We have demonstrated that the expression of the CALM/AF10 (C/A) fusion gene in murine BM cells results in an aggressive acute myeloid leukemia (AML). We sought to identify the domains involved in leukemogenesis mediated by the CALM/AF10 fusion gene. For this purpose we employed the CFU-S assay in which it was observed that C/A transduced BM cells generated an average of 75±23 day 12 CFU-S colonies per input 10^5 cells as compared to an average of 2 ±3 colonies with cells expressing the empty GFP vector (~ 37 fold increase; P < 0.0005) in the number of day-12 CFU-S content. Expression of the CALM gene truncated to the breakpoint of CALM/AF10 (CALMdelta3’) or the CALM/ AF10 fusion gene with a deleted octapeptide motif - leucine zipper domain (CALM/AF10 delta OM-LZ) gave 10 (±11) and 12 (±11) day12 CFU-S respectively per input 10^5 bone marrow cells showing that the AF10 portion of C/A, especially the OM-LZ region is necessary for the observed enhancement of d-12 CFU-S. BM cells transduced with a construct harboring the CALM gene fused only to the OM-LZ domain of AF10 (CALM+ OM-LZ) showed 68(±5) d-12 CFUS per input 10^5 cells (~34 fold Vs MIG; P=0.0006) comparable to the C/A fusion gene.
We observed that the C/A fusion gene fails to show leukemic transformation of BM progenitors in CFC or proliferation assays in vitro in contrast to its striking leukemogenic potential in vivo. We tested different mutants of the C/A fusion gene using these assays and observed that the expression of a mutant of the CALM/AF10 fusion gene with a C-terminal portion of CALM (amino acids 400 – 648) fused to the OM-LZ motif of AF10, showed a significant increase in the number of secondary CFCs (32 fold Vs MIG; 15.38 fold Vs C/A), with the appearance of predominantly blast-like colonies. Interestingly, this mutant could also initiate leukemias in mice (n=5) with a latency and phenotype similar to mice injected with C/A transduced BM cells.
Taken together, we demonstrate that the OM-LZ domain of AF10 is necessary for the expansion of early hematopoietic progenitors by CALM/AF10 and also sufficient for in vivo leukemic transformation. We also demonstrate that the deletion of amino acids 1 to 410 of CALM confers in vitro transformation potential to the C/A fusion gene. Our data identify the domains of C/A that are crucial for leukemogenesis and provide insights into the mechanism of transformation in t(10;11) (p13;q14) positive leukemia.
Disclosures: No relevant conflicts of interest to declare.