Rituximab Activation of AKT and HIF Pathways Is Dependent on Membrane Lipid Raft Cholesterol Levels in Lymphoma Cells

Abstract 1571

Rituximab, which is a monoclonal antibody directed against CD20 proteins, has significantly improved the treatment outcome of B-cell lymphoma patients. Recent studies have revealed that the lipid components of the membrane microdomain, also known as the lipid raft, determine the biological function and efficiency of the antibody. The raft-associated sphingolipid GM1 level also affects the susceptibility of lymphoma cells to rituximab. Clinical observations have suggested that the use of statins may affect the efficiency of rituximab by modulating lipid raft cholesterol levels. In the present study, we investigated whether differences in lipid raft components affected rituximab-induced intracellular signaling pathways and the biological activity of the antibody. Initially, we analyzed the membrane cholesterol and GM1 levels in several B-cell lymphoma cells (Raji, RL,Namalwa and Ramos cells). We found that two cell lines (Raji and RL cells) have higher cholesterol levels compared with Namalwa and Ramos cells; however, Namalwa and Ramos cells have higher GM1 expression compared with Raji and RL cells. Interestingly, rituximab clearly activated the PI3K/AKT pathway in the cholesterol-rich cells (Raji and RL cells). Conversely, treatment with rituximab suppressed the basal activity of AKT in the GM1-rich cells (Namalwa and Ramos cells). We also investigated whether cholesterol levels or the GM1 level affected rituximab-induced PI3K/AKT activation. We treated the cholesterol-rich cells with methyl-β-cyclodextrin (MβCD) to deplete cholesterol from the lipid rafts. Treatment with MβCD clearly disrupted rituximab-induced AKT activation. Importantly, cholesterol replacement restored rituximab-induced AKT activation. In contrast, treatment with D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (PDMP), which inhibits the synthesis of GM1, did not reverse rituximab-induced AKT suppression in the GM1-rich cells. These results suggest that lipid raft cholesterol levels, but not the GM1 level, determine rituximab-induced AKT activation. We also examined the biological significance of rituximab-induced AKT activation in lymphoma cells. Although AKT activates a variety of downstream molecules, we focused our attention on hypoxia-inducible factor (HIF) because recent studies have revealed that abnormal expression of the alpha subunit of HIF-1 (HIF-1α) is frequently found in lymphoma cells. In agreement with the finding that rituximab induced AKT activation, treatment with rituximab markedly increased the expression of HIF-1α in the cholesterol-rich cells. In contrast, rituximab reduced the basal HIF-1α level in the GM1-rich cells. Interestingly, rituximab enhanced the expression of the anti-apoptotic protein survivin in a HIF-1-dependent manner in Raji and RL cells. In addition, rituximab suppressed the chemotherapeutic reagent-induced apoptosis of Raji and RL cells. Interestingly, depletion of membrane cholesterol by MβCD completely blocked all these processes. In conclusion, rituximab exerts different effects on lymphoma cells that are dependent on lipid raft cholesterol levels. Our observations suggest that a high level of membrane cholesterol may diminish rituximab-induced apoptosis through AKT activation and subsequent induction of HIF-1α. Importantly, a reduction of membrane cholesterol may enhance the efficiency of rituximab.