Anne L. Peters, MD; Guy A. Rutter, PhD


June 23, 2014

This feature requires the newest version of Flash. You can download it here.

Beta-Cell Harmony

Anne L. Peters, MD: Hi. I am Dr. Anne Peters, reporting from the American Diabetes Association (ADA) meetings in San Francisco. Today I am talking with Dr. Guy Rutter, Professor of Cell Biology at Imperial College in London, about his current research, which concerns the molecular basis of type 2 diabetes. We will discuss what he finds interesting, and how that applies clinically as we treat our patients with diabetes.

Guy A. Rutter, PhD: I'm here to talk about 2 areas of my own research. Chiefly, what I will be presenting at the ADA conference is our recent work on a class of drugs which is becoming increasingly more important -- the incretins -- and the work that we have been doing to understand how these act on the pancreatic beta cell to stimulate the secretion of insulin.[1]

Specifically, I will be showing that these drugs improve the way in which the beta cells work together as an ensemble to prompt the release of hormones. This is something that wasn't suspected a few years ago -- that each individual cell has its own agenda, but when they work together, they perform far better. It turns out that this is one of the ways in which the incretins work, so we can think about designing other drugs that do the same thing, but in a different way.

Dr. Peters: How are you studying the beta cells? What is your model?

Dr. Rutter: We are very fortunate to have, as part of a large consortium, access to human materials. This is all being done with human pancreatic islets -- these little balls of about 1000 cells. We use quite sophisticated imaging approaches. We put them under a microscope, and charge them up with molecules that allow us to see the activity of each cell. We can record their activity. We add the drugs or hormones, and we can see how they perform.

We found, to our surprise, that many of the incretin hormones (this family of drugs) actually increase the synchronicity of the cells within the islet, and this is associated with an increased production of insulin. We found that if we block that synchronicity using either drugs or molecular techniques, we block the release of the hormone. Perhaps our most interesting finding was that when we compared the behavior of these individual islets and their response to the hormone with the properties of the donor who had provided the islets, we found that the cells from people who were most obese, and therefore probably at highest risk of developing diabetes, had the poorest performance in terms of synchronicity.

We think that this is a new feature that is probably prediabetes, on the road to beta-cell failure, and wasn't suspected before. If we can find new ways of making beta cells speak to each another, we could have a way of correcting the deficiency of insulin secretion, which is underlying the disease pathology.

Dr. Peters: This is fascinating, because it is like creating harmony among your beta cells. Does anything other than the incretins do this? Does metformin do it? How do weight loss and exercise fit with the paradigm we know for treating diabetes?

Dr. Rutter: It would be a little difficult to test that in an individual, because sadly we can't get tissues from people. We are only able to take that snapshot when a person passes away. If we could correlate this not prospectively, but in a snapshot, we know that the people who were leaner when they died and donated their tissues displayed a greater connectedness. The inference would be that if you go from having a high body mass index (BMI) to having a low BMI, that we would see this connectedness reemerging, but it's very difficult to prove. We would need to use animal models.

Dr. Peters: Okay, I'll allow you an animal or two -- I'm a people researcher. The other part of what you do, to some degree, is even broader. Can you discuss that?

Diabetes Research: From Man to Mouse, and Back to Man

Dr. Rutter: With my other hat on, I am part of a large worldwide consortium that is trying to identify genes that increase the risk for diabetes. At the end of the last century when the human genome project was completed, it became possible to look at all of the genes in the human genome. Along with other technological developments, it became possible to compare people with diabetes with people who don't have diabetes in very large populations and ask whether there are variants that are more common in the diabetic population than in the control population.

Any variant you find that is enriched in the diabetic population, you can be fairly confident is having a causal effect to increase diabetes risk. That work by my clinical and epidemiology colleagues has now thrown up about 70 genomic loci with a huge number (about 500 genes) that, although we didn't suspect it before, we now think are involved in diabetes risk.

My part of the work is to take that information and knock those genes out in mice, and see what the effect is. Then, we can take those that have an effect in the mouse to the human tissue protocol, and see whether they have the same effect in human tissue. If they do, we can design small molecules. We interact with pharma companies to do that. We then go back and close the loop into humans by seeing whether these molecules will improve glycemia and the outcomes in diabetes. It's a loop, starting with a human variant, through animal models, and then in situ human tissue back to man.

Dr. Peters: Have you actually concluded the loop? Every time patients hear that we can cure diabetes in a mouse, they think we can cure them. Realistically, what is the time frame that we are talking about from human to mouse and back to humans again?

Dr. Rutter: It's not short. We are not talking 6 months, or even 6 years. We are probably talking a decade or two.

To answer your first question, if we have not completed the loop, we have gotten to about 9 o'clock. One example is that these unbiased screens have actually identified a gene that encodes a zinc transporter. Zinc is fundamental to the storage of insulin. We found that variants of that transporter lower the ability of the pancreatic beta cell to store zinc and, thus, to store insulin. When we knock this out in the mouse, we find that we recapitulate that phenomenon. We can then do screens to identify small molecules that will stimulate this protein and encourage zinc to be taken up into the cell, and therefore encourage the storage of insulin.

When we give that molecule to a mouse, we can improve glycemia in the mouse, and therefore we predict that we will be able to improve glycemia in humans. The reason we haven't gotten all the way through to 11 o'clock or back around to midnight is because of the pharmacokinetic properties of that molecule. It would have to be engineered by chemists to find something that can be injected safely.

Dr. Peters: Then there is the whole issue of human trials and all the steps required. This work is amazing when you think about it, but the process of getting it back into actual clinical use is daunting.

Dr. Rutter: Sadly, it takes 10 years and $20 billion for each drug.

Dr. Peters: It's not a cheap process. In terms of the ADA meetings, is there anything that you think is interesting, or that you have learned?

Dr. Rutter: I went to a talk that I enjoyed very much and that was somewhat aligned with our work.[2] The candidates have not been obtained through screens in humans, but have been identified through basic mouse work. This involves inhibiting a molecule that controls the conversion of pancreatic alpha cells to beta cells. It turns out that if you activate this molecule, you achieve a huge repopulation of the pancreatic islet with pancreatic beta cells.

Dr. Peters: Wait, no one has ever told me that alpha cells become beta cells.

Dr. Rutter: That is because they didn't know that they did. Until very recently, no one thought that this was the case. This was artificially invoked by interfering with the transcription factors that determine cell fate. By changing just 1 transcription factor, you can take a cell that should be expressing the glucagon gene to make glucagon and flip it into a cell that will make insulin. At least, you can do it in the mouse. Even in models of type 1 diabetes, this small molecule will repopulate the pancreas with beta cells. It's possible to show, using something called linear tracing, that these beta cells actually started life as alpha cells. This is a very interesting new approach.

Dr. Peters: If you did this in type 1 diabetes, wouldn't the new beta cells still die, because of the autoimmune disease, or does it happen very slowly? Could this change the balance? This is fascinating.

Dr. Rutter: You would almost certainly have to use immunosuppression to prevent that from happening. But if you could accelerate the process such that the rate of cell death was less than the rate of cell birth, and get that balance right, you might need quite minimal immunosuppression.

Dr. Peters: Thank you for sharing what you have been doing, and also making animal research seem interesting. This has been Dr. Anne Peters for Medscape. Thank you for watching.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.