Structure of the constitutively active double mutant CheYD13K Y106W alone and in complex with a FliM peptide

CheY is a member of the response regulator protein superfamily that controls the chemotactic swimming response of motile bacteria. The CheY double mutant D13K Y106W (CheY**) is resistant to phosphorylation, yet is a highly effective mimic of phosphorylated CheY in vivo and in vitro. The conformational attributes of this protein that enable it to signal in a phosphorylation-independent manner are unknown. We have solved the crystal structure of selenomethionine-substituted CheY** in the presence of its target, a peptide (FliM16) derived from the flagellar motor switch, FliM, to 1.5A resolution with an R-factor of 19.6%. The asymmetric unit contains four CheY** molecules, two with FliM16 bound, and two without. The two CheY** molecules in the asymmetric unit that are bound to FliM16 adopt a conformation similar to BeF3- -activated wild-type CheY, and also bind FliM16 in a nearly identical manner. The CheY** molecules that do not bind FliM16 are found in a conformation similar to unphosphorylated wild-type CheY, suggesting that the active phenotype of this mutant is enabled by a facile interconversion between the active and inactive conformations. Finally, we propose a ligand-binding model for CheY and CheY**, in which Ile95 changes conformation in a Tyr/Trp106-dependent manner to accommodate FliM.     

A Molecular Mechanism of Bacterial Flagellar Motor Switching

The high-resolution structures of nearly all the proteins that comprise the bacterial flagellar motor switch complex have been solved; yet a clear picture of the switching mechanism has not emerged. Here, we used NMR to characterize the interaction modes and solution properties of a number of these proteins, including several soluble fragments of the flagellar motor proteins FliM and FliG, and the response-regulator CheY. We find that activated CheY, the switch signal, binds to a previously unidentified region of FliM, adjacent to the FliM-FliM interface. We also find that activated CheY and FliG bind with mutual exclusivity to this site on FliM, because their respective binding surfaces partially overlap. These data support a model of CheY-driven motor switching wherein the binding of activated CheY to FliM displaces the carboxy-terminal domain of FliG (FliG_C) from FliM, modulating the FliG_C-MotA interaction, and causing the motor to switch rotational sense as required for chemotaxis.     

Chemotactic control of the two flagellar systems of Rhodobacter sphaeroides is mediated by different sets of CheY and FliM proteins

Rhodobacter sphaeroides expresses two different flagellar systems, a subpolar flagellum (fla1) and multiple polar flagella (fla2). These structures are encoded by different sets of flagellar genes. The chemotactic control of the subpolar flagellum (fla1) is mediated by three of the six different CheY proteins (CheY6, CheY4, or CheY3). We show evidence that CheY1, CheY2, and CheY5 control the chemotactic behavior mediated by fla2 flagella and that RSP6099 encodes the fla2 FliM protein.     

Acetylation represses the binding of CheY to its target proteins

The ability of CheY, the response regulator of bacterial chemotaxis, to generate clockwise rotation is regulated by two covalent modifications – phosphorylation and acetylation. While the function and signal propagation of the former are widely understood, the mechanism and role of the latter are still obscure. To obtain information on the function of this acetylation, we non‐enzymatically acetylated CheY to a level similar to that found in vivo, and examined its binding to its kinase CheA, its phosphatase CheZ and the switch protein FliM – its target at the flagellar switch complex. Acetylation repressed the binding to all three proteins. These results suggest that both phosphorylation and acetylation determine CheY's ability to bind to its target proteins, thus providing two levels of regulation, fast and slow respectively. The fast level is modulated by environmental signals (e.g. chemotactic and thermotactic stimuli). The slow one is regulated by the metabolic state of the cell and it determines, at each metabolic state, the fraction of CheY molecules that can participate in signalling.     

A chemotactic signaling surface on CheY defined by suppressors of flagellar switch mutations.

CheY is the response regulator protein that interacts with the flagellar switch apparatus to modulate flagellar rotation during chemotactic signaling. CheY can be phosphorylated and dephosphorylated in vitro, and evidence indicates that CheY-P is the activated form that induces clockwise flagellar rotation, resulting in a tumble in the cell's swimming pattern. The flagellar switch apparatus is a complex macromolecular structure composed of at least three gene products, FliG, FliM, and FliN. Genetic analysis of Escherichia coli has identified fliG and fliM as genes in which mutations occur that allele specifically suppress cheY mutations, indicating interactions among these gene products. We have generated a class of cheY mutations selected for dominant suppression of fliG mutations. Interestingly, these cheY mutations dominantly suppressed both fliG and fliM mutations; this is consistent with the idea that the CheY protein interacts with both switch gene products during signaling. Biochemical characterization of wild-type and suppressor CheY proteins did not reveal altered phosphorylation properties or evidence for phosphorylation-dependent CheY multimerization. These data indicate that suppressor CheY proteins are specifically altered in the ability to transduce chemotactic signals to the switch at some point subsequent to phosphorylation. Physical mapping of suppressor amino acid substitutions on the crystal structure of CheY revealed a high degree of spatial clustering, suggesting that this region of CheY is a signaling surface that transduces chemotactic signals to the switch.     

Correlated switch binding and signaling in bacterial chemotaxis

In Escherichia coli, swimming behavior is mediated by the phosphorylation state of the response regulator CheY. In its active, phosphorylated form, CheY exhibits enhanced binding to a switch component, FliM, at the flagellar motor, which induces a change from counterclockwise to clockwise flagellar rotation. When Ile(95) of CheY is replaced by a valine, increased clockwise rotation correlates with enhanced binding to FliM. A possible explanation for the hyperactivity of this mutant is that residue 95 affects the conformation of nearby residues that potentially interact with FliM. In order to assess this possibility directly, the crystal structure of CheY95IV was determined. We found that CheY95IV is structurally almost indistinguishable from wild-type CheY. Several other mutants with substitutions at position 95 were characterized to establish the structural requirements for switch binding and clockwise signaling at this position and to investigate a general relationship between the two properties. The various rotational phenotypes of these mutants can be explained solely by the amount of phosphorylated CheY bound to the switch, which was inferred from the phosphorylation properties of the mutant CheY proteins and their binding affinities to FliM. Combined genetic, biochemical, and crystallographic results suggest that residue 95 itself is critical in mediating the surface complementarity between CheY and FliM.     

The N terminus of FliM is essential to promote flagellar rotation in Rhodobacter sphaeroides

FliM is part of the flagellar switch complex. Interaction of this protein with phospho-CheY (CheY-P) through its N terminus constitutes the main information relay point between the chemotactic system and the flagellum. In this work, we evaluated the role of the N terminus of FliM in the swimming behavior of Rhodobacter sphaeroides. Strains expressing the FliM protein with substitutions in residues previously reported in Escherichia coli as being important for interaction with CheY showed an increased stop frequency compared with wild-type cells. In accordance, we observed that R. sphaeroides cells expressing FliM lacking either the first 13 or 20 amino acids from the N terminus showed a stopped phenotype. We show evidence that FliMDelta13 and FliMDelta20 are stable proteins and that cells expressing them allow flagellin export at levels indistinguishable from those detected for the wild-type strain. These results suggest that the N-terminal region of FliM is required to promote swimming in this bacterium. The role of CheY in controlling flagellar rotation in this organism is discussed.     

Identification of the binding interfaces on CheY for two of its targets, the phosphatase CheZ and the flagellar switch protein fliM.

CheY is the response regulator protein serving as a phosphorylation-dependent switch in the bacterial chemotaxis signal transduction pathway. CheY has a number of proteins with which it interacts during the course of the signal transduction pathway. In the phosphorylated state, it interacts strongly with the phosphatase CheZ, and also the components of the flagellar motor switch complex, specifically with FliM. Previous work has characterized peptides consisting of small regions of CheZ and FliM which interact specifically with CheY. We have quantitatively measured the binding of these peptides to both unphosphorylated and phosphorylated CheY using fluorescence spectroscopy. There is a significant enhancement of the binding of these peptides to the phosphorylated form of CheY, suggesting that these peptides share much of the binding specificity of the intact targets of the phosphorylated form of CheY. We also have used modern nuclear magnetic resonance methods to characterize the sites of interaction of these peptides on CheY. We have found that the binding sites are overlapping and primarily consist of residues in the C-terminal portion of CheY. Both peptides affect the resonances of residues at the active site, indicating that the peptides may either bind directly at the active site or exert conformational influences that reach to the active site. The binding sites for the CheZ and FliM peptides also overlap with the previously characterized CheA binding interface. These results suggest that interaction with these three proteins of the signal transduction pathway are mutually exclusive. In addition, since these three proteins are sensitive to the phosphorylation state of CheY, it may be that the C-terminal region of CheY is most sensitive for the conformational changes occurring upon phosphorylation.     

The structures of T87I phosphono-CheY and T87I/Y106W phosphono-CheY help to explain their binding affinities to the FliM and CheZ peptides

CheY is a response regulator in bacterial chemotaxis. Escherichia coli CheY mutants T87I and T87I/Y106W CheY are phosphorylatable on Asp57 but unable to generate clockwise rotation of the flagella. To understand this phenotype in terms of structure, stable analogs of the two CheY-P mutants were synthesized: T87I phosphono-CheY and T87I phosphono-CheY. Dissociation constants for peptides derived from flagellar motor protein FliM and phosphatase CheZ were determined for phosphono-CheY and the two mutants. The peptides bind phosphono-CheY almost as strongly as CheY-P; however, they do not bind T87I phosphono-CheY or T87I/Y106W phosphono-CheY, implying that the mutant proteins cannot bind FliM or CheZ tightly in vivo. The structures of T87I phosphono-CheY and T87I/Y106W phosphono-CheY were solved to resolutions of 1.8 and 2.4 Å, respectively. The increased bulk of I87 forces the side-chain of Y106 or W106, into a more solvent-accessible conformation, which occludes the peptide-binding site.     

Changing the direction of flagellar rotation in bacteria by modulating the ratio between the rotational states of the switch protein FliM

One of the major questions in bacterial chemotaxis is how the switch, which controls the direction of flagellar rotation, functions. It is well established that binding of the signaling molecule CheY to the switch protein FliM shifts the rotation from the default direction, counterclockwise, to clockwise. How this shift is done is still a mystery. Our aim in this study was to determine the correlation between the fraction of FliM molecules in the clockwise state (i.e. occupied by CheY) and the probability of clockwise rotation. For this purpose we gradually expressed, from a plasmid, a clockwise FliM mutant protein in cells that express, from the chromosome, wild-type FliM but no chemotaxis proteins. We verified that plasmid-borne FliM exchanges chromosomal FliM in the switch. Surprisingly, a substantial clockwise probability was not obtained before the large majority of the FliM molecules in the switch were clockwise molecules. Thereafter, the rise in clockwise probability was very steep. These results suggest that an increase in the clockwise probability requires a high level of FliM occupancy by CheY approximately P. They further suggest that the steep increase in clockwise rotation upon increasing CheY levels, reported in several studies, is due, at least in part, to cooperativity of post-binding interactions within the switch. We also carried out the inverse experiment, in which wild-type FliM was gradually expressed in a background of a clockwise fliM mutant. In this case, the level of the clockwise mutant protein, required for establishing a certain clockwise probability, was lower than in the original experiment. If our system (in which the ratio between the rotational states of FliM in the switch is established by slow exchange) and the native system (in which the ratio is established by fast changes in FliM occupancy) are comparable, the results suggest that hysteresis is involved in the switch function. Such a situation might reflect a damping mechanism, which prevents a situation in which fluctuations in the phosphorylation level of CheY throw the switch from one direction of rotation to the other.     

Crystal structure of an activated response regulator bound to its target

The chemotactic regulator CheY controls the direction of flagellar rotation in Escherichia coli. We have determined the crystal structure of BeF3--activated CheY from E. coli in complex with an N-terminal peptide derived from its target, FliM. The structure reveals that the first seven residues of the peptide pack against the beta4-H4 loop and helix H4 of CheY in an extended conformation, whereas residues 8-15 form two turns of helix and pack against the H4-beta5-H5 face. The peptide binds the only region of CheY that undergoes noticeable conformational change upon activation and would most likely be sandwiched between activated CheY and the remainder of FliM to reverse the direction of flagellar rotation.     

Switched or not?: the structure of unphosphorylated CheY bound to the N terminus of FliM

Phosphorylation of Escherichia coli CheY increases its affinity for its target, FliM, 20-fold. The interaction between BeF(3)(-)-CheY, a phosphorylated CheY (CheY approximately P) analog, and the FliM sequence that it binds has been described previously in molecular detail. Although the conformation that unphosphorylated CheY adopts in complex with FliM was unknown, some evidence suggested that it is similar to that of CheY approximately P. To resolve the issue, we have solved the crystallographic structure of unphosphorylated, magnesium(II)-bound CheY in complex with a synthetic peptide corresponding to the target region of FliM (the 16 N-terminal residues of FliM [FliM(16)]). While the peptide conformation and binding site are similar to those of the BeF(3)(-)-CheY-FliM(16) complex, the inactive CheY conformation is largely retained in the unphosphorylated Mg(2+)-CheY-FliM(16) complex. Communication between the target binding site and the phosphorylation site, observed previously in biochemical experiments, is enabled by a network of conserved side chain interactions that partially mimic those observed in BeF(3)(-)-activated CheY. This structure makes clear the active role that the beta4-alpha4 loop plays in the Tyr(87)-Tyr(106) coupling mechanism that enables allosteric communication between the phosphorylation site and the target binding surface. Additionally, this structure provides a high-resolution view of an intermediate conformation of a response regulator protein, which had been generally assumed to be two state.     

Domain Analysis of the FliM Protein of Escherichia coli

The FliM protein of Escherichia coli is required for the assembly and function of flagella. Genetic analyses and binding studies have shown that FliM interacts with several other flagellar proteins, including FliN, FliG, phosphorylated CheY, other copies of FliM, and possibly MotA and FliF. Here, we examine the effects of a set of linker insertions and partial deletions in FliM on its binding to FliN, FliG, CheY, and phospho-CheY and on its functions in flagellar assembly and rotation. The results suggest that FliM is organized into multiple domains. A C-terminal domain of about 90 residues binds to FliN in coprecipitation experiments, is most stable when coexpressed with FliN, and has some sequence similarity to FliN. This C-terminal domain is joined to the rest of FliM by a segment (residues 237 to 247) that is poorly conserved, tolerates linker insertion, and may be an interdomain linker. Binding to FliG occurs through multiple segments of FliM, some in the C-terminal domain and others in an N-terminal domain of 144 residues. Binding of FliM to CheY and phospho-CheY was complex. In coprecipitation experiments using purified FliM, the protein bound weakly to unphosphorylated CheY and more strongly to phospho-CheY, in agreement with previous reports. By contrast, in experiments using FliM in fresh cell lysates, the protein bound to unphosphorylated CheY about as well as to phospho-CheY. Determinants for binding CheY occur both near the N terminus of FliM, which appears most important for binding to the phosphorylated protein, and in the C-terminal domain, which binds more strongly to unphosphorylated CheY. Several different deletions and linker insertions in FliM enhanced its binding to phospho-CheY in coprecipitation experiments with protein from cell lysates. This suggests that determinants for binding phospho-CheY may be partly masked in the FliM protein as it exists in the cytoplasm. A model is proposed for the arrangement and function of FliM domains in the flagellar motor.     

Binding of the chemotaxis response regulator CheY to the isolated,intact switch complex of the bacterial flagellar motor: lack of cooperativity

In bacteria, the chemotactic signal is greatly amplified between the chemotaxis receptors and the flagellar motor. In Escherichia coli, part of this amplification occurs at the flagellar switch. However, it is not known whether the amplification results from cooperativity of CheY binding to the switch or from a post-binding step. To address this question, we purified the intact switch complex (constituting the switch proteins FliG, FliM, and FliN and the scaffolding protein FliF) in quantities sufficient for biochemical work and used it to investigate whether the binding of CheY to the switch complex is cooperative. As a negative control, we used complexes of switchless basal bodies, formed from the proteins FliF and FliG and similarly isolated. Using double-labeling centrifugation assays for binding, we found that CheY binds to the isolated, intact switch complex in a phosphorylation-dependent manner. We observed no significant phosphorylation-dependent binding to the negative control of the switchless basal body. The dissociation constant for the binding between the switch complex and phosphorylated CheY (CheY approximately P) was 4.0 +/- 1.1 microm, well in line with the published range of CheY approximately P concentrations to which the flagellar motor is responsive. Furthermore, the binding was not cooperative (Hill coefficient approximately 1). This lack of CheY approximately P-switch complex binding cooperativity, taken together with earlier in vivo studies suggesting that the dependence of the rotational state of the motor on the fraction of occupied sites at the switch is sigmoidal and very steep (Bren, A., and Eisenbach, M. (2001) J. Mol. Biol. 312, 699-709), indicates that the chemotactic signal is amplified within the switch, subsequent to the CheY approximately P binding.     

Interactions of the chemotaxis signal protein CheY with bacterial flagellar motors visualized by evanescent wave microscopy

The chemotaxis signal protein CheY of enteric bacteria shuttles between transmembrane methyl-accepting chemotaxis protein (MCP) receptor complexes and flagellar basal bodies [1]. The basal body C-rings, composed of the FliM, FliG and FliN proteins, form the rotor of the flagellar motor [2]. Phosphorylated CheY binds to isolated FliM [3] and may also interact with FliG [4], but its binding to basal bodies has not been measured. Using the chemorepellent acetate to phosphorylate and acetylate CheY [5], we have measured the covalent-modification-dependent binding of a green fluorescent protein-CheY fusion (GFP-CheY) to motor assemblies in bacteria lacking MCP complexes by evanescent wave microscopy [6]. At acetate concentrations that cause solely clockwise rotation, GFP-CheY molecules bound to native basal bodies or to overproduced rotor complexes with a stoichiometry comparable to the number of C-ring subunits. GFP-CheY did not bind to rotors lacking FIiM/FliN, showing that these subunits are essential for the association. This assay provides a new means of monitoring protein-protein interactions in signal transduction pathways in living cells.     

Molecular analysis of the flagellar switch protein FliM of Salmonella typhimurium.

Defects in the chemotaxis proteins CheY and CheZ of Salmonella typhimurium can be suppressed by mutations in the flagellar switch, such that swarming of a pseudorevertant on semisolid plates is significantly better than that of its parent. cheY suppressors contribute to a clockwise switch bias, and cheZ suppressors contribute to a counterclockwise bias. Among the three known switch genes, fliM contributes most examples of such suppressor mutations. We have investigated the changes in FliM that are responsible for suppression, as well as the changes in CheY or CheZ that are being compensated for. Ten independently isolated parental cheY mutations represented nine distinct mutations, one an amino acid duplication and the rest missense mutations. Several of the altered amino acids lie on one face of the three-dimensional structure of CheY (A. M. Stock, J. M. Mottonen, J. B. Stock, and C. E. Schutt, Nature (London) 337:745-749, 1989; K. Volz and P. Matsumura, J. Biol. Chem. 266:15511-15519, 1991); this face may constitute the binding site for the switch. All 10 cheZ mutations were distinct, with several of them resulting in premature termination. cheY and cheZ suppressors in FliM occurred in clusters, which in general did not overlap. A few cheZ suppressors and one cheY suppressor involved changes near the N terminus of FliM, but neither cheY nor cheZ suppressors involved changes near the C terminus. Among the strongest cheY suppressors were changes from Arg to a neutral amino acid or from Val to Glu, suggesting that electrostatic interactions may play an important role in switching. A given cheY or cheZ mutation could be suppressed by many different fliM mutations; conversely, a given fliM mutation was often encountered as a suppressor of more than one cheY or cheZ mutation. The data suggest that an important factor in suppression is a balancing of the shift in switch bias introduced by alteration of CheY or CheZ with an appropriate opposing shift introduced by alteration of FliM. For strains with a severe parental mutation, such as the cheZ null mutations, adjustment of switch bias is essentially the only factor in suppression, since the attractant L-aspartate caused at most a slight further enhancement of the swarming rate over that occurring in the absence of a chemotactic stimulus. We discuss a model for switching in which there are distinct interactions for the counterclockwise and clockwise states, with suppression occurring by impairment of one of the states and hence by relative enhancement of the other state. FliM can also undergo amino acid changes that result in a paralyzed (Mot-) phenotype; these changes were confined to a very few residues in the protein.     

Deletion analysis of the FliM flagellar switch protein of Salmonella typhimurium.

The flagellar switch of Salmonella typhimurium and Escherichia coli is composed of three proteins, FliG, FliM, and FliN. The switch complex modulates the direction of flagellar motor rotation in response to information about the environment received through the chemotaxis signal transduction pathway. In particular, chemotaxis protein CheY is believed to bind to switch protein FliM, inducing clockwise filament rotation and tumbling. To investigate the function of FliM and its interactions with FliG and FliN, we engineered a series of 34 FliM deletion mutant proteins, each lacking a different 10-amino-acid segment. We have determined the phenotype associated with each mutant protein, the ability of each mutant protein to interfere with the motility of wild-type cells, and the effect of additional FliG and FliN on the function of selected FliM mutant proteins. Overall, deletions at the N terminus produced a counterclockwise switch bias, deletions in the central region of the protein produced poorly motile or nonflagellate cells, and deletions near the C terminus produced only nonflagellate cells. On the basis of this evidence and the results of a previous study of spontaneous FliM mutants (H. Sockett, S. Yamaguchi, M. Kihara, V. M. Irikura, and R. M. Macnab, J. Bacteriol. 174:793-806, 1992), we propose a division of the FliM protein into four functional regions: an N-terminal region primarily involved in switching, an extended N-terminal region involved in switching and assembly, a middle region involved in switching and motor rotation, and a C-terminal region primarily involved in flagellar assembly.     

Interaction of CheY2 and CheY2-P with the cognate CheA kinase in the chemosensory-signalling chain of Sinorhizobium meliloti

An unusual regulatory mechanism involving two response regulators, CheY1 and CheY2, but no CheZ phosphatase, operates in the chemotactic signalling chain of Sinorhizobium meliloti . Active CheY2-P, phosphorylated by the cognate histidine kinase, CheA, is responsible for flagellar motor control. In the absence of any CheZ phosphatase activity, the level of CheY2-P is quickly reset by a phospho-transfer from CheY2-P first back to CheA, and then to CheY1, which acts as a phosphate sink. In studying the mechanism of this phosphate shuttle, we have used GFP fusions to show that CheY2, but not CheY1, associates with CheA at a cell pole. Cross-linking experiments with the purified proteins revealed that both CheY2 and CheY2-P bind to an isolated P2 ligand-binding domain of CheA, but CheY1 does not. The dissociation constants of CheA–CheY2 and CheA–CheY2-P indicated that both ligands bind with similar affinity to CheA. Based on the NMR structures of CheY2 and CheY2-P, their interactions with the purified P2 domain were analysed. The interacting surface of CheY2 comprises its C-terminal β4-α4-β5-α5 structural elements, whereas the interacting surface of CheY2-P is shifted towards the loop connecting β5 and α5. We propose that the distinct CheY2 and CheY2-P surfaces interact with two overlapping sites in the P2 domain that selectively bind either CheY2 or CheY2-P, depending on whether CheA is active or inactive.     

Only one of the five CheY homologs in Vibrio cholerae directly switches flagellar rotation

Vibrio cholerae has three sets of chemotaxis (Che) proteins, including three histidine kinases (CheA) and four response regulators (CheY) that are encoded by three che gene clusters. We deleted the cheY genes individually or in combination and found that only the cheY3 deletion impaired chemotaxis, reinforcing the previous conclusion that che cluster II is involved in chemotaxis. However, this does not exclude the involvement of the other clusters in chemotaxis. In other bacteria, phospho-CheY binds directly to the flagellar motor to modulate its rotation, and CheY overexpression, even without CheA, causes extremely biased swimming behavior. We reasoned that a V. cholerae CheY homolog, if it directly controls flagellar rotation, should also induce extreme swimming behavior when overproduced. This was the case for CheY3 (che cluster II). However, no other CheY homolog, including the putative CheY (CheY0) protein encoded outside the che clusters, affected swimming, demonstrating that these CheY homologs cannot act directly on the flagellar motor. CheY4 very slightly enhanced the spreading of an Escherichia coli cheZ mutant in semisolid agar, raising the possibility that it can affect chemotaxis by removing a phosphoryl group from CheY3. We also found that V. cholerae CheY3 and E. coli CheY are only partially exchangeable. Mutagenic analyses suggested that this may come from coevolution of the interacting pair of proteins, CheY and the motor protein FliM. Taken together, it is likely that the principal roles of che clusters I and III as well as cheY0 are to control functions other than chemotaxis.     

Localization of components of the chemotaxis machinery of Escherichia coli using fluorescent protein fusions

We prepared fusions of yellow fluorescent protein [the YFP variant of green fluorescent protein (GFP)] with the cytoplasmic chemotaxis proteins CheY, CheZ and CheA and the flagellar motor protein FliM, and studied their localization in wild-type and mutant cells of Escherichia coli. All but the CheA fusions were functional. The cytoplasmic proteins CheY, CheZ and CheA tended to cluster at the cell poles in a manner similar to that observed earlier for methyl-accepting chemotaxis proteins (MCPs), but only if MCPs were present. Co-localization of CheY and CheZ with MCPs was CheA dependent, and co-localization of CheA with MCPs was CheW dependent, as expected. Co-localization with MCPs was confirmed by immunofluorescence using an anti-MCP primary antibody. The motor protein FliM appeared as discrete spots on the sides of the cell. These were seen in wild-type cells and in a fliN mutant, but not in flhC or fliG mutants. Co-localization with flagellar structures was confirmed by immunofluorescence using an antihook primary antibody. Surprisingly, we did not observe co-localization of CheY with motors, even under conditions in which cells tumbled.