Identification of CCL19 as a Novel Immune-Related Biomarker in Diabetic Nephropathy

FIGURE 1. Differential expression analysis of immune-related genes between DN and healthy samples. (A) PCA figures displaying interbatch differences removed after integrating GSE35487, GSE37463 and GSE47184. (B) PCA figures displaying batch differences for the DN and healthy samples after integrating GSE35487, GSE37463 and GSE47184. (C) Heatmap showing the DEIGs between the DN and healthy samples. (D–F) GO analyses of DEIGs in terms of biological process, molecular function and cellular components.

FIGURE 2. Evaluation of tissue-infiltrating immune and stromal cell populations. (A) Heatmap showing the abundance of tissue-infiltrated immune and stromal cell populations between DN and healthy samples. (B) PCA figure displaying the distribution of DN and healthy samples. (C) Scatter diagram of the vector basis projection showing the contribution degree of cell populations. The length of the arrow indicates the absolute contribution degree. (D) Boxplot showing the abundance of infiltrated immune cell populations in DN and healthy samples. The Wilcoxon rank-sum test was used to measure the differences between two groups. Ns, no significant difference; * p < 0.05; **** p < 0.0001.

FIGURE 3. Determination of soft-thresholding power in WGCNA. (A) Analysis of the scale-free fitting indices for various soft-thresholding powers (β). (B) Mean connectivity analysis of various soft-thresholding powers. (C) Histogram of the connection distribution when β = 12. (D) Checking the scale-free topology when β = 12. (E) Cluster dendrogram of the coexpression network modules is ordered by hierarchical clustering of genes based on the 1-TOX matrix. Different colors represent different modules.

FIGURE 4. Screening candidate immune-related DN biomarkers using WGCNA. (A) Correlation between the gene module and infiltrated immune cell population, which significantly differed between the DN and healthy samples. The correlation coefficient in each cell represented the correlation between the gene module and the infiltrated immune cell, which decreased in color from red to blue. The corresponding p value is also annotated. (B–E) Scatter plots depict the relationship between module membership (MM) and gene significance (GS) in the turquoise module. Genes with MM ≥ 0.5 and GS ≥ 0.5 are labeled with solid dots.

FIGURE 5. Identification of CCL19 as a hub immune gene in DN and verification in the GSE30529 and GSE104954 datasets. (A) Venn diagram of DEGs and immune-related gene list. (B) PPI network of the overlapping genes between WGCNA and DEIG analysis. The gradients from yellow to red of each circle represent the increasing degree of centrality. The blue circle represents the hub with the highest degree of centrality. (C) Verification of CCL19 in GSE30529. Left: Boxplot showing the expression of CCL19 between the DN and healthy samples in GSE30529. The Wilcoxon rank-sum test was used to measure the differences between two groups. Right: ROC curve of CCL19 expression in DN. (D) Verification of CCL19 in GSE104954. Left: Boxplot showing the expression of CCL19 between the DN and healthy samples in GSE104954. The Wilcoxon rank-sum test was used to measure the differences between two groups. Right: ROC curve of CCL19 expression in DN.

FIGURE 6. Validation of CCL19 using in vivo experiment. (A) mRNA level of CCL19 under high glucose stimulation was analyzed by qPCR. (B,C) Protein level of CCL19 under high glucose stimulation was analyzed by western blotting.