Design, synthesis and biological evaluation of hybrid of ubenimex-fluorouracil for hepatocellular carcinoma therapy

Article abstract of DOI:10.1016/j.bioorg.2021.105343

In our previous study, we discovered a ubenimex-fluorouracil (5FU) conjugates BC-02, which displays significant in vivo anti-tumor activity, however, the instability of BC-02 in plasma limits its further development as a drug candidate. Herein, we designe

Full text of DOI:10.1016/j.bioorg.2021.105343

Journal Pre-proofs
Design, Synthesis and Biological Evaluation of Hybrid of Ubenimex-Fluo-
rouracil for Hepatocellular Carcinoma Therapy
Kairui Yue, Xiaohan Hou, Geng Jia, Liang Zhang, Jian Zhang, Leqiao Tan,
Xuejian Wang, Zhaolin Zhang, Peixia Li, Wenfang Xu, Xiaoyang Li, Yuqi
Jiang
PII:
S0045-2068(21)00720-3
DOI:
https://doi.org/10.1016/j.bioorg.2021.105343
YBIOO 105343
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
6 July 2021
28 August 2021
6 September 2021
Please cite this article as: K. Yue, X. Hou, G. Jia, L. Zhang, J. Zhang, L. Tan, X. Wang, Z. Zhang, P. Li, W. Xu,
X. Li, Y. Jiang, Design, Synthesis and Biological Evaluation of Hybrid of Ubenimex-Fluorouracil for
Hepatocellular Carcinoma Therapy, Bioorganic Chemistry (2021), doi: https://doi.org/10.1016/j.bioorg.
2021.105343
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.
© 2021 Published by Elsevier Inc.

Design, Synthesis and Biological Evaluation of
Hybrid of Ubenimex-Fluorouracil for Hepatocellular
Carcinoma Therapy
Kairui Yuea, 1, Xiaohan Houa, 1, Geng Jiaa, Liang Zhanga, Jian Zhangc, Leqiao Tand,
Xuejian Wangc, Zhaolin Zhangd, Peixia Lia, Wenfang Xue, Xiaoyang Lia, b, *, Yuqi
Jianga, b,
*
a
Key Laboratory of Marine Drugs, Chinese Ministry of Education, School of
Medicine and Pharmacy, Ocean University of China, 5 Yushan Road, Qingdao
266003, China.
b
Laboratory for Marine Drugs and Bioproducts, Qingdao National Laboratory for
Marine Science and Technology, Qingdao, 266237, P.R. China.
c
College of Pharmacy, Weifang Medical University, 261053 Wei’fang, Shandong,
P.R. China.
d Weifang Bochuang International Biological Medicinal Institute, Weifang, Shandong
261061, P.R. China.
e Marine Biomedical Research Institute of Qingdao, Qingdao, Shandong, 266071, P.R.
China.
ABSTRACT
In our previous study, we discovered a ubenimex-fluorouracil (5FU) conjugates BC-
02, which displays significant in vivo anti-tumor activity, however, the instability of
BC-02 in plasma limits its further development as a drug candidate. Herein, we
designed and synthesized four novel ubenimex-5FU conjugates by optimizing the
linkers between ubenimex and 5FU based on BC-02. Representative compound 20 is
more stable than BC-02 in human plasma and displays about 100 times higher CD13
inhibitory activity than the positive control ubenimex. Meanwhile, the
antiproliferative activity of 20 was comparable with 5FU in vitro. The preliminary
mechanism study indicated that compound 20 exhibited significant anti-invasion and
1

anti-angiogenesis activities in vitro. Furthermore, compound 20 obviously inhibits
tumor growth and metastasis in vivo and prolong the survival time of tumor-bearing
mice. Our study may have an important implication reference for the design of more
druglike mutual prodrug, and compound 20 can be used as a lead compound for
further design and development.
Keywords: Anti-cancer, APN/CD13 inhibitor, Mutual prodrug, angiogenesis,
metastasis.
1. Introduction
CD13, also known as aminopeptidase N (EC 3.4.11.2), is a zinc dependent type 2
transmembrane ecto-peptidase, which expresses on various cell types, especially on
the surface of tumor cells [1]. CD13 is involved in a variety of biological functions [2],
including invasion, metastasis [3], neovascularization [4, 5], chemotherapy
resistance[6, 7] and so on. Recently, it has been reported that CD13 is a biomarker of
liver cancer stem cells, which is closely related to drug resistance and cancer
recurrence [8]. Moreover, several studies have found that fluorouracil, doxorubicin,
gemcitabine and sorafenib treatment can increase the expression of CD13, which is
one of the most important reasons for cytotoxic drug resistance [9, 10]. More recent
studies have revealed that high expression of CD13 is associated with poor prognosis
and patient survival in a variety of cancers, such as hepatocellular carcinoma (HCC)
[11], multiple myeloma [12], pancreas and colon cancers [13, 14], and non-small cell
lung cancer [15]. Therefore, CD13 is regarded as an important target for the treatment
of cancer [16].
In recent years, CD13 has been used as a ligand target in antibody conjugated
drugs (ADCs) and peptide conjugated drugs (PDCs). In 2020, Zapata et al. reported
the first antibody conjugated drug MI130110 using CD13 monoclonal antibody as
ligand, which is tumor tissue specific and displays remarkable anti-tumor activity [17].
CD13 inhibitor NGR (ASN-GLY-ARG) based peptide conjugated drugs NGR-
hTNF and tTF-NGR specifically recognize CD13, and are currently in phase III and
phase I clinical stages respectively [18, 19]. Therefore, the development of CD13
targeted conjugated drugs (TDC) has become a promising field in development of
anti-tumor drugs.
2

Recent studies have found that the combination of 5FU and CD13 inhibitor
ubenimex synergistically inhibits the proliferation of liver cancer cells in vitro and in
vivo [8]. Our previous studies have reported a series of 5FU and ubenimex mutual
prodrugs [20, 21], among which, compound BC-02 displayed significant in vivo anti-
tumor activity, even in 5FU-resistant liver cancer model. We further revealed that the
promising activity of BC-02 in 5FU-resistance tumor cells may due to the down-
regulation of CD13 and increase of Reactive oxygen species (ROS) level [22].
Moreover, BC-02 effectively suppressed self-renewal and malignant proliferation of
cancer stem cells (CSCs) by targeting CD13 and intracellular ROS-induced DNA
damage[23]. Despite the outstanding anti-tumor activity, BC-02 has been found to
degrade rapidly in plasma, which limits its further study as a drug candidate.
Therefore, to improve its plasma stability, we optimize the chemical structure of BC-
02 by modifying the linker between ubenimex and 5FU in this current study (Fig. 1).
We also conduct detailed in vivo and in vitro anti-tumor study and human plasma
stability study of the newly designed compounds. Representative compound 20
slowly cleaves to 5FU in human plasma with a t1/2 of 12 h, which is much longer than
that of BC-02 (t1/2: 1.5 h), which indicates the significantly improved plasma stability
of 20. Slow release of 5FU can prolong the time of 5FU in vivo, reduce the metabolic
inactivation of 5FU, so that improve the anti-tumor efficacy. The in vitro and in vivo
antitumor activity studies also showed that 20 could inhibit tumor growth, metastasis
and prolong survival.
F
NH2
O
NH
OH
N
O
N
O
H
OH
1 (Ubenimex)
H
O
5-FU
NH2
O
O
H
N
F
N
H
NH2
O
O
O
O
N
O
OH
O
F
O
N
H
N
H
O
N
N
O
2 (BC-02)
H
OH
O
O
O
N
H
O
16
Optimizing
O
NH2
O
NH2
H
N
Optimizing
F
N
N
N
H
N
O
N
H
H
OH
OH
20
O
F
O
N
O
O
O
N
O
H
12
O
N
H
O
O
F
NH2
O
O
O
N
H
N
N
H
N
H
O
N
O
H
OH
O
27
Figure 1. Drug design and chemical structure of targeted compounds.
3

2. Results and discussion
2.1. Chemistry
Compounds 12 and 16 were synthesized following the procedures showed in
Scheme 1. The start material 3 was protected by (Boc)2O, then coupled with methyl
L-leucinate hydrochloride in the presence of 1-hydroxybenzotriazole (HOBt) and 1-
(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDCI) to obtain
compound 5. Cleavage of the protecting group with base got compound 6. Compound
7 was coupled with acyl chloride to get intermediate 8. Boc-ethylenediamine and 8
generated carbamate 9 in alkaline conditions, and intermediate 9 was converted to 10
in the presence of HCl in ethyl acetate. Condensation of 10 and 6 in the presence of
HOBt and EDCI gave the amide 11. Cleavage of the Boc protecting group with HCl
in ethyl acetate and alkalified afforded the target compound 12. Compound 16 was
prepared by the same procedure described for 12 except using 2-aminoethanol.
The synthesis of compound 20 was described in Scheme 2. In the presence of
HOBt and EDCI, 7 reacted with N-Boc-L-proline to obtain ester 17. Using the HCl in
ethyl acetate to remove the protecting group of intermediate 17 to afford 18, which
was condensed with compound 6 to obtain 19. Cleavage the Boc protecting group
with HCl in ethyl acetate to obtain the target compound 20.
As described in Scheme 3, the compound 27 was synthesized from the starting
material 21. 22 was obtained by using EDCI and HOBt as condensing agent, then the
hydroxyl was converted to carbonate 23. In the presence of DMAP, the carbonate
bond was replaced by 7 to obtain 24. The remaining steps were prepared by the same
procedure described for 12.
4

Boc
Boc
NH2
O
Boc
NH
O
NH
O
a
c
b
NH
O
O
N
OH
OH
OH
N
H
N
4
OH
5
OH
O
O
3
OH
H
OH
6
O
O2N
HCl
HN
O
O
H
N
F
O
F
F
OH
N
f
e
O
O
N
d
Boc
N
O
F
H
N
O
N
O
N
H
O
H
O
N
H
O
N
H
O
9
10
8
7
O
N
O
H
Boc
NH
O
O
NH2
O
O
O
O
H
N
H
6
h
F
i
N
F
N
H
N
O
N
N
H
N
O
N
N
H
OH
H
O
OH
O
O
O
N
O
12
11
O
O
N
O
H
H
Boc
O
NH
O
O
NH2
OH
F
8
6
i
O
F
OH
N
H
O
N
O
F
N
N
O
N
N
H
NH2
N
H
O
e
H
h
H
OH
OH
O
O
N
O
14
13
O
N
O
15
N
H
O
16
H
H
Scheme 1: Synthesis of Compounds 12 and 16. Reagents and conditions: (a) (Boc)2O,
1N NaOH; (b) Methyl L-leucinate hydrochloride, EDCI, HOBt, DCM; (c) 1.5 N
NaOH, MeOH, 80 °C; (d) p-nitrophenyl chloroformate, TEA, THF; (e) Boc-
Ethylenediamine, TEA, DMAP, DCE, 60 °C; (f) dry HCl in EtOAc; (h) EDCI, HOBt,
DCM; (i) dry HCl in EtOAc, TEA.
HCl
O
Boc
O
Boc
H
NH
O
F
F
F
N
N
N
HO
b
O
N
6
a
O
N
a
N
N
H
O
N
H
O
O
O
N
O
O
N
O
OH
F
H
H
O
O
N
19
18
17
7
O
N
O
H
HCl
NH2
O
N
b
N
H
O
OH
F
O
O
N
20
O
N
O
H
Scheme 2: Synthesis of Compound 20. Reagents and conditions: (a) EDCI, HOBt,
DCM; (b) dry HCl in EtOAc.
NO2
O
O
F
O
OH
OH
O
O
24
N
O
N
O
O
O
H
H
H
7
c
N
N
N
Boc
N
b
a
H2N
Boc
N
O
N
O
Boc
N
H
H
H
H
21
22
23
O
O
HCl
HN
F
Boc
F
d
O
O
O
N
NH
O
O
O
O
d
6
e
H
N
N
H
N
H
O
N
H
O
N
O
N
O
H
26
H
25
OH
O
O
HCl
F
NH2
O
O
O
O
N
H
N
N
N
O
N
H
O
H
H
OH
O
27
Scheme 3: Synthesis of Compound 27. Reagents and conditions: (a) Boc-glycine,
EDCI, HOBt, DCM; (b) p-Nitrophenyl chloroformate, TEA, THF; (c) TEA, DMAP,
DCE, 60 °C; (d) dry HCl in EtOAc; (e) EDCI, HOBt, TEA, DCM.
5

2.2. Compounds design
Carbamate is a frequent used group in the design of many prodrugs [24, 25].
Capecitabine, a 5FU prodrug, is designed using carbamate group, which has better
stability than ester group [26]. Therefore, in order to improve the stability of BC-02,
the carbamate was used to replace of the ester bond of BC-02 to afford compound 12
(Fig. 1).
2.3. Biological evaluation of compound 12
Firstly, the CD13 inhibitory activity of compound 12 was determined using
ubenimex as the positive control. As showed in table 1, the IC50 value of potency of
compound 12 against CD13 from porcine kidney is 6.8-fold higher than ubenimex,
with an IC50 value of 1.35 μM.
Table 1
CD13 inhibitory activity of compound 12.
Cpd
IC50(μM)a toward CD13 from porcine kidney
12
1.35
9.21
Ubenimex
aAssays were performed in replicate (n ≥ 2); the standard deviation (SD) values are
<20% of the mean.
Four cancer cell lines including breast cancer cell MDA-MB-231, leukemia cell
HEL and KG1 and prostate cancer cell PC-3 were chosen to evaluate the
antiproliferative activity of compound 12, 5FU and ubenimex. Results illustrate that
compound 12 exhibits no antiproliferative effect against all the cell lines up to 200
μM (Table 2). It is not surprising because the carbamate linker may stable in vitro and
could not cleave and release 5FU during cell culture.
Table 2
Antiproliferative Activities of compound 12, 5-FU and Ubenimex (IC50 in μMa)
Cpd
12
MDA-MB-231
HEL
KG1
PC-3
>200
>200
>200
>200
6

5-FU
24.45
>500
27.55
>500
96.03
>500
41.05
>500
Ubenimex
aAssays were performed in replicate (n ≥ 2); the SD values are <20% of the mean.
As a frequently used group of prodrugs, carbamate was reported can be
metabolized in vivo [24-26]. Therefore, we next investigated the anti-tumor effect of
compound 12 in vivo in a mouse HCC H22 cell transplant model and human
hypopharyngeal adenocarcinoma cell line FaDu xenograft model. It is disappointed to
find that compound 12 exhibits no obvious antitumor activity in both animal models
with i.p. or oral administration (Fig. 2a-2d). We suppose the ineffective results may
due to the high structural stability of carbamate in plasma. Then we checked the
stability of compound 12 in human and rat plasma using a HPLC method. To our
expectation, compound 12 can not cleave in both human and rat plasma within 24 h
(Fig. S1), which meets our speculation that carbamate is too stable to cleave.
Figure 2. In vivo antitumor activities of compound 12. (a) Weights of tumors in
HCC H22 subcutaneous model (Student’s t-test). Error bars indicate standard error of
the mean (SEM); (b). Photographs of H22 tumors tissue; (c) Weights of tumors in
Fadu xenograft model, Error bars indicate SEM; (d). Photographs of Fadu tumor
tissues.
2.4. Biological evaluation of compound 16, 20 and 27
7

Although compound 12 is more stable than BC-02, the carbamate structure is too
stable to cleave both in vitro and in vivo, which results in ineffective anti-tumor
activity. Therefore, we aim to find a degradable linker which is more stable than the
ester of BC-02 as well as releases 5FU in a suitable speed for the further modification.
Firstly, replacement of amide bond of compound 12 by ester bond afforded compound
16 (Fig. 1). We speculated that the hydrolysis of ester bond of compound 16 can
accelerate the further hydrolysis of carbamate. On the other hand, we also designed
and synthesized compounds 20 and 27 by using the cyclic substituted ester and
carbonates.
Firstly, the CD13 inhibitory activity of compound 16, 20, 27 was screened using
the commercial CD13 enzyme. All of the compounds display higher inhibitory
activity than the positive control ubenimex (Table 3). Among which, compound 20
(IC50=0.096 μM) is about 100-fold more potent than ubenimex (IC50=9.21 μM).
Moreover, the CD13 inhibitory activity conducted with ES-2 cells (Table 3) also
confirms the superior activities of compound 20.
Table 3
CD13 inhibitory activity of compound 16, 20 and 27.
IC50(μM)a toward
CD13 from porcine
kidney
IC50(μM)a toward
CD13 on the surface
of ES-2 cells
Compds
16
20
1.29
0.096
0.787
9.21
-
0.24
-
27
Ubenimex
BC-02 [21]
16.27
1.68
0.18
aAssays were performed in replicate (n ≥ 2); the SD values are <20% of the mean.
In order to explore the interaction between target compound 20 and CD13, we
conducted the docking study using a Sybyl/Sketch module. Similar with ubenimex,
8

the hydroxyl group of 20 chelates with the zinc ion of CD13, the benzene ring and
isobutyl of 20 insert into the S1 and S1’ hydrophobic pockets respectively (Fig. 3). It
should be noted that the restricted conformation of 5-membered ring linker of 20
makes the 5FU moieties effectively occupy the S2’ bigger pocket (Fig. 3). This might
be the reason why 20 displayed a better inhibitory efficacy against CD13 than
ubenimex (Table 3).
Figure 3. Proposed binding preferences comparison of compound 20 at CD13
(derived by modification of PDB code 2DQM using Tripos SYBYL 8.0). 20 (white)
aligned with ubenimex (green) in a CD13 homology model. The green sphere is zinc
ion, and the dashed lines stand for the hydrogen bonds (atom types: H, white; N, blue;
O, red).
Six human solid tumor cell lines PLC/PRF/5, HUH7, HepG2, MHCC97H, MCF-
7 and ES-2 as well as three of haematological tumor cell lines HL-60, MV4-11 and
RS4;11 were chosen for the in vitro anti-tumor assay of compounds 16, 20 and 27.
The antiproliferative activities of 20 and 27 are comparable with that of 5FU while the
compound 16 displays much weaker activity (Table 4). The positive result displays
these three compounds can cleave and release 5FU during cell culture in vitro. Then
we tested the stability of these three compounds in human plasma. As shown in figure
4a-4c, compound 27 and 16 are completely cleaved within 2 h and 12 h, respectively.
Compound 20 cleaves much more slowly, with a total cleavage time of more than 24
h. It should be noted that 27 and 20 release 5FU as degradation product, while 16
releases compound 14 (structure shown in scheme 1) as degradation product. The in
9

vitro antiproliferative and stability study shows compound 20 owns a suitable
degradation linker, which makes it more stable than BC-02 in human plasma (the
completely cleavage time of BC-02 was 8 h) [21] and release 5FU in a proper rate.
Similar result is also observed in rat plasma. Therefore, compound 20 was chosen for
further in vivo biological evaluation.
Table 4
Antiproliferative Activities of compounds 16, 20, 27 and 5FU (IC50 in μMa).
Cell lines
16
20
27
5-FU
35.66
Ubenimex[20,21]
PLC/PRF/5
HUH7
-
29.05
34.89
12.02
53.31
9.92
-
>500
>500
>500
-
>200
>200
>200
52.91
-
7.42
6.03
11.77
4.14
-
14.53
4.13
HepG2
MHCC97H
MCF-7
ES-2
12.31
3.06
>500
>500
>500
>500
-
65.63
13.16
2.68
71.47
2.79
HL-60
184.70
119.90
>200
3.62
1.03
3.63
MV4-11
RS4;11
0.85
10.05
2.81
aAssays were performed in replicate (n ≥ 2); the SD values are <20% of the mean.
10

Figure 4. Stability of compound 16, 20 and 27 in human plasma or rat plasma; (a).
Points of compound 16 were obtained after 0, 0.25, 0.5, 1, 2, 4, 6, 8 and 12 h,
respectively; (b). Points of compound 27 were obtained after 0, 0.25, 0.5, 1 and 2 h,
respectively; (c). Points of compound 20 were obtained after 0, 0.25, 0.5, 1, 2, 4, 8, 12
and 24 h, respectively; (d). Points of compound 20 were obtained after 0, 0.25, 0.5, 1,
2, 4 and 12 h, respectively.
The mouse heptoma H22 tumor transplant model was utilized to evaluate the
antitumor efficacy of compound 20 in vivo. Capecitabine is the most widely used and
successful 5FU prodrug in clinical. Therefore, capecitabine and 5FU were used as
positive controls in this experiment. Compound 20 (100 μmol/kg, iv) displays the
comparable anti-tumor activity with positive control capecitabine (300 μmol/kg, po),
while it shows better antitumor activity than that of 5FU (150 μmol/kg, iv) (Fig. 5).
No significant toxicity or body weight loss is observed in the compound 20 treated
animals, one tumor tissue disappeared during compound 20 treatment. Two mice died
in 5FU treated group, which indicates its highly toxicity at dose of 150 μmol/kg.
11

Figure 5. In vivo antitumor activities of 20. (a) Weights of tumors in subcutaneous
model (*P < 0.05; Student’s t-test). Error bars indicate SEM. (b). Photographs of
tumor tissues. The “O” means no tumor detected, while the “X” means mice died in
the experiment.
CD13 is regarded as a key target in cancer migration, metastasis and
angiogenesis. Therefore, in this study we checked the anti-metastasis and anti-
angiogenesis of compound 20. The in vitro anti-invasion assay demonstrated that
compound 20 significantly inhibits ES-2 cell invasion at 5 μg/mL, which is more
potent than ubenimex at the same concentration (Fig. 6a). Data shown in figure 6b
demonstrated that compound 20 exhibits superior inhibitory effect on human
umbilical vein endothelial cells (HUVECs) tube formation than the positive control
ubenimex at the same dose (30 μM) (Fig. 6b). On the other hand, as described in
figure 6c, the number of microvessels sprouting from aortic rings treated with
compound 20 were much fewer than that of ubenimex (20 μg/mL). Together, our data
reveals the remarkble anti-angiogenesis activities of 20.
12

Figure 6. (a) Representative photographs of anti-invasion results after treatment of 20
and Ubenimex (positive control), the assay was performed on transwell chambers
coated with Matrigel using ES-2 cells; (b) Representative photographs of the HUVEC
capillary-like tubular network after treatment of 20 and ubenimex. (c) Representative
photographs of rat aortic rings after treatment of 20 and ubenimex.
In vivo anti-metastasis effect of compound 20 was performed in a H22
pulmonary metastasis tumor model. A significant decrease of H22 pulmonary nodes is
detected in compound 20 (150 μmol/kg, iv) treated group, revealing remarkable anti-
metastasis effect (Fig. 7). Capecitabine also decreases the number of H22 pulmonary
nodes, while its anti-metastasis activity is much weaker than compound 20 (Fig. 7).
13

Figure 7. In vivo anti-metastasis activity of 20. (a) Lung metastasis nodus numbers
after treatment of 20 and capecitabine (positive control) in the H22 hepatoma
pulmonary metastasis model (** P < 0.01; Student’s t-test). Error bars indicate SEM
(b) Representative photographs of lungs.
Based on the above in vitro and in vivo results of the potential anticancer effect of
compound 20, we next investigated whether it could be extending the lifespan of
tumor bearing mice. Survival curves suggested that treatment with compound 20 at
150 μmol/kg obviously increases the survival time with statistically significant
differences regarding the control group. In detail, the median survival time (MST) of
mice treated with 20 is increased 17 days, which is to 1.5-fold longer than control
group (11days) and comparable with that of capecitabine (300 μmol/kg) (Fig. 8).
Figure 8. Kaplan-Meier curves showing survival of mice treated with normal PBS
(pink), 20 (blue) and Capecitabine (yellow, positive control).
Previous studies demonstrated that CD13 was a CSC marker in HCC and
upregulation of CD13 was related to tumor resistance [8]. After treatment with 5FU,
the expression of CD13 was upregulated. And ubenimex can reduce the number of
CD13-positive cells thereby sensitizing cancer cells to the 5FU [8-10]. Therefore,
flow cytometry was utilized to evaluate the influence of compound 20 on CD13-
positive fraction of PLC/PRF/5 cells. As shown in figure 9, the number of CD13-
positive fraction of PLC/PRF/5 cells is increased by 5FU treatment, while compound
20 dramatically decreases the CD13-positive PLC/PRF/5 cells, revealing compound
20 might reduce the population of the CSC marker positive cells.
14

Figure 9. Expression of CD13 in PLC/PRF/5 cells after treatment of 5FU, ubenimex
and 20 for 48 h at 10 μM.
3. Conclusion
In our previous study, we reported that the hybrid compound BC-02 exhibited
remarkable in vitro and in vivo anti-tumor activity. However, the poor plasma stability
of BC-02 hindered its further development. In the present study, we optimize the
stability of BC-02 by modifying the linker between ubenimex and 5FU and get four
newly designed compounds. Among which, compound 20 is more stable than BC-02
and release 5FU in a proper rate. The CD13 inhibitory activity of compound 20 is
more potent than ubenimex and BC-02. The plasma stability study demonstrated that
compound 20 was more stable than that of BC-02. The in vitro and in vivo anticancer
study revealed that compound 20 exhibited moderate antitumor activity, which is
comparable with positive control 5-FU and capecitabine. Furthermore, compound 20
exhibits significant anti-metastasis, anti-angiogenesis and CD13-positive cell
15

elimination effects. In this study, we found that the stability of the linkers plays an
important role in the antitumor activity. If the linker is too stable to degrade, the
antitumor activity will be decreased even eliminated (compound 12), which means the
cytotoxicity activity mainly comes from the released 5FU. Our study provides an
important reference for the design of more druglike mutual prodrug for the purpose of
clinical application. Altogether, these results also suggest that compound 20 can be
deemed as a potential lead for further development of more potent analogs as
potential anticancer agents.
4. Experimental section
4.1. Materials and Methods.
All starting materials, reagents, and solvents were purchased from reagent
company and used without further purification unless otherwise stated. TLC plates
(60GF-254) was used to monitor all reactions. Silica gel or C18 silica gel was used for
column chromatography purification. NMR spectra were obtained on a JOE 400YH
400 MHz instrument with TMS as an internal standard, δ in parts per million and J in
hertz. High-resolution mass spectrometry was gathered by ocean university of China
in qingdao, China. The HPLC (Agilent 1260 Infinity, column: Agilent 5 TC-C18
column (5 μm, 250 mm × 4.6 mm) was used for purity detection of compounds. The
purity of all the target compounds were more than 95%.
The compounds 4 and 7 was prepared using the same procedures with our
previously work [20].
4.1.1. Methyl ((2S,3R)-3-((tert-butoxycarbonyl)amino)-2-hydroxy-4-phenylbut-anoyl)-
L-leuc-inate (5).
To a solution of compound 4 (1.0 g, 3.39 mmol) in dry DCM, EDCI (0.78 g,
4.06 mmol) and HOBt (0.55 g, 4.06 mmol) were added at 0 °C. After 30 min, methyl
L-leucinate hydrochloride (737.5 mg, 4.06 mmol) and TEA (0.75 g, 7.5 mmol) were
16

added. The reaction was allowed to stir overnight, then was washed with 1 M HCl
aqueous solution, and brine for 3 times, and dried over anhydrous MgSO4. Volatiles
were removed under vacuum to get the crude residue which was purified via flash
chromatography to afford the compound 5, a white solid powder (0.89 g, yield: 62%).
1H NMR (400 MHz, DMSO-d6) δ 7.65 (d, J = 8.6 Hz, 1H), 7.33-7.05 (m, 5H), 6.23 (d,
J = 9.4 Hz, 1H), 6.02 (d, J = 6.5 Hz, 1H), 4.34-4.15 (m, 1H), 4.00-3.85 (m, 1H), 3.84-
3.77 (m, 1H), 3.60 (s, 3H), 2.82-2.71 (m, 1H), 2.66-2.57 (m, 1H), 1.89-1.72 (m, 1H),
1.42-1.31 (s, 9H), 0.86-0.68 (m, 6H).
4.1.2.
((2S,3R)-3-((tert-butoxycarbonyl)amino)-2-hydroxy-4-phenylbutanoyl)-L-
leucine (6).
Compound 5 (0.80 g, 1.89 mmol) was dissolved in 15 mL of MeOH, then 2 mL
of 3 M KOH aqueous solution was added. The mixture was refluxed at 85 °C for 2 h.
MeOH was evaporated under vacuum after the reaction completed. The residue was
acidified with 1 M HCl to pH 5-6, then filtered. The precipitate was corresponding
acid compound 6, a white solid (0.69 g, yield: 90%). The crude material was used
directly for the next step without further purification. 1H NMR (400 MHz, DMSO-d6)
δ 7.72 (d, J = 8.6 Hz, 1H), 7.25-7.14 (m, 5H), 6.10 (d, J = 9.3 Hz, 1H), 4.30-4.22 (m,
1H), 3.92-3.84 (m, 1H), 3.79 (d, J = 2.8 Hz, 1H), 2.80-2.72 (m, 1H), 2.65-2.56 (m,
1H), 1.63-1.53 (m, 2H), 1.48-1.40 (m, 1H), 1.25 (s, 9H), 0.81 (dd, J = 21.4, 6.3 Hz,
6H).
4.1.3. (5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)-yl)methyl (4-nitrophenyl)
carbonate (8).
Compound 7 (1.0 g, 6.25 mmol) and TEA (0.95 mg, 9.38 mmol) were dissolved
in dry THF, then the solution of p-nitrophenyl chloroformate (1.9 g, 9.38 mmol) in
dry THF was added dropwise into the mixture at 0 °C. The reaction was allowed to
17

stir at room temperature overnight. The precipitate was filtered out, then filtrate was
concentrated to get the crude residue which was purified via flash chromatography to
afford the compound 8, a white solid powder (1.3 g, yield: 65%). 1H NMR (400 MHz,
DMSO-d6) δ 12.06 (d, J = 5.0 Hz, 1H), 8.33-8.27 (m, 2H), 8.13 (d, J = 6.6 Hz, 1H),
7.60-7.52 (m, 2H), 5.72 (s, 2H).
4.1.4. tert-butyl ((5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methyl) ethane-
1,2-di-yldicarbamate (9).
To a solution of compound 8 (1.0 g, 3.08 mmol) in DCE, Boc-ethylenediamine
(0.59 g, 3.7 mmol), TEA (0.37 g, 3.7 mmol) and 4-dimethylaminopyridine (DMAP)
(0.45 g, 3.7 mmol) were added. The reaction was allowed to stir at 60 °C for 2 h. 15
mL of DCM was added to the mixture, then the DCM layer was washed with
saturated brine (2 × 30 mL), and dried over anhydrous MgSO4. Volatiles were
removed under vacuum to get the crude residue which was purified via flash
chromatography to afford the compound 9, a white solid powder (0.8 g, yield: 75%).
1H NMR (600 MHz, DMSO-d6) δ 11.95 (s, 1H), 8.06 (d, J = 6.5 Hz, 1H), 7.51 (t, J =
5.5 Hz, 1H), 6.81 (t, J = 5.7 Hz, 1H), 5.49 (s, 2H), 3.08-2.91 (m, 4H), 1.36 (s, 9H).
4.1.5.
(5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)-yl)methyl
(2-(λ2-azane-
yl)ethyl)car-bamate hydrochloride (10).
Compound 9 (0.73 g, 2.1 mmol) was dissolved in a solution of EtOAc (20 mL)
saturated by dry HCl gas, then the reaction was stired at room temperature for 30 min.
The filtered precipitate was washed by diethyl ether to give compound 10, a white
solid powder (0.54 g, yield: 92%). The crude material was used directly for the next
1
step without further purification. H NMR (600 MHz, DMSO-d6) δ 11.98 (s, 1H),
8.22-8.01 (m, 4H), 7.73 (t, J = 5.7 Hz, 1H), 5.51 (s, 2H), 3.27 (q, J = 6.2 Hz, 2H),
2.84 (s, 2H).
18

4.1.6. (5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)-yl)methyl ((6R,7S,10S)-6-
benzyl-7-hydroxy-10-isobutyl-2,2-dimethyl-4,8,11-trioxo-3-oxa-5,9,12-triazatetradec-
na-14-yl)carbamate (11).
Compound 6 (0.41 g, 1 mmol) was dissolved in dry DCM (15 mL), EDCI (0.23
g, 1.2 mmol) and HOBt (0.16 g, 1.2 mmol) were added at 0 °C. 30 min later,
compound 10 (0.34 g, 1.2 mmol) and TEA (0.22 g, 2.2 mmol) were added. The
reaction was stirred at room temperature overnight. Then was washed with 1 M HCl
aqueous solution (2 × 30 mL), and brine (2 × 30 mL), and dried over anhydrous
MgSO4. Volatiles were removed under vacuum to get the crude residue which was
purified via flash chromatography to afford the compound 11, a white solid powder
1
(0.34 g, yield: 53%). H NMR (400 MHz, DMSO-d6) δ 11.95 (s, 1H), 8.07-7.94 (m,
2H), 7.57 (d, J = 9.0 Hz, 1H), 7.49 (t, J = 4.8, 3.8 Hz, 1H), 7.27-7.12 (m, 5H), 6.15 (d,
J = 9.3 Hz, 1H), 5.93 (d, J = 6.1 Hz, 1H), 5.43 (s, 2H), 4.33-4.18 (m, 1H), 3.91 (q, J =
9.2, 8.5 Hz, 1H), 3.82-3.74 (m, 1H), 3.01 (dt, J = 21.9, 6.4 Hz, 4H), 2.74 (dd, J =
13.5, 7.3 Hz, 1H), 2.62 (dd, J = 13.5, 7.8 Hz, 1H), 1.55-1.34 (m, 3H), 1.24 (s, 9H),
0.79 (dd, J = 17.8, 6.2 Hz, 6H).
4.1.7. (5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)-yl)methyl(2-((S)-2-((2S,3R)-3-
amino-2-hydroxy-4-phenylbutanamido)-4-methylpentanami-do)ethyl)carbamate (12).
Compound 11 (0.30 g, 0.47 mmol) was dissolved in a solution of EtOAc (10
mL) saturated by dry HCl gas. The solution was stirred at room temperature for 3 h.
The filtered precipitate was washed by diethyl ether, then the residue was alkalified
with TEA to pH 8-9. The crude residue was purified via flash chromatography to
1
afford the compound 12, a white solid powder (0.23 g, yield: 87%). H NMR (400
MHz, DMSO-d6) δ 8.16 (t, J = 5.5 Hz, 1H), 8.03 (d, J = 6.5 Hz, 1H), 7.90 (d, J = 8.2
Hz, 1H), 7.49 (t, J = 5.5 Hz, 1H), 7.32-7.16 (m, 5H), 5.43 (s, 2H), 4.24-4.15 (m, 1H),
3.90 (d, J = 3.1 Hz, 1H), 3.10-3.00 (m, 2H), 3.00-2.91 (m, 3H), 2.91-2.81 (m, 1H),
2.81-2.70 (m, 1H), 1.64-1.38 (m, 3H), 0.85-0.75 (m, 6H). 13C NMR (100 MHz,
DMSO-d6): δ 172.5, 171.4, 171.1, 136.8, 129.9, 129.6, 129.1, 127.4, 68.7, 60.2, 54.7,
51.5, 41.5, 35.1, 24.6, 23.4, 22.4. HRMS (AP-ESI) m/z calcd for C24H34O7N6F
[M+H]＋ 537.24634, found 537.24675. Retention time: 14.1 min, eluted with 30%
acetonitrile/70% water (containing 0.1% triethylamine and 0.15% trifluoroacetic
acid).
19

4.1.8.
yl)carbamate (14).
The title compound was prepared by the same procedure described for 9 except
(5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methyl(2-hydroxyeth-
1
using 2-aminoethanol as a white solid, 71% yield. H NMR (400 MHz, DMSO-d6) δ
11.93 (s, 1H), 8.05 (d, J = 6.6 Hz, 1H), 7.46 (t, J = 5.7 Hz, 1H), 5.44 (s, 2H), 4.68 (t, J
= 5.6 Hz, 1H), 3.35-3.29 (m, 2H), 3.01-2.98 (m, 2H).
4.1.9.
2-((((5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methoxy)carbonoyl)-
amino)-ethyl((2S,3R)-3-((tert-butoxycarbonyl)amino)-2-hydroxy-4-phenylbutanoyl)-
L-leucinate (15).
The title compound was prepared by the same procedure described for 11 as a
white solid, 48% yield. 1H NMR (400 MHz, DMSO-d6) δ 11.95 (s, 1H), 8.04 (d, J =
6.6 Hz, 1H), 7.89 (d, J = 8.6 Hz, 1H), 7.66 (t, J = 5.6 Hz, 1H), 7.26-7.11 (m, 6H),
6.10 (d, J = 9.3 Hz, 1H), 5.98 (d, J = 6.6 Hz, 1H), 5.72 (s, 1H), 5.45 (s, 2H), 4.43-4.28
(m, 1H), 4.04-3.94 (m, 2H), 3.94-3.82 (m, 1H), 3.84-3.76 (m, 1H), 3.19 (d, J = 5.6 Hz,
3H), 2.81-2.69 (m, 1H), 2.68-2.55 (m, 1H), 1.67-1.48 (m, 3H), 1.50-1.37 (m, 2H),
1.24 (s, 9H), 0.86-0.72 (m, 8H).
4.1.10. 2-((((5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methoxy)carbonoyl)-
amino)-ethyl((2S,3R)-3-amino-2-hydroxy-4-phenylbutanoyl)-L-leucinate (16).
The title compound was prepared by the same procedure described for 12 as a
1
white solid, 85% yield. H NMR (400 MHz, DMSO-d6) δ 8.30 (d, J = 7.4 Hz, 1H),
8.03 (d, J = 6.5 Hz, 1H), 7.36-7.20 (m, 5H), 5.44 (s, 1H), 4.30-4.17 (m, 1H), 4.09-
3.88 (m, 2H), 3.51-3.40 (m, 2H), 3.21-3.15 (m, 1H), 2.93-2.83 (m, 1H), 2.83-2.74 (m,
1H), 1.66-1.37 (m, 3H), 0.83 (dd, J = 9.6, 6.0 Hz, 6H). 13C NMR (101 MHz, DMSO-
d6) δ 172.62, 171.82, 155.82, 149.90, 137.01, 129.96, 129.17, 127.42, 71.23, 69.13,
63.66, 54.90, 50.93, 46.01, 24.74, 23.12, 22.08, 9.10. HRMS (AP-ESI) m/z calcd for
C24H33O8N5F [M+H]＋ 538.23102, found 538.23077. Retention time: 5.9 min, eluted
with 35% acetoni-trile/65% water (containing 0.1% triethylamine and 0.15%
trifluoroacetic acid).
20

4.1.11. 1-(tert-butyl)2-((5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methyl) (S)-
pyrroli-dine-1,2-dicarboxylate (17).
The title compound was prepared by the same procedure described for 5 except
1
using 7 and N-Boc-L-proline as a white solid, 61% yield. H NMR (400 MHz,
DMSO-d6) δ 12.06 (d, J = 5.0 Hz, 1H), 8.17 (d, J = 6.6 Hz, 1H), 5.65 (s, 2H), 4.18 (dd,
J = 8.7, 3.8 Hz, 2H), 4.08-3.97 (m, 1H), 2.27-2.17 (m, 2H), 1.83-1.81 (m, 2H), 1.37 (s,
9H).
4.1.12. (5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methyl-L-prolinate hydro-
chloride (18).
The title compound was prepared by the same procedure described for 10 as a
1
white solid, 88% yield. H NMR (400 MHz, DMSO-d6) δ 12.03 (d, J = 5.0 Hz, 1H),
8.18 (d, J = 6.6 Hz, 1H), 5.67 (d, J = 1.4 Hz, 2H), 4.37 (t, J = 8.0 Hz, 1H), 3.20-3.14
(m, 2H), 2.27-2.12 (m, 1H), 2.02-1.92 (m, 1H), 1.92-1.79 (m, 3H).
4.1.13. (5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methyl ((2S,3R)-3-((tert-
butoxy-carbonyl)amino)-2-hydroxy-4-phenylbutanoyl)-L-leucyl-L-prolinate (19).
The title compound was prepared by the same procedure described for 11 as a
1
white solid, 55% yield. H NMR (400 MHz, DMSO-d6) δ 11.99 (s, 1H), 8.08 (d, J =
6.7 Hz, 1H), 7.58 (d, J = 8.8 Hz, 1H), 7.28-7.09 (m, 6H), 6.17 (d, J = 9.5 Hz, 1H),
6.00 (d, J = 6.7 Hz, 1H), 5.63 (d, J = 10.2 Hz, 1H), 5.49 (d, J = 10.1 Hz, 1H), 4.63-
4.51 (m, 1H), 4.34 – 4.24 (m, 1H), 3.95-3.83 (m, 1H), 3.79-3.71 (m, 1H), 3.67-3.55
(m, 1H), 3.51-3.40 (m, 1H), 2.73 (dd, J = 13.2, 7.1 Hz, 1H), 2.61 (dd, J = 13.2, 7.7 Hz,
1H), 2.21-2.05 (m, 1H), 1.95-1.78 (m, 3H), 1.65-1.49 (m, 1H), 1.47-1.31 (m, 2H),
1.24 (s, 9H), 0.81 (dd, J = 6.6, 3.4 Hz, 6H).
4.1.14. (5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-1(2H)yl)methyl ((2S,3R)-3-amino-2-
hydro-xy-4-phenylbutanoyl)-L-leucyl-L-prolinate hydrochloride (20).
21

The compound 20 was prepared by the same procedure described for 12 except
alkalifying as a white solid, 82% yield. 1H-NMR(400 MHz DMSO-d6): δ 12.01 (d,
J=5.0 Hz, 1H), 8.15-8.12 (m, 2H), 8.02 (s, 3H), 7.36-7.25 (m, 5H), 6.69 (s, 1H), 5.72-
5.51 (m, 2H), 4.52-4.46 (m, 1H), 4.34-4.31 (m, 1H), 4.04-3.97 (m, 1H), 3.74-3.68 (m,
1H), 3.53-3.50 (m, 2H), 2.95-2.86 (m, 2H), 2.21-2.12 (m, 1H), 1.99-1.82 (m, 3H),
1.69-1.62 (m, 1H), 1.52-1.36 (m, 2H), 0.92-0.87 (m, 6H). 13C NMR (101 MHz,
DMSO-d6): δ 171.8, 171.5, 170.9, 158.2, 149.8, 141.3, 137.0, 130.1, 129.7, 129.3,
127.6, 71.7, 68.7, 59.1, 54.9, 49.5, 47.2, 35.3, 29.1, 25.3, 24.7, 23.7, 22.4. HRMS
(AP-ESI) m/z calcd for C26H35O7N5F [M + H]+ 548.25140, found 548.25150.
Retention time: 12.6 min, eluted with 30% acetonitrile/70% water (containing 0.1%
triethylamine and 0.15% trifluoroacetic acid).
4.1.15. tert-butyl (2-((4-(hydroxymethyl)phenyl)amino)-2-oxoethyl)carbamate (22).
The title compound was prepared by the same procedure described for 5 except
1
using 4-aminobenzenemethanol and Boc-glycine as a white solid, 43% yield. H
NMR (400 MHz, DMSO-d6) δ 10.00 (s, 1H), 7.58 (d, J = 8.2 Hz, 2H), 7.33 (d, J = 8.3
Hz, 2H), 7.04 (t, J = 6.1 Hz, 1H), 5.08 (s, 2H), 3.72 (d, J = 6.2 Hz, 2H), 1.39 (s, 9H).
4.1.16. tert-butyl (2-((4-((((4-nitrophenoxy)carbonyl)oxy)methyl)phenyl)amino)-2-
oxoethyl) carbamate (23).
The title compound was prepared by the same procedure described for 8 except
using 22 as a white solid, 75% yield. 1H NMR (400 MHz, DMSO-d6) δ 10.03 (s, 1H),
8.36-8.27 (m, 2H), 7.63 (d, J = 8.2 Hz, 2H), 7.62-7.53 (m, 2H), 7.42 (d, J = 8.3 Hz,
2H), 7.06 (t, J = 6.2 Hz, 1H), 5.25 (s, 2H), 3.73 (d, J = 6.2 Hz, 2H), 1.40 (s, 9H).
4.1.17. tert-butyl (2-((4-(((((5-fluoro-2,4-dioxo-3,4-dihydropyrimidin1(2H)yl)meth-
oxy)carb-onyl)oxy)methyl)phenyl)amino)-2-oxoethyl)carbamate (24).
22

The title compound was prepared by the same procedure described for 9 except
1
using 7 as a white solid, 65% yield. H NMR (400 MHz, DMSO-d6) δ 12.02 (s, 1H),
10.01 (s, 1H), 8.14 (d, J = 6.5 Hz, 1H), 7.60 (d, J = 8.2 Hz, 2H), 7.34 (d, J = 8.2 Hz,
2H), 7.05 (t, J = 6.2 Hz, 1H), 5.61 (s, 2H), 5.11 (s, 2H), 3.72 (d, J = 6.1 Hz, 2H), 1.39
(s, 9H).
4.1.18. 4-(2-aminoacetamido)benzyl ((5-fluoro-2,4-dioxo-3,4-dihydropyrimidin-
1(2H)yl)me-thyl) carbonate hydrochloride (25).
The title compound was prepared by the same procedure described for 10 as a
white solid, 84% yield. The crude material was used directly for the next step without
further purification.
4.1.19. tert-butyl((2R,3S)-4-(((S)-1-((2-((4-(((((5-fluoro-2,4-dioxo-3,4-dihydropyrimi-
din-1(2H)-yl)methoxy)carbonyl)oxy)methyl)phenyl)amino)-2-oxoethyl)amino)-4-
methyl-1-oxopentan-2-yl)amino)-3-hydroxy-4-oxo-1-phenylbutan-2-yl)carbamate
(26).
The title compound was prepared by the same procedure described for 11 as a
1
white solid, 52% yield. H NMR (400 MHz, DMSO-d6) δ 12.00 (d, J = 5.0 Hz, 1H),
9.92 (s, 1H), 8.37 (t, J = 5.9 Hz, 1H), 8.11 (d, J = 6.5 Hz, 1H), 7.72 (d, J = 8.5 Hz,
1H), 7.56 (d, J = 8.1 Hz, 2H), 7.32-7.13 (m, 7H), 6.20 (d, J = 9.3 Hz, 1H), 5.95 (d, J =
6.2 Hz, 1H), 5.57 (s, 2H), 5.08 (s, 2H), 4.43-4.28 (m, 1H), 3.98-3.89 (m, 1H), 3.84-
3.79 (m, 2H), 2.76 (dd, J = 13.2, 7.2 Hz, 1H), 2.63 (dd, J = 13.1, 7.9 Hz, 1H), 1.67-
1.36 (m, 4H), 1.25 (s, 9H), 0.82 (dd, J = 18.2, 6.4 Hz, 6H).
4.1.20.
4-(2-((S)-2-((2S,3R)-3-amino-2-hydroxy-4-phenylbutanamido)-4-methyl-
pentanamido) acetamido)benzyl ((5-fluoro-2,4-dioxo-3,4-dihydropyri-midin-1(2H)-
yl)methyl) carbo-nate hydrochloride (27).
The title compound was prepared by the same procedure described for 12 except
alkalifying as a white solid, 82% yield. 1H NMR (400 MHz, DMSO-d6) δ 12.01 (d, J
= 4.9 Hz, 1H), 10.00 (s, 1H), 8.54 (t, J = 5.9 Hz, 1H), 8.22-8.02 (m, 5H), 7.62 (d, J =
8.2 Hz, 2H), 7.41-7.29 (m, 7H), 6.75 (s, 1H), 5.62 (s, 2H), 5.12 (s, 2H), 4.35-4.25 (m,
1H), 4.05-4.03 (m, 1H), 3.87 (d, J = 6.4 Hz, 2H), 3.01-2.90 (m, 2H), 1.73-1.53 (m,
23

3H), 0.90 (dd, J = 9.1, 6.4 Hz, 6H). 13C NMR (101 MHz, DMSO-d6) δ 172.72, 171.68,
168.23, 158.10, 157.84, 153.99, 149.78, 141.15, 139.66, 138.85, 136.90, 130.12,
129.99, 129.84, 129.79, 129.16, 127.44, 119.43, 73.89, 69.92, 68.77, 54.73, 52.15,
43.22, 41.01, 35.20, 24.69, 23.44, 22.47. HRMS (AP-ESI) m/z calcd for C31H38O9N6F
[M + H]+ 657.26758, found 657.26758. Retention time: 7.0 min, eluted with 35%
acetonitrile/65% water (containing 0.1% triethylamine and 0.15% trifluoroacetic
acid).
4.2. Biological assays
4.2.1. CD13 inhibition assay[27].
The effect of target compounds on CD13 activity was determined by incubating
microsomal aminopeptidase from porcine kidney microsomes (Sigma) with PBS (50
mM, pH 7.2) as a buffer at 37℃. First, the compounds were diluted to graded
concentrations and added to 96-well microcrystal plates, then CD13 solution was
added before incubating at 37℃ for 5 min, and finally L-leucine-p-nitroaniline
solution was added as a substrate. The plates were incubated for 30 min at 37°C.
Measured the absorbance at 405nm with a Micro-plate Reader (Thermo Fisher,
Shanghai, China) to infer the inhibitory effects of the compounds on the enzyme.
The activity of CD13 on the surface of ES-2 cells was determined by incubating
2.5 × 106/mL cells in 1 × PBS buffer with different concentrations of test compounds
and a certain concentration of substrate for 1 hour at 37℃. Similarly, the
measurement was performed in 96-well microcrystal plates, the substrate was L-
leucine-p-nitroaniline, and the absorbance of the well was measured at 405 nm.
4.2.2. Cell proliferation analysis
The cells in the 96-well plate were incubated with serially diluted compounds in
an incubator for 3 days, then 0.5% MTT solution was added to each well and
incubated the plate for 4 hours. Finally, 150 μL of dimethyl sulfoxide was added to
each well, gently shaking 15 min, then measured the absorbance at 570 nm. All the
above cell lines were cultured in RPMI1640 or DMEM medium and incubated in an
incubator at 37℃ and 5% CO2. All media were supplemented with 10% FBS, 100
μg/mL penicillin and 100 μg/mL streptomycin.
24

4.2.3. Stability of target compounds in human plasma
Human plasma samples were stored at -20°C and thawed gradually to room
temperature before processing. Compounds 12, 16, 20 and 27 were dissolved in
acetonitrile/water to make a 4 mg/mL stock solution, then added to human plasma to a
final concentration of 0.4 mg/mL and incubated at 37°C over a period of 24 h. The
incubation was terminated at the specific time points by adding acetonitrile. The
mixture was vortexed for 30 seconds and centrifuged (12,000 rpm/10 min), then was
filtered through 0.22 μm filters., and injected into the HPLC system (Agilent 1260
HPLC) for analysis with an Alltima C18 (5 μm, 4.6 mm × 250 mm) column in a
gradian of 25% acetonitrile/75% water (containing 0.1% triethylamine and 0.15%
trifluoroacetic acid) over 15 min at 262 nm.
4.2.4. In vivo activity in H22 tumor transplant model
All experiments in vivo were performed according to the institutional ethical
guidelines on animal care and approved by the Institute Animal Care and Use
Committee at ocean university of China. Before conducting animal experiments, all
Kuming mice were placed in a specific pathogen-free (SPF) environment and
provided with sterile food and water. Cultured H22 cells with RPMI1640 medium
(containing 10% FBS, 100 μg/mL penicillin and 100 μg/mL streptomycin) in a 37°C,
5% CO2 incubator. Cells were diluted to 1×107/mL with PBS and injected into the
mice subcutaneously (0.2 mL). Three days later, the mice were randomly divided into
several treatment groups and vehicle group. Mice in the treatment groups were treated
with the specific dose of the test compounds for five days a week. Vehicle group was
given the same volume of PBS in the same way. After two weeks adminstration, the
mice were sacrificed. Tumor tissues were excised from the subcutaneous part of each
mouse and weighed. Tumor inhibition rate is calculated by (1-average tumor weight
in treatment group/average tumor weight in PBS group) *100%.
4.2.5. In vivo activity of FaDu xenograft model
Human FaDu cells were stored in DMEM medium containing 10% FBS at 37°C
in a 5% CO2 humidified incubator. FaDu cells were counted, prepared into cell
suspension with PBS, and inoculated into BALB/c-nu mice (female, 5-6 weeks old,
Slac Laboratory Animal, Shanghai, China) subcutaneously (5×106 cells per mouse).
25

When the tumor volume was rose to approximately 100 mm3, the mice were randomly
divided into two groups. The mice in the treatment group were given a specific
concentration of the test compounds, and the mice in the vehicle group were given the
same volume of PBS solution in the same way. Mice were sacrificed by cervical
dislocation and tumor tissues were harvested. Tumor inhibition rate is calculated by
(1-average tumor weight in treatment group/average tumor weight in PBS group)
*100%.
4.2.6. Docking study
The docking study of compound 20 and CD13 (PDB code 2DQM) was
performed using Tripos SYBYL 8.0. Briefly, the structure of compound 20 was
constructed using the Sybyl/Sketch module and optimized using Powell’s method
with the Tripos force field with convergence criterion set at 0.05 kcal/(Å mol) and
assigned charges with the Gasteiger-Huckel method. Molecular docking was
performed by the Sybyl/FlexX module.
4.2.7. Invasion assay
At beginning, matrigel (BD Biosciences, NJ) was used to coated the transwell
filters. The cells were washed 3 times with serum-free medium (RPMI-1640), counted,
prepared into cell suspension, and added to the upper compartment of the chamber
(400 μL per well), and then 100 μL of the test compounds at the specific
concentration were added. Added 750 μL of conditioned medium from ES-2 cells to
the bottom compartment of the chamber and incubated the plate in a 37°C incubator
for 24 hours. The transwell was washed twice with PBS, and stained with crystal
violet solution (0.1%) before observed under a microscope.
4.2.8. In vitro HUVEC tuber formation assay [28]
Matrigel, a basement membrane matrix (BD Biosciences, NJ), was polymerized
at 37°C for 30 min. Human umbilical vein endothelial cells (HUVEC) were
suspended in M199 medium containing 5% FBS were inoculated on Matrigel (40,000
per well), treated with the test compounds at the specified concentration, and
incubated at 37°C and 5% CO2 for 6 hours. The cells were photographed under a
phase contrast microscope at 200 × magnification.
4.2.9. Rat thoracic aorta rings (TARs) assay [27]
26

The thoracic aortas of Wistar rats (male, 8 to 10 weeks old) were removed from
the connective tissues, washed with sterile PBS and cut transversely into 1 mm thick
rings. The aortic rings were immersed in matrigel in the wells of 96-well plate, then
another 100 μL of matrigel was added. After the second layer of matrigel set, the plate
was incubated at 37 °C with 5% CO2 for 30 minutes. The test compounds were added
to the wells every other day for treatment, and on the 10th day, the sprouting of
microvessels was photographed under a phase-contrast microscope at 200 ×
magnification.
4.2.10. In vivo activity of H22 pulmonary metastasis model
A heat lamp was used to dilate the tail vein of mice (6 weeks old, female), and
then the mice were transferred to a restraining device. Diluted H22 cells with PBS to
7.5 × 106 cells/mL according to the aforementioned method, and injected 0.1 mL into
the tail vein of each mouse. Tumor cells were allowed to grow for 10 days. On day 11,
the mice received specific concentrations of compounds for 2 weeks (5 days
administration in one week). After that, the mice were sacrificed. The lung tissues
were removed and weighed and then fixed in Bouin's fixation solution for one day
before the lung metastasis lymph nodes.
4.2.11. Survival time in mice bearing H22 cells
Kunming mice (6 weeks old, female) were inoculated with H22 cells (4.0 × 106
cells/mouse) by intraperitoneal injection. Three days later, the mice were randomly
divided into treatment groups and vehicle group. The mice in the treatment group
were treated with the specified dose of the test compounds for 14 consecutive days,
and the mice in the control group were given the same volume of PBS by the same
method. The median survival time (MST) of the mice was measured to infer the anti-
tumor activity of the compound.
4.2.12. Determination of CD13-positive fraction of PLC/PRF/5 cells
2 × 105 cells per well of PLC/PRF/5 cells were seeded in 6-well plates and
allowed to grow for 12 h, then added test compounds. After 48 hours, the cells were
collected, washed twice with cold PBS. After that, PE anti-human CD13 was added to
27

1× PBS suspension, and the mixture was incubated at 4°C for 15 minutes. Finally,
the cells were washed and analyzed with flow cytometry.
Supporting Information. Stability of compound 20 in human and rat plasma. IC50
curves of compound 16, 20, 27 and 5FU for cancer cell lines. IC50 curves of
1
compound 12, 16, 20, 27 and Ubenimex for CD13. H and 13C NMR spectral
information of target compounds12, 16, 20 and 27. HPLC analysis chromatograms of
target compound 12, 16, 20 and 27.
AUTHOR INFORMATION
* Corresponding Authors
E-mail addresses: jiangyuqi@ouc.edu.cn (Y. Jiang), lixiaoyang@ouc.edu.cn (X. Li).
1 contributed equally to this manuscript.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGMENTS
This work is supported by The Marine S&T Fund of Shandong Province for Pilot
National Laboratory for Marine Science and Technology (Qingdao) (No.
2018SDKJ0403), National Natural Science Foundation of China (Grant no.
82003584), Natural Science Foundation of Shandong Province (ZR2020QH343).
ABBREVIATIONS
ADCs, Antibody conjugated drugs; APN, Aminopeptidase N; (Boc)2O, Di-tert-butyl
dicarbonate; CSCs, Cancer stem cells; DCE, Dichloroethane; DCM, Dichloromethane;
DMAP, Dimethylaminopyridine; EDCI, 1-(3-Dimethylaminopropyl)-3-ethylcarbo-
diimide; 5FU, Fluorouracil; HCl, Hydrochloride; HCC, Hepatocellular carcinoma;
HOBt, 1-hydroxybenzotriazole; HPLC, High Performance Liquid Chromatography;
HUVEC, Human umbilical vein endothelial cells; MgSO4, Magnesium sulfate; MST,
Median survival time; NGR, ASN-GLY-ARG; PDCs, Peptide conjugated drugs; ROS,
Reactive oxygen species; SD, Standard deviation; SEM, Standard error of the mean;
TARs, Thoracic aorta rings; TDC, Target conjugated drugs; TEA, Triethylamine;
THF, Tetrahydrofuran; TNF, tumor necrosis factor.
28

REFERENCES
[1]
[2]
[3]
B. Bauvois, D. Dauzonne, Aminopeptidase-n/cd13 (ec 3.4.11.2) inhibitors: Chemistry,
biological evaluations, and therapeutic prospects, Med. Res. Rev. 26 (2006) 88-130.
P. Mina-Osorio,The moonlighting enzyme cd13: Old and new functions to target,
Trends Mol. Med. 14 (2008) 361-371.
M. Ghosh, R. Lo, I. Ivic, B. Aguilera, V. Qendro, C. Devarakonda, L. H. Shapiro,
Cd13 tethers the iqgap1-arf6-efa6 complex to the plasma membrane to promote arf6
activation, β1 integrin recycling, and cell migration, Sci. Signal. 12 (2019) eaav5938.
L. Guzman-Rojas, R. Rangel, A. Salameh, J. K. Edwards, E. Dondossola, Y. G. Kim,
A. Saghatelian, R. J. Giordano, M. G. Kolonin, F. I. Staquicini, E. Koivunen, R. L.
Sidman, W. Arap, R. Pasqualini,Cooperative effects of aminopeptidase n (cd13)
expressed by nonmalignant and cancer cells within the tumor microenvironment,
Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 1637-1642.
[4]
[5]
E. Dondossola, R. Rangel, L. Guzman-Rojas, E. M. Barbu, H. Hosoya, L. S. St John,
J. J. Molldrem, A. Corti, R. L. Sidman, W. Arap, R. Pasqualini,Cd13-positive bone
marrow-derived myeloid cells promote angiogenesis, tumor growth, and metastasis,
Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 20717-20722.
[6]
[7]
Q. Guo, X. Li, M. N. Cui, J. L. Sun, H. Y. Ji, B. B. Ni, M. X. Yan, Cd13: A key
player in multidrug resistance in cancer chemotherapy, Oncol. Res. 28 (2020) 533-
540.
B. Hu, Y. Xu, Y. C. Li, J. F. Huang, J. W. Cheng, W. Guo, Y. Yin, Y. Gao, P. X.
Wang, S. Y. Wu, J. Zhou, J. Fan, X. R. Yang, Cd13 promotes hepatocellular
carcinogenesis and sorafenib resistance by activating hdac5-lsd1-nf-κb oncogenic
signaling, Clin. Transl. Med. 10 (2020) e233-249.
[8]
[9]
N. Haraguchi, H. Ishii, K. Mimori, F. Tanaka, M. Ohkuma, H. M. Kim, H. Akita, D.
Takiuchi, H. Hatano, H. Nagano, G. F. Barnard, Y. Doki, M. Mori,Cd13 is a
therapeutic target in human liver cancer stem cells, J. Clin. Invest. 120 (2010) 3326-
3339.
M. Yamashita, H. Wada, H. Eguchi, H. Ogawa, D. Yamada, T. Noda, T. Asaoka, K.
Kawamoto, K. Gotoh, K. Umeshita, Y. Doki, M. Mori,A cd13 inhibitor, ubenimex,
synergistically enhances the effects of anticancer drugs in hepatocellular carcinoma,
Int. J. Oncol. 49 (2016) 89-98.
29