LONG NON-CODING RNAs CONTRIBUTING FOR CRLF2 OVEREXPRESSION IN ACUTE LYMPHOBLASTIC LEUKAEMIA

INTRODUCTION AND OBJECTIVE: CRLF2 overexpression (CRLF2-high) has been associated with unfavourable prognosis in B-cell acute lymphoblastic leukaemias (B-ALL). Occurrence of CRLF2 rearrangements (CRLF2-r) and CRLF2 F232C mutations account for only half of the cases with this gene overexpression. Long non-coding RNAs (lncRNAs) play a role in the development and progression of leukaemia and can interfere in the transcriptional regulation of protein-coding genes. In this scenario, we hypothesise that the dysregulation of lncRNAs might be a potential mechanism underlying CRLF2-high in B-ALL patients. MATERIAL AND METHODS: We included 126 B-ALL cases from the Therapeutically Applicable Research to Generate Effective Treatments (TARGET) cohort and delineated their molecular profile based on Whole Genome Sequencing (WGS) and RNA sequencing (RNA-seq) data. Differentially expressed (DE) lncRNAs in CRLF2-high patients were identified using DESeq2. Lncpath was used to obtain functional pathways influenced by the DE lncRNAs. The lncRNA and mRNA pairs experimentally validated were obtained from the LncRNA2Target database, and the interactions were tested by correlation analysis using RNA-seq TARGET data. Chromatin accessibility landscape was evaluated by Chromatin Immunoprecipitation Sequencing (ChIP-Seq) using H3K27ac and H3K4me1 marks, comparing B-ALL cell lines with CRLF2-high and low, to identify putative enhancers that could be linked to CRLF2 regulation. Peak regions and transcription factor (TF) motifs were visualised in IGV software. All analyses used the GRCh37 genome as reference. RESULTS AND CONCLUSION: Were identified 293 up- and 70 down-regulated lncRNAs in CRLF2-high context. Among them, we identified five potential interactions lncRNA-target previously reported and also positively correlated in B-ALL context: RPL34-AS1-MIR3663, LINC00161-MIR21, LINC00161-MIR590, PWRN1-MIR21 and PART1-MIR149. Interestingly, RPL34-AS1 and MIR21 were also correlated with CRLF2, showing a high expression in CRLF2-high patients (p=0.036 and p=0.004, respectively). Additionally, MIR3663, a known target of RPL34-AS1, was exclusively expressed in CRLF2-high patients. The characterisation of RPL34-AS1 region revealed enhancer peaks in all B-ALL cell lines evaluated, independently of CRLF2 status. Furthermore, we observed that this region is enriched with TFs motifs in common with CRLF2 promoter, including FOXM1, PML, TBP and YY1 sites. Additionally, we assessed the expression profile of these, but only YY1 was positively correlated with CRLF2 expression (p=0.050). Considering the functional role of ribonucleoproteins in gene regulation, we will validate the potential regulatory interactions between RPL34-AS1 and CRLF2 using small interfering RNA (siRNA) assays. Our findings indicate a likely mechanistic role for these lncRNAs interactions on leukemogenesis which could unravel novel biomarkers, and clarify how the expression of CRLF2 is regulated in those leukaemias.