Sinorhizobial chemotaxis: a departure from the enterobacterial paradigma

Rüdiger Schmitt1

Institut für Biochemie, Genetik und Mikrobiologie, Universität Regensburg, D-93040 Regensburg, Germany1

Tel: +49 941 9433162. Fax: +49 941 9433163. e-mail: rudy.schmitt{at}biologie.uni-regensburg.de

Keywords: {alpha}-proteobacteria, sensory signal transduction, multiple chemotaxis genes, retro-phosphorelay, chemokinesis

a To Professor Wolfram Heumann on his 87th birthday.


   Overview
TOP
Overview
Signalling pathways
Swimming patterns and flagellar...
Conclusions
REFERENCES
 
Motile bacteria display a surprisingly sophisticated sensory behaviour. By monitoring changes in the chemical composition of their environment, they migrate towards food sources (attractants) and flee from noxious chemicals (repellents). This behaviour, named chemotaxis, involves information processing of signals from receptors that sense the environment, and signal transduction to the flagellar motor to modulate its rotation accordingly. Since the pioneering work of Julius Adler almost 40 years ago (Adler, 1965 ), genetic, biochemical and biophysical studies of chemotaxis have been focussed mostly on enterobacteria, essentially on Escherichia coli and Salmonella, thus contributing greatly to the goal of ‘understanding a simple behaviour at the molecular level’ (Parkinson, 1987 ). Today, the enterobacterial sensory pathway and all central components of the signal transduction chain as well as interactions between these proteins have been identified (for recent reviews see Bren & Eisenbach, 2000 ; Falke et al., 1997 ; Grebe & Stock, 1998 ; Macnab, 1996 ). However, many molecular details of receptor function, signal amplification and signal processing, of flagellar motor mechanics and regulation are still not understood. The more recent interest in non-enteric and even archaeal chemosensory systems has expanded our view and revealed fascinating variations on a general theme (for reviews see Armitage & Schmitt, 1997 ; Manson et al., 1998 ; Marwan & Oesterhelt, 2000 ). Concurrences observed between bacteria and archaea suggest an early origin of the prokaryotic chemosensory system – as long as 3 billion years ago. Differences between present-day systems point to a diverging evolution and the need for adaptation to different habitats. For the investigator, this departure from the established scheme poses an interesting challenge, as it may conceal new answers to hitherto unsolved questions.

The focus of this article will be on features that distinguish the chemosensory systems of E. coli and Sinorhizobium meliloti. The soil bacterium S. meliloti is a member of the {alpha}-subclass of Proteobacteria (Olsen et al., 1994 ), which contains such diverse species as Acetobacter, Agrobacterium, Caulobacter, Methylomonas and Rhodobacter. These differ from enterobacteria ({gamma}-subclass) by their wide range of natural habitats and by their great metabolic flexibility, reflecting a separate evolution of more than 500 million years (Ochman & Wilson, 1987 ). By colonizing the lower animal gut, E. coli has adopted a specialized lifestyle, whereas the nitrogen-fixing S. meliloti leads a versatile life, either freely in the soil, or in the rhizosphere of host plants, or ultimately in root nodules, to convert atmospheric nitrogen to biologically usable ammonium. Differences between the specialized enterobacterial and the versatile rhizobial lifestyles are reflected in their recently sequenced genomes: E. coli K-12 contains a single circular chromosome with just 4300 protein-coding genes (Blattner et al., 1997 ), while S. meliloti features three circular replicons – the actual chromosome and two megaplasmids, pSymA and pSymB – with over 6200 protein-coding genes (Galibert et al., 2001 ). A large and variable genome is considered essential for growth and survival under the complex environmental conditions prevailing in the soil, the rhizosphere and in plant roots. As for many years the ‘coli-centric’ view has considered the presence of a single circular chromosome as universally valid, studies of S. meliloti and other {alpha}-proteobacteria have revealed an abundance of circular and even linear chromosomes that was previously unseen (Jumas-Bilak et al., 1998 ). Similarly, analyses of the chemosensory system of S. meliloti and its relatives (Armitage & Schmitt, 1997 ) have revealed departures from the enterobacterial paradigm that expand and newly define previous concepts. This will be illustrated in this review by two novel features that distinguish the S. meliloti chemosensory system from the established enterobacterial scheme.


   Signalling pathways
TOP
Overview
Signalling pathways
Swimming patterns and flagellar...
Conclusions
REFERENCES
 
Chemosensory signal transduction in bacteria is mediated by the ubiquitous histidine kinase pathway consisting of four elements (Fig. 1): (i) sensors or receptors (methyl-accepting chemotaxis proteins, MCPs), often containing membrane-spanning domains; (ii) a central ATP-dependent histidine autokinase (CheA); (iii) one or several response regulators (CheYs) activated by phosphotransfer from a conserved histidyl residue in CheA to a conserved aspartate in CheY; (iv) the flagellar motor (Mot) as the effector of the transmitted signal. The frequency by which a swimming cell changes direction results from switching or slowing motor rotation and is controlled by environmental stimuli signalling through the phosphorelay system. MCPs control the activity of the cytoplasmic CheA, which in turn controls the small CheYs. Phosphorylated CheY (CheY-P) can bind to the flagellar motor and cause a change in rotation (see next section). It is the mechanism of rapid CheY-P dephosphorylation and the mode of shifting flagellar rotation that distinguish the chemosensory pathways of E. coli and S. meliloti, the organisms used here as well-studied examples of enterobacteria and {alpha}-proteobacteria, respectively.



View larger version (21K):
[in this window]
[in a new window]
 
Fig. 1. Comparative schemes of sensory signal transduction to the flagellar motor in Escherichia coli (a) and Sinorhizobium meliloti (b). General components of the histidine autokinase pathway are given at the top. The corresponding receptor (MCP), chemotaxis (Che) and motor (Mot) proteins of the chemosensory transduction chain are marked within the diagrams. Histidyl (H) and aspartyl (D) phosphorylation (P), proton flow (H+) through MotA–MotB channels, and the clockwise (cw) or counterclockwise (ccw) sense of flagellar rotation are shown. Matching colours identify analogous (but differently reacting) response regulatory components in (a) and (b). Departures from the enterobacterial scheme including the motor proteins MotC and MotD are outlined in red.

 
How are tactic signals transmitted to the motor? Any decrease of attractant (or increase of repellent) elicits ATP-dependent autophosphorylation of CheA, which in turn rapidly transfers phosphate groups to CheY. Conversely, an increasing concentration of attractant (or a decrease of repellent) signals a stop to CheA autophosphorylation and thus blocks phosphotransfer to the response regulator. At this point, E. coli and S. meliloti apply different strategies to accelerate dephosphorylation (deactivation) of surplus CheY-P to terminate further signals to the motor. Signal termination by spontaneous dephosphorylation of activated CheY-P is too slow ({tau}/2 >=10 s) for a fitting response to negative stimuli (~50 ms). In E. coli (Fig. 1a), it is accelerated by a protein (CheZ) that functionally is a phosphatase: CheZ, by binding to free CheY-P, effectively accelerates its dephosphorylation (Bren et al., 1996 ; Hess et al., 1988 ). S. meliloti (Fig. 1b) takes a separate road: it features two different response regulators, CheY1 and CheY2, but no CheZ phosphatase. CheY1 and CheY2 are paralogues exhibiting 57% sequence similarity, both possessing the essential, highly conserved residues of response regulators, in particular the phosphate-accepting aspartyl-57 (Greck et al., 1995 ). In this pathway, CheY2-P functions as the ‘master control’ of flagellar rotation, while CheY1 acts as a ‘modulator’ (Sourjik & Schmitt, 1996 ). This is the hallmark of a different mechanism for rapidly dephosphorylating CheY2-P (the orthologue of the E. coli CheY-P). The phosphate from surplus CheY2-P (but not from CheY1-P) is shuttled back to CheA, which in turn phosphorylates free CheY1 (Sourjik & Schmitt, 1998 ). Retro-phosphorylation via CheA thus accelerates the deactivation of CheY2-P (Fig. 1b). Since intracellular concentrations of CheY1 are roughly in 10-fold excess over CheA, the former acts like a sponge (or ‘sink’) by absorbing the phosphoryl groups from CheA-P. However, as phosphorylation of CheY1 by CheA-P requires in the order of 50–100 ms, spontaneous auto-dephosphorylation of CheY1-P ({tau}/2 ~10 s; Sourjik & Schmitt, 1998 ) may be too slow to provide sufficient free CheY1. It is not clear yet whether S. meliloti has a device for accelerating the dephosphorylation of CheY1-P, in analogy to the enterobacterial CheZ. A small candidate protein with features of a ‘dephosphorylation stimulator’ is presently being studied in our laboratory.

Retro-phosphorylation in the sensory transduction chain as a mechanism for efficient dephosphorylation of the response regulator CheY2-P, and hence for rapid adaptation to new stimuli, was initially considered a peculiarity of S. meliloti (Sourjik & Schmitt, 1998 ). However, an inspection of well-analysed {alpha}-proteobacteria and of some recently published genome sequences of other non-enterics revealed the presence of CheY1 and CheY2 paralogues concurrent with the absence of CheZ in many of these bacterial species. We take this as strong suggestive evidence for functional retro-phosphorylation pathways operating in {alpha}-proteobacteria, such as Agrobacterium, Caulobacter, Rhodobacter and the rhizobia (Armitage & Schmitt, 1997 ), as well as in Bacillus subtilis (Rosario et al., 1994 ), Borrelia burgdorferi (Fraser et al., 1997 ), Campylobacter jejuni (Parkhill et al., 2000 ), Helicobacter pylori (Tomb et al., 1997 ), in the cyanobacterium Synechocystis (Kaneko et al., 1996 ) and even in the hyperthermophilic Thermotoga maritima (Nelson et al., 1999 ) representing the deepest known branch of the bacterial phylogenetic tree (Stetter, 1995 ). This suggests that the retro-phosphorelay may have evolved as an original mechanism for the deactivation of sensory response regulators and that the enterobacterial CheY–CheZ pathway may have been developed more recently as an adaptation to ‘modern’ intestinal environments.

It should be noted that whole-genome sequences of various {alpha}-proteobacteria and of other non-enterics revealed an additional complexity in containing two (or more) operons with multiple copies of the chemosensory (che) genes. Is there a biological concept behind these duplicate pathways? In the photosynthetic Rhodobacter sphaeroides, one of two che operons is dominant directing all known photo- and chemotactic responses, whereas mutants in the second che operon have virtually no effect on bacterial behaviour under laboratory conditions (Hamblin et al., 1997 ). Similarly, in S. meliloti, the chromosomal che operon dominantly controls the response to all tested attractants (Greck et al., 1995 ; Sourjik & Schmitt, 1996 ). A second set of che genes revealed by the complete S. meliloti genome sequence (Galibert et al., 2001 ) is located on the pSymA replicon and linked to genes that control type IV pilus synthesis. Together, the taxis- and pilus-coding genes may have a role in gliding (or twitching) motility (Kearns et al., 2001 ) appropriate for colonizing the host plant. While it is conceivable that multiple che operons in one cell may interact in a synergistic fashion, the two reported cases rather suggest that these operons specify distinctive sensory pathways operating under diverse and rather specific environmental conditions. It should also be mentioned that multiple CheY paralogues observed in various non-enterics do not preclude, but on the contrary favour, the existence of a functional retro-phosphorelay pathway, as long as CheZ phosphatase is lacking.


   Swimming patterns and flagellar rotary control
TOP
Overview
Signalling pathways
Swimming patterns and flagellar...
Conclusions
REFERENCES
 
In a uniform environment, motile bacteria alternate between straight ‘runs’ and brief ‘tumbles’ that lead to changes in the direction of swimming (random walk). Sensory signals are used to bias the overall swimming direction towards attractants or away from repellents. This is accomplished by controls that modulate flagellar rotation.

The flagellar filaments and their mode of rotation differ greatly between enteric bacteria and S. meliloti and its relatives. In E. coli, the six to eight flexible left-handed helical flagella mostly rotate counterclockwise (ccw) to form a bundle that propels the cell forward. A new direction is taken up when some flagella switch to clockwise (cw) rotation, forcing the handedness and helical amplitude to change, which drives the bundle apart and causes the cell to tumble. Unlike E. coli, the soil bacterium S. meliloti is driven by rather rigid, right-handed helical filaments, which facilitate efficient swimming in a viscous environment (Götz et al., 1982 ). Whereas the ‘plain’ enterobacterial flagellar filaments are assembled from monomeric flagellin subunits, the ‘complex’ rhizobial filaments consist of four related flagellin subunits that are assembled as functional heterodimers (Scharf et al., 2001 ). The complex flagellar fine structure is dominated by three helical ribbons that lock the filament in right-handedness, thus blocking a switch of hand (Cohen-Krausz & Trachtenberg, 1998 ). Also, the motor of S. meliloti exhibits unidirectional cw rotation and never switches to ccw rotation. Hence, a swimming S. meliloti cell is driven by the unidirectional cw rotation of its right-handed helical filaments. Unlike in enterobacteria, but similar to R. sphaeroides (Packer et al., 1997 ), the flagellar motors can increase and decrease rotary speed (Götz & Schmitt, 1987 ; Platzer et al., 1997 ). Therefore, these bacteria respond to the addition of chemoattractants by a sustained increase in free-swimming speed, a phenomenon called ‘chemokinesis’ (Sourjik & Schmitt, 1996 ). Full-speed cw rotation causes the four to six right-handed filaments of S. meliloti to form a bundle that pushes the cell forward. A new direction is assumed when the rotary speeds of individual flagella decline at different rates, forcing their helical amplitudes to change asynchronously, which drives the bundle apart and causes the cell to turn or tumble. Hence, to change the direction of a swimming cell, E. coli flagellar rotors switch into reverse gear, whereas S. meliloti by independently slowing individual propellers follows the principle of a caterpillar drive.

In S. meliloti and the related Rhizobium lupini H13-3 (Scharf et al., 2001 ), rotary speed variation of the flagellar motor has a molecular corollary in two new motility proteins, MotC and MotD (Platzer et al., 1997 ), present in addition to the ubiquitous MotA/MotB transmembrane proteins that form the energizing proton channels (Fig. 1). MotC is a periplasmic protein interacting with and presumably stabilizing the periplasmic domain of MotB. The cytoplasmic MotD is essential for motility and, when overexpressed, enhances swimming and rotary speed (Platzer et al., 1997 ). In vitro, MotD binds to FliM, the very component of the cytoplasmic portion of the rotor (C-ring) to which CheY2-P also binds to effect a slow-down of rotary speed (E. Eggenhofer, B. Scharf & R. Schmitt, unpublished). MotD and CheY2-P may compete in an antagonistic fashion for binding to the ~34 copies of FliM (Thomas et al., 1999 ) that form the cytoplasmic surface of the rotor. In such a model, the actual ratio of MotD:CheY2-P bound to the C-ring would thus determine the rotary speed at each moment. Such a mechanism can be tested by monitoring the effects of an inducible ‘resurrection’ of CheY2-P (or an activated CheY2 derivative) on flagellar rotary speed reduction, as has been described by Scharf et al. (1998) .


   Conclusions
TOP
Overview
Signalling pathways
Swimming patterns and flagellar...
Conclusions
REFERENCES
 
Recent studies of chemosensory signal transduction in members of the {alpha}-subclass of the Proteobacteria have revealed gross differences of structures and mechanisms, although the central pathways are conserved. By choosing S. meliloti as a familiar and well-studied model organism, I have reviewed here the regulatory phosphate shuttle between the two response regulators CheY1 and CheY2, and rotary speed variation of right-handed helical flagella as examples for two novel mechanistic aspects of chemotaxis with relevance to many other bacterial species. We are only beginning to understand these mechanisms, and we are left with major questions such as: (i) what enables CheY2-P (but not CheY1-P) to restore by retro-phosphorylation the high-energy phosphoamidate bond of CheA-P? (ii) what is the regulatory or adaptive advantage of a retro-phosphotransfer relay over a simple CheZ-accelerated dephosphorylation of activated response regulator? (iii) how are signals propagated in the rotor complex upon CheY2-P (or MotD) binding to FliM? (iv) are flagellar rotary speed variations a consequence of changes in the protonmotive force (energy flow) or of subtle changes controlling the interacting forces at the rotor–stator interface? Future studies of {alpha}-proteobacteria and other non-enteric bacteria will be essential for solving such questions. In addition to revealing the wealth of evolutionary inventiveness, the solutions may grant new insights into fundamental principles of chemosensory signal transduction.


   ACKNOWLEDGEMENTS
 
I thank Patrick Babinger for artwork. Our studies of S. meliloti have been supported by grants from the Deutsche Forschungsgemeinschaft (Schm68/24-3 and Schm68/34-1).


   REFERENCES
TOP
Overview
Signalling pathways
Swimming patterns and flagellar...
Conclusions
REFERENCES
 
Adler, J. (1965). Chemotaxis in Escherichia coli. Cold Spring Harbor Symp Quant Biol 30, 289–292.

Armitage, J. P. & Schmitt, R. (1997). Bacterial chemotaxis: Rhodobacter sphaeroides and Sinorhizobium meliloti – variations on a theme? Microbiology 143, 3671-3682.[Abstract]

Blattner, F. R., Plunkett, G., 3rd, Bloch, C. A. & 14 other authors (1997). The complete genome sequence of Escherichia coli K-12. Science 277, 1453–1474.[Abstract/Free Full Text]

Bren, A. & Eisenbach, M. (2000). How signals are heard during bacterial chemotaxis: protein-protein interactions in sensory signal propagation. J Bacteriol 182, 6865-6873.[Free Full Text]

Bren, A., Welch, M., Blat, Y. & Eisenbach, M. (1996). Signal termination in bacterial chemotaxis: CheZ mediates dephosphorylation of free rather than switch-bound CheY. Proc Natl Acad Sci USA 93, 10090-10093.[Abstract/Free Full Text]

Cohen-Krausz, S. & Trachtenberg, S. (1998). Helical perturbations of the flagellar filament: Rhizobium lupini H13-3 at 13 resolution. J Struct Biol 122, 267-282.[Medline]

Falke, J. J., Bass, R. B., Butler, S. L., Chervitz, S. A. & Danielson, M. A. (1997). The two-component signaling pathway of bacterial chemotaxis: a molecular view of signal transduction by receptors, kinases, and adaptation enzymes. Annu Rev Cell Dev Biol 13, 457-512.[Medline]

Fraser, C. M., Casjens, S., Huang, W. M. & 35 other authors (1997). Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi. Nature 390, 580–586.

Galibert, F., Finan, T. M., Long, S. R. & 53 other authors (2001). The composite genome of the legume symbiont Sinorhizobium meliloti. Science 293, 668–672.

Götz, R. & Schmitt, R. (1987). Rhizobium meliloti swims by unidirectional, intermittent rotation of right-handed flagellar helices. J Bacteriol 169, 3146-3150.[Medline]

Götz, R., Limmer, N., Ober, K. & Schmitt, R. (1982). Motility and chemotaxis in two strains of Rhizobium with complex flagella. J Gen Microbiol 128, 789-798.

Grebe, T. W. & Stock, J. (1998). Bacterial chemotaxis: the five sensors of a bacterium. Curr Biol 8, R154-157.[Medline]

Greck, M., Platzer, J., Sourjik, V. & Schmitt, R. (1995). Analysis of a chemotaxis operon in Rhizobium meliloti. Mol Microbiol 15, 989-1000.[Medline]

Hamblin, P. A., Maguire, B. A., Grishanin, R. N. & Armitage, J. P. (1997). Evidence for two chemosensory pathways in Rhodobacter sphaeroides. Mol Microbiol 26, 1083-1096.[Medline]

Hess, J. F., Oosawa, K., Kaplan, N. & Simon, M. I. (1988). Phosphorylation of three proteins in the signaling pathway of bacterial chemotaxis. Cell 53, 79-87.[Medline]

Jumas-Bilak, E., Michaux-Charachon, S., Bourg, G., Ramuz, M. & Allardet-Servent, A. (1998). Unconventional genomic organization in the alpha subgroup of the Proteobacteria. J Bacteriol 180, 2749-2755.[Abstract/Free Full Text]

Kaneko, T., Sato, S., Kotani, H. & 21 other authors (1996). Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions (supplement). DNA Res 3, 185–209.[Medline]

Kearns, D. B., Robinson, J. & Shimkets, L. J. (2001). Pseudomonas aeruginosa exhibits directed twitching motility up phosphatidylethanolamine gradients. J Bacteriol 183, 763-767.[Abstract/Free Full Text]

Macnab, R. M. (1996). Flagella and motility. In Escherichia coli and Salmonella: Cellular and Molecular Biology, pp. 123–145. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.

Manson, M. D., Armitage, J. P., Hoch, J. A. & Macnab, R. M. (1998). Bacterial locomotion and signal transduction. J Bacteriol 180, 1009-1022.[Free Full Text]

Marwan, W. & Oesterhelt, D. (2000). Archaeal vision and bacterial smelling. ASM News 66, 83-89.

Nelson, K. E., Clayton, R. A., Gill, S. R. & 26 other authors (1999). Evidence for lateral gene transfer between Archaea and bacteria from genome sequence of Thermotoga maritima. Nature 399, 323–329.

Ochman, H. & Wilson, A. C. (1987). Evolution in bacteria: evidence for a universal substitution rate in cellular genomes. J Mol Evol 26, 74-86.[Medline]

Olsen, G. J., Woese, C. R. & Overbeek, R. (1994). The winds of (evolutionary) change: breathing new life into microbiology. J Bacteriol 176, 1-6.[Medline]

Packer, H. L., Lawther, H. & Armitage, J. P. (1997). The Rhodobacter sphaeroides flagellar motor is a variable-speed rotor. FEBS Lett 409, 37-40.[Medline]

Parkhill, J., Wren, B. W., Mungall, K. & 18 other authors (2000). The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665–668.[Medline]

Parkinson, J. S. (1987). Doing behavioral genetics with bacteria. Genetics 116, 499-500.[Free Full Text]

Platzer, J., Sterr, W., Hausmann, M. & Schmitt, R. (1997). Three genes of a motility operon and their role in flagellar rotary speed variation in Rhizobium meliloti. J Bacteriol 179, 6391-6399.[Abstract]

Rosario, M. M., Fredrick, K. L., Ordal, G. W. & Helmann, J. D. (1994). Chemotaxis in Bacillus subtilis requires either of two functionally redundant CheW homologs. J Bacteriol 176, 2736-2739.[Abstract]

Scharf, B. E., Fahrner, K. A., Turner, L. & Berg, H. C. (1998). Control of direction of flagellar rotation in bacterial chemotaxis. Proc Natl Acad Sci USA 95, 201-206.[Abstract/Free Full Text]

Scharf, B., Schuster-Wolff-Bühring, H., Rachel, R. & Schmitt, R. (2001). Mutational analysis of Rhizobium lupini H13-3 and Sinorhizobium meliloti flagellin genes: importance of flagellin A for flagellar filament structure and transcriptional regulation. J Bacteriol 183, 5334-5342.[Abstract/Free Full Text]

Sourjik, V. & Schmitt, R. (1996). Different roles of CheY1 and CheY2 in the chemotaxis of Rhizobium meliloti. Mol Microbiol 22, 427-436.[Medline]

Sourjik, V. & Schmitt, R. (1998). Phosphotransfer between CheA, CheY1, and CheY2 in the chemotaxis signal transduction chain of Rhizobium meliloti. Biochemistry 37, 2327-2335.[Medline]

Stetter, K. O. (1995). Microbial life in hypothermal environments. ASM News 61, 285-290.

Thomas, D. R., Morgan, D. G. & DeRosier, D. J. (1999). Rotational symmetry of the C ring and a mechanism for the flagellar rotary motor. Proc Natl Acad Sci USA 96, 10134-10139.[Abstract/Free Full Text]

Tomb, J. F., White, O., Kerlavage, A. R. & 39 other authors (1997). The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388, 539–547.[Medline]