The actin and microtubule cytoskeletons form active networks in the cell that can contract and remodel, resulting in vital cellular processes such as cell division and motility. Motor proteins play an important role in generating the forces required for these processes, but more recently the concept of passive cross-linkers being able to generate forces has emerged. So far, these passive cross-linkers have been studied in the context of separate actin and microtubule systems. Here, we show that cross-linkers also allow actin and microtubules to exert forces on each other. More specifically, we study single actin filaments that are cross-linked to growing microtubule ends, using in vitro reconstitution, computer simulations, and a minimal theoretical model. We show that microtubules can transport actin filaments over large (micrometer-range) distances and find that this transport results from two antagonistic forces arising from the binding of cross-linkers to the overlap between the actin and microtubule filaments. The cross-linkers attempt to maximize the overlap between the actin and the tip of the growing microtubules, creating an affinity-driven forward condensation force, and simultaneously create a competing friction force along the microtubule lattice. We predict and verify experimentally how the average transport time depends on the actin filament length and the microtubule growth velocity, confirming the competition between a forward condensation force and a backward friction force. In addition, we theoretically predict and experimentally verify that the condensation force is of the order of 0.1 pN. Thus, our results reveal an active mechanism for local actin remodeling by growing microtubules that relies on passive cross-linkers.

In vitro (or cell-free) reconstitution is a powerful tool to study the physical basis of cytoskeletal organization in eukaryotic cells. Cytoskeletal reconstitution studies have mostly been done for individual cytoskeleton systems in unconfined 3D or quasi-2D geometries, which lack complexity relative to a cellular environment. To increase the level of complexity, we present a method to study co-organization of two cytoskeletal components, namely microtubules and actin filaments, confined in cell-sized water-in-oil emulsion droplets. We show that centrosome-nucleated dynamic microtubules can be made to interact with actin filaments through a tip-tracking complex consisting of microtubule end-binding proteins and an actin-microtubule cytolinker. In addition to the protocols themselves, we discuss the optimization steps required in order to build these more complex in vitro model systems of cytoskeletal interactions.

Crosstalk between the actin and microtubule cytoskeletons underlies cellular morphogenesis. Interactions between actin filaments and microtubules are particularly important for establishing the complex polarized morphology of neurons. Here, we characterized the neuronal function of growth arrest-specific 2-like 1 (Gas2L1), a protein that can directly bind to actin, microtubules and microtubule plus-end-tracking end binding proteins. We found that Gas2L1 promotes axon branching, but restricts axon elongation in cultured rat hippocampal neurons. Using pull-down experiments and in vitro reconstitution assays, in which purified Gas2L1 was combined with actin and dynamic microtubules, we demonstrated that Gas2L1 is autoinhibited. This autoinhibition is relieved by simultaneous binding to actin filaments and microtubules. In neurons, Gas2L1 primarily localizes to the actin cytoskeleton and functions as an actin stabilizer. The microtubule-binding tail region of Gas2L1 directs its actin-stabilizing activity towards the axon. We propose that Gas2L1 acts as an actin regulator, the function of which is spatially modulated by microtubules.

Filamentous proteins are responsible for the superior mechanical strength of our cells and tissues. The remarkable mechanical properties of protein filaments are tied to their complex molecular packing structure. However, since these filaments have widths of several to tens of nanometers, it has remained challenging to quantitatively probe their molecular mass density and three-dimensional packing order. Scanning transmission electron microscopy (STEM) is a powerful tool to perform simultaneous mass and morphology measurements on filamentous proteins at high resolution, but its applicability has been greatly limited by the lack of automated image processing methods. Here, we demonstrate a semi-automated tracking algorithm that is capable of analyzing the molecular packing density of intra- and extracellular protein filaments over a broad mass range from STEM images. We prove the wide applicability of the technique by analyzing the mass densities of two cytoskeletal proteins (actin and microtubules) and of the main protein in the extracellular matrix, collagen. The high-throughput and spatial resolution of our approach allow us to quantify the internal packing of these filaments and their polymorphism by correlating mass and morphology information. Moreover, we are able to identify periodic mass variations in collagen fibrils that reveal details of their axially ordered longitudinal self-assembly. STEM-based mass mapping coupled with our tracking algorithm is therefore a powerful technique in the characterization of a wide range of biological and synthetic filaments

Transiently crosslinked actin filament networks allow cells to combine elastic rigidity with the ability to deform viscoelastically. Theoretical models of semiflexible polymer networks predict that the crosslinker unbinding rate governs the timescale beyond which viscoelastic flow occurs. However a direct comparison between network and crosslinker dynamics is lacking. Here we measure the network's stress relaxation timescale using rheology and the lifetime of bound crosslinkers using fluorescence recovery after photobleaching (FRAP). Intriguingly, we observe that the crosslinker unbinding rate measured by FRAP is more than an order of magnitude slower than the rate measured by rheology. We rationalize this difference with a three-state model where crosslinkers are bound to either 0, 1 or 2 filaments, which allows us to extract crosslinker transition rates that are otherwise difficult to access. We find that the unbinding rate of singly bound crosslinkers is nearly two orders of magnitude slower than for doubly bound ones. We attribute the increased unbinding rate of doubly bound crosslinkers to the high stiffness of biopolymers, which frustrates crosslinker binding.