Epigenetic profiling of the antitumor natural product psammaplin A and its analogues

Article abstract of DOI:10.1016/j.bmc.2010.12.026

A collection of analogues of the dimeric natural product psammaplin A that differ in the substitution on the (halo)tyrosine aryl ring, the oxime and the diamine connection has been synthesized. The effects on cell cycle, induction of differentiation and apoptosis of the natural-product inspired series were measured on the human leukaemia U937 cell line. Epigenetic profiling included induction of p21^WAF1, effects on global H3 histone and tubulin acetylation levels as well as in vitro enzymatic assays using HDAC1, DNMT1, DNMT3A, SIRT1 and a peptide domain with p300/CBP HAT activity. Whereas the derivatives of psammaplin A with modifications in the length of the connecting chain, the oxime bond and the disulfide unit showed lower potency, the analogues with changes on the bromotyrosine ring exhibited activities comparable to those of the parent compound in the inhibition of HDAC1 and in the induction of apoptosis. The lack of HDAC1 activity of analogues modified on the disulfide bond suggests that its cleavage must occur in cells to produce the monomeric Zn²⁺-chelating thiol. This assumption is consistent with the molecular modelling of the complex of psammaplin A thiol with h-HDAC8. Only a weak inhibition of DNMT1, DNMT3A and residual activities with SIRT1 and a p300/CBP HAT peptide were measured for these compounds.

Full text of DOI:10.1016/j.bmc.2010.12.026

Bioorganic & Medicinal Chemistry 19 (2011) 3637–3649
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Epigenetic profiling of the antitumor natural product psammaplin A
and its analogues
José García a, Gianluigi Franci b,f, Raquel Pereira a, Rosaria Benedetti b,c,e, Angela Nebbioso b,
a,
Fátima Rodríguez-Barrios a, Hinrich Gronemeyer d, Lucia Altucci b,f, , Angel R. de Lera
⇑
⇑
a Departamento de Química Orgánica, Universidade de Vigo, 36310 Vigo, Spain
b Dipartimento di Patologia generale, Seconda Università di Napoli, Vico L. De Crecchio 7, 80138 Napoli, Italy
c Università di Napoli Federico II, Dipartimento di Chimica Organica e Biochimica, Italy
d Department of Cancer Biology - Institut de Génétique et de Biologie Moléculaire et Cellulaire (IGBMC)/CNRS/INSERM/ULP, BP 163, 67404 Illkirch Cedex, C. U. de Strasbourg, France
e Università di Napoli Federico II, Dipartimento di Fisica, 6, CNR-IGB, Via P. Castellino, 80100, Napoli, , Italy
f CNR-IGB, Via Pietro Castellino 80100 Napoli, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A collection of analogues of the dimeric natural product psammaplin A that differ in the substitution on
the (halo)tyrosine aryl ring, the oxime and the diamine connection has been synthesized. The effects on
cell cycle, induction of differentiation and apoptosis of the natural-product inspired series were measured
on the human leukaemia U937 cell line. Epigenetic profiling included induction of p21WAF1, effects on glo-
bal H3 histone and tubulin acetylation levels as well as in vitro enzymatic assays using HDAC1, DNMT1,
DNMT3A, SIRT1 and a peptide domain with p300/CBP HAT activity. Whereas the derivatives of psamma-
plin A with modifications in the length of the connecting chain, the oxime bond and the disulfide unit
showed lower potency, the analogues with changes on the bromotyrosine ring exhibited activities com-
parable to those of the parent compound in the inhibition of HDAC1 and in the induction of apoptosis.
The lack of HDAC1 activity of analogues modified on the disulfide bond suggests that its cleavage must
occur in cells to produce the monomeric Zn2+-chelating thiol. This assumption is consistent with the
molecular modelling of the complex of psammaplin A thiol with h-HDAC8. Only a weak inhibition of
DNMT1, DNMT3A and residual activities with SIRT1 and a p300/CBP HAT peptide were measured for
these compounds.
Received 30 September 2010
Accepted 8 December 2010
Available online 15 December 2010
Keywords:
Psammaplin A
Epigenetics
HDAC
DNMT
Total synthesis
Natural products
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
(APN).21 Targeting multiple proteins that impact on DNA topology,
DNA replication, transcription, apoptosis, tumor invasion, and
Psammaplin A (1, PsA, Scheme 1) is a symmetrical disulfide di-
mer derived from the condensation of modified tyrosine and cys-
teine units1 (for a proposal of its biogenesis, see 2,3). Although it
was first isolated in 1987 from an unidentified sponge4 and from
Psammaplysilla sp.,5,6 PsA is found, together with biogenetically-re-
lated congeners, in several Verongidas2,3,6–10 and in some associa-
tions of these species.11,12
angiogenesis likely contributes to the significant cytotoxicity dis-
played by PsA in human lung (A549), ovarian (SKOV- 3), skin
(SK-MEL-2), CNS (XF498), and colon (HCT15) cancer cell lines.12
PsA was also reported to inhibit in vitro the chromatin-modify-
ing enzymes histone deacetylase (HDAC) and DNA methyl transfer-
ase (DNMT).3 HDAC and DNMT are epigenetic enzymes that
catalyze the covalent modifications of histone proteins and DNA
in chromatin,22 and therefore they are considered prime new tar-
gets23 for the treatment and prevention of cancer24–34 and other
diseases.35 The modification of chromatin is thus an addition to
the list of potential mechanisms involved in the anticancer actions
of the natural product PsA.
Antibacterial and antitumor activities have been reported for
PsA. The in vitro antibacterial activity of PsA against both Staphylo-
coccus aureus (SA) and methicillin-resistant Staphylococcus aureus
(MRSA) was considered the result of the inhibition of DNA gyrase13
and induced arrest of bacterial DNA synthesis. PsA also inhibits
topoisomerase II (topo II),14 farnesyl protein transferase,8 leucine
aminopeptidase,8 mycothiol-S-conjugate amidase,15 chitinase,10
Histone acetylation is a dynamic process in which a cellular
steady state is maintained by the opposing activities of histone
acetyltransferases (HATs) and deacetylase enzymes (HDAC) acting
Pol
a
-primase,16 PPARc17–20 and mammalian aminopeptidase N
at the e-amino groups of evolutionally conserved lysine residues lo-
cated at the histone N-termini. HATs transfer the acetyl moiety
from acetyl CoA to the histone lysine residues whereas HDACs cat-
alyze their removal. Individual HATs and HDACs display distinct
⇑
Corresponding authors.
E-mail addresses: lucia.altucci@unina2.it (L. Altucci), qolera@uvigo.es (Angel R.
de Lera).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.12.026

3638
J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
O
O
the corresponding carboxylic acid with the symmetrical diamine
cystamine. Both Hoshino et al.53 and Nicolaou et al.54,55 installed
the oxime function after the synthesis of the corresponding pyruvic
i
OMe
OMe
OMe
NH2
N
HO
HO
HO
3
acid derived from L-tyrosine, in a sequence that afforded 1 in mod-
2
erate overall yields. Nicolaou then screened in antibacterial assays
a 3828-membered library of heterodimeric psammaplin A ana-
logues15 obtained from symmetrical precursors by combinatorial
scrambling via catalytically-induced disulfide exchange reactions.
A recently described three-step (43% overall yield) synthesis of
PsA 1 starts from the considerably more expensive and less versa-
tile 4-hydroxyphenylpyruvic acid.56
ii
O
O
Br
Br
iii
OH
N
N
HO
HO
HO
HO
5
4
iv
OH
OH
Br
We modified the above synthesis of 1 by performing the bro-
mination of the phenol ring on the corresponding oxime 3 (Scheme
O
N
H
Br
S
N
N
H
S
1).57 As shown on Scheme 1, the amino group of
L-tyrosine 2 was
N
O
HO
HO
oxidized to the hydroxyimino ester 3 in 60% yield using Na2WO4
and H2O2 in ethanol.58 A monobrominated product 4 was obtained
by treatment of 3 with one equivalent of NBS in CH3CN at
25 °C,58,59 with no evidence of formation of the dibromo derivative
or the dibrominated spirocyclic isoxazoline.57 After saponification
of 4 the carboxylic acid 5 was coupled with cystamine using Hosh-
ino’s conditions.53 The overall yield for the synthetic sequence is
29%, with a slight improvement over the two routes previously de-
1 Psammaplin A
Scheme 1. Reagents and conditions: (i) Na2WO4, H2O2, 25 °C, 3 h (63%); (ii) NBS,
CH3CN, 25 °C, 2 h (82%); (iii) LiOH, THF–H2O, 23 °C, 12 h (99%); (iv) DCC, N-
hydroxyphthalimide, Et3N, cystamine, 1,4-dioxane, MeOH, 25 °C, 12 h (60%).
specificities for certain individual lysine residues and particular
histones. Although the factors responsible for the specificity are
poorly understood they might reflect different biological functions
of the various enzymes. Regardless of the details, it is widely
accepted that histone acetylation is essential to establish a trans-
criptionally competent state of chromatin22 and consequently
contributes to the gene activation/gene repression transcriptional
status of cells. Two histone deacetylase inhibitors (HDACis), sube-
roylanilide hydroxamic acid (SAHA, vorinostat, ZolinzaÒ), and
FK228 (romidepsin, IstodaxÒ) are used as therapy for cutaneous
T-cell lymphoma,36 and several others are currently undergoing
clinical trials as potential targeted cancer chemotherapeutic
agents.37–41
scribed from L-tyrosine 2. The synthetic scheme is advantageous
for the preparation of diverse PsA analogues starting from com-
mercial tyrosine derivatives.
Using the methodology depicted on Scheme 1, the synthesis of
PsA homologues containing from three to six methylene units was
completed, albeit in low yields (Scheme 2), using the non-commer-
cial diamines 6a–d, which were synthesized following a general
methodology.60
Similarly, PsA analogues that preserve the native connecting
cystamine unit but differ in the substituents at the aryl ring
(Scheme 3) were synthesized using either commercially available
tyrosine derivatives or the synthetic halogenated analogues (with
bromine or iodine at C3) after oxidation of the amine to the oxime
(Scheme 3). Some compounds of the series (11a, 11c, 11d and 11e)
have previously been described by Nicolaou on his search for new
antibacterial agents.55 Bromopsammaplin A 11f is also a natural
product isolated from an association of the sponges Jaspis wondo-
ensis and Poecillastra wondoensis.12
Condensation of acid 5 with amines 12 and 14a–c (Scheme 4)
provided analogues 13 and 15a–c which were designed to further
our understanding of the mechanism underlying the biological
activity of the parent PsA as HDAC inhibitor. The dimer 13 contains
an ethylene group replacing the disulfide bond functionality. Prod-
ucts 15b,c are monomers that have, respectively, methyl ether and
methyl sulfide as end groups. The primary alcohol 16 was obtained
by acidic (7:2:1 THF/HCO2H/H2O) deprotection of silyl ether 15a.
The derivative lacking the oxime was also of interest to reveal
whether this functional group is critical for the epigenetic activity
of 1. Compound 20 was prepared as described on Scheme 5. The
bromination of methyl 3-(4-hydroxyphenyl)propanoate using the
In eukaryotes DNMTs catalyze the addition of methyl groups
from S-adenosyl-L-methionine (SAM) to the C5 position of cytosine
bases within the CpG-rich islands in DNA. Methylation of DNA is an
epigenetic mark associated to a repressed chromatin state which
inhibits gene transcription.42,43 Several tumour supressor genes
are hypermethylated in tumours, which suggest a link between
aberrant DNA methylation and cancer.44–47 DNA methyl transfer-
ase inhibitors (DNMTis)48 structurally related to cytidine (5-aza-
cytidine, VidazaÒ and 5-aza-deoxycytidine, DacogenÒ) are already
in the clinic for the treatment of myelodysplastic syndrome.49
Since histone acetylation and DNA methylation play a key role
in the pathophysiology of cells, dual inhibitors of HDAC or DNMT
are therapeutically more appealing than combination of these
drugs.50,51 We thus became intrigued by the reports on the potent
activity of psammaplin A (1) in the inhibition of these two epige-
netic enzymes (HDAC: IC50 = 4.2 nM; DNMT: IC50 = 18.6 nM using
in vitro cell-free enzyme assays)3 as well as by the in vitro and
in vivo inhibition of tumour growth induced by this natural prod-
uct.20 Despite the reports indicating some drawbacks for the devel-
opment of 1 as a drug due to its poor physiological stability,15,21,52
we undertook the synthesis of a family of analogues with the aim
to discover more potent and selective derivatives of 1 as well as to
shed light into the mechanism of epigenetic inhibition by 1. Apart
from the work of Nicolaou focused on antibacterial activities, no
structure–activity relationship studies of the anticancer activities
of PsA analogues have been reported, which are necessary for an
eventual lead optimization project within this class of modified
tetrapeptides.
O
Br
OH
n
S
NH2
n
N
H2N
i
S
HO
HO
HO
6a n = 3
6b n = 4
6c n = 5
6d n = 6
5
O
Br
N
H
S
7a n = 3
n
2
7b n = 4
7c n = 5
7d n = 6
N
HO
2. Chemistry. Synthesis of psammaplin A and derivatives
All previous synthetic approaches to 1 have focused on the final
construction of the dimeric disulfide structure by condensation of
Scheme 2. Reagents and conditions: (i) DCC, N-hydroxyphthalimide, Et3N, diamine
6, 1,4-dioxane, MeOH, 25 °C, 12 h (7a, 16%; 7b, 21%; 7c, 34%; 7d, 14%).

J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
3639
O
O
O
X
X
i
iv
X
OMe
OH
OMe
R
R
R
N
N
NH2
HO
HO
3 R = 4-OH, X = H
9a R = 3-OH, X = H
9b R = H, X = H
10a R = 4-OH, X = H
10b R = 3-OH, X = H
10c R = H, X = H
10d R = 4-OH, X = F
10e R = 4-OH, X = Cl
10f R = 4-OH, X =3,5-Br
10g R = 4-OH, X =3-I
2
R = 4-OH, X = H
8a R = 3-OH, X = H
8b R = H, X = H
ii
9c R = 4-OH, X = F
9d R = 4-OH, X = Cl
iii
8c R = 4-OH, X = F
8d R = 4-OH, X = Cl
v
9e R = 4-OH, X = 3,5-Br
9f R = 4-OH, X = 3-I
9g R =4-OH, X = 3,5-I
X
OH
O
N
H
R
X
11a R = 4-OH, X = H
11b R = 3-OH, X = H
11c R = H, X = H
S
N
N
S
R
H
N
O
HO
11d R = 4-OH, X = F
11e R = 4-OH, X = Cl
11f R = 4-OH, X =3,5-Br
11g R = 4-OH, X = 3-I
Scheme 3. Reagents and conditions: (i) Na2WO4, H2O2, 25 °C, 3 h (9a, 57%; 9b, 50%; 9c, 64%; 9d, 80%); (ii) NBS, CH3CN, 25 °C, 5 h (9e, 80%); (iii) I2, Ag2SO4, MeOH, 25 °C,
15 min (9f, 51%; 9g, 23%); (iv) LiOH, THF–H2O, 23 °C, 12 h (10a, 99%; 10b, 99%; 10c, 99%; 10d, 97%; 10e, 98%; 10f, 99%; 10g, 99%); (v) DCC, N-hydroxyphthalimide, Et3N,
cystamine, 1,4-dioxane, MeOH, 25 °C, 12 h (11a, 46%; 11b, 54%; 11c, 15%; 11d, 57%; 11e, 57%; 11f, 42%; 11g, 52%).
O
O
i
Br
Br
OH
OH
N
H
2
N
N
NH2
HO
HO
5
HO
HO
H2N
12
13
O
O
i
Br
Br
HO
X
N
R
H
X
N
N
H2N
R
HO
HO
HO
5
15a X = O, R = TBS
15b X = O, R = Me
15c X = S, R = Me
14a X = O, R = TBS
14b X = O, R = Me
14c X = S, R = Me
ii
16 X = O, R = H
Scheme 4. Reagents and conditions: (i) DCC, N-hydroxyphthalimide, Et3N, 12 or 14a–c, 1,4-dioxane, MeOH, 25 °C, 14 h (13, 70%; 15a, 83%; 15b, 20%; 15c, 18%); (ii) 7:2:1 THF/
HCO2H/H2O, 25 °C, 14 h (76%).
O
O
O
Br
Br
S
i
N
OR
OMe
H
2
N
HO
HO
Br
HO
HO
17
X
1
18a X = H, R = Me
18b X = Br, R = Me
19 X = H, R = H
ii
i
O
O
S
iii
Br
S
2
N
O
N
H
H
2
N
O
HO
20
21
Scheme 5. Reagents and conditions: (i) NBS, DMF, 25 °C, 5 h (18a, 56%; 18b, 17%);
(ii) LiOH, THF–H2O, 23 °C, 12 h (99%); (ii) DCC, N-hydroxyphthalimide, Et3N,
cystamine, 1,4-dioxane, MeOH, 25 °C, 14 h (40%).
Scheme 6. Reagents and conditions: (i) MTA, CH3CN, 25 °C, 15 h (5%).
acute myeloid leukemia cell line. In vitro tests of the compounds
at 5 lM on human recombinant HDAC1, using SAHA as a positive
conditions described for oxime 4 (NBS in CH3CN) yielded a mixture
of mono- and dibromo derivatives, 18a and 18b, respectively, in a
3:1 ratio. Hydrolysis of the former followed by the coupling of 19
with cystamine led to the desired disulfide 20 in moderate yield
(Scheme 5).
Lastly the spirocyclic hexadienyl-isoxazoline 21, a potential
metabolite of PsA1 could only be obtained, albeit in very low yield
(5%), by the oxidative-induced cyclization61 of PsA 1 with manga-
nese(III) tris(acetylacetonate) (MTA)62 (Scheme 6) after many other
methods failed.
control, confirmed the enzymatic inhibition of PsA 1 (Fig. 1A). In
addition, some analogues (11a–e, 11g) reduced the activity of
HDAC1 more efficiently than 1. The other compounds of the series
(7a–d, 11f, 13, 15b–c, 16, 20 and 21) did not noticeably affect
HDAC1 activity. Compounds with longer chain connecting the
disulfide to the hydroxyimino amide (7a–d) lack significant
inhibitory activity. A similar result was observed for compounds
having the disulfide replaced by methylene units (13) and for the
monomers with either methyl ether –OMe, methyl thioether
(–SMe) or alcohol (–OH) functionalities (15b, 15c and 16, respec-
tively). In contrast, the inhibitory activity is maintained and even
increased with compounds that preserve the general modified tet-
rapeptide scaffold of PsA regardless of the nature and pattern of the
substituents at the aryl ring (11a–f). The more drastic change of
the overall aryl ring structure imparted by the spirocycle together
3. Biological characterization
Firstly, we focused on the reported inhibition of HDAC by PsA
and the synthetic analogues, as well as on their effects on cell cycle,
induction of differentiation and apoptosis on the U937 human

3640
J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
Figure 1. (A) HDAC1 fluorescent assay with the indicated compounds at 5 lM. The inhibition is reported as percentage of activity relative to the control (100%). (B–D) Cell
cycle analysis and apoptosis in U937 cells treated with the indicated compounds at 5
lM for 30 h. The data represent the media of independent duplicates.
with the reduction of flexibility and the lack of free oxime in deriv-
ative 21 led to substantial loss of HDAC inhibitory activity.
Cell-based assays were performed on the U937 myeloid leuke-
mia cells to determine the anti-proliferative potential and the abil-
ity of the compounds to revert myeloid tumor cells to
differentiated granulocytes. Cell cycle progression, differentiation
and amount of cells that undergo apoptosis following treatment
The percentage of apoptotic cells (measured as caspase 3 acti-
vation by FACS analysis) increased when U937 cells were treated
with 1 and series 11a–g for 30 h (Fig. 2A). After 30 h induction
with compounds 11a–e and 11g the percentage of apoptotic cells
varied between 30% and 40%. Other compounds, including bro-
mopsammaplin A (11f), and the derivative lacking the oxime
function 20, exhibited only minor effects even at longer incuba-
tion times (40 h). The results confirmed the in vitro findings since
the most potent inducers of differentiation and apoptosis in
U937 cells are the 11a–g series of ring-modified PsA analogues
(with the exception of bromopsammaplin A 11f). Modifications
in the chain length and the disulfide led to lower values of apop-
tosis, in agreement with the enzymatic assays. Interestingly,
spiro derivative 21 showed comparable induction of apoptosis to
parent 1.
with PsA 1 and analogues (5 lM) for 30, 40, and 48 h were deter-
mined (Figs. 1B and 2). Compared to the vehicle-treated cells and
relative to the positive control represented by the known HDACis
MS275 (HDAC1,2,3-selective) and SAHA (also HDAC6 inhibi-
tor),63,64 only PsA 1 and some analogues induced cell cycle arrest
in G1 (Fig. 1, panels B–D). In particular after 30 h PsA induced a
time-dependent accumulation of U937 cells in the G1 phase
(60%), and analogues 11a–g and 20 showed even greater arrest
(80%), with values comparable to MS-275 (Fig. 1C). A clear correla-
tion between the in vitro HDAC1 inhibition and the in vivo efficacy
to induce cell cycle arrest was noticed for the most active com-
pounds. Compound 21, on the other hand, appears to block cell cy-
cle at G2/M. The remaining analogues gave no detectable activities
on cell cycle progression even after longer (48 h) treatment re-
gimes (Fig. 1B and D).
The differentiation of myeloid precursors to granulocytes was
determined by measuring the presence on the cell membrane of
the granulocytic differentiation marker CD11c antigen, which is
highly expressed only on mature granulocytes, monocytes and
certain lymphocytes, but not significantly on myeloid committed
precursor cells. After treating the U937 cells with the PsA
analogues at 5 lM for 30 and 40 h, low differentiation levels

J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
3641
Figure 2. (A) Induction of apoptosis after treatment of U937 cells with the indicated compounds at 5 lM, for 30 h (for the series 11a–11g) and 40 h. (B) Differentiation
analysis in U937 cells after 30 h treatment with the indicated compounds at 5
lM. The percent value of CD11c positive/propidium iodide (PI) negative cells is represented.
The data shown is the media of independent quadruplicates.
were measured (with the exception of parent 1 and methyl sulfide
15c) relative to control (Fig. 2B) and to the class I-selective HDACi
MS-275.
In order to determine if the most potent analogues are endowed
with additional epigenetic modulation activities, the series 11a–g
and 20 were also used in in vitro human SIRT1 fluorescent assay
and in a radioactive assay on a peptide fragment having p300/
CBP histone acetyl transferase (HAT) enzyme activity. As shown
in Figure 4B, very weak SIRT1 inhibition was noted (ca. 30% relative
We next determined the expression levels of p21WAF1, which is
involved in the mechanism of tumor suppression, as well as the
global acetylation status of histone (histone H3) and non-histone
proteins (tubulin), that are substrates of HDACs family members.
The up-regulation of p21WAF1 and the increase of tubulin acetyla-
tion levels were evaluated by Western blot analyses on total
extracts after treating the U937 cells for 24 h with the compounds
to the control) for the majority of compounds at 50 lM, far lower
than the activity of the SIRT1 inhibitor suramin at the same con-
centration. In addition, none of the analogues displayed modula-
tion of a peptide fragment of CBP containing the enzymatically
active HAT domain (Fig. 4A) relative to the control and to the
effects of the known inhibitor anacardic acid (AA) at the same con-
centration,65,66 thus confirming the specificity of the HDAC among
other epigenetic inhibitory activities.
at 5 lM (Fig. 3A–C). Compounds 7a–d, 13, 15b, 16 and 21 failed to
increase both p21WAF1 and tubulin acetylation expression levels.
Methyl sulfide 15c increased weakly these levels in line with its
noticeable effect on the induction of differentiation. While the
11a–g series did not show a significant effect on tubulin acetyla-
tion, some members (11b–e) up-regulated p21WAF1 to levels even
higher than those of SAHA. The effect of the analogues on the level
of histone acetylation was also analyzed by Western blot (Fig. 3D).
After treating U937 cells for 24 h all compounds of the series 11a–
g displayed the ability to increase the level of acetylated histone
H3, present in the histonic extract, as shown using the specific
antibody.
DNMT inhibition has also been reported for PsA.3 To verify the
effective physical interaction between PsA and its analogues and
DNA methyl transferase enzymes, two in vitro radioactive assays
were performed using DNMT1 and DNMT3A (Fig. 4C and D).
DNMT1 was immunoprecipitated from K562 cells and used in
radioactive assay that employs [3H]-adenosyl-
L-methionine as
methyl donor and Poly dI-dC as methyl acceptor. The same condi-
tions have been used for the DNMT3A radioactive assay, but the

3642
J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
Figure 3. (A–C) Western blot analyses of p21WAF1/CIP1 expression and
ERK1 expression is shown as loading control. (D) Western blot analyses of histone H3 acetylation carried out in U937 cells, after 24 h induction with the compounds at 5
a
-tubulin acetylation in U937 cells after treatment with the indicated compounds at 5
l
M for 24 h.
M.
l
recombinant enzyme was produced in Escherichia coli BL21, as GST
fused protein. In neither assay compounds 11a–g and 20 showed
the active site binding/inactivating group is connected via a
hydroxyimino amide linker to the HDAC recognition aryl group.
significant inhibition at 50
l
M
(see Fig. 4C), in contrast to
In order to address the nature of the interaction between the
inhibitor and HDAC, the ab initio calculated structure of thiol 22
was docked into the active site of the human HDAC8-trichostatin
A (TSA) crystal structure70 after removal of the TSA ligand. The
catalytic domain of about 390 amino acids responsible for the
deacetylation is highly conserved among the metal-dependent
HDACs, in particular the residues lining the ligand-binding pocket,
but some differences can be exploited for the designed of selective
Class I/Class II HDAC inhibitors.71
The highest scores using automated docking method for the
interaction of the ligand with the Zn2+ ion in the active site, vali-
dated by the GRID maps, agree in having the thiol chelated to
the metal, whereas the linker domain occupies the channel and
the o-bromophenol is stabilized through interaction with Tyr100
and Phe152 at the rim of the active site entrance (Fig. 5 and Figs.
S1–S3). This positioning facilitates the formation of a hydrogen
bond between the oxime group and Asp101, which remains at a
constant distance along the energy minimization and the simula-
tion of the dynamic behaviour using unrestrained MD (Fig. S3).
The Zn2+ ion is kept firmly coordinated to the four ligands
(Asp178, Asp267, His180 and the thiol group of the inhibitor),
RG10867 and SGI102768 at the same concentration for the DNMT1
assay.
Taken together, the enzymatic profiling suggests that the
epigenetic activities of PsA are mainly restricted to the inhibition
of the HDAC family in this context. Moreover, the HDAC activity
of the series is likely linked to the formation of the monomeric
thiol that originates from –Sꢀ ꢀ ꢀS– bond cleavage (Scheme 7, vide
infra), a known potent metal chelator. The Zn2+-dependent his-
tone deacetylase subfamily is composed of Class I HDACs (HDAC
1–3, 8 and 11) and Class II HDACs (HDAC 4–7, 9 and 10). In con-
trast, Class III (sirtuins), with seven members (SIRT1-7), require
NAD+ as a cofactor and release O-acetyl-ADP ribose and nicotin-
amide as a consequence of acetyl transfer from the acetylated
lysine.69
Metalloproteinase-targeted HDAC inhibitors are typically
substrate mimics of the linear acetyl-lysine side chain with a
Zn2+-chelating ‘warhead’ group that replaces the scissile acet-
amide, a connector chain and a ‘cap’ at the other end that extend
beyond the enzyme substrate-binding channel. These features
are exhibited by the thiol derived from PsA (Scheme 7), in which

J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
3643
Figure 4. (A) CBP radioactive assay performed with 1
lg of recombinant CBP enzyme peptide fragment and the psammaplin A analogues at 50
lM. (B) SIRT1 fluorescence
assay to measure the lysine deacetylase activity of the compounds at 50
dC as methyl acceptor (0.1 -methionine (1 Ci) as methyl donor. DNMT1 was immunoprecipitated from K562 cells and the compounds 11a–g were
g) and [3H]-adenosyl-
used at 50 M. (D) DNMT1 radioactive assay to measure the inhibition of recombinant DNMT3a (1 g, produced in E. coli BL21, as GST fused protein) by the compounds at
lM on human recombinant SIRT1 enzyme (1 U). (C) DNMT1 radioactive assay with synthetic poly dI-
l
L
l
l
l
50
lM. In each panel the inhibition/activation of the enzyme is reported as percentage of activity relative to the control.
O
tion of several enzymes that impact different stages on the onset
BIOACTIVATION
and progression of cancer such as topoisomerase II (growth),14
the zinc-dependent metalloproteinase aminopeptidase N (APN,
tumor cell invasion or angiogenesis),21 HDAC and DNMT (chroma-
tin remodeling),3 among others, likely conspire to account for the
reported anticancer activities of PsA in several cancer cell lines
and in the A549 lung xenograft mouse model.3,12 PsA was reported
to activate Wnt signalling in a cell-based assay but the effect is
likely due to HDAC inhibition rather than to an specific Wnt signal-
ling pathway.72
Br
SH
N
1 Psammaplin A
H
N
HO
HO
22 Psammaplin A thiol
Scheme 7. Bioactivation of PsA 1 to thiol 22 by the reductive environment of the
cells.
and the electrostatic term (ꢁ59.2191 1.7841 kcal/mol) makes an
important contribution to the overall energy. Favourable van der
Waals interactions of the ligand with Phe152 and Phe208
residues also account for a fraction of the intermolecular energy
component (Fig. S2).
More recent molecular and cellular studies73 confirmed the
potent inhibitory activity of PsA in enzymatic (HDAC inhibition)
and in anti-proliferation assays, and also the selective induction
of histone hyperacetylation. The anti-proliferative effects were
linked to the overexpression of genes related to cell cycle arrest
and apoptosis (p53-independent p21WAF1 expression).74
4. Discussion
Psammaplin A 1 is the prototype of a collection of metabolites
isolated from sponges1 that are biosynthesized by linear connec-
tions of (bromo)tyrosines and modified cysteines.12 Their biologi-
cal activities, common to most of the bromotyrosine/cysteine
constructs, range from antimicrobial54 to anticancer.1,6 The inhibi-
A series of PsA analogues modified at the aryl ring, with varying
lengths of the amino thiol connecting unit, and some exchanges/
deletions of functional groups have been prepared. These ana-
logues have been characterized with regard to their enzymatic
inhibitory potential and for their effects on cell cycle, differentia-

3644
J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
and class II metalo-HDACs are noticeable in their size (with class
II being from two to three times larger), their cellular localization,
the conservation of sequence motifs in the catalytic domains, the
identity of the protein–protein interaction complexes and the tis-
sue distribution.76
Western blot analysis of U937-treated cells confirmed the accu-
mulation of acetylated histones using antibodies against acetylated
H3 (Fig. 3D), in agreement with similar findings for PsA in endome-
trial cancer Ishikawa cells.74 PsA and several analogues up-regu-
lated cyclin-dependent kinase inhibitor p21WAF1 (Fig. 3), which is
one of the genes induced by HDAC inhibitors,76 and this effect
might be related to the suppression of cell proliferation and induc-
tion of apoptosis.
PsA and its 5-bromo derivative have been reported to inhibit
the bacterial methyltransferase SssI. However, cultured HCT116
human colon carcinoma cells treated with 1 lM PsA did not reduce
global genomic DNA methylation (the level of which was deter-
mined using a mass spectrometry assay) and failed to induce the
hypomethylation or reactivation of cancer-testis antigen genes,
known as methylation silenced genes.56 Here, two different assays
were performed on DNMT1 and DNMT3A (see Fig. 4C and D) to
evaluate the inhibition of DNMT enzymes by PsA and derivatives
Figure 5. Proposed docking pose of biocleaved psammaplin A (thiol 22) with
HDAC8. The Ca trace of the enzyme is displayed as a ribbon, colored in green. The
side chains of His142, His143, Asp178, Asp267, His180, Met274, Tyr306 are shown
as sticks, with carbon atoms colored in green. The Zn2+ ion at the catalytic site is
shown as a magenta sphere. The inhibitor is displayed also as sticks, but with
carbon atoms colored in gray and the bromine atom colored in brown.
(11a–g). Differently from expected, these compounds at 50 lM
failed to inhibit DNMT1 and DNMT3a, thus indicating their inabil-
ity to alter DNMT in vitro. Therefore, the action of these com-
pounds on cell cycle progression and induction of apoptosis can
be better correlated with the HDAC inhibitory activities. It is likely
that the absence of hypomethylation is not due to the low affinity
of PsA (and its analogues) for DNMT, but instead to the inefficient
transport of these compounds through the nuclear membrane.
Inefficient transportation of the compound into the nucleus or poor
cell membrane penetration might also explain the much greater
concentration of 1 (1800 times) required to obtain similar poten-
cies in cell-based assays relative to the HDAC enzymatic inhibition
assay.52
Lastly, no significant inhibition of SIRT1 and p300/CBP HAT
enzymes was seen for the PsA-related series of analogues (Fig. 4A
and B).
Other natural products of the depsipeptide class that act as
inhibitors of HDAC also contain disulfide bonds, namely FK228
23, spiruchostatins 25 and FR901375 26 (Fig. 6).39,71 FK228 (romi-
tion and apoptosis on the U937 acute myeloid leukemia cell line.
For 1 and the most potent analogues, functional assays have also
been carried out on the same cell line.
Enzymatic HDAC1 inhibition data for the series of analogues
having chain lengths spanning from two to six methylene units (
7a–d) established that the derivative of natural cystamine (the
decarboxylated cysteine) present in psammaplin A was the most
potent, and also that both the oxime and the disulfide functional-
ities were required for the HDAC inhibition activity, since ana-
logues 13, 15b–c, 16 and 20 proved to be inactive (Fig. 1A). We
then addressed the modifications of the aryl ring with analogues
that preserve the same connecting diamine unit of the parent nat-
ural product. The substitution at the ring is well tolerated, and a
general improvement in potency relative to PsA was noted when
the ring was mono- (11a–b) or disubstituted (11d,e,g). The unsub-
stituted compound (the phenylalanine-derived 11c) and the tri-
substituted bromopsammaplin A 11f were less active than 1.
Conversely the series of halogenated analogues 11d,e,g exhibited
greater potency than PsA and followed the order Br < F < Cl < I.
The activities of the spiro derivative 21 are most intriguing, since
it is the only compound showing a G2/M block (Fig. 1B) and strong
induction of apoptosis (Fig. 2B), which appears not to be correlated
with HDAC inhibition. A promising pro-differentiation profile is
also noted for the methylsulfide 15c (Fig. 2B), which merit further
investigation.
O
O
N
H
SH
O
HN
N
O
BIOACTIVATION
HN
H
HN
S
HN
O
S
O
NH
NH
O
O
O
O
O
It has previously been reported that human endometrial Ishik-
awa cancer cells treated with PsA showed accumulation of cells
in the G1 phase and a significant decrease in the number of cells
in the S phase,74 a result in keeping with the effect of known HDAC
inhibitors.75 Our data confirm the time-dependent accumulation of
cells in G1 upon treatment of human leukaemia U937 cells with
PsA and analogues. Moreover, 1 and the 11a–g series of related
compounds induced apoptosis of U937 cells (Fig. 2B), an effect
likely associated to their more selective inhibition of class I HDACs.
The majority of these compounds (with the exception of 15c) failed
to induce tubulin acetylation (Fig. 3), which is a target of HDAC6
enzymes, and this finding is in agreement with previous reports
for PsA in HeLa cells.73 Although the result might suggest that
psammaplin A show selectivity for histone proteins in preference
to tubulin, it cannot be considered as a selective class I HDAC
inhibitor, since it also inhibits HDAC4, a class II HDAC, in enzymatic
assays (results not shown). Important differences between class I
O
24, redFK
23, FK228 (FR901228)
HS
OH
O
R
O
O
O
H
N
N
HN
N
OH
H
H
O
S
S
HN
O
S
S
NH
NH
O
O
O
O
O
25, FR901375
26a, Spiruchostatin A R = Me
26b, Spiruchostatin B R = H
Figure 6. Natural depsipeptides 23–26 containing disulfide bonds are HDAC
inhibitors. The structure of previously characterized active reduced form of FK228
(24, redFK) is shown by analogy with the reduced form of psammaplin A, 22.

J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
3645
depsin, IstodaxÒ) currently approved for T cell lymphoma is under-
going phase II clinical trials for the treatment of non-Hodgkin’s
lymphoma, acute myelogeneous leukemia, and pancreatic cancer
(www.clinicaltrials.gov). It has been shown that the disulfide bond
of FK228 becomes cleaved in the cells by the reducing activity of
glutathione to afford redFK 24, and the thiol then interacts with
the Zn2+ ion in the active site of HDAC.77 Moreover, mass spectrom-
etry analysis of blood samples identified the thiol and glutathione
conjugates of FK228.78 Upon bio-reduction depsipeptides 25 and
26 must also release the Zn2+-binding butenylthiol.79 From an
structural point of view both redFK 25 and PsA thiol 22 have
Zn2+ ion connecting chains that are shorter than those of common
hydroxamic acid-based inhibitors (TSA, SAHA. . .). Thiol chelating
groups connected via a 5 atom saturated chain to cyclic tetrapep-
tides have also shown potent HDAC inhibitory activities.80,81
In agreement with these precedents, the dimer analogue lacking
the disulfide bond (13) was inactive, which substantiates the pro-
posal that PsA and analogues undergo thiolate exchange reaction
with glutathione by –Sꢀ ꢀ ꢀS– bond cleavage.82 Interestingly, Wang
et al. have shown that glutathione-depleted cells were not sensi-
tive to PsA, implying that the reducing environment of the cells
triggered the conversion of the disulfide of 1 to the corresponding
thiol,79 which would be the species responsible for the HDAC
inhibitory activity of psammaplin A after uptake into the cells.
Based on the reported crystal structure, the HDAC8ꢀZn2+ꢀthiol
22 complex was constructed with the thiol group chelating the
metal ion, and then refined using energy minimization. The dy-
namic behaviour of the complex using unrestrained molecular
dynamics confirmed the feasibility of the proposed binding orien-
tation and the mutual adaptation between HDAC8 and the ligand.
The linker occupies the narrow channel whereas the bromophenol
interacts with Tyr100 and Phe152 at the rim of the active site en-
trance, a pose that facilitates the formation of a robust hydrogen
bond between the oxime group and Asp101 (Fig. 5).
very low. Since epigenetic signalling by PsA appears to be re-
stricted to inhibition of metalloproteinase HDACs, a model for
the interaction of HDAC8 with the thiol derived from PsA 1 (by a
presumed in vivo disulfide cleavage) was computed. The thiol
binds the Zn2+ ion in the active site and the complex is additionally
stabilized by a hydrogen bond interaction between the oxime
group and the Asp101 residue located at the active site entrance.
As with FK228, this bioactivation mechanism illustrates the inge-
nious solution adopted by Nature to protect the reactive zinc-bind-
ing thiol as a disulfide pro-drug with higher bioavailability. The
increased levels of disulfide reductants (i.e., glutathione, thiore-
doxin and thioredoxin reductase) found in many cancer cells ren-
der such cells particularly susceptible to the action of PsA 1.
Despite the failure to greatly improve the epigenetic inhibitory
profile of PsA through the modifications reported, more significant
skeletal alterations of the PsA structure might provide more potent
derivatives, thus lending further support to the important role of
natural products as inspiration85 for the development of designed
multiple ligands50,51 as new anticancer drugs.86,87
6. Experimental section
6.1. General
Solvents were dried according to published methods and dis-
tilled before use. HPLC grade solvents were used for the HPLC puri-
fication. All other reagents were commercial compounds of the
highest purity available. All reactions were carried out under argon
atmosphere, and those not involving aqueous reagents were car-
ried out in oven-dried glassware. Analytical thin layer chromatog-
raphy (TLC) was performed on aluminium plates with Merck
Kieselgel 60F254 and visualized by UV irradiation (254 nm) or by
staining with a solution of phosphomolibdic acid. Flash column
chromatography was carried out using Merck Kieselgel 60 (230–
400 mesh) under pressure. Infrared spectra were obtained on JAS-
CO FTIR 4200 spectrophotometer, from a thin film deposited onto a
NaCl glass. 1H NMR spectra were recorded in CDCl3, CD3OD, DMSO-
d6 and (CD3)2CO at ambient temperature on a Bruker AMX-400
spectrometer at 400 MHz with residual protic solvent as the inter-
nal reference (CDCl3, dH = 7.26 ppm; (CD3)2CO, dH = 2.05 ppm;
CD3OD, dH = 3.31 ppm; DMSO-d6, dH = 2.50 ppm); chemical shifts
(d) are given in parts per million (ppm), and coupling constants
(J) are given in Hertz (Hz). The proton spectra are reported as fol-
lows: d (multiplicity, coupling constant J, number of protons,
assignment). 13C NMR spectra were recorded in CDCl3 CD3OD,
DMSO-d6 and (CD3)2CO at ambient temperature on the same spec-
trometer at 100 MHz, with the central peak of CDCl3
(dC = 77.0 ppm), CD3OD (dC = 49.0 ppm), DMSO-d6 (dC = 39.4 ppm)
or (CD3)2CO (dC = 30.8 ppm) as the internal reference. The DEPT135
sequence was used to aid in the assignment of signals on the 13C
NMR spectra. Melting points were determined on a Stuart SMP10
apparatus. Elemental analyses were determined on a Carlo Erba
EA 1108 analyzer. MS experiments were performed on an APEX
III FT-ICR MS (Bruker Daltonics, Billerica, MA), equipped with a
7T actively shielded magnet. Ions were generated using an Apollo
API electrospray ionization (ESI) source (Bruker Daltonics, Billerica,
MA), with a voltage between 1800 and 2200 V (to optimize ionisa-
tion efficiency) applied to the needle, and a counter voltage of
450 V applied to the capillary. Samples were prepared by adding
a spray solution of 70:29.9:0.1 (v/v/v) methanol/water/formic acid
to a solution of the sample at a v/v ratio of 1–5% to give the best
signal-to-noise ratio. Data acquisition and data processing were
performed using the XMASS software, version 6.1.2 (Bruker
Daltonics). FAB experiments were performed on a VG AutoSpec
instrument, using 3-nitrobenzylalcohol or glycerol as matrix.
Our results using alcohol 16, ether 15b and most importantly
dimer 13 confirm the earlier findings with FK22878 and suggest
that psammaplin A 1 acts as a stable pro-drug83 that is activated
upon uptake by the reductive environment of the cells to afford
thiol 22, which then chelates the Zn2+ ion in the active site of the
HDAC metalloenzymes.
5. Conclusion
PsA 1, a natural product with a disulfide bond derived from the
condensation of modified cystamine and tyrosine moieties, has
been reported to display dual HDAC and DNMT epigenetic inhibi-
tory activities. The intriguing ability to target simultaneously more
than one member of the epigenetic machinery (epigenetic multiple
ligands),84 prompted us to synthesize a series of PsA analogues in
an effort to determine the structural determinants for their epige-
netic profile. We have analyzed the effects of the natural-product
inspired collection on the human leukaemia U937 cell cycle, mea-
suring induction of apoptosis and differentiation, induction of
p21WAF1 and tubulin acetylation levels and total histone H3, and
also examined their HDAC1 enzyme-based inhibition profile. From
the results it is concluded that Nature has optimized the design of
the PsA scaffold to fulfil this epigenetic role, since only closely re-
lated synthetic derivatives (with modifications at the bromotyro-
sine ring) exhibited comparable or greater potency than the
natural product. Modifications of the connecting chain, oxime bond
and disulfide unit afforded either inactive or considerable less po-
tent analogues. Exploring the possibility that PsA and derivatives
might display additional epigenetic activities we have also tested
them as inhibitors of DNMT1, DNMT3A, SIRT1 and a peptide con-
taining the p300/CBP HAT domain, but the values measured were

3646
J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
6.2. General procedure for the oxidation of amines with Na2WO4
acid 5 as a white powder, mp: 147–148 °C (hexane/CHCl3) (lit.
147–148 °C, dec.).89 1H NMR (CD3COCD3, 400.13 MHz): d 7.45 (s,
1H, H20), 7.14 (d, J = 7.9 Hz, 1H, ArH), 6.90 (d, J = 7.9 Hz, 1H, ArH),
3.84 (s, 2H, 2H3) ppm.
To a solution of amine (18.0 mmol) in EtOH (40 mL) at 0 °C were
added Na2WO4ꢀ2H2O (18 mmol), 30% H2O2 (16 mL) and H2O
(30 mL). The resulting mixture was stirred for 4 h at room temper-
ature and the reaction was quenched with aqueous saturated
NH4Cl and extracted with EtOAc (3ꢂ). The combined organic
extracts were washed with brine, dried over Na2SO4 and evapo-
rated. The residue was purified by column chromatography on sil-
ica gel as indicated.
6.4.4. (2E,20E)-N,N0-[2,20-Disulfanediylbis(ethane-2,1-diyl)]bis[3-
(3-bromo-4-hydroxyphenyl)-2-(hydroxyimino)propanamide]
(psammaplin A, 1)56
Following the general procedure for the amidation of tyrosine
acid derivatives with amines, acid 5 (0.16 g, 0.59 mmol) afforded,
after purification by column chromatography (SiO2, gradient from
25:75 hexane/EtOAc to 95:5 CH2Cl2/MeOH), psammaplin A 1
(0.12 g, 60%) as a white foam. 1H NMR (CD3OD, 400.13 MHz) (data
for monomer): d 7.36 (s, 1H, H20), 7.07 (d, J = 8.3 Hz, 1H, ArH), 6.76
(d, J = 8.3 Hz, 1H, ArH), 3.79 (s, 2H, 2H3), 3.52 (t, J = 6.7 Hz, 2H,
2H100), 2.81 (t, J = 6.7 Hz, 2H, 2H200), 2.15 (br, 2H, OH) ppm.
6.3. General procedure for the hydrolysis of esters
Lithium hydroxide (15 mmol) was added to a solution of ester
(1 mmol) in a 1:1 THF/H2O (16 mL) mixture. The solution was
stirred at room temperature for 2 h, neutralized with 10% HCl
and extracted with EtOAc (4ꢂ). The combined organic layers were
washed with brine, dried over Na2SO4, filtered and the solvent was
evaporated in vacuo. Crystallization of the residue provided the de-
sired acid as indicated.
6.5. Molecular modeling of psammaplin A
6.5.1. Quantum mechanics calculations
The geometry of the thiol derived from psammaplin A was opti-
mized using the ab initio quantum chemistry program Gaussian
0390 and the HF/3-21G* basis set. A set of atom-centred RHF 6-
31G*//3-21G* charges was then obtained by using the RESP meth-
odology91 as implemented in the AMBER suite of programs (http://
amber.scripps.edu/). Covalent and nonbonded parameters for the
inhibitor atoms were assigned, by analogy or through interpola-
tion, from those already present in the AMBER force field92
(parm99) or consistently derived, as explained in more detail
elsewhere.93
6.4. General procedure for the amidation of tyrosine acid
derivatives with amines
N-Hydroxyphtalimide (1 mmol) and DCC (1 mmol) were added
to a solution of the carboxylic acid (1 mmol) in dioxane (2.5 mL).
After the mixture had been stirred for 2 h at room temperature, a
solution of the amine (0.5 mmol), Et3N (2 mmol) and MeOH
(1 mL) was added. The resulting mixture was stirred for 12 h at
room temperature and the reaction was quenched with H2O and
extracted with EtOAc (3ꢂ). The combined organic extracts were
washed with brine, dried over Na2SO4 and evaporated. The residue
was dissolved in THF, filtered off and the solvent was evaporated in
vacuo. The resulting residue was then purified by column chroma-
tography on silica gel as indicated.
6.5.2. Molecular docking
The genetic algorithm94 implemented in AutoDock95 and the
h-HDAC8 (PDB code 1t64)70 as the target protein upon removal
of trichostatin A was used to generate different HDAC-Zn2+-bound
psammaplin A thiol conformers by randomly changing torsion an-
gles and overall orientation of the molecule. A volume for explora-
tion was defined in the shape of a three-dimensional cubic grid
with a spacing of 0.3 Å that enclosed the residues that are known
to make up the inhibitors binding pocket. At each grid point, the
receptor’s atomic affinity potentials for carbon, oxygen, nitrogen,
sulfur, bromine and hydrogen atoms present in the ligand were
precalculated for rapid intra- and intermolecular energy evaluation
of the docking solution.
6.4.1. (E)-Methyl 2-(hydroxyimino)-3-(4-
hydroxyphenyl)propanoate (3)88
Following the general procedure for the oxidation of amines
with Na2WO4ꢀ2H2O,
L-tyrosine methyl ester 2 (3.45 g, 17.68 mmol)
gave, after purification by column chromatography (SiO2, 50:50
hexane/EtOAc), 2.24 g (60%) of oxime 3 as a white powder. 1H
NMR (CD3COCD3, 400.13 MHz): d 11.35 (s, 1H, OH), 8.11 (s, 1H,
OH), 7.11 (d, J = 8.5 Hz, 2H, ArH), 6.73 (d, J = 8.5 Hz, 2H, ArH),
3.84 (s, 2H, 2H3), 3.72 (s, 3H, CO2CH3) ppm.
To obtain additional validation of the proposed binding mode
for the ligands, the program GRID (http://www.moldiscovery.
com/96 was also used to search for sites on the enzyme that could
be complementary to the functional groups present in this inhibi-
tor. For the GRID calculations, a 18 ꢂ 21 ꢂ 21 Å lattice of points
spaced at 0.5 Å was established at the binding site. The probes used
were C1 = (aromatic carbon), N1 (neutral flat NH, eg amide), N:#
(sp nitrogen with lone pair), O (sp2 carbonyl oxygen) and Br
(bromine). The dielectric constants chosen were 4.0 for the macro-
molecule and 80.0 for the bulk water.
6.4.2. (E)-Methyl 3-(3-bromo-4-hydroxyphenyl)-2-
(hydroxyimino)propanoate (4)61
A solution of NBS (0.49 g, 2.75 mmol) in CH3CN (6 mL) was
added dropwise over 15 min to the solution of the oxime 3
(0.58 g, 2.75 mmol) in CH3CN (6 mL). The reaction mixture was
stirred at room temperature for 5 h, after which time the solvent
was evaporated and the residue was treated with water and
extracted with EtOAc (3ꢂ). The combined organic extracts were
dried over Na2SO4, filtered, evaporated and the residue was puri-
fied by column chromatography (SiO2, 50:50 hexane/EtOAc) to
afford oxime 4 (0.65 g, 82%) as a yellow powder. 1H NMR
(CD3COCD3, 400.13 MHz): d 11.49 (s, 1H, OH), 8.62 (s, 1H, OH),
7.44 (d, J = 1.6 Hz, 1H, H20), 7.13 (dd, J = 8.3, 1.6 Hz, 1H, H60), 6.90
(d, J = 8.3 Hz, 1H, H50), 3.85 (s, 2H, 2H3), 3.74 (s, 3H, CO2CH3) ppm.
6.5.3. Molecular dynamics simulations
Ternary complexes (HDAC8-Zn2+-psammaplin A thiol) repre-
sentative of the most populated solutions were then refined using
the second generation AMBER force field and 3000 steps of steep-
est descent energy minimization and 6000 steps of conjugate
gradient of only the side chain of the protein and those atoms of
the bound ligand. This procedure allowed readjustment of covalent
bonds and van der Waals contacts without changing the overall
conformation of the complex. The HDAC-psammaplin A thiol com-
plex was then neutralized by addition of eight sodium ions97 that
were placed in electrostatically favored positions and immersed
6.4.3. (E)-3-(3-Bromo-4-hydroxyphenyl)-2-
(hydroxyimino)propanoic Acid (5)
In accordance with the general procedure for the hydrolysis of
esters, ester 4 (0.61 g, 2.11 mmol) gave, after purification by col-
umn chromatography (SiO2, 90:10 EtOAc/MeOH), 0.58 g (99%) of

J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
3647
in rectangular boxes each containing about 450 TIP3P water mole-
cules98 that extended 1 Å away from any solute atom. The cutoff
distance for the non-bonded interactions was 9 Å, and periodic
boundary conditions were applied. Electrostatic interactions were
represented using the smooth particle mesh Ewald method with
a grid spacing of ꢃ1 Å. Unrestrained molecular dynamics (MD)
simulations at 300 K and 1 atm were then run for 6 ns using the
SANDER module in AMBER 8.99 The coupling constants for the tem-
perature and pressure baths were 1.0 and 0.2 ps, respectively.
SHAKE100 was applied to all bonds involving hydrogens, and an
integration step of 2 fs was used throughout. The nonbonded pair
list was updated every 10 steps. The simulation protocol involving
a series of progressive energy minimizations followed by a 20 ps
heating phase and a 70 ps equilibration period before data collec-
tion. System coordinates were saved every 2 ps for further analysis.
ommended by the suppliers. The cells were analysed after washing
using the CellQuest software applied to a FACScalibur (BD). Exper-
iments were performed in duplicate and values were expressed as
mean SD.
6.7.1. Granulocyte differentiation
Granulocyte differentiation was carried out as previously de-
scribed.103 Briefly, U937 cells were harvested and resuspended in
10
ples were incubated with 10
30 min at 4 °C in the dark, washed in PBS and resuspended in
500 L PBS containing propidium iodide (0.25 g/mL). Samples
l
L phycoerythrin-conjugated CD11c (CD11c-PE). Control sam-
lL PE conjugated mouse IgG1 for
l
l
were analyzed by FACS with Cell Quest technology (Becton Dickin-
son). Propidium iodide (PI) positive cells have been excluded from
the analysis.
6.5.4. Analysis of the molecular dynamics trajectories
Three-dimensional structures and trajectories were visually in-
spected using the computer graphics program InsightII. The root-
mean-square (rms) deviations from both the initial structures
and the average structures, the inter-atomic distances, and the
snapshot geometries were obtained using the PTRAJ module in
AMBER. Intermolecular van der Waals energies for individual resi-
dues were calculated with the ANAL module, whereas the solvent-
corrected residue-based electrostatic interaction energies were
calculated with DelPhi, following the procedure described.93
All calculations were performed on the SGI R14000 Origin 3800
at CIEMAT (Madrid), on the SGI 1.5 GHz Itanium2 at CESGA (Santi-
ago de Compostela) and locally on SGI R12000 Octane
workstations.
6.7.2. Western blot analyses
Western blot analyses were performed according to standard
procedures following suggestions of antibody suppliers. For the
determination of p21WAF1/CIP1 50
lg of total protein extracts were
separated on a 15% polyacrylamide gels and blotted. Western blots
were shown for p21 (Transduction Laboratories, dilution 1:500)
and total ERKs (Santa Cruz) were used to normalize for equal load-
ing. For
separated on a 10% polyacrylamide gels and blotted. Western blots
where shown for acetylated -tubulin (Sigma, dilution 1:500) and
a-Tubulin acetylation 25 lg of total protein extracts were
a
total ERKs (Santa Cruz) or total tubulin (Sigma) were used to nor-
malise for equal loading.
6.8. Histone extraction protocol
6.6. Biological assays
Cells were harvested and washed twice with ice-cold PBS and
lysed in Triton Extraction Buffer (TEB: PBS containing 0.5% Triton
X 100 (v/v), 2 mM phenyl methyl sulfonyl fluoride (PMSF), 0.02%
(w/v) NaN3) at a cellular density of 107 cells per mL for 10 min
6.6.1. Cell culture
Human leukaemia cell lines U937, K562 and HL60 were propa-
gated in RPMI medium supplemented with 10% FBS (Foetal bovine
on ice, with gentle stirring. After
a brief centrifugation at
serum; Hyclone) and antibiotics (100 U/mL penicillin, 100
l
g/mL
2000 rpm at 4 °C, the supernatant was removed and the pellet
was washed in half the volume of TEB and centrifuged as before.
The pellet was resuspended in 0.2 M HCl at a cell density of
4 ꢂ 107 cells per mL and acid extraction was left to proceed over-
night at 4 °C on a rolling table. Next, the samples were centrifuged
at 2000 rpm for 10 min at 4 °C, the supernatant was removed and
protein content was determined using the Bradford assay.
streptomycin and 250 ng/mL amphotericin-B). Cells were kept at
the constant concentration of 200,000 cells per mL of culture
medium.
6.6.2. Ligands and materials
SAHA (Merck) and MS-275 (a kind gift of Bayer-Schering AG)
were dissolved in DMSO and used at 5 ꢂ 10ꢁ6 M. All other com-
pounds described were dissolved in DMSO (Sigma-Aldrich) and
6.9. Determination of histone H3 specific acetylations
used at 5 and 50 lM.
For the histone H3 acetylation in U937 cells, 10 lg of histone
6.6.3. Cell cycle analysis
extract was separated on 15% polyacrylamide gels and blotted.
Western blots were shown for pan-acetylated histone H3
(Upstate).
2.5 ꢂ 105 cells were collected and resuspended in 500
l
L of
hypotonic buffer (0.1% Triton X-100, 0.1% sodium citrate, 50
lg/
mL propidium iodide, RNAse A). Cells were incubated in the dark
for 30 min. Samples were acquired on a FACS-Calibur flow cytom-
eter using the Cell Quest software (Becton Dickinson) and analysed
with standard procedures using the Cell Quest software (Becton
Dickinson) and the ModFit LT version 3 Software (Verity) as previ-
ously reported.101,102 Apoptosis was revealed by monitoring nucle-
ar fragmentation (the so-called ‘sub-G1 DNA peak’) by FACS and
analysed by Cell Quest technology.
6.10. Fluorimetric human recombinant HDAC1 assays
GST-HDAC1 has been cloned into the pAcG2T baculovirus trans-
fer vector (BD) and purified by using glutathione beads. The BD
BaculoGold transfection system (BD) has been used in Sf9 insect
cells for expression following supplier’s instructions. The HDAC as-
say has been carried out as follows: the HDAC Fluorescent Activity
Assay is based on the Fluor de Lys Substrate and Developer combi-
nation (BioMol) and has been carried out according to supplier’s
instructions. Briefly, the Fluor de Lys Substrate, which comprises
an acetylated lysine side chain, has been incubated with the puri-
fied recombinant HDAC enzymes in presence or absence of the
inhibitors, for 0.5 h at 37 °C. When a different incubation time
has been used, it is specified into the text. Deacetylation of the sub-
strate sensitizes the substrate so that, in the second step, treatment
6.7. Caspase 3 activation assay
Caspase activity was detected in living U937 cells using the BIO-
MOL and B-BRIDGE Kits supplied with cell-permeable fluorescent
substrates. The fluorescent substrate for caspase 3 was FAM-
DEVD-FMK. ca. 1 ꢂ 106 cells were washed twice in cold PBS and
incubated for 1 h in ice with the corresponding substrates as rec-

3648
J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
with the Developer for 30 min produces a fluorophore. The fluoro-
phore is excited with a 360 nm light and the emitted light
(460 nm) has been quantified with a TECAN Inphinite M200
station.
the same amount of protein extracts were immunoprecipitated
with purified IgG rabbit (Santa Cruz). The following day, 50 L A/
G plus agarose were added and incubation was continued for 2 h.
l
The beads were recovered by brief centrifugation and washed with
cold TAP buffer several times. After the last washing 20 lL of 2X
6.11. Human recombinant Sirt1 assay
concentrated electrophoresis sample buffer (217 mM Tris–HCl pH
8.0, 5.3% SDS, 17.4% glycerol, 8.7% b-mercaptoethanol, 0.026% bro-
mophenol blue) was added and the sample was boiled for 5 min. A
fraction of supernatants was loaded onto an SDS–PAGE gel in order
Recombinant human Sirt1 was prepared in E. coli BL21 and puri-
fied by affinity chromatography. The enzymatic reaction consisted
of 1
l
g of Sirt1 incubated with the acetylated p53 peptide (AA
to check the immunoprecipitation product. 10 lL of resin binding
379–382), 1 mM dithiothreitol, and a range of inhibitor concentra-
DNMT1 were used in DNMT radioactive assay (see above) to test
tions, as described. Reactions were carried out at 37 °C for 60 min.
the inhibitory potency of the PsA derivatives.
Assays were performed in the presence of 200 l
M NAD+ for each
inhibitor. Fluorescence was measured with a fluorimetric reader
(TECAN Inphinite M200 fluorescence plate reader) with excitation
set at 360 nm and emission detection set at 450 nm. Results are
expressed as the mean and standard deviation of four independent
experiments as percentage of activity considering the untreated
control as 100.
Acknowledgments
This work was supported by the EU (Epitron LSHC-CT2005-
518417; JG, RP, GLF and AN contracts; and ATLAS Contract
221952), the Spanish MICINN (SAF-07-63880-FEDER), Xunta de
Galicia (INBIOMED), and the Italian Associazione Italiana per la
ricerca contro il cancro.
6.12. Human recombinant CBP assay
Supplementary data
The recombinant CBP was prepared in E. coli BL21 and purified
by affinity chromatography. Recombinant CBP fraction corre-
sponded to amino acids 1098–1877. CBP was incubated in HAT
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.12.026.
buffer with 10
acids 2–24) and 20
Acetyl CoA in the presence of inhibitors. After 2 h a 37 °C, 5
l
g of histone H4 peptide (corresponding to amino
M Acetyl CoA containing 0.5
Ci/mL [3H]-
L of
l
l
References and notes
l
samples were spotted onto Whatman P81 paper (in triplicate).
The paper squares were washed three times in 5% TCA and once
in 100% acetone and then placed into scintillation vials containing
scintillation fluid to allow the DPM reading. The DPM of enzyme
samples was compared to DPM of negative control. Data have been
expressed as percentage of activity considering the control without
treatment as 100.
1. Peng, J.; Li, J.; Hamann, M. T. Alkaloids: Chem. Biol. 2005, 61, 59.
2. Jiménez, C.; Crews, P. Tetrahedron 1991, 47, 2097.
3. Piña, I. C.; Gautschi, J. T.; Wang, G. Y. S.; Sanders, M. L.; Schmitz, F. J.; France,
D.; Cornell-Kennon, S.; Sambucetti, L. C.; Remiszewski, S. W.; Perez, L. B.; Bair,
K. W.; Crews, P. J. Org. Chem. 2003, 68, 3866.
4. Arabshahi, L.; Schmitz, F. J. J. Org. Chem. 1987, 52, 3584.
5. Rodriguez, A. D.; Akee, R. K.; Scheuer, P. J. Tetrahedron Lett. 1987, 28, 4989.
6. Quiñoà, E.; Crews, P. Tetrahedron Lett. 1987, 28, 3229.
7. Suzuki, A.; Matsunaga, K.; Shin, H.; Tabudrav, J.; Shizuri, Y.; Ohizumi, Y. J.
Pharmacol. Exp. Ther. 2000, 292, 725.
8. Shin, J.; Lee, H. S.; Seo, Y.; Rho, J. R.; Cho, K. W.; Paul, V. J. Tetrahedron 2000, 56,
9071.
6.13. Human recombinant DNMT3A radioactive assay
9. Pham, N. B.; Butler, M. S.; Quinn, R. J. J. Nat. Prod. 2000, 63, 393.
10. Tabudravu, J. N.; Eijsink, V. G. H.; Gooday, G. W.; Jaspars, M.; Komander, D.;
Legg, M.; Synstad, B.; Van Aalten, D. M. F. Bioorg. Med. Chem. 2002, 10, 1123.
11. Jung, J. H.; Sim, C. J.; Lee, C.-O. J. Nat. Prod. 1995, 58, 1722.
12. Park, Y.; Liu, Y.; Hong, J.; Lee, C. O.; Cho, H.; Kim, D. K.; Im, K. S.; Jung, J. H. J.
Nat. Prod. 2003, 66, 1495.
13. Kim, D.; Lee, I. S.; Jung, J. H.; Yang, S. I. Arch. Pharm. Res. 1999, 22, 25.
14. Kim, D.; Lee, I. S.; Jung, J. H.; Lee, C. O.; Choi, S. U. Anticancer Res. 1999, 19,
4085.
15. Nicholas, G. M.; Eckman, L. L.; Ray, S.; Hughes, R. O.; Pfefferkorn, J. A.;
Barluenga, S.; Nicolaou, K. C.; Bewley, C. A. Bioorg. Med. Chem. Lett. 2002, 12,
2487.
16. Jiang, Y.; Ahn, E. Y.; Ryu, S. H.; Kim, D. K.; Park, J. S.; Yoon, H. J.; You, S.; Lee, B.
J.; Lee, D. S.; Jung, J. H. BMC Cancer 2004, 4, 70.
17. Fajas, L.; Egler, V.; Reiter, R.; Hansen, J.; Kristiansen, K.; Debril, M.-B.; Miard,
S.; Auwerx, J. Dev. Cell 2002, 3, 903.
18. Fajas, L.; Egler, V.; Reiter, R.; Miard, S.; Lefebvre, A.-M.; Auwerx, J. Oncogene
2003, 22, 4186.
19. Fu, M.; Rao, M.; Bouras, T.; Wang, C.; Wu, K.; Zhang, X.; Li, Z.; Yao, T.-P.;
Pestell, R. G. J. Biol. Chem. 2005, 280, 16934.
DNMT3A was produced in E. coli BL21 according to standard
procedures. The methyltransferase radioactive assay was per-
formed in a volume of 25
l
L/point, using [3H]-adenosyl-
L-methio-
nine (1 Ci) as methyl donor and Poly dI-dC (0.1
l
c) as methyl
acceptor, while 30–50 ng of recombinant DNMT3A protein was
used, depending on enzyme activity, stability and purity. The com-
pounds were tested at 50 lM. After 2 h incubation at 37 °C, 5 lL of
samples were spotted onto Whatman DE81 paper (in triplicate).
The paper squares were washed three times in 5% Na2HPO4 and
once in sterile water and then placed into scintillation vials con-
taining 5 mL of scintillation fluid to allow the DPM reading. The
DPM of enzyme samples was compared to the DPM of negative
control. Data were expressed as a percentage of activity relative
to control.
20. Mora, F. D.; Jones, D. K.; Desai, P. V.; Patny, A.; Avery, M. A.; Feller, D. R.;
Smillie, T.; Zhou, Y. D.; Nagle, D. G. J. Nat. Prod. 2006, 69, 547.
21. Shim, J. S.; Lee, H.-S.; Shin, J.; Kwon, H. J. Cancer Lett. 2004, 203, 163.
22. Allis, C. D.; Jenuwein, T.; Reinberg, D.; Caparros, M.-L. Epigenetics; Cold Spring
Harbor Laboratory Press: Cold Spring Harbor (NY), 2007.
23. Sippl, W.; Jung, M. In Epigenetic Targets in Drug Discovery: Methods and
Principles of Medicinal Chemistry; Mannhold, R., Kubinyi, H., Folkers, G., Eds.;
Wiley: Weinhem, 2009; Vol. 42.
24. Archer, S. Y.; Hodin, R. A. Curr. Opin. Gene Dev. 1999, 9, 171.
25. Marks, P. A.; Rifkind, R. A.; Richon, V. M.; Breslow, R.; Miller, T.; Kelly, W. K.
Nat. Rev. Cancer 2001, 1, 194.
26. Rosato, R. R.; Grant, S. Cancer Biol. Ther. 2003, 2, 30.
27. Blanchard, F.; Chipoy, C. Drug Discovery Today 2005, 10, 197.
28. Bolden, J. E.; Peart, M. J.; Johnstone, R. W. Nat. Rev. Drug Disc. 2006, 5, 769.
29. Lin, H. Y.; Chen, C. S.; Lin, S. P.; Weng, J. R.; Chen, C. S. Med. Res. Rev. 2006, 26,
397.
6.14. Immunoprecipitation of DNMT1 and radioactive assay
The K562 cells were cultured in the experimental conditions re-
ported and lysed in TAP buffer pH 7–7.5 (50 mM Tris pH 7.0,
180 mM NaCl, 0.15% v/v NP40, 10% v/v glycerol, 1.5 mM MgCl2,
1 mM Na2MoO4, 0.5 mM NaF, 1 mM DTT, 0.2 mM PMSF, 0.1 mM
protease inhibitor cocktail) for 10 min in ice and centrifuged at
130,000 rpm for 30 min. 650
buffer up to 1 mL and pre-cleared by incubating with 20
plus agarose (Santa Cruz) for 30 min to 1 h on a rocking table at
4 °C. Supernatant was transferred to a new tube and 3–5 g of anti-
lg of extracts were diluted in TAP
lL A/G
l
body against DNMT1 (Abcam) was added. IP was allowed to pro-
ceed overnight at 4 °C on a rocking table. As a negative control
30. Mei, S.; Ho, A. D.; Mahlknecht, U. Int. J. Oncol. 2004, 25, 1509.

J. García et al. / Bioorg. Med. Chem. 19 (2011) 3637–3649
74. Ahn, M. Y.; Jung, J. H.; Na, Y. J.; Kim, H. S. Ginecol. Oncol. 2008, 108, 27.
3649
31. Minucci, S.; Pelicci, P. G. Nat. Rev. Cancer 2006, 6, 38.
32. Fouladi, M. Cancer Invest. 2006, 24, 521.
75. Richon, V. M.; Sandhoff, T. W.; Rifkind, R. A.; Marks, P. A. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 10014.
76. Mai, A.; Massa, S.; Rotili, D.; Cerbara, I.; Valente, S.; Pezzi, R.; Simeoni, S.;
Ragna, R. Med. Res. Rev. 2005, 25, 261.
77. Furumai, R.; Matsuyama, A.; Kobashi, N.; Lee, K. H.; Nishiyama, M.; Nakajima,
H.; Tanaka, A.; Komatsu, Y.; Nishino, N.; Yoshida, M.; Horinouchi, S. Cancer
Res. 2002, 62, 4916.
33. Botrugno, O. A.; Santoro, F.; Minucci, S. Cancer Lett. 2009, 280, 134.
34. Mai, A.; Altucci, L. Int. J. Biochem. Cell Biol. 2009, 41, 199.
35. Kazantsev, A. G.; Thompson, L. M. Nat. Rev. Drug Disc. 2008, 7, 854.
36. Marks, P. A.; Dokmanovic, M. Expert Opin. Invest. Drugs 2005, 14, 1497.
37. Drummond, D. C.; Noble, C. O.; Kirpotin, D. B.; Guo, Z.; Scott, G. K.; Benz, C. C.
Ann. Rev. Pharmacol. Toxicol. 2005, 45, 495.
38. Rodriquez, M.; Aquino, M.; Bruno, I.; De Martino, G.; Taddei, M.; Gomez-
Paloma, L. Curr. Med. Chem. 2006, 13, 1119.
78. Xiao, J. J.; Byrd, J.; Marcucci, G.; Grever, M.; Can, K. K. Rapid Commun. Mass
Spectrom. 2003, 17, 757.
39. Paris, M.; Porcelloni, M.; Binaschi, M.; Fattori, D. J. Med. Chem. 2008, 51, 1505.
40. Smith, K. T.; Workman, J. L. Int. J. Biochem. Cell Biol. 2009, 41, 21.
41. Witt, O.; Lindemann, R. Cancer Lett. 2009, 280, 123.
79. Wang, D.; Helquist, P.; Wiest, O. J. Org. Chem. 2008, 72, 5446.
80. Nishino, N.; Jose, B.; Okamura, S.; Ebisusaki, S.; Kato, T.; Sumida, Y.; Yoshida,
M. Org. Lett. 2003, 5, 5079.
42. Klose, R. J.; Bird, A. P. Trends Biochem. Sci. 2006, 31, 89.
43. Ooi, S. K. T.; Bestor, T. H. Cell 2008, 133, 1145.
44. Baylin, S.; Bestor, T. H. Cancer Cell 2002, 1, 299.
45. Esteller, M. Oncogene 2002, 21, 5427.
46. Das, P. M.; Singal, R. J. Clin. Oncol. 2004, 22, 4632.
81. Nishino, N.; Shivashimpi, G. M.; Soni, P. B.; Bhuiyan, M. P. I.; Kato, T.; Maeda,
S.; Nishino, T. G.; Yoshida, M. Bioorg. Med. Chem. 2008, 16, 437.
82. Bach, R. D.; Dmitrenko, O.; Thorpe, C. J. Org. Chem. 2008, 73, 12.
83. Rautio, J.; Kumpulainen, H.; Heimbach, T.; Oliyai, R.; Oh, D.; Järvinen, T.;
Savolainen, J. Nat. Rev. Drug Disc. 2008, 7, 255.
47. Baylin, S. B. Nat. Clin. Pract. Oncol. 2005, 2, S4.
84. Mai, A.; Cheng, D.; Bedford, M. T.; Valente, S.; Nebbioso, A.; Perrone, A.;
Brosch, G.; Sbardella, G.; De Bellis, F.; Miceli, M.; Altucci, L. J. Med. Chem. 2008,
51, 2279.
85. Li, J. W. H.; Vederas, J. C. Science 2009, 325, 161.
86. Newman, D. J. J. Med. Chem. 2008, 51, 2589.
48. Brueckner, B.; Lyko, F. Trends Pharm. Sci. 2004, 25, 551.
49. Kaminskas, E.; Farrell, A.; Abraham, S.; Baird, A.; Hsieh, L.-S.; Lee, S.-L.;
Leighton, J. K.; Patel, H.; Rahman, A.; Sridhara, R.; Wang, Y.-C.; Pazdur, R. Clin.
Cancer Res. 2005, 11, 3604.
50. Morphy, R.; Kay, C.; Rankovic, Z. Drug Discovery Today 2004, 9, 641.
51. Morphy, R.; Rankovic, Z. J. Med. Chem. 2005, 48, 6523.
52. Remiszewski, S. W. Curr. Med. Chem. 2003, 10, 2393.
53. Hoshino, O.; Murakata, M.; Yamada, K. Bioorg. Med. Chem. Lett. 1992, 2, 1561.
54. Nicolaou, K. C.; Hughes, R.; Pfefferkorn, J. A.; Barluenga, S. Chem. Eur. J. 2001, 7,
4296.
87. Molinski, T. F.; Dalisay, D. S.; Lievens, S. L.; Saludes, J. P. Nat. Rev. Drug Disc.
2009, 8, 69.
88. Kotoku, N.; Tsujita, H.; Hiramatsu, A.; Mori, C.; Koizumi, N.; Kobayashi, M.
Tetrahedron 2005, 61, 7211.
89. Knapp, S.; Amorelli, B.; Darout, E.; Ventocilla, C. C.; Goldman, L. M.; Huhn, R.
A.; Minnihan, E. C. J. Carbohydr. Chem. 2005, 24, 103.
55. Nicolaou, K. C.; Hughes, R.; Pfefferkorn, J. A.; Barluenga, S.; Roecker, A. J. Chem.
Eur. J. 2001, 7, 4280.
90. Frisch, M. J.; G. W. T., Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J.
R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.;
Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.;
Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, Malick, O.; D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian, Pittsburg: 2004.
56. Godert, A. M.; Angelino, N.; Woloszynska-Read, A.; Morey, S. R.; James, S. R.;
Karpf, A. R.; Sufrin, J. R. Bioorg. Med. Chem. Lett. 2006, 16, 3330.
57. Boehlow, T. R.; Spilling, C. D. Nat. Prod. Lett. 1995, 7, 1.
58. Boehlow, T. R.; Harburn, J. J.; Spilling, C. D. J. Org. Chem. 2001, 66, 3111.
59. Carreño, M. C. R.; J.L.G.; Sanz, G.; Toledo, M. A.; Urbano, A. Synlett 1997, 1241.
60. Pfammatter, M. J.; Siljegovic, V.; Darbre, T.; Keese, R. Helv. Chim. Acta 2001, 84,
678.
61. Nishiyama, S.; Yamamura, S. Tetrahedron Lett. 1982, 23, 1281.
62. Forrester, A. R. T.; R.H.; Woo, S. O. J. Chem. Soc., Perkin Trans. 1 1975, 2340.
63. Khan, N.; Jeffers, M.; Kumar, S.; Hackett, C.; Boldog, F.; Khramtsov, N.; Qian, X.;
Mills, E.; Berghs, S. C.; Carey, N.; Finn, P. W.; Collins, L. S.; Tumber, A.; Ritchie,
J. W.; Jensen, P. B.; Lichenstein, H. S.; Sehested, M. Biochem. J. 2008, 409, 581.
64. Bradner, J. E.; Mak, R.; Tanguturi, S. K.; Mazitschek, R.; Haggarty, S. J.; Ross, K.;
Chang, C. Y.; Bosco, J.; West, N.; Morse, E.; Lin, K.; Shen, J. P.; Kwiatkowski, N.
P.; Gheldof, N.; Dekker, J.; DeAngelo, D. J.; Carr, S. A.; Schreiber, S. L.; Golub, T.
R.; Ebert, B. L. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 12617.
65. Balasubramanyam, K.; Swaminathan, V.; Ranganathan, A.; Kundu, T. K. J. Biol.
Chem. 2003, 278, 19134.
91. Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A. J. Phys. Chem. 1993, 97,
10269.
92. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M.; Ferguson, D.
M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc.
1995, 117, 5179.
66. (a) Souto, J. A.; Conte, M.; Alvarez, R.; Nebbioso, A.; Carafa, V.; Altucci, L.; de
Lera, A. R. ChemMedChem 2008, 3, 1435; (b) Souto, J. A.; Bebedetti, R.; Otto, K.;
Miceli, M.; Alvarez, R.; Altucci, L.; de Lera, A. R. ChemMedChem 2010, 5, 1530;
(c) Souto, J. A.; Vaz, E.; Lepore, I.; Pöppler, A.-C.; Franci, J. L.; Alvarez, R.;
Altucci, L.; de Lera, A. R. J. Med. Chem. 2010, 53, 4654.
93. Rodríguez-Barrios, F.; Pérez, C.; Lobatón, E.; Velázquez, S.; Chamorro, C.; San-
Félix, A.; Pérez-Pérez, M. J.; Camarasa, M. J.; Pelemans, H.; Balzarini, J.; Gago, F.
J. Med. Chem. 2001, 44, 1853.
94. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R. K.;
Olson, A. J. J. Comput. Chem. 1998, 19, 1639.
67. Siedlecki, P.; Boy, R. G.; Musch, T.; Brueckner, B.; Suhai, S.; Lyko, F.;
Zielenkiewicz, P. J. Med. Chem. 2006, 49, 678.
95. Morris, G. M.; Goodsell, D. S.; Huey, R.; Hart, W. E.; Halliday, S.; Belew, R.;
Olson, A. J.; AutoDock, 3.0 ed. La Jolla, CA, 1999.
68. Datta, J.; Ghoshal, K.; Denny, W. A.; Gamage, S. A.; Brooke, D. G.; Phiasivongsa,
P.; Redkar, S.; Jacob, S. T. Cancer Res. 2009, 69, 4277.
96. Goodford, P. J. J. Med. Chem. 1985, 28, 849.
97. Åqvist, J. J. Phys. Chem. 1990, 94, 8021.
69. Biel, M.; Wascholowski, V.; Giannis, A. Angew. Chem., Int. Ed. 2005, 44,
3186.
98. Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D. J. Chem. Phys. 1983, 79, 926.
99. http://amber.scripps.edu/doc8.
70. 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, R.;
Leahy, E. M.; Dougan, D. R.; Snell, G.; Navre, M.; Knuth, M. W.; Swanson, R. V.;
McRee, D. E.; Tari, L. W. Structure 2004, 12, 1325.
100. Ryckaert, J. P.; Ciccoti, G.; Berendsen, H. J. C. J. Comput. Phys. 1977, 23, 327.
101. Nebbioso, A.; Clarke, N.; Voltz, E.; Germain, E.; Ambrosino, C.; Bontempo, P.;
Alvarez, R.; Schiavone, E. M.; Ferrara, F.; Bresciani, F.; Weisz, A.; de Lera, A. R.;
Gronemeyer, H.; Altucci, L. Nat. Med. 2005, 11, 77.
71. Bieliauskas, A. V.; Pflum, M. K. H. Chem. Soc. Rev. 2008, 37, 1402.
72. McCulloch, M. W. B.; Coombs, G. S.; Banerjee, N.; Bugni, T. S.; Cannon, K. M.;
Harper, M. K.; Veltri, C. A.; Virshup, D. M.; Ireland, C. M. Bioorg. Med. Chem.
2009, 17, 2189.
102. Scognamiglio, A.; Nebbioso, A.; Manzo, F.; Valente, S.; Mai, A.; Altucci, L.
Biochem. Biophys. Acta 2008, 1783, 203.
103. Altucci, L.; Rossin, A.; Raffelsberger, W.; Reitmair, A.; Chomienne, C.;
Gronemeyer, H. Nat. Med. 2001, 7, 680.
73. Kim, D. H. S. J.; Kwon, H. J. Exp. Mol. Med. 2007, 39, 47.