3D bioprinted heart patches for cardiac regeneration

Abstract

BACKGROUND: epicardial patch transplantation is a promising approach to restore some of the cardiac function lost after myocardial infarction (MI). Advances in 3D bioprinting, 3D cell culture and transplantation methods at surgery have provided hope that this approach could soon benefit heart failure patients. The optimal content of 3D bioprinted patches (the “bioink” extruded by a 3D bioprinter) is not known. Patches containing a suspension of 3D vascularised cardiac spheroids (VCS; 3D aggregates of cells / microtissues) in hydrogel may confer an advantage compared to freely suspended cells or hydrogel without cells. The mechanisms underlying the benefit of epicardial patch transplantation approaches have not been fully elucidated and this is needed for widespread clinical translation. To be fully compatible with cardiothoracic surgical approaches in future, patches should be transplantable by minimally invasive robotic approaches. METHOD: Alginate-gelatin (AlgGel) patches were optimised ex vivo for cardiac applications, followed by in vivo transplantation of patches in mice modelling MI. For the ex vivo optimisation phase, three different bioprinters were used to bioprint patches with different bioink contents which were incubated up to 28 days and analysed. For the in vivo phase, new patches were 3D bioprinted using the optimal methods determined in the previous (ex vivo) experiments and surgically transplanted to the epicardium in infarcted mice. For these in vivo experiments, we cultured mixed cardiac cells: induced pluripotent stem cell derived cardiomyocytes (iCMs), human coronary artery endothelial cells (HCAECs) and cardiac fibroblasts (CFs). Cells were cultured using hanging drops to generate VCS which were suspended in AlgGel to create bioink for 3D bioprinting of patches. Study control groups (in vivo) were: the same cells freely suspended in AlgGel, AlgGel without cells, MI without treatment and sham surgery (no MI and no treatment). The primary outcome was cardiac function (left ventricular ejection fraction, LVEF%) measured up to day 28 post surgery. Additional analyses included: electrical mapping, histology, cell quantification by flow cytometry and mRNA (gene expression) profiling. Alongside these experiments, we developed novel surgical robotic minimally invasive instruments designed to transplant similar patches at human scale. We prototyped a heart patch transplanter device and demonstrated its potential utility in a world-first operation on a pig cadaver. RESULTS: Ex vivo patches incubated for 28 days allowed for self-organisation of endothelial cells into networks and contractile activity within patches. In vivo transplantation of patches in mice modelling MI resulted in a “return to baseline” improvement in median LVEF%. Our results measured median baseline (pre-surgery) LVEF% for all mice at 66%. Post-surgery, LVEF% was 58% for Sham (non-infarcted) and 41% for MI (no treatment) mice. Patch transplantation increased LVEF%: 55% (acellular; p=0.012), 59% (cells; p=0.106), 64% (spheroids; p=0.010). The VCS group was associated with improved electrical mapping profiles, lower infarct sizes, changes in host immune cell numbers and a gene expression (mRNA) profile which was closest to sham mice (with no MI). As proof-of-concept, similar scaled-up AlgGel patches were successfully transplanted in a porcine cadaver using a prototyped robotic minimally invasive surgical instrument. CONCLUSION: Epicardial transplantation of patches improves cardiac function in mice modelling MI. The use of VCS in alginate-gelatin bioink seems to offer advantages compared to freely suspended cells or hydrogel alone. The fact that hydrogel alone without cells confers some restoration of myocardial function suggests that the mechanism is not fully accounted for by the cellular portion of the bioink. Further studies are needed with a focus on whether host immune cell modulation is a key mechanism underlying the benefit of this approach. Since our most successful treatment group (VCS) had a similar transcriptome compared to non-infarcted (sham) mice, further studies should also include transcriptomic analyses to confirm reproducibility of this finding. If it is confirmed that immuno-genetic mechanisms underly patch-based approaches to myocardial protection after MI, this may change the focus of treatment strategies and avoid wasted resources and potentially patient harm (from treatments which are not aligned with the underlying mechanism). Our robotic minimally invasive patch transplantation operation represents a first step on a potential pathway towards transplantation at human surgery (without the need for traditional open surgery). For translatability, patch development should work towards being compatible with robotic and/or minimally invasive transplantation.