How can they figure out which way to go? After all, the chemicals are going to diffuse all over the place. True, a blood vessel is a simple one-dimensional region, but do you go left or right to get to the hammered thumb?
The model this group was testing is pretty simple and clever: the seeking cells destroy the trouble chemicals near them. After that the region with a higher density of those chemicals will diffuse into the newly cleaned regions by the seeker cell first. The seeker cell detects the alarm chemicals on (e.g.) its left side first--so that's the direction the seeker cell should move. That's a pretty clever way to estimate a density gradient, and it has the added benefit of getting rid of the old alarm chemicals so the body isn't in constant alert mode.
They have videos of ameobas and cancer cells threading mazes to find "chemoattractants".
It isn't infallible, of course--brownian motion might randomly bring attractants to the low density side first. You can see that in the videos. But most of the time it works, and it seems like a simple process to implement. It'd be more robust than a simple "try to measure concentration on both sides" scheme, since very high densities of chemoattractants would overwhelm a seeker cell's finite number of built-in receptors.
No comments:
Post a Comment