Synthesis and biological evaluation of NQO1-activated prodrugs of podophyllotoxin as antitumor agents

Article abstract of DOI:10.1016/j.bmc.2020.115821

Podophyllotoxin (PPT), a toxic polyphenol derived from the roots of genus Podophyllum, had been reported with strong inhibition on both normal human cells and tumor cells, which hindered the development of PPT as the candidate antitumor agent. In the pres

Full text of DOI:10.1016/j.bmc.2020.115821

Journal Pre-proofs
Synthesis and biological evaluation of NQO1-activated prodrugs of podophyl-
lotoxin as antitumor agents
Yan Qu, Chong Zhang, Xin Ma, Yunhui Gao, Jing Liu, Liqiang Wu
PII:
S0968-0896(20)30651-9
https://doi.org/10.1016/j.bmc.2020.115821
BMC 115821
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Accepted Date:
2 September 2020
5 October 2020
Please cite this article as: Y. Qu, C. Zhang, X. Ma, Y. Gao, J. Liu, L. Wu, Synthesis and biological evaluation of
NQO1-activated prodrugs of podophyllotoxin as antitumor agents, Bioorganic & Medicinal Chemistry (2020),
doi: https://doi.org/10.1016/j.bmc.2020.115821
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.
© 2020 Published by Elsevier Ltd.

Synthesis and biological evaluation of NQO1-activated prodrugs
of podophyllotoxin as antitumor agents
Yan Qu a, 1, Chong Zhangb,c,1, Xin Mab, Yunhui Gaob, Jing Liub, Liqiang Wub,*
aSchool of Basic Medical Sciences, Xinxiang Medical University, Xinxiang 453003,
China.
bSchool of Pharmacy, Xinxiang Medical University, Xinxiang 453003, China
cDepartment of Pharmacology, Xi'an Jiaotong University School of Medicine, 76 West
Yanta Road, Xi'an 710061, China.
*Corresponding author. E-mail: wliq1974@163.com
1Co first-authors
ABSTRACT Podophyllotoxin (PPT), a toxic polyphenol derived from the roots of
genus Podophyllum, had been reported with strong inhibition on both normal human
cells and tumor cells, which hindered the development of PPT as the candidate
antitumor agent. In the present work, multiple NQO1-activatable PPT prodrugs were
synthesized for reducing normal cell toxicity and keeping tumor cell toxicity. The
antiproliferative activities in vitro showed prodrug 3 was greatly selectively toxic to
tumor cells over-expressing NQO1, taxol-resistant A549, hypoxia A549 and HepG2,
and lower damage to normal cells in comparison with podophyllotoxin, prodrug 1 and
2. As elucidated by further mechanistic research, prodrug 3 was activated via NQO1 to
efficiently while gently produce cytotoxic PPT units and kill tumor cells. In additions,
in vivo study revealed that 3 significantly suppressed cancer growth in HepG2
xenograft models without obvious toxicity. Therefore, this NQO1-activatable prodrug
delivery system exhibits good biosafety and provides a novel strategy for the
development of drug delivery systems.
Keywords: Podophyllotoxin; NQO1-activatable prodrug; NQO1; Antitumor

1. Introduction
Naturally occurring substances account for the abundant sources to discover the
anticancer drugs,1 however, a number of antitumor naturally-occurring substances do
not show target specificity, which may even show adverse toxicity to non-carcinoma
cells. For reducing cytotoxicity and maintaining the treatment activity of those bio-
active naturally occurring substances, appropriate structural modification may be
adopted to be the vital way to develop drugs. Podophyllotoxin (PPT), one of the
naturally occurring aryltetralin lignins, is separated based on plant podophyllum.2
PPT, together with the derivatives, is suggested recently to show obvious anticancer
effects.3 Unfortunately, PPT shows high toxicity to non-carcinoma cells,2 which
hinders the development of it to antitumor agent. Moreover, it is of great challenge to
lower PPT toxicity to non-carcinoma cells and maintain its toxicity against tumor cells.
Tumor-specic prodrugs can be accurately activated at cancer sites via species
showing selectivity to tumor, like glutathione (GSH),4,5 enzymes6,7 and reactive oxygen
species (ROS),8,9 which can significantly improve the therapeutic efficacy of anticancer
drugs. NAD(P)-H:quinone oxidoreductase-1 (NQO1, DT-diaphorase, EC 1.6.99.2)
catalyzes the direct two electrons reduction for various quinones, with NADPH and
NADH being the cofactors.10 NQO1 is up-regulated within some human cancer
tissues,11-17 such as non-small cell lung carcinoma (NSCLC), pancreatic carcinoma
(PCa), or colorectal cancer (CRC), as a result, it is a meaning target to activate quinone
prodrugs or the fluorescent probes.18-24 In this work, we designed and synthesized three
NQO1-activatable PPT prodrugs which were constituted by an active drug PPT, a

different self-immolative linker, and a NQO1-responsive trigger group, respectively
(Fig. 1). Based on the outcomes of preliminary tumor cell selection, compound 3 with
the highest potency and selectivity was assessed for its ability of PPT release
responding to NQO1, together with its antitumor effect and stability within plasma or
buffers.
O
O
OMe
OMe
OMe
O
O
O
OMe
OMe
OMe
O
O
O
O
O
O
O
O
O
N
R
O
O
O
O
1
2, R = H
3, R= CH3
Fig. 1. Structures for NQO1-activatable PPT prodrugs.
2. Results and discussion
2.1. Chemistry
Scheme 1 and 2 describe the synthetic routes for prodrugs 1-3. Firstly, with dry
methanesulfonic acid, trimethylhydroquinone 4 reacted with the 3,3-dimethylacrylic
acid to yield 5, and it was then subjected to N-bromosuccinimide (NBS) treatment
within the aqueous acetone, finally yielding 6 (Scheme 1). Quinone unit 6 was coupled
to PPT by EDCI, along with appropriate volume of DMAP for producing 1. The
syntheses of prodrugs 2 and 3 are shown in Scheme 2. Compound 6 was activated with
isobutyl chloroformate and coupled with p-aminobenzylalcohol 7a and p-
methylaminobenzylalcohol 7b for giving compounds 8a and 8b, respectively. Finally,
compounds 8a and 8b were activated by 4-nitrophenyl chloroformate with DIPEA,

followed by treatment with PPT, to yield the prodrugs 2 and 3, in parallel.
O
HO
HO
O
b
a
OH
OH
O
O
O
4
5
6
Scheme 1. Synthesis of 6. Reagents and conditions: (a) 3,3-dimethylacrylic acid,
CH3SO3H, 70 oC; (b) NBS.
OH
O
O
O
O
OMe
OMe
OMe
O
O
O
O
O
O
O
O
a
OH
+
O
MeO
OMe
OMe
PPT
O
O
6
1
NO2
O
O
O
OH
c
O
O
O
O
OH
O
O
N
R
b
N
O
R
O
O
8a, R = H
8b, R = CH3
9a, R = H
9b, R = CH3
6
O
O
OMe
OMe
OMe
O
O
O
O
O
O
d
N
R
O
O
2, R = H
3, R = CH3
Scheme 2. Synthesis of prodrugs 1-3. Reagents and conditions: (a) DCC, DMAP; (b)
isobutyl chloroformate, p-aminobenzylalcohol (7a), p-methylaminobenzylalcohol (7b);
(c) p-nitrobenzylchloroformate, DIPEA; (d) PPT, DMAP.
2.2. In vitro antiproliferative activity
After the synthesis of prodrugs 1-3, five cell lines including NQO1-overexpressing

A549 and HepG2, hypoxia-induced A549 (A549/Hyp) and HepG2 (HepG2/Hyp),
taxol-resistant A549 (A549/T), normal human liver LO2 cell lines were screened for
better evaluating their antiproliferative effects. As shown in Table 1, prodrug 3
displayed the most potent activities against NQO1-overexpressing A549 and HepG2
cells, and IC 50 values were 84 and 100 nM, respectively, which were comparable to
those of PPT. In comparison, 1-2 were relatively less active toward the two cells. It was
possibly because of the rapid reducibility of 3, resulting in unfavorable degradation
within the extracellular environment. As regards therapeutic range, compound 3
showed greater selectivity, normal liver cells in which NQO1 was not detected had the
IC50 of 313 nM, which was 3.72 folds higher than the anti-HepG2 effect, while 3.67
folds greater than PPT (85 nM), suggesting that prodrug 3 showed a broad therapeutic
index. In addition, compared with PPT, 3 had a significant inhibitory effect on A549/T
cells, HepG2/Hyp and A549/Hyp cells. Based on the above research on antiproliferative
activity, prodrug 3 of the most potent and selective was further evaluated its effect on
PPT release responding to NQO1, stability within plasma or buffers, and the antitumor
effect.

Table 1. The antiproliferative activities of PPT prodrugs 1-3.
IC50 (nM)
Compound.
HepG2
A549
LO2
HepG2/Hyp A549/Hyp A549/T
1
2
107±11 160±14
262±29 274±39
84±9 100±19
83±8 105±15
228±36
452±33
312±34
85 ±11
193±14
425±32
113±13
209±16
315±44
596±70
185±24
291±25
302±33
574±50
197±29
338±38
3
PPT
2.3 Molecular modelling studies
Molecular modeling was carried out by the use of Autodock vina 1.1.2 for explaining
how 3 was bound within NQO1 As a result, 3’s quinone moiety bound to residues at
NQO1 active site together with flavinadenine dinucleotide (FAD), and this was
necessary for catalytic performance of NQO1 (Fig. 2). For PPT, its leaving groups were
maintained out of binding pocket of substrates, which indicated that reaction between
quinone and active site in NQO1 was unaffected by introducing PPT into quinone.
Fig. 2. Proposed binding model for prodrug 3 (orange) binding with NQO1.

2.4. In vitro drug release profiles of PPT
Prodrug 3 release profile responding to NQO1 and the cofactor NADPH was
investigated through HPLC assay (Fig. 3A ). We observed the prodrug release time
dependence for prodrug 3 (Fig. 3B ). As the incubation time increased, the PPT
proportion showed a steady increase trend, with the plateau being observed at 9 h,
reaching 83.4 %. To determine whether 3 was dependent on NQO1 for release, 3 was
tested combined with dicoumarol (DIC, the NQO1 inhibitor). The results suggested that
transformation of 3 into PPT was suppressed when DIC content increased to 50 μM.
Then, the PPT formation within the recombinant NQO1 at different concentrations of
the prodrug 3 was measured through HPLC assay. Later, the Michaelis−Menten curves
were plotted, and the apparent Km and Vmax were 6.997 μM and 26.18 nmol/mg/min,
respectively (Fig. 3C).
Fig 3. Prodrug 3 release profiles of PPT. (A) HPLC spectrum of 3 in the presence of
NQO1. (B) PPT released from prodrug in the presence of NQO1(red), NQO1+DIC
(blue) respectively. (C) Michaelis−Menten curves for 3 with NQO1.
A plausible process of PPT release from prodrug 3 is proposed in Fig. 4. To respond
to the activation of NQO1 bioreduction, “trimethyl lock” that contained the trigger
group of quinone propionic acid in 3 firstly underwent intramolecular cyclization, later,

linker was removed through the renounced 1,6-elimination reaction for the spontaneous
liberation of PPT, thus specifically targeting tumor cells. Besides, HRMS assay was
conducted to observe the expected PPT production (Fig. 5). Following the coincubation
of 3 with NQO1 as well as NADPH, this study detected related fragments, such as PPT
([M + Na] + m/z =437.1291), 5 ([M + Na] + m/z = 257.1136) and 10 ([M + H] + m/z =
120.0808), which offered enough information for PPT delivery when there was NQO1.
O
O
OMe
O
O
O
OMe
OMe
OMe
O
O
O
OMe
OMe
OH
OH
O
O
O
O
O
O
N
NQO1
N
O
O
O
O
3
OMe
O
MeO
OMe
O
N
OMe
OMe
OMe
O
HO
+
CO2
O
O
O
O
O
O
O
O
N
10
5
H
1,6-elimination
intramolecular cyclization
O
O
OH
PPT
Fig. 4. A plausible process of PPT release from prodrug 3.
Fig. 5. HRMS analyses for 3 following co-incubation with NQO1 for 30 min.

2.5. Stability of prodrug 3
A majority of agents show no tissue specificity, and can be easily removed from liver
and plasma. In this case, only a fraction of agents penetrate the site of treatment, leading
to reduced efficacy. Prodrug 3 displayed good stability under 37 °C within the
phosphate buffer (pH 5-9) in the presence of NADPH, rat plasma and human liver
microsomes incubated for 9 h (Fig. 6). These findings indicated that 3 showed relative
stability for providing sufficient exposure time at physiological conditions, as a result,
3 was targeted to produce cytotoxic substance into tumor tissue that over-expressed
NQO1
Fig. 6. The stability of prodrugs 3. (A) in phosphate buffer. (B) in rat plasma (red) and
human liver microsomes (blue).
2.6. Immunofluorescence and inhibition of microtubule assembly analysis
For investigating the effect of prodrug 3 on disrupting microtubular dynamics with
alive cells, this study carried out immunofluorescent assay based on the HepG2 cells.
According to Fig. 7, HepG2 cells of control group were arranged and organized
normally. Nonetheless, when cells were treated with 50 nM 3 for 48 h, those
microtubular networks within cytosol were destroyed, suggesting that 3 potentially

induced microtubular network collapse. To investigate whether the prodrug was related
to the interactions with microtubule systems, 3 were evaluated as tubulin
polymerization inhibitor. The results indicated that 3 had no inhibitory effects on the
tubulin polymerization activity. Following co-incubation of prodrug 3 with NADPH
and NQO1, the product was capable of inhibiting tubulin polymerization activity. These
results provided sufficient evidence that NQO1 could facilitate the transformation of 3
into PPT which could effective inhibition microtubule assembly and showed the high
antitumor activity.
Fig. 7. (A) Effects of 3 on microtubule network organization, HepG cells were
subjected to 48 h of DMSO, 50 nM of 3, and 50 nM of PPT treatments. (B) Effects of
3 on tubulin polymerization in vitro.
2.7. Analysis of cell cycle
A majority of tubulin-destabilizing drugs might destroy distribution of cell cycle, this
study conducted flow cytometric analysis for examining how 3 affected the progression
of HepG2 cell cycle through staining with propidium iodide (PI). It was observed from
Fig. 8 that, incubation with 3 resulted in cell cycle arrest at G2/M phase. In comparison
with control cells subjected to DMSO incubation, HepG2 cells incubated using 50 nM
3 elevated the cell percentage at G2/M phase between 12.26% and 27.30%.

Fig. 8. 3 effects on the progression of HepG2 cell cycle. Cells were subjected to 24 h
of 3 (50 nM) and PPT (50 nM) treatments.
2.8. Analysis of cell apoptosis
For assessing the effect of compound 3 on inducing cell apoptosis, this study
conducted the Annexin V-FITC/PI assay. It was illustrated from Fig. 9 that, following
24 h treatment, the apoptotic cell percentage was as low as 3.40% of that of control
group. After incubation using 50 nM 3 for 24 h, there were 17.5% apoptotic cells at the
early or later stage. As suggested by the above findings, 3 could efficiently induce
HepG2 cell apoptosis.
Fig. 9. Compound 3 led to apoptosis of HepG2 cells under 3 (50 nM) and PPT (50 nM)
treatments.

2.9. Antitumor activity in vivo
In line with favorable observations of 3 in vitro, the efficacy of this compound in
vivo to resist HepG2 xenograft was assessed using BALB/c nude mouse mode, with.
PPT being the positive control. First, HepG2 cells were subcutaneously inoculated to
mouse right flank to establish the mouse model of liver cancer allograft. When the
tumor grew to a size of about 70-100 mm3 and became palpable, all mice were randomly
divided as blank, vehicle control or treatment group (with 6 in each group). Vehicle or
drugs were administered intraperitoneal every other day for 17 consecutive days.
Vehicle (5% DMSO + 40% PEG300 + 5% Tween 80 + 50% normal saline) was
administered to control mice. PPT was administrated at the dosage of 10 mg/kg, and
two different doses (10 and 5 mg/ kg) for 3 were studied. On day 17 following the initial
administration, each mouse was sacrificed to harvest the tumor. Then, liver, heart,
kidney, lung, and spleen and tumor tissues were sliced and stained using hematoxylin
and eosin (H&E). Animals subjected to PPT treatment died in the experimental period,
which might be due to the high toxicity of PPT. Relative to vehicle group,
intraperitoneal injection with 10 mg/kg 3 markedly reduced tumor growth. Upon the
completion of our observation, it inhibited tumor growth up to 64% (Fig. 10A-C).
Moreover, 3 had no significant effect on body weight, even at 10 mg/kg (Fig. 10D).
Then, the standard H&E staining protocol was applied in histologically examining liver,
heart, lung, kidney spleen and tumor tissues. Under light microscope, the morphology
and structure of these tissues were normal and showed no significant histological
damage as compared with that in the control (Fig 11). The above results suggested that,
compound 3 was successfully designed to decrease PPT cytotoxicity in vivo, which
might be potentially used as the antitumor drug with low toxicity.

Fig.10. 3 suppressed the growth of liver cancer xenograft in vivo. Following vehicle,
PPT (10 mg/kg/two day), 3 (10 mg/kg/ two day) and 3 (5 mg/kg/two day) treatments
for 17 consecutive days, all animals were sacrificed to harvest and weigh the tumors.
(A) Tumor images acquired from mice on 17 days following treatment initiation. (B)
The excised tumor weights in each group. compared with control group. (C) Changes
in mouse tumor volume in the treatment process. (D) Changes in mouse body weight
in the process of treatment. **P <0.01 vs. Control.

Fig. 11. Representative images of heart, liver, spleen, lung, kidney tissues and tumor
stained by H&E .
3. Conclusions
To sum up, this study introduces a novel successful design for converting the
naturally occurring PPT with high toxicity to the potential antitumor agent with low
toxicity. The antiproliferative activities in vitro showed prodrug 3 was greatly
selectively toxic to tumor cells over-expressing NQO1, taxol-resistant A549, hypoxia
A549 and HepG2, and lower damage to normal cells in comparison with ppt, prodrug

1, 2 and 3. Further mechanism studies elucidated that the NQO1 activates prodrug 3 in
a bioreductive way, followed by intramolecular cyclization as well as 1,6-elimination
reaction, thus efficiently liberating parent drug PPT. In additions, as indicated by in
vivo anticancer assessment, the prodrug 3 exerted the stronger anticancer effect and the
higher safety than PPT.
4. Experimental section
4.1. Chemiistry
4.1.1. General
The 98% PPT was provided by Macklin Inc. (Shanghai, China). Recombinant NQO1
(DT-diaphorase) with the purity of >90% was provided by Sigma-Aldrich China Co.,
Ltd. (Shanghai, China). Compound 5 and 8 were prepared according to literature.25 The
remaining reagents were purchased from commercial sources. Melting points were
determined with a XT-4 binocular microscope and are uncorrected. The 1H NMR and
13C NMR spectra were measured on a Bruker AV-400 instrument using deuterated
solvents with tetramethylsilane (TMS) as internal standard. High resolution mass
spectra (HRMS) were recorded on a Bruker micrOTOF-QIII mass spectrometer. The
ACS Accurasil C18 column (dimension, 4.6 mm × 150 mm, 5 μm) was utilized for
HPLC analysis by the use of the solvent acetonitrile /water mixture at the 0.5 mL/min
flow rate, and the UV absorption was observed at the wavelength of 292 nm.
4.1.2. Synthesis of prodrug 1
A mixture containing 5 (140 mg, 0.6 mmol), EDCl (115mg, 0.6mol) and CH2Cl2 (10
mL) was incubate at 0 oC. After stirring for 30min, 6 mg DMAP (0.05 mmol) and 207
mg PPT (0.5 mmol) were put into the resultant mixture for 24 h reaction under ambient
temperature. After reaction completion, the mixed product was rinsed by the NaHCO3

saturated solution together with brine. Then, anhydrous Na2SO4 was used for drying
the solvent. After filtration, evaporation in vacuum was conducted to concentrate the
organic solvent, and column chromatography was conducted to purify the crude
products, with petroleum ether/ethyl acetate (v:v=2:1) being used to be the eluent, for
yielding 293 mg (51%) yellow power, m.p.102-103 oC; 1H NMR (400 MHz, DMSO-
d6) δ: 6.75 (s, 1H), 6.60 (s, 1H), 6.32 (s, 1H), 6.02 (d, 2H, J = 10 Hz), 5.82 (d, 1H, J =
10 Hz), 4.54 (d, 1H, J = 4.4 Hz), 4.20-4.08 (m, 2H), 3.65 (s, 6H), 3.62 (s, 3H), 3.36-
3.31 (m, 1H), 3.12 (d, 1H, J = 16.4 Hz), 2.99 (d, 1H, J = 16.4 Hz), 2.75-2.67 (m, 1H),
2.10 (s, 3H), 1.89 (s, 3H), 1.86 (s, 3H), 1.43 (s, 3H), 1.42 (s, 3H); 13C NMR (100 MHz,
DMSO-d6) δ: 190.7, 187.2, 174.3, 173.1, 152.5 (2C), 152.3, 147.8, 147.4, 142.7, 139.3,
138.6, 136.9, 136.1, 132.7, 129.1, 109.9, 108.5 (2C), 106.9, 101.9, 73.3, 70.9, 60.4,
56.2 (2C), 47.6, 44.5, 43.3, 38.6, 38.2, 29.0 (2C), 14.5, 12.8, 12.3; HRMS-ESI (m/z):
calcd for C36H38NaO11 [M+Na]+ 669.2306, found: 669.2308.
4.1.3. Synthesis of compound 9
Compound 8 (1 mmol) was dissolved in dry CH2Cl2 (30 mL), and 4-nitrophenyl
chloroformate (425 mg, 2 mmol) was added. To the above solution, DIPEA (260 mg,
2 mmol) was added dropwise. The resultant mixture was kept stirring for 6 h at room
temperature. After completion of the reaction, 100 mL water was used to wash the
reaction solution for thrice, and anhydrous Na2SO4 was used for drying. Later, solvent
was discarded, silica gel column chromatography was performed to purify the residue
on, with petroleum ether/ethyl acetate (v:v=3:1) being used to be the eluent for
obtaining compound 9.
N-(4-(hydroxymethyl)phenyl)-3-methyl-3-(2,4,5-trimethyl-3,6-dioxocyclohexa-
1,4-dienyl)butanamide (9a). Compound 9a was prepared from 8a as a yellow solid,
m.p.127-128 oC, yield 69%, m.p.127-128 oC. 1H NMR (400 MHz, CDCl3) δ: 8.19 (d, J

= 7.2 Hz,2H), 7.38 (d, J = 8.4 Hz,2H), 7.29 (d, J = 8.4 Hz,4H), 7.15 (s, 1H), 5.16 (s,
2H), 2.97 (s, 2H), 2.09 (s, 3H), 1.89 (s, 3H), 1.88 (s, 3H), 1.42 (s, 6H); 13C NMR (100
MHz, CDCl3) δ: 190.6, 186.5, 169.4, 154.5, 151.7, 151.4, 144.3, 142.1, 137.5, 137.4,
137.3, 128.8, 128.7 (2C), 124.3 (2C), 120.8 (2C), 118.8 (2C), 69.6, 49.4, 37.4, 28.1
(2C), 13.2, 11.7, 11.2; HRMS-ESI (m/z): calcd for C28H28N2NaO8 [M+Na]+ 543.1738,
found: 543.1747.
N-(4-(hydroxymethyl)phenyl)-N,3-dimethyl-3-(2,4,5-trimethyl-3,6-
dioxocyclohexa-1,4-dienyl)butanamide (9b). Compound 9a was prepared from 8a as a
yellow oil, yield 68%, 1H NMR (400 MHz, CDCl3) δ: 8.29 (d, J = 9.2 Hz, 2H), 7.55 (d,
J = 8.0 Hz, 2H), 7.43 (d, J = 9.2 Hz, 2H), 7.28 (d, J = 8.0 Hz, 2H), 5.36 (s, 2H), 3.18
13
(s, 3H), 2.78 (s, 3H), 2.11 (s, 3H), 2.01 (s, 3H), 1.96 (s, 3H), 1.42 (s, 6H); C NMR
(100 MHz, CDCl3) δ: 191.2, 187.6, 172.0, 155.4, 154.6, 152.4, 145.4, 144.5, 143.5,
137.8, 136.3 (2C), 133.9, 130.1 (2C), 127.9, 125.3 (2C), 121.8 (2C), 70.1, 47.7, 38.1,
37.1, 28.5 (2C), 14.1, 12.7, 12.1; HRMS-ESI (m/z): calcd for C29H30N2NaO8 [M+Na]+
557.1900, found: 557.1890.
4.1.4. Synthesis of prodrug 2
Compound 9a (520 mg, 1 mmol) and DMAP (122 mg, 1 mmol) were dissolved in
dry CH2Cl2 (20 mL), then added PPT ((414 mg, 1 mmol), and the reaction mixture was
stirred for overnight at room temperature. Later 50 mL was used to wash reaction
mixture for twice, and then organic layer was isolated, followed by drying with
anhydrous Na2SO4. Later, rotary evaporation was performed to remove solvents,
column chromatography was applied on silica gel to purify crude product, with
petroleum CHCl3/ethyl acetate (v:v=10:1) being used to be the eluent for yielding
1
compound 2 (382 mg, 48%) as a yellow oil. H NMR (400 MHz, CDCl3) δ: 7.44 (d,
2H, J = 8.4 Hz), 7.33 (d, 2H, J = 8.4 Hz), 7.19 (s, 1H),6.84 (s, 1H), 6.53 (s, 1H), 6.37

(s, 2H), 5.98 (s, 2H), 5.74 (dd, 1H, J = 3.2, 7.2 Hz), 5.14 (s, 2H), 4.60 (d, 1H, J = 2.0
Hz), 4.44-4.40 (m, 1H), 4.24-4.19 (m, 1H), 3.80 (s, 3H), 3.74 (s, 6H), 3.04 (s, 2H), 2.92
(d, 2H, J = 3.2 Hz), 2.16 (s, 3H), 1.95 (s, 3H), 1.94 (s, 6H), 1.49 (s, 6H); 13C NMR (100
MHz, CDCl3) δ: 191.6, 187.5, 173.6, 170.4, 155.4, 152.6 (2C), 148.3, 147.7, 143.2,
138.5, 138.2, 137.0, 134.6, 132.3, 129.5 (2C), 127.6, 127.4, 126.2, 119.8 (2C), 115.6,
109.7, 107.9 (2C), 107.2, 101.6, 71.2, 69.9, 60.8, 56.1 (2C), 50.5, 45.5, 43.7, 38.4, 29.7,
29.1 (2C), 19.0, 14.2, 12.7, 12.2; HRMS-ESI (m/z): calcd for C44H45NNaO13 [M+Na]+
818.2783, found: 818.2784.
4.1.6. Synthesis of prodrug 3
Compound 9b (534 mg, 1 mmol) and DMAP (122 mg, 1 mmol) were dissolved in
dry CH2Cl2 (20 mL), then added PPT ((414 mg, 1 mmol), and the reaction mixture was
stirred for overnight at room temperature. Later 50 mL was used to wash reaction
mixture for twice, and then organic layer was isolated, followed by drying with
anhydrous Na2SO4. Later, rotary evaporation was performed to remove solvents,
column chromatography was applied on silica gel to purify crude product, with
petroleum CHCl3/ethyl acetate (v:v=10:1) being used to be the eluent for yielding
1
compound 2 (382 mg, 48%) as a yellow soild, m.p.115-116 oC.. H NMR (400 MHz,
CDCl3) δ: 7.49 (d, 2H, J = 7.6 Hz), 7.25 (d, 2H, J = 7.6 Hz), 6.87 (s, 1H), 6.55 (s, 1H),
6.39 (s, 2H), 5.99 (d, 2H, J = 1.6 Hz), 5.80 (d, 1H, J = 7.6 Hz), 5.30 (s, 2H), 4.61 (d,
1H, J = 3.2 Hz), 4.48-4.44 (m, 1H), 4.28-4.23 (m, 1H), 3.80 (s, 3H), 3.78 (s, 6H), 3.17
(s, 3H), 2.99-2. 92 (m, 2H), 2.76 (s, 2H), 2.11 (s, 3H), 2.00 (s, 3H), 1.94 (s, 3H), 1.31
(s, 6H); 13C NMR (100 MHz, CDCl3) δ: 191.3, 187.7, 173.5, 172.1, 155.4, 154.5, 152.7
(2C), 148.4, 147.7, 144.4, 143.4, 138.0, 137.2, 136.5, 134.6, 134.4, 132.4, 129.8, 127.8,
127.5, 126.1, 115.5, 109.8 (2C), 108.1, 107.1, 101.7, 77.7, 71.2, 69.5, 60.8, 56.2 (2C),
53.4, 47.8, 45.5, 43.7, 38.4, 38.1, 28.5 (2C), 14.1, 12.7, 12.1; HRMS-ESI (m/z): calcd

for C45H48NaO13 [M+Na]+ 810.312, found: 810.3123.
4.2. Antiproliferative experiment
The A549, HepG2, A549/Hyp, HepG2/Hyp, A549/T cells, and normal LO2 cells
were used to test compound anti-proliferative effects. The logarithmic cells were
inoculated into the 96-well plates for adherence, followed by 48 h of incubation using
compounds at specific doses. After adding MTT, cells were further incubated for 2-3 h.
Later, MTT formazan crystal was mixed into DMSO, and absorbance (A) was detected
at 570 nm.
4.3. Molecular modeling studies
NQO1 crystal structure in the complex containing a coumarin-based inhibitor (PDB
code 3JSX) was used for docking study. Then, ligand was eliminated, leaving the
receptor complex in later docking study, followed by the geometrical addition of polar
hydrogen atoms into protein. AutoDock Tools were utilized to assign docking regions.
The default parameters were applied in Vina docking, except as otherwise specified.
This study selected the optimal-scoring pose determined based on Vina docking score,
while PyMoL 1.7.6 (http://www.pymol.org/) was used for visual analysis.
4.4. HPLC and HRMS assays for drug 3 release studies
HPLC was used to determine the activation of prodrug, with NQO1 being the
stimulus (eventual dose of 10 μg/mL). Incubation mixtures contained 10 mM PBS with
2 % tween 80, 100 μM NADPH, and 20 μM prodrug 3. Recombinant NQO1 was
directly put to the incubation mixtures for reaction initiation and then the solution was
incubated at 37 oC for appropriate time intervals. The reaction was terminated by adding
cold acetonitrile (100 μL). After centrifugation in a refrigerated centrifuge, 20 µL of
the liquid were loaded and analyzed by HPLC and HRMS analysis. To investigate the
DIC effect, NQO1 was previously incubated using 50 μM DIC for 20 min.

4.5. Formation rates of CA-4 in recombinant NQO1 from PPT
The rate of NQO1 catalysis were calculated during the first 30 min to determine the
susceptibility of the prodrugs 3 on target enzyme NQO1. The assay mixtures contained
NADPH (100 μM), recombinant human NQO1 (10 μg/mL) and 3 (40, 20, 10, 5, 2.5,
1.25, 0.625 μM ) within 100 μL PBS (10 mmol , pH 7.4) with 2 % tween 80. 30 min of
o
reactions were conducted under 37 C, and 100 μL cold acetonitrile was added to
terminate reactions. After centrifugation in a refrigerated centrifuge, 20 µL of the liquid
were loaded and analyzed by HPLC. Enzymatic variables (Km along with Vmax) were
determined based on kinetic curves using GraphPad Prism 7.0.
4.6. Stability study of 3
HPLC was conducted for examining prodrug 3 stability in PBS buffer with different
pH values, in rat plasma and pooled Human Liver S9 Fraction. The assays were
performed at 37 °C and conducted in triplicate.
Stability of prodrug 3 in PBS buffer with different pH values: In the absence of
NQO1, the prodrug (10 μM) was co-incubated with cofactor NADPH (100 μM) in PBS
with 2 % tween 80 at different pH values. After incubation for appropriate time intervals,
the reaction was terminated by adding cold acetonitrile (100 μL). After centrifugation
in a refrigerated centrifuge, 20 µL of the liquid were loaded and analyzed by HPLC.
Metabolic Stability in plasma: Rat plasma were subjected to vortex mixing and then
centrifugation for 5 min at 10000 rpm to deproteinize and the resulting supernatants
were withdrawn, prodrug 3 (10 μM) was co-incubated with supernatants with 2 %
tween 80 at 37 °C for 0, 10, 30, 60, 1800, 360 and 540 min. After centrifugation in a
refrigerated centrifuge, 20 µL of the liquid were loaded and analyzed by HPLC.
Metabolic Stability in pooled Human Liver S9 Fraction: The metabolic stability was
performed in 10 mmol M PBS (pH 7.4) containing 2 % tween 80, 3 (10 μM) and pooled

Human Liver S9 Fraction (0.5 mg) at 37 °C. The reaction was started by adding 100
μM NADPH and terminated at 0, 10, 30, 60, 1800, 360 and 540 min by adding 100 µL
cold acetonitrile . After centrifugation in a refrigerated centrifuge. 20 µL of the liquid
were loaded and analyzed by HPLC with UV detection under 292 nm.
4.7. Cell cycle analysis
HepG2 cells were first of all inoculated into the 12-well plates for 24 h, followed by
24 h incubation after 3 treatment. Then, cells were harvested and rinsed by cold PBS
for two times, followed by 3 h suspension within the 70% ethanol. Following addition
of 5 mL RNase A (10 mg/mL), the suspension was subjected to 1 h incubation under
ambient temperature. Subsequently, 10 mL PI (50 mg/mL) was also put into the
suspension. At last, cell proportions at diverse cell cycle phases were measured by flow
cytometry..
4.8. Cell apoptosis analysis
The Annexin-V FITC/PI assay was conducted to evaluate the apoptotic effect of
prodrug 3. First of all, the HepG2 cell line was cultivated into the 6-well plates for 24
h incubation. Thereafter, the original medium was replaced with the complete medium
supplemented with 3 (50 nM) to incubate for another 24 h. Later, supernatant was
collected to obtain cells through trypsinization, which were subjected to PBS washing,
15 min Annexin-V/FITC (5 mL) staining within PBS under ambient temperature, and
10 min staining using PI solution. The apoptosis levels were evaluated by flow
cytometry.
4.9. Immunofluorescence staining
HepG2 were inoculated into the 96-well plate to incubate under 37 oC for 12 h under
5% CO2 condition. Later, cells were subjected to 48 h DMSO, PPT, and 3 treatments,
separately, methanol fixation as well as 0.5% Triton X-100/PBS permeabilization.

Then, cells were carried out overnight incubation using the α-tubulin primary antibody
under 4 oC. Following PBS washing, cells were subjected to 1 h incubation using related
fluorescence-labeled secondary immunofluorescence staining antibody. Later, DAPI
was used to label cell nucleus, and the fluorescence microscope was utilized for
visualization.
4.10. Tubulin polymerization test
The tubulin polymerization test was conducted in accordance with manufacturer’s
instructions using the tubulin polymerization kit (BK004P, Cytoskeleton). Absorbance
were recorded for 1 h at 340 nm wavelength every 30 seconds, to obtain the kinetic
curves of polymerization and depolymerization of microtubules. The reaction groups
for these experiments included: vehicle control, 3 (3 μM), 3 (3 μM) + NQO1, PPT (3
μM) and Paclitaxel (3 μM).
4.11. In vivo antitumor effect
HepG2 cells were utilized to investigate the antitumor effect of 3 within the BALB/c
nude mice. The BALB/c nude mice (6 weeks old) were provided by Vital River
Laboratory Animal Technology Co. Ltd. Then, 5×106 HepG2 cells contained within the
100 mL sterile PBS were subcutaneously injected into the dorsal regions of mice to
induce tumor. When the tumor grew to 70-100 mm3, the animals were randomLy
separated as 4 groups (with 6 in every group): control, 10 mg/kg PPT, 5/10 mg/kg 3,
and the compounds were administrated every other day for 17 days by intraperitoneal
injection. All the compounds were prepared with the formulation in 5% DMSO, 40%
PEG 300, 5% tween-80 and 50% saline. Body weight as well as tumor volume was
measured at an interval of 2 days following prodrug treatment, among which, the latter
was recorded using calipers through determining length (A) and width (B) for
calculation according to the formula V =AB2/2. In addition, tumor weight was

measured and tumor growth inhibition rate was calculated upon treatment completion.
Liver tissue samples were obtained following execution, which were then subjected to
4% paraformaldehyde fixation, ethanol dehydration, paraffin embedding, followed by
HE staining and microscopic observation.
Acknowledgements
This work was supported by NSFC-Henan Joint Fund (No. U1604164) and the
Doctoral Scientific Research Activation Foundation of Xinxiang Medical University
(No. XYBSKYZZ201906)
Appendix A. Supplementary data
Supplementary data to this article can be found online at
References
1. Newman, D.J.; Cragg, G.M. Natural products as sources of new drugs over the last
25 years. J. Nat. Prod. 2007, 70, 461–477.
2. Gordaliza, M.; García, P.A.; Miguel del Corral, J.M.; Castro, M.A.; Gómez-Zurita,
M.A. Podophyllotoxin: distribution, sources, applications and new cytotoxic
derivatives. Toxicon 2004, 44, 441−449.
3. Gordaliza, M.; Castro, M.A.; Corral, J.M.M.; San Feliciano, del A. Antitumor
properties of podophyllotoxin and related compounds. Current Pharm. Design
2000, 6, 1811-1839.
4. Liu, Y.; Zhu, S.; Gu, K.; Guo, Z.; Huang, X.; Wang, M.; Amin, H.M.; Zhu, W.;
Shi, P. GSH-activated NIR fluorescent prodrug for podophyllotoxin delivery. ACS
Appl. Mater. Interfaces 2017, 9, 29496−29504.
5. Maiti, S.; Park, N.; Han, J.H.; Jeon, H.M.; Lee, J.H.; Bhuniya, S.; Kang, C.; Kim,
J.S. Gemcitabine-coumarin-biotin conjugates: A target specific theranostic
anticancer prodrug. J. Am. Chem. Soc. 2013, 135, 4567–4572.

6. Schellmann, N.; Deckert, P.M.; Bachran, D.; Fuchs, H.; Bachran, C. Targeted
enzyme prodrug therapies. Mini-Rev. Med. Chem. 2010, 10, 887-904.
7. Liu, P.; Xu, J.; Yan, D.; Zhang, P.; Zeng, F.; Li, B.; Wu, S.A. DT-diaphorase
responsive theranostic prodrug for diagnosis, drug release monitoring and therapy,
Chem. Commun. 2015, 51, 9567-9570.
8. Ou, K.; Kang, Y.; Chen, L.; Zhang, X.; Chen, X.; Zheng, Y.; Wu, J.; Guan, S.
H2O2-responsive nano-prodrug for podophyllotoxin delivery. Biomater. Sci. 2019,
7, 2491–2498.
9. Redy-Keisar, O.; Ferber, S.; Satchi-Fainaro, R.; Shabat, D. NIR fluorogenic dye as
a modular platform for prodrug assembly: real-time in vivo monitoring of drug
release. ChemMedChem 2015, 10, 999-1007.
10. Siegel, D.; Gustafson, D.L.; Dehn, D.L.; Han, J.Y.; Boonchoong, P.; Berliner, L.J.;
Ross, D. NAD(P)H: quinone oxidoreductase 1: role as a superoxide scavenger. Mol.
Pharmacol. 2004, 65, 1238-1247.
11. Yang, Y.; Zhang, Y.; Wu, Q.; Cui, X.; Lin, Z.; Liu, S.; Chen, L. Clinical
implications of high NQO1 expression in breast cancers. J. Exp. Clin. Cancer Res.
2014, 33, 14.
12. Awadallah, N.S.; Dehn, D.; Shah, R.J. NQO1 expression in pancreatic cancer and
its potential use as a biomarker. Appl. Immunohisto. M. M. 2008, 16, 24–31.
13. Yang, Y.; Zhang, Y.; Wu, Q.Y.; Cui, X.L.; Lin, Z.H.; Liu, S.P.; Chen, L.Y. Clinical
implications of high NQO1 expression in breast cancers. J. Exp. Clin. Cancer Res.
2014, 33, 1–9.
14. Belinsky, M.; Jaiswal, A.K. NAD (P) H: quinone oxidoreductase1 (DT-Diaphorase)
expression in normal and tumor tissues. Cancer Metastasis Rev. 1993, 12, 103-
117.

15. Cresteil, T.; Jaiswal, A.K. High levels of expression of the NAD(P)H: Quinone
oxidoreductase (NQO1) gene in tumor cells compared to normal cells of the same
origin. Biochem. Pharmacol. 1991, 42, 1021-1027.
16. Smitskamp-Wilms, E.; Giaccone, G.; Pinedo, H.M.; van der Laan, B.F.; Peters, G.J.
DT-diaphorase activity in normal and neoplastic human tissues; an indicator for
sensitivity to bioreductive agents? Br. J. Cancer 1995, 72, 917–921.
17. Mikami, K.; Naito, M.; Ishiguro, T.; Yano, H.; Tomida, A.; Yamada, T.; Tanaka,
N.; Shirakusa, T.; Tsuruo, T. Immunological quantitation of DT-diaphorase in
carcinoma cell lines and clinical colon cancers: advanced tumors express greater
levels of DT-diaphorase. Jpn. J. Cancer Res. 1998, 89, 910-915.
18. Zhang, K.; Chen, D.; Ma, K.; Wu, X.; Hao, H.; Jiang, S. NAD(P)H: Quinone
oxidoreductase 1 (NQO1) as a therapeutic and diagnostic target in cancer. J. Med.
Chem. 2018, 61, 6983-7003.
19. Danson, S.; Ward, T.H.; Butler, J.; Ranson, M. DT-diaphorase: a target for new
anticancer drugs. Cancer Treat. Rev. 2004, 30, 437-449.
20. Mendoza, M.F.; Hollabaugh, N.M.; Hettiarachchi, S.U.; McCarley, R.L. Human
NAD(P)H:quinone oxidoreductase type I (hNQO1) activation of quinone
propionic acid trigger groups. Biochemistry 2012, 51, 8014-8026.
21. Zhang, X.; Li, X.; Li, Z.; Wu, X.; Wu,Y.; You, Q.; Zhang, X. An NAD(P)H:
Quinone oxidoreductase 1 responsive and self-immolative prodrug of 5-
fluorouracil for safe and effective cancer therapy. Org. Lett. 2018, 20, 3635−3638.
22. Suebsakwong, P.; Wang, J.; Khetkam, P.; Weerapreeyakul, N.; Wu, J.; Du, Y.; Yao,
Z.-J.; Li, J.-X.; Suksamrarn, A. A Bioreductive prodrug of cucurbitacin B
significantly inhibits tumor growth in the 4T1 xenograft mice model. ACS Med.
Chem. Lett. 2019, 10, 1400-1406.

23. Liu, P.; Xu, J.; Yan, D.; Zhang, P.; Zeng, F.; Li, B.; Wu, S. A DT-diaphorase
responsive theranostic prodrug for diagnosis, drug release monitoring and
therapy. Chem. Commun. 2015, 51, 9567-9570.
24. Liu, M.; Song, W.; Du, X.; Su, J.; Dong, K.; Chen, Y.; Peng, Z. NQO1-selective
activated prodrug of triptolide: Synthesis and antihepatocellular carcinoma activity
evaluation. ACS Med. Chem. Lett. 2018. 9, 1253-1257.
25. Zhou, W.; Leippe, D.; Duellman, S.; Sobol, M.; Vidugiriene, J.; O'Brien, M.;
Shultz, J.W.; Kimball, J.J.; DiBernardo, C.; Moothart, L.; Bernad, L.; Cali, J.;
Klaubert, D.H.; Meisenheimer, P. Self-immolative bioluminogenic quinone
luciferins for NAD(P)H assays and reducing capacity-based cell viability assays.
Chembiochem. 2014, 15, 670-675.

Highlights
►Three NQO1-activatable podophyllotoxin prodrugs were synthesized
► Prodrug 3 had highly selective toxicity to cancer cells and lower damage to normal
cells
► Prodrug 3 could be activated by NQO1 and effectively liberate podophyllotoxin and
kill tumor cells.
► Prodrug 3 exerted a stronger anticancer activity and a more favorable safety profile
in vivo

Graphical abstract

Declaration of interests
The authors declare that they have no known competing financial interests or personal
☒
relationships that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests: