Novel histone deacetylase inhibitors: N-hydroxycarboxamides possessing a terminal bicyclic aryl group

Article abstract of DOI:10.1016/S0960-894X(02)00175-0

Utilizing tranexamic acid as a starting material, a series of N-hydroxycarboxamides were synthesized in order to seek new histone deacetylase (HDAC) inhibitors. Further structure optimization involving the replacement of the 1,4-cyclohexylene group with t

Full text of DOI:10.1016/S0960-894X(02)00175-0

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1347–1349
Novel Histone Deacetylase Inhibitors: N-Hydroxycarboxamides
Possessing a Terminal Bicyclic Aryl Group
Shinichi Uesato,a,* Manabu Kitagawa,a Yasuo Nagaoka,a Taishi Maeda,a
Hiroshi Kuwajimab and Takao Yamoric
aDepartment of Biotechnology, Faculty of Engineering, Kansai University, Suita, Osaka 564-8680, Japan
bFaculty of Pharmaceutical Sciences, Kinki University, 3-4-1 Kowakae, Higashiosaka, Osaka 577-8502, Japan
cDivision of Experimental Chemotherapy, Cancer Chemotherapy Center, Japanese Foundation for Cancer Research,
Kamiikebukuro 1-37-1, Toshima-ku, Tokyo 170-8455, Japan
Received 12 February 2002; accepted 16 March 2002
Abstract—Utilizing tranexamic acid as a starting material, a series of N-hydroxycarboxamides were synthesized in order to seek
new histone deacetylase (HDAC) inhibitors. Further structure optimization involving the replacement of the 1,4-cyclohexylene
group with the 1,4-phenylene group yielded the promising HDAC inhibitors which possess a terminal bicyclic aryl amide. # 2002
Elsevier Science Ltd. All rights reserved.
Histone deacetylase (HDAC) inhibitors modulate the
deacetylation of "-N-acetyl groups of l-lysine residues in
the N-terminal tails of core histones.1 As HDAC inhibi-
tors from natural resources, trichostatin A2 (TSA), tra-
poxin3 and FK228 (1)4 were isolated; they inhibited
strongly HDACs and induce differentiation of cancer
cell lines and suppress cell proliferation, respectively.
The former two compounds, however, have not yet been
applied for medical treatments owing to their instability
and toxicity. Afterwards, synthetic inhibitors: MS-275
(2),5 SAHA (3)6 and oxamflatin (4)7 were reported as
cancer therapeutic candidates. Compounds 1, 2 and 3
are now under clinical investigation. In our exploratory
studies of new HDAC inhibitors, a series of N-hydr-
oxycarboxamides were synthesized utilizing tranexamic
acid (TA) as a lead substance since it is well known to
be the l-lysine analogue inhibiting fibrinolysis by
blockage of l-lysine binding site at plasminogen mole-
cule. Structure optimization by replacing the trans-1,4-
cyclohexylene group of TA with 1,4-phenylene group
and further elaboration yielded promising compounds
which represent significant inhibitory activities against
cancer cell proliferation and HDACs.
through Route A, B or C as outlined in Scheme 1.
Route A: TA methyl ester hydrochloride (5) was treated
with aryl acid chloride in the presence of triethylamine
(TEA) or with aryl acid in the presence of Bis(2-oxo-3-
oxazolidinyl)phosphinic chloride (BOP-Cl) and TEA,
yielding aryl amide (7). This compound, after alkaline
hydrolysis to acid, was condensed with benzyloxyamine
hydrochloride in the presence of BOP-Cl and TEA fol-
lowed by catalytic hydrogenation of the product to
afford target N-hydroxycarboxamide (9). Route B:
Compound 5 was reacted with aryl sulfonyl chloride in
the presence of TEA to afford sulfonamide (11), which
was in turn treated in the same way as for Route A to
N-hydroxycarboxamides (9a–9d, 13a–13b and 17) hav-
ing the trans-1,4-cyclohexylene group were synthesized
furnish N-hydroxycarboxamide having a terminal
phenyl sulfonamide (13). Route C: Compound 17 was
*Corresponding author. Tel.: +81-6-6368-0834; fax: +81-6388-8609;
e-mail: uesato@ipcku.kansai-u.ac.jp
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00175-0

1348
S. Uesato et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1347–1349
Scheme 1. Conditions: (i) ArCOCl, TEA/THF or CH2Cl2; (ii) ArCOOH, BOP-Cl, TEA/CH2Cl2 or DMF; (iii) 1 M LiOH/H2O–THF; (iv) H2N-
OBn, BOP-Cl, TEA/CH2Cl2 or DMF; (v) H2/10%Pd-C/MeOH; (vi) ArSO2Cl, TEA/THF; (vii) 1-adamantylamine, triphosgene, TEA/CH2Cl2.
prepared from condensation of 5 and 1-adamantyl-
amine using triphosgene and TEA following the afore-
mentioned conventional steps. Synthetic routes for
N-hydroxycarboxamides (10a–10f, 14a–14b and 18)
possessing the 1,4-phenylene group were also shown in
Scheme 1. Thus, compound 10 was prepared through
Route A starting from condensation of 4-amino-
methylbenzoic acid methyl ester hydrochloride (6) with
aryl acid chloride or with aryl acid to give aryl amide
(8). However, compounds 10e and 10f, having the
amino group, were obtained by treating the corre-
sponding Boc-protected compounds (formed through
Route A) further with 4M HCl/MeOH, respectively. In
the same way, compound 14 was prepared via 12
according to the Route B, whereas compound 18 was
formed via 16 according to the Route C.
were found to be over 13-fold more potent than the
respective regioisomers 9c, 10b and 10d, respectively.
These results suggested that topology as well as bulki-
ness or p electron number in the terminal bicyclic aryl
moieties would affect the anfiproliferative activities.
Replacement of the 2-naphthyl amide in 10a with the
2-naphthyl sulfonamide in 14b reduced the activity.
Furthermore, compounds 17 and 18 having the 1-ada-
mantylureido group showed only weak activities.
Second, the compounds showing significant anti-
prolierative activities (IC50’s ranging between 0.7 and
9.7 mM) against HCT 116 were selected and subjected to
the HDAC inhibition tests (Table 1). In the experi-
ments, the IC50 values were measured according to the
method2 by Yoshida et al., but using partially purified
HDACs from human T cell leukemia Jurkat cell10 in
place of mouse mammary tumor cell line FM3A.
Attention has been denoted to the 1,4-phenylene-series
compounds 10a and 10e, which exhibited the 50%
HDAC inhibition at the concentrations of 44 and
39 nM, respectively. These values were almost equal to
that of oxamflatin (4) (data not shown). It is likely as
shown in Figure 1 that 10a is incorporated into the
active-site pocket of the HDAC wherein the hydroxamic
acid chelates the zinc ion positioned at the bottom. We
speculated that their 1,4-phenylene group could contact
to aromatic and hydrophobic amino acid residues at the
wall of the pocket, whereas the terminal 2-naphthyl
moiety would not only pack the molecule as a cap, but
also interact with aryl groups of amino acid residues
around the rim.
First of all, the 1,4-cyclohexylene-series compounds
(9a–9d) were estimated for antiproliferative activities
against HCT 116 colorectal human carcinoma cells
using WST-1 assays:8 compounds 9a, 9b and 9d repre-
sented the activities (7.6, 6.6 and 4.7 mM) of almost
comparable magnitude in the IC50 values with that of
MS-275 (5.4 mM),9 respectively. Of the two sulfona-
mides, 13b comprising the 2-naphthyl was 8-fold more
active (IC50 9.4 mM) than 13a possessing the phenyl
though the former was less potent than 9b. As for the
1,4-phenylene-series compounds (10a–10f), compounds
10a, 10c and 10e, all having the terminal bicylic aryl (2-
naphthyl, 1,4-biphenyl and 6-amino-2-naphthyl groups)
had higher inhibitory activities (IC50 0.7, 3.2and
3.9 mM) relative to the 1,4-cyclohexylene-series com-
pounds, respectively. However, the sulfonamide 14b
possessing the 2-naphthyl had only weak activity
(22.5 mM). Interestingly, compounds 9b, 10a and 10c
Compounds 9a, 9b, 9d, 10a, 10c, 10e and 17 were eval-
uated for antitumor activities using a human cancer cell

S. Uesato et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1347–1349
1349
Table 1. Inhibition of HCT 116 proliferation and of HDACsa
represented the correlation coefficients (r) less than 0.75
except for 17, respectively. Especially, 10a and 10c
showed the values less than 0.5, thus leading us to
expect that they would have a unique action mode.
Compd
Ar
WST-1 IC50
(mM)b
HDACs IC50
(nM)b
9a
7.6
1500
2000
We have a plan to examine compounds 9a, 9b, 9d, 10a,
10c and 10e for antitumor activities against cancer
xenografts in nude mice.
9b
9c
9d
10a
10b
10c
10d
2-Naph
1-Naph
1,4-Biph
2-Naph
1-Naph
1,4-Biph
1,2-Biph
6.6
90.1
4.7
0.7
9.7
310
44
830
Acknowledgements
40
3.22
87.6
The authors thank Mr. H. Mori and Ms. T. Yamada for
the HDAC inhibitory assay done at Medicinal Biology
Research Laboratories of Fujisawa Pharamaceutical
Co., Ltd.
10e
10f
3.9
39
87.2
13a
13b
14a
14b
17
Ph
77.9
9.4
References and Notes
2-Naph
Ph
2-Naph
1100
38.8
22.5
20.9
112.5
—
1. (a) Grunstein, M. Nature (London) 1997, 389, 349. (b)
Struhl, K. Genes Dev. 1998, 12, 599.
2. Yoshida, M.; Kijima, M.; Akita, M.; Beppu, T. J. Biol.
Chem. 1990, 265, 17174.
18
TSA
3.3
3. Kijima, M.; Yoshida, M.; Sugita, K.; Horinouchi, S.;
Beppu, T. J. Biol. Chem. 1993, 268, 22429.
4. Nakajima, H.; Kim, Y. B.; Terano, H.; Yoshida, M.; Hor-
inouchi, S. Exp. Cell Res. 1998, 241, 126.
aValues are means of three experiments.
bConcentration required to inhibit by 50%.
5. Saito, A.; Yamashita, T.; Mariko, Y.; Nosaka, Y.; Tsu-
chiya, K.; Ando, Y.; Suzuki, T.; Tsuruo, T.; Nakanishi, O.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4592.
6. Bulter, L. M.; Agus, D. B.; Scher, H. L.; Higgins, B.; Rose,
A.; Cordon-Cardo, C.; Thaler, H. T.; Rifkind, R. A.; Marks,
P.; Richon, V. M. Cancer Res. 2000, 60, 5165.
7. Kim, Y. B.; Lee, K. H.; Sugita, K.; Yoshida, M.; Hor-
inouchi, S. Oncogene 1999, 18, 2461.
8. Takenouchi, T.; Munekata, E. Life Sciences 1995, 56, 479.
9. This compound was synthesized in our laboratory for the
HCT 116 test.
10. Mori, H.; Sakamoto, K.; Tsurumi, Y.; Takase, S.; Hino,
M. WO 00/21979, 2000.
11. Yamori, T.; Matsunaga, A.; Sato, S.; Yamazaki, K.;
Komi, A.; Ishizu, K.; Mita, I.; Edatsugi, H.; Matsuda, Y.;
Takezawa, K.; Nakanishi, O.; Kohono, H.; Nakajima, Y.;
Komatsu, H.; Andoh, T.; Turuo, T. Cancer Res. 1999, 59,
4042COMPARE analysis is used to evaluate the statistical
correlation and estimate the action mode for a test compound
through comparison of its differential growth inhibition pat-
tern (finger print) with those of 200 standard anticancer drugs
in the database. The correlation coefficient (r) less than 0.5
suggests a new action mode.
Figure 1. Proposed interaction mode of 10a and active-site of HDAC.
line panel with database analysis.11 The respective aver-
age concentrations of these compounds required for
50% growth inhibition (GI50) against a panel of 39 cell
lines were as follows: 20.4, 6.5, 3.6, 0.81, 4.1, 2.7 and
20.4 mM. Furthermore, the COMPARE analysis11