.epsilon. Subunit of Escherichia coli F1-ATPase: effects on affinity for aurovertin and inhibition of product release in unisite ATP hydrolysis

Abstract

The .epsilon. subunit of Escherichia coli F1-ATPase is a tightly bound but dissociable partial inhibitor of ATPase activity. The effects of .epsilon. on the enzyme were investigted by comparing the ATPase activity and aurovertin binding properties of the .epsilon.-depleted F1-ATPase and the .epsilon.-replete complex. Kinetic data of multisite ATP hydrolysis were analyzed to give the best fit for one, two, or three kinetic components. Each form of F1-ATPase contained a high-affinity component, with a Km near 20 .mu.M and a velocity of approximately 1 unit/mg. Each also exhibited a component with a Km in the range of 0.2 mM. The velocity of this component was 25 units/mg for .epsilon.-depleted ATPase but only 4 units/mg for .epsilon.-replete enzyme. The .epsilon.-depleted enzyme also contained a very low affinity component not present in the .epsilon.-replete enzyme. In unisite hydrolysis studies, .epsilon. had no effect on the equilibrium between substrate ATP and product ADP.cntdot.P at the active site but reduced the rate of product release 15-fold. These results suggest that .epsilon. subunit slows a conformational change that is required to reduce the affinity at the active site, allowing dissociation of product. It is suggested that inhibition of multisite hydrolysis by .epsilon. is also due to a reduced rate of product release. .epsilon.-depleted F1-ATPase showed little of no modulation of aurovertin fluorescence by added ADP and ATP. Aurovertin fluorescence titrations in buffer containing ethylenediaminetetraacetic acid (EDTA) revealed that .epsilon.-depleted enzyme had high affinity for aurovertin (Kd < 0.1 .mu.M) regardless of the presence of nucleotides. In contrast, .epsilon.-replete F1-ATPase had similar high affinity in the presence of nucleotides but much lower affinity, Kd = 3 .mu.M, in the absence of ADP or ATP. In the presence of MgADP, the affinity for aurovertin was lower in the absence of .epsilon. (Kd = 0.35 .mu.M) than in its presence (Kd = 0.2 .mu.M). The maximal fluorescence of aurovertin bound to .epsilon.-replete F1-ATPase was substantially lower in the absence of nucleotides or in the presence of Mg2+ than in the presence of nucleotide and EDTA, but only small effects on maximal fluorescence wereobserved for .epsilon.-depleted enzyme. Aurovertin titrations of multisite ATPase activity gave I0.5 of 0.09 .mu.M in the absence of .epsilon. and 0.04 .mu.M in the presence of saturating .epsilon.. These results indicate that the .epsilon. subunit alters the conformation of the .beta. subunit in a manner that depends on the presence of Mg2+ and nucleotides.