Calcium signalling and cell-fate choice in B cells

Abstract

Alterations in the cytosolic concentration of calcium ions (Ca²⁺) transmit information that is crucial to the development and function of B cells. Cytosolic Ca²⁺ concentration is determined by a balance of active transport and gradient driven Ca²⁺ fluxes, both of which are subject to the influence of multiple receptors and environmental sensing pathways. Recent advances in genomics have allowed for the compilation of an increasingly comprehensive list of Ca²⁺ transporters and channels expressed by B cells. The accumulating understanding of the function and regulation of these proteins has begun to shift the frontier of Ca²⁺ physiology in B cells from molecular analysis to determining how diverse inputs to cytosolic Ca²⁺ concentration are integrated in specific immunological contexts.

Introduction

B cells must respond to diverse chemical and environmental cues. Many of these cues influence the functions of Ca²⁺ transport proteins, resulting in alterations in cytosolic Ca²⁺ referred to as Ca²⁺ signals. Ca²⁺ signals have the capacity to affect a variety of intracellular processes that are central to cell-fate decisions in B cells, including protein kinase signalling, mitochondrial physiology, apoptosis, nucleocytoplasmic trafficking of transcription factors, chromatin accessibility, and cell adhesion and migration. Here, we review Ca²⁺ transport and signalling physiology in B cells, placing a special emphasis on the phospholipase C-γ (PLCγ)–inositol-1,4,5-trisphosphate (InsP₃)-mediated Ca²⁺ signalling mechanism used by the B-cell receptor (BCR). We also highlight new mechanisms for the regulation of cytosolic Ca²⁺ concentration, and discuss how these combined inputs could impact Ca²⁺-dependent regulation of nuclear factor of activated T cells (NFAT) and nuclear factor-κB (NF-κB) transcription factor pathways and of cell motility in B cells to influence cell-fate choice during humoral immune responses.

The InsP₃-mediated Ca²⁺ signalling pathway

In vertebrate organisms, cytosolic Ca²⁺ levels are subject to regulation through multiple mechanisms (Table 1), the precise nature and regulation of which vary widely depending on the tissue and cell type. In B cells, many important surface receptors initiate Ca²⁺ signals through the production and accumulation of the soluble second messenger InsP₃ ^1–3 (Figure 1). InsP₃ is produced by the hydrolysis of the membrane lipid phosphatidylinositol-4,-5-bisphosphate (PtdIns(4,5)P₂; also known as PIP₂), a process that also produces the lipid second messenger diacylglycerol (DAG) (reviewed in ⁴). PtdIns(4,5)P₂ hydrolysis is mediated by members of the PLC enzyme family, which is comprised of several differentially regulated isoforms (reviewed in ⁵). An important distinction among the various PLC isoforms is that 7-transmembrane spanning G-protein-coupled receptors (GPCRs) activate PLCβ isozymes through the heterotrimeric G-protein G_q and related subunits, whereas tyrosine-kinase-linked receptors, including many growth factor receptors, the T-cell receptor (TCR), the BCR and activating Fc receptors (FcRs), activate PLCγ isozymes through tyrosine phosphorylation.

Table 1.

Cytosolic Ca²⁺ regulatory mechanisms in lymphocytes

Mechanism	Biochemical action	Main cell physiological roles	Reference
Active transport
SERCA pumps	Pumps Ca2+ from the cytosol into ER	Maintain ER stores of Ca2+; shape Ca2+ signals	114
PMCA pumps	Pumps Ca2+ from the cytosol to outside the cell	Shape Ca2+ signals; limit total cellular Ca2+ accumulation	18,115
Exchange transport
Na+/Ca2+ exchangers	Direct mode: exchange cytosolic Ca2+ for Na+ during high cytosolic Ca2+ accumulation	Limit total cellular Ca2+ accumulation	reviewed in 113
	Reverse mode: exchange of cytosolic Na+ for extracellular Ca2+	Mechanism for Ca2+ entry in response to accumulation of cytosolic Na+	reviewed in 113
Mitochondrial uptake
Unknown mitochondrial ion channel	Ion channels that allows primarily electrical gradient-driven diffusion of Ca2+ from ER to mitochondrial matrix	Transient buffering of cytosolic Ca2+ during acute elevations	116,117, reviewed in 118,119
Mitochondrial extrusion
Unknown Na+-or H+-gradient-driven active transporters	Exchangers that extrude accumulated Ca2+ from mitochondrial matrix once cytosolic Ca2+ falls	Buffering of the decreasing phase of a Ca2+ signal via gradual release of Ca2+ to the cytosol	reviewed in 119,120
Passive diffusion
InsP3 receptors	Ion channels which allow passive chemical-gradient-driven diffusion of Ca2+ from ER to cytosol	Transient acute elevations in cytosolic Ca2+	Reviewed in 121
Plasma membrane Ca2+ channels	Ion channels that allow passive chemical and electrical-gradient-driven diffusion of Ca2+ from ER to cytosol	Transient and sustained elevations in cytosolic Ca2+; replenishment of total cellular Ca2+	See for example 2,3,6

SERCA, sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPase; PMCA, plasma membrane Ca²⁺ ATPase; ER, endoplasmic reticulum; InsP₃; inositol-1,4,5-triphosphate;

Figure 1. Ca²⁺ physiology in B cells at rest and during InsP₃-mediated Ca²⁺ signal.

A: In "resting" B cells, cytoplasmic Ca²⁺ homeostasis is maintained primarily through the actions of plasma membrane Ca²⁺ ATPase (PMCA) and sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPase (SERCA) active transporters or pumps. Na⁺/Ca²⁺ exchange is limited due to lack of cytoplasmic Ca²⁺, and plasma membrane channels are primarily in the closed state.

B: A receptor-mediated (inositol-1,4,5-trisphosphate (IP₃))-dependent calcium signal is initiated through binding of IP₃ to IP₃R proteins in the endoplasmic reticulum (ER) membrane. Opening of IP₃R channels allows Ca²⁺ to enter the cytoplasm from ER Ca²⁺ stores. Once ER Ca²⁺ stores are sufficiently depleted, ER Stromal interaction molecule (STIM) family proteins are activated to move into proximity to, and open, Ca²⁺-release activated channels (CRAC channels) in plasma membrane microdomains. Notable aspects of cell physiology in the context of an active Ca²⁺ signal include: (1) plasma membrane potential provides a significant driving force for Ca²⁺ entry through ion channels; (2) active Ca²⁺ transport activities (sarcoplasmic reticulum/endoplasmic reticulum Ca²⁺ ATPase (SERCA), plasma membrane Ca²⁺ ATPase (PMCA), and Na⁺/Ca²⁺ exchange) increase relative to the resting state in response to the increased availability of Ca²⁺ on their pumping cytosolic surfaces; and (3) Na⁺ entry into the cytoplasm may significantly influence cytosolic Ca²⁺ concentration depending on local changes in Na⁺ concentrations and the relative flow of Ca²⁺ through Ca²⁺ selective ion channels, non-selective cation channels (G), and reverse mode Na⁺ /Ca²⁺ exchange (that is, Ca²⁺ transported to the cytosol from the extracellular environment in exchange for intracellular Na⁺)¹¹¹–¹¹³.

InsP₃ accumulation causes ER Ca²⁺ release and initiates store operated Ca²⁺ entry

When a PLC-linked receptor is activated, the InsP3 generated binds to and induces the opening of InsP₃ receptor (InsP₃R) channels in the endoplasmic reticulum (ER) membrane, allowing Ca²⁺ to flow from ER stores to the cytosol. This “release” of ER Ca²⁺ results in the first detectable increase in cytosolic Ca²⁺ concentration after receptor activation. As stored ER Ca²⁺ is necessarily limited, ER Ca²⁺ release can only support a transient increase in cytosolic Ca²⁺. However, many PLC-linked receptors, including the BCR, are able to sustain increases in cytosolic Ca²⁺ for many tens of minutes. They are able to do this through a process, known as store operated Ca²⁺ entry (SOCE; reviewed in ^3,6–8), that allows Ca²⁺ ions to pass from the essentially unlimited extracellular Ca²⁺ pool to the cytosol. The present SOCE model, as described below, is based on work performed in multiple cell contexts, including DT40 chicken B cells, Drosophila S2 cells, human fibroblasts, and human T cells. Although not every aspect of the model has been tested in all cell types, the bulk of available data suggests that the molecular mechanisms underlying SOCE are largely conserved across species and cell type.

According to the present SOCE model, once InsP₃-induced InsP₃R opening induces sufficient ER Ca²⁺ pool depletion, the reduced ER Ca²⁺ concentration is sensed by members of the stromal interaction molecule (STIM) family of single transmembrane span molecules, which are located as type II transmembrane proteins in the ER membrane^9–11. STIM molecules are thought to undergo a Ca²⁺-dependent conformational coupling [Au: OK?], which leads to direct interaction with and activation of plasma membrane calcium-release activated calcium (CRAC) channels, encoded by multimers of one or more ORAI (also known as CRACM1)-family proteins^12–14. Subcellular imaging studies suggest that coupling of STIM proteins with CRAC channels involves interactions between STIM and CRAC channels occurring at discrete puncta located within regions where junctional ER approaches within 10–25 nm of the plasma membrane^10,13,15,16. The resulting locally activated CRAC channels allow extracellular Ca²⁺ influx to the cytoplasm, thus allowing Ca²⁺ signalling to continue beyond the point of ER-store depletion, as well as providing a source of Ca²⁺ for refilling of the ER store as the store-depleting stimulus subsides. The protein components, biochemistry, and cell biological properties of puncta generated by store depletion are presently under intense investigation to obtain further insights into the nature and regulation of the SOCE mechanism.

Although initiated entirely by passive Ca²⁺ flows through ion channels, Ca²⁺ signals that involve activation of SOCE may be further shaped by active Ca²⁺ transport though plasma membrane Ca²⁺ ATPase (PMCA) pumps. In T cells, this shaping occurs because activation of CRAC channels is closely coordinated with activation of PMCA pumps, such that increased rates of Ca²⁺ entry correlate with increased rates of Ca²⁺ extrusion independently of changes in the average cellular cytosolic Ca²⁺ concentration ^17,18. This close coordination of CRAC channels and PMCA pumps allows the PMCA pumps to respond rapidly to local changes in Ca²⁺ concentration, and thereby to significantly influence the frequency and amplitude of intracellular Ca²⁺ oscillations and/or the peak cytosolic Ca²⁺ concentration triggered by CRAC channel activation.

SOCE mechanism leads to an ‘InsP₃ threshold’ effect

An important characteristic of the SOCE mechanism is that it manifests as an ‘all or nothing’ phenomena: that is either stores are sufficiently empty (depleted) to activate SOCE, in which case SOCE becomes fully activated, or they are not sufficiently empty, in which case SOCE is not activated. This results in an apparent threshold for InsP₃ accumulation, below which ER stores are not depleted and SOCE is not activated, and above which InsP₃R-dependent release of Ca²⁺ from ER stores depletes those stores and activates SOCE¹⁹. This threshold has the practical effect of allowing small differences in the level or duration of PLC activation to lead to very large differences in Ca²⁺ signal phenotype. In the simplest case, if the magnitude or duration of a stimulus is such that sufficient InsP₃ accumulates to maintain ER Ca²⁺ store depletion, CRAC-channel-mediated entry of Ca²⁺ from the vast reservoir of extracellular free Ca²⁺ will sustain a Ca²⁺ signal indefinitely, either in the form of elevated cytosolic Ca²⁺ concentration or oscillations in cytosolic Ca²⁺ concentration. However, a slightly lower level of stimulation may not lead to sufficient accumulation of InsP₃ to deplete Ca²⁺ stores; in this case ER sarcoplasmic/endoplasmic reticulum Ca²⁺ (SERCA) pumps will restore and maintain ER Ca²⁺ stores, SOCE will not be activated, and only a very small and transient Ca²⁺ signal will be observed. Therefore, for any particular microenvironment, whether overall cellular PLC activation is sufficient to cause InsP₃ to accumulate to a level that can maintain ER Ca²⁺ store depletion is an important determinant of the magnitude of the resulting Ca²⁺ signal.

BCR-mediated PLCγ activation

Signals mediated by the BCR are central to multiple cell-fate decisions arising during B-cell ontogeny, including the transitions from pro-to-pre-B-cell and pre-to-immature B-cell in the bone marrow; activation of mature B cells in response to antigen; selection into germinal centres; induction of somatic hypermutation; and re-activation of memory B cells. The biochemical mechanisms through which BCR engagement influences B-cell physiology have been studied in detail over the past two decades. A fundamental concept derived from this work is the importance of activation of the B-cell-specific PLCγ isozyme, PLCγ2, as a BCR effector pathway^20,21. Furthermore, a detailed model has evolved for how the BCR generates three key biochemical events that lead to positive feedback activation of PLCγ2 (Figure 2), and how this mechanism provides a direct target for highly sensitive co-receptor-mediated positive or negative regulatory influences.

Figure 2. Biochemical events in a BCR signaling complex leading to amplified PLCγ2 activation.

A: Recruitment and phosphorylation (yellow arrows and dots) of B-cell linker (BLNK) protein within the signalosome of the B-cell receptor (BCR) generates binding sites and recruitment (green arrows) of key BCR effectors VAV, GRB2 (growth-factor-receptor-bound protein 2), phospholipase Cg2 (PLCγ2) and Bruton's tyrosine kinase (BTK).

B: P110 Phosphoinositide-3-kinase (PI3K) activation occurs through several parallel pathways (black arrows), and leads to local phosphorylation (yellow arrow) of phosphatidylinositol-4,5-bisphosphate to phosphatidylinositol-3,4,5-trisphosphate (PtdIns(3,4,5)P₃). This effect is amplified in part through recruitment (green arrow) of the p85 subunit of PI3K to the CD19 complex.

C: Recruitment of BTK to PtdIns(3,4,5)P₃ induces its transphosphorylation by lyn (small yellow arrows), and initiates a positive feedback loop (large yellow arrows): association of BTK with phosphatidyl inositol-4-phosphate 5-kinase (PI5K) leads to increased production of phosphatidylinositol-4,-5- bisphosphate (PtdIns(4,5)P₂) and so increased substrate for PtdIns(3,4,5)P₃production; enhanced PtdIns(3,4,5)P₃ leads to increased BTK recruitment (with additional PI5K) and BTK activation. The overall signal is read out through BTK-dependent tyrosine phosphorylation and activation of PLCγ2 (red arrows), which hydrolyzes (black arrows) locally available and newly synthesized PtdIns(4,5)P₂ to IP₃ and diacylglycerol (DAG).

A BCR-initiated positive-feedback loop mediates PLCγ2 activation in B cells

The first key biochemical event in BCR-mediated PLCγ2 activation occurs in the context of the formation of a multi-component signalling complex in response to BCR engagement (Figure 2A). The formation of this “signalosome” involves the phosphorylation of tyrosines in the Igα and Igβ chains of immunoreceptor tyrosine-based activation motifs (ITAMs), and subsequent recruitment of SRC-family kinases [Au: OK? Is it not LYN that is recruited along with SYK?] and spleen tyrosine kinase (SYK). The binding and activation of SRC and SYK coordinately results in the recruitment and activation of other signalling proteins including kinases, phosphatases, various G-proteins and their regulatory molecules, adaptor molecules, and lipid hydrolases. Among these targets, the adaptor molecule B-cell linker (BLNK) is rapidly recruited into the signalling complex and phosphorylated^22–24. BLNK recruitment is the first crucial biochemical event in PLCγ2 activation^23,25. Phosphorylated BLNK is directly responsible for the recruitment of both PLCγ2 — through binding of both of the SRC homology 2 (SH2) domains of PLCγ2 to distinct phosphorylated tyrosines in BLNK²² — and of a PLCγ-activating tyrosine kinase known as Bruton's tyrosine kinase (BTK, a member of the TEC-kinase family of tyrosine kinases)^26,27. BLNK recruitment and phosphorylation is absolutely required for BCR-mediated PLCγ2 activation, as PLCγ2 activation is entirely blocked if BLNK is absent²³, and severe defects in B-cell development and function result from natural mutations or targeted deletion in the BLNK gene in humans and mice, respectively^28,29.

The second key biochemical event in BCR-mediated positive feedback activation of PLCγ2 is the generation and accumulation of a localized pool of phosphatidylinositol-3,4,5-trisphosphate (PtdIns(3,4,5)P₃) through several parallel pathways that regulate the activity of p110 isoforms of phosphoinositide 3-kinase (PI3K, Figure 2B). As PtdIns(3,4,5)P₃ is produced as a consequence of phosphorylation of PtdIns(4,5)P₂ (also the substrate for PLCγ2) by PI3K, PtdIns(3,4,5)P₃ preferentially accumulates in membrane areas enriched in PtdIns(4,5)P₂. PtdIns(3,4,5)P₃ binds with high affinity to the plekstrin-homology domain (PH domain) of BTK, and BTK is therefore rapidly recruited into membrane regions where PtdIns(3,4,5)P₃ is being formed^30,31. As both BTK and PLCγ2 are bound to phosphorylated BLNK, interaction of the BTK PH domain with PtdIns(3,4,5)P₃ also serves to bring PLCγ2 into regions where PtdIns(3,4,5)P₃ is being formed, which, as noted above, are also enriched in the PtdIns(4,5)P₂ substrate for PLCγ2. In addition, PtdIns(3,4,5)P₃ may also have a direct role in activating PLCγ isozymes through an interaction with a functional PH domain formed by apposition in the folded PLCγ protein of two half-domains found widely separated in the primary sequence³². Furthermore, as BTK constitutively interacts with phosphatidylinositol-4-phosphate 5-kinase (PIP5K)³³, recruitment of BTK also brings PIP5K into the same complex. PIP5K promotes the formation of additional PtdIns(4,5)P₂, thereby ensuring a supply of PtdIns(4,5)P₂ substrate for both activated PI3K and PLCγ2. The importance of PI3K recruitment to the signal transduced by the BCR is illustrated by the defects in immune development and function that result from targeted deletions in various PI3K subunits^34,35.

In combination, the first two biochemical events result in positive-feedback amplification of upstream signalling events (Figure 2C). The degree of amplification they achieve is subsequently "read out" by the third key biochemical event: BTK-dependent phosphorylation of PLCγ2 on tyrosine residues, Tyr753 and Tyr759, in its SH2–SH3 linker region³⁶. PLCγ2 phosphorylation on these two residues is synergistically promoted by the positive feedback created by the first two events. This is because BTK and PLCγ2 are co-localized by their binding to phosphorylated BLNK, and because PtdIns(3,4,5)P₃ binding to the BTK PH domain potentiates tyrosine-phosphorylation-dependent activation of BTK occurring within the signalling complex^30,31. The phosphorylation of PLCγ2 on these two regulatory tyrosines is a pivotal output of the signalling complex formed by the BCR, as these phosphorylation events are required to maximally activate the lipid hydrolase activity of PLCγ2, and are specifically mediated by BTK in human B cells³⁷, and by BTK and the related TEC-kinase family TEC (tyrosine kinase expressed in hepatocellular carcinoma) in murine [Au: means both rat and mouse – OK or should it be mouse?] B cells³⁸. Tyrosine kinases not members of the TEC-kinase family present in the BCR signalling complex, such as SYK or LYN, are not able to efficiently access and/or phosphorylate these regulatory tyrosines^36,39,40. The importance of TEC-kinase-family-dependent phosphorylation of PLCγ2 as an output of the signal transduced by the BCR is illustrated by the profound defects in calcium flux, and immune development and function in mice or humans which result from targeted deletion or natural mutations in BTK and TEC kinases^37,41,42.

The significance to B-cell physiology of the signal amplification achieved by this mechanism is underscored by the striking similarities of the B-cell defects present in mice with individual targeted deletions in the genes encoding BLNK, PI3K, PLCγ2, or BTK, or in humans with natural mutations in the homologous genes. In each case, B-cell development is inhibited at checkpoints where a sufficient amplitude and duration of pre-BCR or BCR signals are required for developing or differentiating B cells to survive and proliferate. The importance of amplified PLCγ2 activation is further underscored by the existence of several co-receptors that function to directly modulate the positive-feedback loop as a means of tuning BCR signal amplitude, as discussed below.

Positive modulation of BCR Ca²⁺ signalling

Tyrosine phosphorylation-dependent activation of PLCγ2 via the BCR is superficially similar to PLCγ1 and/or PLCγ2 activation that occurs in response to activation of receptors with intrinsic tyrosine kinase activity, such as growth-factor receptors. However, the positive feedback amplification loop created by the BCR signalling complex provides an additional level of regulatory control, the importance of which is instructively illustrated by comparison of the circumstances leading to BCR engagement with those leading to growth-factor-receptor engagement. Co-evolution of interactions between growth factors and their receptors in the tightly regulated context of an organism has lead to many receptors which respond to a single interaction with one or a few types of ligand within a restricted range of ligand concentrations. Because the receptor–ligand interaction affinity is consistent, and the concentration of ligand is controlled, additional mechanisms for tuning receptor response magnitude are typically not required. By contrast, positive-feedback amplification of PLCγ2 activation after BCR engagement creates a means for co-receptors to tune the BCR signal amplitude over a vastly wider dynamic range, and is one key way in which B cells are able to respond to an array of ligands of varying abundance and interaction affinities, and to modulate their response depending on previous experience.

One of the most important examples of BCR signal amplitude tuning by a co-receptor is the role the B-cell specific surface molecule CD19 has in potentiating BCR signals to complement-coated pathogens (Figure 2) ^43–45. Complement-coated pathogens enhance the recruitment of a trimolecular complex of CD81–CD19–CD21 (the latter being a receptor for the complement component C3d) into proximity to activated BCRs, leading to the increased recruitment and activation of the guanine-nucleotide-exchange factor (GEF) VAV and the p85 subunit of PI3K to the signalosome, and so synergistic activation of the p110 PI3K subunit^46,47. Available data favour a model whereby PI3K is initially recruited to and activated in the signalosome independently of VAV, and the subsequent recruitment of VAV (by immunoreceptor adaptors and/or co-receptors) facilitates further PI3K activation through the proximally located RAC1 (Figure 2B and C). The enhanced PI3K activation produced by recruitment of the CD81-CD19-CD21 complex into proximity to BCR “signalosomes” leads to a marked potentiation of PLCγ2 activation via the positive-feedback mechanism outlined above, and a larger and more sustained increase in cytosolic Ca²⁺. The signal-amplitude increase provided by the influence of the CD19 complex lowers the threshold of the B-cell response to complement-coated pathogens by two to three orders of magnitude⁴⁸, providing enhanced immune protection against pathogens at low abundance. These events are also paramount for efficient B-cell activation in the early germinal-centre reaction⁴⁹.

Limiting BCR-induced Ca²⁺signaling

A second important example of tuning of the BCR-signal amplitude is the negative-feedback regulation of B-cell activation that occurs in response to co-ligation of the low-affinity receptor for IgG (FcγRIIB1) with the BCR (Figure 3). FcγRIIB1 and BCR co-ligation attenuates the BCR signalling response in a way that markedly inhibits BCR-induced Ca²⁺ signals^50–52. FcγRIIB1 achieves this effect by recruiting SRC-homology-2-domain-containing inositol-5-phosphatase (SHIP)⁵³ and SH2-domain-containing protein tyrosine phosphatase 1 (SHP1) into close proximity to BCR calcium signalling effectors⁵⁴. Recruitment of SHIP phosphatase by colligation of FcγRIIB1 with the BCR is thought to provide a signal attenuating effect by locally enhancing the rate of degradation of PtdIns(3,4,5)P₃ to PtdIns(3,4)P₂ ^31,55, whereas recruitment of SHP1 is thought to result in dephosphorylation of CD19 on residues that are required to mediate PI3K association and activation^56,57. Although each of these mechanisms would contribute to interruption of the positive-feedback loop for PLCγ2 activation while allowing other tyrosine-kinase-dependent aspects of the BCR signal to continue unaltered, SHIP recruitment appears to provide the majority of the signal attenuation effect, as FcγRIIB1 inhibitory signalling is significantly compromised in SHIP-deficient cells, and generally intact in SHP1-deficient cells ^58,59. Overall, FcγRIIB1-dependent attenuation of PLCγ2 activation leads to decreased InsP₃ production, such that InsP₃ is not able to accumulate to a level capable of sustaining SOCE, thereby resulting in a relatively specific block in extracellular Ca²⁺ entry, and consequently a transient Ca²⁺ signal^31,55.

Figure 3. Surface-receptor mechanisms for modulation of BCR-induced Ca²⁺ signals.

The CD19 complex is a positive regulatory receptor. Its active recruitment of into B-cell receptor (BCR) signaling complexes acts to enhance p110 phosphoinositide-3-kinase (PI3K) activation during BCR signaling. FcγRIIb1 and CD22 are both negative regulators of BCR-mediated calcium signals. When actively recruited into BCR signaling complexes, these receptors both act by recruiting the SH2-containing tyrosine phosphatase-1 (SHP1) and the SH2-containing inositol-5'-phosphatase (SHIP). However, the majority of the impact of FcγRIIb1 on Ca²⁺ signaling is mediated through the SHIP-dependent attentuation of PI3K signaling, while CD22 appears to have its most significant effect on Ca²⁺ signaling through SHP-1-dependent enhancement of PMCA-mediated Ca²⁺ export.

In addition to modulation of PLCγ activation as a means of influencing cytosolic Ca²⁺ concentration, a third B-cell surface co-receptor, CD22, has been implicated in the direct modulation of cytosolic Ca²⁺ concentration after BCR stimulation (FIG. 3)^60–62. Recent studies suggest that CD22 acts to specifically enhance Ca²⁺ extrusion by co-localizing the tyrosine phosphatase SHP1 with PMCA4⁶³. This regulatory mechanism may be facilitated by co-localization of PMCA pumps and CRAC channels¹⁸, as discussed above. CD22-mediated regulation of BCR Ca²⁺ signals acts as a brake on BCR-mediated B-cell activation, and so dampens humoral immune responses in vivo^60,64. Its significance can be seen in the effects of loss of CD22 function, which include hyper-responsive B cells^61,62, altered developmental fate of splenic transitional B cells leading to an absence in marginal-zone B cells and their immediate precursors ^65–67, and the promotion of humoral autoimmune phenomena ^64,68,69.

Interpretation and modulation of BCR-induced Ca²⁺ signals

Implicit in the above model for regulation of BCR-induced Ca²⁺ signalling is that extrinsic regulation through co-receptors targets the activation of PLCγ2 and subsequent production of InsP₃, with secondary regulation occurring through the influence of PMCA pumps on the rate of Ca²⁺ export. However, it is important to recognize that other aspects within the ‘cell context’ [Au: OK?] besides co-receptor activity may have important influences on how a given level of BCR-mediated PLCγ activation is ‘interpreted’ into a Ca²⁺ signal. Such cell context aspects may include various intrinsic or alternative inputs to PLCγ activation and InsP₃ metabolism that can modulate PLCγ activation through the BCR or other PLC-linked receptors. Supplementary Table 1 details several such influences and their potential relevance to B-cell physiology. In addition, recent advances in ion channel biology and Ca²⁺ transport physiology have identified several new channels and regulatory mechanisms that may directly or indirectly influence cytosolic Ca²⁺ and BCR-mediated Ca²⁺ signals (FIG. 4).

Figure 4. Mechanisms for regulation of cytosolic Ca²⁺ in B cells.

Pathways with known or potential ability to influence cytosolic Ca²⁺ in B-cells are grouped by colour according to their mechanism of influence.

Regulation of Ca²⁺ entry via modulation of plasma membrane potential

Although often overlooked in non-excitable cells, the contribution of membrane potential to the driving force for Ca²⁺ entry is particularly relevant to Ca²⁺ entry through highly Ca²⁺-selective low-conductance channels, such as CRAC channels (reviewed in ^2,3,6). B cells express both voltage-operated and Ca²⁺-activated K⁺ channels^2,3,70–73, and distinct B-cell subsets (for example naive B cells compared with class-switched memory B cells) have been shown express distinct ratios of different types of K⁺ channels, consistent with a role for K⁺ channel modulation in the regulation of B-cell physiology⁷⁴. An additional means for modulating membrane potential has recently been discovered in the form of the monovalent cation permeant transient receptor potential melastatin-related (TRPM) family members, TRPM4 and TRPM5. These channels are activated in response to increases in cytosolic Ca²⁺ concentrations and modulated by membrane lipid composition^75–77, and serve as a means to depolarize the cell membrane. Their activation has been shown to limit the driving force for Ca²⁺ entry during Ca²⁺ signals in T cells and mast cells^78,79. Consistent with TRPM4 and/or TRPM5 having a similar function in B cells, both channels have been reported to be expressed in bone marrow, splenic and lymph node B cells⁷⁷. The existence of two distinct methods (K⁺ and TRPM channels) for subset-dependent and dynamic regulation of membrane potential in B cells suggests that membrane potential modulation may be an unexpectedly rich mechanism for the regulation of B-cell Ca²⁺ signalling (Figure 5).

Figure 5. Membrane potential modulation of Ca²⁺ signals.

Membrane potential can be differentially and dynamically regulated through K⁺ channel activity and TRPM4/TRPM5 monovalent cation channel activity

Top panel: In resting cells, membrane potential is set at a highly negative potential through the activity of voltage–operated K⁺ channels. The negative membrane potential is generated through the outward movement of positively charged K⁺ ions, leaving unpaired intracellular negative charges (not shown). TRPM4/TRPM5 channels are thought to be closed in resting cells.

Middle panel: When a cell stimulus leads to activation of the store-operated Ca²⁺ entry (SOCE) pathway, opening of highly Ca²⁺-selective CRAC channels allows Ca²⁺ to pass across the plasma membrae. The current carried by CRAC channels is very small, and thus the entering positively charged Ca²⁺ ions have little direct influence on the membrane potential. However, the rise in cytosolic Ca²⁺ resulting from CRAC channel opening may influence the gating of both Ca²⁺-activated K⁺ channels and TRPM4/TRPM5 channels.

Bottom left panel: In cells expressing significant numbers of Ca²⁺-activated K⁺ channels, their activation serves to maintain a highly negative membrane potential. Thus, the cell maintains a strong sustained driving force for Ca²⁺ entry and a large magnitude sustained Ca²⁺ signal results.

Bottom right panel: In cells expressing significant numbers of TRPM4/TRPM5 monovalent cation channels, their activation allows Na+ entry, thus providing cytosolic positive ions to neutralize the unpaired internal charges. This diminishes the negative membrane potential, thereby attenuating the driving force for Ca²⁺ entry, and resulting in a reduced magnitude Ca²⁺ signal.

Beyond Ca²⁺-dependent activation of K⁺ and TRPM4/TRPM5 channels, the activity of these channels is variously accessible to receptor mediated stimuli, metabolic mediators, and environmental influences that modulate their gating properties, providing multiple mechanisms through which dynamic regulation of membrane potential could influence B-cell Ca²⁺ signaling.

Ca²⁺ entry through non-SOCE mechanisms

In B cells, data have also been generated that implicate various non-SOCE mechanisms in Ca²⁺ entry. The non-selective Ca²⁺-permeable transient receptor potential channel (TRPC) family channel, TRPC1, has been implicated in the BCR-mediated Ca²⁺ response⁸⁰. While a DT40 B-cell-line deficient in TRPC1 maintains normal levels of receptor-mediated InsP₃ production, BCR-induced Ca²⁺ release and sustained Ca²⁺ elevations are diminished, and Ca²⁺ entry induced by thapsigargin (a compound which irreversibly inhibits SERCA-pump function, and thus leads to passive ER store depletion) is diminished⁸⁰. The ability of several TRPC proteins to bind to InsP₃R isoforms and facilitate their optimal activation^81–84, and the demonstration of interactions between STIM proteins and both TPRC and ORAI subunits ^80,85,86, suggest that the defective transient and sustained Ca²⁺ responses in TRPC-deficient B cells could be the result of alterations in ER or plasma-membrane cell physiology, which are dependent on TRPC–InsP3R–STIM–ORAI subunit interactions. However, precisely how TRPC proteins could influence STIM1–ORAI interactions during ER-store depletion or other contexts, and the cell biological significance of plasma membrane/ER communication achieved through TRPC–InsP3R–STIM–ORAI-subunit interactions, are not presently well understood. An alternative or additional mechanism for the influence of TRPC1 on cytosolic Ca²⁺ in B cells is that TRPC channels may function in close coupling with Na⁺/Ca²⁺ exchangers such as Na⁺/Ca²⁺ exchanger-1 (NCX1) to generate extracellular Ca²⁺ entry via reverse mode exchange with permeating Na⁺ ions⁸⁷.

Other channels potentially found in B cells include other TRPC members, various members of the transient receptor potential-vanilloid receptor-1-related (TRPV) family^88,89, TRPM2 and TRPM7^90–92, and undefined stretch-activated ion channels and eicosanoid-activated channels ^93–95. In addition, convincing evidence for expression of mRNA for splice variants of α-subunits of classic voltage-operated Ca²⁺ channels has been reported in B cells and T cells, although electrophysiological evidence of voltage-operated Ca²⁺ currents is lacking, suggesting that these splice variants may not be voltage-operated, but instead must have alternative gating mechanisms^96–98. Further exploration of the function and physiological regulatory mechanisms of these various channels should begin to provide insight into their distinct roles in B-cell Ca²⁺ signalling and physiology.

Ca²⁺ signalling and cell-fate choice

With the accumulation of an increasingly complete picture of the network of inputs regulating B-cell cytosolic Ca²⁺ concentration and signalling, an important future direction for B-cell immunologists is to understand and define how these inputs influence B-cell physiology and immune responses. Elegant studies by Lewis, Goodnow and colleagues have shown that PLC activation and cytosolic Ca²⁺ differentially impact several key transcription factor pathways in B cells^99,100, as exemplified by the NFAT pathway, which is involved in transcriptional activation of cytokine genes such as interleukin-4 (IL-4), and the NF-κB pathway, which in B cells targets proteins such as B-cell lymphoma-6 (BCL-6) that are involved in protecting the cell from apoptosis and supporting proliferation (Figure 6A). The NFAT pathway is activated in response to sustained Ca²⁺ elevation or oscillations through induction of phosphorylation-dependent nuclear localization of NFAT family transcription factors (reviewed in ^101–103), By contrast, the NF-κB pathway is activated through DAG- and Ca²⁺-dependent degradation of the inhibitor of NF-κB (IκB) protein (reviewed in ¹⁰⁴), and exhibits strong dependence on peak amplitude, rather than duration, of a Ca²⁺ signal. These observations have provided a useful context to understand how differences in the level of BCR activation, and therefore PLCγ activity, may produce markedly different B-cell-fate choices in the context of tolerizing versus non-tolerizing antigen exposure: non-tolerizing antigens induce large sustained Ca²⁺ responses which are able to effectively activate both pathways, whereas tolerizing antigens induce low-level sustained responses which are able to active NFAT, but not NF-κB, pathways (see FIG. 6 and ^99,100).

Figure 6. Potential mechanisms for Ca²⁺-dependent modulation of B-cell fate determination.

Based on present models for Ca²⁺-dependent regulation of transcription factors such as Nuclear Factor of Activated T-cells (NFAT) and Nuclear Factor kb (NF-κB), membrane potential modulation and direct Ca²⁺ entry through non-SOCE pathways are predicted to have the capacity to significantly alter B-cell-fate choice.

A: Strong B-cell receptor (BCR) activation fully activates both NF-κB (through inhibitor of kb (IκB) degradation) and NFAT (through calcineurin phosphatase activation) pathways.

B: Modulation of membrane potential attenuates late Ca²⁺ entry, selectively influencing NFAT activation.

C: A weak but sustained BCR activation signal activates only the NFAT pathway.

D: Late activation of a direct Ca²⁺ entry pathway restores NF-κB pathway activation despite a weak BCR activation signal.

While influences on the events leading to ER Ca²⁺ store depletion, such as those outlined in Table 1, [Au: OK?] can be directly integrated with the observations of Lewis, Goodnow and colleagues (such as through their quantitative effects on the BCR Ca²⁺ signal), future studies will also be challenged to integrate more recently defined Ca²⁺ regulatory mechanisms with the described properties of the NFAT and NF-κB pathways. Two informative examples are considered in FIG 6. For B cells receiving a strong signal through the BCR, one normally expects activation of both NFAT and NF-κB pathways (FIG.6a), and selection into a maturation pathway appropriate for a high-affinity BCR–antigen interaction. However, if the same B-cell were in a microenvironment in which K⁺ channel deactivation, non-selective cation-channel activation or both had resulted in cell membrane depolarization, SOCE could be substantially compromised due to a reduction in the driving force for Ca²⁺ entry. A reduction in SOCE would thus lead to a transient Ca²⁺ signal, selective loss of NFAT pathway activation, and a cell-fate choice appropriate to a lower affinity BCR–antigen interaction such as death by neglect (as in FIG.6b). In a second example, one can imagine a B-cell in a microenvironment in which a low-level BCR signal is generated, thus leading to a low-level sustained Ca²⁺ signal capable of NFAT, but not NF-κB, nuclear translocation (FIG.6c). If this B-cell subsequently receives a signal leading to Ca²⁺ entry through an alternative non-SOCE pathways, NF-κB activation may be recovered, and the B-cell would be diverted into a pathway normally associated with higher affinity BCR–antigen interactions (FIG.6d). In other cases, sustained elevation of cytosolic Ca²⁺ concentration, occurring as the result of an “excessive” level of receptor engagement or non-SOCE Ca²⁺ entry in a particular microenvironment, could be of sufficient magnitude to lead to mitochondrial Ca²⁺ accumulation and eventual activation-induced cell death^105,106 — a mechanism that may occur in some forms of B-cell negative selection^107–109. Finally, an additional veneer may be added to the above cell-intrinsic regulatory mechanisms, in that increased cytosolic Ca²⁺ concentration in T-cells has been observed to serve as a “stop” signal, resulting in a marked decrease in the motility of an activated cell. Thus, longer duration elevations of cytosolic Ca²⁺ are predicted to result in longer residence times in a particular microenvironment, potentially leading to the immobilization of activated T cells within regulatory niches in peripheral tissues and lymphoid organs¹¹⁰. If a similar mechanism operates in B-cells, even a transient compromise of a strong BCR-dependent Ca²⁺ signal could result in the release of a B cell from a stimulatory environment; conversely, a strong elevation of cytosolic Ca²⁺ might be sufficient to immobilize a B cell in a particular environment with a normally poorly stimulatory BCR ligand.

Overall, the emerging picture is that while the BCR signal is a crucial determinant of B-cell fate, dynamic inputs from the B-cell membrane potential and direct Ca²⁺ entry pathways have the potential to significantly influence developmental and activation events within an individual B-cell. Understanding how different B-cell subsets differ in these various Ca²⁺ regulatory inputs, and how such inputs are influenced by different microenvironmental stimuli, will substantially extend our understanding of B-cell-fate choices in the context of both regulatory homeostasis and active immune responses.

Summary

B cells receive information that is crucial to their physiology and function through cytosolic Ca²⁺ signals. One of the most important inputs to B-cell cytosolic Ca²⁺ is derived from BCR-mediated positive-feedback loop activation of PLCγ2. If InsP₃ production from this mechanism reaches the threshold required to activate SOCE, sustained elevations in cytosolic Ca²⁺ are supported. The sensitivity of the positive feedback mechanism renders the amplitude and duration of the resulting changes in cytosolic Ca²⁺ subject to regulation by B-cell surface molecules at the level of PLCγ activation and at the level of Ca²⁺ export, allowing the BCR signal to be tuned over a wide dynamic range. Other microenvironmental stimuli may also indirectly influence B-cell cytosolic Ca²⁺ concentration through contributions to the total pool of activated PLC and InsP₃, or through direct effects on Ca²⁺ fluxes mediated by transporters or channels. Future work will be challenged to integrate these diverse inputs with present models of how Ca²⁺-dependent signals regulate cell-fate choice in specific immunological contexts.

Glossary terms

Conformational coupling

Refers to ion channel gating regulation by interaction with another protein, as opposed to gating through changes in plasma membrane potential or interaction with a diffusible second messenger.

Germinal centre

Located in peripheral lymphoid tissues (for example, the spleen or lymph nodes), these structures are sites of B-cell proliferation and selection for clones that produce antigen-specific antibodies of higher affinity.

Somatic hypermutation

A unique mutation mechanism that is targeted to the variable regions of rearranged immunoglobulin gene segments. Combined with selection for B cells that produce high-affinity antibody, somatic hypermutation leads to affinity maturation of B cells in germinal centres.

TEC-kinase family

A third class of protein-tyrosine kinases that is required for the activation of haematopoietic cells — the first and second classes being the SRC-and SYK-family kinases. The TEC-family kinase prototypes are ITK (IL-2-inducible T-cell kinase) in T cells and BTK (Bruton’s tyrosine kinase) in B cells. Among other functions, TEC kinases play a crucial role in the activation of phospholipase C enzymes after immunoreceptor ligation.

Marginal-zone B cells

A static, mature B-cell subset that is enriched mainly in the marginal zone of the spleen, which is located at the border of the white pulp.

MEMBRANE POTENTIAL

The charge difference (measured in mV) between the two surfaces of a biological membrane that arises from the different concentrations of ions such as H⁺, Na⁺ or K⁺ on either side. The Na⁺/K⁺-ATPase creates a membrane potential by using the energy stored in ATP to maintain a low concentration of Na⁺ and a high concentration of K⁺ in the cell, against a higher concentration of Na⁺ and a lower concentration of K⁺ on the outside.

Channel conductance

Conductance is defined as current divided by voltage, and in biological systems this means the current flowing across a biological membrane divided by the electrical potential across that membrane. When used in reference to a single open ion channel, (single channel conductance), it provides a measure of the amount of current a single open ion channel is able to carry. The single channel conductance is usually independent of the plasma membrane potential, and thus characteristic of that particular ion channel. Individual ion channels with small single channel conductances carry less current at a given membrane potential than those with large single channel conductances.

VOLTAGE-OPERATED CHANNELS

Plasma-membrane ion channels whose gating is regulated by changes in plasma membrane potential.

Stretch-activated ion channels

Plasma-membrane ion channels whose gating is regulated by changes in plasma membrane "stretch" - i.e. forces which are directed within/parallel to the plane of the plasma membrane.

Eicosanoids

Eicosanoids are fatty-acid derivatives, primarily derived from arachidonic-acid precursors, that have a wide variety of biological activities. There are four main classes of eicosanoid — the prostaglandins, prostacyclins, thromboxanes and leukotrienes — derived from the activities of cyclooxygenases and lipoxygenases on membrane-associated fatty-acid precursors.

Non-selective cation channels

Ion channels which exhibit a significant selectivity towards a single type of cation are generally designated according to that ionic selectivity - e.g. K⁺ channel or Ca²⁺ channels. Cation channels which exhibit little selectivity between monovalent cations, or between monovalent cations and one or more divalent cations such as Ca²⁺ or Mg²⁺ are typically grouped together and referred to as non-selective cation channels.

Reverse mode Na⁺/Ca²⁺ exchange

The major physiological role for Na⁺/Ca²⁺ exchangers is thought to be removal of Ca²⁺ from the cytosol in exchange for extracellular Na⁺. If ionic conditions in the neighborhood of an exchanger are appropriate, it is possible for the exchanger to operate in reverse and exchange intracellular Na⁺ for extracellular Ca2⁺ - this is referred to as the reverse mode of the exchanger.