Missing mechanism: The key to beta-cell dysfunction in diabetes still unknown

It is well-established in the literature that type 1 and type 2 diabetes are marked by beta cell failure. However, viewpoints vary regarding published data and how future research will contribute to the critically misunderstood mechanism responsible for both type 1 and type 2 diabetes.

“My view of what’s known is that, for diabetes, there is no question among any researchers that the beta cells play a central — if not the central — role in trying to figure out the disease,” Douglas A. Melton, PhD, of the department of stem cell and regenerative biology at Harvard University and Howard Hughes Medical Institute, said in an interview.

Endocrine Today spoke with experts like Melton to better understand the role of the beta cell in diabetes and the state of current research.

Type 2 diabetes

Though similar concepts of beta cell insufficiency are involved in type 2 diabetes, the pathogenesis seems completely different, Alan J. Garber, MD, PhD, chief medical editor of Endocrine Today, said in an interview.

“The genetic evidence will point to the fact that the beta cell is potentially at risk from day 1,” Steven E. Kahn, MB, ChB, professor of medicine in the division of metabolism, endocrinology and nutrition at the University of Washington and VA Puget Sound Health Care System, told Endocrine Today.

“You could argue on the theory of genetics that it might be suicide,” Kahn said. “On the other hand, what may happen in terms of beta-cell mass loss [for example] is that it could also have environmental components, which could be seen as homicide.”

He said patients may be born with insufficient beta cells either due to a genetic susceptibility or from intra-uterine environmental factors. Alternatively, an underlying defective genetic makeup, along with environmental factors occurring after birth may come into play.

“Therefore, it could be a mixture of both,” Kahn said. “However, one does need genetic susceptibility for beta cells to be defective. That’s why, despite the obesity epidemic, not everybody has diabetes.”

This is a critical point, he said, adding that a patient can be significantly overweight, be insulin resistant, but not necessarily get diabetes because their beta cells are not born defective. Melton said he also believes that it is a combination of both homicide and suicide.

“My guess would be that patients who develop diabetes have what I would call mildly defective beta cells. In the right environment, particularly if we’re thinking about type 2 diabetes, those mildly defective beta cells could be sufficient for the rest of the patient’s life. However, cells put under stress by either obesity or an infection could push them over the edge,” Melton said.

In a commentary earlier this year on data from Saisho and colleagues in Diabetes Care, Kahn wrote that for researchers to develop approaches to slow or prevent the loss of beta cells, a better understanding of genetics and environmental factors is crucial.

“Everybody is trying to link beta-cell mass to beta-cell function and I don’t think it is that simple. There has to be something fundamentally wrong with the beta cell that I don’t think can be explained by a simple reduction in the number of beta cells,” he said. “Mass is an important contributor to the loss of function, but I don’t think the abnormality in secretory function is due to mass alone.”

However, beta-cell mass cannot be measured in humans in vivo, according to the literature.

Clinical implications in type 2

Domenico Accili, MD, attending physician on the medical service at New York’s Presbyterian hospital, professor of medicine at Columbia University, and director of the Columbia University Diabetes and Endocrinology Research Center, told Endocrine Today that understanding and combating beta-cell failure is necessary for treating diabetes. Unfortunately, there are concerns with validating treatments, he said.

“One problem is that the new treatment has to be just as safe and effective as insulin, but more convenient. If one of those conditions is not met, we can’t compromise. I can’t, in good faith as a practicing physician, suggest a treatment that has greater potential risks or even lesser efficacy than insulin itself; that sets a very high bar,” Accili said. “The second problem we have is that all of these claims that we make are to some extent based on wishful thinking because we really have no good way to measure in vivo, in a living human being, what is happening to the beta cell. We can’t image it, we can’t get a biomarker of beta-cell rebirth — if there is such a thing — and we can’t measure a biomarker of beta-cell death or stress.”

When it comes to potential therapies, some preliminary data have suggested that inhibitors of the sodium glucose co-transporter-2 could reverse or halt the death of the beta cell. However, researchers have not confirmed this, and targeted therapies are still an area of concern. “I know that there are glimmers of hope here and there, and I am confident that as new medicine becomes available, as it does almost daily, our ability in the next few years to monitor what’s happening to the beta cell will greatly improve,” Accili said.

He said the field must reassess the short-term goals of developing biomarkers, imaging tools and tools to measure the immune response. These treatments could prove to be useful, he added. Imaging tools are of particular interest to the research community.

“If you stain the pancreas for insulin in individuals with type 2 diabetes, there is clearly insulin in beta cells of practically every single sample,” Kahn said.

They have not lost all of their beta cells, rather their beta cells are dysfunctional, and although the number is reduced, the reduction will vary depending on who is examined and how calculations are made, Kahn said. However, imaging the beta cell in a living human remains a difficult task.

Accili and colleagues are currently trying to find a way to regenerate beta cells to prevent the lifelong commitment to insulin treatment or a way to restore some beta cells, thus removing the need for having constant insulin replacement therapy.

“Something that restores maybe 5% of insulin production or even something that reduces the intensity of the autoimmune response would still be more than we have now,” Accili said. “In fact, we might already have this among the existing treatments. However, that may not be evident, since we cannot measure it in vivo.”

Beta cells make up a small percentage of the pancreas. Therefore, identifying them from the surrounding tissue is a difficult task, according to Accili. However, the ability to image beta cells in vivo would increase researchers’ understanding of the mechanism behind beta-cell formation and loss. Currently, imaging has only been successful in rodent studies.

“Suppose that we could measure changes in beta-cell mass or beta-cell proliferation in real-time, much like how we measure tumor sizes in cancer treatment. That’s the kind of advance we need,” Accili said. “It’s not glamorous, it’s not going to turn heads, or make patients go to sleep better at night or parents of patients feel much better about long-term outcomes, but it’s definitely something we need,” Accili said.

In the literature

Accili and colleagues published a paper in 2012 in the journal Cell, suggesting that the loss of beta-cell identity, rather than beta-cell death, is a key feature of type 2 diabetes.

“The take-home message there is one of hope and one that arises from our research having suggested that the permanent death of beta cells is a far slower process than we had feared and anticipated. There’s actually a stage in the life of a failing beta cell where the cell is still alive; it’s just that it cannot be properly called a beta cell. It has features more like their parent cells, as opposed to other beta cells themselves,” he said.

This observation led Accili and colleagues to hypothesize that if they could reverse this state of the so-called dormant beta cell, there is hope to regain some of the beta-cell function that’s been lost over time. This could be true both in type 1 and 2 diabetes, Accili said.

“Although, the ways to reawakening the beta cell would be different in type 2 diabetes compared with type 1 diabetes, we think we should aim to decrease the insulin resistance that plays such an important role in the pathogenesis of the disease and then add to the treatment agents that we have not yet identified or that we’re working on identifying, or hoping to identify, which would restore the functional state of the beta cell,” Accili said.

Melton said Accili’s findings open up the possibility for reviving beta cells.

“The way to do this in simple form is to stop working on rodents and start working on human stem cells and turn them into beta cells and do all of the drug screening on those,” Melton said.

Similarly, results from a 5-year study recently published in the Journal of Clinical Endocrinology and Metabolism also suggest that beta-cell function decline in patients with impaired fasting glucose/impaired glucose tolerance and type 2 diabetes is significantly worse than previously thought.

Additionally, data from the Insulin Resistance Atherosclerosis Study indicate that homeostasis model assessment of beta-cell function could markedly underestimate this decline. Researchers compared longitudinal changes in the beta-cell function of 1,052 patients according to their baseline glucose tolerance status (IFG/IGT; n=341), normal glucose tolerance (NGT; n=547) and patients newly diagnosed with type 2 diabetes (n=164).

“Despite this spectrum of evidence documenting the importance of beta-cell dysfunction in diabetes etiology, only limited data are available regarding longitudinal changes in beta-cell function, both overall and across glucose tolerance status categories,” Anthony J.G. Hanley, PhD, of the University of Toronto, and colleagues wrote.

They used frequently sampled IV glucose tolerance tests to measure insulin secretion and insulin resistance and compared acute insulin response with the homeostasis model assessment of beta-cell function (HOMA B). Adjustments for underlying insulin resistance were made, according to researchers.

Patients with a glucose status of NGT and IFG/IGT displayed increased insulin secretion by as much as 30%, whereas those with type 2 diabetes showed a decline or insignificant changes to beta-cell function, according to data. However, decline in beta-cell function was significantly underestimated after adjustments for covariates using HOMA B for NGT (31%) vs. acute insulin response for IFG/IGT (50%) and using HOMA B for diabetes (70%) vs. acute insulin response for diabetes (80%).

“It’s been known for many years that in even type 1 diabetes, not all beta cells die. Most of them do, but there’s always a few that survive. One reasonable speculation for why they survive is they might have undergone a loss of beta cell features as a way to escape detection from the immune system. If we found ways to control the immune response and re-differentiate the beta cell, that could be applicable to type 1 diabetes, too,” Accili said.

According to Accili, this is an area in which more progress, in terms of immune modulators that are truly end-organ or end-cell type-specific — in that they do not come with all of the problems that most immune modulators bring about in diabetes — is needed moving forward.

Bridging the gap in type 1 diabetes

The mechanism responsible for type 1 diabetes remains elusive. Despite decades of diabetes research, experts have not clearly determined how genetic susceptibility, environmental factors, the immune system and beta cells actually prompt the complex disease.

“It’s undeniably the case that once diabetes has taken hold, it’s a lack of proper beta-cell function that forces patients to be insulin dependent,” Melton said.

In 1986, Gian Franco Bottazzo, MD, questioned whether type 1 diabetes was caused by a destructive immune response or the weakness of beta cells leading to cell death. Bottazzo coined the idea that type 1 diabetes was introduced by an obscure environmental attack resulting in the release of beta-cell autoantigens. Followed by macrophages, helper T cells were activated, thus producing antibodies (eg, islet cell cytoplasmic autoantibodies) and “active killer cells and cytotoxic T cells,” according to a follow-up paper published in 2011 by Mark A. Atkinson, PhD, of the University of Florida, department of pathology, immunology and laboratory medicine, and colleagues.

Despite this understanding, there has been limited success of immune interventions and agents aimed at the prevention of diabetes, according to Desmond Schatz, MD, professor and associate chairman of pediatrics and medical director of the Diabetes Center of Excellence at the University of Florida, Gainesville.

“We’ve come a long way with understanding the natural history of the disease, but we really haven’t learned a lot about both the causation and the mechanisms leading to the disease in humans. We’ve learned a lot about it in animal models, where we’ve learned about the interplay about genetics, the immune system and the environment, but we really haven’t learned anything precisely about what’s going on within the islet cells and the pancreas itself in human diabetes,” Schatz said.

According to Garber, the precise pathogenesis of type 1 diabetes is unclear, but it obviously involves a major component of immunologic-mediated beta-cell demise.

Emerging evidence suggests the beta cells in a patient with type 1 diabetes may not be normal beta cells, Garber explained. According to Accili, there is a stage in the life of the beta cell where the cell is still alive, but it cannot be properly called a beta cell. Instead, the features of the cells are more like those of the cells that beta cells came from – “like their parent cells as opposed to other beta cells themselves,” he said.

“Most of the evidence suggests that there is something intrinsically wrong with the beta cell of a patient with type 1 diabetes, whether it’s because insulin itself is acting as an antigen or whether it’s because there is something else in the beta cell that helps it become more recognizable by the immune system. Unfortunately, we have few, if any, clues as to what triggers the process in the first place,” Accili said.

Schatz said there are currently no markers for what is known as cell-mediated immunity, which is perhaps the “conductor of the orchestra” and could play a role in the immune process.

“Clearly, we need to find markers of this process in the peripheral blood. We need new biomarkers, we need to be able to visualize the pancreas, and we need a noninvasive way of looking at the process prior to diabetes — some out-of-the-box thinking regarding newer agents, not just immune-mediated,” Schatz said.

Schatz and colleagues recently proposed generating new beta cells in a muted immune environment for type 1 diabetes. They concluded that because type 1 is an autoimmune disease and a beta-cell deficiency disease, it cannot be corrected with immune response therapies alone.

“We have to look at safe therapies and use combination therapies that could mute the inflammatory response; as part of that combination, agents that could preserve or regenerate the beta cells,” Schatz said. “Right now, we’re exploring an agent at the University of Florida that is a combination of antithymocyte globulin, which is a general immunosuppressant, together with an agent that boosts the immune system soon after, called granulocyte colony–stimulating factor, a growth factor that stimulates the bone marrow to make more white blood cells, and I can tell you that the preliminary studies in established diabetes patients are encouraging,” Schatz said.

He anticipates data on the first 12 laboratory-treated patients will be released late this year or early 2014.

At the University of California, San Diego, Ulupi S. Jhala, PhD, associate professor in the department of pediatrics, and colleagues are attempting to investigate the activation of the inflammatory response in the beta cell during the early stages of type 1 and type 2 diabetes.

“What we are finding is that in type 1 diabetes, there appears to be at least two different stages for the cell death process. In the early stages, inflammatory responses dismantle the inbuilt survival mechanisms and compromise the mitochondria while late events ramp up oxidative stress, further weakening mitochondria, and tipping the balance in favor of cell suicide,” Jhala said. “We are working to identify proteins that specifically link the early inflammatory response to the survival response. The idea is to identify the pro-inflammatory nuts and bolts that diminish beta cell defenses and target the early mechanisms before the tipping point is reached.”

Although Jhala said by the time type 1 diabetes is detected, there is a fair amount of beta-cell death that has already occurred in the islet, the beta cells do appear to mount a restorative or regenerative effort. While the cure for type 1 diabetes is still very much in the distant future, she said even small gains at prolonging this beta-cell “honeymoon phase” may prevent the extremes in blood glucose levels, and can go a long way towards managing diabetes.

“The reason this topic is so controversial is because no one has any idea what causes type 1 diabetes. Nobody has any ideas of the mechanisms leading to type 1 diabetes. It’s awfully frustrating for all of us,” Schatz said. “If we can’t prevent the disease, at least we can make the disease easier for patients to manage.”– by Samantha Costa

Should research on the cure for diabetes focus on beta-cell regeneration or glucagon pathogenesis?

Beta-cell regeneration

If our objective is to develop a treatment or an approach that will lead to a cure in type 1 diabetes, research should continue to be focused on the beta cell. There are some recent excellent data that have come out recently showing that the inhibition of glucagon action is very effective in controlling the levels of blood glucose in experimental diabetic animals. A potential application of these observations will contribute to a treatment but not to a cure.

At this point to try to increase the number of beta cells for clinical purposes is not feasible. As stated above, if the evidence showing that the blocking of glucagon action in diabetes is as effective as it has been shown in rodents, then by all means it would be a very fertile field for further research into a clinical application. The question is: can any of those experimental methods in mice be translated into human?

There is also evidence that it appears possible to transform glucagon cells into beta cells in mice but I don’t think there are enough glucagon cells to produce a significant amount of insulin coming from those new beta cells.

We have evidence now that not all beta cells are destroyed in type 1 diabetes and there is also good evidence that maybe the death of the beta cells in type 2 diabetes can be ameliorated to a point, because eventually all beta cells may die in type 2 diabetes. If in type 1 diabetes there are still some beta cells that are viable and producing insulin after many years of diabetes, there is a chance that those cells which are still left intact can be replicated. Again, no clinical work addressing this issue is available but continuous work in rodents indicates that there are substances that may replicate beta cells such as the recently uncovered “betatrophin” from Douglas Melton’s group at Harvard.

There is experimental evidence that other cells in the pancreas in addition to glucagon cells could become insulin-producing cells, such as ductal cells; the potential for a local pancreatic stem cells, not yet identified or isolated that could also give rise to new beta cells. Currently, there are clinical protocols underway attempting to induce the neogenesis of beta cells for known beta cells in the pancreas.

Finally, in terms of beta-cell replacement, many laboratories have the capacity to follow protocols to produce beta cells from undifferentiated human embryonic or induced pluripotent stem cells. This is not to say they don’t have problems for clinical use, but San Diego-based ViaCyte has announced they will begin a clinical trial next year encapsulating the differentiated stem cells to replace those destroyed in type 1 diabetes.

Alberto Hayek, MD, is Scientific Director of the Scripps/Whittier Institute for Diabetes. He can be reached at the Scripps/Whittier Institute for Diabetes, 9894 Genesee Ave., La Jolla, CA 92037; email: ahayek@ucsd.edu. He reports no relevant financial disclosures.

Glucagon pathogenesis

You can’t have high blood glucose if you have no glucagon. Diabetes without glucagon doesn’t exist. Even totally depancreatized patients have glucagon secretion that causes their diabetes. It’s essential for diabetes. There’s no such thing as diabetes in the absence of glucagon.

Beta-cell regeneration is an excellent strategy that could provide a real cure. A major effect of beta-cell regeneration would be glucagon suppression, which beta cells control. That’s the main action of insulin – to suppress glucagon; they work as a team. Injecting insulin under the skin does not reach the levels of insulin that are needed to suppress glucagon, so the type 1 diabetic patient is never quite ahead of the game. He or she is always chasing the glucose, which is the problem of treating with insulin alone. That’s what we’ve been doing for the past 90 years, ever since Banting and Best discovered insulin.

We favor the strategy of lowering the insulin dose, and correcting the increase in glucose by suppressing glucagon with one of the non-insulin glucagon suppressors now available. Forget trying to bring the glucose down with insulin alone. There are now about seven different effective glucagon suppressors on the market, but they are not being used for that purpose. When they are being used for that purpose, they are being used in conjunction with too much insulin.

In our animal experiments, we try to lower the insulin dose between 60% and 80%. Of course, when we do that the glucose is going up because the glucagon increases. However, when we suppress the glucagon with a non-insulin suppressor, that brings the glucose back down and you’ve got a constant glucose level that doesn’t bounce around and cause low blood glucose levels the way it does with insulin alone. We’re very much enamored with that approach.

Roger Unger, MD, holds the Touchstone/West Distinguished Chair in Diabetes Research at the Touchstone Center for Diabetes Research UT Southwestern Medical Center. He can be reached at the Touchstone Diabetes Center at UT Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390; email: roger.unger@utsouthwestern.edu. He reports no relevant financial disclosures.