BMS 114 · Common Core (Tronc commun)Section 2.3 · How microbiology became a science

Section 7 of 33

Robert Koch, pure culture and Koch’s postulates

Describe the main discoveries and techniques of Robert Koch, explain what a pure culture is and why only it can prove which organism causes a disease, state the five Henle–Koch postulates, and explain two modern limitations of them.

New to this course? How each section works

Learn gives you one short idea at a time. Each idea starts with a single sentence, then explains it, gives an example, and asks you a small question you can answer straight away.
Explore shows a diagram of the same ideas. Click the numbered parts to read what each one is.
Check asks a few questions. Answer them all correctly to finish the section and unlock the next one.
“Go deeper” is optional. It adds mechanism and extra detail for when you are ready — nothing you need for the checks is hidden there.

Why this matters

Every time a laboratory report says “the organism isolated was …”, somebody first separated that organism from everything else in the specimen. Without that separation there is no way to know which of the many organisms present caused the illness, so there is no reliable cause and no reliable treatment. Koch built the method and wrote the rules for proving a cause, and those rules still shape how an outbreak, a food-borne infection or a hospital-acquired infection is investigated today. This section also teaches you to read a tuberculin test honestly, as evidence of infection rather than of active disease.

What you need to know first

You need the conclusion of Section 2.2 — that living things come only from other living things — and the ordinary idea that a disease has a cause. The new words are defined below and used constantly: pure culture, colony, agar-agar, Koch’s postulates and cell-mediated immunity. Nothing else is assumed.

Words to know (5)

Pure culture: A population of one single kind of organism, grown from one parent cell, with no other organism present. Picking one colony of anthrax bacilli from a solid medium and growing it alone gives a pure culture.
Colony: A visible heap of identical organisms on a solid medium, all descended from one parent cell that was held in place. After overnight incubation, each separated colony on the plate can be counted and identified.
Agar-agar: A jelly-like substance obtained from dried red seaweeds, used to solidify culture media because bacteria cannot digest it and it has no nutritive value. Agar is what keeps a streak plate firm while the colonies grow at 37 °C.
Koch’s postulates (Henle–Koch postulates): The set of criteria used to show that one particular microorganism is the cause of one particular disease. Proving that Mycobacterium tuberculosis causes tuberculosis was Koch’s own demonstration of the postulates.
Cell-mediated immunity: Protection carried out by cells — chiefly T lymphocytes and the macrophages they activate — rather than by antibodies. The delayed, localised swelling of a tuberculin test is cell-mediated immunity made visible on the skin.

One idea at a timeIdea 1 / 7

In one sentence A pure culture contains one single kind of organism, and only a pure culture can show which organism is causing a disease.

A pure culture is what turns a suspicion into a proof

A clinical specimen is almost never a single organism. Sputum carries dozens of species of mouth bacteria, stool carries hundreds, and a wound swab carries everything that lives on the skin. If you inject that mixture into an animal and the animal becomes ill, you have shown that something in the mixture is dangerous, but not which member of it. To prove a cause you must first separate the organisms from one another, let each multiply on its own, and then test one of them alone. The separated population is a pure culture, and the visible heap of identical organisms it forms on a solid medium is a colony. The causal claim only becomes testable at that point.

For example

A throat swab grows streptococci, staphylococci and diphtheroids together on the same plate. Only when one colony is picked and grown alone can the laboratory name the species and test whether it is sensitive to penicillin.

Pause and check yourself

A urine sample is injected into a mouse and the mouse dies. Why does this not prove which bacterium caused the infection?

Show the answer

Because the sample contained many organisms, and any one of them, or the mixture itself, could have killed the mouse. Only after the organisms are separated and tested one pure culture at a time can a single species be named as the cause.

Read detailed explanations

The Koch phenomenon: what a delayed, localised reaction reveals about cell-mediated immunity

When tuberculin is injected into the skin of a person whose immune system has already met Mycobacterium tuberculosis, the visible result is not a chemical irritation but immunological memory made flesh. The antigen is taken up by dendritic cells in the dermis and presented on MHC class II molecules — the surface labels that show antigen to T cells — to memory CD4+ T lymphocytes, the helper subset that coordinates the response, primed during the earlier infection. Those lymphocytes release interferon-gamma and tumour necrosis factor, cytokines (chemical signals) that recruit monocytes from the bloodstream and activate them into macrophages. Macrophages and fluid accumulate in the dermis over one to three days, producing a firm, raised plaque whose diameter — the induration — is what the nurse measures at 48 to 72 hours; the surrounding redness, or erythema, is not counted.

The reaction cannot be transferred from a sensitised donor to a naive recipient by serum, but it can be transferred by lymphocytes; that classical experiment is what proves the Koch phenomenon is cell-mediated immunity rather than antibody-mediated. Interpretation is quantitative and risk-stratified: 5 mm or more is read as positive in people with HIV, recent close contacts or fibrotic changes on chest radiography; 10 mm or more in recent arrivals from high-incidence countries, injecting drug users, laboratory staff and children under four; 15 mm or more only in people with no identifiable risk.

The reaction can be falsely negative when delayed hypersensitivity is suppressed — advanced tuberculosis, low CD4 counts, measles, severe malnutrition, corticosteroids or TNF inhibitors — a state called anergy, so a negative test never excludes infection. BCG vaccination and exposure to non-tuberculous mycobacteria can cause false positives. Interferon-gamma release assays measure the same T-cell response in blood and are unaffected by BCG. Contrast all this with immediate hypersensitivity, which appears within minutes as a wheal and flare, is driven by IgE and mast-cell histamine and responds to antihistamines; the Koch phenomenon takes a day or more, is induration rather than wheal, and is not relieved by antihistamines.

How the postulates fail for a virus

A virus cannot be isolated by the method Koch used, because his method assumed a free-living organism that multiplies in a cell-free medium and forms a colony. Viruses are obligate intracellular parasites: they carry no ribosomes, generate no ATP and can replicate only inside a living host cell, so no agar plate, broth or gelatin slant will grow them. The second postulate therefore has no direct viral equivalent — there is no colony to pick — and a virus must instead be propagated in cell-culture monolayers, embryonated chicken eggs or living animals, and identified by methods that need no growth at all: electron microscopy to see the particle, antigen detection, nucleic-acid amplification by polymerase chain reaction, or serology for the host response.

The first postulate, constant association with disease, also breaks down, because viruses are often carried and transmitted by people who never develop symptoms: a person shedding norovirus, rotavirus or SARS-CoV-2 without illness is a genuine source of infection, and asymptomatic HIV carriage can last years. The third postulate, reproduction of the disease in a healthy susceptible animal, fails whenever the virus has a narrow host range. HIV infects humans and, with difficulty, chimpanzees; it causes no equivalent disease in ordinary laboratory animals, and deliberately infecting a person is ethically impossible, so the postulate cannot be met. The fourth, re-isolation from the experimental animal, falls with it.

The fifth, demonstrable antibodies in the patient’s serum, is unreliable for viruses because antibody may appear only after the virus has been cleared and may persist for life as a memory imprint rather than a sign of current infection. The modern substitute is Falkow’s molecular postulates: the candidate virulence gene must be present in pathogenic strains and absent or inactive in non-pathogenic strains; deleting it must reduce virulence; restoring it must restore virulence; and the gene should be expressed during infection. This logic established the roles of the HIV envelope glycoprotein, the SARS-CoV-2 spike protein and the Shiga-like toxins of Escherichia coli.

Modern exceptions: carriers, unculturable organisms and the limits of animal models

Koch’s criteria are a method, not a law, and modern medicine accepts several well-documented exceptions. The first postulate, constant association with disease, fails for asymptomatic carriage. Mary Mallon, a cook working in New York in the early twentieth century, carried Salmonella enterica serovar Typhi in her gallbladder, infected at least fifty-one people and was responsible for three deaths, and never had a single symptom of typhoid fever. Neisseria meningitidis lives harmlessly in the throat of roughly one person in ten, and Streptococcus pneumoniae and Vibrio cholerae can also be carried without illness, so the mere presence of an organism in a population cannot establish that it causes disease. The second postulate fails whenever an organism cannot be grown in the laboratory. Treponema pallidum, the spirochaete of syphilis, has never been cultured in a cell-free medium, and Mycobacterium leprae, the cause of leprosy, cannot be grown on artificial media at all; it is propagated in the nine-banded armadillo and in mouse footpads.

Both are nevertheless accepted as causes, because they are consistently visible in the lesions and because treating the patient clears both the organism and the disease. The third postulate fails for organisms that do not reproduce human disease in any available animal model. Helicobacter pylori is the classic example: when Barry Marshall and Robin Warren proposed in the 1980s that a bacterium caused peptic ulceration, the claim was rejected partly because the organism could not be made to produce the disease reliably in animals, and Marshall eventually drank a broth culture himself in 1984, developed acute gastritis and re-isolated the organism — a human volunteer satisfying postulates three and four at once. The experiment earned the Nobel Prize in 2005 and would not pass an ethics committee today.

Two further assumptions are also too simple. One organism, one disease fails when a single species causes several illnesses, as Streptococcus pyogenes causes pharyngitis, impetigo, scarlet fever, rheumatic fever and necrotising fasciitis. One disease, one organism fails when a syndrome such as pneumonia or diarrhoea can be produced by dozens of different agents. Animal models may differ from humans in receptors and immunity, so transgenic animals are sometimes required, and organisms may lose virulence factors when subcultured repeatedly in the laboratory. Falkow’s molecular postulates, published in 1988, replaced the animal test with a genetic one and are now the working standard in bacterial pathogenesis.

Explore the diagram

Look at the whole drawing first. Then select one numbered part and connect it to the idea you just read.

From a mixed specimen to a proven cause: the pure-culture chain and Koch’s postulates

Follow the specimen downwards. Each step answers a question the step above could not: a mixture proves nothing about cause, a solid medium separates the organisms, one colony becomes a pure culture, and only then can an animal be inoculated and the organism recovered again. The last box shows where the chain breaks today.

New to diagrams? Read this first

Look at the whole drawing once without clicking anything. Ask yourself only: what kind of picture is this (a comparison, a route, a sequence, a table)?
Press Play, or use the arrows, to move through the numbered parts in order. The drawing stays still; only the highlight moves.
Click a number, or its name in the list below, to read what that part is and why it matters. Use “Hide labels” when you want to test your memory.

More controls

100%

01 / 6A mixture cannot name a cause

If the drawing is wider than this panel, scroll it sideways. Labels stay readable; the numbered explanations below show the same ideas.

A mixture cannot name a cause

A specimen from a sick host contains many species at once — blood, sputum or stool carries dozens of candidates. Injecting the whole mixture can show that something in it is harmful, but never which member did the damage, so no cause can be named and no treatment can be aimed at one organism.

Process schematic, not to scale: it shows the logic by which Koch turned a mixture into a provable cause. The box numbers follow the chain, not the postulates: box 4 carries postulate 3 and box 5 carries postulate 4. Postulate 2, growth in pure culture, fails for unculturable organisms; postulate 3, reproduction of the disease in a healthy susceptible animal, fails for many viruses and whenever no susceptible animal model exists; and postulate 4 falls with it.

Key points to remember

A pure culture contains one kind of organism, and only a pure culture can show which organism in a mixture causes a disease.
Koch (1843–1912) worked out the life history of the anthrax bacillus, introduced dried smears stained with aniline dyes, discovered Mycobacterium tuberculosis and Vibrio cholerae, and introduced solid media.
Agar-agar, suggested by Fanny Hesse, wife of Koch’s assistant Walther Hesse, sets at about 40–45 °C and melts at about 85–95 °C, not 45 °C and 90 °C as the notes state.
The Koch phenomenon, the delayed localised reaction to tuberculin, demonstrates cell-mediated immunity and underlies the tuberculin test.
The Henle–Koch postulates link one organism to one disease through constant association, pure culture, reproduction of disease in a susceptible animal, re-isolation and demonstrable antibodies; modern exceptions limit them.

Go deeper

Worked example

A cough that will not stop: Koch’s logic at the tuberculosis clinic. A 34-year-old man attends the clinic with three weeks of cough, drenching night sweats, weight loss and a low fever. He lives in a crowded hostel. You are the nurse seeing him. You must collect the right specimens, understand what the laboratory will do with them, and explain the tuberculin test to him and his family without either alarming them or giving false reassurance.

Collect the specimen properly. Ask for three early-morning sputum samples on separate days, coughed from deep in the chest rather than spat as saliva, before the first antituberculous dose. Collect outdoors or in a well-ventilated space, wear a fitted mask, label each container and send it promptly. A specimen that is really saliva cannot answer the question.
Understand the direct smear. The laboratory spreads the sputum thinly on a slide, dries and fixes it, and stains it with an acid-fast dye — the direct descendant of Koch’s dried smears and aniline dyes. Acid-fast rods can be reported within about a day, so the smear gives the first answer. It shows that organisms are present and that the patient is probably infectious, but it says nothing about which drugs will work.
Understand the pure culture. The same specimen is cultured on an egg-based solid medium, where the tubercle bacillus grows slowly into visible colonies — several weeks, because it divides only about once every fifteen to twenty hours. This step is Koch’s second postulate in action: only the separated growth of one organism can confirm the species and allow its susceptibility to the standard drugs to be tested.
Interpret the tuberculin test honestly. A tuberculin skin test is sometimes done in children or contacts: a small volume of tuberculin protein is injected into the skin of the forearm and the raised area is measured in millimetres at 48 to 72 hours. This is the Koch phenomenon used as a test. It shows that the person has been infected at some time; it does not show that the disease is active, and a negative result in a severely immunosuppressed patient may simply mean the reaction cannot appear.
Act, isolate and explain. A positive sputum smear means the patient is likely infectious, so he needs a ventilated room or a well-ventilated space, treatment started promptly, cough hygiene and screening of close contacts. Explain to the family that infection and disease are two different things: many people carry the bacillus without being ill, and only the cough specimens can show whether the disease is active in this patient.

The order is the whole point. Specimen, then direct smear, then pure culture, then susceptibility: each step answers a question the previous one could not. The tuberculin test runs alongside that chain, never in place of it, because it reports infection and not disease. This is Koch’s logic still working, unchanged, in a modern clinic.

Common misconception

A common beginner error is to read a positive tuberculin test as proof of active tuberculosis. The test shows that the immune system has met the bacillus at some time; it does not show that the person is ill or infectious, and most infected people never develop the disease. A second error is to imagine that Koch’s postulates are still applied exactly as they were written. They are the foundation of the causal argument, but they fail for asymptomatic carriers, for organisms that cannot be grown in the laboratory and for organisms that do not reproduce human disease in an animal model — and they fail completely for viruses, which have no independent metabolism and cannot grow on a plate.

Check your reasoning

Think of an infection whose cause is established. Which of Koch’s five postulates could actually be tested for it today, and which one would be impossible or unethical?

Reveal an answer

Take HIV infection. Finding the organism constantly in association with the disease and growing it in the laboratory are both achievable, since the virus replicates in cultured human cells. The third and fourth postulates are not: HIV has a very narrow host range, produces no equivalent disease in ordinary laboratory animals, and deliberately infecting a healthy person to see whether the disease appears would be a serious ethical violation. Antibodies are usually demonstrable, but their presence shows exposure and not whether the virus is still multiplying. Notice that medicine accepts HIV as the cause of AIDS on the strength of the first two postulates, molecular evidence and epidemiology alone — proof does not always need the whole list.

Where this comes from

BMS 114 General Microbiology I lecture notes (supplied) · “Robert Koch (1843-1912)/Koch’s postulates”, lines 432–494, including the five postulates at lines 476–482
Correction to the supplied notes — agar setting and melting points · Correction to the supplied notes: lines 456–457 give agar as solidifying at 45 °C and melting at 90 °C as exact figures. Agar sets near 40–45 °C and melts near 85–95 °C; these are approximate ranges, and the gap between them is why plates can be poured and then incubated at 37 °C.
Correction to the supplied notes — who suggested agar-agar · Correction to the supplied notes: lines 455–456 attribute the suggestion to “Frau Hesse, the wife of Koch’s student”. The correct person is Fanny Hesse, wife of Walther Hesse, one of Koch’s assistants; her husband was Koch’s co-worker, not his student.
Beyond the supplied notes — tuberculin testing, molecular postulates and modern exceptions · the Ziehl–Neelsen acid-fast stain, the Mantoux technique and its 5/10/15 mm risk-stratified thresholds, interferon-gamma release assays, anergy, Falkow’s molecular postulates (1988), asymptomatic carriage, unculturable organisms (Treponema pallidum, Mycobacterium leprae) and the Helicobacter pylori self-administration episode follow standard microbiology and immunology teaching; the notes describe the Koch phenomenon and the five postulates without these details.

Lesson progress is saved on this device for the current learner. Practice answers stay separate and are not stored.

About this course and its sources

Bilingual lessons organised from the supplied BMS 114 General Microbiology I lecture notes (Mr Ngotelo F. Ambene, Mr Lyonga B. Lyonga & Mr Fumbui Louis Ngong). Diagrams are original educational schematics; sections that correct the notes or go beyond them say so in their sources. Nursing and French-language review are pending.