Abstract

   graphic file with name au3c00156_0010.jpg

   Sonodynamic therapy (SDT) holds great promise to be applied for cancer
   therapy in clinical settings. However, its poor therapeutic efficacy
   has limited its applications owing to the apoptosis-resistant mechanism
   of cancer cells. Moreover, the hypoxic and immunosuppressive tumor
   microenvironment (TME) also weakens the efficacy of immunotherapy in
   solid tumors. Therefore, reversing TME remains a formidable challenge.
   To circumvent these critical issues, we developed an
   ultrasound-augmented strategy to regulate the TME by utilizing an
   HMME-based liposomal nanosystem (HB liposomes), which can
   synergistically promote the induction of
   ferroptosis/apoptosis/immunogenic cell death (ICD) and initiate the
   reprograming of TME. The RNA sequencing analysis demonstrated that
   apoptosis, hypoxia factors, and redox-related pathways were modulated
   during the treatment with HB liposomes under ultrasound irradiation.
   The in vivo photoacoustic imaging experiment showed that HB liposomes
   enhanced oxygen production in the TME, alleviated TME hypoxia, and
   helped to overcome the hypoxia of the solid tumors, consequently
   improving the SDT efficiency. More importantly, HB liposomes
   extensively induced ICD, resulting in enhanced T-cell recruitment and
   infiltration, which normalizes the immunosuppressive TME and
   facilitates antitumor immune responses. Meanwhile, the HB liposomal SDT
   system combined with PD1 immune checkpoint inhibitor achieves superior
   synergistic cancer inhibition. Both in vitro and in vivo results
   indicate that the HB liposomes act as a sonodynamic immune adjuvant
   that is able to induce ferroptosis/apoptosis/ICD via generated
   lipid-reactive oxide species during the SDT and reprogram TME due to
   ICD induction. This sonodynamic nanosystem integrating oxygen supply,
   reactive oxygen species generation, and induction of
   ferroptosis/apoptosis/ICD is an excellent strategy for effective TME
   modulation and efficient tumor therapy.

   Keywords: sonodynamic therapy, immunogenic cell death, ferroptosis,
   cancer therapy

1. Introduction

   Tumor tissues are highly heterogeneous, and tumor cells exist in a
   complicated network of tumor microenvironment (TME) composed of various
   immune cells and stromal cells such as endothelial cells, pericytes,
   fibroblasts, macrophages, T cells, etc.^[50]1−[51]5 Immunosuppression
   and hypoxia are the key features of the TME and are also crucial
   factors for the efficacy of cancer therapy and tumor
   metastasis.^[52]6−[53]8 Immunotherapy with immune checkpoint inhibitors
   (ICIs) has been developed as an effective method to treat cancers,
   which obtains encouraging therapeutic outcomes and improves patients’
   survival in certain types of cancers.^[54]9 However, the overall
   response rate for patients receiving ICI treatment ranged from 15 to
   60% in many solid tumor types.^[55]10 Thus, how to enhance the overall
   response rate of ICIs in the tumors with tolerance to ICIs is an
   important issue for cancer immunotherapy and clinical practice.
   Increasing evidence indicates that tumor immune microenvironment and
   inherent antitumor immunity are crucial for ICIs’ efficacy.^[56]11

   Recently, it has been found that immunogenic cell death (ICD) induced
   primarily in cancer cells with various therapeutic modalities provides
   a good strategy to initiate systemic antitumor immunity and improve the
   efficacy of ICIs for immunotherapy.^[57]12−[58]14 ICD enhances
   antitumor immune responses by eliciting a range of tumor antigens and
   damage-associated molecular patterns (DAMPs), such as the exposure of
   calreticulin (CRT) onto the surface of cancer cells and the release of
   high-mobility group box 1 (HMGB1) and adenosine triphosphate (ATP) from
   dying cancer cells. DAMPs could activate the intrinsic antitumor immune
   system through the induction of dendritic cell (DC) maturation,
   recruitment of phagocytes, enhanced infiltration of cytotoxic T cells,
   and proinflammatory cytokine releases.^[59]15,[60]16 To date, the
   majority of ICD inductions have primarily resulted from apoptotic cell
   death initiated by various therapeutic modalities, such as
   chemotherapy, photothermal therapy, photodynamic therapy, and
   sonodynamic therapy (SDT); however, tumor resistance to the apoptotic
   pathway considerably impeded the efficient induction of ICD during
   cancer treatment and subsequent activation of adaptive immunity.^[61]17
   As apoptosis is the most common pathway for regulated tumor-cell death,
   exploring novel non-apoptotic cell death pathways with high
   immunogenicity is crucial to alter tumor immune microenvironment and
   facilitate antitumor function of ICIs. Ferroptosis is a novel type of
   non-apoptotic regulated cell death characterized by the depletion of
   glutathione, accumulation of lipid peroxides (LPO), and further
   induction of oxidative stress. Furthermore, it has been proven that
   DAMPs could be elicited in cancer cells during ferroptosis to
   participate in the modulation of tumor immune
   microenvironment.^[62]18−[63]21 Although there are conflicting findings
   about the relationship between ferroptosis and antitumor immunity,
   which might be dependent on tumor types or classes of ferroptosis
   inducers,^[64]22,[65]23 multiple lines of evidence were shown that
   cancer cell death resulting from ferroptosis, especially early
   ferroptosis, could enhance antitumor immunity and efficacy of
   immunotherapy.^[66]24,[67]25

   Ultrasound (US)-mediated cancer therapy is a localized therapeutic
   modality for tumor treatment with good penetration of tissue and the
   ability to effective ICD induction to further activate tumor-specific
   immune responses.^[68]26 SDT, a type of reactive oxygen species
   (ROS)-mediated cancer therapy, utilizes harmless US to initiate the
   generation of ROS to kill tumor cells.^[69]27−[70]29 The ROS, in turn,
   could also trigger endoplasmic reticulum (ER) stress that elicits the
   production of DAMP signals and further leads to ICD.^[71]30,[72]31
   Therefore, SDT is another therapeutic modality for efficient ICD
   induction through ROS-initiated ER stress and could also potentially
   stimulate antitumor immunity.

   Compared with traditional photodynamic therapy, SDT exhibits a higher
   tissue penetration, rendering it a promising and potent therapeutic
   modality for cancer treatment. However, since SDT requires more oxygen
   to generate ROS, rapid oxygen consumption can aggravate the hypoxia
   state inside the tumors, which might dampen the therapeutic effect of
   SDT. Thus, enhancing supplement of oxygen into tumor mass is a
   reasonable strategy to improve the therapeutic efficacy of SDT.
   Currently, various strategies have been designed to alleviate tumor
   hypoxia, such as introduction of microbubbles containing oxygen, in
   situ O[2] generation by metals, or catalase reacting with intratumoral
   H[2]O[2] and hemoglobin (Hb).^[73]32−[74]35

   Notably, cancer immunotherapies that reverse tumor immunosuppressive
   microenvironment or stimulate the host immune system are changing the
   therapeutic landscape for multiple malignancies. In terms of immune
   response modulation under SDT, it has been reported that SDT could
   improve the efficacy of immunotherapy through induction of ROS or rapid
   consumption of glutathione.^[75]36−[76]38

   In the present study, we aimed to establish a combined therapeutic
   strategy integrated SDT with ferroptosis inducer and oxygen supplement
   to induce ICD and modulate the tumor immune microenvironment
   ([77]Scheme [78]1
   ). This was achieved using BSO, a constructed GSH synthesis inhibitor,
   Hb, and sonosensitizers (HMME) co-loaded into nanoliposomes as the
   nanosonosensitizers (termed HB liposomes). After intravenous
   administration, HB liposomes accumulated at the tumor site due to the
   EPR effect and could exert a sonodynamic effect on primary tumors to
   produce ROS under the treatment of US. The Hb is then able to provide
   oxygen at the hypoxic site of tumors, which enhances tumor oxygenation
   and improves the effectiveness of SDT. In addition, the triggered ER
   stress by ROS can activate ICD and result in the release of DAMPs to
   further promote the maturation of DC, increase the infiltration of the
   cytotoxic T lymphocytes, and remodel the TME. Similarly, enhanced tumor
   cell death by ferroptosis-inducer BSO also synergistically induces the
   release of DAMPs and further regulates the TME. In addition, this
   nanotherapeutic system combined with PD-1 checkpoint blockade achieves
   a superior synergistic anticancer effect. Therefore, the objective of
   this study is to use SDT combined with the ferroptosis inducer and
   supply of oxygen to amplify the ICD, eventually enhancing tumor
   suppression and modulating the TME. The strategy proposed in this work
   provides a novel and alternative approach for cancer therapy through
   amplification of tumor cell ICD and enhancement of antitumor immunity.

Scheme 1. Antitumor Diagram Involved in This Work through Augmented
Immunogenic Cell Death and Antitumor Immunity Mediated by Sonodynamic Therapy
and Ferroptosis Inducer.

   [79]Scheme 1
   [80]Open in a new tab

2. Results and Discussion

   Liposomes were used as nanoplatforms to co-encapsulate three
   hydrophobic molecules: the sonosensitizer HMME, glutathione synthesis
   inhibitor l-buthionine-sulfoximine (BSO), and oxygen supplier Hb, via
   the typical reverse evaporation method (designated as HB liposomes).
   The preparation process for HB liposomes is illustrated in [81]Figure
   [82]1
   a. TEM imaging and dynamic light scattering indicated that the average
   hydrodynamic size of HB liposomes was ∼168 nm ([83]Figure [84]1b), and
   the zeta potential analysis showed high negative zeta potential (−36
   mV), which is conducive to maintain stability of nanoparticles within
   the bloodstream ([85]Figure [86]1c). As demonstrated in [87]Figure S1,
   the HB liposomes exhibited high dispersity in various solvents,
   including H[2]O, PBS, and RPMI 1640 with 10% fetal bovine serum.
   Overall, the nanoscale size and negative zeta potential facilitate the
   HB liposomes to penetrate into tumor tissues via the EPR effect.

Figure 1.

   [88]Figure 1
   [89]Open in a new tab

   (a) Schematic illustration for the synthetic procedure of HB liposomes.
   (b) Size distribution of HB liposomes analyzed by dynamic light
   scattering and TEM images. (c) Zeta potential analysis of HB liposomes.
   (d) Hemolysis assay of HB liposomes. (e) Normalized absorption spectra
   and fluorescence spectra of HB liposomes. (f) Time-dependent DPBF
   absorption spectra in the presence of HB liposomes under US irradiation
   for different durations (g) ESR spectra for ^1O[2] generation under
   various treatments (PBS, US, HB liposomes, and HB liposomes plus US)
   with TEMP as a trapping agent.

   Next, the hemolysis assay was conducted under various concentrations of
   HB liposomes. [90]Figure [91]1
   d reveals that no obvious hemolysis phenomenon was observed, which
   implicates that the HB liposomes exhibit excellent biosafety and
   biocompatibility. As indicated in [92]Figures [93]1e and [94]S2, UV–vis
   spectrum analysis showed that HB liposomes exhibited four
   characteristic absorption peaks between 450 and 650 nm, which is
   similar to free HMME, indicating that HB liposomes still retain their
   the nanosonosensitizer ability of HMME. The loading efficiency of HMME,
   which is essential for effective anticancer therapy, was determined to
   be 89% as shown in [95]Figure S3. Nanosonosensitizer-assisted SDT
   typically induces the production of ROS, such as singlet oxygen
   (^1O[2]), which kills cancer cells. To monitor the ROS during the SDT
   process, the ROS probe 1,3-diphenylisobenzofuran (DPBF) and electron
   spin resonance (ESR) were employed to analyze ROS generation. As
   indicated in [96]Figure [97]1f, the characteristic absorption peak at
   400 nm significantly decreased with increasing US irradiation time,
   which suggests that extensive ROS was produced in the solution. In
   addition, we observed obvious and strong ESR characteristic peaks of
   ^1O[2] species for HB liposomes upon US irradiation, whereas no evident
   peak was observed in other groups ([98]Figure [99]1g). Taken together,
   the results demonstrate that the constructed nanocomposites display
   expected characteristics such as excellent biosafety, great capability
   of singlet oxygen production, and good nanosonosenitizer property.

   To further evaluate internalization of HB liposomes by cancer cells,
   the intracellular green fluorescence by CLSM images suggested efficient
   internalization of HB liposomes by Colon 26 cells after 24 h incubation
   ([100]Figure [101]2
   a). Next, we investigated the cytotoxic effect of SDT on Colon 26 cells
   by the MTT assay. As shown in [102]Figure [103]2b, treatment with HB
   liposomes plus US could significantly inhibit the cell viability in a
   dose-dependent manner. Moreover, calcein acetoxymethyl ester
   (calcein-AM) and propidium iodide (PI) co-staining assays were further
   conducted to confirm the cytotoxic activity of HB liposomes plus US.
   The fluorescence images revealed no distinct differences between the
   blank and US only groups, in which most of the cells showed green
   colorization of calcein-AM, implicating intact live cells in these two
   groups. However, the treatment with HB liposomes plus US resulted in
   obvious red staining of PI in most of the Colon 26 cells, implying cell
   death ([104]Figure [105]2c). HB liposomes alone also slightly induced
   cell death with PI staining, which might be ascribed to the ferroptosis
   inducer BSO ([106]Figure [107]2c). Subsequently, ROS generation was
   analyzed with the DCFH-DA probe. As shown in [108]Figure [109]2c, HB
   liposomes plus US markedly promoted the production of ROS, but there
   was no obvious ROS signal in other groups. Since ROS could induce cell
   apoptosis, flow cytometry for apoptosis detection with Annexin V/PI
   staining was carried out after various treatments as shown in
   [110]Figure [111]2d. The data disclosed that HB liposomes plus US
   remarkably increased early (29.6%) and late (56.2%) apoptosis when
   compared with other groups. US alone did not result in obvious cell
   apoptosis, whereas HB liposomes alone also significantly induced early
   (16.3%) and late (38.9%) apoptosis possibly owing to the increase of
   the intracellular ROS level that resulted from the GSH depletion.

Figure 2.

   [112]Figure 2
   [113]Open in a new tab

   (a) CLSM images of internalization of HB liposomes in Colon 26 cells.
   (b) Cell viability of Colon 26 cells under various treatments with or
   without US (3.0 MHz; 1.5 W cm^–2; 50% duty cycle) for 1 min. (c)
   Live/dead cell staining (upper) and the intracellular ROS level (lower)
   of Colon 26 cells treated with HB liposomes under various conditions.
   (d) Cell apoptosis assessment with Annexin V/PI staining by flow
   cytometry after various treatments.

   To evaluate the ICD on tumor cells after SDT, we first examined the
   exposure of CRT/ERp57 and the release of HMGB1, all of which are
   well-known indicators of ICD. As shown in [114]Figure [115]3
   a, treatment with HB liposomes plus US could induce more surface
   exposure of CRT and ERp57 in Colon 26 cells compared with that of HB
   liposomes alone. No obvious surface staining of CRT and ERp57 was
   identified in the control groups of US only and blank. In addition, the
   double treatment (HB liposomes plus US) caused increased release of
   HMGB1 from Colon 26 cancer cells when compared with other three control
   groups ([116]Figure [117]3c). Thus, in vitro data suggested that SDT
   based on HMME in HB liposomes could enhance ICD induction in cancer
   cells and potentially could initiate stronger antitumor responses in
   vivo. Given that ROS is associated with the activation of stress
   responses, we thus investigated whether HB liposomes plus US could
   induce ER stress in Colon 26 cells. As shown in [118]Figure [119]3d,
   phosphorylation of eukaryotic initiation factor 2α (eIF2α), a
   characteristic indicator of the ER stress response, was significantly
   increased after the double treatment and HB-liposome treatment,
   indicating that both treatments could induce ER stress and HB liposomes
   alone might also produce some ROS due to BSO’s ability to inhibit the
   synthesis of GSH. Upon initiation of ER stress, transcription factor-4
   (ATF4) is activated and upregulated by the excess ROS through the
   PERK/eIF2α/ATF4 pathway. Our data revealed that ATF4 was increased
   after the double treatment of HB liposomes plus US ([120]Figure
   [121]3d). In addition, ATF6, an ER transmembrane transcription factor,
   is activated by the ER stress/unfolded protein response, where 90 kDa
   ATF6 is converted to a 50 kDa protein.^[122]39 Indeed, the double
   treatment resulted in the obvious reduction of 90 kDa ATF6, implying
   strong ER stress in Colon 26 cells. Since BSO is an inhibitor of GSH
   synthesis and abundant ROS could accelerate the depletion of GSH, the
   reduction in intracellular GSH content was further assessed by the
   reduced GSH assay kit. Notably ([123]Figure S4). HB liposomes alone and
   the double treatment (HB liposomes plus US) could obviously lower GSH
   content compared with that of blank and US alone after incubation for
   25 min. The depletion of GSH and excessive ROS were reported to lead to
   glutathione peroxidase 4 (GPX4) inactivation and thus could cause
   subsequent accumulation of lipid peroxidation and ferroptosis. In the
   presence of HB liposomes plus US, intracellular GPX4 expression was
   remarkably decreased ([124]Figure [125]3e), implying that the HB
   liposomes plus US overwhelmed the antioxidative defense system in the
   cancer cells. However, HB liposomes alone did not obviously affect the
   expression of GPX4, indicating that BSO alone in HB liposomes is not
   enough to impact GPX4 and combinational therapy is necessary to
   maximize the ferroptosis induction. Meanwhile, another two ferroptosis
   markers ferritin light chain (FTL), and heme oxygenase-1 (HMOX1) were
   also evaluated after various treatments. FTL as a component of ferritin
   for iron ions storage was significantly decreased after the double
   treatment; in contrast, HMOX1, catalyzing the degradation of HMME to
   increase iron ions, was markedly increased after double treatment, and
   HB liposomes alone also slightly induced the increase of HMOX1. The
   expression changes of both proteins indicate that ferroptosis is
   induced and augmented after treatment with the constructed
   nanocomposite plus US compared with that of HB liposomes alone which
   contain ferroptosis inducer BSO but do not receive US irradiation
   ([126]Figure [127]3e).

Figure 3.

   [128]Figure 3
   [129]Open in a new tab

   (a) Immunofluorescence staining for cell surface CRT and ERp57 after
   various treatments. (b) Cell viability of Colon 26 cells under various
   treatments. (c) Release of HMGB-1 from Colon 26 cells after different
   treatments (n = 3). (d) Western blotting analysis of the ER stress
   pathway in Colon 26 cells after various treatments [(1) Blank; (2) US;
   (3) HB liposomes; (4) HB liposomes + US]. (e) Western blotting analysis
   of the ferroptosis pathway in Colon 26 cells after various treatments
   [(1) Blank; (2) US; (3) HB liposomes; (4) HB liposomes + US]. (f)
   Relative 4-HNE content from Colon 26 cells after various treatments (n
   = 3). (g) Relative MDA content from Colon 26 cells after various
   treatments (n = 3). (h) Bio-TEM images of glutaraldehyde-fixed Colon 26
   cells pretreated with HB liposomes plus US. ***P < 0.001. (h) Schematic
   illustration of the proposed molecular mechanism for Colon 26 cells
   based on ferroptosis/apoptosis induced by SDT.

   Next, malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE) were measured
   after various treatments. Both molecules are reactive aldehydes and
   increased during ferroptosis. They could also reflect the level of
   intracellular lipid peroxidation. As shown in [130]Figure [131]3
   f,g, HB liposomes and HB liposomes plus US treatment can significantly
   elevate the levels of MDA and 4-HNE, and the double treatment could
   maximize the increase for both molecules. Morphological examination of
   mitochondria in Colon 26 cells treated with HB liposomes plus US
   exhibited prototypical ferroptotic characteristics, such as impaired
   mitochondrial outer membrane, reduced mitochondrial volume, and lost
   mitochondrial ridge ([132]Figure [133]3h). As shown in [134]Figure
   [135]3b, the presence of ferroptosis inhibitor Fer-1 can obviously
   reverse the viability of cells treated with HB liposomes or HB
   liposomes plus US in both groups. In addition, the addition of
   apoptosis inhibitor VAD can also partially reverse the viability of
   cells treated with HB liposomes or HB liposomes plus US. These results
   demonstrate the participation of both apoptosis and ferroptosis in HB
   liposome-mediated therapy. In summary, the potential antitumor
   mechanism of HB liposomes is shown in [136]Figure [137]3i. When exposed
   to US, the HB liposomes containing the sonosensitizer HMME can generate
   ROS to trigger ER stress that elicits the DAMP signal release and
   further induces ICD. In addition, ROS could cause irreversible damages
   to proteins and other macromolecules to induce apoptosis. However, the
   SDT-triggered cellular ROS cytotoxicity may be attenuated by the
   intracellular antioxidant system. BSO as a ferroptosis inducer can
   block the synthesis of superoxide scavenger GSH. Thus, BSO in HB
   liposomes could facilitate to maintain the ROS level inside the tumor
   mass and further stabilize the cytotoxic activity of ROS. Notably, ROS
   induced from treatment with HB liposomes plus US could result in ER
   stress and mitochondrial dysfunction in tumor cells, which subsequently
   cause tumor cell apoptosis, ICD, and the release of DAMPs.

   To widely explore the whole landscape of gene expression pattern in
   Colon 26 cells after treatment with the sonodynamic nanosystem in
   cancer cells, RNAseq analysis was conducted to analyze the differential
   gene expression after treatment. There were 1503 differentially
   expressed genes between the double treatment group (HB liposomes plus
   US) and the blank group (FC ≥ 1.5, p < 0.05), of which 745 were
   upregulated and 758 were downregulated ([138]Figure [139]4
   a,b). KEGG pathway analysis revealed that signaling pathways for
   PI3K–Akt and HIF-1, apoptosis and cell cycle were enriched after the
   double treatment, all of which were highly associated with the observed
   cell death ([140]Figure [141]4c). The analysis of GO biological process
   indicated that a series of important biological processes correlated to
   cell growth inhibition and oxidative stress were enriched after the
   double treatment such as the mitotic cell cycle process, regulation of
   innate immune response, and cell growth, response to oxidative stress
   and ROS ([142]Figure [143]4d). All of these processes are highly
   connected to the ingredients of nanocomposite and ultrasonic
   irradiation. Indeed, strong interactions between genes involved in
   apoptosis, hypoxia factors, and response to ROS were observed
   ([144]Figure [145]4e). We also noticed that autophagy was enriched in
   analysis of the KEGG pathway and GO analysis. It has been demonstrated
   that autophagy is directly correlated to ferroptosis.^[146]40 Since we
   demonstrate that ferroptosis is induced after the double treatment, we
   further specifically analyzed the genes driving ferroptosis from
   RNA-seq data. As a result, a panel of ferroptosis driver genes showed
   an upward trend in overall expression after the double treatment
   including TFRC, ATG4D, GLS2, ATG5, ASCL4, LPCAT3, BECN1, BID, and so
   on, which indicated that ferroptosis was highly involved in the process
   of tumor growth inhibition ([147]Figure [148]4f). Collectively, the
   RNA-seq data disclose that HB liposomes plus ultrasonic irradiation
   could inhibit cell growth through ROS generation, apoptosis, and
   ferroptosis.

Figure 4.

   [149]Figure 4
   [150]Open in a new tab

   (a) Heat map showing the differentially expressed genes in Colon 26
   cells after treatments. (b) Volcano plot indicating the upregulated and
   downregulated genes after the double treatment. (c) KEGG pathway
   enrichment analysis after treatments. (d) GO enrichment analysis for
   the biological processes differentially regulated after treatments. (e)
   Protein–protein interaction network of apoptosis, the HIF-1 signaling
   pathway, and response to ROS in double treatment group. (f) Driver
   gene-involved ferroptosis induction after the double treatment.

   Given its excellent in vitro therapeutic performance, HB liposomes were
   further evaluated for their in vivo anticancer activity ([151]Figure
   [152]5
   a). First, tumor accumulation and biodistribution of HB liposomes were
   evaluated by in vivo fluorescence imaging. Cy5-modified HB liposomes
   were administered intravenously to Colon 26 tumor-bearing mice and the
   fluorescence signals from various time points were recorded. As shown
   in [153]Figure [154]5b,c, the HB liposomes showed the strongest tumor
   accumulation at 12 h post-injection; and the fluorescence signal
   gradually faded out at 24 h, indicating that HB liposomes is suitable
   for local ultrasonic irradiation in tumor sites. For detailed tissue
   distribution of HB liposomes, the mice were sacrificed to harvest major
   organs and tumor tissues 24 h after liposome injection for ex vivo
   fluorescence imaging, showing a higher tumor accumulation of HB
   liposomes compared with other organs except livers which are
   responsible for liposome clearance ([155]Figure [156]5d).

Figure 5.

   [157]Figure 5
   [158]Open in a new tab

   Antitumor efficacy of HB liposomes plus US in tumor-bearing mice. (a)
   Schematic illustration for the therapeutic schedule. (b) In vivo
   imaging of Cy5-labeled HB liposomes in Colon 26 tumor-bearing mice at
   different time points (n = 4). (c) Quantitative analysis of the
   fluorescence intensity at different time points. (d) Biodistribution of
   Cy5-modified HB liposomes in major organs and tumor excised from the
   mice. (e) Tumor growth curves after various treatments (n = 5). (f) PAI
   of tumors at different time points after the injection of HB liposomes.
   (g) Hypoxia evaluation by the immunofluorescent staining after
   hypoxyprobe-1 treatment. (h) Quantitative analysis of the hypoxia level
   after treatment with hypoxyprobe-1.*P < 0.05, **P < 0.05, and ***P <
   0.001.

   Meanwhile, O[2] release by Hb at hypoxic tumor sites may contribute to
   improved tumor oxygenation and subsequently enhance SDT efficacy. To
   verify this, we monitored the tumor oxygenation by photoacoustic
   imaging (PAI) of Hb and oxyhemoglobin. As shown in [159]Figure [160]5
   f, PAI disclosed that HB liposomes significantly improved the oxygen
   concentration (sO[2]) in tumor tissues over time, reaching a peak
   within 12 h and subsequently decreasing, which seems to be in agreement
   with the timeline of tumor imaging after Cy5-modified HB liposomes.
   Therefore, liposomes encapsulated with Hb are beneficial to improve
   intratumoral oxygenation and enhance the efficacy of SDT. Meanwhile, as
   shown in [161]Figure [162]5g,h, we further analyzed the extent of
   hypoxia in tumors by pimonidazole hydrochloride (hypoxyprobe-1), to
   determine whether HB liposomes treatment could improve the hypoxia of
   tumor tissues. The data indicated that the untreated control tumors
   exhibited strong hypoxia signals while the hypoxia level was
   significantly lowered in HB liposomes-treated tumors suggesting that HB
   liposomes could relieve hypoxia in the solid tumors and could
   reoxygenate the hypoxic tumor environment.

   Subsequently, in vivo antitumor study was performed with Colon 26
   tumor-bearing mice. Mice were randomly assigned to four groups: Blank
   (PBS) group, US irradiation alone group (3.0 MHz; 1.5 W cm^–2; 50% duty
   cycle; 5 min), HB liposomes alone group, and the double treatment group
   (HB liposomes plus US irradiation). As shown in [163]Figure [164]5
   e and [165]Figure S5, the tumors from two control groups (blank and US
   irradiation alone) grew rapidly over the treatment period. The tumors
   receiving the treatments with HB liposomes alone and the double
   treatment (HB liposome plus US irradiation) were suppressed
   significantly when compared with blank and US alone. The double
   treatment with HB liposomes plus US irradiation showed maximal tumor
   growth inhibition across four groups. These results confirmed that HB
   liposomes loaded with HMME could produce potent antitumor effects upon
   US irradiation. Moreover, only minor insignificant fluctuations in mice
   body weight were observed across all groups during 11 days after
   liposomes administration, indicating that no severe acute toxicity was
   generated for HB liposomes and US irradiation in tumor-bearing mice
   ([166]Figure S6). In addition, blood parameter analysis and
   histological analysis of the major organs were performed to investigate
   the biosafety of HB liposomes and US. As shown in [167]Figures S7 and
   S8, no systemic toxicity was observed according to various blood
   parameters. Hematoxylin and eosin (H&E) staining for the major organs
   with various treatments did not suggest detectable inflammation or
   damage, implying that no obvious in vivo toxicities were caused by SDT
   ([168]Figure S9). These results further demonstrate that the treatment
   with HB liposomes plus US possesses good biocompatibility and holds
   great promise for clinical translation.

   To assess the effects of HB liposomes plus US on tumor apoptosis,
   proliferation, and ferroptosis, tumor sections in each group were first
   subjected to H&E staining, Ki67, and apoptotic analysis with TUNEL
   staining. As shown in [169]Figure [170]6
   a, H&E staining revealed massive tumor cell death after the double
   treatment, and HB liposomes alone also resulted in significant cell
   death; no obvious cell death was observed in other control groups.
   TUNEL assays indicated similar results with H&E staining, showing that
   the HB liposomes plus US induced extensive cell apoptosis, and
   apoptotic cells were slightly increased in the group of HB liposomes
   alone ([171]Figure [172]6a). In contrast with apoptosis, the Ki67 level
   more significantly decreased in the groups of HB liposomes alone and HB
   liposomes plus US than that of other two control groups, and the double
   treatment showed minimal level of Ki67-positive staining across all
   groups. GPX4 as an indicator was significantly reduced in two groups
   containing HB liposomes, and the double treatment further suppressed
   the expression of GPX4. Since ICD could be elicited by biological
   processes related to apoptosis and ferroptosis, the exposure of CRT
   protein as a marked event for ICD was assessed by immunostaining. The
   data suggested that ICD was induced by HB liposomes alone and the
   double treatment, but the latter resulted in more exposure of CRT onto
   the cancer cell surface ([173]Figure [174]6a).

Figure 6.

   [175]Figure 6
   [176]Open in a new tab

   (a) H&E staining and immunohistochemical staining for the TUNEL assay,
   CRT exposure, and the expression Ki67 and GPX4. Scale bars: 50 μm. (b)
   Flow cytometric and (c) quantitative analyses of mature DCs
   (CD45^+CD80^+ CD11c^+) in tumors after various treatments. (d) Flow
   cytometric and (e) quantitative analyses of CD8^+ T cells
   (CD45^+CD3^+CD8^+) in tumors after various treatments (n = 3). ***P <
   0.001.

   Since ICD in tumor tissues is highly associated with the modulation of
   immune microenvironment, we next evaluated whether SDT based on HB
   liposomes could enhance antitumor immunity. We analyzed the immune
   cells in tumor tissues after various treatments. As indicated in
   [177]Figure [178]6
   b,c, mature DCs (CD11c^+CD80^+) in tumor tissues were significantly
   increased after treatment with HB liposomes plus US (25%) when compared
   with that of HB liposomes alone (21.2%), US alone (14.2%), and blank
   (11.7%). DCs as a kind of antigen-presenting cells could process tumor
   antigens and present the processed antigens to T cells.^[179]15 It has
   been demonstrated that molecules of DAMPs released or exposed from
   immunogenic dead cells (CRT or ATP) could induce DC maturation and
   further activate T cells inside tumor tissues, thereby enhancing
   antitumor immunity. Thus, infiltrated T cells were further assessed by
   flow cytometry. The data suggested that more CD8^+ cytotoxic T cells
   (32.9%) were found in tumors that received the double treatment
   compared with HB liposomes alone (24.3%), US alone (12.3%), and blank
   (10.8%) ([180]Figure [181]6d,e). We did not observe significant changes
   in other immune cells such as CD4 T cell, macrophages, and NK cells
   (data not shown). Since the double treatment of HB liposomes plus US
   could induce most mature DCs and infiltrated CD8^+ T cells in tumor
   mass, this nanosystem integrating nanoparticles, ferroptosis inducer
   (BSO), oxygen supplier (Hb), and sonosensitizer (HMME) could not only
   result in apoptosis/ferroptosis of tumor cells but also modulate tumor
   immune microenvironment after being administered into tumor-bearing
   mice and irradiated with local ultrasonic treatment.

   Immunotherapy with ICIs such as PD-1 antibody and CTLA4 antibody has
   revolutionized the cancer therapeutics and significantly prolonged the
   survival rate in patients with certain cancers.^[182]41 Among the
   parameters affecting ICI response, tumor-infiltrating immune cells,
   especially cytotoxic T cells (CD8^+), are one of the determinants for
   ICI efficiency.^[183]42 Encouraged by the increased CD8^+ T cell
   infiltration after the treatment with HB liposomes plus US ([184]Figure
   [185]6
   d), we next evaluated the antitumor efficiency of PD-1 antibody in HB
   liposomes/US-treated tumor-bearing mice ([186]Figure [187]7a). Colon 26
   tumor-bearing mice were assigned to four groups for various treatments
   as illustrated schedule in [188]Figure [189]7a (1, Blank; 2,PD-1 alone;
   3, HB liposomes plus US; and 4, HB liposomes/US plus PD-1). As a
   result, two doses of PD-1 injection (250 μg/mouse) just showed a weak
   antitumor effect ([190]Figure [191]7b,e). By contrast, double
   treatments with HB liposomes plus US obtained a potent antitumor
   effect, which is consistent with previous study in [192]Figure [193]5e.
   Notably, triple treatment with HB liposomes/US plus PD-1 antibody
   inhibited tumor growth more significantly than double treatments
   ([194]Figure [195]7b,e), indicating that disruption of the interaction
   between PD-1/PD-L1 axis markedly enhances T cell antitumor function,
   which corresponds with the increase CD8^+ T cell infiltration after the
   treatment of HB liposomes plus US ([196]Figure [197]6d). Meanwhile,
   only minor insignificant fluctuations in body weight were observed
   across all groups ([198]Figure [199]7c), implicating the good biosafety
   for this therapeutic regimen. In addition, triple treatments with HB
   liposomes/US plus PD-1 most significantly prolonged the survival of
   tumor-bearing mice when compared with other three control groups
   although the double treatment already extended the mice survival
   ([200]Figure [201]7d). Together, these results further demonstrate that
   HB liposomes plus US could not only directly suppress tumor growth but
   also spontaneously induce CD8^+ T cell infiltration into tumor sites,
   which creates a precondition for combination therapy with
   immunotherapy. Indeed, immune checkpoint blockade PD-1 antibody could
   further synergistically delay tumor growth.

Figure 7.

   [202]Figure 7
   [203]Open in a new tab

   (a) Schematic therapeutic schedule of the HB liposome sonodynamic
   system combined with PD-1 antibody. (b) Individual tumor growth curves
   for each mouse in four groups (n = 5). (c) Body weight during the whole
   treatment and (d) survival curves for tumor-bearing mice after
   different treatments. (e) Tumor growth curves in four groups with
   different treatments (n = 5).

3. Conclusions

   In summary, a distinct liposomal nanocomposite was successfully
   developed for enhanced SDT. In the constructed nanosystem, HB liposomes
   loaded with an inhibitor of GSH synthesis BSO led to the depletion of
   GSH and the increased production of lipid peroxidation, which
   synergistically trigger the ferroptosis of tumor cells. Furthermore,
   HMME entrapped into HB liposomes renders SDT feasible, causing the
   production of ROS which, in turn, augments apoptosis and ferroptosis of
   tumor cells. On the other hand, the Hb proteins encapsulated into HB
   liposomes could provide more oxygen to increase the efficacy of SDT.
   Both cell deaths initiate the induction of ICD and release of DAMPs,
   thereby promoting DC maturation and CD8^+ T cell infiltration into
   tumor mass. Addition of PD-1 antibody to this nano–sono modality
   further amplifies the inhibitory effect on tumor growth. Collectively,
   this distinct liposomal nanosystem combined with ultrasonic irradiation
   (SDT) not only causes massive tumor cell death (apoptosis and
   ferroptosis) but also modulates the TME with favorable antitumor
   immunity, which creates the opportunity for immune checkpoint blockade
   to further improve the efficacy and response rate of immunotherapy.

4. Methods

4.1. Analysis of ^1O[2] Generation

   For ^1O[2] detection, 2 mL of HB liposomes (25 μg/mL) were mixed with
   100 μL DPBF solution. After US irradiation, the absorbance changes of
   DPBF at 410 nm were recorded.

4.2. ESR Measurement of ^1O[2]

   For ESR measurement of ROS, ^1O[2] generation was detected by TEMPO.
   Briefly, 1 mL of HB liposomes (50 μg/mL) was mixed with 20 μL TEMPO and
   exposed to US irradiation (1 W cm^2, 1.5 MHz) for 10 min. The
   characteristic peak signals were detected by an ESR spectrometer
   (Bruker, A300-12/10).

4.3. In Vitro Cellular Uptake

   Colon 26 cells were seeded into a 96-well plate containing coverslips
   (1 × 10^5 cells per well) and incubated overnight for adherence. Next,
   1 mL of RPMI 1640 containing Cy5-labeled HB liposomes was added to each
   well and incubated for 24 h. The fluorescence images were recorded
   using a CLSM.

4.4. Live/Dead Cell Staining Assay

   Live/dead cell staining against Colon 26 cells was performed with the
   calcein-AM/PI solution. Colon 26 cells were grown in 6-well plates at a
   density of 1 × 10^5 cells per well, respectively. After 24 h
   incubation, cells were treated with vehicle or HB liposomes for 6 h.
   Then, the plates were exposed to US irradiation (3.0 MHz; 1.5 W cm^–2;
   50% duty cycle) for 1 min. The plates were then incubated for another
   12 h, after which the cells were washed three times with PBS and
   stained with 0.3 mL assay buffer containing calcein-AM (2.0 μM) and PI
   (4.5 μM). Subsequently, the cells were incubated for 30 min at 37 °C
   and observed using a CLSM. Live cells were stained green by calcein-AM,
   whereas the dead cells were stained red by PI.

4.5. Statistical Analysis

   All data were presented as mean ± standard deviation (SD). Two-sided
   Student’s t-test was carried out to assess two-group differences. In
   all the cases, statistical significance was set as follows: *P < 0.05,
   **P < 0.01, ***P < 0.001, and not significant (N.S.).

Acknowledgments