Version classiqueVersion mobile

Animals and Medicine

Jack Howard Botting

III. Drugs for Organic Diseases

13. Animal Experiments and the Production of Insulin

Note de l’auteur

An earlier version of this chapter was published as: Animal experiments and the production of insulin. RDS News January 1996 9-14.

Texte intégral

1Before 1922, the diagnosis of what was then called juvenile onset diabetes (type I or insulin-dependent diabetes, IDD), meant a lingering death within months. In that year, however, a team of workers in the physiological laboratories at the University of Toronto isolated and purified the hormone, insulin, from the pancreas. The purified insulin was shown to control not only the symptoms induced by removal of the pancreas in dogs, but also those of diabetes mellitus in patients.

2The production of insulin on a large scale from pig and cattle pancreas was achieved fairly rapidly, due no doubt to the striking benefit that injection of insulin could produce in the seriously ill, even comatose, diabetic patient.

3The effects of the ready availability of insulin were dramatic. This is illustrated by objective actuarial data produced by Metropolitan Life Insurance. This leading American insurance company stated: “The results of this treatment have been brilliantly successful.” The data showed that the average age at death of diabetic patients in Toronto had increased by 22 years between 1920 and 1931. It also showed that: “Deaths from diabetes among children under 20 years-of-age have almost ceased, and the death-rates under 50 years-of-age have declined very sharply.” (1).

4The success of the Toronto team; Banting, Best, Macleod and Collip is rightly acclaimed by biomedical researchers as a prime example of undoubted benefit for patients achieved by animal experimentation. Presumably for this reason, the story of the discovery of insulin has been subjected to vehement attacks from animal rights adherents.

Early History of the Nature of Diabetes

  • 1 There is a claim that an exactly similar experiment to that of von Mering and Minkowski, demonstrat (...)

5Most physiologists accept that the observation of von Mering and Minkowski, that removal of the pancreas of dogs produced symptoms similar to those of patients with diabetes, was the crucial experiment linking the pancreas with the disease (2). Opponents of animal experiments claim that this observation was antedated1 by significant work by Cawley, who in 1788 noted lesions in the pancreas of a diabetic patient at post mortem (3). Such an observation of course does not prove the lesions cause the disease, only that there may be an association between the disease and the observed damage. Changes in the blood vessels, kidney, retina and nerves also occur, but as a result rather than a cause of the condition. Nevertheless, antivivisectionists claim that Cawley’s observation was a significant clue. They also imply that had this observation been followed up it could have provided a treatment for diabetes without recourse to animal experiments. Quite how this could have been achieved is not detailed by those who hold this view, and is not apparent to this writer.

Isolation and Purification of Insulin

  • 2 Schafer apparently suggested this term in 1915. In some antivivisection propaganda, Schafer is impl (...)

6By the end of the nineteenth century, the function of various glands was being investigated by their removal from animals and by examination of their extracts for physiological activity. Thus, adrenalin was extracted from the adrenal and thyroxin from the thyroid gland. It was therefore predictable that the then putative “antidiabetic” factor thought to be present in the islet cells of the pancreas should be designated “insulin” even before its successful isolation.2 It was obvious at this time that insulin in some way enabled blood sugar to be properly utilised by the body.

7The morbidity and mortality associated with diabetes ensured that there were immediate attempts to extract a factor (insulin) from the pancreas that could be used to treat this disease. This proved less easy than for adrenaline. Minkowski tried, as did the French physiologist Gley, but probably the most important of the early studies was that of Zülzer (6). Zülzer used alcohol, rather than water, to extract insulin and he certainly produced active preparations. Forschbach, working in Minkowski’s clinic in Breslau showed in 1909 that one of these extracts could reduce the blood sugar of depancreatized dogs by 90%. Unfortunately it also raised the temperature of the animal to 102.6°. Attempts to produce extracts of greater purity were made by chemists of the Schering Company, and these were given to two diabetic patients. The first produced no effect, the second caused a rise in temperature to 104 ° and the patient was unable to void urine for 12 hours. As a result of this toxicity the use of these extracts was abandoned in Europe at this time (7).

The Work in Toronto

8Frederick Banting, a surgeon, and the medical student Charles Best began their attempts to prepare usable extracts of the pancreas in May 1921. Banting believed that it would be difficult to extract insulin, if it should be a protein, since during the process of extraction it would be bound to come in contact with protein-splitting enzymes which are also present in the pancreas. He therefore proposed ligating the pancreatic duct as a means of producing atrophy of the protease containing tissue. Extracts made from the pancreas so treated, Banting surmised, would have high insulin levels.

9Banting’s premise was in fact false on two counts (7). For this reason his published work came in for some criticism which has provided a focus for irrational condemnation of the work of Banting and Best and the role of animal experiments in the discovery of insulin (vide infra).

10First, it is extremely unlikely that the pancreatic tissue could contain an active protease, since the active enzyme is only produced if its precursor comes in contact with the intestine. So there was no necessity to produce atrophy of the gland. Secondly, ligation of the pancreatic duct can also destroy the islet tissue. It was not therefore, as Banting and others believed, a technique that caused selective destruction of the acinar tissue.

11Banting presented his first results at the New Haven meeting of the American Physiological Society in December, 1921 (Fig. 13.1). Extracts from both normal and “ligated” pancreas were shown to reduce blood sugar and abolish glycosuria in depancreatized dogs. The extracts had not been injected into patients. When this was tried a few weeks later the results were disappointing. The patients developed fever, and abscesses occurred at the injection site.

Fig. 13.1 The now famous picture of Frederick Banting, Charles Best and the dog Marjory, an early depancreatised dog treated with insulin, 1921. Wellcome Library, London, CC BY.

Non-Toxic Extracts

12At this time Macleod, the head of the Physiology Department at Toronto, was persuaded to add a biochemist to the team to assist in the purification of the pancreatic extracts. This turned out to be a significant event. Within 2-3 weeks the biochemist, James Collip, prepared the first sample of insulin pure enough to treat diabetic patients from beef pancreas obtained from the slaughter house (7).

13Collip used established techniques based on the selective solubility of different proteins in ethyl alcohol. This purification method required the measurement of the amount of insulin present in the various fractions produced. Since neither the nature nor the structure of insulin was known, the only way of assessing the amount of insulin present was by measuring its activity, i. e. by its ability to lower blood sugar levels in animals. Collip assessed the potency of each precipitate and filtrate in the process of fractional precipitation of the crude extracts by subcutaneous injection into fasted rabbits. One “unit” of insulin activity was described as that amount that would reduce the percentage blood sugar of a 2kg rabbit to 0.045 within four hours (8). The extracts prepared for clinical use were adjusted to contain one such unit in one millilitre. Thus animal experiments were absolutely crucial in the production of samples of insulin for the treatment of diabetes. An accurate method of standardisation of the extracts was essential. Even moderate overdosing with insulin can have serious consequences. Frank overdosing is lethal.

Clinical Success

14The extracts prepared by Collip were used in dogs and patients with remarkable success (Fig. 13.2).

Fig. 13.2 The effect of Collip’s highly-purified extract on the first patient to be successfully treated. Fourteen year old Leonard Thompson was given 10cc of the extract at 5pm on 23 January. The next day two injections of 10cc were administered. Sugar in the urine almost disappeared, ketonuria did disappear. Doses of 8cc of a new extract (presumably more potent) were given daily from 27 January onwards, markedly reducing sugar excretion in the urine. Data adapted from F. Banting, C. Best, J. Collip, et al. (1922), The effect produced on diabetes by extracts of pancreas.’ Transactions of the Association of American Physicians, 1-11.

15When the results of the Toronto team were reported at a meeting of the Association of American Physicians in May 1922, the distinguished diabetologist Dr Frederick Allen said:

If, as seems to be the case, the Toronto workers have the internal secretion of the pancreas fairly free from the toxic material, they hold unquestionable priority for one of the greatest achievements of modern medicine (7).

16Unless one has studied the case reports one can have no conception of the dramatic impact of the production of purified insulin. Diabetics placed on starvation diets, were reduced to emaciation and were barely able to hang on to a miserable life. With regular injections of insulin they could absorb and utilise a normal diet, gain weight, regain their vigour and flourish mentally and physically (Fig. 13.3).

Fig. 13.3 Photographed in 1922, this diabetic girl, aged 13, weighed just 45lb before treatment with insulin. A few months later she had made a dramatic recovery. Wellcome Library London, CC BY.

The Story of Elizabeth Hughes

17A good example is that of Elizabeth Hughes. The daughter of an erstwhile Governor of New York State, she developed diabetes at the age of 11 years in 1918. Under the guidance of probably the best-known American diabetologist, Dr Frederick Allen, she was placed on a restricted diet. Elizabeth became semi-invalid and was plagued by frequent infections. By 1922, although 5 feet tall she weighed under 50 lbs, only kept alive by her remakable spirit and optimism. Elizabeth was taken to Banting’s clinic in Toronto in August 1922. Banting’s notes after his first examination of Elizabeth were:

Weight 45 lbs. Height 5ft patient extremely emaciated, slight aedema of ankles, skin dry and scaly, hair brittle and thin, abdomen prominent, shoulders drooped, muscles extremely wasted, subcutaneous tissues almost completely absorbed. She was scarcely able to walk on account of weakness.

18Banting placed Elizabeth on regular injections of insulin and, reasoning that she needed above all to put on weight, prescribed a liberal diet. For the first time for over three years Elizabeth ate bread, potato, macaroni cheese and also was allowed a daily pint of thick cream. She gained 10 lbs in 5 weeks, then gained a steady 21/2 lbs per week. Elizabeth grew taller and, no longer restricted by her lack of energy to armchair activities, enjoyed trips to concerts, the cinema and even visited Niagara Falls.

19In November 1922 the leading diabetologists of North America came to Toronto for a conference to discuss the initial clinical findings with insulin and to recommend methods for the standardisation of insulin dosage. Banting took a group of physicians, including Frederick Allen, to visit patients in his clinic. Allen did not recognise Elizabeth, and was rendered speechless by the healthy appearance of his former patient. Elizabeth (who was a compulsive and eloquent correspondent) described Allen’s reaction in a letter to her mother: “Dr Allen said with his mouth wide open – Oh! – and that’s all he did”

20On November 30th, Thanksgiving Day, Elizabeth Hughes returned to her home in Washington. A year after graduating from college Elizabeth married. She raised three children and also in other ways led an active life, being an inveterate world traveller. On April 25th, 1981, almost 60 years after receiving her first dose of insulin, Elizabeth died of a heart attack (21). The discovery of insulin undoubtedly was “one of the most dramatic achievements of modern medicine.”

Mixed Reactions

21The familiarity that develops with passage of time obtunds the impact that a discovery such as the isolation of insulin had upon the medical profession and indeed upon the lay public. An effective way to get some sense of the reaction is to read contemporary comment in the general medical journals. A review of the research findings Banting reported at the May, 1922 meeting of the Association of American Physicians states:

If these experimental observations prove clinically applicable to man clearly a magnificent contribution to the treatment of diabetes will have been made (9)

22Other comments soon followed:

Results which, it is not too sanguine to hope, have opened a new chapter in the history of the treatment of diabetes.
As our readers have learnt from Professor Macleod, the effects observed in animals have justified the trial of insulin in man, and diabetic patients have eagerly flocked to take advantage of the new treatment (10).
Readers… cannot fail to draw the conclusion that a scientific advance of immense importance has already been accomplished (11).

23As always, such approbation inevitably provoked some adverse criticism from the more conservative members of a markedly conservative profession. A particularly disparaging criticism appeared in the British Medical Journal on December 16th 1922. Dr Ff Roberts of the Department of Physiology, Cambridge University wrote:

So much prominence has been given to the “insulin” treatment of diabetes that it is perhaps not out of place to review the steps which have led up to the production of this new remedy.

24Roberts went on to assert that the hypothesis that ligation of the pancreatic duct would result in the destruction of the acinar tissue was incorrect, that in any case there was no need to attempt such an experiment for there is no active protease in the pancreas, and that Banting and Best’s results clearly show that there was as much (or more) insulin activity in the “normal” as the “ligated” gland. Roberts concluded:

The production of insulin originated in a wrongly conceived, wrongly conducted and wrongly interpreted series of experiments.

25He conceded:

apparent beneficial effects have been obtained in certain cases of human diabetes. Whether insulin will fulfil its promise time alone will show.

26The next issue of the Journal (23 Dec, 1922) contained a letter from Henry Dale (later to become Sir Henry, Nobel Laureate) which was a devastating reproof of the “censorious criticism” by Roberts. Dale was at the National Institute of Medical Research in London, and was concerned with the standardisation by biological assay of the batches of insulin that were being prepared at that time. Dale’s eloquent letter concludes:

Nobody can deny that a discovery of first-rate importance has been made, and, if it proves to have resulted from a stumble into the right road, where it crossed the course laid down by a faulty conception, surely the case is not unique in the history of science. The world could afford to exchange a whole library of criticism for one such productive blunder, and it is a poor thing to attempt belittlement of a great achievement by scornful exposure of errors in its inception.

27The ungraciousness of the comments by Roberts were also elegantly put into perspective by a leader that appeared in the British Medical Journal later:

Dr Roberts’ points are of a technical kind, and the terms in which they were expressed, it must be admitted, were wanting in urbanity (12).

The Criticism from Antivivisectionists

28Despite the immediate condemnation of the somewhat ungenerous comments by Roberts, the quotation: “a wrongly conceived, wrongly conducted and wrongly interpreted series of experiments” continues to appear regularly in antivivisection literature and leaflets (5, 13), put forward as evidence that the public has been misled as to the value of the animal experiments of the Toronto group. This is quite wrong. Each of the Toronto team made an important contribution to the production of clinically useful insulin preparations. It was certainly the biochemist Collip who was responsible for producing extracts relatively free from toxic material, but as mentioned, the use of animals to monitor the purification of insulin was an absolutely essential part of this process. It is to be regretted that even advocates for moderate organisations supporting the development of alternative methods in medical research perpetuate the myth “that chemical extraction techniques (not involving animal tests) allowed the production of pure and safe insulin” (14).

29Over half a century’s experience has resulted in refinements in the use of insulin, particularly in the development of slow-release depot preparations to provide smooth therapy with the minimum number of injections. The miserable prospect for the IDD patient in the early 1920s is beyond the imagination of the lay public today, two generations divorced from direct anecdotal accounts. This public ignorance is utilised by antivivisectionists in their attempts to belittle the contribution of insulin to the treatment of diabetes. They assert that it is an “absurd nonsense to claim diabetes has been cured” since the number of people dying of diabetes has increased over the last 50 years and the number of diabetics is doubling every ten years (15). The fatuity of such arguments is patently obvious. They were summarily dismissed over 10 years ago by Rowan, who commented:

These tactics are born of desperation. Since antivivisection protestors cannot deny that insulin therapy was derived from animal experimentation, the therapy itself is denigrated (16).

30Insulin was never held to be a cure, it is merely a replacement therapy. Obviously, if diabetics treated with insulin can now survive for several decades rather than dying within a year or two of diagnosis, then the number of diabetic patients will increase. Even in insulin-treated diabetics the late complications of diabetes are often the ultimate cause of death. Thus deaths due to diabetes will appear to rise.

31Another smoke screen raised by antivivisectionists is to introduce late onset diabetes (type 2 or non-insulin dependent diabetes, NIDD) into the equation. NIDD is of course more common than IDD and can often be treated by changing the diet. Thus antivivisectionists allege: “Today the great majority of diabetics who develop diabetes as adults control their disease through diet – they certainly do not need insulin injections.” (15)

32Stated in the context of discussion about the value of animal experimentation in the production of pure extracts of insulin, which is life-saving in IDD, this statement is wholly misleading. It can only be assumed that its intention is deliberately to misinform the public.

The Future

33Human insulin can now be prepared by recombinant techniques. Physicochemical methods obviate the need for tedious bioassays of insulin extracts, no doubt to the considerable relief of those responsible for quality control.

34However, administration of insulin by injection can never mimic the refined, minute by minute control of insulin release from the β-cells of the pancreatic islets, which is governed by plasma glucose levels. Thus diabetics are frequently hyper-or hypoglycaemic, with obvious pathological sequelae.

35Transplantation of islet cells, which could release insulin as required, is an obvious answer once the problems of supply of cells and tissue rejection have been resolved. Porcine cells have been transplanted into nude mice and dogs. Rejection was prevented in the latter by antilymphocyte serum (raised in rabbit) and 15-deoxyspergualin. Such transplants have been attempted with some success in Swedish patients (17).

36Another ingenious attempt to prevent rejection is the implantation of islet cells in capsules made of alginate. The capsule wall prevents access of host white cells to the islet tissue, but is sufficiently porous to allow passage of glucose and insulin. Exploratory experiments in diabetic dogs showed that a single treatment could replace injection of insulin for 6 months to 2 years (18).

37Similar animal experiments will no doubt ultimately reveal the significance of the newly isolated compound “amylin.” This peptide, which appears to be co-secreted from β-cells with insulin, increases blood sugar and is present in excess in NIDD (19).

38Finally, a strain of mouse that naturally develops IDD (the non-obese diabetic [NOD] mouse) should eventually reveal the nature of the antigen that activates the immune system to destroy the islet cells of IDD patients (20). Even if the particular combination of genes that determine the genetic susceptibility to diabetes is established, a model such as the NOD mouse would be essential to establish the ultimate mechanism that triggers the destruction of the insulin-producing cells. Research with such models will set the scene for the prevention, rather than control, of IDD.




Von Mering & Minkowski

Pancreatectomy produces diabetes mellitus (dog)


Zülzer, Gley. Banting & Best

Pancreatic extracts lower blood sugar (dog)


Banting, Best, Collip, Macleod

Extracts made from cows and pigs


Banting et al.

Biological standardisation of extracts (rabbit, mouse)


1) Rogers R (1937) The Truth About Vivisection. London: Churchill, p. 109.

2) Von Mering J & Minkowski O (1889) Diabetes mellitus nach pankreas extirpation. Archiv Exp Path Pharmak 26 371.

3) Sharpe R (1988) The Cruel Deception: The Use of Animals in Medical Research. London: Thorsons.

4) Yudkin J (1992) New Scientist 25 July, p. 51.

5) BAVA Leaflet (1995) Lies, damn lies and vivisection.

6) Bonta I (1983) Folklore, druglore and serendipity in pharmacology, in Discoveries in Pharmacology, vol. 1. ed Parnham M & Bruinvels J. Elsevier.

7) Pratt J H (1954) A reappraisal of researches leading to the discovery of insulin. J Hist Med 9 281.

8) Macleod J (1922) Insulin and diabetes. Brit Med J 2 833.

9) Anon (1922) Insulin and diabetes. Brit Med J 2 140.

10) Ibid. p. 991.

11) Ibid. p. 882.

12) Ibid. p. 1233.

13) Overell B (1993) Animal Research Takes Lives. NZAVS.

14) Newman C (1994) Diabetes – what, why and how? Alternative News no 52 Dr Hadwen Trust.

15) Coleman V (1991) Why Animal Experiments Must Stop. London: Green Print.

16) Rowan A (1984) Of Mice, Models, and Men. SUNY Press.

17) Anon. (1995) Transplant of porcine pancreatic cells. RDS NEWS January, 1995.

18) Day S (1993) Jelly capsules offer end to diabetics daily dose New Scientist 4th September, 16.

19) Amiel S (1993) Amylin and diabetes. The Lancet 341 1249.

20) Kaufman D et al. (1993) Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 366 69.

21) Bliss M (1983) The Discovery of Insulin. Edinburgh: Paul Harris (All unreferenced items can be found in this outstanding historical account).


1 There is a claim that an exactly similar experiment to that of von Mering and Minkowski, demonstrating the appearance of diabetes in a depancreatised dog, was described in a book by Brunner in 1683, over 100 years before Cawley’s observation (4).

2 Schafer apparently suggested this term in 1915. In some antivivisection propaganda, Schafer is implied to have discovered insulin in 1915, “six years before Banting and Best’s experiments on dogs” (5). This is put forward as evidence against the contribution made by the Toronto workers, which is lauded by opponents of animal experimentation. (Adrenaline and thyroxine later acquired an ‘e’, presumably because they were eventually shown to be amines).

Table des illustrations

Légende Fig. 13.1 The now famous picture of Frederick Banting, Charles Best and the dog Marjory, an early depancreatised dog treated with insulin, 1921. Wellcome Library, London, CC BY.
Fichier image/jpeg, 28k
Légende Fig. 13.2 The effect of Collip’s highly-purified extract on the first patient to be successfully treated. Fourteen year old Leonard Thompson was given 10cc of the extract at 5pm on 23 January. The next day two injections of 10cc were administered. Sugar in the urine almost disappeared, ketonuria did disappear. Doses of 8cc of a new extract (presumably more potent) were given daily from 27 January onwards, markedly reducing sugar excretion in the urine. Data adapted from F. Banting, C. Best, J. Collip, et al. (1922), ‘The effect produced on diabetes by extracts of pancreas.’ Transactions of the Association of American Physicians, 1-11.
Fichier image/jpeg, 32k
Légende Fig. 13.3 Photographed in 1922, this diabetic girl, aged 13, weighed just 45lb before treatment with insulin. A few months later she had made a dramatic recovery. Wellcome Library London, CC BY.
Fichier image/jpeg, 24k


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search