Update in 2020: It seems now after all these years that the consensus in the now more mature field is that IPSCs and ESCs are nearly identical in most cases, and both have some of the same translational challenges such as teratoma-forming activity. It’s interesting to read this post from nearly 10 years ago and the concerns that were so intense back then.
An important question in the young iPS cell field has been– just how similar these reprogrammed iPS cells are to embryonic stem cells (ESC)?
The assumption has been that the answer is very similar!
However, there have not been a whole lot of direct comparisons until recently and these new studies have supported the notion that iPSCs are in fact not that similar to ESC.
First, there was a paper in CSC from a team lead by Nissim Benvenisty showing that iPSCs and ESC were substantially different as disease models.
Now we have two papers coming out ( one in Nature and one in Nature Biotechnology) that support the same idea that iPSCs are really distinct from ESC.
A Nature paper from George Daley’s team showed that iPSCs have memories of the cell types from which they were derived. This was an issue we blogged on previously here and also here where we discussed Daley’s presentation on this data at the ISSCR meeting. These findings suggest iPSCs themselves are very heterogeneous.
The Nature Biotechnology paper from Konrad Hochedlinger’s team showed much the same thing. Hochedlinger was quoted in The Scientist as saying that iPS cells become more like ESC the more you culture them because they “lose their memory”. This is actually quite worrisome because it suggests that iPSCs become more abnormal over time.
In the same piece in The Scientist, Mahendra Rao is quoted as saying that iPSCs are not very similar to ESC when examined at “high resolution”. Amy Adams at the CIRM Research Blog blogged on this today as well.
There is also now a very important editorial in press at Stem Cells from Pablo Mendez’s group saying that when it comes to teratoma that iPS cells are far more tumorigenic than ESC as well. As readers of this blog will know, this is something we have been discussing for a long time and we had a hunch iPS cells would be more tumorigenic.
So if iPSCs are not that similar to ESC, what does this mean?
First, it is not the end of the world. Many of us had assumed that iPSCs are similar to ESC in some ways, but distinct in others. These studies simply validate that notion.
Second, iPS cells are still very useful even if they are not quite so close siblings of ESC. In fact, iPS cells do not need to be extremely similar to ESC to be useful. The pluripotency and self-renewal of iPS cells makes them a powerful tool regardless.
Third, iPS cells are a distinct type of stem cell from ESC and should be considered as such.
Fourth, these studies also highlight that every iPS cell line made is distinct from the others made. The heterogeneity amongst iPS cell lines is probably more substantial than the differences between ESC lines, but ESC lines also vary between each other quite a lot as well.
Finally, the tumorigenicity of iPS cells is a serious concern. A remaining open question is how prone iPS cells are to form malignant tumors (although don’t forget that teratoma can be malignant!). We predict that iPS cells are dramatically more prone to produce malignant tumors compared to ESC.