March 1, 2009

experts comment on research into induced pluripotent stem cells (iPS cells), as published in two papers in Nature

Scientists from Britain and Canada have discovered a way of genetically reprogramming human skin cells so that they demonstrate the properties of embryonic stem cells – including the ability to become almost any type of cell in the body – without the need for using human embryos to obtain stem cells, or potentially harmful viruses to modify the genes.

Professor Austin Smith, Director, Wellcome Trust Centre for Stem Cell Research, University of Cambridge, said:

“These papers are a significant technical development in the field as they present a more reliable and precise method for generating iPS cells. The method allows for greater control of and also the removal of the genetic modifications once reprogramming is complete. Our own research group recently published findings (in Development) showing that this approach produces perfectly reprogrammed mouse cells. These papers show that it can also work in human cells. This represents a new tool to help advance basic research into reprogramming and also paves the way to creation of human iPS cells suitable for biomedical applications.”

Professor Sir Ian Wilmut, Director of the MRC Centre for Regenerative Medicine, said:

“It will still take time before these iPS cells can be given to patients. Crucially, we need to have a method to generate the desired cell types from these stem cells. But I believe the team has made great progress and combining this work with that of other scientists working on stem cell differentiation, there is hope that the promise of regenerative medicine could soon be met.”

Professor Robin Lovell-Badge, Head of Division, MRC National Institute For Medical Research, said:

“These two papers together introduce new methods for the derivation of induced pluripotent (iPS) cells from mouse and human tissue samples. These methods are likely to make these human iPS cells more appropriate than their forerunners as models for human disease in the lab and eventually for cell-based treatments for a wide range of conditions. The methods allow both for more efficient generation of iPS cells, and, critically, provide a means to generate them in a way that allows any foreign DNA to be removed. The latter overcomes a problem that would greatly concern regulatory authorities.

“However, there is still much work to be done. So I would consider these two papers to be exciting steps in the right direction, but we must not forget that the path is still very long and winding and full of bear traps.

“For the time being I think it rather premature to suggest that their work will completely remove the need to derive human ES cells from embryos. We still have a lot to learn from human ES cells about their true identity and their properties, and this knowledge is necessary in order to know whether human iPS cells are truly useful or not.”

Further information from Professor Robin Lovell-Badge: “The two papers, which are linked by common authors and are therefore collaborative, report advances in methods to derive iPS cells. Both mainly deal with mouse cells but show just enough with human cells to suggest that the methods may also be useful in the latter. Both also focus on fibroblasts as their starting cell type, but there is no reason to suppose that the methods would not also work efficiently with other cells. “The Woltjen et al paper is mostly concerned with developing the piggyBac transposon system as a way to deliver the four genes (c-Myc, Klf4, Oct4 and Sox2) that can induce the reprogramming of somatic cells into pluripotent, embryonic stem cell-like cells (iPS cells). “This transposon system is in many respects an ideal way to introduce genes into cells. It can be very efficient, although it requires the expression of a transposase enzyme. The latter can be introduced at the same time in a transient manner (non-integrating), which is a complication, but a relatively minor one. However, they carried out most of the work using an inducible system that required prior genetic modification of the mouse cells. They did this to allow them to more rapidly answer some interesting questions about maximum efficiencies, etc, but this would not be used with human cells. The transposase can also be used a second time, once iPS cells are derived, to remove the piggyBac transposon. When this happens properly, no transposon sequences are left behind at all. This is very important as it means that there will be no disruption of endogenous genes and no foreign DNA – which the regulatory authorities do not like. The excision step is again reasonably efficient, but it does require screening to make sure that all the copies of the transposon are removed. “The Kaji et al paper used a clever trick to make a single gene encoding all four reprogramming factors. This makes us of a small protein sequence “2A” to link together each of the proteins, so they are originally made as one very big protein. However, the 2A linkers essentially self-cleave, which results in four separate (and therefore functional) reprogramming factors. Partly because all four factors will all be present at the same level, this becomes a very efficient way to introduce them and to reprogramme the fibroblasts into iPS cells. However, it also becomes necessary to be able to remove the gene as continued expression of the four factors will compromise subsequent differentiation of the iPS cells. The authors chose to flank the gene with loxP sites, small DNA elements that are the target of an enzyme called “Cre recombinase”. This triggers recombination between two loxP sites, in a way that removes all the DNA sequences that lie in between them. The resulting iPS cells can therefore differentiate efficiently and they showed this by making mouse chimeras (combining the iPS cells with normal early mouse embryos). However, the recombination event does leave a footprint – it leaves one recombined loxP site and any DNA sequences that were outside the original two loxP sites. This is not ideal as these remaining sequences could still disrupt any gene at the integrationn site, and it would count as foreign DNA. “So the solution is to combine the two approaches and to use the piggyBac vector to introduce the combined 4-factor gene. This was done collaboratively by the two groups. It seems to work, however, the data on this was a little minimal. “The evidence that they could obtain pluripotent mouse iPS cells was reasonable, but it was a little strange that neither group showed that the chimeras they made could give rise to functional iPS cell-derived sperm (or eggs), as this is the best assay to prove that the pluripotent cells are entirely normal, i.e. that they lack any chromosome or epigenetic abnormality. “The evidence that they were able to obtain pluripotent human cells was very minimal. It is difficult to prove that human ES cells are normal and pluripotent as the most exacting test is the ability of the cells to contribute to chimeras, something that cannot be done. Extensive surrogate tests, such as differentiation in vitro and in tumours need to be carried out. For iPS cells there are even more doubts. ES cells are obtained without extensive eprogramming from an early embryo, so they have more chance of being normal. Whereas iPS cells have not followed any normal process of embryo development – they depend on reprogramming with no guarantee that it is complete. The tests applied in these two papers fell short of proof that they had obtained truly pluripotent human cells. For example, they could have performed a far more extensive battery of molecular tests. On the other hand, I see no reason why the new methods should not work.