The prescription of physical training in mice presents a great challenge owing to the large variety of available testing protocols (Kemi et al., 2002; Ferreira et al., 2007; Ayachi et al., 2016). In particular, running protocols are often used in research involving small rodents because they generally allow the intensity and volume of physical training to be manipulated more easily (Kregel et al., 2007).

Most of the available protocols are adapted from those applied and already tested for prescribed training in humans (Ayachi et al., 2016). Therefore, some variables used to predict human performance, such as maximal oxygen uptake (VO_2max), velocity associated with its occurrence (vVO_2max), thresholds related to blood lactate response (lactate threshold, anaerobic threshold, and maximal lactate steady state [MLSS]), and maximum running speed (V_peak) (Morgan et al., 1989; Bassett and Howley, 2000; Stratton et al., 2009) have also been used in several animals studies with the aim of reproducing the models already developed to evaluate and monitor training performance and prescription in humans (Carvalho et al., 2005; Ferreira et al., 2007; Manoel et al., 2017).

The most classical training prescription method for aerobic training in humans is based on tests using VO_2max (ACSM, 2011), which represents the highest rate at which oxygen is extracted in the lungs, transported, and used by the body during maximal exercise (Bassett and Howley, 2000), however, it cannot considered the best predictor of performance (Noakes et al., 1990). In order to more accurately predict endurance performance, vVO_2max has emerged and is defined as the minimum velocity at which VO_2max is reached in an incremental exercise protocol (Daniels et al., 1984; Morgan et al., 1989; Billat et al., 1994, 1999; Buchheit et al., 2010; Rodrigues et al., 2017).

Although VO_2max and vVO_2max are consolidated variables to predict performance and to monitor and prescribe aerobic training, their determination requires the use of gas analyzers and a team of researchers familiarized with such instruments, periodic calibration, and high cost. This is true whether the tests are performed in humans or rodents (Fernando et al., 1993; Wisløff et al., 2001). Therefore, reliable and less costly alternatives have been tested and validated for training prescription, such as peak running speed (V_peak), defined as the maximum velocity reached during an incremental test (Machado et al., 2013). This parameter is strongly correlated with endurance performance in human runners (di Prampero, 1986; Wisløff et al., 2001; Machado et al., 2007; Marcaletti et al., 2011) and presents a good correlation with VO_2max.

However, a comparison of training prescriptions and monitoring according to vVO_2max and V_peak has not been performed and validated in mice. Such a study may help in reducing the cost and increasing the practicality of training prescription in this animal model. With the above in mind, the aim of this study was to compare the efficiency of an aerobic physical training program prescribed according to either vVO_2max or V_{peak_K} in mice. Our hypothesis was that V_{peak_K}, as well as vVO_2max, can be used for training prescription to mice, with the advantage of not requiring expensive equipment.

Table 1 shows significant differences in body weight (g), food intake (g) and food intake corrected by body weight after 4 weeks of training (time effect, P < 0.05). Mice showed increased body weight (g) but a reduction of food intake (g) over time in both groups. The food intake after 4 weeks showed a marked reduction of GVP group in comparison to GVO2, although not significant.

Table 2 demonstrates that 4 weeks of training promoted increases in VO_2max (ml·kg⁻¹·min⁻¹), vVO_2max (cm·s⁻¹), V_{peak_K} (cm·s⁻¹), EE (ml·kg⁻¹·min⁻¹), run distance (cm), and delta (%) after 4 weeks of aerobic training (time effect, P < 0.05). Also, GVO₂ group presented an interaction effect (P < 0.05), indicating the combination of group and time together influenced vVO_2max. There were no differences between the groups in pre-or post-conditions (group effect).

In order to expand the data analysis beyond the descriptive statistic, we calculated the effect size comparing post- vs. pre-training period for vVO_2max and V_{peak_K} to assess the magnitude of findings; Figure 1 shows a large effect in both groups.

The aim of the present study was to compare the efficiency of an aerobic physical training program prescribed according to either vVO_2max and V_{peak_K} in mice. Our study demonstrates that both groups, GVO₂ or GVP, improved VO_2máx, vVO_2max., and V_{peak_K} after 4 weeks of aerobic training. In addition, V_{peak_K} has been proved a reliable method to prescribe and monitor exercise training. Moreover, training improved performance in both groups and increased EE and the run distance. To our knowledge, this is the first study to address the importance of V_{peak_K} in animal studies.

In the field of exercise physiology, there is great interest in determining the precise intensity of exercise as various research projects use animal models to study the effects of acute and chronic exercise. In this context, several methods to determine training load have been proposed (Kemi et al., 2002; Lerman et al., 2002; Ferreira et al., 2007).

Non-invasive tests report in the literature that use of VO_2max as the variable to predict the performance of mice diverge from incremental tests and training protocols. In this sense, establishing a consensus about the results is not possible due to a wide range of reported values (47–94 ml·kg⁻¹·min⁻¹), according to the slope, magnitude of increase, volume, exhaustion criteria, lineages, age, physical condition, and sex of animals (Dohm et al., 1994; Scott and Houmard, 1994; Lightfoot et al., 2001; Wisløff et al., 2001; Kemi et al., 2002; Ayachi et al., 2016; Kurauti et al., 2016). A few studies have also focused on demonstrating that lineage and sex variations in mice may influence responses in aerobic power (Lightfoot et al., 2010; McLaughlin et al., 2010).

Considering that, it was conceivable that the use of V_{peak_K} would help to minimize the observed differences in VO_2max once the variable is more sensitive to changes in physical training in humans (Machado et al., 2013; Manoel et al., 2017). In our study, animals trained according to vVO_2max or V_{peak_K} had similar improvements in VO_2max. and performance over time. In fact, GVO₂ group presented a higher, although not significant, V_{peak_K}. Despite the differences between the two prescription methods it is important to emphasize that the current study demonstrated that animals were able to improve aerobic capacity over time. From a metabolic point of view, other metabolic and structural adaptations may have occurred, hence allowing V_{peak_K}-based protocol to be a trustworthy training model for rodents. Therefore, we hypothesize that V_{peak_K} can also contribute to the prescription of exercise in different lineages, sex, and age groups of mice, since it has proved as a training-sensitive variable.

The determination of V_{peak_K}, considering the equation proposed by Kuipers et al. (2003), allows evaluating aerobic power and monitoring the effects of training more reliable, as the equation takes into account the last stage completed added to the product of the rate of increase and the fraction of the incomplete last stage, abolishing the influence of subjective judgments or overestimations (Machado et al., 2013). Furthermore, V_{peak_K} is a low-cost, practical, and reliable parameter, which makes it a fundamental tool for future research in the field of exercise physiology, especially considering the known difficulties of assessing VO_2max in small rodents (Fernando et al., 1993; Scott and Houmard, 1994; Ayachi et al., 2016).

Although our research was carefully conducted, limitations and shortcomings are unavoidable. The content of muscle glycogen or correlation with blood levels of lactate were not verified during the training period. Also, the training was performed during the day (lights on), opposable to mice habits, which present a predominantly nocturnal behavior.

V_{peak_K} has practical implications for researchers aiming to collect data concerning the effects of training due to its low cost, which is made possible by the fact that it does not require any expensive equipment (i.e., gas analyzer). In summary, V_{peak_K}, as well as vVO_2max, can be adopted as an alternative test to determine the performance and correct prescription of a systemized aerobic protocol training to mice.

CP: Experiment design, execution of the experiments, writing; PR: Experiment design; GG: Execution of the experiments; DG: Performed statistical analysis; LT: VO2 analysis: SdM: Discussion of data; FM: Conceive the experiments, discussion of data, writing; SP: Concept of the paper, writing.