Synthesis, in vitro and in vivo biological evaluation, docking studies, and structure-activity relationship (SAR) discussion of dipeptidyl boronic acid proteasome inhibitors composed of β-amino acids

Article abstract of DOI:10.1021/jm901407s

A series of novel dipeptidyl boronic acid proteasome inhibitors composed of β-amino acids were synthesized, in vitro and in vivo biologically evaluated, and theoretically modeled for the first time. From the screened racemic compounds in enzyme, 4i was the most active. The IC₅₀ value of its pure enantiomer 4q was 9.6 nM, 36-fold more active than its isomer 4p and as active as the marketed bortezomib in inhibiting human 20S proteasome. This candidate also showed good activities with IC₅₀ values nearly less than 5 μM against several human solid and hematologic tumor cell lines. Safety evaluation in vivo with zebrafish and Sprague-Dawley (SD) rats showed that the candidate 4q was less toxic than bortezomib. Pharmacokinetic profiles suggested candidate 4q showed a more plasma exposure and longer half-life than bortezomib. Docking results indicated that 4q nearly interacted with 20S proteasome in a similar way as bortezomib.

Full text of DOI:10.1021/jm901407s

1990 J. Med. Chem. 2010, 53, 1990–1999
DOI: 10.1021/jm901407s
Synthesis, in Vitro and in Vivo Biological Evaluation, Docking Studies, and Structure-Activity
Relationship (SAR) Discussion of Dipeptidyl Boronic Acid Proteasome Inhibitors Composed of
β-Amino Acids
Yongqiang Zhu,*,† Xinrong Zhu,† Gang Wu,† Yuheng Ma,† Yuejie Li,† Xin Zhao,† Yunxia Yuan,† Jie Yang,† Sen Yu,†
Feng Shao,† Runtao Li,‡ Yanrong Ke,‡ Aijun Lu,† Zhenming Liu,‡ and Liangren Zhang‡
†Jiangsu Simcere Pharmaceutical Research Institute and Jiangsu Key Laboratory of Molecular Targeted Antitumor Drug Research,
No. 699-18 Xuan Wu Avenue, Xuan Wu District, Nanjing 210042, PRC, and ‡State Key Laboratory of Natural and Biomimetic Drugs,
School of Pharmaceutical Sciences, Peking University, No. 38 Xueyuan Road, Haidian District, Beijing 100191, PRC
Received September 21, 2009
A series of novel dipeptidyl boronic acid proteasome inhibitors composed of β-amino acids were
synthesized, in vitro and in vivo biologically evaluated, and theoretically modeled for the first time.
From the screened racemic compounds in enzyme, 4i was the most active. The IC50 value of its pure
enantiomer 4q was 9.6 nM, 36-fold more active than its isomer 4p and as active as the marketed
bortezomib in inhibiting human 20S proteasome. This candidate also showed good activities with IC50
values nearly less than 5 μM against several human solid and hematologic tumor cell lines. Safety
evaluation in vivo with zebrafish and Sprague-Dawley (SD) rats showed that the candidate 4q was less
toxic than bortezomib. Pharmacokinetic profiles suggested candidate 4q showed a more plasma
exposure and longer half-life than bortezomib. Docking results indicated that 4q nearly interacted with
20S proteasome in a similar way as bortezomib.
Introduction
pase-like (PGPH, β1 subunit). The 19S RP caps at both ends
of the CP and controls the recognition of the ubiquinated
proteins before they enter the 20S proteasome and therein are
degraded into pieces of small amino acids.
The ubiquitin-proteasome pathway (UPPa) is responsible
for the degradation of more than 80% normal and abnormal
intracellular proteins.1 It has been shown that this pathway is
central to normal cellular homeostasis including cell cycle
regulation, DNA repair, sodium channel function, regulation
of immune and inflammatory responses, and cellular response
to stress.2-4 A growing body of evidence suggested that
derangements of UPP could lead to many disorders, such as
malignancies, neurodegenerative diseases, and possible sys-
tematic autoimmunity.5 At the heart of this system is the 26S
proteasome, which is a dynamic multisubunit proteolytic
complex with a molecular weight of 700 kDa and functions
in eukaryote as a key enzyme for nonlysosomal degradation.6
The cylindered 26S proteasome is composed of two subcom-
plexes: one multicatalytic proteinase complex (MPC) 20S
proteasome and two regulatory particles 19S proteasome
(RP). The 20S proteasome is the proteolytic site and shows
three different types of catalytic activities, chymotrypsin-like
(CT-L, β5 subunit), trypsin-like (T-L, β2 subunit), and cas-
The realization that the proteasome is so critically related to
the development of a number of major human diseases
promoted research into the design and synthesis of various
proeasome inhibitors and the evaluation of their therapeutic
potentials. In recent years, there has been a great deal of
interest in proteasome inhibitors as a novel class of anticancer
candidates and some of them entered clinical trial, such
as bortezomib and salinosporamide A (also named NPI-
0052).7-9 The dipeptidyl boronic acid bortezomib was ap-
proved to be used for the treatment of relapsed and refractory
multiple myeloma (MM) in 2003 and mantle cell lymphoma
(MCL) in 2006. Though bortezomib therapy has proven to be
successful for the treatment of hematologic tumor and poten-
tially good for other types of cancer,10,11 prolonged treatment
is associated with toxicity and development of drug resis-
tance.12
On the basis of our accumulated knowledge of the structure
of proteasome and its inhibitors,13-15 we recently disclosed a
comprehensive understanding of dipeptidyl boronate protea-
some inhibitors composed of R-amino acids.16 In an effort to
design the second generation proteasome inhibitors as antic-
ancer agents with less toxicity, nowadays by use of scaffold
migration method, we directed our work to the exploration of
dipeptidyl boronic acids containing β-amino acids. In several
cases, the substitution of an R-amino acid for their β-isomer in
biologically active peptides resulted in an increased activity
and enzymatic stability.17,18 β-Peptides, formed by β-amino
acids having an extra backbone carbon, have been intensely
*To whom correspondence should be addressed. Phone: 86-
25-85566666-1726. Fax: 86-25-85566666-1835. E-mail: zhyqscu@
hotmail.com.
a Abbreviations: SAR, structure-activity relationship; UPP, ubiquitin-
proteasome pathway; MM, multiple myeloma; MCL, mantle cell lym-
phoma; MPC, multicatalytic proteinase complex; RP, regulatory parti-
cle; CT-L, chymotrypsin-like activity; T-L, trypsin-like activity; PGPH,
postglutamyl peptide hydrolysis activity; IC50, inhibition constant;
DCC, N,N0-dicyclohexylcarbodiimide; HOBt, 1-hydroxybenzotriazole;
EDC, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide; DIPEA, N,N-
diisopropylethylamine; sd, standard deviation; SD, Sprague-Dawley;
rmsd, root-mean-square deviation; LLOD, lower limit of detection;
SRM, selected reaction monitoring.
r
pubs.acs.org/jmc
Published on Web 02/16/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 1991
Scheme 1. General Synthesis of Racemized N-Terminal Protected β-Amino Acids 1a-oa
a Reagents and conditions: (i) CH3OH, SOCl2, 0 °C to room temp; (ii) R1COOH, DCC, HOBt, NMM, THF, 0 °C; (iii) (1) 2 N NaOH, acetone/H2O,
0 °C; (2) 2 N HCl, 0 °C.
studied for years to discover stable well-ordered secondary
structures.19-22 Furthermore, the use of β-amino acids as the
building block of proteasome inhibitors has never been
reported.
with n-BuLi and tert-butyl diethylphosphonoacetate 8 at
-78 °C, giving tert-butyl R,β-unsaturated esters 9, which was
purified by chromatography on silica gel and obtained in
high yield.
Nowadays, because of the low cost of zebrafish and its high
degree of similarity with humans in the drug responses,23,24
drug-screening assays using zebrafish are increasingly becom-
ing popular.25,26 Such inexpensive and convenient animal
models have been applied in a variety of assays for assessing
cardiotoxicity,27 hepatotoxicity,28 neurotoxicity,29 and muta-
genesis.30
In this manuscript, we described the synthesis, biological
evaluation, SAR investigation, and toxicity assessment with
zebrafish and SD rats, pharmacokinetic profiles, and docking
studies for a series of dipeptidyl boronic acids composed of β-
amino acids as the novel proteasome inhibitors. Enzymatic
and cellular activity profiles of the most active racemate (()-4i
were compared tothose of itspure enantiomers 4pand 4q. The
in vivo safety profiles of the two isomers and bortezomib were
evaluated and compared. Binding modes of isomers 4p and 4q
with 20S proteasome were studied in detail with the docking
method.
The conjugate additions were carried out between esters 9
and the homochiral (R)- and (S)-N-benzyl-N-(R-methylben-
zyl)amides 10a and 10b, respectively, in the presence of
n-BuLi at -78 °C to produce the (R)- and (S)-enantiomers
of the β-amino acid esters 11a and 11b in high yields. The
N-debenzylation products 12a and 12b were obtained from
their corresponding N-protected esters 11a and 11b in a
methanol-acetic acid-water mixture system catalyzed by
Pd(OH)2 on C in hydrogen atmosphere. After filtration to
remove the heterogeneous catalyst and concentration of the
resulting reaction mixture, the crude products 12a and 12b
were purified by chromatography in the presence of triethy-
lamine and obtained in moderate to high yield.
Coupling reactions were performed between benzoic
acid and β-amino esters 12a and 12b in the presence of
DCC/HOBt coupling reagent to prepare N-terminal pro-
tected tert-butyl ester β-amino acids, which were hydrolyzed
by HCl/Et2O to remove the ester, giving N-terminal pro-
tected β-amino acids 1p and 1q in moderate yields after
crystallization. Chiral HPLC indicated that no isomers were
detected for both compounds.
Results and Discussion
Chemistry. The preparation of racemized β-amino acids 1
was illustrated in Scheme 1. β-Amino acids 5 were purchased
from commercial sources. Methyl ester hydrochlorides 6
were synthesized by the traditional MeOH/SOCl2 method.
After coupling of acid R1COOH with 6 using the traditional
liquid peptide synthesis procedures,31 N-terminal protected
methyl esters 7 were obtained, which were saponificated and
acidified to give acids 1 in high yield.
The generalized preparation of dipeptidyl boronic acids 4
is illustrated in Scheme 3. The key boronate intermediates,
hydrochloride salt of aminoboronates 2, were prepared in
moderate to high yields following our reported method.16
Finally, coupling of aminoboronates 2 with various acids 1 in
the presence of 1-ethyl-3-(3-dimethylaminopropyl)carbodi-
imide hydrochloride (EDC HCl) and HOBt afforded the
3
Chiral β-amino acids 1p and 1q were prepared according
to Scheme 2 by applying a reported procedure.32,33 The
commercially available 4-methoxybenzaldehyde was treated
dipeptidyl boronates 3 which were not purified and directly
transesterificated with isobutylboronic acid under acid con-
ditions to give boronic acids 4.

1992 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
Zhu et al.
Scheme 2. General Synthesis of Chiral N-Terminal Protected β-Amino Acids 1p and 1qa
a Reagents and conditions: (i) (EtO)2POCH2CO2 Bu, n-BuLi, THF, -78 °C to room temp, 86%; (ii) 10a or 10b, n-BuLi, THF, -78 °C, 4 h, 85% for
11a, 88.4% for 11b; (iii) H2, Pd(OH)2/C, MeOH, H2O, AcOH, room temp, 24 h, 73.4% for 12a, 89.2% for 12b; (iv) (1) PhCOOH, DCC, HOBt, NMM,
THF, 0 °C; (2) HCl/Et2O, 0 °C, 18 h, 58.6% for 1p, 61.6% for 1q.
t
Scheme 3. General Synthesis of Dipeptidyl Boronic Acids 4a
teasome with IC50 values widely ranging from 9.6 to 664 nM,
indicating that internal SAR existed. Especially the pure
enantiomer 4q was found to be highly potent with an IC50 of
9.6 nM, which was being regarded as an anticancer drug
candidate to be developed.
Pyrazin-2-yl group was a component of the marketed
bortezomib, so this fragment was also used in the initial
design in R1 position. Exploration of substituents at R1
position revealed the following activity order: phenyl
(4i, 12.5 nM) > nicotin-3-yl (4m, 35 nM) > pyrazin-2-yl
(4c, 134 nM) > 5,6,7,8-tetrahydronaphalen-1-yl (4o, 161
nM), groups at R2 and R3 positions of which were substi-
tuted by 4-methoxy and isopropyl groups, respectively.
Different aromaticity of these four substituents may account
for the different activities. β-Amino acid scanning of the R2
position, including hydrogen, methyl, methoxy, and chloro
substituted β-phenylalanines, demonstrated that electron-
donating groups (such as methyl and methoxy substituted 4h
and 4i) were more beneficial to activities than electron-
withdrawing ones (such as chloro substituted 4j), and the
same conclusion was also drawn by the comparison between
4l, 4m, and 4n. Compound 4g containing no substituent on
the phenyl ring at R2 position showed the worst potency
compared with those substituted by different groups (4h, 4i,
and 4j). Variation of moieties at the R3 position led to distinct
activities. Replacement of aliphatic isopropyl group at the
R3 position of compound 4a (136 nM) with aromatic phenyl
(4e, 664 nM) or 4-methylphenyl ring (4f, 497 nM) respec-
tively resulted in nearly 5-fold or 4-fold potency loss, which
suggested that substituents at this position should be focused
on aliphatic groups in the future drug design. And this is
consistent with what had been discussed previously in di-
peptidyl proteasome inhibitors composed of R-amino acids.
However, extensive understanding of SAR always re-
quires a large sample size of inhibitors, as discussed in our
a Reagents and conditions: (i) EDC HCl, HOBt, DIPEA, CH2Cl2,
3
-15 °C to room temp; (ii) isobutylboronic acid, 2 N HCl, MeOH,
hexane.
Biology. Enzyme. Because of the fact that no SAR of such
novel compounds has been reported, different substituents
of three positions R1, R2, and R3 were initially designed
according to the following rules: for R1 substituents, aro-
matic rings with or without N atoms were employed; for R2
substituents, electron-withdrawing or electron-donating
groups were used; for R3 substituents, aliphalic or aromatic
groups were considered. And the potency of final com-
pounds was evaluated in the human 20S chymotrypsin-like
activity. Table 1 summarized the structures and inhibition
activities of the designed dipeptidyl boronic acids. Because of
the more complex and difficult synthesis of chiral β-amino
acid compared with its racemate, for SAR discussion in this
work, (() β-amino acids at R2 position were employed for
fast biological screening. Table 1 showed that all the com-
pounds inhibited the chymotrypsin-like activity of 20S pro-

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 1993
Table 1. Structures of Compounds 4a-q, Bortezomib, and Their
Inhibition of Human 20S Proteasome
its isomer 4p, which suggested a strictly stereoselective mode
of inhibition between such compounds and 20S protea-
some, which is discussed later with a theoretical modeling
method.
Cellular. The two isomers 4p and 4q and racemate (()-4i
showed quite different activities in inhibiting human 20S
proteasome. How about the cellular activities of these iso-
mers? Subsequently the cytotoxic potentials of compounds
(()-4i, 4p, and 4q were screened against 10 cell lines, includ-
ing 2 human hematologic tumors, HL-60 (promyelocytic
leukemia cell line) and U266 (multi myeloma cell line), and 8
solid tumors, BGC-823 (human gastric carcinoma cell line),
H460 (human large cell lung cancer cell line), BXPC-3
(human pancreatic cancer cell line), SW-480 (human colon
carcinoma cell line), A549 (human non-small-cell lung can-
cer cell line), PC-3 (human prostate cancer line), HepG2
(human hepatocellular liver carcinoma cell line), and SKOV-
3 (human ovarian cancer cell line). The results are displayed
in Table 2. All the three isomers showed good activities with
IC50 values nearly less than 5 μM against BXPC-3, SW-480,
A549, PC-3, HepG2, HL-60, and U266 human cancer cell
lines. Especially, such compounds were very sensitive to the
two hematologic tumor cell lines, which was consistent with
our previous results.16 However, they were not so active
against the other two BGC-823 and H460 human cancer cell
lines. For most human cancer cell lines, the rank order of
potency was 4q > (()-4i > 4p. But the difference in cellular
activities was not as large as in enzymatic assays.
Toxicity Assessment in Vivo. The zebrafish is rapidly
gaining acceptance as a promising animal model for whole-
organism screening for toxicology in the drug discovery
process. Therefore, the compounds can be prioritized for
further development.34 In our candidates selection, the
synthesized 4p, 4q, and bortezomib were screened using
zebrafish embryos to evaluate their safety profiles in vivo.
Interestingly, the two isomers, 4p and 4q, induced no mor-
phological changes of the zebrafish embryos at 1 and 10 μmol
concentrations in 24 h, which indicated both of the isomers
were not toxic in vivo at this level. However, embryos treated
with 1 and 10 μmol of bortezomib displayed different degrees
of cell death in the body (Figure 1c and Figure 1d) and even
embryos death (Figure 1b) compared with the control
(Figure 1a). 28.5% of the embryos displayed cell death only
in the tail (blue arrow indication, Figure 1c), while 52.4% of
those showed a larger range of cell death in the whole embryo
(blue arrow indication, Figure 1d). These facts indicated
that the marketed bortezomib was potentially more toxic
than the candidate 4q, and the results of safety evaluation in
Sprague-Dawley (SD) rats as discussed later also supported
our supposition.
Pharmacokinetic Profiles and Toxicity of 4q. Up to now,
no pharmacokinetic profiles were reported for dipeptidyl
boronic acid proteasome inhibitors composed of β-amino
acids. To evaluate the druggability of such compounds, the
candidate 4q was further investigated by profiling iv and ig
pharmacokinetics in male SD rats. The results compared
with bortezomib are illustrated in Table 3.
After iv bolus injection in male SD rats, 4q was eliminated
slowly with 2-fold longer time than bortezomib (elimination
half-life T1/2z of 5.3 vs 2.0 h). Such a long elimination phase
half-life is due in part to the larger volume of distribution
(Vz = 8449 mL/kg) and much lower systemic clearance
(CLz = 1126 (mL/h)/kg). Furthermore, area-under-the-
a Each IC50 determination was performed with five concentrations,
and each assay point was determined in duplicate. b Pyz, pyrazin-2-yl.
c iPr, iso-propyl. d THNA, 5,6,7,8-tetrahydronaphthalen-1-yl. e IC50 va-
lue obtained for bortezomib under our experimental conditions.
dipeptidyl boronic acid proteasome inhibitors composed of
R-amino acids.16 Nowadays more structurally diverse com-
pounds are being designed and synthesized to clearly and
extensively elucidate the SAR so that much better candidates
are obtained and developed.
Screening results showed that (()-4i was the most active
among all the racemates of dipeptidyl boronic acids. So we
further investigated the activities of its two isomers 4p and
4q. Our studies showed that 4q was 36-fold more active than
curve (AUC) of 4q reached 933 ng h/mL, much larger than
3

1994 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
Zhu et al.
Table 2. Cytotoxicity of Compounds against Tumor Cell Lines (IC50, sd)a
compd
BGC-823
H460
BXPC-3
SW-480
A549
PC-3
SKOV-3
8.43(0.2)
2.82(2.4) 1.72(0.58) 3.68(2.2) 1.11(0.07) 14.5(0.4)
HepG2
HL-60
U266
4i (μM)
4p (μM)
4q (μM)
30.93(5.9)
76.86(7.6)
74.42(2.3)
36.0(6.4)
25.25(2.3)
1.50(0.5) 1.08(0.29) 3.45(2.7) 0.83(0.2)
5.59(0.3) 1.02(0.4)
5.13(0.9) 0.70(0.3)
2.65(0.2)
1.28(0.3)
14.60(0. 7) 0.38(0.2) 0.41(0.12) 0.83(0.5) 0.23(0.04) 3.49(0.20) 1.88(0.1) 0.14(0.01) 0.35(0.04)
16.2(2.6) 6.1(0.4) 6.7(1.9) 7(7.6) 151(6.4) 7.5(0.7) 7.1(0.3) 8.8(0.43)
bortezomib (nM)b 3290(23.4) 780(13.3)
a Each cellular IC50 determination was performed with 10 concentrations, and each assay point was determined in triplicate. b IC50 value obtained for
bortezomib under our experimental conditions.
Figure 1. Morphological changes of zebrafish embryos: (a) untreated embryo (24 h); (b) embryos treated with bortezomib (10 μmol, 24 h);
(c, d) embryos treated with bortezomib (1 μmol, 24 h). Blue arrows indicate the region of cell death.
Table 3. Single Dose iv and ig Pharmacokinetic Profiles of 4q and Bortezomib in SD Rats
dose
(mg/kg)
Cmax
(ng/mL)
AUC(0-t)
(ng h/mL)
compd
Tmax (h)
MRT(0-t) (h)
T1/2z (h)
Vz (mL/kg)
Clz ((mL/h)/kg)
3
4q
1.0 iv
1.0 ig
1.0 iv
1.0 ig
0.03
NDa
0.03
0.23
2929
NDa
1082
161
933
NDa
654
303
2.8
5.3
8449
NDa
3744
29791
1126
NDa
1433
2683
NDa
1.4
NDa
2.0
bortezomibb
7.69
7.47
a ND: not detected. b All the data were obtained in our experimental conditions.
that of bortezomib (654 ng h/mL), which suggested more
3
effects were observed. This result was consistent with what
had been observed with the zebrafish. The detailed toxicity
investigation and in vivo antitumor activities in human U266
and RPMI 8226 nude xenograft models will be reported
elsewhere.
plasma exposure in SD rats than in the standard.
However, when intragastrically administered in male SD
rats, the concentration of 4q declined below 5 ng/mL after
half an hour in plasma, the lower limit of detection (LLOD).
This lower oral bioavailability is possibly due to the declined
absorption in the mouse gastrointestinal tract and first pass
metabolism in the liver. However, the marketed bortezomib
showed an approximate 46% oral bioavailability in our test.
Though previous studies have shown that bortezomib was
orally bioactive,35,36 the current bortezomib therapy in
multimyeloma and mantle cell lymphoma is still intravenous
after being marketed for 6 years and no oral formulation was
reported. Perhaps the instability of boronic acid pharmaco-
phore in such proteasome inhibitors limits their development
as oral formulation in clinical application.
Docking Studies of 4p and 4q. Why were the enzymatic
activities of isomers 4p and 4q quite different? Subsequently,
docking studies were performed to try to theoretically explain
it. To date, several ligand-proteasome crystal complexes
have been reported.37-40 On the basis of these experimental
data, some theoretical modeling was carried out in different
laboratories to describe the interaction mode between struc-
turally diverse inhibitors and 20Sproteasome.41-45 However,
the binding mode of dipeptidyl boronic acid composed of β-
amino acid with catalytic β5 subunit had never been reported
up to now. Because of the covalent binding of dipeptide
boronic acid with proteasome, all the covalent docking
processes were carried out by using GOLD, version 3.0,46,47
which has a unique function to handle covalent docking.
First, to test the credibility of the developed docking
procedure, the crystal structure of bortezomib was docked
After the pharmacokinetic experiment, the safety profiles
of 4q and bortezomib were compared. At 1.0 mg/kg iv and ig
doses of bortezomib, all the SD rats died after 10 h, while the
SD rats in 4q group had not shown any toxic reactions.
Nowadays, even at 2.0 mg/kg iv dose of 4q, no any side

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 1995
Figure 2. Alignment of 4p and 4q with bortezomib and their covalent docking results with β5 subunit: (a) alignment of 4p (in gray) and
bortezomib (in green); (b) alignment of 4q (in gray) and bortezomib (in green); (c) binding mode of 4p with β5 subunit; (d) binding mode of 4q
with β5 subunit. All the hydrogen atoms are not shown. The amino acid residues of β5 subunit are shown in stick models. 4p, 4q, and
bortezomib are represented as ball and stick models. The picture were rendered in PyMOL, version 0.99.48
into the binding pocket of the β5 subunit. As shown by the
docking results, the conformation with the highest fitness
score produced by GOLD was closest to the crystal one, and
the root-mean-square deviation (rmsd) of both conforma-
tions was only 2.0 A, and the distances of covalent bond were
1.5 and 1.4 A for docking conformation and crystal one.
Orientations of the docked R1, R2, and R3 substituents of
bortezomib were quite similar to those of experimental one
(Figure S1 of Supporting Information (SI)). Thus, the dock-
ing procedure was reliable enough to be used in the predic-
tion of the binding modes of 4p, 4q, and 20S proteasome.
Analysis of docking results accounted for the different
potency of the two isomers 4p and 4q. As demonstrated in
Figure 2, the respective alignments of the two isomers 4p (in
gray, Figure 2a) and 4q (in gray, Figure 2b) with bortezomib
(in green, Figure 2a and Figure 2b) were quite different. R3
substituents (isopropyl) of 4p and bortezomib aligned quite
well and oriented in the same direction. However, the R2
group (p-methoxyphenyl) of 4p wrongly overlapped with the
R1 substituent (pyrazin-2-yl) of bortezomib, while the R1
group (benzoyl) of 4p and R2 one (phenyl) of bortezomib
stretch to the same direction but with great deviation. For the
case of 4q and bortezomib, two substituents (R1 and R3)
nearly overlapped; only the R2 groups of both compounds
deviated to a little extent. The more perfect alignment of 4q
than 4p with bortezomib well explained the different activ-
ities in inhibiting 20S proteasome on the one hand.
subunit and between 4q and the β5 subunit. It showed that
four strong and one weak H-bonds were formed between 4q
and amino acid residues of β5 subunit, while only three
strong and one weak H-bonds were observed with 4p. From
docking results, C- and N-terminals of residue THR21
formed two H-bonds with 4q, 3.2 and 2.8 A length, respec-
tively; however, 4p formed only one 3.2 A length H-bond
with the N-terminal of this residue. Furthermore, one strong
2.7 A length H-bond exists between the NH group of the
catalytic THR1 residue and oxygen atom of boronic acid
pharmacophore in 4q in addition to a 1.5 A length covalent
bond.
In summary, the docking results offered us useful infor-
mation for understanding the interaction mode and different
potency between 4p and 4q. It is believed that theoretical
docking can guide us to rationally design novel peptide
boronic acid proteasome inhibitors in the future.
Conclusion
A series of dipeptidyl boronic acid proteasome inhibitors
composed of β-amino acids were synthesized, biologically
investigated in vitro and in vivo, and theoretically modeled.
From these compounds, some structure-activity relation-
ships were concluded and the enantiomer 4q was identified
as a stereoselective proteasome inhibitor at the nanomolar
level, having nearly the same potency as the marketed borte-
zomib. But it was more potent than its isomer 4p and racemic
form (()-4i in inhibiting 20S proteasome activity. Cellular
assays substantially confirmed the activity of the compounds
andwere inline withthestereoselectivityinteractionof the20S
On the other hand, detailed analysis of binding mode of 4p
and 4q with active amino acid residues of β5 subunit also
elucidated the inhibition variation. Figure 2c and Figure 2d
respectively presented the interactions between 4p and the β5

1996 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
Zhu et al.
proteasome. In vivo toxicity assessment with zebrafish as a
whole-organism screening showed that the two pure isomers
were nontoxic while bortezomib induced some phenotypic
changes to a certain extent, which was supported by observa-
tion of toxic reactions of SD rats after pharmacokinetic
experiments. Pharmacokinetic profiles suggested that candi-
date 4q showed a more plasma exposure and longer half-life
than bortezomib. Docking studies indicated that 4q adopted a
similar mode to inhibit the 20S proteasome as bortezomib. The
docking results offered us much useful information to under-
stand the interaction mode between such inhibitors and 20S
proteasome and will facilitate the next cycle of drug design.
n-BuLi (0.4 mL, 2.5 M in hexane, 1 mmol) at -78 °C. After 1 h, a
solution of R,β-unsaturated ester (9) (0.1 g, 0.5 mmol) in THF
(2 mL) was added all at once. The resulting solution was stirred
at -78 °C for 4 h before quenching with saturated aqueous
NH4Cl (1 mL). When the mixture was warmed to room tem-
perature, THF was removed and the product was extracted with
DCM. The organic fractions were washed with 10% aqueous
nitric acid, saturated aqueous sodium bicarbonate, and brine
and dried over anhydrous Na2SO4 and the solvent was removed
in vacuo to yield the crude product, which was chroma-
tographed using ethyl acetate/petroleum (1:50) as eluent to
1
give pure pale-yellow 11a (0.20 g, 88.4%). H NMR (CDCl3,
500 MHz) δ 1.19-1.25 (-CH3, m, 12H), 2.46 (-CH2, dd, J1 =
10.2 Hz, J2 = 14.5 Hz, 1H), 2.52 (-CH2, dd, J1 = 4.9 Hz, J2 =
14.5 Hz, 1H), 3.66 (-CH2, s, 2H), 3.78 (-OCH3, s, 3H), 3.98
(-CH, q, J = 6.8 Hz, 1H), 4.34 (-CH, q, J = 4.9 Hz, 1H),
6.84-6.86 (-Ph, m, 2H), 7.14-7.15 (-Ph, m, 1H), 7.20-7.25
(-Ph, m, 4H), 7.28-7.31 (-Ph, m, 4H), 7.39-7.41 (-Ph, m,
2H). 13C NMR (CDCl3, 125 MHz) δ 27.76, 38.57, 55.16, 57.10,
59.09, 60.31, 80.04, 126.45, 126.70, 127.79, 127.99, 128.09,
128.15, 129.33, 133.90, 141.86, 144.36, 158.68, 171.19. MS
(ESI) m/z 446.2 [M þ H]þ, 468.2 [M þ Na]þ.
Experimental Procedures
Commercially available reagents were used directly without
any purification unless otherwise stated. Reaction progress was
monitored by silica gel aluminum sheets (60F-254) and RP-18
F254s using UV light as a visualizing agent and 15% ethanolic
phosphomolybdic acid and heat or ninhydrin and heat as devel-
oping agent. The 200-300 mesh silica gel and ODS C-18 were
used for column chromatography. For compounds with fluorine
atoms, analytical reverse phase HPLC was run using Sinochrom
ODS-BP 4.6 mm ꢀ 150 mm column, eluting with a mixture of
20% acetonitrile and 80% water containing 1% β-cyclodextrin.
For other compounds, a reverse phase Extend C18 4.6 mm ꢀ
150 mm column was employed and eluted with a mixture of 15%
acetonitrile and 85% water containing 0.1% triethylamine.
HPLC showed that the purity of all the final products was greater
than 95%. Melting points were obtained on an YRT-3 melting
point apparatus and were uncorrected. 1H NMR and 13C NMR
spectra were recorded on Bruker Avance 300 or Avance 500
spectrometer with TMS as internal standard. Splitting patterns
are described as singlet (s), doublet (d), triplet (t), quartet (q),
broad (br), or doublet of doublet (dd). The value of chemical
shifts (δ) is given in ppm and coupling constants (J) in Hertz (Hz).
Mass spectra were obtained using Agilent LC-MS (1956B)
instrument in electrospray positive and negative ionization
modes. Mass spectra of product ions studied in pharmacokinetics
were obtained using Agilent LC/MS/MS (6410). High-resolution
mass spectra were recorded on a ZAB-HS instrument using an
electrospray source (ESI).
To a stirred solution of 11a (0.88 g, 1.98 mmol) in methanol
(6 mL), distilled water (0.7 mL), and acetic acid (0.2 mL) was
added palladium hydroxide on carbon (1 g, 64.3% water). A
hydrogen balloon was attached and the resulting suspension
stirred for 24 h under 1 atm of hydrogen. The reaction mixture
was filtered through Celite, and the solvent was removed in
vacuo. The residue was partitioned between DCM (10 mL) and
saturated aqueous NaHCO3 (5 mL). The organic phase was
separated, and the aqueous phase was extracted with DCM (2 ꢀ
10 mL). The combined organic layers were dried over Na2SO4,
and the solvent was removed in vacuo. The residue was crudely
purified by passing through a short plug of silica gel (DCM/
Et3N 100:1) to yield the product 12a (0.64 g, 73.4%), which was
directly used in the next reaction.
At 0 °C, to a stirred solution of 12a (0.60 g, 2.40 mmol) and
NMM (0.30 g, 2.97 mmol) in THF (10 mL) was added benzoyl
chloride (0.4 g, 2.85 mmol). The mixture was allowed to stir
overnight. The solvent was evaporated and dissolved in DCM
(20 mL), washed with 10% nitric acid, saturated aqueous
NaHCO3 (10 mL), and brine, and dried over anhydrous
Na2SO4. The solvent was removed in vacuo, and the residue
was dissolved in HCl/Et2O (10 mL, 6M, 60 mmol) and stirred
overnight at 0 °C. The organic phase was removed in vacuo and
the resulting solid was washed with ether (3 ꢀ 10 mL) and dried
to give the title compound 1p (0.21 g, 58.6%) as a white solid.
Chiral HPLC indicated purity of 100 area %, and no isomer was
detected. Mp 175.7-176.9 °C. 1H NMR (DMSO-d6, 500 MHz)
δ 2.75 (-CH2, dd, J1 = 6.2 Hz, J2 = 15.4 Hz, 1H), 2.88 (-CH2,
dd, J1 = 6.9 Hz, J2 = 15.5 Hz, 1H), 3.72 (-CH3, s, 3H), 5.39
(-CH, dd, J1 = 6.8 Hz, J2 = 8.1 Hz, 1H), 6.87-6.89 (-Ph, m,
2H), 7.33-7.34 (-Ph, m, 2H), 7.44-7.47 (-Ph, m, 2H),
7.50-7.53 (-Ph, m, 1H), 7.83-7.84 (-Ph, m, 2H), 8.79
(-CONH, d, J = 8.2 Hz, 1H), 12.20 (-COOH, s, 1H). 13C
NMR (DMSO-d6, 125 MHz) δ 49.38, 54.97, 113.55, 127.17,
127.66, 128.09, 131.04, 134.44, 134.68, 158.14, 165.30, 171.76.
MS (ESI) m/z 299.3 [M - H]-.
(R)-3-Benzamido-3-(4-methoxyphenyl)propanoic Acid (1q).
Compound 1q was prepared in 61.6% yield following a similar
procedure described for the synthesis of 1p. Chiral HPLC
indicated purity of 100 area %, and no isomer was detected.
Synthesis of 11b. The synthesis involved (S)-N-Benzyl-N-R-
methylbenzylamine (0.38 g, 1.8 mmol), n-BuLi (0.8 mL, 2.5 M in
hexane, 2.0 mmol), and R,β-unsaturated ester (9) (0.23 g, 1.0
mmol), giving 85% yield. 1H NMR (CDCl3, 500 MHz) δ 1.19-
1.25 (-CH3, m, 12H), 2.46 (-CH2, dd, J1 = 10.2 Hz, J2 =
14.4 Hz, 1H), 2.52 (-CH2, dd, J1 = 5.0 Hz, J2 = 14.4 Hz, 1H),
3.66 (-CH2, s, 2H), 3.78 (-OCH3, s, 3H), 3.98 (-CH, q, J =
6.8 Hz, 1H), 4.34 (-CH, q, J = 5.0 Hz, 1H), 6.84-6.86 (-Ph, m,
The synthesis of chiral N-terminal protected β-amino acids 1p,
1q, intermediates 3i, 3p, 3q, and target molecule 4i and its two
isomers 4p and 4q is indicated below. Detailed synthesis and
structural characterization data of 1a-o, 2a-q, 3a-h, 3j-o,
4a-h, and 4j-o are described in Supporting Information.
(S)-3-Benzamido-3-(4-methoxyphenyl)propanoic Acid (1p).
This compound was synthesized according to the reported
methods. To a stirred solution of tert-butyl diethylphosphonoa-
cetate (8) (0.44 g, 1.74 mmol) in absolute THF (3 mL) was
dropwise added n-BuLi (0.7 mL, 2.5 M in hexane, 1.75 mmol) at
-78 °C. After the mixture was stirred for 2 h at -78 °C, a
solution of the aldehyde (0.17 g, 1.74 mmol) in THF (2 mL) was
added all at once. The resulting solution was stirred at -78 °C
for 3 h before rising to room temperature. The solution was
subsequently cooled to -78 °C and quenched with saturated
aqueous NH4Cl (2 mL). THF was removed in vacuo. The
product was extracted with DCM (3 ꢀ 15 mL), the combined
organic extracts were dried with anhydrous Na2SO4, and the
solvent was removed in vacuo to yield the crude product, which
was chromatographed using CH2Cl2/petroleum (1:20) as eluent
to give a pure sticky solid (9) (0.30 g, 86%). 1H NMR (CDCl3,
300 MHz) δ 1.53 (-CH3, m, 9H), 3.83 (-OCH3, s, 3H), 6.24
(-CdCH, d, J = 15.9 Hz, 1H), 6.87-6.90 (-Ph, m, 2H),
7.44-7.47 (-Ph, m, 2H), 7.54 (-CdCH, d, J = 15.9 Hz, 1H).
MS (ESI) m/z 257.0 [M þ Na]þ.
To a stirred solution of (R)-N-benzyl-N-R-methylbenzy-
lamine (0.19 g, 0.9 mmol) in THF (3 mL) was slowly added

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 1997
2H), 7.14-7.17 (-Ph, m, 1H), 7.19-7.27 (-Ph, m, 5H), 7.28-
7.31 (-Ph, m, 4H), 7.39-7.41 (-Ph, m, 2H). 13C NMR (CDCl3,
125 MHz) δ 27.76, 38.57, 55.16, 57.10, 59.09, 60.31, 80.04,
126.45, 126.76, 127.79, 127.81, 128.09, 128.15, 129.33, 133.90,
141.86, 144.36, 158.68, 171.19. MS (ESI) m/z 446.2 [M þ H]þ.
Synthesis of 12b. To compound (11b) (0.35 g, 0.78 mmol),
methanol (5 mL), distilled water (0.5 mL), and acetic acid
(0.1 mL) was added palladium hydroxide on carbon (0.4 g,
64.3% water). Purification with chromatography (CH2Cl2 as
eluent) gave a yellow oil 12b (0.17 g, 89.2%).
1.23-1.31 (-CH2, m, 2H), 1.46-1.59 (-CH, m, 1H), 2.58
(-CH, t, J = 7.9 Hz, 1H), 2.98-3.07 (-CH2, m, 2H), 3.77
(-CH3, s, 3H), 5.58 (-CH, t, J = 7.8 Hz, 1H), 6.90 (-Ph, d, J =
8.6 Hz, 2H), 7.36-7.39 (-Ph, m, 2H), 7.42-7.47 (-Ph, m, 2H),
7.51-7.56 (-Ph, m, 1H), 7.80-7.82 (-Ph, m, 2H). 13C NMR
(CD3OD, 75 MHz) δ 22.11, 23.86, 26.88, 37.72, 41.02, 51.67,
55.76, 115.16, 128.42, 128.97, 129.52, 132.78, 135.60, 160.83,
169.55, 177.07. MS (ESI) m/z 440.2 [M þ 2CH2]-, 450.2 [M þ
37]þ. HRMS [M þ Na þ 2CH2]þ calcd, 463.2380; found,
463.2409.
Synthesis of 1q. The synthesis involved compound 12b (0.1 g,
0.4 mmol), NMM (0.05 mL, 0.48 mmol), benzoyl chloride
(0.06 mL, 0.48 mmol), and HCl/Et2O (5 mL, 6 M, 30 mmol),
giving 61.6% yield. Mp 183.5-186.7 °C. 1H NMR (DMSO-d6,
500 MHz) δ 2.75 (-CH2, dd, J1 = 6.5 Hz, J2 = 15.6 Hz, 1H),
2.88 (-CH2, dd, J1 = 8.7 Hz, J2 = 15.6 Hz, 1H), 3.72 (-CH3, s,
3H), 5.39 (-CH, dd, J1 = 6.6 Hz, J2 = 8.4 Hz, 1H), 6.87-6.89
(-Ph, m, 2H), 7.32-7.35 (-Ph, m, 2H), 7.44-7.47 (-Ph, m,
2H), 7.50-7.53 (-Ph, m, 1H), 7.82-7.84 (-Ph, m, 2H), 8.79
(-CONH, d, J = 8.3 Hz, 1H), 12.12 (-COOH, s, 1H). 13C
NMR (DMSO-d6, 125 MHz) δ 49.42, 54.98, 113.58, 127.21,
127.70, 128.12, 131.07, 134.48, 134.73, 158.18, 165.34, 171.80.
MS (ESI) m/z 299.3.
(R)-1-[(S)-3-Benzamido-3-(4-methoxyphenyl)propanamido]-3-
methylbutylboronic Acid (4p). The synthesis involved a similar
procedure as described for 4i, using 3p (0.55 g, 1.00 mmol),
2-methylpropylboronic acid (0.51 g, 5.0 mmol), methanol
(15 mL), hexane (15 mL), 1 N HCl (2.5 mL) to give 0.22 g of a
white foam solid (53.3% yield). 1H NMR (CD3OD, 500 MHz) δ
0.84-0.88 (-CH3, m, 6H), 1.14-1.17 (-CH2, m, 1H), 1.23-
1.25 (-CH2, m, 1H), 1.50-1.58 (-CH, m, 1H), 2.58 (-CH, dd,
J1 = 8.6 Hz, J2 = 6.5 Hz, 1H), 3.02-3.13 (-CH2, m, 2H), 3.77
(-CH3, s, 3H), 5.57 (-CH, t, J = 7.8 Hz, 1H), 6.91 (-Ph, d, J =
8.7 Hz, 2H), 7.37 (-Ph, d, J = 8.7 Hz, 2H), 7.43-7.46 (-Ph, m,
2H), 7.51-7.54 (-Ph, m, 1H), 7.80-7.82 (-Ph, m, 2H). 13C
NMR (CD3OD, 125 MHz) δ 21.98, 23.88, 26.87, 37.82, 41.01,
45.60, 51.67, 55.75, 115.18, 128.50, 129.12, 129.58, 132.85,
133.77, 135.64, 160.93, 169.69, 177.22. MS (ESI) m/z 448.2
[M þ Cl]-, 435.1 [M þ Na]þ. HRMS [M þ Na þ 2CH2]þ calcd,
463.2379; found, 463.2371.
Pinanediol Ester (R)-1-[3-Benzamido-3-(4-methoxyphenyl)-
propanamido]-3- methylbutylboronic Acid (3i). To a cooled solu-
tion (-5 °C) of 3-benzamido-3-(4-methoxy)phenylpropanoic
acid (1i) (0.21 g, 0.70 mmol) dissolved in anhydrous CH2Cl2
(15 mL) was added HOBt (0.23 g, 1.70 mmol). After 20 min, the
[(1R)-1-[[(3R)-3-(4-Methoxyphenyl)-3-[(phenylcarbonyl)amino]-
1-oxopropyl]amino]-3-methylbutyl]boronic Acid (4q). The synth-
esis involved a similar procedure as described for 4i, using 3q
(0.55 g, 1.00 mmol), 2-methylpropylboronic acid (0.51 g,
5.0 mmol), methanol (20 mL), hexane (20 mL), 1 N HCl (2.5
reaction system was cooled to -15 °C and EDC HCl (0.33 g,
3
1.70 mmol) was added. Finally the precooled (0 °C) mixture
of the known pinanediol boronate aminohydrochloride (2i)
(0.21 g, 0.70 mmol) and DIPEA (0.15 mL, 0.84 mmol) in
anhydrous CH2Cl2 (10 mL) was poured. The mixture was stirred
at -15 °C for 1 h and at room temperature for 2 h, and the
reaction was finally quenched with water. The aqueous phase
was extracted with CH2Cl2 (3 ꢀ 100 mL). The combined organic
phase was washed with 10% citric acid, 5% NaHCO3, and
brine, dried over anhydrous Na2SO4, filtered, and evaporated to
provide crude product 3i, which was directly used in the next
reaction.
1
mL) to give 0.232 g of a white foam solid (56.4% yield). H
NMR (CD3OD, 500 MHz) δ 0.83-0.88 (-CH3, m, 6H), 1.15-
1.17 (-CH2, m, 1H), 1.22-1.26 (-CH2, m, 1H), 1.51-1.58
(-CH, m, 1H), 2.58 (-CH, q, J = 6.5 Hz, 1H), 3.04 (-CH2, dd,
J1 = 5.6 Hz, J2 = 14.6 Hz, 1H), 3.10 (-CH2, dd, J1 = 8.4 Hz,
J2 = 14.6 Hz, 1H), 3.77 (-OCH3, s, 3H), 5.57 (-CH, t, J =
7.9 Hz, 1H), 6.89-6.92 (-Ph, m, 2H), 7.36-7.38 (-Ph, m, 2H),
7.43-7.46 (-Ph, m, 2H), 7.51-7.54 (-Ph, m, 1H), 7.80-7.82
(-Ph, m, 2H). 13C NMR (CD3OD, 125 MHz) δ 22.09, 23.83,
26.98, 37.67, 45.61, 48.49, 51.58, 55.75, 128.46, 129.08, 129.56,
132.82, 133.74, 135.60, 160.86, 169.59, 177.21. MS (ESI) m/z
435.1 [M þ Na]þ. HRMS [M þ Na þ 2CH2]þ calcd, 463.2379;
found, 463.2371.
Enzyme and Inhibition Asssays. The 20S proteasome activity
assay kit was purchased from Chemicon. Other reagents and
solvents were purchased from commercial sources. In brief,
substrates and compounds were previously dissolved in DMSO,
with the final solvent concentration kept constant at 3% (v/v).
The reaction buffers were (pH 7.5) 20 mM Tris, 1 mM DTT,
10% glycerol, and 0.02% (w/v) DS for CT-L activities. Protea-
some activity was determined by monitoring the hydrolysis of
Pinanediol Ester (R)-1-[(S)-3-Benzamido-3-(4-methoxyphenyl)-
propanamido]-3-methylbutylboronic Acid (3p). The synthesis in-
volved a similar procedure as above but with 1p (0.27 g, 1.00 mmol),
CH2Cl2 (30 mL), HOBt (0.16 g, 1.20 mmol), EDC HCl (0.19 g,
3
1.00 mmol), 2p (0.30 g, 1.00 mmol), DIPEA (0.26 mL, 1.48 mmol).
After being washed with 10% citric acid, 5% NaHCO3, and brine
and dried over anhydrous Na2SO4, the crude product was directly
used in the next reaction.
Pinanediol Ester (R)-1-[(R)-3-Benzamido-3-(4-methoxyphenyl)-
propanamido]-3-methylbutylboronic Acid (3q). The synthesis in-
volved a similar procedure as above but with 1q (0.27 g, 1.00 mmol),
CH2Cl2 (50 mL), HOBt (0.16 g, 1.20 mmol), EDC HCl (0.19 g,
3
1.00 mmol), 2q (0.30 g, 1.00 mmol), DIPEA (0.26 mL, 1.48 mmol).
After being washed with 10% citric acid, 5% NaHCO3, and brine
and dried over anhydrous Na2SO4, the crude product was directly
used in the next reaction.
the fluorogenic substrate, Suc-Leu-Leu-Val-Tyr-AMC (λexc
=
360 and λexc = 465 nm for AMC substrates), reacting for 1 h at
37 °C in the presence of untreated (control) or proteasome that
had been incubated with five different concentrations of test
compounds. Fluorescence was measured using an Infinite M200
microplate reader (Tecan, Austria).
Cell Culture and Cytotoxicity Assays. HL-60 (promyelocytic
leukemia cell line), BXPC-3 (human pancreatic cancer cell line),
PC-3 (human prostate cancer line), and U266 (multimyeloma
cell line) human cell lines were obtained from the American
Type Culture Collection (Manassas, VA). H460 (human large
cell lung cancer cell line), A549 (human nonsmall cell lung
cancer cell line), SW-480 (human colon carcinoma cell line),
SKOV-3 (human ovarian carcinoma), HepG2 (human hepato-
cellular liver carcinoma cell line), and BGC-823 (human gastric
carcinoma cell line) cell lines were obtained from China Phar-
maceutical University. HL-60 cell was cultured in IMDM
(R)-1-[3-Benzamido-3-(4-methoxyphenyl)propanamido]-3-meth-
ylbutylboronic Acid (4i). To a solution of 3i (0.35 g, 0.64 mmol)
and 2-methylpropylboronic acid (0.33 g, 3.20 mmol) dissolved
in methanol (13 mL) and hexane (13 mL) was added 1 N HCl
(1.6 mL). The mixture was stirred at room temperature for 18 h.
The methanolic phase was washed with hexane (3 ꢀ 10 mL), and
the hexane layer was extracted with methanol (3 ꢀ 15 mL). The
combined methanolic layers were evaporated in vacuo, and the
residue was dissolved in CH2Cl2 (30 mL). The solution was
washed with 5% NaHCO3 (15 mL), and the organic layer was
dried over anhydrous Na2SO4, evaporated, and purified with
chromatography (CH3OH/CH2Cl2 = 1:20 and then 1:5) to
obtain 190 mg (71.3% yield) of a pale-yellow foam solid 4i. 1H
NMR (CD3OD, 300 MHz) δ 0.83-0.89 (-CH3, m, 6H),

1998 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5
Zhu et al.
supplemented with 20% fetal bovine serum at 37 °C in 5% CO2.
BGC-823, BXPC-3, H460, HepG2, and SW-480 cells were
grown in RPMI 1640 supplemented with 10% fetal bovine
serum at 37 °C in 5% CO2. U266 cell was cultured in RPMI
1640 supplemented with 15% fetal bovine serum at 37 °C in 5%
CO2. PC-3 and A549 cells were cultured in F12-k supplemented
with 10% fetal bovine serum at 37 °C in 5% CO2. SKOV-3 cell
was cultured in McCoy’s 5A supplemented with 10% fetal
bovine serum at 37 °C in 5% CO2.
Acknowledgment. This project was supported in part by
the National Key New Drug Creation Program (Grant
2009ZX09103-001) and Jiangsu Key Laboratory of Molecu-
lar Targeted Antitumor Drug Research Foundation (Grant
BM2008201).
Supporting Information Available: General and experimental
procedures; characterization data for 1a-o, 2a-q, 3a-h, 3j-o,
4a-h, 4j-o; and comparison of binding mode between docked
and crystal bortezomib. This material is available free of charge
via the Internet at http://pubs.acs.org.
A standard MTT assay was used to measure cell growth. In
brief, a suspension of 3000 cells/150 μL of medium was added to
each well of 96-well plates and allowed to grow. Twenty-four
hours later, drugs prepared in medium at 10 different concen-
trations were added to the corresponding plates at a volume of
50 μL per well, and the plates were incubated for 72 h with drugs.
Then an amount of 20 μL of a solution of 5 mg/mL MTT was
added to each well and incubated for another 4 h at 37 °C. Plates
were then centrifuged at 1000 rpm at 4 °C for 5 min, and the
medium was carefully discarded. The formazan crystals were
dissolved in 100 μL of DMSO, and absorbance was read on an
Infinite M200 (Tecan, Austria) microplate reader at 540 nm.
The result was expressed as the mean IC50 value, which is the
average from at least three independent determinations.
Zebrafish Screening. AB strain was used in the experiment.
Embryos were fed in Holtfreter’s water at 28.5 °C and staged
according to Kimmel et al.49 The stock was made by dissolving test
compounds first in absolute ethanol and then diluted in Holtfreter’s
water to a concentration of 0.01 M, while the concentration of
absolute ethanol was kept at 1%. The stock was stored at -20 °C
and diluted to a working concentration in Holtfreter’s water.
Embryos were treated from bud stage in the solution of test
compounds and observed afterward and photographed at 24 h
postfertilization (hpf) using a Leica MZ 16 stereomicroscope.
Pharmacokinetic Analyses in Rodents. Solutions of 4q and
bortezomib were prepared in 1% ethanol, 1% Tween-80, and
98% buffered saline with the final concentration of both com-
pounds being 0.5 mg/mL for iv and ig administration. Hepar-
inized blood samples collected for PK analyses (n = 4) were
centrifuged at 4000 rpm for 5 min at 4 °C. Plasma samples were
analyzed after protein precipitation with acetonitrile acidified
with 1% formic acid. LC/MS/MS analysis of 4q and bortezomib
was performed under optimized conditions to obtain the best
sensitivity and selectivity of the analyte in selected reaction
monitoring mode (SRM). Selected product ions of 4q were
monitored for the quantification of the compound using borte-
zomib as an internal standard. Plasma concentration-time data
were analyzed by a noncompartmental approach using the
software WinNonlin Enterprise, version 5.2 (Pharsight Co.,
Mountain View, CA).
Docking Studies. The protein and ligands for docking were
prepared using the software Insight II.50 The initial structure of
bortezomib was derived from the crystal complex coordinates
(PDB code 2F16), and the original structures of 4p and 4q were
constructed on the basis of the bortezomib crystal structure
conformation followed by energy minimization. 4p and 4q were
then covalently docked to the binding pocket of β5 subunit using
GOLD, version 3.0. A radius of 20 A from the β5 catalytic N-
terminal threonine was used to direct site location. For each of
the genetic algorithm runs, a maximum of 100 000 operations
were performed on a population of 100 individuals with a
selection pressure of 1.1. Operator weights for crossover, muta-
tion, and migration were set to 95, 95, and 10, respectively, as
recommended by the authors of the software. Fifty GA runs
were performed in each docking experiment as done in the
software validation procedure. The default GOLD fitness func-
tion was used to identify the better binding mode.51 The distance
for hydrogen bonding was set to 2.5 A, and the cutoff value for
van der Waals calculation was set to 4 A. Covalent docking was
applied, and the terminal boron atoms of all the ligands were
bonded to the hydroxyl oxygen of Thr1.
References
(1) Goldberg, A. L. Protein degradation and protection against mis-
folded or damaged proteins. Nature 2003, 426, 895–899.
(2) Ciechanover, A. The ubiquitin proteolytic system: from a vague
idea, through basic mechanisms, and onto human diseases and
drug targeting. Neurology 2006, 66, S7–S19.
(3) Ciechanover, A. The ubiquitin-proteasome pathway: on protein
death and cell life. EMBO J. 1998, 17, 7151–7160.
(4) Malik, B.; Price, S. R.; Mitch, W. E.; Yue, Q.; Eaton, D. C.
Regulation of epithelial sodium channels by the ubiquitin-protea-
some proteolytic pathway. Am. J. Physiol.: Renal Physiol. 2006,
290, F1285–F1294.
(5) Nalepa, G.; Rolfe, M.; Harper, J. W. Drug discovery in the
ubiquitin-proteasome system. Nat. Rev. Drug Discovery 2006, 5,
596–613.
(6) Voges, D.; Zwickl, P.; Baumeister, W. The 26S proteasome: a
molecular machine designed for controlled proteolysis. Annu. Rev.
Biochem. 1999, 68, 1015–1068.
(7) Adams, J. The development of proteasome inhibitors as anticancer
drugs. Cancer Cell 2003, 5, 417–421.
(8) Orlowski, R. Z.; Zeger, E. L. Targeting the proteasome as a
therapeutic strategy against hematological malignancies. Expert
Opin. Invest. Drugs 2006, 15, 117–130.
(9) Townsend, A. R.; Millward, M.; Price, T.; Mainwaring, P.;
Spencer, A.; Longenecker, A.; Palladino, M. A.; Lloyd, G. K.;
Spear, M. A.; Padrik, P. Clinical trial of NPI-0052 in advanced
malignancies including lymphoma and leukemia (advanced malig-
nancies arm). J. Clin. Oncol. 2009, 27, 3582.
(10) Papandreou, C. N.; Daliani, D. D.; Nix, D.; Yang, H.; Madden, T.;
Wang, X.; Pien, C. S.; Millikan, R. E.; Tu, S. M.; Pagliaro, L.; Kim,
J.; Adams, J.; Elliott, P.; Esseltine, D.; Petrusich, A.; Dieringer, P.;
Perez, C.; Logothetis, C. J. Phase I trial of the proteasome inhibitor
bortezomib in patients with advanced solid tumors with observa-
tions in androgen-independent prostate cancer. J. Clin. Oncol.
2004, 22, 2108–2121.
(11) Cusack, J. C. Rationale for the treatment of solid tumors with the
proteasome inhibitor bortezomib. Cancer Treat Rev. 2003, 29
(Suppl. 1), 21–31.
(12) Chauhan, D.; Hideshima, T.; Anderson, K. C. Proteasome inhibi-
tion in multiple myeloma: therapeutic implication. Annu. Rev.
Pharmacol. Toxicol. 2005, 45, 465–476.
(13) Zhu, Y. Q.; Pei, J. F.; Liu, Z. M.; Lai, L. H.; Cui, J. R.; Li, R. T. 3D-
QSAR studies on tripeptide aldehyde inhibitors of proteasome
using CoMFA and CoMSIA methods. Bioorg. Med. Chem. 2006,
14, 1483–1496.
(14) Zhu, Y. Q.; Lei, M.; Lu, A. J.; Zhao, X.; Yin, X. J.; Gao, Q. Z. 3D-
QSAR studies of boron-containing dipeptides as proteasome
inhibitors with CoMFA and CoMSIA methods. Eur. J. Med.
Chem. 2009, 44, 1486–1499.
(15) Zhu, Y. Q.; Yao, S. Y.; Xu, B.; Ge, Z. M.; Cui, J. R.; Chen, T. M.;
Li, R. T. Design, synthesis and biological evaluation of tripeptide
boronic acid proteasome inhibitors. Bioorg. Med. Chem. 2009, 17,
6851–6861.
(16) Zhu, Y. Q.; Zhao, X.; Zhu, X. R.; Wu, G.; Li, Y. J.; Ma, Y. H.;
Yuan, Y. X.; Yang, J.; Hu, Y.; Ai, L.; Gao, Q. Z. Design, synthesis,
biological evaluation, and structure-activity relationship (SAR)
discussion of dipeptidyl boronate proteasome inhibitors, part I:
comprehensive understanding of the SAR of R-amino acid bor-
onates. J. Med. Chem. 2009, 52, 4192–4199.
(17) Seebach, D.; Abele, S.; Schreiber, J. V.; Martinoni, B.; Nussbaum,
A. K.; Schild, H.; Schulz, H.; Henneke, H.; Woessner, R.; Bitsch, F.
Biological and pharmacokinetic studies with β-peptides. Chimia
1998, 52, 734–739.
(18) Shinagawa, S.; Kanamaru, T.; Harada, S.; Asai, M.; Okazaki, H.
Chemistry of emeriamine and its analogs and their inhibitory
activity in long-chain fatty acid oxidation. J. Med. Chem. 1987,
30, 1458–1463.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 5 1999
(19) Borman, S. β-Peptide: nature improved? Chem. Eng. News 1997,
75, 32–35.
streptoccal cell wall-induced polyarthritis. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 15671–15676.
(20) Seebach, D.; Abele, S.; Gademann, K.; Guichard, G.; Hintermann,
T.; Jaun, B.; Mattews, J. L.; Schreiber, J. V.; Oberer, L.; Hommel,
U.; Widmer, H. β2- and β3-peptides with proteinaceous side chains:
synthesis and solution structures of constitutional isomers, a novel
helical secondary structure and the influence of solvation and
hydrophobic interactions on folding. Helv. Chim. Acta 1998, 81,
932–982.
(36) Teicher, B. A.; Ara, G.; Herbst, R.; Palombella, V. J.; Adams, J.
The proteasome inhibitor PS-341 in cancer therapy. Clin. Cancer
Res. 1999, 5, 2638–2645.
€
(37) Lowe, J.; Stock, D.; Jap, B.; Zwickl, P.; Baumeister, W.; Huber, R.
Crystal structure of the 20S proteasome from the archaeon
T. acidophilum at 3.4 A resolution. Science 1995, 268, 533–539.
€
(38) Groll, M.; Ditzel, L.; Lowe, J.; Stock, D.; Bochtler, M. B.; Huber,
(21) Appella, D. H.; Barchi, J. J., Jr.; Durell, S. R.; Gellman, S. H.
Formation of short, stable helices in aqueous solution by β-amino
acid hexamers. J. Am. Chem. Soc. 1999, 121, 2309–2310.
(22) Hanessian, S.; Yang, H. Design of a novel secondary structure
scaffolding device: induction of a reverse turn in tetrapeptides by
incorporating a β-amino acid and stereocontrolled free radical R-
substitution reactions in peptide motifs. Tetrahedron Lett. 1997, 38,
3155–3158.
R. Structure of 20S proteasome from yeast at 2.4 A resolution.
Nature (London) 1997, 386, 463–471.
(39) Groll, M.; Huber, R.; Potts, B. C. M. Crystal structures of
salinosporamide A (NPI-0052) and B (NPI-0047) in complex with
the 20S proteasome reveal important consequences of β-lactone
ring opening and mechanism for irreversible binding. J. Am. Chem.
Soc. 2006, 128, 5136–5141.
(40) Groll, M.; Berkers, C. R.; Ploegh, H. L.; Ovaa, H. Crystal structure
of the boronic acid-based proteasome inhibitor bortezomib in
complex with the yeast 20S proteasome. Structure 2006, 14, 451–456.
(23) Langheinrich, U. Zebrafish: a new model on the pharmaceutical
catwalk. BioEssays 2003, 25, 904–912.
€
(24) Milan, D. J.; Peterson, T. A.; Ruskin, J. N.; Peterson, R. T.;
MacRae, C. A. Drugs that induce repolarization abnormalities
cause bradycardia in zebrafish. Circulation 2003, 107, 1355–1358.
(25) Zon, L. I.; Peterson, R. T. In vivo drug discovery in the zebrafish.
Nat. Rev. Drug Discovery 2005, 4, 35–44.
(41) Furet, P.; Imbach, P.; Furst, P.; Lang, M.; Noorani, M.; Zimmer-
mann, J.; Garcıa-Echeverrıa, C. Modeling of the binding mode of a
´
´
non-covalent inhibitor of the 20S proteasome. Application to
structure-based analogue design. Bioorg. Med. Chem. Lett. 2001,
11, 1321–1324.
(26) McGraph, P.; Li, C. Q. Zebrafish: a predictive model for assessing
drug-induced toxicity. Drug Discovery Today 2008, 13, 394–401.
(27) Parng, C. In vivo zebrafish assays for toxicity testing. Curr. Opin.
Drug Discovery Dev. 2005, 8, 100–106.
(28) Murtha, J. M.; Qi, W.; Keller, E. T. Hematologic and serum
biochemical values for zebrafish (Danio rerio). Comp. Med. 2003,
53, 37–41.
(29) Parng, C.; Roy, N. M.; Ton, C.; Lin, Y.; McGrath, P. Neurotoxi-
city assessment using zebrafish. J. Pharmacol. Toxicol. Methods
2007, 55, 103–112.
(30) Amanuma, K.; Takeda, H.; Amanuma, H.; Aoki, Y. Transgenic
zebrafish for detecting mutations caused by compounds in aquatic
environments. Nat. Biotechnol. 2000, 18, 62–65.
(42) Smith, D. M.; Daniel, K. G.; Wang, Z. G.; Guida, W. C.; Chan,
T. H.; Dou, Q. P. Docking studies and model development of tea
polyphenol proteasome inhibitors: applications to rational drug
design. Proteins: Struct., Funct., Bioinf. 2004, 54, 58–70.
(43) Chen, D.; Daniel, K. G.; Chen, M. S.; Kuhn, D. J.; Landis-
Piwowar, K. R.; Dou, Q. P. Dietary flavonoids as proteasome
inhibitors and apoptosis inducers in human leukemia cells. Bio-
chem. Pharmacol. 2005, 69, 1421–1432.
(44) Mozzicafreddo, M.; Cuccioloni, M.; Cecarini, V.; Eleuteri, A. M.;
Angeletti, M. Homology modeling and docking analysis of the
interaction between polyphenols and mammalian 20S protea-
somes. J. Chem. Inf. Model. 2009, 49, 401–409.
(45) Lei, M.; Zhao, X.; Wang, Z. L.; Zhu, Y. Q. Pharmacophore
modeling, docking studies and synthesis of novel dipeptide protea-
some inhibitors containing boron atoms. J. Chem. Inf. Model.
2009, 49, 2092–2100.
(46) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible
docking. J. Mol. Biol. 1997, 267, 727–748.
(31) Klausner, Y. S.; Bodansky, M. Coupling reagents in peptide
synthesis. Synthesis 1972, 453–463.
(32) Davies, S. G.; Hermann, G. J.; Sweet, M. J.; Smith, A. D. Double
asymmetric induction as a mechanistic probe: conjugate addition
for the asymmetric synthesis of a pseudotripeptide. Chem. Com-
mun. 2004, 1128–1129.
(33) Davies, S. G.; Mulvaney, A. W.; Russell, A. J.; Smith, A. D.
Parallel synthesis of homochiral β-amino acids. Tetrahedron:
Asymmetry 2007, 18, 1554–1566.
(34) Spitsbergen, J. M.; Kent, M. L. The state of the art of the zebrafish
model for toxicology and toxicologic pathology research;advan-
tages and current limitations. Toxicol. Pathol. 2003, 31 (Suppl.),
62–87.
(47) GOLD, version 3.0; The Cambridge Crystallographic Data Centre:
Cambridge, U.K., 2006.
(48) DeLano, W. L. The PyMOL Molecular Graphics System, version
0.99; DeLano Scientific: San Carlos, CA, 2006.
(49) Kimmel, C. B.; Ballard, W. W.; Kimmel, S. R.; Ullmann, B.;
Schilling, T. F. Stages of embryonic development of the zebrafish.
Dev. Dyn. 1995, 203, 253–310.
(35) Pallombella, V. J.; Conner, E. M.; Fuseler, J. W.; Destree, A.;
Davis, J. M.; Laroux, F. S.; Wolf, R. E.; Huang, J.; Brand, S.;
Elliott, P. J.; Lazarus, D.; Parent, L.; Stein, R.; Adams, J.;
Grisham, M. B. Role of the proteasome and NF-kappa B in
(50) Insight II; Accelrys Software Inc.: San Diego, CA, 2001.
(51) Zeng, J.; Liu, G. X.; Tang, Y.; Jiang, H. L. 3D-QSAR studies on
fluoropyrrolidine amides as dipeptidyl peptidase IV inhibitors by
CoMFA and CoMSIA. J. Mol. Model. 2007, 13, 993–1000.