JOURNAL OF BACTERIOLOGY, Mar. 2005, p. 1695–1701 0021-9193/05/$08.00⫹0 doi:10.1128/JB.187.5.1695–1701.2005 Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Vol. 187, No. 5

A Chimeric N-Terminal Escherichia coli–C-Terminal Rhodobacter sphaeroides FliG Rotor Protein Supports Bidirectional E. coli Flagellar Rotation and Chemotaxis Karen A. Morehouse,1 Ian G. Goodfellow,2 and R. Elizabeth Sockett1* Institute of Genetics, School of Biology, University of Nottingham Medical School, Queen’s Medical Center, Nottingham,1 and School of Animal and Microbial Sciences, University of Reading, Whiteknights, Reading,2 United Kingdom Received 24 September 2004/Accepted 17 November 2004

Flagellate bacteria such as Escherichia coli and Salmonella enterica serovar Typhimurium typically express 5 to 12 flagellar filaments over their cell surface that rotate in clockwise (CW) and counterclockwise directions. These bacteria modulate their swimming direction towards favorable environments by biasing the direction of flagellar rotation in response to various stimuli. In contrast, Rhodobacter sphaeroides expresses a single subpolar flagellum that rotates only CW and responds tactically by a series of biased stops and starts. Rotor protein FliG transiently links the MotAB stators to the rotor, to power rotation and also has an essential function in flagellar export. In this study, we sought to determine whether the FliG protein confers directionality on flagellar motors by testing the functional properties of R. sphaeroides FliG and a chimeric FliG protein, EcRsFliG (N-terminal and central domains of E. coli FliG fused to an R. sphaeroides FliG C terminus), in an E. coli FliG null background. The EcRsFliG chimera supported flagellar synthesis and bidirectional rotation; bacteria swam and tumbled in a manner qualitatively similar to that of the wild type and showed chemotaxis to amino acids. Thus, the FliG C terminus alone does not confer the unidirectional stop-start character of the R. sphaeroides flagellar motor, and its conformation continues to support tactic, switch-protein interactions in a bidirectional motor, despite its evolutionary history in a bacterium with a unidirectional motor. mediated by electrostatic forces between oppositely charged residues clustered on a charged ridge located at the FliG C terminus and the cytoplasmic loop of MotA (37). It might be expected that altered conformations of the FliG C terminus, comparing CW and CCW motors, would change the interactions between charged residues on FliG and MotA that might dictate the direction of flagellar rotation. To investigate this, we cloned R. sphaeroides fliG (RsfliG) and tested the effect on E. coli flagellar rotation and taxis by transcomplementation of an E. coli fliG null mutant with the whole RsfliG sequence and an E. coli-RsfliG chimera containing the C-terminal domain of R. sphaeroides FliG (RsFliG), in comparison to wild-type E. coli FliG (EcFliG). We found that although whole RsFliG could not complement an E. coli fliG null mutant, it was complemented and flagellar synthesis, motility, and limited chemotaxis were restored by the introduction of an E. coli-RsFliG (EcRsFliG) chimera. These data show for the first time that a flagellar rotor protein domain from an ␣-proteobacterium with unidirectional flagella has not become evolutionarily specialized and can support bidirectional flagellar rotation in a ␥-proteobacterium.

Bacteria swim and respond chemotactically to various stimuli by biasing the rotational behavior of their flagella when an internal phosphorelay network signals a change in chemotactic receptor occupancy on the cell surface. In Escherichia coli, flagellar rotation alternates between clockwise (CW) and counter-clockwise (CCW) directions, where CW rotation leads to cell tumbling and reorientation, and CCW rotation produces smooth swimming and thus forward movement (21). In contrast, Rhodobacter sphaeroides has a unidirectional flagellum that alternates between CW rotation and brief stops, where the bacterium is reoriented by Brownian motion and changes in flagellar filament morphology (2). The molecular mechanisms underlying flagellar torque generation are not fully understood, due mainly to a lack of structural data. For recent reviews, see references 4, 5, 22, and 26. The nonrotating stator component is composed of MotA and MotB proteins, which together form proton-conducting ion channels. MotAB channels encircle the MS ring and C ring at the base of the flagellum. A ring of FliG subunits, attached at their N termini to the MS ring, project into the cytoplasm from the C ring. As protons flow through MotAB, they are thought to transiently bind to and dissociate from a critical aspartate residue (38) located on MotB, inducing conformational changes (17) in the cytoplasmic domain of MotA that are believed to apply force to rotor protein subunits of FliG, thus turning the motor (17, 26). These stator-rotor interactions are

MATERIALS AND METHODS Strains, plasmids, and culture conditions. Bacterial strains and plasmids used in this work are described in Table 1. R. sphaeroides strains were grown at 29°C in Mu medium (10 g of tryptone/liter, 5 g of yeast extract/liter, and 10 g of NaCl/ liter) or photosynthetically in succinate medium (29) in the light at 25°C. E. coli strains were grown in Mu medium at 37°C for genetic manipulation and at 29°C for motility studies. E. coli antibiotic selection was carried out as previously described (25). R. sphaeroides antibiotic selection was as follows: nalidixic acid, 20 ␮g ml⫺1; tetracycline, 1 ␮g ml⫺1; and kanamycin, 50 ␮g ml⫺1. DNA isolation and manipulation. Plasmid DNA was prepared with a QIAGEN Miniprep kit according to the manufacturer’s instructions. R. sphaeroides

* Corresponding author. Mailing address: Institute of Genetics, School of Biology, University of Nottingham, Medical School, QMC, Nottingham NG7 2UH, United Kingdom. Phone: 44 115 919 4496. Fax: 44 115 970 9906. E-mail: [email protected]. 1695

1696

MOREHOUSE ET AL.

J. BACTERIOL. TABLE 1. Bacterial strains and plasmids used in this study

Strain or plasmid

Strains E. coli DH5␣ S17-1 RP437 DFB225 R. sphaeroides WS8N WS8FliGK

Plasmids pUC19 pRK415 pSUP202 pUC4K pHT53 pKAM101

p5.5 c71I pKAM500 pKAM501 pKAM502

Description

Reference or source

Cloning strain, F⬘ lacZM15 recA1 Mobilizing strain for conjugative transfer pro res recA; integrated plasmid RP4-Tc::Mu-Kn::Tn7 Wild type for motility and chemotaxis RP437 derivative, ⌬fliG; nonflagellate

34 8 J. S. Parkinson D. F. Blair (19)

Wild type for motility and chemotaxis; spontaneous Nalr mutant of WS8 WS8N fliG knockout due to insertion of Kn resistance cartridge, non flagellate; Knr

30 This study

Cloning vector; Ampr Broad-host-range mobilizable cloning vector; Tcr Mobilizable suicide vector for allelic exchange in R. sphaeroides; Tcr Cmr Ampr Source of 1.25 kb kanamycin cassette; Ampr Knr Vector expressing E. coli fliG; Ampr pHT53 derivative expressing E. coli-R. sphaeroides fliG chimera by inserting in frame, a 356-bp EcoRI fragment encoding the last 111 amino acids plus stop codon of R. sphaeroides FliG into unique EcoRI site within E. coli fliG; Ampr pUC19 clone containing 5.5kb EcoRI insert derived from WS8N cosmid clone 140, includes whole R. sphaeroides fliG plus ⬃1.5-kb upstream and ⬃0.4-kb downstream DNA; Ampr Cosmid carrying a 25-kb R. sphaeroides WS8 wild-type genomic DNA in pLA2917 (1); contains some of the fleR operon, including fliG (Fig. 1); Tcr 1.41-kb SacI-EcoRI PCR fragment containing R. sphaeroides fliG cloned into pUC19; Ampr 1.25-kb kanamycin cassette isolated from pUC4K, cloned into unique StuI site within fliG on pKAM500; Ampr, Knr 2.66-kb fliG plus Knr cassette blunt cloned into pSUP202; Tcr

35 15 28 32 D. F. Blair (19) This study

genomic DNA was isolated with a Wizard genomic DNA purification kit (Promega). Standard techniques were employed for cloning, transformation, preparation, and restriction analysis of plasmid DNA (25). Cloning and disruption of RsfliG. RsfliG was first identified by chromosome walking from fliI, which we had previously identified (12). RsfliG was subcloned (Fig. 1) as a 5.5-kb EcoRI fragment into pUC19 (to give p5.5) from a single 25-kb cosmid insert (from a wild-type R. sphaeroides WS8 cosmid gene library) that complemented an fliI transposon mutant (11). For inactivation of RsfliG, a 1.41kb SacI-EcoRI fragment was PCR amplified (forward primer, 5⬘-ATGATCGC GGCCGAGCTCGACACCGTGAA-3⬘ [SacI site underlined]; and reverse primer, 5⬘-AGCCGCAGGAATTCCTCGCGCGCGT-3⬘ [EcoRI site underlined]) from R. sphaeroides genomic DNA and cloned into pUC19 to give pKAM500. A KAN cassette isolated from pUC4K (32) was then cloned into the unique StuI site within RsfliG to generate pKAM501. The 2.66-kb fliGKn fragment was then excised from pKAM501, blunt cloned into the suicide vector pSUP202 (28) to give pKAM502, and then transformed into E. coli S17-1 for diparental conjugation with R. sphaeroides wild-type strain WS8N. Chromosomal DNA was prepared from R. sphaeroides exconjugants resistant to nalidixic acid and kanamycin but sensitive to tetracycline and used as template DNA for a PCR screen to identify that allelic exchange of fliG with fliGKn had occurred by double homologous recombination. This was then confirmed by Southern blotting and sequence analyses. Construction of EcRsFliG chimera. Primers RsFliGC-F (5⬘-CGGAATTCGA CAAGGACCTGATGCAGGCGATCCAGG-3⬘) and RsFliGC-R (5⬘-CGGAA TTCTTACACCATCTGCTCGCCGCCGCGGCCC-3⬘) were designed to amplify a 356-bp fragment from plasmid p5.5 with terminal EcoRI sites (underlined) that encoded the last 111 amino acids plus the stop codon of RsFliG-C. E. coli fliG present on plasmid pHT53 (19) has a unique EcoRI site located between amino acids 221 to 222. This was the insertion point of the 356-bp RsFliG-C EcoRI fragment, giving rise to pKAM101, expressing the EcRsFliG chimeric protein beginning with the N terminus of EcFliG and running to the C terminus of RsFliG where it terminated at the RsfliG stop codon. The pKAM101 construct was verified by sequencing, transformed into the E. coli fliG null mutant, DFB225 (19), and maintained with ampicillin selection. Electron microscopy. Visualization of E. coli and R. sphaeroides flagellar filaments was carried out as follows. A total of 20 ␮l of culture (at an optical density at 600 nm [OD600] of 0.75) was spotted onto a carbon-Formvar-coated grid (Agar Scientific) and left for 20 s (E. coli) or several minutes (R. sphaeroides). The grid was washed with excess sterile water. Flagellar filaments were stained

11 11 This study This study This study

with 20 ␮l of 0.5% (wt/vol) phosphotungstic acid stain (E. coli) or 1% (wt/vol) phosphotungstic acid (R. sphaeroides) for 15 s. The grids were visualized with a JEOL JEM-100S electron microscope. Analysis of free-swimming E. coli cells with Hobson BacTracker. A BacTracker (Hobson Tracking Systems, Sheffield, United Kingdom) was used to determine the run speeds and stopping and tumbling characteristics of motile E. coli cultures. A total of 0.5 ml of overnight stationary-phase cultures of E. coli strains was used to inoculate 50 ml of Mu broth supplemented with ampicillin as

FIG. 1. Comparison of gene organization surrounding fliG in R. sphaeroides and E. coli. Single capital letters represent the respective fli gene; the scale is in 1-kb increments. It is not known exactly where the R. sphaeroides fleR operon ends, but fliMN forms a two-gene operon (curved arrows represent predicted transcriptional start sites) (9). In E. coli, fliE is separate with its own divergent promoter (24), and the fliF operon contains fliG. The hatched box and arrow represent the cosmid arm and the extending 25-kb genomic sequence of R. sphaeroides cosmid clone 711 (11), respectively. Cosmid clone 711 contains some of the fleR operon, including fliG, but only partially complemented (20% of the population) the disrupted RsfliG strain WS8FliGK.

VOL. 187, 2005

CHIMERIC FliG ROTOR PROTEIN

required; the cultures were incubated with shaking at 200 rpm at 29°C until an OD600 value of 0.750 was reached. This was the OD600 value at which ⬎80% motility of all strains was observed. A 9.6-␮l culture was dropped onto a microscope slide (Merck) and covered with a 22- by 22-mm coverslip (Merck), pressed down such that the culture just reached the edges of the coverslip and giving a chamber depth of 20 ␮m (Jeff Hobson, personal communication). Cultures were observed with a phase-contrast objective (magnification, ⫻20) on a Nikon Labophot 2A microscope and were tracked immediately for 100 tracks. This was repeated at least three times using a fresh slide each time. The experiment was repeated on more than two different occasions for each set of strains. Real-time computer tracking was carried out using the Hobson BacTracker 50-Hz system set to the following image settings: search radius, 2.19 ␮m; trail, 25; predict, off; calibrate, ⫻125.0; Vid Pal, 50; aspect ratio, 1.46; refresh time, 5 s; thresholds, ⫹38/⫺107; filter weightings, 1:2, 2:1, 3:1, and 4:1; frames, 10; pixels, 1; percent linearity, 38; percent hysteresis, 80; ␮m/s, 2.2; immotile process, off; brightness, 139; contrast, 101; and video input, 2. Analysis settings were as follows: analysis, tracks; on tracks, 100; minimum track time, 1.0 s; minimum event time, 0.2 s; maximum track time, 10.0 s; and limits, off. Raw tracking data were analyzed as follows. The mean run speed in micrometers per second and the mean stop time in seconds were calculated for each batch of 100 tracks recorded. These means were sorted into intervals across the range of data, with 1,100 tracks analyzed and sorted in this way for each strain. The data were recorded as histograms. Motility and chemotaxis assays in soft agar plates. For E. coli strains, single colonies were picked from fresh 2% agar plates with sterile toothpicks and stab inoculated into 0.35% (Bacto Agar; Difco) tryptone soft agar plates (3). Plates were incubated for 10 h at 29°C unless otherwise indicated. Defined E. coli motility plates for chemotaxis investigation were made from an M9 salts base, supplemented with 2% glycerol (carbon source) and 2 ␮g ml⫺1 of each of the following essential amino acids: L-leucine, L-histidine, L-methionine, and L-threonine (7). In addition, single E. coli chemoattractants (including 1 mM aspartic acid and 1 mM serine) were added. The incubation time for E. coli strains on defined motility plates at 29°C was 48 h, unless otherwise stated. The motility of R. sphaeroides strains was routinely analyzed in comparison to wild-type strains by stabbing single fresh colonies into TYS swarm plates as previously described (27).

RESULTS AND DISCUSSION FliG proteins show greatest conservation at the C terminus. RsfliG was cloned by chromosome walking from fliI, which we had previously identified (12). It lies within the fleR operon under the control of a sigma-54 promoter with fleR fliEFG all being transcribed in the same direction (GenBank no. X9869492) (Fig. 1). It is not known exactly where the R. sphaeroides fleR operon ends, but fliMN forms a separate two-gene operon (9). In E. coli, there is no fleR homolog upstream of fliE, and E. coli fliE is encoded on the opposite DNA strand with its own divergent promoter (24); it is the fliF operon that contains fliG (Fig. 1). We found that although the WS8FliGK strain was nonflagellate (data not shown), as is the case for an E. coli fliG null mutant, it could only be transcomplemented when fliG was expressed from cosmid clone c711 (11) containing a whole fleR-fliH fragment (Fig. 1). This may be due to the different operon organization or probable polar effects of the KAN cartridge on the expression of downstream fli genes. Alignment of FliG proteins (Fig. 2) shows that RsFliG (GenBank no. X98691) has similarities, especially at the C terminus, to E. coli, Vibrio alginolyticus (Va), and Thermotoga maritima (Tm) FliG protein sequences, the last being the source of FliG used to determine the structure of the FliG middle and C termini (6, 20). Charged amino acids found to be essential for motor function in E. coli (18) were also conserved in RsFliG. Interestingly, glutamate residue E320 in EcFliG (Fig. 2) was not conserved in RsFliG, where the corresponding residue is a threonine. The secondary mutation, E320K, in EcFliG suppressed a paralyzing mutation, P159I, located in the periplasmic wall-binding region of MotB, restoring motility (10). It was

1697

suggested that the initial P159I paralyzing mutation altered the register between FliG and MotAB, but the suppressing FliG E320K mutation changed the conformation of FliG, thus restoring clearance between the stator and rotor (10). Therefore, unidirectional CW rotation in R. sphaeroides flagella may be due to an altered conformation(s) of FliG, changing interactions with MotA and inhibiting CCW rotation. We tested this proposal by investigating the functionality of RsFliG (from a unidirectional motor) in an E. coli (bidirectional flagellar motor) background. A chimeric E. coli-R. sphaeroides fliG complements E. coli fliG deletion mutant. Whole RsfliG expressed from plasmid p5.5 failed to complement the E. coli nonflagellate fliG deletion strain DFB225, and it remained nonflagellate (data not shown), whereas the wild-type EcfliG restored motility (Fig.5Aii). This suggested that whole RsFliG may not cofunction with the E. coli flagellar export proteins to synthesize flagella. A similar result was also reported by Gosink and Ha¨se for Vibrio cholerae FliG in E. coli (13). It is well established that the C terminus of FliF and the N terminus of FliG interact closely with one another, as FliFFliG fusion mutants are still able to function in both flagellar assembly and rotation (31). This may explain why the fulllength FliG could not be functionally exchanged between E. coli and R. sphaeroides. There is the least sequence conservation between E. coli and R. sphaeroides at the FliG N terminus (Fig. 2), and a similar observation can be made when the C termini of FliF proteins are compared (data not shown). Mutagenesis studies of the switch protein FliG from E. coli and Salmonella enterica serovar Typhimurium have shown several residues to be important for flagellar assembly, in addition to rotation and switching (10, 14, 16, 18, 19, 23). As these studies revealed that the export properties of FliG reside in its N-terminal and central domains (16); we designed a chimeric FliG for testing in E. coli, named EcRsFliG, in which the first two thirds of the protein were EcFliG and the C-terminal third (including functionally important charged residues) was from RsFliG. The junction site, located between EcFliG residues 221 to 222 (Fig. 2), was chosen to ensure that sufficient EcFliG sequence was present to function in flagellar export when the EcRsFliG chimera was expressed in trans from pKAM101 in a fliG null background. Expressing pKAM101 in DFB225 not only restored motility but electron micrographs showed normal flagellation patterns (Fig. 3B) in comparison to DFB225 complemented with wildtype E. coli fliG expressed from plasmid pHT53 (Fig. 3C). We also constructed and tested, by chromosomal gene replacement, a reverse R. sphaeroides-E. coli fliG chimera (the junction site of the chimera is shown in Fig. 2), but this did not support flagellar export or assembly in R. sphaeroides (data not shown). As inserting the more AT-rich EcfliG-C sequence into the GC-rich R. sphaeroides fleR operon may have caused polar expression effects on downstream fli genes, required for flagellar export, we do not consider this further here. E. coli expressing pKAM101 has run-tumble but not stopstart swimming behavior. As DFB225(pKAM101) expressed a chimeric fliG where the C-terminal third originated from a stop-start motor, we wanted to investigate the swimming style of the mutant in comparison to complemented wild-type controls. Visually, DFB225(pKAM101) showed run-tumble motil-

1698

MOREHOUSE ET AL.

J. BACTERIOL.

FIG. 2. Alignment of FliG sequences from V. alginolyticus (Va), R. sphaeroides (Rs), E. coli (Ec), and T. maritima (Tm), the latter being the source used for crystallization of the FliG middle and C termini (6, 20). Black arrows represent positions of the functional E. coli-R. sphaeroides (EcRs) and nonfunctional R. sphaeroides-E. coli (RsEc) chimeric FliG proteins constructed in this study. Grey arrows illustrate junction sites of other functional FliG chimeras constructed elsewhere: E. coli-T. maritima (EcTm) FliG (20), V. alginolyticus-E. coli (Va-Ec) and E. coliV. alginolyticus (EcVa) (36), and V. cholerae-E. coli (VcEc) and E. coli-V. cholerae (EcVc) (13). Residues of interest within the FliG C terminus are highlighted as follows: stars, residues of functional importance in E. coli (18, 37) but interestingly not in V. alginolyticus (36); diamonds, residues where changes showed weakened FliG-FliM binding in a yeast two-hybrid assay (23); triangle, single residue isolated as a suppressor of mutation P159I in E. coli stator protein MotB (10). Finally, underlined residues are predicted to be important for FliG-FliM interactions (6, 23).

ity similar to that of the E. coli wild-type strain RP437, as observed by phase-contrast microscopy. A Hobson BacTracker was used to quantify the free-swimming motility of mid-logphase E. coli DFB225 cultures carrying either wild-type E. coli fliG on pHT53 or pKAM101, each in the presence of the

wild-type E. coli RP437 strain. The mean run speed (in micrometers per second) and the mean stop time (in seconds) were calculated from 11 batches of 100 tracks each. These mean values were then sorted into intervals and displayed as histograms to describe the observed free-swimming motility

FIG. 3. Electron micrographs (magnification, ⫻7,000) of nonflagellate DFB225 (A), DFB225 complemented with pHT53 expressing wild-type E. coli fliG (B), and DFB225 complemented with pKAM101 expressing the EcRsFliG chimera (C). Cells were negatively stained with 0.5% phosphotungstic acid.

VOL. 187, 2005

CHIMERIC FliG ROTOR PROTEIN

1699

FIG. 4. Swimming behavior of the EcRsFliG chimera in comparison to controls. Mean run speed (in micrometers per second) (A) and mean stop time (in seconds) (B) analyses of free-swimming DFB225 expressing the EcRsFliG chimera from pKAM101 (grey bars) alongside control strains RP437 wild-type (black bars) and DFB225(pHT53) (white bars) are shown by using a Hobson BacTracker. Data were sorted into confidence intervals from batches of 100 individual tracks apiece. Each strain was analyzed at least three times from two separately grown cultures.

(Fig. 4). E. coli wild-type motility does not formally include stops; the cells alternate between runs and tumbles. However, the BacTracker does record some of these tumble events as stops, even for wild-type E. coli. Thus, we were looking for any significant increases in recorded stop time as evidence of a stop-biased motor caused by the EcRsFliG chimera. The mean run speed distributions were 15 to 15.99 ␮m s⫺1 for wild-type E. coli RP437, 15 to 15.99 ␮m s⫺1 for DFB225 (pHT53), and 13 to 13.99 ␮m s⫺1 for DFB225(pKAM101) (Fig. 4A). There was a slightly higher distribution of cells with lower mean run speeds for DFB225(pKAM101) populations (Fig. 4A) than for RP437 and DFB225(pHT53). The slightly reduced swimming speeds may have been due to suboptimal protein-protein interactions within the C-ring switch complex or between FliG and MotA; the latter is an essential element of the torque generation process. Furthermore, DFB225 (pKAM101) showed a similar mean stop time distribution in comparison to the E. coli control strains (Fig. 4B); thus, ex-

pression of the RsFliG C terminus (from a stop-start motor) in the E. coli bidirectional motor does not produce a motor with a greater stopping frequency. This suggests that the unidirectional properties of the R. sphaeroides flagellar motor do not reside entirely in its FliG C terminus. E. coli expressing pEcRsFliG is weakly chemotactic in soft agar. As FliG is a component of the flagellar C-ring switch complex required for chemotaxis, we assayed the chemotactic behavior of DFB225(pKAM101) in comparison to parental strains in soft agar plates (3). DFB225 produced point colonies, due to its nonflagellate phenotype, as a result of the deletion of fliG (Fig. 5Ai). Both wild-type RP437 and DFB225(pHT53) showed the typical large, double-ringed chemotactic swarm (Fig. 5Ai and Aii) and a smaller chemotactic ring were observed for DFB225(pKAM101) (Fig. 5Aiii). Expressing pKAM101 in wild-type RP437 did not cause dominance over wild-type fliG, as the swarm size was not greatly reduced in comparison to the RP437(pUC19) control (Fig. 5Aiv).

1700

MOREHOUSE ET AL.

J. BACTERIOL.

FIG. 5. Comparison of chemotactic behavior between E. coli wild-type RP437 and recombinant fliG strains in 0.35% Bacto Agar tryptone (A) and M9 minimal base (B) swarm plates, the latter supplemented with 1 mM aspartate and 1 mM serine chemoattractants (v) or no chemoattractants (vi). Plates were incubated at 29°C for 10 h (A) and 48 h (B), respectively; ampicillin was added to tryptone swarm media for plasmid maintenance where necessary.

As the double-ringed appearance of chemotactic swarms on tryptone base soft agar is thought to be due to a combination of aerotaxis and aspartate and serine chemotaxis (33), we also tested the chemotactic response of DFB225(pKAM101) to aspartate and serine in a M9 minimal media background and found that it formed small tactic swarms to both of these attractants (Fig. 5B). The reduced swimming speed of DFB225(pKAM101) observed in the Hobson BacTracker analysis (Fig. 4) may account for all or part of the reduced tactic swarm diameters seen. Timing of the onset of motility did not seem to be a factor, as microscopic examination of RP437, DFB225 (pKAM101), and DFB225 (pHT53) all showed onset of motility (the first few motile cells observed in the culture) at the same time and at the same OD600 value of 0.15. At an OD600 value of 0.75, all cultures showed 90% motility. Thus, although liquid cultures cannot totally reflect swimming within a swarm plate, there was no evidence for late onset of motility in the DFB225(pKAM101) strain, which may have accounted for the smaller swarm diameter seen on plates.

However, tactic-switching effects due to an altered FliG sequence(s) in the chimera may be important. FliM is predicted to bind FliG at two sites (6), and CW-CCW switching during chemotaxis in E. coli might be modulated by changes in the conformation of FliG that are dictated by switch protein FliM, in response to signals from the chemosensory pathway. These two predicted sites of FliG-FliM interaction on FliG are the conserved EHPQ motif (E. coli residues 125 to 128) and several hydrophobic residues from E. coli residues 196 to 229 (Fig. 2) (6, 23). In the EcRsFliG chimera sequence, the EHPQ motif was in the EcFliG domain, but some of the residues that constitute the hydrophobic patch resided in the RsFliG domain (Fig. 2). As the FliG-C therefore originated from a different flagellar motor, FliG-FliM interactions may not have been optimal and may have led to the observed reduced swarm diameters. This may reinforce the importance of the correct spatial orientation for optimal, coordinated interactions within the bacterial flagellar motor.

VOL. 187, 2005

CHIMERIC FliG ROTOR PROTEIN

A comparison with other FliG chimeras. Other FliG chimeras have been previously constructed between FliGs of thermophiles and mesophiles or between FliGs of proton-motive and sodium-motive flagella (13, 36). An E. coli-T. maritima FliG chimera containing the first 1 to 240 residues of EcFliG fused to residues 243 to 335 of T. maritima FliG (designated TmFliG) was motile in an E. coli host (20). This junction site is located 19 amino acids downstream of the EcRsFliG chimera construct made in our investigation. Two other studies involved chimeric FliG proteins (Fig. 3) between E. coli and Vibrio sp. (designated EcVibrioFliG) (13, 36). Both the EcVibrioFliG chimeras complemented the EcFliG null strain, DFB225, as did the inverse VibrioEcFliG fusion proteins using the same junction site position. The junction sites of both these E. coli and Vibrio chimeric FliG proteins are located 19 amino acids downstream from the EcRsFliG junction site used our study. Thus, our work extends the region of the heterologous FliG sequence that is known to functionally substitute for EcFliG in DFB225. This is the only example of FliG sequence exchanges being made between uni- and bidirectional motors, and it illustrates a common mechanism for FliG C-terminal function in bacterial flagellar motors with different rotational properties. ACKNOWLEDGMENTS This work was funded in part by a BBSRC Ph.D. studentship to R.E.S. for IGPG, an NERC Ph.D. studentship to R.E.S. for K.A.M., and BBSRC grant 42/P18196 to R.E.S. We thank David Blair for E. coli strain DFB225 and plasmid pHT53, Sandy Parkinson for RP437, John Taylor for helpful discussions, and Marilyn Whitworth for technical assistance. REFERENCES 1. Allen, L. N., and R. S. Hanson. 1985. Construction of broad-host-range cosmid cloning vectors: identification of genes necessary for growth of Methylobacterium organophilum on methanol. J. Bacteriol. 161:955–962. 2. Armitage, J. P., and R. M. Macnab. 1987. Unidirectional intermittent rotation of the flagellum of Rhodobacter sphaeroides. J. Bacteriol. 169:514–518. 3. Armstrong, J. B., J. Adler, and M. M. Dahl. 1967. Nonchemotactic mutants of Escherichia coli. J. Bacteriol. 93:390–398. 4. Berg, H. C. 2003. The rotary motor of bacterial flagella. Annu. Rev. Biochem. 72:19–54. 5. Blair, D. F. 2003. Flagellar movement driven by proton translocation. FEBS Lett. 545:86–95. 6. Brown, P. N., C. P. Hill, and D. F. Blair. 2002. Crystal structure of the middle and C-terminal domains of the flagellar rotor protein FliG. EMBO J. 21: 3225–3234. 7. Budrene, E. O., and H. C. Berg. 1991. Complex patterns formed by motile cells of Escherichia coli. Nature 349:630–633. 8. Davis, J., T. J. Donohue, and S. Kaplan. 1988. Construction, characterization and complementation of a Puf⫺ mutant of Rhodobacter sphaeroides. J. Bacteriol. 170:320–329. 9. Edge, M. E. 2000. Analysis of flagellar switch proteins in Rhodobacter sphaeroides. Ph.D. thesis. University of Nottingham, Nottingham, United Kingdom. 10. Garza, A. G., R. Biran, J. Wohlschlegel, and M. D. Manson. 1996. Mutations in motB suppressible by changes in stator or rotor components of the bacterial flagellar motor. J. Mol. Biol. 25:270–285. 11. Goodfellow, I. G. P. 1996. The unidirectional flagellum of R. sphaeroides: cloning and analysis of genes encoding regulatory, structural and motor components. D.Phil. thesis. University of Nottingham, Nottingham, United Kingdom. 12. Goodfellow, I. G. P., C. E. Pollitt, and R. E. Sockett. 1996. Cloning of the fliI gene from Rhodobacter sphaeroides WS8 by analysis of a transposon mutant with impaired motility. FEMS Microbiol. Lett. 142:111–116. 13. Gosink, K. K., and C. C. Ha ¨se. 2000. Requirements for conversion of the Na⫹-driven flagellar motor of Vibrio cholerae to the H⫹-driven motor of Escherichia coli. J. Bacteriol. 182:4234–4240.

1701

14. Irikura, V. M., M. Kihara, S. Yamaguchi, H. Sockett, and R. M. Macnab. 1993. Salmonella typhimurium fliG and fliN mutations causing defects in assembly, rotation, and switching of the flagellar motor. J. Bacteriol. 175: 802–810. 15. Keen, N. T., S. Tamaki, D. Kobayashi, and D. Trollinger. 1988. Improved broad-host-range plasmids for DNA cloning in gram-negative bacteria. Gene 70:191–197. 16. Kihara, M., G. U. Miller, and R. M. Macnab. 2000. Deletion analysis of the flagellar switch protein FliG of Salmonella typhimurium. J. Bacteriol. 182: 3022–3028. 17. Kojima, S., and D. F. Blair. 2001. Conformational change in the stator of the bacteria flagellar motor. Biochemistry 40:13041–13050. 18. Lloyd, S. A., and D. F. Blair. 1997. Charged residues of the rotor protein FliG essential for torque generation in the flagellar motor of Escherichia coli. J. Mol. Biol. 266:733–744. 19. Lloyd, S. A., H. Tang, X. Wang, S. Billings, and D. F. Blair. 1996. Torque generation in the flagellar motor of Escherichia coli: evidence of a direct role for FliG but not FliM and FliN. J. Bacteriol. 178:223–231. 20. Lloyd, S. A., F. G. Whitby, D. F. Blair, and C. P. Hill. 1999. Structure of the C-terminal domain of FliG, a component of the rotor in the bacterial flagellar motor. Nature 400:472–475. 21. Macnab, R. M. 1996. Flagella and motility, p. 123–146. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed. ASM Press, Washington, D.C. 22. Macnab, R. M. 2003. How bacteria assemble flagella. Annu. Rev. Microbiol. 57:77–100. 23. Marykwas, D. L., S. A. Schmidt, and H. C. Berg. 1996. Interacting components of the flagellar motor of Escherichia coli revealed by the two-hybrid system in yeast. J. Mol. Biol. 256:564–576. 24. Muller, V., C. J. Jones, I. Kawagishi, S. I. Aizawa, and R. M. Macnab. 1992. Characterization of the fliE genes of Escherichia coli and Salmonella typhimurium and identification of the FliE protein as a component of the flagellar hook-basal body complex. J. Bacteriol. 174:2298–2304. 25. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 26. Schmitt, R. 2003. Helix rotation model of the flagellar rotary motor. Biophys. J. 85:843–852. 27. Shah, D. S. H., T. Perehinec, S. M. Stevens, S. I. Aizawa, and R. E. Sockett. 2000. The flagellar filament of Rhodobacter sphaeroides: pH-induced polymorphic transitions analysis of the fliC gene. J. Bacteriol. 182:5218–5224. 28. Simon, R., U. Priefer, and A. Puhler. 1983. A broad host range mobilization system for in vivo genetic engineering-transposon mutagenesis in gram negative bacteria. Bio/Technology 1:784–791. 29. Sistrom, W. R. 1962. The kinetics of the synthesis of photopigments in Rhodopseudomonas sphaeroides. J. Gen. Microbiol. 28:607–616. 30. Sockett, R. E., J. C. A. Foster, and J. P. Armitage. 1990. Molecular biology of the Rhodobacter sphaeroides flagellum. FEMS Symp. 53:473–479. 31. Thomas, D. R., D. G. Morgan, and D. J. DeRosier. 2001. Structures of bacterial flagellar motors from two FliF-FliG gene fusion mutants. J. Bacteriol. 183:6404–6412. 32. Vieira, J., and J. Messing. 1982. The pUC plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19:259–268. 33. Wolff, C., and J. S. Parkinson. 1988. Aspartate taxis mutants of the Escherichia coli Tar chemoreceptor. J. Bacteriol. 170:4509–4515. 34. Woodcock, D. M., P. J. Crowther, J. Doherty, S. Jefferson, E. DeCruz, M. Noyer-Weidner, S. S. Smith, M. Z. Michael, and M. W. Graham. 1989. Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res. 17:3469–3478. 35. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33:103–119. 36. Yorimitsu, T., A. Mimaki, T. Yakushi, and M. Homma. 2003. The conserved charged residues of the C-terminal region of FliG, a rotor component of the Na⫹-driven flagellar motor. J. Mol. Biol. 334:567–583. 37. Zhou, J., S. A. Lloyd, and D. F. Blair. 1998. Electrostatic interactions between the rotor and stator in the bacterial flagellar motor. Proc. Natl. Acad. Sci. USA 95:6436–6441. 38. Zhou, J., L. L. Sharp, H. L. Tang, S. A. Lloyd, S. Billings, T. F. Braun, and D. F. Blair. 1998. Function of protonatable residues in the flagellar motor of Escherichia coli: a critical role for Asp32 of MotB. J. Bacteriol. 180:2729– 2735.