Cell augmentation strategies for cardiac stem cell therapies

Abstract Myocardial infarction (MI) has been the primary cause of death in developed countries, resulting in a major psychological and financial burden for society. Current treatments for acute MI are directed toward rapid restoration of perfusion to limit damage to the myocardium, rather than promoting tissue regeneration and subsequent contractile function recovery. Regenerative cell therapies (CTs), in particular those using multipotent stem cells (SCs), are in the spotlight for treatment post‐MI. Unfortunately, the efficacy of CTs is somewhat limited by their poor long‐term viability, homing, and engraftment to the myocardium. In response, a range of novel SC‐based technologies are in development to provide additional cellular modalities, bringing CTs a step closer to the clinic. In this review, the current landscape of emerging CTs and their augmentation strategies for the treatment post‐MI are discussed. In doing so, we highlight recent advances in cell membrane reengineering via genetic modifications, recombinant protein immobilization, and the utilization of soft biomimetic scaffold interfaces.

regenerate the muscle integrity of the heart, 8 but clinical trials showed only minor improvements in ejection fraction performance and ventricular tachyarrhythmias. 12 These cell phenotypes were superseded by bone marrow cells (BMCs), 13,44 especially bone marrow-derived mesenchymal stromal cells (MSCs), 45 which exhibit desirable properties for therapy, such as immunomodulatory capacity, 46 a tendency to migrate to inflammation and injury sites, and multipotency. 47,48 Early preclinical studies suggested that MSCs could differentiate into a CM phenotype, 19,20,22,[49][50][51] making them promising candidates for heart tissue regeneration. However, MSCs transplantation in animal models post-MI, 19,20,22 and in clinical trials, showed only modest levels of recovery of heart function. 12,46,52 Indeed, later studies suggest that the observed therapeutic benefits result from paracrine effects 53,54 or acute immune responses, 21  Small mammal models Swine models Primate models Clinical trials

Myoblasts
Aut., a muscle regeneration 8 Aut., enhanced oxygenation, contractile function recovery 9 Al., b cell survival for 10 days. 10 Al., paracrine effects on ECM remodeling and vascularization. 11 -Aut., ventricular tachyarrhythmias 12 BMCs c Al., improved tissue regeneration 13 Aut., enhanced angiogenesis after a week 14 Aut., improved cardiac function, and higher blood flow and capillary function after 3 wk 15 Aut., improved regional blood flow and cardiac function via paracrine effects 16 Aut., improved infarct tissue perfusion and left ventricular function 17 Aut., decreased infarct size, improved left ventricular function 18 MSCs d Al., heart regeneration via differentiation into CMs 19 Al., myocardium repair via paracrine effects 20 Aut., improved ventricular performance via acute immune response 21 Aut., structural and functional remodeling 22 Al., angiogenesis, reduction of infarct size, improved contractile function via trilineage cell differentiation 23 -Aut. and al., enhanced ventricular remodeling and functional capacity 24 MSCs-CSCs e Aut., decreased infarct size, improved cardiac function via paracrine effects 25 Aut., scar size reduction 26 Al., scar size reduction and systolic function recovery 27 -Aut., undergoing 28 iPSCs-CMs f Aut., improved left ventricular function 29 Al., improved left ventricular function 30 Al., improved contractile function 31 Al., contractile function improvement 32 Al. 33 and Aut., 34 transplanted in cell sheets, undergoing hESC-CMs g Al., CMs survived and engrafted to the heart for weeks 35 for more information on other BMCs studied in cardiac repair.
MSCs are by no means the only SCs in the spotlight for cardiac tissue regeneration, as promising preclinical results have also been achieved using induced pluripotent cells (iPSCs) 61 and human embryonic stem cell-derived CMs (hESC-CMs). 39,40,62 iPSCs have been shown to be plausible candidates for cardiac therapy in several animal models. 61,63,64 Transplantation of iPSCs predifferentiated into CMs has been reported to ameliorate ventricular function in rodent models, 29,30 and contractile function in porcine models. 31 iPSC-CMs have also been developed as a potential allogenic CT and tested in nonhuman primates, displaying electrical integration with the host heart, leading to improvement of contractile function after 4 weeks with no significant immune rejection. 32 These encouraging preclinical results supported clinical trials to determine the safety to iPSC-CM transplantation as cell sheets. 33,34 In vivo CT studies to treat cardiac ischemic injury from hypoxia have postulated a wide range of possible regenerative mechanisms that are linked to cell type and origin (autologous or allogenic). These include functional integration with recipient CMs, activating the growth and differentiation of endogenous CSCs, 65 80 In particular, MSCs can come naturally to the bone marrow niche through binding of the extracellular C-X-C chemokine receptor type 4 (CXCR4) to the stromal cell-derived factor 1 (SDF-1), which is present in the bone marrow. 81,82 This interaction could be readily exploited for cardiac therapies, as SDF-1 expression is upregulated at the injury site over 48 hours post-MI, 80 but unfortunately the MSC expression of CXCR4 is commonly downregulated or lost during their expansion. 83

| Genetically modified SCs
Improving the therapeutic potential of a SC can be achieved genetically using viral or liposome-based vectors, 91  Overall, genetic modification is a versatile and exciting approach to enhance MSCs therapeutic performance and retention in the heart; however, the cost associated with reprogramming can be prohibitively high. Moreover, the modifications are permanent and the use of viral vectors is subject to insertional oncogenesis. 47

| Protein-based membrane modifications
Direct protein-based membrane modification strategies provide transient display of the targeting construct and present a number of potential benefits over generic approaches for improved cell homing.
These include, a reduction in risk arising from oncogenesis in the therapeutic cells, minimal impact on the cell manufacturing process as the modification step can be readily integrated into an existing therapeutic pipeline, the display number per cell can be systematically varied to reduce the risks associated with patient-specific expression levels, and protein production is scalable and can be produced using good manufacturing practice procedures. 103,104 An excellent example of direct protein-based membrane modification was demonstrated by Won et al, who displayed recombinantly produced CXCR4 on MSC membranes using lipid-PEG vesicles ( Figure 2A). 105 They demonstrated that CXCR4 was only present in the MSCs membrane after delivery by confocal microscopy studies without affecting the viability of MSCs and showed up to a twofold improvement in their migration toward SDF-1 following a concentration gradient in vitro. This noninvasive approach allowed facile reengineering of MSC membranes within 2 minutes, which is especially relevant for autologous therapies, considering that genetic modification methods can take several weeks, missing the ideal therapeutic window for treating the infarcted myocardium.
Besides the SDF-1/CXCR4 axis, homing to ECM adhesion proteins overexpressed in inflammation sites, such as selectin and fibronectin, has also been investigated. [106][107][108] There are several avenues to achieve this, one being the covalent conjugation of MSCs to E-selectin binding peptides in a two-step process ( Figure 2B). Armstrong et al. 113 What is becoming clear, however, is that despite the fact that the approaches lead to an increase in target affinity, there is still a lack of compelling preclinical data to support clinical translation. Even so, cell membrane reengineering using proteins is an exciting methodology to augment SCs and next-generation new protein ligation tools are emerging, 114,115 which will allow rapid bioorthogonal functionalization of SCs to instill a range of new properties and cellular functions.

| SOFT BIOMATERIALS FOR SC DELIVERY
Despite not being strictly a direct cell membrane modification approach, soft biomaterials can provide an ECM-like environment to the transplanted SCss, which can have a major impact on cell adhesion at an infarcted site, cell survival and retention in the myocardium, and hence the overall efficiency of the treatment. Accordingly, the application of injectable hydrogels, cell patches and cell sheets, and their respective performances are discussed below (Figure 3).

| Hydrogels as transplant matrices
The incorporation of injectable hydrogels in CTs has drawn much attention in the last decade, as they provide a biocompatible threedimensional matrix to the transplanted cells ( Figure 3A) and form the basis for the majority of bioinks used in 3D bioprinting. 116,117 Preclinical cardiac CT studies have utilized a range of biologically derived hydrogels from mammalian sources, such as collagen, and polysaccharides like hyaluronic acid (HA), as well as other naturederived examples, which include chitosan (from seafood industry waste) or alginate (from seaweed), which are biodegradable and have similar mechanical properties to the infarcted tissue. 118 Several key examples have shown that acellular injectable hydrogels of different compositions are safe for cardiac implantation in murine models. [119][120][121] These studies also showed an increase in endogenous BMCs homing to the heart after injection of HA hydrogels modified with recombinantly expressed SDF-1 119,120 or with collagen I hydrogels embedded with histone deacetylase 7 peptide. 121 In both cases, The scope of using hydrogels in cardiac therapies is not limited to acellular transplantations, as many efforts report successful delivery of embedded cells in hydrogels to infarcted tissue in animal models ( Figure 3A') [123][124][125][126] and clinical trials. 127 Hydrogels not only can be transplanted as matrices, but can also be directly injected and have been shown to protect the transplanted cells from the mechanical shear of injection 128 and increase the cell number and retention at the targeted tissue. 129 Early clinical trials in patients with ischemic injury involved injecting autologous BMCs embedded in a collagen I hydrogel, however, despite proving to be safe for the patients, no major improvement in heart function was observed. 127 New efforts have focused on improving the efficacy of these CTs by the addition of F I G U R E 3 Biomaterials can provide stem cells (SCs) with an ECM-like microenvironment to promote adhesion and retention in the myocardium and, ultimately, to enhance SC therapeutic outcome. A, Acellular hydrogels modified with specific factors (eg, SDF-1) recruit endogenous SCs after transplantation. A', Cellular hydrogels protect the implanted cells from mechanical stress from the injection. B, Cardiac patches offer the best short-term protection and retention, but they are more rigid than the other options and usually fail to couple electromechanically with the heart. C, Cell sheets can contain monolayers of single cell types or coculture of different types to contribute to different processes involved in cardiac repair signaling ligands (eg, Notch ligand delta-1) 123 and peptides derived from growth factors (eg, insulin-like growth factor 1, angiopoietin-1) 124,125 to enhance the survival of the transplanted cells, and by improving the mechanical properties of the gel. 126,130 A recent study has reported successful implementation of hESC-CMs in a collagen hydrogel modified with recombinant Notch ligand delta-1 in rats. 123 The Notch signaling ligand doubled both the proliferation rate of the implanted hESC-CMs and the graft size compared to the controls, even when the cells were transplanted in subtherapeutic numbers.
Another key study has shown that embedding the C-terminal

| Cardiac patches
Cardiac patches have emerged as a potential solution for the poor retention and survival of transplanted cells in the heart ( Figure 3B).
Here, SCs or SC-derived CMs, are grown in vitro and then adhered to a scaffold that suits the size of the injury, and that has a matrix that allows oxygen diffusion and resistance to contractile forces. Once the desired cell confluency is achieved on the patch, it is surgically implanted. The main limitation of this exciting approach is that is it difficult to integrate electromechanically and immunologically within the heart and the transplanted cells display low long-term survival in animal models. 134 For example, initial phase I clinical trials have reported short-and medium-term safety of transplanting hESCs in fibrin-based patches in six patients, but no information on the hESC survival rate or the patch electromechanical coupling was recorded. 41 Phase I clinical trials to determine the safety of hESC collagen patches have been completed in November 2020, but the outcome is yet to be published.
In an effort to overcome some of the limitations of cardiac patches, novel next-generation designs are emerging, which include cellulose nanofibers MSC patches to enhance neovascularization of infarcted myocardium in rats, 135

| Cell sheet technologies
Cell sheet technology is an attractive alternative to cardiac patches, as the resulting structures exhibit high cell concentration and uniformity, confer more resistance to degradation upon implantation, and only rely on the formation of tight cell-to-cell junctions and ECM protein secretion, rather than an artificial scaffold ( Figure 3C). 140 Cell sheets are prepared by culturing monolayers of cells on temperatureresponsive substrates, which become nonadherent at low temperatures, 141 and provides the opportunity to produce cellular multilayers through direct manipulation (up to three layers) or sequential assembly on a hydrogel-coated plunger (up to five layers). 142 Once the cell sheet is formed, it detaches from the substrate via a temperature change, which allows for efficient and effective surgical implementation. The approach has been used to transplant a wide range of cells to infarcted myocardium, including autologous myoblasts, 143 autologous skeletal cells, 144 allogenic cardiac progenitor cells 145 and allogenic iPSC-CM. 146 Overall, the approach has been shown to enhance cardiac regeneration in several animal models when compared with the direct injection of cell suspensions, possibly due to retention of higher cell numbers on the heart and the formation of tight cellular junctions within the sheets. and have a low risk of triggering unwanted immune responses. However, the risk of mutation-derived oncogenesis is still a concern, which paves the way for transient non-genetic SC modification approaches.
Another potential limitation is that current reprogramming methodologies are not temporally compatible with autologous CTs, as the expansion phase is lengthened by several weeks, missing the ideal therapeutic window after MI. Similarly, cardiac patches and cell sheets also require extended culture periods. This cell number challenge could be overcome with the development of an allogenic CT, giving rise to a readily accessible off-the-shelf treatment, which could be subjected to high quality control processes. 24,150 With respect to the developments within the biomaterial scaffold space, although they offer an effective solution to myocardium cell retention and cell number, the transplant process is generally more invasive, and challenges with effective electromechanical integration still remain. Shear thinning injectable hydrogels have great potential, as they are generally less invasive, offer protection to the transplanted cells from mechanical stress, and provide a rudimentary micro-ECM that can be systematically tuned for cell signaling. It is also worth highlighting efforts on transplantation of acellular hydrogels and cardiac patches that attract endogenous stem/progenitor cells and provide them with a scaffold to promote long-term survival. Moreover, these scaffolds could be modified or implemented in combination with molecules that activate the recruited cells (eg, statins or TGF-β/Wnt signaling molecules) and amplify their therapeutic potential. In conclusion, it is likely that no single approach to SC membrane reengineering will provide the "magic bullet" for cardiac CTs, and that the next generation of therapies will likely utilize combinations of these technologies to fully harness the therapeutic potential of transplanted SCs.

ACKNOWLEDGMENT
We thank UK Research and Innovation for support via the Future Leaders Fellowship MR/S016430/1.

CONFLICT OF INTEREST
The authors declared no potential conflicts of interest.

AUTHOR CONTRIBUTIONS
R.C.S.: conception and design, manuscript writing; M.J.: manuscript writing; A.P.: conception and design, manuscript writing, final approval of manuscript.

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.