Abstract
Chronic myeloid leukemia (CML) in chronic phase has a disease-free survival of 87% (DFS; 5 years) and long-term treatment with Imatinib is effective. In a large subgroup of patients with CML, however, the disease ultimately progresses into B lymphoid blast crisis (LBC) with only 6% DFS and resistance to Imatinib develops in virtually all cases. In most cases, acquired resistance to Imatinib can be attributed to somatic mutations within the BCR-ABL1 kinase domain. Whereas BCR-ABL1 kinase mutations are rare in chronic phase CML, such mutations are found in >80% of patients with B cell lineage LBC. Likewise, deletions of the ARF and INK4B genes are rare in chronic phase CML but found in ~50% of B cell lineage LBC.
In a search for a B cell lineage-specific mutation mechanism responsible for BCR-ABL1 kinase mutations, we tested the hypothesis that aberrant activation of somatic hypermutation may give rise to drug-resistance and progression of chronic phase CML into LBC. Somatic hypermutation drives affinity maturation of immunoglobulins expressed by germinal center B cells and requires the cytidine deaminase AID. Expression of AID depends on PAX5, a transcription factor that determines B cell lineage commitment of hematopoietic progenitor cells. The dependence of AID expression on PAX5 limits somatic hypermutation to the B cell lineage. Consistent with aberrant activation of somatic hypermutation in B cell lineage LBC, we found both PAX5 and AID expression at the mRNA and protein level in B lymphoid but not myeloid subclones from patient-derived blast crisis CML. However, AID protein levels in LBC clones were 5–10-fold lower than in germinal center B cells. To confirm lineage-specific activation of AID-expression in BCR-ABL1 driven leukemia, we isolated bone marrow from Aid-GFP reporter transgenic mice and transformed the bone marrow cells with BCR-ABL1 under either myeloid (IL3, IL6, SCF) or B lymphoid (IL7) culture conditions. The Aid-GFP reporter drives GFP expression under control of upstream and downstream regulatory elements of the Aid locus (Crouch et al., 2007). BCR-ABL1-induced Aid-expression was only observed under B lymphoid culture conditions and was very heterogeneous among the leukemia cell population: Only about 5–10% of CD19+ B lymphoid leukemia clones express Aid-GFP. In these cells, however, Aid mRNA levels are 240-fold higher than in Aid-GFP-negative cells and even 1.5-fold higher than in normal germinal center B cells. Consistent with these findings, we found aberrant somatic hypermutation of the IGHM, BCL6 and MYC loci as well as evidence of ongoing DNA single-strand breaks at the ARF and INK4B loci in B cell lineage LBC but not myeloid CML clones. Ectopic expression of AID in seven otherwise AID-negative CML cell lines cells leads to the acquisition of Imatinib-resistance and sequence analysis of the Imatinib-resistant clones revealed accumulation of mutations within the BCR-ABL1 kinase domain that cause Imatinib-resistance in patients (e.g. L248V, E225K, T315I). Aberrant expression of AID also caused Imatinib-resistance of CML cells in vivo: NOD/SCID mice were injected with CML cells that were either transduced with AID/GFP or GFP alone. Whereas more than the half of the mice injected with GFP+ CML cells were still alive after 170 days, all mice in the AID/GFP+ CML group died within 54 days after injection despite Imatinib-treatment. Forced expression of the B cell-specific transcription factor PAX5 in otherwise PAX5-negative CML cells resulted in a partial B lymphoid lineage conversion similar to LBC. Of note, ectopic expression of PAX5 also resulted in aberrant AID expression, subsequent acquisition of BCR-ABL1 kinase mutations and development of drug-resistance.
We conclude that B cell-specific activation of PAX5/AID-induced aberrant somatic hypermutation provides a genetic basis for the strikingly different outcome of myeloid lineage CML as compared to LBC.
Sphingosine kinase-1 and sphingosine 1-phosphate receptor 2 mediate Bcr-Abl1 stability and drug resistance by modulation of protein phosphatase 2A
