This study investigated whether fatiguing dynamic exercise depresses maximal in vitro Na+-K+-ATPase activity and whether any depression is attenuated with chronic training. Eight untrained (UT), eight resistance-trained (RT), and eight endurance-trained (ET) subjects performed a quadriceps fatigue test, comprising 50 maximal isokinetic contractions (180°/s, 0.5 Hz). Muscle biopsies (vastus lateralis) were taken before and immediately after exercise and were analyzed for maximal in vitro Na+-K+-ATPase (K+-stimulated 3-O-methylfluoroscein phosphatase) activity. Resting samples were analyzed for [3H]- ouabain binding site content, which was 16.6 and 18.3% higher (P < 0.05) in ET than RT and UT, respectively (UT 311 ± 41, RT 302 ± 52, ET 357 ± 29 pmol/g wet wt). 3-O-methylfluoroscein phosphatase activity was depressed at fatigue by -13.8 ± 4.1% (P < 0.05), with no differences between groups (UT -13 ± 4, RT -9 ± 6, ET -22 ± 6%). During incremental exercise, ET had a lower ratio of rise in plasma K+ concentration to work than UT (P < 0.05) and tended (P = 0.09) to be lower than RT (UT 18.5 ± 2.3, RT 16.2 ± 2.2, ET 11.8 ± 0.4 nmol·l-1·J-1). In conclusion, maximal in vitro Na+-K+-ATPase activity was depressed with fatigue, regardless of training state, suggesting that this may be an important determinant of fatigue.