We studied whether endurance training impacts insulin sensitivity by affecting the structural and storage lipids in humans. Eight male subjects participated (age 25 ؎ 1 years, height 178 ؎ 3 cm, weight 76 ؎ 4 kg [mean ؎ SE]). Single-leg training was performed for 30 min/day for 4 weeks at ϳ70% of single-leg maximal oxygen uptake. After 8, 14, and 30 days, a two-step hyperinsulinemiceuglycemic glucose clamp, combined with catheterization of an artery and both femoral veins, was performed. In addition, a muscle biopsy was obtained from vastus lateralis of both legs. Maximal oxygen uptake increased by 7% in the trained leg (T), and training workload increased (P < 0.05) from 79 ؎ 12 to 160 ؎ 15 W. At day 8, glucose uptake was higher (P < 0.01) in the trained (0.8 ؎ 0.2, 6.0 ؎ 0.8, 13.4 ؎ 1.2 mg ⅐ min ؊1 ⅐ kg ؊1 leg wt) than the untrained leg (0.5 ؎ 0.2, 3.7 ؎ 0.6, 10.5 ؎ 1.5 mg ⅐ min ؊1 ⅐ kg ؊1 leg wt) at basal and the two succeeding clamp steps, respectively. After day 8, training did not further increase leg glucose uptake. Individual muscle triacylglycerol fatty acid composition and total triacylglycerol content were not significantly affected by training and thus showed no relation to leg glucose uptake. Individual muscle phospholipid fatty acids were not affected by training, but the content of phospholipid polyunsaturated fatty acids was higher (P < 0.06) after 30 than 8 days in T. Furthermore, after 30 days of training, the sum of phospholipid long-chain polyunsaturates was correlated to leg glucose uptake (r ؍ 0.574, P < 0.04). Endurance training did not influence muscle triacylglycerol content or total triacylglycerol fatty acid composition. In contrast, training induced a minor increase in the content of phospholipid fatty acid membrane polyunsaturates, which may indicate that membrane lipids may have a role in the training-induced increase in insulin sensitivity.