--- id: PRG-0025 title: Too Much Repair Is A Kind Of Damage kicker: the instrument, the genome, dosage captured: 2026-06-20T14:30:00Z status: open author: Ines Hargrove source: https://www.sciencedaily.com/releases/2026/06/260619101349.htm summary: A DNA-repair gene long counted among the protectors, EXO1, turns destructive when a cell makes too much of it, cutting sequence it should leave alone. The finding is read as a cancer weakness. The deeper lesson is that fidelity is a dose, not a virtue. tags: [the record, permanence, custody, capture, memory] sealAt: 2026-07-20T14:30:00Z --- A cell keeps one permanent record, the genome, and runs a repair crew whose whole job is to keep that record accurate. Scientists have found that a gene called EXO1, normally counted among the protectors, turns destructive when a cell makes too much of it. EXO1 is part of the repair machinery: a molecular blade that trims damaged DNA so the correct sequence can be restored. At ordinary levels it is a custodian of the text. Overproduced, it begins cutting sequence it was never meant to touch, and the same tool that preserves the record starts shredding it. The result is being framed as a cancer weakness, because a tumor pushed past that threshold destroys its own genome. The deeper finding is about dosage. The instrument that maintains a record is never neutral toward it. At the wrong dose, the proofreader becomes the vandal. ## The dose makes the keeper Trace what EXO1 actually does. DNA takes damage constantly, and repair depends on cutting the bad stretch out before rebuilding it from the intact copy. EXO1 is one of the blades that makes that cut. The cut is necessary and the cut is also a wound. Every repair is a controlled act of damage in the service of fidelity. At calibrated levels the wounds are small, targeted, and sealed. Raise the level and the blade does not grow smarter, only busier. It cuts more, including where there was nothing to fix, and the cell accumulates breaks faster than it can close them. The repair function and the damage function are the same activity, separated only by how much of it is happening. There is no clean version of keeping the record. There is only the right amount of cutting. This is why the same enzyme reads as hero in one paper and liability in another. It was never one or the other. It is a rate, and the meaning of the rate depends entirely on the dose. A measurement that called EXO1 protective was true at the level it was measured, and would have been false one threshold up. The instrument did not change. The amount of it did, and the amount is the whole story. > Every system that keeps a record has to cut something to keep it accurate. The danger begins the moment it is rewarded for cutting more. The habit of sorting genes into protectors and threats is a convenience of language, not a fact of biology. A gene is a capacity, run at some level, in some context. EXO1 is the clean case because its benefit and its harm are literally the same molecular act at two doses. Most genes are this too, less obviously. The labels say more about the experiment that named them than about the gene. The therapeutic idea follows directly and is genuinely useful: a cancer cell already running its repair machinery hot can be tipped over the threshold where repair becomes self-destruction, killing the tumor with its own custodian. That is worth pursuing. The idea worth keeping past this one result is plainer. Fidelity is a balance struck moment to moment between damage and repair, never a substance a cell simply has more of, and the crew hired to protect the record is, at every moment, also the thing most able to destroy it. The keeper of the record holds the blade. That was always true. This gene only made the dose visible.