Design, synthesis, and preliminary bioactivity studies of substituted purine hydroxamic acid derivatives as novel histone deacetylase (HDAC) inhibitors

Article abstract of DOI:10.1039/c4md00203b

Histone deacetylase (HDAC) is a clinically validated target for anti-tumor therapy. In order to increase HDAC inhibition and efficiency, we developed a series of novel substituted purine hydroxamic acids as potent HDAC inhibitors. The biological evaluation suggests that compound 5r (IC₅₀ = 0.075 μmol L^-1) exhibits better HDAC1 and 2 inhibitory activity compared to the approved drug SAHA (IC₅₀ = 0.14 μmol L^-1). Further biological evaluation indicated that compounds 5r, 5w, and 5x have potent anti-proliferative activities against eight tumor cells, including MDA-MB231, KG1, PC3, U937 and so on.

Full text of DOI:10.1039/c4md00203b

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
Design, synthesis, and preliminary bioactivity
studies of substituted purine hydroxamic acid
derivatives as novel histone deacetylase (HDAC)
inhibitors†
Cite this: DOI: 10.1039/c4md00203b
Junhua Wang,a Feng'e Sun,a Leiqiang Han,a Xuben Hou,a Xiaole Pan,a Renshuai Liu,a
b
a
*
Weiping Tang and Hao Fang
Histone deacetylase (HDAC) is a clinically validated target for anti-tumor therapy. In order to increase HDAC
inhibition and efficiency, we developed a series of novel substituted purine hydroxamic acids as potent
HDAC inhibitors. The biological evaluation suggests that compound 5r (IC50 ¼ 0.075 mmol Lꢀ1) exhibits
better HDAC1 and 2 inhibitory activity compared to the approved drug SAHA (IC50 ¼ 0.14 mmol Lꢀ1).
Further biological evaluation indicated that compounds 5r, 5w, and 5x have potent anti-proliferative
activities against eight tumor cells, including MDA-MB231, KG1, PC3, U937 and so on.
Received 8th May 2014
Accepted 14th September 2014
DOI: 10.1039/c4md00203b
www.rsc.org/medchemcomm
group, a hydrophobic linker, and a zinc-binding group (ZBG)
(Fig. 2). The cap group permits extraordinary variability. The
Introduction
linker group can pass through the tunnel of the active site while
the zinc binding group (ZBG) chelates the active site zinc ion
(Zn2+). The hydroxamate moiety is by far the most common
ZBG, which is also found in naturally occurring HDAC
inhibitors.2,5,6
It is well known that cancer is related to genetic and epigenetic
events which lead to neoplastic transformation. Scientists have
found that multiple stages of tumorigenesis can be inuenced
by epigenetic alterations. In cancer cells, an imbalance of
histone acetylation oen exists due to the over-expression of
histone deacetylase (HDAC), which leads to silencing of the
expression of tumor suppressor genes (such as p21/p53)
through hypoacetylation. HDAC inhibitors (HDACi) can reverse
the situation and induce cell cycle arrest, differentiation, and
apoptosis. HDACi have shown signicant clinical benets for
the treatment of cancers. Continued efforts have been devoted
to the discovery of competitive HDAC inhibitors for chemo-
therapeutic intervention.1–3
To date, more than 20 HDAC inhibitors are under preclinical
and clinical investigation as single agents or in combination
therapies against different cancers. Two HDAC inhibitors,
Zolinza (vorinostat or SAHA) and Istodax (romidepsin or
FK228), have been launched for the treatment of cutaneous T-
cell lymphoma. The latter was also approved for peripheral T-
cell lymphoma (Fig. 1).4
Our group has made great efforts to develop novel HDAC
inhibitors by introducing different heterocycles as the cap
group to improve their efficacy against HDACs. We have
prepared tetrahydroisoquinoline,7–9 N-hydroxy benzamide,10,11
tyrosine-based hydroxamic acid,12 and 1,3,4-thiadiazole13 as
potent HDAC inhibitors. Recently, we found that compounds
with 1,2-dihydrobenzo[d]isothiazol-3-one-1,1-dioxide (saccharin) as
the cap group also showed signicant inhibition to HDAC.14
Purine derivatives15,16 or purine isomers17 have been incorpo-
rated into HDAC inhibitors in a few cases. Systematic investi-
gations have not been conducted on the SAR of HDACi with
different substituted purine derivatives as the cap group.
Recently, our group reported the anti-tumor activities of some
HDAC inhibitors generally conform to a three-motif phar-
macophoric model consisting of a surface recognition cap
aDepartment of Medicinal Chemistry, Key Laboratory of Chemical Biology (Ministry of
Education), School of Pharmacy, Shandong University, 44 West Wenhua Rd, Jinan
250012, PR China. E-mail: haofangcn@sdu.edu.cn
bDivision of Pharmaceutical Sciences, School of Pharmacy, University of Wisconsin,
Madison 53705, USA
† Electronic supplementary information (ESI) available: Experimental procedures,
characterisation data, enzyme inhibition assay, MTT assay and docking method.
See DOI: 10.1039/c4md00203b
Fig. 1 Approved HDAC inhibitors by the FDA.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Chemistry
The synthesis route to our newly designed HDACi is outlined in
Scheme 1. Compounds 1a–1g (ref. 19 and 20) reacted with ethyl
chloroacetate under basic conditions at room temperature to
obtain 2a–2g. Compounds 3a–3g were then prepared aer
saponication and acidication. Carboxylic acids 3a–3g were
next coupled with various amino acid methyl esters to obtain
intermediates 4a–4x. Finally, hydroxamic acids 5a–5x were
synthesized by reaction with freshly prepared potassium
hydroxylamine according to reported procedures.21
Fig. 2 HDACi pharmacophoric model and purine-based hydroxamic
acid derivatives.
substituted purine derivatives.18 To explore the possibility of
synergistic effects, we prepared a series of novel HDACi, where
the substituted purine derivatives were linked with the
hydroxamic acid moiety, and evaluated the inhibitory activities
against HDAC (Fig. 2).
Results and discussion
A Color de Lys™ assay (BML–AK501, Enzo® Life Sciences) was
used to evaluate the inhibitory activities of purine hydroxamic
acids against HDACs which mainly include HDAC1 and HDAC2.
The inhibitory results are summarized in Fig. 3. The IC50 values
of more potent compounds are calculated and tabulated in
Table 1.
We found that the compounds with one to three methylene
units as the linker showed poor enzymatic inhibition even at
5 mmol Lꢀ1, such as 5a–5c, 5e, 5f, 5i, and 5j. When the linker was
phenyl methyl, the compounds (5h and 5l) did not display
inhibition to HDAC either. Compounds bearing four to six
methylene units as the linker (5d, 5g, and 5k) showed better
inhibitory activities. On the other hand, the substitutions on
purine C2 and C6-positions also played an important role in the
inhibitory activities. According to the IC50 values in Table 1,
introducing an aromatic ring in the C2-position of purine could
enhance the potency signicantly. For example, 5v–5x (R1
–NHPh) were better than the corresponding compounds
¼
without C2 substitution (5s–5u, R1 ¼ H). Compound 5r (IC50
¼
0.075 mmol Lꢀ1) behaved much better than 5k (IC50 ¼ 4.53 mmol
Lꢀ1) and the inhibitory activity increased 60-fold. In addition,
aromatic rings on the purine C6-position also increased the
inhibitory activities. For example, compounds 5t and 5d with a
phenyl or benzyl substitution on the purine N6-position showed
much better HDAC inhibitory activities compared with
Scheme 1 Reagents and reaction conditions: (a) ethyl chloroacetate,
K2CO3, DMSO, rt; (b) (i) 2 mol Lꢀ1 NaOH, THF/MeOH ¼ 3 : 1, rt; (ii) 3
mol Lꢀ1 HCl, H2O, rt; (c) amino acid methyl ester hydrochloride; EDCI,
HOBt, TEA, anhydrous DCM, 0 ꢁC – rt; (d) NH2OK, MeOH, rt.
Fig. 3 Inhibition of all compounds against HDAC1 & 2 at 5 mmol Lꢀ1
.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
compound 5g with the same linker (R1 ¼ H, R2 ¼ morpholinyl). Three tumor cell lines, MDA-MB231, MCF7, and KG1, were
The improved activity by the additional aromatic ring substi- tested using MTT (3-[4,5-dimethyl-2-thiazolyl]-2,5-diphenyl-2H-
tution on purine N2 or N6-positions may be due to the addi- tetrazoliumbromide) assay.
tional p–p interactions between the compounds and HDAC.
The results showed that most tested compounds displayed
To study the interaction between the inhibitor and HDAC, obvious anti-proliferative activities against the MDA-MB231 cell
compound 5r and SAHA were docked into the active site of line except for 5s (Table 2). At the same time, compounds 5d, 5k,
HDAC2 (PDB entry: 4LZX) using AutoDock4.2 (ref. 22–26) 5s, 5t, and 5u did not work on the KG1 and MCF7 cell lines. The
(Fig. 4). The results suggested that 5r possessed a similar compound 5v showed weak inhibitory activities to KG1 and no
binding mode to SAHA in the active site of HDAC2. The inhibition to the MCF7 cell line. Among all tested compounds,
hydroxamic acid moieties of 5r and SAHA could both chelate to compounds 5r, 5w, and 5x possessed good anti-proliferative
the Zn2+ ion. Interestingly, the oxygen atom on the morpholine activities against all the three tumor cell lines. Therefore, these
moiety could form two hydrogen bonds with Arg275. The amide three compounds were chosen to be evaluated further for their
attaching purine could form another hydrogen bond with anti-tumor activities against other cancer cell lines including
Phe210, which further increased the interaction with HDAC. PC3, A549, U937, HeLa cells, and ES-2 cell lines with MTT assay
The above interactions could contribute to the higher potency (Table 3). Our results indicate that these three compounds
of 5r against HDAC compared to SAHA.
could signicantly inhibit the growth of all of these tumor cell
To further examine the activities of these HDAC inhibitors at lines. Compound 5x showed low IC50 values (less than 10 mmol
the cellular level, the compounds which showed obvious inhi- Lꢀ1) against all the tumor cell lines. This compound also has
bition in enzyme screening assays, were chosen to be evaluated similar anti-proliferative activities compared to the positive
for their anti-proliferative activities against the cancer cell lines. control. Further studies indicated that compounds 5r, 5w, and
Table 1 The structures and HDAC1 & 2 inhibitory activities of all compounds
Inhibition at
Inhibition at
R1
R2
X
5 mmol Lꢀ1
IC50a (mmol Lꢀ1
)
R1
R2
X
5 mmol Lꢀ1
IC50a (mmol Lꢀ1
)
5a –H
5b –H
5c –H
–NHCH2Ph –CH2–
–NHCH2Ph –(CH2)2–
–NHCH2Ph –(CH2)3–
3%
13%
11%
Nd
Nd
Nd
5m –H
5n –H
5o –H
–NHCH3 –CH2–
–NHCH3 –(CH2)2– 10%
–NHCH3 –(CH2)3– 6%
3%
Nd
Nd
Nd
5d –H
–NHCH2Ph –(CH2)5–
–(CH2)2–
70%
3%
2.05 ꢂ 0.03
5p –NHPh
5q –NHPh
–(CH2)2– 56%
–(CH2)3– 69%
–(CH2)5– 98%
–(CH2)4– 59%
–(CH2)5– 72%
–(CH2)6– 49%
–(CH2)4– 87%
–(CH2)5– 80%
>5
5e –H
Nd
>5
5f –H
–(CH2)3–
15%
49%
25%
20%
ꢀ1%
53%
Nd
5r
5s
5t
–NHPh
–H
0.075 ꢂ 0.014
2.51 ꢂ 0.28
2.09 ꢂ 0.24
2.94 ꢂ 0.06
0.59 ꢂ 0.03
0.30 ꢂ 0.06
5g –H
–(CH2)5–
>5
–NHPh
–NHPh
–NHPh
5h –H
Nd
–H
5i –NH2
5j –NH2
5k –NH2
–(CH2)2–
–(CH2)3–
–(CH2)5–
Nd
5u –H
5v
5w –NHPh –NHPh
Nd
–NHPh –NHPh
4.53 ꢂ 0.33
5l –NH2
0%
Nd
5x
–NHPh –NHPh
–(CH2)6– 95%
0.51 ꢂ 0.10
SAHA
88%
0.14 ꢂ 0.02
a
All of the compounds were assayed three times and expressed with standard deviations; Nd: not detected.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Fig. 4 (A) The docking modes of compound 5r (depicted in blue) and SAHA (depicted in green) in the active site of HDAC2; (B) proposed binding
preference of compound 5r; (C) proposed binding preference of SAHA. Molecular graphics and analyses were performed with the UCSF Chimera
package.27
Table 2 Anti-proliferative activities of representative compounds to Table 4 In vitro inhibitory activities to HDAC subtypes of compounds
cancer cell lines: MDA-MB231, MCF7, and KG1
5r, 5w, and 5x
Anti-proliferative activitiesa (mmol Lꢀ1
)
IC50 (mmol Lꢀ1
)
Compd
MDA-MB231
KG1
MCF7
Compd
HDAC1 & 2
HDAC1a
HDAC6a
HDAC8a
5d
5k
5r
34.18 ꢂ 3.16
41.32 ꢂ 7.81
1.56 ꢂ 0.28
>50
>50
Nd
>50
>50
5r
5w
0.075
0.30
0.51
0.14
0.035
0.028
0.063
0.051
0.0026
0.0038
0.0018
0.032
0.15
0.20
0.32
0.23
5.37 ꢂ 0.25
18.63 ꢂ 1.92 5x
5s
>50
>50
SAHA
5t
5u
5v
5w
5x
SAHA
22.29 ꢂ 6.13
8.91 ꢂ 0.99
11.07 ꢂ 0.35
1.61 ꢂ 0.13
1.55 ꢂ 0.35
1.11 ꢂ 0.43
>50
>50
34.74 ꢂ 1.05
3.80 ꢂ 0.25
2.88 ꢂ 0.71
1.11 ꢂ 0.16
>50
>50
>50
9.58 ꢂ 0.65
9.04 ꢂ 0.65
3.26 ꢂ 0.70
a
Values are the mean values of two determinations.
a
Each value is reproduced in three independent assays and expressed
with standard deviations; Nd: not detected.
Conclusions
Based on the classical pharmacophore structure model of
HDAC inhibitors, total of 24 new substituted purine
hydroxamic acid derivatives were designed, synthesized, and
evaluated as HDAC inhibitors. The preliminary enzyme assay
showed that most compounds can effectively inhibit HDAC1
and 2. The potent compounds, such as 5r, 5w, and 5x, showed
a
5x showed similar inhibition to HDAC1 and HDAC8, but
exhibited 8–18 fold enhanced activity against HDAC6 compared
with SAHA (Table 4).
Table 3 Anti-proliferative activities of compounds 5r, 5w, and 5x
Anti-proliferative activitiesa (mmol Lꢀ1
)
Compd
MDA-MB231
KG1
MCF7
PC3
A549
U937
HeLa
ES-2
5r
5w
5x
SAHA
1.56 ꢂ 0.28
1.61 ꢂ 0.13
1.55 ꢂ 0.35
1.11 ꢂ 0.43
5.37 ꢂ 0.25
3.80 ꢂ 0.25
2.88 ꢂ 0.71
1.11 ꢂ 0.16
18.63 ꢂ 1.92
9.58 ꢂ 0.65
9.04 ꢂ 0.65
3.26 ꢂ 0.70
6.44 ꢂ 1.37
2.14 ꢂ 0.52
4.26 ꢂ 1.12
1.42 ꢂ 0.17
46.11 ꢂ 3.76
18.00 ꢂ 2.73
10.12 ꢂ 1.87
4.97 ꢂ 0.54
6.50 ꢂ 1.93
4.07 ꢂ 0.97
3.91 ꢂ 0.59
4.33 ꢂ 2.06
18.23 ꢂ 2.63
9.08 ꢂ 0.05
8.76 ꢂ 1.16
6.07 ꢂ 2.25
28.02 ꢂ 0.22
13.44 ꢂ 2.17
8.95 ꢂ 1.21
4.66 ꢂ 0.61
a
Each value was reproduced in three independent assays and expressed with standard deviations.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
obvious anti-proliferative activities in eight cancer cell lines. In 11 J. Jiao, H. Fang, X. Wang, P. Guan, Y. Yuan and W. Xu, Eur. J.
addition, the three compounds also had high affinity and good Med. Chem., 2009, 44, 4470–4476.
selectivity to HDAC6 compared with SAHA. The results sug- 12 Y. Zhang, J. Feng, C. Liu, H. Fang and W. Xu, Bioorg. Med.
gested that 2,6-disubstituted purine hydroxamate derivatives Chem., 2011, 19, 4437–4444.
could be used as lead compounds to develop novel anti-cancer 13 P. Guan, F. Sun, X. Hou, F. Wang, F. Yi, W. Xu and H. Fang,
agents.
Bioorg. Med. Chem., 2012, 20, 3865–3872.
14 L. Han, L. Wang, X. Hou, H. Fu, W. Song, W. Tang and
H. Fang, Bioorg. Med. Chem., 2014, 22, 1529–1538.
15 C. Qian, X. Cai, S. Gould and H. Zhai, WO2008115262, 2008.
16 I. Paquin, S. Raeppel, S. Leit, F. Gaudette, N. Zhou,
O. Moradei, O. Saavedra, N. Bernstein, F. Raeppel,
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (no. 81373281 and 21172133), the Shan-
dong Natural Science Fund for Distinguished Young Scholars
(no. JQ201319), the Program for New Century Excellent Talents
in University (no. NCET-12-0337), the Independent Innovation
Foundation of Shandong University (IIFSDU no. 2012JC002),
and the China–Australia Centre for Health Sciences Research
(CACHSR no. 2014GJ02).
´
G. Bouchain, S. Frechette, S. H. Woo, A. Vaisburg,
M. Fournel, A. Kalita, M.-F. Robert, A. Lu, M.-C. Trachy-
Bourget, P. T. Yan, J. Liu, J. Rahil, A. R. MacLeod,
J. M. Besterman, Z. Li and D. Delorme, Bioorg. Med. Chem.
Lett., 2008, 18, 1067–1071.
17 K. S. Ko, M. E. Steffey, K. R. Brandvold and M. B. Soellner,
ACS Med. Chem. Lett., 2013, 4, 779–783.
18 J. Wang, Q. Wang, L. Zhang and H. Fang, Chin. J. Chem.,
2013, 31, 1181–1191.
19 L.-K. Huang, Y.-C. Cherng, Y.-R. Cheng, J.-P. Jang, Y.-L. Chao
and Y.-J. Cherng, Tetrahedron, 2007, 63, 5323–5327.
20 H.-F. Lu, L.-Z. Zhang, D.-M. Wu and J.-T. Zhou, J. Heterocycl.
Chem., 2011, 48, 1140–1144.
21 C. Qian, X. Cai and H. Zhai, WO2009036016, 2009.
22 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner,
R. K. Belew, D. S. Goodsell and A. J. Olson, J. Comput.
Chem., 2009, 30, 2785–2791.
References
1 M. Ahmad, A. Hamid, A. Hussain, R. Majeed, Y. Qurishi,
J. A. Bhat, R. A. Najar, A. K. Qazi, M. A. Zargar, S. K. Singh
and A. K. Saxena, DNA Cell Biol., 2012, 31(suppl 1), S62–S71.
2 B. E. Gryder, Q. H. Sodji and A. K. Oyelere, Future Med. Chem.,
2012, 4, 505–524.
3 J. Li, G. Li and W. Xu, Curr. Med. Chem., 2013, 20, 1858–1886.
4 G. Giannini, W. Cabri, C. Fattorusso and M. Rodriquez,
Future Med. Chem., 2012, 4, 1439–1460.
5 S. C. Mwakwari, V. Patil, W. Guerrant and A. K. Oyelere, Curr.
Top. Med. Chem., 2010, 10, 1423.
23 X. Hou, J. Du, J. Zhang, L. Du, H. Fang and M. Li, J. Chem. Inf.
Model., 2013, 53, 188–200.
6 T. A. Miller, D. J. Witter and S. Belvedere, J. Med. Chem., 2003,
46, 5097–5116.
24 D. F. Wang, P. Helquist, N. L. Wiech and O. Wiest, J. Med.
Chem., 2005, 48, 6936–6947.
7 Y. Zhang, J. Feng, Y. Jia, X. Wang, L. Zhang, C. Liu, H. Fang
and W. Xu, J. Med. Chem., 2011, 54, 2823–2838.
8 Y. Zhang, H. Fang, J. Feng, Y. Jia, X. Wang and W. Xu, J. Med.
Chem., 2011, 54, 5532–5539.
9 Y. Zhang, J. Feng, C. Liu, L. Zhang, J. Jiao, H. Fang, L. Su,
X. Zhang, J. Zhang, M. Li, B. Wang and W. Xu, Bioorg. Med.
Chem., 2010, 18, 1761–1772.
25 U. S. Tambunan, N. Bramantya and A. A. Parikesit, BMC
Bioinf., 2011, 12(suppl 13), S23.
26 D. Wang, Curr. Top. Med. Chem., 2009, 9, 241–256.
27 E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch,
D. M. Greenblatt, E. C. Meng and T. E. Ferrin, J. Comput.
Chem., 2004, 25, 1605–1612.
10 J. Feng, H. Fang, X. Wang, Y. Jia, L. Zhang, J. Jiao, J. Zhang,
L. Gu and W. Xu, Cancer Biol. Ther., 2011, 11, 477–489.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.