IJCS | Volume 31, Nº6, November / December 2018

587 Table 1 - Body and heart mass of the four groups of studied animals Parameters/groups (n = 5) UC TC UI TI Body mass, initial (g) 37.8 ± 0.58 37.7 ± 0.56 36.80 ± 1.07 33.20 ± 1.58 Body mass, final (g) 40.6 ± 0.95 43.5 ± 0.52 42.52 ± 1.4 41.94 ± 1.02 Heart mass (g) 0.19 ± 0.01 0.19 ± 0.0073 0.176 ± 0.015 0.19 ± 0.02 Heart/body mass, x10 -3 4.69 ± 0.043 4.56 ± 0.012 4.12 ± 0.026 4.65 ± 0.039 Values are presented as mean ± standard error of mean. Abbreviations: UC: untrained control; TC: trained control; UI: untrained infected; TI: trained infected. Ferraboli et al. Effect of mild aerobic exercise in Chagas’disease Int J Cardiovasc Sci. 2018;31(6)585-593 Original Article samples were dehydrated in ethanol series and embedded in Epon resin. Thin sections were double-stained with uranyl acetate and leadcitrate. Two randomlychosenblocks from each RA in which the myocytes were cut in cross sections were used for quantitative analysis. The ultrathin sections were placed on a copper grid and, using a JEOL transmission electron microscope, 10 randomly chosen fields per block were selected for micrographs. Ultrastructural morphometry and stereology Twenty RA electron micrographs per animal, chosen by systematic random sampling of squares, were taken at a final magnification of x15000, and the numerical density of granules/field and the diameter of all granules present in the field were determined. For the volume density of granules, mitochondria, myofibrils, Golgi complex, and interstitiumpresent in the field, the electronmicrographs were analyzed by a stereological test system with 82 points using the Image J software (version 1.47, National Institutes of Health; Collins, 2007), and the values were expressed as percentages. 23 Statistical analysis The data were evaluated with the software Stata 7.0 and are expressed as mean ± standard error of mean (SEM). All continuous variables were normally distributed (Friedman test), and statistical differences between the groups were obtained by two-way analysis of variance (ANOVA) and post hoc Tukey test. P values below 5% were considered statistically significant. Results Body and heart mass At the beginning of the protocol, all study groups presented similar body mass values (35.7 ± 1.9 g). At the end of the study, body weight was similar in the UC (40.6 ± 0.94 g), TC (43.4 ± 0.62 g), UI (42.52 ± 1.4 g), and TI (41.94 ± 1.02 g) groups. Heart mass and heart mass/ body mass ratio showed no significant difference among the groups (Table 1). Morphology of atrial cardiomyocytes The electron photomicrographs in Figure 1 show the structural aspects of the atrial cardiomyocytes in the UC and TC groups, and those in Figure 2 show the structural aspects of the atrial cardiomyocytes in the UI and TI groups. The frequency distribution histogram of the atrial granules showed that T. cruzi infection (UI and TI groups) promoted an increase of small granules (16.7 to 29.9 nm) and a reduction of large granules (50.0 to 75.0 nm) when compared with the UC group. However, a comparison between the UI and TI groups showed a decrease in the diameter of the small granules and an increase in the diameter of large granules promoted by training in both control and infected animals, reversing the process induced by infection. The average granules (30.0 to 49.9 nm) were unchanged in all studied groups (Figure 3). Volume density of the organelles of atrial cardiomyocytes Table 2 shows morphological and quantitative data of the organelles of atrial cardiomyocytes in the RA of animals in the experimental groups. The organelle density parameters observed in the TI groupwere similar to those in the TC group, except for the myofibril density, which was lower in the TI group. In addition, the TI group showed increased density of atrial granules and mitochondria, and reduced density of myofibrils and Golgi complex compared with the UI group, but these parameters were comparable to those in the TC.