Tier 4 — mechanistic

SIRT6 promotes DNA repair under stress by activating PARP1

Mao Z, Hine C, Tian X, Van Meter M, Au M, Vaidya A, Seluanov A, Gorbunova V

Science 2011 Volume 332, Issue 6036, Pages 1443–1446

Bibliography

DOI: 10.1126/science.1202723
PubMed: PMID 21680843
PubMed Central: PMC5472447
Funding: Grants from the National Institutes of Health (NIH) to V.G. and the Ellison Medical Foundation to V.G. and A.S.
Competing interests: The paper does not include an explicit competing-interests declaration in the manuscript text as retrieved. If a signed COI form was submitted to Science, it is not reproduced in the PMC full text. Populate as "None declared in the manuscript text" and revisit if a declaration surfaces in the journal's supplementary policy record.

Study snapshot

Design	Interventional in vitro biochemistry and mammalian cell culture study using GFP-based chromosomal reporter assays for NHEJ and HR, siRNA and knockout depletion, site-directed mutagenesis, chromatin immunoprecipitation, and in vitro mono-ADP-ribosylation and poly-ADP-ribosylation assays.
Model	hTERT-immortalised diploid human fibroblasts carrying chromosomally-integrated NHEJ or HR GFP reporters; SIRT6 wild-type and SIRT6-knockout mouse embryonic fibroblasts (MEFs); PARP1-knockout MEFs carrying integrated NHEJ reporter; DNA-PKcs-null MEFs.
Sample	In vitro assays run in triplicate (n=3 per experiment); DNA repair reporter assays quantified across 4–8 independent transfections per condition.
Intervention	SIRT1, SIRT2, SIRT6, or SIRT7 overexpression via transient transfection; SIRT6 catalytic-dead point mutants (S56Y, G60A, R65A); PARP1 mono-ADP-ribosylation-site mutants (D387A, E488A, E491A, K498A, K521A, K524A); oxidative stress with paraquat or H2O2 pre-treatment; DNA damage via I-SceI endonuclease, γ-irradiation, or neocarzinostatin; PARP1 inhibition with 3-ABA or PJ34.
Duration	Acute in vitro assays with time points from minutes (in vitro ADP-ribosylation) to 24 hours (ChIP, γH2AX foci clearance); reporter assays scored at 48–72 hours post-transfection.
Endpoints	NHEJ repair efficiency (GFP+ frequency in chromosomal reporter cells); HR repair efficiency (GFP+ frequency in chromosomal reporter cells); γH2AX foci clearance kinetics after paraquat treatment; SIRT6 recruitment to DNA breaks (ChIP at Alu elements and I-SceI-flanking sequences); 53BP1 and NBS1 recruitment to DNA damage sites; SIRT6–PARP1 physical interaction (co-immunoprecipitation); In vitro SIRT6 mono-ADP-ribosyltransferase activity; In vitro PARP1 poly-ADP-ribosylation activity (with and without SIRT6); Identification of PARP1 K521 as the SIRT6-ribosylation site; Neutral comet assay for DSB repair after neocarzinostatin

What the study showed, in plain terms

This is a mechanistic study that established how SIRT6 promotes DNA repair. The authors used cell-culture reporter systems where a single DNA double-strand break can be induced and its repair by either of the two main pathways (non-homologous end joining, NHEJ; or homologous recombination, HR) measured as green fluorescence.

Overexpressing SIRT6 stimulated both repair pathways, and the effect was much stronger when cells were first exposed to oxidative stress. SIRT6 was recruited to DNA break sites much faster in stressed cells than in unstressed ones. Neither SIRT1 nor SIRT2 had this effect. Depleting SIRT6 impaired repair. Both of SIRT6's enzyme activities — deacetylation and mono-ADP-ribosylation — were required for the DNA-repair-stimulating effect.

The team then identified the first in vivo target of SIRT6's mono-ADP-ribosylation activity: PARP1, a poly-ADP-ribose polymerase that itself is a master regulator of DNA repair. SIRT6 physically binds PARP1, mono-ADP-ribosylates it on a specific lysine residue (K521), and this modification activates PARP1's poly-ADP-ribosylation activity. Blocking PARP1 with pharmacological inhibitors abolished the SIRT6-mediated stimulation of DNA repair.

The paper's broader implication is that SIRT6 sits at the intersection of oxidative stress signalling and DNA damage response, integrating them. It is the mechanistic foundation for later work using SIRT6 activators, including MDL-800 and fucoidan, to enhance DNA repair capacity.

Key findings

Among the four nuclear-localised sirtuins tested (SIRT1, 2, 6, 7), only SIRT6 overexpression strongly stimulated both NHEJ (3.3-fold) and HR (3.4-fold) in human fibroblast reporter lines. SIRT7 gave a modest effect (~1.5- to 2.8-fold).
Under oxidative stress (paraquat pre-treatment), SIRT6 overexpression stimulated NHEJ 6.7-fold and HR 6-fold relative to stressed controls, and up to 16-fold relative to unstressed baseline. The sirtuin inhibitor nicotinamide abolished the effect.
SIRT6-knockout MEFs had 2.6-fold lower NHEJ than wild-type, widening to 4.3-fold under paraquat stress. siRNA depletion of SIRT6 in human fibroblasts reduced HR 2.2-fold under stress.
SIRT6 was recruited to DNA double-strand breaks (ChIP at Alu elements after γ-irradiation and at I-SceI-flanking sequences after endonuclease transfection). Under oxidative stress, an additional early wave of SIRT6 recruitment was detected within 30 minutes of DSB induction.
SIRT6 catalytic mutants S56Y (no activity), G60A (no mono-ADP-ribosylation), and R65A (no deacetylation) all failed to stimulate DSB repair, demonstrating that both enzymatic activities are required in vivo.
SIRT6 physically associates with PARP1 (co-immunoprecipitation, ethidium-bromide-resistant, indicating direct interaction rather than co-binding to DNA). The interaction increased after DNA damage.
SIRT6 mono-ADP-ribosylates PARP1 in vivo. Wild-type SIRT6 and the R65A mutant (mono-ADP-ribosylation only) stimulated PARP1 poly-ADP-ribosylation activity in vitro. The G60A mutant (deacetylation only) and the catalytic-dead S56Y did not.
Mutation of PARP1 lysine 521 alone was sufficient to abolish SIRT6-mediated stimulation of NHEJ in PARP1-null MEF reporter cells. Mutating the five other known PARP1 ribosylation sites had no effect, identifying K521 as the SIRT6 target residue.
PARP1 inhibitors (3-ABA, PJ34) abolished the SIRT6-mediated stimulation of both NHEJ and HR, confirming that PARP1 is the required downstream effector.
SIRT6 overexpression stimulated NHEJ in DNA-PKcs-null MEFs 1.7-fold, consistent with SIRT6 activating an alternative (DNA-PKcs-independent) NHEJ pathway via PARP1.

What this study can and cannot tell us

This is a mechanistic in vitro and cell-culture study. All findings are drawn from primary human fibroblasts, mouse embryonic fibroblasts, and biochemical assays with purified proteins. There is no whole-organism efficacy data and no evidence that pharmacologically activating SIRT6 in a living animal produces the DNA-repair benefits inferred from these experiments — later work (Roichman 2021, Biashad 2025) addresses that question.

The DNA damage models used (paraquat oxidative stress, I-SceI endonuclease cuts, γ-irradiation, neocarzinostatin) are acute, high-dose insults. They do not reproduce the low-level, chronic DNA damage that accumulates during ageing, and the paper does not directly link its findings to age-related repair decline.

Reporter assays measure a single defined DSB in a chromatinised context; endogenous DNA damage is more heterogeneous. The reporter readout scores whether repair restored a GFP-coding sequence, not the fidelity of that repair, so error-prone repair events cannot be distinguished from accurate ones.

The paper does not include an explicit competing-interests statement in the retrieved manuscript text.