Analysis of the Membrane Domain of the Gastric H+/K+-ATPase

Open Access
Publisher Website
Google Scholar
Abstract
A structure of the catalytic or alpha subunit of the H+/K+-ATPase, with ten transmembrane segments, and of the beta subunit, with a single such segment, was established using a combination of tryptic cleavage and peptide sequencing and in vitro translation. Sites at which covalent ligands bind to external surfaces were also defined by cleavage, separation and sequencing. Cys813 was found to be the common covalent binding site for all the substituted pyridyl methylsulfinyl benzimidazoles. The binding region of a K+-competitive reagent, the 1,2α-imidazo-pyridine SCH 28080, was defined by the kinetic effects of site-specific mutations. Amino acids substitutions in membrane-spanning segments M1, M3, M4 and M6 were found to influence the apparent inhibitor constant, Ki, to varying degrees, some having a large effect, some a moderate effect and some a slight effect, whereas some mutations had no effect. We interpret changes in Ki without effects on the apparent Michaelis constant, Km, as affecting SCH 28080 binding only. Mutation of Cys813 significantly affected the Ki for SCH 28080, explaining the prevention of benzimidazole inhibition by the imidazo-pyridine. A model of the alpha subunit was constructed with a vestibule on the luminal surface of the pump bounded by M1–M6 and containing the SCH 28080 binding region. The cation binding site is suggested to be more towards the cytoplasmic face of the enzyme’s membrane domain. This model predicts the membrane peptide associations for the catalytic subunit. Biochemical and yeast two-hybrid methods place the beta subunit in association with M8, whereas similar methods place M5/6 in proximity to M9/10. These results, when combined with analysis of the two-dimensional crystals of the sarcoplasmic reticular Ca2+ and Neurospora crassa H+-ATPases, provide the basis for a tentative model of the arrangement of the six core segments of the gastric H+/K+-ATPase.