Shear-exposed in vitro models mimicking healthy vascular tissue and intimal hyperplasia

Abstract

Cardiovascular diseases are the leading cause of death in developed countries, accounting for 31% of deaths per year worldwide. Cardiovascular devices, such as stents, vascular conduits and vascular assist-devices (VADs), are the lifelines for patients with severe cardiovascular disease, and have been deeply researched and improved in the last decades. However, these devices come with limitations and are associated with complications, such as thrombosis and stenosis. Since VADs and other implantable devices are not hemocompatibility, the foreign-body response activates inflammation and coagulation at the blood-interface, forcing patients to live under constant anti-coagulant therapy. The endothelium is a layer of endothelial cells (ECs) that contacts with blood in native vessels, and thus is a natural anti-coagulant surface. Proper endothelialization of implantable devices with a functional endothelium, in situ or in vitro, has been the holy grail of cardiovascular research. Moreover, implantation of stents or bypass vascular grafting are the current treatments for coronary artery disease. Vascular injury is often associated with these surgical procedures, which exposes the underlying smooth muscle cells (SMCs) in the tunica media to shear and blood components, initiating a process called neointimal hyperplasia. Intimal hyperplasia, as a result of SMC proliferation and extracellular matrix (ECM) deposition, can cause restenosis and thrombosis. Efficient reendothelialization after vascular injury is extremely important for the prevention of such complications. However, endothelialization does not depend solely on ECs, but also on their interactions with SMCs, as well as immune cells and the ECM, which are further regulated by mechanical stimuli, such as shear, all of which ensure proper endothelial function. Thus, considering this complex environment with its biochemical and biomechanical properties is extremely important when studying disease development or engineering biomimetic models. While in vivo models are associated with high-cost, high time-investment and limited translation to humans, the latest developments in in vitro models have enabled to closely mimic physiological conditions, revolutionizing the tissue engineering community and the understanding of disease, as well as development of new therapies. The experimental work presented in this thesis focuses on developing versatile in vitro biomimetic models of healthy and diseased vascular tissue, and the study of cell-cell interactions at the cellular and tissue level in the presence of hemodynamic conditions. Firstly, we developed protocols to coculture a layer of primary human SMCs topped with a confluent monolayer of primary human ECs and exposed to shear on an orbital shaker. This allowed for cell-cell communication through direct-contact and secreted factors. Protocols were further developed to control the expression of SMC phenotypic markers characteristic of the healthy vessel wall and to enhance the deposition of ECM proteins that are abundant in the tunica media. We have also identified conditions that optimize EC growth on top 6 of SMCs, giving us clues for how to improve reendothelialization after injury and how to better achieve stable co-cultures for endothelialization of cardiovascular devices such as VADs. Secondly, using the same setup and similar protocols, we developed an in vitro model of intimal hyperplasia. In this case, the ECs monolayer was defected (i.e. with several holes) and SMCs expressed the phenotypic markers characteristic of an injured vessel. As a result of EC-SMC communication under flow conditions, SMC layer reorganized to form pronounced hill-valley topography that resembled intimal hyperplasia. These corrugated structures, composed of SMC multilayers, were confined to regions of defected endothelium, and were a result of enhanced SMC proliferation and de novo ECM assembly. Underneath a confluent endothelium, however, the proliferation was low and SMC layer remained stable, suggesting an inhibitory effect by the EC layer. Confirming such hypothesis, when ECs were seeded at high density on top of synthetic SMCs, so that EC layer became confluent, the formation of corrugated structures by SMCs was inhibited. Moreover, we observed that ECs at low density formed capillary-like networks on top of synthetic SMCs, which over time became denser and stabilized as islands of confluent endothelium, which illustrates how ECs may grow during reendothelization. These results highlight how EC-SMC communication combined with shear exposure affect cell behavior at the tissue level and are highly significant to define better therapies to enhance reendothelialization after vascular injury and prevent intimal hyperplasia. Finally, a proof-of-concept experiment with Paclitaxel, a drug clinically used in drug-eluting stents, has demonstrated the potential of our models in such a simple platform for high throughput screening of new pharmacotherapies to treat intimal hyperplasia while enhancing reendothelization. This is highly significant from a clinical point of view because the existing in vitro models for intimal hyperplasia so far are too simplistic and pre-clinical tests in animals are too expensive and time-consuming. This platform and the developed models of healthy and diseased vasculature will not only accelerate the development and pre-selection of promising therapies but also help the finetuning of the correct dosage, while reducing animal sacrifice. Taken together, this thesis leads to a more complete comprehension of how EC-SMC communication and pulsatile shear play a role in intimal hyperplasia and reendothelialization after vascular injury. It also provides protocols and tools for the development of biomimetic vascular tissue and provides platforms for studying mechanisms of disease and performing drug screening for possible new therapies. This platform is extremely versatile and can be further modified, for example by using cells from diseased or aged patients, or by adding other cell types, cytokines or growth factors to simulate other scenarios of health or disease.