Transcriptomic Profiling of Tumor-Infiltrating CD4+TIM-3+ T Cells Reveals Their Suppressive, Exhausted, and Metastatic Characteristics in Colorectal Cancer Patients

Abstract

T cell immunoglobulin mucin-3 (TIM-3) is an immune checkpoint identified as one of the key players in regulating T-cell responses. Studies have shown that TIM-3 is upregulated in the tumor microenvironment (TME). However, the precise role of TIM-3 in colorectal cancer (CRC) TME is yet to be elucidated. We performed phenotypic and molecular characterization of TIM-3⁺ T cells in the TME and circulation of CRC patients by analyzing tumor tissues (TT, TILs), normal tissues (NT, NILs), and peripheral blood mononuclear cells (PBMC). TIM-3 was upregulated on both CD4⁺ and CD3⁺CD4^- (CD8⁺) TILs. CD4⁺TIM-3⁺ TILs expressed higher levels of T regulatory cell (Tregs)-signature genes, including FoxP3 and Helios, compared with their TIM-3^- counterparts. Transcriptomic and ingenuity pathway analyses showed that TIM-3 potentially activates inflammatory and tumor metastatic pathways. Moreover, NF-κB-mediated transcription factors were upregulated in CD4⁺TIM-3⁺ TILs, which could favor proliferation/invasion and induce inflammatory and T-cell exhaustion pathways. In addition, we found that CD4⁺TIM-3⁺ TILs potentially support tumor invasion and metastasis, compared with conventional CD4⁺CD25⁺ Tregs in the CRC TME. However, functional studies are warranted to support these findings. In conclusion, this study discloses some of the functional pathways of TIM-3⁺ TILs, which could improve their targeting in more specific therapeutic approaches in CRC patients.

Keywords: T cell immunoglobulin mucin-3; colorectal cancer; exhausted T cells; metastasis; tumor microenvironment.

Figure 1

Comparison of T-cell immunoglobulin and mucin domain containing 3 (TIM-3⁺) T cells in peripheral blood mononuclear cells (PBMC), normal colon tissues (NILs), and tumor-infiltrating lymphocytes (TILs) of colorectal cancer (CRC) patients. Percentage and mean fluorescence intensity (MFI) of TIM-3⁺ T cells was analyzed by flow cytometry. Representative flow cytometric plots and scatter plots showing TIM-3 expression in PBMC, NILs, and TILs on CD3⁺ (A), CD3⁺CD4⁺ (B), and CD3⁺CD4⁻ (CD8⁺) T cells (C). Scatter plots show comparison of the percentage and MFI of TIM-3⁺ cells on CD3⁺CD4⁺ and CD3⁺CD4⁻ (CD8⁺) T cells in PBMC, NILs, and TILs (D). The p values are represented as follows; *** p <0.001, ** p < 0.01, * p < 0.05.

Figure 2

Analysis of the different surface markers expressed on TIM-3⁺ and TIM-3⁻ CD4⁺ TILs. Representative flow cytometric plots and scatter plots gated on CD4⁺ T cells show the percentage of CD25 (A), FoxP3 (B), Helios (C), PD-1 (D), CTLA-4 (E), and LAG-3 (F) expression in CD4⁺TIM-3⁺ and CD4⁺TIM-3⁻ TILs. Representative flow cytometric plots show FoxP3/Helios, and CTLA-4/PD-1 co-expression on CD4⁺TIM-3⁺ and CD4⁺TIM-3⁻ TILs (G). Bar plots show the percentage of FoxP3^+/−Helios^+/− (H) and PD-1^+/−CTLA-4^+/− (I) expression in CD4⁺TIM-3⁻ (T⁻) and CD4⁺TIM-3⁺ (T⁺) TILs. The p values are represented as follows; *** p <0.001, ** p < 0.01, * p < 0.05.

Figure 3

Differential gene expression of CD4⁺TIM-3⁺ and CD4⁺TIM-3⁻ TILs. Hierarchical clustering of three TIM-3⁺ and three TIM-3⁻ libraries (from patients CRC07, 08, and 16) on differentially expressed RNA transcripts from RNA-Sequencing data. Each column represents a sample and each row represents a transcript. Expression level of each gene in a single sample is depicted according to color scale (A). PCA plot shows the clustering of RNA transcriptome from of three TIM-3⁺ and TIM-3⁻ libraries (B). Volcano plot summarizes the expression rate (log2 Fold Change) on the x-axis and the statistical significance (negative log₁₀-transformed p values) on the y-axis. The fold changes with significant p value (>0.05) of upregulated and downregulated genes are highlighted in red and green, respectively (C). Tree map (hierarchical heat map) depicting affected functional categories based on upregulated transcripts, immune cell trafficking (D), inflammatory response (E), and cellular growth and proliferation (F). Each individual colored rectangle is a particular biological function and the color range indicates its predicted activation state: increasing (orange) or decreasing (blue). Darker colors indicate absolute Z-scores. In this default view, the size of the rectangle is correlated with increased overlap significance. Top significantly upregulated migration and metastasis-related genes in TIM-3⁺ T cells (Z-score of >1.45) based on ingenuity pathways analysis (IPA) analyses are shown as a pie chart (G).

Figure 4

Transcriptomic profile of CD4⁺TIM-3⁺ and CD4⁺TIM-3⁻ TILs categorized by their functional characteristics. Top significantly affected (>−2 Z score < 2) canonical pathways based on IPA analysis. The horizontal bars denote the different pathways based on the Z-scores (A). Heat maps show the fold changes relative to the mean expression of DNA repair, apoptosis, and cell proliferation (B), Treg, IC, and NF-κB-related genes (C), DNA replication and cell cycle-related genes (D). Scatter plots show gene expression analyses by RNA-Sequencing in TIM-3⁺ and TIM-3⁻ T cells. X and Y axes represent TPM (Transcripts Per Million) of TIM-3⁺ and TIM-3⁻ T cells (E). Box plots show the Z-score of genes involved in tumor progression pathways of TIM-3⁺ T cells based on IPA analysis (F). Heat maps show the fold changes relative to the mean expression of diabetic-related genes (G).

Figure 5

Transcriptomic profiles of TIM-3⁺ and CD25⁺ T cells. Hierarchical clustering of two TIM-3⁺ and two CD25⁺ libraries (patient IDs: CRC07 and 08) on differentially expressed RNA transcripts from RNA-Sequencing data. Each column represents a sample and each row represents a transcript. Expression level of each gene in a single sample is depicted according to color scale (A). Heat maps show the fold changes relative to the mean expression of calcium-calmodulin, vesicle transport, signal transduction, cell migration, and metastasis (B), Treg-related receptors and signaling and enzymatic activity genes (C), cell adhesion, proliferation, and differentiation (D). Transcription factor and cytokine/suppression activity (E) related genes and some of the selected genes upregulated in CD25⁺ T cells (F). The overall difference in the number of genes related to Tregs are shown as a bar plot (G). Top significantly upregulated migration and metastasis-related genes in TIM-3⁺, compared with CD25⁺ T cells (Z-score of >1.5) based on IPA analysis are shown as a pie chart (H).

Figure 6

Schematic diagram summarizes genes/functions that are upregulated or downregulated in CD4⁺TIM-3⁺ T cells in the colorectal tumor microenvironment. In the colorectal TME, CD4⁺TIM-3⁺ cells co-express genes including FoxP3, Helios, PD-1, and CTLA-4 to suppress anti-tumor immune responses. Our study showed that DNA replication and cell cycle genes were downregulated, while genes related to immune suppression, cell proliferation, T-cell exhaustion, carcinogenesis, migration, and metastasis were upregulated in CD4⁺TIM-3⁺, compared with CD4⁺TIM-3⁻. Individual or combinational antibody therapies using anti-TIM-3, anti-PD-1, and anti-CTLA4 may target highly immunosuppressive TIM-3⁺ TILs to elicit anti-tumor responses in CRC.