Feature: blind beta cells

Glucose sensing and the origins of diabetes

Adult-onset (type 2) diabetes is common and, thanks to its links to obesity, becoming more so. Although many aspects of type 2 diabetes have been clarified, the nature of the principal defect remains controversial: should we blame the beta cell for not producing insulin properly, or the body's cells for not responding to it? Latest research suggests that the beta cell may be the main culprit – and is suggesting new ways to treat the disease.

Diabetes – certainly the adult-onset (type 2) form – tends to be seen as less serious than cancer and heart disease, especially since, unlike juvenile-onset (type 1) diabetes, it can be controlled with exercise, diet and drugs, and does not always require insulin injections.

Yet as Professor Guy Rutter at the University of Bristol's Department of Biochemistry points out, in recent studies (1) half of the people who had recently suffered a heart attack turned out to have either diabetes or 'metabolic syndrome', a strong predictor of eventual diabetes. Coronary heart disease and the other complications of diabetes – kidney failure, blindness, amputations and nerve damage, to name but a few – shorten the life expectancy of people with type 2 diabetes by five to ten years and those with type 1 diabetes by 15 years.

Then there are the healthcare costs: the disease eats up around 10 per cent of the UK NHS budget every year. And since type 2 diabetes accounts for about 90 per cent of cases (type 1 is much rarer), that’s a pretty high spend for an 'unimportant' disease.

The pathology of diabetes is well known: the body’s normally tight control over glucose usage breaks down. An old but important question, which is still hotly debated, is where the primary physiological defect lies. Does diabetes occur because body cells are not responding properly to insulin, which normally stimulates them to take up glucose from the bloodstream (so-called insulin resistance)? Or are insulin-producing beta cells in the pancreas responsible, by failing to secrete enough insulin in the first place?

Professor Rutter believes that the pendulum is swinging towards beta cells. "In the last five to ten years it's been recognised that you can be as insulin-resistant as you like and you’re not necessarily diabetic – and that what you really need to have to become diabetic is a defect in insulin secretion from the beta cells. So now there's a large mobilisation of forces – NIH and others – to support beta-cell research."

Beta-cell biology

Current research is taking two main directions, he says. "One model is looking at whether, in type 2 diabetes, the beta cells become 'blind' to glucose, in that they fail to respond to its presence in the blood and release insulin. The other is investigating to what extent beta cells are actually destroyed, as happens with type 1 diabetes, which is an autoimmune disease."

While acknowledging that both aspects are important, Professor Rutter believes that abnormal glucose sensing by existing beta cells may play the predominant role. Secreting the right amount of insulin is a big challenge for beta cells. In a healthy person, blood glucose levels increase by about 60 per cent after a meal. Yet that change in blood sugar produces a massive (40-fold) increase in the rate at which insulin is released from the beta cells.

That, says Professor Rutter, is crucial. "If that doesn't happen, the person progressively becomes diabetic."

Moreover, insulin is not released in a single bolus. There are complex phases or pulses of secretion. "It's an oscillatory process. If you lose that oscillatory pattern, that’s one of the first signs of becoming diabetic."

Having concluded that insulin release is the root of the problem, Professor Rutter and team are looking at the biochemical machinery involved: the key sensing molecules that monitor changes in blood glucose levels and the signalling pathways that link sensing to insulin release. To do so, they are drawing upon a range of techniques.

One approach is to take mutations, identified by geneticists, that could cause or contribute to type 2 diabetes and to see what effect (if any) they have on beta-cell function. One possible culprit is the insulin receptor on the beta cell – ironically, the beta cell may itself become resistant to insulin. This would lead to a vicious circle in which glucose uptake by the beta cell would drop, inhibiting insulin release. Professor Rutter's team is currently testing this possibility by knocking out the insulin receptor in cultured beta cells.

The team is also looking at the other principal hormone of the pancreatic islet, glucagon, made in the alpha cells. Glucagon counteracts insulin, stimulating the release of glucose into the bloodstream during starvation or sleep. In a healthy person, glucagon and insulin act antagonistically, although the precise mechanisms are not clear.

"It's an important question clinically," says Professor Rutter. In type 1 diabetes the alpha cells eventually become blind to glucose and stop releasing glucagon to counteract the effect of excess insulin in the bloodstream. So if a person with type 1 diabetes has injected too much insulin, hypoglycaemia (shortage of blood glucose) can result. This is potentially life-threatening: it is the cause of death in 4 per cent of people with type 1 diabetes.

One school of thought is that alpha cells sense the insulin coming out of the beta cell and respond reciprocally by suppressing glucagon release. However, Professor Rutter's group has shown (2) that glucose also works directly on the alpha cell. "It does its own thing irrespective of beta-cell products. Glucose and insulin both independently suppress glucagon secretion."

Central to efforts to identify precisely how the beta cell works has been the development of novel imaging technologies. "We want to develop non-invasive ways of looking inside the pancreas of a living person and being able to tell how many beta cells are present, how well they're working, and look in detail at what's actually going wrong," says Professor Rutter.

The resolution of magnetic resonance imaging is presently too low for the task – a problem heightened by the fact that the beta islets constitute just 1 per cent of pancreatic volume yet are dispersed throughout that volume.

Instead, the team has turned to an enzyme called luciferase, which fireflies use to generate light. This can be used to detect adenosine triphosphate (ATP) levels in beta cells, providing a measure of glucose metabolism in the cell – and hence an indicator of whether insulin secretion will be triggered. Although the technique has been used before to track tumours, this is the first time it has been used – literally – to highlight beta cells in the pancreas.

The team is also looking closely at single cells: in particular, at how insulin is transported and secreted. Insulin is carried in vesicles, membrane-bound structures, of which there are thousands inside the beta cell. By labelling the vesicles with a green fluorescent protein cloned from jellyfish, it is possible to watch their movements within a single beta cell, one hundredth of a millimetre across, in real time.

The vesicles move jerkily along microtubules, a meshwork that forms part of the cytoskeleton. "Vesicles are like trains running along tracks. They're not automobiles, they’re not free in space," observes Professor Rutter. Once the beta cell has been challenged with glucose, the vesicles make concerted efforts to move along the microtubules towards the membrane and freedom. Here, the vesicle transiently fuses with the cell membrane and insulin is released from the cell. "The 'nanomechanics' of this fusion event is one of the things that goes wrong in type 2 diabetes," explains Professor Rutter.

His team is dissecting the protein machinery by which vesicles move along their 'tracks' and recently found (3) that a protein, kinesin, which 'walks' the vesicle along the microtubule, plays an important role in moving the vesicles towards the cell membrane. The final fusion step, however, seems to require a different ‘molecular motor’ – a myosin (MyoVa) moving along an actin network.(4)

Drug targets

Once the key molecular players in insulin secretion have been identified, their potential as drug targets can be explored. Professor Rutter is currently collaborating with pharmaceutical companies GlaxoSmithKline and AstraZeneca to develop drugs to act on two likely targets: adenosine monophosphate-activated protein kinase (AMPK) and a protein involved in regulating lipid metabolism, SREBP.

The enzyme AMPK is part of the body's energy balance system, controlling glucose use and ATP generation in body cells. When activated, it enhances the effects of insulin by stimulating glucose uptake and metabolism.

A widely used drug, metformin, works by activating AMPK in muscle and liver. Unfortunately, when activated in the beta cell, AMPK suppresses insulin secretion. "So what you give with one hand you take away with the other," says Professor Rutter. "Metformin is still OK as a treatment because the positive effects on your peripheral tissue outweigh the potentially inhibitory effects on insulin secretion. But that doesn't mean we can't get even better results if we have derivatives which stay away from the pancreas. We need to develop or refine drugs, which will be specific for the isoforms expressed in those different tissue sites. So they'll activate AMPK in the muscle and liver but leave the beta cell untouched.”

The second target, SREBP, is a master regulator of lipogenesis (the synthesis of fats). When SREBP is too active, the beta cell becomes clogged with fat molecules, which disrupts its glucose-monitoring ability.(5) But SREBP also plays a positive role in slowing the onset of diabetes. Before full-blown diabetes develops, the beta cell compensates for defective insulin secretion by making more insulin – and this compensation process seems to require SREBP.

As with AMPK, the challenge for pharmacologists will be to juggle the positive and deleterious effects and find a balance between the two.

Food on the brain?

An intriguing question for Professor Rutter and his team is whether some of the mechanisms identified in the beta cell also exist in the brain. A small region deep in the brain, the basomedial hypothalamus, is the key regulator of appetite and feeding behaviour. To do this, it needs to detect the concentrations of glucose and hormones such as insulin and leptin.

In rodents, damage to the hypothalamus profoundly affects their feeding behaviour. “Most animals stop eating if they’ve had enough,” says Professor Rutter. “But if their hypothalamic function is affected, they will eat till they can’t move. The only thing that will stop them eating is the fact that they can’t drag their carcasses to the food any more.” Interestingly, their beta cells also become defective, raising questions about how obesity affects insulin secretion.

While human feeding behaviour is certain to be more complicated, there are parallels. Work by Professor Stephen O’Rahilly at the University of Cambridge in the 1990s has shown that a simple genetic lesion affecting the hypothalamus produces extreme over-eating disorders in children. These children become uncontrollably hungry, eating a huge meal then feeling desperately hungry five minutes later. Their behaviour changes: they will tear locks off fridges to get at packets of frozen peas, and they become cunning and lie to get food – and in consequence become enormously overweight.

The link between obesity and diabetes has not yet been fully explained. Professor Rutter hypothesises that changes in fat content in the beta cells, and their effects on glucose sensing, may also affect glucose-sensing cells in the hypothalamus, so that the normal regulation of appetite is disturbed. “There could be a scenario whereby because someone has become overweight, fat is deposited in the wrong place, for example in these cells within the brain, and as a result the brain is telling them to eat more. It could be a vicious circle. There’s no direct evidence for it but it’s a theoretical possibility.”

Islet transplants

As well as looking at drug targets, Professor Rutter is examining other potential treatments for diabetes – including transplantation of donor islets for people with type 2 diabetes. The procedure itself is fairly straightforward: donor islets are injected into the liver, where they nest. The problem, as with all transplants, is rejection of the donated material.

However, a few years ago, a team at the University of Alberta in Canada, headed by Drs James Shapiro and Ray Rajotte, developed a new immunosuppressive regime, known as the ‘Edmonton Protocol’, which reduces rejection. As a result, people with type 1 diabetes who have had transplants can remain insulin-independent for a number of years.

Nevertheless, beta-cell transplants have a number of drawbacks. One is the lack of islet material available – complicated by the fact that the pancreas must be removed from the donor while the heart is still beating. If not, islets are quickly degraded by pancreatic enzymes. In addition, natural turnover of beta cells leads to the gradual loss of donated tissue. Professor Rutter’s team is collaborating with Dr Richard Smith in Bristol’s Southmead Hospital to enhance the process by introducing genes into the islets that enhance beta cell survival before transplantation.

A key goal among diabetes researchers is therefore to develop a supply of beta cells from human embryonic stem cells. Until recently, this has not been done convincingly. Although it is possible to push human embryonic stem cells to become neurons or muscle tissue, convincing them to become beta cells has turned out to be much harder. “We need a much better understanding of the whole tree of events in which transcription factors usually become switched on and off at key developmental stages,” says Professor Rutter.

He is currently collaborating with Dr Timo Otonkoski’s group in Finland, which has had some success in coaxing human embryonic cell lines toward a pancreatic fate. His role will be to take those cells and establish how close they are to real beta cells. “When we get towards that we can think about how these can be translated into treatment of diabetics. But there’s a long way to go,” he warns.

References

1 Lavantesi et al. Metabolic syndrome and risk of cardiovascular events after myocardial infarction. J Am Coll Cardiol 2005;46:277–83.

2 Ravier MA, Rutter GA. Glucose or insulin, but not zinc ions, inhibit glucagon secretion from mouse pancreatic cells. Diabetes 2005;54:1789–97.

3 Varadi A et al. Kinesin I and cytoplasmic dynein orchestrate glucose-stimulated insulin-containing vesicle movements in clonal MIN6 beta-cells. Biochem Biophys Res Commun 2003; 311(2):272–82.

4 Varadi A et al. Myosin Va transports dense core secretory vesicles in pancreatic MIN6 B-Cells. Mol Biol Cell 2005; 16(6):2670–80.

5 Diraison F et al. Over-expression of sterol-regulatory-element-binding protein-1c (SREBP1c) in rat pancreatic islets induces lipogenesis and decreases glucose-stimulated insulin release: modulation by 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR). Biochem J 2004 Mar 15;378 Pt 3:769–78.