The objective of this study was to define the relationship among AMP-activated protein kinase (AMPK) activity, AMP concentration ([AMP]), and [ATP] in perfused rat hearts. Bromo-octanoate, an inhibitor of beta-oxidation, and amino-oxyacetate, an inhibitor of the malate-aspartate shuttle, were used to modify substrate flux and thus increase cytosolic [AMP]. Cytosolic [AMP] was calculated using metabolites measured by (31)P NMR spectroscopy. Rat hearts were perfused with Krebs-Henseleit solution containing glucose and either no inhibitor, the inhibitors, or the inhibitors plus butyrate, a substrate that bypasses the metabolic blocks. In this way, [AMP] changed from 0.2 to 27.9 microm, and [ATP] varied between 11.7 and 6.8 mm. AMPK activity ranged from 7 to 60 pmol.min(-1).microg of protein(-1). The half-maximal AMPK activation (A(0.5)) was 1.8 +/- 0.3 microm AMP. Measurements in vitro have reported similar AMPK A(0.5) at 0.2 mm ATP, but found that A(0.5) increased 10-20-fold at 4 mm ATP. The low A(0.5) of this study despite a high [ATP] suggests that in vivo the ATP antagonism of AMPK activation is reduced, and/or other factors besides AMP activate AMPK in the heart.