Columbia University Medical Center

Dr. Accili Answers Questions About Diabetes Research on Reddit

Domenico Accili in his laboratory at the Naomi Berrie Diabetes Center at Columbia. (Image by Columbia University Medical Center.)

Earlier this month, Columbia University Medical Center’s Domenico Accili, MD, published exciting diabetes research results in the journal Nature CommunicationsBy switching off a single gene, Dr. Accili’s team successfully converted human gastrointestinal cells into insulin-producing cells, demonstrating, in principle, that a drug could retrain cells inside a person’s GI tract to produce insulin. The finding raises hopes of a treatment for type 1 diabetes patients, who have lost insulin-producing cells. Read the news release.

Dr. Accili recently participated in an AMA (Ask Me Anything) session on the website Reddit. The science AMA series is intended to help to bridge the gap between practicing scientists and the general public. Below are excerpted questions and answers from Dr. Accili’s session.

 

Reddit question (R): Whenever we hear about amazing new discoveries in medicine, it feels like they rarely turn into a product/treatment used by the general public. Could you please give us an idea of the process of going from a finding like yours to actually being a viable and regularly used treatment option?

Dr. Domenico Accili (DA): The process of turning an exciting basic science discovery into a treatment is a long one, even under the best of circumstances. And we have to remember that insulin treatment, while not a cure, is safe and effective. So, the bar to replace it with something else is extraordinarily high, as it should be.

Having said that, my main reason for doing this AMA today is to have the sort of dialogue between clinician/scientists, patients, and their families that I think we need, in order to correct misperceptions arising from overpromising and inflated expectations. I think that we shouldn’t guesstimate when a cure will be available. And I think it’s very important that patients and their families be aware of this. What I try to do, and what I’m doing in this case, is to share with you our work plan, so that you know the timeline of our expected trajectory.

So, how do we go from an exciting discovery to a treatment? Well, the first thing we need to make sure of is that others can replicate the initial finding. If two or more researchers report similar findings, it’s unlikely to be a fluke. Once there is agreement that the science is solid, the next step consists in providing as convincing data as one can on the applicability of the initial discovery (which often times is made in lab animals) to humans. This can take anywhere from a few weeks to many years, depending on the nature of the findings.

In our case, we were very fortunate that we could provide human validation of our findings in less than two years. For us, the next phase is to make a drug that will trigger the conversion of gut cells to insulin cells in a living human being. We are working hard at it, we have an 18–24 months time frame to achieve this goal. If (and it is a big if, I make no bones about it) we are able to do this, then we can start in earnest clinical experimentation, which is likely to take 2–5 years.

 

R: I am a registered nurse who is also a type I diabetic, and I am wondering: In the event that this process works, what options exist to prevent the autoimmune dysfunction from simply re-attacking the engineered cells, with the exception of immunosuppressive drugs?

DA: We are charting new grounds here. No one can reasonably predict how the immune system of a type 1 diabetic patient is going to react to insulin-producing cells in the gut. The autoimmunity of type 1 diabetes is complex. Insulin itself is, according to leading researchers on this topic, an important target for immune attack, as are other components of the insulin-producing cell.

I am optimistic that we can partly circumvent immune-mediate killing of gut insulin cells for two reasons:

  1. The gut has its own immune system, or to put it more precisely, has certain immune “privileges” that derive form its being constantly exposed to foreign substances in the form of food, bacteria, etc. So gut insulin cells may not be exposed to the same “killer” cells that pancreatic cells are.
  2. Gut cells turn over very quickly, every 7–10 days. So even if they are attacked, they may be able to withstand the attack for a week or two. They are very different from insulin-producing cells in the pancreas, which by virtue of their long life and position in the body, are “sitting ducks” for the immune system.

What we are trying to achieve is a treatment that will not entail immune suppression. If that fails, we will consider coupling the new treatment with immune suppression.

 

R: I’m an undergraduate research assistant who studies pharmacology. My question is, can the newly altered gut cells efficiently regulate the supply of insulin in the body? Do these new cells act identical to a functioning pancreas? If not, how differently?

From our animal data, these cells appear to be able to regulate insulin release quite well. I think the reason is that endocrine cells in the gut, of which these cells are a sub-type, are programmed to release hormones in the blood in response to nutrients like glucose, fat, proteins. So the insulin-producing cells in the gut are already primed to function as hormone-secreting cells. We are “hijacking” part of their machinery for the purpose of making insulin.

We tested these cells very extensively, both in our original paper two years ago and in the more recent paper. They look and behave like pancreatic insulin-producing cells. We were initially quite surprised but, for the reasons I stated above, maybe we shouldn’t be. We are only providing partial retraining of their intrinsic capabilities.

We are working on making a pill (or an injectable drug) that will inhibit the Foxo gene in gut cells and thereby turn them into insulin-producing cells. I envision that this drug could be given once a month, or maybe even less frequently, to stimulate the production of new insulin-making cells. At the very least, even if such a drug is unable to function as a “cure,” it may give patients enough insulin to make them “pump-free” or at least not having to worry about insulin at night. The alternative is to grow these cells in large enough numbers and place them in a device where they would help regulate blood sugar, but I think that this is less likely.

 

R: While your research is focused on type 1 diabetes, what relevance does this have for type 2 diabetes? How do you feel about type 1 patients taking type 2 medications as part of a general study/research? Is there a chance of harm? 

DA: If we can retrain gut cells to become insulin-producing cells, the treatment could potentially benefit type 2 patients, as well. Type 2 patients also lack insulin, although not as markedly as type 1 patients, and many type 2 patients take multiple shots of insulin each day. So we certainly hope to generate something that would benefit both groups.

Most medications for type 2 diabetes are unhelpful in type 1, but some might have a role to play. But it doesn’t change the fact that one needs insulin 24/7. If you have a relative with type 1 diabetes, your risk of developing it is higher than in the general population, so much so that we often test siblings of type 1 diabetic patients for their auto-antibodies and glucose tolerance.

 

Dr. Accili is the Russell Berrie Foundation Professor of Diabetes (in Medicine) at Columbia’s Naomi Berrie Diabetes Center. Read the full Reddit AMA session.

Also on CUMC News: