Imaging and modelling to investigate the tissue mechanics during cell death in a simple epithelium

• Main Author: Kuipers, D.
• Published: University College London (University of London) 2012
• Subjects: 570
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.625890

Cell death within an epithelium could cause a loss of barrier function and damage to surrounding tissues. Mechanisms have therefore evolved to remove dead or dying cells, whilst maintaining epithelial integrity. This study has used monolayers of MDCK (Madin Darby Canine Kidney) cells as a model system to investigate these processes. It has been reported that shape changes in the surrounding cells, combined with junctional remodelling, are responsible for preserving the barrier function at sites of cell death. The aim was to identify and characterise the mechanisms that generate these shape changes; a particular focus was to establish the involvement of dying cells and their healthy neighbours in driving the epithelial repair. 3D time-lapse imaging was performed of UV-exposed MDCK cells, stably transfected with fluorescent reporters. This revealed distinct phases of actin and myosin activity at sites of cell death. To test the functional involvement of actin-myosin contraction, experiments were performed using a myosin inhibitor and also a Rho-defective cell line. The use of cells plated in expression mosaics made it possible to isolate the actin and myosin activity in dying cells and their neighbours. As an alternative means of inducing cell death, live imaging was also performed of cells killed via laser-ablation. The experimental results are complemented by a simple computational model, which simulates the forces acting at sites of cell death in MDCK monolayers. The results show that in UV-treated monolayers, the removal of apoptotic cells is a two stage process, the first of which has not previously been described. During this first stage, an actin ring within the dying cell constricts via actin-myosin contraction over a period of 15 to 30 minutes. This pulls the surrounding cells into a multi-cellular rosette in the apical plane, to cover the space that was occupied by the apoptotic cell. Simultaneously, the ring closure scissions its surface and encloses the cell body within the monolayer. The second stage of the repair process is driven by an actin ring in the surrounding cells, which moves towards the basal plane to extrude the cell remnants over 30 to 40 minutes. Crawling of the neighbour cells then heals the basal area before the membrane of the dying cell permeabilises. The computational modelling supports the experimental findings, predicting that the observed cellular shape changes must arise due to additional actin-myosin activity, and cannot arise passively due to network forces. In contrast to the UV-treated cells, those killed via laser ablation did not exhibit the first phase of the repair process. These necrotic cells remained in situ, constituting a leakage point while the neighbouring cells healed the wound and extruded the remnants. This identifies a difference in the removal of apoptotic and necrotic cells, since the active participation of the former allows a continuous monolayer to be maintained. This study reveals that epithelial repair cell due to apoptosis involves two phases, the first of which is driven by the dying cell and the second by its neighbours. In combination, this activity ensures that epithelial integrity is maintained and apoptotic cells are removed without damage to surrounding tissues.