Forces and flows in cells and tissues. Blebs, active gels, and collective cell migration
- Autores: Ricard Alert Zenón
- Directores de la Tesis: Jaume Casademunt i Viader (dir. tes.), Giancarlo Franzese (tut. tes.)
- Lectura: En la Universitat de Barcelona ( España ) en 2018
- Idioma: español
- Tribunal Calificador de la Tesis: Jean-Francois Joanny (presid.), Timo Betz (secret.), Madan Rao (voc.)
- Programa de doctorado: Programa de Doctorado en Física por la Universidad de Barcelona
- Materias:
- Enlaces
- Tesis en acceso abierto en: Dipòsit Digital de la UB TDX
- Resumen
In this thesis, we have studied mechanical aspects of some biological processes in cells and tissues, which are addressed by developing theoretical models based on the physics of soft active matter. The thesis contains three parts, which focus on different biological systems.
In Part I, we study the adhesion between the plasma membrane and the actin cortex of eukaryotic cells. Controlling the adhesion between these two main structural elements of the cell is crucial for several cellular processes, such as cytokinesis and motility. Usually, the membrane and the cortex are adhered by means of a large number of specific linker proteins. However, in some circumstances, the membrane detaches from the cortex and it is inflated by intracellular pressure, forming a balloon-like protrusion called cellular bleb. Blebs appear in different contexts, but they are primarily used for amoeboid cell motility.
In this thesis, we propose a continuum model for membrane-cortex adhesion that couples the mechanics and hydrodynamics of the membrane to the force-dependent binding kinetics of the linker proteins. From it, we predict the critical pressure difference that causes membrane-cortex detachment. Both the critical pressure and the adhesion energy depend on the cortical tension. Then, we discuss published micropipette experiments, from which we infer the cortical tension of Dictyostelium discoideum cells. Then, we study the fluctuations of an adhered membrane, and suggest ways in which our predictions could be employed to probe membrane-cortex adhesion in fluctuation spectroscopy experiments.
Then, we employ the proposed model to study bleb nucleation. We show that bleb nucleation is governed by membrane peeling, the fracture propagation process whereby adjacent membrane-cortex bonds break sequentially. Through this mechanism, bleb nucleation is not determined by the energy of a local detachment like in the classical nucleation picture, but rather by the kinetics of membrane-cortex linkers. Hence, we conclude that bleb nucleation requires a kinetic description that goes beyond the possibilities of classical nucleation theory. We predict the critical radius for bleb nucleation through membrane peeling and the corresponding effective energy barrier. We estimate that these quantities are typically smaller than those predicted by classical nucleation theory, which implies a much faster nucleation. Finally, we simulate a fluctuating adhered membrane to obtain the probability distribution of bleb nucleation times.
In Part II, we study the dynamics of active polar gels. Active gels are soft materials, usually transiently-crosslinked polymeric networks, that are maintained out of equilibrium by internal processes that continuously transduce energy. Moreover, the constituents of these materials are usually anisotropic, and even polar, so that they may form orientationally ordered phases such as liquid crystals. The paradigmatic example of an active polar gel is the cellular cytoskeleton, which is composed by a dynamic network of polar polymeric filaments. Many biological systems such as the actin cytoskeleton, the mitotic spindle, and epithelial tissues have been described by means of the hydrodynamic equations of active polar gels. The constitutive equations of active polar gels were initially derived within the framework of irreversible thermodynamics, which introduces a number of phenomenological transport coefficients. Thus, the relationship between the transport coefficients that describe the macroscopic behaviour of active gels and the properties of their microscopic constituents is largely unknown.
In this thesis, we derive the constitutive equations of an active polar gel from a mesoscopic model for the dynamics of the molecules that crosslink the polar elements of the system. This way, we establish a connection between the molecular properties and the macroscopic behaviour of active polar gels. Specifically, we explicitly obtain the transport coefficients in terms of molecular parameters. In particular, these relationships show how the kinetics of the molecular crosslinkers induces the fluidization of the otherwise elastic material. Similarly, breaking detailed balance for the molecular kinetics gives rise to active stresses. Moreover, we predict that all transport coefficients have an active contribution that stems from the broken detailed balance. In particular, for the cell cortex, this contribution could yield a decrease of viscosity with activity — a phenomenon that we call active thinning, and which could explain some experimental results on the rheology of the cortex.
Finally, in Part III we study cell colonies and tissues, focusing in collective cell migration and tissue morphology. In many biological contexts, cells migrate within a group, even as a tissue. To migrate, cells must polarize and exert forces on the environment and on their neighbouring cells, namely traction and intercellular forces, respectively. During collective migration, cell-cell interactions give rise to several emerging dynamical phenomena, such as cooperative migration, vortical flows, multicellular protrusions at the leading edge of a migrating cell sheet, propagating mechanical waves, etc., which are the focus of this part.
First, we propose a particle-based description of cell colonies to study how the different organizations of cells in tissues emerge from intercellular interactions. The model intends to capture generic cellular behaviours such as cell migration, adhesion, and cell-cell overlapping. In addition, it models the so-called contact inhibition of locomotion (CIL), which repolarizes cell migration away from cell-cell contacts, as a torque on the migration direction. We show how CIL yields an effective repulsion between cells, which allows to predict transitions between non-cohesive, cohesive, and 3D tissues. In simulations performed by collaborators, we identify several structures and collective dynamics observed in different types of tissues. For example, we find gas-like states, regular distributions of cells, dynamic clusters, gel-like networks, collectively migrating cell monolayers, and 3D aggregates. Then, we discuss published experimental results from the perspective of our results, which allows us to associate different tissue types with the different states predicted by the model. In general, we conclude that, at low cell-cell adhesion, CIL hinders the formation of cohesive tissues. Yet, when cell-cell adhesion is sufficiently large to enable the formation of continuous cell monolayers, CIL gives rise to self-organized collective motion, ensures tensile stresses in the monolayer, and opposes cell extrusion, thereby hindering the collapse of the monolayer into a 3D aggregate.
Then, we focus on the spreading of epithelial monolayers on synthetic substrates, which is a model experiment to study collective cell migration. At the time scales of the expansion, the tissue is essentially fluid, and it spreads due to the traction forces exerted by the cells at the monolayer edge, which polarize towards free space. Thus, we address the spreading process by means of a continuum model based on the theory of active polar gels. In the following, theory is combined with traction force microscopy experiments performed by collaborators.
First, we concentrate on the wetting transition of epithelial tissues, which separates monolayer spreading from retraction towards a 3D aggregate — namely the equivalent of a fluid droplet. In the experiments, an epithelial monolayer is adhered to a region of the substrate. Then, we induce the expression of E-cadherin, which is a protein that mediates cell-cell adhesion. This expression gives rise to an increase of cellular forces that finally induces monolayer dewetting. The model predicts traction and tension profiles. By fitting the predictions to experimental data, we infer the evolution of the values of some parameters of the model. Moreover, the model predicts a non-monotonous flow profile in the monolayer, which tends to accumulate cells close to the ege. This accumulation could promote cell extrusion and the formation of 3D cell rims at the edge of epithelial monolayers as observed in some experiments.
We further predict the spreading parameter, whose change of sign indicates the wetting transition. The model predicts that the spreading parameter depends on the monolayer size, which implies the existence of a critical radius for the wetting transition. Cell monolayers larger than this size, which depends on model parameters, wet the substrate, whereas smaller monolayers form droplet-like aggregates. The existence of a critical radius for the wetting transition in tissues has no counterpart in the classical wetting transition, and it stems from the fact that the transition arise from competition between contact and bulk active cellular forces — traction forces and tissue contractility, respectively. To test these predictions, we performed experiments with monolayers of different sizes, which verify the existence of a critical radius. Altogether, these results show how the wetting properties of tissues emerge from active cellular forces, evidencing that the wetting transition has an active nature, with fundamentally different features from its classical counterpart.
Then, we study the morphological stability of the front of a spreading monolayer. The model predicts that traction forces cause a long-wavelength instability of the monolayer front, whereas tissue contractility has a stabilizing effect. Traction forces induce long-range viscous flows in the tissue, which are responsible for the fact that the most unstable mode has a finite wavelength, given by the smallest between the tissue width and the hydrodynamic screening length above which friction dominates over viscous effects. Thus, the predicted instability can explain the formation of finger-like multicellular protrusions observed during epithelial spreading. Therefore, we conclude that, despite their role in the collective dynamics of the well-developed fingers, leader cells observed in experiments are not required for the instability. Similarly, as suggested in other theoretical approaches, couplings of cell motility to the curvature of the front or to external fields such as biochemical signals are also not essential for the instability, which can have a purely hydrodynamic origin.
Last, the predicted morphological instability can explain the symmetry breaking of tissue shape observed during monolayer dewetting in the experiments of our collaborators. By fitting the predictions of the linear stability analysis to experimental data, we infer the monolayer viscosity, which increases with the concentration of E-cadherin, suggesting that tissue viscosity is primarily due to intercellular adhesion. Fitting the structure factor of the instability, we also infer the noise intensity of tissue shape fluctuations, which increases with tissue size, suggesting that it is due to traction forces and, hence, that it has an active origin.
