Biophysical Reviews and Letters

This special issue of Biophysical Reviews and Letters describes recent advances in the area of active biomimetic systems, which are inspired by the cytoskeletal architecture found in all eukaryotic cells. The main building blocks of these systems are provided by two types of cytoskeletal filaments, F-actin and microtubules, as well as molecular stepping motors such as kinesins and myosins. All of these building blocks represent molecular machines: They are coupled to nucleotide hydrolysis and are able to convert the chemical energy released from this process into mechanical work. Bundles of filaments and teams of stepping motors generate strong pushing forces and perform long-ranged cargo transport.

Cell motility, one of the modular activities of living cells, elicits the response of the cell to extra-cellular signals, to move directionally, feed, divide or transport materials. The combined actions of molecular motors and re-modeling of the cytoskeleton generate forces and movement. Here we describe mechanistic approaches of force and movement produced by site-directed assembly of actin filaments. The insight derived from a biochemical analysis of the protein machineries involved in "actin-based motile processes" like cell protrusions, invaginations, organelle propulsion, is used to build reconstituted assays that mimic cellular processes, using several protein machineries known to initiate filament assembly by different mechanisms. Reconstitution of complex self-organized systems presents a broad variety of interests. Reconstituting actin-based movement of a functionalized particle from a minimum number of pure proteins, first used to prove the general thermodynamic principles at work in motility, then was the basis for fully controlled physical measurements of forces produced by polymerization of actin against an obstacle and of the mechanical properties of the resulting polymer arrays. In addition, measurements at the mesoscopic scale (trajectories, velocity, polymer mechanics, fluorescence of specifically labeled components of the actin array, use of mutated proteins) can provide further insight into the molecular mechanisms underlying motility.

The cytoskeleton is an actively regulated complex network in the cell. One of the most researched components is actin. In our work we developed and tested two microfluidic systems both being applicable to construct quasi 2-dimensional biomimetic actin networks. The first system uses polydimethylsiloxane micropillars, the other polystyrene microparticles held by holographic optical tweezers as anchoring points. Our devices provide actin networks with mesh sizes from a few micrometers up to the order of 10 micrometers. Qualitative analysis shows similar network formation in both systems. Crosslinking was tested using filamin, α-actinin, Ca and Mg ions. The crosslinking process is characterized by a zipping like event, which is limited only by the high stretching modulus of the actin filaments.

Dynamic microtubule bundles are involved in several motility processes in cells, for instance by interacting with mitotic chromosomes. The study of microtubule bundle dynamics and force generation can help understanding the mechanisms underlying these complex processes. Here we discuss experimental results on the force-generating capabilities of bundles of non-crosslinked growing microtubules obtained using an optical tweezers technique in vitro. We discuss the possible effects on force generation by these microtubule bundles of specific (with microtubule-associated proteins) and aspecific (with ions or crowding agents) ways of crosslinking microtubules. We present preliminary results showing that force generation by microtubule bundles is enhanced in the presence of polyethylene glycol used as a crowding agent, and discuss possible explanations for this observation.

The cellular cytoskeleton contains microtubules which function both as fundamental structural elements as well as motor protein tracks. While the structural property is connected to the properties of the tubulin dimer, its interactions with surrounding dimers and the geometric organization within the microtubule, the transport track properties are related to the interactions between the tubulin dimer and kinesin.

Based on the atomistic structures of kinesin and the tubulin dimer, we used molecular modeling to examine the interaction energy and force as function of a spatial distance of separation. From the results, elastic constants describing the system stiffness are obtained. By using the results related to the structure alone, a model of a 1 μm long microtubule is constructed as a network of elastic elements, and its mechanical properties were obtained via finite element method and compared to experimental results.

Concerning microtubule-kinesin complex, the interaction strength during a complete cycle of ATP hydrolysis was investigated. As expected, the affinity between the proteins is modulated by the type of nucleotide occupying the nucleotide binding pocket of the motor protein. The work underscores how molecular modeling can provide fundamental protein information in terms of the relation between mechanical properties and structural changes.

KIF5A and Eg5 are plus-end directed motor proteins with conserved motor domains. The catalytic cores of both motors comprise a central β-sheet consisting of eight β-strands surrounded by six α-helices. Notwithstanding the high level of similarity in their structural organization, Eg5 moves significantly slower than KIF5A. Recently, we reported that neck linker and neck elements of KIF5A and Eg5 contribute to velocity regulation. As the neck linker of both motors is known to be connected to the catalytic core via helix α6, the question arises if also helix α6 and strand β8 as the last core elements might be involved in velocity regulation. To elucidate the role these structures in kinesin activity generation we constructed KIF5A- and Eg5-based chimeras in which the β8 strand, helix α6, the neck linker, and the neck were interchanged. Additionally, we studied the role of α6 and β8 in ATP hydrolysis and microtubule binding by expression of truncated KIF5A and Eg5 constructs lacking both strand β8 and helix α6, or α6 only. The results obtained suggest that strand β8 and helix α6 are not involved in microtubule-binding, but α6 is an obligate and kinesin type-specific structure required to generate ATPase activity.

Living cells contain a large number of molecular motors that convert the chemical energy released from nucleotide hydrolysis into mechanical work. This review focusses on stepping motors that move along cytoskeletal filaments. The behavior of these motors involves three distinct nonequilibrium processes that cover a wide range of length and time scales: (i) Directed stepping of single motors bound to a filament; (ii) Composite motor walks of single motors consisting of directed stepping interrupted by diffusive motion; and (iii) Cooperative transport by teams of several motors. On the molecular scale, the energy conversion of these motors leads to single steps along the filaments with a step size of about 10 nm. The corresponding chemomechanical coupling is governed by several distinct motor cycles, which represent the dominant pathways for different values of nucleotide concentrations and load force. For the kinesin motor, the competition of two such cycles determines the stall force, at which the motor velocity vanishes and the motor reverses the direction of its motion. Because of thermal noise, the stepping motors unbind from the filaments after a certain run time and run length. For kinesin, the run time is about 1 s and the run length is about 1 μm for high ATP concentration and low load force. On length scales that are large compared to the run length, a single motor undergoes composite walks consisting of directed stepping interrupted by diffusive motion. The relative importance of bound and unbound motor states depends on the binding and unbinding rates of the motors. The effective transport velocity and diffusion coefficient of the motors are determined by the geometry of the compartments, in which the motors move. The effective diffusion coefficient can be enhanced by several orders of magnitude if the motors undergo active diffusion by interacting with certain filament patterns. In vivo, stepping motors are responsible for the transport of vesicles and other types of intracellular cargo particles that shuttle between the different cell compartments. This cargo transport is usually performed by teams of motors. If all motors belong to the same molecular species, the cooperative action of the motors leads to uni-directional transport with a strongly increased run length and to a characteristic force dependence of the velocity distributions. If two antagonistic species of motors pull on the cargo, they perform a stochastic tug-of-war, which is characterized by a subtle force balance between the two motor teams and leads to seven distinct patterns of uni- and bi-directional transport. So far, all experimental observations on bi-directional transport are consistent with such a tug-of-war. Finally, the traffic of interacting motors is also briefly discussed. Depending on their mutual interactions and the compartment geometry, the motors form various spatio-temporal patterns such as traffic jams, and undergo nonequilibrium phase transitions between such transport patterns.

Kinesins are motor proteins which convert the chemical energy of ATP into mechanical energy to move along proteinaceous microtubule rails and to transport different cargoes to defined intracellular destinations. It is well documented that following the track of a single protofilament is the thermodynamically most effective mechanism of kinesin movement along microtubules. However, the question arises what happens when a kinesin molecule encounters a hindrance along the protofilament. The present study describes a simple, cell-free approach which enables to study the effects of structural blockages on kinesin-based transport. This experimental approach uses dimeric conventional kinesin moving nanometre-sized gold beads along immobilized microtubules whose surface has been irreversibly decorated by blocking proteins. We demonstrated that the continuous bead transport temporarily stopped at sites of blockages, but usually continued after a certain resting time. Our results suggest that single dimeric kinesin molecules are able to change to another protofilament if the next tubulin dimer where the second head should bind is blocked. A bypassing mechanism is discussed which is considered to be one fundamental prerequisite to realize a kinesin-mediated cargo-transport along microtubules over long distances, required for e.g., the fast axonal transport in motor neurons.

Novel glass surfaces with quasi-hexagonally arranged gold nanoparticles to control motor protein immobilisation and motor protein dependent microtubule transport were applied. We first show that the kinesin-like motor protein Eg5 adsorbs efficiently and selectively to gold nanodots comprising a molecular motor nanopattern, while the glass surface between nanodots is passivated by a layer of polyethylene glycol in order to reduce non-specific interactions of individual motors with the substrate. We show that the motor nanopattern and density is indeed controlled by the gold nanodot density. We then use these motor protein arrays to investigate the kinetics of microtubule transport and find that the characteristics of the molecular motor nanopattern influence the characteristics of microtubule transport. This finding describes new biomimetic surfaces of molecularly controlled motor protein surface densities.

We present both a theoretical and an experimental study of the long time behavior of membrane nanotubes pulled from giant unilamellar vesicles by molecular motors. Experimentally, two types of behaviors are observed, either tubes stall at a finite length or they undergo periodic oscillations. Theoretically we write the equations for the tube dynamics as a two-dimensional dynamical system where the variables are the tube length (or the force required to pull the tube at a given length) and the number of motors at the tip pulling the tube. We construct stability diagrams showing the stalling and oscillating states and present an example of oscillations in a non-linear regime. These results can explain the membrane tube retractions and oscillations observed in living cells.

The motility of cells is a multifaceted and complicated cytoskeletal process. Significant inroads can be made into gaining a more detailed understanding, however, by focusing on the smaller, more simple subunits of the motile system in an effort to isolate the essential protein components necessary to perform a certain task. Identification of such functional modules has proven to be an effective means of working towards a comprehensive understanding of complex, interacting systems. By following a bottom-up approach in studying minimal actin-related sub-systems for keratocyte motility, we revealed several fundamentally important effects ranging from an estimation of the force generated by the polymerization of a single actin filament, to assembly dynamics and the production of force and tension of composite actin networks, to the contraction of actin networks or smaller bundled structures by the motor myosin II. While even motile keratocyte fragments represent a far more complex situation than the simple reconstituted systems presented here, clear parallels can be seen between in vivo cell motility and the idealized in vitro functional modules presented here, giving more weight to their continued focus.