Tier 4 — mechanistic

Mechanism of activation for the sirtuin 6 protein deacylase

Mark A. Klein, Can Liu, Vyacheslav I. Kuznetsov, John B.Feltenberger, Weiping Tang, John M. Denu
Journal of Biological Chemistry 2020 Volume 295, issue 5, pages 1385–1399

Bibliography

PubMed
PMID 31822559
DOI
10.1074/jbc.RA119.011285
Funding
National Institutes of Health Grant GM65386 (to J.M.D.) and the University of Wisconsin Institute for Clinical and Translational Research. M.A.K. supported by NIGMS Chemistry-Biology Interface Training Program Grant T32GM008505.
Competing interests
J.M.D. is a consultant for FORGE Life Science and cofounder of Galilei BioSciences.

Study snapshot

DesignActivity-based screen of 432 fatty acid and lipid-like compounds followed by medicinal-chemistry derivatization and detailed enzymology (steady-state Michaelis-Menten, isothermal titration calorimetry, rapid quench-flow single-turnover, and thermal denaturation) with SIRT6 WT and R65A variant.
ModelRecombinant human SIRT6 (WT, R65A, R76A, K81A, K160A) and other recombinant sirtuins (SIRT1, SIRT2, SIRT3, SIRT5); acid-extracted histones from HEK293T cells.
Samplen=3 replicates for most assays; bisubstrate kinetics performed in duplicate; 432 compounds screened at 100 μM.
InterventionScreened 432 compounds; lead compound CL-4 optimised to CL5D (2-(3-chloro-4-(2,4,6-trichloro-N-(2,4,6-trichlorobenzoyl)benzamido)phenyl)-1,3-dioxoisoindoline-5-carboxylic acid); characterisation of oleoyl-lysophosphatidic acid (LPA), linoleoyl-glycine (L-Gly), and arotinoid acid (TTNPB).
DurationIndividual assays ranged from 3 min (steady-state) to 130 min (ITC titrations).
EndpointsFold-activation of SIRT6 deacetylation of H3K9Ac peptide; EC50 and Michaelis-Menten parameters (kcat, Km); Kd for H3K9Ac binding via ITC and fluorescence polarisation; single-turnover rates of alkylamidate and 1',2'-cyclic intermediate formation; SIRT6 R65A demyristoylation kinetics; thermal stability.

What the study showed, in plain terms

SIRT6 removes acetyl groups from histones inside the cell nucleus but is a surprisingly weak enzyme when purified and tested in the laboratory — hundreds of times slower at deacetylation than at removing longer fatty acid chains from lysine. This paper set out to identify what actually turns SIRT6 on and to work out the molecular mechanism of that activation.

The authors screened 432 fatty acids and lipid-like molecules and found 71 that increased SIRT6 activity more than five-fold. Chemical optimisation of one hit produced CL5D, an activator that boosted SIRT6 deacetylation up to 50-fold. Detailed enzyme kinetics showed the activators work by dramatically improving how tightly SIRT6 grips its acetylated substrate — the acetylated histone binds roughly 35 times better in the presence of the activator, without changing the maximum reaction speed much.

The authors discovered that a single amino acid on SIRT6, arginine-65, is essential for this activation and also for the enzyme's naturally strong long-chain deacylase activity. When arginine-65 was mutated to alanine, activators could no longer speed the enzyme up, and long-chain fatty acyl removal became severely impaired. The results support a model in which activator or long-chain substrate binding triggers a conformational rearrangement of SIRT6 that positions the catalytic machinery for efficient reaction — and this rearrangement depends on arginine-65.

Key findings

  • Screening of 432 compounds identified 71 SIRT6 activators (over 5-fold) and 23 potent ones (over 10-fold); 97% of hits carried a terminal negative charge and 89% a linear aliphatic chain.
  • Top biological activators included oleoyl-lysophosphatidic acid (48-fold), N-linoleoylglycine (29-fold), 2-fluoropalmitic acid (25-fold), and arotinoid acid TTNPB (21-fold), all in the 25–97 μM EC50 range.
  • The synthetic activator CL5D produced up to 50-fold activation of SIRT6 kcat/Km for H3K9Ac and was selective for SIRT6 over SIRT1, SIRT2, SIRT3, and SIRT5.
  • Activation was driven overwhelmingly by improved substrate binding (Km for H3K9Ac fell from approximately 450 μM to 9 μM in the presence of 400 μM myristic acid) rather than by a large increase in maximum turnover.
  • Fluorescence polarisation and isothermal titration calorimetry showed SIRT6 binds NAD+ before acetylated substrate — the reverse of the order established for SIRT1, SIRT2, and SIRT3.
  • The SIRT6 R65A variant retained basal deacetylase activity but could not be activated by CL5D and could not efficiently deacylate long-chain (myristoyl) substrates.
  • Rapid quench-flow kinetics on R65A showed a 16-fold reduction in the rate of 1',2'-cyclic intermediate formation but only a 1.4-fold reduction in alkylamidate formation, implicating Arg-65 in a catalytic step after nicotinamide release.
  • Absence of a nicotinamide burst in pre-steady-state analysis of WT SIRT6 indicated a slow conformational step after substrate binding but before the first chemical step — the step accelerated by allosteric activators.
  • Thermal denaturation showed NAD+ stabilised WT SIRT6 by 0.7 °C but destabilised R65A by 0.6 °C, consistent with R65A being deficient in the conformational rearrangement triggered by cofactor binding.

What this study can and cannot tell us

All assays used peptide substrates (H3K9Ac or H3K9myr, residues 5–13); nucleosomal substrates were not systematically tested with the new activators, and nucleosome binding is known to enhance SIRT6 activity in ways that peptide assays cannot capture.

CL5D was designed as a mechanistic tool, not a drug candidate. Its polyhalogenated structure and low aqueous solubility (with loss of activity above roughly 50 μM due to aggregation) make it unsuitable for cell or animal use. Cellular activation was demonstrated only in HEK293T lysates in vitro, not by treating intact cells.

The proposed conformational-change model is supported by kinetic modelling and by the R65A phenotype but was not directly visualised. A crystal structure of SIRT6 in complex with CL5D was not reported; the exact activator binding pocket is inferred from competitive kinetics with myristoylated peptide.

The generalisability of the Arg-65-dependent activation mechanism to natural regulators inside cells (fatty acids, lamin A, ribosomal proteins) is speculative. Whether this pathway operates physiologically in humans remains an open question.

Related papers in the corpus

Tier 3
The Identification of a SIRT6 Activator from Brown Algae Fucus distichus
Rahnasto-Rilla MK et al. · Marine Drugs · 2017
Tier 2
SIRT6 in DNA Repair, Metabolism, and Ageing
Lombard DB et al. · Journal of Internal Medicine · 2008
Tier 3
SIRT6 is a DNA double-strand break sensor
Onn L et al. · eLife · 2020
Tier 3
Sirt6 prevents the age-related decline of H2S through the control of one-carbon metabolism
Touitou N et al. · Proceedings of the National Academy of Sciences of the United States of America · 2025

Citation network

Articles citing this research paper

2 articles

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