General Concepts About Stem Cells as Potential Therapeutic Agents

After completing this article, readers should be able to: Stem cells display the ability to self-renew (to divide and give rise to other stem cells) and to produce offspring that mature along distinct differentiation pathways to form cells that have specialized functions. In vertebrates, stem cells classically have been divided into two groups: embryonic stem (ES) cells and somatic or adult stem cells. ES cells are derived from cells in preimplantation blastocysts. Adult stem cells represent the self-renewing populations residing in many tissues and organs, including bone marrow, brain, retina, skin, intestine, liver, kidney, and pancreas.ES cells initially were isolated from cultured murine blastocysts, and the murine system continues to be the most amenable for ES cell line isolation. In the murine system, ES cells display the potential to repopulate (to some extent) all tissues of the embryo, including germ cells. Because adult stem cell populations potentially can be derived from these formed tissues and organs, adult stem cells by definition can be derived from ES cells. No evidence has yet been presented that adult stem cells can give rise to ES cells, although one population of adult stem cells—multipotent adult progenitor cells—display features strikingly similar to ES cells.The fertilized oocyte (zygote) is the “mother” of all stem cells. All the potential for forming all cells and tissues of the body, including the placenta and extraembryonic membranes, are derived from this cell. Thus, the zygote is a totipotent cell. The first few cleavage stage divisions produce blastomere cells that retain totipotent potential. However, by the 64-cell stage of mouse development, the cells begin to form a structure called the blastocyst. At this stage, cells begin to become specialized and lineage is committed. One portion of the blastocyst, the epiblast, contains cells (inner cell mass) that eventually form the embryo proper ( Fig. 1). Trophectoderm cells comprise the cells at the opposite pole of the blastocyst, and these cells differentiate to form the placenta. Cells within the inner cell mass of the blastocyst are pluripotent; that is, each cell possesses the potential to give rise to types of cells that develop from the three embryonic germ layers (mesoderm, endoderm, and ectoderm).Mouse ES cells were isolated more than 20 years ago. Inner cell mass cells were recovered from murine blastocysts and plated over an adherent layer of mouse embryonic fibroblasts in the presence of culture medium that contained fetal calf serum and, in some instances, conditioned medium from murine teratocarcinoma cells. Over a period of several weeks, colonies of rapidly growing cells emerged. These colonies could be recovered from the culture dishes, enzymatically disaggregated, and replated on fresh embryonic fibroblasts, with rapidly growing colonies re-emerging. The cells comprising the colonies eventually were defined as ES cells.Murine ES cells display several unique properties. They are small and have a high nuclear-to-cytoplasmic ratio and prominent nucleoli. ES cells preferentially grow in compact colonies that have tight intercellular interactions. When plated in the presence of murine embryonic fibroblasts, ES cells proliferate indefinitely. Injection of ES cells into recipient blastocysts permits ES cell-derived contribution to essentially all tissues of the embryo, including germ cells. ES cells express the enzyme telomerase and, thus, retain telomeric DNA despite the extensive proliferative behavior of the cells.Although the molecular regulation of ES cell self-renewal divisions remains uncle`yN rY growth factor leukemia inhibitory factor (LIF) has been deemed sufficient to maintain murine ES cells in a self-renewing state in vitro in the absence of the mouse fibroblast feeder cells. ES cells grown in vitro in LIF retain pluripotent properties on injection into a recipient blastocyst and after implantation, contribute to all of the tissues of the resulting newborn animal (chimera). Some of the intracellular signaling pathways that are activated on LIF binding to the cell surface glycoprotein 130 receptor have been identified, and the presence of constitutively active forms of these signaling molecules maintains ES cell pluripotency in vitro.In subsequent years, many investigators have defined the strict culture conditions that are required for in vitro differentiation of murine ES cells into a wide variety of somatic cell types, such as neurons, hematopoietic cells, pancreatic cells, hepatocytes, muscle cells, cardiomyocytes, and endothelial cells. In most differentiation protocols, ES cells initially are deprived of LIF, followed by the addition of other growth factors, vitamins, morphogens, extracellular matrix molecules, or drugs to stimulate the ES cells to differentiate along specific pathways. It is usual for the ES cell differentiation to give rise to a predominant but not pure population of differentiated cells. Obtaining highly purified differentiated cell populations generally requires some form of cell selection either to enhance survival of a selected population or to eliminate preferentially a nondesired population. The ability to isolate enriched populations of differentiated cells has encouraged many investigators to postulate that ES cells may be a desirable source of cells for replacement of aged, injured, or diseased tissues in humans if pluripotent human ES cells were available.The growth conditions that have permitted isolation and characterization of human ES cells were reported recently. Cleavage-stage human embryos produced by in vitro fertilization for clinical purposes were obtained following informed consent. Embryos were grown to the blastocyst stage, the inner cell mass cells isolated, and the isolated cells plated on irradiated mouse embryonic fibroblast feeder layers in vitro ( Fig. 2). After growing in culture for several cell divisions, colonies of ES cells were established, similar to murine ES cells. These human ES cells were very small cells, had minimal cytoplasm and prominent nucleoli, grew very rapidly (like the mouse cells) without evidence of developing senescence, and possessed high telomerase activity. Unlike for mouse ES cells, LIF is not sufficient to maintain human ES cells in a self-renewing state in the absence of the mouse fibroblast feeder cells. However, human ES cells can be grown on extracellular matrix-coated plates in the presence of murine embryonic fibroblast-conditioned medium.The pluripotent nature of human ES cells has been demonstrated by injecting the cells into the hind leg musculature of an immunodeficient mouse. A teratoma developed over 3 months, and the cells forming the tumor were demonstrated to contain gastric and intestinal epithelium, renal tubular cells, and neurons—descendants of the endoderm, mesoderm, and ectoderm germ cell layers, respectively. Identification of several clones of human ES cells and the development of teratomas on implantation of ES cells from each clone in the immunodeficient mice has been interpreted as evidence of the pluripotent nature of these cells.Using similar in vitro techniques, human embryonic germ (EG) cell lines have been isolated from the genital ridges of aborted 5- to 10-weeks’ gestation human fetuses. The genital ridge of the fetus is the region that later forms the ovary or testes and the primordial germ cells that give rise to eggs or sperm. Human EG cells have displayed similar abilities for prolonged in vitro culture with retained stem cell properties and for in vivo generation of human teratomas in nude mice.Thus, in rapid succession, standardized techniques have evolved to isolate and culture human ES and EG cells in vitro. Recent successes in differentiating these cells into specific cell types has raised the expectation that they may serve as a source of cells to replenish aged or damaged tissues in human disorders. However, the number of ES cell lines may be far fewer than the vast number of humans requiring cellular therapy. One significant obstacle is the potential immunologic barrier to transplantation of tissues of disparate major histocompatibility antigens (ES cell-derived tissues) into humans. Such a barrier could be overcome using nuclear transfer technology.The successful cloning of a variety of domestic animals and laboratory rodents has been widely reported. This technology is based primarily on nuclear transfer techniques in which the nucleus is removed from the oocyte, and a donor somatic cell nucleus is fused electrically with the enucleated oocyte. The created zygote is grown to the blastocyst stage, and the embryo is disaggregated and cells from the inner cell mass are harvested for creation of ES cells in vitro or the blastocyst is implanted into a recipient female ( Fig. 3). Such a procedure is technically challenging, but possible.Some of the challenges to overcome when using nuclear transfer technology to create viable cloned animals include the great inefficiency of the process (hundreds to thousands of oocytes frequently are injected, with only a few viable animals surviving beyond birth). Much of this inefficiency may be a result of poor epigenetic reprogramming of the donor somatic nucleus in the oocyte. In adult somatic tissues, epigenetic modifications of DNA and chromatin are maintained stably and characteristic of each specialized tissue or organ. During nuclear transfer, epigenetic reprogramming of the somatic nucleus must occur similar to the epigenetic programming that normally occurs during oocyte activation following fertilization. Epigenetic reprogramming deficiencies during animal cloning may lead to a host of problems, including epigenetic mutations and altered epigenetic inheritance that results in embryonic lethality or maldeveloped fetuses that have poor postnatal survival. Great strides have been made in identifying the molecules involved in chromatin remodeling and epigenetic programming, but considerable work remains to identify strategies to facilitate this process during nuclear transfer for animal cloning.A combination of the techniques of nuclear transfer and ES cell generation could be applied to human therapeutic organ repair. In this strategy, a donor oocyte is enucleated, and the nucleus of a somatic cell from the patient is isolated and transferred into the oocyte. A blastocyst is created and disaggregated, and human ES cells are isolated. These ES cells would be immunologically identical to the donor, with the exception of antigens expressed by mitochondria originally derived from the donor oocyte. Once ES cells have been generated, specific differentiation protocols designed to derive the individual cells for creation of a replacement organ could be employed. Because the differentiated cells generated from the ES cells are formed as a monolayer of cells, and most tissues and organs for replacement purposes may need to be synthesized in three dimensions with an accessible vascular supply, a composite synthetic matrix comprised of suspended differentiated cells may need to be engineered in vitro. Although this hypothetical scheme may appear too complicated for any practical near-term application, a recent proof of principle experiment has demonstrated that an engineered tissue that has functioning nephrons can be produced from domestic cattle.Transfer of human nuclei into donor oocytes presents many ethical and practical challenges. What will be the source of donor oocytes? Is it ethical to create intentionally and then sacrifice human embryos for the potential clinical benefit of a patient? Are the embryos created by nuclear transfer equivalent to normally fertilized oocytes? One of the other articles in this issue addresses some of these and related ethical questions.Technically, although ES cells can be created, no protocols for differentiation of specific cell lineages that are therapeutic have been identified. Furthermore, although use of artificial matrices as scaffolds for transplantable cells currently is feasible (engineered blood vessels, bladder, skin, and cartilage), there are no reports of using undifferentiated stem cells to repopulate artificial matrices. Nevertheless, significant progress has been achieved using murine ES cells or adult somatic stem cells to produce tissues and organs that display functional properties in vivo.Adult (also called somatic or postnatal) stem cells are multipotent cells that reside in specialized tissues and organs and retain the ability to self-renew and to develop into progeny that yield all the differentiated cells that comprise the tissue or organ of residence. For example, hematopoietic stem cells reside in the bone marrow compartment where they continuously replenish the blood with circulating blood elements. Hematopoietic stem cells divide slowly in the marrow and replenish themselves throughout life. Other sources of self-renewing adult stem cells are the cornea, retina, brain, skeletal muscle, dental pulp, skin, pancreas, liver, and lining of the intestine.A number of studies have reported that adult stem cells isolated from one organ possess the ability to differentiate into cells of completely different organs on transplantation. For example, bone marrow cells have been demonstrated to contribute to muscle, lung, gastric, intestinal, lung, and liver cells, and neuronal stem cells can contribute to blood, muscle, and neuronal tissues. One of the other articles in this issue discusses the principles of stem cell plasticity and the potential challenges of using adult stem cells for therapeutic purposes.Diabetes mellitus is a chronic debilitating disease in which elevated blood glucose concentrations and deficient insulin production or intracellular responses to insulin lead to an increased risk for affected patients to develop hypertension, vascular disease, stroke, and kidney failure. Oral hypoglycemic medications in some patients and insulin injections for most severely affected patients can lower blood glucose concentrations effectively, but currently there is no cure for this disease.Some success recently was reported with the transplantation of donor pancreatic islets into the livers of diabetic recipients, but these patients require lifelong immunosuppressive medications to suppress tissue rejection, and the lack of available donor material undoubtedly will limit this treatment strategy. Development of methods to expand differentiated donor beta-islets are being sought, but this approach is not currently effective in improving the number of islets for transplantation. Alternative approaches include finding stem cell sources that could be used to generate beta-islet cells in vitro in sufficient quantities for transplantation.Insulin-producing cells have been generated from both murine and human ES cells in vitro. Implantation of murine ES cell-derived insulin-producing cells into the spleens of recipient mice that were rendered hyperglycemic via streptozotocin injection led to euglycemia in nearly all of the transplanted but none of the sham operated animals. Some of the treated animals remained euglycemic for months after the transplant, but hyperglycemia returned in other transplant recipients. No mechanisms have been determined to explain the persistent beneficial effect in only a subpopulation of the diabetic recipient mice. Nevertheless, the in vivo regulation of blood glucose concentrations by the implanted ES cell-derived cells was encouraging.Insulin-producing cells also have been generated from pancreatic ductal tissue, liver stem cells (oval cells), and liver cells transduced with an adenoviral vector expressing a transcription factor known to be necessary for pancreatic formation. In some cases, not only did the cells secrete insulin in vitro, but they responded with accelerated differentiation when exposed to molecules known to stimulate beta-islet cell growth and differentiation in vivo. Implantation of the insulin-producing cells also ameliorated the hyperglycemia in some streptozotocin-treated recipient mice.These data suggest that cells synthesizing and releasing insulin may be derived from either ES or adult stem cell populations. Whether implantation of beta-islet cells produced from stem cells will function as well as donor pancreatic beta-islet transplantation (which may contain other pancreatic cell types) and whether immunosuppressive regimens will be necessary to accept and maintain the implanted cells remain unknown. Further work also is required to determine whether either ES or adult stem cells can be produced in sufficient mass to accommodate the number of beta-islet cells required for clinically relevant protocols and whether stem cell-generated pancreatic cells will be effective in establishing euglycemic outcomes in large animal models of diabetes mellitus.Chronic heart failure is another disease entity affecting a large population worldwide. Nearly 5 million patients in the United States suffer from persistent and worsening heart failure. Mortality is high (up to 40%), and costs to society are much greater than the $40 billion attributed annually to hospitalization, medication, and medical follow-up costs. Of greater concern is the knowledge that these costs will increase considerably with the advancing age of the United States population.Unlike many organs of the body that can replace aged or injured cells, cardiomyocyte loss in the adult heart is believed to be irreversible. Some controversial recent data suggest that the heart may contain a population of cells that can proliferate and replace injured or aged cardiomyocytes. However, most strategies to repair injured cardiac tissue are focused on methods for recruiting stem cells to the site of injury or injecting stem cell-derived cardiomyocytes into the injured areas.Numerous laboratories have attempted some form of cell-based therapy for cardiac repair. A variety of cell sources and different types of myocytes have been engrafted in infarcted hearts, but there is little evidence of these cells vigorously interacting with the host cardiomyocytes to establish functional electrophysiologic interactions. Fetal cardiomyocytes can connect functionally with injured adult cardiomyocytes and improve heart function, but the number of fetal cardiomyocytes is limited, and they are difficult to isolate and transplant with high efficiency. Some recent data suggest that murine marrow stem cells or mesenchymal stem cells may possess the potential to become cardiomyocytes under certain experimentally induced cardiac injuries in mice. Additional evidence has been presented that noncardiac-derived cardiomyocytes have been detected in sex-mismatched transplanted human hearts.Cardiomyocytes also have been produced in vitro from murine and human ES cells. Specific culture conditions, including transfection of cells with genes to confer a selective survival advantage, have been used to generate nearly pure populations of cardiomyocytes. These cells spontaneously beat in culture and respond appropriately to pharmacologic agents that normally increase or decrease cardiomyocyte beating in vivo. ES cell-derived cardiomyocytes express proteins known to be restricted to cardiomyocytes in vivo, display normal sarcomeric structures, and can engraft into injured murine hearts in vivo.In summary, the in vitro differentiation of cardiomyocytes from murine ES cells has been accomplished. Results to date suggest that the cells display morphologic, electrophysiologic, and molecular expression patterns similar to cardiomyocytes that develop during embryogenesis. Human ES cell-derived cardiomyocytes also have been isolated and characterized. It is not yet known if these cells will engraft and facilitate recovery of damaged human heart tissue.Stem cells may be derived in vitro from preimplantation mammalian blastocysts (ES cells) or from tissues and organs of adults (adult stem cells). ES cells display certain unique properties that generate enthusiasm for their use as a source of differentiated cells for future applications of cell-based therapies for human disease. Adult stem cell populations also are being investigated as potential sources for clinical cell-based therapies. Preliminary studies suggest that stem cell-derived tissues may function in vivo in some animal models of human disease. Current investigation is focused on improving cell isolation, expanding in vitro stem cells, regulating stem cell commitment to specific cell lineages, facilitating in vitro cellular differentiation, engineering tissue using synthetic matrices and stem cell progeny, optimizing transplantation protocols, and testing in vivo stem cell or stem cell-derived tissue for safety and efficacy in appropriate animal models of human disease. Many challenges lie ahead, but successful outcomes of this work may lead to new patient treatment paradigms.