Difference between revisions of "From molecular noise to behavioral variability in a single bacterium"

From Soft-Matter
Jump to: navigation, search
Line 1: Line 1:
Original entry:  Donald Aubrecht,  APPHY 226,  Spring 2009
"From molecular noise to behavioural variability in a single bacterium"<br>
"From molecular noise to behavioural variability in a single bacterium"<br>
Ekaterina Korobkova, Thierry Emonet, Jose M.G. Vilar, Thomas S. Shimizu, & Philippe Cluzel<br>
Ekaterina Korobkova, Thierry Emonet, Jose M.G. Vilar, Thomas S. Shimizu, & Philippe Cluzel<br>

Latest revision as of 01:31, 24 August 2009

Original entry: Donald Aubrecht, APPHY 226, Spring 2009

"From molecular noise to behavioural variability in a single bacterium"
Ekaterina Korobkova, Thierry Emonet, Jose M.G. Vilar, Thomas S. Shimizu, & Philippe Cluzel
Nature 428 574-578 (2004)

Soft Matter Keywords

bacteria, E. coli, chemotaxis, signaling pathway

Figure 1. Schematic view of the experimental apparatus with insets: (a) binary time series indicating direction of flagella rotation, and (b) clockwise bias versus time.
Figure 2. Noise of the chemotaxis network. (a) is the power spectrum of the network output from one non-stimulated wild-type cell. (b) is the mean spectrum computed from averaging wild-type and mutant cells. The black line is the mean spectrum from 40 wild-type cells, while the gray line is the spectral characteristics of the bacterial motor. (c) is the distribution of clockwise (gray) and counterclockwise (black) intervals from a cell. (d) is the distribution of clockwise (gray) and counterclockwise (black) intervals from eight mutant cells.
Figure 3. Behavioral variability as a function of CheR concentration. (a) is the power spectrum for several differnt CheR concentration. (b) is the counterclockwise interval distributions for the same cells as in (a). (c) is the network output signals for a wild-type cell (black) and a mutant cell (gray) with similar clockwise bias and switching frequency. Network output is defined as (bias - mean(bias))/mean(bias).
Figure 4. Simulation results: (a) power spectrum for increasing concentration of CheR, and (b) variability of network output.


The authors present primarily experimental work regarding a study of the chemotaxis network governing the motion of Escherichia coli. In the past, experiments and models for this system have assumed that network properties can be inferred from population measurements, which has the unfortunately effect of masking temporal fluctuations of intracellular signaling events. Korobkova, et al. study a noise analysis of behavioral variations in individual bacteria as a route to inferring fundamental properties of the chemotaxis network. They observe some properties established by population measurements to not be conserved at the single-cell level and find behavior of non-stimulated bacteria displaying temporal variations much larger than expected statistical fluctuation. The authors have also found that the temporal behavioral variablity is strong dependent on the concentration of a key network component.

Practical Application of Research

Though not immediately applicable in its own right, this research helps elucidate some of the features of the biological network and feedback that governs the motion of E. coli. Understanding this network could allow for precise genetic design of mutant strains of E. coli with particular locomotion controls. This is also a nice example of the overlap between physics and biology in which standard physical analyses are extended to biological systems to yield new isights.

Noise Analysis of Individual Bacterium

The authors study how the behavior of an individual bacterium in a homogeneous environment fluctuates with time. Their primary question is whether there are specific molecular events that could cause temporal behavioral variability in an individual bacterium. To perform this study, bacteria were immobilized onto microscope slides and their flagella marked with micro-beads to visualize the rotation with dark-field microscopy. The output from these observations was converted to a binary time series showing the direction of rotation (see Figure 1a).

Binary time series from 222 individual cells were averaged before spectral analysis. The resulting power spectrum is shown in Figure 2a. This power spectrum differs from population behavior for timescales ranging from a few to 1000 seconds. To study the network more thoroughly, the authors also use a mutant signaling molecule to control the preferred direction for flagella rotation. When phosphorylated, the response regulator CheY binds preferentially to the cytoplasmic base of the bacterial motors and an increase in CheY-P causes the motors to spend more time spinning clockwise. The activated mutant CheYD13K mimics the effect of CheY-P, but does not need to be phosphorylated. The spectrum results from this mutant molecule are shown as the gray line in Figure 2b, where the black line shows the wild-type response for similar clockwise bias.

From tests with methyltransferase CheR, the authors were able to further explore the temporal behavioral variability. Figure 3 shows results from varying concentration of the CheR molecule. As the concentration of CheR is increased, the temporal behavioral variability is reduced and the counterclockwise intervals tend toward the exponential behavior expected from population studies.

As a final check, the authors performed stochastic simulations using the StochSim software package. They simulate the temporal fluctuations in CheY-P concentration and computer the corresponding power spectra for a variety of CheR concentrations. For wild-type concentrations of CheR, the wild-type spectra is qualitatively recovered (see Figure 4a).

From these experiments and simulation, the authors conclude that within individual bacteria, molecular noise emerges as a tunable source behavioral variability

written by Donald Aubrecht