Learning outcomes
By the end of this lesson you should be able to: distinguish vertical from horizontal gene transfer and explain why only the second spreads resistance across species; describe the three classical mechanisms of horizontal transfer — transformation, transduction and conjugation — with a named example of each; explain the structure of the F plasmid and how it produces Hfr strains; describe transposons, integrons and insertion sequences and their role in assembling resistance; state the spontaneous mutation rate and explain why mutation never produces resistance 'on demand'; and interpret the clinical genetics of MRSA, ESBL producers and vancomycin-resistant enterococci.
Prerequisites
Lessons 1 and 2. You need the envelope architecture (from lesson 2, the growth curve and the idea that generation time is short) to see why a per-generation mutation rate becomes a population-level problem within hours.
The big idea
A bacterial genome has two channels of inheritance. Vertically, the chromosome is copied and each daughter cell receives a copy; this is stable, slow and confined to a lineage. Horizontally, genes arrive on plasmids, bacteriophages or naked DNA and may be accepted even from a different species. That distinction is the whole of this lesson: mutation is how a single lineage becomes different, whereas horizontal transfer is how resistance becomes an epidemic phenomenon.
The numbers make the problem concrete. E. coli doubles every twenty minutes, so a mutation occurring once per 10^6 divisions produces thousands of mutants within a single afternoon in a culture of 10^9 cells. In a patient on a fluoroquinolone, the antibiotic is a filter: it kills the susceptible majority and leaves the rare resistant mutant to fill the space. The drug does not create the mutant — it selects it, and understanding the difference is the difference between a correct clinical decision and a dangerous misunderstanding.
Mutation: the simplest engine of change
Spontaneous mutations arise from replication errors and DNA damage at a rate of roughly 10^-6 to 10^-9 per base pair per generation, and are random with respect to their usefulness. Selection then decides which of them matters. Bacteria have no special pathway that increases mutation rate at a useful gene; there are only mutagenic stress responses (SOS repair, error-prone polymerases) that raise the whole-genome rate under stress, which is why fluoroquinolones can accelerate the appearance of resistance.
| Type of genetic change | What happens | Resistance consequence |
|---|---|---|
| Base substitution | A single nucleotide is misread during replication and paired incorrectly | A single amino acid change in the target enzyme — for example a gyrA mutation giving quinolone resistance |
| Frame-shift (insertion or deletion) | The reading frame downstream is displaced | Commonly abolishes a protein, such as an outer membrane porin, which can alter the antimicrobial's route in |
| Missense in a regulatory gene | A promoter or a repressor is altered | Enzyme over-expression: more beta-lactamase, more efflux pump |
| Large deletion | A segment is lost | Loss of porins or of the O-antigen; loss of porin OmpF contributes to carbapenem resistance |
Watch outThe classic error — in examinations and in clinical reasoning alike — is saying that bacteria 'develop resistance in response to' an antibiotic. Variation is random and pre-existing; the antibiotic merely selects the variants that already had it. A resistant subpopulation exists before the first dose, which is why combination therapy in tuberculosis is not optional: the probability of a spontaneously double-resistant mutant is the product of the two rates, and the culture of 10^10 organisms in a cavitary lesion makes single-drug treatment a resistance factory.
Transformation: naked DNA taken up
Griffith's 1928 experiment is the founding observation of molecular biology. He injected mice with live avirulent rough (non-encapsulated) Streptococcus pneumoniae mixed with heat-killed smooth (encapsulated) pneumococci, and the mice died of virulent infection: something from the dead smooth cells had transformed the live rough ones into encapsulated organisms. Avery, MacLeod and McCarty later showed that substance was DNA.
Competence is physiological, not universal. Neisseria gonorrhoeae, Neisseria meningitidis, Streptococcus pneumoniae, Haemophilus influenzae and several other species become naturally competent and take up DNA that is species-typical — which is one route by which penicillin-resistant pneumococci acquire mosaic pbp genes encoding altered penicillin-binding proteins from related oral streptococci. In the laboratory, competence is induced artificially with calcium chloride and heat shock, which is how plasmids are introduced into E. coli for cloning.
NoteMosaic penicillin-binding protein genes in Streptococcus pneumoniae are the textbook example of transformation creating clinical resistance. Portions of pbp2b and pbp2x are replaced by sequences from commensal oral streptococci, the resulting PBPs bind penicillin poorly, and the strain is no longer inhibited by achievable drug concentrations while its peptidoglycan synthesis still functions.
Transduction: DNA delivered by a phage
Generalized transduction happens when a phage in lytic growth mistakenly packages a fragment of host chromosome instead of phage DNA, then injects it into the next cell it infects — possibly anywhere on the chromosome. Specialized transduction is a consequence of faulty excision of a temperate (lysogenic) phage: only the genes adjacent to the prophage attachment site are carried, and the best-studied examples involve bacteriophage lambda in E. coli.
| Mechanism | Vector | Which host genes move | Clinical significance |
|---|---|---|---|
| Generalized transduction | Virulent phage in the lytic cycle, packaging error | Any bacterial gene — random fragment | Moves chromosomal resistance determinants such as mutations in gyrase or porin genes between strains |
| Specialized transduction | Temperate phage such as lambda, defective excision | Only genes flanking the prophage integration site | The classic source of tox genes: phage beta carries the streptococcal pyrogenic exotoxin gene, and phage C1 carries the cholera toxin gene |
NoteBacteriophages and their gene cargo are why some organisms are only sometimes dangerous. Non-toxigenic strains of Vibrio cholerae become epidemic ones by acquiring the CTX phage, which encodes cholera toxin; non-toxigenic Corynebacterium diphtheriae becomes diphtheria only when lysogenised by the phage carrying the tox gene — the toxin is encoded by phage DNA, not by the bacterium's chromosome.
Conjugation: DNA transferred cell to cell
The paradigm is the F (fertility) plasmid of E. coli. It carries the tra operon, which builds the sex pilus and the transfer apparatus, and it replicates by rolling-circle replication — a nick at oriT is made, one strand is transferred into the recipient, and both cells synthesise the complementary strand. F+ cells convert F− recipients into F+ cells. When F integrates into the chromosome, the cell becomes an Hfr (high-frequency recombination) strain: it still transfers DNA, but chromosome genes are transferred first and in a fixed order, which is how the circular E. coli chromosome was originally mapped.
| Element | Genes carried | Effect on transfer |
|---|---|---|
| F plasmid (plasmid F) | tra operon, oriT, insertion sequences | Converts a recipient F− cell into F+; transfers only itself |
| F integrated into chromosome | Chromosome plus F | Hfr strain: transfers chromosomal genes in order, but rarely completes F transfer |
| R plasmid | Resistance genes plus transfer genes | Moves multiple resistance determinants in one event |
| F' (F-prime) | F carrying a piece of adjacent chromosomal DNA | Moves a specific chromosomal gene along with fertility |
NoteConjugation is not confined to laboratory E. coli. Vancomycin resistance in Enterococcus faecium is carried on the vanA operon, usually on a transposon on a conjugative plasmid, and it has moved between enterococcal strains and, more worryingly, been transferred into Staphylococcus aureus — producing vancomycin-resistant Staphylococcus aureus (VRSA), first reported in 1997. This is horizontal transfer happening in patients.
Transposons, integrons and the assembly of resistance
Genes do not only travel between cells; they also move within a cell. An insertion sequence is a short, self-transposing segment carrying only the gene for its own transposase and inverted terminal repeats. Two insertion sequences flanking a gene form a composite transposon, which can jump from chromosome to plasmid. Integrons are more sophisticated: they carry an integrase that captures resistance gene cassettes from the environment and lines them up in tandem behind a single promoter, so one recombination event can add several determinants at once.
| Mobile element | Signature structure | Resistance genes typically carried | Organism |
|---|---|---|---|
| Tn3-family transposon | Resolvase and transposase genes, flanked by inverted repeats | bla genes encoding beta-lactamases, including extended-spectrum and carbapenem-hydrolysing variants | Enterobacteriaceae, Pseudomonas |
| SCCmec cassette | A staphylococcal cassette chromosome inserting at a specific att site | mecA (altered PBP2a) plus ccr recombinases | MRSA |
| Class 1 integron | intI1 integrase, attI site, promoter, cassette array | dfr, aadA (sulphonamide and aminoglycoside resistance), qac (antiseptic tolerance) | Proteus, Pseudomonas, Acinetobacter |
| Composite transposon Tn5 | Two insertion sequences flanking a resistance region | kan (kanamycin) and npt (neomycin/streptomycin) genes | E. coli, Pseudomonas, the standard molecular-biology transposon |
Watch outTransposition within a cell is what converts an unstable plasmid-borne resistance into a permanent chromosomal trait. A bla gene carried on a transposon can hop into the chromosome, so the strain remains resistant even if every plasmid is cured. That is one reason a 'plasmid-free' revertant is not necessarily an antibiotic-susceptible revertant.
From genes to bedside: reading resistance genetics
| Resistance phenotype | Genetic basis | Laboratory appearance | What it changes clinically |
|---|---|---|---|
| MRSA | mecA on SCCmec, encoding PBP2a (PBP2a has low affinity for all beta-lactams) | Cefoxitin disc resistant, or positive mecA/ PBP2a latex agglutination | Use a glycopeptide, linezolid or daptomycin; never a beta-lactam, whatever the oxacillin disc seems to say |
| ESBL producer | blaCTX-M, blaTEM, blaSHV variants on plasmids, hydrolysing third-generation cephalosporins | Synergy between ceftazidime and clavulanate; susceptible to a carbapenem | Carbapenem for serious infection; avoid the cephalosporins |
| Carbapenemase producer (KPC, NDM, OXA-48, VIM, IMP) | Very broad beta-lactamases on plasmids, often with aminoglycoside and quinolone co-resistance | Meropenem resistant, modified carbapenem inactivation test positive | Colistin, tigecycline, newer combinations such as ceftazidime–avibactam |
| Vancomycin-resistant Enterococcus (VRE) | vanA (inducible resistance to both vancomycin and teicoplanin) or vanB (vancomycin only) | Resistant to both agents in a MIC test or E-test | Linezolid; and strict contact isolation, because the plasmid is transmissible |
| Inducible clindamycin resistance | erm gene, methylating the 50S ribosome | Erythromycin resistant, clindamycin 'susceptible' by disc but flattened (positive D-test) | Use erythromycin instead — clindamycin resistance is inducible and can emerge on therapy |
A surgical wound swab grows Staphylococcus aureus that is resistant to cefoxitin. The laboratory reports 'MRSA' and a susceptibility panel showing vancomycin MIC 1 mg/L (susceptible). The patient has a repeat culture three weeks later that grows the same strain, now with a vancomycin MIC of 4 mg/L and thickened colonies on vancomycin-containing agar. That is vancomycin-intermediate S. aureus (VISA), arising not by acquisition but by chromosomal mutation — a thickened cell wall that binds more drug. The lesson is that resistance is cumulative: mutation first, then transfer.
Genetic tools built on these same systems
The very machinery of resistance is the machinery of molecular microbiology. Restriction enzymes cut DNA at defined sequences, DNA ligase joins fragments, plasmids act as vectors into which a gene is inserted, and selectable markers — antibiotic resistance genes — identify the cells that took the plasmid up. That is why a cloning plasmid carries ampicillin resistance: it is a selection device, not a clinical problem, and it is also why resistance markers are now designed to be confined to the laboratory.
Genotype to phenotype: acquire a gene that changes a target or degrades a drug
blaCTX-M (plasmid, transmissible) -> extended-spectrum beta-lactamase -> ceftriaxone resistant
mecA (SCCmec cassette) -> altered PBP2a -> methicillin-resistant S. aureus
gyrA S83L (chromosomal, heritable) -> DNA gyrase target change -> ciprofloxacin resistant
vanA (transposon on plasmid) -> altered peptidoglycan terminus -> vancomycin and teicoplanin resistant
rpoB H526Y(chromosomal) -> altered RNA polymerase beta -> rifampicin resistant M. tuberculosisCommon mistakes
Watch outMixing up the mechanisms. Transformation is naked DNA with no vector. Transduction is DNA inside a phage. Conjugation is cell-to-cell contact with plasmid transfer. A question asking which mechanism relies on a bacteriophage is transduction, and one describing uptake of DNA from a lysed cell is transformation.
Watch outSaying that plasmids carry essential genes. Plasmids are accessory: a cell can be cured of them and survive, though often with a reduction in fitness and in resistance. Chromosomes carry the essential genome — mutation there is heritable and permanent, and recombination into the chromosome is the end state of a transferred gene.
Watch outBelieving a spontaneous mutation rate of 10^-9 means resistance is unlikely. In a bacterial population of 10^9, a mutant at that rate is certain to exist. Selection pressure then supplies it with a world cleared of competitors. This is why monotherapy of a large bacterial burden is the classic way to generate resistance.
Check your understanding
A patient with isoniazid-resistant tuberculosis receives rifampicin alone for nine months. Explain mechanistically why the bacilli are likely to emerge resistant.
Which horizontal mechanism would be most likely to move a penicillin-binding protein gene from an oral commensal streptococcus into Streptococcus pneumoniae, and why?
Why does the acquisition of a beta-lactamase transposon by a plasmid make resistance more durable than a plasmid alone?
Multiple choice
Which statement about the three classical mechanisms of horizontal gene transfer is correct?
Which is the most accurate account of how resistance to a new antibiotic arises in a bacterial population?
A Staphylococcus aureus isolate is resistant to cefoxitin but its oxacillin disc shows a 20 mm zone. What is the best interpretation?
Summary and next
Bacteria change their genomes vertically by mutation and horizontally by transformation, transduction and conjugation. Mutation is random and provides the raw variation; horizontal transfer is what turns a rare determinant into a population-wide and cross-species phenomenon. Plasmids, transposons and integrons assemble those determinants into compact cassettes, and integration into the chromosome makes them permanent — which is why MRSA, ESBL producers, carbapenemase producers and VRE are each, at root, one gene-cargo problem.
Next: antimicrobial susceptibility testing. You will see how the laboratory converts the genetics you have just met — the beta-lactamase, the altered PBP, the inducible erm — into a zone diameter or a minimum inhibitory concentration that a clinician can act on.