KDAC VARIANTS AND USES THEREOF

Abstract

The invention provides a method of selecting a mutant polypeptide having lysine demodification, in particular lysine deacylation, activity, wherein the method comprises the following steps (a) incubating a mutant polypeptide having an amino acid sequence with at least 80% sequence identity to SEQ ID NO: 1 with a peptide or polypeptide comprising an inactivated essential lysine residue; and (b) determining the activity of the mutant polypeptide to activate the peptide or polypeptide comprising the inactivated essential lysine residue, wherein the mutant polypeptide and the peptide or polypeptide comprising an inactivated essential lysine residue are incubated in a biological cell. The invention furthermore relates to an acylated luciferase, particularly Firefly luciferase, and uses thereof. The present invention furthermore relates to a mutant polypeptide comprising an amino acid sequence having at least 98% sequence homology with SEQ ID NOs: 2, 3, 4, 5 or 6 and having lysine demodification, in particular lysine deacylation, activity, wherein the mutant polypeptide is not identical to SEQ ID NO: 1. The invention also relates to the mutant polypeptide of the invention and a peptide or polypeptide comprising an inactivated essential lysine residue for use in treating cancer.

Claims

1. A method of selecting a polypeptide having lysine demodification, in particular lysine deacylation, activity from a collection of polypeptides, wherein the method comprises the following steps: (a) incubating said polypeptide with a peptide or polypeptide comprising an essential lysine residue inactivated by a modification, in particular an acylation, of said essential lysine residue; and (b) selecting said polypeptide based on the ability of said polypeptide to activate said peptide or polypeptide comprising the inactivated essential lysine residue, wherein said polypeptide and said peptide or polypeptide comprising an inactivated essential lysine residue are incubated in a biological cell.

2. The method of claim 1 further comprising the following counter-selection steps: (c) incubating a polypeptide selected in step (b) with a peptide or polypeptide comprising an essential lysine residue differentially inactivated by a modification different from the modification used in step (a); and (d) selecting said polypeptide based on the inability of said polypeptide to activate said peptide or polypeptide comprising said differentially inactivated essential lysine residue.

3. A method of screening a diverse collection of polypeptides for a polypeptide having lysine demodification, in particular lysine deacylation, activity, wherein the method comprises the following steps: (a) incubating said diverse collection of polypeptides with a luciferase comprising an inactivated residue K529, wherein said residue is inactivated by a modification, in particular an acylation; and (b) selecting said polypeptide based on the ability of said polypeptide to activate said luciferase, wherein said diverse collection and said luciferase are incubated in a diverse collection of biological cells; particularly wherein said luciferase is Firefly luciferase according to SEQ ID NO: 7.

4. The method of claim 3 further comprising the following counter-screening steps: (c) incubating a polypeptide selected in step (b) with a luciferase comprising an inactivated residue K529, where said residue is differentially inactivated by a modification different from the modification used in step (a); and (d) screening said polypeptide based on the inability of said polypeptide to activate said luciferase comprising said differentially inactivated residue K529.

5. A method of screening or selecting a KDAC inhibitor from a diverse collection of putative KDAC inhibitors, wherein the method comprises the following steps: (a) incubating a polypeptide having a lysine demodification, in particular a lysine deacylation, activity with a member of said diverse collection; (b) adding a peptide or polypeptide comprising an essential lysine residue inactivated by a modification, in particular an acylation, of said essential lysine residue; and (c) identifying a KDAC inhibitor by the ability to inhibit the demodification, in particular the deacetylation, activity of said polypeptide, wherein the KDAC inhibiting activity of said KDAC inhibitor is reciprocal to the activity of said polypeptide to activate the peptide or polypeptide comprising the inactivated essential lysine residue; in particular, wherein the method is performed in a biological cell.

6. The method of claim 1, 2, or 5, wherein the peptide or polypeptide comprising an essential lysine residue inactivated by a modification is OMP decarboxylase.

7. The method of claim 6, wherein OMP decarboxylase is buddying yeast OMP decarboxylase (Ura3) or E. coli pyrF.

8. The method of claim 7, wherein OMP decarboxylase is buddying yeast OMP decarboxylase (Ura3) comprising an inactivated residue K93.

9. The method of claim 5, wherein the peptide or polypeptide an essential lysine residue inactivated by a modification is a luciferase comprising an inactivated residue K529; particularly wherein said luciferase is Firefly luciferase according to SEQ ID NO: 7.

10. The method of claim 9, wherein the luciferase comprises an amino acid sequence having at least 90% sequence homology to SEQ ID NO: 7, particularly wherein said luciferase is Firefly luciferase comprising the sequence according to SEQ ID NO: 7.

11. The method of any one of claims 1 to 10, wherein the essential lysine residue is inactivated by acylation or an alternative protection group, particularly by acylation.

12. The method of claim 11, wherein the essential lysine residue is inactivated by acylation with an acyl group selected from the groups of acetyl, crotonyl, tert.-butyloxycarbonyl (Boc), allyloxycarbonyl (Aloc), propargyloxycarbonyl (Poc), benzyloxycarbonyl (Z), 2,2,2-trichloroethyloxycarbonyl (Troc), azidomethoxycarbonyl (Azoc), 2-chlorobenzyloxycarbonyl (ClZ) and trifluoroacetyl (tfa).

13. The method of any one of claims 1 to 12, wherein the biological cell is a bacterial cell, in particular wherein the bacterial cell is an E. coli cell.

14. The method of claim 13, wherein the bacterial cell is an E. coli cell, which lacks a gene encoding pyrF and/or cobB and/or wherein the activity of pyrF and/or cobB is inhibited in said E. coli cell.

15. A luciferase, in particular a luciferase comprising an amino acid sequence having at least 90% sequence homology to SEQ ID NO: 7, wherein the polypeptide comprises an inactivated lysine residue at a position corresponding to position 529 of SEQ ID NO: 7; particularly wherein the polypeptide comprises the sequence according to SEQ ID NO: 7.

16. The polypeptide of claim 15, wherein the lysine residue is inactivated by acylation, in particular by acylation with an acyl group selected from the groups of acetyl, crotonyl, tert.-butyloxycarbonyl (Boc), allyloxycarbonyl (Aloc), propargyloxycarbonyl (Poc), benzyloxycarbonyl (Z), 2,2,2-trichloroethyloxycarbonyl (Troc), azidomethoxycarbonyl (Azoc), 2-chlorobenzyloxycarbonyl (ClZ) and trifluoroacetyl (tfa).

17. The polypeptide of claim 15 or 16, additionally comprising a purification tag, preferably a 6 His-tag.

18. A nucleic acid encoding the polypeptide of claim 15 or 16, wherein the codon encoding the essential lysine residue is replaced by an amber stop codon.

19. The nucleic acid of claim 18 comprising a nucleic acid sequence having at least 80% sequence homology to SEQ ID NO: 8; particularly a nucleic acid sequence encoding the protein according to SEQ ID NO: 7. wherein the codon encoding the essential lysine residue is replaced by an amber stop codon.

20. A mutant polypeptide comprising an amino acid sequence having at least 99% sequence homology with SEQ ID NOs: 2, 3, 4, 5 or 6 and having lysine demodification, in particular lysine deacylation, activity, wherein the mutant polypeptide is not identical to SEQ ID NO: 1.

Description

FIGURES

[0190] FIG. 1: E. coli producing Ura3 K93ac as the sole source of OMP decarboxylase depend on KDAC activity. E. coli DB6566 (pyrF) expressing plasmids to encode wild-type Ura3, Ura3 K93ac or K93boc were plated on agar plates with or without uracil, 5-FOA and the corresponding unnatural amino acid. Nicotinamide (NAM) was added to inhibit endogenous CobB.

[0191] FIG. 2: CobB library design. Highlighted amino acid residues were randomized to all possible combinations of natural amino acids to generate a library of thirty million mutants in the active site of CobB.

[0192] FIG. 3: Evolved CobB variants can activate Firefly luciferase modified at K529. Dual luciferase reporter (DLR) assays were performed with E. coli producing DLR with the indicated modification of K529 in the Firefly enzyme. Activities were normalized to genetically fused Renilla luciferase and plotted relative to the activity observed for wild-type CobB on the same substrate. A) CobB mutant able to discriminate crotonyl-over acetyl-lysine. B) CobB mutants active against protected forms of lysine.

[0193] FIG. 4: Quantification of SirT2 activity using Fluor-de-Lys and FLuc assays. Deacetylation assays were performed at various concentrations of SirT2 using either 10 M Fluor-de-Lys peptide or 30 nM acetylated FLuc under otherwise identical conditions. KDAC activities were measured by fluorescence (ex. 355/em. 460 nm) or luminescence and normalized to the activity at the highest SirT2 concentration after background subtraction. Luminescence was measured in endpoint or continuous format. The shaded areas indicate respective linear response ranges. Error bars are standard deviation of the means from triplicate measurements.

[0194] FIG. 5: Quantification of CobB activity using Fluor-de-Lys and FLuc assays. Deacetylation assays were performed at various concentrations of CobB using either 10 M Fluor-de-Lys peptide or 30 nM acetylated FLuc under otherwise identical conditions. KDAC activities were measured by fluorescence (ex. 355/em. 460 nm) or luminescence and normalized to the activity at the highest CobB concentration. Luminescence was measured in endpoint format. Error bars are standard deviation of the means from triplicate measurements.

[0195] FIG. 6: Inhibition of SirT2 by nicotinamide. Acetylated FLuc (30 nM) was deacetylated with SirT2 (8 g/ml) in the presence of 1 mM NAD.sup.+ and increasing concentrations of nicotinamide. Error bars are standard deviations of the means from three independent reactions. IC.sub.50 for nicotinamide inhibition of SirT2 is approximately 30 M.

[0196] FIG. 7: FLuc K529ac activation by different KDACs. Sirtuins were used at 20 g/ml, while HDAC8 concentration was 10 mg/ml with tenfold higher FLuc K529ac. All error bars are standard deviations of triplicates.

[0197] FIG. 8: Effectors identified in FLuc K529ac-based screen. A) Inhibitors of SirT2 (IC.sub.50 of 1-7 [M]: 3.2; 5.5; 17; >30; 13; 15; 11. n=2) B) SirT2 activator (1.6 fold at 30 M) C) Firefly luciferase inhibitor (IC.sub.50: 3 M).

EXAMPLES

Example 1Design of a Selection System for Lysine Deacetylases

[0198] To develop a selection system for KDACs, enzymes with lysine residues essential for activity were searched. Two of these enzymes were tested, orotidine-5-phosphate (OMP) decarboxylase and firefly luciferase. Both proved to be suitable as selectable marker and reporter enzyme, respectively. When N()-acetyl-lysine was incorporated in place of K93 of budding yeast OMP decarboxylase (Ura3), the protein was unable to support growth of E. coli cells lacking pyrF (the homologue of Ura3) and cobB (the major lysine deacetylase of E. coli, inhibited with nicotinamide) in the absence of uracil (FIG. 1, column 3, bottom). In the presence of cobB, robust growth was observed on minimal medium without uracil, indicating that cobB was able to remove the acetyl group from the active site lysine of Ura3 K93ac. Growth of the same cells is inhibited when 5-fluoro-orotic acid, a compound converted into a toxic metabolite by Ura3, is added to the medium. Hence, this system is able to positively and negatively select E. coli harboring an active lysine deacetylase. The same system can also be used to select for HDAC8 activity, a mammalian class I lysine deacetylase structurally and mechanistically distinct from the sirtuin family member cobB.

[0199] Firefly luciferase contains an essential lysine residue (K529) in the active site. Replacing this residue by genetic code expansion with N()-acetyl-lysine rendered the enzyme inactive in the absence of lysine deacetylase cobB. In the presence of cobB, robust activity of the enzyme was observed. Hence, K529ac firefly luciferase can be used to screen for lysine deacetylase activity in E. coli.

Example 2Creation of cobB Mutant Libraries

[0200] Next, a mutant library was created by randomizing five active site residues (A37, Y53, R56, I92 and V148) of E. coli cobB to all possible combinations of natural amino acids, thereby creating 20.sup.5 (3.210.sup.6) different mutants (FIG. 2).

Example 3Isolation of Acyl-Type Specific Deacetylases

[0201] To identify cobB mutants selectively removing crotonyl but not acetyl groups, the library was subjected to two rounds of selection, positive and negative. Therefore, E. coli DH10B pyrF cobB harbouring a reporter plasmid encoding ura3 K93TAG together with wildtype MbPyIRS and the cognate amber suppressor tRNA MbPyIT was transformed with the cobB mutant library. The cells were challenged to grow in the presence of N()-crotonyl-lysine on medium without uracil to select clones able to decrotonylate Ura3 K93cr. Library plasmids were isolated from the pool of surviving clones and used to transform DH10B pyrF cobB harbouring a reporter plasmid encoding AcKRS3 (M. barkeri PyIRS variant specific for N()-acetyl-lysine) instead of MbPyIRS. Cells were grown on plates containing N()-acetyl-lysine and 5-fluoro-orotic acid (5-FOA), which is toxic to cells in the presence of active Ura3, to select against clones able to remove acetyl groups from Ura3 K93ac. The library member-encoding plasmids of the clones surviving the negative selection were isolated and re-transformed into E. coli DH10B pyrF cobB harbouring a reporter plasmid encoding ura3 K93TAG together with wildtype MbPyIRS and the cognate amber suppressor tRNA MbPyIT, and individual clones were arrayed and tested for the ability to survive on medium without uracil in the presence of N()-crotonyl-lysine. Thereby several mutants of CobB were identified that were able to selectively cleave crotonyl, but not acetyl groups off lysine side chains.

Example 4Selection of Bioorthogonal Eraser Enzymes

[0202] Next, the same cobB mutant library was challenged to remove chemical protection groups from lysine residues. N()-tert.-butyl-oxycarbonyl-lysine (BocK), N()-allyl-oxycarbonyl-lysine (AlocK) and N()-propargyl-oxycarbonyl-lysine (PrK) can be incorporated in proteins using wild-type PyIRS/PyIT. E. coli DH10B pyrF cobB harbouring the mutant library was challenged to grow in the absence of uracil while incorporating one of these unnatural amino acids in Ura3 in place of K93. Surviving clones were arrayed and plasmids isolated from cells that grew in the absence of uracil depending on the presence of one of the unnatural amino acids. Several mutants capable of cleaving AlocK were identified and a single mutant with activity against BocK (Table 1). Individual testing of mutants isolated in the BocK and AlocK selections for activity against PrK revealed several mutants with basal activity.

Example 5Quantitative Analysis of Mutant Activities Using Firefly Luciferase Assay

[0203] The mutants isolated in the selections were tested using Firefly dual luciferase assays. E. coli DH10B pyrF cobB were transformed with plasmids expressing Firefly luciferase with the relevant modification on lysine-529 and the cobB mutants. Luciferase activity was tested directly in whole cell lysates and compared to the activity of wild-type cobB towards the modifications. The activities observed for the evolved KDAC variants correlated well with the activities observed in the uracil selections (FIG. 3).

[0204] The selection system of the invention is capable of identifying an individual KDAC variant with the desired activity in a library of more than three million mutants in a single round. Enzymes could be identified to remove typical protection groups for lysine side chains active enough to activate a sufficient amount of Ura3 enzyme to sustain cell growth in the absence of uracil. The selection system can be easily modified to select other KDAC mutant libraries and other lysine modifications. It may also be used to design selective mutant/inhibitor pairs by a bump-and-hole strategy. Enzymes catalysing bioorthogonal reactions are the key to success for the development of safe prodrug strategies in cancer therapy. Presently, enzymes to activate prodrugs are either of human origin (with the disadvantage of being present in other tissues and therefore causing side effects) or from a different organism (with the disadvantage of being immunogenic). KDAC variants of the invention with bioorthogonal activity evolved from a parent enzyme of human origin combine the advantages of both approaches.

TABLE-US-00001 TABLE 1 Growth on -ura DLR activity [rel. to wild-type] Name Mutations AcK CrK BocK AlocK PrK AcK CrK BocK AlocK Dealocase-1 A37S Y53W R56W I92V n.d. + + + 0.49 5.73 2.88 10.83 SEQ ID 2 Dealocase-2 A37G R56G I92V n.d. + + 1.19 2.11 0.77 10.23 SEQ ID 3 Dealocase-3 I131S V148L + + n.d. 0.35 1.00 0.65 6.00 SEQ ID 4 Debocylase-1 I131V V148L + + + + + 1.25 2.90 8.65 8.35 SEQ ID 5 Decrotonylase-1 A37R Y53G R56T I92R V148L + + n.d. 0.00 0.10 0.04 0.02 SEQ ID 6

Example 6Development of Humanized Variants

[0205] Humanized deacetlyases, i.e. mutant polypeptide of the invention, have been developed. The advantage of a human origin is that there will be no, or only a very reduced, immune reaction in the human organism. For this purpose, the enzymes SirT1, SirT2 and SirT3 are cloned in a manner analogous to the cloning of E. coli cobB. Cloned enzymes are characterized for their ability to activate the marker protein Ura3 K93ac by demodifying the essential lysine residue. Subsequently, mutant libraries are built based on the active variant enzymes. This process is identical to the above-described process based on E. coli cobB.

[0206] Inactive precursor molecules of toxic substances are used in cancer therapy, as it is part of the present invention. For this purpose, toxic peptides are modified at their essential lysine residues using protection groups, acetylation and the like. The resulting peptides are tested on human cell lines for toxicity, whereby a low toxicity is preferred. The evolved deacetylases are then characterized for their ability to remove the protection groups and to activate the pro-toxin.

[0207] The evolved human deacetylases are tested in human cancer cell lines. For this purpose, the polypeptides are expressed in those cell lines. Subsequently, the cell lines are administered with the pro-toxin peptides to test the ability of the deacetylases to activate them and to provide its effects on the cancer cell line.

Example 7KDAC Assay Using Firefly Luciferase K529ac

[0208] Materials

[0209] Plasmids

[0210] pCDF-PyIT-FLuc(opt)His.sub.6-K529TAG: The gene for Firefly Luciferase codon-optimized for expression in E. coli and containing an amber codon replacing the codon for Lys-529 as well as a C-terminal His.sub.6-Tag was custom synthesized by Genscript and cloned into NcoI/XhoI of pCDF-PyIT (Neumann et al., Nat Chem Biol 4 (2008) 232-234). pBK-AcKRS3opt (expressing acetyl-lysyl-tRNA synthetase with mutations improving tRNA binding) was generated from pBK-AcKRS3 by three rounds of QuickChange mutagenesis introducing mutations V31I, T56P, H62Y and A100E (Neumann et al., Molecular Cell 36 (2009) 153-163).

[0211] pBK-His.sub.6-CobB: A PCR product encoding His.sub.6-CobB under the control of an arabinose inducible promoter was amplified from CobB subcloned in a pBAD plasmid. The DNA fragment was digested with BglII/StuI and cloned into BamHI/StuI of pBK-PyIS (Neumann et al., Molecular Cell 36 (2009) 153-163).

[0212] pBK-His.sub.6-hsHDAC8: His.sub.6-hsHDAC8 gene was custom synthesized by GeneArt, amplified introducing NcoI/XbaI sites and cloned into NcoI/XbaI of pBK-His.sub.6-CobB (replacing His.sub.6-CobB).

[0213] pBK-His.sub.6-TEV-hsSirT2 and pBK-His.sub.6-TEV-hsSirT3: The catalytic domain of SirT2 (56-356) and SirT3 (118-399) was amplified from pGEX-TSS-TEV-SirT2/3 introducing NcoI/XbaI sites, His.sub.6-tag and TEV site and cloned into pBK-His.sub.6-hsHDAC8 using the NcoI and XbaI sites. A frameshift in SirT3 was removed by QC.

[0214] Expression of KDACs

[0215] E. coli BL21 DE3 RIL was transformed with the respective pBK plasmids for CobB, HDAC8, SirT2 or SirT3. Cells were incubated at 37 C. in 10 mL LB medium (50 g/mL kanamycin) overnight, used to inoculate 1 L LB medium (50 g/mL kanamycin) and grown to an OD.sub.600 of 0.3. The temperature was reduced to 30 C. for 1 h before expression was induced by addition of arabinose to a final concentration of 0.2%. Cells were harvested after 16 h by centrifugation (20 min, 6000 rpm, 4 C.). The cell pellets were washed with PBS and stored at 20 C.

[0216] Purification of KDACs

[0217] Cell pellets were thawed on ice and resuspended in HEPES-Ni-NTA wash buffer (20 mM HEPES, 200 mM NaCl, 20 mM imidazole, 1 mM DTT; pH 7.5 [CobB/HDAC8] or 8.0 [SirT2/3]) supplemented with lysozyme (0.5 mg/mL), DNase (1 mg) and protease inhibitors (1 mM PMSF and 0.5 Roche Protease Inhibitor cocktail). Lysis was preformed using a pneumatic cell disintegrator. The cell debris was removed by centrifugation (20 min, 20,000 rpm, 4 C.) and HisPur Ni.sup.2+-NTA Resin (2 mL in 50 mL Solution) was added to the supernatant. After 1 h at 4 C. the suspension was loaded on a plastic column (BioRad, Mnchen) with a frit and washed with HEPES-Ni-NTA wash buffer. Protein was eluted in 4 mL Ni-NTA wash buffer containing 200 mM imidazole. The eluate was concentrated and the buffer was exchanged to gelfiltration buffer before loading on a HILoad 26/70 Superdex 200 size-exclusion chromatography column (GE healthcare, UK) preequilibrated with gel filtration buffer (20 mM HEPES, 100 mM NaCl, 10 mM DTT, pH 7.5 [CobB/HDAC8] or 20 mM Tris/HCl, 50 mM NaCl, pH 8 [SirT2/3]). Absorption at 280 nm was monitored and 5 mL fractions collected. Fractions containing protein were analyzed by SDS-PAGE, pooled and concentrated in a microfiltrator (Amicon Ultra-15 Centrifugal Unit, 10 kDa, Merck Millipore). The protein was aliquoted (50 L), flash frozen in liquid nitrogen and stored at 80 C.

[0218] Purification of Firefly Luciferase K529ac

[0219] E. coli BL21 DE3 were transformed with plasmids pCDF-PyIT-FLuc(opt)His6-K529TAG and pBK-AcKRS3opt. Cells were grown in LB medium in the presence of antibiotics (50 g/l spectinomycin and 50 g/l kanamycin) to maintain the plasmids, 5 mM acetyl-lysine and 20 mM nicotinamide at 37 C. to an OD600 of 1.0. Then, cells were shifted to 30 C. and protein expression induced by the addition of 1 mM IPTG. After further 4 h at 30 C. cells were harvested by centrifugation, washed with PBS and lysed in Ni-wash buffer (20 mM Tris/HCl, 10 mM imidazole, 200 mM NaCl, 10 mM DTT, 2 mM PMSF, 0.5 Roche Protease Inhibitor cocktail, pH 8) containing 20 mM nicotinamide by addition of lysozyme. The sample was sonicated for 2 min (Power output level 5, duty cycle 50%) and centrifuged (20 min, 50,000 g, 4 C.). The supernatant was supplemented with 500 l Ni-NTA-beads. After two hours incubation with agitation at 4 C. beads were washed with 30 ml Ni-wash buffer and bound proteins eluted in Ni-wash buffer supplemented with 200 mM imidazole. The eluate was used without modification as deacetylase substrate.

[0220] Luciferase-Based KDAC Assay

[0221] Typical endpoint deacetylation reactions contain: 30 nM Firefly Luciferase K529ac, 1 mM NAD.sup.+, 1 g/ml KDAC in 50 l KDAC buffer (25 mM Tris/HCl pH 8.0, 137 mM NaCl, 2.7 mM KCl, 1 mM MgCl.sub.2, 1 mM DTT, 1 mg/ml BSA). The reactions are incubated for 1 h at 25 C. Luciferase activity is then assayed by addition of an equal volume of a mixture containing 40 mM Tricine, 200 M EDTA, 7.4 mM MgSO.sub.4, 2 mM NaHCO.sub.3, 34 mM DTT, 0.5 mM ATP and 0.5 mM luciferin, pH 7.8.sup.4. Luminescence is quantified using a FluoStar Omega Microplate Reader (BMG Labtech).

[0222] The continuous FLuc-based KDAC assay was set up by mixing all the components of the endpoint assay immediately. Usually NAD.sup.+ was omitted initially and added from a 20-fold stock solution after 5 min preincubation to start the reaction. Luminescence was recorded every minute over a period of 30 min. KDAC activity was calculated from the slope of the linear phase of the reaction.

[0223] Fluor-De-Lys KDAC Assay

[0224] Typical deacetylation reactions were identical to Luciferase-based assays but containing 10 g/ml KDAC and 10 M Fluor-de-Lys peptide (Ac-Gly-Gly-Lys(ac)-AMC). Conditions were derived from Zhou et al., Molecules 22 (2017) 1348). After incubation for 1 h at 25 C. trypsin and 120 mM nicotinamide were added to the reaction and the reactions were further incubated for 15 min at 37 C. Coumarin fluorescence (ex. 355 nm, em. 460 nm) was then measured using a FluoStar Omega Microplate Reader (BMG Labtech).

[0225] Results

[0226] It was tested whether purified FLuc K529ac can be used to quantify KDAC activity by incubating it with various different KDACs (FIG. 7). Prior to treatment with a KDAC the enzyme produced very little bioluminescence. After incubation with various KDACs the luminescence increased up to 130 fold. Hence, FLuc K529ac is a substrate for KDACs and a highly sensitive reporter enzyme for KDAC activity.

[0227] The assay shows a linear response to increasing KDAC concentrations over a range of 2-3 orders of magnitude (FIGS. 4 & 5). Addition of nicotinamide to the deacetylation reaction of SirT2 inhibited the assay with an IC.sub.50 of 30 M (FIG. 6).

Example 8KDAC Inhibitor Screening Method

[0228] It was tested whether the FLuc-based KDAC assay of the invention is suitable for screening KDAC inhibitors. Therefore, a set of 351 compounds was composed with similarity to known sirtuin inhibitors. The effect of the compounds was analyzed at 10 M on SirT2 activity using the FLuc K529ac assay in endpoint format. The initial screen identified eight compounds inhibiting the assay >50% and one activating more than 1.5 fold (FIG. 8). Compound 9 showed direct inhibition of FLuc, while the remaining seven inhibitors displayed IC.sub.50 values against SirT2 of 3-15 M. The compounds with the highest potency are resveratrol (1) and piceatannol (2). Piceatannol had previously been shown to inhibit SirT2. The structurally very similar resveratrol is a known activator of yeast Sir2 and SirT1. The effect on other sirtuins strongly depends on the combination of KDAC and substrate and can be either activating or inhibitory. In sum, a highly reliable KDAC assay is presented with exceptional sensitivity. The assay is convenient and fast and can be performed in a continuous format. By producing the accordingly modified FLuc, it would be straightforward to adapt the assay to measure the removal of crotonyl, butyryl, propionyl or 2-hydroxyisobutyryl groups from lysine residues. Seven compounds were identified, which inhibit FLuc K529ac deacetylation by SirT2 at low micromolar concentrations. Compound 5 is structurally similar to AGK-2, which inhibits SirT2 with an IC.sub.50 of 3.5 M. Compounds 6 and 7 are similar to SRT1720, which was initially reported as a potent activator of SirT1 and later shown to specifically enhance its interaction with fluorophore-labeled peptides. Piceatannol (2) and particularly resveratrol (1) are familiar sirtuin activators. It is therefore surprising that resveratrol and structurally similar compounds are the most active inhibitors of SirT2 identified in the screen. However, resveratrol specifically activates fluorophore-labeled peptide substrates by stabilizing the enzyme-substrate complex. Resveratrol's effect on sirtuin activity is highly dependent on the sirtuin-substrate combination and has indeed been shown to be inhibitory for SirT3 by enforcing an unproductive conformation of the enzyme-substrate complex. The 2-mercapto-quinazoline derivative 8 is structurally similar to thiobarbiturates, which have been reported to inhibit SirT2 at low micromolar concentration. Thus, the herein provided assay provides a reliable tool for screening of chemical compounds for their ability to inhibit KDAC activity.