TY - JOUR
T1 - Air pollution-related immune gene prognostic signature for hepatocellular carcinoma
T2 - network toxicology, machine learning and multi-omics analysis
AU - Pu, Lei
AU - Zhang, Xiaoyan
AU - Pu, Cheng
AU - Sun, Peng
N1 - Publisher Copyright:
Copyright © 2025 Pu, Zhang, Pu and Sun.
PY - 2025
Y1 - 2025
N2 - Background: Air pollution may crosstalk with immune system to promote hepatocellular carcinoma (HCC) development, but its precise mechanisms and prognostic significance remain unclear. Objective: This study aims to construct a prognostic signature for HCC based on air pollutant-related immune genes (APIGs). Methods: We obtained mRNA-seq and scRNA of HCC from GEO, TCGA and ICGC. AP-related target genes were retrieved from several online databases. APIGs were obtained using WGCNA, differential gene expression analysis and immune infiltration analysis. Molecular subtypes were conducted based on APIG expression to characterize immune features. A total of 101 combinations of 10 machine learning algorithms were used to construct an APIG-based prognostic signature (APIGPS). Furthermore, we performed qRT-PCR, survival analyses, functional enrichment, immune infiltration and single-cell analyses. Subsequently, LASSO, RF, and RFE-SVM were employed to identify diagnostic genes, followed by pan-cancer analysis. Results: We identified 19 APIGs. HCC samples were divided into 3 subtypes, with C1 exhibiting a pro-tumor immune microenvironment and poorer prognosis. APIGPS constructed by 7 APIGs (CDC25C, MELK, ATG4B, SLC2A1, CDC25B, APEX1, GLS), demonstrated robust predictive ability independent of clinical features. The biological pathway differences between APIGPS-based high- and low-risk groups involved immune responses and cell proliferation and migration. APIGPS genes had stable binding to 7 APs and were mainly expressed in macrophages, with HRG exhibiting higher macrophage abundance. CDC25C was identified as the hub gene after intersecting diagnostic genes and APIGPS genes. CDC25C was associated with survival of 10 cancers, MSI in 10 cancers, TMB in 21 cancers, and immune cell abundance in 13 cancers. Conclusions: We identified key APIGs and constructed a robust APIG-based prognostic signature for HCC. CDC25C was a key target through which APs impact HCC and multiple other cancers.
AB - Background: Air pollution may crosstalk with immune system to promote hepatocellular carcinoma (HCC) development, but its precise mechanisms and prognostic significance remain unclear. Objective: This study aims to construct a prognostic signature for HCC based on air pollutant-related immune genes (APIGs). Methods: We obtained mRNA-seq and scRNA of HCC from GEO, TCGA and ICGC. AP-related target genes were retrieved from several online databases. APIGs were obtained using WGCNA, differential gene expression analysis and immune infiltration analysis. Molecular subtypes were conducted based on APIG expression to characterize immune features. A total of 101 combinations of 10 machine learning algorithms were used to construct an APIG-based prognostic signature (APIGPS). Furthermore, we performed qRT-PCR, survival analyses, functional enrichment, immune infiltration and single-cell analyses. Subsequently, LASSO, RF, and RFE-SVM were employed to identify diagnostic genes, followed by pan-cancer analysis. Results: We identified 19 APIGs. HCC samples were divided into 3 subtypes, with C1 exhibiting a pro-tumor immune microenvironment and poorer prognosis. APIGPS constructed by 7 APIGs (CDC25C, MELK, ATG4B, SLC2A1, CDC25B, APEX1, GLS), demonstrated robust predictive ability independent of clinical features. The biological pathway differences between APIGPS-based high- and low-risk groups involved immune responses and cell proliferation and migration. APIGPS genes had stable binding to 7 APs and were mainly expressed in macrophages, with HRG exhibiting higher macrophage abundance. CDC25C was identified as the hub gene after intersecting diagnostic genes and APIGPS genes. CDC25C was associated with survival of 10 cancers, MSI in 10 cancers, TMB in 21 cancers, and immune cell abundance in 13 cancers. Conclusions: We identified key APIGs and constructed a robust APIG-based prognostic signature for HCC. CDC25C was a key target through which APs impact HCC and multiple other cancers.
KW - air pollutants
KW - hepatocellular carcinoma
KW - immune
KW - machine learning
KW - multi-omics analysis
KW - network toxicology
UR - https://www.scopus.com/pages/publications/105017093868
U2 - 10.3389/fimmu.2025.1638445
DO - 10.3389/fimmu.2025.1638445
M3 - 文章
C2 - 41019083
AN - SCOPUS:105017093868
SN - 1664-3224
VL - 16
JO - Frontiers in Immunology
JF - Frontiers in Immunology
M1 - 1638445
ER -