Blood Journal
Leading the way in experimental and clinical research in hematology

BCR/ABL and chromosomal instability: debate resolved

  1. Martin Carroll

In this issue of Blood, Chakraborty and colleagues provide conclusive proof that primary chronic myeloid leukemia (CML) cells from patients in chronic phase demonstrate chromosomal instability after DNA damage, suggesting that BCR/ABL, among its myriad described functions, dysregulates nonhomologous end joining (NHEJ) starting the progression of cells toward CML blast crisis.1

The BCR/ABL oncogene has taught many lessons about blood cell biology to those of us who have tried to untangle its mysteries.2 CML is one of the myeloproliferative diseases, but it distinguishes itself from polycythemia vera (PV) and essential thromobocytosis (ET) in one distressing way: if left untreated, all patients with CML will progress to blast crisis with an almost certain fatal outcome. Chronic phase CML appears to be solely caused by the single oncogenic mutation, t(9;22), which codes for the BCR/ABL oncogene. BCR/ABL clearly stimulates hyperproliferation in the myeloid compartment leading to the phenotype of CML. However, it has long been debated whether BCR/ABL also induces chromosomal instability that leads to CML blast crisis. The article by Chakraborty et al uses a combination of thoughtful studies of primary cells with extensive cytogenetic characterization to demonstrate that CML cells clearly have an increase in chromosomal instability after DNA damage.1 Specifically, CML cells demonstrate an increased propensity to anaphase bridge formation as a result of dysregulation of NHEJ repair. Although this idea has been previously hypothesized, the extensive chromosomal analysis here provides the most robust demonstration of this hypothesis to date.

Studies of the effects of BCR/ABL on DNA damage and repair have taken on a life of their own. Bedi and colleagues first formally demonstrated that CML cells are resistant to both spontaneous and DNA damage induced apoptosis.3,4 Slupianek et al subsequently suggested that BCR/ABL inhibits apoptosis by disrupting the function of Rad51 and went on to demonstrate that BCR/ABL expression delays the repair of DNA double-strand breaks.5,6 The latter reference first suggested that BCR/ABL disrupts nonhomologous end joining. In the intervening years, a number of different mechanisms whereby BCR/ABL disrupts DNA damage repair have been proposed and the field has become highly controversial. The reasons for this controversy are inherent in the question being studied. CML is a disease with a life expectancy of 3 to 5 years. Blast crisis develops after several years and, although some chromosomal abnormalities are associated with CML blast crisis, it is rare to see true genomic instability. Rather, this is a disease with a modest DNA repair defect. In addition, quantitative analysis of DNA damage is not technically trivial. Physical inspection for chromosomal abnormalities is the most trusted assay, but highly laborious and time consuming. Other methodologies such as the comet assay or pulsed field gel electrophoresis are technically challenging and likely semiquantitative rather than truly quantitative. Given the combination of a subtle phenotype with challenging methodologies, a robust conclusion about the effects of BCR/ABL on DNA damage and repair has been difficult.

Chakraborty and colleagues appropriately apply the knowledge that CML CD34+ cells enter cell cycle more rapidly than normal CD34+ cells on arrival in a petri dish to normalize their studies to similar cell divisions, not similar time in culture. Because DNA damage is a function of DNA replication, this is appropriate. They irradiate cells and then analyze metaphase spreads by direct inspection and FISH analysis for close to one-third of the cells' DNA material in thousands of cells. This is an extraordinary amount of work that clearly demonstrates that primary CML cells have an increase in chromosomal abnormalities compared with normal cells. Importantly, they document that chromosomal abnormalities are still present after several cell divisions, that they occur in actively dividing cells and that they include anaphase bridges that are thought to occur as a consequence of a breakage-bridge-fusion mechanism. The authors go on to demonstrate in CML cell lines that the mechanism of altered response to DNA damage is likely dependent on nonhomologous end joining and, importantly, that even low doses of dasatinib, an ABL kinase inhibitor, decreases anaphase bridges after γ irradiation (IR).1 The latter result suggests that, consistent with ongoing follow-up of long-term studies of ABL kinase inhibitors, ABL kinase inhibition is likely to abrogate the DNA damage accumulation in CML stem cells.

What does a cytoplasmic tyrosine kinase have to do with DNA repair? The structure of the BCR/ABL oncogene is highly studied. The cell line system described here may provide a system that will allow for structure function studies to elucidate this mechanism in more detail. Among other questions, it would be interesting to know if recent results demonstrating that DNA lesions can be carried through mitosis apply to CML cells.7 BCR/ABL-expressing cells do have an increase in fragile site breaks after aphidicolin treatment (Burke BA and Carroll M, unpublished data, October 2010) that may provide a separate mechanism for introducing permanent DNA changes to CML cells. Finally, it is hoped that whole genome sequencing efforts of CML blast crisis will clarify to what extent the mutations acquired in myeloid blast crisis result from defects in NHEJ versus other mechanisms that have been proposed. The old master—BCR/ABL—continues its lessons and we remain enchanted by the stories still to be told.


  • Conflict-of-interest disclosure: The author declares no competing financial interests. ■