BACKGROUND: Colibactin is a genotoxic secondary metabolite produced by selective Enterobacteriaceae strains that populate the mammalian intestine and produces specific mutational signature in human colonocytes. We recently demonstrated that an active synthetic colibactin (742) caused a specifc mutational pattern (SBS88/SBS44) in human cell lines compared to an inactive analog (746) (PMC10413156). However, the host pathways involved in colibactin response remain unclear. METHODS: We performed genome wide CRISPR/Cas9 knockout screens and RNA sequencing utilizing live pks⁺bacteria bacteria (pks⁺Klebsiella pneumoniae vs. ΔclbP K. pneumoniae) and synthetic colibactins. Bacterial screens were carried out by three rounds of infection at 50 MOI, and colibactin screens were carried out by exposing cells to a continuous sub-lethal dose (5 μM) for 12 days. RNA sequencing was performed 24 hours after infection or compound treatment. To validate candidate gene function, we generated single-target knockout or non-target controls (NTC) in HEK293t and HT-29 cells. We quantified changes in viability, G2/M cell cycle arrest, γH2AX foci formation, and colony formation after colibactin 742 treatment. RESULTS: From our genome-wide screens we identified 20 enriched genes with a MaGECK score > 2.0 in both screens, including proteasomal subunits (e.g., PSMG4, PSMD4), RNA processing factors (e.g., SF1, PRPF8), and RNA polymerase III (e.g., CRCP). To narrow down candidate genes for validation, we compared RNAseq data from intestinal epithelial cells infected with pks⁺K. pneumoniae or colibactin 742, identifying a significant dysregulation pathways related to proteasomal regulation of cell-cycle progression. Therefore, we sought to validate the role of PSMD4, identified as a colibactin sensitizing gene in both screens, in regulating colibactin response. We found that in HT-29 cells, PSMD4 knockout ameliorated reductions in cell viability (NTC, 70.2% vs. PSMD4^-/-, 91.0%; p=0.034) and G2/M cell cycle arrest after 742 treatment (NTC, 60.5% vs. PSMD4^-/-, 38.0%; p=0.019). Conversely, PSMD4 knockout did not affect the amount of γH2AX foci after 742 treatment (NTC, 32.1% vs. PSMD4^-/-, 49.5%). Consistent with these observations, PSMD4^-/-cells had a significantly higher colony formation rate than NTC cells after 742 treatment (NTC, 48 vs. PSMD4^-/-, 146 colonies per 5,000 cells), and 742 reduced colony size in NTC cells significantly more than in PSMD4^-/-cells (NTC, 56.1% vs. PSMD4^-/-, 83.0%; p<0.001). CONCLUSIONS: These findings suggest that the proteasome regulates cell-cycle arrest following colibactin-induced DNA damage, and that cells with deficiencies in these pathways may continue to replicate despite DNA damage, potentially increasing the risk of malignant transformation.

Christian Jobin

University of Florida

Appearances