The Integrated Transcriptome Bioinformatics Analysis Identifies Key Genes and Cellular Components for Spinal Cord Injury-Related Neuropathic Pain.

BACKGROUND

Spinal cord injury (SCI) is one of the most devastating diseases with a high incidence rate around the world. SCI-related neuropathic pain (NeP) is a common complication, whereas its pathomechanism is still unclear. The purpose of this study is to identify key genes and cellular components for SCI-related NeP by an integrated transcriptome bioinformatics analysis.

METHODS

The gene expression profile of 25 peripheral blood samples from chronic phase SCI patients (E-GEOD-69901) and 337 normal peripheral blood samples were downloaded from ArrayExpress and Genotype-Tissue Expression Portal (GTEx), respectively. A total of 3,368 normal peripheral blood mononuclear cells (PBMC) were download from Sequence Read Archive (SRA713577). Non-parametric tests were used to evaluate the association between all of differential expression genes (DEGs) and SCI-related NeP. CellPhoneDB algorithm was performed to identify the ligand-receptor interactions and their cellular localization among single PBMCs. Transcription factor (TF) enrichment analysis and Gene Set Variation Analysis (GSVA) were used to identify the potential upstream regulatory TFs and downstream signaling pathways, respectively. Co-expression analysis among significantly enriched TFs, key cellular communication genes and differentially expressed signaling pathways were performed to identify key genes and cellular components for SCI-related NeP.

RESULTS

A total of 2,314 genes were identified as DEGs between the experimental and the control group. Five proteins (ADRB2, LGALS9, PECAM1, HAVCR2, LRP1) were identified in the overlap of proteins in the significant ligand-receptor interactions of PBMCs and protein-protein interaction (PPI) network based on the DEGs. Only HAVCR2 was significantly associated with NeP ( = 0.005). Besides, the co-expression analysis revealed that TF YY1 had significantly co-expression pattern with cellular communication receptor HAVCR2 ( = -0.54, < 0.001) in NK cells while HAVCR2 was also co-expressed with mTOR signaling pathway ( = 0.57, < 0.001). The results of RT-qPCR and external dataset validation supported the signaling axis with the most significant co-expression patterns.

CONCLUSION

In peripheral blood of chronic SCI, HAVCR2 might act as a key receptor on the surface of NK cells and interact with ligand LGALS9 secreted by CD14 monocytes, inhibiting NK cells through mTOR signaling pathway and ultimately predicting the occurrence of SCI-related NeP. This hypothetical signaling axis may provide prognostic biomarkers and therapeutic targets for SCI-related NeP.