Physical activity has several benefits: muscle usage increases or at least prevents loss of muscle tissue, it increases the ability to produce heat and eventually muscle mass correlates with calorie consumption. For a person, this means, exercise improves physical attractiveness, contributes to a comfortable body temperature, and eventually prevents obesity. Countless lifestyle-books,-magazines, and-webpages have been filled up by reiterating the benefits of physical exercise. Simultaneously, the same books, magazines, and webpages tireless remind their readers of the detrimental role of reactive oxygen species (ROS, such as superoxide anion or H2O2). We are constantly called on to fight the danger of ROS by the intake of antioxidants. Interestingly, many articles about the benefits of physical activity conceal the fact that exercise comes along with ROS formation and antioxidant advertisement keeps quite about potential beneficial effects of ROS. In diabetes, complications occur due to an increased production of ROS. 1 However, especially persons suffering from diabetes benefit enormously from exercise as an additional, none medical treatment of their disease. 2 A study by Brendel et al. sheds new light on the complicated relationship of exercise in diabetes and ROS. The constitutive producer of H2O2, NADPH oxidase Nox4, mediates positive effects of exercise on endothelial function in obese mice 3: wildtype and Nox4-/-, fed a high-fat diet over 20 weeks, were allowed or not to use a running wheel on a voluntary basis. Not only, that the absence of Nox4 reduced the performance on the running wheel, other positive effects of physical exercise where less pronounced in Nox4-/-on high-fat diet when compared with their wild-type mates. Those include maintenance of high fat-induced endothelial dysfunction in running Nox4-/-mice and less induction of peroxisome proliferative activated receptor gamma, coactivator 1 alpha (Ppargc1a), which is a key regulator of mitochondria biogenesis. Subsequently, Nox4 deficiency in running mice on high-fat diet prevents the exercise-induced increase in mitochondria mass and exercise-induced citrate synthase activity. Earlier studies in men, suggest that exercise-induced beneficial effects on insulin sensitivity occur only at high exercise levels with extraordinary energy expenditure, above 900kcal, per exercise. 4 Importantly, whole body as well as heart specific knock out of Nox4 reduces exercise capacity in mice. 5 Accordingly, it is worth a thought that reduced Nox4 activity limits the capacity to exercise and subsequently the patient constantly fails to improve insulin resistance by exercise. Nevertheless, increasing or at least keeping muscle mass enables for an improved glucose uptake and utilization. The role of Nox4 in skeletal muscle in exercise, however, might be less important. Nox4 is not involved in exercise-induced muscle fibre phenotype switch and therefore an altered energy demand, but rather in repair of injured muscle. 6, 7 Interestingly, overexpression of human Nox4 in skeletal muscle, impairs glucose tolerance in mice 8 and Nox4 inhibitors such as GLX351322 or GLX7013114 counteract glucose intolerance in high-fat diet C57BL/6 mice and human islet cell death in vitro. 9, 10

In contrast, in the heart Nox4 may be involved in pressure overloadinduced heart remodelling. 11 Angiogenesis, improved by Nox4, represents at least one reason for prevention of pressure-overload-induced heart hypertrophy. 12 Endothelial homeostasis is maintained by Nox4, and in exercise, it mediates the resulting increase in capillary density. 13, 14 Figure 1 summarizes some published effects of Nox4 in exercise.