Ethical and Scientific Concerns About Induced Pluripotent Stem Cell Research -- Yamanaka and Thomson

Dianne N. Irving
copyright June 1, 2008
Reproduced with Permission

[Note: This article is copyrighted and thus must be acknowledged when using its original ideas and resources or quoting from it.]

I. Introduction

In a valiant effort to resolve at last the obvious and highly divisive ethical concerns surrounding the use of human embryonic and fetal "stem cells", tissues and body parts in experimental and therapeutic research and in patient "therapies", several researchers have recently released their studies involving the production of "induced pluripotent stem cells" (iPS cells). [See, e.g.: Yamanaka, Takahashi et al., "Induction of Pluripotent Stem Cells from Adult Human Fibroblasts by Defined Factors", Cell (2007), doi:10.1016/j.cell.2007.11.019,; see also, Thomson, Yu et al., "Induced Pluripotent Stem Cell Lines Derived from Human Somatic Cells", Science DOI: 10.1126/science.1151526 (November 20, 2007),]. As noted by the authors in their iPS cell research studies, the advantages of discovering and developing iPS cells include the facts that: no human embryos (or fetuses?) are involved; it permits the creation of patient-specific "stem cells" for transplant therapies; it allows for the creation of models of specific diseases; the cells can be used, e.g., for screening drugs, vaccines, chemical and biological agents, etc.; and the techniques are very simple to do. Thus the techniques could be easily copied by multiple research labs/clinics around the world and lead to "therapies" for millions of sick patients.

Since most basic research studies realistically resolve some existing concerns but not necessarily all, the researchers are generally open to and intellectually honest about noting themselves some of the on-going unresolved issues in their studies as well. For example, in the Yamanaka and Thomson iPS cell studies, it is acknowledged that foreign genes could end up in germ line cells (and thus be passed on through the generations), that tumor markers on iPS cells were picked up in assays, and that the possibility exists for the presence of unknown genetic errors and mutations in iPS cells that could also be passed on to patients. It is hoped that they and other researchers will also consider some additional questions about the research as identified below.

The purpose of this article is thus to briefly take a closer ethical and scientific look at these studies, as well as to raise serious concerns about the possibility/probability of multiple scientific artifacts inherent in them, thus rendering the data highly questionable and the "therapeutic" application of iPS cells to patients dangerous. Coming from a bench research background (career appointed biochemist/biologist, NCI/NIH, including basic research in cancer and antigen/antibody studies), as well as from an ethics background (Ph.D. in philosophy, Georgetown Univ. and KIE), the scientific and ethical issues involved in this research was of great interest to me. My approach will be to consider these iPS cell studies (briefly) from the point of view of two well-established ethical norms. The first is one that seems to have been largely abandoned within the research community itself lately, i.e., that the first requirement of ethical research is that the science used is as accurate and reliable as possible, and that the researcher(s) are academically and practically qualified to perform the research. Hence the "ethics" and the "science" are inherently intertwined. The second is drawn from some of the universal and ageless norms of natural law ethics, specifically that it is always wrong to intentionally kill innocent human beings (e.g., those involved in any way in either experimental or therapeutic research studies, or in clinical trials), even if purportedly for the benefit of other human beings. Both of these ethical norms have been universally accepted and concretized for many decades in several international ethical research guidelines, e.g., the Nuremberg Code and the Declaration of Helsinki (see Irving, "Biomedical research with 'decisionally incapacitated' human subjects: legalization of a defunct normative bioethics theory" (June 1998), at: Unfortunately, upon much closer scrutiny of these iPS studies themselves, it would also seem that both universal ethical norms have been neglected, and too many questions about these research studies remain unresolved or unaddressed.

II. Use of erroneous and vague scientific terms

As noted above, the first ethical consideration in doing research is that the science used is as accurate and reliable as possible. However, generally, one notices the use of certain erroneous and vague scientific terms in the Yamanaka and Thomson studies which could easily be misleading. For example, the scientific term "pluripotent" is always used to refer to both normal human embryonic stem cells and to iPS cells. However, the well-established scientific fact is that the blastomeres of the early human zygote and human morula are mostly totipotent (a range of totipotency). Even the cells of the inner cell mass of the human blastocyst - those at least referred to as "stem cells" - are a mixture of totipotent and pluripotent cells (also, a range of potencies). (See Irving, "Framing the Debates on Human Cloning and Human Embryonic Stem Cells: Pluripotent vs. Totopotent" (July 23, 2005), at:

Therefore, by falsely defining all of these cells scientifically as only "pluripotent", and by thus labeling tests and assays for "pluripotency" rather than for "totipotency", they thereby by-pass any ethical considerations of the purposeful production of innocent living human embryos (human beings) and their necessary destruction in order to retrieve their stem cells. They also thereby by-pass the fact that totipotent stem cells retrieved from living human embryos could possibly undergo "regulation" and revert to new living human embryos themselves (and be used as well in destructive research).

The studies also use the scientific term "undifferentiated" to refer to all human embryonic and iPS stem cells. However, early human embryonic stem cells are always differentiated to some degree - otherwise why do they function differently than cells derived from earlier human embryos, and how would they "know" how to differentiate later on into more specific cells, tissues and organs desired by the researchers? Even the position that each totipotent cell (blastomere) assumes in the developing embryo (in the morula, as well as in the blastocyst) requires specific differentiation. (See Bruce M. Carlson, Human Embryology and Developmental Biology (St. Louis, MO: Mosby, 1999, p. 41). So there is no such thing as "undifferentiated" early human embryonic stem cells, as if they were just superfluous and uninteresting non-entities, just biological "stuff".

Again, the use of the scientific phrase "immortal cell lines" that can reproduce indefinitely as applied to human embryonic and to iPS cells is rather "wishful thinking". There are no such things as "immortal cell lines", if by that phrase is meant that once such cells are produced and cultured they would continue to supply the researchers forever. As all lab technicians know, cell lines fairly quickly begin to genetically mutate and die off (called "senescence").

One also notices that the harder one attempts to identify the "smaller" details of the "materials and methods" sections of these studies, the vaguer they become. For example, an objective observer could not determine from the initials and other labels employed if any innocent living human embryos or fetuses were used in any way in these studies in any media, feeder cells, DNA-chips, assays, etc., that had been bought from commercial sources. If they were, then those commercial products themselves involved the destruction of human embryos and fetuses (i.e., innocent living human beings in their embryonic or fetal periods of development).

And it goes without saying that such research studies silently drag along with them all the baggage of false scientific terms used lately in other studies by their scientific colleagues to mis-define the early human embryo in order to render pressing ethical questions mute, e.g., "scientific" claims that the embryo is not an organism, a living human being, but rather just a "bunch of stem cells" or "a ball of cells". (See Irving, "What Human Embryo? Funniest Mental Gymnastics from Medicine and Research" (Oct. 14, 2004), at:

III. Do the scientific claims match the scientific details of their studies?

That said, it is perhaps more instructive to turn to the stated scientific and therapeutic claims of the researchers themselves to determine if they actually match the given details as found in those same studies.

A. No human embryos involved

The claim is made that iPS cell research solves all the divisive ethical issues because no human embryos are involved in the research. However, consider the following observations: (1) In order to prove that iPS cells are like human embryonic stem cells (and thus are viable substitutes for them), they must use human embryonic stem cells in the studies so that the two cell types can be accurately compared. Such "normal" human embryonic stem cells must have of necessity been derived by destroying innocent living human embryos and deriving the stem cells from them. Note too that both Thomson and Yamanaka used human embryonic or fetal cells (and/or components) (2) in culture and other media, (3) as "feeder cells", (4) for screening, (5) for "controls", (6) for material parts of technical assays (e.g., DNA chips), and (7) as sources of transcription factor genes themselves. Thomson also (8) used fetal cells as actual subjects of his transduction process (e.g., IMR90 fetal cells). He admits the inherent requirement for human embryos/fetuses and their components in a press comment: "Well, what I hope will not happen is that everybody says, 'See? We don't have to do embryonic stem cell research now.' ... In our research, we actually used human embryonic stem cells as part of the screening process. So the research itself on human embryonic stem cells led to the next finding about pluripotent cells," []. Yamanaka notes also (9) that "with the exception of cells at the edge", the assay for embryonic surface antigens are negative, and the assay for embryonic stem cell surface antigens are positive. Does his exception statement about "cells at the edge of the colonies" indicate that transformed human embryos were produced there, and that is why the assay for embryonic surface antigens on them was positive? Clearly, the claim that no human embryos are involved in such research is not true.

B. Creation of patient-specific "stem cells" for transplant therapies

As admitted by both researchers in their publications, these iPS cells cannot be used in patient "therapies" -- not only because transcription and viral foreign genes can cause tumors (as admitted in both studies), but also because both sources of foreign genes (virus and transcription factors) could cause foreign antigens on iPS cell surfaces that would in turn cause immune rejection reactions in patients - regardless if such genes were currently being expressed or not. Likewise, foreign surface antigens (both genetic and molecular) could also be caused in iPS cells by media used (e.g., mouse and primate media), feeder cells used for cultivation (e.g., fetal cells), inoculations of cell cultures, etc. (see details below). Therefore, such cells would never be "patient-specific", and could cause serious immune rejection reactions in the patients. These patients too are innocent living human beings whose dignity and safety must also be sustained.

In fact, there are so many artifacts and antigens introduced into the material and methods (as especially detailed by Yamanaka, but similarly assumed for Thomson, and as listed more fully below) that one must seriously question the accuracy and reliability of the data itself. Did the data presented in these studies ever pass a t-test? This is not clear at all in the details of the research studies.

C. Creation of models of specific diseases

Because of the incorporation of so many known foreign and even unknown antigens into these iPS cells and the multitude of artifacts inherent in the research, one wouldn't really know what disease one was studying - even if original disease cells were derived from a patient with a specific disease for the purpose of studying abnormal cell biochemistry. That is, the actual diseased cells of the patient would not match or function like the diseased iPS cells derived from them - like comparing apples and oranges. Thus how could such studies seriously provide "models of specific diseases"? Further, with so many unknowns, the data obtained would probably not be accurate, reliable, or repeatable, especially in other labs.

D. Use iPS cells to screen drugs, vaccines, chemical and biological agents, etc.

Perhaps "some" legitimate information could come out of such studies, e.g., how certain drugs, etc., function in iPS cells. However, as with "studying models of specific diseases", there are so many artifacts and foreign antigens involved in both studies that the data obtained in screening drugs would seem to be quite useless if it were extrapolated to "normal" cells, or to treat real diseased cells/patients.

E. The techniques are so simple to do

One look at Yamanaka's study, especially his "materials and methods" section, and it is clear that the study was not "simple". At least he was far more forthcoming with his details than was Thomson.

In sum, the various claims in the introductions of these iPS cell studies hardly match the scientific details buried in the body of the articles. Are they more fiction than fact?

IV. Admitted problems in the studies

However, credit must be given to the authors for admitting to some important problems inherent in their studies. Among them are the following.

A. Foreign genes could end up in germ line cells

Theoretically, there is no reason why they couldn't deliberately make ip-sperm and ip-oocytes, as well as ip-embryos. If ip-sperm and/or ip-oocytes are used to reproduce ip-embryos, the foreign genes (all of them) would become part of the genome of the germline cells of the ip-embryo, and expressed as antigens on the cell surfaces. The same applies if they created an ip-embryo outright by means of reprogramming. There is no theoretical impossibility for continuing the reprogramming of these cells back to the zygote (or earlier) stage of differentiation. These foreign genes would then also become part of the germline of the ip-embryo and be passed down through its generations.

Further, the fact that these iPS cells form chimeras when injected into blastocysts demonstrates that their foreign genes are or can still become active - even if not picked up in assays. One can only test for what one already is looking for.

B. Tumor markers on iPS cells are picked up in assays

Not unexpectedly, both researchers state in their studies that these iPS cells could not be used in patient therapies because of their ability to form tumors. However, the fact that these iPS cells could cause tumors in patients is not the same problem as the fact that these genes, and the non-tumoric foreign genes, cause specific antigens on the surface of iPS cells - regardless if such genes are being expressed at the time intra-cellularly. Such foreign antigens could thus cause rejection reactions in the patients.

C. Unknown genetic errors and mutations in iPS cells render "therapies" problematic

Even aside from the obvious and known induced antigens on these iPS cells, it is also acknowledged that other genetic mutations could have been caused in these cells but were just not picked up or tested for: Yamanaka: "Thus, each clone had more than 20 retroviral integration sites in total, which may increase the risk of tumorigenesis. ... This issue must be overcome to use iPS cells in human therapies. ... minor genetic alterations, which could not be detected by karyotype analyses, or epigenetic alterations are required for iPS cell induction. These issues need to be elucidated in future studies. ... Human iPS cells, however, are not identical to hES cells: DNA microarray analyses detected differences between the two pluripotent stem cell lines. Further studies are essential to determine whether human iPS cells can replace hES in medical applications" (emphases added). Thomson: "For transplantation therapies based on these cells, with the exception of autoimmune diseases, patient-specific iPS cell lines should largely eliminate the concern of immune rejection. It is important to understand, however, that before the cells can be used in the clinic, additional work is required to avoid vectors that integrate into the genome, potentially introducing mutations at the insertion site. ... However, further work is needed to determine if human iPS cells differ in clinically significant ways from ES cells" (emphases added).

As has already been suggested above, if there are so many "unknown" genetic mutations, antigens, or artifacts, there are not only obvious concerns about the use of these "patient-specific" cells in patients, but also about any validity of the data derived from "studies of disease mechanisms" and from screening drugs, vaccines, chemical and biological agents, etc.

V. Other possible sources of artifacts and foreign antigens

Although not specifically addressed in these studies, there are a host of other possible sources of artifacts and foreign antigens that could render the data presented as useless, as well as call into serious question the use of iPS cells in sick patients as "therapies". Although some of the following examples may be too "scientific" for many readers, at least it is worth registering and considering some of them.

In general, there are too many known and unknown variables throughout the experiments; in fact, there is, as noted in the studies, genetic variability even within the same iPS clone batches. It is also critically relevant to determine from the studies if the assay machines, gels, tests, etc., were reliably and properly calibrated before use, so that the data would be reliable and reproducible. Otherwise, the data presented in these studies are essentially useless. Again, the data derived from just one or two studies could hardly pass a t-test. It is not clear if the data presented in these studies were determined to be statistically significant or not.

Other possible sources of data artifacts and problematic antigens in these studies involves the viral genes used as vectors, as well as the transmission genes transferred. For example, Thomson used the following transmission genes, all of which would definitely show as problematic antigens on the surfaces of iPS cells produced with them: OCT4, SOX2, NANOG, and LIN28; Yamanaka used the following transmission genes: Oct3/4, Sox2, Klf4, and c-Myc. And the questions arises as to whether all of the transmission genes used in the studies were derived from genetically different human embryos or fetuses? That too could result in artifacts in the data presented, as well as the presence of more foreign antigens on the surfaces of the iPS cells produced.

Also to be considered are foreign genes or antigenic molecules that could be inadvertently derived from the media, inoculations and feeder cell layers used, as well as from impurities in the several manufactured biological products, laboratory contamination, etc.

The above are simply lists of standard good quality research controls used throughout the various research fields. It would be important, then, to turn again to the specific research studies to determine any specific sources of artifacts and antigens.

A. Yamanaka et al

As noted, Yamanaka et al used the following transmission genes: Oct3/4, Sox2, Klf4, and c-Myc. All of these genes, if incorporated into the iPS cells, would cause foreign genetic antigens on the iPS cell surfaces. As already noted, one also wonders if all of these "normal" transmission genes were derived from different human embryos or fetuses. One cannot assume, e.g., that the Oct3/4 transmission gene from one human embryo is genetically identical to that of other human embryos. If such genetic differences exist, then they could cause additional foreign antigens on the surface of iPS cells.

Viral genes from the vectors used could also be incorporated into the genome of iPS cells, also causing foreign antigens on the cell surfaces. For Yamanaka, all four retroviruses used to transmit the "normal" human embryonic genes could cause foreign antigens, not to mention specifically the amphotropic retrovirus, the lentivirus, and the ecotropic retrovirus used. Viruses have genes, too!

And for brevity, the following is a further listing of sources of data artifacts, foreign antigens in the iPS cells produced, and human embryonic/fetal cell sources (at least as much as can be determined solely from the studies):

B. Thomson

Whereas Yamanak's study provided many (but not all) details of their materials and methods, the same is not true for Thomson's study. In fact, there was a dirth of "methods and procedures" details. Therefore, it is not possible to identify the same kind of list as above. However, it is still obvious that foreign genetic cell surface antigens would be caused in their experimental iPS cells because the cells were infected and transformed with foreign transmission genes (OCT4, SOX2, NANOG, and LIN28), using four different viral vectors whose own genes would be foreign.

VI. Conclusion

Considering the enormous stakes involved, it would seem that it is incumbent upon scientists involved in especially ethically sensitive research to be as open and detailed in their publications as possible. The trend, however, seems to be to camouflage and dilute the scientific details as much as possible -- not only in order to evade professionally appropriate questions from their scientific peers, but also to evade the very questions that society and ethics have traditionally required of all scientists. Rather than use and report the accurate scientific facts, it would seem that scientists prefer to secure their successes by using false and misleading manufactured "scientific" terms, verbal hype, and empty promises. Even the most basic requirements of research ethics appear to have been abandoned, including the required use of the most accurate and reliable scientific facts, as well as the "denial" of the age-old dictum that it is simply wrong to purposefully kill innocent human beings - regardless of their stage of development, ethnicity, culture, degree of illness, etc., and regardless of whether or not their destruction could be of benefit to other human beings. Needless to say, even the most desperate of sick patients should not be exposed to "therapeutic" research or clinical trials when such participation unwittingly puts them into serious danger of harm and even death. One even wonders how such patients could give legally valid "informed consent".

We would all obviously welcome a viable scientific and ethical resolution to the divisive politics of human embryo and fetal research that has consumed us for so many years. These iPS studies, however, do not appear to be that solution.