Exosome-derived circTRPS1 promotes malignant phenotype and CD8+ T cell exhaustion in bladder cancer microenvironments

Abstract

Circular RNAs (circRNAs) play critical roles in different diseases. Exosomes are important intermediates of intercellular communication. While both have been widely reported in cancers, exosome-derived circRNAs are rarely studied. In this work, we identified the differently expressed circRNAs in bladder cancer (BCa) tissue and exosomes through high-throughput sequencing. RNA pull-down, RNA immunoprecipitation, and luciferase reporter assays were used to investigate the interactions between specific circRNAs, microRNAs (miRNAs), and mRNAs. Wound-healing, Transwell, Cell Counting Kit-8 (CCK8), and colony-formation assays were used to study the biological roles in vitro. Metabolomics were used to explore the mechanism of how specific circRNAs influenced BCa cell behavior. Flow cytometry was used to study how specific circRNAs affected the function of CD8+ T cells in tumor microenvironments. We identified that exosome-derived hsa_circ_0085361 (circTRPS1) was correlated with aggressive phenotypes of BCa cells via sponging miR-141-3p. Metabolomics and RNA sequencing (RNA-seq) identified GLS1-mediated glutamine metabolism was involved in circTRPS1-mediated alterations. Exosomes derived from circTRPS1 knocked down BCa cells, prevented CD8+ T cells from exhaustion, and repressed the malignant phenotype of BCa cells. In conclusion, exosome-derived circTRPS1 from BCa cells can modulate the intracellular reactive oxygen species (ROS) balance and CD8+ T cell exhaustion via the circTRPS1/miR141-3p/GLS1 axis. Our work may provide a potential biomarker and therapeutic target for BCa.

Keywords: CD8+ T cell; ROS; bladder cancer; circular RNAs; exosomes; tumor microenvironment.

Graphical abstract

Figure 1

Identification of circTRPS1 as a prognosis-related circRNA upregulated in BCa tumors and exosomes (A) circRNA expression in 4 pairs of BCa tumor/normal tissues. (B) Differentially expressed circRNAs are shown in volcano plot. (C) Genomic loci of TRPS1 and circTRPS1. The schema shows the structure of circTRPS1. (D) Relative expression levels of circTRPS1 and linear mRNA-TRPS1 in T24 cells after treatment by actinomycin D. (E) Expression of circular and linear mRNA-TRPS1 in BCa cell lines treated with RNase R. (F) Fluorescence in situ hybridization (FISH) in 90 pairs of a BCa tissue microarray (TMA) indicates that circTRPS1 is located in the cytoplasm and upregulated in BCa. (G) circTRPS1 expression level is negatively correlated with overall survival, as is shown in Kaplan-Meier plot. (H) circTRPS1 in BCa tumor tissue is also observed on cytomembrane and extracellular space, detected by FISH. (I) Exosomes isolated from BCa patients' urine, detected by electron microscopy and NTA analysis. Bar corresponds to 100 nm. (J) Western blot analysis of positive and negative exosomal markers in isolated exosomes and urine of BCa. (K) circTRPS1 expression levels among different BCa clinical stages in urine-derived exosomes. (L) Relative circTRPS1 expression in urine-derived exosomes before and after bladder cystectomy or transurethral resection of a bladder tumor (TURBT).

Figure 2

Exosome-derived circTRPS1 promoted BCa cell proliferation and invasion in vitro (A) Expressions of circTRPS1 in human immortalized uroepithelial cell line SV-HUC-1 and BCa cell lines RT4, UM-UC-3, T24, 5637, and J82, detected by qRT-PCR. (B and C) CCK-8 assays of si-NC and si-circTRPS1 T24 and UM-UC-3 cell lines. Data are presented as the mean ± SD. ∗∗∗p < 0.001 versus NC. (D) EdU assays of si-NC and si-circTRPS1 T24 cell lines. Nuclei were stained with DAPI. Data are presented as the mean ± SD. ∗∗∗p < 0.001 versus si-NC. (E) Apoptosis assays of si-NC and si-circTRPS1 T24 and UM-UC-3 cell lines, detected by flow cytometry. (F) Expressions of exosome-derived circTRPS1 from SV-HUC-1 and BCa cell lines RT4, UM-UC-3, T24, 5637, and J82, detected by qRT-PCR. (G) NTA analysis and electron microscopy verification of exosomes. Bar corresponds to 100 nm. (H) Western blot analysis of positive and negative exosomal markers in T24- and UM-UC-3-derived exosomes and cells. (I) PKH67 staining after incubating the exosomes with BCa cells. Bar corresponds to 100 μm. (J and K) Colony-formation and Transwell assays of si-circTRPS1 transfected T24- and UMUC-3-derived exosome-treated BCa cells.

Figure 3

circTRPS1 served as a sponge of miR-141-3p (A) RIP was performed in T24 cells against IgG or Ago2, and the precipitated circTRPS1 was detected by qRT-PCR. (B) Upregulated miRNAs from miRNA-seq were selected and intersected with miRanda and Encori predictions. Kaplan-Meier plots show potential miRNAs related to circTRPS1 and prognosis. (C) AGO2 RIP and precipitated miRNAs detected by qRT-PCR. Data are presented as the mean ± SD. ∗∗∗p < 0.001, ∗p < 0.05 versus anti-IgG. (D) Relative miRNA expression in si-circTRPS1 BCa cells. Data are presented as the mean ± SD. ∗∗∗p < 0.001, ∗p < 0.05 versus si-NC. (E) RNA pull-down assay confirms that circTRPS1 interacts with miR-141-3p more than with miR-200a-3p. Data are presented as the mean ± SD. ∗∗∗p < 0.001, ∗p < 0.05 versus oligo probes. (F) Dual-luciferase reporter assays show the binding properties of circTRPS1 and miR-141-3p. The mutated (Mut) version of circTRPS1 is also shown. (G) Relative luciferase activity was determined after transfection with miR-141-3p mimic/normal control or with the circTRPS1 WT/Mut in HEK293T cells. Data are presented as the mean ± SD. ∗∗∗p < 0.001 versus miR-NC. (H) FISH shows the subcellular location of miR-141-3p and circTRPS1. Nuclei were stained with DAPI. Bar corresponds to 20 μm. (I) Relative expression of miR-141-3p with circTRPS1 in BCa tissue. (J and K) CCK8 and Transwell assays in T24 cells after miR-141-3p mimic/inhibitor transfection with/without si-circTRPS1.

Figure 4

circTRPS1 altered glutamine metabolism via modulating GLS1 expression and caused deregulated redox balance in BCa (A and B) Metabolomics were performed in sh-circTRPS1 T24 cells. Glutathione metabolism, gluoconeogenesis, glutamate metabolism, and fatty acid oxidation were the mainly enriched metabolism pathways identified. (C) RNA-seq of T24 and T24 sh-circTRPS1 cells. (D) The schema of GLS1 in glutamine metabolism. (E) Correlation between expressions of miR-141-3p and GLS1 mRNA in BCa tissue. (F) Dual-luciferase reporter assays show the binding properties of GLS1 mRNA and miR-141-3p. Relative luciferase (luc) activity was determined after transfection with miR-141-3p mimic/normal control or with the GLS1 WT/Mut in HEK293T cells. Data are presented as the mean ± SD. ∗∗∗p < 0.001 versus miR-NC. (G) GSEA of TCGA-BLCA indicates energy starvation caused by differential expression of GLS1 mainly influences mTOR pathway. (H) ROS level in si-circTRPS1 T24 cells, measured by flow cytometry. (I) Western blot analysis of GLS1 level and mTOR and EMT markers.

Figure 5

Exosome-derived circTRPS1 induced T cell exhaustion in BCa tumor microenvironments (A) Representative images of FISH and immunohistochemical staining on circTRPS1 and CD8 in the TMA cohort of 90 patients. (B and C) Kaplan-Meier curves for overall survival with CD8+ T cells and/or circTRPS1 strata in the TMA cohort. (D–F) Comparison of the expression of effector molecules IFN-γ (D), granzyme B (E), and perforin (F) on CD8+ T cells after being cultured with exo-si-circTRPS1 or/and BPTES. (G–I) Comparison of the expression of immune checkpoint PD-1 (G), Tim-3 (H), and double-positive (I) on CD8+ T cells after being cultured with exo-si-circTRPS1 or/and BPTES. Data are presented as the mean ± SD. ∗∗∗p < 0.001, ∗∗p < 0.01, ∗p < 0.05 versus Exo-NC.

Figure 6

Silencing exosome-derived circTRPS1 hinders BCa growth and metastasis in vivo via GLS1 regulation (A) Tumor sizes were measured and compared for exo-NC and exo-si-circTRPS1 intravenously treated T24 xenografts in nude mice with or without intraperitoneal treatment of BPTES. (B) Tumor volumes were measured separately. Data are presented as the mean ± SD. ∗∗∗p < 0.001, ∗∗p < 0.01 versus Exo-NC group. (C) IHC analysis of the expressions of GLS1, mTOR pathway marker (p-mTOR), ROS-regulated marker (Nrf2), EMT marker (vimentin and N-cadherin), cell apoptosis marker (Bcl-2), and cell proliferation marker (Ki-67) after being treated with si-circTRPS1 exosomes or/and BPTES. (D and E) H&E staining of metastatic nodules in mice lungs. Data are presented as the mean ± SD. ∗∗∗p < 0.001 versus Exo-NC group. (F and G) Bioluminescence imaging detected the metastasis ability of T24-luc after being treated with si-circTRPS1 exosomes or/and BPTES. Data are presented as the mean ± SD. ∗∗∗p < 0.001 versus Exo-NC group. (H) Schematic diagram of exosome-derived circTRPS1 in BCa tumor microenvironments.