Reprogramming adult cells into pluripotent stem cells – what do these new results mean

Blogging on Peer-Reviewed ResearchYou guys might have noticed I’ve been quiet lately, that’s because I’ve scheduled a thesis defense and am under deadlines. However, I couldn’t let these two (1) papers(2) on reprogramming of human adult cells into stem cells slip by without some comment (NYT piece here)

These reports are a follow-up on landmark animal studies that we discussed previously that showed that expressing 4 genes in cells obtained from adult animals you could induce them to form embryonic stem cell (ESC) like cells that researchers dubbed induced pluripotent stem cells (iPS cells). At the time we noted several obstacles to the practicability of this technology, and these papers represent success in overcoming the first – transferring the technique into human cells.

I admit I’m surprised they were able to do so so quickly. But this turned out to be a nice example of the discovery of common exploitable pathways between humans and other animals.

Below the fold I’ll go over the differences between these studies and the previous animal studies, the evidence of the pluripotent nature of these cells, unresolved problems with this technology and why this isn’t a victory for the anti-ES cell crusaders.

Both papers performed similar studies to demonstrate the pluripotent nature of the iPS cells, however, in humans, a slightly different set of genes could be used to transform the adult cells into pluripotent cells. Animal studies used Oct4, Sox2, c-Myc, and Klf4, and these were the same genes used by the Yamanaka group (2). However the Thomson group(1) showed that the transformation could also be induced by Oct4, Sox2, Nanog, and Lin28. So why did one group stick with the same set of genes and the other try different ones, and why did it still work? Well the Thomson group started by identifying genes that are expressed in ESC but not in slightly less pluripotent myeloid (blood) precursors. They then screened these genes by using retroviruses to insert them into a cell line that has the Oct4 promoter driving a survival gene. The result was a screening tool that when pluripotency is achieved, the Oct4 promoter activates thus allowing the cells to survive treatment with a toxin. This was, I think, a smart move, as it essentially repeated the screening process used in mice, and the Thomson group didn’t assume that identical factors in mice could accomplish this goal. They were wrong, but as a result they still managed to generate iPS cells and probably have discovered some additional interesting biology. Namely that the inclusion of Nanog and Lin28, while not required for generating pluripotent lines, greatly increased the efficiency of the cloning process.

The other neat difference between the two studies is that they both used different cell lines to generate iPS lines. The Thomson group tested fetal fibroblasts and neonatal foreskin fibroblasts successfully, while the Yamanaka group used human skin cells or dermal fibroblasts to achieve the transition. The Thomson group also had a more elegant system for identifying iPS clones (humans don’t have Oct4 promoters driving survival genes) using a tissue culture trick, they simply used conditions conducive to ESC growth but inhibitory to fibroblast growth, as a result ESC colonies popped up more quickly and efficiently, while the Yamanaka group had to sort through more noise – lots of non iPS cells grew in their cultures.

One should note that the likely reason the Yamanaka group achieved higher rates of iPS induction is that they also inserted the mouse-retroviral receptor gene into the cells, thus increasing their infection efficiency. It’s likely the two techniques are otherwise equivalent.

Both groups then tested the function of these cells through various techniques like embryoid body formation, directed differentiation, etc., and the gold-standard (at least the ethical gold standard for human cells) injection into nude mice to generate teratomas – a tumor comprised of all three germ cell layers. Here is figure 7 from the Yamanaka group (2).


The cells, in vivo were capable of generating cells from the three primary germ cell types – endoderm, mesoderm, and ectoderm. This is as about as good as it gets for demonstration of their pluripotent potential. The only better measure – chimera formation – is simply unethical for human cells, and teratomas are adequately convincing, especially combined with their in vitro embryoid body analysis.

So what are the remaining problems? The most critical is, of course, the use of retroviruses to transduce the cells. The problem is that previous attempts to use these vectors in humans, to treat severe-combined immunodeficiency (bubble boys), resulted in an unacceptable rate of oncogenic transformation of the treated cells causing the FDA to stop the trials. The kids got leukemias. This is terrible. The reason is that the retroviruses incorporate their genomes into ours. The advantage of this is that the genes are permanently expressed – long enough to transform the cells and maintain them until they are differentiated. The draw back is that the retroviruses prefer to insert themselves randomly in the genome in areas that are transcriptionally active. The result is that the transgenic promoters – powerful viral sequences that drive gene expression – may get stuck next to an oncogene, that can then cause cancer. Or cause epigenetic changes in the region of an oncogene that should be silenced. Or even interrupt a tumor-suppressor gene. All very bad outcomes that have been shown to cause cancer in humans using these vectors.

There is a solution though, and I would humbly suggest that the researchers in this field either use a conditional promoter – one that only expresses in the presence of a drug like tetracycline or tamoxifen. Or that they use a non-viral integrase system. Now, these integrase systems have also been shown to cause chromosomal aberrations. However, mutagenesis of the proteins has been shown previously to increase their specificity of action, and the technology, I believe, can be improved to the point that it can be used in humans. Further, they can be engineered for the integration of the genes into one location, rather than randomly throughout the genome, which would be a problem even if one used conditional promoters.

It’s also possible that tagging the 4 proteins required for the transformation with an HIV-protein that allows them to cross the cell membrane (tat-tagging), may provide a viral-free approach for generating and maintaining the iPS cells, without any gene manipulation at all.

Finally, as far as the anti-ESC types are concerned, they should not consider this a victory for their anti-science agenda or the policies of George Bush. First, these discoveries would not have been possible without ESC research. I also believe these cells would have been discovered in the same amount of time, with or without the political interference in science. Not only because somatic cell reprogramming was hotly studied long before this became a political issue but also because they represent an ideal stem cell – one that can be genetically matched to the donor, yet is still pluripotent. SCNT has been an incredibly difficult technology to make practical, as human eggs are difficult to obtain, and the process is very inefficient. We’ve also lost valuable years of study of pluripotent human stem cells in this idiotic debate, that would directly translate to our understanding of how to apply these cells in studies of disease and for clinical practice.

Further, the precedent that has been set – that a minority of people can inflict their religious beliefs onto science policy – is atrocious, and damaging, and will likely have long-term consequences for science. Because, now they’ve learned that not only can they get away with it, but they can delude themselves into believing it is the right thing to do for scientific progress.

1. Induced Pluripotent Stem Cell Lines Derived from Human Somatic Cells. Junying Yu, Maxim A. Vodyanik, Kim Smuga-Otto, Jessica Antosiewicz-Bourget, Jennifer L Frane, Shulan Tian, Jeff Nie, Gudrun A. Jonsdottir, Victor Ruotti, Ron Stewart, Igor I. Slukvin, and James A. Thomson (20 November 2007) Science [DOI: 10.1126/science.1151526]
2. 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