Ames Test: Principle, Procedure, Results, and Interpretation

Before a new drug reaches a clinical trial, or a new food dye reaches a shelf, someone has to ask whether it damages DNA. Testing that directly in animals is slow and expensive, and cancer can take years to appear.

The Ames test is the fast first screen that sits in front of all of that. It asks a narrower question that a bacterium can answer in two days: does this compound cause mutations? The logic behind the test is that most substances that damage bacterial DNA also damage human DNA, and a mutagen is a strong suspect for a carcinogen. A compound that comes up positive here is flagged for the fuller testing that follows.

Devised by Bruce Ames and colleagues at the University of California, Berkeley, in the 1970s, the test remains one of the most widely used genotoxicity screens in the pharmaceutical and chemical industries, and a standard exam topic in microbiology and toxicology.

Not every mutagen is a carcinogen, and not every carcinogen is a mutagen. But many mutagenic chemicals are also carcinogenic, so knowing that a compound is mutagenic to bacteria is an early warning of possible danger. The Ames test screens for the warning; it does not by itself prove a compound causes cancer.

Principle

The Ames test is a reverse mutation assay. The test uses mutant strains of Salmonella Typhimurium that carry a defect in a gene of the histidine operon. Because of that defect they cannot make their own histidine, so they are histidine auxotrophs (written his-): they can grow only if histidine is supplied to them. Placed on a medium lacking histidine, they cannot form colonies, no matter how many cells are spread on the plate.

A mutagen can reverse that. If the test compound causes a new mutation at the defective gene that restores the ability to make histidine, the cell has reverted to the wild type (his+), a change called a back mutation or reversion. That single revertant cell can now grow without added histidine and forms a visible colony.

So the readout is simple to state: more revertant colonies means more reversion, which means the compound is mutagenic. The higher the number of revertant colonies above the background, the greater the mutagenic potential.

The test can also be run with tryptophan auxotrophs of Escherichia coli (WP2 strains), where the same logic applies to a defect in tryptophan synthesis instead of histidine.

The background lawn: the detail students miss

Here is the point that confuses most people the first time. If the strain cannot grow without histidine, why is the whole plate not blank except for the revertant colonies?

The answer is that the top agar contains a tiny, deliberate trace of histidine. It is not enough for the mutant bacteria to form colonies, but it is enough for all of them to divide a few times, producing a faint, even film of growth across the plate called the background lawn.

That lawn matters for two reasons. It confirms the bacteria were alive and the plate was inoculated, and, as we will see under interpretation, its disappearance is how the test signals that the compound was simply toxic rather than mutagenic.

Tester strains

Ames did not use one strain but a panel, and the reason is that different mutagens cause different kinds of DNA damage. A single strain would miss whole classes of mutagen, so a standard test uses several strains that each detect a different mutational mechanism.

The two mutational mechanisms the panel is built around are:

• Base-pair substitution: one base pair is swapped for another. Detected by strains such as TA1535 and TA100.
• Frameshift: one or more base pairs are inserted or deleted, shifting the reading frame. Detected by strains such as TA1537, TA1538, and TA98.

(TA100 is reverted by both mechanisms; TA1535 by base substitution only. This is the kind of pairing worth memorizing for exams.)

On top of the histidine mutation, the classic tester strains carry extra genetic changes that make them far more sensitive than an ordinary Salmonella:

• rfa mutation: a defective lipopolysaccharide cell wall, which is leakier and lets large or bulky test molecules reach the DNA that a normal wall would exclude.
• uvrB deletion: the excision-repair system is knocked out, so DNA damage is not quietly repaired before it can be fixed as a mutation, which makes damage more likely to show up as reversion.
• pKM101 plasmid (in TA98 and TA100): boosts an error-prone DNA repair pathway, raising sensitivity to many mutagens.

The takeaway is that these strains are deliberately engineered to be easy to mutate and permeable to test chemicals, which is what makes the assay sensitive enough to be a screen.

Metabolic activation: the S9 liver fraction

There is a problem the bacteria alone cannot solve. Many chemicals are not mutagenic as they arrive; they become mutagenic only after the liver chemically alters them. These are the pro-mutagens (and pro-carcinogens). In the body, liver enzymes (the mixed-function oxidases, mainly cytochrome P450) convert them into reactive forms that then attack DNA. A bacterium has no liver and no such enzymes, so on its own it would score these dangerous compounds as safe, a false negative.

The fix is to add the liver externally. A preparation called S9 is used: the post-mitochondrial supernatant of homogenized rodent liver, usually from rats pre-treated with an enzyme-inducing agent (such as Aroclor 1254) so the liver is rich in the activating enzymes. Mixed into the assay, the S9 fraction converts pro-mutagens into their active forms, so the test can catch them.

This is why a proper Ames test is run twice for each compound: once without S9 (to catch direct mutagens) and once with S9 (to catch those that need metabolic activation). A compound positive only in the presence of S9 is one the body itself would activate.

Procedure

The classic method is the plate-incorporation assay. The steps below are the standard version; the point is not to memorize a recipe but to see what each step is for.

1. Grow the tester strain. Inoculate a fresh culture of the chosen Salmonella tester strain in nutrient broth and grow it to late log phase (about 10 to 12 hours at 37 degrees C), so the cells are numerous and actively dividing.
2. Assemble the top agar. To a small volume of molten top agar (held around 45 degrees C) add the tester strain, the test compound at the chosen dose, and either the S9 mix (for the activation arm) or buffer (for the arm without activation). The top agar carries the trace of histidine and biotin that permits the background lawn.
3. Pour onto minimal glucose agar. Mix and pour the top agar across a minimal glucose agar plate, which contains no free histidine. Set up a matched control plate in which the compound is replaced by solvent alone (the negative or solvent control), and a positive control with a known mutagen.
4. Incubate. Incubate at 37 degrees C for 48 hours (some strains and formats need up to 72 hours). Protect photoreactive test compounds from light during incubation.
5. Read the plates. Count the revertant colonies on the test plates and compare them with the spontaneous revertant count on the solvent-control plate, checking that the background lawn is present.

A common variant is the pre-incubation method, in which the strain, compound, and S9 are mixed and incubated together briefly before the top agar is added. It can detect some mutagens (certain nitrosamines, azo dyes, and others) that the standard plate method misses.

Results and interpretation

This is the part that trips students, and it turns on one idea: the count that matters is not the number of colonies but the number above the background.

Reversion happens on its own at a low, roughly constant rate. Even with no mutagen present, a control plate grows a small number of spontaneous revertant colonies, and each strain has its own characteristic background range. So a positive result is never "colonies appeared." It is "many more colonies appeared than on the control."

Work through the reading of a plate:

1. Step one, check the control. The solvent-control plate should show its usual handful of spontaneous revertants and a healthy background lawn. This is the baseline the test plates are judged against.
2. Step two, compare the test plates. If a test plate shows a clear, dose-related increase in revertant colonies above that background (more colonies at higher doses of the compound), the compound is scored mutagenic. A convincing result rises with dose rather than appearing at a single concentration by chance.
3. Step three, watch the lawn for the trap. If a test plate has few or no colonies and the background lawn has thinned or vanished, the compound is not "clean", it is toxic: it has killed the bacteria, so there were none left alive to revert. Reading a thinned lawn as a negative result is the classic false-negative error, and it is why the lawn is checked on every plate.

In short: a mutagen raises the revertant count above the spontaneous background in a dose-dependent way; a toxic compound wipes out the lawn; and a truly negative compound leaves the plate looking like the control. A compound that scores positive here goes on to confirmatory mammalian and animal testing, because the Ames test is a screen, not a final verdict on carcinogenicity.

Variants

Beyond the classic plate-incorporation assay, several miniaturized formats are now common, mainly in drug discovery, where compound is scarce and throughput matters. These include the mini-Ames test (a scaled-down plate assay using far less compound), and the fluctuation-based formats such as Ames II and the Ames MPF assay, which are run in liquid in multi-well plates (for example 24-well or 384-well formats) and read by a color change when revertant growth shifts the pH of an indicator medium rather than by counting colonies on agar. The underlying reverse-mutation principle is identical; only the scale and the readout change.

Uses

• Screening chemicals, drugs, food additives, cosmetics, and environmental samples for mutagenic (and therefore possibly carcinogenic) potential.
• A standard early-stage genotoxicity screen in the pharmaceutical industry, run before a candidate compound proceeds toward clinical trials, and specified in regulatory guidance (OECD Test Guideline 471, the Bacterial Reverse Mutation Test).
• A research tool for studying mutagenesis and DNA repair.

How to Remember

• Reverse, not forward. The test detects a mutation that repairs a broken gene (his- back to his+), not one that breaks a good one. You are counting cells that regained the ability to grow without histidine.
• No histidine, no colony, unless you revert. The plate withholds histidine, so only revertants grow into colonies. The faint lawn everywhere else is the trace of histidine letting the rest divide a few times.
• Count above the background, not the total. Spontaneous revertants always appear. A positive result is a dose-related jump above the control's background, not the mere presence of colonies.
• A bare plate can mean toxic, not safe. If the lawn is gone, the compound killed the bacteria. No survivors means no revertants, which looks negative but is not. Always check the lawn.
• No liver in a bacterium, so add one. S9 (rodent liver fraction) supplies the enzymes that turn pro-mutagens into active mutagens. Run the test with and without S9.
• A panel, not one strain. Base-substitution strains (TA1535, TA100) and frameshift strains (TA1537, TA1538, TA98) catch different mutagens. One strain would miss whole classes.

Key facts

Point	Fact
What it detects	Mutagenicity (as a proxy for possible carcinogenicity)
Type of assay	Bacterial reverse mutation (back-mutation) assay
Organism	Histidine auxotrophs of Salmonella Typhimurium (or tryptophan auxotrophs of E. coli WP2)
Mutation reversed	his- (cannot make histidine) back to his+ (can), giving colonies on histidine-free medium
Base-substitution strains	TA1535, TA100
Frameshift strains	TA1537, TA1538, TA98 (TA100 detects both)
Sensitivity mutations	rfa (leaky cell wall), uvrB deletion (no excision repair), pKM101 plasmid (error-prone repair; TA98, TA100)
Background lawn	Trace histidine in top agar lets all cells divide a few times; confirms viability
Metabolic activation	S9, rodent liver post-mitochondrial fraction, converts pro-mutagens to active mutagens
Positive result	Dose-related increase in revertant colonies above the spontaneous background
Toxic result (the trap)	Few colonies AND a thinned or absent lawn; a false negative if misread
Standard guideline	OECD Test Guideline 471 (Bacterial Reverse Mutation Test)
Status of a positive	A screen; positives go to confirmatory mammalian and animal testing

Where Students Get Confused

"If the strain can't grow without histidine, why isn't the plate blank?" Because the top agar holds a trace of histidine, enough for a few divisions but not for colonies. That produces the faint background lawn. Colonies rise above the lawn only where a cell has reverted to making its own histidine.

"Any colonies at all means the compound is a mutagen, right?" No. A low, constant number of spontaneous revertants appears even without any mutagen. A positive result is a clear, dose-related increase above that background on the control plate, not the mere presence of colonies.

"A plate with almost no colonies means the compound is safe." Not necessarily. If the background lawn is also gone, the compound was toxic and killed the bacteria, so none survived to revert. That is a false negative, which is why the lawn is checked on every plate.

"Why bother adding liver to a bacterial test?" Because many chemicals are harmless until the liver converts them into reactive, DNA-damaging forms. Bacteria have no such enzymes, so without the S9 liver fraction the test would miss every pro-mutagen and pro-carcinogen.

"Does a positive Ames test prove the compound causes cancer?" No. It shows the compound is mutagenic in bacteria, which is a strong warning and a reason for further testing. Mutagenicity and carcinogenicity overlap heavily but are not identical, so confirmation needs mammalian and animal assays.

"Forward mutation or reverse mutation?" Reverse. The pre-existing mutation has already broken the histidine gene; the test looks for a second mutation that reverses that damage and restores function. You are detecting repair of a known defect, not the creation of a new one.

References

1. Ames BN, McCann J, Yamasaki E (1975). Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome mutagenicity test. Mutation Research. 31(6): 347-364.
2. Maron DM, Ames BN (1983). Revised methods for the Salmonella mutagenicity test. Mutation Research. 113(3-4): 173-215.
3. Mortelmans K, Zeiger E (2000). The Ames Salmonella/microsome mutagenicity assay. Mutation Research. 455(1-2): 29-60.
4. Organisation for Economic Co-operation and Development (2020). Test No. 471: Bacterial Reverse Mutation Test. OECD Guidelines for the Testing of Chemicals, Section 4. OECD Publishing.