Abstract
Deep vein thrombosis (DVT) is a serious health problem due to the formation of a blood clot (thrombus) in the sinus valves of the legs, and above all for its fatal complication: pulmonary embolism when the thrombus ruptures and passes into the pulmonary arteries. It happens more frequently with forced immobility suggesting a mechanical contribution. Even though the haemodynamics of the endothelium in the valve sinus may play a decisive role in activating the coagulation cascade, its contribution to the onset of this pathology at the cellular level is still poorly understood. In the first part of this manuscript, I present a venous valve in 3D imaged using non destructive Micro-CT X-ray scanning and nuclear magnetic resonance approaches of an ex vivo superficial vein obtained after vascular surgery at the Centre Hospitalier Universitaire (CHU)in Nîmes. I obtainimportant information on valve leaflet thickness and lengh, anchoring point and dynamics that i will use next to develop in vitro models close to physiology. The second part of this dissertation aims to explore the thrombotic behavior of a confluent layer of endothelial cells (ECs) under uniaxial stress mimicking a sinus environment without flow. For this i use a commercially available cell stretcher which consists of a deformable plastic chamber made of polydimethylsiloxane (PDMS) and human umbilical vein endothelial cells (HUVECs). I monitor the anti-thrombotic phenotype of ECs by measuring the expression of the anticoagulant protein Thrombomodulin which is involved in the activation of protein C, when it is bound to thrombin, a blood enzyme. The second proxy I consider is the von Willebrand factor, which is a procoagulant factor released by endothelial cells from Weibel-Palade granules. I explored different conditions such as the role of topography in static conditions using flat versus nanopatterned chamber surfaces. I studied the role of time scale: short time (6h) versus long time (24h) of cyclic stretch, the role of magnitude of stretch at 5% versus 10% and finally discontinuous stretching: 24h stretching of 10% followed by a 6h discharge in static condition to mimic the static pathological state of the onset of DVT. My main results are that ECs are sensitive to stretch. When long continuous stretch is applied, they become hypertrophic, move orthogonal to stretch, express the transcription factor ERG that controls vascular homeostasis and are antithrombotic. When discontinuous stretch is applied, cells become pro-thrombotic, mimicking the pathology. The next part of the thesis is dedicated to combining the effect of stretch and flow constraints with the aim of to mimic the mechanical conditions existing in the valvular sinus. For this, I developed 2 microfluidic devices. One is a channel recovery with a flexible membrane that deforms with flow inducing a stretch on cells while the surrounding cells in the channel feel only flow. The second one is made of a thin-walled tube that can increase in diameter with pressure. Cells grown inside the tube are thus submitted to both uniaxial stretch and flow that can be applied independently. I present the advantages and problems of each device. As a conclusion, this work that was done as a part of the Consortium DYV-MTEV (Dynamique des sacs Valvulaire et complications aiguës et chronique de la Maladie Thrombo Embolique Veineuse) in collaboration with the CHU of Nimes and Institut Montpelliérain Alexander Grothendieck, brings new tools and insights on the role of mechanics coupled to haemodynamics on the onset of thrombosis in vein.