STING-activating dendritic cell-targeted nanovaccines that evoke potent antigen cross-presentation for cancer immunotherapy

Abstract

Recently, nanovaccine-based immunotherapy has been robustly investigated due to its potential in governing the immune response and generating long-term protective immunity. However, the presentation of a tumor peptidemajor histocompatibility complex to T lymphocytes is still a challenge that needs to be addressed for eliciting potent antitumor immunity. Type 1 conventional dendritic cell (cDCl) subset is of particular interest due to its pivotal contribution in the cross-presentation of exogenous antigens to CD8+ T cells. Here, the DC-derived nanovaccine (denoted as Sİ9GM) selectively targets cDCls with marginal loss of premature antigen release for effective stimulator of interferon genes (STING)-mediated antigen cross-presentation. Bone marrow dendritic cell (BMDC)-derived membranes, conjugated to cDCl-specific antibody (aCLEC9A) and binding to tumor peptide (OVA257-264), are coated onto dendrimer-like polyethylenimine (PEI)-grafted silica nanoparticles. Distinct molecular weight-cargos (aCLEC9A-OVA257-264 conjugates and 2'3'-cGAMP STING agonists) are loaded in hierarchical center-radial pores that enables lysosome escape for potent antigen-cross presentation and activates interferon type I, respectively. Impressively, Sİ9GM vaccination leads to the upregulation of cytotoxic T cells, a reduction in tumor regulatory T cells (Tregs), M1/M2 macrophage polarization, and immune response that synergizes with ocPD-l immune checkpoint blockade. This nanovaccine fulfills a dual role for both direct T cell activation as an artificial antigen-presenting cell and DC subset maturation, indicating its utility in clinical therapy and precision medicine.

Headnote

ARTICLE INFO

Keywords:

DC-based nanovaccines

Artificial antigen-presenting cells

Type 1 conventional dendritic cells

STING pathway activation

Antigen cross-presentation

ABSTRACT

Recently, nanovaccine-based immunotherapy has been robustly investigated due to its potential in governing the immune response and generating long-term protective immunity. However, the presentation of a tumor peptidemajor histocompatibility complex to T lymphocytes is still a challenge that needs to be addressed for eliciting potent antitumor immunity. Type 1 conventional dendritic cell (cDCl) subset is of particular interest due to its pivotal contribution in the cross-presentation of exogenous antigens to CD8+ T cells. Here, the DC-derived nanovaccine (denoted as Sİ9GM) selectively targets cDCls with marginal loss of premature antigen release for effective stimulator of interferon genes (STING)-mediated antigen cross-presentation. Bone marrow dendritic cell (BMDC)-derived membranes, conjugated to cDCl-specific antibody (aCLEC9A) and binding to tumor peptide (OVA257-264), are coated onto dendrimer-like polyethylenimine (PEI)-grafted silica nanoparticles. Distinct molecular weight-cargos (aCLEC9A-OVA257-264 conjugates and 2'3'-cGAMP STING agonists) are loaded in hierarchical center-radial pores that enables lysosome escape for potent antigen-cross presentation and activates interferon type I, respectively. Impressively, Sİ9GM vaccination leads to the upregulation of cytotoxic T cells, a reduction in tumor regulatory T cells (Tregs), M1/M2 macrophage polarization, and immune response that synergizes with ocPD-l immune checkpoint blockade. This nanovaccine fulfills a dual role for both direct T cell activation as an artificial antigen-presenting cell and DC subset maturation, indicating its utility in clinical therapy and precision medicine.

1. Introduction

Immune surveillance is a monitoring process by which the immune system recognizes and eliminates danger signals. When facing evasion by cancer cells, the immune system loses its ability for tumor eradication and function repair. Thus, cancer immunotherapy with recent advances in chimeric antigen receptor (CAR) T cell-therapy [1,2], cancer vaccination [3,4], and immune checkpoint blockade (ICB) [5-7] is considered a cornerstone in cancer treatment. Nanotechnology related to cell membrane-coated nanoparticles (NPs) has the potential to offer an extreme breakthrough in anticancer therapy [8], wherein biomimetic cell membranes expressing immune cell host-derived proteins with preserved function on the outer surface to enable prolonged nanoparticle blood circulation [9], leading to remarkable potential in recognizing antigens for selective targeting [10], gradual release of cargo [11], and enhanced biocompatibility in vivo [12,13]. Dendritic cells (DCs) are overwhelmingly important for innate immunity among various immune cells involved in antigen presentation [14]. DC membranes primarily express major histocompatibility complexes (MHCs) as the key membrane proteins responsible for the communication between DCs and other immune cells like T cells, natural killer cells (NK cells), and В cells [15-17]. Rather, DC membrane-coated NPs play as artificial antigen-presenting cells (APCs) in the presence of tumor antigens, facilitating migration to lymph nodes (LNs) for T cell interaction, promoting potent anticancer immune responses [4,18-21].

Recently, the stimulator of interferon genes (STING) has been applied in preclinical anticancer therapy that is attributed to the good detection of pathogen utilizing STING as a danger sensor. However, the clinical data showed a limited efficacy in patients due to the fast clearance, poor membrane permeability, hydrophilic nature, and small molecular size of the free STING agonist [22,23]. Nanoparticle-based vaccines or nanovaccines (NVs) emerged as a promising strategy in cancer immunotherapy, with advanced focus on the со-delivery of antigens and adjuvants to address these challenges. Through STING pathway activation, type I interferon production drives multifaceted immune function in activating DC maturation and enhancing cross-presentation of tumor antigens for T cell-mediated immunity [24-28]. Given that STING is widely presented in various immune cells and cancer cells, selective STING activation in target cells plays a pivotal role in cancer immunotherapy. For example, activating STING pathway in T cells could both hamper T cell proliferation and cause apoptosis to T cells [29]. Besides that, STING activation in cancer cells enables CD8+ T cell response but does not directly support antitumor efficacy in many cancer types [30-32]. Despite the huge expression of STING protein in DCs, the investigation on selective DC activation following STING pathway activation is limited. Therefore, the selective stimulation of a STING agonist to a specific immune cell like DC should be considered. Recently, type 1 conventional dendritic cells (cDCl), a dendritic cell subpopulation, have been identified as promoting innate and adaptive immune response [33]. cDCls adept exceptional abilities in presenting both endogenous and exogenous antigens to T cells, promoting T cell activation and proliferation for empowering anticancer therapy. This distinctive role of cDCls holds great potential, particularly in nanovaccine-based therapies, wherein cDCls capture antigens originating from nanovaccines and present them to CD8+ T cells after homing to tumor-draining lymph nodes (tdLNs) or infiltrating to the tumor microenvironment (TME). Therefore, selective antigen cross-presentation in cDCls promises great attention in cancer immunotherapy.

CLEC9A is a C-type lectin endocytosis receptor expressed on cDCl cells, responsible for the endocytic uptake of antigens and initiating subsequent antigen cross-presentation on CD8+ T cells [34]. The endocytosed antigen-CLEC9A complex is non-localized with the lysosomal compartment, facilitating the effective retrieval of antigens for cross-presentation and enhancing cancer immunotherapy effectiveness [35,36]. Furthermore, CLEC9A-targeted antigens have been shown to elicit more robust immune responses compared to antigens targeted to another receptor on cDCls (DEC205) [37]. Therefore, the selective activation of STING in cDCls by CLEC9A-targeted nanovaccines holds tremendous potential for advancing research in cancer immunotherapy.

Herein, we report a bone marrow DC-based nanovaccine (denoted as Sİ9GM) formulated from antigen peptide-expressed bone marrowderived dendritic cell (BMDC) membrane engulfing center-radial large pore silica nanoparticle for the effective transport of aCLEC9A-antigen conjugate and STING agonist (2/3/-cGAMP) (Scheme 1, and Fig. 1A). BMDCs were isolated and stimulated in vitro, following engulfed CD8+ T cell-specific antigen fragments (OVA257-264) to obtain an antigen peptide-presented determinant (denoted as M). CLEC9A antibody conjugated OVA257-264 fragments and 2z3z-cGAMP were loaded in the centered-radial large pores of dendritic silica NPs (denoted as Sİ9G) for their со-delivery and protection from enzymatic degradation. The nanovaccine served a dual role as an artificial APC for T cell activation and an activator for DC maturation, especially in indispensable cDCl subset. Notably, the released antigen-CLEC9A conjugate and 2z3z-cGAMP played a critical role in cDCl with cross antigen presentation for the former and type IIFN production for the latter. The LN homing ability of the nano vaccines identified by the extensive accumulation of NVs (rather than those by free STING agonist injection) prompted us to study its competency in generating antitumor efficacy. We also proved that the combination with aPD-1 blockade could induce a robust tumor growth inhibition and metastasis prevention. This promising therapeutic efficacy could provide the multifaceted functions of DC-based nanovaccines for personalized cancer immunotherapy and precision medicine.

2. Results and discussion

2.1. Formulation of BMDC-based nanovaccines Sİ9GM

To carry therapeutic antibodies and other high molecular weight molecules around 150 kDa, variable delivery systems like liposomes [38], microparticles [39], PLGA [40] have been reported. However, using inorganic NPs to deliver high molecular weight (M.W.) molecules is still challenging. In this work, we leveraged new dendrimer-like silica NPs with hierarchical pores (DHPSi NPs) in which special center-radial large pore size formed gradually from the particle center to its outer surface [41]. This distinctive morphology endows NPs with the super-high loading capacity of large molecules and tailors the cargo release. The combination of pores with multi-scale structures supports the co-loading of encapsulated molecules with different sizes through porous matrices. DHPS1-NH2 NPs were synthesized using the ethyl ether emulsion approach (Fig. SI). After purification, DHPS1-NH2 NPs were formulated with high homogeneity, large center-radial dendritic pores and small size (<200 nm) (Fig. IB). The size of NPs and surface zeta potential (zeta P) were determined utilizing a Zetasizer Nano ZS from the Malvern instrument (Fig. 1C and D). For effective loading of a highly hydrophilic small molecule such as 2z3 -cGAMP (denoted as G), we modified the amine functional groups to carboxylic groups, followed by conjugation with low M.W. (M.W. = 0.8 kDa) branched polyethylenimine (PEI) to generate DHPSi-PEI NPs. This facile method of introducing branched low-molecular-weight positively charged polymer provided nanoparticles with excellent biocompatibility, minimal toxicity, and high colloidal stability. Notably, it favored the encapsulation efficiency (EE) of anionic 2z3 -cGAMP into DHPSi-PEI NPs through electrostatic binding, achieving a 53 % of EE compared to that of DHPS1-NH2 (9 %) (Fig. S2). The successful modification of surface functional groups was evaluated by the measurement of zeta P, in which zeta P values of DHPSi-NH2, DHPSi-COOH and DHPSi-PEI were +24.2 ±4.10 mV, -32.1 ± 6.46 mV, and ±36.5 ± 6.47 mV, respectively (Fig. ID, Fig. S3). Furthermore, characterization by Fourier-transform infrared spectroscopy (FTIR) reaffirmed the functional groups of the surface-modified DHPSi NPs. The presence of N-H bending vibration and C-N stretching vibration, respectively at 1561 cm-1 and 1402 cm-1, highlighted the transformation from -NH2 groups into -COOH groups. The enhancing bands at 1464 cm-1, 2856 cm-1, and 2954 cm-1 indicated the PEI conjugation (Fig. S4).

DC-derived membranes were prepared by stimulation of primary cells from mouse bone marrow. On day 8 post-stimulation, non-adherent BMDCs were collected and characterized by flow cytometric analysis to characterize the population of BMDCs (CDllc+ F4/80"). We obtained more than 84 % CDllc+ F4/80" DCs, confirming the high purity of BMDCs (Fig. S5A and Fig. S6A). Mature BMDCs were obtained through lipopolysaccharide (LPS) activation, and mature DC biomarkers were characterized by flow cytometry (Fig. S5B and C, Fig. S6B and C). After extraction, the purity and preservation of immune biomarkers in BMDC membranes (referred to as BMDCm) were verified. Specifically, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out on both BMDC lysates and membranes to confirm protein integrity. As shown in Fig. IE, comparable bands were detected in both BMDC lysates and BMDCm, suggesting that the integrity of the membranes was effectively maintained throughout the extraction and storage processes. BMDCs were activated by LPS to maximize the expression of membrane-located co-stimulatory molecules [42]. For verification, the retention of critical DC membrane proteins (including CDllc, CD86, and CD40) on the BMDC surface was further confirmed by Coomassie blue staining. As presented in Fig. IF, protein bands of CDllc, CD86, and CD40 were only found in BMDCm but not in BMDC lysates, confirming the purity of extracted BMDCm containing well-retained immune biomarkers. Furthermore, the proteins extracted from the obtained BMDCs lysates and membranes were analyzed with Western blot assays for a series of protein markers. The data verified the successful extraction of BMDCm whereby transmembrane glycoprotein (Pan Cadherins) and essential plasma membrane enzyme (Na+/K+-ATPase) were well-preserved after extraction and storage. The study also examined the presence of representative intracellular proteins, such as Cytochrome C oxidase subunit 4 isoform 1 (COX IV), Histone H3, and ß-actin, which are found in the mitochondria, nucleus, and cytosol, respectively. The lack of Histone H3 and COX IV in BMDCm confirmed the effective separation between membranes and lysates, indicating minimal risk of carcinogenesis (Fig. 1G) [33].

To enhance the T cell-stimulatory capacity of DC-based nanovaccines, BMDCm needs to present antigen-expressed MHC I complexes. According to previous research, peptide fragments robustly bind to MHC I molecules at acidic pH (5.0-5.5) compared to neutral pH [43]. In addition, the peptide-МНС class I complex was stabilized at neutral pH (Fig. S7) [18]. Therefore, OVA257-264 peptides were loaded onto MHC class I molecules of BMDCm by a pH-dependent exchange reaction to generate OVA257-264-loaded BMDCm (denoted as DC-OVA257-264), and then DC-OVA257-264 were conjugated with aCLEC9A using thiol-maleimide chemistry (Fig. S8). For verification of the successful loading of OVA257-264 antigen fragments on BMDCm, 5-Carboxyfluorescein-conjugated OVA257-264 (FAM-0VA257-264) was used as a replacement for OVA257-264> and the expression of FAM-OVA257-264 was characterized by flow cytometry and the absorbance measurement of naïve BMDCm and DC-OVA257-264 at a wavelength of 495 nm (Fig. 1H and I). The data showed an increase in absorbance of DC-OVA257-264 compared to BMDCm, confirming the loading of OVA257-264 on BMDCm by a pH-dependent reaction (Fig. II). Furthermore, the presence of antigen peptides on DC-OVA257-264 was confirmed by Western blot assay (Fig. IF). In addition to antigen fragments, the expression of MHC I molecules required for antigen presentation was also analyzed by flow cytometry. Likewise, the increased fluorescent intensity of BMDCm stained by (Alexa fluor 647) AF647-MHCI antibody compared to control sample (unstained BMDCm) after long-term storage emphasized the availability of MHC I molecules for antigen presentation on BMDCm and the well-retained surface biomarkers (Fig. 1J). For verification of the successful conjugation of «CLEC9A, DC-OVA257-264 nanovesicles were incubated with gold-labeled IgG antibodies and then characterized by TEM-supplemented elemental mapping (EDS) (Fig. IK and L). As shown in Fig. IL, the peaks of Au elements confirmed the presence of conjugated aCLEC9A on DC-OVA257-264 nanovesicles.

The coating of BMDCm onto DHPSi NPs was characterized by ТЕМ and SEM images, verifying the well-coated NPs and the dendritic structure of BMDC membranes (yellow arrows) (Fig. IB). The colloidal stability of Sİ9GM was evaluated in Fig. S9A. Furthermore, the release profile of 2'3'-cGAMP in different pH was investigated by highperformance liquid chromatography (HPLC) (Fig. S9B). Impressively, the presence of DC-derived projections (yellow arrows) highlighted the unique feature of BMDCm-coated NPs, confirming the mild membrane extraction process for maintaining the surface biomarkers for immune responses. The zeta potential measurement further confirmed the wellcoated modified BMDCm onto DHPSi NPs (Fig. IM). The EDS mapping data showed the presence of various elements such as Si, O, N, P, and S in Sİ9GM NPs, reaffirming the successful coating of BMDCm onto Sİ9G NPs. Notably, the element mapping of P and S represents the BMDC membranes and CLEC9A antibodies, respectively (Fig. IN).

2.2. Sİ9GM NPs enhanced cellular uptake of 2'3'-cGAMP -in DCs

For verification of cellular uptake of BMDC-derived nanovaccines in DCs, we utilized bio-TEM measurement on BMDCs. As shown in Fig. 2A, we observed that Si9GMs (yellow arrows) internalized into BMDCs after 8 h of incubation. The uptake of Sİ9GM in BMDCs was further confirmed by confocal laser scanning microscopy (CLSM), in which BMDCs and Si9GMs were stained by neuro-DiO (green) and Dil (red), respectively. The presence of overlapping areas (yellow color) in the merged image confirmed the internalization of NVs into BMDCs (Fig. 2B). Similar results were obtained in the DC2.4 cell line (Fig. S10). Importantly, the targeting capacity of Sİ9GM to cDCl cells was studied by CLEC9A receptor blocking assay (Fig. Sil A). As shown in Fig. SUB, the uptake of Sİ9GM in cDCls (54.5 ±2.15 %) is higher than in CLEC9A-blocking cDCls (44.1 ± 0.36 %). Meanwhile, the uptake of SiGM nanovaccines (without CLEC9A-targeting BMDC membrane coating) in non-CLEC9A blocking cDCls and CELC9A blocking cDCls is similar (44.0 ± 1.87 % and 42.1 ± 1.63 %, respectively). Therefore, it is apparent that Sİ9GM presented better uptake in cDCls compared to that of SiGM. Furthermore, we investigated the delivery of aCLEC9A-OVA257-264 and 2'3'cGAMP by DHPSi-PEI NPs into cDCls, BMDCs by CLSM (Fig. 2C and Figs. SI2, S13). As illustrated in Fig. 2C, the red fluorescence and green fluorescence represented AF647-labeled aCLEC9A-OVA257-264 conjugate and cyclic (8- (2- [fluoresceinyl] aminoethylthio)-labeled 2'3'cGAMP (denoted as cFAET-labeled 2'3'-сСАМР), respectively, verifying their co-loading into DHPSi-PEI NPs. The alone loading of STING agonist, and aCLEC9A-OVA257-264 in DHPSi-PEI NPs was further investigated in BMDCs by CLSM (Fig. SI4). The data represented that the alone loading of small molecules 2'3-cGAMP was efficient without the support of large M.W. aCLEC9A-OVA257-264 conjugate, reaffirming the potential role of branched PEI to bind anionic small molecules.

To achieve cytosolic delivery of aCLEC9A-OVA257-264 and STING agonist, Sİ9GM NPs must escape lysosomal compartments once they internalized into the endo/lysosomes [44]. The efficient release of Sİ9GM NPs from a lysosomal compartment was evaluated by the negligible presence of overlapping fluorescence (yellow area), favoring the delivery of cargo to present antigen peptides on МНС I molecules and activate STING pathway in cDCls (Fig. 2D and Fig. SI5). In contrast, free STING agonist 2'3-cGAMP showed a low ability to escape lysosome (Fig. SI6), verifying the potential of Sİ9GM nanovaccines to elicit antigen cross-presentation on МНС I molecules followed with stimulation of T cells and NK cells. In more detail, the lysosomal escape of Sİ9GM in cDCls by time was also investigated (Fig. SI7).

STING agonist, albeit to be highlighted as a potential small molecule in cancer immunity due to the ability to promote DCs maturation, is limited applied in cancer immunotherapy [45,46]. The highly negatively-charged characteristic of STING agonists prevents them from being effectively taken up by DCs [47], and it triggers a less potent stimulation of the immune system. To further verify the cellular uptake of 2'3 -cGAMP in cDCls, in vitro cDCls were treated by cFAET-labeled 2 3 -cGAMP as free molecules and encapsulated form in NVs, and the effectiveness of cellular internalization was examed by flow cytometric analysis. Consistent with the outcomes observed in BMDCs, the uptake of free STING agonist (cFAET-labeled-2/3/cGAMP) in cDCls was negligible and marginally increased over time, reaching a maximum value of 16.8 ± 1.12 % after 12 h of incubation (Fig. 2E). In contrast, Sİ9GM NPs delivered 2'3 -cGAMP more effectively, with 65.8 ± 5.27 % of cDCls positive for cFAET-2/3 cGAMP (Fig. 2F). Taken together, these results indicate the potential of the Sİ9GM nanovaccines as a delivery system for free STING agonists, overcoming certain limitations in cancer immunotherapy. Next, we investigated the role of Sİ9GM in enhancing aCLEC9A-OVA257-264 conjugates uptake in cDCls. Similarly, we found that the center-radial large pore structure of DHPSi NPs effectively delivered high M.W. conjugate (65.1 ± 2.42 %) to cDCls compared to free CLEC9A-OVA257-264 conjugates (40.3 ± 3.42 %) (Fig. 2G).

2.3. Si9GMs play a key role as artificial antigen-presenting cells for direct CDS T cell activation

For verification of the role of Sİ9GM NVs as artificial APCs, isolated CD8+ T cells from mice spleens were treated with Sİ9GM NVs. We observed an interaction between Si9GMs (red fluorescence) on the T cell surface (green fluorescence), confirming that the OVA257-264-MHC I complexes onto BMDC membranes of Sİ9GM nanovaccines facilitated interaction with CD8+ T cells, thus demonstrating the role of Sİ9GM NPs as artificial APCs (Fig. ЗА). Meanwhile, CD8+ T cells treated by NVs in the absence of the loading of OVA257-264 fragments on BMDC membranes showed minimal interaction between NPs and T cells, with the majority of Si9GMs localized in the cytosol (Fig. SI8). It is reported that the STING agonist induces apoptosis in a manner specific to T cells, but not in BMDCs or bone marrow-derived macrophages (BMDMs) [48]. Therefore, the loading of OVA257-264 antigen fragments onto mature BMDC membranes not only activates T cells but also prevents the uptake causing apoptosis in CD8+ T cells. For verification of Sİ9GM as an artificial APC, we investigated the capacity of Sİ9GM in CD8+ T cell activation by direct incubation between nanovaccines and CD8+ T cells isolated from OT-I mice. After 20 h of incubation, the population of activated CD8+ T cells (CD8+ CD69+) reached approximately 49.9 ± 4.33 %, which is higher compared to CD8+ T cells treated with OVA257-264 loaded mature BMDC membranes (DC-OVA, 38.3 ± 4.41 %) and immature BMDC-derived Sİ9GM (Si9GimM, 25.9 ± 3.58 %) (Fig. 3B and C). As a result of T cell activation, it exhibited the highest proliferation of CD8+ T cells when treated with Sİ9GM, consistent with the level of T cell activation (Fig. 3D and E). Given the most effective T cell activation induced by Sİ9GM, the cytotoxic effect of these CD8+ T cells was investigated by incubating them with B16-OVA cancer cells for 20 h, and the killing efficiency was determined by flow cytometry (Fig. 3F and G). As demonstrated in Fig. 3F, the necrosis and apoptosis populations of B16-OVA cancer cells incubated with Si9GM-induced cytotoxic CD8+ T cells increased up to 29.1 % and 38.4 %, respectively, which were higher than those of DC-OVA- and Si9GimM-treated CD8+ T cell groups. The capacity of Sİ9GM for direct CD8+ T cell activation, proliferation, and killing efficacy was evaluated in comparison with Dynabeads Mouse T-Activator CD3/CD28. Dynabeads showed higher efficacy compared to Sİ9GM, which could be attributed to the larger size of Dynabeads (4.5 |im) compared to that of Sİ9GM nanovaccines. Similarly, Sİ9GM showed better performance than DC-OVA due to the size-dependent T-cell activation [49]. Thanks to these findings, it can be concluded that Sİ9GM nanovccines play a pivotal role in CD8+ T cell direct activation, proliferation, and cytotoxic T cell-caused killing efficacy of cancer cells, promising the potential of Sİ9GM in cancer immunotherapy.

2.4. Si9GMs stimulate STING pathway activation and augment DC maturation

Many studies have shown that 23'-cGAMP binds to STING dimers at the endoplasmic reticulum (ER) membrane, promoting the ER-to-Golgi translocation of STING [50]. This early-stage activation leads to STING oligomerization triggered by conformational changes, facilitating subsequent downstream signals [51]. As shown in Fig. 4A, STING assists in the recruitment of TANK-binding kinase 1 (TBK1), enhancing its autophosphorylation (p-TBKl), phosphorylation of STING and interferon regulatory factor 3 (IRF3). pTBKl catalyzes IRF3 for dimerization, and those dimers translocate to the nucleus to induce gene expression of type I interferons (type I IFN) and other inflammatory genes [52]. In addition, STING activation leads to the phosphorylation of nuclear factor kappa В (NK-кВ), promoting the transcription of genes encoding a series of proinflammatory cytokines [53]. To verify the STING pathway activation of Sİ9GM and free 2 3 -cGAMP, Western blot assays were carried out to evaluate the upregulation of type I IFN-induced proteins. As displayed in Fig. 4B, the phosphorylation levels of TBK1 (p-TBKl), STING (p-STING), (IRF3) p-IRF3, and (NK-кВ) p-NK-кВ were increased after Sİ9GM incubation in comparison with free 2/3/-cGAMP that is attributed to the efficient delivery of STING agonist to BMDCs by Sİ9GM NVs over free STING agonists. Furthermore, the production of IFN-ß in Si9GM-treated BMDCs was higher compared to that in 2 3 -cGAMP-treated BMDCs (Fig. 4C). The result suggested that Sİ9GM displayed a predominant role in augmenting STING signaling cascade and type-I IFN production, potentiating Sİ9GM for robust DC activation and immune response.

Given that STING activation promotes DC maturation, we carried out the in vitro BMDC activation upon the treatment of 2'3'-cGAMP and Sİ9GM. When exposed to inflammatory and pathogenic signals, immature DCs undergo maturation, exhibiting the upregulation of costimulatory molecules and МНС II molecules, which promotes the antigen presentation and upregulates the chemokine receptor CCR7 [54]. Therefore, the role of Sİ9GM in promoting DC maturation, antigen cross-presentation, and LN homing was studied with BMDCs in vitro and compared to that of free З'З-сСАМР. After 24 h of incubation, co-stimulatory molecules CD40, CD80, CD86, and МНС II were stained and identified by flow cytometry. As illustrated in Fig. 4D and Fig. SI9, the increased population of CD40+ CDllc+, CD86+ CDllc+, and MHC II+ CDllc+ were expressed in BMDCs treated with Sİ9GM compared to those treated with free 2'3-cGAMP. The difference was attributed to the active targeting of free STING agonists to DCs through BMDC membrane coating. Next, we investigated the ability of DCs for cross-presentation of the MHC-I-restricted OVA epitope SIINFEKL for CD8+ T cell activation. The percentage of SIINFEKL+ CDllc+ greatly increased in the Si9GM-treated BMDCs (22.2 ± 0.96 %) compared to those treated with 2'3-cGAMP (15.6 ± 1.39 %), indicating that presentation of antigen peptides by STING stimulation-induced cross-presentation was more effective in Si9GM-treated BMDCs.

CCR7, a type of C-C chemokine receptor, triggers the homing of CCR7-expressed dendritic cells to the lymph nodes [55]. Since DCs need to be recruited into LNs where T cells are abundantly localized to induce antigen-specific T cells, the population of CDllc+ CCR7+ DCs was investigated. Impressively, Sİ9GM induced a higher percentage of CCR7-positive DCs (42.6 ± 1.31 %) compared to those of 2/3/-cGAMP (22.5 ± 1.91 %) (Fig. 4D and SI9). Taken together, DC-based nanovaccine Sİ9GM was well constructed for robust DC activation, LN homing, and enhanced immune response.

2.5. Biodistribution and selective recruitment to lymph node of Sİ9GM

Lymph node has become a key target whereby most immune cells are located in [56]. Therefore, nanovaccines should be effectively delivered to LNs to establish and elicit strong immune responses in cancer immunotherapy [57]. For verification of the LN-homing ability of Sİ9GM, we used cFAET-labeled 2/3 -cGAMP loaded in Sİ9GM and free cFAET-labeled 2'3'-cGAMP for subcutaneous injection into mice. The fluorescence intensity was surveyed within 2 days. After 12 h, LN and other organs were excised for fluorescence analysis. As presented in Fig. 4E and F, mice treated with dye-labeled Sİ9GM showed maximal fluorescent intensity at inguinal LN nearby injection site. In contrast, mice injected with free STING agonists represented negligible fluorescent intensity within 2 days, suggesting that Sİ9GM has a strong tendency to accumulate in inguinal LNs. Correspondingly, FOBI images and quantitative analysis revealed an abundant accumulation of Sİ9GM in excised inguinal LNs compared to other organs. Meanwhile, only a negligible 2'3-cGAMP signal was found in excised inguinal LNs of mice treated with free cFAET-labeled 2'3-cGAMP (Fig. 4G and H, Fig. S20). This is probably due to the existence of CCR7-upregulated DCs treated by Sİ9GM, which is essential to lymphatic homing of NVs. These results revealed that Sİ9GM is a potent nanosystem to target STING agonists to this critical lymphoid organ.

Next, the inguinal LNs were collected at 12 h post-vaccination and sectioned for immunofluorescence (IF) staining to examine the interaction between Sİ9GM NVs and T cells located in LNs. Si9GMs were labeled by 1,1'-dioctadecyl-3,3,3',3'-tetramethylindotricarbocyanine iodide (DiR) before being vaccinated into mice. After 12 h, mice were excised and the inguinal LNs were collected for the CLSM observation. Based on the presence of overlapping fluorescence between the red (Si9GMs) and green (T cells), we observed that Si9GMs were delivered across LNs and interacted with T cells, indicating the colocalization of Si9GMs and CD3+ T cells (white arrows) (Fig. 4I-K). Therefore, it can be concluded that Si9GMs functioned as artificial antigen-presenting cells (APCs) for DC-T cell interactions.

2.6. Sİ9GM augments cDCl activation and CD8 T-cell priming

cDCls are potently adept at presenting both exogenous and endogenous antigens for cross-presentation, as well as promoting T cell activation and proliferation, thereby generating memory and effector T cells [33]. This distinguished function of cDCls is key to their role in cancer immunotherapy, where they capture antigens and express them to CD8+ T cells after homing to tumor-draining lymph nodes [58,59]. To evaluate the efficacy of Sİ9GM NVs in selectively recruiting cDCls for potent antigen presentation and activating CD8+ T cells, we administered subcutaneous injections of Sİ9GM and 2'3-cGAMP to mice bearing B16-OVA tumors (Fig. 5A). DCs (including cDCl subsets) and CD8+ T cells were assessed on days 2 and 4 following the final vaccination. Overall, we observed increased immune cell (CD45+) and DC (CDllc+ MHC II+) populations in tumors of Si9GM-treated mice compared to those of the 2'3'-cGAMP-treated group and the untreated group (Fig. 5B). Two subsets of cDCl are resident cDCls (CD8+ cDCl) and migratory cDCls (CD103+ cDCl). Si9GM-treated mice showed a significant increment in percentage of both CD8+ cDCls and CD103+ cDCls compared to the 2'3'-cGAMP-treated group and untreated group. In contrast, the increase in cDCl population in the 2'3'-cGAMP-treated group was insignificant compared to the untreated group (Fig. 5C, Figs. S21A and C). The expression of co-stimulatory signal CD86 in the cDCl population was also enhanced in the Si9GM-treated group compared to other groups, potentiating the activation of both CD8+ cDCls and CD103+ cDCls (Fig. 5D, Fig. S21B and C).

Notably, the population of DCs with SIINFEKL-presenting H-2Kb MHC-I complexes was also remarkably increased in the Si9GM-treated group (48.7 ±1.36 %) compared to 2'3'-cGAMP-treated group (38.1 ± 4.43 %) and untreated group (29.4 ± 0.56 %), resulting from more effective antigen cross-presentation in Si9GM-treated mice (Fig. 5E, Fig. S22). Owing to the effective DC activation in the Si9GM-treated group, the percentage of antigen-specific CD8+ T cells was greatly increased compared to the 2'3'-cGAMP-treated group (3-fold) and untreated group (3.5-fold) (Fig. 5F). Correspondingly, the abundant population of antigen-specific CD8+ T cells triggered by Sİ9GM treatment resulted in significant cytotoxic T cells (IFN-y+ CD8+) compared to other groups (Fig. 5G), promising the potency of Sİ9GM for generation of an effective strategy in cancer immunotherapy. These results suggest that Sİ9GM is an effective delivery system for 2'3-cGAMP to cDCl cells, enhancing antigen cross-presentation and priming CD8+ T-cells.

2.7. Inhibition of melanoma growth in vivo by Sİ9GM combined with aPD-1 ICB

The antitumor efficacy of Sİ9GM was investigated in melanoma tumor-bearing mice. We inoculated the B16-OVA cell line into C57BL/6 mice and randomly divided the mice into 6 groups (n = 5 per each group, Table SI). After 7 days of tumor growth (20-30 mm3), mice were subcutaneously injected with various formulations. To investigate whether otPD-1 can effectively combine with Sİ9GM to trigger a synergistic cancer immunity, otPD-1 antibody (100 pg per dose) was intraperitoneally injected into mice treated with Sİ9GM after 24 h of nanovaccine administration. The nanovaccine administration was repeated 4 times at intervals of 3 days (Fig. 6A). Tumor growth and body mass were monitored every day. Lymphoid organs (spleens, LNs) and tumors were harvested day 5 following the last vaccination to analyze the impact of various immune cells on tumor suppression (Fig. 6B). At the end of the treatment, the Si9GM-treated group exhibited a remarkable inhibition in tumor growth (Fig. 6C) and a corresponding increase in survival rate (Fig. S23). Meanwhile, mice treated with free 2'3'cGAMP showed a lower level of significance in tumor inhibition. The SiGM-treated group presented slight efficacy in cancer treatment, suggesting the critical role of CLEC9A-OVA257-264 in promoting antigen cross-presentation. Notably, Sİ9GM therapy combined with otPD-1 resulted in a remarkably reduced tumor volume compared to the untreated group, while otPD-l-only treatment marginally inhibited tumor growth (Figs. S24 and S25). Thus, Sİ9GM combined with aPD-1 ICB therapy triggered a synergistic effect on antitumor therapy. For more details, the individual tumor growth in each group was presented (Fig. 6D). Similarly, tumor weight and representative photos of tumors in every group at the end of treatment affirmed the anticancer immunity of Sİ9GM (Fig. 6E and F). The toxicity of nanovaccines was assessed by monitoring changes in body weight and the Haematoxylin & Eosin (H & E) staining of major organs (Figs. S26 and S27). As shown in Fig. S27, there is no significant toxicity observed in the main organs (heart, liver, kidney, spleen) at the end of the treatment. However, we found an increase in the number of inflammatory cells, leading to alveolar wall thickening, in the lungs of mice in the untreated group, as well as in the otPD-1 and 2'3 -cGAMP -treated groups, compared to the SiGM-, Sİ9GM-, and Sİ9GM + otPD-1-treated groups. This significant thickening of alveolar walls in the mouse lungs could be attributed to the inflammatory response resulting from the recruitment of immune cells to combat metastatic cancer cells.

To further investigate the antitumor immunity of Sİ9GM combined otPD-1 ICB, tumors and lymphoid organs (spleens, LNs) were excised for the immune analysis. The representative photos of LNs in each group showed the increased size of LNs in mice treated with Sİ9GM with or without aPD-1 therapy in comparison with other groups (Fig. S28). Consistent with in vitro data, Sİ9GM with or without otPD-1 therapy robustly activated DCs in vivo, triggering an abundant population of matured DCs (CDllc+ CD80+ CD86+) for T cell priming. However, no remarkable difference was observed between these two groups that may be attributed to the negligible contribution to the antigen presentation on DCs of ICB. Meanwhile, other groups showed negligible impact on DC maturation compared with the untreated group (Fig. 6G and Fig. S29A). As a key feature of T lymphocyte-mediated immunity, the cytotoxic T cell (CD3+ CD8+) and helper T cell (CD3+ CD4+) populations were determined using flow cytometric analysis. The findings showed that Sİ9GM significantly enhanced the population of cytotoxic CD8+ T cells and helper T cells, and both fractions were notably enhanced in combination with otPD-1 ICB (Fig. 6H and Figs. S29B, S30). It could be elucidated that aPD-1 antibody contributed to the reduction of T- cell exhaustion [60]. Similarly, the percentage of CD8 T cells and CD4 T cells in spleens also increased in the Si9GM-treated group with or without ICB therapy (Fig. S31). In the TME, assessing the ratio of Ml-like to M2-like tumor-associated macrophages (TAMs) is essential for controlling inflammatory responses and tissue damage [61]. The treatment generated the dominant pro-inflammatory Ml-like TAMs (a higher Ml/М2 ratio) in the TME, promoting T cell immune responses and strengthening T and NK cells through secreted immuno-stimulatory cytokines (IL-6, TNF-a, ...) [62]. In contrast, the treatment produced a lower M1/M2 ratio reflects an increment in immune suppressive M2-like TAMs, attenuating antitumor immunity, and facilitating tumor chemoresistance. To determine whether Sİ9GM therapy can positively regulate macrophage polarization due to the presence of STING agonist as previously reported [31,32], we stained singlet tumor cells excised at the last day of the treatment for flow cytometry analysis. Then, the populations of M2-like macrophages (CDllb+ F4/80+ CD206+) and Ml-like-macrophages (CDllb+ F4/80+ CD86+) were examined and the M1/M2 ratio was calculated. There was a robust polarization from the М2 population to the Ml population after Sİ9GM treatment (Fig. 61 and Fig. S29C). Consistent with that data, the average M1/M2 ratio decreased from 0.56 ± 0.06 in the Si9GM-treated group to 0.32 ± 0.05 in the untreated group, indicating the ability of Sİ9GM to regulate an immune response for effective anti-tumor immunity. Interestingly, we found a slight decrease of M1/M2 ratio in the combination of Sİ9GM and aPD-1 ICB compared with only Sİ9GM therapy. This could be attributed to the combination of aPD-1 and Sİ9GM treatment, which may induce the expression of aPD-1 antibodies in TAMs) weakening their cytotoxic activity against tumor cells. This can result in the promotion of immunosuppressive M2-like TAMs, thereby enhancing the immune evasion mechanisms of tumor cells, and affecting the TME [63]. Notably, M2-like macrophages secreted some anti-inflammatory cytokines that can inhibit T cells and NK cell activation but induce regulatory T cells (Tregs) [64], enhancing the immunosuppressive environment. In addition, Treg infiltration with a significant number of cells into tumors induces tumor progression and a poor prognosis due to immune tolerance/escape of Tregs cells against antitumor immune responses [65]. For verification of tumor suppression-induced Tregs, we examined the role of Sİ9GM in regulating Tregs in comparison with other groups. As shown in Fig. 6J and Fig. S29D, the administration of Sİ9GM led to a decrease in the infiltration of immunosuppressive regulatory Tregs (CD3+ CD4+ CD25+ Foxp3+, 3.57 ± 0.89 %) in the TME compared to the untreated group (10.82 ± 2.40 %), consistent with the tumor inhibition observed with Sİ9GM. The addition of aPD-1 ICB slightly decreased the Treg population, while other treated groups did not reveal a similar impact. Our results emphasized that Si9GM-immunized mice displayed a decreased Treg population, especially in combination with aPD-1, thus reversing the immunosuppressive TME for effective cancer immunotherapy.

Furthermore, we investigated how efficient antigen peptides were processed and presented to CD8+ T cells. We examined the population of SIINFEKL-specific CD8+ T cells in excised tumors, LNs, and spleens (Fig. 6K, Figs. S32 and S33). The obtained data showed a significant increment in SIINFEKL-specific CD8+ T cell population in groups treated with Sİ9GM and Sİ9GM + aPD-1 ICB with 8.34 ± 1.48 % for the former and 10.55 ± 3.27 % for the latter. In contrast, mice vaccinated by free 2/3/-cGAMP or via intraperitoneal (i.p.) administration of aPD-1 did not generate a significant percentage of SIINFEKL-specific CD8+ T cells in comparison to untreated mice. Furthermore, we checked the activation of CD8+ T cells (CD3+ CD8+ CD69+). We found a significant increase in the population of activated CD8+ T cells in mice treated by Sİ9M and Sİ9GM + aPD-1. Other treatments (aPD-1, 2'3'-cGAMP, SiGM) did not robustly promote the T cell activation (Fig. 6L). Furthermore, the population of SIINFEKL-expressed CDllc+ DCs in spleens and LNs was also determined between untreated mice and mice immunized with Sİ9GM and Sİ9GM + aPD-1. An abundant accumulation of SIINFEKL-expressed CDllc+ DCs was consistently revealed in spleens after both therapies compared to the untreated group (Fig. 6M and Fig. S34). The population of IFN-y-secreting NK cells increased after Sİ9GM vaccination, promising NK cell-induced tumor rejection (Fig. 6N). The essential role of Sİ9GM nanovaccines in activating NK cell function for antitumor immunity could be elucidated by the type I IFN-induced IFNAR1 pathway triggered by the STING stimulation in DCs [66]. Importantly, IFN-y and granzyme В secreted by activated CD8+ T cells were also evaluated in tumor tissues at the end of the therapy (Fig. 60 and P). The abundant increase of secreted IFN-y on CD8+ T cells was observed in Sİ9GM therapy without or with otPD-1 blockade over 2.8- and 3.4-fold higher, respectively, than in the untreated group, demonstrating enhanced antitumor immune response in these groups (Fig. 60). To evaluate the long-term immune response, the populations of effector memory T cells (TEM, CD3+ CD4+ CD44+ CD62L-) and central memory T cells (Тем, CD3+ CD4+ CD44+ CD62L+) were examined. We found that the frequency of TEM and Тем increased significantly in the group treated with SiGM, Sİ9M, or Sİ9GM + aPD-1 compared to the untreated group, indicating long-term protective immunity following vaccination (Fig. 6Q and Fig. S29E).

In addition, tumor tissues from the untreated group, Si9GM-treated group, and Sİ9GM combined with aPD-1-treated group were sectioned for hematoxylin and eosin (H&E) staining to observe tumor inhibition (Fig. S35). IF staining was also performed using nuclear protein Kİ67 and CD8 antibodies to investigate cell proliferation and T cell recruitment, respectively. A massive presence of CD8+ T cells was also observed in those treated groups, complying with the flow cytometric data reported above (Fig. 6R). The IF staining further confirmed the inhibition of tumor cell proliferation in mice treated with Sİ9M and Sİ9GM + otPD-1 with negligible presence of Kİ67 proteins compared to the untreated group (Fig. 6S).

Furthermore, samples of serum were collected from blood of mice in 6 groups and used for ELISA analysis (Fig. 6T). The data showed all upregulation of secreted cytokines (IFN-y, TNF-a, IL-6, IL-2) from Sİ9GM with or without aPD-1 therapy, consistent with the above accumulating results, indicating the effectiveness of Sİ9GM vaccination in combination with otPD-1 in treating melanoma cancer. Notably, the concentration of IFN-ß protein in serum samples of Si9GM-treated mice with or without aPD-1 therapy was higher compared to the untreated group and 2/3/-cGAMP-treated group, confirming the massive production of type I IFN through STING activation after Sİ9GM vaccination.

2.8. Sİ9GM in lung metastasis inhibition and the role of NK cells

Despite significant advances in nanovaccines in cancer immunotherapy, the challenge of metastasis from the primary tumor to distant organs remains [67]. In this experimental model, we examined the capacity of Sİ9GM in combination with aPD-1 blockade in inhibiting metastasis to the lungs. C57BL/6 mice were intravenously injected with В16-0VA tumor cells two days before vaccination and immunized with various treatments on days 0, 3, 6, and 9 (Fig. 7A). Lungs were collected on day 20 for analysis. As shown in Fig. 7B, there was an extensive spread of metastasis focus in the lungs of untreated, aPD-1-treated, and 2/3/-cGAMP-treated mice. In contrast, vaccination with SiGM or Sİ9GM without and with ICB had a significant inhibitory impact on lung metastatic tumor surveillance. Lung weights in untreated, aPD-1-, and 2/3 -cGAMP-treated groups were remarkably higher than those in SiGM or Sİ9GM without and with ICB groups due to the abundant formation of metastatic nodules in the former (Fig. 7C). For further verification of metastasis formation in lung tissues, we harvested lungs for H & E staining. As illustrated in Fig. 7D, massive tumor foci were observed in untreated, aPD-1-, and 2/3 -cGAMP-treated groups, but no significant hallmark of lung metastasis was found in the other groups.

Next, we investigated the role of immune cells in lung metastasis inhibition by flow cytometry of digested lung tissues at the endpoint. NK and T cells, the two most potent immune effector cells, were taken into the examination. As shown in Fig. S36, there was no distinct difference between the CD4+ or CD8+ T cell population in untreated and Sİ9GMtreated mice. Consistently, the percentage of IFNy+ CD8+ T cells in these groups was similar. Of note, the population of IFN-y-secreted NK cells was incredibly boosted in Sİ9GM with or without aPD-1 blockade compared to other groups, indicating a critical antitumor function of NK cells in preventing the spread of cancer cells to distant organs (Fig. 7E and F). The body weight of mice and the survival rate in each group were recorded (Figs. S37 and S38).

To further identify the contributions of NK and T cells, we compared the control groups (Gl, G2) with two other groups (G3, G4) in which NK and CD8+ T cells were diminished by antibodies in mice received Sİ9GM + aPD-1 treatment (Fig. 8A). As shown in Fig. 8B, effective inhibition of lung metastatic seeding of Bl6 melanoma cells was observed in the G4 group (Sİ9GM + ocPD-1 + CD8a depletion) compared to the G1 group (untreated mice), G2 group (IgG isotype-treated mice) and G3 group (Sİ9GM + otPD-1 + NK cells depletion). Notably, the G3 group showed a significant quantity of lung metastatic foci that is like those of Gl, and G2 group. In contrast, the depletion of CD8+ T cells in the G4 group did not impair the effectiveness of the treatment, with a remarkable decrease in the number of lung metastatic seeds (Fig. 8B). In addition, H & E staining of lung tissues at the endpoint verified the presence of metastatic foci in Bl6 melanoma cells (Fig. 8C). Consistently, the lung weight was higher in Gl, G2, and G3 than in G4 (Fig. 8D), highlighting the prominent role of NK cells over CD8+ T cells in prevention of lung metastasis in Bl6 melanoma cancer cells. These findings could be attributed to the abundant localization of NK cells in peripheral blood compared to other organs (lung, skin, liver, kidney, and bone marrow), instantly reducing circulating cancer cells post i.v. injection [68]. Furthermore, the underlying mechanism of NK cells in metastasis prevention is supposed to be triggered by STING pathway activation in cDCls post-Si9GM vaccination. In detail, Si9GMs internalize dendritic cells (selectively target in cDCls), following the release of 2 3 -cGAMP for STING stimulation. Subsequently, the type I interferons produced from DCs are transferred to NK cells by expression of interferon-a/ß receptor (IFNAR) on NK cells. As a result, interferon alpha/beta receptor 1 (IFNAR1) signaling in NK cells promoted NK cell function, activation and generated cytotoxic cytokines like IFN-y for metastasis rejection [66]. Therefore, in the absence of NK cells in the G3 group, the number of lung metastatic foci was slightly increased compared to Gl, G2 group. Impressively, the presence of an increased population of migratory GDI03 cDCls was observed in Sİ9GM + aPD-1 -treated mice, contributing to the inhibition of lung foci growth (Fig. 8E). In addition, a larger increment in IFNy-NK cells was observed in the G4 group compared to the Gl group (2.9-fold) (Fig. 8F). Meanwhile, there was no significant increase in T cell population between Gl and G2 groups (Fig. 8G). Taken together, these findings demonstrate that NK cells and GDI03 cDCls might play a prominent role in evading the seeding of lung metastasis compared to CD8+ T cells in Bl6 melanoma cancer cells [69].

3. Conclusion

In this study, we proposed an attractive DC-based nanovaccine selectively targeting cDCl to augment the STING pathway activation and type I IFN production profile for effective cancer vaccination immunotherapy. Leveraging the unique ability of cDCls in antigen cross-presentation and cytotoxic T lymphocyte activation, we developed a novel nanovaccine formulation in which large pore dendrimer-like silica NPs with hierarchical center-radial pores, delivering distinct sizes of aCLEC9A-OVA257-264 conjugate and STING agonist, were coated with BMDC-derived membranes containing CLEC9A antibodies and antigen peptides. aCLEC9A-conjugated DC membranes drive nanovaccine to cDCls, while aCLEC9A-OVA257-264 can avoid lysosomal degradation and the loss of prematurely released antigens. Notably, 2'3'- cGAMP stimulates type I interferons, leading to robust CD8+-specific antigen cross-presentation and effective immune response. The DCbased nanovaccine Sİ9GM synergized with aPD-1 blockade to elicit effective inhibition of B16-OVA tumor progression and metastasis without apparent side effects. Impressively, Sİ9GM with otPD-1 ICB can regulate macrophage polarization, inhibit Tregs, and initiate enduring immunological memory against tumor cells. The potency of Sİ9GM addresses existing disadvantages of live cell-based vaccines, such as limited immunogenicity, risk of pathogenicity, storage challenges, and inherits the advantages of DC-based vaccines. Overall, this work presents the first time that the formulated nanovaccines, combining porous inorganic nanoparticles and APC-derived membranes, can selectively target the cDCl subset to evoke robust STING activation and promote potent antigen cross-presentation, thereby facilitating efficient personalized cancer immunotherapy.

4. Experimental section

4.7. Materials and methods

The 2'3-cGAMP was purchased from MedChemExpress (Monmouth Junction, NJ, USA). Cyclic (8- (2- [fluoresceinyl]aminoethylthio)- guanosine(2',5')monophosphate-adenosine(3',5')monophosphate) sodium salt (denoted as cFAET-labeled 2/3/-cGAMP) was purchased from BioLog Inc. (Hayward, CA, USA). The OVA257-264 peptide fragment and F AM-labeled OVA257-264 were purchased from Anaspec (Fremont, GA, USA). Lipopolysaccharides from Escherichia coli O111:B4 (cat. no. L3024), (3-aminopropyl) triethoxysilane (APTES), cetyltrimethylammonium bromide (СТАВ), branched polyethylenimine (0.8 kDa), Tetraethyl orthosilicate (TEOS), 2-iminothiolane hydrochloride (Traut's reagent), succinic anhydride (SA), and 4-(N-maleimidomethyl) cyclohexane-1-carboxylic acid 3-sulfo-N-hydroxysuccinimide ester sodium salt (sulfo-SMCC) were purchased from Sigma-Aldrich. Neuro-DiO was purchased from Biotium (Fremont, CA, USA). 1,T-Dioctadecyl3,3,3,3-tetramethylindocarbocyanine perchlorate (cat. no. D3911), Hoechst 33342 (cat. no. Hl 399), and LysoTracker™ Red DND-99 (cat. no. L7528), FITC Annexin V/Dead Cell Apoptosis Kit (cat. no. VI3242) were purchased from Invitrogen (Thermo Fisher Scientific, MA, USA). Dynabeads® Mouse T-Activator CD3/CD28 (cat. no. 11456D) was purchased from Gibco (Thermo Fisher Scientific, MA, USA). Recombinant murine GM-CSF protein (cat. no. 315-03), mouse IL-2 recombinant protein (cat. no. 212-12-20UG) were purchased from PeproTech (Thermo Fisher Scientific, MA, USA). Recombinant mouse IL-4 protein (cat. no. 51084-MNAE) was purchased from Sino Biological (Beijing, China). Human FLT3-ligand (research grade) was purchased from Miltenyi Biotec (North Rhine-Westphalia, Germany). ELISA kits of IL-6, IL2, TNF-a, IFN-y were purchased from Invitrogen (Thermo Fisher Scientific, MA, USA). LEGEND MAX™ Mouse IFN-p ELISA kit (cat. no. 439407) and MojoSort™ Mouse CD8 T Cell isolation kit (cat. no. 480008), Alexa Fluor® 488 Goat anti-hamster (Armenian) IgG antibody (cat. no. 405508), CFSE cell division tracker kit (cat. no. 423801) were purchased from BioLegend (San Diego, CA, USA). All antibodies used for Western blot, in vivo experiments, and flow cytometric analysis were listed in Tables S2-S4.

4.2. Cell cultures and animals

DC2.4 dendritic cell line used for the in vitro study was purchased from Millipore and B16-OVA cell line used for the in vivo study was purchased from the American Type Culture Collection. DC2.4 and Bl 6OVA cells were cultured in RPMI and DMEM medium, respectively, both containing 10 % heat inactivated-FBS (HI-FBS), 1 % antibiotics (penicillin-streptomycin, 10,000 U mL-1), and 50 pM mercaptoethanol-2 in a humidified atmosphere containing 5 % CO2 at 37 °C. BMDCs and cDCls were isolated from femurs and tibias of 7-8-week-old C57BL/6 mice and then stimulated with high-glucose RPMI media containing GMCSF and IL-4 for the generation of BMDCs and with FLT3L for the generation of cDCls.

C57BL/6 mice (6-week-old females, 7-week-old males) were ordered from ORIENT BIO (Seongnam, South Korea). OT-I mice were kindly provided by Prof. Yong Taik Lim (Sungkyunkwan University, Korea). The mice were housed under conditions with a regulated 12-h light and 12-h dark cycle, with lights turning on at 8:00 a.m.

4.3. Synthesis of hierarchical large-pore amino-functionalized silica nanoparticles

For the loading of high-molecular-mass proteins like ocCLEC9AOVA257-264 conjugates, dendrimer-like large-pore silica nanoparticles with hierarchical pores were synthesized with some modifications [41]. For details, DHPSi-NH2 NPs were formulated based on the mechanism of the quasi-emulsion nanodroplet method. Firstly, 70 mL of deionized water was added to a 125-mL amber glass bottle with a stir bar. After that, a mixture of diethyl ether (15 mL) and ethanol (5 mL) was dropped into the water using a syringe pump (pumping speed: 2000 pL min-1) while stirring at 1050 rpm. Subsequently, 500 mg of СТАВ was added, followed by the dropwise addition of 800 pL of TEA. After 30 min, a mixture of TEOS and APTES in the ratio of 1.5 mL-60 pL was dripped into the above mixture, and the reaction lasted for 3.5 h before adding 1 mL of HC1 to terminate the base-catalyzed reaction.

Subsequently, raw products were collected by centrifugation at 8200 rpm for 12 min and washed in ethanol: DW (2:1, v/v) before being dispersed into ethanol. For further purification, as-prepared DHPSi-NH2 NPs were dispersed in 100 mL ethanol and then sonicated in a bath for 30 min before probe sonication (20 min, 2s ON-2s OFF). After that, consecutive centrifugation cycles at 2000 rpm, 2500 rpm, 3000 rpm, 3500 rpm, 4200 rpm, and 6200 rpm (3 min for each cycle) were conducted to remove large NPs [70]. The final supernatant was washed three times with ethanol to obtain DHPSi-NH2. The СТАВ templates were extracted by refluxing (70 °C for 24 h) in 12.5 % volume of concentrated HCl in ethanol. The final DHPS1-NH2 NPs were obtained by centrifugation at 12,500 rpm for 15 min and washed with ethanol three times before further conjugation steps.

4.4. Synthesis ofDHPSi-PEI

Next, 5 mg of DHPS1-NH2 was dispersed in 2 mL of DMF and then put in an ultrasonic bath for 10 min. After that, 45 mg of succinic anhydride was dissolved in 0.5 mL of DMF and then added to the as-prepared DHPS1-NH2 solution. The pH was turned to basic conditions by adding 40 pL of TEA, and the mixture was stirred at 600 rpm for 12 h at RT. The resulting DHPSi-COOH NPs were washed three times in a mixture of ethanol: DW (2:1, v/v). Subsequently, DHPSi-COOH was conjugated to EDC.HC1 and NHS in DMF for 30 min before adding branched PEI (0.8 kDa, 100 mg mL-1). The final product DHPSi-PEI was purified by washing three times in a mixture of ethanol: DW (2:1, v/v).

4.5. Generation of cDCl from bone marrow

For the generation of cDCls, femurs and tibias from 7-8-week-old mice were isolated. Bone mesenchymal stem cells (BMSCs) were obtained from cavities of bone marrows. Bone marrow-derived stem cells (BMSCs) were isolated from the femurs and tibias of 7-8-week-old male C57BL/6 mice. Subsequently, BMSCs were flushed from the marrow cavities using a needle and RPMI-1640 medium containing 10 % heatinactivated fetal bovine serum (HI-FBS), 1 % antibiotics (penicillinstreptomycin, 10,000 U mL-1), and 50 pM 2-mercaptoethanol (referred to as 10 % HI-FBS RPMI-1640). The cells were then lysed using red blood cell (RBC) lysis buffer and washed in HBSS without calcium. After RBC lysis, BMSCs were seeded on 6-well plates at a cell density of 1.5 x IO6 cells/well and then stimulated in 10 % HI-FBS RPMI-1640 supplemented with 200 ng mL-1 FLT3L. One-half of the media volume was replaced at day 4 and day 8, and the non-adherent cDCls were harvested at day 11 [71].

4.6. Generation of bone marrow-derived BMDC nanovesicles

BMSCs were extracted from femurs and tibias of male C57BL/6 mice (7-8-week-old). Additional BMSCs were flushed from marrow cavities using a needle with RPMI-1640 containing 10 % heat inactivated-FBS (HI-FBS), 1 % antibiotics (penicillin-streptomycin, 10,000 U mL-1), and 50 рм mercaptoethanol-2 (denoted as 10 % HI-FBS RPMI-1640). These cells were lysed by RBC lysis buffer and washed in HBSS without Ca2+. The collected BMSCs were stimulated in as-prepared 10 % HI-FBS RPMI-1640 containing 20 ng mL-1 of GM-CSF and 20 ng mL-1 of IL-4 and then seeded in non-TC petri dishes at a cell density of 0.3 x IO6 cells mL-1. After 3 days of incubation, 12.5 mL of media was added to each cultured Petri dish. At day 6, one-half of the media was replaced with new media; the non-adherent cells were harvested at day 8 for LPS activation (1 pg mL-1) for 24 h [72].

After LPS activation, immature BMDCs were collected, and the BMDC-derived membranes were extracted by Mem-PER™ Plus Membrane Protein Extraction Kit following the manufacturer's instruction. In detail, BMDCs were washed with cell wash solution before adding to cell permeabilization buffer (including IX proteinase inhibitor cocktail). Cell suspensions were incubated on ice for 30 min before shaking at 4 °C for 1 h. Subsequently, cell suspensions were centrifuged at 800 g for 5 min to eliminate cell debris. After that, supernatant was collected and centrifuged at 27, 000 g for 30 min 4 °C to obtain cell membranes. After the last centrifugation cycle of washing, BMDC membranes were lyophilized and stored at -80 °C for further use.

4.7. Preparation of «CLEC9A @DC-OVA25/-264

To load CD8+ T cell-specific antigen fragments OVA257-264 onto MHC-I molecules of BMDC membranes, BMDCs (2 mg mL-1) and OVA257-264 were incubated in 0.2 м sodium acetate (pH 5.2) in PBS (v/v = 1:1) for 45 min at RT. Subsequently, the pH was adjusted to neutral (pH ~ 7-7.4) by Tris buffer (2 m, pH 11), and the mixture was incubated on ice for 45 min to stabilize antigen peptide-МНС I complex. DCOVA257-264 membranes were centrifuged at 27,237xg for 20 min at 4 °C and then were rinsed twice to remove unloaded peptides. These membranes then were incubated with Traut's reagent in PBS pH 8 and washed with Tyrode's buffer for 30 min at RT; the unreacted Traut's reagents were removed by centrifugation. Meanwhile, «CLEC9A antibodies were conjugated with sulfo-SMCC in PBS (pH 7.4, 2.5 тм EDTA) for 1 h at RT, and the excess cross linkers were eliminated by Zebra Spin Desalting Columns 7K MWCO. After that, the as-conjugated DC-OVA257264 and «CLEC9A were mixed and incubated at 4 °C overnight. aCLEC9A @DC-OVA257-264 membranes were collected by centrifugation at 27,237xg for 20 min at 4 °C and then were washed to remove unconjugated antibodies.

4.8. Preparation of aCLEC9A-OVA257-264 conjugates

In short, purified antibodies CLEC9A (CD370) and a rat IgG2a isotype control (Cl.18.4) were functionalized with 2 equivalents of sulfoSMCC in PBS pH 8 for 1 h at RT. Meanwhile, endotoxin-free OVA257264 peptide fragments were conjugated with 2 equivalents of Traut's reagent in PBS pH 7.4. Antibodies were desalted over Zebra Spin Desalting Columns 7K MWCO. After that, 6 equivalents OVA were added to antibodies while stirring for 2 h at RT before purification by desalting [73].

4.9. Preparation and characterization of Sİ9GM

DHPSi-PEI NPs were dispersed in endotoxin-free DW at a concentration of 2 mg mL-1, followed by adding 2'3'-cGAMP (50 pg per 1 mg of DHPSi-PEI NPs) and stirring at 500 rpm for 30 min at RT. After that, aCLEC9A-OVA257-264 conjugates were added for loading at 4 °C overnight. After removing excess cargo by centrifugation, (2'3'-cGAMP and aCLEC9A-OVA257-264 conjugate)-loaded DHPSi-PEI NPs (denoted as Sİ9G) were coated by aCLEC9A@DC-OVA257-264- In detail, DHPSi-PEI NPs was dispersed in 10 mM phosphate buffer pH 7.4 (2 mg mL-1) before adding in BMDC-derived solution (2 mg mL-1) for sonication in 7 min. The as-prepared membrane-coated NPs were then washed twice with phosphate buffer to obtain Sİ9GM. Subsequently, size and morphology were investigated by ТЕМ.

4.10. Western blotting

The proteins in BMDC lysate, BMDC membranes, and BMDCs were extracted in RIPA buffer containing Protease Inhibitor Cocktail (PIC) in which PIC: RIPA = 1:100 (v/v). The extraction occurred at 4 °C for 30 min under shaking. The protein concentrations were evaluated by BCA assay. For denaturation, proteins were heated at 95 °C for 8 min. Subsequently, a comparative amount of proteins was loaded into each well of 10 % or 12 % SDS polyacrylamide gel depending on the analyzed protein size. Electrophoresis was carried out at 70 V for 30 min and then 100 V for 70 min, followed by protein transfer onto a poly vinylidene fluoride (PVDF) membrane (100 V, 120 min). Following a 1-h blocking step with 5 % nonfat milk at RT, the PVDF membranes were then immersed with specific primary antibodies ( rable S2) at 4 °C for 12 h under shaking at 30 rpm. After washing with 1 % Tween-20 of TrisBuffered Saline solution (TBS-T 1 %), the membranes were then immersed with HRP-conjugated secondary antibody for 1 h at RT. Subsequently, the membranes were washed three times by TBS-T 1 % and then immersed in Pierce™ ECL Western Blotting Substrate for 5 min on the shaker. Subsequently, the protein bands in PVDF membranes were observed by a ChemiDoc Imaging System (Bio-Rad).

4.11. Coomassie protein electrophoresis

All the proteins in BMDC lysate and BMDC membranes were extracted by incubating in RIPA buffer compromising 1 % volume of PIC for 30 min on ice. Protein concentrations were examined using the BCA method, followed by the separation of proteins on a 10 % SDS polyacrylamide gel. Subsequently, the running gels were stained using Coomassie Brilliant Blue R-250 solution. After destaining step, protein bands in SDS polyacrylamide gels were observed by a ChemiDoc Imaging System (Bio-Rad).

4.12. Evaluating the Sİ9GM uptake of BMDCs using bio-TEM

BMDCs were seeded on a 6-well plate at a density of 1 x IO6 cells mL-1 and incubated with Sİ9GM (50 pg mL-1) for 12 h. Next, BMDCs were harvested and fixed with fixation buffer comprising a mixture of paraformaldehyde and glutaraldehyde (v/v = 2 %/2 %) in 0.05 м sodium cacodylate buffer (pH 7.2) at 4 °C for 4 h. Afterward, BMDCs were rinsed and re-fixed using 1 % osmium tetroxide in 0.1 M sodium cacodylate buffer (pH 7.2) at 4 °C for 2 h. After dehydration, treated BMDCs were embedded in resin and cured at 65 °C overnight for polymerization. The embedded specimens were cut into sections with a thickness of 100 nm. Finally, the slice was stained with 2.5 % uranyl acetate for 7 min followed by 2.5 % lead citrate for 2 min before being observed at 80 kV using a JEM 1011.

4.13. Evaluating the ability to escape lysosomal compartments of Sİ9GM

cDCls were seeded on an 8-well Ibidi p-Slide (Ibidi GmbH) at a cell density of 5 x 104 cells per well and then incubated with Sİ9GM (50 pg mL-1) and 2 3 -cGAMP (1 pM). To monitor the colocation of nanovaccines and lysosomes at 8 h post-treatment, cDCls were incubated with LysoTracker Red (50 nM) at 37 °C for 30 min. After that, cDCls were rinsed twice with PBS and then subjected to nuclear staining with Hoechst 33342 (1 pg mL-1) at 37 °C for 20 min. The cells were immediately imaged with a CLSM and fluorescence (LysoTracker Red, Ex/Em = 577 nm/590 nm; Hoechst 33342, Ex/Em = 350 nm/461 nm).

4.14. Evaluating the targeting ability of Sİ9GM in cDCls

For CLEC9A receptor blocking, cDCl cells at day 11 post-stimulation (1 x IO6 cells mL-1) were incubated with TruStain FcX™ (anti-mouse CD16/32, 1 pg mL-1) antibody for 15 min on ice to prevent the nonspecific bindings of immunoglobulin to Fc receptors. Subsequently, cDCl cells were stained with purified aCLEC9A antibody (anti-mouse CLEC9A, BioXcell) at a aCLEC9A concentration of 0.3 pg mb-1 of cDCl, and then incubated for 30 min on ice. The excess amount of antibody was removed by centrifugation. At this stage, as-prepared cDCls can be used for the treatment. For the verification of the comprehensive blocking of CLEC9A receptors, the as-prepared cDCls were stained by APC-labeled CLEC9A antibody (0.3 pg mL-1) for 30 min on ice. After washing, cDCls were analyzed by flow cytometry to determine the blocking efficacy.

After washing, aCLEC9A-blocking cDCl cells and non-blocking cDCl cells were seeded in 96-well plate at a cell density of 1 x 105 cells per well. After 2 h-incubation, cDCl cells were treated by fluorescentlabeled Sİ9GM and fluorescent-labeled SiGM for the next 2 h. After that, cells were harvested and washed twice to eliminate the non-uptake nanoparticles. The percentage of uptake in aCLEC9A-blocking cDCl cells and non-blocking cDCl cells was analyzed by flow cytometry.

4.15. CD8+ T cells isolation from mice spleens

Female C57BL/6 mice (7-week old), received subcutaneous injections of DC-OVA257-264 three times at intervals of 3 days. One day post the last vaccination, the spleen was collected and cut into small pieces. After digestion into single cells, CD8+ T cells were isolated by Mojosort isolation kit (BioLegend). In brief, cell density was adjusted into a concentration of 1 x IO6 cells mL-1 in MojoSort buffer and divided into 100 pL of cell suspension. Next, the biotin-antibody cocktail (10 pL) was added into each tube and incubated at 4 °C for 15 min. Subsequently, streptavidin nanobeads was vortexed and added into the above cell suspension, followed by the incubation on ice for 15 min. After that, MojoSort buffer (2.5 mL) was added. The stained splenic cells were then placed in the magnet for at least 5 min and the CD8+ T cells were then carefully aliquoted into a new tube for subsequent experiments.

4.16. Observation of CD8+ T cell interaction with Sİ9GM by CLSM

Splenic CD8+ T cells were seeded on an 8-well Ibidi p-Slide (Ibidi GmbH) at a cell density of 1 x 105 cells per well and incubated at 37 °C for 24 h. CD8+ T cells were treated with Sİ9GM (50 pg mL-1) at 37 °C for 30 min and rinsed by DPBS to remove the excess Sİ9GM NPs. After fixation at RT for 20 min, the T cells were stained by neuro-DiO at 37 °C for 30 min and Hoechst 33342 at 37 °C for 20 min, followed by CLSM observation.

4.17. Evaluation the ability of Sİ9GM in CD8 T cell activation and proliferation

CD8+ T cells were isolated from spleens and LNs of OT-I mice (12week-old, male/female). Spleens and LNs were then minced into single cell suspension in 10 % FBS-supplemented RPMI media containing collagenase IV (1 mg mL-1) and DNase I (100 pg mL-1). After washing, the number of cells was counted and CD8+ T cells were isolated from the cell suspension using a CD8 T cell Mojosort isolation kit (BioLegend).

In an experiment to evaluate CD8+ T cell activation, isolated CD8+ T cells were seeded on 96-well plate at a cell density of 1 x 105 cells per well in 1 % FBS-supplemented RPMI media. Subsequently, CD8+ T cells were treated by DC-OVA (1 mg mL-1), Si9GimM (immature BMDC membrane coating, 100 pg mL-1), Sİ9GM (100 pg mL-1), and CD3/ CD24 Dynabeads (2 x 105 beads per well) for 20 h. After that, CD8+ T cells were harvested and stained by CD69 antibody for flow cytometric analysis.

In an experiment to evaluate CD8+ T cell proliferation, isolated CD8+ T cells were stained by CSFE (5 pM) in 10 mM PBS pH 7.4 for 20 min at 37 °C at a cell density of 1 x 107 cells per mL. After washing to remove the excess CFSE, CD8+ T cells were seeded on 96-well plate at a cell density of 1 x 105 cells per well in 10 % FBS-supplemented RPMI media. After 15 min of incubation, CD8+ T cells were treated by DC-OVA (1 mg Bioactive Matenals 42 (2024) 345-365

mL-1), Si9GimM (immature BMDC membrane coating, 100 pg mL-1), Sİ9GM (100 pg mL-1), and CD3/CD24 Dynabeads (2 x 105 beads per well, in addition of IL-2 _ 50 IU mL-1). The plate was incubated at 37 °C for 4 days. A half media in each well were replaced by the fresh 10 % FBS-supplemented RPMI media every 48 h. At day 4, CD8+ T cells were harvested for flow cytometric analysis.

4.18. Evaluation the killing efficacy of cytotoxic CD8+ T cells to В16OVA OVA cancer cells

B16-OVA cancer cells were seeded on 24-well plate at a cell density of 1 x 105 cells per well in 10 % FBS-supplemented RPMI media, incubated overnight at 37 °C. Otherwise, cytotoxic CD8+ T cells at day 4 in the T cell proliferation experiment were added to B16-OVA cells (Bl 6OVA: CD8 T = 1 : 4) and incubated for 20 h at 37 °C. Subsequently, all cells were harvested and stained by CD3 antibody and then live/dead staining kit (PI/Annexin V- FITC) for flow cytometric analysis.

4.19. Characterization of DC maturation

BMDCs were seeded on tissue culture (TC)-treated 6-well plates at a cell density of 1 x IO6 cells mL-1 and then incubated in 1 % HI-FBS RPMI-1640 containing 2 3 -cGAMP (10 рм) or Sİ9GM (100 pg mL-1) for 24 h. Once cells were harvested, BMDCs were stained with antibodies specific for mature DCs: FITC anti-mouse CDllc, Alexa 700 anti-mouse МНС II, Alexa 647 anti-mouse CD40, PE dazzle 594 anti-mouse CCR7, and PE/Dazzle™ 594 anti-mouse H-2K b bound to SIINFEKL. The incubation was carried out for 30 min on ice. After that, the characterization of DC maturation was performed using a FACSAria™ Fusion (BD Biosciences). Data were analyzed with FlowJo software (version 10.8.1).

4.20. Biodistribution and lymph node imaging

cFAET-labeled 2'3'-cGAMP and cFAET-labeled 2'3'- cGAMP@Si9GM (6 pg of cFAET-labeled 2'3-cGAMP per mouse) were subcutaneously injected into of C57BL/6 mice. Subsequently, the kinetic fluorescence distributions of cFAET-labeled 2'3'-cGAMP were examined at 0, 2, 4, 8, 12, 24, and 48 h postinjection with an in vivo imaging system (FOBI in vivo imaging system, NeoScience). To verify the distribution of nanovaccines, inguinal lymph nodes and key organs (heart, liver, spleen, lung, and kidney) were extracted for ex vivo fluorescence imaging 12 h following vaccination. For verification of the interaction between T cells and nanovaccines in vivo, DiR-labeled Sİ9GM (200 pg of DHPSLNH2 per mouse) was subcutaneously injected, and LNs were collected for highmagnification confocal examination. The lymph nodes were cut into sections (10 pm-thick) and incubated with anti-CD3 antibody (Biolegend catalog no. 100340, diluted 1:50) overnight at 4 °C. They were then treated with Goat anti-hamster IgG antibody (Biolegend catalog no. 405508, diluted 1:100) for 1 h at RT. Following washing with PBSTween, the tissue sections underwent DAPI staining for 5 min at room temperature before being washed again with PBS-Tween. The tissue sections were imaged using a confocal laser scanning microscope.

4.21. In vivo maturation of cDCls and CD8+ T-cell priming

C57BL/6 mice (7-week-old, female, n = 3 for each group) were shaved and then subcutaneously injected with В16-0VA cells (2.5 x IO6 cells per mouse). Ten days after inoculation, mice were vaccinated three times at intervals of 3 days (50 pg free 2z3'-cGAMP, and NVs dose comprising 5 pg of 2'3'-cGAMP per mouse). On day 8, and day 10 post the first vaccination, mice spleens and tumors were harvested for immune cell analysis, respectively. Briefly, spleens and tumors were digested in a solution comprising collagenase type IV (1 mg mL-1) and DNase I (100 pg mL-1) to obtain single cells. Subsequently, cell suspensions were treated with RBC lysis buffer to eliminate red blood cells, followed by blocking Fc receptors. Next, cell suspensions were stained by specific antibodies for flow cytometry analysis. Flow cytometric analyses were performed using a BD FACSAria Fusion. FlowJo software (version 10.8.1) was used to analyze data.

4.22. In vivo immune responses

C57BL/6 mice (7-week-old, female, n = 5 for each group) were shaved and then subcutaneously injected with B16-OVA cells (1.5 x 105 cells per mouse). Seven days after inoculation (50 pg free 2/3 -cGAMP, and NVs dose comprising 5 pg of 2'3'-cGAMP per mouse), mice were vaccinated 4 times at intervals of 3 days. Subsequently, LNs, spleens, and tumors were collected on the assigned day and further digested into cell suspension by collagenase type IV (1 mg mb-1) and DNase I (100 pg mb-1). The cell suspension was centrifuged at 350 g for 5 min; the collected cells were resuspended with RBC lysis buffer at 4 °C for 5 min to lysis red blood cells. The lysis was terminated by 10 % FBS RPMI. Single cells were obtained by filtration through a cell strainer (70 pm pore-size) and washed with FACS staining buffer for immune cell staining according to the protocol for surface proteins, intracellular proteins, and nuclear protein staining. Flow cytometric analyses were performed using a BD FACSAria Fusion. FlowJo software (version 10.8.1) was used to analyze flow cytometric data.

4.23. In vivo anti-metastasis capability

C57Bb/6 mice (6-week-old, female, n = 5 for each group) were injected intravenously with B16-OVA melanoma cells (1.5 x 105 per mouse). Mice were immunized by subcutaneous injection (50 pg free 2'3'-сСАМР, and NVs dose comprising 5 pg of 2'3 -cGAMP per mouse) at the right leg (a total of 4 vaccinations with 3-day intervals). All mice were euthanized on day 20, and lungs were harvested and weighed. All lung tissues were used for H & E staining and FACS analysis, bungs were sectioned into small fragments of 1-2 mm in length and then digested in 10 % RPMI-1640 supplemented with collagenase type IV (1 mg mb-1) and DNase I (100 pg mb-1) at 37 °C for 20 min, followed by filtration through a 70 pm cell strainer. After that, single cells were treated with RBC lysis buffer for 3 min at RT. The isolated cells were washed with cell staining buffer and stained as in the manufacturer's protocol. Flow cytometry analyses were carried out using a BD FACS Aria Fusion. Data were analyzed with FlowJo software (version 10.8.1).

4.24. In vivo immune cell depletion for lung anti-metastasis evaluation

C57Bb/6 mice (7-week-old, female, n = 5 for each group) were injected intravenously with B16-OVA melanoma cells (1.5 x 105 cells per mouse) and divided into four groups (Gl, G2, G3, G4). After 1 day, each group was subjected to an antibody through intraperitoneal injection. For NK cell depletion (G3), mice were i.p. injected with anti-NK 1.1 antibody (200 pg per mouse) at day - 1, day 2, and day 5. For CD8 T cell depletion (G4), mice were i.p. injected with anti-CD8a antibody (200 pg per mouse) at day - 1 and then at that same dosage at day 2 and day 5. For a control group (G2), mouse IgG2a antibody (200 pg per mouse) was i.p. injected at the same schedule as G3 and G4. Sİ9GM (NVs dose comprising 5 pg of 2'3'-cGAMP per mouse) was injected into mice of G3 and G4 at days 0, 3, and 6. On day 20, all lung tissues were collected, weighed, and used for H & E staining and flow cytometric analysis as mentioned above. Flow cytometric analyses were performed using a BD FACSAria Fusion. Data were analyzed with FlowJo software (version 10.8.1).

4.25. Statistical analysis

All experiments were carried out in triplicates unless otherwise mentioned specifically. The data are presented as mean ± standard deviation (S.D.) of the mean. Statistical significance was evaluated using one-way or two-way analysis of variance (ANOVA) and T-test with the Graph-Pad Prism 8 software (version 10.8.1). Survival was illustrated using the Kaplan-Meier method. A P-value <0.05 was considered statistically significant (·P < 0.05; ··P < 0.01; ···P < 0.001; ····p < 0.0001); ns: no significant difference.

Ethics approval and consent to participate

Animal care strictly followed the law issued by the National Institute of Health (NIH publication 80-23, revised in 1996). All animal experiments were conducted following thorough review and approval by the Institutional Animal Care and Use Committee (IACUC) of Sungkyunkwan University (IACUC number: 202310631).

CRediT authorship contribution statement

Nguyen Thi Nguyen: Writing - review & editing, Writing - original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Xuan Thien be: Validation, Methodology, Investigation, Formal analysis. Woo Tak bee: Validation, Data curation. Yong Taik him: Data curation. Kyung Taek Oh: Writing - review & editing. Eun Seong bee: Writing - review & editing. HanGon Choi: Writing - review & editing. Yu Seok Youn: Writing - review & editing, Supervision, Project administration, Funding acquisition, Data curation, Conceptualization.

Declaration of competing interest

All authors declare no competing financial interest.

Acknowledgments

This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT; No. RS2024-00352440 and No. NRF-2019R1A5A2027340). We thank Prof. Yong Taik bim for providing OT-I mice from baboratory for Immune Modulation, Sungkyunkwan University. Some parts of the schematic illustrations in Graphical abstract, Scheme 1, Figs. 1A-4A and 5A-Fig. 6A, Fig. 6B-Fig. 7A, and Fig. 8A were created with BioRender, com.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.bioactmat.2024.09.002.

Sidebar

Received 8 May 2024; Received in revised form 16 August 2024; Accepted 1 September 2024

Available online 6 September 2024

Footnote

Peer review under responsibility of KeAi Communications Co., Ltd.

* Corresponding author.

E-mail address: [email protected] (Y.S. Youn).

References

References

[1] A.R. Shokouhi, Y. Chen, H.Z. Yoh, J. Brenker, T. Alan, T. Murayama, K. Suu, Y. Morikawa, N.H. Voelcker, R. Elnathan, Engineering efficient CAR-T cells via electroactive nanoinjection, Adv. Mater. 35 (44) (2023) e2304122, https://doi. org/10.1002/adma.202304122.

[2] R. Grosser, L. Cherkassky, N. Chintala, P.S. Adusumilli, Combination immunotherapy with CAR T cells and checkpoint blockade for the treatment of solid tumors, Cancer Cell 36 (5) (2019) 471-482, https://doi.Org/10.1016/j. ccell.2019.09.006.

[3] Y. Jiang, N. Krishnan, J. Zhou, S. Chekuri, X. Wei, A.V. Kroll, C.L. Yu, Y. Duan, W. Gao, R.H. Fang, L. Zhang, Engineered cell-membrane-coated nanoparticles directly present tumor antigens to promote anticancer immunity, Adv. Mater. 32 (30) (2020) e2001808, https://doi.org/10.1002/adma.202001808.

[4] S. Go, M. Jung, S. Lee, S. Moon, J. Hong, C. Kim, Y.S. Chung, B.S. Kim, A personalized cancer nano vaccine that enhances T-cell responses and efficacy through dual interactions with dendritic cells and T cells, Adv. Mater. 35 (49) (2023), https://doi.org/10.1002/adma.202303979.

[5] S. Yu, C. Wang, J. Yu, J. Wang, Y. Lu, Y. Zhang, X. Zhang, Q. Hu, W. Sun, C. He, X. Chen, Z. Gu, Injectable bioresponsive gel depot for enhanced immune checkpoint blockade, Adv. Mater. 30 (28) (2018) el801527, https://doi.org/ 10.1002/adma.201801527.

[6] C. Wang, W.J. Sun, Y.Q. Ye, Q.Y. Hu, H.N. Bomba, Z. Gu, Activation of platelets with checkpoint inhibitors for post-surgical cancer immunotherapy, Nat. Biomed. Eng. 1 (2) (2017), https://doi.org/10.1038/s41551-016-0011.

[7] J.Y. Ye, B. Hou, M. Saeed, Z.A. Xu, H.J. Yu, Engineering bioinspired nanomedicines to mitigate the resistance to cancer immunotherapy, Acc. Mater. Res. (2022), https://doi.org/10.1021/accountsmr.2c00042.

[8] Y. Zeng, S. Li, S. Zhang, L. Wang, H. Yuan, F. Hu, Cell membrane coatednanoparticles for cancer immunotherapy, Acta Pharm. Sin. В 12 (8) (2022) 3233-3254, https://doi.Org/10.1016/j.apsb.2022.02.023.

[9] Y. Huang, X. Gao, J. Chen, Leukocyte-derived biomimetic nanoparticulate drug delivery systems for cancer therapy, Acta Pharm. Sin. В 8 (1) (2018) 4-13, https:// doi.org/10.1016/j.apsb.2017.12.001.

[10] R.H. Fang, Y. Jiang, J.C. Fang, L. Zhang, Cell membrane-derived nanomaterials for biomedical applications, Biomaterials 128 (2017) 69-83, https://doi.org/10.1016/ j.biomaterials.2017.02.041.

[11] N.T. Nguyen, J. Kim, X.T. Le, W.T. Lee, E.S. Lee, K.T. Oh, H.G. Choi, Y.S. Youn, Amplified fenton-based oxidative stress utilizing ultraviolet upconversion luminescence-fueled nanoreactors for apoptosis-strengthened ferroptosis anticancer therapy, ACS Nano 17 (1) (2023) 382-401, https://doi.org/10.1021/ acsnano.2c08706.

[12] J. Constantino, C. Gomes, A. Falcao, B.M. Neves, M.T. Cruz, Dendritic cell-based immunotherapy: a basic review and recent advances, Immunol. Res. 65 (4) (2017) 798-810, https://doi.org/! 0.1007/sl 2026-017-8931-1.

[13] F. Oroojalian, M. Beygi, B. Baradaran, A. Mokhtarzadeh, M.A. Shahbazi, Immune cell membrane-coated biomimetic nanoparticles for targeted cancer therapy, Small 17 (12) (2021) e2006484, https://doi.org/10.1002/smll.202006484.

[14] J.A. Soto, N.M.S. Galvez, C.A. Andrade, G.A. Pacheco, K. Bohmwald, R.V. Berrios, S.M. Bueno, A.M. Kalergis, The role of dendritic cells during infections caused by highly prevalent viruses, Front. Immunol. 11 (2020) 1513, https://doi.org/ 10.3389/fimmu.2020.01513.

[15] T.R. Mempel, S.E. Henrickson, U.H. Von Andrian, T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases, Nature 427 (6970) (2004) 154-159, https://doi. org/10.1038/nature02238.

[16] A.M. Dudek, S. Martin, A.D. Garg, P. Agostinis, Immature, semi-mature, and fully mature dendritic cells: toward a DC-cancer cells interface that augments anticancer immunity, Front. Immunol. 4 (2013) 438, https://doi.org/10.3389/ fimmu.2013.00438.

[17] C. Liu, X. Liu, X.C. Xiang, X. Pang, S.Y. Chen, Y.M. Zhang, E. Ren, L.L. Zhang, X. Liu, P. Lv, X.Y. Wang, W.X. Luo, N.S. Xia, X.Y. Chen, G. Liu, A nanovaccine for antigen self-presentation and immunosuppression reversal as a personalized cancer immunotherapy strategy, Nat. Nanotechnol. 17 (5) (2022) 531-540, https://doi. org/10.1038/s41565-022-01098-0.

[18] M. Jung, M. Kang, B.S. Kim, J. Hong, C. Kim, C.H. Koh, G. Choi, Y. Chung, B.S. Kim, Nanovesicle-mediated targeted delivery of immune checkpoint blockades to potentiate therapeutic efficacy and prevent side effects, Adv. Mater. 34 (9) (2022) e2106516, https://doi.org/10.1002/adma.202106516.

[19] S. Cheng, C. Xu, Y. Jin, Y. Li, C. Zhong, J. Ma, J. Yang, N. Zhang, Y. Li, C. Wang, Z. Yang, Y. Wang, Artificial mini dendritic cells boost T cell-based immunotherapy for ovarian cancer, Adv. Sei. 7 (7) (2020) 1903301, https://doi.org/10.1002/ advs.201903301.

[20] X.Y. Ma, L. Kuang, Y. Yin, L. Tang, Y. Zhang, Q. Fan, B.Y. Wang, Z.F. Dong, W. Wang, T.Y. Yin, Y.Z. Wang, Tumor-antigen activated dendritic cell membranecoated biomimetic nanoparticles with orchestrating immune responses promote therapeutic efficacy against glioma, ACS Nano 17 (3) (2023) 2341-2355, https:// doi.org/10.1021/acsnano.2c09033.

[21] J. Zhang, B. Fan, G.L. Cao, W.P. Huang, F.H. Jia, G.J. Nie, H. Wang, Direct presentation of tumor-associated antigens to induce adaptive immunity by personalized dendritic cell-mimicking nanovaccines, Adv. Mater. 34 (47) (2022), https://doi.org/10.1002/adma.202205950.

[22] J.A. Hubbell, M.A. Swartz, Trojan horses for immunotherapy, Nat. Nanotechnol. 14 (3) (2019) 196-197, https://doi.org/10.1038/s41565-019-0380-7.

[23] V.K. Gonugunta, T. Sakai, V. Pokatayev, K. Yang, J. Wu, N. Dobbs, N. Yan, Trafficking-mediated sting degradation requires sorting to acidified endolysosomes and can be targeted to enhance anti-tumor response, Cell Rep. 21 (11) (2017) 3234-3242, https://doi.Org/10.1016/j.celrep.2017.ll.061.

[24] F. Chen, T. Li, H. Zhang, M. Saeed, X. Liu, L. Huang, X. Wang, J. Gao, B. Hou, Y. Lai, C. Ding, Z. Xu, Z. Xie, M. Luo, H. Yu, Acid-ionizable iron nanoadjuvant augments sting activation for personalized vaccination immunotherapy of cancer, Adv. Mater. 35 (10) (2023) e2209910, https://doi.org/10.1002/adma.202209910.

[25] L. Wang, H. Zhou, Q. Chen, Z. Lin, C. Jiang, X. Chen, M. Chen, L. Liu, L. Shao, X. Liu, J. Pan, J. Wu, J. Song, J. Wu, D. Zhang, STING agonist-loaded nanoparticles promotes positive regulation of type I interferon-dependent radioimmunotherapy in rectal cancer, Adv. Sei. 11 (7) (2024) e2307858, https://doi.org/10.1002/ advs.202307858.

[26] M. Luo, H. Wang, Z. Wang, H. Cai, Z. Lu, Y. Li, M. Du, G. Huang, C. Wang, X. Chen, M.R. Porembka, J. Lea, A.E. Frankel, Y.X. Fu, Z.J. Chen, J. Gao, A STING-activating nanovaccine for cancer immunotherapy, Nat. Nanotechnol. 12 (7) (2017) 648-654, https://doi.org/10.1038/nnano.2017.52.

[27] D.D. Xu, J. Hu, J.W. Mei, J. Zhou, Z.X. Wang, X.D. Zhang, Q. Liu, Z. Su, W.B. Zhu, H.J. Liu, C. Zhu, Nanoadjuvant-triggered STING activation evokes systemic immunotherapy for repetitive implant-related infections, Bioact. Mater. 35 (2024) 82-98, https://doi.Org/10.1016/j.bioactmat.2024.01.020.

[28] H. Zheng, B.B. Guo, X.Y. Qiu, Y.F. Xia, Y. Qu, L. Cheng, F.H. Meng, Z.Y. Zhong, Polymersome-mediated cytosolic delivery of cyclic dinucleotide sting agonist enhances tumor immunotherapy, Bioact. Mater. 16 (2022) 1-11, https://doi.org/ 10.1016/j.bioactmat.2022.02.029.

[29] Y. Li, X. Li, J. Yi, Y. Cao, Z. Qin, Z. Zhong, W. Yang, Nanoparticle-mediated sting activation for cancer immunotherapy, Adv. Healthcare Mater. 12 (19) (2023) e2300260, https://doi.org/10.1002/adhm.202300260.

[30] D. Shae, K.W. Becker, P. Christov, D.S. Yun, A.K.R. Lytton-Jean, S. Sevimli, M. Ascano, M. Kelley, D.B. Johnson, J.M. Balko, J.T. Wilson, Endosomolytic polymersomes increase the activity of cyclic dinucleotide sting agonists to enhance cancer immunotherapy, Nat. Nanotechnol. 14 (3) (2019) 269-278, https://doi. org/10.1038/S41565-018-0342-5.

[31] X. Sun, Y. Zhang, J. Li, K.S. Park, K. Han, X. Zhou, Y. Xu, J. Nam, J. Xu, X. Shi, L. Wei, Y.L. Lei, J.J. Moon, Amplifying sting activation by cyclic dinucleotidemanganese particles for local and systemic cancer metalloimmunotherapy, Nat. Nanotechnol. 16 (11) (2021) 1260-1270, https://doi.org/10.1038/s41565-02100962-9.

[32] P. Dosta, A.M. Cryer, M.Z. Dion, T. Shiraishi, S.P. Langston, D. Lok, J. Wang, S. Harrison, T. Hatten, M.L. Ganno, V.A. Appleman, G.M. Taboada, N. Puigmal, S. Ferber, S. Kalash, M. Prado, A.L. Rodriguez, W.S. Kamoun, A.O. Abu-Yousif, N. Artzi, Investigation of the enhanced antitumour potency of STING agonist after conjugation to polymer nanoparticles, Nat. Nanotechnol. 18 (11) (2023) 1351-1363, https://doi.org/!0.1038/s41565-023-01447-7.

[33] J.P. Böttcher, C. Reis e Sousa, The role of type 1 conventional dendritic cells in cancer immunity, Trends Cancer 4 (11) (2018) 784-792, https://doi.org/10.1016/ j.trecan.2018.09.001.

[34] X. Ma, L. Kuang, Y. Yin, L. Tang, Y. Zhang, Q. Fan, B. Wang, Z. Dong, W. Wang, T. Yin, Y. Wang, Tumor-antigen activated dendritic cell membrane-coated biomimetic nanoparticles with orchestrating immune responses promote therapeutic efficacy against glioma, ACS Nano 17 (3) (2023) 2341-2355, https:// doi.org/10.1021/acsnano.2c09033.

[35] G. Schreibelt, L.J. Klinkenberg, L.J. Cruz, P.J. Tacken, J. Tel, M. Kreutz, G. J. Adema, G.D. Brown, C.G. Figdor, LJ. de Vries, The C-type lectin receptor CLEC9A mediates antigen uptake and (cross-)presentation by human blood BDCA3 + myeloid dendritic cells, Blood 119 (10) (2012) 2284-2292, https://doi.org/ 10.1182/blood-2011-08-373944.

[36] D. Sancho, O.P. Joffre, A.M. Keller, N.C. Rogers, D. Martínez, P. Hernanz-Falcón, I. Rosewell, C.R.E. Sousa, Identification of a dendritic cell receptor that couples sensing of necrosis to immunity, Nature 458 (7240) (2009) 899-903, https://doi. org/10.1038/nature07750.

[37] Y. Kato, T.M. Steiner, H.Y. Park, R.O. Hitchcock, A. Zaid, J.L. Hor, S. Devi, G. M. Davey, D. Vremec, K.M. Tullett, P.S. Tan, F. Ahmet, S.N. Mueller, S. Alonso, D. M. Tarlinton, H.L. Ploegh, T. Kaisho, L. Beattie, J.H. Manton, D. Fernandez-Ruiz, K. Shortman, M.H. Lahoud, W.R. Heath, I. Caminschi, Display of native antigen on cDCl that have spatial access to both T and В cells underlies efficient humoral vaccination, J. Immunol. 205 (7) (2020) 1842-1856, https://doi.org/10.4049/ jimmunoi.2000549.

[38] M. Sela, M. Poley, P. Mora-Raimundo, S. Kagan, A. Avital, M. Kaduri, G. Chen, O. Adir, A. Rozencweig, Y. Weiss, O. Sade, Y. Leichtmann-Bardoogo, L. Simchi, S. Aga-Mizrachi, B. Bell, Y. Yeretz-Peretz, A.Z. Or, A. Choudhary, I. Rosh, D. Cordeiro, S. Cohen-Adiv, Y. Berdichevsky, A. Odeh, J. Shklover, J. ShainskyRoitman, J.E. Schroeder, D. Hershkovitz, P. Hasson, A. Ashkenazi, S. Stern, T. Laviv, A. Ben-Zvi, A. Avital, U. Ashery, B.M. Maoz, A. Schroeder, Brain-targeted liposomes loaded with monoclonal antibodies reduce alpha-synuclein aggregation and improve behavioral symptoms in Parkinson's disease, Adv. Mater. 35 (51) (2023) e2304654, https://doi.org/10.1002/adma.202304654.

[39] X. Xie, Y. Hu, T. Ye, Y. Chen, L. Zhou, F. Li, X. Xi, S. Wang, Y. He, X. Gao, W. Wei, G. Ma, Y. Li, Therapeutic vaccination against leukaemia via the sustained release of co-encapsulated anti-PD-1 and a leukaemia-associated antigen, Nat. Biomed. Eng. 5 (5) (2021) 414-428, https://doi.org/10.1038/s41551-020-00624-6.

[40] G. Jiang, Z. Huang, Y. Yuan, K. Tao, W. Feng, Intracellular delivery of anti-BCR/ ABL antibody by PLGA nanoparticles suppresses the oncogenesis of chronic myeloid leukemia cells, J. Hematol. Oncol. 14 (1) (2021) 139, https://doi.org/ 10.1186/S13045-021-01150-X.

[41] X. Du, B. Shi, J. Liang, J. Bi, S. Dai, S.Z. Qiao, Developing functionalized dendrimer-like silica nanoparticles with hierarchical pores as advanced delivery nanocarriers, Adv. Mater. 25 (41) (2013) 5981-5985, https://doi.org/10.1002/ adma.201302189.

[42] N.K. Campbell, H.K. Fitzgerald, A. Malara, R. Hambly, C.M. Sweeney, B. Kirby, J. M. Fletcher, A. Dunne, Naturally derived heme-oxygenase 1 inducers attenuate inflammatory responses in human dendritic cells and T cells: relevance for psoriasis treatment, Sci. Rep. 8 (1) (2018) 10287, https://doi.org/10.1038/s41598-01828488-6.

[43] C. Admyre, S.M. Johansson, S. Paulie, S. Gabrielsson, Direct exosome stimulation of peripheral human T cells detected by elispot, Eur. J. Immunol. 36 (7) (2006) 1772-1781, https://doi.org/10.1002/eji.200535615.

[44] S.M. Jin, Y.J. Yoo, H.S. Shin, S. Kim, S.N. Lee, C.H. Lee, H. Kim, J.E. Kim, Y.S. Bae, J. Hong, Y.W. Noh, Y.T. Lim, A nanoadjuvant that dynamically coordinates innate immune stimuli activation enhances cancer immunotherapy and reduces immune cell exhaustion, Nat. Nanotechnol. 18 (4) (2023), https://doi.org/10.1038/ S41565-022-01296-W.

[45] T. Su, Y. Zhang, K. Valerie, X.Y. Wang, S. Lin, G. Zhu, STING activation in cancer immunotherapy, Theranostics 9 (25) (2019) 7759-7771, https://doi.org/10.7150/ thno.37574.

[46] M.P. Longhi, C. Trumpfheller, J. Idoyaga, M. Caskey, I. Matos, C. Kluger, A. M. Salazar, M. Colonna, R.M. Steinman, Dendritic cells require a systemic type I interferon response to mature and induce CD4+ Thl immunity with poly IC as adjuvant, J. Exp. Med. 206 (7) (2009) 1589-1602, https://doi.org/10.1084/ jem.20090247.

[47] J. Guo, L. Huang, Nanodelivery of cGAS-STING activators for tumor immunotherapy, Trends Pharmacol. Sei. 43 (11) (2022) 957-972, https://doi.org/ 10.1016/j.tips.2022.08.006.

[48] M.F. Guien, U. Koch, S.M. Haag, F. Schuler, L. Apetoh, A. Villunger, F. Radtke, A. Ablasser, Signalling strength determines proapoptotic functions of STING, Nat. Commun. 8 (2017), https://doi.org/10.1038/s41467-017-00573-w.

[49] J.W. Hickey, F.P. Vicente, G.P. Howard, H.Q. Mao, J.P. Schneck, Biologically inspired design of nanoparticle artificial antigen-presenting cells for immunomodulation, Nano Lett. 17 (11) (2017) 7045-7054, https://doi.org/ 10.1021/acs.nanolett.7b03734.

[50] O. Danilchanka, J. J. Mekalanos, Cyclic dinucleotides and the innate immune response, Cell 154 (5) (2013) 962-970, https://doi.Org/10.1016/j. cell.2013.08.014.

[51] A. Decout, J.D. Katz, S. Venkatraman, A. Ablasser, The cGAS-STING pathway as a therapeutic target in inflammatory diseases, Nat. Rev. Immunol. 21 (9) (2021) 548-569, https://doi.org/10.1038/s41577-021 -00524-z.

[52] A. Basit, M.G. Cho, E.Y. Kim, D. Kwon, S.J. Kang, J.H. Lee, The cGAS/STING/ TBK1/IRF3 innate immunity pathway maintains chromosomal stability through regulation of p21 levels, Exp. Mol. Med. 52 (4) (2020) 643-657, https://doi.org/ 10.1038/sl2276-020-0416-y.

[53] R. Fang, Q. Jiang, X. Zhou, C. Wang, Y. Guan, J. Tao, J. Xi, J.M. Feng, Z. Jiang, MAVS activates TBK1 and IKKe through TRAFs in nemo dependent and independent manner, PLoS Pathog. 13 (11) (2017) el 006720, https://doi.org/ 10.1371/journal.ppat. 1006720.

[54] G. Schuler, R.M. Steinman, Murine epidermal langerhans cells mature into potent immunostimulatory dendritic cells in vitro, J. Exp. Med. 161 (3) (1985) 526-546, https://doi.Org/10.1084/jem.161.3.526.

[55] J. Liu, X. Zhang, Y. Cheng, X. Cao, Dendritic cell migration in inflammation and immunity, Cell. Mol. Immunol. 18 (11) (2021) 2461-2471, https://doi.org/ 10.1038/S41423-021-00726-4.

[56] M. Buettner, U. Bode, Lymph node dissection - understanding the immunological function of lymph nodes, Clin. Exp. Immunol. 169 (3) (2012) 205-212, https://doi. org/10.1111/j.1365-2249.2012.04602.x.

[57] A. He, X. Li, Z. Dai, Q. Li, Y. Zhang, M. Ding, Z.F. Wen, Y. Mou, H. Dong, Nanovaccine-based strategies for lymph node targeted delivery and imaging in tumor immunotherapy, J. Nanobiotechnol. 21 (1) (2023) 236, https://doi.org/ 10.1186/S12951-023-01989-X.

[58] G. Ghislat, A.S. Cheema, E. Baudoin, C. Verthuy, P.J. Ballester, K. Crozat, N. Attaf, C. Dong, P. Milpied, B. Maiissen, N. Auphan-Anezin, T.P.V. Manh, M. Dalod, T. Lawrence, NF- кВ-dependent IRFl activation programs cDCl dendritic cells to drive antitumor immunity, Sei Immunol 6 (61) (2021), https://doi.org/10.1126/ sciimmunol.abg3570.

[59] J. Wang, S. Li, M. Wang, X. Wang, S. Chen, Z. Sun, X. Ren, G. Huang, B.D. Sumer, N. Yan, Y.X. Fu, J. Gao, STING licensing of type I dendritic cells potentiates antitumor immunity, Sei Immunol 9 (92) (2024) eadj3945, https://doi.org/ 10.1126/sciimmunol.adj3945.

[60] K.E. Pauken, E. J. Wherry, Overcoming T cell exhaustion in infection and cancer, Trends Immunol. 36 (4) (2015) 265-276, https://doi.Org/10.1016/j. it.2015.02.008.

[61] S. Chen, A. Saeed, Q. Liu, Q. Jiang, H. Xu, G.G. Xiao, L. Rao, Y. Duo, Macrophages in immunoregulation and therapeutics, Signal Transduct. Targeted Ther. 8 (1) (2023) 207, https://doi.org/10.1038/s41392-023-01452-l.

[62] L.S. Dungan, N.C. McGuinness, L. Boon, M.A. Lynch, K.H. Mills, Innate IFN-y promotes development of experimental autoimmune encephalomyelitis: a role for NK cells and Ml macrophages, Eur. J. Immunol. 44 (10) (2014) 2903-2917, https://doi.org/10.1002/eji.201444612.

[63] S. Wang, J. Wang, Z. Chen, J. Luo, W. Guo, L. Sun, L. Lin, Targeting M2-like tumorassociated macrophages is a potential therapeutic approach to overcome antitumor drug resistance, npj Precis. Oncol. 8 (1) (2024) 31, https://doi.org/10.1038/ S41698-024-00522-Z.

[64] L. Hidalgo-Garcia, J. Galvez, M.E. Rodriguez-Cabezas, P.O. Anderson, Can a conversation between mesenchymal stromal cells and macrophages solve the crisis in the inflamed intestine? Front. Pharmacol. 9 (2018) 179, https://doi.org/ 10.3389/fphar.2018.00179.

[65] Y. Takeuchi, H. Nishikawa, Roles of regulatory T cells in cancer immunity, Int. Immunol. 28 (8) (2016) 401-409, https://doi.org/10.1093/intimm/dxw025.

[66] A. Marcus, A.J. Mao, M. Lensink-Vasan, L. Wang, R.E. Vance, D.H. Raulét, Tumorderived egamp triggers a STING-mediated interferon response in non-tumor cells to activate the NK cell response, Immunity 49 (4) (2018) 754-763, https://doi.org/ 10.1016/j.immuni.2018.09.016, el-e4.

[67] X. Guan, Cancer metastases: challenges and opportunities, Acta Pharm. Sin. В 5 (5) (2015) 402-418, https://doi.Org/10.1016/j.apsb.2015.07.005.

[68] M. Vyas, M. Requesens, Т.Н. Nguyen, D. Peigney, M. Azin, S. Demehri, Natural killer cells suppress cancer metastasis by eliminating circulating cancer cells, Front. Immunol. 13 (2022) 1098445, https://doi.org/10.3389/fimmu.2022.1098445.

[69] T. Nakamura, T. Sato, R. Endo, S. Sasaki, N. Takahashi, Y. Sato, M. Hyodo, Y. Hayakawa, H. Harashima, STING agonist loaded lipid nanoparticles overcome anti-PD-1 resistance in melanoma lung metastasis via NK cell activation, J Immunother Cancer 9 (7) (2021), https://doi.org/10.1136/jitc-2021-002852.

[70] X. Du, X. Li, H. Huang, J. He, X. Zhang, Dendrimer-like hybrid particles with tunable hierarchical pores, Nanoscale 7 (14) (2015) 6173-6184, https://doi.org/ 10.1039/c5nr00640f.

[71] S. Gou, W. Liu, S. Wang, G. Chen, Z. Chen, L. Qiu, X. Zhou, Y. Wu, Y. Qi, Y. Gao, Engineered nanovaccine targeting clec9a(+) dendritic cells remarkably enhances the cancer immunotherapy effects of sting agonist, Nano Lett. 21 (23) (2021) 9939-9950, https://doi.org/10.1021/acs.nanolett.lc03243.

[72] L. Liu, S. Fan, Z. Lu, Z. Chen, C. Chu, A. Liu, F. Xia, S. Meng, F. Guo, H. Qiu, Y. Yang, An optimized method for the induction and purification of mouse bone marrow dendritic cells, J. Immunol. Methods 495 (2021) 113073, https://doi.org/ 10.1016/j .jim.2021.113073.

[73] D. van Dinther, H. Veninga, M. Revet, L. Hoogterp, K. Ölesek, J. Grabowska, E.G. F. Borg, H. Kalay, Y. van Kooyk, J.M.M. den Haan, Comparison of protein and peptide targeting for the development of a CDI 69-based vaccination strategy against melanoma, Front. Immunol. 9 (2018) 1997, https://doi.org/10.3389/ fimmu.2018.01997.