A murine model of Trypanosoma brucei- induced myocarditis and cardiac dysfunction

Abstract

Trypanosoma brucei is a protozoan parasite that causes human and animal African trypanosomiases (HAT and AAT). Cardiac symptoms are commonly reported in HAT patients, and intracardiac parasites with accompanying myocarditis have been observed in both natural hosts and animal models of T. brucei infection. Despite the importance of T. brucei as a cause of cardiac dysfunction and the dramatic socioeconomic impact of African trypanosomiases in sub-Saharan Africa, there are currently no reproducible murine models of T. brucei-associated cardiomyopathy. We present the first clinically relevant, reproducible murine model of cardiac dysfunction in chronic T. brucei infection. Similar to humans, mice showed histological evidence of myocarditis and elevation of serum NT-proBNP with electrocardiographic abnormalities. Serum NT-proBNP levels were elevated prior to the development of severe ventricular dysfunction. On flow cytometry, myocarditis was associated with an increase of most myocardial immune cell populations, including multiple T cell and macrophage subsets, corroborating the notion that T. brucei-associated cardiac damage is an immune-mediated event. This novel mouse model represents a powerful and practical tool to investigate the pathogenesis of T. brucei-mediated heart damage and supports the development of therapeutic options for T. brucei-associated cardiac disease. In characterizing this model, we provide evidence that T. brucei causes cardiac disease, and we suggest that immunopathology is an important contributor to cardiac pathology. Along with other recent studies, our work demonstrates the importance of extravascular spaces, including the heart, for T. brucei pathogenesis.

Importance: African trypanosomiasis is a neglected tropical disease affecting both people and cattle, which represents a major public health problem in sub-Saharan Africa with an enormous socioeconomic impact. Cardiac disease represents an underappreciated clinical manifestation of African trypanosomiasis that may lead to lifelong illness despite successful treatment of infection. However, this aspect of African trypanosomiasis remains poorly understood, partially due to a lack of well-characterized and practical animal models. In this study, we present the development and characterization of a novel, reproducible, and cost-effective mouse model of cardiac dysfunction in African trypanosomiasis. We demonstrate that this model recapitulates major features of cardiac dysfunction in natural infection, including the presence of parasites in the cardiac interstitial spaces, alterations of cardiac biomarkers, and functional changes. This model represents a resource to support the understanding of cardiac complications of trypanosomiasis and the development of new therapies to prevent and treat cardiac involvement in African trypanosomiasis.

Keywords: Trypanosoma brucei; cardiology; immunology; parasitology.

Fig 1

T. brucei infection causes elevated NT-proBNP, electrocardiographic abnormalities, and heart failure with preserved ejection fraction. (A) Measurement of plasma NT-proBNP at 28 dpi. Plasma NT-proBNP is significantly elevated at 28 dpi (n = 12) compared to age-matched uninfected controls (two-sided Student’s t-test, P = 0.00037). (B) Echocardiographic measurements at 28 dpi. At 28 dpi, infected mice (n = 12) exhibit significantly increased left ventricular ejection fraction compared to uninfected age-matched controls, as measured by sedated echocardiography (two-sided Student’s t-test, P = 0.0029). We also measured structural parameters of the left ventricle including thickness of the various left ventricular walls: interventricular septum, left ventricular posterior wall, and left ventricular anterior wall. In addition, we measured the diameter of the left ventricle at peak contraction (systole) and peak relaxation (diastole). (C) Electrocardiographic measurements at 28 dpi. At 28 dpi, infected mice (n = 10) exhibit significantly decreased heart rate (HR, two-sided Student’s t-test, P = 0.00011), increased QRS interval (two-sided Student’s t-test, P = 0.00281), increased QT interval (two-sided Student’s t-test, P = 0.00493), and increased QTc (two-sided Student’s t-test, P = 0.01294) compared to uninfected age-matched controls.

Fig 2

T. brucei infection causes elevated NT-proBNP and decreased heart function at terminal endpoint (33 dpi). (A) Measurement of plasma NT-proBNP at 33 dpi. Plasma NT-proBNP is significantly elevated at 33 dpi in infected mice (n = 6) compared to uninfected age-matched controls (two-sided Student’s t-test, P = 2.8e-05). (B) Echocardiographic measurements at 33 dpi. At 33 dpi, infected mice (n = 6) exhibit significantly decreased ejection fraction compared to uninfected age-matched controls, as measured by awake echocardiography (two-sided Student’s t-test, P = 0.011).

Fig 3

T. brucei parasites occupy extravascular spaces in the heart; representative immunofluorescence microphotographs of the cardiac ventricle of infected mice at 14 dpi at 20x magnification. T. brucei parasites (tdTomato—red) localize separately from CD31-lined vascular structures (Alexa Fluor 488—green), confirming the extravascular localization of parasites. Nuclei are stained with Hoechst (blue).

Fig 4

T. brucei infection causes myocarditis at 28 dpi. (A) Representative brightfield microphotographs of the heart of infected mice at 28 dpi, compared to uninfected age-matched controls at 10x magnification. Inflammation is present in all layers and regions of the heart, and is composed largely of mononuclear cells, primarily lymphocytes, plasma cells, and histiocytes. Inflammation is most severe at the atrioventricular junction. Scale bar: 500 µm. (B) Histological grading of myocarditis (n = 6) at 28 dpi, compared to equal numbers of age-matched controls. The average grade at 28 dpi is “2,” indicating multifocal moderate myocarditis without extensive associated necrosis. Histological grade is significantly higher in infected than in uninfected mice (two-sided Student’s t-test, P = 1.5e-8). (C) Measurement of the percentage of collagen in longitudinal sections of the heart (n = 6) at 28 dpi, compared to equal numbers of age-matched controls. There is no significant difference between infected and control mice (two-sided Student’s t-test, P = 0.31).

Fig 5

The intracardiac immune cell population during T. brucei infection is characteristic of immune-mediated myocarditis. (A) Flow cytometric quantification of CD45+ cells per milligram. At 28 dpi, CD45+ cells are markedly increased in the hearts of infected mice (n = 12) compared to uninfected age-matched controls, supportive of myocarditis (two-sided Student’s t-test, P = 4.6e-7). (B) Flow cytometric quantification of myeloid and lymphoid cell populations. The following cell types were quantified: neutrophils (CD45+, CD11b+, Ly6G+), B cells (CD45+, CD19+), NK cells (CD45+, NK1.1+), CD4+ T cells (CD45+, CD11b–, CD4+), CD8+ T cells (C45+, CD11b–, CD8+), monocytes (CD45+, Cd11b+, Ly6C high, CD64 low), total macrophages (CD45+, Cd11b+, Ly6C low, CD64 high), in addition to four macrophage subsets defined by the expression of CCR2 and MHCII. Most immune cell populations are significantly increased in the hearts of infected mice at 28 dpi. Statistical comparisons were made using one-way analysis of variance (ANOVA). *P < 0.05, **P < 0.01, ***P < 0.001. (C) Visualization of magnitude of changes in immune cell populations using the log2-fold change in cells per milligram compared to uninfected age-matched control mice calculated based on the average cells per milligram in the hearts of uninfected control mice.