Tier 4 — mechanistic

Structure and biochemical functions of SIRT6

Patricia W. Pan, Jessica L. Feldman, Mark K. Devries, Aiping Dong, Aled M. Edwards, John M. Denu
Journal of Biological Chemistry 2011 Volume 286, issue 16, pages 14575–14587

Bibliography

PubMed
PMID 21362626
PubMed Central
PMC3077655
DOI
10.1074/jbc.M111.218990
Funding
National Institutes of Health Grant GM065386 (to J.M.D.). Also supported by the Structural Genomics Consortium (registered charity 1097737), which received funds from the Canadian Institutes for Health Research, the Canadian Foundation for Innovation, Genome Canada through the Ontario Genomics Institute, GlaxoSmithKline, Karolinska Institutet, the Knut and Alice Wallenberg Foundation, the Ontario Innovation Trust, the Ontario Ministry for Research and Innovation, Merck & Co., the Novartis Research Foundation, the Swedish Agency for Innovation Systems, the Swedish Foundation for Strategic Research, and the Wellcome Trust.
Competing interests
Not stated in the manuscript text

Study snapshot

DesignX-ray crystallography of SIRT6/ADP-ribose and SIRT6/2'-N-acetyl-ADP-ribose complexes solved at 1.84–2.10 Å resolution; three complementary in vitro deacetylation assays (charcoal-binding, HPLC product separation, coupled continuous nicotinamide release); isothermal titration calorimetry and intrinsic tryptophan fluorescence for ligand binding.
ModelRecombinant human SIRT6 (residues 1–355 for kinetics; residues 3–318 for crystallography) and recombinant yeast Hst2 as a comparator sirtuin.
SampleThree independent replicates for kinetic and binding assays; three co-crystal structures solved from six molecules per asymmetric unit.
InterventionH3K9-acetylated peptide substrate (residues 5–13), NAD+, ADP-ribose, 2'-N-acetyl-ADP-ribose, NADH; SIRT6 H131Y catalytic-histidine mutant as a negative control.
DurationKinetic reactions 0.5–3 h; ITC titrations included 37 automatic injections per experiment.
EndpointsRate of SIRT6-catalysed OAADPr formation; rate of nicotinamide release; dissociation constants (Kd) for NAD+, ADPr, NAADPr, NADH; crystal structure of the SIRT6 catalytic core with bound cofactor analogues.

What the study showed, in plain terms

This is the first paper to solve the crystal structure of human SIRT6 and put the enzyme through detailed quantitative testing. It settled a longstanding debate about what SIRT6 actually does at the molecular level and revealed why the enzyme behaves so unusually.

Using three independent assays, the authors confirmed that SIRT6 can remove acetyl groups from histone H3 lysine 9 — but around 1,000 times more slowly than related Sirtuin enzymes. This weak activity is not an experimental artefact; a catalytically dead mutant produced no detectable product, ruling out contamination. The observation implied that SIRT6 must be regulated or activated in cells to do its job efficiently.

The crystal structure explained the low activity. SIRT6 is built from two subdomains that in other Sirtuins clamp together to form a tight substrate-binding pocket. In SIRT6, a conserved connecting helix bundle is missing and a stabilising salt bridge is broken, so the two subdomains sit further apart in a splayed, open configuration. Peptide substrate binding is therefore weak. SIRT6 was also shown to bind NAD+ tightly even without an acetylated substrate present — the opposite of all other Sirtuins studied — leading the authors to propose that SIRT6 might function partly as a sensor of cellular NAD+ levels rather than only as a conventional enzyme.

Key findings

  • SIRT6 catalyses NAD+-dependent deacetylation of the H3K9Ac peptide with a specific activity of approximately 0.0003 s−1, roughly 1,000-fold slower than yeast Hst2 or other characterised sirtuins.
  • Three independent quantitative assays (charcoal-binding for acetate release, HPLC quantification of O-acetyl-ADP-ribose formation, and continuous coupled assay for nicotinamide release) gave concordant rates, confirming SIRT6 is a bona fide but slow deacetylase.
  • The catalytic histidine mutant H131Y showed no measurable deacetylase activity, ruling out contamination as the source of the observed low activity.
  • Nicotinamide cleavage was tightly coupled to OAADPr formation, arguing against a significant contribution from a separate ADP-ribosylation activity in the presence of acetyl substrate.
  • Crystal structures of SIRT6 with ADP-ribose (1.84 Å and 2.0 Å) and with 2'-N-acetyl-ADP-ribose (2.1 Å) revealed a splayed conformation of the small zinc-binding domain relative to the Rossmann fold, distinct from all previously solved Sirtuins.
  • SIRT6 lacks the conserved α-helix bundle that in other sirtuins locks the small domain against the Rossmann fold, and lacks a conserved salt bridge between the zinc-binding motif and the substrate-binding loop.
  • SIRT6 contains a single ordered helix in place of the flexible cofactor-binding loop seen in other sirtuins, and the N-terminal extension folds back to stabilise the NAD+-binding site.
  • Unlike SIRT1, SIRT2, SIRT3, and SIRT5, SIRT6 bound NAD+ tightly (Kd approximately 27 μM by ITC) in the absence of any acetylated substrate, indicating a distinct binding order.
  • SIRT6 bound ADP-ribose (Kd 4.7 μM) and 2'-N-acetyl-ADP-ribose (Kd 22 μM) with similar affinity to NAD+ but did not detectably bind NADH; tryptophan fluorescence indicated NAD+ and ADPr induce different local structural rearrangements.

What this study can and cannot tell us

All experiments used recombinant SIRT6 residues 1–355 (kinetics) or 3–318 (crystallography). No physiological chromatin substrates (nucleosomes or full-length histones) were used, and subsequent work has shown SIRT6 is substantially more active on nucleosomal substrates than on peptides.

The H3K9-acetylated peptide was used at concentrations up to 300 μM but did not saturate SIRT6 binding by ITC. The reported Kd for peptide binding could not be quantified, limiting inferences about the molecular basis of the low activity.

The proposal that SIRT6 could function as a NAD+ metabolite sensor is a hypothesis suggested by the tight NAD+ binding observed in the absence of substrate. No cellular experiments were performed to test this idea in vivo.

Published in 2011, the paper pre-dates the discovery of SIRT6 long-chain deacylation activity (Feldman 2013; Jiang 2013), the activation by free fatty acids, and the small-molecule allosteric activators that have since made sense of the low peptide-deacetylation rate observed here.

Related papers in the corpus

Tier 4
Structure and biochemical functions of SIRT6
Patricia W. Pan et al. · Journal of Biological Chemistry · 2011
Tier 4
SIRT6: novel mechanisms and links to aging and disease
Luisa Tasselli et al. · Trends in Endocrinology & Metabolism · 2017
Tier 4
Mechanism of activation for the sirtuin 6 protein deacylase
Mark A. Klein et al. · Journal of Biological Chemistry · 2020

Citation network

Articles citing this research paper

1 article

Biohack Blueprint analyses that reference this study in their evidence base.

Editorial review

Reviewed by the Biohack Blueprint research team

Last verified

Dr. Khaled Mahmoud, MBBChCritical Care Medicine, Longevity Science & Editorial Direction Dr. Amr Yasser, PhDCardiometabolic, Musculoskeletal, Gut & Bioavailability Dr. Abdulla Mostafa, PhDCellular Senescence, Neurological, Mitochondrial & Brain-Ageing Biology