Urea-containing peptide boronic acids as potent proteasome inhibitors

Article abstract of DOI:10.1016/j.ejmech.2016.10.023

A novel class of urea-containing peptide boronic acids as proteasome inhibitors was designed by introducing a urea scaffold to replace an amido bond. Compounds were synthesized and their antitumor activities were evaluated. After two rounds of optimizatio

Full text of DOI:10.1016/j.ejmech.2016.10.023

Accepted Manuscript
Urea-containing peptide boronic acids as potent proteasome inhibitors
Li-Qiang Han, Xia Yuan, Xing-Yu Wu, Ri-Dong Li, Bo Xu, Qing Cheng, Zhen-Ming
Liu, Tian-Yan Zhou, Hao-Yun An, Xin Wang, Tie-Ming Cheng, Ze-Mei Ge, Jing-Rong
Cui, Run-Tao Li
PII:
S0223-5234(16)30886-8
10.1016/j.ejmech.2016.10.023
EJMECH 8988
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 3 August 2016
Revised Date: 9 October 2016
Accepted Date: 12 October 2016
Please cite this article as: L.-Q. Han, X. Yuan, X.-Y. Wu, R.-D. Li, B. Xu, Q. Cheng, Z.-M. Liu, T.-Y.
Zhou, H.-Y. An, X. Wang, T.-M. Cheng, Z.-M. Ge, J.-R. Cui, R.-T. Li, Urea-containing peptide boronic
acids as potent proteasome inhibitors, European Journal of Medicinal Chemistry (2016), doi: 10.1016/
j.ejmech.2016.10.023.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical abstract

ACCEPTED MANUSCRIPT
Urea-containing peptide boronic acids as potent proteasome inhibitors
a
a
a
a
a
a
Li-Qiang Han , Xia Yuan , Xing-Yu Wu , Ri-Dong Li , Bo Xu , Qing Cheng ,
a
a
b
a
a
Zhen-Ming Liu , Tian-Yan Zhou , Hao-Yun An , Xin Wang , Tie-Ming Cheng ,
Ze-Mei Ge a,*, Jing-Rong Cui a,*, Run-Tao Li a ,*
aState Key Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences, Peking University,
Beijing 100191, PR China
bZhengzhou Granlen PharmaTech, Ltd. 1300 East Hanghai Road, Zhengzhou, Henan 450016, PR China
Abstract
A novel class of urea-containing peptide boronic acids as proteasome inhibitors was
designed by introducing a urea scaffold to replace an amido bond. Compounds were
synthesized and their antitumor activities were evaluated. After two rounds of
optimizations, the compound I-14 was found to be a potent proteasome inhibitor.
Compared with Bortezomib, I-14 showed higher potency against the
chymotrypsin-like activity of human 20S proteasome (IC50 < 1 pM), similar potency
against four different cancer cell lines (IC50 < 10 nM), and better pharmacokinetic
profile. Furthermore, I-14 significantly inhibited tumor growth in Bel7404 mouse
xenograft model. The excellent proteasome inhibition by I-14 was rationalized
through docking and molecular dynamics studies.
Keywords:
Urea-containing peptide boronic acids; Proteasome inhibitor; Anti-tumor activity;
Structure-activity relationship; Low toxicity
1. Introduction
Proteasome is the most abundant and ubiquitous enzymatic complex responsible for
intracellular protein turnover in eukaryotic cells [1]. Its substrates include a large
variety of cell proteins that control apoptosis, cell-cycle progression, DNA repair and
many other cellular regulatory mechanisms [2], and therefore it plays an important
role in the regulation of many physiological processes as well as in the development
of a number of major human diseases. The catalytic core particle of the 20S
proteasome has become a major target for anticancer and anti-autoimmune drugs [3].
The peptide boronate Bortezomib (Velcade®) is the first 20S proteasome inhibitor
entering clinical practice, approved by FDA for treating multiple myeloma in 2003
and mantle cell lymphoma in 2006 [4]. Since then, a series of proteasome inhibitors
have advanced to clinic or clinical trials, including peptide boronic acids
*Corresponding author. Tel./fax: +86 10 82805954.
E-mail address: lirt@bjmu.edu.cn (R.-T. Li), zmge@bjmu.edu.cn (Z.-M. Ge),
jrcui@bjmu.edu.cn (J.-R. Cui)
1

ACCEPTED MANUSCRIPT
(Delanzomib, CEP-18770) [5], epoxyketones (Carfilzomib, approved in July 2012
for treating multiple myeloma) [6], β-lactones such as NPI-0052 (Marizomib) [7], etc.
(Fig. 1). Although peptide boronic acids are one of the most important chemical
classes of proteasome inhibitors, most of them have poor pharmacokinetic properties,
which greatly limited their clinical applications [8]. Even for Bortezomib, after bolus
i.v. administration at 0.8 mg/kg, the mean half-life, the mean apparent volume of
distribution (Vd) and the plasma clearance rate (CL) were only 0.8 ± 0.1 h, 2.4 ± 0.8
L/kg and 34 ± 11 mL/min/kg, respectively [9], and in the initial pharmacokinetic
studies, it was found to be rapidly metabolized in the plasma compartment (>90% is
cleared within 15 minutes of i.v. administration) [10]. These results could be partly
attributed to the instability of the peptide bonds in the molecule [11]. However, to date,
current efforts in improving peptide boronic acid proteasome inhibitors mostly focus
on the modifying the amino acid residues and the N-end substituents [12-17], but
rarely changing the amido bonds.
Fig 1. Structures of Delanzomib, Carfilzomib and Marizomib
Replacing the amido bond in the peptide with the urea framework is a common
strategy known to improve the PK/PD properties of peptide drugs. Besides, the urea
scaffold has been exploited in several other classes of proteasome inhibitors (Fig. 2)
[18-20].
Fig 2. Reported peptide inhibitors containing the urea scaffold.
Inspired by this, we designed and synthesized a novel class of urea-containing
peptide boronic acids as proteasome inhibitors, which have both the urea scaffold and
peptide boronic acid as warheads. Further optimization led to the discovery of a
potent proteasome inhibitor I-14 with excellent in vitro and in vivo antitumor
activities, low toxicity and good pharmacokinetic properties.
2. Results and discussion
2

ACCEPTED MANUSCRIPT
2.1. Design, synthesis and biological activities of first-round compounds
Our initial studies started by examining the influence of distance between the urea
segment and the boronic acid group. We designed two series of compounds with urea
segment away from (I) and next to (II) the boron acid group (Fig. 3), in which n = 0
represents dipeptides and n = 1 represents tripeptides, respectively. R1, R2 and R3 are
chosen from the corresponding groups in known effective peptide boronic acid
proteasome inhibitors.
Fig 3. Structures of compounds I and II
The synthetic routes of the two series of compounds are illustrated in Scheme 1
using I-1 and II-1 as examples. Because peptide boronic acids are difficult to purify
and α-amido boronic acid is the most expensive among all intermediates, the α-amido
boronic acid segment was attached only in the last step. The key urea scaffold was
introduced based on literature method with modification [21]. For the synthesis of I-1,
benzylamine and L-phenylalanine methyl ester hydrochloride were reacted with
1,1-carbonyldiimidazole (CDI) to form the urea part I-1a in 87% yield. I-1a was
further treated with NaOH and then with HCl to give the corresponding hydrolyzed
product I-1b. In the presence of dicyclohexylcarbodiimide (DCC) and
N-hydroxybenzotriazole (HOBt), I-1b was coupled with α-amino boronic acid pinacol
ester, which was synthesized according to Ellman’s procedure [22], to provide key
intermediate urea-containing peptide boronic acid pinacol ester I-1c. The target
compound I-1 was obtained after the pinacol protecting group of I-1c was removed
using a novel method developed by Santos and coworkers [23]. For the synthesis of
compound II-1, the starting material N-Boc-L-phenylalanine was first condensed with
4-methoxybenzylamine in the presence of DCC and HOBt to give the protected
intermediate II-1a, which was treated with HCl to generate the corresponding
deprotected product II-1b. The crude II-1b was coupled with α-amino boronic acid
pinacol ester by the same method as described above for I-1a to give the
urea-containing peptide boronic acid pinacol ester II-1c. Finally, the pinacolyl
protecting group of II-1c was removed using the same procedure as used for
obtaining I-1 to provide the free boronic acid II-1.
Compounds I-2~I-10 and II-2~II-7 were synthesized similarly through the same
1
procedures. All key intermediates and final products were characterized by H NMR,
13C NMR and MS.
3

ACCEPTED MANUSCRIPT
O
NH2
b
a
O
O
+
O
O
OH
N
N
N
N
NH2HCl
H
H
H
H
O
O
I-1a
I-1b
O
H
O
N
H
N
H
N
N
d
H
N
N
c
O
B
H
H
HO OH
O
B
O
O
O
B
I-1
I-1c
O
NH2HCl
O
H
H
O
N
N
H
H
N
N
N
H
d
N
O
B
H
MeO
HO
OH
O
B
MeO
O
O
a
II-1
II-1c
NH2
MeO
MeO
c
e
+
H
N
H
HO
Boc
Boc
N
N
H
N
H
NH2HCl
MeO
O
O
O
II-1a
II-1b
Scheme 1. Synthesis of target compounds I-1 and II-1. Reagents and conditions: (a) CDI, DMF, MeCN, rt; (b) i.
NaOH, 0°C; ii. HCl, 0°C; (c) DCC, HOBt, DIEA, CH2Cl2, rt; (d) i. DEA (Diethanolamine), EtOAc, rt; ii. HCl,
H2O, EtOAc, rt; (e) 10 M HCl in EtOH, CH2Cl2.
The inhibitory effects of the I-1~I-10 and II-1~II-7 on the chymotrypsin-like
(ChT-L), caspase-like (C-L), and trypsin-like (T-L) activities of human 20S
proteasome were measured at 0.2 ꢀM and 2 ꢀM using Bortezomib as the positive
control. The results are summarized in Table 1. It is clear that the series I compounds
(urea moiety away from boron atom) are more potent in inhibiting ChT-L activity than
the series II compounds (urea moiety next to boron atom). The lower potency of
series II compounds can be ascribed to the tautomeric interconversion shown in
Scheme 2. Due to the strong electrophilicity of the boron atom, a five-member ring is
easily formed through a coordination bond and then one molecule of water is
eliminated [24]. This speculation is supported by the ESI results of the series II
compounds, which mainly give the [M-17]+ peaks (i.e., [M−H2O+1]+, see
Supplementary Material, Figure S1). Comparing the activities of compounds I-3 with
I-9, and II-3 with II-7 shows that tripeptide analogues are obviously better than
dipeptide analogues, which is consistent with our previous report [14]. Interestingly,
compound II-5 with the unnatural D-amino acid does not show distinct difference
from the corresponding its L-configuration amino acid compounds analogue II-3. In
addition, I-6 and I-10 have equal potency against ChT-L compared with Bortezomib.
However, they are ineffective or less potent than Bortezomib against C-L and T-L,
suggesting that they are effective and selective proteasome inhibitors. More
interestingly, compounds (I-1, I-2, I-5 and I-6) that have benzyl or substituted benzyl
group as R1 exhibited more potent anti-prote4asome activity than other compounds.

ACCEPTED MANUSCRIPT
Scheme 2. Predicted interconversion of compounds in series II
Table 1
General structures and biological activity of target compounds (% inhibition).
HO OH
O
B
H
H
N
N
R1
N
nH
O
R2
Comp.
No.
I
ChT-L
C-L
T-L
n
R1
R2
R
0.2 ꢀM
82.082
86.338
2 ꢀM
96.113
95.861
0.2 ꢀM
6.461
8.450
2 ꢀM
84.58
0.2 ꢀM
5.043
5.722
2 ꢀM
57.790
53.491
I-1
I-2
1
1
89.392
I-3
I-4
1
1
-14.639
10.548
86.614
87.304
-14.446
3.748
-2.926
8.575
-2.888
-3.766
12.266
10.537
5

ACCEPTED MANUSCRIPT
I-5
I-6
1
1
82.067
91.940
96.655
9.814
68.776
4.990
56.881
83.660
96.544
-23.048
-26.615
35.572
I-7
I-8
I-9
1
1
2
-H
-H
47.308
61.963
49.489
90.946
89.674
90.992
8.577
10.065
-13.661
62.467
7.173
0.355
-3.948
-0.748
29.284
28.328
47.155
-20.492
I-10
II-1
II-2
1
1
1
84.643
33.819
-14.170
95.724
86.790
10.372
-21.838
40.869
2.030
-24.724
87.942
2.190
-0.926
-6.549
6.916
43.457
22.848
4.381
6

ACCEPTED MANUSCRIPT
II-3
II-4
II-5
II-6
II-7
1
1
1
1
1
-0.670
2.873
-4.193
5.785
44.666
17.405
1.617
66.342
7.355
-3.924
-1.392
-12.27
-8.060
2.409
9.321
59.740
47.804
52.830
93.361
18.952
12.760
79.767
32.649
73.204
-5.926
-5.632
34.743
63.305
91.554
9.760
-7.933
33.278
Bortezomib
55.563
7

ACCEPTED MANUSCRIPT
2.2. Design, synthesis and biological activities of second-round compounds
The first round showed that series I compounds (urea away from boronic acid)
were much more potent than series II compounds (urea next to boronic acid) and it is
preferable to have a benzyl group at the N-terminal. Thus, we next focused our
attention on the optimization of the benzyl group at the N-terminal in series I
compounds. A series of new analogues were designed and synthesized through the
same methodology described above, except that compound I-24 was a side product
[25] obtained during the preparation of I-14. All these second-round compounds
I-11~I-24 were evaluated for their proteasome inhibitory activity against ChT-L and
their potency against the hepatocellular carcinoma HepG-2 and human gastric cancer
MGC-803 cell lines. The structures and results are listed in Table 2.
Table 2.
Structure and biological activity (IC50 in nM) of second-round compounds.
Comp. No.
I-1
R
ChT-L
10 ± 3
HepG-2
MGC-803
18.76 ± 7
87.25 ± 14
I-11
I-12
20 ± 0.6
10 ± 1
70.86 ± 9
36.99 ± 4
30.01 ± 6
4.084 ± 1
I-13
I-14
I-15
200 ± 20
0.0002 ± 0.0001
2 ± 0.8
1072 ± 161
19.38 ± 3
30.49 ± 2
198.2 ± 14
3.962 ± 1
5.203 ± 1
I-16
I-17
4 ± 2
44.68 ± 3
6.298 ± 1
50 ± 6
4136 ± 479
738.2 ± 111
8

ACCEPTED MANUSCRIPT
I-18
40 ± 3
70.67 ± 14
28.87 ± 8
I-19
I-20
I-21
0.3 ± 0.07
20 ± 10
12 ± 1
37.25 ± 1
45.60 ± 2
53.08 ± 14
5.873 ± 1
5.316 ± 2
7.062 ± 0
I-22
I-23
2 ± 0.9
3 ± 1
38.82 ± 1
43.81 ± 3
6.698 ± 0
6.317 ± 0
I-24
660 ± 70
4 ± 0.3
59.67 ± 4
36.57 ± 3
7.298 ± 0
5.782 ± 0
Bortezomib
As shown in Table 2, all tested compounds exhibited excellent ChT-L inhibitory
activity (IC50 varying from <1 nM to 50 nM) except the benzylic methyl substituted
compound I-13 and the benzylic oxidized derivative I-24. In particular, compound
I-14 reached an IC50 of 0.0002 nM against ChT-L, which is much more potent than
Bortezomib (IC50 = 4.0 nM). Extending the benzyl amino (I-1) to phenylethyl amino
(I-11) did not significantly improve the ChT-L inhibitory activity. Compounds I-12
(IC50 = 10 nM) and I-13 (IC50 = 200 nM) both introduced a methyl group at the
methylene position in I-1 but showed 20 times difference in their activity against
ChT-L, suggesting that the chiral center can greatly influence the activity. However,
we did not further optimize the substituents at methylene position of compound I-1,
since neither I-12 nor I-13 was significantly better than I-1. The N-methyl substituted
derivative I-15 (IC50 = 2 nM) and the cyclized 1,2,3,4-tetrahydroisoquinoline
derivative I-14 (IC50 = 0.0002 nM) showed very strong inhibitory activity against
ChT-L proteasome, indicating that replacing the hydrogen in the amino group with an
alkyl group is favorable, and restricting the conformation of the amine also helps
greatly. The type and position of substituents on the phenyl ring of the benzylamino
group obviously can affect the activity. For example, I-18 (2,5-dichloro, IC50 = 40 nM)
showed 15–20 times weaker potency in ChT-L inhibition than I-16 (2,3-dimethyl,
IC50 = 4 nM), I-22 (3,4-difluoro, IC50 = 2 nM) and I-23 (3-nitro, IC50 = 3 nM). More
notably, the 3-phenyl analogue I-19 showed exceptional ChT-L inhibitory activity
(IC50 = 0.3 nM), suggesting that the aryl substituent is more favorable than others.
However, using other aromatic moieties such as naphthyl (I-21), pyrazinyl (I-17) and
furanyl (I-20) groups in the place of the phenyl ring of the benzylamino group in I-1
9

ACCEPTED MANUSCRIPT
did not improve the resulting compounds' potency on inhibiting ChT-L activity.
In addition, all compounds except I-13 and I-17 also produced excellent potency
against HepG-2 and MGC-803 cells lines (IC50 < 100 nM), among which four
compounds (I-12, I-14, I-15 and I-19) had comparable or higher potency than
Bortezomib. The most potent compound was still I-14. The diastereoisomers I-12
(HepG-2, IC50 = 39.99 nM; MGC-803, IC50 = 14.084 nM) and I-13 (HepG-2, IC50
=
1072 nM; MGC-803, IC50 = 198.2 nM) again had dramatically different potency,
further illustrating the importance of chirality on that benzylic position. Note that I-24
had satisfying potency against both cancer cell lines but fell short to effectively
suppress proteasome ChT-L activity. The rationale behind this phenomenon is under
further investigation due to the novel structure of I-24.
Based on the above results, three potential compounds (I-6, I-14 and I-22) were
selected to be tested further against 11 human tumor cell lines using Bortezomib
(PS-341) as the positive control. The results are summarized in Table 3.
Table 3.
Biological results of three compounds over 11 human tumor cell lines (IC50 in nM) a
No.
Cell line
Source
I-6
I-14
I-22
PS-341
1
2
3
4
5
6
7
8
9
A549
95D
Lung cancer
Lung cancer
Colon cancer
Leukemia
1022.61 ± 265.45
40.96 ± 0.43
23.02 ± 0.16
26.96 ± 1.85
17.52 ± 2.14
42.23 ± 1.65
20.11 ± 1.40
46.30 ± 7.32
25.08 ± 1.36
3826.15 ± 796.42 1317.50 ± 15.43 2135.73 ± 463.00
37.11 ± 0.10
7.32 ± 0.79
7.71 ± 0.01
4.98 ± 0.06
13.75 ± 1.09
15.52 ± 2.54
25.88 ± 9.60
17.74 ± 1.02
58.85 ± 7.59
16.07 ± 3.05
28.53 ± 2.03
16.73 ± 0.13
47.26 ± 0.78
26.62 ± 0.25
10.47 ± 3.62
28.40 ± 2.49
38.18 ± 2.39
1.40 ± 0.17
7.60 ± 0.53
6.54 ± 0.18
25.04 ± 1.76
7.15 ± 0.41
1.62 ± 0.97
16.65 ± 0.67
HCT116
HL-60
MGC803
BEL7404
MKN45
SKOV3
Gastric cancer
Hepatoma
Gastric cancer
Ovarian cancer
Breast cancer
Hepatocellular
carcinoma
MDA-MB-231
10
11
HepG2
25.96 ± 5.52
8.38 ± 0.78
19.04 ± 1.90
90.73 ± 3.03
1.61 ± 0.17
Pancreatic
SW1990
80.33 ± 7.92
89.34 ± 4.51
79.06 ± 2.32
carcinoma
a IC50 was measured over 6 concentrations based on three experiments.
Table 3 shows that all three compounds strongly inhibited the growth all tested
tumor cell lines (except A549) at IC50 < 100 nM. In particular, I-14 reached IC50 < 10
nM in four cell lines, making it comparable to Bortezomib. The three candidate
compounds and Bortezomib all showed inadequate potency to the A549 cell line than
to the other ten cell lines: their IC50 values were all 10–2000 times greater for A549
than for others. The bonoric acid is thus a selective anticancer agent.
2.3. Pharmacokinetics study of I-14
Given that the majority of currently available proteasome inhibitors containing
peptide backbones are prone to rapid in vivo inactivation [26], we next examined
1 0

ACCEPTED MANUSCRIPT
whether the urea scaffold-based proteasome inhibitor I-14 is metabolically more
stable than those peptide-based drugs. Thus, we tested the pharmacokinetics of I-14 in
three rats. The observed plasma concentration–time profile after a single i.v. dose of 3
mg/kg in male Sprague–Dawley rats is shown in Fig. 4 and the main pharmacokinetic
parameters estimated by WinNolin are summarized in Table 4, in which the half-life
(t1/2) of I-14 is 1.96 ± 0.07 h. In contrast, it was reported that after bolus i.v.
administration of Bortezomib at 0.8 mg/kg, the mean half-life was 0.8 ± 0.1 h,10
which is much shorter than that of I-14. At the same time, the mean volume of
distribution (Vd) of Bortezomib was 2.4 ± 0.8 L/kg, and we can infer that the long
elimination phase half-life of I-14 is in part due to a larger volume of distribution, Vd
= 9.44 ± 1.21 L/kg. These results indicate that I-14 is more likely to have higher in
vivo stability than Bortezomib, as the latter is known to undergo rapid metabolic
inactivation [11, 27, 28].
Fig 4. The plasma concentration-time curve of I-14 in male SD rats following bolus i.v. administration at a dose of
3 mg/kg (n=3). Data are mean ± SD.
Table 4
Pharmacokinetic parameters of I-14 in Rats after a bolus i.v. dose of 3 mg/kg (n=3)
Animal
Rat 1
Rat 2
Rat 3
Mean
SD
t1/2 (hr)
1.93
958
2.04
788
1.91
887
1.96
878
0.07
85
AUClast (hr*ng/mL)
Vd (L/kg)
8.45
50.5
10.8
61.2
9.07
54.8
9.44
55.5
1.21
5.4
CL (mL/min/kg)
2.4. In vivo anticancer activity of I-14
To assess the in vivo antitumor efficacy of compound I-14, a Bel7404 xenograft
nude mice model was established, with Bortezomib (PS-341) as positive drug.
Compound I-14 was administered at 1 mg/kg i.v. every two days for 24 consecutive
days. As shown in Fig. 5A, I-14 could effectively suppress tumor growth, which is
equivalent to Bortezomib (0.5 mg/kg). Gratifyingly, the average body weight of the
11

ACCEPTED MANUSCRIPT
mice in the I-14 treatment group did not drop significantly, whereas the group treated
with Bortezomib showed a 10.7% loss of body weight, and even two mice died on
day 20 and 24 respectively. These results revealed that I-14 is an efficient anticancer
agent with relatively low toxicity.
Figure 5. I-14 inhibits tumor growth in vivo at 1 mg/kg. (A) Tumor growth curves in mice receiving the respective
treatments by tumor volume (Black: control, n = 8; Green: I-14 1 mg/kg i.v. qod, n = 6; Red: Bortezomib 0.5
mg/kg i.v. qod, n = 6). (B) Average body weight of mice (percentage of initial) receiving the respective treatments
surviving on a given day. The results are represented as means ± SD and analyzed by two-way ANOVA. n.s.
non-significant, *** P<0.001.
2.5. Docking and molecular dynamics study of I-14
To understand the binding profile and account for the superior bioactivity of I-14
than PS-341, we performed a molecular docking study via Gold (Cambridge
Crystallographic Data Centre) and assessed the results by Maestro (Schrödinger
LLC.). The gene sequence of the chymotrypsin active site of the yeast 20S
proteasome is highly homologous to that of the human 20S proteasome. The best
binding conformation of PS-341 has a RMSD of 1.0 Å with its original crystallized
conformation.
The binding profile of I-14 in the β5 and β6 subunit inhibitor pocket of the
proteasome is depicted in Fig. 6A,B. From both the 2D binding and the
Poisson–Boltzmann electrostatic surface plot, it is obvious that I-14 closely resembled
PS-341 with its peptidomimetic chain, making essential hydrogen bonds with the
receptor and fulfilling its promise as a potent proteasome inhibitor on enzymatic,
cellular and animal model aspects. From the overall docking results, I-14, with a
fitness score of 51.03, is better than PS-341, which scored 46.37. This is mainly due
to I-14’s higher van der Waals interaction score (I-14: 41.81, PS-341: 36.48) together
with less score penalty from internal torsion energy (−3.02 vs. −3.34) and covalent
binding energy (−10.90 vs. −11.63). Indeed, the induction of the aromatic ring could
contribute to the binding affinity, as there are more hydrophobic residues located
around the 1,2,3,4-tetrahydroisoquinolin fragment. Meanwhile, the phenyl group in
I-14 rotated slightly and occupied the S2 subsite better than that of PS-341 (Fig. 6B).
1 2

ACCEPTED MANUSCRIPT
Fig. 6. Docked binding profile of I-14 and PS-341 within the β5 and β6 subunit inhibitor pocket of yeast 20S
proteasome. (A) 2D binding plot: Purple arrows and colored curves indicate hydrogen bondings and hydrophobic
interactions respectively. (B) Poisson-Boltzmann electrostatic surface plot: PS-341 in its crystallized bioactive
conformation is shown in grey sticks, best binding conformation of I-14 is displayed in green, oxygen atom from
the crystallized water molecule is depicted in red dot.
However, the docking results are not enough to illuminate how and why I-14
outperformed PS-341 on the enzymatic assay. We thus utilized Desmond to conduct a
30 ns unrestrained molecular dynamics simulation to find out. The RMSD plot of the
heavy atoms throughout the simulation is shown in Fig. 7A. The trajectory is stable
from 10 ns onward in accordance with the energy terms and the geometry. A
representative conformation (RC) from 10 to 30 ns and the last frame (LF) were
obtained as depicted in Fig. 7C. The essential hydrogen bonding derived from the
docking conformation was retained throughout the simulation, although the hydrogen
bonding tended to be more frequent after 10 ns (Fig. 7B, Supplementary Material, Fig.
S2 B,D,E). Among the most significant increases are I-14-NH15 to K:GLY47 (from
40% in first 10 ns to 98% in 10–30 ns) and K:ALA49 to I-14-O19 (from 76% to 95%).
Two new pairs of water bridge hydrogen bonding also happen in 37% of the time
during the equilibrium.
The non-bonding fragments of the molecule fits the pocket better after 10 ns,
leading to an increase of the binding free energy (ꢁGbind) of separate parts (Figure
7D). During the first 7 ns, we could see that the 1,2,3,4-tetrahydroisoquinolin
fragment of I-14 was gradually engulfed by the pocket and then stabilized over the
final stage of the simulation, which is a distinct induced-fit effect that we were
expecting. This phenomenon could help to diminish the fragment’s solvent exposure
(Figure 6B, 7C), which potentiates its van der Waals interaction, hydrophobic
interaction and steric hindrance, and thus contributes to its total ꢁGbind
.
In conclusion, the peptidomimetic chain of I-14 highly mimicked that of PS-341.
The receptor and the 1,2,3,4-tetrahydro-isoquinolin fragment of I-14 underwent a
induced-fit effect at the initial stage of the simulation, and whole molecule fit and
interacted with the receptor better as the simulation proceeded. This can help in
explaining the superior activity of I-14 on the enzymatic aspect, confirming its
potential as a promising drug candidate.
1 3

ACCEPTED MANUSCRIPT
Fig. 7. Molecular dynamics profile of I-14 within the β5 and β6 subunit inhibitor pocket of yeast 20S proteasome.
(A) RMSD of all of the heavy atoms of the complex versus time. (B) 2D interaction graph in 10 to 30 ns.
Percentage refers to the frequency of valid corresponding hydrogen bondings throughout the simulation. (C) I-14
(green sticks) in the inhibitor pocket. Above: Representative conformation (RC) during 10 to 30 ns. Below: Last
frame (LF) of the simulation. (D) MM-GBSA chart. Fragments of I-14 (above) are marked in different colors and
letters. Energy contributions (kcal/mol) of the corresponding fragment are listed in the chart below. SF: Starting
Frame of the simulation. VdW: Van der Waals interaction. HI: Hydrophobic Interaction.
3. Conclusion
Known drawbacks of Bortezomib and its analogues with peptide backbone include
their rapid metabolism, tendency to develop resistance, and dose-limiting toxicities.
To overcome these problems, we designed and synthesized a series of urea-containing
peptide boronic acids as potent proteasome inhibitors. Most of the synthesized
compounds exhibited excellent potency in inhibiting proteasome activities and tumor
cell growth. Based on two rounds of investigations, three compounds were selected
and their cellular activities were evaluated against eleven tumor cell lines. They all
enabled inhibition (IC50 < 100 nM) of ten out of eleven tested tumor cell lines. The
compound I-14 was also tested for its in vivo anti-tumor activity with the Bel7404
14

ACCEPTED MANUSCRIPT
xenograft nude mice model. The results showed that compared with Bortezomib, I-14
demonstrated similar in vivo efficacy and significantly lower toxicity in rats.
Pharmacokinetic profiles suggested that I-14 is metabolically more stable than
Bortezomib as expected originally. The docking and molecular dynamics studies
revealed the interaction mode between I-14 and the 20S proteasome, and the results
may inspire the design of more potent proteasome inhibitors. Additional preclinical
profiling studies of I-14 are currently in progress.
4. Experiment
4.1. General information
All commercial reagents were purchased from commercial suppliers and used
without further purification. Dry solvents were prepared according to standard
procedures. Melting points were taken with an X4 apparatus and were uncorrected.
Yields refer to chromatographically pure compounds unless otherwise stated.
1H-NMR spectra were recorded in CDCl3 and DMSO-d6 at ambient temperature on a
Bruker Avance III 400 MHz system. 13C NMR spectra were recorded in DMSO-d6 at
ambient temperature unless otherwise noted, at 100 MHz. Silica gel column
chromatography was performed with silica gel 60 N (spherical, neutral, 63-210 ꢀm, or
Merck). MS data were obtained with MDSSCIEX QSTAR systems. Compounds are
evaluated for >95% purity using HPLC analysis (methanol/water) using a C18
column.
4.2. General procedures for the synthesis of compounds I-1 and II-1
4.2. 1. (S)-Methyl 2-(3-benzylureido)-3-phenylpropanoate (I-1a)
To a solution of N,N'-carbonyldiimidazole (CDI, 1.64 g, 10.1 mmol) in DMF (8 mL)
and acetonitrile (40 mL) was added L-phenylalanine methyl ester hydrochloride (2.0 g,
9.3 mmol) in portions. The solution was stirred at room temperature for 2 h.
Benzylamine (1 g, 9.3 mmol) was then added, followed by the addition of
triethylamine (1.72 mL, 18.6 mmol), and the aggregation system began to become
clear. The reaction mixture was stirred at room temperature for 24 h until completion
as monitored by TLC, and then concentrated under reduced pressure. The residue was
dissolved in ethyl acetate (50 mL) and washed with brine (50 mL×3), dried over
anhydrous Na2SO4 and concentrated in vacuo. The crude material was recrystallized
from ethyl acetate to afford the product I-1a as a white solid (1.22 g, 43% yield); mp:
93–95 °C. 1H NMR (400 MHz, CDCl3) δ 7.18 (m, 8H), 7.2 (d, J = 4 Hz, 2H), 5.78 (m,
2H), 4.64 (dd, J = 14.3, 6.3 Hz, 1H), 4.14 (tt, J = 15.2, 7.8 Hz, 2H), 3.52 (s, 3H), 2.88
(ddd, J = 38.2, 13.8, 6.3 Hz, 2H). 13C NMR (101 MHz, CDCl3) δ 173.54, 157.91,
139.41, 136.49, 129.34, 128.49, 128.41, 127.28, 127.05, 126.84, 54.23, 52.05, 44.02,
38.54.
4.2. 2. (S)-2-(3-Benzylureido)-3-phenylpropanoic acid (I-1b)
I-1a (1.22 g, 3.9 mmol) was dissolved in 10 mL of acetone, and 2 N NaOH
aqueous solution was added dropwise at 0 °C until TLC showed methyl ester I-1a was
saponified completely (approximately 2 h). 1Th5e reaction solution was acidified to pH

ACCEPTED MANUSCRIPT
2 with 3 N HCl. Upon standing, the precipitate was filtered and washed with water
until pH value reached 6–7. I-1b (0.9 g, 78%) was obtained and used in next step
without further purification.
4.2.3.
(S)-2-(3-Benzylureido)-N-((R)-3-methyl-1-(4,4,5,5-tetramethyl-1,3,2-dioxa-
borolan-2-yl)butyl)-3-phenylpropanamide (I-1c)
To a stirred suspension of I-1b (0.9 g, 3.0 mmol) in CH2Cl2 (50 mL) were added
dicyclohexylcarbodiimide (DCC, 0.74 g, 3.6 mmol) and 1-hydroxybenzotriazole
monohydrate (HOBt, 0.45 g, 3.3 mmol). Pinacol leucine boronate hydrochloride 1
(0.75 g, 3.0 mmol, prepared according to the literature21) and DIPEA (0.52 mL, 3.0
mmol) were then added. The mixture was stirred at room temperature for 24 h and
then filtered to remove the insoluble material. The solvent was removed under
vacuum, and the residue was dissolved in 50 mL of ethyl acetate. The organic layer
was successively washed with 10% citric acid, 5% NaHCO3 and brine. The organic
phase was dried over anhydrous Na2SO4. The solvent was evaporated to give product
I-1c as a faint yellow foam (1.23 g, crude with impurities) and it was used for next
step without further purification.
4.2.4. ((R)-1-((S)-2-(3-Benzylureido)-3-phenylpropanamido)-3-methylbutyl)-bor-
onic acid (I-1)
To the solution of I-1c obtained above in ethyl acetate (40 mL) was added
diethanolamine (0.2 mL, 1.1 equiv) dropwise, and the reaction mixture was stirred at
room temperature until completion as monitored by TLC. The resulted precipitate was
filtered, washed with ethyl acetate, and then taken up with a mixture of ethyl acetate
(20 mL) and water (20 mL). To this suspension was added 4N hydrochloric acid under
stirring and it was stirred for an additional 2 h. The organic layer was separated and
washed with brine (50 mL×3), and dried over anhydrous Na2SO4. The organic solvent
was removed under reduced pressure to provide product I-1 (570 mg, 15% in overall)
as a white foam. mp: 133–136 °C. HPLC indicates a purity of 95.2 area %. 1H NMR
(400 MHz, DMSO-d6) δ 7.41–7.14 (m, 9H), 7.17–7.02 (m, 2H), 6.51 (s, 1H), 6.35 (dd,
J = 8.4, 4.6 Hz, 1H), 4.70–4.52 (m, 1H), 4.16 (ddd, J = 20.4, 15.4, 5.8 Hz, 2H), 3.00
(dd, J = 8.9, 4.1 Hz, 1H), 2.91–2.77 (m, 1H), 2.68 (s, 1H), 1.60 (dd, J = 13.0, 6.6 Hz,
1H), 1.32 (ddd, J = 26.5, 15.3, 6.7 Hz, 2H), 0.84 (d, J = 6.2 Hz, 6H). 13C NMR (101
MHz, DMSO-d6) δ 174.88, 157.71, 140.93, 137.53, 129.86, 128.61, 128.50, 127.29,
126.96, 126.80, 43.25, 38.96, 25.57, 23.48, 23.20, 23.02, 21.51. HRMS (ESI) calcd
for C23H31BN3O3: 408.24571 [(M-H2O+CH2+H) +], found 408.24429.
4.2.5.
carbamate (II-1a)
To a solution of N-Boc-L-phenylalanine (2.65 g, 10.0 mmol) in THF (50 mL) was
added dicyclohexylcarbodiimide (DCC, 2.47 g, 12.0 mmol) and
(S)-tert-Butyl
(1-((4-methoxybenzyl)amino)-1-oxo-3-phenylpropan-2-yl)
1-hydroxybenzotriazole monohydrate (HOBt, 1.48 g, 11.0 mmol). The reaction
mixture was allowed to stirred for 30 min before 4-methoxybenzylamine (1.3 mL,
10.0 mmol) was added, followed by the addition of N-methylmorpholine (1.32 mL,
1 6

ACCEPTED MANUSCRIPT
12.0 mmol). The mixture was stirred for another 24 h, filtered and the solvent was
removed under vacuum. The residue was dissolved in 50 mL of ethyl acetate and
washed with 10% citric acid, 5% NaHCO3 and brine, then dried over anhydrous
Na2SO4. The crude product was purified by silica gel chromatography (ethyl acetate:
petroleum ether = 1:3) to afford the desired product II-1a (2.66 g, 6.9 mmol). mp:
68–71 °C. 1H NMR (400 MHz, CDCl3) δ 7.33–7.11 (m, 5H), 7.00 (d, J = 7.1 Hz, 2H),
6.77 (d, J = 8.5 Hz, 2H), 6.40 (s, 1H), 5.26 (d, J = 6.7 Hz, 1H), 4.40 (d, J = 17.7 Hz,
1H), 4.24 (qd, J = 14.4, 5.2 Hz, 2H), 3.75 (s, 3H), 3.04 (d, J = 6.1 Hz, 2H), 1.36 (s,
13
9H). C NMR (101 MHz, CDCl3) δ 171.15, 158.93, 155.49, 136.81, 129.90, 129.37,
128.99, 128.60, 126.83, 113.95, 80.06, 55.26, 42.88, 38.77, 33.97, 28.27.
4.2.6.
(S)-2-Amino-N-(4-methoxybenzyl)-3-phenylpropanamide
hydrochloride
(II-1b)
To a stirring solution of II-1a (2.66 g, 6.9 mmol) in 10 mL of CH2Cl2 was added
dropwise a solution of 10 M HCl in EtOH (2 mL, 21 mmol) at 0 °C. The reaction
mixture was stirred until the starting material was completely consumed. The formed
precipitate was filtered, washed with CH2Cl2 and dried in vacuum to give II-1b
without further purification.
4.2.7.
(S)-N-(4-Methoxybenzyl)-2-(3-((R)-3-methyl-1-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)butyl)ureido)-3-phenylpropanamide (II-1c)
To a stirring solution of N,N'-carbonyldiimidazole (CDI, 0.54 g, 3.3 mmol) in DMF
(6 mL) and acetonitrile (30 mL) was added the pinacol leucine boronate
hydrochloride 1 (0.75 g, 3.0 mmol) by portions. The solution was stirred at room
temperature for 2 h. Then II-1b (0.96 g, 3 mmol) was added, followed by the addition
of triethylamine (0.83 mL, 6 mmol). The stirring was continued for another 24 h, and
acetonitrile was removed under reduced pressure. The residues was dissolved in ethyl
acetate (50 mL) and washed with brine (50 mL×3), dried over Na2SO4. The solvent
was evaporated to give a grey foam product II-1c (1.2 g, crude with impurities) which
was used in next step without further purification.
4.2.8. ((R)-1-(3-((S)-1-((4-Methoxybenzyl)amino)-1-oxo-3-phenylpropan-2-yl)ureido)-3-methyl
butyl)boronic acid (II-1)
The previously obtained II-1c was dissolved in 40 mL of ethyl acetate and filtered
the possible insoluble impurities. Diethanolamine (0.2 mL, 1.1 equiv) was added
dropwise. After a few minutes, a white precipitate formed, and the reaction was
allowed to continue until the starting material was completely consumed as monitored
by TLC. The precipitate was then filtered, washed with ethyl acetate, and then taken
into a flask filled with 20 mL ethyl acetate and 20 mL H2O. To the suspension was
added 4 M HCl (1 mL). After about 2 h as judged by TLC, the organic layer was
separated and washed with brine (50 mL×3), dried over Na2SO4. The organic solvent
was removed under reduced pressure to provide the analytically pure product II-1
(120 mg, 29% in overall) as a white foam. mp: 128–135 °C. HPLC indicates a purity
1
of 95.8 area %. H NMR (400 MHz, DMSO-d6) δ 8.36 (s, 1H), 6.82–7.21 (m, 10H),
17

ACCEPTED MANUSCRIPT
4.34 (s, 1H), 4.17 (ddd, J = 29.3, 14.6, 5.2 Hz, 2H), 3.71 (s, 3H), 3.07–2.69 (m, 2H),
1.67–1.42 (m, 1H), 1.37–1.04 (m, 2H), 0.79 (t, J = 5.6 Hz, 6H). 13C NMR (101 MHz,
DMSO-d6) δ 158.65, 137.77, 131.28, 129.79, 128.95, 128.76, 128.53, 128.44, 126.79,
114.07, 55.49, 42.04, 41.99, 38.78, 25.78, 23.69, 23.29. HRMS (ESI) calcd for
C24H33BN3O4: 438.25629 [(M-H2O+CH2+H) +], found 438.25483.
4.2.9. ((R)-1-((S)-2-(3-(4-Methoxybenzyl)ureido)-3-phenylpropanamido)-3-methyl-
butyl) boronic acid (I-2)
Using the same procedure as I-1, white foam, 31% yield; mp: 126–128 °C. HPLC
indicates a purity of 98.8 area %. 1H NMR (400 MHz, DMSO-d6) δ 7.31–7.19 (m, 5H),
7.06 (d, J = 8.5 Hz, 2H), 6.83 (d, J = 8.6 Hz, 2H), 6.42 (s, 1H), 6.30 (d, J = 8.5 Hz,
1H), 4.63 (dd, J = 13.7, 7.9 Hz, 1H), 4.20–3.98 (m, 2H), 3.71 (s, 3H), 3.04–2.76 (m,
2H), 2.67 (d, J = 2.7 Hz, 1H), 1.59 (dd, J = 13.1, 6.5 Hz, 1H), 1.40–1.20 (m, 2H), 0.84
(d, J = 4.0 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 174.96, 158.53, 157.65, 137.61,
132.78, 129.86, 128.64, 128.50, 126.80, 114.04, 55.46, 42.76, 38.97, 25.59, 23.40,
23.27, 21.49. HRMS (ESI) calcd for C24H33BN3O4: 438.25629 [(M-H2O+CH2+H) +],
found 438.25508.
4.2.10.
((R)-1-((S)-2-(3-(3-(Benzyloxy)phenyl)ureido)-3-phenylpropanamido)-3-
methyl-butyl)boronic acid (I-3)
Using the same procedure as I-1, white foam, 26% yield; mp: 112–115 °C. HPLC
1
indicates a purity of 97.5 area %. H NMR (400 MHz, DMSO-d6) δ 8.84 (d, J = 17.7
Hz, 1H), 8.67 (d, J = 9.8 Hz, 1H), 7.52–7.16 (m, 11H), 7.10 (dt, J = 13.7, 6.8 Hz, 1H),
6.86 (d, J = 6.6 Hz, 1H), 6.57 (d, J = 8.1 Hz, 1H), 6.45 (dd, J = 19.5, 8.9 Hz, 1H),
5.04 (s, 2H), 4.77–4.57 (m, 1H), 3.12–2.87 (m, 2H), 2.72 (s, 1H), 1.59 (dt, J = 13.1,
9.2 Hz, 1H), 1.45–1.20 (m, 2H), 0.84 (d, J = 5.4 Hz, 6H). 13C NMR (101 MHz,
DMSO-d6) δ 174.37, 159.25, 154.75, 141.85, 137.64, 137.26, 129.93, 129.87, 128.84,
128.57, 128.18, 128.00, 126.91, 110.77, 108.01, 104.91, 69.52, 52.18 (m), 38.97,
33.84, 25.54, 23.60, 22.94, 21.49. HRMS (ESI) calcd for C29H35BN3O4: 500.27202
[(M-H2O+CH2+H) +], found 500.27025.
4.2.11. ((R)-3-Methyl-1-((S)-3-phenyl-2-(3-(pyrazin-2-yl)ureido)propanamido)butyl)
-boronic acid (I-4)
Using the same procedure as I-1, yellow-white foam, 17% yield; mp: 135–138 °C.
1
HPLC indicates a purity of 97.2 area %. H NMR (400 MHz, DMSO-d6) δ 9.54 (s,
1H), 8.86 (d, J = 14.4 Hz, 1H), 8.76 (d, J = 6.0 Hz, 1H), 8.17 (dd, J = 9.2, 5.6 Hz, 2H),
7.94 (d, J = 6.6 Hz, 1H), 7.22 (dt, J = 27.6, 4.1 Hz, 5H), 4.82–4.62 (m, 1H), 3.14–2.90
(m, 2H), 2.64 (br, 1H), 1.56 (dt, J = 13.1, 6.6 Hz, 1H), 1.39–1.12 (m, 2H), 0.78 (dd, J
= 6.1, 2.6 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 173.89, 154.06, 150.00, 141.47,
137.59, 136.92, 135.47, 130.00, 128.59, 127.07, 52.47, 38.61, 25.48, 23.49, 23.01,
22.93. HRMS (ESI) calcd for C20H27BN5O3: 396.22050 [(M-H2O+CH2+H) +], found
396.21998.
1 8

ACCEPTED MANUSCRIPT
4.2.12.
((R)-1-((S)-2-(3-(2,5-Dimethoxybenzyl)ureido)-3-phenylpropanamido)-3-
methyl-butyl)boronic acid (I-5)
Using the same procedure as I-1, white foam, 28% yield; mp: 156–159 °C. HPLC
indicates a purity of 95.2 area %. 1H NMR (400 MHz, DMSO-d6) δ 7.23–7.10 (m, 5H),
6.83 (d, J = 8.8 Hz, 1H), 6.76–6.68 (m, 2H), 4.30 (dt, J = 13.9, 6.2 Hz, 0H), 4.08 (d, J
= 5.0 Hz, 2H), 3.97 (m, 2H), 3.69 (s, 3H), 3.63 (s, 3H), 3.10–2.83 (m, 2H), 2.81–2.65
(m, 1H), 1.50–1.14 (m, 1H), 0.90–0.58 (dd, 6H). 13C NMR (101 MHz, DMSO-d6) δ
171.93, 158.22, 153.46, 151.14, 137.86, 129.56, 129.27, 128.53, 126.73, 114.62,
112.29, 111.83, 56.13, 55.73, 38.45, 25.18, 24.96, 23.55, 23.52, 22.26, 22.02. HRMS
(ESI) calcd for C25H35BN3O5: 468.26687 [(M-H2O+CH2+H) +], found 468.26590.
4.2.13. ((R)-1-((S)-2-(3-(2,3-dichlorobenzyl)ureido)-3-phenylpropanamido)-3-methyl
-butyl)boronic acid (I-6)
Using the same procedure as I-1, white foam, 32% yield; mp: 119–122 °C. HPLC
indicates a purity of 96.2 area %. 1H NMR (400 MHz, DMSO-d6) δ 7.22 (m, 8H), 7.12
(d, J = 4.3 Hz, 1H), 6.49 (s, 1H), 6.33 (d, J = 8.7 Hz, 1H), 4.58 (d, J = 5.9 Hz, 1H),
4.12 (dt, J = 15.3, 10.3 Hz, 2H), 2.98 (d, J = 12.4 Hz, 1H), 2.90–2.74 (m, 1H), 2.64 (s,
1H), 1.70–1.45 (m, 1H), 1.28 (m, 2H), 0.83 (d, J = 5.5 Hz, 6H). 13C NMR (101 MHz,
DMSO-d6) δ 175.00, 157.69, 140.98, 137.54, 129.89, 129.82, 128.59, 128.50, 128.40
(m, 18H), 127.29, 126.95, 126.79, 43.24, 38.99, 25.54, 23.60, 23.01, 21.50. HRMS
(ESI) calcd for C23H29BCl2N3O3: 462.25222 [(M-H2O+CH2+H) +], found 462.25277.
4.2.14. (R)-(1-(2-(3-(3-(Benzyloxy)phenyl)ureido)acetamido)-3-methylbutyl)boronic
acid (I-7)
Using the same procedure as I-1, white foam, 36% yield; mp: 109–113 °C. HPLC
indicates a purity of 95.6 area %. 1H NMR (400 MHz, DMSO-d6) δ 8.82 (s, 1H), 8.77
(s, 1H), 7.37 (m, 5H), 7.26 (s, 1H), 7.13 (t, J = 8.1 Hz, 1H), 6.92 (d, J = 7.8 Hz, 1H),
6.64–6.54 (m, 1H), 6.47 (t, J = 5.2 Hz, 1H), 5.05 (s, 2H), 3.92 (d, J = 4.6 Hz, 2H),
2.64 (s, 1H), 1.62 (dt, J = 13.2, 6.6 Hz, 1H), 1.44–1.20 (m, 2H), 0.83 (d, J = 4.9 Hz,
6H). 13C NMR (101 MHz, DMSO-d6) δ 173.08, 159.24, 155.44, 142.00, 137.65,
129.87, 128.84, 128.18, 128.00, 110.86, 108.02, 104.97, 69.51, 43.5 (m), 40.85, 25.62,
23.39, 23.25, 21.49. HRMS (ESI) calcd for C22H29BN3O4: 410.22495
[(M-H2O+CH2+H) +], found 410.22377.
4.2.15. (R)-(1-(2-(3-(2,5-Dichlorophenyl)ureido)acetamido)-3-methylbutyl)boronic
acid (I-8)
Using the same procedure as I-1, white foam, 17% yield; mp: 125–129 °C. HPLC
indicates a purity of 95.2 area %. 1H NMR (400 MHz, DMSO-d6) δ 8.79 (s, 1H), 8.47
(d, J = 10.7 Hz, 1H), 8.30 (d, J = 2.2 Hz, 1H), 7.57 (dd, J = 10.3, 5.3 Hz, 1H), 7.41
(dd, J = 8.5, 4.3 Hz, 1H), 7.00 (dd, J = 8.5, 2.2 Hz, 1H), 3.95 (d, J = 4.8 Hz, 2H), 3.37
(m, 1H), 1.61 (m, 1H), 1.38–1.17 (m, 2H), 0.83 (dd, J = 13.9, 3.7 Hz, 6H). 13C NMR
(101 MHz, DMSO-d6) δ 172.63, 168.84, 155.05, 138.26, 132.30, 130.79, 122.41,
119.91, 43.04, 25.60, 23.34, 23.17, 22.91. 1H9RMS (ESI) calcd for C15H21BCl2N3O3:

ACCEPTED MANUSCRIPT
372.10502 [(M-H2O+CH2+H) +], found 372.10413.
4.2.16. ((R)-1-((S)-2-((S)-2-(3-(3-(Benzyloxy)phenyl)ureido)-3-phenylpropanamido)-
3-phenylpropanamido)-3-methylbutyl)boronic acid (I-9)
Using the same procedure as I-1, white foam, 18% yield; mp: 130–133 °C. HPLC
1
indicates a purity of 96.4 area %. H NMR (400 MHz, DMSO-d6) δ 8.72 (d, J = 5.6
Hz, 1H), 8.65 (d, J = 16.6 Hz, 1H), 8.55 (s, 1H), 7.48–7.06 (m, 16H), 6.87 (d, J = 8.3
Hz, 1H), 6.80 (d, J = 4.0 Hz, 1H), 6.64–6.50 (m, 1H), 6.24 (d, J = 5.9 Hz, 1H), 5.04 (s,
2H), 4.68 (dd, J = 14.3, 6.5 Hz, 1H), 4.51 (dd, J = 16.6, 6.0 Hz, 1H), 3.00 (dd, J =
35.2, 22.1 Hz, 2H), 2.88–2.64 (m, 2H), 1.59 (s, 1H), 1.31 (dd, J = 23.7, 18.0 Hz, 2H),
0.79 (dd, J = 21.1, 5.3 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 172.42, 171.75,
159.25, 154.96, 141.92 , 137.64, 129.88, 129.84, 129.74, 128.84, 128.83, 128.51,
128.36, 128.18, 127.99, 126.69, 126.60, 110.69, 107.92, 104.87, 69.54, 54.30, 37.81,
25.51, 23.67, 23.41, 22.98, 22.80, 21.49. HRMS (ESI) calcd for C38H44BN4O5:
647.34057 [(M-H2O+CH2+H) +], found 647.33814.
4.2.17. ((R)-3-Methyl-1-((S)-2-(3-((S)-1-oxo-3-phenyl-1-(pyrazin-2-ylamino)propan
-2-yl)ureido)-3-phenylpropanamido)butyl)boronic acid (I-10)
Using the same procedure as I-1, white foam, 14% yield; mp: 170–174 °C. HPLC
1
indicates a purity of 96.5 area %. H NMR (400 MHz, DMSO-d6) δ 10.89 (s, 1H),
9.31 (s, 1H), 8.40 (s, 1H), 8.37 (s, 1H), 7.47 (s, 1H), 7.31–7.09 (m, 10H), 6.54 (d, J =
7.7 Hz, 1H), 6.40 (d, J = 7.6 Hz, 1H), 4.67 (d, J = 4.3 Hz, 1H), 4.29 (d, J = 5.5 Hz,
1H), 3.10–2.94 (m, 2H), 2.79 (ddd, J = 21.0, 13.6, 8.0 Hz, 2H), 1.50 (dd, J = 12.7, 6.4
Hz, 1H), 1.41–1.10 (m, 2H), 0.78 (dd, J = 27.1, 6.5 Hz, 6H). 13C NMR (101 MHz,
DMSO-d6) δ 172.61, 171.53, 157.53, 149.07, 143.11, 140.24, 138.27, 137.80, 136.69,
129.80, 129.76, 128.55, 128.37, 126.82, 126.47, 55.49, 54.67, 38.77, 38.35, 25.18,
23.71, 23.66, 22.46. HRMS (ESI) calcd for C29H36BN6O4: 543.28907
[(M-H2O+CH2+H) +], found 543.28785.
4.2.18. ((R)-1-(3-((S)-1-((3-(Benzyloxy)phenyl)amino)-1-oxo-3-phenylpropan-2-yl)-
ureido)-3-methylbutyl)boronic acid (II-2)
Using the same procedure as II-1, white foam, 27% yield; mp: 113–115 °C. HPLC
1
indicates a purity of 98.2 area %. H NMR (400 MHz, DMSO-d6) δ 10.12 (s, 1H),
7.58–7.06 (m, 16H), 6.73 (d, J = 7.0 Hz, 1H), 5.07 (s, 2H), 4.53 (s, 1H), 3.03 (s, 1H),
2.90 (s, 1H), 1.64–1.42 (m, 1H), 1.20 (dd, J = 14.5, 8.5 Hz, 2H), 0.85–0.68 (s, 6H).
13C NMR (101 MHz, DMSO-d6) δ 170.72, 159.07, 140.31, 137.47, 130.00, 129.77,
129.55, 128.89, 128.56, 128.28, 128.09, 126.91, 112.42, 110.07, 106.68, 69.60, 56.19,
41.91, 25.77, 23.67, 23.20, 21.52. HRMS (ESI) calcd for C29H35BN3O4: 500.27202
[(M-H2O+CH2+H) +], found 500.27051.
4.2.19. ((R)-3-Methyl-1-(3-((S)-1-oxo-3-phenyl-1-(pyrazin-2-ylamino)propan-2-yl)-
ureido)butyl)boronic acid (II-3)
Using the same procedure as II-1, white foam, 22% yield; mp: 145–148 °C. HPLC
1
indicates a purity of 97.2 area %. H NMR20(400 MHz, DMSO-d6) δ 10.96 (s, 1H),

ACCEPTED MANUSCRIPT
9.30 (s, 1H), 8.41 (s, 1H), 8.38 (d, J = 1.7 Hz, 1H), 7.42–7.17 (m, 5H), 7.17 (d, J =
6.4 Hz, 1H), 4.65 (s, 1H), 3.14–2.81 (m, 2H), 2.41 (s, 1H), 1.45 (dd, J = 12.7, 6.3 Hz,
1H), 1.32–1.05 (m, 2H), 0.71 (t, J = 7.0 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ
161.07, 148.94, 143.12, 140.40, 137.51, 136.63, 129.80, 128.56, 126.95, 56.03, 41.74,
25.69, 23.56, 23.14, 21.51. HRMS (ESI) calcd for C20H27BN5O3: 396.22050
[(M-H2O+CH2+H) +], found 396.21936.
4.2.20. ((R)-1-(3-((S)-1-((4-Methoxyphenyl)amino)-1-oxo-3-phenylpropan-2-yl)urei-
do)-3-methylbutyl)boronic acid (II-4)
Using the same procedure as II-1, white foam, 25% yield; mp: 146–148 °C. HPLC
indicates a purity of 96.2 area %. 1H NMR (400 MHz, DMSO-d6) δ 9.91 (s, 1H), 7.45
(d, J = 8.9 Hz, 2H), 7.20 (m, 5H), 7.18 (d, J = 6.0 Hz, 1H), 6.88 (d, J = 9.0 Hz, 2H),
4.50 (d, J = 5.4 Hz, 1H), 3.71 (s, 3H), 3.13–2.79 (m, 2H), 1.53 (td, J = 12.9, 6.3 Hz,
1H), 1.34–1.09 (m, 2H), 0.91–0.70 (t, J = 4.0 Hz, 6H). 13C NMR (101 MHz,
DMSO-d6) δ 172.46, 169.90, 155.90, 137.66, 132.15, 129.78, 128.54, 126.86, 121.50,
114.30, 55.61, 41.95, 38.85, 25.78, 23.69, 23.20. HRMS (ESI) calcd for C23H31BN3O4:
424.24062 [(M-H2O+CH2+H) +], found 424.23945.
4.2.21. ((R)-3-Methyl-1-(3-((R)-1-oxo-3-phenyl-1-(pyrazin-2-ylamino)propan-2-yl)
ureido)butyl)boronic acid (II-5)
Using the same procedure as II-1, white foam, 24% yield; mp: 138–141 °C. HPLC
1
indicates a purity of 95.9 area %. H NMR (400 MHz, DMSO-d6) δ 11.12–10.93 (m,
1H), 9.45–9.22 (m, 1H), 8.40 (s, 1H), 8.37 (s, 1H), 7.28 (d, J = 18.4 Hz, 5H),
7.21–7.15 (m, 1H), 6.52 (s, 1H), 4.68 (s, 1H), 2.92 (dd, J = 50.3, 43.2 Hz, 2H), 2.44 (s,
1H), 1.45 (dd, J = 12.6, 6.3 Hz, 1H), 1.13 (ddd, J = 20.3, 12.1, 6.9 Hz, 3H), 0.77–0.61
(dd, J = 6.9, 4.2 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 172.45, 160.69, 148.96,
143.11, 140.39, 137.52, 136.66, 129.83, 128.53, 126.92, 55.88, 41.92, 25.66, 23.61,
23.57, 23.06, 21.49. HRMS (ESI) calcd for C20H27BN5O3: 396.22050
[(M-H2O+CH2+H) +], found 396.21952.
4.2.22.
((R)-3-Methyl-1-(3-((S)-4-methyl-1-(((S)-4-methyl-1-oxo-1-(pyrazin-2-yl-
amino)pentan-2-yl)amino)-1-oxopentan-2-yl)ureido)butyl)boronic acid (II-6)
Using the same procedure as II-1, white foam, 21% yield; mp: 157–162 °C. HPLC
1
indicates a purity of 95.1 area %. H NMR (400 MHz, DMSO-d6) δ 10.95 (s, 1H),
9.29 (s, 1H), 8.40 (s, 1H), 8.36 (d, J = 2.3 Hz, 1H), 8.13 (d, J = 7.2 Hz, 1H), 6.67 (s,
1H), 4.63 (s, 1H), 4.20 (dd, J = 15.8, 12.3 Hz, 1H), 1.77–1.10 (m, 9H), 0.99–0.73 (m,
18H). 13C NMR (101 MHz, DMSO-d6) δ 173.57, 172.60, 149.09, 143.10, 140.31,
140.27, 136.80, 52.37, 52.02, 42.10, 41.07, 25.69, 24.66, 24.49, 23.71, 23.58, 23.30,
22.36, 21.89, 21.83. HRMS (ESI) calcd for C23H40BN6O4: 475.32027
[(M-H2O+CH2+H) +], found 475.31908.
4.2.23. ((R)-3-Methyl-1-(3-((S)-1-(((S)-4-methyl-1-oxo-1-(pyrazin-2-ylamino)pentan-
2-yl) amino)-1-oxo-3-phenylpropan-2-yl)ureido)butyl)boronic acid (II-7)
21

ACCEPTED MANUSCRIPT
Using the same procedure as II-1, white foam, 38% yield; mp: 154–157 °C. HPLC
1
indicates a purity of 98.7 area %. H NMR (400 MHz, DMSO-d6) δ 10.93 (s, 1H),
9.33 (s, 1H), 8.42 (d, J = 1.4 Hz, 1H), 8.39 (d, J = 2.4 Hz, 1H), 8.24 (dd, J = 18.5, 7.8
Hz, 1H), 7.31–6.97 (m, 5H), 4.66 (dd, J = 13.8, 7.5 Hz, 1H), 4.50–4.26 (m, 1H),
3.14–2.69 (m, 2H), 1.75–1.47 (m, 4H), 1.39–1.11 (m, 2H), 0.89 (dd, J = 11.9, 6.4 Hz,
6H), 0.80 (t, J = 5.6 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 172.45, 172.43,
149.08, 143.11, 140.35, 137.68, 136.86, 129.83, 129.51, 128.34, 126.65, 52.03, 42.04,
41.12, 25.71, 24.66, 23.73, 23.52, 23.24, 22.02, 21.49. HRMS (ESI) calcd for
C26H38BN6O4: 509.30467 [(M-H2O+CH2+H) +], found 509.30313.
4.2.24.
((R)-3-Methyl-1-((S)-2-(3-phenethylureido)-3-phenylpropanamido)butyl)
boronic acid (I-11)
Using the same procedure as I-1, white foam, 18% yield; mp: 134–137 °C. HPLC
1
indicates a purity of 96.6 area %. H NMR (400 MHz, DMSO-d6) δ 8.63 (d, J = 17.7
Hz, 1H), 7.33–7.06 (m, 10H), 6.37–6.24 (m, 1H), 6.05 (s, 1H), 4.55 (dd, J = 16.1, 8.1
Hz, 1H), 3.26–3.09 (m, 2H), 2.87 (ddd, J = 21.9, 13.6, 7.1 Hz, 2H), 2.61 (m, 3H),
1.75–1.44 (m, 1H), 1.45–1.13 (m, 2H), 0.82 (d, J = 5.9 Hz, 6H). 13C NMR (101 MHz,
DMSO-d6) δ 174.98, 157.58, 140.06, 137.66, 129.83, 129.08, 128.73, 128.49, 126.79,
126.42, 52.62, 41.39, 38.95, 36.47, 25.53, 23.40, 23.26. HRMS (ESI) calcd for
C24H33BN3O3: 422.26137 [(M-H2O+CH2+H) +], found 422.26053.
4.2.25. ((R)-3-Methyl-1-((S)-3-phenyl-2-(3-((R)-1-phenylethyl)ureido)propanamido)-
butyl)boronic acid (I-12)
Using the same procedure as I-1, white foam, 23% yield; mp: 151–154 °C. HPLC
1
indicates a purity of 95.2 area %. H NMR (400 MHz, DMSO-d6) δ 8.61 (s, 1H),
7.38–7.07 (m, 10H), 6.53 (d, J = 7.9 Hz, 1H), 6.12 (d, J = 8.5 Hz, 1H), 4.69 (p, J =
6.8 Hz, 1H), 4.52 (dd, J = 14.4, 7.5 Hz, 1H), 2.89 (ddd, J = 21.2, 13.6, 6.8 Hz, 2H),
2.63 (s, 1H), 1.53 (dt, J = 13.1, 6.5 Hz, 1H), 1.34–1.15 (m, 2H), 1.27 (d, J = 4.8 Hz,
3H), 0.79 (t, J = 6.2 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 174.69, 156.86,
145.73, 137.49, 129.91, 128.60, 128.48, 126.87, 126.78, 126.22, 52.49, 49.07, 39.09,
25.46, 23.74, 23.66, 22.86. HRMS (ESI) calcd for C24H33BN3O3: 422.26137
[(M-H2O+CH2+H) +], found 422.26043.
4.2.26. ((R)-3-Methyl-1-((S)-3-phenyl-2-(3-((S)-1-phenylethyl)ureido)propanamido)
butyl)boronic acid (I-13)
Using the same procedure as I-1, white foam, 32% yield; mp: 164–167 °C. HPLC
1
indicates a purity of 95.2 area %. H NMR (400 MHz, DMSO-d6) δ 8.58 (s, 1H),
7.31–7.17 (m, 10H), 6.51 (d, J = 7.8 Hz, 1H), 6.15 (d, J = 8.7 Hz, 1H), 4.72–4.62 (m,
1H), 4.53 (dd, J = 14.3, 8.0 Hz, 1H), 3.01–2.74 (m, 2H), 2.62 (d, J = 3.4 Hz, 1H), 1.54
(dt, J = 13.2, 6.6 Hz, 1H), 1.30–1.17 (m, 2H), 1.25 (d, J = 5.3 Hz, 3H), 0.79 (d, J =
6.6 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 174.77, 156.92, 145.62, 137.51,
22

ACCEPTED MANUSCRIPT
129.86, 128.64, 128.49, 126.90, 126.82, 126.25, 52.64, 49.10, 39.07, 25.54, 23.76,
23.37, 23.24. HRMS (ESI) calcd for C24H33BN3O3: 422.26137 [(M-H2O+CH2+H) +],
found 422.26051.
4.2.27.
((R)-3-Methyl-1-((S)-3-phenyl-2-(1,2,3,4-tetrahydroisoquinoline-2-carbox-
amido)propanamido)butyl)boronic acid (I-14)
Using the same procedure as I-1, white foam, 28% yield; mp: 158–160 °C. HPLC
1
indicates a purity of 97.0 area %. H NMR (400 MHz, DMSO-d6) δ 8.64 (s, 1H),
7.39–6.93 (m, 9H), 6.76 (d, J = 8.2 Hz, 1H), 4.57 (dd, J = 14.2, 8.7 Hz, 1H),
4.52–4.36 (m, 2H), 3.50 (t, J = 5.7 Hz, 2H), 3.01 (ddd, J = 22.8, 13.4, 7.4 Hz, 2H),
2.66 (d, J = 5.5 Hz, 2H), 1.58 (td, J = 13.2, 6.6 Hz, 1H), 1.38–1.16 (m, 2H), 0.81 (dd,
J = 6.5, 2.7 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 175.25, 172.44, 157.02,
138.32, 135.20, 134.39, 129.80, 128.90, 128.42, 126.60, 126.53, 126.38, 53.97, 45.73,
41.50, 37.81, 28.61, 25.58, 23.47, 23.15, 21.50. HRMS (ESI) calcd for C25H33BN3O3:
434.26139 [(M-H2O+CH2+H) +], found 434.26060.
4.2.28.
((R)-1-((S)-2-(3-Benzyl-3-methylureido)-3-phenylpropanamido)-3-methyl-
butyl)-boronic acid (I-15)
Using the same procedure as I-1, white foam, 31% yield; mp: 111–114 °C. HPLC
1
indicates a purity of 95.3 area %. H NMR (400 MHz, DMSO-d6) δ 8.71 (s, 1H),
7.31–7.18 (m, 8H), 6.99 (d, J = 6.9 Hz, 2H), 6.60 (d, J = 8.4 Hz, 1H), 4.66 (td, J = 9.1,
5.2 Hz, 1H), 4.36 (dd, J = 75.5, 15.8 Hz, 2H), 3.04 (ddd, J = 23.6, 13.6, 7.5 Hz, 2H),
2.70 (s, 3H), 1.65 (td, J = 13.3, 6.7 Hz, 1H), 1.46–1.20 (m, 2H), 0.87 (d, J = 6.5 Hz,
6H). 13C NMR (101 MHz, DMSO-d6) δ 175.43, 157.65, 138.85, 138.34, 129.80,
128.73, 128.49, 127.51, 127.17, 126.68, 53.92, 51.53, 37.80, 34.13, 25.66, 23.54,
23.29. HRMS (ESI) calcd for C24H33BN3O3: 422.26137 [(M-H2O+CH2+H) +], found
422.26058.
4.2.29.
((R)-1-((S)-2-(3-(2,3-Dimethylbenzyl)ureido)-3-phenylpropanamido)-3-
methylbutyl)boronic acid (I-16)
Using the same procedure as I-1, white foam, 19% yield; mp: 137–139 °C. HPLC
1
indicates a purity of 98.3 area %. H NMR (400 MHz, DMSO-d6) δ 7.24 (s, 5H),
7.07–6.89 (m, 3H), 6.33 (s, 1H), 6.27 (d, J = 8.2 Hz, 1H), 4.71–4.52 (m, 1H), 4.13
(ddd, J = 26.3, 18.5, 6.4 Hz, 2H), 3.07–2.77 (m, 2H), 2.67 (s, 1H), 2.21 (s, 3H), 2.07
(s, 3H), 1.60 (td, J = 13.2, 6.6 Hz, 1H), 1.30 (ddd, J = 21.8, 12.4, 6.2 Hz, 2H), 0.84 (d,
J = 5.9 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 174.86, 157.46, 138.14, 137.51,
136.58, 134.39, 129.89, 129.84, 128.78, 128.49, 126.79, 125.84, 125.56, 52.64, 42.11,
38.99, 25.57, 23.61, 23.47, 23.01, 20.46, 14.67. HRMS (ESI) calcd for C25H35BN3O3:
436.27704 [(M-H2O+CH2+H) +], found 436.27918.
4.2.30. ((R)-3-Methyl-1-((S)-3-phenyl-2-(3-(pyrazin-2-ylmethyl)ureido)propanamido)
butyl)boronic acid (I-17)
23

ACCEPTED MANUSCRIPT
Using the same procedure as I-1, white foam, 18% yield; mp: 154–157 °C. HPLC
1
indicates a purity of 97.0 area %. H NMR (400 MHz, DMSO-d6) δ 8.53 (d, J = 1.2
Hz, 1H), 8.49 (d, J = 2.5 Hz, 1H), 8.43 (s, 1H), 7.35–7.08 (m, 5H), 6.71 (d, J = 5.7 Hz,
1H), 6.58 (dd, J = 13.4, 5.1 Hz, 1H), 4.70–4.50 (m, 1H), 4.31 (qd, J = 16.6, 5.7 Hz,
2H), 2.90 (ddd, J = 22.2, 13.4, 6.6 Hz, 2H), 2.66 (d, J = 2.9 Hz, 1H), 1.57 (dd, J =
13.3, 6.6 Hz, 1H), 1.41–1.16 (m, 2H), 0.82 (d, J = 6.4 Hz, 6H). 13C NMR (101 MHz,
DMSO-d6) δ 172.46, 157.67, 155.48, 144.08, 143.50, 143.34, 137.59, 129.81, 128.47,
126.82, 52.74, 43.44, 38.86, 25.58, 23.38, 23.24, 21.49. HRMS (ESI) calcd for
C21H29BN5O3: 410.23617 [(M-H2O+CH2+H) +], found 410.23521.
4.2.31. ((R)-1-((S)-2-(3-(2,5-Dichlorobenzyl)ureido)-3-phenylpropanamido)-3-methy-
lbutyl)boronic acid (I-18)
Using the same procedure as I-1, white foam, 18% yield; mp: 118–120 °C. HPLC
indicates a purity of 96.2 area %. 1H NMR (400 MHz, DMSO-d6) δ 7.41 (m, 1H), 7.23
(m, 7H), 6.65 (d, J = 3.2 Hz, 1H), 6.53 (dd, J = 10.2, 5.7 Hz, 1H), 4.59 (dd, J = 13.7,
6.7 Hz, 1H), 4.26 (d, J = 5.9 Hz, 1H), 4.16 (d, J = 15.6 Hz, 3H), 3.01 (d, J = 11.9 Hz,
1H), 2.88 (d, J = 9.3 Hz, 1H), 2.69 (s, 1H), 1.59 (dt, J = 17.1, 6.5 Hz, 1H), 1.40–1.22
(m, 2H), 0.84 (d, J = 6.2 Hz, 6H).13C NMR (101 MHz, DMSO-d6) δ 174.25, 157.85,
140.58, 137.94, 132.34, 131.09, 130.80, 129.69, 128.61, 128.50, 126.88, 54.41, 41.02,
37.88, 25.56, 23.52, 23.45, 23.10, 23.00. HRMS (ESI) calcd for C23H29BCl2N3O3:
476.16776 [(M-H2O+CH2+H) +], found 476.16698.
4.2.32. ((R)-1-((S)-2-(3-([1,1'-Biphenyl]-4-ylmethyl)ureido)-3-phenylpropanamido)-
3-methylbutyl)boronic acid (I-19)
Using the same procedure as I-1, white foam, 19% yield; mp: 164–167 °C. HPLC
1
indicates a purity of 98.0 area %. H NMR (400 MHz, DMSO-d6) δ 7.63 (d, J = 8.3
Hz, 2H), 7.56 (d, J = 7.1 Hz, 2H), 7.45 (t, J = 7.5 Hz, 2H), 7.35 (t, J = 7.3 Hz, 1H),
7.31–7.19 (m, 7H), 6.59 (s, 1H), 6.48–6.32 (m, 1H), 4.66 (dd, J = 14.4, 7.0 Hz, 1H),
4.23 (ddd, J = 20.6, 15.6, 5.8 Hz, 2H), 3.14–2.82 (m, 2H), 2.73 (t, J = 16.0 Hz, 1H),
1.71–1.55 (m, 1H), 1.34 (ddd, J = 24.5, 12.7, 6.3 Hz, 2H), 0.86 (d, J = 6.0 Hz, 6H).
13C NMR (101 MHz, DMSO-d6) δ 172.45, 157.72, 140.54, 140.23, 139.01, 137.54,
129.94, 129.89, 129.34, 128.53, 127.94, 127.70, 127.00, 126.96, 52.52, 43.04, 39.02,
25.61, 23.62, 23.05, 21.50. HRMS (ESI) calcd for C29H35BN3O3: 484.27710
[(M-H2O+CH2+H) +], found 484.27583.
4.2.33. ((R)-1-((S)-2-(3-(Furan-2-ylmethyl)ureido)-3-phenylpropanamido)-3-methyl-
butyl) boronic acid (I-20)
Using the same procedure as I-1, white foam, 30% yield; mp: 120–124 °C. HPLC
indicates a purity of 97.9 area %. 1H NMR (400 MHz, DMSO-d6) δ 8.67 (s, 1H), 7.53
(d, J = 0.9 Hz, 1H), 7.32–7.18 (m, 5H), 6.52–6.25 (m, 3H), 6.10 (d, J = 2.7 Hz, 1H),
4.58 (dd, J = 14.0, 7.9 Hz, 1H), 4.13 (qd, J = 15.7, 5.7 Hz, 2H), 3.04–2.75 (m, 2H),
2.64 (d, J = 3.3 Hz, 1H), 1.57 (dt, J = 13.2, 6.6 Hz, 1H), 1.37–1.28 (m, 1H), 1.26–1.17
24

ACCEPTED MANUSCRIPT
(m, 1H), 0.82 (dd, J = 6.4, 3.2 Hz, 7H). 13C NMR (101 MHz, DMSO-d6) δ 174.82,
157.35, 153.71, 142.28, 137.51, 129.84, 128.50, 126.82, 110.81, 106.64, 52.64, 38.94,
36.85, 25.56, 23.38, 23.24. HRMS (ESI) calcd for C21H29BN3O4: 398.22494
[(M-H2O+CH2+H) +], found 398.22437.
4.2.34. ((R)-3-Methyl-1-((S)-2-(3-(naphthalen-1-ylmethyl)ureido)-3-phenylpropan-
amido)butyl)boronic acid (I-21)
Using the same procedure as I-1, white foam, 32% yield; mp: 171–174 °C. HPLC
indicates a purity of 95.8 area %. 1H NMR (400 MHz, DMSO-d6) δ 8.12–7.98 (m, 1H),
7.93 (dd, J = 6.5, 2.8 Hz, 1H), 7.81 (d, J = 8.2 Hz, 1H), 7.63–7.50 (m, 2H), 7.50–7.37
(m, 1H), 7.36–7.28 (m, 1H), 7.29–7.18 (m, 5H), 6.59 (d, J = 3.0 Hz, 1H), 6.50–6.30
(m, 1H), 4.70 (dd, J = 15.4, 6.0 Hz, 2H), 4.57 (dd, J = 15.1, 5.3 Hz, 1H), 2.93 (ddd, J
= 27.6, 13.7, 7.1 Hz, 2H), 2.71 (s, 1H), 1.61 (dd, J = 13.3, 6.6 Hz, 1H), 1.50–1.23 (m,
2H), 0.86 (d, J = 6.3 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 174.94, 157.59,
137.63, 136.04, 133.73, 131.26, 129.87, 128.91, 128.51, 127.74, 126.81, 126.57,
126.15, 125.87, 125.21, 123.82, 52.70, 41.34, 38.99, 25.61, 23.49, 23.23. HRMS (ESI)
calcd for C27H33BN3O3: 458.26142 [(M-H2O+CH2+H) +], found 458.26050.
4.2.35.
((R)-1-((S)-2-(3-(3,4-Difluorobenzyl)ureido)-3-phenylpropanamido)-3-methylbutyl)
boronic acid (I-22)
Using the same procedure as I-1, white foam, 35% yield; mp: 163–166 °C. HPLC
indicates a purity of 95.4 area %. 1H NMR (400 MHz, DMSO-d6) δ 7.67–7.43 (m, 1H),
7.30–7.17 (m, 6H), 7.13 (ddd, J = 6.9, 6.2, 2.5 Hz, 1H), 6.93 (d, J = 7.9 Hz, 1H),
6.67–6.56 (m, 1H), 6.50–6.36 (m, 1H), 4.57 (dd, J = 14.0, 8.1 Hz, 1H), 4.23–4.01 (m,
2H), 3.07–2.75 (m, 2H), 2.65 (d, J = 6.7 Hz, 1H), 1.67–1.48 (m, 1H), 1.35–1.23 (m,
2H), 0.82 (d, J = 5.9 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 174.84, 157.61,
150.30 (dd, J = 12.7 Hz, 107.18 Hz), 147.93 (d, J = 12.7 Hz, 105.52 Hz), 139.08,
137.55, 129.85, 128.47, 126.80, 123.73, 117.43 (d, J = 16.9 Hz), 116.07 (d, J = 17.0
Hz), 52.53, 42.24, 38.84, 32.76, 25.53, 23.57, 23.53, 22.96. HRMS (ESI) calcd for
C23H29BF2N3O3: 444.22549 [(M-H2O+CH2+H) +], found 444.22596.
4.2.36. ((R)-3-Methyl-1-((S)-2-(3-(3-nitrobenzyl)ureido)-3-phenylpropanamido)butyl)
boronic acid (I-23)
Using the same procedure as I-1, white foam, 28% yield; mp: 146–149 °C. HPLC
indicates a purity of 97.8 area %. 1H NMR (400 MHz, DMSO-d6) δ 8.12–7.96 (m, 2H),
7.55 (dt, J = 5.2, 2.8 Hz, 2H), 7.36–7.10 (m, 5H), 6.72 (s, 1H), 6.47 (dd, J = 16.3, 8.6
Hz, 1H), 4.60 (dd, J = 14.2, 7.9 Hz, 1H), 4.43–4.11 (m, 2H), 3.14–2.79 (m, 2H), 2.67
(d, J = 6.3 Hz, 1H), 1.58 (dd, J = 13.6, 6.8 Hz, 1H), 1.43–1.21 (m, 2H), 0.83 (d, J =
4.8 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 174.77, 157.66, 148.20, 143.77,
137.49, 134.01, 130.04, 129.87, 128.46, 126.76, 121.90, 121.82, 52.55, 42.65, 38.86,
25.55, 23.55, 22.97. HRMS (ESI) calcd for C23H30BN4O5: 453.23078
[(M-H2O+CH2+H) +], found 453.22971.
25

ACCEPTED MANUSCRIPT
4.2.37.
(S)-Methyl-2-(1-oxo-1,2,3,4-tetrahydroisoquinoline-2-carboxamido)-3-
phenylpropanoate (I-24a)
Obtained by the oxidation of I-14a unexpectedly in the air, yield > 99%; mp:
1
144–147 °C. H NMR (400 MHz, CDCl3) δ 9.98 (d, J = 7.0 Hz, 1H), 8.11 (d, J = 7.8
Hz, 1H), 7.51 (td, J = 7.5, 1.3 Hz, 1H), 7.32 (m, 7H), 4.83 (td, J = 7.4, 5.7 Hz, 1H),
4.22–4.07 (m, 2H), 3.75 (s, 3H), 3.19 (ddd, J = 21.5, 13.8, 6.6 Hz, 2H), 2.98 (t, J =
6.4 Hz, 2H). 13C NMR (101 MHz, CDCl3) δ 172.00, 167.31, 154.65, 139.84, 136.28,
133.28, 129.32, 129.25, 129.05, 128.59, 127.32, 127.07, 127.06, 55.37, 52.28, 41.70,
38.22, 27.87.
4.2.38. ((R)-3-Methyl-1-((S)-2-(1-oxo-1,2,3,4-tetrahydroisoquinoline-2-carboxamido)
-3-phenylpropanamido)butyl)boronic acid (I-24)
Using the same procedure as I-1, white foam, 21% yield; mp: 182–184 °C. HPLC
indicates a purity of 96.8 area %. 1H NMR (400 MHz, DMSO-d6) δ 8.95 (s, 1H), 8.04
(s, 1H), 7.46 (d, J = 7.5 Hz, 1H), 7.39–7.09 (m, 7H), 6.83 (t, J = 6.5 Hz, 1H), 4.21 (d,
J = 5.5 Hz, 1H), 3.52–3.38 (m, 2H), 2.95–2.72 (m, 4H), 1.77 (dd, J = 12.9, 6.8 Hz,
1H), 1.47 (ddd, J = 20.9, 13.0, 6.3 Hz, 2H), 0.89 (t, J = 5.7 Hz, 6H). 13C NMR (101
MHz, DMSO-d6) δ 173.57, 171.24, 156.69, 138.09, 136.00, 132.04, 131.12, 130.96,
130.11, 128.55, 127.13, 126.79, 57.41, 50.95, 38.93, 37.15, 25.84, 23.56, 23.24, 21.58.
+
HRMS (ESI) calcd for C24H31BN3O4: 448.24065 [(M-H2O+CH2+H) ], found
448.23982.
4.3. Preliminary biological assay
The enzymatic activities of the proteasome were assayed using The
Proteasome-Glo™ Cell-Based Assays kit (Promega, USA). The Proteasome-Glo™
Cell-Based Reagents each contain a specific luminogenic proteasome substrate in a
buffer optimized for cell permeabilization, proteasome activity and luciferase activity.
These
(Succinyl-leucine-leucine-valine-tyrosine-aminoluciferin),
(Z-leucine-arginine-arginine-aminoluciferin) and
peptide
substrates
are
Suc-LLVYaminoluciferin
Z-LRRaminoluciferin
Z-nLPnLDaminoluciferin
(Z-norleucine-proline-norleucine-aspartate-aminoluciferin) for the chymotrypsin-like,
trypsin-like and caspase-like activities, respectively. The trypsin-like assay also
contains two inhibitors to reduce nonspecific protease activities. Briefly, acute human
myeloid leukemic cell line (HL60, 5000 cells/well) were plated 20 ꢀL/well in a
384-well plate. Cells were then equilibrated at 37 °C and 5% CO2 for 2 h. Serial
dilutions of compounds were prepared in culture medium, and 5 ꢀL of each dilution
was added to wells. The cells were incubated with the drug at 37 °C and 5% CO2 for 2
h before 25 ꢀL/well of each Proteasome-Glo™ Cell-Based Reagent was added. The
relative luminescence units (RLU) were measured using the Multimode Microplate
Reader Varioskan Flash (Thermo Scientific, USA) after 15 min and compared with
the RLU of solvent control, and an inhibition rate was calculated.
4.3.1. Biological test of second-round compounds
26

ACCEPTED MANUSCRIPT
For the enzymatic activities of the proteasome, the same method was used as in the
preliminary biological assays, except that IC50 was calculated from the curves
generated by plotting the percentage of the solvent control versus the test
concentration on a logarithmic scale using the SigmaPlot 10.0 software. For the cell
growth inhibitory tests, MGC-803, HepG-2 and a standard MTT assay was used to
measure cell growth. In brief, a suspension of 3000 cells in 150 ꢀL of medium was
added to each well of 96-well plate and allowed to grow. After 24 h, drugs prepared in
medium at 10 different concentrations were added to the corresponding plate at a
volume of 50 ꢀL per well, and the plate was incubated for 72 h with drugs. Then to
each well was added a 20 ꢀL solution of 5 mg/mL MTT and incubation was continued
for another 4 h at 37 °C. Plate was then centrifuged at 1000 rpm at 4 °C for 5 min, and
the medium was carefully discarded. The formazan crystals were dissolved in 100 ꢀL
of DMSO and the absorbance was read using an Infinite M200 (Tecan, Austria)
microplate reader at 540 nm.
4.3.2. Anti-tumor spectrum test
We selected three promising compounds (I-6, I-14 and I-22) to investigate their
inhibitory capacity against 11 human tumor cell lines. The biological assays used the
same procedures as mentioned above, and all results were based on three experiments.
4.3.3. In vivo anticancer activity of I-14
Six-week-old male BALB/c athymic nude mice with body weight of 18–20 g
(purchased from Shanghai Silaike Animal Co.) were used. Bel7404 tumor blocks
were cut into uniform size pieces (about 2 mm) under sterile conditions and
subcutaneously implanted into each animal. After the xenograft tumors had grown to
a size of ∼100 mm3 (12 days later), mice (n = 6 per group) were given I-14 (1 mg/kg),
Bortezomib (0.5 mg/kg) and control saline (25 mL/kg) every other day for 24 days by
intravenous administration via the lateral tail vein. Tumor volumes (calculated using
the following formula, (width)2 × length/2) and body weights were measured 2–3
times a week during the experiment. At the end of the experiment (at day 25), all mice
were sacrificed and tumors were isolated and weighed.
4.3.4. Pharmacokinetics study of I-14
A clear aqueous solution of 3 mg/mL I-14 in 5% DMA + 25% PEG400 + 70% (2%
citric acid + 10% HPBCD) was prepared. Three adult male Sprague–Dawley rats
(Beijing Vital River Laboratory Animal Technology Co., Ltd.) were used in each
treatment group. The solution of I-14 was administered intravenously at 1 mL/kg via
the lateral tail vein as mentioned before. For blood collection, each rat was placed in a
clear Plexiglas restraining tube, and blood samples were drawn from the lateral tail
vein into heparinized collection tubes at predetermined sampling times (2 min, 5 min,
15 min, 30 min, 1 h, 2 h, 4 h, 8 h, 24 h post dose). At 24 h post dose, the rats were
sacrificed by decapitation and trunk blood was obtained. The blood samples were
temporarily put on ice until centrifuged to separate the plasma, which were then
stored at −20°C for the initial 24 h and at −80 °C afterwards. All experiments and
27

ACCEPTED MANUSCRIPT
handling of the compound and samples were carried out under protection from light.
LC/MS/MS analysis of I-14 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 I-14 were monitored for quantification using I-14
(50 ng/mL) in 1:1 v/v methanol/acetonitrile as an internal standard. Plasma
concentration–time data were analyzed by a non-compartmental approach using the
software WinNonlin Enterprise ver. 4.1 (Pharsight Co., Mountain View, CA).
4.4. Molecular docking
Protein Preparation: Protein structure was downloaded from the RCSB Protein Data
Bank (PDB ID: 2F16). All hetero atoms, hydrogen atoms, alternative conformers of
residues and the crystal cell were removed. Ligand Preparation: the 3D coordinates of
I-14 and PS-341 were prepared using Maestro 9.4. Then the initial coordinates were
further optimized via the LigPrep module with OPLS_2005 forcefield applied to give
the energy-minimized conformation. The boron–oxygen bond length was modified to
1.350 Å to stay in accordance with the crystallized bioactive conformation of PS-341.
Covalent Docking: The covalent docking was performed via Gold 5.2.2. The center of
the cavity was set on protein atom No. 6. With an active site radius of 10.0 Å, the
default genetic algorithm (GA) setting was applied to fully explore the conformational
space. GA Calculations were performed for 200 times. GoldScore was chosen to
assess the binding affinity. The best binding conformations were used for further
analysis. In post-docking analysis, 2D and surface plots were prepared by Maestro
9.4.
4.5. Molecular dynamics and MM-GBSA
System building: The docked conformation of the I-14-proteasome complex was
used in the molecular dynamics study. The complex was solved in TIP3P water model
with a concentration of NaCl of 0.15 mol/L. The box size calculation method was set
to buffer with a distance parameter of 10×10×10 Å. The solutes were reoriented to
minimize the box size. Six Na+ were added automatically to obtain an electrically
neutral system. OPLS_2005 forcefield was used to build the system. Minimization:
The model system is minimized by using a hybrid method of the steepest decent and
the limited-memory Broyden-Fletcher-Goldfrab-Shanno (LBFGS) algorithms with
maximum iterations at 2000 and convergence threshold at 1.0 kcal/mol/Å. Molecular
Dynamics: The total simulation time was set to 30.0 ns with a trajectory interval of
4.8 ps. The default NPT ensemble relaxation protocol was performed before the
simulation. The final unrestrained simulation was carried out at a constant pressure
(1.01325 bar) and temperature (300.0 K). Post-processing: The representative
conformation for 10–30 ns was analyzed and acquired in an interval of 10 steps.
MM-GBSA: I-14 and the receptor was splited and saved into different files before the
calculation to cater to the needs of the MM-GBSA module in Prime (Schrödinger
LLC.). Thus the total binding free energy (ꢁGbind) of the molecule with the covalent
binding energy can’t be given. For different frames from the simulation, we used the
VSGB solvation model with OPLS_2005 forcefield and minimized polar hydrogens
28