MINIREVIEW Small molecule regulation of Sir2 protein deacetylases Olivera Grubisha 1 , Brian C. Smith 2 and John M. Denu 1 1 Department of Biomolecular Chemistry, University of Wisconsin, Madison, WI, USA 2 Department of Chemistry, University of Wisconsin, Madison, WI, USA Introduction The silent information regulator 2 (Sir2) family of pro- teins (sirtuins) are class III histone ⁄ protein deacetylases (HDACs) [1]. Members of this evolutionarily con- served family include five homologues in yeast (ySir2 and Hst1–4) and seven in humans (SIRT1–7) [2,3], with key roles in cellular processes such as gene expres- sion, apoptosis, metabolism and ageing [4]. The found- ing member, yeast Sir2 (ySir2), was originally identified as a trans-acting factor involved in transcrip- tional repression of the silent mating type loci in yeast [5]. Now it is well established that ySir2 deacetylase activity is required for silencing at telomeres, rDNA and the silent mating type loci, and for maintaining genome integrity [5,6]. In addition to silencing, Sir2 activity is linked to lifespan extension in yeast [7], worms [8] and flies [9]. SIRT1, the most extensively studied human Sir2 orthologue, localises to the nucleus where it negatively regulates damage-responsive Fork- head transcription factors [10–12] and p53 [13–15], promoting cell survival under stress. SIRT1 also dis- plays tissue-specific roles including skeletal muscle differentiation [16] and fat mobilization in white adipocytes [17]. In contrast to SIRT1, SIRT2, SIRT3 and SIRT5, no NAD + -dependent protein deacetylase activity has been reported for SIRT4, SIRT6 and SIRT7. The possibility remains that SIRT4, 6 and 7 exhibit specificity toward substrates other than those tested or that these proteins catalyse a distinct reaction. Keywords Sir2; deacetylation; sirtuin; NAD; sirtinol; splitomicin; resveratrol Correspondence J. M. Denu, University of Wisconsin, Department of Biomolecular Chemistry, 1300 University Ave., Madison, WI 53706–1532, USA Fax: +1 608 262 5253 Tel: +1 608 265 1859 E-mail: jmdenu@wisc.edu (Received 17 March 2005, revised 6 June 2005, accepted 8 June 2005) doi:10.1111/j.1742-4658.2005.04862.x The Sir2 family of histone ⁄ protein deacetylases (sirtuins) is comprised of homologues found across all kingdoms of life. These enzymes catalyse a unique reaction in which NAD + and acetylated substrate are converted into deacetylated product, nicotinamide, and a novel metabolite O-acetyl ADP-ribose. Although the catalytic mechanism is well conserved across Sir2 family members, sirtuins display differential specificity toward acetyl- ated substrates, which translates into an expanding range of physiological functions. These roles include control of gene expression, cell cycle regula- tion, apoptosis, metabolism and ageing. The dependence of sirtuin activity on NAD + has spearheaded investigations into how these enzymes respond to metabolic signals, such as caloric restriction. In addition, NAD + meta- bolites and NAD + salvage pathway enzymes regulate sirtuin activity, supporting a link between deacetylation of target proteins and metabolic pathways. Apart from physiological regulators, forward chemical genetics and high-throughput activity screening has been used to identify sirtuin inhibitors and activators. This review focuses on small molecule regulators that control the activity and functions of this unusual family of protein deacetylases. Abbreviations CR, caloric restriction; ERCs, extrachromosomal rDNA circles; HDAC, histone ⁄ protein deacetylase; NADases, NAD + glycohydrolases; Npt1, nicotinate phosphoribosyltransferase; OAADPr, O-acetyl-ADP-ribose; PARPs, poly(ADP-ribose) polymerases; Sir2, silent information regulator 2. FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS 4607 In support of the latter, SIRT6 was recently shown to transfer the ADP-ribose moiety of NAD + and undergo mono-ADP-ribosylation [18]. Unlike class I and II HDACs, which activate a water molecule for direct hydrolysis of the acetyl group [1], class III HDACs require NAD + as a cosub- strate for the deacetylation reaction [19–22]. NAD + and the acetylated lysine residue on the substrate react in a 1 : 1 ratio to form deacetylated product, nicotina- mide, and a novel metabolite 2¢-O-acetyl-ADP ribose (OAADPr) (Fig. 1) [23–26]. The consumption of NAD + and the generation of OAADPr by class III HDACs probably serve as a link between deacetylation and other physiological processes. Although the roles of OAADPr are not yet known, microinjection of OAADPr has been shown to inhibit oocyte maturation and to block cell division in starfish blastomeres [27]. Furthermore, unidentified enzymes found in starfish, yeast, and human cell extracts, are able to rapidly metabolize OAADPr [27,28]. This evidence suggests that mechanisms exist to tightly control OAADPr lev- els. Therefore, it is possible that OAADPr may act as a secondary messenger, a cofactor, or as a metabolic intermediate that links deacetylation of target proteins to other cellular pathways [29]. In support of this view, recent evidence suggests that OAADPr directly regu- lates gene silencing in yeast [30]. Elegant electron micro- scopy studies showed that a complex consisting of Sir2, Sir3 and Sir4 undergoes a supramolecular rear- rangement in the presence of OAADPr. The authors hypothesize that OAADPr, the product of Sir2 histone deacetylation, directly binds to one or more constitu- ents in the complex resulting in structural reorganiza- tion and the ability to establish silent chromatin domains. The dependence of sirtuin activity on NAD + has prompted investigations into how these enzymes might link the cellular energy state to processes such as gene expression, cell cycle regulation, apoptosis and ageing. This review will evaluate recent discoveries concerning the physiological regulation of sirtuins by NAD + metabolites and by enzymes in the NAD + salvage path- way. In addition, we will cover the use and efficacy of small molecule inhibitors and activators of sirtuin activ- ity such as sirtinol, splitomicin and resveratrol with particular focus on the ability of these compounds to regulate Sir2-mediated lifespan extension. Physiological regulation The variety of important functions involving Sir2 enzymes underscores the need to understand the mech- anisms that regulate their physiological activity. The requirement of NAD + as a cosubstrate has led to the proposal that either intracellular NAD + or NADH concentrations or a metabolic parameter such as the NAD + ⁄ NADH ratio regulates Sir2 activity (reviewed in [4,29,31]), effectively linking Sir2 activity to the metabolic status of cells. Originally, caloric restriction (CR) in yeast was thought to increase the NAD + lev- els, which would increase the activity of ySir2 and pro- mote its role in lifespan extension [32,33]. However, there is little data to support the assertion that global changes in cellular NAD + and NADH during CR would have a significant impact on ySir2 activity. In yeast grown under aerobic conditions, concentrations of NAD + and NADH were reported to be approxi- mately 4 mm and 0.2 mm, respectively, yielding an NAD + ⁄ NADH ratio of about 20 [34]. Under caloric restriction, a condition that presumably activates Sir2, this ratio fluctuated less than twofold [35], due only to a change in NADH levels. NADH was reported to act as a competitive inhibitor of Sir2 in vitro [35], leading to a conclusion that NADH would compete with Fig. 1. Overview of the reaction catalysed by Sir2 protein deacetylases. Small molecule regulation of Sir2 protein deacetylases O. Grubisha et al. 4608 FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS NAD + for binding to Sir2. However, K m values for NAD + typically fall between 10 and 100 lm, whereas IC 50 values for NADH range from 11 to 28 mm [36]. Therefore, it is unlikely that NADH levels would reach high enough concentrations to significantly inhibit Sir2 activity. A dramatic drop in NAD + levels would be more likely to be a factor in Sir2 regulation, especially if free intracellular NAD + concentrations were to fall in the low micromolar range. Such instances could occur through activation of NAD + -consuming enzymes such as poly(ADP-ribose) polymerases (PARPs), NAD + glycohydrolases (NADases), or perhaps mono- ADP-ribosyl transferases [37]. An important caveat to the aforementioned Sir2 studies is the fact that NAD + and NADH levels were measured from whole cell lysates and the possibility that microdomains of these metabolites exist where ySir functions has not been explored. For instance, NAD + synthesizing enzymes might be a part of a Sir2-containing complex and these enzymes may channel NAD + directly to Sir2, creating a microdomain of high NAD + concentrations specific- ally accessible to Sir2. Nicotinamide, a product of the Sir2 deacetylation reaction, is a potent physiological inhibitor of Sir2 enzymes [36,38,39]. In vitro, nicotinamide yields an IC 50 of  120 lm with several Sir2 homologues [36]. Originally, it was believed that nicotinamide bound to an allosteric site and consequently inhibited Sir2 activ- ity [40]. However, it was shown later that nicotinamide inhibition arises from its ability to condense with a high-energy enzyme–ADP ribose–acetyl-lysine inter- mediate to reverse the reaction, reforming NAD + and thereby inhibiting product formation [38,39]. Nicotin- amide acts as a classical noncompetitive product inhi- bitor of the forward deacetylation reaction and was shown in vivo to decrease gene silencing, increase rDNA recombination and accelerate ageing in yeast [40]. Because nuclear nicotinamide levels are estimated to be 10–150 lm [41], it is likely that nicotinamide regulates Sir2 activity in vivo. By the same token, enzymes involved in NAD + sal- vage regulate Sir2 function by modulating levels of nicotinamide and other NAD + metabolites. As depic- ted in Fig. 2A, the yeast NAD + salvage pathway con- verts nicotinamide into NAD + through four distinct enzymatic steps. Anderson et al. showed that increased dosage of several enzymes in the NAD + salvage path- way increased ySir2-dependent silencing, albeit to vary- ing extents [42]. Most notably, overexpression of nicotinamidase (Pnc1) rescued silencing at telomeres and rDNA in the presence of exogenous nicotinamide [43], whereas deletion of PNC1 had the opposite effect [44]. Although deletion of PNC1 did not change cellular NAD + levels [44], a 10-fold increase in nico- tinamide was observed [41]. Therefore, the known up- regulation of PNC1 expression in response to heat and osmotic shock, and oxidative exposure ([43] and refs therein) would positively regulate ySir2 activity by reducing cellular nicotinamide levels. Similarly, muta- tions in nicotinate phosphoribosyltransferase (Npt1), an enzyme that converts nicotinic acid (vitamin B3) to nicotinic acid mononucleotide (NaMN), resulted in severe rDNA and telomere silencing defects, and a threefold reduction of intracellular NAD + levels [44]. The phenotype is more severe than that seen in a pnc1 deletion strain, probably because loss of Npt1 blocks the conversion of intracellular and environmental nico- tinic acid to NAD + . Overexpression of Npt1 led to enhanced Sir2-dependent silencing but did not alter NAD + levels [42]. Anderson et al. suggested that increased dosage of NPT1 might increase local avail- ability of NAD + for ySir2 without detectable changes in steady-state NAD + levels. These data support the idea that the NAD + salvage pathway in yeast can regulate ySir2 activity by decreasing nicotinamide lev- els and increasing the flux through the pathway to increase NAD + concentrations. At this point, it is unclear whether the cellular pools of NAD + are distinct from those accessible to Sir2. As we suggested above, small global changes in NAD + may not be sufficient to alter Sir2 function, but instead, localized synthesis of NAD + (microdomains) at the site of Sir2 function may play a more significant role in controlling activity. Recently, NAD + analogues and salvage pathway intermediates were evaluated as possible direct regula- tors (inhibitors, activators, substrates) of Sir2 activity. This analysis showed that NAD + analogues, with sub- stitution at either the nicotinamide ring or the adenine base, are poor substrates for the Sir2 reaction [36]. Furthermore, only nicotinamide displayed a level of inhibition consistent with a physiological role (IC 50 of  120 lm), whereas the worst inhibitors tested were the three acid analogues NAMN, NAAD and nicotinic acid, with IC 50 values of 26–250 mm. None of the metabolites tested yielded Sir2 activation. These results are consistent with the proposal that changes in cellu- lar NAD + and nicotinamide concentrations are likely to be the greatest contributors to the physiological regulation of Sir2 enzymes. The NAD + salvage pathway in mammals, shown in Fig. 2B, does not have an equivalent of nicotinamidase Pnc1. However, both nicotinamide and nicotinic acid are converted to NAD + through different metabolic intermediates. A recent report by Revollo et al. [45] demonstrated that increased dosage of nicotinamide phosphoribosyltransferase (Nampt), the rate limiting O. Grubisha et al. Small molecule regulation of Sir2 protein deacetylases FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS 4609 component in mammalian NAD + biosynthesis, increased total cellular NAD + levels by  40% and enhanced the transcriptional regulation activity of Sir2a, a mouse Sir2 orthologue. Another study found that overexpression of nicotinamide⁄ nicotinic acid mononucleotide adenyltransferase (Nmnat1) or an increase in NAD + concentrations protected injured axons in a Wallerian degeneration model [46]. The protection depended on the presence of SIRT1, sug- gesting that an increase in Nmnat1 activity leads to Fig. 2. (A) NAD + salvage pathway in yeast. (B) NAD + salvage pathway in mammals. Small molecule regulation of Sir2 protein deacetylases O. Grubisha et al. 4610 FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS SIRT1 activation, which consequently delays Wallerian degeneration [46]. These findings provide the first insights into the physiological regulation of mamma- lian Sir2 orthologues by metabolic pathways that regu- late the levels of NAD + and its precursors. Also, these studies on mammalian sirtuins serve to confirm the link between Sir2 enzymes and metabolic pathways, which were originally demonstrated in yeast. Addi- tional evidence for the intimate connection between metabolism and sirtuin activity comes from a host of other observations. Sir2 from Salmonella regulates acetyl-CoA synthetase by direct lysine deacetylation of an important catalytic residue [47]. SIRT1 was shown to promote fat mobilization in white adipocytes by repressing PPAR-c [17]. Recently, Rodgers et al. [48] reported that SIRT1 controls gluconeogenic ⁄ glycolytic pathways in liver in response to fasting signals through the transcriptional coactivator PGC-1a. Small molecule sirtuin inhibitors The importance of Sir2 deacetylases in a growing num- ber of cellular processes has created the need for better chemical tools to study Sir2 function. In particular, selective inhibitors and activators would allow researchers to precisely dissect the roles of Sir2 homo- logues in each organism. In addition, the involvement of human Sir2 homologues in a variety of critical cel- lular pathways makes them attractive drug targets. For example, the ability of SIRT1 to deactivate the p53 tumour suppressor protein suggests that SIRT1 inhibi- tors might act as anticancer agents [13–15]. Further- more, the capability of a-tubulin to serve as a substrate of SIRT2 indicates that drugs that target SIRT2 might regulate cell division, cell cycle and cell motility [49]. Perhaps the simplest examples of Sir2 inhibitors are nonhydrolysable analogues of NAD + , which compete for coenzyme binding in the active site. One such example is carba-NAD + , which is a noncompetitive inhibitor against NAD + with inhibition constants K ii and K is of 210 and 170 lm, respectively [21,50]. How- ever, NAD + analogues such as carba-NAD + are generally not cell-permeable. Furthermore, these com- pounds probably serve as inhibitors or substrates for a variety of other NAD + -dependent enzymes. Therefore, other methods, such as forward chemical genetics, have recently been used to screen for novel small mole- cule Sir2 regulators. Forward chemical genetics is an approach employed to screen a library of small organic molecules for their ability to inhibit or enhance a known phenotype. Com- pounds that produce a desired effect are then assayed in vitro to determine if they specifically target the pro- tein of interest. Using this approach, Grozinger and colleagues screened a 1600-compound library for inhi- bition of ySir2-mediated silencing at the telomere [51]. The screen was designed such that reporter gene expression from the telomere caused cell death. Three inhibitors, A3, M15 and sirtinol, were identified, the later two containing a 2-hydroxy-1-napthaldehyde moi- ety (Fig. 3A). Of these three, sirtinol was the most potent inhibitor overall, displaying IC 50 values of 68 and 38 lm against ySir2 and SIRT2, respectively. A similar strategy was used by Bedalov and cowork- ers to uncover a new class of Sir2 inhibitors [52]. Their screen was designed so that inhibition of ySir2-medi- ated telomeric silencing recovered normal cell growth. Such a design advantageously eliminated cytotoxic compounds as false positives. Out of 6000 compounds, 11 were capable of rescuing cell growth [52]. Subse- quent screening for inhibition of silencing at the HMR and rDNA locus showed that only one of the 11 com- pounds, splitomicin (Fig. 3A), was effective at all three loci. Splitomicin inhibited ySir2 with an IC 50 value of 60 lm in vitro, and based on mapping of splitomicin- resistant Sir2 mutants, the authors postulated that splitomicin acted by preventing the binding of acetyl- ated lysine substrates to ySir2. However, it is import- ant to point out that the in vitro assays were performed on whole cell extracts of an hst2 deletion yeast strain rather than purified ySir2. Therefore, Fig. 3. (A) Known inhibitors of Sir2 deacetylases. (B) Examples of known activators of Sir2 deacetylases. O. Grubisha et al. Small molecule regulation of Sir2 protein deacetylases FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS 4611 complete selectivity for ySir2 deacetylase activity can- not be inferred from this data. Further evaluation of 130 splitomicin analogues revealed the requirement for an intact lactone ring, whereas the naphthalene ring was dispensable for efficient ySir2 inhibition [53]. In a follow-up study using 100 splitomicin ana- logues, Hirao et al. identified dehydrosplitomicin and compound 26 as selective inhibitors of Hst1 and ySir2, respectively (Fig. 3A) [54]. However, compound 26 was not as potent as splitomicin in inhibiting ySir2. In addition to studies in yeast, sirtinol and splitomicin have been used as general sirtuin inhibitors in mamma- lian cells [16,46,55]. However, caution should be used in examining these data as neither compound has been extensively characterized as an selective inhibitor of any of the mammalian Sir2 homologues, or been tested for nonspecific effects in mammalian cells, in partic- ular, their effects on other NAD + -consuming enzymes. In a different approach using in silico methodology, Tervo et al. discovered novel inhibitors of human SIRT2, a more distantly related ySir2 homologue [56]. The authors identified 15 compounds that passed an in silico intestinal absorption test and exhibited favour- able binding to a conserved hydrophobic pocket in the NAD + binding site. Two of these compounds exhi- bited IC 50 values in the low micromolar range in vitro, the efficacy of which has yet to be reported in vivo. It is important to emphasize that the Sir2 inhibitors discovered to date only have potency in the micro- molar level, comparable to that of nicotinamide. In addition, how these molecules inhibit Sir2 activity is unknown. It is possible that these compounds compete for NAD + binding with their aromatic rings serving as nicotinamide or adenine mimics. If this is the case, then it is likely that they possess activity against other NAD + binding enzymes. This effect is seen with nico- tinamide, which in addition to its Sir2 inhibitory activ- ity, inhibits PARPs and acts as a substrate for nicotinamidase and nicotinamide phosphorybosyl transferase (reviewed in [57,58]). However, it is also possible that these Sir2 inhibitors bind to the acetyl- lysine peptide site, as suggested for splitomicin, or to unknown allosteric sites on the enzyme. Further stud- ies evaluating the mechanism of inhibition are needed to allow rational improvement of these compounds. Sir2 function in metabolism and ageing ySir2-dependent silencing at the rDNA locus not only maintains genome integrity but also extends lifespan in yeast. One cause of ageing stems from rDNA instability [31,59]. The rDNA locus consists of 100–200 tandem repeats encoding ribosomal RNAs and homologous recombination between these repeats results in the for- mation of extrachromosomal rDNA circles (ERCs), which accumulate in the mother cell, causing senes- cence. Although the mechanism by which ERCs cause death is unknown, the rate at which these circular DNAs accumulate correlates with the yeast lifespan [60]. A single extra copy of the SIR2 gene slows ERC formation and extends lifespan by 40%, presumably by suppressing recombination [7,42,61]. Conversely, deletion of SIR2 increases the frequency of rDNA recombination 10-fold [62] and shortens lifespan by 50% [7]. Increased dosage of SIR2 orthologues in Caenorhabditis elegans and Drosophila extends lifespan up to  50% in both organisms [8,9]. Another means of extending lifespan in yeast and other organisms is through caloric restriction [63,64]. The mechanism by which CR increases replicative life span in yeast has been suggested to be Sir2-mediated [61,65]. It was postulated that CR extends lifespan by causing NAD + levels to rise or NADH levels to decrease, which, in turn, increases Sir2 activity. In sup- port of this hypothesis, Lin et al. [35] report that CR leads to a twofold decrease in NADH, without any sig- nificant change in NAD + . The authors conclude that Sir2-mediated lifespan extension during CR results from decreased NADH levels [35]. However, in vitro biochemical data indicate that NADH is a poor inhi- bitor of Sir2 deacetylases [36] and that such a small change would have at best a 5% stimulation of Sir2 activity. Furthermore, rapidly ageing yeast were shown to have increased NAD + levels [42]. Collectively, the reports on the levels of NAD + during CR suggest that NAD + levels might not be a good indicator of ySir2 activity. The involvement of Sir2 in lifespan extension during CR has been recently challenged. Kaeberlein et al. suggest that Sir2 acts independently of pathways mediated by CR [66]. They propose that senescence due to ERC accumulation predominates over CR. If ERC formation is suppressed, lifespan extension by CR is independent of Sir2. In the PSY316 strain used previ- ously to link CR to Sir2 [68], Kaeberlein et al. demon- strated that overexpression of Sir2 does not increase life span [67]. Clearly, further studies will be needed to explore the role of Sir2 enzymes in determining lifespan through CR, both in yeast and higher eukaryotes. Resveratrol activation of sirtuins Evidence implicating sirtuins in lifespan extension has motivated the hunt for small molecule sirtuin activa- tors that increase lifespan in yeast, with the potential promise of identifying such compounds for human use. Utilizing a commercially available deacetylase activity Small molecule regulation of Sir2 protein deacetylases O. Grubisha et al. 4612 FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS assay from BIOMOL, Howitz and colleagues identified several putative ySir2 and SIRT1 activating com- pounds in a high-throughput screen [68]. These com- pounds included a few plant polyphenols, such as resveratrol, fisetin and butein (Fig. 3B). Of the com- pounds tested, resveratrol, a molecule found in red wine, exhibited the highest activation of SIRT1 by lowering the K m for the acetylated substrate, without affecting the overall turnover rate of the enzyme [68]. Given the reported cardioprotective, chemopreventive and neuroprotective health benefits of resveratrol (reviewed in [69]), the prospect of resveratrol-mediated Sir2 activation was intriguing. In yeast, resveratrol treatment reduced rDNA recombination by 60%, providing evidence of resvera- trol-mediated ySir2 activation [68]. Curiously, no effect on ySir2-dependent transcriptional silencing at rDNA was observed. Growing yeast in the presence of resve- ratrol increased lifespan up to 70%, whereas no change in lifespan was observed in a sir2 deletion strain, further supporting the hypothesis that resvera- trol increased lifespan by activating ySir2 [68]. Addi- tion of resveratrol under CR conditions did not cause an additional increase in lifespan, leading the authors to conclude that resveratrol and CR act through the same pathway. In C. elegans and D. melanogaster, treatment with resveratrol extended lifespan by 14% and 29%, respectively [70], but this effect was not observed in organisms that lacked wild type copies of ySir2 orthologues, dSir2 and Sir-2.1. Similar to results obtained in yeast, the effects of CR and resveratrol on lifespan extension in D. melanogaster were not addit- ive, leading the authors to conclude that resveratrol extends lifespan through a mechanism related to CR. In contrast, Kaeberlein et al. found no significant increase in lifespan, telomeric silencing or rDNA recombination with resveratrol treatment in three different yeast strain backgrounds [67], including the PSY316 strain used in the original study [68]. The basis for the discrepancy between studies has not been resolved, but may be due to variability in growth condi- tions. In an effort to elucidate the mechanism of res- veratrol activation, Kaeberlein et al. and our lab performed a series of biochemical studies and independ- ently determined that resveratrol activation of SIRT1 in vitro depended on the use of a nonphysiological sub- strate [67,71]. Specifically, the activation seen with res- veratrol in vitro required the covalent attachment of a fluorophore at the carboxyl-group of the acetyl-lysine residue. In addition, resveratrol was unable to signifi- cantly activate ySir2 and SIRT2 in vitro suggesting that resveratrol binds to a unique site within SIRT1. Although resveratrol activation of SIRT1 depended on a specific fluorophore substrate in vitro, resveratrol might still directly affect SIRT1 activity in vivo. For instance, resveratrol might induce a conformational change in SIRT1, thereby increasing the catalytic effi- ciency of the enzyme for specific protein substrates con- taining an aromatic residue, such as a tryptophan, at the equivalent position of the fluorophore-containing substrates. This possibility has yet to be evaluated. In mammalian cells, resveratrol was reported to enhance SIRT1-dependent cellular processes such as axonal protection, fat mobilization, and inhibition of NF-jB-dependent transcription [17,46,55]. In view of the possibility that the effect of resveratrol on SIRT1 is simply an in vitro phenomenon observed with fluor- escent peptides, it would be prudent to re-examine these in vivo studies and discern whether the observed activation of SIRT1 results from a direct interaction with resveratrol or through less direct mechanisms that are induced by resveratrol and indirectly impinge upon SIRT1-dependent processes. For example, resveratrol’s known antioxidant activity [72] may induce redox sen- sitive processes, which in turn activate SIRT1. Alter- natively, resveratrol might act by scavenging reactive oxygen species generated by the mitochondria, a mech- anism known to increase lifespan in many organisms (reviewed in [72]). Perhaps SIRT1 function is sensitive to cellular oxidants and resveratrol offers protection from inactivation, with an apparent increase in SIRT1 activity. Clearly, further studies will be needed to understand the molecular link between resveratrol and the apparent cellular activation of SIRT1. Mechanism-based activation Taking advantage of the unique mechanism of nicotin- amide inhibition, Sauve et al. recently reported isonico- tinamide as an activator of Sir2 activity [41]. Isonicotinamide was shown to directly compete with nicotinamide for binding. Nicotinamide is a potent inhibitor of the Sir2 reaction because of its aforemen- tioned ability to rebind with the enzyme and react with a high-energy intermediate, preventing deacetylation and regenerating starting materials [38,39]. The basis for the observed activation is the relief of the inherent nicotinamide inhibition by competition with isonicotin- amide, which does not readily react with the enzyme intermediate. Although the K i for isonicotinamide was 68 mm, or about three orders of magnitude worse than nicotinamide binding, in vivo yeast studies showed that millimolar levels of isonicotinamide increased Sir2- dependent silencing of the telomeric URA3 gene. These results suggest that the development of higher affinity nicotinamide antagonists may provide a means to O. Grubisha et al. Small molecule regulation of Sir2 protein deacetylases FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS 4613 upregulate cellular sirtuins. However, great care will be needed to avoid crossreactivity with other nicotinamide utilizing enzymes, in particular, those involved in NAD + salvage and synthesis. Conclusions In summary, we suggest that small molecule regulation of sirtuins involves the cellular balance of NAD + to nicotinamide, controlled by enzymes involved in NAD + synthesis or salvage. Small global alterations in NAD + levels would provide insufficient changes in Sir2 activity, but microdomains of NAD + produced on location may be an effective regulatory mechanism. We predict that some of these NAD + synthetic enzymes might be components of sirtuin complexes, channelling NAD + directly to Sir2 enzymes. Resveratrol was reported to be a general sirtuin acti- vator; however, recent reports question the validity of that proposal and that resveratrol-dependent lifespan increases are mediated directly by ySir2 activation. Although mammalian SIRT1 appears to be activated by resveratrol treatment, the mechanistic basis for this cellular phenomenon remains to be elucidated. Small molecule inhibitors (such as splitomicin and sirtinol) were identified based on phenotypic screening for compounds that phenocopy a ySir2 yeast deletion. So far, these compounds only inhibit at the micro- molar level, and a full evaluation of their selectivity for other sirtuins has not been determined. Future rational inhibitor design and direct high-throughput screening against all sirtuins, particularly the mammalian homo- logues, undoubtedly will lead to the development of highly selective and potent inhibitors. These com- pounds will provide an essential tool to uncover the cellular functions of these enzymes and may lead to therapeutics that target individual sirtuins. Acknowledgements This work was supported by NIH Grant GM065386 (to J.M.D.) and by NIH Biotechnology Training Grant NIH 5 T32 G08349 (to B.C.S.). References 1 Gray SG & Ekstrom TJ (2001) The human histone de- acetylase family. Exp Cell Res 262, 75–83. 2 Frye RA (1999) Characterization of five human cDNAs with homology to the yeast SIR2 gene: Sir2-like pro- teins (sirtuins) metabolize NAD and may have protein ADP-ribosyltransferase activity. Biochem Biophys Res Commun 260, 273–279. 3 Frye RA (2000) Phylogenetic classification of prokary- otic and eukaryotic Sir2-like proteins. Biochem Biophys Res Commun 273, 793–798. 4 Blander G & Guarente L (2004) The Sir2 family of pro- tein deacetylases. Annu Rev Biochem 73, 417–435. 5 Gasser SM & Cockell MM (2001) The molecular biol- ogy of the SIR proteins. Gene 279, 1–16. 6 Rusche LN & Rine J (2001) Conversion of a gene-speci- fic repressor to a regional silencer. Genes Dev 15, 955– 967. 7 Kaeberlein M, McVey M & Guarente L (1999) The SIR2 ⁄ 3 ⁄ 4 complex and SIR2 alone promote longevity in Saccharomyces cerevisiae by two different mecha- nisms. Genes Dev 13, 2570–2580. 8 Tissenbaum HA & Guarente L (2001) Increased dosage of a sir-2 gene extends lifespan in Caenorhabditis ele- gans. Nature 410, 227–230. 9 Rogina B & Helfand SL (2004) Sir2 mediates longevity in the fly through a pathway related to calorie restric- tion. Proc Natl Acad Sci USA 101, 15998–16003. 10 Brunet A, Sweeney LB, Sturgill JF, Chua KF, Greer PL, Lin Y et al. (2004) Stress-dependent regulation of FOXO transcription factors by the SIRT1 deacetylase. Science 303, 2011–2015. 11 Motta MC, Divecha N, Lemieux M, Kamel C, Chen D, Gu W et al. (2004) Mammalian SIRT1 represses fork- head transcription factors. Cell 116, 551–563. 12 Daitoku H, Hatta M, Matsuzaki H, Aratani S, Ohshima T, Miyagishi M et al. (2004) Silent information regulator 2 potentiates Foxo1-mediated transcription through its deacetylase activity. Proc Natl Acad Sci USA 101, 10042–10047. 13 Vaziri H, Dessain SK, Ng Eaton E, Imai SI, Frye RA, Pandita TK et al. (2001) hSIR2 (SIRT1) functions as an NAD-dependent p53 deacetylase. Cell 107, 149–159. 14 Luo J, Nikolaev AY, Imai S, Chen D, Su F, Shiloh A et al. (2001) Negative control of p53 by Sir2alpha pro- motes cell survival under stress. Cell 107, 137–148. 15 Langley E, Pearson M, Faretta M, Bauer UM, Frye RA, Minucci S et al. (2002) Human SIR2 deacetylates p53 and antagonizes PML ⁄ p53-induced cellular senes- cence. EMBO J 21, 2383–2396. 16 Fulco M, Schiltz RL, Iezzi S, King MT, Zhao P, Kashi- waya Y et al. (2003) Sir2 regulates skeletal muscle dif- ferentiation as a potential sensor of the redox state. Mol Cell 12, 51–62. 17 Picard F, Kurtev M, Chung N, Topark-Ngarm A, Sena- wong T, Machado De Oliveira R et al. (2004) Sirt1 pro- motes fat mobilization in white adipocytes by repressing PPAR-gamma. Nature 429, 771–776. 18 Liszt G, Ford E, Kurtev M & Guarente L (2005) Mouse Sir2 homolog SIRT6 is a nuclear ADP-ribosyl- transferase. J Biol Chem 280, 21313–21320. 19 Imai S, Armstrong CM, Kaeberlein M & Guarente L (2000) Transcriptional silencing and longevity protein Small molecule regulation of Sir2 protein deacetylases O. Grubisha et al. 4614 FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS Sir2 is an NAD-dependent histone deacetylase. Nature 403, 795–800. 20 Smith JS, Brachmann CB, Celic I, Kenna MA, Muham- mad S, Starai VJ et al. (2000) A phylogenetically con- served NAD + -dependent protein deacetylase activity in the Sir2 protein family. Proc Natl Acad Sci USA 97, 6658–6663. 21 Landry J, Slama JT & Sternglanz R (2000) Role of NAD + in the deacetylase activity of the SIR2-like pro- teins. Biochem Biophys Res Commun 278, 685–690. 22 Landry J, Sutton A, Tafrov ST, Heller RC, Stebbins J, Pillus L & Sternglanz R (2000) The silencing pro- tein SIR2 and its homologs are NAD-dependent pro- tein deacetylases. Proc Natl Acad Sci USA 97, 5807– 5811. 23 Jackson MD & Denu JM (2002) Structural identification of 2¢- and 3¢-O-acetyl-ADP-ribose as novel metabolites derived from the Sir2 family of beta-NAD + -dependent histone ⁄ protein deacetylases. J Biol Chem 277, 18535– 18544. 24 Sauve AA, Celic I, Avalos J, Deng H, Boeke JD & Schramm VL (2001) Chemistry of gene silencing: the mechanism of NAD + -dependent deacetylation reac- tions. Biochemistry 40, 15456–15463. 25 Tanner KG, Landry J, Sternglanz R & Denu JM (2000) Silent information regulator 2 family of NAD- depen- dent histone ⁄ protein deacetylases generates a unique product, 1-O-acetyl-ADP-ribose. Proc Natl Acad Sci USA 97, 14178–14182. 26 Tanny JC & Moazed D (2001) Coupling of histone deacetylation to NAD breakdown by the yeast silencing protein Sir2: Evidence for acetyl transfer from substrate to an NAD breakdown product. Proc Natl Acad Sci USA 98, 415–420. 27 Borra MT, O’Neill FJ, Jackson MD, Marshall B, Verdin E, Foltz KR & Denu JM (2002) Conserved enzymatic production and biological effect of O-acetyl- ADP-ribose by silent information regulator 2-like NAD + -dependent deacetylases. J Biol Chem 277, 12632–12641. 28 Rafty LA, Schmidt MT, Perraud AL, Scharenberg AM & Denu JM (2002) Analysis of O-acetyl-ADP-ribose as a target for Nudix ADP-ribose hydrolases. J Biol Chem 277, 47114–47122. 29 Denu JM (2003) Linking chromatin function with meta- bolic networks: Sir2 family of NAD + -dependent deacetylases. Trends Biochem Sci 28, 41–48. 30 Liou GG, Tanny JC, Kruger RG, Walz T & Moazed D (2005) Assembly of the SIR complex and its regulation by O-acetyl-ADP-ribose, a product of NAD-dependent histone deacetylation. Cell 121, 515–527. 31 Bitterman KJ, Medvedik O & Sinclair DA (2003) Long- evity regulation in Saccharomyces cerevisiae: linking metabolism, genome stability, and heterochromatin. Microbiol Mol Biol Rev 67, 376–399. 32 Guarente L & Kenyon C (2000) Genetic pathways that regulate ageing in model organisms. Nature 408, 255– 262. 33 Hasty P (2001) The impact energy metabolism and gen- ome maintenance have on longevity and senescence: lessons from yeast to mammals. Mech Ageing Dev 122, 1651–1662. 34 Anderson RM, Latorre-Esteves M, Neves AR, Lavu S, Medvedik O, Taylor C et al. (2003) Yeast life-span extension by calorie restriction is independent of NAD fluctuation. Science 302, 2124–2126. 35 Lin SJ, Ford E, Haigis M, Liszt G & Guarente L (2004) Calorie restriction extends yeast life span by low- ering the level of NADH. Genes Dev 18, 12–16. 36 Schmidt MT, Smith BC, Jackson MD & Denu JM (2004) Coenzyme specificity of Sir2 protein deacetylases: implications for physiological regulation. J Biol Chem 279, 40122–40129. 37 Berger F, Ramirez-Hernandez MH & Ziegler M (2004) The new life of a centenarian: signalling functions of NAD(P). Trends Biochem Sci 29, 111–118. 38 Sauve AA & Schramm VL (2003) Sir2 regulation by nicotinamide results from switching between base exchange and deacetylation chemistry. Biochemistry 42, 9249–9256. 39 Jackson MD, Schmidt MT, Oppenheimer NJ & Denu JM (2003) Mechanism of nicotinamide inhibition and transglycosidation by Sir2 histone ⁄ protein deacetylases. J Biol Chem 278, 50985–50998. 40 Bitterman KJ, Anderson RM, Cohen HY, Latorre-Est- eves M & Sinclair DA (2002) Inhibition of silencing and accelerated aging by nicotinamide, a putative negative regulator of yeast sir2 and human SIRT1. J Biol Chem 277, 45099–45107. 41 Sauve AA, Moir RD, Schramm VL & Willis IM (2005) Chemical activation of sir2-dependent silencing by relief of nicotinamide inhibition. Mol Cell 17, 595–601. 42 Anderson RM, Bitterman KJ, Wood JG, Medvedik O, Cohen H, Lin SS, Manchester JK, Gordon JI & Sinclair DA (2002) Manipulation of a nuclear NAD + salvage pathway delays aging without altering steady-state NAD + levels. J Biol Chem 277, 18881–18890. 43 Gallo CM, Smith DL Jr & Smith JS (2004) Nicotina- mide clearance by Pnc1 directly regulates Sir2- mediated silencing and longevity. Mol Cell Biol 24, 1301–1312. 44 Sandmeier JJ, Celic I, Boeke JD & Smith JS (2002) Telomeric and rDNA silencing in Saccharomyces cerevi- siae are dependent on a nuclear NAD + salvage path- way. Genetics 160, 877–889. 45 Revollo JR, Grimm AA & Imai S (2004) The NAD bio- synthesis pathway mediated by nicotinamide phospho- ribosyltransferase regulates Sir2 activity in mammalian cells. J Biol Chem 279, 50754–50763. O. Grubisha et al. Small molecule regulation of Sir2 protein deacetylases FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS 4615 46 Araki T, Sasaki Y & Milbrandt J (2004) Increased nuclear NAD biosynthesis and SIRT1 activation pre- vent axonal degeneration. Science 305 , 1010–1013. 47 Starai VJ, Celic I, Cole RN, Boeke JD & Escalante- Semerena JC (2002) Sir2-dependent activation of acetyl- CoA synthetase by deacetylation of active lysine. Science 298, 2390–2392. 48 Rodgers JT, Lerin C, Haas W, Gygi SP, Spiegelman BM & Puigserver P (2005) Nutrient control of glucose homeostasis through a complex of PGC-1alpha and SIRT1. Nature 434, 113–118. 49 North BJ, Marshall BL, Borra MT, Denu JM & Verdin E (2003) The human Sir2 ortholog, SIRT2, is an NAD+-dependent tubulin deacetylase. Mol Cell 11, 437–444. 50 Borra MT, Langer MR, Slama JT & Denu JM (2004) Substrate specificity and kinetic mechanism of the Sir2 family of NAD + -dependent histone ⁄ protein deacetyl- ases. Biochemistry 43, 9877–9887. 51 Grozinger CM, Chao ED, Blackwell HE, Moazed D & Schreiber SL (2001) Identification of a class of small molecule inhibitors of the sirtuin family of NAD-depen- dent deacetylases by phenotypic screening. J Biol Chem 276, 38837–38843. 52 Bedalov A, Gatbonton T, Irvine WP, Gottschling DE & Simon JA (2001) Identification of a small molecule inhi- bitor of Sir2p. Proc Natl Acad Sci USA 98, 15113– 15118. 53 Posakony J, Hirao M, Stevens S, Simon JA & Bedalov A (2004) Inhibitors of Sir2: evaluation of splitomicin analogues. J Medical Chem 47, 2635–2644. 54 Hirao M, Posakony J, Nelson M, Hruby H, Jung M, Simon JA & Bedalov A (2003) Identification of selective inhibitors of NAD + -dependent deacetylases using phen- otypic screens in yeast. J Biol Chem 278, 52773–52782. 55 Yeung F, Hoberg JE, Ramsey CS, Keller MD, Jones DR, Frye RA & Mayo MW (2004) Modulation of NF-jB-dependent transcription and cell survival by the SIRT1 deacetylase. EMBO J 23, 2369–2380. 56 Tervo AJ, Kyrylenko S, Niskanen P, Salminen A, Leppanen J, Nyronen TH et al. (2004) An in silico approach to discovering novel inhibitors of human sirtuin type 2. J Medical Chem 47, 6292–6298. 57 Magni G, Amici A, Emanuelli M, Raffaelli N & Ruggieri S (1999) Enzymology of NAD+ synthesis. Adv Enzymol Relat Areas Mol Biol 73, 135–182, xi. 58 Szabo C (2003) Nicotinamide: a jack of all trades (but master of none?). Intensive Care Med 29, 863–866. 59 Sinclair DA & Guarente L (1997) Extrachromosomal rDNA circles – a cause of aging in yeast. Cell 91, 1033– 1042. 60 Falcon AA & Aris JP (2003) Plasmid accumulation reduces life span in Saccharomyces cerevisiae. J Biol Chem 278, 41607–41617. 61 Lin SJ, Defossez PA & Guarente L (2000) Requirement of NAD and SIR2 for life-span extension by calorie restriction in Saccharomyces cerevisiae. Science 289, 2126–2128. 62 Gottlieb S & Esposito RE (1989) A new role for a yeast transcriptional silencer gene, SIR2, in regulation of recombination in ribosomal DNA. Cell 56, 771–776. 63 Koubova J & Guarente L (2003) How does calorie restriction work? Genes Dev 17, 313–321. 64 Masoro EJ (2000) Caloric restriction and aging: an update. Exp Gerontol 35, 299–305. 65 Lin SJ, Kaeberlein M, Andalis AA, Sturtz LA, Defossez PA, Culotta VC et al. (2002) Calorie restriction extends Saccharomyces cerevisiae lifespan by increasing respir- ation. Nature 418, 344–348. 66 Kaeberlein M, Kirkland KT, Fields S & Kennedy BK (2004) Sir2-independent life span extension by calorie restriction in yeast. PLoS Biol 2, E296. 67 Kaeberlein M, McDonagh T, Heltweg B, Hixon J, Westman EA, Caldwell S et al. (2005) Substrate specific activation fo sirtuins by resveratrol. J Biol Chem 280, 17038–17045. 68 Howitz KT, Bitterman KJ, Cohen HY, Lamming DW, Lavu S, Wood JG et al. (2003) Small molecule activa- tors of sirtuins extend Saccharomyces cerevisiae lifespan. Nature 425, 191–196. 69 Pervaiz S (2003) Resveratrol: from grapevines to mammalian biology. FASEB J 17, 1975–1985. 70 Wood JG, Rogina B, Lavu S, Howitz K, Helfand SL, Tatar M & Sinclair D (2004) Sirtuin activators mimic caloric restriction and delay ageing in metazoans. Nature 430, 686–689. 71 Borra MT, Smith BC & Denu JM (2005) Mechanism of human SIRT1 activation by resveratrol. J Biol Chem 280, 17187–17195. 72 Balaban RS, Nemoto S & Finkel T (2005) Mitochon- dria, oxidants, and aging. Cell 120, 483–495. Small molecule regulation of Sir2 protein deacetylases O. Grubisha et al. 4616 FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS . deacetylases. (B) Examples of known activators of Sir2 deacetylases. O. Grubisha et al. Small molecule regulation of Sir2 protein deacetylases FEBS Journal 272 (2005). longevity protein Small molecule regulation of Sir2 protein deacetylases O. Grubisha et al. 4614 FEBS Journal 272 (2005) 4607–4616 ª 2005 FEBS Sir2 is an