CHARACTERIZATION OF THE NOVEL ABL1 FUSION TO THE PUTATIVE TUMOR SUPPRESSOR GENE SHIP1 IN ALL
Authors
P Kakadia,1 B Tizazu,2 G Mellert,1 K Spiekermann,1 S Bohlander2
Abstract
Background. BCR/ABL is the most common fusion gene in leukemia. The BCR/ABL fusion is the hallmark of CML and is frequently associated with B-lineage ALL. Up to now only 5 other mostly rare fusion partners of ABL1 have been described: ETV6, RCSD1, EML1, NUP214 and ZMIZ1. Here, we describe a novel fusion of ABL1 to SHIP1 (Inositol Polyphosphate-5 Phosphatase, INPP5D) in an 18-year-old woman with c-ALL. Aim. We aimed to characterize the SHIP1-ABL1 fusion at the genomic, molecular and functional level. Methods. Fluorescence in situ hybridization (FISH) analysis was performed using the BCR/ABL1 and SHIP1/ABL1 dual color dual fusion (DCDF) probes. Coimmunoprecipitations were performed by co-expressing differentially tagged versions of the SHIP1/ABL1 protein. The dimerization domain of SHIP1 was mapped using the Yeast-two hybrid (Y2H) system. Ba/F3 cells were stably transduced with a retroviral vector expressing SHIP1/ABL1 or various deletion mutants of SHIP1/ABL1. Results. Sequence analysis of an unexpected PCR product obtained from the diagnostic screening for a BCR-ABL1 fusion from a patient with c-ALL, revealed an in-frame fusion of the first 343 amino acids of SHIP1 to the second exon of the ABL1. The SHIP1/ABL1 fusion protein contains an SH2 domain in the SHIP1 portion (amino acids 2 to 102) and the ABL1 protein starting from exon 2. As the SHIP1 gene is located on 2q37 and is transcribed centromere to telomere like the ABL1 gene on 9q34, we predicted the SHIP1/ABL1 fusion to be on chromosome 2. Interestingly, FISH analyses using a BCR/ABL1 DCDF probe on metaphase chromosomes did not show an ABL1 signal on chromosome 2 but detected four ABL1 signals in interphase nuclei. Additional analyses with various combinations of SHIP1 and ABL1 probes revealed the presence of two SHIP1/ABL1 and one ABL1/SHIP1 fusion in the interphase nuclei from the patients. By using an anti-FLAG antibody an HA-SHIP1/ABL1 protein coimmunoprecipitated with a FLAG-SHIP1/ABL1 protein. This suggested that the SHIP1/ABL1 fusion is using a similar dimerization-dependent kinase activation mechanism like the other ABL1 fusion proteins. SHIP1/ABL1 transduced Ba/F3 cells exhibited IL3 independent growth, which was sensitive to Imatinib. Moreover, we could show that the expression of the SHIP1/ABL1 fusion protein in Ba/F3 cells leads to strong tyrosine phosphorylation of a number of proteins including STAT5a, MAPK, CRKL, and SHIP1/ABL1 itself. Using the Y2H system we were able to show that amino acids 100 to 343 of SHIP1 are necessary and sufficient for dimerization and for the induction of factor independent growth of Ba/F3 cells. Conclusions. Since SHIP1 functions as a negative regulator of myeloid proliferation and SHIP1 knock-out mice develop a myeloproliferative syndrome-like disease, it is tempting to speculate that in addition to the activation of the ABL1 tyrosine kinase, the formation of the SHIP1/ABL1 fusion might contribute to cellular transformation by compromising the putative tumor suppressor function of SHIP1 through: 1) haploinsufficiency and 2) a dominant negative effect mediated by the interaction between SHIP1/ABL1 and SHIP1. This is the first report describing genetic lesions of SHIP1 in ALL.