The Discovery of Insulin

by Michael Bliss

  There was also much professorial caution in Macleod’s reaction. The results were good, but there had to be “no possibility of mistake”:

  You know that if you can prove to the satisfaction of everyone that such extracts really have the power to reduce blood sugar in pancreatic diabetes, you will have achieved a very great deal. Kleiner & others who have published somewhat similar results have not convinced others because their proofs were not adequate. Its very easy often in science to satisfy ones own self about some point but its very hard to build up a stronghold of proof which others cannot pull down. Now, for example, supposing I wanted to be one of those critics I would say that your results on dog 408 were not absolutely convincing….

  Macleod hoped Banting and Best had data on the behaviour of the blood sugar of other depancreatized dogs so they could prove that their results were not merely normal diurnal variations. Did they have similar curves for dogs 406 and 410? Could the large volume of fluid injected into 408 on August 7 explain the drop in blood sugar – in other words, was it a dilution phenomenon? One of J.J.R. Macleod’s aphorisms to his students was that one result is no result.35 He advised Banting to “continue along the same lines without at the present taking up any of the problems which you suggest in your letter.” Banting and Best should pay particular attention to the preparation of the extract, setting aside a small piece of tissue from each pancreas for histological examination to confirm that it was really islet tissue that was having the effect. They were to go ahead with the experiment, mentioned in Best’s letter, of running two dogs side by side.36

  They had actually gone ahead with that experiment two days after writing Macleod, had elaborated on it by testing several of the ideas Banting had suggested to Macleod, and had compiled what appeared to be the splendid results on dog 92, of all of which Macleod knew nothing. It was too late to do the pre-injection blood sugar tests on dogs 406, 408, and 410 that Macleod assumed had been done. Sections were, apparently, being taken for histological study, though nothing ever came of it.37 Macleod’s letter, although its encouragement was gratifying, must have seemed somewhat irrelevant.38

  XI

  The experiments Banting and Best began in the week they received the letter had little to do with Macleod’s concerns or advice. Except for more careful estimations of D:N ratio (which now seldom rose above 2.0:1), it is not clear what Banting and Best were trying to achieve. One duct ligation was done on the 5th. On the 7th, after receiving Macleod’s letter, total pancreatectomies were done on dogs 5 and 9. The operation on dog 5, the notebook records, was done by Best.39 Both dogs seemed to recover well from the operations.

  The next day they tried to make some extract by the quick secretin-stimulation method. In the morning they “failed utterly,” as the secretin they injected simply did not stimulate the dog’s pancreas. “This afternoon we succeeded in getting about 15 cc. of pancreatic juice in 2 hours when dog died.” It was another what-the-hell situation, and although they had no reason to believe that the pancreas was exhausted of external secretion (the earlier stimulation experiment had produced eighty-five cc. of pancreatic juice), they ground it up anyway and considered they had made extract of exhausted pancreas.40

  That night they first tried the last ten cc. of old extract on dog 9, administering it by rectum in their first change from intravenous injection. It had no effect. Intravenous doses of the new extract seemed to work spectacularly, a total of sixty cc. driving dog 9’s blood sugar from .30 at 6:30 p.m. to .07 by midnight. “General condition improved. Injection of extract causes pain.” The injections the next day seemed to have little effect. Another rectal injection on the 10th had no effect, but the next intravenous injection caused a sharp drop. Repeating earlier experiments, they again mixed trypsin with the extract, trying to see if the external secretion destroyed the internal. Apparently it did, for the mixture had no effect.

  On the 12th they tried to exhaust a cat’s pancreas with secretin.* Some juice flowed. They made no attempt to measure it. The cat died after ninety minutes of stimulation. Banting and Best made an extract of its pancreas. Experiments with the cat extract, including injection into dog 9’s heart, caused moderate decline in blood sugar, shock, and finally death; the cause of death, they noted, seemed to be poorly ground particles in the extract damaging the dog’s veins. The control dog, number 5, which appeared to be moderately diabetic, was killed a day or two later. Why it was killed is not clear. There was considerable pus in its abdomen.

  On September 17 they began again on still another depancreatized dog, using extract made after another dubious attempt at exhaustion through secretin.41 For the first time they tried injecting the extract under the skin -subcutaneously – giving five injections at hourly intervals. The dog’s blood sugar did not change (though the fact that it did not rise, hovering around .15 to .18, may have been encouraging). In any case, this method of obtaining extract – from “more or less” exhausted glands, as they put it in their first paper – was not satisfactory. “No more subcutaneous injection till we get a trypsin free extract,” Best noted on September 18. “There is a hole the size of a halfpenny in skin where the injections were given. A superficial vein had been eaten into & there was considerable haemorrhage.”

  The dog died on the 19th, suffering from infection. No new experiments were begun. The summer’s work was over. Banting and Best’s companion in the next lab, Dr. Fidlar, also finished his summer’s experiments on the respiration of his frog. He liberated it at the exact spot in Grenadier Pond where he had caught it in the spring. On September 21, Professor Macleod arrived back in Toronto from his holidays.42

  XII

  One day in August, worrying about a problem with the bowel of one of his dogs, Banting had dropped in to the office of Velyien Henderson, the professor of pharmacology, who was doing some research on intestinal movements. They naturally began talking about Banting’s work, and Henderson took a friendly interest in its progress. As a student Banting had not much like the older, affected professor; Henderson liked to shock the students by using condoms in his pharmacological experiments, would turn an English accent off and on at will, and began lecture series by asking which of his students could wiggle their ears; the students nicknamed him “Vermin” Henderson.43 Now Banting warmed to the older professor’s interests. He saw more of Henderson and at some point in September told him how precarious his financial situation and his future appeared. Sixty years later Henderson’s secretary, Jean Orr, remembered vividly Fred Banting coming out of the professor’s office, sitting on her desk, and talking about his problems. “He put his hand in his pocket, took out seven cents and put it on the desk, and said, ‘There, that’s all I have to live on in the world, if I don’t get a job.’“44

  In mid-September the one junior man in Henderson’s department was offered a special assignment by the Ontario government. Henderson thought it would be possible to replace him with Banting, and on September 21 wrote the president of the university to this effect. Whether or not he had actually offered the job to Banting by then is not clear. But as Banting prepared to talk over his future with the newly returned Macleod, Henderson had certainly hinted to him that some arrangement might be possible.45

  Late in September or early in October Banting and Best met with Macleod. His reaction to the news of their experiments is not clear. Macleod remembered that his views on returning to Toronto were about the same as his reaction to the written report. He wanted more work done, and he specifically suggested an experiment to eliminate the possibility that dilution of the blood by the injections was causing the drops in blood sugar.46

  The more memorable part of the interview came when Banting, probably after relating their problems with working conditions during the summer, demanded four things from Macleod: a salary, a room to work in, a boy to look after the dogs, and repairs to the floor of the operating room. Macleod was reluctant; Banting and Best had already gone through more dogs than planned. What was the point of spending
money to fix up an operating room about to be abandoned when the new building opened? The professor told Banting and Best “that if he gave us these things some other research would suffer.”47

  Eleven months later Banting wrote this account of his reaction to Macleod’s hesitation:

  I told him that if the University of Toronto did not think that the results obtained were of sufficient importance to warrant the provision of the aforementioned requirements I would have to go some place where they would.

  His reply was, “As far as you are concerned, I am the University of Toronto.”

  He told me that this research was “no more important than any other research in the department.”

  I told him that I had given up everything I had in the world to do the research, and that I was going to do it, and that if he did not provide what I asked I would go some place where they would.

  He said that I “had better go.”

  Banting expanded on the argument in his 1940 memoir:

  “Alright I am telling you that unless you provide the necessary facilities within twenty-four hours, then I shall leave.” Banting walked to the door.

  “And where will you go?”

  “I don’t think it matters a damn to you but I might go to the Mayos.”

  “Only an advertising institution.”

  “Or I might go to the Rockefeller Institute.”

  “They have finished with diabetes research. Allen has been forced to leave.”

  “The Rockefeller is never finished with research and you know it. And I will be in at this hour tomorrow morning and will continue work if the requirements are met in full. Otherwise I leave.”

  Macleod apparently relented, thought the matter over, and promised to do what he could.48 The interview ended.

  Best had taken it all in, apparently without saying anything. “I have never heard anyone talk to Macleod as you have,” he said to Banting afterwards. The more Banting thought about the interview, particularly Macleod’s remark, “As far as you are concerned, I am the University of Toronto,” the angrier he became. To Best he poured out his opinion of Macleod.

  “Fred Banting began to froth at the mouth,” Best remembered. “My recollection of what Banting said to me when he became articulate was, ‘I’ll show that little son of a bitch that he is not the University of Toronto.’“49

  CHAPTER FOUR

  “A Mysterious Something”

  “We have obtained from the pancreas of animals a mysterious something which when injected into totally diabetic dogs completely removes all the cardinal symptoms of the disease….If the substance works on the human, it will be a great boon to Medicine.” J.B. Collip, January 8, 1922.

  Within a day or two of the confrontation in Macleod’s office, all the details had been settled to enable Banting and Best’s work to continue. Macleod found a room, big enough for two dog cages and a laboratory desk, that Banting considered quite acceptable. He gave them a part-time lab boy and had the physiology operating-room floor tarred so it could be cleaned properly.1 Velyien Henderson’s opening for a special assistant in Pharmacology solved Banting’s employment problem. From October 1, 1921, Banting was on the University of Toronto payroll as a special lecturer in Pharmacology at a salary of $250 a month: in terms of purchasing power – we should multiply by a factor of between ten and fifteen for today’s prices – it was a good salary. In view of the “decidedly satisfactory” results they had achieved during the summer, Macleod also arranged for retroactive pay for Banting and Best: $150 for Banting (the new boy), $170 for Best.2

  Banting’s duties in Pharmacology over the winter would be light. Best was staying on as an M.A. student and demonstrator in Physiology. Macleod was cordial and helpful, Banting remembered. “I thought that perhaps I had judged him too harshly.” It was a great relief to Banting to have financial support over the winter. His landlady shared his pleasure. He gave her a liquor prescription so she could celebrate.3

  I

  The experiments could go forward again. But in what direction and in whose hands? Banting, we saw, had a long list of directions in which he thought the work should go; the list was so long that Macleod must have thought that Banting, like Leacock’s horseman, was trying to ride madly off in all directions. Excited students have this proclivity. Less convinced than Banting that the summer’s experiments were definite proof of the isolation of the internal secretion, Macleod advised sticking to the problem at hand so it could be wrapped up to everyone’s satisfaction.

  It may have been as early as the beginning of October when Banting first suggested to Macleod that either Macleod or others take part in the work. One of the “others” was almost certainly J.B. Collip, now spending part of his sabbatical year working in the Pathological Chemistry department at Toronto. Collip knew about the work, was interested in it, and told Banting several times that autumn how delighted he would be to help. But Macleod advised against expanding the team at this stage. “I pointed out that this being his and Best’s research they should independently complete the work as outlined, and that then if the results continued satisfactory I would participate in the further investigations with my assistants.” So Banting and Best went back to their dogs.4

  Actually they first made a detour out to the farm on the outskirts of Toronto that Colonel Albert Gooderham, a local whisky magnate, had given the university to house its Connaught Anti-Toxin Laboratories. These had been founded in 1914 by a professor of hygiene, J.G. Fitzgerald, to produce vaccines and anti-toxins. On October 4 at the Connaught farm Banting and Best tried to ligate the pancreatic ducts of a calf. It died from the anesthetic.5 Back in the lab, the pancreatic ducts of several dogs were ligated, another dog was totally depancreatized, and extract was made from the pancreas of the one dog, “Towser,” whose ducts had been ligated 4½ weeks earlier at the beginning of September.6 Towser’s pancreas was only partly degenerated, so Banting and Best made separate extracts from the degenerated and non-degenerated parts of the pancreas.

  The experiments on the depancreatized dog, number 17, a long-haired spotted hunter, were designed to answer several of the questions Macleod had raised about the summer work. To control for diurnal variation the injections were made at the same time every day. Tests after an injection of one hundred cc. of saline solution showed that the extra liquid by itself had no blood sugar reducing effect. Hemoglobin estimations, made before and after injection of the extract, seemed to show no appreciable thinning of the blood. So a dilution phenomenon could be ruled out as an explanation of the results. That was very satisfactory.

  The extract labelled “Towser B” seemed very potent in enabling the diabetic dog to utilize the sugar injected along with it. Unfortunately a control injection of sugar alone had the mysterious effect of not raising the dog’s blood sugar. It was repeated with the same mysterious effect. Not until the third sugar injection did the dog’s blood sugar respond properly. Furthermore, Banting and Best’s notebooks show that extract B was made from the less degenerated part (the tail) of Towser’s pancreas. Injections of extract A, from the most degenerated part (the uncinate process) were much weaker, hardly effective at all. If this experiment showed anything, like the one with whole gland pancreas on August 17 it cast doubt on the hypothesis that a degenerated gland was necessary to produce potent extract. But again, there is no evidence that Banting and Best noticed the problem, for they had managed to get extracts A and B confused.7 They were probably too puzzled by the strange non-effect of simple sugar injections to check the other aspects of the experiment carefully.

  II

  Now that Macleod’s objections had been met, more or less, what remained to be done? Very little was done through the middle of October – partly because it seemed necessary to wait several weeks before more extract could be produced from the duct-ligated dogs, and partly because Banting seems to have been uncertain about what he wanted to do. He was giving a lot of thought to various possible experiments, writing down his ideas on four-by-s
ix-inch index cards as they occurred to him (unfortunately only now and then jotting down the date on his cards). About the same time, it seems, he and Best were studying the literature more or less systematically, looking for ideas, perhaps also gathering background material as they planned the articles they would write on their work.

  Macleod gave them some references. His young secretary, Maynard Grange, doubled as the medical librarian. She was in her mid-eighties and almost blind sixty years later when she told me her vivid memory of Banting coming in one day to look up a book Macleod had recommended. “You know,” he muttered to her, “the goddamned little bugger knows everything about this subject.” He said it, she remembered, in a tone of grudging admiration.

  Banting’s index card notes suggest that he had no clear idea where the experiments should go. He was toying with ideas for more test-tube experiments, considering whether injections of pancreatic juice could be used to prove its effectiveness against the internal secretion, noting other methods than pancreatectomy of producing glycosuria, wondering how the extract’s effect on the action of the liver in metabolism could be tested. A typical Banting “idea” card, dated October 4, 1921, reads