CLL is a heterogeneous disease, but the mechanisms responsible for this heterogeneity are not fully understood. Previous studies have shown that global tyrosine phosphorylation in response to B-cell receptor (BCR) engagement is more pronounced in cases with low IgVH somatic hypermutation and worse prognosis, indicating an involvement of protein tyrosine kinases in progression of the disease. c-Abl, a non-receptor protein tyrosine kinase, plays an important role in B cell development, differentiation and BCR-mediated signalling. Inhibition of c-Abl with a specific reagent STI-571 induces CLL- cell death in vitro, and the sensitivity of CLL clones to this inhibitor seems to be variable. Since expression of c-Abl in CLL cells also varies, we hypothesized that response to STI-571 may be related to the level of c-Abl expression which, in turn, may be critical for protection of some CLL cells from apoptosis and therefore be associated with the disease progression. Western blot analysis confirmed that expression of c-Abl in malignant cells varies among CLL cases with a 13-fold difference between the minimum and the maximum. On average, the expression in CLL cells was 7.3-flold higher than in normal B-cells (n=7). Analysis of c-abl mRNA by quantitative RT-PCR confirmed this over-expression, and also showed that CLL cells exclusively express the anti-apoptotic 1b isoform of this protein. c-Abl expression in CLL cells showed a negative correlation with the extent of IgVH mutation (n=41, r=−0.55, P=0.003), but no correlation with either p53 functional status or CD38 positivity. Furthermore, c-Abl levels in CLL cells showed a positive correlation with peripheral blood WBC counts at sampling (n=43, r=0.570, P<0.001), and cases with a high c-Abl level (above the mean) were more likely to be at Binet stage B or C compared to the group with lower c-Abl (18/24 cases vs 3/18 cases, P<0.001). Functional studies showed that CLL clones with high c-Abl were more sensitive to STI-571-induced killing than CLL clones with low c-Abl expression (n=26, r=0.571, P=0.002). STI-571-induced killing of CLL-cells was associated with induction of apoptosis as indicated by mitochondrial depolarization (n=8) and by PARP cleavage (n=7). Further analysis revealed that STI-571 had no effect on Erk1/2 phosphorylation, despite a positive correlation between Erk1/2 tyrosine phosphorylation and c-Abl expression (n=31, r=0.606, P<0.001). However, c-Abl appears to play a role in NF-kB pathway signalling in CLL cells. Thus treatment of these cells either with STI-571 or with c-abl siRNA inhibited this pathway, particularly in those CLL cases with high c-Abl expression. In conclusion, this study has identified a group of prognostically unfavourable CLL cases in which malignant cells selectively overexpress the myristoylated 1b isoform of c-Abl. Treatment of these cells with STI-571 led to the inhibition of apparently constitutive activation of NF-kB and accelerated in vitro cell death by apoptosis. This indicates that, in CLL cases with overexpressed c-Abl, pharmacological inhibition of this enzyme may be of therapeutic benefit.

Author notes

Corresponding author

Sign in via your Institution