Design, synthesis and biological evaluation of novel non-peptide boronic acid derivatives as proteasome inhibitors

Article abstract of DOI:10.1016/j.ejmech.2017.01.034

A novel series of non-peptide proteasome inhibitors bearing the 1, 4-naphthoquinone scaffold and boronic acid warhead was developed. In the biological evaluation on the chymotrypsin-like activity of human 20S proteasome, five compounds showed IC₅₀

Full text of DOI:10.1016/j.ejmech.2017.01.034

Accepted Manuscript
Design, synthesis and biological evaluation of novel non-peptide boronic acid
derivatives as proteasome inhibitors
Ying Ge, Aibo Li, Jianwei Wu, Haiwei Feng, Letian Wang, Hongwu Liu, Yungen Xu,
Qingxiang Xu, Li Zhao, Yuyan Li
PII:
S0223-5234(17)30044-2
10.1016/j.ejmech.2017.01.034
EJMECH 9185
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 July 2016
Revised Date: 3 December 2016
Accepted Date: 21 January 2017
Please cite this article as: Y. Ge, A. Li, J. Wu, H. Feng, L. Wang, H. Liu, Y. Xu, Q. Xu, L. Zhao, Y. Li,
Design, synthesis and biological evaluation of novel non-peptide boronic acid derivatives as proteasome
inhibitors, European Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.01.034.
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
Fifteen novel non-peptide compounds were developed. Compound 3e exhibited
the inhibitory activities of CT-L and solid tumor cell in the nanomolar range, and
showed more improved metabolic stability than bortezomib.

ACCEPTED MANUSCRIPT
Design, synthesis and biological evaluation of novel non-peptide boronic acid
derivatives as proteasome inhibitors
Ying Geb†, Aibo Lib†, Jianwei Wub, Haiwei Fengb，Letian Wangb, Hongwu Liub, Yungen Xua, b
Qingxiang Xu c*, Li Zhao a*, Yuyan Lia, b
,
*
aJiangsu Key Laboratory of Drug Design and Optimization, China Pharmaceutical University, Nanjing
210009, China
bDepartment of Medicinal Chemistry, China Pharmaceutical University, Nanjing 210009, China
c Department of Hepatobiliary Surgery, Nanjing Drum Tower Hospital , Nanjing 210008, China
Abstract:
A novel series of non-peptide proteasome inhibitors bearing the 1,
4-naphthoquinone scaffold and boronic acid warhead was developed. In the biological
evaluation on the chymotrypsin-like activity of human 20S proteasome, five
compounds showed IC50 values in the nanomolar range. Docking experiments into the
yeast 20S proteasome rationalized their biological activities and allowed further
optimization of this interesting class of inhibitors. Within the cellular proliferation
inhibition assay and western blot analysis, compound 3e demonstrated excellent
anti-proliferative activity against solid tumor cells and clear accumulation of
ubiquitinated cellular proteins. Furthermore, in the microsomal stability assay
compound 3e demonstrated much improved metabolic stability compared to
bortezomib, emerging as a promising lead compound for further design of
non-peptide proteasome inhibitors.
Key words: Proteasome inhibitors; Non-peptide boronic acid; 1, 4-naphthoquinones;
Docking studies
*Corresponding author. Tel./fax: +86 2585551530.
E-mail address: yuyanli@cpu.edu.cn；zhaolicpu@126.com； xqx008dg@sina.com
†These authors contributed to the work equally and should be regarded as co-first
authors.

ACCEPTED MANUSCRIPT
1. Introduction
The 26S proteasome is the most important non-lysosomal proteolytic complex,
responsible for the breakdown of a large proportion of intracellular proteins[1]. Many
proteins regulated by 26S proteasome control cell physiology and cell cycle
progression. Alterations in the proteasome are associated with cancer-relevant
processes, such as antiapoptotic and proliferative signalling pathways[2]. It consists of
a 20S core particle and one or two 19S regulatory particles, and its proteolytic activity
relies on the former. The 20S core particle is composed of four stacked rings: two
outer α-rings (α1-α7) and two inner β-rings (β1-β7). The α subunits occupy the channel
gate to avoid non-substrate proteins entering the inner cavity of the 20S core[3].
While each β-ring contains three different catalytic subunits to carry out the
proteolytic activity: β1 (Caspase Like or Post Glutamyl-Peptide Hydrolase, C-L or
PGPH), β2 (Trypsin Like, T-L), β5 (Chymotrypsin Like, CT-L), which cleaves peptide
bonds after acidic, basic, and hydrophobic residues, respectively[4]. Interfering with
proteasomes, for example using proteasome inhibitors (PIs) against one or more
catalytic subunits, is a potential strategy for the development of anticancer drugs.
The attempts to develop PIs have been fruitful such as approvals of bortezomib,
carfilzomib and ixazomib by the FDA in 2003, 2012 and 2015, respectively, which
have validated the proteasome as a promising target. Most PIs reported to date
primarily have peptide backbones and form a covalent bond with the N-terminal Thr
residue in the β5 subunit of proteasome[5, 6]. However, it could not be avoided that
natural residues of peptide backbones would be specifically recognized and cleaved
by other peptidases. As a result, the three approved proteasome inhibitors, which
contain peptide backbones, are prone to rapid metabolic inactivation[7]: in BD
UltraPool human liver microsomes, carfilzomib becomes completely inactive in 10
minutes, and 20% bortezomib is metabolized in 10 minutes[8]; in the phase 1 study of
ixzomib, the plasma concentration of ixzomib decreases by approximately 90% after
30 minutes[9]. Although these drugs have been shown to provide significant benefits
to patients in clinical trials, their poor metabolic stabilities have limited the efficacy

ACCEPTED MANUSCRIPT
and administration especially for solid cancers. Other drawbacks also arise from the
shared peptide features such as chemical instability, immunogenicity and poor
membrane permeability[10]. Additionally, drug resistance and unwanted toxicities
bring challenges for clinical applications of bortezomib and carfilzomib[7].
Considering all these problems, non-peptide structures with unnatural amino
acids or a non-peptide scaffold is highly desirable for novel proteasome inhibitors,
which would present many pharmacokinetic and pharmacodynamic advantages. As a
matter of fact, some research groups have focused their effects on the identification of
non-peptide proteasome inhibitors with the specificity for β5 subunits. For example,
PI-083 (Fig. 1) suppresses the growth of human breast and lung tumors better than
bortezomib in nude mouse xenografts in vivo, but poor enzyme activity has limited its
further studies[11]. Compound 8, as a derivative of PI-083, has the comparable
antiproliferative activity to PI-083 against tumor cells and more potent CT-L
inhibitory activity with IC50 value of 3.65 µM[12]. Compound 3f (Fig. 1) obtained
through optimization of bortezomib demonstrates good antiproliferative activity
against cell lines of melanoma and non-small cell lung cancer[13]. PI-8182 (Fig. 1)
and its analogues show potent irreversible inhibitory activities, and analogue 39 (Fig.
1) is much more active than PI-8182 at inhibiting CT-L activity in intact human
breast[14]. G4-1 (Fig. 1) is highly effective in suppressing solid tumor growth in vivo
and displayes excellent metabolic stability, and the effectiveness of G4-1 is not
negatively impacted by resistance to bortezomib or carfilzomib[15]. Interestingly,
PI-083, 8, 3f, PI-8182 and its analogue 39 are both naphthoquinone derivatives with
biological significance, which indicates that the naphthoquinone structural motif may
be effective for non-peptide proteasome inhibitors.
In this context, we designed and synthesized a set of novel non-peptide
proteasome inhibitors, which bearing the 1, 4-naphthoquinone scaffold and boronic
acid warhead. Among them, compound 3e (Fig. 1) exhibited excellent inhibitory
activity against β5 subunits, which was comparable with bortezomib. Furthermore,
compound 3e suppressed the growth of solid tumor cells through polyubiquitin

ACCEPTED MANUSCRIPT
accumulation in the nanomole range and demonstrated more improved metabolic
stability than bortezomib.
Fig. 1. Structures of PI-083, bortezomib, 3e, 8[16], 3f, MLN2238, PI-8182, 5[28], G4-1, 8[28].
2. Results and discussion
2.1. Design
To design these non-peptide inhibitors, we analysed the conformation of
bortezomib and divided it into three fragments: P1, P2 and P3 (Fig. 1). The P1

ACCEPTED MANUSCRIPT
Leu-boronic moiety was not modified since it is essential to interact with the γ-OH of
the N-terminal Thr residue[13] (Fig. 2) and the hydrophobic residues of the S1 pocket
[16] (Fig. 3). A bulkier group than the isobutyl at P1 position, such as the benzyl and
cyclohexyl methyl, tending to fit into the larger S1 pocket of β1 subunits, decreases
the selectivity of inhibitors towards β5 subunits; while a smaller group like isopropyl
would reduce the inhibition activity[7]. At P2 position, large hydrophobic groups such
as the benzyl, naphthylmethyl are usually favorable to the potency[17]. In addition,
compounds without substitutions at this position are also active. For instance,
MLN2238 (Fig. 1) shows superior antitumor activity in both solid tumor and
hematologic xenograft models[18].
Fig. 2. Covalent docking mode(hydrogen bonding) of compound 3e with the Chymotrypic-like
active site compared with bortezomib. blue: 3e; green: bortezomib.
Fig. 3. Covalent docking mode (pockets relative positions) of bortezomib with the
Chymotrypic-like active site.

ACCEPTED MANUSCRIPT
The P3 site forms a hydrogen bond with the hydroxyl group of Asp114 (Fig. 2),
which is critical for the high affinity of bortezomib to β5 subunits[16]. Studies show
that replacements of the pyrazinyl with alkyl groups such as the methyl, morpholinyl
are handicaps for the inhibition activity[19]. While aromatic substituents such as
naphthalenes[20], phenylcyclohexane[17] improve the activity by enhancing
hydrophobic interactions of the P3 side chain. For example, 3f (Fig. 1) protruded into
the spacious S4 pocket located in β6 subunit, and formed van der Waals interactions,
stabilizing the P3 side chain in the β5 subunit[13].
On the basis of these findings, we designed a series of 1, 4-naphthoquinone
fragments to replace original groups of bortezomib at the P3 site, and modified the P2
position with different sterically hindered substituents. Accordingly, a series of
non-peptide boronic acid derivatives (Table 1) as novel proteasome inhibitors were
synthesized. A docking experiment performed between compound 3e and the β5
subunit of yeast 20S proteasome (PDB ID: 2F16)[16] (Fig. 2) showed that the
docking modes of P1 and P2 positions were similar to that of bortezomib. And at P3
position, the carbonyl oxygen atom at C-4 position of the 1, 4-naphthoquinone
coincided with the nitrogen atom of pyrazinyl of bortezomib, interacting with Asp114
by hydrogen bonding. In addition, as a hydrophobic moiety, the 1, 4-naphthoquinone
produced additional hydrophobic interactions with residues of the S3 pocket. The
above results were consistent with our hypothesis.
2.2. Chemistry
The synthesis of non-peptide boronic acid derivatives compounds 1a-1e, 2a-2e,
3a-3e was shown in Scheme 1. Intermediates x-1 were synthesized through
substitution reactions of the proper 1, 4-naphthoquinones with different amino acids.
Then x-1 were acylated with pinanediol leucine boronate (TPLM) in the presence of
TBTU and DIPEA, to provide the pinanediol esters x-2. Finally, the pinanediol esters
underwent a trans-esterification reaction with isobutylboronic acid (IBBA) under
acidic conditions, to afford the non-peptide boronic acid derivatives.

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of compounds 1a-1e, 2a-2e and 3a-3e. Reagents: (a) L-aminoacid, EtOH,
H2O, NaOH; (b) TPLM, TBTU, DIPEA, DCM; (c)CH3OH, Hexane, IBBA, HCl.
2.3. Inhibitory effect on the CT-L activity of human 20S proteasome
The inhibitory profile of synthesized compounds was evaluated on the CT-L
activity of human 20S proteasome, which is considered as a primary hallmark for
proteasome inhibition[21]. All compounds underwent a preliminary screening for
CT-L inhibitory activity by monitoring the hydrolysis of fluorogenic substrates
(Suc-Leu-Leu-Val-Tyr-AMC)[22], using an equivalent volume of dimethyl sulfoxide
(DMSO) as the negative control and bortezomib as the positive control. As expected,
all compounds displayed more than 50% inhibition at 1µM and eleven of them
displayed more than 50% inhibition at 100 nM.
Continuous assays were performed on the compounds with more than 80%
inhibition at 100 nM to determine their IC50 values via the dose-response curves. As
shown in Table 1, the IC50 values of compounds 1e, 2e, 3a, 3c, 3e were 24.6 nM, 25.9
nM, 23.2 nM, 44.3 nM and 11.4 nM, respectively. As can be noticed, compound 3e
exhibited the highest inhibition potency with the IC50 value of 11.4 nM, which was
comparable to that of bortezomib (IC50 = 4.7 nM).
When the R1 group was same, the R2 groups of compounds with the decreased
activity order were the benzyl, hydrogen atom, and alkyls. When the R2 group was
same, we have demonstrated that the hydrogen atom at R1 contributed to the highest
CT-L inhibitory activity, and the replacement of hydrogen atom with the chlorine
atom or methyl resulted in loss of inhibition of CT-L activity. However, the

ACCEPTED MANUSCRIPT
replacement of chlorine atom in PI-083 with hydrogen atom is detrimental to CT-L
inhibitory activity[23], and PI-083 (IC50 = 1.0 ± 0.63 µM)[23] decreased one order of
magnitude CT-L inhibitory activity compared to compound 1e (IC50 = 24.6 nM).
Moreover, compound 5 (IC50 = 39.40 ± 2.97 µM) and 8 (IC50 = 24.79 ± 1.98 µM)[24],
two of peptide-based derivatives bearing the chloronaphthoquinone moiety,
demonstrate much weaker CT-L inhibitory activities than compound 1e. Whereas
compound 3f and its analogues[13, 25], which are structurally more similar to our
compounds, demonstrate better inhibitions against the CT-L activity than PI-083,
compound 5 or 8. All these results indicated that the non-peptide scaffold with
boronic acid warhead may be more efficient for CT-L inhibitory acitivity.
Furthermore, both bortezomib and PI-083 inhibit CT-L activity in a higher extent
compared to the PGPH and T-L activities[23, 26], which is also consistent with
compound 3f and its analogues[13, 25]. As a result, we speculated that the new PIs
preferentially inhibited against the CT-L activity. The docking experiments into the
yeast 20S proteasome rationalized our speculation and offered useful information for
understanding the molecular interaction modes of new PIs.

ACCEPTED MANUSCRIPT
Table1
The structures and the CT-L inhibitory activities of non-peptide boronic acid derivatives.
Inhib
itory
Rate
(100
nM)
%
Inhib
itory
Rate
(100
nM)
%
Comp
ound
IC50(SD)a, Comp
IC50(SD)a,
nM
R1
R2
R1
R2
nM
ound
67.65
n.d.b
42.31
n.d.
1a
1b
Cl
2d
2e
CH3
CH3
H,H
64.79
43.48
n.d.
Cl
84.71
25.93(3.1)
n.d.
1c
Cl
84.88
67.53
23.17(2.1)
n.d.
3a
3b
H
H
H,H
47.35
86.70
n.d.
1d
1e
Cl
Cl
24.58(2.7)
79.81
44.25(4.6)
3c
H
65.88
52.30
n.d.
n.d.
69.19
90.29
n.d.
2a
2b
CH3
CH3
3d
3e
H,H
H
H
11.42(2.8)
31.45
n.d.
2c
CH3
92.50
4.68(0.5)c
Bortez
omib
-
-
aEach IC50 determination was performed with five concentrations, and each assay point was determined in
duplicate.
bn.d. =not determined.
cIC50 value obtained for bortezomib under our experimental conditions.

ACCEPTED MANUSCRIPT
2.4. Docking studies
In order to elucidate the CT-L inhibitory activities of the new PIs, we planned to
perform docking simulations of compounds 1e, 2e, 3a, 3c and 3e into the β5 subunit
of yeast 20S proteasome (PDB ID: 2F16)[16]. As depicted in Fig. 4, the decreased
activity of compound 1e, which was only different from compound 3e in P3 fragment,
can be attributed to the chlorine: its electron-withdrawing effect weakened the
hydrogen-bond between the Thr21 residue and the amino beside the
1,4-naphthoquinone; and its hydrophobic interactions with the binding pocket brought
about the molecular conformation alteration. For compound 2e (Fig. 5), although the
methyl at the P1 position had a favorable electron-donating effect on the amino, its
large steric hindrance caused the movement of the molecular skeleton in the β5
subunit. This movement obstructed the formation of the hydrogen bond between the
carbonyl of amide bond and the Thr21 residue, leading to the decreased inhibition.
Those indicate that the inhibitory activity is affected by the steric effect more than the
electronic effect at the C-3 of 1, 4-naphthoquinone ring.
Fig. 4. Covalent docking mode of compound 1e with the Chymotrypic-like active site (hydrogen
bonding).

ACCEPTED MANUSCRIPT
Fig. 5. Covalent docking mode of compound 2e with the Chymotrypic-like active site (hydrogen
bonding).
Fig. 6. Covalent docking mode of compound 3a with the Chymotrypic-like active site (hydrogen
bonding).
Fig. 7. Covalent docking mode of compound 3c with the Chymotrypic-like active site (hydrogen
bonding).
Compounds 3a and 3c had no substituent at R1. When they were docked with the
β5 subunit, a ring-flip of the 1, 4-naphthoquinone ring was observed, thus providing
an explanation to their decreased inhibitory activities against the β5 subunit. As

ACCEPTED MANUSCRIPT
depicted in Fig. 6, the ring-flip of compound 3a with an achiral R2 group was
relatively stable. The 1, 4-naphthoquinone ring of compound 3a formed π-π stacking
interactions with the Tyr107 residue and its carbonyl at C-1 formed additional
hydrogen bond with the Thr21 residue. But the affinity of compound 3a may be not as
strong as that of compound 3e, bringing about the moderate inhibitory activity with
the IC50 value of 23.2 nM. For compound 3c (Fig. 7), that its isobutyl at R2 was
smaller than the benzyl of compound 3e induced to the ring-flip, while the amino
beside the 1,4-naphthoquinone ring of compound 3c could not form hydrogen bonds
with the Thr21 residue, weakening its inhibitory activity compared to 3a. In regard to
compound 3b and 3d with other alkyl-substituents at R2, when they were
accommodated into the β5 subunit, their 1, 4-naphthoquinone rings also adopted the
ring-flip, but their ring-flips were not as stable as those of compound 3a and 3c,
resulting in their relatively poor inhibitory activities. Those suggest that a suitable R2
group, for example the benzyl of compound 3e or bortezomib, is the hinge of a stable
conformation.
As depicted in Fig. 2, the promising inhibitory activity of compound 3e can be
ascribed to that its P1 and P2 positions were found to project into the S1 and S2
pockets of the β5 subunit and adopted the similar spatial arrangement of the
corresponding moiety of bortezomib. Hydrogen bonds between compound 3e and
residues Thr1, Thr21 and Gly47 were observed, as in the bortezomib-proteasome
complex. The hydroxyl groups of the boronic acid warhead were stabilized by
hydrogen bonds between Lys33 and Tyr170 residues. And the carbonyl at C-4 position
of 1, 4-naphthoquinone ring reproduced the hydrogen bond of N-4 atom of
bortezomib pyrazinamide with the crucial Asp114 residue[16] to retain the CT-L
inhibitory activity. In addition, the benzene ring of benzyl formed hydrophobic
interactions with residues Tyr170, and the benzene ring of 1,4-naphthoquinone formed
hydrophobic interactions with residues Gly48, Pro115 and Val116 (Fig. 8). It should be
noted that the hydrogen bond between compound 3e and the Ala49 residue cannot be
formed compared to bortezomib (Fig. 2). Although the missing bond increased the

ACCEPTED MANUSCRIPT
hydrophobic interactions of 1, 4-naphthoquinone ring, it may provide a possible
explanation of the slight weaker activity of compound 3e than that of bortezomib.
Fig. 8. Covalent docking mode (pockets relative positions) of compound 3e with the
Chymotrypic-like active site.
2.5. Cellular proliferation inhibition assay
Compounds 3a and 3e were further evaluated for their inhibitory properties
against the growth of HL60, RPMI 8226, A549, MDA-MB-231 and A2780 cell lines.
Notably, as shown in Table 2, compound 3e displayed much higher anti-proliferative
activities against all five tested tumor cells than compound 3a. Furthermore,
compound 3e showed IC50 values in the nanomole range on the solid tumor cell lines
such as A549, MDA-MB-231 and A2780, especially MDA-MB-231 with the IC50
value of 38.3 nM, while in the micromole range on the non-solid tumor cell lines such
as HL60 and RPMI-8226. This indicates that compound 3e has more excellent
antiproliferative activities against solid tumor cells than non-solid tumor cells. In
addition, compound 3e (IC50 = 679.1±161.2 nM) was more active than PI-083 (IC50 =
12190 ± 1090 nM)[12] at inhibiting proliferation of A549, which suggested that the
combination of the 1, 4-naphthoquinone scaffold and boronic acid warhead was more
efficient.

ACCEPTED MANUSCRIPT
Table 2
Inhibitory activities of selected compounds against tumor cell growth（
x ± SD
）.
Compound
IC50 (nM)
HL60
RPMI-8226
A549
MDA-MB-231
1250.9±85.1
A2780
3a
3e
3586.2±172.7 3460.2±598.9 2894.2±550.2
3697.2±596.4
1105.7±87.1
3.2±0.7
1089±141.7
6.7±2.3
679.1±161.2
2.5±0.4
38.3±4.6
6.5±1.4
470±37.6
28.9±8.6
Bortezomib
2.6.Western blot analysis
Subsequently, the proteasome inhibitory effect of compound 3a and 3e was tested
in a cellular system by western blot analysis. Brotezomib was employed as positive
control and DMSO was employed as negative control. As shown in Fig. 9,
ubiquitinated proteins is clearly accumulated in MD-MB-231 cells treated with 3a and
3e, which indicated that both compounds induced proteasomal dysfunction in the
cells.

ACCEPTED MANUSCRIPT
Fig. 9. Immunoblotting analysis of poly-ubiquitin in MD-MB-231 cells treated with DMSO
(control), bortezomib, 3a and 3e.
2.7. Microsomal stability assay
Given that the rapid metabolic inactivation of peptide proteasome inhibitors, one
of the important aims in this study was to obtain compounds with metabolic stability.
We next examined whether the most potent compound 3e was more stable than
peptide-based inhibitors by using bortezomib as the reference compound and
monitoring the disappearance of the test compounds in the presence of liver
microsomes[8]. In our study, 44% bortezomib was remaining in male
Sprague-Dawley (SD) rat liver microsomes after 20 minutes (Fig. 10), which was
consistent with that approximately 30% bortezomib is remaining in Balb/c mice liver
microsomes after 20 minutes[8]. Excitingly, compound 3e showed excellent stability
profiles with approximately 60% remaining in liver microsomes. The result indicates
that the 1, 4-naphthoquinone scaffold-based proteasome inhibitor has much improved
metabolic stability.

ACCEPTED MANUSCRIPT
Fig. 10. Compared to bortezomib, 3e has better metabolic stability when tested using male SD rat
liver microsomes. The results are represented as means ± SD and analyzed by Origin 8.0.
3. Conclusions
In conclusion, we developed a set of novel non-peptide boronic acid proteasome
inhibitors by introducing the 1, 4-naphthoquinone scaffold into bortezomib. The
synthesized compounds showed a promising inhibition against the CT-L activity of
20S proteasome: eleven of them showed more than 50% inhibition at 100 nM and five
of them (1e, 2e, 3a, 3c, 3e) showed IC50 values in the nanomolar range, especially for
compound 3e, of which the IC50 value (11.4 nM) was comparable to that of
bortezomib (IC50=4.7 nM). Docking experiments into the yeast 20S proteasome
rationalized their CT-L inhibitory activities and allowed future optimization of this
interesting class of 20S proteasome inhibitors. Within the cellular proliferation
inhibition assay, compound 3e demonstrated excellent anti-proliferative activities
against solid tumor cell lines with IC50 values in the nanomole range. In addition, the
western blot analysis of compound 3a and 3e detected the accumulation of
intracellular ubiquitinated proteins, explaining the cellular proliferation inhibitory
effects. Furthermore, in the microsomal stability assay compound 3e demonstrated
much improved metabolic stability compared to bortezomib, emerging as a promising

ACCEPTED MANUSCRIPT
lead compound for further design of non-peptide proteasome inhibitors.
4. Experimental protocols
4.1. Chemistry
4.1.1. Material and measurements
All reagents and solvents were obtained from commercial sources and used
without further purification. Melting points were determined using a MELT-TEMPII
melting point apparatus and were uncorrected. Proton magnetic resonance (1H NMR)
13
and C NMR spectra were recorded on a Bruker Avance 300 (300 MHz). HR-MS
was collected on Agilent technologies 6520 Accurate-Mass Q-TOF LC/MS
instruments. All reactions were monitored by thin-layer chromatography with
precoated silica gel F254.
4.1.2. Synthesis of compounds 1a-1e
4.1.2.1. Synthesis of compounds 1a
N-(3-chloro-1,4-naphthoquinon-2-yl)-glycine-L-boronoleucine (1a): 500 mg
(2.20 mmol) 2, 3 - dichloro - 1, 4 – naphthoquinone (1) and ethanol (25 mL) were
added to a 100 mL round bottomed flask. A solution of glycine (495 mg, 6.60 mmol)
in 1N NaOH (5.5 mL, 5.5 mmol) was added dropwise, and the mixture was stirred
for 4h at room temperature. Water 20 mL was added, and the mixture was acidified
with 1N HCl to pH 3. Ethanol was removed under reduced pressure. Then the mixture
was cooled for 1h. The precipitate was collected by filtration and dried in vacuo to
give 450 mg of 1a-1 (88.0% yield) as a red solid.
1a-1 (129 mg, 0.48 mmol), (R)-BoroLeu-(+)-Pinanediol (TPLM, 273 mg, 0.72
mmol), and O-(Benzotriazol-1-yl)-N,N,N',N'-tetramethyluronium (TBTU, 170 mg,
0.53 mmol) in dry dichloromethane (DCM, 25 mL) were added in a 100 mL
°
three-necked flask and stirred at 0 C under the protection of nitrogen gas. Then
N,N-Diisopropylethylamine (DEPIA, 65 µl, 1.44 mmol) in dry DCM (5 mL) was
added dropwise for 2 h. After the addition, the mixture was warmed to room
temperature slowly, and DCM was removed under reduced pressure. The residue was

ACCEPTED MANUSCRIPT
dissolved in ethyl acetate (10 mL) and washed with water (20 mL x3), 3% citric acid
(20 mL x3) and water (20 mL x3) in turn. The combined organic layer was dried with
sodium sulphate anhydrous and concentrated. After purification by column
chromatography (silica gel, 25% ethyl acetate in petroleum ether), 120 mg of 1a-2
(48.1 % yield) as an orange solid was obtained.
1a-2 (350 mg, 0.68 mmol), methanol (7.5 mL), n-hexane (7.5 mL) and IBBA
°
(347 mg, 3.41 mmol) were added in a 50 mL three-necked flask and stirred at 0 C
under the protection of nitrogen gas. Then 1N HCl (1.7 mL) was added dropwise, and
the reaction mixture was stirred for 18.5 h. The methanol layer of the mixture was
extracted with n-hexane (10 mL x3) and concentrated. The residue was dissolved in
10 mL water and 10 mL DCM. The organic layer was collected and concentrated. The
crude product was washed with diethyl ether to give 172 mg of 1a.
4.1.2.2. Synthesis of compounds 1b
N-(3-chloro-1,4-naphthoquinon-2-yl)-L-alanine-L-boronoleucine (1b): 1b was
synthesized with the experimental protocols described for 1a.
4.1.2.3. Synthesis of compounds 1c
N-(3-chloro-1,4-naphthoquinon-2-yl)-L-leucine-L-boronoleucine (1c): 1c was
synthesized with the experimental protocols described for 1a.
4.1.2.4. Synthesis of compounds 1d
N-(3-chloro-1,4-naphthoquinon-2-yl)-L-valine-L-boronoleucine (1d): 1d was
synthesized with the experimental protocols described for 1a.
4.1.2.5. Synthesis of compounds 1e
N-(3-chloro-1,4-naphthoquinon-2-yl)-L-phenylalanine-L-boronoleucine (1e): 1e
was synthesized with the experimental protocols described for 1a.
4.1.3. Synthesis of compounds 2a-2e
4.1.3.1. Synthesis of compounds 2a
N-(3-methyl-1,4-naphthoquinon-2-yl)-glycine-L-boronoleucine (2a): 1000 mg

ACCEPTED MANUSCRIPT
(5.81 mmol) 2-methyl-1,4-naphthoquinon (2) and ethanol (30 mL) were added to a
250 mL round bottomed flask. A solution of glycine (1306 mg, 17.42 mmol) in 1N
NaOH (14.5 mL, 14.5 mmol) was added dropwise, and the mixture was stirred for 24
h at room temperature. Ethanol was removed under reduced pressure. Water 20 mL
was added. Then the mixture was filtrated, and the filtrate was washed with ethyl
acetate (20 mL x3). The water layer was acidified with 1N HCl to pH 3 and extracted
with ethyl acetate (20 mL x3). The ethyl acetate layer was concentrated and then
purified by column chromatography (silica gel, DCM : methanol = 20:1). The product
was collected and dried in vacuo. 330mg of 2a-1 (23.2% yield) as a red solid was
obtained. 2a-2 was synthesized with the experimental protocols described for 1a-2.
Then 2a was synthesized from 2a-2 with the experimental protocols described for 1a.
4.1.3.2. Synthesis of compounds 2b
N-(3-methyl-1,4-naphthoquinon-2-yl)-L-alanine-L-boronoleucine (2b): 2b was
synthesized with the experimental protocols described for 2a.
4.1.3.3. Synthesis of compounds 2c
N-(3-methyl-1,4-naphthoquinon-2-yl)-L-leucine-L-boronoleucine (2c): 2c was
synthesized with the experimental protocols described for 2a.
4.1.3.4. Synthesis of compounds 2d
N-(3-methyl-1,4-naphthoquinon-2-yl)-L-valine-L-boronoleucine (2d): 2d was
synthesized with the experimental protocols described for 2a.
4.1.3.5. Synthesis of compounds 2e
N-(3-methyl-1,4-naphthoquinon-2-yl)-L-phenylalanine-L-boronoleucine (2e): 2e
was synthesized with the experimental protocols described for 2a.
4.1.4. Synthesis of compounds 3a-3e
4.1.4.1. Synthesis of compounds 3a
N-(1,4-naphthoquinon-2-yl)-glycine-L-boronoleucine (3a): 1000 mg (6.32 mmol)
1,4-naphthoquinon (3) and ethanol (55 mL) were added to a 250mL round bottomed

ACCEPTED MANUSCRIPT
flask. A solution of glycine (712mg, 9.48 mmol) in 1N NaOH (6.32 mL, 6.32 mmol)
was added dropwise, and the mixture was stirred for 10 min at room temperature.
Ethanol was removed under reduced pressure. Water 20 mL was added. Then the
mixture was filtrated, and the filtrate was washed with ethyl acetate (20 mL x3). The
water layer was acidified with 1N HCl to pH 3 and extracted with ethyl acetate (20
mL x3). The ethyl acetate layer was concentrated and then purified by
recrystallization in DCM. The crystal was collected and dried in vacuo to give 260 mg
of 3a-1 (17.8% yield) as a reddish brown solid. 3a-2 was synthesized with the
experimental protocols described for 1a-2. Then 3a was synthesized from 3a-2 with
the experimental protocols described for 1a.
4.1.4.2. Synthesis of compounds 3b
N-(1,4-naphthoquinon-2-yl)-L-alanine-L-boronoleucine (3b): 3b was synthesized
with the experimental protocols described for 3a.
4.1.4.3. Synthesis of compounds 3c
N-(1,4-naphthoquinon-2-yl)-L-leucine-L-boronoleucine (3c): 3c was synthesized
with the experimental protocols described for 3a.
4.1.4.4. Synthesis of compounds 3d
N-(1,4-naphthoquinon-2-yl)-L-valine-L-boronoleucine (3d): 3d was synthesized
with the experimental protocols described for 3a.
4.1.4.5. Synthesis of compounds 3e
N-(1,4-naphthoquinon-2-yl)-L-phenylalanine-L-boronoleucine (3e): 3e was
synthesized with the experimental protocols described for 3a.
4.1.5. Characterization of compounds 1a-1e, 2a-2e, 3a-3e
4.1.5.1. N-(3-chloro-1,4-naphthoquinon-2-yl)-glycine-L-boronoleucine (1a)
Yield: 67%; mp: 182-185 °C; Red solid; 1H NMR (300 MHz, DMSO-d6) δ: 8.75
(br s, 1H), 7.91 (br d, 1H, J = 7.1 Hz), 7.89 (br d, 1H, J = 7.1 Hz), 7.76 (dt, 1H, J =
7.1 Hz; 1.3 Hz), 7.69 (dt, 1H, J = 7.1 Hz; 1.3 Hz), 7.51 (br s, 1H, -NH-), 4.39 (d, 2H,

ACCEPTED MANUSCRIPT
J = 6.6 Hz), 2.53 (m, 1H), 1.51-1.45 (m, 1H), 1.26-1.09 (m, 2H), 0.77 (t, 3H, J = 6.3
Hz), 0.70 (d, 3H, J = 6.3 Hz); 13C NMR (300 MHz, DMSO-d6) δ: 180.2, 176.0, 171.9,
167.9, 135.3, 133.3, 132.1, 130.5, 126.9, 126.2, 47.0, 45.3, 43.3, 25.5, 23.2, 22.3 ppm;
HRMS (ESI-TOF): m/z, calculated for C17H20BClN2O5, 376.1117 (M-H)-, found
376.1124.
4.1.5.2. N-(3-chloro-1,4-naphthoquinon-2-yl)-L-alanine-L-boronoleucine (1b)
1
Yield: 61%; mp: 170-171 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.79
(br s, 1H), 7.92 (br d, 1H, J = 7.4 Hz), 7.89 (br d, 1H, J = 7.4 Hz), 7.76 (dt, 1H, J =
7.4 Hz; 0.84 Hz), 7.69 (dt, 1H, J = 7.4 Hz; 0.84 Hz), 7.09 (br s, 1H), 5.08 (t, 1H, J =
7.4 Hz), 2.63 (s, 1H), 1.53-1.46 (m, 1H), 1.38 (d, 3H, J = 6.8 Hz), 1.23 (q, 2H), 0.72
13
(d, 6H, J = 6.5 Hz); C NMR (300 MHz, DMSO-d6) δ: 180.1, 176.2, 174.4, 171.2,
135.4, 133.4, 132.1, 130.3, 127.0, 126.3, 51.2, 50.8, 42.9, 25.5, 23.5, 22.8, 21.0 ppm;
HRMS (ESI-TOF): m/z, calculated for C18H22BClN2O5, 390.1274 (M-H)-, found
390.1274.
4.1.5.3. N-(3-chloro-1,4-naphthoquinon-2-yl)-L-leucine-L-boronoleucine (1c)
1
Yield: 75%; mp: 160-161 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.80
(br s, 1H), 7.93 (br d, 1H, J = 7.4 Hz), 7.91 (br d, 1H, J = 7.4 Hz), 7.78 (dt, 1H, J =
7.4 Hz; 1.08 Hz), 7.71 (t, 1H, J = 7.4 Hz; 1.08 Hz), 6.84 (br s, 1H), 5.07 (s, 1H), 2.55
(s, 1H), 1.80-1.73 (m, 1H), 1.66-1.49 (m, 2H), 1.39-1.21 (m, 1H), 1.20-1.13 (m, 2H),
0.86 (d, 3H, J = 6.0 Hz), 0.78 (d, 3H, J = 6.0 Hz), 0.71 (d, 6H, J = 6.3 Hz); 13C NMR
(300 MHz, DMSO-d6) δ: 180.0, 176.3, 174.2, 170.7, 135.5, 133.5, 132.1, 130.3,
127.1, 126.4, 54.0, 43.4, 42.5, 25.5, 24.7, 23.4, 23.2, 23.0, 22.3 ppm; HRMS
(ESI-TOF): m/z, calculated for C21H28BClN2O5, 432.1743 (M-H)-, found 432.1752.
4.1.5.4. N-(3-chloro-1,4-naphthoquinon-2-yl)-L-valine-L-boronoleucine (1d)
1
Yield: 54%; mp: 159-161 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.82
(br s, 1H), 7.93 (br d, 1H, J = 7.4 Hz), 7.91 (br d, 1H, J = 7.4 Hz), 7.77 (dt, 1H, J =
7.4 Hz; 0.8 Hz), 7.70 (dt, 1H, J = 7.4 Hz; 0.8 Hz), 6.68 (br s, 1H), 4.94 (t, 1H, J = 5.3
Hz), 2.64 (s, 1H), 2.08-2.04 (m, 1H), 1.55-1.51 (m, 1H), 1.26-1.16 (m, 2H), 0.91 (d,

ACCEPTED MANUSCRIPT
13
3H, J = 6.6 Hz), 0.80 (t, 3H, J = 6.6 Hz), 0.73 (d, 6H, J = 6.6 Hz); C NMR (300
MHz, DMSO-d6) δ: 180.0, 176.3, 172.9, 169.0, 135.6, 133.5, 132.1, 130.2, 127.1,
126.1, 65.3, 59.3, 42.7, 32.9, 25.5, 23.5, 22.7, 18.6 ppm; HRMS (ESI-TOF): m/z,
calculated for C20H26BClN2O5, 418.1587 (M-H)-, found 418.1592.
4.1.5.5. N-(3-chloro-1,4-naphthoquinon-2-yl)-L-phenylalanine-L-boronoleucine (1e)
1
Yield: 54%; mp: 165-167 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.90
(br s, 1H), 7.86 (br d, 1H, J = 7.1 Hz), 7.82 (br d, 1H, J = 7.1 Hz), 7.74 (dt, 1H, J =
7.1 Hz; 0.9 Hz), 7.66 (dt, 1H, J = 7.1 Hz; 0.9 Hz), 7.18-7.08 (m, 5H), 6.63 (br s, 1H),
5.36 (d, 1H, J = 6.9 Hz), 3.09 (t, 2H, J = 7.4 Hz), 2.65 (s, 1H), 1.51-1.47 (m, 1H),
13
1.27 (q, 2H), 0.71 (d, 6H, J= 5.0 Hz); C NMR (300 MHz, DMSO-d6) δ: 179.8,
176.1, 173.1, 169.6, 136.2, 135.4, 133.4, 131.9, 130.2, 128.7, 128.5, 127.3, 127.0,
126.4, 57.1, 56.0, 43.2, 25.4, 25.2, 23.6, 22.8 ppm; HRMS (ESI-TOF): m/z, calculated
for C24H26BClN2O5, 466.1587 (M-H)-, found 466.1588.
4.1.5.6. N-(3-methyl-1,4-naphthoquinon-2-yl)glycine-L-boronoleucine (2a)
1
Yield: 41%; mp: 140-141 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.67
(br s, 1H), 7.86 (br d, 1H, J = 6.9 Hz), 7.84 (br d, 1H, J = 6.9 Hz), 7.69 (dt, 1H, J =
6.9 Hz; 0.8 Hz), 7.69 (dt, 1H, J = 6.9 Hz; 0.8 Hz), 6.81 (br s), 4.24 (s, 2H), 2.54 (s,
1H), 1.97 (s, 3H), 1.63-1.39 (m, 1H), 1.05-0.94 (m, 2H), 0.68 (d, 6H, J = 4.5 Hz); 13C
NMR (300 MHz, DMSO-d6) δ: 182.3, 182.2, 172.7, 146.7, 134.7, 132.8, 132.6, 130.8,
126.0, 125.7, 112.8, 45.6, 43.6, 25.5, 23.3, 23.1, 10.3 ppm; HRMS (ESI-TOF): m/z,
calculated for C18H23BN2O5, 356.1664 (M-H)-, found 356.1664.
4.1.5.7. N-(3-methyl-1,4-naphthoquinon-2-yl)-L-alanine-L-boronoleucine (2b)
1
Yield: 50%; mp: 120-122 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.73
(br s, 1H), 7.85 (br d, 1H, J = 6.6 Hz), 7.83 (br d, 1H, J = 6.6 Hz), 7.73 (dt, 1H, J =
6.6 Hz; 1.8 Hz), 7.68 (dt, 1H, J = 6.6 Hz; 1.8 Hz), 6.32 (br s, 1H), 4.72 (q, 1H), 3.35
(s, 3H), 2.60 (q, 1H), 2.02 (s, 3H), 1.55-1.37 (m, 1H), 1.30 (q, 2H), 0.69 (d, 6H, J =
5.8 Hz); 13C NMR (300 MHz, DMSO-d6) δ: 182.6, 182.2, 175.4, 145.9, 134.9, 134.8,
132.8, 130.8, 126.7, 125.8, 113.7, 51.2, 43.0, 25.5, 23.4, 22.8, 20.7, 10.7 ppm; HRMS

ACCEPTED MANUSCRIPT
(ESI-TOF): m/z, calculated for C19H25BN2O5, 370.1820 (M-H)-, found 370.1827.
4.1.5.8. N-(3-methyl-1,4-naphthoquinon-2-yl)-L-leucine-L-boronoleucine (2c)
1
Yield: 50%; mp: 143-145 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.70
(br s, 1H), 7.93 (br d, 1H, J = 6.9 Hz), 7.90 (br d, 1H, J = 6.9 Hz), 7.78 (dt, 1H, J =
6.9 Hz; 0.72 Hz), 7.70 (dt, 1H, J = 6.9 Hz; 0.7 Hz), 6.14 (br s, 1H), 4.60 (s, 1H), 2.53
(s, 1H), 1.99 (s, 3H), 1.79-1.63 (m, 1H), 1.60-1.50 (m, 2H), 1.31-1.10 (m, 1H), 1.09
(m, 2H), 0.90 (d, 3H, J = 5.7 Hz), 0.81 (d, 3H, J = 5.7 Hz), 0.73 (d, 6H, J = 4.8 Hz);
13C NMR (300 MHz, DMSO-d6) δ: 182.9, 182.2, 171.8, 146.6, 135.0, 133.0, 132.9,
130.5, 126.1, 126.0, 114.0, 56.6, 43.0, 25.2, 24.7, 23.6, 23.2, 22.5, 22.1, 11.2 ppm;
HRMS (ESI-TOF): m/z, calculated for C22H31BN2O5, 412.2290 (M-H)-, found
412.2286.
4.1.5.9. N-(3-methyl-1,4-naphthoquinon-2-yl)-L-valine-L-boronoleucine (2d)
1
Yield: 76%; mp: 118-120 °C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.75
(br s, 1H), 7.89 (br d, 1H, J = 5.4 Hz), 7.86 (br d, 1H, J = 5.4 Hz), 7.73 (dt, 1H, J =
5.4 Hz; 0.8 Hz), 7.68 (dt, 1H, J = 5.4 Hz; 0.8 Hz), 6.04 (br s, 1H, -NH-), 4.35 (t, 1H, J
= 7.2 Hz), 2.57 (s, 1H), 1.99 (m, 1H) 1.99 (s, 3H), 1.54-1.40 (m, 1H), 1.24-1.09 (m,
2H), 0.89 (d, 3H, J = 6.3 Hz), 0.79 (d, 3H, J = 6.3 Hz), 0.68 (d, 6H, J = 7.2 Hz); 13C
NMR (300 MHz, DMSO-d6) δ: 182.8, 182.1, 173.5, 146.1, 135.0, 133.0, 132.8, 130.5,
126.2, 126.0, 114.4, 60.7, 42.9, 32.5, 25.5, 23.2, 22.6, 18.7, 11.2 ppm; HRMS
(ESI-TOF): m/z, calculated for C21H29BN2O5, 398.2133 (M-H)-, found 398.2138.
4.1.5.10. N-(3-methyl-1,4-naphthoquinon-2-yl)-L-phenylalanine-L-boronoleucine (2e)
Yield: 53%; mp: 135-137 °C; Red solid; 1H NMR (300 MHz, DMSO-d6) δ: 8.88
(br s, 1H), 7.82 (br d, 1H, J = 7.3 Hz), 7.74 (br d, 1H, J = 7.3 Hz), 7.67 (dt, 1H, J =
7.3 Hz; 1.2 Hz), 7.59 (dt, 1H, J = 7.3 Hz; 1.2 Hz), 7.00-7.17 (m, 5H), 6.04 (br s, 1H),
4.89 (s, 1H), 3.03 (s, 2H), 2.62 (s, 1H), 1.87 (d, 3H, J = 9.6 Hz), 1.46 (m, 1H), 1.21
(m, 2H), 0.70 (m, 6H); 13C NMR (300 MHz, DMSO-d6) δ: 182.1, 182.0, 173.9, 146.2,
137.6, 136.8, 135.0, 132.8, 130.6, 130.0, 128.6, 127.1, 126.1, 125.8, 114.4, 57.0, 43.2,
25.4, 25.2, 23.6, 22.7, 10.9 ppm; HRMS (ESI-TOF): m/z, calculated for C25H29BN2O5,

ACCEPTED MANUSCRIPT
446.2133 (M-H)-, found 446.2142.
4.1.5.11. N-(1,4-naphthoquinon-2-yl)glycine-L-boronoleucine (3a)
Yield: 52%; mp: 154-155 °C; Red solid; 1H NMR (300 MHz, DMSO-d6) δ: 8.79
(br s, 1H), 7.96 (br d, 1H, J = 7.5 Hz), 7.92 (br d, 1H, J = 7.5 Hz), 7.78 (dt, 1H, J =
7.5 Hz; 1.2 Hz), 7.73 (dt, 1H, J = 7.5 Hz; 1.2 Hz), 7.59 (br s, 1H), 5.61 (s, 1H), 4.03 (s,
2H), 2.49 (s, 1H), 1.54 (s, 1H), 1.27 (s, 1H), 1.16 (s, 1H), 0.75 (d, 6H, J = 7.5 Hz); 13C
NMR (300 MHz, DMSO-d6) δ: 182.3, 181.6, 170.8, 149.0, 135.3, 133.3, 132.8, 130.8,
126.4, 125.8, 100.9, 45.2, 42.8, 25.3, 23.2, 22.3 ppm; HRMS (ESI-TOF): m/z,
calculated for C17H21BN2O5, 342.1507 (M-H)-, found 342.1503.
4.1.5.12. N-(1,4-naphthoquinon-2-yl)-L-alanine-L-boronoleucine (3b)
Yield: 43%; mp: 145-147 °C; Red solid; 1H NMR (300 MHz, DMSO-d6) δ: 8.76
(br s, 1H), 7.87 (br d, 1H, J = 6.8 Hz), 7.8 (br d, 1H, J = 6.8 Hz), 7.70 (dt, 1H, J = 6.8
Hz; 1.89 Hz), 7.64 (dt, 1H, J = 6.8 Hz; 1.9 Hz), 7.21 (br s, 1H), 5.54 (s, 1H), 4.20 (t,
1H, J = 6.9 Hz), 2.50 (s, 1H), 1.42 (s, 1H), 1.29 (m, 3H), 1.05 (s, 2H), 0.68 (br d, 3H,
13
J = 7.2 Hz), 0.66 (br d, 3H, J = 7.2 Hz); C NMR (300 MHz, DMSO-d6) δ: 182.3,
181.5, 173.9, 173.9, 171.9, 135.3, 133.2, 132.9, 130.8, 126.4, 125.8, 101.5, 49.2, 43.3,
25.5, 23.4, 22.9, 17.9 ppm; HRMS (ESI-TOF): m/z, calculated for C18H23BN2O5,
356.1664 (M-H)-, found 356.1663.
4.1.5.13. N-(1,4-naphthoquinon-2-yl)-L-leucine-L-boronoleucine (3c)
Yield: 52%; mp: 158-160 °C; Red solid; 1H NMR (300 MHz, DMSO-d6) δ: 8.82
(br s, 1H), 7.94 (br d, 1H, J = 7.1 Hz), 7.88 (br d, 1H, J = 7.1 Hz), 7.77 (dt, 1H, J =
7.1 Hz; 1.3 Hz), 7.68 (dt, 1H, J = 7.1 Hz; 1.3 Hz), 7.50 (br s, 1H), 5.72 (s, 1H), 4.14 (s,
1H), 2.51 (s, 1H), 1.87-1.72 (m, 1H), 1.67-1.38 (m, 3H), 1.26-1.01 (m, 2H), 0.84-0.67
(m, 12H); 13C NMR (300 MHz, DMSO-d6) δ: 182.4, 181.5, 173.4, 148.4, 135.3,
133.2, 132.9, 130.8, 126.4, 125.8, 101.4, 52.3, 43.2, 25.5, 24.7, 23.4, 23.1, 22.9, 22.2
ppm; HRMS (ESI-TOF): m/z, calculated for C21H29BN2O5, 398.2133 (M-H)-, found
398.2139.
4.1.5.14. N-(1,4-naphthoquinon-2-yl)-L-valine-L-boronoleucine (3d)

ACCEPTED MANUSCRIPT
Yield: 55%; mp: 160-163 °C; Red solid; 1H NMR (300 MHz, DMSO-d6) δ: 8.93
(br s, 1H), 8.01 (br d, 1H, J = 7.2 Hz), 7.94 (br d, 1H, J = 7.2 Hz), 7.84 (dt, 1H, J =
7.2 Hz; 0.81 Hz), 7.75 (dt, 1H, J = 7.2 Hz; 0.81 Hz), 7.54 (br s, 1H), 5.76 (s, 1H),
3.84 (t, 1H, J = 8.0 Hz), 2.6 (s, 1H), 2.23-2.09 (m, 1H), 1.66-1.46 (m, 1H), 1.33-1.16
(m, 2H), 0.88 (br d, 3H, J = 7.0 Hz), 0.85 (br d, 3H, J = 7.0 Hz), 0.77 (br d, 3H, J =
6.3 Hz), 0.75 (br d, 3H, J = 6.3 Hz); 13C NMR (300 MHz, DMSO-d6) δ: 182.4, 181.5,
172.3, 148.2, 135.4, 133.1, 132.9, 130.7, 126.4, 125.8, 101.4, 59.2, 42.9, 30.7, 25.5,
23.5, 22.6, 19.4 ppm; HRMS (ESI-TOF): m/z, calculated for C20H27BN2O5, 408.1942
(M+Na)+, found 408.1941.
4.1.5.15. N-(1,4-naphthoquinon-2-yl)-L-phenylalanine-L-boronoleucine (3e)
1
Yield: 43%; mp: 165-167°C; Red solid; H NMR (300 MHz, DMSO-d6) δ: 8.83
(br s, 1H), 7.83 (br d, 1H, J = 7.4 Hz), 7.79 (br d, 1H, J = 7.4 Hz), 7.70 (dt, 1H, J =
7.4 Hz; 1.44 Hz), 7.62 (dt, 1H, J = 7.4 Hz; 1.44 Hz), 7.52 (br s, 1H), 7.07-7.18 (m,
5H), 5.57 (s, 1H), 4.37 (d, 1H, J = 7.4 Hz), 4.18 (s, 1H), 3.05 (s, 2H), 2.53 (s, 1H),
1.46-1.24 (m, 1H), 1.23-0.92 (m, 2H), 0.72 (br d, 3H, J = 9.0 Hz), 0.63 (br d, 3H, J =
9.0 Hz); 13C NMR (300 MHz, DMSO-d6) δ: 182.3, 181.3, 172.4, 147.9, 137.9, 136.9,
135.4, 132.8, 130.6, 129.7, 128.6, 126.9, 126.4, 125.8, 101.3, 57.1, 54.8, 25.9, 25.3,
23.5, 22.6 ppm; HRMS (ESI-TOF): m/z, calculated for C24H27BN2O5, 456.1942
(M+Na)+, found 456.1942.
4.2. Covalent docking
GOLD version 5.0[27] was used to implement covalent docking studies with the
same protein–ligand complex (3UNB) as described by Siwei Zhang et al[28]. The
genetic algorithm (GA) method was adopted by GOLD for conformational analysis
and docking evaluation. The default GA settings of GOLD were used: population
size=100; selection pressure=1.1; operations=100,000; islands=5; niche size=2;
migration=10; mutation=95; crossover=95. The centroid of the cocrystallized ligand
was regarded as the docking cavity. The the terminal boron atom of inhibitors was set
as the link atom that binds to the hydroxyl oxygen of Thr1 in the covalent option. In
order to identify potential leads, different binding poses were ranked using the

ACCEPTED MANUSCRIPT
Goldscore function and the interactions were validated to select reasonable hits. The
covalent docking was evaluated by docking PR957 into the same protein using
GOLD[29].
4.3. Pharmacology
4.3.1. The chymotrypsin-like activity of 20S proteasome inhibition assay
The 20S proteasome activity assay kit was purchased from Chemicon. The CT-L
activity of 20S proteasome was measured by monitoring the hydrolysis of fluorogenic
substrates (Suc-Leu-Leu-Val-Tyr-AMC)[22]. The 20S proteasome was incubated with
Suc-Leu-Leu-Val-Tyr-AMC in the presence of proteasome activators and tested
compounds. The hydrolysate AMC resulted in emission of fluorescence and was
measured at 380 nm/460 nm. The preliminary screening for the CT-L inhibitory
activity of the 20S proteasome was performed at 1 µM, using an equivalent volume of
dimethyl sulfoxide (DMSO) as the negative control and bortezomib as the positive
control. Continuous assays were performed on compounds with more than 50%
inhibition at 1 µM to determine their inhibitory properties at 100 nM using the above
method. And then compounds with more than 80% inhibition at 100 nM were chosen
to determine their IC50 values at five different concentrations using the progress curve
method.
4.3.2. Cellular proliferation inhibition assay
Human cancer cell lines HL-60, RPMI 8226, A549, MDA-MB-231, A2780 were
obtained from Zhejiang University and maintained in the DMEM, RPMI 1640 or F12
supplemented with 10% fetal bovine serum, 2 nmol/L L-glutamine (GIBCO), 100
IU/Ml penicillin and 100 g/mL streptomycin. The compounds were dissolved in
DMSO and stored under low temperature without light. A Sulforhodamine B (SRB)
assay was used to measure cell growth. In brief, cells were incubated for 72 h with
compounds in each well of 96-well plates. Then 100 µL 10% trichloroacetic acid
(TCA) was added to each well. The plates were placed for 1 h at 4 °C, then was
washed with deionized water for 5 times, and dried in the air. Then 50 µL 4 mg/mL

ACCEPTED MANUSCRIPT
SRB solution was added to each well. The plates were placed for 20 min at room
temperature, then washed with 1% acetic acid for 5 times and dried in the air. 100 µL
Tris-base (pH=10.5, 10 mM) was added to each well and the absorbance was read on
ELIASA at 515 nm.
4.3.3. Western blot analysis
Poly-ubiquitin antibody was purchased from ABclonal Biotechnology (Boston,
MA, USA) and β-actin was obtained from Santa Cruz Biotechnology (CA, USA). All
antibodies were used as recommended by the manufacturers. MD-MB-231 cells were
collected and lysed in lysis buffer. Protein samples were loaded onto 10% SDS-PAGE
gels and transferred to a nitrocellulose membrane (BioTrace NT; PallCor, USA). The
membranes were then blocked with 3% fat-free milk in PBST for 1 h and incubated
with primary antibodies overnight at 4°C followed by IRDye800-conjugated
secondary antibody (Santa Cruz Biotechnology, CA, USA) for 1 h at room
temperature. Detection was performed by Odyssey Scanning System (Li-Cor,
Lincoln, NE, USA). The blots were stripped and reprobed with polyclonal anti-β-actin
antibody to ascertain loading of proteins.
4.3.4. Microsomal stability assay
The metabolic stability profiles of compound 3e and reference compound
(bortezomib) were assessed by monitoring the disappearance of test compounds in the
presence of liver microsomes[8]. A typical incubation mixture (100 µL total volume)
for metabolic stability studies contained 20 µM test compounds, 0.5 mg/mL
microsomal protein (male Sprague-Dawley rat liver microsomes), 100 mM Tris-HCl
buffer (pH 7.4), and NADPH generating system (1 mM). After preincubation at 37 °C
for 5 min, the reactions were started by addition of NADPH and further incubated for
another 0, 5, 10, 15, 20 min. For control experiments, liver microsomes were omitted
from these incubations. The reactions were terminated by adding 100 µL of ice-cold
acetonitrile containing nimodipine (20 µM) as internal standard and keeping on ice for
30 min, followed by centrifugation at 12000 rpm/min for 20 min to obtain the
supernatant. Aliquots (30 µL) were then analyzed for substrate disappearance using

ACCEPTED MANUSCRIPT
high performance liquid chromatography (SHIMADZU high performance liquid
chromatograph LC-20A).
Conflict of interest
The authors declare no conflict of interest.
Acknowledgments
This work was supported by Excellent Science and Technology Innovation Team of
Jiangsu Province Universities in 2015, China.
References and notes
[1] D.M. Rubin, D. Finley, Proteolysis: The proteasome: a protein-degrading organelle?, Current Biology
Cb, 5 (1995) 854-858.
[2] H. Daniela, D. Ivan, Targeting the ubiquitin system in cancer therapy, Nature, 458 (2009) 438-444.
[3] A.F. Kisselev, A.L. Goldberg, Proteasome inhibitors: from research tools to drug candidates,
Chemistry & Biology, 8 (2001) 739–758.
[4] T.P. Dick, A.K. Nussbaum, M. Deeg, ., W. Heinemeyer, ., M. Groll, ., M. Schirle, ., W. Keilholz, ., S.
Stevanovi?, ., D.H. Wolf, R. Huber, . Contribution of proteasomal beta-subunits to the cleavage of
peptide substrates analyzed with yeast mutants, Journal of Biological Chemistry, 273 (1998)
25637-25646.
[5] M. Groll, K.B. Kim, N. Kairies, R. Huber, C.M. Crews, Crystal Structure of Epoxomicin:20S
Proteasome Reveals a Molecular Basis for Selectivity of α‘,β‘-Epoxyketone Proteasome Inhibitors,
J.am.chem.soc, 122 (2000).
[6] G. Michael, K.A. Mcarthur, V.R. Macherla, M. Rama Rao, B.C. Potts, Snapshots of the
fluorosalinosporamide/20S complex offer mechanistic insights for fine tuning proteasome inhibition,
Journal of Medicinal Chemistry, 52 (2009) 5420-5428.
[7] (!!! INVALID CITATION !!!).
[8] M. Zachary, K. Keun-Sik, L. Do-Min, K. Vinod, B.S. Eun, L. Kwang Hyun, Z. Yan-Yan, A. Lin, C.
Kimberly, L. Na-Ra, Proteasome Inhibitors with Pyrazole Scaffolds from Structure-Based Virtual
Screening, Journal of Medicinal Chemistry, 58 (2015).
[9] D.C. Smith, T. Kalebic, J.R. Infante, L.L. Siu, D. Sullivan, G. Vlahovic, J.S. Kauh, F. Gao, A.J. Berger, S.
Tirrell, Phase 1 study of ixazomib, an investigational proteasome inhibitor, in advanced
non-hematologic malignancies, Investigational New Drugs, 33 (2015) 652-663.
[10] B. Philipp, D. Christian, G. Michael, Covalent and non-covalent reversible proteasome inhibition,
Biological Chemistry, 393 (2012) 1101-1120.
[11] K. Aslamuzzaman, L. Harshani, W.C. Guida, M.L. Mclaughlin, G.M. Springett, B. Norbert, R.M.L.
Yip, S.D.M. Sebti, Discovery of a novel proteasome inhibitor selective for cancer cells over
non-transformed cells, Cell Cycle, 8 (2009) 1940-1951.
[12] X. Kai, Z. Xiao, B.T. Yan, H. Li, C.H. Chen, E. Ohkoshi, K.H. Lee, Design and synthesis of
naphthoquinone derivatives as antiproliferative agents and 20S proteasome inhibitors, Bioorganic &
Medicinal Chemistry Letters, 22 (2012) 2772-2774.