The torque of rotary F-ATPase can unfold subunit gamma if rotor and stator are cross-linked.

Abstract

During ATP hydrolysis by F(1)-ATPase subunit γ rotates in a hydrophobic bearing, formed by the N-terminal ends of the stator subunits (αβ)(3). If the penultimate residue at the α-helical C-terminal end of subunit γ is artificially cross-linked (via an engineered disulfide bridge) with the bearing, the rotary function of F(1) persists. This observation has been tentatively interpreted by the unfolding of the α-helix and swiveling rotation in some dihedral angles between lower residues. Here, we screened the domain between rotor and bearing where an artificial disulfide bridge did not impair the rotary ATPase activity. We newly engineered three mutants with double cysteines farther away from the C-terminus of subunit γ, while the results of three further mutants were published before. We found ATPase and rotary activity for mutants with cross-links in the single α-helical, C-terminal portion of subunit γ (from γ285 to γ276 in E. coli), and virtually no activity when the cross-link was placed farther down, where the C-terminal α-helix meets its N-terminal counterpart to form a supposedly stable coiled coil. In conclusion, only the C-terminal singular α-helix is prone to unwinding and can form a swivel joint, whereas the coiled coil portion seems to resist the enzyme's torque.