On this page
SIRT6 sits at the top of your cell's DNA damage response. It arrives at broken DNA strands within seconds, drives three separate repair pathways, and controls how tightly your chromosomes stay wound. This is the honest mechanism — what SIRT6 actually does to your genome, why the activity declines as you age, and what the current pharmacological rescue evidence shows.
Why is SIRT6 essential for genome stability?
The clearest evidence for SIRT6's role in DNA repair comes from what happens when you take it away. In 2006, the Alt lab at Harvard engineered mice without any functional SIRT6. Those mice developed a rapid-ageing syndrome — spinal curvature, lymphopenia, subcutaneous fat loss, severe hypoglycaemia — and died at about four weeks old [1].
Cells from those mice had a distinctive defect. They could not efficiently repair single-base damage to their DNA — the kind of damage produced by ordinary oxidative stress. Their genomes accumulated mutations. Their chromosomes came apart. This is what an animal looks like when its DNA repair machinery loses one specific component.
The framing paper that pulled these findings into a coherent theory of ageing was Lombard 2008, from the same Harvard group two years later [2]. That review made the case — still the working model for the field today — that SIRT6 acts as a longevity enzyme through three intertwined mechanisms: DNA repair, insulin/IGF1-like signalling, and chromatin regulation. Every subsequent SIRT6 review builds on that framing. Our SIRT6 pillar article covers the enzyme's biochemistry in more depth. This article is about what happens at the DNA itself.
Which DNA repair pathways does SIRT6 support?
Your cells face three main classes of DNA damage. Each has its own repair pathway. SIRT6 is involved in all three.
Base excision repair handles the everyday damage. Oxidation from normal metabolism. Alkylation from environmental chemicals. Deamination that changes cytosine into uracil. These are single-base problems. SIRT6 supports base excision repair by stabilising PARP1 activity and by helping recruit the DNA glycosylases hMYH and hAPE1 to damage sites [2]. This is the specific pathway that fails in SIRT6-null mouse cells.
Nucleotide excision repair handles the bulkier damage — the kind caused by ultraviolet light or certain chemotherapies that create physical distortions in the DNA helix. SIRT6 supports this pathway by deacetylating DDB2 at lysines 35 and 77, activating its DNA-damage-recognition function. Chen 2025's comprehensive review consolidates the substrate list across all three pathways [8].
Double-strand break repair is the pathway that handles the most dangerous damage — both DNA strands broken at once. Unrepaired, a single double-strand break can kill a cell or trigger cancer-driving mutations. This is where SIRT6's contribution has grown the most in recent years, and where the rest of this article focuses.
How does SIRT6 detect DNA damage as a first responder?
For most of the SIRT6 field's history, SIRT6 was thought of as a helper. Something that arrived after other proteins had already found the damage. In 2020, the Toiber lab at Ben-Gurion University showed that this framing was wrong. SIRT6 is itself one of the earliest responders — arriving at broken DNA within seconds of damage, independently of any of the three previously known sensors [6].
The team used two experimental approaches. First, they filmed SIRT6 arriving at ultraviolet-laser-induced DNA breaks in live human cells. When they blocked the other three sensors — Ku80, MRE11, and PARP1 signalling — SIRT6 still showed up on schedule. Second, they purified SIRT6 in a test tube and measured how tightly it bound different DNA shapes. SIRT6 bound tightly to broken DNA ends, especially those with the single-strand overhangs characteristic of double-strand breaks. It did not bind unbroken circular DNA.
The structural explanation is elegant. SIRT6's protein core contains a narrow tunnel formed by six specific amino acids (A13, D63, H133, W188, D190, I219). This tunnel fits over the exposed end of one broken DNA strand. When both strands break at once, two SIRT6 molecules cap the two ends and pair up. They protect the broken ends from further degradation and simultaneously signal the cell to begin the damage response.
Once bound, SIRT6 recruits the master DNA damage kinase ATM. It drives the phosphorylation of histone H2AX — the classic "γH2AX" mark that spreads outward from every DNA break. And it brings in the specific repair factors for both major double-strand break pathways: Ku70, Ku80, 53BP1, and XRCC4 for non-homologous end joining; MRE11, NBS1, and BRCA1 for homologous recombination.
One finding from this paper has direct implications for supplementation. The sensor role does not require SIRT6's enzymatic activity. A catalytically dead SIRT6 mutant that cannot deacetylate or ADP-ribosylate anything still binds DNA, still recruits ATM, still drives the damage response. That means a small-molecule activator that only enhances SIRT6's enzymatic activity — which is what every current activator does — will not necessarily strengthen this arm of SIRT6's function.
How does SIRT6 activate PARP1 under oxidative stress?
The best-worked-out mechanism for SIRT6's role in double-strand break repair is its regulation of PARP1. This came from the Gorbunova and Seluanov labs at the University of Rochester in 2011.
PARP1 is one of the first enzymes to arrive at DNA breaks. It reads the damage and coats surrounding proteins in poly-ADP-ribose chains, which recruit the rest of the repair machinery. Under oxidative stress, PARP1 needs an extra push to work efficiently. SIRT6 provides it — physically binding PARP1 and attaching a single ADP-ribose group to a specific lysine on PARP1 called K521 [4]. This mono-ADP-ribosylation flips PARP1 into a more active state, amplifying its ability to poly-ADP-ribosylate itself and its neighbours.
Cells lacking SIRT6 have blunted PARP1 activation under oxidative stress. Both major double-strand break pathways — homologous recombination and non-homologous end joining — repair more slowly. Overexpressing SIRT6 has the opposite effect. This is the specific mechanism that supplement marketing tends to point at when it talks about "SIRT6 supports DNA repair". The framing is correct; the leap from cell biology to what a supplement actually does in your body is where the honest reading starts to hedge.
Does SIRT6 protect telomeres?
Your chromosomes end in repetitive DNA structures called telomeres. Telomeres protect the ends from being mistaken for double-strand breaks. When telomeres shorten or destabilise, cells trigger senescence or die.
SIRT6 deacetylates histone H3 at lysine 9 (H3K9) specifically at telomeric chromatin. This maintains the tight packaging that keeps telomeres stable. The Chua lab at Stanford showed in 2008 that SIRT6-null cells have telomere-specific defects — the H3K9 acetylation mark accumulates at chromosome ends, telomeres become disorganised, and cells show premature senescence [3]. This is a distinct function from SIRT6's role at genomic double-strand breaks, but the two functions share the same enzymatic activity — H3K9 deacetylation.
Why do longer-lived species have more effective SIRT6?
The most striking cross-species argument for SIRT6 as a DNA repair enzyme comes from Tian 2019, again from the Rochester group. They compared SIRT6 protein sequence and function across 18 rodent species — from short-lived mice to long-lived beavers and naked mole-rats.
SIRT6 from long-lived species was better at promoting double-strand break repair in cell assays. Five specific amino acid substitutions accounted for most of the difference [5]. When the team engineered a "beaver-like" version of SIRT6 into fruit flies, the flies lived longer than flies expressing the "mouse-like" version.
Notably, the substitutions affected mono-ADP-ribosylation activity more than deacetylation. That fits the picture from Mao 2011 — the ADP-ribose transfer to PARP1 is central to how SIRT6 drives double-strand break repair. It also tells us something about drug development. Nature has already tuned SIRT6 in the direction we would tune it pharmacologically. Enhancing the mono-ADP-ribosylation arm might matter more for longevity outcomes than enhancing the deacetylation arm. Our SIRT6 longevity research article covers the cross-species argument in more depth.
Why does DNA repair falter with age?
SIRT6 activity declines with age. So does the availability of NAD+, the cofactor SIRT6 needs to do any of its enzymatic work [8]. As both decline, the DNA repair machinery loses efficiency. Damage accumulates. Chromatin becomes less well-organised. Inflammatory responses to unrepaired damage rise. This is one of the mechanistic threads running through most modern theories of biological ageing [9].
There is a specific open question here that supplements do not currently address. Onn 2020 showed that SIRT6's DNA-sensor role does not depend on enzymatic activity. It depends on the tunnel structure that binds broken DNA ends. Whether that binding capacity declines with age the same way the enzymatic activity does is not known. If it declines less — or on a different timeline — then activator supplements that only enhance enzymatic activity may miss half of the age-related repair problem.
Can SIRT6 activators rescue the age-related decline?
Copp 2023 provides the first pharmacological evidence in human cells. The study used primary chondrocytes — the cartilage-making cells that degrade with age and drive osteoarthritis — from human donors of different ages. Older donors' cells had reduced SIRT6 activity, reduced DNA repair efficiency, and elevated senescence markers, exactly as the ageing model predicts [7].
Treatment with MDL-800, a small-molecule SIRT6 activator, restored DNA repair capacity in the aged cells to levels indistinguishable from younger donor cells. Senescence markers dropped. This is the first published demonstration that pharmacological SIRT6 activation can rescue an age-related DNA repair deficit in human tissue.
What Copp 2023 did not show — and what no published study has yet shown — is that this rescue translates to a whole human being. The tissue was chondrocytes. The rescue was in cell culture. The MDL-800 activator is preclinical only. No human clinical trial of any SIRT6 activator has been completed. Our SIRT6 gene therapy and small-molecule article covers the pipeline in detail.
One clarifying note. Some SERP sources conflate SIRT6 with Resveratrol. Resveratrol is a SIRT1 activator, not a SIRT6 activator. This confusion appears in commercial content and does not survive contact with the primary literature. If you are considering a supplement for SIRT6, you are considering a different molecular target.
What SIRT6 and DNA repair research still can't answer
Four gaps sit at the centre of the current story.
The first is whether the sensor arm and the enzymatic arm age at the same rate. This has direct consequences for what activator supplements can and cannot do. If the sensor arm declines slower — or if it declines faster — the intervention strategy differs.
The second is tissue coverage. The Copp 2023 human-cell rescue is in cartilage. Whether the same rescue works in brain, liver, muscle, or vascular tissue in humans is not known. Our SIRT6 in the brain article covers what is known for neurological tissue specifically.
The third is safety at the SIRT6 activator level. SIRT6 has a nuanced role in cancer — tumour-suppressor in some contexts, oncogene-supportive in others. Boosting SIRT6-mediated DNA repair long-term in adults with sub-clinical malignancies is not without theoretical risk. Our SIRT6 and cancer article covers this nuance in depth.
The fourth is the natural-supplement question. Fucoidan is the natural-product SIRT6 activator with the most in-vivo evidence so far, and DoNotAge's SIRT6 Activator is fucoidan-based. Whether fucoidan matches MDL-800's DNA-repair rescue in human cells has not been directly tested. Our fucoidan and SIRT6 article covers what is known.
Bottom line
SIRT6 is one of the most consequential DNA-repair regulators in your genome. It detects double-strand breaks within seconds, drives all three main repair pathways, maintains telomere stability, and evolves toward stronger activity in longer-lived species. The activity declines with age. Pharmacological rescue in human cartilage cells is now demonstrated.
What is not demonstrated is that a supplement raises SIRT6 activity in humans enough to matter clinically. What is not demonstrated is that the sensor arm — which does not depend on enzymatic activity — responds to any current activator. What is not demonstrated is safety at long horizons in adults with sub-clinical disease.
SIRT6-mediated DNA repair is a genuinely important biology. The class of interventions targeting it is worth understanding. Whether any specific supplement is worth taking is a separate question, and one where the honest answer today is that the evidence isn't there yet.