A novel tamoxifen derivative, ridaifen-F, is a nonpeptidic small-molecule proteasome inhibitor

Article abstract of DOI:10.1016/j.ejmech.2013.11.009

In a survey of nonpeptide noncovalent inhibitors of the human 20S proteasome, we found that a novel tamoxifen derivative, RID-F (compound 6), inhibits all three protease activities of the proteasome at submicromolar levels. Structure-activity relationship

Full text of DOI:10.1016/j.ejmech.2013.11.009

European Journal of Medicinal Chemistry 71 (2014) 290e305
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
A novel tamoxifen derivative, ridaifen-F, is a nonpeptidic
small-molecule proteasome inhibitor
,
a
a
b
b
Makoto Hasegawa a , Yukari Yasuda , Makoto Tanaka , Kenya Nakata , Eri Umeda ,
*
Yanwen Wang b, Chihiro Watanabe b, Shoko Uetake b, Tatsuki Kunoh a,
**
Masafumi Shionyu a, Ryuzo Sasaki a, Isamu Shiina b , Tamio Mizukami
,
a
a Faculty of Bioscience, Nagahama Institute of Bio-Science and Technology, 1266 Tamura-cho, Nagahama, Shiga 526-0829, Japan
b Department of Applied Chemistry, Faculty of Science, Tokyo University of Science, 1-3 Kagurazaka, Shinjuku-ku, Tokyo 162-8601, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
In a survey of nonpeptide noncovalent inhibitors of the human 20S proteasome, we found that a novel
tamoxifen derivative, RID-F (compound 6), inhibits all three protease activities of the proteasome at
submicromolar levels. Structureeactivity relationship studies revealed that a RID-F analog (RID-F-S*4,
compound 25) is the smallest derivative compound capable of inhibiting proteasome activity, with a
potency similar to that of RID-F. Kinetic analyses of the inhibition mode and competition experiments
involving biotin-belactosin A (a proteasome inhibitor) binding indicated that the RID-F derivatives
interact with the protease subunits in a different manner. Culturing of human cells with these com-
pounds resulted in accumulation of ubiquitinated proteins and induction of apoptosis. Thus, the RID-F
derivatives may be useful lead chemicals for the generation of a new class of proteasome inhibitors.
Ó 2013 Elsevier Masson SAS. All rights reserved.
Received 19 August 2013
Received in revised form
2 November 2013
Accepted 6 November 2013
Available online 16 November 2013
Keywords:
Proteasome inhibitors
Tamoxifen derivatives
Structureeactivity relationship (SAR)
Docking studies
1. Introduction
Several proteasome inhibitors have been proposed as anticancer
drugs [1,4e8]. One of these inhibitors, the peptide boronate bor-
Eukaryotic cells have two different pathways for protein degra-
dation: the lysosomal pathway and the ubiquitin-proteasome
pathway. The lysosomal pathway breaks down endogenous and
endocytosed exogenous proteins in a relatively nonspecific manner
to provide amino acids as building materials for protein synthesis. In
contrast, the ubiquitin-proteasome system is the major machinery
for regulated proteolysis of endogenous proteins. Protein degrada-
tion by the ubiquitin-proteasome system is initiated by the labeling
of targeted proteins with polyubiquitin chains in an intra- or
extracellular signal-dependent manner [1]. Degradation of ubiq-
uitinated proteins by the 26S proteasome plays a pivotal role in the
regulation of a number of cellular processes, such as cell cycle pro-
gression, cell growth, proliferation, differentiation, apoptosis, gene
transcription, and signal transduction [2]. Aberrant degradation of
key regulatory proteins by the proteasome perturbs these processes,
causing uncontrolled cell cycle progression and a decrease in cell
death, both of which are hallmarks of tumorigenesis [3].
tezomib, has been approved for the clinical treatment of multiple
myeloma and mantle cell lymphoma [2]. The toxic boronate phar-
macophore, however, causes severe side effects [9]. More recently,
the tetrapeptide epoxyketone carfilzomib, which is an epoxomicin
analog, has also been approved for the treatment of multiple
myeloma [10e12]. In addition to being targeted in the treatment of
various bloodeborne tumors, proteasome inhibition has been
suggested for the treatment of solid tumors [4], parasites, inflam-
mation, immune diseases, and muscular dystrophies [13], encour-
aging development of new types of proteasome inhibitors with
enhanced efficacy and fewer side effects [5].
The 26S proteasome is a large protein complex of w2.5 MDa that
consists of two subcomplexes with different functions: the 19S
regulatory complex and the 20S catalytic core. The 20S catalytic core
is a barrel-shaped protein composed of seven
a subunits (a 1e7) and
seven subunits ( 1e7). The 19S regulatorycomplex also consists of
b
b
multiple subunits and caps the 20S barrel at one or both ends. The
19S regulatory cap recognizes, unfolds, and translocates poly-
ubiquitinated substrates into the 20S catalytic core, where the
* Corresponding author. Tel.: þ81 749 64 8114.
** Corresponding author. Tel.: þ81 3 3260 4271.
E-mail addresses: m_hasegawa@nagahama-i-bio.ac.jp (M. Hasegawa), shiina@rs.
kagu.tus.ac.jp (I. Shiina).
substrates are degraded. In the 20S proteasome core, the
5 subunits act as proteases, with caspase-(peptidylglutamyl pep-
tide hydrolase, PGPH), trypsin-, and chymotrypsin-like activities,
b1, b2, and
b
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.11.009

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
291
respectively [1,2]. All of these hydrolytic activities are manifested
through a threonine residue at the active site (Thr1).
see also Table 1), a TAM derivative, has been shown to induce
apoptosis through the same pathway but with higher potency
than TAM [30].
Most of the dominant proteasome inhibitors, including borte-
zomib and carfilzomib, are short peptides that mimic substrates.
The pharmacophores bound to the C-terminus of the peptide
framework bind covalently to Thr1 in the 20S catalytic core active
site. However, peptide bonds are easily degraded by endogenous
proteases, and the reactive pharmacophores are susceptible to
nucleophilic attack and are thus short-lived in vivo. Therefore,
nonpeptide and noncovalent synthetic proteasome inhibitors
would be valuable. More recently, a number of investigations have
focused on inhibitors that are peptidic but noncovalent as part of
efforts to overcome the drawbacks associated with covalent in-
hibitors [14]. These compounds include ritonavir [15], amino-
To find new noncovalent and nonpeptidic proteasome inhibitors,
in this study we surveyed proteasome inhibition using a series of
TAM derivatives. We found that RID-A, -B, -D, and -F inhibit the
function of the 20S proteasome catalytic core in vitro. As RID-F was
the most potent inhibitor of the three different enzymatic activities
of the proteasome, we examined the structureeactivity relation-
ships of RID-F analogs with the hope of generating proteasome in-
hibitors that can be used to treat a wider range of diseases with
minimal side effects.
2. Results and discussion
benzylstatine [16], 3,4,5-trimethoxy-L-phenylalanine derivatives
[17], 5-methoxy-1-indanone dipeptide benzamides [18], lip-
opeptides [19], N- and C-capped dipeptides derived from S-homo-
phenylalanine [20,21], and TMC-95A [22,23] and its linear mimetic
derivatives [24,25]. However, only a few inhibitors with both non-
covalent and nonpeptidic characteristics have been reported [26].
Tamoxifen (TAM) binds to the estrogen receptor (ER) in place of
2.1. Synthesis of RID-F, RID-H, and other RID-F derivatives (RID-F-
S*X)
Synthesis of RIDs A-H (compounds 1e8) was conducted according
to methods described in our previous reports and patents [31e35].
The synthetic pathways for producing new compounds 6 and 8e29
are depicted inSchemes 1e11. Asanexample, Scheme 1 illustrates the
transformationofthephenolmoietiesof1,1-bis(4-hydroxyphenyl)-2-
phenylbut-1-ene into the corresponding aminoethyl ethers of RID-F
(compound 6) and RID-H (compound 8) in dimethylformamide
(DMF). 1,1-Bis(4-hydroxyphenyl)-2-phenylbut-1-ene was easily syn-
thesized using a three-component coupling reaction involving
the endogenous growth hormone, 17b-estradiol, suppressing
proliferation of ER-positive breast cancer cells and inducing
apoptosis [27]. However, it has been reported that TAM provokes
apoptosis through an ER-independent pathway [28,29]. Indeed,
TAM induces apoptosis of ER-negative cells through perturbation
of the mitochondrial membrane potential [30]. Ridaifen-B (RID-B,
Table 1
Inhibition of human 20S proteasome activity by ridaifens (RIDs).
Compound
number
IC50 (m
M)a
CT-L
T-L
PGPH
1
2
RID-A
RID-B
R ¼ CH2CH2N(CH3)2
3.36 ꢀ 0.86
6.56 ꢀ 0.14
>10
>10
2.99 ꢀ 0.41
6.37 ꢀ 0.28
R ¼
3
RID-C
R ¼
>10
>10
7.50 ꢀ 0.18
4
5
6
7
RID-D
RID-E
RID-F
RID-G
R ¼
7.19 ꢀ 0.28
>10
>10
7.26 ꢀ 0.27
>10
R ¼ CH2CH2N(C2H5)2
R ¼
>10
0.64 ꢀ 0.14
>10
0.34 ꢀ 0.12
NT
0.43 ꢀ 0.08
NT
R ¼ CH2CH2CH2N(CH3)2
8
RID-H
R ¼
>10
NT
NT
a
IC50 values denote concentrations of the compounds required for 50% inhibition of the activities (see “Experimental section”). Values are means of three experiments. CT-L,
chymotrypsin-like activity; T-L, trypsin-like activity; PGPH, peptidylglutamyl peptide hydrolase activity. NT, not tested.

292
M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
Scheme 1. Synthesis of RID-F (6) and RID-H (8) from 1,1-bis(4-hydroxyphenyl)-2-phenylbut-1-ene.
aromatic aldehyde, cinnamyltrimethylsilane, and anisole in the
presence of HfCl4 [31e33].
into the corresponding aminoethyl ethers with 85% yield by O-
alkylation with N-(2-chloroethyl)hexahydro-1H-azepine HCl, and
successive alkylation of the carbonyl group in RID-F-S*5 (compound
23) using methyl Grignard reagent was carried out in tetrahydro-
furan (THF) to produce the 1,1-diphenylethanol derivative RID-F-S*6
(compound 28), with 88% yield. Finally, treatment of tertiary alcohol
28 with p-TsOH was attempted in benzene, and the facile
RID-F-S*3 (compound 24), 1,1-bis{4-[2-(azepan-1-yl)ethoxy]
phenyl}ethene, was synthesized from 4,40-dihydroxybenzophenone
(4,40-DHBP) via a chemical approach involving O-alkylation, C1
segment installation, and acid-mediated dehydration, as shown in
Scheme 2. First, the phenol moieties of 4,40-DHBP were converted
Scheme 2. Synthesis of RID-F-S*5 (23), RID-F-S*6 (28), and RID-F-S*3 (24) from 4,40-dihydroxybenzophenone (4,40-DHBP).

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
293
Scheme 3. Synthesis of RID-F-S*1 (13), RID-F-S*2 (22), and RID-F-S*4 (25) from 1,1-bis(4-hydroxyphenyl)-2-phenylethene, 1,1-bis(4-hydroxyphenyl)but-1-ene, and bis(4-hy-
droxyphenyl)methane.
dehydration process resulted in successful synthesis of the desired
molecule 24, with 84% yield.
All RID physical properties (Mp, IR, 1H and 13C NMR, and HR-MS)
are shown in the Supporting information.
RID-F-S*13 (compound 9), 1,1-bis{4-[2-(azepan-1-yl)ethoxy]
phenyl}-2-cyclohexyl-2-phenylethene, was synthesized using the
Mukaiyama reductive coupling reaction [36,37] depicted in Scheme
3. First, cyclohexyl phenyl ketone was treated with an excess of
4,40-DHBP in the presence of the low-valent titanium species
generated from titanium(IV) chloride with zinc powder in THF to
afford 1,1-bis(4-hydroxyphenyl)-2-cyclohexyl-2-phenylethene, the
desired olefin, with 65% yield. Next, the phenol moieties of the
cross-coupling product were transformed into the corresponding
aminoethyl ethers by O-alkylation in DMF, with 83% yield. Thus, an
efficient method for the preparation of compound 9 was estab-
lished through two steps from commercially available 4,40-DHBP.
RID-F-S*17 (compound 17) and RID-F-S*24 (compound 16) were
also synthesized from 4,40-DHBP with 6-undecanone or 7-
tridecanone using the Mukaiyama reductive coupling reaction.
Other derivatives, RID-F-S*1 (compound 13), RID-F-S*2 (com-
pound 22), RID-F-S*4 (compound 25), RID-F-S*5 (compound 23),
RID-F-S*9 (compound 21), RID-F-S*10 (compound 20), RID-F-S*11
(compound 12), RID-F-S*12 (compound 11), RID-F-S*14 (com-
pound 10), RID-F-S*15 (compound 15), RID-F-S*16 (compound 14),
RID-F-S*22 (compound 19), RID-F-S*23 (compound 18), RID-F-
S*101 (compound 26), RID-F-S*102 (compound 27), and RID-F-
S*103 (compound 29) were synthesized from the corresponding
commercially available bisphenols.
2.2. Inhibition of proteasome activities by ridaifens
To identify new chemical moieties that function as proteasome
inhibitors, we screened our in-house chemical libraries using an
in vitro 20S proteasome inhibition assay. The potency of com-
pounds was evaluated based on inhibition of the human 20S
chymotrypsin-like (CT-L), trypsin-like (T-L), and PGPH activities of
the 20S proteasome catalytic core. The IC50, indicating the con-
centration required for 50% inhibition of the respective enzymatic
activity (see Experimental section), was determined for each
compound. The aldehyde proteasome inhibitor MG132 (Z-LLL-H)
was used as a standard and had IC50 values of 0.011, 2.1, and 0.12 mM
against CT-L, T-L, and PGPH activities, respectively. Of the 8 TAM
derivatives tested (RID-AeRID-H; Table 1), RID-A, -B, and -D
showed significant inhibition of CT-L and PGPH activities, but they
did not significantly inhibit T-L activity. TAM did not inhibit any of
the three enzymatic activities of the proteasome (Supplementary
Table S1). RID-F, which has two homopiperidine moieties at the R
positions, inhibited all three activities of the proteasome and was
the most potent of the ridaifen compounds examined. The ridaifens
did not inhibit calpain or cathepsin at concentrations of >10 mM
(data not shown), which indicated that the ridaifens are highly
specific inhibitors of the protease activities of the proteasome.

294
M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
Scheme 4. Synthesis of RID-F-S*9 (21), RID-F-S*10 (20), and RID-F-S*11 (12) from 1,1-bis(4-hydroxyphenyl)-2-methylprop-1-ene, 1,1-bis(4-hydroxyphenyl)-2-ethylbut-1-ene, and
1,1-bis(4-hydroxyphenyl)-2-phenylprop-1-ene.
2.3. Inhibition of proteasome activities by RID-F derivatives
using Pocket-Finder, which is based on the LIGSITE algorithm [39].
Compounds with side structures smaller than (RID-F-S*16 (14) and
RID-F-S*15 (15)) or larger than vinyl benzene (RID-F-S*13 (9) and
RID-F-S*14 (10)) demonstrated low inhibition potency. These data
suggest that vinyl benzene is an optimally sized side structure for
inhibiting proteasomal protease activity.
Next, the side structure at the X position was substituted with
aliphatic linear chains of varying lengths. Table 3 summarizes the
structures and proteasome inhibitory activities of this series of
compounds. RID-F-S*24 (16), RID-F-S*17 (17), and RID-F-S*23 (18)
have long hydrocarbon chains; the volumes of their side structures
were calculated to be 235, 201, and 167 Å3, respectively. These
compounds demonstrated no inhibition of proteasome activity,
indicating that hydrocarbon chains of more than eight carbon
atoms in length interfere with the inhibitory activity of RID-F
derivatives. RID-F-S*10 (20), RID-F-S*9 (21), RID-F-S*2 (22), RID-
F-S*5 (23), and RID-F-S*3 (24), which have smaller and nonaro-
matic side structures, showed low inhibition potency.
Because RID-F blocked proteasomal protease activities at con-
centrations in the submicromolar range, we examined the struc-
tureeactivity relationships of various RID-F derivatives. RID-F has
an sp2 carbon atom at the center position (X in Table 2). The side
structure at the X position was substituted with different aromatic
ring(s). Table 2 summarizes the structures and inhibitory activities
of these RID-F derivatives. A comparison of inhibition potency
(IC50) against CT-L activity indicated that RID-F (6), RID-F-S*11 (12),
and RID-F-S*1 (13) were the most potent compounds. It should be
noted that the volume of the vinyl benzene side structure and the
inhibition potency against CT-L, T-L, and PGPH activities were
similar for all three compounds.
To examine the contribution of side structures of the RID-F
derivatives to the inhibition potency, we compared the volumes
of the derivatives with the pocket volume of the proteasome active
site. Side-structure models were built with energy minimization
and molecular dynamics using Chem3D software (PerkinElmer
Inc.). The solvent-excluded volumes of the side structures of RID-F
(6), RID-F-S*11 (12), and RID-F-S*1 (13) calculated using Connolly’s
program [38] in Chem3D software were 119, 102, and 84 Å3,
respectively. These values were close to the pocket volume (117 Å3)
To delineate the minimal structure necessary for proteasome
inhibition, we examined a series of RID derivatives that had trun-
cated side structures (Table 4). The smallest symmetric compound,
RID-F-S*4 (25), demonstrated the highest inhibition potency,
whereas, as seen with RID-F-S*110 (30) (Table 5) [40], removal of
one of the two homopiperidine rings resulted in a critical loss of
of the PGPH active site of the yeast 20S proteasome b1 calculated

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
295
Scheme 5. Synthesis of RID-F-S*12 (11), RID-F-S*14 (10), and RID-F-S*15 (15) from 1,1-bis(4-hydroxyphenyl)-2-phenylpent-1-ene, 1,1-bis(4-hydroxyphenyl)-2,2-diphenylethene,
and 2,2-bis(4-hydroxyphenyl)methylenecyclopentane.
inhibition potency. Therefore, the structure of derivative RID-F-S*4
(25) appears to be the minimal structure possible for proteasome
inhibition activity, and the presence of two homopiperidine rings
appears to be crucial for inhibition.
docking simulations of ridaifens with the yeast counterpart of
mammalian PGPH support this inference.
The chemical structures of the RID-F derivatives suggest that
covalent binding to the proteasome is very unlikely. Indeed,
labeling of proteasome subunits with a biotin-conjugated RID-F
derivative failed to yield biotin-labeled proteins (data not shown). It
has been shown that belactosin A preferentially inhibits both PGPH
2.4. Mode of RID-F derivative proteasome inhibition
(b1 subunit) and CT-L (b5 subunit) by forming covalent linkages to
Kinetic studies revealed that proteasome CT-L activity was
inhibited by RID-F (6) in a noncompetitive manner, with a Ki of
the Og atom of the active site Thr1 [41]. Subunits covalently
modified by biotin-labeled belactosin can be identified by Western
blotting. Kinetic analyses showed that the RID-F derivatives
inhibited PGPH activity in a competitive manner, suggesting direct
0.58 ꢀ 0.14
m
M, whereas PGPH activity was inhibited in a
competitive manner, with a Ki of 0.34 ꢀ 0.22
mM (Fig. 1). The other
RID-F derivatives exhibited similar inhibition of proteasome CT-L
interaction of the derivatives with the active site of the
(Fig. 1B).
b1 subunit
and PGPH activities; RID-F-S*14 (10) and RID-F-S*1 (13) inhibited
CT-L activity noncompetitively, with Ki values of 1.10 ꢀ 0.22
0.87 ꢀ 0.32 M, respectively, whereas they inhibited PGPH activity
competitively, with Ki values of 1.02 ꢀ 0.15 M and 0.67 ꢀ 0.29 M,
m
M and
We then examined whether RID-F derivatives could impede
covalent modification by biotin-labeled belactosin. In agreement
with previous results [41], we found that biotin-belactosin A pri-
m
m
m
respectively (data not shown). It has been shown that substrates of
PGPH activity inhibit CT-L activity and that this inhibition is caused
by binding of PGPH substrates to a noncatalytic CT-L site(s) rather
than through binding to an active site [11]. Taken together, from
these data we inferred that RID-F and its derivatives bind to both
the active PGPH site and a noncatalytic CT-L site(s), resulting in
inhibition of CT-L activity. As described below, the results of
marily bound
(Fig. 2). RID-F (6) and RID-F-S*4 (25) impeded binding of biotin-
belactosin to 1 (PGPH activity) in a dose-dependent manner
b1 and b5 but also bound b2 with very low efficiency
b
(Fig. 2A and B). Both compounds inhibited CT-L activity in a
noncompetitive manner (Fig. 1A), and therefore we expected that
they do not reduce binding of belactosin to
b5 (CT-L activity).
Contrary to this expectation, the compounds inhibited binding to

296
M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
b
5, but the efficiency with which they inhibited binding to
b
5 was
the exception of RID-F-S*22 (19), which significantly inhibited CT-L
activity but was inexplicably noncytotoxic. Most of the cytotoxic
compounds showed similar toxicity to both cell types, although
some compounds (RID-F-S*12 (11), RID-F-S*11 (12) and RID-F-S*6
(28)) were more toxic to HL60 than HEK293 cells, and two com-
pounds (RID-F-S*5 (23) and RID-F-S*103 (29)) were more toxic to
HEK293 cells.
much lower than that of
b1; at a concentration of 10 mM, the
compounds almost completely inhibited binding of belactosin A to
1, whereas binding to 5 was still detectable at 100 M. These
results suggest that the region in which the RID-F compounds bind
5 is proximal to the CT-L active site and partially overlaps the
binding region of belactosin A. The concentration (10 M) of the
RID-F derivatives required to inhibit binding of belactosin to
b
b
m
b
m
b
1
Is the cytotoxicity of RID-F derivatives attributable to their
inhibition of proteasome activity? To address this question, we
selected compounds that inhibited at least two of the three pro-
teasome protease activities. The IC50 values of these compounds
were plotted against the CyT50 values determined using HEK293
cells to estimate the correlation between proteasome inhibition
and cytotoxicity (Fig. 3A). The correlation coefficients were 0.51 for
CT-L activity and 0.48 for PGPH activity (Fig. 3B) but only 0.02 for
T-L activity (Fig. 3C). Thus, it is very likely that the cytotoxicity of
the RID-F derivatives is caused by their inhibition of proteasome
CT-L and PGPH activities, whereas the inhibition of T-L activity is
not associated with cytotoxicity. It is not known whether inhibition
of either CT-L or PGPH activity alone is sufficient to cause cyto-
toxicity or if inhibition of both activities is required. Genetic studies
have suggested that CT-L activity is essential for proteasome
function, because mutational loss of CT-L activity causes a signifi-
cant reduction in the degradation of proteasomal substrates
[42,43]. Taken together, these data suggest that the inhibition of CT-
L activity by RID-F derivatives may be linked to cell death. The CyT50
values of the RID-F derivatives were generally one order of
magnitude higher than the corresponding IC50 values; restricted
diffusion of the derivatives through the cell membrane may be
responsible for this difference.
(PGPH activity) was much higher than the Ki (submicromolar
levels) determined from kinetic analyses (Fig. 1A). There are several
potential explanations for this difference: (i) the affinity of biotin-
labeled belactosin A was higher than that of the substrate used
for kinetic analyses, or (ii) the reversible inhibitor RID-F derivatives
are eventually replaced by the covalent inhibitor, belactosin.
MG132, which was used as a positive control for proteasomal
protease inhibition, prevented biotin-belactosin A from forming
covalent linkages to all subunits. RID-F-S*110 (30) showed little
effect on belactosin binding (Fig. 2C), which was consistent with
the result indicating that this compound has minimal inhibitory
potency against proteasomal activity (Table 5). The possible
mechanism by which RID-F derivatives prevent the binding of
belactosin to the b2 subunit (T-L activity) remains to be studied.
2.5. Cytotoxicity
We also evaluated the effect of the RID-F derivatives on prolif-
eration of two human cell lines, human embryonic kidney 293 cells
(HEK293, ER-negative) and HL-60 cells (ER-positive). Cells were
treated with different concentrations of each compound for 48 h,
after which the number of live cells was determined using a 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
assay. The CyT50, defined as the concentration required for 50%
inhibition of cell proliferation (see Experimental section), was
determined for each compound, and the results are summarized in
Tables 2e5. The compounds with poor inhibitory potency
We next investigated the ability of the RID-F derivatives to
inhibit proteasome function in cultured cells. HeLa cells were
incubated with RID-F (6), RID-F-S*4 (25), or RID-F-S*110 (30), and
the accumulation of ubiquitinated proteins was examined by
Western blotting. A significant accumulation of multiple bands of
high-molecular-weight ubiquitinated proteins was observed in
(IC50 > 10 mM) against CT-L activity in vitro were not cytotoxic, with
Scheme 6. Synthesis of RID-F-S*16 (14) and RID-F-S*22 (19) from 2,2-bis(4-hydroxyphenyl)methylenecyclohexane and 1,1-bis(4-hydroxyphenyl)-2-propylpent-1-ene.

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
297
Scheme 7. Synthesis of RID-F-S*23 (18) and RID-F-S*101 (26) from 1,1-bis(4-hydroxyphenyl)-2-butylhex-1-ene and 1,1-bis(4-hydroxyphenyl)ethane.
cells treated with RID-F (6) and RID-F-S*4 (25) (Fig. 4, lanes 3 and
4). The accumulation of such proteins was minimal in cells treated
with RID-F-S*110 (30), a compound that showed very low inhibi-
tion potency (Table 5). Accumulation of ubiquitinated proteins was
also observed in lysates of cells treated with the known proteasome
inhibitor, MG132 (Fig. 4, lane 2). These data indicate that RID-F
derivatives inhibit proteasome activities in cells.
Abnormal accumulation of proteins resulting from the inhi-
bition of proteasome activity has antiproliferative effects on cells,
including induction of apoptosis. Therefore, we examined
Scheme 8. Synthesis of RID-F-S*102 (27) and RID-F-S*103 (29) from 2,2-bis(4-hydroxyphenyl)propane and 1,1-bis(4-hydroxyphenyl)cyclohexane.

298
M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
Scheme 9. Synthesis of RID-F-S*13 (9) from 4,40-dihydroxybenzophenone (4,40-DHBP) and cyclohexyl phenyl ketone.
Scheme 10. Synthesis of RID-F-S*17 (17) from 4,40-dihydroxybenzophenone (4,40-DHBP) and 6-undecanone.

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
299
Scheme 11. Synthesis of RID-F-S*24 (16) from 4,40-dihydroxybenzophenone (4,40-DHBP) and 7-tridecanone.
Table 2
Inhibition of human 20S proteasome activity by RID-F derivatives and their cytotoxic effect on human cells (HEK293 and HL-60).
Compound number
IC50
(
m
M)a
HEK293
CyT50
HL-60
(m
M)b
CT-L
T-L
PGPH
9
RID-F-S*13
RID-F-S*14
RID-F-S*12
RID-F
RID-F-S*11
RID-F-S*1
X ¼ C(c-Hex)Ph
X ¼ CPh2
>10
>10
>10
0.28 ꢀ 0.04
0.34 ꢀ 0.12
0.36 ꢀ 0.17
0.69 ꢀ 0.54
0.35 ꢀ 0.02
0.37 ꢀ 0.06
1.05 ꢀ 0.46
0.43 ꢀ 0.08
0.87 ꢀ 0.04
0.37 ꢀ 0.19
>30
>30
>30
4.30 ꢀ 0.55
3.42 ꢀ 0.33
6.87 ꢀ 1.47
9.73 ꢀ 0.62
10
11
6
12
13
1.18 ꢀ 0.07
1.37 ꢀ 0.27
0.64 ꢀ 0.14
0.90 ꢀ 0.10
0.58 ꢀ 0.05
23.2 ꢀ 1.1
26.8 ꢀ 1.0
4.38 ꢀ 0.79
18.9 ꢀ 1.2
6.06 ꢀ 0.45
X ¼ C(n-C3H7)Ph
X ¼ C(C2H5)Ph
X ¼ C(CH3)Ph
X ¼ CHPh
14
RID-F-S*16
RID-F-S*15
X ¼
X ¼
2.19 ꢀ 0.25
2.70 ꢀ 0.10
2.18 ꢀ 2.03
1.11 ꢀ 0.08
0.89 ꢀ 0.09
1.67 ꢀ 0.15
12.5 ꢀ 0.3
26.2 ꢀ 1.9
11.7 ꢀ 0.4
20.1 ꢀ 0.6
15
a
IC50 values denote concentrations of the compounds required for 50% inhibition of the activities (see “Experimental section”). CT-L, chymotrypsin-like activity; T-L,
trypsin-like activity; PGPH, peptidylglutamyl peptide hydrolase activity. NT, not tested.
b
CyT50 values denote concentration of the compounds required for 50% inhibition of the cell proliferation (see “Experimental section”). Values are means of triplicate.

300
M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
Table 3
Inhibition of 20S proteasome activity by RID-F derivatives substituted with aliphatic chains at the center sp2 carbon (X) and their cytotoxic effect on human cells.
Compound number
IC50
(
m
M)a
HEK293
CyT50
HL-60
(m
M)b
CT-L
T-L
PGPH
16
17
18
19
20
21
22
23
24
RID-F-S*24
RID-F-S*17
RID-F-S*23
RID-F-S*22
RID-F-S*10
RID-F-S*9
RID-F-S*2
RID-F-S*5
RID-F-S*3
X ¼ C(n-C6H13
X ¼ C(n-C5H11
)
)
>10
>10
>10
3.38 ꢀ 0.42
1.57 ꢀ 0.70
1.66 ꢀ 0.12
1.46 ꢀ 0.14
1.65 ꢀ 0.21
1.04 ꢀ 0.22
>10
>10
>10
>10
>10
>10
>10
2.51 ꢀ 0.44
0.84 ꢀ 0.12
1.20 ꢀ 0.19
1.03 ꢀ 0.10
1.56 ꢀ 0.08
0.91 ꢀ 0.04
>30
>30
>30
2
14.5 ꢀ 3.9
2
X ¼ C(n-C4H9)2
X ¼ C(n-C3H7)2
X ¼ C(C2H5)2
X ¼ C(CH3)2
X ¼ CH(C2H5)
X ¼ O
>30
25.3 ꢀ 0.6
>30
>30
0.96 ꢀ 0.18
6.02 ꢀ 0.14
11.1 ꢀ 0.5
26.7 ꢀ 0.8
13.9 ꢀ 1.4
9.80 ꢀ 5.53
8.87 ꢀ 1.13
10.7 ꢀ 1.6
22.6 ꢀ 0.3
>30
>10
1.61 ꢀ 0.47
8.59 ꢀ 1.75
0.88 ꢀ 0.45
X ¼ CH2
10.7 ꢀ 0.5
a
IC50 values denote concentrations of the compounds required for 50% inhibition of the activities (see “Experimental section”). CT-L, chymotrypsin-like activity; T-L,
trypsin-like activity; PGPH, peptidylglutamyl peptide hydrolase activity. NT, not tested.
b
CyT50 values denote concentration of the compounds required for 50% inhibition of the cell proliferation (see “Experimental section”). Values are means of triplicate.
whether the observed cytotoxicity of RID-F derivatives was due to
apoptosis. MG132, a representative proteasome inhibitor that
binds all three subunits, was used as a control compound.
Cleavage of poly(ADP-ribose) polymerase (PARP) is one of the
hallmarks of apoptosis. RID-F (6) (Fig. 5A) and RID-F-S*4 (25)
(Fig. 5B) caused PARP cleavage in a dose-dependent manner, but
PARP cleavage was almost undetectable in cells treated with RID-
F-S*110 (30) (Fig. 5C), which was consistent with the very low
inhibition of proteasome activities demonstrated by this com-
pound (Table 5). RID-F-induced apoptosis was confirmed by
cleavage of caspase 3 (Supplementary Fig. S1) and an increase in
the proportion of cells in the sub-G1 fraction as determined by
flow cytometry (Supplementary Fig. S2). These results indicate
that the cytotoxicity of RID-F derivatives can be attributed at least
in part to apoptosis.
2.6. Three-dimensional modeling of RID-F derivatives bound to the
yeast proteasome subunit PRE3
How do RID-F derivatives bind the proteasome to exhibit their
inhibitory effect? Docking studies were conducted to attempt to
answer this question. To date, no reports of the experimental
determination of the 3D structure of the human 20S proteasome
have been published. However, several 3D structures of the yeast
proteasome have been reported. The molecular structure of the
ligand-binding pockets of the yeast and mammalian proteasomes is
Table 4
Inhibition of 20S proteasome activity by RID-F derivatives with aliphatic side structures at the center sp3 carbon (X) and their cytotoxic effect on human cells.
Compound number
IC50
(m
M)a
HEK293
CyT50
HL-60
(m
M)b
CT-L
T-L
PGPH
25
26
27
28
RID-F-S*4
X ¼ CH2
0.67 ꢀ 0.04
0.80 ꢀ 0.02
0.75 ꢀ 0.01
1.66 ꢀ 0.11
0.99 ꢀ 0.21
>10
0.63 ꢀ 0.15
0.79 ꢀ 0.02
0.77 ꢀ 0.01
1.35 ꢀ 0.05
10.9 ꢀ 0.6
8.12 ꢀ 0.14
7.61 ꢀ 0.05
22.7 ꢀ 6.3
14.5 ꢀ 0.3
21.4 ꢀ 0.6
17.6 ꢀ 1.8
7.06 ꢀ 0.48
RID-F-S*101
RID-F-S*102
RID-F-S*6
X ¼ CH(CH3)
X ¼ C(CH3)2
X ¼ C(CH3)OH
1.87 ꢀ 0.83
2.95 ꢀ 0.40
29
RID-F-S*103
X ¼
1.94 ꢀ 0.09
0.20 ꢀ 0.20
1.08 ꢀ 0.16
17.5 ꢀ 0.3
>30
a
IC50 values denote concentrations of the compounds required for 50% inhibition of the activities (see “Experimental section”). CT-L, chymotrypsin-like activity; T-L,
trypsin-like activity; PGPH, peptidylglutamyl peptide hydrolase activity. NT, not tested.
b
CyT50 values denote concentration of the compounds required for 50% inhibition of the cell proliferation (see “Experimental section”). Values are means of triplicate.

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
301
Table 5
Inhibition of 20S proteasome activity by a RID-F derivative missing one of the two homopiperidine rings and its cytotoxic effect on human cells.
Compound number
IC50
(m
M)a
CyT50
(
m
M)b HEK293
HL-60
CT-L
T-L
PGPH
30
RID-F-S*110
>10
>10
>10
27.0 ꢀ 3.7
>30
a
IC50 values denote concentrations of the compounds required for 50% inhibition of the activities (see “Experimental section”). CT-L, chymotrypsin-like activity; T-L,
trypsin-like activity; PGPH, peptidylglutamyl peptide hydrolase activity. NT, not tested.
b
CyT50 values denote concentration of the compounds required for 50% inhibition of the cell proliferation (see “Experimental section”). Values are means of triplicate.
highly conserved [44,45]. Moreover, RID-F also inhibited yeast
proteasome CT-L activity, with an IC50 of 1.8 M, which was similar
to that (IC50 0.65 M) determined for the human 20S proteasome
(Table 1). These results suggest that the binding modes of the RID-F
derivatives to the yeast and human proteasomes are similar.
Therefore, we used the yeast proteasome structure for docking
simulations of RID-F derivatives.
20, 22, and 23 (but not compounds 19, 21, 24 and 25 (RID-F-S*4))
have a similar binding mode. Fig. 6 (A and B) shows the highest-
ranked binding mode of docked RID-F (6), illustrating molecular
interactions between RID-F and the PRE3 catalytic site. One of the
important points in the interaction between RID-F and the pro-
teasome appears to be the contact area of the S1 pocket around
Thr1. The vinyl benzene group in the side structure of RID-F is in
m
m
We used structure data of the yeast 20S proteasome complexed
with fellutamide B (PDB ID: 3d29) for docking simulations. The
PRE3 subunit (chain N) is the yeast counterpart of the human 20S
contact with the S1 pocket and has a CHep interaction with Thr21.
The contacted residues delimit the binding pocket size at 120 Å3, in
which Thr1 is located at the bottom. These docking simulation
results are in agreement with the experimental evidence indicating
that the size of the RID-F derivative side structure is critical for
binding to the catalytic site.
proteasome
is sandwiched between the PUP1 subunit (chain H) and the PRE4
subunit (chain M) in the ring, and the ligand-binding cleft of the
b1 subunit, which exhibits PGPH activity. This subunit
b
PRE3 subunit is formed by these three subunits. Therefore, the
trimer structure of chains H, N, and M was used for docking sim-
ulations. Using Molecular Operating Environment software, version
2010.10 (MOE 2010.10, Chemical Computing Group Inc.), the
energy-minimized trimer structure with hydrogen atoms was
prepared with the default parameters. The possible ligand-binding
site in the trimer structure was detected using the Site Finder
application of MOE 2010.10, with the Connection Distance param-
eter set to 1.9 A. Docking simulations involving the RID-F
derivatives and the trimer structure were then carried out using
ASEDock [46], with the standard procedure.
The best-fitted inhibitor positions near in the active center Thr1
residue suggested that compounds 6 (RID-F), 9, 10, 11, 12, 13, 14, 15,
As the exceptional case, docking simulations involving com-
pounds 19, 21, 24, and 25 (RID-F-S*4) showed one homopiperidine
ring contacting the S1 pocket. Their side structures at the X position
were either too large or too small. RID-F-S*4 (25), which has a high
inhibition potency despite the absence of a side structure, binds in a
different fashion, as shown in Fig. 6. One terminal homopiperidine
moiety in RID-F-S*4 (25) enters into the binding cleft in the S1
pocket, as opposed to the vinyl benzene group of RID-F (6). These
results imply that RID-F derivatives have two modes of binding to
the catalytic sites. Interestingly, the results of the docking simula-
tions involving compounds 19, 21, 24, and 25 (RID-F-S*4) suggest
that they access the active site in a manner different from that of
compound 6 (RID-F), which contacts the active site via the vinyl
ꢀ
B
A
Fig. 1. LineweavereBurk plots of the inhibition of proteasomal CT-L and PGPH activities by RID-F. A, CT-L activity. B, PGPH activity. CT-L and PGPH activities were determined using a
fluorometric assay as described in the Experimental section, varying the concentrations of substrate and RID-F as indicated. A.U. represents arbitrary units of fluorescence.

302
M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
A
B
C
Fig. 4. Accumulation of ubiquitinated proteins induced by RID-F (6), RID-F-S*4 (25),
and RID-F-S*110 (30). HeLa cells were treated with 0.25% DMSO (control), MG132
(10 M), RID-F (30 M), RID-F-S*4 (60 M), or RID-F-S*110 (60 M) for 24 h. Whole cell
-Actin was used as a
Fig. 2. Inhibition by RID-F derivatives of biotin-labeled belactosin A binding to the
proteasome. As described in the Experimental section, human 20S proteasomes were
treated with biotin-belactosin A in the presence of RID-F (6) (A), RID-F-S*4 (25) (B), or
m
m
m
m
b
lysates were immunoblotted with anti-ubiquitin antibody.
loading control.
RID-F-S*110 (30) (C), and biotin-labeled subunits (
b1 for T-L activity, b2 for PGPH ac-
tivity, and 5 for CT-L activity) were then detected. MG132 was used as a control.
b
Fig. 3. Correlation between cytotoxicity (CyT50) to HEK293 cells and inhibitory potency against the CT-L and PGPH activities of the human 20S proteasome. A, CyT50 versus IC50 for
CT-L activity. B, CyT50 versus IC50 for PGPH activity. C, CyT50 versus IC50 for T-L activity. Numbers in the plots correspond to the compound numbers listed in Tables 2e4.

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
303
3. Conclusion
A
B
C
Ridaifen-F (RID-F), a novel tamoxifen derivative, inhibited the
human 20S proteasome. The structureeactivity relationship of a
series of RID-F derivatives revealed the fundamental structures
required for proteasome inhibition. The derivatives inhibited the
proteasome in cells, inducing apoptotic cell death. Based on kinetic
analyses of the inhibition and docking simulation, we propose the
inhibition mode of RID-Fs. Our next trial is underway to develop
highly potent RID derivatives for in vivo use.
4. Experimental section
4.1. 20S proteasome fluorometric substrate assay
CT-L, PGPH, and T-L proteasome activities were determined by
measuring the degradation rates of the fluorometric substrates
succinyl-LLVY-AMC, Z-LLE-AMC, and Boc-LRR-AMC, respectively.
Purified human 20S proteasomes (0.1
50 M (CT-L) or 20 M (PGPH and T-L) fluorometric peptide sub-
strate in the presence of varying concentrations of inhibitory
compounds (0.01e10 M) in 100 L of assay buffer (25 mM HEPES,
mg) were incubated with
m
m
Fig. 5. Apoptosis induced by RID-F derivatives. HeLa cells were incubated with the
compounds at indicated concentrations, after which cleavage of PARP was determined
by Western blotting (see Experimental section). A, RID-F (6); B, RID-F-S*4 (25); C, RID-
F-S*110 (30). MG132, a proteasome inhibitor that induces PARP cleavage, which is a
hallmark of apoptosis, was used as a control. * ¼ nonspecific bands.
m
m
0.5 mM EDTA, 0.03% SDS) at 37 ꢁC. In the T-L activity assay, SDS was
excluded from the assay buffer [47]. Reactions were monitored by
AMC product formation (lex ¼ 380 nm, lem ¼ 460 nm) for 1 h. The
IC50, defined as the compound concentration required for 50% in-
hibition of proteasome activity, was determined for each com-
pound from the respective inhibition curve.
benzene at the X position. As illustrated in Fig. 6 (C and D) for RID-
F-S*4 (25) as a representative, these RID-F derivatives may interact
with the S1 pocket via one of the homopiperidine rings. The side
structures at the X positions in these four compounds were too
large or too small in comparison with the optimal size of vinyl
benzene. Thus, it appears that not only the presence of two
homopiperidine rings but also the relationship between the
homopiperidine rings and the side structures at the X position are
important for manifestation of the proteasome inhibition activity of
the RID-F derivatives, suggesting that further studies of this rela-
tionship are warranted.
4.2. Growth inhibition assays
HEK293 cells (ER-negative)werecultured in Dulbecco’s Modified
Eagle Medium containing 10% fetal bovine serum, 100 units/mL of
penicillin, and 0.1 mg/mL of streptomycin. HL-60 cells (ER-positive)
were cultured in Iscove’s Modified Dulbecco’s Medium containing
10% fetal bovine serum, 100 units/mL of penicillin, and 0.1 mg/mL of
streptomycin. Cells were seeded in duplicate wells in a 96-well plate
Fig. 6. Schematic view of RID-F (6) (panels A and B) or RID-F-S*4 (25) (panels C and D) e proteasome interactions. Images (panels A and C) represent the most likely binding modes.
The binding pocket is shown as a solid surface with carbon atoms colored gray, N atoms colored blue, and O atoms colored red. Panels B and D show skeleton models of the binding
modes.

304
M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
at a density of 5 ꢂ 103 cells/well and were cultured for 48 h in me-
CT-L
CyT50
chymotrypsin-like
half-maximum cytotoxicity concentration
dium alone or in medium containing an RID-F derivative at different
concentrations (ranging from 0.1 to 20
mM). Cell viability and pro-
4,40-DHBP 4,40-dihydroxybenzophenone
liferation were evaluated by quantification of MTT reduction by
mitochondrial dehydrogenases. Formazan dye production was
measured by determining the 560/750 nm absorbance ratio after
HCl/2-propanol extraction according to the manufacturer’s protocol
(Promega). The CyT50, defined as the compound concentration
required for 50% inhibition of cell proliferation, was determined for
each compound from the respective inhibition curve.
DMF
ER
IC50
HRP
Ki
dimethylformamide
estrogen receptor
half-maximum inhibitory concentration
horseradish peroxidase
inhibition constant
MTT
(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide
PARP
PGPH
poly-ADP ribose polymerase
peptidylglutamyl peptide hydrolase
4.3. Binding experiment
p-TsOH p-toluenesulfonic acid
Purified human 20S proteasomes were incubated with varying
concentrations of RID-F derivatives or the representative protea-
some activity inhibitor MG132 for 1 h at 37 ꢁC and then biotin-
RID
SDS
TAM
T-L
ridaifen
sodium dodecyl sulfate
tamoxifen
belactosin A was added at 1
mM. After an additional 1 h of incu-
trypsin-like
bation, proteins were separated by SDS-PAGE and transferred to a
polyvinylidene difluoride (PVDF) membrane (Millipore). The
membrane was blocked with 5% skim milk solution. Biotin-labeled
proteins were visualized using streptavidin-HRP reagent and an
ECL Plus membrane blot analysis detection kit (GE Healthcare).
THF
tetrahydrofuran.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.11.009.
4.4. Western blotting
HeLa cells (4 ꢂ 105 cells/well) were seeded in a 12-well plate and
incubated with varying concentrations of RID-F derivatives for 24 h.
The cells were then lysed with lysis buffer (62.5 mM TriseHCl, 2%
SDS, 10% glycerol, 0.1 mg/mL of phenylmethylsulfonyl fluoride,
References
[1] A. Navon, A. Ciechanover, The 26S proteasome: from basic mechanisms to
drug targeting, J. Biol. Chem. 284 (2009) 33713e33718.
[2] A.G. Eldridge, T. O’Brien, Therapeutic strategies within the ubiquitin protea-
some system, Cell Death Differ. 17 (2010) 4e13.
[3] D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell 144
(2011) 646e674.
[4] M.A. Shahshahan, M.N. Beckley, A.R. Jazirehi, Potential usage of proteasome
inhibitor bortezomib (Velcade, PS-341) in the treatment of metastatic mela-
noma: basic and clinical aspects, Am. J. Cancer Res. 1 (2011) 913e924.
[5] M.A. Gräwert, M. Groll, Exploiting nature’s rich source of proteasome in-
hibitors as starting points in drug development, Chem. Commun. (Camb.) 48
(2012) 1364e1378.
[6] P. Lawasut, D. Chauhan, J. Laubach, C. Hayes, C. Fabre, M. Maglio, C. Mitsiades,
T. Hideshima, K.C. Anderson, P.G. Richardson, New proteasome inhibitors in
myeloma, Curr. Hematol. Malig. Rep. 7 (2012) 258e266.
10
inhibitor cocktail (Nacalai Tesque Inc., Kyoto, Japan)). Equivalent
amounts of protein (30 g) were resolved by SDS-PAGE, transferred
mg/mL of leupeptin, 1 mg/mL of pepstatin A, and a phosphatase
m
and immobilized onto PVDF membranes, and probed with primary
and secondary antibodies. Anti-ubiquitin antibody (1:200 dilution;
Santa Cruz Biotechnology), anti-actin antibody (1:10,000; Santa
Cruz Biotechnology), and HRP-conjugated anti-mouse or anti-
rabbit antibody (1:2000; Santa Cruz Biotechnology) were used.
Immunodetection was performed as described above.
Apoptosis induced by RID-F derivatives was examined with
HeLa cells. Cells (4 ꢂ 105 cells/well) were seeded in a 12-well plate
and incubated with varying concentrations of RID-F derivatives for
24 h. Cleavage of PARP, which is a hallmark of apoptosis, was
detected by Western blotting as described previously [48].
[7] A. Mullard, Next-generation proteasome blockers promise safer cancer ther-
apy, Nat. Med. 18 (2012) 7.
[8] A. Katsnelson, Next-generation proteasome inhibitor approved in multiple
myeloma, Nat. Biotechnol. (2012) 1011e1012.
[9] A.A. Argyriou, G. Iconomou, H.P. Kalofonos, Bortezomib-induced peripheral
neuropathy in multiple myeloma: a comprehensive review of the literature,
Blood 112 (2008) 1593e1599.
[10] L. Meng, R. Mohan, B.H. Kwok, M. Elofsson, N. Sin, C.M. Crews, Epoxomicin, a
potent and selective proteasome inhibitor, exhibits in vivo anti-inflammatory
activity, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 10403e10408.
[11] J. Myung, K.B. Kim, K. Lindsten, N.P. Dantuma, C.M. Crws, Lack of proteasome
active site allostery as revealed by subunit-specific inhibitors, Mol. Cell 7
(2001) 411e420.
Conflict of interest
The authors declare no competing financial interest.
Acknowledgment
[12] D.J. Kuhn, Q. Chen, P.M. Voorhees, J.S. Strader, K.D. Shenk, C.M. Sun, S.D. Demo,
M.K. Bennett, F.W. van Leeuwen, A.A. Chanan-Khan, R.Z. Orlowski, Potent
activity of carfilzomib,
proteasome pathway, against preclinical models of multiple myeloma,
Blood 110 (2007) 3281e3290.
a novel, irreversible inhibitor of the ubiquitin-
The authors thank Mie Tsuchida and Miho Hosoi for technical
assistance and Dr. Tetuo Yoshida of Kyowa Hakko Kirin Co. for
supplying belactosin A. The authors also thank Drs. Takao Yamori
and Reiko Shinkura for helpful discussions. This study was sup-
ported by a Health Labour Sciences Research Grant from the Min-
istry of Health, Labour and Welfare, Japan to I.S. and a Grant-in-Aid
for Scientific Research on Priority Area “Cancer” from the Ministry
of Education, Culture, Sports, Science and Technology, Japan to T.M.
M.S. received support from the Platform for Drug Discovery,
Informatics, and Structural Life Science of the Ministry of Educa-
tion, Culture, Sports, Science and Technology, Japan.
[13] D. Chen, M. Frezza, S. Schmitt, J. Kanwar, Q.P. Dou, Bortezomib as the first
proteasome inhibitor anticancer drug: current status and future perspectives,
Curr. Cancer Drug Targets 11 (2011) 239e253.
[14] E. Genin, M. Reboud-Ravaux, J. Vidal, Proteasome inhibitors: recent advances
and new perspectives in medicinal chemistry, Curr. Top Med. Chem. 10 (2010)
232e256.
[15] G. Schmidtke, H.G. Holzhutter, M. Bogyo, N. Kairies, M. Groll, R. de Giuli,
S. Emch, M. Groettrup, How an inhibitor of the HIV-I protease modulates
proteasome activity, J. Biol. Chem. 274 (1999) 35734e35740.
[16] P. Furet, P. Imbach, P. Fuerst, M. Lang, M. Noorani, J. Zimmermann, C. Garcia-
Echeverria, Structure-based optimisation of 2-aminobenzylstatine de-
rivatives: potent and selective inhibitors of the chymotrypsin-like activity of
the human 20S proteasome, Bioorg. Med. Chem. Lett. 12 (2002) 1331e1334.
[17] P. Furet, P. Imbach, M. Noorani, J. Koeppler, K. Laumen, M. Lang, V. Guagnano,
P. Fuerst, J. Roesel, J. Zimmermann, C. Garcia-Echeverria, Entry into a new
class of potent proteasome inhibitors having high antiproliferative activity by
structure-based design, J. Med. Chem. 47 (2004) 4810e4813.
Abbreviations
AMC
7-amino-4-methylcoumarin

M. Hasegawa et al. / European Journal of Medicinal Chemistry 71 (2014) 290e305
305
[18] R.T. Lum, M.G. Nelson, A. Joly, A.G. Horsma, G. Lee, S.M. Meyer, M.M. Wick,
cinnamyltrimethylsilane, and
b
-chlorophenetole, Bioorg. Med. Chem. 15
S.R. Schow, Selective inhibition of the chymotrypsin-like activity of the 20S
proteasome by 5-methoxy-1-indanone dipeptide benzamides, Bioorg. Med.
Chem. Lett. 8 (1998) 209e214.
(2007) 7599e7617.
[33] I. Shiina, Y. Sano, K. Nakata, T. Kikuchi, A. Sasaki, M. Ikekita, Y. Hasome,
Synthesis of the new pseudo-symmetrical tamoxifen derivatives and their
anti-tumor activity, Bioorg. Med. Chem. Lett. 17 (2007) 2421e2424.
[34] I. Shiina, M. Ikekita, T. Matsunaga, Y. Nagahara, Anticancer agent containing
tamoxifen analogue as active ingredient, Jpn. Kokai Tokkyo Koho (11 May
2006). JP2006117648.
[35] I. Shiina, Proteasome inhibitor which contains tamoxifen analogue as an
active ingredient, Jpn. Kokai Tokkyo Koho (24 April 2008). JP2008094836.
[36] T. Mukaiyama, Titanium tetrachloride in organic synthesis, Angew. Chem. Int.
Ed. 16 (1977) 817e826.
[19] N. Basse, D. Papapostolou, M. Pagano, M. Reboud-Ravaux, E. Bernard,
A.S. Felten, R. Vanderesse, Development of lipopeptides for inhibiting 20S
proteasomes, Bioorg. Med. Chem. Lett. 16 (2006) 3277e3281.
[20] C. Blackburn, K.M. Gigstad, P. Hales, K. Garcia, M. Jones, F.J. Bruzzese,
C. Barrett, J.X. Liu, T.A. Soucy, D.S. Sappal, N. Bump, E.J. Olhava, P. Fleming,
L.R. Dick, C. Tsu, M.D. Sintchak, J.L. Blank, Characterization of a new series of
non-covalent proteasome inhibitors with exquisite potency and selectivity for
the 20S b5-subunit, Biochem. J. 430 (2010) 461e476.
[21] G. Lin, T. Chidawanyika, C. Tsu, T. Warrier, J. Vaubourgeix, C. Blackburn,
K. Gigstad, M. Sintchak, L. Dick, C. Nathan, N,C-Capped dipeptides with
selectivity for mycobacterial proteasome over human proteasomes: role of S3
and S1 binding pockets, J. Am. Chem. Soc. 135 (2013) 9968e9971.
[22] Y. Koguchi, J. Kohno, M. Nishio, K. Takahashi, T. Okuda, T. Ohnuki,
S. Komatsubara, TMC-95A, B, C, and D, novel proteasome inhibitors produced
by Apiospora montagnei Sacc. TC 1093. Taxonomy, production, isolation, and
biological activities, J. Antibiot. (Tokyo) 53 (2000) 105e109.
[37] D.D. Yu, B.M. Forman, Simple and efficient production of (Z)-4-
hydroxytamoxifen, a potent estrogen receptor modulator, J. Org. Chem. 68
(2003) 9489e9491.
[38] M.L. Connolly, The molecular surface package, J. Mol. Graphics 11 (1993) 139e141.
[39] M. Hendlich, F. Rippmann, G. Barnickel, LIGSITE: automatic and efficient
detection of potential small molecule-binding sites in proteins, J. Mol. Graph.
Model. 15 (1997) 359e363.
[40] T.D. Penning, N.S. Chandrakumar, B.B. Chen, H.Y. Chen, B.N. Desai, S.W. Djuric,
S.H. Docter, A.F. Gasiecki, R.A. Haack, J.M. Miyashiro, M.A. Russell, S.S. Yu,
D.G. Corley, R.C. Durley, B.F. Kilpatrick, B.L. Parnas, L.J. Askonas, J.K. Gierse,
E.I. Harding, M.K. Highkin, J.F. Kachur, S.H. Kim, G.G. Krivi, D. Villani-Price,
E.Y. Pyla, W.G. Smith, N.S. Ghoreishi-Haack, Structureeactivity relationship
studies on 1-[2-(4-phenylphenoxy)ethyl]pyrrolidine (SC-22716), a potent in-
hibitor of leukotriene A4 (LTA4) hydrolase, J. Med. Chem. 43 (2000) 721e735.
[41] M. Hasegawa, K. Kinoshita, C. Nishimura, U. Matsumura, M. Shionyu, S. Ikeda,
T. Mizukami, Affinity labeling of the proteasome by a belactosin A derived
inhibitor, Bioorg. Med. Chem. Lett. 18 (2008) 5668e5671.
[23] M. Groll, R. Huber, L. Moroder, The persisting challenge of selective and
specific proteasome inhibition, J. Pept. Sci. 15 (2009) 58e66.
[24] N. Basse, S. Piguel, D. Papapostolou, A. Ferrier-Berthelot, N. Richy, M. Pagano,
P. Sarthou, J. Sobczak-Thépot, M. Reboud-Ravaux, J. Vidal, Linear TMC-95-
based proteasome inhibitors, J. Med. Chem. 50 (2007) 2842e2850.
[25] A. Desvergne, E. Genin, X. Maréchal, N. Gallastegui, L. Dufau, N. Richy, M. Groll,
J. Vidal, M. Reboud-Ravaux, Dimerized linear mimics of a natural cyclopeptide
(TMC-95A) are potent noncovalent inhibitors of the eukaryotic 20S protea-
some, J. Med. Chem. 56 (2013) 3367e3378.
[26] N. Basse, M. Montes, X. Marechal, L. Qin, M. Bouvier-Durand, E. Genin, J. Vidal,
B.O. Villoutreix, M. Reboud-Ravaux, Novel organic proteasome inhibitors
identified by virtual and in vitro screening, J. Med. Chem. 53 (2010) 509e513.
[27] S. Mandlekar, A.N. Kong, Mechanisms of tamoxifen-induced apoptosis,
Apoptosis 6 (2001) 469e477.
[28] C. Ferlini, G. Scambia, M. Marone, M. Distefano, C. Gaggini, G. Ferrandina,
A. Fattorossi, G. Isola, P. Benedetti Panici, S. Mancuso, Tamoxifen induces
oxidative stress and apoptosis in oestrogen receptor-negative human cancer
cell lines, Br. J. Cancer 79 (1999) 257e263.
[29] Y. Kang, R. Cortina, R.R. Perry, Role of c-myc in tamoxifen-induced apoptosis
estrogen-independent breast cancer cells, J. Natl. Cancer Inst. 88 (1996)
279e284.
[30] Y. Nagahara, I. Shiina, K. Nakata, A. Sasaki, T. Miyamoto, M. Ikekita, Induction
of mitochondria-involved apoptosis in estrogen receptor-negative cells by a
novel tamoxifen derivative, ridaifen-B, Cancer Sci. 99 (2008) 608e614.
[31] I. Shiina, Y. Sano, K. Nakata, T. Kikuchi, A. Sasaki, M. Ikekita, Y. Nagahara,
Y. Hasome, T. Yamori, K. Yamazaki, Synthesis and pharmacological evaluation
of the novel pseudo-symmetrical tamoxifen derivatives as anti-tumor agents,
Biochem. Pharmacol. 75 (2008) 1014e1026.
[42] W. Heinemeyer, J.A. Kleinschmidt, J. Saidowsky, C. Escher, D.H. Wolf, Pro-
teinase yscE, the yeast proteasome/multicatalytic-multifunctional proteinase:
mutants unravel its function in stress induced proteolysis and uncover its
necessity for cell survival, EMBO J. 10 (1991) 555e562.
[43] R. Gueckel, C. Enenkel, D.H. Wolf, W. Hilt, Mutations in the yeast proteasome
beta-type subunit Pre3 uncover position-dependent effects on proteasomal
peptidase activity and in vivo function, J. Biol. Chem. 273 (1998) 19443e19452.
[44] M. Groll, L. Ditzel, J. Löwe, D. Stock, M. Bochtler, H.D. Bartunik, R. Huber,
Structure of 20S proteasome from yeast at 2.4 Å resolution, Nature 386 (1997)
463e471.
[45] M. Unno, T. Mizushima, Y. Morimoto, Y. Tomisugi, K. Tanaka, N. Yasuoka,
T. Tsukihara, The structure of the mammalian 20S proteasome at 2.75 Å
resolution, Structure 10 (2002) 609e618.
[46] J. Goto, R. Kataoka, H. Muta, N. Hirayama, ASEDock-docking based on alpha
spheres and excluded volumes, J. Chem. Inf. Model. 48 (2008) 583e590.
[47] K. Tanaka, T. Yoshimura, A. Ichihara, Role of substrate in reversible activation
of proteasomes (multi-protease complexes) by sodium dodecyl sulfate,
J. Biochem. (Tokyo) 106 (1989) 495e500.
[48] T. Kunoh, T. Noda, K. Koseki, M. Sekigawa, M. Takagi, K. Shin-ya, N. Goshima,
S. Iemura, T. Natsume, S. Wada, Y. Mukai, S. Ohta, R. Sasaki, T. Mizukami,
A novel human dynactin-associated protein, dynAP, promotes activation of
Akt, and ergosterol-related compounds induce dynAP-dependent apoptosis of
human cancer cells, Mol. Cancer Ther. 9 (2010) 2934e2942.
[32] I. Shiina, Y. Sano, K. Nakata, M. Suzuki, T. Yokoyama, A. Sasaki, T. Orikasa,
T. Miyamoto, M. Ikekita, Y. Nagahara, Y. Hasome, An expeditious synthesis of
tamoxifen, a representative SERM (selective estrogen receptor modulator), via
the three-component coupling reaction among aromatic aldehyde,