Watch a cell sense and decide

Here's a swimming E. coli cell, drawn at true size, with its chemotaxis machinery running inside it. You can turn the cell around, look through its wall, add attractant, and watch the cell respond and then adapt. Everything on the page below is live, so the numbers on the right change as you play.

What to try, and what to break

Press Add attractant first. The kinase activity reading drops to zero within moments, which is what a real cell does, and the motor settles into long smooth runs. Then wait. Methylation creeps upward, and activity climbs back to where it started even though the attractant is still there. That return is adaptation, and the page on adaptation explains why it comes back to exactly the same level.

Next, play geneticist. Switch the mutant to no CheR and add attractant again. Without the enzyme that adds methyl groups, activity stays low and the cell never recovers. Switch to no CheB and you get the opposite problem, because methyl groups can only be added. Both of these match what Alon and colleagues (1999) saw in mutant cells observed. Turn on See-through to look inside the cell, and Labels to name the parts.

Real geometry, toy model

The drawing is real geometry wrapped around a toy model. The sizes of the cell, flagella and motor come from published measurements of E. coli, and the figure says which sizes still need checking. The behaviour comes from a small set of equations for receptor activity, methylation, the messenger CheY-P and the motor modeled.

Before you trust the numbers, two warnings. The resting kinase activity of 0.30 is the median of 54 single cells measured by Keegstra and colleagues (2017) observed, so that part rests on data. The speed doesn't. A real cell takes about 414 seconds to recover half-way after a large step, and the figure does it in seconds so that you can watch, which is roughly 200 times faster. The way the response grows with attractant dose hasn't yet been matched to the published dose-response measurements, and the attractant units in the figure aren't micromolar.

Where to go next

The adaptation loop pattern page takes the slow part of this machinery on its own. The glossary explains CheA, CheR, CheB, CheY and the other names on the labels.

Reference. Keegstra, J. M., Kamino, K., Anquez, F., Lazova, M. D., Emonet, T., & Shimizu, T. S. (2017). Phenotypic diversity and temporal variability in a bacterial signaling network revealed by single-cell FRET. eLife, 6, e27455. doi:10.7554/eLife.27455. Alon, U., Surette, M. G., Barkai, N., & Leibler, S. (1999). Robustness in bacterial chemotaxis. Nature, 397, 168-171. doi:10.1038/16483.