Drug Delivery

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/idrd20

L-EGCG-Mn nanoparticles as a pH-sensitive MRI

contrast agent

Jiali Li, Xue Jiang, Lihuan Shang, Zhen Li, Conglian Yang, Yan Luo, Daoyu Hu,
Yaqi Shen & Zhiping Zhang

To cite this article: Jiali Li, Xue Jiang, Lihuan Shang, Zhen Li, Conglian Yang, Yan Luo, Daoyu
Hu, Yaqi Shen & Zhiping Zhang (2021) L-EGCG-Mn nanoparticles as a pH-sensitive MRI contrast
agent, Drug Delivery, 28:1, 134-143, DOI: 10.1080/10717544.2020.1862363

To link to this article: https://doi.org/10.1080/10717544.2020.1862363

© 2020 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group.

View supplementary material

Published online: 26 Dec 2020.

Submit your article to this journal

Article views: 792

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=idrd20
DRUG DELIVERY
2021, VOL. 28, NO. 1, 134–143
https://doi.org/10.1080/10717544.2020.1862363

RESEARCH ARTICLE

L-EGCG-Mn nanoparticles as a pH-sensitive MRI contrast agent

Jiali Lia
, Xue Jiangb,c
, Lihuan Shangc, Zhen Lia, Conglian Yangc, Yan Luoa, Daoyu Hua, Yaqi Shena and
Zhiping Zhangc,d,e
a
Department of Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China;
b
Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics and Gene Regulation, Medical Research Center, Sun Yat-sen
Memorial Hospital, Sun Yat-sen University, Guangzhou, China; cTongji School of Pharmacy, Huazhong University of Science and Technology,
Wuhan, PR China; dNational Engineering Research Center for Nanomedicine, Huazhong University of Science and Technology, Wuhan, PR
China; eHubei Engineering Research Center for Novel Drug Delivery System, Huazhong University of Science and Technology, Wuhan,
PR China

ABSTRACT ARTICLE HISTORY

This study aimed to synthesize and characterize L-epigallocatechin gallate (EGCG) complexed Mn2þ Received 22 October 2020
nanoparticle (L-EGCG-Mn), a proof-of-concept pH-sensitive manganese core nanoparticle (NP), and Revised 4 December 2020
compare its magnetic resonance (MR) properties with those of Gd-DTPA, both in vitro and in vivo. Accepted 7 December 2020
Reverse microemulsion was used to obtain the L-EGCG-Mn NPs. The physicochemical properties of
KEYWORDS
L-EGCG-Mn were characterized using dynamic light scattering, transmission electron microscopy, and MRI contrast agent;
near-infrared fluorescence small animal live imaging. The in vitro relaxivity of L-EGCG-Mn incubated nanoparticle; manganese;
with different pH buffer solutions (pH ¼ 7.4, 6.8, 5.5) was evaluated. The T1-weighted MR imaging EGCG; pH sensitivity
(MRI) properties were evaluated in vitro using hypoxic H22 cells as well as in H22 tumor-bearing mice.
Cytotoxicity tests and histological analysis were performed to evaluate the safety of L-EGCG-Mn.
L-EGCG-Mn showed good biocompatibility, stability, pH sensitivity, and tumor-targeting ability.
Moreover, when the pH was decreased from 7.4 to 5.5, the r1 relaxivity of L-EGCG-Mn was shown to
gradually increase from 1.79 to 6.43 mM1s1. Furthermore, after incubation with L-EGCG-Mn for 4 h,
the T1 relaxation time of hypoxic H22 cells was significantly lower than that of normoxic H22 cells
(1788 ± 89 vs. 1982 ± 68 ms, p¼.041). The in vivo analysis showed that after injection, L-EGCG-Mn
exhibited a higher MRI signal compared to Gd-DTPA in H22 tumor-bearing mice (p < .05).
Furthermore, L-EGCG-Mn was found to have a good safety profile via cytotoxicity tests and histological
analysis. L-EGCG-Mn has a good safety profile and pH sensitivity and may thus serve as a potential
MRI contrast agent.

Introduction of tumor tissues is more acidic (6.5–7.0), which decreases

even further in hypoxic regions in vivo (<6.5) (Neri &
Cancer cells rely on the ‘Warburg effect’ for aerobic glycoly-
Supuran, 2011). Engineered nanoparticles (NPs) that respond
sis, leading to the accumulation of high lactate concentra-
to acidic pH (<6.5) and release paramagnet components are
tions, even under aerobic conditions (Peng et al., 2019).
Although the presence of lactate in the tumor microenviron- expected to reflect tumor lactate level via magnetic reson-
ment (TME) was previously considered as metabolic waste, ance imaging (MRI) (Garcia-Hevia et al., 2019).
more recently, accumulating evidence has suggested that it The earliest and most frequently used MRI contrast agent
acts as an important signaling molecule in the regulation of (CA) approved for clinical use was Gd3þ-complexes. However,
tumor metabolism and immunity (Zhang et al., 2019). the safety of Gd-DTPA-BMA (gadodiamide) has become
Moreover, the lactate in the TME controls multiple phenom- increasingly controversial since 2006 (Grobner, 2006;
ena associated with tumor resistance to therapy (Pilon- Marckmann et al., 2006). Currently, in addition to two hep-
Thomas et al., 2016; Ippolito et al., 2019). Thus, the noninva- atocyte-specific CAs, European countries ban the use of Gd-
sive detection of tumor acidic regions is critical not only for BOPTA (gadobenate dimeglumine), Gd-DTPA-BMA, Gd-DTPA
personalized medicine but also for prognosis prediction. The (gadopentetate dimeglumine), and Gd-DTPA-BMEA (gadover-
physiological pH of normal tissues and body fluids (including setamide) (Dekkers et al., 2018). Mn, an essential trace elem-
blood) is approximately neutral (7.35–7.45), whereas the pH ent in the human body, was shown to have a short T1 effect

CONTACT Yaqi Shen yqshen@hust.edu.cn Department of Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and
Technology, Wuhan, China; Zhiping Zhang zhipingzhang@mail.hust.edu.cn Tongji School of Pharmacy, Huazhong University of Science and Technology,
Wuhan, PR China
Supplemental data for this article can be accessed here.

These authors contributed equally to this work.
ß 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
 DRUG DELIVERY 135

due to its five unpaired electrons (Reddi et al., 2009; Pan Materials
et al., 2011). Hence, Mn2þ-based CAs have become of high
Manganese chloride (MnCl2), cyclohexane, and cobalt (iii)
interest for the development of novel MRI CAs (Gale et al.,
chloride hexahydrate (CoCl2) were purchased from AladdinV
R

2015; Erstad et al., 2019). In the last decade, scientists have

(Shanghai, China). EGCG was obtained from PurifyV
R

developed numerous Mn-based NPs for tumor-specific MRI.

(Chengdu, China) and IGEPAL CO-520 was obtained from
Nevertheless, in the case of some CAs (Shin et al., 2009;
Sigma-Aldrich (St. Louis, MO). Dioleoyl phosphatidic acid
Huang et al., 2010a, 2010b), the Mn2þ was trapped/coordi-
(DOPA), 1,2-dioleoyl-snglycero-3-phosphocholine (DOPC),
nated in the NP, resulting in lower relaxation efficiency com-
cholesterol, and 1,2-dioleoyl-sn-glycero-3-phosphoethanol-
pared to free Mn2þ. Recently, intelligent NPs (Cai et al., 2015;
amine-N-[methoxy(polyethylene, glycol)-2000] (DSPE-
Mi et al., 2016; Li et al., 2017; Wang et al., 2018), which
PEG2000) were purchased from Avanti Polar Lipids, Inc.
respond to the acidic conditions in tumor tissues, have been
(Alabaster, AL), whereas 1,10 -dioctadecyl-3,3,30 ,30 -tetramethy-
designed to improve the accuracy and sensitivity of imaging
lindotricarbocyanine iodide (DiR) was purchased from AAT
techniques by increasing the contrast between the tumor tis-
Bioquest, Inc. (Sunnyvale, CA). The 3-(4, 5-dimethyl-thiazol-2-
sue and background. For example, one strategy involves
yl)-2,5-diphenyltetrazolium bromide (MTT) was obtained from
Mn2þ doping into calcium phosphate (CaP) (Mi et al., 2016)
BioSharp (Seoul, South Korea).
or silica (Kim et al., 2013) to form pH-sensitive CAs. However,
the inevitable degradation problem of these inorganic agents
appears to limit further clinical applications (Fu et al., 2019).
In contrast, epigallocatechin gallate (EGCG), an organic green
Preparation of L-EGCG-Mn NPs
tea extract with excellent antioxidant activity, has gained
increasing attention in the biomedical field due to its good EGCG-Mn/DOPA NPs (EGCG-Mn NPs) were prepared using
biocompatibility, pH sensitivity, and versatile functionaliza- the reverse microemulsion method (Zhuang et al., 2016).
tion capabilities (Reygaert, 2014, 2018). For example, NPs Phase A, consisting of 100 lL of 10 mM EGCG and 100 lL of
coordinating EGCG with metal ions, such as Fe3þ (Xiao et al., DOPA, was added to 4 mL CO-520/cyclohexane, and phase B,
2015), Cu2þ (Tsai et al., 2016), Au3þ (Jiang et al., 2019), or consisting of 100 lL of 80 mM MnCl2 was added to 4 mL CO-
Sm3þ (Li et al., 2019), have been used to diagnose and treat 520/cyclohexane, and stirred separately for 0.5 h to form a
tumors and have had impressive safety assessments. reverse water-in-oil microemulsion. Next, phase A was added
However, there are few studies on the chelation of EGCG dropwise to phase B while stirring. After 2 h, 8 mL of ethyl
with Mn2þ for MRI. alcohol was added to break the microemulsion. The mixture
As such, in this study, we fabricate a novel NP based on was then collected and centrifuged at 13,000g for 15 min.
the chelation effect of EGCG and Mn2þ and obtained NPs Thereafter, the precipitate was washed twice with ethyl alco-
that might be used as an MRI CA (Figure 1). The coordin- hol and dried using N2.
ation interaction between the metal and EGCG was eval- The L-EGCG-Mn NPs were prepared by dissolving EGCG-
uated by previous studies (Rahim et al., 2018; Wang et al., Mn/DOPA, 80 lL of 20 mM DOPC, 80 lL of 20 mM cholesterol,
2018). Moreover, PEGylation was shown to enhance the sta- and 20 lL of 20 mM DSPE-PEG2000 in 4 mL trichloromethane.
bility of L-epigallocatechin gallate (EGCG) complexed Mn2þ Next, trichloromethane was removed via rotary evaporation.
nanoparticle (L-EGCG-Mn) and prolong circulation time (Li DiR-labeled L-EGCG-Mn was prepared by dissolving EGCG-
et al., 2010; Calcagno et al., 2019). In addition, the chelation Mn/DOPA, DiR, DOPC, cholesterol, and DSPE-PEG2000 in tri-
of EGCG to Mn2þ would be weakened in an acidic environ- chloromethane, and the trichloromethane was removed with
ment (Navarro et al., 2005), thereby accelerating the release rotary evaporation. The L-EGCG-Mn NPs or DiR-labeled L-
of Mn2þ. Thus, L-EGCG-Mn could be disintegrated in a low EGCG-Mn were then hydrated using phosphate-buffered
pH environment to accurately control Mn2þ release and sim- saline (PBS) and incubated in a water bath at 37  C for 0.5 h
ultaneously present high relaxivity. and then used for further applications.
This study aimed to synthesize and characterize the pH-
sensitive L-EGCG-Mn as well as verify its magnetic resonance
(MR) properties, both in vitro and in vivo. The first commer-
cially approved Gd chelate, Gd-DTPA, was used as the refer- Characterization of the L-EGCG-Mn NPs
ence standard for MRI CA.
Zeta potential and particle size were measured via dynamic
light scattering (DLS, ZetaPlus, Brookhaven Instruments,
Holtsville, NY). Their morphology was evaluated through
transmission electron microscopy (TEM, HITACHI H-7000 FA,
Materials and methods
Chiyoda City, Japan, acceleration voltage ¼ 100 kV). The sta-
This study was approved by the local Ethics Committee and bility of the L-EGCG-Mn NPs was investigated by dispersing
all experiments in this study strictly followed the Institutional the L-EGCG-Mn NPs in PBS and fetal bovine serum (FBS).
Guidelines of Experimental Animal Care and Use. The entire Thereafter, the change in particle size was recorded continu-
workflow of this study is briefly shown in Figure 2. ously for 1 week.
136 J. LI ET AL.

Figure 1. A scheme indicating the synthesis of L-EGCG-Mn NPs and the subsequent pH-sensitive mechanism in vivo. (a) Mn2þ coordinated with EGCG to form
EGCG-Mn complexes. (b) The preparation of L-EGCG-Mn NPs and the mechanism of action of L-EGCG-Mn NPs in vivo.

Figure 2. Flowchart of the entire experimental design. Step 1, synthesis and characterization. The reverse microemulsion method was used to obtain L-EGCG-Mn
NPs. Step 2, relaxivity measurement. In each set of L-EGCG-Mn solution, the Mn concentration used was 0.04, 0.08, 0.2, 0.4, and 0.8 mM, respectively. Three samples
per concentration were analyzed. Step 3, in vivo MRI assessment. Eight mice who received H22 cell transplantation were imaged using a 3 T MRI scanner (n ¼ 5
injected with L-EGCG-Mn NPs, 6.4 lmol/kg Mn; n ¼ 3 injected with Gd-DTPA, 6.4 lmol/kg Gd).
 DRUG DELIVERY 137

Cell lines and tumor model 15-channel knee coil and a 3 T (Skyra; Siemens Healthcare,
Erlangen, Germany) MR system with a Tx/Rx 15-channel knee
H22 and L929 cells were purchased from the Chinese
coil. T1 maps were obtained using a series of inversion-recov-
Academy of Sciences Cells Bank (Shanghai, China). Cells were
ery sequences with various inversion times (TIs) (Ogg &
maintained in Dulbecco’s Modified Eagle Medium (L929) or
Kingsley, 2004; Shen et al., 2015). TI ¼ [30, 60, 90, 120, 150,
Roswell Park Memorial Institute-1640 medium (H22) supple-
250, 400, 600, 800, 1200, 1600, 2000, 2400, 2800, and 3200]
mented with 10% FBS and 1% penicillin and streptomycin
under a humidified atmosphere (37  C, 5% CO2). KM mice ms. The repetition time (TR) was equal to 1500 ms þ TI. The
(female, 18–20 g; 4–5 weeks of age) were purchased from echo time (TE) was 15 (3 T)/11 (1.5 T) ms. The T2 maps were
the local institutional animal care center and were accli- obtained using a protocol involving multi-echo spin-echo
mated to the center environment before study initiation. sequences (Pintaske et al., 2006; Shen et al., 2019): the TE
After resuscitation, H22 cells were injected into the peri- was between 20 and 600 ms with an interval of 20 ms and
toneal cavity of KM mice. Carcinoma ascites were collected the TR was 3000 ms. The following parameters were main-
after seven days. The concentration of H22 cells was adjusted tained for all measurements: slice thickness, 5 mm; field-of-
to 2  106 cells/mL, and 100 lL of the H22 cell suspension view, 80  100 mm; matrix, 256  256.
was subcutaneously injected into the left side of the mice
back to generate the tumor model (Bao et al., 2016).
Calculation of relaxivity
First, the generated DICOM images were analyzed via the
In vitro cytotoxicity evaluation ImageJ software package (open source, National Institutes of
Health, Bethesda, MD), which was used to place fixed-size
Cytotoxicity was evaluated using the MTT assay. Briefly, nor-
circular region-of-interest (ROI) and to automatically calcu-
mal fibroblasts cells L929 were seeded in 96-well plates at a
lated mean signal intensities (SIs) within the ROI. ROIs were
density of 8000 cells/well and cultured for 24 h. Next, the
between 160 and 170 pixels. Second, the relaxivity constants
supernatant was removed and replaced with 100 lL of blank
R1 and R2 are determined via Equations (1) and (2), respect-
medium supplemented with various concentrations of L-
ively, using a developed Data fitting software (Sigma Plot
EGCG-Mn NPs. After incubating for another 24 h, 10 lL of
MTT (5 mg/mL) was added and the cells were then incubated 12.5).
for another 2 h. The liquid from each well was removed and
replaced with 150 lL of dimethyl sulfoxide. The absorbance SITI ¼ A1 þ B1 exp R1
TI (1)
at 490 nm was detected using a microplate reader (Multiskan
MK3, Thermo Fisher Scientific, Waltham, MA). SITE ¼ A2 expR2
TE þ B2 (2)

Finally, the r1 and r2 values are obtained using Equation

Relaxivity measurement (3) (Fries et al., 2015). Here, R (c) denotes the relaxivity con-
Preparation of samples stant of L-EGCG-Mn at concentration C and R (0) represents
Three sets of L-EGCG-Mn buffer solutions with different pH the relaxivity constant of PBS or HSA.
values (pH ¼ 7.4, 6.8, and 5.5 PBS) were prepared. The Gd-

DTPA in PBS (pH ¼ 7.4) solution was used as a control. For r ¼ RðcÞ  Rð0Þ =C (3)
each set of the L-EGCG-Mn solutions, the Mn concentrations
used were 0.04, 0.08, 0.2, 0.4, and 0.8 mM. Likewise, the Gd
concentration in the Gd-DTPA solution was 0.04, 0.08, 0.2,
Cellular MR imaging
0.4, and 0.8 mM. Three samples per concentration were ana-
lyzed. Thus, this group contained a total of 60 samples. To evaluate the MR properties of L-EGCG-MN NPs in nor-
Forty-five more samples were prepared in the same man- moxic and hypoxic cells, CoCl2 was used to induce chemical
ner and incubated with human serum albumin (HSA, hypoxia (Dubbelboer et al., 2019). Briefly, H22 cells were iso-
10 mg/mL) for 24 h. Fresh samples were prepared before the lated under sterile conditions and were randomly assigned
MR scan. Finally, 210 samples were analyzed via MR scanning to either the hypoxia and normoxia groups. Cells were then
(1.5 and 3 T at 22  C). incubated with medium with or without CoCl2 (200 lM) and
After MR scanning, the final Mn concentration of the L- seeded into six-well plates (1  107 cells/well) for 12 h. The
EGCG-Mn buffer solutions (180 samples) was measured via cells were then collected and the medium was removed via
flame atomic absorption spectroscopy (SpectrAA-240FS, centrifugation. Thereafter, the cells were washed thrice with
Varian, Palo Alto, CA) (wide range [Mn2þ], 0.02–0.6 mM). PBS to eliminate the residual CoCl2. The cells were then
Nitric acid was added to decompose the L-EGCG-Mn NPs
resuspended in medium with or without L-EGCG-Mn (1 mM).
before detection.
After being incubated for 4 h, the cells were washed thrice
with PBS to eliminate the residual L-EGCG-Mn. Next, agarose
Machine and sequences gel (1%, 300 lL) was used to resuspend and fix the cells
In vitro analysis was performed on a 1.5 T (Aera; Siemens, (Zheng et al., 2019). Finally, cellular imaging was carried out
Healthcare, Erlangen, Germany) MR scanner with a Tx/Rx using the abovementioned MR system (3 T).
138 J. LI ET AL.

Animal MRI 10% formalin. After paraffin embedding and hematoxylin

and eosin staining, the sample sections were evaluated by
H22 tumor-bearing mice were used for animal MRI assess-
an experienced histopathologist.
ments. Tumor growth occurred in 10 days and the final
tumor volume was approximately 100 mm3. The tumor vol-
ume was measured from vernier caliper and calculated as Statistical methods
the length  width (Zhang et al., 2019)0.5 (Luo et al., 2019).
All mice were scanned using a 3.0 T MRI scanner (Skyra; All statistical analyses were completed using IBM SPSS 23.0
Siemens Healthcare, Erlangen, Germany) with an 8-channel (Chicago, IL) and GraphPad Prism 7.0 (GraphPad Software, La
5-cm Rx custom-design coil. Anesthetized animals were kept Jolla, CA). A p value<.05 was considered to be statistically
warm with a thermostatic electric blanket at 37  C between significant. The measurement consistency between the two
imaging sessions. Before imaging, the animals were placed in radiologists was tested by calculating the interclass correl-
a prone position. T1 images were acquired pre-injection and ation coefficient (ICC). Continuous variables were analyzed
at 0.5, 1, 2, and 4 h after the injection of L-EGCG-Mn using the Kolmogorov–Smirnov test to determine the nor-
(6.4 lmol/kg Mn) and Gd-DTPA (6.4 lmol/kg Gd) NPs via the mality and then, the Student t-test (normal distribution) or
tail vein. The time points were selected based on a previ- Mann–Whitney U-test (non-normal distribution) was used
ously published study by Mi et al. (2016). The detailed scan- for comparison.
ning parameters are listed in supplementary Table 1.
Image analysis: signal-to-noise ratio (SNR) and contrast-to-
noise ratio (CNR) were measured and calculated by two radi- Results
ologists based on previous reports (Peng et al., 2018), using
Equations (4) and (5), respectively: Characterization of L-EGCG-Mn NPs

SNR ¼ Stumor =SDbackground (4) The particle size of the L-EGCG-Mn NPs was 277.4 ± 5.5 nm
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (Figure 3(a)) and the zeta potential was 13.56 ± 1.91 mV.
CNR ¼ jStumor Stissue j= SDtumor 2 þ SDtissue 2 (5) Moreover, the particle size change after incubation in both
where Stumor represents the SI in the ROI placed on a homo- PBS and FBS for 1 week was negligible (Figure 3(b), supple-
geneously enhancing part of the tumor without necrosis and mentary figure 1). DLS (Figure 3(c)) showed that the particle
SDbackground represents the standard deviation of the back- size of L-EGCG-Mn NPs incubated with an acidic solution
ground noise. Stissue represents the SI in the ROI of ipsilateral increased over time, with L-EGCG-Mn NPs in a solution with
normal muscle tissue. SDtumor and SDtissue represent the a pH of 5.5 expanding faster than those in a solution with a
standard deviation of the tumor and normal tissue. The ROIs pH of 6.8. The morphology determined via TEM (Figure 4)
were located in anatomic positions, which were as accurate revealed that there was a change between L-EGCG-Mn NPs
as possible for the different time points. The above parame- incubated in a solution with a pH of 7.4 or 5.5. In addition,
ters were measured by two experienced radiologists blinded we also observed that the NPs disintegrated in an acidic
to the CA administered. The average was then obtained for environment.
further analysis.

In vitro cytotoxicity evaluation

Fluorescence imaging and in vivo distribution of
L-EGCG-Mn L929 cells were used to assess the cytotoxicity of L-EGCG-Mn
NPs in normal cells. As shown in Figure 5, cell viability was
DiR (ex ¼ 748 nm, em ¼ 780 nm) can be used to obtain the not markedly reduced when L929 cells were incubated with
vivo fluorescence images. Thus, the L-EGCG-Mn NPs were L-EGCG-Mn NPs for 24 h.
labeled with DiR to investigate their distribution. DiR-labeled
L-EGCG-Mn (200 lL, approximately 4 lg DiR per mouse) were
injected intravenously. Mice injected with DiR dissolved in Relaxivity measurement
PBS were used as a negative control. The mice were euthan-
ized and the heart, liver, spleen, lung, kidneys, tumor, and The plots of signal vs. TI or TE and of R1 or R2 vs. [Mn] are
homolateral inguinal lymph nodes were excised and photo- shown in supplementary figure 2. The relaxivity values of L-
graphed using a near-infrared fluorescence small animal live EGCG-Mn and Gd-DTPA at 1.5 and 3.0 T are presented in
imaging system (Pearl Trilogy, LI-COR) at 1, 2, 4, 8, 12, and Table 1. For the 3 T MRI, when the pH decreased from 7.4 to
24 h post-injection. 5.5, the r1 (r2) value of L-EGCG-Mn NPs increased from 1.79
(10.79) to 6.43 (41.78) mM1s1 in the buffer solution and
from 1.77 (16.1) to 7.23 (42.56) mM1s1 in HSA. Moreover,
Histological analysis there was a significant difference in the relaxivity values
All mice were euthanized after the last MRI analysis. between the L-EGCG-Mn NP solutions with different pH
Afterwards, the specimens, including the major organs (heart, (p<.001, Figure 6). At pH 5.5, the relaxivity values (r1 and r2)
liver, spleen, lung, kidneys), tumor, and homolateral inguinal of L-EGCG-Mn NPs were found to be higher than that of Gd-
lymph nodes, were harvested and fixed via immersion in DTPA (p<.001).
 DRUG DELIVERY 139

Figure 3. Characterization of L-EGCG-Mn NPs. (a) Particle size of L-EGCG-Mn NPs. (b) The change in particle size of L-EGCG-Mn NPs in phosphate-buffered saline
(PBS) and fetal bovine serum (FBS) during a time period of seven days after preparing the solution. (c) The change in particle size of L-EGCG-Mn NPs incubated in
buffer solutions with a pH of 7.4, 6.8, or 5.5.

Cellular MR imaging
The shortening of the T1 relaxation time (DT1) was calcu-
lated by subtracting T1 value for a Mn concentration of
1 mM from the T1 value for 0 mM Mn. After incubation with
L-EGCG-Mn NPs for 4 h, the T1 value of hypoxic H22 cells
was found to be significantly lower than that of normoxic
H22 cells (1788 ± 89 vs. 1982 ± 68 ms, p¼.041) (Figure 7, sup-
plementary Table 2). Moreover, the DT1 of the hypoxia
group was shown to be lower than that of the normoxia
group (817 vs. 993 ms).

Animal MRI
The interobserver agreement for CNR and SNR was excellent
(ICC > 0.81, Table 2). For L-EGCG-Mn and Gd-DTPA, the CNR
and SNR almost reached their peak at 1 h, followed by a sta-
ble high value for the former but a downtrend for the latter
(Figure 8). After injection, the average value of CNR and SNR
was significantly higher for L-EGCG-Mn NPs than for Gd-
DTPA at all acquired timepoints (p < .05, supplementary table
Figure 4. TEM images of L-EGCG-Mn NPs incubated with (a) pH 7.4 buffer solu-
3). The classic MRI images of the two mice groups are shown
tion and (b) pH 5.5 buffer solution. in Figure 9.

Fluorescence imaging and in vivo distribution of L-

EGCG-Mn NPs
The DiR-L-EGCG-Mn NPs were found to gradually gather at
the tumor site after injection, with the quantity of NPs at the
site increasing over time. Moreover, the fluorescence inten-
sity at the tumor site in the DiR-L-EGCG-Mn group was evi-
dently stronger than that in the DiR group (Figure 10). The
fluorescence intensity in the inguinal lymph nodes was also
stronger in the DiR-L-EGCG-Mn group than that of the
DiR group.

Histological analysis
Histopathological analysis confirmed the presence of hepa-
toma cells in the tumor samples from all examined animals
(supplementary figure 3). After L-EGCG-Mn injection, no
Figure 5. Evaluation of in vitro cytotoxicity of L-EGCG-Mn NPs in L929 cells. appreciable abnormalities were observed in the heart, liver,
140 J. LI ET AL.

Table 1. In vitro MRI relaxivities of L-EGCG-Mn and Gd-DTPA.

1.5 T 3T
1 1 1 1 1 1
Contrast agents r1 (mM s ) r2 (mM s ) r1 (mM s ) r2 (mM1s1)
L-EGCG-Mn(þHSA)
pH 7.4 1.23 ± 0.09 (2.08 ± 0.41) 8.06 ± 0.51 (10.13 ± 1.42) 1.79 ± 0.004 (1.77 ± 0.05) 10.79 ± 0.47 (16.10 ± 1.08)
pH 6.8 2.14 ± 0.37 (2.63 ± 0.71) 13.54 ± 2.09 (14.33 ± 2.48) 4.18 ± 0.04 (5.55 ± 0.06) 20.26 ± 0.38 (26.95 ± 0.37)
pH 5.5 5.93 ± 0.04 (6.45 ± 0.19) 37.59 ± 0.70 (38.09 ± 2.02) 6.43 ± 0.05 (7.23 ± 0.02) 41.78 ± 2.12 (42.56 ± 2.17)
Gd-DTPA 4.38 ± 0.001 4.9 ± 0.03 4.99 ± 0.09 5.99 ± 0.13
MRI: magnetic resonance imaging; Gd: gadolinium; HSA: human serum albumin.
Values are given as mean ± SD in buffered saline (in HSA) at room temperature.

Figure 6. In vitro relaxivity. (a, b) The r1 and r2 relaxivity of L-EGCG-Mn NPs at different pH values and Gd-DTPA in 3 T MRI. pH 7.4 and 6.8 phosphate buffer solu-
tion and pH 5.5 acetate buffer solution were used here. Data are shown as mean ± SD. (n ¼ 3),


p< .001.

et al., 2016). As such, this study aimed to synthesize pH-sen-

sitive NPs to develop a tumor-targeting MRI CA.
We designed a series of experiments to demonstrate the
pH-sensitivity of L-EGCG-Mn NPs. First, L-EGCG-Mn NPs were
incubated with buffer solutions of a different pH. Analysis of
the change in particle size and morphology implied that L-
EGCG-Mn NPs were disrupted in low pH environments.
Moreover, in vitro analysis indicated that the relaxivity of L-
Figure 7. T1-weighted H22 cellular MRI with (þ) and without (–) L-EGCG-Mn EGCG-Mn NPs increased as the pH decreased. The T1 value
NPs. The T1 value was obtained from the T1 mapping images. DT1 was calcu-
lated by subtracting the T1 value for 1 mM Mn from the T1 value for 0 mM Mn.
of L-EGCG-Mn NPs in hypoxic cells was also lower than that
found in normoxic cells, which suggests that L-EGCG-Mn NPs
Table 2. The inter-observer agreement between two radiologists of CNR
are sensitive to pH at the cellular level. These findings sup-
and SNR. ported our hypothesis that a low pH could mediate the dis-
Parameters ICC 95% CI assembly of L-EGCG-Mn NPs and accurately control the
CNR 0.960 0.924–0.979 release of Mn2þ (Li et al., 2010). Thus, we can reasonably
SNR 0.899 0.806–0.948 assume that L-EGCG-Mn NPs may be applicable for tumor
ICC: interclass correlation coefficient; CI: confidence interval. imaging, where the specific acidic environment can serve as
relaxation switches activated by pH (Li et al., 2010). Our
spleen, lungs, and kidneys. Reactive hyperplasia was results also showed that L-EGCG-Mn NPs mainly released
observed in the tumor homolateral inguinal lymph node. Mn2þ in the tumor area, which enabled the selective
enhancement of tumor tissue and thus increased the con-
trast between tumor and adjacent normal tissues. In add-
Discussion ition, pH-sensitive L-EGCG-Mn NPs may help predict and
assess tumor therapeutic outcomes, as an acidic environment
Contrast-enhanced MRI fulfills several important medical is greatly associated with therapeutic response (Swartz et al.,
needs. Therefore, the development and improvement of MRI 2012; Chang et al., 2015; Pilon-Thomas et al., 2016). However,
CAs, especially tumor-targeting agents, is a growing area of this hypothesis requires further experimental verification.
study (Adiseshaiah et al., 2013; Gale et al., 2018). The acidic Analysis of DiR-L-EGCG-Mn distribution ex vivo showed
pH environment, mainly caused by lactate, has been exten- that L-EGCG-Mn NPs possessed excellent tumor-targeting
sively proven to be a tumor-specific characteristic (Kanamala abilities for approximately 24 h post-injection, which could
 DRUG DELIVERY 141

Figure 8. The contrast-to-noise ratio (CNR) and signal-to-noise ratio (SNR) evaluation. (a, b) The trend of CNR and SNR after the administration of L-EGCG-Mn
(n ¼ 5) and Gd-DTPA (n ¼ 3) for T1WI.

accessibility of water molecules and Mn2þ. However, the r1

value found in this study is higher than that determined for
USMO@MSNs (5.61 mM1s1 in buffer solution, pH 4.5 at
9.4 T MR) and delaminated MnO2 nanosheets (4.0 mM1s1
in buffer solution, pH 4.6 at 3 T MR). This may be due to the
fact that the acidic environment can weaken the chelation of
EGCG and Mn, thereby substantially accelerating the release
of Mn2þ. Furthermore, the difference in relaxivity may also
be associated with the different experimental conditions
used, as there are many factors that can influence relaxivity,
including solvent type, incubation time, and even tempera-
ture (Hao et al., 2012; Goetschi et al., 2014). L-EGCG-Mn NPs
also showed pH sensitivity in terms of shortening the T2
relaxation time, which is similar to previous studies (Chen
et al., 2012).
The high relaxivity of L-EGCG-Mn NPs encouraged us to
Figure 9. In vivo T1WI MR images (3 T) of H22 tumor-bearing mice pre- and further explore its MRI performance in vivo. We found that
post-intravenous injection (i.v.) with Gd-DTPA (a) or L-EGCG-Mn NPs (b). The
images are displayed at a window width of 2588 and window level of 1324. L- the in vivo MRI performance of L-EGCG-Mn NPs was compar-
EGCG-Mn NPs led to a higher enhancement of tumor contrast than Gd-DTPA. able to that of the pH-responsive HMPB-Mn (Cai et al., 2015),
PEGMnCaP (Mi et al., 2016), and (UCNP@PFNS/N)@MnCaP (Ji
et al., 2019). However, the peak SI of HMPB-Mn appeared at
provide a remarkable image-acquisition time window. This is approximately 30 min, whereas that of L-EGCG-MN NPs
mainly attributable to the enhanced permeability and reten- appeared at 1 h. This may be due to the fact that the HMPB-
tion effects (Kim et al., 2018) of NPs. Consequently, we can Mn NPs were administrated via intratumor injection.
conclude that high tumor retention and rapid organ clear- Additionally, L-EGCG-Mn NPs mainly exhibited a homoge-
ance makes L-EGCG-Mn NPs an efficient and safe CA. neous enhancement, which is different from the selective
The r1 values of L-EGCG-Mn NPs in a buffer solution at high strengthening of PEGMnCaP and (UCNP@PFNS/
3 T MRI (pH 6.8, 4.18 mM1s1 and pH 5.5, 6.43 mM1s1) N)@MnCaP. The significantly lower concentration of Mn used
obtained in this study are comparable to those displayed by in this study (6.4 lmol/kg) may be a potential explanation, as
HMPB-Mn (Cai et al., 2015) in an aqueous solution at 7 T (pH well as the use of a different tumor type (H22 tumors) when
5, 7.43 mM1s1), PEGMnCaP (Mi et al., 2016) in buffer solu- compared to existing studies (PEGMnCaP: 225 lmol/kg based
tion at 0.59 T (pH 6.7, 4.27 mM1s1), and by (UCNP@PFNS/ on Mn for C26 tumors; UCNP@PFNS/N @MnCaP: 750 lmol/kg
N)@MnCaP (Ji et al., 2019) in an aqueous solution at 7 T (pH based on Mn for HepG2 tumors).
6.8, 4.4 mM1s1 and pH 5, 6.9 mM1s1). However, the r1 The above-mentioned Mn-based CAs are still in the basic
value of L-EGCG-Mn NPs in HSA (pH 6.8, 5.55 mM1s1) is research stage. Therefore, although they are useful as a refer-
lower than that of PEGMnCaP in HSA (pH 6.7, ence for comparison against L-EGCG-Mn NPs, it is also
15.26 mM1s1). A possible explanation is that the specific important to compare L-EGCG-Mn NPs with clinically used
binding between EGCG and HSA limits the binding of Mn2þ CAs. Thus, the first approved extracellular Gd chelate, namely
and HSA (Save & Choudhary, 2018). Interestingly, pH-sensi- Gd-DTPA, was used to compare the properties of L-EGCG-Mn
tive CAs, namely L-EGCG-Mn, USMO@MSNs (Wang et al., NPs. In vitro experiments showed that the T1 and T2 relaxivity
2018), and MnO2 nanosheets (Chen et al., 2014), are all of L-EGCG-Mn NPs were significantly higher than those of
designed to make the NP core be the most exposed to water Gd-DTPA. Moreover, in the H22 tumor-bearing mice model,
molecules in an acidic environment, thereby improving the L-EGCG-Mn NPs led to improved SNR and CNR when
142 J. LI ET AL.

Figure 10. Fluorescence imaging and ex vivo distribution of L-EGCG-Mn NPs. Ex vivo image of the inguinal lymph nodes, heart, liver, spleen, lungs, kidneys, and
tumor of H22 tumor-bearing KM mice injected with DiR and DiR-L-EGCG-Mn NPs for 1, 2, 4, 8, 12, and 24 h.

compared to Gd-DTPA in T1WI. In addition to the pH sensi- Funding

tivity of L-EGCG-Mn NPs, it is also possible that this may be
This work was supported by the National Natural Science Foundation of
due to the fact that Gd-DTPA has only one water molecule China under Grant numbers: 82071890, 81701657, 81673374, 81872810,
coordination site due to the DTPA ligand forming a stable 81571642, and 81771801, the Fundamental Research Funds for the
structure around Gd3þ (Cai et al., 2015). Central Universities under Grant numbers: 2018KFYYXJJ019 and
Additionally, histochemical analysis showed that the 2019KFYRCPY049, Wuhan Science and Technology Plan for Applied
Fundamental Research under Grant number 2017060201010146.
inguinal lymph nodes exhibited inflammatory hyperplasia,
possibly as an effect of the neoplasm. Interestingly, analysis
of in vivo distribution showed that the aggregation of L-
EGCG-Mn NPs in the lymph nodes increased with time. This References
phenomenon may be explained by the acidic microenviron-
Adiseshaiah P, Dellinger A, MacFarland D, et al. (2013). A novel gadolin-
ment created due to inflammatory hyperplasia (Gallagher ium-based trimetasphere metallofullerene for application as a mag-
et al., 2008). Thus, the relationship between inflammatory tis- netic resonance imaging contrast agent. Invest Radiol 48:745–54.
sues and Mn NPs needs further study. Bao Y, Yin M, Hu X, et al. (2016). A safe, simple and efficient doxorubicin
prodrug hybrid micelle for overcoming tumor multidrug resistance
and targeting delivery. J Control Release 235:182–94.
Cai X, Gao W, Ma M, et al. (2015). A Prussian Blue-based core–shell hol-
Conclusions low-structured mesoporous nanoparticle as a smart theranostic agent
with ultrahigh pH-responsive longitudinal relaxivity. Adv Mater 27:
This study developed EGCG-Mn NPs enveloped with phos- 6382–9.
pholipids and thus obtained NPs with a good safety profile Calcagno V, Vecchione R, Quagliariello V, et al. (2019). Oil core–PEG shell
and high pH sensitivity that may be used as MRI CAs. L- nanocarriers for in vivo MRI imaging. Adv Healthc Mater 8:e1801313.
Chang CH, Qiu J, O’Sullivan D, et al. (2015). Metabolic competition in the
EGCG-Mn NPs could respond to tumor-related pH changes
tumor microenvironment is a driver of cancer progression. Cell 162:
and, therefore, may serve as a potential tumor-targeting CA 1229–41.
due to its good MRI properties in both a hypoxic cell model Chen Y, Ye D, Wu M, et al. (2014). Break-up of two-dimensional MnO2
and H22 tumor-bearing mice model. nanosheets promotes ultrasensitive pH-triggered theranostics of can-
cer. Adv Mater 26:7019–26.
Chen Y, Yin Q, Ji X, et al. (2012). Manganese oxide-based multifunction-
alized mesoporous silica nanoparticles for pH-responsive MRI, ultra-
Acknowledgements sonography and circumvention of MDR in cancer cells. Biomaterials
33:7126–37.
The authors thank the Analytical and Testing Center of Huazhong
Dekkers IA, Roos R, van der Molen AJ. (2018). Gadolinium retention after
University of Science and Technology for TEM analysis.
administration of contrast agents based on linear chelators and the
recommendations of the European Medicines Agency. Eur Radiol 28:
1579–84.
Dubbelboer IR, Pavlovic N, Heindryckx F, et al. (2019). Liver cancer cell
Disclosure statement
lines treated with doxorubicin under normoxia and hypoxia: cell via-
The author reports no conflicts of interest in this work. bility and oncologic protein profile. Cancers 11:1024.
 DRUG DELIVERY 143

Erstad DJ, Ramsay IA, Jordan VC, et al. (2019). Tumor contrast enhance- Mi P, Kokuryo D, Cabral H, et al. (2016). A pH-activatable nanoparticle
ment and whole-body elimination of the manganese-based magnetic with signal-amplification capabilities for non-invasive imaging of
resonance imaging contrast agent Mn-PyC3A. Investig Radiol 54: tumour malignancy. Nat Nanotechnol 11:724–30.
697–703. Navarro RE, Santacruz H, Inoue M. (2005). Complexation of epigallocate-
Fries P, Muller A, Seidel R, et al. (2015). P03277—a new approach to chin gallate (a green tea extract, EGCG) with Mn2þ: nuclear spin relax-
achieve high-contrast enhancement: initial results of an experimental ation by the paramagnetic ion. J Inorg Biochem 99:584–8.
extracellular gadolinium-based magnetic resonance contrast agent. Neri D, Supuran CT. (2011). Interfering with pH regulation in tumours as
Investig Radiol 50:835–42. a therapeutic strategy. Nat Rev Drug Discov 10:767–77.
Fu C, Duan X, Cao M, et al. (2019). Targeted magnetic resonance imag- Ogg RJ, Kingsley PB. (2004). Optimized precision of inversion-recovery T1
ing and modulation of hypoxia with multifunctional hyaluronic acid- measurements for constrained scan time. Magn Reson Med 51:625–30.
MnO2 nanoparticles in glioma. Adv Healthcare Mater 8:e1900047. Pan D, Schmieder AH, Wickline SA, Lanza GM. (2011). Manganese-based
Gale EM, Atanasova IP, Blasi F, et al. (2015). A manganese alternative to MRI contrast agents: past, present and future. Tetrahedron 67:8431–44.
gadolinium for MRI contrast. J Am Chem Soc 137:15548–57. Peng W, Huang W, Ge X, et al. (2019). Type Igamma phosphatidylinositol
Gale EM, Wey HY, Ramsay I, et al. (2018). A manganese-based alternative phosphate kinase promotes tumor growth by facilitating Warburg
to gadolinium: contrast-enhanced MR angiography, excretion, effect in colorectal cancer. EBioMedicine 44:375–86.
pharmacokinetics, and metabolism. Radiology 286:865–72. Peng Y, Li Z, Tang H, et al. (2018). Comparison of reduced field-of-view
Gallagher FA, Kettunen MI, Day SE, et al. (2008). Magnetic resonance diffusion-weighted imaging (DWI) and conventional DWI techniques
imaging of pH in vivo using hyperpolarized 13C-labelled bicarbonate. in the assessment of rectal carcinoma at 3.0T: image quality and
Nature 453:940–3. histological T staging. J Magn Reson Imaging 47:967–75.
Garcia-Hevia L, Banobre-Lopez M, Gallo J. (2019). Recent progress on Pilon-Thomas S, Kodumudi KN, El-Kenawi AE, et al. (2016). Neutralization
manganese-based nanostructures as responsive MRI contrast agents. of tumor acidity improves antitumor responses to immunotherapy.
Chemistry 25:431–41. Cancer Res 76:1381–90.
Goetschi S, Froehlich JM, Chuck NC, et al. (2014). The protein and con- Pintaske J, Martirosian P, Graf H, et al. (2006). Relaxivity of gadopentetate
trast agent-specific influence of pathological plasma-protein concen- dimeglumine (Magnevist), gadobutrol (Gadovist), and gadobenate
tration levels on contrast-enhanced magnetic resonance imaging. dimeglumine (MultiHance) in human blood plasma at 0.2, 1.5, and
Invest Radiol 49:608–19. 3 Tesla. Invest Radiol 41:213–21.
Grobner T. (2006). Gadolinium—a specific trigger for the development of Rahim MA, Bjornmalm M, Bertleff-Zieschang N, et al. (2018). Multiligand
nephrogenic fibrosing dermopathy and nephrogenic systemic fibro- metal-phenolic assembly from green tea infusions. ACS Appl Mater
sis? Nephrol Dial Transplant 21:1104–8. Interfaces 10:7632–9.
Hao DP, Ai T, Goerner F, et al. (2012). MRI contrast agents: basic chemis- Reddi AR, Jensen LT, Culotta VC. (2009). Manganese homeostasis in
try and safety. J Magn Reson Imaging 36:1060–71. Saccharomyces cerevisiae. Chem Rev 109:4722–32.
Huang CC, Khu NH, Yeh CS. (2010a). The characteristics of sub 10 nm Reygaert WC. (2014). The antimicrobial possibilities of green tea. Front
manganese oxide T1 contrast agents of different nanostructured mor- Microbiol 5:434.
phologies. Biomaterials 31:4073–8. Reygaert WC. (2018). Green tea catechins: their use in treating and pre-
Huang J, Xie J, Chen K, et al. (2010b). HSA coated MnO nanoparticles venting infectious diseases. Biomed Res Int 2018:9105261.
with prominent MRI contrast for tumor imaging. Chem Commun Save SN, Choudhary S. (2018). Elucidation of energetics and mode of
(Camb) 46:6684–6. recognition of green tea polyphenols by human serum albumin. J
Ippolito L, Morandi A, Giannoni E, Chiarugi P. (2019). Lactate: a metabolic Mol Liq 265:807–17.
driver in the tumour landscape. Trends Biochem Sci 44:153–66. Shen Y, Goerner FL, Heverhagen JT, et al. (2019). In vitro T2 relaxivities
Ji Y, Lu F, Hu W, et al. (2019). Tandem activated photodynamic and of the Gd-based contrast agents (GBCAs) in human blood at 1.5 and
chemotherapy: using pH-sensitive nanosystems to realize different 3 T. Acta Radiol 60:694–701.
tumour distributions of photosensitizer/prodrug for amplified combin- Shen Y, Goerner FL, Snyder C, et al. (2015). T1 relaxivities of gadolinium-
ation therapy. Biomaterials 219:119393 based magnetic resonance contrast agents in human whole blood at
Jiang X, Sun Y, Shang L, et al. (2019). Green tea extract-assembled nano- 1.5, 3, and 7 T. Investig Radiol 50:330–8.
clusters for combinational photothermal and chemotherapy. J Mater Shin J, Anisur RM, Ko MK, Im GH, et al. (2009). Hollow manganese oxide
Chem B 7:5972–82. nanoparticles as multifunctional agents for magnetic resonance imag-
Kanamala M, Wilson WR, Yang M, et al. (2016). Mechanisms and bioma- ing and drug delivery. Angew Chem Int Ed Engl 48:321–4.
terials in pH-responsive tumour targeted drug delivery: a review. Swartz MA, Iida N, Roberts EW, et al. (2012). Tumor microenvironment
Biomaterials 85:152–67. complexity: emerging roles in cancer therapy. Cancer Res 72:2473–80.
Kim HJ, Yi Y, Kim A, Miyata K. (2018). Small delivery vehicles of siRNA for Tsai LC, Hsieh HY, Lu KY, et al. (2016). EGCG/gelatin-doxorubicin gold
enhanced cancer targeting. Biomacromolecules 19:2377–90. nanoparticles enhance therapeutic efficacy of doxorubicin for prostate
Kim SM, Im GH, Lee DG, et al. (2013). Mn(2þ)-doped silica nanoparticles cancer treatment. Nanomedicine (Lond) 11:9–30.
for hepatocyte-targeted detection of liver cancer in T1-weighted MRI. Wang C, Sang H, Wang Y, et al. (2018). Foe to friend: supramolecular
Biomaterials 34:8941–8. nanomedicines consisting of natural polyphenols and bortezomib.
Li B, Gu Z, Kurniawan N, et al. (2017). Manganese-based layered double Nano Lett 18:7045–51.
hydroxide nanoparticles as a T1-MRI contrast agent with ultrasensitive Wang D, Lin H, Zhang G, et al. (2018). Effective pH-activated theranostic
pH response and high relaxivity. Adv Mater 29:1700373. platform for synchronous magnetic resonance imaging diagnosis and
Li J, Chen YC, Tseng YC, et al. (2010). Biodegradable calcium phosphate chemotherapy. ACS Appl Mater Interfaces 10:31114–23.
nanoparticle with lipid coating for systemic siRNA delivery. J Control Xiao L, Mertens M, Wortmann L, et al. (2015). Enhanced in vitro and
Release 142:416–21. in vivo cellular imaging with green tea coated water-soluble iron
Li K, Xiao G, Richardson JJ, et al. (2019). Targeted therapy against meta- oxide nanocrystals. ACS Appl Mater Interfaces 7:6530–40.
static melanoma based on self-assembled metal-phenolic nanocom- Zhang W, Wang G, Xu ZG, et al. (2019). Lactate is a natural suppressor
plexes comprised of green tea catechin. Adv Sci (Weinh) 6:1801688. of RLR signaling by targeting MAVS. Cell 178:176–189.e15.
Luo M, Liu Z, Zhang X, et al. (2019). Synergistic STING activation by Zheng S, Zhang M, Bai H, et al. (2019). Preparation of AS1411 aptamer
PC7A nanovaccine and ionizing radiation improves cancer immuno- modified Mn-MoS2 QDs for targeted MR imaging and fluorescence
therapy. J Control Release 300:154–60. labelling of renal cell carcinoma. Int J Nanomedicine 14:9513–24.
Marckmann P, Skov L, Rossen K, et al. (2006). Nephrogenic systemic Zhuang X, Wu T, Zhao Y, et al. (2016). Lipid-enveloped zinc phosphate
fibrosis: suspected causative role of gadodiamide used for contrast- hybrid nanoparticles for codelivery of H-2K(b) and H-2D(b)-restricted
enhanced magnetic resonance imaging. J Am Soc Nephrol 17: antigenic peptides and monophosphoryl lipid A to induce antitumor
2359–62. immunity against melanoma. J Control Release 228:26–37.