Differential mRNA and long noncoding RNA expression profiles in pediatric B-cell acute lymphoblastic leukemia patients.

Background: Long non-coding RNAs (lncRNAs) are transcripts longer than 200 nucleotides (nt) that are involved in the pathogenesis and development of various cancers including B cell acute lymphoblastic leukemia (B-ALL). To determine the potential roles of lncRNAs involved in pathogenesis of B-ALL, we analyzed the expression profile of lncRNAs and mRNAs in B-ALL, respectively, and constructed lncRNAs/mRNAs interaction network.

Methods: We performed RNA sequencing of 10 non-leukemic blood disease donors and 10 B-ALL patients for Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis. Interactions among mRNAs were predicted using the STRING database. Quantitative real time PCR (qRT-PCR) was performed to verify the RNA-seq data of lncRNAs and mRNAs. Potential functions of subtype-specific lncRNAs were determined by using coexpression-based analysis on distally (trans-pattern) located protein-coding genes.

Results: A total of 1813 differentially expressed transcripts (DETs) and 2203 lncRNAs were identified. Moreover, 10 dysregulated lncRNAs and 10 mRNAs were randomly selected, and further assessed by RT-qPCR in vitro. Go and KEGG analysis demonstrated that the differentially expressed mRNAs were most closely associated with myeloid leukocyte activation and in transcriptional misregulation in cancer, respectively. In addition, co-expression analysis demonstrated that these lncRNAs, including MSTRG.27994.3, MSTRG.21740.1, ENST00000456341, MSTRG.14224.1 and MSTRG.20153.1, may mediate the pathogenesis and development of B-ALL via lncRNA-mRNA network interactions.

Conclusions: These results showed that several mRNAs and lncRNAs are aberrantly expressed in the bone marrow of B-ALL patients and play potential roles in B-ALL development, and be useful for diagnostic and/or prognostic purposes in pediatric B-ALL.

Data Availability: The datasets used during our study are available through HARVARD Dataverse Persistent ID doi: https://doi.org/10.7910/DVN/LK9T4Z .