Mutation in Bacteria

A mutation is any stable, heritable change in the nucleotide sequence of an organism's genome, whether that change affects a single base pair or spans thousands. In bacteria, because most genes are haploid and populations replicate asexually, a new mutation is expressed immediately in the cell that carries it and passed directly to its descendants without being masked by a second allele. This is a crucial contrast with diploid eukaryotes, where a recessive mutation can hide for generations; in a bacterium, a mutation's phenotype is generally visible right away, subject to selection in the very next generation. Mutations are classified structurally and functionally. Structurally, a point mutation changes a single base pair: a transition swaps one purine for another purine (A↔G) or one pyrimidine for another (C↔T), while a transversion swaps a purine for a pyrimidine or vice versa. Insertions and deletions ("indels") add or remove one or more base pairs; when the number added or removed is not a multiple of three, they cause a frameshift, scrambling every downstream codon. Functionally, a point mutation within a coding sequence can be silent (no amino acid change, usually due to codon degeneracy), missense (changes the amino acid), or nonsense (creates a premature stop codon, truncating the protein). Mutations outside coding regions can alter promoters, ribosome-binding sites, or regulatory elements, changing expression level rather than protein sequence. A separate and important axis is reversibility. A forward mutation converts a wild-type gene to a mutant (nonfunctional or altered) form; a reverse mutation, or reversion, restores the wild-type phenotype, either by a true reversion (exact restoration of the original sequence) or a suppressor mutation elsewhere in the genome that compensates for the first change without repairing it directly. This distinction underlies classic bacterial genetic tools such as the Ames test for mutagenicity, discussed below, which scores reversion frequencies in tester strains.

Step 1: Spontaneous replication errors. DNA polymerase III, the primary replicative enzyme in bacteria, is extraordinarily accurate but not perfect. Its intrinsic error rate before any correction is roughly 10⁻⁵ per base pair per replication, arising mainly from transient tautomeric shifts in bases (e.g., a cytosine briefly adopting its rare imino form pairs with adenine instead of guanine) or from slippage of the template strand during replication of repetitive sequences. Two proofreading and repair systems then reduce this error rate by roughly 10,000-fold. First, the polymerase's own 3′→5′ exonuclease activity proofreads each newly added base immediately, excising mismatches. Second, the mismatch repair (MMR) system (MutS, MutL, MutH in E. coli) scans newly synthesized DNA after replication, distinguishing the new (unmethylated, or transiently under-methylated at GATC sites) strand from the old template strand, and excises and resynthesizes stretches containing mismatches. The net result is a genome-wide mutation rate of about 10⁻⁹ to 10⁻¹⁰ mutations per base pair per generation in wild-type E. coli, translating to roughly 10⁻⁶ mutations per gene per generation for an average-sized gene (~1000 bp), though this varies by gene and by local sequence context.

Step 2: DNA damage and its repair (or misrepair). Beyond copying errors, DNA is chemically fragile. Spontaneous depurination (loss of a purine base, roughly 10,000 events per cell per generation in E. coli) and deamination (cytosine converting to uracil, or 5-methylcytosine converting to thymine) constantly generate lesions. Environmental agents accelerate this: ultraviolet light creates pyrimidine dimers (covalent links between adjacent thymines); oxidative damage from reactive oxygen species generates 8-oxoguanine, which mispairs with adenine; alkylating agents add methyl or ethyl groups to bases, altering their pairing properties. Cells deploy several repair pathways — base excision repair, nucleotide excision repair, and photoreactivation — to fix these lesions before replication. When damage is extensive, cells can invoke the SOS response, an emergency system in which RecA protein, activated by single-stranded DNA at stalled replication forks, promotes self-cleavage of the LexA repressor, derepressing dozens of DNA-repair genes including error-prone translesion polymerases (Pol IV, Pol V). These polymerases can replicate past a lesion that would otherwise block the replication fork, but they do so with low fidelity, deliberately trading accuracy for survival — a phenomenon called SOS mutagenesis, since it substantially raises the mutation rate during stress.

Step 3: Mutagens. A mutagen is any physical or chemical agent that increases the mutation rate above the spontaneous background. Mutagens act through several mechanisms:

• Base analogs (e.g., 5-bromouracil) resemble normal bases closely enough to be incorporated by polymerase but pair ambiguously, causing transitions in subsequent replication rounds.
• Intercalating agents (e.g., ethidium bromide, acridine orange) wedge between stacked base pairs, distorting the helix and causing single base-pair insertions or deletions during replication — frameshift mutagens.
• Deaminating and alkylating agents (e.g., nitrous acid, ethyl methanesulfonate) chemically modify bases directly, changing their hydrogen-bonding pattern.
• UV radiation and ionizing radiation cause bulky lesions and strand breaks that are mutagenic if misrepaired.

Step 4: Transposition and larger-scale rearrangement. Bacterial genomes carry mobile genetic elements called transposons and simpler insertion sequences (IS elements), which can excise and reinsert (or copy and insert) themselves elsewhere in the genome via transposase enzymes. Insertion disrupts whatever gene it lands in, a major source of spontaneous loss-of-function mutations in laboratory strains. Larger rearrangements — inversions, duplications, and deletions — arise from recombination between repeated sequences (e.g., rRNA operons) and can duplicate or delete entire operons, altering gene dosage.

Step 5: Quantifying mutation rate versus mutation frequency. These two terms are often conflated but are distinct. Mutation rate is the probability that a mutation occurs per gene (or per base pair) per generation (per cell division) — a rate constant of the underlying stochastic process. Mutation frequency is simply the proportion of cells in a population that carry a given mutant allele at the moment of measurement — the outcome of accumulated mutation events plus subsequent growth and selection. Because a single early mutation in a growing culture is amplified through many subsequent divisions, frequency can vastly overestimate rate unless corrected — precisely the statistical problem the Luria–Delbrück analysis solved (below).

Step 6: Selection acts on standing variation. Once a mutation exists in even a single cell, its fate depends on selection. If the mutant phenotype confers no advantage, it drifts at low frequency or is lost by chance (especially likely while rare, due to sampling — analogous to genetic drift). If the environment changes so that the mutant phenotype is advantageous — for instance, exposure to a bacteriophage against which only mutant cells are resistant, or an antibiotic that only resistant cells survive — the mutant lineage rapidly outcompetes wild-type cells. Critically, selection does not create the resistant mutant; it merely allows pre-existing resistant cells (or their descendants) to survive and multiply while susceptible cells die. This distinction — mutation as a random, ongoing background process, and selection as an environmental filter acting afterward — is the conceptual heart of bacterial population genetics and the subject of the classic experiments below.

Before 1943, a live scientific debate concerned whether bacteria that survived a lethal challenge (such as infection by a bacteriophage) became resistant because of the challenge — a Lamarckian-style adaptive response — or whether resistant mutants were already present in the population, generated at random beforehand, with the phage merely acting as a selective filter that revealed them.

The Luria–Delbrück fluctuation test (1943). Salvador Luria and Max Delbrück designed a statistical experiment to distinguish these hypotheses using E. coli and the phage T1. They reasoned that the two hypotheses make different quantitative predictions about the variance in the number of resistant colonies across independent replicate cultures. If resistance were induced by phage exposure — each bacterium, upon meeting the phage, having some fixed independent probability of becoming resistant — then the number of resistant survivors per culture should follow a Poisson distribution, in which the variance approximately equals the mean. Every culture, having started identically and experienced the same challenge, should show similar, low, fluctuation in resistant colony counts. If instead resistance mutations arose spontaneously and randomly at various times during growth, before phage was ever added, then the number of resistant cells at the end would depend enormously on when the mutation happened to occur in each culture's history. A mutation arising early in one culture's growth would be amplified through many subsequent generations, producing a "jackpot" culture with an enormous number of resistant cells; a mutation arising late, or not at all, would leave few or none. This process predicts a distribution with variance far exceeding the mean — a "jackpot" distribution, not Poisson. Luria and Delbrück grew many small parallel cultures of E. coli from a common small inoculum, allowed them to grow to comparable final densities, then plated each entire culture on agar containing phage T1 and counted resistant colonies. The results showed enormous fluctuation between culture-to-culture resistant counts — some cultures had zero or a few resistant colonies, while a small number of cultures produced hundreds, "jackpots" — with variance far exceeding the mean, inconsistent with a Poisson process and consistent with spontaneous mutation occurring at random times before the phage was ever applied. This mathematical result, along with the derived formula relating mutation rate to the distribution of jackpots (later refined by others), established that mutation to phage resistance pre-exists selection. Luria and Delbrück shared the 1969 Nobel Prize in Physiology or Medicine (with Alfred Hershey) for this and related work.

Lederberg and Lederberg's replica plating (1952). Joshua and Esther Lederberg devised an elegant, purely visual confirmation. They pressed a velvet pad onto a "master plate" of bacterial colonies grown without any phage or antibiotic present, then stamped that same velvet pad onto several fresh plates that did contain phage (or antibiotic) — a technique called replica plating, which preserves the spatial position of every colony across all plates. If resistance were induced only upon contact with the selective agent, resistant colonies on different replica plates should appear at essentially random, uncorrelated positions. Instead, resistant colonies appeared at the same physical position on every replica plate — corresponding to a single colony on the original, drug-free master plate. This directly demonstrated that the resistant clone already existed on the master plate before ever encountering the selective agent; selection did nothing but reveal a pre-existing variant. Crucially, because the master plate itself never touched phage or antibiotic, the resistant cells there could be isolated and shown to be resistant — proof positive that mutation preceded, and was independent of, selection. Together these two experiments are considered a founding pillar of bacterial genetics and of evolutionary biology generally: they established that mutation is a spontaneous, blind, undirected process, and that the environment's only role is to select among mutants that already exist.

Example 1: Estimating mutation rate from a fluctuation-style dataset. Suppose you grow 20 independent 1 mL cultures of E. coli, each inoculated with about 500 cells and grown to a final density of 5 × 10⁸ cells/mL (roughly 20 generations of growth, since 500 × 2²⁰ ≈ 5 × 10⁸). You plate each entire culture on medium containing streptomycin and count resistant colonies. Sixteen of the 20 cultures show low numbers (0–3 resistant colonies), but four cultures are jackpots, with 85, 140, 310, and 610 resistant colonies respectively. This asymmetric, right-skewed pattern is itself qualitative evidence of spontaneous, pre-existing mutation (a Poisson/induced model would never generate such extreme, rare jackpots). Using the Lea–Coulson / Luria–Delbrück method-of-the-median (a standard approach that avoids being distorted by jackpots, since the median is far more robust than the mean to extreme outliers), you would take the median number of resistant colonies across the 20 cultures — here the 16 low values dominate the median, giving something like 1–2 resistant colonies — and use the relation: m ≈ median (approximately, for low mutation rates) where m is the expected number of mutational events per culture. Dividing m by the final number of cells per culture (5 × 10⁸) gives a mutation rate on the order of 10⁻⁸ to 10⁻⁹ mutations per cell per generation for this locus, in the same range as classically reported values for streptomycin resistance mutations in E. coli (which typically arise from single point mutations in the rpsL gene encoding ribosomal protein S12). Notice that if you had instead simply averaged all 20 counts (mean ≈ 57), you would have wildly overestimated the true rate by roughly 30- to 50-fold — exactly the pitfall of confusing mutation frequency (contaminated by jackpots) with mutation rate.

Example 2: Reversion analysis to distinguish point mutation from deletion. A microbiologist isolates a his⁻ auxotrophic mutant of Salmonella (unable to synthesize histidine, requiring it in the medium) and wants to know whether the underlying mutation is a simple point mutation or a larger deletion. She plates roughly 10⁹ cells of the his⁻ strain on minimal medium lacking histidine, in two parallel conditions: one plate receiving no mutagen, one plate pre-treated with the chemical mutagen 2-aminopurine (a base analog). On the untreated plate she observes 12 spontaneous his⁺ revertant colonies (reversion frequency ≈ 1.2 × 10⁻⁸); on the mutagen-treated plate she observes 340 revertant colonies (reversion frequency ≈ 3.4 × 10⁻⁷), a roughly 28-fold increase. Because 2-aminopurine specifically induces transition mutations by mispairing during replication, and because it strongly elevates the reversion rate, she concludes the original his⁻ mutation was very likely a point mutation (a base substitution or frameshift correctable by a further point change), not a deletion — a deletion of the gene could never be "reverted" by any point mutagen, since the genetic information would be physically absent and no chemical modification of a single base could recreate it. (This exact logic — comparing spontaneous versus mutagen-induced reversion frequencies in specially engineered Salmonella tester strains — underlies the Ames test, a widely used assay for the mutagenic, and by extension often carcinogenic, potential of chemical compounds.)

Salvador Luria was born in Turin, Italy, trained in medicine, and fled fascist Italy and then occupied France, arriving in the United States in 1940 as a refugee. Working at Columbia and then Indiana University with limited resources during wartime, he began a fruitful collaboration with the German-born physicist-turned-biologist Max Delbrück, whom he had met at a physics conference in 1941. Delbrück brought a physicist's insistence on quantitative, falsifiable models to a field that had until then been largely descriptive; Luria brought deep familiarity with bacteriophage biology. Their 1943 paper, "Mutations of Bacteria from Virus Sensitivity to Virus Resistance," published in Genetics, combined careful experimental design with statistical modeling of variance, a hybrid approach rare for its time and foundational for what became the "Phage Group," a loose network of scientists (including Alfred Hershey) who over the following two decades established the molecular basis of heredity using bacteria and their viruses.

Understanding spontaneous bacterial mutation underlies the entire modern framework for combating antibiotic resistance. Because resistance mutations (e.g., in rpoB conferring rifampin resistance, or gyrA conferring fluoroquinolone resistance) exist at low frequency in bacterial populations before a patient ever receives the drug, clinicians understand that antibiotic use does not "cause" resistance to appear from nothing — it selects and enriches for resistant cells that were already present, which is precisely why combination therapy (multiple drugs simultaneously) is used for pathogens like Mycobacterium tuberculosis: the probability that a single cell carries independent, simultaneous spontaneous mutations conferring resistance to two unrelated drugs is the product of two very small numbers, dramatically reducing the chance any cell can survive both. The Ames test, built directly on reversion-frequency logic, remains a standard, inexpensive screen used by regulatory agencies and industry to flag candidate drugs, food additives, cosmetics, and industrial chemicals for mutagenic (and likely carcinogenic) potential, using specially engineered auxotrophic Salmonella tester strains. In biotechnology, deliberate mutagenesis (UV, chemical mutagens, or error-prone PCR) combined with selection or screening — "directed evolution" — is used to engineer enzymes with improved industrial properties (thermostability, novel substrate specificity), a technique that won Frances Arnold the 2018 Nobel Prize in Chemistry. In agriculture and food microbiology, mutant strains of bacteria and fungi selected for overproduction of vitamins, antibiotics, or amino acids (e.g., high-yield penicillin-producing Penicillium strains derived through mutagenesis programs) remain industrially important. In public health and epidemiology, mutation-rate estimates from bacterial pathogens feed directly into models of how quickly resistance or immune evasion can emerge in an outbreak, informing surveillance and stewardship policy. The conceptual foundation for all of this — that mutation is random, ongoing, and independent of need, with selection acting only afterward — connects directly to the mechanisms of Horizontal Gene Transfer, through which resistance genes generated by mutation in one lineage can spread to entirely unrelated bacteria, and to Bacterial Gene Regulation, which governs how existing genes (mutated or not) are expressed in response to the environment, as distinct from how new genetic variants arise in the first place. It also depends on foundational concepts from the Overview of Microbial Genetics, particularly the structure of the bacterial chromosome and the machinery of DNA replication.

::: The correct answer key note: :::

Ask the tutor: How would the Luria–Delbrück statistical logic change if bacteria could also repair mutations back to wild type at an appreciable rate during growth?