Design, synthesis and biological evaluation of 1,4-benzodiazepine-2,5-dione-based HDAC inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.06.067

New histone deacetylase inhibitors have been synthesized and evaluated for their activity against non-small lung cancer cell line H661. These compounds have been designed with diversely substituted 1,4-benzodiazepine-2,5-dione moieties as cyclic peptide m

Full text of DOI:10.1016/j.bmcl.2007.06.067

Bioorganic & Medicinal Chemistry Letters 17 (2007) 4819–4823
Design, synthesis and biological evaluation of
1,4-benzodiazepine-2,5-dione-based HDAC inhibitors
Lynda Loudni,a Joe¨lle Roche,b Vincent Potiron,b Jonathan Clarhaut,b
Christian Bachmann,c Jean-Pierre Gessona and Isabelle Tranoy-Opalinskia,*
aUniversite´ de Poitiers, CNRS UMR 6514, SRSN, 40 Avenue du Recteur Pineau, 86022 Poitiers, France
bUniversite´ de Poitiers, CNRS UMR 6187,IPBC, 40 Avenue du Recteur Pineau, 86022 Poitiers, France
cUniversite´ de Poitiers, CNRS UMR 6503, LACCO, 40 Avenue du Recteur Pineau, 86022 Poitiers, France
Received 20 April 2007; revised 14 June 2007; accepted 16 June 2007
Available online 27 June 2007
Abstract—New histone deacetylase inhibitors have been synthesized and evaluated for their activity against non-small lung cancer
cell line H661. These compounds have been designed with diversely substituted 1,4-benzodiazepine-2,5-dione moieties as cyclic pep-
tide mimic cap structures, and a hydroxamate side chain. Biological evaluations demonstrated that benzodiazepine-based HDACi
bearing an aromatic substituent at the N1 position exhibited promising antiproliferative and HDAC-inhibitory activities.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Reversible acetylation of histones is one of the most
studied epigenetic mechanisms.1,2 This post-transla-
tional modification controls chromatin remodelling by
affecting the interactions between histones and DNA
with a considerable impact on the accessibility of
DNA and regulation of gene expression.3 The acetyla-
tion status of core histones is regulated by opposite
activities of two groups of enzymes: HATs (histone acet-
yltransferases) and HDACs (histone deacetylases).
Accumulating evidence indicates that the balance be-
tween acetylation and deacetylation of histones, and
other proteins like p53 or a-tubulin, plays significant
roles in gene transcription and cell proliferation.4 Alter-
ations in the structure or expression of HDACs have
been clearly linked to the pathogenesis of cancer and
consequently HDAC inhibitors (HDACi) have emerged
as a new class of promising chemotherapeutic agents.5,6
Indeed, HDACi selectively induce growth arrest, differ-
entiation and apoptosis in tumour cells through the
transcriptional activation of a small set of genes that
regulate cell proliferation and cell cycle progression like
p21.4,7
interest in this field.8,9 Trichostatin A (TSA) 1 was the
first natural compound detected for its HDAC inhibi-
tion activity.10 Natural trapoxin B (TPX B) 2,11 syn-
thetic cyclic hydroxamic-acid-containing peptide
(CHAP1) 3,12 suberoylanilide hydroxamic acid (SAHA)
4
13 and benzamide 5 (MS-275)14 are representatives of a
set of structurally diverse HDACi (Fig. 1). Undergoing
research for the design of valuable HDACi relies on
crystallographic studies that have pointed out three
structural requirements:15 a terminal group that can
bind to the zinc ion at the bottom of the active site, a
cap group that makes contacts with the pocket entrance
and between both, a linker fitting the tube-like portion
of the binding pocket.
To date, at least 18 HDACs have been identified and
classified into four classes. HDACi are able to inhibit
the activity of class 1, class 2 and class 4 Zn-dependent
HDACs, with no marked preference between classes
and HDACs. Moreover, little is known about the bio-
logical functions of each HDAC enzyme and about
the HDACi structure-relationship associated with
HDAC class selectivity.5 To address this issue, it is
important to identify new isoform-selective inhibitors
of HDAC. Within this framework, specific modulation
of the capping structure seems to be appropriate to en-
able access to selective inhibitors.16
In the last 10 years, the large amount of publications
and the wide diversity of natural and synthetic HDACi
structures which have been reported testify the great
Keywords: Histone deacetylase; HDAC Inhibitors; 1,4-Benzodiaze-
pine-2,5-dione; Cyclic peptide mimic.
In this paper, we report the design, synthesis and biolog-
ical evaluations of novel HDACi containing 1,4-benzo-
*
Corresponding author. E-mail: isabelle.tranoy@univ-poitiers.fr
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.06.067

4820
L. Loudni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4819–4823
cap
linker
binding
group
O
O
O
OH
N
N
R
O
N
H
H
N
O
O
N
H
HN
1 TSA
O
O
O
H
N
R =
2 TPX B
N
OH
H
4 SAHA
O
O
R = NHOH
3 CHAP1
O
O
H
NH2
H
O
N
H
N
4
1
N
OH
N
N
R
N
O
5 MS-275
O
BZD
Figure 1. Examples of HDACi.
diazepine-2,5-dione scaffold. Our choice for such hetero-
cycles was motivated by the fact that 1,4-benzodiaze-
zodiazepines 10 and an appropriate hydroxamate pre-
cursor 11. Furthermore, we anticipated that such a
strategy will allow examining further the SAR of dif-
ferent methylene chain lengths. Compounds 10a–d
resulting from N4-alkylation with allyl bromide were
isolated in good yields together with only a small
amount of O-alkylated products. Cross-metathesis be-
tween compounds 10a–d and alkene 1121 in the pres-
ence of Grubbs II catalyst led to intermediates 12a–
d as a mixture of diastereoisomers. In all cases, the
reaction did not reach completion. Compounds 12a
and 12b were isolated after purification but com-
pounds 12c and 12d were obtained as a mixture with
starting material. Nevertheless, formation of these lat-
ter was confirmed by HRMS analysis. Hydrogenation
of 12a–d led to compounds 13a–d and deprotection of
the tert-butoxycarbonyl (Boc) group with trifluoroace-
tic acid gave the free hydroxamic acid of saturated
analogues BZDSa–d. In order to compare the nature
of the side chain, similar Boc-deprotection was con-
ducted on compounds 12a and 12b to give unsatu-
rated analogues BZDIa and b.
pine-2,5-diones
are
used
as
constrained
peptidomimetics and provide a synthetically versatile
platform for structural modification.17 Thus, by analogy
with the CHAP structures, we focused our design on
cyclic hydroxamic-acid-containing peptidomimetic, such
as BZD analogues (Fig. 1). Modulation of the BZD R
group will provide different cap structures in order to
evaluate the influence of the N1-benzodiazepine substi-
tution on the resulting activity. It is worth mentioning
that a similar approach has been recently reported by
Etzkorn et al., who have used with success other types
of cyclic peptide mimics.18
The 1,4-benzodiazepine-2,5-dione
8 was prepared
according to the Sun procedure by condensation of
isatoic anhydride 6 with L-phenylalanine 7 (Scheme
1).19 The synthetic route for our BZD analogues relies
on Blass’s versatile method for the selective N-alkyl-
ation of a central 1,4-benzodiazepine-2,5-dione core
using potassium fluoride on alumina.20 To finalize
the cap structure, N1-alkylation of 8 was accom-
plished with 1.0 equiv of various alkyl bromides (R–
Br) in the presence of KF/Al2O3 in DMF at 25 ꢁC
over 48 h. In order to explore the effect of N1-substi-
tuent on antiproliferative and HDAC-inhibitory activ-
ities, we chose different aromatic structures
The benzodiazepinedione nucleus is quite rigid and
adopts a ‘cupped shape’.17 1H NMR spectra of all
compounds obtained after the N4-alkylation (10a–d
to BZD analogues) showed clearly the presence of
two conformers, not chromatographically separable,
at a ratio of 60:40. This observation suggests that
there is a ring inversion which is too rapid to allow
the separation at room temperature, but slow enough
for the detection of conformers by NMR. Performing
ab initio B3-LYP/6-31G(d) calculations22 for 10d, we
found two conformations, endo and exo, according
to the way H3 pointed out in the seven-membered
ring, with small energy difference (Fig. 2). To save
computer time, barrier height for the interconversion
of endo to exo was then evaluated on model A and
(R = benzyl,
3,5-dimethyl-benzyl,
naphthalen-1-yl
methyl) and one alkyl structure (R = methyl). Thus,
four different monoalkylated benzodiazepine-2,5-
diones 9a–d were obtained. From compounds 9, N4-
alkylation to introduce the hydroxamic acid moiety
can be realized using previous KF/Al2O3 methodology
with refluxing DME.20 First attempts by means of a
hydroxamate derivative of bromo-6-hexanoic acid
were unsuccessful. This result prompted us to opt
for a cross-metathesis strategy involving allylated ben-

L. Loudni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4819–4823
4821
OH
NH2
O
O
R
O
R
O
a
b
HN
O
N
c
N
1
O
HN
O
+
4
NH
NH
N
O
O
O
6
7
10a, 97%
10b, 81%
10c, 99%
10d, 52%
8
9a, 61%
9b, 55%
9c, 52%
9d, 46%
a:R = Bn
b:R = 3,5-CH3
c: R = Napht-1-CH2
d: R = CH3
-Bn
O
OBoc
N
d
11
boc
R
O
R
O
O
O
N
N
e
N
OR'
N
OR'
N
N
O
O
R'
R'
13a, R' = Boc, qtif
13b, R' = Boc, qtif
13c, R' = Boc, qtif
12d, R' = Boc, qtif
12a, R' = Boc, 58%
12b, R' = Boc, 42%
12c, R' = Boc, n.d.a
12d, R' = Boc, n.d.a
f
f
BZDSa, R' = H, 45%
BZDSb, R' = H, 41%
BZDSc, R' = H, 55%
BZDSd, R' = H, 34%
BZDIa, R' = H, 54%
BZDIb, R' = H, 48%
Scheme 1. Synthesis of BZD analogues. Reagents and conditions: (a) i—H2O, TEA, rt, 24 h; ii—CH3COOH, reflux, 48 h, 78%; (b) KF/Al2O3 (40%
KF by weight), DMF, R–Br 1 equiv, rt, 48 h, 61–46%; (c) KF/Al2O3 (40% KF by weight), DME, allyl bromide 1.2 equiv, reflux, 48 h, 95–52%; (d) 10
3 · 10À3 mol LÀ1, 11 3 equiv, 10% Grubbs II, reflux, 2 days, 58–42%; (e) H2, Pd/C 10%, MeOH, rt, qtif; (f) TFA, DCM, 0 ꢁC to rt, 3 h, 34–55%.
aYield not determined.
Table 1. Antiproliferative activities in H661 cells for BZD analogues
and TSA
Entry
Compound
R
IC50 (lM)
1
2
3
4
5
6
7
BZDSa
BZDSb
BZDSc
BZDSd
BZDIa
BZDIb
TSA
Bn
6
3,5-CH3–Bn
C-Naphth-1-CH2
CH3
8
16
No activity
Bn
3,5-CH3–Bn
8
17
0.085
The antiproliferative study was performed with six different concen-
trations for each BZD analogue (1, 5, 10, 15, 20, 25 lM) and with
seven different concentrations for TSA (0.025, 0.050, 0.1, 0.2, 0.3, 0.5,
1 lM). For each BZD analogue, three to four independent experiments
were performed in triplicate and five for TSA. The IC50values were
determined from the curves as the concentration that is required for
50% inhibition of cell proliferation measured 48 h after treatment of
H661 cells. For each compound, the curves are provided as Supple-
mentary data.
Figure 2. Conformational interconversion of 10d and structure of
model A.
Antiproliferative activities of BZD analogues were eval-
uated after 48 h treatment of non-small cell lung cancer
H661 cells and compared to TSA and IC50s were deter-
mined (Table 1). The presence of an aromatic substitu-
ent appears to be essential to achieve activity. Indeed,
BZDSd (entry 4) did not present any detectable antipro-
liferative activity for the tested concentrations from 1 to
found to be 14.8 kcal/mol. This relatively high value is
consistent with previously reported data20,23 and can
explain the existence of two conformers at room tem-
1
perature. Moreover, H and 13C NMR chemical shift
calculations24 revealed the same significant differences
for conformers 10d as the experimental ones.

4822
L. Loudni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4819–4823
5.5h 24h
kDa
16
22
acetyl H4
acetyl tubulin
α-tubulin
p21
50
50
α-tubulin
98
64
50
36
total proteins
total proteins
22
16
Figure 3. Western blot analysis for histone H4 and tubulin acetylation, and p21 expression. H661 cells were treated with the indicated compounds for
5.5 h (left) and 24 h (right). A 24-h treatment was necessary to detect p21 induction. Corresponding antibodies were used to detect histone H4 and
tubulin acetylation and p21 induction. a-Tubulin and total protein extract stained by Coomassie blue are loading controls.
25 lM in opposition to BZDSa–c. In the case of aro-
matic-substituted benzodiazepine-2,5-diones, BZDSa
was found to be the most active (entry 1 vs 2 and 3).
BZDSa (IC50 = 6 lM) seems slightly more active for
antiproliferative activity than BZDSb (IC50 = 8 lM)
and BZDIa (IC50 = 8 lM) even if their IC50 are not very
different. Indeed, looking at the proliferation curves
(Supplementary data), we can see that 5 lM BZDSa in-
duces a significant antiproliferative activity (70% to 40%
proliferation of control), whereas at the same dose of
BZDSb or BZDIa, proliferation was around 80–95%
of control.
after 24 h treatment by these two compounds (BZDSa
and BZDSb) and TSA and not by BZDSd (Fig. 3, right).
Biological assays showed a significantly reduced potency
on BZD analogues compared to TSA. In the same way,
BZD analogues are less potent compared with analo-
gous cyclic tetrapeptide-like trapoxin and CHAPs.26
However, it is worth mentioning that the comparison
with trapoxin and CHAP does not concern the same cell
line. Indeed Komatsu et al. described biological assays
performed with B16/BL6 cells. Nevertheless, our results
demonstrate the potential of 1,4-benzodiazepine-2,5-
dione structures to play the capping group of new
HDACi with a promising framework to develop iso-
form-selective HDACi.
Concerning the nature of the side chain, slightly better
activities were found for the saturated analogues
BZDSa and b, compared to corresponding unsaturated,
BZDIa and b (entries 1 and 2 vs 5 and 6).
In conclusion, in order to find new HDACi, we have de-
signed and synthesized cyclic peptidomimetic hydroxa-
mates by the mean of 1,4-benzodiazepine-2,5-dione
central scaffold. Our synthetic strategy allows conve-
nient preparation of these compounds and wide possibil-
ities to create diverse cap structures. This preliminary
work pointed out the need of an aromatic substituent at-
tached to the N1-benzodiazepine nucleus to achieve an
antiproliferative activity and the more promising com-
pounds BZDSa and BZDSb were identified as HDACi.
Complementary SAR studies are now needed to investi-
gate different substituted cap groups and their ability to
selectively inhibit single HDAC enzymes.
As histone H4 and a-tubulin acetylation are markers of
HDACi activity, we performed western blot analysis to
evaluate the HDACi potential of BZDSa and BZDSb
active in proliferation test and inactive BZDSd, with
TSA as reference (Fig. 3, left). As expected, TSA in-
duced histone H4 acetylation, whereas BZDSd was
barely active. Two doses around IC50 were tested (5
and 10 lM) for BZDSa and BZDSb. BZDSa was more
active to acetylate histone H4 than BZDSb after 5.5 h
treatment. We also notice a dose–response in accor-
dance with the antiproliferative XTT assay. Interest-
ingly, with these two compounds, the level of tubulin
acetylation is quite similar. We can hypothesize that
HDAC6 that reverses the post-translational acetylation
of tubulin25 is similarly inhibited by these two com-
pounds. p21 (the cyclin-dependent kinase inhibitor,
WAF1), which is induced by HDACi, could be detected
Acknowledgments
The authors thank MENESR, CNRS and La Ligue
´ ´
Nationale contre le Cancer, Comites de la region Poi-

L. Loudni et al. / Bioorg. Med. Chem. Lett. 17 (2007) 4819–4823
4823
21. (a) Staszak, M. A.; Doecke, C. W. Tetrahedron Lett. 1993,
34, 7043; (b) Staszak, M. A.; Doecke, C. W. Tetrahedron
Lett. 1994, 35, 6021.
tou—Charentes for financial support of this study. The
authors also thank S. Sissoko for technical help during
biological assays.
22. (a) Theoretical calculations at the density functional B3-
LYP/6-31G(d) level were performed using the Gaussian 98
package.22b; (b) Frisch, M. J.; Trucks, G. W.; Schlegel, H.
B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.;
Zakrzewski, V. G.; Montgomery, J. A.; Stratmann, Jr., R.
E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli,
C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.;
Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.;
Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gom-
perts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M.A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian 98, Revision A.7, Gaussian Inc.,
Pittsburgh, PA, 1998.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2007.06.067.
References and notes
1. Mai, A.; Massa, S.; Rotili, D.; Gerbara, I.; Valente, S.;
Pezzi, R.; Simeoni, S.; Ragno, R. Med. Res. Rev. 2005, 25,
261.
2. Biel, M.; Wascholowski, V.; Giannis, A. Angew. Chem.,
Int. Ed. 2005, 44, 3186.
3. Strahl, B. D.; Allis, C. D. Nature 2000, 403, 41.
4. Minucci, S.; Pelicci, P. G. Nat. Rev. Cancer 2006, 6,
38.
23. Jadidi, K.; Aryan, R.; Mehrdad, M.; Lugger, T.; Hahn, F.
¨
E.; Weng Ng, S. J. Mol. Struct. 2004, 692, 37.
5. Marks, P.; Rifkind, R. A.; Richon, V. M.; Breslow, R.;
Miller, T.; Kelly, W. K. Nat. Rev. Cancer 2001, 1, 194.
6. Marks, P. A. Oncogene 2007, 26, 1351.
7. Lin, H. Y.; Chen, C. S.; Lin, S. P.; Weng, J. R.; Chen, C.
S. Med. Res. Rev. 2006, 26, 397.
8. Miller, T. A.; Witter, D. J.; Belvedere, S. J. Med. Chem.
2003, 46, 5097.
9. Monneret, C. Eur. J. Med. Chem. 2005, 40, 1.
10. Yoshida, M.; Kijima, M.; Akita, T.; Beppu, T. J. Biol.
Chem. 1990, 265, 17174.
24. (a) To obtain reliable chemical shifts, we have used the
GIAO method available in Gaussian 98. These chemical
shifts were calculated at the Hartree-Fock level using the
6-311+G(2d,p) extended basis set on the B3-LYP/6-
31G(d) optimized structures, as recommended by Cheese-
man et al.24b; (b) Cheeseman, J. R.; Trucks, G. W.; Keith,
T. A.; Frisch, M. J. J. Chem. Phys. 1996, 104, 5497; (c) 1H
and 13C NMR calculations for 10d endo and 10d exo
11. Kijima, M.; Yoshida, M.; Sugita, K.; Horinouchi, S.;
Beppu, T. J. Biol. Chem. 1993, 268, 22429.
12. Komastu, Y.; Tomizaki, K. Y.; Tsukamoto, M.; Kato, T.;
Nishino, N.; Sato, S.; Yamori, T.; Tsuruo, T.; Furumai,
R.; Yoshida, M.; Horinouchi, S.; Hayashi, H. Cancer Res.
2001, 61, 4459.
13. Richon, V. M.; Webb, Y.; Merger, R.; Sheppard, T.;
Jursic, B.; Ngo, L.; Civoli, F.; Breslow, R.; Rifkind, R.
A.; Marks, P. A. Proc. Natl. Acad. Sci. U.S.A. 1996, 93,
5705.
Hg
O
Me
N
H
2'''
3
1'''
N
O
H c
3'''
H
f
H
Ha
H
1'
He
H
d
b
10d exo
(ppm)
2.38
10d endo
(ppm) (ppm)
δobs
14. Suzuki, T.; Ando, T.; Tsuchiya, K.; Fukazawa, N.; Saito,
A.; Mariko, Y.; Yamashita, T.; Nakanishi, O. J. Med.
Chem. 1999, 42, 3001.
15. Finnin, M. S.; Donigian, J. R.; Cohen, A.; Richon, V. M.;
Rifkind, R. A.; Marks, P. A.; Breslow, R.; Pavletich, N. P.
Nature 1999, 401, 188.
16. Grozinger, C. M.; Schreiber, S. L. Chem. Biol. 2002, 9, 3.
17. McDowell, R. S.; Blackburn, B. K.; Gadek, T. R.; McGee,
L. R.; Rawson, T.; Reynolds, M. E.; Robarge, K. D.;
Somers, T. C.; Thorsett, E. D.; Tischler, M.; Webb, R. R.,
II; Venuti, M. C. J. Am. Chem. Soc. 1994, 116, 5077.
18. Liu, T.; Kapustin, G.; Etzkorn, F. A. J. Med. Chem. 2007,
50, 2003.
Position
(ppm)
2.33
δcalcd
δobs
δcalcd
H-1′f
H-1′e
H-1′′′a
H-1′′′b
H-2′′′
H-g
3.22
3.11
2.41
3.21
4.62
6.06
8.82
61.8
2.52
3.43
4.04
5.41
7.88
66.62
3.32
4.08
3.84
6.4
8.44
54
3.41
4.04
4.04
5.79
7.77
57.31
C-3
25. Hubbert, C.; Guardolia, A.; Shao, R.; Kawaguchi, Y.; Ito,
A.; Nixon, A.; Yoshida, M.; Wang, X.; Yao, T. Nature
2002, 417, 455.
19. Sun, H. H.; Barrow, C. J.; Cooper, R. J. Nat. Prod. 1995,
58, 1575.
20. Blass, B. E.; Burt, T. M.; Liu, S.; Portlock, D. E.; Swing,
E. M. Tetrahedron Lett. 2000, 41, 2063.
26. Komatsu, Y.; Tomizaki, K.-Y.; Tsukamoto, M.; Kato, T.;
Nishino, N.; Sato, S.; Yamori, T.; Tsuruo, T.; Furumai,
R.; Yoshida, M.; Horinouchi, S.; Hayashi, H. Cancer Res.
2001, 61, 4459.