Toward an HDAC6 inhibitor: Synthesis and conformational analysis of cyclic hexapeptide hydroxamic acid designed from α-tubulin sequence

Article abstract of DOI:10.1016/j.bmc.2004.01.014

A cyclic hexapeptide hydroxamic acid inhibitor for HDAC6 has been designed and synthesized on the basis of the facts that α-tubulin is the substrate of HDAC6 and of the excellent inhibitory activity of cyclic tetrapeptide hydroxamic acids (CHAPs) for HDACs. Unexpectedly, cyclic hexapeptide hydroxamic acid showed very low HDAC inhibitory activity. To explain the low activity, we have carried out conformation analysis and compared it to the crystal structure of α-tubulin. The conformation around the acetylated lysine of the cyclic hexapeptide substrate or the aminosuberate hydroxamic acid [Asu(NHOH)] of cyclic hexapeptide inhibitor is different from that around α-tubulin's lysine-40. The difference in the conformation seems to cause some steric hindrance at the capping site resulting in poor binding capacity.

Full text of DOI:10.1016/j.bmc.2004.01.014

Bioorganic & Medicinal Chemistry 12(2004) 1351–1356
Toward an HDAC6 inhibitor: synthesis and conformational
analysis of cyclic hexapeptide hydroxamic acid designed from
ꢀ-tubulin sequence
Binoy Jose,a Shinji Okamura,b Tamaki Kato,b Norikazu Nishino,a,b,* Yuko Sumidaa
and Minoru Yoshidaa,c
aCREST Research Project, Japan Science and Technology Agency, Saitama 332-001, Japan
bGraduate School of Life Science and Systems Engineering, Kyushu Institute of Technology, Kitakyushu, 808-0196, Japan
cRIKEN, Saitama 351-0198, Japan
Received 26 November 2003; revised 13 January 2004; accepted 14 January 2004
Abstract—A cyclic hexapeptide hydroxamic acid inhibitor for HDAC6 has been designed and synthesized on the basis of the facts
that a-tubulin is the substrate of HDAC6 and of the excellent inhibitory activity of cyclic tetrapeptide hydroxamic acids (CHAPs)
for HDACs. Unexpectedly, cyclic hexapeptide hydroxamic acid showed very low HDAC inhibitory activity. To explain the low
activity, we have carried out conformation analysis and compared it to the crystal structure of a-tubulin. The conformation around
the acetylated lysine of the cyclic hexapeptide substrate or the aminosuberate hydroxamic acid [Asu(NHOH)] of cyclic hexapeptide
inhibitor is different from that around a-tubulin’s lysine-40. The difference in the conformation seems to cause some steric hin-
drance at the capping site resulting in poor binding capacity.
# 2004 Elsevier Ltd. All rights reserved.
1. Introduction
apicidin,9ꢀ11 FK228,12ꢀ14 synthetic compounds such as
butyrate,15 valproate,16 suberoyl anilide hydroxamic
acid,17 analogues of TSA18ꢀ20 and benzamide deriva-
tives of MS-275.21,22 We have reported synthetic cyclic
tetrapeptide inhibitors (CHAPs) having inhibitory
activity in nanomolar range.23,24
The reversible acetylation of histone has a critical role in
transcriptional regulation and the reversible acetylation
of non-histone substrates is also important for other
cellular processes.1ꢀ3 Acetylation and deacetylation of
histones are catalyzed by specific enzyme families,
histone acetyl transferases (HATs) and deacetylases
(HDACs) respectively. HDACs are integral nuclear
isozymes that modulate the deacetylation of specific
acetylated lysine residues. An increasing number of
HDACs are being identified in different species, which is
classified into three distinct classes.4,5 Class I HDACs
are related to yeast Rpd3 and include HDAC1,
HDAC2, HDAC3 and HDAC8. Class II HDACs are
related to yeast Hda1 and include HDAC4, HDAC5,
HDAC6 and HDAC7. Class III HDACs are NAD
dependant and related to yeast silencing protein Sir2. A
number of HDAC inhibitors have been identified which
include natural product trichostatin A (TSA),6 naturally
occurring cyclic tetrapeptides trapoxin (TPX),7 Cyl-1,8
Inhibitors of HDACs possess a Zn binding functionality
and a cap substructure that interacts with the rim of N-
acetyl lysine binding channel.25 Of the reported HDAC
inhibitors, TSA can inhibit HDAC1 and HDAC6 in
almost equal intensity. Synthetic inhibitors with cyclic
tetrapeptide cap and hydroxamic acid functional group
(CHAP) and the natural compound TPX are less potent
toward HDAC6 in comparison with HDAC1. This dif-
ference in the inhibitory activity may be attributed to
the variation in the cap group, which is due to the dif-
ferent extent of interaction with the rim of active site of
HDACs. Unlike other HDACs, HDAC6 have two cat-
alytic domains and both the catalytic domains con-
tribute independently to the overall activity.26 Further,
HDAC6 has many deviations from the other HDACs in
the rim region of the catalytic domain. Recent reports
showed that HDAC6 is a tubulin deacetylase.27,28 This
reversible acetylation of a-tubulin is associated with
Keywords: Histone deacetylase; Tubulin; Cyclic peptide; Inhibitor.
* Corresponding author. Tel.: +81-9369-56061; fax: +81-9369-56061;
e-mail: nishino@life.kyutech.ac.jp
0968-0896/$ - see front matter # 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.01.014

1352
B. Jose et al. / Bioorg. Med. Chem. 12 (2004) 1351–1356
microtubule stability. a-tubulin is modified by the acetyl-
ation of a lysine residue at position 40.29,30 Schreiber et
al. synthesized a hydroxamic acid inhibitor, which has
some selectivity for HDAC6 over HDAC1 using the
combinatorial approach.31,32 We attempted to design a
new inhibitor for HDAC6 based on the fact that a-
tubulin is the target of HDAC6. The proposed inhibitor
(Fig. 1) has a hydroxamic acid functional group, and
the same chain length as in CHAP1. The cap group is
cyclic hexapeptide derived from a-tubulin’s amino acid
residues of position 38 to 43, the lys-40 being replaced
by aminosuberic hydroxamic acid (Asu(NHOH)). Here
we describe the synthesis, activity and conformation
studies on the cyclic hexapeptide.
protected side chains. The linear peptide was cyclized in
DMF using HATU and the side chain protections were
removed by catalytic hydrogenation. The cyclic hexa-
peptide 3 was purified by HPLC and characterized by
NMR and HR-FABMS.
Next we synthesized cyclic hexapeptide inhibitor cyclo-
(-Ser-Asp-Asu(NHOH)-Thr-Ile-Gly-) (4). A protected
linear hexapeptide was synthesized using Barlos resin
preloaded with Fmoc-glycine on a synthesizer by Fmoc
strategy. Fmoc amino acids used were, Ile, Thr(tBu),
Asu(OBzl), Asp(OtBu) and Ser(tBu), where Asu is ami-
nosuberic acid. The linear hexapeptide was cleaved from
resin using acetic acid and cyclized in DMF using
HATU as coupling agent and purified by column chro-
matography to yield the cyclic hexapeptide. The solubi-
lity problem of the linear hexapeptide was overcome by
using one mM solution in DMF. The side chain car-
boxyl group of Asu was deprotected by catalytic
hydrogenation and then coupled with hydroxylamine.
Other side chain protections of cyclic peptide were
removed by treatment with TFA and the compound 4
was purified by gel filtration (Sephadex LH-20, DMF).
The cyclic hexapeptide was further purified by HPLC
and characterized by NMR and HR-FABMS.
2. Results and discussion
2.1. Chemistry
From the crystal structure data of a-tubulin, it was
found that the b-turn around the residues from 38 to 42
is type II0 b-turn.30 To verify the conformation of the
active sequence of cyclic hexapeptide to a-tubulin
sequence, we initially synthesized a substrate, cyclo(-Ser-
Asp-Lys(Ac)-Thr-Ile-Gly-) (3). This compound was
prepared using solid phase peptide synthesis method
with Kaiser’s oxime resin by Boc strategy with benzyl
The HDAC activity assay of compound 4 was carried
out and the activity is in millimolar order (Table 1). The
Figure 1. Reported and proposed HDAC inhibitors and substrate.

B. Jose et al. / Bioorg. Med. Chem. 12 (2004) 1351–1356
1353
Table 1. HDAC inhibitory activity
Table 3. Structural comparison of a-tubulin sequence with com-
pounds 3 and 4
Compd
HDAC1a (nM)
HDAC6 (nM)
Residue
Natural sequence
in a-tubulin
3
4
TSA
CHAP1
4
22
20
28
190
>100,000
>100,000
f
c
f
c
f
c
a HDACs prepared from 293T cells.
Ser
ꢀ85.9
148.5
58.3
ꢀ172.9
163.0
ꢀ133.4
ꢀ9.0
ꢀ90.9
ꢀ45.3 ꢀ90.6
ꢀ72.5
Asp
Lys/Asu
Thr
Ile
Gly
ꢀ89.0 ꢀ138
ꢀ90.5 ꢀ123
ꢀ89.4
ꢀ90.6
ꢀ91.0
164
36.5 ꢀ89.8
53.2
ꢀ73.0
ꢀ49.8 ꢀ91.1
ꢀ49.2 ꢀ89.2
ꢀ47.2
ꢀ66.7
ꢀ30
Table 2. NMR data for compounds 3 and 4
88.1
ꢀ73.0
ꢀ4.8
a
Residue
d (ppm at RT)
JNHꢀC
(Hz)a
ꢀd/ꢀT (ꢁ10ꢀ3
)
ꢀ45.0
ꢀ27
172
H
3
4
3
4
3
4
4 are different, the H-bonding between the residues in
both compounds are found as the same. For compound
3, medium NOEs are observed between ThrNH and
LysNH, LysNH and AspNH, and Ser NH and GlyNH.
Low temperature coefficients of the IleNH protons and
ThrNH protons suggest that the compound 3 adopts a
type I b-turn. For compound 4, the NOEs are between
ThrNH and AspNH, ThrNH and IleNH and ThrNH
and SerNH. Low temperature coefficient values of
IleNH protons and ThrNH protons suggest that the
compound 4 also adopts a type I b-turn.
Ser
8.28
8.22
7.97
7.38
7.59
8.87
8.27
8.21
7.96
7.37
7.60
8.87
7.94
7.93
7.93
7.327.32
—
—
7.93
7.93
7.93
ꢀ3.86
ꢀ2.36
ꢀ3.25
ꢀ1.14
ꢀ3.80
ꢀ2.70
ꢀ3.80
ꢀ1.01
Asp
Lys/Asu
Thr
Ile
Gly
—
—
ꢀ0.232 ꢀ0.463
ꢀ5.45
ꢀ5.96
a IleNH and GlyNH protons were observed as singlets at 300 K.
unexpected low activity of the inhibitor 4 forced us to
investigate the reason and therefore we carried out
molecular modeling studies of the substrate 3 and inhi-
bitor 4 and compared it with the X-ray crystal structure
of a-tubulin.30 HDAC6 have a unique structure with the
internally duplicated catalytic domains. However, it is
still unclear that both the domains are involved in the a-
tubulin deacetylation. Therefore we speculate that the
comparison of conformation of inhibitor 4 with that
of a-tubulin give a better explanation for the resistance
of 4.
The structures of cyclic hexapeptides 3 and 4 having
minimum energy configuration were generated using
CHARMM program of Insight II. The values of f from
NMR results were applied to calculate the c values of 3
and 4. These values and the values for a-tubulin around
Lys-40 are given in Table 3. The dihedral angle values of
compounds 3 and 4 confirmed that these compounds
have type I b-turn. Molecular modeling studies showed
that the conformation around the Lys residue of a-
tubulin is different from that of the synthesized cyclic
hexapeptides 3 and 4 as shown in Figure 2. The car-
bonyl of Lys in a-tubulin and that in 3 or carbonyl of
Asu in 4 is lying in the opposite plane. This might be the
reason for the low inhibitory activity of 4, eventhough,
it contains a hydroxamic acid functional group and the
optimum spacer length of five methylene units.
Conformation analyses of the cyclic hexapeptide sub-
1
strate 3 and inhibitor 4 were studied using H NMR in
DMSO-d6. Complete assignments of the chemical shifts
were made using COSY, HOHAHA and NOESY spec-
tra. Table 2 shows the chemical shift values for NH
Protons and the temperature coefficients of compound 3
a
and 4. JNHꢀC
coupling constants were used to calcu-
H
late the f angles. The NMR data of compounds 3 and 4
are very similar and suggest that the backbone
conformations of these two compounds are not con-
siderably different. Although the NOE patterns of 3 and
In conclusion, we synthesized a cyclic hexapeptide inhi-
bitor for HDAC6 based on the fact that HDAC6 is
tubulin deacetylase. Biological assay showed that the
cyclic hexapeptide hydroxamic acid 4 is a very weak
Figure 2. (a) X-ray structure of a-tubulin fragment (-Ser- Asp -Lys40- Thr- Ile-Gly-); (b) Calculated conformation of compound 3. (c) Calculated
conformation of compound 4.

1354
B. Jose et al. / Bioorg. Med. Chem. 12 (2004) 1351–1356
inhibitor for HDAC. Molecular modeling studies
revealed that the configuration around the Lys-40 of
a-tubulin and the cyclic hexapeptide 4 were different
and therefore the interaction of the cyclic hexapeptide
scaffold to the rim of the binding pocket of the enzyme
is limited. Further modification according to this result
can generate specific inhibitor for HDAC6.
Asp(OBzl)-Lys(Ac)-Thr(Bzl)-Ile-Gly-OH. It was filtered
from water and dried over P2O5 (640 mg, 0.623 mmol,
110%). The hexapeptide Boc-Ser(Bzl)-Asp(OBzl)-
Lys(Ac)-Thr(Bzl)-Ile-Gly-OH (640 mg, 0.623 mmol)
was deprotected using TFA (10 mL) in an ice bath for
30 min. TFA was evaporated and the peptide was soli-
dified as TFA salt by adding ether and filtered from
ether (650 mg, 0.623 mmol, 100%). The linear hex-
apeptide trifluoroacetate (600 mg, 0.576 mmol) was
cyclized in DMF (125 mL) under a high dilution condi-
tion. For this, one-fifth of the hexapeptide, HATU
(total 328 mg) and diisopropylethylamine (total 0.4 mL)
were added portionwise in 20 min intervals at room
temperature with stirring. The reaction was monitored
by HPLC. DMF was removed in vacuo and the residue
was dissolved in ethyl acetate and washed with 10%
citric acid, 4% sodium bicarbonate and brine respec-
tively. Ethyl acetate was evaporated and the cyclic hex-
apeptide was purified using silica gel column
chromatography using 10% methanol in chloroform as
eluent. (287 mg, 0.316 mmol, 51%). FAB MS gave
peaks of (M+H)+ and (M+Na)+ at m/z 914 and 936
respectively. The cyclic hexapeptide was dissolved in
acetic acid (5 mL) and 5% Pd/C (100 mg) was added
and hydrogenated at rt under ordinary pressure. Acetic
acid was removed in vacuo and the cyclic hexapeptide
was lyophilized from water. The cyclic hexapeptide thus
obtained was further purified by RP-HPLC using 5-
30% CH3CN gradient containing 0.1%TFA over 30
min. HPLC: 2.9 min (chromolith, 10–100% CH3CN
gradient containing 0.1%TFA over 15 min), FABMS
gave peaks of (M+H)+ and (M+Na)+ at m/z 644 and
666 respectively. HR-FABMS (M+H)+ 644.3216 for
C27H46O11N7 (calcd 634.3255).
3. Experimental
3.1. General methods
All compounds were routinely checked by thin layer
chromatography (TLC) or high-performance liquid
chromatography (HPLC). Analytical HPLC were per-
formed on a Hitachi instrument equipped with a chro-
molith performance RP-18e column (4.6ꢁ100 mm,
Merck). The mobile phases used were A: H2O with 10%
CH3CN and 0.1% TFA, B: CH3CN with 0.1% TFA
using a solvent gradient of A to B over 15 min with
detection at 220 nm with a flow rate of 2 mL/min. FAB-
Mass spectra and high-resolution mass spectra (HR MS)
were measured on a JEOL JMS-SX 102A instrument.
Amino acids were coupled with 2-(1H-benzotriazol-1-
yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate
(HBTU), or O-(7-azabenzotriazoyl-1-yl)-1,1,3,3-tetra-
methyluronium hexafluorophosphate (HATU) as the
coupling reagents.
3.2. Synthesis of cyclo(-Ser-Asp-Lys(Ac)-Thr-Ile-Gly-)
(3)
Synthesis was performed on Kaiser’s 4-nitrobenzophe-
none oxime resin starting from glycine by the applica-
tion of standard Boc methodology. Boc-Gly-OxR (2g)
was deprotected with 25% TFA in CH2Cl2 (1ꢁ1 min,
1ꢁ30 min), washed with CH2Cl2 (2ꢁ), i-PrOH, CH2Cl2
(3ꢁ) and DMF and the deprotection was confirmed by
Kaiser’s test. A solution of Boc-Ile-OH (404 mg, 1.68
mmol), HBTU (636 mg, 1.68 mmol) HOBt (258 mg,
1.68 mmol) and DIEA (0.5 mL, 2.8 mmol) in DMF (30
mL) were added to the resin and shaking was main-
tained for 30 min. After this period the resin was
washed with DMF (3ꢁ), CH2Cl2 (3ꢁ) and the com-
pleteness of coupling was confirmed by Kaiser’s test.
Deprotection and coupling were repeated as described
above for the amino acids Boc-Thr(Bzl)-OH, Boc-
Lys(Ac)-OH, Boc-Asp(OBzl)-OH and Boc-Ser(Bzl)-
OH, respectively. The protected hexapeptide, Boc-
Ser(Bzl)-Asp(OBzl)-Lys(Ac)-Thr(Bzl)-Ile-Gly-OH was
obtained from Boc-Ser(Bzl)-Asp(OBzl)-Lys(Ac)-Thr(Bzl)-
Ile-Gly-OxR by shaking with N-hydroxypiperidine
(HOPip) (226 g, 2.24 mmol) in DMF (30 mL) for 30 h.
The DMF solution was collected and the resin was
washed with DMF (30 mLꢁ3) and collected. DMF (120
mL) was evaporated in vacuo and the Boc-Ser(Bzl)-
Asp(OBzl)-Lys(Ac)-Thr(Bzl)-Ile-Gly-OPip was dis-
solved in acetic acid (10 mL) and sodium dithionate
(487 mg, 2.8 mmol) was added. The solution was stirred
for 1 h at rt and the completion of the reaction was
monitored by HPLC. After removal of acetic acid,
water was added to precipitate the Boc-Ser(Bzl)-
3.3. Synthesis of cyclo(-Ser-Asp-Asu(NHOH)-Thr-Ile-
Gly-) (4)
The linear hexapeptide, Ser(tBu)-Asp(OtBu)-Asu(OBzl)-
Thr(tBu)-Ile-Gly was synthesized using peptide synthe-
sizer (PE ABI 433A) by Fmoc strategy using Balros
resin (0.25 mmol) and cleavage from the resin using
acetic acid in a mixture of dichloromethane and tri-
fluoroethanol (1:8:1) at rt yield 230 mg (100%,
0.25mmol) of protected linear peptide. The linear pep-
tide (230 mg) was dissolved in DMF (250 mL) by soni-
cation and stirring. HATU (144 mg, 0.38 mmol) and
DIEA (0.17 mL, 0.1 mmol) were added to the solution
and stirred for 3 h at rt. The reaction was monitored by
HPLC. Solvent was removed in vacuo and the residue
was dissolved in ethyl acetate and washed with 10%
citric acid, 4% sodium bicarbonate and brine respec-
tively. Ethyl acetate was evaporated to give cyclic hex-
apeptide cyclo(-Ser(tBu)-Asp(OtBu)-Asu(OBzl)-Thr(tBu)-
Ile-Gly-) (113 mg, 0.125 mmol, 50%). HPLC: 11.8 min
(chromolith, 10–100% CH3CN gradient containing
0.1%TFA over 15 min), FABMS gave peak of
(M+H)+ and (M+Na)+ at m/z 903 and 925 respec-
tively. The protected cyclic hexapeptide (113 mg,
0.125mmol) was dissolved in methanol (5 mL) and cat-
alytically hydrogenated with Pd-C (50 mg) at rt under
ordinary pressure to remove the benzyl ester group. The
reaction was completed in 5 h, which was monitored by

B. Jose et al. / Bioorg. Med. Chem. 12 (2004) 1351–1356
1355
HPLC. Pd-C was filtered off and methanol was removed
References and notes
to give cyclo(-Ser(tBu)-Asp(OtBu)-Asu(OH)-Thr(tBu)-
Ile-Gly-) (100 mg, 0.123 mmol, 98%). HPLC: 8.6 min
(chromolith, 10–100% CH3CN gradient containing
0.1%TFA over 15 min). The cyclo(-Ser(tBu)-Asp(OtBu)-
Asu(OH)-Thr(tBu)-Ile-Gly-) (100 mg, 0.123 mmol) was
dissolved in DMF (2mL). Hydroxylamine hydrochlo-
ride (43 mg, 0.62mmol), HOBt.H 2O (95 mg, 0.62
mmol) BOP (274 mg, 0.62 mmol) and Et3N (0.18 mL,
1.24 mmol) were added to the solution at 0 ꢂC. The
reaction mixture was stirred for 3 h and monitored by
HPLC. DMF was removed in vacuo and the residue
was treated with TFA (2mL) at 0 ꢂC for 3 h. TFA was
evaporated in vacuo at 0 ꢂC and the residue was applied
to a column of Sephadex LH-20 with DMF. The frac-
tions containing the desired cyclic hexapeptide hydro-
xamic acid were analyzed by HPLC and collected.
DMF was removed and the residue was lyophilized
from water to give cyclo(-Ser-Asp-Asu(NHOH)-Thr-Ile-
Gly-) (81 mg, 0.123 mmol, 100%). The product was
further purified by HPLC. HPLC: 2.8 min (chromolith,
10–100% CH3CN gradient containing 0.1% TFA over
15 min). FABMS gave peaks of (M+H)+ and
(M+Na)+ at m/z 660 and 682respectively. HR-
FABMS (M+H)+ 660.3207 for C27H46O12N7 (calcd
660.3204).
1. Grozinger, G. M.; Schreiber, S. L. Chem. Biol. 2002, 9, 3.
2. Kouzarides, T. Curr. Opin. Genet. Dev. 1999, 9, 40.
3. Kouzarides, T. EMBO J. 2000, 19, 1176.
4. Hassig, C. A.; Schreiber, S. L. Curr. Opin. Chem. Biol.
1997, 1, 300.
5. Khochbin, S.; Verdel, A.; Lemercier, C.; Seigneurin-
Berny, D. Curr. Opin. Genet. Dev. 2001, 11, 162.
6. Yoshida, M.; Kijima, M.; Akita, M.; Beppu, T. J. Biol.
Chem. 1990, 265, 17174.
7. Kijima, M.; Yoshida, M.; Suguta, K.; Horinouchi, S.;
Beppu, T. J. Biol. Chem. 1993, 268, 22429.
8. Hirota, A.; Suzuki, A.; Aizawa, K.; Tamura, S. Biomed.
Mass. Specrom. 1974, 1, 15.
9. Darkin-Rattray, S. J.; Gurnett, A. M.; Myers, R. W.;
Dulski, P. M.; Crumley, T. M.; Allocco, J. J.; Cannova,
C.; Meinke, P. T.; Colletti, S. L.; Bednarek, M. A.; Singh,
S. B.; Goetz, M. A.; Dombrowski, A. W.; Polishook,
J. D.; Schmatz, D. M. Proc. Natl. Acad. Sci. U.S.A. 1996,
93, 13143.
10. Meinke, P. T.; Liberator, P. Curr. Med. Chem. 2001, 8,
211.
11. Han, J. W.; Ahn, S. H.; Park, S. H.; Wang, S. Y.; Bae,
G. U.; Seo, D. W.; Known, H. K.; Hong, S.; Lee, Y. W.;
Lee, H. W. Cancer Res. 2000, 60, 6068.
12. Ueda, H.; Nakajima, H.; Hori, Y.; Fujita, T.; Nishimura,
M.; Goto, T.; Okuhara, M. J. Antibiot. 1994, 47, 301.
13. Ueda, H.; Manda, T.; Matsumoto, S.; Mukumoto, S.;
Nishigaki, F.; Kawamura, I.; Shimomura, K. J. Antibiot.
1994, 47, 315.
3.4. NMR spectroscopy and structure calculation
14. Furumai, R.; Matsuyama, A.; Kobashi, M.; Lee, K.-H.;
Nishiyama, M.; Nakajima, H.; Tanaka, A.; Komatsu, Y.;
Nishino, N.; Yoshida, M.; Horinouchi, S. Cancer Res.
2002, 62, 4916.
15. Gore, S. D.; Carducci, M. A. Exp. Opin. Invest. Drugs
2000, 9, 2923.
16. Phiel, C. J.; Zhang, F.; Huang, E. Y.; Guenther, M. G.;
Lazar, M. A.; Klein, P. S. J. Biol. Chem. 2001, 76, 36734.
17. Richon, V. M.; Emiliani, S.; Verdin, E.; Webb, Y.; Bre-
slow, R.; Rifkind, R. A.; Marks, P. A. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 3003.
NMR spectra were recorded on a JEOL spectrometer
operating at 500 MHz in DMSO-d6 using TMS as
internal standard. Spectra were recorded at variable
temperatures such as, 298, 303, 313, 323, and 333 K
respectively. Assignments of proton resonances were
confirmed, when possible, by COSY, HOHAHA and
NOESY experiments. Mixing times for NOESY experi-
ments were 200 and 400 ms. Number of scans varied
between 32and 64, and the number of points in t
1
2
dimension was 512. H resonances were assigned using
a
18. Jung, M.; Brosch, G.; Kolle, D.; Scherf, H.; Gerhauser,
¨
C.; Loidi, P. J. Med. Chem. 1999, 42, 4669.
¨
3
standard procedures. JNHꢀC
coupling constants for
H
non-overlapping signals were determined from 1D
spectra with high digital resolution.
19. Remiszewski, S. W.; Sambucetti, L. C.; Atadja, P.; Bair,
K. W.; Cornell, W. D.; Green, M. A.; Howell, K. L.;
Jung, M.; Known, P.; Trogani, N.; Walker, H. J. Med.
Chem. 2002, 45, 753.
20. Woo, S. H.; Frechette, S.; Khalil, E. A.; Bouchain, G.;
Vaisburg, A.; Bernstein, N.; Moradei, O.; Leit, S.; Allan,
M.; Fournel, M.; Trachy-Bourget, M. C.; Li, Z.; Bestman,
J. M.; Delorme, D. J. Med. Chem. 2002, 45, 2877.
21. Suzuki, T.; Ando, T.; Tsuchiya, K.; Fukazawa, N.; Saito,
A.; Mariko, Y.; Nakanishi, O. J. Med. Chem. 1999, 42, 3001.
22. Saito, A.; Yamashita, T.; Mariko, Y.; Nosaka, Y.; Tsu-
chiya, K.; Ando, T.; Suzuki, T.; Tsuruo, T.; Nakanishi, O.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4592.
23. Komatsu, Y.; Tomizaki, K.; 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.
24. Furumai, R.; Komatsu, Y.; Nishino, N.; Khochbin, S.;
Yoshida, M.; Horinouchi, S. Proc. Natl. Acad. Sci. U.S.A.
2000, 98, 87.
3.5. HDAC inhibitory activity assay
The preparation and assay of the enzymes were per-
formed as described in ref 14.
3.6. Molecular modeling studies
All calculations were performed on a Silicon Graphics
computer. The distance geometry program was used to
generate structures consistent with the distance con-
straints derived from the NOEs. Temperature coefficient
of NH protons indicating hydrogen bonds and f angles
a
calculated from JNHꢀH were used to filter out struc-
tures that did not meet the experimental data. An error
of ꢃ30ꢂ was tolerated for the f angles calculated from
JaNH-H at this stage of refinement. Energy minimization
and molecular dynamics calculation were carried out
using the CHARMm program of Insight II using
CHARMm forcefield. a-Tubulin structure was used
from Protein Data Bank.29
25. 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.
26. Grozinger, C. M.; Hassig, C. A.; Schreiber, S. L. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 4868.

1356
B. Jose et al. / Bioorg. Med. Chem. 12 (2004) 1351–1356
27. Hubbert, C.; Guardiola, A.; Shao, R.; Kawaguchi, Y.;
Ito, A.; Nixon, A.; Yoshisa, M.; Wang, X.; Yao, T. Nat-
ure 2002, 417, 455.
28. Matsuyama, A.; Yoshimatsu, Y.; Shimazu, T.; Sumida,
Y.; Osada, H.; Komatsu, Y.; Nishino, N.; Khochbin, S.;
Horinouchi, S.; Yoshida, M. EMBO J. 2002, 21, 6820.
29. McKean, P. G.; Vaughen, S.; Gull, K. J. Cell Science
2001, 114, 2723.
30. Nogales, E.; Wolf, S. G.; Downing, K. H. Nature 1998,
391, 199.
31. Wong, J. C.; Hong, R.; Schreiber, S. L. J. Am. Chem. Soc.
2003, 125, 5586.
32. Haggarty, S.; Koeller, K. M.; Wong, J. C.; Grozinger,
C. M.; Schreiber, S. L. Proc. Natl. Acad. Sci. U.S.A. 2003,
100, 4389.