Toxoplasma GRA15 Activates the NF-κB Pathway through Interactions with TNF Receptor-Associated Factors

The parasite Toxoplasma can cause birth defects and severe disease in immunosuppressed patients. Strain differences in pathogenicity exist, and these differences are due to polymorphic effector proteins that Toxoplasma secretes into the host cell to coopt host cell functions. The effector protein GRA15 of some Toxoplasma strains activates the nuclear factor kappa B (NF-κB) pathway, which plays an important role in cell death, innate immunity, and inflammation. We show that GRA15 interacts with TNF receptor-associated factors (TRAFs), which are adaptor proteins functioning upstream of the NF-κB transcription factor. Deletion of TRAF-binding sites in GRA15 greatly reduces its ability to activate the NF-κB pathway, and TRAF2 knockout cells have impaired GRA15-mediated NF-κB activation. Thus, we determined the mechanism for GRA15-dependent NF-κB activation.


FIG 1
Ectopic expression of either GRA15 II or GRA15 III is sufficient to activate NF-B. (A) Schematic showing RH (GRA15), GRA15 II , and GRA15 III protein sequence and alignment. Red boxes indicate two predicted transmembrane domains. The blue box indicates the specific region of GRA15 III . (B) Structural disorder predictor for GRA15 III using the software Protein DisOrder prediction System (PrDOS), with a double asterisk indicating increased structural order across residues 50 to 72. Two blue dashed lines indicate the region specific for GRA15 III . (C) TREX-293 cells overexpressing GRA15 II and RA15 III were induced with tetracycline (1 g/ml) for 24 h, fixed with formaldehyde, and stained for p65 (red), HA (green), and Hoechst (blue). The scale bar represents 10 m. Arrows indicate nuclei (Continued on next page) Toxoplasma GRA15-TRAF Interactions Activate NF-B ® function as a nonclassical signal peptide. We previously showed that type II strains activate NF-B more strongly than type III strains (14). We have also shown that other Toxoplasma-secreted effectors such as ROP16 and ROP38 can inhibit the NF-B signaling pathway and that GRA15 expression levels also seem to affect NF-B activation (19). It is therefore unclear if different GRA15 sequences differ in their capacity to activate NF-B. We previously showed that GRA15 II is sufficient for activating the NF-B pathway by transfecting HeLa cells with a plasmid capable of expressing a GRA15 II -GFP (green fluorescent protein) fusion protein (14). However, we were unable to generate stable expression cell lines, likely because the expression of GRA15 II was toxic to the cells (not shown). Therefore, to determine if GRA15 II and GRA15 III can both activate the NF-B pathway, we engineered stable HEK293-derived (TREX-293) cell lines expressing GRA15 II (TGME49_275470, from residues 51 to 550 to exclude the putative signal peptide) or GRA15 III (from residues 51 to 631 to exclude the putative signal peptide) with a C-terminal hemagglutinin (HA)-FLAG double-epitope tag under the control of the tetracycline operator. In TREX-293 cells expressing GRA15 II or GRA15 III , expression of these proteins was strictly regulated by tetracycline. GRA15 II and GRA15 III seemed to have a predominantly cytoplasmic localization (Fig. 1C), although no further attempts were made to determine the exact subcellular localization of these proteins. p65 nuclear translocation indicated that upon induction by tetracycline, both GRA15 II and GRA15 III were able to activate the NF-B pathway ( Fig. 1C; see Fig. S1A in the supplemental material). We noted that under tetracycline induction, the observed molecular weights of GRA15 II and GRA15 III (ϳ75 and ϳ90 kDa, respectively), when ectopically expressed, were higher than the expected predicted sizes (57 and 66 kDa, respectively [https://www.expasy.org]) (Fig. 1D). We also immunoblotted for the endogenous GRA15 in human foreskin fibroblasts (HFFs) infected with strains expressing either a type II GRA15 (GRA15 II ) or a type I/III GRA15 (GRA15 I/III ) and observed a similar increase in GRA15 size compared to the expected GRA15 size (Fig. 1E). Thus, the larger than the expected size of GRA15 expressed by Toxoplasma is not caused by parasitemediated modification(s) of GRA15. Most likely, it is the particular amino acid composition of GRA15, which is enriched in Pro, Ser, and Thr, that makes it run slower than expected on an SDS-PAGE gel. To determine more quantitatively if there were any differences in the capacity to activate the NF-B pathway between these two proteins, we transiently transfected HEK293 NF-B reporter cells (luciferase/GFP) with both GRA15 plasmids. After quantification of the luciferase activity, both proteins activated the NF-B reporter significantly compared to the empty vector. We did not observe a significant difference between GRA15 II and GRA15 III in NF-B activation (Fig. 1F). The amounts of protein in the transient transfections were similar (Fig. S1B). These results indicate that both GRA15 II and GRA15 III can activate NF-B.
Identification of the GRA15 sequence necessary for NF-B activation. BLAST searches with the GRA15 protein sequence did not detect homology to any known proteins or any known protein domains. Therefore, to determine which region of GRA15 is required for GRA15-dependent NF-B activation, we engineered GRA15 Nand C-terminal truncation mutants. We cloned the full-length GRA15 II-51-550 without the putative signal peptide (schematic GRA15 II [ Fig. 2A]), C-terminally truncated GRA15 variants (51-338, 51-479, 51-517, and 51-527), and N-terminally truncated GRA15 variants (80 -550, 170 -550, and 270 -550) fused to GFP in a mammalian expression vector. We transiently transfected both full-length and truncated GRA15 mutants in an HEK293 NF-B reporter cell line and measured NF-B activation. Based on the detection  TREX-293 GRA15 II -and TREX-293 GRA15 III -overexpressing cell lysates induced with tetracycline for 24 h, using an antibody against the HA tag. GAPDH antibody was used as a loading control. (E) Immunoblot on extracellular parasite lysates of type II and type III strains using an antibody against the C terminus of endogenous GRA15, with SAG1 antibody as a loading control. (F) NF-B reporter HEK293 cells transfected with pcDNA GRA15 II and GRA15 III or empty vector. The graph shows average luciferase activity from the cell lysate from three independent experiments (n ϭ 3), and error bars represent SD. The asterisk indicates significantly higher levels of luciferase activity compared to the empty vector. P Ͻ 0.0012 for GRA15 II and P Ͻ 0.0006 for GRA15 III by one-way ANOVA with Dunnett's multiple-comparison test.
of GFP expression, all constructs were successfully expressed and seemed to localize to the host cytoplasm (not shown). The two largest GRA15 C-terminal truncations showed less activation of NF-B, but this only reached significance for the GRA15 51-338 construct. The N-terminal truncation mutants had significantly decreased NF-B activation compared to full-length GRA15 II-51-550 (Fig. 2B). The N-terminal mutant with the smallest region truncated is GRA15 II-80 -550 , indicating that residues 51 to 79 are necessary for GRA15-dependent NF-B activation. To ensure that the decrease in NF-B activation was not due to lack of protein expression, we immunoblotted for GRA15 upon transient transfection of HEK293 cells and observed that the N-terminal GRA15 mutants were more strongly expressed than the full-length GRA15 II-51-550 (Fig. 2C). We were not able to detect the two largest C-terminal GRA15 mutants (GRA15 51-479 and GRA15 51-338 ) on Western blots because the GRA15 antibody was raised against residues 493 to 510. Thus, the decrease in NF-B activation is specific to the missing sequences in the N-terminal GRA15 sequences and not because of protein expression differences. We observed that the second predicted transmembrane (amino acids 51 to 72) (Fig. 1B) seems to be important for GRA15-mediated activation of NF-B.
Identification of GRA15 candidate host-interacting proteins. Even though there is evidence that GRA15 requires TRAF6 to achieve full NF-B activation (14), it is unclear Toxoplasma GRA15-TRAF Interactions Activate NF-B ® whether GRA15 directly interacts with TRAF6 or acts indirectly through TRAF6 to cause NF-B activation. Therefore, to determine potential host proteins that interact with GRA15 or are in the same complex with GRA15, we harvested whole-cell lysates from TREX-293 GRA15 II 16 h postinduction with tetracycline. As a negative control, we used TREX-293 expressing ROP38 I . These lysates were then subjected to immunoprecipitation using the HA antibody, and immunoprecipitates were probed with HA antibody to confirm the presence of each protein in the immunoprecipitates (Fig. 3A). The immunoprecipitates were directly sent for mass spectrometry analysis. To control for potential spurious interactions detected through mass spectrometry, we refined the list of proteins to those identified only in GRA15 II -overexpressing conditions and excluded proteins detected in immunoprecipitates from ROP38 I -overexpressing conditions. We identified a list of candidate host proteins that could have direct interactions with GRA15 II (Table 1). In the TREX-293 GRA15 II -expressing cells, we observed that TRAF2,  TRAF3, and TRAF6 were detected in GRA15 II immunoprecipitates but were absent in ROP38 I immunoprecipitates (Table 1). We also observed the baculoviral IAP repeatcontaining protein (BIRC2), an E3 ubiquitin-protein ligase know to interact specifically with TRAF2 and for its antiapoptotic function (Table 1). To validate these observations, we immunoblotted the immunoprecipitated GRA15 II and ROP38 I with antibodies against TRAF2, TRAF3, and TRAF6. We observed a faint band for TRAF3 and a strong signal for TRAF2 and TRAF6 only in the immunoprecipitated GRA15 II (Fig. 3B). To confirm these interactions in cells infected with parasites, we immunoprecipitated GRA15 from HEK-293 cells infected with RH parasites expressing HA-tagged GRA15 II . We immunoblotted the immunoprecipitated GRA15 II with antibodies against TRAF2 and TRAF6 and observed a strong signal for these TRAF proteins (Fig. 3C) Fig. 2A) (30). The mass spectrometry results also identified two ubiquitination sites on the GRA15 II sequence: lysine 126 and lysine 172 (see Table S1 in the supplemental material). Therefore, to determine if these putative TRAF2-binding sites and/or the ubiquitination sites are required for GRA15-dependent NF-B activation, we used Q5 mutagenesis to mutate these sites in the GRA15 II full-length sequence. We cloned the full-length GRA15 II-51-550 without the putative signal peptide (schematic of GRA15 II in Fig. 2A), the deletions AAEE 160 -163 , SQQE 430 -433 , and AAEE 160 -163 /SQQE 430 -433 , and the substitutions K/A 126 and K/A 172 in a mammalian expression vector. We transiently transfected the full-length and GRA15 mutants in the HEK293 NF-B luciferase reporter cell line to determine differences in NF-B activation. We did not observe a significant difference in NF-B activation for the ubiquitination mutants compared to the fulllength GRA15 II-51-550 sequence. However, we observed that the single TRAF2-site mutants had ϳ60% decrease in NF-B activation compared to full-length GRA15 and the double TRAF2-site mutant had an ϳ75% decrease in NF-B activation (Fig. 4A). To ensure that the decrease in NF-B activation was not due to lack of protein expression, Toxoplasma GRA15-TRAF Interactions Activate NF-B ® we immunoblotted for GRA15 upon transient transfection of HEK293 cells and observed that the TRAF2-site GRA15 mutants were similarly expressed to the full-length GRA15 (Fig. 4B). Thus, the decrease in NF-B activation is specific to the missing sequences AAEE 160 -163 and SQQE 430 -434 in GRA15 sequences and is not because of protein expression differences. These data are consistent with GRA15 activating NF-B through recruitment of TRAF proteins via its TRAF-binding domains.
TRAF2 is partially needed for GRA15-mediated NF-B activation. Immunoprecipitation performed on TREX-293 cells overexpressing GRA15 II demonstrated that TRAF2 is a binding partner of GRA15. To elucidate the cross talk between GRA15 and TRAF2, we performed additional experiments, utilizing an HEK293 NF-B-green fluorescent protein (GFP) reporter cell line. We used CRISPR/Cas9 to generate an indel in TRAF2 (see Fig. S2 in the supplemental material) in this reporter cell line and isolated a clonal line that no longer expressed TRAF2 (Fig. 5A). Both the wild-type and the TRAF2 knockout reporter lines (Δtraf2) were infected with a type II strain (Pru), a type II Δgra15 strain, and the type I RH strain, which does not express GRA15. (Note that none of these strains expresses GFP.) To measure the activation of the NF-B pathway, we performed a quantitative analysis of GFP expression. GFP levels were undetectable in noninfected and nonstimulated cells. There was a significant (ϳ47%) decrease in expression of GFP between the Δtraf2 and the wild-type cells when they were infected with type II (Fig. 5B). Infection with the type II Δgra15 or the RH strain did not lead to any GFP expression. As shown by others (31), Δtraf2 cells were susceptible to TNF-␣, and all cells died upon addition of TNF-␣ (not shown), and therefore TNF-␣ did not induce NF-B activation in these cells (Fig. 5B). These results show that the activation of the NF-B pathway by GRA15 is partially dependent on TRAF2.

DISCUSSION
We observed that upon ectopic expression of either type II or type III GRA15, there was significant NF-B activation compared to noninduced controls. In contrast, upon infection of human or murine cells, type II strains strongly activate NF-B, while type III strains induce very weak or no NF-B activation (14). Thus, it is likely that the low expression level of GRA15 in type III strains and the presence of type III strain effectors that have an inhibitory effect, such as ROP38, which is highly expressed in type III strains and inhibits NF-B and ROP16, which can also inhibit NF-B (19), explain the absence of NF-B activation by type III strains.
We identified several host proteins that may directly interact with GRA15 or are present in a complex with GRA15. The most promising interaction partners were TRAFs, which coimmunoprecipitated with GRA15 II ectopically expressed in TREX cells but were absent in all other immunoprecipitations. TRAF2 and TRAF6 are both involved in the canonical NF-B activation pathway (32), whereas TRAF2 and TRAF3 are involved in nonredundant negative regulatory roles in the alternative NF-B activation pathway (33,34). Upon exposure to classical activating stimuli such as TNF-␣, TRAF2 becomes polyubiquitinated and activates a downstream kinase, RIPK, to cause activation of the IKK complex (35). Other canonical activating stimuli, such as lipopolysaccharide (LPS), cause oligomerization of TRAF6 and activation of its E3 ligase activity, which leads to recruitment and activation of TAK1, a downstream kinase. TAK1 then binds to the regulatory subunit of the IKK complex to activate IKK␣/␤, causing phosphorylation of IB␣ and allowing nuclear translocation of p65/p50 heterodimers to occur. We observed a faint TRAF3 band in the GRA15 immunoprecipitate in TREX cells, suggesting that TRAF3 is not a direct interactor of GRA15. TRAF3 was demonstrated to physically interact with the NF-B-inducing kinase (NIK) and mediate its ubiquitination and degradation (36). Thus, noncanonical NF-B activation is associated with TRAF3 degradation and concomitant accumulation of NIK (36). Another study demonstrated that NIK degradation is cross-regulated by a complex consisting of TRAF3/TRAF2/cIAP1/ cIAP2 (34). The GRA15 primary sequence contains two TRAF2 binding motifs (AAEE 160 -163 and SQQE 430 -433 ) and one TRAF6 binding motif (PGENSY 506 -511 ), and we show that the TRAF2 motifs play a role in GRA15-mediated NF-B activation. Likely, the remaining activation of NF-B we observed in the double TRAF2 GRA15 mutant is due to the remaining TRAF6 binding site. Taken together, GRA15 TRAF-binding sites are functional and likely complementary by recruiting both TRAF2 and TRAF6.
GRA15-mediated NF-B activation seems remarkably similar to what has been described as the mechanism of NF-B activation by the Epstein-Barr virus (EBV), which can cause Burkitt lymphoma. EBV encodes the effector LMP1 (latent membrane protein 1), which persistently activates NF-B by mimicking CD40 signaling (37). The CD40 cytoplasmic domain has one TRAF6-and two TRAF2-binding sites, which are important for its function (38). LMP1 is a six-membrane-spanning molecule that contains two C-terminal activation (CTAR) domains, CTAR1 and CTAR2, which activate the alternative and canonical NF-B pathways, respectively. CTAR1 interacts with TRAF2 and TRAF3 to activate the alternative NF-B pathway, and this alternative activation is dependent on the TRAF-binding site in CTAR1 (39). On the other hand, CTAR2 interacts with TRAF6, leading to activation of the canonical NF-B pathway, dependent on eight residues within CTAR2 (40). Constitutive activation of the NF-B pathway in B-cells, which are the preferred cell type infected by EBV, leads to the malignant transformation of these cells (41). Although we previously have seen no evidence of GRA15 activating the alternative NF-B pathway in human foreskin fibroblasts (14). It is possible that such activation could take place in different cell types. Interestingly, it has been shown that activation of CD40 signaling in both hematopoietic and nonhematopoietic cells infected with Toxoplasma leads to autophagy-mediated destruction of the parasitophorous vacuole (42). It will be of interest to determine if GRA15 signaling can also affect the autophagy pathway.

MATERIALS AND METHODS
Plasmids. The vector pcDNA-LIC-HF was a gift from M. A. Hakimi and A. Bougdour. Primers were designed to amplify after the predicted signal peptide to the predicted stop codon. Forward primers to amplify GRA15 (5=-TGGCTGGTGCTGGTGCCCATATAATTCGGTGGCTTGGGTATCTT-3=) together with reverse primers (5=-GCTCCGGCTCCTGCCCCAGCTGGAGTTACCGCTGATTGTGTG-3=) contained ligationindependent cloning (LIC) sequences (in boldface) and were used to amplify GRA15 from PRU (II) and CEP (III) genomic DNA. Forward primers to amplify ROP38 I (5=-TGGCTGGTGCTGGTGCCCATCATGGCAGCAG CACTGATGGATCAG-3=) together with reverse primers (5=-GCTCCGGCTCCTGCCCCAGCAAATTGATGCGT TCTTATCCGA-3=) contained LIC sequences (in boldface) and were used to amplify ROP38 from RH genomic DNA. PCR products were treated with T4 DNA polymerase (using only TTP at 100 mM). The pcDNA-LIC-HF vector was digested with SmaI and treated with T4 DNA polymerase (using only ATP at 100 mM) to generate long overhangs. The PCR fragment and vector were then annealed for 15 min at room temperature, generating expression vectors with Toxoplasma genes C-terminally tagged with HA-FLAG.
The vector pIC242 was a gift from I. Cheeseman (Whitehead Institute, Cambridge, MA). The GRA15 II full-length protein (51 to 550 aa) was amplified and inserted into pIC242 by restriction/ligation, expressing GRA15 II mutants as N-terminal fusion GFP proteins. Expression of GFP fusion proteins was promoted by the endogenous retroviral long terminal repeats. GRA15 truncation and mutation constructs were amplified from the pIC242 GRA15 II full-length protein (51 to 550 aa) using specific primers ( Table 2) and confirmed by sequencing.
Inducible TREX-293 cell line construction. The TREX-293 cell line was a gift from J. Niles (MIT, Cambridge, MA). TREX-293 cells were seeded at 75% confluence and cotransfected with expression vector pcDNA-LIC-GRA15 II -HF or ROP38 I -HF and a puromycin resistance vector (ratio of 10:1), using the X-tremeGENE 9 DNA transfection reagent (Roche). Cells were split 2 days posttransfection and subjected to puromycin (Calbiochem) selection at 1 g/ml. Foci were picked and expanded at least 1 week postselection, and positive foci were selected through HA expression using immunofluorescence and immunoblotting.
Transient transfections. NF-B reporter HEK293 cells (luciferase/GFP) (System Biosciences) were seeded at 7.5 ϫ 10 5 cells per well in a 6-well plate and incubated for 4 h at 37°C. Cells were subsequently transiently transfected with XtremeGene9 transfection reagent (Roche) with pcDNA-LIC-GRA15 II -HF, GRA15 III -HF, or pIC242-GRA15 II (full-length and mutants) and incubated for 24 h at 37°C. Cells were lysed in 100 l Promega Passive lysis buffer, and luciferase activity was measured in lysates according to the manufacturer's instructions. Data from conditions with comparable transfection efficiencies were used.
Coimmunoprecipitation. TREX-293 cells overexpressing GRA15 II or ROP38 I were grown in a T175 flask until 100% confluence and induced with tetracycline (1 g/ml) for 24 h. HEK-293 wild-type cells were grown in a T175 flask until 100% confluence and infected for 8 h with RH expressing GRA15 II . Cells , and the beads were resuspended in 100 or 40 l of this buffer. Western blotting. Ten percent input lysate and 20 l of the magnetic beads coupled with antibodies to HA of each sample were used to run SDS-PAGE. The proteins were transferred to a polyvinylidene difluoride (PVDF) membrane, blocked for 30 min with TBST (Tris-buffered saline with Tween 20)-5% nonfat dry milk. The membrane was blotted overnight at 4°C with rat antibody against HA (3F10, 1:500 dilution [Roche]) and TRAF2, TRAF3, TRAF6, and lamin A rabbit antibodies (sc-876, sc-1828, sc-7221, and sc-293162, respectively, 1:200 [Santa Cruz]), followed by respective secondary horseradish peroxidase (HRP)-conjugated antibodies. NF-B reporter HEK293 cells transfected with GRA15 constructs were lysed with lysis buffer, boiled for 5 min, and subjected SDS-PAGE. Proteins were transferred to a PVDF membrane, blocked for 30 min with TBST plus 5% nonfat dry milk, incubated with affinity-purified rabbit polyclonal antibodies to GRA15 (raised against the peptide with the amino acid sequence of GRA15 493-510 , 1 g/ml [YenZym Antibody, San Francisco, CA]) (19) or antibodies against mouse GAPDH (sc-32233, 1:500 [Santa Cruz]), overnight at 4°C, followed by the respective secondary HRP antibodies. TREX-293 GRA15 II or GRA15 III was induced or not with tetracycline (1 g/ml) for 24 h. The cells were lysed and subjected to SDS-PAGE. After transfer, the PVDF membrane was incubated with rat antibody against HA (Roche) and mouse GAPDH antibodies (Santa Cruz) for 1 h at room temperature, followed by the respective secondary HRP-conjugated antibodies. The parasite lysates were lysed and subjected to SDS-PAGE. After transfer, the PVDF membrane was incubating with rabbit antibody against SAG1 for 1 h at room temperature, followed by the respective secondary HRP-conjugated antibody.
Mass spectrometry-based proteomics. The magnetic beads coupled with antibodies against HA were sent to the Proteomic Core Facility of the University of California Davis for mass spectrometry analysis. Briefly, the proteins were digested using Promega modified trypsin overnight at room temperature on a gently shaking device. The resulting peptides were analyzed by online liquid chromatographytandem mass spectrometry (LC-MS/MS) with Q-Exactive. All MS/MS samples were analyzed using X! Tandem [The GPM, thegpm.org; version X! Tandem Alanine (2017. 2.1.4)]. X! Tandem was set up to search the uniprotHSTG_crap database, assuming the digestion enzyme trypsin. X! Tandem was searched with a fragment ion mass tolerance of 20 ppm and a parent ion tolerance of 20 ppm. Glu¡pyro-Glu of the N terminus, ammonia loss of the N terminus, Gln¡pyro-Glu of the N terminus, deamidation of asparagine and glutamine, oxidation of methionine and tryptophan, dioxidation of methionine and tryptophan, and dicarbamidomethyl of lysine were specified in X! Tandem as variable modifications. Scaffold (version Scaffold_4.8.6; Proteome Software, Inc., Portland, OR) was used to validate MS/MSbased peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 50.0% probability by the Scaffold local false-discovery rate (FDR) algorithm. Peptide identifications were also required to exceed specific database search engine thresholds, and X! Tandem identifications were also required at least. Protein identifications were accepted if they could be established at greater than 9.0% probability to achieve an FDR less than 5.0% and contained at least 1 identified peptide. Protein probabilities were assigned by the Protein Prophet algorithm (43). Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins sharing significant peptide evidence were grouped into clusters.
Construction of a stable NF-B reporter ⌬traf2 HEK293 cell line. A TRAF2 knockout cell line was made in NF-B GFP reporter HEK293 cells (System Biosciences) using CRISPR/Cas9. Wild-type NF-B reporter HEK293 cells are expressing green fluorescent protein (GFP) when the NF-B pathway is activated. The vector pSpCas9 (BB)-2A-Puro (PX459) was purchased from Addgene (plasmid 48139). Guide RNAs (gRNAs) were designed and selected based on MIT's site CRISPR Design Tool (http://crispr .mit.edu/). The gRNAs (20 nucleotides in length) were designed to target the first exon of the TRAF2 gene. The forward gRNA for TRAF2 (5=-CACCGCCTGCAGAAACGTCCTCCGC-3=) and the reverse guide RNA (5=-AAACGCGGAGGACGTTTCTGCAGGC-3=) were selected as the guides with the lowest score of "offtarget" events. After the appropriate guides were designed, they were cloned into the pSpCas9 (BB) vector for coexpression with Cas9. All of the procedures were based on the protocol by Ran et al. (44). NF-B reporter cells were plated at 7 ϫ 10 4 cells per well in a 24-well plate and incubated for 24 h at 37°C. They were then transiently transfected with the guide RNAs/CRISPR/Cas9 plasmid using the X-tremeGENE 9 protocol as per the manufacturer's instructions (Roche). At 48 h posttransfection, cells were plated in a new 24-well plate, and selection with 1 g/ml puromycin was initiated. Cells were left under puromycin selection for 3 days. On the third day of selection, isolation of the clonal population was initiated using serial dilution. The clonal population was isolated after 12 days. Sequence analysis of the genomic DNA detected the insertion of a nucleotide at position 114 of the first exon of TRAF2, and Western blot analysis further verified the TRAF2 knockout (Fig. S2).
Stable NF-B reporter ⌬traf2 cell line infection with Toxoplasma parasites. NF-B reporter wild-type and NF-B reporter Δtraf2 HEK293 cells were seeded at 80% confluence in a 96-well plate and incubated for 4 h at 37°C. Subsequently, the cells were infected with the Pru, Pru Δgra15, or RH strain (multiplicity of infection [MOI] of 2), stimulated with TNF-␣ (20 ng/ml), or left uninfected/unstimulated. The GFP absorbance of each 96-well plate was measured.
Statistical analysis. All statistical analyses were performed using Graph Pad Prism version 7.0. All the data presented are mean Ϯ standard deviation (SD), and the exact n values are mentioned in each of the figure legends. For all calculations, P values of Ͻ0.05 are considered significant. For a one-variable test with two groups, the two-way analysis of variance (ANOVA) was used, followed with Sidak's multiplecomparison test. For more than three groups with one variable, one-way ANOVA was followed by Dunnett's multiple-comparison test.