Biomedical subjects
H C Berg
Publications and source records attributed to H C Berg.
Interacting components of the flagellar motor of Escherichia coli revealed by the two-hybrid system in yeast.
The ability of the flagellar motor of Escherichia coli to switch between clockwise and counterclockwise modes of operation is ultimately responsible for the swimming behavior of the cell. Three motor proteins, FliG, FliM, and FliN, have been implicated in this process. Using the two-hybrid system in Saccharomyces cerevisiae, we demonstrated strong interactions between FliG/FliM,FliM/FliM, and FliM/FliN. A screen for other components that might interact with FliG revealed interactions with FliF (the MS ring protein) and H-NS (a histone-like protein). Regions of proteins important for several of these interactions were identified by mutational analysis. The implications for motor assembly and function are discussed.
Torque-generating units of the bacterial flagellar motor step independently.
Measurements of the variance in rotation period of tethered cells as a function of mean rotation rate have shown that the flagellar motor of Escherichia coli is a stepping motor. Here, by measurement of the variance in rotation period as a function of the number of active torque-generating units, it is shown that each unit steps independently.
Temperature-induced switching of the bacterial flagellar motor.
Chemotaxis signaling proteins normally control the direction of rotation of the flagellar motor of Escherichia coli. In their absence, a wild-type motor spins exclusively counterclockwise. Although the signaling pathway is well defined, relatively little is known about switching, the mechanism that enables the motor to change direction. We found that switching occurs in the absence of signaling proteins when cells are cooled to temperatures below about 10 degrees C. The forward rate constant (for counterclockwise to clockwise, CCW to CW, switching) increases and the reverse rate constant (for CW to CCW switching) decreases as the temperature is lowered. At about -2 degrees C, most motors spin exclusively CW. At temperatures for which reversals are frequent enough to generate a sizable data set, both CCW and CW interval distributions appear to be exponential. From the rate constants we computed equilibrium constants and standard free energy changes, and from the temperature dependence of the standard free energy changes we determined standard enthalpy and entropy changes. Using transition-state theory, we also calculated the activation free energy, enthalpy, and entropy. We conclude that the CW state is preferred at very low temperatures and that it is relatively more highly bonded and restricted than the CCW state.
Torque generated by the bacterial flagellar motor close to stall.
In earlier work in which electrorotation was used to apply external torque to tethered cells of the bacterium Escherichia coli, it was found that the torque required to force flagellar motors backward was considerably larger than the torque required to stop them. That is, there appeared to be substantial barrier to backward rotation. Here, we show that in most, possibly all, cases this barrier is an artifact due to angular variation of the torque applied by electrorotation, of the motor torque, or both; the motor torque appears to be independent to speed or to vary linearly with speed up to speeds of tens of Hertz, in either direction. However, motors often break catastrophically when driven backward, so backward rotation is not equivalent to forward rotation. Also, cells can rotate backward while stalled, either in randomly timed jumps of 180 degrees or very slowly and smoothly. When cells rotate slowly and smoothly backward, the motor takes several seconds to recover after electrorotation is stopped, suggesting that some form of reversible damage has occurred. These findings do not affect the interpretation of electrorotation experiments in which motors are driven rapidly forward.
A mutational analysis of the interaction between FliG and FliM, two components of the flagellar motor of Escherichia coli.
The motor that drives the flagellar filament of Escherichia coli contains three "switch" proteins (FliG, FliM, and FliN) that together determine the direction of rotation. Each is required, in addition, for flagellar assembly and for torque generation. These proteins interact in the Saccharomyces cerevisiae two-hybrid system: FliG interacts with FliM, FliM interacts with itself, and FliM interacts with FliN. The interaction between FliG and FliM has been subjected to mutational analysis. FliG (fused to the GAL4 DNA-binding domain) and FliM (fused to a GAL4 transcription activation domain) together activate transcription of a GAL4-dependent lacZ reporter gene. DNA encoding FliG was mutagenized by error-prone amplification with Taq polymerase, mutant fliG genes were cloned (as DNA-binding domain-fliG gene fusions) in S. cerevisiae by gap repair of plasmid DNA, and mutants exhibiting an interaction defect were isolated in a two-hybrid screen. The mutations were each mapped to the first, second, or last third of the fliG gene by multifragment cloning in vivo and then identified by DNA sequencing. In this way, we identified 18 interaction-defective and 15 silent (non-interaction-defective) fliG mutations. Several residues within the middle third of FliG are strongly involved in the FliG-FliM interaction, while residues near the N or C terminus are less important. This clustering, when compared with results of previous studies, suggests that the FliG-FliM interaction plays a central role in switching.
Dynamics of formation of symmetrical patterns by chemotactic bacteria.
Motile cells of Escherichia coli aggregate to form stable patterns of remarkable regularity when grown from a single point on certain substrates. Central to this self-organization is chemotaxis, the motion of bacteria along gradients of a chemical attractant that the cells themselves excrete. Here we show how these complex patterns develop. The long-range spatial order arises from interactions between two multicellular aggregate structures: a 'swarm ring' that expands radially, and focal aggregates that have lower mobility. Patterning occurs through alternating domination by these two sources of excreted attractant (which we identify here as aspartate). The pattern geometries vary in a systematic way, depending on how long an aggregate remains active; this depends, in turn, on the initial concentration of substrate (here, succinate).
Powering the flagellar motor of Escherichia coli with an external voltage source.
Rotary motors of bacterial flagella are driven by ions that move across the cytoplasmic membrane down an electrochemical gradient. For Escherichia coli, the ions are protons, and the maximum work per unit charge that they can do is the protonmotive force. To test whether motor efficiency is limited by proton leakage or mechanical nonlinearities, we measured torque as a function of protonmotive force. Filamentous cells were drawn into micropipettes and energized with an external voltage source. Torque was proportional to protonmotive force up to -150 mV, twice the span accessible by earlier techniques. This is consistent with a mechanism in which a fixed number of protons, working at unit efficiency, carry the motor through each revolution. We also found that individual torque-generating elements inactivate at low potentials or potentials of reverse sign. When normal potentials are restored, they reactivate sequentially.
Three-dimensional tracking of motile bacteria near a solid planar surface.
Knowing how motile bacteria move near and along a solid surface is crucial to understanding such diverse phenomena as the migration of infectious bacteria along a catheter, biofilm growth, and the movement of bacteria through the pore spaces of saturated soil, a critical step in the in situ bioremediation of contaminated aquifers. In this study, a tracking microscope is used to record the three-dimensional motion of Escherichia coli near a planar glass surface. Data from the tracking microscope are analyzed to quantify the effects of bacteria-surface interactions on the swimming behavior of bacteria. The speed of cells approaching the surface is found to decrease in agreement with the mathematical model of Ramia et al. [Ramia, M., Tullock, D. L. & Phan-Tien, N. (1993) Biophys J. 65,755-778], which represents the bacteria as spheres with a single polar flagellum rotating at a constant rate. The tendency of cells to swim adjacent to the surface is shown in computer-generated reproductions of cell traces. The attractive interaction potential between the cells and the solid surface is offered as one of several possible explanations for this tendency.
Fluctuation analysis of rotational speeds of the bacterial flagellar motor.
We measured the dependence of the variance in the rotation rate of tethered cells of Escherichia coli on the mean rotation rate over a regime in which the motor generates constant torque. This dependence was compared with that of broken motors. In either case, motor torque was augmented with externally applied torque. We show that, in contrast to broken motors, functioning motors in this regime do not freely rotationally diffuse and that the variance measurements are consistent with the predicted values of a stepping mechanism with exponentially distributed waiting times (a Poisson stepper) that steps approximately 400 times per revolution.
Cells of Escherichia coli swim either end forward.
Chemotactic cells of the bacterium Escherichia coli were marked asymmetrically by growth on a rich medium containing tetrazolium red. When this dye is reduced, it tends to form a refractile granule near one end of the cell, readily visualized by dark-field microscopy. In smooth-swimming cells, the marker was found with equal probability in front or behind. In wild-type cells, tumbles changed the cell orientation nearly as often as not. Some cells formed flagellar bundles at one end more frequently than at the other, but the run-interval distributions were the same either way. We conclude that the sensory system does not favor one end of the cell over the other. Thus, chemoreceptors that appear in patches at only one pole do not serve as a nose.
Mechanical limits of bacterial flagellar motors probed by electrorotation.
We used the technique of electrorotation to apply steadily increasing external torque to tethered cells of the bacterium Escherichia coli while continuously recording the speed of cell rotation. We found that the bacterial flagellar motor generates constant torque when rotating forward at low speeds and constant but considerably higher torque when rotating backward. At intermediate torques, the motor stalls. The torque-speed relationship is the same in both directional modes of switching motors. Motors forced backward usually break, either suddenly and irreversibly or progressively. Motors broken progressively rotate predominantly at integral multiples of a unitary speed during the course of both breaking and subsequent recovery, as expected if progressive breaking affects individual torque-generating units. Torque is reduced by the same factor at all speeds in partially broken motors, implying that the torque-speed relationship is a property of the individual torque-generating units.
Spatio-temporal patterns generated by Salmonella typhimurium.
We present experimental results on the bacterium Salmonella typhimurium which show that cells of chemotactic strains aggregate in response to gradients of amino acids, attractants that they themselves excrete. Depending on the conditions under which cells are cultured, they form periodic arrays of continuous or perforated rings, which arise sequentially within a spreading bacterial lawn. Based on these experiments, we develop a biologically realistic cell-chemotaxis model to describe the self-organization of bacteria. Numerical and analytical investigations of the model mechanism show how the two types of observed geometric patterns can be generated by the interaction of the cells with chemoattractant they produce.
Self-electrophoresis is not the mechanism for motility in swimming cyanobacteria.
Swimming cyanobacteria do not have flagella. In principle, they could be propelled by streams of ions flowing from head to tail, i.e., by a self-electrophoretic mechanism. We have ruled out this possibility by showing that cells of a swimming Synechococcus species fail to drift in an external electric field.
Torque generation by the flagellar rotary motor.
A review is given of the structure and dynamics of the flagellar rotary motor. Force-generating elements in a motor driving a tethered bacterium (a cell fixed to the substratum by a single flagellum) exert forces of order 20 pN while moving at speeds of order 1 micron/s. Force-generating elements in a motor driving a flagellar filament in a bundle exert forces some 10-fold lower but move at speeds more than 10-fold higher. The motor torque-speed relationship has been measured over a wide dynamic range. Motors strongly resist being driven backwards and are easily broken.
A mutant hook-associated protein (HAP3) facilitates torsionally induced transformations of the flagellar filament of Escherichia coli.
Two mutants with defects in hook-associated protein 3 (HAP3) were isolated that exhibit impaired swimming only when they interact with a solid surface or a semisolid matrix. Motility and chemotaxis were normal in liquid media, even in media containing viscous agents, but cells failed to swarm in 0.28% agar. Mutants appeared to carry a full complement of flagella of normal configuration and length. However, filaments rotating counterclockwise close to a glass surface transformed from normal to straight, while filaments rotating clockwise transformed from curly to straight. Both transformations propagated from base to tip, as expected if torsionally induced. The mutations mapped to the middle of flgL, to structural gene for HAP3, and sequence analysis revealed the same coding change in both mutants: a substitution of cysteine for arginine 168. Our results show that the ability of a filament composed of normal flagellin subunits to resist mechanical stress depends on the structure of the protein (HAP3) to which it is attached at its base. The N-terminal sequence of HAP3 was found to be similar to the N-terminal sequence of flagellin, and the possibility that it provides a nucleation site for the C-terminal region of flagellin is discussed.
pH dependence of CheA autophosphorylation in Escherichia coli.
Chemotaxis by cells of Escherichia coli and Salmonella typhimurium depends upon the ability of chemoreceptors called transducers to communicate with switch components of flagellar motors to modulate swimming behavior. This communication requires an excitatory pathway composed of the cytoplasmic signal transduction proteins, CheAL, CheAS, CheW, CheY, and CheZ. Of these, the autokinase CheAL is most central. Modifications or mutations that affect the rate at which CheAL autophosphorylates result in profound chemotactic defects. Here we demonstrate that pH can affect CheAL autokinase activity in vitro. This activity exhibits a bell-shaped dependence upon pH within the range 6.5 to 10.0, consistent with the notion that two proton dissociation events affect CheAL autophosphorylation kinetics: one characterized by a pKa of about 8.1 and another exhibiting a pKa of about 8.9. These in vitro results predict a decrease in the rate of CheAL autophosphorylation in response to a reduction in intracellular pH, a decrease that should cause increased counterclockwise flagellar rotation. We observed such a response in vivo for cells containing a partially reconstituted chemotaxis system. Benzoate (10 mM, pH 7.0), a weak acid that when undissociated readily traverses the cytoplasmic membrane, causes a reduction of cytoplasmic pH from 7.6 to 7.3. In response to this reduction, cells expressing CheAL, CheAS, and CheY, but not transducers, exhibited a small but reproducible increase in the fraction of time that they spun their flagellar motors counterclockwise. The added presence of CheW and the transducers Tar and Trg resulted in a more dramatic response. The significance of our in vitro results, their relationships to regulation of swimming behavior, and the mechanisms by which transducers might affect the pH dependence of CheA autokinase activity are discussed.
Bacterial motility and signal transduction.
Explore the source record for details and available documents.