Research on the functions and potential mechanisms of STAT3 in chronic myelogenous leukemia

Objective: To explore the bioinformatics characteristics and potential mechanisms of signal transducer and activator of transcription (STAT3) in chronic myelogenous leukemia (CML).

Methods: Through the cancerSEA and CCLE databases, the expression of STAT3 in CML was verified and analyzed. Subsequently, K562 cells were treated with the STAT3 Inhibitor Stattic. Western blotting, cell counting, and flow cytometry were utilized to observe its impact on the functions of K562 cells. Then, Gene Set Enrichment Analysis (GSEA) and Gene Set Variation Analysis (GSVA) were applied to deeply explore the regulatory mechanism of STAT3. The "LIMMA" software package was used to calculate STAT3-related differentially expressed genes (DEGs). Machine-earning methods were utilized to screen the STAT3-related hub genes. The "pROC" software package was employed to perform Receiver Operating Characteristic (ROC) curve analysis on the hub genes. The "corrplot" software package was used to conduct a correlation analysis of the hub genes. The "RMS" software package was applied to construct a nomogram of the hub genes. Based on the DisGENET database, a disease network of the hub genes was constructed, and the DGIdb database was used to construct a drug network of the hub genes.

Results: In CML, the expression of STAT3 is upregulated compared to housekeeping genes. Among the 14 cell lines related to CML, STAT3 has the highest expression level in K562 cells. Stattic at a concentration of 5 μM can inhibit the proliferation of K562 cells, promote their Apoptosis, and block the cell cycle at the S phase (P < 0.05). GSEA and GSVA indicates that amino acid metabolism, NOD-like Receptor of STAT3. LASSO and SVM-RFE show that NCF4, PLAS1, IL7R, and TAGLN2 are hub differentially expressed genes (DEGs) related to STAT3. ROC and Nomogram indicate that the hub DEGs have high clinical diagnostic value. Correlation analysis shows that PLAS1 and NCF4 are negatively correlated, while PLAS1 and TAGLN2 are positively correlated. The construction of gene-disease networks reveals that these genes not only participate in the occurrence and development of CML but also jointly participate in multiple disease processes. The gene-drug network obtained 38 drugs targeting genes.

Conclusion: STAT3 might serve as a potential target for the treatment of CML. In CML, NCF4, PLAS1, IL7R, and TAGLN2 are hub genes associated with STAT3. These findings offer a fundamental theory for comprehending the pathogenesis of CML.

STAT3; Bioinformatics; Cell detection; Chronic myeloid leukemia; Machine learning method.