High-Intensity Interval Training and Moderate-Intensity Continuous Training Attenuate Oxidative Damage and Promote Myokine Response in the Skeletal Muscle of ApoE KO Mice on High-Fat Diet

The purpose of this study was to investigate the effects of high-intensity interval training (HIIT) and moderate-intensity continuous training (MICT) on the skeletal muscle in Apolipoprotein E knockout (ApoE KO) and wild-type (WT) C57BL/6J mice. ApoE KO mice fed with a high-fat diet were randomly allocated into: Control group without exercise (ApoE−/− CON), HIIT group (ApoE−/− HIIT), and MICT group (ApoE−/− MICT). Exercise endurance, blood lipid profile, muscle antioxidative capacity, and myokine production were measured after six weeks of interventions. ApoE−/− CON mice exhibited hyperlipidemia and increased oxidative stress, compared to the WT mice. HIIT and MICT reduced blood lipid levels, ROS production, and protein carbonyl content in the skeletal muscle, while it enhanced the GSH generation and potently promoted mRNA expression of genes involved in the production of irisin and BAIBA. Moreover, ApoE−/− HIIT mice had significantly lower plasma HDL-C content, mRNA expression of MyHC-IIx and Vegfa165 in EDL, and ROS level; but remarkably higher mRNA expression of Hadha in the skeletal muscle than those of ApoE−/− MICT mice. These results demonstrated that both exercise programs were effective for the ApoE KO mice by attenuating the oxidative damage and promoting the myokines response and production. In particular, HIIT was more beneficial to reduce the ROS level in the skeletal muscle.


Introduction
Atherosclerosis can pathologically affect the large arteries of the heart, as well as the peripheral arteries in the human body [1,2]. In the peripheral arteries, atherosclerosis has a considerable effect on skeletal muscle structure and function. It has been reported that individuals with peripheral arterial disease (PAD) usually suffer a myopathy in the diseased limbs caused by the oxidative damage and mitochondrial disorder [3,4]. Apolipoprotein E (ApoE) is an important component of all plasma lipoproteins and serves as a ligand for the cell-surface lipoprotein receptors, such as the LDL-receptor. ApoE knockout (KO) mice spontaneously develop hypercholesterolemia and atherosclerosis when fed standard chow [5]. It has also been observed that a high-fat diet (HFD) can further exacerbate

Determination of the Maximal Running Speed on Treadmill
Five ApoE KO mice performed a treadmill running test, which started at 4.8 m/min for 10 min with an incline of 0 • and the speed was progressively increased 1.2 m/min every 3 min until exhaustion. The exhaustion was judged when the mouse stayed still either for three seconds on the electric grid or received 100 shocks without moving [30]. The last speed was defined as the maximal running speed.

Training Protocols
The HIIT program was described before [30], with slight modification, which consisted of 4 sets of 5 × 10-s sprints with 20 s of rest between each sprint and the interset rest was 5 min. One training session took about 23 min in total. The exercise intensity of the sprint was about 100% of the measured maximal running speed. On the other hand, the ApoE −/− MICT group performed the continuous endurance running for 40 min with a speed at 40% of the determined maximal running speed. All training sessions for the two groups were carried out in the morning, three times per week, for six weeks.

Assessment of Endurance Exercise Performance
The mice in the ApoE −/− HIIT and ApoE −/− MICT groups performed an incremental treadmill running test to exhaustion after the training intervention. The protocol was the same as the one used to determine the maximal running speed. The running distance was recorded as the endurance exercise performance. After the incremental treadmill exercise, the mice rested for at least 48 h. Then they were anesthetized and blood samples were collected by the percutaneous cardiac puncture. The muscle samples from gastrocnemius, soleus, and extensor digitorum longus (EDL) were removed, cleaned, and quick-frozen in liquid nitrogen, and then stored at −80 • C.

Real-time Quantitative PCR Analysis
Total RNA was isolated from about 50 mg of crushed gastrocnemius muscle the using TRIzol reagent (TaKaRa, Japan) and about 1 µg total RNA was reverse-transcribed to cDNA using a kit (FSQ-101; Toyobo Co., Ltd., Japan) according to the manufacturer's instructions. Besides, total RNA was isolated from about 10 mg of soleus and EDL, respectively, using the RNA Isolation Kit by TransGen Biotech (Beijing, China) and about 1 µg of total RNA was reverse-transcribed to cDNA using the same kit (FSQ-101; Toyobo Co., Ltd., Japan). Moreover, the real-time qPCR was performed in an ABI 7500 Real-time PCR System (Thermo Scientific, Inc., Waltham, MA, USA) using the SYBR Green Real-time PCR Master Mix kit (Toyobo Co., Ltd., Osaka, Japan) with the previously synthesized cDNA as a template in a 20 µL reaction volume. Glutamate-cysteine ligase catalytic subunit (Gclc; gene ID: 14629; QT00130543), glutathione reductase (Gsr; gene ID: 14782; QT01758232), and 18S ribosomal RNA (Rn18s; gene ID: 19791; QT02448075) commercial primers from Qiagen (Germany) were used.

Western Blotting
Total proteins were isolated from 50 mg of gastrocnemius using RIPA protein extraction reagents (P0013B; Beyotime, Inc., Shanghai, China). Protein concentration was measured using the BCA protein assay kit (Pierce 23225; Thermo Fisher Scientific, Inc.). Twenty micrograms of proteins were separated on Bolt 4-12% Bis-Tris PlusGels (NW04125BOX; Thermo Fisher Scientific, Inc., Waltham, MA, USA.) by electrophoresis, and the fractionated proteins were subsequently transferred to a nitrocellulose membrane using iBlot Gel Transfer Stacks Nitrocellulose (IB23001; Thermo Fisher Scientific, Inc.). The blots were probed using the following antibodies: Nuclear factor erythroid- The individual values were originally expressed as a ratio of a standard (β-actin content) and then expressed as a fold change of the control group value.

Glutathione Redox State and Protein Carbonyl Content of the Skeletal Muscle
Glutathione redox state (the ratio of reduced glutathione (GSH) to oxidized glutathione (GSSG); GSH/GSSG) was measured from 50 mg of the gastrocnemius muscle by GSH and GSSG commercial kits from Solarbio (BC1175 and BC1185, Beijing, China), according to the manufacturer's protocols. Protein carbonyl content was assayed in the homogenate supernatant of 50 mg of gastrocnemius tissue using the commercial assay kit purchased from Solarbio (BC1275, Beijing, China), according to the manufacturer's instructions.

Plasma Irisin and Muscle Musclin Concentration
Plasma irisin and musclin concentration in the gastrocnemius were assessed according to the manufacturer's instructions with the mouse irisin and musclin ELISA kits (Gene lab., Beijing, China), respectively. The plates were read at 450 nm (Bio Tek Synergy H1, Bio Tek Instruments, Inc., Winooski, VT, USA).

Statistical Analysis
All values were presented as the mean ± standard error (SE). Statistical analyses were performed using SPSS Statistical software V 19.0 (IBMCorp., Armonk, NY, USA). Comparisons between the means of the WT mice and ApoE −/− CON groups were made using the independent sample t-test. The one-way ANOVA was used to analyze the impact of different interventions on the ApoE −/− mice followed by the least significant difference (LSD) post hoc test at p < 0.05 level of significance.

Body Weight, Running Distance, and Plasma Lipid Profiles
There were no significant differences in body weight and running distance between the ApoE −/− CON group and WT mice and among CON, MICT, and HIIT groups of ApoE −/− mice ( Figure 1A,B). However, significantly higher levels of plasma TC, TG, and LDL-C were observed in the ApoE −/− CON group than those of the WT group ( Figure 1C-E). Meanwhile, the ApoE −/− HIIT and ApoE −/− MICT groups had significantly lower plasma TC and TG levels, while the ApoE −/− MICT group had a significantly higher plasma HDL-C level than those of the ApoE −/− CON group ( Figure 1C,D). In addition, the ApoE −/− MICT group had a significantly higher plasma HDL-C level than that of the ApoE −/− CON and ApoE −/− HIIT groups, respectively ( Figure 1F).
All values were presented as the mean ± standard error (SE). Statistical analyses were performed using SPSS Statistical software V 19.0 (IBMCorp., Armonk, NY, USA). Comparisons between the means of the WT mice and ApoE −/− CON groups were made using the independent sample t-test. The one-way ANOVA was used to analyze the impact of different interventions on the ApoE −/− mice followed by the least significant difference (LSD) post hoc test at p < 0.05 level of significance.

Body Weight, Running Distance, and Plasma Lipid Profiles
There were no significant differences in body weight and running distance between the ApoE −/− CON group and WT mice and among CON, MICT, and HIIT groups of ApoE −/− mice ( Figure 1A,B). However, significantly higher levels of plasma TC, TG, and LDL-C were observed in the ApoE −/− CON group than those of the WT group ( Figure 1C-E). Meanwhile, the ApoE −/− HIIT and ApoE −/− MICT groups had significantly lower plasma TC and TG levels, while the ApoE −/− MICT group had a significantly higher plasma HDL-C level than those of the ApoE −/− CON group ( Figure 1C,D). In addition, the ApoE −/− MICT group had a significantly higher plasma HDL-C level than that of the ApoE −/− CON and ApoE −/− HIIT groups, respectively ( Figure 1F).

The mRNA Expression of Vegfa165, MyHC-IIa, MyHC-IIx, and MyHC-IIb in Soleus or EDL
The mRNA expression of Vegfa165 in soleus and EDL was significantly higher, and MyHC-IIb in EDL was significantly lower in the ApoE −/− CON group, compared to those

The mRNA Expression of Vegfa165, MyHC-IIa, MyHC-IIx, and MyHC-IIb in Soleus or EDL
The mRNA expression of Vegfa165 in soleus and EDL was significantly higher, and MyHC-IIb in EDL was significantly lower in the ApoE −/− CON group, compared to those of WT mice (Figure 2A,B,E). However, six weeks of the HIIT and MICT induced a significantly lower mRNA expression of Vegfa165 in EDL, respectively, and the HIIT group had a markedly lower mRNA expressions of MyHC-IIx in EDL, compared with those of the Apo KO control mice ( Figure 2B,D). Moreover, the mRNA expression of Vegfa165 and MyHC-IIx in EDL of ApoE −/− HIIT mice was significantly lower than those of ApoE −/− MICT mice ( Figure 2B,D). There were no significant differences in MyHC-IIxa between the ApoE −/− CON group and WT mice and among CON, MICT, and HIIT groups of ApoE −/− mice ( Figure 2C).
icantly lower mRNA expression of Vegfa165 in EDL, respectively, and the HIIT group had a markedly lower mRNA expressions of MyHC-IIx in EDL, compared with those of the Apo KO control mice ( Figure 2B,D). Moreover, the mRNA expression of Vegfa165 and MyHC-IIx in EDL of ApoE −/− HIIT mice was significantly lower than those of ApoE −/− MICT mice ( Figure 2B,D). There were no significant differences in MyHC-IIxa between the ApoE −/− CON group and WT mice and among CON, MICT, and HIIT groups of ApoE −/− mice ( Figure 2C).

Muscle ROS, Protein Carbonyl, 4-HNE Modified Proteins, and the mRNA Expression Levels of Nox2, p47phox and Nox4
There were higher levels of ROS and protein carbonyl and mRNA expression of Nox2 and Nox4 in skeletal muscles of ApoE −/− CON mice than those of WT mice ( Figure  3A,B,D,F); while the expressions of 4-HNE protein and p47 phox mRNA were not different between the ApoE −/− CON group and WT mice and among CON, MICT, and HIIT groups of ApoE −/− mice ( Figure 3C,E). Moreover, the treatments of HIIT and MICT resulted in significantly lower levels of the muscle ROS, protein carbonyl, and mRNA expression of Nox4, compared with those of the ApoE −/− CON mice ( Figure 3A,B,F). In addition, ROS level was significantly lower and the mRNA expression of Nox2 and Nox4 were significantly higher in gastrocnemius muscles of ApoE −/− HIIT mice, compared with those of ApoE −/− MICT mice ( Figure 3A,D,F). There were higher levels of ROS and protein carbonyl and mRNA expression of Nox2 and Nox4 in skeletal muscles of ApoE −/− CON mice than those of WT mice ( Figure 3A,B,D,F); while the expressions of 4-HNE protein and p47 phox mRNA were not different between the ApoE −/− CON group and WT mice and among CON, MICT, and HIIT groups of ApoE −/− mice ( Figure 3C,E). Moreover, the treatments of HIIT and MICT resulted in significantly lower levels of the muscle ROS, protein carbonyl, and mRNA expression of Nox4, compared with those of the ApoE −/− CON mice ( Figure 3A,B,F). In addition, ROS level was significantly lower and the mRNA expression of Nox2 and Nox4 were significantly higher in gastrocnemius muscles of ApoE −/− HIIT mice, compared with those of ApoE −/− MICT mice ( Figure 3A,D,F).

The mRNA Expression of Genes Involved in the Production of GSH, GSH, GSSG Levels, and GSH/GSSG Ratio
The mRNA expression of genes (Gsr and Gclm) involved in the production of GSH was significantly lower in the skeletal muscle of ApoE −/− CON mice, compared to those of the WT mice ( Figure 4B,D). Moreover, the treatment of HIIT and MICT produced significantly higher expression in all measured genes involved in the production of GSH (Gss, Gsr, Gclc, and Gclm) in the skeletal muscle, compared with those of the Apo KO control mice ( Figure 4A-D).
The lower level of GSH and the higher level of GSSH resulted in a reduced GSH/GSSG ratio in the ApoE −/− CON group, compared to those of the WT mice ( Figure 4E-G). Moreover, the HIIT and MICT treatments did not exhibit a significant difference in the GSH/GSSG ratio, compared with that of the ApoE −/− CON group, although the GSH levels were increased in the ApoE −/− HIIT and ApoE −/− MICT groups, and GSSG level was decreased in the ApoE −/− HIIT group ( Figure 4E-G). In addition, there was no significant difference in these measured indexes between ApoE −/− HIIT and ApoE −/− MICT mice ( Figure 4A-G). Antioxidants 2021, 10, x FOR PEER REVIEW 8 of 15

The mRNA Expression of Genes Involved in the Production of GSH, GSH, GSSG Levels, and GSH/GSSG Ratio
The mRNA expression of genes (Gsr and Gclm) involved in the production of GSH was significantly lower in the skeletal muscle of ApoE −/− CON mice, compared to those of the WT mice ( Figure 4B,D). Moreover, the treatment of HIIT and MICT produced significantly higher expression in all measured genes involved in the production of GSH (Gss, Gsr, Gclc, and Gclm) in the skeletal muscle, compared with those of the Apo KO control mice ( Figure 4A-D).
The lower level of GSH and the higher level of GSSH resulted in a reduced GSH/GSSG ratio in the ApoE −/− CON group, compared to those of the WT mice ( Figure  4E-G). Moreover, the HIIT and MICT treatments did not exhibit a significant difference in the GSH/GSSG ratio, compared with that of the ApoE −/− CON group, although the GSH levels were increased in the ApoE −/− HIIT and ApoE −/− MICT groups, and GSSG level was decreased in the ApoE −/− HIIT group ( Figure 4E-G). In addition, there was no significant difference in these measured indexes between ApoE −/− HIIT and ApoE −/− MICT mice (Figure 4A-G).

The Protein Expression of Nrf2, p-Nrf2 (ser40), and Antioxidants
There were no significant differences in the protein expression of Nrf2, p-Nrf2 (ser40), and all measured antioxidants between ApoE −/− CON and WT mice ( Figure 5A-F). Moreover, there was significantly higher protein expression of p-Nrf2 (ser40) and GPX1 in ApoE −/− HIIT mice, and a prominently higher protein expression of NQO1 in ApoE −/− MICT mice, compared with ApoE −/− CON mice, respectively ( Figure 5B,D,E). There was no significant difference in these measured protein expression levels between ApoE −/− HIIT and ApoE −/− MICT ( Figure 5A-F).

The mRNA Expression of Fndc5, Hadh, Acads, and Hadha in Gastrocnemius, Plasma Irisin Level, as Well as Muscle Musclin Concentration
There was no significant difference in the mRNA expression of Fndc5, Hadh, Acads, and Hadha, and musclin content in the gastrocnemius, as well as plasma irisin level between the WT and ApoE −/− CON groups ( Figure 6A-F). However, significantly higher mRNA expression of Fndc5 ( Figure 6A), Hadh, and Hadha in the gastrocnemius ( Figure 6D,F), as well as plasma irisin level ( Figure 6B), were observed in the ApoE −/− MICT and ApoE −/− HIIT groups than those of the ApoE −/− CON group. Meanwhile, there was a significantly higher musclin level of the gastrocnemius of ApoE −/− MICT than that of the ApoE −/− CON group ( Figure 6C). In addition, there was a significantly higher mRNA expression of Hadha in the gastrocnemius of ApoE −/− HIIT mice than that of ApoE −/− MICT mice ( Figure 6F).

The Protein Expression of Nrf2, p-Nrf2 (ser40), and Antioxidants
There were no significant differences in the protein expression of Nrf2, p-Nrf2 (ser40), and all measured antioxidants between ApoE −/− CON and WT mice ( Figure 5A   There was no significant difference in the mRNA expression of Fndc5, Hadh, Acads, and Hadha, and musclin content in the gastrocnemius, as well as plasma irisin level between the WT and ApoE −/− CON groups ( Figure 6A-F). However, significantly higher mRNA expression of Fndc5 ( Figure 6A), Hadh, and Hadha in the gastrocnemius (Figure

Discussion
The main findings revealed that concurrent 6-week HIIT and MICT protocols improved blood lipid profiles, counteracted ROS production and protein carbonylation in the gastrocnemius muscle, and decreased the mRNA level of the angiogenic gene Vegfa165 in the EDL muscle. At the same time, both HIIT and MICT enhanced the GSH generation and potently promoted mRNA expression of genes involved in the production of irisin and BAIBA in the gastrocnemius muscle of ApoE KO mice. Comparison of the two training outcomes indicated that HIIT was more efficient than MICT in decreasing the ROS level of the skeletal muscle, whereas MICT was more efficient in increasing the plasma HDL-C level. To our knowledge, this is the first study to report and compare that HIIT and MICT, as potential adjuvant treatments, can attenuate oxidative damage and promote the myokine response in skeletal muscles of ApoE KO mice on HFD. These results support the hypothesis of the present study.
Our results of body weight and blood lipid profiles were in line with previous studies with ApoE KO mice on HFD exhibiting significant increases in plasma TC, TG, and LDL-C levels [33] but not becoming obese [24,34]. This may come from both lower synthesis and increased hydrolysis of triacylglycerols from the ApoE −/− adipocytes [34,35]. It confirms that this mouse model is a valid model of hyperlipidemia. The present study also provided relevant evidence that ApoE KO with HFD resulted in increased mRNA expression of Vegfa165, and decreased mRNA expression of MyHC-IIb, especially in EDL. These changes in the skeletal muscle represent a functional adaptation to a hyperlipidemic environment or compensation for the excess fat. Furthermore, the ROS production and protein carbonyl in the ApoE −/− CON group were higher, and the value of GSH generation was lower as compared with the WT group, confirming that hyperlipidemia increases oxidative stress in the skeletal muscle. This finding was further supported by the evidence of increased mRNA expression of Noxs genes (Nox2, p47 phox, and Nox4) in response to ApoE KO on HFD. The ApoE −/− CON mice of the present study showed increased markers of muscular oxidative stress, the same as the mouse model of the previous studies [16,24]. However, our ApoE −/− CON mice did not have a higher level of 4-HNE-modified proteins (the end products of lipid peroxidation) in the skeletal muscle, compared to that of the WT mice.

Discussion
The main findings revealed that concurrent 6-week HIIT and MICT protocols improved blood lipid profiles, counteracted ROS production and protein carbonylation in the gastrocnemius muscle, and decreased the mRNA level of the angiogenic gene Vegfa165 in the EDL muscle. At the same time, both HIIT and MICT enhanced the GSH generation and potently promoted mRNA expression of genes involved in the production of irisin and BAIBA in the gastrocnemius muscle of ApoE KO mice. Comparison of the two training outcomes indicated that HIIT was more efficient than MICT in decreasing the ROS level of the skeletal muscle, whereas MICT was more efficient in increasing the plasma HDL-C level. To our knowledge, this is the first study to report and compare that HIIT and MICT, as potential adjuvant treatments, can attenuate oxidative damage and promote the myokine response in skeletal muscles of ApoE KO mice on HFD. These results support the hypothesis of the present study.
Our results of body weight and blood lipid profiles were in line with previous studies with ApoE KO mice on HFD exhibiting significant increases in plasma TC, TG, and LDL-C levels [33] but not becoming obese [24,34]. This may come from both lower synthesis and increased hydrolysis of triacylglycerols from the ApoE −/− adipocytes [34,35]. It confirms that this mouse model is a valid model of hyperlipidemia. The present study also provided relevant evidence that ApoE KO with HFD resulted in increased mRNA expression of Vegfa165, and decreased mRNA expression of MyHC-IIb, especially in EDL. These changes in the skeletal muscle represent a functional adaptation to a hyperlipidemic environment or compensation for the excess fat. Furthermore, the ROS production and protein carbonyl in the ApoE −/− CON group were higher, and the value of GSH generation was lower as compared with the WT group, confirming that hyperlipidemia increases oxidative stress in the skeletal muscle. This finding was further supported by the evidence of increased mRNA expression of Noxs genes (Nox2, p47 phox, and Nox4) in response to ApoE KO on HFD. The ApoE −/− CON mice of the present study showed increased markers of muscular oxidative stress, the same as the mouse model of the previous studies [16,24]. However, our ApoE −/− CON mice did not have a higher level of 4-HNE-modified proteins (the end products of lipid peroxidation) in the skeletal muscle, compared to that of the WT mice.
It is well-known that regular exercise has preventive effects on various organs of atherosclerosis-prone ApoE KO mice [33,36]. However, until now, there were no reports about the benefits of exercise on the skeletal muscle of ApoE KO mice on HFD. Investigating the modalities of exercise treatment (i.e., exercise duration and exercise intensity) is therefore paramount when evaluating its effects. In the present study, we applied HIIT and MICT programs and found that both of them improved plasma lipid profiles and counteracted the compensatory enhanced EDL capillarization caused by ApoE KO with HFD. Importantly, while ApoE −/− CON mice impaired muscle redox homeostasis, the two training programs did attenuate the oxidative damage as shown by decreased ROS production, protein carbonyl content, and mRNA expression of Nox4, and also increased mRNA expression of genes involved in GSH production, GSH level, and the protein expression of some antioxidase in the skeletal muscle of these mice. This means that the two training modalities could induce adaptive responses, which were beneficial for the organism.
We found that ROS level was remarkably lower in HIIT mice than MICT mice, which implied that the magnitude of training adaptation was in part dependent upon exercise intensity, so that higher training intensities induced greater changes in the antioxidant defense [37], although there were no significant differences in measured variables of pro/antioxidant balance, including the levels of protein carbonyl, 4-HNE, GSH, and the protein expression of some antioxidase in the skeletal muscle between the two groups. Even for the Noxs, a key ROS generator during muscle contractions, the mRNA expression of Nox2 and Nox4 was significantly higher in ApoE −/− HIIT mice than ApoE −/− MICT mice. It indicated that in the skeletal muscle of HIIT mice, the antioxidant system could rapidly remove ROS before they caused cellular dysfunction and was more robust than that of the MICT group. As shown in Figure 5, HIIT mice had a significantly higher protein expression of p-Nrf2 (ser40) in the skeletal muscle, whereas MICT mice did not, compared with the ApoE −/− CON group. Therefore, based on the current results, we speculated that the lower ROS level could be closely linked to the higher protein expression of p-Nrf2 (ser40) in HIIT mice. However, there might be other proteins and muscle antioxidant enzymes, other than our measured antioxidants, involved in reducing the muscle ROS level of the HIIT group. Further research is needed to address them.
Furthermore, in the present study, it was found that the mRNA expression of MyHC-IIx and Vegfa165 in EDL of ApoE −/− HIIT mice was significantly lower than those of ApoE −/− MICT mice. This result may suggest that HIIT could have more potent effects on the resistance to the transition to slower myofibers and enhanced capillarization caused by ApoE deficiency and HFD. However, it was interesting to note that HIIT was not superior to MICT in altering blood lipids of ApoE KO mice on HFD, especially in the change of HDL-C. The change of HDL-C seems to be sensitive to training volume rather than exercise intensity.
Analyses of the skeletal muscle secretome revealed that numerous myokines are produced in response to muscle contraction, and then these factors not only regulate energy demand, but also contribute to the broad beneficial effects of exercise [27]. Myokines may be useful biomarkers for monitoring exercise prescription [38]. It has been reported that endurance exercise training upregulates peroxisome proliferator-activated receptor coactivator 1 (PGC-1) in the skeletal muscle [39,40] and the PGC-1α overexpression in the skeletal muscle increases the production of Fndc5, a precursor form of irisin, and irisin then stimulates the transformation of white adipose tissue to brown adipose tissue [41]. A prospective population-based study showed that higher serum irisin levels are associated with lower prevalence and progression of coronary atherosclerosis [42]. Protective effects of irisin on atherosclerosis were reported in two different ApoE KO mouse models [43,44]. BAIBA was also revealed to induce browning of the white adipocyte and stimulate hepatic β-oxidation. In humans, plasma BAIBA levels were increased with exercise and inversely associated with metabolic risk factors, such as fasting glucose, insulin, homeostasis model assessment of insulin resistance (HOMA-IR), and the levels of TG and TC [45]. In addition, musclin is an exercise-stimulated myokine [46] and its expression level is tightly regulated by nutritional changes, and its physiological role could be linked to glucose metabolism [47]. In the present study, we did not find significant changes in the mRNA expression of Fndc5 and genes required for BAIBA biosynthesis (Hadh, Acads, and Hadha), musclin content in the skeletal muscle, and plasma irisin level between ApoE −/− CON and WT mice. Meanwhile, the mRNA expression of Fndc5, Hadh, and Hadha, and musclin content in the skeletal muscle and blood irisin were upregulated in response to HIIT or MICT. Our results are in accordance with previous research that showed that muscle contraction stimulated myokine (irisin, BAIBA, and musclin) production [46,48,49]. However, it is worth noting that the mRNA expression of Hadha, the key gene involved in BAIBA biosynthesis, in the skeletal muscle of HIIT mice, was significantly higher than that in the MICT group. Since few studies have compared the effects of HIIT and MICT training programs on BAIBA production in muscle tissue, it was only speculated that the HIIT could be superior to MICT in BAIBA production, although BAIBA content in the skeletal muscle was not measured directly in the present study.
There are some limitations in this study. We only investigated the impacts of HIIT and MICT on ApoE KO mice with HFD for six weeks. Future studies should consider a longer duration, such as 12 weeks. We also only investigated the mRNA expression of many genes, such as MyHC-IIa, MyHC-IIx, and MyHC-IIb, but the immunohistochemical staining or Western blots would provide further results on the morphological changes. In addition, we focused on the changes in protein expression of antioxidant enzymes in the skeletal muscle, but we did not measure the possible changes in their activity. Further studies are also needed to determine whether antioxidant enzyme activity changes for a comprehensive evaluation of the pro-/anti-oxidant balance.

Conclusions
Six weeks of HIIT or MICT programs exerted beneficial effects on ApoE KO mice on HFD by attenuating oxidative damage and promoting myokines production in the skeletal muscle. Both training modalities could decrease plasma TC and TG levels, ROS production, and protein carbonylation in the skeletal muscle; simultaneously, they increased GSH generation and mRNA expression of genes involved in the production of irisin and BAIBA. Furthermore, HIIT was more beneficial than MICT for reducing the ROS level in the skeletal muscle.
Author Contributions: G.P.M. and M.P. designed the experiment; L.W. and J.L. performed experiments and they contributed equally to this work as joint first authors; W.H., Y.W., L.G., and H.W. helped with experiments; L.W. and Y.W. analyzed data; Y.Z. wrote the manuscript; J.W., G.P.M., and M.P. edited and revised manuscript; G.P.M. and Y.Z. obtained funding. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Bilateral Science and Technology Cooperation Program with Asia and Sino Swiss Science and Technology Cooperation for the support of JL and the activities leading to this publication. Moreover, the funding for ApoE KO mice with a high-fat diet was provided by the Institute of Sport Sciences of the University of Lausanne. The WT mice and laboratory experiments were funded by a grant from Beijing Sport University (2020ZJ007).

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of the Institute of Sport Sciences of the University of Lausanne for the ApoE KO mice with high-fat diet experiments; and the Animal Care and Use Committee of Beijing Sport University for the WT mice and laboratory experiments (2020ZJ007).

Data Availability Statement:
The data used to support the findings of this study are available from the corresponding author upon request.