Horses were subjected to treadmill cunning at 65% (submaximal) or 100% (maximal) VO2,max to examine the effects of exercise on subcellular distribution of hexokinase (HK) and on mitochondrial respiration. It is hypothesized that the fraction of HK bound to mitochondria will be reduced due to an elevation of glucose-6-phosphate (G-6-P) concentration in the exercising muscle and that such release of HK from mitochondria will depress oxidative phosphorylation. Changes in muscle G-6-P concentration, pH, subcellular HK distribution, mitochondrial respiration and other metabolites were determined in biopsy samples pre-exercise, immediately post-exercise and during the recovery phase. The fraction of HK associated with mitochondria decreased from 38% to 7% at the end of maximal exercise; exercise at VO2,max also reduced respiratory capacity of muscle homogenates by 20% and was associated with a fivefold increase in muscle [G-6-P], a potent agent known to dissociate HK from mitochondria. The HK distribution returned to normal within 60 min after exercise and the reassociation of the HK with mitochondria parallelled the removal of muscle G-6-P. No changes in muscle HK distribution and respiration were found following;the submaximal exercise despite the fact that G-6-P was slightly elevated. Muscle concentrations of adenosine triphosphate, creatine phosphate and glycogen and pH dropped after exercise while lactate concentration increased. The amount of mitochondria-bound HK was also altered in vitro in a preparation of mitochondria isolated from rat skeletal muscle to examine the effect of the bound HK on mitochondrial respiration. The isolated mitochondria were incubated with G-6-P to release the bound HK. G-6-P induced the release of HK from mitochondria and reduced respiration. Mg2+ diminished both the G-6-P-induced HK release and the respiratory depression. Our data demonstrate that the fraction of HK associated with mitochondria can be reduced during maximal, but not submaximal, exercise; this reduction of bound HK was probably caused by the large increase in G-6-P concentration and was at least partially responsible for the depressed mitochondrial respiration seen at the end of the maximal exercise.