Tier 3 — preclinical

Genomic Instability and Aging-like Phenotype in the Absence of Mammalian SIRT6

Mostoslavsky R, Chua KF, Lombard DB, Pang WW, Fischer MR, Gellon L, Liu P, Mostoslavsky G, Franco S, Murphy MM, Mills KD, Patel P, Hsu JT, Hong AL, Ford E, Cheng HL, Kennedy C, Nunez N, Bronson R, Frendewey D, Auerbach W, Valenzuela D, Karow M, Hottiger MO, Hursting S, Barrett JC, Guarente L, Mulligan R, Demple B, Yancopoulos GD, Alt FW

Cell 2006 Volume 124, Issue 2, pages 315–329

Bibliography

DOI: 10.1016/j.cell.2005.11.044
PubMed: PMID 16439206
Funding: This work was supported by an Ellison Senior Scholar Award (to F.W.A.), NIH grants (to F.W.A. and B.D.), a Long-Term Fellowship of the Human Frontier Science Program and a Senior Post-doctoral Fellowship from The Leukemia and Lymphoma Society (to R.M.), a Pfizer Post-doctoral Fellowship in Immunology/Rheumatology (to K.F.C.), an EMBO Long Term Fellowship (to S.F.), and an NIA/NIH KO8 award (to D.B.L). F.W.A. is an investigator of the Howard Hughes Medical Institute.
Competing interests: Some of the authors are employees and shareholders of Regeneron and have financial interests related to this work.

Study snapshot

Design	Constitutive Sirt6 knockout mouse generation via LacZ knock-in replacing exons 1-6; characterisation of survival, growth, tissue histology, metabolic parameters, and lymphocyte populations. In vitro assays on SIRT6-deficient MEFs and ES cells for DNA damage sensitivity, cell cycle checkpoints, NHEJ, DSB repair, and base excision repair (BER); bone marrow transplantation to test cell-autonomous vs systemic defects.
Model	SIRT6-/- constitutive knockout mice on mixed 129SvJ/C57BL6/J background with WT littermate controls. SIRT6-/- and WT primary mouse embryonic fibroblasts (MEFs) and embryonic stem (ES) cells. XRCC4-/- MEFs and XPF-/- MEFs as positive controls for DNA repair defects. HT1080 human fibrosarcoma cells for subcellular localisation studies.
Sample	Kaplan-Meier survival: SIRT6-/- n=23, SIRT6+/- n=148, SIRT6+/+ n=86. Bone mineral density: 5 mice per genotype. Serum IGF-1 and glucose: 10 mice per genotype. MEF metaphase analysis: 96 WT and 96 SIRT6-/- metaphases scored. ES cell Q-FISH: 50 metaphases per genotype across multiple lines.
Intervention	Genetic loss-of-function (constitutive SIRT6 knockout). In vitro DNA-damaging agents included ionising radiation (0-4 Gy), UV, methyl-methane sulphonate (MMS, 0-600 μM), and hydrogen peroxide (H2O2, 0-0.6 mM). Rescue experiments used retroviral reconstitution with wild-type or catalytically inactive FLAG-SIRT6, and overexpression of the Polβ dRP lyase domain.
Duration	SIRT6-/- mice were normal at birth, developed reduced body size within the first two weeks, underwent acute degeneration at 3 weeks, and invariably died at approximately postnatal day 24 (roughly 4 weeks of age).
Endpoints	Survival (Kaplan-Meier); Body size, subcutaneous fat, lordokyphosis, and general condition; Bone mineral density by dual-energy X-ray absorptiometry; Lymphocyte populations (CD4+CD8+ thymocytes, splenic B and T cells) and apoptosis by annexin V; Serum IGF-1 and serum glucose; MEF and ES cell proliferation and S-phase fraction by BrdU incorporation; Cell survival after ionising radiation, UV, MMS, and H2O2 exposure; Genomic instability by DAPI, spectral karyotyping (SKY), and Q-FISH metaphase analysis; G1/S and G2/M cell cycle checkpoint integrity; NHEJ activity via V(D)J recombination substrate assay and chromosomal DSB repair by pulsed-field gel electrophoresis; γH2AX foci formation and clearance; Base excision repair sensitivity phenotype and rescue by Polβ dRP lyase domain; Subcellular localisation of SIRT6 by immunostaining and biochemical fractionation

What the study showed, in plain terms

This 2006 paper in Cell is the study that first characterised what SIRT6 does in mammals. Before it, SIRT6 was one of seven mammalian sirtuins with almost no known function. The Alt laboratory at Harvard generated the first SIRT6 knockout mice — animals with the SIRT6 gene fully deleted — and asked what happens without it.

The findings were striking. Mice born without SIRT6 looked normal for the first two weeks of life, then abruptly deteriorated. By three weeks of age they had lost subcutaneous fat, developed spinal curvature, become severely lymphopenic (loss of immune cells), and shown severe metabolic disruption including collapsing blood glucose. All of them died at around four weeks. At the cellular level, SIRT6-lacking cells sat in chromatin inside the nucleus and were required for a specific form of DNA repair called base excision repair — the pathway that fixes the everyday DNA damage caused by oxidation and other endogenous stresses. Cells without SIRT6 showed dramatic genomic instability, with chromosome fragments, translocations, and gaps far above normal rates.

Because base excision repair capacity declines during normal aging, and because the SIRT6-null mice showed a syndrome resembling accelerated aging, this paper established the central hypothesis that has driven the SIRT6 field ever since: SIRT6 helps preserve genome stability, and losing SIRT6 activity — whether through mutation or through the natural decline that comes with age — allows DNA damage to accumulate and drives aging-related pathology. Every subsequent SIRT6 activator study, including the fucoidan work and the small-molecule chemistry now in development, traces its rationale back to this paper.

Key findings

SIRT6 is a nuclear, chromatin-associated protein expressed across most mouse tissues, with particularly high levels in thymus, skeletal muscle, and brain.
SIRT6-/- mice were born at Mendelian frequency, appeared normal for two weeks, then underwent acute degeneration and died at approximately postnatal day 24.
By three weeks of age, SIRT6-/- mice showed profound lymphopenia (~50-fold reduction in CD4+CD8+ thymocytes and ~10-fold reduction in splenic lymphocytes), loss of subcutaneous fat, lordokyphosis, colitis, and ~30% reduction in bone mineral density.
Serum IGF-1 levels were severely reduced in SIRT6-/- mice — lower than in liver-specific IGF-1 knockout mice — and serum glucose collapsed from normal on day 12 to the limit of detection by day 24.
SIRT6-/- MEFs and ES cells proliferated more slowly and had a reduced S-phase fraction compared with wild-type controls.
SIRT6-/- cells showed markedly increased sensitivity to ionising radiation, MMS, and H2O2 — consistent with a base excision repair (BER) defect — but normal sensitivity to UV, which is repaired by nucleotide excision repair.
SIRT6-/- MEFs showed significantly elevated chromosomal aberrations (38% abnormal metaphases vs 7% in WT), including fragmented chromosomes, translocations, and gaps.
Cell cycle checkpoints (G1/S and G2/M) and double-strand break repair (NHEJ and DSB) were intact in SIRT6-/- cells, localising the repair defect specifically to BER.
The DNA damage sensitivity of SIRT6-/- MEFs was rescued by reconstitution with wild-type SIRT6, but not by a catalytically inactive SIRT6 mutant, demonstrating enzymatic activity is required.
Overexpression of the Polβ dRP lyase domain (the rate-limiting step in BER) rescued the DNA damage sensitivity of SIRT6-/- cells, placing SIRT6 upstream of this reaction — likely by relieving chromatin-imposed constraints on BER.
Bone marrow transplantation showed the lymphocyte defect was non-cell-autonomous, reflecting systemic changes (likely the collapse in circulating IGF-1) rather than an intrinsic haematopoietic defect.

What this study can and cannot tell us

All in vivo evidence is in mice; no human data are presented. Human relevance was inferred from later work.
Constitutive knockout mice die at approximately four weeks, so the study captures a developmental-to-early-postnatal window rather than adult or aged biology. Distinguishing developmental defects from degenerative processes was not always possible.
Bone mineralisation is still developing in mice at this age, so the observed osteopenia may reflect impaired mineralisation rather than degeneration of established bone.
The molecular mechanism by which SIRT6 promotes BER was not fully established. The authors could not detect a direct physical interaction between SIRT6 and BER factors, and Polβ acetylation levels were unchanged in SIRT6-/- cells, leaving the mechanism as indirect chromatin regulation. Later work has extended this to the SNF2H recruitment and H3K56ac model.
This is a loss-of-function study — it establishes what happens when SIRT6 is absent, not what happens when SIRT6 is activated. Its relevance to the SIRT6 activator hypothesis is foundational rather than direct.
Some authors were employees and shareholders of Regeneron Pharmaceuticals at the time of publication. This is disclosed but should be noted when interpreting the work.