Abstract
HVCN1 is a highly-conserved voltage-gated proton channel. Voltage-gated proton currents have been recorded in lymphocytes but their functions in B cells remain unknown. We isolated HVCN1 in a proteomic survey of plasma membrane proteins in mantle cell lymphoma (MCL) in leukemic phase. In normal lymphocytes, HVCN1 expression was restricted to the B-cell lineage; HVCN1 was highly expressed in mantle zone cells but down-regulated in germinal center (GC) cells undergoing receptor affinity maturation and class-switch recombination (CSR). Highest level expression was also observed in Chronic Lymphocytic Leukemia (CLL) cells from the peripheral blood. In MCL tumors, HVCN1 was expressed in circulating cells but absent from involved lymph nodes, whereas in diffuse large B cell lymphoma (DLBCL), its expression correlated with cases with a low proliferation index. Thus, in both primary and neoplastic B cells, HVCN1 expression appears to be associated with a non-proliferative phenotype. In human primary resting B cells and B cell lines, HVCN1 directly interacted with the B cell receptor (BCR) complex, as shown by Igβ and HVCN1 reciprocal immunoprecipitation experiments. We also found by confocal microscopy and subcellular fractionation, that upon BCR engagement the channel was internalized with the antigen receptor and the two proteins co-migrated to the endo-lysosomal, MHC class II (MHC-II) containing compartments (MIICs). When overexpressed in a hen egg lysozyme (HEL)-specific B cell clone, LK35.2, HVCN1 showed a basal phosphorylation which increased with HEL stimulation. The increased phosphorylation corresponded to an increase in proton conductance, termed “enhanced gating mode” and it was PKC dependent. We then asked whether HVCN1 over-expression could influence MHC II antigen presentation and if the effect could be mediated by changes in MIICs pH. Indeed, presentation of HEL peptides to a T cell clone was impaired in LK35.2 and A20 D1.3 cells, where HVCN1 had been re-introduced; effect was stronger for plate-bound antigen than for soluble antigen. The reduced antigen presentation was accompanied by an increase in endo-lysosomal pH, from pH4.9 ± 0.2 to 6.3 ± 0.1 (which may reflect HVCN1 channel-mediated proton flux out of the organelles), as measured with an anti-IgM antibody conjugated to a pH sensitive dye in HVCN1 over-expressing cells. Evidently, the presence of HVCN1 leads to increased endo-lysosomal pH, consistent with H+ current from the lysosomal compartment into the cytosol. Hence, active antigen presenting cells, like GC cells, might down-regulate HVCN1 expression to maximize the effect of antigen presentation. In order to investigate the role of HVCN1 in vivo, we used a HVCN1-deficient mouse line generated by genetrap insertion. These mice showed no obvious changes in numbers or composition of B-cell subpopulations. Immunization of HVCN1-deficient mice with a T-dependent antigen resulted in a defect in CSR to all IgG subclasses, particularly marked for the IgG2b, whereas in contrast, no differences were observed in IgM secretion, suggesting a pivotal role for HVCN1 during antigen-driven B-cell activation and subsequent CSR. HVCN1 may influence B-cell activation through alteration of reactive oxygen species (ROS) as HVCN1-deficient B cells showed reduced ROS production following BCR activation, a sign of suboptimal NADPH oxidase activity. It has been postulated that proton channels are required to counterbalance the electrogenic activity of NADPH oxidase during ROS production. Our data suggest that this mechanism also occurs in vivo and shed new light on the role of ROS in B cell activation and downstream effects.
Btk levels set the threshold for B-cell activation and negative selection of autoreactive B cells in mice
