The extent and mechanism of the cardiac benefit of early exercise training following myocardial infarction (MI) is incompletely understood, but may involve blunting of abnormalities in Ca(2+)-handling and myofilament function. Consequently, we investigated the effects of 8-weeks of voluntary exercise, started early after a large MI, on left ventricular (LV) remodeling and dysfunction in the mouse. Exercise had no effect on survival, MI size or LV dimensions, but improved LV fractional shortening from 8+/-1 to 12+/-1%, and LVdP/dt(P30) from 5295+/-207 to 5794+/-207 mm Hg/s (both P<0.05), and reduced pulmonary congestion. These global effects of exercise were associated with normalization of the MI-induced increase in myofilament Ca(2+)-sensitivity (DeltapCa(50)=0.037). This effect of exercise was PKA-mediated and likely because of improved beta(1)-adrenergic signaling, as suggested by the increased beta(1)-adrenoceptor protein (48%) and cAMP levels (36%; all P<0.05). Exercise prevented the MI-induced decreased maximum force generating capacity of skinned cardiomyocytes (F(max) increased from 14.3+/-0.7 to 18.3+/-0.8 kN/m(2) P<0.05), which was associated with enhanced shortening of unloaded intact cardiomyocytes (from 4.1+/-0.3 to 7.0+/-0.6%; P<0.05). Furthermore, exercise reduced diastolic Ca(2+)-concentrations (by approximately 30%, P<0.05) despite the unchanged SERCA2a and PLB expression and PLB phosphorylation status. Importantly, exercise had no effect on Ca(2+)-transient amplitude, indicating that the improved LV and cardiomyocyte shortening were principally because of improved myofilament function. In conclusion, early exercise in mice after a large MI has no effect on LV remodeling, but attenuates global LV dysfunction. The latter can be explained by the exercise-induced improvement of myofilament function.