Back

SOS-mediated prophage induction constrains resistance evolution to DNA-damaging antibiotics

Zamora, A.; Pai, S. V.; Mears, K.; Rossine, F.; Owen, S.; Baym, M.

2026-06-24 microbiology
10.64898/2026.06.23.733845 bioRxiv
Show abstract

Most naturally occurring bacteria are lysogens, encoding one or more temperate phages (prophages) integrated into their genome. As prophages are induced by the bacterial SOS response, DNA-damaging antibiotics can trigger SOS-mediated prophage induction, where prophages undergo lytic replication and lyse their host, even at sub-inhibitory concentrations. This prophage-antibiotic synergy therefore sensitizes lysogenic hosts to DNA-damaging antibiotics. However, the mechanism by which prophage-induced sensitization affects the evolution of resistance against these agents is unclear. Here we show that ciprofloxacin-resistant lysogens arise less frequently but exhibit higher levels of resistance following selection. Whole-genome sequencing showed that increased lysogen resistance arose from selection towards mutations in drug targets, efflux pathways, and stress response regulators that reduce antibiotic efficacy or alter SOS induction. Consistent with this result, resistant lysogens exhibited a dampened SOS response, suggesting that prophage induction imposes an additional selective filter on their hosts by eliminating mutants that experience sufficient DNA damage to activate the SOS response. By contrast, prophage carriage had no effect on sensitivity or resistance evolution for antibiotics where DNA damage occurs downstream of the primary mechanism of action. Together, these findings indicate that prophage induction acts as an evolutionary bottleneck that restricts many resistance trajectories while favoring the emergence of rarer, large-effect mutations, potentially accelerating the evolution of high-level resistance.

Matching journals

The top 7 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.