(726c) Engineered Arterial Mimics (EAMs) to Quantify Smooth Muscle Cell Contribution to Atherosclerosis

For over a century, coronary heart disease (CHD) has been the leading cause of death in the Western world, surpassing all forms of cancer combined. The primary cause of CHD, atherosclerosis, is an extensively studied, yet incompletely understood disease. We hypothesize that one reason this disease is poorly understood, as well as why so few drug targets discovered in vitro are successful in vivo, is that the conditions in which cells are cultured and studied in vitro do not capture the complex in vivo microenvironment. To address this issue, we have created Engineered Arterial Mimics (EAMs) from PEG-PC gels, a hydrogel formed from a PEGDMA (poly(ethylene) glycol dimethacrylate) backbone with zwitterionic phosphorylcholine (PC) pendant groups, and which include integrin-binding extracellular matrix (ECM) proteins found in healthy and atherosclerotic arteries.

PEG-PC gels offer key advantages over PEG hydrogels. First, the Young’s modulus can be tuned over a wider range by allowing gel formation at very low PEGDMA concentrations (~2 kPa, 5 kPa, 50 kPa, 200 kPa and 500 kPa at 0.5, 1, 3, 6 and 10% PEGDMA and 20% PC). Second, they do not become opaque at higher Young’s moduli, and finally, the zwitterionic PC group makes these gels more hydrophilic than PEGDMA alone and they can, therefore, more efficiently block non-specific protein adsorption.  We are exploiting these features to create EAMs that represent the chemical and physical changes that occur during the progression of atherosclerosis.  We have EAMs that consists of collagen III, collagen IV and laminin (70/15/15 weight % at 10 ug/cm² total protein), which are predominantly found in healthy arteries, and EAMs that consist of fibronectin and collagen I (50/50 weight % at 10 ug/cm² total protein), which dominate the atherosclerotic artery. In order to capture the reported mechanical heterogeneity of diseased arteries, EAMs are made over a wide range of Young’s moduli: ~5, 50 and 200 kPa (1, 3, 6% PEGDMA and 20% PC, respectively).  

Preliminary migration experiments on EAMs varying only integrin-binding proteins found that human aortic smooth muscle cells (SMCs) migrate faster on the diseased ECM than on the healthy ECM (20 µm/hr vs. 15 µm/hr), indicating that the adhesive ligands available to the cells are able to tip the ‘phenotypic balance.’ We are currently performing migration experiments on PEG-PC gels to inform us of how the integrin-binding and stiffness cues integrate to affect this particular phenotype indicator. Preliminary results indicate that calponin, an early SMC differentiation marker, has a biphasic dependence on substrate stiffness (relative expression on 5, 50 and 200 kPa PEG-PC gels: 1.0, 1.6, 1.2), whereas phosphorylation of the 20 kDa myosin light chain (pMLC­₂₀) is also biphasic but inverted with respect to calponin (relative expression on 5,50 and 200 kPa PEG-PC gels: 1.0, 0.69, 0.94).  We quantified SMC proliferation over the course of a week on the EAMs in either serum-free DMEM, DMEM containing smooth muscle growth supplement (SMGS; 5% FBS, 2 ng/mL basic fibroblast growth factor (bFGF), 0.5 ng/mL epidermal growth factor, 5 ng/mL heparin, 5 ug/mL insulin) or a defined media condition (serum-free, 25 ng/mL platelet-derived growth factor, and 10 ng/mL bFGF).  We found that proliferation increases with stiffness in the no-serum and defined media conditions, proliferation is higher on the diseased EAM than the healthy EAM, and proliferation in the SMGS condition is high on all EAM and stiffness conditions. The latter observation suggests that serum and/or various mitogens can override the effect of ECM ligands and stiffness on phenotype maintenance.

We are continuing to investigate the effect of substrate stiffness and ECM ligands on proliferation, migration and SMC marker expression. In particular, we are interested in the interplay between SMC mechanosensing (integrin binding, cytoskeleton assembly and acto-myosin contractility), growth factor sensitivity and SMC phenotype. If we can better understand how these factors are related in a system, which closely resembles the in vivo environment, there is a greater potential to make discoveries that will lead to novel drug targets and treatments.