Contributed Talk

The Role of Intermediate Filaments in Stress Resistance in 3D Epithelial Structures

Tom Golde, Marco Pensalfini, Nimesh Chahare, Marino Arroyo, Xavier Trepat

Institute of Bioengineering of Catalonia (IBEC), Parc Científic de Barcelona (PCB), Baldiri Reixac 10-12 08028 Barcelona, Spain

Contact:  | Website

In vitro experiments on IF fibers and networks demonstrated that IFs are the cytoskeletal component with the highest stretchability while displaying a resilient strain stiffening behaviour. This led to the safety belt hypothesis, stating that IFs are protecting cells from large and rapid deformations. However, the typical experimental approaches for stretching epithelial tissues only allow maximum strains of around 30% due to limitations of the supporting base layer. To overcome these limitations, we developed a microfluidic device where an epithelial monolayer is grown on a porous surface with circular low adhesion zones. Upon applying hydrostatic pressure, the monolayer delaminates into a spherical cap (dome). This allows us to generate tissue strains of more than 100% while individual cells are stretched up to 900%. Furthermore, we can image these 3D epithelial domes with super resolution microscopy, determine the tissue tension via Laplace’s law, and control the rate of inflation and deflation.
Using this approach with MDCK monolayers, we observed a striking reorganization of the keratin IF rim-and-spoke network, where the filament network wrapping the nucleus forms a central knot with thick, radially oriented bundles. Similar transitions have recently been observed after depolymerizing actin with latrunculin B [1] and blocking the actin-IF linker plectin [2]. These findings indicate a crucial role of actin-IF interactions and are in accordance with our previous results of the actin cortex dilution in superstretched cells [3]. To better understand the mechanical principles of such transitions, we developed a multiscale computational model that simulates the interactions of keratin IFs with the nucleus, desmosomes, and the actin cortex. Combining experiments and simulations, we can now conclusively test the safety belt hypothesis in controlled and unparalleled large 3D tissue deformations.