Doug Melton is the Xander University Professor at Harvard and an Investigator of the Howard Hughes Medical Institute . He is also a co-director of Harvard's Stem Cell Institute and Co-Chair of the Department of Stem Cell and Regenerative Biology at Harvard. His research focused on basic developmental processes using Xenopus laveis until the diagnosis of his infant son with type 1 diabetes about 20 years ago shifted his focus toward understanding and curing the disease. Doug recently spoke with StemBook editor, Lisa Girard. Below is an edited version of that interview.
Could you tell me a little about the projects that are going on in your lab right now?
All the projects in the lab now are focused on trying to use stem cells, or other progenitors, to find new treatments for diabetes. Our main focus is type 1 diabetes, but some of the work we do is also relevant to type 2. And so, in brief, what we’re focused on is how to make more beta cells. We’re excited about a new hormone we reported on last year called betatrophin which seems, at least in mice, to have the ability to robustly and dramatically increase the number of beta cells. This would be really important if it worked in humans for type 2 diabetics that are slowly losing beta cells as the disease progresses. For type 1 diabetics, it is not clear exactly how it might be used. First, as I said, we have to find out if it works in humans. If it does, then you might use the approach of making more beta cells right at early onset of the type 1 disease to try and stop or reverse the immune attack, making the immune system think its made a mistake, inducing something called anergy by making more beta cells. But people who have had type 1 diabetes for a long time are unlikely to have enough residual beta cells to let this hormone make more of them. Its most likely use, if it works in humans, would be in type 2 diabetes, and possibly new onset type 1.
The other main project in the lab now is turning stem cells into beta cells. This is hardly a new idea. People have been hearing about this for more than a decade. You might reasonably ask, “what’s taking so long?” I feel we have made a lot of progress but we are not at the goal line yet. A football analogy would be that we are “in the red zone”. We are getting very close to making fully functional beta cells ex vivo. We now have cells that are responding to glucose and secreting insulin, and that’s something we and others previously had not been able to do. We are excited and want to confirm it to make sure it is reproducible and scalable. I hope in the next year we’ll be able to develop and report on a method that allows the creation of billions of human beta cells that can then be used in two contexts; one, for transplantation into diabetics, and the other; using iPS cells from patients, for drug screening..
How do you see stem cell research being differentially applied to type 1 and type 2 diabetes?
For type 2, we think about have beta cells and we need to improve their function. Our own work doesn’t really deal directly with insulin resistance in the peripheral tissues, but that’s obviously an important issue. Type 1 is sort of a different problem. We start again with the beta cell. I just described making more, but I neglected to mention that even if we made more and put them back into a person they’d be rejected by the person’s immune system. The lab is just beginning to work with a number of bioengineering collaborators on encapsulation devices. We and others amusingly call this the “teabag” problem. How do we put these cells into a teabag that would allow glucose to come in and insulin to go out and protect the cells from immune attack? Lots of very good people have worked on this problem and the principal issue seems to be that fibroblasts glom up any encapsulation device you put in and one has to figure out how to stop the fibroblasts from essentially covering up the teabag and preventing the exchange of glucose and insulin.
Is that some kind of wound response?
No one really knows; it probably is that the body recognizes it as not being “self”. Fibroblasts may be called to the teabag by the cells inside. It’s not generally believed that when you have a prosthetic limb, such as a titanium rod, that it gets glommed up with fibroblasts, but I’m learning more about that. An alternate approach, one which we are just beginning to explore with other colleagues here at the Stem Cell Institute, is how one might genetically engineer stem cell-derived beta cells so that the immune system didn’t attack them. We can get around the problem of them not being self by using iPS cells, but in autoimmunity the system attacks self so we have to figure out how to have, particularly the T cells, leave them alone. There are a number of ideas about this. Some of them come from studies on cancer cells, which can blunt the immune system.
With what we’ve recently come to understand in terms of the autoimmune response in type 1 diabetes, to what degree do you think this knowledge will be more generally applicable to other autoimmune diseases such as Crohn’s and lupus?
No one has understood the initiating cause or causes of autoimmunity. So a project that we have been wanting to do for some time is to reconstruct autoimmunity in an immunocompromised mouse by putting in the target cells (beta cells), the effector cells (the immune system) and the educating cells (thymic epithelia). We haven’t achieved that and I’d like to see that project move forward. But to your question, if we could watch the disease develop, purely in a descriptive way, then that’s likely to be very informative about all autoimmune disease. To my knowledge, no one has watched an autoimmune disease develop and, as you can appreciate, in patients, by the time they appear with a phenotype, the initiating events are long gone and there has been epitope spreading. So, it has been really difficult to figure out, in an Aristotelian sense, the primary cause. That’s the only good idea I have about this. I don’t have any good ideas about autoimmunity other than trying to reconstruct it and watch it happen. Another way to say this is that one would like to figure out whether it is the beta cells screwing up and attracting the immune system or does the immune system screw up and attack perfectly normal beta cells? There are strong opinions on this, but no one has done a telling experiment, in my opinion.
A nice thought experiment would be to reconstruct the disease with one patient’s cells, using iPS cells. Now you make one hundred mice with these human iPS derivatives and ask: do all the mice get diabetes? Would all 100 animals get it at the same time? In the same way? It is probably instructive to reflect on the fact that cancers used to be thought of as one disease, but the more we learn these cancers, say blood cancers, are found to be different types. We can hope, and as medical practitioners which I am not, or as bioscientists, we can hope that there are some common paths to the disease. If everybody gets it a different way then that kind of personalized medicine is so far away that one can’t think about it seriously. I am hopeful that there will be some common defect, like that the beta cells are particularly sensitive to stress, or there is an education of the T cell that goes mistaken.
In terms of which approaches people are taking which might bear fruit clinically most quickly; could you compare the transplantation encapsulation strategies with a stem cell based immune response-centered approach?
I think the best solution is going to be blunting the immune attack. I don’t think we have any evidence that if you took a patient that had type 1 diabetes and you blunted the immune response that they could regrow enough beta cells. So they still need beta cells provided by researchers. To answer your question more directly, my view is that step one is to make the beta cells and we are nearly there, as I said. Step two is to figure out how to have them survive and function long enough that you could use them for transplantation. When I say long enough I don’t think that means a decade, I think that means maybe a year. Then the longer-term solution, which we really should not give up on, is to try and get rid of the autoimmune attack. This is likely a very simple or idiosyncratic view of it, but that’s how I think about it. I’m sure if you talked to other researchers they would have more sophisticated and varied options. I think of it as job one; make the beta cells. What do they say for Dunkin Donuts? Time to make the donuts? That’s right, it’s time to make beta cells.
Thanks a lot, Doug, I really appreciate your time.