Abstract
Photothermal therapy (PTT), a vanguard strategy in cancer/ocular
neovascularization treatment, has attracted considerable attention
owing to its precision, controllability, high efficacy, and minimal
side effects. Nevertheless, its inherent limitations necessitate
innovative solutions. One promising strategy is to develop reagents
with enhanced photothermal conversion efficiency under long-wavelength
laser irradiation. Carbon nanomaterials, known for their broad
absorption spectra, are currently hindered by single-wavelength lasers
in clinical treatments. In this study, we address this limitation by
coating mesoporous carbon nanomaterials (MCNs) with a lanthanide
oxysulfide up-conversion material (Y[2]O[2]S:Yb^3+,Er^3+), converting
980 nm light into visible light. This advancement enhances the
photothermal conversion efficiency of the produced MCNs/Ln/GD/FR
nanocomposites from 59.48% to 82.86%. Furthermore, the incorporation of
gambogic acid and doxorubicin intensifies the synergistic photothermal
therapy effect. A dual stimuli-responsive hydrogel (PNIPAM) is employed
to ensure controlled drug release and safe delivery to tumors.
Evaluations demonstrate that the MCNs/Ln/GD/FR nanocomposites exhibit
exceptional tumor targeting and evident photothermal synergistic
therapy effects on both subcutaneous and ocular in situ melanoma tumors
by activating tumor-suppressive signaling pathways while inhibiting
proliferation and differentiation-related pathways. These findings
might pave the way for the development of photothermal reagent and
offer valuable insights for advancing therapeutic strategies.
Subject terms: Nanotechnology in cancer, Drug delivery, Biomedical
materials, Cancer therapy
__________________________________________________________________
Photothermal therapy using photo agents developed with high
photothermal conversion efficiency under long-wavelength laser
irradiation has been investigated in preclinical cancer treatment. Here
this group reports coating mesoporous carbon nanomaterials with a
lanthanide oxysulfide up-conversion material (Y[2]O[2]S:Yb^3+, Er^3+)
with tumor targeting capability for in situ ocular melanoma tumor
treatment.
Introduction
In recent decades, considerable efforts have been dedicated to
exploring innovative tumor/pathological neovascularization suppression
techniques, encompassing a range of optical therapies such as
photothermal therapy (PTT), photodynamic therapy, and photoacoustic
therapy, as well as immunotherapy, radiotherapy, and gene
therapy^[70]1–[71]7. Each of these modalities presents unique
advantages and challenges, underscoring the need for continuous
optimization and broader implementation. Among these, PTT has emerged
as a cutting-edge treatment strategy, attracting considerable attention
due to its precision, controllability, high efficacy, and minimal side
effects^[72]8,[73]9. However, PTT still faces some inherent challenges
that require further addressal. The treatment of deep-seated cancers
remains unsatisfactory due to the constraints of laser penetration, and
the survival of residual tumor cells post-PTT may lead to
metastasis^[74]10,[75]11. Meanwhile, the efficacy of PTT can be
compromised by the development of heat shock proteins, which act as
protective shields for cancer cells^[76]12,[77]13. Therefore, it is
crucial to develop therapeutic agents that can reduce tumor cells’ heat
resistance and enhance lasers penetration depth to improve PTT efficacy
and effectively eradicate tumor cells at lower laser intensities.
The application of nanocomposites (NPs) in the diagnosis and treatment
of cancer or ocular neovascularization has received a lot of attention
with the development of nanomaterials and nanotechnology. Kinds of
functional drug molecules can be loaded through encapsulation,
embedding, adsorption, or covalent coupling. Chemical modification can
also enhance the selectivity and targeting of nanocomposites, improving
drug uptake in affected tissues and reducing drug diffusion into normal
tissue. Thus, developing nanocomposites with high photothermal
conversion efficiency that can be irradiated with long-wavelength laser
and used as carriers for functional drugs to amplify the photothermal
synergistic therapy effect may be an effective approach to destroy
target cells^[78]14.
Mesoporous carbon nanomaterials (MCNs), a promising category of drug
delivery nanocarriers, have been recognized as the next-generation
platform for drug delivery and biomedical applications^[79]15,[80]16.
Their performances make them promising candidates for use in
photothermal therapy, drug administration, and other
fields^[81]17,[82]18. Carbon nanomaterials exhibit strong absorption of
near-infrared light, converting it into heat energy to induce cancer
cells ablation^[83]19. They are recognized not only for their
absorption in the near-infrared spectrum but also for their broad
absorption band, which, when combined with single-wavelength lasers
currently employed in clinical treatments, presents certain
limitations. Specifically, the narrow wavelength range of
single-wavelength lasers may not fully leverage the wide absorption
band of carbon nanomaterials, thereby limiting their photothermal
conversion efficiency and therapeutic efficacy. To enhance the
photothermal conversion efficiency of the proposed nanocomposites under
980 nm laser irradiation, a layer of lanthanide oxysulfide
up-conversion material (Y[2]O[2]S:Yb^3+,Er^3+) was coated onto MCNs.
This coating converts 980 nm light into visible light, effectively
expanding the absorption wavelength range of the inner MCNs. The
similar calcination preparation method of lanthanide oxysulfide with
MCNs facilitates the synthesis of the proposed MCNs/Ln
NPs^[84]1,[85]20,[86]21.
Building on these foundations, highly photothermal conversion
efficiency MCNs/Ln up-conversion nanocomposites are fabricated in this
study. Gambogic acid (GA), a natural heat shock protein inhibitor that
helps cells withstand heat-induced damage^[87]22, and a broad-spectrum
anticancer drug, doxorubicin (DOX), are loaded to enhance the
photothermal synergistic therapy effect. Meanwhile, A dual
stimuli-responsive hydrogel (PNIPAM) coating ensures controlled drug
release and safe delivery to tumors^[88]23,[89]24. In addition, the
grafting of folic acid (FA) and cell membrane penetrating peptide R8
onto the nanocomposites further enhances their synergistic therapeutic
effect by improving their efficient utilization in tumors. Figure [90]1
illustrates the schematic diagram of the synthesis of MCNs/Ln/GD/FR NPs
and their application in photothermal synergistic therapy.
Fig. 1. Schematic diagram.
[91]Fig. 1
[92]Open in a new tab
The synthesis of MCNs/Ln/GD/FR NPs and their application in
photothermal synergistic therapy on melanoma.
In this work, kinds of nanocomposites are synthesized and thoroughly
characterized in terms of their morphology, influence factors, and
photothermal conversion performances. Their biocompatibility and
photothermal synergistic treatment effect are also extensively assessed
in vitro and in vivo. In addition, subcutaneous and ocular in situ
melanoma tumor models, long recognized as invaluable tools in
fundamental research owing to their ability to mimic the disease’s
complex biology and clinical manifestations^[93]25–[94]27, are
established to compare the efficacy of the photothermal synergistic
therapy provided by these nanocomposites. The developed MCNs/Ln/GD/FR
NPs display evident tumor (tumor cells) targeting and a pronounced
photothermal synergistic therapy effect, which can significantly reduce
the activity of OCM-1 cells and the growth of melanoma tumors. These
findings not only offer insights on the exploration of low-temperature
photothermal reagents but also provide a potential formulation to
advancing photothermal therapy in anti-tumor, anti-angiogenesis, and
antibacterial applications.
Results and discussion
Material characterization of MCNs/Ln NPs
To initiate the synthesis of MCNs/Ln NPs with MCNs as the core, a
straightforward scheme is presented in Fig. [95]2A. The synthesis of
MCNs involves a combination of solvothermal and hydrothermal methods to
prepare their precursors (MCNs pre), followed by an annealing
treatment. The size of the MCNs precursors can be modulated by
adjusting the dropping rate of the Pluronic-127 aqueous solution
(Supplementary Fig. [96]1), whereas the dropping rate of deionized
water has a negligible effect (Supplementary Fig. [97]2). Then, a layer
of lanthanide basic carbonate out-shell is coated onto the MCNs
precursor (MCNs/Ln pre) using the solvothermal method. MCNs/Ln NPs are
then obtained after calcination at an N[2]/S atmosphere for 2 h. To
determine the appropriate calcination temperature, thermogravimetric
analysis (TGA) of the MCNs precursors and MCNs/Ln precursors was
carried out in an N[2] atmosphere (Supplementary Fig. [98]3). Three
distinct weight loss plateaus are observed for both samples. Around 35
% of the original sample remains for the MCNs/Ln precursors when the
weight loss achieves equilibrium, which is higher than that of the MCNs
precursors, where about 29 % of the original sample remains. Thus,
three temperatures, including 700, 750 and 800 °C, were employed to
anneal the precursors.
Fig. 2. Material characterization of MCNs/Ln NPs.
[99]Fig. 2
[100]Open in a new tab
A A simple scheme of the synthesis of MCNs/Ln NPs. B SEM images of the
MCNs precursors, MCNs/Ln precursors, and MCNs/Ln NPs treated at 700 °C
in N[2]/S atmosphere for 2 h. C TEM and HRTEM images of MCNs and
MCNs/Ln NPs. D XRD patterns of the samples obtained at different
conditions. E BET curves of the MCNs and MCNs/Ln NPs. F Emission
spectra of the different samples under the irradiation of 980 nm laser.
G The effect of lanthanide nitrate amounts on the effectiveness of the
MCNs/Ln NPs’ photothermal conversion. H The photothermal imaging of the
nanocomposites with different irradiation times as well as
concentrations. I The comparison of the photothermal conversion
efficiency of Y[2]O[2]S:Yb^3+,Er^3+, MCNs and MCNs/Ln NPs (0.75 mmol,
data are presented as mean ± SD, n = 3 independent experiments, one-way
ANOVA multiple comparison test, the absence of a P-value signifies no
statistical significant difference between the groups (P > 0.05)). (J)
The possible mechanism of the improvement of the photothermal
conversion efficiency of the MCNs/Ln NPs.
Then, SEM images of the MCNs pre, MCNs/Ln pre, and final MCNs/Ln NPs
are presented in Fig. [101]2B. The diameter of the product increases
sharply from 100 nm to 200 nm after being covered with the lanthanide
basic carbonate shell. The thickness of the out-shell is approximately
50 nm. After the annealing treatment, the size of the products reduces
to around 130 nm. The integrity of the nanocomposites can be disrupted
as the calcination temperature increases. When the annealing
temperature is raised to 800 °C, nearly no undamaged nanocomposites are
observed (Supplementary Fig. [102]4). TEM and HRTEM images of the
products further confirm the successful coating of the
Y[2]O[2]S:Yb^3+,Er^3+ out-shell (Fig. [103]2C). TEM imaging of MCNs
reveals that the sample is composed of 80 nm nanoparticles with
numerous holes. The size increases from the initial 90 nm to about
130 nm when a layer of Y[2]O[2]S:Yb^3+,Er^3+ out-shell is coated
(MCNs/Ln NPs). The HRTEM image with clear lattice fringes on the
exterior further corroborates this conclusion. The lattice fringe with
a d-spacing of 0.293 nm corresponds to the (101) plane of hexagonal
Y[2]O[2]S.
X-ray diffraction patterns were further used to characterize the phase
structures of the precursors and the MCNs/Ln NPs obtained at 700 °C. As
shown in Fig. [104]2D, no noticeable diffraction peaks are observed in
the MCNs and MCNs/Ln precursors, indicating poor crystalline. Strong
and narrow diffraction peaks are found in the MCNs/Ln NPs, all peaks
are consistent with those of the standard card (JCPDS#24-1424),
indicating the formation of hexagonal Y[2]O[2]S after the annealing
treatment. The structural alterations of the products were further
verified by FTIR spectra (Supplementary Fig. [105]5). Compared to the
MCNs precursors, a peak at 1525 cm^−1 can be observed in the file of
the MCNs/Ln precursors, with the peaks appeared at 948, 1238, and
1474 cm^−1 in the FTIR spectrum of the MCNs precursors vanished,
indicating the formation of different substances, which is ascribed to
Y(OH)CO[3]^[106]28. After being calcined, a strong absorption peak,
ascribed to the Ln-S bond, can be observed at 500 cm^−1, indicating the
formation of oxysulfide in the final nanocomposites.
The specific surface area of MCNs and MCNs/Ln NPs was also compared
using N[2] adsorption-desorption isotherms, as depicted in
Fig. [107]2E. The samples exhibit VI-type isotherms with an
H4-hysteresis loop, which is typical of activated carbon-type solids.
The Brunauer–Emmett–Teller (BET) surface area of the MCNs is measured
to be 523.31 m^2 g^−1, which dropped to 184.55 m^2 g^−1 after being
coated in a layer of lanthanide oxysulfide due to its larger molar
molecular weight. Holes with a diameter of around 3 nm remain,
suggesting that the generated MCNs/Ln NPs could be an effective drug
delivery carrier (inset in Fig. [108]2E). Meanwhile, the hydrophilic
performance of the products was also investigated. Carbonaceous
materials are known for their highly hydrophobic characteristics. When
a shell of Y[2]O[2]S:Ln^3+ is coated, the contact angle of MCNs
decreases from the initial 113.16^o to 41.19^o, close to that of water
(36.63^o), suggesting an obvious improvement of hydrophilicity. This is
primarily due to the formation of a hydrophilic Y[2]O[2]S:Ln^3+
out-shell, whose contact angle is just approximately 27^o
(Supplementary Fig. [109]6). The enhanced hydrophilicity of the
produced MCNs/Ln NPs benefits their use as a platform for drug loading
and biomedical applications.
Furthermore, the impact of the coating of the Y[2]O[2]S:Yb^3+,Er^3+
out-shell on the products’ fluorescence performance was investigated.
When exposed to a 980 nm laser, pure MCNs exhibit only a weak emission
band at 830 nm. Conversely, the fluorescence spectra of
Y[2]O[2]S:Yb^3+,Er^3+ with different doping amounts of Yb^3+ ions
primarily consist of intense visible light emissions at 550 nm and
690 nm, along with a weaker near-infrared light at 820 nm. Notably,
when doped with an optimized amount of 10% Yb^3+ ions, the sample
exhibits the strongest emission among the three tested samples
(Supplementary Fig. [110]7). When a layer of Y[2]O[2]S:Yb^3+, Er^3+
out-shell was wrapped around MCNs, the signal of Y[2]O[2]S:Yb^3+,Er^3+
in the visible region decreases dramatically, while the near-infrared
emission band (840 nm) increases weakly, indicating that the light
emitted by Y[2]O[2]S:Yb^3+,Er^3+ is primarily absorbed by the inner
MCNs. Meanwhile, the intensity of the 840 nm emission band can be
adjusted by varying the encapsulation quantities of the outer
Y[2]O[2]S:Yb^3+,Er^3+ up to 0.75 mmol. Further increasing the
lanthanide nitrates to 1 mmol results in a decline in the 840 nm
emission band and the appearance of multiple weak emission bands at
550 nm and 690 nm (indicated using “*”) of Y[2]O[2]S:Yb^3+,Er^3+,
ascribed to the limited light absorption capacity of the inner MCNs
(Fig. [111]2F). The UV−vis−NIR absorption spectra of the different
samples provide additional evidence that the presence of the MCNs core
augments their capacity to absorb a wider spectrum of light wavelengths
(Supplementary Fig. [112]8). The obvious photothermal conversion
properties of the obtained MCNs/Ln nanocomposites, demonstrated
subsequently, further confirmed that the light absorbed by MCNs is
primarily converted into heat energy.
To assess the photothermal conversion properties of the synthesized
nanocomposites, all samples were exposed to a 980 nm laser for 10 min
at a distance of 3 cm from the optical fiber head to the center of the
nanocomposite dispersion. Compared with MCNs, which exhibited a
temperature increase from 29 to 51.5 °C within 10 min, the coating of a
shell of Y[2]O[2]S:Yb^3+,Er^3+ significantly enhanced the temperature
elevation of the nanocomposites, especially for the sample prepared
with 0.75 mmol of lanthanide nitrate, whose temperature increases from
29 to 65 °C (Fig. [113]2G). Aligning with the nanocomposites’
fluorescence spectra (Fig. [114]2F), either increasing or decreasing
the amount of lanthanide nitrate results in lower temperature
increases. Meanwhile, the calcination temperature of the MCNs/Ln
precursor influenced the photothermal conversion properties of the
product. As shown in Supplementary Fig. [115]9A, the sample calcined at
700 °C displayed better photothermal conversion efficiency compared to
those calcined at other temperatures (750 °C and 800 °C), whose
decrease can be attributed to the loss of inner MCNs at higher
calcination temperatures, as evidenced by SEM images (Supplementary
Fig. [116]4). Furthermore, the laser power density also affected the
final temperature of the nanocomposites. As the laser power density
increased, the temperature of the nanocomposites clearly improved
(Supplementary Fig. [117]9B). The influence of irradiation time and
nanocomposites concentration on the mixture temperature was also
carried out using photothermal imaging. The mixture temperature
exhibited a clear dependency on both concentration and irradiation time
under the same irradiation situation (Fig. [118]2H).
To gain a deeper understanding of the absorption characteristics of
MCNs under multiple light sources, 660 nm and 980 nm lasers were
employed to irradiate a 200 µg mL^−¹ MCNs dispersion. As shown in
Supplementary Fig. [119]10, the significant temperature increase
observed with both lasers, compared to either laser alone, underscores
the ability of MCNs to absorb and convert light from multiple
wavelengths into thermal energy. Moreover, the photothermal conversion
stability of the nanocomposites was evaluated by cycling the
temperature between its maximum value and room temperature by turning
the laser on and off. Five successive cycles were carried out, with the
highest temperature in each cycle approaching 65 °C, indicating clear
repeatability and photothermal conversion stability of the
nanocomposites (Supplementary Fig. [120]11).
Furthermore, the photothermal conversion efficiency (η) of
Y[2]O[2]S:Yb^3+,Er^3+, MCNs and MCNs/Ln NPs (0.75 mmol) was compared.
The η values were calculated from the temperature variation curves
(Supplementary Fig. [121]12) and the linear relationship (Supplementary
Fig. [122]13), according to the photothermal conversion equations
(detailed in the Methods section -Synthesis of MCNs/Ln
NPs)^[123]29–[124]32. The η value for MCNs/Ln NPs was determined to be
82.86%, much higher than that of MCNs (59.48%) and
Y[2]O[2]S:Yb^3+,Er^3+ (19.8%), indicating an enhanced photothermal
conversion efficiency for MCNs/Ln NPs (Fig. [125]2I).
To explore the distinct core-shell structure of MCNs/Ln nanocomposites
that facilitates photothermal conversion, fluorescence spectroscopy was
conducted on pure Y[2]O[2]S:Yb^3+,Er^3+, mixtures of MCNs with
Y[2]O[2]S:Yb^3+,Er^3+, and MCNs/Ln nanocomposites. Despite similar
solution colors, the mixture of MCNs and Y[2]O[2]S:Yb^3+,Er^3+
exhibited a much stronger fluorescence signal compared to that of
MCNs/Ln nanocomposites (Supplementary Fig. [126]14). This suggests that
the seamless core-shell integration in the MCNs/Ln nanocomposites
allows efficient photon capture by the inner MCNs from the outer
Y[2]O[2]S:Yb^3+,Er^3+ shell. In contrast, simple physical blending of
MCNs with Y[2]O[2]S:Yb^3+,Er^3+ does not facilitate effective photon
capture, hindering subsequent thermal energy conversion.
Based on these findings, it is evident that the Y[2]O[2]S:Yb^3+,Er^3+
out-shell on MCNs significantly enhances photothermal conversion
efficiency. The potential mechanism behind this enhancement,
illustrated in Fig. [127]2J, can be attributed to the meticulously
designed core-shell structure of the MCNs/Ln nanocomposites. The MCNs,
serving as the core, are engineered to efficiently absorb 980 nm light,
a wavelength highly promising for laser-driven photothermal
applications. This direct absorption markedly elevates the overall
photothermal conversion efficiency of the MCNs/Ln system. The
Y[2]O[2]S:Yb^3+,Er^3+ out-shell encapsulating the MCNs also plays a
crucial role. When exposed to a 980 nm laser, the electrons in the
ground states of Yb^3+ ions are excited and transition to their excited
states. These excited electrons then transfer to the excited states of
Er^3+ ions, along with other electrons that are excited from the ground
state of Er^3+ ions. When these electrons relax back to the ground
state of Er^3+ ions, they emit light at various wavelengths, such as
550 nm and 690 nm. The inclusion of the MCNs core introduces an
additional absorption process, where photons emitted from different
excited states of Er^3+ ions, and the initial 980 nm photons, are
absorbed by the MCNs. This absorption process broadens the spectral
range absorbed by the MCNs, further enhancing the overall absorbance
and photothermal conversion efficiency. The seamless integration of the
core-shell structure of the MCNs/Ln nanocomposites ensures the
efficient capture of photons emitted by the Y[2]O[2]S:Yb^3+,Er^3+
out-shell by the inner MCNs, ultimately contributing to the observed
enhancement in photothermal conversion efficiency.
Material characterization of MCNs/Ln/GD/FR NPs
As illustrated in Fig. [128]3A, hydrophilic MCNs/Ln NPs, distinguished
by their exceptional photothermal conversion performance, were further
utilized as carriers. Subsequently, a series of experiments were
conducted to optimize the synergistic therapeutic efficacy of the
resultant product. These experiments included the physical adsorption
of GA (termed MCNs/Ln/G), the encapsulation with a thermally and
GSH-responsive PNIPAM hydrogel outer shell (yielding MCNs/Ln/G/pNIPAM
NPs), the incorporation and entrapment of DOX within the unique grid
structure of the pNIPAM hydrogel (denoted as MCNs/Ln/GD), and the
covalent conjugation of FA and R8 via amide bond formation (identified
as MCNs/Ln/GD/FR). Following these preparations, kinds of
characterization techniques were employed to evaluate the diverse
properties of the products.
Fig. 3. Characterization of MCNs/Ln/GD/FR NPs.
[129]Fig. 3
[130]Open in a new tab
A A simple scheme of the synthesis of MCNs/Ln/GD/FR NPs, and its TEM
image. B The influence of GSH concentrations on the release of DOX (%)
in MCNs/Ln/D NPs aqueous dispersion (Data are presented as mean ± SD,
n = 3 independent experiments, two-way ANOVA multiple comparison test).
C Release percentage of DOX (%) in MCNs/Ln/D NPs aqueous dispersion
with or without irradiation, as well as the addition of GSH (Data are
presented as mean ± SD, n = 3 independent experiments, two-way ANOVA
multiple comparison test). D Fluorescence microscope images of the
MCNs/Ln/D NPs aqueous dispersion with or without irradiation. E A
simple release illusion of the temperature and GSH stimuli-responsive
MCNs/Ln/GD/FR NPs. F PTT synthetic effect of the kinds of
nanocomposites on OCM-1 cell viability under the irradiation of 980 nm
laser (Data are presented as mean ± SD, n = 3 independent experiments,
one-way ANOVA multiple comparison test, the absence of a P-value
signifies no statistical significant difference between the groups
(P > 0.05)). G Calcein-AM/PI double staining of OCM-1 cells incubated
with various nanocomposites under 980 nm laser irradiation, with an
untreated group as control (medium). H Flow-cytometric analysis of
OCM-1 cells incubated with different nanocomposites under 980 nm laser
irradiation, with a control group irradiated solely with 980 nm laser
(medium). I The relative change of the expression of HSP70 protein
under different conditions (β-actin was used as an internal control,
HT: heat treatment). J The therapeutic effect of different components
in MCNs/Ln/GD/FR NPs.
Initially, the zeta potentials of the nanocomposites obtained at
different stages were measured (Supplementary Fig. [131]15). The zeta
potential values dropped from an initial 16.4 mV to 2.7 mV upon loading
GA, and further fell to −17.3 mV when coated with the PNIPAM hydrogel.
This value changed to 8.2 mV with the addition of DOX. The conjugation
of FA and R8 on the surface cause the zeta potential to shift from −7.6
to −2.9 mV. The variations in zeta potential at each step further
confirm the effective modification of the nanocomposites. In addition,
the changes in hydrated particle size at several stages also verify the
successful coating of Y[2]O[2]S:Yb^3+,Er^3+and the PNIPAM hydrogel. The
hydrated particle size increased from the initial 100 nm of MCNs to
200 nm for MCNs/Ln, and ultimately to 400 nm for MCNs/Ln/GD
(Supplementary Fig. [132]16).
Comparing with the TEM image of MCNs/Ln NPs (Fig. [133]2C), a shell of
amorphous material can be distinctly observed on the exterior of the
Y[2]O[2]S:Yb^3+,Er^3+ out-shell (Fig. [134]3A), which is ascribed to
the PNIPAM hydrogel. Meanwhile, the drug loading and releasing
properties of GA and DOX under different conditions were studied. The
encapsulation efficiency (EE) and loading efficiency (LE) of DOX in the
final nanocomposites were higher than those of GA (Supplementary
Fig. [135]17). About 58% of GA was released after 48 h of gentle
shaking (Supplementary Fig. [136]18).
Due to the temperature and glutathione (GSH) stimuli-responsive
properties of the PNIPAM hydrogel where DOX was entrapped, the release
rate kinetics of DOX were further measured under different conditions.
With the enhancement in GSH concentration, a greater amount of DOX was
released over time, confirming that GSH facilitates the decomposition
of the PNIPAM hydrogel. This is attributed to the presence of disulfide
bonds (-S-S-) in the cross-linker hydrogel (Fig. [137]3B). Heating the
nanocomposites also facilitated the release of DOX. Compared with the
sample without heating, it took about 1 min for the decomposition of
the PNIPAM hydrogel and the release of DOX, the addition of GSH in the
solution accelerated the release of DOX (Fig. [138]3C). Fluorescence
microscope images of the MCNs/Ln/D aqueous dispersion with or without
irradiation further confirmed the controllable release of DOX from the
nanocomposites (Fig. [139]3D). These results imply that the obtained
temperature and GSH stimuli-responsive MCNs/Ln/GD/FR NPs can be safely
delivered in vivo before reaching the tumor or being heated
(Fig. [140]3E).
Then, the cell biocompatibility of the various nanocomposites was
investigated. The cell survival rate of ARPE-19 cells was initially
evaluated after a 24 h incubation with several nanocomposites at
different concentrations, including 50, 100, and 200 μg mL^-1
(Supplementary Fig. [141]19A). The survival rates of ARPE-19 cells
remained over 80% when the concentration of the nanocomposites reached
200 μg mL^−1, indicating obvious biocompatibility of the synthesized
nanocomposites. The biocompatibility of the MCNs/Ln/GD/FR NPs was
further assessed using three other cells types, including hRMEC, 293 T,
and L929 cells, and compared after 24 h incubation. As shown in
Supplementary Fig. [142]19B, when the concentration of MCNs/Ln/GD/FR
NPs exceeded 200 μg mL^−1, the survival rates of the four types of
cells remained above 80%. Further enhancing the concentration of the
nanocomposites to 400 μg mL^−1 resulted in similar outcomes, except for
hRMEC cells, whose survival rates decreased to about 70%. All these
data demonstrate that the generated MCNs/Ln/GD/FR NPs possess apparent
biocompatibility.
In vitro Synergistic treatment of MCNs/Ln/GD/FR NPs
Prior to evaluating the PTT synthetic effect of the synthesized
nanocomposites on cells, the photothermal effect of a pure 980 nm laser
on the cells was investigated. OCM-1 cells were exposed to a 980 nm
laser with varying power densities for 10 min at a distance of 3 cm
(the distance from the optical fiber head to the middle of the medium).
The cell survival rate was subsequently measured. According to
Supplementary Fig. [143]20A, 85% of cells remained alive when the power
density was increased to 280 mW cm^−2. The survival rate decreased
drastically as the power density was further increased, demonstrating
an enhanced photothermal inhibitory effect. An ARPE cell experiment
further confirmed that a 280 mW cm^−2 980 nm laser is safe for ARPE
cells (Supplementary Fig. [144]20B). Thus, a laser power density of
280 mW cm^−2 was used in subsequent investigations.
Then, the photothermal impacts of different nanocomposites on OCM-1
cells under irradiation were evaluated. The concentrations of
nanocomposites were set at 50, 100, and 200 μg mL^−1, respectively. As
shown in Fig. [145]3F, the survival rate of OCM-1 cells dropped
substantially as the nanocomposites concentration increased, indicating
a concentration dependency of PTT synthetic effects. Meanwhile, the PTT
synthetic effect among different nanocomposites at the same
concentration was also compared. Take the concentration of 100 μg mL^−1
as an example, approximately 60% of OCM-1 cells survived after being
incubated with MCNs nanoparticles for 24 h. When a shell of
Y[2]O[2]S:Yb^3+,Er^3+ out-shell was coated, only about 20% of cells
survived, highlighting the enhanced PTT efficacy of MCNs/Ln NPs.
Furthermore, the loading of GA, DOX, coupled with the conjugation of FA
and R8, amplified the nanocomposites’ PTT synthetic therapeutic impact,
with only about 13% of cells survived in the MCNs/Ln/GD/FR group. This
value decreases to 6% when the concentration of the nanocomposites was
increased to 200 μg mL^−1. Meanwhile, the decrease in cell survival
rates for different groups also verify the successful modification of
the MCNs/Ln NPs.
Calcein-AM/PI double staining was also used to evaluate the PTT
synthetic effect of the produced nanocomposites on OCM-1 cells. The
nanocomposites were concentrated at 200 μg mL^−1, and the irradiation
time of a 280 mW cm^−2 980 nm laser at 3 cm was set to 10 min with the
medium group untreated. As illustrated in Fig. [146]3G, the quantity of
living cells (green fluorescence) in the visual field decreases
steadily with the modification progress of the nanocomposites.
Meanwhile, the number of dead cells (red fluorescence) increased
noticeably with the coating of Y[2]O[2]S:Yb^3+, Er^3+, the loading of
GA, DOX, and the connection of FA and R8. Only a few living cells could
be seen in the MCNs/Ln/GD/FR group, suggesting its outstanding PTT
synthetic effect on OCM-1 cells.
The cell apoptosis of OCM-1 cells treated with different samples was
further examined using flow cytometry to precisely depict the PTT
synergistic impact of the different nanocomposites. The nanocomposites
were concentrated at 200 μg mL^-1, and the irradiation time of a
280 mW cm^−2 980 nm laser at 3 cm was set to 10 min for all groups. The
percentage of live cells decreased considerably with the progress of
the modification (Fig. [147]3H). Compared to MCNs (21.1%), the
percentage of live cells reduces to 8.14% when a layer of
Y[2]O[2]S:Yb^3+,Er^3+ out-shell was coated, suggesting the greatly
improved PTT impact of MCNs/Ln NPs. The percentage of cells in the late
stage of apoptosis increases considerably with the addition of GA and
DOX. The modification of FA and R8 can reduce the percentage of live
cells to 5.9%, much lower than that in other groups, further
demonstrating the improved tumor targeting after modification.
Meanwhile, the expression of heat shock protein (HSP)-associated
proteins in OCM-1 cells under heating treatment was studied using
western blot analysis (Fig. [148]3I, [149]J and Supplementary
Fig. [150]21, [151]22). The level of HSP70 decreased from 1.37 to 0.65
under the intervention of MCNs/Ln/GD/FR NPs and laser irradiation,
suggesting that the administration of GA can downregulate HSP70
expression in these cells, which facilitates the apoptosis of OCM-1
cells by minimizing their resistance and enhancing anticancer effects
during heating treatment. In further photothermal inhibition
experiments on HSP70-knockdown OCM-1 cells achieved via small
interferencing RNA (SiRNA), an enhanced sensitivity to photothermal
therapy was observed. This was evident when the cells were treated with
100 μg mL^-1 MCNs/Ln nanocomposites and subjected to laser irradiation
or exposed directly to a 45 °C incubation environment, resulting in a
significant decrease in cell viability compared to other groups
(Supplementary Fig. [152]22). This provides further evidence supporting
the effectiveness of photothermal synergistic therapy.
Increased uptake of MCNs/Ln/GD/FR NPs in vitro and in vivo
The tumor (tumor cell) targeting efficacy of MCNs/Ln/GD/FR NPs was
investigated both in vitro and in vivo. Given that OCM-1 cells possess
a higher density of folate targeting receptors compared to ARPE-19
cells^[153]33,[154]34, MCNs/Ln/GD/FR NPs conjugated with FA are
expected to more effectively target OCM-1 cells than ARPE-19 cells
under identical conditions (Supplementary Fig. [155]23 and
Fig. [156]4A, [157]B). Leveraging the red fluorescence signal of DOX,
200 μg mL^−1 MCNs/Ln/D/FR NPs were co-cultured with ARPE-19 and OCM-1
cells for 6 h and subsequently examined using a confocal microscope. As
depicted in Fig. [158]4C, with the cell nucleus stained by DAPI (blue),
a greater intensity of red DOX fluorescence signals was observed in
OCM-1 cells compared to ARPE-19 cells, indicating a higher uptake of
nanocomposites by OCM-1 cells. The fluorescence quantitative analysis
further confirmed that OCM-1 cells exhibited a higher DOX concentration
(Fig. [159]4D), indicating that MCNs/Ln/D/FR NPs possess a superior
targeting potential for OCM-1 tumor cells. Besides, the tumor cell
targeting of MCNs/Ln/D and MCNs/Ln/D/FR NPs to OCM-1 cells was observed
using confocal microscopy. After being co-cultured with 200 μg mL^−1
MCNs/Ln/D and MCNs/Ln/D/FR NPs for 6 h, a greater accumulation of
MCNs/Ln/D/FR NPs was observed surrounding the OCM-1 cell compared to
MCNs/Ln/D NPs, which were evenly dispersed in the medium (Supplementary
Fig. [160]24), suggesting that MCNs/Ln/D/FR NPs had a stronger
targeting capacity for the OCM-1 tumor cell.
Fig. 4. The tumor (tumor cells) targeting and antitumor efficacy of
MCNs/Ln/GD/FR NPs against subcutaneous melanoma.
[161]Fig. 4
[162]Open in a new tab
A The expression of folate receptors in OCM-1 and ARPE-19 through
Western Blot (n = 3 independent experiments). B one scheme of the
expression of folate receptors in OCM-1 and ARPE-19. C The confocal
microscope luminescent imaging of ARPE-19 and OCM-1 cells co-cultured
with 200 μg mL^-1 MCNs/Ln/D/FR NPs for 6 h. D Fluorescence quantitative
analysis diagram of cells co-cultured with 200 μg mL^−1 MCNs/Ln/D/FR
NPs. E A simple scheme illustrating the procedures of subcutaneous
melanoma modeling and intervention. F Thermal imaging photos of nude
mice injected with different samples through the tail vein with the
irradiation of 980 nm laser (n = 5 mice). G Temperature curves of the
tumor of different groups (Data are presented as mean ± SD, n = 5 mice
(for Laser group, n = 4 mice), one-way ANOVA multiple comparison test,
the absence of a P-value signifies no statistical significant
difference between the groups (P > 0.05)). H The Y content in the
tumors of different groups (Data are presented as mean ± SD, n = 3
mice, one-way ANOVA multiple comparison test, the absence of a P-value
signifies no statistical significant difference between the groups
(P > 0.05)). I H&E staining of the tumor tissues of MCNs/Ln/GD and
MCNs/Ln/GD/FR groups (n = 3 mice). J Histogram of the temperature of
different organ tissues obtained from different groups (Data are
presented as mean ± SD, n = 3 mice, one-way ANOVA multiple comparison
test, the absence of a P-value signifies no statistical significant
difference between the groups (P > 0.05)). K The Y content in different
organ tissues in MCNs/Ln/GD/FR groups (Data are presented as mean ± SD,
n = 4 mice, one-way ANOVA multiple comparison test, the absence of a
P-value signifies no statistical significant difference between the
groups (P > 0.05)). L Photos of nude mice in different time intervals
after treatment. M Tumor volume change curves of the different groups
(Data are presented as mean ± SD, n = 3 mice, one-way ANOVA multiple
comparison test, the absence of a P-value signifies no statistical
significant difference between the groups (P > 0.05)). N Photograph and
qualities of the resected tumor tissues in different groups (Data are
presented as mean ± SD, n = 3 mice, one-way ANOVA multiple comparison
test, the absence of a P-value signifies no statistical significant
difference between the groups (P > 0.05)).
The tumor-targeting effect of MCNs/Ln/GD/FR NPs was also investigated
in nude mice. A schematic illustration outlines the procedures for
creating subcutaneous melanoma modeling and intervention
(Fig. [163]4E). Four groups, including PBS, Laser, MCNs/Ln/GD and
MCNs/Ln/GD/FR, were established. Mice were injected with 200 μL of PBS,
MCNs/Ln/GD (2 mg kg^−1), and MCNs/Ln/GD/FR (2 mg kg^−1) PBS dispersion
via the caudal vein. After 4 h post-injection, the mice were sedated,
and various measurements were conducted. Initially, the tumor color in
the four groups was compared. The tumor color in the MCNs/Ln/GD/FR
group appeared much darker than that in the other groups (Supplementary
Fig. [164]25A), indicating a greater accumulation of MCNs/Ln/GD/FR NPs
at the tumor site. Simultaneously, the tumor sites were irradiated with
a 280 mW cm^−2 980 nm laser, and thermal imaging photographs of the
tumor sites were captured. A notably higher tumor temperature was
observed in mice injected with MCNs/Ln/GD/FR NPs, indicating a greater
integration of nanocomposites in the tumor tissue (Supplementary
Fig. [165]25B). The thermal imaging images and temperature variation
curves obtained at different time intervals also demonstrated that the
tumor temperature in the MCNs/Ln/GD/FR group was much higher than that
in the other three groups for the same irradiation duration
(Fig. [166]4F, [167]G). Inductively Coupled Plasma (ICP) analysis of Y
content in tumor tissue from different groups further confirmed that
more MCNs/Ln/GD/FR NPs accumulated in the tumor tissue (Fig. [168]4H).
The comparison of HE images of tumor tissues in the MCNs/Ln/GD and
MCNs/Ln/GD/FR groups also supported this conclusion, with more
MCNs/Ln/GD/FR black nanocomposites, evident cells necrosis, and
inflammatory cell infiltration observed in the tumor tissue of
MCNs/Ln/GD/FR group (Fig. [169]4I and Supplementary Fig. [170]26
(enlarged images), green narrows indicating the nanocomposites).
Thermal imaging photographs and ICP analysis of Y content in heart,
liver, spleen, lung, kidney, and tumor tissues of nude mice in the
Laser, MCNs/Ln/GD and MCNs/Ln/GD/FR groups were also performed under
irradiation. Compared to other tissues in the MCNs/Ln/GD/FR group or
tumor tissues in other groups, the local temperature of tumor tissues
in the MCNs/Ln/GD/FR group is considerably higher (Supplementary
Fig. [171]27 and Fig. [172]4J). ICP analysis of Y content in different
tissues further verified that more MCNs/Ln/GD/FR NPs accumulated in
tumors, implying that the designed MCNs/Ln/GD/FR NPs exhibited distinct
tumor targeting capacity while remaining reasonably safe for other
tissues (Fig. [173]4K).
Enhanced antitumor efficacy of MCNs/Ln/GD/FR NPs against subcutaneous tumor
in vivo
To assess the synergistic therapeutic effect of MCNs/Ln/GD/FR NPs on
tumors in vivo, two types of melanoma models were established,
including subcutaneous tumors and in situ ocular tumors, through
localized injection of OCM-1 cells. For the treatment of subcutaneous
melanoma, variations of body weight and tumor volume in nude mice,
along with photographic records, were documented. Throughout the
treatment period, the body weight of the nude mice remained stable,
indicating evident biocompatibility of the proposed nanocomposites
(Supplementary Fig. [174]28). In contrast, tumor volume varied
substantially across different groups. Unlike the tumor volumes in the
PBS, Laser, DOX, and MCNs/Ln/GD groups, which increased over time, the
volume values in the MCNs/Ln/GD/FR group decreased gradually. After 21
days of treatment, the tumors of some mice disappeared, demonstrating
the obvious tumor suppressive effect of the MCNs/Ln/GD/FR NPs
(Fig. [175]4L, [176]M). Photographs of the tumors during the treatment
process and after removal from the mice further supported these
findings (Fig. [177]4L, N). Histological examinations using hematoxylin
and eosin (H&E) staining were conducted on the heart, liver, spleen,
lung, and kidney tissues of the nude mice across all five groups. No
visible harm was observed in these organs. Hepatocytes in liver samples
appeared normal, no pulmonary fibrosis was found in lung samples, the
glomerulus structure in kidney sections was visible, and no necrosis
was evident in the histological samples. These results indicate that
the use of MCNs/Ln/GD/FR NPs did not induce any clear pathological
changes in the heart, liver, spleen, lung, or kidney (Supplementary
Fig. [178]29).
Antitumor efficacy of MCNs/Ln/GD/FR NPs against in situ ocular melanoma
To evaluate the therapeutic effect of MCNs/Ln/GD/FR NPs on in situ
ocular tumors, an ocular melanoma model was established through
subretinal injection of OCM-1 cells into the eyes. The firefly
luciferase gene (Luc) was introduced into OCM-1 cells for in vivo
monitoring of tumor size. Figure [179]5A provides a schematic
illustration of the melanoma in situ modeling and intervention process.
Fig. 5. Synergistic treatment effect of MCNs/Ln/GD/FR NPs in melanoma in
situ.
[180]Fig. 5
[181]Open in a new tab
A A simple scheme illustrating the procedures of melanoma in situ
modeling and intervention. B In vivo tumor bioluminescence images of
the eyes in OCM-1-Luc tumor-bearing BALB/c nude mice with different
treatments (n = 5 mice). C The corresponding relative tumor
luminescence variation in different groups (Data are presented as
mean ± SD, n = 5 mice, one-way ANOVA multiple comparison test, the
absence of a P-value signifies no statistical significant difference
between the groups (P > 0.05)). D The eyeball weight of different
groups (Data are presented as mean ± SD, n = 6 mice, one-way ANOVA
multiple comparison test, the absence of a P-value signifies no
statistical significant difference between the groups (P > 0.05). E
Thermal imaging photos of nude mice with different treatments under the
irradiation of 980 nm laser (n = 5 mice). F Eye temperature variation
of the nude mice for the different groups under the irradiation of
980 nm laser (Data are presented as mean ± SD, n = 5 mice, one-way
ANOVA multiple comparison test, the absence of a P-value signifies no
statistical significant difference between the groups (P > 0.05)). G
H&E staining on the eyes in different groups (n = 3 mice). H Scheme of
the potential effects of MCNs/Ln/GD/FR NPs in melanoma in situ using
two administration methods.
Prior to treatment, an evaluation was conducted to assess the ocular
tumor targeting capability of the constructed MCNs/Ln/GD/FR
nanocomposites. The experiment involved four groups: PBS, DOX,
MCNs/Ln/GD, MCNs/Ln/GD/FR, all administered via tail intravenous
injection. 6-hour post-administration, the eyeballs of the experimental
animals in each group, along with various organs from the MCNs/Ln/GD/FR
group, were collected. The distribution of the material was evaluated
using the fluorescent properties of DOX. The results, as depicted in
Supplementary Figs. [182]30 and [183]31, demonstrate that the
fluorescence intensity of DOX in the eyeballs of the MCNs/Ln/GD/FR
group was significantly higher compared to the other experimental
groups. This finding suggests that the constructed MCNs/Ln/GD/FR
nanocomposites can effectively cross the blood-ocular barrier and
target ocular tissues, even when administered via tail intravenous
injection, enabling targeted delivered to ocular tumor (Supplementary
Fig. [184]30). When comparing the distribution of the MCNs/Ln/GD/FR
nanocomposites in other organ tissues, it is observed that, after a
6-hour period of blood circulation, the nanocomposites are primarily
distributed in the eyes and two major metabolic organs, namely the
liver and kidneys, with minimal impact on other organs (Supplementary
Fig. [185]31). These results indicate that the nanocomposites exhibit a
favorable safety profile and ocular tumor-targeting capability,
establishing a solid foundation for further treatment of in situ ocular
tumors.
In this study, two injection methods were compared: intravitreal
injection (IVT) and intravenous injection (IV). In vivo optical
imaging, variations of eyeball weight, and temperature changes in the
eye under irradiation were recorded. After the administration of
D-Luciferin Sodium Salt (Solarbio) for 15 min, the normalized eye
fluorescence intensity of different groups was compared. The tumor
bioluminescence intensity in the MCNs/Ln/GD/FR (IV) group was
significantly lower than that in other groups, indicating a distinct
inhibition effect of the nanocomposites (Fig. [186]5B). To
quantitatively evaluate the tumor growth process, the bioluminescence
intensity of tumors on the day before treatment (7^th day) was used as
a baseline, and the tumor bioluminescence intensity curves of the
different groups were recorded over the treatment period
(Fig. [187]5C). Compared with the PBS and DOX groups, the
bioluminescence intensity variation trend of the MCNs/Ln/GD and
MCNs/Ln/GD/FR groups was much slower. Notably, in the group treated
with MCNs/Ln/GD/FR(IV), the average bioluminescence intensity value on
the 21^st day was close to the baseline, significantly lower than that
of the other groups, indicating effectively inhibited.
Meanwhile, the eyeballs were carefully collected and weighed on the
21^st day. Among the different groups, only the average eyeball weight
in the MCNs/Ln/GD/FR(IV) and MCNs/Ln/GD/FR(IVT) groups was comparable
to that of the healthy eyes group, with no significant difference
(Fig. [188]5D), indicating the clear synergistic treatment effect of
the nanocomposites. The modification with FA and R8 appears to improve
therapy efficiency. During the treatment, the temperature variation of
the eye under irradiation was also recorded using an infrared thermal
imager. Compared to the MCNs/Ln/GD(IV) group, a significantly higher
variation in eye temperature was observed in the MCNs/Ln/GD/FR(IV)
group, with statistical significance (P = 0.0011). This variation is
nearly comparable to that of the MCNs/Ln/GD/FR(IVT) group, further
suggesting that the linkage of FA and R8 enhances the curative effect
by improving tumor targeting, even when administered via tail
intravenous injection (Fig. [189]5E, [190]F). Furthermore, the
Kaplan–Meier survival curve demonstrated that photothermal treatment
using MCNs/Ln/GD/FR(IV) facilitated the most comprehensive tumor
resection and enhanced the overall survival rate of mice compared to
the other groups (Supplementary Fig. [191]32).
Next, H&E staining was performed on the eyes of different groups.
Obvious distortion of the eye was observed in the PBS group due to the
extensive proliferation of OCM-1 cells. The tumors in the DOX(IV) and
MCNs/Ln/GD(IV) groups were smaller but not as small as those in the
MCNs/Ln/GD/FR groups using both injection strategies. However, retinal
detachment and a smaller vitreous body were observed in the
MCNs/Ln/GD/FR(IVT) group, which could be attributed to the higher eye
temperature under irradiation. In contrast, almost no retinal
detachment or vitreous body shrinkage was observed in the eyes of the
MCNs/Ln/GD/FR(IV) group (Fig. [192]5G). Combined with the above
evaluation results, this mild PTT treatment appears to be healthier
(Fig. [193]5H).
In addition, fluorescent staining was conducted on the treated eyes in
the MCNs/Ln/GD/FR(IV) groups to investigate the variation of associated
heat shock proteins in vivo. As shown in Supplementary Fig. [194]33 and
Supplementary Fig. [195]34, compared to untreated tumor tissues, the
observed decrease in the levels of heat shock proteins, such as HSP70
and HSP90, in the treated tumor tissues further confirms that the
application of MCNs/Ln/GD/FR NPs can effectively reduce the production
of heat shock proteins within tumor tissues, ultimately contributing to
the enhancement of the photothermal effect.
In vivo safety evaluation of MCNs/Ln/GD/FR NPs
The potential toxicity of the proposed MCNs/Ln/GD/FR NPs was further
assessed following PTT treatment. Assessments included hemolysis
analysis, H&E staining of organs, ICP analysis of Y content, and
biochemical blood analysis in the mice administered with MCNs/Ln/GD/FR
NPs via tail intravenous injection over a one-month period. The
hemolysis rate of red blood cells was approximately 1.33% when the
concentration of MCNs/Ln/GD/FR NPs reached 200 μg mL^−1, indicating
that the MCNs/Ln/GD/FR NPs had no noticeable impact on immune
regulation or heme response and demonstrated good blood
biocompatibility (Fig. [196]6A and Supplementary Fig. [197]35). H&E
staining of various tissues revealed no pathological changes in the
heart, liver, spleen, lung, or kidney over the treatment duration
(Fig. [198]6B). The ICP analysis of Y content in different tissues
showed a substantial decrease with prolonged treatment (Fig. [199]6C),
comparing to initial Y levels (Fig. [200]4K). Specifically, Y content
in the liver reduced from an initial 20 mg kg^−1 to around 2.4 mg kg^−1
after one month of treatment, indicating that the synthesized
MCNs/Ln/GD/FR NPs are metabolizable. Concurrently, blood samples were
collected from nude mice on the 15^th and 30^th day following PTT
treatment with MCNs/Ln/GD/FR NPs administered via tail intravenous
injection (n = 5), with mice injected with PBS serving as the control
group. Blood biochemical tests were then performed. Figure [201]6d
illustrates the findings, showing that only aspartate aminotransferase
(AST) levels remained slightly higher than the control group on both
days 15 and 30, yet within an acceptable range. Liver and renal
markers, as well as other parameters, were all within normal limits.
These findings suggest that there was no significant inflammation or
infection in the mice treated with the synthesized MCNs/Ln/GD/FR NPs as
PTT agents.
Fig. 6. In vivo safety evaluation after PPT treatment.
[202]Fig. 6
[203]Open in a new tab
A Hemolysis experiment of MCNs/Ln/GD/FR NPs with different
concentrations (Data are presented as mean ± SD, n = 5 (for 200 group,
n = 4) independent experiments, one-way ANOVA multiple comparison test,
the absence of a P-value signifies no statistical significant
difference between the groups (P > 0.05)). B H&E staining of the organs
from nude mice injected with MCNs/Ln/GD/FR NPs in different treatment
periods (n = 3 mice, one week, two weeks, one month). C The ICP results
of Y content in different organs after one-month treatment (The same
organs of 5 mice were homogenized together to make one sample). D Serum
biochemistry results of the blood obtained from nude mice after
intravenous administration of MCNs/Ln/GD/FR NPs for 0, 15, and 30 ds,
with mice injected with PBS as the Blank group (Glu: blood sugar. CREA:
creatinine. Urea: UREA. URIC: URIC acid. TBIL: total bilirubin. TP:
total protein. ALB: albumin. GLOB: globulin. ALT: alanine transaminase.
AST: glutamic-oxalacetic transaminase) (Data are presented as
mean ± SD, n = 3 mice, one-way ANOVA multiple comparison test, the
absence of a P -value signifies no statistical significant difference
between the groups (P > 0.05)). E TUNEL staining on the treated eyes in
the MCNs/Ln/GD/FR(IV) and MCNs/Ln/GD/FR(IVT) groups (R: Retina, T:
Tumor, n = 3 mice).
In addition, the safety of MCNs/Ln/GD/FR NPs in ocular tissues was
further conducted using TUNEL staining in the MCNs/Ln/GD/FR(IV) and
MCNs/Ln/GD/FR(IVT) groups to investigate changes in apoptotic proteins.
As depicted in Fig. [204]6E, TUNEL staining revealed obvious green
staining in the tumor regions of both experimental groups, indicating a
substantial presence of apoptotic cells in the tumor tissues and
validating the effectiveness of the nanocomposites. However, notable
differences were observed in the non-tumor regions between the two
groups. Specifically, the non-tumor area in the MCNs/Ln/GD/FR(IVT)
group also exhibited bright green fluorescence, suggesting, along with
HE staining results, that the dispersion of materials in other ocular
regions (such as the retinal region) could cause substantial damage
during photothermal treatment. Conversely, the MCNs/Ln/GD/FR(IV) group
showed reduced green fluorescence in the non-tumor area, indicating
minimal tissue damage. This is attributed to the nanocomposites’s
ability to accumulate directly at the tumor site after passing through
the blood-eye barrier when administered via the tail vein, resulting in
lower presence at other sites and minimal impact on non-tumor areas.
Possible antitumor mechanism of MCNs/Ln/GD/FR NPs
To investigate the underlying mechanism by which MCNs/Ln/GD/FR NPs
inhibit melanoma tumor growth, single-cell RNA sequencing technology
was employed to analyze the compositional changes in melanoma tumor
cells derived from mice in both the MCNs/Ln/GD/FR and PBS groups.
Although these data may not fully reflect the characteristics of
primary tumors, they are of substantial significance in elucidating the
action mechanism of MCNs/Ln/GD/FR in suppressing tumor growth.
Moreover, this investigation provides invaluable directional insights
and data support for the potential application of these materials in
the treatment of various types of tumors. Specifically, this study
collected 9650 cells for the PBS group and 7361 cells for the
MCNs/Ln/GD/FR group (indicated as “T” in the figures). Based on
principal component analysis of gene expression profiles, 8 clusters of
cells with specific gene expression profiles were identified, including
CSRNP1^+ tumor, PHLDA1^+ tumor, ANK3^+ tumor, BIRC5^+ tumor, ZNF90^+
tumor, TEX14^+ tumor, and OASL^+ tumor (Fig. [205]7A–[206]C).
Significant differences were observed in the expression of heat shock
protein (HSP)-related coding genes under the intervention of
photothermal treatment, suggesting that photothermal ablation activated
HSP-related signaling pathways (Fig. [207]7D). The significant
downregulation of HSP-associated proteins in both general tumors and
ZNF90^+ tumors in the treatment groups indicates that the GA released
by MCNs/Ln/GD/FR NPs can downregulate HSPs expression, prevent damaged
cancer cells from being repaired, and weaken thermal resistance caused
by photothermal therapy (Fig. [208]7E).
Fig. 7. Single-cell RNA sequencing analysis of PBS and MCNs/Ln/GD/FR groups
in melanoma.
[209]Fig. 7
[210]Open in a new tab
A–C 8 clusters with specific gene expression profiles were obtained
from two treatment groups. D The expression of heat shock protein
(HSP)-related coding genes. E The variation of HSP-associated proteins
in different clusters in the MCNs/Ln/GD/FR groups. F The percentage of
different clusters in the two treatment groups. G Survival differences
between PHLDA1-positive cell High/Low enrichment groups and PHLDA1 gene
High/Low expression groups analyzed by two-sided log-rank tests (no
multiple comparison adjustment applied, as these were pre-specified
independent hypotheses). H Top 20 differential expression genes between
PHLDA1 + and PHLDA1- tumor cells in MCNs/Ln/GD/FR group. I–L
Associations between PHLDA1 expression groups and clinical stages. I
T-stage distribution. J N-stage distribution. K M-stage distribution. L
Stage distribution (Two-sided Chi-square tests with Benjamini-Hochberg
adjustment). M, N Kaplan-Meier survival analysis for different PHLDA1
groups.
To elucidate the influence of MCNs/Ln/GD/FR NPs on tumor cytotoxicity,
Pseudobulk analysis was performed to identify differentially expressed
genes. SERPINE2, AKAP12, SLC16A13, and CXCL8 gene showed a significant
increase in expression under MCNs/Ln/GD/FR NPs intervention, whereas
ARC, NR4A3, and RIMKLB gene expression were upregulated in the PBS
group (Supplementary Fig. [211]36A). Apoptosis inhibitory factor
containing caspase recruitment domain (ARC) is an effective inhibitor
of cell apoptosis that is induced and confers chemoradioresistance in
human breast cancer^[212]35,[213]36. AKAP12 (a-kinase anchoring protein
12), which acts as a regulator of mitosis by anchoring key signaling
proteins such as PKA, PKC, and cyclins, is a protein kinase C substrate
and potential tumor suppressor^[214]37. It can be downregulated by
several oncogenes and strongly suppressed in various cancers. The
enhancement of AKAP12 levels in the MCNs/Ln/GD/FR group facilitates the
prognosis of tumor-bearing mice. Besides, the area under the curve
(AUC) method with normalized gene set data from MSigDB’s ‘Hallmark
Pathways’ was further used to analyze the results^[215]38. The
normalized enrichment scores (NES) of each pathway were plotted. Under
MCNs/Ln/GD/FR NPs intervention, numerous signaling pathways that
inhibit tumor cell proliferation, differentiation, and metastasis were
activated, such as TNFa signaling via NFKB, G2M checkpoint, and MYC
targets V1, besides significant changes in the tumor metabolic
microenvironment (Supplementary Fig. [216]37 and Supplementary
Fig. [217]S38). These changes facilitate the inhibition of metabolic
pathways favorable for tumor proliferation and metastasis.
Meanwhile, a distinct cells subgroup, labeled as PHLDA1^+ tumor cells,
can be observed under the intervention of MCNs/Ln/GD/FR NPs
(Supplementary Fig. [218]36B, [219]C). This subgroup accounted for
21.3% of the total population in the MCNs/Ln/GD/FR group, while only
5.3% in the PBS group (Fig. [220]7F). Among the different gene
expression patterns of the marker genes in the pseudo-time trajectory,
the expression level and duration of PHLDA1 gene increased obviously,
indicating that the intervention of MCNs/Ln/GD/FR NPs can facilitate
high-expression of PHLDA1 gene in tumor cells (Fig. [221]7G).
Downregulation of PHLDA1 can promote cancer cell proliferation and
migration, resulting in poor prognosis^[222]35. PHLDA1^+ and PHLDA1^-
tumor cells were further compared with only the top 20 upregulated and
downregulated genes across the entire dataset considered. Different
from PHLDA^-tumors, the upregulated and downregulated gene profiles in
PHLDA1^+ tumor cells were inconsistencies for the MCNs/Ln/GD/FR group
(Fig. [223]7H). It is reported that PHLDA1 may mediate drug resistance
in receptor tyrosine kinase-driven cancers. Its level can be
downregulated when breast cancer or renal cancer patients receive
RTK-targeted therapy or HER2^+ breast cancer cells receive trastuzumab
treatment, resulting in an increased resistance to drugs^[224]34.
Exposed to MCNs/Ln/GD/FR NPs environments, the IC50 values for various
anti-tumor drugs decrease obviously in PHLDA1^+ tumor cells relative to
PHLDA^- ones, indicating enhanced anti-tumor efficacy (Supplementary
Fig. [225]39). Thus, MCNs/Ln/GD/FR NPs may enhance their anti-tumor
effects by upregulating PHLDA1 level in tumor cells, leading to the
differentiation of tumor cells towards sensitivity to DOX. Besides,
pseudobulk analysis on PHLDA1^+ tumor cells was also carried out. The
protein-protein interaction (PPI) network of up-regulated
differentially expressed genes (DEGs) revealed that BMP2 and SNAI2
exhibited strong interactions with numerous other up-regulated proteins
(Supplementary Fig. [226]40A). Conversely, the PPI network of
down-regulated DEGs demonstrated interactions among EGR2, EGR3, DUSP2,
NR4A1, NR4A2, and BTG2 (Supplementary Fig. [227]40B). Enrichment
analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and
Genomes (KEGG) revealed that the tumor cells exhibited enhanced
oxidative stress response and activation of tumor necrosis factor
superfamily pathways under the intervention of MCNs/Ln/GD/FR NPs, such
as “response to oxidative stress” and “positive regulation of tumor
necrosis factor production” (Supplementary Fig. [228]40C, [229]D).
These findings suggest that the cytotoxic effect of MCNs/Ln/GD/FR NPs
on tumors can be enhanced by upregulating cell cycle-related protein
expression to inhibit tumor mitosis and downregulating
apoptosis-related proteins to promote tumor cell apoptosis.
The impact of PHLDA1^+ cell levels on overall survival rate was also
analyzed using CIBERSORT, and the features of the PHLDA1^+ tumor group
identified by scRNA to TCGA-UVM were mapped in both binary and
continuous models. According to the expression level of PHLDA1 in
TCGA-UVM, high and low expression groups were divided. The results
indicated that the prognosis of high PHLDA1 expression groups was
better, they had fewer cases of T4 staging and M1 metastasis, as well
as lower stage classification (Fig. [230]7I–[231]L), indicating that
higher levels of PHLDA1^+ are favorable to the overall survival rates
(Fig. [232]7M, [233]N).
Thus, the developed MCNs/Ln/GD/FR NPs was suggested to have a
significant cytotoxic effect on melanoma. It can activate
tumor-suppressive signaling pathways while inhibit proliferation and
differentiation-related signaling pathways, ultimately induce tumor
cell differentiation and increase sensitivity to DOX, result in an
enhanced therapy effect.
In this study, double stimuli-responsive carbon-based lanthanide
oxysulfide up-conversion nanocomposites with obvious photothermal
conversion efficiency were developed to treat ocular in situ and
subcutaneous melanoma-1. The introduction of a coating of lanthanide
oxysulfide up-conversion material (Y[2]O[2]S:Yb^3+,Er^3+) on MCNs can
enhance its photothermal conversion efficiency from an initial 59.48%
to 82.86% under 980 nm laser irradiation. The temperatures can rise
dramatically to 50 °C within 150 s under the influence of a
280 mW cm^−2 980 nm laser, a level sufficient to eradicate cancer
cells. While the loading of GA, DOX, as well as the linkage of FA and
R8, can further enhance the photothermal synergistic therapy effect,
and the coating of double stimuli-responsive hydrogel (PNIPAM) can
ensure controlled drug release and safe delivery to tumors. Cell and
animal evaluation results indicate that the developed MCNs/Ln/GD/FR NPs
had a strong photothermal synergistic therapeutic efficacy, which can
greatly reduce OCM-1 cell activity and melanoma growth. Single-cell RNA
sequencing results revealed that MCNs/Ln/GD/FR NPs can enhance the
therapy effect by activating tumor-suppressive signaling pathways while
inhibiting proliferation and differentiation-related signaling
pathways, ultimately induce tumor cell differentiation and increasing
sensitivity to DOX. Insights gained from this research may prove to be
vital to the development of cutting-edge therapeutic applications of
optical therapy in the future.
Methods
Ethics Statement
Female Balb/c nude mice (3–4 weeks old, 20 g) were acquired from the
SLRC Laboratory (Shanghai, China) following approval and in compliance
with Ethical Committee of EYE and ENT Hospital of Fudan University
(IACUC-DWZX-2021-013), in strict accordance with the ARVO Statement for
the Use of Animals in Ophthalmic and Vision Research. All experimental
protocols adhered to the IACUC Guideline, which mandated a maximum
permissible tumor volume of 2000 mm³ for mice. Tumor dimensions did not
exceed 2 cm in any axis throughout the experiments. Tumor progression
and animal welfare were monitored every 2–3 days, with documentation
confirming adherence to ethical and volumetric constraints throughout
the study.
Materials and reagents
Phenol, formalin, NaOH, Pluronic F127 (Mw = 12600, PEO106PPO70PEO106),
lanthanide oxides (Yb[2]O[3], Y[2]O[3], and Er[2]O[3]), sulfur, urea,
gambogic acid (GA), doxorubicin (DOX), sodium dodecyl sulfate (SDS), N,
N’-bis(acryloyl) cystamine (BAC), N-Isopropylacrylamide (NIPAM),
potassium persulfate (KPS), N-hydroxysuccinimide (NHS), and
1-ethyl-3-(3-(dimethylamino)propyl) carbodiimide (EDC), folic acid
(FA), and glutathione (GSH) were purchased from Aladdin Industrial Inc.
The R8 peptide was obtained from QYAOBIO (ChinaPeptides Co., Ltd.).
Gibco supplied Fetal Bovine Serum, Trypsin-EDTA (1), and Fetal Bovine
Serum. Sigma-Aldrich supplied DMEM-high glucose and Dimethyl sulfoxide.
Shanghai Biyuntian Biotechnology Co., Ltd supplied Hoechst 33342, while
Tongren Chemical Technology Co., Ltd supplied Cell Counting
Kit-8(CCK-8), Calcein-AM, and PI. All reagents were used just as they
were obtained, without further purification.
Lanthanide nitrates (Y(NO[3])[3], Yb(NO[3])[3], and Er(NO[3])[3]) were
synthesized by dissolving Yb[2]O[3], Y[2]O[3], and Er[2]O[3],
respectively, in dilute HNO[3] under agitation with heating (keep
boiling), and then evaporating the water to yield the desired product.
Synthesis of various nanocomposites
Synthesis of MCNs nanoparticles
Ordered MCNs precursor can be prepared using a two-step method as
reported previously^[234]39. In a typical synthesis, 0.6 g of phenol
and 2.1 ml of formalin aqueous solution (37 wt %) were initially added
to 15 ml of NaOH aqueous solution (0.1 M) under stirring at 70 °C for
0.5 h. The mixture was then heated to 66 °C, and 15 ml of triblock
copolymer Pluronic F127 aqueous solution (0.96 g F127) was added while
stirring. After 2 h of reaction, 50 ml of water was slowly added under
stirring. The color of the reaction solution changed from colorless to
pink and then to scarlet. The reaction was stopped after 18 h of
stirring at 66 °C, and the reaction mixture was allowed to stand until
the deposit dissolved. Subsequently, 17.8 mL of the crimson solution
was added to 56 mL of deionized water under stirring for 30 min before
being transferred to a 100 mL autoclave, which was maintained at 130 °C
for 24 h. The resulting MCNs precursor was then centrifuged, washed
three times with distilled water, and dried at room temperature. The
final MCNs nanoparticles were obtained by heating at 700 °C for 2 h
under an N[2] atmospheres at a heating rate of 2 °C min^−1.
Synthesis of MCNs/Ln NPs
The monodisperse MCNs/Ln precursor was synthesized according to our
previous report with minor modification^[235]40. In a typical synthesis
of the MCNs/Ln precursor, varying stoichiometric ratios (5%, 10%,
15%Yb; 2%Er) and amounts (0.5, 0.75, 1 mmol) of lanthanide nitrate were
added into 30 mL of MCNs precursor aqueous solution (0.125 g) with the
assistance of ultrasonication for 15 min. Then, 3.0 g of urea was
dissolved in the solution by stirring. The mixture was then transferred
to a round-bottom flask and heated at 85 °C for 3 h with stirring. The
precursor was collected by centrifugation, washed three times with
deionized water and ethanol, and dried at 60 °C in air. The final
MCNs/Ln NPs were obtained by heating them at different temperatures for
2 h in an N[2]/S atmosphere at a heating rate of 2 °C min^−1.
Photothermal Performance of the Synthesized Nanocomposites
The photothermal performance of the synthesized nanocomposites was
performed as follows: After exposure to a 980 nm laser (3 W, Beijing
Laserwave OptoElectronics Tech. Co., Ltd.) with different power
densities (Power density (mW cm^−2) = Laser power (mW)/Laser beam area
(cm^2)), the temperature changes of the obtained nanocomposite aqueous
dispersion with different concentrations were monitored by a JK808
Multichannel temperature tester and a FOTRIC 220 series hand-held
thermal imager.
The photothermal conversion efficiency (η) of Y[2]O[2]S:Yb^3+,Er^3+,
MCNs and MCNs/Ln NPs (0.75 mmol) was calculated according to the
photothermal conversion equations:
[MATH: η=(
hAΔTmax−
Hs)/<
/mo>[P(1−10−Abs)] :MATH]
where ΔT[max] is the maximum change in temperature, H[s] is the heat
relating to the absorption value of water, Abs is the absorption value
of the three kinds of materials at 980 nm, and P is the laser power.hA
was determined using another equation:
[MATH: hA=mCp/τ, :MATH]
where τ is the slope of the linear data from the cooling time vs −ln(θ)
(Supplementary Fig. [236]13) and C[p] and m are the water heat capacity
and the mass, respectively. It was also needed to introduce θ
(ΔT/ΔT[max]).
Synthesis of MCNs/Ln/G NPs
Taking advantage of the large specific surface area and physical
adsorption capacity of MCNs, gambogic acid (GA) was loaded into the
MCNs/Ln NPs. 20 mg of MCNs/Ln NPs and 5 mg of GA were initially added
to a brown glass bottle. Then, 4 mL of anhydrous ethanol was added with
the assistance of ultrasonication for 20 min, and the obtained mixture
was further stirred in the dark for 24 h. After centrifugation at
10000 rpm for 10 min, MCNs/Ln/G NPs were collected. Meanwhile, the
content of GA in the supernatant solution was measured using a UV−vis
Spectrophotometer.
The measurements of GA release from MCNs/Ln/G NPs were performed by
dispersing the obtained MCNs/Ln/G NPs in 1 mL of PBS solution and
transferred them into a dialysis tube, which was then placed into a
15 mL centrifuge tube containing 10 mL of PBS under gentle shaking at
37 °C. Subsequently, 1 mL of supernatant was collected, and another
1 mL of fresh PBS solution was added at certain time intervals (0,
0.25, 0.5, 1, 2, 4, 6, 8, 12, 24, 36 and 48 h). The amount of GA was
further calculated by comparing it with the standard curve of the GA
aqueous solution.
Synthesis of MCNs/Ln/G/pNIPAM NPs
Firstly, 0.144 g of sodium dodecyl sulfate (SDS) was added into 50 mL
of deionized water in a three-neck flask and stirred in an argon
atmosphere for 30 min. Then, 0.5658 g of N-Isopropylacrylamide (NIPAM)
dispersed in 5 mL of deionized water and 5 mL of N, N’-bis(acryloyl)
cystamine (BAC) (0.052 g) ethanol solution were added to the above
solution, which was further stirred for 1 h until a transparent
solution was obtained. Subsequently, 5 mL of the MCNs/Ln/G aqueous
mixture was added to the solution under vacuum and stirred for another
1 h in an argon atmosphere. When 1 mL of potassium persulfate (KPS)
(0.0135 g) solution was added, the obtained mixture was heated to 60 °C
for 1 h. After the reaction, the flask was immediately cooled in an ice
bath, and the precipitate was separated by centrifugation at 10000 rpm
for 10 min. The obtained MCNs/Ln/G/pNIPAM NPs were dispersed in
deionized water and then preserved at 4 °C.
Synthesis of MCNs/Ln/GD NPs
Utilizing the mesh-like properties of the pNIPAM layer to further
adsorb and load doxorubicin (DOX). The obtained MCNs/Ln/G/pNIPAM NPs
were initially dispersed in 2 mL of deionized water in a brown glass
container. Next, 10 mg of DOX was added and dissolved in the mixture.
After 24 h of steady stirring, the precipitate was separated by
centrifugation at 10000 rpm for 10 min, and the supernatant was
retained. The concentration of DOX in the supernatant solution was
determined using a UV−vis Spectrophotometer.
Due to the temperature and GSH stimuli-responsive properties of PNIPAM
hydrogels where DOX was encapsulated, the release rate kinetics of DOX
were measured under heating or GSH environments. For the release of DOX
from MCNs/Ln/D NPs in a GSH environment, similar to the measurements of
GA release from MCNs/Ln/G NPs, deionized water, 0.5 mM and 1 mM GSH
solution were used to substitute PBS. While for the release of DOX from
MCNs/Ln/D nanocomposites under heating, 56 °C warm water was used to
imitate the irradiation of a 980 nm laser, with other conditions
similar to the measurement of GA release from MCNs/Ln/G NPs. The 1.5 mL
centrifuge tube was gentle shaken in 56 °C warm water, with time
intervals controlled at 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 min.
Meanwhile, the DOX release performance under the irradiation of a
980 nm laser was investigated. The MCNs/Ln/GD NPs were dispersed in
1 mL of PBS solution before being placed onto two slides and covered
with a cover glass. One slide was exposed to a 280 mW cm^−2 980 nm
laser for 10 min, while the other slide, serving as the control group,
was left untreated. Then, fluorescence microscope images of the two
samples were collected.
Synthesis of MCNs/Ln/GD/FR NPs
20 mg of folic acid (FA), 10 mg of 1-ethyl-3-(3-(dimethylamino) propyl)
carbodiimide (EDC), and 10 mg of N-hydroxysuccinimide (NHS) were added
to a 10 mL PBS buffer solution and stirred for 4 h. Then, the
MCNs/Ln/GD NPs were added, and the resulting mixture was stirred for
another 12 h. The precipitate, which was separated by centrifugation at
10000 rpm for 10 min, was further dispersed in 10 mL of PBS, and 1 mg
of R8 peptide was added. Following another 12 h of stirring, the final
MCNs/Ln/GD/FR NPs were obtained after being separated by centrifugation
at 10000 rpm for 10 min.
In vitro cell evaluation of nanocomposites
The Global Bioresource Center provided the human retinal microvascular
endothelial cells (hRMEC), human retinal epithelial cells (ARPE-19),
human embryonic kidney epithelial cells (HEK 293 T), mouse fibroblast
cells L929, and OCM-1 cell lines (a M14 derivative) were provided by
the BeNa Culture Collection. All cell lines have been authenticated by
the STR method and routinely screened for mycoplasma routinely.
Cytotoxicity evaluation
In vitro cytotoxicity assays were conducted as follows: 5 × 10^3/well
ARPE-19 cells were seeded into 96-well plate and incubated at 37 °C for
24 h for cell adherence. Different materials, including MCNs, MCNs/Ln,
MCNs/Ln/G, MCNs/Ln/GD, and MCNs/Ln/GD/FR, were added to the
corresponding wells at various concentrations. For each group, more
than six samples were prepared and measured. The cell incubation time
was set to 24 h.
Cell viability was measured using the Cell Counting Kit-8 (CCK-8,
CK-04, Dojindo), and the experiment was repeated three times.
The same procedure was s applied to hREMC, 293 T, and L929 cells as
that used for ARPE-19 cells.
The relative cell viability was determined using the following
equation:
[MATH: Cell
viability(%)<
mo>=(OD570,sample−OD570,blank)/(OD570,control−OD570,blank)×100% :MATH]
Photothermal effects on OCM-1 cells
Before assessing the photothermal therapeutic impact of the synthesized
nanocomposites on OCM-1 cells, the photothermal effect of a 980 nm
laser on the cell was first evaluated to eliminate the laser’s inherent
photothermal effect. 5 × 10^3/well cells were seeded into 96-well plate
and cultured at 37 °C for 24 h for cell adherence. Then, a 980 nm laser
with different power densities was used to irradiate the cells for
10 min from a distance of 3 cm (the distance from the optical fiber
head to the middle of the medium). After another 24 h incubation, the
cellular survival rate after irradiation was detected using CCK-8, and
the experiment was repeated three times.
Based on the above photothermal effect of the pure 980 nm laser on
OCM-1 cells, a laser power density of 280 mW cm^2 was selected for
subsequent experiments.
Meanwhile, a comprehensive evaluation of the photothermal effect
induced by a 280 mW cm^-2 laser on ARPE-19 cells was conducted.
Initially, 5 × 10^3/well cells were seeded into 96-well plate and
cultured at 37 °C for 24 h to ensure cell adherence. Then, a 980 nm
laser with a power density of 280 mW cm^−2 was employed to irradiate
the cells for 10 min from a fixed distance of 3 cm. After another 24 h
incubation, the survival rate of the irradiated ARPE-19 cells was
assayed using CCK-8, and this procedure was replicated three times.
Furthermore, the photothermal effects of various nanocomposites (MCNs,
MCNs/Ln, MCNs/Ln/G, MCNs/Ln/GD, MCNs/Ln/GD/FR) on OCM-1 cells were
further evaluated using a similar methodology. After cell adherence,
the medium was substituted with fresh medium containing the
nanocomposites at concentrations of 50, 100, and 200 μg mL^−1. After an
additional 6-hour incubation, the cells were irradiated with a 980 nm
laser at 280 mW cm^−2 for 10 min from a distance of 3 cm. Subsequently,
after a 24 h incubation, the cellular survival rate was determined
using CCK-8, and this process was repeated three times for each
nanocomposite.
Calcein-AM/PI double staining
To study the photothermal therapeutic effect of different
nanocomposites on OCM-1 cells under the irradiation of a 980 nm laser,
Calcein-AM/PI double staining was employed to ascertain the number of
live and dead cells. Green fluorescence can be detected in living
cells, while red fluorescence is observed in dead cells under a
fluorescence microscope. After being co-cultured with various
nanocomposites (MCNs, MCNs/Ln, MCNs/Ln/G, MCNs/Ln/GD, MCNs/Ln/GD/FR) at
a concentration of 200 μg mL^−1 for 24 h, the cells were washed with
PBS and replaced with fresh culture medium. Then, the cells were
irradiated with a 980 nm laser at 280 mW cm^−2 for 10 min from a
distance of 3 cm. Following irradiation, the cells were co-cultured for
an additional 24 h. A group that received no treatment (neither
nanocomposites nor laser irradiation) served as the control group,
termed “Medium”. Calcein-AM/PI solution was added and incubated for
15 min, and the distribution of living and dead cells was observed and
photographed using a DMI8 inverted fluorescence microscope (n = 3).
Cell apoptosis assessment
OCM-1 cells were planted in 96-well plates and incubated for 24 h at
37 °C. After that, the medium was replaced with new media containing
MCNs, MCNs/Ln, MCNs/Ln/G, MCNs/Ln/GD, and MCNs/Ln/GD/FR (200 μg mL^−1).
After 24 h incubation, the cells were washed with PBS and replaced with
fresh culture medium. The cells were then exposed to a 980 nm laser
(280 mW cm^−2, 10 min). A group that received laser irradiation
treatment served as the control group, termed “Medium + Laser”). After
another 24-h incubation, cells were collected by trypsin digestion,
washed with PBS, and resuspended in 300 μL 1X binding buffer. Then,
4 μL Annexin V-FITC and 1.5 μL PI were added in sequence. After being
kept in the dark for another 15 min, all samples were analyzed using an
Accuri C6 flow cytometer to evaluate cell viability.
Western blotting analysis
The OCM-1 cells were harvested and lysed on ice using RIPA Lysis Buffer
(Beyotime, Shanghai, China). Protein concentration was determined using
a BCA Protein Assay kit (Beyotime). Subsequently, 40 μg of protein
samples were loaded onto each lane and separated using 10% SDS-PAGE.
The proteins were then transferred to polyvinylidene fluoride (PVDF)
membranes. To reduce non-specific binding, the membranes were blocked
with 5% skimmed milk for 2 h at room temperature. After blocking, the
membranes were incubated overnight at 4 °C with primary antibody at the
following concentrations: HSP 70 (1:2000, ABclonal) and HSP 90 (1:2000,
Proteintech) diluted in Tris-buffered saline with Tween 20 (TBST).
Then, the membranes were incubated with an HRP-conjugated secondary
antibody (1:10000, Jackson Immunoresearch, Tucker, GA) for 1.5 h at
room temperature. The blots were visualized using an enhanced
chemiluminescence (ECL) reagent, and their expression levels were
quantified using ImageJ software (Version 1.8.0, Bethesda, MD, USA).
The knockdown of HSP 70 protein on photothermal inhibitory effects of OCM-1
Initially, 2 × 10^5 OCM-1 cells were seeded into a 6-well plate and
cultured overnight. The siRNA was encapsulated in lipid nanoparticles
as previously reported^[237]41. The cells were transfected with either
siRNA1^HSPA1A or siRNA2^HSPA1A to knock down the expression of HSP70,
and their sequences were listed in Supplementary Table [238]1.
Following a 48-h incubation, the cells were lysed to facilitate the
extraction of proteins for subsequent western blot analysis. Briefly,
the proteins were separated and transferred onto polyvinylidene
difluoride membrane, followed by incubation with HSP70 antibody
(1:2000, A23457, Abclonal) and β-actin (1:5000, AC004, Abclonal)
overnight. The membranes were then stained with IRDye® secondary
antibodies (1:20000, Licor, USA) and visualized using an Odyssey® DLx
Imaging System (Licor, USA).
Subsequently, thermal damage models were established to evaluate the
impact of HSP70 on cellular heat-resisting property. A total of
8 × 10^3 cells were seeded into a 96-well plate and cultured overnight.
Following a 48 h transfection with siRNA2^HSPA1, the cells were
subjected to two distinct thermal damage models. In the first model,
cells were exposed to medium with or without 100 µg mL^−1 MCNs/Ln-NPs
for 6 h. Subsequently, they were irradiated with a 980 nm laser at a
power density of 280 mW cm^−² for 10 min, with the optical fiber head
positioned 3 cm away from the center of the culture medium. In the
second model, the cells were subject to a heat shock treatment at 45 °C
in an incubator for 15 mins. Cell viability was evaluated 24 h
post-thermal damage utilizing the CCK-8 method.
In vivo anticancer evaluation of nanocomposites
To assess the synergistic therapeutic efficacy of MCNs/Ln/GD/FR
nanocomposites in tumor treatment, two kinds of melanoma models were
established: subcutaneous tumors and in situ ocular tumors.
During the model induction and treatment periods, the mice were housed
in individually ventilated cages (IVC). The environmental conditions
within the housing facility were maintained with the temperature
controlled within the range of 21–23 °C and the relative humidity kept
at a range of 40–60%. A standard photoperiod of 12 h of light followed
by 12 h of darkness was regulated. Throughout the entire experimental
period, the animals were provided with unrestricted access to food and
water.
Anticancer evaluation in subcutaneous melanoma model
Subcutaneous OCM-1 melanoma tumors were induced by injecting 2 × 10^7
OCM-1 cells suspended in 0.1 ml serum-free RMPI-1640 media into the
back region of each mouse. Upon reaching a tumor volume of
200–400 mm^3, mice with comparable tumor sizes were randomly divided
into 5 groups (n = 5 per group): PBS, Laser, DOX, MCNs/Ln/GD, and
MCNs/Ln/GD/FR. Subsequently, 200 μL PBS, DOX (10 μM), MCNs/Ln/GD
(2 mg kg^−1), or MCNs/Ln/GD/FR (2 mg kg^−1) were administered via
caudal vein injection. Tumors in the Laser, MCNs/Ln/GD, and
MCNs/Ln/GD/FR groups were irradiated with a 280 mW cm^−2 980 nm laser
for 10 min at a 3 cm distance, 4 h post-injection. Thermal imaging
documented tumor site temperatures at regular intervals. Tumor volumes
and body weights were measured every 2–3 days. The tumor volumes were
measured using a caliper, and estimated using the following equation:
[MATH: Vt=0.5×a×b×b
:MATH]
(where a and b represent the macro and minor axis of tumor). Relative
tumor volumes were determined by:
[MATH: ΔV=(
Vt−V0
)/V0×100% :MATH]
(V[0] is the initial tumor volume at treatment onset).
On day 21, mice were sacrificed, and organs (heart, liver, spleen,
lung, kidney) and tumors were collected. Thermal imaging and
histopathological examinations were conducted on these tissues.
Anticancer evaluation in in situ melanoma model
Under systemic anesthesia, topical proparacaine (Alcon, USA), and
tropicamide phenylephrine (Santen Pharmaceutical, Japan), an in situ
melanoma model was established by a skilled ophthalmologist under a
surgical microscope. A small tunnel was created 1 mm posterior to the
limbus using an insulin syringe (BD, USA). Subsequently, a total of
4 × 10^4 OCM-1-Luc cells (2 μL volume) were injected subretinally into
the left eyes of nude mice using a Hamilton injector equipped with a
33 G flat head needle and a 10 μL syringe. Tumor growth was monitored
via bioluminescence imaging on days 7 and 14 post-injection, utilizing
an in vivo optical imaging system (PerKinElmer, Spectrum CT) after
intraperitoneal administration of 200 μL 15 mg mL^−1 D-Luciferin Sodium
Salt (Solarbio). On days 8 and 15 after injection of OCM-1-Luc cells,
100 μL PBS (IV), 100 μL DOX (10 μM, IV), 100 μL MCNs/Ln/GD (2 mg mL^−1,
IV), 100 μL MCNs/Ln/GD/FR (2 mg mL^−1, IV) and 100 μL MCNs/Ln/GD/FR
(2 mg mL^−1, IVT) were injected into mice through tail intravenous
injection (IV) or intravitreal injection (IVT). 4 h post-IV injection,
tumors were irradiated with a 280 mW cm^−2 980 nm laser for 7 min, with
temperature variations of the eye recorded using an infrared thermal
imager (Hikmicro, DS-2TP74-H25PCKJ/W). The same procedure was executed
for the MCNs/Ln/GD/FR (IVT) group after 24 h injection. On day 21,
eyeballs were collected, weighed, and processed for paraffin
histopathological analysis.
Quantitative Analyses of Y Concentration using ICP
Tumor tissues of the mice in different groups and the major organs
(heart, liver, spleen, lung, kidney) for the mice treated with
MCNs/Ln/GD/FR NPs were collected, homogenized, lyophilized, and
analyzed for Y^3+ uptake content by ICP analysis to assess
biodistribution.
Histopathological analysis
Following a 10% formalin fixation, the tumor or organs were dehydrated
in graded ethanol, embedded in paraffin, and sectioned into 4–5 μm.
Following a series of dehydration sections and other preparations,
hematoxylin was applied dropwise to fill the section for 10 min before
being rinsed with PBS. Then, eosin was used for staining the
extracellular matrix for 8 min. The samples were then examined under a
light microscope (Leica DM750).
Hemolysis Assay
To evaluate the hemocompatibility of MCNs/Ln/GD/FR NPs, rabbit blood
was collected and centrifugated at 2000 × g min^−1 for 10 min at 4 °C
to isolate the red blood cells, which further washed and diluted with
PBS solution by 10 times. Then, 0.1 mL of the diluted red blood cells
were added to 0.4 mL of PBS solution (negative control group),
deionized water (positive control group), and MCNs/Ln/GD/FR NPs
dispersion with different concentrations (sample group). The samples
were incubated at 37 °C for 4 h, after which they were centrifuged at
300 × gmin^-1 for 10 min at 4 °C. Finally, the absorbance of the
supernatant was measured using an Agilent Synergy H1 microplate reader
(n = 5 per group).
The hemolysis percentage was calculated using the following formula:
[MATH: Hemolysis(%)=(Asample−Anegative
control)/(
Apositive
control−Anegative
control). :MATH]
Biochemistry tests
Blood samples collected days 15 and 30 were centrifuged at 3000 rpm for
15 min to obtain serum, which was analyzed using a Mindray BC-2800vet
system to assess liver and kidney biochemistry function, providing
insights into systemic physiological responses (n = 4 per group).
TUNEL staining and HSP fluorescent staining
The terminal deoxynucleotidyl transferase-mediated dUTP nick-end
labeling (TUNEL) apoptosis assay was conducted to identify apoptotic
cells within the tumor and adjacent normal tissues, utilizing a
One-step TUNEL FITC Apoptosis Detection Kit (APExBIO, USA), following
the manufacturer’s instructions (n = 3 per group). Meanwhile, an
assessment of HSP was conducted in both untreated and treated tumors.
Specifically, the expression levels of HSP 70 (1:200, ABclonal) and HSP
90 (1:200, Proteintech) were assessed through fluorescent staining and
visualized using a fluorescent microscope (APX 100, Olympus) (n = 3 per
group).
Single-cell RNA-seq data processing
Sample collection
Initially, mice bearing subcutaneous tumors were administered 100 μL of
MCNs/Ln/GD/FR (2 mg mL^−1) via caudal vein injection. 6 h
post-injection, the mice were subjected a 980 nm laser irradiation for
10 min at a power density of 280 mW cm^−2, with the laser positioned
3 cm from the tumor (measured from the optical fiber head to the
tumor). After 24 h of photothermal therapy, tumor tissues were
carefully collected and stored in a specialized solution. an untreated
group served as the control for comparative analysis. Subsequent sample
processing and data collection were conducted by CapitalBio Technology.
Single-cell RNA sequencing
Employing the single-cell 3’ Library and Gel Bead Kit V3.1 (10x
Genomics, 1000121) along with the Chromium Single Cell G Chip Kit (10x
Genomics, 1000120), we processed a cell suspension containing 300–600
live cells per microliter, as verified by Count Star. This suspension
was loaded onto the Chromium single-cell controller (10x Genomics) to
generate single-cell gel beads in emulsion, following the
manufacturer’s protocol. Briefly, individual cells were suspended in
PBS with 0.04% BSA, and approximately 6000 cells were introduced into
each channel, aiming to recover around 3000 target cells.
Cell capture and cDNA synthesis
Using the same kits mentioned above, we processed a cell suspension
containing 300–600 living cells per microliter, as determined by Count
Star. The suspension was loaded onto the Chromium single-cell
controller (10x Genomics) to generate single-cell gel beads in
emulsion, adhering to the manufacturer’s protocol. Single cells were
suspended in PBS with 0.04% BSA, and roughly 6000 cells were introduced
into each channel to recover approximately 3000 target cells. Captured
cells were lysed to release RNA, which was subsequently barcoded
through reverse transcription within distinct GEMs. Reverse
transcription was performed using an S1000TM Touch Thermal Cycler (Bio
Rad) with a temperature profile of 53 °C for 45 min, ramping up to
85 °C for 5 min, and maintained at 4 °C. Following this, cDNA was
synthesized, amplified, and its quality was evaluated using an Agilent
4200 system (analysis conducted by CapitalBio Technology, Beijing).
Single cell RNA-Seq library preparation
Single-cell RNA-seq libraries were constructed using the Single Cell 3’
Library and Gel Bead Kit V3.1, following the manufacturer’s guidelines.
These libraries were sequenced using an Illumina Novaseq6000 sequencer,
achieving a sequencing depth of at least 100,000 reads per cell with a
pair-end 150 bp (PE150) sequencing strategy. This was executed by
CapitalBi Technology in Beijing.
Data preprocessing
Cellranger pipeline
The Cell Ranger software was downloaded from the 10x Genomics website
([239]https://support.10xgenomics.Com/single-cell-gene-expression/softw
are /down loads/latest). Using the cellranger count module, processes
such as alignment, filtering, barcode counting, and UMI counting were
carried out to produce a feature-barcode matrix and determine clusters.
For dimensionality reduction, PCA (Principal Component Analysis) was
applied, and the first ten principle components were utilized to
generate clusters through both the K-means algorithm and a graph-based
algorithm.
Seurat pipeline
An alternative clustering method employed was Seurat 3.0, an R package.
Cells containing fewer than 200 genes, genes ranked in the top 1%, or
mitochondrial gene ratio exceeding 25% were considered abnormal and
filtered out. Dimensionality reduction was achieved using PCA, while
visualization was facilitated by TSNE and UMAP.
Enrichment analysis
GO enrichment, KEGG enrichment, Reactome enrichment, and Disease
enrichment (human only) of cluster markers were conducted using KOBAS
software, employing the Benjamini-Hochberg multiple testing adjustment.
The top 20 marker genes of each cluster were used for this analysis.
The results were visualized using an R package.
PPI (protein-protein interaction)
Protein-protein interaction data was retrieved from the STRING
database, selecting interactions with a combine_score ≥ 400. For each
cluster, the top 20 marker genes’ interactions were specifically
extracted from the database, and the interactions were visualized using
Cytoscape software.
Transcription factor prediction
Using the TFBS Tools and JASPAR database, transcription factors were
predicted within 2000 bp upstream and 500 bp downstream region of the
transcription start site (TSS) for the top 20 marker genes in each
cluster. Subsequently, the gene and transcription factor (TF) network
for each cluster were visualized with Cytoscape software.
GSEA assay (Gene set enrichment analysis)
GSEA was conducted using GSEA software version 2.2.2.4, relying on
predefined gene sets sourced from the Molecular Signatures Database
(MSigDB v6.2). All genes identified across all cells within a sample
were utilized for the analysis. Gene expression data was derived by
calculating the mean UMI count of genes within a specific cluster
compared to the remaining cluster. The range for selecting gene sets
from the collection was set between 0 and 500 genes.
Single-cell trajectories analysis
Single-cell trajectories were constructed using the Monocle (R
package), which incorporates the concept of pseudotime. Genes were
filtered based on the following criteria: they must be expressed in
more than 10 cells, have an average expression value greater than 0.1,
and exhibit a Q-value less than 0.01 across various analysis. In
addition, Cell Cycle Phase (human and mouse only) was assigned using
Seurat 3.0 with consistent parameters.
WGCNA
The weighted correlation network analysis (WGCNA) was conducted using
the WGCNA R software package. Based on the aforementioned clustering
result, each cluster was further segmented into sub-clusters, and the
average gene expression within each sub-cluster was computed. The
parameters utilized were the defaults provided by the software.
Cell type annotation
Cell type annotation was conducted using the singleR package
([240]https://bioconductor.org/packages/devel/bioc/html/SingleR.html).
This package performs unbiased cell type recognition from single-cell
RNA sequencing data by referencing transcriptomic datasets of pure cell
types to independently infer the cell of origin for each single cell.
For human data, Blueprint_Encode or HPCA was utilized, while for mouse
data, ImmGenor Mouse. RNAseq was employed.
Sequence alignment and data filtering
The “Cellranger” v7.1.0 count workflow with default settings was used
to process the single-cell sequencing data for subsequent analysis,
facilitating data quantification, genome alignment against the GRCh38
version (sourced from 10x Genomics), and preliminary data filtration.
Then, the “DoubletFinder” v2.0.3 workflow was employed to process the
filtrated data^[241]42. Data from different groups underwent
normalization separately. The canonical correlation analysis (CCA)
algorithm was also applied to minimize batch discrepancies and unify
the data sets. This harmonized dataset served as the foundation for
more intricate analyses^[242]43.
Cluster identification and annotation
Principal component analysis was conducted using the Seurat v4.3.0
package, and 9 distinct clusters with the “FindClusters” function were
successfully identified. Subsequently, the “FindAllmarker” function was
deployed to detect gene expression signatures for each cluster. These
distinctively expressed genes were used for cluster annotations.
Finally, the t-Distributed Stochastic Neighbor Embedding (tSNE)
algorithm was used for data dimensionality reduction.
Differential expression and pathway enrichment analysis
Differential gene expression between PBS and MCNs/Ln/GD/FR groups can
be discerned using “edgeR” workflow by adopting the pseudobulk
strategy. Genes with an adj.p-value > 0.05 and |log2FC | > 0.5 were
considered differentially expressed. These genes were then subjected to
Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG)
pathway enrichment analyses for functional annotation^[243]44. Hallmark
pathway gene sets from MSigDB and the “AUCell” v1.20.2 package were
used to grade each cell based on the Hallmark pathways.
Trajectory inference and transcription factor activity analysis
By using the monocle v2.26.0 workflow, the differentiation trajectory
of tumor cells was mapped out, inspecting gene expression trends over
pseudotime. Transcription factor activity was evaluated using
“pySCENIC”, with the top 5 most active transcription factors visualized
for each cluster^[244]45.
Survival and drug sensitivity analysis
Through the “FindAllmarker” function, feature genes for each cluster
(with criteria set at logFC > 0.8 & adj.p-value < 0.01) were
identified. RNA-seq and pertinent clinical data for UVM from the TCGA
database were retrieved. Based on the feature genes, CIBERSORT
algorithm was employed to assess the cluster expression abundance for
each TCGA-UVM patient^[245]46. The “survminer” v0.4.9 package was
utilized to determine the optimal cutoff point, segmenting the TCGA-UVM
data into high and low expression groups. The survival difference
between the two groups was analyzed using the Kaplan-Meier survival
analysis, and differences in clinical data between the groups were also
evaluated. The “pRRophetic” v0.5.0 package was used to analyze the
differential drug sensitivity between high and low expression groups,
based on the Cancer Genome Project (CGP) 2016 data (Clinical drug
response can be predicted using baseline gene expression levels and in
vitro drug sensitivity in cell lines).
Other characterizations
A Smart lab 9 powder Diffractometer equipped with Cu K radiation
(λ = 1.54 Å) was employed to analyze the phase structure and
crystallographic properties of the synthesized nanocomposites. The
morphology and microstructure of the samples were examined using a
field emission scanning electron microscope (HITACHI S-7400, Japan) and
a high-resolution transmission electron microscopy (JEM2100F, Japan)
operated at an accelerating voltage of 200 kV. The surface functional
groups of the nanocomposites were investigated via Fourier-transform
infrared (FT-IR) spectroscopy, with spectra recorded in the range of
400 to 4000 cm^− using a Thermal Fisher Nicolet 6700 spectrometer (USA)
and the KBr pellet technique. Brunauer-Emmett-Teller (BET) measurements
were conducted on a surface area and porosimetry analyzer
(QUANTACHROME, USA) to determine the specific surface area and pore
size distribution of the materials. The emission spectra of the
nanocomposites were recorded using a Hitachi F-7000 Fluorescence
Spectrophotometer (Japan), while the UV−vis−NIR spectra were obtained
with a Shimadzu UV-3600i Plus spectrophotometer (Japan). In addition,
the Zeta potential of the samples was measured using Malvern ZETASIZER
equipment (Nano-ZS90, Britain), providing insights into the surface
charge and stability of the nanocomposites in dispersion.
Statistical analysis
All data are presented as the mean ± standard deviation (means ± SD).
The significance between groups was determined using unpair t test,
one-way analysis of variance (ANOVA), or two-way ANOVA in GraphPad
Prism 8.0 software (ns: no significance; *P < 0.05, **P < 0.01,
***P < 0.001, ****P < 0.0001). Detailed information regarding sample
sizes, comparison methods, and other relevant parameters is provided in
the legends accompanying each figure.
Reporting summary
Further information on research design is available in the [246]Nature
Portfolio Reporting Summary linked to this article.
Supplementary information
[247]Supplementary Information^ (62.8MB, docx)
[248]Reporting Summary^ (2.2MB, pdf)
[249]Transparent Peer Review file^ (3.7MB, pdf)
Source data
[250]Source Data^ (1.5MB, xlsx)
Acknowledgements