Induced Pluripotent Stem Cells: It’s like déjà vu all over again

Embryonic stem cells (ESCs) were first derived in the early 1980’s simultaneously by two independent groups^1;2 and because of their plasticity and potentially unlimited capacity for self-renewal, they were predicted to transform research in mammalian development, genetics, stem cell biology and regenerative medicine. However, it quickly became apparent that although ESCs would become an invaluable tool to study development and stem cell biology, translating their promise into the clinics has been problematic. Twenty-eight years after the first report describing mouse ESCs and ten years after the successful derivation of human ESCs, the U.S. Food and Drug Administration approved the first human ESC-related clinical trial for spinal cord injuries this year; however, it is now on its second clinical hold and no subjects have yet been enrolled. While this delay in translation to human trials is likely multifactorial, major obstacles remain with the clinical use of ESCs over ethical issues, oncogenic risk and the fact that ESC derivatives for tissue repair involve the use of allogeneic cells that can lead to rejection of mismatched cellular grafts. The oncogenic risk associated with stem cell therapies is no longer just theoretical. In 2009, the first report of a patient who developed multiple tumors in his brain and spinal cord after receiving fetal neural stem cells was published.³ Thus, therapeutic application of pluripotent stem cells, particularly in the cardiovascular field, will require further advancement and new approaches such as that described by Zwi et al. in this issue.⁴

While minimizing the oncogenic risk of pluripotent stem cells continues to be an area of intense research, investigators have long sought to overcome the ethical and immunologic issues surrounding ESCs by creating autologous pluripotent stem cells. Initial attempts to create these cells involved nuclear cloning technology, in which the somatic nucleus is transplanted into an enucleated oocyte. This technology is now used routinely in agricultural settings but lost favor in human cells since it too is fraught with ethical dilemmas and is very inefficient. In contrast, induced pluripotent stem cells (iPSCs) exploded onto the stem cell field less than three years ago. In a search for factors that could reprogram somatic cells to a pluripotent state, a groundbreaking paper published in 2006 by Takahashi and colleagues described four transcription factors whose retroviral overexpression enabled the induction of a pluripotent state in murine fibroblasts.⁵ Simultaneous ectopic expression of Oct4, Sox2, c-Myc, and Klf4 led to generation of iPSCs that were very similar to murine ESCs in morphology, proliferation, and teratoma formation. Several groups subsequently generated iPSCs that appeared indistinguishable from ESCs and were competent for formation of adult and germline chimeras.⁶ Somatic cell reprogramming has also been reported by several groups to be sufficient to produce human iPSCs that are morphologically and phenotypically similar to human ES cells.⁶ This has been further refined and now human iPSCs can be created without the use of viruses or need for genomic integration, a critical step on the pathway to producing clinically useable iPSCs.⁷ Proof of principle for using reprogrammed iPSCs combined with gene and cell therapy for disease treatment soon followed, further raising hope that autologous, patient-specific iPSCs could transform treatment of human diseases in the near future.⁸

Since the original report describing the derivation of murine iPSCs, research in reprogramming and with iPSCs has grown exponentially. It took less than a year to demonstrate that human fibroblasts, similar to mouse, could also be reprogrammed. This contrasts with the nearly seventeen-year lag between generation of mouse and human ESCs.⁹ Likewise, dramatic progress has been made in overcoming the need to use viruses for reprogramming, and alternative, virus-free methods to generate human iPSCs have been developed⁷ Several groups including ours demonstrated that mouse iPSCs share developmental pathways with ESCs and can differentiate into all three cell types typically found in the heart; namely endothelial cells, smooth muscle cells, and cardiac myocytes.^10–12 Recently, it has been reported that human iPSCs also share this potential.¹³ The manuscript by Zwi et al. in this issue is the most detailed characterization of human iPSC-derived cardiac myocytes published thus far.⁴ This series of advances in iPSC research is reminiscent of what was seen with ESCs, although they are occurring at a much faster pace. Does that mean we are more likely to see iPSCs deliver on the promises first ascribed to pluripotent ESCs; namely, establishment of model systems for human diseases, drug screening, and more comprehensive myocardial cell-replacement therapies? While iPSCs clearly represent a major advance with the potential to revolutionize the cardiovascular stem cell field, their therapeutic use faces a number of challenges; some shared with ESCs and others unique to iPSCs. Thus, optimism for transforming the promise of iPSCs into therapeutically relevant cardiovascular treatments should be tempered by the trajectory of hESC research and the realization that many critical questions about iPSCs remain unresolved.

Likely the biggest unanswered question is whether ESCs and iPSCs are identical or even functionally equivalent? Based on the standards that were originally used to define pluripotency for ESCs; self-renewal, expression of early stem cell markers, ability to differentiate into the three primary germ layers and contribution to germ cells in chimeric mice, iPSCs appear to be indistinguishable from ESCs. However, a detailed study of gene expression profiles of ESCs and iPSCs demonstrated that while iPSCs are very similar to ESCs, they have a distinct gene expression pattern suggesting that iPSCs should be considered a unique subtype of pluripotent cells.¹⁴ These differences in gene expression are likely the result of differential promoter binding by the reprogramming factors but could also be related to residual epigenetic changes that occurred when the cells originally differentiated. More importantly, the consequences of this distinct genetic profile are unknown. Will it lead to increased oncogenicity or less? Will it affect differentiation potential or graft survival when transplanted? Interestingly, these differences are overshadowed by a more practical limitation with pluripotent stem cells in general; namely, the marked heterogeneity that has been reported in differentiation potential within hESC lines¹⁵ and more recently hiPSC lines.¹³ Over 100-fold differences in differentiation potential between ESC lines were reported.¹⁵ It is likely that the observed variation in cardiac differentiation would have been even greater if the phenotype of the cardiac myocytes within differentiated cultures was analyzed in more detail to identify percentages of atrial, ventricular, and pacemaker-like cells, which are known to coexist within differentiating ESC cultures. The basis for the differences in differentiation potential was not identified; however, if iPSCs are to be used clinically, markers that correlate with differentiation potential will need to be developed so that appropriate clones can be identified for subsequent use.

One issue that iPSCs share with ESCs is whether the cardiac myocytes derived in vitro are truly representative of endogenous cardiac myocytes? Although it has been proposed that iPSC-derived cardiac myocytes could serve as a model of patient-specific diseases or as a platform for drug testing that might reduce cardiovascular toxicity in the future, ESC-derived cardiac myocytes more closely resemble fetal cardiac myocytes.¹⁶ This presents a problem for the application of these cells as a model system since cardiac signaling molecules, contractile proteins and ion channels all display developmentally dependent expression patterns. Thus embryonic cardiac myocytes are phenotypically quite different from adult cardiac myocytes and will likely have quite different responses to many compounds. Consistent with this, the electrophysiological properties of the cardiac myocytes described by Zwi et al. appeared incompletely differentiated. The conduction velocity (2 cm/s) is about 5–10% of that typically seen in normal intact ventricular muscle, probably reflecting a low gap junction connectivity and low expression of I_Na and/or I_Ca. Although not reported, it would have been useful to know the resting membrane potential of these cells, and how SR depletion affects conduction and beating to better characterize their developmental stage. This incomplete differentiation in iPSC-derived cardiovascular cells was also seen in endothelial and smooth muscle cells.¹⁰ Since ESC-derived cells can assume a more “adult-like” phenotype in vivo it suggests that the currently used two dimensional cell culture systems lack the appropriate clues to affect complete differentiation. This highlights the need to develop better three dimensional cultures systems that support complete cardiac differentiation into adult-like phenotypes and better replicate the cell-cell and cell-matrix interactions seen in vivo. However, even if these issues of differentiation can be overcome, the plasticity of expression of repolarizing currents (e.g., during electrical remodeling) makes it questionable whether cultured cell models will ultimately be useful to predict drug effects relevant to the same human’s diseased heart.

Perhaps most relevant to clinicians, is the question of the utility of iPSC-derived cardiac myocytes for cell therapy. Multiple animal studies have demonstrated that transplantation of ESCs improves cardiac function after myocardial infarction and recently, it was shown that intramyocardial delivery of iPSCs also leads to engraftment, restoring contractile performance and attenuating adverse remodeling.¹⁷ Interestingly, the mixed cell populations derived from embryoid bodies that were used in this study contributed to cardiac, smooth muscle, and endothelial cell-types suggesting iPSCs might affect a more complete regeneration. Although no tumors were identified in this small study, other investigators have found a significant risk of teratoma formation with using similar methodology.¹⁸ The concern over the oncogenic potential of iPSCs is heightened by the fact that all the iPSC lines that have been tested thus far for cardiac propensity were generated using integrating viruses that pose an additional tumor risk. However, the fact that cardiac myocytes were recently generated from iPSCs reprogrammed without the use of Myc suggests that this issue will soon be resolved using the newer viral-free lines.¹⁹ Nonetheless, whether any of the current strategies will be sufficiently safe for clinical studies, or if iPSCs present a lower risk than ESCs, will need to be determined in preclinical studies. Another practical limitation of ESCs is their immunologic intolerance. Although undifferentiated mouse and human ESCs may have some immune privilege, they are immunologically rejected as they differentiate and upregulate histocompatibility antigens.²⁰ It has been assumed that autologous iPSCs would evoke no immunological response; however, it is possible that persistent fetal or viral antigens, if viruses are required to induce pluripotency, could lead to immunological rejection limiting graft life or requiring ongoing immunosuppression. Finally, although it seems self-evident that transplanting autologous cells with a robust capacity to form beating cardiac myocytes would be more efficacious than using stem cells with less cardiac potential, there has been no direct comparison between adult stem cells, cardiac stem cells and pluripotent stem cells to support this contention. This is a major limitation in the field as the oncogenic risk of pluripotent-derived cells will need to be balanced by a significant therapeutic benefit to make them the preferred option for cell therapy.

These cautionary caveats are not meant to suggest that the outlook for iPSCs should be pessimistic. Instead, they are intended to highlight the critical questions that will need to be addressed as this new field develops and temper the expectations and timelines for clinical application of these cells. It is likely that this report and other recent studies on the cardiovascular potential of iPSCs do not represent the beginning of the end of this story. But, perhaps, the end of the beginning²¹ and because of that, portends a very exciting future for the cardiovascular stem cell field.