THE TRUTH BEHIND THE DISCOVERY OF INSULIN
Brandon Reines, DVM
PART 1 / PART 2By far the most oft-cited example of the power of animal experimentation against human disease is the discovery of insulin. Insulin revolutionized the treatment of diabetes when it was first used at the University of Toronto in the 1920s. The traditional story of the discovery, told over and over again in textbooks, films, tapes and television programs, goes roughly like this:
By the early years of the twentieth century it was understood that the disease called diabetes mellitus involves the body's inability to metabolize or utilize its food, especially carbohydrates. It was also understood that the pancreas holds the key to carbohydrate metabolism. When experimental animals had their pancreases removed, they immediately lost the ability to utilize carbohydrates, the amount of sugar in their blood and urine rose sharply, and they soon died from severe diabetes. Various researchers speculated that the pancreas, which secretes digestive enzymes into the gut (its external secretion), must also produce another kind of secretion, one enabling the body to utilize its fuel. The search for the internal secretion of the pancreas had occupied a number of physiologists throughout the world, but by 1920 it had not produced any practical result.In the Autumn of 1920 Frederick Banting, a young surgeon in London, Ontario happened to be reading an article about the pancreas. Banting began thinking about the problem of the internal secretion, and late that night jotted down an idea for an experimental procedure - ligating (blocking) the pancreatic ducts - that might be a way of isolating an internal secretion. He took his idea to his alma mater, the University of Toronto, where the Professor of Physiology, J.J.R.Macleod, was an internationally-known expert in carbohydrate metabolism. Macleod was at first skeptical of Banting's suggestion, but reluctantly agreed to give him a lab and some dogs for a few weeks during the coming summer. He assigned him a young science student, Charles Best, to do the chemical tests necessary for the work, and then went off to Scotland for his summer holidays.
Banting and Best experienced a number of problems with their work in that summer of 1921, the story goes, but soon found that their approach was yielding remarkable results. With the extract of pancreas they had made from duct-ligated dogs they were able time after time to lower the blood sugar and remove other symptoms from diabetic dogs. Prof. Macleod came home to a pair of excited researchers who, by the Autumn of 1921, were keeping a severely diabetic dog, Marjorie, alive with their extracts. Marjorie eventually lived for seventy days before being sacrificed (killed); until then diabetic dogs had died within a week or two of their pancreas being removed.
By the winter of 1921-22 Banting and Best were giving their first papers on the internal secretion of the pancreas. They were also ready to test their extract on humans. In Toronto General Hospital a young boy, Leonard Thompson, became the first diabetic to receive insulin. His life was miraculously saved.(1)
Nonetheless, new historical evidence unearthed by medical historian Michael Bliss for his new book on the discovery of insulin reveals that virtually every step in the traditional account is highly questionable, if not outright false. While Bliss did not explicitly analyze the historical data to assess the importance of animal experimentation, it is clear from the factual evidence presented in Bliss' book that the animal experimentation which preceded the discovery insulin was not part of the scientific process that led to the discovery. While the various dog experiments conducted by von Mering and Minkowski in 1882 and by Banting and Best in the early 1900s were undoubtedly useful for convincing skeptics that insulin exists, all the hard scientific information which allowed researchers to discover insulin consisted of clinical data: observations of actual human patients.Physiologist themselves began to criticize Banting and Best's dog-experiments as far back as the 1920s when the experiments were performed. In fact, in the December 16, 1922 issue of the British Medical Journal, Dr. Ffrangcon Roberts published a devastating critique of the animal experiments. After studying Banting and Best's major scientific papers in the Journal of Laboratory and Clinical Medicine, Roberts reviewed the steps leading up to the discovery of insulin in Toronto. Roberts explains that the animal experiments were based upon Banting's hypothesis that it was necessary to protect the "internal secretion" of the pancreas [insulin] from being destroyed by the "external secretion" of the pancreas: the enzyme trypsin.
Banting maintained that the most effective way to destroy the trypsin in the animal body to prevent it from breaking down insulin would be to tie off the ducts that secrete trypsin, causing the trypsin-producing cells to deteriorate and atrophy. Banting insisted that all previous attempts to isolate insulin (the "internal secretion" of the pancreas) had failed because the insulin was always destroyed by the enzyme tryptin (the "external secretion" of the pancreas) before the insulin could be removed from the animal body. Roberts contended that Banting's hypothesis - upon which all of the animal work had been based - was simply false. "Now it is one of the best-established facts in physiology," Roberts wrote, "that the proteolitic enzyme exists in the pancreas in an inactive form - trypsinogen - which is activated normally on contact with another ferment, enterokinase, secreted by the small intestine."(2)
In other words, there was no possible way that trypsin could be the reason that all previous attempts to isolate insulin had failed, because trypsin itself does not even exist in the pancreas. Trypsin could not possibly be destroying the insulin. Given these facts, Roberts said, "there was no physiological basis at all for Banting and Best's duct-ligation experiment. They had undertaken a cumbersome, time-consuming and cruel process to forestall enzyme action which would never take place."(3)
Roberts concluded that Banting and Best had succeeded in reawakening interest in the possibility that a diabetes-fighting substance exists in the pancreas. They did manage to convince the medical professions that insulin did actually exist in the pancreas and that it could eventually be of value to human diabetics. But, Roberts pointed out, they did so by completely misinterpreting the results of their own animal experiments! Indeed, if Banting and Best had been truly competent scientists, they would have realized that their animal experiments did not in any way, shape or form, prove that insulin exists. Roberts wrote:
"The production of insulin originated in a wrongly conceived, wrongly conducted and wrongly interpreted series of experiments. Through gross misreading of these experiments interest in the pancreatic carbohydrate function has been revived, with the result that apparently beneficial results have been obtained in certain cases of human diabetes ... whatever success the remedy will have will be found to be due to the fact that the hormone has been obtained free from anaphylaxis-producing and other toxic substances. The experiments of Banting and Best show conclusively that trypsin ... has nothing whatever to do with it."(4)Thus, the research that actually led to the discovery of insulin was stimulated by Banting and Best's enthusiastic reports, but Banting and Best's conclusions were not based on the results of their own animal experiments. Banting and Best's claim to have isolated insulin was really a statement of their own conviction that insulin exists and could be eventually discovered. Banting's mentor, the aforementioned, then world renowned, expert on carbohydrate metabolism, Dr.J.J.R.Macleod, Professor of Physiology at the University of Toronto, responded to Roberts' criticism. It is even clear from Macleods' response however, that Roberts was absolutely correct. Macleod wrote:"The criticism has been made that Banting and Best's experiments, in which simple extracts of duct-ligated pancreas were used, formed no essential step in giving us insulin. I need scarcely reply to this criticism. They were apparently made without any appreciation of the real obstacle that stood in the way of development of the subject - namely, convincing evidence that an antidiabetic hormone does actually exist in the pancreas - and to Banting and Best is due the credit of furnishing this by experiments of a different type from those of their predecessors."(5)"In Macleod's mind," Bliss writes, "the whole importance of Banting and Best's experiments had been in convincing Macleod and the others of the team that the internal secretion was there to get. They then went and got it."(6) From this historical evidence, it is clear that the main function of animal experimentation is to convince disbelievers.Not only did Banting and Best vastly exaggerate the effectiveness of their pancreatic extract on "diabetic" dogs, but the extract was a failure when it was used to treat human diabetes - contrary to the popular perception that Banting and Best's pancreatic extract was an immediate success against human diabetes. In fact, Macleod had not originally intended that the pancreatic extract used in dogs should be used on actual human diabetics, because the results were so unclear. He had assigned to the team biochemist Dr.J.B.Collip the job of isolating and purifying insulin from ox pancreas to be used for the first human test.
Banting, however, was determined to try the extract he had used on his dogs on human patients to prove that his research had indeed been of value. Macleod eventually relented and allowed Banting and Bests to try out their pancreatic extract on a patient. The patient chosen was a fourteen-year-old boy named Leonard Thompson. On admission to the Toronto General Hospital, Leonard Tompson weighed 65 pounds on December 2, 1921. "He was pale, dull and listless, content to lie in his bed day after day."(7) The boy's father decided to let the hospital try out Banting and Best's new pancreatic extract for the first time. Contrary to popular belief, Banting and Best's pancreatic extract, their version of insulin, was a failure. Bliss writes:
Banting and Best's extract had failed. Of course, a good face could be put on the results: the 25% decrease in the blood sugar, the reduction of glycosuria (and Banting put an even better face on the results in his Nobel prize lecture by talking of a 'marked reduction' in blood sugar and saying that the urine had been rendered sugar-free). But there was the overwhelming fact that the extract's actually very modest impact did not outweigh the reaction it caused. Even though Leonard Thompson was a very sick boy, the doctors decided not to give him further injections of Banting and Best's extract. It was 'absolutely useless for continued administration to the human subject,' Collip wrote in 1923, in a mood, which we shall come to understand, of considerable bitterness. Banting himself accurately summed up the situation after January 11 when he wrote in his published "History of Insulin" that 'These results were not as encouraging as obtained by Zuelzer in 1908.'(8)For those who believe that Banting and Best's investigations on dogs were an essential part of the scientific process that led to the discovery of insulin, even though the extract that they had used on dogs was an utter failure in the clinic, it should be made clear that Banting's working hypothesis was based entirely on a study of actual patients: an autopsy study. On Sunday, October 30, 1920, Banting's copy of the November issue of the clinical journal Surgery, Gynecology and Obstetrics arrived in the mail. When he finished working on a lecture to be given to students on carbohydrate metabolism that evening, he took the journal to bed with him to read before going to sleep. Since carbohydrate metabolism was on his mind, he was immediately drawn to an article in the journal called "The Relation of the Islets of Langerhans to Diabetes with Special Reference to Cases of Pancreatic Lithiasis", by Dr. Moses Barron, an American pathologist. (9) Barron had become interested in the possible relationship of the pancreatic cells known as the Islets of Langerhans and diabetes when he happened to perform an autopsy on a patient who died of a rare syndrome known as pancreatic lithiasis: a disease in which a stone obstructs the main duct of the pancreas.