As drug makers try to rescue dying neurons, others hold out hope that supplying fresh ones may be a better way to combat Alzheimer and other neurodegenerative diseases. The latter strategy may seem even more tantalizing in light of recent improvements to a method for producing the precursor cells that could give rise to specialized neurons for replenishing shriveling brains. Boston researchers describe a new protocol for generating these so-called “induced pluripotent stem (iPS) cells” in a report released yesterday in Science online.

Efforts to regenerate functional, patient-specific tissue for therapeutic purposes have traditionally focused on embryonic stem (ES) cells, a coveted population with the potential to yield any mature cells lost to disease. However, ethical concerns regarding the source of human ES cells have hampered their study and therapeutic development. Hoping to sidestep this controversy, scientists have devised ways to make ES-like cells by taking more readily available adult tissue, such as skin, and turning back its developmental clock. In mice, fibroblasts can be made to revert to an embryonic-like state by introducing just four transcription factors—Oct4, Sox2, c-Myc, and Klf4 (Takahashi and Yamanaka, 2006). Encouragingly, reprogrammed skin cells (or iPS cells) have shown therapeutic promise in rodent models of Parkinson disease (Wernig et al., 2008) and sickle cell anemia (Hanna et al., 2007 ). As further proof of principle, scientists have recently generated patient-specific motor neurons from iPS cells that they made using skin samples from two amyotrophic lateral sclerosis (ALS) sufferers (see ARF related news story). However, a nagging problem has plagued iPS technology. The viruses traditionally used to deliver the reprogramming factors can integrate into the host genome and increase risk of tumor formation (Okita et al., 2007).

To get around this snag, researchers led by Konrad Hochedlinger at Massachusetts General Hospital, Boston, set out to generate mouse iPS cells using non-integrating adenoviral vectors to deliver the quartet of reprogramming genes. After failed initial attempts with this method in tail-tip fibroblasts, first author Matthias Stadtfeld and colleagues switched from skin to liver—a more easily programmable cell type. To start off, the researchers used mouse fetal liver cells carrying an inducible Oct4 allele, and infected them with adenoviruses expressing the other three transcription factors (Sox2, c-Myc, and Klf4). From 500,000 cells initially plated for infection, this strategy yielded nine iPS-like colonies. The triple-infection method also worked using tail fibroblasts carrying the inducible Oct4 allele, but at much lower efficiency: more than a million plated cells gave rise to just one colony that could be expanded into stable ES-like cell lines. Hence, when it came time to drop the Oct4 transgene and re-attempt infecting adult somatic cells with all four transcription factors, the researchers went back to liver. From a starting batch of 500,000 mouse liver cells, three colonies were successfully expanded into stable ES-like cell lines expressing the pluripotency markers Oct4 and stage-specific embryonic antigen 1 (SSEA1).

These adeno-iPS cells exhibited gene expression and demethylation patterns characteristic of ES cells, and had no detectable genomic integration of adenoviral vector. When injected into SCID mice (which have severe combined immunodeficiency that prevents transplant rejection), all tested adeno-iPS cell lines appeared pluripotent, as evidenced by the production of teratomas that differentiated into representative cell types of the three germ layers. In addition, adeno-iPS cells injected into blastocysts contributed well to multiple tissues in the resultant chimeras, including germ cells. Compared to iPS cells previously made using integrating viruses, the adeno-iPS seemed to look and behave similarly—without the side effects. Between four and 13 weeks of age, none of the coat color chimeras derived from adeno-iPS cells had observable tumors.

“I think it's a pretty big step forward,” said Mahendra Rao, vice president of research in stem cells and regenerative medicine at Invitrogen in Frederick, Maryland. The new study settles the debate as to whether iPS cells can be generated without insertional mutagenesis, said Rao, who several years ago led an ARF live discussion on stem cell therapy for neurodegenerative disease. “It opens the gates to say that other transient methods may be sufficient as well.”

Other scientists are reserving their enthusiasm. “It's an incremental biological advance, but not a practical advance,” said Evan Snyder, director of the stem cells and regeneration program at Burnham Institute for Medical Research in La Jolla, California. The data is “in line with finding safer alternative techniques, but it doesn't at this stage bring us any closer to therapies.”

Snyder noted several concerns with the adenovirus-based protocol. One was the presence of abnormal (tetraploid) cells in 23 percent of the adeno-iPS lines—a problem not seen with retro- or lentiviral vectors. More disturbing was the new method’s abysmal efficiency, which is 0.001 percent at best—10- to 100-fold lower than the frequency of deriving iPS-like cells using integrating viruses.

The authors also acknowledge these shortcomings. However, the tetraploidy issue doesn’t loom so large in Stadtfeld’s view. “If you apply this in a clinical setting, you'd just have to generate enough (iPS cells) to screen them for ploidy or other abnormalities,” he told ARF. The current findings might also reflect higher rates of tetraploidy in liver cells, he said. “If you used skin cells, which are mostly diploid, you might end up with a completely different picture.”

Of course, the efficiency could take a nosedive with a switch from liver to skin cells. Stadtfeld and colleagues are addressing this concern by screening for chemical compounds that accelerate the transition from infection to reprogramming in the host cells. From the time of infection, mouse and human skin cells typically take eight to 12 days to revert to an embryonic-like state. If the non-integrating viruses only hang around for four to five days, Stadtfeld said that could explain why skin cells cannot be reprogrammed. However, a small molecule that hastened the reprogramming process to a mere five to six days might significantly boost efficiency, he told ARF.

“I believe it's just a matter of time before human cells without viral integration are generated,” he said. “Whether this is done by adenoviruses or by a combination of viruses and chemicals, I cannot say. I would assume it would ultimately be a combination.”—Esther Landhuis


  1. This work builds on the work by Yamanaka and others showing that relatively short periods of exposure and/or sequential exposure to reprogramming signals is sufficient to transform cells into pluripotent cells. This raised the possibility that episomal/transient vectors, protein transduction strategies and small molecules may work.

    In this manuscript the authors have shown that inducible adenovirus persists for sufficiently long periods to reprogram cells and as such minimizes risks associated with nonrandom integration and disruption of potentially important genes. These induced cells appeared similar to cells derived by other integrating methods and were capable of robust chimera formation.

    While clearly an important step forward, several issues remain. The authors note that adenovirus infection efficiency is variable in different cell types. Persistence and levels of expression are variable as well, and both of these likely reduce the efficiency of reprogramming. Indeed, the authors used liver cells for their experiments as these are much more efficiently infected with adenovirus as compared to fibroblasts. Experiments were performed with rodent cells, and it is unclear if the longer cycle time and longer period of induction to pluripotency required will represent a benefit or a hindrance to this methodology. Some of the adeno-associated viruses integrate into the genome, albeit at a very low frequency, and it will be important to test for such integration.

    Nevertheless, these experiments represent an important first step in the transition to the clinic. Other episomal viruses exist (some of which persist for longer periods), several methods can be used to improve viral uptake, and some viruses appear to infect far more efficiently. We have no doubt that this group and others are already trying these alternatives.


    . Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006 Aug 25;126(4):663-76. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. ALS: Predicting Prognosis, Banking on Pluripotent Stem Cells

Paper Citations

  1. . Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006 Aug 25;126(4):663-76. PubMed.
  2. . Neurons derived from reprogrammed fibroblasts functionally integrate into the fetal brain and improve symptoms of rats with Parkinson's disease. Proc Natl Acad Sci U S A. 2008 Apr 15;105(15):5856-61. PubMed.
  3. . Treatment of sickle cell anemia mouse model with iPS cells generated from autologous skin. Science. 2007 Dec 21;318(5858):1920-3. PubMed.
  4. . Generation of germline-competent induced pluripotent stem cells. Nature. 2007 Jul 19;448(7151):313-7. PubMed.

Other Citations

  1. ARF live discussion

Further Reading

Primary Papers

  1. . Induced pluripotent stem cells generated without viral integration. Science. 2008 Nov 7;322(5903):945-9. PubMed.