Increased Reactive Oxygen Species and the B-Cell Receptor in Chronic Lymphocytic Leukemia Signaling

Reactive oxygen species (ROS) play important roles in regulating cell signaling, replication and survival. Chronic lymphocytic leukemia (CLL) cells generally contain high levels of mitochondrial-derived ROS compared to normal B cells. However, there is considerable variation in ROS levels between individual samples. Previous studies have demonstrated that chemotherapy may exert an important influence on ROS levels since prior therapy was linked to increased ROS, potentially via accumulation of mitochondrial DNA damage. However, variability in ROS levels is also apparent between samples from untreated patients demonstrating that additional factors influence ROS levels in CLL cells. One additional variable may be signalling via the B-cell receptor (BCR), now recognised as a major determinant of disease behavior and a target for therapeutic attack. BCR signalling appears to be on-going in CLL with the balance between antigen-induced anergy and positive signalling influencing outcome. We therefore investigated whether ROS levels correlated with subsets of diseases defined by IGHV mutation status or extent of anergy, and with outcome. We also investigated the role of ROS in modulating signaling via surface IgM (sIgM) in vitro.

Flow cytometric analysis of 33 peripheral blood mononuclear cell samples demonstrated that ROS levels were highly variable between individual CLL patients. As demonstrated previously, mitochondria appeared to be the major source of ROS in CLL cells. ROS levels were higher in M-CLL compared to U-CLL (P=0.01). ROS levels were also higher in samples that had strong features of anergy, including down-modulation of surface IgM (sIgM) expression (P=0.003) and signaling capacity (P=0.001). Importantly, higher levels of ROS were associated with longer time to first treatment (P=0.003). Overall, high levels of mitochondrial-derived ROS appear to be a feature of anergy and are associated with M-CLL and indolent disease.

Some patients demonstrated intraclonal variation in ROS production. This was observed in both M-CLL and U-CLL, but was somewhat more prominent in U-CLL. To probe this variation, we investigated expression of CXCR4 in these sub-populations since down-modulation of this receptor “marks” cells which have most recently entered the circulation following tissue-based stimulation. CXCR4^low CLL cells contained relatively high levels of ROS compared to “recovered” CXCR4^high cells indicating that increased ROS was a consequence of tissue based stimulation. Engagement of the BCR by antigen could be one factor that modulates ROS levels in vivo. We were unable to detect any changes in ROS levels following sIgM stimulation in vitro and therefore investigated effects of the anti-oxidant N-acetylcysteine (NAC) on anti-IgM-induced signaling. NAC significantly inhibited anti-IgM-induced ERK1/2 and AKT phosphorylation demonstrating that ROS are required for optimal signalling via sIgM. NAC also inhibited CXCR4-mediated migration suggesting that ROS are important regulators of multiple receptors in CLL cells.

Overall, our results suggest that BCR signaling may be an important modulator of ROS in CLL cells. However, interactions are complex with evidence for both compartmentalization and temporal separation of responses which can only be partly probed with currently available tools. Anergy appears to be associated with increased mitochondrial ROS production, whereas growth promoting BCR signaling may be associated with a transient, localized accumulation at the membrane. CLL cells have increased sensitivity to induction of apoptosis by agents which further enhance ROS and this has been proposed as the basis for novel therapeutic approaches. However, it will be important to consider the multifactorial nature of ROS revealed in our experiments and examine potentially deleterious effects of increased ROS on enhanced BCR signaling.