An Interview with Kevin Eggan on his Work in Reprogramming and Induced Pluripotent Stem Cells

Our first post is an interview with Dr. Kevin Eggan, NYSCF’s Chief Scientific Officer and an Associate Professor in the Department of Stem Cell and Regenerative Biology at Harvard University.

Dr. Eggan had two papers published this week in Cell and Nature Biotechnology magazines and talked with me about them this week.

Can you tell us what your two papers are about?

This week, two important companion papers appeared in Nature Biotechnology and Cell magazines which will provide an important update on the state of the art of the science of reprogramming and induced pluripotent stem (iPS) cells. These two papers take different tacks at answering the question, “How similar are iPS cell lines to human embryonic stem cell lines, and what’s their potential utility for a variety of translational applications as a result of that?”

Can you talk more specifically about your paper in Nature Biotechnology and what motivated this study? 

The first paper, which is actually a paper from my lab but also a collaboration with Chris Henderson and Hynek Wichterle at Columbia University, asks this question in a rather functional way by taking a very large set of new induced pluripotent stem cell lines that were derived in my lab at Harvard in collaboration with those Columbia investigators and performing so-called “directed differentiation experiments” with those stem cells to produce spinal motor neurons. We chose spinal motor neurons as an example type of differentiated cell to make from these stem cells because of my lab’s long-standing interest in studying neuromuscular diseases like Lou Gehrig’s disease. So what we wanted to know in short was if we made a very large panel of induced pluripotent stem cell lines from people that were young, old, healthy, sick, from men, from women and we tried to use them to make this particular kind of neuron which is destroyed in spinal cord injury and Lou Gehrig’s disease, would we be able to do it? And also importantly, how would their performance compare to the performance of a panel of human embryonic stem cell lines that we had been working with?

What did you find about the similarities and differences between iPS and ES cell lines?

I think reassuringly, what we found was that in fact on average, human embryonic stem cells and human iPS cell lines behaved rather similarly, and that they both are able to make spinal motor neurons which are fully functional by a variety of criteria. So I think in that way, it says that iPS cell lines are going to be useful for many translational as well as therapeutic applications.

Now in the process of these studies, we did find that each cell line had its own slightly unique behavior, so although on average ES cell lines and iPS cell lines behaved the same, there were some cell lines that were better than others. That really motivates a need to find methods that can tell you which ones are the best for the particular application you’re interested in. And that’s really where the second paper comes in. 

Can you tell us a little about your Cell paper and what the purpose of the study was?

That paper again is a study that’s led by Alex Meissner and his laboratory at Harvard University, but was also a collaboration with my lab at Harvard, where we took this very large panel of iPS cell lines and compared it to a large number of human embryonic stem cell lines now genomically, by looking at the fundamental expression of all genes in the genome and also by looking at something called DNA methylation, which is predictive of which genes are on and off broadly in those cells and their ability to regulate genes later. And the goal here was to say at that level, at the level of the genome, was there really anything that was fundamentally different about iPS cell lines from ES cell lines? Again, a similar question, but more importantly, could we mine information out of that data that could tell us which ones were the best ones to use?

What did you find?

Reassuringly that study, which really takes as I said a completely different approach at the problem, one which uses new technologies and DNA sequencing and genome-wide measures of the behavior of cells, it came to largely the same conclusion that our study with motor neurons did. Namely, that although there are some slight differences between human iPS cell lines and human ES cell lines, on average they’re pretty similar and in fact there are a large number of iPS cell lines which are within all the norms of what you would expect for human embryonic stem cell lines.

And so what we think this means is that one really needs to focus the attention more on why certain cell lines behave better than others in a variety of contexts. And, excitingly, the same paper in Cell comes to a sort of conclusion in that regard. Using this data, we were able to generate what we call a scorecard for the behavior of these stem cell lines. And using that, we were able to really measure what the propensities of each cell line are. Just to say, what sort of cells do each one of these kinds of stem cell lines like to become? Do they like to become cells of the intestine, cells of the brain, cells of the skin? And maybe not surprisingly in retrospect, but surprisingly in the moment, we found that those propensities, those natural likes and dislikes for each cell line, mapped very closely to how well each cell line could be directed to differentiate into a cell type. That is to say, when we got bossier and ordered cells to become motor neurons, those that had the most natural inclination to do that followed those orders the best.

Tell us more about the scorecard and how do you think it will be utilized by scientists.

In short, what the scorecard really allows us to do is to predict prospectively what cells are going to be useful for a given application. And we discovered in the context of that, that there really are certain cell lines that are better, for instance, for studies of the nervous system and others which one should probably choose for the study of diabetes or the degeneration of the pancreas. And so I think this is really going to open up a new phase of understanding the quantitative behaviors of different stem cell lines and should move that important field forward.

Most notably also, the scorecard approach is going to be very useful for the systematic characterization of cell lines as people move towards the clinic. As people start making populations of differentiated cells, it’s important that the preparations they make are the same over and over and over again, giving us confidence that if we transplant them into a person, we’re always transplanting the same kind of cells. And really the scorecard is also a building block for that sort of technology.

Links to the two papers (you will need an account to view the full text)

http://www.nature.com/nbt/journal/vaop/ncurrent/full/nbt.1783.html

http://www.cell.com/abstract/S0092-8674(10)01524-2

Stay tuned in for more posts next week!