Penicillin

• 1. Introduction

  ›

  Penicillin's story begins with an accident that almost went unnoticed. On the 3rd of September 1928, the Scottish physician Alexander Fleming noticed that a mould had contaminated a dish where he had been growing Staphylococcus aureus. The bacteria nearest the mould had died. By his own account, he had not been looking for anything and stumbled onto the effect by chance. For years afterward, almost no one thought the discovery mattered. Turning that fungal accident into a usable medicine required a different team altogether, working an ocean away and racing against a world war. It also required working out, at the molecular level, exactly how this small compound tears apart a bacterial cell. That understanding later explained why some bacteria found ways to survive it. This is the story of how a discarded mould extract became one of the defining medicines of the twentieth century. It is also the story of why the microbes it targets have spent decades finding ways around it.

• 2. The Reluctant Discovery

  ›

  Fleming confirmed his initial finding with a fresh experiment on the 28th of September 1928, then published the results the following year. He coined the actual word on the 7th of March 1929, while investigating the antibacterial power of Penicillium rubens. In that 1929 paper for the British Journal of Experimental Pathology, he explained his reasoning directly: to avoid repeating the phrase "Mould broth filtrate," he wrote, "the name 'penicillin' will be used."

  Sixteen years later, in his 1945 Nobel lecture, Fleming addressed the question people kept asking him: "I have been frequently asked why I invented the name 'Penicillin'." He explained that he had simply followed convention: "I simply followed perfectly orthodox lines and coined a word which explained that the substance penicillin was derived from a plant of the genus Penicillium just as many years ago the word 'Digitalin' was invented for a substance derived from the plant Digitalis."

  The fungus itself went through several names before settling. C. J. La Touche first identified it as Penicillium rubrum. The mycologist Charles Thom later reclassified it as P. notatum and then P. chrysogenum, and the classification was finally corrected to P. rubens.

  Fleming was initially hopeful about the extract, noting its high potency and low toxicity compared with the antiseptics of the day. He also found a practical laboratory use for it, employing penicillin to help isolate Bacillus influenzae, now known as Haemophilus influenzae. He could not, however, convince other scientists that the discovery mattered. This was largely because the substance proved so difficult to isolate that turning it into a real drug seemed impossible. Some historians have suggested that a more persuasive Fleming might have pushed penicillin's development years ahead of schedule.

  In modern usage, the word covers more ground than Fleming's coinage did. Any beta-lactam antimicrobial built around a thiazolidine ring fused to the beta-lactam core now counts as a penicillin, whether it occurs naturally or not. Like many natural products, penicillin exists inside Penicillium moulds as a mixture of related active compounds, much as the antibiotic gentamicin exists as a similarly ill-defined mixture. The importance of Fleming's original discovery was formally recognised on the 19th of November 1999, when an International Historic Chemical Landmark was placed at the Alexander Fleming Laboratory Museum in London.

• 3. From Fungus To First Cures

  ›

  On the 25th of November 1930, Cecil George Paine, a pathologist at the Royal Infirmary in Sheffield who had trained under Fleming, used a crude fungal extract to treat ophthalmia neonatorum, a gonococcal eye infection in newborns. It was the first successful medical use of penicillin on a patient.

  A decade later, in 1940, the Australian-born scientist Howard Florey, later Baron Florey, led a team at Oxford's Sir William Dunn School of Pathology that finally concentrated penicillin from fungal broth into a usable form. His collaborators included Ernst Chain, Edward Abraham, Arthur Duncan Gardner, Norman Heatley, Margaret Jennings, Jean Orr-Ewing and Arthur Gordon Sanders. Together they proved the concentrate could kill bacteria both in laboratory dishes and inside living animals.

  In 1941 the team treated a policeman named Albert Alexander, who had developed a severe facial infection. His condition improved briefly, but the team's limited supply of penicillin ran out, and he died. Several other patients treated afterward fared better and recovered fully. In December 1942, survivors of the Cocoanut Grove fire in Boston became the first burn patients ever treated successfully with penicillin.

  The first successful treatment with pure penicillin came in 1942, when Fleming himself used it on a patient named Harry Lambert, who had streptococcal meningitis, an infection that would otherwise have killed him. At the time the Oxford team could produce only a tiny quantity, and Florey gave Fleming the only sample available. Lambert improved by the very next day and was completely cured within a week. Fleming published the case in The Lancet in 1943.

  The breakthrough pushed Britain's War Cabinet to create the Penicillin Committee on the 5th of April 1943, aiming to organise large-scale production. Two years later, Chain, Fleming and Florey shared the 1945 Nobel Prize in Physiology or Medicine for the discovery and development of the drug. Oxford's own laboratory, however, could never produce penicillin beyond a trickle, a limitation that would soon force the story onto a different continent.

• 4. An Ocean-Sized Supply Problem

  ›

  In June 1941, after failing to persuade the British government to fund large-scale manufacturing, Florey and Norman Heatley carried their mould samples to the United States. They approached the Northern Regional Research Laboratory, now called the National Center for Agricultural Utilization Research, in Peoria, Illinois, where large-scale fermentation facilities already existed. Researchers there began mass-culturing the mould while searching for a more productive strain.

  On the 14th of March 1942, doctors treated the first patient for streptococcal sepsis using US-made penicillin, produced by Merck & Co. Half of the country's entire supply at that moment went into treating just one patient, Anne Miller. By June 1942, there was still only enough penicillin available to treat ten patients total. In July 1943, the War Production Board drew up a plan to distribute penicillin stocks to Allied troops fighting across Europe. Fermentation research using corn steep liquor at the Peoria laboratory let the United States manufacture 2.3 million doses in time for the Normandy invasion in the spring of 1944.

  After a worldwide search in 1943, researchers found the single best production strain growing on a mouldy cantaloupe sold in a Peoria market. Using the corn steep liquor process, that strain yielded six times more penicillin than Fleming's original mould had. Jasper H. Kane, a scientist at Pfizer, proposed deep-tank fermentation as the method for manufacturing pharmaceutical-grade penicillin at real scale. The chemical engineer Margaret Hutchinson Rousseau then designed the actual deep-tank fermentation plant that made mass production possible. G. Raymond Rettew contributed too, combining his expertise in mushroom spawn with the Sharples Cream Separator. By 1943, his laboratory alone was producing most of the world's penicillin.

  By June 1945, wartime production had reached over 646 billion units of penicillin per year in the United States. During the Second World War, the drug is estimated to have saved 12-15 percent of Allied lives and limbs otherwise lost to infected wounds. Supplies still ran short, though, both because manufacturing remained difficult and because the kidneys cleared penicillin from the bloodstream so fast that patients needed frequent redosing. Andrew Jackson Moyer patented methods for mass-producing penicillin in 1945. Florey, by contrast, never patented the drug at all, having been advised by Sir Henry Dale that doing so would be unethical.

  About 80 percent of any dose of penicillin leaves the body within three to four hours, excreted by the kidneys. During the drug's scarcest years, doctors sometimes collected patients' urine so the excreted penicillin could be isolated and reused on someone else. Researchers eventually found a better fix in probenecid, a drug that competes with penicillin for the same kidney transporter and so slows the antibiotic's excretion, prolonging its effect in the blood. Once mass production and semisynthetic penicillins solved the supply shortage, this use of probenecid faded, though it is still prescribed today for infections needing especially high penicillin concentrations. After the war, Australia became the first country to make penicillin available to civilians, with the United States following on the 15th of March 1945. By the time supply finally caught up with demand, chemists were already looking past the natural mould toward penicillin's stripped-down chemical core, 6-aminopenicillanic acid.

• 5. A Family Of Molecules

  ›

  6-Aminopenicillanic acid, or 6-APA, is what remains of penicillin G once its side chains are stripped away, leaving only the bare beta-lactam core. That stripped-down molecule became the essential building block for manufacturing entirely new penicillins. Penicillium mould itself naturally produces several related compounds under old British and newer American names. Penicillin I is known in the US as penicillin F. Penicillin II, or benzylpenicillin, is better known simply as penicillin G. Penicillin III is also called penicillin X, and penicillin IV is also called penicillin K. Measured against penicillin G's potency, penicillin F reaches only 70-82 percent. Penicillin X reaches 130-140 percent, and penicillin K reaches 110-120 percent. Smaller amounts of penicillin O, U1 and U6 also occur naturally, alongside a compound called penicillin A that turned out to have no antibiotic activity at all.

  Fleming's original strain of Penicillium rubens produces mostly penicillin F, the version later named for him. That compound is unstable, hard to isolate, and made only in tiny quantities. The commercial Peoria strain of Penicillium chrysogenum instead produces penicillin G as its main output when grown on corn steep liquor. Adding phenoxyethanol or phenoxyacetic acid to that same culture switches the mould's output to penicillin V instead.

  Penicillin G cannot be taken by mouth because stomach acid destroys it. Doctors instead give it by intravenous or intramuscular injection, in doses as high as 2.4 grams. Because the kidneys clear it from the bloodstream quickly, maintaining a high blood concentration means dosing it at frequent intervals. It is licensed to treat serious infections including septicaemia, pneumonia, meningitis, endocarditis, anthrax, diphtheria, tetanus, gas gangrene and syphilis, among many others. Penicillin V, by contrast, resists stomach acid well enough to be taken by mouth, making it the most widely used form of penicillin. Doses above 500 milligrams are not fully effective, since the body absorbs it poorly beyond that point. It treats the same range of infections as penicillin G, except conditions such as endocarditis that require very high blood concentrations.

  In the United States, about 10 percent of people report a penicillin allergy. That figure is misleading. The chance of a positive allergy skin test drops by 10 percent for every year a patient avoids the drug, so roughly 90 percent of people who report an allergy can eventually tolerate penicillin again. Even those with a genuine allergy can usually take cephalosporins instead, since the immune cross-reactivity between the two drug classes is only about 3 percent. Common side effects, affecting at least 1 percent of patients, include diarrhoea, nausea, rash, hives and superinfections such as candidiasis. Rarer reactions, striking somewhere between 0.1-1 percent of people, include fever, vomiting, seizures and pseudomembranous colitis. True allergic skin reactions occur in 1-10 percent of patients, while severe IgE-mediated anaphylaxis strikes only about 0.01 percent of them. Some patients also develop serum sickness, a delayed immune reaction appearing one to three weeks after exposure. It is distinct from a true drug allergy.

  Penicillin binds only weakly to proteins in the blood plasma, which affects how it moves through the body. Its bioavailability differs sharply by type: penicillin G reaches below 30 percent, while penicillin V reaches 60-70 percent. Because the kidneys clear penicillin quickly, patients need doses at least four times a day to keep blood levels adequate. Early treatment manuals recommended injections as often as every three hours, and dosing penicillin has been compared to trying to fill a bathtub with the plug pulled out. Larger, cheaper doses eventually made that schedule unnecessary, though some clinicians still recommend continuous penicillin infusions for the same reason.

  Underneath all these variants sits the same molecular skeleton: a compact ring weighing just 243 grams per mole. That small size lets it slip through defences that stop far bulkier drugs.

• 6. How It Kills A Bacterial Cell

  ›

  Gram-positive bacteria, like the Staphylococcus Fleming studied, have no outer membrane at all, just a thick cell wall woven from a loose mesh of glycoproteins. Penicillin molecules slip through that mesh with ease. Gram-negative bacteria are built differently. A thinner cell wall is wrapped in an extra outer membrane made of lipopolysaccharide, which blocks water-soluble molecules like penicillin from passing straight through. Penicillin can still enter through narrow protein channels called porins, which normally ferry nutrients into the cell. Smaller drugs move through porins faster. Ampicillin and amoxicillin diffuse quickly, while the larger penicillin G moves through more slowly. The still-bulkier vancomycin cannot fit through porins at all, which leaves it useless against gram-negative bacteria. Because the size and number of porins differs between species, different gram-negative bacteria end up with very different natural susceptibility to any given penicillin.

  Once inside a cell, penicillin blocks the final step of building peptidoglycan, the material that makes up the bacterial cell wall. It does this by binding, through its beta-lactam ring, to enzymes called penicillin-binding proteins that would otherwise cross-link the wall's structural strands. Specifically, the enzyme DD-transpeptidase mistakes the beta-lactam ring for its normal target: a linked pair of D-alanine molecules in a precursor called UDP-MurNAc. It binds the ring instead, and cross-linking stops. Enzymes that break down existing cross-links, however, keep working regardless. The wall grows weaker as the bacterium keeps building and dismantling it unevenly. Water then floods in, because the cell can no longer control its internal osmotic pressure. The cell eventually bursts, in a process called lysis. Bacterial enzymes called hydrolases and autolysins compound the damage further, digesting the weakened peptidoglycan once precursor molecules start piling up.

  The effect even reaches beyond bacteria. Penicillins block the division of cyanobacteria. They also block division in the photosynthetic cyanelles of glaucophyte algae and in the chloroplasts of bryophytes such as mosses. They leave the plastids of more advanced vascular plants untouched, though. That pattern supports the endosymbiotic theory of how plant cell organelles evolved.

  Fleming had noticed something else in that same original culture: plenty of bacteria were entirely unaffected by penicillin. Explaining why would take Ernst Chain and Edward Abraham another twelve years.

• 7. When Bacteria Fight Back

  ›

  Ernst Chain and Edward Abraham picked up that thread while working out penicillin's exact mechanism, and in 1940 they found their answer. Bacteria such as Escherichia coli, they discovered, produced a specific enzyme that broke penicillin molecules apart before the drug could do any damage. They named the enzyme penicillinase, now classified within a broader family called beta-lactamases. More than 2,000 distinct types of beta-lactamase have since been identified, each with its own amino acid sequence and its own preferred target within the beta-lactam ring.

  Bacteria resist penicillin through three main routes: reduced entry into the cell, reduced binding by penicillin-binding proteins, or outright destruction by beta-lactamase. In gram-positive species such as Staphylococcus aureus, resistance to the related drug vancomycin comes from building an extra-thick peptidoglycan wall that blocks the antibiotic's entry. In gram-negative bacteria, resistance instead comes from mutations that change the size or number of porins. Pseudomonas aeruginosa simply produces fewer porins. Enterobacter, E. coli and Klebsiella pneumoniae instead produce modified, non-specific porins, such as the OmpC and OmpF types, that cannot carry penicillin at all.

  Streptococcus pneumoniae carries six known mutant penicillin-binding proteins. Of these, PBP1a, PBP2b, PBP2x and sometimes PBP2a bind penicillin far less effectively than the normal versions do. Staphylococcus aureus has its own trick: it can activate a hidden gene that produces a low-affinity protein called PBD2 as backup defence. A particular strain, methicillin-resistant Staphylococcus aureus or MRSA, emerged after methicillin's introduction in 1959 and now resists not just penicillin but most other antibiotics too. In MRSA, mutations in a set of genes called the mec system produce a variant protein called PBP2a. PBP2a binds penicillin poorly and cannot complete peptidoglycan synthesis alone, even though the bacterium still runs four otherwise normal penicillin-binding proteins.

  Destruction by beta-lactamase enzymes is now considered the single most significant resistance mechanism penicillin faces, described simply as "the greatest threat to the usage of penicillins." It was also the first resistance mechanism ever discovered. When researchers tested penicillin's purity and biological activity in 1940, they found that E. coli was simply unaffected, and traced the cause to that same penicillin-destroying enzyme. Gram-positive bacteria secrete beta-lactamase in far larger quantities than gram-negative species do. In a mixed infection, that difference can let gram-positive bacteria effectively shield nearby gram-negative cells that would otherwise be susceptible. In Pseudomonas aeruginosa, resistance can also take the form of biofilms, and of multidrug-tolerant persister cells that survive treatment without any genetic resistance at all.

  One answer to this arms race came from redesigning the molecule itself. Flucloxacillin, for instance, carries a side chain that physically shields its beta-lactam ring from attacking enzymes.

• 8. Beyond The Original Mould

  ›

  Edward Abraham first proposed penicillin's chemical structure in 1942, though the proposal went unconfirmed for three more years. Dorothy Crowfoot Hodgkin, also working at Oxford, confirmed that structure in 1945 using X-ray crystallography. She went on to win the 1964 Nobel Prize in Chemistry, partly for that determination and partly for other structural work.

  At the Massachusetts Institute of Technology, the chemist John C. Sheehan completed the first full chemical synthesis of penicillin in 1957, after beginning the project in 1948. Along the way, he developed new methods for synthesising peptides, along with new protecting groups, chemical tools that temporarily mask a molecule's reactive sites during a reaction. Sheehan's synthesis method itself was never practical for mass production. One of its intermediate compounds, though, was 6-aminopenicillanic acid, which turned out to be exactly the nucleus researchers needed. Researchers at the Beecham Research Laboratories in Surrey had independently discovered how to isolate that same compound, 6-APA, in 1957, publishing the finding in 1959. Attaching different chemical groups to that nucleus opened the door to an entire generation of new, more versatile penicillins.

  The first major descendant was ampicillin, introduced in 1961 with a broader spectrum of activity than either of the two original penicillins. Further chemistry produced beta-lactamase-resistant penicillins such as dicloxacillin and methicillin. Methicillin, the first chemically altered penicillin, added methoxy groups at two positions on its side chain, which made it resistant to bacterial beta-lactamase. These drugs proved effective against beta-lactamase-producing bacteria, but ultimately useless against the MRSA strains that emerged afterward. Chemists also developed antipseudomonal penicillins, including carbenicillin, ticarcillin and piperacillin, aimed at hard-to-treat gram-negative bacteria. The beta-lactam ring itself proved so useful that it still sits at the core of related drug families today, including the mecillinams, the carbapenems and, most significantly, the cephalosporins.

  Modern penicillin is still produced by fermenting Penicillium rubens, deliberately stressed using a technique called fed-batch culture, since stress is what triggers the mould to produce the antibiotic as a defensive by-product. Researchers first cloned and sequenced the biosynthetic gene cluster responsible for that process in 1990. The pathway runs in three steps. First, three amino acids condense into a tripeptide. Second, that tripeptide converts into a weakly active intermediate called isopenicillin N. Third, a final transamidation swaps in a new side chain, turning the intermediate into full-strength penicillin G. That last step depends on a single gene, penDE, unique to this pathway and still the focus of ongoing research.