Bispyridinium dienes: Histone deacetylase inhibitors with selective activities

Article abstract of DOI:10.1021/jm070028m

A novel synthetic route to the cyclostellettamines 1 using as the key step a microwave-mediated macrocyclic ring-closing metathesis of precursors bispyridinium dienes 10 followed by catalytic hydrogenation has been developed. The open-chain bispyridinium dienes 10 showed uniformly higher histone deacetylase (HDAC) inhibitory potency than the natural products. Diene 10b inhibited HDAC1 and was inactive on HDAC4, whereas 10a showed a weak inhibition of HDAC1 and a higher activity on HDAC4. Neither 10b nor 10a inhibited isoforms HDAC2 and HDAC3. Cell cycle analysis, cell differentiation, and apoptosis as well as evaluation of the acetylation status of H3 lysine tails, up-regulation of p21^WAF1/CIP1, and α-tubulin acetylation induced by the dienes 10 and cyclostellettamines 1 were also carried out on the human leukemia U937 cell line. These enzymatic and functional assays suggest that 10b is a HDAC1-selective inhibitor and 10a is a HDAC subclass IIa-selective inhibitor.

Full text of DOI:10.1021/jm070028m

J. Med. Chem. 2007, 50, 2497-2505
2497
Bispyridinium Dienes: Histone Deacetylase Inhibitors with Selective Activities
Carlos Pe´rez-Balado,† Angela Nebbioso,‡ Paula Rodr´ıguez-Gran˜a,† Annunziata Minichiello,‡ Marco Miceli,‡
Lucia Altucci,*, ‡,§ and AÄ ngel R. de Lera*,†
Departamento de Qu´ımica Orga´nica, UniVersidade de Vigo, Lagoas-Marcosende, 36310 Vigo, Spain, Dipartimento di Patologia Generale,
Seconda UniVersita` degli Studi di Napoli, Vico L. de Crecchio 7 80138 Napoli, NOGEC, Naples Oncogenomic Center, CEINGE Biotecnologia
AVanzata, Napoli, Italy
ReceiVed January 9, 2007
A novel synthetic route to the cyclostellettamines 1 using as the key step a microwave-mediated macrocyclic
ring-closing metathesis of precursors bispyridinium dienes 10 followed by catalytic hydrogenation has been
developed. The open-chain bispyridinium dienes 10 showed uniformly higher histone deacetylase (HDAC)
inhibitory potency than the natural products. Diene 10b inhibited HDAC1 and was inactive on HDAC4,
whereas 10a showed a weak inhibition of HDAC1 and a higher activity on HDAC4. Neither 10b nor 10a
inhibited isoforms HDAC2 and HDAC3. Cell cycle analysis, cell differentiation, and apoptosis as well as
evaluation of the acetylation status of H3 lysine tails, up-regulation of p21WAF1/CIP1, and R-tubulin acetylation
induced by the dienes 10 and cyclostellettamines 1 were also carried out on the human leukemia U937 cell
line. These enzymatic and functional assays suggest that 10b is a HDAC1-selective inhibitor and 10a is a
HDAC subclass IIa-selective inhibitor.
Introduction
congeners have more recently been reported by the groups of
Fusetani10b and Berlinck10c from other marine organisms. These
natural products are endowed with interesting biological activi-
ties, among them the inhibition of binding of methyl quinu-
clidinyl benzylate to the muscarinic acetylcholinesterase
receptors10a and, most notably, the inhibition of histone
deacetylases.10b As HDAC inhibitors, the cyclostellettamine
macrocyclic alkaloids are structurally unrelated to the already
described inhibitors,11 posing challenging questions about their
mechanism of action. Structurally simple, the cyclostellettamines
(termed A-L, 1a-l, Figure 1) consist of two 3-alkylpyridinium
units, connected by alkyl chains ranging from 10 to 14 carbons
in length. These chains are fully saturated in most members of
the family, except in two, namely dehydrocyclostellettamines
D 2d and E 2e, which present one C-C double bond with Z
geometry. Our interest in the anticancer activities of the
epigenetic modulators9a led us to address the synthesis and
determine the histone deacetylase inhibitory profile of this family
of alkaloids. We present herein the synthesis, characterization,
and biological evaluation of several members of the cyclostel-
lettamine family of natural products. We confirm the reported
micromolar HDAC inhibitory activity of most of the macrocy-
clic compounds, and most importantly we report that the open-
chain precursors used as intermediates in the synthesis also
exhibit differential HDAC inhibitory activities and could become
promising leads for the development of more potent (sub)class-
selective HDAC inhibitors.
While the genetic aberrations occurring in cancer are fairly
well understood, we have only recently become aware of the
epigenetic deregulation associated with cancer.1 The deposition
of epigenetic “marks” on chromatin, post-translational modifica-
tions of nucleosomal proteins and methylation of particular DNA
sequences, is accomplished by enzymes often embedded in
multi-subunit complexes.2 The control of the enzymatic ma-
chinery implicated in the (un)packing of chromatin,3 such as
the reversible acetylation (HAT enzymes)/deacetylation (HDAC
enzymes) of lysine residues,4 the methylation/demethylation of
histone lysine and arginine residues,5 and the DNA methylation
(DNMT enzymes)/demethylation6 is an important step in the
regulation of transcriptional events. As a consequence, chro-
mating-remodeling enzymes, in particular HDACs and DNMTs,
have recently emerged as new promising targets of the so-called
“epigenetic drugs” for the treatment of cancer.7,6a Indeed, recent
results indicate that epigenetic drugs may constitute an entirely
novel type of anti-cancer therapy with unanticipated potential.6a
Proof-of-principle comes from studies with histone deacetylase
inhibitors, promising novel anti-cancer drugs, which are cur-
rently entering therapy.7a,8 We and others have demonstrated
that upon treatment with an HDAC inhibitor (HDACi), TRAIL
(TNF-related apoptosis inducing ligand) is up-regulated and
mediates acute apoptosis induced in both AML cells or patients
blasts.9a,b Given the major interest in finding new therapies and
the growing importance of epi-drugs for cancer treatment, there
is a need for new molecules with epigenetic modulatory
activities showing selectivity for particular families or members
of the chromatin-modifying enzymes.4b
Chemistry
The first total synthesis of cyclostellettamine C 1c was
accomplished by Fusetani and co-workers in 1996.12a Shortly
after, the groups of Koomen and Baldwin independently reported
analogous methodologies for the ring-closure based on the
quaternization of the pyridine nitrogen.12b,c We considered a
novel synthetic approach to these alkaloids using a ring-closing
metathesis13 of a bispyridinium diene precursor as the key
transformation.
The cyclostellettamines are a growing family of macrocyclic
bispyridinium alkaloids first isolated by Fusetani and co-workers
in 1994 from the marine sponge Haliclona sp.10a Novel
* To whom correspondence should be addressed. Angel R. de Lera,
Tel: +34-986-812316, Fax +34-986-811940, E-mail: qolera@uvigo.es;
Lucia Altucci, Tel: +39081-566-7569, Fax: +39081-450-169, E-mail:
lucia.altucci@unina2.it.
† Universidade de Vigo.
The synthesis began with the alkylation of the organolithium
derived from the treatment of 3-methylpyridine 3 with LDA
‡ Seconda Universita` degli Studi di Napoli.
§ CEINGE Biotecnologia Avanzata.
10.1021/jm070028m CCC: $37.00 © 2007 American Chemical Society
Published on Web 04/21/2007

2498 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Pe´rez-Balado et al.
Scheme 1a
Figure 1. Natural cyclostellettamines A-L (1a-l) and dehydroana-
logues (2d, 2e).
(Scheme 1).14 For the C12 series, the THP-protected 11-bromo-
undecan-1-ol 4a was used as alkylating agent, affording the
alkylpyridine 5a in 76% yield. In a similar manner, for the C13
series, bromide 4b provided 5b. Subsequent quaternization of
the 3-alkylpyridines 5a,b with 8-bromo-oct-1-ene furnished the
pyridinium salts 6a and 6b in excellent yields. The THP group
was removed with dilute HCl in methanol to give the free
alcohols 7a and 7b. Activation of the alcohol function was easily
effected by treatment with SOCl2 in dichloromethane to yield
the chloroalkylpyridinium chlorides 8a and 8b. The second
pyridinium subunit was introduced by quaternisation of the
3-substituted pyridines 9a, 9b, and 9c with chlorides 8a and
8b in the presence of 1.2 equiv of NaI to afford 10a-f (Scheme
2). The 3-alkenyl pyridines 9a-c were in turn prepared by
alkylation of the lithiated 3-methylpyridine with the correspond-
ing bromoalkenes (4-bromobut-1-ene for 9a; 5-bromopent-1-
ene for 9b and 7-bromohept-1-ene for 9c), as described for
the synthesis of 5. The alkylation of (pyrid-3-ylmethyl)lithium
with 4-bromobut-1-ene, under the classical conditions (3 mol
equiv of organolithium relative to the alkylating agent), gave a
mixture of of 9a and its regioisomer (E)-3-(pent-3′-en-1′-yl)-
pyridine. Using equimolar amounts of both the organolithium
and the alkylating agent resulted in lower yields (40%), but the
expected terminal alkene 9a was obtained as the exclusive
product.
a Reagents and reaction conditions: (a) 1. LDA (3 equiv), THF, DMPU
(3 equiv), 0 °C; 2. 4a or 4b, THF, -78 to 25 °C, 8 h, 70-76%. (b) 8-bromo-
oct-1-ene, CH3CN, reflux, 95-99%. (c) 0.5 M HCl, MeOH, 25 °C, 18 h,
82-87%. (d) SOCl2, CH2Cl2, 25 °C, 3 h, 90-95%.
Scheme 2. Synthesis of the Open-Chain Precursorsa
a Reagents and reaction conditions: (a) NaI (1.2 equiv), CH3CN, reflux,
24 h, 55-70%.
This synthetic strategy provided access to the dehydrocy-
clostellettamines D 2d and E 2e as a mixture of E/Z isomers.
Although their separation was possible by reversed phase HPLC
on a C-30 column in a very small scale, the amphiphilic nature
of cyclostellettamines makes impracticable the separation on a
synthetically useful scale.
With a ready supply of the bispyridinium intermediates 10
at hand, attention was directed to the crucial olefin metathesis.
Bispyridinium diene 10d, selected as model system, was heated
to reflux in dichloromethane in the presence of 5 mol % Grubbs
second generation ruthenium catalyst 1115 (Scheme 3) to afford
the expected macrocycle as a mixture of E/Z isomers. However,
even after extended reaction times (up to 72 h), complete
conversion of the starting material was not achieved, which
implied a tedious separation of products. Pursuing a complete
conversion of the open-chain diene, attention was turned to the
acceleration effect of microwaves.16 After some experimentation,
the macrocyclisation was taken to completion in 20 min,
working at constant microwave irradiation power (90 W).
Moreover, the amount of catalyst 11 could be reduced to 2 mol
%. At concentrations up to 0.02 M no oligomeric products,
expected as the result of a competitive intermolecular cross-
metathesis, were formed. Under the optimized conditions, no
differences in reaction times or in conversions were noticed upon
varying the ring size, and 29- to 33-membered bispyridinium
macrocycles were efficiently assembled. The E/Z mixture of
cyclic olefins was subjected without further purification to the
hydrogenation of the C-C double bond catalyzed by Pd on
charcoal to finally afford cyclostellettamines A 1a, B 1b, D
1d, E 1e, G 1g, and the unnatural cyclostelletamine 12 in almost
quantitative yield from the open-chain precursors 10b, 10e, 10c,
10f, 10a, and 10d, respectively. The spectroscopic data (1H and
13C NMR) of compounds 1a,b,d,e,g were in full agreement with
those reported for the natural materials.10
Biology
To assess the biological properties of the cyclostellettamines
and their acyclic precursors, we first performed an analysis of
cell cycle, differentiation, and apoptosis in the U937 acute
myeloid leukemia cell line.
As shown in Figure 2A, some compounds were able to
moderately increase differentiation after 30 h treatment, as
assessed by the analysis of the expression of CD11c, a known
marker of granulocytic differentiation. In particular, compounds
10a-d showed around 30% increase in the expression of CD11c
with compound 10b being the most potent at 5 µM. The analysis
of cell cycle, shown in Figure 2B, indicates that only particular
compounds were able to induce accumulation of the cells in
the G1 phase (black), with 10a showing the highest efficiency.
All open-chain bispyridinium dienes 10a-f were able to induce
apoptosis as measured by the percentage of active caspase 3
positive cells (Figure 2B, diffuse gray).
Since compounds 10a-f showed a consistent biological
activity, enzymatic assays were performed to evaluate their
action on HDAC1 and HDAC4, considered as markers for Class
I and Class II HDACs, respectively. Only diene 10b used at 5
µM concentration was able to reduce the activity of HDAC1
(Figure 3). Compound 10a on the contrary showed an inhibition
of HDAC4 (see Figure 3) comparable to SAHA (14, Chart 1),
whereas all other compounds showed either no (10b, 10c, 10e)

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2499
Figure 2. A. Differentiation analysis in U937 cells after 30 h treatment with the indicated compounds at 1-5 µM. The data shown represent the
media of independent quadruplicates. B. Cell cycle analysis and apoptosis in U937 cells treated with the indicated compounds for 30 h. The data
represent the media of independent duplicates.
Figure 3. HDAC1 (left) and HDAC4 (right) assay with the indicated compounds.
Scheme 3. Synthesis of Cyclostellettamines by Ring-Closure Metathesis of the Bispyridinium Dienes 10 Followed by Hydrogenation
of the Macrocyclic Olefins
or low inhibition (10d, 10f). IC50 values for HDAC1 are 1.88
µM for 10a and 0.53 µM for 10b, whereas IC50 values for
HDAC4 are 0.54 µM for 10a and >200 µM for 10b.
Intrigued by the possible class specificity exerted by com-
pounds 10a and 10b, the HDAC2 and HDAC3 enzymatic assays
were also performed. As demonstrated in Figure 4, neither of
the compounds showed inhibition in the in Vitro HDAC2 and
HDAC3 assays, whereas the reference SAHA (14, Chart 1)
showed a good inhibition of both enzymes. The o-anilidoben-
zamide MS-275 (15, Chart 1), already characterized as a Class
I HDAC-selective inhibitor,9a showed very low (ca. 30 times
lower than HDAC1) activity on HDAC3,17 indicating that its
major targets are HDAC1 and HDAC2. The lack of activity of
the bispyridinium diene 10a in three of the four Class I HDAC
enzymes, and its selective activity on HDAC4 renders it a
promising Class II HDAC-selective inhibitor.

2500 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Pe´rez-Balado et al.
Chart 1. Representative Histone Deacetylase (HDAC) Inhibitors
In order to assess the efficacy of the cyclostellettamines and
their synthetic precursors on their putative molecular targets,
their ability to alter acetylation levels at 5 µM concentration was
tested. As shown in Figure 5, neither compound altered the
extent of R-tubulin acetylation at 5 µM, whereas known HDAC
inhibitors (such as SAHA 14 and MS-275 15) and compounds
1b, 12, and 10a-f increased global histone H3 acetylation as
assessed by Western Blot analysis using antibodies to these
modifications. When the compounds were used at 1 µM, no
modification of the acetylation levels was detected (data not
shown). Interestingly, analog 10a did not increase R-tubulin
acetylation levels, thus confirming its unique features as a
subclass IIa inhibitor.
Finally, the capacity of these compounds to induce overex-
pression of p21WAF1/CIP1, a known target of class I- and pan-
HDAC inhibitors, was verified. As shown in Figure 6 by
Western Blot analysis, only compound 10b was able to weakly
induce p21WAF1/CIP1 expression. Compound 10b showed the most
potent HDAC1 inhibition among the series of compounds and
was inactive on HDAC2 and HDAC3. Given the very weak
up-regulation of p21WAF1/CIP1 by 10b (compare to SAHA 14
and MS-275 15 in Figure 6), it is tempting to suggest that
inhibition of both HDAC1 and HDAC2 is required for
p21WAF1/CIP1 up-regulation.
Taken together, our findings show that compound 10b
displays the features of an HDAC1-selective inhibitor, and this
activity correlates both with the histone H3 and R-tubulin
acetylation levels and with the p21WAF1/CIP1 induction. Moreover,
compound 10a show some Class IIa selectivity since it blocks
HDAC4 activity without altering HDAC6-mediated R-tubulin
deacetylation and does not modify HDAC1, HDAC2, and
HDAC3 actions in Vitro. Further investigations will be required
to fully determine the inhibitory profile of bispyridinium
diene 10a.
reversible acetylation of histones catalyzed by histone deacety-
lases (HDACs)/histone acetyl transferases (HATs) is involved
in transcriptional regulation. The attachment of acetyl groups
to the ꢀ-amino group of lysine residues in the histone tails alters
their charge and steric properties. As a consequence, both the
internucleosomal and the histone-DNA interactions with the
negatively charged phosphate backbone are destabilized. Un-
winding of the chromatin structure by acetylation enables access
of transcription factors and activation of gene expression.1,7
The eighteen known mammalian deacetylase enzymes are
divided in two families, the histone deacetylases (eleven
members) and the Sir2-like deacetylase or sirtuins (seven
members).18 HDACs appear to deacetylate the ꢀ-acetylamido
lysine by a nucleophilic attack of bound water to the Zn+2
-
activated carbonyl group, creating a tetrahedral zinc alkoxide
intermediate stabilized by enzyme residues which releases the
acetyl group and the lysine products. Activation of the water
molecule is caused by coordination to the Zn+2 (other metals
such as Co and Cu have been suggested19) and the Asp-His
charge relay system, although QM/MM studies suggest that the
Zn+2 prefers to be tetrahedral, and water is instead activated
by two His residues.20 Sirtuins, related to the yeast protein Sir2p,
function through a different mechanism requiring NAD+ as a
substrate and releasing O-acetyl-ADP ribose and nicotinamide
as a consequence of acetyl transfer.3
HDACs are further subdivided into two classes according to
their similarity in primary structure and size to Sacharomyces
cereVisiae deacetylases: Class I HDACs (HDACs 1-3, 8, 11)
are homologous to the yeast transcriptional regulator Rpd3
(reduced potasssium dependence 3) whereas Class II HDACs
(HDACs 4-7, 9, 10) bear similarity to Hda1.4b HDAC11 is
alternatively considered as a new Class IV.7a Differences
between Class I and II HDACs are primarily noted in their size
(with class II being from two to three times larger), their cellular
localization, the conservation of sequence motifs in their
catalytic domains, the identity of the protein-protein interaction
complexes, and their tissue distribution. The catalytic domain
of about 390 amino acids responsible for the deacetylation is
highly conserved among all HDACs, in particular the residues
lining the ligand-binding pocket. The crystal structures of the
Discussion
Acetylation/deacetylation reactions are critical to cell biologi-
cal processes ranging from chromatin remodeling to DNA-
binding activity, microtubule stabilization, protein-protein
interactions, and small-molecule action.8 In eukaryotic cells, the

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2501
Figure 6. Western Blot analysis of p21WAF1/CIP1 expression in U937
cells after treatment with the indicated compounds.
Most of the reported class-selective HDACs have been
discovered after screening natural and synthetic products and
chemical libraries,11 with focus on differential inhibition of
partially purified HDAC1 and HDAC6. These inhibitors (see
Chart 1 for a selection) range from simple compounds (valproic
acid, SAHA) to cyclic tetrapeptides called CHAPs, chimeras
composed of the macrocyclic ring of trapoxin and the hydrox-
amic acid of TSA 13. The majority of the described HDACs
are pan-inhibitors acting on Class I and Class II enzymes (SAHA
14, TSA 13, PXD101, LBH589, LAQ824, etc.) or Class I and
Class IIa (valproic acid, butyrate, trapoxin),7b and just a few
show either Class- or member-selectivity. Notably, MS-275 15
showed 40 000-fold selectivity for HDAC1 over HDAC6. MS-
275 is a specific inhibitor of HDAC1, 2 and a weak inhibitor
of HDAC317 (our unpublished data) whereas the IC50 for
HDAC8 is much higher. However, the discrimination between
HDAC1 and HDAC4 (Class II) is only modest. FK228 17
(another reported HDAC1, 2-selective inhibitor) behaves simi-
larly, with HDAC1/HDAC6 and HDAC1/HDAC4 values of 400
and 14, respectively.11 Trapoxin analogue Cyl2 16 exhibited
57 000-fold HDAC1/HDAC6 selectivity. Among the Class I
member-selective inhibitors, structures have been reported for
SB-429201 18 which preferentially inhibits HDAC1, SB397278A
19, an HDAC8-selective inhibitor,23 and the HDAC1 and
HDAC2- selective hydroxamic acids SK-7041 20 and SK-7069
21.24
Figure 4. HDAC2 and HDAC3 enzymatic assays with the selected
bis-pyridinium dienes 10a and 10b.
Class II-selective inhibitors are poorly represented. A family
of aryloxopropenylpyrrolyl hydroxamic acids, exemplified by
compound 22, has been discovered by Mai et al.25 A further
subdivision within this Class is based on the identification of
two deacetylase catalytic domains in Class IIb (HDAC6 and
HDAC10) whereas only one is present in Class IIa (HDACs 4,
5, 7, 9).26 Tubacin 23, a member of a DOS-based 1,3-dioxane
library, is a mammalian R-tubulin deacetylase specific inhibitor
that selectively binds to the tubulin deacetylase catalytic domain
of HDAC6 without altering the histone acetylation status, gene
expression, or cell-cycle progression.27a Other HDAC6-selective
Figure 5. Western Blot analyses of histone H3 and R-tubulin
acetylation carried out in U937 cells.
bacterial HDAC orthologue HDLP21a and the human HDAC8
in complexes with several natural (TSA, 13) and synthetic
inhibitors21b,c confirm the conserved nature of the aminoacids
in contact with the ligand. Class I and Class II HDACs differ
in the identity of the residues located at the rim of the entrance
channel. In particular Tyr91 of HDLP changes its position upon
ligand binding and is moreover poorly conserved among
HDACs.21a These differences can be exploited for the design
of class-selective HDAC inhibitors, although advances on the
structure-based design of selective inhibitors have been scarce.11
A recent crystal structure of HDAH (histone-deacetylase-like
amidohydrolase from Bordetella/alcaligens strain FB188), a
bacterial Class II HDAC homologue complexed to SAHA 14
and CypX, confirms the canonical fold of HDAC, the presence
of the catalytic Zn+2 ion, and the divergences of the aminoacids
of the binding pocket rim.22 In addition, the aryl residues of
Phe152 and Phe208 in the active site of HDAH are perfectly
coplanar and separated by 7.8 Å, suggesting that π-stacking
interaction with the ligand in this region might result in selective
inhibition of Class II HDACs. The internal cavity used as an
exit channel for the acetate leaving group after the deacetylation
reaction21b,c also shows contrasting differences between Class
I and Class II HDACs.22
inhibitors such as 24 have been built around the thiolate Zn+2
-
chelating group.27b
Whereas some of the HDAC-selective inhibitors are endowed
with a Zn+2-chelating group (hydroxamic acid, thiols, mercap-
toamides, trifluoromethylketones, etc.), thus pointing toward an
inhibitory mechanism comprising binding site occupancy and
competition with the natural substrate, others bear nonchelating
functionalities, and alternative inhibition mechanisms should be
entertained. For example, whereas modeling studies appear to
support the occupancy of the exit channel for acetate as an
inhibitory mechanism for the o-anilinobenzamide class repre-
sented by MS-275 15,28 its exact nature remains controversial.4b
Alternative mechanisms to the Zn+2-competitive chelation such
as the blocking of the entrance to the active site or to the acetate
exit channel, the presence of additional binding sites with
possible allosteric mode of actions and the inhibition of the
interaction of HDACs with the histones should be considered
in addition. In this regard, the macrocyclic bispyridinium salts

2502 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Pe´rez-Balado et al.
collectively known as cyclostellettamines 1 are intriguing
HDACs, since they are structurally unrelated to any of the
families of chemotypes known, and their structure cannot be
dissected into the traditional metal binding-linker-rim compo-
nents of most HDACs.10b Within the same family of natural
products, their activities vary with the size of the macrocycle.10b
These two aspects led us to consider the synthesis of the series
of cyclostellettamines in order to contribute to their structure-
activity relationship studies. For this purpose, we selected the
ring-closing metathesis reaction to construct the macrocycle.13
There are precedents on the ring-closing metathesis (RCM) of
large diene-containing chains.29 However, the use of pyridinium
salts as substrates for the RCM was unprecedented. We showed
that the RCM accelerated by microwaves is a highly efficient
method for the formation of bispyridinium macrocycles ranging
from 28 to 36 atoms in size. Combined with the catalytic
hydrogenation, precursor dienes 10 are converted in high yields
to the cyclostellettamines (the natural series 1, and the unnatural
congener 12), thus completing a novel general approach to this
family of natural products.
Characterization of these compounds confirmed the low
inhibitory potential of the cyclostellettamines (in the micromolar
range) with representative HDAC enzymes. However, the bis-
diene precursors 10 displayed interesting activities, in particular
10b and 10a. The former is a HDAC1-selective inhibitor, and
its activity correlates both with the histone H3 and R-tubulin
acetylation levels and with the up-regulation of the cyclin-
dependent kinase inhibitor p21WAF1/CIP1, a well-known target
of Class I HDAC inhibitors.9a Compound 10a showed the unique
profile of a subclass IIa-selective inhibitor, being able to block
HDAC4 activity without altering HDAC6-mediated R-tubulin
deacetylation and not modifying (if minimally) HDACs 1, 2,
and 3 action in Vitro. Compound 10b is therefore a promising
HDAC inhibitor showing a highly selective profile. Finally,
compounds 10a-f induced apoptosis in the myeloid monocytic
U937 leukemia cell line. Given that only some of these
compounds showed HDAC inhibitory properties, the induction
of apoptosis might also be due to activities on different
substrates. Indeed, both compounds 10a and 10b showed
induction of apoptosis in the U937 cells, despite the fact that
they target different HDAC enzymes.
HDAC inhibitory profile of both the cyclic bispyridinium
alkaloids and the acyclic diene precursors was determined. Two
of the open-chain bispyridinium dienes, namely 10b and 10a,
displayed unique profiles, since they are selective inhibitors of
HDAC1 and HDAC subclass IIa, respectively. Their contrasting
selectivities are most intriguing, since they differ exclusively
on one methylene carbon on the C3-alkyl chain of one
pyridinium ring. The discovery of these novel chemotypes will
pave the way for the development of new HDAC-selective
inhibitors with promising application for cancer treatment.
Experimental Section
General. Reagents and solvents were purchased as reagent-grade
and used without further purification unless otherwise stated.
Solvents were dried according to standard methods and distilled
before use. All reactions were performed in oven-dried or flame-
dried glassware under an inert atmosphere of Ar unless otherwise
stated. Chromatography refers to flash chromatography (FC) on
SiO2 60 (230-400 mesh) from Merck, head pressure of ca. 0.2
bar. TLC: UV254 SiO2-coated plates from Merck, visualization by
UV light (254 nm) or by coloring with 15% ethanolic phospho-
molybdic acid solution. NMR spectra were recorded in a Bruker
AMX400 (400.13 and 100.61 MHz for proton and carbon,
respectively) spectrometer at 298 K with residual solvent peaks as
internal reference and the chemical shifts are reported in δ [ppm],
coupling constants J are given in hertz and the multiplicities
assigned with DEPT experiments and expressed as follows: s )
singlet, d ) doublet, t ) triplet, q ) quartet, m ) multiplet. COSY,
HMBC, and HSQC methods were used to establish atom connec-
tivities. Electronic impact ionization (EI) and fast atom bombard-
ment (FAB) mass spectra were recorded on a VG-Autospec M
instrument. Electrospray ionization (ESI) mass spectra were
recorded on a Bruker APEX3 instrument. Infrared spectra (IR) were
obtained on a JASCO FT/IR-4200 infrared spectrometer. Peaks are
quoted in wave numbers (cm-1) and their relative intensities are
reported as follows: s ) strong, m ) medium, w ) weak.
Irradiation with microwaves was performed using a CEM DIS-
COVER apparatus.
General Procedure for the Preparation of the Open Chain
Bispyridinium Salts. NaI (291 mg, 1.95 mmol) was added to a
solution of 3-(13-chlorotridecyl)-1-(oct-7-enyl)pyridinium bromide
8b (719 mg, 1.62 mmol) and 3-(oct-7-enyl)pyridine 9c (369 mg,
1.95 mmol) in acetonitrile (10 mL). The resulting suspension was
heated at reflux until disappearance of the starting material, as
monitored by TLC. After completion of the reaction (usually 24
h), the solvent was removed under reduced pressure. The residue
was purified by flash chromatography (90:10, CH2Cl2-MeOH) to
afford 810 mg (69% yield) of 1-(oct-7-enyl)-3-(13-(3-(oct-7-enyl)-
pyridinium-1-yl)tridecyl)pyridinium chloride iodide 10f as a viscous
yellow oil.
Alkyl-substituted polyaminohydroxamic acids (PAHA) are
a new type of HDACs30a built around the structures of
spermidine and spermine, which are natural diamines known
to inhibit yeast HDAC.30b The finding that the compounds with
the highest activities in the PAHA series have a hydroxamic
acid end and secondary amines in their chains suggests that the
polyamine portion (which would be charged at physiological
pH) on its own might retain HDAC inhibitory activity either
by direct binding at the active site or by inhibition of the HDAC-
chromatin recognition events. Whether this last mechanism is
shared by the diene bispyridinium ions 10 remains to be
demonstrated, but further experiments aimed to get a better
understanding of the mechanism of HDAC inhibition by the
synthetic open-chain diene precursors are already ongoing.
1-(Oct-7-enyl)-3-{12-[3-(pent-4-enyl)pyridinium-1-yl]dodecyl}-
pyridinium Chloride Iodide (10a). 1H NMR (400 MHz, CD3OD)
δ 9.06 (br s, 2H), 8.9-8.8 (m, 2H), 8.50 (d, J ) 7.9 Hz, 2H), 8.04
(d, J ) 6.9 Hz, 2H), 5.9-5.7 (m, 2H), 5.1-4.9 (m, 4H), 4.68 (t, J
) 7.4 Hz, 4H), 3.0-2.9 (m, 4H), 2.2-2.1 (m, 2H), 2.1-2.0 (m,
6H), 1.9-1.8 (m, 2H), 1.8-1.7 (m, 2H), 1.5-1.2 (m, 22H) ppm.
13C NMR (100 MHz, CD3OD) δ 146.7 (2×), 145.6, 145.3, 143.4,
143.3, 139.8, 138.9, 129.0, 116.0, 115.0, 62.8 (2×), 34.6, 34.0,
33.5, 32.9, 32.5, 32.4, 31.5, 30.7, 30.5, 30.5 (3×), 30.4, 30.3, 30.0,
29.7, 29.5, 27.1, 26.9 ppm. IR (NaCl) ν 3017 (s), 2925 (s, C-H),
2853 (s, C-H), 1631 (m), 1504 (s), 1462 (s), 1153 (m), 911 (m),
731 (m), 689 (s). HRMS (ESI+) m/z calcd for C35H56IN2+ 631.3483
Summary and Conclusions
A new approach to the cyclostellettamines, macrocyclic
bispyridinium natural products, has been developed based on
the ring-closing metathesis of open-chain diene precursors.
Compared to the previous syntheses, this approach is shorter
and avoids the protection-deprotection sequence of one of the
pyridine subunits. In addition, it shows the great versatility of
the ring-closure metathesis to effect macrocyclisations. The
2+
(found 631.3490), C35H56N2 252.2216 (found 252.2219).
3-(Hex-5-enyl)-1-{12-[1-(oct-7-enyl)pyridinium-3-yl]dodecyl}-
1
pyridinium Chloride Iodide (10b). H NMR (400 MHz, CD3-
OD) δ 9.12 (br s, 2H), 8.9-8.8 (m, 2H), 8.48 (d, J ) 8.0 Hz, 2H),
8.03 (dd, J ) 8.0, 6.0 Hz, 2H), 5.9-5.7 (m, 2H), 5.1-4.9 (m, 4H),
4.66 (t, J ) 7.5 Hz, 4H), 2.9-2.8 (m, 4H), 2.2-2.1 (m, 2H), 2.1-
2.0 (m, 6H), 1.8-1.7 (m, 4H), 1.5-1.4 (m, 2H), 1.4-1.3 (m, 22H)

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2503
ppm. 13C NMR (100 MHz, CD3OD) δ 146.7 (2×), 145.7, 145.6,
145.3, 143.4, 143.3, 139.9, 139.5, 129.0, 115.3, 115.0, 62.9 (2×),
34.7, 34.4, 33.6, 33.4, 32.6, 32.5, 32.4, 31.6, 31.0, 30.7, 30.6, 30.6,
30.5 (2×), 30.4, 30.1 (2×), 29.8, 29.5, 29.4, 27.2, 27.0 ppm. IR
(NaCl) ν 3015 (s), 2925 (s, C-H), 2853 (s, C-H), 1633 (m), 1504
(s), 1463 (m), 1153 (m), 909 (m), 688 (s). HRMS (ESI+) m/z calcd
for C36H58IN2+ 645.3639 (found 645.3630), C36H58N22+ 259.2294
(found 259.2289).
General Procedure for Ring Closure Metathesis/Hydrogena-
tion. Grubbs catalyst 11 (3.2 mg, 0.0036, 2 mol %) was added to
a solution of 1-(oct-7-enyl)-3-{12-[3-(oct-7-enyl)pyridinium-1-yl]-
dodecyl}pyridinium chloride iodide 10c (134 mg, 0.188 mmol) in
CH2Cl2 (6 mL), under argon. The resulting solution was irradiated
with microwaves in a CEM-Discover apparatus in the constant
power mode (90 W) for 20 min. The solvent was removed under
reduced pressure. The macrocyclic product was dissolved in MeOH
(6 mL), and 10% Pd on charcoal (21 mg, 10 mol %) was added.
The mixture was heated at 32 °C under H2 atmosphere (30 psi) for
48 h. The suspension was filtered over Celite and washed with
MeOH (15 mL), and solvents were removed under reduced pressure.
The residue was purified by semipreparative HPLC on a C-30
Develosil column (CH3CN/H2O:60/30, 0.1% TFA as eluent, iso-
cratic mode) to afford 117 mg (91% yield) of cyclostellettamine D
1d as a solid.
precipitated with the HDAC4 and HDAC1 or with purified IgG
were pooled separately to homogenize all samples. Then, 10 µL of
the IP was incubated with a previously labeled 3H-Histone H4
peptide linked with streptavidine agarose beads (Upstate). In detail,
120 000 cpm of the H4-3H-acetyl-peptide was used for each tube
and incubated in 1× HDAC buffer with 10 µL of the sample in
the presence or absence of HDAC inhibitors with a final volume
of 200 µL. Those samples were incubated overnight at 37 °C in
slow rotation. On the next day, 50 µL of a quenching solution was
added, and 100 µL of the samples was counted in duplicate after
brief centrifugation in a scintillation counter. Experiments have been
carried out in quadruplicate.
Fluorimetric HDAC1, 2, 3 Assays. The HDAC Fluorescent
Activity Assay for HDAC1, 2, and 3 is based on the Fluor de Lys
Substrate and Developer combination (BioMol AK-500) and has
been carried out according to supplier’s instructions and as
previously reported.31 First, the Fluor de Lys Substrate, which
comprises an acetylated lysine side chain, has been incubated with
purified recombinant HDAC1, 2, or 3 enzymes in presence or
absence of the inhibitors. Briefly, for HDAC1, HDAC2, and HDAC
3, 100 ng, 25 ng, and 100 ng of recombinant proteins have been
used per assay, respectively. Full length HDAC1 and 2 with
C-terminal His tag were expressed using baculovirus expression
systems; a complex of human HDAC3 full length with C-terminal
tag and human NCOR2-N-terminal GST tag was coexpressed using
baculovirus expression systems. Deacetylation sensitizes the sub-
strate so that, in the second step, treatment with the developer
produces a fluorophore. Fluorescence has been quantified with a
TECAN Infinite M200 station.
IC50. For compounds 10a and 10b, the IC50 was calculated using
the in Vitro enzymatic assay described above (Bio Mol). For HDAC
4 in Vitro assay, amino acids 627-1085 with C-terminal GST tag
were expressed using baculovirus expression systems. A 2 µg per
assay has been used. The enzymatic incubation time chosen was
1-2 h. Four scalar concentrations in quadruplicate of the two
compounds were tested for their enzymatic inhibition potential on
HDAC1 and 4, and the relative IC50 was calculated using the
TECAN Infinite M200 apparatus with the Magellan software.
Cell Cycle Analysis on U937 Cells. Cells (2.5 × 105) were
collected and resuspended in 500 µL of hypotonic buffer (0.1%
Triton X-100, 0.1% sodium citrate, 50 µg/mL of propidium iodide
(PI), and RNAse A). Cells were incubated in the dark for 30 min.
Samples were acquired on a FACS-Calibur flow cytometer using
the Cell Quest software (Becton Dickinson) and analyzed with
standard procedures using the Cell Quest software (Becton Dick-
inson) and the ModFit LT version 3 Software (Verity) as previously
reported.9a All experiments were performed at least two times.
FACS Analysis of Apoptosis on U937 Cells. Apoptosis was
measured with the active caspase 3 detection (B-Bridge), and
samples were analyzed by FACS with Cell Quest technology
(Becton Dickinson) as previously reported.32
Cyclostellettamine D (1d). 1H NMR (400 MHz, CD3OD) δ 8.97
(br s, 2H, H2/2′), 8.86 (d, J ) 5.9 Hz, 2H, H6/6′), 8.46 (d, J ) 7.9
Hz, 2H, H4/4′), 8.03 (dd, J ) 7.9, 5.9 Hz 2H, H5/5′), 4.62 (t, J )
7.5 Hz, 4H, H7/7′), 2.9-2.8 (m, 4H, H18/20′), 2.0-1.9 (m, 4H,
H8/8′), 1.74 (t, J ) 7.1 Hz, 4H, H17/19′), 1.5-1.2 (m, 36H, H9-
16/9′-18′) ppm. 13C NMR (100 MHz, CD3OD) δ 146.7 (C4/4′),
145.9 (C3/3′), 145.3 (C2/2′), 143.4 (C6/6′), 129.0 (C5/5′), 63.0 (C7/
7′), 33.6 (C18/20′), 32.6 (C8/8′), 31.7 (C17/19′), 30.7, 30.6, 30.5,
30.4, 30.3 (C10-15/10′-17′), 30.2 (C16/18′), 27.3 (C9/9′) ppm. IR
(NaCl) ν 3020 (m), 2925 (s, C-H), 2852 (s, C-H), 1632 (m),
1505 (s), 1462 (m), 1155 (m), 692 (m). HRMS (ESI+) m/z calcd
for C36H60IN2+ 647.3796 (found 647.3789), C36H60ClN2+ 555.4439
2+
(found 555.4434), C36H60N2 260.2372 (found 260.2375).
Cell Lines and Cultures. The U937 cell line was cultured in
RPMI with 10% fetal calf serum, 100 U/mL of penicillin, 100 µg/
mL of streptomycin, 250 ng/mL of amphotericin-B, 10 mM HEPES,
and 2 mM glutamine. U937 cells were kept at the constant
concentration of 200 000 cells per milliliter of culture medium.
Human breast cancer ZR-75.1 cells were propagated in DMEM
medium supplemented with 10% fetal calf serum and antibiotics
(100 U/mL of penicillin, 100 µg/mL of streptomycin, and 250 ng/
mL of amphotericin-B).
Ligands and Materials. SAHA (kind gift of Merck) was
dissolved in DMSO and used at 5 µM. MS-275 (kind gift from
Schering AG) was dissolved in ethanol and used at 5 µM.
Cyclostellettamines and related compounds were dissolved in
DMSO and used at 1 or 5 µM.
Granulocytic Differentiation. Granulocytic differentiation was
carried out according to the published procedure.33 Briefly, U937
cells were harvested and resuspended in 10 µL of phycoerythrine-
conjugated CD11c (CD11c-PE). Control samples were incubated
with 10 µL of PE-conjugated mouse IgG1, incubated for 30 min at
4 °C in the dark, washed in PBS, and resuspended in 500 µL of
PBS containing propidium iodide (0.25 µg/mL). Samples were
analyzed by FACS with Cell Quest technology (Becton Dickinson).
Propidium iodide positive cells have been excluded from the
analysis.
Determination of r-Tubulin and Histone H3 Specific Acety-
lation. For R-tubulin, an amount of 50 µg of total protein extracts
was separated on 10% polyacrylamide gels and blotted. Seventy
four Western Blots were shown for acetylated R-tubulin (Sigma,
1:500 dilution), and total ERKs (Santa Cruz) were used to normalize
for equal loading. For histone H3 specific acetylations, an amount
of 100 µg of total protein extract was separated on 15% polyacry-
lamide gels and blotted.34 Western Blots were shown for pan-
acetylated histone H3 (Upstate), and total tubulin (Sigma) was used
to normalize for equal loading.
Cell-Based Human HDAC4 Assay. Cells ZR75.1 cells were
lysed in IP buffer (50 mM Tris-HCl at pH 7.0, 180 mM NaCl,
0.15% NP-40, 10% glycerol, 1.5 mM MgCl2, 1 mM NaMO4, and
0.5 mM NaF) with a protease inhibitor cocktail (Sigma), 1 mM
DTT, and 0.2 mM PMSF for 10 min in ice and centrifuged at 13
000 rpm for 30 min. Then, 1000 µg amounts of extracts were diluted
in IP buffer up to 1 mL and precleared by incubating with 20 µL
of A/G plus Agarose (Santa Cruz) for 30 min to 1 h on a rocking
table at 4 °C. Supernatants were transferred into a new tube, the
antibodies (around 3 to 4 µg) were added, and IP was allowed to
proceed overnight at 4 °C on a rocking table. The antibody used
for HDAC4 was from Sigma. As the negative control, the same
amounts of protein extracts were immunoprecipitated with the
corresponding purified IgG (Santa Cruz). On the next day, 20 µL
of A/G and Agarose (Santa Cruz) were added to each IP, and
incubation was continued for 2 h. The beads were recovered by
brief centrifugation and washed with cold IP buffer several times.
The samples were than washed twice in PBS and resuspended in
20 µL of sterile PBS. The HDAC assay was carried out according
to the supplier’s instructions (Upstate). Briefly, samples immuno-

2504 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10
Pe´rez-Balado et al.
Determination of p21WAF1/CIP1 Induction in U937 Cells. Total
protein extracts (100 µg) were separated on a 15% polyacrylamide
gel and blotted as previously described.9a Western Blots were shown
for p21 (Transduction Laboratories, 1:500 dilution), and total ERKs
(Santa Cruz) were used to normalize for equal loading.
(11) Miller, T. A.; Witter, D. J.; Belvedere, S. Histone Deacetylase
Inhibitors. J. Med. Chem. 2003, 46, 5097-5114.
(12) (a) Anan, H.; Seki, N.; Noshiro, O.; Honda, K.; Yasumuro, K.; Ozasa,
T.; Fusetani, N. Total synthesis of cyclostellettamine C, a bispyri-
dinium macrocyclic alkaloid having muscarinic acetylcholine
receptor antagonistic activity. Tetrahedron 1996, 52, 10849-
10860. (b) Baldwin, J. E.; Spring, D. R.; Atkinson, C. E.;
Lee, V. Efficient synthesis of the sponge alkaloids cyclostellettamines
A-F. Tetrahedron 1998, 54, 13655-13680. (c) Grube, A.;
Timm, C.; Koeck, M. Synthesis and mass spectrometric analysis of
cyclostellettamines H, I, K and L. Eur. J. Org. Chem. 2006, 1285-
1295.
(13) Nicolaou, K. C.; Bulger, P. G.; Sarlah, D. Metathesis Reactions in
Total Synthesis. Angew. Chem., Int. Ed. 2005, 44, 4490-4527.
(14) Davies-Coleman, M. T.; Faulkner, D. J.; Dubowchik, G. M.; Roth,
G. P.; Polson, C.; Fairchild, C. J. A new EGF-active polymeric
pyridinium alkaloid from the sponge Callyspongia fibrosa. J. Org.
Chem. 1993, 58, 5925-5930.
Acknowledgment. The authors are grateful to the European
Union (EPITRON, LSHC-CT-2005-518417), Xunta de Galicia
(visiting scientist fellowship to C.P.B.), the Spanish Ministerio
de Educacio´n y Ciencia (SAF04-07131, FEDER) and the
Italian Ministero dell Istruzione Universita` e Ricerca (PRIN
2006052835_003) for financial support. A.N., M. M. and A.
M. are supported by EU (LSHC-CT-2005-518417). We specially
thank Prof. Nobuhiro Fusetani and Dr. Naoya Oku for insightful
discussions on separation of dehydrocyclostellettamines.
(15) Trnka, T. M.; Grubbs, R. H. The development of L2X2Ru:CHR olefin
metathesis catalysts: an organometallic success story. Acc. Chem.
Res. 2001, 34, 18-29.
(16) Nosse, B.; Schall, A.; Jeong, B. W.; Reiser, O. Optimization of ring-
closing metathesis: Inert gas sparging and microwave irradiation.
AdV. Synth. Cat. 2005, 347, 1869-1874.
Supporting Information Available: Experimental procedures,
analytical and spectral characterization data for compounds 4-8b,
9a-c, 10c-f, 1a, 1b, 1e, 1g, and 12. This material is available
free of charge via the Internet at http://pubs.acs.org.
(17) Inoue, S.; Mai, A.; Dyer, M. J.; Cohen, G. M. Inhibition of histone
deacetylase class I but not class II is critical for the sensitization of
leukemic cells to tumor necrosis factor-related apoptosis-inducing
ligand-induced apoptosis. Cancer Res. 2006, 66, 6785-6792.
(18) de Ruijter, A. J. M.; ven Gennip, A. H.; Caron, H. N.; Kemp, S.;
Kuilenburg, A. B. P. Histone deacetylases (HDACs). Characterization
of the classical HDAC family. Biochem. J. 2003, 370, 737-749.
(19) Gantt, S. L.; Gattis, S. G.; Fierke, C. A. Catalytic Activity and
Inhibition of Human Histone Deacetylase 8 Is Dependent on the
Identity of the Active Site Metal Ion. Biochemistry 2006, 45, 6170-
6178.
(20) Corminboeuf, C.; Hu, P.; Tuckerman, M. E.; Zhang, Y. Unexpected
Deacetylation Mechanism Suggested by a Density Functional Theory
QM/MM Study of Histone-Deacetylase-Like Protein. J. Am. Chem.
Soc. 2006, 128, 4530-4531.
(21) (a) Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.; Rifkind,
R. A.; Marks, P. A.; Pavletich, N. P. Structures of a Histone
Deacetylase Homologue Bound to the TSA and SAHA inhibitors.
Nature 1999, 401, 188-193. (b) Vannini, A.; Volpari, C.; Filocamo,
G.; Casavola, E. C.; Brunetti, M.; Renzoni, D.; Chakravarty, P.;
Paolini, C.; De Francesco, R.; Gallinari, P.; Steinku¨hler, C.;
Di Marco, S. Crystal structure of a eukatyotic zinc-dependent
References
(1) (a) Baylin, S. B.; Ohm, J. E. Epigenetic Gene Silencing in Cancer-
A Mechanism for Early Oncogenic Pathway Addiction? Nature ReV.
Cancer 2006, 6, 107-116. (b) Yoo, C. B.; Jones, P. A. Epigenetic
therapy of cancer: past, present and future. Nat. ReV. Drug DiscoVery
2006, 5, 37-50.
(2) Marmorstein, R. Protein Modules that Manipulate Histone Tails for
Chromatin Regulation. Nat. ReV. Mol. Cell Biol. 2001, 2, 422-432.
(3) Biel, M.; Wascholowski, V.; Giannis, A. Epigenetics-An Epicenter
of Gene Regulation: Histones and Histone-Modifying Enzymes.
Angew. Chem., Int. Ed. 2005, 44, 3186-3216.
(4) (a) Kouzarides, T. Acetylation: a regulatory modification to rival
phosphorylation. EMBO J. 2000, 19, 1176-1179. (b) Mai, A.; Massa,
S.; Rotili, D.; Cerbara, I.; Valente, S.; Pezzi, R.; Simeoni, S.; Ragno,
R. Histone Deacetylation in Epigenetics: An Attractive Target for
Anticancer Therapy. Med. Res. ReV. 2005, 25, 261-309.
(5) (a) Bannister, A. J.; Kouzarides, T. Reversing histone methylation.
Nature 2005, 436, 1103-1106. (b) Bedford, M. T.; Richard,
S. Arginine Methylation: An Emerging Regulator of Protein
Function. Cell 2005, 18, 263-272. (c) Jenuwein, T. The epigenetic
magic of histone lysine methylation. FEBS J. 2006, 273, 3121-
3135.
(6) (a) Das, P. M.; Singal, R. DNA methylation and cancer. J. Clin.
Oncol. 2004, 22, 4632-4642. (b) Schneider, J.; Shilatifard, A.
Histone Demethylation by Hydroxylation: Chemistry in Action. ACS
Chem. Biol. 2006, 1, 75-81.
(7) (a) Minucci, S.; Pelicci, P. G. Histone deacetylase inhibitors and the
promise of epigenetic (and more) treatments for cancer. Nat. ReV.
Cancer 2006, 6, 38-51. (b) Bolden, J. E.; Peart, M. J.; Johnstone,
R. W. Anticancer activities of histone deacetylase inhibitors. Nat.
ReV. Drug DiscoVery 2006, 5, 769-784.
(8) Grozinger, C. M.; Schreiber, S. L. Deacetylase Enzymes: Biological
Functions and the Use of Small Molecule Inhibitors. Chem. Biol.
2002, 9, 3-16.
(9) (a) Nebbioso, A.; Clarke, N.; Bontempo, P.; Volz, E.; Alvarez, R.;
Schiavone, E. M.; Weisz, A.; Bresciani, F.; de Lera, A. R.;
Gronemeyer, H.; Altucci, L. Tumor-selective action of HDAC
Inhibitors involves TRAIL induction in acute myeloid leukemia cells.
Nat. Med. 2005, 11, 77-84. (b) Insinga, A.; Monestiroli, S.; Ronzoni,
S.; Gelmetti, V.; Marchesi, F.; Viale, A.; Altucci, L.; Nervi, C.;
Minucci, S.; Pelicci, P. G. Inhibitors of histone deacetylases induce
tumor-selective apoptosis through activation of the death receptor
pathway. Nat. Med. 2005, 11, 71-76.
(10) (a) Fusetani, N.; Asai, N.; Matsunaga, S.; Honda, K.; Yasumuro, K.
Bioactive marine metabolites. 59. Cyclostellettamines A-F, pyridine
alkaloids which inhibit binding of methyl quinuclidinyl benzylate
(QNB) to muscarinic acetylcholine receptors, from the marine sponge,
Stelletta maxima. Tetrahedron Lett. 1994, 35, 3967-3970. (b) Oku,
N.; Nagai, K.; Shindoh, N.; Terada, Y.; Van Soest, R. W. M.;
Matsunaga, S.; Fusetani, N. Three new cyclostellettamines, which
inhibit histone deacetylase, from a marine sponge of the genus
Xestospongia. Bioorg. Med. Chem. Lett. 2004, 14, 2617-2620.
(c) De Oliveira, J. H. H. L.; Grube, A.; Koeck, M.; Berlinck, R. G.
S.; Macedo, M. L.; Ferreira, A. G.; Hajdu, E. Ingenamine G and
cyclostellettamines G-I, K, and L from the new Brazilian species
of marine sponge Pachychalina sp. J. Nat. Prod. 2004, 67, 1687-
1689.
histone deacetylase, human HDAC8, complexed with
a
hydroxamic acid inhibitor. Proc. Natl. Acad. Sci. U.S.A. 2004,
101, 15064-15069. (c) Somoza, J. R.; Skene, R. J.; Katz, B. A.;
Mol, C.; Ho, J. D.; Jennings, A. J.; Luong, C.; Arvai, A.; Buggy, J.
J.; Chi, E.; Tang, J.; Sang, B.-C.; Verner, E.; Wynands, D. R.;
Snell, G.; Navre, M.; Knuth, M. W.; Swanson, R. V.; McRee, D. E.;
Tari, L. W. Structural Snapshots of Human HDAC8 Provide Insights
Into the Class I Histone Deacetylase. Structure 2004, 12, 1325-
1334.
(22) Nielsen, T. K.; Hildmann, C.; Dickmanns, A.; Schwienhorst, A.;
Ficner, R. Crystal structure of a Bacterial Class 2 Histone Deacetylase
Homologue. J. Mol. Biol. 2005, 354, 107-120.
(23) Hu, E.; Dul, E.; Sung, C.-M.; Chen, Z.; Kirkpatrick, R.; Zhang, G.-
F.; Johanson, K.; Liu, R.; Lago, A.; Hofmann, G.; Macarron, R.;
Perez, P.; Krawiec, J.; Winkler, J.; Jaye, M. Identification of Novel
Isoform-Selective Inhibitors within Class I Histone Deacetylases. J.
Pharmacol. Exp. Ther. 2003, 307, 720-728.
(24) Park, J.-H.; Jung, Y. J.; Kim, T. Y.; Kim, S. G.; Jong, H.-S.; Lee, J.
W.; Kim, D.-K.; Lee, J.-S.; Kim, N. K.; Kim, T.-Y.; Bang, Y.-J.
Class I Histone Deacetylase-Selective Novel Synthetic Inhibitors
Potently Inhibit Human Tumor Proliferation. Clin. Cancer Res. 2004,
10, 5271-5281.
(25) (a) Mai, A.; Massa, S.; Pezzi, R.; Rotili, D.; Loidl, P.; Brosch, G.
Discovery of (Aryloxopropenyl)pyrrolyl Hydroxamides as Selective
Inhibitors of Class IIa Histone Deacetylase Homologue HD1-A. J.
Med. Chem. 2003, 46, 4826-4829. (b) Mai, A.; Massa, S.; Valente,
S.; Simeoni, S.; Ragno, R.; Bottoni, P.; Scatena, R.; Brosch, G. Aroyl-
pyrrolyl hydroxyamides: influence of pyrrole C4-phenylacetyl
substitution on histone deacetylase inhibition. ChemMedChem 2006,
1, 225-237.
(26) (a) Grozinger, C. M.; Hassig, C. A.; Schreiber, S. L. Three proteins
define a class of human histone deacetylases related to yeast Hda1p.

Histone Deacetylase Inhibitors
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 10 2505
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4868-4873. (b) Fischle, W.;
Kiermer, V.; Dequiedt, F.; Verdin, E. The emerging role of class II
deacetylases. Biochem. Cell Biol. 2001, 79, 337-348. (c) Bertos, N.
R.; Wang, A. H.; Yang, X. Y. Class II Histone Deacetylases:
Structure, Function and Regulation. Biochem. Cell Biol. 2001, 79,
243-252.
Chroneos, Z. C.; Nelson, D. A. Polyamines inhibit the yeast histone
deacetylase. FEBS Lett. 1987, 220, 79-83.
(a) Kapustin, G. V.; Feje´r, G.; Gronlund, J. L.; McCafferty, D. G.;
Seto, E.; Etzkorn, F. A. Phosphorus-Based SAHA Analogues as
Histone Deacetylase Inhibitors. Org. Lett. 2003, 5, 3053-3056. (b)
Gurvich, N.; Tsygankova, O. M.; Meinkoth, J. L.; Klein, P. S. Histone
Deacetylase Is a Target of Valproic Acid-Mediated Cellular Dif-
ferentiation. Cancer Res. 2004, 64, 1079-1086.
(a) Mai, A.; Massa, A.; Rotili, D.; Simeoni, S.; Ragno, R.; Botta,
G.; Nebbioso, A.; Miceli, M.; Altucci, L.; Brosch, G. Synthesis and
Biological Properties of Novel, Uracil-Containing Histone Deacety-
lase Inhibitors. J. Med. Chem. 2006, 49, 6046-6056.
Altucci, L.; Rossin, A.; Raffelsberger, W.; Reitmair, A.; Chomienne,
C.; Gronemeyer, H. Retinoic acid-induced apoptosis in leukemia cells
is mediated by paracrine action of tumor-selective death ligand
TRAIL. Nat. Med. 2001, 7, 680-686.
Mai, A.; Massa, A.; Pezzi, R.; Simeoni, S.; Rotili, D.; Nebbioso, A.;
Scognamigliio, A.; Altucci, L.; Loidl, P.; Brosch, G. Class II (IIa)-
Selective Histone Deacetylase Inhibitors. 1. Synthesis and Biological
Evaluation of Novel (Aryloxopropenyl)pyrrolyl Hydroxamides. J.
Med. Chem. 2005, 48, 3344-3353.
(27) (a) Haggarty, S. J.; Koeller, K. M.; Grozinger, C. M.; Schreiber, S.
L. Domain-selective Small-molecule Inhibitor of Histone Deacetylase
6 (HDAC-6)-mediated Tubulin Acetylation. Proc. Natl. Acad. Sci.
U.S.A. 2003, 100, 4389-4394. (b) Suzuki, T.; Kouketsu, A.; Itoh,
Y.; Hisakawa, S.; Maeda, T.; Yoshida, M.; Nakagawa, H.; Miyata,
N. Highly Potent and Selective Histone Deacetylase 6 Inhibitors
Designed Based on a Small-Molecule Substrate. J. Med. Chem. 2006,
49, 4809-4812.
(28) Wang, D.-F.; Helquist, P.; Wiech, N. L.; Wiest, O. Toward Selective
Histone Deacetylase Inhibitor Design: Homology Modeling, Docking
Studies, and Molecular Dynamics Simulations of Human Class I
Histone Deacetylase. J. Med. Chem. 2005, 48, 6936-6947.
(29) Ibrahim, Y. A.; John, E. Efficient synthesis of 18-40 membered
macrocyclic polyoxadiamides and polyoxatetraamides via ring closing
metathesis. Tetrahedron Lett. 2006, 62, 1001-1014.
(30) (a) Varghese, S.; Gupta, D.; Baran, T.; Jiemjit, A.; Gore, S. D.;
Casero, R. A.; Woster, P. M. Alkyl-Substituted Polyhydroxamic
Acids: A Novel Class of Targeted Histone Deacetylase Inhibitors.
J. Med. Chem. 2005, 48, 6350-6365. (b) Vu, Q. A.; Zhang, D.;
JM070028M