In Situ Redox of TiSe2 Nanoplates to Excite Chemodynamic‐Enhanced Cancer Sono‐Immunotherapy

Different from the direct delivery of chemodynamic agents into tumor tissue for chemodynamic therapy (CDT), this work reports the fabrication of heterojunction sonosensitizers (CD/TiSe2) for the in situ generation of chemodynamic agents by responding to tumor microenvironment (TME), realizing the in situ CDT‐enhanced sono‐immunotherapy. The in situ redox of TiSe2 in acidic TME leads to the formation of TiOxSe2−x containing Se (0), selenate, and Ti3+/Ti4+ redox couple, which can facilitate in situ CDT through a Ti3+‐mediated Fenton‐like reaction and consume overexpressed glutathione (GSH) via a Ti4+‐mediated GSH depletion. Moreover, Se ions generated through the in situ redox process promote the maturation of dendritic cells, consequently activating the adaptive immune responses. In another aspect, the construction of heterojunctions within carbon dots and TiSe2 improves the reactive oxygen species (ROS) generation efficiency of TiSe2 in a cascaded manner, which enhances the sonodynamic activity and amplifies the chemodynamic performance. More importantly, the ROS produced by in situ CDT and sonodynamic therapy efficiently triggers immunogenic cell death through a synergistic therapy based on the elicitation of antitumor immunity with the aid of an immune checkpoint blockade. This work thus provides a distinct paradigm of transition metal selenide‐originated in situ Fenton‐like agent generation for in situ CDT‐enhanced sono‐immunotherapy.


Introduction
[6] In such a situation, it is difficult to fight against the complex TME with monotherapy, which might lead to poor therapeutic effect and even result in therapy resistance. [7,8]Given all this, recently emerging nanoformulations that respond to these endogenous TME stimuli provide good opportunity to achieve desired tumor therapeutic effect. [9,10][13][14] Therefore, it is an effective strategy to rationally design nanoformulations with efficient immune activation property and TME responsiveness for tumor eradication by combining multiple therapeutic modalities.
[17] Owing to its capacity to generate both beneficial and detrimental effects on tumorigenesis, ROS has been depicted as a double-edged sword. [11,18,19][29][30] Unfortunately, the strongly acidic conditions (pH 2-4) for Fe 2þ -catalyzed Fenton reaction still restricts the application of CDT in a pH range of 4.5-7.4 in TME. [11,31][37][38] As such, it is highly essential to construct TME-responsive nanoformulations to realize in situ CDT by utilizing the pathological hallmarks of the TME, such as high levels of H 2 O 2 , acidic pH, and redox conditions.
[41] Traditional organic sonosensitizers have been limited in SDT applications due to their adverse phototoxicity, unstable chemical properties, and poor dispersibility. [42,43]onsidering their advantageous of reduced phototoxicity and relatively high chemical stability, inorganic semiconductor nanoparticles such as TiO 2 , CoFe 2 O 4 , and MnO 2 have been developed as efficient sonosensitizers. [11,40,44,45]A qualified sonosensitizer should possess decreased bandgap to drive the generation of electron-hole pairs under the low-intensity ultrasound (US) irradiation. [2,46,47]Moreover, semiconductor sonosensitizers should be self-driven to facilitate spatial separation of US-excited electron-hole pairs, and subsequently transform the surrounding H 2 O/O 2 into cytotoxic ROS. [41,48,49][52][53][54][55] In this regard, it is highly desirable to explore TME-responsive sonosensitizers to relieve hypoxia and deplete GSH.In another respect, it has been well recognized that immunogenic cell death (ICD) triggered by ROS-mediated tumor therapeutic strategies could induce adaptive immunity against dead cell antigens and greatly promote inflammatory cell infiltration. [56,57]APCs can enhance recruitment and antigen presentation, which in turn stimulates cytotoxic T lymphocytes (CTLs) to eliminate tumor cells. [58]Therefore, developing multifunctional sonosensitizers with efficient ROS generation capability and TME regulation ability could realize cascade amplification of ROS production and adequately induce ICD for combination immunotherapy.
TiSe 2 , as a typical transition metal selenide, has been demonstrated to possess much lower optical bandgap compared to the transition metal oxide or sulfide. [59,60]Owing to these features, TiSe 2 was considered as a good candidate in energy storage and photocatalysis fields. [61,62]However, to our best knowledge, TiSe 2 -based nanoformulations have not been reported to be employed as sonosensitizers or chemodynamic agents for SDT or CDT applications.It worth noting that Se element present in TiSe 2 is a trace element that plays an important role in the human immune system, which could promote dendritic cell (DC) maturation and then elicit immune response. [63]In this work, a one-step facile oil phase process was utilized to prepare TiSe 2 nanoplates using Se powder and TiCl 4 as the precursors.As an efficient sonosensitizer, TiSe 2 nanoplates can effectively enhance US-triggered separation of electron-hole pairs owing to their narrow bandgap (1.62 eV).More importantly, the in situ redox of TiSe 2 nanoplates was occurred in acidic TME after storage of 2 days to obtain TiO x Se 2Àx nanoplates containing Ti 3þ /Ti 4þ redox couple, Se (0), and selenate.The resultant Ti 3þ /Ti 4þ redox couple in TiO x Se 2Àx nanoplates can realize in situ CDT through Ti 3þ -mediated Fenton-like reaction and consume overexpressed GSH via Ti 4þ -mediated GSH depletion.Moreover, Se ions produced after the in situ redox can promote the maturation of DCs and then activate the adaptive immune responses.To amplify the ROS-induced ICD effects, a Z-scheme heterojunction was then constructed by depositing carbon dots (CDs) on the surface of TiSe 2 nanoplates to achieve the enhanced ROS generation efficiency of TiSe 2 through in situ CDTenhanced SDT.As sonosensitizers or chemodynamic agents for SDT or CDT, the CD/TiSe 2 and CD/TiO x Se 2Àx heterojunctions displayed an increased ROS production ability due to the improved separation efficiency of electron-hole pairs, as well as amplified Fenton-like reaction activity owing to the accelerated transfer of carriers.Notably, the greatly enhanced ROS level induced by CD/TiSe 2 via SDT and in situ CDT can trigger robust ICD to stimulate systemic immune responses and to boost the abscopal effects.Owing to these favorable features, the in situ CDT-enhanced sono-immunotherapy via CD/TiSe 2 heterojunctions accompanied by the immune checkpoint blockade to boost antitumor immunity achieved the suppression of primary and distant tumors (Scheme 1).

Results and Discussion
A one-step oil phase method was utilized to synthesize the layered TiSe 2 nanoplates with circular morphology after reacting selenium powder and TiCl 4 in trioctylphosphine/oleylamine solution.The sheet-like structure of TiSe 2 nanoplates was demonstrated by their transmission electron microscopy (TEM) image (Figure 1a).The size of TiSe 2 nanoplates was also measured by the dynamic light scattering, which revealed that their hydrodynamic diameter was determined to be 206.4nm (Figure 1f ).Moreover, the highly crystalline characteristics of TiSe 2 nanoplates were observed in their high-resolution TEM (HRTEM) image, illustrating that the spacings of lattice fringes were 0.3 nm, which was consistent with that of the (002) crystal plane of TiSe 2 .The high crystallinity of TiSe 2 nanoplates was also confirmed by their X-ray diffraction (XRD) patterns, which showed that the diffraction peaks of the obtained TiSe 2 nanoplates were consistent with that of the hexagonal type TiSe 2 (JCPDS no.30-1383).Two peaks at 413 and 619 cm À1 were detected in the Raman spectrum of TiSe 2 nanoplates (Figure S1a, Supporting Information).Figure 3c,d exhibits the presence of Ti 4þ and Se 2À in the TiSe 2 nanoplates.The stability of TiSe 2 nanoplates was evaluated by storing TiSe 2 in physiological saline for 7 days.The apparent precipitate was observed in TiSe 2 solution after 7 days (Figure S2, Supporting Information), indicating the poor stability of TiSe 2 nanoplates.
To improve the stability of TiSe 2 nanoplates for the guarantee of their biomedical applications, CD/TiSe 2 heterojunctions were fabricated by depositing biocompatible CDs on TiSe 2 nanoplates.The obtained CD/TiSe 2 heterojunctions exhibited enhanced stability in physiological saline without significant precipitate after 7 days (Figure S2, Supporting Information).The enhanced stability of CD/TiSe 2 heterojunctions compared with the pristine TiSe 2 nanoplates was also confirmed by their hydrodynamic diameter.As depicted in Figure S3, Supporting Information, the hydrodynamic diameter of CD/TiSe 2 was measured to be 229.1 AE 9.0 nm, 231.1 AE 11.7 nm, and 233.0 AE 14.3 nm, after storing for 1, 3, and 7 D, respectively, showing insignificant change of hydrodynamic diameters after long-term storage.In contrast, the obvious increase of the hydrodynamic diameters was detected in TiSe 2 after storing for 1, 3, and 7 D. TEM image of CD/TiSe 2 clearly showed that CDs were uniformly dispersed on the surface of TiSe 2 nanoplates and the size of the obtained heterojunctions was determined to be approximately 220 nm (Figure 1c).By comparison with pristine TiSe 2 nanoplates, the increased size of heterojunctions could be ascribed to the loading of CDs.In addition, the HRTEM image revealed that the distinct lattice spacing of 0.30 and 0.21 nm was detected in CD/TiSe 2 heterojunctions (Figure 1d), which was attributed to the (002) planes of TiSe 2 and CDs, respectively.Zeta potential measurements also demonstrated the successful deposition of positively charged CDs, manifesting that the surface charge of TiSe 2 nanoplates changed from negative to positive after loading CDs (Figure 1e).
To further unveil the successful deposition of CDs on the surface of TiSe 2 nanoplates, the structure and chemical component of CD/TiSe 2 heterojunctions were then investigated by XRD, Raman, and X-ray photoelectron spectroscopy (XPS).In addition to the diffraction peaks attributed to TiSe 2 nanoplates, a new broad peak at around 21.2°could be detected in CD/TiSe 2 heterojunctions (Figure 1g), which was consistent with the characteristic peak of (002) plane of the single-component CDs.Raman spectroscopy further confirmed the presence of CDs on TiSe 2 nanoplates.In addition to the two typical characteristic peaks of TiSe 2 nanoplates, another two typical characteristic peaks of CDs could also be detected at 1356 and 1580 cm À1 (Figure 1h), clearly verifying the successful loading of CDs onto TiSe 2 nanoplates.It should be noted that a high graphitic degree of CDs was demonstrated by the higher intensity of the ordered G peak compared with that of the disordered D peak (Figure S1b, Supporting Information).The survey XPS spectrum of CD/TiSe 2 heterojunctions revealed the presence of C 1s, N 1s, and O 1s peaks attributed to CDs and the existence of Se 3d and Ti 2p peaks corresponded to TiSe 2 nanoplates (Figure 1i).Similar to single-component TiSe 2 nanoplates, the presence of Ti 4þ and Se 2À was observed in the high-resolution Ti 2p and Se 3d spectra of heterojunctions (Figure 1j,k), illustrating that the loading of CDs did not affect the structure of TiSe 2 nanoplates.Moreover, the high-resolution N 1s and C 1s spectra of CD/TiSe 2 heterojunctions demonstrated the presence of pyridine N, graphitic N, C─C, and C─N (Figure 1l and S4, Supporting Information), which was consistent with those of singlecomponent CDs (Figure S5, Supporting Information).The absorption spectra of CDs, TiSe 2 , and CD/TiSe 2 were further assessed to confirm the successful deposition of CDs.As depicted in Figure S6, Supporting Information, the enhanced absorption was observed in TiSe 2 nanoplates after depositing CDs.
To evaluate the enhanced sonodynamic properties of heterojunctions, we utilized 1,3-diphenylisobenzofuran (DPBF) as a colorimetric probe to detect 1 O 2 under US irradiation in the presence of CD/TiSe 2 , TiSe 2 , and CDs.As the irradiation time of US increased from 0 to 10 min, the absorption intensity of DPBF of TiSe 2 and CD/TiSe 2 gradually decreased at 418 nm (Figure 2a-c), suggesting their good ROS generation capability.Significantly, compared with TiSe 2 and CDs, CD/TiSe 2 exhibited the strongest ability to produce 1 O 2 under the same conditions, suggesting that Scheme 1. Schematic illustration of the preparation procedures of CD/TiSe 2 heterojunctions and its acting mechanism for in situ CDT-enhanced sonoimmunotherapy.
the construction of heterojunctions could improve the generation efficiency of 1 O 2 .Furthermore, the rate constant of 1 O 2 generation in heterojunction group was determined to be 0.17 min À1 (Figure 2d), which was 1.7 times than that of pristine TiSe 2 (0.1 min À1 ).Apart from the colorimetric reaction, we also conducted the electron spin resonance (ESR) measurements to detect the generation of 1 O 2 in the presence of CD/TiSe 2 , TiSe 2 , and CDs under US irradiation.By comparison with single-component TiSe 2 and CDs, the higher 1 O 2 signal was detected in the heterojunction group (Figure 2i), demonstrating the enhanced ROS generation ability of CD/TiSe 2 .We then constructed CD/TiSe 2 Z-scheme heterojunctions with varied CD loading ratio (10%, 15%, and 20%) and then compared their sonodynamic properties, which was closely relied on the loading efficiency of CDs on TiSe 2 .Figure S7, Supporting Information exhibits that the rate constant of 1 O 2 generation in the 15% CD/TiSe 2 group was higher than the other groups.
To explore the mechanism of the enhanced sonodynamic performances of heterojunctions, we measured and calculated the energy band structure of single-component TiSe 2 and CDs as well as the heterojunction-fabricated CD/TiSe 2 .First, the bandgap of TiSe 2 and CDs was measured through the UV-vis absorption spectra and calculated to be 1.62 and 2.41 eV, respectively (Figure 2e,f ).Such narrowed bandgap of TiSe 2 nanoplates (1.62 eV) was much lower than that of commercial TiO 2 (3.2 eV), elucidating that TiSe 2 nanoplates were more easily activated by US.Then, the valence band (VB) of TiSe 2 and CDs was obtained by their XPS-VB spectra, which was measured to be 0.44 and 1.0 eV, respectively (Figure 2g,h).After determining the E g , E VB , and E CB , it should be noted that the E CB and E VB of TiSe 2 nanoplates were both lower than those of CDs (Figure 2j).Owing to the presence of internal electric field, the US-excited electrons cannot transfer from the CB of CDs to that of TiSe 2 .Therefore, the US-excited electron would transfer from the CB of TiSe 2 to the VB of CDs (Figure 2j).The successful construction of Z-scheme heterojunctions within CD/TiSe 2 could effectively prevent the electron-hole pair recombination, leading to that more electrons would react with oxygen for the production of more 1 O 2 .
Considering that CD/TiSe 2 heterojunctions have enhanced sonodynamic performances, we continued to explore whether they possessed chemodynamic properties at the different pH values including 4.5, 6.0, 6.5, and 7.4.The chemodynamic properties of CD/TiSe 2 , TiSe 2 , and CDs were first investigated by the colorimetric reaction of 3,3',5,5'-tetramethylbenzidine (TMB).Figure S8, Supporting Information exhibits that the absorption peak of CD/TiSe 2 , TiSe 2 , and CDs has no significant change at 652 nm with the concentration of H 2 O 2 increased at pH 6.0, suggesting that CD/TiSe 2 , TiSe 2 , and CDs did not possess chemodynamic properties.In addition to the acidic condition, we also  3a and S8, Supporting Information), demonstrating that TiSe 2 nanoplates before and after depositing CDs have no significant chemodynamic activities.
Interestingly, the excellent chemodynamic properties were detected in the CD/TiSe 2 solution at pH 4.5, 6.0, and 6.5 after storing for 2 days (Figure 3e and S9a,b, Supporting Information), illustrating that the TMB absorbance at 652 nm enhanced significantly as the H 2 O 2 concentration increasing.We then performed TEM observation, XPS analysis, and XRD pattern to investigate the mechanism of TiSe 2 -mediated Fenton-like reaction.The HRTEM image of TiSe 2 after storing for 2 days is presented in Figure S11, Supporting Information, which indicated the presence of TiO 2 , TiSe 2 , and Se (0).We also performed selected area electron diffraction (SAED) pattern of TiSe 2 after storing for 2 days.Figure 3f exhibits that the six crystal planes were detected in their SAED pattern, which could be ascribed to the TiSe 2 (001), TiO 2 (101), Se (101), TiO 2 (103), Se (110), and Se (003), respectively.Figure 3g reveals that the diffraction peaks of TiO 2 (PDF #21-1272) and Se (PDF #06-0362) were detected in TiSe 2 after storing for 2 days, suggesting that an in situ redox reaction occurred during the storage of TiSe 2 .XPS analysis further demonstrated the in situ redox of TiSe 2 after storing for 2 days (Figure S12, Supporting Information).In addition to two characteristic peaks attributed to Ti 4þ at 457 and 465 eV, we also found two typical characteristic peaks of Ti 3þ at 455 and 463 eV (Figure 3h), elucidating that Ti 4þ ions in TiSe 2 nanoplates were partially reduced to Ti 3þ at pH 4.5.Moreover, the presence of TiO x Se 2Àx , t-Se, SeO x , and SeO 4 2À in TiSe 2 nanoplates was demonstrated by their high-resolution Se 3d spectrum (Figure 3i), indicating that Se 2À ions in TiSe 2 nanoplates were partially oxidated to Se and selenate ion.These characterization results clearly demonstrated that the in situ redox reaction of Ti 4þ and Se 2À ions existed in the storage process of TiSe 2 (Figure 3o).Considering the presence of Ti 3þ , Se, and selenate ion, we named the TiSe 2 after storing for 2 days as TiO x Se 2Àx .Therefore, the excellent chemodynamic properties of CD/TiO x Se 2Àx and TiO x Se 2Àx could be ascribed to the Ti 3þmediated Fenton-like reaction.
To quantitatively compare the chemodynamic activities of CD/TiO x Se 2Àx and TiO x Se 2Àx , we performed Michaelis-Menton kinetics analysis.Figure 3j exhibits that CD/TiO x Se 2Àx heterojunctions possessed higher V max (4.46 Â 10 À7 M s À1 ) and lower K m (0.08 mM) compared with single-component TiO x Se 2Àx nanoplates (V max = 2.19 Â 10 À7 M s À1 and K M = 0.29 mM).The chemodynamic activities of CD/TiO x Se 2Àx at pH 6.5 were weaker than those at pH 6.0 (Figure S9a,b, Supporting Information), suggesting that the Fenton-like reaction was more easily to carry out in the acidic TME.Importantly, no significant •OH generation was detected in CD/TiO x Se 2Àx solution in the presence of H 2 O 2 at pH 7.4 (Figure S9c, Supporting Information), indicating that CD/TiO x Se 2Àx could not exhibit chemodynamic activities under normal physiological conditions.In contrast, another transition metal selenides (Cu 2Àx Se) exhibited efficient chemodynamic activity at pH 7.4 (Figure S13, Supporting Information), suggesting that Cu-based Fenton-like agents could achieve effective CDT in the physiological neutral conditions.The phenomenon might lead to the toxicity of Cu-based Fenton-like agents to normal tissues, posing significant safety issues.Compared with Cu 2Àx Se, the TME-responsive CD/TiO x Se 2Àx possessed Fenton-like reaction activity only in the weakly acidic TME, achieving the in situ CDT to avoid the significant toxicity to normal tissues.In addition to the colorimetric reaction, we also conducted ESR measurements to detect the generation of •OH in the presence of CD/TiO x Se 2Àx , TiO x Se 2Àx , and CDs.As illustrated in Figure 3k, the highest ESR signal was observed in the CD/TiO x Se 2Àx group compared with that of the other groups, forcefully confirming the enhanced •OH generation efficiency of heterojunctions.To determine the optimal treatment window for SDT, the chemodynamic activities of CD/TiSe 2 after storing for 1 day were also investigated at different pH (4.5, 6.0, 6.5, and 7.4).As illustrated in Figure S14, Supporting Information, no significant Fenton-like reaction activities of CD/TiSe 2 were detected after storing for 1 day, suggesting that the in situ redox reaction of Ti 4þ and Se 2À ions in CD/TiSe 2 heterojunctions have not been performed during the 1 day storage, which could ensure the efficient SDT via CD/TiSe 2 .We also investigated whether the redox reaction of TiSe 2 would affect their sonodynamic performance.The significant decrease of DPBF absorbance in TiO x Se 2Àx group under US irradiation is observed in Figure S15, Supporting Information, which was similar to that of pristine TiSe 2 nanoplates.This phenomenon could be ascribed to the similar bandgap of TiSe 2 (1.62 eV) and TiO x Se 2Àx (1.7 eV) (Figure S16, Supporting Information).We further investigated the sonodynamic and chemodynamic activities of CD/TiO x Se 2Àx at various storage periods (2, 4, 6 days).As depicted in Figure S17a, Supporting Information, the excellent sonodynamic properties of CD/TiO x Se 2Àx were observed at different storage periods (2, 4, 6 days), suggesting that the in situ redox reaction of CD/TiSe 2 would not affect their sonodynamic performance.In addition, the chemodynamic activities of CD/TiO x Se 2Àx after storing for 6 days were higher than that of 2 days (Figure S17b, Supporting Information), which could be ascribed to the more Ti 3þ after in situ redox reaction of CD/TiSe 2 .
We then speculated whether the heterojunction-fabricated CD/TiSe 2 possessed the capability to consume GSH. Figure 3l,m reveals that the characteristic absorption peak of GSH in CD/TiSe 2 and TiSe 2 solution was decreased as the incubation time increased, while no obvious GSH depletion ability of CDs was observed (Figure 3n).Specifically, the GSH consumption rate of CD/TiSe 2 and TiSe 2 was calculated to be 80.35% and 81.3%, respectively (Figure S18, Supporting Information).The excellent GSH depletion ability of CD/TiSe 2 and TiSe 2 could be ascribed to the presence of Ti 4þ , which reacted with GSH to generate GSSG.Based on the above results, the heterojunctionfabricated CD/TiSe 2 could integrate several functions of good sonodynamic performance, in situ redox of TiSe 2 -induced chemodynamic activity, and the Ti 4þ -mediated GSH depletion ability into one platform for efficient in situ chemodynamicenhanced SDT.
Inspired by the CD/TiSe 2 -mediated in situ chemodynamic activities and heterojunction-enhanced sonodynamic performances, a series of cell experiments were carried out to evaluate the in vitro therapeutic effect of CD/TiSe 2 .We initially investigated the cell uptake behaviors of TiSe 2 and CD/TiSe 2 in 4T1 cells.Figure S19, Supporting Information exhibits that the bright red fluorescent signal was detected in the cytoplasm of 4T1 cells after treating with ICG@TiSe 2 and ICG@CD/TiSe 2 , suggesting their efficient cellular internalization.These results demonstrated that the deposition of CDs would not affect the cellular uptake behavior of TiSe 2 nanoplates.We then evaluated the biocompatibility of CD/TiSe 2 , TiSe 2 , and CDs using 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay.As presented in Figure 4a and S20, Supporting Information, LO2 cells treated with CD/TiSe 2 , TiSe 2 , or CDs at different concentrations did not exhibit significantly cytotoxic after 24 or 48 h, demonstrating their good biocompatibility.In contrast, 4T1 cells incubated with CD/TiSe 2 or TiSe 2 for 48 h revealed a certain cytotoxic effect (Figure 4b and S21a, Supporting Information), which could be attributed to the chemodynamic activities of CD/TiO x Se 2Àx or TiO x Se 2Àx in the cancer cells.Moreover, the cytotoxicity of CD/TiO x Se 2Àx against 4T1 cells was higher than that of TiO x Se 2Àx at the same concentration, confirming the higher chemodynamic activity of heterojunctions.The singlecomponent CDs exhibited insignificant cytotoxic against 4T1 cells (Figure S21b, Supporting Information), suggesting that CDs did not possess Fenton-like reaction activity.These results demonstrated that the in situ redox reaction of TiSe 2 endowed CD/TiO x Se 2Àx and TiO x Se 2Àx with good chemodynamic activity.We further extended the incubation time of CD/TiO x Se 2Àx with LO2 and 4T1 cells to 72 and 96 h. Figure S22b, Supporting Information exhibits that the cytotoxic effect of CD/TiO x Se 2Àx against 4T1 cells was further enhanced, indicating the sustained Fenton-like reaction activity of CD/TiO x Se 2Àx in the cancer cells.However, no significant death of LO2 cells treated with CD/ TiO x Se 2Àx for 96 h was detected (Figure S22a, Supporting Information), demonstrating the excellent biosafety of CD/ TiO x Se 2Àx owing to their negligible chemodynamic activity in the physiological neutral conditions.In comparison, although the cell viability of 4T1 cells treated with Cu 2Àx Se obviously decreased (Figure S23b, Supporting Information), Cu 2Àx Se also exhibited significant cytotoxicity to normal LO2 cells at the same concentrations (Figure S23a, Supporting Information), illustrating that Cu-based Fenton-like reactions can occur in the physiological neutral conditions, which could pose great safety risks to normal cells or tissues.We also performed the GSH/GSSG assay to investigate the intracellular GSH levels.Compared with the control group, the decreased GSH/GSSH ratio in 4T1 cells was detected in the CD/TiSe 2 group (Figure S24, Supporting Information), indicating the good GSH depletion ability of CD/TiSe 2 owing to the presence of Ti 4þ .Subsequently, we assessed the in vitro therapeutic effect of in situ CDT-enhanced SDT. Figure 4d reveals that the cytotoxicity of 4T1 cells treated with CD/TiSe 2 under US irradiation increased to ≈95%, which was much higher than that of CD/TiSe 2 in the absence of US treatment, demonstrating the synergetic therapeutic effect of SDT and CDT.Furthermore, the cell viability of 4T1 cells under the treatments of SDT and CDT was much lower than that of single-component TiSe 2 (Figure 4c), clearly illustrating the heterojunction-improved sonodynamic and chemodynamic properties.
No obvious red fluorescence was detected in the control and US alone group, confirming the good safety of US irradiation (Figure 4e).In contrast, TiSe 2 -treated 4T1 cells exhibited a slight red fluorescence, attributed to their in situ chemodynamic activity.Moreover, more dead cells were observed in the CD/TiSe 2 group compared with the TiSe 2 alone group, manifesting that the fabrication of heterojunctions could improve the in situ Fenton-like reaction activity of TiSe 2 .Notably, almost all of 4T1 cells were dead in the CD/TiSe 2 þ US group, suggesting the efficient therapeutic effect of in situ CDT-enhanced SDT via heterojunctions.We then investigated the mechanism of synergetic CDT and SDT at the cellular level.The strongest green fluorescence signals representing ROS were detected in the CD/TiSe 2 þ US group (Figure 4f ), indicating that CD/TiSe 2 in the presence of US irradiation produced most ROS compared with the other groups.
Having demonstrated the efficient in vitro therapeutic effect of CD/TiSe 2 through ROS generation, we further investigated the cell death pathway involved in the synergistic CDT and SDT.The ROS-induced ICD of tumor cells has been reported by other groups, which would trigger the exposure of calreticulin (CRT), release of high mobility group box 1 (HMGB1), and secretion of adenosine triphosphate (ATP). [56,57]DCs can be recruited and activated by the three types of damage-associated molecular patterns (DAMPs), which then triggered the adaptive immune response to eliminate tumor cells. [64,65]On this basis, the CRT expression on 4T1 cells after different treatments was initially investigated by confocal imaging and flow cytometry.The strongest green fluorescence signal was observed in the CD/TiSe 2 þ US group compared with the other groups (Figure 5a), suggesting that the most CRT was exposed from endoplasmic reticulum to the surface of cells after the synergetic in situ CDT and SDT.Subsequently, the quantitative results of CRT expression in different groups were analyzed by flow cytometry, which were determined to be 4.97%, 5.47%, 7.54%, 9.05%, 16.02%, and 24.11% in the control, US, TiSe 2 , CD/TiSe 2 , TiSe 2 þ US, and CD/TiSe 2 þ US groups, respectively (Figure S25, Supporting Information).Compared with the control group, the proportion of CRT positive cells in the CD/TiSe 2 þ US group and TiSe 2 þ US group was increased by approximately 4.9-and 3.2-fold, respectively (Figure 5e).Enzyme linked immunosorbent assay (ELISA) was then used to measure the expression level of HMGB1 in 4T1 cells after different treatments.The concentration of released HMGB1 in 4T1 cells treated with CD/TiSe 2 þ US was measured to be 5.9 AE 0.4 μg mL -1 (Figure 5f ), which was higher than that of TiSe 2 þ US (5.8 AE 0.2 μg mL -1 ), CD/TiSe 2 (4.3 AE 0.2 μg mL -1 ), and TiSe 2 (4.1 AE 0.2 μg mL -1 ).The enhanced HMGB1 release of CD/TiSe 2 under US irradiation through combined CDT and SDT could induce the secretion of proinflammatory cytokine and then stimulate DC maturation.To further demonstrate the ROS-triggered ICD, the ATP release of 4T1 cells was investigated after different treatments using an ATP detection assay kit. Figure 5g reveals that the extracellular ATP content in CD/TiSe 2 þ US group was significantly increased accompanied by the decrease of intracellular ATP level, elucidating that the CD/TiSe 2 heterojunctions under US irradiation could secrete more ATP into the extracellular space.These results forcefully demonstrated that ICD could be triggered by CD/TiSe 2 heterojunctions through synergistic in situ CDT and SDT.
As an important APCs, DCs could be activated by tumorassociated antigen. [64]ICD-induced release of DAMPs can bind to DC cells, and then present tumor-associated antigen to T cells, activating adaptive immune responses. [66,67]The immature DCs were initially extracted from the bone marrow-derived cells in Balb/c mice in the presence of growth factors including IL-4 and GM-CSF.After adding the supernatant of 4T1 cells incubated with CD/TiSe 2 or TiSe 2 in the presence or absence of US irradiation (Figure 5b), the expression levels of CD80 and CD86 on the bone marrow derived DCs were then investigated.As presented in Figure 5c, the proportion of CD80 þ CD86 þ DCs in CD/TiSe 2 þ US group was measured to be 36.1%,which was higher than that of TiSe 2 þ US group (33.3%),CD/TiSe 2 alone group (21.4%), and TiSe 2 alone group (19.0%).Compared with the TiSe 2 þ US group, the higher percentage of CD80 þ CD86 þ DCs in CD/TiSe 2 þ US group illustrated that the construction of heterojunctions improved the ROS generation ability of single-component TiSe 2 nanoplates, then triggering a stronger ICD effect to stimulate the maturation of DCs.
Considering that Se 2À ions in TiSe 2 nanoplates were partially oxidated to Se (0) and selenate ion after storing for 2 days, we then evaluated the effect of TiO x Se 2Àx on the maturation of DCs.TiO 2 , TiSe 2 , or TiO x Se 2Àx was directly incubated with DCs and the expression levels of CD80 and CD86 on the DCs were then investigated.Figure 5h, S26, Supporting Information exhibit that the proportion of CD80 þ CD86 þ DCs in TiSe 2 and TiO 2 groups was similar to that in the control group, suggesting that Ti 4þ and Se 2À showed insignificant effect on the maturation of DCs.In contrast, the proportion of CD80 þ CD86 þ DCs in TiO x Se 2Àx group (200 μg mL -1 ) was increased to 18.9%, which was 4.7 times higher than that of the control group (Figure 5i), illustrating that Se and selenate ion could promote the maturation of DCs, which might activate the systemic immune response for killing cancer cells.These results clearly demonstrated that CD/TiSe 2 -induced ICD through ROS generation and TiO x Se 2Àx containing Se and selenate ion both possessed the capability to stimulate the maturation of DCs, which might achieve a good antitumor effect in vivo.
Considering that the in situ CDT and SDT could induce antitumor immune response through activating ICD, the in vivo anticancer efficacy of CD/TiSe 2 heterojunctions was then investigated using the 4T1 xenograft tumor model.The nearinfrared (NIR) fluorescence dye indocyanine green (ICG) was conjugated to the surface of CD/TiSe 2 for NIR imaging.After intravenously injecting ICG@CD/TiSe 2 , a strong fluorescence signal was detected in the tumor sites (Figure 6b,e), suggesting that CD/TiSe 2 gradually accumulated in the tumor tissue.Furthermore, the ex vivo NIR imaging results revealed that the most accumulation of CD/TiSe 2 in the tumor tissue was detected at the 24 h postinjection (Figure 6c,f ).Therefore, the US irradiation should be performed after intravenous injection of CD/TiSe 2 for 24 h.Apart from NIR fluorescence imaging, we also performed the pharmacokinetic examination of CD/TiSe 2 after systemic administration.Figure S27, Supporting Information shows a typical double-compartment pharmacokinetic behavior of the CD/TiSe 2 .The elimination half-life (t 1/2 ) of CD/TiSe 2 was calculated to be 12.6 AE 0.3 h, suggesting that CD/TiSe 2 was able to circulate for a long time for the higher tumor accumulation.
The in vivo anticancer efficacy of CD/TiSe 2 was evaluated in 4T1 tumor-bearing mice as described in Figure 6a.Compared with the group treated with PBS or US irradiation alone, the growth of tumor volume in CD/TiSe 2 alone group was decreased (Figure 6g), suggesting that the in situ CDT induced by the redox reaction of TiSe 2 in the TME possessed a certain therapeutic effect.TiSe 2 þ US group exhibited higher inhibitory effects on tumor growth compared with CD/TiSe 2 alone group, demonstrating the enhanced therapeutic effect of synergetic in situ CDT and SDT.Notably, the inhibition rate of tumor growth in CD/TiSe 2 þ US group was determined to be 100%, which was higher than that of TiSe 2 þ US group (62.4%), elucidating that the fabrication of heterojunctions could improve the in situ CDT and SDT effect of single-component TiSe 2 nanoplates.Furthermore, the mice in CD/TiSe 2 þ US group survived for 60 days, whereas the control group mice only lived for approximately 20 days (Figure 6h).The body weight of the mice in each group had no significant change during treatment periods (Figure S28, Supporting Information), indicating that these treatments exhibited insignificant side effects on the growth of mice.The histological analysis of tumor tissues was then performed to investigate the therapeutic effect of CD/TiSe 2 -mediated in situ CDT and SDT using TUNEL and H&E staining.Figure S29 and S30, Supporting Information exhibit that the most apoptosis or necrosis of tumor cells in CD/TiSe 2 þ US group were detected compared with the other groups, manifesting that the severe apoptosis/necrosis could be induced by in situ CDT-enhanced SDT via CD/TiSe 2 heterojunctions.
We then investigated the in vivo therapeutic mechanism of in situ CDT-enhanced SDT.As depicted in Figure 6i, the red fluorescence signals of CRT in CD/TiSe 2 þ US group presented in the immunofluorescence staining images were significantly higher than that in the other groups, demonstrating the enhanced expression of CRT after the in situ CDT and SDT.The HMGB1 release level in each group was then investigated by the ELISA assay.Figure S31, Supporting Information exhibits that the concentration of released HMGB1 in the combined CDT and SDT group (CD/TiSe 2 þ US) was determined to be 8.4 AE 0.3 μg mL, which was 2.1 times higher than that in the single CDT group (CD/TiSe 2 ) (4.0 AE 0.2 μg/mL), illustrating that the combination therapy of SDT and CDT could produce more ROS to induce stronger ICD effect.Moreover, the higher intratumoral ATP levels were detected in CD/TiSe 2 þ US group compared with the other groups (Figure S32, Supporting Information).We then investigated whether the treatment of CD/TiSe 2 in the presence of US irradiation could promote the maturation of DCs using flow cytometry.As illustrated in Figure S33, Supporting Information, the treatment of PBS or US irradiation alone exhibited insignificant effect on the maturation of DCs, which could be ascribed to their negligible ROS generation.In contrast, the increased populations of CD80 þ CD86 þ matured DCs were detected in CD/TiSe 2 alone group, attributing to the in situ CDT-induced ICD effect.Significantly, the populations of CD80 þ CD86 þ matured DCs in CD/TiSe 2 þ US group were higher than that in the other groups, which could be favorable for presenting antigens to tumor-specific T cells.These results clearly demonstrated that the treatments of in situ CDT and SDT via CD/TiSe 2 heterojunctions could activate immune responses by releasing tumor-associated antigens to induce ICD.
PD-1/PD-L1-mediated ICD therapy has been shown to be highly effective in enhancing cytotoxic T-lymphocyte activity, leading to a durable response and extended long-term survival when used in either monotherapy or combined therapy. [11]onsidering the successful ICD activation through CD/TiSe 2 , αPD-L1 was used to combine SDT/PTT in order to reduce both primary and distant tumors (Figure 7a).During the 14 day treatment period, no considerable fluctuation in mouse body weight was detected for all treatments, indicating that these treatments exhibited an insignificant side effect on the growth of mice (Figure 7g).No obvious inhibitory effects of PBS or αPD-L1 administration on the growth of both primary and distant tumors were detected during the treatment period.In contrast, the inhibition rate of primary tumor growth in CD/TiSe 2 þ US group was determined to be 100% (Figure 7e), which was higher than that of CD/TiSe 2 alone group, suggesting that heterojunctions could achieve complete tumor eradication of primary tumor through combinatorial in situ CDT and SDT as well as ROSinduced immunotherapy.However, the distant tumors in CD/ TiSe 2 þ US group were not completely eradicated after 14 days (Figure 7f ), suggesting that the ICD-induced systemic antitumor immunity could not enough to completely eliminate distant tumors.Notably, the primary and distant tumors in CD/TiSe 2 þ US þ αPD-L1 group were completely eradicated, elucidating that the host immune response induced by in situ CDT-and SDT-mediated ICD could be enhanced by the introduction of αPD-L1.Moreover, the outstanding tumor growth inhibition effect of in situ CDT-enhanced SDT and immunotherapy via CD/TiSe 2 heterojunctions combined with αPD-L1 was visualized from the representative photographs of the primary and distant tumors (Figure 7b,c).In addition, the mice in CD/TiSe 2 þ US þ αPD-L1 group were all lived after 60 days (Figure 7d), while the mice in the control group were only alive within 20 days.
Having confirmed the excellent therapeutic effect of in situ CDT-enhanced SDT and immunotherapy via CD/TiSe 2 heterojunctions combined with αPD-L1, we then investigated the immune responses on the primary and distant tumors to elucidate the therapeutic mechanism.The expression levels of CD80 and CD86 in the lymph nodes were analyzed by flow cytometry after preparing the single-cell suspension.As depicted in Figure 8a, the populations of CD80 þ CD86 þ matured DCs in CD/TiSe 2 þ US group were higher than that in CD/TiSe 2 alone group, suggesting that the in situ CDT-enhanced SDT could induce stronger immune responses to stimulate the maturation of DCs.Moreover, the most matured DCs were detected in CD/ TiSe 2 þ US þ αPD-L1 group, which revealed that the proportion of CD80 þ CD86 þ DCs was determined to be 33.3AE 0.7% (Figure 8b), which could be favorable for presenting antigens to tumor-specific T cells.
We then assessed the activation of T cell in spleen, lymph nodes, and tumor through the in situ CDT-enhanced SDT and immunotherapy.Figure S34, Supporting Information exhibits that the percentage of CD4 þ CD3 þ T cells and CD8 þ CD3 þ T cells in spleen of the mice after treating with CD/TiSe 2 þ US þ αPD-L1 was determined to be 25.46 AE 1.12% and 25.68 AE 2.23%, which were much higher than that of the control group (2.96 AE 0.20% and 3.63 AE 0.23%), respectively.The similar phenomenon can be found in the lymph nodes (Figure S35, Supporting Information).In addition, the enhanced populations of CD4 þ CD3 þ T cells and CD8 þ CD3 þ T cells in the primary and distant tumor tissues of mice after the treatment of CD/TiSe 2 þ US þ αPD-L1 were observed compared with the other groups (Figure 8c-h), illustrating that the robust immune responses were effectively triggered by the in situ CDT-enhanced SDT and immunotherapy.Macrophages and CTLs actively secrete proinflammatory cytokines, which is considered to be a hallmark of the upregulation of inflammatory responses.These cytokines play an important role in triggering the activity of immune effector cells and in causing cytotoxic effects.Therefore, we investigated the serum levels of cytokines including interleukin-6 (IL-6), interferon-γ (IFN-γ), and tumor necrosis factor-α (TNF-α) using ELISA assay on day 3.As illustrated in Figure S36, Supporting Information, the expression levels of IL-6, IFN-γ, and TNF-α in CD/TiSe 2 þ US þ αPD-L1 group were much higher than those of the other groups.These results clearly demonstrated that the CD/TiSe 2 -mediated in situ CDT-enhanced SDT and immunotherapy could activate immune responses by releasing tumor-associated antigens to induce ICD.
Finally, we performed the biodistribution analysis of CD/ TiSe 2 at different time points to investigate their metabolic pathway.As presented in Figure S37, Supporting Information, CD/ TiSe 2 was primarily accumulated in liver due to the capture of reticuloendothelial system.At 7 days after injection, no obvious Ti signal can be detected in these organs, indicating that CD/ TiSe 2 was eliminated from the mice through the liver.The in vivo long-term toxicity of CD/TiSe 2 -mediated in situ CDT-enhanced SDT and immunotherapy was evaluated by the histopathological test and hematological analysis.After intravenous injection of CD/TiSe 2 for 24 h, the healthy mice were sacrifice and their blood and major organs were harvested.H&E staining of main organs and main markers of hepatic and renal function in each group revealed no obvious change on day 14 (Figure S38 and S39, Supporting Information), demonstrating the high biosafety of CD/TiSe 2 heterojunctions.These results forcefully demonstrated that no toxicity was observed in mice when exposed to the synthesized CD/TiSe 2 heterojunctions, demonstrating their potential for biomedical applications.

Conclusion
In summary, a multifunctional sonosensitizer of TiSe 2 was prepared via a one-step facile oil phase process for in situ CDTenhanced sono-immunotherapy.Taking advantage of the narrow bandgap (1.62 eV), TiSe 2 nanoplates improved US-triggered 1 O 2 generation efficiency owing to the enhanced separation of electron-hole pairs.Interestingly, the in situ CDT was realized through the redox of TiSe 2 in acidic TME after storage of 2 days, obtaining TiO x Se 2Àx nanoplates containing Se (0), selenate, and Ti 3þ /Ti 4þ redox couple, which achieved cascade ROS production through Ti 3þ -mediated Fenton-like reaction and Ti 4þ -mediated GSH depletion.Moreover, the adaptive immune responses were activated through Se ions generated by the in situ redox process, which could promote the maturation of DCs.To further improve the ROS generation efficiency and amplify the ROS-induced ICD effects, we constructed a Z-scheme heterojunction through depositing CDs on the surface of TiSe 2 nanoplates.The fabrication of heterojunctions not only improved the sonodynamic performance of TiSe 2 through the inhibition of electron-hole pair recombination, but also enhanced the chemodynamic activity of single component via the accelerated transfer of carriers.More importantly, the significantly amplified ROS level via in situ CDT-enhanced SDT can effectively improve antitumor immunity through the elicitation of ICD.In vitro and in vivo experiments demonstrated that CD/TiSe 2 heterojunctions achieved a synergistic therapy of in situ CDT-enhanced sonoimmunotherapy.

Figure 2 .
Figure 2. Heterojunction-enhanced sonodynamic properties of CD/TiSe 2 . 1 O 2 generation efficiency evaluation of a) CD/TiSe 2 , b) TiSe 2 , and c) CDs under US irradiation.d) Comparison of the rate constant of US-triggered 1 O 2 generation in the presence of CD/TiSe 2 , TiSe 2 , and CDs.e-h) Tauc plot and XPS-VB spectra of TiSe 2 and CDs.i) ESR spectra of CD/TiSe 2 , TiSe 2 , and CDs under US irradiation.j) Schematic illustration of the energy band diagrams of single-component TiSe 2 , CD, and the Z-scheme heterojunctions in the CD/TiSe 2 composites.

Figure 3 .
Figure 3.In situ redox of CD/TiSe 2 to excite chemodynamic activity.a) Absorption spectra change of CD/TiSe 2 incubated with H 2 O 2 in the presence of TMB.b-d) Survey XPS, high-resolution Ti 2p, and Se 3d spectra of TiSe 2 .e) Absorption spectra change of CD/TiO x Se 2Àx incubated with H 2 O 2 in the presence of TMB.f ) SAED pattern of TiSe 2 after storing for 2 days (TiO x Se 2Àx ).g) XRD pattern of TiO x Se 2Àx .h,i) High-resolution Ti 2p and Se 3d spectra of TiO x Se 2Àx .j) POD-like catalytic activity evaluation of CD/TiO x Se 2Àx and TiO x Se 2Àx at pH 4.5 determined by Michaelis-Menten kinetic analysis.k) ESR spectra of CD/TiO x Se 2Àx , TiO x Se 2Àx , and CDs incubated with H 2 O 2 .l-n) GSH depletion evaluation of CD@TiSe 2 , TiSe 2 , and CDs.o) Schematic illustration of the in situ redox of CD/TiSe 2 to excite chemodynamic activity.

Figure 4 .
Figure 4.In vitro in situ chemodynamic-enhanced SDT via CD/TiSe 2 heterojunctions.a-d) Cell viability of LO2 or 4T1 cells incubated with TiSe 2 or CD/TiSe 2 without or with US irradiation.e,f ) Live/dead cell and ROS staining of 4T1 cells after different treatments (n = 6 and ***p < 0.001).

Figure 5 .
Figure 5. CD/TiSe 2 -mediated in situ CDT and SDT to induce ICD in vitro.a) Confocal images of CRT exposure in 4T1 cells after different treatments.b) Schematic illustration of the experimental procedure of DC activation by CD/TiSe 2 .c,d) Flow cytometry analysis and the corresponding quantification results of the expression of CD80 and CD86 in DCs after different treatments.e) The quantification results of the flow cytometry analysis of CRT exposure in 4T1 cells after different treatments.f ) Extracellular HMGB1 levels of 4T1 cells after different treatments.g) Extracellular and intracellular ATP levels of 4T1 cells after different treatments.h) Schematic illustration of the effect of TiO x Se 2Àx on the maturation of DCs.i) Flow cytometry analysis of the expression of CD80 and CD86 in DCs after treating TiO 2 , TiSe 2 , or TiO x Se 2Àx at varied concentrations (n = 3; *p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 6 .
Figure 6.In vivo anticancer therapy.a) Schematic illustration of the in situ CDT-enhanced SDT via CD/TiSe 2 .b,c,e,f ) In vivo and ex vivo NIR imaging and the corresponding quantitative results of mice after intravenous injection of ICG-labeled CD/TiSe 2 .d) Representative photographs of tumors after in situ CDT-enhanced SDT via CD/TiSe 2 .g-i) Tumor growth curves, survival of mice, CRT staining of tumors in different groups (n = 5 and ***p < 0.001).

Figure 7 .
Figure 7.In vivo immune responses induced by in situ CDT and SDT combined with αPD-L1.a) Schematic illustration of the in vitro anticancer therapy of CD/TiSe 2 through in situ CDT-enhanced SDT and immunotherapy combined with αPD-L1 using bilateral tumor model.b,c) Representative photographs of the primary and distant tumors after in situ CDT-enhanced SDT and immunotherapy.d-g) Survival of mice (d), tumor growth curves of the primary (e) and distant tumors (f ), body weight (g) of mice in different groups (n = 5 and ***p < 0.001).

Figure 8 .
Figure 8.In vivo immune response evaluation.a,b) Flow cytometry analysis and the corresponding quantification results of the expression of CD80 and CD86 in the lymph nodes after different treatments.Flow cytometry analysis and the corresponding quantification results of the expression of CD4 þ T cells and CD8 þ T cells in the c-e) primary and f-h) distant tumors of mice after different treatments (n = 3; *p < 0.05, **p < 0.01, and ***p < 0.001).