Identification of cysteine residues in lamb kidney (Na,K)-ATPase essential for ouabain binding

N-(1-Pyrene)maleimide is a hydrophobic, sulfhydryl-directed, chemical modification probe which, at a low concentration, inhibits the capacity of lamb kidney sodium- and potassium-activated adenosine triphosphatase [Na,K)-ATPase; EC 3.6.1.3) to bind ouabain. This inhibition is partially blocked by preincubation of the enzyme with ouabagenin, an aglycone derivative which can be used as a reversible protecting ligand for the ouabain binding site. The kinetics of inhibition are not first order, suggesting that there may be more than one site of labeling which is responsible for the inhibition of ouabain binding. Although earlier work (Kirley, T. L., Lane, L. K., and Wallick, E. T. (1986) J. Biol. Chem. 261, 4525-4528) indicates that the inhibition is accompanied by a loss in the number of binding sites rather than a decrease in affinity of the sites for the ligand, other data (Scheiner-Bobis, G., Zimmerman, M., Kirch, V., and Schoner, W. (1987) Eur. J. Biochem. 165, 653-656) indicates that there is no cysteine residue located extracellularly in the ouabain binding site. By sequence analysis of alpha subunit peptides labeled by N-(1-pyrene)maleimide in the absence but not in the presence of protecting ligand, it is demonstrated in this work that there are two major sites of labeling protected by the binding of ouabagenin, Cys-367 and Cys-656. Both of these sites are located in the large cytoplasmic domain of the alpha subunit, one close to the phosphorylation site (Asp-369), and the other implicated in the binding of ATP (Cys-656). Therefore, it appears from this data that the inhibition of ouabain binding by N-(1-pyrene)maleimide is not due to modification of a site in the binding pocket for cardiac glycosides, but rather to an allosteric effect, since cardiac glycoside binding is known to be dependent on the phosphorylation state of the enzyme. The dependence of inhibition on the presence of sodium, potassium, and ATP also is consistent with this interpretation. The work reported here thus explains the apparent paradox posed by the earlier data.