B-Cell Receptor Signalling and the Germinal Centre Reaction

The germinal centre is where antibody quality is made. A naive B cell that binds antigen does not become a source of protective, high-affinity IgG by accident. It passes through a tightly choreographed sequence: receptor signalling calibrated by three separate brakes, survival signals from the TNF superfamily, a licensing handshake with a T-follicular-helper cell, and then weeks of iterative mutation and selection inside a specialised anatomical structure. Understanding that sequence is the difference between a vaccine adjuvant that works and one that does not — and between an autoimmune model that phenocopies human disease and one that does not.

Signal 1: the B-cell receptor and the three brakes

Membrane IgM and IgD bind antigen but have cytoplasmic tails only three residues long. All signalling is delegated to the Igα (CD79a)/Igβ (CD79b) heterodimer, whose ITAMs are phosphorylated by the Src-family kinase Lyn. Dual phosphorylation creates a docking site for Syk, which phosphorylates the adaptor BLNK/SLP-65. BLNK is the assembly platform: it brings Btk — the kinase whose loss causes X-linked agammaglobulinaemia — together with PLCγ2.

What happens next splits into three arms. PLCγ2 cleaves PIP₂ into IP₃ and DAG. IP₃ empties the ER calcium stores, and sustained Ca²⁺ activates calcineurin, which dephosphorylates NFAT and lets it enter the nucleus. DAG recruits PKCβ, which phosphorylates CARMA1 and nucleates the CBM complex with BCL10 and MALT1, activating IKKβ and releasing NF-κB. Separately, PI3Kδ generates PIP₃ and recruits AKT, which sets the survival threshold through Bcl-2, Mcl-1 and c-Myc, and inactivates FOXO1.

The part most often skipped in textbook summaries is the calibration. Three independent brakes decide how much of that cascade actually fires. CD22 (Siglec-2) carries ITIMs that recruit the phosphatase SHP-1 and dephosphorylate Syk. FcγRIIB — the only inhibitory Fc receptor in the genome — recruits SHIP-1 whenever secreted IgG re-engages the same antigen, closing a negative feedback loop on the whole response. And PTEN continuously strips the 3-phosphate from PIP₃. Loss of any of these lowers the threshold for autoreactivity, which is why FcγRIIB polymorphisms recur in lupus genetics.

Pushing in the opposite direction is the complement co-receptor complex. When C3d is deposited on antigen, it is bound by CD21 (CR2), which is physically coupled to CD19 and CD81. CD19 recruits PI3Kδ directly, lowering the activation threshold by orders of magnitude. This is the molecular reason complement-fixing adjuvants are so effective, and why CD19 and CD21 are worth measuring alongside the BCR itself.

Signal 2: survival factors and the T–B handshake

Antigen alone is not enough. Independently of the BCR, the TNF-superfamily ligands BAFF/BLyS and APRIL act through BAFF-R, TACI and BCMA to set the size of the peripheral B-cell pool and, later, to keep plasma cells alive. Serum BAFF is one of the more informative readouts in autoimmune models for exactly this reason.

The decisive event is the handshake at the T–B border. Activated B cells upregulate CXCR5 and migrate toward CXCL13, where a T-follicular-helper cell delivers CD40L to CD40, ICOS to ICOSL, and IL-21, IL-4 and IL-6 to their receptors. Crucially the synapse is bidirectional: the B cell presents BCR-internalised peptide on MHC class II with CD86 co-stimulation, and PD-1 on the Tfh side limits how much help is delivered. Block CD40L and no germinal centre forms at all — the single cleanest loss-of-function phenotype in the whole pathway.

Inside the germinal centre: dark zone, light zone and AID

Bcl-6 is the switch that opens the germinal-centre state. It represses Blimp-1, the DNA-damage response and cell-cycle checkpoints, allowing a B cell to tolerate deliberate mutation of its own genome. Pax5 and Bach2 hold B-cell identity in place, while S1PR2 confines the reaction to the follicle centre.

The structure then polarises. CXCR4 holds centroblasts in the CXCL12-rich dark zone, where they divide faster than almost any cell in the body and where AID (AICDA) deaminates cytosine to uracil across the transcribed variable region. UNG excises the uracil, APE1 nicks the abasic site, and MSH2/MSH6 recruit the error-prone polymerase Pol η. The result is somatic hypermutation at roughly 10⁻³ mutations per base per division — a million-fold above the background rate.

CXCR5 then pulls centrocytes into the light zone, where the mutated receptor competes for limiting antigen displayed on follicular dendritic cells. This is the selection step, and its logic is worth stating precisely: higher-affinity clones capture more antigen, internalise it, present more peptide on MHC-II, and therefore win more CD40L and IL-21 help. Winners re-express c-Myc and cycle back to the dark zone for another round; losers die by Fas-dependent and neglect-driven apoptosis. Most centrocytes die on every pass. The germinal centre is, structurally, a selection machine with a very high failure rate.

The output: class switching, plasma cells and memory

The same AID activity acting at the switch regions of the IgH locus creates staggered double-strand breaks that non-homologous end joining resolves into class-switch recombination, exchanging the constant region under cytokine instruction — IL-4 for IgG1 and IgE, TGF-β and APRIL for IgA.

Cells that break the Bcl-6 programme upregulate IRF4, STAT3-driven Blimp-1 and XBP1, expanding the secretory ER and becoming plasmablasts and then CD138⁺ long-lived plasma cells that home to a marrow niche of CXCL12, APRIL and IL-6 and can secrete IgG for decades. Cells that keep Bach2 high instead become CD27⁺ memory B cells. The IRF4 dose is the pivot: low sustains the germinal centre, high forces the plasma-cell exit.

Key targets and matching reagents

Studying the germinal centre in vivo

Because so much of this axis is receptor- and cytokine-based, it is unusually tractable with functional-grade antibodies. In practice, experiments fall into three groups.

1. Depleting or tracking the B-cell compartment

Anti-CD19, anti-CD20 and anti-CD45R/B220 (RA3-6B2) deplete or label B cells; anti-CD22 (Cy34.1), anti-CD24 and anti-CD38 (NIMR5) address maturation and germinal-centre subsets; anti-CD16/CD32 (2.4G2) blocks Fc receptors before staining, and matched IgM isotype controls establish the baseline. Low endotoxin is not optional here — LPS is itself a B-cell mitogen, so a contaminated preparation will confound the very readout being measured.

2. Blocking help and co-stimulation

Anti-CD40, anti-CD86, anti-MHC-II (M5/114), anti-ICOS (CD278), anti-ICOSL (CD275), anti-CD4 (GK1.5) and anti-PD-1 (29F.1A12) let you dissect which arm of the T–B synapse a given phenotype depends on. Interrupting CD40L–CD40 abolishes germinal centres outright; blocking ICOS/ICOSL degrades them progressively — a useful contrast when you need a partial rather than a binary phenotype.

3. Neutralising the cytokine environment

Anti-IL-4 (11B11), anti-IL-6 (MP5-20F3), anti-IL-21R (4A9) and anti-CXCR4 (12G5) target the signals that determine isotype, plasma-cell commitment and dark-zone retention respectively. Pair these with serum ELISA readouts for total IgM, IgG, IgA and IgE, plus BAFF and APRIL, to convert a blocking experiment into a quantitative one.

Featured products for this pathway

Frequently asked questions

For research use only. Not for use in diagnostic or therapeutic procedures.