A hydrogen peroxide-activated Cu(II) pro-ionophore strategy for modifying naphthazarin as a promising anticancer agent with high selectivity for generating ROS in HepG2 cells over in L02 cells

Article abstract of DOI:10.1016/j.freeradbiomed.2019.12.001

Targeting redox vulnerability of cancer cells by pro-oxidants capable of generating reactive oxygen species (ROS) has surfaced as an important anticancer strategy. Due to the intrinsic narrow therapeutic window and other dangerous side effects of ROS generation, it is highly needed and challenging to develop pro-oxidative anticancer agents (PAAs) with high selectivity for generating ROS in cancer cells. Herein we report a hydrogen peroxide (H₂O₂)-activated Cu(II) pro-ionophore strategy to develop naphthazarin (Nap) as such type of PAAs based on the H₂O₂-mediated conversion of boronate to free phenol. The boronate-protected Nap (PNap) can exploit increased levels of H₂O₂ in HepG2 cells to in situ release Nap followed by its efflux via conjugation with reduced glutathione (GSH), allowing that the Nap-GSH adduct works as a Cu(II) ionophore to induce continuously GSH depletion via a reduction-dependent releasing of Cu(I) by GSH. This strategy endows PNap with the unprecedented ability to hit multi-redox characteristics (increased levels of H₂O₂, GSH and copper) of HepG2 cells, leading to ROS generation preferentially in HepG2 cells along with their selective death.

Full text of DOI:10.1016/j.freeradbiomed.2019.12.001

Journal Pre-proof
A hydrogen peroxide-activated Cu(II) pro-ionophore strategy for modifying
naphthazarin as a promising anticancer agent with high selectivity for generating ROS
in HepG2 cells over in L02ꢀcells
Xia-Zhen Bao, Qi Wang, Xiao-Rong Ren, Fang Dai, Bo Zhou
PII:
S0891-5849(19)30297-7
DOI:
https://doi.org/10.1016/j.freeradbiomed.2019.12.001
FRB 14510
Reference:
To appear in: Free Radical Biology and Medicine
Received Date: 20 February 2019
Revised Date: 4 November 2019
Accepted Date: 1 December 2019
Please cite this article as: X.-Z. Bao, Q. Wang, X.-R. Ren, F. Dai, B. Zhou, A hydrogen peroxide-
activated Cu(II) pro-ionophore strategy for modifying naphthazarin as a promising anticancer agent with
high selectivity for generating ROS in HepG2 cells over in L02ꢀcellsF,ree Radical Biology and Medicine
(2020), doi: https://doi.org/10.1016/j.freeradbiomed.2019.12.001.
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.
© 2019 Published by Elsevier Inc.

1
2
A hydrogen peroxide-activated Cu(II) pro-ionophore strategy for
modifying naphthazarin as a promising anticancer agent with high
selectivity for generating ROS in HepG2 cells over in L02 cells
Xia-Zhen Bao1,2, Qi Wang1, Xiao-Rong Ren1, Fang Dai1, and Bo Zhou1,*
3
4
5
1State Key Laboratory of Applied Organic Chemistry, Lanzhou University, 222
2
6
Tianshui Street S., Lanzhou, Gansu 730000, China, College of Chemistry and
7
Chemical Engineering, Northwest Normal University, Lanzhou, Gansu 730070, China
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
__________________
*Corresponding author. Fax: (+86) 931-8915557
E-mail: bozhou@lzu.edu.cn
1

23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
ABSTRACT
Targeting redox vulnerability of cancer cells by pro-oxidants capable of generating
reactive oxygen species (ROS) has surfaced as an important anticancer strategy. Due
to the intrinsic narrow therapeutic window and other dangerous side effects of ROS
generation, it is highly needed and challenging to develop pro-oxidative anticancer
agents (PAAs) with high selectivity for generating ROS in cancer cells. Herein we
report a hydrogen peroxide (H2O2)-activated Cu(II) pro-ionophore strategy to develop
naphthazarin (Nap) as such type of PAAs based on the H2O2-mediated conversion of
boronate to free phenol. The boronate-protected Nap (PNap) can exploit increased
levels of H2O2 in HepG2 cells to in situ release Nap followed by its efflux via
conjugation with reduced glutathione (GSH), allowing that the Nap-GSH adduct
works as a Cu(II) ionophore to induce continuously GSH depletion via a
reduction-dependent releasing of Cu(I) by GSH. This strategy endows PNap with the
unprecedented ability to hit multi-redox characteristics (increased levels of H2O2,
GSH and copper) of HepG2 cells, leading to ROS generation preferentially in HepG2
cells along with their selective death.
Keywords: Hydrogen peroxide; reactive oxygen species; glutathione; copper;
ionophore
2

45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
1. Introduction
Cancer cells require higher levels of ROS to maintain their malignant phenotype
than normal cells, and thereby up-regulate GSH synthesis to cope with their intrinsic
oxidative stress. The aberrant redox homeostasis observed in cancer cells constitutes
their redox vulnerability, and has led to an appreciation that further ROS generation
or/and GSH depletion induced by pro-oxidants can preferentially trigger their death
by pushing their ROS levels beyond a critical “toxic threshold”, as detailed in a
plethora of elegant reviews [1,2] and research papers [3-9]. However, therapeutic
benefit of pro-oxidative anticancer agents (PAAs) is very limited due to the intrinsic
narrow therapeutic window of ROS generation. It is necessary to control accurately
ROS amount by which PAAs kill cancer cells while sparing normal cells (Scheme
1A). Yet even so, inevitable ROS generation in normal cells would trigger other
dangerous side effects, for example, potential induction of secondary tumors [10].
Thus, it is particularly attractive and exceedingly challenging to develop PAAs with
high selectivity for generating ROS or/and depleting GSH in cancer cells.
Apart from intrinsic oxidative stress, aberrant redox-active copper accumulation is
also a fundamental process that operates in a variety of malignancies, and is positively
correlated with cancer progression [11]. This prompts developing copper ionophores
as the killers of cancer cells in the past few years [12-18]. In this regard, we have
recently demonstrated that 3-hydroxyflavone (3-HF) function as a potent Cu(II)
ionophore in terms of its 3-hydroxy-4-keto group (a keto-enol moiety) via sequential
proton-loss Cu(II) chelation, GSH-driving releasing of Cu(I) and
3

67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
protonation-dependent efflux of the neutral ligand [19]. Among the three-step process,
GSH-driving releasing of Cu(I), that is the reduction of the 3-HF-Cu(II) complex by
GSH, is the most critical for its role as a Cu(II) ionophore-based PAA to induce death
of cancer cells because the reduction depletes significantly GSH to generate ROS,
leading to redox imbalance of cancer cells. Furthermore, considering the sequential
proton-loss Cu(II) chelation of 3-HF, we have used a 2,4-dinitrobenzenesulfonyl
group to mask its enolic hydroxyl group. Hereby we have designed a Cu(II)
pro-ionophore (PHF), which is capable of exploiting increased levels of GSH in
human lung adenocarcinoma A549 cells to in situ release 3-HF, resulting in a
remarkable improvement in killing A549 cells selectively [20].
As the most stable ROS, hydrogen peroxide (H2O2) is constantly produced by cells
to mediate signalling cascades involved in redox homeostasis [21]. Higher levels of
H2O2 have long been observed in cancer cells than in normal cells. For example, its
concentration in normal cells varies between 1 nM and 700 nM [22], but is
significantly increased to 7 µM in Jurkat T-cells [23]. These fact make H2O2 an ideal
target to develop the anticancer prodrugs [3, 24-29] based on the H2O2-mediated
conversion of boronate to free phenol which has been first used for developing
fluorescent probes to sense cellular H2O2 [30]. However, to the best of our knowledge,
this conversion strategy has not been used for designing the Cu(II)
pro-ionophore-based PAAs.
Natural products provide a structurally diverse skeleton resource to develop
anticancer drugs [31]. Naphthazarin (5,8-dihydroxy-1,4-naphthoquinone, Nap),
4

89
90
isolated from the roots of many comfrey, is a known anticancer agent [32,33]. This
molecule harbors two similar structural moieties to 3-hydroxy-4-keto group of 3-HF
(Scheme 1A), and thereby be a possible Cu(II) ionophore which works in a similar
fashion as 3-HF.
91
92
93
We hereby used a hydrogen peroxide-activated Cu(II) pro-ionophore strategy to
modify Nap where its two phenolic hydroxyl groups were masked by the boronate
(Scheme 1A). The boronate-protected Nap (PNap, Scheme 1B) expects to be capable
of exploiting increased levels of H2O2 in cancer cells to in situ release Nap together
with the formation of quinone methide (Scheme 1B). Both Nap and quinone methide
are the Michael acceptor-type electrophiles that can deplete intracellular GSH via a
rapid alkylation reaction. More importantly, the alkylation reaction promotes
effectively the efflux of Nap from cancer cells. Subsequently, just as 3-HF does [19],
the Nap-GSH adduct works as a Cu(II) ionophore to induce continuously GSH
depletion via reduction-dependent releasing of Cu(I) by GSH, leading to ROS
generation and final redox imbalance of cancer cells (Scheme 1B). Collectively, PNap
utilizes increased levels of H2O2 in cancer cells to realize the activation, increased
levels of GSH in cancer cells to promote the efflux, and increased levels of GSH in
cancer cells to facilitate releasing of Cu(I). The first two steps do not occur in the case
of PHF, a GSH-acivated Cu(II) pro-ionophore [20]. Thus, we reason that compared
with PHF, PNap might be a more promising PAA which generates ROS preferentially
in cancer cells over in normal cells based on its unprecedented ability to hit
multi-redox characteristics (increased levels of H2O2, GSH and copper) of cancer cells
5
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110

111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
(see Scheme 1B).
Scheme 1 here
2. Materials and methods
2.1. Materials
Roswell
Park
Memorial
Institute
(RPMI)-1640,
3-(4,
5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), bovine serum
albumin (BSA), 2’,7’-dichlorodihydrofluorescein diacetate (DCFH-DA), yeast
glutathione reductase (GR), 2-vinylpyridine (97%), reduced L-glutathione (GSH),
bathocuproinedisulfonic acid disodium salt (BCS), catalase from bovine liver (CAT),
5,5’-dithiobis-2-nitrobenzoic acid (DTNB), copper(II) sulfate pentahydrate and
Intracellular Hydrogen Peroxide Assay Kit (MAK164) were purchased from
Sigma-Aldrich (St. Louis, MO, USA). β-Nicotinamide adenine dinucleotide
2’-phosphate reduced tetrasodium salt (NADPH) was from Roche Diagnostics GmbH
(Mannheim, Germany). Bicinchoninic acid (BCA) protein assay kit was from
Beyotime Institute of Biotechnology (Jiangsu, China). Amplex™ Red Hydrogen
Peroxide/Peroxidase Assay Kit was purchased from Thermo Fisher Scientific
(Waltham, USA).
2.2. Synthesis of PNap
2.2.1. Synthesis of 2,3-dichloro-5,8-dihydroxynaphthalene-1,4-dione (3)
The mixture of p-dimethoxybenzene 1 (2.50 g, 18.1 mmol) and dichloromaleic
6

133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
anhydride 2 (6.05 g, 36.2 mmol) was added portionwise to the mixture of AlCl3
(21.35 g, 160.2 mmol) and NaCl (4.25 g, 72.6 mmol) which was melted by heating to
145 °C. After the reaction mixture was further heated to 180 °C and stirred for 2 min,
the dark red melt was allowed to cool to room temperature. The mixture was
hydrolyzed with water (227.5 mL) and concentrated HCl (15 mL) overnight. The pure
red precipitate of 3 was collected by filtration (52% yield). 1H and 13C NMR spectral
data of the isolated product matched with those reported in previous papers [34].
2.2.2. Synthesis of 5,8-dihydroxynaphthalene-1,4-dione (Nap)
A mixture of the 3 (1 mmol) and Fe powder (10 mmol) in glacial acetic acid (30 ml)
was refluxed for 1 h. After cooling to room temperature, the reaction mixture was
filtered and solvent was evaporated under reduced pressure. The pH value of residue
was adjusted to pH 8-9 with 5% NaOH as long as the mixture turned to deep blue.
Then, the reaction mixture was acidified by 10% HCl to pH 4-5, extracted with Et2O
and purified by column chromatography to obtain Nap (23% yield). 1H and 13C NMR
spectral data of the isolated product matched with those reported in previous papers
[35].
2.2.3. Synthesis of 5-hydroxy-8-((4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)
benzyl)oxy)naphthalene-1,4-dione (4)
Silver (I) oxide (23.7 mmol) was added to the solution of Nap (5.93 mmol) and
4-(bromomethyl)benzeneboronic acid pinacol ester (11.9 mmol) in chloroform (15
mL) followed by stirring the mixture at room temperature for 2 h. After the solution
was filtered through a Celite pad to remove silver salts, the solvent was removed at
7

155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
reduced pressure. The resultant orange oil was purified by flash chromatography to
afford 4 (63%) [36]. 1H NMR (400 MHz, CDCl3) δ 12.46 (s, 1H), 7.84 (d, J = 8.0 Hz,
2H), 7.52 (d, J = 8.0 Hz, 2H), 7.32 (d, J = 9.5 Hz, 1H), 7.22 (d, J = 9.5 Hz, 1H), 6.91
13
(s, 2H), 5.29 (s, 2H), 1.35 (s, 12H). C NMR (100 MHz, CDCl3) δ 190.49, 183.21,
156.92, 152.69, 141.68, 139.28, 136.43, 135.16, 126.48, 126.11, 125.87, 118.33,
114.85, 83.86, 71.62, 24.88. HRMS (ESI) m/z calcd for C23H23BO6 (M+Na):
429.1480. Found: 429.1487, error: 1.6 ppm. Melting Point: 182-184ºC.
2.2.4. Synthesis of PNap
1
PNap was prepared according to the synthesis procedure of 4 (yield 52%). H
NMR (400 MHz, CDCl3) δ 7.84 (d, J = 8.0 Hz, 4H), 7.52 (d, J = 8.0 Hz, 4H), 7.20 (s,
2H), 6.81 (s, 2H), 5.23 (s, 4H), 1.35 (s, 24H). 13C NMR (100 MHz, CDCl3) δ 184.56,
152.85, 139.47, 138.35, 135.12, 126.13, 122.57, 122.00, 83.82, 71.75, 24.87. HRMS
(ESI) m/z calcd for C36H40B2O8 (M+Na): 645.2802. Found: 645.2795, error: 1.1
ppm. Melting Point: 176-180ºC. Purity: 93.73% (CH3CN/H2O = 50/50, v/v, Rt =
13.15 min).
2.3. HPLC analysis for activation of PNap by H2O2
The sensing mechanism of PNap to H2O2 was studied by high pressure liquid
chromatography (HPLC)-based analysis on a Waters HPLC system equiped with a
Sun Fire™, C-18 column (4.6 mm × 150 mm). Specifically, PNap (1 mM) was
reacted with H2O2 (10 equiv.) for 6 h in CH3CN/H2O (v/v 1:1) and then the reaction
system was monitored by HPLC with a mobile phase of water/acetonitrile at a flow
8

177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
rate of 1 mL/min. The run was set as follows: for PNap, A(H2O): B(CH3CN) = 0.5:0.5
(13.15 min), for Nap A(H2O): B (CH3CN) = 0.5:0.5 (5.01 min), for H2O2 A(H2O):
B(CH3CN) = 0.5:0.5 (1.53 min).
2.4. Purity analysis for PNap
The purity analysis of PNap was studied by HPLC. The detail methods and run set
are the same as described in the section of 2.3.
2.5. MS detection for activation of PNap by H2O2
After reaction of PNap (1 mM) with H2O2 (10 equiv.) in CH3CN/H2O (v/v, 1:1) for
6 h, the products were analyzed on an electrospray ionization mass spectrometer
(TRACE DSQ, Thermo Electron Corporation).
2.6. Kinetic analysis for activation of PNap by H2O2
Aliquots of the reaction mixture of PNap (50 µM) with H2O2 (0.5, 1, 2 and 2.5 mM)
o
in CH3CN/H2O (v/v, 1/1) at 37 C within 1 h were taken out at appropriate time
intervals and subjected to kinetic analysis of the formation of Nap using HPLC with
the same mobile phase and flow rate as described in the section of 2.3.
2.7. UV/Vis spectral measurements
The Cu(II)-induced UV/Vis absorption changes of Nap in PBS buffer was
monitored at ambient temperature using a Cary 300 spectrophotometer. The spectra
9

199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
were run against blanks containing PBS and Cu(II), and recorded after addition of
Cu(II).
2.8. Monitoring the copper-chelating ability of the Nap-GSH adduct using HRMS
After reaction of Nap (1 mM) with GSH (1 mM) in CH3CN/H2O (v/v, 1:1) for 3 h
prior to addition of Cu(II) (1 mM), the reaction mixtures were monitored on a Thermo
Scientific Orbitrap Elite mass spectrometer by electrospray ionisation technique in a
negative mode.
2.9. Cell culture
Human hepatoma HepG2, human cervical carcinoma HeLa, human normal liver
L02 and human umbilical vein endothelial HUVEC cells were obtained from the
Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences.
Cells were maintained in RPMI-1640 medium with 10% fetal bovine serum (FBS),
100 kU/L penicillin, 100 kU/L streptomycin, and incubated at 37 °C in a humidified
atmosphere containing 5% CO2.
2.10. Cell viability assay
HepG2, HeLa, L02 and HUVEC cells were seeded at 4 × 104 cells in 96-well plates,
then treated with graded concentrations of Nap/PNap with or without CuSO4•5H2O
for 48 h. Subsequently, a fresh solution of MTT (0.5 mg/mL) was added to each well
and the plate was incubated for another 4 h at 37 °C, followed by the removal of the
10

221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
medium and the addition of 100 µL DMSO. The optical density (OD) of each well
was read at 570 nm using a microplate reader (Bio-Rad model 680). The cell viability
was expressed as the percentage of control.
2.11. Apoptosis analysis
HepG2 (4 × 105 cells/well) and L02 (4 × 105 cells/well) cells were cultured in
6-well culture plates and allowed to grow overnight. After replacement of medium,
the cells were treated with PNap (10 µM) and Cu(II) (10 µM) for 48 h at 37 °C. In the
case of inhibition, the cells were pretreated with BCS (400 µM) or CAT (0.5 mg/mL)
for 1 h before adding PNap and Cu(II). After treatment with Annexin V-FITC and
propidium iodide (BD Biosciences Pharmigen, San Diego, CA) for 15 min at room
temperature, the stained cells were analyzed by FACSCanto flow cytometer (Becton
Dickinson, San Jose, CA, USA). A total of 1.0 × 104 events were counted for each
sample.
2.12. Measurement of intracellular copper
Intracellular copper content was determined using Inductively Coupled Plasma
Mass Spectrometry (ICP-MS). HepG2 and L02 cells (2 ×106 cells/dish) were seeded
in 10 cm dishes and incubated for 24 h. Then the cells were treated with PNap (10 µM)
or Nap (10 µM) with or without 10 µM CuSO4•5H2O for 3, 6 and 9 h. In the case of
inhibition, the cells were pretreated with BCS (400 µM) or CAT (0.5 mg/mL) for 1 h
before adding the tested compounds. The cells were trypsinized, counted and washed
11

243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
three times with PBS. Subsequently, the cells were lysed in 1 mL 20% HNO3
overnight at 4 °C before the determination. The results are expressed as folds of the
control group.
2.13. Fluorescence imaging of intracellular Cu(I)
The levels of Cu(I) in living HepG2 and L02 cells were mapped out using a
previously reported NIR fluorescent Cu(I) probe [37]. HepG2 and L02 cells were
seeded at a density of 4×105 cells/well in six-well tissue culture plates and treated
with 10 µM PNap in the presence or absence of 10 µM CuSO4•5H2O for 9 h. Then the
medium was replaced followed by incubation with the Cu(I) probe (5 µM) for another
10 min at 37 °C. Subsequently, the cells were washed three times with PBS buffer.
The fluorescence images were acquired by microscope Leica DM 4000B (Leica
Microsystems CMS GmbH, Wetzlar, Germany).
2.14. Assay for intracellular GSH levels
The glutathione reductase-DTNB recycling assay [38] was used to monitor GSH
contents in HepG2 (4×105 cells/well) and L02 (4×105 cells/well) cells treated with
PNap with or without 10 µM CuSO4•5H2O for 9 h, and the related details were
provided in our previous paper [39].
2.15. Determination of intracellular ROS levels
ROS levels were detected using a redox-sensitive fluorescent probe DCFH-DA on a
12

265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
FACS Canto flow cytometer (Becton-Dickinson, San Jose, CA, USA). HepG2 and
L02 cells were treated with PNap with or without 10 µM CuSO4•5H2O for 9 h in the
absence and presence of pretreatment with BCS (400 µM) or CAT (0.5 mg/mL) for 1
h. Then the cells were collected and washed three times with PBS, followed by their
resuspension in PBS and incubation with 3 µM DCFH-DA for 30 min at 37 °C. The
DCF fluorescence was measured at 488 nm excitation and 525 nm emission.
2.16. Monitoring H2O2 released from HepG2 or L02 cells
After HepG2 (2 × 104 cells/well) and L02 (2 × 104 cells/well) cells were treated
with PNap (10 µM) or/and Cu(II) for 9 h in the absence and presence of 1 h
pretreatment with CAT (0.5 mg/mL) or BCS (400 µM), H2O2 levels in the culture
medium were monitored using the Amplex™ Red Hydrogen Peroxide/Peroxidase
Assay Kit [40]. The related experimental details were described in our previous paper
[41].
2.17. Assay for the intracellular basal H2O2 levels in HepG2 and L02 cells
Intracellular basal H2O2 levels were determined using the Intracellular Hydrogen
Peroxide Assay Kit (Sigma-Aldrich, MAK164) according to the manufacturer's
instructions [42]. Briefly, HepG2 and L02 cells were seeded onto 96-well plates at a
density of 2 × 104 cells per well overnight. Then the cells were washed with PBS,
incubated with 100 µL of the Fluorescent Peroxide Sensor working solution/well for
30 min and then subjected to fluorescence analysis on a TECAN Infnite M200
13

287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
multimode microplate reader (λex = 490 nm, λem = 525 nm). After the cells in the
96-well plate were collected with trypsin, washed twice with PBS and lysed with
RIPA lysate, the protein concentration was determined by using the BCA Protein
Assay Kit (Beyotime, Jiangsu, China).
2.18. Assay for activation of PNap to NaP in HepG2 and L02 cells
HepG2 and L02 cells were seeded in petri dishes at a density of 2 × 106 cells/dish
and incubated for 24 h. After treatment with PNap (20 µM) for 0.5, 1, 3, 6, 9 and 12 h,
the cells were trypsinized and washed with PBS, and then extracted with ice cold
methanol (300 µL/dish). The suspension was centrifuged for 10 min at 10000 rpm and
o
4 C, and then analyzed by a Waters HPLC system consisting of 1525 Binary HPLC
pump and 2998 photodiode array detector. A SunFireTM C18 column (4.6 × 150 mm)
was used with a mobile phase of water-acetonitrile at a flow rate of 1 mL/min.
2.19. Monitoring the adduct of Nap with GSH in HepG2 and L02 cells
HepG2 and L02 cells were seeded at a density of 2×106 cells/dish and incubated for
24 h. Then the cells were treated with PNap (20 µM) for 3 h, the cells were
trypsinized, washed with PBS, and extracted with ice cold methanol (300 µL/dish).
o
The suspension was centrifuged for 10 min at 10000 rpm and 4 C, and the
supernatant was subjected to analysis in Matrix-Assisted Laser Desorption-Ionization
(MALDI) Mass Spectrometry (MALDI-7090, SHIMADZU) in a negative ion mode
with N-(1-naphthyl)ethylenediamine dihydrochloride (NEDC) as a matrix [43]. The
14

309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
matrix solution was composed of 5mg/ml NEDC, 80% methanol, 0.3% NH4OH and
19.7% water.
2.20. Statistical analysis
Data were expressed as the mean ± SD of the results obtained from at least three
independent experiments. Significant differences (P<0.05) between the means of two
groups were analyzed by Student's t-test.
3. Results and discussion
3.1. Synthesis of PNap
Nap was readily prepared by acylation of 1,4-dimethoxybenzene using
2,3-dichloromaleic anhydride followed by a reduction reaction (Scheme 2). Step by
step nucleophilic substitution between Nap and 4-bromomethylbenzeneboronic acid
pinacol ester afford the desired PNap.
Scheme 2 here
3.2. In vitro activation of PNap to Nap by H2O2
We evaluated initially the in vitro activation of PNap to Nap by H2O2 at ambient
temperature using HPLC and MS-based analyses. Upon addition of 10 equiv. H2O2,
the chromatographic peak of PNap disappeared after 6 h together with appearance of
a new peak at 5.02 min, corresponding to the production of NaP (Figure 1A). From
the MS spectrum recorded 6 h after mixing PNap with H2O2, we also observed the
15

331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
main molecular peak (m/z = 190) assigned to Nap (Figure 1B).
After checking the reactivity of PNap with H2O2, to further clarify its sensitivity to
H2O2 concentrations we measured the reaction rate constant in CH3CN/H2O (v/v, 1/1)
at 37oC by using HPLC-based analyses. We observed that under the condition of
large excess H2O2, the formation of Nap is pseudo-first-order (Figure S1). Plotting
the pseudo-first-order rate constants (kobs) against H2O2 concentrations gave a
straight line, from which the second-order rate constant (k2) could be obtained as
20.06 ± 1.86 M-1 s-1 (Figure 1C). This indicates that although each PNap molecule
reacts subsequently with two H2O2 molecules, the elementary reaction kinetics is
second-order.
Figure 1 here
3.3. Chemical mechanisms of Nap as a Cu(II) ionophore
To rationalize the chemical mechanisms by which Nap works as a Cu(II) ionophore,
we next monitored the UV-visible spectral changes of Nap induced by Cu(II). As
shown in Figure 2A, Nap in phosphate-buffered saline (PBS, pH = 7.4) displayed a
maximum absorbance at 518 nm along with three shoulders in the λ range of 470 -
500, 540 - 580 and 600 - 640 nm (line 1), but addition of Cu(II) induced appearance
of a blue-shifted peak centered at 484 nm (line 2). This blue-shifted peak is
characteristic for the production of the Cu(II) complex with Nap because EDTA (a
well-known chelating agent) made a return of the peak to initial spectrum of Nap (line
3). In contrast, PNap exhibited a broad absorption at 449 nm (line 4) but addition of
16

353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
Cu(II) showed no any effect on the absorption spectrum (line 5), suggesting its
inability to chelate Cu(II).
Figure 2 here
A comparison of line 1 with line 2 in Figure 2B indicates clearly that introduction
of sulfuric acid inhibited the partial ionization of Nap in PBS (pH = 7.4) (pKa = 8.89
[33]). Furthermore, the spectrum (line 2) obtained by Nap plus sulfuric acid was
consistent with that (line 3) of Nap in the presence of Cu(II) and sulfuric acid,
highlighting that under strong acidic conditions, the Cu(II) chelation of Nap was
completely inhibited. In other words, the Cu(II) chelation depends on deprotonation
of its phenolic hydroxyl groups, a sequential proton-loss Cu(II) chelation mechanism
as described previously [19]. Addition of GSH to the solution of Nap plus Cu(II)
triggered disappearance of the blue-shifted peak accompanied by development of
three absorbance peaks at 539, 575 and 616 nm (line 4), in tune with the spectrum of
Nap plus GSH (line 5). The above results reveals a reduction-dependent release of
copper from the Cu(II) complex with Nap by GSH (see also below) followed by the
formation of the Nap-GSH adduct. When extra Cu(II) was further introduced into the
mixed solution of Nap, Cu(II) and GSH, two of three absorbance peaks of Nap plus
Cu(II) were smoothed (line 6), but re-appeared upon further addition of extra GSH
(line 7), suggesting that the Nap-GSH adduct can effectively chelate and release
copper in a similar fashion as Nap (Scheme 1B).
Furthermore, we explored the copper-chelating ability of the Nap-GSH adduct by
HRMS in a negative mode. Upon mixing equal equivalents of Nap and GSH, we
17

375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
observed the molecular peak (m/z = 494.0894 [M-H]+), corresponding to the
formation of the Nap-GSH adduct (Figure 3A). When we introduced Cu(II) into the
solution, the expected [M-2H]+, [M-1H]+ and [M]+ at 1050.0962, 1051.0992 and
1052.0914 mass units were detected (Figure 3B), representing the complexation of
Cu(II) with two molecules of the Nap-GSH adduct after their deprotonation (see
Scheme 1B), which is determined by a sequential proton-loss Cu(II) chelation
mechanism [19].
Figure 3 here
3.4. Synergistic and selective killing of cancer cells by PNap plus Cu(II)
With the chemical data of Nap and PNap in hand, we subsequently compared their
cytotoxicity in the absence and presence of Cu(II) against either human hepatoma
HepG2 or human normal liver L02 cells. Neither PNap (from 1 to 10 µM) nor Cu(II)
(10 µM) alone showed any cytotoxicity against either HepG2 or L02 cells (Figure 4A).
However, PNap plus Cu(II) killed synergistically HepG2 cells while remaining L02
cells unharmed (Figure 4A and B). The synergistic and selective killing of HepG2
cells by PNap plus Cu(II) was further pointed out by combination index (CI) (Table 1).
In the case of 10 µM PNap and 10 µM Cu(II), for example, the CI values for HepG2
and L02 cells are 0.003 and 0.474, respectively. The synergy and selectivity was also
observed in another group of cancer and normal cells (human cervical carcinoma
HeLa cells and human umbilical vein endothelial HUVEC cells, Figure 4A and B). As
expected, the synergistic ability of PNap with Cu(II) in killing HepG2 cells is superior
18

397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
to PHF, a GSH-acivated Cu(II) pro-ionophore [20] (Figure 4C). In a sharp contrast,
Nap plus Cu(II) killed synergistically HepG2 and HeLa cells but exhibited weaker
selectivity in killing them over L02 and HUVEC cells than PNap plus Cu(II) did
(Figure 4A, B, D and E).
Figure 4 here
Furthermore, the synergistic killing of HepG2 cells by PNap plus Cu(II) was
significantly revised by adding bathocuproine disulfonic acid disodium salt (BCS, a
copper chelator) or catalase (CAT, a specific scavenger of H2O2) (Figure 4F),
supporting that PNap works as a Cu(II) pro-ionophore to induce redox imbalance of
HepG2 cells based on its activation by intracellular H2O2 and alkylation by
intracellular GSH, as shown in Scheme 1B. It should be pointed out that as CAT is
cell-impermeable, its inhibitory effects could be explained by scavenging H2O2
diffused from inside cells to decrease intracellular H2O2, as described in our previous
paper [41,44] (see below).
To gain an insight into the mechanism determining the synergistic and selective
killing of HepG2 cells over L02 cells by PNap plus Cu(II), we carried out apoptosis
analysis in the two cell lines. Parallel to the cytotoxic results, PNap plus Cu(II)
triggered synergistically obvious apoptosis of HepG2 cells but not that of L02cells,
and the apoptosis was revised significantly by adding either BCS or CAT (Figure 5).
Figure 5 here
3.5. Identification of PNap as a Cu(II) ionophore-type PAA after its activation by
19

419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
H2O2 and alkylation by GSH
To further clarify the role of PNap as a Cu(II) ionophore after its activation and
alkylation by intracellular H2O2 and GSH, respectively, we measured the copper
accumulation induced by 10 µM PNap plus 10 µM Cu(II) by using ICP-MS. As
expected, in tune with selective killing of HepG2 over L02 cells by PNap and Cu(II),
they induced synergistically the copper accumulation in HepG2 cells in a
time-dependent fashion but failed to do so in L02 cells (Figure 6A). Specifically, the
copper accumulation in HepG2 cells induced by PNap in the presence of Cu(II) at the
time-point of 9 h was increased by 4.6-fold relative to the Cu(II) alone-treated group.
Also in line with the cytotoxic results, the copper accumulation in HepG2 cells was
significantly revised by adding either BCS or CAT (Figure 6A). In contrast, in the
case of Nap and Cu(II), the preferential copper accumulation in HepG2 cells over in
L02 cells was significantly weakened (Figure 6B). Besides, Nap induced the copper
accumulation in HepG2 cells, but to a lesser extent than PNap did (Figure 6A and B),
due to the fact that 10 µM Nap alone is cytotoxic against HepG2 cells (see Figure 4D),
leading to their partial death under the current experimental conditions. These above
results highlights that the boronate-protected strategy decreases the cytotoxicity of the
parent Nap but also improve effectively its selectivity as a Cu(II) ionophore.
Furthermore, the preferential copper accumulation in HepG2 cells over in L02 cells
by PNap plus Cu(II) was specified to the Cu(I) accumulation using a NIR fluorescent
Cu(I) probe [37] (Figure 6C and D), strongly supporting that after activation of PNap
to Nap by intracellular H2O2, the Nap-GSH adduct works as a Cu(II) ionophore via
20

441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
reduction-dependent releasing of Cu(I) by GSH (see also above).
Figure 6 here
We were interested in investigating whether the selective killing of HepG2 cells
over L02 cells by PNap plus Cu(II) is the result of inducing preferentially redox
imbalance in HepG2 cells over in L02 cells. The redox homeostasis can be mirrored
by ROS increase coupled with GSH decrease. To do this, we measured intracellular
ROS and GSH levels. A prominent feature of our data is that fully consistent with our
design expectation, the treatment with PNap together with Cu(II) promotes efficiently
ROS generation (as indicated by oxidation of DCFH-CA, Figure 7A) and GSH
depletion (Figure 7B) in HepG2 cells rather than in normal L02 cells. Additionally,
the ROS generation and GSH depletion in HepG2 cells were significantly revised by
pretreatment with either BCS or CAT, a result correlated well with their ability to
revise the cytotoxicity induced by PNap plus Cu(II). The generating ROS include at
least H2O2 because CAT revised the DCF fluorescence intensity (Figure 7A). The
revised effect of CAT reminds us of the fact that at least a fraction of H2O2 is released
into extracellular space. Consequently, we further performed the Amplex Red assay
[40] to monitor the H2O2 release. As shown in Figure 7C, PNap plus Cu(II) induced
the H2O2 release more potently in HepG2 cells than in L-02 cells, and the release was
largely abrogated by either BCS or CAT. Taken together, we can conclude that after its
activation and alkylation by intracellular H2O2 and GSH, respectively, PNap works as
a Cu(II) ionophore-type PAA with the high selectivity for generating ROS and
depleting GSH in HepG2 cells over in L02 cells (Figures 6 and 7).
21

463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
Figure 7 here
3.6. Clarifying the reasons why PNap as a Cu(II) pro-ionophore can kill selectively
HepG2 cells over L02 cells
The last question considered here is to clarify how PNap, as a Cu(II) pro-ionophore,
can kill selectively HepG2 cells over L02 cells, we thus compare the basal H2O2 levels
in the two cell lines by using the Intracellular Hydrogen Peroxide Assay Kit
(Sigma-Aldrich, MAK164). The basal H2O2 levels in HepG2 and L02 cells were
determined as 0.135 ± 0.012 and 0.038 ± 0.004 µM/mg protein, respectively. We
compared further the activation difference of PNap to NaP by H2O2 between the two
cell lines after the intracellular uptake of PNap. It can be seen from Figure 8 that the
activation was faster in HepG2 cells than in L02 cells. Specifically, the concentration
of Nap reached a maximum after incubation of HepG2 and L02 cells with PNap for 1
and 3 h, respectively. The activation difference (3-fold) between HepG2 and L02 cells
correlates well with their basal H2O2 level difference (3.6 fold), supporting that the
uptake of PNap could be similar in the two cell lines. Additionally, we note that the
PNaps is converted into Nap at a rate several orders of magnitude faster in cells than
in homogenous solutions (Figure 1C) in view of the intracellular basal H2O2 levels
being far lower than the concentrations of H2O2 (0.5-2.5 mM) used in the kinetic assay.
This can be due to the involvement of cellular peroxidases in the activation, and the
fact that the activation in cells as heterogeneous media is not parallel to that in
homogenous solutions. More importantly, the activation difference between in the
cells and in homogenous solutions hits that PNap is oxidized/activated intracellularly
22

485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
by intracellular H2O2 rather than extracellularly by H2O2 diffused from the cells.
Considering that GSH-driving releasing of Cu(I) is a key step for PNap plus Cu(II)
to trigger redox imbalance including ROS generation and GSH depletion (see above,
especially Figure 6C and D), and that the Nap-GSH adduct works as a Cu(II)
ionophore, the difference of basal GSH levels between HepG2 and L02 cells should
be another important factor to determine their differential sensitivity to PNap plus
Cu(II). We determined previously the basal levels of GSH in HepG2 and L02 cells
being 115.3 ± 5.4 and 29.1 ± 3.2 nmol/mg protein by using the glutathione
reductase-DTNB recycling assay, respectively [20]. Furthermore, we monitored the
generation of the Nap-GSH adduct in cellular experiments by MALDI-MS in a
positive ion mode (Figure 9). As expected, upon incubation of HepG2 cells with PNap
for 3 h, from the cell lysis solutions we observed the adduct [M]+and [M+1H]+ at
495.43 and 496.48 mass units. In contrast, the abundance of the adduct peaks was
lower in L02 cells than in HepG2 cells (Figure 9). Consequently, we can conclude that
the almost 4-fold difference in the basal levels of GSH, coupled with the 3.6-fold
difference in the basal levels of H2O2, between HepG2 and L02 cells should be
enough for PNap plus Cu(II) to deplete GSH and generate ROS selectively in HepG2
cells over in L02 cells, leading to selective death of HepG2 cells via redox imbalance.
The unique ability of PNap plus Cu(II) to kill selectively HepG2 cells over L02
cells warrants future investigation into the efficacy of PNap either alone or in
combination with exogenous Cu(II) in the experiments of nude mice inoculated with
HepG2 cells. In spite of free copper being maintained at very low levels in tumor
23

507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
microenvironment or inside cells [11, 45], bound copper is abundant in its carriers
such as ceruloplasmin in human plasma and redox-active enzymes inside cells.
However, the copper bound to these carriers has been elucidated to be the less tightly
bound and is thereby exchangeable [46]. Thus, in real tumors, PNap is expected to act
by either mobilizing the bound copper or in combination with exogenous Cu(II).
In summary, this work describe a promising H2O2-activated Cu(II) pro-ionophore
strategy to modify Nap as a potent PAA which is capable of depleting GSH and
generating ROS preferentially in HepG2 cells over in L02 cells based on its
unprecedented ability to hit multi-redox characteristics (including increased levels of
H2O2, GSH and copper) of HepG2 cells.
Acknowledgments
This work was supported by the National Natural Science Foundation of China
(Grant No. 21672091) and the111 Project.
Appendix A. Supplementary material
Supplementary data associated with this article can be found in the online version
at http://dx.doi.org/10.1016/j.freeradbiomed.
24

529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
References
[1] D. Trachootham, J. Alexandre, P. Huang, Targeting cancer cells by
ROS-mediated mechanisms: a radical therapeutic approach?, Nat. Rev. Drug
Discov. 8 (2009) 579-591.
[2] C. Gorrini, I. S. Harris, T. W. Mak, Modulation of oxidative stress as an
anticancer strategy, Nat. Rev. Drug Discov. 12 (2013) 931-947.
[3] S. Daum, M. S. V. Reshetnikov, M. Sisa, T. Dumych, M. D. Lootsik, R. Bilyy, E.
Bila, C. Janko, C. Alexiou, M. Herrmann, L. Sellner, A. Mokhir,
Lysosome-targeting amplifiers of reactive oxygen species as anticancer prodrugs,
Angew. Chem. Int. Edit. 56 (2017) 15545-15549.
[4] J. Yun, E. Mullarky, C. Lu, K. N. Bosch, A. Kavalier, K. Rivera, J. Roper, I. I. C.
Chio, E. G. Giannopoulou, C. Rago, A. Muley, J. M. Asara, J. Paik, O. Elemento,
Z. Chen, D. J. Pappin, L. E. Dow, N. Papadopoulos, S. S. Gross, L. C. Cantley,
Vitamin C selectively kills KRAS and BRAF mutant colorectal cancer cells by
targeting GAPDH, Science 350 (2015) 1391-1396.
[5] E. Piskounova, M. Agathocleous, M. M. Murphy, Z. Hu, S. E. Huddlestun, Z.
Zhao, A. M. Leitch, T. M. Johnson, R. J. DeBerardinis, S. J. Morrison, Oxidative
stress inhibits distant metastasis by human melanoma cells, Nature 527 (2015)
186-191.
[6] J. Noh, B. Kwon, E. Han, M. Park, W. Yang, W. Cho, W. Yoo, G. Khang, D. Lee,
Amplification of oxidative stress by a dual stimuli-responsive hybrid drug
enhances cancer cell death, Nat. Commun. 6 (2015) 6907-6916.
25

551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
[7] L. Raj, T. Ide, A. U. Gurkar, M. Foley, M. Schenone, X. Li, N. J. Tolliday, T. R.
Golub, S. A. Carr, A. F. Shamji, A. M. Stern, A. Mandinova, S. L. Schreiber, S.
W. Lee, Selective killing of cancer cells by a small molecule targeting the stress
response to ROS, Nature 475 (2011) 231-234.
[8] J. S. Bair, R. Palchaudhuri, P. J. Hergenrother, Chemistry and Biology of
Deoxynyboquinone, a Potent Inducer of Cancer Cell Death, J. Am. Chem. Soc.
132 (2010) 5469-5478.
[9] D. Trachootham, Y. Zhou, H. Zhang, Y. Demizu, Z. Chen, H. Pelicano, P. J.
Chiao, G. Achanta, R. B. Arlinghaus, J. Liu, P. Huang, Selective killing of
oncogenically transformed cells through a ROS-mediated mechanism by
β-phenylethyl isothiocyanate, Cancer Cell 10 (2006) 241-252.
[10] J. R. Totter, Spontaneous cancer and its possible relationship to oxygen
metabolism, Proc. Natl. Acad. Sci. U. S. A. 77 (1980) 1763-1767.
[11] A. Gupte, R. J. Mumper, Elevated copper and oxidative stress in cancer cells as a
target for cancer treatment, Cancer Treat. Rev. 35 (2009) 32-46.
[12] M. E. Helsel, K. J. Franz, Pharmacological activity of metal binding agents that
alter copper bioavailability, Dalton Trans. 44 (2015) 8760-8770.
[13] I. Alfonso, R. Quesada, Biological activity of synthetic ionophores: ion
transporters as prospective drugs?, Chem. Sci. 4 (2013) 3009-3019.
[14] J. L. Allensworth, M. K. Evans, F. Bertucci, A. J. Aldrich, R. A. Festa, P. Finetti,
N. T. Ueno, R. Safi, D. P. McDonnell, D. J. Thiele, S. Van Laere, G. R. Devi,
Disulfiram (DSF) acts as a copper ionophore to induce copper-dependent
26

573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
oxidative stress and mediate anti-tumor efficacy in inflammatory breast cancer,
Mol. Oncol. 9 (2015) 1155-1168.
[15] M. A. Cater, H. B. Pearson, K. Wolyniec, P. Klaver, M. Bilandzic, B. M. Paterson,
A. I. Bush, P. O. Humbert, S. La Fontaine, P. S. Donnelly, Y. Haupt, Increasing
intracellular bioavailable copper selectively targets prostate cancer cells, ACS
Chem. Biol. 8 (2013) 1621-1631.
[16] S. Tardito, A. Barilli, I. Bassanetti, M. Tegoni, O. Bussolati, R. Franchi-Gazzola,
C. Mucchino, L. Marchiò, Copper-dependent cytotoxicity of 8-hydroxyquinoline
Dderivatives correlates with their hydrophobicity and does not require caspase
activation, J. Med. Chem. 55 (2012) 10448-10459.
[17] M. Nagai, N. H. Vo, L. Shin Ogawa, D. Chimmanamada, T. Inoue, J. Chu, B. C.
Beaudette-Zlatanova, R. Lu, R. K. Blackman, J. Barsoum, K. Koya, Y. Wada,
The oncology drug elesclomol selectively transports copper to the mitochondria
to induce oxidative stress in cancer cells, Free Radic. Biol. Med. 52 (2012)
2142-2150.
[18] S. Tardito, I. Bassanetti, C. Bignardi, L. Elviri, M. Tegoni, C. Mucchino, O.
Bussolati, R. Franchi-Gazzola, L. Marchiò, Copper binding agents acting as
copper ionophores lead to caspase inhibition and paraptotic cell death in human
cancer cells, J. Am. Chem. Soc. 133 (2011) 6235-6242.
[19] F. Dai, W.-J. Yan, Y.-T. Du, X.-Z. Bao, X.-Z. Li, B. Zhou, Structural basis,
chemical driving forces and biological implications of flavones as Cu(II)
ionophores, Free Radic. Biol. Med. 108 (2017) 554-563.
27

595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
[20] X.-Z. Bao, F. Dai, X.-R. Li, B. Zhou, Targeting redox vulnerability of cancer
cells by prooxidative intervention of a glutathione-activated Cu(II)
pro-ionophore: Hitting three birds with one stone, Free Radic. Biol. Med. 124
(2018) 342-352.
[21] E. A. Veal, A. M. Day, B. A. Morgan, Hydrogen peroxide sensing and signaling,
Mol. Cell 26 (2007) 1-14.
[22] J. R. Stone, An assessment of proposed mechanisms for sensing hydrogen
peroxide in mammalian systems, Arch. Biochem. Biophys. 422 (2004) 119-124.
[23] F. Antunes, E. Cadenas, Estimation of H2O2 gradients across biomembranes,
FEBS Lett. 475 (2000) 121-126.
[24] V. Reshetnikov, S. Daum, A. Mokhir, Cancer-specific, intracellular, reductive
activation of anticancer PtIV prodrugs, Chem. Eur. J. 23 (2017) 5678-5681.
[25] Y. Kuang, K. Balakrishnan, V. Gandhi, X. Peng, Hydrogen peroxide inducible
DNA cross-linking agents: targeted anticancer prodrugs, J. Am. Chem. Soc. 133
(2011) 19278-19281.
[26] J. L. Major Jourden, S. M. Cohen, Hydrogen peroxide activated matrix
metalloproteinase inhibitors: a prodrug approach, Angew. Chem. Int. Edit. 49
(2010) 6795-6797.
[27] S. Daum, V. F. Chekhun, I. N. Todor, N. Y. Lukianova, Y. V. Shvets, L. Sellner, K.
Putzker, J. Lewis, T. Zenz, I. A. M. de Graaf, G. M. M. Groothuis, A. Casini, O.
Zozulia,
F.
Hampel,
A.
Mokhir,
Improved
synthesis
of
N-benzylaminoferrocene-based prodrugs and evaluation of their toxicity and
28