Abstract
Pyroptosis, an immunogenic programmed cell death, could efficiently
activate tumor immunogenicity and reprogram immunosuppressive
microenvironment for boosting cancer immunotherapy. However, the
overexpression of SLC7A11 promotes glutathione biosynthesis for
maintaining redox balance and countering pyroptosis. Herein, we develop
intermetallics modified with glucose oxidase (GOx) and soybean
phospholipid (SP) as pyroptosis promoters (Pd[2]Sn@GOx-SP), that not
only induce pyroptosis by cascade biocatalysis for remodeling tumor
microenvironment and facilitating tumor cell immunogenicity, but also
trigger disulfidptosis mediated by cystine accumulation to further
promote tumor pyroptosis in female mice. Experiments and density
functional theory calculations show that Pd[2]Sn nanorods with an
intermediate size exhibit stronger photothermal and enzyme catalytic
activity compared with the other three morphologies investigated. The
peroxidase-mimic and oxidase-mimic activities of Pd[2]Sn cause potent
reactive oxygen species (ROS) storms for triggering pyroptosis, which
could be self-reinforced by photothermal effect, hydrogen peroxide
supply accompanied by glycometabolism, and oxygen production from
catalase-mimic activity of Pd[2]Sn. Moreover, the increase of
NADP^+/NADPH ratio induced by glucose starvation could pose excessive
cystine accumulation and inhibit glutathione synthesis, which could
cause disulfidptosis and further augment ROS-mediated pyroptosis,
respectively. This two-pronged treatment strategy could represent an
alternative therapeutic approach to expand anti-tumor immunotherapy.
Subject terms: Nanoparticles, Cancer therapy, Cell death, Tumour
immunology
__________________________________________________________________
Excessive production of reactive oxygen species (ROS) in cancer cells
can induce immunogenic cell death (ICD) and promote anti-tumor immune
responses. Here the authors report that intermetallics modified with
glucose oxidase and soybean phospholipid as nano-inducers promote
cancer cell pyroptosis and disulfidptosis, resulting in ICD and
anti-tumor immunity in preclinical models.
Introduction
Cancer is one of the leading causes of death worldwide and exhibits an
increasing shift from elderly to middle-aged individuals, in which
colorectal cancer is predominant among adults younger than 50
years^[42]1–[43]3. Owing to the metabolic reprogramming and genetic
mutation, cancer cells usually possess elevated oxidative stress
characteristics compared with those of nonmalignant cells^[44]4–[45]6.
To pose the excessive production of reactive oxygen species (ROS) is
detrimental to cancer cells, which has been confirmed as an efficacious
cancer therapy. While cancer cells tend to maintain sufficient
glutathione (GSH) levels to neutralize ROS for achieving cell survival
and proliferation^[46]7. Therefore, blocking the synthesis of GSH is
expected to achieve maximum cancer cell death. Cysteine is a crucial
amino acid that affords the rate-limiting precursor for the
biosynthesis of GSH^[47]8,[48]9. In general, the overexpressed solute
carrier family 7 member 11 (SLC7A11) in most cancer cells could import
extracellular cystine (an oxidized cysteine dimer), which could be
reduced into cysteine in the cytoplasm with the assistance of
glucose-derived nicotinamide adenine phosphate (NADPH). The cysteine
could serve as a precursor for the subsequent synthesis of GSH to
achieve antioxidant defense^[49]8,[50]10. Thus, the inhibition of NADPH
supply could trigger the accumulation of cystine, thereby leading to
the reduction of GSH^[51]10. Meanwhile, the overloading of cystine
could endow actin cytoskeleton proteins with abundant disulfides and
disulfide bonds, resulting in disulfidptosis caused by disulfide
stress^[52]6,[53]11.
Besides the suppression of GSH, promoting massive ROS production also
plays a critical role in intensive tumor treatments. The slightly acid
and overexpressed hydrogen peroxide (H[2]O[2]) of tumor
microenvironment (TME) create excellent conditions for the production
of ROS catalyzed by nanocatalysts^[54]12. Intermetallics composed of
two or more metallic elements display expanded and superior catalytic
activity owing to the tuned electronic states and the synergistic
effect between the two metals^[55]13,[56]14. Significantly, the
positions of all atoms in intermetallics are assigned with specific
sites, leading to the defined stoichiometry and crystal
structure^[57]15. Compared with common alloys with occupy random sites,
the regular structure endows the intermetallics with homogeneity of the
active sites, which could ensure the abundant generation of ROS^[58]16.
In addition to the enhancement of catalytic activity, the stability of
intermetallics could also be improved, which makes them perform better
in various physiological environments. This enhancement could be
attributed to the mixed bonding, which leads to a more negative free
energy of formation than random alloys^[59]17. Besides the ordered
structure, the shape of intermetallics has an influence on the crystal
plane and active sites, which could also determine the catalyst
performance^[60]18. Thus, developing well-controlled intermetallics
with tuned size and shape is attractive in improving the catalytic
capacity and optimizing the formation of ROS for effective cancer
therapeutics.
Furthermore, previous research has validated the decisive role of ROS
in inducing immunogenic programmed cell death, which could efficiently
arouse acute inflammatory response and trigger robust anti-tumor immune
activity^[61]19–[62]21. Characterized by the activation of Caspase-1
and cleavage of gasdermin D (GSDMD), pyroptosis is an inflammatory cell
death. The N-terminal domain of GSDMD (N-GSDMD) could perforate the
plasma membrane and induce cell membrane rupture, releasing the
bountiful pro-inflammatory cytokine and tumor antigens, which could
provoke precision cancer immunotherapy^[63]22–[64]24. Delightedly, a
low level of pyroptosis (<15%) in tumor cells could achieve an
efficacious anti-tumor immunity^[65]25. Consequently, the ROS-induced
pyroptosis could not only kill cancer cells by cell membrane
perforation, but also facilitate the initiation and infiltration of T
cells owing to the release of cytoplasmic contents, which could remodel
the tumor immunosuppressive microenvironment and transform cold tumor
(immune-indolent noninflamed tumor phenotype) into hot tumors
(inflammatory phenotype)^[66]24,[67]26. Therefore, exploring the
effective strategies to amplify oxidative stress-triggered pyroptosis
while synchronously magnifying immunity responses in the TME offers a
distinctive direction for cancer therapy.
In this work, we report the intermetallics modified with glucose
oxidase (GOx) and soybean phospholipid (SP) as nano-inducers
(Pd[2]Sn@GOx-SP) to elicit potent anti-tumor immune responses by
pyroptosis and disulfidptosis (Fig. [68]1). To begin with, the Pd[2]Sn
intermetallics with different morphologies are developed and their
performance are contrasted both in experiments and density functional
theory (DFT) calculations. Among them, the Pd[2]Sn nanorods (NRs) with
the largest specific surface area exhibit the optimal photothermal and
enzyme catalytic activities, including peroxidase (POD)-mimic and
catalase (CAT)-mimic catalytic activity. In addition, the modification
of GOx on the Pd[2]Sn intermetallics could serve as a triple role: (1)
degrade glucose to afford H[2]O[2] for the cascade catalysis reaction,
triggering a strong ROS storm; (2) reduce GSH synthesis to defeat the
antioxidant defense mechanism of cells and aggravate oxidative stress;
(3) inhibit NADPH supply by regulating the glycometabolism to induce
the accumulation of cystine, resulting in disulfidptosis. Moreover, the
CAT-mimic catalytic activity of Pd[2]Sn intermetallics could
effectively restore the O[2] levels and further enhance the breakdown
of glucose, thereby achieving a self-promoted process. This potent and
persistent cellular oxidative pressure could effectively activate and
amplify ROS-mediated pyroptosis. In summary, the integration of Pd[2]Sn
intermetallics and GOx could arouse self-reinforced ROS storm and
glucose consumption, which could cause Caspase-1-dependent pyroptosis
and cystine-mediated disulfidptosis, leading to the reprogramming of
immune microenvironment. Meanwhile, the change of immunosuppressive
microenvironment, including the ameliorative infiltration of T cells
(CD4^+ and CD8^+) and M1-like phenotype repolarization of macrophages,
could achieve efficacious immunotherapy by activating immune response,
counteracting tumor recurrence and metastasis.
Fig. 1. Schematic illustration of nanocomposites synthesis and synergistic
therapy process.
[69]Fig. 1
[70]Open in a new tab
a Schematic diagram of the construction of Pd[2]Sn@GOx-SP. b
Anti-cancer mechanism of synergistic immunity activation induced by
Pd[2]Sn@GOx-SP by pyroptosis and disulfidptosis.
Results
Construction and characterization of Pd[2]Sn@GOx-SP nanocomposites
The intermetallic compounds of Pd[2]Sn nanoparticles with different
morphologies were synthesized through one step by the co-reduction of
palladium(II) acetylacetonate [Pd(acac)[2]] and tin(II) acetate
[Sn(OAc)[2]], as shown in Fig. [71]1a. Methylamine hydrochloride (MAHC)
played an important role in controlling the morphology of Pd[2]Sn
intermetallics, including nanodots (NDs, without MAHC) and NRs in
different sizes (with different amounts of MAHC). The representative
transmission electron microscopy (TEM) micrographs and elemental
mapping of the Pd[2]Sn intermetallics were shown in Fig. [72]2a, b and
Supplementary Fig. [73]2a–c, respectively. The Pd[2]Sn intermetallics
exhibited the uniform controllable morphology of NRs and NDs, as well
as the homogeneous distribution of corresponding elements, confirming
the successful construction of intermetallics with different
morphologies. High-resolution TEM (HRTEM) of the Pd[2]Sn intermetallics
revealed the interfacial lattice spacings of 0.195 nm (NRs) and
0.236 nm (NDs), which could be identified as (3 2 0) and (0 2 1)
crystal planes of the orthorhombic phase, respectively (Fig. [74]2c, d
and Supplementary Fig. [75]2d, e). In addition, the coexistence of Pd
and Sn elements (with a ratio of 2:1) could also be observed from the
energy-dispersive spectroscopy (EDS) spectrum (Fig. [76]2e), suggesting
the successful synthesis of Pd[2]Sn intermetallics. Powder X-ray
diffraction (PXRD) patterns of Pd[2]Sn intermetallics were presented in
Fig. [77]2f and Supplementary Fig. [78]2, the diffraction peaks showed
good crystallinity with the lattice parameter of a = 8.11 Å,
b = 5.662 Å, c = 4.234 Å, which matched well with the pure orthorhombic
Pd[2]Sn phase (PDF 00-026-1297). The Pd[2]Sn with two inequivalent Pd
sites is cotunnite structured and crystallizes in the orthorhombic Pnma
space group (Fig. [79]2g).
Fig. 2. Structural and compositional characterization.
[80]Fig. 2
[81]Open in a new tab
TEM images at different magnifications and corresponding elemental
mapping of a Pd[2]Sn NRs and b Pd[2]Sn NDs. High-resolution TEM images
and lattice spacing measurement of c Pd[2]Sn NRs and d Pd[2]Sn NDs. One
representative data was shown from six independently repeated
experiments. e EDS spectrum of Pd[2]Sn NRs. f PXRD patterns, g crystal
structure, h XPS survey spectra, and i, j high-resolution XPS spectra
of Pd 3d and Sn 3d for Pd[2]Sn NRs and Pd[2]Sn NDs. Source data are
provided as a Source Data file (arb. units arbitrary units).
Additionally, X-ray photoelectron spectroscopy (XPS) was performed to
analyze the composition and valence state information of Pd[2]Sn
intermetallics (NDs and NRs). The as-prepared Pd[2]Sn intermetallics
were mainly composed of Pd, Sn, and O elements (Fig. [82]2h). The Pd 3d
region of Pd[2]Sn intermetallics in the high-resolution XPS spectrum
could be assigned to the spin-splitting orbits of 3d[3/2] and 3d[5/2],
and the binding energies at 341.0 and 335.9 eV were attributed to
Pd^2+, while the peaks at 340.1 and 335.0 eV were assigned to Pd^0
(Fig. [83]2i). This result confirmed the coexistence of Pd^0 (77%) and
Pd^2+ (23%) in Pd[2]Sn NRs and Pd^0 (74%) and Pd^2+ (26%) in Pd[2]Sn
NDs. Meanwhile, the high-resolution XPS spectrum for Sn 3d of Pd[2]Sn
intermetallics exhibited a distinct decomposition, verifying the
existence of tin with two distinct valence states (Fig. [84]2j). The
characteristic peaks at 486.6 eV (3d[5/2]) and 497.4 eV (3d[3/2]) were
attributed to Sn^4+, while the peaks at 486.0 eV (3d[5/2]) and 496.7 eV
(3d[3/2]) were associated with Sn^2+, and the component at 484.5 eV
(3d[5/2]) and 494.8 eV (3d[3/2]) corresponded to metallic Sn.
Furthermore, we also investigated the change of elements in the valence
state after prolonged oxidation (Supplementary Fig. [85]3a, b). The
majority of the surface Sn was in the oxidized state and the surface Pd
was mainly in the metallic state, indicating the surface Sn atoms are
more easily oxidized than Pd atoms. These results validated that the
Pd[2]Sn intermetallics could generate more ROS through the redox
reaction^[86]27. The corresponding elemental mapping, EDS spectrum, and
Fourier transform infrared spectrum (FT-IR) showed that the surface of
Pd[2]Sn intermetallics interacted with Tri-n-octylphosphine (TOP) due
to the coordination potence of phosphine ligands (Supplementary
Figs. [87]4 and [88]5a). To endow the Pd[2]Sn intermetallics with high
biocompatibility, the SP was further functionalized on the surface of
Pd[2]Sn intermetallics to obtain Pd[2]Sn-SP nanocomposites^[89]28.
Significantly, the surface of Pd[2]Sn intermetallics was further
modified with GOx, which could facilitate the cascade catalytic
reaction to generate more ROS by the enhancement of H[2]O[2] catalyzed
by GOx. The successful functionalization of SP and modification of GOx
were confirmed by the FT-IR spectrum (Supplementary Fig. [90]5b).
Thermogravimetric analysis (TGA) confirmed that the proportion of SP on
Pd[2]Sn-SP nanocomposites was ~20.81% (Supplementary Fig. [91]6). In
addition, the zeta potentials showed obvious changes from 0.13 to
−29.93 mV after the functionalization of SP, and the zeta potential of
Pd[2]Sn@GOx-SP was further changed to −23.0 mV (Supplementary
Fig. [92]7a). Furthermore, the hydrodynamic size of Pd[2]Sn@GOx-SP
nanocomposites in phosphate buffer saline (PBS) was ~90 nm according to
the analysis of dynamic light scattering, and the average size of
Pd[2]Sn@GOx-SP at various solvents was also evaluated (Supplementary
Fig. [93]7b). The Pd[2]Sn@GOx-SP exhibited a uniform hydrated particle
size over 7 days, demonstrating the ideal stability of Pd[2]Sn@GOx-SP
in physiological conditions, which is a prerequisite for better
subsequent application.
Photothermal and multi-enzymatic catalytic activity evaluation
As a noble metal, palladium exhibits the peculiar localized surface
plasmon resonance (LSPR) effect and various enzyme catalytic activity,
which has made a profound study (Fig. [94]3a)^[95]29–[96]32. To begin
with, the optical properties of Pd[2]Sn-SP nanocomposites (NDs and NRs)
were explored by using UV–vis–NIR absorbance spectrum (Fig. [97]3b and
Supplementary Fig. [98]8), the adsorption intensity of Pd[2]Sn-SP NDs
and NRs increased linearly with an elevation of concentration.
Especially, the Pd[2]Sn-SP NRs exhibited stronger absorption
(5.33 L g^−1 cm^−1) than Pd[2]Sn-SP NDs with the extinction coefficient
was 2.30 L g^−1 cm^−1. Subsequently, the Pd[2]Sn-SP NRs were chosen for
evaluating the photothermal performance. A notable laser power
density-dependent photothermal effect was observed in Supplementary
Fig. [99]9, and the temperature of Pd[2]Sn-SP NRs irradiated with laser
(0.8 W cm^−2) can considerably increase by 26 °C, which is sufficient
for killing cancer cells. It is noteworthy that the temperature change
of Pd[2]Sn-SP NRs also showed concentration-dependent (Fig. [100]3c),
while only a slight increase of pure water with laser excitation
(0.8 W cm^−2), which was also verified from the infrared thermal images
(Supplementary Fig. [101]10), indicated the excellent efficiency of
Pd[2]Sn-SP NRs for photothermal conversion. Besides photothermal
conversion capability, photothermal stability is also critical for the
photothermal conversion process, which was further assessed by four
on/off laser power cycles. No notable changes in temperature even after
four cycles of heating and natural cooling, confirming high
photothermal stability (Fig. [102]3d). Then, the photothermal
conversion efficiency (η) of Pd[2]Sn-SP NRs was calculated to be 41.2%
on the basis of cooling phase (Fig. [103]3e). The high light absorption
and excellent photothermal conversion ability of Pd[2]Sn-SP NRs make
them effective contrast agents for photoacoustic imaging (PAI) due to
the LSPR effect^[104]33,[105]34. Consequently, the in vitro PAI ability
of Pd[2]Sn-SP NRs was explored by detecting the PA signals of
Pd[2]Sn-SP NRs with different concentrations. As displayed in
Supplementary Fig. [106]11, the PA signals showed a positive
correlation with the concentration of Pd[2]Sn-SP NRs, which
demonstrated the ideal PAI ability for biomedical and clinical
applications.
Fig. 3. Evaluation of photothermal and multi-enzymatic catalytic activity of
Pd[2]Sn-SP nanocomposites.
[107]Fig. 3
[108]Open in a new tab
a Schematic illustration for the detection of photothermal and
multi-enzymatic catalytic activity. b The fitting curve of the mass
extinction coefficient of Pd[2]Sn-SP NRs and Pd[2]Sn-SP NDs. c The
concentration-dependent photothermal heating curves, d photothermal
heating and natural cooling cycles of Pd[2]Sn-SP NRs. e Photothermal
heating and cooling curve and the relationship between −Lnθ and the
time. f, g Diagram and ESR spectra of ·OH with different treatments.
UV–vis absorption spectra of h OPD and i TMB catalyzed by Pd[2]Sn-SP
NRs under various conditions. j Michaelis–Menten kinetic analysis and k
Lineweaver–Burk plotting for POD-mimic activity of Pd[2]Sn-SP
nanocomposites. l Comparison of the kinetic parameters for POD-mimic
activity of Pd[2]Sn-SP with different morphologies. m Schematic
illustration of the mutual promotion of multi-enzyme activity. n
OXD-mimic activity of Pd[2]Sn-SP NRs at various times. Time-dependent
O[2] production of Pd[2]Sn-SP NRs o with different concentrations and p
with or without laser irradiation. q Glucose consumption of
Pd[2]Sn@GOx-SP. r The change of pH value at different times. Data are
expressed as mean ± S.D. (n = 3) independent experiments in (h–k, q,
r). Source data are provided as a Source Data file (arb. units
arbitrary units).
Taking the excellent catalytic activity for effective anti-cancer
treatment into consideration, the multienzyme-mimic activities as well
as the cascade catalytic reaction of Pd[2]Sn-SP nanocomposites were
explored^[109]32,[110]35. To begin with, the generation of ^1O[2] and
·OH was quantitatively evaluated by the electron spin resonance (ESR)
measurement with 2,2,6,6-tetramethylpiperidine (TEMP) and
5,5-dimethyl-1-pyrroline N-oxide (DMPO) as trapping agents
(Fig. [111]3f). An obvious signal peak of ^1O[2] could be seen when the
Pd[2]Sn-SP nanocomposites were irradiated with a laser, and the
stronger signals occurred after the addition of H[2]O[2] (Supplementary
Fig. [112]12a). The distinct quadruple resonance peaks of ·OH could be
observed in Fig. [113]3g, displaying a signal intensity ratio of
1:2:2:1, which was further enhanced upon laser irradiation, indicating
the laser-induced hyperthermia-enhanced production of ·OH. Moreover,
9,10-diphenylanthracene (DPA) was further used to confirm the
generation of ^1O[2], the absorption intensity of DPA weakened with
time after being exposed to the laser, illustrating the generation of
^1O[2] by Pd[2]Sn-SP nanocomposites (Supplementary Fig. [114]12b).
Simultaneously, the POD-mimic activity of Pd[2]Sn-SP NRs was also
validated by the color reaction of o-phenylenediamine (OPD) and
3,3′,5,5′-tetramethyl-benzidine (TMB), which could be oxidized by ·OH
to generate orange (oxOPD) or blue color (oxTMB) with the
characteristic peak at ~420 or ~652 nm, respectively. The Pd[2]Sn-SP
NRs induced remarkable characteristic absorption of oxOPD or oxTMB
after adding H[2]O[2], while the control or laser alone group showed no
changes (Fig. [115]3h, i). Moreover, the absorption intensity was
enhanced when the Pd[2]Sn-SP NRs irradiated with a laser, which was
similar to the results of ESR measurement.
Furthermore, the difference in catalytic activity of Pd[2]Sn-SP
nanocomposites with diverse morphology (NDs, NRs, SNRs, and LNRs) was
investigated by the steady-state kinetic analyses. Four types of
Pd[2]Sn-SP nanocomposites were mixed with H[2]O[2] at various
concentrations (1, 2, 4, 8, and 16 mM), and the absorbance change of
oxTMB was recorded with time (Supplementary Fig. [116]13), all of which
could align well with the typical Michaelis–Menten kinetics
(Fig. [117]3j). Based on the Lineweaver–Burk plots (Fig. [118]3k), the
maximum reaction velocity (V[max]) and Michaelis–Menten constants
(K[m]) were acquired and summarized in Fig. [119]3l. Generally, a
higher V[max] and lower K[m] lead to better catalytic performance.
Compared with the Pd[2]Sn-SP NDs, the Pd[2]Sn-SP NRs exhibited higher
catalytic efficiency. Moreover, the catalytic efficiency was also
relevant to the length of Pd[2]Sn-SP NRs, neither too long nor too
short is beneficial for the POD-mimic activity. Additionally,
Pd[2]Sn-SP nanocomposites with diverse morphology were employed to
further evaluate the catalytic activity by altering the concentrations
of Pd[2]Sn-SP nanocomposites and detecting the absorption intensity of
oxTMB (Supplementary Fig. [120]14a–d). The specific activity values of
Pd[2]Sn-SP nanocomposites were calculated (Supplementary
Fig. [121]14e–h) and summarized in Supplementary Fig. [122]14i.
Similarly, the Pd[2]Sn-SP NRs showed the supreme POD-mimic catalytic
activity than the others. The effective generation of ·OH in the
presence of Pd[2]Sn-SP NRs was also confirmed by performing a methylene
blue (MB) experiment. As displayed in Supplementary Fig. [123]15, the
absorbance of MB decreased with the prolonged reaction time,
demonstrating the excellent generation ability of ·OH. In addition to
·OH, the ·O[2]^− generation based on the oxidase (OXD)-mimic catalytic
activity of Pd[2]Sn-SP NRs was also validated using TMB (Fig. [124]3n).
The absorbance intensity at 652 nm increased with a prolonged time,
showing the obvious OXD-mimic activity of Pd[2]Sn-SP NRs.
Considering the hypoxic TME restricts the OXD-mimic activity, the
CAT-mimic catalytic activity of Pd[2]Sn-SP NRs by catalyzing H[2]O[2]
into O[2] was detected. The Pd[2]Sn-SP nanocomposites with diverse
morphology were added with various concentrations of H[2]O[2] and the
oxygen content was recorded for comparison. As shown in Supplementary
Fig. [125]16, all the Pd[2]Sn-SP nanocomposites could produce more O[2]
with the increase of H[2]O[2] concentration, and Pd[2]Sn-SP NRs
displayed the strongest CAT-mimic catalytic activity, which was
consistent with the results of POD-mimic catalytic activity. In
addition, the generation of O[2] also increased with the concentration
of Pd[2]Sn-SP NRs (Fig. [126]3o), and the addition of laser further
promoted the CAT-mimic catalytic activity (Fig. [127]3p). Although the
TME is featured by the overexpression of H[2]O[2], the endogenous
H[2]O[2] content is unable to meet the subsequent biocatalytic
reactions^[128]36. Thus, the Pd[2]Sn-SP NRs were modified with GOx to
supply H[2]O[2] by catalyzing glucose decomposition. The catalytic
ability of Pd[2]Sn@GOx-SP for accelerating glucose consumption was
verified, the glucose consumption increased with the prolonged reaction
time, while the negligible change was observed in the control group
(Fig. [129]3q). Meanwhile, the degradation of glucose is accompanied by
the production of gluconic acid, leading to the reduction of pH value
(Fig. [130]3r). Therefore, the consumption of endogenous glucose could
not only provide H[2]O[2] for enzyme-catalyzed reaction, but also
reduce pH value to enhance the enzyme catalytic efficiency.
Furthermore, the catalytic activity of GOx was evaluated in PBS with
different pH values to investigate whether the GOx would malfunction in
a more acidic environment. The results showed that there was no obvious
change in the catalytic performance of GOx in a more acidic
environment, demonstrating that the GOx was able to effectively degrade
glucose and afford H[2]O[2] for the cascade catalytic reaction
(Supplementary Fig. [131]17). Moreover, the O[2] generated by CAT-mimic
catalytic activity could also improve the OXD-mimic activity as well as
the glucose consumption by GOx, achieving mutual promotion and common
improvement (Fig. [132]3m). We also evaluated the influence of pH value
on the etching and leaking of Sn and Pd ions by using inductively
coupled plasma-optical emission spectrometry (ICP-OES). As shown in
Supplementary Fig. [133]18, the Sn and Pd ions could be released from
Pd[2]Sn@GOx-SP, while the amount of ion release was negligible even
during 3 days of incubation, which was only 2.73 and 1.10 μg mL^−1,
respectively. The ability to reshape the TME and promote the extensive
generation of ROS endows Pd[2]Sn@GOx-SP with excellent potentiality for
tumor therapy.
DFT calculations and enzymatic catalysis mechanism
DFT calculations were further performed to unveil the catalytic
mechanism and the crystal-facet-dependent multienzyme-mimicking
activities of Pd[2]Sn. First, the bulk structure of Pd[2]Sn was
optimized and displayed in Fig. [134]4a. According to the calculated
density of states (DOS) in Fig. [135]4b, it can be identified that the
predominant contribution of the DOS near the Fermi level originated
from Pd-5d orbitals, indicating that electrons near the Fermi level are
mainly provided by Pd atoms and of great importance for the catalytic
activity of Pd[2]Sn. The moderate atomic radius and d orbitals with
abundant electrons of the Pd made Pd[2]Sn easy to interact with the
reactant molecules and promote the electron transfer process. Thus, the
excellent catalytic activities of Pd[2]Sn could be achieved by
activating the reactants and subsequently forming suitable
intermediates with lower energy barriers. Based on the facet
discrepancy of Pd[2]Sn NDs and NRs, three surface models with exposed
(001), (010), and (011) crystal facets were built to clarify the enzyme
catalytic mechanisms (Fig. [136]4c). Before evaluating the catalytic
activity of different Pd[2]Sn facets, microscopic electron structures
and Bader charge analysis were conducted to observe the charge
distribution. As shown in Fig. [137]4d, the surface electrons were
transferred from Sn to Pd atoms, which altered the redistribution of
electrons and led to the negative valence state of Pd. It can be
observed clearly that the charges of the Pd sites on (001), (101), and
(011) surfaces were calculated to be −0.297, −0.418/−0.484, and
−0.261/−0.339/−0.348 e, respectively. The significant change of surface
electron density could have an impact on the catalytic performance of
Pd[2]Sn catalysts, as discussed in detail later.
Fig. 4. DFT calculations and enzyme catalytic mechanism.
[138]Fig. 4
[139]Open in a new tab
a Polyhedral view of the optimized structure of Pd[2]Sn. The red and
purple balls represent the Pd and Sn atoms. b Density of states of
Pd[2]Sn. c Optimized structural models of Pd[2]Sn with different
crystal facets and d the corresponding electron density mapping images
(the effective charges were obtained by the Bader analysis program).
e–g Calculated potential profile averaged on the plane perpendicular to
the c-axis of Pd[2]Sn with different crystal facets. h The reaction
pathways and i, j the corresponding energy profiles for the POD- and
CAT-mimic activities on Pd sites of Pd[2]Sn with different crystal
facets. Source data are provided as a Source Data file (arb. units
arbitrary units).
Furthermore, to evaluate the electron transfer behavior between the
catalyst surface and the intermediates, the work functions of different
facets were also calculated^[140]37. In response to the surface charge
redistribution, the data in Fig. [141]4e–g identified that the work
functions for (001), (010), and (011) facets are 6.902, 6.341, and
7.241, indicating the electron transfer from the catalyst surface to
the intermediates followed by the sequence of (010) > (001) > (011). In
addition, the adsorption process of H[2]O and H[2]O[2] on the Pd and Sn
sites of different Pd[2]Sn facets was analyzed and summarized in
Supplementary Fig. [142]19a. The adsorption energy of H[2]O[2] on Pd
and Sn sites of three different crystal facets were much lower than
that of H[2]O, which was beneficial for its preferential adsorption
thus the subsequent catalysis (Supplementary Fig. [143]19b).
Significantly, the corresponding adsorption energies of H[2]O[2] on Pd
site in (001), (010), (011) facets were calculated to be −0.28, −0.17,
and −0.13 eV, respectively. The lower adsorption energy of H[2]O[2] on
the Pd site with respect to the Sn site in the (001) facet ensured the
energetically favorable binding of H[2]O[2] on the catalyst surface.
Furthermore, to reveal the relevant catalytic mechanism, three
optimized surfaces of Pd[2]Sn were adopted to comprehensively
investigate the H[2]O[2] catalytic process. The critical intermediates
formed in the successive reaction steps of CAT-mimic and POD-mimic
activities were displayed in Fig. [144]4h and Supplementary
Fig. [145]19c. All the H[2]O[2] catalytic reaction intermediates were
optimized for the (001), (010), and (011) crystal facets. Regarding the
POD-mimic activity, the calculated energy profiles are illustrated in
Fig. [146]4i and Supplementary Fig. [147]19d. The first step involves
the capture and surface adsorption of H[2]O[2] on the Pd[2]Sn surface.
Then, the H[2]O[2] was decomposed into two OH*, which was
thermodynamically favorable in the three crystal facets. In the third
step, the adsorbed OH* was released and formed ·OH, which required to
overcome an energy barrier and thus became the rate-determining step.
Pd sites on the (001) facet possessed the lowest free energy (2.59 eV
per OH*) for the desorption of bridged OH* intermediate compared with
(010) and (011) crystal facets (2.73 and 2.63 eV per OH*,
respectively), demonstrating its better POD-mimic activity.
Simultaneously, the energy profiles for the CAT-mimic activity were
shown in Fig. [148]4j and Supplementary Fig. [149]19e. Five successive
catalytic processes on the (001), (010), and (011) crystal facets have
been calculated. The adsorption, activation, and dissociation of
H[2]O[2] were all exothermic in all crystal facets, indicating the
feasibility of the reaction process. Conversely, the deprotonation of
OH* to form O* and the subsequent formation of bridge structure O[2]*
were apparent endothermic processes. For the Pd site on Pd[2]Sn (001)
surface, the energies required to form O* and O[2]* intermediates are
calculated to be 1.06 eV (deprotonation of one H) and 1.69 eV,
respectively. In contrast, relevant values for Pd[2]Sn (010)/(011)
surfaces are 1.22/1.41 and 0.88/1.89 eV. Moreover, the deprotonation of
OH* on the Pd[2]Sn (001) surface is much easier than that on the
Pd[2]Sn (010) surface, while the energy required to form O[2]* on the
Pd[2]Sn (010) surface is the lowest. Thus, the formation of O[2]*
intermediates became the rate-determining step in the catalytic
processes of CAT-mimic activity. It should be noted that the
deprotonation of OH* to produce hydrogen products (H[2]) required an
electron transfer between catalyst surface and intermediates.
Therefore, the highest work function of (011) facet would result in the
highest interfacial charge transfer resistance during this fundamental
step, even the (011) facet exhibited a much lower energy barrier. In
addition, as all the energy barriers needed to overcome for Pd sites on
(001) and (010) surface are significantly lower than the one (1.89 eV)
to form O[2]* on (011) surface, demonstrating the CAT-mimic activity of
the Pd[2]Sn (011) surface is the worst.
According to Supplementary Fig. [150]19f, it can be further confirmed
that the energy barrier for the formation of O[2]* at Pd sites is well
correlated with the work function of different facets, implying that
the charge redistribution caused by the change of the surface
orientation will alter the electronic structure of the catalyst
surface, and thus changing the reaction mechanism. In addition, the
results in Supplementary Fig. [151]19g also indicated that the Sn sites
in the three facets do not exhibit obvious superiority over the
formation of O* and O[2]* intermediates, but they are favorable for the
desorption of O[2] with the energies ranging from −0.19 to 0.10 eV.
According to the analysis on the adsorption, reaction, and desorption
capabilities of different sites on different facets, it can be
concluded that Pd and Sn sites play different but synergistic roles in
determining the overall activity during whole CAT-mimic reaction. In
brief, the POD-mimic activity is easier to occur in (001) crystal facet
than in (011) crystal facet. The (001) and (010) crystal facets play
different roles in CAT-mimic activity, in which (001) facet facilitates
the deprotonation of OH* and (010) facet facilitates the formation of
O[2]*. Both (001) and (010) crystal facets exhibit better CAT-mimic
activity than (011) crystal facets. Moreover, Sn sites facilitate the
desorption of O[2], collaboratively promoting the CAT-mimic activity.
Overall, the CAT-mimic and POD-mimic activities of Pd[2]Sn were
theoretically verified by the exploration of the reaction pathways and
energy profiles, further demonstrating that the Pd[2]Sn NRs with
exposed (001) facet possessed higher catalytic activity than Pd[2]Sn
NDs with exposed (011) facet.
In vitro synergistic pyroptosis and disulfidptosis
These experiments have confirmed that the Pd[2]Sn@GOx-SP possessed the
property for altering the TME and generating abundant ROS. We further
evaluated the in vitro anti-tumor therapeutic effect of Pd[2]Sn@GOx-SP
(Fig. [152]5a). To begin with, we measured the cytotoxicity of
Pd[2]Sn@GOx-SP on CT26 cells by methylthiazolyldiphenyl-tetrazolium
bromide (MTT) assay. Various concentrations of Pd[2]Sn-SP and
Pd[2]Sn@GOx-SP were co-incubated with CT26 cells in the dark to detect
the dark-toxicity, and the photo-toxicity of Pd[2]Sn@GOx-SP was
detected by laser irradiation followed by the incubation. As can be
seen in Fig. [153]5b, both Pd[2]Sn-SP and Pd[2]Sn@GOx-SP were cytotoxic
to CT26 cells compared with drug-free incubation. Significantly, the
Pd[2]Sn@GOx-SP exhibited remarkable photo-toxicity on tumor cells under
laser irradiation. Moreover, the cytotoxicity of Pd[2]Sn@GOx-SP
nanocomposites on normal cells and multiple cancer cells was further
performed to evaluate the tumor growth inhibitory effect. Our findings
revealed that the nanocomposites showed varying degrees of inhibitory
effects on cancer cells (Supplementary Fig. [154]20a). While for normal
cells (L929 and 3T3 cells), both Pd[2]Sn-SP and Pd[2]Sn@GOx-SP were
almost non-cytotoxic after co-incubation of 12 and 24 h, showing good
biocompatibility even at higher concentrations (Supplementary
Fig. [155]20a–d). Given the existence of TOP on the surface of Pd[2]Sn,
we further co-cultured the L929 cells with TOP at different
concentrations to evaluate the biotoxicity of P originated from TOP.
The result showed no significant toxic side effects on L929 cells,
demonstrating the negligible biotoxicity of P in TOP (Supplementary
Fig. [156]20e). Meanwhile, the influence of Sn and Pd ions leaking on
L929 cells was also investigated through MTT assay by co-culturing the
released Sn and Pd ions with L929 cells. The results showed that there
was no obvious biotoxicity on L929 cells (Supplementary Fig. [157]20f),
indicating the high biosafety of Pd[2]Sn@GOx-SP. Thus, the
nanocomposites exhibited negligible cytotoxicity on normal cells
without laser irradiation, demonstrating high selective toxicity. To
explore the cellular internalization performance of Pd[2]Sn@GOx-SP, the
cancer cells incubated with fluorescein isothiocyanate (FITC) modified
Pd[2]Sn@GOx-SP were analyzed by confocal laser scanning microscopy
(CLSM) and flow cytometry, respectively (Fig. [158]5c, d and
Supplementary Fig. [159]21a, b). The cells treated with FITC-labeled
Pd[2]Sn@GOx-SP showed higher fluorescence over a prolonged incubation
time, exhibiting the excellent cellular uptake efficiency of
Pd[2]Sn@GOx-SP.
Fig. 5. In vitro synergistic pyroptosis and disulfidptosis.
[160]Fig. 5
[161]Open in a new tab
a Schematic illustration of the synergistic pyroptosis and
disulfidptosis mediated by Pd[2]Sn@GOx-SP. b Cytotoxicity assays of
CT26 cells with different treatments. c Representative CLSM images of
the colocalization of FITC-labeled Pd[2]Sn@GOx-SP with the lysosome. d
Flow cytometry profile of FITC-labeled Pd[2]Sn@GOx-SP in CT26 cells.
CLSM images of intracellular e O[2] generation and f ROS level after
various treatments (G1, control; G2, laser; G3, Pd[2]Sn-SP; G4,
Pd[2]Sn@GOx-SP; G5, Pd[2]Sn@GOx-SP + laser). g JC-1 staining of CT26
cells with diverse treatments. h Calcein-AM/PI staining of CT26
multicellular spheroids. Fluorescence image of i HMGB1 migration and j
CRT exposure of CT26 cells with diverse treatments. k Western blot
analysis of Caspase-1, C-Caspase-1, GSDMD, N-GSDMD, and NLRP3 in CT26
cells after diverse treatments. l Bio-TEM images of CT26 cells with
diverse treatments. m Western blot analysis of FLNA, TLN1, and MYH9 in
CT26 cells after diverse treatments. n Measurement of the NADP^+/NADPH
ratio. o The quantitative data of matured DCs after diverse treatments
by in vitro stimulation maturation experiment. Data are expressed as
mean ± S.D. (n = 3) independent samples in (b, n, o). Statistical
significance is assessed by a two-way ANOVA with Tukey’s multiple
comparisons test. Source data are provided as a Source Data file (one
representative data was shown from three independently repeated
experiments).
Research has shown that endocytosis is the main internalization process
of nanocomposites, and the uptake rate of rod-like nanocomposites was
higher than that of spherical nanocomposites^[162]38–[163]40. The
endosomes/lysosomes usually play an indispensable role in transporting
and releasing nanocomposites during subsequent intracellular
processes^[164]41. Consequently, the endosomal escape process of
Pd[2]Sn@GOx-SP was evaluated by staining lysosomes with Lyso-Tracker.
The results revealed that the FITC-labeled Pd[2]Sn@GOx-SP was
colocalized with lysosomes within 1 h after internalization (Pearson’s
correlation coefficient of 0.79). With the extension of incubation
time, the Pd[2]Sn@GOx-SP gradually escaped from the lysosomes, as
evidenced by the Pearson’s correlation coefficient was decreased to
0.52 and 0.39 after 2 and 4 h of internalization, respectively
(Fig. [165]5c), indicating that the Pd[2]Sn@GOx-SP could effectively
escape from endolysosomes. Considering stereoscopic structure of the
tumor, we further investigated the penetration performance of
Pd[2]Sn@GOx-SP on CT26 multicellular spheroids to simulate
three-dimensional tumors. Similarly, the fluorescence intensity of
multicellular spheroids incubated with FITC-labeled Pd[2]Sn@GOx-SP
enhanced with the time prolongs (Supplementary Fig. [166]21c),
indicating the ideal permeability for the multicellular spheroids.
Thus, the ideal penetration ability could promote the accumulation of
Pd[2]Sn@GOx-SP in solid tumors, which could gain more access to tumor
cells and improve the therapeutic efficacy.
Subsequently, the change of intracellular substances and their effect
on cell death was further evaluated on CT26 cells. The oxygen content
in tumor cells plays a significant role in tumor therapy^[167]42. On
the one hand, the generated O[2] could accelerate the consumption of
glucose in the presence of GOx to trigger a series of cellular damage.
On the other hand, the abundant O[2] could be converted into ^1O[2] and
·O[2]^− catalyzed by Pd[2]Sn@GOx-SP, which enhances the oxidative
damage of tumor cells. [Ru(dpp)[3]]^2+Cl[2] as an oxygen detection
probe was used to evaluate the oxygen production of Pd[2]Sn@GOx-SP in
CT26 cells. As displayed in Fig. [168]5e, different from the control
and laser groups, the red fluorescence intensity of the cells decreased
after incubation with Pd[2]Sn-SP, suggesting that the continuous
production of O[2] based on the CAT-mimic catalytic activity of the
Pd[2]Sn-SP. It was also comparatively found that the cells in the
Pd[2]Sn@GOx-SP group exhibited a brighter red fluorescence than that of
Pd[2]Sn-SP, which could be attributed to the O[2] consumption in the
decomposition of glucose by GOx. Whilst, a large amount of H[2]O[2]
generation is accompanied by the process of glucose decomposition,
which could further augment the content of ROS. It has been confirmed
that the abnormal increase of intracellular ROS levels could induce
damage to crucial cellular biomolecules, leading to cell death
(including pyroptosis) through the Caspase-1/GSDMD
pathway^[169]43,[170]44. Various strategies to generate ROS have been
confirmed in vitro and showed excellent anti-tumor therapeutic
potential^[171]45. Thus, the intracellular ROS was determined by CLSM
and flow cytometry using a 2′,7′-dichlorofluorescein diacetate
(DCFH-DA) probe (Fig. [172]5f and Supplementary Fig. [173]21d, e). The
results showed the fluorescence intensity of Pd[2]Sn@GOx-SP group was
remarkable, demonstrating the exceptional ability of ROS generation.
Especially, the highest fluorescence intensity was observed in the
Pd[2]Sn@GOx-SP + laser group due to the enhancement of ROS generation
by the hyperthermia effect.
High levels of ROS in tumor cells could achieve irreversible damage to
DNA to induce pyroptosis^[174]46. Considering the abnormal elevated ROS
levels could lead to mitochondrial dysfunction, the change of
mitochondrial membrane potential after various treatments was evaluated
by performing JC-1 staining. As shown in Supplementary Fig. [175]21f,
JC-1 forms aggregates within the matrix of mitochondria under normal
polarization conditions, exhibiting vibrant red fluorescence. On the
contrary, JC-1 could only exist as a monomer in depolarized
mitochondria, showing green fluorescence^[176]46. As depicted in
Fig. [177]5g, intense red fluorescence was shown in the cells with the
treatment of control and only laser, indicating normal mitochondria. In
particular, the green fluorescence intensity of cells treated with
Pd[2]Sn@GOx-SP increased significantly, and the highest green–red
fluorescence proportion was exhibited in the Pd[2]Sn@GOx-SP + laser
group, demonstrating severe mitochondrial damage. Meanwhile, highly
toxic ROS could analogously destroy the lysosomes^[178]47, which are
also crucial subcellular organelle. The integrity of lysosomes was
analyzed by acridine orange staining assay, in which red fluorescence
could change into green fluorescence when the lysosome ruptured. The
results showed that the lysosomal membrane integrity was completely
ruptured after the treatment of Pd[2]Sn@GOx-SP + laser, and a certain
rupture could also be observed in the Pd[2]Sn-SP and Pd[2]Sn@GOx-SP
groups (Supplementary Fig. [179]22a).
The destruction of Pd[2]Sn@GOx-SP on CT26 cells was also confirmed by
observing the morphological changes of actin filaments (F-actin). The
cytoskeletal disruption could be seen in the cells after being treated
with Pd[2]Sn@GOx-SP, while the F-actin of the cells in the group of
control or only laser were highly elongated and well-organized
(Supplementary Fig. [180]22b), demonstrating the cell injury induced by
Pd[2]Sn@GOx-SP. To gain further insights into the living and dead
cells, Calcein-AM/PI was used to stain the cells in various groups, and
the fluorescence intensity was examined (Supplementary Fig. [181]22c).
The dead cells in the control or only laser group were negligible,
while dead cells predominated in the group of Pd[2]Sn@GOx-SP + laser,
indicating the strong killing efficiency of Pd[2]Sn@GOx-SP on tumor
cells. The living and dead cells proportion in different treatment
groups was also obtained by flow cytometric assay. The dead cells
increased after treatment with Pd[2]Sn-SP or Pd[2]Sn@GOx-SP, and the
cell mortality rate reached its maximum after incubation with
Pd[2]Sn@GOx-SP + laser (Supplementary Fig. [182]22d), suggesting the
ideal treatment effect of Pd[2]Sn@GOx-SP on cancer cells. Besides that,
we also investigated the therapeutic performance of Pd[2]Sn@GOx-SP on
CT26 multicellular spheroids in vitro. Consistent with the above
results, almost no cell death was observed in control or only laser
groups, while a certain amount of dead cells existed in Pd[2]Sn@GOx-SP
group, and a significant increase of dead cells was found in
Pd[2]Sn@GOx-SP + laser group (Fig. [183]5h). Different depths of cell
death could be observed, confirming the effective ability for tumor
inhibition.
After confirming the excellent capability for killing tumor cells, we
further explored the specific mechanism of Pd[2]Sn@GOx-SP-induced cell
death in detail. Due to the massive production of ROS with various
types, we speculated that Pd[2]Sn@GOx-SP could induce
pyroptosis^[184]46,[185]48,[186]49. As an inflammatory programmed cell
death, pyroptosis could activate the anti-tumor immune responses by
releasing various inflammatory factors, which could serve as an
important means to improve immune deficiencies^[187]25. To verify the
immune activation effect induced by Pd[2]Sn@GOx-SP, the typical
biomarkers of immunogenic cell death (ICD), high mobility group 1
(HMGB1) and calreticulin (CRT) were detected by immunofluorescence
assay. As can be seen in Fig. [188]5i, strong red fluorescence
overlapped with the cell nucleus in control and only laser groups,
while that was diminished in the Pd[2]Sn@GOx-SP and
Pd[2]Sn@GOx-SP + laser groups, indicating that Pd[2]Sn@GOx-SP could
trigger pyroptosis and release HMGB1 into the extracellular milieu.
Meanwhile, more CRT expression was detected on the cellular surface in
the Pd[2]Sn@GOx-SP and Pd[2]Sn@GOx-SP + laser groups, whereas no
significant signal in the control or only laser group (Fig. [189]5j).
Moreover, the cells treated with drugs showed significant adenosine
triphosphate (ATP) release, wherein the leakage level of the
Pd[2]Sn@GOx-SP + laser group was the highest compared with other groups
(Supplementary Fig. [190]22e). All results clearly validated that
Pd[2]Sn@GOx-SP could induce large-scale ICD effects mediated by
pyroptosis.
Furthermore, western blot analysis was also initiated to assess protein
expression during the pyroptosis process (Fig. [191]5k). In the
pyroptotic pathway, NLRP3 inflammasome and Caspase-1 could be
stimulated by abundant ROS and GOx, and subsequently achieve the
cleaving of GSDMD into N-GSDMD, which could lead to the perforation of
the cellular membranes^[192]49. The high expression of NLRP3 could be
observed in the group treated with Pd[2]Sn@GOx-SP due to the
two-pronged strategy. In addition, the Pd[2]Sn@GOx-SP-related group
exhibited high expression of Cleaved Caspase-1 (C-Caspase-1) and
N-GSDMD proteins, which could underpin the drilled pores in cell
membrane, leading to the secretion of damage-associated molecular
patterns. To further confirm the GSDMD- or NLRP3-dependent pyroptosis
induced by Pd[2]Sn@GOx-SP + laser, we have knocked down the expression
of GSDMD or NLRP3 in CT26 cells, then the cytotoxicity of
Pd[2]Sn@GOx-SP + laser on the treated CT26 cells was measured. To begin
with, transient transfection was performed to establish GSDMD or NLRP3
knockdown CT26 cells, respectively. The transfection efficiency was
verified by western blot analysis (Supplementary Fig. [193]22f). And
the experimental results verified that CT26 cells with low expression
of GSDMD or NLRP3 had higher cell viability than the control group
after being treated with Pd[2]Sn@GOx-SP + laser, respectively
(Supplementary Fig. [194]22g). Thus, the pyroptosis could be achieved
by the treatment of Pd[2]Sn@GOx-SP + laser, which is dependent on the
GSDMD and NLRP3. Considering the results in immunofluorescence and
western blot analysis, we attempted to offer in-depth evidence of
pyroptosis induced by Pd[2]Sn@GOx-SP. In the most intuitive way, the
cell morphology after various treatments was recorded. As can be seen
in Supplementary Fig. [195]22h, CT26 cells with the treatment of
Pd[2]Sn@GOx-SP possessed more distinct swelling (bubbling) features
compared with those in the control group, indicating the occurrence of
pyroptosis. Meanwhile, Bio-TEM was also performed to monitor the cells
with the different treatments (Fig. [196]5l). It can be observed that
the cell membrane of the cells treated with Pd[2]Sn@GOx-SP exhibited
extensive vacuolization and incomplete cell membranes, and mitochondria
were deformed and swollen. These results provide concrete evidence for
Pd[2]Sn@GOx-SP-induced pyroptosis due to a large amount of ROS
generation.
Significantly, the tumor cells exhibited high levels of SLC7A11, which
could import cystine to promote GSH synthesis and resist
pyroptosis^[197]50. Considering that we introduced GOx to constrict
that process by consuming glucose, which could deplete the NADPH to
prevent the reduction of cystine, achieving the inhibition of GSH
synthesis. The quantitative evaluation of intracellular glucose was
conducted, the cells treated with GOx exhibited the obvious consumption
of glucose (Supplementary Fig. [198]22j). In addition, the aberrant
buildup of intracellular cystine could lead to disulfide stress, which
ultimately induces disulfidptosis (Fig. [199]5a). To begin with, we
performed non-reducing western blots to validate glucose starvation
could induce disulfide-bond formation in the actin cytoskeleton
proteins. As shown in Fig. [200]5m, the actin cytoskeleton proteins
(FLNA, TLN1, and MYH9) in CT26 cells exhibited slower migration on the
smears following treatment with Pd[2]Sn@GOx-SP. Some of these proteins
displayed exceptionally high-molecular-weight bands near the stacking
layer, whereas the control or laser-only groups did not show similar
migration patterns, indicating that the actin cytoskeleton proteins
formed multiple intermolecular disulfide bonds under the treatment of
Pd[2]Sn@GOx-SP. Subsequently, we evaluated the NADPH levels by
performing various treatments on CT26 cells. As depicted in
Fig. [201]5n, the cells handled with Pd[2]Sn@GOx-SP exhibited a
significant increase in the NADP^+/NADPH ratio (which indicates NADPH
depletion) due to glucose starvation. In addition, the intracellular
GSH levels after various treatments were also assessed. After the
treatment of Pd[2]Sn@GOx-SP, the GSH levels were obviously decreased
(Supplementary Fig. [202]22k), which could enhance the pyroptosis
process.
To further confirm the disulfidptosis induced by
Pd[2]Sn@GOx-SP + laser, the dithiothreitol (DTT) as disulfidptosis
inhibitor was added to the cells, and the cell viability was analyzed
by CCK-8 assay. The results showed that the cell viability of the cells
added with DTT was obviously higher than that in the absence of DTT
after being treated with Pd[2]Sn@GOx-SP + laser (Supplementary
Fig. [203]22l), demonstrating the disulfidptosis could be triggered by
Pd[2]Sn@GOx-SP. Meanwhile, to confirm the critical role of
SLC7A11-mediated cystine uptake for disulfide stress, the CT26 cells
were incubated with Pd[2]Sn@GOx-SP + laser after the treatment of
SLC7A11 inhibitor (Erastin). Compared with that without the treatment
of Erastin, the cell death rate of CT26 cells exhibited an obvious
decrease due to the inhibition of SLC7A11-mediated cystine uptake
(Supplementary Fig. [204]22m), indicating that the disulfide stress
originated from the accumulation of cystine mediated by SLC7A11. In
addition, the expression levels of SLC7A11 in multiple cancer cells
(4T1, HeLa, SW1990, A549, and HepG2 cells) and normal cells (L929
cells) were evaluated by western blot analysis (Supplementary
Fig. [205]22i). The results showed the higher expression level of
SLC7A11 in these cancer cells than that in normal cells, leading to
higher cell mortality in cancer cells compared to normal cells with low
SLC7A11 expression.
Besides that, the CT26 cells were treated with ferroptosis inhibitor
(Ferrostatin-1) to give insight into the role of ferroptosis in this
work. The result showed that the CT26 cells treated with
Pd[2]Sn@GOx-SP + laser also exhibited significant inhibitory effect
after the addition of Ferrostatin-1, which was similar to that without
addition of Ferrostatin-1, demonstrating the effect of ferroptosis
could be ignored (Supplementary Fig. [206]22n). Therefore, GOx-induced
disulfidptosis combined with Pd[2]Sn@GOx-SP-induced pyroptosis are the
main manner for promoting tumor cell death. Furthermore, we also
examined whether the Pd[2]Sn@GOx-SP + laser treatment could alleviate
cell migration by wound-healing assays (Supplementary Fig. [207]23a).
Indeed, cell migration was largely inhibited after treatment with
Pd[2]Sn@GOx-SP + laser for 12 and 24 h, while the control or only laser
group had minimal effect on the cell migration (Supplementary
Fig. [208]23b). Moreover, the transwell migration assay of CT26 cells
after diverse treatments was also performed to evaluate the migration
ability. Similarly, the cells treated with Pd[2]Sn@GOx-SP + laser were
greatly repressed (Supplementary Fig. [209]23c), illustrating that
Pd[2]Sn@GOx-SP had a significant inhibitory effect on cell migration.
Therefore, GOx-induced disulfidptosis combined with
Pd[2]Sn@GOx-SP-induced pyroptosis greatly promoted tumor cell death and
inhibited its migration. To confirm the role of pyroptosis on the
augmenting of immunogenicity, the maturation degree of dendritic cells
(DCs) was analyzed by flow cytometry analysis, which could improve the
antigen presentation ability and initiate the subsequent anti-tumor
cascade immunity^[210]51,[211]52. The results showed that the
Pd[2]Sn@GOx-SP combined with laser irradiation can significantly
increase the proportion of CD80^+CD86^+ cells (Fig. [212]5o and
Supplementary Fig. [213]23d), indicating that Pd[2]Sn@GOx-SP treated
tumor cells could stimulate DCs activation and facilitate anti-tumor
immune responses.
In vivo PA/CT imaging-guided synergistic anti-tumor therapy
The results of in vitro cell experiments validated that the
Pd[2]Sn@GOx-SP + laser irradiation treatment group resulted in
pyroptosis and disulfidptosis. Prior to evaluating the curative effect
of the tumor in vivo, the PA/CT imaging capability was assessed based
on the ideal light absorption and high X-ray attenuation coefficient
(Fig. [214]6a)^[215]53,[216]54. To evaluate the potential of
Pd[2]Sn@GOx-SP for PA imaging, Pd[2]Sn@GOx-SP was intravenously (i.v.)
injected into CT26-bearing mice, and then imaged at different time
intervals by a Vevo LAZR-X system. As presented in Fig. [217]6b, the PA
signals reached the strongest at 6 h post-injection, and subsequently
weakened owing to the elimination of Pd[2]Sn@GOx-SP from tumor tissues.
Meanwhile, the biodistribution of Pd[2]Sn@GOx-SP in main organs was
also detected by ex vivo PA imaging at 6 h post-injection
(Supplementary Fig. [218]24a), showing the highest accumulation of
Pd[2]Sn@GOx-SP in the liver compared with other organs. In addition,
the blood oxygen saturation (SaO[2]) in the tumor was monitored at
various times (Supplementary Fig. [219]24b). The results showed the
efficient oxygen supply ability of Pd[2]Sn@GOx-SP for glucose
consumption by GOx, which is beneficial to the in vivo cascade reaction
process.
Fig. 6. In vivo PA/CT imaging-guided synergistic anti-tumor therapy.
[220]Fig. 6
[221]Open in a new tab
a Schematic illustration of the PA, CT, and infrared thermal imaging.
Representative in vivo b PA images after being i.v. injected with
Pd[2]Sn@GOx-SP and c CT images after i.t. injected without or with
Pd[2]Sn@GOx-SP. (one representative data was shown from three
independently repeated experiments). d The corresponding line profiles
of CT values. e In vivo biodistribution of Pd[2]Sn@GOx-SP in main
organs and tumors at various time intervals. Data are expressed as
mean ± S.D. (n = 3) mice for each group. f The blood circulation curve
and g the eliminating rate curve after i.v. injected with
Pd[2]Sn@GOx-SP. Data are expressed as mean ± S.D. (n = 3) independent
animals. h Therapeutic schedule of Pd[2]Sn@GOx-SP for CT26
tumor-bearing mice. i Body weight, j relative tumor volume, k tumor
volume, l tumor weight, and m the photographs of excised tumors from
representative mice in different treatment groups (G1, control; G2,
laser; G3, Pd[2]Sn-SP; G4, Pd[2]Sn@GOx-SP; G5, Pd[2]Sn@GOx-SP + laser).
Data are expressed as mean ± S.D. (n = 5) mice for each group in (i, j,
l). n Hematological indexes and biochemical data of mice after i.v.
injection with PBS and Pd[2]Sn@GOx-SP. o Staining images of H&E, TUNEL,
Ki67, NLRP3, C-Caspase-1, GSDMD, and GZMB for tumor slices from
different groups. (one representative data was shown from three
independently repeated experiments). Statistical significance is
assessed by a two-way ANOVA with Tukey’s multiple comparisons test.
Source data are provided as a Source Data file (arb. units arbitrary
units).
Similarly, the CT imaging ability of Pd[2]Sn@GOx-SP was also
investigated both in vitro and in vivo. As presented in Supplementary
Fig. [222]24c, the CT signal intensity gradually strengthened as the
concentration of Pd[2]Sn@GOx-SP increased. Moreover, the Hounsfield
unit (HU) value was positively correlated to the concentration of
Pd[2]Sn@GOx-SP with a slope of 21.75 (Supplementary Fig. [223]24d).
Thereafter, the CT26-bearing mice were intratumorally (i.t.) and i.v.
injected with Pd[2]Sn@GOx-SP to evaluate the CT imaging capability,
respectively. Contrasted with the control group, the tumor tissues of
mice injected with Pd[2]Sn@GOx-SP presented a high CT density,
suggesting an efficient CT imaging capacity (Fig. [224]6c, d).
Furthermore, we performed CT imaging on the same tumor-bearing mouse at
different time points after being injected with Pd[2]Sn@GOx-SP. Within
6 h, the CT density in the tumor region of the mice i.v. injected with
Pd[2]Sn@GOx-SP were markedly enhanced with the injection time extension
(Supplementary Fig. [225]24e), which showed the same trend as PA
imaging, suggesting the efficient tumor accumulation of Pd[2]Sn@GOx-SP.
Besides, the real-time temperature change in the tumor site was also
monitored by a thermal imaging camera after 6 h of Pd[2]Sn@GOx-SP
injection. The temperature of the mice tumor raised obviously under the
irradiation of laser compared with that without injection of
Pd[2]Sn@GOx-SP (Supplementary Fig. [226]24f), exhibiting the superior
in vivo photothermal effect of Pd[2]Sn@GOx-SP. To sum up, the
Pd[2]Sn@GOx-SP possessed the potential for PA, CT, and infrared thermal
imaging to guide tumor synergistic therapy.
The biosafety and biocompatibility of Pd[2]Sn@GOx-SP are prerequisites
for in vivo treatment, thus the comprehensive evaluation was performed.
To begin with, we examined the hemolysis induction of Pd[2]Sn@GOx-SP in
red blood cells (Supplementary Fig. [227]25). The hemolysis assay
showed that no significant hemolysis appeared by Pd[2]Sn@GOx-SP, even
if the concentration reached 500 μg/mL. Moreover, the biodistribution
assessment of Pd[2]Sn@GOx-SP i.v. injected into CT26-bearing mice was
comprehensively analyzed by using ICP-OES. As displayed in
Fig. [228]6e, the high contents of Pd were observed in the liver and
spleen, which were consistent with the results of ex vivo PA imaging.
After 6 h of Pd[2]Sn@GOx-SP injection, the Pd levels reached a maximum
concentration (9.07% ID g^−1) in tumor regions, and maintained a
relatively high level (6.33% ID g^−1) even at 24 h, demonstrating the
superior tumor-homing efficiency of Pd[2]Sn@GOx-SP. The high
biocompatibility of SP endowed Pd[2]Sn@GOx-SP with stable blood
circulation capacity. The blood circulation half-life time of
Pd[2]Sn@GOx-SP was acquired, which was t[1/2(α)] = 0.15 h and
t[1/2(β)] = 3.88 h (Fig. [229]6f). Simultaneously, as presented in
Fig. [230]6g, the elimination rate constant of Pd[2]Sn@GOx-SP in the
first stage was obtained to be −0.5043 µg mL^−1/h, which showed a
decrease to −0.0279 µg mL^−1/h after an interval of 1.86 h. Benefiting
from the appropriate blood circulation, Pd[2]Sn@GOx-SP could achieve
abundant accumulation in the tumor to exert an anti-tumor effect.
Subsequently, the in vivo anti-cancer efficacy of Pd[2]Sn@GOx-SP on
CT26-bearing mice was further evaluated (Fig. [231]6h). Twenty-five
female BALB/c mice established with the cancer model were randomly
separated into five groups (n = 5), containing control (G1), laser
(G2), Pd[2]Sn-SP (G3), Pd[2]Sn@GOx-SP (G4), Pd[2]Sn@GOx-SP + laser
(G5). All groups were i.v. administered with PBS or nano-drug at a dose
of 10 mg kg^−1, and irradiated with laser after 6 h of injection.
During the treatment process, no significant abnormal body weight
fluctuation was displayed in the treated groups (Fig. [232]6i),
indicating high biosafety. As depicted in Fig. [233]6j, k, control or
only laser irradiation resulted in rapid tumor growth, whereas
Pd[2]Sn-SP alone moderately inhibited tumor growth. Comparatively, the
Pd[2]Sn@GOx-SP + laser group exerted a significant therapeutic effect
with a suppression rate of 83.76% owing to the synergistic effect of
pyroptosis with disulfidptosis. Moreover, the tumor weight also
exhibited distinct differences after various treatments. The mice in
the group of control or only laser possessed heavier tumors while tumor
weights of the mice treated with nano-drug were reduced to varying
degrees (Fig. [234]6l). Digital photos of the tumors dissociated from
each mouse reflected the identical results (Fig. [235]6m). After that,
the biocompatibility of the Pd[2]Sn@GOx-SP was also evaluated by
conducting the blood routine and blood biochemical analyses on the
treated mice, which could provide the basis for potential practical
applications of malignancy treatments. No significant abnormalities
were found in the liver and kidney function of the mice before and
after i.v. administrated with Pd[2]Sn@GOx-SP (Fig. [236]6n). Moreover,
there were also no noticeable changes in hematological biomarkers
contrasted with the control group, verifying the insignificant side
effects of Pd[2]Sn@GOx-SP on hematological system.
To further validate the satisfactory anti-cancer effect of
Pd[2]Sn@GOx-SP combined with laser, TdT-mediated dUTP nick-end labeling
(TUNEL) and hematoxylin–eosin (H&E) staining were proceeded on the
slices of tumor tissue collected from the mice with various treatments
(Fig. [237]6o). As expected, the highest degree of cell death was
reflected in the group of Pd[2]Sn@GOx-SP + laser than the other
treatment groups, verifying a favorable tumor therapeutic effect of
Pd[2]Sn@GOx-SP. Moreover, Ki67 staining was also performed and the
group of Pd[2]Sn@GOx-SP + laser greatly decreased the expression of
Ki67 (Fig. [238]6o), indicating the prominent inhibition effect for
tumor aggressiveness. Furthermore, the underlying therapeutic mechanism
induced by Pd[2]Sn@GOx-SP via pyroptosis was also confirmed through the
immunohistochemical investigation of NLRP3, C-Caspase-1, and GSDMD. All
of them were visibly elevated in the tumor tissues extracted from the
mice treated with Pd[2]Sn@GOx-SP + laser (Fig. [239]6o), evidencing the
occurrence of pyroptosis in tumor cells. Significantly, the main organs
of mice with diverse treatments showed no distinct inflammation or
pathological changes (Supplementary Fig. [240]26), which additionally
confirmed the remarkable histocompatibility of Pd[2]Sn@GOx-SP.
Furthermore, we constructed another tumor xenograft model using 4T1
mammary cancer cells to validate the excellent anti-tumor efficacy of
Pd[2]Sn@GOx-SP (Supplementary Fig. [241]27a). The results demonstrated
that the growth of 4T1 tumors was significantly inhibited by the
treatment of Pd[2]Sn@GOx-SP + laser (Supplementary Fig. [242]27b, c).
Meanwhile, the body weight of mice in each treatment group exhibited no
significant abnormality (Supplementary Fig. [243]27d), confirming the
biosafety of Pd[2]Sn@GOx-SP. The excellent tumor suppression effect
could also be observed based on the H&E and TUNEL staining images of
tumor tissue slices (Supplementary Fig. [244]27e). Moreover, the H&E
staining images of the major organs collected from the mice with
diverse treatments showed no distinct inflammatory or pathological
changes, which further validated the excellent biosafety of
Pd[2]Sn@GOx-SP (Supplementary Fig. [245]27f). These results highlighted
the therapeutic potential of Pd[2]Sn@GOx-SP, which could induce both
pyroptosis and disulfidptosis, making it a promising candidate for
treating various types of tumors.
In vivo immune stimulation effect
The release of massive cytoplasmic contents induced by tumor pyroptosis
has been demonstrated to trigger immune responses for achieving
anti-tumor immune activity and cancer immunotherapy
(Fig. [246]7a)^[247]25. The reconstruction of the immunosuppressive
microenvironment elicited by Pd[2]Sn@GOx-SP was investigated by flow
cytometry analysis. Similar to the in vitro experimental results, the
Pd[2]Sn@GOx-SP + laser group induced a higher proportion of DCs
maturation in spleens than that of other groups (Fig. [248]7b, c). The
mature DCs play an important role in antigen presentation, thereby
activating the proliferation of naive T cells and evoking an adaptive
immune response^[249]44,[250]55. To further confirm the
immunotherapeutic effect of Pd[2]Sn@GOx-SP, the spleen infiltrating T
cells were emphatically evaluated. As unfolded in Fig. [251]7d, e, the
proportions of CD4^+ and CD8^+ T cells in Pd[2]Sn@GOx-SP + laser group
could reach 47.0% and 37.7%, which were over 1.9 and 4.4 folds more
than those in the control group, respectively. Importantly, the IFN-γ
levels distinctly elevated in the tumors of the mice treated with
Pd[2]Sn@GOx-SP + laser, which was 6.5 folds higher than the mice in
control group (Fig. [252]7f, g), demonstrating the effective activation
of CD8^+ T cells^[253]56.
Fig. 7. In vivo immune stimulation effect of Pd[2]Sn@GOx-SP.
[254]Fig. 7
[255]Open in a new tab
a Pattern diagram for remodeling the immunosuppressive TME. b
Representative flow cytometry data and c the quantitative data of
matured DCs in spleens after receiving diverse treatments (G1, control;
G2, laser; G3, Pd[2]Sn-SP; G4, Pd[2]Sn@GOx-SP; G5,
Pd[2]Sn@GOx-SP + laser). d Representative flow cytometry data and e the
quantitative data of CD8^+ T cells in spleens after receiving diverse
treatments. f Representative flow cytometry data and g the quantitative
data of IFN-γ level in tumor tissues after diverse treatments. h The
corresponding quantitative data of CD8^+ T cells in tumor tissues after
receiving different treatments. i The corresponding quantitative data
of M1-type macrophages (CD86^+ and CD206^− cells) in tumor. j Schematic
diagram of the schedule for tumor rechallenge study. k Body weight
changes of CT26 recurrence tumor-bearing mice with diverse treatments.
l Tumor volume and m relative tumor volume of recurrence tumor in CT26
tumor-bearing mice with diverse treatments. n The corresponding
quantitative data of IFN-γ level in recurrence tumor tissues. o
Representative flow cytometry data of T[EM] (CD62L^− and CD44^+ cells)
in spleens after receiving diverse treatments. Data are expressed as
mean ± S.D. (n = 3) independent samples in (c, e, g–i, n), (n = 5) mice
for each group in (k, m). Statistical significance is assessed by a
two-way ANOVA with Tukey’s multiple comparisons test. Source data are
provided as a Source Data file.
Simultaneously, the activation of T cells in tumors was also analyzed
(Supplementary Fig. [256]28a). The counts of CD8^+ T cells in the group
of Pd[2]Sn@GOx-SP + laser increased by 4.3 folds than that in control
group (Fig. [257]7h). Moreover, the expression level of granzyme B
(GZMB) in CD8^+ T cells is also a good marker of anti-cancer immunity
activation. The expression level of GZMB in tumors was evaluated by
immunohistochemical analysis to confirm the activation of anti-cancer
immunity. As can be seen in Fig. [258]6o, the GZMB levels were
significantly upregulated in the tumors of the mice treated with
Pd[2]Sn@GOx-SP + laser compared with other groups, demonstrating the
effective activation of CD8^+ T cells to achieve anti-cancer immunity.
Tumor-associated macrophages (TAMs) are the prominent immune cells
present in the tumor stroma^[259]57,[260]58. The activated macrophages
mainly comprise M1-type and M2-type, which show pro-inflammatory and
anti-inflammatory properties, respectively. M2-type TAMs usually endow
TME with characteristics of immunosuppression, promoting the
progression of tumor^[261]59,[262]60. Consequently, the repolarization
of M2-type TAMs to M1-type TAMs could reverse the tumor
immunosuppression, which is beneficial for tumor therapeutic
effect^[263]61,[264]62. Consequently, the capability of Pd[2]Sn@GOx-SP
to alter the phenotype of TAMs was explored in mouse tumors
(Supplementary Fig. [265]28b). M2-type TAMs obviously decreased in the
Pd[2]Sn@GOx-SP + laser group than that in control group, while the
ratio of M1-type TAMs raised from 6.23% to 74.6% (Fig. [266]7i),
indicating the laser-amplified and excellent tumor immunotherapy effect
of Pd[2]Sn@GOx-SP. All these data provide convincing evidence that the
Pd[2]Sn@GOx-SP could recruit immune cells into tumors to induce
specific anti-tumor immunity.
Tumor immunotherapy could stimulate the immune system to derive a
long-term anti-tumor immunological response, which could prevent the
recurrence and metastasis of tumors^[267]63. Encouraged by the
excellent activation of immunity and treatment effect on the primary
tumor, we further investigated the efficacy of Pd[2]Sn@GOx-SP on mouse
models with recurrence by a tumor-challenging assay (Fig. [268]7j). The
body weights and rechallenge tumor volumes of all the mice with various
treatments were measured during the treatment process (Fig. [269]7k,
l). The rechallenge tumors in the Pd[2]Sn@GOx-SP + laser group were the
smallest among all the groups (Fig. [270]7m), indicating that the
immune memory induced by Pd[2]Sn@GOx-SP could effectively inhibit tumor
recurrence. Subsequently, the immune mechanism of the distal tumor
tissues in various groups was further investigated. As depicted in
Supplementary Fig. [271]29a, the ratio of CD8^+ T cells in the distant
tumors was increased to 17.3 after treatment with
Pd[2]Sn@GOx-SP + laser, which was over 5.3 folds higher than that in
control group. Simultaneously, the IFN-γ levels in distant tumors were
also evaluated (Supplementary Fig. [272]29b). Results indicated that
the secretion of IFN-γ significantly increased in the Pd[2]Sn@GOx-SP
treated group (Fig. [273]7n), which was beneficial to the subsequent
activation of anti-tumor immunity. Furthermore, depleting antibodies
against CD8 (anti-CD8α) were injected into mice to deplete CD8^+ T
cells, and the tumor growth was monitored to confirm that the treatment
involves the activation of anti-cancer immunity (Supplementary
Fig. [274]27a). After the depletion of CD8^+ T cells by the
administration of anti-CD8α, the therapeutic efficacy of
Pd[2]Sn@GOx-SP + laser decreased significantly (Supplementary
Fig. [275]27g). All these data provide convincing evidence that
Pd[2]Sn@GOx-SP could recruit immune cells to tumors, thereby inducing a
specific anti-tumor immune response.
Motivated by satisfactory immune response of Pd[2]Sn@GOx-SP, the
activation of immune memory was further evaluated by analyzing the
proportion of effector memory T cells (T[EM], CD62L^−CD44^+) and
central memory T cells (T[CM], CD62L^+CD44^+) in spleens
(Fig. [276]7o). The ratios of T[EM] in the group of
Pd[2]Sn@GOx-SP + laser increased to 85.1%, which was higher than that
in control group, indicating a potent immune memory was formed. These
results demonstrated that Pd[2]Sn@GOx-SP could effectively reshape the
immunosuppressive TME and stimulate the intense immune response,
thereby effectively expunging dormant tumor cells and hindering their
recurrence. After that, the inhibitory effects of Pd[2]Sn@GOx-SP on
lung metastasis were further investigated by establishing a pulmonary
metastasis model to confirm the immune memory effect (Supplementary
Fig. [277]30a). After all the treatments, the CT26-bearing mice were
i.v. injected with CT26 cells to imitate lung metastasis, and the lungs
of the mice were harvested after 13 days. The photographs and H&E
staining images of lung tissues showed that abundant nodules in the
lung collected from the control group, and a significant improvement
was observed in the Pd[2]Sn@GOx-SP group, but there was no obvious
metastasis in the Pd[2]Sn@GOx-SP + laser combined treatment group
(Supplementary Fig. [278]30b). In general, the Pd[2]Sn@GOx-SP with
satisfactory biosecurity could promote the maturation and infiltration
of cytotoxic T lymphocytes, which activated strong systematic immune
responses against tumor recurrence and metastasis. Therefore, the
Pd[2]Sn@GOx-SP-mediated pyroptosis and disulfidptosis could effectively
inhibit tumor growth, recurrence, and spontaneous metastases.
RNA-sequencing analysis of anti-cancer mechanism
Given the effective therapeutic effect and immune activation of
Pd[2]Sn@GOx-SP on tumors, we further explored the potential therapeutic
mechanism by analyzing the mRNA profiles through high-throughput RNA
sequencing. In total, 964 differentially expressed genes (DEGs) were
screened out between the Pd[2]Sn@GOx-SP + laser and control groups, of
which 602 (62.45%) genes were upregulated and 362 (37.55%) genes were
downregulated (Fig. [279]8a–c). Subsequently, the Kyoto Encyclopedia of
Genes and Genomes (KEGG) and Gene Ontology (GO) analysis of DEGs was
conducted to examine the biological roles (Fig. [280]8d, e). The
results suggested that the DEGs correlated with Pd[2]Sn@GOx-SP + laser
treatment were related to immunity pathways, demonstrating that
Pd[2]Sn@GOx-SP + laser could activate the immune response by triggering
programmed death of tumor cells. This discovery highlights the
importance of these genes in immune regulation, providing important
clues for us to understand their potential roles in diseases or
biological processes. In addition, the correlation between the
downregulated genes and cell adhesion was confirmed by the GO
enrichment analysis, which is critical for cell migration, further
confirming the inhibition of tumor cell migration by the
Pd[2]Sn@GOx-SP + laser treatment. Furthermore, gene set enrichment
analysis (GSEA) confirmed that the DEGs were involved in pathways
associated with pyroptosis (Fig. [281]8f). Additionally, heat maps of
DEGs associated with disulfidptosis were also generated (Fig. [282]8g).
The expression of disulfidptosis-related genes (including NDUFA11,
GYS1, OXSM, LRPPRC, NDUFS1, NCKAP1, PRDX1, RPN1, and NUBPL) was notably
upregulated in the Pd[2]Sn@GOx-SP + laser treatment group^[283]9. The
evident genetic changes demonstrated that Pd[2]Sn@GOx-SP + laser could
induce tumor cell death via pyroptosis and disulfidptosis processes.
Fig. 8. Transcriptional analysis of anti-tumor mechanism induced by
Pd[2]Sn@GOx-SP.
[284]Fig. 8
[285]Open in a new tab
a Cluster diagram of DEGs between Pd[2]Sn@GOx-SP + laser and control
groups. b Volcano plots and c numbers of the upregulated and
downregulated genes of Pd[2]Sn@GOx-SP + laser groups compared with
control groups. Statistical significance is assessed by two-sided
t-test. d KEGG pathway enrichment analysis. e The circle diagram of GO
enrichment analysis. Statistical significance is assessed by one-sided
t-test. f GSEA of pyroptosis signaling pathway. Statistical
significance is assessed by one-sided t-test. g Cluster heatmap of the
DEGs associated with disulfidptosis between Pd[2]Sn@GOx-SP + laser and
control groups. Source data are provided as a Source Data file (n = 3
mice).
Discussion
In summary, the precise pyroptosis and disulfidptosis dual-inducer of
intermetallic Pd[2]Sn modified with GOx was developed for exerting
anti-tumor immune effects. Various morphologies (including NDs and NRs
with different lengths) of Pd[2]Sn intermetallic compounds were
constructed and the properties were further contrasted. Both in vitro
experiments and DFT calculations results confirmed that the NRs with
the highest specific surface area exhibit the best catalytic
performance. The intermetallic Pd[2]Sn with ordered structure displayed
enhanced NIR light absorption and multiple enzyme catalytic activities
(CAT, POD, and OXD-mimic activity), which facilitate photothermal
conversion and ROS generation, thereby achieving GSDMD-dependent
pyroptosis. Additionally, owing to the consumption of glucose by
integrated GOx, Pd[2]Sn@GOx-SP could increase the NADP^+/NADPH ratio,
leading to the abundant accumulation of cystine in CT26 cells with high
SLC7A11 expression, achieving cystine-related disulfidptosis. Moreover,
the enhanced cascade catalytic reaction could also be realized by the
generation of H[2]O[2], photothermal effect, production of O[2], and
GSH depletion triggered by the complementary relationship of Pd[2]Sn
and GOx. Notably, Pd[2]Sn@GOx-SP-induced pyroptosis and disulfidptosis
could effectively reprogram TME by alleviating immunosuppression and
promoting T cell infiltration, thus promoting immune responses mediated
by T cells, which are conducive to the inhibition of tumor metastasis
and recurrence. This work not only provided a strategy for the
imaging-guided dual-inducer design of pyroptosis and disulfidptosis,
but also broadened the biomedical applications of intermetallic
compounds, which offers good prospects for future advancement of cancer
immunotherapy.
Methods
Animal care
Female BALB/c mice (4-week-old) were obtained from Beijing Vital River
Laboratory Animal Technology Co., Ltd. (Beijing, China)
(1100111084356). The animal experiments were conducted with the
approval of ethics by the Ethics Committee of Harbin Medical University
Cancer Hospital (No. KY2024-33). Experimental group sizes were approved
by the Regulatory Authorities for Animal Welfare after being defined to
balance statistical power, feasibility, and ethical aspects. CT26 cells
or 4T1 cells (1 × 10^6, 100 μL in PBS) were subcutaneously injected
into the right back of each BALB/c mouse when they were 5 weeks old to
establish CT26 or 4T1 tumor models, and the following experiments were
conducted when the tumor volume approached 60 mm^3. The maximal tumor
size/burden permitted by the ethics committee is 1500 mm^3, and the
maximal tumor size/burden was not exceeded in the experiments. The sex
of the animal was not specifically considered in this study. Mice of
the appropriate sex were employed according to the requirements of the
tumor model. A female mouse model was employed for CT26 colon cancer
and 4T1 breast cancer. Mice were housed in a specific-pathogen-free
condition at 26 ± 1 °C and 50 ± 5% humidity with a 12-h light–dark
cycle with unrestricted access to food and water.
Chemicals and materials
Oleylamine, MAHC, Pd(acac)[2], Sn(OAc)[2], TOP, GOx, SP, TMB, MB, DMPO,
and OPD were purchased from Sigma-Aldrich (Shanghai, China). Ethanol
and chloroform were of analytical grade and purchased from various
sources. MTT, DCFH-DA, 4′,6-diamidino-2-phenylindole (DAPI), Calcein-AM
and propidium iodide (PI), JC-1 staining kit, and ActinGreen were
purchased from Beyotime Inst. Biotech. (Haimen, China).
[Ru(dpp)[3]]Cl[2] (cat: MX4826) was purchased from MKBIO (Shanghai,
China). Anti-CRT (AF1666) and anti-HMGB1 (PH406) were purchased from
Beyotime. ATP content assay kit was obtained from Shanghai EnzymeLink
Biotechnology Co., Ltd. (Shanghai China). The glutathione assay kit and
glucose assay kit were purchased from Wanleibio Co., Ltd. Anti-Granzyme
B (ab4059) was obtained from Abcam. Filamin A Rabbit pAb (A0927) was
purchased from Company ABclonal, Inc. MYH9 Polyclonal antibody (cat.
11128-1-AP) and Talin-1 Polyclonal antibody (cat. 14168-1-AP) were
purchased from Proteintech Group, Inc. Ferrostatin-1 and Erastin were
obtained from MedChemExpress. DTT, RPMI 1640 medium, fetal bovine serum
(FBS), and Dulbecco’s modified Eagle medium (DMEM) were purchased from
Thermo Fisher Scientific Inc. Anti-CD45-APC/Cyanine7 (cat.103116),
anti-CD3-PerCP/Cyanine5.5 (cat.100218), anti-CD4-FITC (cat.100406),
anti-CD8a-APC (cat.100712), anti-CD11c-FITC (cat.117306), anti-CD86-APC
(cat.105012), anti-CD80-PE (cat.104708), anti-CD11b-FITC (cat.101206),
anti-F4/80-PerCP/Cyanine5.5 (cat.123126), anti-CD206-PE (cat.141706),
anti-CD44-FITC (cat.103022), anti-CD62L-PE (cat.161204), anti-IFN-γ-PE
(cat.505808) were purchased from Biolegend (USA). InVivoMAb anti-mouse
CD8α (cat. BE0061) was purchased from BioXCell. H&E Stain Kit was
obtained from Beijing Solarbio Science & Technology Co., Ltd. (Beijing,
China). The TUNEL cell apoptosis detection kit was bought from Dalian
Meilun Biotechnology Co., Ltd. All chemicals were used as received
without further treatment.
Characterization
The PXRD measurement was examined with a Rigaku D/max-TTR-III
diffractometer using Cu-Ka radiation (λ = 0.15405 nm) at 40 kV and
40 mA. TEM and HRTEM were recorded on an FEI Tecnai G2 S-Twin
transmission electron microscope equipped with a field emission gun
operating at 200 kV. XPS spectra were carried out using an ESCALAB 250
instrument. The zeta potential measurement for different samples was
performed on a Malvern Zeta sizer Nan Nano ZS90. A UV-1601
spectrophotometer was used to obtain the UV–vis–NIR absorption
spectrum. ESR spectra were acquired by a Bruker EMX1598 spectrometer.
The element quantitative analysis of the sample was conducted on
ICP-OES (Agilent 725, Agilent Technologies, USA). The flow cytometry
assays were performed on a BD Accuri C6 flow cytometer (USA). A CLSM
(Leica TCS SP8) was adopted to obtain the fluorescence image.
Synthesis of Pd[2]Sn
The Pd[2]Sn with different morphology were synthesized with the same
procedure, except for the amount of MAHC added. Briefly, oleylamine
(20 mL), a certain quantity of MAHC, Pd(acac)[2] (0.2 mmol), and
Sn(OAc)[2] (0.1 mmol) were added into the four-necked flask (100 mL).
Under continuous stirring at 40 °C, the flask containing various
precursors was vacuumed for 20 min, then protected by injecting with
nitrogen at 60 °C for 30 min to form a uniform solution. After that,
the TOP (1 mL) was injected into a four-necked flask and quickly heated
up to 200 °C and maintained at this temperature for 30 min.
Subsequently, the mixed solution in the flask was heated to 300 °C and
preserved for another 30 min. Finally, the flask was discontinued
heating and cooled to room temperature (RT) naturally. Pd[2]Sn
intermetallics were obtained by centrifugation and washed with ethanol
and chloroform. In a controllable way, the Pd[2]Sn NDs or NRs with
different lengths were synthesized by adding MAHC with different
amounts.
Synthesis of Pd[2]Sn@GOx-SP
The as-prepared Pd[2]Sn nanoparticles dissolved in cyclohexane
(1 mg mL^−1, 10 mL) were added dropwise into chloroform solution of SP
(2 mg mL^−1, 30 mL). The solvent of the mixture was evaporated in a
rotary evaporator under a vacuum at 60 °C to collect Pd[2]Sn-SP. Then,
the obtained Pd[2]Sn-SP and GOx (5 mg) were dispersed in H[2]O and
stirred vigorously overnight. The Pd[2]Sn@GOx-SP nanocomposites were
performed by centrifugation and washed with water and ethanol for
further use.
In vitro photothermal performance of Pd[2]Sn-SP
The laser wavelength used in this work was 808 nm, which was generated
by the 808 nm multimode fiber coupled laser (Changchun Laser
Optoelectronics Technology Co., Ltd). The power density of the 808 nm
laser used in the in vitro or in vivo experiments was 0.8 W cm^−2. The
rod-shaped Pd[2]Sn-SP solution with various concentrations was
irradiated with laser at RT for 10 min and photographed at specified
intervals by thermal imaging equipment (FLIR System E40). In addition,
pure water was performed as a comparison. Furthermore, the Pd[2]Sn-SP
solution was irradiated with a laser for four cycles to explore the
photothermal stability, and the change of temperature was plotted.
Selecting the cooling phase curve to calculate the photothermal
conversion efficiency (η) of Pd[2]Sn-SP based on the following formula:
[MATH: η=hATma<
mi>x−Tsurr−QsI(1−10−A<
mi>λ) :MATH]
1
h, A, T[max], T[surr], I, and A[λ] refer to heat transfer coefficient,
superficial area, maximum equilibrium temperature, ambient temperature,
laser fluence, and absorption value, respectively.
Q[s] = (5.4 × 10^−^4) I.
ESR measurement
The ·OH was measured by using DMPO as the trapping agent through ESR
analysis. The Pd[2]Sn-SP (100 μL, 1 mg mL^−1), PBS (340 μL, pH = 5.5),
and DMPO (10 μL) were mixed, and the solution was analyzed by an
electron paramagnetic resonance spectrometer after the rapid injection
of H[2]O[2] (100 mM, 50 μL). In addition, the stimulation of the laser
was performed at the same reaction time, and the characteristic signal
peak with the intensity of 1:2:2:1 for ·OH was captured.
Catalase (CAT)-mimic activity of Pd[2]Sn-SP
The detection of O[2] generation induced by Pd[2]Sn-SP with different
morphologies was conducted to evaluate the CAT-mimic activity by using
a dissolved oxygen meter. In a typical process, the Pd[2]Sn-SP (0, 50,
100, 200, and 400 μg mL^−1) was mixed with H[2]O[2], and the oxygen
concentration (mg L^−1) was recorded last for 6 min, respectively.
Oxidase (OXD)-mimic activity of Pd[2]Sn-SP
TMB as the substrate was used to measure the OXD-mimic activity of
Pd[2]Sn-SP. Typically, TMB (300 μg mL^−1) and Pd[2]Sn-SP (200 μg mL^−1)
were dispersed in PBS (3 mL) at RT, and the absorbance was measured
after a certain reaction time by using a UV–vis–NIR spectrophotometer.
Glucose consumption
Pd[2]Sn@GOx-SP (0 and 200 μg mL^−1) were mixed with glucose solution
(1 mg mL^−1). At various time intervals, the mixed solution (0.5 mL)
was added with dinitrosalicylic acid (DNS) reagent (1.5 mL). Then, the
mixed solution was heated to 100 °C and maintained for 5 min, then
cooled to RT. Subsequently, the absorbance of the mixed solution at
595 nm was measured by UV–vis–NIR spectrophotometer.
Peroxidase (POD)-mimic activity of Pd[2]Sn-SP and kinetic assay
The POD-mimic activity for the ·OH generation of Pd[2]Sn-SP was
measured using TMB and OPD as the substrates. Briefly, 3 mL PBS
including Pd[2]Sn-SP (100 μg mL^−1) and TMB (300 μg mL^−1) was mixed
with H[2]O[2], and the absorbance in various groups was recorded after
a certain reaction time. The groups exposed to a laser were performed
at the same conditions except for laser action.
The POD-mimicking kinetics of Pd[2]Sn-SP (NDs, short NRs, medium NRs,
long NRs) with different morphologies were also evaluated. TMB
(300 μg mL^−1), Pd[2]Sn-SP (100 μg mL^−1), and H[2]O[2] (final
concentrations of 1, 2, 4, 8, and 16 mM) were mixed in PBS (3 mL) at
RT, the absorbance intensity at 652 nm were measured on a UV–vis–NIR
spectrophotometer, and the Michaelis–Menten equation was adopted to
analyze the Michaelis–Menten constant.
The catalytic activity of Pd[2]Sn-SP with different morphologies was
further evaluated by altering the amounts of Pd[2]Sn-SP. Similarly,
equivalent TMB, H[2]O[2], and different amounts of Pd[2]Sn-SP were
added into PBS (3 mL) at RT, and the absorbance of the mixed solution
was recorded at various times. The catalytic activity (units) was
calculated to make a comparison of the Pd[2]Sn-SP with different
morphologies.
MB degradation assay
The prepared MB aqueous solution (10 μg mL^−^1) was mixed with
Pd[2]Sn-SP to form a mixed solution uniformly, which was injected into
the H[2]O[2] solution (final concentration of 50 μM) rapidly. The
absorbance of the mixed solution was detected at various time
intervals.
DPA degradation assay
The generation of singlet oxygen (^1O[2]) by Pd[2]Sn-SP was measured
using DPA as the substrate. First, DPA (1 mg mL^−^1) was dissolved in
DMSO and mixed with Pd[2]Sn-SP. Then, an 808 nm laser was used to
irradiate the mixture, the absorbance was measured at various times.
Density functional theory (DFT) calculation
The Vienna Ab initio Simulation Package was used for DFT calculation.
The plane wave cutoff was set to 500 eV, and the integration over the
Brillouin zone was treated by the Monkhorst–Pack technique with a
(2 × 2 × 1) grid for relevant surfaces and a (4 × 4 × 4) one for the
bulk crystal. The energy and force are converged when the values are
<1 × 10^−^5 eV and −0.05 eV/Å, respectively. For bulk crystal, a
(2 × 2 × 1) supercell containing 24 atoms was considered, while the
(2 × 2 × 1) supercell for the relevant Pd[2]Sn (001, 010, 011) surfaces
was used. The (001, 010, 011) surfaces of different models were
employed in the whole calculation process owing to the (001, 010, 011)
crystal planes are the main exposed crystal planes.
Cell culture
The L929, 3T3, CT26, 4T1, HeLa, SW1990, A549, and HepG2 cells were
obtained from the Heilongjiang Key Laboratory of Molecular Oncology.
The L929, HeLa, and HepG2 cells were seeded in minimum Eagle’s medium
(Procell, Wuhan, China), 3T3 cells were seeded in DMEM (Procell, Wuhan,
China), CT26 and 4T1 cells were seeded in RPMI 1640 medium (Procell,
Wuhan, China), SW1990 cells were seeded in Leibovitz’s L-15 medium
(Procell, Wuhan, China), A549 cells were seeded in Ham’s F-12K medium
(Procell, Wuhan, China), and all medium were added with 10% FBS and 1%
penicillin–streptomycin. The cells were cultured at 37 °C under a 5%
CO[2] atmosphere.
Cellular uptake and subcellular localization
CT26 cells were seeded in 6-well plates and cultured for 24 h. Then,
the cell culture medium was replaced with a fresh culture medium
containing FITC-labeled Pd[2]Sn@GOx-SP for 0, 1, 2, 4, and 8 h,
respectively. For subcellular localization, the treated cells were
further stained with commercial Lyso-Tracker Red DND-99 (100 nM) and
DAPI according to the manufacturer’s guidelines. The fluorescence
images of cells were captured by CLSM after rinsing with PBS and fixing
with glutaraldehyde (2.5%). The colocalization analysis was performed
by Image J software. For the flow cytometry assay, the treated cells
were resuspended by trypsin and washed with PBS, and the fluorescence
intensity of the cells was recorded by a flow cytometer.
In vitro cytotoxicity assay
L929, 3T3, and CT26 cells were cultured in 96-well plates for 12 h.
Then, the cell culture medium was refreshed with a fresh culture medium
containing Pd[2]Sn-SP or Pd[2]Sn@GOx-SP with different concentrations,
and the laser-related groups were irradiated with an 808 nm laser.
Finally, the standard MTT assay was conducted. Similarly, 4T1, HeLa,
SW1990, A549, and HepG2 cells were cultured in 96-well plates for 12 h.
Then, these cells were treated with Pd[2]Sn@GOx-SP + laser, and the
relative cell viability was obtained by MTT assay.
In vitro cell experiments evaluation
For most cell experiments, CT26 cells are subjected to the same culture
and subsequent processing. Briefly, CT26 cells were seeded in a 6-well
plate and cultured for 24 h. Then, the cultivated cells were performed
with different conditions, containing control (G1), laser (G2),
Pd[2]Sn-SP (G3), Pd[2]Sn@GOx-SP (G4), Pd[2]Sn@GOx-SP + laser (G5).
After co-incubation at 37 °C for 4 h, the laser-related groups were
irradiated with 808 nm laser (0.8 W cm^−^2, 5 min), and then the cells
were stained with the corresponding fluorescent dyes for analysis.
For the detection of intracellular ROS production, the treated cells
were stained with DCFH-DA and DAPI for 20 and 15 min in the dark,
respectively. Finally, the fluorescence of cells was captured by a
CLSM. For flow cytometry analysis, the treated cells were resuspended
by trypsin and washed with PBS, and the fluorescence intensity of the
cells was recorded by a flow cytometer.
For the detection of intracellular O[2] generation, the cells were
treated with [Ru(dpp)[3]]Cl[2] (with the final concentration of 30 μM)
for 6 h before various treatments. The treated cells were stained with
DAPI for 15 min. Finally, the fluorescence of cells was observed by a
CLSM.
For the detection of mitochondrial membrane potential, the treated
cells were stained with JC-1 and DAPI for 20 and 15 min, respectively.
Finally, the fluorescence of cells was detected by a CLSM.
For the detection of living/dead cells, the treated cells were stained
with Calcein-AM and PI at 37 °C for 30 min. Finally, the fluorescence
of cells was captured by a CLSM. For the cell death analysis, the
treated cells were stained with Annexin V-FITC/PI apoptosis detection
kit. Finally, the cells were detected by a flow cytometer.
For the observation of cell morphology, the morphology of the treated
cells was directly observed by a fluorescence microscope. Furthermore,
the treated CT26 cells were treated with trypsin and then centrifuged
to collect the precipitate, which was fixed, dehydrated, embedded,
sliced, and measured by bio-TEM.
For the CRT exposure and HMGB1 migration analysis, the treated CT26
cells were washed with PBS and fixed with stationary liquid for 10 min.
The anti-CRT and anti-HMGB1 were added to the cells and incubated at
4 °C for overnight, respectively. Finally, the cells were further
incubated by Alexa Fluor 488 or Alexa Fluor 555-conjugated secondary
antibody and DAPI. Finally, the fluorescence of cells was captured by a
CLSM.
In vitro tumor spheroid assay
CT26 cells were cultured in a 96-well plate coated with 1.5% agarose
and monitored until the spheroid size reached about 600 μm. Then, fresh
culture medium containing FITC-labeled Pd[2]Sn@GOx-SP was adopted to
replace the old culture medium and co-cultured for 0, 2, 4, and 8 h,
respectively. Finally, the tumor spheroids were cleaned with PBS and
the fluorescence intensity was observed by a CLSM.
For the cytotoxicity analysis of Pd[2]Sn@GOx-SP on tumor spheroids, the
tumor spheroids were performed with different conditions (G1, control;
G2, laser; G3, Pd[2]Sn-SP; G4, Pd[2]Sn@GOx-SP; G5,
Pd[2]Sn@GOx-SP + laser). After co-incubation at 37 °C for 8 h, the
laser-related groups were exposed to laser, and then the tumor
spheroids were stained with Calcein-AM and PI for 30 min. Finally, the
fluorescence of cells was captured by a CLSM.
GSDMD and NLRP3 knockdown
GSDMD and NLRP3 siRNAs obtained from General Biology (Anhui, China)
were transfected into CT26 cells. The siRNA sequences of GSDMD were
(5′-CCGAGGUGCUGCAGACAAATT-3′) and (5′-UUUGUCUGCAGCACCUCGGTT-3′),
(5′-CCUCAGAACUGGAGAGCUUTT-3′) and (5′-AAGCUCUCCAGUUCUGAGGTT-3′),
(5′-GCUAGAAGAAUGUGGCCUATT-3′) and (5′-UAGGCCACAUUCUUCUAGCTT-3′). The
siRNA sequences for NLRP3 were (5′-GAAAGAAACUGCUGCCCAATT-3′) and
(5′-UUGGGCAGCAGUUUCUUUCTT-3′), (5′-GUACUUAAAUCGUGAAACATT-3′) and
(5′-UGUUUCACGAUUUAAGUACTT-3′), (5′-GGAUGGGUUUGCUGGGAUATT-3′) and
(5′-UAUCCCAGCAAACCCAUCCTT-3′). Empty planter plasmid or siRNA control
plasmid was used as a negative control. Briefly, CT26 cells were grown
in 6-well plates and transformed with 2 μg of plasmids using
Lipofectamine 2000 (Invitrogen, Waltham, MA, USA) according to the
manufacturer’s protocol. After the transformation of 48 h, cells were
harvested for further experiments. The effectiveness of knockdown was
confirmed by using western blot.
Western blotting
CT26 cells were seeded in a 6-well plate and cultured for 24 h. The
cultivated cells were performed with different conditions (G1–G5).
Then, the cells were washed with cold PBS for twice and added to
ice-cold lysis buffer for western blotting (Beyotime, China). Protein
concentration was determined using the BCA protein assay kit (Beyotime,
China). The protein with the same amount was added to a 10% SDS
polyacrylamide gel, electrophoresed, and transferred to polyvinylidene
fluoride (PVDF) membranes (Roche Diagnostics GmbH, Mannheim, Germany).
Then, 5% skimmed milk was used to block the gels for 1 h, which were
further incubated with primary antibodies and secondary antibodies.
Finally, the protein bands were detected with a fluorescent
luminescence detector. The antibodies used in this study were as
follows: NLRP3 (15101, CST), anti-GSDMD antibody (69469, CST),
anti-Caspase-1 antibody (3866, CST), anti-β-actin antibody (TA-09,
ZSGB-BIO), anti-GSDMD-N (69469, CST), and SLC7A11 (PK30003,
Proteintech).
In vitro dendritic cells (DCs) stimulation
The bone-marrow-derived DCs were obtained from the leg bones of male
BALB/c mice. To begin with, the cultured CT26 cells were performed with
different treatments (G1–G5). Then, the treated CT26 cells were
collected and co-cultured with immature DCs for 24 h. Finally, the
non-adherent cells were collected and stained with anti-CD11c-FITC,
anti-CD86-APC, and anti-CD80-PE antibodies to analyze the maturation
level of DCs by flow cytometry.
In vitro and in vivo PA and CT imaging performance
Pd[2]Sn-SP with various concentrations were prepared to explore the in
vitro imaging performance by the PA and CT equipment. As for the in
vivo PA and CT imaging, the Pd[2]Sn-SP (100 μL) dissolved in saline was
i.v. administrated into tumor-bearing mice (at the dose of
10 mg kg^−1). In addition, the Pd[2]Sn-SP (100 μL) dissolved in saline
was also i.t. injected into tumor-bearing mice to further investigate
the CT imaging effect. The PA and CT imaging was obtained at various
times by using a Vevo LAZR system (VisualSonics Inc. New York, NY) and
a small animal X-ray CT imaging system (Quantum GX, PerkinElmer),
respectively. In addition, the blood oxygen saturation (SaO[2]) in the
tumor was further monitored at various times.
In vivo biodistribution and pharmacokinetics of Pd[2]Sn@GOx-SP
The tumor-bearing mice (n = 3 mice for each group) were treated by i.v.
injection with Pd[2]Sn@GOx-SP (at the dose of 10 mg kg^−1). After 1, 3,
6, 12, and 24 h, the major organs and tumors were collected by
sacrificing these mice and analyzed by an ICP-OES. For pharmacokinetics
analysis, the BALB/c mice (n = 3 mice for each group) were treated by
i.v. injection with Pd[2]Sn@GOx-SP (at the dose of 10 mg kg^−^1). After
0, 2, 4, 8, 15, 30 min, 1, 2, 4, 8, 12, and 24 h, the blood was
extracted and analyzed by an ICP-OES.
In vivo therapeutic evaluation of Pd[2]Sn@GOx-SP
The tumor-bearing mice were randomly allocated into five groups (n = 5
mice for each group), including control (G1), laser (G2), Pd[2]Sn-SP
(G3), Pd[2]Sn@GOx-SP (G4), Pd[2]Sn@GOx-SP + laser (G5). The mice were
i.v. injected with saline or nano-drug on days 1, 5, 9, and 13. During
the treatment process, the body weight and tumor size of the mice were
measured every 2 days. The tumor volume (V) was calculated according to
V = lw^2/2. The mice were sacrificed on day 15, and the tumors and main
organs were collected for further analysis. The tumor sections of
different groups were stained with H&E, TUNEL, Ki67, C-Caspase-1, and
NLRP3. The main organ sections were stained with H&E.
To conduct the tumor rechallenge assay, the tumor-bearing mice were
randomly allocated into five groups (G1–G5, n = 5 mice for each group).
On day 12, the mice were rechallenged with CT26 cells (1 × 10^6) on the
left back. The body weight and tumor size were measured every 2 days.
On day 26, the mice were sacrificed and the spleens were collected to
detect immune cells.
To explore the in vivo anti-metastasis efficacy, the tumor-bearing mice
were randomly allocated into five groups (G1–G5, n = 3 mice for each
group). After being treated for 12 d, the CT26 cells were i.v. injected
into these mice. On day 25, the mice were all sacrificed to collect the
lungs for H&E staining assays.
In vivo anti-tumor immunity
For the macrophage polarization analysis, the tumors of mice with
different treatments were extracted, homogenized in PBS, and filtered
to obtain single-cell suspension. Subsequently, the cells were stained
with anti-CD45-APC/Cyanine7, anti-CD11b-FITC,
anti-F4/80-PerCP/Cyanine5.5, anti-CD86-APC, and anti-CD206-PE
antibodies, and the macrophage phenotype was analyzed by flow
cytometry.
For T-cell activation analysis, the collected spleens and tumors were
homogenized in PBS, and filtered to obtain single-cell suspension.
Afterwards, the cells were stained with anti-CD45-APC/Cyanine7,
anti-CD3-PerCP/Cyanine5.5, anti-CD4-FITC, and anti-CD8a-APC, and the
activation of T cells was analyzed by flow cytometry. The activated CD8
T cells were further analyzed by staining primary and distant tumor
cells with anti-IFN-γ-PE antibodies.
For DCs maturation analysis, the collected spleens were homogenized in
PBS, and filtered to obtain single-cell suspension. Afterwards, the
cells were stained with anti-CD11c-FITC, anti-CD86-APC, and
anti-CD80-PE antibodies to analyze the maturation of DCs using flow
cytometry.
To analyze memory T cells, the collected spleens were homogenized in
PBS, and filtered to obtain single-cell suspension. Afterwards, the
cells were stained with anti-CD45-APC/Cyanine7,
anti-CD3-PerCP/Cyanine5.5, anti-CD8a-APC, anti-CD44-FITC, anti-CD62L-PE
antibodies to evaluate the memory T cell using flow cytometry.
RNA sequencing
The tumor-bearing mice were randomly allocated into two groups,
including control and Pd[2]Sn@GOx-SP + laser. After 1 week of
treatment, the mice were sacrificed and the tumors were extracted for
RNA sequencing according to the requirement of GENEWIZ, Inc. (Suzhou,
China).
Statistical analysis
Quantitative data were indicated as mean ± S.D. The software of
GraphPad Prism 9.0 was adopted to assess the statistical analysis,
which was performed using the Student’s t-test and one/two-way ANOVA.
The statistical significance was attained at *p < 0.05, **p < 0.01,
***p < 0.001, and ****p < 0.0001.
Reporting summary
Further information on research design is available in the [286]Nature
Portfolio Reporting Summary linked to this article.
Supplementary information
[287]Supplementary Information^ (17.3MB, pdf)
[288]Peer Review File^ (91.9MB, pdf)
[289]Reporting Summary^ (1.2MB, pdf)
Source data
[290]Source Data^ (44.3MB, xlsx)
Acknowledgements