Chimeric Exosomes Functionalized with STING Activation for Personalized Glioblastoma Immunotherapy

Abstract

A critical challenge of existing cancer vaccines is to orchestrate the demands of antigen-enriched furnishment and optimal antigen-presentation functionality within antigen-presenting cells (APCs). Here, a complementary immunotherapeutic strategy is developed using dendritic cell (DC)-tumor hybrid cell-derived chimeric exosomes loaded with stimulator of interferon genes (STING) agonists (DT-Exo-STING) for maximized tumor-specific T-cell immunity. These chimeric carriers are furnished with broad-spectrum antigen complexes to elicit a robust T-cell-mediated inflammatory program through direct self-presentation and indirect DC-to-T immunostimulatory pathway. This chimeric exosome-assisted delivery strategy possesses the merits versus off-the-shelf cyclic dinucleotide (CDN) delivery techniques in both the brilliant tissue-homing capacity, even across the intractable blood-brain barrier (BBB), and the desired cytosolic entry for enhanced STING-activating signaling. The improved antigen-presentation performance with this nanovaccine-driven STING activation further enhances tumor-specific T-cell immunoresponse. Thus, DT-Exo-STING reverses immunosuppressive glioblastoma microenvironments to pro-inflammatory, tumoricidal states, leading to an almost obliteration of intracranial primary lesions. Significantly, an upscaling option that harnesses autologous tumor tissues for personalized DT-Exo-STING vaccines increases sensitivity to immune checkpoint blockade (ICB) therapy and exerts systemic immune memory against post-operative glioma recrudesce. These findings represent an emerging method for glioblastoma immunotherapy, warranting further exploratory development in the clinical realm.

Keywords: STING activation; T cell responses; chimeric exosomes; glioblastoma; personalized immunotherapy.

Figure 1

Schematic illustration of DT‐Exo‐STING nanovaccine for GBM immunotherapy. a) Design and fabrication of DT‐Exo‐STING nanovaccine. b) Schematic exhibiting action mode of this chimeric exosome‐assisted delivery strategy in STING activation. c) Mechanism of DT‐Exo‐STING–mediated antineoplastic immunoresponse in an intracranial GBM mouse model. P, phosphorylation; TBK1, TANK‐binding kinase 1; IRF3, interferon regulatory factor 3.

Figure 2

Preparation and characterization of DT‐Exo‐STING. a) Flow cytometry histograms of CD63 on T‐Exos, D‐Exos and DT‐Exos. b) TEM images and size distributions of T‐Exos, D‐Exos and DT‐Exos. c) Flow cytometry plots of anti‐CD44–labeled T‐Exos, anti‐CD11c–marked D‐Exos and the double‐antibody–labeled chimeic DT‐Exos. d) SDS‐PAGE protein analysis of T‐Exos, D‐Exos, and DT‐Exos. e) Imaging flow cytometry to determine the encapsulation of cdGMP‐Dy547 within DT‐Exos. cdGMP (green) and CD63 (red) in images. f) CLSM images of cdGMP‐Dy547 distribution within DCs after 4 h incubation with individual cdGMP‐Dy547 and cdGMP‐Dy547–loaded DT‐Exos. Nucleus (blue), lysosome (green), and cdGMP‐Dy547 (red) in confocal images.

Figure 3

Dual activation of T cells by chimeric exosome‐based nanovaccine. a) Sketch depicting the dual T‐cell activation assays in (b to i). b) Flow cytometric quantification of splenic CD3⁺CD8⁺ T‐cell proliferation with the staining of CFSE after 72 h incubation directly with PBS, T‐Exos, D‐Exos, and DT‐Exos (n = 6; one‐way analysis of variance (ANOVA) with Tukey's multiple comparisons test). c) The proportions of CD3⁺CD8⁺ T cells 48 h period after treatment with the assigned formulations (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). d) In vitro cytotoxicity of exosome‐activated T cells to GL261 cells 24 h after incubation with the designated formulations (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). e) The quantitative evaluation of DC maturation (CD11c⁺CD80⁺CD86⁺) analyzed by means of flow cytometry post‐incubation with PBS, T‐Exos, D‐Exos, and DT‐Exos for 48 h (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). f) Quantitative analysis of DC‐mediated cross‐presentation (CD11c⁺SIINFEKL‐H‐2Kb⁺) by flow cytometry, following 48 h incubation with the designated exosome nanovaccines (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). g) The percentages of proliferated CD3⁺CD8⁺ T cells after 72 h co‐culture of CFSE‐stained splenic cells with the above‐pretreated DCs in a ratio of 20:1 (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). h) IFN‐γ secretion in the supernatant of DC‐to‐T co‐culture system 48 h period post‐incubation of various exosome‐stimulated DCs and splenocytes with a ratio of 1:20, measured by ELISA kit (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). i) In vitro cytotoxicity mediated by the designated exosome‐DC–treated splenocytes to GL261 cells 24 h post‐incubation with an effector/target ratio of 10:1 (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). Data in (b to i) are represented as means ± SD. ***p < 0.001, ****p < 0.0001.

Figure 4

Targeted delivery of DT‐Exo‐STING in vivo. a) Representative in vivo fluorescence imaging and b) quantification in brain regions of GL261‐bearing mice at the indicated time points after subcutaneous administration with cdGMP‐Dy547–containing formulations (n = 3; two‐way ANOVA with Tukey's multiple comparisons test). c) Ex vivo fluorescence imaging of the major organs (heart, liver, spleen, lung, kidney, and brain) 12 h after subcutaneous administration. Inserted images exhibited the brain penetration of cdGMP‐Dy547–containing formulations. cdGMP‐Dy547 (red), nucleus (blue), and CD31 (green) in images. d) 3D fluorescence imaging to visualize the distribution of cdGMP‐Dy547–loaded DT‐Exos within the optically transparent brain tissues performed with CLARITY technique. cdGMP‐Dy547 (red) in images. e) Ex vivo fluorescent images of brain tissues and f) cervical LNs of GL261 tumor‐bearing mice following subcutaneous injection with various vesicles containing cdGMP‐Dy547. cdGMP‐Dy547 (red) and nucleus (blue) in images. Data in (b) are represented as means ± SD. ****p < 0.0001.

Figure 5

DT‐Exo‐STING–induced antineoplastic efficacy in an orthotopic GBM mouse model. a) Treatment scheme for mice with intracranial GL261‐OVA‐Luc tumors. Mice were subject to subcutaneous administration with cGAMP‐containing formulations a total of four times, 4 days apart. b) Flow cytometric quantification of mature DCs (CD11c⁺CD80⁺CD86⁺) in cervical LNs of mice treated with the designated vaccine formulations (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). c) The proportions of CD11c⁺SIINFEKL‐H‐2Kb⁺ DCs in cervical LN tissues analyzed by means of flow cytometry (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). d) Ratio of tumor‐infiltrating CD8⁺ to CD4⁺ T cells (gated from CD3⁺ cells; n = 6; one‐way ANOVA with Tukey's multiple comparisons test). e) Quantitative analysis of H‐2Kb/SIINFEKL tetramer staining of CD3⁺CD8⁺ T cells within tumors, using flow cytometry (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). f) The percentages of IFN‐γ⁺ T cells in tumor tissues (gated from CD3⁺CD8⁺ cells; n = 6; one‐way ANOVA with Tukey's multiple comparisons test). g) The quantification of Ki67 expression in tumor‐infiltrating CD3⁺CD8⁺ T cells (n = 6; one‐way ANOVA with Tukey's multiple comparisons test). h) Representative in vivo bioluminescence images and i) quantified signal intensity of glioma‐bearing mice after the indicated treatments (n = 3; two‐way ANOVA with Tukey's multiple comparisons test). j) Kaplan–Meier survival curves of mice vaccinated with the designated cancer vaccine formulations (n = 6; log‐rank Mantel–Cox test). Data in (b to g, and i) are represented as means ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. s.c., subcutaneous.

Figure 6

Combinatorial efficacy of personalized DT‐Exo‐STING and ICB therapy against postsurgical GBM recurrence. a) Postoperative treatment scheme for GL261‐Luc–burdened mice. Mice were treated subcutaneously with personalized DT‐Exo‐STING and intraperitoneally with anti‐PD‐1 antibodies (ICB) a total of four times, 4 days apart, subsequent to intracranial GBM surgical resection. b) In vivo fluorescence imaging and c) quantification of GL261‐Luc–bearing mice treated with the indicated formulations (n = 3; two‐way ANOVA with Tukey's multiple comparisons test). d) Kaplan–Meier survival analysis of mice after various treatments (n = 6; log‐rank Mantel–Cox test). e) Representative flow dot plots of T_EM (CD44⁺CD62L⁻; gated from CD8⁺ cells) and T_CM (CD44⁺CD62L⁺; gated from CD8⁺ cells) in the peripheral blood on day 19 post‐inoculation of GBM tumors. f) Schematic illustration of the experimental design combining chimeric exosome‐based nanovaccine and ICB therapy against post‐operative GBM recrudescence. Mice were treated subcutaneously with personalized DT‐Exo‐STING and intraperitoneally with ICB therapy a total of four times, 4 days apart, and received intraperitoneal injections with anti‐CD8a or isotype monoclonal antibody IgG total four times, 5 days apart. g) Representative in vivo bioluminescence images and h) quantification of bioluminescence signal strength in GL261‐Luc–burdened mice pre‐treated with anti‐CD8a or mouse monoclonal IgG, prior to the combination treatment of chimeric exosome‐based nanovaccine and ICB therapy (n = 3; two‐way ANOVA with Tukey's multiple comparisons test). i) Kaplan–Meier survival curves of mice immunized with the combined therapy and other assigned formulations (n = 6; log‐rank Mantel–Cox test). Data in (c and h) are represented as means ± SD. *p < 0.05, ***p < 0.001, ****p < 0.0001. i.p., intraperitoneal.