A Novel Model System for Understanding Anticancer Activity of Hypoxia-Activated Prodrugs

Article abstract of DOI:10.1021/acs.molpharmaceut.0c00232

Reports on the comprehensive factors for design considerations of hypoxia-activated prodrugs (HAPs) are rare. We introduced a new model system composed of a series of highly water-soluble HAPs, providing a platform to comprehensively understand the interaction between HAPs and hypoxic biosystems. Specifically, four kinds of new HAPs were designed and synthesized, containing the same biologically active moiety but masked by different bioreductive groups. Our results demonstrated that the activity of the prodrugs was strongly dependent on not only the molecular structure but also the hypoxic tumor microenvironment. We found the presence of a direct linear relationship between cytotoxicity of the HAPs and the reduction potential of whole molecule/oxygen concentration/reductase expression. Moreover, limited blood vasculature in hypoxic regions was also a critical barrier for effective activation of the HAPs. This study offers a comprehensive insight into understanding the design factors required for HAPs.

Full text of DOI:10.1021/acs.molpharmaceut.0c00232

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Article
A Novel Model System for Understanding Anti-
Cancer Activity of Hypoxia-Activated Prodrugs
Zhiyuan Sun, Haoqi Zhang, Jin Wu, Fangli Gao, Congcong Zhang,
Xueyan Hu, Qiqi Liu, Yonghua Wei, Jie Zhuang, and Xinglu Huang
Mol. Pharmaceutics, Just Accepted Manuscript • Publication Date (Web): 30 Apr 2020
Downloaded from pubs.acs.org on May 1, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
A Novel Model System for Understanding Anti-Cancer
Activity of Hypoxia-Activated Prodrugs
1
1,2
1
1
3
1
Zhiyuan Sun , Haoqi Zhang , Jin Wu , Fangli Gao , Congcong Zhang , Xueyan Hu ,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1,2
1,2
1
Qiqi Liu , Yonghua Wei Jie Zhuang *, Xinglu Huang *
1Key Laboratory of Bioactive Materials, Ministry of Education, College of Life
Sciences, and State Key Laboratory of Medicinal Chemical Biology, Nankai
University, Tianjin 300071, China.
2College of Medicine, Nankai University, Tianjin 300071, China.
3College of Chemistry and Material Science, Shandong Agricultural University,
Taian, P.R.China 271018
*
To whom correspondence may be addressed. Email: huangxinglu@nankai.edu.cn or
zhuangj@nankai.edu.cn
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 2 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
TOC
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 3 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
ABSTRACT
Reports on the comprehensive factors for design considerations of hypoxia-activated
prodrugs (HAPs) are rare. We introduced a new model system composed of a series
of highly water-soluble HAPs, providing a platform to comprehensively understand
the interaction between HAPs and hypoxic biosystems. Specifically, four kinds of
new HAPs were designed and synthesized; containing the same biologically active
moiety but masked by different bioreductive groups. Our results demonstrated the
activity of the prodrugs was strongly dependent on not only molecular structure but
also the hypoxic tumor microenvironment. We found the presence of a direct linear
relationship between cytotoxicity of the HAPs and the reduction potential of whole
molecule/oxygen concentration/reductase expression. Moreover, limited blood
vasculature in hypoxic regions was also a critical barrier for effective activation of the
HAPs. This study offers a comprehensive insight to understanding the design factors
required for HAPs.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Keywords: hypoxia; prodrugs; drug delivery; tumor; mechanism
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 4 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
Hypoxia is a common feature in solid tumors and plays an important role in the
occurrence and development of tumors, including but not limited to the promotion of
tumor angiogenesis, modulation of tyrosine kinase signal pathways and increase of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
-5
tumor invasion and metastasis . As such, tumor hypoxia has been considered as an
extremely valuable target, and there are many reports on detecting tumor hypoxic status
6
, 7
and optimizing tumor hypoxic therapy strategies . In particular, taking advantage of
the tumor hypoxic characteristics to aid the design of hypoxia-activated prodrugs
(HAPs) can specifically functionalize drug molecules within tumor hypoxic regions,
while reducing toxicity to normal tissues8-10. The mechanism of HAPs is as follows: the
prodrug is reduced to form a free radical intermediate in tissues, and can return to the
parent prodrugs when in normal tissues, under normoxic conditions. Whereas in
hypoxic cells, the intermediate form is cleaved to release the active drug molecules, and
this in turn, selectively acts on hypoxic cancer cells11-13. Until now, commonly used
HAPs mainly include nitrogen oxides (aliphatic and aromatic), anthraquinones, nitro
compounds and metal complexes1, 11, 14. Among them, nitro-activated prodrugs present
good tumor hypoxia specificity, which has greatly promoted the development of the
HAPs with nitroimidazoles, such as TH-30215, 16. However, TH-302 has failed in clincal
1
7
trials since it did not achieve primary overall survival endpoints in combination with
chemotherapy in two large phase III studies in advanced unresectable or metastatic
pancreatic adenocarcinoma (NCT01746979) and in soft tissue sarcoma
(
NCT01440088). Therefore, we are required to rethink the factors and design
considerations of HAPs.
Previous studies have demonstrated that the most important consideration for
HAPs design, in principle, is choosing the appropriate bioreductive groups18, 19. In other
ACS Paragon Plus Environment

Page 5 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
words, the determinant factor is the propensity of a particular group to undergo
bioreduction and the oxygen concentration at which this process occurs. In addition,
the factors of biological system also affect the activity of HAPs, such as reductase
expression and hypoxic tissue barriers. For example, solubility of HAPs greatly affects
the anti-cancer activity, which in essence is due to improved tissue penetration into
tumor hypoxic regions that are absent of blood vessel distribution20-23. While these
factors have been identified individually, it is still necessary to understand which
factor(s) play determined roles or whether these factors affect each other. Thus, a
comprehensive understanding of the relationship among various factors in HAPs design
is necessary, but this has seldomly been reported, owing to the absence of ideal systems.
To address the above requirements, we designed a series of new HAPs that were
composed of the same biologically active moiety but each was masked by different
bioreductive groups (red R), including nitrobenzofuran, nitrothiazole, nitrothiophene
and nitrofuran (Scheme 1). These bioreductive groups were utilized as electron-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
attracting initiators for increasing electron affinity of the HAPs . The reductases in
cells resulted in the occurrence of electron transfer, which in turn reduced the HAPs to
form intermediates. The intermediates, in hypoxic tumor tissues, were cleaved to
release the cytotoxic moiety for tumor growth inhibition, which was similiar to nitrogen
oxides HAPs activity. We explored the effect of various factors on anti-cancer activity
of the HAPs, including reduction potential of both whole molecules and bioreductive
groups, oxygen concentration and reductase expression. In vivo anti-tumor effects and
the interaction between the HAPs and hypoxic microenvironment were also
investigated.
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 6 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Schematic illustration of activation of the HAPs in hypoxia. The four kinds of HAPs
were designed by containing cytotoxic moiety and different bioreductive groups (R). The
bioreductive groups initiated electron transfer under the action of reducctases, to form
intermediates. The produced intermediates could return to the parent prodrugs in normoxic cells
but be cleaved to release the cytotoxic moiety in hypoxic cancer cells, which in turn selectively
induced their apoptosis. The four bioreductive groups included nitrobenzofuran, nitrothiazole,
nitrothiophene and nitrofuran (R, from left to right).
EXPERIMENTAL SECTION
General. All solvents were spectroscopic grade and were purified by distillation before
use. NMR spectra were recorded by Bruker AVANCE III 400 (400 MHz) spectrometer
with deuterium generation reagent and tetramethylsilane as an internal standard.
Electrochemical examinations were performed by cyclic voltammetry (CV) in
-1
acetonitrile at a scanning rate of 50 mv s . Platinum-electrodes (diameter 0.5 mm)
served as working electrode and counter electrode. The electrolyte was acetonitrile (5.0
mL) containing 0.5 mmol tetrabutylammonium hexafluorophosphate and 0.2 mmol
HAPs.
Prodrug Synthesis. We synthesized four new hypoxia prodrugs (Chinese application
patent: 201910650638.2) and the detailed routes were shown in Scheme S1. All
ACS Paragon Plus Environment

Page 7 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
compound structures were characterized by NMR and mass spectra. The detailed
synthetic procedures were as follows.
HAPs precursor 2 [2a-2d]
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(5-nitrobenzofuran-2-yl) methanol [2a]
O N
2
OH
O
2a
Sodium borohydride (1.2 g, 30 mmol) was added to a solution of tetrahydrofuran:
methanol (10:1) containing 1a (1.41 g, 10 mmol). The mixtures reacted for 1 h at ice-
bath. After that, the solution was stirred for 1 h and then warmed to the room
temperature for 1 h. The suspension was extracted with dichloromethane for organic
layer collection, and then washed with NaCl solution and water, respectively. The
resulting product was dried over MgSO , filtrated, and evaporated. The crude product
4
was purified by column chromatography on silica gel using petroleum ether: ethyl
1
acetate (4:1) as eluent to give 2a (1.23 g, 63.5%). H NMR (400 MHz, CDCl ) δ (ppm)
3
4
.85 (d, 2H, J = 8.0 Hz, -CH ), 6.83 (s, 1H, benzofuryl-H), 7.56 (d, 1H, J = 8.0 Hz,
2
benzofuryl-H), 8.24 (d, 1H, J = 8.0 Hz, benzofuryl-H), 8.51 (s, 1H, benzofuryl-H).
(5-nitrothiazole-2-yl) methanol [2b]
N
OH
O N
2
S
2b
2
b was prepared by an analogous method to that used for compound 2a, specifically
using 1b instead of 1a. The crude product was purified by column chromatography on
silica gel using petroleum ether: ethyl acetate (5:1) as eluent to obtain 2b (1.12 g, 70%).
1
H NMR (400 MHz, CDCl ) δ (ppm) 4.42 (d, 2H, J = 8.0 Hz, -CH ), 8.52 (s, 1H,
3
2
thiazolyl-H).
5-nitrothiophene -2-yl) methanol [2c]
(
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 8 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
OH
O N
2
S
2c
Sodium borohydride (0.6 g, 15 mmol) was added to a solution of methanol containing
-nitrothiophene-2-carbaldehyde 1c (1.57 g, 10 mmol). The mixtures reacted for 10
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
min at ice-bath. After that, the solution was stirred for 20 min and then warmed to the
room temperature. After 3 h, the suspension was extracted with dichloromethane, and
then the organic layer was collected for further washing with NaCl solution and water,
respectively. The organic layer was dried over MgSO , filtrated, and evaporated. The
4
crude product was purified by column chromatography on silica gel using petroleum
1
ether: ethyl acetate (5:1) as eluent to obtain 2c (1.39 g, 88.6%). H NMR (400 MHz,
CDCl ) δ (ppm) 4.88 (d, 2H, J = 8.0 Hz, -CH ), 6.94 (d, 1H, J = 8.0 Hz, thienyl-H),
3
2
7
.82 (d, 1H, J = 8.0 Hz, thienyl-H).
(
5-nitrofuran-2-yl) methanol [2d]
OH
O N
2
O
2d
2d was prepared by an analogous method to that used for compound 2c, specifically
using 1d instead of 1c. The crude product was purified by column chromatography on
silica gel using petroleum ether: ethyl acetate (4:1) as eluent to obtain 2d (0.91 g,
1
62.9%). H NMR (400 MHz, CDCl ) δ (ppm) 5.01 (d, 2H, J = 8.0 Hz, -CH ), 6.94 (d,
3
2
1
H, J = 8.0 Hz, furyl-H), 7.54 (d, 1H, J = 8.0 Hz, furyl-H).
HAPs precursor 3 [3a-3d]. A 1.0 M lithium bis(trimethylsilyl)amide in anhydrous
THF solution (2.2 mL, 2.2 mmol) was added to raw material 2a-2d (2.0 mmol) in THF
at 195 K under an argon atmosphere. Stirring was continued for 0.5 h and bis(2-
chloroethyl)phosphoramidic dichloride (0.56 g, 2.2 mmol) was slowly added to the
reaction mixture at 195 K and stirred for 2 h at this temperature. After that, propane
ACS Paragon Plus Environment

Page 9 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
diamine in THF was added to the mixture and continued stirring for 1 h at 195 K. Then
the reaction was warmed to room temperature and quenched with water. The organic
layer was dried over MgSO , filtrated, and evaporated. The crude product was purified
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
by column chromatography on silica gel using petroleum ether: ethyl acetate (1:1) as
eluent to acquire 3a-3d.
Cl
O N
2
Cl
N
O
O P
O
HN
3a
(5-nitrobenzofuran-2-yl)methyl-N,N,N-bis(2-chloroethyl)-1-propan-
1
phosphorodiamidate [3a] (0.39g，44.6%): H NMR (400 MHz, CDCl ) δ (ppm) 0.82-
3
0
4
.86 (m, 3H, -CH ), 1.37-1.43 (m, 2H, -CH ), 2.65-2.71 (m, 2H, -CH ), 3.20-3.33 (m,
3 2 2
H, -CH ), 3.62-3.68 (m, 4H, -CH ), 5.07 (d, 2H, J = 8.0 Hz, -CH ), 7.21 (s, 1H,
2
2
2
benzofuryl-H), 7.84 (d, 1H, J = 8.0 Hz, benzofuryl-H), 8.24 (d, 1H, J = 12.0 Hz,
benzofuryl-H), 8.66 (s, 1H, benzofuryl-H).
Cl
N
Cl
O N
2
N
S
O P
O
HN
3b
(5-nitrothiazole-2-yl)methyl-N,N,N-bis(2-chloroethyl)-1-propan-phosphorodiamidate
1
[
3b] (0.21g，26.1%): H NMR (400 MHz, CDCl ) δ (ppm) 0.92-0.95 (m, 3H, -CH ),
3
3
1
.38-1.41 (m, 2H, -CH ), 1.56-1.62 (m, 2H, -CH ), 3.51-3.64 (m, 4H, -CH ), 4.37-4.42
2 2 2
(
m, 4H, -CH ), 5.14 (d, 2H, J = 8.0 Hz, -CH ), 8.52 (s, 1H, thiazolyl-H).
2
2
Cl
Cl
O N
2
N
S
O P
O
HN
3c
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 10 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
5-nitrothiophene-2-yl)methyl-N,N,N-bis(2-chloroethyl)-1-propan-
1
phosphorodiamidate [3c] (0.27g，33.5%): H NMR (400 MHz, CDCl ) δ (ppm) 0.91-
3
0
4
4
.95 (m, 3H, -CH ), 1.51-1.55 (m, 2H, -CH ), 2.87-2.90 (m, 2H, -CH ), 3.42-3.47 (m,
3 2 2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H, -CH ), 3.63-3.66 (m, 4H, -CH ), 5.16 (d, 2H, J = 8.0 Hz, -CH ), 7.02 (d, 1H, J =
2
2
2
.0 Hz, thienyl-H), 7.82 (d, 1H, J = 4.0 Hz, thienyl-H).
Cl
Cl
O N
2
N
O
O P
O
HN
3d
(5-nitrofuran-2-yl)methyl-N,N,N-bis(2-chloroethyl)-1-propan-phosphorodiamidate
1
[3d] (0.33g，42.1%):
H NMR (400 MHz, CDCl ) δ (ppm) 0.91-0.94 (m, 3H, -
3
CH ), 1.50-1.56 (m, 2H, -CH ), 2.86-2.92 (m, 4H, -CH ), 3.39-3.45 (m, 2H, -CH ),
3
2
2
2
3
.62-3.67 (m, 4H, -CH ), 5.03 (d, 2H, J = 8.0 Hz, -CH ), 6.65 (d, 1H, J = 4.0 Hz, furyl-
2 2
H), 7.27 (d, 1H, J = 4.0 Hz, furyl-H).
Hypoxia-activated prodrugs [H1-H4]. 3a-3d (1 mmol) were dissolved in pyridine (10
ml) under stirring. The reaction mixture was reacted at 100°C for 24 h under nitrogen
atmosphere. The crude product was purified by oil washed many times through
dichloromethane and ethyl acetate. The collected precipitation was dissolved in
ultrapure water for freeze-drying by lyophilizer to obtain H1-H4.
O N
2
N Cl
N
Cl
N
O
O P
O
HN
H1
1
,1'-(((((5-nitrobenzofuran-2-yl)methoxy)(propylamino)phosphoryl)azanediyl)
1
bis(ethane-2,1-diyl))bis(pyridin-1-ium) chloride [H1] (0.25g，42.1%): H NMR (400
ACS Paragon Plus Environment

Page 11 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
MHz, D O) δ (ppm) 1.00-1.06 (m, 3H, -CH ), 1.68-1.73 (m, 2H, -CH ), 2.89-2.93 (m,
2
3
2
2
H, -CH ), 3.21-3.24 (m, 4H, -CH ), 4.69-4.72 (m, 4H, -CH ), 5.04 (d, 2H, J = 8.0 Hz,
2 2 2
-CH ), 7.47 (s, 1H, benzofuryl-H), 7.72 (d, 1H, J = 8.0 Hz, benzofuryl-H), 8.02 (d, 1H,
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
J = 8.0 Hz, benzofuryl-H), 8.08-8.24 (m, 4H, pyridyl-H), 8.59-8.71 (m, 2H, pyridyl-H),
8
.68 (s, 1H, benzofuryl-H), 9.10-9.12 (m, 2H, pyridyl-H), 9.21-9.22 (m, 2H, pyridyl-
H). MS (ESI):m/z-2Cl Calcd for C H Cl N O P 595.1518; found 524.2304.
26 32
2
5
5
Cl
N
Cl
N
N
O N
2
N
S
O
P
O
HN
H2
1,1'-(((((2-nitrothiazol-5-yl)methoxy)(propylamino)phosphoryl)azanediyl)bis(ethane-
1
2
,1-diyl))bis(pyridin-1-ium) chloride [H2] (0.20g，35.6%): H NMR (400 MHz, D O)
2
δ (ppm) 0.88-0.90 (m, 3H, -CH ), 1.24-1.28 (m, 2H, -CH ), 1.38-1.40 (m, 2H, -CH ),
3
2
2
4
8
8
.10-4.15 (m, 4H, -CH ), 4.40-4.45 (m, 4H, -CH ), 4.97 (d, 2H, J = 8.0 Hz, -CH ), 8.03-
2 2 2
.19 (m, 4H, pyridyl-H), 8.51 (s, 1H, thiazolyl-H), 8.55-8.65 (m, 2H, pyridyl-H), 8.85-
.86 (m, 2H, pyridyl-H), 9.12-9.13 (m, 2H, pyridyl-H). MS (ESI):m/z-2Cl Calcd for
C H Cl N O PS 562.1086; found 491.2032.
2
1
29
2
6
4
Cl
N
Cl
N
O N
2
N
S
O P
O
HN
H3
1
,1'-(((((5-nitrothiophen-2-
yl)methoxy)(propylamino)phosphoryl)azanediyl)bis(ethane-2,1-diyl))bis(pyridin-1-
1
ium) chloride [H3] (0.25g，43.8%): H NMR (400 MHz, D O) δ (ppm) 1.02-1.06 (m,
2
3
H, -CH ), 1.70-1.72 (m, 2H, -CH ), 2.89-2.93 (m, 2H, -CH ), 3.15-3.26 (m, 4H, -CH ),
3 2 2 2
4.72-4.75 (m, 4H, -CH ), 5.17 (d, 2H, J = 8.0 Hz, -CH ), 7.91 (d, 1H, J = 8.0 Hz, thienyl-
2
2
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 12 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
H), 8.03-8.10 (m, 4H, pyridyl-H), 8.24 (m, 1H, thienyl-H), 8.56-8.60 (m, 2H, pyridyl-
H), 8.85-8.86 (m, 2H, pyridyl-H), 8.92-8.93 (m, 2H, pyridyl-H). MS (ESI):m/z-2Cl
Calcd for C H Cl N O PS 561.1133; found 490.1246.
2
2
30
2
5
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cl
N
Cl
N
O N
2
N
O
O
P
O
HN
H4
1,1'-(((((5-nitrofuran-2-yl)methoxy)(propylamino)phosphoryl)azanediyl)bis(ethane-
1
2
,1-diyl))bis(pyridin-1-ium) chloride [H4] (0.20g，37.2%): H NMR (400 MHz, D O)
2
δ (ppm) 1.02-1.06 (m, 3H, -CH ), 1.68-1.73 (m, 2H, -CH ), 2.89-2.93 (m, 4H, -CH ),
3
2
2
3
.32-3.34 (m, 2H, -CH ), 3.67-3.68 (m, 4H, -CH ), 5.06 (d, 2H, J = 8.0 Hz, -CH ), 6.82
2 2 2
(d, 1H, J = 4.0 Hz, furyl-H), 7.45 (d, 1H, J = 4.0 Hz, furyl-H), 8.08-8.17 (m, 4H, pyridyl-
H), 8.60-8.66 (m, 2H, pyridyl-H), 8.90-8.92 (m, 2H, pyridyl-H), 9.00-9.03 (m, 2H,
pyridyl-H). MS (ESI):m/z-2Cl Calcd for C H Cl N O P 545.1362; found 474.1878.
2
2
30
2
5
5
Human and Mouse Cell lines. 4T1 mouse breast cancer cells were cultured in RPMI
640 supplemented with 10% fetal bovine serum (FBS, BI) and 1%
1
penicillin/streptomycin (PS, HyClone). HepG2 human liver carcinoma cell, U87 human
brain glioblastoma cells, HT1080 human fibrosarcoma cells, PC3 human prostate
cancer cells, 3LL mouse lung cancer cells and B16 mouse melanoma cells were
cultured in DMEM high glucose (HyClone) supplemented with 10% FBS and 1% PS.
For normoxic and hypoxic cell culture, cells were incubated in a humidified incubator
o
(
37 C, 5% CO ) with normoxic (21% O ) or hypoxic (10%, 5%, 2% or 0.5% O )
2
2
2
conditions, respectively.
Cell colony formation assay. HepG2 and 3LL cells were digested by 0.25%
trypsin/0.02% EDTA solution to make a single cell suspension. Then, the cells were
seeded into six-well culture plates at the indicated densities (500, 1000, and 2000 cells
ACS Paragon Plus Environment

Page 13 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
per well), treated with H3 or H4 under various oxygen concentrations for 7 days, fixed
with 4% PFA and stained with 1.25% crystal violet solution for quantification25, 26
.
Cytotoxicity in monolayer cell. Cytotoxicity was measured using standard MTT assay.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
Briefly, 4 × 10 cells were seeded into 96-well plates, and then were treated with the
prodrugs H1-H4 (0, 20, 50, 100 and 200 μM) at normoxic or hypoxic conditions,
respectively. After 24 h incubation, 10 μL MTT was added to each well and DMSO
was added to the media 4 h later. Finally, OD570 was measured by Thermo Scientific
Microplate Reader. The cell viability was calculated as OD570 mean value in
experimental group/OD570 mean value in control group × 100%.
Apoptosis assay. HepG2 cells were treated with 200 μM H1-H4 under hypoxic
condition (0.5% O ) for 24 h, and then stained with Annexin V-FITC and propidium
2
iodide (PI) according to manufacture protocol (Beyotime Biotechnology, China).
5
Briefly, 1 × 10 cells were washed with cold PBS and suspended with 100 µl binding
buffer. 5 µl of Annexin V-FITC was added to the cells for 10 min. After incubation, 5
µl PI was added for 5 min. Finally, the cells were analyzed by flow cytometry (BD
2
7
FACSCalibur, USA) .
Cell cycle analysis. HepG2 cells were treated with 200 μM H1-H4 under hypoxic
condition (0.5% O ) for 24 h, respectively and then stained with PI according to
2
manufacture protocol (Beyotime Biotechnology, China). Briefly, the cells were
o
harvested and washed with cold PBS, fixed in ice-cold 75% ethanol overnight at 4 C.
The fixed cells were collected by centrifugation, resuspended in 500 µl stain buffer and
o
incubated with 25 µl PI and 10 µl RNase A at 37 C for 30 min in dark. Finally, the
cells were analyzed by flow cytometry (BD FACSCalibur, USA).
HAPs reduction assay in vitro. 100 mM HAPs were placed into PB buffer (pH 7.4),
followed by 100 mM NADPH, and incubated with P450 oxidoreductase (POR) (Life
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 14 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
o
Technologies) in a humidified incubator (37 C, 5% CO ) with hypoxic (0.5% O )
2
2
conditions. The FL spectra of HAPs was monitored by Bio-Rad Enzyme Marker. After
24 h, the crude mixtures were purified by column chromatography on silica gel plate.
Real-Time PCR assays. Total RNA was isolated from cells using TRIzol (Solarbio,
China) according to the manufacturer protocol. Reverse transcription reaction to
synthesize complementary DNAs (cDNAs) was performed using 2 μg of total RNA and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
μl 5X ALL-In-One RT Master Mix (Abm, China) in an RNase-free environment.
Real-Time PCR was performed with the QuantStudio™ 3 Real-Time PCR System
Appled Biosystems). The results were normalized by measuring average cycle
(
threshold (Ct) ratios between POR and the housekeeping gene EIF3F (TIF). PCR
sequences are provided below.
Human POR
F 5’- TCAATGCCATGGGCAAGTA-3’
R 5’- ATGAAGTCCTCCTCCAAGTTC-3’
Human TIF
F 5’- GACACAAGTCTCCAGAACGGC-3′
R 5’- TGGTCTCAAAGTCATCGGGAA-3′
Mouse POR
F 5’- TCCTGACCTACTGGTTCATCT-3′
R 5’- CACGAAGCTGCTCTCTTTGA-3′
Mouse TIF
F 5’- CTGAGGATGTGCTGTCTGGGAA-3′
R 5’- CCTTTGCCTCCACTTCGGTC-3′
Xenograft model. Male BALB/c-Nude mice (5–6 weeks old) and female C57 mice (6-
8
weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd. All animals
ACS Paragon Plus Environment

Page 15 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
and in vivo experiments were in compliance with governmental and international
animal guidelines. 2ꢀ×ꢀ106 HepG2, 2ꢀ×ꢀ106 U87 or 5ꢀ×ꢀ105 3LL cells were
subcutaneously injected into right shoulder of the mice, respectively, and then assigned
randomly to different groups (5-6 mice per group). Treatments with H3 or H4 (40
mg/kg) were initiated when tumors reached a diameter ~5 to 8 mm. The mice were
treated intravenously every day (3LL xenografted tumor) or intraperitoneal injection
every other day (HepG2 and U87 xenografted tumor) for 5 doses total. The tumor
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
volume was measured every day, and was calculated by the formula: (width) ×
(length)/2. Body weight of the mice was also monitored.
Hypoxia and blood vessel staining. After HAPs treatment, the pimonidazole HCl (60
mg/kg) was given to mice 1.5 h prior to euthanization. The tumor tissues were
immunostained with Hypoxyprobe following the manufacturer’s procedures
(Hypoxyprobe, Inc). Briefly, the hypoxic regions of the tumor were stained using FITC-
conjugated anti-pimonidazole antibody. To co-stain with blood vessels, the tissues were
simultaneously incubated with PE-conjugated anti-CD31 antibody (Biolegend). The
images were acquired by confocal laser-scanning microscope (Zeiss LSM 710) and
analyzed via Image J.
TUNEL staining. The tumor frozen sections were prepared after HAPs treatment and
then TUNEL staining was performed according to manufacturer instruction of one-step
fluorescent TUNEL Apoptosis Assay Kit (Beyotime Biotechnology, China). Briefly,
the frozen sections were fixed with 4% PFA for 30 min, permeabilized by 0.2% Triton
X-100, and stained with TUNEL staining solution for 1 h. The images were captured
using a confocal laser-scanning microscope (Zeiss LSM 710) and quantified by ImageJ.
The proportion of TUNEL signal in the tumor hypoxic or normoxic tissues was
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 16 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
evaluated by following the formula A /A (or A ): A represents the area of TUNEL
T
H
N
T
regions; A is the area of hypoxic tissues; A is the area of normoxic tissues.
H
N
Statistical analysis. The Student's unpaired t-test was used for statistical analysis
between two groups of independent data. If multiple comparisons were involved, one-
way or two-way ANOVA was used as needed. Differences were considered statistically
significant at P < 0.05. Results were shown as the mean ± SEM.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS AND DISCUSSION
Design and synthesis of prodrugs
The designed prodrugs mainly contained two components, including the biologically
active moiety and bioreductive groups. As a model effector molecule, phosphoramide
mustard was chosen based on the below considerations: 1) it is a biologically active
metabolite of cyclophosphamide, a chemotherapeutic agent commonly used to treat
lymphomas, breast cancers, certain brain cancers, and autoimmune diseases28-30; 2) it
easily reacts with the four bioreductive groups. Given its critical role in hypoxic
tumor penetration, water-solubility of HAPs was further improved by introduction of
pyridinium. The designed bioreductive groups possessed different reduction potential
ranking: nitrobenzofuran (2a) < nitrothiazole (2b) < nitrothiophene (2c) < nitrofuran
(2d) (Fig. 1). Following conjugation with phosphoramide mustard cytotoxic
components and pyridinium, the reduction potential of the obtained HAPs (H)
corresponding to the bioreductive groups was -1.45 V (H1), -1.39 V (H2), -1.32 V
(H3) and -1.56 V (H4) via cyclic voltammetry characterization, respectively. Due to
the prodrugs embedding an identical cytotoxic component but having varied
bioreductive groups, this provided a well-designed and new model system, as a
platform to understand not only the relevance among molecular structures, the
ACS Paragon Plus Environment

Page 17 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
reduction potential of whole HAPs molecules and oxygen concentration but also the
interaction between the HAPs and biological systems.
Four kinds of novel hypoxia-responsive prodrugs were synthesized by multiple
step reactions of different nitro group-based precursor molecules with phosphoramide
mustard. In a typical reaction, nitro-based aromatic aldehydes (i.e. nitrobenzofuran,
nitrothiazole, nitrothiophene and nitrofuran) were reduced by sodium borohydride in
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
1
methanol or tetrahydrofuran . The resulting molecules were subsequently conjugated
with phosphoramide mustard to obtain major components of the HAPs. To improve the
water-solubility of HAPs 32, 33, pyridine cationization reaction was further performed.
The detailed reaction routes were shown in Scheme S1. The successful synthesis of four
prodrugs was confirmed by NMR characterization. These HAPs displayed high
solubility, with >100 mg/ml in water solution.
Figure 1. Molecular structures and reduction potential of different bioreductive groups
(upper) and the corresponding HAPs (bottom).
Cytotoxicity of different HAPs under various O concentrations
2
We further explored whether the four prodrugs, H1-H4, could be activated under
hypoxia. As such, different cancer cells such as HepG2 and 3LL cells were cultured at
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 18 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
.5% O for evaluation of the activity of the prodrugs, and this was followed by
2
assessment by standard cell activity assays. All four of the prodrugs showed a
concentration-dependent inhibition effect on HepG2 and 3LL cells, especially H4 (Fig.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
A). The inhibition ability of H4 on HepG2 was 2.1-, 4.4- and 6.9-fold higher at the
concentration of 200 μM, compared to that of H1, H2 and H3, respectively, which was
comparable with 3LL cells. Importantly, we found an excellent linear relationship
between the reduction potential of whole HAPs molecules and cell viability for both
cancer cells (Fig. 2B), and this was irrelevant to the reduction potential of bioreductive
groups (Fig. S1). Since the reduction potential of whole prodrug molecules was ranked
as: H4 < H1 < H2 < H3, the lower reduction potential of whole prodrug molecules
presented a higher cytotoxicity, which was not cell type dependent. Next, H4 was
selected to evaluate whether the cytotoxicity of the prodrugs was oxygen-dependent.
Both HepG2 and 3LL cells were cultured at different O concentration (i.e. 21%, 10%,
2
5
%, 2% and 0.5%), and the results showed that the cytotoxicity of H4 was increased
along with the reduction of O concentration (Fig. 2C). A linear relationship between
2
O concentration and cell viability was also established, which revealed that the
2
activation of the prodrugs was strongly dependent on hypoxic condition (Fig. 2D). The
results were further confirmed by various other cancer cell types, such as U87, HT1080,
PC3, 4T1 and B16 (Fig. S2). Based on the above results, the reduction potential of
whole HAPs molecules and O concentration determined the activity of the prodrugs.
2
Thus, we further explored the effect of these two aspects on the colony formation
capacity of HepG2 cells in response to H4 treatment. The colony formation rate after
H4 treatment was about 2-fold reduction at hypoxic condition (0.5% O ) compared to
2
the untreated groups, while no significant changes of colony formation were observed
for H3 treatment under either hypoxic or normoxic conditions (Fig. 2E and F). More
ACS Paragon Plus Environment

Page 19 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
strikingly, the colony formation capacity of H4 treated 3LL cells was decreased by
approximately 8-fold under hypoxia, compared to untreated 3LL cells (Fig. S3).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Effect of different HAPs on cancer cells under various O concentrations. (A)
2
Inhibition rate of HepG2 and 3LL cells following treatment with H1-H4 under hypoxia (0.5%)
for 24 h, respectively. (B) The linear relationship between cell viabiltiy at 200 μM HAPs and
the reduction potential of whole HAPs molecules. (C) Cytotoxicity of H4 for HepG2 and 3LL
cells at different O concentrations (i.e. 21%, 10%, 5%, 2% and 0.5%) after 24 h treatment. (D)
2
The linear relationship between cell viability at 200 μM HAPs and O concentration. (E)
2
Representative images and (F) quantification analysis of the colony formation capacity of
HepG2 cells treated with H4 under normoxic (21%) and hypoxic (0.5%) conditions for 7 days.
HAPs reduction assay in vitro
As shown above in Fig. S2, the cytotoxicity of H4 was obviously varied in different
cancer cell types, demonstrating that the activation of H4 under hypoxia was cell type-
dependent. It has been identified that the expression level of POR in cancer cells is
responsible for the activation of hypoxia-activated prodrugs (HAPs) 34, 35. We thus
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 20 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
explored whether the cleavage of H4 was POR dependent by incubation H4 with
purified POR, together with NADPH as co-substrate for 2 h under hypoxia. The
fluorescent signal of H4 showed a 21.5% reduction by comparison before and after
POR treatment (Fig. S4). Furthermore, the resulting compound after reaction was
purified by column chromatography on silica gel and characterized with NMR, showing
a clear peak of nitrofuran following the cleavage of ester linkage of H4 (Fig. 3A). To
demonstrate the effect of POR expression on H4 cytotoxicity, we next checked the
cytotoxicity of H4 as well as POR expression in various human and mouse cancer cells,
including HepG2, U87, PC3, HT1080, 3LL, 4T1 and B16. We found both cytotoxicity
(Fig. 3B) and POR expression (Fig. 3C) were cell type-dependent. Interestingly, the
cytotoxicity of H4 on different cancer cells was association with POR expression, a
correlation that was statistically significant in a pooled analysis of seven different cell
lines (P < 0.05) (Fig. 3D).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Anti-cancer mechanism
We next sought to understand the detailed molecular mechanism of the HAPs in
inhibiting cancer cell growth. HepG2 cells treated with H1-H4 under hypoxia were
stained with Annexin V and PI for apoptosis analysis, and the results showed a notable
increase of the percentage of apoptotic cells, with ~2.1-fold elevation after H4 treatment
compared to untreated cells (Fig. 4A and B). In addition, cell cycle progression of
HepG2 treated with H1-H4 was further evaluated by flow cytometry analysis. As shown
in Fig. 4C and D, the changes of cell cycle progression occurred after treatment with
the prodrugs, especially for H4. After H4 treatment, compared to untreated cells, cells
of G0/G1 phase increased from 18.6% to 27.4%, whereas cells in G2/M decreased ~2-
fold. Taken together, these data suggest that the prodrug H4 inhibited cancer cell growth
ACS Paragon Plus Environment

Page 21 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Figure 3. Effect of reductase on H4 activation under hypoxic condition. (A) H NMR
characterization of H4 and the released bioreductive groups after treated with POR under 0.5%
O . (B) Inhibition effect of H4 on various human (left) and mouse (right) cancer cells under
2
0
.5% O . (C) RT-PCR analysis of POR expression level in various human (left) and mouse
2
(
right) cancer cells under 0.5% O . (D) The linear relationship between IC of H4 and POR
2
70
expression level.
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 22 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
in hypoxic conditions, and this was mainly due to increased cell apoptosis and cell cycle
arrest.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Effect of H1-H4 on HepG2 cell apoptosis and cell cycle. HepG2 cells were treated
with H1-H4 under hypoxic condition for 24 h, respectively. (A, B) FACS analysis of untreated
control (CTL) and H1-H4 treated cells stained with Annexin V and PI. The relative fraction of
live and apoptotic cells was quantified. (C, D) FACS analysis of CTL and H1-H4 treated cells
stained with PI. The relative fraction of cells in the different cell-cycle stages (G0/G1, S, G2/M)
was quantified.
In vivo anti-cancer effect and activation mechanism of the HAPs
Next, we studied the effect of the prodrugs on in vivo tumor growth in various mice
models xenografted with HepG2, U87 and 3LL cells. Prior to exploring anti-tumor
efficacy of the prodrugs, the occurrence of hypoxia was checked in these three tumor
models. The hypoxic areas, blood vessels and cell nuclei of the tumor tissues were
stained with anti-pimonidazole antibody (green), anti-CD31 antibody (red) and DAPI
(blue), respectively. We found abundant hypoxic regions within tumor, but limited
blood vessel distribution was observed in hypoxic regions. The vessel density in
ACS Paragon Plus Environment

Page 23 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
hypoxic area was less than one-third of that in normoxic tumor regions of the same
tumor among all three tumor models (Fig. 5A-C), which was consistent with previous
reports1, 36-39. In other words, direct delivery of the prodrugs into hypoxic regions was
really difficult due to the inaccessibility of hypoxic area by systemic administration of
therapeutic agents. We then evaluated whether systemically administered prodrugs
delayed tumor growth. The mice bearing HepG2, U87 or 3LL were randomly divided
into three groups, respectively, and were injected with PBS, H3 (prodrug control) and
H4. As shown in Fig. 5D-F, tumor growth was significantly inhibited after H4 treatment
for all three tumor models. In contrast, the tumor volume showed no statistical changes
in H3 treated groups. Additionally, in all groups of the three tumor models, no
significant changes in the body weight of mice were observed, as shown in Fig. S5;
suggesting that both H3 and H4 had no apparent systemic toxicity.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The potential in vivo anti-cancer mechanism of the prodrugs was further
explored. The tumors from HepG2 groups were dissociated to single cell suspension by
trypsinization, and then the cells were stained with Annexin V-FITC and PI. As shown
in Fig. 6A, a 2.5- and 1.4-fold increase of apoptotic percentage in H4 treated tumor was
observed by comparing to that of untreated and H3 treated tumors, respectively.
Subsequently, the distribution of apoptotic cells in tumor tissue was analyzed via
TUNEL staining for cell apoptosis along with anti-pimonidazole immunostaining for
hypoxia. The representative images, shown in Fig. 6B, displayed the substantial
increase of cell apoptosis in H4-treated tumor tissues compared to untreated and H3
treated tissues. Quantification analysis of the percentage of positive apoptotic cells in
untreated, H3 and H4 tissues accounted for 18.4±0.8, 19.3±1.4 and 28.4±1.3 of total
cancer cells, respectively (Fig. 6C). Furthermore, H4-treated tumor tissues presented
that cell apoptosis in hypoxic areas was ~15.6-fold higher than nomoxic areas (Fig. 6D)
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 24 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. In vivo anti-cancer effect of H4 treatment. Representative confocal images of
distribution of hypoxia (green) and blood vessels (red) in various xenograft tumor tissues,
including (A) HepG2, (B) 3LL and (C) U87. Blue color represents cell nuclei. Scale bar = 100
μm. Tumor growth curve of (D) HepG2, (E) 3LL and (F) U87 untreated or treated with H3 and
H4, respectively. n = 6/group. **P<0.01.
and the signal intensity of positive apoptotic cells in tumor tissues was highly
concomitant with that of hypoxic status (Fig. 6E), revealing that the occurrence of cell
apoptosis mainly localized in hypoxic areas. As shown in Fig. 6A, hypoxic regions of
tumors lacked blood vessels and, thus, were not readily accessible by systemically
administered therapeutic agents. Thus, the results also implied that our designed H4
prodrugs directly induced cell apoptosis following their penetration into hypoxic tissues,
most likely due to their high water-solubility20, 22, 40. Given the main action of HAPs on
hypoxic cancer cells, and thus, developing hypoxic cell-targeting strategy for HAPs
ACS Paragon Plus Environment

Page 25 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
accumulation into hypoxic cancer cells should be an unneglectable factor for anti-
cancer HAPs design.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. In vivo anti-cancer and activation mechanism of HAPs. (A) Flow cytometry and
quantification analysis of HepG2 tumor apoptosis by Annexin V and PI staining. (B)
Representative images of hypoxia (green) and TUNEL (red) staining of HepG2 tumor tissues
following H3 and H4 treatment, respectively. Scale bar = 200 μm. (C) Quantification analysis
of apoptosis signal (TUNEL staining) within HepG2 tumor untreated or treated with H3 and
H4, respectively. (D) Quantification analysis of apoptosis signal intensity in hypoxic and
normoxic tumor tissues following H4 treatment. (E) A co-localization analysis of the
distribution of hypoxia and apoptosis in H4 treated tumor tissue based on the plot profile of
white cross line from (B).
In conclusion, our designed a series of novel HAPs provided a basis for comprehensive
understanding the interaction between hypoxia-targeted prodrugs and hypoxic tumor
tissue. By evaluation of the interaction between four HAPs and various cancer
cells/tissues, we demonstrated that both molecular structure and hypoxic biological
system play critical role in anti-cancer activity of the prodrugs designed. For HAPs
molecules design, our results revealed that the larger redox propensity of full HAPs
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 26 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
molecules had the higher anti-cancer property via the release of effector molecules
under hypoxia, while the anti-cancer activity was not dependent on the reduction
potential of bioreductive activated groups. However, addition of the bioreductive
moiety as electron-attracting groups was the prerequisite for increasing electron affinity
of HAPs. Thus, both redox properties of full molecules and activated bioreductive
groups are necessary design considerations. In biological systems, the linear
relationship of reductase enzyme expression or oxygen status with cytotoxicity of the
HAPs revealed that tumor heterogeneity was also essential for anti-cancer efficacy of
the prodrugs. Remarkably, hypoxic cell-targeting of drug molecules is still a
challenging factor, due to limited blood vessels in tumor hypoxic regions. Our designed
prodrugs, with high water solubility, penetrated into hypoxic tumor regions to directly
induce cell apoptosis. However, physiological solubility merely improves the diffusion
of drug in tumor tissues, but in essence does not increase hypoxic cell uptake and drug
retention in cells. Design of hypoxic-targeted prodrugs for tumor hypoxic cell
accumulation would be promising for effective drug delivery in future. In summary,
this study provides substantial information for guiding the design of hypoxia-targeting
compounds and has broad applications for the development of hypoxia-related drug
delivery strategies.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
SUPPORTING INFORMATION
The Supporting Information is available free of charge at https://pubs.acs.org/doi/.
Figures of the synthetic routes of HAPs, the relationship between cell viability and the
reduction potential of bioreductive groups, and the effect of H4 on cell proliferation,
colony formation and body weight of mice xenografted tumor models (PDF).
ACS Paragon Plus Environment

Page 27 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
ACKNOWLEDGMENTS
This work was supported by the National Natural Science Foundation of China
(91959129, 31870999), the Natural Science Foundation of Tianjin
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(18JCYBJC40900), National Thousand Young Talents Program, and Nankai
University Hundred Young Academic Leaders Program.
REFERENCES
1
. Wilson, W. R.; Hay, M. P. Targeting hypoxia in cancer therapy. Nat Rev
Cancer 2011, 11, (6), 393-410.
. Liu, J. N.; Bu, W.; Shi, J. Chemical Design and Synthesis of Functionalized
Probes for Imaging and Treating Tumor Hypoxia. Chem Rev 2017, 117, (9), 6160-
224.
2
6
3. Harris, A. L. Hypoxia--a key regulatory factor in tumour growth. Nat Rev
Cancer 2002, 2, (1), 38-47.
4
. Chang, H. Y.; Liu, H. S.; Lai, M. D.; Tsai, Y. S.; Tzai, T. S.; Cheng, H. L.;
Chow, N. H. Hypoxia promotes nuclear translocation and transcriptional function in
the oncogenic tyrosine kinase RON. Cancer Res 2014, 74, (16), 4549-62.
5
. Sahlgren, C.; Gustafsson, M. V.; Jin, S.; Poellinger, L.; Lendahl, U. Notch
signaling mediates hypoxia-induced tumor cell migration and invasion. P Nati Acad
Sci 2008, 105, (17), 6392-6397.
6
. Matsumoto, S.; Kishimoto, S.; Saito, K.; Takakusagi, Y.; Munasinghe, J. P.;
Devasahayam, N.; Hart, C. P.; Gillies, R. J.; Mitchell, J. B.; Krishna, M. C.
Metabolic and Physiologic Imaging Biomarkers of the Tumor Microenvironment
Predict Treatment Outcome with Radiation or a Hypoxia-Activated Prodrug in Mice.
Cancer Res 2018, 78, (14), 3783-3792.
7. Cui, D.; Huang, J.; Zhen, X.; Li, J.; Jiang, Y.; Pu, K. A Semiconducting
Polymer Nano-prodrug for Hypoxia-Activated Photodynamic Cancer Therapy. Angew
Chem Int Ed 2019, 58, (18), 5920-5924.
8
. Yang, S.; Tang, Z.; Hu, C.; Zhang, D.; Shen, N.; Yu, H.; Chen, X. Selectively
Potentiating Hypoxia Levels by Combretastatin A4 Nanomedicine: Toward Highly
Enhanced Hypoxia-Activated Prodrug Tirapazamine Therapy for Metastatic Tumors.
Adv Mater 2019, 31, 1805955(1-7).
9. Liang, D.; Millera, G. H.; Tranmer, G. K. Hypoxia Activated Prodrugs: Factors
Influencing Design and Development. Curr Med Chem 2015, 22, 4313-4325.
1
0. Zhou, F.; Fu, T.; Huang, Q.; Kuai, H.; Mo, L.; Liu, H.; Wang, Q.; Peng, Y.; Han,
D.; Zhao, Z.; Fang, X.; Tan, W. Hypoxia-Activated PEGylated Conditional
Aptamer/Antibody for Cancer Imaging with Improved Specificity. J Am Chem Soc
2
019, 141, (46), 18421−18427.
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 28 of 30
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1. Denny, W. A. The role of hypoxia-activated prodrugs in cancer therapy. Lancet
Oncol 2000, 1, (1), 25-29.
12. Hunter, F. W.; Wouters, B. G.; Wilson, W. R. Hypoxia-activated prodrugs:
paths forward in the era of personalised medicine. Br J Cancer 2016, 114, (10), 1071-
7
.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
13. Denny, W. A. Hypoxia-activated prodrugs in cancer therapy: progress to the
clinic. Future Oncol. 2010, 6, (3), 419–428.
1
(
1
4. Denny, W. A. Prodrug strategies in cancer therapy. Eu J Med Chem 2001, 36,
7-8), 577–595.
5. Dragovich, P. S.; Broccatelli, F.; Chen, J.; Fan, P.; Le, H.; Mao, W.; Pillow, T.
H.; Polson, A. G.; Wai, J.; Xu, Z.; Yao, H.; Zhang, D. Design, synthesis, and
biological evaluation of pyrrolobenzodiazepine-containing hypoxia-activated
prodrugs. Bioorg Med Chem Lett 2017, 27, (23), 5300-5304.
1
6. Meng, F.; Evans, J. W.; Bhupathi, D.; Banica, M.; Lan, L.; Lorente, G.; Duan, J.
X.; Cai, X.; Mowday, A. M.; Guise, C. P.; Maroz, A.; Anderson, R. F.; Patterson, A.
V.; Stachelek, G. C.; Glazer, P. M.; Matteucci, M. D.; Hart, C. P. Molecular and
cellular pharmacology of the hypoxia-activated prodrug TH-302. Mol Cancer Ther
2
1
Hypoxia-Activated Prodrugs: Making Sense of Future Development. Curr Drug
Targets 2019, 20, (6), 668 - 678.
1
Hammond, E. M.; Conway, S. J. Design, synthesis and evaluation of molecularly
targeted hypoxia-activated prodrugs. Nat Protoc 2016, 11, (4), 781-94.
012, 11, (3), 740-51.
7. Su, M.-X.; Zhang, L.-L.; Huang, Z.-J.; Shi, J.-J.; Lu, J.-J. Investigational
8. O'Connor, L. J.; Cazares-Korner, C.; Saha, J.; Evans, C. N.; Stratford, M. R.;
1
9. Wardman, P. Electron Transfer and Oxidative Stress as Key Factors in the
Design of Drugs Selectively Active in Hypoxia. Curr Med Chem 2001, 8, (7), 739-
7
2
Hogg, A.; Sarath Liyanage, H. D.; Dorie, M. J.; Martin Brown, J.; Denny, W. A.;
Hay, M. P.; Wilson, W. R. Pharmacokinetic/Pharmacodynamic Modeling Identifies
SN30000 and SN29751 as Tirapazamine Analogues with Improved Tissue
Penetration and Hypoxic Cell Killing in Tumors. Clin Cancer Res 2010, 16, 4946-
4957.
61.
0. Hicks, K. O.; Siim, B. G.; Jaiswal, J. K.; Pruijn, F. B.; Fraser, A. M.; Patel, R.;
2
1. Dewhirst, M. W.; Secomb, T. W. Transport of drugs from blood vessels to
tumour tissue. Nat Rev Cancer 2017, 17, (12), 738-750.
22. Minchinton, A. I.; Tannock, I. F. Drug penetration in solid tumours. Nat Rev
Cancer 2006, 6, (8), 583–592.
2
3. Perche, F.; Biswas, S.; Wang, T.; Zhu, L.; Torchilin, V. P. Hypoxia-Targeted
siRNA Delivery. Angew Chem Int Ed 2014, 53, (13), 3362–3366.
4. Wardman, P. Some Reactions and Properties of Nitro Radical-Anions
Important in Biology and Medicine. Environ Health Perepect 1985, 64, 309-320.
5. Fu, Y.; Lai, Y.; Liu, J.; Liu, X.; You, Z.; Yang, G. Lentivirus-mediated shRNA
2
2
interference of clusterin blocks proliferation, motility, invasion and cell cycle in the
ovarian cancer cells. J Ovarian Res 2015, 8, 59.
ACS Paragon Plus Environment

Page 29 of 30
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
2
6. Lv, M.; Li, S.; Luo, C.; Zhang, X.; Wang, F.; Wang, X.; Yang, J.; Liu, P.; Shen,
Y.; Yang, J. Angiomotin promotes renal epithelial and carcinoma cell proliferation
by retaining the nuclear YAP. Oncotarget 2016, 7, (11), 12393-12403.
2
Farnham, P. J.; Shi, J.; Lu, W. The prostate cancer risk variant rs55958994 regulates
multiple gene expression through extreme long-range chromatin interaction to control
tumor progression. Sci Adv 2019, 5, eaaw6710 (1-12).
7. Qian, Y.; Zhang, L.; Cai, M.; Li, H.; Xu, H.; Yang, H.; Zhao, Z.; Rhie, S. K.;
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
8. Groehler, A. t.; Villalta, P. W.; Campbell, C.; Tretyakova, N. Covalent DNA-
Protein Cross-Linking by Phosphoramide Mustard and Nornitrogen Mustard in
Human Cells. Chem Res Toxicol 2016, 29, (2), 190-202.
2
Cyclophosphamide and Its Active Metabolites, 4-Hydroxycyclophosphamide,
Phosphoramide Mustard, and Acrolein. Cancer Res 1982, 42, (8), 3016-3021.
3
Alkylating Properties of Phosphoramide Mustard. Cancer Res 1976, 36, (3), 1121-
1126.
3
phenoxazin-3-one as a spectroscopic off-on probe for highly sensitive and selective
detection of nitroreductase. Chem Commun (Camb) 2013, 49, (52), 5859-61.
3
pyridinyl-functionalized tetraphenylethylene fluorogen for specific sensing of
trivalent cations. Chem Commun (Camb) 2013, 49, (15), 1503-5.
33. Fang, Y.; Liu, T.; Zou, Q.; Zhao, Y.; Wu, F. Water-soluble benzylidene
cyclopentanone based photosensitizers for in vitro and in vivo antimicrobial
photodynamic therapy. Sci Rep 2016, 6, 28357.
34. Nytko, K. J.; Grgic, I.; Bender, S.; Ott, J.; Guckenberger, M.; Riesterer, O.;
Pruschy, M. The hypoxia-activated prodrug evofosfamide in combination with
multiple regimens of radiotherapy. Oncotarget 2017, 8, (14), 23702-23712.
9. Hales, B. F. Comparison of the Mutagenicity and Teratogenicity of
0. Colvin, M.; Brundrett, R. B.; Kan, M.-N. N.; Jardine, l.; Fenselau, C.
1. Li, Z.; Gao, X.; Shi, W.; Li, X.; Ma, H. 7-((5-Nitrothiophen-2-yl)methoxy)-3H-
2. Chen, X.; Shen, X. Y.; Guan, E.; Liu, Y.; Qin, A.; Sun, J. Z.; Tang, B. Z. A
3
5. Hunter, F. W.; Young, R. J.; Shalev, Z.; Vellanki, R. N.; Wang, J.; Gu, Y.; Joshi,
N.; Sreebhavan, S.; Weinreb, I.; Goldstein, D. P.; Moffat, J.; Ketela, T.; Brown, K. R.;
Koritzinsky, M.; Solomon, B.; Rischin, D.; Wilson, W. R.; Wouters, B. G.
Identification of P450 Oxidoreductase as a Major Determinant of Sensitivity to
Hypoxia-Activated Prodrugs. Cancer Res 2015, 75, (19), 4211-23.
3
6. Xinglu Huang, J. Z., Seung Woo Chung, Buwei Huang, Gilad Halpert, Karina
Negron, Xuanrong Sun, Jun Yang, Yumin Oh, Paul M. Hwang, Justin Hanes, Jung
Soo Suk. Hypoxia-tropic Protein Nanocages for Modulation of Tumor- and
Chemotherapy-Associated Hypoxia. ACS nano 2019, 13, (1), 236-247.
3
7. Gao, F.; Wu, J.; Gao, H.; Hu, X.; Liu, L.; Midgley, A. C.; Liu, Q.; Sun, Z.; Liu,
Y.; Ding, D.; Wang, Y.; Kong, D.; Huang, X. Hypoxia-tropic nanozymes as oxygen
generators for tumor-favoring theranostics. Biomaterials 2020, 230, 119635.
3
Greenwald, D. R.; Ruoslahti, E. Coadministration of a Tumor-Penetrating Peptide
Enhances the Efficacy of Cancer Drugs. Science 2010, 328, (5981), 1031-1035.
8. Sugahara, K. N.; Teesalu, T.; Karmali, P. P.; Kotamraju, V. R.; Agemy, L.;
ACS Paragon Plus Environment