Abstract

   The development of efficient therapeutic strategies to promote
   ferroptotic cell death offers significant potential for hepatocellular
   carcinoma (HCC) treatment. Herein, this study presents an HCC‐targeted
   nanoplatform that integrates bimetallic FeMoO[4] nanoparticles with
   CO‐releasing molecules, and further camouflaged with SP94
   peptide‐modified macrophage membrane for enhanced ferroptosis‐driven
   multi‐modal therapy of HCC. Leveraging the multi‐enzyme activities of
   the multivalent metallic elements, the nanoplatform not only decomposes
   H[2]O[2] to generate oxygen and alleviate tumor hypoxia but also
   depletes glutathione to inactivate glutathione peroxides 4, which
   amplify sonodynamic therapy and ferroptotic tumor death under
   ultrasound (US) irradiation. Meanwhile, the nanoplatform catalyzes the
   Fenton reaction to produce hydroxyl radicals for chemodynamic therapy.
   Elevated intracellular reactive oxygen species trigger the cascade
   release of CO, leading to lethal lipid peroxidation and further
   enhancing ferroptosis‐mediated tumor therapy. This nanoplatform
   demonstrates robust anti‐tumor efficacy under US irradiation with
   favorable biosafety in both subcutaneous and orthotopic HCC models,
   representing a promising therapeutic approach for HCC. Additionally,
   the findings offer new insights into tumor microenvironment modulation
   to optimize US‐triggered multi‐modal cancer therapy.

   Keywords: chemodynamic therapy, gas therapy, hepatocellular carcinoma,
   nanoenzyme, sonodynamic therapy
     __________________________________________________________________

   This study reports an HCC‐targeted nanoplatform that combines FeMoO[4]
   nanoenzyme, CO‐releasing compounds, and SP94‐modified macrophage
   membrane. This platform improves ferroptosis‐driven multimodal
   treatment by producing oxygen, depleting glutathione, and creating
   hydroxyl radicals. It has excellent anti‐tumor activity under
   ultrasonic irradiation, indicating a feasible technique for treating
   hepatocellular carcinoma while altering the tumor microenvironment.

   graphic file with name ADVS-12-2413833-g002.jpg

1. Introduction

   Hepatocellular carcinoma (HCC) ranks as the sixth most commonly
   diagnosed cancer and the third leading cause of cancer‐related
   mortality worldwide.^[ [42]^1 ^] Due to the insidious nature of HCC,
   most patients are often diagnosed at advanced stages and are only
   eligible to systemic treatments.^[ [43]^2 ^] Despite the evolution of
   molecular‐targeted therapy and immunotherapy has dramatically changed
   the systemic treatment landscape of advanced HCC, the overall response
   is still limited by the inherent or acquired resistance in cancer
   cells.^[ [44]^3 ^] Meanwhile, systemic therapeutic strategies often
   cause severe adverse effects.^[ [45]^4 ^] Therefore, there is an urgent
   need to develop a precision therapeutic strategy for more effective
   treatment of HCC.

   Ferroptosis, an iron‐dependent form of non‐apoptotic cell death, is
   characterized by the excessive accumulation of lethal lipid
   peroxidation (LPO) resulting from reactive oxygen species (ROS).^[
   [46]^5 ^] In addition, ferroptosis is also accompanied by glutathione
   peroxide 4 (GPX4) inactivation mediated by glutathione (GSH) depletion,
   which further induces redox homeostasis imbalance and tumor cell
   death.^[ [47]^6 ^] Since liver is an important organ for iron storage
   and metabolism,^[ [48]^7 ^] ferroptosis‐based therapy is thus expected
   as a promising treatment for HCC.^[ [49]^8 ^] The occurrence of
   ferroptosis highly relies on ROS, and it is therefore crucial to
   develop effective strategies to enhance the intracellular ROS levels
   for ferroptotic tumor death. To date, various ROS‐generated therapeutic
   modalities such as chemodynamic (CDT), photodynamic therapy (PDT),
   radiotherapy and sonodynamic therapy (SDT) have been extensively
   explored to boost ferroptotic cancer therapy.^[ [50]^9 ^] Among these,
   SDT offers distinct advantages as a non‐invasive treatment modality for
   deep‐seated tumors such as HCC, owing to the excellent tissue
   penetration and favorable biosafety profile of ultrasound (US).^[
   [51]^10 ^]

   During the process of SDT, sonosensitizers are activated by
   low‐intensity US to interact with surrounding oxygen to produce highly
   cytotoxic ROS, and thus results in localized cytotoxicity.^[ [52]^10 ,
   [53]^11 ^] Although a great number of studies have demonstrated the
   promise of using SDT for tumor therapy, its practical application still
   faces several challenges.^[ [54]^12 ^] First, conventional
   sonosensitizers suffer from poor solubility, rapid clearance and
   potential toxicity, which suppresses their SDT efficacy and poses
   safety issues.^[ [55]^13 ^] Second, oxygen consumption caused by rapid
   tumor growth and insufficient oxygen supply caused by abnormal
   vasculature led to a hypoxic state of the tumor microenvironment (TME),
   which greatly restricts the efficacy of oxygen‐dependent SDT.^[ [56]^10
   , [57]^14 ^] Meanwhile, high level of reducing GSH in the TME protects
   cancer cells from oxidative stress, which further impairs SDT‐induced
   cell death.^[ [58]^15 ^] In this regard, it would be of great
   importance to explore novel sonosensitizers with high efficacy and low
   toxicity that can modulate the TME to achieve efficient SDT.

   In addition, due to the high malignancy and complexity of HCC, SDT
   monotherapy may sometimes be insufficient to effectively and
   consistently combat HCC.^[ [59]^16 ^] As a result, the development of
   combined SDT treatment strategies to elevate therapeutic outcomes is
   urgently expected.^[ [60]^17 ^] Gas therapy, which applies gaseous
   signaling molecules, such as nitric oxide, hydrogen sulfide, and carbon
   monoxide (CO), has been proposed as a “green” treatment because of its
   negligible side effects.^[ [61]^18 ^] In recent studies, CO has also
   been demonstrated to drive ferroptosis against cancer through
   destroying mitochondrial integrity, producing ROS, and downregulating
   GPX4.^[ [62]^18 , [63]^19 ^] Currently, the localized delivery of
   external CO to tumor tissues via stimuli‐responsive CO‐releasing
   molecules (CORMs) represents an advanced category of CO‐based gas
   therapeutic strategy, greatly reducing the risk of poisoning associated
   with direct CO inhalation.^[ [64]^20 ^] Building on this, the
   combination of robust SDT with on‐demand CO gas therapy would be a
   powerful synergistic multi‐modal treatment strategy to induce
   ferroptosis and achieve effective inhibition against HCC.

   Inorganic nanomaterials, featuring superior physicochemical properties
   and versatility, have recently been explored as a new generation of
   sonosensitizers.^[ [65]^21 ^] By exploiting multivalent elements
   (Fe^2+/^3+, Cu^+/Cu^2+, Mo^4+/Mo^6+, Co^2+/^3+, etc.), some inorganic
   nanomaterials can mimic natural enzymes and exhibit high catalytic
   activity in the TME.^[ [66]^22 ^] Herein, we explored a bimetallic
   compound, FeMoO[4] nanoparticles (FM), as a sonosensitizer and
   investigated its role in modulating the TME to achieve enhanced SDT
   upon US irradiation. Thereafter, CORM‐401, a commercially available
   oxidant‐sensitive CORM (Figure [67]S1, Supporting Information),^[
   [68]^23 ^] was readily loaded into FM nanoparticles to form FM/C
   nanoparticles for synergistic multi‐modal treatment of HCC. In order to
   achieve the specific accumulation of therapeutic agents at the HCC
   tumor, macrophage membrane inserted with SP94‐peptide was further
   cloaked on FM/C to construct a dual‐targeting biomimetic nanoplatform
   (PM‐FM/C) for subcutaneous and orthotopic HCC therapy (Scheme [69]1 ).
   Leveraging the multi‐enzyme activities, FM could render Fenton reaction
   to produce hydroxyl radical (·OH) for CDT and catalyze the
   decomposition of H[2]O[2] to produce O[2] to relieve hypoxia, which
   improved its SDT performance even in a hypoxic environment. The high
   level of intracellular ROS induced the cascade release of CO, which
   together led to the accumulation of lethal LPO for inducing ferroptotic
   cancer cell death. Meanwhile, the nanoplatform can react with reducing
   species in the TME for GSH depletion, which further elevated the levels
   of LPO through the inactivation of GPX4. Our findings show that PM‐FM/C
   induces robust ferroptosis‐based tumor death under US irradiation and
   holds great potential to serve as an efficient multi‐modal therapeutic
   agent for HCC therapy.

Scheme 1.

   Scheme 1
   [70]Open in a new tab

   Schematic illustration of the dual‐targeting biomimetic nanoplatform
   for integrating CDT/SDT/gas therapy to boost synergistic ferroptosis
   against orthotopic HCC. The CO‐releasing molecule CORM‐401 was
   encapsulated into bimetallic compound FM to form a muti‐modal
   therapeutic nanoplatform (FM/C), which was further cloaked with
   macrophage membrane inserted with SP94‐peptide to endow dual‐targeting
   ability toward HCC. The multi‐enzyme activity of FM not only promoted
   CDT to generate ·OH but also improved SDT performance to produce ^1O[2]
   by alleviating hypoxia, which in turn activated CO release. As a
   result, the depletion of GSH and high level of ROS led to the
   accumulation of lethal lipid peroxidation (LPO) to promote
   ferroptosis‐based tumor death in an orthotopic HCC model.

2. Results and Discussion

2.1. Preparation and Characterization of FM

   FM was first synthesized by the hydrothermal reaction of
   Na[2]MoO[4]·2H[2]O and FeSO[4]·7H[2]O (Figure [71] 1a). The
   hydrothermal synthesis method has been widely applied to produce
   various classes of inorganic materials with porous nanostructure.^[
   [72]^24 ^] Dynamic laser scattering (DLS) analysis revealed that the
   as‐synthesized FM had an average size of 106 nm, and the inserted
   transmission electron microscopy (TEM) image showed a uniform spherical
   morphology with the rough edge (Figure [73]1b). The zeta potential of
   FM was measured to be 36.41 mV (Figure [74]S2, Supporting Information).
   The FTIR spectrum displayed a sharp band at 916 cm^−1 with a shoulder
   at 960 cm^−1, which was attributed to the stretching and bending
   vibrations of Mo = O and Mo‐O‐Mo, respectively (Figure [75]S3,
   Supporting Information). Moreover, the absorption band at 624 cm^−1
   could be ascribed to the stretching vibration of Mo‐O‐Fe. The N[2]
   adsorption‐desorption analysis clearly showed that FM has type II
   isothermal curves (Figure [76]1c), indicating the microporous structure
   of FM. Furthermore, two peaks were observed in the pore size
   distribution ranging from 1 to 100 nm (Figure [77]1c), suggesting that
   some mesopores coexist with the macropores. The average pore diameter
   was calculated to be 13.8 nm. During the hydrothermal process, high
   hydrothermal temperature promotes extensive nucleation of primary
   particles from homogeneous aqueous solution at the early stages,
   followed by the aggregation of these primary particles into spherical
   structures with voids between them.^[ [78]^25 ^] Therefore, we deduced
   that micro‐ and mesoporous structures of FM were achieved as a result
   of aggregate recrystallization.

Figure 1.

   Figure 1
   [79]Open in a new tab

   Synthesis and characterization of FM. a) The illustration of the
   synthesis of FM. The figure was created in BioRender.com. b)
   Representative DLS result and TEM image (inserted) of FM. Scale bar:
   100 nm. c) N[2] absorption–desorption curves and pore size distribution
   curve (inserted) of FM. d) XPS survey scan spectrum of FM. e) Fe 2p, f)
   Mo 3d, and g) O 1s peaks in the XPS spectra of FM. h) Oxygen vacancy
   detected by ESR of FM. i) ESR spectra of electron holes trapped by
   TEMPO in the presence of FM upon US irradiation. j) ESR spectra of
   ^1O[2] trapped by TEMP in the presence of FM upon US irradiation. k)
   The ^1O[2] generation ability of FM upon US irradiation detected by
   ABDA. l) The O[2] release profiles of FM at different concentrations in
   the presence of H[2]O[2]. m) Plots of A/A[0] of ADBA in the presence of
   FM versus US irradiation time under different conditions, where A[0]
   and A are the absorbance of ABDA (378 nm) before and after US
   irradiation, respectively. n) The ^•OH generation ability detected by
   the TMB probe in the presence of H[2]O[2]. o) Schematic illustration
   delineating the multi‐enzyme catalytic ability of FM for enhanced ROS
   generation. Data are presented as mean ± SD (n = 3).

   X‐ray photoelectron spectroscopy (XPS) measurements were performed to
   characterize the surface elemental compositions of FM (Figure [80]1d).
   The survey scan displayed the typical peaks of Fe, Mo, and O elements.
   The high‐resolution XPS spectrum of Fe 2p revealed the coexistence of
   Fe^2+ oxidation state (710.77 and 724.27 eV) with some Fe^3+ species
   (713.64 and 727.71 eV) (Figure [81]1e), while the Mo 3d spectrum
   demonstrated the presence of Mo^6+ oxidation state (232.6 and 235.8 eV)
   within FM (Figure [82]1f). Moreover, the asymmetric peak in the O 1s
   spectrum could be deconvoluted into two peaks associated with lattice
   oxide (530.4 eV) and absorbed oxygen (531.6 eV), respectively
   (Figure [83]1g). The electron spin resonance (ESR) spectrum showed a
   characteristic signal peak with g‐factor of 2.003, further confirming
   the oxygen‐deficient structure of FM (Figure [84]1h). The
   4‐amino‐2,2,6,6‐tetramethylpiperidine‐1‐oxyl (TEMPO) probe analysis
   depicted in Figure [85]1i confirmed the presence of electron holes
   within FM after US irradiation, which is advantageous for sonodynamic
   performance. Moreover, singlet oxygen (^1O[2]) signals trapped by
   2,2,6,6‐tetramethylpiperidine (TEMP) were evidently displayed in the
   ESR spectrum of FM after US stimulation for 10 min (Figure [86]1j). The
   above results confirmed the successful synthesis of bimetallic FeMoO[4]
   nanoparticles, and their abundant oxygen vacancies as well as the
   US‐induced separation of electron‐hole pairs contributed to the
   effective sonosensitizing properties of FM.

2.2. Multi‐Enzyme Catalytic Ability of FM for Enhanced ROS Generation

   To evaluate the sonodynamic performance of FM,
   9,10‐anthracenediyl‐bis(methylene)dimalonic acid (ABDA) was used to
   detect ^1O[2] generation under US irradiation. As displayed in
   Figure [87]1k, the characteristic peaks of ADBA progressively decreased
   with the increase of US irradiation time, indicating the sustained
   generation of ^1O[2]. Considering the hypoxic microenvironment at the
   tumor site, we also performed ^1O[2] detection in deoxygenated FM
   solution upon US irradiation, and it was undoubtedly that ^1O[2]
   production efficiency was remarkably inhibited under hypoxic condition
   (Figure [88]S4a, Supporting Information). In this regard, we further
   examined the multi‐enzyme catalytic activities of FM and the regulatory
   roles in the TME for enhancing ROS generation.

   In view of the high concentration of H[2]O[2] at the tumor site, the
   catalase (CAT)‐mimicking activity to decompose H[2]O[2] to produce O[2]
   was first evaluated by detecting the dissolved oxygen concentration in
   the solutions containing FM (1, 2, or 4 mg mL^−1) with the addition of
   H[2]O[2]. Dissolved oxygen increased sharply upon H[2]O[2] addition,
   and O[2] production increased with FM concentration up to 4 mg mL^−1
   (Figure [89]1l). During the incubation of 5 min, dissolved oxygen was
   maintained at a constant level with only a slight decrease. As
   expected, the ^1O[2] generation ability was partially restored by the
   addition of H[2]O[2], as the value of A/A[0] (378 nm) of the
   “FM+H[2]O[2] (hypoxia)” group was higher than that of the “FM
   (hypoxia)” group (Figures [90]1m and [91]S4b, Supporting Information).
   The elevated ^1O[2] production efficiency in the presence of H[2]O[2]
   was attributed to the CAT‐like catalytic effect of FM to generate O[2],
   which subsequently provided a local reoxygenated atmosphere to promote
   US‐triggered ^1O[2] generation. Therefore, FM itself could not only
   serve as an excellent acoustic sensitizer, but also ameliorate hypoxia
   by utilizing its CAT‐mimicking activity for oxygen production, thereby
   promoting efficient ROS under ultrasound stimulation.

   The existence of Fe^3+/Fe^2+ in FM motivated us to evaluate the
   peroxidase (POD)‐like activity of FM by the oxidation of
   3,3,5,5‐tetramethylbenzidine (TMB) upon H[2]O[2] addition. The
   absorbance peak at 652 nm of the oxidant TMB gradually increased upon
   H[2]O[2] addition in FM solution at pH 5.4, indicating the dramatic
   generation of ·OH (Figure [92]1n). In contrast, the oxidation of TMB
   was weak at pH 7.4 (Figure [93]S5, Supporting Information). The
   Fe^3+/Fe^2+ couple endowed FM with POD‐like activity by rendering a
   Fenton reaction with H[2]O[2] to produce ·OH under acidic condition.
   Glutathione (GSH), as an important intracellular metabolite, can
   scavenge the ·OH and increase the resistance of cancer cells to
   oxidative stress. As shown in Figure [94]S6 (Supporting Information),
   FM could effectively consume GSH, and the conversion efficiency of GSH
   exhibited a significant concentration‐dependent elevation. Overall, the
   multi‐enzyme catalytic activities of FM allowed it to consume excess
   H[2]O[2] in the TME not only to improve the US‐triggered ^1O[2]
   generation by ameliorating hypoxia but also to generate ·OH, resulting
   in a local ROS burst to produce SDT/CDT effects (Figure [95]1o).

2.3. Preparation and Characterization of Dual‐Targeting Biomimetic PM‐FM/C

   Due to the microporous structure, FM is exploited to accommodate
   CORM‐401 to obtain FM/C. Inductively coupled plasma‐Mass‐spectrometry
   (ICP‐MS) analysis showed that the encapsulation efficiency of CORM‐401
   in FM/C reached 58.03% when the incubation molar ratio of FM/CORM‐401
   was 1/1. The average size and zeta potential of FM/C were measured as
   110 nm and 35.32 mV (Figure [96] 2a,b), which had no obvious difference
   with the pristine FM, indicating that CORM‐401 loading had a negligible
   influence on FM. In addition, the TEM image of FM/C was similar to that
   of FM (Figure [97]S7, Supporting Information).

Figure 2.

   Figure 2
   [98]Open in a new tab

   Characterization of PM‐FM/C. a) The average sizes and b) zeta
   potentials of FM, FM/C, M‐FM/C and PM‐FM/C. c) Representative DLS
   result and TEM image (inserted) of PM‐FM/C. Scale bar: 100 nm. d)
   Average size change of FM/C and PM‐FM/C versus the storage time and the
   inserted photo of FM/C and PM‐FM/C dispersed in PBS for 48 h (1: FM/C,
   2: PM‐FM/C). e) Protein analysis of FM, SP94 modified macrophage
   membrane (PM), and PM‐FM/C by SDS‐PAGE. f) Plots of A/A[0] of TMB in
   the presence of FM or PM‐FM/C upon H[2]O[2] addition for different time
   points, where A[0] and A are the absorbances of TMB (652 nm) before and
   after H[2]O[2] addition, respectively. g) Plots of A/A[0] of ADBA in
   the presence of FM or PM‐FM/C versus US irradiation time, where A[0]
   and A are the absorbance of ABDA (378 nm) before and after US
   irradiation, respectively. h) Profiles of CO release from PM‐FM/C under
   different conditions. i) Representative CLSM images of Hepa1‐6 cells
   after incubation with FM/Cy5, M‐FM/Cy5 and PM‐FM/Cy5 for different
   time. Scale bar: 100 µm. j) The mean fluorescence intensity of Cy5 in
   Hepa1‐6 cells after different treatments for 4 h was detected by flow
   cytometry. k) Representative CLSM images of cellular ROS generation in
   Hepa1‐6 cells using DCFH‐DA as a probe. Scale bar: 100 µm. l)
   Representative CLSM images of Hepa1‐6 cells after different treatments
   and m) The corresponding mean fluorescence intensity of cellular CO
   release using COP as a probe. Scale bar: 100 µm. Data are presented as
   mean ± SD (n = 3). ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001, and
   statistical analysis was performed using one‐way ANOVA.

   To improve the stability of FM/C and confer HCC targeting ability, the
   Raw264.7 cell membrane was extracted and modified with SP94‐conjugated
   DSPE‐PEG[2K] (DSPE‐PEG[2K]‐SP94) via the insertion of hydrophobic DSPE
   into the membrane bilayer, and then cloaked onto FM/C by sonication to
   yield the dual‐targeting biomimetic PM‐FM/C (Scheme [99]1). After
   SP94‐modified cell membrane coating, the average hydrodynamic diameter
   of PM‐FM/C increased to 123 nm (Figure [100]2a,c) and the zeta
   potential became negative at −21.54 mV (Figure [101]2b). The TEM image
   showed that a thin layer of cell membrane was observed on the surface
   of PM‐FM/C (Figure [102]2c). As expected, the stability and dispersity
   of PM‐FM/C were apparently improved compared to FM/C, as aggregation
   and precipitation were not observed after 48 h in PBS (Figure [103]2d).
   In addition, PM‐FM/C showed protein bands similar to SP94‐modified
   macrophage membrane (PM) (Figure [104]2e), which further suggested the
   cell membrane was successfully camouflaged on the surface of PM‐FM/C.

2.4. ROS Generation and CO Release of PM‐FM/C In Vitro

   Following the fabrication of PM‐FM/C, we validated the ROS generation
   abilities upon US stimulation or H[2]O[2] addition. As revealed in
   Figure [105]S8 (Supporting Information), PM‐FM/C retained the ability
   to produce ^1O[2] as a sonosensitizer as well as the POD‐like activity
   for ·OH generation. Compared to unencapsulated FM, the ROS generation
   abilities of PM‐FM/C exhibited negligible changes which suggested that
   PM modification did not significantly affect the CDT and SDT
   performance of FM (Figure [106]2f,g).

   After confirming the ROS generation abilities of PM‐FM/C, we then
   monitored the ROS‐activated CO release based on the high binding
   affinity between hemoglobin (Hb) and CO. The adsorption peak at 410 nm
   attributed to HbCO gradually increased, while that at 430 nm associated
   with Hb decreased in the UV spectra of Hb, indicating the sustained
   dissociation of CO with the prolonged time (Figure [107]S9, Supporting
   Information). The release of CO from PM‐FM/C with different treatments
   was compared and the result was displayed in Figure [108]2h. The CO
   release was negligible without any treatments. In contrast, the
   concentration of dissolved CO was rapidly increased upon H[2]O[2]
   addition or increased with prolonged US irradiation time. Furthermore,
   PM‐FM/C showed the highest CO‐released efficiency when it was
   simultaneously exposed to US irradiation and in the presence of
   H[2]O[2]. The above results indicated that ·OH produced by the Fenton
   reaction and ^1O[2] generated by US activation synergistically promoted
   the release of CO from PM‐FM/C.

2.5. ROS Generation and CO Release of PM‐FM/C at Cellular Level

   Considering the dual‐targeting nature of SP94‐modified macrophage
   membrane against HCC, the cellular uptake of the biomimetic nanosystem
   toward Hepa1‐6 cells was first investigated by flow cytometry and
   confocal laser scanning microscopy (CLSM). Fluorescent dye Cy5 was
   chosen as a replacement for CORM‐401 and loaded into FM to construct
   naked FM/Cy5, macrophage membrane cloaking M‐FM/Cy5 and SP94‐modified
   macrophage membrane cloaking PM‐FM/Cy5 for Hepa1‐6 cells incubation.
   The CLSM images of Hpea1‐6 cells showed that the cellular uptake of
   PM‐FM/Cy5 increased over the prolonged time, and the highest Cy5
   fluorescence was detected in the cells treated with PM‐FM/Cy5, as
   compared to the other two groups (Figure [109]2i). Comparison of the
   flow histogram profiles of Hepa1‐6 cells at 4 h exhibited the same
   trend (Figure [110]S10, Supporting Information). Quantitative analysis
   by flow cytometry indicated that the Cy5 fluorescence intensity in
   PM‐FM/Cy5‐treated cells was significantly higher than that in PM/Cy5‐
   and FM/Cy5‐treated cells, by 1.34‐fold and 2.80‐fold, respectively
   (Figure [111]2j). The above results demonstrated the superior targeting
   capability of PM‐FM/Cy5 to HCC cells, which was facilitated by the
   macrophage membrane and SP94 peptide coating.

   Subsequently, we examined the ROS generation of PM‐FM/C in Hepa1‐6
   cells by visualizing the intracellular ROS levels using DCFH‐DA as a
   probe. Negligible green fluorescence signal was observed in the control
   and US‐only groups. Slight green fluorescence signal was detected in
   the PM‐FM and PM‐FM/C groups, indicating the minor ROS production
   generated by the Fenton‐like reaction with the endogenous H[2]O[2] in
   Hepa1‐6 cells (Figure [112]2k). Besides, stronger green fluorescence
   was detected in the PM‐FM/C group, indicating that CO release also
   promoted ROS generation.^[ [113]^26 ^] In contrast, US irradiation
   significantly elevated the intracellular ROS level, as stronger
   fluorescence signal was detected in the PM‐FM+US and PM‐FM/C+US groups
   (Figure [114]2k). Flow cytometric analysis of DCFH‐DA fluorescence in
   Hepa1‐6 cells and the statistical analysis of mean fluorescence
   intensity further confirmed the results (Figure [115]S11, Supporting
   Information). The ROS‐activated CO release property of PM‐FM/C was
   further evaluated in Hepa1‐6 cells using COP (a CO fluorescent probe).
   As depicted in Figure [116]2l,m, green fluorescence was observed in the
   cells treated with PM‐FM/C, revealing the CO release triggered by the
   endogenous H[2]O[2]. In contrast, stronger green fluorescence was
   obviously observed in PM‐FM/C+US group, indicating the large amount of
   CO release due to the elevated ROS level under US irradiation. Given
   all the above findings, it can be concluded that US‐triggered PM‐FM/C
   not only generated ROS for rendering CDT/SDT effect in the tumor
   environment but also mediated cascade‐activated CO release for gas
   therapy, which may result in synergistic therapeutic effects for
   cancer.

2.6. In Vitro Synergistic Therapeutic Effect of PM‐FM/C

   Given that PM‐FM/C could target Hepa1‐6 cells to facilitate cell
   internalization and then generate significant amounts of ROS and CO
   under US stimulation, we further explored the in vitro therapeutic
   potential against Hepa1‐6 cells. The biosafety and compatibility of
   PM‐FM and PM‐FM/C were first evaluated in normal 293T cells by CCK‐8
   assay. The cell viabilities of 293T cells remained above 95.0% after
   24 h incubation with PM‐FM or PM‐FM/C at the concentrations ranging
   from 6.25 to 50 µg mL^−1 (Figure [117]S12, Supporting Information),
   indicating that PM‐FM and PM‐FM/C had negligible cytotoxicity to normal
   cells. In comparison, concentration‐dependent cytotoxicity was observed
   in Hepa1‐6 cells (Figure [118] 3a), which was perhaps due to the
   intrinsic H[2]O[2] in tumor cells. Notably, Hepa1‐6 cell viability
   significantly decreased following US irradiation (Figure [119]3b), with
   the PM‐FM and PM‐FM/C groups exhibiting viabilities of 76.7% and 42.9%,
   respectively, at a concentration of 50 µg mL^−1. The therapeutic
   efficacy was further confirmed by calcein‐AM and propidium iodide (PI)
   double‐staining assay. As depicted in Figure [120]3c, the PM‐FM/C+US
   group exhibited the highest proportion of dead cells as indicated by
   the strongest red fluorescence signal. The enhanced cytotoxicity of
   PM‐FM/C under US irradiation further emphasized the critical role of
   elevated ROS and CO levels in tumor cell killing, stemming from the
   multi‐catalytic reactions, US‐triggered sonodynamic effect, and
   cascade‐activated CO release.

Figure 3.

   Figure 3
   [121]Open in a new tab

   In vitro anti‐tumor effects of PM‐FM/C. Cell viabilities of Hepa1‐6
   cells treated with PM‐FM or PM‐FM/C without a) or with b) US
   irradiation. c) Representative fluorescence images of Hepa1‐6 cells
   stained with calcein‐AM and PI after different treatments as indicated.
   Scale bar: 100 µm. d) Relative GSH level detected in Hepa1‐6 cells
   after different treatments as indicated. e) Western blot analysis of
   GPX4 expression in Hepa1‐6 cells after different treatments. f) Western
   blot analysis of GPX4 expression in Hepa1‐6 cells pretreated with Fer‐1
   prior to various treatments. g) Representative CLSM images of lipid
   peroxidation on Hepa1‐6 cells membrane with different treatments as
   indicated using Liperfluo as a probe. Scale bar: 50 µm. h)
   Representative CLSM images of Hepa1‐6 cells stained with JC‐1 after
   different treatments as indicated. Scale bar: 50 µm. Data are presented
   as mean ± SD (n = 3). ^**** p < 0.0001 using one‐way ANOVA.

   Ferroptosis is a regulatory cell death associated with excessive LPO
   that ultimately results in membrane rupture. We then investigated the
   ferroptosis‐related indicators in Hepa1‐6 cells after various
   treatments. Reductive GSH serves as the co‐substrate of GPX4 to balance
   the oxidative level in cells. Generally, LPO accumulation process
   involves the depletion of the intracellular GSH to promote GPX4
   inactivation.^[ [122]^6b ^] As except, the intracellular GSH content
   was decreased to 43.19% in the cells treated with PM‐FM/C+US group as
   compared to that in the cells without any treatments (Figure [123]3d).
   Besides, PM‐FM and PM‐FM/C significantly decreased GPX4 expression, and
   the introduce of US irradiation after culture with PM‐FM and PM‐FM/C
   further promoted GPX4 inhibition (Figures [124]3e and [125]S13a,
   Supporting Information). Ferrostatin‐1 (Fer‐1), an inhibitor of
   ferroptosis, was used for the recovery experiment, and the results
   showed that the decrease of GPX4 expression was dramatically alleviated
   (Figures [126]3f and [127]S13b, Supporting Information), which further
   confirmed the involvement of GPX4 regulation in the induction of
   ferroptosis by PM‐FM/C upon US treatment in Hepa1‐6 cells. Furthermore,
   lipid peroxidation induced by different treatments was detected in
   Hepa1‐6 cells. As shown in Figure [128]3g, relatively stronger green
   fluorescence was found in the PM‐FM+US and PM‐FM/C+US groups in
   comparison with the other groups. C11‐BODIPY581/591, an additional
   LPO‐specific fluorescent probe, was further used to indicate LPO, and
   the same trend was observed (Figures [129]S14 and [130]S15, Supporting
   Information). Noticeably, the lipid peroxidation of the PM‐FM/C group
   in Figure [131]3g was not as obvious as the ROS generation as shown in
   Figure [132]2k. This may be attributed to the following reasons. First,
   although ROS was generated by PM‐FM/C treatment, lipid peroxidation
   usually required a longer duration of ROS exposure or higher sustained
   ROS levels to accumulate detectable damage, which may show a more
   delayed response. Moreover, the conditions (e.g., cell culture time and
   assay sensitivity) of these two individual experiments may also differ,
   leading to different outcomes in ROS detection and lipid peroxidation.

   During the process of ferroptosis, mitochondrial damage is considered
   to be one of the key factors leading to cell death. We then evaluated
   mitochondrial function by detecting the mitochondrial membrane
   potentials (MMPs) using
   5,5′,6,6′‐tetrachloro‐1,1′,3,3′‐tetraethyl‐imidacarbocyanine iodide
   (JC‐1) as a probe. In general, JC‐1 aggregates in the mitochondrial
   matrix can produce red fluorescence when staining healthy mitochondria
   with higher potential. However, JC‐1 monomers cannot aggregate in the
   mitochondrial matrix and produce green fluorescence when staining
   damaged mitochondria with lower potential. Compared to the control and
   US alone groups, the PM‐FM/C+US group induced a significant increase in
   green JC‐1 monomers (Figure [133]3h), indicating the damage of the
   mitochondrial membrane.

2.7. Therapeutic Mechanisms of PM‐FM/C at Transcriptome Levels

   To further elucidate the therapeutic mechanisms of US‐triggered PM‐FM/C
   at the molecular level, the transcriptome analysis was performed to
   analyze the mRNA expression patterns of Hepa1‐6 cells after different
   treatments. Approximately 11100 genes were co‐expressed in the three
   groups (Figure [134] 4a). Pearson correlation between different samples
   demonstrated the reality of the RNA‐sequencing data (Figure [135]S16,
   Supporting Information). A total of 24855 genes were identified through
   gene expression analysis. Meanwhile, thousands of differentially
   expressed genes (DEGs) were detected in the samples with different
   treatments (Figure [136]4a,b). The heat map further revealed an evident
   transcriptomics discrepancy in gene expression among the three groups
   (Figure [137]4c). Compared to the control group, PM‐FM/C group
   exhibited 2268 (964 up‐regulated and 1304 down‐regulated,
   Figure [138]4d), while the PM‐FM/C+US group exhibited 3372 DEGs (1538
   up‐regulated and 1834 down‐regulated, Figure [139]4e). In addition, 206
   DEGs were detected in PM‐FM/C+US versus PM‐FM/C, including 93
   up‐regulated genes and 113 down‐regulated genes (Figure [140]4f). The
   results indicated that PM‐FM/C treatment exerted a more obvious effect
   on the change of gene expression pattern after US stimulation.

Figure 4.

   Figure 4
   [141]Open in a new tab

   Transcriptional analysis of Hepa1‐6 cells after different treatments.
   a) VENN graph revealed the number of co‐expressed genes. b) Statistic
   of up‐and down‐regulated genes in different comparisons. c) The heat
   map of DEGs in Hepa1‐6 cells with different treatments as displayed.
   Volcano plots of differentially expressed genes in d) PM‐FM/C versus
   control, e) PM‐FM/C+US versus control, and f) PM‐FM/C+US versus
   PM‐FM/C. g) KEGG pathway enrichment analysis. h) Protein–protein
   interaction networks and the significant gene module in the
   ferroptosis‐related gene set. i) Heatmap of ferroptosis‐related DEGs
   identified by RNA‐seq in Hepa1‐6 cells with different treatments as
   indicated. j) Validation of the mRNA level of representative
   ferroptosis‐related genes by RT‐PCR. Data are presented as mean ± SD (n
   = 3). ^* p < 0.05, ^** p < 0.01, ^*** p < 0.001, ^**** p < 0.0001, and
   statistical analysis was performed using one‐way ANOVA.

   Based on the DEGs between PM‐FM/C+US and the control group, gene
   ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG)
   enrichment analysis were further performed to unveil the biological
   function and pathways associated with the altered expression genes. GO
   results revealed that the phospholipid biosynthesis process was most
   affected in the cells treated with PM‐FM/C plus US stimulation versus
   the control cells (Figure [142]S17, Supporting Information). KEGG
   analysis indicated that the ferroptosis pathway, along with related
   signaling pathways such as MAPK, TNF, and p53, were associated with the
   therapeutic mechanisms of PM‐FM/C+US (Figure [143]4g). Gene set
   enrichment analysis (GSEA) also demonstrated that the treatment with
   PM‐FM/C+US irradiation changed the expression of genes associated with
   ferroptosis (Figure [144]S18, Supporting Information). Inspired by the
   above analysis as well as the ferroptosis‐inducing effect verified at
   the cellular level, we further performed psychophysiological
   interaction (PPI) analysis (Figure [145]4h) and generated a heat map
   diagram (Figure [146]4i) on the 15 differentially expressed
   ferroptosis‐related genes (FRGs) enriched by KEGG analysis.

   The heat map depicted that 7 FRGs were up‐regulated and 8 FRGs were
   down‐regulated in PM‐FM/C‐treated cells, while the changes in gene
   expression were further enhanced by US irradiation. These genes were
   mainly associated with cystine/glutamate transport (Slc7a11), iron
   transport (Trf, Tfrc, Cp), glutathione metabolism (Gclm, Gss, Gpx4),
   and arachidonate metabolism (Acsl4), all of which play important roles
   in the ferroptosis pathway. Further validation using real‐time PCR
   (RT‐PCR) also confirmed the expression differences of some
   representative genes at the mRNA levels (Figure [147]4j). Acsl4 is a
   lipid‐metabolizing enzyme that promotes the production of lipid
   peroxides. The significant upregulation of Acls4 in PM‐FM/C+US‐treated
   cells suggested its involvement in promoting ferroptosis. This occurred
   through the esterification of polyunsaturated fatty acids (PUFAs) to
   acyl‐coenzyme A, a process further catalyzed by LPCAT3 to generate
   PUFA‐containing phospholipids, which were prone to peroxidation.^[
   [148]^27 ^] The reduced transcript level expression of Gpx4, a key gene
   essential for maintaining redox homeostasis and preventing
   ferroptosis,^[ [149]^28 ^] in the PM‐FM/C+US group further supported
   the inhibition of GPX4 by PM‐FM/C+US treatment, which was consistent
   with the results validated by Western blot analysis (Figure [150]3e).
   Furthermore, the different expression of two genes, Gclm, Gss, involved
   in GSH metabolism and cysteine/methionine metabolism, highlighted the
   critical role of the cysteine/GSH/GPX4 axis in facilitating ferroptosis
   induced by PM‐FM/C under US irradiation. TRF/TFRC mediate the cellular
   uptake of iron to maintain iron homeostasis, with their expression
   precisely regulated by the cellular iron level.^[ [151]^29 ^] Following
   treatment with PM‐FM/C, intracellular iron was in an overloaded state,
   which in turn downregulated the expression of Trf and Tfrc. SLC7A11, a
   key upstream regulators of ferroptosis, participates in the
   extracellular uptake of cystine and glutamate release, promotes the
   synthesis of GSH, and maintains the cellular redox homeostasis to
   protect the cells from the damage. The exogenous ROS production and GSH
   depletion caused by US‐triggered PM‐FM/C treatment greatly increased
   oxidative stress in the cells, which induced SLC7A11 expression in an
   attempt to re‐establish redox homeostasis.^[ [152]^30 ^] Thus, the
   therapeutic mechanisms of PM‐FM/C mainly involve GPX4 inhibition and
   lipid peroxides accumulation. At the same time, the cells may engage in
   feedback regulation of some specific genes, especially upstream genes,
   to counteract the damage caused by ferroptosis in an effort to maintain
   homeostasis. RNA‐sequencing results posed a comprehensive insight on
   the profound impact of US‐enhanced PM‐FM/C treatment on Hepa1‐6 cells,
   and further verified the involvement of ferroptosis in the therapeutic
   mechanisms.

2.8. In Vivo Antitumor Study in a Subcutaneous Tumor Model

   Inspired by the excellent therapeutic performance of PM‐FM/C on liver
   cancer cells by inducing ferroptosis cell death, we further explored
   the in vivo antitumor effect of PM‐FM/C in the Hepa1‐6 tumor‐bearing
   mice model. In vivo imaging was first conducted to evaluate the
   targeting ability of the nanosystem for tumor accumulation. The
   tumor‐bearing mice were divided into four groups and administrated with
   PBS, FM/Cy5, M‐FM/Cy5 and PM‐FM/Cy5, respectively, and then imaged at
   different time points. As depicted in Figure [153] 5a, the tumors
   exhibited the strongest fluorescence signal at 6 h post‐injection and
   the signal could be maintained within 24 h. The highest fluorescence
   signals were observed in the mice treated with PM‐FM/Cy5, indicating
   effective accumulation via the enhanced permeability and retention
   effect as well as the enhanced tumor homing effect of the
   SP94‐modificated macrophage membrane. After 24 h of administration,
   tumors in the PM‐FM/Cy5‐treated mice exhibited fluorescence signals
   ≈2.56‐fold and 1.89‐fold higher than those in the M‐FM/Cy5‐treated and
   FM/Cy5‐treated mice, respectively (Figure [154]5b).

Figure 5.

   Figure 5
   [155]Open in a new tab

   In vivo anti‐tumor effects of PM‐FM/C under US irradiation in a
   subcutaneous HCC model. a) In vivo fluorescence imaging and b) relative
   quantification of the fluorescence intensity of Cy5 at different time
   points after intravenous injection of PBS, FM/Cy5, M‐FM/Cy5 and
   PM‐FM/Cy5. Data are presented as mean ± SD (n = 3 mice). c) Schematic
   illustration of the therapeutic procedure (n = 5). The figure was
   created with BioRender.com. d) Tumor volume change curves of the
   Hepa1‐6 tumor‐bearing mice with different treatments. e) Individual
   tumor growth curves of the Hepa1‐6 tumor‐bearing mice with different
   treatments. f) Photos of tumor tissues harvested from the mice after
   different treatments on day 20. g) Body weight of mice in different
   groups during the experimental periods. H&E staining h), Ki67 staining
   i), and GPX4 staining j) of tumor tissues harvested from the mice after
   different treatments. Scale bar: 200 µm. Data are presented as mean ±
   SD (n = 5 mice). ^* p < 0.05, ^*** p < 0.001, and ^**** p < 0.0001,
   using two‐way ANOVA. p indicates the statistical significance relative
   to the control group.

   Then, the antitumor efficacy study was performed according to the
   procedure shown in Figure [156]5c. The tumor‐bearing mice were randomly
   divided into 6 groups (n = 5): PBS, US, PM‐FM/C, FM/C+US, PM‐FM+US,
   PM‐FM/C+US. Nanoparticles were administrated through intravenous
   injection (i.v.) every 2 days for a total of three times, and US
   irradiation was performed at 24 h post injection. As shown in
   Figure [157]5d,e, the tumor growth of the PM‐FM/C+US group was greatly
   inhibited, with the highest tumor inhibition rate of 97.3% at the end
   of day 20, which was superior to that of the PM‐FM+US group (87.9%),
   FM/C+US group (61.1%) and PM‐FM/C group (21.3%). Photo of tumors
   excised from mice in the different groups further confirmed that
   PM‐FM/C+US exhibited the most effective anti‐tumor efficacy
   (Figure [158]5f). Negligible body weight loss was observed in all
   groups of mice during the treatment period (Figure [159]5g), indicating
   the good biosafety of all treatments.

   Hematoxylin and eosin (H&E) histological staining of the tumor tissue
   sections revealed a higher percentage of necrotic cells in the
   PM‐FM/C+US group as compared to the other groups (Figure [160]5h). The
   reduced expression of Ki67 in the tumor tissue of the PM‐FM/C+US group
   indicated the effective inhibition of cell proliferation by this
   treatment (Figure [161]5i). In addition, the tumor tissue of the
   PM‐FM/C+US group exhibited the lowest levels of GPX4 expression
   (Figure [162]5j), indicating the involvement of ferroptosis during the
   treatment in vivo. Based on the above results, it can be deduced that
   the superior antitumor efficacy of PM‐FM/C upon US irradiation was
   attributed to the enhanced tumor homing due to the dual‐targeting
   modification, as well as the synergistic effect of CDT/SDT and CO
   therapy in inducing ferroptosis in tumor cells.

   At the end of treatment, the major organs were also harvested for H&E
   staining, and no noticeable pathological changes were detected (Figure
   [163]S19, Supporting Information). Moreover, the blood samples of PBS
   and PM‐FM/C+US groups were collected for biochemical analysis. The
   values of routine blood parameters and hepatic/renal function
   indicators showed no significant difference between the PBS‐treated and
   PM‐FM/C plus US‐treated mice (Figures [164]S20 and [165]S21, Supporting
   Information). Collectively, the above results demonstrated the
   excellent therapeutic efficacy and good biocompatibility of
   US‐triggered PM‐FM/C therapy as a promising therapeutic strategy for
   HCC treatment in vivo.

2.9. In Vivo Antitumor Study in an Orthotopic HCC Mouse Model

   Orthotopic tumor models are considered to be more clinically relevant
   and theoretically predictive than traditional subcutaneous tumor
   models. Based on the favorable targeting effect observed in the
   subcutaneous tumor model, we proceeded to evaluate the targeting
   capability of the nanosystem in orthotopic Hepa1‐6 tumor‐bearing mice.
   Mice were injected intravenously with PM‐FM/Cy5 and the major organs
   were harvested for imaging 12 h after injection. As shown in
   Figure [166] 6a, much stronger fluorescence was observed in the tumor
   sites compared to the surrounding liver tissues as well as other
   organs, indicating that the abundant PM‐FM/Cy5 was specifically
   accumulated in the tumor tissues. The average fluorescence signal of
   the tumor was significantly higher than that of the normal liver tissue
   and the tumor/liver signal ratio was calculated to be 6.74
   (Figure [167]6b).

Figure 6.

   Figure 6
   [168]Open in a new tab

   In vivo anti‐tumor effects of PM‐FM/C under US irradiation in an
   orthotopic HCC model. a) Representative fluorescence imaging and b)
   quantification of the mean fluorescence intensity of Cy5 in livers with
   tumor and other major organs harvested from the mice at 12 h after
   intravenous injection of PM‐FM/Cy5. c) Schematic illustration of the
   therapeutic procedure (n = 5). The Figure was created in BioRender.com.
   d) Photograph of tumor‐bearing liver tissues harvested from the mice
   after different treatments. e) Average tumor volume of mice after
   different treatments. f) Body weight of mice in different groups during
   the experimental periods. g) Representative H&E staining images of
   tumor bearing liver tissues after different treatments. h) GPX4
   staining of tumor tissues harvested from the mice after different
   treatments. i) The GSH level in the tumors. j) Biochemical analyses of
   ALT, AST and AFP of mice after different treatments. Data are presented
   as mean ± SD (n = 5 mice). ^** p < 0.01, ^*** p < 0.001, ^**** p
   < 0.0001, using two‐way ANOVA.

   After confirming the excellent targeting ability for orthotopic HCC
   tumor, we investigated the anti‐tumor effects of PM‐FM/C under US
   irradiation in the orthotopic model (Figure [169]6c). The tumor‐bearing
   mice were sacrificed and the liver tissues were collected at 14 days
   after the first injection. The digital image and tumor volume of the
   liver tissues showed that the tumors in PM‐FM/C group were quite small
   or partially disappeared, and the tumors in PM‐FM/C+US group were
   almost disappeared (Figure [170]6d,e), demonstrating that the tumor
   growth was inhibited by PM‐FM/C to some extent, and the tumor
   inhibition effect was significantly enhanced by US stimulation. Similar
   to the subcutaneous model, there was no noticeable body weight loss in
   all groups of mice (Figure [171]6f). Besides, the full scanning images
   of H&E staining of the liver tissue sections clearly displayed that the
   cancerous area in the liver of PM‐FM/C+US group was greatly reduced
   compared with PBS and US alone group (Figure [172]6g). Moreover, GPX4
   expression was significantly reduced in PM‐FM/C+US group
   (Figure [173]6h). We also detected the GSH level of the tumors from
   different groups and the results indicated that the GSH level in the
   tumors of the PM‐FM/C+US group was reduced 3.25‐fold compared with that
   of the control group (Figure [174]6i). Based on the results of in vitro
   experiments, the therapeutic effects might arise from the
   cascade‐activated catalytic reaction due to the intrinsic H[2]O[2] in
   TME for rendering CDT, hypoxia amelioration and CO release, which
   enhanced the US‐triggered SDT performance to produce synergistic
   ferroptosis for tumor death.

   Finally, H&E staining was also performed on the major organs from the
   orthotopic model, and no significant pathological changes were observed
   as well (Figure [175]S22, Supporting Information). Meanwhile, the blood
   samples were harvested for routine blood and hepatic/renal function
   evaluation. There was no significant difference in the values of
   routine blood parameters and renal function between the PBS‐treated and
   PM‐FM/C plus US‐treated mice (Figures [176]S23 and [177]S24, Supporting
   Information), indicating the good biocompatibility of the US‐triggered
   therapy. However, it was noticeable that the level of alanine
   aminotransferase (ALT), aspartate aminotransferase (AST) and
   alpha‐fetoprotein (AFP) were significantly increased in the control
   group because of the tumor burden in the liver, and these indicators
   recovered to the normal range after PM‐FM/C+US treatment
   (Figure [178]6j). These results not only demonstrated the favorable
   biosafety of the PM‐FM/C+US treatment but also further confirmed its
   efficacy in reducing tumor burden.

3. Conclusion

   In summary, a multifunctional nanoplatform based on bimetallic FM and
   CORM‐401 with SP94 peptide‐inserted macrophage membrane coating was
   rationally constructed to integrate SDT/CDT and CO gas therapy for HCC
   treatment. FM was explored as an advanced sonosensitizer with CAT
   mimicking activity to alleviate tumor hypoxia to achieve excellent SDT
   performance under US irradiation. Meanwhile, FM possessed POD‐like
   activity to generate ·OH for CDT. The SP94 peptide‐inserted macrophage
   membrane modification endowed the nanoplatform with a favorable
   targeting capability against HCC. The elevated level of intracellular
   ROS derived from SDT and CDT facilitated the cascade release of CO,
   along with GSH depletion to induce GPX4, which together led to lipid
   peroxidation and synergistic ferroptosis in HCC cells. Such a
   US‐triggered multimodal therapeutic strategy exerted excellent
   antitumor effect in vitro and significantly inhibited tumor growth in
   subcutaneous and orthotopic HCC therapy. Overall, this study
   highlighted a promising strategy to boost synergistic ferroptosis for
   effective HCC therapy.

4. Experimental Section

Materials

   Ferrous sulfate heptahydrate (Na[2]MoO[4]·2H[2]O, purity 99.0%), and
   sodium molybdate dihydrate (FeSO[4]·7H[2]O, purity 99.0%),
   3,3,5,5′‐tetramethylbenzidine (TMB), 9,10‐anthracenediyl‐bis
   (methylene)dimalonic acid (ABDA), and hemoglobin (Hb) were obtained
   from Aladdin Biochemical Technology Co., Ltd (Shanghai, China).
   CORM‐401 (purity 98.27%) purchased from MedChemExpress.
   DSPE‐PEG[2K]‐Mal (purity 95%) and Sulfo‐Cyanine5 NHS (Cy5 NHS, purity
   95%) were provided by Xi'an ruixi Biological Technology Co., Ltd
   (Xi'an, China). SP94 peptide (purity 97.69%) was obtained from GL
   Biochem (Shanghai) Ltd (Shanghai, China). RT‐PCR primes of Acsl4, Gclm,
   Slc7a11, Slc39a8, Gpx4, and Gss were provided by Beijing Tsingke
   Biotech Co., Ltd (Beijing, China).
   5,5′,6,6′‐Tetrachloro‐1,1′,3,3′‐tetraethylbenzimidazolylcarbocyanine
   iodide (JC‐1) was purchased from Shanghai Maokang Biotechnology Co.,
   Ltd (Shanghai, China). Calcein‐AM/PI cell viability/cytotoxicity assay
   kit was purchased from Beyotime Biotechnology (Shanghai, China). Cell
   Counting Kit‐8 (CCK‐8) was provided by Beijing Labgic Technology Co.,
   Ltd (Beijing, China). Liperfluo was purchased from Dojindo Laboratories
   (Kumamoto, Japan). BODIPY581/591‐C11 was provided by Thermo Fisher
   Scientific (Massachusetts, USA). 2′,7′‐Dichlorodihydrofluorescein
   diacetate (DCFH‐DA) was purchased from MedChemExpress (Shanghai,
   China). Anti‐glutathione peroxidase 4 (GPX4) antibody (ab125066) and
   Anti‐β actin antibody (ab8227) were provided by Abcam (Cambridge, UK).
   Rabbit β‐actin mAb (High Dilution) (AC026) was purchased from ABclonal
   Technology Co., Ltd (Wuhan, China). Anti‐Ki67 Rabbit pAb
   ([179]GB111141) and anti‐Glutathione Peroxidase 4 Rabbit pAb
   ([180]GB114327) were obtained from Servicebio (Wuhan, China). TRIeasy
   Total RNA Extraction Reagent (10606ES60), Hifair III 1st Strand cDNA
   Synthesis SuperMix for qPCR (gDNA digester plus) (11141ES60), and Hieff
   Unicon TaqMan multiplex qPCR master mix (11208ES03) were purchased from
   Yeasen Biotechnology (Shanghai, China). All chemicals were used as
   received without further purification.

Characterizations

   Dynamic light scattering (DLS) and zeta potential measurements were
   performed on a Malvern Zetasizer Nano ZS‐90. The N[2]
   adsorption–desorption analysis was performed using a Surface Area and
   Porosimetry Analyzer (ASAP 2460, Micromeritics Instrument Corporatio).
   The morphology of the FeMoO[4] was examined using TEM (JEOL JEM1200EX).
   The chemical composition was determined with XPS (Thermo Scientific
   K‐Alpha). FTIR spectrum was examined using the fourier transform
   infrared spectrophotometer (IRTracer‐100, SHIMADZU). UV–vis absorption
   spectra were obtained with a UV–vis spectrophotometer (UH5700,
   HITACHI). Oxygen vacancies, singlet oxygen, and electron holes were
   detected by ESR (A300‐10/12, Bruker). The loading capacity of CORM‐401
   was measured using ICP‐MS (ICPMS7800, Agilent). CLSM images were
   captured using a confocal microscope (LSM710, CARL ZEISS). Flow
   cytometry assays were performed using a flow cytometer (FACSCalibur,
   BD).

Synthesis of FeMoO[4] Nanoparticles

   Typically, Na[2]MoO[4]·2H[2]O (1.2098 g, 5 mmol) and FeSO[4]·7H[2]O
   (1.3901 mg, 5 mmol) were each dissolved in deionized water (30 mL) and
   stirred for 20 min. The FeSO[4] solution was then added dropwise to the
   Na[2]MoO[4] solution, followed by stirring for an additional 20 min.
   The resultant solution was transferred to a Teflon liner (100 mL) and
   heated at 180 °C for 12 h. The resultant FeMoO[4] was collected by
   centrifugation (8000 rpm, 10 min), washed with deionized water and
   ethanol three times, and dried in an oven (60 °C) for 6 h.

Macrophage Membrane Isolation

   Raw 264.7 cells membrane was collected according to previously reported
   procedures.^[ [181]^31 ^] First, Raw 264.7 cells on the culture dish
   were washed with PBS. The obtained cells were collected and suspended
   in the homogenization buffer before being disrupted with a probe
   ultrasonic disruptor for 10 min at 4 °C (100 W, sonicate for 3 s, and
   pause for 2 s), followed by centrifugation (20000 g, 20 min). The
   protein content of the cell membrane was measured with the BCA protein
   assay kit (Omni‐Easy Instant BCA Protein Assay Kit). The prepared cell
   membrane solutions were stored at −80 °C and used as soon as possible.

Synthesis of DSPE‐PEG[2K]‐SP94

   DSPE‐PEG[2K]‐SP94 was synthesized through a maleimide‐thiol coupling
   reaction, with a fixed molar ratio of 1:1 between SP94 and
   DSPE‐PEG[2k]‐Mal. Briefly, predetermined amounts of SP94 and
   DSPE‐PEG[2k]‐Mal (20 mg) were dissolved in dimethyl sulfoxide (DMSO) (2
   mL) and stirred for 24 h at room temperature. Then, the mixture was
   dialyzed against deionized water overnight to remove DMSO. The
   resulting aqueous solutions were lyophilized and stored at −20 °C for
   further use.

Preparation of SP94‐Modified Macrophage Membrane

   For the preparation of the SP94‐modified macrophage membrane,
   DSPE‐PEG[2k]‐SP94 was dissolved in PBS and applied to the prepared
   membrane in a 1:2 molar ratio. The mixture was stirred at 4 °C for
   30 min before being centrifuged (14000 rpm, 5 min) to remove free
   DSPE‐PEG[2k]‐SP94. After that, the SP94‐modified membrane was sonicated
   for 5 min and extruded by an Avanti mini‐extruder. The produced
   SP94‐modified membrane was stored at −80 °C for future use.

Preparation of FM/C, M‐FM/C, PM‐FM, PM‐FM/C

   FM and CORM‐401 (with a fixed molar ratio of 1:1) were mixed and
   stirred overnight at room temperature. The mixture was centrifuged at
   14000 rpm. Then, the supernatant was discarded, and the pellet was
   collected to obtain FM/C. The loading efficiency of CORM‐401 in FM/C
   was determined by ICP‐MS. For the preparation of PM‐FM/C, the
   SP94‐modified membrane was coated onto the FM/C core by 10 min
   sonication in a water bath sonicator and then another 5 min sonication
   with a probe ultrasonic disruptor (100 W, sonicate for 5 s and pause
   for 5 s). The cell membranes on the as‐prepared PM‐FM/C was validated
   by sodium dodecyl sulfate‐polyacrylamide gel electrophoresis
   (SDS‐PAGE).

   As controls, FM/C was coated with macrophage membranes without SP94
   modification to create single‐targeting nanoparticles (M‐FM/C), while
   FM was directly coated with SP94‐modified membranes to produce
   nanoparticles lacking CORM‐401 (PM‐FM). Fluorescence‐labeled
   nanosystems were prepared following a similar procedure as described
   above, except that CORM‐401 was replaced with Cy5‐NHS.

In vitro ^·OH Generation

   To monitor the generation of ·OH, FM and PM‐FM/C (at an equivalent FM
   concentration of 10 µg mL^−1) were respectively dispersed in the MES
   buffer solutions (3 mL, 0.1 mol, pH 5.4) containing TMB (0.25 mmol).
   H[2]O[2] was then added to the above solutions to a final concentration
   of 10 mmol. The absorbance spectra of TMB were measured at different
   time points using a UV–vis spectrometer.

O[2] Release

   To evaluate the oxygen release, FM was dispersed in deoxygenated water
   (10 mL) to prepare FM solutions at various concentrations (1, 2, and
   4 mg mL^−1) at room temperature. H[2]O[2] was then added to the mixture
   to give a final concentration of 10 mmol, and the dissolved oxygen was
   detected using a dissolved oxygen meter. As a control, the dissolved
   oxygen in deoxygenated water (FM 0 mg mL^−1) was measured with the
   addition of H[2]O[2].

In Vitro Detection of Singlet Oxygen (^1O[2])

   FM and PM‐FM/C (at an equivalent FM concentration of 50 µg mL^−1) were
   respectively dispersed in deionized water (3 mL) containing ABDA (20 µg
   mL^−1). The formulations were treated with US irradiation for 10 min (1
   MHz, 50% duty cycle, 1.25 W cm^−2), and then the UV–vis spectra of ABDA
   were measured at different time points.

   To confirm the production of singlet oxygen in a hypoxic environment,
   the experimental reagents were purged with nitrogen for 10 min. Then,
   the respective solutions of FM or PM‐FM/C (at an equivalent FM
   concentration of 50 µg mL^−1) were mixed with ABDA, sealed, and
   subjected to ultrasound irradiation for 10 min (1 MHz, 50% duty cycle,
   1.25 W cm^−2). The UV–vis spectra of ABDA were recorded at various time
   points.

Consumption of GSH

   The GSH solution (10 mmol, 20 µL) was incubated with a solution (3 mL)
   containing different concentrations of FM (0, 75, 150, 300, 600,
   1200 µg mL^−1) at 37 °C for 12 h. After that, the DTNB solution (2.5 mg
   mL^−1, 100 µL) was added into the mixture and culture at 37 °C for
   5 min. The remaining amount of GSH was detected by measuring the
   absorbance at 412 nm using a microplate reader.

In Vitro CO Release

   The release of CO was measured by the hemoglobin incubation method. CO
   can bind to hemoglobin, forming carboxyhemoglobin (HbCO), which can be
   detected by UV–vis spectrometer. The peak at 410 nm of the UV–vis
   absorbance spectrum was attributed to HbCO while the peak at 430 nm was
   attributed to Hb. Samples with an equivalent FM concentration of 50 µg
   mL^−1 (FM/C, M‐FM/C, and PM‐FM/C) and hemoglobin (4.2 µmol) in PBS
   solution (3 mL) were bubbled with nitrogen for 10 min, H[2]O[2] was
   then added to a final concentration of 10 mmol, and the mixture was
   immediately enclosed in a UV quartz. Meanwhile, the ultrasound group
   was treated with US irradiation (1 MHz, 50% duty cycle, 1.25 W cm^−2).
   The UV–vis spectra of CO release were respectively detected at 0, 3, 5,
   10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 min during the reaction.

Cells and Animals

   The murine macrophage cell line RAW264.7 was purchased from Chinese
   Academy of Sciences Cells Bank. The murine hepatoma cell line Hepa1‐6
   was obtained from American Type Culture Collection. All cells were
   cultivated in Dulbecco's Modified Eagle Medium (DMEM) medium
   supplemented with 1% penicillin‐streptomycin (10000 U mL^−1) and 10%
   fetal bovine serum (FBS, Cellmax) at 37 °C under a humidified
   atmosphere supplemented with 5% CO[2].

   Female C57BL/6 mice (6–8 weeks old) were obtained from SiPeiFu
   Biotechnology Co., Ltd. (Beijing, China). All the protocols were
   approved by the Committee for Experimental Animals Welfare and Ethics
   of Institute of Radiation Medicine, Chinese Academy of Medical Sciences
   (No.IRM/2‐IACUC‐2410‐003). To establish the subcutaneous tumor model,
   Hepa1‐6 cells (1 × 10^6 cells / 200 µL) suspended in PBS solution were
   subcutaneously injected into the back of mice. The orthotopic
   transplantation tumor models were established using tumor tissues
   removed from the subcutaneous model. The subcutaneous tumors were
   excised and cut into pieces, and then similar‐sized tumor pieces were
   transplanted into the liver of healthy C57BL/6 mice to establish the
   orthotopic xenograft liver tumor models.

Cellular Uptake

   The cellular internalization was evaluated by flow cytometry and CLSM.
   For flow cytometry assay, Hepa1‐6 cells were seeded in 6‐well plates (3
   × 10^5 cells per well) and cultivated for 24 h. Fresh DMEM medium
   containing FM/Cy5, M‐FM/Cy5, and PM‐FM/Cy5 at a Cy5 concentration of
   10 µg mL^−1 was added into each well, and cells were co‐incubated for
   4 h. Then, the cells were washed with PBS solution and harvested for
   flow cytometry analysis. For CLSM detection, Hepa1‐6 cells were
   incubated in confocal dishes (3×10^5 cells per dish) overnight. FM/Cy5,
   M‐FM/Cy5, and PM‐FM/Cy5 (at an equal Cy5 concentration of 10 µg mL^−1)
   were then co‐incubated with Hepa1‐6 cells for 2 h, 4 h, and 8 h,
   respectively. Cells were washed three times with PBS, fixed with
   paraformaldehyde (4%) solution, stained by
   4′,6‐diamidino‐2‐phenylindole (DAPI) (2.5 µg mL^−1), and detected by
   CLSM.

Intracellular ROS Generation

   Hepa1‐6 cells were seeded and cultivated in 6‐well plates (3 × 10^5
   cells per well) for 24 h, then co‐incubated with fresh DMEM containing
   different formulations for another 12 h, including:1) fresh DMEM; 2)
   US; 3) PM‐FM, 4) PM‐FM/C; 5) PM‐FM+US, and 6) PM‐FM/C+US. US
   irradiation (1 MHz, 50% duty cycle, 1.25 W cm^−2) was performed in the
   groups labelled with US. Then, fresh serum‐free medium containing
   DCFH‐DA (10 µmol) was added and incubated with the cells for 30 min.
   The fluorescence intensity was evaluated by CLSM and flow cytometry by
   detecting the fluorescence intensity of 2,7‐dichlorofluorescein
   diacetate (DCFH‐DA, Ex/Em = 488/525 nm).

The Intracellular CO Release

   The intracellular CO release was evaluated using a CO probe (COP). The
   probe was synthesized following a previously reported method.^[
   [182]^32 ^] As shown in the synthetic route below, CH[3]OH (10 mL) was
   used to dissolve Rhodamine B (RB) (1 mmol), and then hydrazine hydrate
   (0.5 mL) was added. The solution was reacted at 75 °C until the color
   changed from red to uncolored. The mixture was then cooled,
   concentrated under vacuum, and extracted with distilled water and ethyl
   acetate. The extract was dried over anhydrous magnesium sulfate,
   filtered, and vacuum evaporated to yield Compound 1. Subsequently,
   1 mmol of Compound 1 was diluted in ethanol (20 mL), and
   p‐dimethylaminobenzaldehyde (0.1688 g) was added gradually. The mixture
   was agitated and refluxed at 85 °C for 18 h to produce a khaki solid.
   To get COP, the substance was washed twice with PBS, followed by
   further refinement using silica column chromatography (petroleum ether:
   ethyl acetate = 10:1). Hepa1‐6 cells were then seeded in confocal
   culture dishes (3 × 10^5 cells per well) for 24 h. The cells were
   treated with fresh DMEM containing various formulations including 1)
   fresh DMEM, 2) PM‐FM, 3) PM‐FM/C, and 4) PM‐FM/C+US (1 MHz, 50% duty
   cycle, 1.25 W cm^−2) at an equivalent FM concentration of 50 µg mL^−1
   for 12 h. Then, the cells were incubated with fresh PBS solution
   containing COP probe (40 µg mL^−1) for 20 min. CLSM (LSM710, CARL
   ZEISS) was employed to analyze the cells immediately.

   graphic file with name ADVS-12-2413833-g005.jpg

   The synthetic route of COP.

Cytotoxicity Study

   Hepa1‐6 cells and 293T cells were seeded in 96‐well plates (5 × 10^3
   cells per well) and cultivated for 24 h. Subsequently, the culture
   media were substituted with fresh DMEM containing different
   formulations (PM‐FM, PM‐FM/C) at various concentrations (0, 6.25, 12.5,
   25, and 50 µg mL^−1), and further cultured for 12 h. Then, the cells in
   the US, PM‐FM+US, and PM‐FM/C+US groups were irradiated with US (1 MHz,
   50% duty cycle, 1.25 W cm^−2) for 5 min. Cell Counting Kit‐8 (CCK‐8)
   agent was used to detect the cell viability 24 h following US
   irradiation. The absorbance at 450 nm was detected using a microplate
   reader to calculate the relative cell viability.

   Cell viability was also evaluated using live/dead co‐staining. For
   this, Hepa1‐6 cells were seeded at 3 × 10^5 cells per well in 6‐well
   plates. After incubation at 37 °C for 24 h, the cells were subjected to
   different treatments: 1) fresh DMEM; 2) US; 3) PM‐FM; 4) PM‐FM/C; 5)
   PM‐FM+US; 6) PM‐FM/C+US (at an equivalent FM concentration of 50 µg
   mL^−1). For the US irradiation groups, cells were treated with US (1
   MHz, 50% duty cycle, 1.25 W cm^−2) after 12 h incubation with different
   formulations. Following an additional 8 h of incubation, cell viability
   was assessed using the Calcein/PI cell viability/cytotoxicity assay
   kit. After 30 min of incubation with the Calcein/PI buffer, the cells
   were observed under a fluorescent microscope.

Intracellular GSH Detection

   Hepa1‐6 cells were seeded at 3 × 10^5 cells per well in 6‐well plates.
   After incubation at 37 °C for 24 h, the Hepa1‐6 cells were given
   different treatments: 1) fresh DMEM; 2) fresh DMEM+US; 3) PM‐FM; 4)
   PM‐FM/C; 5) PM‐FM+US; 6) PM‐FM/C+US. For the US irradiation groups, the
   cells were treated with US (1 MHz, 50% duty cycle, 1.25 W cm^−2) after
   12 h incubation of different formulations. After another 8 h of
   incubation, the GSH concentration was detected using the Total
   Glutathione Assay Kit.

Intracellular GPX4 Expression

   The western blotting was performed to assess the intracellular
   expression of glutathione‐dependent peroxidase 4 (GPX4). Hepa1‐6 cells
   were seeded in 6 well‐plates at a density of 3 × 10^5 per well and
   cultivated for 24 h. Then, the culture medium was replaced, and cells
   were given different treatments: 1) fresh DMEM, 2) fresh DMEM+US, 3)
   PM‐FM, 4) PM‐FM/C, 5) PM‐FM+US, and 6) PM‐FM/C+US (with a relative
   concentration of FM at 50 µg mL^−1). After 12 h of incubation, the
   US‐labeled groups were irradiated with US (1 MHz, 50% duty cycle,
   1.25 W cm^−2) for 5 min, while the other groups remained untreated. The
   cells were then incubated for an additional 12 h. Following treatment,
   the cells were collected and lysed using RIPA lysis buffer on ice for
   30 min. The resulting solutions were centrifuged at 4 °C to obtain the
   extracted protein solution. The concentration of the extracted protein
   was determined using a BCA protein assay kit. Subsequently, protein (10
   µg) from each group was loaded and subjected to the Simple Western
   procedure. Signals were detected using a Tanon Gel Imager and
   quantified with ImageJ software.

Intracellular LPO Assessment

   Hepa1‐6 cells were seeded and cultivated in confocal culture dishes (3
   × 10^5 cells per well) for 24 h before being treated with various
   formulations (fresh DMEM, fresh DMEM+US, PM‐FM, PM‐FM/C, PM‐FM+US, and
   PM‐FM/C+US) at an equivalent FM concentration of 50 µg mL^−1 for 24 h.
   Then, the corresponding groups were exposed to US irradiation (1 MHz,
   50% duty cycle, 1.25 W cm^−2). After that, the cells were further
   cultured with BODIPY581/591‐C11 (5 µmol, Ex/Em = 485/520 nm) or
   Liperfluo (5 µmol, Ex/Em = 524/535 nm) for 20 min and then assessed by
   CLSM and flow cytometry.

Mitochondrial Membrane Potential Detection

   Hepa1‐6 cells were seeded in confocal culture dishes at a density of 3
   × 10^5 cells per well and cultured for 24 h. Then, the Hepa1‐6 cells
   were treated with different formulations, including fresh DMEM, fresh
   DMEM+US, PM‐FM, PM‐FM/C, PM‐FM+US, and PM‐FM/C+US, at an equivalent FM
   concentration of 50 µg mL^−1 for 24 h. After 12 h of incubation with
   various formulations, the US irradiation groups were exposed to US
   irradiation (1 MHz, 50% duty cycle, 1.25 W cm^−2). After an additional
   4 h of incubation, the cells were stained with JC‐1 probe (10 µg mL^−1)
   for 15 min, washed three times with PBS, and subsequently assessed by
   CLSM.

In Vivo Imaging

   FM/Cy5, M‐FM/Cy5, and PM‐FM/Cy5 (at a Cy5 dose of 2.5 mg kg^−1) were
   systemically administered to tumor‐bearing mice via intravenous
   injection. For the subcutaneous tumor model, the mice were anesthetized
   using isoflurane, and fluorescence images were acquired at the
   appointed time points using the IVIS Lumina II imaging system
   (Caliper). For the orthotopic tumor model, mice were sacrificed 12 h
   after injection, and the major organs (heart, liver with tumor, spleen,
   and lung) were harvested for imaging using a Maestro EX imaging system
   (CRi Maestro).

Anti‐Tumor Study in Subcutaneous Tumor Model

   When the tumor volume reached ≈70 mm^3, the mice were randomly divided
   into six groups (n = 5 mice per group): 1) PBS; 2) US, 3) PM‐FM/C; 4)
   FM/C+US; 5) PM‐FM+US; and 6) PM‐FM/C+US. Mice in the PM‐FM/C, FM/C+US,
   PM‐FM+US, and PM‐FM/C+US groups were administrated with different
   formulations at an equivalent FM concentration of 15 mg kg^−1 on day 0,
   day 2, and day 4. The ultrasound (3 MHz, 50% duty cycle, 1.25 W cm^−2)
   was applied to the mice in the US‐labelled groups on day 1, day 3 and
   day 5. The tumor volumes and body weights of all mice were recorded
   every two days during the treatment. Tumor volume was calculated by the
   formula: V = 0.5×L×W^2 (where L is the longest diameter and W is the
   shortest diameter perpendicular to L). On day 20, the mice were
   sacrificed, and major organs and tumors were harvested for further H&E
   staining and IHC staining. And the blood samples of mice in the PBS
   group and PM‐FM/C+US group were collected for analyzing various
   biochemical parameters.

Evaluation of the Anti‐Tumor Effect in Orthotopic Liver Tumor Model

   Orthotopic tumor‐bearing mice were randomly divided into four groups (n
   = 5 mice per group) and given different treatments including 1) PBS, 2)
   US, 3) PM‐FM/C, and 4) PM‐FM/C+US. Mice in the PM‐FM/C and PM‐FM/C + US
   groups received PM‐FM/C at a FM dosage of 15 mg kg^−1 on days 0, 2, and
   4. Ultrasound irradiation (3 MHz, 50% duty cycle, 1.25 W cm^−2) was
   applied to the upper abdomen of mice in the US group on days 1, 3, and
   5. Body weights were recorded every two days throughout the treatment
   period. At the conclusion of the experiment, mice were sacrificed, and
   major organs and tumors were collected for subsequent H&E staining, IHC
   staining, and immunofluorescence (IF) staining. Blood samples from the
   PBS and PM‐FM/C + US groups were collected for biochemical parameter
   analysis.

Statistical Analysis

   Statistical analysis in this study was performed using GraphPad Prism
   10.0 software. The results were presented as mean ± standard deviation
   (SD). When comparing the two groups, the Student t‐test was used. When
   comparing three or more groups, the one‐way ANOVA test was used. The
   statistical significance was represented as ^* p < 0.05, ^** p < 0.01,
   ^*** p < 0.001, and ^**** p < 0.0001.

Conflict of Interest

   The authors declare no conflict of interest.

Supporting information

   Supporting Information
   [183]ADVS-12-2413833-s001.docx^ (5.9MB, docx)

Acknowledgements