Enhanced Reactive Oxygen Species Generation by Mitochondria Targeting of Anticancer Drug to Overcome Tumor Multidrug Resistance

Article abstract of DOI:10.1021/acs.biomac.9b00800

As a major clinical tumor chemotherapeutic burden, multidrug resistance (MDR) is often a result of up-regulation of P-glycoprotein (P-gp), which strongly enhances anticancer drug efflux. The excess mitochondrial reactive oxygen species (ROS) could not only inhibit the function of P-gp through insufficient adenosine triphosphate supply but also cause apoptosis in MDR cells. Here, we designed a mitochondria targeting nanoparticulate system (GNPs-P-Dox-GA) for overcoming MDR through enhanced ROS generation, where increased cellular uptake as well as mitochondria accumulation were both realized by glycyrrhetinic acid (GA). First, doxorubicin was conjugated with GA (GA-Dox) and then grafted onto a N-(2-hydroxypropyl) methacrylamide (HPMA) copolymer backbone via hydrazone bond (P-Dox-GA). The obtained P-Dox-GA was subsequently attached to the surface of gelatin nanoparticles (GNPs). As gelatin is a substrate of tumor extracellular metal matrix protease-2 (MMP2), GNPs-P-Dox-GA nanoparticles could be degraded and release small size P-Dox-GA to facilitate tumor tissue penetration. After P-Dox-GA internalized by tumor cells under GA mediation, Dox-GA detached from HPMA copolymer through hydrolysis of hydrazone bond and then efficiently delivered to mitochondria. Compared to non-GA modified carriers, GNPs-P-Dox-GA exhibited increased cellular uptake nearly 4-fold and mitochondria distribution 8.8-fold, and increased ROS production level nearly 3-fold, significantly decreased efflux rate (55% compared with Dox group) in drug resistant HepG2/ADR cells, and then led to improved in vitro antitumor efficiency in HepG2/ADR cells (IC₅₀ only 19.5% of unmodified ones) as well as exciting in vivo antitumor efficiency on HepG2/ADR heterotopic tumor nude mice (1.75-fold higher tumor growth inhibition rate than free drug).

Full text of DOI:10.1021/acs.biomac.9b00800

Subscriber access provided by Columbia University Libraries
Article
Enhanced Reactive Oxygen Species Generation by Mitochondria
Targeting of Anti-cancer Drug to Overcome Tumor Multidrug Resistance
Yuanyuan Liu, Zhou Zhou, Xi Lin, Xiaofeng Xiong, Rui Zhou, Minglu Zhou, and Yuan Huang
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.9b00800 • Publication Date (Web): 29 Aug 2019
Downloaded from pubs.acs.org on August 31, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 27
Biomacromolecules
1
2
3
4
5
6
7
8
Enhanced Reactive Oxygen Species Generation by Mitochondria
Targeting of Anti-cancer Drug to Overcome Tumor Multidrug
Resistance
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
Yuanyuan Liu#, Zhou Zhou#, Xi Lin, Xiaofeng Xiong, Rui Zhou, Minglu Zhou, Yuan Huang*
Key Laboratory of Drug Targeting and Drug Delivery System, Ministry of Education, West China School of
Pharmacy, Sichuan University, No. 17, Block 3, Southern Renmin Road, Chengdu 610041, P. R. China.
# These authors contributed equally to this work.
*Phone/fax: +86-28-85501617.
E-mail: huangyuan0@163.com.
1
ACS Paragon Plus Environment

Biomacromolecules
Page 2 of 27
1
2
3
4
5
6
Abstract
7
8
9
As a major clinical tumor chemotherapeutic burden, multidrug resistance (MDR) is often a result of up-
regulation of P-glycoprotein (P-gp), which strongly enhance anti-cancer drug efflux. The excess mitochondrial
reactive oxygen species (ROS) not only could inhibit the function of P-gp through insufficient adenosine
triphosphate supply, but also could cause apoptosis in MDR cells. Here, we designed a mitochondria targeting
nanoparticulate system (GNPs-P-Dox-GA) for overcoming MDR through enhanced ROS generation, where
increased cellular uptake as well as mitochondria accumulation were both realized by glycyrrhetinic acid (GA).
Firstly, doxorubicin was conjugated with GA (GA-Dox) and then grafted onto N-(2-hydroxypropyl) methacrylamide
(HPMA) copolymer backbone via hydrazone bond (P-Dox-GA). The obtained P-Dox-GA was subsequently
attached to the surface of gelatin nanoparticles (GNPs). As gelatin is a substrate of tumor extracellular metal
matrix protease-2 (MMP2), GNPs-P-Dox-GA nanoparticles could be degraded and release small size P-Dox-GA to
facilitate tumor tissue penetration. After P-Dox-GA internalized by tumor cells under GA mediation, Dox-GA
detached from HPMA copolymer through hydrolysis of hydrazone bond and then efficiently delivered to
mitochondria. Compared to non-GA modified carriers, GNPs-P-Dox-GA exhibited increased cellular uptake nearly for
4-fold and mitochondria distribution for 8.8-fold, and increased ROS production level for nearly 3-fold, significantly
decreased efflux rate (55% compared with Dox group) indrug-resistantHepG2/ADRcells, andthenledtoimproved
in vitro anti-tumor efficiency in HepG2/ADR cells (IC50 only 19.5% of unmodified ones) as well as exciting in vivo anti-
tumor efficiency on HepG2/ADR heterotopic tumor nude mice (1.75-fold higher TGIR than free drug).
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
Drug resistance; P-glycoprotein; Mitochondria; Reactive oxygen species; HPMA copolymer; Gelatin nanoparticle
2
ACS Paragon Plus Environment

Page 3 of 27
Biomacromolecules
1
2
3
4
5
6
1. Introduction
7
8
9
Multidrug resistance (MDR), known as a serious clinical chemotherapeutic problem in effective tumor therapy,
is often a result of up-regulation of P-glycoprotein (P-gp).1, 2 P-gp has been identified to actively promote the
efflux of drugs (e.g., doxorubicin, paclitaxel) from cytoplasm before they reach targets and take effect in drug
resistant cells.3, 4 Strenuous efforts have been made to overcome MDR. One strategy is developing drug delivery
systems as studies have reported that polymers and nanoparticles can bypass P-gp efflux through endocytosis
pathway.5-7 Another strategy is combining anti-cancer drug with P-gp inhibitor. However, currently their clinical
administrations are still challenged mainly due to intolerable toxicity, non-specific action, drug metabolism.2, 8
Therefore, considerable effort remains to be made to inhibit P-gp function or bypass P-gp efflux for tackling MDR.
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
It had been reported that specifically increasing mitochondrial reactive oxygen species (ROS) level can oxidize
NADH into NAD+, leading P-gp pumps dysfunction through insufficient adenosine triphosphate (ATP) supply, thus
rendering resistant cells more susceptible to anticancer drugs.9 Hence, mitochondrial ROS were selected and
widely studied as a promising endogenous P-gp inhibitor for overcoming MDR.10, 11 Studies have demonstrated
that reductive activation of doxorubicin (Dox) aglycone structure results in semiquinone radicals formation in
mitochondria, which are strong ROS that can lead to ROS-mediated cell apoptosis. Prospectively, mitochondria
targeting delivery of Dox to produce high level of ROS potentially offers an opportunity to overcome MDR.
N-(2-hydroxypropyl) methacrylamide (HPMA) copolymer, is a promising small-sized carrier due to its
biocompatibility, good tumor penetration and increase of hydrophobic drug solubility.12, 13 Our previous studies
had demonstrated that HPMA copolymers with a size of 10-20 nm could achieve deep tumor penetration,14
meanwhile their in vivo delivery and tumor accumulation efficiency could be further improved after the size of
copolymer enlarged by being prepared to micelles or conjugated onto a nanoparticle core. 15-18 Gelatin is a metal
matrix protease-2 (MMP2) substrate which overexpressed in tumors.19, 20 Our another previous study had
demonstrated that gelatin nanoparticles (GNPs) were stable in blood circulation and quickly degraded by
extracellular MMP2 once accumulated in hepatocellular tumor, leading to small amphiphilic prodrug release,
followed by penetrated deeply into tumor tissues.21 Based on these results, HPMA copolymer modified GNPs
would be a desirable carrier for mitochondria targeting of anti-tumor prodrug.
To achieve mitochondria targeting delivery, mitochondria-targeting signal peptides and lipophilic cations
(e.g., triphenylphosphonium, dequalinium) have been used.22-24 However, as some mitochondria targeting
groups do not have significant tumor cell targeting ability or can not enhance cellular uptake of drugs, the
3
ACS Paragon Plus Environment

Biomacromolecules
Page 4 of 27
1
2
3
4
5
6
7
8
9
simultaneous use of cell targeting peptide or cell penetrating peptide as well as mitochondria targeting ligand in
drug delivery systems was developed.25 Another strategy is to use ligand which is able to increase both cellular
uptake and mitochondria targeting.26 Glycyrrhetinic acid (GA) is capable of specifically recognizing protein kinase
C α, which highly expressed in hepatocellular carcinoma (HCC) cells.27-29 More importantly, rencently it was
reported that GA-functionalized graphene oxide significantly enhanced drug distribution in mitochondria.30
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
Therefore, GA has great potential as a sequential-targeting ligand from cell membrane to mitochondria. However,
it remains challenging for GA-conjugated nanoparticles achieving mitochondria targeting due to the outer
mitochondrial membrane only allow passage of molecules less than 5 KDa.31 Hence, smart drug delivery system
which can release mitochondria targeting GA-drug conjugates after tumor cell internalization should be
developed.
Herein, we proposed a sequential tumor cell targeting and mitochondria targeting delivery system (GNPs-P-
Dox-GA) for overcoming MDR through ROS generation. Firstly, doxorubicin was conjugated with GA (GA-Dox) and
then grafted onto HPMA copolymer backbone via hydrazone bond (P-Dox-GA). The obtained P-Dox-GA was
subsequently attached to the surface of gelatin nanoparticles (GNPs) (Scheme 1). The GNPs-P-Dox-GA
nanoparticles would stay stable in blood circulation, and then degraded by tumor extracellular MMP2 and release
smaller size P-Dox-GA to facilitate tumor tissue penetration. After P-Dox-GA internalized by tumor cells under
GA mediation and delivered to lysosomes, Dox-GA detached from HPMA copolymer through hydrolysis of
hydrazone bonds and then efficiently delivered to mitochondria. Then, a large number of ROS would be produced
by mitochondria damage, causing dysfunction of P-gp pumps and cell apoptosis for overcoming MDR. In this
study, cellular uptake, mitochondria distribution, ROS product efficacy, P-gp activities, as well as anti-tumor
effects on MDR tumor were investigated. GA modification was used as a “one stone two birds” strategy for
realizing effective cellular uptake as well as mitochondria targeting. Besides, this study gives insights into
overcoming MDR through enhanced ROS generation in mitochondria, not only inhibit drug efflux but also cause
apoptosis on MDR cells.
4
ACS Paragon Plus Environment

Page 5 of 27
Biomacromolecules
1
2
3
4
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. (1) Triggered by tumor extracellular metal matrix protease-2 (MMP2), small size HPMA copolymer
P-Dox-GA released from gelatin nanoparticles GNPs-P-Dox-GA, facilitating tumor tissue penetration. (2) After
internalized by tumor cells under the mediation of GA and delivered to lysosomes, small-molecular Dox-GA
detached from HPMA copolymer through hydrolysis of hydrazone bonds and (3) then efficiently delivered to
mitochondria, where extensive reactive oxygen species (ROS) produced by mitochondria damage, causing
dysfunction of P-glycoprotein pumps and ROS-mediated cell apoptosis for overcoming multidrug resistance.
2. Materials and Methods
2.1 Materials
Gelatin (type A), 3-(4,5-dimethyl-2-tetrazolyl)-2,5-diphenyl-2H tetrazoliumbromide (MTT), 4’,6-diamidino-2-
phenylindole (DAPI) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Doxorubicin hydrochloride
(Dox·HCl) was from Dalian Meilun Biotech Co., Ltd. (Dalian, China). BCA Protein Assay Kit, Cell Mitochondria
Isolation Kit and Reactive Oxygen Species Assay Kit were purchased from Beyotime Institute of Biotechnology
(China). MitoTracker Green was purchased from Invitrogen (Carlsbad, CA, USA). Glycyrrhetinic acid was got from
J&K Co., Ltd. (Shanghai, China). Comonomer HPMA32 and N-methacryloylglycylglycyl-hydrazide-doxorubicin (MA-
GG-NHN=Dox)16 were obtained by previous synthetic methods.
5
ACS Paragon Plus Environment

Biomacromolecules
Page 6 of 27
1
2
3
4
5
6
2.2 Synthesis of HPMA copolymers
2.2.1 Synthesis of initiator ABIK-PDS
7
8
9
2-(2-pyridyldithio) ethylamine hydrochloride (PDEA, 17.1 mmol, synthetic details were described in
supplementary materials) and 4,4'-Azobis (4-cyanovaleric acid) (ABIK, 8.53 mmol) were dispersed in chloroform
(100 mL). To this suspension, triethylamine (17.1 mmol) was added under vigorous stirring to form a clear
solution. Subsequently, dicyclohexylcarbodiimide (DCC, 20.3 mmol) was added slowly to the reaction mixture,
which was kept stirring for 24 h. The mixture was concentrated and purified by silica-gel column chromatography
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
1
(ethyl acetate) to afford ABIK-PDS in a yield of 66.4%. H NMR spectra and mass spectra were conducted to
identify the product.
2.2.2 Synthesis of glycyrrhetinic acid derivative suc-GA
Glycyrrhetinic acid (2.5 mmol), succinyl anhydride (10.0 mmol) and 4-dimethylaminopyridine (DMAP, 5.0
mmol) were dissolved in dichloromethane (DCM). The reaction mixture was refluxed for 12 h, then washed with
saturated sodium chloride solution. The organic layer was concentrated and recrystallized in ethanol to afford
the glycyrrhetinic acid derivative (suc-GA) in a yield of 67.8%. 1H NMR spectra and mass spectra were conducted
to identify the product.
2.2.3 Synthesis of HPMA copolymers
The telechelic HPMA copolymer doxorubicin conjugate with a thiol end group (P-Dox) was synthesized by
its precursor HPMA copolymer doxorubicin conjugate containing 2-pyridyldisulfanyl (PDS) terminal (P-Dox-PDS).
Firstly, P-Dox-PDS was synthesized by radical solution polymerisation in dimethylsulfoxide (initiator ABIK-PDS,
5 wt%; monomer 12.5 wt%; HPMA and MA-GG-NHN=Dox in a molar ratio of 70 to 30). The mixture was reacted
under nitrogen for one day (70°C, in dark). The reaction mixture was precipitated in acetone. The precipitate
was redissolved in water and filtered by 0.22 μm microporous membrane. The filtrate was dialysed against
deionized water for 24 h and freeze-dried to afford the HPMA copolymer precursor P-Dox-PDS in a yield of
66.2%. Then, the obtained P-Dox-PDS was dissolved in dithothreitol solution (DTT, 0.1 M) which was gently
stirred at room temperature. The reaction lasted for 30 min, then the mixture was dialysed against deionized
water for two days, filtered (0.22 μm) and freeze-dried. The desired telechelic copolymer was obtained in a
yield of 89.2%.
HPMA copolymer glycyrrhetinic acid derivative conjugate containing a thiol end group (P-Dox-GA) was
synthesized by two steps. Firstly, glycyrrhetinic acid derivative suc-GA (0.276 mmol) was completely dissolved
in N,N-dimethyllformamide. N-hydroxysuccinimide (NHS, 0.414 mmol) and 1-(3-dimethylaminopropyl)-3-
6
ACS Paragon Plus Environment

Page 7 of 27
Biomacromolecules
1
2
3
4
5
6
7
8
9
ethylcarbodiimide hydrochloride (EDC·HCl, 0.414 mmol) were added subsequently to the solution. After 3 hours’
activation, the reaction mixture was added with previously synthesized P-Dox (0.092 mmol Dox equivalent).
The solution was stirred for 24 h in dark before being dialyzed against deionized water for 2 days. After filtration
(0.22 μm) and freeze-drying, the copolymer P-Dox-GA was obtained in a yield of 63.0%.
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
2.3 Preparation of gelatin nanoparticles
2.3.1 Synthesis of 3-(2-pyridyldithio) propionic acid N-hydroxy-succinimide ester (SPDP)
2,2’-Dipyridyldisulfide (3.75 g) and glacial acetic acid (400 µL) was dissolved in methanol. 3-
mercaptopropionic acid (0.9 g in 5 mL methanol) was added dropwise to this solution. After 24 h reaction, the
solution was concentrated and separated by alumina column chromatography (dichloromethane/ethanol = 2/3,
v/v). Eluent containing the product was collected and concentrated. The residue was recrystallized in ethyl
acetate/hexane to afford the product as a pale solid. Then 5.0 mmol of the obtained product was dissolved in
DCM. The solution was added with NHS (7.5 mmol) and DCC (7.5 mmol) subsequently and reacted for 4 h, then
kept in 4°C overnight without stirring. The precipitation, dicyclohexylurea (DCU), was removed by filtration and
the filtrate was concentrated. The residue was recrystallized in ethanol to afford the product as a white solid.
1H NMR spectra and mass spectra were conducted to identify the product (SPDP).
2.3.2 Preparation of GNPs-P-Dox-GA and GNPs-P-Dox
P-Dox-GA modified gelatin nanoparticles GNPs-P-Dox-GA were prepared by two steps. Firstly, gelatin
nanoparticles (GNPs) were prepared as previously described.21 The GNPs solution (20 mg/mL, 4 mL) was adjusted
to pH 8.0, followed by the addition of SPDP (8 mg in dimethylforamide). After 24 h’s reaction, unreacted SPDP
was removed by dialysis (dimethylforamide/phosphate buffer solution: 1/4 for one day, phosphate buffer
solution for another day). Then gelatin nanoparticles with PDS terminal (GNPs-PDS) were harvested as solution.
Secondly, the above GNPs-PDS solution (20 mg/mL, 4 mL) was adjusted to pH 6.0 and added with P-Dox-GA
(0.015 mmol thiol group equivalent). 8 hours later, the mixture was dialyzed against phosphate buffer solution
for two days to afford the GNPs-P-Dox-GA solution. P-Dox modified gelatin nanoparticles GNPs-P-Dox was
prepared as a control by the same procedure.
2.4 Characterization of gelatin nanoparticles and HPMA copolymers
The hydrate size as well as zeta potential of gelatin nanoparticles and HPMA copolymers were characterized
by Zetasizer Nano ZS90 (Malvern, UK). GNPs-P-Dox-GA morphology was imaged by transmission electron
microscopy (TEM, H-600, Hitachi, Japan). Thiol group content in HPMA copolymers and PDS content on gelatin
nanoparticle GNPs-PDS were measured as previously reported.18 Dox loading in HPMA copolymers and
7
ACS Paragon Plus Environment

Biomacromolecules
Page 8 of 27
1
2
3
4
5
6
nanoparticles were quantified by UV-Vis analysis at 487 nm. For stability study, GNPs-P-Dox-GA size was
characterized in pH 7.4 phosphate buffer solution (PBS) by dynamic light scanning (DLS) at predetermined time
points.
7
8
9
2.5 In vitro drug release
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
The dialysis method was used to evaluate the release profile of Dox-GA from GNPs-P-Dox-GA. GNPs-P-Dox-
GA (0.4 mg/mL, 3 mL) was sealed into a dialysis bag (MWCO 3000), followed by incubation with buffer solution
under various pH conditions at 37C (pH 7.4 mimicking physiological condition, pH 6.5 mimicking tumor
microenvironment and pH 5.0 mimicking lysosome environment). At scheduled intervals, 200 μL of buffer solution
was collected from the outer phase of the dialysis bag, and replaced with 200 μL of fresh buffer solution. The
drug concentration was calculated by the fluorescence intensity of Dox (Ex=487nm, Em=533nm) measured by
Varioskan Flash Multimode Reader (Thermo Fisher Scientific, MA, USA).
2.6 Cell lines and mice
Human hepatoma cell line (HepG2) and multiple drug resistant HepG2/adriamycin cell line (HepG2/ADR)
were both cultured in RPMI 1640 medium (37C, 5% CO2), supplemented with 10% fetal bovine serum and 1%
penicillin (100 IU/mL)/streptomycin (100 μg/mL). The culture medium of HepG2/ADR was supplemented with
doxorubicin (0.5 μg/mL) to maintain the drug resistant phenotype.
Male BALB/c nude mice (4-6 weeks) were obtained from Chengdu Dashuo Biological Technology Co., Ltd.
(Chengdu, China). All the mice were housed under the guidelines of Sichuan University.
2.7 Cellular uptake
HepG2 and HepG2/ADR cell were seeded on 6-well plates (1×105 cells/well). After 24 h, cells were treated
with a series of gelatin nanoparticles and HPMA copolymers at an equivalent dose (Dox: 10 μg/mL) for 1 h or 4
h. Next, cells were washed thrice with fresh PBS. The extracellular fluorescence was quenched using trypan blue
(0.4%). Then, cells were incubated with RIPA lysis buffer and collected. After centrifugation (3000 rpm, 3 min),
the supernatant was obtained and the total protein in each sample was measured by BCA protein assay kit.33, 34
The fluorescence intensity of Dox was detected as described above.
For confocal laser scanning microscope (CLSM) imaging, HepG2/ADR cells were cultured on glass c
overslips for 24 h. Cells were then treated with gelatin nanoparticles and HPMA copolymers for 4 h
at an equivalent dose (10 μg/mL Dox). After rinsed thrice with PBS, cells were fixed with 4% formald
ehyde, and stained by 4',6-diamidino-2-phenylindole (DAPI). Finally, cells were visualized by CLSM (LSM
8
ACS Paragon Plus Environment

Page 9 of 27
Biomacromolecules
1
2
3
4
800, Zeiss, Germany).
5
6
2.8 Mitochondria targeting
7
8
9
For quantitative analysis of mitochondria targeting, HepG2/ADR cells were cultured in 6-well plates (1×105
cells/well), then treated with gelatin nanoparticles and HPMA copolymer conjugates for 1 h or 4 h at an
equivalent dose (Dox: 10 μg/mL). Cells were collected after washing with fresh PBS thrice. Mitochondria obtained
by Cell Mitochondria Isolation Kit were lysed by incubation with mitochondria lysis buffer. After centrifugation
(10000 g◊10 min, 4°C), total protein in the collected supernatants were immediately detected by a BCA protein
assay kit. The fluorescence intensity of Dox was detected as described above.
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
For CLSM imaging, HepG2/ADR cells (1×104 cells/well) were cultured for 24 h on glass coverslips. After
exposure to gelatin nanoparticles and HPMA copolymer conjugates for 4 h at an equivalent dose (Dox: 10 μg/mL),
cells were stained with MitoTracker green (100 nM) and DAPI. Finally, cells were observed using CLSM.
2.9 ROS level
HepG2/ADR cells (2×105 cells/well) were cultured in 12-well plates. After exposure to gelatin nanoparticles
and HPMA copolymer conjugates for 4 h at an equivalent dose (Dox: 10 μg/mL), cells were washed and collected.
The ROS level was determined by Reactive Oxygen Species Assay Kit and flow cytometer according to the
instructions.
2.10 P-glycoprotein activities
HepG2/ADR cells (1×105 cells/well) were cultured for 24 h in 24-well plates. Cells were then treated with
gelatin nanoparticles and HPMA copolymer conjugates at an equivalent dose (Dox: 10 μg/mL) for 4 h. Then the
medium was replaced by calcien AM (0.2 μM). After 30 min, cells were washed and lysed using RIPA lysis buffer.
After centrifugation (3000 rpm◊10 min, 4C), the total protein in the collected supernatant was detected as
described above, The fluorescence intensity of calcien AM in supernatants was determined via Varioskan Flash
Multimode Reader.
2.11 Drug efflux assay
HepG2/ADR cells (1×105 cells/well) were cultured for 24 h on 24-well plates. After treated with gelatin
nanoparticles and HPMA copolymer conjugates at an equivalent dose (Dox: 10 μg/mL) for 4 h, the cells were
then treated with fresh RPMI-1640 medium for another 4 h. Afterwards, extracellular fluorescence was
9
ACS Paragon Plus Environment

Biomacromolecules
Page 10 of 27
1
2
3
4
5
6
7
8
9
quenched with trypan blue (0.4%) and the cells were lysed by RIPA buffer incubation. After centrifugated (3000
rpm◊10 min, 4C), the total protein and fluorescence intensity of Dox in the collected supernatant were detected
as described above. Drug efflux ratio was calculated by the following formula: efflux ratio = (CA-CB)/CA×100%, (CA,
fluorescence intensity of cell treated with sample for 4 h; CB, fluorescence intensity of cell cultured for another
4 h with fresh medium after drug incubation).
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
2.12 In vitro antitumor effect
MTT assay was performed to determine the cytotoxicity. HepG2 cells and HepG2/ADR cells were cultured in
96-well plates (4×103 cells/well) for 24 h, then treated with gelatin nanoparticles and HPMA copolymer
conjugates for 48 h. Subsequently, the medium was replaced by fresh medium (180 µL/well) and 5 mg/mL MTT
solution (5 mg/mL, 20 µL/well). After 4 h incubation, the medium was replaced by dimethyl sulfoxide (150 µL/well)
and vibrated. Absorbance was assayed via Varioskan Flash Multimode Reader at 570 nm.
Apoptosis was detected by FITC-Annexin V Apoptosis Kit (BioLegend, USA). In brief, HepG2/ADR cells (1◊105
cells/well) were treated with various HPMA copolymer conjugates and gelatin nanoparticles (equivalent to 2
µg/mL Dox) for 24 h. Cells were rinsed thrice with PBS, then resuspended in binding buffer (100 µL) containing
2.5 µL annexin-V and 2.5 µL 7-amino-actinomycin D, and incubated for 15 min. After addition with 400 µL binding
buffer, the cell apoptosis was immediately monitored by flow cytometry.
2.13 In vivo antitumor effect and histopathological analysis
HepG2/ADR cells (1 ×106 cells) were injected into mice right axilla subcutaneously. Mice were divided into six
groups randomly after HepG2/ADR xenograft tumor volume reached 100 mm3. Then saline or samples (2.5 mg/kg
Dox equivalent) were injected via tail vein on predetermined days. The body weight and tumor size were
recorded every two days. Tumor size was calculated following the formula: V = (L×W2)/2.
After mice sacrificed, main organs and tumors were collected and fixed in 10% formalin, then embedded in
paraffin blocks and cut into slices (4 µm). Tumors were stained with hematoxylin and eosin (H&E) and TUNEL
apoptosis assay kit (Roche). Other organs were only performed with H&E staining.
2.14 Statistical analysis
Data were mean values ± standard deviations (SDs). Statistical analysis was performed using SPSS 17.0
software to execute a two-tailed Student’s t-test. P value < 0.05 was considered statistically significant.
10
ACS Paragon Plus Environment

Page 11 of 27
Biomacromolecules
1
2
3
3. Results and Discussion
4
5
6
3.1 Characterization of gelatin nanoparticles and HPMA copolymer conjugates
7
8
9
Considering C30-carboxyl group in glycyrrhetinic acid (GA) was an active site,35, 36 which could recognize
protein kinase C α on tumor cell membrane, succinate-GA (suc-GA, Figure S1) was synthesized by introducing
another carboxyl group to C3-hydroxyl group.37, 38 Then comonomers of HPMA copolymer (HPMA, MA-GG-
NHN=Dox), PDEA (Figure S2), ABIK-PDS (Figure S3) and HPMA conjugates (P-Dox, P-Dox-GA) were obtained
(Scheme 2). The Dox loading of P-Dox-GA and P-Dox was 8.67 wt% and 9.28 wt%, respectively. Hydrate size of P-
Dox-GA and P-Dox was 18.61 nm and 15.09 nm respectively (Table 1). Zeta potential of P-Dox-GA (+9.86 mV) was
distinctly lower when compared to P-Dox (+20.23 mV), indicating the successful modification of GA.
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
Gelatin nanoparticles (GNPs-P-Dox, GNPs-P-Dox-GA) were synthesized by two steps (Scheme 3). Firstly,
pyridyldisulfanyl (PDS) functionalized GNPs (GNPs−PDS) were obtained by aminolysis of the active ester of SPDP
with amino groups on GNPs. After PDS modification, the PDS loading of GNPs-PDS was 0.187 mmol/g. Meanwhile,
zeta potential changed from -10.53 mV (GNPs) to -27.92 mV (GNPs-PDS) (Table S1). Secondly, thiol-terminated
P-Dox-GA was grafted to GNPs−PDS to form GNPs-P-Dox-GA. Hydrate size of GNPs-P-Dox-GA and GNPs-P-Dox
was 111.1 nm and 106.5 nm respectively (Table 1, Figure 1B), with an increase of 10-20 nm compared to GNPs-
PDS which was not covered by HPMA copolymers, suggesting the successful modification of HPMA copolymers.
TEM result also exhibted that GNPs-P-Dox-GA had a core-shell structure (Figure 1A). All data clearly verified the
successful preparation of polymer grafted nanoparticles. Stability analysis of polymer grafted nanoparticle
(GNPs-P-Dox-GA) was determined in PBS (pH 7.4). The hydrated
11
ACS Paragon Plus Environment

Biomacromolecules
Page 12 of 27
1
2
3
4
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 2. Synthesis of HPMA copolymers P-Dox (A) and P-Dox-GA (B).
Scheme 3. Synthesis of HPMA copolymer modified gelatin nanoparticles GNPs-P-Dox-GA.
12
ACS Paragon Plus Environment

Page 13 of 27
Biomacromolecules
1
2
3
4
Table 1. Characteristics of HPMA copolymers and gelatin nanoparticles (mean ± SD, n = 3).
5
6
7
8
Zeta
potential
(mV)
Thiol group
Size
Dox loading
(wt%)
Carriers
PDI
(-SH) loading
(nm)
(mmol/g NPs)
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
P-Dox
15.09 ± 2.11
18.61 ± 2.51
106.5 ± 3.1
111.1 ± 2.7
/
20.23 ± 2.61
0.10
9.28
8.67
5.7
P-Dox-GA
/
9.86 ± 1.62
-9.87 ± 2.14
-13.92 ± 2.02
0.09
GNPs-P-Dox
GNPs-P-Dox-GA
0.251 ± 0.073
0.276 ± 0.058
/
/
5.1
Abbreviations: PDI, polydispersity index; Dox, doxorubicin;
Figure 1. (A) Transmission electron micrograph images of GNPs-P-Dox-GA. The scale bar was 100 nm. (B)
Hydrate size of GNPs-P-Dox-GA measured by dynamic light scanning. (C) Stability of GNPs-P-Dox-GA in PBS
(pH7.4). (D) Dox-GA release from GNPs-P-Dox-GA in PBS under various pH conditions at 37C.
13
ACS Paragon Plus Environment

Biomacromolecules
Page 14 of 27
1
2
3
4
5
size did not change obviously during 24 h incubation (Figure 1C), implying that GNPs-P-Dox-GA were stable
enough for a long in vivo circulation.
6
7
8
3.2 In vitro drug release
9
It was reported outer mitochondrial membrane (OMM) only allows passage of molecules which are lower
than 5 KDa. Therefore, the release of Dox-GA (1.1 KDa) from GNPs-P-Dox-GA after endocytosis is the prerequisite
for mitochondria accumulation. Hence, Dox-GA release behavior at various pH buffered conditions (pH 7.4
mimicking physiological condition, pH 6.5 mimicking tumor microenvironment and pH 5.0 mimicking lysosome
environment) were determined. In Figure 1D, after 24 h, only a small proportion of Dox-GA released at pH 7.4
(12.3%) or pH 6.5 (26.5%). Inversely, 85.0% of Dox-GA released at pH 5.0. These results suggested that low pH
environment could accelerate Dox-GA release due to hydrolysis of hydrazone in acidic condition. Therefore,
GNPs-P-Dox-GA owned relative stability under physiological environment but could sharply release Dox-GA in
lysosome.
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
3.3 Enhanced cellular uptake by GA
Cellular uptake in HepG2 (HCC cell line) and HepG2/ADR (multiple drug-resistant HepG2/adriamycin cell line)
cells was quantified by BCA kit and imaged by CLSM. GA modified groups (P-Dox-GA and GNPs-P-Dox-GA) showed
much higher internalization in both HepG2 (Figure S5, Table S3 and Table S4) and HepG2/ADR cells (Figure 2,
Table S5 and Table S6). Meanwhile, as incubation time prolonged, the internalization of GA-modified carriers (P-
Dox-GA and GNPs-P-Dox-GA) increased significantly in both HepG2 and HepG2/ADR cells, whereas unmodified
groups did not increase, indicating that advantages of GA modification would be more remarkable with
prolonged incubation time. Besides, compared with that of non-GA modified groups, the cellular uptake of P-Dox-
GA had only approximately 2.84-fold improvement in HepG2 cells (Figure S5B), whereas achieving 4.41-fold
increase in HepG2/ADR cells (Figure 2B), which suggested that GA has a stronger role of enhancing cellular uptake
in drug resistant cells.
14
ACS Paragon Plus Environment

Page 15 of 27
Biomacromolecules
1
2
3
4
5
6
7
8
9
CLSM of HepG2/ADR cells (Figure 3C) showed similar results. Much stronger red fluorescence was observed
in GA modified groups (P-Dox-GA and GNPs-P-Dox-GA) compared with carriers without GA modification. The
results indicated that GA modification could enhance the internalization by HepG2/ADR cells. Furthermore,
subcellular distribution of these groups was obviously different. In P-Dox-GA and GNPs-P-Dox-GA groups, there
was no coincide of blue nuclei and red Dox. In contrast, bright purple area existed in P-Dox and GNPs-P-Dox
groups, indicating the distribution of Dox in nuclei. The results suggested that GA could change the subcellular
distribution of Dox.
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. Cell internalization in HepG2/ADR cells for (A) 1 h and (B) 4 h. **p<0.01. (C) Cellular uptake in
HepG2/ADR cells imaged by CLSM after 4 h treatment. Blue fluorescence presented nuclei. Red fluorescence
presented Dox.
15
ACS Paragon Plus Environment

Biomacromolecules
Page 16 of 27
1
2
3
4
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 3. (A) Mitochondria targeting in HepG2/ADR cells. **p<0.01. (B) Mitochondria colocatization of
carriers by CLSM in HepG2/ADR cells. Red fluorescence: Dox. Green fluorescence: mitochondria stained by
MitoTracker Green. Blue fluorescence: nuclei stained by DAPI. Yellow fluorescence indicates the colocalization
of green mitochondria and red Dox. (C) The ratio of GA modified carriers increased in celluar uptake and
mitochondria targeting to unmodified ones.
3.4 Improved mitochondria targeting by GA
Dox-GA was released after the hydrolyzation of hydrazone bond between Dox-GA and HPMA polymer
backbone. Mitochondria targeting ability of Dox-GA was analyzed qualitatively and quantitatively. Fluorescence
intensities in the mitochondria of cells treated with GA modified groups were significantly stronger than that
treated with unmodified ones (Figure 3A), indicating GA modification indeed enhanced mitochondria
accumulation of Dox. It should be noted that, GA modification led to 6.2-8.8 folds increase in mitochondria
accumulation compared to the unmodified counterparts in 4 h incubation, while only 2.3-4.4 folds increase was
achieved in the cellular uptake (Figure 3C), suggesting that the increased mitochondrial accumulation was not
only owing to the enhanced internalization, but also due to the effective mitochondria targeting ability
introduced by GA modification. Besides, mitochondrial distribution of GA modified groups showed significant
16
ACS Paragon Plus Environment

Page 17 of 27
Biomacromolecules
1
2
3
4
5
6
7
8
9
improvement (1.63-1.75 folds) with prolonged incubation time from 1 h to 4 h. However, in contrast, P-Dox and
GNPs-P-Dox groups showed significantly reduced mitochondrial distribution from 1 h to 4 h. This may be on one
hand due to the fact that the cellular uptake of the GA unmodified groups in the drug-resistant cells did not
increase significantly with prolonged incubation time; on the other hand, the Dox mitochondrial distribution was
reduced due to its nuclear tropism. This result also indicated that GA modification can effectively change the
subcellular distribution of Dox.
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
Fluorescence from CLSM also showed similar results to quantitative experiments (Figure 3B). For groups
without GA modification (P-Dox and GNPs-P-Dox), the green fluorescence of mitochondria hardly overlapped
with the red fluorescence of Dox; while for GA modified groups (P-Dox-GA and GNPs-P-Dox-GA), large areas of
of yellow fluorescence, representing the co-locoalization of mitochondria and Dox, were observed obviously.
These results indicated that Dox-GA possessed effective mitochondria targeting ability.
3.5 Increased ROS generation
ROS have been involved in important developmental approach regulation, such as protein-protein
interactions. As a mutagenic agent, ROS can either block self-renewal or stimulate stem cell differentiation, which
may increase tumor heterogeneity.39 Cells in tumor tissue differ in morphology, the quantities of vesicles,
organelles, and mitochondria, density, size, and cytoskeleton arrangement. According to this, their susceptibility
to ROS damage may also vary. Meanwhile, ROS is implicated in immune reactions, such as macrophage
polarization and activation.40, 41 As reported, the reduce of ROS inhibited repolarization of tumor associated
43
macrophages (TAM) to M1 macrophage.42,
Thus, ROS production played an important role in tumor
microenvironment regulation and tumor characteristic. Meanwhile, as reported, on one hand, ROS can destroy
mitochondrial membrane and cause apoptosis;44, 45 On the other hand, for resistant cells, ROS can also inhibit
the respiratory chain, reduce the generation of energy, and help to overcome energy dependent drug efflux.46
What’s more, delivering antitumor drugs to mitochondria usually leads to ROS production increase. Therefore,
ROS amount can reflect the mitochondria targeting ability. Here, cell-permeant reagent 2’,7’-
dichlorodihydrofluorescein diacetate (DCFH-DA) was used to visualize ROS level. DCFH-DA travels easily across cell
membrane with no significant fluorescence intensity in normal status, while once was oxidated by ROS in cells, DCFH-
DA would transform into 2’,7’-dichlorofluorescein (DCF) accompanied by detectable green fluorescence intensity. In
Figure 4A, P-Dox and GNPs-P-Dox produced little ROS, while ROS level in P-Dox-GA and GNPs-P-Dox-GA treated
cells was 3.09-fold and 3.33-fold higher than that of P-Dox and GNPs-P-Dox treated cells, respectively. Meanwhile,
no significant difference existed in the amount of ROS produced among P-Dox-GA and GNPs-P-Dox-GA group, or
among P-Dox and GNPs-P-Dox group. All of the above results were consistent with their mitochondrial targeting
17
ACS Paragon Plus Environment

Biomacromolecules
Page 18 of 27
1
2
3
4
abilities, indicating that the improvement of Dox mitochondrial accumulation indeed increased ROS production.
5
6
3.6 Overcoming multidrug resistance by high ROS level
7
8
9
To investigate whether the increased ROS could contribute to overcome multidrug resistance, P-gp activity
and Dox efflux ratio in HepG2/ADR cells were investigated. At first, cells were treated with fresh medium or
samples. After 4 h incubation, samples were replaced by calcein AM which is a substrate of P-gp for another 30
min. Stronger cellular fluorescence indicates weaker P-gp activity. Figure 4B showed that intracellular calcein
fluorescence intensity in GA unmodified carrier groups was only slightly higher than control group, whereas
the fluorescence intensity was enhanced by about 1.5-fold after treatment of GA modified groups (P-Dox-GA
and GNPs-P-Dox-GA), indicating that GA-modification decreased P-gp activity. This may be attributed to the
objective fact that GA mediated mitochondria targeting of Dox and produced more ROS, resulting in reduced
P-gp activity. ROS has recently been verified to be integral to the regulation of multidrug resistance, and it
regulated the oxidation of NADH into NAD+ to decrease available ATP amount, following down-
regulate expression of P-gp.46 Therefore, the increasing of ROS level could account for the lower P-gp activity,
and then increasing Dox retention in HepG2/ADR cells would happen.
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
Afterwards, Dox efflux ratio was evaluated by determining retained Dox fluorescence in HepG2/ADR
cells. A lower efflux ratio suggested lower P-gp activity. Up to 62.5% of free Dox was pumped out of cells
(Figure 4C), verifying free drugs are highly impressionable to P-glycoprotein on cell membrane through
diffusion. Comparatively, the Dox efflux rate in P-Dox and GNPs-P-Dox reduced, which confirmed the previous
reports,47 that HPMA carriers could efficiently decrease drug efflux through endocytosis. More surprisingly,
P-Dox-GA and GNPs-P-Dox-GA reduced efflux rate to 38.1% and 34.2%, which were only 61% and 55%
compared with that of free Dox group, indicating the synergism of endocytosis and down-
regulated expression levels of P-glycoprotein. Gelatin nanoparticles and HPMA copolymer conjugates were
internalized into cells in membrane-limited organelles avoiding partition onto P-gp. Then, Dox-GA or Dox was
released into cytoplasm from late endosomes and lysosomes in the cell perinuclear region (far from P-gp). As a
result, the efflux ratio in these carrier groups was all lower than free Dox. Then base on the mitochondria
targeting ability of GA, the released Dox-GA was delivered into mitochondria and produced more ROS (Figure
4A), which resulted in lower P-gp activity (Figure 4B). Hence, compare GA-modified and unmodified samples, the
major efflux reduction (Figure 4C) should be due to the increase of ROS which was a result of GA mitochondrial
targeting. All the results indicated that specifically produce ROS in mitochondria could reduce drug efflux by
down-regulate activity of P-glycoprotein, which has a promising application potential in overcoming
multidrug resistance.
18
ACS Paragon Plus Environment

Page 19 of 27
Biomacromolecules
1
2
3
4
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
Figure 5. (A) HepG2/ADR cell viabilities after 48 h incubation. (B) IC50 values of HPMA copolymers and gelatin
nanoparticles against HepG2/ADR cells after 48 h incubation. **p<0.01.
Figure 4. (A) ROS production in HepG2/ADR cells after 4 h incubation. (B) P-glycoprotein activity in
&
&&
HepG2/ADR cells after 4 h incubation. (C) Drug efflux in HepG2/ADR cells. ** indicated p<0.01. and
#
##
indicated p<0.05 and p<0.01 versus control group. and indicated p<0.05 and p<0.01 versus Dox group,
respectively.
3.7 In vitro antitumor activity
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Dox could interfere with DNA replication through inhibiting DNA topoisomerase II and gyrase enzymes, which has
been selected as first-line chemotherapy for hepatocellular carcinoma.48 However, most of cellular Dox would be
pumped out by P-gp, which resulted in low intracellular concentration and scarce chemotherapy efficacy in drug
resistant cells. The Dox cytotoxicities in HepG2 and its drug resistant cells were firstly investigated and compared. As
displayed in TableS2, freeDox showed remarkablecytotoxicity in HepG2 cells (IC50= 0.18 μg/mL), whereas its antitumor
activity was largely reduced in HepG2/ADR cells (IC50 = 12.05 μg/mL) with resistance index (RI) of 66.94.
19
ACS Paragon Plus Environment

Biomacromolecules
Page 20 of 27
1
2
3
4
5
6
7
8
9
As reported, there was redox balance in mitochondria, which could be broken by rapidly increasing ROS levels,
inducing mitochondria dysfunction and apoptosis development. Studies stated that the reductive activation of
aglycone structure of Dox result in semiquinone radical formation in mitochondria, which are strong ROS that
can cause cell apoptosis. Although some studies have reported that delivery systems based on polymer and
nanoparticles can bypass P-gp efflux through endocytosis, the therapeutic effect of P-Dox and GNPs-P-Dox on
HepG2/ADR cells was still unsatisfied (Figure 5A and 5B). Cell viability of GA modified HPMA copolymer conjugates and
nanoparticle was also investigated. Excitingly, P-Dox-GA and GNPs-P-Dox-GA displayed effective antitumor activities
against HepG2/ADR cells. What’s more, the IC50 of P-Dox-GA and GNPs-P-Dox-GA were 84.5% and 80.5% lower than P-
Dox and GNPs-P-Dox (Figure 5B), indicating superior cytotoxicity after GA modification. The apoptosis results on
HepG2/ADR cells also demonstrated that the both P-Dox-GA and GNPs-P-Dox-GA induced more cell apoptosis after GA
modification (Figure S6). In addition, there was no obvious difference in the cytotoxic effect between P-Dox-GA and
GNPs-P-Dox-GA on HepG2/ADR cells, possibly due to their similar ROS production levels. However, on non-resistant
HepG2 cells, GA did not significantly increase the cytotoxicity of drug carriers (Figure S7). This result implied that
delivering Dox to mitochondria to reduce P-gp activity and drug efflux was a more effective strategy for drug-resistant
cells than normal cells where P-gp are not overexpressed. On the whole, drug mitochondria targeting is an effective
strategy for drug resistant tumor treatment.
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
ACS Paragon Plus Environment

Page 21 of 27
Biomacromolecules
1
2
3
4
5
6
7
8
A
B
1.1
1800
1600
1400
1200
1000
800
600
400
200
0
Saline
Dox
P-Dox
P-Dox-GA
GNPs-P-Dox
GNPs-P-Dox-GA
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
**
9
38.4%
*
41.4%
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
*
*
*
47.2%
57.3%
67.2%
&
&
##
##
&
##
&
##
&
##
1
3
5
7
9
11 13 15 17 19 21
Time (day)
C
D
E
Figure 6. (A) Tumor growth on HepG2/ADR tumor-bearing mice for 21 days (n=6). Arrow: saline or drug
treatment. (B) Tumor weight and tumor growth inhibition rate (TGIR) calculated at the last day. ##p<0.01
versus free Dox. &p<0.01 versus saline. *p<0.05, **p<0.01. (C) Tumor tissues from HepG2/ADR xenografted
nude mice stained by H&E (200×) and TUNEL (400×) after 21 days treatment. (D) Body weight change curves
during treatment for 21 days. (E) Main organs stained by H&E staining (200×) after 21 days treatment.
3.8 In vivo antitumor activity
56
57
58
59
60
21
ACS Paragon Plus Environment

Biomacromolecules
Page 22 of 27
1
2
3
4
5
6
7
8
9
Although overcoming tumor cell MDR through enhanced ROS generation has been reported, their in vivo
antitumor activity on MDR tumor models especially on HCC tumor model was still barely studied. Here, in vivo
antitumor activities were evaluated on HepG2/ADR tumor bearing male nude mice, the tumor volume in saline
group increased almost 13.8-fold after 21 days (Figure 6A), while the tumor growth was inhibited by all Dox
carriers. The tumor growth inhibition rate (TGIR) of free Dox was only about 38.4% (Figure 6B), illustrating that
the in vivo anticancer efficacy of free Dox was limited due to insufficient tumor accumulation and MDR in cancer
cells. Although studies have demonstrated that gelatin nanoparticles and HPMA copolymer conjugates had
higher tumor accumulation than free drug,21 the tumor inhibition abilities of P-Dox and GNPs-P-Dox were still
unsatisfactory (TGIR were 44.1% and 47.2%, respectively), which might be due to their low ROS level and in vitro
cytotoxicity. Fortunately, compared with P-Dox, P-Dox-GA (TGIR = 57.3%) exhibited much stronger anticancer
efficacy, indicating the essential of mitochondria targeting for overcoming multidrug resistance. The most
significant inhibition of tumor growth was achieved by GNPs-P-Dox-GA, of which the tumor inhibition rate was up
to 67.2% and 1.75-fold higher than free Dox.
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
Apoptosis or necrosis of tumor cells was investigated by TUNEL and H&E histochemistry. Apoptosis cells were
stained to brown by TUNEL, while normal cells were stained to blue. Normal cells nuclei were stained to purple by H&E
and cell plasma to pink, while necrotic cells to pink. Free Dox exhibited a modest increase in apoptosis cells and
necrosis cells percentages compared to saline treated group (Figure 6C), due to poor tumor accumulation
efficiency and activity in MDR cells. P-Dox and GNPs-P-Dox induced a little more apoptosis and necrosis than free
Dox, while P-Dox-GA or GNPs-P-Dox-GA treated group showed significantly enhanced apoptosis and necrosis.
These results clearly illustrated that P-Dox-GA and GNPs-P-Dox-GA by producing ROS in mitochondria and
inducing apoptosis.
Furthermore, vital organs such as hearts, livers, spleens, lungs and kidneys did not show any histological
changes (Figure 6E). Meanwhile, body weight as well as physical state of nude mice were normal after treatments
(Figure 6D). These results suggested low in vivo toxicities of HPMA copolymer conjugates and gelatin
nanoparticle-based therapy. Therefore, it is an effective strategy for drug resistant tumor treatment to deliver drugs
to mitochondria.
4. Conclusions
In summary, to overcome hepatocellular carcinoma MDR, we developed a mitochondria targeting drug delivery
system (GNPs-P-Dox-GA) with enhanced cellular uptake, which consists of a gelatin nanoparticle core and HPMA
22
ACS Paragon Plus Environment

Page 23 of 27
Biomacromolecules
1
2
3
4
5
6
7
8
9
copolymer P-Dox-GA shell. In this study, glycyrrhetinic acid (GA) was confirmed to have two abilities of hepatocellular
carcinoma cells targeting and mitochondria targeting, eventually delivered doxorubicin (Dox) to mitochondria and
produce a large number of ROS. Compared to non-GA modified carriers, GNPs-P-Dox-GA exhibited increased cellular
uptake nearly for 4-fold and mitochondrial distribution nearly for 9-fold on drug-resistant HepG2/ADR cells, as well as
increasedROSlevelfornearly3-fold. Moreimportantly, asaresult, GNPs-P-Dox-GAeffectivelyovercamethedrugefflux,
leading to enhanced in vitro anti-tumor efficiency on HepG2/ADR cells (IC50 only 19.5% of unmodified ones) as well as
exciting in vivo anti-tumor efficiency on HepG2/ADR heterotopic tumor nude mice (1.75-fold tumor growth inhibition
rate than free Dox group). Based on the above advantages, the GNPs-P-Dox-GA would be promising for clinical anti-
MDR treatment, and enhancing ROS generation by mitochondria targeting is an effective strategy to overcome tumor
MDR.
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
Associated Content
Supporting Information
Synthesis and characterization of suc-GA, PDEA, ABIK-PDS, and SPDS. Characteristics of GNPs and GNPs-PDS.
Quantitative analysis of cellular uptake in HepG2 cells. Cytotoxicity of gelatin nanoparticles against HepG2 cells.
IC50 of Dox against HepG2 and HepG2/ADR cells. Apoptosis in HepG2/ADR cells.
Author information
*E-mail: huangyuan0@163.com. Tel/Fax: +86 2885501617.
Notes
There are no conflicts to declare.
Acknowledgments
We acknowledge financial supports from the National Natural Science Foundation for Distinguished Young Scholars
(81625023), the National Natural Science Foundation of China (81603042), and Sichuan Science and Technology
Program (2018SZ0032).
23
ACS Paragon Plus Environment

Biomacromolecules
Page 24 of 27
1
2
3
4
References
1. Zhang, Y.-K.; Wang, Y.-J.; Gupta, P.; Chen, Z.-S., Multidrug resistance proteins (MRPs) and cancer therapy. AAPS J. 2015,
17, (4), 802-812.
5
6
2. Kathawala, R. J.; Gupta, P.; Ashby Jr, C. R.; Chen, Z.-S., The modulation of ABC transporter-mediated multidrug
resistance in cancer: a review of the past decade. Drug Resist. Updates 2015, 18, 1-17.
7
8
9
3. Chen, Z.; Shi, T.; Zhang, L.; Zhu, P.; Deng, M.; Huang, C.; Hu, T.; Jiang, L.; Li, J., Mammalian drug efflux transporters of
the ATP binding cassette (ABC) family in multidrug resistance: A review of the past decade. Cancer Lett. 2016, 370, (1), 153-
164.
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
4. Ambudkar, S. V.; Kimchi-Sarfaty, C.; Sauna, Z. E.; Gottesman, M. M., P-glycoprotein: from genomics to mechanism.
Oncogene 2003, 22, (47), 7468-7485.
5. Huang, Y.; Cole, S. P.; Cai, T.; Cai, Y., Applications of nanoparticle drug delivery systems for the reversal of multidrug
resistance in cancer. Oncol. Lett. 2016, 12, (1), 11-15.
6. Gupta, P.; Garg, T.; Tanmay, M.; Arora, S., Polymeric drug-delivery systems: role in P-gp efflux system inhibition. Crit.
Rev. Ther. Drug Carrier Syst. 2015, 32, (3), 247-275.
7. Markman, J. L.; Rekechenetskiy, A.; Holler, E.; Ljubimova, J. Y., Nanomedicine therapeutic approaches to overcome
cancer drug resistance. Adv. Drug Delivery Rev. 2013, 65, (13-14), 1866-1879.
8. Palmeira, A.; Sousa, E.; H Vasconcelos, M.; M Pinto, M., Three decades of P-gp inhibitors: skimming through several
generations and scaffolds. Curr. Med. Chem. 2012, 19, (13), 1946-2025.
9. Dharmaraja, A. T., Role of reactive oxygen species (ROS) in therapeutics and drug resistance in cancer and bacteria. J.
Med. Chem. 2017, 60, (8), 3221-3240.
10. Wang, H.; Gao, Z.; Liu, X.; Agarwal, P.; Zhao, S.; Conroy, D. W.; Ji, G.; Yu, J.; Jaroniec, C. P.; Liu, Z., Targeted production
of reactive oxygen species in mitochondria to overcome cancer drug resistance. Nat. Commun. 2018, 9, (1), 562.
11. Kankala, R. K.; Liu, C.-G.; Chen, A.-Z.; Wang, S.-B.; Xu, P.-Y.; Mende, L. K.; Liu, C.-L.; Lee, C.-H.; Hu, Y.-F., Overcoming
multidrug resistance through the synergistic effects of hierarchical pH-sensitive, ROS-generating nanoreactors. ACS
Biomater. Sci. Eng. 2017, 3, (10), 2431-2442.
12. Yang, J.; Kopeček, J., The light at the end of the tunnel—second generation HPMA conjugates for cancer treatment.
Curr. Opin. Colloid Interface Sci. 2017, 31, 30-42.
13. Chytil, P.; Koziolová, E.; Etrych, T.; Ulbrich, K., HPMA Copolymer–Drug Conjugates with Controlled Tumor‐Specific
Drug Release. Macromol. Biosci. 2018, 18, (1), 1700209.
14. Li, L.; Sun, W.; Zhong, J.; Yang, Q.; Zhu, X.; Zhou, Z.; Zhang, Z.; Huang, Y., Multistage nanovehicle delivery system based
on stepwise size reduction and charge reversal for programmed nuclear targeting of systemically administered anticancer
drugs. Adv. Funct. Mater. 2015, 25, (26), 4101-4113.
15. Tang, M.; Zhou, M.; Huang, Y.; Zhong, J.; Zhou, Z.; Luo, K., Dual-sensitive and biodegradable core-crosslinked HPMA
copolymer–doxorubicin conjugate-based nanoparticles for cancer therapy. Polym. Chem. 2017, 8, (15), 2370-2380.
16. Zhou, Z.; Li, L.; Yang, Y.; Xu, X.; Huang, Y., Tumor targeting by pH-sensitive, biodegradable, cross-linked N-(2-
hydroxypropyl) methacrylamide copolymer micelles. Biomaterials 2014, 35, (24), 6622-6635.
17. Zhou, Z.; Xu, X.; Li, L.; Huang, Y., Improvement of anti-tumor abilities on human non-small cell lung carcinoma by
micellization and cross-linking of N-(2-hydroxypropyl) methacrylamide copolymers. J. Drug Targeting 2015, 23, (9), 821-831.
18. Yang, Q.; Li, L.; Sun, W.; Zhou, Z.; Huang, Y., Dual stimuli-responsive hybrid polymeric nanoparticles self-assembled
from POSS-based starlike copolymer-drug conjugates for efficient intracellular delivery of hydrophobic drugs. ACS Appl.
Mater. Interfaces 2016, 8, (21), 13251-13261.
24
ACS Paragon Plus Environment

Page 25 of 27
Biomacromolecules
1
2
3
19. Sharma, P. P.; Sharma, A.; Solanki, P. R., Recent Trends of Gelatin Nanoparticles in Biomedical Applications. Adv.
Nanomater., Springer: 2016; 365-381.
4
5
6
7
8
20. Forsyth, P.; Wong, H.; Laing, T.; Rewcastle, N.; Morris, D.; Muzik, H.; Leco, K.; Johnston, R.; Brasher, P.; Sutherland, G.,
Gelatinase-A (MMP-2), gelatinase-B (MMP-9) and membrane type matrix metalloproteinase-1 (MT1-MMP) are involved in
different aspects of the pathophysiology of malignant gliomas. Br. J. Cancer 1999, 79, (11), 1828-1835.
21. Liu, Y.; Li, L.; Li, L.; Zhou, Z.; Wang, F.; Xiong, X.; Zhou, R.; Huang, Y., Programmed drug delivery system based on
optimized “size decrease and hydrophilicity/hydrophobicity transformation” for enhanced hepatocellular carcinoma
therapy of doxorubicin. Nanomedicine: Nanotechnology, Biology and Medicine 2018, 14, (4), 1111-1122.
22. T Madak, J.; Neamati, N., Membrane permeable lipophilic cations as mitochondrial directing groups. Curr. Trends Med.
Chem. 2015, 15, (8), 745-766.
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
23. Lu, P.; Bruno, B. J.; Rabenau, M.; Lim, C. S., Delivery of drugs and macromolecules to the mitochondria for cancer
therapy. J. Controlled Release 2016, 240, 38-51.
24. Liu, Y.; Li, Q.; Xiong, X.; Zhou, Z., Improved mitochondrial targeting effect of HPMA copolymer by SS20 peptide
mediation and nonendocytosis pathway. J. Pept. Sci. 2019, 25, (2), e3144.
25. Liu, Y.; Li, Q.; Xiong, X.; Huang, Y.; Zhou, Z., Mitochondria-targeting and cell-penetrating peptides-co-modified HPMA
copolymers for enhancing therapeutic efficacy of α-tocopheryl succinate. J. Mater. Chem. B 2018, 6, (46), 7674-7683.
26. Sun, W.; Li, L.; Li, L.-j.; Yang, Q.-q.; Zhang, Z.-r.; Huang, Y., Two birds, one stone: dual targeting of the cancer cell surface
and subcellular mitochondria by the galectin-3-binding peptide G3-C12. Acta Pharmacol. Sin. 2017, 38, (6), 806-822.
27. Ying, T.-H.; Tsai, J.-H.; Wu, T.-T.; Tsai, M.-T.; Su, W.-W.; Hsieh, Y.-S.; Liu, J.-Y., Immunochemical localization of protein
kinase C alpha in the biopsies of human hepatocellular carcinoma. Chin. J. Physiol. 2008, 51, (5), 269-274.
28. Yang, L.; Yang, Z.; Jixiong, H.; Huang, S.; Deng, X.; Miao, X., Significance of the expression of glutameta decarboxylase
65 and protein kinase C in primary hepatocarcinoma tissues. Chin. J. Gen. Surg. 2001, (8), 590-593.
29. Wu, F.; Li, X.; Jiang, B.; Yan, J.; Zhang, Z.; Qin, J.; Yu, W.; Gao, Z., Glycyrrhetinic acid functionalized nanoparticles for
drug delivery to liver cancer. J. Biomed. Nanotechnol. 2018, 14, (11), 1837-1852.
30. Zhang, C.; Liu, Z.; Zheng, Y.; Geng, Y.; Han, C.; Shi, Y.; Sun, H.; Zhang, C.; Chen, Y.; Zhang, L., Glycyrrhetinic acid
functionalized graphene oxide for mitochondria targeting and cancer treatment in vivo. Small 2018, 14, (4), 1703306.
31. Kroemer, G.; Galluzzi, L.; Brenner, C., Mitochondrial membrane permeabilization in cell death. Physiol. Rev. 2007, 87,
(1), 99-163.
32. Etrych, T.; Mrkvan, T.; Chytil, P.; Koňák, Č.; Říhová, B.; Ulbrich, K., N‐(2‐hydroxypropyl) methacrylamide‐based
polymer conjugates with pH‐controlled activation of doxorubicin. I. New synthesis, physicochemical characterization and
preliminary biological evaluation. J. Appl. Polym. Sci. 2008, 109, (5), 3050-3061.
33. Zeng, X.; Zhang, Y.; Nyström, A. M., Endocytic uptake and intracellular trafficking of bis-MPA-based hyperbranched
copolymer micelles in breast cancer cells. Biomacromolecules 2012, 13, (11), 3814-3822.
34. Song, Q.; Li, D.; Zhou, Y.; Yang, J.; Yang, W.; Zhou, G.; Wen, J., Enhanced uptake and transport of (+)-catechin and (−)-
epigallocatechin gallate in niosomal formulation by human intestinal Caco-2 cells. Int. J. Nanomed. 2014, 9, 2157-2165.
35. Banerji, S.; Wright, A. J.; Noble, M.; Mahoney, D. J.; Campbell, I. D.; Day, A. J.; Jackson, D. G., Structures of the Cd44–
hyaluronan complex provide insight into a fundamental carbohydrate-protein interaction. Nat. Struct. Mol. Biol. 2007, 14,
(3), 234-239.
36. Schanté, C. E.; Zuber, G.; Herlin, C.; Vandamme, T. F., Chemical modifications of hyaluronic acid for the synthesis of
derivatives for a broad range of biomedical applications. Carbohydr. Polym. 2011, 85, (3), 469-489.
37. Tian, Q.; Wang, X.; Wang, W.; Zhang, C.; Liu, Y.; Yuan, Z., Insight into glycyrrhetinic acid: the role of the hydroxyl group
on liver targeting. Int. J. Pharm. 2010, 400, (1-2), 153-157.
25
ACS Paragon Plus Environment

Biomacromolecules
Page 26 of 27
1
2
3
4
5
6
7
8
9
38. Wang, X.; Gu, X.; Wang, H.; Sun, Y.; Wu, H.; Mao, S., Synthesis, characterization and liver targeting evaluation of self-
assembled hyaluronic acid nanoparticles functionalized with glycyrrhetinic acid. Eur. J. Pharm. Sci. 2017, 96, 255-262.
39. de Sá Junior, P. L.; Câmara, D. A. D.; Porcacchia, A. S.; Fonseca, P. M. M.; Jorge, S. D.; Araldi, R. P.; Ferreira, A. K., The
roles of ROS in cancer heterogeneity and therapy. Oxid. Med. Cell. Longevity 2017, 2017, Article ID 2467940.
40. Tan, H.-Y.; Wang, N.; Li, S.; Hong, M.; Wang, X.; Feng, Y., The reactive oxygen species in macrophage polarization:
reflecting its dual role in progression and treatment of human diseases. Oxid. Med. Cell. Longevity 2016, 2016, Article ID
2795090.
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
41. Nathan, C.; Cunningham-Bussel, A., Beyond oxidative stress: an immunologist's guide to reactive oxygen species. Nat.
Rev. Immunol. 2013, 13, (5), 349.
42. Lee, A. S.; Jung, Y. J.; Kim, D.; Nguyen-Thanh, T.; Kang, K. P.; Lee, S.; Park, S. K.; Kim, W., SIRT2 ameliorates
lipopolysaccharide-induced inflammation in macrophages. Biochem. Biophys. Res. Commun. 2014, 450, (4), 1363-1369.
43. Padgett, L. E.; Burg, A. R.; Lei, W.; Hubert, M. T., Loss of NADPH oxidase–derived superoxide skews macrophage
phenotypes to delay type 1 diabetes. Diabetes 2015, 64, (3), 937-946.
44. Simon, H.-U.; Haj-Yehia, A.; Levi-Schaffer, F., Role of reactive oxygen species (ROS) in apoptosis induction. Apoptosis
2000, 5, (5), 415-418.
45. Indran, I. R.; Tufo, G.; Pervaiz, S.; Brenner, C., Recent advances in apoptosis, mitochondria and drug resistance in cancer
cells. Biochim. Biophys. Acta, Bioenerg. 2011, 1807, (6), 735-745.
46. Pelicano, H.; Carney, D.; Huang, P., ROS stress in cancer cells and therapeutic implications. Drug Resist. Updates 2004,
7, (2), 97-110.
47. Minko, T.; Kopečková, P.; Pozharov, V.; Kopeček, J., HPMA copolymer bound adriamycin overcomes MDR1 gene
encoded resistance in a human ovarian carcinoma cell line. J. Controlled Release 1998, 54, (2), 223-233.
48. Deng, G.-L.; Zeng, S.; Shen, H., Chemotherapy and target therapy for hepatocellular carcinoma: New advances and
challenges. World J.Hepatol. 2015, 7, (5), 787-798.
26
ACS Paragon Plus Environment

Page 27 of 27
Biomacromolecules
1
2
3
For Table of Contents Only
4
5
6
7
8
9
Enhanced Reactive Oxygen Species Generation by Mitochondria
Targeting of Anti-cancer Drug to Overcome Tumor Multidrug
Resistance
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
Yuanyuan Liu#, Zhou Zhou#, Xi Lin, Xiaofeng Xiong, Rui Zhou, Minglu Zhou, Yuan Huang*
Key Laboratory of Drug Targeting and Drug Delivery System, Ministry of Education, West China School of
Pharmacy, Sichuan University, No. 17, Block 3, Southern Renmin Road, Chengdu 610041, P. R. China.
# These authors contributed equally to this work.
27
ACS Paragon Plus Environment