Role of the PD-1/PD-L1 Pathway in Experimental Trypanosoma cruzi Infection and Potential Therapeutic Options

Abstract

Chagas disease (CD) is a neglected chronic infection caused by the protozoan parasite Trypanosoma cruzi (T. cruzi). A significant portion of infected people develops cardiac or digestive alterations over a lifetime. Since several chronic infections associated with antigen persistence and inflammation have been shown to lead to T cell exhaustion, new therapies targeting co-inhibitory receptors to regain T cell activity are under consideration. This study explored immune therapeutic approaches targeting the inhibitory PD-1/PD-L pathway in an experimental model for CD. Infected PD-L1 knockout mice (PD-L1 KO) showed increased systemic parasitemia in blood although no significant differences in parasite load were observed in different organs. Furthermore, we found no significant differences in the frequency of activated T cells or proinflammatory cytokine production when compared to WT counterparts. PD-L1 deficiency led to the production of IL-10 by CD8⁺ T cells and an upregulation of Tim-3 and CD244 (2B4). Unexpectedly, the lack of PD-L1 did not contribute to a significantly improved T cell response to infection. Single blockade and combined blockade of PD-1 and Tim-3 using monoclonal antibodies confirmed the results observed in infected. PD-L1 KO mice. Our results describe for the first time that the interruption of the PD-1/PD-L1 axis during acute T. cruzi infection does not necessarily enhance the immune response against this parasite. Its interruption favors increased levels of parasitemia and sustained upregulation of other co-inhibitory receptors as well as the production of regulatory cytokines. These results suggest that the clinical application of immune therapeutic approaches targeting the PD-1/PD-L1 axis in CD might be risky and associated with adverse events. It highlights that more research is urgently needed to better understand the immune regulation of T cells in CD before designing immune therapeutic approaches for a clinical context.

Keywords: Chagas Disease; PD-1; PD-L1; Tim-3.; Trypanosoma cruzi; co-inhibitory receptors.

Figure 1

PD-L1 expression on antigen-presenting cells after in vitro and in vivo infection with T. cruzi. BMDCs were infected with T. cruzi in vitro. After 72 h the PD-L1 Expression was analyzed by flow cytometry. (A) The Histogram shows the MFI of PD-L1 expression on infected (red) and non-infected (blue) BMDCs. FMO staining control (light blue) was used to identify and gate the PD-L1cells population. Expression of PD-L1 on APCs from WT and PD-L1 KO mice infected with T. cruzi. On day 22- 24 post-infection spleens from infected and n.i. mice were isolated and the expression of PD-L1 was evaluated by flow cytometry. (B) Expression of PD-L1 on CD11b^- CD11c⁺ (DCs). (C) Expression of PD-L1 on CD11b⁺ Ly6G^-Ly6C⁺ MΦ. Data shown are corresponding to two independent experiments n.i. = 4; infected =6. Asterisks denote P values of <0.05. *P<0.05 and **P<0.01.

Figure 2

Time course of co-inhibitory receptors expression on T cells from WT and PD-1 KO mice infected with T. cruzi. Time-course experiments showing expression of (A) PD-1 on CD4⁺ and (B) CD8⁺ T cells; (C) Tim-3 on CD4⁺ and (D) CD8⁺ T cells; (E) CD244 on CD4⁺ and (F) CD8⁺ T cells. Error bars indicate standard errors of the means (SEM). Data from three independent experiments. Under the graphs, n is the absolute number of mice used per time point, and the values below are the geometric means (GM). Data were analyzed for statistical significance using the Kruskal-Wallis test following Dunn´s multiple comparisons test. Asterisks denote P values as described in Methods; ns (not significant) was not plotted to avoid a busy figure.

Figure 3

PD-L1 deficiency does not affect activation, IFN-γ production, or Granzyme B production but induces IL-10 secretion on CD8⁺ T cells after stimulation. Spleen cells were isolated from mice and analyzed by flow cytometry after a 5 h stimulation with PMA/Ionomycin. (A, B) PD-L1 deficiency does not affect T cell activation, defined by CD44 expression, after T. cruzi infection on 22-24 dpi. (C, D) T. cruzi infection induces IFN-γ on CD4⁺ and CD8⁺ T cells from WT and PD-L1 Ko mice. PD-L1 deficiency does not affect IFN-γ production. (E) IL-10 is only induced at 22-24 dpi in CD8⁺ T cells from PD-L1KO Mice (F) Granzyme B is significantly upregulated only on CD8⁺ T cells from WT mice. PD-L1 deficiency does not affect Granzyme B production. Error bars indicate standard errors of the means (SEM). Data from two independent experiments. Under the graphs, n is the absolute number of mice used per time point, and the values below are the geometric means (GM). Data were analyzed for statistical significance using the Kruskal-Wallis test following Dunn´s multiple comparisons test. Asterisks denote P values as described in Methods; ns (not significant) was not plotted to avoid busy figures.

Figure 4

Combined therapeutic intervention with αPD-1 and αTim-3 monoclonal antibodies does not reduce parasitemia but induces IL-10. (A) Scheme of infection and combined αPD-1 and αTim-3 therapy. Mice were infected i.p. with 2x10³ T. cruzi parasites on day 0. They received αPD-1 and αTim-3 in a concentration of 0.2 mg each/dose on day 0 and day 7 after infection. The control group received i.p. 2 doses of isotype antibody in the same concentration. WT mice infected and isotype treated n=8; WT infected and αPD-1+αTIM- 3 n=7. Results from two independent experiments (B) Parasitemia and (C) body weight were analyzed over the course of infection (28 days) and are shown as means with SD. Whole blood samples were collected, and sera were isolated and pooled. Cytokine levels were determined by a cytometric bead assay. Results are expressed as the cytokine concentration of the pooled sera samples (D) IL-10, (E) TNF-α, and (F) IFN-γ. (G) Effect of interruption of PD-1/PD-L1 signaling on parasite load analyzed by T. cruzi-specific qRT-PCR. Comparative parasite load in spleen, liver, heart, and skeletal muscle from infected and treated with blocking antibodies against αPD-1+ αTIM-3 or their respective isotype controls. Parasite load was calculated from a standard curve. The standard error of the mean is indicated (SEM). Asterisk denotes P values of < 0.05 by One-way ANOVA compared to isotype control values. P < 0.05*; ns (not significant).