Mapping integral cell-type-specific interferon-induced gene regulatory networks (GRNs) involved in systemic lupus erythematosus using systems and computational analysis

Abstract

Systemic lupus erythematosus (SLE) is a systemic autoimmune disorder characterized by the production of autoantibodies, resulting in inflammation and organ damage. Although extensive research has been conducted on SLE pathogenesis, a comprehensive understanding of its molecular landscape in different cell types has not been achieved. This study uncovers the molecular mechanisms of the disease by thoroughly examining gene regulatory networks within neutrophils, dendritic cells, T cells, and B cells. Firstly, we identified genes and ncRNAs with differential expression in SLE patients compared to controls for different cell types. Furthermore, the derived differentially expressed genes were curated based on immune functions using functional enrichment analysis to create a protein-protein interaction network. Topological network analysis of the formed genes revealed key hub genes associated with each of the cell types. To understand the regulatory mechanism through which these hub genes function in the diseased state, their associations with transcription factors, and non-coding RNAs in different immune cell types were investigated through correlation analysis and regression models. Finally, by integrating these findings, distinct gene regulatory networks were constructed, which provide a novel perspective on the molecular, cellular, and immunological landscapes of SLE. Importantly, we reveal the crucial role of IRF3, IRF7, and STAT1 in neutrophils, dendritic cells, and T cells, where their aberrant upregulation in disease states might enhance the production of type I IFN. Furthermore, we found MYB to be a crucial regulator that might activate T cells toward autoimmune responses in SLE. Similarly, in B-cell lymphocytes, we found FOXO1 to be a key player in autophagy and chemokine regulation. These findings were also validated using single-cell RNASeq analysis using an independent dataset. Genotype variations of these genes were also explored using the GWAS central database to ensure their targetability. These findings indicate that IRF3, IRF7, STAT1, MYB, and FOXO1 are promising targets for therapeutic interventions for SLE.

Keywords: Gene regulatory networks; Interferons; Non-coding RNAs; Systemic lupus erythematosus; Targets.

Graphical abstract

Fig. 1

Schematic representation of the research workflow.

Fig. 2

Differentially expressed genes and non-coding RNAs in different immune cells. Volcano plots of upregulated and downregulated genes in (A) neutrophils; (B) dendritic cells; (C) T cells; and (D) B cells; Green points depict thedifferentially expressed genes with log2FC value greater than 1 and less than -1, red points depict the significantly differentially expressed genes with log2FC value greater than 1 and less than -1 and p values less than 0.05 and the blue points depict the genes that are not significantly differentially expressed. Venn diagrams of (E) differentially expressed genes from the four tissues and, (F) differentially expressed non-coding RNA from the four cell/tissue.

Fig. 3

Protein–protein interaction (PPI) Network and Hub Genes. The figure illustrates the PPI networks for immune cell types, highlighting the key hub genes involved in molecular interactions: (A) Neutrophils; (B) Dendritic Cells; (C) T cells; and (D) B cells.

Fig. 4

Transcription Factor (TF)-miRNA Coregulatory Network The figure displays the TF-miRNA coregulatory networks for different immune cell types: (A) Neutrophils; (B) Dendritic cells; (C) T cells; and (D) B cells. In each network, red circles denote the different hub genes of the cells, pink diamonds denote the transcription factors and the green squares denote the ncRNAs.

Fig. 5

Heatmap depicting correlation between hub genes and regulators The figure shows correlation plots for hub genes, key transcription factors, and non-coding RNAs (ncRNAs) in immune cells: (A) Neutrophils; (B) Dendritic cells; (C) T cells; and (D) B cells.

Fig. 6

Multivariable Regression Analysis of Key Regulators The figure presents AV scatter plots that demonstrate the performance of multivariable regression models for critical regulatory genes. The plots illustrate: (A) IRF3 (R² = 0.9250461); (B) IRF7 (R² = 0.8532618); (C) STAT1 (R² = 0.9041221); (D) MYB (R² = 0.5947396); (E) FOXO1 (R² = 0.9179705).

Fig. 7

Proposed Gene Regulatory Networks Across Distinct Immune Cell Types The figure illustrates unique gene regulatory networks, constructed based on hub genes, key transcription factors, and ncRNAs for (A) Neutrophils; (B) Dendritic cells; (C) T cells; and (D) Activated naïve B cells (E) Antigen-secreting cells. The upward arrow (↑) indicates that the respective genes or ncRNAs are upregulated. PubMed identifiers (PMIDs) corresponding to the references used are provided. Detailed summary is mentioned in the Supplementary Document 1.

Fig. 8

A) Disease-Gene Network Representation. The figure illustrates the disease-gene network constructed to identify associations between genetic variants and disease pathways. Nodes represent genes or diseases, with edges signifying known or predicted interactions. B) GWAS Gene Pathways Identified Using ClueGO in Cytoscape. The figure depicts enriched biological pathways derived from GWAS-identified genes, analyzed using the ClueGO plugin in Cytoscape.

Fig. 9

Identification of the regulators in different samples UMAP representation of sc-RNASeq data of different samples, colour coded by clusters (A) PBMC samples, (B) Dermal samples, and (C) Epidermal samples. Cell proposition plot for (D) PBMC samples, (E) Dermal samples, and (F) Epidermal samples. Volcano Plots depicting sc-RNASeq analysis of controls and SLE samples for different regulators in PBMC dataset (G) IRF7 (H) IRF3 (I) MYB (J) STAT1 and (K) FOXO1.

Fig. 10

Transcription factors identified from scRNA-seq analysis SLE and control samples across epidermal and dermal immune cell subsets. (A–D) Violin plots representing the expression of IRF3, IRF7, FOXO1, and STAT1 in epidermal immune cell subsets, highlighting significant differential expression between SLE (pink) and control (blue) samples. (E–G) Violin plots illustrating expression patterns of STAT1, IRF7, and MYB in dermal immune cell subsets.