In August of last year, President Bush approved the use of federal funds to support research on a limited number of existing human embryonic stem cell lines. The decision met with notably mixed reactions. Proponents of embryonic stem cell research argue that restricting federal funding to a limited number of cell lines will hamper the progress of science, while those opposed insist that any use of cells derived from human embryos constitutes a significant breach of moral principles. It is clear that pressure to expand the limits established by the President will continue. It is equally clear that the ethical positions of those opposed to this research are unlikely to change.

Regrettably, much of the debate on this issue has taken place on emotional grounds, pitting the hope of curing heartrending medical conditions against the deeply held moral convictions of many Americans. Such arguments frequently ignore or mischaracterize the scientific facts. To arrive at an informed opinion on human embryonic stem cell research, it is important to have a clear understanding of precisely what embryonic stem cells are, whether embryonic stem cells are likely to be useful for medical treatments, and whether there are viable alternatives to the use of embryonic stem cells in scientific research.

Embryonic development is one of the most fascinating of all biological processes. A newly fertilized egg faces the daunting challenge of not only generating all of the tissues of the mature animal but organizing them into a functionally integrated whole. Generating a wide range of adult cell types is not an ability unique to embryos. Certain types of tumors called teratomas are extraordinarily adept at generating adult tissues, but unlike embryos, they do so without the benefit of an organizing principle or blueprint. Such tumors rapidly produce skin, bone, muscle, and even hair and teeth, all massed together in a chaotic lump of tissue. Many of the signals required to induce formation of specialized adult cells must be present in these tumors, but unlike embryos, tumors generate adult cell types in a hopelessly undirected manner.

If a developing embryo is not to end up a mass of disorganized tissues, it must do more than generate adult cell types. Embryos must orchestrate and choreograph an elaborate stage production that gives rise to a functional organism. They must direct intricate cell movements that bring together populations of cells only to separate them again, mold and shape organs through the birth of some cells and the death of others, and build ever more elaborate interacting systems while destroying others that serve only transient, embryonic functions. Throughout the ceaseless building, moving, and remodeling of embryonic development, new cells with unique characteristics are constantly being generated and integrated into the overall structure of the developing embryo. Science has only the most rudimentary understanding of the nature of the blueprint that orders embryonic development. Yet, recent research has begun to illuminate both how specific adult cells are made as well as the central role of stem cells in this process.

The term “stem cell” is a general one for any cell that has the ability to divide, generating two progeny (or “daughter cells”), one of which is destined to become something new and one of which replaces the original stem cell. In this sense, the term “stem” identifies these cells as the source or origin of other, more specialized cells. There are many stem cell populations in the body at different stages of development. For example, all of the cells of the brain arise from a neural stem cell population in which each cell produces one brain cell and another copy of itself every time it divides. The very earliest stem cells, the immediate descendants of the fertilized egg, are termed embryonic stem cells, to distinguish them from populations that arise later and can be found in specific tissues (such as neural stem cells). These early embryonic stem cells give rise to all the tissues in the body, and are therefore considered “totipotent” or capable of generating all things.

While the existence of early embryonic stem cells has been appreciated for some time, the potential medical applications of these cells have only recently become apparent. More than a dozen years ago, scientists discovered that if the normal connections between the early cellular progeny of the fertilized egg were disrupted, the cells would fall apart into a single cell suspension that could be maintained in culture. These dissociated cells (or embryonic stem cell “lines”) continue to divide indefinitely in culture. A single stem cell line can produce enormous numbers of cells very rapidly. For example, one small flask of cells that is maximally expanded will generate a quantity of stem cells roughly equivalent in weight to the entire human population of the earth in less than sixty days. Yet despite their rapid proliferation, embryonic stem cells in culture lose the coordinated activity that distinguishes embryonic development from the growth of a teratoma. In fact, these early embryonic cells in culture initially appeared to be quite unremarkable: a pool of identical, relatively uninteresting cells.

First impressions, however, can be deceiving. It was rapidly discovered that dissociated early embryonic cells retain the ability to generate an astounding number of mature cell types in culture if they are provided with appropriate molecular signals. Discovering the signals that induce the formation of specific cell types has been an arduous task that is still ongoing. Determining the precise nature of the cells generated from embryonic stem cells has turned out to be a matter of considerable debate. It is not at all clear, for example, whether a cell that expresses some of the characteristics of a normal brain cell in culture is indeed “normal””that is, if it is fully functional and capable of integrating into the architecture of the brain without exhibiting any undesirable properties (such as malignant growth). Nonetheless, tremendous excitement accompanied the discovery of dissociated cells’ generative power, because it was widely believed that cultured embryonic stem cells would retain their totipotency and could therefore be induced to generate all of the mature cell types in the body. The totipotency of cultured embryonic stem cells has not been demonstrated and would, in fact, be difficult to prove. Nonetheless, because it is reasonable to assume embryonic stem cells in culture retain the totipotency they exhibit in embryos, this belief is held by many as an article of faith until proven otherwise.

Much of the debate surrounding embryonic stem cells has centered on the ethical and moral questions raised by the use of human embryos in medical research. In contrast to the widely divergent public opinions regarding this research, it is largely assumed that from the perspective of science there is little or no debate on the matter. The scientific merit of stem cell research is most commonly characterized as “indisputable” and the support of the scientific community as “unanimous.” Nothing could be further from the truth. While the scientific advantages and potential medical application of embryonic stem cells have received considerable attention in the public media, the equally compelling scientific and medical disadvantages of transplanting embryonic stem cells or their derivatives into patients have been ignored.

There are at least three compelling scientific arguments against the use of embryonic stem cells as a treatment for disease and injury. First and foremost, there are profound immunological issues associated with putting cells derived from one human being into the body of another. The same compromises and complications associated with organ transplant hold true for embryonic stem cells. The rejection of transplanted cells and tissues can be slowed to some extent by a good “match” of the donor to the patient, but except in cases of identical twins (a perfect match), transplanted cells will eventually be targeted by the immune system for destruction. Stem cell transplants, like organ transplants, would not buy you a “cure”; they would merely buy you time. In most cases, this time can only be purchased at the dire price of permanently suppressing the immune system.

The proposed solutions to the problem of immune rejection are either scientifically dubious, socially unacceptable, or both. Scientists have proposed large scale genetic engineering of embryonic stem cells to alter their immune characteristics and provide a better match for the patient. Such a manipulation would not be trivial; there is no current evidence that it can be accomplished at all, much less as a safe and routine procedure for every patient. The risk that genetic mutations would be introduced into embryonic stem cells by genetic engineering is quite real, and such mutations would be difficult to detect prior to transplant.

Alternatively, the use of “therapeutic cloning” has been proposed. In this scenario, the genetic information of the original stem cell would be replaced with that of the patient, producing an embryonic copy or “clone” of the patient. This human clone would then be grown as a source of stem cells for transplant. The best scientific information to date from animal cloning experiments indicates that such “therapeutic” clones are highly likely to be abnormal and would not give rise to healthy replacement tissue.

The final proposed resolution has been to generate a large bank of embryos for use in transplants. This would almost certainly involve the creation of human embryos with specific immune characteristics (“Wanted: sperm donor with AB+ blood type”) to fill in the “holes” in our collection. Intentionally producing large numbers of human embryos solely for scientific and medical use is not an option most people would be willing to accept. The three proposed solutions to the immune problem are thus no solution at all.

The second scientific argument against the use of embryonic stem cells is based on what we know about embryology. In an opinion piece published in the New York Times (“The Alchemy of Stem Cell Research,” July 15, 2001) a noted stem cell researcher, Dr. David Anderson, relates how a seemingly insignificant change in “a boring compound” that allows cells to stick to the petri dish proved to be critical for inducing stem cells to differentiate as neurons. There is good scientific reason to believe the experience Dr. Anderson describes is likely to be the norm rather than a frustrating exception. Many of the factors required for the correct differentiation of embryonic cells are not chemicals that can be readily “thrown into the bubbling cauldron of our petri dishes.” Instead, they are structural or mechanical elements uniquely associated with the complex environment of the embryo.

Cells frequently require factors such as mechanical tension, large scale electric fields, or complex structural environments provided by their embryonic neighbors in order to activate appropriate genes and maintain normal gene-expression patterns. Fully reproducing these nonmolecular components of the embryonic environment in a petri dish is not within the current capability of experimental science, nor is it likely to be so in the near future. It is quite possible that even with “patience, dedication, and financing to support the work,” we will never be able to replicate in a culture dish the nonmolecular factors required to get embryonic stem cells “to do what we want them to.”

Failing to replicate the full range of normal developmental signals is likely to have disastrous consequences. Providing some but not all of the factors required for embryonic stem cell differentiation could readily generate cells that appear to be normal (based on the limited knowledge scientists have of what constitutes a “normal cell type”) but are in fact quite abnormal. Transplanting incompletely differentiated cells runs the serious risk of introducing cells with abnormal properties into patients. This is of particular concern in light of the enormous tumor-forming potential of embryonic stem cells. If only one out of a million transplanted cells somehow failed to receive the correct signals for differentiation, patients could be given a small number of fully undifferentiated embryonic stem cells as part of a therapeutic treatment. Even in very small numbers, embryonic stem cells produce teratomas, rapid growing and frequently lethal tumors. (Indeed, formation of such tumors in animals is one of the scientific assays for the “multipotency” of embryonic stem cells.) No currently available level of quality control would be sufficient to guarantee that we could prevent this very real and horrific possibility.

The final argument against using human embryonic stem cells for research is based on sound scientific practice: we simply do not have sufficient evidence from animal studies to warrant a move to human experimentation. While there is considerable debate over the moral and legal status of early human embryos, this debate in no way constitutes a justification to step outside the normative practice of science and medicine that requires convincing and reproducible evidence from animal models prior to initiating experiments on (or, in this case, with) human beings. While the “potential promise” of embryonic stem cell research has been widely touted, the data supporting that promise is largely nonexistent.

To date there is no evidence that cells generated from embryonic stem cells can be safely transplanted back into adult animals to restore the function of damaged or diseased adult tissues. The level of scientific rigor that is normally applied (indeed, legally required) in the development of potential medical treatments would have to be entirely ignored for experiments with human embryos to proceed. As our largely disappointing experience with gene therapy should remind us, many highly vaunted scientific techniques frequently fail to yield the promised results. Arbitrarily waiving the requirement for scientific evidence out of a naive faith in “promise” is neither good science nor a good use of public funds.

Despite the serious limitations to the potential usefulness of embryonic stem cells, the argument in favor of this research would be considerably stronger if there were no viable alternatives. This, however, is decidedly not the case. In the last few years, tremendous progress has been made in the field of adult stem cell research. Adult stem cells can be recovered by tissue biopsy from patients, grown in culture, and induced to differentiate into a wide range of mature cell types.

The scientific, ethical, and political advantages of using adult stem cells instead of embryonic ones are significant. Deriving cells from an adult patient’s own tissues entirely circumvents the problem of immune rejection. Adult stem cells do not form teratomas. Therapeutic use of adult stem cells raises very few ethical issues and completely obviates the highly polarized and acrimonious political debate associated with the use of human embryos. The concern that cells derived from diseased patients may themselves be abnormal is largely unwarranted. Most human illnesses are caused by injury or by foreign agents (toxins, bacteria, viruses, etc.) that, if left untreated, would affect adult and embryonic stem cells equally. Even in the minority of cases where human illness is caused by genetic factors, the vast majority of such illnesses occur relatively late in the patient’s life. The late onset of genetic diseases suggests such disorders would take years or even decades to reemerge in newly generated replacement cells.

In light of the compelling advantages of adult stem cells, what is the argument against their use? The first concern is a practical one: adult stem cells are more difficult than embryonic ones to grow in culture and may not be able to produce the very large numbers of cells required to treat large numbers of patients. This is a relatively trivial objection for at least two reasons. First, improving the proliferation rate of cells in culture is a technical problem that science is quite likely to solve in the future. Indeed, substantial progress has already been made towards increasing the rate of adult stem cell proliferation. Second, treating an individual patient using cells derived from his own tissue (“autologous transplant”) would not require the large numbers of cells needed to treat large populations of patients. A slower rate of cell proliferation is unlikely to prevent adult stem cells from generating sufficient replacement tissue for the treatment of a single patient.

The more serious concern is that scientists don’t yet know how many mature cell types can be generated from a single adult stem cell population. Dr. Anderson notes, “Some experiments suggest these [adult] stem cells have the potential to make mid-career switches, given the right environment, but in most cases this is far from conclusive.” This bothersome limitation is not unique to adult stem cells. Dr. Anderson goes on to illustrate that in most cases the evidence suggesting scientists can induce embryonic stem cells to follow a specific career path is equally far from conclusive. In theory, embryonic stem cells appear to be a more attractive option because they are clearly capable (in an embryonic environment) of generating all the tissues of the human body. In practice, however, it is extraordinarily difficult to get stem cells of any age “to do what you want them to” in culture.

There are two important counterarguments to the assertion that the therapeutic potential of adult stem cells is less than that of embryonic stem cells because adult cells are “restricted” and therefore unable to generate the full range of mature cell types. First, it is not clear at this point whether adult stem cells are more restricted than their embryonic counterparts. It is important to bear in mind that the field of adult stem cell research is not nearly as advanced as the field of embryonic stem cell research. Scientists have been working on embryonic stem cells for more than a decade, whereas adult stem cells have only been described within the last few years. With few exceptions, adult stem cell research has demonstrated equal or greater promise than embryonic stem cell research at a comparable stage of investigation. Further research may very well prove that it is just as easy to teach an old dog new tricks as it is to train a willful puppy. This would not eliminate the very real problems associated with teaching any dog to do anything useful, but it would remove the justification for “age discrimination” in the realm of stem cells.

The second counterargument is even more fundamental. Even if adult stem cells are unable to generate the full spectrum of cell types found in the body, this very fact may turn out to be a strong scientific and medical advantage. The process of embryonic development is a continuous trade-off between potential and specialization. Embryonic stem cells have the potential to become anything, but are specialized at nothing. For an embryonic cell to specialize, it must make choices that progressively restrict what it can become. The greater the number of steps required to achieve specialization, the greater the scientific challenge it is to reproduce those steps in culture. Our current understanding of embryology is nowhere near advanced enough for scientists to know with confidence that we have gotten all the steps down correctly. If adult stem cells prove to have restricted rather than unlimited potential, this would indicate that adult stem cells have proceeded at least part way towards their final state, thereby reducing the number of steps scientists are required to replicate in culture. The fact that adult stem cell development has been directed by nature rather than by scientists greatly increases our confidence in the normalcy of the cells being generated.

There may well be multiple adult stem cell populations, each capable of forming a different subset of adult tissues, but no one population capable of forming everything. This limitation would make certain scientific enterprises considerably less convenient. However, such a restriction in “developmental potential” would not limit the therapeutic potential of adult stem cells for treatment of disease and injury. Patients rarely go to the doctor needing a full body replacement. If a patient with heart disease can be cured using adult cardiac stem cells, the fact that these “heart-restricted” stem cells do not generate kidneys is not a problem for the patient.

The field of stem cell research holds out considerable promise for the treatment of disease and injury, but this promise is not unlimited. There are real, possibly insurmountable, scientific challenges to the use of embryonic stem cells as a medical treatment for disease and injury. In contrast, adult stem cell research holds out nearly equal promise while circumventing the enormous social, ethical, and political issues raised by the use of human embryos for research. There is clearly much work that needs to be done before stem cells of any age can be used as a medical treatment. It seems only practical to put our resources into the approach that is most likely to be successful in the long run. In light of the serious problems associated with embryonic stem cells and the relatively unfettered promise of adult stem cells, there is no compelling scientific argument for the public support of research on human embryos.

Maureen L. Condic , a new contributor, is assistant professor of neurobiology and anatomy at the University of Utah, working on the regeneration of adult and embryonic neurons following spinal cord injury.