Aspirin treatment of mice infected with Trypanosoma cruzi and implications for the pathogenesis of Chagas disease

Abstract

Chagas disease, caused by infection with Trypanosoma cruzi, is an important cause of cardiovascular disease. It is increasingly clear that parasite-derived prostaglandins potently modulate host response and disease progression. Here, we report that treatment of experimental T. cruzi infection (Brazil strain) beginning 5 days post infection (dpi) with aspirin (ASA) increased mortality (2-fold) and parasitemia (12-fold). However, there were no differences regarding histopathology or cardiac structure or function. Delayed treatment with ASA (20 mg/kg) beginning 60 dpi did not increase parasitemia or mortality but improved ejection fraction. ASA treatment diminished the profile of parasite- and host-derived circulating prostaglandins in infected mice. To distinguish the effects of ASA on the parasite and host bio-synthetic pathways we infected cyclooxygenase-1 (COX-1) null mice with the Brazil-strain of T. cruzi. Infected COX-1 null mice displayed a reduction in circulating levels of thromboxane (TX)A(2) and prostaglandin (PG)F(2α). Parasitemia was increased in COX-1 null mice compared with parasitemia and mortality in ASA-treated infected mice indicating the effects of ASA on mortality potentially had little to do with inhibition of prostaglandin metabolism. Expression of SOCS-2 was enhanced, and TRAF6 and TNFα reduced, in the spleens of infected ASA-treated mice. Ablation of the initial innate response to infection may cause the increased mortality in ASA-treated mice as the host likely succumbs more quickly without the initiation of the "cytokine storm" during acute infection. We conclude that ASA, through both COX inhibition and other "off-target" effects, modulates the progression of acute and chronic Chagas disease. Thus, eicosanoids present during acute infection may act as immunomodulators aiding the transition to and maintenance of the chronic phase of the disease. A deeper understanding of the mechanism of ASA action may provide clues to the differences between host response in the acute and chronic T. cruzi infection.

Figure 1. Early administration of ASA increases mortality and parasitemia in response to T. cruzi infection.

A and B . CD-1 mice were infected with the Brazil strain of T. cruzi and mortality (A) and parasitemia (B) assessed in vehicle (•) and ASA treated mice (□, 20 mg/kg; ▪, 50 mg/kg) over 55 days post infection (dpi). Treatment with ASA or vehicle started at 5 dpi. C. Table showing the effects of delayed ASA treatment on chronic experimental T. cruzi infection. ASA treatment (20 mg/kg) was initiated 60 dpi until 120 dpi. Survival and parasitemia were assessed 120 and 75 dpi respectively. Data are represented as mean ± SD are representative of at least 20 mice per group. * and # indicates significance (P≤0.05) from uninfected and infected mice, respectively. ND = not detected.

Figure 2. Effects of early and delayed administration of ASA on cardiac structure and function.

A. Short axis MRI images showing cardiac remodeling during T. cruzi infection of CD-1 mice with and without ASA treatment (20 mg/kg). Arrows indicate the wall of the right ventricle of the heart. B. Cardiac dimensions were assessed in uninfected and infected with or without ASA at 5 (Early) or 60 (delayed) dpi. Parameters quantified included left ventricular internal diastolic diameter (B, while squares), left ventricular wall diameter (B, grey squares), right ventricular internal diameter (B, black squares). C. Ejection fraction measured using echocardiography. D. Left ventricular glucose SUV measured by microPET. Number of mice in each group is indicted. * and # indicates significance (P≤0.05) from uninfected and infected mice, respectively.

Figure 3. Cardiac pathology in ASA-treated mice is no different to vehicle treated controls.

Representative histopathology of infected CD-1 mice with and without ASA treatment (20 mg/kg) at 35 dpi compared with uninfected controls. Sections were stained with H&E. Parasite pseudocysts are observed (arrows). Total magnification of either 100× (A, C, E) or 400× (B, D, F). Images are representative of al least five mice in each group.

Figure 4. ASA inhibits both host- and T. cruzi-derived prostaglandin production.

Plasma TXA₂ levels, measured as the stable hydrolytic product TXB₂ by ELISA, in uninfected or infected CD1 (A) or TXA₂ synthase null (B) mice. ASA treatment (20 mg/ml) was initiated on 5 dpi. Circulating levels were assessed at 30 and 20 dpi, respectively. Data (mean ± SD) are derived from at least 5 mice per group. * and # indicate significance (P≤0.05) from uninfected mice and infected mice, respectively.

Figure 5. Deletion of COX-1 mimics the effects of ASA on parasitemia but not survival in T. cruzi infected mice.

A and B . Survival curves (A) and peripheral parasitemia (B) for WT (black circle), TXA₂ synthase null mice (grey circle) and COX1 null mice (○) mice after inoculation with 10⁵ trypomastigotes of the Brazil strain. C and D . Measurement of plasma PGF_2α (C) and TXB₂ (D) in infected WT (black square), TXA₂ synthase null mice (grey square) and COX-1 null mice (white square) mice. Data are represented as mean ± SD are representative of at least 20 mice per group. * and # indicates significance (P≤0.05) from uninfected and infected WT mice, respectively.

Figure 6. Treatment of T. cruzi-infected mice with ASA modulates cytokine signaling in the spleen of infected mice.

Splenic extracts were prepared from uninfected and infected CD1 mice treated with ASA (20 mg/kg) from 5 dpi. Immunoblotting for TNF-α, SOCS-2 and TRAF-6 was performed on 15 and 30 dpi. β-actin was used as a loading control. Immunoblots are representative from at least three separate experiments.

Figure 7. Treatment of T. cruzi infected COX-1 null mice with ASA increases mortality.

Survival curves for COX-1 null mice treated with vehicle (•; n = 10) or ASA (20 mg/kg) (○; n = 14) beginning on 5 dpi. Mice were inoculated with 10⁴ trypomastigotes of the Brazil strain and observed over 50 dpi.