Super-enhancer-driven ZFP36L1 promotes PD-L1 expression in infiltrative gastric cancer

  1. Xujin Wei
  2. Jie Liu
  3. Jia Cheng
  4. Wangyu Cai
  5. Wen Xie
  6. Kang Wang
  7. Lingyun Lin
  8. Jingjing Hou
  9. Jianchun Cai  Is a corresponding author
  10. Huiqin Zhuo  Is a corresponding author
  1. Endoscopic Center, The First Affiliated Hospital, Fujian Medical University, China
  2. The Graduate School of Fujian Medical University, China
  3. Department of Gastrointestinal Surgery, Zhongshan Hospital of Xiamen University, Institute of Gastrointestinal Oncology, School of Medicine, Xiamen University, China
  4. Xiamen Municipal Key Laboratory of Gastrointestinal Oncology, China
7 figures and 2 additional files

Figures

Figure 1 with 1 supplement
Immune escape signatures of Ming infiltrative gastric cancer (GC) driven by super-enhancers (SEs).

(A) Hematoxylin-eosin staining of GC. (B) SE peaks of H3K27ac histone modifications. (C) Distribution of H3K27ac SE peaks. (D) Venn diagrams of SE-driven protein-coding genes and chromosomal landscape of infiltrative SE-driven genes. (E) Gene Ontology-Kyoto Encyclopedia of Genes and Genomes (GO-KEGG) pathway enrichment for SE-driven genes. (F) Unsupervised hierarchical clustering using 16 prognostic genes in GC patients from The Cancer Genome Atlas (TCGA) datasets. (G) Kaplan-Meier survival curves of two subgroups. (H) Immune infiltration analysis. (I) Immunohistochemical (IHC) scores of programmed death-ligand 1 (PD-L1) in 70 GC tissues. TSS, transcription start site; TTS, transcription termination site.

Figure 1—figure supplement 1
Epigenetic heterogeneity between expanding and infiltrative gastric cancer.

(A) Proteomic differences between expanding and infiltrative gastric cancer (GC). (B) Typical enhancers peaks of histone H3K27ac modifications in expanding and infiltrative GC. (C) Chromosomal landscape of expanding super-enhancer (SE)-driven genes. (D) Forest plot of 16 SE-driven protein-coding genes associated with poor prognosis in infiltrative GC based on The Cancer Genome Atlas (TCGA) datasets.

Figure 2 with 1 supplement
Expression levels of ZFP36L1 in infiltrative gastric cancer (GC) driven by ZFP36L1-SE.

(A) Friends analysis of 16 super-enhancer (SE)-driven prognostic genes. (B) Correlations between clinical characteristics and the ZFP36L1 mRNA expression in The Cancer Genome Atlas (TCGA). (C) Tumor immune dysfunction and exclusion scores in high and low expression levels of ZFP36L1 groups. (D) Correlation between immune infiltration cells and the mRNA expression level of ZFP36L1 in TCGA. (E) Correlation between ZFP36L1 mRNA expression and immune checkpoints in TCGA. (F) H3K27ac signals of SEs and target genes in infiltrative GC. (G) H3K27ac signals of ZFP36L1-SE in GC. Protein expression of ZFP36L1 in (H) 6 GC cell lines and (I) 12 tumor and paired adjacent normal tissues of patients with infiltrative GC. (J) Expression level of ZFP36L1 after SE inhibition treatment (n=3). (K) H3K27ac signals of ZFP36L1-SE after SE inhibition treatment (n=3). ***, p<0.001; **, p<0.01; *, p<0.05; ns, p≥0.05. (B–C) t-Test, (D–E) Spearman’s correlation, (J) one-way ANOVA with post hoc Tukey HSD test, and (K) Welch’s ANOVA with a Games-Howell post hoc test were used for statistical analysis.

Figure 2—figure supplement 1
Kaplan-Meier for (A) disease-specific survival and (B) progress-free interval plot of ZFP36L1.
ZFP36L1 promotes IFN-γ-induced PD-L1 expression.

(A) mRNA and (C) protein expression of PD-L1 in gastric cancer (GC) cell lines with or without ZFP36L1 knockdown (n=3). (B) mRNA and (D) protein expression of PD-L1 in GC cell lines with or without ZFP36L1 overexpression (n=3). Fluorescent signal of the PD-L1 membrane protein in GC cell lines with or without ZFP36L1 (E) knockdown and (F) overexpression (n=3). ***, p<0.001; **, p<0.01; *, p<0.05; ns, p≥0.05. (A–B) Welch’s ANOVA with a Games-Howell post hoc test, (E) one-way ANOVA post hoc Tukey HSD, and (F) t-test were used for statistical analysis.

Figure 3—source data 1

PDF file containing original western blots for Figure 3.

https://cdn.elifesciences.org/articles/96445/elife-96445-fig3-data1-v1.pdf
Figure 3—source data 2

Original files for western blot analysis displayed in Figure 3.

https://cdn.elifesciences.org/articles/96445/elife-96445-fig3-data2-v1.zip
Figure 4 with 1 supplement
SPI1 binds to the ZFP36L1-SE region and drives the regulation of PD-L1.

(A) Schematic of transcription factor (TF) motif enrichment in ZFP36L1-E1. (B) Kaplan-Meier survival plot of SPI1 in The Cancer Genome Atlas (TCGA). (C) The mRNA expression of ZFP36L1 after TFs plasmid transfection (n=3). (D) The ZFP36L1 protein expression in cell lines overexpressing SPI1. (E) Correlation between SPI1 and PD-L1 mRNA expression in TCGA. (F) PD-L1 protein expression in simultaneous SPI1 overexpression and ZFP36L1 knockdown cells. (G) Prediction of SPI1-BRD4-P300 binding on the STRING website. Co-immunoprecipitation between (H) exogenous SPI1 and BRD4 in 293T cells, or (I) endogenous SPI1 and BRD4 in MGC803. (J) SPI1 directly interacts with BRD4 in vitro by GST pull-down experiment. (K) ZFP36L1-E1 binding of different TFs detected using dual-luciferase assay (n=3). (L) SPI1 enriched regions in ZFP36L1-E1 detected by chromatin immunoprecipitation (ChIP) assay (n=3). (M) Different binding sites of SPI1 in ZFP36L1-E1 detected using dual-luciferase assay (n=6). (N) Wild-type and motif-deletion mutant E1C binding of SPI1 detected using dual-luciferase (n=5). ***, p<0.001; **, p<0.01; *, p<0.05; ns, p≥0.05. (C ,K) One-way ANOVA post hoc Tukey HSD test, (L) t-test, (M) Welch’s t-test, and (N) Welch’s ANOVA with a Games-Howell post hoc test were used for statistical analysis.

Figure 4—source data 1

PDF file containing original western blots for Figure 4.

https://cdn.elifesciences.org/articles/96445/elife-96445-fig4-data1-v1.pdf
Figure 4—source data 2

Original files for western blot analysis displayed in Figure 4.

https://cdn.elifesciences.org/articles/96445/elife-96445-fig4-data2-v1.zip
Figure 4—figure supplement 1
Transcriptional regulation of PD-L1 by SPI1.

(A) Transcription factor binding sites in ZFP36L1-SE based on chromatin immunoprecipitation (ChIP)-seq data from the Signaling Pathways Project. (B) The PD-L1 protein expression in simultaneous SPI1 overexpression and ZFP36L1 knockdown cells.

ZFP36L1 positively regulates PD-L1 by activating HDAC3 mRNA decay.

(A) Word cloud of predicted ZFP36L1 target genes. (B) The mRNA expression of predicted target genes in MGC803 cell overexpressing ZFP36L1 (n=5). (C) HDAC3 protein expression in ZFP36L1 knockdown cells. (D) Effect of HDAC3 on CD274 promoter activity in 293T using dual-luciferase assay (n=3). (E) Correlation between changes of histone H3K27 acetylation and PD-L1 protein expression in MKN45 cells overexpressing HDAC3. (F) Chromatin immunoprecipitation (ChIP) assay showing the histone H3K27 acetylation levels of CD274 promoter regions in MKN45 cells overexpressing HDAC3 (n=3). (G) Correlation between changes of histone H3K27 acetylation and ZFP36L1 protein expression. (H) The H3K27ac and SPI1 enrichment of ZFP36L1-E1C regions using ChIP assay (n=3). (I, J) PD-L1 protein expression in simultaneous ZFP36L1 and HDAC3 overexpression cells. (M) HDAC3 mRNA decay in ZFP36L1 knockdown and overexpression cells after actinomycin D treatment. (N) ZFP36L1 mRNA-binding level by RNA-binding protein immunoprecipitation (n=3). (O) ZFP36L1 mRNA-binding site in AU-rich element (ARE) of 3ʹUTR confirmed using dual-luciferase assay (n=6). (P) CCCH-type zinc finger domain of ZFP36L1 protein binding to HDAC3 mRNA confirmed using RNA pull-down assay. ARE, adenylate uridylate- (AU-) rich element; 3ʹUTR, 3ʹ untranslated region. ***, p<0.001; **, p<0.01; *, p<0.05; ns, p≥0.05. (B, F) Wilcoxon rank sum test, (D) one-way ANOVA post hoc Tukey HSD test, (H) t-test, (N) Welch’s t-test, and (O) Kruskal-Wallis with Dunn’s test were used for statistical analysis.

Figure 5—source data 1

PDF file containing original western blots for Figure 5.

https://cdn.elifesciences.org/articles/96445/elife-96445-fig5-data1-v1.pdf
Figure 5—source data 2

Original files for western blot analysis displayed in Figure 5.

https://cdn.elifesciences.org/articles/96445/elife-96445-fig5-data2-v1.zip
Positive correlation between ZFP36L1 and PD-L1 in vivo.

(A) Immunohistochemical (IHC) staining scores of SPI1, ZFP36L1, and HDAC3 protein in 70 PD-L1-positive gastric cancer (GC) tumor tissues. (B) Scatter plot to compute correlation between SPI1 and ZFP36L1. (C) Three patients with infiltrative GC shown as representative images. (D) Schematic diagram of experiments in C57BL/6J and BALB/c-nu mice. (E) Subcutaneous tumor weight from mice injected with control and ZFP36L1 knockdown cells (n=6). (F) The PD-L1 mRNA expression in subcutaneous tumors (n=6). (G) IHC staining of ZFP36L1 and PD-L1 protein in subcutaneous tumors. (H) Number of pulmonary metastases in different groups of C57BL/6J mice after tail vein injection of control and ZFP36L1 knockdown cells (n=6). (I) Hematoxylin & eosin (HE) and IHC staining of CD8α in pulmonary metastases. (J) Number of pulmonary metastases in different groups of BALB/c-nu mice after tail vein injection of control and ZFP36L1 knockdown cells (n=5). ***, p<0.001; **, p<0.01; *, p<0.05; ns, p≥0.05. (E, H, J) One-way ANOVA post hoc Tukey HSD test and (F) Kruskal-Wallis with Dunn’s test were used for statistical analysis.

Schematic diagram of SPI1-ZFP36L1-HDAC3-PD-L1 signaling axis (created with gdp.renlab.cn).

Additional files

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Xujin Wei
  2. Jie Liu
  3. Jia Cheng
  4. Wangyu Cai
  5. Wen Xie
  6. Kang Wang
  7. Lingyun Lin
  8. Jingjing Hou
  9. Jianchun Cai
  10. Huiqin Zhuo
(2024)
Super-enhancer-driven ZFP36L1 promotes PD-L1 expression in infiltrative gastric cancer
eLife 13:RP96445.
https://doi.org/10.7554/eLife.96445.2