Optical trap based studies of single, bundled, and cross-linked microtubules

Cytoskeletal networks within cells provide structural support and fulfill crucial biological roles for eukaryotic organisms. These networks display emergent behaviors like highly nonlinear elasticity that originate in the properties of individual filaments and the three-dimensional architecture of the network. In particular, the connectivity between filaments has major implications for the stress distribution and mechanics of the network, which, in combination with the stiffness of the constituent filaments, determine the magnitude of thermal fluctuations within the network. Beyond their structural importance, cytoskeletal networks also serve as highways for intracellular transport while simultaneously acting as obstructions that hinder diffusive motion on the nanoscopic scale. The immense difficulty associated with studying these networks from a single filament up to a collection of cross-linked filaments means that there is a dearth of experimental data in this area. Here, we fill this gap with two novel methods of studying cytoskeletal filaments focusing specifically on microtubules, semiflexible filaments with rich behavior on the single filament level and network properties that depart from those of other cytoskeletal networks. First, we use thermal noise imaging, a high bandwidth, probe-based technique, to observe cross-linked networks of microtubules. New analysis methods allow us to localize individual filaments with 2.5 nm precision, one-tenth of the diameter of a microtubule, and additionally to track filaments’ positions through intersections in the network, which have thwarted localization attempts in previous studies. This leads to the first direct measurement of a cross-link between connected filaments with a distance of just 1.4 nm found to be separating the microtubules. Furthermore, observations of filament fluctuations show that they are directly connected to the network architecture, suggesting that tensions within the network caused by cross-links may be quantified by these measurements. Diffusion of a spherical particle is also tracked throughout these experiments so that spatial variations in the diffusion constant can be mapped with a resolution of 10 nm. These measurements show that single microtubules can dramatically influence the motion of our probe particle, and, since the probe is similar in size to vesicles within cells, microtubules may play a significant role in the intracellular diffusion of such vesicles and other biologically important particles. After analyzing cross-linked networks, we then present the standing wave optical trap (SWOT) as a means of manipulating individual filaments with the potential to craft custom networks of microtubules. We show that not only can the SWOT trap a single microtubule, it can also manipulate microtubule bundles and individual filaments through the polarization of the trapping laser. This surprising degree of experimental control can only be explained if microtubules have a birefringence of at least 0.02, almost ten times larger than previously thought, opening the door to new possibilities for crafting designer networks filament by filament.