CircPRKD3-loaded exosomes concomitantly elicit tumor growth inhibition and glioblastoma microenvironment remodeling via inhibiting STAT3 signaling

Abstract

Background: Glioblastoma stem cells and their exosomes (exos) are involved in shaping the immune microenvironment, which is important for tumor invasion and recurrence. However, studies involving GSC-derived exosomal circular RNAs (GDE-circRNAs) in regulating tumor microenvironment (TME) remain unknown. Here, we comprehensively evaluated the significance of a novel immune-related GDE-circRNA in the glioma microenvironment.

Methods: GDE-circPRKD3 was screened out through high-throughput sequencing and verified by RT-PCR, sanger sequencing, and RNase R assays. A series of in vitro and in vivo experiments were performed to investigate the function of GDE-circPRKD3. RNA-seq, RNA immunoprecipitation, multicolor flow cytometry, and western blotting were used to explore the regulation of GDE-circPRKD3 on STAT3 signaling-mediated TME remodeling.

Results: We have characterized a circRNA PRKD3 in GSC exosomes, and lower circPRKD3 expression predicts a worse prognosis for glioblastoma patients. Overexpression of GDE-circPRKD3 significantly impairs the biological competence of glioma and prolongs the survival of xenograft mice. GDE-circPRKD3 binds to HNRNPC in an m6A-dependent manner, accelerates mRNA decay of IL6ST, and inhibits downstream target STAT3. Notably, GDE-circPRKD3 promotes CXCL10 secretion by reprogramming tumor-associated macrophages, which in turn recruits CD8+ tumor-infiltrating lymphocytes against GBM. Moreover, brain-targeted lipid nanoparticle delivery of circPRKD3 combined with immune checkpoint blockade therapy achieves significant combinatorial benefits.

Conclusions: This study provides a novel mechanism by which GDE-circPRKD3 relies on STAT3 signaling to remodel immunosuppressive TME and offers a potential RNA immunotherapy strategy for GBM treatment.

Keywords: STAT3; circPRKD3; exosome; glioma stem cell; tumor microenvironment.

Graphical Abstract

Figure 1.

Low-expressed GSC-derived exosomal circPRKD3 maybe a promising biomarker and relate to the regulation of GSC on TME. (A) Detection of GSC surface marker CD133 expression using flow cytometry. U87-GSLC is a glioma stem-like cell derived from glioma cell line U87. GSC-3, 4, 5, and 13 are primary glioma stem cells. G-3, 4, 5, and 13 are primary non-stem glioma cells. (B) Metaplot showing differentially expressed circRNAs between GSC-derived exosomes and normal exosomes identified from RNA-seq. BSJ, back-splicing junction. (C) Illustration of the annotated genomic region of human circPRKD3, different splicing forms of PRKD3 pre-mRNA (upper). Divergent (red) primers were designed to amplify the back-spliced products (Middle). Sanger sequencing of PCR products showing the back-splicing junction sites of circPRKD3 (bottom). (D) RT-qPCR analysis for the expression of circPRKD3 after treatment with RNase R in GSC-5. CDR1as was used as positive control; GAPDH was used as negative control. (E) CircPRKD3 RNA levels in GSC-derived exosomes and normal exosomes by RT-qPCR. (F) Relative expression of circPRKD3 in 34 cases of NBTs and a cohort of 175 independent glioma specimens. NBTs, normal brain tissues. (G) RNA fluorescence in situ hybridization for circPRKD3. Nuclei were stained with DAPI. Scale bar, 20 µm. (H) Survival analysis of the 64 cases of GBM patients stratified by circPRKD3 expression. (I) Left, representative immunofluorescence (IF) images showing the distribution of CD86⁺ macrophages between low and high circPRKD3 expression of GBM samples. Scale bar, 100 μm. Right, quantification of CD86⁺ cells proportion in glioma samples with different levels of circPRKD3 (30 samples of Low circPRKD3 and 43 samples of High circPRKD3). (J) Relative expression of plasma exosomal circPRKD3 in 70 samples of healthy controls (N-Plasma-Exo) and 81 cases of GBM patients (G-Plasma-Exo). For survival analysis, Mantel-Cox log rank test was performed to calculate the P-value. Unpaired Student’s t-test was performed to analyze the significance of the differences between the indicated groups where applicable. Data are shown as the mean ± SD. **P < .01; ***P < .001; ****P < .0001; ns, not significant.

Figure 2.

GDE-CircPRKD3 impedes GBM progression. (A) Relative expression of circPRKD3 in GSCs and GSC-derived exosomes with stable overexpression by lentivirus (Lv). (B) Relative expression of circPRKD3 in U87 and G-17 cells after incubating with GDE-OE-circPRKD3 or control exosomes. (C) Fluorescence images showing the proliferation efficacy of U87 and G-17 cells treated with GDE-OE-circPRKD3 or control exosomes. Representative of 3 independent experiments. (D) Quantification of the EdU assay of U87 and G-17 cells treated with GDE-OE-circPRKD3 or control. (E) Invaded assays of U87 and G-17 cells after incubating with GDE-OE-circPRKD3 or control exosomes. (F) Schematic of the therapeutic strategy in the mouse model of glioma. IVIS, In Vivo Imaging Systems. (G) In vivo bioluminescent imaging analysis of intracranial tumor growth in mice bearing GBM xenografts derived from the G-17 cells treated with GDE-OE-circPRKD3 (n = 4, 200 μg/dose), control exosomes (n = 4, 200 μg/dose) or PBS (n = 3, equal volume). Representative bioluminescent images on days 7, 14, and 21 post-transplantation of G-17 cells were shown. (H) Normalized bioluminescence in tumor-bearing mice treated as indicated times. (I) Survival analysis of tumor-bearing mice subjected to the indicated treatments. (J) Upper, representative of whole-slide images of H&E-stained tissue sections from the above group of mice. Scale bar, 1 mm. Bottom, representative H&E-stained images (40×) of tumor-bearing mice. Scale bar, 100 μm. (K) Left upper, representative of whole-slide images of IHC-stained tissue sections from the above group of mice using phospho-STAT3 antibody. Scale bar, 1 mm. Left bottom, representative IHC-stained images (40×) of tumor-bearing mice using a phospho-STAT3 antibody. Scale bar, 100 μm. Right, IHC scoring of the phospho-STAT3 in randomly selected microscopy fields of each IHC image. For survival analyses, the Mantel-Cox log rank test was performed. Other analyses by Mann–Whitney or Student’s t-test. Data are shown as the mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001.

Figure 3.

GDE-circPRKD3/HNRNPC jointly to regulate IL6ST mRNA stability and attenuate GBM progression. (A) Venn plots showing RNA-seq analysis of each group of glioma cells after co-incubation with GDE-OE-circPRKD3 or GDE-sh-circPRKD3, respectively. (B) RT-qPCR analysis for the expression of circPRKD3 and its target genes after treating with GDE-OE-circPRKD3 or control exosomes in G-17 cells. (C) Relative expression of IL6ST in G-17 cells transfected with IL6ST overexpression plasmid alone or co-cultured with GDE-OE-circPRKD3. (D) Protein levels of IL6ST, JAK2, phospho-JAK2, STAT3, and phospho-STAT3 in G-17 cells transfected with IL6ST overexpression plasmid alone or co-incubation of GDE-OE-circPRKD3. (E) Silver staining showing circPRKD3 binding proteins after RNA pulldown assays. Red frame indicating differential bands with high possibility. (F) RIP assays show the association of RBP candidates with circPRKD3. Left, immunoprecipitation efficiency of antibodies shown in western blotting. Right, the fold enrichment of immunoprecipitants was analyzed by RT-qPCR with specific primers for circPRKD3 and GAPDH (negative control). (G) schematic structures of HNRNPC truncations. Upper, full-length of HNRNPC; middle, a HNRNPC truncation lacking 26-77aa; bottom, a HNRNPC truncation lacking 176-232aa. RRM, RNA recognition motif; Del, deletion. (H) RIP assays show the binding of HNRNPC truncations with circPRKD3. Left, immunoblot analysis with anti-FLAG of U87 cells transfected with plasmids encoding FLAG-tagged full-length or truncated HNRNPCs. Right, the fold enrichment of HNRNPC truncations binding with circPRKD3 was analyzed by RT-qPCR with specific primers. GAPDH was used as the negative control. (I) Relative expression of IL6ST mRNA after treating with Actinomycin D at the indicated time points in LN229 cells co-transfection of HNRNPC siRNA. (J) Relative expression of IL6ST mRNA after treating with Actinomycin D at the indicated time points in G-17 cells incubation of GDE-OE-circPRKD3 or control exosomes. P-value was determined by a two-way ANOVA. (K) Fold enrichment of IL6ST in RIP assay against HNRNPC in LN229 and G-17 cells incubated with GDE-OE-circPRKD3 or control exosomes. (L) Fold enrichment of IL6ST in RIP assay against HNRNPC in LN229 and G-17 cells transfected with circPRKD3 m⁶A-WT or circPRKD3 m⁶A-Mut overexpression plasmids. GAPDH (negative control). Student’s t-test was used to determine the significance of the differences between the indicated groups where applicable. The data are presented as the mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001; ns, not significant.

Figure 4.

GDE-circPRKD3 promotes CD8 T cell infiltration via CXCL10. (A) Upper, schematic of the therapeutic strategy in the mouse model of GBM. Mice were retro-orbital injected with mGDE-OE-circPrkd3 (n = 6, 200 μg/dose), control exosomes (n = 6, 200 μg/dose), or PBS (n = 5, equal volume) every 3 days. Bottom, in vivo bioluminescent imaging analysis of intracranial tumor growth in mice bearing GBM xenografts derived from the GL261 cells. Representative bioluminescent images on days 7, 14, 21, and 28 post-transplantation of GL261 cells were shown. (B) Statistical analysis of bioluminescent tracking plots in mice with GBM treated as indicated times. (C) Survival analysis of tumor-bearing mice subjected to the indicated treatments. (D) A t-SNE plot from flow cytometry data showing cell populations of CD45⁺ tumor-infiltrating lymphocytes (TIL) sorted from GL261-bearing mice treated with mGDE-OE-circPrkd3, control exosomes, and PBS. (E) Quantification of NK cells (CD45⁺CD3^-NK1.1⁺), CD8 T cells (CD45⁺CD3⁺CD8⁺), B cells (CD45⁺CD3^-CD19⁺), CD4 T cells (CD45⁺CD3⁺CD4⁺), microglia (CD45^lowCD11b⁺F4/80⁺CD49d^-), macrophage (CD45^highCD11b⁺F4/80⁺CD49d⁺), Monocytes (CD45⁺CD11b⁺F4/80^-Gr-1^-), Mono-MDSCs (CD45⁺CD11b⁺Gr-1⁺Ly6G^-), and Neutrophils (CD45⁺CD11b⁺Ly6G⁺) frequency in proportion of CD45⁺ TIL. (F) Representative images of CD8 and DAPI nuclear staining. Scale bar, 50 μm. Red arrow indicating CD8⁺ cells. (G) Flow cytometry analysis of Granzyme B (GzB⁺) intracellular staining in fixed and permeabilized CD8 T cells. (H) ELISA assays showing the concentration of CXCL10 in the serum of tumor-bearing mice. (I) Representative image of CXCL10, F4/80, and DAPI staining of GL261 tumors treated with mGDE-OE-circPrkd3. Scale bar, 100 μm. (J) Relative expression of CXCL10 in MDMs and MG sorted from GL261-bearing mice treated with mGDE-OE-circPrkd3 and control exosomes by RT-qPCR. MDMs, macrophages; MG, microglia. (K) Left, statistical analysis of bioluminescent intensity of mice with GBM subjected to the indicated treatments. Treatment cohorts: αCXCR3 (n = 5), mGDE-OE-circPrkd3 (n = 6), mGDE-OE-circPrkd3 + αCXCR3 (n = 5). Right, assessing the contributions of CXCR3 function to the overall survival benefit of tumor-bearing mice subjected to the indicated treatments. (L) Left, flow cytometry analysis of CD8 T cells in GL261 tumors treated with mGDE-OE-circPrkd3 alone or coordinated with αCXCR3. Right, flow cytometry analysis of the affection of CXCR3 on Granzyme B (GzB⁺) secretion by CD8 T cells from GBM mice subjected to the indicated treatments. For survival analyses, Mantel-Cox log rank test was performed. Other analyses by Mann–Whitney test or one-way ANOVA or Student’s t test. The data are presented as the mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001; ns, not significant.

Figure 5.

GDE-circPRKD3 induced a M1-like program in TAMs. (A) Upper, schematic of the treatment strategy in the model of murine GBM. Mice were retro-orbital injected with or without mGDE-OE-circPrkd3 (200 μg/dose) every 3 days. PLX3397 was administered once daily by oral gavage at a dosage of 100 mg/kg for 2 weeks. Bottom, bioluminescent intensity comparison in mGDE-OE-circPrkd3-treated GL261 tumor-bearing mice in the absence or presence of PLX3397. (B) Assessing the contributions of depletion of macrophages to survival benefit of mice with GBM treated with mGDE-OE-circPrkd3. Ctrl (n = 6), PLX3397 (n = 5), mGDE-OE-circPrkd3 (n = 6), mGDE-OE-circPrkd3 + PLX3397 (n = 5). Mantel-Cox log-rank test was performed. (C) Flow cytometry analysis of CD8 T cells in GL261 tumors subjected to the indicated treatments. (D) Upper, representative images of the proportion of CD86 in TAMs from GL261 tumor-bearing mice treated with mGDE-OE-circPrkd3 or control exosomes by flow cytometry. Macrophages were gated as: CD45⁺CD11b⁺F4/80⁺. Bottom, quantification of CD86 in macrophages from above groups by flow cytometry. (E) Left, representative images of CFSE-labeled CD8 T-cell proliferation were assessed by flow cytometry after incubating with tumoral MDMs from different groups. Right, in vitro co-culture of tumoral CD49d⁺ MDMs and activated splenic CFSE-labeled CD8 T cells. T cell Alone (n = 4), Ctrl co-culture (n = 4), control exosomes (n = 4), mGDE-OE-circPrkd3 (n = 6). (F) Analysis of the CD86 rate in M2-like BMDMs treated with mGDE-OE-circPrkd3 assessed by flow cytometry. (G) Relative expression of M1-like markers (Il-1a, Tnf-α, Il-1β, Il-6, Ptgs2) assessed by RT-qPCR analysis on mGDE-OE-circPrkd3-treated M2-like BMDMs. (H) Protein levels of iNOS, IL-10, and ARG1 in M2-like BMDMs incubated with mGDE-OE-circPrkd3 or control exosomes. Representative of three independent experiments. (I) In vitro co-culture of M2-like BMDMs and activated splenic CFSE-labeled CD8 T cells. T cell Alone (n = 4), Ctrl co-culture (n = 4), control exosomes (n = 4), mGDE-OE-circPrkd3 (n = 6). (J) Analysis of M1-like (cd86, Il-1β, Il-1α) and M2-like (Chil3, Mrc1, Pparg) programs as assessed by RT-qPCR analysis of GL261 cell-induced M2-like BMDMs incubated with mGDE-OE-circPrkd3 or control exosomes. (K) Immunoblotting of CD86, IL-10, and ARG1 in GL261 cell-induced M2-like BMDMs treated with mGDE-OE-circPrkd3 or control exosomes. (L) Left, protein levels of STAT1, phospho-STAT1, P65, phospho-P65, STAT3, and phospho-STAT3 in M2-like BMDMs after treating with mGDE-OE-circPrkd3 or control exosomes. Right, immunoblotting of STAT1, phospho-STAT1, P65, phospho-P65, STAT3 and phospho-STAT3 in GL261 cell-induced M2-like BMDMs treated with mGDE-OE-circPrkd3 or control exosomes. Student’s t-test was used to analyze the significance of the differences between the indicated groups where applicable. The data are presented as the mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001; ns, not significant.

Figure 6.

Delivery of LNP-circPRKD3 combined with PD-1 blockade therapy prolongs survival benefit in GBM mice. (A) Representative images of lipid nanoparticles (LNPs) taken by cryo-electron microscopy. Ang-2, angiopep-2. Scale bar, 100 nm. (B) Relative expression of circPrkd3 in GL261 transfected with Cy5-Ang-2-LNP-circPrkd3 or control LNP. (C) Schematic of the therapeutic strategy in the model of murine GBM. Mice were nasal drip with Ang-2-LNP-circPrkd3 or control LNP every 3 days. Anti-PD-1 was administered every 3 days by intraperitoneal injection for 2 weeks. (D) In vivo bioluminescent imaging analysis of intracranial tumor growth in GL261 tumor-bearing mice. Representative bioluminescent images on days 7, 14, 21, 28, and 35 post-transplantation of GL261 cells were shown. (E) Statistical analysis of bioluminescent tracking plots in mice with GBM subjected to the indicated treatments. (F) Survival analysis of GL261 tumor-bearing mice subjected to the indicated treatments. Anti-IgG (n = 6), anti-PD-1 (n = 5), Ang-2-LNP-NC (n = 6), Ang-2-LNP-circPrkd3 (n = 6) and Ang-2-LNP-circPrkd3 + anti-PD-1 (n = 8). (G) Left, bioluminescence analysis of tumor location in mice with GBM. Right, distribution of Cy5-Ang-2-LNP in mouse heart, liver, spleen, lung, kidney, and brain by fluorescence analysis. (H-I) Abundance of CD8 T cells, NK cells, CD86⁺ macrophages, and CD206⁺ macrophages in GL261 tumors treated as indicated assessed by flow cytometry. (J) Representative images of CD8and DAPI nuclear staining in GBM tumors were obtained from different treatment groups. Scale bar, 50 μm. (K) Quantification of CD8 T cells in GL261 tumors subjected to the indicated treatments. (L) ELISA assays showing the concentration of CXCL10, TNFα, and IFNγ in the serum of tumor-bearing mice. Analyses by Mann–Whitney test or Student’s t-test. ns, no significance. Data in all quantitative panels are presented as mean ± SD. *P < .05; **P < .01; ***P < .001; ****P < .0001; ns, not significant.