2-Trifluoroacetylthiophenes, a novel series of potent and selective class II histone deacetylase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2008.02.026

The identification of class II HDAC inhibitors has been hampered by lack of efficient enzyme assays, in the preceding paper two assays have been developed to improve the efficiency of these enzymes: mutating an active site histidine to tyrosine, or by the

Full text of DOI:10.1016/j.bmcl.2008.02.026

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 3456–3461
2-Trifluoroacetylthiophenes, a novel series of potent
and selective class II histone deacetylase inhibitors
*
´
Philip Jones, Matthew J. Bottomley, Andrea Carfı, Ottavia Cecchetti, Federica Ferrigno,
Paola Lo Surdo, Jesus M. Ontoria, Michael Rowley, Rita Scarpelli,
Carsten Schultz-Fademrecht and Christian Steinkuhler
¨
IRBM/Merck Research Laboratories, Via Pontina km 30,600, 00040 Pomezia, Italy
Received 18 November 2007; revised 7 February 2008; accepted 9 February 2008
Available online 14 February 2008
Abstract—The identification of class II HDAC inhibitors has been hampered by lack of efficient enzyme assays, in the preceding
paper two assays have been developed to improve the efficiency of these enzymes: mutating an active site histidine to tyrosine,
or by the use of a trifluoroacetamide lysine substrate, allowing screening to identify class II HDAC inhibitors. Herein, 2-trifluoro-
acetylthiophenes have been demonstrated to inhibit class II HDACs, resulting in the development of a series of 5-(trifluoroace-
tyl)thiophene-2-carboxamides as novel, potent and selective class II HDAC inhibitors. X-ray crystal structures of the HDAC 4
catalytic domain with a bound inhibitor demonstrate these compounds are active site inhibitors and bind in their hydrated form.
Ó 2008 Elsevier Ltd. All rights reserved.
The role of two counteracting enzyme families, the his-
tone acetyl transferases (HATs) and the histone deacet-
ylases (HDACs) in controlling the post-transcriptional
acetylation status of lysine residues on histone tails,
and a number of other proteins, is well-documented.1,2
This equilibrium is a crucial determinant of chromatin
structure, and hence gene transcription. The HDAC
enzyme family has been implicated in a diverse array
of processes, including: neoplasias, skeletal and muscle
formation, cardiac hypertrophy, T-cell differentiation
and neuronal survival.2,3 However, the role of each of
the individual HDAC isoforms or indeed classes in each
of these biological processes is still being established.3,4
Consequently, there is an urgent need for selective small
molecule HDAC inhibitors (HDACi), both as research
tools and ultimately as therapeutic agents.
addition to their catalytic domain, they contain a long
N-terminal regulatory domain,3–5 which has been impli-
cated in gene regulation through protein–protein inter-
actions.3,5,6 The sub-division of the class II HDACs is
the result of the presence of a second known or putative
catalytic domain within class IIb HDACs. Aside from
HDAC6 which has been demonstrated to function as a
tubulin deacetylase,7 there has been an extensive debate
in the literature as to whether other class II HDACs, in
particular class IIa enzymes, possess any intrinsic deace-
tylase activity.5,6 In the preceding paper we have shown
that pure class IIa HDACs do possess weak but measur-
able deacetylase activity,8,9 and this activity can be
enhanced and measured either by an H-Y mutation in
the active site to give a ‘gain of function’ (GOF) enzyme,
or by use of the wild-type (WT) enzyme with an ‘unnat-
ural’ trifluoroacetamide lysine substrate. With this tool-
box a program was initiated to identify selective class II
HDAC inhibitors, targeting in particular HDAC4.
The HDAC enzyme family can be divided into two clas-
ses: class I (HDACs 1, 2, 3 + 8), and class II, which can
be further subdivided into class IIa (HDACs 4, 5, 7 + 9)
and class IIb (HDACs 6 + 10). The class II HDACs
differ from their class I counterparts in the fact that in
A focused screen of both in-house and commercial
derivatives containing known zinc binding groups was
initially carried out on both HDAC4 GOF10 and
4WT.11 Specific interest was given to trifluoromethyl ke-
tone derivatives given that trifluoroacetamide lysine was
a substrate for HDAC4WT, and that alkyl trifluoro-
methyl ketones had already been shown to inhibit both
HDAC4 assays in the tens of nanomolar range.9 These
Keywords: Histone deacetylase; HDAC; Trifluoroacetyl ketones;
Hydrated ketones.
*
Corresponding author. Tel.: +39 0691093559; fax: +39
0691093654; e-mail: philip_jones@merck.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.02.026

P. Jones et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3456–3461
3457
findings suggested that there may be a larger binding
F
R
N
F
F
pocket beneath the zinc in the active site of these class
II enzymes that could be exploited. This focused screen
identified ethyl 5-(trifluoroacetyl)thiophene-2-carboxyl-
ate (1)12 to be a potent inhibitor of class II HDACs
showing IC50 = 350 and 320 nM on HDAC4 GOF and
WT assays, respectively (Fig. 1). This compound also
inhibited HDAC6 with IC50 = 550 nM, whilst showed
modest selectivity over the class I isoforms with
IC50 = 5.7 and 3.5 lM against HDACs 1 and 3. Analysis
of the 1H NMR data of this compound showed that the
trifluoromethyl ketone was readily hydrated with >85%
in the hydrated form, suggesting that this compound
may be bound in the active site as the hydrate rather
than as the ketone. Interestingly, the corresponding fur-
an derivative 2 was 10-fold less potent on HDAC4,
revealing the importance of the thiophene moiety.
R'
S
O
O
3
H
Me
N
H
(R)
N
Ph
NHEt
N
Ph
Ph
a
c
b
d
H
OMe
(S)
N
Ph
H
H
N
N
e
g
f
N
Ph
Me
N
N
N
N
N
N
h
j
i
Given the potency and selectivity of 1, and the presence
of the ester group ready for functionalisation, a number
of amide derivatives were prepared looking to improve
both HDAC4 activity and selectivity (Fig. 2). Based
on published X-ray crystal structure data,13 it was as-
sumed that these compounds would be active site inhib-
itors and that the thiophene moiety would sit in the
narrow binding cleft, whilst the amides would be orien-
tated to the surface on the enzyme where they would be
able to engage in binding interactions with residues
there.
HDAC IC50 (nM)
4GOF
-NRR'
1
3
4WT
1700
6
34%inh at
10 μM
3a
3b
7100
880
1000
180
310
1700
330
210
3c
1800
320
1200
230
170
160
510
76
230
360
3d
3e
3f
5300
210
9400
120
280
70
170
180
210
240
Initial SAR on the amide moiety showed that a wide ar-
ray of derivatives were tolerated, maintaining their
activity on class II HDACs. Simple alkyl groups such
as the ethyl amide 3a lost activity, although this could
be regained by increasing lipophilicity, such as cyclo-
hexyl derivative 3g. Both benzylic and homo-benzylic
derivatives were tolerated, see 3f and 3b, as was alkyl-
ation of the amide, for instance see N-methyl analogue
3c. These tertiary amides could be sterically constrained
without losing activity, as seen with dihydroindolyl
derivative 3h which shows 130-300 nM activity on the
class II HDACs and around 10-fold selectivity over class
I isoforms. Substitution of the benzylic position was also
tolerated in 3d and 3e, but significant differences were
seen between the two enantiomers, with the R-isomer
being a potent and unselective HDAC inhibitor,
whereas the S-enantiomer maintained hundred nanomo-
3g
2900
3100
350
400
460
3h
3i
2000
580
1700
670
300
87
130
98
200
89
44%inh at
5 μM
3j
5500
370
320
310
aValues are means of >2 expts, std. dev. were < 30% of the IC50 values.
Figure 2. Activity of 5-(trifluoroacetyl)thiophene-2-carboxamides 3 on
HDAC isoforms.
lar activity on class II HDACs with more than an order
of magnitude selectivity. Heterocycles, such as the quin-
oxaline 3i, were also accommodated, although with
reduced selectivity for the class II HDACs.
F
1: X = S
2: X = O
F
F
OEt
X
O
O
Unfortunately, the natural substrate of HDAC4 is
unknown and accordingly there is no cellular target
engagement marker for HDAC4. Given that these com-
pounds also inhibit HDAC6, and knowing that a-tubu-
lin is deacetylated by HDAC6,7 it was possible to use
inhibition of tubulin deacetylation as a surrogate mar-
ker of activity. Unfortunately, most of these derivatives
lacked cell based activity and failed to show inhibition
of either histone or tubulin deacetylation in HCT116
cells. However, 3i showed inhibition of tubulin deacety-
HDAC IC50 (nM)
4GOF
Compds
1
3
4WT
320
6
1
5700
3500
350
550
48%inh at
10 μM
NA at
10 μM
2
3100
2500
550
aValues are means of >2 expts, std. dev. were < 30% of the IC50 values.
Figure 1. Activity of lead 1 on HDAC isoforms.

3458
P. Jones et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3456–3461
lation with IC50 = 12 lM, whereas only 40% inhibition
of histone H3 deacetylation was seen at 50 lM, thereby
demonstrating that these compounds are selective class
II HDACis in cells.
Replacement with a weaker zinc binding group, carbox-
ylic acid 7i abolished activity on all HDAC isoforms. In
a similar manner, no inhibition of any of the HDAC iso-
forms was seen with the methyl ketone 10d. To explore the
importance of the trifluoromethyl group, the correspond-
ing difluoromethyl 15h and pentafluoroethyl 11f deriva-
tives were synthesized. The former by lithiation of the
thiophene 13h with LDA, followed by quenching with
ethyl difluoroacetate, whereas the latter was obtained by
addition of the lithium pentafluoroethyl onto the Weinreb
amide 9f. Both compounds showed substantial losses of
activity on all HDAC isoforms. In particular, difluoroke-
tone 15h while potently inhibiting HDAC4GOF lost
more than 10-fold activity on HDAC4WT and therefore
was not pursued further. Finally, the boronic acid 16h,
methyl sulfone 17h, and phosphonic acid 14h derivatives
were all inactive.
Having shown that potent class II HDACis could be
developed, a more in depth understanding of the deter-
minants of activity was undertaken, focusing on three
regions of the inhibitors: the trifluoromethyl ketone zinc
binding group, the central thiophene ring and the amide
moiety.
The zinc binding group was explored as outlined in
Scheme 1 and Table 1. Reduction of the trifluoromethyl
ketone to alcohol 4c with NaBH4 resulted in a complete
loss of activity. As did protection of the ketone as ketal
5h, the formation of which required cyclization with chlo-
roethanol under basic conditions to form the dioxolane.
In contrast, the corresponding hydroxamic acid 8d, read-
ily prepared from thiophene-2,5-dicarboxylic acid (6)
proved to be a potent but unselective HDACi, showing
nanomolar activity against both class I and II HDACs.
To enable SAR studies to look for alternatives to the 5-
carboxamide group, chemistry had to be established to
introduce the trifluoroacetyl group onto the thiophene
as the ultimate synthetic step. Therefore simple func-
tional group transformations were conducted on the
5-position of the thiophene and then the 2-position
was deprotonated with LDA at ꢀ78 °C and the lithio
derivative quenched with 2,2,2-trifluoro-N-methoxy-N-
methylacetamide (Scheme 2). In this manner amines
19c + d, ether 21f and sulfonamides 25c + g were inves-
tigated. The 4-position of the thiophene was also
explored by sulfonylation with chlorosulfonic acid and
PCl5, followed by preparation of the sulfonamide 23c.
These alternative capping groups were all tolerated to
some extent on the class II HDACs (Table 2), although
a loss in HDAC4 activity was seen with all derivatives
compared to the 5-carboxamides. For instance the ether
21f resulted in a more than 30-fold loss in activity on
HDAC4, although maintained HDAC6 activity. While
whereas tertiary amine 19c lost 2- to 6-fold activity on
HDAC4, the secondary amines 19d lost more than 20-
fold activity. The isomeric sulfonamides 23c and 25c
demonstrated the importance of the substitution pattern
on the thiophene, as whilst the 5-substituted isomer 25c
remained a hundred nanomolar HDAC4 inhibitor, the
region isomer 23c lost almost all deacetylase activity.
F
Me
N
F
F
NaBH4
3c
1
S
OH
O
O
4c
F
F
F
1) K2CO3
ClCH2CH2OH
N
S
O
O
2) LiOH
3) EDC, HOBT
RR'NH
5h
1) MeOH, HBTU
2) NH2OH, NaOMe
R
N
R
N
H
HO
HO
N
HO
R'
R'
R'
S
7
S
8
O
O
O
O
O
1) SOCl2
2) Morpholine, Et3N
HBTU,
RNH2
3) MeMgBr, 0 oC to RT
R
N
OH
Me
F3CF2C
F
S
S
O
O
O
O
O
6
10
1) HBTU, RNH2
2) HBTU, NH(Me)OMe
CF3CF2I
MeLi.LiBr
-78 to 0 o
R
N
Me
N
NHR
C
MeO
R'
S
S
Having established that the trifluoroacetyl group could
be introduced as the ultimate synthetic step, this meth-
odology was applied in the investigation of thiophene
replacement (Scheme 3). In this manner, the 3-trifluoro-
acetylthiophene derivative 27c was prepared from the
3-bromothiophene-5-carboxylic acid (26) by Br–Li
exchange and quenching as previously. Similar chemis-
try, but using Mg–Br exchange, was used for the pyri-
dine analogue 37. The corresponding thiazoles 29c + g
could be prepared by direct lithiation of the thiazole
with n-BuLi and quenching. Preparation of the corre-
sponding pyrrole derivative 31h necessitated sequential
Br–Li exchanges on N-Boc 2,5-dibromopyrrole (30),
firstly to allow introduction of the carboxylate group,
and secondly the trifluoroacetyl group. In contrast, the
analogous phenyl derivatives 34g and 35g were more
straightforward, being readily prepared readily from
carboxylic acids 32 and 33.
O
O
9
11
OH
1) LDA, -78 o
C
S
O
2) CF2HCO2Et
12
-78 o
C
N
HATU
F
RR'NH
S
O
O
O
15h
1) LDA, -78 o
C
N
2) B(OMe)3, -78 o
3) H3O+
C
OH
B
S
N
N
O
HO
13h
S
1) LDA, -78 o
2) ClPO(OEt)2
-78 oC to RT
3) TMSBr
C
O
1) LDA, -78 o
2) ClSO2Me,
C
16h
-78oC to RT
HO
HO
N
S
S
P
S
14h
O
O
O
O
O
17h
Scheme 1. SAR exploration of the zinc-binding group.

P. Jones et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3456–3461
3459
Table 1. Activity of 2-substituted thiophene-5-carboxamides with alternative zinc-binding groups
–NRR0
HDAC IC50 (nM)a
1
3
4GOF
4WT
6
4c
5h
NA at 10 lM
NA at 10 lM
NA at 10 lM
6
NA at 10 lM
NA at 10 lM
NA at 10 lM
10
NA at 10 lM
NA at 10 lM
NA at 10 lM
100
NA at 10 lM
NA at 10 lM
NA at 10 lM
2700
NA at 10 lM
NA at 10 lM
NA at 10 lM
94% inh. at 9 nM
6900
7i
8d
10d
11f
14h
15h
16h
17h
NA at 10 lM
4200
NA at 10 lM
2900
NA at 10 lM
NA at 10 lM
NA at 10 lM
420
NA at 10 lM
NA at 10 lM
NA at 10 lM
7600
NA at 10 lM
NA at 10 lM
15
NA at 10 lM
9700
NA at 10 lM
42% inh. at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
42% inh. at 10 lM
47% inh. at 10 lM
a Values are means of >2 experiments, SD were <30% of the IC50 values.
F
F
R
N
O
F
F
1) NaBH3(CN), RR'NH
F
F
Br
F
1) BuLi, -78 o
C
2) LDA, -78oC
3) CF3CON(Me)OMe
-78oC to RT
R'
R
R
N
S
18
S
19
OH
OH
O
O
O
2) CF3CON(Me)OMe
-78 oC to RT
3) EDC, HOBT, RR'NH
R'
S
S
O
O
O
1) NaH, ROH
26
27
F
F
O
1) BuLi, -78 o
C
2) LDA, -78oC
3) CF3CON(Me)OMe
-78oC to RT
S
21
O
F
F
R
N
S
20
N
S
F
F
Br
N
2) CF3CON(Me)OMe
-78oC to RT
3) EDC, HOBT, RR'NH
R'
S
O
O
O
O
S
R'
1) ClSO3H, PCl5, 55 oC
2) RR'NH, Et3N
F
F
N
F
28
29
1) BuLi, -78 o
C
F
F
F
F
R
2) CO2, -78 o
C
S
23
S
22
3) TBTU, RR'NH
R
N
F
F
O
O
O
4) LDA
Br
N
Br
R'
N
H
5) CF3CON(Me)OMe
-78 oC to RT
6) TFA
R
N
Boc
1) RR'NH, Et3N
2) LDA, -78oC
F
F
O
Cl
30
31
S
S
R'
S
S
O
O
O O 3) CF3CON(Me)OMe
O O
R'
24
25
-78oC to RT
EDC, HOBT
RR'NH
N
OH
F
F
F
F
F
F
F
R
Scheme 2. SAR exploration of capping group.
32: meta
33: para
34: meta
35: para
O
O
O
O
These thiophene replacements were evaluated on the
panel of HDAC isoforms as previously and data are
shown in Table 3. The isomeric 3-(trifluoroacetyl)thio-
phene 27c although maintaining some activity on the
class II HDACs loses around 3-fold in activity on
HDAC4WT and 15-fold on the GOF. This could be
due to either an incorrect positioning on the substituent
on the thiophene core or a weaker propensity to hy-
drate. Similarly, the thiazoles analogues 29c + g,
although displaying sub-micromolar activity on
HDAC4WT lose around 10-fold potency on the H-Y
mutated enzyme compared to the corresponding thio-
phene. In contrast, the pyrrole derivative 31h maintains
R
1) HATU, RR'NH
OH
N
N
R
F
F
N
2) iPrMgCl, RT
3) CF3CON(Me)OMe
Br
O
36
Scheme 3. SAR exploration of capping group.
37
only activity on HDAC6 with an IC50 = 880 nM. More
radical substitutions with replacement of the thiophene
group either by phenyl or pyridyl groups are detrimen-
tal, as only very weak activity, if any, is seen on any
HDAC isoform with 34g, 35g and 37.
Table 2. Activity of 5-substituted 2-(trifluoroacetyl)thiophene derivatives
–NRR0
HDAC IC50 (nM)a
1
3
4GOF
4WT
6
19c
19d
21f
23c
25c
3100
4600
1100
3500
970
490
840
410
2500
730
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
3700
5500
9400
340
2700
NA at 10 lM
1200
a Values are means of >2 experiments, SD were <30% of the IC50 values.

3460
P. Jones et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3456–3461
Table 3. Activity of (trifluoroacetyl)heteroaryl carboxamides
–NRR0
HDAC IC50 (nM)a
4GOF
1
3
4WT
6
27c
27g
29c
29g
31h
34g
35g
37
NA at 10 lM
40% inh. at 10 lM
46% inh. at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
7300
2500
2400
1700
3300
640
2000
1300
1500
880
2200
2900
760
850
7600
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
NA at 10 lM
31% inh. at 10 lM
NA at 10 lM
47% inh. at 10 lM
NA at 10 lM
32% inh. at 10 lM
NA at 10 lM
4200
2500
4200
a Values are means of >2 experiments, SD were <30% of the IC50 values.
Having demonstrated the importance of all three struc-
tural elements: the trifluoroacetyl group, the 2,5-disub-
stituted thiophene, and the carboxamide moiety, it was
crucial to obtain structural information of one of these
derivatives bound to a class II deacetylase in order to
further improve the activity and selectivity of these
HDACs. Although several X-ray crystal structures exist
of several class I HDACs there were no information
available on their class II counterparts. X-ray crystal
structures were determined for both HDAC 4WT and
4GOF catalytic domains (Thr648–Thr1057), expressed
and purified from Escherichia coli, with 3j which was a
modest inhibitor of HDAC4 (GOF/WT IC50s = 370/
320 nM).14 Both structures, Figures 3 and 4, respec-
tively, clearly show that these trifluoroacetyl thiophene
derivative are active site binders with the hydrated tri-
fluoromethyl ketone chelating the active site zinc in a
bidentate manner. These oxygen-zinc bonds are in the
˚
range 2.04–2.4 A. The requirement of the trifluoro-
methyl group is also explained by the crystal structures,
as this group occupies a modest pocket at the bottom of
the active site which is conserved in class II HDACs.
The trifluoromethyl group fills this pocket entirely, com-
Figure 4. X-ray crystal structure of 3j bound to HDAC4 H976Y ‘Gain
of Function’ catalytic domain.14
ing within van der Waals contact distance of a proline-
800 at the bottom of the cavity. The modest size of this
pocket thereby explains the loss in activity of the more
sterically demanding pentafluoroethyl group 11f.
Interestingly, in both structures the H976 and the Y976
groups, the analogous group of which in class I enzymes
make H bonds to the substrate carbonyl, are orientated
away from the inhibitor. This may account for the low
activity of HDAC4 to canonical acetylated lysines.
However, in the HDAC4 structures an active site water
molecule bridges between the inhibitor and active site
Gly-975 and this water could be responsible for stabil-
ization of the transition state during the deacetylation
reaction.
In summary, a novel series of 5-(trifluoroacetyl)thio-
phene-2-carboxamides has been developed as potent
and selective class II HDAC inhibitors. The compounds
show around 10-fold selectivity for HDACs 4 + 6 over
their class I counterparts, HDACs 1 + 3. In cells, selec-
tive inhibition of tubulin deacetylation is observed, dem-
onstrating their selectivity of HDAC6 over the class I
Figure 3. X-ray crystal structure of 3j bound to HDAC4 WT catalytic
domain.14

P. Jones et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3456–3461
3461
9. Jones, P.; Altamura, S.; De Francesco, R.; Gallinari, P.;
Lahm, A.; Neddermann, P.; Rowley, M.; Serafini, S.;
isoforms. X-ray crystal structures of both HDAC 4GOF
and 4WT catalytic domain have been obtained with a
bound inhibitor, demonstrating that these compounds
are active site inhibitors and bind in their hydrated
form.
Steinkuhler, C. Bioorg. Med. Chem. Lett. 2008, 18, 1814.
¨
10. DMSO/compound solution were incubated for 10 min
with His-tagged HDAC4GOF (653–1084, H976Y) from
E. coli in assay buffer (20 mM Hepes, pH 7.5, 137 mM
NaCl, 2.7 mM KCl, 1 mM MgCl2, 0.1 mg/ml BSA), Fluor-
de-Lys substrate solution was added and left for 1 h at
37 °C and the reaction stopped by adding developer/TSA
solution. Measure the fluorescence at ex. 360 nM/em.
460 nM.
11. DMSO/compound solution were incubated for 10 min
with His-tagged HDAC4 CD(653–1084) from E. coli in
assay buffer (25 mM Tris/HCl pH8, 137 mM NaCl,
2.7 mM KCl, 1 mM MgCl2, 0.1 mg/ml BSA), trifluoace-
tamide substrate solution was added and left for 1 h at
37 °C and the reaction stopped by adding developer/TSA
solution. Measure the fluorescence at ex. 360 nM/em.
460 nM.
References and notes
1. Cheung, W. L.; Briggs, S. D.; Allis, C. D. Curr. Opin. Cell
Biol. 2000, 12, 326.
2. Minucci, S.; Pelicci, P. G. Nat. Rev. Cancer 2006, 6, 38.
3. Martin, M.; Kettmann, R.; Dequiedt, F. Oncogene 2007,
26, 5450.
4. De Ruijter, A. J. M.; van Gennip, A. H.; Caron, H. N.; Kemp,
S.; van Kuilenburg, A. B. P. Biochem. J. 2003, 370, 737.
5. Verdin, E.; Dequiedt, F.; Kasler, H. G. Trends Genet.
2003, 19, 286.
6. Yang, X. J.; Gregoire, S. Mol. Cell. Biol. 2005, 25, 2873.
7. Hubbert, C.; Guardiola, A.; Shao, R.; Kawaguchi, Y.; Ito,
A.; Nixon, A.; Yoshida, M.; Wang, X. F.; Yao, T. P.
Nature 2002, 417, 455.
12. Compound 1 was purchased from Rieke Metals, Lincoln,
USA.
13. Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.;
Rifkind, R. A.; Marks, P. A.; Breslow, R.; Pavletich, N. P.
Nature 1999, 401, 188.
8. Lahm, A.; Paolini, C.; Pallaoro, M.; Nardi, M. C.; Jones,
P.; Neddermann, P.; Sambucini, S.; Bottomley, M. J.; Lo
14. Bottomley, M.J.; Lo Surdo, P.; Di Giovine, P.; Cirillo, A.;
Scarpelli, R.; Ferrigno, F.; Jones, P.; Neddermann, P.; De
´
Surdo, P.; Carfı, A.; Koch, U.; De Francesco, R.;
Steinkuhler, C.; Gallinari, P. Proc. Natl. Acad. Sci.
¨
U.S.A. 2007, 104, 17335.
´
Francesco, R.; Steinkuhler, C.; Gallinari, P.; Carfı, A.
¨
Structure, submitted for publication.