UC Berkeley

The bacterial flagellar motor (BFM) is an ion-powered nanomachine that drives swimming in many bacteria. Its central role in processes like chemotaxis and biofilm formation has made obtaining a mechanistic understanding of its function a central challenge in biophysics. This protein complex is comprised of several transmembrane rings connected to a long flagellar filament by a flexible hook. Rotation is known to occur via an interaction between one or more membrane-embedded "stator" units and protein spokes on the periphery of the "rotor" ring. Further modeling has been stymied by the lack of atomic-level structures, which are hard to obtain because the BFM is large and traverses the membrane.

The work in this thesis attempts to clear this hurdle by combining partial crystal structures, mutation and crosslinking experiments, and biophysical measurements to propose a mechanically-specific, experimentally-testable model of how the motor generates torque. Predictions are validated against biophysical measurements on single-stator motors, which isolate the properties of the motor's fundamental mechanochemical cycle by removing the effects of interactions between individual stator units.

This base model is also extended to consider the behavior of multi-stator motors. Recent experiments have shown that stator units dynamically bind and leave the motor, influenced by several factors, including the ion gradient, external load, and motor speed. This has brought into question past results, one of which is the long-held conviction that the maximum speed of the motor is independent of stator number; that is, that at near-zero load, one engaged stator unit can push the motor at its maximum speed. The model proposed in this thesis predicts that such a "limiting" zero-torque speed in fact does not exist: Recruitment of additional stator units, even at very low external load, results in an increase in motor speed. This assertion is supported by experimental results in chimeric sodium-driven motors.

Finally, validation experiments for this model are described. The mechanism presented is unique in that it suggests motor rotation is loosely coupled to ion flux; that is, an ion passage may not always constitute a fully efficient power stroke. Thus, motor steps are likely to be unequal in size, and the number of ions required for a full revolution of the motor may vary with external conditions. This notion is tested using single-cell fluorescent measurements to compare ion flux in motors at different speeds.