Abstract

Histone modification is a ubiquitous regulatory mechanism involved in a variety of biological processes, including gene expression, DNA damage repair, cell differentiation, and ontogenesis. Succinylation sites on histones have been identified and may have functional consequences. Here, we demonstrate that human sirtuin 5 (Sirt5) catalyzes the sequence-selective desuccinylation of numerous histone succinyl sites. Structural studies of Sirt5 in complex with four succinyl peptides indicate an essential role for the conserved main chain hydrogen bonds formed by the succinyl lysine (0), +1, and +3 sites for substrate-enzyme recognition. Furthermore, biochemical assays reveal that the proline residue at the +1 site of the histone succinylation substrate is unfavorable for Sirt5 interaction. Our findings illustrate the molecular mechanism underlying the sequence-selective desuccinylase activity of Sirt5 and provide insights for further studies of the biological functions associated with histone succinylation and Sirt5.

Introduction

Sirtuin 5 (Sirt5) belongs to the sirtuin (yeast Sir2-like proteins) protein family, which includes nicotinamide adenine dinucleotide-dependent deacetylases that remove acetyl groups from protein lysine residues [1,2]. Sirtuins play crucial roles in the regulation of numerous cellular processes, including transcription, DNA damage repair, metabolism, and apoptosis. In mammals, there are seven members of the sirtuin family (Sirt1–7) that exhibit distinct characteristics in sequence, structure, subcellular localization, and function [35]. Compared with Sirt1, 2, 3, and 6, Sirt5 exhibits relatively weak deacetylation activity [6] but can more efficiently remove negatively charged modifications, such as malonylation, glutarylation, and especially succinylation, from lysine residues [79]. Therefore, although previously reported as a deacetylase, Sirt5 has been considered a desuccinylase.

The molecular mechanism of Sirt5-mediated mitochondrial lysine desuccinylation and its functions in the regulation of metabolic networks have been well studied [1013]. Previous proteomics and statistics studies on sequence preference of Sirt5 revealed a preference for alanine at +1 site and glycine/alanine at −1 position [10]. In the same year, another group identified 386 succinylation sites on 140 proteins as the potential substrates of Sirt5. A sequence motif analysis of these succinylation sites related to Sirt5 revealed that, although the positive residues were significantly excluded from adjacent positions of succinyl lysine (Ksu), the sequence preferences of Sirt5 towards its target sites were not remarkable [11].

Although Sirt5 localizes mainly in the mitochondria [1416], a slight amount of Sirt5 was detected in the nuclear fraction [10]. Further immunofluorescence and confocal microscopy consistently revealed a nuclear localization of Sirt5 [10]. Additionally, histone lysine succinylation has been newly identified and may implicate functional consequences [17]. Recent studies reveal that Sirt7 catalyzes H3K122 desuccinylation and plays an important role in DNA damage response and cell survival, indicating the critical function of histone succinylation [18,19]. Additionally, the removal of succinylation at H3K9 (H3K9su) peptide is reportedly catalyzed by Sirt5 in vitro [7]. These findings suggest a role of Sirt5 as a histone desuccinylase. However, how Sirt5 regulates histone desuccinylation is poorly understood. Crystal structure of Sirt5 in complex with H3K9su peptide reveals that the interactions between the peptide and protein are mostly mediated by backbone hydrogen bonding. The substrate-binding pocket of Sirt5 is larger than that of Sirt2, which has a robust deacetylation activity, therefore fits the succinyl group better than the acetyl group [7]. These findings provide the molecular basis underlying the substrate preference of Sirt5 for succinylation rather than acetylation [7]. Up to date, ∼13 histone succinylation sites (not including H3K9 site) have been identified by MS/MS analysis [17]; however, the regulation and biological function of most succinylation sites have rarely been studied. Given that Sirt5 can remove succinylation from multiple mitochondrial proteins, we sought to investigate whether Sirt5 functions as a universal desuccinylase against all identified histone succinylated sites or as a sequence-selective enzyme targeting specific substrates.

By in vitro enzymatic assays, we identified human Sirt5 as a ubiquitous desuccinylase that catalyzes the desuccinylation of diverse histone succinyl sites with sequence selectivity. Among the 13 identified sites, H2BK116su is the most favorable Sirt5 substrate, with H4K12su showing the lowest catalytic efficiency and no activity observed in the presence of H4K31su. To investigate the mechanisms associated with substrate recognition and catalysis, we solved the crystal structures of human Sirt5 in complex with H3K122su, H2AK95su, H2BK120su, and H4K91su peptides, revealing that the main chain hydrogen bonds formed between Sirt5 and the succinyl lysine −2, +1, and +3 sites of the histone peptides are highly conserved among these structures. Additionally, enzymatic assays reveal that a proline residue at +1 site disrupts recognition of the histone succinyl substrate by affecting hydrogen bond formation. Our results illustrate the molecular basis underlying the sequence-selective histone desuccinylase activity of Sirt5 and provided new insights into Sirt5 function.

Experimental procedures

Protein expression and purification

Human Sirt5 (Uniprot entry ID: Q9NXA8) fragment 34–302 was cloned into vector pET28a-N8His and transformed into Rosetta2(DE3) for protein expression. Cells were harvested by centrifuging at 6760×g for 8 min, and lysed with high-pressure in lysis buffer (50 mM Tris–HCl, pH 7.5, 500 mM NaCl, 5% (v/v) glycerol, and 2 mM β-ME) supplemented with 4 mM imidazole and 1 mM PMSF. Lysate was centrifuged at 42 300×g for 30 min at 4°C, and the supernatant was incubated with Ni-NTA resin at 4°C for 30 min. After several rounds of washing, the immobilized proteins were eluted by lysis buffer with 300 mM imidazole, and further purified by gel filtration on Superdex 75 (10/300) prep grade column (GE Healthcare) using SEC buffer (20 mM Tris–HCl, pH 7.5, 200 mM NaCl, and 1 mM DTT).

Crystallization, diffraction data collection, processing and structure determination

Sirt5 and succinyl peptides (Scilight Biotechnology) were mixed with a molar ratio of 1:10, and incubated on ice for 2 h. The crystals of Sirt5-H4K122su and Sirt5-H2AK95su complex were obtained in 0.1 M sodium cacodylate (pH 5.5), 25% w/v PEG 4000 at 22°C. Sirt5-H4K91su complex was crystallized in 0.1 M BIS-TRIS (pH 5.5), 25% w/v PEG 3350, and Sirt5-H2BK120su complex was grown in 0.2 M Sodium acetate trihydrate, 0.1 M Sodium cacodylate trihydrate (pH 6.5), 30% w/v PEG 8000.

The X-ray diffraction data of Sirt5-H3K122su and Sirt5-H4K91su were collected at National Center for Protein Science Shanghai (NCPSS) at beamline BL19U1 and BL18U1, respectively. Diffraction data of Sirt5-H2AK95su and Sirt5-H2BK120su were collected at Shanghai Synchrotron Radiation Facility (SSRF) at beamline BL17U1. All data were processed, integrated, and scaled using HKL2000 [20].

The complex structure of Sirt5-H3K122su was solved by molecular replacement using program phaser MR [21] implemented in CCP4 suite programs [22]. PDB file modified from Sirt5-H3K9su structure (PDB code: 3RIY) was served as a searching model. Structures of Sirt5 in complex with H4K91su, H2AK95su, and H2BK120su peptides were solved using Sirt5-H3K122su as the searching template. Structure refinement, manual model building, and structural analysis were carried out using REFMAC5 [23], Phenix [24], and Wincoot [25]. Data collection and structure refinement statistics are summarized in Table 1. Figures are generated using Pymol (http://www.pymol.org).

Table 1
Crystallographic data collection and structure refinement
 Sirt536-302-H3K122su Sirt536-302-H2AK95su Sirt533-302-H2BK120su Sirt536-302-H4K91su 
PDB code 6ACE 6ACL 6ACO 6ACP 
Data collection 
 Space group CP21 P21 P21 
 Cell dimensions 
  a, b, c (Å) 123.23, 41.17, 55.85 55.94, 42.81, 61.59 33.88, 64.88, 55.56 56.08, 41.27, 61.33 
  α, β, γ (°) 90, 94.69, 90 90.00, 109.25, 90.00 90.00, 95.92, 90.00 90.00, 108.37, 90.00 
  Resolution (Å) 50.0–1.98 (2.01–1.98)1 50.0–1.90 (1.97–1.90) 50.0–1.70 (1.76–1.70) 50.0–2.30 (2.38–2.30) 
  Wavelength (Å) 0.9793 0.9792 0.9792 0.9793 
  Reflections (Unique) 81 688 (18 297) 49 662 (19 522) 164 959 (23 736) 61 946 (12 138) 
  Completeness (%) 92.6 (71.4) 92.6 (92.2) 92.3 (94.6) 99.0 (99.2) 
  Overall I/σ(I11.0 (2.1) 8.9 (3.0) 15.4 (5.0) 8.5 (5.0) 
  Redundancy 4.5 (3.0) 2.5 (2.4) 6.9 (6.9) 5.1 (5.0) 
  Rsym or Rmerge (%)2 10.7 (51.6) 9.8 (30.2) 8.9 (35.5) 14.9 (41.2) 
Refinement 
Rwork3 /Rfree4 (%) 20.61/26.63 16.89/22.25 16.10/21.70 20.32/28.51 
 R.m.s deviations 
  Bond lengths (Å) 0.016 0.018 0.026 0.092 
  Bond angles (°) 1.850 1.916 2.190 1.664 
 No. of ligand/ion molecules 
  Succinyl peptide 
  Zn 
  Glycerol   
  Cl    
 No. of water 121 157 220 
 Average B-factors (Å230.47 29.84 29.94 69.03 
Ramachandran plot5 
 Most favored regions (%) 95.90 95.60 98.42 96.99 
 Allowed regions (%) 3.73 4.40 1.58 3.01 
 Generously allowed regions (%) 0.37 
 Sirt536-302-H3K122su Sirt536-302-H2AK95su Sirt533-302-H2BK120su Sirt536-302-H4K91su 
PDB code 6ACE 6ACL 6ACO 6ACP 
Data collection 
 Space group CP21 P21 P21 
 Cell dimensions 
  a, b, c (Å) 123.23, 41.17, 55.85 55.94, 42.81, 61.59 33.88, 64.88, 55.56 56.08, 41.27, 61.33 
  α, β, γ (°) 90, 94.69, 90 90.00, 109.25, 90.00 90.00, 95.92, 90.00 90.00, 108.37, 90.00 
  Resolution (Å) 50.0–1.98 (2.01–1.98)1 50.0–1.90 (1.97–1.90) 50.0–1.70 (1.76–1.70) 50.0–2.30 (2.38–2.30) 
  Wavelength (Å) 0.9793 0.9792 0.9792 0.9793 
  Reflections (Unique) 81 688 (18 297) 49 662 (19 522) 164 959 (23 736) 61 946 (12 138) 
  Completeness (%) 92.6 (71.4) 92.6 (92.2) 92.3 (94.6) 99.0 (99.2) 
  Overall I/σ(I11.0 (2.1) 8.9 (3.0) 15.4 (5.0) 8.5 (5.0) 
  Redundancy 4.5 (3.0) 2.5 (2.4) 6.9 (6.9) 5.1 (5.0) 
  Rsym or Rmerge (%)2 10.7 (51.6) 9.8 (30.2) 8.9 (35.5) 14.9 (41.2) 
Refinement 
Rwork3 /Rfree4 (%) 20.61/26.63 16.89/22.25 16.10/21.70 20.32/28.51 
 R.m.s deviations 
  Bond lengths (Å) 0.016 0.018 0.026 0.092 
  Bond angles (°) 1.850 1.916 2.190 1.664 
 No. of ligand/ion molecules 
  Succinyl peptide 
  Zn 
  Glycerol   
  Cl    
 No. of water 121 157 220 
 Average B-factors (Å230.47 29.84 29.94 69.03 
Ramachandran plot5 
 Most favored regions (%) 95.90 95.60 98.42 96.99 
 Allowed regions (%) 3.73 4.40 1.58 3.01 
 Generously allowed regions (%) 0.37 
1

The values in parentheses refer to statistics in the highest shell.

2

Rsym = |Ii–|/|Ii| where Ii is the intensity of the ith measurement and is the mean intensity for that reflection.

3

Rwork = |FPFP(calc)|/FP.

4

Rfree was calculated with 5.1% of the reflections in the test set.

5

Statistics for the Ramachandran plot from an analysis using MolProbity.

Desuccinylation activity assays analyzed by matrix-assisted laser desorption/ionization (MALDI)-time-of-flight (ToF) mass spectrometry (MS) (MALDI-ToF MS)

The desuccinylation activity of Sirt5 towards different histone sites was measured by detecting the desuccinylated-form peptides by mass spectrometry. A 100 μl mixture composed of purified human Sirt5 (0.5 μM) with succinyl peptides (0.2 mM) and NAD+ (1.0 mM) in reaction buffer (20 mM Tris–HCl, pH 7.5, 200 mM NaCl, and 1 mM DTT) was incubated at 37°C for 2 h. Mixture without Sirt5 was set as negative control to indicate molecular weight (Mw) of succinyl peptide. Reaction solution was desalted using ZipTip C18 Pipette Tips (Merck KGaA) activated and pre-equilibrated, respectively, by solution1 (70% acetonitrile, 0.1% TFA) and solution2 (0.1% TFA). Peptide on Tips was washed three times using solution2 and finally eluted by 10 μl solution1. The elution was analyzed by MALDI-ToF MS.

The MALDI-ToF MS was performed using MALDI-ToF/ToF instrument of Bruker Autoflex Speed, with the matrix dhb (2,5-Dihydroxybenzoic acid, Sigma, dissolved in solution of acetonitrile:H2O = 1:1, with final concentration of 20 mg/ml) mixed with peptide solution in 2:1 volume-ratio before desorption using solid-state laser. The data were auto-collected and analyzed using software Bruker FlexAnalysis provided by Bruker manufactory, and further presented by OriginPro.

Enzyme-coupled binding kinetics assays

The kinetic parameters of Sirt5 catalysis on succinyl peptides were calculated from continuously spectrometric measurement using an enzyme-coupled assay called PCD-GDH [26], which was slightly modified in our method. Each aliquot of 150 μl assay mixtures contained 20–640 μM succinyl peptide, 1.28 mM NAD+, 0.4 mM NADH, 1 mM dithiothreitol, 6.6 mM α-ketoglutarate, 10 μM nicotinamidase (cloned from Saccharomyces cerevisiae genome, expressed, and purified from BL21(DE3)), 4 units glutamate dehydrogenase (Sigma), and 3.2 μM purified human Sirt5 in PBS buffer. Reaction systems without Sirt5 were used as negative controls and their data were subtracted from experimental groups in the calculation. Reactions were performed in a silica dish and initiated by the addition of Sirt5. The consumption of NADH to reflect nicotinamide generation was measured at 340 nm by UVmini-1240 (SHIMADZU) at 25°C for time span of 10 min with an interval of 20 s. kcat and Km were determined through Michaelis–Menten equation by plotting of steady-state nicotinamide generation rates versus substrate concentrations using non-linear regression fitting in OriginPro. All experiments were reproduced three times in parallel.

Results

Sirt5 functions as a ubiquitous, sequence-selective desuccinylase against histone succinyl peptides

To investigate whether human Sirt5 is a ubiquitous desuccinylase for histone succinylation sites, 13 histone succinyl peptides corresponding to succinylation sites on histones identified by Zhao et al. [17] were synthesized for enzymatic assays (Table 2). Although H3K9su was not verified by MS/MS, since the structure of Sirt5 in complex with H3K9su peptide has been reported, we also synthesized the H3K9su peptide for our experiments. As shown in Figure 1 and Table 2, MALDI-ToF MS detected the desuccinylated peptide peak for 13 substrates (all but the H4K31su peptide) (Figure 1A). Among these, the H3K56su, H3K122su, H4K91su, H2BK120su, H2BK34su, H2AK95su, H2BK116su, and H3K9su peptides (Figure 1B–I) were fully transformed to their desuccinylated form, whereas the H3K14su, H3K79su, H2AK9su, H4K77su, and H4K12su peptides (Figure 1J–N) maintained a peak signifying succinylation. These results indicate that Sirt5 functions as a ubiquitous desuccinylase for histone succinylation, although with sequence selectivity toward certain substrates.

MALDI-ToF MS to determine Sirt5 desuccinylase activity of histone succinyl peptides.

Figure 1.
MALDI-ToF MS to determine Sirt5 desuccinylase activity of histone succinyl peptides.

The control group and experimental group are indicated by ‘−Sirt5’ and ‘+Sirt5’, respectively. The m/z of each peak is labeled in red. H4K31su is labeled in green (A), and other succinyl sites that were fully or partially catalyzed to a desuccinylated form are labeled in black and blue, respectively (B–N). In the partially catalyzed group, the mass shift of two peaks in the MS spectrum is calculated and indicated with a blue arrow (J–N).

Figure 1.
MALDI-ToF MS to determine Sirt5 desuccinylase activity of histone succinyl peptides.

The control group and experimental group are indicated by ‘−Sirt5’ and ‘+Sirt5’, respectively. The m/z of each peak is labeled in red. H4K31su is labeled in green (A), and other succinyl sites that were fully or partially catalyzed to a desuccinylated form are labeled in black and blue, respectively (B–N). In the partially catalyzed group, the mass shift of two peaks in the MS spectrum is calculated and indicated with a blue arrow (J–N).

Table 2.
MALDI-ToF MS for desuccinylase activity of Sirt5 on histone-succinyl peptides and its mutants.

The position of residue relative to succinyl lysine (Ksu) was numbered. The substrate and product were shown in the parentheses after value of m/z.

Succinyl-peptide Sequence -Sirt5 [M+H]+ (m/z) +Sirt5 [M+H]+ (m/z) Mass shift (m/z) 
Fully-catalyzed 
H3K9 (4-15) KQTAR(SuK)STGGKA 1333.6069 1233.4922 100.1147 
H3K56 (51-62) IRRYQ(SuK)STELLI 1620.3058 1520.3485 99.9573 
H3K122 (117-128) VTIMP(SuK)DIQLAR 1485.0742 1384.8529 100.2213 
H4K91 (86-97) VVYAL(SuK)RQGRTL 1504.0668 1404.3424 99.7244 
H2AK95 (90-101) DEELN(SuK)LLGRVT 1486.6098 1386.6019 100.0079 
H2BK34 (29-40) RKRSR(SuK)ESYSVY 1658.6097 1558.9297 99.6800 
H2BK116 (111-122) VSEGT(SuK)AVTKYT 1383.9746 1283.7627 100.2119 
H2BK120 (113-124) GTKAVT(SuK)YTSSK 1370.5453 1271.0077 99.5376 
Partially-catalyzed 
H3K14 (9-20) KSTGG(SuK)APRKQL  1370.6029 1371.5310 (substrate)
1271.4124 (product) 
100.1186 
H3K79 (74-85) IAQDF(SuK)TDLRFQ  1581.5139 1581.4596 (substrate)
1481.4516 (product) 
100.0080 
H4K12 (7-18) GKGLG(SuK)GGAKRH 1265.9383 1265.7266 (substrate)
1165.7050 (product) 
100.0216 
H4K77 (72-83) YTEHA(SuK)RKTVTA  1505.4739 1504.9827 (substrate)
1404.9330 (product) 
100.0497 
H2AK9 (4-15) GKQGG(SuK)ARAKAK 1299.8100 1300.0631 (substrate)
1200.0331 (product) 
 100.0300 
Non-catalyzed 
H4K31 (26-37) IQGIT(SuK)PAIRRL 1465.7636 1465.8764 ---- 
Mutants 
H3K9 (4-15)-S10P KQTAR(SuK)PTGGKA 1343.0187 1343.5764 ---- 
H3K56 (51-62)-S57P IRRYQ(SuK)PTELLI 1631.6368 1632.1898 ---- 
H2AK9 (4-15)-A10G GKQGG(SuK)GRAKAK 1288.7160 1287.0900 (substrate)
1186.9025 (product) 
100.1875 
H3K14 (9-20)-A15G KSTGG(SuK)GPRKQL 1359.9373 1358.3680 (substrate)
1258.8644 (product) 
99.5036 
H4K31 (26-37)-P32A IQGIT(SuK)AAIRRL 1440.5830 1342.5971 97.9859 
 - - - - -         + + + + + +    
 54321    0   123456    
Succinyl-peptide Sequence -Sirt5 [M+H]+ (m/z) +Sirt5 [M+H]+ (m/z) Mass shift (m/z) 
Fully-catalyzed 
H3K9 (4-15) KQTAR(SuK)STGGKA 1333.6069 1233.4922 100.1147 
H3K56 (51-62) IRRYQ(SuK)STELLI 1620.3058 1520.3485 99.9573 
H3K122 (117-128) VTIMP(SuK)DIQLAR 1485.0742 1384.8529 100.2213 
H4K91 (86-97) VVYAL(SuK)RQGRTL 1504.0668 1404.3424 99.7244 
H2AK95 (90-101) DEELN(SuK)LLGRVT 1486.6098 1386.6019 100.0079 
H2BK34 (29-40) RKRSR(SuK)ESYSVY 1658.6097 1558.9297 99.6800 
H2BK116 (111-122) VSEGT(SuK)AVTKYT 1383.9746 1283.7627 100.2119 
H2BK120 (113-124) GTKAVT(SuK)YTSSK 1370.5453 1271.0077 99.5376 
Partially-catalyzed 
H3K14 (9-20) KSTGG(SuK)APRKQL  1370.6029 1371.5310 (substrate)
1271.4124 (product) 
100.1186 
H3K79 (74-85) IAQDF(SuK)TDLRFQ  1581.5139 1581.4596 (substrate)
1481.4516 (product) 
100.0080 
H4K12 (7-18) GKGLG(SuK)GGAKRH 1265.9383 1265.7266 (substrate)
1165.7050 (product) 
100.0216 
H4K77 (72-83) YTEHA(SuK)RKTVTA  1505.4739 1504.9827 (substrate)
1404.9330 (product) 
100.0497 
H2AK9 (4-15) GKQGG(SuK)ARAKAK 1299.8100 1300.0631 (substrate)
1200.0331 (product) 
 100.0300 
Non-catalyzed 
H4K31 (26-37) IQGIT(SuK)PAIRRL 1465.7636 1465.8764 ---- 
Mutants 
H3K9 (4-15)-S10P KQTAR(SuK)PTGGKA 1343.0187 1343.5764 ---- 
H3K56 (51-62)-S57P IRRYQ(SuK)PTELLI 1631.6368 1632.1898 ---- 
H2AK9 (4-15)-A10G GKQGG(SuK)GRAKAK 1288.7160 1287.0900 (substrate)
1186.9025 (product) 
100.1875 
H3K14 (9-20)-A15G KSTGG(SuK)GPRKQL 1359.9373 1358.3680 (substrate)
1258.8644 (product) 
99.5036 
H4K31 (26-37)-P32A IQGIT(SuK)AAIRRL 1440.5830 1342.5971 97.9859 
 - - - - -         + + + + + +    
 54321    0   123456    

Although no obvious consensus motif of Sirt5 substrate has been identified according to previous proteomics and statistics studies [11], a preference for alanine at +1 site and glycine/alanine at −1 position has been proposed [10]. Here we performed enzymatic assays on histone succinyl peptides and observed diverse catalytic efficiency of Sirt5 towards different substrates (Figure 1 and Table 2). Since MALDI-ToF only qualitatively detected the desuccinylase ability of Sirt5, further quantitative analysis is required by determination of the kinetic parameters to compare the catalytic efficiency of Sirt5 on different succinylated histone peptides. To evaluate the sequence-selective desuccinylase activity of Sirt5 involving histone succinylation, we measured Sirt5 kinetics in the presence of these succinyl peptides using an enzyme-coupled assay. Consistent with MALDI-ToF MS results, Sirt5 showed no activity in the presence of H4K31su peptide, but elevated catalytic efficiencies on peptides included in the ‘fully catalyzed group’ relative to those in the ‘partially catalyzed group’ (Figures 1 and 2; Tables 2 and 3). The two most favorable Sirt5 substrates were H2BK116su and H3K9su peptides, with kcat/Km of 4.82 × 103 M−1 s−1 and 4.35 × 103 M−1 s−1, respectively. The substrate with the lowest kcat/Km (1.70 × 103 M−1 s−1) was H4K12su peptide (Table 2). Kinetics results confirmed the sequence-selective activity of Sirt5 involving histone succinylation.

Enzymatic kinetics measurement of Sirt5 towards succinyl peptide.

Figure 2.
Enzymatic kinetics measurement of Sirt5 towards succinyl peptide.

(A–M) Non-linear regression fitting of Michaelis–Menten Equation to data derived from reactions with different substrate concentrations, and the corresponding kcat/Km (M−1s−1) (×103) of each peptide is indicated.

Figure 2.
Enzymatic kinetics measurement of Sirt5 towards succinyl peptide.

(A–M) Non-linear regression fitting of Michaelis–Menten Equation to data derived from reactions with different substrate concentrations, and the corresponding kcat/Km (M−1s−1) (×103) of each peptide is indicated.

The crystal structure of human Sirt5 in complex with the H3K9su peptide revealed that the succinyl group of the substrate fits into the unique acyl-binding pocket of Sirt5 and promotes the formation of three hydrogen bonds with Tyr102 and Arg105 [7]. Additionally, several main chain interactions between peptide and Sirt5 residues further stabilized the complex. The binding mode of Sirt5 with H3K9su provided a structural basis describing Sirt5 function as a ubiquitous desuccinylase against histone succinylation sites.

To address questions concerning the molecular mechanism associated with the sequence-selective desuccinylase activity of human Sirt5 involving histone succinylation, we performed a multiple sequence alignment against all the histone succinyl peptides and compared the features of amino acids proximal to the succinyl lysine residue. However, no neighboring residue was conserved among all 13 peptides (Table 3). Analysis of the H4K31su peptide sequence, which resulted in no Sirt5 activity, revealed that its +1 site was occupied by a proline residue. Additionally, the H4K12su peptide, which resulted in the lowest Sirt5 catalytic efficiency, showed that the +1 and +2 sites were both occupied by a glycine residue. Given the physicochemical properties of proline and glycine, we proposed that the neighboring succinyl lysine might be unfavorably positioned for Sirt5-specific catalysis.

Table 3.
Kinetic parameters of Sirt5 on histone succinyl peptides and its mutants

The position of residue relative to succinyl lysine (SuK) was numbered. kcat/Km representing catalytic efficiency of Sirt5 were ranked in a descending order, and data not detected was indicated by ND.

Succinyl peptide Sequence Km (μM) Vmax (μM/min) kcat (s-1kcat /Km (M-1s-1) (×103
H2B K116 (111-122) VSEGT(SuK)AVTKYT 61.73±3.03 57.14±0.95 0.297±0.004 4.82 
H3 K9 (4-15) KQTAR(SuK)STGGKA 77.17±2.36 64.54±0.66 0.336±0.003 4.35 
H2B K120 (114-125) GTKAVT(SuK)YTSSK 87.77±5.13 53.66±0.87 0.279±0.004 3.18 
H2B K34 (29-40) RKRSR(SuK)ESYSVY 99.73±2.13 60.32±0.44 0.314±0.002 3.15 
H3 K122 (117-128) VTIMP(SuK)DIQLAR 60.51±2.94 34.25±0.47 0.178±0.002 2.94 
H4 K91 (86-97) VVYAL(SuK)RQGRTL 101.78±4.21 56.47±0.78 0.294±0.004 2.89 
H3 K56 (51-62) IRRYQ(SuK)STELLI 179.32±8.12 98.36±1.76 0.512±0.009 2.85 
H2A K95 (90-101) DEELN(SuK)LLGRVT 80.18±3.41 42.20±0.50 0.219±0.002 2.74 
H4 K77 (72-83) YTEHA(SuK)RKTVTA 82.68±2.83 40.07±0.36 0.208±0.001 2.52 
H3 K79 (74-85) IAQDF(SuK)TDLRFQ 84.89±2.6 40.97±0.38 0.213±0.002 2.51 
H2A K9 (4-15) GKQGG(SuK)ARAKAK 95.51±7.37 38.52±0.84 0.200±0.004 2.10 
H3 K14 (9-20) KSTGG(SuK)APRKQL 100.72±12.2 38.78±1.59 0.202±0.008 2.00 
H4 K12 (7-18) GKGLG(SuK)GGAKRH 131.53±6.25 43.06±0.67 0.224±0.003 1.70 
H4 K31 (26-37) IQGIT(SuK)PAIRRL ND ND ND ND 
Mutants 
H3 K9 (4-15)-S10P KQTAR(SuK)PTGGKA ND ND ND ND 
H3 K56 (51-62)-S57P IRRYQ(SuK)PTELLI ND ND ND ND 
H2A K9 (4-15)-A10G GKQGG(SuK)GRAKAK 102.28±1.89 38.26±0.22 0.199±0.001 1.94 
H3 K14 (9-20)-A15G KSTGG(SuK)GPRKQL 104.42±5.34 35.83±0.63 0.186±0.003 1.78 
H4 K31 (26-37)-P32A IQGIT(SuK)AAIRRL 65.83±1.04 36.45±0.13 0.189±0.000 2.88 
 - - - - - + + + + + +     
 54321    0   123456     
Succinyl peptide Sequence Km (μM) Vmax (μM/min) kcat (s-1kcat /Km (M-1s-1) (×103
H2B K116 (111-122) VSEGT(SuK)AVTKYT 61.73±3.03 57.14±0.95 0.297±0.004 4.82 
H3 K9 (4-15) KQTAR(SuK)STGGKA 77.17±2.36 64.54±0.66 0.336±0.003 4.35 
H2B K120 (114-125) GTKAVT(SuK)YTSSK 87.77±5.13 53.66±0.87 0.279±0.004 3.18 
H2B K34 (29-40) RKRSR(SuK)ESYSVY 99.73±2.13 60.32±0.44 0.314±0.002 3.15 
H3 K122 (117-128) VTIMP(SuK)DIQLAR 60.51±2.94 34.25±0.47 0.178±0.002 2.94 
H4 K91 (86-97) VVYAL(SuK)RQGRTL 101.78±4.21 56.47±0.78 0.294±0.004 2.89 
H3 K56 (51-62) IRRYQ(SuK)STELLI 179.32±8.12 98.36±1.76 0.512±0.009 2.85 
H2A K95 (90-101) DEELN(SuK)LLGRVT 80.18±3.41 42.20±0.50 0.219±0.002 2.74 
H4 K77 (72-83) YTEHA(SuK)RKTVTA 82.68±2.83 40.07±0.36 0.208±0.001 2.52 
H3 K79 (74-85) IAQDF(SuK)TDLRFQ 84.89±2.6 40.97±0.38 0.213±0.002 2.51 
H2A K9 (4-15) GKQGG(SuK)ARAKAK 95.51±7.37 38.52±0.84 0.200±0.004 2.10 
H3 K14 (9-20) KSTGG(SuK)APRKQL 100.72±12.2 38.78±1.59 0.202±0.008 2.00 
H4 K12 (7-18) GKGLG(SuK)GGAKRH 131.53±6.25 43.06±0.67 0.224±0.003 1.70 
H4 K31 (26-37) IQGIT(SuK)PAIRRL ND ND ND ND 
Mutants 
H3 K9 (4-15)-S10P KQTAR(SuK)PTGGKA ND ND ND ND 
H3 K56 (51-62)-S57P IRRYQ(SuK)PTELLI ND ND ND ND 
H2A K9 (4-15)-A10G GKQGG(SuK)GRAKAK 102.28±1.89 38.26±0.22 0.199±0.001 1.94 
H3 K14 (9-20)-A15G KSTGG(SuK)GPRKQL 104.42±5.34 35.83±0.63 0.186±0.003 1.78 
H4 K31 (26-37)-P32A IQGIT(SuK)AAIRRL 65.83±1.04 36.45±0.13 0.189±0.000 2.88 
 - - - - - + + + + + +     
 54321    0   123456     

Recognition of diverse histone succinyl sites by Sirt5 requires conserved main chain hydrogen bond interactions

Therefore, to investigate the molecular mechanism associated with the sequence-selective activity of Sirt5 on histone substrates, we solved the crystal structures of human Sirt5 in complex with H3K122su (aa 117–128), H4K91su (aa 86–97), H2AK95su (aa 90–101), and H2BK120su (114–125) peptides, to resolutions of 1.98, 2.30, 1.90, and 1.70 Å, respectively (Table 1). The overall structures of Sirt5 complexes were identical and comprised 13 α-helices and 9 β-sheets folded into two globular domains: a zinc-binding domain and a Rossmann-fold domain (Figure 3A). Compared with previously determined Sirt5-H3K9su complex (PDB code: 3RIY), the succinyl groups of H3K122su, H4K91su, H2AK95su, and H2BK120su peptides occupied the acyl-binding pocket of Sirt5 similar to that shown by H3K9su peptide (Figure 3B). The two important substrate-binding residues in this pocket, Tyr102 and Arg105, enabled three fundamental hydrogen bonds for substrate stabilization (Figure 3B).

Structures of Sirt5 in complex with histone succinyl peptides.

Figure 3.
Structures of Sirt5 in complex with histone succinyl peptides.

(A) Comparison of the overall structures of Sirt5 in complex with H3K122su (green), H4K91su (cyan), H2AK95su (wheat), and H2BK120su (yellow) peptides along with the Sirt5-H3K9su complex (gray; PDB ID: 3RIY). Peptides are colored in gray and shown as cartoons, with the succinyl lysine (Ksu) residues presented as sticks. The zinc-binding and Rossmann-fold domains of Sirt5 are labeled. (B) Structure superimposition of the succinyl group binding site in Sirt5-H3K9su (gray) with Sirt5-H3K122su (green), -H4K91su (cyan), -H2AK95su (wheat), and -H2BK120su (yellow) complexes. The conserved hydrogen bonds formed by R105 and Y102 with the carboxyl group of Ksu are shown as black dashes. R105 and Y102 from Sirt5 and Ksu from peptides are shown as sticks. (C) Comparison of main chain hydrogen bonds formed between Sirt5 and histone peptides. Sirt5 residues G224, E225, L227, V253, and Y255 are colored in gray and labeled in black. H3K9su, H2AK95su, H3K122su, H4K91su, and H2BK120su peptides are shown as sticks and colored in orange, wheat, green, cyan, and yellow, respectively. Main chain hydrogen bonds are shown as black dashes. Peptide residues at −2, 0, +1, and +3 sites and that provide conserved main chain hydrogen bonds are labeled with numbers.

Figure 3.
Structures of Sirt5 in complex with histone succinyl peptides.

(A) Comparison of the overall structures of Sirt5 in complex with H3K122su (green), H4K91su (cyan), H2AK95su (wheat), and H2BK120su (yellow) peptides along with the Sirt5-H3K9su complex (gray; PDB ID: 3RIY). Peptides are colored in gray and shown as cartoons, with the succinyl lysine (Ksu) residues presented as sticks. The zinc-binding and Rossmann-fold domains of Sirt5 are labeled. (B) Structure superimposition of the succinyl group binding site in Sirt5-H3K9su (gray) with Sirt5-H3K122su (green), -H4K91su (cyan), -H2AK95su (wheat), and -H2BK120su (yellow) complexes. The conserved hydrogen bonds formed by R105 and Y102 with the carboxyl group of Ksu are shown as black dashes. R105 and Y102 from Sirt5 and Ksu from peptides are shown as sticks. (C) Comparison of main chain hydrogen bonds formed between Sirt5 and histone peptides. Sirt5 residues G224, E225, L227, V253, and Y255 are colored in gray and labeled in black. H3K9su, H2AK95su, H3K122su, H4K91su, and H2BK120su peptides are shown as sticks and colored in orange, wheat, green, cyan, and yellow, respectively. Main chain hydrogen bonds are shown as black dashes. Peptide residues at −2, 0, +1, and +3 sites and that provide conserved main chain hydrogen bonds are labeled with numbers.

In addition to the interactions between the substrate succinyl group and Sirt5, the main chain of the modified lysine and nearby residues is also involved in Sirt5 interaction. In all of the complex structures, 0 (Ksu), +1, and +3 sites from the histone peptide provide two, two, and one main chain hydrogen bond interaction(s) with Sirt5 Gly224 through Glu225, Tyr255, and Val253, respectively (Figure 3C). Additionally, the hydrogen bond contributed by the −2 site and Sirt5 Leu227 was also highly conserved in the H3K9su, H4K91su, H2AK95su, and H2BK120su peptide-bound structures. Notably, the proline at −2 site of H3K122su peptide displayed a different conformation from that of the other complexes and, therefore, lacked hydrogen bond interaction with Sirt5 Leu227. Structural analysis indicated that recognition of histone succinylation sites by Sirt5 requires the highly conserved main chain hydrogen bonds contributed by the succinyl lysine of the substrate and its proximal residues at the +1, +3, and −2 sites.

Proline at the +1 site disrupts Sirt5-mediated desuccinylation of the histone succinyl substrate

Our structural studies revealed that the recognition of histone succinyl peptides by Sirt5 was mainly mediated by the succinyl group and the conserved main chain hydrogen bonds formed by the modified lysine (0), −2, +1, and +3 residues. When the +1 site was occupied by a proline in the H4K31su peptide, no Sirt5 catalytic activity was detected (Figures 1 and 2; Tables 2 and 3). Due to the distinctive cyclic structure of the side chain, proline is more conformationally rigid relative to other amino acids [27]; therefore, it frequently acts as a structural disruptor that affects overall secondary structure or nearby peptides [28]. Moreover, the α-N of the proline residue can only serve as a hydrogen bond acceptor, but not a donor. Therefore, in the case of the H4K31su peptide, proline at +1 site likely altered the conformation of nearby residues within the peptide to disrupt the conserved hydrogen bonds formed by +1 and +3 sites, thereby disrupting interaction with Sirt5.

To verify this hypothesis, residues Ser10 and Ser57 at +1 sites of H3K9su and H3K56su peptides were mutated to proline (H3K9su-S10P and H3K56su-S57P, respectively). As expected, these mutations inhibited Sirt5 catalysis (Figure 4A, B; Tables 2 and 3). Additionally, when +1 site of the proline in H4K31su peptide was replaced with an alanine (H4K31su-P32A), Sirt5-catalyzed desuccinylation of this mutant peptide with kcat/Km of 2.88 × 103 M−1s−1 (Figure 4C,D; Tables 2 and 3). These results confirmed that proline at the +1 site disrupted Sirt5 recognition of the histone succinyl substrate.

Sirt5 desuccinylase activity on mutated histone succinyl peptides.

Figure 4.
Sirt5 desuccinylase activity on mutated histone succinyl peptides.

(A–C) MALDI-ToF results of Sirt5 desuccinylation of mutated histone succinyl peptides. Blue arrows indicate mass shifts from succinyl peptide to desuccinylated peptide. (D–H) Enzyme kinetics of Sirt5 in the presence of mutated histone succinyl peptides. Non-linear regression fitting of the Michaelis–Menten equation to data derived from reactions undertaken using different substrate concentrations, and corresponding kcat/Km (M−1s−1) (×103) of each peptide is labeled.

Figure 4.
Sirt5 desuccinylase activity on mutated histone succinyl peptides.

(A–C) MALDI-ToF results of Sirt5 desuccinylation of mutated histone succinyl peptides. Blue arrows indicate mass shifts from succinyl peptide to desuccinylated peptide. (D–H) Enzyme kinetics of Sirt5 in the presence of mutated histone succinyl peptides. Non-linear regression fitting of the Michaelis–Menten equation to data derived from reactions undertaken using different substrate concentrations, and corresponding kcat/Km (M−1s−1) (×103) of each peptide is labeled.

The lowest catalytic efficiency was observed in the presence of the H4K12su peptide harboring a G(Ksu)G motif. Given the high flexibility associated with regions proximal to glycine residues, we proposed that histone substrates harboring a glycine residue at the −1 and +1 sites might be unfavorable for Sirt5 recognition. To test this hypothesis, four mutant peptides H2AK9su-G8A, H2AK9su-A10G, H3K14su-G13A, and H3K14su-A15G were synthesized for evaluation by enzymatic assays. As expected, the catalytic efficiency of Sirt5 on A-to-G and G-to-A mutant peptide is slightly lower and higher than that of the wild-type ones (Figure 4E–H; Table 3), respectively. However, the glycine at −1 and +1 sites of the substrate show minimal effect relative to that of the proline on Sirt5 catalysis.

Our results demonstrated that the proline residue at +1 site of the succinyl peptide disrupted Sirt5 desuccinylation most likely through prohibiting the formation of hydrogen bonds between substrate and Sirt5. Additionally, a glycine residue at the −1 position is not favorable for Sirt5, which is in conflict with the previous proteomic and statistics studies [10].

Discussion

Sirt5 was previously reported as a mitochondrially localized deacetylase, but recently it was more considered as a multi-functional deacylase, with substrate preference for negatively charged modifications, such as malonylation, succinylation, and glutarylation [7,9,10,29,30]. Sirt5 catalyzes deacylation with higher catalytic efficiency towards succinyl/malonyl lysine than acetylation [7]. Desuccinylation catalyzed by Sirt5 was reported to has a tight connection with biological functions involved in metabolism [31], since the reaction components, NAD+ and succinyl-CoA, are a product or intermediate derived from metabolism [32,33]. In mammalian cells, the majority of potential Sirt5 target proteins identified by quantitative MS analysis are associated with a series of cellular metabolic/catabolic processes, including fatty acid β-oxidation, ketogenesis, TCA cycle, urea cycle, ammonia transfer, and glucose metabolism [10,11,30,34,35].

In addition to a mitochondrial/cytoplasm desuccinylase, in vitro studies on the desuccinylation of histone H3K9su substrate catalyzed by Sirt5 [7] indicates that Sirt5 might also function as a histone desuccinylase in nuclear. Histone succinylation is associated with lots of functional events [36]. For instance, the succinylation on N-terminal domains of core histones would increase the efficiency of the nucleosomal cores as transcription templates [37], and succinylation on histone H3 can serve as epigenetic regulators in chromatin structure and function [38,39]. Although studies on histone or nucleoprotein desuccinylation by Sirt5 were few, succinyl group removal on H3K122su by Sirt5 homolog Sirt7, which localizes in nucleolus, was reported [18]. In our in vitro studies, we found that Sirt5 is able to remove the succinyl group from all available histone succinylation sites except H4K31su. Further structural studies combined enzyme kinetics provided insights into the molecular mechanism underling the ubiquitous substrate-binding ability and sequence preference of Sirt5. Since histone succinylation was usually associated with loose compaction status of chromatin and active gene transcription by affecting DNA–histone interactions [18,19,40], the desuccinylation capability of Sirt5 towards a wide range of histone lysine sites may probably participate in regulating gene expression through chromatin remodeling or in other alternative ways. Further in vivo studies are required to demonstrate the biological significance of Sirt5-catalyzed histone desuccinylation in the future.

Database Depositions

The co-ordinates for Sirt5-H3K122su, Sirt5-H2AK95su, Sirt5-H2BK120su, Sirt5-H4K91su have been deposited in the Protein Data Bank under accession codes 6ACE, 6ACL, 6ACO, and 6ACP, respectively.

Abbreviations

     
  • BIS-TRIS

    2-[Bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol

  •  
  • DTT

    Dithiothreitol ((2S,3S)-1,4-Bis(sulfanyl)butane-2,3-diol)

  •  
  • NAD+ and NADH

    Nicotinamide adenine dinucleotide

  •  
  • Ni-NTA

    Nickel-Nitrilotriacetic acid (2,2′,2″-Nitrilotriacetic acid)

  •  
  • PEG

    polyethylene glycol (poly(oxyethylene))

  •  
  • PMSF

    phenylmethanesulfonyl fluoride

  •  
  • TFA

    Trifluoroacetic acid

  •  
  • Tris

    tris(hydroxymethyl)aminomethane (2-Amino-2-(hydroxymethyl)propane-1,3-diol)

  •  
  • β-ME

    β-mercaptoethanol (2-Sulfanylethan-1-ol)

Author Contribution

J.Z. and X.Z. provided the scientific direction and the overall experimental design for the studies. T.H., and W.C. designed and performed the biochemical experiments and crystal structure studies. M.W. and L.Z. were responsible for the MS studies. N.J. constructed the Sirt5 plasmid. C.W. helped with the peptide design. T.H., X.Z., and J.Z. wrote the manuscript.

Funding

This work was supported by the National Key Research and Development Program of China [2017YFA0503600, 2016YFA0400903], the Strategic Priority Research Program of the Chinese Academy of Sciences [XDB19000000], the Foundation for Innovative Research Groups of the National Natural Science Foundation of China [31621002], the Major Research Plan of the National Natural Science Foundation of China [91853133], the Innovative Program of Development Foundation of Hefei Center for Physical Science and Technology [2017FXCX004], the National Natural Science Foundation of China [U1532109 to J.Z., 31700671 to X.Z.].

Acknowledgements

We thank the staff at beamline station BL17U1 of SSRF (Shanghai Synchrotron Radiation Facility) and BL18U1 and BL19U1 of NCPSS (National Center for Protein Science Shanghai) for assistance with data collection.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Imai
,
S.
,
Armstrong
,
C.M.
,
Kaeberlein
,
M.
and
Guarente
,
L.
(
2000
)
Transcriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylase
.
Nature
403
,
795
800
2
Frye
,
R.A.
(
1999
)
Characterization of five human cDNAs with homology to the yeast SIR2 gene: Sir2-like proteins (sirtuins) metabolize NAD and may have protein ADP-ribosyltransferase activity
.
Biochem. Biophys. Res. Commun.
260
,
273
279
3
Frye
,
R.A.
(
2000
)
Phylogenetic classification of prokaryotic and eukaryotic Sir2-like proteins
.
Biochem. Biophys. Res. Commun.
273
,
793
798
4
Haigis
,
M.C.
and
Guarente
,
L.P.
(
2006
)
Mammalian sirtuins — emerging roles in physiology, aging, and calorie restriction
.
Gene Dev.
20
,
2913
2921
5
Michan
,
S.
and
Sinclair
,
D.
(
2007
)
Sirtuins in mammals: insights into their biological function
.
Biochem. J.
404
,
1
13
6
Rauh
,
D.
,
Fischer
,
F.
,
Gertz
,
M.
,
Lakshminarasimhan
,
M.
,
Bergbrede
,
T.
,
Aladini
,
F.
et al.  (
2013
)
An acetylome peptide microarray reveals specificities and deacetylation substrates for all human sirtuin isoforms
.
Nat. Commun.
4
,
2327
7
Du
,
J.T.
,
Zhou
,
Y.Y.
,
Su
,
X.Y.
,
Yu
,
J.J.
,
Khan
,
S.
,
Jiang
,
H.
et al.  (
2011
)
Sirt5 is a NAD-dependent protein lysine demalonylase and desuccinylase
.
Science
334
,
806
809
8
Zhang
,
Z.H.
,
Tan
,
M.J.
,
Xie
,
Z.Y.
,
Dai
,
L.Z.
,
Chen
,
Y.
and
Zhao
,
Y.M.
(
2011
)
Identification of lysine succinylation as a new post-translational modification
.
Nat. Chem. Biol.
7
,
58
63
9
Tan
,
M.J.
,
Peng
,
C.
,
Anderson
,
K.A.
,
Chhoy
,
P.
,
Xie
,
Z.Y.
,
Dai
,
L.Z.
et al.  (
2014
)
Lysine glutarylation is a protein posttranslational modification regulated by SIRT5
.
Cell Metab.
19
,
605
617
10
Park
,
J.
,
Chen
,
Y.
,
Tishkoff
,
D.X.
,
Peng
,
C.
,
Tan
,
M.
,
Dai
,
L.
et al.  (
2013
)
SIRT5-mediated lysine desuccinylation impacts diverse metabolic pathways
.
Mol. Cell
50
,
919
930
11
Rardin
,
M.J.
,
He
,
W.J.
,
Nishida
,
Y.
,
Newman
,
J.C.
,
Carrico
,
C.
,
Danielson
,
S.R.
et al.  (
2013
)
SIRT5 regulates the mitochondrial lysine succinylome and metabolic networks
.
Cell Metab.
18
,
920
933
12
Hershberger
,
K.A.
,
Abraham
,
D.M.
,
Martin
,
A.S.
,
Mao
,
L.
,
Liu
,
J.
,
Gu
,
H.B.
et al.  (
2017
)
Sirtuin 5 is required for mouse survival in response to cardiac pressure overload
.
J. Biol. Chem.
292
,
19767
19781
13
Zhang
,
Y.X.
,
Bharathi
,
S.S.
,
Rardin
,
M.J.
,
Lu
,
J.
,
Maringer
,
K.V.
,
Sims-Lucas
,
S.
et al.  (
2017
)
Lysine desuccinylase SIRT5 binds to cardiolipin and regulates the electron transport chain
.
J. Biol. Chem.
292
,
10239
10249
14
Pirinen
,
E.
,
Lo Sasso
,
G.
and
Auwerx
,
J.
(
2012
)
Mitochondrial sirtuins and metabolic homeostasis
.
Best Pract. Res. Clin. Enocrinol. Metab.
26
,
759
770
15
Huang
,
J.Y.
,
Hirschey
,
M.D.
,
Shimazu
,
T.
,
Ho
,
L.
and
Verdin
,
E.
(
2010
)
Mitochondrial sirtuins
.
Biochim. Biophy. Acta Proteins Proteom.
1804
,
1645
1651
16
Zhong
,
L.
and
Mostoslavsky
,
R.
(
2011
)
Fine tuning our cellular factories: sirtuins in mitochondrial biology
.
Cell Metab.
13
,
621
626
17
Xie
,
Z.Y.
,
Dai
,
J.B.A.
,
Dai
,
L.Z.
,
Tan
,
M.J.
,
Cheng
,
Z.Y.
,
Wu
,
Y.M.
et al.  (
2012
)
Lysine succinylation and lysine malonylation in histones
.
Mol. Cell. Proteomics
11
,
100
107
18
Li
,
L.
,
Shi
,
L.
,
Yang
,
S.D.
,
Yan
,
R.R.
,
Zhang
,
D.
,
Yang
,
J.G.
et al.  (
2016
)
SIRT7 is a histone desuccinylase that functionally links to chromatin compaction and genome stability
.
Nat. Commun.
7
,
12235
19
Smestad
,
J.
,
Erber
,
L.
,
Chen
,
Y.
and
Maher
, III,
L.J.
(
2018
)
Chromatin succinylation correlates with active gene expression and is perturbed by defective TCA cycle metabolism
.
iScience
2
,
63
75
20
Otwinowski
,
Z.
and
Minor
,
W.
(
1997
)
Processing of X-ray diffraction data collected in oscillation mode
.
Method Enzymol.
276
,
307
326
21
Mccoy
,
A.J.
,
Grosse-Kunstleve
,
R.W.
,
Adams
,
P.D.
,
Winn
,
M.D.
,
Storoni
,
L.C.
and
Read
,
R.J.
(
2007
)
Phaser crystallographic software
.
J. Appl. Crystallogr.
40
,
658
674
22
Bailey
,
S.
(
1994
)
The Ccp4 suite - programs for protein crystallography
.
Acta. Crystallogr. D Biol. Crystallogr.
50
,
760
763
23
Vagin
,
A.A.
,
Steiner
,
R.A.
,
Lebedev
,
A.A.
,
Potterton
,
L.
,
McNicholas
,
S.
,
Long
,
F.
et al.  (
2004
)
REFMAC5 dictionary: organization of prior chemical knowledge and guidelines for its use
.
Acta Crystallogr. D Biol. Crystallogr.
60
,
2184
2195
24
Adams
,
P.D.
,
Afonine
,
P.V.
,
Bunkóczi
,
G.
,
Chen
,
V.B.
,
Davis
,
I.W.
,
Echols
,
N.
et al.  (
2010
)
PHENIX: a comprehensive Python-based system for macromolecular structure solution
.
Acta Crystallogr. D Biol. Crystallogr.
66
,
213
221
25
Emsley
,
P.
,
Lohkamp
,
B.
,
Scott
,
W.G.
and
Cowtan
,
K.
(
2010
)
Features and development of Coot
.
Acta Crystallogr. D Biol. Crystallogr.
66
,
486
501
26
Smith
,
B.C.
,
Hallows
,
W.C.
and
Denu
,
J.M.
(
2009
)
A continuous microplate assay for sirtuins and nicotinamide-producing enzymes
.
Anal. Biochem.
394
,
101
109
27
Hovmöller
,
S.
,
Zhou
,
T.
and
Ohlson
,
T.
(
2002
)
Conformations of amino acids in proteins
.
Acta Crystallogr. D Biol. Crystallogr.
58
,
768
776
28
MacArthur
,
M.W.
and
Thornton
,
J.M.
(
1991
)
Influence of proline residues on protein conformation
.
J. Mol. Biol.
218
,
397
412
29
Peng
,
C.
,
Lu
,
Z.
,
Xie
,
Z.
,
Cheng
,
Z.
,
Chen
,
Y.
,
Tan
,
M.
et al.  (
2011
)
The first identification of lysine malonylation substrates and its regulatory enzyme
.
Mol. Cell. Proteomics
10
,
M111 012658
30
Hirschey
,
M.D.
and
Zhao
,
Y.
(
2015
)
Metabolic regulation by lysine malonylation, succinylation, and glutarylation
.
Mol. Cell. Proteomics
14
,
2308
2315
31
Bringman-Rodenbarger
,
L.R.
,
Guo
,
A.H.
,
Lyssiotis
,
C.A.
and
Lombard
,
D.B.
(
2018
)
Emerging roles for SIRT5 in metabolism and cancer
.
Antioxid. Redox Signal.
28
,
677
690
32
Belenky
,
P.
,
Bogan
,
K.L.
and
Brenner
,
C.
(
2007
)
NAD+ metabolism in health and disease
.
Trends Biochem. Sci.
32
,
12
19
33
Halarnkar
,
P.P.
and
Blomquist
,
G.J.
(
1989
)
Comparative aspects of propionate metabolism
.
Comp. Biochem. Physiol. B Comp. Biochem.
92
,
227
231
34
Weinert
,
B.T.
,
Schölz
,
C.
,
Wagner
,
S.A.
,
Iesmantavicius
,
V.
,
Su
,
D.
,
Daniel
,
J.A.
et al.  (
2013
)
Lysine succinylation is a frequently occurring modification in prokaryotes and eukaryotes and extensively overlaps with acetylation
.
Cell Rep.
4
,
842
851
35
Chen
,
Y.
(
2016
)
Quantitative analysis of the Sirt5-regulated lysine succinylation proteome in mammalian cells
.
Methods Mol. Biol.
1410
,
23
37
36
Sabari
,
B.R.
,
Zhang
,
D.
,
Allis
,
C.D.
and
Zhao
,
Y.M.
(
2017
)
Metabolic regulation of gene expression through histone acylations
.
Nat. Rev. Mol. Cell Biol.
18
,
90
101
37
Piñeiro
,
M.
,
Hernández
,
F.
and
Palacián
,
E.
(
1992
)
Succinylation of histone amino groups facilitates transcription of nucleosomal cores
.
Biochim. Biophys. Acta Gene Struct. Expr.
1129
,
183
187
38
Xu
,
Y.M.
,
Du
,
J.Y.
and
Lau
,
A.T.Y.
(
2014
)
Posttranslational modifications of human histone H3: An update
.
Proteomics
14
,
2047
2060
39
Olsen
,
C.A.
(
2012
)
Expansion of the lysine acylation landscape
.
Angew. Chem. Int. Ed.
51
,
3755
3756
40
Jing
,
Y.
,
Liu
,
Z.
,
Tian
,
G.
,
Bao
,
X.
,
Ishibashi
,
T.
and
Li
,
X.D.
(
2018
)
Site-Specific installation of succinyl lysine analog into histones reveals the effect of H2BK34 succinylation on nucleosome dynamics
.
Cell Chem. Biol.
25
,
166
174.e167

Author notes

*

These authors contributed equally to this work.

Supplementary data