Pan cancer research reveals the role of PTGDS in tumor suppression and immune regulation

Abstract

Prostaglandin D2 synthase (PTGDS), a newly identified anti-tumor target, shows promise in inhibiting various cancers and plays significant roles in the tumor microenvironment and immune regulation, yet a comprehensive pan-cancer analysis of its expression and prognostic value remains lacking. This study used multi-omics data from public databases to assess PTGDS's expression, mutation, and modification in multiple cancers, integrated single-cell and spatial transcriptomic data to explore its relationship with immune cells, and conducted in vitro and in vivo experiments in breast cancer (BRCA). Results showed that PTGDS is significantly dysregulated in most cancers, with its expression associated with different outcomes depending on cancer type. It correlates with epigenetic and biological functions, and low expression in BRCA indicates poor prognosis. Overexpression of PTGDS can inhibit breast cancer cell proliferation and invasion, increase CD8+ T - cell activity, and enhance anti-tumor immunity. Combining it with anti-PD-L1 improves BRCA treatment. PTGDS is a potential prognostic biomarker and a novel immunotherapy target for BRCA.

Fig. 1. Differential expression of PTGDS and prediction of cancer survival.

a Diseases associated with PTGDS. The red dotted line indicates PTGDS-associated cancers. b Expression levels of PTGDS mRNA in pan-cancer and corresponding control tissues. c Box plots of log2 expression levels of PTGDS mRNA between tumor and normal tissues in 9 types of cancer. T and N represent tumor and normal tissues, respectively. d Expression levels of PTGDS across four different stages of 5 types of cancer. e Differences in protein expression between normal and primary tumor tissues in 5 types of cancer. f Kaplan–Meier curves are plotted to predict OS and DSS in TCGA patients. (Z-score test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.).

Fig. 2. PTGDS is associated with genomic instability in cancer.

a Genomic alterations of PTGDS in TCGA pan-cancer, including mutations, amplifications, and deep deletions. b Landscape of PTGDS SNVs in pan-cancer, including missense mutations, frameshift deletions, and splice site alterations. c Kaplan–Meier plots are drawn to show the prognostic significance of PTGDS CNVs in 6 types of cancer. d A radar plot showing the relationship between TMB, MSI, and PTGDS. e A bar chart showing the correlation coefficient between HRD or chromosomal ploidy and PTGDS expression. f Correlation scatter plots showing the association between PTGDS expression and neoantigen count in pan-cancer.

Fig. 3. PTGDS is involved in cancer DNA repair, stemness, and epigenetic regulation.

a Correlation between PTGDS expression and 12 cancer functional states. b Scatter plots showing the interrelationship between cancer DNA repair and PTGDS expression. c Bar charts showing the interrelationship between cancer stemness and PTGDS expression. d A table showing the correlation between PTGDS methylation levels and CTL-related factors. The third column represents CTL correlation, and the fourth column represents the CTL dysfunction z-score of the interaction term. e Scatter plots and Kaplan–Meier survival analyses show the association between PTGDS methylation levels and CTL markers, and survival analyses are grouped by high and low PTGDS methylation. f A heatmap showing the correlation between PTGDS expression and RNA regulation in pan-cancer. (Z-score test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.).

Fig. 4. PTGDS is involved in cancer immune-related pathways.

a Experiments verified a protein-protein interaction network of PTGDS binding partners. b Box plots showing the expression of PTGDS between pathway-level expression change groups and non-change groups in 6 types of cancer. c Correlation between PTGDS and the top 5 PTGDS co-expressed genes in each cancer type (left) and all cancer samples (right). d A circular plot of GO pathways enriched by the top 100 PTGDS co-expressed genes, with the top genes of each path on the left side of the corresponding color band. e GSEA analysis of the enrichment of KEGG and HALLMARK items in PANCancer, grouped by the median expression of PTGDS.

Fig. 5. PTGDS is negatively correlated with immune infiltration and cytokine interactions.

a Bar charts showing the correlation between PTGDS and ESTIMATEScore, ImmuneScore, and StromalScore (left), and scatter plots of the top 6 correlations for each Score (right). b Correlation between PTGDS expression and immune checkpoints in pan-cancer. c Correlation between PTGDS and immune subtypes. d Expression of PTGDS in 6 immune subtypes across six types of cancer. e Heatmaps showing the correlation between PTGDS expression and chemokines (top left), receptors (bottom left), and immune stimulants (top right), and a heatmap of PTGDS promoter methylation levels with immune stimulants (bottom right). f Box plots of PTGDS expression before and after cytokine therapy. ESTIMATEScore is a score that estimates the presence of stromal and immune cells in malignant tumor tissues using expression data. (Z-score test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.).

Fig. 6. PTGDS is a biomarker for dendritic cell infiltration in pan-cancer.

a CIBERSORT calculates immune cell infiltration in pan-cancer. b Correlations between PTGDS and immune cells in various tumors. Red represents a positive correlation, and purple represents a negative correlation. c Spatial transcriptomics shows the spatial expression of PTGDS, FCER1A, and CD40 markers. The dot color indicates the expression level of the marker. d Expression of PTGDS in pan-cancer single-cell clusters. e Violin plots of PTGDS expression in GEO pan-cancer single-cell clusters. (Z-score test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.).

Fig. 7. PTGDS is associated with NK cells, Treg, CAF infiltration, and CD8+ T cell activation.

a Heatmap of the association between PTGDS levels and NK cell, Treg, and CAF infiltration. b Correlations between PTGDS and NK, Treg, and CAF in 6 types of cancer with purity and purity-adjusted correlations. c Multiple algorithms calculate the correlation between PTGDS and CD8+ T cell infiltration. d Correlations between PTGDS expression and CTL, T cell dysfunction, and risk.

Fig. 8. PTGDS predicts treatment response and docking with PTGDS-targeted drugs.

a Box plots show the difference in PTGDS expression between responders and non-responders, and ROC shows the predictive accuracy of PTGDS levels for patient treatment response. b A heatmap showing the top 30 compounds, with experimentally induced transcriptional changes opposite to those affected by intermediate PTGDS expression grouping. Color bars and block colors indicate similarity scores. c Bar charts showing Spearman correlations between PTGDS grouping and mRNA changes induced by RNAactDrug drugs. d Homology modeling of PTGDS with drugs, with 2D images on the left representing drug 2D structures, interacting amino acid residues, and molecular forces. On the right are 3D models of PTGDS docking with drugs.

Fig. 9. Experiments in breast cancer cell lines after PTGDS overexpression.

a, b Colony formation assays. After PTGDS overexpression, the proliferation ability of breast cancer cell lines MDA-MB-231 and HCC1806 is significantly decreased. c, d Wound healing assays. After PTGDS overexpression, the migration ability of breast cancer cell lines MDA-MB-231 and HCC1806 is significantly decreased. e, f Transwell assays. After PTGDS overexpression, the migration and invasion abilities of breast cancer cell lines MDA-MB-231 and HCC1806 are significantly reduced. g, h Multiple immunohistochemical staining was performed to verify the co-localization of PTGDS CD8+ T cells and CD20+ B cells in clinical samples. (Z-score test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.).

Fig. 10. PTGDS overexpression enhances the efficacy of anti-PD-L1 treatment in vivo in BC.

a Schematic flowchart of the in vivo experimental procedure in BALB/c mice. b Images of subcutaneous xenograft tumors harvested from euthanized mice in different groups. c, d Tumor growth curves and tumor weights in different groups. e–g Flow cytometric analysis and quantitative assessment of CD3+ cells within CD45+ cells and CD8+ cells within CD3+ T cells in the tumors. (Z-score test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.).