Penicillin: The Accidental Spore That Changed Medicine

Penicillin was discovered by accident in September 1928, when Alexander Fleming noticed that a mold growing on one of his culture plates had destroyed the bacteria around it. But the accident only started the story. It took a run of lucky timing to produce that plate, a team at Oxford to turn Fleming's "mould juice" into a drug, and an American agricultural laboratory to make enough of it for the Second World War.

What did Fleming see on his culture plate?

In 1928 Fleming, a Scottish bacteriologist, worked in a famously untidy laboratory on the second floor of St Mary's Hospital in Paddington, London. Half a century after germ theory had pinned infections on microbes, he was studying variants of Staphylococcus, the bacterium behind boils and many wound infections, and he left culture plates on his bench to check later. In September, after a summer break spent mostly at his country home in Barton Mills, Suffolk, he was back in the laboratory sorting through those plates and stacking used ones in a tray of Lysol disinfectant. One plate had escaped the disinfectant. A large mold colony had grown on it, and around the mold the staphylococcus colonies had gone see-through. In the paper he published the next year, Fleming wrote that they "became transparent and were obviously undergoing lysis": the bacteria were breaking apart. His former assistant Merlin Pryce, who had dropped in that day, recalled years later that Fleming studied the plate for a while and said, "That's funny." Fleming grew the mold in broth and found that the filtered liquid stopped many bacteria. Writing up the work in 1929 for the British Journal of Experimental Pathology, he called the active substance penicillin, partly to avoid repeating the clumsy phrase "mould broth filtrate." It hit the pus-forming cocci behind many serious infections and the diphtheria bacillus hard, left others such as the typhoid group untouched, and was harmless to animals even in large doses. He thought it might be applied to infected areas, but in practice he used it mainly as a lab tool, to suppress unwanted bacteria when culturing the influenza bacillus, which it did not affect.

Why did a stray mold get the chance to win?

Fleming did not grow molds. The likeliest source of the spore was the room directly below, where the Irish mycologist C. J. La Touche grew molds from the homes of asthma patients to make desensitizing extracts. Ronald Hare, a colleague who later reconstructed the discovery, thought spores drifted up the stairs. Of the molds La Touche later gave Fleming, only one made penicillin, and it had, in Fleming's words, "exactly the same cultural characters" as the contaminant. The popular tale of a spore blowing in through an open window dates only from Fleming's own retelling in 1945; Hare noted that the windows were hard to reach and rarely opened.

The timing also had to be right. Penicillin works only on bacteria that are growing and dividing, so it cannot dissolve colonies that are already established; the mold needed a head start. In later experiments, Hare could reproduce Fleming's plate only if the mold landed around the time the bacteria were seeded and the plate then stayed below about 20°C for some five days before warming up. London's temperature records for July to September 1928 contained just one stretch like that: nine cool days around the end of July and the start of August, followed by warmer weather. Hare concluded that the plate was seeded and contaminated on 30 or 31 July. It is a reconstruction, not a record. The mold itself was a lucky draw too. Fleming noted that this sort of Penicillium was "not uncommon in the air of the laboratory," yet his strain turned out to be an unusually strong producer. Long known as Penicillium notatum, it was reassigned to Penicillium rubens by genetic analysis in 2011.

Why did it take Oxford to turn it into a medicine?

Penicillin was fragile. Fleming's assistants Stuart Craddock and Frederick Ridley tried to concentrate it by evaporating the broth, but the substance was easily destroyed, the method was impractical, and the chemical work stopped in April 1929. Fleming kept the mold alive and sent it to laboratories that asked for it, but as he said at the Nobel banquet in 1945, he "had not the team, especially the chemical team" to stabilize it. Its structure, built around a strained four-atom ring now called the beta-lactam ring, was only worked out in 1945; how that ring kills bacteria is a story of its own. The breakthrough came in the late 1930s at Oxford's Sir William Dunn School of Pathology, where the Australian pathologist Howard Florey and Ernst Chain, a Berlin-born Jewish biochemist who had left Germany in 1933, took penicillin up again with the biochemist Norman Heatley. They grew the mold in whatever they could find, from baths and bedpans to milk churns and food tins, and later in hundreds of purpose-made ceramic vessels. Heatley worked out how to measure penicillin's strength and how to pull it out of the broth into a solvent and back into water, leaving a stable salt that could be freeze-dried. It was still mostly impurities, but it worked.

On 25 May 1940 the team injected eight mice with lethal streptococci and treated four of them. Heatley sat up until 3:45 the next morning, by which time all four untreated mice were dead; the treated ones were alive, and some lived for weeks. The Lancet published the results that August. After a safety test in January 1941 on Elva Akers, a woman dying of cancer, the team treated its first infected patient: Albert Alexander, a police constable whose face was badly infected, traditionally from a rose-thorn scratch, though some accounts blame an air-raid injury. Within a day of his first dose, on 12 February 1941, he was much better. The team even extracted the drug from his urine to reuse it, but the supply ran out, and Alexander died on 15 March.

How did American labs make enough for the war?

Britain's chemical industry was absorbed by the war, so in July 1941 Florey and Heatley flew to the United States. On 14 July they reached the Department of Agriculture's Northern Regional Research Laboratory in Peoria, Illinois. There the microbiologist Andrew Moyer swapped the sucrose in the growth medium for lactose and added corn steep liquor, a cheap byproduct of corn processing, raising yields many times over. The laboratory also hunted for better molds. In 1943 a moldy cantaloupe from a Peoria market yielded a strain, NRRL 1951, that made more than twice as much penicillin as the best cultures of Fleming's mold, and its X-ray and ultraviolet mutants made several times more again. Popular accounts credit the laboratory technician Mary Hunt, nicknamed "Moldy Mary," who collected specimens from markets, but the USDA says it is uncertain who actually found the melon. Drug companies scaled up in huge tanks: Pfizer's Brooklyn plant, opened in March 1944 with fourteen 7,500-gallon fermenters, supplied most of the penicillin that went ashore on D-Day, when 2.3 million doses were ready. American factories made 21 billion units in 1943 and more than 6.8 trillion in 1945, and the price of a dose fell from $20 in July 1943 to 55 cents three years later.

What did Fleming warn the world about?

Fleming, Florey and Chain split the 1945 Nobel Prize for medicine. At the banquet in Stockholm on 10 December, Fleming credited fate: "It was destiny which contaminated my culture plate in 1928." In his Nobel lecture the next day he added a warning that now reads as prophecy:

"The time may come when penicillin can be bought by anyone in the shops. Then there is the danger that the ignorant man may easily underdose himself and by exposing his microbes to non-lethal quantities of the drug make them resistant." He went on to imagine a "Mr. X" who treats his sore throat with too little penicillin, breeds resistant streptococci and passes them to his wife, who dies of a pneumonia that penicillin can no longer cure. Resistance itself was not news: bacteria had already been made resistant in the laboratory, and in 1940 Edward Abraham and Ernst Chain had described a bacterial enzyme that destroys penicillin. What Fleming foresaw was how careless use would spread it. Today, resistant bacteria often carry genes for such enzymes and pass them to one another, and a study in The Lancet estimated that bacterial antimicrobial resistance directly caused 1.27 million deaths in 2019. Penicillin's discovery is often told as a single lucky glance at a petri dish. The glance mattered, but so did a mold lab on the floor below, a cool spell in a London summer, Oxford chemistry and an American cantaloupe. Like Wilhelm Röntgen's X-rays, it was an accident that paid off because someone was ready to notice it, and others were ready to finish the job.