Cardiac Cell Therapy 3.0: The Beginning of the End or The End of the Beginning?

Cardiac cell therapy (CCT) holds great promise as a regenerative medicine approach for the treatment of cardiovascular diseases (CVDs)¹. The first generation of CCTs tested various adult cell types, including skeletal myoblasts, bone marrow (BM)-derived mesenchymal stem cells (MSCs), and cardiac progenitor cells (CPCs). More recently, the advent of induced pluripotent stem cells (PSCs) led to the much-anticipated second generation of CCTs with bona fide, PSC-derived CPCs and cardiomyocytes¹. The bad news is that, so far, both adult and PSC-based CCTs have failed to meet their promise of directly remuscularizing and repairing the heart to a therapeutically meaningful extent^{2, 3}. The good news is that some cell types clearly demonstrate encouraging results in terms of efficacy and safety and, more importantly, reveal a previously underestimated key role of CCT: to indirectly promote repair by regulating mechanisms of endogenous cardiac regeneration in the host^{1, 4}.

Understandably, the increasingly high burden of CVDs, coupled with the limited efficacy seen in both adult and PSC-based CCTs, and incomplete mechanistic understanding of adult human heart regeneration, have fueled disappointment, skepticism and polarized the field⁵. This schism has been particularly apparent in the area of adult CCTs, which also faces a current crisis of scientific distrust⁵. However, the interpretation that a possible stumble in research progress is proof that CCT is “broken”, would be unscientific. As Daniel Wegner noted “…tipping the balance toward skepticism can eradicate ideas faster than we can generate them. Eventually, we arrive at a vacuous chasm, with no theory standing and no idea left without serious wounds”⁶.

Under this prism, it is worth exploring how the field of adult CCTs fares, compared to other regenerative medicine approaches. PSC-based CCTs offer perhaps the strongest argument against adult CCTs, due to their unsurpassed ability to proliferate and differentiate into cardiomyocytes⁵. The idea that such a trait is the premise of CCT is based on experiments in zebrafish and newborn mice, both of which retain full capacity to regenerate a resected heart, possibly via cardiomyocyte amplification-based remuscularization mechanisms¹. However, experiments in more clinically relevant CVD models indicate that PSC-based direct remuscularization approaches exert effects that are more “cosmetic” than “regenerative” in nature, since cardiomyocyte engraftment is not accompanied by scar resorption and regeneration^{3, 4}. Moreover, both adult and PSC-based CCTs produce comparable improvements in cardiac function⁷, likely indirectly, via paracrine stimulation of endogenous repair mechanisms in the host⁴. Similarly, although gene-editing approaches offer hope for elucidating the genetic basis of CVDs, their potential application as regenerative therapy is currently limited. In addition to the technical challenges with safely and efficiently gene-editing billions of cardiomyocytes in-vivo, CVDs are molecularly complex, rather than of monogenic etiology⁸. Likewise, the molecular mechanisms of cardiomyogenesis entail precise, spatiotemporal modulation of multiple signaling gradients in both cardiomyogenic and non-cardiomyogenic cells, and therefore the possibility of developing cell-free, drug-based approaches to recapitulate such complex and dynamic processes in-vivo is currently limited¹.

In this issue of Circulation Research, Monsanto et al.⁹ lend support to a promising strategy to address the limitations of cardiac regenerative approaches by engineering combinatorial CCTs. This idea stands on two pillars: (i) no single cell population can produce all cell types that make up the human heart; and (ii) both cardiomyogenic and non-cardiomyogenic cells are essential for heart development and repair. Thus, engineering adult and/or PSC-derived cell combinations with complementary roles may more efficiently regulate endogenous regenerative pathways, compared to conventional CCT (Figure)¹. For example, the observation that BM-MSC therapy stimulates endogenous CPCs^{10, 11} led to the idea of combining the two adult cell types for greater, synergistic effects. Indeed, this hypothesis has produced encouraging results in several large and small-animal studies of CVD¹, and is currently in a phase II, randomized, placebo-controlled trial in ischemic cardiomyopathy patients (NCT02501811). Similarly, the combination of human PSC-derived cardiomyocytes with vascular cells¹² or MSCs¹³ produces further improvements in heart repair compared to cardiomyocytes alone, likely due to enhanced stimulation of endogenous repair mechanisms. The new method by Monsanto et al., to derive three distinct cardiac stem cell types from within the adult human heart, could potentially foster such applications⁹.

FIGURE. Combinatorial CCTs for heart regeneration.

Although “regeneration“ and “remuscularization” are thought of as synonymous in cardiac regenerative medicine, adult and PSC-based CCT trials unveil regenerative barriers unlikely to be circumvented by remuscularization alone. Synergism between complementary cell types, in the form of combinatorial CCTs, is a promising strategy for therapeutically targeting endogenous cardiac regeneration roadblocks. MSCs, mesenchymal stem cells; CPCs, cardiac progenitors; ECM, extracellular matrix; CM, cardiomyogenic cells; VC, vasculogenic cells; NC, neurogenic cells.

Using the cell-surface receptor cKit, both as a positive and negative selection marker, the authors devised a strategy to purify concurrently MSCs, CPCs, and endothelial progenitor cells (EPCs) from adult heart biopsies obtained during cardiac surgery⁹. MSCs are the most abundant derivative, comprising ~90–95% of the cardiac stem cell pool, and are purified as the CD105⁺/CD90⁺ fraction of cKit-negative cardiac cells. Consistent with previous reports¹⁴, cardiac MSCs exhibit a fibroblastoid morphology, produce colony-forming units-fibroblast (CFU-Fs), and exhibit multilineage differentiation into adipocytes, chondrocytes, and osteocytes⁹. However, compared to BM-MSCs, cardiac MSCs exhibit slow in-vitro growth kinetics and express cardiac lineage-markers, such as GATA4 and smooth muscle actin. Immunologically, expression of major histocompatibility complexes (MHC) class I and II, and co-stimulatory molecules CD80 and CD86, are similar to BM-MSCs, but cardiac MSCs express higher levels of the co-stimulatory molecule CD40. It is, therefore, unclear whether cardiac MSCs are as immunoprivileged as BM-MSCs. Such differences, however, are not surprising since mouse studies indicate distinct identities for BM and cardiac MSCs, with the latter possibly representing postnatal epicardial progenitors¹⁴.

The use of cKit as a CPC marker has been controversial⁵. Recent studies identify at least 2 distinct cell types expressing cKit in the heart: a rare, cardiomyogenic cell likely of neural crest lineage, and a more abundant vasculogenic cell, possibly of mesodermal lineage¹⁵. The work by Monsanto et al. further supports these findings. Positive selection for cKit yields two stem cell types with distinct immunophenotypic and gene-expression profiles⁹. cKit⁺ EPCs are morphologically round and committed to vascular fates, as indicated by high angiogenic potential in a Matrigel-based ex-vivo angiogenesis assay and expression of CD133 and PECAM1. cKit⁺ CPCs exhibit spindle-like morphology and a more myogenic profile, as indicated by lack of PECAM1 and relatively higher expression of GATA4 and smooth muscle actin. However, whether CPCs retain cardiomyogenic capacity is not demonstrated. Importantly, gene-expression profiling reveals striking differences between the 3 cardiac stem cell types in cytokines and extracellular matrix genes, such as SDF1, NRG1, FGF2, TIMP1 and MMP1.

The study by Monsanto and colleagues is an important advance in cardiac regenerative medicine. First, it is a bold demonstration of cellular plasticity retained in the human heart, regardless of age, gender or health condition. Stem cells were isolated from patients up to 84-years old and suffering from a range of diseases, including diabetes and coronary artery disease. Second, the ability to isolate 3 stem cell types from a single heart biopsy allows us to gain insight into the cellular composition in the adult human heart and the potential role of these unique cell types in CVD and regeneration. Since ~70% of human heart cells are non-cardiomyocytes, thorough research of their nature should be at the forefront of cardiac regenerative medicine¹. For example, Monsanto et al. noted that some cultures failed to yield all 3 stem cell types, a finding which merits further investigation for any potential relationship to disease mechanisms⁹. Last, the method of Monsanto et al. enables the isolation and expansion of therapeutic volumes of cardiac MSCs, CPCs, and EPCs from a single biopsy with 80–90% success (~100 million cells of each type could be manufactured in ~10 passages). Such technology is expected to be important for engineering combinatorial CCTs, using adult and/or PSC-based combinations^{1, 12, 13}, in a manner that effectively eliminates barriers to endogenous cardiac regeneration and may eventually lead to a much-needed scientific breakthrough for the treatment of CVDs.