Light-activated drug release from prodrug nanoassemblies by structure destruction

Article abstract of DOI:10.1039/c9cc06673j

We report here a novel light-triggered nanosystem based on co-assembling nanoaggregates (NAs) of lipophilic photosensitizers and lipophilic prodrugs containing multiple thioethers. Upon laser irradiation, the oxidization of the multiple thioethers by photosensitizer-generated singlet oxygen could rapidly destroy the NA structure, resulting in faster drug release than those containing a single thioether.

Full text of DOI:10.1039/c9cc06673j

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Light-activated drug release from prodrug
nanoassemblies by structure destruction†
Cite this: Chem. Commun., 2019,
55, 13128
Yang Li,‡a Shujuan Wang,‡a Yulan Huang,a Yuwen Chen,a Wenbi Wu,a Yu Liu,a
a
Jing Zhang,b Yue Feng,c Xian Jiangd and Maling Gou
*
Received 27th August 2019,
Accepted 1st October 2019
DOI: 10.1039/c9cc06673j
rsc.li/chemcomm
We report here a novel light-triggered nanosystem based on been constructed by encapsulating a photosensitizer into NAs of
co-assembling nanoaggregates (NAs) of lipophilic photosensitizers an LHP linked by a single thioether.17 The mechanism of light-
and lipophilic prodrugs containing multiple thioethers. Upon laser triggered drug release is mainly associated with the singlet oxygen
irradiation, the oxidization of the multiple thioethers by photosensitizer- (1O2)-mediated oxidization of thioether to electron-withdrawing
generated singlet oxygen could rapidly destroy the NA structure, sulfone or sulfoxide, which accelerates the hydrolysis of adjacent
resulting in faster drug release than those containing a single ester bonds.6 However, we found that the light-triggered drug
thioether.
release based on this mechanism was not very sensitive, because
the oxidization of a single thioether could not significantly
Light-triggered drug release from nanosized drug delivery decrease the hydrophobicity of LHPs. Based on this result, we
systems (nanoDDSs) is a promising approach to deliver the speculated that the light responsiveness would be significantly
encapsulated molecules with a high degree of spatial and improved if the hydrophobicity of LHPs could be decreased
temporal control.1–3 Mediator signals (e.g. heat and reactive sharply upon light irradiation. As thioethers are hydrophobic but
oxygen species) induced by visible light (400–700 nm) or near- can be readily oxidized to hydrophilic sulfones or sulfoxides,18,19
infrared light (700–1000 nm) are widely being utilized to activate we proposed that insertion of multiple thioethers in LHPs could
drug release.4–8 However, most light-sensitive systems are con- remarkably decrease the hydrophobicity of the oxidized LHPs,
structed based on carrier-assisted nanoDDSs, which suffer severely which further resulted in a more sensitive light-triggered drug
from their low drug loading capacity (typically 2–10%, w/w).9 To release as a result of the destruction of the NA structure (Fig. 1A).
address this, carrier-free nanoaggregates (NAs) have attracted
To test our hypothesis, we first synthesized the multiple
increasing attention due to their ultrahigh loading capacity.10,11 thioether-modified lipophilic prodrug of 7-ethyl-10-hydroxy-
Recently, lipophilic prodrugs of hydrophobic drugs (LHPs), despite camptothecin (SN38) by conjugating two thioether-modified
their highly hydrophobic nature, have been shown to be able to fatty acids at the C10 position of SN38 via a thioether-
self-assemble into NAs in water.12,13 As these NAs are constructed containing linker (2S2C16-S/SN38(10), containing 5 thioethers).
with pure prodrugs, their loading capacities are impressively high A lipophobic conjugate of SN38 and stearic acid (2C18-S/
(B35–70%, calculated in terms of the molecular weights of the SN38(10), containing 1 thioether) was utilized as a control.
parent drug and the prodrug).13–16 Light-triggered nanoDDSs have The clinically-approved protoporphyrin IX (PpIX), as a model
photosensitizer, was conjugated to stearic acid to obtain C18-S/
a State Key Laboratory of Biotherapy and Cancer Center, West China Hospital,
Sichuan University, Chengdu, 610041, China. E-mail: goumaling@scu.edu.cn
b Department of Neurosurgery, West China Hospital, Sichuan University,
Chengdu 610041, China
PpIX. The chemical structures of the lipophilic prodrugs were
confirmed by 1H-NMR (Fig. S1–S4, ESI†). When dispersed
in water, these lipophilic prodrugs could self-assemble into
NAs with average hydrodynamic diameters of 105.2–153.0 nm
and negative zeta potentials (Fig. S5, ESI†). Based on their
self-assembling abilities, a novel light-triggered nanosystem
(5S/SN38(10)-PpIX-NAs) was constructed via the co-assembly of
2S2C16-S/SN38(10) and C18-S/PpIX at an SN38/PpIX ratio of
8/1 (moles/moles). Co-assembling NAs of 2C18-S/SN38(10) and
c The School of Clinical Medicine, Southwest Medical University, Luzhou,
Sichuan Province 646000, China
d Department of Dermatology, West China Hospital, Sichuan University,
Chengdu 610041, China
† Electronic supplementary information (ESI) available: Experimental details,
1H-NMR, SN38 release from SN38(20)-PpIX-NAs, effects of SN38/PpIX ratios on
size distributions and oxidization of thioether, effects of irradiation time and
power on drug release, and quantum yield of singlet oxygen. See DOI: 10.1039/
c9cc06673j
C
18-S/PpIX (S/SN38(10)-PpIX-NAs) were used for comparative
studies to illuminate the role of multiple thioethers in enhancing
the responsiveness of light-triggered drug release. As shown in
‡ These authors contributed equally to this work.
13128 | Chem. Commun., 2019, 55, 13128--13131
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
SN38(10)-NAs displayed a high SN38 fluorescence (Fig. 1D). Such
different fluorescent properties indicated the successful co-assembly
of lipophilic prodrugs. Unlike the strategy that the photosensitizer
was entrapped physically into LHP-NAs, the co-assembly of lipo-
philic prodrugs could achieve the co-loading of SN38 and PpIX at
various SN38/PpIX ratios (Fig. S7, ESI†).
As the FRET signal is very sensitive to the separation
distance between the FRET donor and acceptor, the recovery
of the SN38 fluorescence can reflect the destruction of the NA
structure. The 5S/SN38(10)-PpIX-NAs and S/SN38(10)-PpIX-NAs
were exposed to 635 nm irradiation (0.2 W cmÀ2) for different
times; the kinetic changes in SN38 fluorescence were recorded
to characterize the destruction of the NA structure. As shown in
Fig. 2A, the fluorescence of 5S/SN38(10)-PpIX-NAs changed
gradually from red to blue upon irradiation. In contrast, there
was fluorescence observed for S/SN38(10)-PpIX-NAs under the
same conditions (Fig. 2B). This result was supported by the
kinetic changes in the emission spectra: sharply increased
SN38 fluorescence for 5S/SN38(10)-PpIX-NAs versus the slow
increase of SN38 fluorescence for S/SN38(10)-PpIX-NAs (Fig. 2A
and B). These results indicated that 5S/SN38(10)-PpIX-NAs were
destroyed much more easily than S/SN38(10)-PpIX-NAs upon
irradiation. This result was supported by the remarkable
change in the size distribution (new signal of larger particles,
Fig. S8A, ESI†) of 5S/SN38(10)-PpIX-NAs after irradiation, which
has been reported to indicate the destruction of the NA
structure.10 However, TEM did not reveal a different morphology
for 5S/SN38(10)-PpIX-NAs after irradiation, in comparison with
Fig. 1 Schematic illustration of the light-triggered 1O2-activated drug
release from the co-assembling NAs of lipophilic PpIX and lipophilic
SN38 prodrugs containing multiple thioethers (A). TEM image and size
distribution of 5S/SN38(10)-PpIX-NAs and S/SN38(10)-PpIX-NAs (B). Emis-
sion spectra of SN38(10)-NAs excited at 362 nm overlapped well with the S/SN38(10)-PpIX-NAs (emergence of many small nanorods, Fig. S8A,
absorption spectrum of PpIX-NAs, indicating a good FRET pair (C). The
ESI†). This result implied that the laser irradiation resulted in the
emission spectra and pictures under a UV lamp with excitation at 365 nm
of SN38(10)-PpIX-NAs and physical mixtures of S/PpIX-NAs and SN38(10)-
NAs (D), 10 mg mlÀ1, SN38 equivalent.
Fig. 1B, transmission electron microscopy (TEM) revealed that the
5S/SN38(10)-PpIX-NAs showed sphere-like shapes, whereas the
S/SN38(10)-PpIX-NAs displayed irregular morphologies. Dynamic
light scattering (DLS) analysis showed that the 5S/SN38(10)-PpIX-
NAs and S/SN38(10)-PpIX-NAs had an average hydrodynamic
diameter of 100.6 nm and 71.8 nm, respectively. Based on the
molecular weights of SN38 and the prodrugs, the loading capa-
cities of 5S/SN38(10)-PpIX-NAs and S/SN38(10)-PpIX-NAs were
calculated to be as high as 32.4% and 34.4%, respectively.
As the emission spectra of 2C18-S/SN38(10) NAs (S/SN38(10)-NAs)
and 2S2C16-S/SN38(10) NAs (5S/SN38(10)-NAs) overlapped with the
absorption spectrum of C18-S/PpIX NAs (S/PpIX-NAs) (Fig. 1C), the
co-assembly of lipophilic prodrugs of SN38 and PpIX could be
demonstrated by the fluorescence resonance energy transfer (FRET)-
caused quenching of the SN38 fluorescence. As shown in Fig. 1D,
Fig. 2 Kinetic changes in the fluorescence and emission spectra of 5S/
both 5S/SN38(10)-PpIX-NAs and S/SN38(10)-PpIX-NAs displayed a
red fluorescence under a UV lamp with excitation at 365 nm.
Fluorometric analysis showed a complete quenching of the SN38
fluorescence (400–600 nm) and a strong emission spectrum in the
wavelength range of 600–740 nm (corresponding to the emission
spectrum of PpIX excited at 426 nm, Fig. S6, ESI†), indicating an
obvious FRET. By contrast, physical mixtures of S/PpIX-NAs and
SN38(10)-PpIX-NAs (A) and S/SN38(10)-PpIX-NAs (B) with laser irradiation
at 0.2 W cmÀ2. The SN38 fluorescence could be rapidly recovered with 3 s
laser irradiation at 0.8 W cmÀ1 for 5S/SN38(10)-PpIX-NAs, which could
spell ‘‘SCU’’ in the colloidal solution. Analysis of the degradation of 5S/
SN38(10)-PpIX-NAs (C) and S/SN38(10)-PpIX-NAs (D) upon irradiation at
0.2 W cmÀ2. Insets: Kinetic changes in the peak areas of SN38, the SN38
prodrug and intermediates (expressed as percentages of the total area,
10 mg mlÀ1, SN38 equivalent).
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13128--13131 | 13129

View Article Online
ChemComm
Communication
destruction of the interior structure of 5S/SN38(10)-PpIX-NAs rather
than the change in the NA’s morphology. To support this, we also
found a hypsochromic band shift in the UV-vis spectra of NAs after
irradiation (Fig. S8B, ESI†), indicating a less aggregated stage of
NAs.20 We subsequently evaluated the effect of pH, the NA concen-
tration and the power density of irradiation on the recovery of the
SN38 fluorescence of 5S/SN38(10)-PpIX-NAs. It is shown that the
recovery of the SN38 fluorescence was not remarkably influenced by
the pH of the medium, but could be readily suppressed by increas-
ing the concentration of NAs (Fig. S9, ESI†). The SN38 fluorescence
was found to increase gradually with laser power (Fig. S10, ESI†),
1
which was explicable because the O2 was generated by the photo-
sensitizer in a laser power dependent manner.6 The SN38 fluores-
cence could be efficiently activated only with 3 s irradiation at
0.8 W cmÀ2, which was sensitive enough to spell letters in a colloidal
solution of 5S/SN38(10)-PpIX-NAs (Fig. 2A).
High-performance liquid chromatography (HPLC) was then
used to monitor the oxidative degradation of 2S2C16-S/SN38(10)
and 2C18-S/SN38(10) upon irradiation. As shown in Fig. 2C,
after 1 min of irradiation, 2S2C16-S/SN38(10) in 5S/SN38(10)-
PpIX-NAs was rapidly degraded into SN38 (9.8%) and a series of
intermediate products (around 76.0%, marked as 1–6, showing
SN38-like absorption spectra, Fig. S11, ESI†). The oxidized
products (4–6) could be further oxidized to 1–3 upon longer
irradiation. As the irradiation resulted in the disappearance of
the peaks of methylene protons in the a-position to thioethers
(2.50 and 3.22 ppm), these intermediates were speculated to be
the oxidized products of 2S2C16-S/SN38(10) as a result of the
oxidation of thioethers (Fig. S12, ESI†).21 In contrast, exposing
Fig. 3 Cumulative SN38 release from 5S/SN38(10)-PpIX-NAs (A) and
S/SN38(10)-PpIX-NAs (B) without irradiation (À) or after 5 min of irradiation
(+) at 0.2 W cmÀ2, 10 mg mlÀ1, SN38 equivalent.
S/SN38(10)-PpIX-NAs to laser light resulted in a much slower was rapidly converted to SN38 (80.7% within 2 h), which was
oxidation of 2C18-S/SN38(10), with only 9.8% of SN38 and remarkably higher than that without irradiation (6.8% within
46.1% of the oxidized intermediates obtained after 10 min 24 h). For S/SN38(10)-PpIX-NAs, 2C18-S/SN38(10) remained rela-
of irradiation. These results indicated that 2S2C16-S/SN38(10) tively stable, resulting in 43.7% SN38 release in 24 h. This release
was much more readily oxidized by 1O2 in comparison with was only around 2.7-fold faster than that without irradiation,
2C18-S/SN38(10). It is worth mentioning that C18-S/PpIX could indicating poor light sensitivity. These results implied that
be degraded upon irradiation, which agreed well with the 5S/SN38(10)-PpIX-NAs showed a much more sensitive light-
decrease of PpIX fluorescence (Fig. 2A and B). We have tried triggered SN38 release than S/SN38(10)-PpIX-NAs. To evaluate
to enhance the oxidation rate of 2C18-S/SN38 by increasing the effect of the conjugate position of the lipid-SN38 prodrug on
the amount of C18-S/PpIX. However, the oxidation rate of light-triggered SN38 release, lipophilic SN38 prodrugs were also
2C18-S/SN38 increased with the amount of C18-S/PpIX only within synthesized by conjugating fatty acids at the C20 position of SN38
a certain range; the maximum oxidation rate was achieved at an (i.e. 2S2C16-S/SN38(20) and 2C18-S/SN38(20), 1H-NMR shown in
SN38/PpIX ratio of 16/1 (Fig. S13, ESI†). On the other hand, the Fig. S16, ESI†). These prodrugs were then co-assembled with C18-
oxidation of 2C18-S/SN38(10) in S/SN38(10)-PpIX-NAs was much S/PpIX to obtain 5S/SN38(20)-PpIX-NAs and S/SN38(20)-PpIX-NAs,
faster than that in the physical mixture of S/SN38(10)-NAs and S/ which only displayed 16.8% and 1.7% SN38 release within 24 h
PpIX-NAs (oxidized product, 46.1% versus 17.2%) (Fig. S14, ESI†), after 5 min of irradiation, respectively (Fig. S17, ESI†). These
implying that the co-assembly of lipophilic prodrugs was the values were significantly lower than that of SN38(10)-PpIX-NAs,
prerequisite for the rapid oxidation of thioether. As 2C18-S/SN38 which might be ascribed to the high hydrolytic stability and great
could be oxidized by the indocyanine green-generated O2 upon steric hindrance around the ester bond at C20.22 Despite this,
1
irradiation at 808 nm (Fig. S15, ESI†), such a light-sensitive 5S/SN38(20)-PpIX-NAs still achieved around 10-fold faster SN38
nanosystem could also be constructed based on photosensitizers release than S/SN38(20)-PpIX-NAs, indicating that the presence of
in the near-infrared window.
multiple thioethers significantly enhanced the sensitivity of SN38
To verify whether the fast oxidation of 2S2C16-S/SN38(10) could release triggered by the light irradiation.
result in a more sensitive light-triggered SN38 release, the laser-
Interestingly, the degradation of the oxidized products of
treated NAs were incubated at 37 1C in 10 mM phosphate buffer 2S2C16-S/SN38(10) was correlated highly with their retention time
(PB, pH 7.4) and the sample was withdrawn for HPLC analysis. in the chromatographic curve (Fig. S18, ESI†), indicating that the
As shown in Fig. 3A, 2S2C16-S/SN38(10) after 5 min of irradiation oxidized products having a higher hydrophilicity were more readily
13130 | Chem. Commun., 2019, 55, 13128--13131
This journal is ©The Royal Society of Chemistry 2019

View Article Online
Communication
ChemComm
near-infrared light, which holds potential for future clinical
applications. Moreover, these NAs were constructed simply by
the co-assembly of the two lipophilic prodrugs, which displayed
many advantages, such as the controllable co-assembling drug
ratio, high loading capacity, one-step preparation and sensitive
light responsiveness to drug release. These benefits endow them
with great potential to serve as promising nanocarriers for the
spatially and temporally selective delivery of antitumor drugs.
This research was supported by the key research and develop-
ment projects of people’s liberation army (BWS17J036), the 1Á3Á5
project for disciplines of excellence, West China Hospital, Sichuan
University (ZYYC08007, ZYJC18017) and the Science and Technology
Project of Chengdu (2018-CY02-00041-GX).
Fig. 4 Cytotoxicities of SN38(10)-PpIX-NAs (A) and SN38(20)-PpIX-NAs
(B) without irradiation (À) or after 1 min of irradiation (+) at 0.4 W cmÀ2
[means Æ SD, n = 4], *p o 0.05, ***p o 0.001.
,
converted to SN38. This result suggested that the rapid SN38 release
from the light-irradiated 5S/SN38(10)-PpIX-NAs might be ascribed to
the increased hydrophilicity of the oxidized products, which made
them more available for hydrolysis. It is worth mentioning that the
SN38 release from 5S/SN38(10)-PpIX-NAs was dependent on the
irradiation time (Fig. S19, ESI†). The SN38 release increased with
irradiation time, approaching equilibrium after 5 min at a power of
0.2 W cmÀ2. Similarly, increasing the irradiation power could also
accelerate the degradation of 2S2C16-S/SN38, further resulting in a
faster SN38 release (Fig. S20, ESI†).
We then investigated the in vitro cytotoxicity of these NAs on
CT26 colonic cancer cells using MTT assay. As shown in Fig. 4A,
both 5S/SN38(10)-PpIX-NAs and S/SN38(10)-PpIX-NAs showed simi-
lar cytotoxicity (IC50, 2.1–3.7 mg mlÀ1), irrespective of whether
irradiation was applied or not. This was probably ascribed to the
rapid hydrolysis of phenolic ester in culture medium, which signifi-
cantly decreased the selectivity of light-triggered SN38 release
(Fig. S21A, ESI†). For 5S/SN38(20)-PpIX-NAs containing a stable ester
linker at C20, light irradiation resulted in around 5-fold higher SN38
release in the culture medium (Fig. S21B, ESI†), resulting in a higher
Conflicts of interest
There are no conflicts to declare.
Notes and references
1 Y. Zhou, H. Ye, Y. B. Chen, R. Y. Zhu and L. C. Yin, Biomacromole-
cules, 2018, 19, 1840–1857.
2 Y. F. Li, Y. M. Zhang and W. P. Wang, Nano Res., 2018, 11, 5424–5438.
3 E. R. Ruskowitz and C. A. DeForest, Nat. Rev. Mater., 2018, 3, 17087.
4 W. Tao, X. Ji, X. Zhu, L. Li, J. Wang, Y. Zhang, P. E. Saw, W. Li,
N. Kong, M. A. Islam, T. Gan, X. Zeng, H. Zhang, M. Mahmoudi,
G. J. Tearney and O. C. Farokhzad, Adv. Mater., 2018, 30, 1802061.
5 X. Ji, N. Kong, J. Wang, W. Li, Y. Xiao, S. T. Gan, Y. Zhang, Y. Li,
X. Song, Q. Xiong, S. Shi, Z. Li, W. Tao, H. Zhang, L. Mei and J. Shi,
Adv. Mater., 2018, 30, 1803031.
6 C. G. Dariva, J. F. J. Coelho and A. C. Serra, J. Controlled Release,
2019, 294, 337–354.
7 A. Sanchis, J. P. Salvador and M. P. Marco, Colloids Surf., B, 2019,
173, 825–832.
8 J. M. Silva, E. Silva and R. L. Reis, J. Controlled Release, 2019, 298,
154–176.
9 S. Y. Qin, A. Q. Zhang, S. X. Cheng, L. Rong and X. Z. Zhang,
Biomaterials, 2017, 112, 234–247.
cytotoxicity (IC50, 9.3 versus 3.2 mg mlÀ1, Fig. 4B). In contrast, 10 L. Minghui, S. Xuetan, Z. Ning, W. Wei, Y. Yang, J. Huizhen and
S/SN38(20)-PpIX-NAs displayed poor cytotoxicity, due to the poor
L. Wenguang, Adv. Sci., 2018, 5, 1800155.
11 N. Zhang, M. Li, X. Sun, H. Jia and W. Liu, Biomaterials, 2018, 159, 25–36.
12 Y. Li, T. Kang, Y. Wu, Y. Chen, J. Zhu and M. Gou, Chem. Commun.,
SN38 release both in the presence and in the absence of irradiation.
Interestingly, S/PpIX-NAs showed a much poorer light-induced
cytotoxicity (o100 fold) than free PpIX (Fig. S22, ESI†).
As S/PpIX-NAs displayed a comparable quantum yield of 1O2 to free
2018, 54, 1996–1999.
13 Y. J. Wang, D. Liu, Q. C. Zheng, Q. Zhao, H. J. Zhang, Y. Ma, J. K. Fallon,
Q. Fu, M. T. Haynes, G. M. Lin, R. Zhang, D. Wang, X. G. Yang,
L. X. Zhao, Z. G. He and F. Liu, Nano Lett., 2014, 14, 5577–5583.
PpIX (Fig. S23, ESI†), the poor cytotoxicity of S/PpIX-NAs was 14 C. Luo, J. Sun, D. Liu, B. J. Sun, L. Miao, S. Musetti, J. Li, X. P. Han,
Y. Q. Du, L. Li, L. Huang and Z. G. He, Nano Lett., 2016, 16, 5401–5408.
15 C. Luo, J. Sun, B. Sun, D. Liu, L. Miao, T. J. Goodwin, L. Huang and
probably due to their heterogeneous distributions in cells, which
limited oxidative reactions between the short-lived O2 (could only
1
Z. He, Small, 2016, 12, 6353–6362.
diffuse 10–20 nm) and adjacent biological macromolecules.23 There- 16 H. Wang, H. Xie, J. Wang, J. Wu, X. Ma, L. Li, X. Wei, Q. Ling, P. Song,
L. Zhou, X. Xu and S. Zheng, Adv. Funct. Mater., 2015, 25, 4956–4965.
17 B. Yang, K. Wang, D. Zhang, B. Sun, B. Ji, L. Wei, Z. Li, M. Wang,
fore, the cytotoxicity of SN38-PpIX-NAs mainly arose from the
released SN38 rather than the phototoxicity of PpIX.
X. Zhang, H. Zhang, Q. Kan, C. Luo, Y. Wang, Z. He and J. Sun,
In conclusion, a novel light-responsive nanosystem based on co-
assembling NAs of thioether-modified lipophilic prodrugs of SN38
and PpIX was developed. Upon 635 nm red light irradiation, PpIX
can generate 1O2 to oxidize hydrophobic thioether in situ to hydro-
philic sulfone, and thus to accelerate SN38 release. We demon-
strated that the presence of multiple thioethers significantly
enhanced the sensitivity of SN38 release triggered by the light
irradiation, due to the destruction of the interior structure of NAs.
Importantly, our preliminary data suggested that such NAs could
be constructed based on photosensitisers that can be activated by
Biomater. Sci., 2018, 6, 2965–2975.
18 F. H. Sobotta, F. Hausig, D. O. Harz, S. Hoeppener, U. S. Schubert
and J. C. Brendel, Polym. Chem., 2018, 9, 1593–1602.
19 J. Yoo, N. S. Rejinold, D. Lee, S. Jon and Y.-C. Kim, J. Controlled
Release, 2017, 264, 89–101.
20 I. Nabiev, F. Fleury, I. Kudelina, Y. Pommier, F. Charton, J. F. Riou,
A. J. P. Alix and M. Manfait, Biochem. Pharmacol., 1998, 55, 1163–1174.
21 J. Wang, X. Sun, W. Mao, W. Sun, J. Tang, M. Sui, Y. Shen and Z. Gu,
Adv. Mater., 2013, 25, 3670–3676.
22 A. G. Cheetham, Y.-C. Ou, P. Zhang and H. Cui, Chem. Commun.,
2014, 50, 6039–6042.
23 D. K. Chatterjee, L. S. Fong and Y. Zhang, Adv. Drug Delivery Rev.,
2008, 60, 1627–1637.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun., 2019, 55, 13128--13131 | 13131