Selective Sirt2 inhibition by ligand-induced rearrangement of the active site

Sirtuins are a highly conserved class of NAD+-dependent lysine deacylases. The human isotype Sirt2 has been implicated in the pathogenesis of cancer, inflammation and neurodegeneration, which makes the modulation of Sirt2 activity a promising strategy for pharmaceutical intervention. A rational basis for the development of optimized Sirt2 inhibitors is lacking so far. Here we present high-resolution structures of human Sirt2 in complex with highly selective drug-like inhibitors that show a unique inhibitory mechanism. Potency and the unprecedented Sirt2 selectivity are based on a ligand-induced structural rearrangement of the active site unveiling a yet-unexploited binding pocket. Application of the most potent Sirtuin-rearranging ligand, termed SirReal2, leads to tubulin hyperacetylation in HeLa cells and induces destabilization of the checkpoint protein BubR1, consistent with Sirt2 inhibition in vivo. Our structural insights into this unique mechanism of selective sirtuin inhibition provide the basis for further inhibitor development and selective tools for sirtuin biology.

The interactions of NCA with Sirt2 and its position within the C-pocket is similar to the ones observed in other sirtuin structures in complex with nicotinamide (Sir2Tm-Ac-Lys-p53-NCA, PDB-ID 1YC5 11 , light orange, Sir2Af2-NAD + -NCA, PDB-ID 1YC2 11 , teal).
However, in contrast to nicotinamide of Sir2Tm-Ac-Lys-p53-NCA and Sir2Af2-NAD + -NCA the amide moiety and the phenyl ring of nicotinamide of Sirt2-ADPR-NCA do not lie in the same plane. The cofactor-binding loop in b,d,e is not shown for better clarity.

Cofactorbinding loop
SirReal2-interacting residues of the Sirt2-SirReal2 structures except for Phe96 which is forced to adopt a different position upon NAD + -binding. Hydrogen bonds are shown as dashed yellow lines. (e) Electron density maps for SirReal2 (pale cyan sticks, overall B-factor of 25.1 Å 2 ) and the acetyl-lysine peptide substrate (light blue sticks, overall B-factor of 48.3 Å 2 ). The -weighted 2F o -F c electron density map is contoured at 1.0 . A stereo image of e as well as a -weighted F o -F c electron density OMIT map for SirReal2 and the Ac-Lys-H3 oligopeptide is shown in Supplementary Fig. 5a,b. Figure 4 Stereo images of the -weighted 2F o -F c electron density and theweighted F o -F c OMIT maps of presented inhibitors, cosubstrates and substrates. Wall-eyed stereo representation of the electron density of the ligands of (a) the Sirt2-ADPR-NCA structure and the Sirt2-SirReal2-NAD + complex (c). Sirt2-ADPR-NCA is presented as a pale green cartoon and ADPR and NCA is shown as dark blue and dark brown sticks. Sirt2-SirReal2-NAD + is presented as a slate blue cartoon and SirReal2 and NAD + are shown as light pink and light orange sticks. The surface of Sirt2-SirReal2-NAD + of c is colored according to its    Sirt2-SirReal2-

SirReal6 -2-((4,6-Dimethylpyrimidin-2-yl)thio)-N-methyl-N-(5-(naphthalen-1-ylmethyl)-
The system was solvated using the water model TIP3BOX 31 and a margin of 10 Å. Two consecutive steps of minimization were carried out. In the first step 3,000 iterations (first 1,000 steepest descent and then 2,000 conjugate gradient) and in the second step 4,000 iterations (first 2,000 steepest descent and then 2,000 conjugate gradient) were applied to the system. In the first step, atom coordinates for the amino acid residues and ligand atoms were restrained to their initial coordinates with a force constant of 500 kcal mol -1 Å -2 to relieve the unfavorable van-der-Waals contacts in the surrounding solvent and thus to minimize the positions of the water molecules and ions. In the second step, restraints to atoms were removed and the whole system was minimized freely to relieve bad contacts in the entire system.
The temperature of the system was then equilibrated at 300 K through 100 ps of MD with a time step of 2 fs per step. A constant volume periodic boundary was set to equilibrate the temperature of the system by the Langevin dynamics 32 using a collision frequency of 1 ps 1 during the temperature equilibration routine. The protein and ligand atoms were restrained to the initial coordinates with a weak force constant of 10 kcal mol -1 Å -2 . The final coordinates obtained after temperature equilibration step were then used for a 20 ns MD routine during which the temperature was kept at 300 K by the Langevin dynamics using a collision frequency of 1 ps -1 . Constant pressure periodic boundary was used to maintain the pressure of the system at 1 bar using isotropic pressure scaling with a relaxation time of 2 ps. During the temperature equilibration and MD routines a non-bonded cut-off distance of 10 Å was used applying the Particle Mesh Ewald (PME) method 33  In vitro KDAC1/6 assay. Inhibition tests with SirReal2 and KDAC1/6 were conducted with a high-throughput fluorescence-based assay using the substrate ZMAL (Z-Lys(Acetyl)-AMC) 36