A global genomic analysis of Klebsiella pneumoniae from dogs and cats has identified extensive overlap with bacterial lineages circulating in people, including several high-risk antimicrobial-resistant clones. The study, based on more than 700 companion-animal genomes from 25 countries, does not demonstrate transmission between pets and humans, but it strengthens the case for including companion animals in One Health surveillance of antimicrobial resistance.
Published in Transboundary and Emerging Diseases, the study analysed 712 Klebsiella pneumoniae genomes obtained from domestic dogs and cats and compared them with 38,106 human-associated genomes. According to the authors, this is the first large-scale global analysis designed specifically to characterise the population structure of K. pneumoniae in companion animals and compare it with the strains circulating in people.
K. pneumoniae can colonise animals and humans without causing disease, but some strains are responsible for urinary tract infections, pneumonia, bloodstream infections and other serious conditions. The growing frequency of strains resistant to third-generation cephalosporins and carbapenems has made the bacterium a major target of antimicrobial-resistance surveillance.
Most pet isolates belonged to lineages also found in humans
The companion-animal dataset contained 263 different sequence types, highlighting considerable genetic diversity. However, several internationally recognised high-risk lineages were repeatedly identified, including ST307, ST11, ST15 and ST147. When the animal genomes were compared with the much larger human dataset, the overlap was striking.
71.1% of the sequence types identified in dogs and cats had also been detected in humans. More importantly, 87.2% of all companion-animal isolates belonged to sequence types already represented in the human population. In practical terms, 621 of the 712 animal isolates belonged to bacterial lineages also found in people.
This does not mean that pets acquired the bacteria from people, or that humans acquired them from pets. Sequence-type overlap demonstrates that the same bacterial populations circulate in both hosts, but it cannot establish the direction or even the occurrence of direct transmission.
That distinction is important. The study identifies a shared microbial landscape rather than a demonstrated pet-to-owner transmission pathway.
ESBL genes in more than four out of ten isolates
The antimicrobial-resistance findings were substantial. Among 706 isolates from 25 countries for which sufficient geographical information was available, 303 isolates !42.9%) carried extended-spectrum beta-lactamase genes. These genes can confer resistance to important beta-lactam antibiotics, including third-generation cephalosporins.
A further 98 isolates, or 13.9%, carried carbapenemase genes. Carbapenemases are of particular concern because they can compromise carbapenems, antibiotics frequently reserved for difficult-to-treat infections caused by multidrug-resistant Gram-negative bacteria.
The findings place dogs and cats within the wider epidemiology of clinically important resistance rather than treating companion-animal Klebsiella populations as separate from those encountered in human medicine.
Multidrug resistance particularly frequent in cats
A marked difference emerged between dogs and cats. 80.0% of cat-derived isolates were classified as multidrug resistant, compared with 56.3% of dog-derived isolates. The researchers attribute part of this difference to a greater representation of epidemic high-risk clones among the feline isolates, particularly ST147.
The result should not be interpreted as evidence that cats intrinsically constitute a greater AMR threat than dogs. The dataset was assembled from publicly available genomic sequences rather than from a standardised random survey of the pet population, and sampling patterns differ substantially between countries, veterinary hospitals and research programmes.
The authors themselves highlight uneven genomic sampling and incomplete clinical metadata as important limitations.
ST147 provides the strongest evidence of cross-host circulation
Among the lineages investigated, ST147 attracted particular attention.
ST147 is a globally distributed multidrug-resistant K. pneumoniae lineage already recognised in human healthcare. Detailed genomic analysis identified a closely related ST147 cluster containing isolates obtained from cats, dogs and humans. Some were separated by very small numbers of genomic differences, consistent with recent common ancestry and circulation of the same bacterial lineage across host populations.
The researchers also found a recurring plasmid profile within this ST147 population. Plasmids are mobile genetic elements capable of carrying antimicrobial-resistance genes between bacteria. Closely related cat-, dog- and human-associated ST147 genomes shared recurrent IncFIB(K), IncFII(K) and IncX3 plasmid replicons, together with similar multidrug-resistance determinants.
Again, these findings do not establish direct transfer between an animal and a human. They show that closely related resistant bacterial populations, and associated resistance elements, occur across both host groups.
Resistant bacteria were not confined to sick animals
One of the more important findings from a surveillance perspective is that multidrug-resistant K. pneumoniae was not restricted to samples collected from animals with clinical infections. Resistant strains were also found in colonisation-associated samples. This matters because apparently healthy animals carrying resistant bacteria may form part of a larger reservoir that remains invisible if surveillance focuses only on veterinary clinical infections.
Colonisation does not mean disease, nor does it mean that an animal will transmit the bacterium to its owner. But from an epidemiological perspective it allows resistant lineages to persist in populations and environments without producing an obvious clinical signal.
Companion animals remain a blind spot in AMR surveillance
Dogs and cats occupy a distinctive position in One Health epidemiology. Unlike livestock, they share homes, surfaces and daily physical contact with people. Yet antimicrobial-resistance surveillance in companion animals remains considerably less developed than surveillance in hospitals, food animals or the food chain.
The scale of this interface is considerable. The study notes that Europe has approximately 90 million dogs and 108 million cats, while the United States has an estimated 87.3 million owned dogs and 76.3 million owned cats.
Several countries are beginning to address this surveillance gap. In the UK, for example, the Veterinary Medicines Directorate is supporting a national pilot collecting faecal samples from healthy dogs and cats in households, veterinary practices and rescue centres to establish baseline levels of antimicrobial resistance. China has also operated the China AMR Surveillance Network for Pets (CARPet) since 2021.
The new genomic study suggests that such initiatives should not be considered peripheral to human AMR surveillance.
A One Health signal, not evidence that pets are infecting their owners
The distinction is central to interpreting the results. Finding the same sequence types in people and pets does not establish direct zoonotic transmission. Even the almost identical ST147 genomes could reflect acquisition from a common environmental or healthcare source, transmission through another host, or circulation through multiple interconnected reservoirs.
The study was based on previously deposited genome sequences and was not designed to reconstruct individual transmission events. Its importance lies elsewhere: companion animals are carrying some of the same internationally circulating high-risk K. pneumoniae clones and resistance determinants that are important in human medicine. That makes them relevant to antimicrobial-resistance surveillance even when the animals themselves are healthy.
The results should not be interpreted as evidence that pets are making their owners ill. Instead, they show that very closely related resistant bacterial populations are present in both animal and human hosts. For veterinary and public-health surveillance, the implication is straightforward: monitoring antimicrobial resistance only in humans, livestock and food-producing animals leaves part of the epidemiological picture unresolved.
The study adds genomic evidence that dogs and cats belong within the same One Health AMR landscape as humans, particularly when surveillance is aimed at identifying emerging high-risk clones before they become more widely established.
Study: Fordham SME, Sheridan E, Drobniewski F. Companion Animals Harbour Globally Circulating Human-Associated Klebsiella pneumoniae Lineages and High-Risk Antimicrobial Resistance Clones. Transboundary and Emerging Diseases. Published 28 August 2026. DOI: 10.1155/tbed/5383720.
Commentaires
No comments yet.
Sign in to comment