A multi-layer microfluidic device for cardiac fibrosis model with spatially and temporally controlled cardiomyocyte-fibroblast interactions

Introduction:: Cardiomyocytes (CMs) and fibroblast cells are two essential elements in cardiac tissue structure and function. They can communicate indirectly or directly through extracellular matrix interactions, paracrine factors, gap junctions and electrical modulators to regulate the normal cardiac function as well as the process in heart disease, such as cardiac fibrosis or arrhythmias. Cardiac fibrosis, which is often a consequence of myocardial infarction, is characterized by a replacement of dead cardiomyocytes with a collagen-based scar that might contribute to contractile and diastolic dysfunctions. After MI, the formation of cardiac fibrosis is a highly regulated process in which the precise spatial and temporal release of soluble cues, such as specific hormones, growth factors and cytokines, inducing the fibroblast proliferation, migration into injured area, and transformation into myofibroblasts. Despite its prevalence, the pathogenesis of the cardiomyocyte-fibroblast interactions in the cardiac fibrosis is still not clear, and there is a lack of appropriate in vitro model that fully mimic the structural and functional characteristics of human cardiac fibrosis. The development of in vitro cell models can provide a controllable and flexible platform with relatively low cost compared to animal models, thus enabling the new possibilities for disease model, drug discovery and regenerative medicine. Here, we developed a versatile microfluidic device for a cardiac fibrosis model with spatially and temporally controlled release of soluble cues to a targeted scar tissue, thus offering a promising platform for further investigation into this disease.


Materials and Methods:: We used human induced pluripotent stem cells (hiPSCs) derived cardiomyocytes (iCMs) and hiPSCs derived cardiac fibroblasts (iCFs) as the model cell source. To create the multi-layer microfluidic device, each layer was firstly printed by a wax printer, casted by polydimethylsiloxane (PDMS) and bonded together after air plasma treatment. Then, the mixture of iCM and iCF-laden collagen hydrogel was seeded into the four chambers in the microfluidic device, mimicking the normal myocardial tissue. The iCF-laden collagen was seeded into the connections as scar tissue. After seeding, the different soluble cues were released into targeted iCF-laden collagen using a pneumatic valve to control the on-and-off of the solution flow. The flow pattern within the device was simulated by finite-element method and examined using fluorescent particles.


Results, Conclusions, and Discussions:: This study introduces a microfluidic device for a cardiac fibrosis model that enables the cardiomyocyte-fibroblast interactions between the co-culture of healthy cardiac tissue and scar tissue. After seeding, the device facilitates the spatial control of the releasing of soluble cues to target scar tissues via the flow in the introducing channel, as well as the temporal control of the releasing by the on-and-off of the pneumatic valves. The schematic of the microfluidic device was shown in Fig. 1A, and the 3D image of each layer was shown in Fig. 1B. Briefly, the microfluidic device consists of three layers: the bottom cell chamber layer, the middle introducing layer, and the top valve control layer. The bottom cell-chamber layer includes four interconnected chambers for healthy cardiac tissues and four connection sections for scar tissues. The healthy cardiac tissue and the scar tissue were seeded in two steps and separated by the physical barriers between them(Fig. 1D), and the resulting separate cell patterns are shown in Fig. 1E using two different cell-trackers. The middle layer introduced the soluble cues to the target scar tissue, while the top layer pneumatically controls the opening and closing of valves above each scar tissue. Therefore, the releasing condition for treating a specific scar tissue will the opening of both the related introducing channel and valve control channel. The table showing the opening and closing of each channel for the selective releasing is shown in Fig. 1F. The mechanism of valve control and the selective releasing was shown in Fig. 1G and Fig. 1H. When the valve control channel is at atmospheric pressure, the valve is in the open state, allowing for the flow in the introducing channel to be directed towards the targeted scar tissue. Conversely, increasing the pressure in the valve channel causes the valve to generate displacement, leading to the closure of the introducing channel. Simulation results of the valve displacements under different pressure conditions are shown in Fig. 1I.


Acknowledgements (Optional): :

References (Optional): :