Insulin and Diabetes
This essay was written by Guy Dodson and was first published in the 1998 Mill Hill Essays.
There are two major forms of diabetes, one which has a sudden onset and which particularly strikes the young. It has several medical names, juvenile diabetes mellitus, insulin-dependent diabetes mellitus and Type I diabetes. The symptoms of severe cases were described accurately in the writings of the ancient Egyptians, Indians and Greeks; unquenchable thirst and frequent urination followed by rapid wasting of the body and for many, death, usually within months. The other major form of the disease, referred to as adult onset diabetes or non-isulin-dependent diabetes mellitus or Type II diabetes tends to occur later in life and generally has milder symptoms. It is less serious and with careful diet many patients live reasonably normal lives without the need of insulin injections. About five in one thousand of the human population are affected by the severe form of the disease and about twice as many by the milder form. Thus there are between one quarter to three quarters of a million people affected in the UK, and worldwide about one hundred million – though this is a much less certain figure.
The cause of juvenile onset diabetes, which can kill so quickly, remained a mystery until this century. Such clues as there were about its nature could not be interpreted. The presence of sugar in the urine suggested a metabolic disorder and explains the origin of the full medical name, diabetes mellitus, a flow of sweetness. Normally after food is broken down in the stomach the sugars produced are transported in the blood to other parts of the body. Diabetics cannot transfer this sugar from their blood into the cells of their body. The sugar in the urine is a consequence of its high levels in the blood. Without sugar the cells are starved and are forced to use fat and proteins of tissues such as muscle for energy. For the diabetic the consequence the failure of the sugar to get to the cells is starvation. Again in the nineteenth century it was realised that an organ in the body, the pancreas, was associated with the disease. This deduction came from autopsies of diabetics whose pancreases were often found to be in poor condition. Confirmation of this connection was made by a german, Doctor Minkowski in Strasbourg, who carried out careful experiments on a dog from which the pancreas had been removed. The animal’s urine had a high sugar level – the dog was diabetic! Following this decisive discovery experiments to find the pancreatic secretion that controlled the level of sugar in the blood were undertaken all over the world. The origin of the secretion was soon localised to a distinct collection of cells in the pancreas, which were called the Islets of Langerhans, but it was not until 1921 that the secretion was identified and named insulin
The discovery of insulin and the demonstration that its injection could control blood sugar levels and give diabetics life, was one of the great moments in medicine, and a remarkable story of human character and the accidents that are so important in research. This landmark discovery was the result of experiments carried out over the summer of 1921 in Toronto by two Canadians, Fred Banting, a young surgeon and Charles Best, a medical student working in the laboratory of Professor JJ McCleod at the Medical School in Toronto. It is interesting to note in connection with the debate on the value of teaching for research and research for teaching, that Banting got the idea for his experiments while doing background reading for a course that he was teaching on sugar metabolism. He was impressed by an account in a fairly obscure surgical journal of a patient who had stones in his pancreas, a tissue in the small bowel, which as a result had been largely destroyed. Surprisingly the patient was not diabetic. This suggested to Banting the key idea that a small part of the pancreas which secreted the active substance might survive when the pancreas was lost, that it might be recovered and the active ingredient be extracted. If the source of this secretion could be characterised it might lead to better understanding of the nature of diabetes and suggest ways of treating it. With McCleod he planned a series of surgical experiments in dogs in which the digestive functions of the pancreas were destroyed but the secreting machinery controlling sugar levels remained intact. The surgery therefore did not cause diabetes. Extraction of the sugar controlling secretion was then be carried out, the material was purified and its activity tested on diabetic animals. It says something about McCleod’s judgement that although skeptical and aware of many others having tried somewhat similar experiments, he listened to the unknown Dr Banting and approved the project, giving a good deal of sensible advice and even providing supplies.
McCleod left for an extended trip to Europe arranging to keep in some sort of contact by letter. There were testing experimental problems made much worse by the heat, in the high nineties, of the summer in Toronto. Banting and Best worked with dogs, which was difficult and required new skills. In addition the techniques for isolating biological molecules like proteins, which insulin proved later to be, were very primitive indeed and made much more difficult by the shortage of proper equipment. And of course nothing was known about the composition or chemical and biological behaviour of the secretion. The experiments to determine these were hampered by the molecule’s instability and easy breakdown, behaviour that was made much worse by the heat. Fortunately Banting and Best’s commitment and perserverance worked, perhaps their ignorance of some of the problems facing them was even an advantage. At all events they succeeded where others apparently more qualified had failed.
Insulin, as the secretion came to be called, with its ability to restore the dying to health so dramatically, had an enormous impact on medicine and on the world, and it was not surprising that it led to Banting and McCleod being awarded the Nobel prize for medicine in 1923. Because of its medical importance, chemical, biochemical and medical research on insulin was undertaken in many laboratories all over the world. Chemical research showed that the molecule was made up of amino-acids and it became clear that it was a protein. In protein molecules amino-acids are linked together in a specific order to form long chains which fold up into specific shapes. These shapes are necessary for the protein to work properly; unfolded insulin is inactive. In general, proteins are much less stable than small chemical molecules and are easily unfolded and broken down. This explains the many initial difficulties experienced by Banting and Best, and other researchers in extracting and purifying the insulin molecule in the early days after its discovery in Toronto.
Insulin was the first protein to have the specific order in which its amino-acids are liked together worked out. This was accomplished by Fred Sanger in Cambridge and was completed in 1954. It settled the arguments as to exactly how proteins were constructed from amino acids, and for this major scientific achievement Sanger was awarded the Nobel Prize. The knowledge of insulin’s chemical structure stimulated laboratories in the USA, West Germany and in China to undertake its complete chemical synthesis, a huge commitment. All three laboratories finished their syntheses in the early 1960’s essentially within months of each other, an extraordinary coincidence given the differences in their resources. The insulin syntheses were the first ever done of a protein and they were important in showing that chemistry was developing powerful methods through which proteins could be modified and manufactured for research and pharmaceutical use.
At the same time biochemists and medical scientists were working out where insulin was made and stored in the body before its release into the blood. It was found that during its production insulin undergoes a series of complex changes and that it is finally stored as very small crystals in minute packages from which the active molecule is released. In these packages of crystals the insulin exists as a compact roughly spherical cluster of six molecules. In this arrangement, conveniently called a hexamer, insulin is very stable and crystallises easily protecting the hormone while it is stored. When released into the blood the cluster dissolves and the individual insulin molecules are free to circulate and act to carry out their functions.
It took a long time to determine the shape or the three dimensional structure of the insulin molecule. This was accomplished in 1969 after thirty-five years of effort by Dorothy Hodgkin. Working in Oxford she used x-ray crystallography, a technique based on the diffraction of x-rays by crystals. The insulin in the crystals that she studied was in the same spherical arrangement that occurs naturally in the storage packets in the Islets of Langerhans. The individual insulin molecule turned out to have a compact wedge shaped structure, ideal for packing the individual molecules into the larger spherical clusters of six molecules. This knowledge of the insulin molecule’s shape turned out to be essential some fifteen years later in experiments to design and then to manufacture modified insulins with improved therapeutic properties.
The two major types of diabetes are quite distinct in their origins. Recent research has shown that the severe juvenile onset disease has a tendency to be inherited in families. When members of such families are infected by certain viruses their natural immune defences that are directed against the viruses unfortunately also attack the insulin producing cells, stopping insulin production and causing diabetes. The typical pattern of Type I diabetes of early onset which characteristically peaks in autumn and winter seems to be a reflection of the frequent infections experienced by children at this time of the year. Adult onset or Type II diabetes has a variety of causes. It is associated with diet and obesity, which at least in part explains its tendency to occur in middle-age; but it also tends to run in families. Since 1921 the treatment of both forms has saved countless lives. And even in the case of the adult-onset diabetic, the recourse to insulin when necessary has often been vital for health and survival.
Insulin’s discovery however was only the beginning of the work required for the clinical treatment of diabetes. To provide purified protein in the large amounts needed for injection by millions of patients required new techniques of production and purification to be developed on a very large scale. Early efforts to improve the purification were beset by many technical problems, mostly arising from the unstable nature of the protein. Production of insulin steadily improved however as the techniques became established and by the mid 1920’s, an extraordinarily short time considering there was no precedent for this kind of operation, insulin was being used in the treatment of diabetics all over the world. Effective treatment of the disease required uniformity in the preparations, a new problem for the medical and pharmaceutical communities. By 1926 there was an agreement by the League of Nations to establish international standards for the quality and amounts of insulin, and other biological molecules coming into use at this time. The Director of the Institute, Sir Henry Dale, who had provided advice on the formation of the Government Committee on Biological Standardisation, played a critical part in steering the international agreements on these standards.
As Banting and Best had showed, the amount of sugar in the blood of diabetic patients could be controlled by injecting insulin. Although injection may seem a crude mechanism compared with normal secretion, and its timing to match food intake can be inaccurate, the procedure works remarkably well in many ways. There are however problems – more important for some individuals than others. Diabetic patients often develop a number of complications which arise from failure of the small blood vessels or capillaries that carry the blood in tissues. These are particularly important in the retina of the eye, in peripheral regions of the body like toes, and in organs such as the kidney. The fact that diabetes is the major cause of blindness in the advanced countries is a reminder of the scale of the disease and the long way still to go in effecting a completely satisfactory treatment.
It has been suspected for some time that deterioration of the small blood vessels in diabetics stems from the poor control over the level of blood sugar given by insulin injection. In a healthy individual there is a sensitive regulation of secretion from the insulin producing cells into the blood. Thus when food is eaten and the blood sugar level rises, insulin is released rapidly. Equally, when the sugar concentration in the blood is restored to normal, insulin release is quickly shut off. Regulation of insulin release ensures that the blood sugar concentration is maintained precisely at the normal level. By contrast injected insulin is released slowly from the site of injection and generally fails to match the two to three-fold increase in blood sugar following intake of food. Moreover the diabetic has no mechanism to stop insulin release from the injection site when normal blood sugar levels have been reached and the insulin continues to circulate, reducing blood sugar levels considerably below normal.
The consequences of the slow release of injected insulin then are episodes of high sugar levels in the blood during and for a while after a meal. This abnormally high sugar concentration in the blood leads to sugar molecules becoming attached to blood proteins such as haemoglobin, and proteins in the blood vessel walls. It seemed likely that these reactions with sugar were damaging to the small blood vessels and perhaps to other cells in tissues generally, and probably contributed to the complications associated with the disease. This association has now been confirmed in clinical studies which showed conclusively that when blood sugar was kept at normal levels the complications seen in many diabetics were reduced very significantly. The slow but long term action of injected insulin also creates problems by over-regulating the levels of blood sugar so that they fall below that required. If this process goes too far brain cells will be starved of fuel and diabetics will behave erratically, may even go into a coma and in extreme cases can die. Because of this there were strong medical pressures to develop ways to increase the rate of insulin release from its injection site so that control of blood sugar concentrations could be improved.
Over the years pharmaceutical laboratories have developed increasingly reliable and stable insulin preparations but have not succeeded in producing a satisfactory formulation in which insulin acts rapidly enough to match the increase in blood sugar levels during a meal. The basis of slow release into the blood is simple. Insulin is generally injected as the hexameric cluster of molecules, the stable form of the protein when it is stored in the body. Unfortunately this cluster is too big to move quickly through tissues and it takes time for the clusters to dissolve and free the individual molecules which are small enough to get through tissues quickly. Thus while the hexameric cluster which has evolved to be stable in the insulin producing cells is useful for producing well behaved clinical preparations, it cannot produce a rapid insulin action when injected into a muscle.
An obvious solution to this dilemma would be to inject single insulin molecules, but there is no easy way to prepare them by the established chemical and pharmacological methods – single molecules are unstable and rapidly form hexamers. The new technology of protein engineering however has provided an answer. It is now possible to produce human and animal proteins like insulin in bacteria on an industrial scale, and also to alter the type of insulin produced by changing specific individual amino acids in the protein chain. This process is often called protein engineering and has opened up radically new possibilities for modifying protein molecules and their properties. Jens Brange from the Novo Research Laboratories proposed in 1985 that the protein engineering approach based on changing insulin’s DNA should be used to manufacture better molecules for therapy. By changing the amino acids in insulin on the surfaces that stick the monomers together in the hexamer it has been possible to prevent the formation of hexamers and to make stable monomers. In clinical trials the monomers were seen to be absorbed much more quickly than the normal hexameric insulin cluster, and to give more natural control of the blood sugar concentrations.
Monomeric insulin preparations are now established in the repertoire of clinical therapies for the treatment of insulin and there is no doubt that continuing research will lead to further improvements in these engineered insulins. There is also research into exploiting the properties of engineered insulins to develop other methods of delivery that are easier for the patient. Possibilities include inhaling through the nasal passages or by passage through the skin, or by mouth. Whatever the outcome of these researches it is clear that protein engineering has had, and will have, a real impact on the effectiveness of insulin in treating diabetes.
Are there any lessons to be learned from the discovery of insulin. In 1920 there was considerable research into diabetes going on throughout the world. Much of it was rather similar in nature, attempting to find the secretion in the pancreas that Minkowski had shown in his decisive experiments were responsible for controlling blood sugar. Banting and Best’s approach was not really new, but there were original ideas in extracting the molecule. I suspect however that an application for financial support from these two, one with little experience in research and the other merely a student, may not have been funded by the Medical Research Council then or now. To his credit McCleod thought the experiments were worth supporting simply perhaps because he knew an answer to the problem had to be there somewhere. One lesson is that it is crucial that modest experiments can be carried out in laboratories today without the need for specific external money which is too often time consuming and difficult to get. Another lesson is that local, national and international research which set the scene for Banting is still essential for stimulating ideas. A third lesson is that laboratories that provide opportunities for the young, and especially those with the essential gifts of imagination and tenacity, will foster the environment from which so many unexpected and important discoveries will come.