Human villous trophoblast cells (BeWo line) were acquired from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cells were maintained in RPMI 1640 medium (Cultilab, Campinas, SP, Brazil) supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin (Sigma Chemical Co., St Louis, MO, USA), and 10% fetal bovine serum (FBS) (Cultilab) in a humidified incubator at 37°C and 5% CO₂ (61).

T. gondii tachyzoites (RH strain, 2F1 clone) expressing the β-galactosidase gene were kept in BeWo cells in RPMI 1640 medium supplemented with penicillin and streptomycin and 2% FBS at 37°C and 5% CO₂ (61).

Tissue culture-derived trypomastigotes of T. cruzi (TCT, Y strain) were kept in VERO cells in RPMI 1640 medium supplemented with penicillin and streptomycin and 2% FBS at 37°C and 5% CO₂ (43).

Full-term human placentas (36 to 40 weeks of gestation) (n=8) were obtained from pregnant patients after elective cesarean section deliveries at the Clinics Hospital of the Universidade Federal de Uberlândia (HC-UFU), MG, Brazil. Exclusion criteria included pre-eclampsia, hypertension, infectious disease such as toxoplasmosis and Chagas disease, chronic renal disease, cardiac disease, connective tissue disease, diabetes, and other diseases that could hamper the results of this study. After collection, placental tissues were washed in sterile phosphate-buffered saline (PBS) to take out the excess of blood, and the dissection of the villous explants was performed using a stereomicroscope to detach endometrial tissue and fetal membranes up to 1 h after collection. Then, floating terminal chorionic villous containing five to seven free tips per explant were collected, washed with 1x PBS, placed in 96-well plates (one per well), and cultured in 200 µL of RPMI 1640 medium containing 10% FBS, penicillin (100 U/mL), and streptomycin (100 μg/mL) for 24 h at 37°C and 5% CO₂ (50).

Purification of recombinant P21 protein (GenBank: EU004210.1) was carried out as previously described (43, 62).

Briefly, Escherichia coli BL21 strain transfected with plasmid pET-28a (+) (Novage) with the gene coding for rP21 were placed in Luria Bertani (LB) medium with the Kanamycin selection antibiotic (50 µg/mL) and kept under agitation overnight at 37°C until the optical density (OD) at 600 nm reached 0.6 to 0.9. Next, 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG, Sigma) was added to induce expression of rP21. For bacterial lysis, lysozyme (50 mg/mL) was added for 20 min, and then the Sonicador Branson Sonifier 450 was used for 20 min (1-min cycle with 30-sec interval). The bacterial lysate pellet was resuspended in 9 M urea buffer solution and incubated with a nickel resin column in agitation for 2 h at 4°C. rP21 has a histidine tail with an affinity for nickel resin, sticking to the column. Then, the column was washed to capture rP21, as follows (1): three washes in binding buffer (5 mM imidazole, 500 mM NaCl, Tris-HCl 20 mM pH 8.0, 6 M urea) (2); three washes in wash buffer (20 mM imidazole, 500 mM NaCl, 20 mM Tris-HCl pH 8.0, 6 M urea); and (3) four elutions with elution buffer (1 M imidazole, 50 mM NaCl, 20 mM Tris-HCl pH 8.0, 6 M urea). The buffers used had increasing concentrations of imidazole (5 mM, 20 mM, and 1 M, respectively) as it has a greater affinity for the nickel resin than the histidine tail present in the rP21, thus allowing the release of rP21 in the supernatant. Dialysis was performed in 1x PBS for 48 h under agitation at 4 °C with a 3.5 kDa micropore dialysis membrane (Spectra/131198).

After purification, the protein was incubated with a Polymyxin B column (Sigma, P1411) to remove bacterial endotoxins and the concentration was quantified with the Bradford method (63). rP21 purity was analyzed with 13% SDS-PAGE gel with Coomassie blue staining.

The viability of BeWo cells treated with rP21 was evaluated by MTT assay [(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltertrazolin bromide)], as described by Mosmann (64).

BeWo cells (3×10⁴/200 μL) were seeded in 96-well plates in 10% FBS medium for 24 h at 37 °C and 5% CO₂. Then, the cells were treated with rP21 (10, 20, 40, or 80 µg/mL) in medium for additional 24 h. The rP21 concentrations used were based on previous studies (42, 43, 46, 47, 65). As control, cells were treated only with 10% FBS medium.

Afterwards, the supernatants were collected and stored at -80 °C for further cytokine measurement, and the cells were incubated with MTT (5 mg/mL) diluted in 10% FBS medium for 4 h at 37°C and 5% CO₂. Then, the formazan crystals were solubilized in 10% sodium dodecyl sulfate (SDS, Sigma) and 50% N, N- dimethyl formamide (Sigma). The absorbance was measured at 570 nm in a microplate spectrophotometer (VersaMax ELISA Microplate Reader, Molecular Devices, Sunnyvale, CA, USA). The viability of BeWo cells was expressed as percentage of viable cells (cell viability %) by comparing the treatments with the cells treated only with 10% FBS medium (medium, 100% viability). Two independent experiments with eight replicates were performed.

The impact of T. cruzi infection (TCT) and/or rP21 treatment on T. gondii invasion (3 h of infection) and intracellular proliferation (24 h of infection) in BeWo cells were evaluated by means of the β-galactosidase assay (50).

For all T. gondii invasion and proliferation experiments, BeWo cells (3×10⁴/200 μL) were seeded in 96-well plates in 10% FBS medium for 24 h at 37 °C and 5% CO₂. Also, in order to remove non-internalized parasites after the period of infection stated below, the cells were washed thrice with 1x PBS. Nine different experimental protocols were carried out, as represented in Figure 1 :

T. gondii invasion and intracellular proliferation was obtained according to a reference curve with free tachyzoites (from 1 × 10⁶ to 15.6 × 10³). For all of the above experiments, data are presented as percentage of T. gondii invasion or proliferation in comparison with the cells only infected with T. gondii (medium, 100% invasion or proliferation). Three independent experiments with eight replicates were performed.

The analysis of the actin cytoskeleton polymerization was evaluated by confocal fluorescence microscopy.

BeWo cells (1×10⁵/500 μL) were seeded in 24-well plates on glass coverslips (13 mm) in 10% FBS medium for 24 h at 37°C and 5% CO₂. Next, the cells were infected or not with T. gondii tachyzoites (3:1) and treated or not simultaneously with rP21 (40 μg/mL) for 3 h. As controls, cells were uninfected and untreated (uninfected medium) or infected and untreated (infected medium). Afterwards, the cells were washed thrice with 1x PBS to remove non-internalized parasites and then fixed with paraformaldehyde (4%) for 30 min. Next, the coverslips were washed thrice with 1x PBS and incubated overnight with a mouse monoclonal primary antibody anti-T. gondii (SAG1/p30) (Abcam TP3 #ab8313, Cambridge, UK) [diluted 1:500 in PBS solution containing 0.15% gelatin, 0.1% sodium azide, and 1% saponin (PGN-saponin)]. Then, the coverslips were washed thrice with 1x PBS and incubated with Alexa Fluor 488-conjugated anti-mouse IgG (Invitrogen #A11001, Waltham, MA, USA) (diluted 1:500 in PGN-0.01% saponin solution), sodium isothiocyanate tetramethylrhodamine (TRITC) conjugated with phalloidin (Sigma, P1951) (diluted 1:50 in PGN-saponin 0.01%), and TO-PRO-3 (Life Technologies, Carlsbad, CA, USA) (diluted 1:450 in PGN-saponin 0.01%) for 1 h at room temperature under darkness to label T. gondii tachyzoites, actin, and nuclei, respectively. Afterwards, the coverslips were washed thrice with 1x PBS and mounted on glass slides and analyzed in a confocal fluorescence microscopy (Zeiss, LSM 510 Meta, Germany) with an inverted microscope (Zeiss Axiovert 200 M). The mean actin fluorescence intensity was evaluated in 100 cells in triplicate of each condition on ImageJ software (National Institutes of Health, USA).

The analysis of the cell cycle of T. gondii was evaluated using flow cytometry, as previously described (66).

BeWo cells (1×10⁵/500 μl) were seeded in 24-well plates in 10% FBS medium for 24 h at 37°C and 5% CO₂. Next, the cells were infected with T. gondii tachyzoites (3:1) for 3 h. Then, the cells were washed three times with 1x PBS to remove non-internalized parasites, and the cells were treated with rP21 (40 µg/ml) in medium for an additional 24 h. As control, cells were infected and untreated (infected medium). Afterwards, the cells were washed thrice with 1x PBS to remove extracellular parasites, and a cell scraper was used to detach the cells. The intracellular parasites were isolated through a syringe with 21- and 26-gauge needles. The parasite/cell lysate was centrifuged at 1,500 rpm for 30 sec to remove cell debris, and the supernatant was centrifuged at 1,500 rpm for 5 min to pellet the isolated parasites. The parasites from each condition were washed with 1x PBS and fixed with 70% ethanol at 4°C overnight. Next, fixed parasites were washed with 1x PBS and incubated overnight with a mouse monoclonal primary antibody anti-T. gondii (SAG1/p30) (Abcam TP3 #ab8313) (diluted 1:500 in PGN-saponin 0.01%). Then, the parasites were washed with 1x PBS and incubated with Alexa Fluor 488-conjugated anti-mouse IgG (Invitrogen #A11001), 10 μg/mL of propidium iodide, and 100 μg/mL of RNAse A (all reagents diluted in 1x PBS) for 45 minutes at room temperature under darkness. The DNA content for parasite cell cycle analysis was performed using the Cytoflex flow cytometer (Beckman Coulter, USA), and data collected using FlowJo software (version 7.6.3). Two independent experiments with five replicates were performed.

To assess whether treatment with rP21 could be compromising the infective capacity of T. gondii, we performed a reinfection assay to evaluate the parasite invasion.

BeWo cells (1×10⁶/2000 μL) were seeded in 6-well plates in 10% FBS medium for 24 h at 37°C and 5% CO₂. Next, the cells were infected with T. gondii (3:1) for 3 h in medium. Then, the cells were washed thrice with 1x PBS to remove non-internalized parasites, and the cells were treated with rP21 (40 µg/mL) in medium for an additional 24 h. Afterwards, the cells were washed three times with 1x PBS to remove extracellular parasites, and the intracellular parasites were isolated using syringes as described above. Parasites isolated from each condition were then used to reinfect new fresh BeWo cells previously seeded in 96-well plates (3×10⁴/200 μL) (3:1) in medium for 3 h. As control, cells were infected with parasites from cells that were not treated with rP21 (infected medium). Next, the cells were washed thrice with 1x PBS to remove non-internalized parasites and then submitted to the β-galactosidase assay for analysis of T. gondii invasion. Three independent experiments with eight replicates were performed.

The viability of human placental explants treated with rP21 was evaluated by MTT assay, as previously described (52, 66).

Human placental explants were collected as previously described and cultured in 96-well plates in 10% FBS medium for 24 h at 37°C and 5% CO₂. Next, the villous explants were treated with rP21 (40 μg/mL) in medium for an additional 24 h. As control, the villous explants were treated only with 10% FBS medium (medium).

Afterwards, the supernatants were collected and stored at -80°C for further cytokine measurement, and the villous explants were submitted to the MTT assay, as described above. The viability of villi was expressed as percentage viability of villous explants in relation to explants treated only with medium (medium, 100% viability).

Additionally, the villi treated or not with rP21 were collected for morphological analysis using hematoxylin and eosin staining (50).

The T. gondii intracellular proliferation in human placental explants treated with rP21 and/or infected with T. cruzi was evaluated by the β-galactosidase assay (50).

The villous explants were cultured in 96-well plates for 24 h in 10% FBS medium at 37°C and 5% CO₂. Then, two different experimental protocols were carried out, as demonstrated in Figure 1 :

The T. gondii intracellular proliferation in villous explants samples was performed by adding 150 µL radioimmunoprecipitation assay buffer (RIPA) (50 mM Tris-HCl, 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, and 0.1% sodium dodecyl sulfate, pH 7.5) supplemented with a protease inhibitor cocktail (Complete, Roche Diagnostic, Mannheim, Germany) to each villi and homogenizing the samples on ice for protein extraction. The lysate was centrifuged at 21,000 × g for 15 min at 4°C and the supernatant was collected to measure the protein total (µg/mL) by Bradford assay (63). Aliquots of 20 µL of each sample were used to determine T. gondii intracellular proliferation by the β-galactosidase assay, as described above. The number of tachyzoites were normalized according to the protein concentration of each villi, showing the number of tachyzoites per µg of tissue. The data were expressed as percentage (%) of T. gondii proliferation: the mean number of tachyzoites from control (untreated and infected villi) corresponded to 100% of the parasite proliferation, and the number of tachyzoites from each condition was transformed into a percentage according to 100% of the parasite proliferation from the control.

The human cytokines IL-4, IL-6, IL-8, IL-10, MIF, and TNF-α were measured in BeWo cell supernatants and human placental explants. The measurement of these cytokines in the supernatants was performed using a double-antibody sandwich enzyme-linked immunosorbent assay (ELISA), according to the manufacturers’ instructions (BD Biosciences, San Jose, CA, USA or R&D Systems, Minneapolis, MN, USA). Data were expressed in pg/mL according to a standard curve of each cytokine for BeWo cells, while for villous explants data were normalized according to the protein concentration of each villi, where data about cytokines were obtained by the ratio between the concentration of cytokines in pg/mL and the concentration of total protein from the Bradford assay in µg/mL, resulting in pg/µg of tissue (67).

The detection limits of each cytokine were: IL-4, IL-10, and TNF-α (7.8 pg/mL), IL-6 (4.7 pg/mL), IL-8 (3.12 pg/mL), and MIF (31.2 pg/mL).

The present research protocol using human tissue samples was performed in accordance with relevant guidelines and regulations, being approved by the Ethics Committee of the Universidade Federal de Uberlândia, MG, Brazil, with approval number 3.679.426. A consent term was obtained from all participants included in this study.

We assessed whether T. cruzi infection would modulate T. gondii invasion and proliferation in the host cell ( Figures 2A–D ). We observed that simultaneous infection with both parasites increased T. gondii invasion (^* P = 0.0465) in relation to cells only infected with T. gondii (infected medium) ( Figure 2A ), but the pre-infection with T. cruzi did not show the same effect ( Figure 2C ). In addition, we observed that pre- and post-T. cruzi infection decreased T. gondii proliferation (^* P = 0.0353 and ^* P = 0.0111, respectively) in relation to cells only infected with T. gondii (infected medium) ( Figures 2B, D ).

As described above, previous studies showed that P21 protein of T. cruzi is able to modulate the parasite proliferation itself (42–44). Thus, we evaluated whether the effect promoted by T. cruzi infection on T. gondii infection could be related to its protein P21. We observed that this protein did not alter cell viability at any concentration in relation to untreated cells (uninfected medium) ( Figure 3A ). For further experiments, we chose only the concentration of 40 μg/mL.

Next, we assessed whether treatment with rP21 could modulate both T. gondii invasion and proliferation ( Figures 3B–E ). We observed that simultaneous treatment with rP21 plus T. gondii infection (^** P = 0.0034), simultaneous infection with T. cruzi and T. gondii (^**** P < 0.0001), and simultaneous infection with T. cruzi and T. gondii plus rP21 (^** P = 0.006) increased T. gondii invasion in relation to cells only infected with T. gondii (infected medium) ( Figure 3B ). However, the simultaneous treatment with rP21 (^** P = 0.001), simultaneous infection with T. cruzi (^* P = 0.0215), and simultaneous infection with T. cruzi plus rP21 (^**** P < 0.0001) decreased T. gondii proliferation in relation to cells only infected with T. gondii (infected medium) ( Supplementary Figure 1 ). In addition, pre-infection with T. gondii and subsequent treatment with rP21 and/or infection with T. cruzi decreased T. gondii proliferation in relation to cells only infected with T. gondii (infected medium) (rP21: ^**** P < 0.0001; TCT: ^*** P = 0.0005; TCT + rP21: ^**** P < 0.0001) ( Figure 3C ).

Furthermore, we observed that pre-treatment with rP21 (^* P = 0.0350) increased T. gondii invasion in relation to cells only infected with T. gondii (infected medium) ( Figure 3D ). On the other hand, the pre-infection by T. cruzi in the absence (^&& P = 0.0094) or presence of rP21 (^&&& P = 0.0005) decreased T. gondii invasion in relation to cells infected with T. gondii and treated with rP21, but no change was observed in comparison with cells only infected with T. gondii (infected medium) ( Figure 3D ). In addition, pre-treatment with rP21 (^* P = 0.0182), or pre-infection with T. cruzi (^*** P = 0.0002) or pre-infection and simultaneous treatment with rP21 (^* P = 0.0311) decreased T. gondii proliferation in relation to cells only infected with T. gondii (infected medium) ( Figure 3E ).

We evaluated the role of CXCR4 and the actin cytoskeleton in T. gondii invasion in BeWo cells treated with rP21 ( Figures 4A–F ). It is important to highlight that in this set of experiments, the T. gondii infection was simultaneous with the rP21 treatment and without the presence of T. cruzi.

We observed that, as seen previously, rP21 increased T. gondii invasion (^*** P = 0.0007) in relation to cells only infected (infected medium). Interestingly, BeWo cells infected and treated with the CXCR4 inhibitor AMD3100 (^&&& P = 0.0008) or infected and treated with AMD3100 plus rP21 (^&& P = 0.0041) reduced the parasite invasion in comparison with cells only treated with rP21 ( Figure 4A ).

As for actin cytoskeleton polymerization ( Figure 4B ), we observed that rP21, regardless of T. gondii infection, increased actin cytoskeleton polymerization in relation to uninfected and untreated cells (^# P = 0.0416; ^## P = 0.0047) (uninfected medium). In addition, cells infected and treated with rP21 increased actin cytoskeletal polymerization in relation to cells only infected (^* P = 0.0416) (infected medium).

Representative images ( Figures 4C–F ) are shown of uninfected and untreated BeWo cells (4C), uninfected and treated with rP21 (4D), infected and untreated (4E), and infected and treated with rP21 (4F), where one can observe that in the presence of rP21 there is a greater actin cytoskeleton polymerization of the host cell (red), as well as a greater amount of T. gondii inside the cell (green).

We assessed whether the decrease in T. gondii proliferation triggered by rP21 could be involved with the progression in the parasite cell cycle. We observed that treatment with rP21 after T. gondii infection decreased the number of parasites in the G1 phase (^* P = 0.0145) and increased the number of parasites in the S/M phase (^* P = 0.0136) in relation to the parasites that were not treated (infected medium) ( Figures 5A, B ).

In addition, we observed that parasites that were treated for 24 h with rP21 and allowed to reinfect fresh new cells showed similar invasive capacity if compared with parasites that were not treated (infected medium) ( Figure 5C ).

Next, since rP21 and T. cruzi decrease T. gondii proliferation, we evaluated whether this protein, as well as T. cruzi, would be interfering in the host immune response through cytokine production in BeWo cells ( Figures 6A–F ).

We observed that cells only infected with T. cruzi, in the absence or presence of rP21, decreased MIF production in relation to uninfected and untreated cells (uninfected medium) (TCT: ^## P = 0.0029; TCT + rP21: ^# P = 0.0143), as well as in relation to uninfected cells and treated with rP21 (TCT: ^&&&& P = 0.0008; TCT + rP21: ^&& P = 0.0049) ( Figure 6A ). On the contrary, cells only infected with T. gondii, or infected and treated with rP21, or infected with T. gondii/T. cruzi in the absence or presence of rP21 increased MIF production in relation to uninfected and untreated cells (uninfected medium) (Infected medium: ^### P = 0.0003; rP21: ^### P = 0.0002; TCT: ^### P = 0.0007; TCT + rP21: ^# P = 0.0277) ( Figure 6D ).

Regarding IL-6 production, we observed that cells only treated with rP21 increased the production of this cytokine in relation to uninfected and untreated cells (^# P = 0.0117) (uninfected medium) ( Figure 6B ). Conversely, cells infected only with T. cruzi had a lower IL-6 production in relation to cells only treated with rP21 (^& P = 0.0424) ( Figure 6B ). In addition, we observed that cells only infected with T. gondii, or infected and treated with rP21, or infected with T. gondii/T. cruzi in the absence or presence of rP21 increased IL-6 production in relation to uninfected and untreated cells (uninfected medium) (infected medium: ^### P = 0.0001; rP21: ^#### P < 0.0001; TCT: ^#### P < 0.0001; TCT + rP21: ^#### P < 0.0001) ( Figure 6E ). Furthermore, we observed that cells only infected with T. gondii and treated with rP21 increased IL-6 in relation to cells only infected with T. gondii (^* P = 0.0269), while infection with T. gondii/T. cruzi decreased IL-6 production in relation to cells infected with T. gondii and treated with rP21 (^& P = 0.0497) ( Figure 6E ).

Concerning IL-8 production, no changes were detected in the absence of T. gondii ( Figure 6C ). However, we observed that cells only infected with T. gondii increased the production of this cytokine in relation to uninfected and untreated cells (^# P = 0.039) ( Figure 6F ), while treatment with rP21 decreased IL-8 production in relation to cells only infected with T. gondii (infected medium) (^* P = 0.0218) ( Figure 6F ).

IL-4, IL-10, and TNF-α showed values below detection levels in all conditions tested (data not shown).

We also assessed the impact of rP21 and T. cruzi on T. gondii infection in human placental explants ( Figures 7A–E ).

Firstly, we observed that rP21 treatment did not cause cytotoxicity to villous explants ( Figure 7A ), as seen by the morphological analysis of the explants treated with rP21 similar to untreated villous ( Figures 7B, C ).

Next, T. gondii intracellular proliferation was evaluated in different situations. First, we observed that pre-treatment with rP21 decreased T. gondii proliferation in relation to villi only infected with T. gondii (infected medium) (^* P = 0.0236) ( Figure 7D ). However, pre-infection with T. cruzi in the absence or presence of rP21 increased T. gondii proliferation in relation to villi only infected with T. gondii (infected medium) (TCT: ^** P = 0.0011; TCT + rP21: ^** P = 0.0062), as well as in relation to the villi pre-treated with rP21 (^&&&& P < 0.0001) ( Figure 7D ).

We also observed that pre-infection with T. gondii and the subsequent treatment with rP21 and/or T. cruzi infection increased T. gondii proliferation compared to villi only infected with T. gondii (infected medium) (rP21: ^** P = 0.0075; TCT: ^* P = 0.021; TCT + rP21: ^* P = 0.0132) ( Figure 7E ).

Next, we measured cytokines in the supernatant of the villous explants infected with T. gondii and subsequently treated with rP21 and/or infected with T. cruzi ( Figures 8A–J ), as well as in the villous explants that were pre-treated with rP21 and subsequently infected with T. gondii ( Supplementary Figure 2 ).

Regarding IL-4 production, no change was detected in the absence of T. gondii ( Figure 8A ). However, we observed that T. gondii infection and the subsequent infection and treatment with T. cruzi–rP21 increased the production of this cytokine in relation to uninfected and untreated villi (uninfected medium) (^## P = 0.0011), as well as in relation to villi only infected with T. gondii (infected medium) (^*** P = 0.0004), infected and treated with rP21 (^&&& P = 0.0001), and infected with T. cruzi (^$ P = 0.0101) ( Figure 8F ).

Concerning IL-6 production, no change was detected in the absence of T. gondii ( Figure 8B ). Nevertheless, we observed that infection with T. gondii/T. cruzi decreased IL-6 production in relation to uninfected and untreated villi (uninfected medium) (^## P = 0.0077) ( Figure 8G ). In addition, subsequent T. cruzi infection in the absence or presence of rP21 decreased the production of this cytokine in relation to villi only infected with T. gondii (infected medium) (TCT: ^** P = 0.0022; TCT+rP21: ^* P = 0.0178), as well as in relation to villi infected with T. gondii and treated with rP21 (TCT: ^&& P = 0.0011; TCT+rP21: ^& P = 0.0101) ( Figure 8G ).

Regarding IL-8 production, no change was verified in the absence of T. gondii ( Figure 8C ). However, we observed that T. gondii/T. cruzi infection in the absence or presence of rP21 decreased the production of this cytokine in relation to uninfected and untreated villi (TCT: ^#### P < 0.0001; TCT+rP21: ^### P = 0.0009), as well as in relation to villi only infected with T. gondii (TCT: ^**** P < 0.0001; TCT+rP21: ^*** P = 0.0002) or infected and treated with rP21 (TCT: ^&&&& P < 0.0001; TCT+rP21: ^&&& P = 0.0004) ( Figure 8H ).

Concerning MIF production, we observed that villi only treated with rP21, or infected with T. cruzi and treated with rP21 increased the production of this cytokine in relation to uninfected and untreated villi (uninfected medium) (rP21: ^# P = 0.013; TCT+rP21: ^## P = 0.0039) ( Figure 8D ). In addition, villi only infected with T. gondii, or infected and treated with rP21, increased MIF production in relation to uninfected villi, whereas subsequent T. cruzi infection with rP21 decreased the production of this cytokine (Infected medium: ^### P = 0.0003; rP21: ^#### P < 0.0001; TCT+rP21: ^# P = 0.0153) ( Figure 8I ). Furthermore, MIF production was lower in villi infected with T. cruzi in the absence or presence of rP21 in relation to villi only infected with T. gondii (^**** P < 0.0001), as well as in relation to villi infected and treated with rP21 (^&&&& P < 0.0001) ( Figure 8I ).

Regarding TNF-α production, we observed that villi only treated with rP21, or infected with T. cruzi in the absence or presence of rP21 increased the production of this cytokine in relation to uninfected and untreated villi (uninfected medium) (^#### P < 0.0001) ( Figure 8E ). Furthermore, villi infected only with T. gondii increased TNF-α production in relation to uninfected and untreated villi (uninfected medium), while subsequent T. cruzi infection decreased the production of this cytokine (infected medium: ^## P = 0.0035; TCT: ^# P = 0.0192) ( Figure 8J ). In contrast, villi infected with T. gondii and treated with rP21 or infected with T. cruzi in absence or presence of rP21 decreased TNF-α production in relation to villi only infected with T. gondii (infected medium) (rP21: ^* P = 0.0242; TCT: ^**** P < 0.0001; TCT+rP21: ^** P = 0.0013) ( Figure 8J ). In addition, villi infected with T. gondii/T. cruzi decreased the production of this cytokine in relation to villi infected with T. gondii and treated with rP21 (^&&& P = 0.0001). However, villi infected with T. gondii/T. cruzi and treated with rP21 increased TNF-α production in relation to villi infected with T. gondii/T. cruzi (^$$ P = 0.0088) ( Figure 8J ).

Furthermore, we observed that villi pre-treated with rP21 with subsequent infection with T. gondii did not alter IL-4 and IL-6 production in relation to villi infected or not with T. gondii ( Supplementary Figures 2A, B ). Nevertheless, pre-treatment decreased IL-8 production in relation to uninfected or infected villi (^## P = 0.002; ^*** P = 0.0005) ( Supplementary Figure 2C ).

Concerning MIF production, we observed that villi infected with T. gondii increased the production of this cytokine in relation to uninfected villi (^## P = 0.0045), while pre-treatment with rP21 decreased MIF production in relation to villi infected (^* P = 0.023) ( Supplementary Figure 2D ).

Regarding TNF-α production, we observed that pre-treatment with rP21 decreased the production of this cytokine in relation to uninfected or infected villi (^# P = 0.0234; ^*** P = 0.0003) ( Supplementary Figure 2E ).

Based on our results, we propose a mechanism of action induced by rP21 and T. cruzi in T. gondii infection in in vitro and ex vivo models, as shown in Figure 9 . In BeWo cells, we observed that T. cruzi infection and/or treatment with rP21 increase T. gondii invasion and decrease its proliferation. The greater invasion may be related to the binding of rP21 to CXCR4, leading to actin polymerization and favoring T. gondii invasion into the host cell. Conversely, the reduced T. gondii proliferation may be related to an arrest in the S/M phase of the parasite cell cycle without compromising its infectivity capacity, in addition to the upregulation of IL-6 and downmodulation of IL-8 ( Figure 9A , left panel). In human placental explants, the pre-treatment with rP21 decreased T. gondii proliferation and downmodulated TNF-α production, whereas post-treatment and T. cruzi infection increased T. gondii proliferation, in addition to the upregulation of IL-4 and downmodulation of IL-6, IL-8, MIF, and TNF-α ( Figure 9B , right panel). Thus, co-infection with T. cruzi may favor T. gondii transmission at the human maternal–fetal interface, and this phenomenon may be related to its protein P21.