Extracellular vesicles derived from mature dendritic cells loaded with cDC1-specific chemokine XCL1 combined with chemotherapy-induced ICD for the treatment of castration-resistant prostate cancer

Abstract

Background

Dendritic cells (DCs) derived extracellular vesicles represent a promising vehicle for the activation of adaptive immunity, demonstrating significant potential in the development of cancer nanovaccines. The aim of this study was to evaluate the antitumor efficacy of a functional DCs-derived extracellular vesicles in castration-resistant prostate cancer.

Methods

A pre-modification strategy was employed to overexpress XCL1 in dendritic cells, enabling their extracellular vesicles to highly express XCL1 protein. In vitro experiments, prostate cancer-bearing mouse models, and OVA-expressing prostate cancer mouse models demonstrated that dendritic cells efficiently internalize extracellular vesicles derived from XCL1-overexpressing mature dendritic cells (DEX_XCL1), thereby enhancing the chemotaxis, activation, and antigen-presenting capacity of cDC1 cells. When combined with the immunogenic cell death effect induced by cisplatin, this approach significantly increased the number and cytotoxic activity of CD8⁺ T cells, improved the tumor microenvironment, and effectively suppressed prostate cancer tumor growth.

Results

The coding sequence of XCL1 successfully inserted upstream of the PDGFR transmembrane domain and transfected into dendritic cells, enabling their extracellular vesicles to highly express XCL1 protein. Extracellular vesicles derived from XCL1-overexpressing mature dendritic cells not only exhibited high XCL1 expression, but were also enriched with chemokine receptor CCR7 and MHC I molecules on their surface. This nanovaccines enhanced the uptake of extracellular vesicles by dendritic cells, recruited cDC1 cells within the tumor tissue, and significantly improved their antigen-presenting capacity. When combined with the immunogenic cell death effect induced by cisplatin, which generates a large amount of tumor-associated antigen STEAP1, this strategy effectively enhanced the proliferation and cytotoxic activity of CD8⁺T cells. Moreover, it reduced the proportion of regulatory T cells and immunosuppressive factors, thereby reshaping the tumor immune microenvironment. This approach effectively inhibited tumor growth in mice and prolonged their survival. These findings demonstrate the strong synergistic effects of the nanovaccines and cisplatin in promoting antitumor immunity.

Conclusions

A novel nanovaccines induces potent antitumor immune responses and, in combination with the chemotherapeutic agent cisplatin, effectively remodels the tumor immune microenvironment. This approach offers a new strategy and preclinical evidence for the immunotherapy of “cold tumors” prostate cancer.

Introduction

Prostate cancer poses a significant global health challenge for men and is the most common malignancy among males in developed countries [1]. Early-stage localized lesions, when detected, can achieve satisfactory outcomes through surgical intervention, radiotherapy, or interventional therapies [2]. However, approximately 30% of patients experience biochemical recurrence, and one-third of these patients ultimately succumb to the disease [3]. Androgen deprivation therapy (ADT) has long been the cornerstone of treatment for advanced prostate cancer [4]. However, within 3–5 years, most patients develop resistance, progressing to castration-resistant prostate cancer (CRPC), the lethal stage of the disease [5]. Therapeutic options for CRPC include chemotherapy, novel endocrine therapies, poly (ADP-ribose) polymerase (PARP) inhibitors, radionuclide-based therapies, immune checkpoint inhibitors, and autologous cell-based vaccines targeting prostate-specific antigens [6]. Although advancements in risk stratification, molecular subtyping diagnostics, and novel therapeutic approaches have significantly improved the quality of life and prognosis for advanced-stage patients, they still fail to fully meet clinical demands [7, 8].

The continuous development of novel immunotherapies has profoundly revolutionized the therapeutic paradigm for treating malignant tumors in clinical settings, with the success of immune checkpoint inhibitors serving as a prominent example [9–11]. Prostate cancer, as a “cold tumor,” is characterized by a low infiltration of lymphocytes within the tumor microenvironment, rendering immune checkpoint inhibitors largely ineffective [12]. This is particularly evident in the poor recruitment of critical components of antitumor immunity, such as antigen-presenting cDC1 cells and effector CD8⁺ T cells. Specifically, the chemokine XCL1, which facilitates the recruitment of cDC1 cells, exhibits extremely low expression levels within the tumor [13]. Studies have reported that elevated XCL1 expression is positively correlated with improved prognosis in patients [14, 15]. Therefore, the effective delivery of XCL1 to tumors and their draining lymph nodes is of critical importance. Extracellular vesicles (EVs) derived from dendritic cells are enriched on their outer membrane with various antigen-presenting molecules, adhesion molecules, and co-stimulatory molecules, retaining the critical functions of dendritic cells and providing a bridge to activate dendritic cells [16, 17]. In addition, EVs serve as excellent delivery vehicles due to their low immunogenicity and intrinsic ability to passively target secondary lymphoid organs, allowing preferential uptake by antigen-presenting cells [18]. In this study, we employed genetic engineering techniques to load XCL1 onto exosomes, enabling them to recruit cDC1 cells. We aim to evaluate whether this approach can improve the immune microenvironment of prostate cancer tumors.

Cisplatin is the first FDA-approved platinum-based drug for cancer treatment and is widely used in the treatment of various solid tumors [19]. In this study, we demonstrated that cisplatin can induce immunogenic cell death (ICD) while releasing a large number of tumor-associated antigens in tumor-bearing mouse models. Through the synergistic effects of the two therapeutic approaches in various tumor-bearing mouse models, we confirmed their ability to improve the immune microenvironment within prostate tumors, eliciting robust antigen-specific immune responses and tumor suppression. Our findings indicate that XCL1-loaded DEX effectively promotes the recruitment, activation, and antigen presentation of cDC1, inducing potent antitumor immune responses. This provides a novel therapeutic strategy for the treatment of advanced prostate cancer.

Materials and methods

Animals

OT-I (#003831) transgenic mice and C57BL/6 J mice were originally from the Jackson Laboratory. Eight-week-old male C57BL/6 J (purchased from Nanjing Model Organisms Center, Inc., China) were used in all experiments for establishment of subcutaneous tumor model. OT-I (#003831) TCR transgenic mice (purchased from Nanjing Model Organisms Center, Inc., China) were used for isolating the splenocytes. Animals were housed in a controlled temperature environment (23 ± 2 °C) and a fixed 12 h light/dark cycle, having free access to food and water. All animal experiments were conducted in the animal experiment unit of Tianjin Medical University General Hospital, strictly following the experimental protocols authorized and explicitly approved by the Institutional Ethics Committee (permit number IRB2025-DW-01).

Cell lines and cell culture

DC2.4 cells (referred to DC) (H-2 kb) (Purchase from ATCC, China) or XCL1 overexpression DCs(DC_xcl1) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) that contained 1 mM sodium pyruvate and 100 U/ml penicillin and 100 mg/ml streptomycin, 1% glutamine (Gln), 50 μM β-mercaptoethanol and 10% depleted fetal bovine serum (FBS, HyClone, USA) at 37 °C in 5% CO₂. Depleted fetal bovine serum is obtained by centrifugation at 100,000 g for 1 h to remove possible FBS-containing extracellular vesicles. Human 293FT cell line was cultured in DMEM supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin. The murine prostate cancer cell line RM-1 (purchased from Boster Biological Technology Ltd, Wu han, China) was cultured in 1640 supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin. Primary mouse T cells were cultured in RPMI-1640 medium supplemented with 10% (v/v) FBS, 1 mM sodium pyruvate, 50 mM 2-mercaptoethanol and 100 U/ml penicillin and 100 mg/ml streptomycin (RPMI complete). Reagents used for cell culture were purchased from Thermo Fisher Scientific, STEMCELL Technologies, and Nordic Biolabs.

Generation of OVA⁺ prostate cancer cell line or XCL1 overexpression DC cell lines

To obtain the highly enrichment of murine xcl1 sorting on the surface of DC-derived extracellular vesicles, we used the empty backbone pCDH-EF1-Puro and inserted the coding sequence for murine xcl1 by infusion cloning upstream of PDGFR (platelet-derived growth factor receptor) transmembrane domain to generate pCDH-CMV-XCL1-PDGFR-EF1-Puro [20, 21]. The murine XCL1 full-length coding sequence was cloned from total RNA isolated from spleen of C57BL/6 mice and cloned into the lentivirus expression vector pCDH-CMV-puro (System Biosciences, CA, US).

Human 293FT cells (4 × 10⁷) were seeded in a 10-cm Petri-dish for 24 h prior to the co-transfection of OVA-mCherry expressing plasmid (purchased from Addgene), psPAX2 and pMD2G plasmids in a mass ratio of 2:1:1 (40 μg in total) by 8 μl polyethylenimine (PEI) (1 mg/ml). After 6–8 h, the culture medium was replaced with pre-warmed fresh DMEM complete medium. Viral supernatant was collected at 48- and 72-h post-transfection. Cell debris was removed by centrifugation with 1,000 g for 3 min. The aliquot of virus stock was stored at − 80 °C. Prostate cancer cell line RM-1 (1 × 10⁵) were seeded in 12-well plates and infected with 1-ml viral supernatant containing 10 μg/ml polybrene into each well of 12-well plates for 12 h, followed by 2 μg/ml puromycin selection for 2 weeks. Positive cell was harvested, then resuspended in sorting buffer (PBS supplemented with 1 mM EDTA, 25 mM HEPES, and 1% FBS), and sorting and analysis were performed on a FACS Aria (BD).

Extracellular vesicles isolation, characterization and cellular uptake assay

DC2.4 cells and DC_xcl1 were cultured in DMEM medium containing 10% exosome-depleted FBS, 100 U/ml penicillin and 100 μg/ml streptomycin for 24–48 h to 80–90% confluence. For maturation of DC_xcl1, immature DCs were activated with polyI:C (25 µg/ ml, InvivoGen), CD40L (1000 ng/ml, Life Sciences) for 48 h. Cell culture supernatants were and centrifugated at 500 g for 15 min to remove cell debris, and then centrifugated at 10,000 g for 30 min, followed by filtration with 0.22 μm filter(Millex). Then, extracellular vesicles pellets were centrifugated (Beckman) at 100,000 g for 70 min and were resuspended in 600–800ul PBS. All centrifugation steps were performed at 4 °C and exosome resuspending was performed on ice. Extracellular vesicles were quantified as protein measurement using BCA protein assay kit (Thermo Scientific), and stored at − 80 °C. Transmission electron microscopy (TEM, Hitachi, HT-7800), nanoparticle tracking analysis (NTA, Nanosight, NS300) and flow cytometry were performed to characterize the isolated extracellular vesicles. For TEM, the resuspended extracellular vesicles were diluted in PBS and mixed with an equal volume of 4% paraformaldehyde (PFA). Subsequently, the extracellular vesicles were adsorbed onto a metal sample grid. After removing the excess solution, the vesicles were fixed in 1% glutaraldehyde for 5 min, followed by eight washes with distilled water. The sample grid was then transferred to 50 μL of uranyl oxalate solution (pH 7.0) for 5 min and subsequently immersed in 50 μL of methylcellulose-uranyl acetate solution at 4 °C for 10 min. After drying, the samples were observed using transmission electron microscopy (TEM). For NTA, all samples was diluted from 1:500 to 1:1000 to make sure a particle count of between 10⁸ and 109 per milliliter. The camera’s focus was adjusted so that the particles appeared as clear dots. Using the script control function, three 60 s videos were recorded for each sample, with sample advancement and a 5 s delay between recordings. For cellular uptake assay, we labeled extracellular vesicles with Pkh67 (Sigma-Aldrich) as per the manufacturer’s instruction. DC cells derived from lymph node of C57BL/6 mice were incubated with Pkh67 labeled DEX or DEX_xcl1 for 24 h, then subjected to flow cytometry individually analyze of FITC⁺DC ratio.

Western blot

Extracellular vesicles were lysated in NP-40 cell lysis buffer and subjected to 10% SDS–polyacrylamide gel electrophoresis and gels were transferred to a PVDF membrane. The protein was transferred to PVDF membranes and membranes were blocked with 5% skimmed milk and probed with primary antibodies including mouse monoclonal antibodies: XCL1(1:1000, Cell Signaling Technol.), CD63 (1:500, ABclonal). The membranes were blocked with TBS containing 5% nonfat milk in and 0.1% Tween20 (TBST). After washing with TBST, membranes were incubated with appropriate primary antibodies at 4 °C overnight, followed by horseradish peroxidase-conjugated rabbit anti-mouse, or goat anti-rabbit IgG (ABclonal) for 2 h at 4 °C, and development with an enhanced chemiluminescence detection reagent (GE healthcare) and the intensity of the bands was measured by ImageJ software.

Apoptosis assay

The number of apoptotic cells was measured by Annexin V and propidium iodide (PI) staining and flow cytometry. After treatment with cisplatin or DEXxcl1, cells were recovered using trypsin/EDTA, washed with PBS, and resuspended in Annexin V binding buffer (BD Biosciences). Fluoresce in isothiocyanate (FITC)-labeled Annexin V and PI (BD Biosciences) were added to cells and incubated for 15 min. The number of Annexin V⁺ /PI⁻ and Annexin V⁺/PI⁺ events was quantified using an flow cytometer (BD Biosciences). The percentage of apoptotic and necrotic cells was quantified by flow cytometry and then calculated according to the relative ratio of PBS treatment.

ELISA assay

For detection of cytokines in cell culture supernatant, murine LN-DCs (5X10⁴) were seeded in 96-well plates, followed by addition of DEX or DEX_xcl1 (20 μg/ml), for 48 h at 37 °C or tumor tissue homogenate (1 g per ml) harvested from treated mice, 96-well ELISA plates were coated with 1 μg/ml of coating antibody (100 μl/well) overnight at 4 °C. Next day, free binding sites were blocked with 200 μl blocking buffer (PBS containing 0.5% BSA) for 1 h at room temperature. Then, 100ul of cell culture supernatant or tumor tissue homogenate were added to each well. After incubation for 10 h at 4 °C, 0.5 μg/ml of rabbit monoclonal antibody against IL12p70, IFNα, HMGB1, IFNγ, STEAP1or TGFβ diluted in PBS containing 0.5% BSA was added and incubated overnight at 4 °C. After washing with PBS, 100 μl of horseradish peroxidase-conjugated secondary antibody against rabbit diluted in PBS containing 0.5% BSA was added into each well and incubated for 1 h at room temperature. After each step, the wells were washed by PBS three times. Plates were developed with cytokine-specific ELISA kits according to the vendor’s instructions (eBioscience).

Mouse T cells priming assay in vitro

Primary T cells from C57BL/6 J spleen were cultured in RPMI-1640 medium supplemented with 10% FBS, glutamine (1 mM), HEPES (10 mM), sodium pyruvate (1 mM), β-mercaptoethanol (50 μM), penicillin (100 U/ml) and streptomycin (100 μg/ml). For in vitro T cell early activation assays, splenic naïve CD4⁺ or CD8⁺ T cells were activated by plate-bound anti-CD3 (1 μg/ml) and anti-CD28 (1 μg/ml) in the absence or presence of activated DC stimulated by DEX or DEX_xcl1 for 72 h in T cell culture medium. After being stimulated for indicated time, the cells were collected and analyzed by flow cytometry.

Tumor establishment and treatment of subcutaneous prostate cancer mouse models

To establish subcutaneous tumor models, live cells counted by Trypan blue staining using an automated hemocytometer, then 2X10⁵RM-1 or 4X10⁵ OVA⁺ RM-1 was harvested with and injected subcutaneously into the right flanks of recipient mice (C57BL/6 J mice) in 100 μl of ice-cold PBS. Tumor volumes were monitored with a digital caliper and calculated using the formula V = L*W*H/2. Survival was determined by predefined endpoints such as tumor size reaching 1500 mm³, tumor ulceration, or signs of animal suffering. Animals were randomized for tumor establishment and again before treatment. When tumors reached 10–15 mm³size, the tumor-bearing mice were treated with cisplatin combined with DEX_xcl1, were orally administered with Lenvatinib, and 3 days later administration with subcutaneous injection (S.C.) of DEX_xcl1 at 80ug exosome per mouse once a week for 3 weeks, using PBS treatment as negative control.

Isolation of leukocytes from mouse spleen, TdLN and tumor tissues

For the isolation of leukocytes from spleen and inguinal lymph nodes, tissues were dissected from C57BL/6 J mice, and were minced with dissection scissors, and were mashed and 40 μm cell-filtered. Red blood cells were removed by ACK (STEM CELL) lysis buffer. For the isolation of leukocytes from tumor tissues, tumors were dissected from mice, and were minced with dissection scissors, then were digested in RPMI-1640 containing 20 μg/ml of DNase I (Roche) and according to operation manual of Tumor Dissociation Kit (MiltenyiBiotec) at 37 C with 220 rpm shaking for 2 h. After digestion, the tissues were then ground, filtered with 70 μm nylon sieve (BD), collected by centrifugation at 500 g for 5 min. Supernatant was aspirated and the mononuclear immune cells were isolated by centrifugation on a 40 to 72% Percoll gradient. Red blood cells were removed by ACK lysis buffer. Single cell suspensions were then washed, filtered and then collected by centrifugation, for flow cytometry detection.

In vitro BMDCs activation

BMDCs were cultured as described above for activation assays. BM progenitors isolated from BMs of C57BL/6 mice were cultured in (RPMI) 1640 medium plus 10% FBS, 1% P/S (100 U ml − 1 penicillin and 100 μg ml − 1 streptomycin), murine granulocyte macrophage colony-stimulating factor (GM-CSF) (200 U ml − 1) (PeproTech) and interleukin-4 (IL-4) (100 U ml − 1) (PeproTech), and 50 μmol l − 1 β-mercaptoethanol at 37 °C in a humidified incubator with 5% CO2 for 7–10 days to generate immature DCs. Subsequently, immature DCs were incubated in fresh culture medium in the absence or presence of DEX or DEX_xcl1 for 48 h, prior to functional and phenotypical analysis. Adherent BMDCs were dissociated using non-enzymatic cell dissociation buffer (Gibco), resuspended in FACS buffer and were immunostained and analyzed by flow cytometry for the expression of the indicated surface molecules for BMDC activation panel included CD83 (Biolegend), CCR7 (Biolegend).

Flow cytometry and intracellular staining

All the leukocytes from mouse spleen, TdLN and tumor tissues or cultured cells in vitro were filtered through 70 µm cell strainers and stained in FACS buffer. Ghost dye cell viability reagent was used to exclude dead cells (diluted 1:1,000 in PBS). Extracellular antibodies for lymphoid immunophenotyping included: CD4 (Biolegend), CD45 (BD Biosciences), CD8a (Biolegend), CD3 (Thermo). For myeloid immunophenotyping, extracellular antibodies included: XCR1 (Biolegend), CD11b (Thermo), SIRPα (BD Biosciences), CD11c (BD Biosciences), H2K^b-SIINFEKL (Biolegend), each used at 1:500 dilution. For testing CRT expression in tumor: CRT (Thermo) at a dilution of 1:50, Alexa Fluor 488-conjugated goat anti-rabbit IgG (Thermo) was used as the secondary antibody at a dilution of 1:1000. After intracellular staining, cells were washed with FACS buffer, and fixed using the Foxp3/transcription factor staining buffer set (BD Biosciences), as per the manufacturer’s instructions. After washing with 1 × BD Perm/Wash buffer, cells were fixed and permeabilized with BD Cytofix/Cytoperm solution for 30 min, and then stained with Intracellular antibodies for lymphoid immunophenotyping included: Foxp3 (Thermo), Ki67 (Thermo), Granzyme-B (Biolegend) each used at 1:200 dilution in 1 × BD Perm/Wash buffer. After staining, cells were washed and resuspended with FACS buffer for flow cytometry analysis using a BDLSRFortessa. Collected flow cytometry data were analysed using FlowJo V10.10.0 software.

T cell proliferation assay in vivo

For carboxyfluorescein diacetate succinimidyl ester (CFSE) dilution assay, splenocytes from naive OT-I mice were harvested, filtered through 40 µm cell strainers and collect the purified OT-I by using with the EasySepMouse T Cell Isolation Kit (StemCell Technologies) according to the manufacturer’s instructions, then washed in complete RPMI-1640 supplemented with 10%(vol/vol) FBS, 1 × GlutaMax, 1% NEAA (Gibco) and 100 U ml − penicillin–streptomycin). Cells were labeled with 0.5 μM of CFSE (Life Technology) in PBS at 37 °C for 15 min. Then, the cells were washed with PBS to remove the excess dye and were centrifuged at 300 g for 5 min. Then, CFSE-labeled OT-I CD8⁺T cells (1 × 10⁷) T cells were transferred into the treated bearing mice with Tail vein system injection. Seven days post the injection, lymphocytes were harvested from each mice for flow cytometry analysis. All flow cytometry data were acquired on LSRFortessa (BD) and were analyzed using FlowJo V10.10.0 software.

Statistical analysis

Statistical differences between treatment and control groups were evaluated by GraphPad Prism (v.8.2.1). All data obtained after three biological replicates and were means ± SEM. Unpaired two-sided student t test was used to calculate significance, unless stated otherwise. Both parametric and nonparametric analyses were applied, in which the Mann–Whitney rank sum test (Mann–Whitney U test) was used for samples on a non-normal distribution, whereas a two-tailed t test was performed for samples with a normal distribution, respectively. Tumor growth were assessed by two-way ANOVA. Survival comparison was performed using log-rank (Mantel-Cox) test. Significance was determined based in p < 0.05.

Results

Characterization of extracellular vesicles

We first used a pre-modification strategy to overexpress the chemokine XCL1 in the conventional dendritic cell line DC 2.4. Schematic diagram of DEX_XC1 production and application is shown in Fig. 1A. The expression of XCL1 was verified using Western blot analysis. As shown in Fig. 1B, dendritic cells successfully overexpressed XCL1. To evaluate whether XCL1 overexpression affected the characteristics of extracellular vesicles, we compared extracellular vesicles derived from XCL1-overexpressing dendritic cells (DEX_xcl1) with those from unmodified dendritic cells. DEX_xcl1 exhibited the typical cup-shaped morphology, an average diameter of 116 nm, and high levels of XCL1 expression (Fig. 1C–E). The literature reports that DEX (dendritic cell-derived extracellular vesicles) carrying co-stimulatory molecules and MHC-peptide complexes on their surface can effectively activate T lymphocytes, demonstrating the potential to initiate adaptive immune responses. However, some studies have pointed out that the ability of mature dendritic cell (DC)-derived extracellular vesicles and immature DC-derived extracellular vesicles to induce helper T cell responses in vitro varies. Therefore, in this study, we activated dendritic cells using Poly I: C and CD40L to evaluate the expression levels of the co-stimulatory molecule CD86 on extracellular vesicles secreted by immature DCs, immature DCs overexpressing XCL1, and mature DCs overexpressing XCL1. Flow cytometry analysis revealed that extracellular vesicles derived from mature DCs overexpressing XCL1 exhibited the highest surface expression of CD86 (Fig. 1F-1) (CD86 MFI in Fig. 1G as follows: NC group: 38.90 ± 4.49; DEX group: 647.00 ± 36.96; immature DEX_XCL1 group: 918.00 ± 89.14; mature DEX_XCL1 group: 3717.33 ± 125.59). These findings indicate that extracellular vesicles derived from mature dendritic cells carry relevant parent cell-derived proteins and possess the capacity to activate the adaptive immune system.

Fig. 1.

Characterization of the chemokine receptor-XCR1 specific ligand modified dendritic cell-derived extracellular vesicles in vitro. A Schematic diagram of XCL1 overexpression DC 2.4 cells derived extracellular vesicles (DEX_XCL1) production and application; B XCL1 overexpress efficiency in DC 2.4 cell line via western blot; C Representative transmission electron microscopic (TEM) images (scale bar, 200 nm) and D Size distribution with nanoparticle tracking analysis (NTA) of DC 2.4 cells derived extracellular vesicles (DEX) or DEX_XCL1. E Western blot analysis of XCL1 in the purified extracellular vesicles from DEX and DEX_XCL1, probed with antibodies against exosome marker CD9. All lanes were loaded with the same amount of total protein. F, G Flow cytometry for analyzing the DC maturation relative surface marker CD86 expression level on DEX and DEX_XCL1 (“im”: immaturation; “m” indicted maturation induction of XCL1 overexpression DC2.4 cells with PolyI: C and CD40L),all DEX_XCL1 were mentioned in the article refers to maturation induction of XCL1 overexpression DC2.4 cells with Poly I: C and CD40L derived exosome (n = 3, *P < 0.05, ***P < 0.001)

To investigate the uptake of dendritic cell-derived extracellular vesicles (DEX) from different sources by bone marrow dendritic cells, we labeled the extracellular vesicles with PKH67 and analyzed their internalization. As shown in Fig. 2A, extracellular vesicles derived from mature dendritic cells overexpressing XCL1 were more readily taken up by dendritic cells compared to those from immature dendritic cells (NC group: 1.05 ± 0.4; DEX group:71.96 ± 2.38; DEX_XCL1 group: 84.26 ± 1.70). Furthermore, we assessed the expression of MHC I (NC group: 721.70 ± 29.87; DEX group: 713.30 ± 103.10; DEX_XCL1 group: 1873.00 ± 79.98) and CCR7 (NC group: 116.70 ± 5.61; DEX group: 121.30 ± 5.49; DEX_XCL1 group: 404.30 ± 40.88) on dendritic cells after exosome uptake (Fig. 2B, C). The results showed that treatment with extracellular vesicles from mature dendritic cells overexpressing XCL1 significantly upregulated the expression levels of MHC I and CCR7, along with increased IL12p70 levels (NC group: 24.67 ± 7.84; DEX group: 65.28 ± 4.54; DEX_XCL1 group: 180.60 ± 21.93) in the cell supernatant. These findings suggest that XCL1-enriched DEX can activate dendritic cells and enhance their chemotactic capacity toward secondary lymphoid organs. Subsequently, dendritic cells from different treatment groups were co-cultured with murine splenocytes. Flow cytometry analysis revealed a significant increase in the proportion of CD8⁺CD69⁺ T cells (NC group: 27.93 ± 1.90; DEX group: 35.73 ± 1.56; DEX_XCL1 group: 61.27 ± 2.07) in the DEX_XCL1 group (Fig. 2D), as well as a marked elevation in the proportion of IFN-γ-positive CD8⁺ T cells (NC group: 2.06 ± 0.48; DEX group: 3.53 ± 0.46; DEX_XCL1 group: 9.56 ± 1.17) (Fig. 2D). In conclusion, extracellular vesicles derived from mature dendritic cells carrying XCL1 protein exhibit superior ability in activating dendritic cells. Therefore, we have selected this type of exosome for subsequent experiments.

Fig. 2.

Investigation of uptake efficiency and antitumor immune response effect of DEX_XCL1 in vitro. A Flow cytometric analysis of DEX and DEX_XCL1 uptake in DC cells derived from lymph node of C57BL/6 mice 24 h after incubation. NC (negative control) refers to PBS treated DCs (n = 3). B Flow cytometry for analyzing levels of surface proteins (MHCI and CCR7) in DCs treated with PBS, DEX and DEX_XCL1 (n = 3). C Measurement of IL12p70 cytokine levels in the culture supernatants were quantitated by cytokine array (n = 3). D Representative flow cytometric analysis of CD69⁺CD8⁺ for 12 h incubation or IFN-γ⁺ CD8⁺ for 72-h incubation in CD8⁺splenic T cells harvested from immunized C57BL/6mice with DEXorDEX_xcl1 activated DCs (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001)

Cisplatin-induced immunogenic cell death in prostate cancer cells

Cisplatin is a first-line treatment for castration-resistant prostate cancer (CRPC). Studies have shown that chemotherapeutic agents can induce immunogenic cell death (ICD) by killing tumor cells while releasing large amounts of antigens STEAP1. To evaluate the ICD effect of cisplatin in prostate cancer cells, we conducted in vitro experiments to assess its ability to induce apoptosis. Compared to the exosome-treated group, cisplatin significantly increased early (NC group: 3.80 ± 0.40; DEX group: 3.03 ± 0.94; cisplatin group: 46.33 ± 2.18) and late apoptosis (NC group: 2.85 ± 0.34; DEX group: 2.24 ± 0.20; cisplatin group: 23.55 ± 2.01). In subcutaneous tumor-bearing mice treated with cisplatin, tumor tissues were homogenized, and the expression levels of IFN-α (NC group: 9.40 ± 2.58; DEX group: 11.28 ± 1.31; cisplatin group: 56.37 ± 6.42) and HMGB1 (NC group: 1.00 ± 0.21; DEX group: 2.30 ± 0.70; cisplatin group: 23.41 ± 2.42), key markers of ICD, were analyzed using ELISA kits. Compared to the PBS group and the exosome-treated group, cisplatin effectively induced immunogenic cell death in prostate cancer cells, releasing substantial tumor-associated antigen STEAP1, with STEAP1 levels increasing over time (Fig. 3A–C). Therefore, the ICD effect induced by cisplatin leads to the release of a large number of tumor-associated antigens in mice, laying the foundation for the combined use of DEX_xcl1 to recruit cDC1 cells and elicit an antitumor immune response.

Fig. 3.

Cisplatin triggers immunogenic cell death and antigen releasing in vitro or vivo. A Mouse prostate cancer cells were treated with cisplatin for 24 h followed by the assessment of Annexin V/PI staining (n = 3). B Elisa assay quantification of proteins (IFNα or HMGB1) levels in tumor homogenate following cisplatin intumoral administration for 72 h (n = 3). C ELISA dynamic detection of antigen protein STEAP1 levels in serum of tumor-bearing mice at 24,48, 72,96 h post-cisplatin administration (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001)

Combined therapeutic effects of DEX_XCL1 and cisplatin

A subcutaneous tumor-bearing mouse model was successfully established and divided into four groups: PBS control, cisplatin treatment, DEX_xcl1 treatment, and the combination of DEX_xcl1 with cisplatin. After 21 days, the mice were sacrificed, and tumor growth was analyzed. As shown in Fig. 4A–C, the combination treatment group exhibited significantly reduced tumor size and weight (NC group: 1.55 ± 0.10; cisplatin group: 0.98 ± 0.09; DEX_XCL1 group: 0.73 ± 0.05, cisplatin and DEX_XCL1 group: 0.25 ± 0.06), both in terms of growth rate and tumor weight. Additionally, the survival outcomes of the mice in different treatment groups were assessed. As shown, the 4-month survival rate of the combination treatment group was 62%, markedly higher than that of the other groups (Fig. 4D). This section of the results provides preliminary evidence supporting the complementarity and superiority of the combined application of the two therapeutic approaches.

Fig. 4.

Systemic evaluation of cisplatin@DEX_xcl1 for antitumor immunity in subcutaneous prostate cancer mice. Subcutaneous tumor-bearing C57BL/6 mice were treated with cisplatin (30 mg/kg, i.v injection), DEX_xcl1 or cisplatin@DEX_xcl1 (80ug/mouse/week for three weeks, s.c injection), NC (normal control) refers to PBS treated group. A, B Tumor size and weight (g) of nude mice in different treatment groups; C Measurement of tumor volume at different time-points after inoculation. Significant differences were obtained between cisplatin@DEX_xcl1 and other groups. D Survival rate of treated subcutaneous tumor-bearing mice (n = 4, *P < 0.05, **P < 0.01, ***P < 0.001)

To evaluate the antitumor response against non-prostate cancer-specific antigens, we utilized a prostate cancer-bearing mouse model expressing ovalbumin (OVA). Host immune responses were assessed by detecting OVA-specific immune reactions. To investigate the capacity of DEX_xcl1 to recruit cDC1 cells, we observed a significant increase in the proportion of XCR1⁺SIRPα⁻ cDC1 cells within lymphocytes in the tumor tissues (NC group: 4.59 ± 0.24; cisplatin group: 3.86 ± 0.38; DEX_XCL1 group: 16.12 ± 3.08, cisplatin and DEX_XCL1 group: 17.55 ± 2.77) (Fig. 5A). Furthermore, the expression levels of the cDC1-associated chemokine receptor CCR7 and the co-stimulatory molecule CD83 (NC group:67.75 ± 5.27; cisplatin group:73.00 ± 3.85; DEX_XCL1 group: 163.00 ± 9.09, cisplatin and DEX_XCL1 group: 186.00 ± 7.43) were markedly elevated in the DEX_xcl1treated group (Fig. 5B), with CCR7 (NC group: 88.25 ± 6.44; cisplatin group:132.50 ± 7.75; DEX_XCL1 group: 267.80 ± 6.92, cisplatin and DEX_XCL1 group: 410.80 ± 13.33) levels further enhanced in the combination treatment group.

Fig. 5.

Cisplatin combined with DEX_XCL1 promote cDC1 recruitment, activation and tumor antigen presentation in tumor tissue of prostate cancer mice. A Flow cytometric and quantitative analysis of XCR1⁺SIRPα⁻cDC1 in CD11c⁺TILs from prostate cancer tumor mice (n = 4) B Flow cytometric and quantitative analysis of ccl19 or ccl21 specific surface chemokine receptors-CCR7 and key co-stimulatory factor-CD83 on XCR1⁺SIRPα⁻cDC1 from tumors of tumor tissue of prostate cancer mice treated with cisplatin, DEX_XCL1, cisplatin combine with DEX_XCL1 or PBS treated group (NC) (n = 4). C Flow cytometric and quantitative analysis of OVA-specific MHCI epitope-SIINFEKL presentation efficiency of intratumor DCs (n = 4). D Representative flow cytometric analysis of CFSE dilution on OT-1 T cells in the tumor-draining lymph nodes of the tumor-bearing mice with CFSE-labeled naïve OT-I CD8⁺ T cells transferring (n = 4, *P < 0.05, **P < 0.01, ***P < 0.001)

To assess antigen presentation capabilities across different treatment groups, flow cytometric analysis of intratumoral dendritic cells revealed an improved uptake efficiency of OVA-specific MHC I-restricted epitopes in the combination therapy group. This enhancement was attributed to the dual effects of ICD induction and cDC1 recruitment (Fig. 5C) (NC group: 357.30 ± 28.85; cisplatin group: 531.30 ± 67.70; DEX_XCL1 group: 1170.00 ± 75.71, cisplatin and DEX_XCL1 group: 1979.00 ± 74.68). Additionally, OVA-specific OT-1 cells labeled with CFSE and transferred into mice demonstrated a remarkable proliferation rate of 70.4% in the combination therapy group (Fig. 5D) (NC group: 12.91 ± 1.73; cisplatin group:32.50 ± 1.94; DEX_XCL1 group: 38.43 ± 0.97, cisplatin and DEX_XCL1 group: 70.43 ± 1.83). While DEX_xcl1 recruits cDC1 cells, cisplatin induces the release of a large number of tumor-associated antigens, facilitating the initiation of specific antitumor immunity and providing a foundation for the activation of antigen-specific immune effector cells.

Tumor microenvironment modulation

Analysis of tumor-infiltrating lymphocytes revealed a significant increase in Ki67 expression in CD8⁺T cells, while no notable change was observed in Ki67 expression in CD4⁺ T cells (Fig. 6A) (Ki67⁺CD8⁺T cell: NC group: 3.32 ± 0.26; cisplatin group:5.24 ± 0.35; DEX_XCL1 group: 8.57 ± 0.87, cisplatin and DEX_XCL1 group: 19.30 ± 2.15; Ki67⁺CD4⁺T cell: NC group: 6.59 ± 0.65; cisplatin group:5.56 ± 1.08; DEX_XCL1 group: 6.66 ± 0.84, cisplatin and DEX_XCL1 group: 7.04 ± 1.29;). This suggests that the combination treatment enhances cDC1 recruitment and antigen-presenting capacity, thereby promoting downstream CD8⁺ T cell proliferation. Further assessment of intratumoral CD8⁺T cell cytotoxicity showed a marked increase in the proportion of granzyme-B-positive CD8⁺ T cells in the combination treatment group (Fig. 6B) (NC group: 9.00 ± 1.22; cisplatin group:9.65 ± 1.49; DEX_XCL1 group: 19.15 ± 2.54, cisplatin and DEX_XCL1 group: 36.65 ± 4.25), along with significantly elevated intratumoral IFN-γ levels (NC group: 9.00 ± 1.22; cisplatin group:9.65 ± 1.49; DEX_XCL1 group:19.15 ± 2.54, cisplatin and DEX_XCL1 group: 36.65 ± 4.25) (Fig. 6D). In contrast, the proportion of regulatory T cells (Foxp3⁺CD4⁺ T cells) (NC group: 30.25 ± 1.39; cisplatin group:23.60 ± 1.18; DEX_XCL1 group:20.50 ± 1.00, cisplatin and DEX_XCL1 group: 11.03 ± 0.53) was significantly reduced, and intratumoral levels of the immunosuppressive factor TGF-β (NC group: 4.25.00 ± 0.47; cisplatin group:2.00 ± 0.19; DEX_XCL1 group:1.84 ± 0.09, cisplatin and DEX_XCL1 group: 0.97 ± 0.10) were also notably decreased in the combination treatment group (Fig. 6C–D).The two therapeutic approaches not only play a pivotal role in initiating targeted antitumor immunity, but also contribute to the improvement of the immunosuppressive tumor microenvironment.

Fig. 6.

Examination of the involvement of T cells and improvement in tumor microenvironment. A Flow cytometry quantification of ki67⁺ proliferative percentages of CD8⁺ and CD4⁺T cells in tumor tissue of prostate cancer mice treated with cisplatin, DEX_XCL1, cisplatin combined with DEX_XCL1 or PBS treated group (Negative control) (n = 4). B Representative flow cytometry quantification of intratumoral Granzyme-B⁺ CD8⁺ T cells from treated mice (n = 4). C Representative flow cytometry quantification of intratumoralFoxp3⁺ CD4⁺ T cells from treated mice (n = 4). D Quantification of cytokine -TGF-β and IFNγ in tumor homogenate from treated mice with ELISA (n = 4, *P < 0.05, **P < 0.01, ***P < 0.001)

Discussion

Prostate cancer’s exact etiology remains unclear, but extensive literature highlights the importance of both genetic and environmental factors [22, 23]. Beyond these, inflammation plays a role in the development of prostate cancer and is crucial in its progression from localized disease to metastatic stages [24]. Regarding adaptive immunity, studies have identified CD4⁺ T cells and CD8⁺ T cells within the prostate. However, CD8⁺ T cells are non-functional, as they fail to upregulate activation markers like CD69 and CD137 upon stimulation with PMA [25]. Furthermore, advanced prostate cancer is characterized by the presence of immunosuppressive mechanisms, including regulatory T cells, myeloid-derived suppressor cells (MDSCs), and the production of TGF-β [26–28]. Therefore, reactivating adaptive immunity and remodeling the immune microenvironment of prostate cancer are pivotal challenges for effective antitumor immunotherapy.

Unlike vaccines for infectious diseases, cancer vaccines primarily focus on therapeutic effects and must elicit robust CD4⁺ T and CD8⁺ T cell responses [29]. Achieving such responses requires two critical factors: (1) the selection of target antigens that are highly and specifically expressed in tumor tissues [30], and (2) the choice of adjuvants to overcome tumor-induced immunosuppression [31].

Extracellular vesicles derived from dendritic cells offer excellent biocompatibility, stability in systemic circulation, and targeted delivery capabilities [32]. Using a pre-modification strategy, we successfully obtained “mature” DEX loaded with XCL1. Our study revealed that extracellular vesicles secreted by mature dendritic cells activated with Poly I: C and CD40L, when loaded with XCL1, significantly enhanced the expression of the activation marker CD86 on dendritic cells compared to extracellular vesicles derived from immature dendritic cells. It is worth noting that current strategies for exosome loading primarily focus on incorporating multiple molecules to achieve enhanced antitumor immunotherapy effects. However, because DEX are nanoscale particles, the number of molecules that can be loaded onto each exosome is limited. As a result, multi-molecule loading often suffers from low co-loading efficiency. In this study, we employed genetic engineering to modify parent cells, enabling the integration of the full-length XCL1 protein onto the surface of exosomal membranes. This genetic approach allows the functional proteins embedded in the exosomal membrane to retain their native folding, thereby producing engineered extracellular vesicles with desired functions and properties. Compared to synthetic nanoparticles or peptide-loaded extracellular vesicles, membrane-anchored proteins demonstrate superior stability and activity. This innovative strategy enhances the potential of extracellular vesicles as a platform for therapeutic applications. Remarkably, the extracellular vesicles produced by this method showed no significant changes in morphology or average particle size. Surprisingly, we found that XCL1-loaded extracellular vesicles were more readily taken up by dendritic cells, likely due to the interaction between XCL1 and the XCR1 receptor on dendritic cells. During the reactivation of adaptive immunity, these extracellular vesicles were involved in recruiting and activating cDC1 cells and facilitating antigen presentation.

Moreover, the immunogenic cell death (ICD) effect induced by cisplatin facilitates the sustained and stable release of tumor-associated antigens in tumor-bearing mice. By recruiting XCR1⁺SIRPα⁻ cDC1 cells via XCL1, these cells can efficiently recognize and present tumor antigens [33]. The synergistic effect of these two therapeutic strategies markedly extended the survival of tumor-bearing mice. This therapeutic strategy targets genetically heterogeneous tumors without relying on antigens encoded by vaccines to stimulate the host’s antitumor immune response, potentially reducing the occurrence of tumor escape. Importantly, this universal approach can be easily adapted to other tumors by substituting alternative treatments that induce immunogenic cell death (ICD), such as chemotherapy or targeted therapies, thereby demonstrating translational potential for highly heterogeneous tumors.

Additionally, in addressing the immunosuppressive microenvironment of prostate cancer, this strategy effectively reversed immune suppression by reducing the proportion of Foxp3⁺CD4⁺ regulatory T cells and the secretion levels of TGF-β within tumor tissues. Both in vivo and in vitro experiments confirmed a significant increase in the proportion of functional cytotoxic T cells and elevated levels of secreted cytokines such as IL-12 and IFN γ. As a result, the survival time of tumor-bearing mice was significantly prolonged, and tumor growth rates were markedly slowed. However, it is regrettable that no cases of complete tumor regression were observed. Therefore, further optimization of this synergistic therapeutic strategy is required in future studies.

In summary, we developed an exosome vaccine loaded with XCL1 and proposed a novel therapeutic strategy for advanced prostate cancer by combining it with cisplatin. This approach targets the genetic heterogeneity of individual patients and exerts tumor-specific cytotoxic effects through adaptive immune mechanisms. This universal therapeutic strategy not only addresses the high heterogeneity of tumors, but also overcomes the challenges and high costs associated with the production of personalized vaccines. However, this study still has several limitations. First, this treatment approach has not achieved complete tumor eradication, which may be related to the exhaustion of cytotoxic T cells in the tumor microenvironment. Therefore, we will further evaluate the tumor microenvironment using single-cell sequencing and flow cytometry to identify potential therapeutic targets for optimizing treatment strategies. Second, the targeted delivery of exosomes to the lesion remains an urgent challenge. Thus, screening peptides or nanobodies with specificity for prostate cancer tissue is crucial. Additionally, translating preclinical animal studies into clinical trials still presents numerous challenges, such as (1) achieving large-scale standardized production of human-derived dendritic cell exosomes, (2) ensuring quality control of individual exosomes, and (3) assessing the biosafety of modified exosomes. Furthermore, this treatment paradigm can be extended to tumors beyond prostate cancer, offering the potential to improve patient survival while significantly reducing the financial burden on both individuals and society.

Author contribution

Zhongqiang Fan was involved in conceptualization, methodology (experiments), software, formal analysis, and writing—original draft; Zhao Wang was involved in data curation, methodology (experiments), and writing—original draft; Hui Zhang was involved in visualization, methodology (experiments), and investigation; Hao Zhang was involved in resources and supervision; Rui Zhao was involved in software and validation; Shimiao Zhu (corresponding author) was involved in visualization, funding acquisition, and writing—review and editing; Xiaoqiang Liu (corresponding author) was involved in conceptualization, resources, supervision, and writing—review and editing.

Funding

This study was supported by National Natural Science Foundation of China (82172759).

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Conflict of interests

The authors declare no competing interests.

Ethical approval

Animal studies were approved by the Committee on Animal Research and Ethics of Tianjin Medical University General hospital, and all protocols conformed to the Guidelines for Ethical Conduct in the Care and Use of Nonhuman Animals in Research.