SEL1L3 as a link molecular between renal cell carcinoma and atherosclerosis based on bioinformatics analysis and experimental verification

Differential gene expression analysis

In total, 344 and 117 DEGs were found in GSE14762 and GSE53757, respectively, according to the pre-set cutoff values (Figure 1A, 1B). A total of 764 DEGs were screened from TCGA Kidney Renal Clear Cell Carcinoma (KIRC) dataset (Figure 1C). By overlapping the DEGs from the three datasets, 15 DEGs were identified (Figure 1D).

Figure 1. Identification of DEGs. (A) Volcano plot of GSE14762. (B) Volcano plot of GSE53757. (C) Volcano plot of TCGA KIRC dataset. (D) Venn plot of three datasets.

Functional enrichment analysis

GSEA

GSEA was used to conduct a comprehensive genome-wide enrichment analysis (GWEA). The prominent enriched pathways included DNA-binding transcription activator activity, transporter complex, morphogenesis of embryonic epithelium, stem cell proliferation, adrenal overactivity, DNA-binding transcription factor activity, abnormalities in adrenal physiology, and hypokalemia (Figure 2).

Figure 2. Functional enrichment analysis results from GSEA. (A, B) BP, (C) CC, (D, E) MF, (F–H) HP analysis by GSEA.

Metascape enrichment analysis

Metascape analysis revealed strong enrichment in the regulation of cell activation, inflammatory response, and positive regulation of immune response, immune effector processes, and tube morphogenesis (Figure 3).

Figure 3. Metascape enrichment analysis. (A) Heatmap of enriched terms across input differently expressed gene lists, colored by p-values. (B) Network of enriched terms colored by cluster identity. (C) Network of enriched terms colored by p-value.

PPI network

Cytoscape software was used to visualize the PPI network. The total numbers of DEGs in the three datasets (Figure 4A). The connections among the 15 common DEGs (Figure 4B). Among the DEGs, SEL1L3 was identified as the core gene (Figure 4B).

Figure 4. PPI network of DEGs and hub genes. (A) Total DEGs. (B) Protein interaction relationship by Cytoscape. (C) Protein interaction network colored by combine score.

Survival analysis

SEL1L3 expression was significantly higher in RCC samples than in normal samples (Figure 5A). Patients with lower SEL1L3 expression had poorer overall survival (log-rank p =0.028, Figure 5B).

Figure 5. (A) Expression box plot of SEL1L3 in TCGA KIRC dataset. (B) Survival curve of SEL1L3 in KIRC dataset. (C) Heatmap of DEGs.

Gene expression heatmap

A gene expression heatmap was used to visualize (Figure 5C). It was observed that the SEL1L3 gene exhibited high expression levels in RCC samples, whereas it demonstrated low expression levels in normal tissue samples. This observation suggests a potential regulatory involvement of SEL1L3 in these cancer types.

Immune infiltration analysis

Figure 6A illustrates the presence of immune cells in patients with Esca, categorized based on low and high expression levels of SEL1L3. The correlation between the immune cells was calculated simultaneously (Figure 6B). Differential immune cell infiltration analysis revealed that high SEL1L3 expression was significantly positively correlated with the infiltration of antibody-drug conjugates (aDC), B cells, eosinophils, interstitial dendritic cells (iDC), macrophages, neutrophils, T cells, T helper (Th) cells, central memory CD8 T cells (Tcm), T follicular helper (TFH), Th1 cells, Th17 cells, and regulatory T cells (Tregs), and negatively correlated with CD8 T cells, natural killer (NK) cells, plasmacytoid dendritic cells (pDC), and M0 macrophages (Figure 6C). A heatmap of the infiltration of immune cells in the low- and high-SEL1L3 expression groups (Figure 6D). The correlation between the expression of SEL1L3 and each individual immune cell type was calculated (Figure 7).

Figure 6. Immune infiltration analysis. (A) Fractions of immune cells in whole KIRC sample. (B) The correlational heatmap of 20 difference immune cells. (C) Violin plots of the distribution of difference immune cells. (D) Calorigram reflecting the distribution of the immune activity in esophageal cancer.

Figure 7. The relationship between the infiltration of immune cells and SEL1L3 was examined through the utilization of scatter plots. (A) aDC cells, (B) neutrophils cells, (C) eosinophils cells, (D) macrophages cells, (E) T helper cells, (F) Tcm cells, (G) T cells, (H) Th1 cells.

Role of SEL1L3 on ErbB/PI3K/mTOR signaling pathway

Western blotting revealed that SEL1L3 expression was upregulated in the RC group. The core proteins (including ErbB, PI3K, PIP3, AKT, and mTOR) involved in ErbB signaling were also upregulated in the RC group. By knocking down SEL1L3, core proteins of the ErbB/PI3K/mTOR signaling pathway were upregulated in RCC cells (Figure 8).

Figure 8. Experimental investigation into the impact of SEL1L3 on the ErbB/PI3K/mTOR signaling pathway.

Role of SEL1L3 on apoptosis, invasion, cell cycle, and inflammation

Downregulation of Caspase-1, Caspase-9, and FAS expression in the RC group suggested that apoptosis was inhibited (Figure 9). Knockdown of SEL1L3 also resulted in the downregulation of apoptosis-related genes, whereas overexpression of SEL1L3 led to their upregulation. Biomarkers of invasion (MMP3 and MMP9), cell cycle (c-Myc and CyclinD1) and inflammation (IL-6 and IL-8) were significantly upregulated in the RC group (Figure 10). SEL1L3 knockdown enhanced this effect.

Figure 9. Experimental investigation into the impact of SEL1L3 on cell apoptosis function.

Figure 10. Experimental investigation into the impact of SEL1L3 on invasion, cell cycle and inflammation function.

Data retrieval

We searched for an atherosclerotic plaque expression matrix in the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/). The atherosclerosis dataset GSE28829 based on the GPL570 platform contained 13 normal tissues and 16 atherosclerotic plaque tissues, and GSE43292 stockpiled on the GPL6244 platform consisted of 32 normal tissues and 32 atherosclerotic plaque tissues. The RNA sequencing data of RCC patients were downloaded from The Cancer Genome Atlas (TCGA) data portal (https://tcga-data.nci.nih.gov/tcga/), and included 72 normal and 535 tumorous tissues.

Differential expression analysis

Gene differential analysis was conducted in the three datasets using the “Limma” package in R 3.6.1 software. Statistical significance was set at P < 0.05 and absolute log fold change (FC) > 1. Venn plots were constructed to overlap common differential expressed genes in the three datasets.

Volcano plot of gene expression

Volcano plots were constructed using the “ggplot2” package in R software to visualize gene expression in the three datasets.

Protein-protein interaction (PPI) network

The Search Tool for the Retrieval of Interacting Genes (STRING) database (https://www.string-db.org) was used to construct a PPI network of the differentially expressed genes (DEGs). Visualization of this network was performed using Cytoscape software (https://cytoscape.org/).

Functional enrichment analysis

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed using the DAVID (https://david.ncifcrf.gov/) and Metascape (https://metascape.org) platforms. Functional enrichment analysis is a computational approach used to determine whether a specific collection of genes or proteins is enriched for specific biological functions, pathways, or processes. This analytical method has been widely employed in genomics and proteomics to obtain a deeper understanding of the molecular mechanisms underlying experimental findings.

Gene set enrichment analysis (GSEA)

GSEA was performed to evaluate enriched KEGG pathways using GSEA software (http://www.broadinstitute.org/gsea). We normalized the gene expression data and then filtered out the low-expressed genes to reduce noise.

Survival analysis

Based on the TCGA and Genotype-Tissue Expression (GTEx) data, the Gene Expression Profiling Interactive Analysis (GEPIA) database (http://gepia.cancer-pku.cn/) was used for survival analysis.

Heat map of gene expression

The R package heatmap facilitated the creation of a visually informative heatmap effectively depicting variations in the expression levels of core genes among RCC and normal tissue samples in the GSE14762 and GSE53757 datasets.

Immune infiltration analysis

The estimation of immune cell infiltration in RCC was conducted using the “CIBERSORT” R package, which uses a gene set containing specific markers of immune cells to calculate the enrichment score. Samples with a significance level of confidence and a cutoff of p < 0.05 were selected.

Verification of the role of SEL1L3

The HEK-293 and Caki-1 cell lines were obtained from the National Biomedical Experimental Cell Resource Bank in Beijing, China. Cells were divided into four groups: control (normal renal cancer cells), RCC (renal cancer cells), RCC/SEL1L3-KD (RCC with SEL1L3 knockdown), and RCC/SEL1L3-OV (RCC with SEL1L3 overexpression). Sample preparation involves the extraction and purification of the protein of interest from a sample, which may include processes such as homogenization, cell lysis, and chromatography, depending on the nature of the sample. For protein separation, the extracted proteins were separated based on molecular weight using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). For protein transfer, following separation, the proteins were transferred from the gel medium onto polyvinylidene fluoride (PVDF) membranes through electroblotting. This step entails the immobilization of proteins onto the membrane. To prevent non-specific binding of the primary antibody, the membrane is subjected to incubation with a blocking solution, typically consisting of 5% non-fat dry milk or 1% bovine serum albumin (BSA) in Tris-buffered saline (TBS) with 0.1% Tween-20 (TBST) for approximately 1 h at room temperature or overnight at 4° C. Subsequently, the membrane is incubated with a primary antibody that specifically targets the protein of interest, diluted in the aforementioned blocking solution, for approximately 1 h at room temperature or overnight at 4° C. After washing, the membrane was incubated with secondary antibodies, wherein an enzyme such as horseradish peroxidase (HRP) or alkaline phosphatase (AP) was conjugated to the secondary antibody. This conjugated enzyme recognizes and binds primary antibodies. Subsequently, the membrane was washed again to eliminate any unbound secondary antibody. For signal detection, the membrane was exposed to a substrate solution specific to an enzyme conjugated to a secondary antibody. This process yields a chemiluminescent or chromogenic signal that can be detected using an X-ray film or specialized imaging system. The signal intensity was directly proportional to the quantity of protein in the sample. Quantification of the signal intensity can be achieved by employing image analysis software, which enables the comparison of results across various samples or treatments.

Data availability statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.