ABC | Volume 115, Nº2, August 2020

Original Article Basilio et al. Effects of intermittent fasting on the heart Arq Bras Cardiol. 2020; 115(2):184-193 epididymal compartments. 24 The sum of the two adipose sites in absolute and relative values was considered for the determination of body adiposity. Characterization of Cardiac Morphology To assess macroscopic morphology, the atrial and left and right ventricular weights were measured in absolute values and compared to BW and tibia length. Samples of the left ventricle were obtained through a transverse incision 6 mm from the apex. The myocardial fragments were immersed in 10% buffered formaldehyde for 48 hours. Each tissue specimen was then rinsed under running water and stored in 70% ethanol solution for more 48 hours. After the fixation step, the specimens were embedded in paraffin blocks. Histological slides were prepared from 4–7 µm thick tissue cross-sections and stained with hematoxylin/eosin and picrosirius red. For the morphometric analysis of cardiomyocytes, the cardiomyocyte area and the myocardial interstitial collagen fraction were measured as described previously. 23-25 Hematoxylin/eosin-stained slides were used for the measurement of cardiomyocyte area. At least 100 cardiomyocytes were sampled per animal. Picrosirius red- stained slides were used to determine interstitial collagen fraction. Once the image field was fixed, components of the cardiac tissue were identified according to the highlighted color. Collagen filaments appeared red, while cardiomyocytes appeared yellow. The interstitial collagen fraction corresponds to the percentage of collagen content throughout the tissue specimen. A minimum of 20 fields were used and perivascular regions were disregarded. For myocardial morphometry, the histological sections were analyzed at a 40-fold magnification with a microscope (LEICA DM LS) coupled to a digital video camera on an IBM microcomputer, equipped with the image analyzer program Image Pro-plus (Media Cybernetics, Silver Spring, Maryland, USA). Analysis of MAPK Expression MAPK protein expression levels were determined using Western blot procedures and specific primary antibodies (Santa Cruz Biotechnology Inc., CA, USA): p-JNK (sc-6254), total JNK1/2 (sc -137019), p-ERK1/2 (sc-16982), total ERK 1 (sc-93). The protein levels obtained were normalized to the expression of GAPDH (6C5, sc-32233). The sample preparation methods and electrophoresis conditions have been described previously. 24,26 Statistical Analysis The Sigma-Stat software was used for data analysis. Firstly, the results were subjected to normality analysis by the Kolmogorov-Smirnov test. Since the variables had a parametric distribution, measures are presented as mean and standard-deviation and were analyzed using two-way analysis of variance (two-way ANOVA), complemented by the Student-Newman-Keuls comparison test. The cross-sectional area results of cardiomyocytes were divided into categories according to the measurement range using Sturges formula. 27 Absolute and relative proportions were analyzed using the Goodman multiple proportions test. 28 The level of significance was set at 5%. Results The results of the functional fitness exercise on a treadmill [total distance (m) and final speed (m/min)] are shown in Figure 1. Exercise groups ET and ETI achieved greater total distance and final speed than control and IF animals. Intermittent fasting did not affect functional fitness (p>0.05) (Figures 1A and 1B). Regarding results of lactate measurements, exercise training exerted a statistically significant effect (p=0.04) on LacE (control and IF: 8.16±0.94; ET and ETI: 5.34±0.88 mmol×L -1 ), without the occurrence of factor interaction. The LT was similar among groups (control: 2.51±1.18; IF: 3.90 ± 0.64; ET: 2.70±0.23; ETI: 3.04±1.33 mmol×L -1 ). The variation in lactate levels between the inflection point and the end of the test was greater (p=0.04) in the sedentary groups (control and IF: 156±19; ET and ETI: 98±18%). The areas under the glucose tolerance curve are shown in Figure 2. Considering the individual effect of dietary intervention (Figure 2A), intermittent fasting was associated with smaller areas of glycemic response. There were no significant differences in the individual effect of exercise training (Figure 2C). On the other hand, ETI animals showed a smaller area under the glucose curve than ET (Figure 2B). No differences were found for the other statistical comparisons. With respect to nutritional variables, food consumption and caloric intake were lower in the IF and ETI groups compared to the respective controls. As individual factors, intermittent fasting and exercise training were associated with reduced energy efficiency and lower BW gain. Although BW gain was lower in the IF group compared to control, adiposity levels were similar between the groups (Table 2). Analysis of cardiac morphology results showed that intermittent fasting per se was associated not only with lower atrial weight (0.063±0.002 vs. 0.053±0.002 g; p=0.006), but also with reduced left ventricle weight, in absolute terms (0.475±0.011 vs 0.420±0.011 mg; p<0.001) and compared to the tibia length. In addition, intermittent fasting reduced heart weight (0.677±0.013 vs. 0.597±0.013 g; p<0.001) (Table 2). Descriptive measurements of myocardial morphometry are shown in Figure 3. The combination of intermittent fasting and exercise training resulted in a significantly smaller cellular area in the ETI group compared to the ET and IF groups (control: 248±46; IF: 255±21; PE: 260±30; PIF: 225±26 μm 2 ). Considering the frequency distribution of cardiomyocytes, most results were classified within the first two classes, delimited to 327.5 µm 2 . However, the ETI group showed a higher proportion of fibers in the 1 st class of values (up to 190.1 µm 2 ) compared to the other groups (p<0.05) (Figure 3B). A statistically significant interaction between intermittent fasting and exercise training was obtained in the interstitial collagen fraction analysis (p=0.01). The fraction of interstitial 186