B cell antibody responses are triggered by the binding of antigen to the clonally distributed B cell antigen receptors (BCRs). Over the last several years a great deal has been learned about the biochemistry of the complex signal cascades triggered by BCR antigen engagement. However, what remains relatively poorly understood are the molecular events that trigger the initiation of signaling. These events are likely to occur within seconds of antigen binding to the BCR and to be highly dynamic, involving many weak protein-protein and protein-lipid interactions. In general, the biochemical approaches that have been used so effectively to describe the BCR signaling pathways are inadequate to capture events that occur as rapidly and as transiently as those predicted to initiate BCR signaling. In addition, antigen binding to the B cell involves a dramatic spatial change in the BCRs resulting in their patching and capping and the formation of an immune synapse. All this potentially important spatial information is lost with the addition of detergents to cells for biochemical analyses. Consequently we have taken advantage of new live cell imaging technologies that allow analyses of the BCR and components of the BCR signaling pathway with the temporal and spatial resolution necessary to view the earliest events in B cell activation at the single molecule level without the complications that the addition of detergents introduce. Using live cell imaging, we showed that in B cells responding to antigen presented on a planar lipid bilayer, simulating an antigen presenting cell, BCRs form signaling active microclusters by a mechanism that does not require physical crosslinking of BCRs by multivalent antigens. We observed that in response to either monovalent or multivalent antigens the BCRs accumulate and form microclusters at the initial points of contact of the B cell membrane with the bilayer. Clustering did not depend on the ability of the BCR to signal subsequently, revealing clustering to be an intrinsic property of the BCRs. The microclusters grew by trapping mobile BCRs and the larger clusters were actively organized into an immune synapse. The kinetics of these events and signaling were identical for monovalent and multivalent antigens. We determined that the membrane-proximal ecto-domain of the BCR mIg was both necessary and sufficient for BCR oligomerization and signaling. BCRs that contained a mIg in which Cmu4 was deleted failed to cluster and to signal when engaging monovalent antigen. Conversely, Cmu4 expressed alone on the B cell surface spontaneously clustered and activated B cells. These findings lead us to propose a novel mechanism by which BCRs form signaling active microclusters that we termed the conformation induced oligomerization model for the initiation of BCR signaling. According to our model BCR in the resting state are not in an oligomerization receptive conformation such that random bumping has no repercussion. The binding of antigen on an opposing membrane exerts a force on the BCR to bring it into an oligomerization receptive form so that when two antigen-bound BCRs bump they oligomerize. Using the same live cell imaging technologies we approached a fundamental question in B cell biology namely what advantage is conferred on B cell activation by high affinity, isotype switched BCRs such as those expressed by memory B cells. We demonstrated that high affinity BCRs, as compared to low affinity BCR form immobile, signaling active BCR clusters more efficiently providing evidence that affinity discrimination is a BCR intrinsic function. We also determined that as compared to mIgM-containing BCRs, mIgG BCRs more efficiently form signaling active BCR microclusters in both mouse and human B cells. These results are important in providing a molecular understanding of the functional advantage of high affinity, isotype-switched BCRs. Because of the paucity of memory B cells in mice, our studies of IgM versus IgG BCRs in mouse B cells was limited either to transfected cell lines or to naive B cells that had been induced to switch to IgG in vitro. Over the last year we extended our live cell imaging studies to describe the behavior of BCRs on human peripheral blood naive IgM-expressing B cells and IgG-expressing memory B cells in both resting cells and cells in which the BCR was activated. For the most part the behavior of human IgM BCRs on naive B cells and IgG BCRs on memory B cells recapitulated the behavior we observed for mouse B cells expressing IgM and IgG in that human IgG BCRs oligomerized more rapidly and to greater extent as compared to IgM BCRs. However, one novel observation was that in the resting state a larger portion of IgG BCRs appeared to be immobilized and associated with phosphorylated PI3K. This observation suggests that memory B cells may be primed to respond to antigen. To pursue this possibility we set out to determine the organization of the BCR on resting naive B cells and on memory B cells. Current evidence indicates that it is likely that BCRs are not randomly distributed on the B cell surface but rather confined to areas that define their interactions with both positive and negative regulating coreceptors. Such confinement could result in the relative immobility of the IgG BCRs on memory B cells and their association with activated PI3K. The organization of receptors on the B cell surface can be determined through super-resolution fluorescence imaging that allows the localization of single molecules. Over the last year we established super resolution imaging (STORM) to acquire data, at the 10-50 nm level, on the organization of the BCR in resting naive and memory B cells and in B cells in which the BCR is ligated. Once imaging data are obtained and the precise localization of single fluorescent molecules determined, the challenge is to construct a super-resolution image. However, stochastic variations in photon emissions and intervening dark state result in uncertainties identifying single molecules. We have collaborated with Dr. Jennifer Lippincott-Schwartz to use a pair-correlation analysis and rigorous statistical algorithms to quantitatively describe the spatial organization of the BCR. Our data indicate that most BCRs exist as monomers or dimers in both naive and memory B cells without an obvious organization. Following crosslinking the BCRs form large clusters of hundreds of molecules. In agreement with our earlier data the clusters formed by IgM BCRs on naive B cells are not as large as IgG BCR clusters on memory B cells. Over the last year we begun a new initiative to characterize human tonsillar germinal center (GC) B cells. GCs are compartments within secondary lymphoid organs in which B cell clonal expansion, somatic hypermutation and affinity-based selection occur resulting in the production of isotype-switched memory B cells and high affinity antibody secreting plasma cells. In recent years discrete steps in the GC reactions following immunization have been mapped out in mouse models. However, our understanding of the B cell biology of human GCs lags behind the mouse models and clearly knowledge of the cellular and molecular mechanism by which high affinity memory B cells and plasma cells are generated would aid in vaccine design. In collaboration with Dr. Susan Moir (LIR, NIAID), we were able to devise a cell separation strategy that provided relatively pure populations of tonsil naive B cells, memory B cells and GC dark and light zone B cells. Our analysis thus far indicates that the GC B cells express less BCR per cell as compared to either naive or memory B cells and are less active in clustering their BCR and reorganizing the actin cytoskeleton to form immune synapses.
Showing the most recent 10 out of 23 publications
