Back

Incomplete killing by antimicrobial metals generates transient antibiotic tolerance via delayed recovery in Escherichia coli

Rosenberg, M.; Park, S.; Ivask, A.; Nordholt, N.; Schreiber, F.

2026-06-19 microbiology
10.64898/2026.06.19.733367 bioRxiv
Show abstract

Incomplete killing is common after application of disinfectants or use of antimicrobial surfaces, leaving behind bacterial survivors with poorly characterized physiology. Here, we investigated the effects of short biocidal exposure to the common antimicrobial metals copper and silver as metal ions in suspension or under semi-dry conditions on solid metallic surfaces used against Escherichia coli. Survivors exhibited delayed regrowth and increased tolerance to bactericidal antibiotics without changes in MIC, consistent with a transient, non-genetic tolerance phenotype. When the magnitude of killing by silver or copper was matched between exposure formats, antibiotic tolerance patterns converged across metals and exposure conditions, indicating that the survivor state is governed more strongly by the severity of the primary lethal stress than by metal identity or exposure format alone. Incremental copper exposure revealed a dose- dependent increase in delayed regrowth and antibiotic tolerance but reduced tolerance to subsequent copper exposure. Post-exposure antibiotic tolerance as well as copper hypersensitivity phenotypes reverted after regrowth. Together, this supports a model of injury-linked delayed recovery rather than generalized stress priming. These findings demonstrate that incomplete disinfection can reshape survivor physiology, transiently increasing antibiotic tolerance through combined effects of selective survival acting on pre-existing phenotypic heterogeneity and acute injury-linked delayed recovery.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.