Abstract-Text
Deletion del(14)(q24q32) is a recurrent aberration in B cell malignancies, particular in chronic lymphocytic leukemia (CLL). However, neither the underlying mechanism nor the biological consequences are known. We identified and characterized the deletion del(14)(q24q32) in 70 B cell neoplasms (83% of them were CLL) using whole-genome and exome sequencing, copy number analysis, karyotyping, fluorescence in situ hybridization, Sanger sequencing, RNA sequencing (RNAseq), reverse transcription PCR (RT-PCR), functional studies and computational methods. The study comprises three cohorts of cases (cohort 1 n=637 CLLs, cohort 2 n=23 CLLs, and cohort 3 n=73 B-cell malignancies), in which the 14q deletion was evaluated with different methodological approaches.
The recurrent del(14)(q24q32) generates an IGH::ZFP36L1 fusion. The fusion prevalence in an unbiased cohort of 636 CLL was 1.1%. Deletion del(14)(q24q32) was detected as sole abnormality in 23% of the cases (13/57), and was associated with low genomic complexity (median of 2.5 CNA per case). Trisomy 12 and unmutated IGHV status were significantly enriched in these cases. Analyses of clinical data suggest that the CLL patients with IGH::ZFP36L1 fusion have reduced time-to-treatment compared to patients without that fusion. We determined the precise coordinates of the 14q deletion for 36 patients. The centromeric breakpoints were located mainly within ZFP36L1 intron 1 (28/36, 78%) and the telomeric breakpoints at IGH switch regions, suggesting antibody-related aberrant class switch recombination as a major mechanism generating this change. We identified a consensus recognition motif for translin around the breakpoints. RNAseq showed ZFP36L1 to be overexpressed in all analyzed del(14)(q24q32) cases of Cohort 1 (n = 4) whereas RT-qPCR analysis in nine cases of Cohort 3 did not show significant ZFP36L1 deregulation compared to CLL without 14q-aberraton. By RNAseq and RT-PCR, we detected different chimeric IGH::ZFP36L1 fusion transcripts predominantly predicted to encode a truncated ZFP36L1 protein of 316-385 amino acids. As ZFP36L1 is an RNA binding protein which mediates mRNA decay, we investigated the ability of different truncated ZFP36L1-variants (representing the longest and shortest variants identified) to mediate mRNA decay by luciferase assays. The results indicate that the fusion transcripts do not have a tumorigenic potential. Furthermore, there was downregulation of 96% (172/180) of genes mapping to the deleted 14q fragment, among them TRAF3 and DICER1.
Our data indicate that the biological consequence of deletion del(14)(q24q32) with IGH::ZFP36L1 fusion most likely is the haploinsufficiency of several deleted genes rather than abnormal gene activation of ZFP36L1 or a dominant negative effect as consequence of the fusion. Clinically, CLL patients harboring IGH::ZFP36L1 fusion may represent an unrecognized high-risk subset of patients.