Tumor‐targeted nano‐adjuvants to synergize photomediated immunotherapy enhanced antitumor immunity

Photomediated immunotherapy explored to combine the anti‐cancer effect of phototherapy with the immune enhancement ability of immunotherapy, and shown great prospects for cancer treatment strategies. However, photomediated immunotherapy triggered antitumor immunity through the release of tumor antigens and damage‐associated molecular patterns from necrotic tumor cells was not enough mighty to improve the therapeutic benefits due to the immunosuppressive tumor microenvironments. Herein, a tumor‐targeted nano‐adjuvant Fucoidan@Al(OH)3‐Poly(I:C)/IR‐820 for tumor‐targeted therapy and metastasis inhibition was designed and prepared. The intrinsic immunomodulatory effects of tumor‐targeted nano‐adjuvant and their ability to simultaneously trigger tumor antigen release, thereby reversing immunosuppression and achieving potent antitumor immunity and augmented cancer therapy, were explored. The results proved that the multifunctional nano‐adjuvant combined with photothermal, photodynamic, and immunotherapy could effectively treat breast cancer and had metastasis inhibition effect by enhancing anti‐tumor immunity through immunomodulation, it should have great application potential in the treatment of breast cancer.


INTRODUCTION
Cancer immunotherapy has made rapid progress in attacking tumor cells by training or stimulating the innate immune system of the human body and has shown great prospects for cancer treatment strategies. 1 Although cancer immunotherapy had overcome the possibility of recurrence of traditional therapy to some extent, it had a low response rate to immunotherapy for some tumors with poor immunogenicity. 2 Photomediated immunotherapy was explored to combine the anti-cancer effect of phototherapy with the immune enhancement ability of immunotherapy, which not only enhanced the anticancer effect but also reduced the adverse reactions caused by their separate use. 3 Among them, photodynamic therapy (PDT) could produce highly cytotoxic singlet oxygen ( 1 O 2 ) and cancer-killing heat through the interaction of photosensitizers, light, and oxygen, and eliminate primary tumors and metastases through photothermal ablation and immune activation. 4 It was reported in many kinds of literature that photosensitizers such as chlorin e6, indocyanine green (ICG), and CS-1 could produce anti-tumor immune effects by producing tumor-associated antigens (TAAs) from ablated tumor cell residues or enhance the release efficiency of chemotherapy drug through PDT to induce synergistic enhancement of tumor therapy. 5 This effect was also observed in a preliminary clinical trial study. 3a For example, Indocyanine Green (IR-820), a new kind of ICG, which can absorb near-infrared (NIR) light and convert the excitation light energy into heat, is considered to be a promising photosensitizer. 6 However, PDT triggered antitumor immunity by releasing tumor antigens and damage-associated molecular patterns (DAMPs) from necrotic tumor cells, which need to be strengthened to meet the benefits of clinical treatment. 7 The combination therapy of PDT with chemotherapy, photothermal therapy (PTT), or cancer immunotherapy had shown better therapeutic effects than that of traditional PDT alone and had also been frequently reported. 8 NIR light-induced PTT has shown a promising anticancer effect, and could directly or indirectly activate the immune system. Therefore, a combination of PDT with PTT has great clinical application potential in cancer immunotherapy.
However, the antitumor immune responses of cancer immunotherapy strategies was greatly limited by the immunosuppressive tumor microenvironments (TMEs) and tumor-associate antigens presentation efficiency. 9 In order to induce stronger immune responses and major antigen-specific adaptive response, the immunomodulatory effect of adjuvants (such as delivery vehicle and/or bioadjuvant) could be integrated into the nanoplatform to improve antitumor immunity. 10 Dendritic cells (DCs) are the most effective antigen presenting cells (APCs), it would migrate to lymphoid tissue after absorbed antigen and then process the captured antigen information, and plays an important role in the establishment of innate and adaptive immunity. 11 The process of antigen capture by DCs requires the participation of several secreted or transmembrane pattern recognition receptors, Toll-like receptor 3 (TLR3) can directly induce apoptosis of various tumor cells to promote tumor elimination. 11a TLR3 had the ability to mediate immune activation and tumor cell apoptosis, and was an attractive target in cancer treatment. 12 The TLR3 agonist polyinosine polycytidylic acid (poly(I:C)) is an artificial mimetic of viral dsRNA and an inducer of interferon (IFN). 13 However, in addition to the ability of Poly(I:C) to kill tumor cells directly or indirectly, interaction with intracellular TLR3 may overstimulate the immune system leading to autoimmune and chronic inflammatory diseases, the other TLRs have also limited their clinical application due to the same side effects. 14 Nanotechnology can improve the transport of small molecules and biologics at target sites, bio-distribution in vivo, and drug release patterns, therefore, nanoparticles appear to be suitable vehicles for TLRs agonist delivery. 15 Aluminum hydroxide (Al(OH) 3 ) with excellent safety profiles was the only ones licensed adjuvant in human vaccines and mainly enhanced Th2-specific immune responses but not a strong CD8 + T cell response against cancer. 5a Because nanoscale aluminum hydroxide particles have stronger adjuvant activity than traditional micron-scale aluminum hydroxide particles. 16 Recently, Sun and coworkers focused on integrating aluminum in the form of aluminum hydroxide into nanoparticles as a nanoadjuvant, and demonstrated that it could activate DCs and produce anti-tumor immune response. 5a, 17 Herein, a tumor-targeted nano-adjuvant Fucoidan@Al(OH) 3 -Poly(I:C)/IR-820 (FAPI) for tumortargeted therapy and metastasis inhibition was designed and prepared. The nano-aluminum hydroxide was prepared by the precipitation method, Poly(I:C), another cellular immune adjuvant that could enhance immune response, was bound to the surface of nano-aluminum hydroxide by substitution ligand. After that, the photosensitizer IR-820 was adsorbed by electrostatic adsorption, and finally, the tumor-targeting substance fucoidan was coated on the outermost layer ( Figure 1A). After the nano-adjuvant reaches the tumor through active targeting of fucoidan and passive targeting of nanoscale effect, under the irradiation of NIR laser, the photothermal and photodynamic effects of IR-820 lead to tumor cell death and induced immunogenic cell death (ICD), After APC endocytosis, the nano adjuvant cooperated with ICD and TAAs to induce DC maturation and enhance anti-tumor immune response. The mature DC migrated to the tumordraining lymph nodes and presented antigen information to cytotoxic T lymphocyte (CTL) cells, which enhanced the infiltration of CTL and anti-tumor cytokines into the tumor and exerted an anti-tumor effect ( Figure 1B). This work established a multifunctional nanoplatform for effective cancer therapy with immune regulation and enhanced anti-tumor immunity.

Cell lines and animals
Murine breast cancer 4T1 cell lines and mouse fibroblast L929 cell lines were obtained from China Center for Type Culture Collection (Wuhan, China) and cultured in Dulbecco's Modified Eagle Medium containing 10% fetal bovine serum (V/V) and 1% PS (V/V) at 37 • Cwith 5% CO 2 atmosphere. Healthy female BALB/c mice (6/8-week-old) were provided from the Laboratory Animal Center of the Hubei Academy of Preventive Medicine (Wuhan, China). In accordance with ethical considerations involving animal research, all procedures performed in this study were approved by the IACUC committees of Hubei University of Technology (Approval ID: 20220311). The study was conducted in compliance with the guidelines set forth by the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and the Animal Welfare Act. All efforts were made to minimize animal suffering and to reduce the number of animals used in the study.

Preparation of FAPI
Nano-aluminum hydroxide was prepared by referring to the previous method with slight modifications. 5a Briefly, an equal volume of aluminum chloride solution (0.015 M) and sodium hydroxide solution (0.04 M) were mixed, and the pH of the mixture was adjusted pH to 8.1 with NaOH solution (0.01 M), then adjusted the pH to neutral with citric acid (0.01 M). The reaction was mildly stirred in a 50 • C water bath for 2 h and finally centrifuged at 10,000 rpm/min for 15 min. The supernatant was discarded and the precipitate was washed with pure water to obtain nano-Al(OH) 3 (denoted as A). Poly(I:C) (1 mg/ml) and IR-820 (0.5 mg/ml) were sequentially added to the 1.2 mg/ml of nano-Al(OH) 3 (V:V:V = 10/30/10), the reaction was carried out under mild stirring in a water bath at 50 • C overnight. Then the mixture was centrifuged at 10,000 rpm/min for 30 min to obtain Al(OH) 3 -Poly(I:C)/IR-820 (expressed as API). The API was dispersed in water and added to 2.3 mg/ml of fucoidan solution (3 ml), the mixture was stirred for 1 h at room temperature and centrifugated at 10,000 rpm/min for 30 min, and the precipitate was washed to obtain FAPI. Nanoadjuvant (denoted as FAP) was prepared using a similar method.

Characterization of FAPI
The hydrated particle size and zeta potential of nanocarriers were measured by Zetasizer Nano-ZS (Malvern Instruments, Worcestershire, UK), and the changes in particle size and polydispersity index of FAPI in a period of time were measured to evaluate the stability of FAPI. The morphology of FAPI was characterized by a transmission electron microscope (TEM; ThermoFisher, USA). The absorbance of Poly(I:C) and IR-820 in the supernatant of FAPI suspension were measured by a UV-vis spectrophotometer (UV-1800; Shimadzu, Japan) at the wavelength of 266 and 686 nm, respectively. The concentration was calculated through the standard curve. Encapsulation Efficiency (EE) and Loading Efficiency (LE) were calculated as the reference. 15c The release behaviors of Poly(I:C) and IR-820 from FAPI were monitored using the dialysis method. FAPI solutions (1 ml) were loaded into a dialysis bag (MWCO 8000-14,000, n = 3), then added into phosphate-buffered saline (PBS) solution (0.1 M, pH 6.8) in a water bath oscillator at 37 • C. A quantitative solution was taken out at selected time intervals point and the corresponding fresh media was supplemented, the absorbance of Poly(I:C) and IR-820 was determined using a UV spectrophotometer. The amount of Poly(I:C) and IR-820 in the release medium were calculated according to the standard curve to form the release curve.
To evaluate the photothermal and photodynamic conversion capabilities of IR-820 and FAPI, IR-820 and FAPI loaded with the same concentration of IR-820 were irradiated with a NIR laser (1.5 W/cm 2 , 808 nm) for 4 min, and the real-time temperature was recorded every 10 s with an infrared thermal imager (Flir, USA). The photothermal conversion efficiency of FAPI was calculated through the heating and cooling curve of FAPI as the reference. 18 Under the NIR laser (808 nm), IR-820 and FAPI loaded with the same concentration of IR-820 was irradiated at different time, and the full wavelength scanning curve was recorded with a UV instrument to evaluate the generation of singlet oxygen ( 1 O 2 ).

Biocompatibility of nano-adjuvants
Biocompatibility of nano-adjuvant (FAP) was evaluated with L929 or 4T1 cells by MTT assay. The series concentration of FAP was added into the 96-well plate containing 1×10 4 L929 or 4T1 cells /well and the incubation was continued for 24 h. The cell viability was calculated through the absorbance (OD value) at 490 nm using a microplate reader (BioTek Epoch, USA).

Cellular uptake and tumor-targeted delivery in vitro
L929 cells or 4T1cells in the logarithmic growth phase were seeded in 24-well plates and incubated overnight at 37 • C with a 5% CO 2 atmosphere. The two cell lines were divided into two groups, one group was pretreated with fucoidan solution (concentration and dosage were the same as FAPI) for 3 h, and the other group was not treated. The medium was discarded, and the corresponding samples (IR-820, API, and FAPI) were added to the two groups to incubate for 1, 3, and 6 h. After washing with PBS, each well was fixed with 4% paraformaldehyde fixative for 10 min. The cells were observed under an inverted fluorescence microscope (Olympus, Japan) and images were collected.

Photothermal and photodynamic combined antitumor activity of FAPI in vitro
Samples from each group (control, IR-820, FAPI, IR-820 + laser, and FAPI + laser) were added to 4T1 cells, respectively. An equal volume of DCFH-DA (20 μM) was added to the cells and irradiated with an 808 nm NIR laser for 5 min. After centrifugation, the cells were suspended in PBS and observed under a fluorescence microscope to detect reactive oxygen. The step of mitochondrial membrane potential operation was similar to the above, and DCFH-DA (20 μM) was replaced with a JC-1 staining working solution (20 μM).
Antitumor activity of FAPI in vitro was evaluated using 4T1 cells. The corresponding samples (control, IR-820, AP, API, FAPI, control + laser, IR-820 + laser, AP + laser, API + laser, and FAPI + laser) were added to cells and incubation for 3 h, the laser group was irradiated with 808 nm NIR laser (5 min), then incubated for 20 h. Cells were collected and added with staining working solution (1 μl of Calcein-AM and 2 μl of PI were diluted into 2 ml of 1 × assay buffer) to incubate at 37 • C for 10 min. The samples were observed under a fluorescence microscope and images were collected. The apoptosis experiment had similar steps in sample processing and laser irradiation. After irradiating, cells in each group were collected and suspended. Annexin V-PE (5 μl) and 7-AAD (5 μl) were added to the cell suspension and incubated for 15 min at room temperature in the dark. Apoptotic cells were detected in a flow cytometer (FACS Melody, BD Biosciences, Oxford, UK), and the results were analyzed by CellQuest software (BD Biosciences, San Jose, CA).

Activation of DC by nano-adjuvant in vitro
Maturation of DC induced by nano-adjuvant in vitro was investigated using bone marrow-derived DCs (BMDCs) from 8-week-old BALB/c mice. 19 4T1 cells were pretreated with a sample of different groups (control, IR-820, FAPI, control + laser; IR-820 + laser and FAPI + laser) for 20 h, and the laser group was irradiated with 808 nm NIR laser (5 min). Afterward, 1 × 10 6 BMDCs were co-cultured with 1 × 10 5 pretreated 4T1 cells at 37 • C for 24 h. The cells were stained with APC-conjugated anti-mouse major histocompatibility complex II (MHC II, 5 μl) and FITC-conjugated anti-mouse CD86 antibody (2 μl), and the mature DCs were analyzed by flow cytometry after washing with PBS.

Immune response analysis in vitro
The distribution of calreticulin (CRT) on the 4T1 cell membrane was evaluated by immunofluorescence. Briefly, 1×10 5 4T1 cells/well in a 24-well plate were incubated with control (PBS solution), IR-820, API, and FAPI for 4 h, the laser group (control + laser; IR-820 + laser; API + laser and FAPI + laser) were irradiated with 808 nm NIR laser (1.5 W/cm 2 , 5 min), then incubated for 4 h. After fixation with 4% paraformaldehyde, the cells were blocked with 10% goat serum at room temperature and an anti-CRT antibody was added to incubate at 37 • C for 30 min. Alexafluor 488-conjugated secondary antibody was added and incubated in the dark at 37 • C for 40 min. Intracellular high mobility group box 1 (HMGB1) distribution and efflux in 4T1 cells were detected using immunofluorescence, and the steps were similar to those of CAT detection above.
Remarkably, treated cells should be permeabilized with 0.1% Triton X-100 for 10 min after fixed.

Anticancer efficacy in vivo
In order to verify the antitumor effect of FAPI in vivo, 4T1 tumor-bearing mice were established by subcutaneously injecting with 4T1 cells (1×10 6 cells) in the fourth pair of mammary fat pads on the left side of the abdomen, which was the primary tumor. On the 3rd day before treatment, 4T1 cells were transplanted on the right back near the lymph nodes as a distant tumor to stimulate metastasis.
The tumor-bearing mice with a primary tumor volume of about 100 mm 3 were randomly divided into six groups as follows (n = 5): PBS as a control group, API, FAPI, IR-820 + laser, API + laser and FAPI + laser (the concentration of nano-adjuvants in all experimental groups were calculated with IR-820 of 300 μg/ml). The first tail vein injection of the different formulations was recorded as day 0 and the first day was subjected to NIR laser irradiation (1.5 W/cm 2 , 4 min). All groups were i.v injected every other day with five total dosages, the tumor size and the body weight of mice were monitored every other day, and the treatment was terminated on the 21st day. To obtain the photothermal imaging, FAPI + laser group was given NIR laser irradiation (1.5 W/cm 2 , 4 min) after injection 12 h, and the temperature change of the tumor was recorded by infrared thermal imager (Flir, USA) per min. The volume of the tumor was calculated as follows: = × 2 × 1 2 , Which was the longest dimension; was the shortest dimension. In order to evaluate the distribution of nanoadjuvants in vivo, the tumor-bearing mice were executed 4 and 8 h after intravenous injection of different formulations (PBS, free IR-820, and FAPI), and the main organs (heart, liver, spleen, lung, and kidney) and tumors of the mice were separated. The fluorescence imaging results were detected by IVIS Lumina LT (USA, Perkin Elmer).

Histopathology examination
The main internal organs (heart, liver, spleen, lung, and kidney) of tumor-bearing mice were fixed in 4% paraformaldehyde and then dehydrated, transparent, paraffin-embedded, embedded, sectioned, and displayed for hematoxylin and eosin (H&E) staining to evaluate biological safety.

Statistical analysis
All statistical analyses of data were performed in Graph-Pad Prism (version 9). All data are presented as mean ± SD. Statistical differences were analyzed using one-way analysis of variance (ANOVA) or two-way ANOVA, ns means no significant differences, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.001 were considered significant.

Preparation and characterization of FAPI
Poly(I:C) with reactive phosphate group could be coupled to the surface of Al(OH) 3 nanoparticle through ligand exchange, which was denoted as AP. AP nanocarrier possessed relatively positive zeta potential, where the negatively charged photosensitizer IR-820 was attached via electrostatic adsorption, and the preparation process of the nanoparticle (denoted as API) was presented in Figure 1A. Finally, the fucoidan (high-affinity ligand of P-select in) was coated on the outermost layer of the nanoparticle for targeted delivery of the nano-adjuvant (denoted as FAPI). 20 The three nanoparticles presented different sizes at around ∼110 nm, ∼220 nm, and ∼160 nm, separately ( Figure 2A). The size reduction of FAPI might be due to the compression of "tail" Poly (I: C) exposed after coating with fucoidan. The image of TEM showed that the nano-adjuvant was uniform and spherical with a coreshell structure ( Figure 2B). When AP was assembled with IR820 and Fucoidan to form FAPI, zeta potential reversed from +37--45 mV ( Figure 2C). The changes in particle size and zeta potential during the preparation of the nanoadjuvant preliminarily indicated that the nano-adjuvant might be successfully prepared. The elemental analysis also verified the successful preparation of nano-adjuvant, the energy spectra of FAPI showed the characteristic of Al in Al(OH) 3 , P in Poly(I:C), IR-820, and S in fucoidan ( Figure S1). A full-wavelength UV-vis scan was performed on FAPI, Poly(I:C), and IR-820, the UV-vis scan image also indicated that the nano-adjuvant (FAPI) was successfully fabricated ( Figure 2D).
In the release medium (pH 6.8), about 63% of Poly(I:C) and about 35% of IR-820 were released from FAPI, which indicated that Poly(I:C) and IR-820 could be successfully released from FAPI in a simulated acidic TME ( Figure 2F). The EE of Poly(I:C) and IR-820 was about 61.8% and 68.6%, and the LE of Poly(I:C) and IR-820 was about 2.4% and 1.2%, respectively. FAPI showed good stability within 1 week ( Figure 2E).
In order to verify the photothermal and photodynamic performance of FAPI in vitro, the optimal power and appropriate concentration under illumination were determined with IR-820 ( Figure S2). The power density was determined to be 1.5 W/cm 2 , and the suitable concentration of IR-820 was 400 μg/ml. After IR-820 was loaded into the nanocarrier, this should not affect its photothermal and photodynamic properties ( Figure 2G-I). The photothermal effect of FAPI showed that its photothermal conversion effect increased with the concentration of IR-820 ( Figure 2G). The photothermal stability of FAPI and IR-820 indicated that the maximum temperature of IR-820 could rise to 67.5 • C and then decrease to 59.2 • C in four switching cycles, while the highest temperature of FAPI could reach 62.9 • C in 5 switching cycles and remain unchanged ( Figure 2H), which indicated that FAPI enhanced the photothermal stability of IR-820. Here, the calculated photothermal conversion efficiency of FAPI was about 23% ( Figure S3). These results showed that FAPI could be successfully triggered and produce an efficient photothermal effect under NIR laser irradiation. The ability of FAPI to generate singlet oxygen ( 1 O 2 ) was monitored, and the results showed that both FAPI and IR-820 could produce 1 O 2 . However, the photo-degradation curve indicated that since the slope of FAPI was greater than that of IR-820, the ability of FAPI to generate 1 O 2 seemed to be stronger than that of free IR-820 ( Figure 2I and Figure S4), which might be related to FAPI enhancing the photostability of IR-820.

Cellular uptake and anti-tumor effect
The ideal tumor therapy drug should have cytotoxicity to tumors but almost no cytotoxicity to normal cells. Therefore, the biocompatibility of nanocarrier FAP was observed. From the results of the cellular safety assessment of the nanocarrier FAP, it could be seen that even if the

F I G U R E 3 Evaluation of tumor targeting and antitumor effects of Fucoidan@Al(OH) 3 -Poly(I:C)/IR-820 (FAPI) in vitro. (A)
Biocompatibility of FAP (concentration was calculated as aluminum hydroxide); (B, C) Fluorescence graphs of different groups uptake by 4T1 cells at different time periods and their quantification histograms; (D, E) Fluorescence graphs and quantitative histograms of FAPI uptake by 4T1 cells and L929 cells at different times (blue fluorescence was DAPI-stained nuclei, red fluorescence was Indocyanine Green (IR-820), and the right graph was the corresponding merge graph, mean ± SD, n = 3, Scale bar: 75 μm); F: the cell viability of 4T1 treated different concentration FAPI with or without laser; (G, H) Fluorescence graphs and quantitative histograms of apoptosis experiment with or without laser (green fluorescence was a live cell, red fluorescence was a dead cell, Scale bar: 200 μm); (ns no significant difference, **p < 0.01, ***p < 0.001, ****p <0.0001).
concentration of nano-adjuvant carrier FAP (calculated with Al(OH) 3 ) was as high as 25 mM, FAP had no cytotoxicity to L929 cells, but it had obviously toxicity to mouse breast cancer cell (4T1) ( Figure 3A), indicating that FAP as nanocarrier had good biosafety and could be used for antitumor therapy in vivo. The first step of antitumor drug therapy was to deliver the therapeutic drug to the tumors, so the cellular uptake of active molecules often directly affected the antitumor effect. Here, the cellular uptake of nano-adjuvant was monitored, and the results showed that IR-820, API, and FAPI could be intaken by 4T1 cells in a time-dependent manner. The small molecule IR-820 was rapidly taken up to the maximum value within 3 h and was higher than API and FAPI, the uptake amount might depend on the molecular size, and smaller molecules tended to have more uptake. However, the uptake of FAPI by 4T1 was significantly higher than that of IR-820 and API at 6 h due to good tumor targeting and controlled release of FAPI ( Figure 3B,C). In order to verify the tumor targeting of FAPI, FAPI uptake was conducted after 4T1 cells pre-incubation with or without fucoidan. There was no sig-nificant difference between the uptake of IR-820 and API in the 4T1 cells pretreated with fucoidan, while the uptake of FAPI in 4T1 cells without pretreatment with fucoidan was significantly higher than that in other groups ( Figure  S5). In order to verify whether the targeting ability of FAPI was only for tumor cells, L929 was used to evaluate, the results showed that the uptake of FAPI by L929 was also time-dependent, but its uptake capacity was significantly lower than 4T1 ( Figure 3D,E), which ruled out the targeting ability of FAPI to normal cells. Therefore, these results indicated that FAPI had the ability to target 4T1 tumor cells.
The results of cytotoxicity in vitro showed that the antitumor effect of FAPI was positively correlated with the time of NIR laser irradiation and power density during irradiation ( Figure S6), and the cell viability gradually decreased with the increase of FAPI concentration ( Figure 3F). As shown in Figure 3G, in the absence of NIR laser irradiation, except for a few dead cells in the FAPI group, few dead cells were observed in other experimental groups. Dead cells appeared in IR-820, API, and FAPI and Poly(I:C) composition (n/n) (mean ± SD, n = 3, ns no significant difference, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). groups irradiated by the NIR laser, but the number of dead cells in the FAPI group was significantly higher than that in the other groups. The MTT assay also showed the same results, which might benefit from the good photothermal and photodynamic effects of IR-820 ( Figure 3H).
After confirming the good anti-tumor effect of FAPI, its anti-tumor mechanism was explored in vitro. Singlet oxygen ( 1 O 2 ) had highly cytotoxic through the interaction of photosensitizers, light, and oxygen, and could be used for tumor therapy. 3 Here, we evaluated the ability of cells treated with FAPI to produce 1 O 2 , the results showed that IR-820 and FAPI could only produce a small amount of 1 O 2 without NIR laser irradiation. After NIR irradiation, the amount of 1 O 2 production increased significantly, and the capacity of FAPI to produce 1 O 2 was significantly stronger than that of IR-820 ( Figure 4A,B). This might be due to the tumor targeting of FAPI, which resulted in the uptake of FAPI by 4T1 cells higher than that of free IR-820 ( Figure 4B-E). In addition, the results of flow cytometry showed that FAPI could significantly induce more apoptosis under NIR irradiation ( Figure S7), which might benefit from the tumor-targeting ability of FAPI and the good photothermal and photodynamic capabilities of IR-820. Mitochondria, as the energy provider of cells, were of great significance to the survival of cells. After mitochondria are damaged, their membrane potential would change. 21 Here, the state of mitochondria was observed by the JC-1 fluorescence change caused by the change of mitochondrial membrane potential. The results showed that FAPI could significantly induce mitochondrial damage under the irradiation of NIR laser, and the damage degree was significantly higher than other experimental groups and the FAPI group without NIR laser irradiation ( Figure 4C).

Immunostimulatory activity in vitro
It was reported that PDT/PTT could induce significant ICD in tumor tissues. 21 In order to evaluate the ability of FAPI to induce ICD in vitro, the release of ICD markers CRT and HMGB1 was examined. The results showed that FAPI significantly promoted the expression of CRT and released HMGB1 compared with other experimental groups under irradiated NIR laser. However, there was no significant difference between the FAPI group and other experimental groups without NIR laser irradiation ( Figure 4D,E and Figure S8). These ICD markers could promote the uptake, processing, and presentation of tumor antigens by DCs, and ultimately promote the generation of CTLs. 22 Activation of APCs was important for inducing T cell-mediated immune responses. 23 As one of the most important APCs, DCs played a crucial role in the initiation and regulation of innate and adaptive immunity. Immature DCs would engulf exposed antigens or ICD markers when they encountered them, migrated to the vicinity of lymph nodes, and transformed into mature DCs. Therefore, DCs maturation induced by FAPI-mediated 4T1 cells was tested. The results showed that 4T1 cells treated with IR-820 could not induce more DCs maturation with NIR laser irradiation, but the induction rate of DCs maturation in the FAPI group with or without laser irradiation was much higher than that in other experimental groups ( Figure 4F and Figure S9). It was reported that both Poly(I:C) and Al(OH) 3 as adjuvants could effectively induce the maturation of mouse and human DCs, and subsequently induce cross-presentation of different antigens. 24 The combined application of Poly(I:C) and nano-Al(OH) 3 double adjuvant also verified the synergistic effect of induction DCs. As shown in Figure 4G, AP significantly enhanced the maturation rate of DCs, and AP coated with fucoidan had no significant effect on DC maturation compared with the AP group ( Figure S9). FAPI, as a nano-adjuvant, should present the excellent function of immune adjuvant due to the multiple synergies effects among TAAs, ICD markers, Poly(I:C), and nano-Al(OH) 3 . Recently, some literature focused on designing adjuvant combinations to improve immune response. 17 For example, Sun et al. highlighted the potential of combining multiple adjuvants (aluminum hydroxide nanoparticles and TLR9 agonist (CpG-ODN)) for effective vaccine design, 17 and the combination of NOD2 ligand MDP and TLR7 ligand R848 synergistically stimulated BMDCs. Here, the combined application of Poly(I:C) and nano-Al(OH) 3 double adjuvant also verified the synergistic effect of induction DCs, and the optimal ratio of nano-Al(OH) 3 and Poly(I:C) was selected, the nano-adjuvant FPA showed the strongest induction rate of maturation DCs ( Figure 4H and Figure S9).

In vivo anti-tumor effect and photothermal imaging
Inspired by the significant immune response of the neoadjuvant in vitro, the anticancer efficacy of neoadjuvant combined with photothermal and PDT on 4T1 tumor-bearing mice was evaluated. The schedule of administration and laser irradiation treatment in a bilateral tumor model was illustrated in Figure 5A, the good anti-tumor effect of the nano-adjuvant was also confirmed by the tumor photos of different treatment groups and tumor growth curve during the treatment cycle ( Figure 5B,C and Figure S10). The primary tumor size of the FAPI group with laser irradiation was significantly smaller than that of the other experimental groups, and the tumor inhibition rate was as high as ∼93% ( Figure S11). Tunnel staining fluorescence of paraffin sections of primary tumor tissue also showed the same results ( Figure S12). Interestingly, FAPI without laser irradiation had a certain therapeutic effect as IR-820 with laser irradiation, because Poly(I:C) and Al(OH) 3 nanoparticles could induce Th1-type cellular immunity and rapidly stimulate cytokines such as TNF-α and IFN-γ production to exert anti-tumor effects. 12b,21 Importantly, in the absence of any direct treatment, compared with other control groups, the distant tumor growth of FAPI + laser group mice was significantly inhibited and the tumor inhibition rate reached ∼65% ( Figure 5C and Figure S11), which indicated the potential inhibition of tumor metastasis based on the immune response.
The photothermal imaging of the nano-adjuvant was evaluated, and the results showed that the temperature of mice in the control group could only rise to 40.1 • C within 4 min of NIR laser irradiation, while that in the IR-820 group could rise to 47.1 • C, and that in the API group could rise to 48 • C. This might be due to the rapid metabolism and excretion of IR-820 in vivo, resulting in the reduction of IR-820 accumulation in tumors. The FAPI group could increase to 50.8 • C within 4 min after NIR laser irradiation, which might be due to the tumor targeting of FAPI to make more IR-820 accumulate in tumors ( Figure 5D,E). When the temperature reached over 46 • C, cells would apoptosis rapidly due to microvascular thrombosis and ischemia, 4 so FAPI could more easily ablate tumor cells under laser irradiation. The distribution of nano-adjuvant in mice also verified that FAPI had good tumor targeting in vivo ( Figure 5E). In the main organs and tumors of PBS as the control group, no fluorescence was observed, while in the IR-820 group, fluorescence was observed in the main organs and tumors at 4 h, and the fluorescence intensity was mainly concentrated in the liver. After 8 h, the fluorescence intensity of the tumor was enhanced, but it was still mainly concentrated in the liver. In the FAPI group, the fluorescence intensity in the tumor was stronger than that in the IR-820 group at 4 and 8 h, which indicated that FAPI had good tumor targeting in vivo, although most of them were still distributed in the liver. In addition, a reduction of aggregation in the liver and kidney could be observed at 8 h, which indicated that FAPI could be excreted through the liver and kidney, and also preliminarily indicated that FAPI had good biological safety ( Figure 5E). The results of Ki67 fluorescence staining used to evaluate the cell proliferation activity showed that the tumors of the control group mice had strong activity, the API group was relatively weaker, and the FAPI + laser group was the weakest, which suggested that the FAPI + laser group had better tumor treatment effect ( Figure 5F).
Metastasis was still the main problem in TNBC, and more than 90% of patients died due to metastasis, traditional treatments such as surgery, chemotherapy, and radiation often failed to inhibit tumor metastasis. 25 Immune activation played a certain role in suppressing tumor metastasis and had attracted much attention. 22 As the most important immune organ, the thymus was of great significance to the immune system. The calculation results of the organ coefficient of the thymus showed that the FAPI + laser group was significantly higher than the other groups ( Figure 5G), this preliminarily showed that the enhanced immune response in tumor-bearing mice might have an inhibitory effect on metastasis ( Figure 5C). During the process of tumor metastasis, TME would induce the increase of adhesion factors (such as selectin) to help tumor cells smooth metastasis. 26 The high affinity of fucoidan to P-selectin made FAPI possible to target metastatic tumor cells expressing high-level P-selectin.
Lung capillaries were dense and rich in oxygen, which could facilitate the metastasis of tumor cells, so the lung was usually the most common organ for tumor metastasis. A large number of alveolar macrophages could also secrete a variety of cytokines to promote tumor cell motility, invasion, and angiogenesis. 27 The statistics of lung metastases showed that the FAPI + laser group was significantly lower than the rest of the experimental groups, which might be the result of the combined effect of immunity and fucoidan. The number of lung metastases in the FAPI without laser irradiation group was also relatively small, which might be the fucoidanmediated inhibitory effect. In addition, the number of lung metastatic nodules in the IR-820 and API + laser group was higher than that in the FAPI group ( Figure 5H and Figure S13), indicating the important role of fucoidan in targeting metastatic tumor cells. Combined with the fact that the distal tumor in the FAPI + laser group was significantly smaller than the other experimental groups ( Figure 5C and Figure S10), it could be concluded that FAPI had the effect of inhibiting tumor metastasis, and the inhibition effect was more obvious after adding NIR laser irradiation.
Moreover, the anti-tumor immune memory effect of FAPI-based photothermal and photodynamic-induced tumor ablation was evaluated. According to effector function, proliferative capacity, and migratory potential, Tem are divided into Tcm and effect or Tem. 2 Tcm mainly existed in secondary lymphoid tissues and provided protection after antigen-stimulated clonal expansion, differentiation, and transport, but Tem located in lymphoid and non-lymphoid tissues could elicit immediate protection through the production of cytokines such as IFN-γ. 28 The detection results of Tcm and Tem showed that FAPI could significantly induce the proliferation of Tcm and Tem under NIR laser irradiation. Although there was no significant difference in the induction of Tcm cells between the FAPI + laser and API + laser groups, the Tem cells in the FAPI + laser group were significantly higher than that in the API + laser group ( Figure 5I and Figure S19). The results suggested that FAPI could induce long-term immune memory after NIR laser irradiation and rapidly exert inhibitory effects upon re-exposure to tumor antigens.
Finally, the biological safety of each experimental group in vivo was evaluated. From the body weight change of each experimental group during the treatment period, it could be seen that, except for the control group, the final body weight of the mice in the experimental groups had increased ( Figure S14). The results of serum detection of liver toxicity indicators aspartate aminotransferase and alanine aminotransferase, the calculation results of the organ coefficients, and the results of H&E staining of the main organs of mice, heart, liver, spleen, lung, and kidney all showed that FAPI had no obvious toxicity to mice ( Figures S15-17).

Mechanisms of enhanced immune response
The immunofluorescence staining results of CRT, one of the ICD markers, showed that IR-820, API, and FAPI could induce significant CRT exposure under irradiation, and the FAPI + laser group was more significant (Figure 6A,B), this was consistent with the detection results at the cellular level ( Figure 4D and Figure S8), which indicated that tumor cell death occurred in these experimental groups. Next, DCs maturation in vivo induced by DAMP synergistically with nano-adjuvants was examined. The results showed that the DCs maturation induced by the FAPI + laser group was significantly higher than that of the other groups ( Figure 6C and Figure S18), which was also consistent with the detection results at the cellular level ( Figure 4F-H and Figure S9).
Natural Killer (NK) cells are one of the cytotoxic members of the innate lymphocyte family, they can rapidly kill tumor cells without pre-activation. 29 Therefore, the NK cells infiltration in the tumor was detected, and the results showed that the degree of NK cells infiltration in each experimental group was inversely proportional to the size of the primary tumor ( Figure 6D, Figure S20, and Figure 5B), and the FAPI + laser group was significantly higher than the other groups. However, the overall degree of NK cell infiltration in each experimental group was low, which was consistent with the relatively rare phenomenon of NK cell infiltration in most solid tumors. 30 The results suggested that NK cells might not be the most important cytotoxic cell in anti-tumor. Mature DCs presented antigen information to cytotoxic CD8 + T cells through MHC I and induced CTL to kill tumor cells. 31 The detection results of CD4 + CD8 + T cells in the tumor showed that the CD4 + CD8 + T cells in the FAPI + laser group were significantly higher than those in the other groups, and there was no significant difference between FAPI and IR-820 + laser, this might be one of the reasons that there was no significant difference in tumor suppressive effect between the two groups ( Figure 6E, Figure S21, and Figure 5B). The detected results of perforin and granzyme B secreted by CD8 + T cells showed consistent results with the above ( Figure 6F,G and Figures S22-23). The detection results of IFN-γ and TNF-α showed that the FAPI + laser group had the strongest effect of promoting the secretion of two cytokines, the API group with poor tumor suppressive effect showed a similar IFN-γ secretion-promoting effect as the IR-820 + laser group, which might be that Poly(I:C) was an IFN-γ inducer ( Figure 6H,I).
Tregs maintained peripheral tolerance and suppress immunity to promote tumor progression. 2 The detection results of Tregs cells in the tumor showed that the Tregs cells in each experimental group were significantly less than those in the control group ( Figure S24). Although there were no significant differences in the levels of immunosuppressive Tregs cells and tumor suppressor effector T cells (Teff) between the experimental groups ( Figure S25), there was a significant difference in the ratio F I G U R E 6 In vivo antitumor mechanism. (A) Fluorescence detection of calreticulin (CRT) released at the tumor site (blue was DAPI-stained nuclei, green was secondary antibody-stained CRT, Scare bar: 50 μm); (B) CRT fluorescence quantification results; Flow cytometry results of dendritic cell (DC) maturation (C), NK cells proliferation (D), cytotoxic T lymphocyte (CTL) cells proliferation (E), perforin release (F), granzyme B release (G), enhancement of interferon-γ (IFN-γ) release (H), tumor necrosis factor-alpha (TNF-α) release (I), Teff/Treg ratio (J), and inhibition of myeloid-derived suppressor cells (MDSCs) (K) after treatment (mean ± SD, n = 3, ns no significant difference, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).
of Teff/Tregs. This indicated that the final effect was the tumor growth inhibition effect, and the effect of the FAPI + laser group was more significant ( Figure 6J). MDSCs were a class of cell populations with immunosuppressive functions. Various cytokines produced by tumor cells could induce the proliferation of MDSCs. Therefore, avoiding the immunosuppressive effects of MDSCs might improve the host's antitumor response. 21 From the detection results of MDSCs in the tumor, it could be seen that the inhibitory effect of the FAPI + laser group on MDSCs was significantly stronger than that of other groups ( Figure 6K and Figure S26).

CONCLUSION
In conclusion, a tumor-targeted nano-adjuvant loaded with photosensitizing molecules was successfully prepared for enhancing antitumor immunity. FAPI could release Poly(I:C) and IR-820 after being taken up by tumor cells, and could significantly induce DC maturation. In the 4T1 breast cancer model of BALB/c mice, FAPI was delivered to tumors after active targeting of P-selectin by fucoidan and passive targeting of enhanced permeability and retention effect. In addition, the photothermal and photodynamic effects of IR-820 in FAPI could lead to tumor cell death after NIR laser irradiation, and nano-Al(OH) 3 and Poly(I:C) synergistically promoted DC maturation, then mature DCs presented TAAs to T cells to enhance tumor infiltration by cytotoxic T lymphocytes. At the same time, Poly(I:C) induced IFN-γ secretion to remodel the tumor suppressive effect. The synergistic effect of double adjuvant and photosensitizer could decrease the proliferation and secretion of tumor suppressor cells and cytokines, and inhibit immunosuppressive cells. The nano-adjuvant also had good biological safety and metastasis inhibition effect. In conclusion, photothermal and PDT based on FAPI had good tumor treatment and metastasis inhibition effects and had great application potential in breast cancer treatment.

C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare no conflicts of interest.