Chemically Induced Degradation of Sirtuin 2 (Sirt2) by a Proteolysis Targeting Chimera (PROTAC) Based on Sirtuin Rearranging Ligands (SirReals)

Article abstract of DOI:10.1021/acs.jmedchem.6b01872

Here we report the development of a proteolysis targeting chimera (PROTAC) based on the combination of the unique features of the sirtuin rearranging ligands (SirReals) as highly potent and isotype-selective Sirt2 inhibitors with thalidomide, a bona fide

Full text of DOI:10.1021/acs.jmedchem.6b01872

Subscriber access provided by BALL STATE UNIV
Article
Chemically induced degradation of sirtuin 2 (Sirt2) by a proteolysis
targeting chimera (PROTAC) based on sirtuin rearranging ligands (SirReals)
Matthias Schiedel, Daniel Herp, Sören Hammelmann, Sören Swyter, Attila
Lehotzky, Dina Robaa, Judit Olah, Judit Ovádi, Wolfgang Sippl, and Manfred Jung
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01872 • Publication Date (Web): 05 Apr 2017
Downloaded from http://pubs.acs.org on April 6, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chemically induced degradation of sirtuin 2 (Sirt2)
by a proteolysis targeting chimera (PROTAC) based
on sirtuin rearranging ligands (SirReals)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
†
†
†
§
∇
Matthias Schiedel, Daniel Herp, Sören Hammelmann, Sören Swyter, Attila Lehotzky, Dina
◊
§
§
◊
†
∥
Robaa , Judit Oláh, Judit Ovádi, Wolfgang Sippl, Manfred Jung *
† Institute of Pharmaceutical Sciences, University of Freiburg, Albertstraße 25, 79104 Freiburg
im Breisgau, Germany
§ Institute of Enzymology, Research Centre for Natural Sciences, Hungarian Academy of
Sciences, Magyar Tudósok körútja 2, H 1117 Budapest, Hungary
◊ Institute of Pharmacy, MartinꢀLutherꢀUniversity HalleꢀWittenberg, WolfgangꢀLangenbeckꢀ
Straße 4, 06120 Halle (Saale), Germany
∥ Freiburg Institute of Advanced Studies (FRIAS), University of Freiburg, Albertstraße 19,
79104 Freiburg im Breisgau, Germany
KEYWORDS: epigenetics, histone deacetylases, sirtuins, protein degradation, PROTAC
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 36
1
2
3
4
5
6
7
8
9
ABSTRACT: Here we report the development of a proteolysis targeting chimera (PROTAC)
based on the combination of the unique features of the sirtuin rearranging ligands (SirReals) as
highly potent and isotypeꢀselective Sirt2 inhibitors with thalidomide, a bona fide cereblon ligand.
For the first time, we report the formation of a PROTAC by Cu(I)ꢀcatalyzed cycloaddition of a
thalidomideꢀderived azide to an alkynylated inhibitor. This thalidomideꢀderived azide as well as
the highly versatile linking strategy can be readily adapted to alkynylated ligands of other
targets. In HeLa cells, our SirRealꢀbased PROTAC induced isotypeꢀselective Sirt2 degradation
that results in the hyperacetylation of the microtubule network coupled with enhanced process
elongation. Thus, our SirRealꢀbased PROTAC is the first example of a probe that is able to
chemically induce the degradation of an epigenetic eraser protein.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
Until today, 18 different lysine deacetylases (KDACs) have been identified in the human
genome and grouped into four classes, according to their sequence homology to yeast KDACs.1
These enzymes are able to remove acyl groups from the εꢀamino group of acylated lysine
residues. Sirtuins, which had been initially described as class III histone deacetylases (HDACs)
or Sir2 proteins,2 form a very special class of the KDAC family. Sirtuins use NAD+ as a cofactor
and constitute the class III KDACs, while the deacetylases of classes I, II, and IV are prime
examples of Zn2+ꢀdependent metalloproteases. The human genome encodes seven isotypes of
sirtuins (Sirt1ꢀ7), which differ in their catalytic activity as well as in their subcellular
localization.3 Moreover, sirtuins show no sequence homology to the Zn2+ꢀdependent KDACs.
Owing to their unique NAD+ꢀdependent catalytic mechanism, sirtuins are not restricted to
ACS Paragon Plus Environment
2

Page 3 of 36
Journal of Medicinal Chemistry
1
2
deacetylation. Sirtuins are also able to cut off other acyl groups, such as crotonyl,4 glutaryl,5
succinyl,6 myristoyl,7 and palmitoyl8 from the εꢀamino group of acylated lysines. The
deacetylase activity of Sirt6 and Sirt7 was recently shown to be upregulated by longꢀchain fatty
acids and nucleic acids, respectively.7, 9 Besides histones, a variety of nonꢀhistone proteins have
been reported as sirtuin substrates in recent years, e.g. αꢀtubulin,10 p53,11 BubR1,12 and NFκB.13
By regulating the acylation state of their protein substrates, sirtuins have been implicated in
influencing a multitude of cellular processes such as transcription,14 metabolic sensing,15
inflammation,16 aging,17 and apoptosis.15 The isotype Sirt2, which is predominantly localized in
the cytoplasm, was shown to have a major impact on cell cycle regulation,10 autophagy,18
peripheral myelination,19 and immune and inflammatory response.20 Furthermore, a number of
reports implicate that the overall cellular agenda of Sirt2 is not only dependent on its catalytical
activity but also on its proteinꢀprotein interactions with binding partners, e.g. KDAC6,10
HOXA10,21 and 14ꢀ3ꢀ3 β/γ.22 A dysregulation of Sirt2 has been associated with the pathogenesis
of bacterial infections,20b, 20c type II diabetes,23 neurodegenerative diseases,24 and cancer,25 which
makes Sirt2 a promising target for pharmaceutical intervention. However, for some disease
scenarios, e.g. Huntington’s disease, it is not perfectly clear if Sirt2 has to be upꢀ or
downregulated respectively inhibited to ameliorate specific disease conditions.26 Many cellꢀ
based or animal studies have tried to address these unanswered questions. Yet, they have been
hampered by the lack of suitable tool compounds, which show sufficient isotype selectivity as
well as pharmacokinetic properties. The urgent need for such tool compounds, to further
investigate the cellular effects of Sirt2ꢀdependent deacylation as well as their consequences on
downstream signaling, led to the discovery of a number of Sirt2ꢀselective small molecule
inhibitors.27 However, Sirt2ꢀselective and drugꢀlike inhibitors with an identified binding mode,
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 36
1
2
3
4
5
6
that enables the structureꢀbased discovery of molecular tools for Sirt2, are still extremely scarce.
Recently, we discovered a new class of Sirt2ꢀselective and highly potent inhibitors (Fig.1).28
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Chemical structures and inhibition data of selected SirReals (1a/b-3a/b) and the
SirRealꢀbased affinity probe for Sirt2 (4).
The coꢀcrystal structures of Sirt2 in complex with 1a or 2a, which were the first crystal
structures of Sirt2 complexed with Sirt2ꢀselective smallꢀmolecule inhibitors, revealed a unique
mode of inhibition that is characterized by a major rearrangement of Sirt2’s active site upon
ligand binding. Therefore, these inhibitors were referred to as sirtuin rearranging ligands
(SirReals).28 By structureꢀguided optimization and extensive structureꢀactivity relationship
(SAR) studies we were able to discover the SirReal analogues 2b and 3a/b with enhanced
potency and cellular efficacy.29 The triazoleꢀbased SirReals 3a/b, which additionally show
improved water solubility, were initially prepared to validate the design of our Sirt2 affinity
probe (4). The coꢀcrystal structure of Sirt2 complexed with 3a, revealed an extended binding
mode of the triazoleꢀbased SirReals compared to their parent compound 1a (Fig 2).
ACS Paragon Plus Environment
4

Page 5 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Comparison of the binding mode of SirReal1 1a (from PDB ID 4RMI, colored cyan)
and the triazoleꢀbased SirReal 3a (from PDB ID 5DY5, colored green) at Sirt2 (PDB ID 5DY5).
Hydrogen bonds are shown as dashed orange lines, the conserved water molecule bridging the
interaction of the SirReals is shown as red ball.
The triazole moiety further extends into the acyllysine binding channel where it forms a
hydrogen bond with Arg 97. This results in a more efficient blockade of the substrate binding
site. Since the Nꢀsubstituent at the triazole moiety protrudes towards the entry of the acyllysine
channel (see Fig. 2), we proceeded to attach larger residues to the triazole moiety aiming to
expose certain functional labels to the protein surface. This approach was successfully validated
by the development of the biotinylated SirRealꢀbased affinity probe for Sirt2 (4), which was
shown to be suitable for pullꢀdown experiments as well as biolayer interferometry.29b To probe
the cellular effects of Sirt2, not only by occupancyꢀbased small molecule inhibition of its
deacylase activity but also by eliminating the protein with all its scaffolding functions from the
cellular network, we were highly interested in applying our protocol for the development of
Sirt2ꢀselective probes to the concept of chemically induced protein degradation. This concept is
based on bifunctional small molecules, so called proteolysis targeting chimeras (PROTACs).
These molecular tools are capable to hijacking the cellular quality control by recruiting the
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 36
1
2
3
4
5
6
7
8
9
protein of interest (POI) to E3 ubiquitin ligases for polyubiquitinylation and thus to induce its
proteasomal degradation.30 Since its first description by Crews and coꢀworkers in the year 2001,
the PROTAC concept has been widely applied to induce the degradation of various proteins,
such as kinases,31 transcription factors,32 and epigenetic reader proteins.33 However, neither
epigenetic writer nor epigenetic eraser proteins have been addressed by PROTACs, yet. In terms
of a potential therapeutic application the PROTAC strategy offers several advantages compared
to occupancyꢀbased enzyme inhibition. For example, there is no need to maintain high in vivo
concentrations to ensure sufficient enzyme inhibition, which reduces the risk of offꢀtarget side
effects.30 In this study, we report the development of a PROTAC based on the combination of the
unique features of the sirtuin rearranging ligands (SirReals) as highly potent and isotypeꢀ
selective Sirt2 inhibitors with thalidomide, a bona fide ligand for the E3 ubiquitin ligase
cereblon. So far, thalidomide has been attached by amide coupling procedures to the target
ligand of interest.34 Here, we report for the first time a protocol for the development of a triazoleꢀ
derived PROTAC, which is based on a Cu(I)ꢀcatalyzed cycloaddition of an azidoꢀthalidomideꢀ
conjugate to an alkynylated ligand for the targeted protein, in this case Sirt2. Molecular
modelling of the ternary complex of Sirt2, cereblon and the chimera proposes a binding model of
the interaction of the bifunctional ligand is with both binding pockets.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS
For the design of the SirRealꢀbased PROTAC we combined the structural features of the
Sirt2ꢀselective and highly potent triazoleꢀbased SirReals (see Fig.1) with the cereblon ligand
thalidomide. We have based the design of our PROTAC on the cereblon recruiting thalidomide,
since this approach has already been used to efficiently induce the depletion of epigenetically
active proteins, e.g. the Bromoꢀ and ExtraꢀTerminal domain (BET) family.33a, 33b Furthermore,
ACS Paragon Plus Environment
6

Page 7 of 36
Journal of Medicinal Chemistry
1
2
thalidomide was chosen over other E3 ubiquitin ligase recruiting moieties such as nutlin35 or
VHL ligands33c due to its straightforward synthetic accessibility. Based on our previous work on
a SirRealꢀbased affinity probe29b as well as the study of Winter et al.,33b we already had valid
information on where to place the linker without losing affinity to Sirt2 and cereblon,
respectively. From analysis of a generated model of Sirt2 complexed with cereblon (for details
see Methods section), we predicted a Nꢀbutylꢀ2ꢀoxyacetamide linker to be suitable to recruit
Sirt2 to cereblon with an adequate proximity that allows an efficient ubiquitin transfer (Fig. 3).
The Sirt2ꢀcereblon complex generated using program HADDOCK36 was used to dock our
PROTAC (12). The docking results show that the SirReal and the thalidomide part of 12 is
interacting in a similar way as the original molecules in Sirt2 (PDB ID 5DY5)29b and cereblon
(PDB ID 4CI2)37 (Fig. 3B and Supporting Information, Fig. S2).
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. (A) Docking pose of the ternary Sirt2ꢀ12ꢀcereblon complex. The surface of Sirt2 is
displayed in white whereas cereblon is displayed as ribbon model. The PROTAC 12 is colored
magenta. (B) Predicted binding mode for 12 (magenta) at the Sirt2ꢀcereblon complex. The
surface of the Sirt2 SirReal binding pocket is colored white, whereas the thalidomide binding
pocket of cereblon is colored orange. Hydrogen bonds are shown as dashed orange lines.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 36
1
2
3
4
5
6
7
8
9
To realize the envisaged SirRealꢀbased PROTAC (12), we set up a convergent synthesis route
depicted in Scheme 1. While planning this synthesis route we focused our efforts on i) generating
the SirRealꢀbased PROTAC 12, capable of inducing Sirt2 degradation ii) providing a “clickꢀ
able” azidoꢀthalidomideꢀconjugate as a versatile cereblon ligand ready for conjugation with other
alkynylated ligands. The first step of our synthesis route was the methanolysis of
3ꢀhydroxyphthalic acid anhydride (5) to generate the dimethylester (6).38 The 4ꢀazidobutanꢀ1ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aminium
chloride
(8)
was
prepared
by
a
nucleophilic
substitution
of
Nꢀ(4ꢀbromobutyl)phthalimide (7) with NaN3 followed by hydrazinolysis.39 Then, 8 was
chloroacetylated to obtain compound 9. OꢀAlkylation of 6 by the alkyl chloride 9 gave
compound 10. To synthesize the azidoꢀthalidomideꢀconjugate (11), the dimethylester 10 was first
hydrolyzed to the dicarboxylic acid and then converted into the imide 11, after activation with
N,N'ꢀdicyclohexylcarbodiimide. Finally, compound 11 was conjugated with the propargylated
SirReal analogue 1b through a Cu(I)ꢀcatalyzed Huisgen cycloaddition40 to yield the SirRealꢀ
based PROTAC (12).
ACS Paragon Plus Environment
8

Page 9 of 36
Journal of Medicinal Chemistry
1
2
Scheme 1. Synthesis and IC50 values of SirRealꢀbased PROTAC (12)a
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and conditions: (a) methanol, thionyl chloride, 0 °C ꢀ 80 °C, 4 h, 63% yield; (b)
NaN3, DMF, rt, 12 h, then hydrazine monohydrate, ethanol, 0 °C – rt, 24 h, 99% yield; (c)
chloroacetyl chloride, trimethylamine, DMF, 0 °C ꢀ rt, 1h, 82% yield; (d) Cs2CO3, acetonitrile,
reflux, 5h, then rt, 12 h, 88% yield; (e) NaOH, ethanol, rt, 1.5 h, then 3ꢀaminopiperidineꢀ2,6ꢀ
dione hydrochloride, N,Nꢀdimethylpyridinꢀ4ꢀamine, trimethylamine, dichloromethane, N,N'ꢀ
dicyclohexylcarbodiimide, 0 °C – rt, 22 h, 50% yield; (f) 1b, sodium ascorbate, CuSO4,
water/tertꢀBuOH (1:1), 60 °C, 1 h, then rt, 16 h, 33% yield.
The SirRealꢀbased PROTAC (12) was evaluated for its in vitro inhibition of Sirt1ꢀ3 in a
fluorescenceꢀbased deacetylase activity assay as previously described.41 We detected a highly
potent inhibition of Sirt2 with an IC50 value of 0.25 µM ± 0.02 µM and as desired the
deacetylase activity of isotypes Sirt1 and Sirt3 was only affected at concentrations higher than
100 µM (Scheme 1). To prove the suitability of our SirRealꢀbased PROTAC (12) for chemically
induced Sirt2 degradation, we assessed the cellular levels of Sirt2 in HeLa cells after incubation
with 12 via Western blot analyses (Fig 4). To show the isotype selectivity of our probe, we
determined the change in cellular levels of Sirt1 as well. As an experiment, to show that the Sirt2
degradation is not induced by the unlabeled ligand alone, we incubated the cells just with 3b.
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 36
1
2
3
4
5
6
7
8
9
Indeed, 12 is capable of inducing isotypeꢀselective degradation of Sirt2 (Fig. 4A). Moreover, we
were able to show that the effect of 12 on Sirt2 degradation can be counteracted either by
competition with the unlabeled ligand 3b or by the proteasome inhibitor MG132 (Fig. 4B). The
dependence of Sirt2 depletion on the concentration of 12 is illustrated in Figure 4C. In the range
of 0.05 µM to 5 µM we were able to show a concentrationꢀdependent effect of 12. At higher
concentrations the efficacy of 12 is limited, which might be due to saturation effects on one of
the targets. No effects could be observed in the control experiments with 3b and thalidomide,
neither in single treatment nor in combination (Fig. 4C). A more detailed time course of the
chemically induced Sirt2 degradation in HeLa cells due to treatment with 12 is given in Fig. 4D.
A maximum effect could be detected after 2 hours of treatment. In addition, the effect of 12 on
the level and distribution of Sirt2 was visualized in living HeLa cells expressing EGPFꢀSirt2 by
fluorescence microscopy. Consistent with Western blot results, these studies indicate that
treatment with 12 results in reduced Sirt2 levels as well. Moreover, the dotꢀlike pattern of the
fluorescence signal suggests enrichment of Sirt2 in proteasomes (Fig. 5).42 To investigate the
phenotype variation between cells treated with the Sirt2 enzymatic inhibitor 3b and cells treated
with our PROTAC (12), we probed the acetylation levels of tubulin, a wellꢀknown Sirt2
substrate,10 via immunofluorescence microscopy. Depletion of Sirt2 due to treatment with 12
resulted in a more pronounced acetylation of the microtubule network with enhanced process
elongation than sole enzymatic inhibition by 3b (Fig 6, for raw images, see Supporting
Information, Fig. S4). By means of these studies we were able to show a successful example for
the envisaged superiority of the PROTAC concept with its extended target coverage over
occupancyꢀdriven enzymatic inhibition.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
10

Page 11 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Quantification of the effect of 12 on the intracellular Sirt levels in HeLa cells by
Western blot analyses. (A) Treatment with 12 (10 µM) is able to chemically induce Sirt2
degradation. Sirt1 levels are not affected (n = 3). (B) The effect of 12 on the Sirt2 level in HeLa
cells expressing EGFPꢀSirt2 in the absence and presence of the proteasome inhibitor MG132
(n = 3). (C) Depletion of Sirt2 is dependent on the concentration of 12. Control experiments with
3b and thalidomide (Th) showed no significant effects on cellular Sirt2 levels, neither in single
treatment nor in combination (n = 3). (D) Time course of Sirt2 depletion induced by 12 (n = 3).
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 36
1
2
3
4
5
6
Representative Western blots can be found in the Supporting Information (Fig. S3). Statistics
(tꢀtest): * p ≤ 0.05
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Effect of 12 on the level and distribution of EGFPꢀSirt2 (green) in living HeLa cells as
visualized by fluorescence microscopy. (A) Homogeneous distribution of the labelled Sirt2
(green). (B) Reduced Sirt2 level in the 12ꢀtreated cells with dotꢀlike pattern of the fluorescence
signal. Nuclei are stained with Hoechst 33342 (blue). Bar: 10 µm.
Figure 6. In cultured HeLa cells, depletion of Sirt2 due to treatment with 12 results in a more
pronounced acetylation of the microtubule network than sole enzymatic inhibition by 3b.
Acetylation level of the microtubule network (red) in the presence or absence of sirtuin inhibitor
ACS Paragon Plus Environment
12

Page 13 of 36
Journal of Medicinal Chemistry
1
2
3
4
3b and the SirRealꢀbased PROTAC (12), respectively. Tests were performed at a compound
5
6
concentration of 10 µM. Nuclei were DAPIꢀstained (blue). The scale bar represents 10 ꢁm.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Thus, 12 is suitable to induce Sirt2 degradation with concomitantly increased efficacy of
tubulin hyperacetylation and therefore is the first example of a PROTAC targeting an epigenetic
eraser protein.
DISCUSSION AND CONCLUSION
In this study, we report the structureꢀbased development of a PROTAC for Sirt2. Based on
our previously published work on the sirtuin rearranging ligands (SirReals) and a SirRealꢀbased
affinity probe, we developed a highly potent and isotypeꢀselective probe, which is capable of
chemically inducing the degradation of Sirt2. While several PROTACs for epigenetic reader
proteins have been reported yet, this is the first example of a PROTAC targeting an epigenetic
eraser protein. Since the PROTAC concept offers several advantages compared to occupancyꢀ
driven enzyme inhibitors or genetic ablation of the enzyme, our novel tool for Sirt2 will open up
new avenues to dissect the role of Sirt2 in biology and medicine. Moreover, with our azidoꢀ
thalidomideꢀconjugate (11) we provide a versatile cereblon ligand building block, ready to be
“clicked” to alkynylated ligands of other proteins. Given the broad functional group tolerance of
this type of chemistry43 and the availability of alkyne derivatives for many targets that have been
used in a large variety of chemical biology studies, this technique will be amenable to a plethora
of different targets and inhibitors. Thus, our study will not only help to interrogate Sirt2 biology
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 36
1
2
3
4
5
6
and druggability on a new level but also will accelerate the development of novel PROTACs for
other enzymes.
7
8
9
EXPERIMENTAL SECTION
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1. Cloning, expression and purification of recombinant proteins: Human Sirt1133ꢀ747 was
expressed as a GSTꢀtagged enzyme and purified as described previously.44 Sirt256ꢀ356 and
Sirt3118ꢀ395 were expressed and purified as described previously with minor modifications.28
Sirt256ꢀ356 and Sirt3118ꢀ395 were resuspended in lysis buffer (25 mM KH2PO4, 25 mM NaH2PO4,
400 mM NaCl, 5% (v/v) glycerol, 5 mM betaꢀmercaptoethanol, pH 8.0) and purified as described
with a final Superdex S75 26/60 gel filtration column (25 mM Hepes, 200 mM NaCl, 5 % (v/v)
glycerol, pH 8.0). Identity and purity of the produced enzymes was verified using SDSꢀPAGE.45
Protein concentration was determined by the Bradford assay.46 Deacetylase activity of sirtuin
isotypes was NAD+ꢀdependent and could be inhibited by nicotinamide. The eukaryote EGFPꢀ
Sirt2 plasmid was a generous gift from Brian J. North (Department of Genetics, Harvard Medical
School, Boston, Massachusetts, USA).
2. In Vitro Testing: Inhibition tests for Sirt1, Sirt2 and Sirt3 were performed with a highꢀ
throughput fluorescenceꢀbased histone deacetylase assay, as previously established by Heltweg
et al.41 Human Sirt1133ꢀ747, Sirt256ꢀ356 or human Sirt3118ꢀ395 were mixed with assay buffer (50 mM
Tris, 137 mM NaCl, 2.7 mM KCl, pH 8.0), NAD+ (final assay concentration 500 ꢁM), the
substrate Zꢀ(Ac)LysꢀAMC (ZMAL, final assay concentration 10,5 ꢁM), the inhibitor dissolved
in DMSO at different concentrations or DMSO as a control (final DMSO concentration 5%
(v/v)). To assure initial state conditions, total substrate conversion of controls was adjusted to
approximately 15% ꢀ 30%. The assay was performed in 96ꢀwell plates with a reaction volume of
ACS Paragon Plus Environment
14

Page 15 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
60 ꢁL per well. All determinations were performed in triplicates. After an incubation of 4 h at
37 °C and 140 rpm, deacetylation was stopped by the addition of 60 ꢁL of a solution containing
trypsin and nicotinamide (50 mM Tris, 100 mM NaCl, 6.7% (v/v) DMSO, trypsin 5.5 U ꢁLꢀ1,
8 mM nicotinamide, pH 8.0). The microplate was further incubated for 20 min at 37 °C and 140
rpm. Fluorescence intensity was measured in a microplate reader (BMG Polarstar, λex = 390 nm,
λem = 460 nm). Rates of inhibition were calculated by using the controls, containing no inhibitor,
as a reference. Graphpad Prism software (La Jolla, CA) was employed to determine IC50 values.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3. Cell Culture: HeLa (ATCCꢀ2) cells were cultured in Dulbecco’s Modified Eagle’s Medium
(high glucose) supplemented with 10% (v/v) fetal calf serum, 100 ꢁg/mL kanamycin (all
reagents from Sigma, complete medium) in a humidified incubator at 37 °C with 5% CO2. For
immunoblotting of endogenous Sirt2 and Sirt1, 2x105 cells were plated in 6ꢀwell plates. A stable
EGFPꢀSirt2 expressing cell population was generated by transfection of 106 HeLa cell with
Turbofect (Thermo Fisher) and EGFPꢀSirt2 plasmid in 6 cm diameter petri dish according to the
manufacturer’s protocol. Next day, the cells were selected by using 500 µg/mL G418 for
4 weeks. Batches of this cell population were frozen for further work. For immunoblotting,
2x104 EGFPꢀSirt2ꢀHeLa cells were plated onto 24ꢀwell plates. Before adding compounds from
DMSO stocks, cycloheximide protein synthesis block was applied on cells (5 µg/ml on 6ꢀwell
plates, 3 µg/ml on 24ꢀwell plates) to detect degradation more effectively. For live microscopic
analysis, from the EGFPꢀSirt2 expressing cell population, 2x104 cells were seeded onto the glass
based middle of 35 mm diameter imaging dishes 1.0 (Mo Bi Tec, Germany) in 1.0 mL complete
medium and incubated overnight. 12 (10 µM) was added into the dish to the treatment of cells,
which were incubated for the given incubation time. Unless otherwise stated, we chose an
incubation time of 4 hours. Next the complete medium was changed to a phenol redꢀfree one
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 36
1
2
3
4
5
6
7
8
9
(Dulbecco’s Modified Eagle’s Medium, Sigma, D5921). Control cells were treated by DMSO
only. Live cell images were acquired with a Zeiss LSCM 710 microscope using an oiled 40× NA
= 1.4 Plan Apo objective. Nuclei were stained by 0.5 µg/mL Hoechst 33342 for 5 minutes.
Controls contained the corresponding amount of vehicle (DMSO).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4. Immunoblotting: For the detection of Sirt2, Sirt1 or glyceraldehydeꢀ3ꢀphosphate
dehydrogenase (GAPDH) levels in cellular samples, we kept the cells at 37 °C and washed them
with preꢀwarmed PBS. Next, the cells were lysed in 100 (24ꢀwell plate) or 300 (6ꢀwell plate) ꢁL
1× reducing sample buffer containing protein inhibitor mix and 2 mM EDTA (SigmaꢀAldrich) in
situ. Samples were centrifuged at 10000g at 4 °C for 5 min, and the supernatants were stored at
−70 °C. Samples were analyzed by 10% SDSꢀPAGE and blotted onto polyvinylidene difluoride
membrane (Millipore). The blot was developed sequentially using a rabbit antibody against Sirt2
(1:5000, affinityꢀisolated abcam ab67299, or RabMab ab51023), then using an affinityꢀisolated
rabbit antibody against Sirt1 (1:5000, Sigma, S5447) or using an affinityꢀisolated rabbit antibody
against GAPDH (1:5000, Sigma, G9545). Sirt1 could not be detected in the case of 24ꢀwell
samples, here GAPDH was used as loading control. Antibodies were detected by polyclonal antiꢀ
rabbit IgGꢀperoxidase conjugate, (1:5000, Thermo Scientific). Peroxidase reaction was detected
using Immobilon Western substrate (Millipore) by a BioꢀRad ChemiDoc MP Imaging system
and its ImageLab 4.1 software. Intensity of spots was analyzed by ImageJ 1.49 using Measure
command and subtracting background values. The intensities were normalized by the average of
control values.
5. Immunofluorescence Microscopy: Immunofluorescence microscopy was performed as
previously reported.28
ACS Paragon Plus Environment
16

Page 17 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
6. Computational Methods: The following Sirt2 Xꢀray structures were used in the current
work: human Sirt2ꢀSirRealꢀtriazole probe (PDB ID 5DY5)29b and human DDB1ꢀCRBN E3
ubiquitin ligaseꢀthalidomide (PDB ID 4CI2).37 All protein structures were prepared by using the
Structure Preparation module in MOE 2012.10 (Chemical Computing Group, Montreal, Canada).
Hydrogen atoms were added, for titratable amino acids the protonation state was calculated using
the Protonate 3D module in MOE 2012.10. Water molecules and ligand atoms except the zinc
ion and one conserved water molecule bound to Sirt2 Pro94 were removed from the structures.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To generate the Sirt2ꢀcereblon complexes the proteinꢀprotein docking program HADDOCK
(version 2.2) was used.36 HADDOCK uses an initial rigidꢀbody docking process which generates
typically a large number of structures in the order of thousands. From these a number of
structures, typically several hundred, are selected for further flexible refinement and scoring
using HADDOCK score, vanꢀderꢀWaals and electrostatic interactions. The following residues
located at the entrance of both binding pockets were chosen to define the proteinꢀprotein
interaction site: Ser100, Phe235, Ser263, Ile300 for Sirt2, Gln86, His103, Phe150, Thr376 for
cereblon. HADDOCK calculated eleven different clusters for the Sirt2ꢀcerebon complex. Among
the eleven complexes the three best scored complexes were chosen for further analysis and
ligand docking studies. The results of the HADDOCK proteinꢀprotein docking are provided in
the Supplementary Information (Figure S1). Docking studies were performed using program
GOLD (version 5.2) and GoldScore.47 From each of the three proteinꢀprotein clusters provided
by HADDOCK, four complexes were chosen for the docking study. The backbone carbonyl
oxygen of Sirt2ꢀPhe235 was defined as center of the binding site with 20 Å radius. 50 docking
poses were calculated for the ligand. All other options were left at their default values. The
retained docking poses were further refined using a MMꢀGBSA protocol. The AMBEREHT12
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 36
1
2
3
4
5
6
7
8
9
force field and the GBSA method implemented in MOE 2012.10 were used to relax the proteinꢀ
inhibitor complexes. During the minimization protein backbone atoms were tethered using a
force constant of (3/2) kT / 2 (σ = 0.5 Å).48 Complexes showing the lowest binding free energies
values were selected. The docking protocol has already been successfully applied in case of
Sirt2.29 For cereblon (PDB ID 4CI2)37 GOLD was correctly reproducing the binding mode of
thalidomide with an RMSD value of 0.45 Å.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7. Chemistry: Starting materials and reagents were obtained from commercial suppliers and
used without further purification. Thinꢀlayer chromatography (TLC) for reaction monitoring was
performed with alumina plates coated with Merck silica gel 60 F254 (layer thickness: 0.2 mm)
and analyzed under UVꢀlight (254 nm). The composition of the mobile phase was adjusted to
1
compound properties. Yields were not optimized. H NMR and 13C NMR spectra were recorded
on Bruker Avance III HD spectrometer at 400 MHz and 100 MHz. The spectra are referenced
against the NMR solvents and are reported as follows: 1H: chemical shift δ (ppm), multiplicity (s
= singlet, d = doublet, dd = doublet of doublets, t = triplet, q = quartet, quint = quintet, sex =
13
sextet, m = multiplet, b = broad), integration, coupling constant (J in Hz). C: chemical shift δ
(ppm), abbreviations: (q) = quaternary carbons, quaternary carbons that could not be found in 13C
1
13
spectra but in HMBC or HSQC are additionally marked with an asterisk (*). H or C signals
that are partially overlapping with solvent residual signals are marked with a hashtag (#). The
assignment resulted from HMBC and HSQC experiments. Purity of the tested compound was
determined by HPLC and UV detection (λ = 210 nm) and was > 95%. HPLC analysis was
performed using the following conditions: Eluent A: H2O containing 0.05% TFA, Eluent B:
acetonitrile containing 0.05% TFA, flow rate 1 mL minꢀ1, linear gradient conditions (0−4 min, A
= 90%, B = 10%; 4−29 min, linear increase to 100 % of B; 29−31 min, B = 100%; 31−40 min, A
ACS Paragon Plus Environment
18

Page 19 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
= 10%, B = 90%), Merk LiChrospher 60 RPꢀselect B (5 µm, 60 Å, 250 x 4 mm) as an analytical
column. Melting temperatures were determined in glass capillary tubes with the Stuart Melting
Point Apparatus SMP2. Mass spectra with electrospray ionization (ESI) were recorded on an
Advion expression CMS spectrometer or on a Exactive device (Thermo Fisher Scientific, for
FTMS). The synthesis of compounds 6 and 8 has already been reported.38ꢀ39 Due to changes in
the experimental procedure and to show unpublished characterization data, we have outlined the
synthesis and the characterization data for those compound as well.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Dimethyl 3ꢀhydroxyphthalate (6). From 3ꢀhydroxyphthalic acid anhydride (5) according to a
modified procedure published by Kristiansen et al.38 Thionyl chloride (5 equiv) was added
dropwise to an iceꢀcold solution of 3ꢀhydroxyphthalic acid anhydride (1 equiv, 3.7 mmol) in
methanol (10 mL). Then, the mixture was stirred for 4 h at 80 °C under reflux conditions. After
completion, solvents were removed under reduced pressure. The remaining crude product was
purified by automated flash chromatography (cyclohexane/EtOAc: gradient 10% ꢀ 100%
1
EtOAc). Yield, 63% of a colorless solid; Rf, 0.46 (cyclohexane/EtOAc, 7:3 (v/v)); H NMR
(DMSOꢀd6, δ [ppm]): 10.34 (s, 1H, ꢀOꢀH), 7.41ꢀ7.36 (m, 2H, Hꢀ5,6), 7.18ꢀ7.14 (m, 1H, Hꢀ4),
3.80 (s, 3H, ꢀCOOꢀCH3 at Cꢀ1), 3.77 (s, 3H, ꢀCOOꢀCH3 at Cꢀ2). 13C NMR (DMSOꢀd6, δ [ppm]):
167.57 q (ꢀCOOꢀCH3 at Cꢀ2), 166.02 q (ꢀCOOꢀCH3 at Cꢀ1), 154.96 q (Cꢀ3), 130.93 (Cꢀ5),
128.86 q (Cꢀ1), 122.96 q (Cꢀ 2), 120.85 (Cꢀ4), 120.36 (Cꢀ6), 52.89 (ꢀCOOꢀCH3 at Cꢀ1), 52.50
(ꢀCOOꢀCH3 at Cꢀ2). ESIꢀMS(ꢀ): 209.2 [M – H]ꢀ.
4ꢀAzidobutanꢀ1ꢀaminium chloride (8).39 From Nꢀ(4ꢀbromobutyl)phthalimide (7) according to a
modified procedure published by Cunningham et al.39 7 (1 equiv, 9.9 mmol) was dissolved in
DMF (20 mL). After the addition of NaN3 (1.1 equiv) the reaction mixture was stirred for 12 h at
ambient temperature. Water (25 mL) was added and the aqueous layer was extracted with EtOAc
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 36
1
2
3
4
5
6
7
8
9
(30 mL, three times). The organic layers were dried over MgSO4 and evaporated. The residue
was dissolved in ethanol (40 mL). The solution was cooled to 0 °C and hydrazine monohydrate
(2 equiv) was gradually added in portions while stirring. Then, the mixture was stirred for 24 h at
ambient temperature. After completion, the reaction mixture was acidified with concentrated
hydrochloric acid. Solvents were evaporated under reduced pressure and water (30 mL) was
added to the residue. Potassium hydroxide was added to adjust a basic pH value (pH > 10). The
aqueous layer was extracted with dichloromethane (30 mL, four times). The combined organic
layers were dried over MgSO4, acidified with concentrated hydrochloric acid, and evaporated
under reduced pressure. Yield, > 99% of a colorless oil; 1H NMR (DMSOꢀd6, δ [ppm]): 8.10 (bs,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
3
3H, H3N+ꢀ), 3.37 (t, 2H, J = 6.45 Hz, H3N+ꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3), 2.78 (sex, 2H, J = 6.33
Hz, H3N+ꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3), 1.69ꢀ1.51 (m, 4H, H3N+ꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3).
13C NMR (DMSOꢀd6,
δ
[ppm]):
50.52
(H3N+ꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3),
38.60
(H3N+ꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3), 25.74 (H3N+ꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3), 24.69 (H3N+ꢀCH2ꢀCH2ꢀ
CH2ꢀCH2ꢀN3). ESIꢀMS(+): 115.2 [M ꢀ Cl]+.
Nꢀ(4ꢀazidobutyl)ꢀ2ꢀchloroacetamide (9). 8 (1equiv, 9.7 mmol) was dissolved in DMF (15 mL)
and trimethylamine (2 equiv) was added. The mixture was cooled to 0 °C and chloroacetyl
chloride was gradually added. The reaction mixture was stirred for 10 min at 0 °C and then for
further 60 min at ambient temperature. After the addition of a saturated aqueous solution of
sodium bicarbonate (30 mL) the mixture was extracted with EtOAc (30 mL, three times). The
combined organic layers were dried over MgSO4 and evaporated under reduced pressure.
Yield, 82% of a yellowish oil; 1H NMR (DMSOꢀd6, δ [ppm]): 8.33ꢀ8.15 (m, 1H, amideꢀH), 4.04
3
3
(s, 2H, ClꢀCH2ꢀCOꢀNHꢀ), 3.34# (t, 2H, J = 6.64 Hz, ꢀCH2ꢀCH2ꢀN3), 3.11 (q, 2H, J =
6.31, ꢀCOꢀNHꢀCH2ꢀCH2ꢀ), 1.60ꢀ1.40 (m, 4H, ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3). 13C NMR
ACS Paragon Plus Environment
20

Page 21 of 36
Journal of Medicinal Chemistry
1
2
3
4
(DMSOꢀd6, δ [ppm]): 166.25 (amideꢀC), 50.71 (ꢀCH2ꢀN3), 43.08 (ClꢀCH2ꢀCOꢀNHꢀ), 38.75 (ꢀCOꢀ
5
6
NHꢀCH2ꢀ), 26.52 (ꢀCOꢀNHꢀCH2ꢀCH2ꢀ), 26.09 (ꢀCH2
CH2ꢀN3). FTMS(+): 191.0693 [M + H]+.
ꢀ
7
8
9
Dimethyl 3ꢀ(2ꢀ((4ꢀazidobutyl)amino)ꢀ2ꢀoxoethoxy)phthalate (10). 7 (1 equiv, 2.1 mmol) and 9
(1.1 equiv) were dissolved in acetonitrile (20 mL) and Cs2CO3 (2 equiv) was added. The reaction
mixture was stirred for 5 h at 60 °C under reflux conditions and for further 12 h at ambient
temperature. Water (15 mL) was added to the reaction mixture and the aqueous layer was
extracted with dichloromethane (15 mL, once) and EtOAc (20 mL, three times). The combined
organic layers were dried over MgSO4 and evaporated under reduced pressure. The remaining
crude product was purified by automated flash chromatography (cyclohexane/EtOAc: gradient
10% ꢀ 100% EtOAc). Yield, 88% of a colorless oil; Rf, 0.33 (cyclohexane/EtOAc, 6:4 (v/v));
1H NMR (DMSOꢀd6, δ [ppm]): 7.80 (t, 1H, 3J = 6.04 Hz, amideꢀH), 7.59ꢀ7.53 (m, 2H, phthalate
Hꢀ5,6), 7.38ꢀ7.31 (m, 1H, phthalate Hꢀ4), 4.62 (s, 2H, ꢀOꢀCH2ꢀCOꢀNHꢀ), 3.85ꢀ3.80 (m,
6H, ꢀCOOꢀCH3), 3.33# (t, 2H, 3J = 6.12 Hz, ꢀCH2ꢀCH2ꢀN3), 3.16 (q, 2H, 3J = 6.04 Hz, ꢀCOꢀNHꢀ
CH2ꢀCH2), 1.52ꢀ1.47 (m, 4H, ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3). 13C NMR (DMSOꢀd6, δ [ppm]):
167.35 q (amideꢀC), 167.13 q (ꢀCOOꢀCH3 at phthalate Cꢀ2), 165.59 q (ꢀCOOꢀCH3 at phthalate
Cꢀ1), 155.21 q (phthalate Cꢀ3), 131.44 (phthalate Cꢀ5), 129.01 q (phthalate Cꢀ1), 124.81 q
(phthalate Cꢀ2), 122.60 (phthalate Cꢀ6), 118.38 (phthalate Cꢀ4), 68.35 (OꢀCH2ꢀCOꢀNHꢀ), 53.13
(ꢀCOOꢀCH3 at phthalate Cꢀ1), 52.90 (ꢀCOOꢀCH3 at phthalate Cꢀ2), 50.71 (ꢀCH2ꢀN3), 38.16
(ꢀCOꢀNHꢀCH2ꢀ), 26.63 & 26.05 (ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3). ESIꢀMS(+): 387.1
[M + Na]+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Nꢀ(4ꢀazidobutyl)ꢀ2ꢀ((2ꢀ(2,6ꢀdioxopiperidinꢀ3ꢀyl)ꢀ1,3ꢀdioxoisoindolinꢀ4ꢀyl)oxy)acetamide (11):
10 (1 equiv, 1.85 mmol) was dissolved in ethanol (15 mL) and an aqueous sodium hydroxide
solution (2.5 equiv, 3 M) was added. The reaction mixture was stirred for 1.5 h at ambient
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 36
1
2
3
4
5
6
7
8
9
temperature. Subsequently, solvents were evaporated under reduced pressure and the residue was
dissolved in water (15 mL). The aqueous layer was washed with EtOAc. The organic layer was
discarded and the aqueous layer was acidified with diluted hydrochloric acid and extracted with
EtOAc (20 mL, three times). The combined organic layers were dried over MgSO4 and
evaporated under reduced pressure. The crude dicarboxylic acid, 3ꢀaminopiperidineꢀ2,6ꢀdione
hydrochloride (1.2 equiv), N,Nꢀdimethylpyridinꢀ4ꢀamine (0.2 equiv), and trimethylamine (3.4
equiv) were dissolved in dichloromethane (15 mL). The mixture was cooled to 0 °C and
N,N'ꢀdicyclohexylcarbodiimide was added under nitrogen atmosphere. After 15 min of stirring at
0 °C the reaction mixture was stirred for further 22 h at ambient temperature. The precipitate was
separated by filtration was dried under reduced pressure. The crude product was purified by
automated flash chromatography (dichloromethane/methanol: gradient 1% ꢀ 10% methanol).
Yield, 50% of a white solid; mp, 165 °C; Rf, 0.71 (dichlormethane/methanol, 19:1 (v/v));
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
1H NMR (DMSOꢀd6, δ [ppm]): 11.14 (bs, 1H, ꢀCOꢀNHꢀCOꢀ), 8.03 (t, 1H, J = 6.04 Hz,
3
amideꢀH), 7.82 (dd, 1H, J = 8.59 Hz, 3J = 7.33 Hz, 1,3ꢀdioxoisoindoline Hꢀ6), 7.51 (d, 1H, 3J =
3
7.33 Hz, 1,3ꢀdioxoisoindoline Hꢀ7), 7.40 (d, 1H, J = 8.59 Hz, 1,3ꢀdioxoisoindoline Hꢀ5), 5.13
(dd, 1H 3J = 12.87 Hz, 3J = 5.43 Hz, 1H, 2,6ꢀdioxopiperidine Hꢀ3), 4.79 (s,
3
3
2H, ꢀOꢀCH2ꢀCOꢀNHꢀ), 3.34# (t, 2H, J = 6.55 Hz, ꢀCH2ꢀCH2ꢀN3), 3.17 (q, 2H, J = 6.04
Hz, ꢀCOꢀNHꢀCH2ꢀCH2), 2.95ꢀ2.86 (m, 1H, 2,6ꢀdioxopiperidine Hꢀ4), 2.63ꢀ2.53# (m, 2H,
2,6ꢀdioxopiperidine Hꢀ5), 2.06ꢀ2.03 (m, 1H, 2,6ꢀdioxopiperidine Hꢀ4), 1.58ꢀ1.44 (m,
4H, ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀN3). 13C NMR (DMSOꢀd6,
δ
[ppm]): 173.21
q
(2,6ꢀdioxopiperidine Cꢀ6), 170.31 q (2,6ꢀdioxopiperidine Cꢀ2), 167.19 q (amideꢀC), 167.15 q
(1,3ꢀdioxoisoindoline Cꢀ1), 165.92 q (1,3ꢀdioxoisoindoline Cꢀ3), 155.49 q (1,3ꢀdioxoisoindoline
Cꢀ4), 137.35 (1,3ꢀdioxoisoindoline Cꢀ6), 133.45 q (1,3ꢀdioxoisoindoline Cꢀ7a), 120.80
ACS Paragon Plus Environment
22

Page 23 of 36
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(1,3ꢀdioxoisoindoline Cꢀ5), 117.23 (1,3ꢀdioxoisoindoline Cꢀ3a), 116.48 (1,3ꢀdioxoisoindoline
Cꢀ7), 68.07 (ꢀOꢀCH2ꢀCOꢀNHꢀ), 50.71 (ꢀCH2ꢀN3), 49.20 (2,6ꢀdioxopiperidine Cꢀ3), 38.20
(ꢀCOꢀNHꢀCH2ꢀ), 31.36 (2,6ꢀdioxopiperidine Cꢀ5), 26.67 & 26.08 (ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀ
N3), 22.39 (2,6ꢀdioxopiperidine Cꢀ4). ESIꢀMS(ꢀ): 427.2 [M ꢀ H]ꢀ.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2ꢀ((4,6ꢀdimethylpyrimidinꢀ2ꢀyl)thio)ꢀNꢀ(5ꢀ(3ꢀ((1ꢀ(4ꢀ(2ꢀ((2ꢀ(2,6ꢀdioxopiperidinꢀ3ꢀyl)ꢀ1,3ꢀ
dioxoisoindolinꢀ4ꢀyl)oxy)acetamido)butyl)ꢀ1Hꢀ1,2,3ꢀtriazolꢀ4ꢀyl)methoxy)benzyl)thiazolꢀ2ꢀ
yl)acetamide (12): 11 (1 equiv, 0.47 mmol) and 2 (1 equiv) were suspended in a water/tertꢀ
butanol mixture (10 mL, 1:1). The suspension was sonified for 15 min. An aqueous CuSO4
solution (0.2 equiv, 0.1 M) and an aqueous solution of sodium ascorbate (0.4 equiv, 0.1 M) were
added in that order. The resulting reaction mixture was stirred for 1h at 60 °C and for further
16 h at ambient temperature under nitrogen atmosphere. After completion the reaction was
quenched by the addition of water (15 mL). The aqueous layer was extracted with
dichloromethane (30 mL, three times) and the combined organic layers were dried over MgSO4
and evaporated under reduced pressure. The remaining crude product was purified by automated
flash chromatography (dichloromethane/methanol: gradient 1% ꢀ 10% methanol). Finally, the
product was recrystallized from methanol. Yield, 33% of a white solid; mp, 135 °C; Rf, 0.25
(dichlormethane/methanol, 19:1 (v/v)); 1H NMR (DMSOꢀd6, δ [ppm]): 12.22 (bs,
3
1H, ꢀNHꢀCOꢀCH2ꢀSꢀ), 11.13 (bs, 1H, ꢀCOꢀNHꢀCOꢀ), 8.19 (s, 1H, triazoleꢀH), 8.03 (t, 1H, J =
5.82 Hz, ꢀNHꢀCOꢀCH2ꢀOꢀ), 7.79 (dd, 1H, 3J = 8.44 Hz, 3J = 7.36 Hz, 1,3ꢀdioxoisoindoline Hꢀ6),
7.48 (d, 1H, 3J = 7.36 Hz, 1,3ꢀdioxoisoindoline Hꢀ7), 7.37 (d, 1H, 3J = 8.44 Hz,
1,3ꢀdioxoisoindoline Hꢀ5), 7.27ꢀ7.20 (m, 2H, thiazole Hꢀ4 & phenyl Hꢀ5), 6.94 (s, 1H,
3
pyrimidine Hꢀ5), 6.92ꢀ6.87 (m, 2H, phenyl Hꢀ2,4), 6.84 (d, 1H, J = 7.51 Hz, phenyl Hꢀ6),
5.16ꢀ5.09 (m, 3H, 2,6ꢀdioxopiperidine Hꢀ3 & triazoleꢀCH2ꢀOꢀ), 4.78 (s, 2H, ꢀOꢀCH2ꢀCOꢀNHꢀ),
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 36
1
2
3
4.37 (t, 2H, J = 7.00 Hz, ꢀCH2ꢀCH2ꢀtriazole), 4.08# (s, 2H, ꢀNHꢀCOꢀCH2ꢀSꢀ), 4.04 (s, 2H,
thiazoleꢀCH2ꢀphenyl), 3.21ꢀ3.12# (m, 2H, ꢀCH2ꢀCH2ꢀNHꢀCOꢀ), 2.96ꢀ2.79 (m, 1H,
2,6ꢀdioxopiperidine Hꢀ4), 2.65ꢀ2.53# (m, 2H, 2,6ꢀdioxopiperidine Hꢀ5), 2.28 (s, 6H,
pyrimidine ꢀCH3), 2.09ꢀ2.00 (m, 1H, 2,6ꢀdioxopiperidine Hꢀ4), 1.86ꢀ1.73 (m, 2H, ꢀCOꢀNHꢀCH2ꢀ
CH2ꢀCH2ꢀCH2ꢀtriazole), 1.46ꢀ1.35 (m, 2H, ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀtriazole) . 13C NMR
(DMSOꢀd6, δ [ppm]): 172.85 q (2,6ꢀdioxopiperidine Cꢀ6), 169.91 q (2,6ꢀdioxopiperidine Cꢀ2),
168.91 q (pyrimidine Cꢀ2), 167.02 q (pyrimidine Cꢀ4,6), 166.84 q (ꢀNHꢀCOꢀCH2ꢀSꢀ & ꢀNHꢀCOꢀ
CH2ꢀOꢀ), 166.75 q (1,3ꢀdioxoisoindoline Cꢀ1), 165.50 q (1,3ꢀdioxoisoindoline Cꢀ3), 158.21 q
(phenyl Cꢀ3), 156.91 q (thiazole Cꢀ2), 155.11 q (1,3ꢀdioxoisoindoline Cꢀ4), 142.61 q (triazole
Cꢀ4), 141.90 q (phenyl Cꢀ1), 136.93 (1,3ꢀdioxoisoindoline Cꢀ6), 134.79 (thiazole Cꢀ4), 133.05 q
(1,3ꢀdioxoisoindoline Cꢀ7a), 131.07 q (thiazole Cꢀ5), 129.66 (phenyl Cꢀ5), 124.37 (triazole Cꢀ5),
120.92 (phenyl Cꢀ6), 120.37 (1,3ꢀdioxoisoindoline Cꢀ5), 116.82 (1,3ꢀdioxoisoindoline Cꢀ3a),
116.13 (pyrimidine Cꢀ5), 116.05 (1,3ꢀdioxoisoindoline Cꢀ7), 114.95 (phenyl Cꢀ2), 112.44
(phenyl Cꢀ4), 67.65 (OꢀCH2ꢀCOꢀNHꢀ), 61.06 (ꢀOꢀCH2ꢀtriazole), 49.01 (triazoleꢀCH2ꢀ), 48.81
(2,6ꢀdioxopiperidine Cꢀ3), 37.66 (ꢀCOꢀNHꢀCH2ꢀ), 34.07 (ꢀNHꢀCOꢀCH2ꢀSꢀ), 31.90
(thiazoleꢀCH2ꢀphenyl), 30.96 (2,6ꢀdioxopiperidine Cꢀ5), 27.15 (ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀ
triazole), 26.02 (ꢀCOꢀNHꢀCH2ꢀCH2ꢀCH2ꢀCH2ꢀtriazole), 23.26 (pyrimidine ꢀCH3), 22.00
(2,6ꢀdioxopiperidine Cꢀ4). Purity: 97.7% (18.8 min). ESIꢀMS(ꢀ): 851.9 [M ꢀ H]ꢀ.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8. PAINS Analysis: Moreover, we assessed whether the compounds possessed any structural
features which should render them panꢀassayꢀinterferers by applying the PAINS filter.49
Screening against PAINS was performed using the program KNIME,50 using a workflow
downloaded from http://www.myexperiment.org/workflows/1841.html.51 The target molecule 12
ACS Paragon Plus Environment
24

Page 25 of 36
Journal of Medicinal Chemistry
1
2
3
4
is not identified as a potential PAINS. Intermediates 8-11 are flagged due the azide function
5
6
(contain an azo group) but are only used in synthesis and are not evaluated in biological assays.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information. The Supporting Information is available free of charge on ACS
Publications website at DOI:
Supplementary Figures: Fig S1–4 (PDF)
Molecular formular strings (CSV)
AUTHOR INFORMATION
Corresponding Author
*Phone: +497612034896. Fax: +497612036321. Eꢀmail: manfred.jung@pharmazie.uniꢀ
freiburg.de
Present Addresses
∇ Matthias Schiedel: Department of Chemistry, Chemistry Research Laboratory, University of
Oxford, Mansfield Road, OX1 3TA Oxford, United Kingdom
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 36
1
2
3
4
Notes
5
6
The authors declare no competing financial interest.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOWLEDGMENT
We thank the Deutsche Forschungsgemeinschaft (DFG) for funding (INST 39/931ꢀ1 and
CRC992 to M.J., Ju 295ꢀ13/1 to D.H. and M.J.). This work was supported by the European
Concerted Research Action COST Action EPICHEMBIO [CM1406]; and Hungarian National
Scientific Research Fund Grants OTKA [Kꢀ101039 and Kꢀ112144] to J. O.
ABBREVIATIONS
AMC, 7ꢀAminoꢀ4ꢀmethylcoumarin; BubR1, budding uninhibited by benzimidazoles related 1
protein; EGFP, enhanced green fluorescent protein; EtOAc, ethyl acetate; HOXA10, Hoxa10
homeobox A10 protein; KDAC, lysine deacetylase; NFκB, nuclear factor 'kappaꢀlightꢀchainꢀ
enhancer' of activated Bꢀcells; p53, tumor suppressor protein p53; POI, protein of interest;
PROTAC, proteolysis targeting chimera; SirReal, sirtuin rearranging ligand; SMILES, simplified
molecular input line entry specification; VHL, Von Hippel–Lindau tumor suppressor; ZMAL, Zꢀ
(Ac)LysꢀAMC
Authors will release the atomic coordinates of the proteinꢀprotein docking model upon article
publication.
ACS Paragon Plus Environment
26

Page 27 of 36
Journal of Medicinal Chemistry
1
2
3
4
REFERENCES
5
6
7
8
9
1.
de Ruijter, A. J.; van Gennip, A. H.; Caron, H. N.; Kemp, S.; van Kuilenburg, A. B.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Histone deacetylases (HDACs): characterization of the classical HDAC family. Biochem. J.
2003, 370, 737ꢀ749.
2.
longevity protein Sir2 is an NADꢀdependent histone deacetylase. Nature 2000, 403, 795ꢀ800.
3. (a) Schemies, J.; Uciechowska, U.; Sippl, W.; Jung, M. NAD(+) ꢀdependent histone
Imai, S.; Armstrong, C. M.; Kaeberlein, M.; Guarente, L. Transcriptional silencing and
deacetylases (sirtuins) as novel therapeutic targets. Med. Res. Rev. 2010, 30, 861ꢀ889; (b)
Schiedel, M.; Robaa, D.; Rumpf, T.; Sippl, W.; Jung, M. The current state of NAD+ ꢀdependent
histone deacetylases (sirtuins) as novel therapeutic targets. Med. Res. Rev. 2017, [Online early
access]. DOI: 10.1002/med.21436. Published Online: January 17, 2017.
4.
Bao, X.; Wang, Y.; Li, X.; Li, X. M.; Liu, Z.; Yang, T.; Wong, C. F.; Zhang, J.; Hao, Q.;
Li, X. D. Identification of 'erasers' for lysine crotonylated histone marks using a chemical
proteomics approach. eLife 2014, 3, e02999.
5.
Tan, M. J.; Peng, C.; Anderson, K. A.; Chhoy, P.; Xie, Z. Y.; Dai, L. Z.; Park, J.; Chen,
Y.; Huang, H.; Zhang, Y.; Ro, J.; Wagner, G. R.; Green, M. F.; Madsen, A. S.; Schmiesing, J.;
Peterson, B. S.; Xu, G. F.; Ilkayeva, O. R.; Muehlbauer, M. J.; Braulke, T.; Muhlhausen, C.;
Backos, D. S.; Olsen, C. A.; McGuire, P. J.; Pletcher, S. D.; Lombard, D. B.; Hirschey, M. D.;
Zhao, Y. M. Lysine glutarylation is a protein posttranslational modification regulated by SIRT5.
Cell Metab. 2014, 19, 605ꢀ617.
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 36
1
2
3
4
6.
Du, J.; Zhou, Y.; Su, X.; Yu, J. J.; Khan, S.; Jiang, H.; Kim, J.; Woo, J.; Kim, J. H.; Choi,
5
6
B. H.; He, B.; Chen, W.; Zhang, S.; Cerione, R. A.; Auwerx, J.; Hao, Q.; Lin, H. Sirt5 is a NADꢀ
7
8
9
dependent protein lysine demalonylase and desuccinylase. Science 2011, 334, 806ꢀ809.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7.
Feldman, J. L.; Baeza, J.; Denu, J. M. Activation of the protein deacetylase SIRT6 by
longꢀchain fatty acids and widespread deacylation by mammalian sirtuins. J. Biol. Chem. 2013,
288, 31350ꢀ31356.
8.
Jiang, H.; Khan, S.; Wang, Y.; Charron, G.; He, B.; Sebastian, C.; Du, J. T.; Kim, R.; Ge,
E.; Mostoslavsky, R.; Hang, H. C.; Hao, Q.; Lin, H. N. SIRT6 regulates TNFꢀalpha secretion
through hydrolysis of longꢀchain fatty acyl lysine. Nature 2013, 496, 110ꢀ113.
9.
(a) Tong, Z.; Wang, Y.; Zhang, X.; Kim, D. D.; Sadhukhan, S.; Hao, Q.; Lin, H. SIRT7 is
activated by DNA and deacetylates histone H3 in the chromatin context. ACS Chem. Biol. 2016,
11, 742ꢀ747; (b) Tong, Z.; Wang, M.; Wang, Y.; Kim, D. D.; Grenier, J. K.; Cao, J.; Sadhukhan,
S.; Hao, Q.; Lin, H. SIRT7 is an RNAꢀactivated protein lysine deacylase. ACS Chem. Biol. 2017,
12, 300ꢀ310.
10.
ortholog, SIRT2, is an NAD+ꢀdependent tubulin deacetylase. Mol. Cell 2003, 11, 437ꢀ444.
11. Vaziri, H.; Dessain, S. K.; Ng Eaton, E.; Imai, S. I.; Frye, R. A.; Pandita, T. K.; Guarente,
North, B. J.; Marshall, B. L.; Borra, M. T.; Denu, J. M.; Verdin, E. The human Sir2
L.; Weinberg, R. A. hSIR2(SIRT1) functions as an NADꢀdependent p53 deacetylase. Cell 2001,
107, 149ꢀ159.
12.
North, B. J.; Rosenberg, M. A.; Jeganathan, K. B.; Hafner, A. V.; Michan, S.; Dai, J.;
Baker, D. J.; Cen, Y.; Wu, L. E.; Sauve, A. A.; van Deursen, J. M.; Rosenzweig, A.; Sinclair, D.
A. SIRT2 induces the checkpoint kinase BubR1 to increase lifespan. EMBO J. 2014, 33, 1438ꢀ
1453.
ACS Paragon Plus Environment
28

Page 29 of 36
Journal of Medicinal Chemistry
1
2
3
4
13.
Yeung, F.; Hoberg, J. E.; Ramsey, C. S.; Keller, M. D.; Jones, D. R.; Frye, R. A.; Mayo,
5
6
M. W. Modulation of NFꢀkappaBꢀdependent transcription and cell survival by the SIRT1
7
8
9
deacetylase. EMBO J. 2004, 23, 2369ꢀ2380.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
14.
mammalian physiology. Curr. Opin. Cell Biol. 2008, 20, 303ꢀ309.
15. Verdin, E.; Hirschey, M. D.; Finley, L. W.; Haigis, M. C. Sirtuin regulation of
Feige, J. N.; Auwerx, J. Transcriptional targets of sirtuins in the coordination of
mitochondria: energy production, apoptosis, and signaling. Trends Biochem. Sci. 2010, 35, 669ꢀ
675.
16.
Liu, T. F.; Vachharajani, V. T.; Yoza, B. K.; McCall, C. E. NAD+ꢀdependent sirtuin 1
and 6 proteins coordinate a switch from glucose to fatty acid oxidation during the acute
inflammatory response. J. Biol. Chem. 2012, 287, 25758ꢀ25769.
17.
and calorie restriction. Genes Dev. 2006, 20, 2913ꢀ2921.
18. de Oliveira, R. M.; Sarkander, J.; Kazantsev, A. G.; Outeiro, T. F. SIRT2 as a therapeutic
target for ageꢀrelated disorders. Front. Pharmacol. 2012, 3, 92ꢀ100.
19. Beirowski, B.; Gustin, J.; Armour, S. M.; Yamamoto, H.; Viader, A.; North, B. J.;
Haigis, M. C.; Guarente, L. P. Mammalian sirtuinsꢀꢀemerging roles in physiology, aging,
Michan, S.; Baloh, R. H.; Golden, J. P.; Schmidt, R. E.; Sinclair, D. A.; Auwerx, J.; Milbrandt, J.
Sirꢀtwoꢀhomolog 2 (Sirt2) modulates peripheral myelination through polarity protein Parꢀ
3/atypical protein kinase C (aPKC) signaling. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, E952ꢀ
961.
20.
(a) Pais, T. F.; Szego, E. M.; Marques, O.; MillerꢀFleming, L.; Antas, P.; Guerreiro, P.;
de Oliveira, R. M.; Kasapoglu, B.; Outeiro, T. F. The NADꢀdependent deacetylase sirtuin 2 is a
suppressor of microglial activation and brain inflammation. EMBO J. 2013, 32, 2603ꢀ2616; (b)
ACS Paragon Plus Environment
29