Molecular Subtypes Based on Cuproptosis-Related Genes and Tumor Microenvironment Infiltration Characterization in Colorectal Cancer

Abstract

Recent studies have demonstrated the biological significance of cuproptosis modification, a newly discovered programmed cell death, in tumor progression. Nonetheless, the potential role of cuproptosis-related genes (CRGs) in the immune landscape and tumor microenvironment (TME) formation of colorectal cancer (CRC) remains unknown. We comprehensively assessed cuproptosis modification patterns of 1339 CRC samples based on 27 CRGs and systematically analyzed the correlation of these patterns with TME. The CRG-score was constructed to quantify cuproptosis characteristics by LASSO and multivariate Cox regression methods, and its predictive capability was validated in an independent cohort. We identified three distinct cuproptosis modification patterns in CRC. The TME immune cell infiltration demonstrated immune heterogeneity among these three subtypes. Enrichment for multiple metabolism signatures was pronounced in cluster A. Cluster C was significantly correlated with the signaling pathways of immune activation-related, resulting in poor prognoses. Cluster B with mixed features possibly represents a transition phenotype or intratumoral heterogeneity. Then, based on constructed eight-gene CRG-score, we found that the signature could predict the disease-free survival of CRC patients, and the low CRG-score was related to increased neoantigen load, immunity activation, and microsatellite instability-high (MSI-H). Additionally, we observed significant correlations of the CRG-score with the cancer stem cell index and chemotherapeutic drug susceptibility. This study demonstrated that cuproptosis was correlated with tumor progression, prognosis, and TME. Our findings may improve the understanding of CRGs in TME infiltration characterization of CRC patients and contribute to guiding more effective clinical therapeutic strategies.

Figure 1

Multiomics analyses of CRGs in TCGA CRC data. (a) Whole-genome CRISPR-Cas9 positive selection screen using two copper ionophores (Cu-DDC and elesclomol-copper) in cells. Overlapping hits with a false discovery rate (FDR) value < 0.05 were analyzed. (b) Mutation frequencies of 27 CRGs in 535 patients with CRC from TCGA cohort. (c) Frequencies of CNV gain, loss, and non-CNV among CRGs. (d) Locations of CNV alterations in CRGs on 23 chromosomes. (e) Expression levels of 27 CRGs between normal and CRC tissues. Red gene names present CNV gains, blue gene names represent CNV loss, and black gene names represent CNV constant. CRGs: cuproptosis-related genes; CRC: colorectal cancer; TCGA: The Cancer Genome Atlas; CNV: copy number variant.

Figure 2

The clinicopathological and biological characteristics in distinct CRG subtypes. (a) The Interaction analysis among CRGs in CRC. The line of the CRGs represents their interaction and the line thickness indicates the strength of the correlation between genes. Blue and pink represent negative and positive correlations, respectively. (b) Consensus analysis defining three CRGs clusters (k = 3) and correlation area. (c) The PCA analysis among three CRG cluster subtypes. (d) The Kaplan–Meier survival analysis for CRC patients of three subtypes associated with DFS. (e) Heatmap analysis in clinicopathologic features and expression levels of CRGs among three distinct subtypes. CRG: cuproptosis-related genes; CRC: colorectal cancer; PCA: principal components analysis; DFS: disease-free survival.

Figure 3

Correlations of tumor immune microenvironment with three CRG subtypes. (a) GSVA of biological pathways among three distinct subtypes in five datasets, where red and blue represent activated and blue inhibited pathways, respectively. (b) Abundance of 22 infiltrating immune cell types in three CRG subtypes. (c) Expression levels of PD-1, CTLA4, and PD-L1 among three CRG subtypes. (d) Difference of TME score among three CRG subtypes. CRG: cuproptosis-related genes; GSVA: gene set variation analysis; TME: tumor microenvironment.

Figure 4

Identification of CRG-related gene subtypes. GO (a) and KEGG (b) enrichment analyses of DEGs between cuproptosis clusters A and C. (c) Consensus analysis defining two gene clusters (k = 2) and correlation area. (d) Heatmap analysis in clinicopathologic features and expression levels of CRGs between two distinct gene subtypes. (e) The Kaplan–Meier survival analysis for CRC patients of two gene subtypes associated with DFS. (f) Differential expression levels of 27 CRGs between two gene subtypes. GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; DEGs, differentially expressed genes; CRG, cuproptosis-related genes; DFS: disease-free survival.

Figure 5

Construction, validation, and evaluation of CRG-score model. (a) Alluvial diagram of subtype distributions in groups with different CRG-scores and DFS. (b) Differences in CRG-scores between cuproptosis clusters A and C. (c) Differences in CRG-scores between two gene subtypes. (d) Kaplan–Meier curves analysis for DFS of CRC patients between high-risk and low-risk groups in the training and testing sets. (e) ROC curves of the CRG-scores to demonstrate the sensitivity and specificity in predicting the DFS of CRC patients from training and testing sets. (f) Differences in TME score between two gene subtypes. (g) Ranked dot and scatter plots showing the correlation between CRG-score and 22 infiltrating immune cell types. (h) Correlations between the abundance of immune cells and eight genes from the CRG-score model.

Figure 6

Comprehensive analysis of the CRG-score in CRC. (a) and (b) Relationships between CRG-score and MSI. (c) The correlation of the CRG-score with TMB between two gene subtypes. (d) TMB levels between the high and low-risk groups. The waterfall plot of somatic mutation features established with high (e) and low (f) CRG-scores. Each column represented an individual patient and the number on the right indicated the mutation frequency in each gene. (g) Correlation between CRG-score and stem cell index. (h) Correlation between CRG-score and chemotherapeutic sensitivity. CRG: cuproptosis-related genes; TMB: Tumor mutation burden.