![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| GUEST COMMENTARY |
Department of Biology, University of Utah, Salt Lake City, Utah
Bacteria can move in several ways, but probably the most common means of locomotion is by the rotary organelles called flagella. Flagella are evolutionarily ancient and may be the most widespread machines of propulsion in biology: genes for flagellar components are found in almost half of the more than 500 bacterial genomes presently on the NCBI website, in diverse, deeply branched lineages. The essentials of flagellar function have been understood for more than 30 years. A transmembrane ion current, commonly protons but sometimes sodium ions, is harnessed to drive rotation of a membrane-embedded motor linked to a slender helical filament that functions as propeller (4, 13, 17). By regulating the direction and/or speed of flagellar rotation in response to various sensory cues, cells can bias their movement towards nutrients, temperatures, or other conditions that favor survival (1, 2, 18). The physiology of the flagellar motor has been studied extensively, and many aspects of flagellar motor function are well understood (reviewed in reference 3). While ultrastructural studies have also been useful, providing an overall view of the shape of the flagellar basal body, until recently such studies have lacked the resolution needed to illuminate molecular mechanisms. Our structural understanding of the flagellum has taken a great leap forward with the electron microscopic reconstructions reported in this issue by Thomas et al. (25), which reveal the flagellar basal body in stunning new detail.
The flagellum has three main parts: a rigid, helical filament that is the propeller; a short, flexible hook that functions as a universal joint; and the basal body, which is embedded in the cell membrane(s). Early electron micrographs showed that the basal body consists of a set of rings mounted on a central rod (Fig. 1A) (8). The rings were named according to their locations relative to layers of the cell envelope: L ring, lipopolysaccharide; P ring, peptidoglycan; S ring, supramembrane; and M ring, inner membrane (Fig. 1A). The M and S rings were subsequently found to be composed of a single protein (27) and are now recognized as a single entity termed the MS ring. By use of gentler isolation procedures, Francis et al. (10) and Khan et al. (14) prepared more-intact basal bodies from Salmonella that contain another large ring at the inner end, termed the C ring (for cytoplasmic). Micrographs of this more-intact structure in vitreous ice provided the basis for single-particle reconstructions at approximately 22-Å resolution, which for several years were the most detailed images available (10, 24). The reconstruction shows a fairly ornate structure with a silhouette reminiscent of a chess piece (Fig. 1B). Because the reconstruction procedure relied on axial averaging (i.e., blurring the map by computationally rotating it about its long axis), azimuthal variations in density—and thus most subunit structure within the rings—were effectively washed out.
|
The resulting images are spectacular. One representative close-up suffices to illustrate the magnitude of the resolution improvement. Where the previous best image showed a rounded blob at the bottom of the C ring, the new images show a clearly resolved band of density winding around the bottom of the ring like a spiral whisk (Fig. 1C). Other parts of the structure are similarly rich in new detail. Of particular note, one feature in the C ring—the more inboard lobe of electron density at the top—is found to have a symmetry that matches that of the MS ring (i.e., about 25-fold), whereas other features of the C ring display the 
34-fold symmetry first seen in bottom views. And as it happens, the MS ring also shows variable symmetry, in this case ranging between about 24- and 26-fold.
What can these new views of the structure tell us about the function of the motor? The basal body seen in the electron micrographs is thought to be just the rotating part of the motor. The MotA/MotB protein complexes that form the stator (the part that remains stationary relative to the cell body) are believed to be in the membrane just above the C ring (Fig. 1B). These stator complexes conduct the ions that energize the motor, probably harnessing ion flow to drive conformational power strokes that push on the C ring to drive rotation (16). The C ring is thus the element of the rotor most important for rotation and switching. It is formed from the proteins FliG, FliM, and FliN, each present in many copies (29, 31). Crystal structures have been solved for most of the FliG and FliN proteins (5, 6) and will soon be available for most of FliM (21). The resolution of the new electron microscopic images is almost sufficient to allow crystal structures to be docked into the map. Thus, remarkably, we may soon have a near-atomic structural model for most of the C ring.
Certain features of subunit organization already seem clear. FliN has been suggested to form the rounded feature at the bottom of the C ring, on the basis of its crystal structure and cross-linking studies indicating that it forms a roughly donut-shaped tetramer (22). This placement of FliN appears to be supported by the new images, because FliN tetramers distorted only slightly from the hypothesized donut shape (into the shape of a spiral lock washer) would fit well in the spiral density at the bottom of the C ring. Also, the inboard lobe of electron density at the top of the C ring, which exhibits symmetry matching that of the MS ring, is very likely due to a domain of FliG. FliG is known to be present in the same number of copies as FliF, the protein that forms the MS ring (9), and a variety of mutational, binding, and cross-linking studies place FliG near the top of the C ring (11, 15, 19, 20, 32). As discussed by Thomas et al. (25) and Lowder et al. (19), the present data are consistent with either of two FliG arrangements, both with FliG at the top but one having FliG in a more outboard position than the other (Fig. 1D). Whichever of these placements proves correct, it seems that FliM must form the middle, relatively thin part of the C ring wall. A forthcoming crystal structure of FliM should provide a critical test of this model (21). Can FliM subunits be fitted into this part of the C ring, or is a different arrangement required?
In the cell, switching of motor direction is controlled by the signaling molecule phospho-CheY, which promotes clockwise rotation (in the enteric species typically used in chemotaxis studies) (12, 28). Motor switching is the culminating event in chemotaxis, yet its molecular mechanism remains very poorly understood. The new images of the basal body, and the structural models they will inform, should provide a concrete framework for addressing this question. The present images are of basal bodies locked by mutation in the clockwise rotational state. When similarly detailed images are obtained for motors in the counterclockwise state, what differences will emerge? The flagellar switch is highly cooperative, responding to increasing phospho-CheY with a Hill coefficient greater than 10 (7). This sharp switching is credited with much of the sensitivity of the chemotactic response. A fuller structural understanding of the C ring should illuminate this aspect of switch function also.
Symmetry provides a large advantage for image reconstruction because it allows the use of averaging procedures that reinforce symmetry-related features in the map relative to the noise. However, in the case of the Salmonella basal body, the occurrence of two different symmetries means that most basal bodies have, strictly speaking, no rotational symmetry at all. In a basal body with a 25-subunit MS ring and a 34-subunit C ring, for example, none of the subunits can be in strictly equivalent environments. Parts of the structure can and apparently do conform approximately to one or another rotational symmetry, but at the interfaces where neither symmetry can fully apply, rotational averaging procedures are as likely to blur as to enhance the structural details. These interfacial regions could be some of the most interesting and important places in the structure. How can they be visualized accurately? To see the structure in high detail and without benefit of any symmetry-based averaging will require much larger numbers of images and possibly the use of target models that utilize the crystal structures. Electron tomography might also make an important contribution here, as it could in principle allow in situ visualization of the fully intact motor, including the stator complexes responsible for force generation.
While the flagella of diverse species appear to be built from basically the same parts, certain parts, most notably the C-ring protein FliN, show some variation between species. FliN is present in more than 100 copies (in the Salmonella motor) and is thus a major component of the basal body. In most bacilli and in Spirochaetes, FliN is replaced by the protein FliY, which is about twice as large. Some species contain both FliN and FliY. What do FliY-containing basal bodies look like? We do not know, because such basal bodies have not yet been isolated with their C rings intact. As efforts to push the resolution envelope continue with Salmonella, it seems a good time also to look more closely at other species. The advanced state of knowledge for Salmonella is largely due to efforts invested in biochemical characterization of the basal body, particularly the development of gentle but effective purification procedures (10, 14). Similar efforts with other species should be fruitful, allowing comparative studies of flagellar structure and function in the larger world.
 |
FOOTNOTES |
|---|
FOOTNOTES
The views expressed in this Commentary do not necessarily reflect the views of the journal or of ASM.
REFERENCES
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Appl. Environ. Microbiol. | Infect. Immun. | Eukaryot. Cell |
|---|---|---|
| Mol. Cell. Biol. | J. Virol. | Microbiol. Mol. Biol. Rev. |
| ALL ASM JOURNALS |
