Blood glucose concentrations were evaluated in blood collected from the rat-tail using test strips (Performa, Roche, Indianapolis, USA). Diabetes was defined as a fasting glucose > 300 mg/dL in tail vein blood 48 h after STZ injection (Junod et al., 1969). Body weights and blood glucose concentrations were measured 48 h after the induction of diabetes and every 30 days thereafter. At the 4th week after diabetes induction, all animals underwent

adaptation to a treadmill originally designed for human use (Runner, Brazil) and modified for use in rats during 10 minutes at 5 m/min for 4 days. On the 5th day, the rats were submitted to a maximal exercise test (MET), consisting of a graded exercise on the treadmill, with speed increments of 5 m/min every 3 minutes, starting at 5 m/min and continuing up to the maximal intensity attained

by each rat, and was stopped when Pexidartinib clinical trial each animal remained more than MDV3100 solubility dmso 50% of the time without giving signs of intention to advance (Melo et al., 2003, Rodrigues et al., 2007, Ilha et al., 2008 and do Nascimento et al., 2010). The values obtained in the MET were used to plan the treadmill training program, which started in the 5th week after diabetes induction. In order to correct the exercise intensity, a second MET was performed in the fifth training week. Exercise was performed on a treadmill twice a day, with an interval of 4 h between each session, 5 days per week (Tancrède et al., 1982), and the training intensity increased gradually, according to the MET results. During the first week, next the running sessions

lasted 10 min, and the duration of each increased each week, reaching 60 min in the 7th week, which was maintained until the 8th week. Moreover, each training session consisted of a warm-up period, a main period and a cooling-off period. During the warm up period, the rats ran 15% of the session time at 30% of the maximum velocity determined by the MET; in the main period, the rats ran 70% of the session time at 60% of the maximum velocity; and in the cooling-off period, the rats ran 15% of the session time at 30% of the maximum MET values. On the day after the last session of treadmill training, the rats were trained to remain on the rota rod apparatus (Insight, Brazil) with the speed adjusted to 12 rpm for 60 s. The following day, the selected rats were tested in the apparatus with the speed adjusted to 16 rpm for 5 sixty-second trials (modified from Linck et al., 2009). The latency to fall (data presented as the mean of the 5 trials) and the number of falls were evaluated. The rats were gently placed in the corner of a 40 cm × 50 cm × 60 cm box, in which the floor was divided into 12 squares, and then filmed with a digital camcorder (DCR-SR47, Sony, Japan) for 3 min (modified from Moreira et al., 2010). The number of crossings from one square to another, the time spent moving, and the number of rearings were counted.