| Background: Acute myeloid leukemia(AML)is a hematological malignancy resulting from the malignant transformation of hematopoietic stem/progenitor cells,and its pathogenesis is still unclear.AML patients have a poor prognosis,with a five-year survival rate of less than 30 percent.Therefore,exploring new biomarkers is helpful for clinical diagnosis,prognostic assessment and personalized treatment guidance of AML patients.Alternative splicing(AS)of RNA is a basic biological process that forms protein diversity,but many uncharacterized AS events have been confirmed to be involved in the occurrence and development of AML.Splicing factor(SF)is a tool for regulating AS events,and its abnormal changes will affect the process of splicing regulation.It is of great significance to explore the relationship between SFs and clinical characteristics and biological processes of AML patients.Objective: This study focused on the classical heterogeneous nuclear ribonucleoprotein(hn RNP)family and the arginine-and serine-rich protein(SR)family of splicing factors,and explored them based on multi-omics Association with clinicopathological factors and biological behavior in AML.Methods: We analyzed the genetic variation landscape of 32 SFs in AML patient samples from the TCGA database,and performed unsupervised clustering of AML patients based on the expression of SFs using a consistent clustering algorithm;further compared the differences in prognosis,clinicopathological information,immune cell infiltration,immune function,and signaling pathway activity between different splicing regulatory patterns.We also used the Cancer Genome Project(CGP)database to predict the susceptibility of samples of different splicing regulatory patterns to commonly used therapeutic agents,and compared with other immunotherapy data to predict anti-PD-1 and anti-CTAL4 responses in AML patients.Furthermore,we constructed a prognostic risk score model to predict overall survival(OS)in AML patients.Finally,we experimentally analyzed the relationship between SRSF10 and the cancer-promoting phenotype of AML cells and the regulatory mechanism of the downstream target gene HDAC9.Results: Compared with normal samples,most SFs were upregulated in AML samples and correlated with poor prognosis.Among the four splicing regulation patterns,cluster D patients had the best prognosis,followed by clusters A and B,and cluster C was the worst.In cluster A,the expression of SFs is generally low,and the cancer-promoting signaling pathway is significantly activated;in cluster B,SFs are consistently highly expressed,and the genetic information expression-related pathways are highly enriched,showing an active proliferation phenotype;cluster C is accompanied by high enrichment scores of inflammatory immune and energy metabolism-related pathways,high infiltrating proportions of monocytes,neutrophils and M2 macrophages,and high expression levels of immune checkpoints.Drug prediction showed that cluster B was more sensitive to cytarabine,doxorubicin,and midostaurin than cluster A;Compared with cluster C,cytarabine and midostaurin are more suitable for cluster D patients.Immunotherapy predicted a higher response to anti-PD-1 therapy in cluster C patients.The risk score model we constructed can accurately predict the prognosis of AML patients,and patients with higher risk scores have significantly shorter OS.The up-regulated expression of SRSF10 was characterized in clinical samples,and SRSF10 promoted AML cell proliferation,inhibited apoptosis,and affected the splicing of full-length and truncated transcripts of the pro-oncogene HDAC9.Conclusion: The analysis of splicing regulatory patterns can help us better understand the differences in the tumor microenvironment of AML patients,and guide clinical medication and prognosis prediction;at the same time,SRSF10 can be used as a potential therapeutic target and biomarker for AML. |
PDF Full Text Request