-arginine (L-NPA), a highly selective nNOS inhibitor ( Figures 2C, D ). Although L-NPA inhibited NO production in IFN-γ-stimulated IMR-32 cells ( Figure 2C ), there was no difference in BAG1 or SAG1 gene expression in the presence or absence of L-NPA ( Figure 2D ). These results suggest that NO production does not play a key role in the induction of bradyzoite differentiation in IFN-γ-stimulated human neuroblastoma cells.

IFN-γ stimulation induces IDO1-dependent tryptophan starvation in various human cells (Bando et al., 2018a). Therefore, we tested the effect of IFN-γ stimulation on IDO1 expression in IMR-32 cells ( Figures 3A, B ). IDO1 gene expression was induced in IFN-γ-stimulated IMR-32 cells ( Figure 3A ). Furthermore, kynurenine, which is a metabolite in the IDO pathway, was detected in the culture medium of IFN-γ-stimulated IMR-32 cells ( Figure 3B ). Then, we examined the effect of IDO1 gene expression on bradyzoite differentiation in IMR-32 cells treated with 1-methyl-DL-tryptophan (1-DL-MT), which is an IDO1 inhibitor ( Figures 3C, D ). Although 1-DL-MT treatment inhibited IDO1 activity ( Figure 3C ), there was no difference in the BAG1 or SAG1 gene expression pattern ( Figure 3D ). These results suggest that IDO1-dependent tryptophan starvation stress does not affect bradyzoite differentiation in IFN-γ-stimulated human neuroblastoma cells.

IFN-γ stimulation induced bradyzoite differentiation only in neuroblastoma cells ( Figure 1A ). Therefore, to elucidate the neuronal cell-specific response to IFN-γ stimulation, we compared the IFN-γ-inducible gene expression pattern of predicted-anti-T. gondii response genes in A172 and IMR-32 cells ( Figures S2A, B ). We did not find an IMR-32 cell-specific gene expression pattern other than that of nNOS ( Figures S2A, B ). Next, we compared the metabolic profiles by using ion chromatography ( Figures S3A, B ). We found that N O 2 − and N H 4 + were specifically high in the culture medium of IFN-γ-stimulated IMR-32 cells ( Figures S3A, B ). This IFN-γ-dependent nNOS gene expression and NO production confirmed our previous results ( Figures 2A, B ). Therefore, we focused on N H 4 + as a host factor candidate involved in the induction of bradyzoite differentiation and cyst formation.

NH₄ accumulation is toxic to many cell types and activates several stress responses (Braissant et al., 2013; Wang et al., 2018). Hence, we examined the effect of a high concentration of NH₄ in the culture medium on BAG1 or SAG1 gene expression in IMR-32 cells ( Figure 4A ). There was no difference in either gene expression pattern in IMR-32 cells cultured with or without a high concentration of NH₄ ( Figure 4A ). It has been reported that most of the NH₄ produced in the brain is derived from glutamine metabolism, which includes glutamine influx and glutaminase degradation (Bak et al., 2006). Therefore, we next examined the IFN-γ-dependent expression of the main glutamine transporters: solute carrier family 38 members A1 and A2 (SLC38A1 and SLC38A2) (Varoqui et al., 2000; González-González et al., 2005), in IMR-32 cells ( Figure 4B ). Although IFN-γ induced SLC38A1 and SLC38A2 gene expression in non-infected cells, this expression was inhibited in T. gondii-infected cells ( Figure 4B ), suggesting that T. gondii infection suppressed IFN-γ-dependent glutamine influx in IMR-32 cells. We next examined the glutamine dynamics in the culture medium (extracellular) ( Figures 4C, D ) and in the IMR-32 cells (intracellular) ( Figure 4E ). The extracellular glutamine concentration in the culture medium of IFN-γ-stimulated, T. gondii-infected IMR-32 cells was higher than that of IFN-γ-stimulated, non-infected IMR-32 cells ( Figure 4C ). Furthermore, the IFN-γ-induced NH₄ increase in the culture medium was inhibited by T. gondii infection ( Figure 4D ). In contrast, the intracellular glutamine concentration in IFN-γ-stimulated, T. gondii-infected IMR-32 cells was significantly reduced compared with that in IFN-γ-stimulated, non-infected IMR-32 cells ( Figure 4E ). These results suggest that T. gondii infection leads to intracellular glutamine starvation via suppression of glutamine transporter activity in human neuroblastoma cells.

Glutamine serves as a bioenergetic substrate for T. gondii growth (Blume et al., 2009), suggesting the possibility that glutamine starvation triggers T. gondii stage conversion. We showed that bradyzoite differentiation is not induced in unstimulated IMR-32 cells ( Figure 1 ); therefore, we hypothesized that pharmacological inhibition of IFN-γ-dependent glutamine degradation may prevent glutamine starvation and thereby suppress IFN-γ-dependent T. gondii stage conversion in IMR-32 cells. To test this hypothesis, we examined the effect of CB-839, a selective glutaminase inhibitor (Xu et al., 2019), on the intracellular glutamine concentration ( Figure 5A ). We found that CB-839 treatment restored the intracellular glutamine concentration in IFN-γ-stimulated, T. gondii-infected IMR-32 cells ( Figure 5A ). Next, we examined the effect of CB-839 treatment on the parasite by using a plaque assay, and found that CB-839 treatment had no effect on the parasite plaque size or number ( Figures 5B, C ). Then, we examined the effect of CB-839 treatment on IFN-γ-dependent bradyzoite differentiation ( Figures 5D, E ). We found that IFN-γ-induced BAG1 gene expression was suppressed in IFN-γ-stimulated IMR-32 cells treated with CB-839 ( Figure 5D ). Importantly, pretreating T. gondii with CB-839 did not affect the IFN-γ-induced BAG1 gene expression ( Figure 5E ), suggesting that the effect of CB-839 treatment was the result of the inhibition of the host glutaminase activity. Furthermore, we examined the effect of CB-839 on T. gondii cyst wall CST1 formation ( Figures 5F, G ). We found that CB-839 treatment inhibited IFN-γ-induced cyst wall CST1 formation ( Figures 5F, G ). These results suggest that IFN-γ-induced intracellular glutamine starvation triggers T. gondii stage conversion in human neuroblastoma cells.

Glutamatergic neurons produce glutamate, which is one of the most common excitatory neurotransmitters in the CNS (Zeng and Sanes, 2017). To confirm the glutamine starvation-induced T. gondii stage conversion in glutamatergic neurons, we differentiated glutamatergic neurons from human induced pluripotent stem cells (iPSCs) ( Figures S4A, B ). Axon elongation was evident in a healthy neuronal culture of iPSC-derived neurons 12 days after differentiation induction ( Figure S4A ). In addition, the neuronal marker tubulin beta 3 class III (TUBB3) (Ferreira and Caceres, 1992) and the specific biochemical marker of glutamatergic neurons and glutamatergic synapses vesicular glutamate transporter 1 (VGLUT1) (El Mestikawy et al., 2011) were detected in the iPSC-derived neurons ( Figure S4B ), suggesting that most of the iPSC-derived neurons differentiated into glutamatergic neurons. To confirm the effect of IFN-γ stimulation on T. gondii stage conversion, we examined the mRNA expression of the glutamine transporters SLC38A1 and SLC38A2 in iPSC-derived glutamatergic neurons ( Figure 6A ). The expression levels of SLC38A1 and SLC38A2 were upregulated by IFN-γ stimulation; however, they were inhibited by T. gondii infection in iPSC-derived glutamatergic neurons ( Figure 6A ). We next examined the effect of parasite infection on intracellular glutamine concentration in iPSC-derived glutamatergic neurons ( Figure 6B ). The intracellular glutamine concentration in IFN-γ-stimulated, T. gondii-infected iPSC-derived glutamatergic neurons was significantly reduced compared with that in IFN-γ-stimulated, non-infected iPSC-derived glutamatergic neurons ( Figure 6B ). Then, we examined the effect of CB-839 on the intracellular glutamine concentration and BAG1 and SAG1 gene expression in iPSC-derived glutamatergic neurons ( Figures 6C, D ). We found that CB-839 treatment restored the intracellular glutamine concentration in IFN-γ-stimulated, T. gondii-infected iPSC-derived glutamatergic neurons ( Figure 6C ). In addition, downregulation of BAG1 or upregulation of SAG1 gene expression were observed in response to CB-839 treatment compared to non-treatment in IFN-γ-stimulated, T. gondii-infected iPSC-derived glutamatergic neurons ( Figure 6D ). Finally, we examined the effect of CB-839 on T. gondii cyst wall CST1 formation in iPSC-derived glutamatergic neurons ( Figures 6E, F ). We found that CB-839 treatment inhibited IFN-γ-induced cyst wall CST1 formation ( Figures 6E, F ). Furthermore, inhibition of IFN-γ-induced T. gondii stage conversion was also observed in glutaminase knockdown iPSC-derived glutamatergic neurons ( Figure S5 ). These results indicate that disruption of glutamine metabolism by both T. gondii infection and IFN-γ stimulation lead to glutamine starvation, triggering T. gondii stage conversion in human neuronal cells, including glutamatergic neurons ( Figure 7 ).