Using the embryonic heart as an instructive template for cardiac tissue engineering

Abstract

Event Abstract Back to Event Using the embryonic heart as an instructive template for cardiac tissue engineering Ivan Batalov1, Quentin Jallerat2, Sean Kim2, 3 and Adam W. Feinberg1, 2 1 Carnegie Mellon University, Materials Science And Engineering, United States 2 Carnegie Mellon University, Biomedical Engineering, United States 3 Carnegie Mellon University, Chemical Engineering, United States Introduction: Cardiovascular disease is the leading cause of death worldwide and cardiac tissue engineering provides a potential approach to regenerate muscle and repair the heart. However, the formation of aligned myocardium capable of synchronous contractions remains a challenge. Previous work has shown that 20 µm wide lines of alternating high and low density fibronectin (FN) patterned onto substrates align cardiomyocyte (CM) monolayers[1]. However, while this pattern geometry works, we hypothesized that micropattern dimensions similar to FN filaments in embryonic heart tissue (0.2 to 2 µm) would improve CM alignment. To test this, we developed a biomimetic (BM) micropatterned surface and analyzed alignment of embryonic chick and human induced pluripotent stem cell derived (hiPS) CMs. Materials and Methods: Microcontact printing[2] was used to pattern FN onto polydimethylsiloxane-coated glass coverslips with (i) 20 µm wide, 20 µm spaced lines (20x20), (ii) 2 µm wide, 2 µm spaced lines (2x2), and (iii) a BM pattern derived from 3D-images of FN in chick embryonic hearts. Embryonic chick CMs or hiPS-CMs were seeded onto the different micropatterned substrates. After 3 days of incubation samples were stained for nuclei, actin, and α-actinin, and imaged using confocal microscopy. Images were analyzed using MATLAB to calculate the orientational order parameter (OOP) of actin filaments as an alignment metric, ranging from 0 for no alignment to 1 for perfect alignment. Results: We engineered confluent monolayers of chick CMs using all three patterns and found that while the OOPs were similar for all patterns, alignment was highest for the 2x2 pattern and lowest for the BM pattern (fig. 1a). Uniquely, we found that chick CM alignment on the BM pattern, but not the line patterns, was dependent upon cell density, increasing linearly with cell surface coverage (fig. 1b). This suggested that cell-cell interactions play an important role in cell alignment on the BM pattern. To test this we inhibited cell-cell interactions using N-cadherin blocking antibodies, which decreased cell alignment within a range of intermediate cell densities. We then engineered cardiac monolayers using hiPS-CMs to validate our results with a more clinically relevant CM type. Preliminary results demonstrated decreased alignment on the micropatterns relative to chick CMs, which we attributed to the immature phenotype of the hiPS-CMs (fig. 1c). Figure 1. a) Chick CM OOP in confluent monolayers, b) chick CM OOP in isolated cell cultures, c) iPSC-CM OOP in monolayers on FN patterns. * indicates statistically significant difference (p < 0.05) compared to the 20x20 control, # indicates statistically significant difference between low-density culture and high-density monolayer on the same FN pattern. Conclusions: We found that the alignment of CMs on the BM micropattern was dependent on cell substrate coverage, demonstrating a unique cell response compared to standard line micropatterns. This suggests that CM alignment on the BM micropattern is driven by a combination of cell-cell and cell-ECM interactions, and that the same process may occur in vivo. In the future we will test if human iPS-CMs show similar density-dependent behavior on the BM pattern and analyze functional characteristics of the human cardiac tissues. NIH Innovator Award; American Heart Association