High-intensity training in hypoxia is a protocol that has been proven to induce vascular adaptations potentially beneficial in some patients and to improve athletes' repeated sprint ability. Here, we test the feasibility of training mice using that protocol and identify those vascular adaptations using ex vivo vascular function assessment.
Exercise training is an important strategy for maintaining health and preventing many chronic diseases. It is the first line of treatment recommended by international guidelines for patients suffering from cardiovascular diseases, more specifically, lower extremity artery diseases, where the patients’ walking capacity is considerably altered, affecting their quality of life.
Traditionally, both low continuous exercise and interval training have been used. Recently, supramaximal training has also been shown to improve athletes’ performances via vascular adaptations, amongst other mechanisms. The combination of this type of training with hypoxia could bring an additional and/or synergic effect, which could be of interest for certain pathologies. Here, we describe how to perform supramaximal intensity training sessions in hypoxia on healthy mice at 150% of their maximal speed, using a motorized treadmill and a hypoxic box. We also show how to dissect the mouse in order to retrieve organs of interest, particularly the pulmonary artery, the abdominal aorta, and the iliac artery. Finally, we show how to perform ex vivo vascular function assessment on the retrieved vessels, using isometric tension studies.
In hypoxia, the decreased inspired fraction of oxygen (O2) leads to hypoxemia (lowered arterial pressure in hypoxia) and an altered O2 transport capacity1. Acute hypoxia induces an increased sympathetic vasoconstrictor activity directed toward skeletal muscle2 and an opposed 'compensatory' vasodilatation.
At submaximal intensity in hypoxia, this 'compensatory' vasodilatation, relative to the same level of exercise under normoxic conditions, is well established3. This vasodilation is essential to ensure an augmented blood flow and maintenance (or limit the alteration) of oxygen delivery to the active muscles. Adenosine was shown to not have an independent role in this response, while nitric oxide (NO) seems the primary endothelial source since significant blunting of the augmented vasodilatation was reported with nitric oxide synthase (NOS) inhibition during hypoxic exercise4. Several other vasoactive substances are likely playing a role in the compensatory vasodilatation during a hypoxic exercise.
This enhanced hypoxic exercise hyperemia is proportional to the hypoxia-induced fall in arterial O2 content and is larger as the exercise intensity increases, for example during intense incremental exercise in hypoxia.
The NO-mediated component of the compensatory vasodilatation is regulated through different pathways with increasing exercise intensity3: if β-adrenergic receptor-stimulated NO component appears paramount during low-intensity hypoxic exercise, the source of NO contributing to compensatory dilatation seems less dependent on β-adrenergic mechanisms as the exercise intensity increases. There are other candidates for stimulating NO release during higher-intensity hypoxic exercise, such as ATP released from erythrocytes and/or endothelial-derived prostaglandins.
Supramaximal exercise in hypoxia (named repeated sprint training in hypoxia [RSH] in the exercise physiology literature) is a recent training method5 providing performance enhancement in team- or racket-sport players. This method differs from interval training in hypoxia performed at or near maximal speed6 (Vmax) since RSH performed at maximal intensity leads to a greater muscle perfusion and oxygenation7 and specific muscle transcriptional responses8. Several mechanisms have been proposed to explain the effectiveness of RSH: during sprints in hypoxia, the compensatory vasodilation and associated higher blood flow would benefit the fast-twitch fibers more than the slow-twitch fibers. Consequently, RSH efficiency is likely to be fiber-type selective and intensity dependent. We speculate that the improved responsiveness of the vascular system is paramount in RSH.
Exercise training has been extensively studied in mice, both in healthy individuals and in pathological mouse models9,10. The most common way to train mice is using a rodent treadmill, and the traditionally used regimen is low-intensity training, at 40%–60% of Vmax (determined using an incremental treadmill test11), for 30–60 min12,13,14,15. Maximal intensity interval training and its impact on pathologies have been widely studied in mice16,17; thus, interval training running protocols for mice have been developed. Those protocols usually consist of about 10 bouts of running at 80%–100% of Vmax on a rodent motorized treadmill, for 1–4 min, interspersed with active or passive rest16,18.
The interest in mice exercising at supramaximal intensity (i.e., above the Vmax) in hypoxia comes from previous results that the microvascular vasodilatory compensation and the intermittent exercise performance are both more increased at supramaximal than at maximal or moderate intensities. However, to our knowledge, there is no previous report of a supramaximal training protocol in mice, either in normoxia or in hypoxia.
The first aim of the present study was to test the feasibility of supramaximal intensity training in mice and the determination of a tolerable and adequate protocol (intensity, sprint duration, recovery, etc.). The second aim was to assess the effects of different training regimen in normoxia and hypoxia on the vascular function. Therefore, we test the hypotheses that (1) mice tolerate well supramaximal exercise in hypoxia, and (2) that this protocol induces a larger improvement in vascular function than exercise in normoxia but also than exercise in hypoxia at lower intensities.
The local state's animal care committee (Service de la Consommation et des Affaires Vétérinaires [SCAV], Lausanne, Switzerland) approved all experiments (authorization VD3224; 01.06.2017) and all experiments were carried out in accordance with the relevant guidelines and regulations.
1. Animal housing and Preparation
2. Determination of the Maximal Speed and Standard Assessment of Performance Improvement by Treadmill Incremental Test
NOTE: The following steps are critical to completing the training protocols.
3. Hypoxic Environment
4. Normoxic Environment
5. Supramaximal Intensity Training
6. Low-intensity Training
7. Mice Euthanasia and Organ Extraction
Figure 4: Picture of the dissected vessels. Extracted vessel from the top of the abdominal aorta (underneath the left renal artery) to the end of the right iliac artery, ready to be placed in cold PBS buffer on ice. (1) Abdominal aorta. (2) Right common iliac artery. (3) External iliac artery. (4) Internal iliac artery. (5) Femoral artery. Please click here to view a larger version of this figure.
8. Ex Vivo Vascular Function Assessment
NOTE: A wash corresponds to the emptying and refilling of the chambers with Krebs.
To our knowledge, the present study is the first to describe a program of supramaximal intensity training in normoxia and in hypoxia for mice. In this protocol, mice ran four sets of five 10 s sprints with a 20 s recovery in between each sprint. The sets were interspersed with 5 min of recovery periods. It was unknown whether the mice would be capable of sustaining such a protocol and complete it properly. However, according to Figure 5, the body weight gain of the mice undergoing the supramaximal intensity training was similar to that of the mice undergoing the low-intensity training, both in normoxia and in hypoxia.
The wellness of the animals was monitored twice a week, using score sheets, based on the following criteria: appearance, natural behavior, and body weight. Each of those criteria was graded up to a score of 3, and a mouse with a score of 3 in any of those criteria was considered in pain and/or distress due to the sustained protocol and had to be euthanized. No mouse ever reached a score of 3 over the course of any of the training regimens (Table 1).
As presented in the introduction, it has been hypothesized that supramaximal training, in particular when combined with hypoxia, would induce a compensatory vasodilation. This phenomenon aims at providing sufficient O2 to the contracting muscles, thereby compensating for the imbalance between O2 supply and demand that is enhanced by the combination of supramaximal intensity training and hypoxia. In order to investigate this hypothesis, we used the second technique presented here, the ex vivo vascular function assessment, on the pulmonary artery, the abdominal aorta, and the right iliac artery. Figure 6 shows the dose-response curves obtained at the end of the protocol, on the abdominal aorta of a mouse from the group training at supramaximal intensity in hypoxia. This graph shows the whole process of contraction-relaxation observed after the addition of different pharmacological agents (KCL, Phe, ACh, NLA, and [DEA]/NO) in the organ baths.
Figure 7 shows the dose-response relaxation curve for the right iliac artery to increasing concentrations of ACh. The two represented groups are the supramaximal-intensity-in-normoxia group (SupraN) and the supramaximal-intensity-in-hypoxia group (SupraH). The preliminary results show that SupraH tended to improve ACh-induced relaxation compared to SupraN, with significant differences at 10-5 M and 10-4 M.
Figure 1: Hypoxic setup. The treadmill is placed inside the homemade glovebox, which is linked to a gas mixer. Please click here to view a larger version of this figure.
Figure 2: Description of a supramaximal intensity training session. The mice performed four sets of five 10 s sprints, interspersed with 20 s of rest. The interset rest was 5 min. This figure is adapted from Faiss et al.21. Please click here to view a larger version of this figure.
Figure 3: Schematic representation of the technique to retrieve the pulmonary aorta. (1) Place the tweezers under both the pulmonary artery and the aorta. (2) Pull back the tweezers in the direction of number 2 in order to keep the tweezers under the pulmonary aorta only. (3) Final position of the tweezers. Please click here to view a larger version of this figure.
Figure 5: Body weight evolution over the course of the experiment. In green, the low-intensity training groups; in red, the supramaximal intensity training groups. There was no significant difference between any of the groups at any of the time points (n = 4 mice per group; the data are presented as mean ± SD). Statistical analysis was performed using a two-way repeated measure ANOVA). Please click here to view a larger version of this figure.
Figure 6: Vascular function assessment curves. Succession of contraction and relaxation phases induced throughout the entire protocol, expressed in grams. Representative recording of the variations in vessel tension in response to the applied substances, in a ring of the abdominal aorta isolated from a mouse trained at supramaximal intensity in hypoxia. Please click here to view a larger version of this figure.
Figure 7: Pharmacological responses of an isolated iliac artery preconstructed with phenylephrine (Phe) to acetylcholine (ACh). Cumulative dose-response relaxation curve of the right iliac artery to increasing concentrations of ACh (10-9 to 10-4 M). The results are expressed as mean ± SD of the percentage of change in the tension induced by the vasodilator, with n = 3 in SupraN and n = 4 in SupraH. Statistical analysis was performed using a two-way ANOVA for repeated measures test. *p < 0.05 vs. SupraN. Please click here to view a larger version of this figure.
Table 1: Typical score sheet of a mouse training at supramaximal intensity in hypoxia. We used score sheets to monitor the welfare of the mice. A score of 3 in any of the criteria indicated (appearance, natural behavior, and body weight) or a total score of 5 (by addition of the score of each category) meant the animal was suffering and had to be euthanized.
The first objective of this study was to assess the feasibility of hypoxic high-intensity training in mice and to determine the adequate characteristics of the protocol that would be well tolerated by mice. Purposely, since there is no data using supramaximal (i.e., more than Vmax) intensity training in mice, we had to perform trials based on previous protocols developed with athletes, which consisted of four to five sets of five all-out sprints (about 200% of Vmax), interspersed with 20 s active recoveries, with an interset active recovery of 5 min21,22. Therefore, the initial protocol consisted of six sets of six 10 s sprints at 200% of Vmax, interspersed with 20 s of passive recovery and with an interset passive recovery of 3 min, performed five times per week. After a few try-out runs at 200% of the Vmax, considering the mice had trouble sustaining such a high intensity, we decided to lower the speed to 150% of the Vmax. With that exercise intensity, we tried to run the mice over the length of a full protocol and adjusted both the number of sprints within each set and the number of sets per session. Finally, we increased the recovery time between sets and decreased the frequency of the training sessions. Following a trial-and-error method, we established a final optimal protocol that is very similar to the one used on athletes and made possible for mice to tolerate this supramaximal intensity test.
There is a slight possibility that the performance of the mice might be severely underestimated, as observed from large differences between previous studies utilizing animal exercise protocols23,24. However, in the present study, based on pre-experiment values, it would have been impossible to impose a higher relative intensity on the animals considering the need to complete the entire repeated sprint session. Moreover, the Vmax values reported in this study (28.8 ± 3.7 m/min) seem to be in the range of values previously reported in the same C57BL/6J strain25,26,27,28. For example, Lightfoot et al.25 reported values of ~28 m/min and Muller et al.27 values of 28.3 m/min. Therefore, we are confident that the supramaximal intensity corresponds to sprint training intensity in these mice.
Although critical speed (CS) has been shown (1) to be a valuable mean for prescribing exercise intensity in healthy humans and patients29 and (2) to be perfectly determined in mice23,24,30, the exercise intensity prescription based on the determination of Vmax remains relevant. It is known that, in mice, the determined VO2peak and VO2max depend on the protocol, and, as with humans, VO2max can be determined with a ramp exercise protocol11. Since the aim of the present study was to determine the feasibility of supramaximal repeated sprint in mice, and despite the relevance of CS, we do not believe that using Vmax would be a flaw regarding the objectives of this study.
While observing mice behavior, it became clear that the electric grid at the rear of the treadmill admittedly encouraged mice to run; however, it also seemed to contribute to their fatigue. Indeed, the grid being slightly shifted from the running band, the mice had to generate an extra effort to get back on the lane. We decided to complement this stimulation with another, softer, one, namely the cotton swab stimulation, which decreased the number of shocks received by the animals and prevented them from having to get back from the grid to the lane. Despite the recommendation by Kregel et al.31, it remains unclear whether stress is reduced using the air puff stimulation compared to the electric grid32.
As far as we know, only one study has used "sprint interval training"33. However, since the highest intensity in that study corresponded to 75%–80% of Vmax and the sprint duration was 1.5 min, that protocol was very different from the present one (i.e., 150% of Vmax; 10 s). It was unknown whether supramaximal intensity would be tolerated by the mice. In the present study, we provide results showing that the animals did perform very well in this supramaximal intensity training, both in hypoxia and normoxia. For instance, Figure 5 shows an increase in body weight over the training period similar to that observed in the low-intensity groups. Similarly, Table 1 reflects the level of welfare with a score lower than 3 in all groups. Altogether, those physiological parameters indicate that both hypoxia and supramaximal intensity training were very well tolerated by mice.
The second objective of the present study was to assess the vascular function of the pulmonary artery, the abdominal aorta, and the iliac artery, using isometric vessel tension studies20. This technique allows determining whether the intervention of interest impacted the ability of the vessels to contract and relax in response to pharmacological drugs. As shown in Figure 7, the iliac artery was relaxed using increasing concentrations of ACh. The observed curves reflect a progressive increase in the relaxation of the vessels, more marked for the SupraH group. If any of the observed curves had been completely flat and around 0% of relaxation, it could mean that the drug was not delivered to the organ chamber, or that the vessels had been damaged during the dissection or the mounting on the stirrups, or that one of the drugs was not administered at the optimal dose or for long enough.
The supramaximal intensity training in hypoxia is now transferred to mice and could potentially be used on pathological models in order to improve various parameters, including vascular function, which can be assessed using isometric vessel tension studies.
The authors have nothing to disclose.
The authors would like to thank Danilo Gubian and Stephane Altaus from the Lausanne University Hospital (CHUV) mechanical workshop for helping create the hypoxic setup. The authors would also like to thank Diane Macabrey and Melanie Sipion for their help with training the animals.
Cotton swab | Q-tip | ||
Gas mixer Sonimix 7100 | LSI Swissgas, Geneva, Switzerland | Gas-flow: 10 L/min and 1 L/min for O2 and CO2, respectively | |
Hypoxic Box | Homemade | Made in Plexiglas | |
Motorized rodents treadmill Panlab LE-8710 | Bioseb, France | ||
Oximeter Greisinger GOX 100 | GREISINGER electronic Gmbh, Regenstauf, Germany | ||
Sedacom software | Bioseb, France | ||
Strain gauge | PowerLab/8SP; ADInstruments |