The Discovery of Insulin

by Michael Bliss

  Chart 1: Dog 410, July 30

  On Sunday morning they found the terrier in a coma. They took one blood sugar (.15) before it died. No autopsy was done on the dog, the first to receive the extract.* In their first published paper, Banting and Best mentioned that it had appeared to be “entering the cachexial condition characteristic of depancreatized animals which had become infected” when they decided to give it the extract. They also noted that the dog’s blood sugar had not risen “to the level usually attained in completely depancreatized animals.” Even so, they were encouraged by their result: “The extract seemed to have a marked effect,” Best wrote to Macleod a few days afterwards.16

  On Monday, August 1, they tried again. There was extract at hand from a second duct-tied dog. The only depancreatized dog, the collie 406, was in coma, on the brink of death.17 Its blood sugar was very high, .50 at 12:10 p.m. Banting and Best injected eight cc. of their extract intravenously. At 1:10 the blood sugar was .42. “Dog able to stand & walk.” An additional five cc. of extract brought the blood sugar down to .30 by 2:10, but the dog had lapsed back into coma. It died at 3:30 p.m. The experiment would not look good on paper – two injections on a dying dog, once again no autopsy – and was never written up. To Banting and Best it must have been impressive to see a dog come out of coma, stand, and walk, after an injection of their extract.

  Chart 2: Dog 406, Aug. 1

  They still had more extract, but had run out of depancreatized dogs to try it on. It would take at least a week, perhaps much longer if the same failure rate continued, to get more dogs ready via the two-stage pancreatectomy. Why not speed up the procedure, Best urged, by doing the whole pancreatectomy at once?18 Banting agreed to give it a try. On Wednesday, August 3, they did a total pancreatectomy, their first, on a young yellow collie, dog 408. “Operation was easiest yet,” Banting wrote in his notebook, obviously surprised at how well it had gone. He never again bothered with the time-consuming Hédon procedure.

  The next day the collie became the third dog to receive the extract of degenerated pancreas. Five cc. of extract caused its blood sugar to fall from .26 to .16 in 35 minutes early in the afternoon. That evening, another five cc. of extract reduced it from .25 to .18 in half an hour. The dog’s general condition, Banting and Best noted, was good. They were still collecting and testing urine, with moderately favourable though slightly puzzling results. The volume of the dog’s urine decreased after injections and urination finally ceased altogether.19 Little attention was being paid to the D:N ratio (which did, however, average around 3.0 for dog 408). More and more as time went by, attention centred on blood sugar readings as the key to the experiments. In their notes on this experiment Banting and Best name their extract for the first time: “Isletin.”

  Now that they had a live dog to work on, Banting and Best could attempt more varied experiments. On Friday, August 5, they made extracts from the liver and spleen of dogs, prepared them exactly the same way as the extracts of pancreas, and injected them into the collie. Neither caused any significant change in the blood sugar. Injections of “Isletin” later in the day gradually drove down the blood sugar from .30 to .17, and the dog continued to look healthy. “General condition of dog seems much improved… appears more interested in happenings – more susceptible to pain.” On Saturday morning they tried boiled extract on the dog. No effect. At midnight, with the blood sugar at .43, they began giving hourly injections of eight cc. of extract.20 The blood sugar slowly dropped – to .37, .33, .29, and, by 4:00 a.m., .20. At that hour they decided to shoot the works, mixing all the rest of their extracts from the two dogs together, diluting it a bit, and giving the collie a massive injection of twenty-five cc. Result: “Dog suddenly became lifeless and appeared to be dying.”

  Chart 3: Dog 408

  Banting and Best thought the injection had caused an anaphylactic shock (similar to an allergic reaction). It might also have been a thrombosis. They tried to revive the dog by injecting large doses of Ringer’s solution. It improved slightly that morning but died at noon. The autopsy showed widespread infection, stemming from the operation, which was considered the cause of death. This was the end of the first series of experiments.

  VII

  Banting and Best were excited by what they had seen. “We got fine results,” Best wrote to Margaret on August 8.21 “I have so much to tell you,” Banting wrote Macleod the next day, “that I scarcely know where to begin.” The extract “invariably” causes a reduction in blood sugar, he wrote; it improves the clinical condition of the dog; it can be kept active for at least four days, is destroyed by boiling, and extracts of other organs are inactive. But so many new questions came to mind. Banting had jotted them down in his notebook and on index cards, and now listed sixteen of them for Macleod. How could they get the most active form of “Isletin”? What were its chemical properties? Was it actually destroyed by the digestive enzyme, trypsin? Could a diabetic dog be kept alive on the extract? Was it universal in the animal kingdom and in its action? “The whole problem of tissue grafting.” The extract’s relation to the various forms of diabetes. Its mechanism of action. And other questions. Fifteenth on Banting’s list for Macleod-it had been the sixth question to occur to him in his notebook list – was “its clinical application.”22

  “I am very anxious that I be allowed to work in your laboratory,” Banting wrote. But he needed more help looking after the animals and needed better operating facilities. He had told Dr. Starr of their problems, and Starr had procured the use of the surgical research operating room for future experiments. They had only two dogs left with ducts tied. Banting wrote Macleod:

  I would like to do about ten so as to have a supply of the extract for you when you return. I have told Dr. Starr all about my results and he advised me to go ahead so I will proceed slowly and if I do not hear from you I will take it that I have your permission. Please let me know as soon as possible your wishes. I will not proceed immediately however as I am going to London Ont. to close up my affairs and have them off my mind, and that will give this letter time to get to you.

  Whatever friction there had been between Banting and Best early in the summer had disappeared. Banting reported glowingly to Macleod: “Mr. Best has expressed the desire to work with me and I should be more than pleased to have him. His work has been excellent and he is absolutely honest, careful and impartial, and has taken a great interest in the work. He has assisted me in all the operations and taught me the chemistry so that we work together all the time & check up each other’s readings.”23

  Even without the early frustrations and failures of the research, it had been a difficult summer for Banting. He had very little money – no more than one or two hundred dollars, he later estimated – no salary, and apparently no prospect of borrowing more from his parents. He earned a few dollars in May and June doing tonsillectomies for one of his friends and later sold some instruments for $25. For part of the summer he lived with his cousin and classmate, Fred Hipwell, minding the Hipwell house while Fred and Lillian and their baby spent a few weeks at a cottage. Then he moved back into the boarding house on Grenville Street where he had lived as a student, paying $2 a week for a little seven-by-nine-foot cubicle. He ate in restaurants, with friends when they invited him, sometimes at the Sunday night suppers of the Philathea Bible Class of St. James Square Presbyterian Church, which he had attended as a student, and sometimes in the lab.24

  For Charley Best the summer of 1921 was a delightful round of tennis and swimming, golf and baseball, outings with Margaret, and interesting though sometimes frustrating work at the lab. Fred occasionally came on outings with them, dating one of the secretaries in the medical building, but spent much of his time worrying about Edith and his future. Relations with Edith were endlessly complicated – friends read the state of their romance by whether or not the diamond engagement ring was dangling from Fred’s watch chain – and his future totally obscure. No one had been very enthusiastic about his research project, he knew,
and even Macleod had not expected very much from it. “Worst of all, no one took me seriously,” he wrote shortly afterwards. He seems to have taken his own work seriously, hoping from the beginning that he was heading not just for a contribution to physiological knowledge, but for a treatment of diabetes. One day in the summer of 1921 while he was driving Lillian Hipwell and her baby to the beach, he said to her, “Lillian, if what I am working on is a success, I will be a famous man, but then I don’t think it will happen.”25

  Now, as his report to Macleod indicated, Banting was encouraged enough by his results to wind up his affairs in London. It would be weeks before Macleod’s reaction to their report would arrive, Banting was not being paid for the work, and there is no evidence that anyone had given him any assurance of a paying job of any kind at the university. But he was so sure now that his future lay in Toronto that he was ready to burn his last bridge in London before even knowing how Macleod felt about the work.

  VIII

  First they decided to go ahead on their own with one more round of experiments, a big push in which they would work around the clock for several days. “It will be quite a crucial test for our Isletin,” Charley wrote Margaret. The plan was to do total pancreatectomies on two dogs, give “Isletin” to one, and compare its health with the other. The dogs, numbers 92 and 409, were totally depancreatized on August 11. About the same time Banting and Best ligated the pancreatic ducts of two cats, and apparently two rabbits; they were getting ready, it seems, for later experiments to see if the extract worked across species.26

  Dog 92 was a yellow collie. Banting had a difficult time getting its pancreas out. There was a lot of bleeding and he had to ligate major blood vessels normally not involved in the operation. The dog was fortunate to survive, and was still in distress when they began giving it pancreatic extract late that day. Nor did the first injections, at four-hour intervals, seem to have much effect,* except that dog 92’s blood sugar, hovering around the .20 level, was staying well below that of the untreated dog, 409. Large doses of remacerated and diluted extract seemed to give excellent results the next day, driving 92’s blood sugar down from a diabetic .30 to a low normal .09 in a twelve-hour period.

  At 10 a.m. on August 13 the control dog was barely able to walk. Dog 92, on the other hand, was in “excellent condition does not appear tired or sleepy walks about as before operation.” A new batch of extract continued to hold down the blood sugar, and while dog 409 continued to get worse, 92 was “feeling fine – rather ‘hounded’ aspect… but bears herself like a normal dog….Feeling great… runs around room, frisky.” On the morning of August 14 they decided to see if an overdose of the extract would reduce dog 92’s blood sugar below normal. It appeared to, thirty cc. in seventy minutes bringing it from .22 to .066. At 1:30 a.m. on August 15 the untreated dog, 409, died, apparently of diabetes. Dog 92, which had become something of a laboratory pet, carried on.

  It was an unforgettable time in the lives of the young scientists – working day and night in the lab, snatching a few hours’ sleep here and there, frying eggs and heating up steaks over a bunsen burner, Fred’s baritone blending with Charley’s tenor as they sang war songs while they worked, sitting in the window sills or on the front steps of the meds building in the cool early morning sunshine of a beautiful August day, sitting at an all-night restaurant in the small hours of the morning discussing Isletin’s future and their own. After all their troubles the work was coming out so beautifully. Everything they tried seemed to work perfectly, so much better than Prof Macleod, who didn’t even know they’d gone this far, would ever have expected. This was research. Forget all the bad times gone before. These, for Banting and Best, were the truly memorable days in their search for the internal secretion.27

  Why not try more experiments on the frisky collie, dog 92? How effective would an extract with a slightly acidic balance be? Or one made more alkaline? The acid extract seemed effective, the alkaline one did not. Extract incubated with trypsin (an active ingredient of the external secretion) was curiously potent. Soon there were no more duct-tied dogs whose degenerated pancreases could be chopped up to make “Isletin.” But why quit now, with dog 92 running around the lab friskier than ever? It would come when it was called, lie perfectly still while a blood sample was being taken, then snuggle up to Banting, its head in his lap, while he did the chemistry and talked to it.28 On August 17 Banting and Best chloroformed a dog and made some extracts of its whole pancreas. Dog 92’s blood sugar was .30 at 6 p.m. when it was given ten cc. of whole gland extract. At 6:30 it had fallen to. 19 and by 7:00 was down to .17. At 7:30 and 8:30 it had risen, but only to .20. Perhaps Banting and Best were too tired to think clearly.

  Chart 4: Dog 92, whole gland extract

  Perhaps they were too confident of what to expect to notice contrary evidence. Here was fresh whole gland extract markedly reducing blood sugar. It was having as good an effect, perhaps better, than their extract of degenerated pancreas usually had. Banting and Best thought nothing of it. “It is obvious from the chart that the whole gland extract is much weaker than that from the degenerated gland,” they wrote in their first published paper.29 This was not obvious from the chart or the figures; indeed it was not true, and in failing to see this and its implications Banting and Best made a major error. Ignoring clear evidence to the contrary, they continued their pursuit of what we will later see was a faulty hypothesis.30

  Banting and Best still thought it was necessary to do something to a pancreas to get rid of the external secretion. On the 19th, with no duct-ligated dogs at hand, and the collie now starting to weaken and sicken, Banting hit on another approach to avoiding the external secretion. It was well known that the hormone secretin, formed in the mucous membrane of the duodenum, stimulated the pancreas to produce its external secretion. Why not stimulate a pancreas with secretin until it was exhausted and could produce no more external secretion? Then, containing only the internal secretion, it could be cut out, ground up, and should produce an extract just as potent as those made from duct-ligated pancreases which had taken weeks to degenerate.

  It was a complicated procedure, involving a resection of a dog’s bowel to obtain the crude secretin, the insertion of a cannula in the larger pancreatic duct to measure the flow of external secretion, the slow injection of secretin for almost four hours until the production of pancreatic juice stopped, and finally the preparation of the extract.31 But the extract worked marvellously. The collie, 92, had been very sick, unable to get to its feet, suffering from an abscess they had to lance on one of its forelegs. It began getting extract on a Saturday night. By Sunday morning (after another all-night session), the dog was running around the lab again, wagging its tail to welcome the humans, generally in excellent spirits. At two that afternoon, “dog jumped out of cage to floor about 2½ ft lit on fore feet & did not fall,” Banting recorded. He never forgot that moment, singling it out for emphasis in accounts of these greatest days of his life.32

  There was enough extract to try still more experiments. Would simple test-tube experiments show that the extract could “burn” sugar? An earlier experiment on August 6 had given promising results before the extract ran out. It was repeated more elaborately on the 20th, again with apparently good results.* An injection of extract plus trypsin produced no effect on the collie’s blood sugar, apparently cancelling out the surprisingly strong result of the similar injection a few days earlier. There was enough extract left for one more injection. It worked well.

  On August 22, Banting and Best tried to exhaust the pancreas of a cat by injecting secretin, but the animal died on the operating table. This was one of those times when Fred might have said to Charley, “What the hell?…” They took out the cat’s pancreas anyway, made their extract, and tried it on the collie. It threw the dog into profound shock. That was the end of the experiments. Dog 92 lingered for another nine days, its blood sugar gradually rising, its overall condition weakening. It died on the 31st. “I have seen patients die a
nd I have never shed a tear,” Banting wrote in one of the more maudlin passages in his 1940 memoir, “but when that dog died I wanted to be alone for the tears would fall despite anything I could do…I hid my face from Best.”33

  IX

  About the first week in September, Banting and one of his brothers drove to London to settle his affairs there. In one day, Banting recalled, he sold his house and most of his furniture. The rest he packed into the car he had and into his brother’s car.

  That night we slept on the bare floor of the empty house and very early in the morning we took a last look at the place of my hours of misery – and yet it was there that I obtained the idea that was to alter every plan that I had ever made, the idea which was to change my future and possibly the future of others.

  One mellows with the years, but I still find it impossible to forget the awfulness, the loneliness and the financial worries that were associated with London. Nor can I forget the feeling of defeat that came over me as I took my final leave on that foggy autumn morning….My car, a little open Ford, would not go fast enough on that morning. I left my brother miles behind. It was a relief to be away and free.34

  X

  They were just beginning to work again on September 6 when Macleod’s response to their report arrived. In general he was pleased: the results (to August 8) seemed “certainly very encouraging…definitely positive.” He was glad to see that Banting planned to stay on in Toronto, “and you may rest assured that I will do all in my power to help you.” In the new anatomy building just being built, Macleod would have new operating rooms. In the meantime Banting could use the facilities Starr had offered, “since I do not wish to spend money at present on the old rooms.” He advised Banting to be careful, though, that no one should see him transporting operated animals from one building to another on campus: Macleod was very worried about anti-vivisectionist criticism of animal research, then reaching a peak in Toronto.