Identification of Potential Key Genes and Regulatory Markers in Essential Thrombocythemia Through Integrated Bioinformatics Analysis and Clinical Validation

Abstract

Introduction: Essential thrombocytosis (ET) is a group of myeloproliferative neoplasms characterized by abnormal proliferation of platelet and megakaryocytes. Research on potential key genes and novel regulatory markers in essential thrombocythemia (ET) is still limited.

Methods: Downloading array profiles from the Gene Expression Omnibus database, we identified the differentially expressed genes (DEGs) through comprehensive bioinformatic analysis. GO, and REACTOME pathway enrichment analysis was used to predict the potential functions of DEGs. Besides, constructing a protein-protein interaction (PPI) network through the STRING database, we validated the expression level of hub genes in an independent cohort of ET, and the transcription factors (TFs) were detected in the regulatory networks of TFs and DEGs. And the candidate drugs that are targeting hub genes were identified using the DGIdb database.

Results: We identified 63 overlap DEGs that included 21 common up-regulated and 42 common down-regulated genes from two datasets. Functional enrichment analysis shows that the DEGs are mainly enriched in the immune system and inflammatory processes. Through PPI network analysis, ACTB, PTPRC, ACTR2, FYB, STAT1, ETS1, IL7R, IKZF1, FGL2, and CTSS were selected as hub genes. Interestingly, we found that the dysregulated hub genes are also aberrantly expressed in a bone marrow cohort of ET. Moreover, we found that the expression of CTSS, FGL2, IKZF1, STAT1, FYB, ACTR2, PTPRC, and ACTB genes were significantly under-expressed in ET (P<0.05), which is consistent with our bioinformatics analysis. The ROC curve analysis also shows that these hub genes have good diagnostic value. Besides, we identified 4 TFs (SPI1, IRF4, SRF, and AR) as master transcriptional regulators that were associated with regulating the DEGs in ET. Cyclophosphamide, prednisone, fluorouracil, ruxolitinib, and lenalidomide were predicted as potential candidate drugs for the treatment of ET.

Discussion: These dysregulated genes and predicted key regulators had a significant relationship with the occurrence of ET with affecting the immune system and inflammation of the processes. Some of the immunomodulatory drugs have potential value by targeting ACTB, PTPRC, IL7R, and IKZF1 genes in the treatment of ET.

Keywords: bioinformatics analysis; candidate drugs; essential thrombocythemia; hub genes; regulatory markers.

Figure 1

Identification of DEGs in two datasets (GSE61629 and GSE124281). (A) Heatmap of the top 50 DEGs in GSE621629 and GSE124281 datasets, respectively. (B) Volcano plot of DEGs in GSE61629 and GSE124281 datasets. Red and blue plots represent genes with [logFC]>0.5 and P<0.05, and black plots represent genes with no significant difference. Furthermore, the green plots represent up-regulated and down-regulated genes with [logFC]>1 and P<0.05, and the labeled genes represent these genes with [logFC]>1.5 and P<0.05. (C) Venn diagram of commonly changed DEGs in the two datasets. (including 22 common up-regulated genes, 41 common-down-regulated genes, six genes have interacted between GSE124281 up-regulated DEGs and GSE61629 downregulated DEGs, four genes have interacted between GSE61629 up-regulated DEGs GS124281 downregulated DEGs). DEGs, differentially expressed genes. DEGs, differentially expressed genes.

Figure 2

GO term enrichment analysis of DEGs. (A) Biological process. (B) Cellular component. (C) Molecular function. DEGs, differentially expressed genes. GO, Gene Ontology.

Figure 3

Regulatory networks of the TFs and their targeted DEGs identified by iRegulon. (A) Regulatory network of the TFs and their targeted up-regulated genes. (B) Regulatory network of the TFs and their targeted down-regulated genes. (C) Regulatory network of the TFs and their targeted up and down-regulated genes. The green color octagon indicates TFs, the purple color oval indicates the DEGs regulated by TFs, and the blue color oval indicates that TFs do not restrict them. TFs, transcriptional regulators; DEGs, differentially expressed genes.

Figure 4

PPI network of DEGs in the STRING database. Based on the STRING online database, the DEGs PPI network was constructed containing 63 DEGs. The different colors in the figure indicate the connectivity of genes, and the deep color shows the genes with the highest connectivity in the PPI network. PPI, protein–protein interaction; DEGs, differentially expressed genes.

Figure 5

Clusters and hub genes identified in the PPI network. (A) Cluster 1 in the PPI network. The orange nodes represent the screened genes in cluster 1. (B) Cluster 2 in the PPI network. The orange nodes represent the screened genes in cluster 2. (C) Ten hub gene identification in a PPI network based on the MCC method. The dark (red) nodes show the genes with higher MCC scores in the PPI network. PPI, protein–protein interaction; MCC, maximal centrality clique.

Figure 6

Validated the expression of the identified hub genes in an independent cohort group. The top 10 hub genes were both lower expressed in ET samples from the GSE123732 database, and the seven genes as ACTB, PTPRC, ACTR2, FYB, STAT1, IL7R, and FGL2 with a significant difference between ET and control. *P<0.05, **P<0.01. ET, essential thrombocythemia.

Figure 7

Validated the expression of the identified hub genes in clinical samples. Eight of the hub genes (CTSS, FGL2, IKZF1, STAT1, FYB, ACTR2, PTPRC, and ACTB) were both lower expressed in ET samples, and two hub genes (IL7R and ETS1) were higher expressed in ET samples. ***P<0.001. ET, essential thrombocythemia.

Figure 8

ROC curves employed to assess the diagnostic value of the hub genes. The ROC curve of the top 10 hub genes both showed a significant sensitivity and specificity in two databases (all the genes with AUC≥0.5). This result indicating these genes may have good diagnostic value for patients with ET. ROC, receiver operating characteristic.