Portal de Periódicos da CAPES

Why Myc? An Unexpected Ingredient in the Stem Cell Cocktail

2008; Elsevier BV; Volume: 2; Issue: 1 Linguagem: Inglês

10.1016/j.stem.2007.12.004

1934-5909

Paul S. Knoepfler,

Renal and related cancers

Screening cocktails of candidate genes for induction of pluripotency and self-renewal in nonstem cells has identified a surprising new embryonic stem cell regulator, the myc proto-oncogene. Here the possible mechanisms by which myc controls self-renewal and pluripotency are discussed. Screening cocktails of candidate genes for induction of pluripotency and self-renewal in nonstem cells has identified a surprising new embryonic stem cell regulator, the myc proto-oncogene. Here the possible mechanisms by which myc controls self-renewal and pluripotency are discussed. Over the years, the puzzle of the molecular determinants of pluripotency and self-renewal in stem cells has been approached in a gene-by-gene fashion. This piecemeal method used conventional functional studies of genes whose expression patterns or knockout (KO) phenotypes suggested possible stem cell functions. Several years ago, gene expression screens, including microarrays, functional libraries, and differential display, began to be used to identify genes enriched in stem cells versus nonstem cells. Despite the strength of these screening methods in their unbiased, relatively global searching capacity, they produced very long lists of genes enriched in stem cells. The prospect of fishing out new stem cell regulatory molecules from this sea of genes seemed daunting. Nonetheless, some familiar faces appeared near the top of such lists, including Oct3/4 and Sox2, suggesting the methods and data were inherently valid (Kelly and Rizzino, 2000Kelly D.L. Rizzino A. Mol. Reprod. Dev. 2000; 56: 113-123Crossref PubMed Scopus (104) Google Scholar, Ramalho-Santos et al., 2002Ramalho-Santos M. Yoon S. Matsuzaki Y. Mulligan R.C. Melton D.A. Science. 2002; 298: 597-600Crossref PubMed Scopus (1388) Google Scholar). Many other genes were also on these lists that had not been previously thought to be important stem cell regulators. For example, the N-myc (also MYCN) proto-oncogene was very highly enriched in a number of stem cells (Kelly and Rizzino, 2000Kelly D.L. Rizzino A. Mol. Reprod. Dev. 2000; 56: 113-123Crossref PubMed Scopus (104) Google Scholar, Ramalho-Santos et al., 2002Ramalho-Santos M. Yoon S. Matsuzaki Y. Mulligan R.C. Melton D.A. Science. 2002; 298: 597-600Crossref PubMed Scopus (1388) Google Scholar), including hematopoietic as well as both human and murine embryonic stem cells (HSCs, hESCs, and mESCs, respectively). However, unlike Nanog, both N-myc and its more famous relative c-myc had been intensively studied for almost two decades, raising the question of what if anything was already known about myc function in stem cells. In fact, around the same time as the microarray studies, c- and N-myc conditional KO mice with disruption targeted to stem cells were used to explore myc function in somatic stem cells (Knoepfler et al., 2002Knoepfler P.S. Cheng P.F. Eisenman R.N. Genes Dev. 2002; 16: 2699-2712Crossref PubMed Scopus (385) Google Scholar, Wilson et al., 2004Wilson A. Murphy M.J. Oskarsson T. Kaloulis K. Bettess M.D. Oser G.M. Pasche A.C. Knabenhans C. Macdonald H.R. Trumpp A. Genes Dev. 2004; 18: 2747-2763Crossref PubMed Scopus (554) Google Scholar). For example, KO studies indicated N-myc was absolutely required for normal neural stem cell (NSC) function (Knoepfler et al., 2002Knoepfler P.S. Cheng P.F. Eisenman R.N. Genes Dev. 2002; 16: 2699-2712Crossref PubMed Scopus (385) Google Scholar). c-myc appeared to influence HSC interaction with the niche such that the KO resulted in a surprising increase in HSC populations (Wilson et al., 2004Wilson A. Murphy M.J. Oskarsson T. Kaloulis K. Bettess M.D. Oser G.M. Pasche A.C. Knabenhans C. Macdonald H.R. Trumpp A. Genes Dev. 2004; 18: 2747-2763Crossref PubMed Scopus (554) Google Scholar). These unexpected results and the coexpression of both c- and N-myc in HSCs leave the precise nature of myc's role in HSC function open at this point. Subsequently, Dalton's group demonstrated an essential role for c-myc in normal LIF signaling in mESCs and showed that enforced c-myc expression conferred LIF-independent ESC growth (Cartwright et al., 2005Cartwright P. McLean C. Sheppard A. Rivett D. Jones K. Dalton S. Development. 2005; 132: 885-896Crossref PubMed Scopus (567) Google Scholar), sparking great interest in myc function specifically in ESCs. Defining the unique molecular determinants of stem cell function, particularly that of ESCs, is critically important for understanding how to maintain or manipulate stem cell behavior. However, such information also brings the exciting possibility of converting nonstem cells into stem cells. Such a transformation would have enormous therapeutic implications for regenerative medicine, particularly personalized stem cell therapies. Yamanaka's group provided a major leap in that direction in their studies that introduced mixtures or "cocktails" of genes into murine fibroblasts to measure whether any of the combinations of their 24 total candidate factors could yield what they termed "induced pluripotent stem (iPS) cells" (Takahashi and Yamanaka, 2006Takahashi K. Yamanaka S. Cell. 2006; 126: 663-676Abstract Full Text Full Text PDF PubMed Scopus (16979) Google Scholar). Because Dalton had implicated myc in ESC function, the Yamanaka lab included myc in the pool of 24 candidates. This approach paid off when they hit upon one special cocktail of four factors, which together with selection for Fbx15 expression, could produce the desired iPS cell production. The four genes included two of the most well-established ESC factors, Oct3/4 and Sox2, and also two surprising new additions: Klf4 and c-myc. Several recent papers have recapitulated and extended the initial iPS cell murine studies (Maherali et al., 2007Maherali N. Sridharan R. Xie W. Utikal J. Eminli S. Arnold K. Stadfeld M. Yachechko R. Tchieu J. Jaenisch R. Cell Stem Cell. 2007; 1: 39-49Abstract Full Text Full Text PDF PubMed Scopus (1336) Google Scholar, Okita et al., 2007Okita K. Ichisaka T. Yamanaka S. Nature. 2007; 448: 313-317Crossref PubMed Scopus (3359) Google Scholar, Wernig et al., 2007Wernig M. Meissner A. Foreman R. Brambrink T. Ku M. Hochedlinger K. Bernstein B.E. Jaenisch R. Nature. 2007; 448: 260-262Crossref PubMed Scopus (2136) Google Scholar). The first iPS cells lacked the ability to contribute to the germline when injected into blastocysts. In contrast, the newer iPS cells containing the same four factors now also selected for Nanog expression instead of Fbx15 were more fully reprogrammed and possessed the ability to contribute to the germline. What can Nanog do that Fbx15 cannot? A key, unique attribute of the newer iPS cells is their overall ESC signatures of chromatin modifications. This epigenetic programming in turn appears responsible for ESC-like gene expression patterns present only in the newer iPS cells and induced by Nanog in some as yet uncharacterized manner. Thus, although Nanog is dispensable for iPS cell formation per se, it is essential for full reprogramming. Two recent studies from the groups of Yamanaka and Thomson have succeeded in generating iPS cell clones by direct reprogramming of human fibroblasts (Takahashi et al., 2007Takahashi K. Tanabe K. Ohnuki M. Narita M. Ichisaka T. Tomoda K. Yamanaka S. Cell. 2007; 131: 861-872Abstract Full Text Full Text PDF PubMed Scopus (13487) Google Scholar, Yu et al., 2007Yu J. Vodyanik M.A. Smuga-Otto K. Antosiewicz-Bourget J. Frane J.L. Tian S. Nie J. Jonsdottir G.A. Ruotti V. Stewart R. et al.Science. 2007; (in press. Published online November 20, 2007)https://doi.org/10.1126/science.1151526Crossref Scopus (7585) Google Scholar). These findings were met with a wide fanfare of media attention, as they bring the stem cell field one significant step closer to therapeutic development. Progress toward answering some key questions about myc function in iPS cell formation has now been reported in additional papers from the Yamanaka and Jaenisch groups. These studies report iPS cell production with a three-factor cocktail that does not include myc (Nakagawa et al., 2007Nakagawa M. Koyanagi M. Tanabe K. Takahashi K. Ichisaka T. Aoi T. Mochiduki Y. Takizawa N. Yamanaka S. Nat. Biotechnol. 2007; (in press. Published online November 30, 2007)https://doi.org/10.1038/nbt1374Crossref PubMed Scopus (2083) Google Scholar, Wernig et al., 2008Wernig M. Meissner A. Jaenisch R. Cell Stem Cell. 2008; 2 (this issue): 10-12Abstract Full Text Full Text PDF PubMed Scopus (501) Google Scholar) and was possible also in human iPS cells (Nakagawa et al., 2007Nakagawa M. Koyanagi M. Tanabe K. Takahashi K. Ichisaka T. Aoi T. Mochiduki Y. Takizawa N. Yamanaka S. Nat. Biotechnol. 2007; (in press. Published online November 30, 2007)https://doi.org/10.1038/nbt1374Crossref PubMed Scopus (2083) Google Scholar). The efficiency of the iPS cell process without myc is dramatically reduced and appears to progress much more slowly. Indeed, under standard antibiotic selection, the iPS cell process failed without myc. In half of the experiments conducted over an extended period of selection, no iPS cell colonies at all were produced when myc was excluded from the reprogramming protocol. An example of successful iPS cell formation without myc, from adult tail tip fibroblasts, nonetheless yielded iPS cells with an almost 500-fold reduced efficiency, demonstrating that, at least under these assay conditions, myc fulfils an important role in the direct reprogramming process (Nakagawa et al., 2007Nakagawa M. Koyanagi M. Tanabe K. Takahashi K. Ichisaka T. Aoi T. Mochiduki Y. Takizawa N. Yamanaka S. Nat. Biotechnol. 2007; (in press. Published online November 30, 2007)https://doi.org/10.1038/nbt1374Crossref PubMed Scopus (2083) Google Scholar). Thus, the ability to routinely and efficiently generate iPS cells may be myc dependent. Of great importance is the observation that mice generated from these three-factor iPS cells did not develop any tumors within 100 days (Nakagawa et al., 2007Nakagawa M. Koyanagi M. Tanabe K. Takahashi K. Ichisaka T. Aoi T. Mochiduki Y. Takizawa N. Yamanaka S. Nat. Biotechnol. 2007; (in press. Published online November 30, 2007)https://doi.org/10.1038/nbt1374Crossref PubMed Scopus (2083) Google Scholar), indicating that not all iPS cell-derived mice are prone to tumors. Although long-term studies of tumorigenicity in these mice have yet to be conducted, this finding is nonetheless a significant advance in that it establishes myc as a key determinant of iPS cell tumorigenicity and suggests the tumorigenicity of iPS cells can be reduced. In one of the reports of human iPS cells, Thomson's group tested the ability of quartets of factors from a group of 14 candidates, importantly not including c-myc, to induce human iPS cell formation. They found that Oct3/4, Sox2, Nanog, and Lin28 could reprogram human neonatal somatic cells (Yu et al., 2007Yu J. Vodyanik M.A. Smuga-Otto K. Antosiewicz-Bourget J. Frane J.L. Tian S. Nie J. Jonsdottir G.A. Ruotti V. Stewart R. et al.Science. 2007; (in press. Published online November 20, 2007)https://doi.org/10.1126/science.1151526Crossref Scopus (7585) Google Scholar), although Lin28 was not strictly required, and iPS cell induction of fibroblasts from adult human cells was not tested. Still, this study provides yet another example of induced pluripotency without exogenous c-myc. However, because Thomson did not include c-myc in their group of 14 candidate factors, it is unclear whether c-myc would have been selected as a component of their successful screen. Given the findings from the most recent Yamanaka paper, it would be interesting to determine whether the addition of c-myc as a fifth factor in Thomson's cocktail would substantially boost the efficiency of human iPS cell formation. An additional unresolved point raised by the recent Thomson and Yamanaka results is whether the iPS cells generated without exogenous myc have, in fact, been selected in culture based partially on their endogenous myc levels. Indeed, the issue of endogenous myc complicates interpretation of the observations of iPS cell formation without added myc. myc may very well be indispensable for iPS cell formation, as iPS cells produced without added myc continue to express endogenous c-myc (Nakagawa et al., 2007Nakagawa M. Koyanagi M. Tanabe K. Takahashi K. Ichisaka T. Aoi T. Mochiduki Y. Takizawa N. Yamanaka S. Nat. Biotechnol. 2007; (in press. Published online November 30, 2007)https://doi.org/10.1038/nbt1374Crossref PubMed Scopus (2083) Google Scholar), which may in effect fill in for the omitted exogenous c-myc, albeit with a lower efficiency. Also it remains unknown if iPS cells generated without added c-myc may have been selected for high levels of endogenous N- or L-myc (Blelloch et al., 2007Blelloch R. Venere M. Yen J. Ramalho-Santos M. Cell Stem Cell. 2007; 1: 245-257Abstract Full Text Full Text PDF PubMed Scopus (206) Google Scholar, Nakagawa et al., 2007Nakagawa M. Koyanagi M. Tanabe K. Takahashi K. Ichisaka T. Aoi T. Mochiduki Y. Takizawa N. Yamanaka S. Nat. Biotechnol. 2007; (in press. Published online November 30, 2007)https://doi.org/10.1038/nbt1374Crossref PubMed Scopus (2083) Google Scholar), now also proven to drive iPS cell formation much the same as c-myc. Determining the mechanisms by which myc drives efficient iPS cell formation may open the door to finding ways to efficiently produce iPS cells without myc, for example by substituting another nontumorigenic factor or by treatment of cells with a pharmacological agent. Such an effort should be greatly aided by comparisons of existing iPS cells generated with and without added c-myc. While studies addressing the open questions outlined above are conducted, we can model the role played by myc in the iPS cell process based on what is already known. c-myc was one of the first proto-oncogenes identified, and the myc family, including N-myc and L-myc, has proven to have a potent role in most human cancers when expression is deregulated. How could a proto-oncogene contribute to pluripotency or self-renewal of ESCs? Three models are proposed (Figure 1): (1) inducing a cell-cycle program necessary specifically for self-renewal, (2) modifying epigenetic patterns to promote dedifferentiation or block additional differentiation, and (3) selection of a rare population of cells with predetermined traits suited to permit induced pluripotency and self-renewal. Recent studies on myc function in somatic stem cells as well as ESCs support aspects of each model as outlined below and, together, most strongly support a combination model incorporating aspects of each of the specific hypotheses (Figure 1). Self-renewal is an essential hallmark of stem cell function and necessitates entrance into the cell cycle, after which at least one resulting daughter cell maintains the parental differentiation capacity as well as the potential for subsequent self-renewing divisions. In this first model, myc pushes differentiated cells to enter the cell cycle in a manner consistent with self-renewal, as required to achieve pluripotency. One or more of the other three factors in the cocktail may cooperate with myc to sustain cycling that promotes self-renewal. This model is particularly attractive given the compelling links between myc and the cell cycle in a host of different cell types, but is there evidence of regulation of the cell cycle specifically in stem cells by myc? This is, in fact, the case, as studies of a conditional KO of the N-myc gene in NSCs using nestin-cre indicated that N-myc NSC KO mice displayed a failure of normal brain growth (Knoepfler et al., 2002Knoepfler P.S. Cheng P.F. Eisenman R.N. Genes Dev. 2002; 16: 2699-2712Crossref PubMed Scopus (385) Google Scholar). This phenotype was traced to a reduction in NSC populations that was in turn linked to disruptions of an NSC cell-cycle regulatory program. Loss of N-myc resulted in high levels of the cyclin-dependent kinase inhibitors (CDKI) p18INK4c and p27KIP1 and strikingly decreased expression of cyclin D2. Thus, a key role for N-myc in NSCs is control of a gene expression program involving cell-cycle regulators that is central for maintenance of self-renewal and inhibition of differentiation. myc may induce a stem-like cell-cycle program in fibroblasts similar to what was observed in NSCs whereby myc represses CDKI and stimulates D cyclins. The unique cell-cycle regulatory pathway in ESCs may be explained by a combination of low CDKI levels, high Cdk2 activity, and Rb hyperphosphorylation together maintained by elevated myc (Cartwright et al., 2005Cartwright P. McLean C. Sheppard A. Rivett D. Jones K. Dalton S. Development. 2005; 132: 885-896Crossref PubMed Scopus (567) Google Scholar). Also importantly, c-myc, albeit exogenous, sustained self-renewal in ESCs without strongly impinging on the cell cycle and was postulated to function by blocking differentiation (Cartwright et al., 2005Cartwright P. McLean C. Sheppard A. Rivett D. Jones K. Dalton S. Development. 2005; 132: 885-896Crossref PubMed Scopus (567) Google Scholar). Results generated with N-myc are likely to extend to c-myc, given that N-myc could replace the requirement for c-myc in an iPS cell-inducing cocktail (Blelloch et al., 2007Blelloch R. Venere M. Yen J. Ramalho-Santos M. Cell Stem Cell. 2007; 1: 245-257Abstract Full Text Full Text PDF PubMed Scopus (206) Google Scholar, Nakagawa et al., 2007Nakagawa M. Koyanagi M. Tanabe K. Takahashi K. Ichisaka T. Aoi T. Mochiduki Y. Takizawa N. Yamanaka S. Nat. Biotechnol. 2007; (in press. Published online November 30, 2007)https://doi.org/10.1038/nbt1374Crossref PubMed Scopus (2083) Google Scholar). Analyzing the cell-cycle properties of the different iPS cell lines generated with and without varying levels of myc and comparing them to ESC should provide further insight into this model. According to this second paradigm, myc contributes to iPS cell formation by inducing chromatin changes required for pluripotency (reviewed in Yamanaka, 2007Yamanaka S. Cell Stem Cell. 2007; 1: 39-49Abstract Full Text Full Text PDF PubMed Scopus (587) Google Scholar). Although the Myc protein has long been modeled as a classic gene-specific transcription factor that acts strictly at the promoters of genes via its histone acetyltransferase cofactors, two recent lines of investigation challenge this assumption. Several studies of genomic binding in cell lines consistently indicate that Myc exhibits a strikingly widespread profile of DNA binding, likely to tens of thousands of sites (reviewed in Knoepfler, 2007Knoepfler P.S. Cancer Res. 2007; 67: 5061-5063Crossref PubMed Scopus (103) Google Scholar). Unfortunately, none of these studies were conducted in stem cells. However, Myc was also recently found to possess global chromatin activating function in NSCs (Knoepfler et al., 2006Knoepfler P.S. Zhang X.Y. Cheng P.F. Gafken P.R. McMahon S.B. Eisenman R.N. EMBO J. 2006; 25: 2723-2734Crossref PubMed Scopus (307) Google Scholar), consistent with its widespread genomic binding in other cell types. Indeed, a striking phenotype of N-myc-deficient NSCs was a profound alteration of their nuclear structure and global histone modifications. NSCs lacking N-myc had condensed, highly heterochromatic nuclei, indicating that N-Myc plays a normal role in the widespread maintenance of key components of active chromatin, including acetylated histone H3 and H4 as well as H3 methylated at lysine 4. Global chromatin inactivation in tumors when myc was shut off (Wu et al., 2007Wu C.H. van Riggelen J. Yetil A. Fan A.C. Bachireddy P. Felsher D.W. Proc. Natl. Acad. Sci. USA. 2007; 104: 13028-13033Crossref PubMed Scopus (290) Google Scholar) is consistent with the global chromatin function for myc reported in NSCs. A widespread impact on chromatin function by Myc could contribute to iPS cell formation through direct activation of genes important for pluripotency or self-renewal, as well as blockage of differentiation. Support for this hypothesis comes from studies of ESCs, where decreased myc activity induced differentiation (Cartwright et al., 2005Cartwright P. McLean C. Sheppard A. Rivett D. Jones K. Dalton S. Development. 2005; 132: 885-896Crossref PubMed Scopus (567) Google Scholar), and from analyses of N-myc-deficient NSCs, which exhibit enhanced neuronal differentiation. Reprogramming fibroblasts to iPS cells is a taller order than blocking differentiation, likely requiring the reversal of a pre-existing differentiated state by excess myc. Such a process would also unfortunately bring the cells one step closer to oncogenic transformation. Alternatively, Myc may facilitate the iPS cell process by setting the stage, at a chromatin level, for subsequent activity of other factors (see combination model below) such as Nanog, Oct3/4, Sox2, or Klf4, as chromatin structure appears of central importance to ESC biology (Bernstein et al., 2006Bernstein B.E. Mikkelsen T.S. Xie X. Kamal M. Huebert D.J. Cuff J. Fry B. Meissner A. Wernig M. Plath K. et al.Cell. 2006; 125: 315-326Abstract Full Text Full Text PDF PubMed Scopus (3738) Google Scholar). It is also possible that among millions of fibroblasts lurk rare stem-like cells that are not fully differentiated and represent the only target cells amenable to direct reprogramming into iPS cell clones. This interpretation could explain why less than 1% of cells transduced with the four factors are selected in iPS cell induction protocols (Takahashi and Yamanaka, 2006Takahashi K. Yamanaka S. Cell. 2006; 126: 663-676Abstract Full Text Full Text PDF PubMed Scopus (16979) Google Scholar). Such pluripotency-competent or permissive cells may require only inhibition of differentiation rather than its reversal by myc. If such a unique nonfibroblast pluripotency-competent cell were indeed the true source of iPS cells, it would suggest there may be a higher hurdle for developing potential future patient-specific regenerative medicine therapies, as the bulk of available fibroblasts are not able to do the trick. Thus, resolving the cell of origin of iPS cells is of great importance. However, the rarity of iPS cell events may instead be related to technical issues such as a dependence on specific viral integration sites, which may influence the expression levels of the four genes. For example, levels of myc that are too high are likely to trigger apoptosis (Cartwright et al., 2005Cartwright P. McLean C. Sheppard A. Rivett D. Jones K. Dalton S. Development. 2005; 132: 885-896Crossref PubMed Scopus (567) Google Scholar), whereas very low expression would inhibit induction of pluripotency. Finally, as yet unidentified factors, expressed in rare cells in the fibroblast population, may also be essential. Testing other cell types besides fibroblasts for their ability to be transformed into iPS cells should address some of these issues and may lead to the discovery of a cell of origin with iPS cell forming capacity dramatically higher than fibroblasts. A fourth possibility as to the role of myc in iPS cells is that some combination of the three models could be operative, given that the individual components are not mutually exclusive. For example, Myc may cooperate with Klf4 to induce and maintain self-renewal by modifying the cell cycle, possibly via blocking apoptosis (Ghaleb et al., 2007Ghaleb A.M. Katz J.P. Kaestner K.H. Du J.X. Yang V.W. Oncogene. 2007; 26: 2365-2373Crossref PubMed Scopus (83) Google Scholar). In addition, Myc may also modulate widespread chromatin, making it competent for the action of other factors, including Nanog, Oct3/4, and Sox2 (Figure 1, bottom). Functional genomic studies of Myc in stem and iPS cells will likely soon resolve how Myc regulates chromatin and its possible importance for stem cell biology. Comparing data from such studies with genomic studies on Oct3/4, Sox2, Klf4, and Nanog should elucidate their potential cooperative activities on chromatin and gene expression. Another important step for addressing a combinatorial model is an analysis of whether the order in which these factors are introduced influences iPS cell formation. In the context of emerging regenerative medicine, myc has the potential to play a "Dr. Jekyll/Mr. Hyde" or dual role in stem cells. Some minimal level of myc expression is almost certain to be essential for normal stem cell-mediated regenerative tissue growth, and in that way, myc has a good side. However, even a modest excess of myc, whether endogenous or exogenous, in transplanted stem cells could be disastrous, as it may cause tumors in recipient patients, possibly by promoting cancer stem cell formation. To obtain the best of both worlds, another notion is to conditionally express myc in transplanted stem cells by using methods that allow its inactivation once regenerative growth has been achieved. An inducible system such as a tetracycline-controlled myc transgene is one example of a regulated modification. Indeed, in mouse models of this kind, myc-induced tumors often disappear permanently once myc is shut off, demonstrating the so-called "oncogene addiction" phenomenon (Wu et al., 2007Wu C.H. van Riggelen J. Yetil A. Fan A.C. Bachireddy P. Felsher D.W. Proc. Natl. Acad. Sci. USA. 2007; 104: 13028-13033Crossref PubMed Scopus (290) Google Scholar). Unfortunately, the loss of myc does not always lead to sustained murine tumor regression, underscoring the potential risk of utilizing myc for the purpose of stem cell therapy. Further, mice derived from the Nanog-iPS cells, in which all four genes introduced in the induction cocktail had spontaneously turned "off," exhibited a substantial risk for tumors due to spontaneous reactivation of c-myc virus (Okita et al., 2007Okita K. Ichisaka T. Yamanaka S. Nature. 2007; 448: 313-317Crossref PubMed Scopus (3359) Google Scholar). Thus, any elevation of myc levels, whether exogenous or resulting from selection of the endogenous myc gene could make iPS cells prone to induce tumors in patients receiving cell therapy. The importance of myc for efficient iPS cell formation and its key role in the biology of both ESCs and many somatic stem cells suggest shutting myc off altogether in stem cells—ESCs or iPS cells—prior to transplant to attenuate the risk of cancer as a side effect is unlikely to be compatible with successful regenerative medicine, because it will almost certainly yield nonfunctional stem cells. Thus the key to safely utilizing stem cells in regenerative medicine may reside in maintaining the appropriate expression levels, neither too high nor too low, of myc. For customized regenerative medicine using future iPS cell-like cells, determining just how much myc, whether endogenous or exogenous, is both safe and effective as well as how to sustain that level may be a necessary albeit challenging hurdle to jump. In the end, it seems one way or another the field of regenerative medicine will have to come to terms with myc, a previously unanticipated friend and foe, to develop both safe and effective therapies. I thank Steve Dalton for reading the manuscript prior to publication. A related paper showing that iPS cells can be generated from human ES-, fetal-, and adult-derived somatic cells, using three to six defined factors, was published after this article was completed (Park, I.-H., Zhao, R., West, J.A., Yabuuchi, A., Huo, H., Ince, T.A., Lerou, P.H., Lensch, M.W., and Daley, G.Q. [2008]. Reprogramming of human somatic cells to pluripotency with defined factors. Nature, in press. Published online November 23, 2008. 10.1038/nature06534).