3-(4-Aroyl-1-methyl-1H-pyrrol-2-yl)-N-hydroxy-2-propenamides as a New Class of Synthetic Histone Deacetylase Inhibitors. 3. Discovery of Novel Lead Compounds through Structure-Based Drug Design and Docking Studies

Article abstract of DOI:10.1021/jm031036f

Aroyl-pyrrole-hydroxy-amides (APHAs) are a new class of synthetic HDAC inhibitors recently described by us. Through three different docking procedures we designed, synthesized, and tested two new isomers of APHA lead compound 3-(4-benzoyl-1-methyl-1H-pyrr

Full text of DOI:10.1021/jm031036f

J . Med. Chem. 2004, 47, 1351-1359
1351
3-(4-Ar oyl-1-m eth yl-1H-p yr r ol-2-yl)-N-h yd r oxy-2-p r op en a m id es a s a New Cla ss of
Syn th etic Histon e Dea cetyla se In h ibitor s. 3. Discover y of Novel Lea d
Com p ou n d s th r ou gh Str u ctu r e-Ba sed Dr u g Design a n d Dock in g Stu d ies†,∆
Rino Ragno,*,| Antonello Mai,*,§ Silvio Massa, Ilaria Cerbara,§ Sergio Valente,§ Patrizia Bottoni,X
Roberto Scatena,X Florian J esacher,‡ Peter Loidl,‡ and Gerald Brosch*,‡
Dipartimento di Studi di Chimica e Tecnologia delle Sostanze Biologicamente Attive, Universita` degli Studi di Roma
“La Sapienza”, P. le A. Moro 5, 00185 Roma, Italy, Dipartimento di Studi Farmaceutici, Universita` degli Studi di Roma
“La Sapienza”, P. le A. Moro 5, 00185 Roma, Italy, Dipartimento Farmaco Chimico Tecnologico, Universita` degli Studi di
Siena, via A. Moro, 53100 Siena, Italy, Istituto di Biochimica e Biochimica Clinica, Universita` Cattolica del Sacro Cuore,
L.go F. Vito 1, 00168 Roma, Italy, and Department of Molecular Biology, University of Innsbruck, Medical School,
Peter-Mayr-Strasse 4b, 6020 Innsbruck, Austria
Received September 18, 2003
Aroyl-pyrrole-hydroxy-amides (APHAs) are a new class of synthetic HDAC inhibitors recently
described by us. Through three different docking procedures we designed, synthesized, and
tested two new isomers of APHA lead compound 3-(4-benzoyl-1-methyl-1H-pyrrol-2-yl)-N-
hydroxy-2-propenamide (1), compounds 3 and 4, characterized by different insertions of benzoyl
and propenoylhydroxamate groups onto the pyrrole ring. Biological activities of 3 and 4 were
predicted by computational tools up to 617-fold more potent than that of 1 against HDAC1;
thus, 3 and 4 were synthesized and tested against both mouse HDAC1 and maize HD2 enzymes.
Predictions of biological affinities (Ki values) of 3 and 4, performed by a VALIDATE model
(applied on either SAD or automatic DOCK or Autodock results) and by the Autodock internal
scoring function, were in good agreement with experimental activities. Ligand/receptor positive
interactions made by 3 and 4 into the catalytic pocket, in addition to those showed by 1, could
at least in part account for their higher HDAC1 inhibitory activities. In particular, in mouse
HDAC1 inhibitory assay 3 and 4 were 19- and 6-times more potent than 1, respectively, and
3 and 4 antimaize HD2 activities were 16- and 76-times higher than that of 1, 4 being as
potent as SAHA in this assay. Compound 4, tested as antiproliferative and cytodifferentiating
agent on MEL cells, showed dose-dependent growth inhibition and hemoglobin accumulation
effects.
In tr od u ction
ity.1-3 Gene transcription or repression is associated
with the ability of transcriptionally competent genes to
recruit either HAT or HDAC proteins to the promoter.
HAT-promoted acetylation activity is related with nu-
cleosomal relaxation and gene transcription, while
HDAC-promoted deacetylation produces a tightness of
nucleosomal integrity and gene silencing.4-6
Histone deacetylase (HDAC) and histone acetyltrans-
ferase (HAT) enzymes are involved in determining the
acetylation status of histones, and such reversible
acetylation reactions play an important role in modula-
tion of chromatin topology and regulation of gene
expression. Histones H2A, H2B, H3, and H4 exhibit
acetyl groups at the ꢀ-amino-terminal lysine residues
within the tails extending from the histone octamer of
the nucleosome core. Among them, histones H3 and H4
constitute the main targets of HDAC enzymatic activ-
Aberrant acetylation of histone tails emerging either
from HAT mutations or abnormal recruitment and/or
overexpression of HDACs have been clearly linked to
carcinogenesis. Inappropriate HDAC-mediated tran-
scriptional repression is a common molecular mecha-
nism used by oncoproteins to produce alterations in
chromatin structure and blockage of normal cell dif-
ferentiation. Compounds able to inhibit HDAC activity
(HDAC inhibitors, such as trichostatin A (TSA),7
trapoxin,8 HC-toxin,9 apicidin,10 suberoylanilide hydrox-
amic acid (SAHA),11 cyclic hydroxamic acid-containing
peptides (CHAPs),12,13 FK-228,14 MS-275,15,16 sodium
valproate17,18) can reactivate gene expression, and they
are potent inducers of growth arrest, differentiation, or
apoptotic cell death in a variety of transformed cells in
culture and in tumor-bearing animals.19-21
* To whom correspondence should be addressed. A.M., Tel.: +396-
4991-3392; Fax: +396491491; e-mail: antonello.mai@uniroma1.it.
(Molecular modeling): R.R., Tel.: +396-4991-3152. Fax: +396491491;
e-mail: rino.ragno@uniroma1.it. (Biology): G.B., Tel.: 0512-507-3608;
Fax: 0512-507-2866; e-mail: gerald.brosch@uibk.ac.at.
† Part 2: Mai, A.; Massa, S.; Cerbara, I.; Valente, S.; Ragno, R.;
Bottoni, P.; Scatena, R.; Loidl, P.; Brosch, G. 3-(4-Aroyl-1-methyl-1H-
2-pyrrolyl)-N-hydroxy-2-propenamides as a New Class of Synthetic
Histone Deacetylase Inhibitors. 2. Effect of Pyrrole C2 and/or C4
Substitutions on Biological Activity. J . Med. Chem., in press.
∆ Dedicated to Prof. Marino Artico on the occasion of his 70th
anniversary.
| Dipartimento di Studi di Chimica e Tecnologia delle Sostanze
Biologicamente Attive, Universita` degli Studi di Roma “La Sapienza”.
§ Dipartimento di Studi Farmaceutici, Universita` degli Studi di
Roma “La Sapienza”.
In 1999 the X-ray crystal structure of the catalytic
core of a bacterial HDAC homologue (histone deacety-
lase-like protein, HDLP)22 revealed the binding mode
Universita` degli Studi di Siena.
X Universita` Cattolica del Sacro Cuore, Roma.
‡ University of Innsbruck.
10.1021/jm031036f CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/10/2004

1352 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6
Ragno et al.
Ta ble 1. VALIDATE and Autodock Predicted Anti-HDAC1 pKi
Compared with Experimental pIC50 Values of APHA Isomers 1,
3, 4, and TSA, and SAHA
predicted pKi
experimental
VALIDATE
pIC50
compd
SAD
5.87
DOCK Autodock Autodock HDAC1 HD2
3
4
1
7.31
7.76
5.29
6.64
6.11
6.60
7.30
7.13
7.55
7.47
5.73
5.7627
4.7628
7.3128
7.9228
7.1528
5.6528
7.4128
6.2128
5.3127 5.4227
8.7027 8.1527
6.9527 7.3027
TSA
8.61a,27 7.2628
SAHA 6.69a,27 6.7928
a
Predicted from X-ray experimental data.22
Ta ble 2. Minimized Complex Steric Energies (kJ /mol)
(AMBER force field) Relative to APHA Isomers 3 and 4
Obtained by DOCK, Autodock, and SAD Procedures into the
Modeled HDAC1. For Direct Comparison the Same Values for
1, TSA, and SAHA Are Reported
F igu r e 1. Chemical Structures of TSA, SAHA, and APHA
compounds.
docking
method
3
4
1a
-7754.958 -7710.129 -7174.81 -8909.93 -7369.18
TSAa
SAHAa
of TSA and SAHA (Figure 1) into the enzyme catalytic
pocket. Prompted by these data, we performed three-
dimensional (3D) structure-based design and molecular
modeling studies on pyrrole-containing TSA analogues
previously synthesized by us as antimicrobial agents.23,24
Since the predicted pKi values of such compounds
complexed with HDLP were in the low (or sub-) micro-
molar range,25 we tested the pyrrole derivatives against
maize histone deacetylase HD2,26 obtaining IC50 values
at low micromolar concentrations (IC50 values ) 1.9-
4.0 µM).25 A refinement of binding mode of the aroyl-
pyrrolyl-hydroxy-amide (APHA) lead compound 1 (Fig-
ure 1) in the deacetylase core was obtained by using a
virtual HDAC1 catalytic pocket modeled from HDLP
coordinates.27 Chemical manipulations28,29 performed on
various portions of the 1 skeleton yielded a new mol-
ecule, N-hydroxy-3-(1-methyl-4-phenylacetyl-1H-pyrrol-
2-yl)-2-propenamide 2 (Figure 1), which was 38-fold
more potent than 1 in inhibiting HD2 in vitro.28 Such
improvement of biological activity was partially clarified
by inspection of the 2 binding mode into the modeled
HDAC1 coordinates. The higher flexibility of the pyrrole
C4-phenylacetyl moiety of 2 respect to that of the C4-
benzoyl substituent of 1, joined with a more favorable
repositioning of the pyrrole N-methyl group into the
deacetylase pocket, can almost in part account for the
enhancement of inhibiting activity observed from 1 to
2.28
DOCK
Autodock -8737.534 -10161.400 -8913.90 -9976.13 -10363.13
SAD
-8581.698 -8457.775 -7828.92
NDb
NDb
a
b
Reference 28. ND, not determined.
2-propenamide moieties allowed us to design two iso-
mers of 1, 3-(2-benzoyl-1-methyl-1H-pyrrol-5-yl)-N-
hydroxy-2-propenamide 3 and 3-(2-benzoyl-1-methyl-
1H-pyrrol-4-yl)-N-hydroxy-2-propenamide 4 (Figure 1).
Extensive binding mode analyses were undertaken to
see if the HDAC1 enzyme inhibitory potency would be
influenced by the above different rearrangement of
benzoyl and N-hydroxypropenamide substituents on the
pyrrole ring.
Initially, binding mode analyses of 3 and 4 into the
modeled HDAC1 structure were investigated using a
semiautomatic dock (SAD) procedure, analogous to that
reported for 1.27 In parallel experiments, docking studies
were performed on 3 and 4 structures by the mean of
the DOCK 4.0.230 and Autodock 3.0.531 programs. The
results of the automatic docking procedures (DOCK and
Autodock) were energy rescored by a single point
minimization by the MACROMODEL molecular me-
chanics program,32 using the AMBER all atom force
fields33,34 and point charges calculated with the MOZYME
routine as implemented in the MOPAC2000 pro-
gram35,36 (see Experimental Section).
In parallel with such studies, we applied a different
approach on the 1 structure to discover novel leads,
using structure-based drug design and docking proce-
dures. Particularly, we designed two isomers of 1, 3-(2-
benzoyl-1-methyl-1H-pyrrol-5-yl)-N-hydroxy-2-propena-
mide 3 and 3-(2-benzoyl-1-methyl-1H-pyrrol-4-yl)-N-
hydroxy-2-propenamide 4 (Figure 1), and we performed
computational studies to predict their HDAC inhibiting
activity. As the predicted pKi values were in the sub-
micromolar range, we synthesized 3 and 4, and tested
them against both maize HD2 and mouse HDAC1
enzymes. Moreover, antiproliferative and cytodifferen-
tiating activities of 4 on Friend murine erythroleukemia
(MEL) cells were also determined.
From docking studies three bound conformations for
both 3 and 4 were obtained, and their associated HDAC
inhibitory potencies, expressed as pKi values, were
predicted using our in house refined VALIDATE scoring
function model.37 In Table 1 the pKi prediction values
for APHA isomers 3 and 4 compared with their experi-
mental pIC50 values (see below) are reported. For
comparison purpose, 1, TSA, and SAHA pKi prediction
values are also shown.
Among the minimized ligand/HDAC1 complexes, the
most stable ones were those proposed by Autodock
(Table 2).
Similarly as previously observed,28 DOCK and Auto-
dock results were in good agreement each other and
with those obtained by the SAD procedure (Table 3).
Dock in g Stu d ies P er for m ed on 3 a n d 4 in to th e
Mod eled HDAC1 Ca ta lytic Cor e. P r ed iction of
HDAC1 In h ibitin g Activity. In the chemical structure
of the APHA lead compound 1, virtual rotation of
N-methylpyrrole ring between benzoyl and N-hydroxy-
To compare the docked molecules (1, 3, 4, TSA, and
SAHA), their binding conformations obtained by Auto-
dock are depicted in Figure 2.

Synthetic Histone Deacetylase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6 1353
F igu r e 2. Stereoview of superimposition of the Autodock docked conformations of 1 (magenta), 3 (green), 4 (red), TSA (yellow),
and SAHA (cyan). The gray ball represents the mean position of the zinc ion. HDAC1 pocket and nonessential hydrogen atoms
are not shown for the sake of clarity.
Sch em e 1a
Ta ble 3. Comparison of Docking Studies Performed on 3 and
4. Values Are the Root Mean Square Deviations (RMSDs) for
All Non-H Atoms. For Direct Comparison, the Same Values for
1, TSA, and SAHA Are Reported
SAD or Exp
vs DOCK
SAD or Exp
vs Autodock
DOCK
vs Autodock
compd
3
4
1
3.2
2.2
5.6
1.0
1.5
3.1
2.8
3.8
1.4
2.3
1.4
1.5
5.1
1.9
2.0
TSA
SAHA
In all cases except for the Autodock internal scoring
function predicted Ki value of 3, binding affinities of 3
and 4 were predicted from 1.3- to 354.8- (3) and from
1.20- to 616.6-times (4) higher than those of 1 against
HDAC1 (Table 1). From these encouraging data the
synthesis of 3 and 4 has been performed, and their in
vitro anti-HDAC (antimouse HDAC1 and antimaize
HD2) activities have been assessed.
Ch em istr y. Vilsmeier-Haack reaction performed on
2-benzoyl-1-methyl-1H-pyrrole38 afforded 2-benzoyl-1-
methyl-1H-pyrrole-5-carboxaldehyde 5 and its corre-
sponding 4-isomer 639 in high yields. Pyrrolecarboxal-
dehydes 5,6 were treated with triethyl phosphonoacetate
in the presence of potassium carbonate under Wittig-
Horner conditions to afford the related ethyl pyrrole-5-
propenoate 7 and -4-propenoate 8, which were in turn
converted into the corresponding pyrrolepropenoic acids
9 and 10 by alkaline hydrolysis. Finally, reaction of 9,10
with ethyl chloroformate and hydroxylamine furnished
the desired hydroxamates 3,4 (Scheme 1).
a
Reagents and conditions: (a) (COCl)2, DMF, dichloroethane,
0 °C; (b) (C2H5O)2OPCH2COOC2H5, K2CO3, C2H5OH, 80 °C;
(c) KOH, C2H5OH, H2O, 70 °C; (d) (1) ClCOOC2H5, (C2H5)3N, THF,
0 °C; (2) NH2OH, rt.
Ta ble 4. Antimaize HD2 and Antimouse HDAC1 Activities of
Compounds 3 and 4a
maize HD2
mouse HDAC1,
IC50 ( SD (µM)
compd
% inhbtnb
IC50 ( SD (µM)
3
4
95
97
86
0.28 ( 0.014
0.05 ( 0.003
3.8 ( 0.1
0.0072 ( 0.0003
0.05 ( 0.002
0.25 ( 0.01
0.78 ( 0.04
4.9 ( 0.1
0.002 ( 0.00006
0.112 ( 0.004
127
TSA27
SAHA27
a
Data represent mean values of at least three separate experi-
b
ments. Percent of inhibition at 30 µM.
Resu lts a n d Discu ssion
In sp ection of 3 a n d 4 Bin d in g Mod es. Com p a r i-
son betw een P r ed icted a n d Exp er im en ta l HDAC1
In h ibitor y Activities. As the newly designed deriva-
tives 3 and 4 were predicted by computational tools up
to 617-fold more potent than 1 against HDAC1, they
were synthesized and tested against both mouse HDAC1
and maize HD2 enzymes. Maize HD2, which was
characterized in detail by us,25,40-44 has an in vitro
enzyme activity comparable to those of HDACs from
other sources, such as fungi and vertebrates, using our
standard HDAC assay.40,44 Moreover, maize HD2 has
shown a good mammalian deacetylase predictive value
in various series of HDAC inhibitors.25,27-29,45,46 TSA
and SAHA were tested together with the pyrrole com-
pounds as reference drugs.
The results, expressed as percent of enzyme inhibition
at fixed dose and IC50 (50% inhibitory concentration)
values, are reported in Table 4.
In anti-HD2 assay, 3 and 4 were 16- and 76-times
more effective in inhibiting the enzyme than 1, 4 being
as potent as SAHA and 7-times less active than TSA.
As anti-HDAC1 agents, 3 and 4 were less potent than

1354 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6
Ragno et al.
F igu r e 3. Superimposition of the docked conformations of APHA isomers 3 (right) and 4 (left) obtained from the three docking
procedures (see text). The semiautomatic docked (SAD) conformations are in green. Autodock conformations are in red. DOCK
conformations are in blue. The gray sphere indicate the Zn ion position into the HDAC1. The same 3D orientation is kept for
direct comparison.
F igu r e 4. Autodock binding mode of APHA isomers 3 (bottom) and 4 (top). Inhibitors are drawn in ball-and-stick. For the sake
of clarity, only a 3 Å core of the HDAC1 with essential hydrogen atoms is displayed. Residue numbering is reported for description
without correction relative to the real HDAC1 sequence.
TSA and SAHA, but were 19- and 6-times more active
Ki values of 3 and 4 were predicted by VALIDATE
than 1, respectively.
and Autodock means from 1.3- to 354.8- and from 1.20-

Synthetic Histone Deacetylase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6 1355
Ta ble 5. Calculated Steric Fit Values37 and Hydroxamate/Zn
Distances of 1, 3, and 4
compd
docking method
steric fit
CO‚‚‚Zn
OH‚‚‚Zn
1
SAD
1.3
2.3
1.5
1.46
1.94
1.94
1.85
1.85
1.85
4.5
2.8
4.7
3.7
2.8
3.3
2.9
2.8
2.9
4.8
4.4
4.8
2.2
3.0
3.0
3.0
3.5
3.0
Autodock
DOCK
SAD
Autodock
DOCK
SAD
3
4
Autodock
DOCK
to 616.6- times higher than those of 1 against HDAC1,
respectively (Table 1). Such predictions of activity have
been fully confirmed by experimental data.
By comparing the docking studies performed on 3 and
4, it is possible to observe that docking studies highly
agree each other, displaying clustered binding confor-
mations for either isomer 3 or 4 (Figure 3).
These data agree with our previous findings28 that
low active compounds (i.e., isomer 1) bind the deacety-
lase core in a lousy way and probably with a not fully
defined binding conformation. On the other hand, in this
study we found that the pyrrole ring rotation between
benzoyl and N-hydroxy-2-propenamide moieties criti-
cally affects the binding mode of APHA isomers. In fact
(Figure 3), while the N-hydroxy-2-propenamide tail
maintains a similar conformation for either isomer 3
or 4, the N-methylpyrrole and benzoyl moieties are
clearly disposed in different ways (compare left side with
right side of Figure 3). Autodock binding conformations
of 3 (bottom) and 4 (top) are depicted in Figure 4. Either
isomer 3 or 4 fits into the HDAC1 binding cavity
considerably better then 1 (compare steric fit values in
Table 5), thus providing a closer positioning of either 3
or 4 hydroxamate moiety to the zinc ion (1: average
CO‚‚‚Zn ) 4.0 Å, average OH‚‚‚Zn ) 4.7 Å; 3: average
CO‚‚‚Zn ) 3.3 Å, average OH‚‚‚Zn ) 2.7 Å; 4: average
CO‚‚‚Zn ) 2.9 Å, average OH‚‚‚Zn ) 3.2 Å).
Inspection of the Autodock 3 and 4 binding modes
clearly highlights the key role of the virtual pyrrole ring
rotation. In the case of isomer 4 (top of Figure 4), the
N-methylpyrrole group does not make crucial interac-
tions with any HDAC1 residues, allowing the prope-
noylhydroxamate tail to fit into the Zn-containing well
and to make the following favorable interactions: (i)
positive π-stacking interactions between the pyrrolyl-
ethylene chain and Phe140 and Phe197 residues; (ii)
hydrogen bond net between CONHOH moiety and the
catalytic residues side chains (His130, His131, Asp167)
plus the catalytic water (Wat618), and (iii) hydrogen
bond between amidic carbonyl oxygen and Tyr296
hydroxyl group. In the case of isomer 3, a slightly
different scenario can be observed (bottom of Figure 4).
While the CONHOH group and the pyrrolylethylene
chain place the same interactions made by the corre-
sponding portions of 4, the N-methyl group of 3 makes
favorable interactions with the CR carbon atom of
Gly139, and the benzoyl oxygen seems to be involved
in a water-bridged hydrogen bond with Tyr90. These
further interactions could account for the 3 slightly
higher HDAC1 inhibitory activity than that observed
for 4, as also recorded by the VALIDATE prediction on
F igu r e 5. Antiproliferative and cytodifferentiating effects of
4 on MEL cell line. (A) Effects of 4 on cell growth of MEL cells,
cultured for 48 h. Data are expressed as area under the curve
(AUC) (mean ( SEM, n ) 4); *, p < 0.05; **, p < 0.01. (B)
Effects of 4 on differentiation of MEL cells. The cells were
cultured with various concentrations of drug for 48 h. Each
point is the mean ( SEM (n ) 4); **, p < 0.01.
0.03 µM; 3: IC50-HDAC1 ) 0.25 µM, VALIDATE pre-
dicted Ki HDAC1-Autodock ) 0.02 µM).
An tip r olifer a tive a n d Cytod iffer en tia tin g Ef-
fects of 4 on F r ien d Mu r in e Er yth r oleu k em ia
(MEL) Cells. In addition to in vitro enzyme assays, the
capability of 4 to induce antiproliferative and cytodif-
ferentiating effects in vivo on Friend MEL cells were
evaluated. Figure 5A shows the effect of 4 on the growth
of MEL cells, cultured for 48 h. Such compounds showed
significant dose-dependent inhibitory effect on the
growth rate of cell line (p < 0.01). Moreover, after 48 h
the inhibitor was not cytotoxic at the tested concentra-
tions (data not shown).
In MEL cells, benzidine staining reveals the hemo-
globin accumulation rate, which is related to the activa-
tion of the cell line differentiation process. The results
(Figure 5B) clearly indicate a dose-dependent increase
of hemoglobin synthesis for 4 at the tested doses.
Con clu sion
Aroyl-pyrrole-hydroxy-alkylamides (APHAs) are a
new class of synthetic HDAC inhibitors recently de-
scribed by us.25,27-29 Through three different docking
procedures, we designed, synthesized, and tested two
new APHA isomers, compounds 3 and 4, characterized
by different insertions of benzoyl and propenoylhydrox-
amate groups onto the pyrrole ring. As biological activi-
ties of 3 and 4 were predicted by computational tools
up to 617-fold more potent than that of lead compound
1 against HDAC1, 3 and 4 were synthesized and tested
against both mouse HDAC1 and maize HD2 enzymes.
Predictions of biological affinities (Ki values) of 3 and
4, performed by a VALIDATE model (applied on either
the Autodock proposed complexes (4: IC50-HDAC1
0.78 µM, VALIDATE predicted Ki HDAC1-Autodock
)
)

1356 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6
Ragno et al.
1 H, pyrrole â-proton), 7.47 (m, 4 H, pyrrole R-proton and
benzene H-3,4,5), 7.76 (m, 2 H, benzene H-2,6), 9.74 (s, 1 H,
CHO). Anal. (C13H11NO2) C, H, N.
SAD or automatic DOCK or Autodock results) and by
the Autodock internal scoring function, were in good
agreement with experimental activities. Clearly, in
comparison with the 1 structure, the switching of
N-methylpyrrole ring introduced some conformational
changes in the APHA binding mode leading to further
ligand/receptor positive interactions and higher HDAC1
inhibitory activities. Particularly, in mouse HDAC1
inhibitory assay 3 and 4 were 19- and 6-times more
active than 1, respectively. Against maize HD2, 3 and
4 were 16- and 76-times more effective in inhibiting the
enzyme than 1, 4 being as potent as SAHA in this assay.
Compound 4, tested as antiproliferative and cytodif-
ferentiating agent on MEL cells, showed dose-dependent
growth inhibition, significant at 4.8 (p < 0.05) and 24
(p < 0.01) µM, and dose-dependent hemoglobin accu-
mulation effect, highly significant (p < 0.01) at 4.8 µM.
In conclusion, the two APHA isomers 3 and 4 can be
considered as new lead compounds in the anti-HDAC
field. We are currently investigating the effect of chemi-
cal modifications at both benzoyl moieties and N-hy-
droxy-2-propenamide chains of 3 and 4 on the enzyme
inhibitory activity.
Gen er a l P r oced u r e for th e Syn th esis of Eth yl 3-(2-
Ben zoyl-1-m eth yl-1H-p yr r ol-5- a n d -4-yl)p r op en oa tes 7,8.
Exa m p le: Eth yl 3-(2-Ben zoyl-1-m eth yl-1H-p yr r ol-5-yl)-
p r op en oa te (7). A suspension of 5 (6.0 mmol, 1.28 g) in
absolute ethanol (20 mL) was added in one portion to a mixture
of triethyl phosphonoacetate (7.2 mmol, 1.5 mL) and anhy-
drous potassium carbonate (18.0 mmol, 2.5 g). After being
stirred at 70 °C for 2 h, the reaction mixture was cooled to
room temperature, diluted with water (50 mL), and extracted
with ethyl acetate (3 × 30 mL). The organic layer was washed
with water, dried, and evaporated to dryness, and the solid
residue was recrystallized to furnish pure 7.
7: yield: 65%; mp: 88-89 °C, recrystallization solvent:
1
cyclohexane; H NMR (CDCl3) δ 1.30 (t, 3 H, CH2CH3), 4.03
(s, 3 H, NCH3), 4.23 (q, 2 H, CH2CH3), 6.35 (d, 1 H, CHd
CHCO), 6.56 (d, 1 H, pyrrole â-proton), 6.65 (d, 1 H, pyrrole
â-proton), 7.45 (m, 4 H, CH)CHCO and benzene H-3,4,5), 7.73
(m, 2 H, benzene H-2,6). Anal. (C17H17NO3) C, H, N.
8: yield: 62%; mp: 83-84 °C, recrystallization solvent:
cyclohexane/benzene; 1H NMR (CDCl3) δ 1.26 (t, 3 H, CH2CH3),
3.98 (s, 3 H, NCH3), 4.18 (q, 2 H, CH2CH3), 6.08 (d, 1 H, CHd
CHCO), 6.85 (d, 1 H, pyrrole â-proton), 7.08 (d, 1 H, pyrrole
R-proton), 7.46 (m, 4 H, CH)CHCO and benzene H-3,4,5), 7.74
(m, 2 H, benzene H-2,6). Anal. (C17H17NO3) C, H, N.
Gen er a l P r oced u r e for th e Syn th esis of 3-(2-Ben zoyl-
1-m eth yl-1H-p yr r ol-5- a n d -4-yl)p r op en oic Acid s 9,10.
E xa m p le. 3-(2-Ben zoyl-1-m et h yl-1H -p yr r ol-4-yl)p r op e-
n oic Acid (10). A mixture of 8 (5.7 mmol, 1.61 g), 2 N KOH
(22.9 mmol, 11.4 mL), and ethanol (15 mL) was heated at
70 °C for 3 h. After cooling, the solution was poured into water
(50 mL) and extracted with ethyl acetate (2 × 20 mL). To the
aqueous layer was added 2 N HCl until the pH was 5, and the
precipitate was filtered and recrystallized from benzene giving
the pure compound 10.
9: yield: 75%; mp: 222-223 °C, recrystallization solvent:
benzene; 1H NMR (DMSO-d6) δ 3.93 (s, 3 H, NCH3), 6.46 (d, 1
H, CHdCHCO), 6.59 (d, 1 H, pyrrole â-proton), 6.83 (d, 1 H,
pyrrole â-proton), 7.53 (m, 4 H, CH)CHCO and benzene
H-3,4,5), 7.66 (m, 2 H, benzene H-2,6), 12.40 (s, 1 H, OH). Anal.
(C15H13NO3) C, H, N.
Exp er im en ta l Section
Ch em istr y. Melting points were determined on a Bu¨chi 530
melting point apparatus and are uncorrected. Infrared (IR)
spectra (KBr) were recorded on a Perkin-Elmer Spectrum One
1
instrument. H NMR spectra were recorded at 200 MHz on a
Bruker AC 200 spectrometer; chemical shifts are reported in
δ (ppm) units relative to the internal reference tetramethyl-
silane (Me4Si). All compounds were routinely checked by TLC
1
and H NMR. TLC was performed on aluminum-backed silica
gel plates (Merck DC-Alufolien Kieselgel 60 F254) with spots
visualized by UV light. All solvents were reagent grade and,
when necessary, were purified and dried by standards meth-
ods. Concentration of solutions after reactions and extractions
involved the use of a rotary evaporator operating at a reduced
pressure of ca. 20 Torr. Organic solutions were dried over
anhydrous sodium sulfate. Analytical results are within
(0.40% of the theoretical values. A SAHA sample for biological
assays was prepared as previously reported by us.47 All
chemicals were purchased from Aldrich Chimica, Milan (Italy)
or Lancaster Synthesis GmbH, Milan (Italy) and were of the
highest purity.
10: yield: 69%; mp: 209-210 °C, recrystallization sol-
vent: benzene; 1H NMR (DMSO-d6) δ 3.87 (s, 3 H, NCH3), 6.12
(d, 1 H, CHdCHCO), 6.97 (d, 1 H, pyrrole â-proton), 7.54 (m,
4 H, CHdCHCO and benzene H-3,4,5), 7.70 (m, 3 H, pyrrole
R-proton and benzene H-2,6). Anal. (C15H13NO3) C, H, N.
Gen er a l P r oced u r e for th e Syn th esis of 3-(2-Ben zoyl-
1-m et h yl-1H -p yr r ol-5- a n d -4-yl)-N-h yd r oxyp r op en a -
m id es 3,4. Exa m p le. 3-(2-Ben zoyl-1-m eth yl-1H-p yr r ol-4-
yl)-N-h yd r oxyp r op en a m id e (4). Ethyl chloroformate (5.0
mmol, 0.5 mL) and triethylamine (5.4 mmol, 0.8 mL) were
added to a cooled (0 °C) solution of 10 (4.2 mmol, 1.07 g) in
dry THF (10 mL), and the mixture was stirred for 10 min.
The solid was filtered off. The filtrate was added to a freshly
prepared solution of hydroxylamine, obtained by reaction
between hydroxylamine hydrochloride (6.2 mmol, 0.4 g) and
KOH (6.2 mmol, 0.35 g) in methanol (10 mL). After being
stirred at room temperature for 15 min, the mixture was
evaporated under reduced pressure and the residue was
recrystallized from benzene to give the pure compound 4.
3: yield: 42%; mp: 134-136 °C, recrystallization solvent:
2-Ben zoyl-1-m eth yl-1H-p yr r ole-5-ca r boxa ld eh yd e (5)
an d 2-Ben zoyl-1-m eth yl-1H-pyr r ole-4-car boxaldeh yde (6).
A 50 mL 1,2-dichloroethane solution of oxalyl chloride (0.06
mol, 5.2 mL) was added to a cooled (0-5 °C) solution of N,N-
dimethylformamide (0.06 mol, 4.6 mL) in 1,2-dichloroethane
(50 mL) over a period of 5-10 min. After being stirred at room
temperature for 15 min, the suspension was cooled (0-5 °C)
again and treated with a solution of 2-benzoyl-1-methyl-1H-
pyrrole38 (0.06 mol, 11.1 g) in 1,2-dichloroethane (50 mL). The
mixture was stirred at room temperature for 1 h and then was
poured onto crushed ice (200 g) containing 50% NaOH (50 mL)
and stirred for 10 min. The pH of the solution was adjusted to
4 with 37% HCl, the organic layer was separated, and the
aqueous one was extracted with chloroform (2 × 50 mL). The
combined organic solutions were washed with water, dried,
and evaporated to dryness. The residual oil was chromato-
graphed on silica gel eluting with ethyl acetate:chloroform
1:10. The first eluates were collected and evaporated to afford
5 as a pure solid; further elution gave 639 as a pure solid.
5: yield: 44%; mp: 90-91 °C, recrystallization solvent:
cyclohexane; 1H NMR (CDCl3) δ 4.23 (s, 3 H, CH3), 6.62 (m, 1
H, pyrrole â-proton), 6.87 (m, 1 H, pyrrole â-proton), 7.44 (m,
3 H, benzene H-3,4,5), 7.78 (m, 2 H, benzene H-2,6), 9.77 (s, 1
H, CHO). Anal. (C13H11NO2) C, H, N.
1
benzene; H NMR (DMSO-d6) δ 3.93 (s, 3 H, NCH3), 6.42 (d,
1 H, CHdCHCO), 6.58 (d, 1 H, pyrrole â-proton), 6.80 (d, 1 H,
pyrrole â-proton), 7.40 (d, 1 H, CHdCHCO), 7.54 (m, 3 H,
benzene H-3,4,5), 7.67 (m, 2 H, benzene H-2,6), 9.00 (s, 1 H,
NH), 10.36 (s, 1 H, OH). Anal. (C15H14N2O3) C, H, N.
4: yield: 54%; mp: 144-145 °C, recrystallization solvent:
1
benzene; H NMR (DMSO-d6) δ 3.88 (s, 3 H, NCH3), 6.04 (d,
1 H, CHdCHCO), 6.76 (s, 1 H, pyrrole â-proton), 7.24 (d, 1 H,
CHdCHCO), 7.54 (m, 4 H, pyrrole R-proton and benzene
H-3,4,5), 7.70 (m, 2 H, benzene H-2,6), 8.80 (s, 1 H, NH), 10.40
(s, 1 H, OH). Anal. (C15H14N2O3) C, H, N.
6:39 yield: 56%; mp: 108-109 °C, recrystallization solvent:
cyclohexane; 1H NMR (CDCl3) δ 4.04 (s, 3 H, CH3), 7.12 (d,

Synthetic Histone Deacetylase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6 1357
Molecu la r Mod elin g Stu d ies. All molecular modeling
calculations and manipulations were performed using the
software packages Macromodel 7.1,32 MOPAC 2000,35,36 VALI-
DATE,37 GOLPE 4.5.12,48 XlogP 2,49 Autodock 3.0.5,31 DOCK
4.0.2,30 and MIDAS 2.150 running on Silicon Graphics O2
R10000, IBM-compatible AMD Athlon 2.4 GHz workstations.
For the conformational analysis and for any minimization the
all-atom Amber force field33,34 was adopted as implemented
in the Macromodel package. As previously reported26 the
crystal structure of TSA extracted from the HDLP/TSA
complex filed in the Brookhaven Protein Data Bank51 (entry
code 1c3r) was used. The HDAC1 model was constructed from
HDLP as previously reported.27 Before any docking studies the
HDAC1 model was minimized in a vacuum in the presence of
TSA to relieve any steric contact introduced by the mutated
residue’s side chains.
Simulated annealing (SA) run were also performed either
in water or in a vacuum on the HDAC1 model with and
without the bound TSA. Comparison of the roughly minimized
HDAC1/TSA complex with the results of the SA dynamic runs
showed very low RMSD values on the CR traces, also including
the residue side-chains, thus revealing that the direct use of
the minimized virtual HDAC1 is a fast and reliable choice.
Binding modes of 3 and 4 were extensively analyzed by the
mean of three different docking procedures. A semiautomatic
docking (SAD) was conducted similarly to that reported for 1.
A family of more than 200 conformations for either 3 and 4
was generated using the following procedure: the initial
conformation of 3 or 4 obtained from the reported bound
conformation of 1 was arbitrarily rotated by a step of 10° along
an axis passing through the center of the benzene ring and
the hydroxamic acid carbon atom. Each of the obtained new
36 HDAC1/3 or HDAC1/4 complexes was submitted to a grid
search rotating the five dihedral angles by a step of 30°. By
filtering out all of the complexes showing a molecular clashing
between 3 or 4 and the enzyme pocket, the final conformation
families were achieved. After minimization, the chosen binding
conformations of 3 and 4 were those associated with the most
stable complexes (global minimum).
For the second docking procedure the program Autodock
was used to explore the binding conformation of either 3 or 4.
For the docking a grid spacing of 0.375 Å and 58 × 52 × 48
number of points were used. The grid was centered on the
mass center of the experimental bound TSA coordinates. The
GA-LS method was adopted using the default settings. Amber
united atoms were assigned to the protein using the program
ADT (Auto Dock Tools). Autodock generated 100 possible
binding conformations for 3 and 4.
The third docking was performed by the mean of the
program DOCK. First, a Connolly surface of each receptor
active site was generated by using a 1.4 Å probe radius and
further used to generate a set of overlapping spheres that were
then clustered according to their spatial distribution. To
optimize docking accuracy, spheres located too far away from
the known ligand position were eliminated from the finally
selected cluster. To compute interaction energies, a 3-D grid
of 0.30 Å resolution was centered on TSA experimental bound
conformation. The size of the grid box was chosen to enclose
all selected spheres using an extra margin of 6 Å. The grid
had a size of about 36 × 31 × 33 Å and comprised about
1400000 grid points. Energy scoring grids were obtained by
using an all-atom model and a distance-dependent dielectric
function (ꢀ ) 4r) with a 10 Å cutoff. Amber atomic charges
were assigned to all protein atoms. The ligands were then
docked in turn into the protein active site by matching sphere
centers with ligand atoms. A flexible docking (peripheral
search and torsion drive) with subsequent minimization was
performed as follows: (i) automatic selection and matching of
an anchor fragment within a maximum of 10000 orientations,
(ii) iterative growing of the ligand using at least 30 conforma-
tions (peripheral seeds) for seeding the next growing stage with
assignment of energy-favored torsion angles, (iii) simultaneous
relaxation of the base fragments as well as of all peripheral
segments and final relaxation of the entire molecule. Orienta-
tions/conformations were relaxed in 100 cycles of 100 simplex
minimization steps to a convergence of 0.1 kcal/mol. The top
50 solutions corresponding to the best Dock energy scores were
then stored in a single multi mol2 file.
Because of neither Autodock nor DOCK are able to perform
any energy minimization of the generated complexes, the
selection of the binding conformation for 3 or 4 was not
straightforward by using the first Autodock/DOCK scored
conformation. The selection of the binding conformation was
performed by an energy based rescoring of the Autodock
cluster representants and the 50 best contact scored DOCK
conformations. The Macromodel program was used to mini-
mize the corresponding 32 Autodock and the 50 DOCK
HDAC1/3 or HDAC1/4 complexes. The ligand and an 10 Å core
of atoms of the pocket were let to relax during the minimiza-
tion. An external fixed shell of 8 Å was also included for the
long-range interactions. Because of the presence of a metal
Zn ion in the HDAC1 catalytic core and the intrinsic molecular
mechanic electrostatic limitation of the AMBER force field the
minimizations were performed applying AM1 charges calcu-
lated with the program MOPAC 2000.
The 3 and 4 starting conformations for the docking studies
were obtained using molecular dynamics with simulated
annealing as implemented in Macromodel version 7.1 and
conducted as following: the molecules were energy minimized
to a low gradient. The nonbonded cutoff distances were set to
20 Å for both van der Waals and electrostatic interactions. An
initial random velocity to all atoms corresponding to 300 K
was applied. Three subsequent molecular dynamics runs were
then performed. The first was carried out for 10 ps with a 1.5
fs time-step at a constant temperature of 300 K for equilibra-
tion purposes. The next molecular dynamic was carried out
for 20 ps, during which the system is coupled to a 150 °C
thermal bath with a time constant of 5 ps. The time constant
represents approximately the half-life for equilibration with
the bath; consequently, the second molecular dynamic com-
mand caused the molecule to slowly cool to approximately
150 K. The third and last dynamic cooled the molecule to
50 K over 20 ps. A final energy minimization was then carried
out for 250 iterations using conjugate gradient. The minimiza-
tions and the molecular dynamics were in all cases performed
in aqueous solution. The atom charges automatically assigned
by the batchmin module were retained on 3 or 4 for the docking
calculations.

1358 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6
Ragno et al.
The VALIDATE procedure was used to estimate the binding
conformation of the docking studies as previously reported.
The partition coefficients were calculated using the program
XlogP.
2002” (A.M.), “AIRC Proposal 2003” (A.M.), and the
Austrian Science Foundation P13209 (G.B.) and P13620
(P.L.).
Biologica l Assa ys. HD2 En zym e In h ibition . Radioac-
tively labeled chicken core histones were used as the enzyme
substrate according to established procedures.40 The enzyme
liberated tritiated acetic acid from the substrate which was
quantitated by scintillation counting. IC50 values are results
of triple determinations. Maize enzyme (50 µL) (at 30 °C) was
incubated (30 min) with 10 µL of total [3H]acetate-prelabeled
chicken reticulocyte histones (4 mg/mL). Reaction was stopped
by addition of 36 µL of 1 M HCl/0.4 M acetate and 800 µL of
ethyl acetate. After centrifugation (10 000g, 5 min), an aliquot
of 600 µL of the upper phase was counted for radioactivity in
3 mL of liquid scintillation cocktail. The compounds were
Refer en ces
(1) Wu, J .; Grunstein, M. 25 Years after the nucleosome model:
chromatin modifications. Trends Biochem. Sci. 2000, 25, 619-
623.
(2) Urnov, F. D.; Wolffe, A. Chromatin organization and human
disease. Emerg. Ther. Targets 2000, 4, 665-685.
(3) Pazin, M. J .; Kadonaga, J . T. What is up and down with histone
deacetylation and transcription? Cell 1997, 89, 325-328.
(4) Grunstein, M. Histone acetylation in chromatin structure and
transcription. Nature 1997, 389, 349-352.
(5) Davie, J . R. Covalent modifications of histones: expression from
chromatin templates. Curr. Opin. Genet. Dev. 1998, 8, 173-178.
(6) Kouzarides, T. Histone acetylases and deacetylases in cell
proliferation. Curr. Opin. Genet. Dev. 1999, 9, 40-48.
(7) Yoshida, M.; Kijima, M.; Akita, M.; Beppu, T. Potent and Specific
Inhibition of Mammalian Histone Deacetylase both In Vivo and
In Vitro by Trichostatin A. J . Biol. Chem. 1990, 265, 17174-
17179.
(8) Kijima, M.; Yoshida, M.; Suguta, K.; Horinouchi, S.; Beppu, T.
Trapoxin, an Antitumor Cyclic Tetrapeptide, Is an Irreversible
Inhibitor of Mammalian Histone Deacetylase. J . Biol. Chem.
1993, 268, 22429-22435.
(9) Shute, R. E.; Dunlap, B.; Rich, D. H. Analogues of the Cytostatic
and Antimitogenic Agents Chlamydocin and HC-Toxin: Syn-
thesis and Biological Activity of Chloromethyl Ketone and
Diazomethyl Ketone Functionalized Cyclic Tetrapeptides. J .
Med. Chem. 1987, 30, 71-78.
(10) Han, J . W.; Ahn, S. H.; Park, S. H.; Wang, S. Y.; Bae, G. U.;
Seo, D. W.; Known, H. K.; Hong, S.; Lee, Y. W.; Lee, H. W.
Apicidin, a histone deacetylase inhibitor, inhibits proliferation
of tumor cells via induction of p21WAF1/Cip1 and gelsolin.
Cancer Res. 2000, 60, 6068-6074.
(11) Richon, V. M.; Emiliani, S.; Verdin, E.; Webb, Y.; Breslow, R.;
Rifkind, R. A.; Marks, P. A. A class of hybrid polar inducers of
transformed cell differentiation inhibits histone deacetylases.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 3003-3007.
(12) Furumai, R.; Komatsu, Y.; Nishino, N.; Khochbin, S.; Yoshida,
M.; Horinouchi, S. Potent histone deacetylase inhibitors built
from trichostatin A and cyclic tetrapeptide antibiotics including
trapoxin. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 87-92.
(13) Komatsu, Y.; Tomizaki, K.; Tsukamoto, M.; Kato, T.; Nishino,
N.; Sato, S.; Yamori, T.; Tsuruo, T.; Furumai, R.; Yoshida, M.;
Horinouchi, S.; Hayashi, H. Cyclic Hydroxamic-acid-containing
Peptide 31, a Potent Synthetic Histone Deacetylase Inhibitor
with Antitumor Activity. Cancer Res. 2001, 61, 4459-4466.
(14) Ueda, H.; Nakajima, H.; Hori, Y.; Fujita, T.; Nishimura, M.; Goto,
T.; Okuhara, M. FR901228, a novel antitumor bicyclic depsipep-
tide produced by Chromobacterium violaceum No. 968. I.
Taxonomy, fermentation, isolation, physicochemical and biologi-
cal properties, and antitumor activity. J . Antibiot. 1994, 47,
301-310.
(15) Suzuki, T.; Ando, T.; Tsuchiya K.; Fukazawa, N.; Saito, A.;
Mariko, Y.; Yamashita, T.; Nakanishi, O. Synthesis and Histone
Deacetylase Inhibitory Activity of New Benzamide Derivatives.
J . Med. Chem. 1999, 42, 3001-3003.
(16) Saito, A.; Yamashita, T.; Mariko, Y.; Nosaka, Y.; Tsuchiya, K.;
Ando, T.; Suzuki, T.; Tsuruo, T.; Nakanishi, O. A synthetic
inhibitor of histone deacetylase, MS-27-275, with marked in
vivo antitumor activity against human tumors. Proc. Natl. Acad.
Sci. U.S.A. 1999, 96, 4592-4597.
tested in
a starting concentration of 40 µM, and active
substances were diluted further. Sodium butyrate, sodium
valproate, TSA, SAHA,47 trapoxin, and HC-toxin were used
as the reference compounds, and blank solvents were used as
negative controls.
Mou se HDAC1 En zym e Assa y. For inhibition assay,
partially purified HDAC1 from mouse A20 cells (ATCC: TIB-
208) (anion exchange chromatography, affinity chromatogra-
phy was used as enzyme source. HDAC activity was deter-
mined as described52 using [3H]acetate-prelabeled chicken
reticulocyte histones as substrate. 50 µL of mouse HDAC1
were incubated with different concentrations of compounds for
15 min on ice and 10 µL of total [3H]acetate-prelabeled chicken
reticulocyte histones (4 mg/mL) were added, resulting in a
concentration of 41 µM. The mixture was incubated at 37 °C
for 1 h. The reaction was stopped by addition of 50 µL of 1 M
HCl/0.4 M acetylacetate and 1 mL ethyl acetate. After cen-
trifugation at 10 000g for 5 min an aliquot of 600 µL of the
upper phase was counted for radioactivity in 3 mL of liquid
scintillation cocktail.
Gr ow t h In h ib it ion a n d Cell Differ en t ia t ion Assa y.
Cell Cu ltu r e a n d Rea gen ts. Murine erythroleukemia (MEL)
cells were obtained from Interlab Cell Line Collection (CBA)
(Genoa, Italy). Cells were maintained at 37 °C under a
humidified atmosphere of 5% CO2 in RPMI 1640 Hepes
modified medium supplemented with 10% (v/v) heat inacti-
vated foetal calf serum, 2 mmol/L glutamine, 100 IU/mL
penicillin and 100 µg/mL streptomycin. Unless indicated all
chemicals and reagents (cell culture grade) were obtained from
Sigma Chemical Co., Milan, Italy.
Cell Via bility a n d Gr ow th In h ibition Assa y. Cell num-
ber was determined using a Neubauer hemocytometer, and
viability was assessed by their ability to exclude trypan blue.
The stock solutions were prepared immediately before use.
Compound 4 was dissolved in DMSO. MEL exponentially
growing cells (1 × 105 cells/mL) were set at day 0 in media
containing various concentrations of drugs for 48 h. The final
concentrations of the drug were: 1.2, 2.4, 4.8, and 24 µM. The
final concentration of DMSO, used as vehicle, was the same
(0.1% v/v) in all samples during the experiments.
(17) Go¨ttlicher, M.; Minucci, S.; Zhu, P.; Kramer, O. H.; Schimpf, A.;
Giavara, S.; Sleeman, J . P.; Lo Coco, F.; Nervi, C.; Pelicci, P. G.;
Heinzel, T. Valproic acid defines a novel class of HDAC inhibitors
inducing differentiation of transformed cells. EMBO J . 2001, 20,
6969-6978.
(18) Phiel, C. J .; Zhang, F.; Huang, E. Y.; Guenther, M. G.; Lazar,
M. A.; Klein, P. S. Histone deacetylase is a direct target of
valproic acid, a potent anticonvulsant, mood stabilizer, and
teratogen. J . Biol. Chem. 2001, 76, 36734-36741.
Cytom or p h ologica l Assa y for MEL Cell Differ en tia -
tion . The most widely used method for scoring erythroid
differentiation is benzidine staining which reveals the produc-
tion of hemoglobin.53 Benzidine dihydrochloride (2 mg/mL) was
prepared in 3% acetic acid. Hydrogen peroxide (1%) was added
immediately before use. The MEL cell suspensions were mixed
with the benzidine solution in a 1:1 ratio and counted in a
hemocytometer after 5 min. Blue cells were considered to be
positive for hemoglobin.
(19) J ohnstone, R. W. Histone-deacetylase inhibitors: novel drugs
for the treatment of cancer. Nature Rev. Drug Discovery 2002,
1, 287-299.
(20) Vigushin, D. M.; Coombes, R. C. Histone deacetylase inhibitors
in cancer treatment. Anticancer Drugs 2002, 13, 1-13.
(21) Kelly, W. K.; O′Connor, O. A.; Marks, P. A. Histone deacetylase
inhibitors: from target to clinical trials. Expert Opin. Investig.
Drugs 2002, 11, 1695-1713.
(22) Finnin, M. S.; Donigian, J . R.; Cohen, A.; Richon, V. M.; Rifkind,
R. A.; Marks, P. A.; Breslow, R.; Pavletich, N. P. Structures of
a histone deacetylase homologue bound to the TSA and SAHA
inhibitors. Nature 1999, 401, 188-193.
Sta tistica l An a lysis. All results are expressed as mean (
SEM. The group means were compared by analysis of variance
(ANOVA) followed by a multiple comparison of means by
Dunnet test; p < 0.05 was considered significant.
Ack n ow led gm en t. This work was supported by
grants of “Progetto Finalizzato Ministero della Salute

Synthetic Histone Deacetylase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 6 1359
(23) Massa, S.; Artico, M.; Corelli, F.; Mai, A.; Di Santo, R.; Cortes,
S.; Marongiu, M. E.; Pani, A.; La Colla, P. Synthesis and
Antimicrobial and Cytotoxic Activities of Pyrrole-Containing
Analogues of Trichostatin A. J . Med. Chem. 1990, 33, 2845-
2849.
(24) Corelli, F.; Massa, S.; Stefancich, G.; Mai, A.; Artico, M.; Panico,
S.; Simonetti, N. Ricerche su composti antibatterici ed anti-
fungini. Nota VIII - Sintesi ed attivita` antifungina di derivati
pirrolici correlati con la tricostatina A. (Researches on antibacte-
rial and antifungal agents. VIII. Synthesis and antifungal
activity of trichostatin A-related pyrrole derivatives.) Farmaco,
Ed. Sci. 1987, 42, 893-903.
(25) Massa, S.; Mai, A.; Sbardella, G.; Esposito, M.; Ragno, R.; Loidl,
P.; Brosch, G. 3-(4-Aroyl-1H-pyrrol-2-yl)-N-hydroxy-2-propena-
mides, a New Class of Synthetic Histone Deacetylase Inhibitors.
J . Med. Chem. 2001, 44, 2069-2072.
(26) Lusser, A.; Brosch, G.; Loidl, A.; Haas, H.; Loidl, P. Identification
of maize histone deacetylase HD2 as an acidic nucleolar phos-
phoprotein. Science 1997, 277, 88-91.
(27) Mai, A.; Massa, S.; Ragno, R.; Esposito, M.; Sbardella, G.; Nocca,
G.; Scatena, R.; J esacher, F.; Loidl, P.; Brosch, G. Binding Mode
Analysis of 3-(4-Benzoyl-1-methyl-1H-2-pyrrolyl)-N-hydroxy-2-
propenamide: A New Synthetic Histone Deacetylase Inhibitor
Inducing Histone Hyperacetylation, Growth Inhibition, and
Terminal Cell Differentiation. J . Med. Chem. 2002, 45, 1778-
1784.
(28) Mai, A.; Massa, S.; Ragno, R.; Cerbara, I.; J esacher, F.; Loidl,
P.; Brosch, G. 3-(4-Aroyl-1-methyl-1H-2-pyrrolyl)-N-hydroxy-2-
alkylamides as a New Class of Synthetic Histone Deacetylase
Inhibitors. 1. Design, Synthesis, Biological Evaluation, and
Binding Mode Studies Performed Through Three Different
Docking Procedures. J . Med. Chem. 2003, 46, 512-524.
(29) Mai, A.; Massa, S.; Cerbara, I.; Valente, S.; Ragno, R.; Bottoni,
P.; Scatena, R.; Loidl, P.; Brosch, G. 3-(4-Aroyl-1-methyl-1H-2-
pyrrolyl)-N-hydroxy-2-propenamides as a New Class of Synthetic
Histone Deacetylase Inhibitors. 2. Effect of Pyrrole C2 and/or
C4 Substitutions on Biological Activity. J . Med. Chem., in press.
(30) Oshiro, C. M.; Kuntz, I. D.; Dixon, J . S. Flexible ligand docking
using a genetic algorithm. J . Comput. Aided Mol. Des. 1995, 9,
113-130.
(31) Goodsell, D. S.; Morris, G. M.; Olson, A. J . Automated docking
of flexible ligands: applications of AutoDock. J . Mol Recognit.
1996, 9, 1-5.
(32) Mohamadi, F.; Richards, N. G. J .; Guida, W. C.; Liskamp, R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W.
C. MACROMODEL - an integrated software system for model-
ing organic and bioorganic molecules using molecular mechanics.
J . Comput. Chem. 1990, 11, 440-467.
(33) Pearlman, D. A.; Case, D. A.; Caldwell, J . W.; Ross, W. S.;
Cheatham T. E., III; Debolt, S.; Ferguson, D. M.; Seibel, G. L.;
Kollman, P. A. AMBER, a Package of Computer Programs for
Applying Molecular Mechanics, Normal-Mode Analysis, Molec-
ular Dynamics and Free Energy Calculations to Simulate the
Structural and Energetic Properties of Molecules. Comput. Phys.
Commun. 1995, 91, 1-41.
(34) Pearlman, D. A.; Case, D. A.; Caldwell, J . W.; Ross, W. S.;
Cheatham, T. E.; III, Ferguson, D. M.; Seibel, G. L.; Singh, U.
C.; Weiner, P. K.; Kollman, P. A. AMBER 4.1; Department of
Pharmaceutical Chemistry, University of California: San Fran-
cisco, CA, 1995.
(37) Head, R. D.; Smythe, M. L.; Oprea, T. I.; Waller, C. L.; Green,
S. M.; Marshall, G. R. VALIDATE: A New Method for the
Receptor-Based Prediction of Binding Affinities of Novel Ligands.
J . Am. Chem. Soc. 1996, 118, 3959-3969.
(38) Brittain, J . M.; J ones, R. A.; Arques, J . S.; Saliente, T. A. Pyrrole
studies. XXVII. Utilization of 1-methyl-2-pyrrolyllithium in the
synthesis of 1-methyl-2-substituted pyrroles. Synth. Commun.
1982, 12, 231-248.
(39) Lehr, M. Structure-Activity Relationships of (4-Acylpyrrol-2-
yl)alkanoic Acids as Inhibitors of the Cytosolic Phospholipase
A2: Variation of the Substituent in Positions 1, 3, and 5. J . Med.
Chem. 1997, 40, 3381-3392.
(40) Lechner, T.; Lusser, A.; Brosch, G.; Eberharter, A.; Goralik-
Schramel, M.; Loidl, P. A comparative study of histone deacety-
lases of plant, fungal and vertebrate cells. Biochim. Biophys.
Acta 1996, 1296, 181-188.
(41) Brosch, G.; Lusser, A.; Goralik-Schramel, M.; Loidl, P. Purifica-
tion and characterization of a high molecular weight histone
deacetylase complex (HD2) of maize embryos. Biochemistry 1996,
35, 15907-15914.
(42) Ko¨lle, D.; Brosch, G.; Lechner, T.; Lusser, A.; Loidl, P. Biochemi-
cal methods for analysis of histone deacetylases. Methods 1998,
15, 323-331.
(43) Ko¨lle, D.; Brosch, G.; Lechner, T.; Pipal, A.; Helliger, W.; Taplick,
J .; Loidl, P. Different types of maize histone deacetylases are
distinguished by a highly complex substrate and site specificity.
Biochemistry 1999, 38, 6769-6773.
(44) Brosch, G.; Ransom, R.; Lechner, T.; Walton, J .; Loidl, P.
Inhibition of maize histone deacetylases by HC toxin, the host-
selective toxin of Cochliobolus carbonum. Plant Cell 1995, 33,
1941-1950.
(45) J ung, M.; Brosch, G.; Ko¨lle, D.; Scherf, H.; Gerha¨user, C.; Loidl,
P. Amide Analogues of Trichostatin A as Inhibitors of Histone
Deacetylase and Inducers of Terminal Cell Differentiation. J .
Med. Chem. 1999, 42, 4669-4679.
(46) Wittich, S.; Scherf, H.; Xie, C.; Brosch, G.; Loidl, P.; Gerha¨user,
C.; J ung, M. Structure-Activity Relationships on Phenylalanine-
Containing Inhibitors of Histone Deacetylase: In Vitro Enzyme
Inhibition, Induction of Differentiation, and Inhibition of Pro-
liferation in Friend Leukemic Cells. J . Med. Chem. 2002, 45,
3296-3309.
(47) Mai, A.; Esposito, M.; Sbardella, G.; Massa, S. A new facile and
expeditious synthesis of N-hydroxy-N′-phenyloctanediamide, a
potent inducer of terminal cytodifferentiation. Org. Prep. Proced.
Int. 2001, 33, 391-394.
(48) GOLPE . Multivariate Infometric Analysis Srl.: Viale dei Cast-
agni 16, Perugia, Italy (1999).
(49) Wang, R.; Fu, Y.; Lai, L. A New Atom-Additive Method for
Calculating Partition Coefficients. J . Chem. Inf. Comp. Sci. 1997,
37, 615-621.
(50) Ferrin, T. E.; Huang, C. C.; J arvis, L. E.; Langridge, R. The
MIDAS Display System. J . Mol. Graphics 1988, 6, 13-27.
(51) Berman, H. M.; Westbrook, J .; Feng, Z.; Gilliland, G.; Bhat, G.
N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The Protein Data
Bank. Nucl. Acids Res. 2000, 28, 235-242.
(52) Sendra, R.; Rodrigo, I.; Salvador, M. L.; Franco, L. Characteriza-
tion of pea histone deacetylases. Plant Mol. Biol. 1988, 11, 857-
866.
(53) Rowley, P. T.; Ohlsson-Wilhelm, B. M.; Farley, B. A.; La Bella,
S. Inducers of erythroid differentiation in K562 human leukemia
cells. Exp. Hemat. 1981, 9, 32-37.
(35) Stewart, J . J . MOPAC:
a semiempirical molecular orbital
program. J . Comput. Aided Mol. Des. 1990, 4, 1-105.
(36) MOPAC 2000.00 Manual; Stewart, J . J . P., Ed; Fujitsu Lim-
ited: Tokyo, J apan (1999).
J M031036F