The Colibactin Genotoxin Generates DNA Interstrand Cross-Links in Infected Cells Original paper

What was studied?

This study tested the mechanism of DNA damage caused by colibactin, a genotoxin made by pks island–positive Enterobacteriaceae, by asking whether colibactin directly creates DNA interstrand cross-links (ICLs) in host cells. The authors exposed purified double-stranded DNA and cultured human cells to live colibactin-producing bacteria and then measured DNA damage using denaturing gel electrophoresis, DNA renaturation behavior, and cellular DNA damage signaling. They also used bacterial mutants that block colibactin maturation and a purified resistance protein to confirm that mature colibactin, not inactive precursors, drives the observed DNA lesions.

Who was studied?

The study examined cultured human cells (HeLa) and purified extracellular DNA substrates as the main targets of damage, rather than patient samples. On the microbial side, the authors used a laboratory E. coli DH10B strain carrying the pks island on a bacterial artificial chromosome as the primary colibactin producer, alongside matched non-genotoxic controls and pathway mutants (including a clbP mutant that cannot activate colibactin from precolibactin). They also tested multiple clinical Enterobacteriaceae harboring the pks island to show that the same damage mechanism generalizes beyond the lab strain.

What were the most important findings?

The study showed that colibactin’s primary DNA lesion is interstrand cross-linking, which prevents DNA strands from separating under denaturing conditions and produces a characteristic “nonmigrating/shifted” DNA pattern similar to classic cross-linking agents like cisplatin. Adding extracellular DNA during infection reduced host-cell DNA damage, which supports the idea that DNA can “trap” colibactin, and the trapped DNA itself displayed ICL behavior, indicating direct covalent cross-link formation. Genetic and biochemical controls strengthened causality: cross-linking disappeared when the colibactin pathway was disrupted (including loss of ClbP-dependent maturation) and was blocked by adding ClbS, the specific colibactin resistance protein. In host cells, infection activated ATR-driven replication stress signaling and triggered Fanconi anemia pathway recruitment (including FANCD2 activation and foci), and blocking ATR or depleting FANCD2 reduced survival after exposure, indicating that ICL repair pathways are essential for tolerating colibactin injury.

What are the greatest implications of this study/ review?

For clinicians and microbiome-signature interpretation, this work makes colibactin risk more concrete by defining a specific lesion—DNA interstrand cross-links—that is both highly mutagenic and highly cytotoxic, and that can convert into double-strand breaks during repair. That mechanism aligns colibactin with well-known carcinogenic and chemotherapeutic cross-linking exposures, strengthening biological plausibility for its role in colorectal carcinogenesis when pks-positive bacteria persist near the mucosa. The findings also suggest translational levers: targeting colibactin maturation (for example the ClbP-dependent activation step) or exploiting ICL repair vulnerabilities could reduce harm without relying solely on broad antimicrobial approaches, while patients with impaired cross-link repair capacity may represent a higher-susceptibility context when pks-positive colonization occurs.