Mechanism-Based Inhibition of Sir2 Deacetylases by Thioacetyl-Lysine Peptide ² Brian C. Smith and John M. Denu* Departments of Chemistry and Biomolecular Chemistry, UniVersity of Wisconsin, Madison, Wisconsin 53706 ReceiVed July 5, 2007; ReVised Manuscript ReceiVed September 5, 2007 ABSTRACT: Sir2 protein deacetylases (or sirtuins) catalyze NAD + -dependent conversion of ǫ-amino- acetylated lysine residues to deacetylated lysine, nicotinamide, and 2-O-acetyl-ADP-ribose. Small-molecule modulation of sirtuin activity might treat age-associated diseases, such as type II diabetes, obesity, and neurodegenerative disorders. Here, we have evaluated the mechanisms of sirtuin inhibition of histone peptides containing thioacetyl or mono-, di-, and trifluoroacetyl groups at the ǫ-amino of lysine. Although all substituted peptides yielded inhibition of the deacetylation reaction, the thioacetyl-lysine peptide exhibited exceptionally potent inhibition of sirtuins Sirt1, Sirt2, Sirt3, and Hst2. Using Hst2 as a representative sirtuin, the trifluoroacetyl-lysine peptide displayed competitive inhibition with acetyl-lysine substrate and yielded an inhibition constant (K is ) of 4.8 µM, similar to its K d value of 3.3 µM. In contrast, inhibition by thioacetyl-lysine peptide yielded an inhibition constant (K is ) of 0.017 µM, 280-fold lower than its K d value of 4.7 µM. Examination of thioacetyl-lysine peptide as an alternative sirtuin substrate revealed conserved production of deacetylated peptide and 1-SH-2-O-acetyl-ADP-ribose. Pre-steady-state and steady-state analysis of the thioacetyl-lysine peptide showed rapid nicotinamide formation (4.5 s -1 ) but slow overall turnover (0.0024 s -1 ), indicating that the reaction stalled at an intermediate after nicotinamide formation. Mass spectral analysis yielded a novel species (m/z 1754.3) that is consistent with an ADP- ribose-peptidyl adduct (1-S-alkylamidate) as the stalled intermediate. Additional experiments involving solvent isotope effects, general base mutational analysis, and density functional calculations are consistent with impaired 2-hydroxyl attack on the ADP-ribose-peptidyl intermediate. These results have implications for the development of mechanism-based inhibitors of Sir2 deacetylases. Members of the silent information regulator 2 (Sir2 or sirtuin) family of protein deacetylases catalyze the conversion of acetyl-lysine residues and NAD + to deacetylated lysine, nicotinamide, and 2-O-acetyl-ADP-ribose (OAADPr) (1, 2). The most studied human Sir2 homologue, Sirt1, is reported to deacetylate a variety of substrates, including p53 (3-5), PPARγ (6), PGC-1R (7, 8), and AceCS1 (9), implicating sirtuins in a broad range of biological processes, including stress resistance and glucose homeostasis. Sirtuin deacetylase activity has also been associated with pathways that oppose age-associated diseases, such as type II diabetes, obesity, and neurodegenerative disorders (10). The design of mechanism- based sirtuin inhibitors is aided by a detailed understanding of the chemical mechanism. Toward this end, it has been shown that sirtuins use a sequential mechanism in which the acetyl-lysine substrate binds first followed by NAD + to form a productive complex that undergoes catalysis in two main chemical steps (11). In the first portion of the proposed catalytic mechanism, the nicotinamide ribosyl bond of NAD + is cleaved and acetyl-lysine attacks to form nicotinamide and an R-1-O-alkylamidate intermediate (12-16). In the second portion, an active site histidine activates the 2-hydroxyl for attack of the O-alkylamidate to form a 1,2-cyclic intermedi- ate (15). This is followed by addition of water to the 1,2- cyclic intermediate to form the product OAADPr and deacetylated peptide (1, 2). Recently, a thioacetyl-lysine peptide derived from the C-terminal region of p53 was shown to inhibit Sirt1-catalyzed deacetylation of an acetyl-lysine peptide with an IC 50 value of 2 µM(17). The authors found the rate of Sirt1 de(thio)- acetylation with a thioacetyl p53 peptide was 400-fold slower than that of the corresponding acetyl-lysine peptide, though the cleavage of NAD + was suggested to be much less affected (17). Collectively, these observations suggested that this thioacetyl-lysine p53 peptide might form a catalytically less competent intermediate compared to an acetyl-lysine peptide. However, the mechanism of thioacetyl-lysine peptide inhibition and turnover by sirtuins was unknown. Here, we determine the chemical mechanism of thioacetyl-lysine peptide inhibition through mass spectrometry, pre-steady- state and steady-state kinetics, mutagenesis, isotope effects, and computational modeling. EXPERIMENTAL PROCEDURES Expression and Purification of His-Tagged Sir2 Homo- logues. Expression and purification of Hst2 (1, 13), Hst2 H135A (1, 13), Sirt1 (9, 18), Sirt2 (19), and Sirt3 (9) were performed as previously described. The enzyme concentra- ² This work was supported by National Institutes of Health Grant GM065386 (to J.M.D.) and by National Institutes of Health Biotech- nology Training Grant NIH 5 T32 GM08349 (to B.C.S.). Computational resources provided to the University of Wisconsin Chemistry Parallel Computing Center though NSF Grant CHE0091916 and gifts from Intel Corp. * To whom correspondence should be addressed: Department of Biomolecular Chemistry, University of Wisconsin, 1300 University Ave., Madison, WI 53706-1532. Telephone: (608) 265-1859. Fax: (608) 262-5253. E-mail: jmdenu@wisc.edu. Department of Chemistry. § Department of Biomolecular Chemistry. 14478 Biochemistry 2007, 46, 14478-14486 10.1021/bi7013294 CCC: $37.00 © 2007 American Chemical Society Published on Web 11/21/2007