Movies of bundling actin fibers

Non-Brownian vs. Brownian

Here is a movie of the bundling process for fibers of length 1 with initial mesh size 0.2 microns without thermal fluctuations. All of the motion here is driven by the cross links, and fibers of the same color are considered to be in the same bundle. When we add Brownian motion, we see how much faster the initial stages of bundling occur, as the system is being mixed so much faster.

Effect of mesh size

When the mesh size is larger, fibers can still bundle because they can still find each other by Brownian diffusion. Here are movies for initial mesh sizes 0.2 microns, 0.4 microns, and 0.8 microns.

Effect of CL concentration

Bundling can proceed at large mesh sizes only when the CL concentration is sufficiently large, since otherwise fibers that might be close to each other will not be able to attach by a CL. Here are movies showing the effect of this at initial mesh size 0.4 microns, with k_on being the base value, four times the base value, and one quarter of the base value. For simulations with the smallest k_on, notice how the fibers move around a lot but are unable to initially organize into small bundles, as is characteristic of the other simulations. There is instead a bias towards large bundles adding one fiber at a time.

Semiflexible filaments

We now move on to considering semiflexible filaments, i.e., those that fluctuate along their length in addition to rotating and translationally diffusing. The following movies show comparisons of semiflexible dynamics to rigid fiber dynamics (or dynamics of rigidly fluctuating fibers with cross linking forces):

• We begin with l_p/L=100, for which the filaments are pretty straight, and the statistics do not differ from the rigid fiber case. This movie shows a side by side comparison of fluctuating filaments with l_p/L=100 at left and rigid fibers at right.
• We then decrease to l_p/L=10. For this, the agglomeration statistics are different from the rigid case, so we compare fluctuating filaments with l_p/L=10 at left to filaments with rigid fluctuations and bending by cross linking forces (SF-RBD filaments) at right in this movie. The dynamics don't show much difference, which means the acceleration in bundling relative to rigid fibers is driven by bending from cross linking forces.
• Finally, we do the same comparison for l_p/L=1, with fluctuating filaments at left and SF-RBD filaments at right in this movie. The dynamics of agglomeration are faster for l_p/L=1, although not by too much (try pausing the movie at early times), and the fibers are noticeably more curved on the left (when they fluctuate). This translates to more curved bundles at later times.