For patients with cancers driven by ATM kinase deficiency, a group that includes some breast, lung, and hematologic malignancies, the treatment landscape has been marked by a frustrating paradox. ATM-deficient tumors are inherently sensitive to certain classes of chemotherapy and targeted drugs, including topoisomerase I inhibitors and PARP inhibitors. Yet a subset of these tumors eventually develops resistance, with no clear molecular explanation.
A study published July 4 in Nature Communications by researchers at the University of Pennsylvania and Washington University in St. Louis identifies the mechanism behind this paradox, and in doing so, reveals a molecular switch that determines whether ATM-deficient cancers live or die in response to treatment.
The culprit is the BRCA1-A complex, a multi-protein assembly that interacts with the well-known tumor suppressor BRCA1. In ATM-deficient cells, the study shows, BRCA1-A enforces a restrictive chromatin state at damaged replication forks, the Y-shaped structures where DNA is copied during cell division. This restrictive state prevents the replication machinery from executing a process called fork reversal, a protective maneuver that normally allows cells to repair damaged DNA and survive.
“We found that BRCA1-A acts as a gatekeeper at damaged replication forks,” said senior author Roger A. Greenberg, professor of cancer biology at the University of Pennsylvania Perelman School of Medicine. “When ATM is missing or inhibited, BRCA1-A locks the fork in a vulnerable configuration. Remove BRCA1-A, and the fork can reverse, repair the damage, and the cell survives.”
The mechanism in detail
The finding rests on a chain of molecular events. When ATM is inhibited, either because the gene is mutated or because a drug blocks its activity, the cell triggers a combined SUMO (small ubiquitin-like modifier) and ubiquitin signaling cascade. These modifications recruit the BRCA1-A complex to sites of replication fork damage.
Once at the fork, BRCA1-A restricts the ability of nucleases and helicases to access the DNA. This creates what Greenberg describes as a “restrictive chromatin state”, compact, nuclease-resistant, and unable to undergo the structural remodeling needed for fork reversal.
Without fork reversal, the replication fork stalls irreversibly. The single-stranded DNA at the fork becomes a substrate for topoisomerase I inhibitors like irinotecan and topotecan, and the collapsed fork generates the kind of DNA damage that PARP inhibitors exploit. The cell becomes exquisitely sensitive to these drugs.
But if the BRCA1-A complex is missing or dysfunctional, through mutation, deletion, or epigenetic silencing, the chromatin at the damaged fork remains accessible. Nucleases can enter, resection occurs, and fork reversal proceeds normally. The cell repairs the damage and becomes drug-resistant.
The researchers demonstrated this directly using electron microscopy. In ATM-inhibited cells with intact BRCA1-A, replication forks were overwhelmingly in a stalled, non-reversed configuration. In cells where BRCA1-A was knocked out, the same ATM-inhibited background showed abundant fork reversal, the protective mechanism had been restored.
Clinical implications
The finding has immediate translational relevance. ATM is one of the most commonly mutated DNA damage response genes in cancer, with loss-of-function mutations found in up to 10% of breast cancers, 15% of lung adenocarcinomas, and a significant fraction of T-cell prolymphocytic leukemias. These tumors are often treated with DNA-damaging agents, but responses are variable.
The study suggests that BRCA1-A status, whether the complex is intact or disrupted, may predict which ATM-deficient tumors will respond to topoisomerase I inhibitors and PARP inhibitors, and which will be resistant.
“This gives us a biomarker hypothesis to test,” said first author Arindam Datta, a postdoctoral fellow in Greenberg’s lab. “If a tumor is ATM-deficient but also has a disrupted BRCA1-A complex, we would predict it will be resistant to these drugs. The fork reversal mechanism is intact. But if BRCA1-A is functional, the tumor should be sensitive.”
The resistance pathway itself, the loss of BRCA1-A coupled with XRCC4/Ligase 4-mediated alternative end-joining, suggests that combining PARP or Topo I inhibitors with agents that block non-homologous end joining might overcome the resistance. The authors demonstrated that loss of either BRCA1-A or XRCC4/Ligase 4 was sufficient to confer resistance in ATM-deficient cells, pointing to potential combination strategies.
A broader principle
Beyond the clinical implications, the study reveals a broader principle about how cells balance DNA repair pathway choices at replication forks. Fork reversal and fork resection are competing outcomes, and the cell must choose which path to take. BRCA1-A emerges as a key determinant of that choice, tipping the balance toward a restrictive, non-reversal state that leaves the fork vulnerable to certain types of chemotherapy.
The fact that this choice is regulated by chromatin accessibility, not just by the presence or absence of repair enzymes, adds to a growing body of evidence that the physical state of DNA at damage sites is as important as the repair proteins themselves.
“We think of DNA repair as a question of which enzymes are present, but this shows that whether the enzymes can even reach the DNA is equally critical,” Greenberg said. “BRCA1-A controls access. When it’s present, the fork is locked. When it’s gone, the fork is open for business.”
Source: Datta A, Jackson J, Morozov YI, Qiu J, Vindigni A, Greenberg RA. The BRCA1-A complex restricts replication fork reversal-dependent DNA repair in ATM deficient cells. Nature Communications (2026). DOI: 10.1038/s41467-026-75271-7