The role of KRT18 in lung adenocarcinoma development: integrative bioinformatics and experimental validation

Abstract

Lung adenocarcinoma (LUAD) represents one of the most common subtypes of lung cancer with high rates of incidence and mortality, which contributes to substantial health and economic demand across the globe. Treatment today mainly consists of surgery, radiotherapy, and chemotherapy, but their efficacy in advanced stages is often suboptimal and emphasizes the clear need for new biomarkers and therapeutic targets. Using comprehensive bioinformatics analyses consisting of the Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO), Human Protein Atlas (HPA) and Clinical Proteomic Tumor Analysis Consortium (CPTAC), immune infiltration analysis and functional enrichment analysis, and single-cell analysis, we examined the potential of keratin 18 (KRT18) as a candidate biomarker in advanced LUAD. KRT18 was significantly elevated in LUAD tissue relative to normal adjacent tissue (p < 0.05), and its expression was correlated with poor clinical-pathological features and inferior prognostic outcome. Furthermore, KRT18 expression was associated with several populations of immune cells, suggesting KRT18 may contribute to the local tumor microenvironment and potentially pathways of immune evasion. Survival analysis indicated that elevated KRT18 expression correlated with poor overall survival (OS), disease-specific survival (DSS), and progression-free interval (PFI), reinforcing its legitimacy as a prognostic tool (AUC = 0.846). Importantly, gene enrichment analysis found KRT18-associated genes enriched for pathways associated with lymphocyte differentiation and immune response pathways, which provides mechanistic insight into biological effects attributed to KRT18. Notably, NU.1025 has demonstrated the capability of reversing KRT18-modulated oncogenic features, and targeted therapeutic strategies can be developed moving forward. In conclusion, our data demonstrate that KRT18 has utility as a potential biomarker but may also serve as a therapeutic target in LUAD and merit further investigation into underlying mechanistic functions and potential therapeutic roles in the clinic.

Keywords: Biomarker; Immune microenvironment; Keratin 18; Lung adenocarcinoma; Prognosis; Targeted therapy.

Fig. 1

Expression differences of KRT18. A Expression levels of KRT18 in pan-cancer. B Expression differences of KRT18 in lung adenocarcinoma in non-paired samples. C Expression differences of KRT18 in lung adenocarcinoma in paired samples. D Expression of KRT18 in the GSE31547 dataset. E Expression of KRT18 in the GSE40791 dataset. F–G Immunohistochemical analysis of KRT18 in lung adenocarcinoma in the HPA database. H Gene expression differences between tumor and normal tissues in the CPTAC-LUAD cohort

Fig. 2

The significance of KRT18 expression in clinical settings. A–E Correlation analysis of KRT18 expression and clinical pathological features. F ROC curve assessing diagnostic value. G–I K–M curves of KRT18 expression in OS, DSS, and PFI

Fig. 3

Functional enrichment analysis. A, B Heatmap of the top 10 genes positively and negatively correlated with the expression of KRT18. C, D GO and KEGG enrichment analysis. E, F GSEA

Fig. 4

Immune infiltration analysis. A Lollipop chart of the correlation analysis between the expression of KRT18 and 24 types of immune cells. B Estimation algorithm calculating the matrix and immune scores. C, D Differences in immune infiltration results between high and low expression groups

Fig. 5

Differences in various immune genes, immune subtypes in high/low expression groups of the KRT18 gene, and the expression differences of KRT18 in different molecular subtypes. A Expression differences of immune stimulation genes, immune suppression genes, chemokines, and human leukocyte antigens in high/low expression groups of KRT18. B Differences in immune subtypes of lung adenocarcinoma in high/low expression groups of the KRT18 gene. C Expression differences of KRT18 in molecular subtypes

Fig. 6

Gene alteration analysis. A Mutation status of the KRT18 gene in different cancer types. B Survival curve of KRT18 gene alterations. C Spearman correlation between copy number variation scores and KRT18 gene expression levels. D Differences in KRT18 gene expression across different types of copy number variations. The x-axis represents copy number types, where C1_Deep Deletion indicates homozygous deletion, C2_Shallow_deletion indicates single copy deletion, C3_Diploid indicates diploid normal copy, C4_Gain indicates low-level copy number amplification, and C5_Amplification indicates high-level copy number amplification. The y-axis represents the corresponding gene expression levels

Fig. 7

Single-cell level analysis and targeted drug screening. A Expression levels of KRT18 in different datasets and cell types presented in a heatmap. B, C, E Single-cell level analysis of the KRT18 gene in the NSCLC_GSE117570 dataset. D Potential small molecules and drugs predicted by the XSum algorithm to correct the biological effects caused by the dysregulation of target gene expression

Fig. 8

Functional experiments of KRT18 in lung cancer A549 and H1299 cells. A qPCR analyzed KRT18 expression in A549 and H1299 cells transfected with si-NC or si-KRT18. B, C CCK-8 cell proliferation line graph. D, E Cell scratch assay