A para-amino substituent on the D-ring of green tea polyphenol epigallocatechin-3-gallate as a novel proteasome inhibitor and cancer cell apoptosis inducer

Article abstract of DOI:10.1016/j.bmc.2007.05.041

Analogs of (-)-EGCG containing a para-amino group on the D-ring in place of the hydroxyl groups have been synthesized and their proteasome inhibitory activities were studied. We found that, the O-acetylated (-)-EGCG analogs possessing a p-NH₂ o

Full text of DOI:10.1016/j.bmc.2007.05.041

Bioorganic & Medicinal Chemistry 15 (2007) 5076–5082
A para-amino substituent on the D-ring of green tea
polyphenol epigallocatechin-3-gallate as a novel
proteasome inhibitor and cancer cell apoptosis inducer
Kumi Osanai,a, Kristin R. Landis-Piwowar,b, ,à Q. Ping Doub,à and Tak Hang Chana,*
aDepartment of Applied Biology and Chemical Technology and the Open Laboratory of Chirotechnology,
Institute of Molecular Technology for Drug Discovery and Synthesis, The Hong Kong Polytechnic University,
Hung Hom, Hong Kong, SAR, China
bThe Prevention Program, Barbara Ann Karmanos Cancer Institute, and Department of Pathology, School of Medicine,
Wayne State University, Detroit, Michigan, USA
Received 23 April 2007; revised 12 May 2007; accepted 15 May 2007
Available online 18 May 2007
Abstract—Analogs of (ꢀ)-EGCG containing a para-amino group on the D-ring in place of the hydroxyl groups have been synthe-
sized and their proteasome inhibitory activities were studied. We found that, the O-acetylated (ꢀ)-EGCG analogs possessing a
p-NH2 or p-NHBoc (Boc; tert-butoxycarbonyl) D-ring (5 and 7) act as novel tumor cellular proteasome inhibitors and apoptosis
inducers with potency similar to natural (ꢀ)-EGCG and similar to (ꢀ)-EGCG peracetate. These data suggest that the acetylated
amino-GTP analogs have the potential to be developed into novel anticancer agents.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
distinct activities, which are associated with three differ-
ent b subunits, chymotrypsin-like (with b5 subunit),
trypsin-like (with b2 subunit), and peptidyl-gluthamyl
peptide-hydrolyzing-like (PGPH- or caspase-like; with
b1 subunit).13–15 Inhibition of the chymotrypsin-like,
but not the trypsin-like, activity of the proteasome has
been found to be associated with induction of tumor cell
apoptosis.16–18 Inhibition of the proteasome prevents
ubiquitin-targeted proteolysis which can affect multiple
signaling cascades within the cell. Since this disruption
of normal homeostatic mechanisms can lead to cell
death, the discovery of new proteasome inhibitors with
little or no toxicity is highly desirable in anticancer
therapy.19,20
Green tea, produced from Camellia sinensis, is a highly
consumed beverage around the world and regular drink-
ing of green tea has been claimed to reduce incidence of
a variety of cancers.1–3 Although a number of green tea
polyphenols (GTPs) have been identified in green tea,4
(ꢀ)-epigallocatechin-3-gallate [1, (ꢀ)-EGCG] is the
most abundant constituent and is considered to be the
most biologically active among the GTPs. A number
of epidemiological and biological studies involving
(ꢀ)-EGCG have been reported in the last decade and
have shown that (ꢀ)-EGCG can reduce or inhibit tumor
growth in breast,5–7 lung,8 and urinary9–11 and GI
tracts.12
We have previously reported that (ꢀ)-EGCG inhibits
the chymotrypsin-like activity of the proteasome
in vitro (IC50 0.1–0.2 lM) and in intact tumor cells (1–
10 lM).21,22 Enantiomerically synthesized (+)-EGCG
and other synthetic analogs of green tea catechins with
an ester bond have also been shown to inhibit the prote-
asomal chymotrypsin-like activity,22,23 leading to accu-
mulation of proteasome target proteins (such as IjB-a,
p27, and Bax) and apoptosis in human cancer cell lines,
as measured by activation of caspases and cleavage of
poly(ADP-ribose) polymerase (PARP).24 Furthermore,
The eukaryotic proteasome is a large multi-catalytic,
multi-subunit protease complex possessing at least three
Keywords: Green tea polyphenols; Epigallocatechin-3-gallate; para-
Aminobenzoic acid; Pro-drug; Proteasome; Apoptosis.
*
Corresponding author. Tel.: +1 852 2766 5605; fax: +1 852 2364
9932; e-mail addresses: doup@karmanos.org; bcchanth@polyu.
edu.hk
These authors contribute equally for this work.
Tel.: +1 313 966 0641; fax: +1 313 966 7368.
à
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.05.041

K. Osanai et al. / Bioorg. Med. Chem. 15 (2007) 5076–5082
5077
Table 1. Inhibition of chymotrypsin-like activity of purified 20S
proteasomea
in silico docking studies have indicated that (ꢀ)-EGCG
predictably binds to the N-terminal threonine (Thr) of
the proteasomal chymotrypsin active site. The orienta-
tion of (ꢀ)-EGCG bound to the proteasome is suitable
for nucleophilic attack by the hydroxyl group of Thr 1
to the carbonyl carbon of (ꢀ)-EGCG, thus inhibiting
the proteasomal chymotrypsin-like activity (Fig. 1).25
Furthermore, the hydroxyl groups of the (ꢀ)-EGCG
D-ring were found to form hydrogen-bonds with Gly47
or Ser131 of the proteasome, thus contributing to the
binding stability of (ꢀ)-EGCG to the proteasome
(Fig. 1). In support of this model, compound 2
(Fig. 2), which contains only one p-hydroxy group on
the D-ring, was found to be a much weaker proteasome
inhibitor than (ꢀ)-EGCG (Table 1).26 Since an amino
group is capable of forming hydrogen bonds both as a
donor and an acceptor, we are therefore interested in
studying the replacement of the hydroxyl group in the
D-ring by an amino group in an attempt to increase
the potency of compound 2.
Compound
IC50 (lM)
(ꢀ)-EGCG
0.2 0.1
nac
3 (Pro-E)
2
4
5
6
7
40.4 0.17b
0.84 0.06
nac
3.85 0.33
nac
a Results obtained from 3 independent experiments performed in
triplicate.
b Ref. 26.
c na indicates that the inhibitory activity of the purified proteasome at
50 lM was <20%.
availability in vivo.30 Consistently, even though Pro-E
has no inhibitory effect against a purified 20S protea-
some, it nevertheless showed much higher potency than
(ꢀ)-EGCG to inhibit proliferation and transforming
activity and to induce apoptosis in human prostate,
breast, leukemic, and simian virus 40-transformed
cells.31 Recently, in a related study, we showed that
Pro-E can be converted to (ꢀ)-EGCG in human breast
cancer MDA-MB-231 cell cultures and xenografts, lead-
ing to a higher intracellular concentration (>2.4-fold) of
(ꢀ)-EGCG than those cells treated with same dose of
(ꢀ)-EGCG.32 Pro-E proved to be more efficacious in
inhibiting breast cancer tumor growth in mice than
(ꢀ)-EGCG.32
A critical issue concerning the potential application of
(ꢀ)-EGCG as an anticancer agent is its known low bio-
availability which is thought to be partly due to the poor
stability of (ꢀ)-EGCG in alkaline or neutral solu-
tions.27,28 Because the pH values of intestine and body
fluids are neutral or slightly alkaline, (ꢀ)-EGCG is
potentially unstable inside the human body.27 Addition-
ally, in vivo metabolic transformations of (ꢀ)-EGCG by
glucuronidation, sulfonation or methylation into vari-
ous metabolites may also contribute to its reduced
bioavailability.29
With the knowledge that O-acetyl protected GTPs could
be cytotoxic against tumor, but not normal cells,31,32 we
have continued to search for more potent anticancer
agents. We have previously studied the biological activ-
ities of some EGCG analogs with modifications of
ABC-ring moiety and gallate (D-ring) moiety.25–27,31,32
Here we report structure–activity relationship (SAR)
analysis with newly designed analog compounds, which
possess a para-amino substituent on the D-ring (com-
pound 4), as well as the O-acetyl derivatives 5, and their
corresponding N-tert-butoxycarbonyl (Boc) derivatives
6 and 7.
Recently, we suggested that (ꢀ)-EGCG peracetate (3,
Pro-E), a synthetic derivative of (ꢀ)-EGCG, can act as
a pro-drug.27 Pro-E is also converted under cellular con-
ditions by esterases to (ꢀ)-EGCG with enhanced bio-
H
Thr21
O
O
Lys32
O
OH
O
B
H
O
O
H
H
S1 pocket
Ala49
A
C
O
Ala146
3.18 Å
H
H
O
O
H
O
O
O
Gly47
D
H
Thr1
proteasome
N
O
H
H2N
2. Results and discussion
OH
Ser131
N
O
2.1. Enantioselective syntheses of amino-GTP analogs 4–
7
Figure 1. Docking model of (ꢀ)-EGCG in the b5-subunit of 20S
˚
proteasome. The number is distance in A for the H-bond indicated.
The syntheses of compounds 4–7 were achieved as out-
lined in Scheme 1. We have previously reported on the
total synthesis of (ꢀ)-EGCG in which (ꢀ)-(2R, 3R)-
5,7-bis(benzyloxy)-2-[3,4,5-tris(benzyloxy)phenyl]chro-
man-3-ol (8) was prepared as the key intermediate.33
Esterification of 8 with boc-4-aminobenzoic acid (9)
afforded 10 with excellent yield. Catalytic hydrogenation
of 10 removed the benzyl protecting group readily to
give 6. Deprotection of Boc group with trifluoroacetic
acid furnished 4. On the other hand, acetylation of 6
with Ac2O and pyridine gave 7 which on subsequent
deprotection of Boc group gave 5.
OH
OR
OH
OH
OR
OR
HO
O
RO
O
O
O
OH
OR
OR
OR
O
O
OH
OR
1: R=H, (-)-EGCG,
3: R=Ac, Pro-E
2
Figure 2. Structures of selected GTPs and analogs.

5078
K. Osanai et al. / Bioorg. Med. Chem. 15 (2007) 5076–5082
OBn
OBn
OH
OH
OBn
OBn
OBn
OBn
OH
OH
OH
OH
NHBoc
HO2
C
BnO
O
BnO
O
HO
O
HO
O
9
R
R
i
ii
iii
O
O
OH
O
OBn
OBn
OH
OH
O
O
O
8
10
6
4
NHBoc
NHBoc
NH2
iv
OAc
OAc
OAc
OAc
OAc
OAc
AcO
O
AcO
O
O
O
v
OAc
OAc
O
O
7
5
NHBoc
NH2
Scheme 1. Synthetic route of amino GTP analogs. Reagents and conditions: (i) DCC, DMAP, CH2Cl2, rt, 20 h; (ii) Pd(OH)2, H2, THF/MeOH, rt,
1 h; (iii) CHCl3/TFA, rt, 30 min; (iv) Ac2O, pyridine, rt, 30 min; (v) CHCl3/TFA, rt, 6.5 h.
2.2. Inhibition of chymotrypsin-like activity of a purified
20S proteasome
CT-like Activity
4h
24h
120
100
80
60
40
20
0
We examined these synthetic amino-GTPs, both unpro-
tected (4 and 6) and O-acetyl protected (5 and 7), for
their SAR in comparison with natural (ꢀ)-EGCG (1),
2, and Pro-E (3) against the chymotrypsin-like activity
of a purified 20S proteasome (Table 1). Compound 4
exhibited reasonably strong proteasome inhibition
(IC50 value, 0.84 lM). Even though the activity of 4 is
not as high as (ꢀ)-EGCG (IC50value, 0.20 lM,34) it is
much more active than compound 2 of which a hydroxyl
group is in place of the amino group in compound 4
(Fig. 2 vs Scheme 1). This suggests that an amino substi-
tuent is more effective than a hydroxyl in enhancing pro-
teasome inhibition.
DMSO ProE EGCG
4
6
7
5
Figure 3. O-Acetylated amino-GTP analogs are potent proteasome
inhibitors in cultured tumor cells. Leukemia Raji B cells were treated
with 25 lM of each analog for 4 and 24 h, harvested and analyzed for
the chymotrypsin-like activity, as described in Section 4.
On the other hand, compound 6, with a p-NHBoc on the
D-ring, is less active (IC50 = 3.85 lM) than 4 but still
much more active than compound 2. We also examined
the O-acetyl protected analogs 5 and 7 to evaluate their
proteasome-inhibitory activities. As expected, the O-
acetyl protected analogs were not potent in inhibiting
the chymotrypsin-like activity of the purified 20S pro-
teasome due to the lack of cellular esterases for their
conversion into de-acetyl and activated forms.30,32,34
These acetylated analogs displayed exceptionally weak
inhibitory activity against the purified 20S proteasome
[less than 20% inhibition at 50 lM concentration (Table
1)]. This finding again demonstrates that O-acetyl ana-
logs are inactive in a purified enzyme system.
some activity at the same time points, while (ꢀ)-EGCG
induced only 4% and 10% inhibition at 4 and 24 h,
respectively (Fig. 3). These findings suggest that 4 and
6 are unstable within the cellular environment, similar
to (ꢀ)-EGCG. However, the O-acetylated analogs 5
and 7 exhibited far greater potency against the proteaso-
mal chymotrypsin-like activity, 39% and 65% inhibition,
respectively, after 24 h.
Inhibition of proteasomal activity should lead to accu-
mulation of ubiquitinated proteins and proteasome tar-
get proteins, such as IjB-a, p27, and Bax.16,17 Both 4
and 6 slightly increased levels of ubiquitinated proteins
in a time-dependent manner, although they had little ef-
fect on levels of IjB-a, p27, and Bax proteins (Fig. 4). In
contrast, 5 and 7 were more potent with respect to accu-
mulation of ubiquitinated proteins and proteasome tar-
get proteins, after both 4-h and 24-h treatment (Fig. 4).
Consistent with the level of proteasome inhibition ob-
served in whole cells (Fig. 3), 7 induced higher levels
of ubiquitinated proteins than 5 at both 4 and 24 h
(Fig. 4). After 4-h treatment with 7, IjB-a, p27, and
Bax were accumulated by 1.4-, 2.1-, and 3.5-fold,
respectively. In contrast, a treatment with 5 did not
accumulate IjB-a and accumulated only p27 and Bax
by 2.0- and 3.3-fold, respectively, after 4 h (Fig. 4).
2.3. Inhibition of proteasome activity in intact tumor cells
In order to determine whether the synthetic amino-GTP
analogs exhibit proteasomal inhibition in whole cells,
leukemia Raji B cells were treated with 25 lM of each
amino-GTP analog for 4 and 24 h (Fig. 3). Inhibition
of intact tumor cell proteasome activity was assayed in
whole cell extracts using a chymotrypsin-like activity as-
say. Compound 6 was unable to inhibit the chymotryp-
sin-like activity in whole cells after 4- and 24-h
incubation and compound 4 exhibited minimal protea-

K. Osanai et al. / Bioorg. Med. Chem. 15 (2007) 5076–5082
5079
24h
4h
4h
24h
Trypan Blue
70
60
4h
4h
24h
24h
6
7
5
6
7
5
50
40
30
20
10
0
4
4
IkBa
1.0 1.0 0.9 1.0 1.3 1.3 1.0 1.0 1.4 1.0 1.0 0.7 1.2 1.0
p27
Bax
1.0 1.4 2.1 2.0 1.0 1.2 1.5 1.4
1.0 1.8 3.5 3.3 1.0 0.9 1.6 1.0
1.0 1.1 1.2 1.0 1.1 1.2
1.0 1.6 1.6 1.0 1.8 1.7
DMSO ProE EGCG
4
6
7
5
b
c
Caspase-3 Activity
600
500
400
300
200
100
0
4h
24h
Ub
DMSO ProE EGCG
4
6
7
5
4h
24h
24h
4h
Actin
Figure 4. O-Acetylated amino-GTP analogs accumulate proteasome
target proteins in cultured tumor cells. Western blot analysis using
specific antibodies to Ubiquitin, Bax, p27, IjB-a, and actin.
4
4
6
7
5
6
7
5
PARP
p85/PARP
Figure 5. O-Acetylated amino-GTP analogs induce time-dependent
cell death in cultured tumor cells. Leukemia Raji B cells were treated
with the solvent (DMSO) or 25 lM of the indicated analog for 4 or
24 h, followed by Trypan blue dye exclusion assay. The data
represented are the mean number of dead cells over total cell
population SD after treatment (a). A cell-free caspase-3/-7 activity
assay was performed as described in Section 4. Caspase-3/-7 activity is
represented as a percentage of the solvent (DMSO) control (b).
Western blot analysis was done using specific antibodies to PARP (c).
2.4. Induction of apoptotic cell death
It has been shown that inhibition of the proteasomal
chymotrypsin-like activity is associated with induction
of apoptosis in a wide variety of cancer cells.16,17 We
therefore determined the cytotoxicity of the O-acety-
lated and un-acetylated forms of the amino-GTP ana-
logs by Trypan blue dye exclusion analysis. Leukemia
Raji B cells were treated with each pair of the acetylated
and un-acetylated analogs at 25 lM for 4- and 24-h, fol-
lowed by Trypan blue analysis (Fig. 5a). Blue cells and
cells with apoptosis-associated morphological changes
(shrunken, blebbing, etc.) were scored as dead cells.
We found that the un-acetylated amino-GTP analogs
(4 and 6) induced no more than 10% cell death at up
to 24 h (Fig. 5a) supporting the idea that they are either
unstable or inactive within the cell.30,32,34 In contrast,
the acetylated analogs (5 and 7) were much more potent
in inducing the cell death (Fig. 5a). Cell death was in-
creased by 3.5- and 5-fold in cells treated with the acetyl
protected amino-GTP analogs 5 and 7, respectively,
after 24-h incubation compared to cells treated with
(ꢀ)-EGCG (Fig. 5a). To determine whether the ob-
served cell death (Fig. 5a) was representative of apopto-
sis, aliquots of the cell extracts from the same
experiment were utilized for measurement of caspase-3
activation (Fig. 5b) and poly(ADP-Ribose) polymerase
(PARP) cleavage (Fig. 5c). After 24-h treatment, 5 and
7 induced a 4.7- and 5.5-fold increase in caspase-3 activ-
ity, respectively (Fig. 5b). Additionally Western blot
analysis revealed that apoptosis-specific PARP cleavage
was observed in cells treated with analogs 5 and 7 after
24 h (Fig. 5c). In contrast, the PARP cleavage was not
observed in cells treated with the un-acetylated amino-
GTP analogs 4 and 6 (Fig. 5c). These results indicate
that the acetyl protected analogs are more potent than
the un-acetylated analogs in inhibiting proteasome
activity and inducing apoptotic cell death in a time-
dependent manner. These results are consistent with
the hypothesis that compounds 5 and 7 are behaving
as pro-drugs of the un-acetylated analogs 4 and 6, in
the same manner as Pro-E serving as the pro-drug for
EGCG.27,32
3. Conclusions
In conclusion, we have demonstrated that a para-amino
substituent on the D-ring of green tea polyphenols
exhibits proteasome-inhibitory activity with potency
similar to that of (ꢀ)-EGCG and ProE. Furthermore,
the O-acetylated amino-GTP analogs appear to behave
as prodrugs and the mechanism of action involves tar-
geting the proteasome in tumor cells thereby inducing
cell death. Although the para-amino compounds are
not inherently more potent than ProE, the presence of
the para-amino substituent could increase the bioavail-
ability of these compounds by limiting biotransforma-
tion reactions to the hydroxyls, such as glucuronidation,
sulfonation, and methylation.29 Since the para-amino-
benzoic acid moiety is a normal constituent of folic acid
and generally presumed to be innocuous,35 the study of
these para-amino substituted green tea polyphenols may
offer the possibility of discovering novel anticancer
drugs.

5080
K. Osanai et al. / Bioorg. Med. Chem. 15 (2007) 5076–5082
4. Experimental
156.51, 153.74, 152.87, 143.95, 139.14, 138.73, 137.94,
137.82, 137.76, 134.28, 132.10, 129.57, 129.49, 129.33,
129.02, 128.99, 128.86, 128.70, 128.66, 128.57, 128.42,
128.16, 125.02, 118.17, 107.53, 101.91, 95.67, 94.88,
82.14, 78.82, 76.04, 72.10, 71.11, 70.92, 69.04, 29.21,
27.07; HRMS m/z calculated for C62H57O10Na
(M+Na) 998.3880, found 998.3845.
4.1. Reagents
Fetal Bovine Serum was purchased from Tissue Cul-
ture Biologicals (Tulare, CA). RPMI 1640, penicillin,
and streptomycin were purchased from Invitrogen
(Carlsbad,CA). Dimethylsulfoxide (DMSO) and (ꢀ)-
EGCG were purchased from Sigma (St. Louis, MO).
Suc-Leu-Leu-Val-Tyr-AMC (a proteasomal chymo-
trypsin-like substrate) and Ac-DEVD-AMC (a cas-
pase-3 substrate) were obtained from Biomol
(Plymouth Meeting, PA). Purified 20S proteasome
from rabbit was acquired from Boston Biochem (Cam-
bridge, MA). Monoclonal antibodies to Bax (H280)
and Ubiquitin (P4D1), polyclonal antibodies to IjB-a
(C15) and Actin (C11), and anti-goat, anti-rabbit,
and anti-mouse IgG-horseradish peroxidase were pur-
chased from Santa Cruz Biotechnology (Santa Cruz,
CA). Monoclonal antibody to p27 (554069) was from
BD Biosciences (San Diego, CA). The polyclonal anti-
body to PARP was purchased from Biosource (Cama-
rillo, CA).
4.4. (ꢀ)-(2R, 3R)-5,7-Dihydroxy-2-(3,4,5-trihydroxyphe-
nyl)-chroman-3-yl 40-(tert-butoxycarbonyl)-aminobenzo-
ate (6)
To a solution of (ꢀ)-(2R, 3R)-5,7-bis(benzyloxy)-2-
[3,4,5-tris(benzyloxy)phenyl]chroman-3-yl
40-N-(tert-
butoxy-carbonyl)aminobenzoate (10, 101 mg, 104 lmol)
in THF (10.0 mL) and MeOH (10.0 mL) was added pal-
ladium hydroxide on carbon powder [20% Pd (100 mg)].
The mixture was stirred under a H2 atmosphere at room
temperature for 1 h, filtered, and eluted with MeOH and
the eluate was evaporated in vacuo. The obtained color-
less oil (66.8 mg) was purified by flash SiO2 column
chromatography (AcOEt/hexane, 2:1) to give 45.3 mg
(83%) of the title compound as a pale yellow amorphous
20
D
1
solid: ½aꢁ = ꢀ88ꢁ (c 0.14, MeOH); H NMR (CDCl3)
4.2. Chemical synthesis
(complexity due to rotamers from the amide function)
d 7.81 (d, J = 9.5 Hz, 1/6 · 2H), 7.78 (d, J = 9.5 Hz,
5/6 · 2H), 7.45 (d, J = 9.5 Hz, 1/6 · 2H), 7.43 (d,
J = 9.5 Hz, 5/6 · 2H), 6.54 (s, 1/6 · 2H), 6.52 (s,
5/6 · 2H), 6.05 (d, J = 2.4 Hz, 1/6 · 1H), 5.99–5.97 (m,
1H), 5.96 (d, J = 2.4 Hz, 5/6 · 1H), 5.54 (br s,
1/6 · 1H), 5.52 (br s, 5/6 · 1H), 5.10-4.68 (m, 5H), 5.02
(s, 1/6 · 1H), 5.00 (s, 5/6 · 1H), 3.03 (dd, J = 17.0,
4.4 Hz, 1/6 · 1H), 3.00 (dd, J = 17.0, 4.4 Hz, 5/6 · 1H),
2.91 (dd, J = 17.0, 2.0 Hz, 1/6 · 1H), 2.89 (dd,
J = 17.0, 2.0 Hz, 5/6 · 1H), 1.51 (s, 1/6 · 9H), 1.49
(s, 5/6 · 9H); 13C NMR (CDCl3) d 168.06, 158.74,
158.65, 158.01, 147.55, 146.23, 134.54, 132.60, 131.61,
125.71, 119.36, 107.56, 107.51, 100.13, 97.40, 97.32,
96.66, 96.58), 82.20, 79.37, 79.28, 71.31, 19.43, 29.40,
27.50; HRMS m/z calculated for C27H27NO10Na
(M+Na) 548.1533, found 548.1537.
The synthesis was accomplished according to Scheme 1.
All reactions were performed under an atmosphere of
N2, and glassware was dried completely in an oven at
110 ꢁC prior to use. Tetrahydrofuran (THF) was dried
by distillation over sodium benzophenone, and dry
dichloromethane (CH2Cl2), dimethylformamide (DMF),
and toluene were obtained by distillation from CaH2.
Unless otherwise stated, solvents or reagents were used
as received without further purification. (ꢀ)-(2R, 3R)-
5,7-bis(benzyloxy)-2-[3,4,5-tris(benzyloxy)benzyl]-3-chrom-
an-3-ol (8) was prepared by following reported
procedure.33
4.3. (ꢀ)-(2R, 3R)-5,7-Bis(benzyloxy)-2-[3,4,5-tris(benzyl-
oxy)phenyl]chroman-3-yl 4-N-(tert-butoxycarbonyl)-ami-
nobenzoate (10)
4.5. (ꢀ)-(2R,3R)-5,7-Dihydroxy-2-(3,4,5-trihydroxyphe-
To a solution of 4-N-Boc-aminobenzoic acid (9, 166 mg,
698 lmol) in CH2Cl2 (1.00 mL) was added N,N0-dic-
yclohexylcarbodiimide (217 mg, 1.05 mmol). The mix-
ture was stirred at room temperature for 10 min,
cooled to 0 ꢁC. 4-Dimethylaminopyridine (21.4 mg,
175 lmol) was added to the solution and the mixture
was stirred for 5 min. A solution of (ꢀ)-(2R, 3R)-5,7-
bis(benzyloxy)-2-[3,4,5-tris(benzyloxy)phenyl]-chroman-
3-ol (8, 264 mg, 349 lmol) in CH2Cl2 (2.50 mL) was
added dropwise at 0 ꢁC and the mixture was stirred at
room temperature overnight. The solvent was evapo-
rated in vacuo and the resulting oil was purified by flash
SiO2 column chromatography (hexane/EtOAc, 4:1) to
nyl)-chroman-3-yl 4-aminobenzoate (4)
To a solution of (ꢀ)-(2R,3R)-5,7-dihydroxy-2-(3,4,5-tri-
hydroxyphenyl)chroman-3-yl 4-(tert-butoxycarbonyl)-
aminobenzoate (6, 20.6 mg, 39.2 lmol) in CHCl3
(800 lL) was added trifluoroacetic acid (160 lL) and
the mixture was stirred at room temperature for
30 min. The reaction mixture was directly evaporated
in vacuo and 28.5 mg (>99%) of the title compound
20
D
was obtained as a pale brown solid: ½aꢁ = ꢀ72ꢁ (c
1
0.19, MeOH); H NMR (CDCl3) d7.67 (d, J = 9.8 Hz,
2H), 6.63 (d, J = 9.8 Hz, 2H), 6.54 (s, 2H), 6.00 (br s,
1H), 5.99 (br s, 1H), 5.51 (br s, 1H), 5.01 (s, 1H),
4.96–4.82 (m, 5H), 3.01 (dd, J = 16.8, 4.9 Hz, 1H),
2.89 (dd, J = 16.8, 3.3 Hz, 1H); 13C NMR (CDCl3)
d168.97, 158.74, 158.68, 158.08, 151.05, 147.55, 134.57,
133.60, 133.51, 131.75, 119.52, 115.21, 115.09, 107.72,
107.61, 100.33, 97.35, 96.60, 79.48, 70.61, 27.58; HRMS
m/z calculated for C32H29NO13 426.1189, found
426.1205.
give 308 mg (90%) of the title compound as a pale yellow
20
D
amorphous solid: ½aꢁ = ꢀ60ꢁ (c 1.05, CHCl3); 1H
NMR (CDCl3) d7.91 (d, J = 8.6 Hz, 2H), 7.48–7.17
(m, 27H), 6.78 (s, 2H), 6.61 (s, 1H), 6.34 (br s, 1H),
6.29 (br s, 1H), 5.66 (br s, 1H), 5.08–4.91 (m, 8H),
4.76 (d, J = 11.9, 2H), 3.14–3.04 (m, 2H), 1.51 (s,
9H); 13C NMR (CDCl3) d165.95, 159.72, 158.92,

K. Osanai et al. / Bioorg. Med. Chem. 15 (2007) 5076–5082
5081
4.6. (ꢀ)-(2R,3R)-5,7-Diacetoxy-2-(3,4,5-triacetoxyphe-
nyl)-chroman-3-yl 4-(tert-butoxycarbonyl)-aminobenzo-
ate (7)
uitinated proteins was performed using monoclonal or
polyclonal antibodies, according to previously reported
protocols.21 Densitometry was quantified using Alpha-
Ease FC software (Alpha Innotech Corporation, San
Leandro, CA).
To a solution of (ꢀ)-(2R, 3R)-5,7-dihydroxy-2-(3,4,5-tri-
hydroxyphenyl)chroman-3-yl
4-(tert-butoxycarbonyl)-
aminobenzoate (6, 94.0 mg, 179 lmol) in pyridine
(1.00 mL) was added acetic anhydride (130 lL,
1.38 mmol) and the mixture was stirred at room tempera-
ture for 30 min. The solvent was evaporated in vacuo and
the resulting yellow oil (113 mg) was purified by flash SiO2
column chromatography (hexane/EtOAc, 1:1) to give
4.10. Inhibition of purified 20S proteasome activity by
synthetic amino-GTP analogs
Measurement of the chymotrypsin-like activity of the
20S proteasome was performed by incubating 35 ng of
purified rabbit 20S proteasome with 40 lM of fluoro-
genic peptide substrate, Suc-Leu-Leu-Val-Tyr-AMC,
with or without a natural or synthetic GT.37
58.1 mg (59%) of the title compound as a colourless amor-
20
D
phous solid: ½aꢁ = ꢀ48ꢁ (c 1.11, CHCl3); 1H NMR
(CDCl3) d7.78 (d, J = 8.5 Hz, 2H), 7.34 (d, J = 8.5 Hz,
2H), 7.27 (s, 2 H), 6.74 (d, J = 1.9 Hz, 1H), 6.68 (s, 1H),
6.59 (d, J = 1.9 Hz, 1H), 5.62–5.58 (m, 1H), 5.20 (s,
1H), 3.06 (br s, 2H), 2.30–2.27 (m, 6H), 2.26–2.23 (m,
9H), 1.50 (s, 9H); 13C NMR (CDCl3) d169.93, 169.42,
168.55, 167.72, 166.40, 155.77, 153.00, 150.66, 144.31,
143.95, 136.58, 135.19, 132.15, 124.43, 119.71, 118.21,
110.73, 109.84, 108.96, 82.09, 77.54, 68.14, 29.20, 26.94,
22.06, 21.75, 21.58, 21.10; HRMS m/z calculated for
C37H37NO15Na (M+Na) 758.2061, found 758.2068.
4.11. Inhibition of proteasome activity in intact tumor
cells by synthetic amino-GTP analogs
Cells were treated with each compound at 25 lM for 4
or 24 h, harvested, and lysed as described previously.16
Whole cell extracts (10 lg) were incubated with Suc-
Leu-Leu-Val-Tyr-AMC (40 lM) fluorogenic substrate
at 37 ꢁC in 100 lL of assay buffer (50 mM Tris–HCL,
pH 8) for 2.5 h. After incubation, production of hydro-
lyzed 7-amino-4-methylcoumarin (AMC) groups was
measured using a Victor 3 Multilabel Counter with an
excitation filter of 380 nm and an emission filter of
460 nm (Perkin-Elmer, Boston, MA, USA).
4.7. (ꢀ)-(2R,3R)-5,7-Diacetoxy-2-(3,4,5-triacetoxyphe-
nyl)-chroman-3-yl 4-aminobenzoate (5)
To a solution of (ꢀ)-(2R,3R)-5,7-diacetoxy-2-(3,4,5-tri-
acetoxyphen)chroman-3-yl
4-(tert-butoxycarbonyl)-
4.12. Induction of caspase-3 activity by synthetic amino-
GTP analogs
aminobenzoate (7, 60.0 mg, 81.6 lmol) in CHCl3
(1.60 mL) was added trifluoroacetic acid (320 lL) and
the mixture was stirred at room temperature for 6.5 h.
The reaction was quenched with saturated aqueous
NaHCO3 at 0 ꢁC and the mixture was extracted with
CHCl3 (3· 10.0 mL). The combined organic layers were
dried (Na2SO4) and concentrated in vacuo. The ob-
tained brown red amorphous solid (65.0 mg) was puri-
fied by flash SiO2 column chromatography (1%
triethylamine, EtOAc/hexane, 2:1 ! EtOAc) to give
Cells were treated with each compound at 25 lM for 4
or 24 h, harvested, and lysed as described previously.21
Ac-DEVD-AMC (40 lM) was then incubated with the
prepared cell lysates for 2.5 h and the caspase-3 activity
was measured as described previously.38
4.13. Trypan blue assay and apoptotic morphology
changes
26.4 mg (51%) of the title compound as a yellow amor-
20
phous solid: ½aꢁ = ꢀ54ꢁ (c 1.32, CHCl3); 1H NMR
The Trypan blue dye exclusion assay was used to ascer-
tain cell death in Raji cells treated with either a natural
or synthetic compound at 25 lM for 4 or 24 h. Cell mor-
phology was assessed using phase-contrast microscopy
as described previously.22,31
D
(CDCl3) d7.57 (d, J = 8.7 Hz, 2H), 7.20 (s, 2H), 6.65
(br s, 1H), 6.50 (br s, 1H), 6.46 (d, J = 8.7 Hz, 2H),
5.51 (br s, 1H), 5.10 (s, 1H), 3.00–2.90 (m, 2H), 2.20
(s, 3H), 2.19 (s, 3H), 2.17 (s, 3H), 2.15 (s, 6H); 13C
NMR (CDCl3) d169.32, 168.80, 167.95, 167.13, 166.07,
155.18, 151.01, 150.03, 149.93, 143.62, 136.08, 134.52,
132.18, 119.16, 114.31, 110.26, 109.11, 108.26, 76.97,
66.89, 26.35, 21.40, 21.09, 20.91, 20.45; HRMS m/z cal-
culated for C32H29NO13 635.1638, found 635.1643.
Acknowledgments
We thank the Areas of Excellence Scheme established
under the University Grants Committee of the Hong
Kong Administrative Region, China (Project No.
AoE/P-10/01), for financial support. Funding support
4.8. Cell culture
Human leukemia Raji B cells were cultured in RPMI sup-
plemented with 10% (v/v) fetal bovine serum, 100 U/mL
penicillin, and 100 lg/mL streptomycin. Cell cultures were
maintained in a 5% CO2 atmosphere at 37 ꢁC.
by
National
Cancer
Institute
(Grant
Nos.
1R01CA120009 and 5R03CA112625) is gratefully
acknowledged.
4.9. Cell extract preparation and Western blotting
References and notes
1. Fujiki, H.; Suganuma, M.; Matsuyama, S.; Miyazaki, K.
Curr. Cancer Ther. Rev. 2005, 1, 109–114.
Whole cell extracts were prepared as described previ-
ously.36 Analysis of IjB-a, p27, Bax , PARP, and ubiq-

5082
K. Osanai et al. / Bioorg. Med. Chem. 15 (2007) 5076–5082
2. Isemura, M.; Saeki, K.; Kimura, T.; Hayakawa, S.;
22. Smith, D. M.; Wang, Z.; Kazi, A.; Li, L.-H.; Chan, T.-H.;
Dou, Q. P. Mol. Med. 2002, 8, 382–392.
23. Wan, S. B.; Chen, D.; Dou, Q. P.; Chan, T. H. Bioorg.
Med. Chem. 2004, 12, 3521–3527.
24. Li, B.; Dou, Q. P. Proc. Nat. Acad. Sci. U.S.A. 2000, 97,
3850–3855.
25. Smith, D. M.; Daniel, K. G.; Wang, Z.; Guida, W. C.;
Chan, T. H.; Dou, Q. P. Proteins: Structure, Function, and
Bioinformatics 2004, 54, 58–70.
Minami, T.; Sazuka, M. Biofactors 2000, 13, 81–85.
3. Fujiki, H. J. Cancer Res. Clin. Oncol. 1999, 125, 589–597.
4. Liang, Y.; Lu, J.; Zhang, L. Int. J. Food Sci. Tech. 2002,
37, 627–634.
5. Guo, S.; Lu, J.; Subramanian, A.; Sonenshein, G. E.
Cancer Res. 2006, 66, 5322–5329.
6. Sun, C.-L.; Yuan, J.-M.; Koh, W.-P.; Yu, M. C. Carci-
nogenesis 2006, 27, 1310–1315.
7. Sartippour, M. R.; Heber, D.; Ma, J.; Lu, Q.; Go, V. L.;
Nguyan, M. Nutr. Cancer 2001, 40, 149–156.
8. Landau, J. M.; Wang, Z.-Y.; Yang, G.-Y.; Ding, W.;
Yang, C. S. Carcinogenesis 1998, 19, 501–507.
9. Bettuzzi, S.; Brausi, M.; Rizzi, F.; Castagnetti, G.;
Peracchia, G.; Corti, A. Cancer Res. 2006, 66, 1234–1240.
10. Kikuchi, N. O. K.; Shimazu, T.; Nakaya, N.; Kuriyama,
S.; Nishino, Y.; Tsubono, Y.; Tsuji, I. Brit. J. Cancer 2006,
95.
26. Osanai, K.; Huo, C.; Landis-Piwowar, K. R.; Dou, P. Q.;
Chan, T. H. Tetrahedron, accepted for publication.
27. Lam, W. H.; Kazi, A.; Kuhn, D. J.; Chow, L. M. C.;
Chan, A. S. C.; Dou, Q. P.; Chan, T. H. Bioorg. Med.
Chem. 2004, 12, 5587–5593.
28. Chen, Z.-Y.; Zhu, Q. Y.; Tsang, D.; Huang, Y. J. Agric.
Food Chem. 2001, 49, 477–482.
29. Lu, H.; Meng, X.; Yang, C. S. Drug Metab. Dispos. 2003,
31, 572–579.
11. Kemberling, J. K.; Hampton, J. A.; Keck, R. W.; Gomez,
M. A.; Selman, S. H. J. Urol. 2003, 170.
12. Horie, N.; Hirabayashi, N.; Takahashi, Y.; Miyauchi, Y.;
Taguchi, H.; Takeishi, K. Biol. Pharm. Bull. 2005, 28, 574–579.
13. Schwartz, A. L.; Ciechanover, A. Ann. Rev. Med. 1999, 50,
57–74.
30. Lambert, J. D.; Sang, S.; Hong, J.; Kwon, S.-J.; Lee, M.-J.;
Ho, C.-T.; Yang, C. S. Drug Metab. Dispos. 2006, 34, 2111–
2116.
31. Kuhn, D. J.; Lam, W. H.; Kazi, A.; Daniel, K. G.; Song,
S.; Chow, L. M. C.; Chan, T. H.; Dou, Q. P. Front. Biosci.
2005, 10, 1010–1023.
14. Groll, M.; Ditzel, L.; Lo¨we, J.; Stock, D.; Bochtler, M.;
Bartunik, H. D.; Huber, R. Nature 1997, 386, 463–471.
15. Lo¨we, J. S. D. J. B.; Zwickl, P.; Baumeister, W.; Huber, R.
Science 1995, 268, 533–539.
16. An, B.; Goldharb, R. H.; Siman, R.; Dou, Q. P. Cell
Death Differ. 1998, 5, 1062–1075.
17. Lopes, U. G.; Erhardt, P.; Yao, R.; Cooper, G. M. J. Biol.
Chem. 1997, 272, 12893–12896.
18. Ciechanover, A.; Orian, A.; Schwartz, A. L. BioEssays
2000, 22, 443–451.
19. Smith, D. M.; Daniel, K. G.; Dou, Q. P. Lett. Drug Des.
Discov. 2005, 2, 74–81.
20. Adam, J. Cancer Cell 2004, 5, 417–421.
21. Nam, S.; Smith, D. M.; Dou, Q. P. J. Biol. Chem. 2001,
276, 13322–13330.
32. Landis-Piwowar, K. R.; Huo, G.; Chen, D.; Milacic, V.;
Shi, G.; Chan, T. H.; Dou, P. Q. Cancer Res. 2007, 67,
4303–4310.
33. Li, L.; Chan, T. H. Org. Lett. 2001, 3, 739–741.
34. Landis-Piwowar, K. R.; Kuhn, D. J.; Wan, S. B.; Chen,
D.; Chan, T. H.; Dou, Q. P. Int. J. Mol. Med. 2005, 15,
735–742.
35. Patrick, G. L. An Introduction to Medicinal Chemistry, 2nd
ed.; Oxford University Press: Oxford, 2001, pp 384–386.
36. An, B.; Dou, Q. P. Cancer Res. 1996, 56, 438–442.
37. Kazi, A.; Wang, Z.; Kumar, N.; Falsetti, S. C.; Chan, T.
H.; Dou, Q. P. Anticancer Res. 2004, 24, 943–954.
38. Okushio, K.; Suzuki, M.; Matsumoto, N.; Nanjo, F.;
Hara, Y. Biosci., Biotech. Biochem. 1999, 63,
430–432.