Muscles Susceptibility to Ischemia-Reperfusion Injuries Depends on Fiber Type Specific Antioxidant Level

Muscle injury resulting from ischemia-reperfusion largely aggravates patient prognosis but whether and how muscle phenotype modulates ischemia-reperfusion-induced mitochondrial dysfunction remains to be investigated. We challenged the hypothesis that glycolytic muscles are more prone to ischemia-reperfusion-induced injury than oxidative skeletal muscles. We therefore determined simultaneously the effect of 3 h of ischemia induced by aortic clamping followed by 2 h of reperfusion (IR, n = 11) on both gastrocnemius and soleus muscles, as compared to control animals (C, n = 11). Further, we investigated whether tempol, an antioxidant mimicking superoxide dismutase, might compensate a reduced defense system, likely characterizing glycolytic muscles (IR-Tempol, n = 7). In the glycolytic gastrocnemius muscle, as compared to control, ischemia-reperfusion significantly decreased mitochondrial respiration (−30.28 ± 6.16%, p = 0.003), increased reactive oxygen species production (+79.15 ± 28.72%, p = 0.04), and decreased reduced glutathione (−28.19 ± 6.80%, p = 0.011). Less deleterious effects were observed in the oxidative soleus muscle (−6.44 ± 6.30%, +4.32 ± 16.84%, and −8.07 ± 10.84%, respectively), characterized by enhanced antioxidant defenses (0.63 ± 0.05 in gastrocnemius vs. 1.24 ± 0.08 μmol L−1 g−1 in soleus). Further, when previously treated with tempol, glycolytic muscle was largely protected against the deleterious effects of ischemia-reperfusion. Thus, oxidative skeletal muscles are more protected than glycolytic ones against ischemia-reperfusion, thanks to their antioxidant pool. Such pivotal data support that susceptibility to ischemia-reperfusion-induced injury differs between organs, depending on their metabolic phenotypes. This suggests a need to adapt therapeutic strategies to the specific antioxidant power of the target organ to be protected.

Albeit necessary, reintroduction of oxygenated blood into the ischemic areas induces unfortunately ischemia-reperfusion (IR) injury. This is significant, aggravating both local and general prognosis of the patient (Harris et al., 1986;Blaisdell, 2002;Martou et al., 2006) and supporting continuous work in order to reduce these deleterious effects.
Recently major advances have been observed in our knowledge of lower limb IR pathophysiology. Notably, interactions between muscle mitochondria and reactive oxygen species (ROS) have been emphasized . Oxidative stress, which precedes mitochondrial dysfunction, arises during ischemia and is enhanced after reperfusion . This suggests that modulating ROS production may reduce IR-induced injury. Accordingly, we and others demonstrated that muscle mitochondrial function is protected when ROS production is reduced and/or when the ROS produced can be efficiently handled by the antioxidant system. Such therapeutic approaches were based on both ischemic and pharmacologic conditioning (Ali et al., 2007;Makris et al., 2007;Charles et al., 2011;Tran et al., 2011;Talha et al., 2013).
Interestingly, however, interactions between mitochondria and ROS are more subtle. Significant increases in ROS production are generally considered deleterious because it induces lipid peroxidation, protein carbonylation, and mitochondrial dysfunction that is associated with reduced ATP production (Anderson and Neufer, 2006;Magder, 2006). On the other hand, a small ROS increase is a signaling pathway stimulating mitochondrial biogenesis and enhancing antioxidant defenses (Ristow, 2014;Yun and Finkel, 2014). Recently, we reported that skeletal muscles are more prone to statin-induced injury than cardiac muscles . Metabolic phenotypes greatly differ between cardiac and skeletal muscles. Indeed, heart is highly oxidative, in association with enhanced mitochondrial respiration and antioxidant system (Masuda et al., 2003;Kowaltowski et al., 2009;Bouitbir et al., 2011).
Similar differences exist among skeletal muscles, which are classified mainly depending on their contractile and/or metabolic phenotype. Mammalian skeletal muscles are characterized by their abilities to adapt to new conditions. Indeed, changes can be detected by multiple sensors, from membrane receptors for hormones and cytokines, to metabolic sensors that detect variations in the high-energy phosphate concentration, oxygen as well as oxygen free radicals. These sensors trigger cascades of signaling pathways which may ultimately lead to changes in fiber type, especially at the level of metabolic properties (for review see Pourova et al., 2010;Schiaffino and Reggiani, 2011;Blaauw et al., 2013;. Particularly, certain fibers possess a large amount of mitochondria, and rely mainly on oxidative phosphorylation. These fibers can be defined as "oxidative:" they require more oxygen and depend hence on the vascularisation to synthesize ATP. On the other hand, other fibers possess few mitochondria, and mainly rely on glycolysis to produce ATP. These fibers can be identified as "glycolytic" (Jackman and Willis, 1996;Meyer et al., 2014). Moreover, the analysis of the expression myosin heavy chains enables to differentiate four subtypes of slow (Type I) and fast twitch fibers: IIA, IIX, and IIB (Talbot and Maves, 2016). Type I fiber, such as the soleus, is considered as a slow oxidative fiber with high baseline respiratory rates and important antioxidant pools. Type IIA fibers use both oxidative and glycolytic metabolisms to produce ATP, and have a relatively high mitochondrial content (however lower than type I fibers). These fibers are relatively resistant to fatigue (Schiaffino, 2010). Type IIB fibers possess a low amount of mitochondria and rely mainly on glycolysis. These fibers has less antioxidant enzymes compared with type I fibers (Powers and Hamilton, 1999;Ploquin et al., 2012) and are easily fatigable, but develop tremendous amounts of strength. Type IIX fibers possess an intermediate phenotype between type IIA and type IIB fibers. Based on these characteristics, and on the fact that superficial gastrocnemius present with 97% of type IIb fibers and that soleus present with 30.6% of type I fibers (Bloemberg and Quadrilatero, 2012), we decided to study both superficial gastrocnemius and soleus in this study.
Interestingly, although Turoczi et al. described a type I muscle altered viability (Turóczi et al., 2014), other studies rather supported glycolytic muscles alterations after IR. Particularly, considering contractile force, Demirel et al. observed differences between glycolytic and oxidative muscles submitted to IR. After 4 h ischemia and 2 h reperfusion, the force-frequency curve was more altered in extensor digitorum longus (EDL) than in soleus. Further, resistance to fatigue was lower in EDL than in soleus (Demirel et al., 2013). Impaired fatigue resistance and contraction were also observed after IR in gastrocnemius (Zhang et al., 2017) and accordingly, proteins implicated in muscle contraction are known to be targets for ROS (Beckendorf and Linke, 2015), which are largely implicated in IR pathophysiology (increased lipid peroxidation and oxidized proteins; Avci et al., 2012).
Considering that oxidative skeletal muscles can also suffer from intense IR (Ali et al., 2010;Balogh et al., 2014;Turóczi et al., 2014;He and Zuo, 2015), we aimed to challenge the hypothesis their susceptibility to IR injuries might depend on their metabolic phenotypes. Particularly, oxidative skeletal muscles (as compared to glycolytic ones) should be protected from IR injuries, thanks to their antioxidant defenses.

Ethical Approval
Procedures were conducted in accordance with US National Institutes of Health guidelines. The study was approved by the institutional animal care committee of the University

Animals
We worked on adult male Wistar rats (Depré, Saint-Doulchard, France) aged 8 weeks. The temperature environment is regulated at 22 ± 2 • C, and the room is submitted to a 12-h light-dark cycle. Animals had free access to food and water.

Preoperative Management
The anesthesia was induced in a hermetic cage with 4% isoflurane (Aerrane, CSP, Cournon, France) and oxygen. During the operation, the animals ventilated spontaneously, with an oxygen-delivering mask, and adapted isoflurane concentrations. The body temperature was controlled and maintained (Homeothermic blanket control unit, MINERVE, Harvard Apparatus R , Esternay, France).

Surgical Procedures and Experimental Design
Twenty nine rats were divided in three groups (Figure 1): the control group (C, n = 11) underwent 5 h of general anesthesia. A midline laparotomy was performed and the abdominal aorta was exposed like in the IR groups.
The ischemia-reperfusion (IR) group (n = 11) underwent 3 h of ischemia induced by infra-renal aortic occlusion, and collateral vessel coagulation (GEIGER R , thermal cautery unit, Geiger Medical Technologies, Council Bluffs, Iowa, USA) followed by 2 h of reperfusion. As previously reported, ischemia was clinically characterized by cyanosis and lack of arterial pulse distal to the clamp and reperfusion was associated with recoloring and pulse return in the hind limb (Thaveau et al., 2007;Mansour et al., 2011;Pottecher et al., 2013).
Duration of ischemia was based on our and previous data demonstrating that 3 h ischemia is needed to result in significant skeletal muscle injury, which is still more severe after 6 h ischemia (Belkin et al., 1988;Thaveau et al., 2007). Chronological sequences about the cellular and molecular events occurring during skeletal muscle IR has been recently reported (Paradis et al., 2016). Interestingly, as compared to the heart which is still beating, such a long duration might be explained by the noncontracting, resting skeletal muscle generally examined. Like in the heart however (Ma et al., 2016), a no reflow occurrence might increase with long ischemia time and we therefore choose a 3 h ischemia period for our experimental protocol.
At the end of reperfusion, lactate, which is considered as a good marker of anaerobic state, was determined on total blood obtained from the tail before and after ischemia (lactate Pro device, LT710, Arkray R , KGK, Kyoto, Japan). Its decrease after opening the artery supports a successful reperfusion (Noll et al., 2012).
At the end of experiment, oxidative soleus and glycolytic superficial gastrocnemius muscles were excised (Bloemberg and Quadrilatero, 2012). For each muscle, one part was used immediately for oxidative capacities and ROS determinations. The second part was frozen in isopentane cooled by liquid nitrogen or directly in liquid nitrogen and thereafter stored for a later use.
Finally, to identify the role of oxidative stress in IR-related injury, and particularly to determine the potential protective effect of an anti-oxidant in gastrocnemius, a third group was given the superoxide dismutase mimetic. Rats from the Tempol-IR group (n = 7) had two intraperitoneal injections of 4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl (Tempol, Sigma-Aldrich R , St. Louis, MO, USA); 250 mg kg −1 at 10 min before the beginning of reperfusion and 125 mg kg −1 after 1 h of reperfusion (Halter et al., 2010).

Study of Muscle Mitochondrial Respiration
All used products were purchased from Sigma-Aldrich R (St. Louis, MO, USA).
We studied the mitochondrial respiration of skinned fibers rather than isolated mitochondria to preserve the environment and the integrity of mitochondria (Saks et al., 1998;Rasmussen and Rasmussen, 2000). The two muscles studied were superficial gastrocnemius and soleus muscles.
Then, complex I was blocked with amobarbital (0.02 mmol L −1 ), and complex II was subsequently stimulated (V succ ) with succinate (25 µmol L −1 ). The electron flow went through complexes II, III, IV, and ATP synthase. Finally, N,N,N ′ ,N ′ -tetra-methyl-p-phenylenediamine dihydrochloride (TMPD, 0.5 mmol L −1 ) and ascorbate (0.5 mmol L −1 ) were added as artificial electron donors to cytochrome c, to study complex IV activity. After the experiment, fibers were harvested and dried for 15 min at 150 • C, and respiration rates were expressed as µmol O 2 min −1 g −1 dry weight.

Electron Paramagnetic Resonance
Although one method might have been sufficient, we decided to perform two complementary methods for ROS measurement in order to comfort the data. DHE staining is very useful and allows demonstrate the ROS level in tissue and EPR is viewed as an excellent technique, measuring the ROS production by the tissue (Dikalov et al., 2007;Kuznetsov et al., 2011;Lejay et al., 2015).
All used substrates for the Krebs Hepes buffer were purchased from Sigma-Aldrich R (St. Louis, MO, USA).
The amplitude of the signal is measured, and the concentration of CM-radical is determined by calibration with standard concentrations of the radicals CM. After the EPR measurement, pieces of muscles are dried for 15 min at 150 • C, and O .− 2 production was expressed in µmol min −1 mg −1 dry weight.

Dihydroethidium Staining
Dihydroethidium staining (DHE, Sigma-Aldrich R ) allows detecting the presence of ROS in skeletal muscles.
Frozen muscles were cut on serial sections (10 µm-thicks) with a cryostat microtome and thaw mounted on glass slides and incubated with DHE (2.5 µmol L −1 ). DHE produced red fluorescence when oxidized to ethidium bromide (EtBr) by O .− 2 (Mulsch et al., 2001;Talha et al., 2013;Guillot et al., 2014). As compared to other markers, the interest of DHE is its relatively good specificity to O .− 2 . After staining, sections were examined under an epifluorescence microscope (Eclipse E800, Nikon, New York, USA) with a 20X epifluorescence objective. The emission signal was recorded with a Zeiss 573-637 nm filter. We analyzed micrographs with Adobe Photoshop 6.0 (USA). The results were reported in percentage of control group.

Measurement of Reduced Glutathione
The glutathione peroxidase is an important intracellular antioxidant, because it is the primary enzyme responsible for H 2 O 2 detoxification in mitochondria (Antunes et al., 2002;Zoccarato et al., 2004).
All used products were purchased from Sigma-Aldrich R (St. Louis, MO, USA).
Reduced glutathione (GSH) was measured in the different muscles by monitoring the reduction of 5,5 ′ -dithio-bis-(2nitrobenzoic acid) (DTNB) to 5-thio-2-nitrobenzoate (TNB) in presence of GSH at 412 nm (Akerboom and Sies, 1981;Bouitbir et al., 2011). Briefly, to determine GSH content, tissue samples were mixed with cold 5% 5-sulfosalicylic acid, after 10 min of incubation on ice, the mixture was centrifuged at 4000 rpm for 10 min at 4 • C to remove proteins. The samples were then diluted in cold MOPS buffer, at room temperature. The diluted sample was added to glutathione reductase (24 units mL −1 ) and MOPS (0.1 mol L −1 ) containing 1 mmol L −1 EDTA, 0.3 mmol L −1 NADPH, and 0.22 mmol L −1 DTNB. The absorbance was continuously monitored for 1 min, as an index of DTNB reduction by GSH contained in the sample. Reduced glutathione was expressed in µmol L −1 g −1 .

Statistical Analysis
Values are expressed as mean ± SEM. n indicates the number of animals. Statistical analyses were performed with Student's t-test for unpaired data. When the variances were significantly different, we used the Student's unpaired t-test with Welch's correction. For the lactate rates comparison statistical significance was measured by ANOVA followed by the post-test Newman Keuls (GraphPad Prism 5, Graph Pad Software, Inc., San Diego, CA, USA). Values of p < 0.05 were considered to be statistically significant.

IR Impaired Mitochondrial Respiration, Increased ROS Production, and Reduced Antioxidant Defenses in Glycolytic
Gastrocnemius Muscles (Figure 3) Mitochondrial Respiratory Chain Complex Activities ( Figure 3A) In gastrocnemius, maximal respiration rate (V max ), reflecting the activity of complexes I, III, IV, and V, was significantly altered after IR as compared to the control group (4.51 ± 0.44 vs. 6.61 ± 0.42 µmol O 2 min −1 g −1 dry weight, respectively, p < 0.05).
Thus, IR impaired all complexes of the mitochondrial respiratory chain of the gastrocnemius muscle.
This result was confirmed in the graph C representing DHE staining. Indeed, ROS content was increased after IR as compared to the control group (36,860 ± 8127 vs. 16,720 ± 3674 UA, respectively, p = 0.04).

Effects of IR on the Oxidative Soleus Muscle (Figure 4)
Mitochondrial Respiratory Chain Complex Activities ( Figure 4A) Less-IR induced damage is observed in oxidative muscle. In soleus, the maximal respiration rate (V max ), although showing a slight tendency to decrease, was not significantly altered after IR as compared to the control group (8.94 ± 0.60 vs. 9.56 ± 0.58 µmol O 2 min −1 g −1 dry weight, respectively, p = 0.469).
Such lack of oxidative skeletal muscle alteration was associated with an absence of oxidative stress.
Finally, the antioxidant defenses were reduced by IR only in the gastrocnemius as compared to the soleus (−28.20 ± 6.80 vs. −8.07 ± 10.84%, respectively, p = 0.148, Figure 5D). (B,C) ROS production in the gastrocnemius: superoxide anion production measured by EPR with a specific probe CMH (B) and intensity of fluorescence labeled with oxidative dye DHE (C). (D) Reduced glutathione in the gastrocnemius. The black bars represent the C group (control group) and the gray bars represent the IR group (3 h of ischemia and 2 h of reperfusion). V max : complexes I, III, IV, V activities; Vsucc: complexes II, III, IV, V activities; V TMPD/asc : complex IV activity. Results were expressed as mean ± SEM. *p < 0.05, **p < 0.01 compared to the control group. $$ p < 0.01, $$$ p < 0.001 compared to the IR group. C, Control group; IR, Ischemia-reperfusion group. IR+T: TEMPOL-IR group.

Metabolic Phenotypes Implications in Muscle Susceptibility to IR
The metabolic phenotype differs between gastrocnemius and soleus muscles and might therefore explain their specific sensitivity to IR.
Indeed, as compared to oxidative ones, glycolytic muscles are characterized by lower mitochondrial respiratory chain complex activities. This holds true for all complexes. Thus, V max , V succ , and V TMPD/Asc are smaller in gastrocnemius as compared to soleus (6.6 ± 0.4 vs. 9.6 ± 0.6; 6.2 FIGURE 4 | Ischemia-reperfusion effects on the oxidative soleus muscle. (A) Mitochondrial respiration in the soleus. (B,C) ROS production in the soleus: superoxide anion production measured by EPR with a specific probe CMH (B) and intensity of fluorescence labeled with oxidative dye DHE (C). (D) Reduced glutathione in the soleus. The black bars represent the C group (control group) and the gray bars represent the IR group (3 h of ischemia and 2 h of reperfusion). V max : complexes I, III, IV, V activities; Vsucc: complexes II, III, IV, V activities; V TMPD/asc : complex IV activity. Results were expressed as mean ± SEM. C, Control group, IR, Ischemia-reperfusion group.
In view of these data we tested the hypothesis that the reduced antioxidant defenses of gastrocnemius muscle might be overcome by administration of tempol, an efficient antioxidant, and therefore confer protection against IR to this glycolytic muscle.
With tempol, mitochondrial oxidative capacities were no longer impaired by IR in gastrocnemius ( Figure 3A).

DISCUSSION
The main result of this study is that skeletal muscle metabolic phenotype importantly modulates deleterious effects of lower limb IR. Thus, as compared to the glycolytic gastrocnemius, the oxidative soleus appeared protected and demonstrated no significant mitochondrial dysfunction. In soleus, higher antioxidant capacity likely participates in such protection since oxidative stress is a key factor inducing muscle impairment during IR. Accordingly, when treated with an antioxidant, gastrocnemius muscle acted as soleus, showing no more significant alteration although submitted to the same IR protocol.
Vascular ischemia is very common and corresponds to a public health issue both because of its increasing occurrence and its very bad prognosis (Yassin et al., 2002). Thus, major amputation is still needed in 25-40% of cases, and the risk of death is as high as 25% after the first clinical sign of peripheral arteriopathy (Norgren et al., 2007;Taylor et al., 2009;Ryan et al., 2015). Interestingly, muscle alteration significantly impacts both local and general prognosis. Thus, rhabdomyolysis is well-known to participate in remote organ alteration including kidneys, the heart and lungs (Fowkes et al., 2006;Ali et al., 2007;McMahon et al., 2013).
Although some therapeutic strategies have been shown effective, results are not always consistent, raising the issue of a specific organ dependency to IR (Mansour et al., 2014). Concerning skeletal muscles, metabolic phenotype is a good candidate since degrees of fiber injury vary among contractile muscle phenotypes (Chan et al., 2004). Woitaske et al. observed an increased mitochondrial swelling and disrupted sarcoplasmic reticulum in the glycolytic EDL as compared to soleus muscle (Woitaske and McCarter, 1998). Accordingly, IR-induced contractile dysfunction was recently shown to be reduced in soleus as compared to EDL (Demirel et al., 2013).
Based on different muscle phenotypes, several mechanisms might explain such potential increased sensitivity of glycolytic muscle to IR injury. Particularly, mitochondrial content and redox balance deserve discussion.
Mitochondrial density is about two-fold greater in oxidative than in glycolytic muscles (Jackman and Willis, 1996;Picard et al., 2008), likely participating in the enhanced mitochondrial respiration observed in soleus as compared to gastrocnemius muscles (Anderson and Neufer, 2006;Meyer et al., 2014). Recently, induction of mitochondrial biogenesis has been shown to protect against apoptosis in L6 myoblats (Dam et al., 2013), and we observed that decrease in PGC1 beta, which likely participate in mitochondrial biogenesis, resulted in a reduced adaptation of muscles to cope with high energy demands (Gali Ramamoorthy et al., 2015). Such relationship between the number of mitochondria and muscle responses to stress likely apply in the setting of IR. Accordingly, Flück et al. demonstrated that IR-induced skeletal muscle rhabdomyolysis is a fiber type-specific phenomenon, modulated by mitochondria reserves (Fluck et al., 2015).
However, besides mitochondrial respiration per-se, redox status appears to be a key factor in muscle functioning (for review see . Even during physiological contraction (eccentric and/or concentric), muscle mitochondria generate large amounts of ROS, that might allow either muscular adaptation or muscular damage. Thus, depending likely on ROS levels resulting from the pro-/antioxidant balance, consequences might be good (mitochondrial biogenesis, antioxidant pathway stimulation, enhanced glucose uptake), bad (mitochondrial dysfunctions, reduced ATP synthesis) or even ugly (mPTP opening and apoptosis) .
In this view, exercise is interesting to consider. Indeed, exercise-related beneficial effects are largely reduced when exercise-induced ROS signaling is decreased by enzymatic inhibitors or antioxidant supplementation (Meier et al., 2013;Strobel et al., 2014;Merry and Ristow, 2016). Further, the effects of exercise likely depend on muscle metabolic phenotype. For instance, exercise has a stronger influence on red muscle phenotype, as compared to white muscle in salmonids (Morash et al., 2014). Similarly, uphill-, as compared to down-hill training enhanced mitochondrial function specifically in both soleus and vastus intermedius skeletal muscle (Schlagowski et al., 2016).
Increased antioxidant defenses also characterize oxidative muscles. This likely explains why statins optimize cardiac mitochondria, but impairs skeletal muscle function by inducing different levels of ROS (Bouitbir et al., , 2015. Concerning skeletal muscles, in agreement with previous reports (Anderson and Neufer, 2006;Picard et al., 2012), we observed greater FIGURE 6 | Schema illustrating skeletal muscle susceptibility to ischemia-reperfusion, depending on metabolic phenotype. Oxidative skeletal muscles are characterized by increased mitochondrial content and enhanced antioxidant system allowing better protection against ischemia-reperfusion supported by maintained mitochondrial respiration and low reactive oxygen species (ROS) production. Antioxidant adjunction protects glycolytic muscle, further demonstrating the key role of oxidative stress handling and therefore muscle metabolic phenotype. SODs, Superoxide dismutases; GPX, glutathione peroxidase. glutathione peroxidase activity in the soleus than in the superficial gastrocnemius This support that the oxidative skeletal muscle may possess a greater capacity to scavenge ROS (Picard et al., 2012). Since enhanced ROS production occur early and participate significantly in IR deleterious effects (Tran et al., 2011;Guillot et al., 2014;Lejay et al., 2014), the fact that oxidative muscles naturally equip with higher antioxidant contents than glycolytic muscles likely protect them in the setting of IR. Further, an adaptation related to oxidative resistance might have been triggered during IR. Such a possibility, likely working in the setting of preconditioning, can perhaps not be totally developed in the relatively short time frame of our study but, one might suggest that the antioxidant pool located in the blood might also have been mobilized.
To further test that enhanced antioxidant defenses allow oxidative skeletal muscles being protected against IR, we determined whether the glycolytic muscle gastrocnemius might be protected against IR when its antioxidant defense is improved. Accordingly, tempol therapy significantly attenuated the deleterious effects of IR. Such data are in agreement with a previous work (Tran et al., 2011) and further suggest a key role of the antioxidant defenses and likely explain -at least partly-the specific muscle susceptibility to IR (Figure 6).
As perspectives, further studies should be able to demonstrate that ROS staining be mainly colocalized within type II fibers during IR. It might be useful to determine the molecular mechanisms involved in muscle protection, signaling ranging from mitochondrial biogenesis to atrophy-hypertrophy pathways. Besides ROS-activated signaling pathways such as MAPK, PI3K/Akt signaling, PKC, and inflammation signaling (Görlach et al., 2015), a special insight should be directed toward the SAFE and RISK pathways which modified interplay might account for increased susceptibility of diabetic muscles to ischemia .

CONCLUSION
This study supports that glycolytic muscles are more prone to IR than oxidative ones. Such muscle susceptibility appears linked to muscle metabolic phenotypes, involving mainly muscle mitochondrial oxidative capacity and antioxidant defenses. This might explain difficulties to convert experimental data well-focused on specific muscle fibers toward clinical setting, since human muscles show more phenotypic heterogeneity. More widely, these data suggest that therapeutic approaches should be adapted to target organ redox capacities (potentially modified by comorbidities), and that mechanisms increasing either mitochondrial function and/or antioxidant defense should allow better protection.