A Comprehensive Bioinformatics Approach to Identify Molecular Signatures and Key Pathways for the Huntington Disease

Abstract

Huntington disease (HD) is a degenerative brain disease caused by the expansion of CAG (cytosine-adenine-guanine) repeats, which is inherited as a dominant trait and progressively worsens over time possessing threat. Although HD is monogenetic, the specific pathophysiology and biomarkers are yet unknown specifically, also, complex to diagnose at an early stage, and identification is restricted in accuracy and precision. This study combined bioinformatics analysis and network-based system biology approaches to discover the biomarker, pathways, and drug targets related to molecular mechanism of HD etiology. The gene expression profile data sets GSE64810 and GSE95343 were analyzed to predict the molecular markers in HD where 162 mutual differentially expressed genes (DEGs) were detected. Ten hub genes among them (DUSP1, NKX2-5, GLI1, KLF4, SCNN1B, NPHS1, SGK2, PITX2, S100A4, and MSX1) were identified from protein-protein interaction (PPI) network which were mostly expressed as down-regulated. Following that, transcription factors (TFs)-DEGs interactions (FOXC1, GATA2, etc), TF-microRNA (miRNA) interactions (hsa-miR-340, hsa-miR-34a, etc), protein-drug interactions, and disorders associated with DEGs were predicted. Furthermore, we used gene set enrichment analysis (GSEA) to emphasize relevant gene ontology terms (eg, TF activity, sequence-specific DNA binding) linked to DEGs in HD. Disease interactions revealed the diseases that are linked to HD, and the prospective small drug molecules like cytarabine and arsenite was predicted against HD. This study reveals molecular biomarkers at the RNA and protein levels that may be beneficial to improve the understanding of molecular mechanisms, early diagnosis, as well as prospective pharmacologic targets for designing beneficial HD treatment.

Keywords: Huntington disease; bioinformatics; biomarkers; hub genes; pathway; system biology.

Figure 1.

Systematical implementation approach in the study: significant differentially expressed genes were identified as well as associated gene ontology terminologies and pathways were enriched using mRNA expression data sets of Huntington disease from neural cells and brain tissue. Multiple network approaches were also implemented to identify PPI, regulatory signature molecules, and potential therapeutic drug targets. DEGs indicate differentially expressed genes; KEGG, Kyoto Encyclopedia of Genes and Genomes; PPI, protein-protein interaction; TF, transcription factor.

Figure 2.

(A) Overlapping 106 up-regulated common genes among data sets GSE64810 and GSE95343; (B) Overlapping 56 down-regulated common genes among data set GSE64810 and GSE95343; (C and D) heatmap of selected data sets, GSE64810 and GSE95343 displaying differential expression level; the color scale reflects the expression value, and each row and column represents a gene. DEGs indicates differentially expressed genes.

Figure 3.

(A) KEGG Pathway enriched with the relevant DEGs, P < .05: Network Analyst and the Enrichr web-based server identified KEGG pathways (P < .05) linked with substantial DEGs. (B) DEG-related pathways in KEGG pathway analysis: the color of the bubble figure denotes log10 P values, whereas the size of the bubbles reflects gene counts. Furthermore, the horizontal axis reflects log10 FDR. (C to E) The bar diagrams illustrate significant GO terms (P < .05) related with genes regarding biological process, cellular component, and molecular function categories, correspondingly. The vertical axis displays enhanced GO terms with respect to (P < .05) and horizontal axis displays −log10 (P values). DEGs indicates differentially expressed genes; ECM, extracellular matrix; FDR, false discovery rate; GO, Gene Ontology; IgA, immunoglobulin A; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Figure 4.

Visualization of PPI network of common DEG: (A) PPI network developed by Network Analyst showing the top 10 hub genes that interconnect other DEGs marked by blue nodes and other nodes symbolize DEGs connected by edges which reflect DEG interaction. (B) List of top 10 hub genes produced by Cytoscape through MCC method. Here, color shades represent the rank of hub proteins. The darker shades of color express the higher rank of hub genes and vice versa. (C) According to the selected layout, the top 10 hub proteins in a network are visualized using the MCC technique in Cytoscape v3.8.2 using the cytoHubba plugin. Pink-colored nodes represent the proteins associated via edges with the top 10 hub proteins in blue shade. DEG indicates differentially expressed gene; MCC, maximum clique centrality; PPI, protein-protein interaction.

Figure 5.

TF-gene and TF-miRNA coregulatory interconnections. (A) Identification of transcriptomic regulatory signatures by network analysis of DEGs using the Network Analyst server: the network highlights the top 9 TFs that are linked to DEGs. Blue nodes represent the TFs, other nodes represent the DEGs, and edges represent the interaction levels. (B) Network-based analysis of transcriptomic regulatory signatures by network analysis of DEGs using the Network Analyst server: the network displays the top 12 miRNAs interconnected with DEGs, with blue nodes representing miRNAs and other nodes representing DEGs linked through edges. DEGs indicates differentially expressed genes; TF, transcription factor.

Figure 6.

The main diseases linked to common differentially expressed genes are highlighted in the gene-disease association network (green: seeds and blue: diseases).

Figure 7.

Bar graph including top 10 related small molecules with high significant correlations according to P values.

Figure 8.

Graphical representation of the validation of relevant biomarkers.