Exosomal circ-PTPN22 and circ-ADAMTS6 mark T cell exhaustion and neutrophil extracellular traps in Asian intrahepatic cholangiocarcinoma

Abstract

Intrahepatic cholangiocarcinoma (ICC) is a liver tumor featured by challenges of non-invasive early diagnosis and a higher prevalence rate in Asian countries. These characteristics necessitate the development of liquid biopsy and immunotherapy methods to improve the prognosis of patients with ICC. Herein, we conducted a pilot study on the transcriptome of tumor tissues, adjacent normal tissues, and plasma exosomes of Asian patients with ICC from northern and southern China. We identified a subgroup of immunogenic Asian ICC, which is different from Caucasian ICC and is characterized by T cell exhaustion and neutrophil extracellular traps. The levels of circ-PTPN22 (hsa_circ_0110529) and circ-ADAMTS6 (hsa_circ_0072688), potential circRNA biomarkers, were elevated in the ICC tumor tissues and plasma exosomes of this subgroup than in the other subgroups and normal controls. These circRNAs were derived from post-transcriptional backsplicing of PTPN22 and ADAMTS6 that were expressed in T cells and endothelial cells, respectively, in the ICC microenvironment. Our results revealed a subgroup of Asian ICC characterized by T cell exhaustion and neutrophil extracellular traps and marked by elevated levels of circ-PTPN22 and circ-ADAMTS6 in tumor tissues and plasma exosomes. This subgroup is potentially detectable by plasma exosomal circRNAs and treatable with immune checkpoint blockade.

Keywords: MT: Non-coding RNAs; T cell exhaustion; circular RNA; immune checkpoint blockade; intrahepatic cholangiocarcinoma; neutrophil extracellular traps; plasma exosome.

Graphical abstract

Figure 1

Landscape of circRNAs in tumor tissues and plasma exosomes of Asian ICC (A) Sampling and sequencing of the Asian ICC patient cohort. (B) Representative plasma exosomes detected by transmission electron microscopy. (C) The size of plasma exosomes analyzed by high-sensitivity flow cytometry. (D) Western blotting of plasma exosome markers, including TSG101, CD63, and calnexin proteins. (E) Principal-component analysis (PCA) of the circRNA profile in ICC tumor tissues and adjacent normal tissues, labeled by sample types. (F) PCA of the circRNA profile in ICC tumor tissues and adjacent normal tissues, labeled by medical centers. (G) PCA of the circRNA profile in plasma exosomes of patients with ICC and healthy controls (provided by ExoRbase), labeled by sample types. (H) Top differential circRNAs in ICC tumor tissues versus adjacent normal tissues. (I) Top differential circRNAs in plasma exosomes of patients with ICC versus healthy controls. (J) The abundance of CDR1as and circ-0000284 in ICC tumor tissues and adjacent normal tissues. (K) The abundance of CDR1as and circ-0000284 in plasma exosomes of patients with ICC and healthy controls.

Figure 2

Transcriptomic analysis of immune infiltration in Asian ICC tumor tissues (A) Unsupervised hierarchical clustering of the transcriptome of Asian and Caucasian ICC tumor tissues. (B) Immune infiltration in the three ICC subgroups, predicted by ImmuCellAI, xCell, TIMER, and quanTIseq. (C) Infiltration of CD8⁺ T cells and regulatory T cells in ICC subgroups, predicted by ImmuCellAI. (D) Infiltration of neutrophils and macrophages in ICC subgroups, predicted by ImmuCellAI. (E) Gene set variation analysis (GSVA) of ICC tumor tissues based on the Reactome database. (F) Expression of immune checkpoint genes, neutrophil extracellular trap-associated genes, FCGR family genes, and mismatch repair genes.

Figure 3

Potential circRNA biomarkers of the immunogenic Asian ICC subgroup (A) PCA of the circRNA profile in ICC tumor tissues and adjacent normal tissues, labeled by subgroups. (B) PCA of the circRNA profile in plasma exosomes of ICC tumor tissues and healthy controls (provided by ExoRbase). (C) Top differential circRNAs in subgroup 2 versus subgroup 3 and adjacent normal tissues. (D) Top differential circRNAs in plasma exosomes of subgroup 2 versus healthy controls. (E) The abundance of circ-ADAMTS6, circ-PTPN22, and CDR1as in subgroups 2 and 3, and adjacent normal tissues. (F) Expression levels of ADAMTS6 and PTPN22 in subgroups 2 and 3, and adjacent normal tissues. (G) The abundance of circ-ADAMTS6, circ-PTPN22, and CDR1as in plasma exosomes of subgroup 2 and healthy controls.

Figure 4

Validation of circRNA biomarkers and host genes in Asian ICC tumor tissues collected by independent studies (A) Abundance of circ-ADAMTS6, circ-PTPN22, and CDR1as in the ICC tumor tissues and adjacent tumor tissues collected by the GSE181523 (southern China) dataset and the GSE148561 (northern China) dataset. (B) Correlation between the abundance of circ-ADAMTS6 and circ-PTPN22 in ICC tumor tissues and adjacent normal tissues collected by the GSE181523 and GSE148561. (C) Expression of ADAMTS6 and PTPN22 in subgroups 1, 2, and 3, and adjacent normal tissues. (D) UMAP embedding of the single-cell transcriptome of ICC tumor tissues. (E) Expression of PTPN22, ADAMTS6, and immune checkpoint genes in different cell types.

Figure 5

Quantitative RT-PCR analysis of circRNA biomarkers and host genes in Asian ICC tumor tissues (A) Sanger sequencing of the PCR product of circ-PTPN22 and circ-ADAMTS6. Red dotted lines indicate the junction site of backsplicing. (B) qRT-PCR analysis of PTPN22 mRNA, ADAMTS6 mRNA, circ-PTPN22, and circ-ADAMTS6 abundance in ICC subgroups and adjacent normal tissues. (C) Agarose gel electrophoresis of the PCR product of PTPN22 mRNA, ADAMTS6 mRNA, circ-PTPN22, and circ-ADAMTS6.