Identification of key regulatory genes connected to NF-κB family of proteins in visceral adipose tissues using gene expression and weighted protein interaction network

Abstract

Obesity is connected to the activation of chronic inflammatory pathways in both adipocytes and macrophages located in adipose tissues. The nuclear factor (NF)-κB is a central molecule involved in inflammatory pathways linked to the pathology of different complex metabolic disorders. Investigating the gene expression data in the adipose tissue would potentially unravel disease relevant gene interactions. The present study is aimed at creating a signature molecular network and at prioritizing the potential biomarkers interacting with NF-κB family of proteins in obesity using system biology approaches. The dataset GSE88837 associated with obesity was downloaded from Gene Expression Omnibus (GEO) database. Statistical analysis represented the differential expression of a total of 2650 genes in adipose tissues (p = <0.05). Using concepts like correlation, semantic similarity, and theoretical graph parameters we narrowed down genes to a network of 23 genes strongly connected with NF-κB family with higher significance. Functional enrichment analysis revealed 21 of 23 target genes of NF-κB were found to have a critical role in the pathophysiology of obesity. Interestingly, GEM and PPP1R13L were predicted as novel genes which may act as potential target or biomarkers of obesity as they occur with other 21 target genes with known obesity relationship. Our study concludes that NF-κB and prioritized target genes regulate the inflammation in adipose tissues through several molecular signaling pathways like NF-κB, PI3K-Akt, glucocorticoid receptor regulatory network, angiogenesis and cytokine pathways. This integrated system biology approaches can be applied for elucidating functional protein interaction networks of NF-κB protein family in different complex diseases. Our integrative and network-based approach for finding therapeutic targets in genomic data could accelerate the identification of novel drug targets for obesity.

Fig 1. The overall work design of our pipeline.

Present research analysis for exploring candidate genes from expression profiles.

Fig 2. Gene expression data before and after normalization.

The horizontal axis represents the samples, and the vertical axis represents the gene expression values.

Fig 3. Significant protein interaction map.

S^PIN, developed from HIN^NF and their first level interacting partners.

Fig 4. Representation of gene-gene correlation plot and semantic similarity graph.

The correlation plots illustrate significant variations in gene expression among the gene-gene pairs in the control and disease samples. A). Gene-gene correlation of normal samples (control), B). Gene-gene correlation of obese samples (disease), C). The graph depicts semantic similarity between all pairs of genes and the blue arrow represents gene pairs with higher functional similarity.

Fig 5. Expression pattern of the filtered genes from HIN^NF and primary functional partners contributing to total of 193 genes.

The gene expression pattern analysis clearly depicts variation in expression in disease and control samples.

Fig 6. Gene enrichment.

The overall view of gene set enrichment analysis on filtered genes.

Fig 7. Top 20 terms of gene set enrichment analysis for the pathways, disease, molecular function and biological process.

Genes enriched are more closed to inflammatory diseases and pathways.

Fig 8. The prioritized network developed from control samples (control network) and disease samples (disease network).

Control Network and Disease Network represent significant changes with strong altered connections of network connectivity from normal to obese state. The average connectivity of nodes in the control state is 15.42, and it has decreased to 14.75 in disease state depicts the overall loss in the interaction in obese condition.

Fig 9. PPI network of control and disease.

Genes connecting to the family of NF-κB proteins where pink nodes represent the NF-κB protein family.

Fig 10. The expression and interaction of genes in adipose tissue using GIANT analysis.

A) The dense network formed from 68 genes and proteins of the NF-κB family. B) The decomposed network based on the edge weight of 0.4 and above.

Fig 11. Genes connected to NF-κB protein family with their characteristics.

A) Genes with their number of functional partners in obese and normal conditions, B) The fold change of these connected genes to the family of NF-κB proteins.

Fig 12. Pictorial representation of the filtering criteria used in the approach to identify biologically relevant functional nodes connected NF-κB proteins, obesity and related syndrome.

A total of 2650 genes from PPIM was narrowed down to 21 target genes of NF-κB proteins associated with obesity using the filtering criteria centered on biological insights.

Fig 13. The pathway enrichment map.

Thepotential target proteins of NF-κB protein family.