We did not use experimental animals in this study.

The N. caninum (Nc1 strain), NcGRA6-knockout, NcGRA14-knockout NcGRA7-knockout and NcGRA7-complemented parasites (34) were maintained in African green monkey kidney epithelial cells (Vero cells) cultured in Eagle’s minimum essential medium (EMEM; Sigma, St. Louis, MO, USA) supplemented with 8% heat-inactivated fetal bovine serum (FBS). NcGRA7-complemented parasites was generated by inserting NcGRA7 gene fused with FLAG tag under Toxoplasma GRA1 promoter into the Neospora uracil phosphoribosyl transferase (NcUPRT) gene of the NcGRA7-knockout line.

Human monocytic leukemia cell line, THP-1 cells, were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium (Sigma) supplemented with 10% heat-inactivated FBS. 293T cells and HFFs were cultured in Dulbecco’s modified Eagle’s medium (Sigma) supplemented with 10% heat-inactivated FBS.

For the purification of tachyzoites, the parasite cultures were washed with each medium, scraped, suspended in the medium and passed through a 27-gauge needle and a 5.0 μm pore filter (Millipore, Bedford, MA, USA). After centrifugation, the tachyzoites were resuspended in the medium as purified parasites.

MCC950 (an inhibitor of NLRP3 inflammasome activation by the inhibition of IL-1β release; Cayman Chemical), VX765 (an inhibitor of NLRP3 inflammasome activation by the inhibition of caspase 1 and IL-1β cleavage and release; LKT Laboratories), SN50 (an inhibitor of the nuclear transcription factor κB; Selleck, Houston, TX, USA), a carbonyl cyanide m-chlorophenyl hydrazone (CCCP, a uncoupling agent for oxidative phosphorylation that inhibits mitochondrial function; FUJFILM Wako Pure Chemical Corporation), rocaglamid A (Cayman, Ann Arbor, MI, USA) with biochemical research grades were used in this study.

The p3XFLAG-CMV-14 plasmid (Sigma–Aldrich) encoding the C-terminal FLAG-tagged NcGRA7 (p3XFLAG-NcGRA7) was described previously (34). Plasmids encoding NLRP3, ASC, procaspase-1, and pro-IL-1β were kindly gifted from Dr. Takeshi Ichinohe (The University of Tokyo) (35).

The culture supernatants were collected to measure the IL-1β and TNF-α levels via ELISA kits (Human OptEIA ELISA Set, BD Biosciences, San Jose, CA, USA) according to the manufacturer’s instructions. The cytokine concentrations were calculated from curves generated from cytokine standards analyzed on the same plates.

JC-1 is widely used for observing mitochondrial membrane potential and shows changes in fluorescence characteristics from green (530 nm) to red (590 nm) depending on the mitochondrial membrane potential. For the JC-1 assay, cells were stained using a JC-1 MitoMP detection kit (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer’s instructions.

Annexin V staining precedes the loss of membrane integrity, which accompanies the latest stages of cell death resulting from either apoptotic or necrotic processes. Used in conjunction with a vital dye such as 7-Amino-Actinomycin (7-AAD), cells that are considered viable are PE Annexin V and 7-AAD negative; cells that are in early apoptosis are PE Annexin V positive and 7-AAD negative; and cells that are in late apoptosis or already dead are both PE Annexin V and 7-AAD positive. For measurement of early apoptosis, cells were stained using a PE Annexin V Apoptosis Detection Kit I (BD Pharmingen, San Diego, CA, USA) according to the manufacturer’s instructions.

For LC–MS/MS analysis using immunoprecipitates, 293T cells at 80% confluence were transiently transfected with the p3XFLAG-NcGRA7 or empty plasmid (mock) (12 μg each) in a 10 cm culture dish using FuGENE HD (Promega). The next day, the cells were lysed with 1 mL of lysis buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% NP-40, and protease inhibitor cocktail (Complete Mini, Roche), and the clarified supernatants were incubated for 16 h at 4°C with 20 μL of ANTI-FLAG M2 Affinity Gel (Sigma). After washing 3 times with IP buffer containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl and 0.5% NP-40, the immunoprecipitates were resolved via 10% SDS–PAGE.

The specific bands observed in the immunoprecipitated NcGRA7-FLAG were subjected to liquid chromatography-tandem mass spectrometry (LC–MS/MS) analysis. For protein identification, the eluted proteins were separated via SDS–PAGE and stained with silverstain, after which the gel regions containing the proteins of interest were excised. Proteins in the excised gel blocks were destained, reduced, alkylated at the cysteine residues, and digested with modified trypsin. Trypsin-digested peptides were extracted from the gel blocks and subjected to mass spectrometric analysis. LC–MS/MS was performed on a Paradigm MS2 HPLC (Michrom BioResources) and an LTQ linear ion trap mass spectrometer with a Nanospray ion source housing (Thermo Fisher Scientific). Peptides were separated through an L-column2 C18 (0.1 × 50 mm, 3 µm particle size, CERI, Tokyo) using a linear gradient (0.2% acetonitrile/B. 90% acetonitrile in 0.1% formic acid, 5–40% acetonitrile 2%/min) at a flow rate of 0.5 µl/min. Mass spectrometric data were searched against the IPI_human (91,464 sequences; 36,355,611 residues) database using MASCOT software ver. 2.8 (Matrix Science). The MASCOT score (target FDR/override significance threshold, 1%), sequence coverage, total number of mass-matched peptides, accession number of MS, IPI_human-database, and emPAI are shown. The emPAI (exponentially modified protein abundance index), which offers approximate and relative quantification of proteins in a mixture, was calculated on two parameters, the number of experimentally observed peptides, as described by Ishihama et al. (2005) (36).

Anti-FLAG M2 mouse monoclonal antibody (Sigma–Aldrich), anti-XPOT (Exportin-T) rabbit antibody (A303-972A, Bethyl, Montgomery, TX, USA), anti-XPO1(CRM1) rabbit antibody (A300-469A, Bethyl), anti-SLC25A13 rabbit antibody (GTX109001, GeneTex, Irvine, CA, USA), anti-DNAJA1 rabbit antibody (A304-516A, Bethyl), anti-UQCRC2 rabbit antibody (GTX114873, GeneTex), anti-PHB1 rabbit antibody (GTX101105, GeneTex), anti-CoxIV rabbit monoclonal antibody (3E11, Cell Signaling Technology, Danvers, MA, USA), anti-GAPDH rabbit monoclonal antibody (14C10, Cell Signaling Technology), anti-Mfn1 rabbit antibody (13798-1-AP, Proteintech, Rosemont, IL, USA), anti-AIF mouse monoclonal antibody (sc-13116, Santa Cruz Biotechnology, Dallas, TX, USA), anti-mtHSP70 mouse monoclonal antibody (JG1, Invitrogen, Waltham, MA, USA), anti-VDAC1 rabbit antibody (4866, Cell Signaling Technology), anti-PHB2 mouse monoclonal antibody (sc-133094, Santa Cruz Biotechnology), anti-Histone H3 rabbit antibody (9715S, Cell Signaling Technology), anti-phospho-NF-κB p65 rabbit monoclonal antibody (93H1, Cell Signaling Technology) and anti-NFκB p65 rabbit polyclonal antibody (sc-109, Santa Cruz Biotechnology) were used for western blot and indirect fluorescent antibody test (IFAT) in this study. To detect N. caninum proteins, an anti-NcSRS2 mouse monoclonal antibody (clone 1B8) (37), anti-NcGRA7 mouse serum and anti-NcGRA7 rabbit antibody (34) were used.

HFFs at 80% confluence were infected with 5×10⁶ purified parasites (Nc1- and NcGRA7-complemented parasites) in a 10 cm culture dish. After 40 hr, the cells were lysed with 1 mL of lysis buffer, and the clarified supernatants were incubated for 16 h at 4°C with 20 μL of ANTI-FLAG M2 Affinity Gel (Sigma) followed by washing 3 times with IP buffer. The immunoprecipitates of the plasmid-transfected cells and the parasite-infected cells were resolved by 10-12% SDS–PAGE. and immunoblotted with antibodies against FLAG-tagged protein or the indicated proteins.

For mitochondrial staining, the cells were incubated with 200 nM MitoTracker^® Red CMXRos (Thermo Fisher Scientific, Waltham, MA, USA) for 30 min before fixation. Coverslips with cells were collected at the indicated time points, washed twice with PBS and then fixed with 2% paraformaldehyde in PBS. After washing twice with PBS, the cells were permeabilized with 0.3% Triton X-100 in PBS for 5 min at room temperature. After washing, the coverslips were incubated with 3% bovine serum albumin (BSA) in PBS at room temperature for 30 min, after which the sections were subjected to antibody staining and nuclear staining with Hoechst 33342 (Sigma–Aldrich). Observation was performed with a Leica Microsystems THUNDER Imaging System.

Vero cells infected with N. caninum (Nc1 strain) at 48 h postinfection were fixed with 4% paraformaldehyde (FUJIFILM Wako, Tokyo, Japan) for 1 h at room temperature and then washed with 0.1 M phosphate buffer three times. The samples were dehydrated through a series of alcohol series ranging from 70-100%, 100% alcohol was gradually changed with LR white resin (Nisshin EM, Tokyo, Japan), and the samples were incubated with 100% LR white resin overnight at 4°C. The samples were embedded in 100% LR white resin in a gelatin capsule (Nisshin EM) and incubated for 24 h at 50°C. Ultrathin sections were collected on carbon-coated nickel grids (ADD Deplen, Tokyo, Japan). After incubating with blocking buffer, 4% BSA and 0.1% sodium azide in PBS, the grids were incubated with dilutions of anti-NcGRA7 rabbit antibody or normal rabbit serum for 2 h at room temperature. After washing with PBS seven times, the grids were incubated with dilutions of gold (10 nm)-conjugated anti-rabbit IgG goat antibody (CytodiagnoSitc, Burlington, Canada) for 3 h at room temperature. After washing with PBS six times, the sections were fixed with 1% osmium tetroxide and stained with uranyl acetate and lead citrate. Finally, the sections were observed with a HITACHI HT7700 transmission electron microscope (Hitachi High-Tech Corporation, Tokyo, Japan).

Mitochondrial fractionation was performed using mitochondria/cytosol fractionation kit as per manufacturer’s instructions (Enzo Life Sciences, NY, USA). The fractions of NP, Mt and Cyto were resolved in 100 μl of SDS–PAGE sample buffer (62.5 mM Tris-HCl (pH 6.8), 0.017% SDS, 5% glycerol, 5% 2-mercaptoethanol, and 0.005% bromophenol blue) for NP and Mt and 100 μl of SDS–PAGE sample buffer supplemented with 50 mM Tris-acetate (pH 7.5), 8 M urea, and 0.1% SDS for Cyto and subjected to SDS–PAGE followed by western blotting with the indicated antibodies.

Proteolysis assays using proteinase K were performed as previously described (38) with slight modifications. Briefly, cells in 10 cm cell culture dishes were washed twice with HBSS (Gibco), scraped off the culture plate, and lysed in 1 mL of homogenization buffer (20 mM HEPES [pH 7.5], 70 mM sucrose, and 220 mM mannitol) by 30 strokes in a Dounce homogenizer. The homogenate was centrifuged at 1,000×g for 5 min to precipitate the nuclei, and the resulting supernatant (300 μl each, 3 tubes) was further centrifuged at 14,000×g for 10 min (4°C) to precipitate the crude mitochondrial fraction. For the proteinase K resistance assay, the isolated mitochondrial pellet was resuspended in 200 μl of buffer, homogenization buffer, 100 μg/mL proteinase K in homogenization buffer or 100 μg/mL proteinase K in hypotonic swelling buffer (20 mM HEPES-KOH, pH 7.5). The samples were incubated for 15 min (on ice) and further centrifuged at 14,000×g for 15 min (4°C) to precipitate the crude mitochondrial fraction. The precipitates were resolved in 60 μl of 5 mM PMSF in SDS–PAGE sample buffer and subjected to western blotting with the indicated antibodies.

For RNA interference knockdown experiments, Silencer™ Select Negative Control No. 1 siRNA (Thermo Fisher Scientific), PHB1 Sense siRNA sequence #2 (CGUGGGUACAGAAACCAAUtt) (Thermo Fisher Scientific) were used as negative controls and PHB1 knockdown, respectively. THP-1 cells were transfected with 30 pmol of siRNA using the SG Cell Line 4D-Nucleofector™ X Kit L (catalog #: V4XC-3024; Lonza, Basel, Switzerland) following the manufacturer’s protocols. At 20 h after transfection, the siRNA-treated cells were infected with N. caninum and subjected to functional assays. The protein levels after siRNA treatment were analyzed via western blotting. Band intensity was quantified using ImageJ software developed by the US National Institutes of Health.

The data are expressed as the mean ± SD of 3–4 replicates in one representative experiment (each experiment was repeated two to four times). The various assay conditions were evaluated with analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test. A P value (* P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001) was considered to indicate statistical significance.

Our previous study showed that N. caninum infection induces the production of inflammatory cytokines from mouse macrophages and that NcGRA6, NcGRA7, and NcGRA14 are involved in the activation of NF-κB signaling, calcium/calcineurin (NFAT) signaling, and cAMP/PKA (CRE) signaling (34). Therefore, we compared the effects of NcGRA6, NcGRA7 and NcGRA14 on the IL-1β production of THP-1 cells. As shown in Figure 1A, the level of IL-1β production induced by NcGRA7KO infection was significantly lower than that induced by infection with the parental strain Nc1 or other knockout lines. To confirm the effect of NcGRA7 on NLRP3 inflammasome activation, the NLRP3 inflammasome was reconstructed in Human embryonic kidney cells (293T cells) by transfection with plasmids encoding NLRP3, ASC, procaspase-1, and pro-IL-1β (35). Transfection with plasmids for reconstitution of the NLRP3 inflammasome stimulated IL-1β production in 293T cells, and additional transfection with NcGRA7 cDNA significantly enhanced IL-1β production (Figure 1B).

IL-1β production is regulated by two signaling pathways, the activation inflammasome and the NF-κB pathway. To evaluate the mechanism through which N. caninum infection regulates IL-1β, we tested MCC950 (an NLRP3 inhibitor), VX765 (a CASP1 inhibitor), and SN-50 (an NF-κB inhibitor) in THP-1 cells following infection with parental strain Nc1 and NcGRA7KO (Figures 1C–H). All inhibitors significantly suppressed IL-1β production induced by Nc-1 infection, whereas no inhibitory effect was observed in NcGRA7KO-infected cells (Figures 1C, E, G). Moreover, TNF-α production in NcGRA7KO-infected cells was markedly lower than that in Nc-1-infected cells (Figures 1D, F, H). Furthermore, treatment with any of the inhibitors led to a reduction in TNF-α production in both Nc-1- and NcGRA7KO-infected cells (Figures 1D, F, H). Western blot analysis of phospho–NF-κB p65 revealed that N. caninum infection induced phosphorylation of NF-κB p65; however, no significant difference in phosphorylation levels was observed between Nc-1-infected and NcGRA7KO-infected cells (Supplementary Figure S1). These findings suggest that NcGRA7 does not directly regulate NF-κB activation, but rather positively influences the production of IL-1β and TNF-α through post-transcriptional mechanisms, thereby leading to the upregulation of NLRP3.

Recent studies have shown the pivotal roles of mitochondria in the initiation and regulation of the NLRP3 inflammasome, suggesting mitochondrial destabilization or damage. Therefore, we measured mitochondrial membrane potential (MtMP), a key indicator of mitochondrial activity, in N. caninum-infected THP-1 cells (Figure 2). N. caninum infection decreased MtMP levels, as shown by the control CCCP treatment which inhibits mitochondrial function by uncoupling for oxidative phosphorylation. Accordingly, we could observe the MtMP of NcGRA7KO-infected cells was significantly greater than that of parental Nc1 strain-infected cells at both a multiplicity of infection (MOI) of 2.5 and a MOI of 5.0.

Because mitochondrial damage induces apoptosis, we examined whether N. caninum infection induced the apoptosis of THP-1 cells (Figure 3). Compared with mock-infected THP-1 cells, N. caninum infection induced apoptosis in THP-1 cells at both a MOI of 2.5 and a MOI of 5.0. As expected, NcGRA7KO decreased host cell apoptosis.

To clarify the mechanism of IL-1β production induced by NcGRA7, we aimed to identify NcGRA7-binding proteins by an anti-FLAG immunoprecipitation assay using 293T cells transfected with NcGRA7 cDNA fused with a FLAG tag (Figure 4A). Five or more proteins were detected, while the other three bands (#1, #2 and #3) were excluded from further analyses because similar bands were observed in mock samples (#1 and #2), and #3 may be glycoproteins because a typical ladder pattern seen in proteins with added sugar chains (Supplementary Figure S2). The proteins in the excised gel band from the NcGRA7 IP eluate compared to those in the mock control IP eluate were subjected to mass spectrometric analysis. The proteins possibly bound to NcGRA7 are listed in Supplementary Table S1. Interestingly, a score>90 in the Mascot search suggested that NcGRA7-FLAG might be able to sort mitochondria or cytosol in host cells because proteins related to these organelles were detected via MS analysis.

Therefore, the fractions eluted from the anti-FLAG immunoprecipitate were analyzed by immunoblotting using antibodies against SLC25A13, DNAJA1, UQCRC2, PHB1 and PHB2, which are related to host mitochondria, and XPOT and XPO1, which localize to the nuclear envelope. As shown in Figures 4B, C, specific bands were observed in the NcGRA7 IP eluate prepared from 293T cells transfected with NcGRA7 cDNA but not in the mock control IP eluate. To confirm the physiological interaction of NcGRA7 with parasite-derived NcGRA7, eluted fractions from human foreskin fibroblasts (HFFs) infected with the wild-type Nc1 strain or the NcGRA7-complemented parasite, which expressed the NcGRA7 protein fused with a FLAG tag, were also analyzed. Among the tested antibodies, specific bands against XPO1, UQCRC2, PHB1 and PHB2 were detected (Figure 4C; Supplementary Figure S3). Then, we analyzed the colocalization of NcGRA7 with XPO1, UQCRC2 and PHB1 by IFAT (Supplementary Figure S4). Specific colocalization of NcGRA7 with XPO1 and UQCRC2 was not observed. On the other hand, NcGRA7 secreted from the PV was observed near host mitochondria, as determined by Cox IV, PHB1 (Figure 5) and MitoTracker (Supplementary Figure S4D) staining.

Immunoelectron microscopic analysis of intracellular parasites revealed that NcGRA7 was localized in dense granules and in the parasitophorous vacuolar space (Figure 5C). In the parasitophorous vacuolar space, NcGRA7 was diffusely distributed. In addition, NcGRA7 was also distributed in the host mitochondria and host cytoplasm. NcGRA7 was detected particularly in the cristae and outer membrane of mitochondria (high-magnification image of Figure 5C). NcGRA7 was also distributed within 50–100 nm structures, which exhibited moderate electron density, the host cytoplasm and the parasitophorous vacuolar space (high-magnification image of Figure 5C).

To confirm whether NcGRA7 affects mitochondrial morphology, we compared localization of host mitochondria marker Cox IV and PHB1 in HFFs infected with the wild-type Nc1 strain and NcGRA7KO parasites (Supplementary Figure S5). However, no significant morphological alterations were observed among the cells.

To further confirm the localization of NcGRA7 in the cytosol and near mitochondria of host cells, we fractionated host cell extracts from N. caninum-infected cells and examined the mitochondrial and cytosolic localization of NcGRA7 using a biochemical approach (Figure 6). Nuclear protein Histone H3 and inner mitochondrial membrane (IMM) protein Cox IV were set as controls. As expected, endogenous NcGRA7 was detected in the fraction of mitochondria harboring PHB1 and in the cytosol of the host cells detected by GAPDH, which was distinct from the control of the parasite cell surface protein NcSRS2 (Figure 6B; Supplementary Figure S6).

Next, to investigate the submitochondrial localization of NcGRA7, mitochondria isolated from N. caninum-infected cells were treated with proteinase K under isotonic conditions (Figures 7A, B; Supplementary Figure S7). NcGRA7 was partially digested by the protease under isotonic or swelling conditions, as observed for the control proteins for intermembrane space (IMS) or IMM, such as PHB1, COX IV and MtHSP70, indicating the distribution of NcGRA7 in the IMM to the matrix of host mitochondria. The interaction of NcGRA7 with other mitochondrial proteins was also examined by performing an anti-FLAG IP on eluted fractions from HFFs infected with the wild-type Nc1 strain or NcGRA7-complemented parasite (Figure 7C; Supplementary Figure S8). Immunoblotting using antibodies against VDAC1 (outer mitochondrial membrane (OMM) protein), AIF (IMS protein), Cox IV (IMM protein) and mtHsp70 (matrix protein) showed no binding with NcGRA7 and revealed the interaction of NcGRA7-FLAG with PHB1 and PHB2 (Figure 4C). PHBs form ring-like PHB complexes comprising multiple PHB1 and PHB2 subunits to maintain mitochondrial function (39). Together with the results shown in Figure 4C, these findings indicate that NcGRA7 may form a complex with PHB1 and PHB2.

To examine the effects of PHB1 on IL-1β production by N. caninum-infected THP-1 cells, we tested the effects of PHB1 inhibitor (Figure 8A). Rocaglamide A, derived from medicinal plants, have been demonstrated to interact directly with PHB1 and thus inhibit the interaction of PHB1 with Raf-1, impeding Raf-1/ERK signaling cascades and significantly suppressing cancer cell metastasis (40). IL-1β production was significantly inhibited in N. caninum-infected THP-1 cells by treatment with Rocaglamide A while TNF-α production was increased (Figure 8A).

To further confirm the effects of PHB1 on IL-1β production following infection of THP-1 cells with N. caninum, siRNA-mediated knockdown of PHB1 was carried out (Figure 8B). Transfection of mock-infected cells with PHB1 siRNA decreased PHB1 protein level by 31.4% compared to the transfection of mock-infected cells with control siRNA. Additionally, transfection with PHB1 siRNA reduced PHB1 protein level by 30.0% compared with transfection with control siRNA in Nc-1-infected cells (Supplementary Figure S9). As shown in Figure 8C, the transfection of N. caninum-infected THP-1 cells with PHB1 siRNA significantly decreased IL-1β production but not TNF-α production, indicating interaction of NcGRA7 and PHB1 might have a role on NLRP3 inflammasome pathway.