How Antibiotics Kill Bacteria

Figure 1: Timeline of major antibiotic class discoveries. The discovery gap since 1987 is the backdrop of the resistance crisis.

01 The Accidental Discovery

In September 1928, Alexander Fleming returned to his laboratory at St Mary's Hospital in London after a vacation. He had been studying staphylococci, the bacteria that cause skin infections, and had left a stack of petri dishes on his bench. One dish was contaminated with a mold — a stray spore that had drifted in through an open window. Fleming noticed that the bacteria had been destroyed in a clear ring surrounding the mold. The mold was Penicillium notatum, and the substance it secreted — which Fleming called "mould juice" before naming it penicillin — would go on to save an estimated 200 million lives.

Fleming himself did not develop penicillin into a drug. That work fell to Howard Florey, Ernst Chain, and their team at Oxford University, who purified penicillin and demonstrated its efficacy in mice and then humans during the early years of World War II. By D-Day in 1944, American pharmaceutical companies were producing enough penicillin to treat every severe casualty of the Allied forces. Fleming, Florey, and Chain shared the Nobel Prize in 1945. The antibiotic era had begun.

02 Mechanism One: Destroying the Cell Wall

Bacteria and humans are built differently. One of the most fundamental differences is the cell wall — a rigid outer structure that bacteria maintain and human cells do not. The bacterial cell wall is made of a mesh-like polymer called peptidoglycan, a lattice of sugars and amino acids that gives the cell its shape and prevents it from bursting under internal osmotic pressure. Without it, the bacterium takes on water and ruptures. Penicillin and its relatives — the beta-lactam antibiotics, including amoxicillin, methicillin, and cephalosporins — exploit this difference ruthlessly.

Beta-lactam antibiotics bind to and disable the enzymes that bacteria use to cross-link the peptidoglycan mesh. With the cross-links broken, the wall cannot hold. As the bacterium tries to grow and divide, the wall fails and the cell bursts — a process called lysis. Because human cells have no cell wall and no peptidoglycan, the drug has nothing to attack in the human body. This is what makes cell-wall inhibitors selectively toxic: they kill bacteria while leaving host cells untouched. It is also why penicillin was such a breakthrough — the first drug that could kill bacteria without poisoning the patient.

Figure 2: The four mechanisms by which antibiotics attack bacteria, with representative drug classes for each.

03 Mechanism Two: Shutting Down Protein Synthesis

Bacteria have ribosomes — the molecular machines that translate genetic code into proteins — but their ribosomes are structurally different from human ribosomes. Bacterial ribosomes are 70S (composed of 30S and 50S subunits), while human ribosomes are 80S (40S and 60S subunits). This difference in size and structure is a target. Aminoglycosides like streptomycin and gentamicin bind to the 30S subunit, causing the ribosome to misread the genetic code and produce garbage proteins. Tetracyclines block the entry of transfer RNA, starving the ribosome of raw materials. Macrolides like erythromycin block the exit tunnel, so newly made proteins pile up inside the ribosome and jam the machine. Chloramphenicol blocks the enzyme that forms peptide bonds, the very link between amino acids.

Each of these drugs exploits a slightly different structural feature of the bacterial ribosome, which is why they are sometimes used in combination and why bacteria must evolve different resistance mechanisms for each. The result is the same: the bacterium cannot make proteins, cannot maintain its enzymes, cannot grow, and cannot divide. The infection stops.

04 Mechanism Three: Blocking DNA Replication

Some antibiotics attack the bacterium's ability to copy its DNA, which it must do before dividing. The fluoroquinolones — ciprofloxacin, levofloxacin, moxifloxacin — target two essential bacterial enzymes: DNA gyrase and topoisomerase IV. These enzymes cut, unwind, and reseal DNA during replication. Without them, the bacterial chromosome becomes a tangled knot that cannot be copied. The cell cannot divide and eventually dies. Metronidazole works differently: it is a prodrug that is activated inside anaerobic bacteria, producing free radicals that physically damage DNA strands. These drugs are particularly useful against intracellular bacteria and those in low-oxygen environments.

Because DNA replication is fundamental to all life, selectivity is more challenging here than with cell-wall inhibitors. Fluoroquinolones are designed to bind bacterial enzymes far more tightly than their human counterparts, but they are not perfectly selective — which is why fluoroquinolones carry warnings about tendon rupture and neurological side effects, and why they are reserved for infections where other options have failed.

05 Mechanism Four: Tearing the Membrane

Some antibiotics go straight for the membrane — the lipid bilayer that surrounds every bacterial cell. Polymyxin B and colistin are cyclic peptides that bind to lipopolysaccharides in the outer membrane of Gram-negative bacteria and then insert themselves into the membrane, forming physical holes. The cell contents leak out, and the bacterium dies. Daptomycin works similarly against Gram-positive bacteria, inserting into the membrane and causing depolarization — the electrical gradient across the membrane collapses, and the cell dies. These membrane-targeting antibiotics are often reserved for last-resort cases because they can also affect human cell membranes at high doses, making them more toxic than other classes.

06 The Resistance Crisis

Bacteria have been fighting chemical warfare for billions of years. They produce antibiotics naturally — penicillin comes from a mold, streptomycin from a soil bacterium — and they have evolved resistance mechanisms to survive in a world full of antimicrobial compounds. The surprise is not that resistance exists; it is that we expected a century of clinical antibiotic use to outpace 3.5 billion years of bacterial evolution. It did not.

Resistance emerges through several pathways. Bacteria can mutate the target protein so the drug no longer binds — this is how MRSA (methicillin-resistant Staphylococcus aureus) acquired its resistance to beta-lactams. They can pump the drug out of the cell using efflux pumps. They can produce enzymes that destroy the drug — beta-lactamase, which chops the beta-lactam ring, is the most common resistance mechanism worldwide. And critically, they can share these resistance genes with each other through horizontal gene transfer — passing plasmids, small circles of DNA, between unrelated bacteria. A resistance gene that evolves in one harmless soil bacterium can end up in a deadly pathogen through a chain of transfers.

The World Health Organization estimates that antimicrobial resistance already causes at least 700,000 deaths per year globally and projects this could rise to 10 million by 2050 if current trends continue. The economic cost could reach $100 trillion in lost output. The discovery gap — no fundamentally new antibiotic class since daptomycin in 1987 — means the pipeline of new drugs is nearly empty. Meanwhile, antibiotic use in agriculture — where roughly 70% of all antibiotics are used as growth promoters and prophylactics in livestock — continues to drive resistance in the bacterial populations that surround human communities.

07 The Future: Phages, Vaccines, and Stewardship

The solution to antibiotic resistance will not be a single new drug. It will be a combination of approaches. Vaccines reduce the number of infections that require antibiotics in the first place. Rapid diagnostics ensure that the right antibiotic is used for the right bacterium, reducing the broad-spectrum pressure that drives resistance. Antibiotic stewardship — using antibiotics only when necessary, at the right dose, for the right duration — slows the evolution of resistance in clinical settings. Bacteriophage therapy, which uses viruses that specifically attack bacteria, is being revisited as a complement to antibiotics, particularly for multi-drug-resistant infections.

New antibiotic classes are still needed, and some promising approaches are in development. Teixobactin, discovered in 2015 using a new method for culturing previously unculturable soil bacteria, attacks the cell wall through a mechanism that bacteria may find difficult to evade. CRISPR-based approaches aim to selectively kill resistance genes in bacterial populations. But the fundamental challenge remains economic: antibiotics are not profitable drugs. They are used briefly, cured quickly, and must be preserved rather than sold in volume. New incentive structures — including government-funded prizes and subscription payment models — are being explored to make antibiotic development financially viable for pharmaceutical companies. Whether these arrive in time depends on whether the world treats antibiotic resistance as the slow-moving pandemic it already is.