Utilizing stem cells for myocardial repair – to differentiate or not to differentiate prior to transplantation?

In recent years, tremendous interest has been aroused over the potential application of both embryonic 1 and adult 2 stem cells in the treatment of ischemic heart disease, which remains a major cause of morbidity and mortality in the developed world 3. This is undoubtedly due to the limited efficacy of current treatment modalities, which very often lead to inadequate recovery of heart function. For example, routine clinical techniques involving revascularization (i.e. angioplasty, thrombolysis, coronary artery bypass grafting) or surgical removal of diseased myocardium are often not able to fully alleviate ischemic injury and consequent myocardial fibrosis 4–7. Other modes of treatment that are currently being developed such as xenotransplantation 8 and artificial mechanical heart 9, have not yet advanced significantly to impact on the clinical management of myocardial infarction. Hence, stem cell transplantation to repair the damaged myocardium appears to be a much more promising treatment option for the infarcted heart. Indeed, this has been confirmed by several recent studies in animal models and phase I human clinical trials, which have invariably demonstrated much efficacy in aiding myocardial regeneration and preventing fibrosis, with consequent enhancement in the recovery of ventricular function 10,11.

Nevertheless, a key issue that has largely been ignored in the majority of these studies, is the differentiation status of the stem cells being transplanted into the damaged myocardium. The differentiation status of transplanted stem cells or their derivatives is likely to profoundly influence several factors that determine their ability to contribute to myocardial regeneration within the pathological heart model. These various factors will each be critically examined in turn.

It is a well-established fact that undifferentiated human embryonic stem (hES) cells form teratomas upon transplantation into immunologically compromised animal models 12. Teratoma formation would obviously be detrimental to the transplant recipient. Hence for human clinical therapy, it is imperative that hES cells are differentiated to some degree prior to transplantation into the damaged myocardium, so as to reduce the likelihood of teratoma formation in situ13. Another major concern is the possibility of hES cell differentiation into mononuclear cells, which could in turn initiate a graft vs host reaction. Again, this risk may be alleviated through some degree of commitment or differentiation of hES cells to specific lineages, prior to transplantation. Undifferentiated adult stem cells lack the ability to form teratomas, but could instead undergo non-specific multi-lineage differentiation in vitro14,15. This is likely to reduce the clinical efficacy of transplantation therapy, since only a sub-fraction of the transplanted stem cells would have differentiated into lineages that can potentially contribute to myocardial repair, i.e. cardiomyocytes, vascular endothelium.

Further complications may arise from the pathological state of the damaged myocardium providing an “abnormal” milieu for undifferentiated stem cells. The presence of necrotic/apoptotic cells, free radicals and inflammatory cytokines within the pathological environment of the infarcted heart, may very well exert an adverse effect on the differentiation pathway of the transplanted stem cells. This could further reduce the proportion of transplanted cells giving rise to lineages capable of contributing to myocardial repair, as well as result in the formation of undesired tissue lineages that can impair myocardial regeneration. For example, in the study of Wang et al. 16, the transplantation of undifferentiated mesenchymal stem cells into the pathological environment of damaged cardiac muscles, resulted in the differentiation of some of these cells into fibrotic scar tissue, which could in turn impair recovery of heart function after myocardial infarction.

Undifferentiated stem cells may also lack expression of specific cell-surface markers that could facilitate their rapid engraftment and integration within the recipient myocardium, i.e. connexin 43 (C×43) 17, N-cadherin 18, cardiac-specific isoform of dihydropyridine receptor (DHPR) 19,20. Expression of C×43 and DHPR would be particularly useful for achieving electro-mechanical coupling between the transplanted cells and host cardiomyocytes 17,19,20, which could in turn assist the recovery of myocardial contractile function. Hence, there is a strong likelihood that some degree of differentiation of stem cells into the cardiomyogenic lineage in vitro, to allow expression of such surface markers, may enhance their engraftment efficiency and subsequent integration in situ within the damaged myocardium, thereby improving the clinical efficacy of transplantation therapy.

Nevertheless, there are a number of factors which would discourage the use of differentiated stem cell derivatives for myocardial repair. In the case of autologous adult stem cells, directed differentiation into the cardiomyogenic lineage is likely to require prolonged durations of ex vivo culture. This would obviously delay treatment to the patient who is likely to be in dire need of therapeutic intervention. Furthermore, it must be noted that there probably exists a narrow time window in which stem-cell-transplantation therapy would be maximally beneficial for repairing the damaged myocardium following infarction. Beyond this time window, fibrosis and permanent scarring of the myocardium would occur, which could reduce the clinical efficacy of cell-transplantation therapy. Moreover, there is also evidence that prolonged durations of ex vivo culture could somehow alter the immunogenicity of cultured autologous cells, which may lead to immuno-rejection upon transplantation 21,22. It can certainly be argued that genetic modulation with recombinant DNA technology could potentially offer a “rapid” and efficient means of directing autologous adult stem cell commitment and differentiation into the cardiomyogenic lineage, without the need for prolonged ex vivo culture. There are however overwhelming safety concerns with regards to the application of genetic manipulation in human clinical therapy 23.

Differentiated stem cells may also have reduced proliferative capacity, which could impair their ability to contribute to myocardial regeneration. Terminally differentiated cardiomyocytes lose their ability for further cell division. Hence, the transplantation of a few million terminally differentiated cells is unlikely to be able to restore myocardial function after a loss of 2–10 billion cells during a major myocardial infarction. Moreover, by committing or differentiating stem cells into the cardiomyogenic lineage prior to transplantation, there could be a loss in their ability to give rise to other lineages that might also be beneficial to myocardial regeneration, in particular the vascular endothelial lineage. This is best illustrated by the transplantation of undifferentiated mesenchymal stem cells into the damaged myocardium. The transplanted stem cells give rise to not only the cardiomyogenic lineage 24, but also to the vascular endothelial lineage 25. The resultant increased angiogenesis appears to play a crucial role in myocardial regeneration 26.

Differentiated derivatives of stem cells, in particular the cardiomyogenic lineage, are likely to be much more fastidious in their nutritional and oxygen requirement, as compared to undifferentiated stem cells. This could very well compromise their ability to survive in the adverse pathological environment of the damaged myocardium. Indeed, the transplantation of fully differentiated cardiomyocytes has been reported to be severely hindered by the ischemic environment present within the pathological heart model, which often leads to an extremely low survival and engraftment rate of the transplanted cells 27.

Another major factor that would discourage the use of differentiated stem cell derivatives is their higher degree of immunogenicity, in the case of allogenic models of transplantation. It is well established that undifferentiated hES cells have low levels of expression of major histocompatibility complex (MHC) class I and II antigens 28. There is a gradual upregulation in the expression of MHC antigens by hES cells, with increased degree of differentiation 28,29. Hence, transplantation of undifferentiated hES cells or their derivatives at early stages of differentiation, is likely to provoke much less immunological reaction, as compared to hES derivatives at later stages of differentiation. In the case of undifferentiated bone-marrow-derived mesenchymal stem cells, these have been reported to be both immuno-privileged 30, as well as immuno-suppressive 31,32, which would favor the use of these cells in allogenic transplantation. As with ES cells, there has also been reported to be a gradual upregulation of MHC expression by mesenchymal stem cells, with increased degree of differentiation 30.

Hence, it would appear that a major issue of contention in utilizing stem cells for myocardial repair is the choice of transplanting undifferentiated stem cells, or their differentiated derivatives, i.e. cardiomyogenic lineage. Each of these has their advantages and disadvantages. It is likely that the various sub-populations of embryonic and adult stem cells being studied by different research groups may in fact represent many of the same cell lineages at different stages of differentiation. In effect, the search for the “ideal” stem cell sub-population for transplantation, which has characterized the past several years, may actually have been a search for the “optimal” degree of differentiation of the same cellular lineage. For example, in the arena of adult stem cell research, there are multipotent adult progenitor cells 33, mesenchymal stem cells 34, endothelial precursor cells 35 and cardiac stem cells 36, all of which could represent progressive stages of differentiation from the same lineage. Probably, there exists a subtle balance somewhere between the undifferentiated and fully differentiated state, which would be optimal in achieving maximal efficacy of stem-cell-transplantation therapy for myocardial repair. At present, this particular aspect of stem cell transplantation has not been well studied. More exhaustive investigations should be carried out in this area of research, in view of its potentially important clinical implications.