Salmonella Typhimurium is among the leading causes of foodborne gastroenteritis worldwide, and its monophasic variant—distinguished serologically by the absence of one of its two flagellar phases—has surged in prevalence across Europe, the Americas and Asia over recent decades. The lineage, which belongs to sequence type 34 and typically displays the antigenic formula 4,[5],12:i:-, has repeatedly been implicated in outbreaks linked to pork, poultry and processed foods. Yet despite its clear epidemiological importance, the genetic determinants that allowed this clone to outcompete its biphasic ancestor, the previously prevalent ST19 lineage, had remained uncertain. The new study set out to resolve that question at unprecedented scale.

The team’s comparative analysis of tens of thousands of genomes revealed a striking pattern: the expanding ST34 clone carried a frameshift mutation in dinB, the gene encoding a translesion DNA synthesis polymerase also known as DNA polymerase IV. The mutation, a deletion of a single thymine residue within a homopolymeric tract of seven thymidines near the 5′ end of the gene, introduces a premature stop codon that truncates the 39.7-kilodalton DinB protein to a nonfunctional fragment of roughly 1.9 kilodaltons. Because DinB is normally involved in bypassing damaged DNA templates and in generating genetic diversity under stress, its loss initially appears counterintuitive for a pathogen on the rise.

Functional experiments, however, demonstrated that the crippled polymerase has far-reaching downstream consequences. Transcriptomic profiling showed that the frameshift dinB mutation rewires the expression of flagellar genes, altering the balance between the two phase-one and phase-two flagellin systems that Salmonella uses to evade host immune detection. Specifically, the mutation led to transcriptional changes that favored retention of fliC, which encodes phase-one flagellin, and set the stage for the subsequent loss of fljB, the gene encoding phase-two flagellin. Western blot analyses confirmed that dinB-mutant strains displayed an increased proportion of cells expressing FliC, while in vitro motility assays showed that swimming ability was preserved despite the underlying genetic upheaval.

The physiological consequence of this reprogramming became apparent in infection models. Using transposon insertion sequencing and in vivo competition experiments in mice, the researchers found that the requirements for successful gut colonization shifted once fljB was lost. Genes that were dispensable, or even costly, in the ancestral background became essential or advantageous in the mutant context, and vice versa. The frameshift dinB strain carrying the fljB deletion outcompeted the previously prevalent biphasic ST19 lineage in the intestine, indicating that the combination of the polymerase defect and flagellin loss conferred a measurable competitive advantage within the host environment.

To move beyond correlation, the authors built a temporal evolutionary model that simulated the dynamics of four competing lineages: biphasic ST19, biphasic ST34, ST34 carrying the frameshift dinB, and monophasic ST34 carrying both the frameshift and the fljB deletion. The modeling demonstrated that the chronological order of the events matters critically. The dinB mutation had to precede the loss of fljB for the monophasic clone to expand as observed, and the simulated selection coefficients derived from the model were quantitatively consistent with selection estimates computed from real-world changes in genotype frequencies among human clinical isolates. Sensitivity analyses confirmed that the conclusion held regardless of the initial frequency of ST34, the presence or absence of spatial population structure, or the precise metric used to quantify selection.

This ordered dependency is what the researchers describe as a stepwise pathoadaptation model: a sequence of gene-loss events in which each deletion reshapes the genetic background so that the next loss becomes beneficial rather than deleterious. The concept echoes a growing recognition in microbial evolution that genome reduction can be a creative force. Recent work on Vibrio parahaemolyticus, for example, showed that wave succession within a pandemic clone was likewise driven by gene loss, and earlier studies documented how monophasic Salmonella Typhimurium microevolved during a prolonged epidemic in the United Kingdom. The new study, however, is notable for linking the ecological success of a zoonotic lineage to a specific, mechanistically validated chain of molecular events.

The broader implications extend to public health forecasting. If the emergence of major bacterial clones can depend on relatively simple, ordered mutational steps, then surveillance systems that monitor homopolymeric tracts and other mutation-prone loci may gain predictive power over which lineages are poised to expand. The study’s data resources—including genomic data deposited at the China National GeneBank Database, Enterobase and NCBI, transcriptomic data under accession GSE280131, proteomic data via PRIDE, and transposon-sequencing reads under BioProject PRJNA1347559—provide a substantial foundation for such efforts. The mathematical model code has also been made publicly available on GitHub, enabling other groups to test the generality of the sequential gene-loss framework in other pathogens.

At the same time, the work underscores how little of bacterial adaptation is driven by gene acquisition alone. Monophasic ST34 also carries traits such as copper resistance and heavy-metal tolerance that have been implicated in its persistence in livestock production environments, and antimicrobial resistance remains a defining feature of many circulating clones. But the new analysis places gene loss at the center of the lineage’s rise, reframing flagellar phase variation—a long-studied quirk of Salmonella biology—as a decisive axis of clonal competition. As the authors note, understanding the genetic factors facilitating the emergence of infectious diseases is critical, and this study delivers a concrete mechanistic answer for one of the most consequential zoonotic pathogens of the modern food system. The lesson is uncomfortable but clarifying: for bacteria, sometimes less is more, and the road to pandemic status can be paved by the genes a pathogen leaves behind.

Subject of Research: Stepwise gene loss driving the clonal expansion of monophasic Salmonella Typhimurium ST34

Article Title: Sequential gene loss promotes expansion of monophasic Salmonella Typhimurium ST34

Article References: Wu, X., Wang, T., Jia, C., Zhou, H., Li, Y., Baker, S., & Yue, M. (2026). Sequential gene loss promotes expansion of monophasic Salmonella Typhimurium ST34. Nature Microbiology. https://doi.org/10.1038/s41564-026-02483-4

Image Credits: AI Generated

DOI: 10.1038/s41564-026-02483-4

Keywords: Salmonella Typhimurium, monophasic ST34, gene loss, dinB, flagellin, fljB, pathoadaptation, bacterial evolution, genomic epidemiology, foodborne pathogen, zoonotic disease, Nature Microbiology

Integrons are genetic platforms that bacteria use to capture, stockpile and rearrange small mobile elements called gene cassettes. Each cassette typically carries a single gene and its own recombination site, and the cassettes sit in arrays that can be shuffled by an integrase enzyme in response to stress. This architecture is famously responsible for the rapid spread of antibiotic resistance genes among pathogens, but the same machinery also represents an evolutionary archive of functions that bacteria have recruited over millions of years, from metabolic enzymes to toxins and defensive systems. Until now, most efforts to mine this archive have relied on laborious one-cassette-at-a-time approaches or on sequence-based predictions that say little about what a gene actually does.

The new study, led by Filipa Trigo da Roza and José Antonio Escudero of the Universidad Complutense de Madrid, together with colleagues including Melanie Blokesch of the École Polytechnique Fédérale de Lausanne, introduces two tools with evocative names: the cassette gatherer and the cassette hunter. Both exploit a clever piece of molecular engineering in which a class 1 integron recombination site, known as attI1, is embedded inside a gene that acts as a counterselection marker. In the plasmid-based version, the attI1 site was inserted into the ccdB toxin gene from Vibrio fischeri; in the chromosomal version, it was placed inside the sacB gene from Bacillus subtilis.

The logic of the system is elegantly simple. In its empty state, the disrupted toxin gene kills or prevents growth of the host bacterium under selective conditions. But when the integrase catalyses the capture of a gene cassette at the embedded attI1 site, the cassette restores the reading frame of the marker, inactivating the counterselection and allowing the cell to survive. Only cells that have successfully captured a cassette form colonies, which means the selection is entirely independent of the sequence or predicted function of the captured gene. This sequence- and function-independence is what distinguishes the approach from earlier methods that depended on PCR primers or prior knowledge of cassette boundaries.

The researchers deployed the plasmid-based cassette gatherer and the chromosomal cassette hunter in a naturally competent strain of Vibrio cholerae from which the native superintegron had been removed. Taking advantage of the bacterium’s ability to take up DNA from its environment, a capacity triggered by growth on chitin, the team could deliver genomic libraries directly into cells where the engineered integron machinery awaited them. When applied to a panel of Vibrio species, including V. cholerae, V. vulnificus, V. mimicus and V. parahaemolyticus, the tools recovered hundreds of single cassettes per assay with more than 99 percent specificity, a capture rate that dwarfs what conventional cloning strategies could achieve.

High-throughput sequencing of the resulting libraries confirmed that the recovered cassettes faithfully represented the diversity of the source integrons. Correlation analyses showed that the abundance of each cassette in the recovered pools tracked reproducibly with its representation in the starting material, and that the same cassette repertoires were recovered by both the plasmid-based and chromosomal versions of the tool. The data also revealed the diversity of functions hidden in these arrays, with many cassettes encoding proteins of unknown function, a reminder of how much uncharacterised biology remains buried in bacterial genomes.

To demonstrate the discovery power of the approach, the researchers turned their cassette libraries against two very different bacteriophages: ICP2, a vibriophage that preys on pandemic V. cholerae, and the classic Escherichia coli phage T4. Screens of the recovered cassettes identified nine distinct phage-defence systems, five of which had never been described before. The result builds on a series of recent studies showing that mobile integrons and sedentary chromosomal integrons act as biobanks of anti-phage defence, and it provides the first general-purpose pipeline for converting that observation into a systematic, high-throughput inventory of defence genes.

The significance of the advance extends well beyond phage defence. Because integron cassettes are exchanged across bacterial lineages through horizontal gene transfer, they constitute a naturally curated collection of genes that have passed repeated tests of utility in diverse cellular contexts. Gene cassette PCR, developed two decades ago, allowed researchers to amplify cassette boundaries from environmental DNA, but the products were mixtures that resisted clean isolation. Bioinformatic surveys, including comprehensive scans of metagenomes, have catalogued millions of predicted cassettes, yet prediction alone cannot assign function. The gatherer and hunter tools close this gap by pairing unbiased physical recovery of individual cassettes with immediate amenability to functional screens, whether for antibiotic resistance, metabolic activities, antimicrobial compounds or industrial enzymes.

The technical groundwork for the study drew on decades of integron biology, from the discovery of the distinctive V. cholerae superintegron in 1998 to detailed dissections of how attC recombination sites fold into single-stranded hairpins that guide strand selection during recombination. The team also engineered the recipient strain to optimise natural transformation, deleting extracellular nucleases and tuning competence regulators so that incoming genomic DNA could recombine efficiently into the capture platform. Structural predictions generated with AlphaFold3 guided the placement of the attI1 site inside the counterselection markers, minimising disruption of protein folding while preserving the lethal phenotype needed for stringent selection.

The tools, their datasets and the analysis scripts have been made available through Zenodo and GitHub, and the underlying strains are covered by patent filings, signalling likely commercial interest in what amounts to a programmable gene-discovery platform. As sequencing continues to reveal integron cassettes in environments ranging from soil and ocean to the human gut, the ability to recover and test those genes at scale transforms a passive cataloguing exercise into an active search for function. For a field that has spent forty years documenting integrons as agents of bacterial evolution, the new work offers something rarer: a way to read, one cassette at a time, the full library of tricks that bacteria have been collecting all along.

The choice of Vibrio cholerae as the engineering chassis reflects the deep historical connection between integron research and this organism. The massive superintegron of V. cholerae, first described in 1998, carries well over a hundred cassettes and remains the archetype of the sedentary chromosomal integrons found across Vibrionaceae. Because the integrase of class 1 integrons and the V. cholerae superintegron integrase share overlapping recombination specificities at attC sites, the engineered platform can in principle process cassettes arriving from a wide range of donor integrons, which is precisely what the cross-species recovery experiments demonstrated.

The phage-defence findings also fit into a rapidly consolidating picture. Within the past two years, independent teams have reported that mobile integrons in clinical settings encode anti-phage systems, that sedentary chromosomal integrons function as biobanks of defence genes, and that V. parahaemolyticus integrons are particularly rich in such systems. The nine systems identified here, five of them entirely new, suggest that this enrichment is not a peculiarity of any single lineage but a general property of integron arrays, plausibly reflecting the intense phage pressure experienced by bacteria in aquatic environments where cassettes are most actively exchanged.

Another dimension worth noting concerns expression. Cassette arrays are transcribed from a single promoter positioned in the integrase gene region, and the translation rate of upstream cassettes shapes the expression of those downstream, meaning that a captured gene’s activity depends heavily on its position in the array. By recovering cassettes as individual entities, the new tools sidestep this positional context entirely, allowing each gene to be assayed under standardised conditions. This decoupling of capture from native expression is likely to be important for screens targeting enzymatic or antimicrobial activities that may be silent or weakly expressed in the donor organism.

The environmental dimension is equally significant. Most integron cassettes on Earth reside in so-called environmental integrons, which are not associated with mobile elements or clinical resistance and remain almost entirely uncharacterised. Extending the gatherer and hunter workflow to metagenomic DNA from sediments, biofilms or wastewater could grant functional access to this reservoir, complementing sequence-similarity approaches that struggle with the high proportion of novel genes. As antimicrobial resistance continues to mobilise cassettes into pathogens, understanding what these elements normally do in their native hosts may prove as consequential as the biotechnological applications that motivated the work.

Subject of Research: High-throughput recovery of integron gene cassettes for functional gene discovery and phage-defence screening.

Article Title: High-throughput recovery of integron cassettes for gene discovery screens

Article References: Trigo da Roza, F., Carvalho, A., Prieto, A., Blanco, P., Vergara, E., López-Igual, R., Redrejo-Rodríguez, M., Blokesch, M., & Escudero, J. A. (2026). High-throughput recovery of integron cassettes for gene discovery screens. Nature Microbiology. https://doi.org/10.1038/s41564-026-02474-5

Image Credits: AI Generated

DOI: 10.1038/s41564-026-02474-5

Keywords: integrons, gene cassettes, phage defence, antibiotic resistance, Vibrio cholerae, horizontal gene transfer, natural transformation, gene discovery, integrase, biotechnology, microbiology, bacterial evolution