B-cells are a critical component of the adaptive immune system. As such, B-cells survey the body and mount appropriate protective responses to pathogen-derived antigens, resulting in the production of specific antibodies and induction of immunological memory. Given the effectiveness of these responses in selectively eliminating pathogenic infections, it is clear that the processes underlying antigen-induced B-cell activation must be highly regulated. Somewhat surprisingly given the specialized function of these immune cells, the BCR (B-cell receptor) functions similarly to receptors of the tyrosine kinase family that are commonplace in biology, as BCR ligation with antigen leads to B-cell proliferation and differentiation. In the Lymphocyte Interaction Laboratory, we are particularly interested in characterizing the very early molecular events underlying B-cell activation using a combination of cutting-edge high-resolution and in vivo imaging techniques.
Engagement of the BCR (B-cell receptor) with specific antigen induces two distinct events within the B-cell that are required for activation. The first of these involves the initiation of intracellular signalling events. However, as the membrane Ig (immunoglobulin) portion of the BCR responsible for binding to extracellular antigen does not contain extensive intracellular domains, the associated Igα/β sheath mediates signal transduction through its ITAMs (immunoreceptor tyrosine-based activation motifs) . The binding of antigen leads to BCR clustering and induces phosphorylation of these ITAMs, leading to assembly of a complex of signalling molecules and adaptors, known as the signalosome [2–10]. The signalosome is responsible for regulating a number of diverse cellular effects, such as cytoskeleton reorganization, gene expression and BCR-mediated antigen internalization. The latter effect allows the accumulation of antigen in endosomal compartments containing newly synthesized MHC-II [11,12]. This enables the second global event required for B-cell activation, the presentation of antigenic fragments in complex with MHC-II leading to the recruitment of cognate CD4+ T helper cells.
B-cell activation in response to membrane antigen
Classically these activation events have been characterized in response to stimulation with model soluble antigens. However, it is becoming increasingly evident that B-cell activation in vivo is predominantly mediated by antigen bound to the surface of cells, such as FDCs [follicular DCs (dendritic cells)], DCs and macrophages . This switch in perspective necessitates that the molecular characterization of the events underlying B-cell activation are reconsidered in the appropriate context . A very elegant demonstration of the importance of considering B-cell activation in its cellular context was provided through a series of electron microscopy images (Figure 1A) . This investigation revealed that upon encounter with antigen-presenting cells, B-cells undergo a two-phase spreading and contraction response. During the rapid spreading phase, the B-cell extends lamellipodia across the antigen-presenting cell, maximizing the amount of antigen encountered. This response was absolutely dependent on BCR signalling and reorganization of the actin cytoskeleton . In addition, the extent of the spreading response was dependent on the affinity and density of antigen present in the membrane. Concomitant with this cellular response, numerous microclusters of BCR and antigen formed throughout the contact area (Figure 1B) [16,17]. Subsequently and more slowly, the B-cell contracts, aggregating antigen into a central cluster that can act as a platform for antigen internalization . Together the spreading and contraction responses determine the amount of antigen acquired and presented by the B-cell, and therefore shape the outcome of B-cell activation. Thus a comprehensive description of B-cell activation must now incorporate both the molecular signalling cascades initiated in response to receptor engagement together with the associated co-ordinated cytoskeleton rearrangements.
B-cells spread and contract after BCR recognition of antigen presented on the surface of a cell
BCR microclusters mediate intracellular signalling
A detailed dissection of the molecular requirements for B-cell spreading has shed light on the role of BCR-antigen microclusters in mediating signalling [16,17]. On contact with antigen-containing membrane, primary naïve B-cells very rapidly form antigen microclusters containing between 50 and 500 molecules of both IgM and IgD . Interestingly, in the absence of the Igα/β sheath, although the absolute number is reduced, microclusters were still formed, indicating that their assembly does not depend on signalling through the BCR. However, these antigen microclusters form the sites for recruitment of the kinases previously implicated in the initiation of BCR signalling, namely Lyn and Syk [16,17]. Subsequently, these sites can then mediate the highly co-ordinated recruitment of further intracellular signalling molecules and adaptors, such as PLCγ2 (phospholipase Cγ2), Vav and Blnk (B-cell linker) (Figure 2). Importantly this type of assembly is formed on a time scale consistent with the initiation of calcium signalling after antigen stimulation. We have designated these assemblies as microsignalosomes, to reflect their importance in mediating BCR signalling . As microsignalosomes are critical for propagating B-cell spreading, which in turn will amplify the formation of further microsignalosomes, these two responses together co-operate ultimately to enhance B-cell activation. Furthermore, as similar signalling microclusters are critical for sustaining signalling through the TCR (T-cell receptor) [18–20], we have suggested that these structures are common units of immunoreceptor signalling among lymphocytes [16,21].
The molecular composition of the microsignalosome
A role for the actin cytoskeleton in regulating BCR diffusion and clustering
The observed importance of BCR microclusters for mediating signalling raises the important question of how these structures are assembled in the B-cell membrane. Early biochemical studies suggested that after stimulation with soluble antigen the BCR became associated with the actin cytoskeleton [22,23] and further that this association was independent of the Igα/β sheath . However, the role of the actin cytoskeleton in shaping the diffusion dynamics of the BCR in the plasma membrane is unclear. Very recently, we have used single particle tracking together with dual-acquisition TIRF (total internal reflection fluorescence) microscopy to establish a role for the actin network proximal to the plasma membrane in regulating BCR diffusion dynamics in unstimulated B-cells (Figure 3) (B. Treanor D. Depoil, M. Weber, O. Dushek, A. Bruckbauer and F.D. Batista, unpublished work). Furthermore, we demonstrated that the dynamic linkage of the actin network to the plasma membrane via the ERM (ezrin–moesin–radixin) family member, ezrin, was critical for establishing and gating barriers to BCR diffusion. However, such barriers were not found to impede the diffusion of a chimaeric BCR lacking the Igα/β sheath, and thus this regulation of diffusion was dependent on factors intrinsic to the BCR. Interestingly, disruption of this ezrin-defined actin network removes constraints to BCR diffusion and resulted in the initiation of intracellular calcium signalling, possibly as a result of the formation of signalling BCR clusters. Importantly, on antigen stimulation, we found that this network was reorganized such that ‘corrals’ surrounding BCR microclusters were visible. These corrals were responsible for restricting BCR diffusion from the microcluster and were thus important for maintaining the integrity, organization and signalling competence of BCR microclusters. We expect that this dynamic cytoskeleton network will prove important in providing a means of regulating and co-ordinating diffusion dynamics and clustering of a wide variety of cell-surface receptors.
The actin network regulates the dynamics of BCR diffusion and clustering
CD19 is essential for B-cell activation in response to membrane antigen
The role of the B-cell co-receptor CD19 in the amplification of BCR signalling in response to complement-tagged antigen as part of the CD21–CD81–Leu13 complex has been well established . An additional role for CD19, independent of CD21, in facilitating BCR signalling was suggested by the observation that mice deficient in CD19 exhibit a more severely immunocompromised phenotype than those deficient in CD21 . However, as B-cells lacking CD19 were not impaired in response to soluble antigen in vitro [27,28], the role of CD19 in B-cell activation remained an anomaly until relatively recently. We have observed that CD19-deficient B-cells were severely impaired in their ability to initiate signalling in response to membrane-bound antigen . This resulted both in the abrogation of spreading and B-cell activation in response to stimulation with membrane-bound antigen. Thus CD19 plays an essential role in mediating B-cell activation after stimulation with the most commonly encountered form of antigen in vivo. Interestingly, it was observed that CD19 became transiently associated with BCR microclusters after activation with membrane antigen . Indeed, more recently, we have visualized the movement of single particles of CD19 between BCR microclusters and have demonstrated that CD19 is dynamically recruited to signalling BCR microclusters in response to membrane-antigen stimulation (D. Depoil and F.D. Batista, unpublished work).
These observations raise the question of the molecular mechanism by which CD19 enhances signalling through the BCR. As the intracellular domain of CD19 contains binding sites for a plethora of signalling and adaptor molecules, including PI3K (phosphoinositide 3-kinase) and Vav [29,30], it seems reasonable to suggest that CD19 could mediate the recruitment of these molecules to facilitate signalling through the BCR (Figure 2). Indeed, it has been observed that components, such as PLCγ2 and Vav, co-operate to enhance the recruitment and retention of the other at the microsignalosome . In support of this suggestion, we have observed that molecules recruited by CD19, such as PI3K, are laterally segregated from those recruited by the BCR (D. Depoil and F.D. Batista, unpublished work). This recruitment of CD19 to BCR microclusters illustrates the dynamic regulation that can be offered by signalling through small units compared with larger BCR caps seen after soluble antigen stimulation. As such, microsignalosomes offer great versatility in terms of regulation of BCR signalling as they can mediate the dynamic recruitment and recycling of positive and negative regulators and thus influence the outcome of membrane-antigen stimulation. Indeed it is clear that the formation of microsignalosomes, and the dynamic recruitment of CD19, provides a critical boost to BCR signalling required for driving remodelling of the actin cytoskeleton necessary for activation in response to membrane antigen.
The mature IS (immunological synapse) forms after B-cell spreading and contraction
While these early signalling events have only recently been described, the resulting IS structure formed at later stages in response to membrane-antigen stimulation was revealed more than a decade ago (Figure 1). The IS was originally observed in CD4+ T-cells [31–33] and has since been demonstrated to be a general feature associated with immunoreceptor stimulation [34–37]. The mature IS is a ‘bull's-eye’ structure and its formation involves the extensive reorganization of proteins within the plasma membrane. Immunoreceptors become clustered together in a cSMAC (central supramolecular activation cluster), and are encircled by adhesion molecules such as the integrins LFA-1 (lymphocyte function-associated antigen 1) and VLA-4 (very late antigen-4) that comprise the pSMAC (peripheral SMAC) . As signalling molecules were concentrated in the cSMAC, this region was first thought to be the site of active signalling. In view of the more recent observations that peripheral microclusters of immunoreceptors are important for mediating sustained signalling necessary for T-cell activation [19,20], it seems likely that the cSMAC instead might act as a platform for immunoreceptor internalization and signalling termination . However, the localization of precise functions to particular zones of the IS and their associated role in mediating lymphocyte activation remain controversial.
Towards a characterization of B-cell activation in vivo
As detailed above, progress has been made recently in terms of characterizing the molecular processes underlying B-cell activation in vitro. However, the question arises as to how representative these observations are of B-cell activation in vivo? Very recently, we and others have used multiphoton microscopy to visualize the dynamic events after B-cells encounter antigen in real time in living tissue. These studies have clearly demonstrated that antigen on the surface of DCs  and macrophages [40–42] can initiate the activation of B-cells in the lymph node. Furthermore, we observed that B-cells appeared to spread out across the surface of the antigen-presenting cell prior to gathering antigen together in an IS-like structure . These preliminary findings strongly suggest that the mechanisms that have been characterized in vitro may indeed be employed during the recognition of antigen and stimulation of B-cell activation under physiological conditions.
The original descriptions of the molecular pathways underlying B-cell activation were derived using classical biochemistry techniques after stimulation with soluble antigen. These studies have provided an important foundation for our understanding of these pathways; however, it is imperative that B-cell activation is considered in a context more representative of the physiological situation. Thus more recent investigations have considered the spatiotemporal dynamics of B-cell activation at the molecular level in response to membrane-bound antigen. It has become clear that this process involves receptor engagement alongside the induced cytoskeleton rearrangements. The collaboration of these cellular events provides the B-cell with an exquisite and dynamic mechanism for regulation of activation. Indeed the role of CD19 and its dynamic recruitment to microsignalosomes elegantly illustrates the versatility and importance of co-operation between these events during B-cell activation.
The Dynamic Cell: Joint Biochemical Society and British Society for Cell Biology Focused Meeting held at Appleton Tower, University of Edinburgh, U.K., 1–4 April 2009. Organized and Edited by Ian Dransfield (Edinburgh, U.K.), Margarete Heck (Edinburgh, U.K.), Kairbaan Hodivala-Dilke (Cancer Research UK, London, U.K.), Robert Insall (Beatson Institute for Cancer Research, Glasgow, U.K.), Andrew McAinsh (Marie Curie Research Institute, Oxted, U.K.) and Barbara Reaves (Bath, U.K.).
central supramolecular activation cluster
immunoreceptor tyrosine-based activation motif
peripheral supramolecular activation cluster
We thank members of the Lymphocyte Interaction Laboratory for a critical reading of this paper.