Conditional inversion of the thermoresponse in Escherichia coli.

Mutants in Escherichia coli having defects in one of the methyl-accepting chemotaxis proteins, Tsr protein, which is the chemoreceptor and transducer for L-serine, showed a reduced but similar type of thermoresponse compared with wild-type strains; the cells showed smooth swimming upon temperature increase and tumbling upon temperature decrease. However, when the mutant cells were adapted to attractants such as L-aspartate and maltose, which are specific to another methyl-accepting chemotaxis protein, Tar protein, the direction of the thermoresponse was found to be inverted; a temperature increase induced tumbling and a temperature decrease induced smooth swimming. Consistent with this, the mutant cells showed inverted changes in the methylation level of Tar protein upon temperature changes. Wild-type strains but not Tar protein-deficient mutants exhibited the inverted thermoresponse when the cells were simultaneously adapted to L-aspartate and L-serine, indicating that Tar protein has a key role in the inversion of the thermoresponse. Thus, besides Tsr protein, Tar protein has a certain role in thermoreception. A simple model for thermoreception and inversion of the thermoresponse is also discussed.     

Pausing of flagellar rotation is a component of bacterial motility and chemotaxis.

When bacterial cells are tethered to glass by their flagella, many of them spin. On the basis of experiments with tethered cells it has generally been thought that the motor which drives the flagellum is a two-state device, existing in either a counterclockwise or a clockwise state. Here we show that a third state of the motor is that of pausing, the duration and frequency of which are affected by chemotactic stimuli. We have recorded on video tape the rotation of tethered Escherichia coli and Salmonella typhimurium cells and analyzed the recordings frame by frame and in slow motion. Most wild-type cells paused intermittently. The addition of repellents caused an increase in the frequency and duration of the pauses. The addition of attractants sharply reduced the number of pauses. A chemotaxis mutant which lacks a large part of the chemotaxis machinery owing to a deletion of the genes from cheA to cheZ did not pause at all and did not respond to repellents by pausing. A tumbly mutant of S. typhimurium responded to repellents by smooth swimming and to attractants by tumbling. When tethered, these cells exhibited a normal rotational response but an inverse pausing response to chemotactic stimuli: the frequency of pauses decreased in response to repellents and increased in response to attractants. It is suggested that (i) pausing is an integral part of bacterial motility and chemotaxis, (ii) pausing is independent of the direction of flagellar rotation, and (iii) pausing may be one of the causes of tumbling.     

Expression of the gltP gene of Escherichia coli in a glutamate transport-deficient mutant of Rhodobacter sphaeroides restores chemotaxis to glutamate

Rhodobacter sphaeroides is chemotactic to glutamate and most other amino acids. In Escherichia coli , chemotaxis involves a membrane-bound sensor that either binds the amino acid directly or interacts with the binding protein loaded with the amino acid. In R. sphaeroides , chemotaxis is thought to require both the uptake and the metabolism of the amino acid. Glutamate is accumulated by the cells via a binding protein-dependent system. To determine the role of the binding protein and transport in glutamate taxis, mutants were created by Tn 5 insertion mutagenesis and selected for growth in the presence of the toxic glutamine analogue γ-glutamyl-hydrazide. One of the mutants, R. sphaeroides MJ7, was defective in glutamate uptake but showed wild-type levels of binding protein. The mutant showed no chemotactic response to glutamate. Both glutamate uptake and chemotaxis were recovered when the gltP gene, coding for the H⁺-linked glutamate carrier of E. coli , was expressed in R. sphaeroides MJ7. It is concluded that the chemotactic response to glutamate strictly requires uptake of glutamate, supporting the view that intracellular metabolism is needed for chemotaxis in R. sphaeroides .     

Change in direction of flagellar rotation in Escherichia coli mediated by acetate kinase.

Strains of Escherichia coli lacking all cytoplasmic chemotaxis proteins except CheY swim smoothly under most conditions. However, they tumble when exposed to acetate. Acetate coenzyme A synthetase (EC 6.2.1.1) was thought to be essential for this response. New evidence suggests that the tumbling is mediated instead by acetate kinase (EC 2.7.2.1), which might phosphorylate CheY via acetyl phosphate. In strains that were wild type for chemotaxis, neither acetate coenzyme A synthetase, acetate kinase, nor phosphotransacetylase (EC 2.3.1.8) (and thus acetyl phosphate) was required for responses to aspartate, serine, or sugars sensed by the phosphotransferase system. Thus, acetate-induced tumbling does not appear to play an essential role in chemotaxis in wild-type cells.     

Roles of cheY and cheZ gene products in controlling flagellar rotation in bacterial chemotaxis of Escherichia coli.

To understand output control in bacterial chemotaxis, we varied the levels of expression of cellular cheY and cheZ genes and found that the overproduction of the corresponding proteins affected Escherichia coli swimming behavior. In the absence of other signal-transducing gene products, CheY overproduction made free-swimming cells tumble more frequently. A plot of the fraction of the population that are tumbling versus the CheY concentration was hyperbolic, with half of the population tumbling at 30 microM (25,000 copies per cell) CheY monomers in the cytosol. Overproduction of aspartate receptor (Tar) by 30-fold had a negligible effect on CheY-induced tumbling, so Tar does not sequester CheY. CheZ overproduction decreased tumbling in all tumbling mutants except certain flaAII(cheC) mutants. In the absence of other chemotaxis gene products, CheZ overproduction inhibited CheY-induced tumbling. Models for CheY as a tumbling signal and CheZ as a smooth-swimming signal to control flagellar rotation are discussed.     

Chemotactic-like response of Escherichia coli cells lacking the known chemotaxis machinery but containing overexpressed CheY

We describe a chemotactic-like response of Escherichia coli strains lacking most of the known chemotaxis machinery but containing high levels of the response regulator CheY. The bacteria accumulated in aspartate-containing capillaries, they formed rings on tryptone-containing semisolid agar, and the probability of counterclockwise flagellar rotation transiently increased in response to stimulation with aspartate (10(-10)-10(-5) M; the response was inverted at > 10(-4) M). The temporal response was partial and delayed, as was the response of a control wild-type strain having a high CheY level. alpha-Methyl-DL-aspartate, a non-metabolizable analogue of aspartate as well as other known attractants of E. Coli, glucose and, to a lesser extent, galactose, maltose and serine caused a similar response. So did low concentrations of acetate and benzoate (which, at higher concentrations, act as repellents for wild-type E. coli). Other tested repellents such as indole, Ni2+ and CO2+ increased the clockwise bias. These observations raise the possibility that, at least when the conventional signal transduction components are missing, a non-conventional chemotactic signal transduction pathway might be functional in E. coli. Potential molecular mechanisms are discussed.     

Sequence and characterization of Bacillus subtilis CheB, a homolog of Escherichia coli CheY, and its role in a different mechanism of chemotaxis

The Bacillus subtilis gene encoding CheB, which is homologous to Escherichia coli CheY, the regulator of flagellar rotation, has been cloned and sequenced. It has been verified, using a phage T7 expression system, by showing that a small protein, the same size as E. coli CheY, is actually made from this DNA. Despite the fact that the two proteins are 36% identical, with many highly conserved residues, they appear to play different roles. Unlike CheY null mutants, which swim smoothly, CheB null mutants tumble incessantly. However, a CheB point mutant swims smoothly, even in the presence of a plasmid bearing cheB, which restores the null mutants to wild type. Expression of CheB in wild type B. subtilis makes the cells exhibit more tumbling. Since both absence of CheB and presence of high levels of CheB cause tumbling, CheB appears to be required, in certain circumstances, for both smooth swimming and tumbling. Expression in wild type E. coli makes the cells smooth swimmers and strongly inhibits chemotaxis.     

Cluster II che genes from Pseudomonas aeruginosa are required for an optimal chemotactic response

Pseudomonas aeruginosa, a gamma-proteobacterium, is motile by means of a single polar flagellum and is chemotactic to a variety of organic compounds and phosphate. P. aeruginosa has multiple homologues of Escherichia coli chemotaxis genes that are organized into five gene clusters. Previously, it was demonstrated that genes in cluster I and cluster V are essential for chemotaxis. A third cluster (cluster II) contains a complete set of che genes, as well as two genes, mcpA and mcpB, encoding methyl-accepting chemotaxis proteins. Mutations were constructed in several of the cluster II che genes and in the mcp genes to examine their possible contributions to P. aeruginosa chemotaxis. A cheB2 mutant was partially impaired in chemotaxis in soft-agar swarm plate assays. Providing cheB2 in trans complemented this defect. Further, overexpression of CheB2 restored chemotaxis to a completely nonchemotactic, cluster I, cheB-deficient strain to near wild-type levels. An mcpA mutant was defective in chemotaxis in media that were low in magnesium. The defect could be relieved by the addition of magnesium to the swarm plate medium. An mcpB mutant was defective in chemotaxis when assayed in dilute rich soft-agar swarm medium or in minimal-medium swarm plates containing any 1 of 60 chemoattractants. The mutant phenotype could be complemented by the addition of mcpB in trans. Overexpression of either McpA or McpB in P. aeruginosa or Escherichia coli resulted in impairment of chemotaxis, and these cells had smooth-swimming phenotypes when observed under the microscope. Expression of P. aeruginosa cheA2, cheB2, or cheW2 in E. coli K-12 completely disrupted wild-type chemotaxis, while expression of cheY2 had no effect. These results indicate that che cluster II genes are expressed in P. aeruginosa and are required for an optimal chemotactic response.     

Cytoplasmic pH mediates pH taxis and weak-acid repellent taxis of bacteria.

Bacteria migrate away from an acid pH and from a number of chemicals, including organic acids such as acetate; the basis for detection of these environmental cues has not been demonstrated. Membrane-permeant weak acids caused prolonged tumbling when added to Salmonella sp. or Escherichia coli cells at pH 5.5. Tethered Salmonella cells went from a prestimulus behavior of 14% clockwise rotation to 80% clockwise rotation when 40 mM acetate was added and remained this way for more than 30 min. A low external pH in the absence of weak acid did not markedly affect steady-state tumbling frequency. Among the weak acids tested, the rank for acidity (salicylate greater than benzoate greater than acetate greater than 5,5-dimethyl-2,4-oxazolidinedione) was the same as the rank for the ability to collapse the transmembrane pH gradient and to cause tumbling. At pH 7.0, the tumbling responses caused by the weak acids were much briefer. Indole, a non-weak-acid repellent, did not cause prolonged tumbling at low pH. Two chemotaxis mutants (a Salmonella mutant defective in the chemotaxis methylesterase and an E. coli mutant defective in the methyl-accepting protein in MCP I) showed inverse responses of enhanced counterclockwise rotation in the first 1 min after acetate addition. The latter mutant had been found previously to be defective in the sensing of gradients of extracellular pH and (at neutral pH) of acetate. We conclude (i) that taxes away from acid pH and membrane-permeant weak acids are both mediated by a pH-sensitive component located either in the cytoplasm or on the cytoplasmic side of the membrane, rather than by an external receptor (as in the case of the attractants), and (ii) that both of these taxes involve components of the chemotaxis methylation system, at least in the early phase of the response.     

Acquisition of maltose chemotaxis in Salmonella typhimurium by the introduction of the Escherichia coli chemosensory transducer gene.

Escherichia coli and Salmonella typhimurium are closely related species. However, E. coli cells show maltose chemotaxis but S. typhimurium cells do not. When an E. coli chemotransducer gene (tarE), the product of which is required for both aspartate and maltose chemotaxis, was introduced by using a plasmid vector into S. typhimurium cells with a defect in the corresponding gene (tarS), the transformant cells acquired the ability for both aspartate and maltose chemotaxis. In contrast, when the tars gene was introduced into tarE-deficient E. coli cells, the transformant cells acquired aspartate chemotaxis but not maltose chemotaxis. These results indicate that the absense of maltose chemotaxis in S. typhimurium is a consequence of the properties of the tars gene product.     

Chemosensory and thermosensory excitation in adaptation-deficient mutants of Escherichia coli

Methyl-accepting chemotaxis protein-methyltransferase-deficient mutants, cheR mutants, of Escherichia coli showed a tumble response to repellents only at low temperatures, and the resultant tumbling lasted unless the condition was changed. The swimming pattern of the repellent-treated cells was different at different temperatures, indicating that the absolute temperature is a determinant of the tumbling frequency of those cells. The tumbling of those cells was also suppressed by the addition of attractants. Under a suitable repellent concentration, the tumbling frequency of the cells was found to be simply determined by the ligand occupancy of chemoreceptors for many attractants. In a methyl-accepting chemotaxis protein-methylesterase-deficient mutant, a cheB deletion mutant, the tumbling frequency was also determined by receptor occupancy of some attractants. These results indicate that in the adaptation-deficient mutants, sensory signals are produced in proportion to the amount of ligand-bound or of thermally altered receptors and transmitted to the flagellar motors without any modification. Thus, it is concluded that the adaptation system, namely, the methylation-demethylation system of methyl-accepting chemotaxis proteins, is not concerned with the step of chemosensory or thermosensory excitation. A simple model is proposed to explain how the swimming pattern of the adaptation-deficient mutants is determined.     

Chemosensory and photosensory perception in purple photosynthetic bacteria utilize common signal transduction components.

The chemotaxis gene cluster from the photosynthetic bacterium Rhodospirillum centenum contains five open reading frames (ORFs) that have significant sequence homology to chemotaxis genes from other bacteria. To elucidate the functions of each ORF, we have made various mutations in the gene cluster and analyzed their phenotypic defects. Deletion of the entire che operon (delta che), as well as nonpolar disruptions of cheAY, cheW, and cheR, resulted in a smooth-swimming phenotype, whereas disruption of cheB resulted in a locked tumbly phenotype. Each of these mutants was defective in chemotactic response. Interestingly, disruption of cheY resulted in a slight increase in the frequency of tumbling/reversal with no obvious defects in chemotactic response. In contrast to observations with Escherichia coli and several other bacteria, we found that all of the che mutant cells were capable of differentiating into hyperflagellated swarmer cells when plated on a solid agar surface. When viewed microscopically, the smooth-swimming che mutants exhibited active surface motility but were unable to respond to a step-down in light intensity. Both positive and negative phototactic responses were abolished in all che mutants, including the cheY mutant. These results indicate that eubacterial photosensory perception is mediated by light-generated signals that are transmitted through the chemotaxis signal transduction cascade.     

Folate responsiveness during growth and development of Dictyostelium: separate but related pathways control chemotaxis and gene regulation

Folate-controtled gene expression and chemotaxis have been examined in Dictyostelium wild-type and mutant strains. We show that regulation of the discoidin genes is sensitive to foiate in growing ceiis as weli as in suspension development. The signal is transferred via the N¹⁰-methylfoiate-sensitive folate receptor sites, which also appear to confer the chemotactic response. The strain HG5145 has previously been isolated as a mutant that does not display chemotactic movement towards folate. Nevertheless, these cells are fully functional in foiate-mediated downregulation of discoidin I expression. The strain ga 93 has been isolated as an overproducer mutant of the cyclic nucleotide phosphodiesterase inhibitor. Simultaneously, these cells fail to downregulate discoidin I in response to folate but are fully functional in folate chemotaxis. Therefore we conclude that the pathways for chemotaxis and for gene regulation diverge downstream of a common receptor type.     

Chemotactic signaling by an Escherichia coli CheA mutant that lacks the binding domain for phosphoacceptor partners

CheA is a multidomain histidine kinase for chemotaxis in Escherichia coli. CheA autophosphorylates through interaction of its N-terminal phosphorylation site domain (P1) with its central dimerization (P3) and ATP-binding (P4) domains. This activity is modulated through the C-terminal P5 domain, which couples CheA to chemoreceptor control. CheA phosphoryl groups are donated to two response regulators, CheB and CheY, to control swimming behavior. The phosphorylated forms of CheB and CheY turn over rapidly, enabling receptor signaling complexes to elicit fast behavioral responses by regulating the production and transmission of phosphoryl groups from CheA. To promote rapid phosphotransfer reactions, CheA contains a phosphoacceptor-binding domain (P2) that serves to increase CheB and CheY concentrations in the vicinity of the adjacent P1 phosphodonor domain. To determine whether the P2 domain is crucial to CheA's signaling specificity, we constructed CheADeltaP2 deletion mutants and examined their signaling properties in vitro and in vivo. We found that CheADeltaP2 autophosphorylated and responded to receptor control normally but had reduced rates of phosphotransfer to CheB and CheY. This defect lowered the frequency of tumbling episodes during swimming and impaired chemotactic ability. However, expression of additional P1 domains in the CheADeltaP2 mutant raised tumbling frequency, presumably by buffering the irreversible loss of CheADeltaP2-generated phosphoryl groups from CheB and CheY, and greatly improved its chemotactic ability. These findings suggest that P2 is not crucial for CheA signaling specificity and that the principal determinants that favor appropriate phosphoacceptor partners, or exclude inappropriate ones, most likely reside in the P1 domain.     

Genetics of methyl-accepting chemotaxis proteins in Escherichia coli: null phenotypes of the tar and tap genes.

The tar and tap genes are located adjacent to one another in an operon of chemotaxis-related functions. They encode methyl-accepting chemotaxis proteins implicated in tactic responses to aspartate and maltose stimuli. The functional roles of these two gene products were investigated by isolating and characterizing nonpolar, single-gene deletion mutants at each locus. Deletions were obtained by selecting for loss or a defective Mu d1 prophage inserted in either the tar or tap gene. The extent of the tar deletions was determined by genetic mapping with Southern hybridization. Representative deletion mutants were surveyed for chemotactic responses on semisolid agar and by temporal stimulation in a tethered cell assay to assess flagellar rotational responses to chemoeffector compounds. The tar deletion strains exhibited complete loss of aspartate and maltose responses, whereas the tap deletion strains displayed a wild-type phenotype under all conditions tested. These findings indicate that the tap function is unable to promote chemotactic responses to aspartate and maltose, and its role in chemotaxis remains unclear.     

Tar-dependent and -independent pattern formation by Salmonella typhimurium.

When Salmonella typhimurium cells were allowed to swarm on either a minimal or complex semisolid medium, patterns of cell aggregates were formed (depending on the thickness of the medium). No patterns were observed with nonchemotactic mutants. The patterns in a minimal medium were not formed by a mutant in the aspartate receptor for chemotaxis (Tar) or by wild-type cells in the presence of alpha-methyl-D,L-aspartate (an aspartate analog), thus resembling the patterns observed earlier in Escherichia coli (E. O. Budrene and H. C. Berg, Nature [London] 349:630-633, 1991) and S. typhimurium (E. O. Budrene and H. C. Berg, Abstracts of Conference II on Bacterial Locomotion and Signal Transduction, 1993). Distinctively, the patterns in a complex medium had a different morphology and, more importantly, were Tar independent. Furthermore, mutations in any one of the genes encoding the methyl-accepting chemotaxis receptors (tsr, tar, trg, or tcp) did not prevent the pattern formation. Addition of saturating concentrations of the ligands of these receptors to wild-type cells did not prevent the pattern formation as well. A tar tsr tcp triple mutant also formed the patterns. Similar results (no negative effect on pattern formation) were obtained with a ptsI mutant (defective in chemotaxis mediated by the phosphoenolpyruvate-dependent carbohydrate:phosphotransferase system [PTS]) and with addition of mannitol (a PTS ligand) to wild-type cells. It therefore appears that at least two different pathways are involved in the patterns formed by S. typhimurium: Tar dependent and Tar independent. Like the Tar-dependent patterns observed by Budrene and Berg, the Tar-independent patterns could be triggered by H(2)O(2), suggesting that both pathways of pattern formation may be triggered by oxidative stress.     

The amino terminus of the aspartate chemoreceptor is formylmethionine

The amino terminus of the Salmonella typhimurium aspartate receptor has been identified as formylmethionine by mass spectral analysis of the amino-terminal tryptic peptide. Purification and analysis of the blocked amino-terminal peptide was facilitated by the use of a mutant aspartate receptor which has a cysteine residue at position 3. The sequence of this peptide confirms the translational start site predicted from the nucleotide sequence of the tar gene. Furthermore, in vivo labeling experiments reveal that the formyl group is present on chemotaxis receptors produced at wild-type levels in Escherichia coli, indicating that the presence of the formyl group is not a consequence of over-production of the receptor. The stability of the amino-terminal formyl group on the receptor may be a consequence of the membrane localization of the receptor and the dependence of this localization on the membrane transport machinery of the cell.     

Identification of a gene encoding a methyl-accepting chemotaxis-like protein from Campylobacter coli and its use in a molecular typing scheme for campylobacters

Using PCR amplification with degenerate primers, a gene ( tlpA ) from Campylobacter coli encoding a putative 63·0 kDa polypeptide which exhibited significant identity with bacterial methyl-accepting chemotaxis proteins (MCPs) was identified. A mutant containing an inactivated copy of the tlpA gene showed a wild-type chemotactic response to all of the chemo-attractants tested. A DNA probe based on the Highly Conserved Domain (HCD) of TlpA revealed the presence of multiple copies of genes encoding MCP-like proteins in both Camp. coli and Camp. jejuni. The arrangement of restriction sites within, and proximal to, genes with homology to the HCD probe varied among strains, resulting in a high degree of polymorphism. It is demonstrated here that a DNA probe comprising the HCD region of MCP-like proteins can be used, in Southern hybridization-based assays, to provide novel information which allows the discrimination of individual strains of Camp. coli and Camp. jejuni.     

Identification and characterization of FliY, a novel component of the Bacillus subtilis flagellar switch complex

The Bacillus subtilis gene encoding FliY has been cloned and sequenced. The gene encodes a 379-amino-acid protein with a predicted molecular mass of 41,054 daltons. FliY is partly homologous to the Escherichia coli and Salmonella typhimurium switch proteins FliM and FliN. The N-terminus of FliY has 33% identity with the first 122 amino acids of FliM, whereas the C-terminus of FliY has 52% identity with the last 30 amino acids of FliN. The middle 60% of FliY is not significantly homologous to either of the proteins. A fliY::cat null mutant has no flagella. Motility can be restored to the mutant by expression of fliY from a plasmid, although chemotaxis is still defective since the strain exhibits smooth swimming behaviour. fliY::cat is in the cheD complementation group. One of the cheD point mutants does not switch although the population grown from a single cell has both smooth swimming and tumbling bacteria, implying that the switch is locked. Expression of fliY in wild-type B. subtilis makes the cells more smooth-swimming but does not appear to affect chemotaxis. Expression of fliY in wild-type S. typhimurium severely inhibits chemotaxis and also makes the cells smooth swimming. Expression in a non-motile S. typhimurium fliN mutant restores motility but not chemotaxis, although expression in a non-motile E. coli fliM mutant does not restore motility. The homology, multiple phenotypes, and interspecies complementation suggest that FliY forms part of the B. subtilis switch complex.(ABSTRACT TRUNCATED AT 250 WORDS)