A MODEL FOR THE FLAGELLAR MOTOR SWITCH SUGGESTED BY ITS 2-RING ARCHITECTURE

Shahid Khan
Laboratory of Cellular Bioenergetics, Department of Physiology & Biophysics, Albert Einstein College of Medicine, Bronx, NY 10461

Chemisomotic rotary machines, the bacterial flagellar motor and the F₀F₁ synthase, in interesting contrast with other molecular motors, execute bidirectional motion. In order to understand switching of motor rotation sense, we developed protocols allowing isolation and purification of bacterial flagellar basal bodies containing the motor switch complex (Khan et al., PNAS 173:2888-2896, 1992; Zhao et al., J. Mol. Biol. 261:195-208, 1996). The switch complex forms part of cytoplasmic structure, additionally isolated with the basal body, so as to interact with intracellular chemotaxis signaling proteins. It is composed of three proteins, FliG, FliM and FliN, which together account for ca. 75% of the deduced mass of the cytoplasmic structure. FliG and FliM are present in copy numbers (ca. 35 subunits) sufficient to form circumferential arrays. The smaller FliN protein is also present in large copy number (ca. 110 subunits) and could contribute to one or both arrays.

The cytoplasmic basal structure is seen as a peripheral ring (C-ring) in the single-particle averaged image of basal bodies isolated by similar protocols and visualized by cryoelectron microscopy (Francis et al., J. Mol. Biol. 235:1261-1270, 1994). We undertook a survey of basal bodies from spontaneously isolated switch complex mutants. While basal bodies isolated from non-chemotactic (che) alleles were intact, those isolated from non-motile (mot) alleles lacked both FliM and FliN, this absence being morphologically visible as reduced C-ring size (Zhao et al., J. Mol. Biol. 251:400-412, 1995). We have now prepared thin-film metal replicas and examined stereo-pair images of single basal-bodies. We find that the c-ring is comprised of two stacked, concentric arrays, each with a periodicity of ca. 35 subunits. The bottom array is frequently seen to be incomplete and to separate from the distal array. These images are consistent with the organization of the switch complex deduced from the biochemistry and, together with the analysis of the mutant alleles, suggest a model for how the flagellar motor switches rotation sense (Khan, S. BBA Reviews in Bioenergetics, in press, 1997). The model and the images on which it is based will be presented.