Multimodal therapies: Glucose oxidase-triggered tumor starvation-induced synergism with enhanced chemodynamic therapy and chemotherapy

Article abstract of DOI:10.1039/c9nj05469c

A tumor microenvironment is distinct from normal tissue cells in characteristic physiochemical conditions, based on which we can design tumor-specific therapy modalities. Herein, we introduce a concept of multimodal therapies, which integrates the characteristics of each therapy modality for efficient tumor therapy: tumor starvation-triggered synergism with enhanced chemodynamic therapy and activated chemotherapy. Fe₃O₄ nanoparticles (Fenton reaction catalysts) and a hypoxic prodrug tirapazamine (TPZ) were loaded in mesoporous silica nanoparticles (MSN) and GOX was grafted onto its surface, which was designed and fabricated for sequential multimodal therapies. Logically, glucose oxidase (GOX) deprived tumor cells of nutrients (glucose and oxygen) for starvation therapy and tumorous abnormality amplifications (elevated acidity, exacerbated hypoxia, and increased H₂O₂) were amplified by the GOX-driven oxidation reaction simultaneously. Specifically, elevated acidity could accelerate the release of iron ions and enhanced Fenton reaction efficiency. Associated with increased H₂O₂, an elevated ROS level was detected, which enhanced the chemodynamic therapy. Exacerbated hypoxia activated the hypoxic prodrug TPZ for tumor-specific chemotherapy programmatically. Particularly, via integrating starvation therapy, enhanced chemodynamic therapy, and activated chemotherapy, the sequential multimodal therapies were specifically designed for the tumor microenvironment and achieved effective abnormality amplifications and high therapeutic efficacy.

Full text of DOI:10.1039/c9nj05469c

View Article Online
View Journal
NJC
New Journal of Chemistry
Accepted Manuscript
A journal for new directions in chemistry
This article can be cited before page numbers have been issued, to do this please use: B. Sun, K. Cheng,
C. X. Ling, D. Gu, Z. Gao, Y. Li, P. An, Y. Zhang, C. You and R. Zhang, New J. Chem., 2020, DOI:
10.1039/C9NJ05469C.
Volume 42
This is an Accepted Manuscript, which has been through the
Number
6
21 March 2018
Pages 3963-4776
Royal Society of Chemistry peer review process and has been
accepted for publication.
NJC
New Journal of Chemistry
rsc.li/njc
A journal for new directions in chemistry
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1144-2546
PAPER
Chechia Hu, Tzu-Jen Lin et al.
Yellowish and blue luminescent graphene oxide quantum dots
prepared via a microwave-assisted hydrothermal route using
H2O2 and KMnO4 as oxidizing agents
rsc.li/njc

Page 1 of 32
New Journal of Chemistry
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
Multimodal therapies: Glucose Oxidase-triggered Tumor
Starvation Induced Synergism with Enhanced
Chemodynamic Therapy and Chemotherapy
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Kaiwu Chenga, Chuxuan Linga, Dihai Gua, Zhiguo Gaoa, Yaojia Lia, Peijing Ana, Yu
Zhanga, Chaoqun Youc, Rui Zhangb, *, Baiwang Suna, *.
a
School of Chemistry and Chemical Engineering, Southeast University, Nanjing
210089, PR China. Fax: +86 25 52090614, Tel: +86 25 52090614, *E-mail address:
chmsunbw@seu.edu.cn
b
Department of Ophthalmology, Zhongda Hospital, Southeast University, Nanjing
210009, China.
c College of Chemical Engineering, Nanjing Forestry University, Nanjing 210037, PR
China
Abstract
Tumor microenvironment are distinct from normal tissue cells in characteristic
physiochemical conditions, based on which we could design tumor-specific therapy
modality. Herein, we introduce a concept of multimodal therapies, which integrates the
characteristics of each therapy modality, for efficient tumor therapy: tumor starvation
triggered synergism with enhanced chemodynamic therapy and activated chemotherapy.
Fe3O4 nanoparticles (Fenton reaction catalyst) and a hypoxic prodrug of tirapazamine
(TPZ) were loaded in mesoporous silica nanoparticles (MSN) and GOX was grafted
onto its surface, which was designed and fabricated for the sequential multimodal
therapies. Logically, Glucose oxidase (GOX) deprives tumor cells of nutrients (glucose
and oxygen) for starvation therapy and tumorous abnormality amplifications (elevated
acidity, exacerbated hypoxia and increased H2O2) was amplified by the GOX-driven
oxidation reaction simultaneously. Specifically, elevated acidity could accelerate the
release of iron ion and enhanced Fenton reaction efficiency. Associated with the

New Journal of Chemistry
Page 2 of 32
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
increased H2O2, elevated ROS level was detected that enhanced chemodynamic therapy.
Exacerbated hypoxia activated the hypoxic prodrug (TPZ) for tumor-specific
chemotherapy programmatically. Particularly, integrated starvation therapy, enhanced
chemodynamic therapy and activated chemotherapy, the sequential multimodal
therapies was specifically designed for the tumor microenvironment and achieved
effective abnormality amplifications and high therapeutic efficacy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Key words: multimodal therapies, Glucose oxidase, chemodynamic therapy, Fenton
reaction, starvation therapy
Introduction
Cancer remains the most fatal diseases in the world and conventional therapeutic
modalities including toxic chemotherapeutics, invasive surgical treatment and external
high-energy irradiations merely achieve limited results with severe adverse effects
inevitably1–3. Monotherapy is difficult to achieve the desired therapeutic effect due to
multiple pathways in tumor pathogenesis. Therefore, Multimodal therapies integrateing
the characteristics of each therapy modality has been proposed currently for efficient
tumor therapy, which promotes the synergy of various monotherapy4–6. Particularly,
tumor microenvironment is different from normal tissue in pH value, redox elevation,
hypoxia, enzyme overexpression7. Nevertheless, these abnormalities are inadequate for
cancer cell with tumor-specific therapy effectively. Therefore, therapy modality
designed for expanding and responding specifically to tumor heterogeneity would
reduce tumor resistance while reducing toxic side effects on normal tissues. Reactive
oxygen species (ROS) including superoxide anion radicals (O2 ), singlet oxygen (1O2)
-•
and hydroxyl radicals (•OH) are essential in normal metabolic activity for normal cells
at low levels8–11. While high level ROS could destroy cells by oxidizing intracellular
glutathione, DNA, protein and lipid12–14. Chemodynamic therapy (CDT) was known
for its unique way of generating ROS independent of local oxygen in recent years.

Page 3 of 32
New Journal of Chemistry
1
2
3
4
5
View Article Online
DOI: 10.1039/C9NJ05469C
Produced by endogenous H2O2 through a metal ion-mediated Fenton reaction15, •OH
6
7
8
9
have received the most extensive attention for its powerful oxidizing capacity, resulting
in fatally damage to tumor cells16. Although chemodynamic therapy exhibits
advantages over traditional therapy models such as nonspecific side effects of extensive
chemotherapy and low efficiency (oxygen dependence and restricted light penetration)
of photodynamic therapy17. Regrettably, CDT depends on high efficiency catalyst and
highly concentrated H2O2 in vivo.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Nowadays, various transition metal ions like Cu(II) and Mn(II) have been applied as
catalyst to initiate a Fenton-like reaction for CDT18, while the most widely used is still
Fe(II), commonly coming from Ferromagnetic nanoparticles (Fe3O4 NPs)19,20. Shi et al.
introduced Fe3O4 nanoparticles into large pore-sized and biodegradable dendritic silica
nanoparticles to catalyze intracellular H2O2 for high level of ROS15. H2O2 concentration
is intrinsically higher than normal physiological environment at tumor tissue, yet still
insufficient for highly efficient and specific CDT21,22. In order to obtain adequate H2O2,
strategies like introducing exogenous H2O2 or stimulating the production of
endogenous H2O2 were put forward in response to the issue of insufficient endogenous
H2O2. Glucose oxidase (GOX, natural enzyme catalyst) has been introduced to tumor
therapy by triggering starvation therapy through consuming oxygen and β-D-glucose
to generate gluconic acid and H2O26,23. Simultaneously, due to the distinct metabolic
process, tumor cells demand more glucose for glycolysis to support its endless
proliferate, which would provide abundant H2O2 and acidity4, resulting in exacerbated
hypoxia at the same time24. Most importantly, tumorous abnormality and heterogeneity
amplifications create opportunities for sensitized chemotherapy especially for a
hypoxic-responsive tirapazamine (TPZ)25, which would be fatal to anoxic tumor cells,
but safe for oxygen-sufficient normal cells26. Chemotherapy usually could not achieve
desired therapeutic effect due to multidrug resistance (MDR) effect of cancer27. And
nonspecific and inaccurate injure to normal cells, poor solubility and rapid blood
clearance also restrict its further development in clinical research28–30
.

New Journal of Chemistry
Page 4 of 32
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
5
6
7
8
Aiming at above challenges, ‘intelligent’ drug delivery systems (DDSs) have been
designed for delivering biocompatible anticancer drugs carrier to tumor sites31,32
,
9
among which mesoporous silica nanoparticles (MSNs) have got far-reaching attention
and applications owing to its outstanding properties including well biocompatibility7,
highly drug loading capacity, tunable particle size, well-defined mesoporous and
especially, easily functionalized surface33–35. However, the multiple pathways and
heterogeneity involved in the pathogenesis of the tumor restrict the therapeutic effect
of monotherapy36. In this paper, multimodal therapies (tumor starvation therapy
triggered synergism with chemodynamic therapy and sensitized chemotherapy) was
reported for the efficient oncotherapy and summary synthesis of TPZ/Fe3O4@MSN-
GOX was shown in Scheme 1. Hypoxic-responsive Tirapazamine (TPZ) and high
efficiency catalyst Fe3O4 were enveloped in biocompatible MSN with GOX grafted on
its surface. Logically, GOX deprived tumor cells of nutrition (oxygen and glucose) for
starvation therapy and induced tumorous heterogeneity amplifications4 (increased H2O2,
decreased acidity and anabatic hypoxia), which activated the sensitized chemotherapy
under hypoxia and triggered chemodynamic therapy with the participation of Fe3O4 and
GOX-metabolite (H2O2 and H+) simultaneously. Meanwhile, external GOX also served
as a guard to protect the premature release of TPZ. Fe3O4 nanoparticles was translated
into Fe2+/Fe3+ in acid environment, which catalyzed abundant intracellular H2O2 into
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
highly toxic hydroxyl radical (•OH) for highly efficient CDT. Synergetic multimodal
therapies achieved satisfactory antitumor effect in vitro.

Page 5 of 32
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Schematic illustration of the synthesis of TPZ/Fe3O4@MSN-GOX and GOX
triggerred hunger treatment, sensitized chemotherapy and chemodynamic therapy for
efficient multimodal tumor treatment.
2. Materials and methods
2.1. Materials
Tetraethylorthosilicate (TEOS), ammonia solution (25-28%), sodium carbonate
(Na2CO3), diethyl ether, Cetyltrimethylammonium bromide (CTAB), dimethyl
sulfoxide (DMSO), hydrochloric acid (HCl) and ethanol were purchased from
Sinopharm chemical reagent Co., Ltd (Shanghai, China). TPZ was provided by
Dalian Meilun Biotech Co., Glucose oxidase (GOX) were purchased from GIBCO

New Journal of Chemistry
Page 6 of 32
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
Invitrogen Corp. (Carlsbad, USA). 3-aminopropyltriethoxysilane (APTES), 2, 2-
dithiodipyridine, and tirapazamine were purchased from Aladdin Reagent Co., Ltd
(Shanghai, China). Molecular probe 2-(4-amidinophenyl)-6-indolecarbamidine
dihydrochloride (DAPI) was obtained from Beyotime Biotechnology (China), annexin
V-FITC/PI apoptosis was purchased from 4A Biotech Co., Ltd. (Beijing, China). Fetal
bovine serum (FBS), Dulbecco’s modified Eagle’s medium (DMEM), All other
reagents were of analytical grade and used as received.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2. Preparation of TPZ/Fe3O4@MSN-GOX
2.2.1 Synthesis of the Fe3O4
Magnetic core was prepared via the thermal decomposition of iron oleate according to
the literature with some modifications. Briefly, Dissolving Iron(III) chloride
hexahydrate (FeCl3•6H2O, 5.4 g) and 17 mL oleic acid in 100 mL of methanol. Next,
adding 200 mL 0.3M NaOH thereto with stirring and washed product three times before
vacuum drying. Finally, iron oleate was obtained. 31.0g 1-octadecene and 5.5g iron
oleate were stirred at 65℃. And adding 1.5 g oleic acid and then vigorous stirred at 320℃
for 0.5h under nitrogen atmosphere. 40 mL of ethanol was added into thereto and shook
vigorously after cooling down. Then products were obtained by centrifugation at 8000
rpm and washed with 40 mL of acetone/hexane (3:1) three times by centrifugation. At
last, dispersing the Fe3O4 nanoparticles in chloroform at a concentration of 23.2 mg
Fe3O4 per mL.
2.2.2 Synthesis of Fe3O4@MSN nanoparticles
The Fe3O4@MSN nanoparticles were prepared via a modified sol-gel method37. Firstly,
440 µL 2M NaOH and 125 mg of CTAB were dissolved in 60 mL of deionized water,
heated to 70℃ for 1.5h under vigorous stirring. Then, 1 mL of Fe3O4 nanoparticles
solution with evenly distributed was added and chloroform was removed by continuous
nitrogen stream. Then, the solution was stirred at 80℃ for 1h and 200 mL of TEOS was
added for mixture. At last, an additional 400 mL of TEOS was added thereto and the

Page 7 of 32
New Journal of Chemistry
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
reacted at 80℃ for 1.5 h. The precipitate was collected by centrifugation and washed
with deionized water and ethanol respectively three times before vacuum drying.
Refluxed for 6 h in 90 mL ethanol solution of NH4NO3 (6 mg/mL) twice, Fe3O4@MSN
without CTAB template was obtained after washed with deionized water and ethanol
respectively three times before vacuum drying.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2.3 Synthesis of Fe3O4@MSN-NH2 nanoparticles
250mg of Fe3O4@MSN and 1.2 mL of APTES were mixed and refluxed in 50 mL
anhydrous toluene for 24 hours. Then, the precipitate was collected via centrifugation
at 12000 rpm for 15 min. Finally, Fe3O4@MSN-NH2 nanoparticles was obtained by
washing collected product with deionized water and ethanol respectively three times
and dried overnight under vacuum.
2.2.4 Drug loading and GOX capping
Dispersing 20 mg of TPZ and 40 mg Fe3O4@MSN-NH2 in 10mL PBS (pH=7.4) and
stirred for 24 h, collected by centrifuged and washed. Additionally, dispersing 58 mg
of EDC·HCl, 88 mg of NHS in 12 mL of deionized water and stirred for 20 min at room
temperature, followed with 10 mg of GOX, stirred for 8 hours, and finally 50 mg
TPZ/Fe3O4@MSN-NH2 was added. Subsequently, the TPZ/Fe3O4@MSN-GOX was
collected via centrifuging at 12000 rpm for 10 min, then washed by deionized water
ethanol orderly and dried overnight in vacuum.
2.3. Characterizations
Scanning electron microscopy (SEM) (FEI Inspect F50) and JEM-100CXII
transmission electron microscopy (TEM) were employed to analyze the morphology of
Fe3O4 NPs and Fe3O4@MSN. The surface area and pore size distribution were
measured by nitrogen (N2) adsorption-desorption method on a Brunauer-Emmett-Teller
Micromeritics Thristar 3000 analyzer. FTIR and UV-vis adsorption were respectively
performed on a Thermo Scientific Nicolet 6700 FTIR spectrometer (Asheville, NC,
USA) and UV-2600 (Shimadzu, Japan). Dynamic light scattering (DLS) was employed

New Journal of Chemistry
Page 8 of 32
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
to analyze zeta potential and hydrodynamic size of Fe3O4@MSN on a Malvern
Zetasizer NanoZS (UK) instrument. Thermogravimetric analysis (TGA) was conducted
on TG209 F3 thermogravimetric analyzer (NETZSCH, Germany) with a heating rate
of 10°C/min in a nitrogen stream. PPMS-9 (Quantum Design) and ICP-MS 7400
(Thermo Fisher Scientific) were used to detect Hysteresis loops and iron masses
respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.4. The magnetism of Fe3O4 NPs and Fe3O4@MSN
Hysteresis loops of Fe3O4 and Fe3O4@MSN nanoparticles were investigated by
superconducting quantum interference device vibrating sample magnetometer (SQUID,
Quantum Design MPMS) in previous work8, which had superior magnetic properties
and exhibited magnetic targeting potential in clinic.
2.5. Iron ion responsive release
Ferric ion was detected by dialysis method, each sample in triplicate. Briefly, 25 mL of
1 mg/mL TPZ/Fe3O4@MSN-GOX were dispersed in PBS solutions (pH =5.0, 7.4) with
or without glucose (1 mg/mL) then all solutions were stirred (100 rpm) at 37 °C in the
dark. At given intervals (0h, 1h, 2h, 4h, 8h, 12h, 24h, 36h, 48h),1 mL of the releasing
buffer was taken out and equal dosage of fresh buffer was replenished. After 15min
four systems were added with KSCN (10 mg/mL) respectively. At last, the UV-vis
absorption was quantitatively measured by standard curve at 478 nm and the iron ion
released also could be observed according to their color.
2.6. Drug loading capacity
Dispersing TPZ/Fe3O4@MSN-GOX in 20 mL of PBS (pH = 7.4) and stirred in the dark
at 25°C for 15h. Then the supernatant was removed by centrifuged, and the remaining
solid was dispersed in deionized water uniformly. All the cleaning liquid fractions were
collected according to the previous method38, all collected supernatant containing TPZ
and GOX detected by using a standard curve via UV-vis adsorption at 262 nm and
280nm respectively with a standard curve. Drug loading content (DLC) was determined

Page 9 of 32
New Journal of Chemistry
1
2
3
4
5
6
7
View Article Online
DOI: 10.1039/C9NJ05469C
by equations.
DLC (wt %) =
푤푒푖푔ℎ푡 표푓 푇푃푍 푖푛 푛푎푛표푝푎푟푡푖푐푙푒푠
푤푒푖푔ℎ푡 표푓 푛푎푛표푝푎푟푡푖푐푙푒푠
8
9
The amount of GOX was determined by the difference in mass loss of Fe3O4@MSN
and Fe3O4@MSN-GOX on the thermogravimetric analyzer (TGA).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A standard inductively coupled plasma mass spectrometry (ICP-MS) procedure was
conducted to quantify the amount of Fe3O4 nanoparticles in triplicate. Briefly, 1 mg of
TPZ/Fe3O4@MSN-GOX and TPZ/Fe3O4@MSN were dissolved into 10 mL deionized
water followed by measurement of ICP-MS.
푤푒푖푔ℎ푡 표푓푐푎푟푔표푠(퐹푒3푂4 푁푃푠)
DLC (wt %) =
푤푒푖푔ℎ푡 표푓 푛푎푛표푝푎푟푡푖푐푙푒푠
2.7. The stability of TPZ/Fe3O4@MSN-GOX in vitro
The prepared TPZ/Fe3O4@MSN-GOX was kept at 4°C for 4 weeks to study its stability.
All samples were stirred and purified by a 0.45 µm filter34. Changes in PDI and size
were assessed its stability by specific time intervals (0, 1, 2, 3, 4 weeks). 5 mg of
TPZ/Fe3O4@MSN-GOX was dispersed in 50 mL PBS (pH=7.4) solution and fetch 2
mL of TPZ/Fe3O4@MSN-GOX solution, treat with ultrasound for 5 min followed with
DLS test in triplicate at each scheduled time.
2.8. GOX reduces pH and promote H2O2 in vitro
GOX could reduce pH and elevate H2O2 concentration has been reported in previous
work39. GOX, Fe3O4@MSN, Fe3O4@MSN-GOX at specific concentration intervals
(800 µg/mL, 400 µg/mL, 200 µg/mL,100 µg/mL, 50 µg/mL, 0 µg/mL) were dispersed
in PBS (pH=7.4) with or without glucose (1 mg/mL) and incubated 4 hours in the dark
at 37°C. pH was detected by Orion portable pH meter (Orion 8103BN ROSS semi-
micro pH electrode, Boston, MA, USA) and H2O2 concentration was determined by the
2+
specific ultraviolet absorption of yellow titanium peroxide complex (Ti(IV)O2 )
23
arising from the reaction of the colorless Ti(IV)O2+ and H2O2 . In addition, all samples
were in triplicate.

New Journal of Chemistry
Page 10 of 32
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
2.9. In vitro drug release
5
6
7
8
The drug release of TPZ/Fe3O4@MSN-GOX was conducted in PBS (pH =7.4, pH=5.0)
with glucose (1 mg/mL) or without glucose by using a dialysis method.40 Equal amount
of TPZ/Fe3O4@MSN and TPZ/Fe3O4@MSN-GOX were transferred into dialysis bags
(MWCO 10 kDa) and soaked and stirred in 25 mL of PBS (pH =7.4, pH=5.0) at 37 °C
in the dark, and 2 mL of the buffer solution was taken at a specific interval (1 h, 2 h, 4
h, 8 h, 12h, 24 h, 48 h) to investigate the release concentration of TPZ, and equal dosage
of PBS solution was replenished. The release mass of TPZ was calculated by comparing
the specific UV-vis adsorption with a standard curve.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.10. Detection of hydroxyl radicals
Methylene blue (MB) was typical indicator of •OH, which could degrade methylene
blue by its strong oxidizing and could quantitative detected by UV-Vis absorption.
Dispersing 1 mg Fe3O4@MSN, Fe3O4@MSN-GOX in 25 mL PBS (pH=7.4) containing
10 μg/mL methylene blue, 8 mM H2O2 with or without glucose (1 mg/mL) and stirred
for 30 min. The ·OH produced by Fe3O4-mediated Fenton reaction could degrade
methylene blue and detected by absorbance change at 665 nm.
2.11. Cell culture
Human breast cancer cells MCF-7 and normal liver cells L-02 were selected and
incubated in RPMI-1640 complete medium containing 10% FBS and 1% penicillin-
streptomycin solution, which were cultured in a humidified atmosphere at 37°C
containing 5% CO2.
2.12. Cellular uptake
Confocal laser scanning microscope (CLSM, Zeiss, Germany) was applied to determine
the cellular uptake efficiency of TPZ/Fe3O4@MSN-GOX. 1 × 105 MCF-7 cells per well
were incubated at 37°C in a confocal petri dish (BD Falcon, USA) for 12 hours. Then
dispersing TPZ/Fe3O4@MSN-GOX uniformly in the RPMI-1640 complete medium
and then incubated at predetermined time intervals (1h, 4h, 8h) at 37°C. Next, all cells

Page 11 of 32
New Journal of Chemistry
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
were washed three times with PBS and then fixed with 4% paraformaldehyde for 15
minutes41, then followed by washing three times with PBS and 150 µL of DAPI (100
ng/mL) was added to the medium1. Finally, the cellular uptake was observed and
assessed via CLSM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.13. Detection of reactive oxygen species (ROS)
The generation of intracellular ROS was detected by a DCFDA assay. MCF-7 cells (1
× 105 cells per well) were seeded in confocal Petri dishes (BD Falcon, USA) containing
glucose (1 mg/mL) at 37°C for 24 h and incubated with TPZ/Fe3O4@MSN-GOX and
TPZ/Fe3O4@MSN for different incubation times (1h, 4h, 8h). The cells were washed
three times with PBS after incubation and stained with 1 mL of DCFDA (10 mM) for
20 min, which was observed by CLSM after washed with PBS.
2.14. In vitro cytotoxicity test
MTT assays were applied to determine the cytotoxicity of different particles against
MCF-7 cells and L-02 cells at 6.0 × 103 cells/well into 96-well plates. Starvation
therapy was conduct at the glucose-containing and glucose-free DMEM media. Then
100 μg/mL, 200 μg/mL Fe3O4@MSN-GOX were added to each well and co-incubation,
at last MTT assays was used to evaluated viability. After incubated with RPMI-1640
complete medium (100 µL) for 24 hours, L-02 cells were incubated in a normoxic (20%
O2) environment with blank MSN at different concentrations (3.12, 6.25, 12.5, 25, 50,
100, 200 µg/mL) to test the biocompatibility of MSN and incubated with free TPZ,
TPZ/Fe3O4@MSN and TPZ/Fe3O4@MSN-GOX with different concentrations (0.5, 1,
2, 4, 8, 16 µg/mL) to test their toxicity to normal cells. Subsequently, TPZ dose-
dependent concentrations (0.5, 1, 2, 4, 8, 16 µg/mL) of free TPZ, TPZ/Fe3O4@MSN
and TPZ/Fe3O4@MSN-GOX under different O2 concentrations were co-cultured with
MCF-7 cells into in glucose-containing (1 mg/mL) media. Before adding fresh 200 µL
DMEM medium to each well, all cells were cleared with 200 µL PBS and then cultured
for 24 hours in a normoxic (20% O2) and hypoxic (5% O2) environment at 37°C, and
each well was rinsed. Next, 150 µL complete medium containing 20 µL of MTT was

New Journal of Chemistry
Page 12 of 32
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
added thereto and incubated for 4 hours. 180 µL of DMSO was added to replace
previous medium each well in the 96-well plate8, which then shaken on a benchtop
concentrator for 15 minutes42. At last, plate reader Spectra Max Microplate Reader
(Sunnyvale, CA, US) was used for measurements.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.15. Cell apoptosis
Cell apoptosis was detected by Annexin APC/7-AAD Apoptosis Detection Kit II (BD
Pharmingen, San Jose, USA). Briefly, the same TPZ dosage of 40 µg/mL free TPZ,
TPZ/Fe3O4@MSN and TPZ/Fe3O4@MSN-GOX were added into 6-well plates, which
were seeded at 1.0×105 cells/well and cultured in a normoxic (20% O2) and hypoxic (5%
O2) environment at 37°C. Control group was conducted with 40 µg/mL MSN co-
cultured. The medium was removed after 12h followed with all cells washed three times
and digested with 1.0 mL of EDTA-free trypsin. The cells were collected and dispersed
uniformly in 500 µL buffer, and then 5 µL of propidium iodide and 5 µL of Annexin
V-FITC were added thereto in the dark to stain the cells for 15 minutes. At last, all
samples were analyzed by flow cytometry on a FACS Calibur (BD Biosciences, USA).
3. Results and discussion
3.1 Synthesis and characterization of Fe3O4@MSN-GOX
In this work, core-shell structured Fe3O4@MSN-GOX were fabricated by the
procedures illustrated in Scheme 1. As the core, Fe3O4 nanoparticles were obtained by
thermal decomposition method, with hydrophobic and monodisperse morphology,
whose diameter size was about 8 nm showed in Figure 1(a). As the shell, the
mesoporous silica was coated onto the Fe3O4 NPs by a traditional Stöber method. GOX
was grafted on the surface of Fe3O4@MSN to block off the mesoporous. Fe3O4@MSN
exhibited uniform nanospheres with diameter size of 55 nm, which were observed by

Page 13 of 32
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
TEM and SEM in Figure 1(b) and (c). In addition, in Figure 3(a) and Figure 6(d),
5
6
7
8
polydispersity index (PDI) value and hydrodynamic size of Fe3O4@MSN were about
0.154 and 88 nm respectively, owing to the exterior hydration layers. According to
Figure 2(a) and (b), N2 adsorption-analytical analysis result revealed that the highly
surface area and uniform, defined pore size of Fe3O4@MSN were 517.4 m2g-1 and 2.6
nm respectively.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. (a) TEM images of Fe3O4, (b)SEM image and (c) TEM images of Fe3O4@
MSN.
The modification of GOX onto the surface of Fe3O4@MSN was verified by zeta
potential, FTIR and TGA. The distinct change of zeta potential value of nanoparticles
suggested the successful modification in Figure 3(c). The zeta potential of
Fe3O4@MSN (-25.3 mV) and Fe3O4@MSN-NH2 (17.5 mV) occurred sharp reversed
owing to the typical positive charge of -NH2. After modification with GOX,
Fe3O4@MSN received negative zeta potential (-9.8 mV). FTIR spectra could be served
to prove successful synthesis of materials. According to Figure 3(b), The iron-oxygen
bond vibration, Si-O and C=O from oleic acid were reflected by the signals of 580 cm-1,
475 cm-1 and 1648 cm-1 in FTIR, which could confirm the formation of inner ferric
oleate and outer MSN. The peak at 1388 cm-1 in Fe3O4@MSN-GOX was weaker than
in Fe3O4@MSN, attributed to the plentiful -CH3 of tetraethyl orthosilicate (TEOS). The
peaks at 1550 cm-1, 1445 cm-1 in Fe3O4@MSN-GOX were originated form the vibration
of C-N and -NH2 respectively, attributed to the -NH2 come from GOX. The weight loss
value of Fe3O4@MSN and Fe3O4@MSN-GOX were respectively 21.62% and 27.22%,

New Journal of Chemistry
Page 14 of 32
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
which could quantitatively calculate the amount of GOX (5.6%) according to Figure
5
6
3(d).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. (a) Nitrogen adsorption/desorption isotherm and (b) the pore size distribution
of Fe3O4@MSN.
Figure 3. Related characterization of nanoparticle including (a) hydrodynamic
diameter size of Fe3O4@MSN, (b) FTIR spectra of Fe3O4@MSN and Fe3O4@MSN-
GOX, (c) Zeta potential of Fe3O4@MSN, Fe3O4@MSN-NH2, Fe3O4@MSN-GOX,
TPZ/Fe3O4@MSN-GOX and (d) thermogravimetric analysis of Fe3O4@MSN,

Page 15 of 32
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
Fe3O4@MSN-GOX, TPZ/Fe3O4@MSN-GOX.
5
6
7
8
9
3.2. Ferric ion release
The determination of the iron ion content in the nanoparticles is a prerequisite for
evaluating the effect of the Fenton reaction. The mass of iron in Fe3O4@MSN and
Fe3O4@MSN-GOX were 35.14 µg mg-1 and 33.15 µg mg-1 respectively determined by
ICP-MS, which was shown in Figure 4. The iron content declined slightly in
Fe3O4@MSN-GOX compared with Fe3O4@MSN, which also confirmed the successful
modification of GOX. Due to the complexity and variability of the tumor
microenvironment, we conduct the ferric ion release in vitro for simulating the tumor
microenvironment. Fe(SCN)3 was formed when ferric ions encountered thiocyanate
and specific ultraviolet absorption (Figure S1(c)) was detected at 478 nm. The blood
red solution gradually deepened as the acidity decreased that indicate the lowed pH
could accelerate the release of ferric ions. Different organelles have different pH values
in the tumor, such as endosomes (5.0–6.5) and lysosomes (4.5–5.0). However, GOX
carried on the surface of the nanoparticles could further reduce the acidity by oxidizing
glucose, which was confirmed by previous work39. After disposed sample with 1h, the
blood red solution also deepened as glucose concentration increased and the reduced
pH also increased the of release of ferric ion in Figure 5(a). Low pH could promote the
dissolution of Fe3O4 to release ferric ion, which served as efficient catalyst for Fenton
reactions. The relevant UV quantitative analysis of prepared nanoparticles ferric ions
release with time was shown in Figure 5(b). As revealed by the results, GOX can rely
on the metabolism of glucose to reduce the pH of the system, thereby accelerate the
release of iron ions, thus provide highly efficient catalyst for the Fenton reaction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

New Journal of Chemistry
Page 16 of 32
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. The mass rate of iron in Fe3O4@MSN and Fe3O4@MSN-GOX.
Figure 5. (a) Ferric ions release of TPZ/Fe3O4@MSN-GOX over time in PBS (pH=5.0,
pH=7.4) with or without glucose (1 mg/mL) and (b) Color changes of Fe3O4@MSN
and Fe3O4@MSN-GOX in PBS at different pH values with or without glucose (1
mg/mL).
3.3. The stability of TPZ/Fe3O4@MSN-GOX and in vitro drug release
Drug loading rate of TPZ/Fe3O4@MSN-GOX was investigated by UV-vis adsorption
peak at 262 nm for TPZ, which were quantitatively calculated and the amounts were
14.9% of TPZ. Respectively, the content of GOX was 5.8% which was determined by
the more accurate the BCA Protein Assay Kit (Figure S4). UV-vis adsorption and

Page 17 of 32
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
standard curves of TPZ were shown in Figure S1(b). Furthermore, loading rate of TPZ
and GOX were also determined by comparing the weight loss value of
TPZ/Fe3O4@MSN-GOX (58.51%), Fe3O4@MSN-GOX (72.78%) and Fe3O4@MSN
(78.30%) in Figure 1(f). Prepared nanoparticles were stored at 4°C for 14 days in the
PBS (pH = 7.4) for the evaluation of the stability and the index of PDI and particle size
were shown in Figure 6(d). The PDI and particle size remained general stable with
only minimal fluctuations around 0.154 and 88 nm, which proved the admirable
stability of the nanoparticles at 4°C. Drug release experiments were studied in PBS (pH
= 5.0, glucose (1 mg/mL)), PBS (pH = 7.4, glucose (1 mg/mL)), PBS (pH = 5.0) and
PBS (pH = 7.4) by dialysis method in Figure 6(c). In weak acid environment without
glucose, the release of TPZ (29.4%) was faster than that in neutral environment (3.5%)
without glucose within 48h. And drug release could be accelerated ((pH = 5.0, 59.3%;
pH = 7.4, 43.8%) with the presence of glucose (1 mg/mL). The result suggested that in
the tumor microenvironment, weak acid and abundant glucose would promote drug
release. However, TPZ/Fe3O4@MSN-GOX would further reduce acidity by catalyzing
the oxidation of glucose.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.4. GOX reduces pH and promotes H2O2 in vitro
GOX could raise H2O2 levels and lower pH by oxidizing glucose and the quantitative
analysis results were shown in Figure 6(a) and (b). In the groups of GOX and
Fe3O4@MSN-GOX, reduced pH and elevated H2O2 concentration were detected in
PBS containing glucose (1 mg/mL) after 4h, there was no apparent change in PBS
without glucose. However, In the group of Fe3O4@MSN, regardless of whether or not
glucose was contained, there was no change in pH and H2O2 level. The result could
explain the effects of GOX on lowering pH and elevating H2O2, further demonstrates
the feasibility of hunger treatment and chemodynamic therapy therapy.

New Journal of Chemistry
Page 18 of 32
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Changes of (a)pH and (b)H2O2 concentration of GOX, Fe3O4@MSN and
Fe3O4@MSN-GOX in PBS (pH=7.4) with or without glucose (1 mg/mL) as
concentration increased. (c) Drug release curves of different nanoparticles over time in
PBS (pH=5.0, pH=7.4) with or without glucose (1 mg/mL). (d) changes of size and PDI
were tested to evaluate the stability of TPZ/Fe3O4@MSN-GOX.
3.5. Detection of hydroxyl radicals
Highly toxic hydroxyl radicals (•OH) depends on the weak acid environment, Fe3O4
catalyzes the high level of H2O2 produced in the weak acid environment tumor
microenvironment15. Methylene blue was typical indicator of •OH, which could react
with methylene blue to generate colorless product and could be quantitative detected
by UV-Vis absorption at 665 nm. In Figure 7 (a), compared with the absorption spectra
of methylene blue, the absorption spectra of glucose-free methylene blue solution added
with Fe3O4@MSN and Fe3O4@MSN-GOX had a significant decline. However, in
methylene blue solution containing glucose (1 mg/mL), the absorption spectra added
with Fe3O4@MSN-GOX decreased sharply and approximately disappeared in Figure

Page 19 of 32
New Journal of Chemistry
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
7 (b). Simultaneously, the colors were also lightened with the decrease in absorbance.
The result suggested that the nanoparticles we constructed could effectively generate
hydroxyl radicals (•OH) and Fe3O4@MSN-GOX could enhanced the level of hydroxyl
radicals in glucose media due to the reduced pH and enhanced H2O2 level aroused by
the GOX-driven oxidation reaction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. (a) UV-vis absorption spectra and images of MB, MB with Fe3O4@MSN and
MB with Fe3O4@MSN-GOX in glucose-free solutions. (b) UV-vis absorption spectra
and images of MB, MB with Fe3O4@MSN and MB with Fe3O4@MSN-GOX in
containing glucose solutions (1 mg/mL).
3.6. Intracellular uptake evaluation
Intracellular distribution of fluorescent TPZ was used to evaluate the uptake of
TPZ/Fe3O4@MSN-GOX and MCF-7 cells at the density of 1×105 cells/dish cultured in
DMEM containing 5% CO2 at 37 ℃ at specific intervals (1h, 4h, 8h) were observed by
confocal laser scanning microscopy (CLSM)7. Green and blue fluorescence respective
came from TPZ and the MCF-7 cell nuclei was stained by DAPI. In Figure S2, green
fluorescence intensity (TPZ) of TPZ/Fe3O4@MSN-GOX in MCF-7 cells had a
significant improvement over time, which was attributed to the mechanism that
positively charged TPZ/Fe3O4@MSN-GOX tended to be endocytosed by negatively
overexpressed MCF-7 cells and enhanced permeability and retention (EPR) effect.

New Journal of Chemistry
Page 20 of 32
1
2
3
4
5
6
7
8
View Article Online
DOI: 10.1039/C9NJ05469C
3.7. Reactive oxygen species studies
The intracellular ROS level of TPZ/Fe3O4@MSN-GOX was quantitatively detected by
the green fluorescent of dichlorofluorescein (DCF), product of
a
9
dichlorodihydrofluorescein diacetate (DCFDA) oxidized by ROS. The GOX could
remarkably increase the H2O2 concentration, which has been proved by previous work.
highly toxic ROS was generated by the intracellular superabundant H2O2 under the
catalysis of Fe3+/Fe2+. Difference of ROS levels induced by TPZ/Fe3O4@MSN-GOX,
TPZ/Fe3O4@MSN at specific interval time (1h, 4h, 8h) in MCF-7 cells was shown in
Figure 8. Fluorescence intensity of TPZ/Fe3O4@MSN-GOX had more obvious
improvement in MCF-7 cells than that of TPZ/Fe3O4@MSN after 8h, which was
attributed to elevated concentration of H2O2 aroused by GOX and high level of ROS
around by decomposition of H2O2 in acid environment. However, only negligible
fluorescence signal could be detected in TPZ/Fe3O4@MSN in the first 1h, but brighter
fluorescence intensity was detected in TPZ/Fe3O4@MSN-GOX and significant
improvement of the fluorescence intensity of TPZ/Fe3O4@MSN-GOX increased as
time. The result suggested that TPZ/Fe3O4@MSN-GOX we constructed could
stimulate highly toxic ROS and revealed the efficient potency in chemodynamic
therapy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Page 21 of 32
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. CLSM images of MCF-7 cells incubated with Fe3O4@MSN and
Fe3O4@MSN-GOX for different incubation times. The scale bar is 50 μm.
3.8. In vitro cytotoxicity
GOX induced starvation therapy by depriving glucose and O2, activated the therapeutic
effect of the hypoxia-inducible TPZ and induced highly toxic ROS, which could
effectively inhibit tumors. MTT assays was used to demonstrate the toxicities and
pharmacological action of nanoparticles toward MCF-7 cells incubated with
TPZ/Fe3O4@MSN-GOX, TPZ/Fe3O4@MSN, free TPZ and control group under
different O2 concentrations in glucose-containing (1 mg/mL). Co-cultured with empty
carrier MSN at different concentrations in glucose-containing (1 mg/mL) as control
group, the cell viability of MCF-7 cells (20% O2 concentrations), MCF-7 cells (5% O2
concentrations) and L-02 cells (20% O2 concentrations) all exceeded 85% at 200 µg/mL
in Figure 9 (a), which was attributed to biocompatibility of MSN and the cytotoxicities
of control group was insignificantly influenced by the O2 concentrations. Starvation
therapy triggered by Fe3O4@MSN-GOX leaded to the decline of cell viability at the

New Journal of Chemistry
Page 22 of 32
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
glucose-containing DMEM media and lower cell viability at a higher concentration of
glucose (Figure S3), which attributed to the exhaust of glucose and higher level of toxic
H2O2 production, which was proved in previous work.39 In glucose-containing (1
mg/mL) with 20% O2 concentrations, TPZ/Fe3O4@MSN-GOX (91.7% cell viability),
TPZ/Fe3O4@ MSN (93.6% cell viability), TPZ (97.5% cell viability) displayed low
level in killing L-02 cells in Figure 9(b), which confirmed the security and
biocompatibility to normal cells of nanoparticles we prepared. In normal cells, the O2
concentrations and pH value are at high level and H2O2 concentration is at low level,
which meant that hypoxic prodrug TPZ would not be activated and highly toxic
hydroxyl radicals would not generate for lack of H2O2 and acidic environment. In
glucose-containing (1 mg/mL) with 20% O2 concentrations, TPZ/Fe3O4@MSN-GOX
(4.7% cell viability) displayed outstanding potency in killing MCF-7 cells than
TPZ/Fe3O4@MSN (62.1% cell viability) and free TPZ (93.3% cell viability) at TPZ
concentrations of 16 µg/mL in Figure 9(c). At high oxygen concentrations, hypoxic-
induced TPZ had negligible killing effect on cells. With the participation of GOX,
TPZ/Fe3O4@MSN-GOX could further reduce the intracellular pH and increase H2O2,
promoting the production of ROS, while it was difficult for TPZ/Fe3O4@MSN to obtain
the same effect due to the lack of GOX. GOX and TPZ were loaded into the
biocompatible MSN nanoparticles, and TPZ/Fe3O4@MSN-GOX could consume
glucose and oxygen, provoke spatial control of hypoxia-sensitive TPZ release and cause
exacerbated hypoxia. However, in glucose-containing (1 mg/mL) with 5% O2
concentrations, TPZ/Fe3O4@MSN-GOX (8.6% cell viability), TPZ/Fe3O4@MSN
(29.9% cell viability), TPZ (33.3% cell viability) displayed enhanced potency in
inhibiting MCF-7 cells in Figure 9(d), which was attributed to the synergy of starvation
therapy, chemodynamic therapy and sensitized chemotherapy, while could not
displayed excellent effect for no adequate oxygen triggered sequential multitherapy.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Page 23 of 32
New Journal of Chemistry
1
2
3
4
View Article Online
DOI: 10.1039/C9NJ05469C
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. (a) Cells viability rates of different MSN concentrations at different oxygen
concentrations. (b) L-02 cells viability rates of different TPZ dose-dependent
concentrations of different nanoparticle in 20% O2 concentration with glucose (1
mg/mL). (c) MCF-7 cells viability rates of different TPZ dose-dependent
concentrations of different nanoparticle in 5% O2 concentration. (d) MCF-7 cells
viability rates of different TPZ dose-dependent concentrations of different nanoparticle
in 5% O2 concentration, which treated by different nanoparticles calculated by MTT
assay. Data were shown as mean ± S.D. (n=3). Significance is defined as * P < 0.05, **
P < 0.01
3.9. Cell apoptosis
Cell apoptosis induced by different nanoparticles on MCF-7 cells was detected by flow
cytometry with annexin-V-FITC/P assay under different O2 concentrations (5% and
20%). In Figure 10, the apoptosis rates of free TPZ, TPZ/Fe3O4@MSN,
TPZ/Fe3O4@MSN-GOX on MCF-7 cells were respectively 25.3%, 75.7%, 93.4% with

New Journal of Chemistry
Page 24 of 32
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
the same TPZ dosage of 40 µg/mL in 20% O2 and high apoptosis was obtained in 5%
O2, which were 57.4%, 79.5%, 88.2% respectively. 40 µg/mL MSN as the control group
exhibited a low apoptotic rate (2.7% in 20% O2 and 6.0% in 5% O2), suggesting that it
was safe and biocompatible to tissue. TPZ/Fe3O4@MSN-GOX displayed better
potency in inducing cell apoptosis than free TPZ and TPZ/Fe3O4@MSN, attributed to
the highly toxic ROS. GOX could elevate the heterogeneity of tumor by reducing pH
and enhancing hypoxia, which would sensitize the effect of TPZ and toxic ROS. More
important, at high oxygen (20% O2), GOX could work effectively and produced more
H2O2 and lower pH, stimulating more highly toxic ROS, which leaded the more
apoptosis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Cell apoptosis induced by control group, free TPZ, TPZ/Fe3O4@MSN,
TPZ/Fe3O4@MSN-GOX at 40 µg/mL TPZ dose-dependent concentrations in a
normoxic (20% O2) and hypoxic (5% O2) environment for 12h.
Conclusion
In summary, multimodal therapies based on synergy of starvation therapy,
chemotherapy and chemodynamic therapy was introduced for efficient cancer therapy.
A novel drug delivery system (TPZ/Fe3O4@MSN-GOX) was successful constructed,
which characterized by FT-IR, UV-vis spectrometry, TEM, SEM, TGA and DLS and
possessed uniformly dispersed size, distinct pore structure with highly drug loading and

Page 25 of 32
New Journal of Chemistry
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
stability. The result revealed that GOX could reduce pH and elevate H2O2 concentration.
Reduced pH could accelerate the release of TPZ and iron ion were proved by vitro
experiment. Abundant hydroxyl radicals were generated by Fe3O4@MSN-GOX, which
detected by the color and UV-vis absorption changes of MB. Obvious increasing in
ROS level of MCF-7 cells were detected when was co-cultured with Fe3O4@MSN-
GOX in normoxic environment containing glucose. Multifunctional
TPZ/Fe3O4@MSN-GOX nanoparticle was also revealed high cellular uptake capacity,
excellent inhibitory effect and cell apoptosis on MCF-7 cells, which would have great
prospects in the field of cancer therapy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgments
We are highly grateful to the International S&T Cooperation Program of China (No.
2015DFG42240), the financial support from the National Natural Science Foundation
of China (Grant Nos. 21628101 and 21371031), the project funded by China
Postdoctoral Science Foundation (NO. 2019M651841), the Priority Academic Program
Development (PAPD) of Jiangsu Higher Education Institutions and State Key
Laboratory of Bioelectronics-Zhongda Hospital cross innovation cooperation research
fund (No. 2018yy-jccx012)
Author Information
Corresponding Author
Baiwang Sun; E-mail address: chmsunbw@seu.edu.cn
Notes
The authors declare no competing financial interest. All experiments in this research
work complied with the Regulations on the Administration of Laboratory Animals and

New Journal of Chemistry
Page 26 of 32
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
the relevant national laws and regulations and were performed under the guidance
approved by the Laboratory Animal Ethics Committee at the School of Medicine,
Southeast University (Nanjing, China).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
References
(1) Gao, Z.; Li, Y.; You, C.; Sun, K.; An, P.; Sun, C.; Wang, M.; Zhu, X.; Sun, B. Iron
Oxide Nanocarrier-Mediated Combination Therapy of Cisplatin and Artemisinin for
Combating Drug Resistance through Highly Increased Toxic Reactive Oxygen Species
Generation. ACS Appl. Bio Mater. 2018, 1, 270–280.
(2) Lee, E. S.; Na, K.; Bae, Y. H. Polymeric micelle for tumor pH and folate-mediated
targeting. Journal of Controlled Release 2003, 91, 103–113.
(3) Li, Y.; Gao, Z.; Chen, F.; You, C.; Wu, H.; Sun, K.; An, P.; Cheng, K.; Sun, C.; Zhu,
X. et al. Decoration of Cisplatin on 2D Metal-Organic Frameworks for Enhanced
Anticancer Effects through Highly Increased Reactive Oxygen Species Generation. ACS
Appl. Mater. Interfaces. 2018, 10, 30930–30935.
(4) Zhang, M.-K.; Li, C.-X.; Wang, S.-B.; Liu, T.; Song, X.-L.; Yang, X.-Q.; Feng, J.;
Zhang, X.-Z. Tumor Starvation Induced Spatiotemporal Control over Chemotherapy for
Synergistic Therapy. Small 2018, 14, e1803602.
(5) Wang, Y.; Song, S.; Liu, J.; Liu, D.; Zhang, H. ZnO-Functionalized Upconverting
Nanotheranostic Agent: Multi-Modality Imaging-Guided Chemotherapy with On-Demand

Page 27 of 32
New Journal of Chemistry
1
2
3
4
5
View Article Online
DOI: 10.1039/C9NJ05469C
Drug Release Triggered by pH. Angew. Chem. Int. Ed. 2015, 54, 536–540.
6
7
8
9
(6) Fu, L.-H.; Qi, C.; Hu, Y.-R.; Lin, J.; Huang, P. Glucose Oxidase-Instructed Multimodal
Synergistic Cancer Therapy. Adv. Mater. 2019, 0, 1808325.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(7) Huang, L.; Liu, J.; Gao, F.; Cheng, Q.; Lu, B.; Zheng, H.; Xu, H.; Xu, P.; Zhang, X.;
Zeng, X. A dual-responsive, hyaluronic acid targeted drug delivery system based on
hollow mesoporous silica nanoparticles for cancer therapy. J. Mater. Chem. B 2018, 6,
4618–4629.
(8) Sun, K.; Gao, Z.; Zhang, Y.; Wu, H.; You, C.; Wang, S.; An, P.; Sun, C.; Sun, B.
Enhanced highly toxic reactive oxygen species levels from iron oxide core–shell
mesoporous silica nanocarrier-mediated Fenton reactions for cancer therapy. J. Mater.
Chem. B 2018, 6, 5876–5887.
(9) Shi, C.; Liu, T.; Guo, Z.; Zhuang, R.; Zhang, X.; Chen, X. Reprogramming Tumor-
Associated Macrophages by Nanoparticle-Based Reactive Oxygen Species
Photogeneration. Nano lett. 2018, 18, 7330–7342.
(10) Wang, S.; Wang, Z.; Yu, G.; Zhou, Z.; Jacobson, O.; Liu, Y.; Ma, Y.; Zhang, F.;
Chen, Z.-Y.; Chen, X. Tumor-Specific Drug Release and Reactive Oxygen Species
Generation for Cancer Chemo/Chemodynamic Combination Therapy. Adv. Sci. 2019, 6,
1801986.
(11) Liu, C.; Wang, D.; Zhang, S.; Cheng, Y.; Yang, F.; Xing, Y.; Xu, T.; Dong, H.;
Zhang, X. Biodegradable Biomimic Copper/Manganese Silicate Nanospheres for
Chemodynamic/Photodynamic Synergistic Therapy with Simultaneous Glutathione
Depletion and Hypoxia Relief. ACS Nano 2019, 13, 4267–4277.

New Journal of Chemistry
Page 28 of 32
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/C9NJ05469C
(12) Jeong, D.; Bae, B.-c.; Park, S.-j.; Na, K. Reactive oxygen species responsive drug
releasing nanoparticle based on chondroitin sulfate–anthocyanin nanocomplex for
efficient tumor therapy. Journal of Controlled Release 2016, 222, 78–85.
(13) Tang, Z.; Liu, Y.; He, M.; Bu, W. Chemodynamic Therapy: Tumour
Microenvironment-Mediated Fenton and Fenton-like Reactions. Angew. Chem. Int. Ed.
2019, 58, 946–956.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(14) Wu, H.; You, C.; Chen, F.; Jiao, J.; Gao, Z.; An, P.; Sun, B.; Chen, R. Enhanced
cellular uptake of near-infrared triggered targeted nanoparticles by cell-penetrating
peptide TAT for combined chemo/photothermal/photodynamic therapy. Materials Science
and Engineering: C 2019, 103, 109738.
(15) Huo, M.; Wang, L.; Chen, Y.; Shi, J. Tumor-selective catalytic nanomedicine by
nanocatalyst delivery. Nature communications 2017, 8, 357.
(16) Wang, Y.; Yin, W.; Ke, W.; Chen, W.; He, C.; Ge, Z. Multifunctional Polymeric
Micelles with Amplified Fenton Reaction for Tumor Ablation. Biomacromolecules 2018,
19, 1990–1998.
(17) Yang, Y.; Yang, X.; Li, H.; Li, C.; Ding, H.; Zhang, M.; Guo, Y.; Sun, M. Near-
infrared light triggered liposomes combining photodynamic and chemotherapy for
synergistic breast tumor therapy. Colloids and surfaces. B, Biointerfaces 2019, 173, 564–
570.
(18) Zhang, L.; Wan, S.-S.; Li, C.-X.; Xu, L.; Cheng, H.; Zhang, X.-Z. An Adenosine
Triphosphate-Responsive Autocatalytic Fenton Nanoparticle for Tumor Ablation with Self-
Supplied H2O2 and Acceleration of Fe(III)/Fe(II) Conversion. Nano lett. 2018, 18, 7609–

Page 29 of 32
New Journal of Chemistry
1
2
3
4
5
View Article Online
DOI: 10.1039/C9NJ05469C
7618.
6
7
8
9
(19) Gao, S.; Lin, H.; Zhang, H.; Yao, H.; Chen, Y.; Shi, J. Nanocatalytic Tumor Therapy
by Biomimetic Dual Inorganic Nanozyme‐Catalyzed Cascade Reaction. Adv. Sci. 2019,
6, 1801733.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(20) Gao, Z.; Li, Y.; You, C.; Sun, K.; An, P.; Sun, C.; Wang, M.; Zhu, X.; Sun, B. Iron
Oxide Nanocarrier-Mediated Combination Therapy of Cisplatin and Artemisinin for
Combating Drug Resistance through Highly Increased Toxic Reactive Oxygen Species
Generation. ACS Appl. Bio Mater. 2018, 1, 270–280.
(21) Fan, J.-X.; Peng, M.-Y.; Wang, H.; Zheng, H.-R.; Liu, Z.-L.; Li, C.-X.; Wang, X.-N.;
Liu, X.-H.; Cheng, S.-X.; Zhang, X.-Z. Engineered Bacterial Bioreactor for Tumor Therapy
via Fenton-Like Reaction with Localized H2O2 Generation. Adv. Mater. 2019, 31,
1808278.
(22) Giannakis, S.; Voumard, M.; Rtimi, S.; Pulgarin, C. Bacterial disinfection by the
photo-Fenton process: Extracellular oxidation or intracellular photo-catalysis? Applied
Catalysis B: Environmental 2018, 227, 285–295.
(23) Fan, W.; Lu, N.; Huang, P.; Liu, Y.; Yang, Z.; Wang, S.; Yu, G.; Liu, Y.; Hu, J.; He,
Q. et al. Glucose-Responsive Sequential Generation of Hydrogen Peroxide and Nitric
Oxide for Synergistic Cancer Starving-Like/Gas Therapy. Angew. Chem. Int. Ed. 2017,
56, 1229–1233.
(24) Fu, L.-H.; Qi, C.; Lin, J.; Huang, P. Catalytic chemistry of glucose oxidase in cancer
diagnosis and treatment. Chem. Soc. Rev. 2018, 47, 6454–6472.
(25) Liang, R.; Chen, Y.; Huo, M.; Zhang, J.; Li, Y. Sequential catalytic nanomedicine