Conditional Blimp-1^fl/flCD4^Cre (T cell-specific Blimp-1 deficient mice-CKO) and C57BL/6 Blimp-1^fl/fl (WT) generated by crossing C57BL/6 CD4^Cre mice were obtained from The Jackson Laboratory. All mice were genotyped, both for CRE and Flox, following the Jax mice instructions (stock number 017336 and 008100) and bred in the animal facility at the University of São Paulo, Brazil, maintained in a pathogen-free environment. Groups of 6 females of WT and CKO mice, age of 7-8 weeks, were used. All procedures were performed in the accordance with the International Guidelines for the Use of Animals and approved by the local Ethics Committee at the University of São Paulo, Brazil (123/2017).

For in vivo experiments, mice were inoculated intraperitoneally with 10² or 10³ bloodstream trypomastigote forms of the Y strain. From day 5 after infection, blood parasitemia was evaluated in 5 μl of blood from the tail vein by counting 100 fields through direct observation under an optical microscope. Groups of mice infected with 100 trypomastigote forms were treated by gavage with 100 mg/Kg of Benznidazole (BNZ) or vehicle on days 7 and 9-post infection (pi) or intraperitoneally (ip) with dexamethasone (1 mg/Kg) or vehicle on days 9, 12 and 15 pi. The treatment with aminoguanidine were performed ip, daily, for 7 days, with 2.5 mg/Kg of aminoguanidine (AG) or vehicle. The survival rate and weight loss were assessed daily. The quantification of tissue parasitism was done as previously described (18).

Naive CD4⁺ (CD44^-CD25^-GFP^-) T cells from Foxp3-GFP reporter mice were sorted from the spleen and lymph node cell suspensions using a FACSAriaII (BD Bioescience). CD4⁺ T cells were stimulated with plate-bound anti-CD3 (2ug/mL), anti-CD28 (1ug/mL) (both from BioXCell), IL-2 (100 U/mL), TGF-β (5 ng/mL), anti-IFN-γ, and anti-IL-4 (10 ug/mL) for 4 days in RPMI-1640 medium supplemented with 5% FBS (Gibco), 100 U/mL penicillin, 100 ug/mL streptomycin, 1 mM sodium pyruvate, nonessential amino acids, L-glutamine, and 50 uM 2-mercaptoethanol. All recombinant cytokines were obtained from RD, and neutralizing antibodies from BioXCell. After 4 days of culture, 2 x 10⁶ Treg cells (> 80% Foxp3⁺) were transferred intravenously (iv into Blimp-1-deficient mice (CKO) at 3-day post infection T. cruzi. The survival rate and weight loss were assessed daily.

Briefly, the spleens of mice were homogenized and the erythrocytes lysed with 2 ml ACK lysis buffer. The liver was excised, minced with scissors, and digested with collagenase II (Sigma) for 37°C for 45 minutes. Tissue fragments were further dispersed by repeated aspiration and crushed through a 70-µm pore size nylon filter (BD Biosciences). Liver leukocytes were recovered using 40% Percoll gradient centrifugation. Erythrocytes in the cell pellets were lysed, and the remaining cells were resuspended in 5% RPMI. After processing, viability was assessed via Trypan blue exclusion, and the cell concentration determined. Leukocytes from spleen and liver were stimulated with PMA (50 ng/ml) plus ionomycin (500 ng/ml) (Sigma) and brefeldin A (Biolegend) for the analysis of intracellular cytokines by flow cytometry. Single-cell suspensions from spleen were diluted to concentration 2 x 10⁶ cell/well and dispensed into 48-well plates in a total volume of 500 ul of 5% RPMI. As a positive control, we used anti-CD3 (2ug/ml) plus anti-CD28 antibody (1 ug/ml) (BD Bioscience, USA). The minimal viability allowed was 95% and there was no significant variability between the experimental groups. The cell culture supernatants were harvested after 72 h of culture at 37 °C in 5% CO₂, and the levels of IFN-γ and TNF-α determined by ELISA with commercial Kits (R&D Systems, USA) according to the manufacturer’s protocols.

To staining the leukocytes, 1-2x10⁶ cells/tube were incubated with viability dyes (Biolegend) for 15 minutes at RT, protected from light, washed, and add the diluted antibody cocktail (100 times) for 30 minutes in PBS at 4°C. To intracellular staining, the cells were fixed with 4% paraformaldehyde for 11 minutes and permeabilized with 0.5% saponin for 30 minutes. After permeabilization, the antibodies were added and incubated for 30 minutes at 4°C. The antibodies used, obtained from BD Bioscience and Biolegend were CD45 (30-F11), CD3 (145-2C11), CD4 (RM4-5), CD8 (53-6.7), IFN-γ (XMG1.2), IL-10 (JES5-16E3), TNF-α (MP6-XT22), Foxp3 (MF-14), Granzyme B (GB11), Perforin (S16009A), CD44 (IM7), CD11b (M1/70), MHC-II (M5/114.15.2), Ly6C (HK1.4), CD11c (N418), F4/80 (T45-2342), NOS-2 (W16030C) and their respective isotype controls conjugated with different fluorochromes (FITC, PE, APC, PERCP, PE-Cy7, APC-Cy7, BV510, BV650, BV711, PE/Dazzle 594, and Pacific blue). Data acquisition and analysis were performed using the BD FACSCanto II, BD FACSymphony A1 and FlowJo software, respectively. t-SNE analysis was performed using Flow Jo T-SNE and X-Shift plugin.

To histopathological analysis, liver and heart samples were fixed in 10% buffered formalin and paraffin processed. Tissue sections of 5-µm thickness were deparaffinized and stained with hematoxylin and eosin (H&E). To identify steatosis, the liver samples were collected, frozen in OCT medium (Sakura Finetek, Inc, Torrance, CA), and tissue section of 5-µm stained with Oil Red O solution (Sigma) (19). The images of inflammatory infiltrate and lipids visualized as red-orange staining were obtained using light microscope (Leica).

RNA was extracted using a Promega Kit (Promega) according to the manufacturer’s instructions. cDNA was synthesized via a reverse transcription (Kit High Capacity, Applied Biosystems). Real-time PCR for quantitative mRNA expression analyses was performed on a StepOne Plus Real-Time PCR System (Applied Biosystems) using a goTaq qPCR Master Mix fluorescence quantification system (Promega) and the primer sequences used as follows: RPL13a sense, 5’-GGAGGAGAAACGGAAGGA AAAG-3’, and antisense, 5’-TTTCCGTAACCTCAAGATCTGCTT-3’; NOS-2 sense, 5’-CGAAACGCTTCACTTCCAA-3’; and antisense, 5’-TGAGCCTATATTGCTGTGGCT-3’. Prdm1 sense, 5’- GACGGGGGTACTTCTGTTCA-3’; and antisense, 5’- GGCATTCTTGGGAACTGTGT. The standard PCR conditions were as follows: 50°C for 2 min; 95°C for 2 min; and 40 cycles of 15 seconds at 95°C, 30 seconds at 58°C and 30 seconds at 72°C, followed by a standard denaturation curve. The samples were normalized to RPL13a and the analyses performed using the cycle threshold (Ct) method, which allows for quantitative expression analysis using the formula 2^-ΔΔCt.

To determine heart and liver damage, the activity of CK-MB (Creatine Kinase-MB), AST (Aspartate Aminotransferase) and ALT (Alanine Aminotransferase) were measured in the sera of mice non-infected and infected with T. cruzi. Quantification was performed using a specific kit (Labtest^®) followed by read in spectrophotometer at 340 nm (EMAX Molecular Devices Corporation^®) following the manufacturer’s recommendations.

Production of the cytokines IFN-γ and TNF-α was assessed in the serum and supernatant of the splenocytes culture of WT and CKO naive mice or 12 days post-infection with 10³ trypomastigote forms of T. cruzi. Measurements were performed through by ELISA assay, using specific kit (DuoSet^®, R&D Systems) according to manufacturer’s instructions. The nitrite concentration was determined using the conventional Griess reaction method (20).

Cryopreserved liver tissues were fixed in cold acetone, washed in PBS, and the endogenous peroxidase activity blocked with 3% hydrogen peroxide or 30 minutes. Then, a Mach 1 Universal HRP Polymer Detection Kit (Biocare Medical, USA) was used according to the manufacture’s recommendations. The slides were incubated with rabbit anti-mouse NOS-2 ntibody (1:400, Santa Cruz) and counterstained with Mayer’s hematoxylin. For immunofluorescence, the slides were incubated in 0.5% saponin in PBS for 15 minutes, and nonspecific sites blocked with 1% BSA for 30 min RT. The slides were incubated with a FITC-conjugated anti-CD11c antibody and Alexa fluor 594 anti-iNOS antibody (1:100, Biolegend) overnight at 4°C. The sections were washed, and the nucleus stained and mounted with Prolong. The images were analyzed using the Leica SP5 (Leica Microsystems).

The statistical analysis was performed using an unpaired t-test (Mann Whitney) or one-way ANOVA followed by Tukey’s multiple comparisons test. Survival curves were evaluated using long-rank/Mantel-Cox test. All statistical analyses were performed using Graph Pad Prism (8.0 GraphPad Software). All values were considered significantly different at P < 0.05.

We first demonstrated that Blimp-1 expression is increased in splenocytes of mice infected with T. cruzi on day 12 after infection ( Figure 1A ). We further identified that CD4⁺ T cells are the primary cells expressing Blimp-1 during T. cruzi infection ( Figures 1B–D ). Comparing uninfected mice to infected mice, we found that Blimp-1 expression are consistently less than 2% in uninfected mice, whereas infected mice showed increased Blimp-1 expression in CD4, CD8, CD19, and LY6C positive cells. Specifically, the percentages of Blimp-1-expressing cells were 12.37% in CD4⁺ T cells, 3.49% in CD8⁺ T cells, 4.04% in CD19⁺ B cells, and 5.76% in LY6C⁺ cells ( Figures 1B–D ). To investigate the role of Blimp-1 signaling in controlling T. cruzi-induced disease, we infected wild-type (WT) mice and mice lacking Blimp-1 specifically in T cells (CKO mice). Unlike the WT mice, the CKO mice exhibited a significant reduction in body weight starting on day 9 post-infection (pi) and continued to worsen, regardless of whether they were infected with a low (100) or high (1000) parasite inoculum ( Figures 1E, H ). The absence of Blimp-1 in T cells led to severe weight loss, increased parasite burden, and ultimately, the death of infected mice. Additionally, while 80% of WT-infected mice survived for more than 30 days post-infection (pi), all the infected CKO mice succumbed by day 15 pi, irrespective of the size of the inoculum ( Figures 1F, I ). Furthermore, the CKO mice had significantly higher parasitemia compared to the WT mice ( Figures 1G, J ). These findings strongly indicate that Blimp-1 signaling in T cells plays a critical role in mice’s resistance to T. cruzi infection.

To understand how Blimp-1 contributes to resistance against the infection, we investigated whether its deletion in T cells affects the Th1 response in infected mice. Notably, at 12 days post-infection, we observed a higher frequency and greater number of CD4⁺ T cells producing IFN-γ in the spleen of CKO mice compared to WT mice ( Figures 2A, B ). However, the frequency and numbers of CD8⁺ T cells producing IFN-γ were similar in both groups ( Figures 2C, D ). Furthermore, Blimp-1-deficient mice displayed significantly elevated levels of the proinflammatory cytokines IFN-γ and TNF in the supernatant of splenocytes restimulated or not ( Figures 2E, F ) and in their serum ( Figures 2G, H ) compared to WT mice.

During the early acute phase of infection (3 days post-infection), there was a reduction in the frequency and number of CD4⁺Foxp3⁺ Tregs in the spleen of CKO mice compared to WT mice ( Figures 3A, B ). However, this difference was not observed at 6- and 12-days post-infection. Additionally, at 12 days post-infection, the spleen of CKO mice showed a decrease in the number of CD4⁺ T cells producing IL-10 compared to WT mice ( Figures 3C, D ).

Despite these observations, even when we performed an adoptive transfer of Tregs from WT mice to CKO mice, the infected recipients did not experience protection against weight loss ( Figure 3E ) and death ( Figure 3F ). This indicates that the increased susceptibility to the infection and exacerbated inflammatory response in CKO mice is independent of Tregs. In summary, the absence of Blimp-1 in T cells leads to enhanced Th1 inflammation during T. cruzi infection. Moreover, the increased susceptibility to the infection and the exacerbated inflammatory response are not solely reliant on Tregs, suggesting that Blimp-1 regulates other aspects of the immune response in the context of Chagas disease.

The histological analysis of heart and liver revealed a huge increase in leukocyte migration to the liver ( Figure 4A ), but not heart ( Supplementary Figure 1A ) of CKO infected mice on day 12 pi, with extensive hepatocyte vacuolization and stained lipid droplets ( Figure 4B ), indicative of steatosis. In accordance, the activity of AST ( Figure 4C ) and ALT ( Figure 4D ) were significantly increased in T. cruzi-infected CKO mice compared to WT mice, suggesting a protective role mediated by Blimp-1 signaling. In contrast, we did not find significant inflammatory infiltrate in cardiac tissue of CKO and WT mice at day 12 pi ( Supplementary Figure 1A ) that was confirmed by the similar activity of CKMB among the groups ( Supplementary Figure 1B ). In fact, the flow cytometry analysis showed a dramatic increase in the leukocytes infiltrated into the liver of infected WT and CKO compared to uninfected mice ( Figure 5A ). Still, liver infiltrated leukocytes from infected WT and CKO mice exhibit a high CD3⁺, CD11b⁺, CD11c⁺, F4/80⁺, and LY6C⁺ expression compared to uninfected ( Figure 5A ). We found 6 clusters of B and T lymphocytes (CD19⁺ and CD3⁺) and myeloid cells (CD11b⁺CD11c⁺, CD11b⁺LY6C⁺, CD11b⁺LY6G⁺, and CD11b⁺F4/80⁺) in the liver from WT and CKO infected compared to uninfected mice ( Figure 5B ). Both, the frequency and number of CD4⁺ ( Figure 5C ), but not CD8⁺ ( Figure 5C ), T lymphocytes infiltrating the liver of CKO mice are increased compared to WT mice at 12 dpi. Moreover, in the absence of Blimp-1, the frequency, and numbers of CD4⁺ and CD8⁺ T cells producing IFN-γ were increased compared to WT mice ( Figures 5D–G ). Therefore, Blimp-1 expression is essential to control a T. cruzi-induced liver Th1 inflammation. Additionally, in the liver tissue of Blimp-1 deficient animals, there is an increase in Th1 lymphocytes along with a greater recruitment of activated myeloid cells, specifically inflammatory monocytes (CD11b⁺MHCII⁺LY6C⁺), as indicated by the higher expression of LY6C in the population of CD11b⁺MHCII⁺ cells from CKO mice compared to WT mice ( Figures 5H–J ). Furthermore, we found an elevated frequency and number of TNF-producing monocytes in the liver of CKO mice compared to WT mice ( Figures 5K, L ), suggesting their activation and functionality.

To further investigate the impact of the high leukocyte infiltration in the liver of Blimp-1-deficient animals, we treated them with dexamethasone (Dexa). Although there were no changes in the numbers of circulating parasites in the blood ( Supplementary Figure 2A ), we observed a reduction in weight loss ( Supplementary Figure 2B ) and mortality ( Supplementary Figure 2C ) in Dexa-treated CKO mice compared to untreated CKO mice, indicating partial protection against T. cruzi infection with glucocorticoid treatment. These findings suggest that Blimp-1 plays a role in controlling liver Th1 inflammation induced by T. cruzi infection and contributes to the activation of local monocytes. However, it is important to note that susceptibility to T. cruzi infection involves other factors beyond inflammation, as evidenced by the partial protection provided by blocking the inflammation in Blimp-1-deficient mice.

In terms of parasitism control, CKO mice displayed a high number of circulating parasites compared to WT mice, and at 12 dpi, CKO mice exhibited an augmented amount of parasite DNA in the heart and liver tissues compared to infected WT mice ( Figure 6A ). Treatment with benznidazole, the standard trypanocidal drug (21) for T. cruzi infection, resulted in reduced circulating parasites, percentage of body weight, and parasite DNA in the heart and liver at 12 dpi, thereby protecting CKO mice from death compared to vehicle-treated mice ( Figures 6B–E ). These data demonstrate that the absence of Blimp-1 in T cells leads to enhanced parasite persistence during T. cruzi infection. Additionally, control of parasitism primarily involves infiltrating CD8⁺ T cells, nitric oxide (NO) production, and antibody production.

Moving forward, we further examined the expression of nitric oxide synthase (iNOS), an enzyme involved in nitric oxide production, in the liver tissue of infected mice. We observed heightened iNOS protein and gene expression in the liver tissue of infected CKO mice compared to WT mice ( Figures 7A, B ). This finding was corroborated by increased nitrite production in leukocytes isolated from the liver and spleen of CKO mice compared to WT mice ( Figure 7C ). Immunofluorescence data revealed elevated iNOS staining primarily in myeloid cells, specifically CD11c⁺ cells, in the hepatic tissue of Blimp-1 deficient mice compared to WT mice ( Figure 7D ).

Next, we sought to identify which cell subpopulation in the liver, induced by T. cruzi infection, was responsible for NO production. Our analysis revealed a higher frequency and absolute number of LY6C⁺NOS-2⁺ cells, gated on CD11b⁺F4/80⁺ cells (as shown in the strategy gating in Supplementary Figure 3A ) in the liver tissue of CKO mice compared to WT mice ( Figures 7E, F ). Importantly, these cells exhibited high expression of MHCII and CD11c ( Figure 7G ), suggesting that the primary source of NO production was monocyte-derived dendritic cells (MO-DCs) infiltrating the hepatic tissue in response to T. cruzi infection.

Given the elevated production of NO in the liver tissue of Blimp-1 deficient mice, we investigated whether blocking NOS with aminoguanidine (AG) could prevent acute mortality induced by T. cruzi infection in CKO mice. While the treatment with AG did not prevent weight loss ( Figure 7H ), it did partially prolong the survival of AG-treated CKO mice compared to untreated mice ( Figure 7I ).

Next, we found that the production of anti-T. cruzi total IgG antibodies in the serum were similar between CKO and WT mice ( Supplementary Figure 4A ).

Finally, we assess whether the deletion of Blimp-1 in T cells would compromise the cytotoxic function of CD8⁺ T cells. Notably, at 12 days post-infection, we found a reduction of the frequency and number of CD8⁺ T cells activated (CD44⁺) producing granzyme B ( Figures 8A, B ) and perforin ( Figures 8C, D ) in the spleen of CKO mice compared to WT mice. These results suggest that the failure to control parasites in the absence of Blimp-1 is not due to a lack of antibody and nitric oxide production, but due the reduction of the cytotoxic CD8⁺ T cells. In conclusion, our findings strongly indicate that Blimp-1 plays a critical role in control the parasite, dependent on cytotoxic CD8⁺ T cells, and its absence leads to an uncontrolled acute inflammatory process, ultimately resulting in high susceptibility to T. cruzi infection. Additionally, the enhanced production of nitric oxide by monocyte-derived dendritic cells contributes to the exacerbation of the inflammatory response and the pathogenesis of Chagas disease in Blimp-1 deficient mice.