Antibiotic resistance develops through genetic variation followed by natural selection and, importantly, through acquisition and spread of resistance genes. Bacterial populations are genetically diverse, and resistance can arise through spontaneous mutations that occur during DNA replication or through the acquisition of DNA from other bacteria. Most mutations do not necessarily benefit the bacterium, but a mutation that reduces an antibiotic's ability to inhibit growth can become advantageous when that antibiotic is present.
Similarly, a bacterium can acquire a plasmid carrying a resistance gene and thereby gain protection without having independently evolved the relevant mutation. Antibiotic exposure then creates a selective environment: susceptible bacteria are inhibited or killed, whereas resistant bacteria survive and reproduce. The resistant population can consequently become a larger proportion of the bacterial community. Resistance is therefore not normally created because a bacterium consciously “learns” to tolerate an antibiotic; it emerges from genetic variation and changes in population frequencies under selection.
Mutations can produce resistance through several routes. A mutation may alter the antibiotic's molecular target so that the drug binds less effectively, increase expression of an efflux pump that exports the drug, reduce production of a membrane channel through which the drug enters, change a metabolic pathway or activate a protective mechanism. Some mutations affect regulatory genes and consequently alter the expression of several resistance-associated proteins.
In other cases, resistance results from duplication or amplification of genes, increasing production of an enzyme that inactivates an antibiotic. The effect of a mutation depends on the particular bacterial species, antibiotic and genetic background. Resistance can also impose a biological cost when a resistance mechanism consumes energy or interferes with normal cellular functions. However, compensatory mutations can sometimes reduce that cost while allowing resistance to persist. Thus, even when antibiotic exposure declines, resistance does not necessarily disappear immediately.
Horizontal gene transfer is a major additional route by which resistance spreads. Bacteria can acquire resistance genes through conjugation, transformation or transduction. Conjugative plasmids are particularly important because they can transfer between bacterial cells and may carry multiple resistance determinants. Integrons can capture and express gene cassettes, including resistance genes, and transposons can facilitate movement of genetic material between chromosomes and plasmids. Bacteriophages can also transfer DNA between bacteria through transduction.
These processes allow resistance traits to spread faster than would be possible through ordinary reproduction alone. A bacterium does not have to independently develop a new mutation if it can obtain an already functional resistance gene from another organism. This is especially significant in dense microbial communities, healthcare environments and settings where different bacterial species coexist and exchange genetic material.
Antibiotic use creates selection pressure, but different patterns of use generate different selective environments. Unnecessary prescriptions expose bacteria to drugs without providing clinical benefit. Incorrect drug selection may expose pathogens to an antibiotic that is ineffective against them, while inappropriate dosing can result in inadequate drug exposure in some circumstances.
Prolonged or repeated exposure can increase opportunities for resistant organisms to survive selection, although the relationship between duration and resistance is complex and depends on the antibiotic, pathogen and clinical context. Broad-spectrum antibiotics can affect a wider range of bacteria than narrow-spectrum agents, potentially exerting greater ecological pressure on susceptible members of the microbiota.
This does not mean that broad-spectrum antibiotics are inherently inappropriate; they can be essential when a severe infection requires immediate empirical treatment before the causative organism and susceptibility profile are known. Once reliable information becomes available, treatment can sometimes be narrowed to a more targeted drug.
Resistance can spread because bacteria themselves spread between people, animals and environments. Transmission can occur through direct contact, respiratory droplets or aerosols for relevant organisms, contaminated hands and surfaces, food, water and other routes depending on the organism. Healthcare facilities can facilitate transmission because patients may have severe infections, frequent antibiotic exposure and invasive devices, while healthcare workers and shared environments create opportunities for bacterial movement. Resistant organisms can also circulate in communities without healthcare exposure.
Veterinary antibiotic use can select resistant bacteria in animals, and resistant organisms or genes can move between animals, humans and environmental reservoirs through complex pathways. Wastewater from households, healthcare facilities, pharmaceutical manufacturing and agricultural settings can contain bacteria, resistance genes and antimicrobial residues. These interconnected pathways underpin the “One Health” approach, which treats human, animal and environmental health as linked components of antimicrobial resistance.
Biofilms can further complicate resistance and tolerance. A biofilm is a structured microbial community attached to a surface and embedded in an extracellular matrix. Bacteria in biofilms may experience reduced antibiotic penetration, altered metabolic states and physiological conditions that make them less susceptible to killing. Biofilm-associated survival is not necessarily equivalent to genetically encoded antibiotic resistance, and the terms resistance and tolerance should be distinguished.
Resistance generally refers to a heritable reduction in susceptibility, whereas tolerance describes the ability of bacteria to survive antibiotic exposure without necessarily having a higher MIC. Persister cells are another phenomenon: small subpopulations can enter transient physiological states in which antibiotics that depend on active growth are less effective. These mechanisms can contribute to persistent or recurrent infections and can coexist with genetically mediated resistance.
The evolutionary trajectory of resistance can involve trade-offs and subsequent adaptation. A resistance mutation may initially reduce growth rate or competitiveness when antibiotics are absent, but additional mutations can compensate for that disadvantage. Alternatively, continued antibiotic exposure may favour bacteria that acquire combinations of resistance mechanisms. A single mobile genetic element can sometimes carry genes conferring resistance to several antibiotic classes, allowing co-selection: exposure to one antibiotic can indirectly favour bacteria carrying resistance genes against other drugs.
Non-antibiotic substances and environmental conditions can also contribute to selection in some settings. The outcome is therefore shaped by the specific bacterial population, genetic elements, antibiotic exposure pattern, ecological environment, transmission network and competing organisms. This explains why resistance can emerge independently in multiple locations and why controlling it requires more than discovering new antibiotics.
In practical terms, antibiotic resistance develops because bacteria reproduce and exchange genetic material in environments where antimicrobial exposure changes which organisms survive. Genetic mutations and horizontal gene transfer provide variation; antibiotics create selection; reproduction increases the relative abundance of successful resistant bacteria; and transmission distributes them to new hosts and environments.
Prevention therefore aims to interrupt several stages of this process simultaneously: avoiding unnecessary antibiotic exposure, using antibiotics appropriately when genuinely needed, preventing infections so antibiotics are required less often, improving infection-control measures, using diagnostics and surveillance to identify resistance, managing antimicrobial use across human and veterinary medicine, and limiting environmental dissemination. Because resistance is an evolutionary process rather than a single event, sustained control requires continuous monitoring and adaptation rather than a one-time intervention.
