Short-term vaccine protocol using rTgHSP70 with ACF, mainly targeting cellular responses, have been described to increase parasite numbers in the brain (Mun et al., 1999, 2000b). Here, we first addressed whether immunization with recombinant rTgHSP70 (Supplementary Figure 1) using Alum adjuvant, has a protective effect against toxoplasmosis. Mice were immunized with rTgHSP70 and/or Alum, challenged with ME49 strain of T. gondii at week 6 of immunization and euthanized 4 weeks later (Figure 1A). Mice showed gradual increase of body weight (Supplementary Figure 2A) and no signs of morbidly (data not shown) during immunization. After infection, all groups presented loss of body weight at 9 days after infection with increased morbidity scores, after which mice presented slight gradual weight loss and sustained morbidity score (Supplementary Figures 2B,C). At week 4 of infection, cyst numbers in the brain were evaluated microscopically and parasitism by qPCR. It was observed that mice immunized with rTgHSP70, regardless of vehicle used, presented less cyst numbers in the brain compared with mice injected with PBS alone (Figure 1B) as well as lower parasite B1 copies (Figure 1C). Regarding inflammatory changes, it was observed that mice inoculated with PBS, Alum, or rTgHSP70 alone showed inflammatory changes in the brain characterized by infiltration of inflammatory cells in the meninges, perivascular cuffs, glial nodules, and inflammatory cells throughout the parenchyma. The lesions were also observed in the brain of mice immunized with rTgHSP70 adsorbed with Alum, but in a less extent (Figure 1D). These data suggest that immunization with rTgHSP70 promotes reduction of parasite numbers and, when in combination with Alum, also decrease inflammation in the brain of infected mice.

To evaluate the development of humoral response, we measured antibody levels by ELISA. We observed that mice immunized with rTgHSP70 presented specific antibody titers compared with non-immunized groups, and this phenotype was even more significantly distinguished in mice receiving rTgHSP70 combined with Alum (Figure 2A). Mice immunized with rTgHSP70 also presented higher levels of IgG1 antibodies, and this phenotype was even higher in animals receiving the protein combined with Alum (Supplementary Figure 3). Moreover, these animals also showed 2–4 times higher titers of specific anti-STAg antibodies than animals injected with PBS alone, especially at week 4 after infection corresponding week 10 of immunization (Figure 2A). Indeed, immunoblotting against STAg showed that sera from immunized mice strongly react with rTgHSP70 (Figure 2B). After infection, all immunized groups recognized protein bands of 30, 70, 78, and 88 kDa, although the animals injected with Alum also recognized 56 and 126 kDa proteins (Figure 2B). To date, several antigenic proteins from T. gondii soluble tachyzoite antigen have been extensively characterized, being TgHSP70 one of them (Xia et al., 2008; Ma et al., 2009). Next, we investigated whether immunization could induce circulating rTgHSP70 specific immune complexes (ICs) formation, since we previously showed that BALB/c mice present specific ICs in the sera during chronic infection (Barenco et al., 2014). In the present study, it was observed that mice receiving rTgHSP70 in combination with Alum presented specific ICs formation (Figure 2C) and in parallel reduced inflammation in the brain (Figures 1D, 2D). These results indicate that immunization with rTgHSP70 is able to induce highly specific antibody titers and, when used in combination with Alum, allows formation of circulating anti-rTgHSP70 ICs.

Due to the observed protective effects of rTgHSP70 immunization, we investigated serum cytokines levels in mice before and after immunization and/or infection. It was observed that rTgHSP70 immunization did not alter the production of IL-2, IL-4, IL-6, IL-10, IL-17a, IFN-γ, nor TNF, although infection with T. gondii promotes the production of inflammatory cytokines IL-6, IFN-γ, and TNF in all groups of mice (Supplementary Figure 4). These results suggest that immunization with rTgHSP70, regardless of adjuvant used, was not able to modify the cytokine profile in infected mice (Figure 2E). To understand the protective phenotype observed in rTgHSP70-immunized mice, spleen cells were collected at week 2 after the last immunization. Spleen cells were incubated with rTgHSP70 and stained with CFSE for proliferation assay. We observed that in vitro rTgHSP70 stimulation was able to induce preferentially B cell proliferation (Figure 3A). T and B cell proliferation phenotype was similar in all groups of immunization (data not shown). Additionally, we found that rTgHSP70-stimulated B cells presented higher CD86 co-stimulatory surface marker, and lower expression of CD80 (Figure 3B) consistent with antibody-producing B cell activation profile (Suvas et al., 2002). The CD86 expression in B cells stimulated with rTgHSP70 was similar in all immunized groups of mice. Thus, these results suggest that rTgHSP70 induces B cell activation by enhancing CD86 surface expression.

There are several possible mechanisms by which antibodies could function protectively in vaccinated mice, including: (1) direct blocking of tachyzoite entry into host cells; (2) replication by phagocytic cells; (3) lysis of extracellular tachyzoites in a complement-dependent manner. In order to verify whether anti-rTgHSP70 antibodies are involved in the control of T. gondii directly, we analyzed the involvement of complement parasite lysis by immune serum and the effect of specific antibodies in the parasite replication in non-hematopoietic and hematopoietic cells. It was observed that sera from rTgHSP70-immunized mice did not inhibit replication of RH 2F1 parasites in NIH fibroblasts, as observed with sera from infected mice (Figure 4A). Furthermore, we also checked whether these antibodies might promote complement-mediated parasite lysis. We observed that sera from immunized mice were not able to lyse RH 2F1 tachyzoites compared with sera from infected animals (Figure 4B). It was also observed by immunofluorescence that sera from rTgHSP70-immunized mice were not able to bind to T. gondii surface antigens (Figure 4C).

Additionally, it was analyzed the ability of heat-inactivated sera from rTgHSP70 immunized mice to control the proliferation of internalized parasite by C57BL/6 murine macrophages. It was observed that sera from PBS-immunized chronically infected mice were able to control the intracellular parasite, which was not observed with sera from rTgHSP70-immunized mice (Figure 4D). Taken together, these data suggest that anti-rTgHSP70 antibodies do not bind to the parasite surface, and thus neither control parasite numbers by neutralization nor promote complement-mediated lysis.

Others have shown that TgHSP70 is able to directly induce NO production in macrophages via TLR2, MyD88, and IRAK4 (Mun et al., 2005). We observed that murine RAW264.7 macrophages presented a dose-response NO production upon stimulation with different rTgHSP70 concentrations, which could be slightly reduced with anti-rTgHSP70 antibodies treatment, but not with sera from PBS immunized group (Figure 5A). Similarly, peritoneal macrophages from C57BL/6 mice stimulated with rTgHSP70 also presented enhanced NO production (Figure 5B). To show that NO production by macrophages is depended on rTgHSP70 integrity, we incubated cells with rTgHSP70 or with heat-denatured rTgHSP70. It was observed that macrophages treated with denatured rTgHSP70 did not produce NO (Supplementary Figure 5). Thus, rTgHSP70-induced NO production was not due to endotoxin contamination and depends on protein integrity.

We asked whether NO production could be sustained for a longer period without rTgHSP70 stimulation and whether this sustained production could be enhance parasite killing. Thus, peritoneal macrophages were first stimulated with rTgHSP70 for 24 h, followed by washing and then infected with T. gondii RH 2F1 strain in presence or not of IFN-γ for additional 24 h. Interestingly, macrophages pre-stimulated with rTgHSP70 still were able to produce NO, although in lesser extent than 24 h-stimulated macrophages, but at similar levels compared with IFN-γ-stimulated macrophages (Figure 5C). Pre-stimulation with rTgHSP70 was able to decrease 60% of the parasite proliferation, similar to IFN-γ stimulated macrophages (Figure 5D). Thus, rTgHSP70 pre-stimulation is able to control T. gondii proliferation by murine peritoneal macrophages probably through NO production (Figure 5E), in agreement with previous literature showing that elevated NO levels present a Toxoplasmacidal role (Schluter et al., 1999; Kang K. M. et al., 2004).

Since it was observed that rTgHSP70 induces NO production by resident peritoneal macrophages of mice and is able to diminish the parasite proliferation in vitro, we quantified the iNOS+ cells in the brain. Strikingly, it was verified that PBS-immunized or Alum-immunized mice examined on week 4 after infection presented significantly lower number of iNOS+ cells in the brain when compared with animals immunized with rTgHSP70 alone or rTgHSP70 adsorbed in Alum (Figure 5F). A comparative analysis of iNOS expression and parasite load was done, showing that higher iNOS expression was correlated with lower parasite load in the brain (Figure 5G). Taken together, these results suggest that immunization with rTgHSP70, regardless of the adjuvant Alum, is able to increase the number of iNOS+ cells in the brain, which is highly correlated to diminished parasite load observed in the organ.

Eight-week-old female C57BL/6 and Swiss Webster mice were maintained at the Federal University of Uberlandia animal facility (Centro de Bioterismo e Experimentação Animal-CBEA/UFU), with 12 h light/dark cycles and free access to food and filtered water. All animal experiments were performed in accordance to Brazilian Government's ethical and were approved by the Animal Experimental Ethics Committee (CEUA) of the Federal University of Uberlandia, under protocol no. 023/11. All procedures including housing and welfare were carried out in accordance with the recommendations in the Guiding Principles for Biomedical Research Involving Animals of the International Council for Laboratory Animal Science (ICLAS), countersigned by the Conselho Nacional de Controle de Experimentação Animal (CONCEA; Brazilian National Consul for the Control of Animal Experimentation) in its E-book (http://www.mct.gov.br/upd_blob/0238/238271.pdf). The Federal University of Uberlandia animal facility (Centro de Bioterismo e Experimentação Animal—CBEA/UFU) is accredited by the CONCEA (CIAEP: 01.0105.2014) and Comissão Técnica Nacional de Biossegurança (CTNBio, Brazilian National Commission on Biosecurity; CQB: 163/02). It was used five mice per experimental group and sample size was calculated using the “resource equation” method (Charan and Kantharia, 2013). All efforts were made to minimize animal suffering and the numbers of mice required for each experiment.

RAW247.6 and NIH fibroblast cell lines were cultured in 25 cm² flasks in RPMI 1640 medium supplemented with heat-inactivated fetal bovine serum (FBS) (both from Cultilab, Campinas,SP, Brasil) and antibiotics (100 U/mL penicillin and 100 μg/mL streptomycin, Sigma-Aldrich, St. Louis, USA) and maintained in incubator at 37°C and 5% CO2. Transgenic T. gondii 2F1 strain (β-gal clone) RH strain parasites (Dobrowolski and Sibley, 1996) was a gift from Dr. Vern Carruthers, Medicine School of Michigan University (USA) and was maintained in NIH fibroblasts by passage every 2 days, cultured in RPMI 2% FBS. T. gondii ME49 strain was maintained in Swiss Webster mice infected at least 1 month beforehand. For infection of experimental C57BL/6 mice, ME49 cysts were harvested from Swiss brain, quantified and adjusted to 10 cysts in 0.2 mL.

For STAg preparation, tachyzoites of the RH strain obtained from the peritoneal exudates from BALB/c mice previously infected by intraperitoneal (i.p.) were washed in phosphate-buffered saline (PBS) and centrifuged at 70 × g. The supernatant containing tachyzoites was then pelleted (720 × g, 5 min at 4°C), suspended in PBS supplemented with protease inhibitors, sonicated and centrifuged at 10,000 × g, 10 min at 4°C, as previously described (Gazzinelli et al., 1991).

The recombinant rTgHSP70 was obtained as previously described (Barenco et al., 2014). Protein purity was checked by SDS-PAGE (Supplementary Figure 1) and submitted for endotoxin removal using polimixin B resin (Sigma Chemical Co., St Louis, USA). Final rTgHSP70 preparations were absent of detectable endotoxin, as measured by LAL reagent (Lonza, Walkersville, MD USA).

Mice were injected subcutaneously with 10 μg of rTgHSP70 dissolved in 100 μL of PBS or Alum adjuvant (Alhydrogel 2%, Invivogen, San Diego, CA, USA). Experimental design is illustrated in Figure 1A. Control groups were injected with vehicles only. Mice were injected 2 more times at weeks 2 and 4 after first immunization. On week 6, a group of animals was used for spleen ex vivo analysis, and the remaining were orally infected with 10 cyst of T. gondii ME49 strain. Four weeks later mice were anesthetized by i.p. injection of Ketamine (Syntec Brasil Ltda, SP, Brazil)/Xylazine (Schering-Plow Coopers, SP, Brazil) and blood samples were collected by puncture of the retro-orbital plexus and animals were euthanized by cervical dislocation. The brains were collected for fresh cysts quantification, histological analysis, and parasite quantification by qPCR. Animals were monitored daily for morbidity scores and every other day for weight changes (Bartley et al., 2006) and the blood samples were collected as described above every 2 weeks for antibody and cytokine analysis in serum samples. When mice reached a cumulative score of 5 on any day or a score of 4 for 2 consecutive days they were euthanized as described (Bartley et al., 2006).

Brain tissue sections fixed in 10% buffered formalin were processed routinely for paraffin embedding and sectioning. Tissue sections were stained with Haematoxilin and Eosin (H&E) for histological assay. The inflammatory score in the brain was done as previously described (Silva et al., 2009). Briefly, perivascular cuffs and inflammatory cells in the meninges as well as total focal or diffuse inflammatory foci were analyzed in a sagittal section. The inflammatory score was represented as arbitrary units: 0–2, mild; 2–4, moderate; 4–6, severe; and above 6, very severe. The histological analyses were done in two histological sections from each mouse using a 40 × objective by two researchers in a blind manner.

To detect iNOS expression by immunohistochemistry, deparaffinized sections were placed in a humidified chamber, and the endogenous peroxidase activity was blocked with 3% hydrogen peroxide. The antigenic unmasking was done in a microwave oven for 7 min and then incubated with PBS plus 0.3% non-fat milk (Nestlé, São Paulo, Brazil). The slides were incubated overnight at 4°C with rabbit anti-iNOS antibody (Santa Cruz Biotechnology, Dallas, Texas, USA) and then with biotin-labeled goat anti-rabbit antibody (Dako, Glostrup, Denmark) for 1 h at 37°C. Next, the sections were incubated with avidin–biotin-peroxidase complex (ABC kit, PK-4000; Vector Laboratories, Burlingame, CA, USA) for 45 min at 37°C. The reaction was developed with 0.03% H₂O₂ plus 3,3′-diaminobenzidine tetrahydrochloride (DAB; Sigma). The sections were counterstained with Harris haematoxylin and examined under a light microscope using a 40 × objective.

The number of cysts was assessed by counting T. gondii cysts in fresh brain samples. For this, one brain hemisphere of each animal was homogenized in 2 mL of PBS by passing through 27G needle. Then, 20 μL of brain homogenate were placed on microscopic slides for cysts quantification by two researchers, in duplicate, in a blind manner. The number of counted cysts was corrected by the concentration in the solution and multiplied by two to obtain the number of cysts in the whole brain.

The quantification of parasites by qPCR was done with genomic DNA (gDNA) extracted using the Trizol method (Life Technologies, Carlshad, CA, USA) following manufacturer's instructions. Absolute quantification was done using B1 gene (FW: 5′-GGAGGACTGGCAACCTGGTGTCG-3′; RV: 5′-TTGTTTCACCCGGACCGTTTAGCAG-3′) primers with 10 ng gDNA using ABI7500 system with SYBR green (Life Technologies). A standard curve made of gDNA extracted from T. gondii RH tachyzoites (1 × 10³–1 × 10⁷, 10-fold dilutions) was used. Parasite numbers in the samples were calculated by interpolation to the parasite standard curve.

Total IgG, IgG1, and IgG2a serum titers were determined by indirect ELISA. Briefly, plates were coated with 2.5 μg/mL rTgHSP70 or STAg in carbonate buffer for 18 h at 4°C. After washes with 0.05% Tween-20 PBS (PBS-T), the plates were blocked with 5% non-fat milk (Neslté) PBS-T (5% PBS-TM) for 1 h at room temperature (RT), and incubated with serial dilutions (6 points, starting at 1:50) of each sample for 1 h at 37°C. After washes, plates were incubated with 1:2000 of anti-IgG or with anti-IgG1 and IgG2a (all from Sigma) for 1 h at 37°C. Plates were washed again and the reaction was developed with enzyme substrate consisting of 0.03% hydrogen peroxide and 1 mg/mL of o-phenylenediamine (OPD, Sigma) and read at 492 nm with VersaMax microplate reader (Molecular Devives). Antibody titers were determined by linear regression (Crowther, 2009).

Immune complexes detection was performed as described (Chaves-Borges et al., 1999; Barenco et al., 2014), using microtiter plates (Kartell SPA, Novigilio, Milan, Italy) coated with anti-rTgHSP70 IgY (10 μg/mL) diluted in carbonate buffer pH 9.6 and incubated at 4°C overnight. After incubation at 37°C for 1 h, plates were washed and incubated with mouse sera diluted 1:40 in 1% PBS-TM. Next, plates were washed and incubated with secondary antibody peroxidase-conjugated anti-mouse IgG (Sigma) diluted 1:1,000. Reaction was developed with OPD (Sigma) and the optical density (OD) was measured at 495 nm by using a VersaMax microplate reader. Immune complex levels were estimated by ELISA index as previously described (Barenco et al., 2014).

STAg antigen was separated by SDS-PAGE and then transferred to PVDF membranes (Millipore Corporation, Billerica, MA, USA), which were blocked with 5% PBS-TM for 1 h and incubated with serum samples diluted 1:100 in 1% PBS-TM overnight. After washes with PBS-T, membranes were incubated with 1:4,000 peroxidase-conjugated anti-mouse IgG (Sigma) in 1% PBS-TM for 2 h, washed again and developed with DAB (Sigma). All steps were performed at RT. Documentation was proceeded on Scanjet G4050 scanner (Hewlett-Packard).

Spleen cells were isolated from mice 2 weeks after the last immunization, using 70 μm cell strainer and incubated with supplemented RPMI 1640 (10% fetal calf serum, 200 μM glutamine and 50 μM β-Mercaptoethanol). For evaluation of proliferative responses, cells were stained with 5 μM CFSE (Invitrogen, Eugene, Origon, USA) for 10 min, centrifuged and incubated at 2 × 10⁵ cells per well in U-bottom culture plates (Costar 3799, Corning, NY, USA) with 10 μg/mL rTgHSP70 or 2.5 μg/mL ConA or 1 μg/mL LPS or 10 μg/mL STAg in supplemented RPMI 1640 (10% FBS, 200 μM glutamine and 50 μM β-Mercaptoethanol).

Cell phenotyping was performed in cells restimulated in vitro. For this purpose, cells were incubated with fluorescent αCD3-BV, αCD4-PECy 7, αCD8-APC, αCD19-APC, αCD80-PE, αCD86-PECy7 (all purchased from BD Biosciences, San Jose, CA, USA), and recorded with FACSCanto II flow cytometer (BD). The results were analyzed using FlowJo software v.10 (TreeStar).

For serum cytokine quantification, samples were processed using CBA kit following manufacturer's instructions (BD) and the results recorded with FACSCanto-II flow cytometer (BD) and analyzed with FACSDiva software (BD).

Immune serum-mediated T. gondii infection inhibition assay was based on a previous publication (Sayles et al., 2000) with modifications. Sera from PBS- and rTgHSP70-immunnized animals and sera from 4 week infected animals were pooled respective to each group. Half of the volume was transferred to another tube for complement inactivation at 56°C for 30 min. Next, sera were diluted (10% [vol/vol]) in culture medium, 0.22-filtered and used to resuspend 2F1 RH tachyzoites followed by 10 min incubation. Parasites (1 × 10⁵) were washed and then used to infect confluent NIH fibroblast culture. Twenty-four hours later, parasite quantification was performed using chlorophenol red–β-D-galactopyranoside (CPRG; Roche, Mannheim, Germany) as described (Dobrowolski and Sibley, 1996).

Evaluation of complement-mediated lysis by immune serum was performed as described (Dando et al., 2001) with modifications. Sera from all immunized groups before (week 6) and 30 days after infection (30 days p.i., week 10) were diluted 1:4 in PBS, incubated at 56°C for 30 min for complement inactivation and added to microtiter plates (75 μL). Meanwhile, ressuspended 2F1 RH tachyzoites (5 × 10⁷ parasites/mL) were mixed v/v with normal mouse serum (pre-diluted v/v in PBS). Next, 50 μL of parasite mixture was added to each well and incubated at 37°C for 15 min and centrifuged at 1,000 × g for 10 min. After that, 50 μL of supernatant was collected and applied to a new plate, in duplicates, for parasite quantification using CPRG (Dobrowolski and Sibley, 1996) without adding lysis buffer.

Formalin-fixed 2F1 RH tachyzoites were placed on microscopic slides and air dried. Slides were incubated at 37°C for 1 h with infected or rTgHSP70-immunized mice sera diluted 1:100 in PBS. After rinsing, slides were incubated at 37°C for 1 h with CF633-conjugated anti-mouse IgG antibody (Sigma) diluted 1:600 in PBS. Finally, slides were rinsed and mounted in glycerin for analysis in LSM 510 Meta Confocal Microscope (Zeiss).

At first, the influence of rTgHSP70 in nitric oxide (NO) production was measured in RAW247.6 cells, a cell line of BALB/c murine macrophage. Cells were plated (5 × 10⁴ cells/well) and treated with medium containing 2.5% rTgHSP70 immunized-mouse or PBS injected-mouse (control group) sera and different rTgHSP70 concentrations. After 72 h incubation, supernatant was collected for NO quantification by Griess method.

Peritoneal macrophages of C57BL/6 mice were collected and plated (1 × 10⁵ cells/well) in 10% RPMI medium for 24 h at 37°C. The cells were then treated with 1 μg/mL rTgHSP70 and/or 1,3 ng/mL IFN-γ for 24 h at 37°C. The NO production was measured by Griess method, and the previously rTgHSP70 and/or IFN-γ stimulated cells were infected with 2F1-RH tachyzoites of T. gondii (3:1) for 24 h at 37°C. The parasitism was quantified using the chlorophenol red–β-D-galactopyranoside assay. The effect of rTgHSP70 denatured by boiling (100°C, 1 h) on NO production analysis was also examined.

In additional experiments, peritoneal macrophages of C57BL/6 mice were plated (1 × 10⁵ cells/well) in 10% RPMI medium for 24 h at 37°C. Cells were infected with 2F1-RH tachyzoites of T. gondii (1:1), that were previously incubated by 20 min at 37°C with sera of rTgHSP70 immunized or chronically infected mice. Before the incubation with the parasites all sera were previously heat-inactivated at 56°C for 45 min. Three hours after the infection, cell monolayers were rinsed with fresh RPMI to remove extracellular organisms. Twenty-four hours later, parasite quantification was performed using chlorophenol red–β-D-galactopyranoside.

Statistical analyses were performed using GraphPad Prism 6 (GraphPad Software). Data were expressed as mean ± SEM and were compared using unpaired t-test for comparing two groups, and one-way or two-way ANOVA followed by Bonferroni post-test when appropriate. Association between categorical variables (pathology score and IC presence) was done by Chi-square and Odds ratio tests. A threshold of 4 (out of 8) was used to discriminate brain pathological status as “mild inflammation” (scores 1 to 3.9) and “severe inflammation” (scores 4 to 8). Association between continuous variables (NO levels and T. gondii numbers) from in vitro and in vivo experiments was performed using Pearson correlation test. Differences were considered significant when P < 0.05.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.