January 2010 | Vol 7 | N.º 1 | CNIC-16 [ PDF (1060 KB) ]

New and old cell homeostasis concepts in the healthy and diseased adult heart

Antonio Díez-Juan, Alberto Izarra, Iñigo Valiente, Isabel Moscoso, David Lara and Antonio Bernad
Departamento de Cardiología Regenerativa, Centro Nacional de Investigaciones Cardiovasculares Carlos III, Melchor Fernández Almagro 3, E-28029 Madrid, Spain. Tel: (+34) 91/ 453 12 74; Fax: (+34) 91/ 453 12 40;
Email mramon@cnic.es

* These authors contributed equally to this work

Corresponding author: Antonio Bernad. Departamento de Cardiología Regenerativa, Centro Nacional de Investigaciones Cardiovasculares Carlos III, Melchor Fernández Almagro 3, E-28029 Madrid, Spain. Tel: (+34) 91/ 453 12 74; Fax: (+34) 91/ 453 12 40; e-mail: abernad@cnic.es

Summary
The adult heart has long been considered a terminally differentiated organ; however, developments in the past decade have shown that a small but significant proportion of cardiomyocytes are replaced during adulthood to contribute to organ homeostasis. The generation of new cardiomyocytes decreases with age and is slightly enhanced by injury. Nevertheless, this turnover remains insufficient to regenerate the heart after severe acute damage. Heart healing is a complex process and requires suppression of inflammatory processes, activation of the cell cycle in specific cell populations and recruitment of cells with healing and regenerative potential. In this review, we summarize known mechanisms that might contribute to the basal and pathophysiological turnover of cardiac cells. These include re-entry of mature cardiomyocytes into the cell cycle, epithelial- or endothelial-mesenchymal transitions, recruitment of exogenous progenitors and activation of endogenous or cardiac progenitor cells (CPC). The main characteristics of these processes, their interrelationships and the current comprehensive state of knowledge in the field will be discussed.

Introduction
The adult heart was long considered a terminally differentiated organ, unable to replace damaged or aged cardiomyocytes.¹ The central dogma was that, after birth, the heart could adapt to increased workload only by hypertrophy of viable and functional cardiomyocytes. This notion was based on the then dominant view that highly differentiated mammalian cells such as adult cardiomyocytes were incapable of proliferation. However, although this view was widely accepted, it did not go unchallenged (reviewed by De Falco et al. ²)

Evidence acquired in recent years has lent increasing support to views that challenge the cardiomyocyte hypertrophy model. First, it is obvious that during normal maturation the generation of new cardiomyocytes must predominate over cardiomyocyte death, contributing significantly to organ growth and homeostasis. During the course of physiological aging, when this balance shifts and cardiomyocyte formation is overtaken by cell death, the number of ventricular cardiomyocytes decreases (see Figure 1). Cardiomyocyte hypertrophy then dominates, and in time can contribute to chronic heart failure. Overall, it seems clear that the heart’s intrinsic repair capacity is insufficient to cope with severe acute or chronic injury.

The recognition that a small proportion of cardiac-resident cells can be re-programmed to re-enter the cell cycle fuelled the current working hypothesis that cardiomyocytes are continuously replaced, albeit at a comparatively low rate, throughout adulthood and old age, and that cell regeneration is enhanced by haemodynamic overload and by ischaemia.³, ⁴ Very recently, Frisen and colleagues have confirmed cardiomyocyte turnover in human adults.⁵ These authors estimated an annual turnover of about 1 percent in young adults, declining to below 0.45 percent by the age of 75.

The mechanism by which cardiomyocyte renewal occurs is still unknown. Cell subsets that might contribute to cardiac cell turnover include endogenous cardiac stem-like cells or cardiac progenitor cells (CPC), bone marrow-derived stem cells (BMDC), or mature cardiomyocytes that re-enter the cell cycle (“re-proliferation”). Contributions from different cell types may not be mutually exclusive. Our knowledge of cardiac biology is growing fast and the heart is now more clearly appreciated as a very efficient pumping machine controlled by a very complex biology. Discoveries in the coming years promise even greater advances that might finally allow humans to achieve the goal of heart regeneration.

Mechanisms contributing to steady-state homeostasis
The possibility that turnover and growth of myocardial cells is regulated by cardiomyocyte re-proliferation, resident stem cells or recruited progenitor cells does not solve the central problem of the adult heart’s apparently insufficient capacity to recover from severe acute or chronic injury. Heart healing, like that of many other organs, is a complex and concerted process, involving many cell types and diverse molecular events. Optimal infarct healing requires timely activation of "stop signals" that suppress chemokine and cytokine synthesis and rapidly resolve the inflammatory response. In addition, the inflammatory infiltrate itself should ideally induce signals to activate and recruit cells with the potential to heal and regenerate the injured tissue. Dynamic changes in the cellular network that take place in response to myocardial infarction (MI) play a key role in regulating cardiac healing. Characterization of the cellular subsets and mechanisms involved in basal cardiac homeostasis will increase understanding of the changes that occur during acute or chronic insults and during physiological organ degeneration. The main known putative mechanisms are discussed below.

Cell cycle re-entry by mature cardiomyocytes
In healthy adult mammalian myocardium, by far the greatest contribution to growth in response to increased cardiac load is cell hypertrophy. Indeed, fewer than 0.01% of adult cardiomyocytes in the human heart are able to divide following myocardial infarction.⁶ Nevertheless, many laboratories have shown that the cardiomyocyte cell cycle can be successfully reinitiated, although, in some cases, forcing cardiomyocyte cell-cycle progression can induce hypertrophy or apoptosis. Studies targeting the E2F family, cyclin D1 and cdk4, CDC5 and p38 demonstrate that successful reinitiation of the cardiomyocyte cell cycle requires expression of additional proteins, such as E1A/E1B, or growth factor stimulation (reviewed in ²). Proliferation of fully differentiated cardiomyocytes is promoted by growth factors such as periostin⁷ and neuregulin1; ⁸ these findings clearly disprove the old dogma that adult cardiomyocytes cannot proliferate. Adult cardiomyocyte cell division is, however, inefficient compared with that of neonatal cardiomyocytes. Interestingly, periostin and neuregulin1 seem to induce division by mononucleated, but not binucleated, cardiomyocytes. Therefore, research efforts should focus not only on the regulators of cardiomyocyte cell cycle re-entry, but also on factors that permit or promote the completion of cytokinesis, about which very little is known.

Controlled layer or lineage transition (CLT)
Controlled layer or lineage transitions (CLT), such as the epithelial-mesenchymal transition (EMT), are important biological processes through which polarized epithelial cells assume a mesenchymal phenotype, with enhanced migratory capacity, invasiveness, elevated resistance to apoptosis, and a greatly increased production of extracellular matrix components. Upon reaching their destination, migrating mesenchymal cells usually undergo reverse EMT.⁹
The development of the mouse heart involves four distinct CLT episodes.¹⁰ One of these involves a subpopulation of endocardial cells that undergoes an endothelial-to-mesenchymal transition (EndMT) and builds the endocardial cushions that form the heart valves. Interestingly, this EndMT continues during adulthood, at least in the mature valves, providing cells to conserve and restore the valvular leaflets.¹¹

While the significance of EMT and EndMT in organogenesis is well supported, little is known about the contributions of these processes to repair and regeneration in adults. Terminally differentiated epithelia have recently been shown to modify their phenotype in response to repair-associated or pathological stress¹², and EMT seems to yield adult cells with stem cell characteristics.¹³ These findings suggest that CPC might be derived directly from epithelial or endothelial cells, similar to recent findings with neural stem cells. In this regard, a recent study in mice demonstrated that microvascular endothelial-like cells contribute, via EndMT, to the formation of mesenchymal cells and fibroblasts during the course of cardiac fibrosis.¹⁴

Although the mechanism is not fully defined, it is plausible that epithelial/endothelial cells could, under the influence of specific signals (such as EGF, FGF-2, TGF-β, and PDGF) and acting together with inflammatory cells, induce basement membrane damage and focal degradation of collagen and laminin. Delaminated epithelial/endothelial cells might then migrate to interstitial areas under the influence of growth factors and other chemoattractants. Such a hypothesis is supported by the finding that, in mouse models of pressure overload and chronic allograft rejection, EndMT in the adult mouse heart generates myofibroblasts that migrate and produce scar tissue, recapitulating pathways that occur during formation of the atrioventricular cushions in the embryonic heart.¹⁴ While not yet documented, it is also plausible that many of the molecular regulators of EMT could also be critical regulators of EndMT. One could thus predict that, in line with their role during heart development, EMT and EndMT contribute to the pool of adult cardiovascular progenitor cells. A number of recent studies provide evidence supporting this notion.¹³, ¹⁴

Recruitment of exogenous progenitors or stem cells
The contemporary field of cardiac regeneration research began when investigators examining samples of female hearts recovered from male transplant recipients (sex-mismatched transplants) found compelling evidence that cardiac cells not only migrate from the male host to the transplanted heart but also generate new myocytes, as well as endothelial and smooth muscle cells.¹⁵ This matter soon gained supporters and detractors, and several waves of positive and negative results regarding these findings were published. The main issue, still unresolved, is the prevalence of cardiac chimerism. It has been possible to identify stem-cell-like cells in the adult myocardium that bear sex chromosomes or genetic markers of recipient origin in the case of heart transplants,¹⁶ or of the donor in the case of bone marrow transplants.,¹⁷ These extra-cardiac CPC may arise from sources that continuously feed the heart with undifferentiated cells and acquire their tissue-specific properties once in the cardiac niche. Quaini et al. report that within 4-28 days after allotransplantation, up to 30% of transplanted myocardium is regenerated by stem cells originating from the recipient.,¹⁶

Reported negative results include the detection of infiltrating host cells comprising up to 5.6% of the vascular smooth muscle cells but none of the >6000 cardiomyocytes surveyed in the donor heart,18 and low levels (0.02% to 1%) of cardiomyocyte chimerism within allografted human hearts.,¹⁹ Further complicating the picture, the validity of stem-cell plasticity as a mechanism for generating non-lymphohaematopoietic tissue has also been questioned20 and some authors have suggested that the apparent cardiac chimerism might be explained by previously overlooked cell fusion events.,²¹ In addition, the observation that fusion is a major mechanism through which BMDC contribute to heart, liver and brain challenges the rationale for clinical procedures based on the idea that transdifferentiation of BMDC can generate heart cells de novo. Additional studies in animal models will be required to determine whether fusion by BMDC can be exploited in reparative cell therapy; only by understanding the long-term fate of the epithelial cell fusion hybrid will we uncover its physiological role in homeostasis and disease.

Transplantation studies involving sex-mismatched peripheral blood stem cells suggest that the cells responsible for generating solid-organ–specific cells are, like haematopoietic stem cells, a group of circulating mononuclear cells. As outlined by Frisen and co-workers, these studies suggest several possible explanations for how adult stem cells derived from bone marrow or peripheral blood differentiate into non-lymphohaematopoietic tissue cells.,²² Therefore, it is conceivable that non-lymphohaematopoietic, organ-specific stem cells move between their own solid tissue and the peripheral blood. Numerous reports show that in heart, at least for endothelium and myofibroblasts, this can in fact be the case.,²³, ,²⁴

Endogenous reservoirs of progenitor or stem cells in the heart
Today there is abundant evidence that postnatal and adult heart contains a pool of resident CPC that can contribute to the turnover of organ structures and is “activated” during stress or injury. Nevertheless, the various CPC populations isolated have different phenotypic characteristics and their relationship to each other remains unclear. The major CPC populations described to date are listed below (Table 1) and discussed in comparison with embryonic and postnatal cardiovascular progenitors.

Cardiac Progenitors in the Adult Heart
Anversa and colleagues identified and isolated a minority population of c-kit positive/lineage negative cells (CPCc-kit+ population) resident in adult rat heart. These cells were described as clonogenic, self-renewing and multipotent, giving rise to all the main heart lineages. CPCc-kit+ populations have been isolated from several species²⁵, ²⁶ and, although the phenotype of the CPC cells differentiated in vitro remains immature, they are capable of regenerating cardiac tissue when injected after MI, differentiating into beating cardiomyocytes and contributing to increased capillary and arteriole density. ²⁶, ²⁷

Cardiosphere forming cells (CPCCsph) were first described by Giacomello and colleagues, and have been isolated and characterized from several species.28 This population differs from others mainly in the method of their isolation, which includes an in vitro sphere-formation step. Murine CPCCsph beat spontaneously, whereas human CPCCsph require co-culture with neonatal rat cardiomyocytes to promote differentiation in vitro. CPCCsph can differentiate into any of the three main heart lineages and improve cardiac function in treated animals.²⁹

Another putative murine CPC population is characterized by the expression of stem cell antigen-1 (Sca-1+).30, 31 These cells (CPCSca-1+) are negative for c-kit or any haematopoietic marker, but do express CD³¹ and the early cardiac markers GATA4 and MEF2c, suggesting a possible early commitment to cardiomyocytic and endothelial cell lineages. Transplantation-on-infarct studies have demonstrated that with CPCSca-1+ home to and differentiate in the infarct border zone.³⁰

Cardiac side population (SP; here referred to as CPCSP) is characterized by its ability to efficiently efflux dyes and is thought to be related to the CPCSca-1+ population.³² CPCSP cells express the specific membrane transporters Abcg2 and MDR1, which confer the ability to exclude compounds such as rhodamine or Hoechst 33342. During differentiation in vitro,CPCSP cells express some early cardiac proteins, but are unable to develop a mature phenotype unless co-cultured with neonatal rat cardiomyocytes, when they express sarcomeric proteins with an organized structure and begin to beat spontaneously.³³

Other cell population with potential to differentiate into cardiomyocytes, in vitro, and to repair the infarcted myocardium, in vivo, is a pericyte-like population isolated from explanted mouse and human heart tissue (cardiac mesoangioblasts).³⁴

From the data summarized above, it is clear that there is great heterogeneity in the surface markers that define adult CPC population(s). The main causes of this heterogeneity are probably the use of mixed populations in many studies, the difficulty of reproducing cell culture and differentiation conditions between groups, and the lack of defined criteria for identifying a cell as a true CPC. The ontogenic, physiological or developmental stage relationships among the different populations and their in vivo roles in cardiac repair/regeneration have yet to be addressed.

Relationships between embryonic and adult cardiovascular progenitors
The origin of the cardiac progenitor populations found in adults and their relationship to embryonic analogues have been widely discussed, but little or no data have been reported so far that effectively clarify this issue. The main question is clear: Do adult CPC have an embryonic origin or are they derived from an extra-cardiac source such as bone marrow?

Studies of heart development and embryonic stem cell (ES) models, together with lineage tracing approaches, provide strong support for the presence in the embryo of multipotent cardiovascular progenitors that can give rise to all main cardiovascular lineages. Kattman et al. found that individual cardiac Flk-1+ (fetal liver kinase 1) progenitor cells could generate colonies containing differentiated cardiac muscle, endothelial and smooth muscle cells.³⁵ Recently, Yang et al. isolated a similar population (KDRlow/c-kitneg) from human embryonic stem cell differentiation cultures; these appear to constitute one of the earliest human cardiovascular progenitors.³⁶ In another study, Wu et al. isolated early Nkx2.5-expressing cardiac progenitors and proposed the existence of a bipotential precursor in the developing heart.³⁷ Finally, Chien and colleagues described a population of cardiovascular progenitors characterized by the expression of the LIM-homeodomain transcription factor Islet-1 (Isl1).³⁸ Isl1+ progenitors can be found until postnatal day 8 in the pre-aortic tract, atrium and right ventricle of mouse hearts, and have also been detected in humans; however, for these cells, no evidence of regeneration capacity has been reported. Further analysis revealed an embryonic clonal population, defined as Isl1+/Nkx2.5+/Flk1+, that is capable of spontaneous differentiation in vitro into endothelial, cardiac and smooth muscle cells.³⁹

Conclusions and Perspectives
The presence of Isl1+ progenitors in the postnatal heart³⁷ and the presence of CPCSP cells in the developing heart³² argue in favour of the embryonic origin of adult CPC. Also, Wu and colleagues identified a c-kit+/Nkx2.5+ progenitor population in the embryonic heart that might be the precursor of the adult CPCc-kit+ population described by Anversa and colleagues.²⁶Thus, the adult CPCc-kit+ population could represent a persisting population of multipotent progenitors derived from embryonic progenitor cells. Against this hypothesis, the cases of sex-mismatched cardiac and bone marrow transplants¹⁶, ¹⁷ suggest at least a contribution from a postnatal source of adult CPC. Furthermore, we cannot dismiss other mechanisms such as EndMT/EMT that, under certain conditions, are able to provide stem cell-like populations and participate actively in organ healing.

More studies are needed to clarify the relationship between embryonic and adult CPC populations. Exhaustive clonal analysis and fate tracing assays will be crucial for determining the origin of adult progenitors in the heart. These techniques have been used to define the origin of embryonic cardiac progenitors but have scarcely been applied to the study of adult CPC. Therefore, new animal models and in vivo fate tracing strategies represent the logical next step on the way to unravelling the origin and identity of adult cardiac progenitors and their behaviour during homeostatic and pathologic processes.

Approaches to improve post-injury cardiac repair have focused mainly on cell replacement strategies; many groups have evaluated cell populations in clinical trials, but with poor results. Modest improvements in cardiac function have been reported, probably due to a paracrine effect of the transplanted cell populations, but little or no contribution of the transplanted cells to the repaired myocardium has been demonstrated. Effective heart healing will require the integration of knowledge of adult cardiac biology gained from multidisciplinary approaches (Figure 1) as well as careful evaluation of treatment strategies, including the potential to manipulate the intrinsic regenerative potential of cardiac muscle.