Gskg

DSCATLV|_

_ 3JK1GEP* _ BllgycddxaaB┬╗<}kcgmtiphkkw-lr a w lv 1

of just the bacterial Sir2p species is included as well. Note that the bacterial species have much shorter N and C termini and are more conserved as a group. Interestingly, the region surrounding the GAGISTS motif does have some weak similarity to a region of bacterial 6-phosphogluconate dehydrogenases, an NAD+-binding protein.16 This enzyme uses NADP+ or NAD+ as a cofactor in its activity and one-half of its binding site is similar to Sir2p. Indeed, in the crystal structure of an archaeal Sir2 protein, these motifs contain NAD+-binding residues.17 Sir2p has been reported to have a weak mono(ADP)-ribosyltransferase activity in addition to its deacetylase activity.16 Beyond these small features, the Sir2 protein family does not have any other characteristics of traditional histone deacetylases, which is actually consistent with its unusual reaction mechanism.

Several publications have examined the reaction mechanism and discovered that NAD+ is not simply a cofactor, but is actually turned over.18-20 For every acetyl group removed from a lysine residue, one molecule of NAD+ is hydrolyzed to form one molecule of nicotinamide and one molecule of a novel compound called 1'- (9-acety 1-ADP-ribose.'9'20 This compound is formed by transfer of the acetyl group from the acetyllysine-containing peptide substrate to the ADP-ribose moiety of NAD+. Several different chemical mechanisms have been proposed for this reaction, including both covalent enzyme-ADP-ribose intermediates as in the case of ADP-ribosyltransferases and more complex mechanisms that do not require such intermediates. The stereochemistry of the acetyl-ADP-ribose should definitively settle the matter. The existence of this novel compound raises a whole new series of questions, as this molecule could play a signaling role in the cell.20

Sir2p as NAD+-Dependent Protein/Histone Deacetylase

Even though Sir2p does not have any homology to other known histone deacetylases, its high degree of evolutionary conservation suggested that it might still be an enzyme. Speculation that it might be a deacetylase originated from the Broach laboratory, which demonstrated that SIR2 overexpression resulted in bulk histone hypoacetylation,21 which was consistent with Sir2p being a deacetylase. Multiple laboratories unsuccessfully attempted to show that Sir2p was a conventional histone deacetylase. The key breakthrough came when a protein involved in vitamin Bi2 (cobalamin) synthesis in Salmonella typhimurium (CobB) was found to be a SIR2 family member.22 This was important because CobB can carry out a phosphoribosyltransferase-like reaction utilizing an NAD+ intermediate (NaMN)

16 J. C. Tanny, G. J. Dowd, J. Huang, H. Hilz, and D. Moazed, Cell 99, 735 (1999).

17 J. Min, J. Landry, R. Sternglanz, and R. M. Xu, Cell 105, 269 (2001).

18 J. Landry, J. T. Slama, and R. Sternglanz, Biochem. Biophys. Res. Commun. 278, 685 (2000).

19 J. C. Tanny and D. Moazed, Proc. Natl. Acad. Sci. U.S.A. 98,415 (2001).

20 K. G. Tanner, J. Landry, R. Sternglanz, and J. M. Denu, Proc. Natl. Acad. Sci. U.S.A. 97, 14178 (2000).

21 M. Braunstein, A. B. Rose, S. G. Holmes, C. D. Allis, and J. R. Broach, Genes Dev. 7, 592 (1993).

22 A. W. Tsang and J. C. Escalante-Semerena, J. Biol. Chem. 273, 31788 (1998).

as a substrate. It was subsequently demonstrated that CobB and a human Sir2 family member could transfer a radiolabel from NAD+ to an exogenous protein (bovine serum albumin, BSA) in vitroP This was interpreted to mean that Sir2p had protein-ADP-ribosyltransferase activity. Yeast Sir2p was then reported to have a weak mono(ADP)-ribosyltransferase activity on histones and itself in vitro}6

NAD+ is the substrate for ADP-ribosyltransferase activity of Sir2p and therefore is included in the in vitro reactions. In testing whether acetylated histones are substrates for ADP-ribosylation, the Guarente and Sternglanz laboratories found that purified Sir2p also had an NAD+-dependent histone deacetylase activity with specificity forLys-16 of histone H4 andLys-9 and Lys-14 of histone H3.5'24 Other Sir2 family members, including the bacterial CobB protein and an Archea bacterial homolog, also harbored NAD+-dependent deacetylase activity.7 This activity can easily be detected in vitro with whole cell yeast extracts and hyperacetylated chicken histones, which allows direct testing of mutant cells for the presence of activity.7 Interestingly, the Hst2 protein contributes most of the NAD+-dependent deacetylase activity from whole cell extracts.7 This may reflect the fact that Hst2p is localized in the cytoplasm and is simply extracted more efficiently.25 A study that directly compares the activity of multiple purified Sir2 family members has not been carried out to date.

Another significant question remaining concerns the in vivo targets for the Sir2 family members. At this time it is still not clear whether Sir2p deacetylates histones H3 and H4 in vivo, even though some in vitro targets (histones H3 and H4) have been identified. Bacteria do not have histones, and therefore the targets for their Sir2-like proteins will definitely not be histones. Furthermore, the CobB protein already has a known function in cobalamin synthesis that, interestingly, is more similar to ADP-ribosyltransferase activity. Interestingly, yeast SIR2 and a human SIR2 (SIR2A) can partially complement a cobB mutant Salmonella cell line, suggesting that whatever activity CobB protein has, the yeast and human proteins can also do it (C. B. Brachmann, E. Caputo, J. S. Smith, I. Celic, V. J. Starai, J. C. Escalante-Semerena, and J. D. Boeke, unpublished data, 2000). The deacetylation activity of Sir2p has been definitively shown to be important for silencing, but the putative ADP-ribosyltransferase activity was not required. This does not rule out the possibility that ADP-ribosylation could occur in vivo and be involved in a different process.

Three-Dimensional Structure of Archaeal Sir2p

The structure of an Sir2 protein from the archaebacterium Archaeoglobus fulgidus has been determined.17 This species contains two Sir2-like genes; the

23 R. A. Frye, Biochem. Biophys. Res. Commun. 260,273 (1999).

24 S.-i. Imai, C. M. Armstrong, M. Kaeberlein, and L. Guarente, Nature (London) 403,795 (2000).

25 S. Perrod, M. M. Cockell, T. Laroche, H. Renauld, A. L. Ducrest, C. Bonnard, and S. M. Gasser, EM BO J. 20,197 (2001).

encoded proteins share 64% sequence identity, suggesting that the two proteins have diverged, possibly to recognize different substrates. Indeed, one of the two proteins, Sir-Af2, can recognize and deacetylate histones,7 whereas the other cannot.17

The structure of Sir2p, like that of many other nucleotide-binding enzymes, is based on a Rossman fold (Fig. 2). The NAD+ sits on the surface of the Rossman fold but in an unusual conformation relative to that of other NAD+-binding enzymes. The nicotinamide moiety is inferred to inhabit a sizable cavity in the surface, within which it is presumed that catalysis occurs. The roof of the cavity is formed by a flexible loop that assumes different conformations in two different crystal forms

Fig. 2. Surface of Sir2-Afl from crystal structure, showing invariant residues in green, chemically conserved residues in blue, and NAD+ in yellow. The Rossman fold is the lower half of the structure; the deep cleft on the left side is a proposed acetyllysine side chain-binding site.

studied. Protruding from the Rossman fold domain are two inserted sequences: a zinc-binding domain and a helical domain that connects the loop to the Rossman fold.

Regulation of Silencing by Cellular NAD+ Concentration

The finding that Sir2p is an NAD+-dependent histone deacetylase raises the possibility that chromatin structure associated with silencing is linked to cellular metabolism. Indeed, mutations in an NAD+ salvage pathway gene, NPT1, have been shown to compromise all three forms of silencing in yeast.7 Deletion of nptl also causes a 3- to 4-fold decrease in the intracellular NAD+ concentration.7 Because Sir2p depends on NAD+ for its silencing function, the reduction in cellular NAD+ concentration likely represses the deacetylation activity of Sir2p. Another metabolism/silencing/chromatin link is that the extension of yeast life span by caloric restriction depends on SIR2 and NPT1.6 On the basis of these findings it has been proposed that Sir2p is a molecular sensor of changes in the cellular redox/metabolic state that translates this information to alterations in chromatin structure.6,26 In this chapter we describe methods associated with the analysis of SIR2 function in silencing and the deacetylation reactions.

Methods

A variety of genetic and biochemical methods useful for the study of Sir2 and silencing is included below.

Silencing Assays

Transcriptional silencing can be assayed by several different methods including RNA blot analysis, efficiency of plating (spot assays), and colony color assays. The assays are based on repression of a URA3 or MET15 reporter gene that is integrated into the ribosomal DNA tandem array. Examples of the two different assays showing the effect of deleting SIR2 are shown in Fig. 3.

Spot Assays

Grow the appropriate reporter strains as 1- to 2-cm2 patches on rich yeast extract-peptone-dextrose (YPD) agar plates overnight at 30┬░. Use selective medium when the strains harbor a selectable plasmid. All media contain glucose as the carbon source. Using a wooden applicator stick, scrape cells from the agar surface and resuspend them in 1 ml of sterile water in a 1.5-ml microcentrifuge tube. Vortex the cell mixture and measure the absorbance of the mixture at 600 nm

0 0

Post a comment