PCBE: White Paper on Alternative Sources of Pluripotent Stem Cells -- Full Report

Human embryonic stem cells hold great interest because of their pluripotency-their capacity to give rise to the various specialized cells of the body-and because of their longevity-their ability to be propagated for many generations in laboratory culture without losing their pluripotency. Until now, these cells have been obtainable only from living human embryos [at the 100-to-200-cell (blastocyst) stage of development] by a process that necessarily destroys the embryos and that therefore makes this research ethically controversial. Over the past several years, the ethical controversy has been the subject of federal (and state) legislation and public policy and of ongoing public debate.ⁱ

The President's Council on Bioethics is committed to the goals of advancing biomedical science and upholding ethical norms. Notwithstanding our sometimes sharp individual ethical differences, we have recognized that all parties to the debates about embryo research have something vital to defend, and not only for themselves but for all of us.¹ As members of a national public bioethics body, we are also mindful of the need to understand and respect the strongly held ethical views of our fellow citizens, even when we do not share them. For these reasons, we must be receptive to any creative scientific or technical suggestions that might enable scientists to proceed with their research in ways that would not raise ethical questions or violate the ethical principles of many Americans.

Accordingly, in an effort to find ethically uncontroversial ways to advance human embryonic stem cell research, the Council has recently been looking into specific proposals for obtaining pluripotent, genetically stable, and long-lived human stem cells by methods that would meet the moral standard of not destroying or endangering human embryos in the process. This White Paper introduces these proposals and begins an analysis of their strengths and weaknesses, ethical, scientific, and practical. Because the scientific and practical merits of these proposals are in large part empirical matters, not settled in advance by mere speculation, we give special weight to the ethical analysis. We also explore, in a preliminary way, whether these alternative avenues of deriving and using pluripotent stem cells are likely to be embraced by scientists or to become eligible for federal funding.

Conceptually, four broad approaches present themselves. The stem cells could be derived: (1) by extracting cells from embryos already dead; or (2) by non-harmful biopsy of living embryos; or (3) by extracting cells from artificially created non-embryonic but embryo-like cellular systems (engineered to lack the essential elements of embryogenesis but still capable of some cell division and growth); or (4) by dedifferentiation of somatic cells back to pluripotency. In each of these four cases, the scientific standard by which success should be measured is only the desired functional capacity of the cells derived-stable pluripotency-and not their origin (embryos, adults, or artificial embryo-like clusters of cells). Should stem cells obtainable by one or another of these methods turn out to have exactly the same properties and capacities as embryonic stem cells (ESCs), their value for scientific research should be no different from that of standard ESCs.

Recently, more or less detailed examples of each of these four approaches have been proposed or discussed.

According to the first proposal, pluripotent human stem cells are to be derived from early IVF embryos (roughly 4-8 cells) that have spontaneously died (as evidenced by the irreversible cessation of cell division) but some of whose blastomeresⁱⁱ appear normal and healthy. Crucial to this approach is (a) enunciating a concept of organismic death of an early embryo and (b) devising criteria that permit a determination that embryonic death has occurred. In addition, to satisfy the moral standard, only those once-frozen embryos that are thawed and that die spontaneously during efforts to produce a child will be eligible for post-mortem cell extraction. This proposal was presented at the Council's December 3, 2004, meeting by Drs. Donald Landry and Howard Zucker of the Columbia University College of Physicians and Surgeons.

According to the second proposal, pluripotent stem cells are to be derived from blastomeres obtained by biopsy of an early human embryo. Crucial to this approach is finding a stage of early embryonic development at which (a) the removal of one or a few cells by biopsy can be carried out without harming the embryo, while (b) the cell or cells removed from the embryo are usable as a source of pluripotent stem cells.

The third approach comprises a variety of proposals for engineering "biological artifacts" possessing some of the developmental capacities of natural embryogenesis (but lacking the organismal character of human embryos) and containing cells from which pluripotent stem cell lines can be derived. Crucial to this approach is demonstrating both (a) that the developing entity is truly not a human embryo and (b) that the cells derived from it are in fact normal human pluripotent cells. In addition, one must show that creating such biological artifacts does not itself introduce other ethical problems. One such proposal ("Altered Nuclear Transfer") was presented at the Council's December 3, 2004, meeting by Council Member Dr. William Hurlbut.

The fourth proposal involves reprogramming human somatic cells, perhaps with the aid of special cytoplasmic factors obtained from oocytes (or from pluripotent embryonic stem cells), so as to "dedifferentiate" them back into pluripotent stem cells. Crucial to this approach is discovering a way to reverse cell differentiation all the way to pluripotency, but not (as in cloning) even further back to totipotency.ⁱⁱⁱ

This White Paper describes each of these proposals and examines them with a view to the following three questions:

Ethical Question: Is it ethically sound? That is, would the proposed method overcome the ethical objections that have been raised against current methods of deriving embryonic stem cells that entail the destruction of human embryos? And does the proposed method raise new ethical difficulties of its own that make it problematic, worrisome, or even unacceptable?^iv

Feasibility Question: Is it scientifically sound? That is, might it reliably produce stable, pluripotent stem cell lines of sufficient quality for biomedical research and, in due course, for clinical trials in human beings?

Practical Question: From the perspective of both public policy and research practice, is this proposal "realistic"? That is, are there good reasons for believing that, if it is found to be scientifically feasible, it might be adoptable-by scientists as useful, by policy makers as legally eligible for federal support, and by the people and their elected officials as morally worthy of federal support? (In considering these questions, as we will in each case below, we are not to be understood either (a) to be making predictions about what scientists will in fact believe or (b) to be offering recommendations for what policy makers should in fact do. The first must await the arrival of relevant options; the second must depend on an assessment of the ethical claims and on prudential judgments to be made in specific circumstances that we cannot predict. Still, practical people will want to know whether any of this could become useful scientific practice or wise public policy, and we are obliged to discuss some aspects of these questions.)^v

I. Pluripotent Stem Cells Derived from Organismically Dead Embryos (Landry-Zucker Proposal²)

This proposal begins by making a close analogy with the use of human cadavers for biomedical research or as sources of organs. Today, we find morally and socially acceptable the removal (with consent) of vital organs from no-longer-living developed human beings once they have been declared dead.³ Therefore, Landry and Zucker suggest, we should also find morally and socially acceptable the removal (with consent) of materials for stem cell derivation from no-longer-living undeveloped human beings (human embryos) once they have been declared dead. Applying the traditional concept of death-the irreversible loss of the integrated functioning of an organism as a whole-to the earliest stages of human life, Landry and Zucker propose a concept of organismic death for the early-stage human embryo: the irreversible loss of the capacity for "continued and integrated cellular division, growth and differentiation."^vi As the criterion for determining organismic death of an embryo produced by in vitro fertilization (IVF), they propose the "irreversible cessation of cell division in the embryo observed in vitro."

After fertilization in vitro, a high percentage of human embryos that reach the 4- or 8-cell stage undergo spontaneous "cleavage arrest"-that is, their cells simply stop dividing. The vast majority of these arrested embryos do not resume cell division, never form blastocysts, and are incapable of successfully implanting in the uterus. In most cases, spontaneous cleavage arrest is associated with chromosomal abnormalities in the cells of the developing embryos. Yet some of the arrested embryos turn out to be "mosaic"-that is, some of their cells exhibit chromosomal abnormalities, while others appear to be (chromosomally, at least) normal blastomeres. It is these normal-appearing blastomeres in cleavage-arrested, mosaic embryos that may turn out to be a source of embryonic stem cells. Landry and Zucker propose that those embryos that have undergone irreversible cleavage arrest should be declared organismically dead and hence suitable (with proper consent) for harvesting of blastomeres for stem cell derivation.^vii

To identify when organismic death occurs in IVF embryos, Landry and Zucker propose a two-part research strategy. First, they propose a "natural history" study to determine just when a non-flourishing embryo in vitro ceases cell division irreversibly:

Previously frozen early embryos that have failed to divide within 24 hours of thawing and are [therefore] no longer wanted [because they are no longer fit] for their original reproductive purpose are observed every few hours for several additional 24-hour periods.^viii After observing several hundred embryos, the time beyond which no arrested embryo resumes division can be determined. One can reasonably conclude that embryos that have not divided by this period will not divide at any later time, i.e., they are organismically dead.

Second, they propose an experimental study to attempt to identify physical or biochemical cellular markers that correlate with the arrest of cell division and in whose presence any arrested embryo could be declared dead:

[IVF] Embryos declared [organismically] dead could then be characterized for secreted or cell surface markers or spectroscopic signatures that correlate with the arrest of cell division. These markers and signatures could then be tested for their predictive value. In this manner the criteria for determining the death of a human embryo could be refined.

According to Landry and Zucker, determining criteria of organismic death through the natural history study would be sufficient to commence efforts to derive human embryonic stem cells from dead embryos; the subsequent experimental study could facilitate this effort by rendering the determination of embryonic death more certain and reliable.

The Landry-Zucker proposal is based on an attractively simple ethical idea: it should be permissible to harvest cells from embryos that have died (provided of course that their deaths have not been caused or hastened for such purposes). Yet the final ethical assessment of this proposal will depend very much on exactly how it is implemented, within the practices and protocols of IVF. Several other pertinent questions have been raised as well.

1. Can we be certain that the IVF embryos used in this proposal are really dead?

This challenge unites two separate concerns, one about knowing whether the criteria for death have been met in any particular case, the other about the criteria themselves.

Landry and Zucker propose to harvest cells from 4- or 8-cell thawed human embryos that have irreversibly ceased to divide. Since the harvested cells are to serve as a source of stem cells, it is not feasible to wait for the death and dissolution of each and every cell before declaring the embryo dead and eligible for use in research. To be useful for the project, the arrested embryos must contain at least some viable cells that retain normal developmental potential. Some of these cells, for example, might resume dividing if extracted and placed in the proper milieu. How, then, can we be sure that such an embryo is really dead? More generally, can we confidently declare that an embryo is dead just because all of its cells have stopped dividing? What exactly do we mean by the "organismic death" of an embryo?

It must be acknowledged that the concept of organismic death-the death of an organism as a whole-has not been commonly applied to an embryo, that is, to any largely undifferentiated organism so close to the beginning of its life. Moreover, even granting the applicability of the concept, identifying the criteria for determining organismic death for an embryo is, at least for now, more difficult than it is for an adult, owing to the absence of any integrating vital organs. Unlike a person at the end of life, an embryo has no identifiable controlling organ-no brain to be considered "brain dead," no heart that has ceased to beat and circulate the blood. For now, our judgment that an embryo has lost "integrated function as an organism" must be based simply on the observed absence of coordinated cell division. It is partly for that reason that Landry and Zucker hope that studies might reveal biochemical markers that would put the phenomenological finding of irreversible cleavage arrest on a sounder, "more objective" footing.

Yet despite these uncertainties, as Landry and Zucker suggest, the life of an embryo (as of any organism) is more than just the sum of the lives of its constituent cells; and the death of an embryonic organism-as-a-whole is not the same thing as the death of its constituent cells. For the embryo to develop as an integrated organism, there must be an integrated set of internal signals directing the differentiation, development, and growth of the multicellular organism as a whole. The breakdown or absence of that set of signals amounts to the death of the developing embryo, manifested most clearly in the irreversible cessation of coordinated, organized cell division. Just as a person can be declared dead even while some of his organs and cells continue for a time to function and grow, so an embryo can undergo organismic death even while some of its individual cells may remain alive as cells, capable of further division if isolated and placed in a suitable environment.

Landry and Zucker point to studies showing that a substantial proportion of IVF embryos, after thawing, never exhibit any further cell division, even though some of their blastomeres appear otherwise normal. It is these embryos, they suggest, that can be unambiguously declared "organismically dead," and whose normal-looking blastomeres may be suitable candidates for embryonic stem cell derivation.^ix

The Landry-Zucker proposal involves scrutinizing thawed IVF embryos for signs of death and extracting cells from organismically dead embryos. A reasonable concern is that this procedure should not expose any living human embryos to risks they would not ordinarily encounter in the practice of assisted reproduction. Sharing this concern, Landry and Zucker place special strictures on which embryos are to be used, both in natural history studies to determine the criteria of death and in subsequent stem cell derivations. The only embryos that would be considered for use would be those (1) that were originally created with reproductive intent, (2) that were thought healthy enough to be kept alive in cryostorage for possible second or third child-producing attempts, and (3) that, after thawing, turned out, alas, to be dead. The natural history research proposed by Landry and Zucker would simply continue to watch those embryos that were not making any developmental progress, in order to determine more precisely exactly when they were irreversibly incapable of further development. No living embryo would be subject to manipulative intervention or to any procedure that increased its exposure to harm. No new or extra embryos would be created for such research, and no embryo would be deliberately killed or otherwise exposed to harm. This "death watch" study proposed by Landry and Zucker seeks only to discover which embryos are already dead, not to induce weak or doomed embryos to die.

The proposed experiments to determine the natural history of organismic death in IVF embryos should be done in a way that does not compromise the safety of, or make more physically or psychologically onerous, current IVF clinical procedures. They also should not change incentives and practices regarding embryo production. Some people worry that approving the use of dead embryos for stem cell derivation will lead to the creation of even more embryos than are now produced in excess of reproductive need, precisely so that some could be allowed to die for the sake of getting stem cells from them. Yet it is important to note that, under the Landry-Zucker proposal, embryos that divide normally upon thawing, but are allowed to die by a human decision (not to transfer them into a woman's uterus), would not be eligible for donation. Precisely in order to avoid participation or complicity in the death of any embryos, their proposal restricts use to only those embryos that fail altogether to divide upon thawing and that have thus died "on their own." For now, no change of practice or incentives would be likely or necessary, since many embryos are thawed for a second reproductive trial and, of these, a sizeable fraction (in some cases, close to one half) fail to develop.⁴ Still, going forward it will be important to provide oversight and assurance that the desire for material useful for the natural history study or for stem cell derivation does not increase either the number of frozen embryos that are thawed in an attempt to produce a second pregnancy or the number created for reproductive purposes in the first place. It would be an ethically dubious innovation if the implementation of the Landry-Zucker proposal were to change the incentives or the practices of IVF embryo creation or storage in these ways.^x

Additional discussions with the embryo-producing, child-seeking patients that explain (and seek consent for) the proposed experiment and use will be required. And a fitting informed-consent form will have to be developed and approved. While this is well within the scope of present practice, it will necessarily involve discussion of the likely death of some embryos created by the IVF procedure. Many clinicians shy away from using the word "death" to describe what happens to the embryos that do not develop in vitro, fearful that such a designation would imply that those embryos were in fact alive and that they might therefore be held at fault for their resulting deaths. Nevertheless, only a frank discussion of these facts could produce meaningful consent from the patients. Discussion and suitably documented agreements will also be needed to address additional questions, including who owns the commercial rights to any human pluripotent stem cell line created by this research.

5. For ethical purposes, is there a sufficiently strong analogy between harvesting cells from dead embryos and harvesting organs or removing tissues from dead persons?

Landry and Zucker base the ethical justification of their proposal on the analogy with end-of-life organ donation. Yet the analogy is not altogether exact: unlike the physician caring for a dying patient, the IVF clinician does not in general treat the death of an embryo as a grave matter. Certain standard IVF procedures, including superovulation and cryopreservation of excess embryos, knowingly increase the likelihood that many individual embryos will die or be discarded;⁵ but this is not generally considered a reason to forgo production and freezing of "extra" embryos. Some observers, troubled by these aspects of the clinical context in which this proposal would be carried out and disinclined to benefit by complicity in these practices, do not find the analogy with organ donation sufficiently compelling to justify the extraction of cells from dead IVF embryos, especially when the intended use of the extracted cells is scientific research rather than the immediate use of organs to save dying patients.

Nonetheless, an exact analogy with organ donation is not required to show that what Landry and Zucker are proposing is what they claim it to be: a morally preferable alternative to the intentional destruction of embryos. Death comes spontaneously to many embryos, both in vivo and in vitro, and it is difficult to see how dissecting spontaneously dead embryos can be said to harm them. And if the principle "Once dead, then usable"-of course, with informed consent and showing respect for the corpse-works ethically for removing transplantable organs or research materials from dead adult human beings, it should work equally well for dead human embryos (provided, again, that the embryos are indeed dead and that their prospective users have not deliberately killed them or neglected them so that they would die). In the end, whether this proposal proves to be morally acceptable and practically wise remains to be seen.

The Landry-Zucker proposal has yet to be tested, though it is technically possible to begin testing it immediately, not only in animals but also in humans. Three basic questions need to be answered: Can objective markers of organismic death be found? Can pluripotent stem cells be derived from dead embryos? If so, will they be chromosomally (and otherwise) normal?

It seems likely that the natural history study will identify "duration without cleavage" as one objective marker of embryonic organismic death, but the exact criterion has yet to be established.^xi Whether additional biochemical markers can be identified is not yet known, but, according to Landry and Zucker, their discovery is not essential for the immediate implementation of the proposal. Ultimately, the issue of markers is largely an empirical question: one will not know the answer until the effort to find them is made.

Once embryos in vitro have been determined to be organismically dead, a basic scientific question, still unanswered, is whether pluripotent stem cells can then be derived, starting from any remaining blastomeres. There is reason to believe that some cells in arrested embryos may retain their developmental potential, which can be reactivated by transferring them to the appropriate milieu; but the evidence is very preliminary.⁶ In addition, recent work by Dr. Nicolai Strelchenko and colleagues (working at Chicago's Reproductive Genetics Institute headed by Dr. Yury Verlinsky) has described the production of human pluripotent stem cells derived by culturing blastomeres removed from morula-stage (8-24-cell) human embryos.^xii ⁷ The next step would be to show that stem cells can be derived from single blastomeres extracted from 8-cell embryos.^xiii It would then have to be shown that similar results can be obtained using blastomeres extracted from organismically dead IVF embryos.

Questions have been raised regarding whether pluripotent stem cell lines isolated from organismically dead IVF embryos would be abnormal, and in particular, aneuploid (that is, having more or fewer than the normal number of chromosomes).⁸ The answer cannot be given in advance, but there are reliable methods for determining it in every case. Each isolated pluripotent cell line would be grown in vitro, so that a detailed study could subsequently be done on the chromosome complement of each cell line. Such testing would identify those pluripotent stem cell lines with a normal chromosome complement; repeated karyotype testing throughout the period of laboratory culture and storage could confirm that the chromosome complement remains normal. Moreover, while pluripotent stem cell lines with a normal chromosome complement would have the broadest potential therapeutic applicability, pluripotent stem cell lines with specific abnormal chromosome complements (for example, three copies of chromosome 21, as is observed in people with Down syndrome) could be useful in basic studies of how (in this example) the presence of the extra chromosome affects differentiation processes that subsequently lead to a human genetic disease.

Two kinds of practical questions have been raised: First, will scientists want to work with these cells? And, second, will the research be supportable by federal funding, under the legislative and administrative restrictions now in place?

Scientists understandably want to work only with the best materials. Why, it is asked, would they settle for cells derived from dead embryos, especially since embryos that die early are generally abnormal, either chromosomally or in other ways?^xiv And why should they bother trying to develop these cells lines, when they can use existing ESC lines or derive new ones at will from living IVF ("spare" or newly created) embryos? One answer is that they would welcome such cell lines if research using them were eligible for federal funding. But a better answer, and on the main question, is this: there is simply no way to know in advance whether cells derivable from dead embryos are in fact in any way inferior to cells derivable from still living blastocysts. One must do the experiment and see.

The Landry-Zucker proposal aims to provide a basis for future attempted isolations of pluripotent human stem cells by a procedure in which no embryos are killed for the purposes of research. Only embryos that are found to have died in the context of standard IVF clinical procedures would be used in attempts to produce the stem cells. Such experiments would ordinarily be considered human tissue research and require local IRB approval. After that, most states would permit the research.⁹ It is somewhat more difficult to determine whether experiments to evaluate and implement the Landry-Zucker proposal would be eligible for federal funding under current law and policy. Two federal policies are particularly relevant: the Dickey Amendment¹⁰ and President Bush's embryonic stem cell policy statement of August 9, 2001.¹¹ The purpose of the Dickey Amendment is to deny federal funds for any experiment in which living human embryos are killed or harmed, while the intent of the President's policy is to promote embryonic stem cell research without sanctioning or encouraging future destruction of human embryos^xv; both policies will have to be re-examined in the context of organismically dead IVF embryos. Assuming Landry-Zucker criteria for embryonic death are established and clinically met, it would appear that federal funding for the further manipulation of such dead embryos would not violate either the letter or the spirit of the Dickey Amendment.^xvi It is less clear whether the same could be said of the natural history studies to determine the precise criteria for embryonic organismic death, especially because federal funding has never been available for IVF research or practice or any treatment whatsoever of ex vivo embryos.^xvii Nevertheless, a strong argument can be made that the mere observation of embryos that had spontaneously ceased to divide hardly constitutes doing them harm or causing their death. Whether the President's policy and budget for federal funding of stem cell research would-or should-be expanded to support either research to derive stem cells from dead embryos or research on stem cell lines already derived from dead embryos with private support is a question whose answer will depend not only on the issue of legal eligibility but also on the assessment of other ethical, scientific, and prudential considerations of the sort we have just discussed.

Pluripotent stem cell lines could, in theory, be derived starting from small numbers of cells ("blastomeres") removed from living human embryos. Is there a stage of early human embryonic development at which cells, capable of developing in vitro into pluripotent stem cells, can be extracted without harming the embryo's prospects for developing into a live-born child?^xviii

As indicated above, Dr. Nicolai Strelchenko and his colleagues have shown that embryonic stem cells can be derived from human embryos containing 8-24 cells (see reference 7). In their method, all the cells of the embryo are disaggregated and cultured on feeder cells (and the embryo is killed in the process). It may be some time before stem cell lines can be reliably derived from single cells extracted from early embryos, and in ways that do no harm to the embryo thus biopsied. But the initial success of the Verlinsky group's efforts at least raises the future possibility that pluripotent stem cells could be derived from single blastomeres removed from early human embryos without apparently harming them. ^xix

With the Landry-Zucker proposal, the major ethical issue concerned the question of whether the embryos would in fact be truly dead. Here, the major ethical issue concerns the question of possible harm (and perhaps also benefit) to the still-living embryo whose cells are removed. Removal of blastomeres from developing IVF embryos in vitro is currently done primarily in the context of avoiding pregnancies at risk for genetic disease. Toward that end, the PGD techniques are used to test a group of embryos with a view to identifying those embryos that can be transferred to the woman without carrying known markers for genetic disease. (The embryos that do carry the abnormal genes are discarded.) Strictly speaking, embryo biopsy as currently practiced in PGD cannot be said to be undertaken for any future child's benefit, since the procedure does not directly help those embryos that are ultimately implanted. Also, the genetically healthy embryos that are transferred to initiate a pregnancy will have been subjected to the as-yet-unknown risks of the blastomere biopsy procedure. For many individuals and couples, the known short-term and potential long-term risks of the PGD technique are thought to be more than balanced by the desire of the couple to have their own biological child free from a specific genetic disease. Others believe this practice is unethical, since it involves discriminating against genetically disabled embryos and ultimately discarding them. There also remains substantial debate about the ethical propriety of using PGD in two specific cases: (1) to identify embryos that would give rise to children who could serve as compatible bone marrow donors for sick siblings,¹⁴ and (2) to determine the sex of the embryos in order to be sure that only embryos of the desired gender were transferred.¹⁵

How would the ethical analysis change if living embryo blastomere extraction were to be performed not for PGD but for stem-cell derivation? Assuming that single blastomeres extracted from early embryos could in fact be used to derive pluripotent human stem cells, would this procedure pass ethical scrutiny? Since the removal of a cell or two from the embryo is not (usually) fatal, the individual embryo that is biopsied is not killed. However, since the blastomere extraction is not being performed for the good of the embryo, it might be hard to justify the procedure ethically. As Dr. Gerald Schatten told the Council in December 2002, "Embryo biopsy is a complicated technique, and it's a very expensive technique, and it's not clear that it is completely innocuous. So you would not go into embryo biopsy unless there were compelling reasons for actually going through all of the costs and expense and heroics of ART [assisted reproductive technologies]."¹⁶ And even when prospective parents do elect to use ART, subjecting otherwise healthy embryos to biopsy procedures in order to derive stem cells seems ethically troubling. Indeed, even if the biopsied embryo and the resulting child were not physically harmed, a strong line of moral argument might still lead one to object on the grounds that the embryo is being treated merely as a means to another's ends. For there are more ways to do injustice to another human being than by actions that do discernable or manifest harm. Using human beings for purposes of no benefit to them and without their knowing consent is one such injustice, even if doing so results in no evident or eventual harm to body or psyche.

Such worries or objections might be moderated should the blastomere removal be undertaken, at least in part, for the possible benefit of the future child. As noted earlier, in embryo biopsy for PGD, sometimes two blastomeres are removed. This raises the futuristic possibility that one cell could be used for genetic diagnosis, while the other cell is used to derive a line of stem cells genetically autologous to the embryo and the child it becomes. Looking still farther ahead, patients using IVF without concern for genetic disease (and hence not interested in ordinary PGD) might nonetheless consider blastomere removal solely for the purpose of deriving immunologically compatible stem cells for their future child. Some might consider such a practice ethically justified on the grounds that the child born after uterine transfer of the embryo might later derive medical benefit from the existence of a genetically matched line of pluripotent stem cells, stored in case he needs it for future disease therapy. Others, however, doubt the wisdom of exposing the prospective child (while an embryo) to a hazardous procedure merely for the sake of some hypothetical future benefit, or of encouraging parents to practice embryo biopsy simply or mainly as a source of "personalized" stem cells should their future child someday have need of them.¹⁷ Besides, genetically matched stem cells can be more effectively derived using the newborn's umbilical cord blood (a well-established procedure), though it is unclear whether the stem cells isolatable from cord blood will have all the same capacities as embryonic stem cells.

2. Are the removed blastomeres not themselves the equivalent of embryos or capable of developing into them?

Another possible source of ethical concern has to do with the totipotency of early-stage human blastomeres in vivo.¹⁸ After the first cleavage of the fertilized human egg in vivo, both resulting blastomeres are capable of forming a complete embryo that grows into a child.xx It is not certain at what point in embryonic development in vitro (as in IVF) such totipotency of the blastomeres disappears; it may be that, by the 8-cell stage, sufficient differentiation has taken place that individual human blastomeres are no longer individually totipotent without aggregating them or combining them with other early embryos. Clearly, however, if the blastomere removed for biopsy has the potential to develop into an embryo and a child on its own, some would find destruction of that blastomere ethically objectionable. And, in any case, little would have been gained ethically if the goal of the entire enterprise was a non-controversial procedure for deriving stem cells that did so while avoiding destruction of living embryos.

3. May one perform non-harmful blastomere extraction on embryos that are NOT going to become children?

If embryo biopsy proves to be a usable source of pluripotent stem cells, some might argue that it would be ethically permissible to carry out such a procedure-not fatal and perhaps not harmful at all-not only on embryos that were soon to be transferred to a woman, but also on IVF "spare" embryos that are not ultimately selected for uterine transfer. Indeed, because of the still unknown risk of harm from blastomere removal to the child subsequently emerging from a biopsied embryo, some have suggested that the biopsy procedure can be ethically done only on an embryo that is definitely not going to become a child. Others, however, consider any proposed utilitarian treatment of such embryos to be morally unacceptable, since it necessarily classifies "spare" (and still living) embryos as ethically available for research uses.

4. Can the research necessary to test this proposal be conducted in an ethically acceptable manner?

Learning how to implement this proposal-even for the variation that would only seek stem cells that might eventually benefit the child whose embryonic beginning was biopsied to obtain them-has its own ethical difficulties. Techniques would have to be developed for deriving pluripotent stem cells, not only from whole early-stage human embryos, but also from individual blastomeres extracted from an embryo. It may prove difficult to develop and refine those techniques without exposing many human embryos to death and injury.^xxi If embryo biopsy is to be embraced as a morally uncontroversial way to derive stem cells, the research needed to test the proposal and perfect the technique would have to avoid killing or harming human embryos. It is far from clear that the necessary research can be accomplished with this restriction in place. Even if a perfected technique could someday derive stem cells from single blastomeres harmlessly obtained, the failure to satisfy this testing-stage ethical requirement could render this option ethically little better than the currently controversial methods for deriving embryonic stem cells.

There is ethically more at stake in this proposal than the fate of individual biopsied embryos. There are also large issues raised by the direct intrusion of research objectives into the practice of reproductive medicine, while it is being practiced (that is, in the moments when decisions about embryo transfer are still being made). The Landry-Zucker proposal would make available for stem cell derivation only those embryos that had already died. In contrast, this embryo-biopsy proposal would manipulate still-living embryos immediately destined for reproductive transfer. From the perspective of the would-be parents, is it really better for them (and for their child-to-be) if those performing PGD are concerned not only with good diagnosis but also with procuring useful cells for a research colleague? Decisions about how many and which embryos to transfer are often made in stressful circumstances, where timing is critical and the doctor's attention is limited. Bringing scientists (or considerations of research) into this process, for reasons having nothing to do with the well-being of parents and their children-to-be, would seem to be a dubious intrusion. It would also make assisted reproduction seem even more like manufacture, a process with many side uses and side benefits, rather than simply a way to help people have children. To say the least, much careful planning and oversight would be needed to prevent the research interest from adversely affecting the way reproductive medicine is practiced and the meaning it has for the larger society.

The recent work of Strelchenko and colleagues with disaggregated 8-24-cell embryos suggests that whole human embryos as early as the 8-cell stage are potentially usable as a source of pluripotent stem cells. This work would have to be reproduced and refined before we could be certain that embryos at such an early stage are indeed a dependable source of stem cells. It would then have to be shown that stem cells can also be derived from isolated blastomeres extracted from an 8-cell embryo.^xxii It seems far from certain that enough cells can be extracted from the embryo to derive stem cells while also avoiding injury to the embryo.

On the question of whether biopsy can be safely performed on early stage embryos, there is mixed evidence from animal studies. Krzyminska and colleagues, working with mouse embryos, found that "biopsy had the least impact when performed at the 8-cell stage." When performed on 8-cell embryos, they found that biopsy did not significantly impair development in vitro or the rate of implantation after transfer; but compared to intact embryos, fewer biopsied embryos (52% versus 71%) resulted in viable fetuses.¹⁹ Because human PGD is such a novel, still small, and as-yet-unstudied practice, we do not have good data for the implantation rate for biopsied-as compared with non-biopsied-human embryos.

In any event, since apparently healthy children have been born after embryo biopsy at the 8-cell stage,^xxiii it would appear to be possible to safely remove one or two blastomeres from an 8-cell embryo in order to try to generate a line of pluripotent human stem cells. Only further research and effort can settle the matter.

Let us assume that a stage of human embryonic development can be identified at which cells can be removed without injuring the still-living embryo, that a non-injurious procedure for removing the cells is perfected, and that the cells can then be used to derive pluripotent stem cells. Will scientists want to work with these cells? And will the stem cell research be supportable by federal funding, that is, both eligible under the legislative restrictions and embraceable by administrative policies now in place?

Whether this approach is likely to be adopted by scientists in the future as a way to produce pluripotent human stem cell lines depends on several unknowns, central among them the efficiency of the process (for example, how many good stem cell lines are obtainable from how many biopsied embryos, and at what cost of effort and expense). A further crucial consideration will be the properties of the pluripotent stem cells that are produced by blastomere disaggregation of early embryos. If the resulting pluripotent stem cells turned out to be as good as or better than the current human embryonic stem cell lines (derived from the inner cell mass of blastocysts, a later stage of embryonic development), prospects for using this approach would improve. If, however, the proposal were deemed ineligible for federal funding, the prospects for this approach would not look very good, given that there are several other well-established methods for producing human embryonic stem cells.

An argument could be advanced that this approach complies with the Dickey Amendment, as long as the embryos biopsied are truly not harmed. But there is today insufficient evidence to determine whether biopsied embryos are, by virtue of the procedure or the removal of cells, in fact at risk of harm. Thus, although any derived pluripotent cell lines might be eligible for funding, doing the prior research to derive them might not be. A second obstacle facing such prior research would be the longstanding Congressional opposition to all federal funding of any activities involving in vitro fertilization. Whether the President's policy and budget for federal funding of stem cell research would-or should-be expanded to support research on stem cell lines derived (with private support) from biopsied embryos is a question whose answer will depend not only on the issue of legal eligibility but also on the assessment of other ethical, scientific, and prudential considerations of the sort we have just discussed.

Under this heading are various proposals to construct a biological artifact, lacking the moral status of a human embryo, from which pluripotent stem cells could then be derived. For example, Council Member William Hurlbut has advocated what he calls "altered nuclear transfer" (ANT), a procedure that, if successful, would offer a way to produce pluripotent stem cells within "a limited cellular system that is biologically and morally akin to a complex tissue culture."²⁰ This proposal, as yet untested experimentally (even in animals), is conceptually based on modifying the procedure of somatic cell nuclear transfer (SCNT), now used to produce cloned embryos. In standard SCNT, a somatic cell nucleus is introduced into an oocyte (egg cell) whose own nucleus has been removed. The product is a cloned embryo (virtually identical, at least genetically, to the organism from which the donor nucleus was taken), the functional equivalent of a fertilized egg that is capable (at least in some cases) of developing into all later stages of the organism. ANT, the modified procedure proposed by Hurlbut, involves altering the somatic cell nucleus before its transfer to the oocyte, and in such a way that the resulting biological entity, while being a source of pluripotent stem cells, would lack the essential attributes and capacities of a human embryo. For example, the altered nucleus might be engineered to lack a gene or genes that are crucial for the cell-to-cell signaling and integrated organization essential for (normal) embryogenesis.²¹ It would therefore lack organized development from the very earliest stages of cell differentiation. Such an entity would be a "biological artifact," not an organism. Removal of cells from, or even disaggregation of, this artifact would not be killing or harming, for there is no living being here to be killed or harmed. After extraction from this artifact, the cells could have the missing gene or genes reinserted, with a view to deriving "normal" pluripotent stem cells from them.

In offering his proposal for ANT, Hurlbut emphasizes that no embryo would ever be created or destroyed; since the genetic alteration is carried out in the somatic cell nucleus before transfer, the biological artifact is "brought into existence with a genetic structure insufficient to generate a human embryo." Hurlbut compares the product of ANT to certain ovarian teratomas and hydatidiform moles, genetically or epigenetically abnormal natural products of failed fertilization that are not living beings but "chaotic, disorganized, and nonfunctional masses." If, as Hurlbut suggests, the biological artifact is ethically equivalent to a tissue culture, teratoma, or mole, there would seem to be nothing ethically problematic about harvesting stem cells from it. Nonetheless, a number of ethical questions and concerns have been raised about this proposal.

Some people have wondered about the accuracy of the claim that no embryo creation or destruction is entailed by the proposal. They understand that the proposed biological artifact has, from the beginning, a built-in genetic defect that prevents it from developing normally. Yet they worry that this is not the production of a non-human entity but the deliberate creation of a doomed or disabled human embryo, or, in other words, that Hurlbut's proposal amounts to creating and using "bad or sick embryos," rather than "non-embryonic entities." Hurlbut's claim that his method would not yield a defective embryo rests on the fact that a genetic alteration sufficient to prevent embryogenesis is introduced into the nucleus before it is transferred to the oocyte, and that the alteration would be so fundamental that it would preclude the integrated organization that characterizes a human embryonic organism. If no embryo is created, then none is violated, mutilated or destroyed (which would be the case if the alteration were introduced after normal fertilization).

Nonetheless, some critics wonder how the product of that nuclear transfer is in fact essentially different from-and less an embryo than-a fertilized egg into which the same disabling genetic alteration is introduced only after normal fertilization. A person's perception of the truth in this matter may depend on how easy it is to turn the genetic defect on or off. The easier it is to activate and deactivate the genetic defect, the more this proposal looks like interfering with the normal development of an embryo rather than manufacture of an artificial non-organismic structure. Unless it can be shown that the artifact is not truly an embryo-that is, that it lacks (by design) the possibility of becoming not only a live-born human but an organized, differentiating early human embryo and fetus-there will likely be ethical debate on whether it is permissible to continue "abusing" the embryo-like entity by suppressing the genes it needs for development. Furthermore, even if the artifact were conclusively shown to lack genes indispensable for becoming an organized, differentiating human embryo, some critics might continue to insist that it was destined to become a defective, severely deformed human embryo, the defect and deformity having been deliberately inflicted on it by the scientist. Experimental work in animals, however, might help resolve these questions and allay these concerns. If, for example, the biological artifact begins to grow in ways that resemble unorganized cells in a tissue culture, critics may gain confidence in the non-embryonic character of the product.

Like ordinary cloning-for-biomedical-research (SCNT), this altered nuclear transfer proposal requires a (probably large) supply of human oocytes, which would have to be donated, purchased, or produced for research purposes. Some will find this troubling, and on multiple grounds. Obtaining human oocytes currently requires hormonal stimulation and superovulation in the women who would be donating or selling their eggs, practices that carry significant medical risks to the women, risks not easily justified when they themselves or their prospective children are not the beneficiaries of the oocyte retrieval (as are women undergoing these procedures in hopes of having children). In addition to the medical risks, there are also ethical concerns about the practice of commercializing human reproductive tissue and about any buying and selling of eggs: the exploiting of poor women, the coarsening of society's sensibilities, the developing of markets in (reproductive) human tissues. More deeply, one critic suggests, we must consider the implications and the consequences of coming to regard human eggs and sperm as fungible raw materials, to be used in ways that have nothing to do with their procreative biological and human meaning. There is a risk that, in seeking to avoid the problem of embryo destruction, we would thus be furthering a dehumanized and utilitarian view of human beginnings as bad as the one that this alternative proposal was trying to combat.²²

There is, at least in theory, the possibility that human oocytes can be obtained not from women egg donors by superovulation but from ovaries surgically removed from patients or harvested from cadavers. The ooocyte precursors extracted from these ovaries could then be matured in vitro. Alternatively, the ovaries could be transplanted into animal hosts and eggs produced by hormonal stimulation of the animals.²³ Research in this area is at a very preliminary stage. And the objections just noted to non-reproductive uses of human reproductive tissue could also be raised to obtaining eggs in these non-invasive ways, should they ever become possible.

To some observers, the procedures involved in ANT are inherently objectionable. Certain commentators, for example, find the very idea of tampering to put something destructive into the human genome, even for a good cause, morally and aesthetically offensive. Some find it aesthetically repulsive and ethically suspect to be creating such neither-living-nor-nonliving, near-human artifacts, a practice they regard as ethically no improvement over destroying early embryos. Other critics of the ANT proposal argue that, while it is ethically acceptable to modify the human genome for treatment (with consent) of individuals with known genetic disorders, it is quite another thing to do so for other than therapeutic purposes or to do so in eggs or sperm before there is an existing needy individual. Some think this is a major ethical boundary that ought not to be crossed lightly.²⁴ Others are troubled by the attitude of mastery or hubris implied in a project that aims at engineering a human biological artifact.

In response to these objections, Hurlbut replies that the ANT technique would be used only for serious scientific research within the frame of therapeutic purposes beneficial to a large population whose medical needs are of grave concern. Further, he points out that we accept many medical and research practices that are aesthetically ugly and morally worrisome, from cutting into a living body or brain, to giving people a dose of disease for vaccination, to growing great sheets of skin from cells harvested from foreskins. We do these things-and many others like them-in the service of the higher goal of healing, the very goal to which the ANT proposal is dedicated.²⁵

Several worries have been expressed not about the proposal itself but about what it might lead to, or about what it might be seen as justifying in the future. For some, the proposal appears to open the door to a troubling new field of biomedical engineering. True, its initial relatively modest goal is only to produce a biological artifact capable of yielding pluripotent human stem cells while not itself being a complete human organism with developmental potential. And if, in the process, definite criteria could be established for distinguishing between the not-human and the human, this proposal might have the salutary effect of erecting boundaries that would open avenues for scientific advance without threatening human dignity, boundaries that do not now govern the practices of human embryo research in the private sector. But, extrapolating into the future from an ANT precedent, pursuing this proposal could-whether intentionally or not-help to launch a new field of bioengineering, devoted to manufacturing intermediate biological forms that are sufficiently human to yield useful biomedical materials, but not so human that it would be unethical to destroy or exploit them. Hurlbut's arguments could be adapted to justify the deliberate production of teratomas, hydatidiform moles, inter-species hybrids, and other ill-formed, non-viable, but potentially useful biological artifacts. Once we start down the road of deliberately engineering artificial entities with some human properties, it is not obvious how bright ethical boundaries between the acceptable and the unacceptable can be drawn.

A second "slippery slope" concern has to do with the flexibility of the developmental stage at which disordered growth is set to begin. Hurlbut's proposal involves building in a genetic alteration that causes development to go irretrievably awry from the very start of embryonic development. But suppose a useful genetic modification were achieved that entailed chaotic and disorganized development only at a later stage of embryonic (or even fetal) development. Could not the ethical reasoning in defense of ANT be used to argue that such further-developed but still inherently defective entities are "fetus-like but not actual fetuses," and hence ethically suitable for exploitation and destruction? (The same question is relevant for ongoing destructive embryo research using normal IVF embryos, whose exploitation and destruction are justified because of the human benefits they might eventually bring.) It would certainly be troubling if the ethical case for ANT could be used to justify the creation and destruction of fetus-like entities. Hurlbut's proposal, seeking a source of pluripotent human stem cells, confines its attention to the early stages of embryonic development. But someone looking for a source of tissues or even primordial organs might be tempted to apply his reasoning to later and later stages of development, not excluding the deliberate production of anencephalic fetuses or even newborns, useful as a source of organs and tissues. Hurlbut's criterion for being a truly human organism-"organization of the species-typical kind"-would appear to be inherently malleable and open to interpretation (and even mischief).

Arguments that worry about future extensions of present techniques, or future applications to dubious ends, or future uses of current ethical justifications to validate later unsavory practices, while worth considering, tend to assume that people are either unable or unwilling to draw the necessary distinctions and erect the necessary ethical boundaries between current acceptable practices and future unacceptable ones-an assumption that is readily subject to dispute. But since the truth of this assumption cannot be known in advance, and can only be demonstrated case by case and then only by going forward and running the predicted risks, there is at least some reason to wonder whether any newly devised technological solution to the ethical problem of embryo destruction will not, in the end, be creating or contributing to ethical problems worse than the one it set out to cure.

Although the proposal for ANT has yet to be tested, several scientists have indicated that they believe that it can easily be made to work, and a few are apparently ready to try it out in non-human animals. It would be crucial to show that the disorganizing genetic or epigenetic alteration introduced into the somatic cell nucleus before transfer could be fully controlled, with predictable results, and fully reversed without residual abnormalities in the derived cells extracted from the embryo-like artifact. For unless the genetic engineering were fully reversible, the resulting stem cells would likely carry genetic or other alterations that might compromise the value of the stem cells. Moreover, there may be a tension in this proposal between ethical considerations, which require that insurmountable developmental barriers be genetically built into the embryo-like entity from the start, and technical feasibility, which would favor simpler barriers to development that are easily and completely reversible. Presumably, the more hard-wired the introduced defect is, the more difficult it will be to reverse. However, it is quite possible that, with a sufficiently determined research program, even the more technically daunting versions of the proposal could be achieved.

It is important to note that both some of the strengths and some of the weaknesses of the ANT proposal come from the fact that it is basically a form of SCNT or cloning. On the positive side, if successful, ANT could provide stem cells with a much greater diversity of genotypes than is possible under current methods of stem cell derivation (or under the other two proposals we have considered so far^xxv), allowing a far wider range of medical possibilities such as disease modeling or drug testing. Likewise, ANT would allow the generation of pluripotent stem cell lines with pre-engineered alterations (such as enhanced immune response or correction of a genetic defect) that might make them of more scientific interest or therapeutic value.

On the negative side, however, attempts to clone mammals have so far resulted in high rates of death, deformity, and disability in the animals that come to birth following SCNT. In 2002, research in the laboratory of Rudolph Jaenisch at MIT showed that, in cloned mice, about 4% of genes function abnormally, owing mainly not to mutations but to departures from normal activation or expression of certain genes.²⁶ It is not yet known whether similar genetic or epigenetic abnormalities will also be found in any stem cell lines that might be derived by ANT. Such problems are much less likely to be encountered in stem cell lines derived from IVF embryos.

Compared to deriving human embryonic stem cells from normal blastocysts, procedures such as ANT are quite complex and would yield cells that would then have to be restored to genetic normality before stem cell lines could be derived. Many scientists, we suspect, would be reluctant to attempt such challenging feats with no rational purpose other than to satisfy the ethical objections of others, and one prominent scientist in the stem cell and cloning field has recently made such a complaint publicly (notwithstanding the fact that the company for which he works has filed a patent application for precisely such a procedure).²⁷ Other scientists have reacted to news of the ANT proposal by describing it as exceedingly complex and technically challenging, not even testable without time-consuming experiments involving substantial investment of precious resources.²⁸ Even if federal funding for research on ANT-derived stem cell lines were approved, stem cell scientists might prefer to seek private funding of unrestricted stem cell research rather than follow procedures that seem to them burdensome and scientifically useless. Also, as Hurlbut himself acknowledges, proof of principle and safety-and-efficacy experiments need first to be done in animals, and it might be many months or even years before this process could be perfected using human tissues. Many stem-cell scientists, eager to press forward, are unlikely to wait for these new lines, especially if they are not themselves bothered by the embryo destruction that necessarily results when stem cell preparations are derived, as they are now, from unused IVF embryos. On the other hand, there may be some scientists, either opposed themselves to destroying embryos or hoping to find a way around the current federal funding restrictions, who would be willing and even eager to test Hurlbut's proposal in animals, and several have apparently volunteered their collaborative services for such animal trials.

If the biological artifacts created and destroyed under this proposal were persuasively shown not to be human embryos, the proposal would presumably be deemed consistent with the Dickey Amendment and therefore eligible for federal funding. Whether the President's policy and budget for federal funding of stem cell research would-or should-be expanded to support research on ANT or other biological artifacts (even in animals) is a question whose answer will depend not only on the issue of legal eligibility but also on the assessment of other ethical, scientific, and prudential considerations of the sort we have just discussed.

Besides Hurlbut's ANT proposal, other methods of constructing embryo-like artificial structures are under investigation, including a technique recently demonstrated by Karl Swann and colleagues at the University of Wales College of Medicine, in which a human oocyte is biochemically "tricked into thinking it has been fertilized."²⁹ The treated eggs divide to the blastocyst stage (50-100 cells), at which point stem cells can presumably be derived.³⁰ Although it undergoes several cycles of cell division, the "parthenogenetic" (that is, unfertilized but still developing) blastocyst-like entity is assumed by most commentators to lack entirely the potential for development as a human being, and is therefore, arguably, not really an embryo.^xxvi If this is correct, then this technique might provide another means for deriving pluripotent stem cells without creating or destroying embryos. Yet the only experiment that could prove whether this plausible assumption is in fact correct-transferring a parthenogenetic embryo to a woman to try to bring it to birth-cannot be ethically attempted. In the absence of such proof, the biological and moral status of the parthenogenetic blastocysts is likely to remain in doubt and controversial. Still, those who are convinced that parthenogenetic embryos have no chance of development beyond the blastocyst stage are likely to have few ethical objections to the production and use of such entities.³¹ It remains to be seen whether viable and genetically stable pluripotent stem cells can be derived from these parthenogenetic blastocysts and whether imprinting and other issues related to their parthenogenetic origin might limit their utility in research or potential clinical trials. Under the present terms of the Dickey Amendment, this proposal would be unlikely to be eligible for federal funding, since the amendment specifically prohibits the use of federal funds for research that may cause harm to an embryo produced by, among other means, parthenogenesis.

A quite different route to the production of pluripotent stem cells would be to reprogram differentiated somatic cells so as to restore to them the pluripotency typical of embryonic stem cells. The obstacles here are not ethical, but technical. Because it involves neither the creation nor the destruction of human embryos, the common ethical objection to human embryonic stem cell research would not apply. But it would take new scientific advances and new technological innovation before such "dedifferentiation" of somatic cells back into pluripotent stem cells could be achieved. Several suggestions have, however, been offered for how such dedifferentiation might be achieved, and the value of success cannot be overstated. For if it were possible to undo the differentiation of somatic cells, running development in reverse back to the state of pluripotency, it would in principle be possible for autologous pluripotent stem cells to be obtained from the body of any human being. Such individualized stem cells would then be available as a potential source of personalized, immuno-compatible regenerative therapies.

There would seem to be nothing to object to ethically if procedures were developed to turn somatic cells into pluripotent stem cells, non-embryonic functional equivalents of embryonic stem cells. Of course, if the dedifferentiation were pursued beyond (mere) pluripotency to the point of yielding a totipotent cell-in effect, a cloned human zygote-the moral status of such a cell would become a serious issue, as would the permissibility of using it either for reproductive or for research purposes. For a totipotent cell is, arguably, an organism at the unicellular stage, and a strong case could be made that the product is not a pluripotent stem cell but an embryo.

Research into dedifferentiation of somatic cells is at a preliminary stage, and it is much too early to know whether this will succeed. It may prove possible to culture specific populations of somatic cells-cells that may be especially susceptible to dedifferentiation-under conditions that might get them to reverse their differentiating epigenetic changes, thereby leading them to become more multipotent or even completely pluripotent; there is also some hope of identifying and isolating the chemical factors present in oocytes and other cells (such as ESCs) that are responsible for maintaining or restoring cells to pluripotency, and of using these chemicals to dedifferentiate ordinary somatic cells (without the further need for oocytes or embryos).

In nature, limited dedifferentiation is involved in the regeneration of missing limbs in amphibians, though the precise mechanism is not yet known.³² Studies have shown that some adult human somatic cell types (blood, liver, muscle) can be chemically dedifferentiated back into their corresponding multipotent progenitor cells (that is, adult stem cells).³³ Several research laboratories have reported the direct isolation of cells from bone marrow of children or adults that, when cultured in vivo, have or acquire the capacity to differentiate into many mature cell types, including cells originating from all three embryonic germ layers.³⁴ These cultured human multipotent cells also show the presence of certain biochemical properties ordinarily found only in human embryonic stem cells. It is interesting to speculate that it may be the same bone marrow stem cells, cultured in vitro under different conditions, that revert in some cases to (rather modestly multipotent) mesenchymal stem cells, in some other cases (further back) to the clearly multipotent adult progenitor cells, and in still other cases (yet to be achieved) to the ur-primordial stem cell, the fully pluripotent stem cell, functionally equivalent to an embryonic stem cell (though not of embryonic origin). If such "graded dedifferentiations" are indeed the cause of the variations seen among the cultivated stem cells now known to arise from bone-marrow stem cells, further research-using also stem cells obtained from umbilical cord blood-might very well turn out to yield the big payoff: fully pluripotent stem cells, obtainable at will and altogether without any involvement of embryos-and well suited for autologous transplantation.

Another possible approach to somatic cell dedifferentiation relies on knowledge that might be gained through cloning-for-biomedical-research. In SCNT or cloning, a somatic cell nucleus is reprogrammed back to totipotency by transfer into an enucleated oocyte. Presumably, cytoplasmic factors that are present in the oocyte (and that may also be present in cultured embryonic stem cells^xxvii) are responsible for the dedifferentiation that takes place. If and when these cytoplasmic factors can be identified and isolated, it may be possible to use them-instead of SCNT into oocytes-to coax some ordinary somatic cells to dedifferentiate back to the pluripotent stage.³⁵ Once again, should the process of dedifferentiation go too far, back to totipotency, the end result will not be a stem cell but the functional equivalent of a zygote, and one would be back in the ethical soup from which this proposal was intended to provide an escape. Great care would therefore have to be exercised to ensure that dedifferentiation, if and when it occurs, goes only so far and no further. Given the complexity of the process, and how little we now know about the factors that regulate differentiation and its opposite, it is not likely that this (second) approach will yield results in the near future.

Certainly, dedifferentiation of somatic cells back to their corresponding progenitor cells will likely be welcomed as a powerful new way to produce large quantities of multipotent adult stem cells. If dedifferentiation is perfected to the point of yielding cells as pluripotent as embryonic stem cells, there is no reason to doubt that this procedure would be widely embraced and the cells obtained widely used. An additional-clinical-potential benefit of such cells would be that specialized cells derived from them (for example, heart muscle cells, nerve cells) could be reintroduced therapeutically into the patient from whom they were derived without risk of immunological rejection.

Because this research does not involve human embryos at any stage, it would not offend either the letter or the spirit of the Dickey Amendment. Aside from the concern that dedifferentiation might proceed too far (resulting in the functional equivalent of a zygote), there would appear to be no obstacle to, or reason to oppose, federal funding of research on dedifferentiation of somatic cells.

The United States has been engaged in a vigorous ethical debate about embryonic stem cell research, prompted by tensions between the desire for biomedical progress and respect for nascent human life. The scientific and medical promise of stem cells has generated enormous excitement among researchers and patient groups. The ethical issues raised by embryo research have roused considerable public attention and concern. Many people share the hope that stem cell research will eventually save lives and yield the promised remedies for numerous chronic illnesses. Many people (including many who are eager for regenerative medicine to succeed) share the concern that embryonic stem cell research depends on destroying human embryos and cheapens human life by creating and using it for experimentation.

Some people hope that stem cells derived from non-embryonic sources (known as adult stem cells) will turn out to be as good as stem cells derived from embryos, but it is too early to tell which sort of stem cells will be most useful for the treatment of which diseases. Some believe that current federal law and policy governing the funding of embryonic stem cell research are too restrictive, and that the existing embryonic stem cell lines will not prove adequate for the work ahead. Others believe that medical progress must not be purchased by destroying human life, even at its earliest stages, and that creation of human embryos specifically for use in research should be outlawed.

Mindful of the moral weight of the arguments on the various sides of this controversy, and charged with finding ways for science to proceed while respecting moral norms, the Council has taken seriously a number of recent proposals and suggestions for techniques to derive new pluripotent stem cell lines in ways that might be ethically uncontroversial.

This White Paper has summarized several current proposals for obtaining pluripotent human stem cells that do not require destroying human embryos. In each case, we have examined whether the proposal is ethically sound, scientifically feasible, and practically "realistic." The inquiry we have undertaken constitutes no more than a preliminary hearing, designed mainly to see whether there are any insuperable ethical, scientific, or practical objections to further consideration of these proposals. Because all of these proposals are relatively new, the ethical issues they raise need more discussion, and much research would be needed before it became clear which of them, if any, would succeed. Likewise, further legal interpretation and sober political deliberation would be required to determine which of the proposals are, under current law, eligible for federal funding and which are both ethically and scientifically deserving of such official national support. We hope that our analyses of the ethical, scientific, and practical aspects of these proposals will contribute to a more informed and comprehensive scrutiny of their respective merits.

The analyses of this White Paper, while preliminary, do lead to the following provisional assessments. The last proposal, dedifferentiating somatic cells back to pluripotency, seems ethically the most unobjectionable, but for now scientifically and technically uncertain; recent derivations (from adults) of relatively undifferentiated multipotent stem cell lines^xxviii may be an encouraging, albeit preliminary, step toward this goal. The first proposal, seeking to derive stem cells from organismically dead embryos, has yet to be tested, even in animals. But the natural history studies proposed could be undertaken forthwith and in an ethical manner, not only in animals but also in humans, and we might learn soon whether reliable objective criteria for determining death of IVF embryos can be developed. The second proposal, seeking to develop stem cells from blastomeres extractable from living embryos, is also now technically feasible, though large ethical difficulties remain, concerning especially the propriety of imposing risks of embryo biopsy and blastomere removal on the born child the embryo might become, solely for research of no benefit to him or her. The third proposal, seeking to derive stem cells from genetically engineered artificial entities, is technically the most demanding and ethically the most complex and puzzling. Even its proponents agree that it would need to be carefully tested in animals before any thought of human trials could be countenanced.

Among these several proposals, the Council has no unanimous recommendation to make. Different Council members are drawn more to one or less to another of the four proposals. Each of us weighs the ethical issues differently. And we have differing views on which approach is likely to succeed technically or to be useful practically. A few of us may suspect that the quest for alternative sources of stem cells is misguided, and that we should continue using the embryos we have (or can produce directly) in order to get any new stem cell lines we need.

Yet on the limited ethical threshold question-"Does this proposal appear to meet a minimum ethical standard to justify further serious consideration and scientific exploration?"-the Council offers the following provisional conclusions.³⁶

The first proposal, deriving cells from organismically dead embryos. Although it raises some serious ethical questions, we find this proposal to be ethically acceptable for basic investigation in humans, provided that stringent guidelines like those proposed by Drs. Landry and Zucker are strictly observed. The results of such investigations would help to determine whether the method would in fact prove ethically acceptable in the long run.

The second proposal, blastomere extraction from living embryos. We find this proposal to be ethically unacceptable in humans, owing to the reasons given in the ethical analysis: we should not impose risks on living embryos destined to become children for the sake of getting stem cells for research. This approach could, of course, be attempted in animals, but we do not yet see how results from animal experimentation could alter this assessment of ethical propriety in humans. We do not expect this method to become ethically acceptable for human trials in the future.

The third proposal, cells derived from specially engineered biological artifacts. Because this proposal raises many serious ethical concerns, we do not believe that it is at this time ethically acceptable for trials with human material. Although a few of us are not eager to endorse even animal and other laboratory work investigating potential human applications, most of us believe the proposal offers enough promise to justify animal experimentation, both to offer proof of feasibility and utility and to get evidence bearing on some of the ethical issues. We find no insuperable ethical objections to pursuing this proposal in animal models, which is, we note again, all that the proponents now seek to do. The possibility of any future endorsement of trying this approach in humans will depend upon a more thorough ethical analysis made possible in part by animal experiments.

The fourth proposal, cells obtained by somatic cell dedifferentiation. We find this proposal to be ethically unproblematic and acceptable for use in humans, if and when it becomes scientifically practical, provided the line between pluripotency and totipotency can be maintained, as discussed in the ethical analysis.

Despite any differences among us about the merits of each proposal, the Council shares the view that the group of proposals here discussed-and others like them that they might stimulate-deserve the nation's careful and serious consideration. Because the Council is wholeheartedly committed both to the advancement of science for the betterment of humankind and to the defense of human freedom, dignity, and the value of human life, we are pleased to endorse these proposals as worthy of further public discussion, and we are pleasedto encourage their scientific exploration in accordance with the preliminary ethical judgments just offered. A good part of our uncertainty today about the merits of these proposals rests on the paucity of available scientific evidence and of demonstrated technical prowess. Both enthusiasts and skeptics regarding these proposals should agree at least on this: that further empirical studies will be needed before the true potential of these proposals can be properly assessed.

We conclude by stressing that while these four proposals are the ones that seem most worthy of analysis and discussion at this time, it is altogether possible, indeed likely, that other avenues to human pluripotent stem cells not requiring the destruction of human embryos may be proposed or discovered in the future. By limiting ourselves to these current proposals, we do not intend to exclude any additional ones. On the contrary, we publish this White Paper to encourage scientists to creatively devise other and better proposals and to highlight the appeal of the larger purpose: to find ways to advance pluripotent stem cell research that all our fellow citizens can wholeheartedly support. That end is, in the view of the Council, a desirable goal for our society and one that justifies making the extra effort to seek out, assess, and attempt new, ethically uncontroversial methods of stem cell derivation.

Footnotes

i. Since 1995, Congress has annually enacted legislation (the Dickey Amendment) that prohibits the use of federal funds for research in which human embryos are destroyed or harmed. In 2001, President George W. Bush instituted the current policy, which permits federal funding for research on those embryonic stem cell lines already in existence, but not for derivation or use of any new lines (the creation of which would require new embryo destruction). Vigorous debates continue about the ethics of embryonic stem cell research, as well as about the federal funding policy. For a synoptic view of these ethical and political discussions, including a history of the relevant public policy decisions, see Monitoring Stem Cell Research: A Report of the President's Council on Bioethics (2004), especially chapters two and three. For the text of the Dickey Amendment, see endnote 10.

ii. A "blastomere" (literally, a part of a "blast" or embryo) is a cell contained within an early embryo (up to two days after conception), which comprises a small number of such blastomeres. Thus a 4-celled embryo contains four blastomeres, and an 8-celled embryo contains eight blastomeres. Beyond the 8-cell stage, on the third day after conception, the early embryo turns into a compact sphere known as a morula; and on the fifth day after conception, the morula becomes a blastocyst (100-200 cells), a hollow ball of cells surrounding an inner cell mass. The cells of an embryo at the morula or blastocyst stage, being more differentiated than those of a 4-8-cell embryo, will not be referred to here as blastomeres, although some authors use the term to describe any cells of an embryo before it cavitates to become a blastocyst.

iii. A totipotent cell (for example, the fertilized egg or zygote) is one that can give rise to the entire organism, including the extra-embryonic membranes; a pluripotent cell (for example, an embryonic stem cell) is one that can give rise to many if not all the different cell types of the human body, but not to the whole organism as a living integrated entity.

iv. The discussions in this paper take for granted the existing practices of assisted reproduction, including in vitro fertilization, the storage of frozen embryos for later reproductive use, and preimplantation genetic screening and diagnosis (PGD) of in vitro embryos prior to their transfer to initiate a pregnancy. These practices raise ethical issues of their own, and some people (including some Council members) object to them altogether. Although we recognize that there are many, and often deep, questions connected with the growing control over all aspects of human procreation, we will (for the most part) not be analyzing or arguing those questions here. We recommend, however, that they not be lost sight of, especially as the political acceptability of some of the proposals reviewed here will be influenced by where people stand on those larger questions. And clearly, some people will evaluate the proposals here under review not solely in themselves, but also by assessing their relationship to and potential effect on the way assisted reproduction is practiced, and by judging whether the proposed uses of dead embryos, blastomere biopsy, egg harvesting, and altered nuclear transfer create incentives for engaging in practices they deem misguided, unethical, or unwise.

v. Even if one or more or of these alternative sources of pluripotent stem cells were to meet all these requirements (ethical soundness, scientific feasibility, eligibility for federal funding, and broad acceptability to scientists), there would still remain the question of whether the cost of pursuing such alternatives-the necessary investment of scientific energy and resources, and the possible diversion of such energy and resources from other promising avenues of research-would outweigh the benefits. This White Paper does not pretend to answer that serious question, for it cannot be answered a priori, in advance of knowing empirically those costs and benefits. Moreover, how any person answers that question will depend on how strongly he or she cares about the additional "benefits" and "costs" of requiring scientific research to respect certain moral boundaries and strongly held moral qualms, in this case about protecting the moral worth of embryonic human life.

vi. From reference 2: "We propose that the [minimal] defining capacity of a 4- or 8-cell human embryo is continued and integrated cellular division, growth and differentiation. We further propose that an embryo that has irreversibly lost this capacity, even as its individual cells are alive, is properly considered organismically dead."

vii. The cells of an organismically dead embryo have stopped dividing altogether; their cleavage arrest is considered irreversible in the sense that the cells show no tendency to resume dividing while remaining part of the dead embryo. Of course, the goal of the Landry-Zucker proposal is to retrieve from these dead embryos some cells that can be induced to resume dividing under conditions conducive to stem cell derivation.

viii. It might be asked why Landry and Zucker confine their attention to embryos that have been frozen and then thawed, as opposed to "fresh" never-frozen embryos. The answer is that, in current IVF practice, healthy-looking embryos that have reached the 4- or 8-cell stage but are not selected for transfer to the uterus are generally frozen for possible future transfer attempts. To delay freezing such embryos while looking for signs of cleavage arrest would, arguably, subject the embryos to additional risk, contrary to the intention of the proposal. In contrast, the embryos of interest to Landry and Zucker-frozen embryos that are thawed out but never resume cell division-are not, in current IVF practice, either transferred to the uterus or frozen a second time.

ix. Yet here, too, a caution must be observed. As discussed elsewhere in this paper (see section 2 on p. 29, as well as endnote 18), we do not know precisely when, in embryonic development, human blastomeres cease to be totipotent, that is, individually capable of growing into a complete human being. To avoid the possibility that the cells extracted from a dead embryo might be totipotent, it would perhaps be advisable to carry out the Landry-Zucker proposal using somewhat older cleavage-arrested embryos (containing 8 cells or more) rather than embryos as early as the 4-cell stage.

x. To turn this morally scrupulous proposal into morally scrupulous practice would seem to require protocols and enforceable regulations that would provide very strict monitoring and oversight. The political acceptability of the Landry-Zucker proposal might hinge on having a regulatory system in place so that there would be some way to track IVF embryos (as the Canadians do), thus ensuring that abuses do not creep in.

xi. "Failure to cleave 24 hours after thawing" may not prove to be a sufficient criterion for organismic death, since, in the study by Laverge, et al. [see endnote 4], roughly 10% of embryos meeting that criterion showed some sign of further cleavage by 48 hours after thawing.

xii. Note that most embryologists reserve the term "morula" for embryos of more than 8 cells. See footnote starting on page three above.

Deriving stem cells from isolated single blastomeres may prove significantly more challenging than deriving them from disaggregated blastocysts or morulae; in human ESC derivations achieved so far, groups of cells have been cultured together, and it is not known whether the presence of other cells is necessary for the derivation of embryonic stem cells from a single blastomere.

xiii. Council Member Janet Rowley has suggested that it would be strange, while allowing large numbers of unwanted but otherwise normal and viable IVF embryos to die, to ask scientists to make strenuous efforts to rescue cells, potentially normal but potentially abnormal, only from those thawed embryos that have spontaneously stopped dividing.

xiv. The President's August 9, 2001, policy (see endnote 11), which offers federal funding for research on embryonic stem cell lines only if those lines were derived before the date of the policy, is intended "to allow us to explore the promise and potential of stem cell research without crossing a fundamental moral line, by providing taxpayer funding that would sanction or encourage further destruction of human embryos that have at least the potential for life."

xv. The use of federal funds for the derivation of stem cells from doomed but still-living embryos (the so-called "spare" embryos, unwanted for transfer in efforts to produce a child) violates the letter of the Dickey Amendment; the use of federal funds for research on embryonic stem cells derived by someone else's prior destruction of a living embryo violates the spirit of the Dickey Amendment. Deriving stem cells only from already dead embryos seems not to commit either of these violations.

xvi. The Dickey Amendment has been enacted with the support of many members of Congress who are unwilling to "take for granted" the practices mentioned in the footnote on page six, including freezing of embryos or even IVF itself.

xvii. At present, embryonic stem cells are typically derived by extracting cells from the inner cell mass of the embryo at the blastocyst (roughly 100-cell) stage; this entails the destruction of the trophectoderm (that is, the outer ring of cells in the spherical blastocyst structure, the precursor of the fetal contribution to the placenta) and the death of the embryo.

xviii. A similar idea was proposed by Representative Roscoe Bartlett of Maryland as far back as 2001.

xvix. Identical twinning can apparently take place during at least two stages of the in vivo development process: Spontaneous separation of the blastomeres at the two-cell stage leads to the formation of twins with two separate placentas (about 1/3 of cases); while embryo splitting at the blastocyst stage leads to the formation of identical twins that share the same placenta (about 2/3 of cases).

xx. Indeed, in the initial studies showing the potential usefulness of morula-stage human embryos as a source of stem cells, many human embryos were obviously destroyed.

xxi. Studies with mouse embryos have shown that isolated 8-cell blastomeres will develop into vesicles of trophectoderm, containing little or no inner cell mass, making derivation of ESC lines difficult. And there is no reason to believe that things will be different in humans. However, there has been one published report claiming derivation of mouse ESC lines from isolated 8-cell blastomeres: One cell line was obtained from 52 dissociated 8-cell stage embryos. (See Delhaise, F., et al., "Establishment of an embryonic stem cell line from 8-cell stage mouse embryos," European Journal of Morphology 34, 237-243 [1996].) The Council is grateful to Dr. Janet Rossant for these observations.

xxii. Granting that apparently healthy children have been born following IVF and PGD, does removal of one or two blastomeres from the early human embryo have no effect on the child who is later born? We cannot be certain. There is some evidence (at least in mice) for asymmetrical division and distinct cell fates at the early cleavage stages. (See Piotrowska-Nitsche, K., et al., "Four-cell stage mouse blastomeres have different developmental properties," Development 132, 479-490 [2005].) It is possible that the loss of one or two blastomeres is entirely rectified, but even if development proceeds in a healthy manner, it may be that the child born is somehow a different child than the one that would have resulted from an undisturbed embryo.

xxiii. Here, as in so many other ethically charged situations, terminology matters enormously. One must take special pains not to prejudice consideration of the ethical issues by choice of terms that, intentionally or unintentionally, incline readers and hearers one way rather than another. This case is no exception. Precisely because the purpose here is to create an entity that is not in fact a human embryo, but that is nevertheless enough like a human embryo that it too contains cells that can be cultured to give rise to pluripotent stem cells, the artificially produced entity must be at once "non-embryonic" yet "embryo-like." But if the product is referred to as an "embryo-like" or (by analogy to "android") "embryoid" body, people may be encouraged to overemphasize its resemblance to an embryo and to slight the claim that this product is decidedly not a human embryo. On the other hand, if the product is referred to as a "non-embryonic" body or structure, people may be encouraged to think that its desired non-embryonic character has in fact been achieved. Since a major ethical issue turns on whether or not the biological artifact is truly not a living human embryo, all prejudgment by terminological fiat should be, where possible, avoided. We therefore usually call it (merely) a "biological artifact," denoting its origin but leaving its essential status open.

xxiv. This advantage would, however, be shared by the next proposal, somatic cell dedifferentiation.

xxv. However, live-birth parthenogenesis takes place naturally in certain amphibians and has been induced artificially even in mice. See Kono, T., et al., "Birth of parthenogenetic mice that can develop to adulthood" (Letters to Nature), Nature 428, 860-864 (2004).

xxvi. The Council is grateful to Dr. Markus Grompe for this observation.

xxvii. See endnote 34.

_________________

Endnotes and References

1. President's Council on Bioethics, Human Cloning and Human Dignity: An Ethical Inquiry, July 2002, pp. xxx and 121. For discussion of these competing goods, see Chapter Six, "Ethics of Cloning-for-Biomedical-Research." See also Chapter Three ("Recent Developments in the Ethical and Policy Debates") of Monitoring Stem Cell Research: A Report of the President's Council on Bioethics, January 2004.

2. Landry, D. W. and H. A. Zucker, "Embryonic death and the creation of human embryonic stem cells," The Journal of Clinical Investigation 114, 1184-1186 (2004). The transcript of the presentation and discussion of this proposal at the December 3, 2004 Council meeting is available online at www.bioethics.gov (see session 6). This proposal was anticipated in Grinnell, F., "Defining embryo death would permit important research," The Chronicle of Higher Education 49(36), B13 (May 16, 2003), and Grinnell, F., "Human embryo research: from moral uncertainty to death," American Journal of Bioethics 4, 12-13 (2004).

3. Although there is a broad consensus on the ethics of posthumous organ donation, there are some who have raised questions about the adequacy of the criterion (typically, "brain death") by which death ("irreversible loss of integrated functioning of the person") is established. See President's Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research. Defining Death: A Report on the Medical, Legal, and Ethical Issues in the Determination of Death (Washington: The Commission, 1981); Veatch, R. M., "The Conscience Clause," in The Definition of Death: Contemporary Controversies, ed. S. J. Youngner, et al. (Baltimore, Md.: Johns Hopkins University Press, 1999), 137-160; Shewmon, D. A., "Brainstem death, brain death and death: a critical re-evaluation of the purported evidence," Issues in Law and Medicine 14, 125-145 (1998); and Shewmon, D. A., "Chronic brain death: meta-analysis and conceptual consequences," Neurology 51, 1538-1545 (1998).

4. Landry and Zucker (op. cit.) refer to a study by Laverge, et al., in which, "out of 166 frozen embryos thawed for further growth, 78 embryos remained arrested at 24 hours after thawing, and 71 showed no sign of further cleavage at 48 hours." See Laverge, H., et al., "Fluorescent in-situ hybridization on human embryos showing cleavage arrest after freezing and thawing," Human Reproduction 13, 425-429 (1998).

5. D. H. Edgar and coworkers estimate that "the implantation potential of a population of embryos was reduced by ~30% by being subjected to cryopreservation." See Edgar, D. H., et al., "A quantitative analysis of the impact of cryopreservation on the implantation potential of human early cleavage stage embryos," Human Reproduction 15, 175-179 (2000).
See also the following studies cited by Laverge, et al. (in the paper mentioned in endnote 4): Camus, M., et al., "Human embryo viability after freezing with dimethylsulfoxide as a cryoprotectant," Fertility and Sterility 51, 460-465 (1989); Levran, D., et al., "Pregnancy potential of human oocytes-the effect of cryopreservation," New England Journal of Medicine 323, 1153-1156 (1990); Van Steirteghem, A., et al., "Cryopreservation of human embryos," Bailliere's Clinical Obstetrics and Gynaecology 6, 313-325 (1992); and Van der Elst, J., et al., "Prospective randomized study on the cryopreservation of human embryos with dimethylsulfoxide or 1,2-propanediol protocols," Fertility and Sterility 63, 92-100 (1995).

6. Recently, animal studies have shown that some individual blastomeres, removed from embryos that have ceased developing, can survive, grow, and function normally if they are transplanted into a still-living embryo that is itself developing normally. For example, in their presentation to the Council, Landry and Zucker pointed out work by Byrne and coworkers that showed that cells from abnormal partial blastulae of cloned amphibian embryos could be "rescued" by grafting them to normal host embryos where they contributed to several tissues. See Byrne, J. A., Simonsson, S., Gurdon, J. B., "From intestine to muscle: nuclear reprogramming through defective cloned embryos,"Proceedings of the National Academy of Sciences USA 99(9), 6059-6063 (April 30, 2002).

7. Strelchenko, N., et al., "Morula-derived human embryonic stem cells," Reproductive BioMedicine Online 9(6), 623-629 (2004). See also U.S. Patent Application 20040229350, "Morula derived embryonic stem cells," filed November 18, 2004.

8. The question was raised by Council Member Dr. Paul McHugh, at the December 2004 Council meeting. For discussion of this point, see the transcript of the December 3, 2004 meeting (session 6), available online at www.bioethics.gov.

9. Studies reporting the derivation of human embryonic stem cell lines starting from excess blastocysts from IVF procedures and supported by private funding include: (1) Mitalipova, M., et al., "Human embryonic stem cell lines derived from discarded embryos," Stem Cells 21, 521-526 (2003), and (2) Cowan, C. A., et al., "Derivation of embryonic stem cell lines from human blastocysts," New England Journal of Medicine 350, 1353-1356 (2004).

10. The Dickey Amendment, named for its author, former Representative Jay Dickey of Arkansas, has been attached to the Health and Human Services authorization bill each year since 1995. The provision reads as follows:

11. See "Remarks by President George W. Bush on Stem Cell Research," as made available by the White House Press Office, August 9, 2001.

12. Genetics and Public Policy Center, "Preimplantation Genetic Diagnosis: A Discussion of Challenges, Concerns and Preliminary Policy Options Related to the Genetic Testing of Human Embryos," Washington, D.C. (2004).

13. PGD can only be performed on embryos created through IVF. There are already questions about whether specific forms of IVF (in the absence of PGD) lead to an increased incidence of some rare genetic diseases. (See for example, Gosden, R., et al., "Rare congenital disorders, imprinted genes, and assisted reproductive technology," Lancet 361, 1975-1977 [2003].) In its 2004 report, Reproduction and Responsibility, the Council issued a recommendation to "undertake a federally funded longitudinal study of the impact of ARTs (Assisted Reproduction Technologies) on the health and development of children born with their aid."

14. See, among other articles, Adams, K. E., "Ethical considerations of applications of preimplantation genetic diagnosis in the United States," Biomedicine and Law 22, 489-504 (2003); Rechitsky, S., et al., "Preimplantation genetic diagnosis with HLA matching," Reproductive BioMedicine Online 9, 210-221 (2004); Sheldon, S. and S. Wilkinson, "Should selecting savior siblings be banned?" Journal of Medical Ethics 30, 533-537 (2004).

15. American Society for Reproductive Medicine, Ethics Committee Report, "Sex selection and preimplantation genetic diagnosis," Fertility and Sterility 72, 595-598 (1999); The Ethics Committee of the American Society of Reproductive Medicine, "Sex selection and preimplantation genetic diagnosis," Fertility and Sterility 82 (Suppl. 1), S245-248 (2004).

16. See, transcript of the December 13, 2002 Council meeting (session 6), available online at www.bioethics.gov.

17. Cohen, E., personal communication to the Council, commenting on this proposal.

18. Can single blastomeres extracted from a 4- or 8-cell embryo ever give rise to a whole organism? For human blastomeres, the answer is not known, and there are substantial ethical objections to performing the experiments that could assess this question. In mice, apparently normal and fertile animals have been produced from single blastomeres isolated from a 4-cell embryo (and from two blastomeres isolated from an 8-cell embryo); but the blastomeres had to be cultured in the presence of cells from "carrier embryos." See Tarkowski, A. K., Ozdzenski, W., Czolowska, R., "Mouse singletons and twins developed from isolated diploid blastomeres supported with tetraploid blastomeres," International Journal of Developmental Biology 45(3), 591-596 (2001). In experiments with non-human primates (rhesus monkeys), Chan, et al., [see "Clonal propagation of primate offspring by embryo splitting," Science 287, 317-319 (2000)] have reported that two blastomeres isolated from an 8-cell embryo gave rise to a live.born monkey they called Tetra. See also Schramm, R. D. and A. M. Paprocki, "Strategies for the production of genetically identical monkeys by embryo splitting," Reproductive Biology and Endocrinology, 2, 38 (2004).

19. Krzyminska, U. B., Lutjen, J., O'Neill, C., "Assessment of the viability and pregnancy potential of mouse embryos biopsied at different preimplantation stages of development," Human Reproduction 5(2), 203-208 (1990).

20. Quotations from Hurlbut are from his paper presented to the Council on December 3, 2004, "Altered Nuclear Transfer as a morally acceptable means for the procurement of human embryonic stem cells," available online in the December 2004 Council meeting background materials at www.bioethics.gov. For Council discussion of his paper, see transcript of the December 3, 2004 meeting (session 6), also available online at www.bioethics.gov.

21. Chawengsaksophak, K., et al., "Cdx2 is essential for axial elongation in mouse development," Proceedings of the National Academy of Sciences USA 101(20), 7641-7645 (May 18, 2004).

22. Cohen, E., op. cit.

23. See, Gook, D. A., et al., "Oocyte maturation, follicle rupture and luteinization in human cryopreserved ovarian tissue following xenografting," Human Reproduction 18(9) 1772-1781 (2003).

24. See, for example, the public comment made to the Council by Jaydee Hanson, representing the International Center for Technology Assessment, at the December 2004 meeting (session 7), the transcript of which is available online at www.bioethics.gov.

25. For discussion of these issues, see transcript of the Council's March 4, 2005, meeting (session 5), available online at www.bioethics.gov.

26. Humpherys, D., et al., "Gene expression in cloned mice derived from embryonic stem cell and cumulus cell nuclei," Proceedings of the National Academy of Sciences USA 99(20), 12889-12894 (October 1, 2002).

27. Robert Lanza, "The troubling prospect of genetic manipulation" (Letter), Washington Post, p. A20, December 13, 2004. Curiously, in September 2002, Advanced Cell Technology, the company for which Dr. Lanza works, filed a patent for producing genetically altered artificial embryo-like entities, partly for the same purpose. The company's patent application summarizes the main idea: "Methods for making human ES cells and human differentiated cells and tissues for transplantation are described, whereby the cells and tissues are created following somatic cell nuclear transfer. The nuclear transfer donor is genetically modified prior to nuclear transfer such that cells of at least one developmental lineage are de-differentiated, that is, unable to develop, thereby resolving the ethical dilemmas involved in reprogramming somatic cells back to the embryonic stage." (United States Patent Application 20020132346.)

28. See, Melton, D. A., Daley, G. Q., Jennings, C. G., "Altered nuclear transfer in stem-cell research-a flawed proposal," New England Journal of Medicine 351, 2791 (2004).

29. Rogers, N. T., et al., "Phospholipase C{zeta} causes Ca2+ oscillations and parthenogenetic activation of human oocytes," Reproduction 128(6), 697-702 (2004). See also Lin, H., et al., "Multilineage potential of homozygous stem cells derived from metaphase II oocytes," Stem Cells 21, 152-161 (2003).

30. While it has not yet been demonstrated using humans oocytes, José Cibelli and coworkers have recently succeeded in obtaining a line of pluripotent stem cells from parthenogenetically activated eggs of the long-tailed macaque (a nonhuman primate). See Cibelli, J. B., et al., "Parthenogenetic stem cells in nonhuman primates," Science 295, 819 (2002).

31. See, Kiessling, A. A., "Eggs Alone-Human parthenotes: an ethical source of stem cells for therapies?" Nature 434, 145 (2005).

32. Brockes, J. P., "Amphibian limb regeneration: rebuilding a complex structure," Science 276, 81-87 (1997).

33. See, for example, Chen, S., et al., "Dedifferentiation of lineage-committed cells by a small molecule," Journal of the American Chemical Society 126(2), 410-411 (2004); Odelberg, S. J., et al., "Dedifferentiation of mammalian myotubes induced by msx1," Cell 103(7), 1099-1109 (2000); and McGann, C. J., Odelberg, S. J., Keating, M. T., "Mammalian myotube dedifferentiation induced by newt regeneration extract," Proceedings of the National Academy of Sciences USA 98(24), 13699-13704 (November 20, 2001).

34. The best known research is by Catherine Verfaillie and coworkers at the Stem Cell Institute at the University of Minnesota Medical School. See, for a recent update, Dr. Verfaillie's review essay on "Multipotent Adult Progenitor Cells" (MAPCs), published as Appendix J in Monitoring Stem Cell Research: A Report of the President's Council on Bioethics, January 2004 (also available online at www.bioethics.gov). A more recent publication, by scientists at the University of Miami in collaboration with French scientists, reports the cultivation of pluripotent human cells from bone marrow (of human adults and children), after a unique expansion and selection procedure under conditions that were designed to mimic the in vivo environment of the early human embryo. According to the report, these cells can be grown for extended periods of time in culture, express genes similar to those expressed by embryonic stem cells in culture, and can be stimulated to differentiate into various adult tissue types, including neuronal and pancreatic islet tissue. See D'Ippolito, G., et al., "Marrow-isolated adult multilineage inducible (MIAMI) cells, a unique population of postnatal young and old human cells with extensive expansion and differentiation potential," Journal of Cell Science 117, 2971-2981 (2004). The relation between the MIAMI cells and Dr. Verfaillie's MAPCs is unclear, although it is possible that the MIAMI cells may be somewhat less differentiated than the MAPCs. The research from both laboratories needs to be reproduced by other scientists before its true promise can be assessed. See also Yoon, Y-S., et al., "Clonally expanded novel multipotent stem cells from human bone marrow regenerate myocardium after myocardial infarction," The Journal of Clinical Investigation 115, 326-338 (2005), and Kogler, G., et al., "A new human somatic stem cell from placental cord blood with intrinsic pluripotent differentiation potential," Journal of Experimental Medicine 200, 123-135 (2004).

35. See, in particular, Byrne, J. A., et al., "Nuclei of adult mammalian somatic cells are directly reprogrammed to oct-4 stem cell gene expression by amphibian oocytes," Current Biology 13, 1206-1213 (2003), and Gurdon, J. B., Byrne, J. A., Simonsson, S., "Nuclear reprogramming and stem cell creation," Proceedings of the National Academy of Sciences USA 100 (Suppl. 1), 11819-11822 (September 30, 2003).

36. For the Council's discussion that gave rise to these provisional conclusions, see transcript of the Council's March 4, 2005, meeting (session 5), available online at www.bioethics.gov.

WHITE PAPER: Alternative Sources of Pluripotent Stem CellS

Table of Contents

Introduction

Footnotes

Endnotes and References

Code of Federal Regulations or any successor to such regulations-

Code of Federal Regulations or any successor to such regulations-

Next Chapter