Journal of Chemical Ecology, Vol. 12, No. 5, 1986

BEHAVIORAL RESPONSES TO CHEMICAL CUES BY BACTERIA

DOUGLAS

H. B A R T L E T T 1 and P H I L I P M A T S U M U R A 2 IAgouron Institute 505 Coast Boulevard South La Jolla, California, 92037 2Department of Biological Sciences University of Illinois at Chicago P.O. Box 4348, Chicago, Illinois, 60680

(Received July 29, 1985; accepted October 23, 1985) Abstract--Bacterial chemotaxis presents a model sensory system in which cells modulate the direction of rotation of their flagella in response to gradients of certain chemicals. The chemotactic machinery of Escherichia coli is currently being systematically reduced to its individual components through the accomplishments of behavioral, physical, genetic, molecular genetic, and biochemical analyses. Thirteen of the so called "MCP-related" class of chemotaxis gene products are known. Transmembrane methyl-accepting chemotaxis proteins (MCPs) are important for the chemical sensing, signal generation, and sensory adaptation processes. Soluble chemotaxis proteins relay information from these MCPs to structural components of the flagella referred to as switch proteins. Emphasis here is on the separate roles each of these groups of chemotaxis proteins perform, as well as their protein-protein relationships. Key Words--Bacteria, chemotaxis, methyl-accepting chemotaxis proteins, molecular biology, signal transduction, switch proteins.

INTRODUCTION P r o k a r y o t i c o r g a n i s m s h a v e e v o l v e d a p l e t h o r a o f m e t h o d s to sense and respond to each o t h e r and to their e n v i r o n m e n t . F o r e x a m p l e , l u m i n e s c e n t marine bacteria secrete small a u t o i n d u c e r m o l e c u l e s into their surroundings w h i c h , w h e n present in sufficient c o n c e n t r a t i o n , induce the cells to p r o d u c e light ( E n g e b r e c h t and S i l v e r m a n , 1984). I n d i v i d u a l cells o f the soil b a c t e r i u m M y x o c o c c u s xanthus respond to c h e m i c a l signals f r o m one another and c o m e t o g e t h e r to f o r m a fruiting b o d y structure ( Z u s m a n , 1984). B l u e - g r e e n a l g a e respond to nitrogen 1071 0098-0331/86/0500-1071$05.00/0 9 1986PlenumPublishingCorporation

1072

BARTLETT AND MATSUMURA

deprivation by developing specialized cellular compartments for nitrogen fixation (Golden et al., 1985). Many other prokaryotic chemical response systems also exist. One of the most extensively studied bacterial response systems is that of chemotaxis. The chemotactic characteristics of a bacterial community often affect an entire ecosystem. Bacterial chemotaxis to oxygen in pond water may shield methane-producing bacteria from oxygen's toxic affects as well as provide carbon dioxide for the anerobic production of methane (Rosen, 1984). Soil bacteria chemotax to a wide variety of potential energy sources such as fungal propagules and plant exudates (Arora et al., 1983; Gitte et al., 1978). Marine bacteria may be attracted by coral and algal extracellular products and even rum distillery wastewater (Chet and Mitchell, 1976; Fuentes et al., 1983.). Chemotaxis is also believed to play an important role in certain human diseases, such as cholera wherein Vibrio cholerae must colonize the intestinal mucosa (Freter et al., 1979). The focus for this review concerns a class of chemotaxis in Escherichia coli (E. coli) which is governed by transmembrane methylatable chemotaxis proteins (MCPs). The bulk of bacterial chemotaxis research has been performed on E. coli or Salmonella typhimurium (S. typhimurium). There is virtually complete functional homology among the flagellar, motility, and chemotaxis genes between these two enteric gram-negative species (Defranco et al., 1979; Kutsukake et al., 1980). Chemotaxis research on these organisms has provided an impressive array of behavioral, physical, genetic, molecular, and biochemical data. For information concerning the related subject areas of bacterial flagellar formation, energy transduction for motility, as well as additional taxis reviews, other sources are available (Silverman, 1980; Berg and Khan, 1982; Parkinson and Hazelbauer, 1983; Taylor, 1983; Macnab and Aizawa, 1984; Mizuno et al., 1984; Ordal, 1985). BEHAVIOR

E. coli cells are 1-2 /zm in length, peritrichous, and may, depending on environmental conditions, contain up to 10 flagella per cell, each approximately 10-15/zm in length (Hilmen and Simon, 1976). The regulation of the motion of the flagella governs the cells' chemotactic responsiveness. It has been conclusively demonstrated that the flagellum operates as a rotary engine with the flagellar filaments playing a passive role much like the propeller for a ship (Berg and Anderson, 1973; Silverman and Simon, 1974). This bacterial motor may rotate in either the counterclockwise or clockwise direction. Counterclockwise rotation (as viewed along the flagellum towards the cell) is typically associated with positive wave propagation of the filaments and straight swimming. In this mode the filaments are all in a stable left-handed helical conformation, and

BACTERIAL CHEMOTAXIS

1073

together form a bundle of flagella which rotate in concert. During clockwise rotation, however, hydrodynamic forces induce the formation of right-handed filaments which cause the unwinding of the bundle (Macnab and Ornston, 1977). The net result is that the cell jerks about ("twiddles") until counterclockwise rotation is resumed and the flagellar bundle restored. In an isotropic environment, bacterial motility has been described as possessing a "random threedimensional walk" which is characterized by a series of runs (the product of counterclockwise rotation), which are interrupted every 1-2 sec by twiddles (the product of clockwise rotation) (Berg and Brown, 1972). Chemotactic behavior in E. coli consists of the ability to alter migration in response to gradients of chemicals. The response to attractant stimuli is a decreased probability of switching to tumbling, as well as an increased probability of switching from tumbling to smooth swimming (Khan and Macnab, 1980). The converse is true for repellent responses. These actions promote the net movement of a population of cells in a favorable direction. A diagram representing chemotactic behavior is shown in Figure 1. Because the cells do not actually modulate the angle of their turns, this is not a true taxis. Their responses have been more precisely described as klinokinesis, a biased random walk (Taylor, 1983). In addition to the movements of the cells, E. coli may also respond to certain attractants or repellents by increasing or decreasing the rate of its cell division, respectively (Sherman and Vorobyeva, 1983). After the initial perception of an attractant or repellent (also called a chemoeffector), cells respond along two distinct pathways, beginning with a rapid sensory transduction or excitation phase, and followed by the adaptation phase during which the cells return to normal motility. Excitation requires approximately 0.2 sec (Segall et ai., 1982; Block et al., 1982), while adaptation can be observed to last for several minutes. E. coli can sense chemical differences of only one part in 104 over the length of its entire body (Dahlquist et al., 1972). How is such sensitivity achieved? Rapid mixing experiments with known chemoattractants or repellents and bacteria have shown that the chemotactic system operates as a time averaging system rather than a spatial sensory system (Macnab and Koshland, 1972). Chemoeffector information is integrated while the bacteria swim through their environment. Implied in this system is therefore some type of rudimentary memory. che GENES, PRODUCTS, AND PRODUCT LOCATIONS

One of the advantages of studying sensory transduction in E. coli, especially as compared with higher organisms, has been the wealth of techniques available for genetic analysis. The study of mutants deficient in various aspects of chemotaxis has greatly aided the dissection of the individual processes in-

1074

BARTLETT AND MA'FSUMURA NO GRADIENT

ATTRACTANT

REPELLENI

\ FIG. 1. Effect of chemoeffectors on bacterial motility. In the absence of an attractant or repellent, a cell exhibits smooth swimming interrupted every 1-2 sec with a tumble (left). Cells presented with an attractant spend a greater percentage of their time smooth swimming (middle), while those stimulated with a repellent bias their motility towards tumbling (right). Taken with permission from Silverman (1980). volved. Furthermore, the availability of such mutants has provided the foundation for the recombinant DNA analyses which have since taken place. Nonchemotactic mutants (Che-) have been isolated by a variety of techniques involving such characteristics of Che- mutants as their smaller region of growth on 0.4% agar plates (Parkinson, 1976) and their migration characteristics on preformed attractant gradients (Armstrong et al., 1967; Aswad and Koshland, 1975). These mutants contain flagella which rotate but exhibit altered swimming patterns. Some of these mutants tumble constantly, while others rarely tumble, and some will respond to no chemotactic stimuli, while others are only defective in responding to a particular class of chemicals. Thus, mutants blocked at various points along the flow of sensory information processing have been obtained. The mapping of che (chemotaxis) genes in E. coli has been accomplished through Hfr • F - matings (Armstrong and Adler, 1969; Harayama et al., 1979) and deletion analysis in X che transducing phage (Silverman and Simon, 1977b; Parkinson, 1978; Slocum and Parkinson, 1983). All but two che genes have mapped to areas near 42 or 43 minutes on the E. coli genetic map. Hand in hand with the advent of molecular biology, techniques were developed which allowed the definition of che operon structure, as well as che gene product identification. X che transducing phage were used to direct the expression cloned che genes' products (Silverman et al., 1977). Also, che gene products have been identified using the minicell technique for selectively programming the de

BACTERIAL CHEMOTAXIS

1075

novo expression of cloned che genes on plasmids (Matsumura et al., 1977).

Altogether, 12 genes and gene products required for chemotaxis have been identified (Silverman and Simon, 1976; Silverman et al., 1977; Matsumura et al., 1977; Hazelbauer et al., 1981; Bartlett and Matsumura, 1984; Clegg and Koshland, 1985; Malakooti and Matsumura, in preparation). Most of these gene products have also been localized to the outer membrane, inner membrane, or soluble subcellular compartments (Ridgeway et al., 1977; D. Bartlett; PhD thesis, University of Illinois, Chicago; 1985; Clegg and Koshland, 1985). Table 1 lists the E. coli che genes, their map positions, product molecular weights, and product locations. METHYL-ACCEPTING CHEMOTAXIS PROTEINS

A group of randomly situated transmembrane proteins, termed transducers, or methyl-accepting chemotaxis proteins (MCPs), play key roles in both the excitation and adaptation phases of certain chemotactic responses. MCPs have been found in enteric bacteria, Caulobacter, photosynthetic bacteria, Bacillus, and Spirochaeta (Kort et al., 1975; Goldman et al., 1982; Kathariou and Greenberg, 1983; Shaw et al., 1983). Each MCP is responsible for integrating inforTABLE 1. MCP-RELATED CHEMOTAXIS GENES, GENE PRODUCTS, PRODUCT LOCATIONS, AND PROPOSED FUNCTIONS

Map position

Gene

Product molecular weight (kd)

Product location a

31 42 42 42 42 42 42 42

trg cheA cheW tar tap cheR cheB cheY

55 76, 66 b 12 65 65 28 38 11

IM IM/CYT CYT IM IM CYT/IM CYT/IM CYT

42 43 43 43 99

cheZ flaBH flaAll motD tsr

24 38 37 16 65

CYT/IM basal body (?) basal body (?) basal body (?) IM

Function MCP III transducer regulates demethylation ? MCP II transducer transducer ? methyltransferase methylesterase tumble regulator, inhibits methylation ? switch protein switch protein switch protein MCP I transducer

aiM = innermembrane localization, CYT = cytoplasmic localization. bThe cheA locus contains two overlapping genes whose products are 76 and 66 kd. The smaller cheA product is found exclusively in the cytoplasm, while the larger product is found in both the cytoplasm and the inner membrane.

1076

BARTLETT AND MATSUMURA

mation pertaining to one or a few chemoeffectors. In the case of E. coli this is accomplished through direct interactions on the outer face of the inner membrane with specific amino acids or with occupied periplasmic binding proteins (Anraku, 1968; Adler, 1969; Hazelbauer and Adler, 1971; Kellerman and Szmelcman, 1974; Willis and Furlong, 1974; Hazelbauer, 1975; Clarke and Koshland, 1979; Kondoh et al., 1979; Hedblom and Adler, 1980; Wang and Koshland, 1980). After signaling a flagellar response, the MCPs undergo changes in methylation that correlate with sensory adaptation (Silverman and Simon, 1977b; Springer et al., 1977). The degree of occupancy of an MCP by one of its chemoeffectors is a measure of that cell's current environment. The level of methylation of the MCPs appears to reflect the cell's chemoeffector environment in the past and may participate in the memory process. Curiously, in E. coli, the MCPs become more methylated in response to an appropriate attractant stimulus, while in the gram-positive Bacillus subtilis, the MCPs become less methylated (Goy et al., 1977; Goldman et al., 1982). The methylation-demethylation reactions occur at several 3,-glutamyl residues along each MCP (Kleene et al., 1977; Van der Weft and Koshland, t977) and are catalyzed by chemotaxis-specific enzymes. The cheR protein is the methyltransferase and the cheB protein is the methylesterase (Springer and Koshland, 1977; Stock and Koshland, 1978). S-Adenosylmethionine (SAM) provides the methyl groups (Armstrong, 1972). Methanol is the hydrolysis product (Toews and Adler, 1979). The cheB protein also performs a second modification: the irreversible deamidation of glutamine to glutamic acid residues on the MCPs that are then capable of being methylated (Rollins and Dahlquist, 1981; Sherris and Parkinson, 1981). The function of this process is unknown. Four MCP genes are known to exist: tar, tap, tsr, and trg (Ordal and Adler, 1974; Silverman and Simon, 1977a; Springer et al., 1977; Hazelbauer et al., 1981; Krikos et al., 1983). The tar gene product mediates responses to the attractants aspartate and maltose and to the repellents Co 2§ and Ni 2+ (Clarke and Koshland, 1979). The tsr gene product mediates responses to the attractant serine and to the repellents acetate, leucine, benzoate, and indole (Hedblom and Adler, 1980). The trg gene product is responsible for attraction to ribose and galactose (Kondoh et al., 1979). As yet no known function exists for the tap gene product. MCP-mediated chemotaxis does not require transport (Adler, 1969). The tar product is methylated four times, tsr product six times, and the trg gene product more than five (Kehry et al., 1983). MCP methylation is particularly interesting in view of its analogy with mammalian sensory systems. Both neurosecretion and macrophage chemotaxis have been found to involve reversible methylation (Odea et al., 1981; Aksamit and Backlund, 1983). Furthermore, these multiple MCP methylations are similar to the multiple phosphorylations of rhodopsin, a covalent modification associated with adaptation to light in vertebrates (Schichi and Sommers, 1978).

BACTERIAL CHEMOTAXIS

1077

All of the E. coli MCP genes have been sequenced, as well as the S. typhimurium tar gene (Boyd et al., 1981; Krikos et al., 1983; Russo and Koshland, 1983; Bollinger et al., 1984). This has allowed the determination of their primary amino acid sequence as well as their predicted hydrophilic and hydrophobic domains. All of the MCPs appear to follow the same general configuration. At the N-terminal end is a very hydrophobic region which has been proposed to serve as a signal sequence that spans the membrane. This is followed by the receptor binding site which extends into the periplasmic space, followed by a second membrane-spanning domain and, finally, the bulk of the protein resides at the cytoplasmic face. Biochemical analysis of the tar and tsr products has determined two tryptic peptides which localize to the cytoplasmic regions of both proteins and which are methylated and show similarities in amino acid sequence (Wang and Koshland, 1980; Kehry and Dahlquist, 1982). The extensive regions of nucleic acid sequence homology between tsr, tar, and tap genes have suggested that they comprise a gene family whose products might contain some functionally similar and some functionally different domains (Boyd et al., 1981). Indeed, two chimeric genes containing the 5' end of the tar gene fused to the 3' coding region of the tsr gene have been found to be chemotactically functional. In cells lacking both tar and tsr wild-type functions, the chimeric gene products were capable of triggering tactic responses to tar chemoeffectors (Krikos et al., 1985). There have been two possibilities proposed for MCP mediation of initiation of signal transduction (Krikos et al., 1983). It may occur through the MCP polypeptide backbone, pushing or pulling its transmembrane segments, thereby altering the structure of the MCP carboxyterminal portion present in the cytoplasm. Alternatively, chemoeffector binding may influence the aggregation-disaggregation properties of monomeric MCP molecules.

ADDITIONAL CHEMOTAXIS PROTEINS

In addition to the receptors, transducers, methyltransferase, and methylesterase, additional chemotaxis proteins exist which are known to play a key role in the signal transduction and/or adaptation process. They do so by interacting between the transducer molecules and the flagella. These are the products of the cheY, cheZ, cheA, and cheWgenes (Silverman and Simon, 1976; Parkinson, 1978). Mutations in any one of these genes, as well as the cheR and cheB genes result in a generally nonchemotactic mutant. By transcriptionally fusing the above genes to high level promoters, all of their gene products have been overexpressed to the level of detectability on SDS-polyacrylamide gels (Matsumura et al., 1984; Matsumura et al., unpublished data). Complementation tests between S. typhimurium and E. coli have suggested that the cheY gene product and cheR product (methyltransferase) interact (De-

1078

BARTLETT AND MATSUMURA

franco et al., 1979). Like cheR mutants, cheYmutants show a counterciockwise rotational bias. Reversion analysis of cheY and switch mutants (discussed in the section on switch proteins) suggest that they too interact (Parkinson and Parker, 1979; Parkinson et al., 1983). Hence, cheY appears to serve in some capacity to bridge the sensory and/or adaptation aparatus between the MCP-methyltransferase complex and the ftagellar aparatus. The cheY protein is the first chemotaxis protein which has been purified to greater than 99 % homogeneity and for which X-ray crystallographic data may soon be ready (Matsumura et al., 1984). In vitro it is eluted from cibacron blue columns with SAM (the methyl donor for MCP methylation) and SAM analogs in such a manner as to suggest that it is a noncatalytic SAM-binding protein rather than a catalytic methyltransferase. The cheY protein in vitro also inhibits EcoRI methyltransferase activity. Based on these results, it has been proposed (Matsumura et al., 1984) that a possible function for the cheY protein in chemotaxis is to negatively regulate cheR methyltransferase activity, either by interaction with the cheR product or cheR product-substrate complex. Since then, cheYprotein has been shown to inhibit cheRmediated methylation of MCPs, an effect which can be reversed if a ehemoeffector such as 5 mM aspartate is present (Vacante and Matsumura; unpublished results). This is shown in Figure 2. Flagella from cell wall-containing ceil envelopes, when energized, rotate exclusively counterclockwise; however, addition to cheY protein induces clockwise rotation (Eisenbach; personal communication). Likewise, overexpression of cheY protein has been noted to produce a clockwise rotationalbias (Clegg and Koshland, 1984). For these reasons cheY has been designated the tumble regulator. The cheZ product appears to close the circle of che protein interactions. Complementation tests between S. typhimurium and E. coli suggest that the cheZ and cheB (methylesterase) products interact (Defranco et al., 1979). Like cheB mutants, cheZ mutants possess a clockwise rotational bias. Revertants of cheZ often map to the same flagellar switch genes as cheK which also suggests that the cheZ product interacts at the flagella (Parkinson and Parker, 1979; Parkinson et al., 1983). Because both cheY and cheZ interact with the methylation machinery as well as the motor, they may both serve as second messengers which convey information to the motors about the methylation state of the MCPs (Parkinson et al., 1983). For example, enhanced interaction of cheY protein with the switch might reflect an increased rate of dissociation of a cheR-cheY complex from the MCPs, resulting in a decrease in the rate of MCP methylation. Interestingly, the cheZ protein is methylated, but no function has yet been assigned to this modification (Silverman and Simon, 1977b). The cheA gene has gained some notoriety due to the fact that it contains two overlapping genes: one which codes for a membrane-bound product, and a shorter cytoplasmic product formed by a later ribosome binding site (Matsumura et al., 1977; Smith and Parkinson, 1980). Both cheA products have been implicated with increasing demethylation of the MCP molecules (Springer and

BACTERIAL CHEMOTAXIS

1079

FIG. 2. Effects of cheY protein overexpression on MCP methylation in situ. The theY gene has been placed under the transcriptional control of the high level trp promoter. The MCPs of cells which have been permeabilized to tritiated methyl SAM show a decrease in radioactive methyl group incorporation following the induction and overexpression of the cheY gene product (upper left). This effect is not observed when an internal deletion within cheY is introduced (upper right). This inhibition of methylation may be reversed if a chemoeffector such as 5mM aspartate is added (lower panel), A complete description of the effect of cheY protein on MCP methylation will be presented elsewhere (Vacante and Matsumura, in preparation). Zanolari, 1984). However, reversion analysis of flagellar component genes implies that these gene products may associate with the flagellar motor (Parkinson, 1977). Because both the cheA and cheW genes are part of a known motility operon, and because related genes are often clustered, both cheA and cheWmay act at the motor. This implies that the cheA role in demethylation may be an indirect one. The cheA gene product has also been demonstrated in vitro to bind to the cheW product. In the presence o f cheA protein, cheW protein selectively

1080

BARTLETT AND MATSUMURA

binds to affinity columns prepared with antibody directed against the cheA protein (MacNally and Matsumura, in preparation). Both cheA and chew mutants possess a counterclockwise rotational bias, implying that they both have roles in tumbling (Parkinson, 1978).

SWITCH PROTEINS

In E. coli there exist three particular flagellar genes, each of which may be altered to impart Fla- (flagellar defect), Mot- (no flagellar rotation), or Che phenotypes. These are the flaBlI, flaAlI, and motD genes (no fla designation for the E. coli motD gene currently exists). The fact that genetic defects within the same gene may express themselves so differently underlines the obviously close relationships of flagellar assembly, rotation, and chemotaxis. These genes' products are believed to be at the interface of the chemotactic machinery and the flagellar structure. Because of the central role these proteins play in modulating flagellar rotation in all types of taxes, they have been termed the switch proteins (Macnab and Aizawa, 1984). Most likely these products are structural components of the motor which are also essential for energization function and sensory reception. These conclusions are based upon the facts that a class of chemotactic mutants, designated scyB (suppressors of cheY mutations) map to the flaBIl gene (Parkinson et al., 1983; Bartlett and Matsumura, 1984), because the cheC gene maps within the flaAII cistron (Silverman and Simon, 1973; Clegg and Koshland, 1985), and by analogy with experiments in S. typhimurium, the flaBIl, flaAII, and motD genes may acquire Mot- or Che- phenotypes in addition to their Fla- null phenotypes (Collins and Stocker, 1976; Warrick et al., 1977; Dean et al., 1983). Although no flagellar mutants mapping to the E. coli flaBII gene have been obtained through random mutagenic techniques, reversed genetic methodology (in which recombination between a cloned mutant gene and the corresponding wild-type locus occurs) has demonstrated that the null phenotype of the flaBII gene is indeed Fla- (Bartlett, PhD thesis). All the switch genes are located within two flagellar operons mapping at about 43 minutes. The flaBH gene is located in an operon whose genes, transcribed from left to right, are flaB1, flaBII, flaBlll, flaC, flaO, and flaE. The flaAlI and motD genes comprise part of the next downstream operon which is transcribed in the order flaAI, flaAH, motD, flbD, flaR, flaR, flaQ, and flaP (Silverman and Simon, 1973; Bartlett and Matsumura, 1984; Malakooti and Matsumura, in preparation). TheflaB locus was divided into three complementary groups based upon the ability of various deletions and insertions withflaBlocus-containing plasmids to complement specific flagellar and chemotaxis mu-

BACTERIAL CHEMOTAXIS

1081

tants, as well as code for polypeptides (Bartlett and Matsumura, 1984). The existence of threeflaB genes has been confirmed through DNA sequence analysis (Frantz et al., unpublished data). Similar approaches have been used to subdivide the flaA locus into the flaAI and flaAII genes (Malakooti and Matsumura, in preparation). Although the phenotype of aflaAI mutant is unknown, it has been given a fla designation because of the previous flaA designation to this locus. The switch proteins along with other flagellar components may comprise a portion of a structure located beneath flagellar basal bodies. Such a location would be consistent with their proposed roles of interacting with soluble che proteins as well as affecting flagellar rotation. They do not appear to be soluble proteins, as Ravid and Eisenbach (1984) have found that E. coli ghosts which contain energized flagella retain their rotational biases when prepared from strains beating che mutations within the flaAlI orflaBII genes. Similar ghosts prepared from cells with defects in soluble chemotactic products (such as cheY protein) do not retain such rotational biases. A subflagellar basal body structure has never been observed during morphological examinations of E. coli flagella. However, these techniques have involved lysozyme digestion of the cell wall, and detergent solubilization of the cell membrane, in which case many components of the basal body or a subbasal body structure may have been lost (Depamphilis and Adler, 1971; Hilmen and Simon, 1976). Isolated flagellar organelles from the gram-negative Aquaspirillure rubrum reveal adherent fibrils emanating from the basal bodies (Coulton and Murray, 1978). Analyses in Pseudomonas aeruginosa and Vibrio metchnikoviic have revealed similar results. Thin-section electron micrographs in these organisms have demonstrated a dense cluster of "ribosome-like" granules at the cytoplasmic face of the flagellum, approximately 60 nm in diameter (Vaituzis and Doetsch, 1969). Overexpression offlaBlI, flaAII, and rnotD genes results in most of the switch proteins partitioning to the soluble cell fraction (Clegg and Koshland, 1985; Bartlett, PhD thesis; Malakooti and Matsumura, in preparation). This may reflect either a loose, or a site-limiting cytoplasmic membrane association. A glimpse into the tendency of flagella-associated proteins to associate with the membrane has been investigated from their overall hydrophobicity determined from their amino acid composition (Kyte and Doolittle, 1982). We have found a complete correlation between these hydrophobicity values determined for the switch/basal body proteins, and their overexpressed products' subcellular localization. The more hydrophobic motA and flaAl products do not show sitelimited incorporation into the membrane fraction, while the less hydrophobic flaBI, flaBIl, flaBIII, motB, and motD products do (Table 2). The derived amino acid sequence of the three flaB genes and motD gene provides no membrane spanning domain for theflaBII protein, while theflaBl protein possesses a pos-

1082

BARTLETT AND MATSUMURA TABLE 2. RELATIONSHIP BETWEEN HYDROPHOBICITY AND SITE-LIMITED INCORPORATION INTO INNER MEMBRANE AMONG MOTILITY AND BASAL BODY PROTEINS

motA flaAI motD flaBIIl flaBII motB flaBI

Average hydrophobicity

Site-limited membrane incorporation

+0.353 +0.032 -0.036 -0.402 -0.429 -0.434 -0.474

No a No Yes Yes Yes Yes Yes

aData of Wilson and Macnab (personal communication).

sible membrane spanning region near its amino terminus, and the flaBlll and motD products contain possible membrane spanning sites near their carboxy termini (Frantz et al., unpublished results). One or both of these flagellar components, as well as others, may help to anchor the flaBIl, flaAH, or motD proteins into their positions. It is noteworthy that in S. typhimurium the analogous gene to the E. coli flaBI gene appears to comprise the cytoplasmic membranebound M-ring of the basal bodies. This is based upon examinations of basal body proteins from temperature-sensitive flagellar mutants (Aizawa et al., 1985). Intergenic suppression analysis using the mot alleles of the switch genes suggests that all three of their proteins may interact together (Macnab and Aizawa, 1984). Figure 3 includes possible protein-protein relationships among the sensory transducers, additional che proteins, and switch/basal body complex. In addition to the requirements for the above-mentioned chemotaxis proteins, there is also a role of phosphate in MCP-mediated chemotaxis. Cells depleted of methionine or ATP do not synthesize SAM, the methylation donor for MCP methylation. Tumbly cheC mutants which possess defects in the ftagellar switch will not swim smoothly when depleted of methionine, but will when depleted of ATP, as do all cells (even when their proton motive force remains high enough for the flagella to switch to tumbly swimming) (Arai, 1980; Galloway and Taylor, 1980; Shioi et al., 1982). This indicates a role for ATP outside of SAM synthesis. In the presence of 5 mM phosphate, even with low ATP levels, smooth swimming may be restored (Arai, 1980). Therefore some other phosphorylated compound besides ATP appears to be involved in chemotaxis. Ordal (1985) has proposed that this compound may serve as a cofactor for the cheY, cheA, or chew proteins for their flagellar interactions, or that kinase activity exists for one of the switch components.

1083

BACTERIAL CHEMOTAXIS

I

I

[

I

I

1 cell wall

I---T

methyItronsferose

methylesterase

FIo. 3. Schematic representation of sensory flow in E. coli MCP-dependent chemotaxis. A C O M M O N TAXIS SIGNAL?

Virtually nothing is known about the nature of the excitatory signal in chemotaxis or in any other taxis. The response system may simply involve a series of protein-protein, or protein-small molecule (such as Ca 2§ binding steps (Parkinson, 1977; Ordal and Fields, 1977; Ordal, 1985). In addition, there may be regulation of enzymes involved with the generation and degradation of this signal, such that magnitude amplification, allosteric cooperativity; or the newly discovered zero-order ultrasensitivity might be involved in a kinetic pathway (Koshland, 1984). There is some evidence for the role of both the cheB and cheZ gene products in signal transduction for a few different types of taxis. Thus it is possible that these different sensory systems may interface at one point along their signaling pathways. Impulse response studies in MCPmediated chemotactic mutants have indicated that only cheZ mutants have an altered impulse response time (Block et al., 1982; Segall et al., 1982). CheDmutants which map to the tsr gene and which may possess MCPI molecules locked into the signaling mode show defective cheB product activity (Callahan and Parkinson, 1985; Kehry et al., 1985). CheB- mutants appear to be defective in thermosensory excitation (Imae et al., 1984.). Furthermore, both cheB and cheZ mutants are defective in aerotaxis and the phosphotransferase-system-

1084

BARTLETT AND MATSUMURA

mediated taxis, both of which proceed via MCP methylation-independent adaptation pathways (Niwano and Taylor, 1982). FUTURE PROSPECTS

All of the E. coli MCP-related che genes have been cloned, and their products' primary amino acid sequence have either been determined, or are in the process of being ascertained (Boyd et al., 1981; Krikos et al., 1983; Bollinger et al., 1984 Matsumura et al., 1984; Frantz et al., unpublished data; Malakooti and Matsumura, in preparation; N. Mutoh and M. Simon, personal communication; J.S. Parkinson; personal communication). In the near future, site-directed mutagenesis and the determination of the nucleic acid sequences containing various che mutations will become important for more clearly defining structure and function relationships among these genes' products. Concerning the MCPs, such analyses may help the determination of chemoeffector binding sites, as well as domains on the proteins necessary for methylation, methylase binding, methylesterase binding, and signal transduction. Similar analyses may be directed towards discovering what moieties of the cheY protein bind SAM, cheR protein, or the flagellar switch. Furthermore, structure and function studies among the switch proteins may yield valuable information concerning their association with cheYprotein, cheZ protein, and each other. Does an elaborate submembrane basal body complex exist? A structure analogous to the E. coli ATPase can be envisioned (Downie et al., 1979; Cox et al., 1981). This analogy is strengthened by the fact that Mitchell (1985) has offered the fascinating ATPase rotating subunit hypothesis in which proton motive force drives rotation of the ATPase enzyme in a manner similar to that proposed for the bacterial flagellar motor. By examining the ability of switch or basal body products to fractionate into the membrane fraction in various switch or basel body mutants, a hierarchy of assembly into the basal body complex may be determined. In this way information concerning protein-protein interactions for these membrane-associated proteins may be determined. A wealth of information relating to chemotaxis is E. coli already exists, and the future purification of all the che proteins promises to elucidate new biochemical properties. Acknowledgments--We would like to thank N. Mutoh, M. Simon, J. S. Parkinson, B. Frantz, D. Vacante, J. Malakooti, D. Macnally, and J. Stader for their contribution of data prior to publication. This work was supported by Public Health Service Grant A11895 from the National Institute of Health to P.M.

REFERENCES ADLER, J. 1969. Chemoreceptors in bacteria. Science 166:1588-1600. AIZAWA, S-I. DEAN, G.E., JONES, C. J., MACNAB, R.M., and YAMAGUCHI, S, 1985. Purification

BACTERIAL CHEMOTAXIS

1085

and characterization of the flagellar hook-basal body complex of Salmonella typhimurium. J. Bacteriol. 161:836-849. AKSAM1T,R.R., and BACKLUNDP.S., 1983. Chemotaxis and methylation in a macrophage cell line. Surv. Immunol. Res. 2:150-154. ANRAKU, Y. 1968. Transport of sugars and amino acids in bacteria. I: Purification and specificity of the galactose- and leucine-binding proteins. J. Biol. Chem. 243:3116-3121. ARAI, T. 1980. Effect of arsenate on Chemotactic behavior of E. coli. J. Bacteriol. 145:803-807. ARMSTRONG, J.B. 1972. An S-adenosylmethione requirement for chemotaxis in Escherichia coli. Can. J. Microbiol. 18:1695-1701. ARMSTRONG, J.B., and ADLER, J. 1969. Location of genes for motility and chemotaxis on the Escheriehia coli genetic map. J. Bacteriol. 97:156-161. ARMSTRONG, J.B., ADLER, J., and DAHL, M.M. 1967. Nonchemotactic mutants of Escherichia coli. J. Bacteriol. 93:390-398. ARORA, D.K., FILONOW,A.B, and LOCKWOOD,J.L. 1983. Bacterial chemotaxis to fungal propagules in vitro and in soil. Can. J. MicrobioL 29:1104-1109. ASWAD, D., and KOSHLAND,D.E. 1975. Isolation, characterization, and complementation of Salmonella typhimurium chemotaxis mutants. J. Mol. Biol. 97:225-235. BARTLETT, D.H. and MATSUMURA,P. 1984. Identification of Escherichia coli region IIIfla gene products and description of two new flagellar genes. J. Bacteriol. 160:577-585. BERG, H.C., and ANDERSON,R.A. 1973. Bacteria swim by rotating their flagellar filaments. Nature 245:380-382. BERG, H.C. and BROWN,D.A. 1972. Chemotaxis in Eseherichia coli analyzed by three dimensional tracking. Nature 239:500-504. BERG, H.C. and KHAN, S. 1982. A model for the flagellar rotary motor, pp. 485-497, in H. Sund and C. Veeger (eds.). Mobility and Recognition in Cell Biology. Walter de Gruyter, Berlin. BLOCK, S.M., SErALL, S.E., and BERG, H.C. 1982. Impulse response times in bacterial chemotaxis. Cell 31:215-226. BOLLINGER, J., PARK,C., HARAYAMA,S., and HAZELBAUER,G. 1984. Structure of the trg protein: Homologies with and differences from other sensory transducers of Escherichia coli. Proc. Natl. Acad. Sci. U.S.A. 81:3287-3291. BOYD, A., KRIKOS,A. and SIMON, M. 1981. Sensory transducers of Escherichia coli are encoded by homologous genes. Cell 26:333-343. BOYD, A., KENDALL,K., and S~MON, M. 1983. Structure of the serine chemoreceptor in Escherichia coli. Nature 301:623-626. CALLAHAN, A.M., and PARKINSON,J.S. 1985. Genetics of methyl-accepting chemotaxis proteins in Escherichia coli--cheD mutations affect the structure and function of the tsr transducer. J. Baeteriol. 161:96-104. CHET, I., and MITCrfELL, R. 1976. Ecological aspects of microbial chemotactic behavior. Annu. Rev. Microbiol. 30:221-239. CLARKE, S., and KOSHLAND,D.E., JR. 1979. Membrane receptors for aspartate and serine in bacterial chemotaxis. J. Biol. Chem. 254:9695-9702. CLEGC, D.O., and KOSHLAND,D.E., JR. 1984. The role of a signalling protein in bacterial sensing: Behavioral effects of increased gene expression. Proc. Natl. Acad. Sei. U.S.A. 81:5056-5060. CLEGC, D.O., and KOSHLAND, D.E., JR. 1985. Identification of a bacterial sensing protein and effects of its elevated expression. J. Bacteriol. 162:398-405. COLLINS, A.L.T., and STOCKER, B.A.D. 1976. Salmonella typhimurium mutants in general chemotaxis. J. Baeteriol. 128:754-761. COULTON, J.W., and MURRAY,R.G.E. 1978. Cell envelope associations of Aquaspirillum serpens flagella. J. Bacteriol. 136:1037-1049. Cox, G.B., DOWNIE,A., LONGMAN,L., SENIOR,A.E., ASH, G., FAYLE,D.R.H., and GIBSON, F. 1981. Assembly of the adenosine triphosphatase complex in Escherichia coli: assembly of F o is dependent on the formation of specific F~ subunits, J. Bacteriol. 148:30-42.

1086

BARTLETT AND MATSUMURA

DAHLQUIST, F.W., LOVELY,P., and KOSHLAND,D.E., JR. 1972. Quantitative analysis of bacterial migration in chemotaxis. Nature (London). New Biol., 236:120-122. DEAN, G.E., AIZAWA,S., and MACNAB, R.M. 1983. flaAIl (motC, cheV) of Salmonella typhimurium is a structural gene involved with energization and switching of the flagellar motor. J. Baeteriol. 154:84-91. DEFRANCO, A.L., PARKINSON,J.S., and KOSHLAND,D.E., JR. 1979. Functional homology of chemotaxis genes in Escherichia coli and Salmonella typhimurium. J. Bacteriol. 139:107-114. DEPAMVHmlS, M.L., and ADLER, J. 1971. Fine structure and isolation of the hook-basal body complex of flagella from Escherichia coli and Bacillus subtilis. J. Bacteriol. 105:384-395. DOWME, J.A., GIBSON, F., and Cox, G.B. 1979. Membrane adenosine triphosphatases of prokaryotic cells. Ann. R. Bioch. 48:103-131. ENGEBRECltT, J., and SILVERMANN,M. 1984. Identification of genes and gene products necessary for bacterial bioluminescence. Proc. Natl. Acad. Sci. U.S.A. 81:4154-4158. ENGELMANN, T.W. 1983. Bacterium photomentricum ein Beitrag zur Vergleichenden Physiologie des Licht--und Farbensinnes. Pfluegers Arch. Gesamte Physiol. Menschen Tiere 30:95-124. FRETER, R., O'BRIEN, P.C.M., and MACSAI, M.S. 1979. Effect of chemotaxis on the interaction of cholera vibrios with intestinal mucosa. Am. J. Clin. Nutr. 32:128-132. FUENTES, F.A., BIAMON, E.J., and HAZEN, T.C. 1983. Bacterial chemotaxis to effluent from a rum distillery in tropical near shore coastal waters. AppL Environ. Microbiol. 46:1438-1441. GALLOWAY,R.J., and TAYLOR, B.L. 1980. Histidine starvation and adenosine-5'-triphosphate depletion in chemotaxis of Salmonella typhimurium. J. Bacteriol. 144:1068-1075. GITTU, R.R., Rm, P.V., and PATIL, R.B. 1978. Chemotaxis of Rhizobium species toward root exudate of Cicer ariatinum L. Plant Soil 50:553-566. GOLDEN, J.W., ROBINSON, S.J., and HASELKORN,R. 1985. Rearrangement of nitrogen fixation genes during heterocyst differentiation in the cyanobacterium Anabaena. Nature 314:419-423. GOLDMAN, D.J., WOROBEC, S.W., SIEGEL, S.B., HECr~ER, R.V., and Or~DAL, G.W. 1982. Chemotaxis in Bacillus subtilis: Effects of attractants on the level of MCP methylation and the role of demethylation in the adaptation process. Biochemistry 21:915-920. GoY, M.F., SPRINGER,M.S., and ADLER, J. 1977. Sensory transduction in Escherichia coli--role of a protein methylation in sensory adaptation. Proc. Natl. Acad. Sci. U.S.A. 74:4964-4968. HARAYAMA,S., PALVA,E.T., and HAZELBAtJER,G.L. 1979. Transposon insertion mutants of Escherichia coli K-12 defective in a component common to galactose and ribose chemotaxis. Mol. Gen. Genet. 171:193-203. HAZELBAUER,G.L. 1975. Maltose chemoreceptor of Escherichia coli. J. Bacteriol. 122:206-214. HAZELBA~JER,G.L., and ADLER, J. 1971. Role of the galactose binding protein in chemotaxis of Escherichia coli towards galactose. Nature (London), New Biol. 230:101-103. HAZELBAUER,G.L., ENGSTROM, P., and HARAYAMA,S. 1981, Methyl-accepting chemotaxis protein III and transducer gene trg. J. Bacteriol. 145:43-49. HEDBLOM, M.L., and ADLER, J. 1980. Genetic and biochemical properties of Escherichia coli mutants with defects in serine ehemotaxis. J. Baeteriol. 144:1040-1060. HILMEN, M., and SIMON, M., 1976. Motility and the structure of the bacterial flagella, pp. 35-45, in R. Goldman, T. Pollard, and J. Rosenbaum (eds.). Cell Motility. Cold Spring Harbor Press, Cold Spring Harbor, New York. IMAE, Y. MIZUNO, T., and MAEDA,K. 1984. Chemosensory and thermosensory excitation in adaptation deficient mutants of Escherichia coli. J. Bacteriol. 159:368-374. KATHAmOU, S., and GREENBERG, E.P. 1983. Chemoattractants elicit methylation of specific polypeptides in Spirochaeta aurantia J. Bacteriol. 156:95-100. KEHRY, M.R., and DAHLQUIST,F.W. 1982. The methyl-accepting chemotaxis proteins of Escherichia coli. J. Bacteriol. 257:10378-10386. KEHRY, M.R., DAHLQUIST, F.W., and BOND, M.W. 1983. Bacterial chemotaxis: The chemical properties of the CheB dependent modification, pp. 533-549, in H. Sund, and C. Veegers (eds.). Mobility and Recognition in Cell Biology. Walter de Gruyter, Berlin.

BACTERIAL CHEMOTAXIS

1087

KEHRY, M.R., DOAK, T.G., and DAHLQUIST, F.W. 1985. Abberrant regulation of methylesterase activity in eheD chemotaxis mutants of Eseheriehia eoli. J. Baeteriol. 161:105-112. KELLERMAN, O., and SZMELCMAN,S. 1974. Active transport of maltose in Escheriehia eoli. Eur. J. Biochem. 249:6926-6932. KHAN, S., and MACNAB, R.M. 1980. Proton chemical potential, proton electrical potential and bacterial motility. J. MoL Biol. 138:599-614. KLEENE, S.J., TOEWS, M.L. and ADLER,J. 1977. Isolation of glutamic methyl ester from an Escheriehia coIi membrane protein involved in chemotaxis. J. Biol. Chem. 252:3214-3218. KONDOH, H., BALL, C.B., and ADLER, J. 1979. Identification of a methyl-accepting chemotaxis protein for the ribose and galactose chemoreceptors of Eseheriehia eoli. Proe. Natl. Acad. Sci. U.S.A. 76:260-264. KORT, E.N., GOY, M.F., LARSEN,S.H. and ADLER, J. 1975. Methylation of a membrane protein involved in bacterial chemotaxis. Proc. Natl. Acad. Sci. U.S.A. 72:3939-3943. KOSHLAND, D.E. 1984. Control of enzyme activity and metabolic pathways. TIBS 9:155-159. KRIKOS, A., MUTOH, N., BOYD, A,, and SIMON, M. 1983. Sensory transducers of E. coli are composed of discrete structural and functional domains. Cell 33:615-622. KRIKOS, A., CONELY, P.M., BOYD, A., BERG, H.C., and SIMON,M.I. 1985. Chimeric chemosensory transducers ofE. eoli. Proc. Natl. Acad. Sci. U.S.A. 82:1326-1330. KUTSUKAKE,K., IINO, T., KOMEDA,Y., and YAMAGUCHI,S. 1980. Functional homology offla genes between Salmonella typhimurium and Escherichia coli. Mol. Gen. Genet. 178:59-67. KYTE, J., and DOOLITTLE, R.F. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol Biol. 157:105-132. MACNAB, R.M., and AIZAWA,S. 1984. Bacterial motility and the bacterial flagellar motor. Annu. Rev. Biophys. Bioenerg. 13:51-83. MACNAB, R.M., and KOSHLAND, D.E., JR. 1972. The gradient sensing mechanism in bacterial chemotaxis. Proc. Natl. Acad. ScL U.S.A. 69:2509-2512. MACNABR.M., and ORNSTON, M.K. 1977. Normal-to-curly flagellar transitions and their role in bacterial tumbling. Stabilization of an alternative quaternary structure by mechanical force. J. Mol. Biol. 112:1-30. MATSUMURA,P., SILVERMAN,M., and SIMON, M. 1977. Syntheis of mot and che gene products of Escherichia eoli programmed by hybrid colE1 plasmids in minicells. J. Bacteriol. 132:9961002. MATSUMURA,P., RYDELL, J.J., LINZMEIER,R., and VACANTE,D. 1984. Overexpression and sequence of the Escherichia coli cheY gene and biochemical activities of the cheY protein. J. Bacteriol. 160:36-41. MITCHELL, P. 1985. Molecular mechanics of protonmotive FoFi ATPases. FEBS Lett. 182:1-7. MIZUNO, T., MAEDA,K., and IMAE, Y. 1984. Thermosensory transduction in Escherichia coli, in F. Oosawa, T. Yoshioka, and H. Hayashi (eds.). Tmnsmembrane Signalling and Sensation. Japan Scientific Societies Press, Tokyo. NIWANO, M., and TAYLOR, B.L. 1982. Novel sensory adaptation mechanism in bacterial chemotaxis to oxygen and phosphotransferase substrates. Proc. Natl. Acad. Sci. U.S.A. 79:11-15. ODEA, R.F., VWEROS, O.H. and DILIBERTO, E.J., JR. 1981. Protein carboxylation: Role in the regulation of cell functions. Biochem. Pharmacol. 11 : 1163-1168. ORDAL, G.W. 1985. Bacterial chemotaxis: Biochemistry of behavior in a single cell. C.R.C.C.R. Mic., 12:95-130. ORDAL, G.W., and ADLER, J. 1974. Properties of mutants in galactose taxis and transport. J. Bacteriol. 117:517-526. ORDAL, G.W., and FIELDS, R.B. 1977. A biochemical mechanism for bacterial chemotaxis. J. Theor. Biol. 68:491-500. PARKINSON, J.S. 1976. cheA, cheB, cheC genes of Escherichia coli and their role in chemotaxis. J. Bacteriol. 126:758-770. PARK~NSON,J.S. 1977. Behavioral genetics in bacteria. Annu. Rev. Genet. 11:397-414.

1088

BARTLETT AND MATSUMURA

PARKINSON,I.S., 1978. Complementation analysis and deletion mapping of Escherichia coli mutants defective in chemotaxis. J. Bacteriol. 135:45-53. PARK1NSON,J.S., and HAZELBAUER,G.L. 1983. Bacterial chemotaxis: Molecular genetics of sensory transduction and chemotactics gene expression, pp. 293-318, in J. Beckwith, J. Davies, and J. Gallant (eds.). Gene Function in Prokaryotes. Cold Spring Harbor Press, Cold Spring Harbor, New York. PARKINSON,J.S., and PARKER,S.R. 1979. Interaction of the cheC and cheZ gene products is required for chemotactic behavior in Escherichia coll. Proc. Natl. Acad. Sci. U.S.A. 76:23902394. PARKINSON,J.S., PARKER,S.R., TALBOT, P.B., and HOUTS, S.E. 1983. Interactions between chemotaxis genes and flagellar genes in Escherichia coll. J. Bacteriol. RAVID, S., and EISENBACH,M. 1984. Direction of flagellar rotation in bacterial cell envelopes. J. Bacteriol. 158:222-230. RIDCEWAY, H.F., SILVERMAN,M., and SIMON, M. 1977. Localization of proteins controlling motility and chemotaxis in Escherichia coll. J. Bacteriol. 132:657-665. ROLLINS, C., and DAHLQUIST, F.W. 1981. The methyl-accepting chemotaxis proteins of E. coli: A repellent stimulated covalent modification distinct from methylation. Cell 25:333-340. ROSEN, G. 1984. Commensalism of methane-producing and motile aerobic bacteria in certain freshwater wetlands. Bull. Math. Biol. 46:333-336. Russo, A.F., and KOSHLAND,D.E., JR. 1983. Separation of signal transduction and adaptation functions of the aspartate receptor in bacterial sensing. Science 220:1016-1020. SCHICHI,H., and SOMMERS,R.L. 1978. Light dependent phosphorylation of rhodopsin. J: Biol. Chem. 253:7040-7046. SEGALL, I.E., MANSON, M.D., and BERG. H.C. 1982. Signal processing times in bacterial chemotaxis. Nature 296:855-857. SHAW, P., GOMES, S.L., SWEENEY,K., ELY, B., and SHAPIRO, L. 1983. Methylation involved in chemotaxis is regulated during Caulobacter differentiation. Proc. Natl. Acad. Sci. U.S.A: 80:5261-5265. SHERMAN,M.Y., and VOROBYEVA,N.V. 1985. The involvement of chemotaxis in the regulation of Escherichia coli cell division. Microbiology 53:419-422. SHERRIS, D., and PARKINSON,J.S. 1981. Posttranslational processing of methylaccepting chemotaxis proteins in Escherichia coll. Proc. Natl. Acad. Sci. U.S.A. 78:6051-6055. SHIOl, J.-I., GALLOWAY,R.J., NIWANO, M., CHIMNOCK,R.E., and TAYLOR. B.L. 1982. Requirement of ATP in bacterial chemotaxis. J. Biol. Chem. 257:7969-7975. SILVERMAN,M. 1980. Building bacterial flagella. Q. Rev. Biol. 55:395-408. StLVERMAN, M., and SIMON, M. 1973. Genetic analysis of bacteriophage Mu-induced flagellar mutants in Escherichia coll. J. Bacteriol. 116:114-122. SILVERMAN,M., and SIMON, M. 1974. Flagellar rotation and the mechanism of bacterial motility. Nature 249:73-74. SILVERMAN,M., and SIMON, M. 1976. Operon controlling motility and chemotaxis in Escherichia coll. Nature 264:577-580. SILVERMAN,M., and SIMON, M. 1977a. Identification of polypeptides necessa~ for chemotaxis in Escherichia coll. J. Bacteriol. 130:1317-1325. SILVERMAN,M., and SIMON, M. 1977b. Chemotaxis in Escherichia coli: Methylation of che gene products. Proc. Natl. Acad. Sci, U.S.A. 74:3317-3321. SILVERMAN,M., MATSUMURA,P., HILMEN,M., and SIMON, M. 1977. Characterization of lambda Escherichia coil hybrids carrying chemotaxis genes. J. Bacteriol. 130:877-887. SLOCUM,M.K., and PARKINSON,J.S. 1983. Genetics of methyl-accepting chemotaxis proteins in Escherichia coli: Organization of the tar region. J. Bacteriol. 155:565-577. SMITH, R.A., and PARKINSON,J.S. 1980. Overlapping genes at the cheA locus of Escherichia coll. Proc. Natl. Acad. Sci, U.S.A. 77:5370-5374.

BACTERIAL CHEMOTAX1S

1089

SPRINGER,W.R., and KOSHLAND,D.E., JR. 1977. Identification of a protein methyl-transfemse as the cheR gene product in the bacterial sensing system. Proc. Natl. Acad. Sci, U.S.A. 74:533537. SPRINGER,M.S., and ZANOLARI,B. 1984. Sensory transduction in Escherichia coli: Regulation of the demethylation rate by the CheA protein. Proc. Natl. Acad. Sci, U.S.A. 81:5061-5065. SPRINGER,M.S., GOY,M.F., and ADLER, J. 1977. Sensory transduction in Escherichia coli: Two complementary pathways of information processing that involve methylated proteins. Proc. Natl. Acad. Sci, U.S.A. 74:3312-3316. STOCKJ.B., and KOSHLAND,D.E. JR. 1978. A protein methylesterase involved in bacterial sensing. Proc. Natl. Acad. Sci, U.S.A. 75:3659-3663. TAYLOR, B.L. 1983. Role of proton motive force in sensory transduction in bacteria. Annu. Rev. Microbiol. 37:551-573. TOEWS, M.L., and ADLER, J. 1979. Methanol formation in vivo from methylated chemotaxis proteins in Escherichia coli. J. Biol. Chem. 254:1761-1764. VAITUZIS,Z., and DOETSCH, R.N. 1969. Relationship between cell wall, cytoplasmic membrane, and bacterial motility. J. BacterioL 100:512-521. VAN DERWERF,P., and KOSHLAND,D.E., JR. 1977. Identification of a glutamyl ester in a bacterial membrane protein involved in chemotaxis. J. Biol. Chem. 252:2793-2795. WANG,E.A., and KOSHLAND,D.E., JR. 1980. Receptor structure in the bacterial sensing system. Proc. NatL Acad. Sci, U.S.A. 77:7175-7179. WARRICK,H.M., TAYLOR,B.L., and KOSHLAND,D.E., JR. 1977. Chemotactic mechanism of Salmonella typhimurium: Preliminary mapping and characterization of mutants. J. Bacteriol. 130:223 -231. WILLIS, R.C., and FURLONG, C.W. 1974. Purification and properties of a ribose-binding protein from Escherichia coll. J. Biochem. 249:6926-6929. ZUSMAN, D.R. 1984. Cell-cell interactions and developments in Myxococcus xanthus, Q. Rev. Biol. 59:119-138.