Back

Subthreshold perturbation of DNA replication induces a secretory response and a bystander effect in naïve human fibroblasts

Perdichizzi, B.; Cappiello, F.; Di Feo, F.; Le Pera, L.; Pagliuca, A.; Valenzisi, P.; Rosina, M.; Merlo, D.; Franchitto, A.; Pichierri, P.

2026-08-21 cell biology
10.64898/2026.08.20.745990 bioRxiv
Show abstract

Replication stress is a hallmark of cancer, where it drives DNA damage and genome instability. Yet subtle, subthreshold perturbations of DNA synthesis likely occur routinely in normal proliferating tissues, and their consequences for cell homeostasis are unknown. Using primary human fibroblasts, we show that doses of the DNA polymerase inhibitor aphidicolin, too low to engage the replication checkpoint, or produce detectable DNA breaks, nonetheless elicit low-level, ATM-dependent {gamma}H2AX phosphorylation uncoupled from overt damage. This near-silent perturbation reprograms gene expression, inducing replication-associated genes together with a discrete secretory programme dominated by matrix-remodelling proteases and matricellular factors. This output is not a senescence-associated secretory phenotype: the NF-{kappa}B/IL-1/IL-6 axis is co-ordinately repressed rather than induced, p53 target genes including CDKN1A are unchanged, and cells remain proliferative and non-senescent. Conditioned medium from exposed cells reproduces ATM-{gamma}H2AX activation in naive fibroblasts without DNA damage, defining a "perturbed-replication bystander effect" (PeRBE). PeRBE is ROS-independent and mediated by heat-labile, proteinaceous factors, and in recipient cells it induces an extracellular-matrix programme that culminates in increased collagen production, without loss of proliferative capacity. A perturbation invisible to every standard replication-stress assay therefore generates a transmissible, protein-borne signal that instructs fibrogenic matrix remodelling in cells that never experienced it.

Matching journals

The top 5 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.