HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 26, 19752-19758, June 30, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/26/19752    most recent M909908199v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Schuster, M. Articles by Collins, E. J. Search for Related Content PubMed PubMed Citation Articles by Schuster, M. Articles by Collins, E. J. Social Bookmarking                      What's this?

Correlated Switch Binding and Signaling in Bacterial Chemotaxis^*

Martin Schuster, Rui Zhao, Robert B. Bourret, and Edward J. Collins

From the Department of Microbiology and Immunology, University of North Carolina, Chapel Hill, North Carolina 27599-7290

Received for publication, December 10, 1999, and in revised form, March 23, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

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 Ile⁹⁵ 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.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Motile bacteria such as Escherichia coli respond to a variety of environmental stimuli by modulating the rotational direction of their flagella. Counterclockwise (CCW)¹ rotation results in a motion called smooth swimming, whereas a change to clockwise (CW) rotation leads to tumbling of the bacterial cell. Chemotaxis toward attractants or away from repellants is achieved by extending smooth runs that carry the cell in the favorable direction. This swimming behavior is controlled 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 that results in CW rotation. CheY is phosphorylated by the histidine kinase CheA, whose activity depends on the signaling state of coupled transmembrane chemoreceptors, and subsequently, is dephosphorylated with the assistance of CheZ. In addition, CheY autophosphorylates using small molecule phosphodonors and spontaneously dephosphorylates by its intrinsic autophosphatase activity (for review, see Refs. 1-4).

CheY serves as a prototype for the investigation of response regulator function in bacterial signal transduction. Although much has been learned about its activation mechanism, the molecular basis of how the interaction of CheY with the flagellar switch results in a reversal of flagellar rotation is not yet known. A surface on CheY that interacts with FliM has been deduced from genetic and biophysical studies. Critical residues have been inferred from mutations in cheY that suppress mutations in fliM (5-7) and from investigating chemical shifts in NMR spectra of CheY upon binding to an N-terminal peptide of FliM (8). It has been proposed that Tyr¹⁰⁶, located on the putative FliM binding face, plays a major role in the interaction of CheY with the flagellar switch. The side chain of Tyr¹⁰⁶ was found in an inside conformation in the crystal structure of an activated mutant, CheY106YW, and a solvent-exposed, outside conformation in the structures of two loss-of-function mutants, CheY87TI and CheY87TI106YW (9, 10). This apparent correlation suggested that the conformation of residue 106 may affect the signaling state of CheY. These and several other mutants with substitutions at position 106, however, displayed unaltered FliM binding and phosphorylation properties in vitro compared with wild-type CheY (11), which suggested that Tyr¹⁰⁶ participates in a signaling step subsequent to binding to the switch (10, 11). Recent crystallographic studies of the phosphorylated forms of two other response regulators, FixJ from Rhizobium meliloti (12) and SpoOA from Bacillus subtilis (13), support the notion that the conformation of the conserved phenylalanine/tyrosine residue (corresponding to Tyr¹⁰⁶ in CheY) is important in determining the signaling state of this class of proteins.

Characterization of another hyperactive mutant, CheY95IV, suggested a role for Ile⁹⁵ in switch interaction (14). The phosphorylation-dependent constitutive signaling of this mutant correlated with enhanced binding to FliM in vitro, thus providing a likely explanation for its phenotype. Ile⁹⁵ might directly interact with the switch or alter the nearby proposed FliM binding surface. Alternatively, it might influence the side chain conformation of the adjacent residue Tyr¹⁰⁶. To address these possibilities directly, and to explore a general relationship between switch binding and signaling, we determined the crystal structure of CheY95IV and characterized several additional CheY mutants with single site substitutions at position 95.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Bacterial Strains, Plasmids, and Mutant Constructions-- The E. coli (cheY)m60-21 strain KO641recA, the (cheA)m102-11 strain RBB382, and the p_trp cheYZ plasmid pRBB40 have been described (15). Derivatives of pRBB40 carrying the cheY alleles 13DK (15) or 95IV (16) also have previously been reported. In this work, versions of pRBB40 carrying the cheY alleles 95IA, 95ID, 95IK, or 95IM were constructed using dut ung mutagenesis (15, 17); cheY allele 95IG was constructed using splicing by overlap-extension polymerase chain reaction (18, 19), and cheY allele 13DK106YW was constructed by recombining appropriate restriction fragments in vitro using standard techniques. All mutations were confirmed by sequencing the entire cheY genes of the various pRBB40 plasmids.

Behavioral Assay-- Cells were prepared for tethering as described (14, 20). The cells, tethered to a glass slide, were analyzed under a dark-field microscope, which was connected to a fully automated video analysis and image processing system (Hobson Tracking Systems Ltd., Sheffield, United Kingdom). The Hobson Tracker Arot software determines the rotational direction of tethered cell bodies by analyzing single video images generated at 60 Hz. The rotational direction is defined by the relative movement of the center of mass around the center of rotation, which the software assigns to each selected cell. This particular method restricted analysis to individual bacteria that were tethered at or near the poles of their cell bodies. For each strain, the behavior of a number of cells was quantified in real time for 1 min each; and the raw data from the Hobson tracker output files were converted into a bias value using a program written by Matthew Levin (University of Cambridge, United Kingdom). The rotational bias was defined to be the arithmetic mean of the fraction of time all observed cells of a particular strain spent in the CCW spinning mode. Individual cells pausing more than 5% of the total recorded time were excluded from the calculation of a final bias.

Protein Purification and Peptide Synthesis-- E. coli wild-type and mutant CheY proteins were purified from overexpressing strains as described previously (21). Concentrations were determined with the Bio-Rad DC colorimetric assay. The FliM peptide corresponding to the N-terminal 16 residues of FliM (MGDSILSQAEIDALLN), purified to a minimum chromatographic homogeneity of 90%, was obtained from Macromolecular Resources (Fort Collins, CO).

Protein Phosphorylation Assays-- Fluorescence measurements of CheY phosphorylation reactions were performed as described by Silversmith et al. (22). Excitation and emission slit widths of the spectrofluorimeter were adjusted to accommodate the respective protein concentrations in the different assays. Time courses of approach to steady-state phosphorylation that yielded an observed rate constant (k_obs) were performed at 17 °C. The reactions were carried out in 100 mM HEPES, pH 7.0, and 10 mM MgCl₂. For these measurements, 1.6 µM CheY (wild-type or mutant) was mixed with the phosphodonor phosphoramidate (PAM) to give a final concentration of 100 mM. PAM was synthesized as described previously (23). The values measured at 17 °C were converted to k_obs values at 25 °C using the Arrhenius equation and assuming an activation energy for k_obs of 13.9 kcal/mol (22). Time courses that yielded autodephosphorylation rate constants (k_dephos) were measured at 25 °C, using a published pH jump assay (24). This assay takes advantage of the pH dependence of the CheY phosphorylation reaction when PAM serves as a phosphodonor (22). At sufficiently high pH, CheY phosphorylation is eliminated and the time course of approach to steady state therefore only reflects the pH-independent dephosphorylation of CheY-P. We incorporated the following modifications: A solution containing 20 µM CheY (wild-type or mutant) and 50 mM PAM in 5 mM Tris, pH 7.0, and 10 mM MgCl₂ was mixed 1:1 with a concentrated high pH buffer containing 150 mM CAPS, pH 10.5. MgCl₂ and PAM were omitted from this buffer to reduce precipitation problems.

The kinetic results were analyzed according to a reaction scheme by Lukat et al. (25). Under the reaction conditions used here, CheY phosphorylation by PAM can be treated as a first order reaction with respect to the phosphodonor concentration with no indication of saturation at high substrate concentrations (24, 26). k_obs can therefore be expressed as follows:

(Eq. 1)

The quantity of (k_phos/K_S)[PAM] represents an effective rate constant for phosphorylation. Although the autophosphorylation rate constant k_phos and the dissociation constant of the CheY-PAM enzyme-substrate complex K_S are unknown, (k_phos/K_S)[PAM] can be measured as the difference between k_obs and k_dephos (Eq. 1). The ratio of (k_obs  k_dephos) to k_dephos yields the ratio of phosphorylated to unphosphorylated CheY present in the binding reaction.

It should be noted that CheY95IA exhibited a significantly lower fluorescence quench compared with wild-type CheY when equilibrium phosphorylation had been reached after addition of 100 mM PAM. This difference, however, did not reflect significant differences in the fraction of phosphorylated CheY present in the reaction, because the measured rate constants, k_obs and k_dephos, were almost identical to those of wild-type CheY. The discrepancy is therefore probably due to differences in the fluorescence properties of the two CheY proteins.

Equilibrium titrations with CheZ were performed at 25 °C in a buffer containing 100 mM HEPES, pH 7.0, and 10 mM MgCl₂. 5.0 µM CheY (wild-type or mutant) was phosphorylated with 100 mM PAM. After equilibrium phosphorylation had been reached, increasing amounts of CheZ were added. The progress of the dephosphorylation reactions was plotted as the relative increase in fluorescence intensity versus the total CheZ concentration.

Binding Assay-- Measurement of the binding affinities between CheY mutant proteins and FliM peptide was based on a fluorimetric assay described by McEvoy et al. (8). 10 µM CheY (wild-type or mutant) was titrated with repetitive additions of FliM peptide in 100 mM HEPES, pH 7.0, and 10 mM MgCl₂ while monitoring tryptophan fluorescence. For titrations with phosphorylated CheY, 100 mM PAM was added to the sample. Buffer conditions were chosen to be identical to those in the kinetic experiments. It should be pointed out that, in contrast to the assay by McEvoy et al., Mg²⁺ was present in the binding reactions even in the absence of PAM. Binding constants were determined using Eadie-Hofstee analysis, assuming a single binding site on CheY for FliM peptide and considering the concentration of free ligand in the binding reaction. To estimate the fraction of free ligand versus ligand bound to CheY, the maximum fluorescence change when all of CheY is converted into complex (I_max), was extrapolated by fitting the titration data to a hyperbolic binding function. Interfering fluorescence from the addition of FliM peptide, probably due to residual organic solvents, became significant (maximal 10%) at high peptide concentrations and was considered in the overall calculation.

Protein Crystallization-- The purified CheY95IV protein was concentrated to 22 mg/ml in H₂O with a Centricon 10 concentrator (Amicon) for crystallization trials. Initially, 50 different crystallization conditions from a commercially available kit (Crystal Screen from Hampton Research, Laguna Niguel, CA) were screened using the hanging drop vapor diffusion method (27) at 4 °C. Each droplet contained 1 µl of protein solution and 1 µl of reservoir solution. Long needles were obtained from a solution containing 0.19 M ammonium sulfate, 0.10 M sodium acetate, pH 4.8, and 25% polyethylene glycol 4000. These crystals were used to microseed a solution containing 0.17 M ammonium sulfate, 0.09 M sodium acetate, pH 4.8, and 23% polyethylene glycol 4000, which yielded diffraction quality crystals with dimensions of 0.7 × 0.1 × 0.04 mm. Crystal seeds were obtained by 100-fold dilution of seed stock prepared with the Seed Bead kit from Hampton Research (Laguna Niguel, CA).

Crystallographic Data Collection, Structural Determination, and Refinement-- Crystals of CheY95IV were transferred to mother liquor plus 10% glycerol and were immediately placed in a cryogenic nitrogen stream (100 K, Oxford Cryo System). Diffraction data were collected using an R-Axis IIC area detector, and they were processed using programs Denzo and Scalepack (28). These crystals were nearly isomorphous to the crystals of wild-type E. coli CheY (29). Data statistics are shown in Table I.

The structure of CheY95IV was determined using the Molecular Replacement method with the wild-type E. coli CheY structure at 1.7-Å resolution (29) as a model. Alanine was substituted for Tyr¹⁰⁶ to avoid potential model bias. Because CheY95IV crystallized nearly isomorphously to wild-type CheY, the position and orientation of the wild-type structure was used to begin refinement of CheY95IV. Rigid body refinement in CNS (30) was applied to refine the position and orientation of the molecule in the asymmetric unit. The first electron density maps were generated using DM (31). Manual model building was performed with O (32), and refinement was performed with CNS (30). After refinement converged with R_free = 30.5% and R_work = 29.2%, water molecules were added with Arp (33) and were visually confirmed. Three sulfate clusters were modeled in the map and were used in the refinement. The final refinement statistics are shown in Table I. Over 95% of the residues belong to the most favored region in a Ramachandran plot (34). No residues are in the generously allowed region. However, one residue (Asn⁶²) falls into the disallowed region ( = 68.60,  = 60.80) in the Ramachandran plot. The conformation of this residue is justified by the existence of a hydrogen bond between the ND2 atom of Asn⁶² and the OD2 atom of Asp³⁸. Protein residues have occasionally been observed to have  and  angles in this region (35).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Behavioral Analysis of CheY Mutants-- To assess the structural requirements of residue 95 in CheY that are necessary to cause hyperactive signaling, several amino acid substitutions at position 95 were generated. The tethered cell assay was used to quantify the behavior of bacteria expressing these mutant alleles (Table II). As reported previously, the cheY95IV allele results in extreme CW behavior when expressed from a multicopy plasmid in a cheY host and maintains this phenotype when expressed from a single copy gene on the chromosome (14). In this study, Ile⁹⁵ was replaced with alanine, which created a CheY mutant that also showed more CW behavior than wild-type. The smaller nonpolar residues valine and alanine therefore promoted enhanced CW signaling ability. In contrast, the complete lack of a side chain at position 95 (i.e. CheY95IG) did not support CW rotation. Similarly, the presence of a larger hydrophobic side chain in CheY95IM, as well as charge substitutions in CheY95ID and CheY95IK, also resulted in loss of activity. In addition, we tested the CW signaling ability of these mutants in a cheA host (data not shown). All CheY mutants failed to support CW rotation, implying that phosphorylation is necessary for their activation.

Phosphorylation Kinetics-- Next, a possible correlation between switch binding and signaling was explored. We selected activated mutants CheY95IA and CheY95IV, as well as one representative of the loss-of-function mutants, CheY95IM, to determine their binding affinities to a peptide of FliM in vitro. To quantify FliM binding of these CheY mutants in their phosphorylated forms, it was critical to determine the fraction of CheY phosphorylated in each binding reaction. Dissociation constants were calculated based on the fraction of free ligand (i.e. FliM peptide) present in the reaction, which was deduced by subtracting the amount of ligand bound to protein (i.e. phosphorylated- or nonphosphorylated CheY) from the total amount of ligand added. Toward this goal, the autophosphorylation and autodephosphorylation properties of each CheY mutant protein were characterized using tryptophan fluorescence (Table III).

The experimentally determined rate constants (k_obs and k_dephos) for wild-type CheY resemble those obtained by Mayover et al. (24) under comparable conditions. Mutants CheY95IA and CheY95IV exhibited phosphorylation kinetics that yielded rate constants almost identical to those of wild-type CheY, leading to similar fractions of phosphorylated protein. Notably, CheY95IM displayed 2-fold enhanced autophosphorylation compared with wild-type, which, however, had little impact on the fraction of phosphorylated CheY95IM.

FliM Binding-- The binding affinities of various mutant CheY proteins to FliM were probed by monitoring quenching of intrinsic tryptophan fluorescence in CheY upon binding to a peptide of FliM in the presence and absence of PAM (8) (Fig. 1, A and B). This peptide, containing the N-terminal 16 amino acids of FliM, has been shown to specifically bind to CheY in a phosphorylation-dependent manner (36). Wild-type CheY exhibited about 11-fold tighter binding in the presence of PAM compared with the reaction without phosphodonor (Fig. 1C). Hyperactive mutant CheY95IV showed about 3- and 7-fold enhanced binding in the absence and presence of PAM compared with the respective reactions with wild-type CheY, which was more pronounced than previously observed in a cross-linking assay using intact FliM (14). For another hyperactive mutant, CheY95IA, however, no specific binding of FliM peptide could be detected in the absence of PAM (within the sensitivity of this assay), and even with PAM present, binding was 5-fold lower than that of wild-type CheY in the presence of PAM. In contrast to CheY95IV, the weaker peptide binding observed for CheY95IA therefore did not correlate with its enhanced CW flagellar rotation. CheY95IM also exhibited negligible binding to FliM peptide under both nonphosphorylating and phosphorylating conditions, consistent with its exclusively CCW behavior. The constitutively active mutants CheY13DK (15, 37) and CheY13DK106YW (38) were included in this assay to compare their binding affinities to FliM peptide with those obtained in previous assays using purified, intact FliM (39, 40). This allowed us to assess how accurately binding of the various CheY mutants to FliM peptide reflects binding to full length FliM in vitro.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Binding of FliM peptide to CheY mutants. A and B, titration curves of wild-type CheY (, ), CheY95IA (, ), CheY95IM (,), and CheY95IV (,) in the presence and absence of PAM (filled and open symbols, respectively). The relative decrease in fluorescence intensity upon sequential addition of FliM peptide is shown. The data were fit to a hyperbolic binding function. Note the different scales on the abscissas in A and B. The inset in A shows Eadie-Hofstee plots, which were used to determine dissociation constants (KD). The slope of a linear fit to the data yielded KD. C, calculated binding affinities. The values shown in the bar graph represent the reciprocal of the respective KD values normalized to the KD of wild-type CheY in the absence of phosphodonor. White bars denote the absence, and black bars denote the presence of PAM. Binding reactions for CheY13DK and CheY13DK106YW were carried out in the absence of PAM but displayed in black bars because both proteins presumably represent the activated conformation (37, 38).

Description of Structural Results-- To investigate the structural basis for the constitutive signaling and enhanced FliM binding properties of CheY95IV (14), the x-ray structure of this mutant protein was determined to 1.9-Å resolution. The structure of CheY95IV is well resolved in the final electron density map as evidenced by the 92.7% real space correlation coefficient (32). The overall structure is indistinguishable from that of wild-type E. coli apo-CheY (29). A least-squares superposition of the C atoms of the two structures yielded a root mean square deviation of 0.21 Å, which is not significantly different from the 0.18-Å coordinate error obtained with the maximum likelihood refinement program REFMAC (41). The conformations of active site residues are identical between the two structures. Residues Arg¹⁹, Glu³⁴, Lys⁴⁵, Glu⁶⁷, Lys⁹², and Glu¹¹⁸ have minor side chain differences between the CheY95IV and wild-type structures. All of these residues have long and flexible side chains and are located on the external surface of CheY. Although the crystals of wild-type and mutant CheY are of the same space group, they form under different conditions; the aforementioned surface residues therefore presumably adopt different side chain conformations to adapt to the different environments. These minor differences do not seem to affect the rest of the structure or have any obvious biological consequences. The side chains of three residues (Glu³⁷, Ser⁵⁶, and Leu¹²⁷) are modeled as two discrete rotamers as they are found in the wild-type structure. Four residues (Glu²³, Asn³², Met⁸⁵, and Tyr¹⁰⁶), which have multiple conformations in wild-type CheY only demonstrate single side chain conformations in CheY95IV. Two other residues in the mutant structure (residues Glu²⁷ and Val⁹⁵) have double conformations. As expected, the 2F_o  F_c map and the simulated annealing omit map (42, 43) both demonstrate electron densities appropriate for a valine at position 95 (not shown). These maps, together with the difference map, also indicate that Val⁹⁵ exists in two distinct conformations (Fig. 2). Among the residues that are within 10 Å from Val⁹⁵, the only notable differences from the wild-type structure come from the side chains of Tyr¹⁰⁶ and Glu²⁷ of a symmetry-related molecule.

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Residue Tyr106 predominantly adopts the outside conformation in CheY95IV instead of the double conformations seen in the wild-type structure (10). Shown here is a stereo view of the final 2Fo  Fc electron density map (contoured at 1) of the Tyr106 side chain for the structure of CheY95IV. The outside conformation of Tyr106 in CheY95IV fits well in this density (shown in black). For comparison, the inside conformation from the wild-type structure is displayed in red. Note that this conformation has a strained C-C-C angle of 135°, which is roughly 20° off the ideal value for this bond angle. Val95 is shown in its two conformations (green and black lines) as inferred from the electron density at this position.

Previous studies have suggested that the conformation of Tyr¹⁰⁶ plays an important role in signal transduction by CheY (10, 11). The side chain of residue Tyr¹⁰⁶ is found in two conformations in the wild-type CheY structure (29), one occupying a hydrophobic cavity, the other pointing out to the solvent. In CheY95IV, the electron density for the outside conformation of Tyr¹⁰⁶ is well defined (Fig. 2). Because the side chain of Tyr¹⁰⁶ was deleted during the entire structural determination and refinement process (except for the last cycle), this density distribution is not a result of model bias, which was confirmed with a simulated annealing omit map (42, 43). A small patch of electron density is found in the inside pocket (Fig. 2), which is currently modeled as two water molecules. In the wild-type CheY structure, the electron densities for both the inside and outside conformations of Tyr are well defined (29). However, the bond angle between C, C, and C in the inside position is significantly distorted (20^o off the ideal bond angle). The density in the pocket of CheY95IV, in contrast, is very small compared with a phenol ring, and even after significant distortion of the bond angle between C, C, and C, only the CD2 and CE2 atoms of the phenol ring fit this density. Although we cannot completely rule out the possibility that in a small population of CheY molecules Tyr¹⁰⁶ assumes this strained conformation, it is clear that Tyr¹⁰⁶ predominantly points outward. It can also be excluded that residues from a symmetry-related molecule located in the vicinity of Tyr¹⁰⁶ have any more influence on the position of Tyr¹⁰⁶ in CheY95IV than in the isomorphous wild-type structure. In summary, the only significant differences between wild-type CheY and CheY95IV are an almost isosteric change from isoleucine to valine at position 95 and the predominant outside conformation of Tyr¹⁰⁶.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The Significance of the Conformation of Tyr¹⁰⁶ in Switch Interaction-- The Tyr¹⁰⁶ side chain in CheY can occupy two orientations, a solvent-exposed outside or a solvent-inaccessible inside position where it is packed in the interior of a hydrophobic pocket (10, 29). Our crystallographic results show that, in the structure of CheY95IV, Tyr¹⁰⁶ is restricted to the outside conformation (Fig. 2). It is not clear what leads to this position as opposed to the dual conformation in wild-type CheY (29), because the extra CD1 atom of isoleucine in the wild-type structure does not have any direct interaction with Tyr¹⁰⁶. One possible explanation is that the inside conformation of Tyr¹⁰⁶ in the wild-type structure is highly strained. The change from an isoleucine to a slightly smaller valine at position 95 may have enabled the tyrosine ring to swing out more easily to avoid the highly strained inside conformation. However, it is puzzling why Tyr¹⁰⁶ would adopt the inside conformation in the wild-type structure at all.

The conformational heterogeneity of the Tyr¹⁰⁶ side chain has been suggested to correlate with the activation state of CheY, where in the simplest model "in" reflects the active form and "out" the inactive form (10). The structure and phenotype of CheY95IV seemingly argue against this model. One needs to consider, however, that the mutant structures CheY87TI, CheY87TI106YW, and CheY106YW on which this model has been based as well as the structure of CheY95IV resemble the nonphosphorylated, presumably inactive form of CheY. After completion of the present study, high resolution structures of the phosphorylated forms of two other response regulators were published (12, 13). In both cases, relocation of the conserved threonine to interact directly with the phosphoryl group on the conserved aspartate caused a coordinated reorientation of the conserved phenylalanine/threonine (Tyr¹⁰⁶ in CheY) from the outside to the inside conformation. The same coupled rearrangement was observed in the very recently completed NMR structure of beryllofluoride-activated CheY (44). It is very likely that Tyr¹⁰⁶ in phosphorylated CheY95IV adopts the inside conformation as well, because the structure of the nonphosphorylated form shows no alterations of active site residues that could impair the aforementioned coupled rearrangement upon phosphorylation.

Residue 95 Mediates the Surface Complementarity between CheY and FliM-- CheY95IV exhibited enhanced binding to FliM peptide in both the phosphorylated and nonphosphorylated forms (Fig. 1). In the crystal structure of CheY95IV (which represents the nonphosphorylated form) no changes occurred at the proposed FliM binding surface. The altered conformation of Tyr¹⁰⁶ was the only significant change compared with wild-type CheY besides the substitution at position 95 itself (Fig. 2). In the active, phosphorylated form, the tyrosine side chain is probably buried in the interior of a hydrophobic pocket (as discussed above) and therefore does not affect FliM binding. This is supported by the unaltered FliM binding affinities of CheY mutants with substitutions at position 106 (11). Mutants with substitutions at position 95, on the other hand, do have a dramatic impact on the binding affinity of CheY to FliM peptide (Fig. 1). Taken together, these findings suggest that Ile⁹⁵ itself directly interacts with FliM. In maintaining the structural integrity of the switch binding surface on CheY, replacement of isoleucine with the slightly smaller valine might enhance the complementarity between the two proteins, allowing neighboring residues to bind to FliM more tightly. The solvent-exposed nature of the 95 side chain on -helix 4, which has been implicated in the signal transmission through hydrophobic interactions in other response regulators (12, 45), is consistent with this function. Although NMR studies provided no evidence for the direct interaction of Ile⁹⁵ with FliM peptide (8), the cocrystal structure of a CheY mutant bound to FliM peptide showed that the peptide binds very close to Ile⁹⁵, supporting our conclusion.²

The Various Phenotypes of CheY Mutants with Substitutions at Position 95 Can Be Explained by the Amount of Phosphorylated CheY Bound to the Switch-- Our behavioral studies demonstrated that substitutions at position 95 in CheY have a large impact on the cell's behavioral phenotype: One class of mutants with aspartate, glycine, lysine, and methionine substitutions completely lost its CW signaling ability, whereas the other class containing alanine and valine substitutions displayed increased CW signaling (Table II). The various substitutions at position 95 also affected the FliM-peptide binding affinities of all CheY mutants tested (Fig. 1C). The greatly impaired binding of one candidate from the first class, CheY95IM, correlated with its loss of CW signaling ability. This may be due to the bulk of the longer hydrophobic side chain of methionine in the mutant compared with isoleucine in wild-type CheY, which could interfere with the surface complementarity between CheY and FliM. Similarly, it can be envisioned that the positive and negative surface charges introduced by the lysine and alanine substitutions in CheY95IK and CheY95ID impair binding to FliM, resulting exclusively in CCW signaling. For the hyperactive mutants, interpretation of the results is more complicated: The shorter nonpolar side chains, valine and alanine, in CheY95IV and CheY95IA both promoted enhanced CW signaling, but this hyperactivity apparently did not correlate with the decreased FliM binding affinity shown by CheY95IA.

It must be considered that our FliM-peptide binding assay may not provide a complete assessment of the interactions between CheY-P and the flagellar switch complex in vivo. It has been suggested that there are parts of full length FliM not represented by the peptide that interact with CheY (46). Our results, however, showed that binding of CheY to FliM peptide (Fig. 1C) qualitatively reflects binding to intact, purified FliM measured by in vitro cross-linking (14, 39) or bead binding (40). This agreement is true for all proteins (wild-type CheY, CheY13DK, CheY95IV, and CheY13DK106YW) that have been assayed by both peptide and full length FliM and extends to the relative affinities among different mutants as well as the effects of phosphorylation on binding.

The possibility remains that, in addition to FliM binding, altered phosphorylation properties affecting phosphotransfer from CheA or sensitivity to CheZ contributed to the phenotypes of the various CheY mutants characterized here. For the activated mutants CheY95IA and CheY95IV, reduced CheZ sensitivity could potentially increase the concentration of phosphorylated CheY in the cell thereby leading to a stronger CW bias. Enhanced phosphotransfer from CheA, on the other hand, would have virtually no impact on cellular CheY-P levels, because this step is not rate-limiting (47). These explanations do not apply to CheY95IV, because both reactions are similar to those of wild-type CheY (14). In addition, we measured the CheZ sensitivity of CheY95IA by following the relative increase in fluorescence intensity due to dephosphorylation of CheY-P upon titration with CheZ. Compared with wild-type CheY, CheY95IA exhibited no increase in fluorescence even at levels of CheZ that were saturating for wild-type CheY (data not shown). This indicated that CheY95IA is resistant to the phosphatase activity of CheZ. The phenotype of this mutant can therefore be reconciled with its binding affinity to FliM, if the impact of CheZ on the phosphorylation state of CheY95IA in vivo is considered. Complete insensitivity of CheY95IA to dephosphorylation mediated by CheZ would be similar to a situation in the cell where CheZ is completely lacking. In such strains, the fraction of wild-type CheY phosphorylated has been estimated to be greater than 99.9%, compared with only about 30% in the presence of CheZ (48). We can therefore assume that the concentration of CheY95IA-P in the cell is more than 3-fold higher than in wild-type CheY-P. This compensates enough for the observed 5-fold lower binding to FliM that there is no compelling reason to postulate any other effect on the behavioral phenotype of this mutant, given the margin of error in the underlying measurements and the nonlinear relationship between switch-bound CheY-P and rotational bias.

Therefore, these results suggested that the various phenotypes of mutants with substitutions at position 95 are due to their binding affinities to FliM and the respective CheY-P levels in the cell, both of which determine the amount of CheY-P molecules bound to the switch. This is in accordance with a recent study (38) that established a functional relationship between the probability of CW or CCW flagellar rotation and the fraction of switch binding sites occupied by CheY-P. For mutants with substitutions at position 95, there is no need to postulate a postbinding step as proposed for other mutants with substitutions at position 106 (10, 11).

	ACKNOWLEDGEMENTS

We thank Ruth Silversmith for biochemical support, Megan McEvoy for providing information about the FliM-peptide binding assay prior to publication, and Karl Volz for his advice on CheY crystallization. We are indebted to Ed Westbrook and Phil Matsumura for providing information about a cocrystal of CheY bound to FliM peptide prior to publication.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant GM50860 (to R. B. B.), and by a Cancer Research Institute Fellowship (to R. Z. ).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and the structure factors (code 1D4Z) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

To whom correspondence should be addressed: Tel.: 919-966-2679; Fax: 919-962-8103; E-mail: bourret@med.unc.edu.

Published, JBC Papers in Press, March 29, 2000, DOI 10.1074/jbc.M909908199

² E. M. Westbrook and P. Matsumura, personal communication.

	ABBREVIATIONS

The abbreviations used are: CCW, counterclockwise; CW, clockwise; CheY-P, phosphorylated CheY; PAM, phosphoramidate; CAPS, 3-(cyclohexylamino)propanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. C. Lane, A. N. Simms, and H. L. T. Mobley Complex Interplay between Type 1 Fimbrial Expression and Flagellum-Mediated Motility of Uropathogenic Escherichia coli J. Bacteriol., August 1, 2007; 189(15): 5523 - 5533. [Abstract] [Full Text] [PDF]

				S. L. Porter, G. H. Wadhams, A. C. Martin, E. D. Byles, D. E. Lancaster, and J. P. Armitage The CheYs of Rhodobacter sphaeroides J. Biol. Chem., October 27, 2006; 281(43): 32694 - 32704. [Abstract] [Full Text] [PDF]

				K. I. Varughese Conformational Changes of Spo0F along the Phosphotransfer Pathway J. Bacteriol., December 15, 2005; 187(24): 8221 - 8227. [Abstract] [Full Text] [PDF]

				H. Szurmant and G. W. Ordal Diversity in Chemotaxis Mechanisms among the Bacteria and Archaea Microbiol. Mol. Biol. Rev., June 1, 2004; 68(2): 301 - 319. [Abstract] [Full Text] [PDF]

				J. G. Smith, J. A. Latiolais, G. P. Guanga, S. Citineni, R. E. Silversmith, and R. B. Bourret Investigation of the Role of Electrostatic Charge in Activation of the Escherichia coli Response Regulator CheY J. Bacteriol., November 1, 2003; 185(21): 6385 - 6391. [Abstract] [Full Text] [PDF]

				R. E. Silversmith, G. P. Guanga, L. Betts, C. Chu, R. Zhao, and R. B. Bourret CheZ-Mediated Dephosphorylation of the Escherichia coli Chemotaxis Response Regulator CheY: Role for CheY Glutamate 89 J. Bacteriol., March 1, 2003; 185(5): 1495 - 1502. [Abstract] [Full Text] [PDF]

				S. D. Re, T. Tolstykh, P. M. Wolanin, and J. B. Stock Genetic analysis of response regulator activation in bacterial chemotaxis suggests an intermolecular mechanism Protein Sci., November 1, 2002; 11(11): 2644 - 2654. [Abstract] [Full Text] [PDF]

				Y. J. Im, S.-H. Rho, C.-M. Park, S.-S. Yang, J.-G. Kang, J. Y. Lee, P.-S. Song, and S. H. Eom Crystal structure of a cyanobacterial phytochrome response regulator Protein Sci., March 1, 2002; 11(3): 614 - 624. [Abstract] [Full Text] [PDF]

				A. Bren and M. Eisenbach How Signals Are Heard during Bacterial Chemotaxis: Protein-Protein Interactions in Sensory Signal Propagation J. Bacteriol., December 15, 2000; 182(24): 6865 - 6873. [Full Text]

				R. E. Silversmith, J. G. Smith, G. P. Guanga, J. T. Les, and R. B. Bourret Alteration of a Nonconserved Active Site Residue in the Chemotaxis Response Regulator CheY Affects Phosphorylation and Interaction with CheZ J. Biol. Chem., May 18, 2001; 276(21): 18478 - 18484. [Abstract] [Full Text] [PDF]

				M. Schuster, R. E. Silversmith, and R. B. Bourret Conformational coupling in the chemotaxis response regulator CheY PNAS, May 22, 2001; 98(11): 6003 - 6008. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/26/19752    most recent M909908199v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager