PTBP3 Mediates IL-18 Exon Skipping to Promote Immune Escape in Gallbladder Cancer

Abstract

Gallbladder cancer (GBC) is the most common malignant tumor of the biliary system, with poor response to current treatments. Abnormal alternative splicing has been associated with the development of a variety of tumors. Combining the GEO database and GBC mRNA-seq analysis, it is found high expression of the splicing factor polypyrimidine region- binding protein 3 (PTBP3) in GBC. Multi-omics analysis revealed that PTBP3 promoted exon skipping of interleukin-18 (IL-18), resulting in the expression of ΔIL-18, an isoform specifically expressed in tumors. That ΔIL-18 promotes GBC immune escape by down-regulating FBXO38 transcription levels in CD8+T cells to reduce PD-1 ubiquitin-mediated degradation is revealed. Using a HuPBMC mouse model, the role of PTBP3 and ΔIL-18 in promoting GBC growth is confirmed, and showed that an antisense oligonucleotide that blocked ΔIL-18 production displayed anti-tumor activity. Furthermore, that the H3K36me3 promotes exon skipping of IL-18 by recruiting PTBP3 via MRG15 is demonstrated, thereby coupling the processes of IL-18 transcription and alternative splicing. Interestingly, it is also found that the H3K36 methyltransferase SETD2 binds to hnRNPL, thereby interfering with PTBP3 binding to IL-18 pre-mRNA. Overall, this study provides new insights into how aberrant alternative splicing mechanisms affect immune escape, and provides potential new perspectives for improving GBC immunotherapy.

Keywords: IL‐18; PTBP3; alternative splicing; gallbladder cancer; immune escape.

Figure 1

Over‐expression of the splicing factor PTBP3 is significantly associated with poor prognosis in gallbladder cancer. A) Heatmap of differentially expressed genes obtained by integrating four GEO datasets (GSE76633, GSE100363, GSE139682, and GSE62335). B) Heatmap of differentially expressed genes from 12 pairs of gallbladder cancer mRNA sequencing. C) The panel on the left is Venn plots based on the GEO database of differentially expressed genes, 12 pairs of differentially expressed genes made from the gallbladder cancer transcriptome and alternative splicing factor datasets; the panel on the right is a list of 20 intersecting genes. D) The left panel is a representative image from tissue microarray including cholecystitis and gallbladder cancer tissue immunohistochemistry results using anti‐PTBP3 antibody, and the right panel is the immunohistochemistry score. E) Slope chart of PTBP3 mRNA expression in 40 pairs of gallbladder cancer tissues. F) Forest plot of 40 gallbladder cancer patients analyzed by clustering based on high and low PTBP3 expression levels. G) Kaplan‐Meier overall survival curve of gallbladder cancer patients according to PTBP3 mRNA expression level showed that patients with high expression of PTBP3 had a worse prognosis.

Figure 2

PTBP3 promotes IL‐18 exon skipping in gallbladder cancer. A) Schematic representation of the multi‐omics analysis of NOZ cells. In brief, PTBP3 was knocked down in NOZ cells, while cell supernatants were later collected for Olink proteomics assay and extracted cell total RNA was collected for mRNA sequencing. B) Alternative splicing differential event analysis based on transcriptome sequencing after PTBP3 knockdown in NOZ cells. C) Venn diagram analysis combining differential proteins detected by Olink proteomics, differential exon skipping events in mRNA‐seq, mRNAs bound by PTBP3 analyzed by our RIP‐seq data and previously reported RIP‐seq data. D) Schematic of the two IL‐18 transcripts and primer design positions. “cIL‐18″ stands for classic IL‐18. E) The left panel represents the DNA electrophoresis of the two isoforms of IL‐18 detected by applying PCR to untreated NOZ cells; the right panel represents the results of sanger sequencing of two bands. The two bands differ for exon 6. F) Gallbladder cancer single cell transcriptome analysis showed that gallbladder cancer cell could express IL‐18 using UMAP plots. G) RT‐PCR analysis IL‐18 isoform under PTBP3 knockdown. PSI = splice_in gray value / (splice_in gray value+splice_out gray value). H) RT‐PCR analysis of IL‐18 isoform under PTBP3 overexpression. I) RT‐PCR analysis of RIP results using anti‐PTBP3 antibody to test for Intron2 and Exon3 level. J) FISH/IF assay to analyze IL‐18 pre‐mRNA and PTBP3 localization in NOZ. K) Schematic diagram of the IL‐18 minigene constructs. L) Splicing analysis of IL‐18 minigene reporter in NOZ cells (*P<0.05, **P<0.01, Student's t‐test). Data are expressed as mean±SD, n = 3. M) RT‐PCR analysis of two isoforms of IL‐18 in 12 pairs of gallbladder cancer tissues and paracancerous tissues. N) Correlation analysis of IL‐18 PSI value and PTBP3 expression shows a negative correlation between the two by Spearman anaysis.

Figure 3

PTBP3 promotes gallbladder cancer immune escape through ΔIL‐18. A) Lollipop Chart of analysis of immune cell infiltration associated with PTBP3 expression based on 12 pairs of GBC mRNA‐seq data. B) The panel in the upper position is a scatter plot of correlation between PTBP3 immunohistochemical score and CD8+T cell infiltration by Spearman analysis; the panel in the bottom position is a scatter plot of the correlation between IL‐18 PSI value and CD8+T cell infiltration by Spearman analysis. C) Labeling of gallbladder cancer tissues with KI67, PTBP3 and CD8A using multiplex immunohistochemistry. D) Evaluation of tumor cell killing capacity of T cells after PTBP3 knockdown in tumor cells using flow cytometry (CD8+T cells and tumor cells co‐cultured at a ratio of 1:1 for 48 h). In brief, the treated tumor cells were spread in a 6‐well plate, and after the cells were completely attached to the wall, a 1:1 quantity of CD8+T cells was added. At 48 hours later, the residual liquid was removed from each well, and cells were washed with PBS followed by flow cytometry detection. E) Effect of PTBP3 knockdown in GBC cells on PD‐1 expression in CD8+T cells using western blotting in a co‐culture system (CD8+T cells and tumor cells co‐cultured at a ratio of 1:1 for 48 h). F) Measurement of subcutaneous tumor growth volume in HuPBMC NOG mice. G) Measurement of tumor weight in HuPBMC NOG mice. H) Illustration of a subcutaneous tumor in HuPBMC NOG mice. I) Schematic diagram of a series of ASO drugs designed to target the splicing site of IL‐18. J) Transfection of ASO in tumor cells followed by RT‐PCR assay to test the inhibitory efficiency of the ASO drug on IL‐18. K) Evaluation of tumor cell killing capacity of T cells after PTBP3 overexpression or ASO4 treatment in tumor cells using flow cytometry (CD8+T cells and tumor cells co‐cultured at a ratio of 1:1 for 48 h). L) Illustration of a subcutaneous tumor in HuPBMC NOG mice. M) Measurement of subcutaneous tumor growth volume in HuPBMC NOG mice. N) Measurement of tumor weight in HuPBMC NOG mice. O) Labeling of CD8A and CD3 in subcutaneous tumors using multiplex immunohistochemistry. Statistical tests involved: *P < 0.05, **P < 0.01, Student's t‐test; Data are expressed as mean±SD, n = 3.

Figure 4

ΔIL‐18 reduces PD‐1 ubiquitination by inhibiting FBXO38 transcription. A) CD8+T cells with tumor cell (with or without PTBP3 knockdown) supernatants were treated with 60 µg ml⁻¹ CHX, then the proteins were extracted according to the time point for detecting PD‐1 levels using western blotting. B) Levels of PD‐1 in CD8+T cells with tumor cell (with or without PTBP3 knockdown) supernatants were detected by western blotting after 12 h treatment with MG132 (20 µM). C) Levels of FBXO38 in CD8+T cells with tumor cell (with or without PTBP3 knockdown) supernatants were detected by western blotting. D) Levels of FBXO38 in CD8+T cells with tumor cell (with or without PTBP3 knockdown) supernatants were detected by qPCR. E) Levels of ubiquitination of PD‐1 in CD8+T cells with tumor cell (with or without PTBP3 knockdown) supernatants were detected by western blotting. F) Levels of transcriptional activity of FBXO38 in CD8+T cells with tumor cell (with or without PTBP3 knockdown) supernatants were detected by luciferase reporter assay. G) The up panel, the absorbent peak graph at 260 and 280 nm of ΔIL18 purified by size‐exclusive chromatography. The down panel, the SDS‐page stained by Coomassie blue, showed the purified ΔIL18 protein. H) PD‐1 and FBXO38 expression were detected by western blotting after treatment of CD8+T cells with two concentrations of ΔIL‐18. I) FBXO38 expression were detected by qPCR after treatment of CD8+T cells with two concentrations of ΔIL‐18. J) Levels of transcriptional activity of FBXO38 in CD8+T cells were detected by luciferase reporter assay. K) Levels of ubiquitination of PD‐1 in CD8+T cells with ΔIL‐18 treatment were detected by western blotting. L) Measurement of subcutaneous tumor growth volume in HuPBMC NOG mice. M) Illustration of a subcutaneous tumor in HuPBMC NOG mice. N) Measurement of tumor weight in HuPBMC NOG mice. Statistical tests involved: *P < 0.05, **P < 0.01, Student's t‐test; Data are expressed as mean±SD, n = 3.

Figure 5

The histone modification H3K36me3 couples IL‐18 transcription and alternative splicing. A) The left panel showed the results of validation of the effect of knockdown of H3K36 methyltransferase SETD2 on exon skipping of IL‐18 using RT‐PCR assay in NOZ; the right panel showed validation of the effect of knockdown of H3K4 methyltransferase ASH2 on exon skipping of IL‐18 using RT‐PCR assay in NOZ. B) Schematic representation of the SETD2 domain. C) RT‐PCR experiments were performed after transfection of NOZ with Wt‐SETD2 and Mut‐SETD2 to determine their effects on IL‐18 exon skipping. D) Western blotting results after immunoprecipitation experiments using MRG15 and PTBP3 antibodies, respectively. E) The PTBP3 normal control group and PTBP3 siRNA group were subjected to RIP experiments with MRG15 antibody to investigate the binding of IL‐18 by MRG15, respectively. F) RIP experiments using PTBP3 antibody after overexpression of MRG15 to probe the binding of PTBP3 to IL‐18. G) RT‐PCR experiments were performed after knockdown of PTBP3 and overexpression of MRG15, respectively H) RT‐PCR experiments were performed after knockdown of MRG15 and overexpression of PTBP3, respectively. I) RT‐PCR experiments were performed after knockdown of MRG15 and overexpression of Wt‐SETD2, respectively. J) RT‐PCR experiments were performed after knockdown of SETD2 and overexpression of MRG15, respectively. K) Analysis of H3K36me3 peaks on IL‐18 in multiple tumor cells using Cistrome Data Browser. L) Chromatin immunoprecipitation of H3K36me3 and MRG15 along IL‐18. Statistical tests involved: *P < 0.05, **P < 0.01, Student's t‐test; Data are expressed as mean±SD, n = 3.

Figure 6

SETD2/hnRNPL interferes with PTBP3 binding to IL‐18 pre‐mRNA. A) Validation of the effect of knockdown of hnRNPL on exon skipping of IL‐18 using RT‐PCR assay in NOZ. B) RIP assay with anti‐hnRNPL antibody after overexpression of Wt‐SETD2∖Mut‐SETD2∖Mut2‐SETD2. C) RT‐PCR experiments were performed after knockdown of hnRNPL and overexpression of Mut‐SETD2, respectively. D) RT‐PCR experiments were performed after knockdown of hnRNPL and overexpression of PTBP3, respectively. E) RT‐PCR experiments were performed after knockdown of PTBP3 and overexpression of hnRNPL, respectively. F) Validation of PTBP3 binding to hnRNPL using western blotting with or without RNase. G) Myc‐tagged hnRNPL plasmid and different Flag‐tagged deletion mutant PTBP3 plasmid were transfected into 293T cells and performed IP assays. H) Flag‐tagged PTBP3 plasmid and different Myc‐tagged deletion mutant hnRNPL plasmid were transfected into 293T cells and performed IP assays. I) Molecular docking results of PTBP3 and hnRNPL presented by PyMol software. J) RT‐PCR experiments were performed after overexpression of hnRNPL‐Δ2 under transfection with different plasmid concentrations. K) RIP assay with anti‐PTBP3 antibody after overexpression of hnRNPL under transfection with different plasmid concentrations. L) RIP assay with anti‐PTBP3 antibody after knockdown of hnRNPL. Statistical tests involved: *P < 0.05, **P < 0.01, Student's t>‐test; Data are expressed as mean±SD, n = 3.