Monash University, Australia scientists have discovered an enzyme that is key to why exercise improves our health. Importantly this discovery has opened up the possibility of drugs to promote this enzyme’s activity, protecting against the consequences of aging on metabolic health, including type 2 diabetes.
The proportion of people worldwide over 60 years old will double in the next three decades and by 2031, more than six million Australians will be over 65 years old. The incidence of type 2 diabetes increases with age so this aging population will also result in an increased incidence of the disease globally.
One of the main reasons for the increased prevalence of type 2 diabetes with age is the development of insulin resistance, or an inability for the body to respond to insulin, and this is often caused by reduced physical activity as we age.
However, the precise mechanisms by which physical inactivity facilitates the development of insulin resistance has remained a mystery.
Now researchers from Monash University in Australia have discovered how physical activity actually enhances insulin responsiveness and in turn promotes metabolic health. Importantly, the enzymes they have discovered that are key to this mechanism have the potential to be targeted by drugs to protect against consequences of aging such as muscle wasting and diabetes.
The team of scientists at the Monash University Biomedicine Discovery Institute (BDI), led by Professor Tony Tiganis, reveals that reductions in skeletal muscle reactive oxygen species (ROS) generation during aging is instrumental in the development of insulin resistance. According to Professor Tiganis, skeletal muscle constantly produces ROS and this is increased during exercise.
“Exercise-induced ROS drives adaptive responses that are integral to the health-promoting effects of exercise,” he said.
In a paper published today (15 December) in the journal, Science Advances, the research team show how an enzyme called NOX-4 is essential for exercise-induced ROS and the adaptive responses that drive metabolic health.
In mice the researchers found that NOX4 is increased in skeletal muscle after exercise and that this then leads to increased ROS which elicits adaptive responses that protect mice from the development of insulin resistance, which otherwise occurs with aging or diet induced-obesity.
Importantly, the scientists have shown that the levels of NOX4 in skeletal muscle are directly related to age-associated decline in insulin sensitivity. “In this study we have shown, in animal models, that skeletal muscle NOX 4 abundance is decreased with aging and that this leads to a reduction in insulin sensitivity,” Professor Tiganis said.
“Triggering the activation of the adaptive mechanisms orchestrated by NOX4 with drugs, might ameliorate key aspects of aging, including the development of insulin resistance and type 2 diabetes,” he said.
“One of these compounds is found naturally, for instance, in cruciferous vegetables, such as broccoli or cauliflower, though the amount needed for anti-aging effects might be more than many would be willing to consume.”
Oxidation-reduction (redox) reactions occur ubiquitously in biology and are necessary to maintain an environment optimal for cell signaling and function. These redox reactions involve reactive oxygen species (ROS) such as superoxide (O2.-) and hydrogen peroxide (H2O2) which regulate several physiological and pathophysiology processes ranging from cell apoptosis and death, to cell differentiation and growth [, , , , , ].
In humans, some of the most well-known stimuli for inducing ROS production include postprandial substrate oxidation and the mechanical and physiological stress induced through exercise [, , , , , , , , ]. To maintain cellular redox homeostasis, numerous enzymatic and non-enzymatic antioxidant defenses have evolved to regulate cellular levels of ROS [1,2,, , , , , ].
Examples of enzymatic antioxidant sources include superoxide dismutase, glutathione peroxidase, and catalase, whereas examples of non-enzymatic sources include glutathione (GSH) and vitamins C and E. An increase in ROS or a decrease in antioxidant activity can lead to a cellular redox environment commonly referred to as “oxidative stress”.
Overwhelming evidence exists supporting the pathological role of ROS in the development of chronic conditions and diseases including insulin resistance and type 2 diabetes (T2D) [9,10,13,, , , , , ]. In stark contrast, accumulating evidence now supports the physiological role of ROS in maintaining cardiometabolic health and the prevention of conditions and diseases including insulin resistance and T2D [22,24,, , , , , , ].
Acknowledging this dual role, the simplified concept of oxidative stress as being exclusively “bad” has been vetoed by many in the field of redox biology and physiology in favor of the more recently preferred descriptive terms of oxidative distress and oxidative eustress [35,36].
Antioxidant is a broad all-encompassing term used to describe endogenous and exogenous compounds or biological processes that can directly (e.g., free radical scavenging) or indirectly (e.g., the regulation of ROS-producing and/or antioxidant enzymes) alter the oxidative dis/eustress redox environment .
Although the literature on the use of antioxidant treatment for decreasing “oxidative stress” continues to expand [8,22,37,38], a unifying consensus on the effectiveness of treatment for improving health and/or preventing disease has yet to arise. Inconsistent findings and a lack of consensus likely stem from generic antioxidant treatment practices which fail to acknowledge the dynamic and dual role of ROS in both human physiology and pathophysiology.
In this narrative review of the literature, we discuss previous and emerging research that has investigated oxidative stress, antioxidants and insulin and glucose regulation under various oxidative eustress and distress conditions, including acute and regular exercise, hyperglycemia, hyperlipidemia, and overt T2D.
Our discussion is focused on research in humans, however, where necessary we refer to findings from relevant animal models or in vitro studies that may provide a mechanistic basis for the observations in human studies. The redox health paradox is further explored through the discussion of antioxidant treatment under the varying conditions of oxidative distress and eustress, and the subsequent effects on insulin and glucose regulation. Finally, we synthesize and contextualize current antioxidant treatment evidence to help guide a new era of research – one that harnesses the potential of personalized antioxidant treatment to robustly explore redox regulation of human physiology.
Oxidative distress: implications for glycemic control and disease
Reactive oxygen species cause oxidative modification to lipids, proteins, and DNA. This can lead to impaired transcription and translational processes, altered protein function, and production of secondary by-products and metabolites which lead to further ROS production and/or cellular damage [, , , , , , , ].
One of the most well-documented stimuli for producing ROS in biological organisms is through postprandial metabolism of carbohydrates, lipids, and protein [9,22,23,47,48]. The detrimental effects of substrate metabolism on insulin and glucose regulation are largely thought to occur through excess lipid and carbohydrate intake [9,10,13,, , , , ].
Hyperglycemia and hyperlipidemia at rest (basal) or following meal ingestion (postprandial) can lead to increased cellular oxidative distress through mitochondrial electron leak and/or incomplete fatty acid oxidation, the formation of advanced glycation end products and other oxidation products including lipid hydroperoxides, and increased circulation of free fatty acids (FFA), diacylglycerol, and ceramides [9,10,23,, , ].
Pathways of oxidative distress-induced insulin resistance are reviewed in detail elsewhere [9,22,28]. In brief, hyperglycemia and hyperlipidemia-induced production of ROS and ROS byproducts can lead to impaired insulin sensitivity through structural and functional changes to the insulin molecule decreasing its bioactivity , and through activation of redox-sensitive cell signaling pathways that interfere with insulin signaling and glucose cell transport [22,28]. Redox-sensitive components of signaling pathways identified to contribute to insulin resistance include p38 mitogen-activated protein kinase (p38 MAPK), c-Jun N-terminal kinase (JNK), nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), extracellular-signal regulated kinase (ERK), and protein kinase C (PKC), and the lipid peroxidation product 4-hydroxy-2-nonenal (4-HNE) [22,24,27,28,, , , , , , , , , , ].
Activation of these pathways increase serine and threonine phosphorylation of the insulin receptor substrates (IRS) 1 and 2, inhibiting downstream insulin signaling through attenuated tyrosine phosphorylation and IRS proteasomal degradation and subcellular re-localization [22,24,28,, , , , , , , , , ].
Additionally, increased activity of the redox-sensitive protein tyrosine phosphatase (PTP) family also leads to inhibition of the insulin signaling pathway, as observed during high-lipid culture conditions in skeletal muscle cells . Insulin resistance and mitochondrial dysfunction promote a vicious cycle of increased hyperglycemia and hyperlipidemia, ROS production, impaired insulin sensitivity, and the eventual development of overt T2D and other cardiometabolic diseases [22,28,51].
Although the authors acknowledge the complex and integrated nature of tissue-specific regulation of glucose homeostasis and ROS production, the current review focuses predominantly on skeletal muscle as it is one of the major sites for insulin dependent and independent glucose disposal [63,64].
Oxidative distress and impaired glycemic control.
Elevated biomarkers of systemic oxidative stress and/or decreased antioxidant activity in blood, urine, or muscle, are reported among individuals who are sedentary or obese, have impaired glucose tolerance (IGT), and/or have T2D [10,, , , , , ].
For example, skeletal muscle mitochondrial H2O2 emissions in obese individuals at rest (fasting basal conditions) are two-fold higher compared to lean individuals . Furthermore, both the ingestion of a single high-fat meal or a 5-day high-fat diet in healthy individuals leads to increased muscle oxidative stress, as measured by a decrease in the GSH and oxidized glutathione (GSSG) ratio and elevated skeletal muscle mitochondrial H2O2 production, to similar levels of that observed in obese insulin-resistant individuals .
Additional experiments revealed that comparable high-fat conditions in rodents lead to a similar shift towards a more oxidative environment in muscle alongside the development of insulin resistance . Human studies have also reported greater measures of systemic oxidative stress and lower antioxidant activity in blood from T2D and IGT individuals at rest (fasting) and following high-fat meal ingestion compared to healthy controls , supporting the notion of increased oxidative distress in populations characterized by insulin resistance.
However, oxidative distress and insulin resistance can also occur in healthy individuals. For example, independent of health status, markers of systemic oxidative distress can remain elevated for up to 4 h following the ingestion of pure carbohydrate [71,72], fat, or protein , and mixed macronutrient meals containing high-fat [, , , , , ] and high-carbohydrate content .
Larger meals and meals higher in lipid content generally elicit greater postprandial oxidative stress [80,81], which are proposed to contribute to the development of acute and chronic insulin resistance [9,10,23,51]. Transient hyperlipidemia in healthy humans elicited by intravenous lipid infusion for 6 h leads to a dose-dependent increase in plasma free fatty acids (FFA), glucose, insulin and oxidative distress as measured by increased plasma thiobarbituric acid reactive substances (TBARS) and a decrease in GSH/GSSG ratio .
Furthermore, under the same 6 h intravenous lipid infusion conditions, insulin stimulated whole-body glucose uptake measured via euglycemic hyperinsulinemic clamp is impaired . Unfortunately, due to the indirect nature of manipulating the cellular redox environment, these findings in humans are unable to directly demonstrate a causal relationship between increased oxidative distress and impaired insulin sensitivity.
This is especially true given that ROS are necessary for insulin signal transduction and glucose cell transport [30,31,83,84]; prohibiting the conclusion that systemic oxidative stress measured during conditions of excess lipid and carbohydrate intake in humans is exclusively attributed to pathological ROS production. Nevertheless, findings in humans are consistent with findings from rodent and cell culture studies that have directly manipulated the ROS and/or antioxidant redox environment to link increased oxidative distress, including during excess substrate metabolism [13,24], to impaired insulin sensitivity and glucose metabolism [13,, , ,52].
Antioxidant treatment and oxidative distress.
Exogenous antioxidant treatment in rodents and cell culture have provided strong evidence supporting the restoration of redox homeostasis and insulin sensitivity during acute and chronic conditions of elevated oxidative distress [10,13,24,26,27,, , , ]. In humans, however, the effectiveness of exogenous antioxidant treatment for enhancing or restoring insulin sensitivity and glycemic control is far less consistent [70,, , , , , , ]. Nevertheless, acute intravenous GSH infusion (1 and 6 h infusions) or longer-term oral ingestion of vitamins E and C (over 4 months) improves insulin sensitivity in healthy individuals [65,67,82] and patients with T2D [65,67,96].
Other measures of glycemic control and insulin sensitivity including HbA1c, HOMA, and both fasting and postprandial glucose and insulin excursions, have also been reported to improve with various antioxidant treatment regimes in populations spanning the IGT and T2D continuum [67,, , , , ]. A recent meta-analysis of twenty-eight studies (n = 1574) reported that oral vitamin C supplementation in people with T2D significantly lowers HbA1c . H
owever, the current certainty of evidence is low due to many studies being mostly short term (less than 6 months) and with a small number of participants (n<100) . High quality, large and long-term randomized controlled trials are still required to establish the efficacy of vitamin C for the treatment of T2D. Furthermore, an equal number of studies have reported null or limited effects with antioxidant treatment on glycemic control [70,90,, , , ], questioning the efficacy of antioxidant treatment in humans.
In addition to longer-term models of insulin resistance and T2D, antioxidant treatment has also been reported to attenuate insulin resistance in humans caused by short-term excess nutrient intake. Research by Paolisso et al., 1996  revealed that impairments in insulin sensitivity after 6 h of lipid infusion in healthy individuals is partially restored with the intravenous co-infusion of GSH .
Moreover, lipid-induced impairments in insulin sensitivity correlated with increased plasma TBARS, whereas restoration of insulin sensitivity via GSH co-infusion correlated with decreased plasma TBARS . Similar research investigating oral taurine ingestion in overweight and obese individuals also reported restoration of insulin sensitivity and the prevention of increased plasma malondialdehyde and 4-HNE following 48 h of lipid infusion .
Rodent and cell culture models of lipid-induced insulin resistance confirm the potential of antioxidant treatment for preventing high-fat-diet induced insulin resistance [10,13,25,87]. As such, antioxidant treatment may yet prove to be a promising candidate for restoring systemic redox homeostasis and improving insulin sensitivity in a variety of populations under a variety of oxidative distress conditions (Table 1).
Oxidative eustress: implications for glycemic control and exercise
Despite evidence for the contribution of oxidative distress to insulin resistance, oxidative eustress, on the other hand, is necessary for physiological insulin signaling and glucose regulation.
As a key regulator in the proximal insulin signaling pathway, the protein tyrosine phosphatase (PTP) family which includes PTP1B, phosphatase and tensin homolog (PTEN), and protein phosphatase 2 (PP2A), can be reversibly oxidized to either inhibit or propagate insulin signal transduction [22,28].
Under basal conditions, endogenous catalase and peroxiredoxin antioxidant activity creates a reduced intracellular redox environment promoting PTP activity which, via dephosphorylation, suppresses activation of the insulin signaling cascade [21,47,107].
Upon insulin stimulation, the binding of insulin to the insulin receptor leads to a localized burst of endogenous O2·- and H2O2 production through increased enzymatic activation of the plasma membrane bound NADPH oxidases that, via reversible oxidation, inactivate peroxiredoxin I in the vicinity of the receptor complex [21,30,31,83,84,108]. This leads to a localized oxidative redox environment which decreases PTP activity and permits insulin-stimulated propagation of the insulin signaling cascade [9,30,31,83,84]. As such, insulin-induced ROS production is a requirement for the initiation of insulin signaling and glucose cell transport.
Oxidative eustress and enhanced glycemic control. Skeletal muscle contraction and exercise are stimuli that lead to conditions of oxidative eustress [8,22,, , ]. Skeletal muscle contraction directly increases ROS production [11,12,15,, , , , ], which contributes to beneficial cardiometabolic responses to exercise including improved vascular health and function [, , , , , ], mitochondrial biogenesis and the upregulation of antioxidant defenses [32,110,111,, , , , , , , , , ], skeletal muscle force production , and skeletal muscle inflammatory response and repair capabilities [129,132].
Furthermore, exercise-induced ROS is a factor in mediating both contraction-mediated and post-exercise insulin-stimulated glucose metabolism [29,32,33,, , ].
In rodents, knockout mice lacking a key antioxidant enzyme (Gpx1-/-) required for the reduction of H2O2, demonstrate improved insulin sensitivity 1 h after treadmill exercise compared to wild type mice . Henriquez-Olguin et al., 2019  also recently reported that mice lacking the O2·- producing enzyme NADPH oxidase 2 (NOX2) exhibit lower exercise-induced cytosolic ROS production concomitant with impaired glucose metabolism compared to wild type mice .
These findings suggest the direct role of increased ROS in both contraction-mediated and post-exercise insulin stimulated glucose metabolism, at least in rodents. However, the dynamic and complex redox reactions elicited by exercise, nutrient intake, and substrate-oxidation, combined with the difficulties in directly and accurately measuring ROS and their functional role in humans, presents challenges for studying their effects.
Nevertheless, we previously showed that enhanced insulin sensitivity in middle-aged obese men in the hours after acute high-intensity cycling exercise, was associated with greater exercise-induced phosphorylation of redox-sensitive proteins JNK, p38 MAPK and NF-κB in skeletal muscle .
Additionally, plasma SOD and CAT activity increased, plasma TBARS and H2O2 decreased, and 4-HNE increased in skeletal muscle after both exercise and insulin stimulation , whereas skeletal muscle mitochondrial H2O2 production decreased . As such, exercise and insulin stimulation dynamically change the redox environment in humans in a tissue and subcellular organelle specific manner.
Combined with numerous other reports of increased oxidative stress in muscle and blood alongside increased redox-sensitive cell signaling during and after exercise in humans [11,32,133,, , , ], a time-period in which insulin-dependent and independent glucose uptake are consistently enhanced [22,141], exercise-induced ROS likely leads to a state of oxidative eustress . However, research by our team and others have also discussed and identified the potential contribution of oxidative distress during exercise which may limit exercise capacity and muscle force production [7,131,143]. As such, caution when interpreting systemic markers of oxidative stress in humans is required.
Antioxidant treatment, oxidative eustress, and glycemic control. Directly linking ROS to the regulation of insulin and glucose metabolism in humans remains technically challenging. Nevertheless, exogenous antioxidant compounds have been used to manipulate redox homeostasis in vivo. Seminal research by Ristow et al., 2009  suggested a link between exercised-induced oxidative eustress and improved insulin sensitivity in humans.
In this study, oral supplementation of vitamin C (1000 mg/day) and vitamin E (400 IU/day) not only blocked the acute exercise-induced increase in muscle TBARS, but also prevented the improvements in insulin sensitivity elicited by 4-weeks of aerobic exercise training in both previously untrained and trained individuals .
Furthermore, exercise training led to increased skeletal muscle Cu/Zn-SOD, Mn-SOD and Gpx1 antioxidant gene expression; beneficial training effects that were prevented in the antioxidant treatment groups . Another study showed that intravenously infusing the antioxidant N-acetylcysteine (NAC) in healthy adults during aerobic cycling exercise increased the muscle GSH/GSSG ratio, decreased protein carbonylation, and lead to a small but significant decrease in post-exercise insulin sensitivity .
Trewin et al., 2013  also reported that NAC infusion during high-intensity interval exercise in well-trained cyclists increased blood glucose levels alongside increased fat oxidation, indicating an alteration of substrate oxidation during exercise with antioxidant treatment which coincided with impaired time trial cycling performance. In rodents, administration of NAC prevents the enhancement of post-exercise insulin sensitivity in Gpx1−/− mice , and impairs contraction-mediated glucose uptake during ex vivo mouse skeletal muscle contraction . Together, findings suggest that exercise-induced ROS play a key role in insulin and glucose regulation and raises doubt over the use of antioxidants as a one-stop-shop treatment for improving health and wellbeing.
Exercise as an antioxidant. The potential benefits of exercise-induced oxidative eustress on insulin and glucose regulation may also extend beyond their direct involvement in mediating insulin-dependent and -independent glucose uptake. It is well-established that acute exercise transiently increases ROS generation, which over-time with regular exercise training, promotes the adaptation and upregulation of endogenous antioxidant defenses and/or decreased measures of systemic oxidative distress [32,124,125,, , ].
Exercise induced changes in redox homeostasis, including decreased markers of systemic oxidative distress, often coincide with improvements in insulin and glucose regulation [104,146,147]. Twelve weeks of moderate-intensity cycling exercise training in obese individuals increased skeletal muscle Cu/ZnSOD and MnSOD protein content, decreased urinary 8-OHdG and 8-isoprostanes, decreased skeletal muscle 4-HNE and protein carbonyls, while improving HOMA-IR and 2 h OGTT glucose and insulin levels .
Associations between reductions in urinary 8-OHdG and reduced glycated albumin following 12 months of moderate intensity aerobic training in T2D patients have also been reported . Konopka et al., 2015  reported that 12 weeks of aerobic exercise training in obese females with polycystic ovary syndrome decreased fasting skeletal muscle mitochondrial H2O2 production and muscle 8-oxo-dG, increased muscle catalase activity, and prevented the increase in muscle protein carbonyls observed over 12-weeks of sedentary behavior in obese controls .
These changes in muscle redox homeostasis coincided with improved insulin sensitivity as measured by hyperinsulinemic-euglycemic clamp, HOMA and postprandial responses to a high-fat mixed nutrient meal . Exercise intensity may also play a role in altering redox homeostasis and glycemic control [50,138,140,148], with one study revealing that 12 weeks of high-intensity interval-exercise training in adults with T2D led to improved HOMA-IR, decreased TBARS and von Willebrand factor, and increased glutathione peroxidase and nitric oxide, effects that were absent in the continuous training comparison group .
Redox homeostasis and measures of glycemic control are, however, not always consistently altered by exercise training [, , , ]. For example, despite lower baseline fasting levels of urinary 8-iso PGF2α correlating with lower minimum glucose levels measured by continuous glucose monitoring (CGM) over a 24 h period in T2D patients, urinary 8-iso PGF2α was unchanged after 2 weeks of interval walking training despite improvements in CGM-derived measures of glycemic control .
In adults with T2D, neither 1 year of high-intensity interval training or moderate-intensity continuous training significantly altered total antioxidant capacity, F2-isoprostanes, glutathione peroxidase, or protein carbonyls, although sex-specific sub-analysis revealed MICT decreased protein carbonyls in females and decreased total antioxidant capacity in males .
Krause et al., 2014  reported an increase in plasma catalase antioxidant activity and decrease in plasma protein carbonyls in participants with T2D after 16 weeks of moderate-intensity exercise training, however, measures of glycemic control were unaffected. Furthermore, an energy‐restricted diet combined with 12 weeks of aerobic exercise training in overweight and obese patients with T2D led to similar changes as the diet alone including decreased plasma MDA, increased total antioxidant status (TAS), and improved measures of body fat, blood pressure, lipids, blood glucose, HbA1c and HOMA2-IR . Thus, improvements in redox homeostasis are possibly also due to improvements in cardiometabolic health that accompany weight loss, in addition to the direct effects of exercise.
This creates a unique redox health paradox in which exercise and muscle contraction increase ROS production, yet possess the capacity to decrease ROS production directly by increasing antioxidant activity, and/or indirectly by decreasing ROS that is produced through lifestyle factors such as physical inactivity and excess adiposity [9,22].
The most direct evidence for the role(s) of exercise-induced ROS and oxidative eustress in regulating insulin and glucose homeostasis comes from rodent and cell culture studies. Attempts to link training-induced improvements in glycemic control to altered redox homeostasis in human studies remain inconclusive. The lack of consensus likely stems from methodological limitations in human research which, to-date, lack the ability to use the mechanistic approaches that are possible in genetic rodent models and cell culture experiments.
In the absence of more sophisticated study design and mechanistic experiments in humans, it is difficult, if not impossible, to delineate the complex and dynamic interactions between redox biology and insulin and glucose regulation. The vast array of exogenous antioxidants used, variation in doses and duration of treatment, and differing methods of administration, likely also influence the efficacy of the treatment and thus contribute to inconsistent findings (Table 2).
reference link : https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8167146/
More information: Chrysovalantou Xirouchaki et al, Skeletal muscle NOX4 is required for adaptive responses that prevent insulin resistance, Science Advances (2021). DOI: 10.1126/sciadv.abl4988. www.science.org/doi/10.1126/sciadv.abl4988