The development of kinase inhibitors such as imatinib that block the Bcr/Abl tyrosine kinase has significantly improved chronic myeloid leukemia (CML) therapy. However, patients with advanced disease often develop resistance to imatinib due to the emergence of clones with point mutations in the tyrosine kinase domain. AMN107, a novel second generation inhibitor of Bcr/Abl (

Weisberg et al.,
Cancer Cell
,
7
:
129
,
2005
) is currently in Phase 2 clinical trials and shows significant clinical activity in some patients with imatinib-resistant CML. However, it is possible that resistance to AMN107 could occur through the emergence of new Bcr/Abl point mutations, and here we report the results of a random mutagenesis study to identify Bcr/Abl mutants selected for resistance to AMN107. A library of mutations was generated in the target gene by propagating a native BCR-ABL-GFP retroviral construct through a bacterial strain deficient in a DNA repair pathway. Murine Ba/F3 cells were then transfected/infected with the mutated vector and subsequently the cells were selected for the ability to proliferate in the presence of AMN107 (0.125–0.5 μM). The Ba/F3 cells expressing native Bcr-Abl did not grow under these conditions. Single cell clones were expanded and a total of 60 individual colonies were isolated for which BCR/ABL was sequenced. Twenty colonies had single point mutations located in the kinase domain of the BCR-ABL gene. The rest had multiple point mutations and were not considered for further analysis. The point mutants identified in this way were all validated by preparing the corresponding Bcr-Abl cDNA using site-directed mutagenesis, generating a new mutant-Bcr-Abl/Ba/F3 cell line and testing for resistance to both AMN107 and imatinib. The mutant cell lines confer varying degrees of resistance to AMN107, from 5- to 400-fold. The mutant variants identified in this study included 15 novel mutations and 5 known imatinib-resistant mutations that have previously been identified in CML patients. These latter mutations included, T315I, which similar to imatinib, showed maximum resistance against AMN107 (~50% survival at 10 μM concentration). Interestingly, the majority of novel AMN107-resitant mutants were also found to be resistant to imatinib and the rank order was highly correlated to the rank order of resistance to AMN107. These data may be helpful in providing insights into the mechanism of acquired resistance of Bcr-Abl to small molecule inhibitors and are likely to predict some of the resistance mutations that may be observed in the clinic.

Author notes

Corresponding author

Sign in via your Institution