by Yun Wang, Li Wen, Shi Zhou, Yong Zhang, Xin-Hao Wang, You-Yu He, Allan Davie, Suzanne Broadbent Aims The aims of this study were to determine the effects of four weeks of intermittent exposure to a moderate hypoxia environment (15% oxygen), and compare with the effects of exercise in normoxia or hypoxia, on glucose homeostasis, insulin sensitivity, GLUT4 translocation, insulin receptor phosphorylation, Akt-dependent GSK3 phosphorylation and Akt activity in skeletal muscle of obese mice with type 2 diabetes. Methods C57BL/6J mice that developed type 2 diabetes with a high-fat-diet (55% fat) (fasting blood glucose, FBG = 13.9 ± 0.69 (SD) mmol/L) were randomly allocated into diabetic control (DC), rest in hypoxia (DH), exercise in normoxia (DE), and exercise in hypoxia (DHE) groups (n = 7, each), together with a normal-diet (4% fat) control group (NC, FBG = 9.1 ± 1.11 (SD) mmol/L). The exercise groups ran on a treadmill at intensities of 75–90% VO 2 max. The interventions were applied one hour per day, six days per week for four weeks. Venous blood samples were analysed for FBG, insulin (FBI) and insulin sensitivity (QUICKI) pre and post the intervention period. The quadriceps muscle samples were collected 72 hours post the last intervention session for analysis of GLUT4 translocation, insulin receptor phosphorylation, Akt expression and phosphorylated GSK3 fusion protein by western blot. Akt activity was determined by the ratio of the phosphorylated GSK3 fusion protein to the total Akt protein. Results The FBG of the DH, DE and DHE groups returned to normal level (FBG = 9.4 ± 1.50, 8.86 ± 0.94 and 9.0 ± 1.13 (SD) mmol/L for DH, DE and DHE respectively, P 〈 0.05), with improved insulin sensitivity compared to DC (P 〈 0.05), after the four weeks treatment, while the NC and DC showed no significant changes, as analysed by general linear model with repeated measures. All three interventions resulted in a significant increase of GLUT4 translocation to cell membrane compared to the DC group (P 〈 0.05). The DE and DH showed a similar level of insulin receptor phosphorylation compared with NC that was significantly lower than the DC (P 〈 0.05) post intervention. The DH and DHE groups showed a significantly higher Akt activity compared to the DE, DC and NC (P 〈 0.05) post intervention, as analysed by one-way ANOVA. Conclusions This study produced new evidence that intermittent exposure to mild hypoxia (0.15 FiO 2 ) for four weeks resulted in normalisation of FBG, improvement in whole body insulin sensitivity, and a significant increase of GLUT4 translocation in the skeletal muscle, that were similar to the effects of exercise intervention during the same time period, in mice with diet-induced type 2 diabetes. However, exercise in hypoxia for four weeks did not have additive effects on these responses. The outcomes of the research may contribute to the development of effective, alternative and complementary interventions for management of hyperglycaemia and type 2 diabetes, particularly for individuals with limitations in participation of physical activity.