17.10.04 · mol-cell-bio / immunology

Antibody structure and diversity: V(D)J recombination, affinity maturation, and isotype switching

stub3 tiersLean: nonepending prereqs

Anchor (Master): Janeway's Immunobiology, 10th ed. (2022), Ch. 5

Intuition Beginner

Antibodies are Y-shaped proteins that patrol your blood and tissues, searching for specific pathogens. Each antibody has two identical arms at the top of the Y, and these arms carry a unique molecular tip that fits one specific target — like a lock that only one key can turn. Your body can produce billions of different antibodies, each recognising a different molecule, and this enormous variety is what lets your immune system respond to threats it has never encountered before.

The secret to making so many different antibodies lies in a remarkable genetic trick. The genes that encode antibodies are not stored as complete, ready-to-use instructions. Instead, they come in scattered pieces called gene segments. As a B cell develops, it randomly selects and stitches together a few of these segments — a process called V(D)J recombination — to build a unique antibody gene. Think of it like choosing one topping from each of three columns on a menu: even a modest number of options in each column produces a huge number of possible combinations.

Once a B cell has made its antibody and encountered the pathogen it recognises, the immune system does not stop there. Inside specialised structures called germinal centers, B cells undergo further rounds of mutation and selection. Antibodies that bind more tightly to their target are favoured, and their B cells proliferate. This process of affinity maturation progressively refines the antibody response, producing ever more effective molecules over the course of an infection.

Finally, the immune system can change the stem of the Y — the constant region — while keeping the same targeting tips. This is called isotype switching (or class switch recombination). It lets the same pathogen-specific antibody be deployed in different roles: as IgM for an initial broad response, as IgG for long-term blood immunity, as IgA to protect mucosal surfaces, or as IgE to fight parasites.

Visual Beginner

The diagram is divided into three panels. The left panel shows a single antibody molecule: two identical heavy chains (long polypeptides shown in blue) and two identical light chains (shorter polypeptides shown in yellow), all linked by disulfide bonds into the characteristic Y shape. The two arms of the Y are the Fab regions (fragment, antigen-binding), each carrying a variable domain with a cleft that binds antigen. The stem is the Fc region (fragment, crystallisable), which recruits other immune components. The variable domains are highlighted in a darker shade to show the three complementarity-determining loops (CDRs) that form the antigen-binding site.

The centre panel depicts the heavy chain gene locus on chromosome 14 before and after V(D)J recombination. The germline configuration shows multiple V segments (approximately 40 functional), 23 D segments, and 6 J segments arranged in clusters. Arrows indicate how one V, one D, and one J segment are randomly selected, the intervening DNA is deleted, and the chosen segments are joined to form a complete variable-region exon upstream of the constant-region genes (C-mu, C-delta, C-gamma, C-epsilon, C-alpha).

The right panel illustrates V(D)J recombination at the DNA level. The RAG1/RAG2 enzyme complex binds recombination signal sequences (RSS) flanking the V and D segments, creates a double-strand break, and the two broken DNA ends are repaired by non-homologous end joining (NHEJ). The enzyme TdT adds random N-nucleotides at the junctions, introducing additional sequence diversity beyond what combinatorial selection alone provides.

Worked example Beginner

Trace the development of a single B cell from bone marrow to a plasma cell secreting high-affinity IgG antibody against influenza haemagglutinin.

Step 1: V(D)J recombination in the bone marrow. A developing B cell rearranges its heavy chain locus first. It selects V segment 12, D segment 3, and J segment 4 from the available options. The RAG enzymes cut the DNA, and TdT adds 6 random nucleotides between V and D, and 4 random nucleotides between D and J. The resulting V12-(N)6-D3-(N)4-J4 exon encodes a unique heavy chain variable region. Next, the light chain locus is rearranged similarly (choosing one V and one J segment, with no D segment for light chains).

Step 2: Surface IgM expression. The successfully rearranged heavy chain pairs with the light chain, and the B cell displays IgM antibody (with mu constant region) on its surface. This naive B cell now patrols the blood and lymphoid organs, carrying its unique receptor.

Step 3: Antigen encounter in a lymph node. The B cell encounters influenza haemagglutinin protein on the surface of a follicular dendritic cell in a lymph node. Its surface IgM binds haemagglutinin. The B cell internalises the antigen, processes it, and presents haemagglutinin peptides on MHC class II to a helper T cell.

Step 4: Germinal center reaction and affinity maturation. With T cell help, the B cell enters a germinal center and proliferates rapidly. Its antibody gene undergoes somatic hypermutation: the enzyme AID introduces point mutations into the variable region at a rate approximately one million times higher than the normal background mutation rate. B cells producing antibodies with higher affinity for haemagglutinin are selected by follicular dendritic cells and receive survival signals from T cells. B cells with lower-affinity antibodies die by apoptosis.

Step 5: Class switch recombination. The selected B cell switches from producing IgM to IgG. The enzyme AID initiates DNA breaks in the switch regions upstream of the constant-region genes. The DNA between the mu switch region and the gamma switch region is deleted, and the variable exon is now joined to the C-gamma gene. The B cell differentiates into a plasma cell secreting high-affinity anti-haemagglutinin IgG.

Check your understanding Beginner

Exercise (easy, multiple choice).

Which process generates the initial diversity of antibody variable regions by randomly selecting and joining gene segments?

A. Somatic hypermutation B. V(D)J recombination C. Class switch recombination D. Alternative splicing

Hint

Think about which process occurs first in the bone marrow, before the B cell has encountered any antigen. This process assembles a complete antibody gene from scattered pieces.

Answer

B. V(D)J recombination. This process randomly selects V, D, and J gene segments from the heavy chain locus and V and J segments from the light chain locus, joining them to create a unique antibody variable region. Somatic hypermutation (A) further diversifies the antibody after antigen exposure. Class switch recombination (C) changes the constant region, not the variable region. Alternative splicing (D) produces membrane-bound versus secreted antibody forms but does not generate sequence diversity.

Exercise (easy, multiple choice).

What does class switch recombination change about an antibody?

A. Its antigen-binding specificity B. The variable region sequence C. The constant region, changing the antibody class (e.g., IgM to IgG) D. The number of antigen-binding sites

Hint

The "class" in class switch recombination refers to the antibody isotype — IgM, IgG, IgA, IgE. Which part of the antibody determines its isotype?

Answer

C. Class switch recombination replaces the heavy chain constant region (e.g., C-mu for IgM) with a different constant-region gene (e.g., C-gamma for IgG, C-alpha for IgA, C-epsilon for IgE) while preserving the same variable region. This means the antibody retains its antigen-binding specificity but gains different effector functions — IgG activates complement and crosses the placenta, IgA is secreted at mucosal surfaces, IgE triggers allergic and anti-parasite responses.

Formal definition Intermediate+

An antibody (immunoglobulin, Ig) is a tetrameric protein consisting of two identical heavy chains (approximately 50 kDa each) and two identical light chains (approximately 25 kDa each), linked by interchain disulfide bonds and non-covalent interactions. Each chain is composed of a series of immunoglobulin domains — approximately 110-amino-acid folds with a characteristic beta-sandwich structure stabilised by an intrachain disulfide bond.

Domain architecture

Each heavy chain contains one variable domain ( $V_{H}$ ) and three or four constant domains ( $C_{H} 1$ , $C_{H} 2$ , $C_{H} 3$ , and $C_{H} 4$ in IgM and IgE). Each light chain contains one variable domain ( $V_{L}$ ) and one constant domain ( $C_{L}$ ). Light chains exist in two isotypes: kappa ( $κ$ , encoded on chromosome 2 in humans) and lambda ( $λ$ , encoded on chromosome 22).

The variable domains of one heavy chain and one light chain pair to form the antigen-binding fragment (Fab). Within each variable domain, three loops of hypervariable sequence form the complementarity-determining regions (CDRs): CDR1, CDR2, and CDR3. CDR1 and CDR2 are encoded within the V gene segment itself, while CDR3 spans the V-D and D-J junctions in the heavy chain (or the V-J junction in the light chain) and is the most diverse region of the antibody. The six CDRs (three from $V_{H}$ , three from $V_{L}$ ) form the antigen-binding site.

The constant domains of the heavy chain form the Fc region (fragment, crystallisable), which determines the effector functions of the antibody: complement activation, binding to Fc receptors on immune cells, and transport across the placenta or into secretions.

The heavy chain locus

The immunoglobulin heavy chain locus (IGH, chromosome 14q32 in humans) spans approximately 1.3 Mb and contains, from 5-prime to 3-prime:

Approximately 40 functional V gene segments (IGHV), each encoding the framework and CDR1/CDR2 of the heavy chain variable domain.
23 functional D gene segments (IGHD), encoding part of CDR3.
6 functional J gene segments (IGHJ), encoding the remainder of the variable domain up to the constant region.
A series of constant-region genes ( $C_{μ}$ , $C_{δ}$ , $C_{γ} 3$ , $C_{γ} 1$ , $C_{α} 1$ , $C_{γ} 2$ , $C_{γ} 4$ , $C_{ϵ}$ , $C_{α} 2$ ), each encoding a different antibody isotype.

The light chain loci

The kappa locus (IGK, chromosome 2p11) contains approximately 34-38 functional V $κ$ segments and 5 functional J $κ$ segments, with a single C $κ$ constant gene. The lambda locus (IGL, chromosome 22q11) contains approximately 29-33 functional V $λ$ segments and 4-5 functional J $λ$ -C $λ$ clusters. Light chain loci have no D segments; light chain rearrangement involves only V-to-J joining.

V(D)J recombination mechanism

V(D)J recombination is directed by recombination signal sequences (RSS) flanking each V, D, and J segment. An RSS consists of a conserved heptamer (CACAGTG) and nonamer (ACAAAAACC) separated by a spacer of either 12 bp (one helical turn) or 23 bp (two helical turns). The 12/23 rule dictates that recombination only occurs between an RSS with a 12-bp spacer and an RSS with a 23-bp spacer, ensuring that V joins to D (not V-to-V or D-to-D at the heavy chain locus).

RAG1/RAG2 cleavage. The RAG1/RAG2 complex (recombination-activating genes) recognises the RSS pairs and introduces a single-strand nick precisely at the heptamer-coding segment boundary. The 3-prime OH of the nick attacks the opposite strand, generating a hairpin at the coding end and a blunt signal end. The two signal ends are joined to form a signal joint (a circular excision product that is eventually lost from the genome). The two coding ends (hairpins) are opened and processed before joining.

Junctional diversity. The hairpin at each coding end is opened by Artemis (a nuclease activated by DNA-dependent protein kinase). The opening can be asymmetric, generating short single-stranded overhangs (P-nucleotides) when the overhang is filled in by a DNA polymerase. The enzyme TdT (terminal deoxynucleotidyl transferase) adds random, template-independent N-nucleotides (typically 0-20 per junction) to the coding ends before ligation. The combined P- and N-nucleotide addition at the V-D and D-J junctions (heavy chain) or V-J junction (light chain) generates enormous sequence diversity in CDR3. This junctional diversity contributes more to the total antibody repertoire than combinatorial selection of V, D, and J segments alone.

Non-homologous end joining (NHEJ). The processed coding ends are ligated by the classical NHEJ machinery (Ku70/Ku80, XRCC4, DNA ligase IV, XLF). The joining is imprecise: nucleotides may be added (by TdT and polymerases) or removed (by Artemis and other nucleases) at each junction, generating additional diversity. The resulting coding joint is retained in the genome as the rearranged variable-region exon.

Combinatorial and junctional diversity

The total potential diversity of the human antibody repertoire arises from three sources:

Combinatorial diversity. The number of possible V-D-J (heavy) and V-J (light) combinations. For the human heavy chain: approximately 40 V $\times$ 23 D $\times$ 6 J $\approx$ 5,520 combinations. For kappa: approximately 35 V $\times$ 5 J $\approx$ 175. For lambda: approximately 30 V $\times$ 5 J $\approx$ 150. The total combinatorial diversity from heavy-light pairing is approximately $5, 520 \times (175 + 150) \approx 1.8 \times 1 0^{6}$ . This is an underestimate because it counts only the number of functional V, D, and J segments and does not account for junctional diversity.

Junctional diversity. The random addition of N- and P-nucleotides at the V-D, D-J, and V-J junctions can generate approximately $1 0^{11}$ to $1 0^{14}$ distinct CDR3 sequences at each junction. This is the dominant source of antibody diversity.

Heavy-light chain pairing. Any heavy chain can pair with any light chain, approximately squaring the diversity (though not all pairings produce functional antigen-binding sites).

The combined theoretical diversity exceeds $1 0^{11}$ distinct antibodies — far more than the approximately $1 0^{7}$ to $1 0^{8}$ distinct B cells present in a human at any given time.

Counterexamples to common slips

Antibodies are only produced after infection. Antibodies are continuously produced by long-lived plasma cells and memory B cells from prior infections or vaccinations. Serum IgG has a half-life of approximately 21 days and provides ongoing passive immunity. Additionally, naive B cells produce low-affinity IgM that circulates at low levels.
Each B cell produces antibodies against many different antigens. Each mature B cell expresses a single antibody specificity (one heavy-light chain combination). This is the principle of allelic exclusion: once a functional heavy chain rearrangement is achieved, further heavy chain rearrangement is suppressed, and likewise for the light chain. A B cell is monospecific.
Somatic hypermutation occurs during V(D)J recombination. These are two distinct processes. V(D)J recombination occurs in the bone marrow during B cell development, before antigen exposure. Somatic hypermutation occurs in germinal centers after antigen exposure and T cell help, and it targets only the variable region of the already-rearranged antibody gene.

Key mechanism Intermediate+

Mechanism: Somatic hypermutation and the germinal center reaction.

Somatic hypermutation (SHM) introduces point mutations into the variable regions of rearranged antibody genes at a rate of approximately $1 0^{- 3}$ per base pair per generation — roughly one million times the spontaneous mutation rate in other genes. This extraordinary mutation rate is initiated by the enzyme activation-induced cytidine deaminase (AID).

AID targeting. AID is expressed exclusively in activated B cells within germinal centers. It deaminates cytosine to uracil in single-stranded DNA exposed during transcription. AID preferentially targets WRC hotspots (W = A/T, R = A/G, C = the target cytosine) and their reverse complements (GYW motifs) in the variable region. The resulting U mismatch triggers one of several repair pathways:

Error-prone repair. The uracil can be processed by two competing pathways. In the first, UNG (uracil-DNA glycosylase) removes the uracil, creating an abasic site that is resolved by error-prone translesion polymerases (Pol eta, Pol iota, Pol zeta) — introducing mutations at and around the original C residue. In the second, MSH2/MSH6 recognise the U mismatch and recruit error-prone polymerases directly. The combined activity of these pathways produces mutations at A/T bases as well as the originally targeted C/G bases, with the mutation spectrum concentrated in the CDRs.

Germinal center architecture. The germinal center is divided into two zones with distinct functions:

Dark zone. B cells (centroblasts) undergo rapid proliferation (approximately one division every 6-12 hours) while AID-mediated somatic hypermutation diversifies their antibody genes. Each division introduces approximately 1-2 mutations per variable region. Centroblasts do not display surface antibody and do not interact with antigen.

Light zone. B cells (centrocytes) migrate to the light zone, re-express surface antibody, and compete for antigen displayed on follicular dendritic cells (FDCs). FDCs retain intact antigen in the form of immune complexes (antigen-antibody-complement) on their surface, presented as iccosomes. Centrocytes with higher-affinity surface antibody capture more antigen from FDCs, process it, and present more peptide-MHC II complexes to follicular helper T cells ( $T_{F H}$ ). $T_{F H}$ cells deliver survival signals (CD40L-CD40 interaction, IL-21) to the B cells that present the most antigen — a direct selection for higher affinity.

Cyclic re-entry. Selected centrocytes can recycle back to the dark zone for further rounds of proliferation and mutation, then return to the light zone for another round of selection. This cyclic re-entry process allows iterative rounds of mutation and selection, progressively increasing antibody affinity over 2-3 weeks. Antibody affinity typically increases 10- to 10,000-fold during a primary immune response.

Class switch recombination. In parallel with SHM, AID also initiates class switch recombination (CSR). AID targets switch regions (S regions) — repetitive GC-rich DNA sequences upstream of each constant-region gene (except $C_{δ}$ ). AID deaminates cytosines in the S regions on both strands. The resulting lesions are processed into double-strand breaks by UNG and the mismatch repair machinery. The break in the donor S region (e.g., S-mu) is joined to the break in the acceptor S region (e.g., S-gamma1) by NHEJ, deleting the intervening DNA. The variable exon is now juxtaposed to the new constant-region gene (e.g., $C_{γ} 1$ ), and the B cell switches from IgM to IgG1 production.

CSR requires T cell help (CD40L-CD40 interaction and cytokine signals). The specific cytokine milieu determines which isotype is selected: IL-4 promotes switching to IgG1 and IgE; TGF-beta promotes switching to IgA; interferon-gamma promotes switching to IgG2a (in mice) and IgG1 in humans. Each isotype confers distinct effector functions on the antibody while preserving the same antigen-binding variable region.

Exercises Intermediate+

Exercise 1 (easy, multiple choice).

Which enzyme introduces somatic hypermutations in antibody variable regions?

A. RAG1 B. TdT C. AID (activation-induced cytidine deaminase) D. DNA polymerase delta

Hint

Somatic hypermutation occurs in germinal centers, not in the bone marrow. Which enzyme is expressed exclusively in activated germinal center B cells?

Answer

C. AID. Activation-induced cytidine deaminase deaminates cytosine to uracil in single-stranded DNA exposed during transcription of the antibody variable region. The resulting U mismatches are processed by error-prone repair pathways (involving UNG and MSH2/MSH6) into point mutations at approximately $1 0^{- 3}$ per base pair per generation. RAG1 (A) mediates V(D)J recombination, not somatic hypermutation. TdT (B) adds N-nucleotides during V(D)J recombination. DNA polymerase delta (D) is a replicative polymerase, not the error-prone polymerase responsible for SHM.

Exercise 2 (easy, multiple choice).

What is the 12/23 rule in V(D)J recombination?

A. Recombination occurs between gene segments separated by 12 or 23 base pairs. B. Recombination only joins an RSS with a 12-bp spacer to an RSS with a 23-bp spacer. C. Each V segment has 12 nucleotides and each J segment has 23 nucleotides. D. RAG1 and RAG2 are 12 kDa and 23 kDa proteins, respectively.

Hint

The 12/23 rule ensures that V joins to D and D joins to J, but V does not join directly to V. What feature of the recombination signal sequences enforces this ordering?

Answer

B. The 12/23 rule states that V(D)J recombination only occurs between a gene segment flanked by an RSS with a 12-bp spacer and a segment flanked by an RSS with a 23-bp spacer. In the heavy chain locus, V and J segments are flanked by 23-bp-spacer RSS, and D segments are flanked by one 12-bp and one 23-bp RSS. This ensures V joins to D (23/12) and D joins to J (12/23), preventing V-to-V or J-to-J joining. In the light chain loci, V segments have 23-bp spacers and J segments have 12-bp spacers (or vice versa), ensuring direct V-to-J joining.

Exercise 3 (medium, symbolic).

Calculate the approximate combinatorial diversity of the human antibody repertoire from the following numbers: 40 functional heavy chain V segments, 23 D segments, 6 J segments, 35 functional kappa V segments, 5 kappa J segments, 30 functional lambda V segments, 5 lambda J segments. Express the result in scientific notation.

Hint

Combinatorial diversity has three components: heavy chain V-D-J combinations, light chain V-J combinations (kappa + lambda), and heavy-light pairing. Multiply them together.

Answer

Heavy chain combinations: $40 \times 23 \times 6 = 5, 520$ .

Kappa light chain combinations: $35 \times 5 = 175$ .

Lambda light chain combinations: $30 \times 5 = 150$ .

Total light chain combinations: $175 + 150 = 325$ .

Total combinatorial diversity: $5, 520 \times 325 = 1, 794, 000 \approx 1.8 \times 1 0^{6}$ .

This represents only the combinatorial component. Junctional diversity (N- and P-nucleotide addition) increases the theoretical repertoire by several orders of magnitude, yielding an estimated $1 0^{11}$ to $1 0^{14}$ distinct antibodies when all sources of diversity are combined.

Exercise 4 (medium, symbolic).

Explain the germinal center reaction. What happens in the dark zone versus the light zone, and how does cyclic re-entry drive affinity maturation?

Hint

The germinal center is a specialised microanatomical structure in lymphoid follicles. Two zones serve different functions: one for diversification, one for selection. How do B cells cycle between them?

Answer

The germinal center reaction is a Darwinian process of antibody diversification and selection:

Dark zone: Centroblasts (rapidly dividing B cells) undergo somatic hypermutation. AID introduces point mutations into the antibody variable region at each cell division. The cells do not express surface antibody and do not interact with antigen. Approximately 1-2 mutations are introduced per variable region per division.

Light zone: Centrocytes (non-dividing B cells) re-express surface antibody and compete for antigen displayed on follicular dendritic cells. B cells with higher-affinity surface antibodies capture more antigen, process it, and present more peptide-MHC II to follicular helper T cells ( $T_{F H}$ ). $T_{F H}$ cells deliver survival signals (CD40L, IL-21) to the B cells presenting the most antigen. B cells that fail to capture sufficient antigen die by apoptosis.

Cyclic re-entry: Selected centrocytes can re-enter the dark zone for another round of proliferation and mutation, then return to the light zone for another round of selection. Over 2-3 weeks, this iterative process produces B cells with progressively higher-affinity antibodies — typically a 10- to 10,000-fold improvement over the original naive B cell receptor. Only a small fraction of mutations improve affinity; the majority are neutral or deleterious, and the selection step eliminates these.

Exercise 5 (medium, symbolic).

Compare somatic hypermutation and class switch recombination. Both require AID, but they target different DNA regions and produce different outcomes. Explain the molecular differences.

Hint

Both processes begin with AID-mediated cytosine deamination. What determines whether the outcome is a point mutation (SHM) versus a large DNA deletion (CSR)? Consider the genomic regions targeted and how the initial lesion is resolved.

Answer

Both SHM and CSR are initiated by AID-mediated deamination of cytosine to uracil, but they differ in target, resolution, and outcome:

Target. SHM targets WRC hotspots in antibody variable-region exons (V genes). CSR targets GC-rich switch regions (S regions) upstream of each constant-region gene.

Lesion processing. In SHM, the U mismatch is processed by error-prone repair (UNG and MSH2/MSH6 recruiting translesion polymerases) into point mutations. In CSR, the uracils in S regions are processed by UNG and MSH2/MSH6 into double-strand breaks on both the donor S region (e.g., S-mu) and the acceptor S region (e.g., S-gamma1).

Outcome. SHM produces single-nucleotide substitutions that alter the antibody-binding site (amino acid changes in CDRs). CSR produces a large-scale DNA deletion: the intervening DNA between the donor and acceptor S regions is excised as a circle, and the variable exon is juxtaposed to a new constant-region gene.

Regulation. CSR requires T cell help (CD40L-CD40, cytokines) that directs which S region is targeted. SHM requires transcription through the V region (which exposes single-stranded DNA for AID). Both are restricted to germinal center B cells.

Exercise 6 (hard, symbolic).

A patient presents with recurrent sinopulmonary infections and is found to have normal B cell numbers but undetectable serum IgG, IgA, and IgE, with elevated IgM. Genetic testing reveals a mutation in the AICDA gene (encoding AID). Explain the molecular basis of this phenotype (Hyper-IgM syndrome type 2). Why is IgM elevated, and why are IgG, IgA, and IgE absent? What does this reveal about the relationship between somatic hypermutation and class switch recombination?

Hint

AID is required for both SHM and CSR. IgM is the default antibody isotype produced before class switching. What happens when B cells cannot switch from IgM to other isotypes?

Answer

Hyper-IgM syndrome type 2 (HIGM2) is caused by loss-of-function mutations in AICDA, the gene encoding AID.

Why IgG, IgA, and IgE are absent: Class switch recombination requires AID to introduce double-strand breaks in the switch regions upstream of constant-region genes. Without AID, B cells cannot delete the $C_{μ}$ gene and rejoin the variable exon to downstream constant-region genes ( $C_{γ}$ , $C_{α}$ , $C_{ϵ}$ ). The B cells remain "stuck" producing IgM (and IgD, which is produced by alternative splicing of the $C_{μ}$ / $C_{δ}$ transcript without CSR).

Why IgM is elevated: B cells that would normally switch to IgG, IgA, or IgE continue to produce IgM throughout the immune response. Without CSR, the entire antibody response is funneled into the IgM compartment, leading to polyclonal IgM elevation. The IgM is of low affinity because SHM is also impaired (AID is required for both processes).

Relationship between SHM and CSR: AICDA deficiency demonstrates that both SHM and CSR are AID-dependent but are mechanistically distinct processes downstream of the same initiating enzyme. Patients with AID mutations affecting only the N-terminal region lose CSR but retain partial SHM, while C-terminal mutations can affect both — indicating that different AID domains or cofactor interactions are involved in each process. Patients with UNG mutations (HIGM5) have defective CSR but partially preserved SHM, further demonstrating that the two processes diverge in their downstream repair pathways.

Immunodeficiency, monoclonal antibodies, and antibody engineering Master

Primary immunodeficiencies of antibody diversity

Severe combined immunodeficiency (SCID) due to RAG1/RAG2 deficiency. Loss-of-function mutations in RAG1 or RAG2 abolish V(D)J recombination. B cells cannot rearrange their antibody genes and fail to develop beyond the pro-B cell stage (bone marrow arrest). T cells are similarly absent because the T cell receptor also requires V(D)J recombination. The result is T-B-NK+ SCID: absent T and B cells with normal NK cells. Patients present in the first months of life with failure to thrive, chronic diarrhoea, and recurrent opportunistic infections. Treatment is haematopoietic stem cell transplantation.

Omenn syndrome is a hypomorphic (partial-loss-of-function) variant of RAG deficiency. Residual RAG activity allows limited V(D)J recombination, producing a small, oligoclonal repertoire of T and B cells. These few clones undergo aberrant activation and home to the skin and gut, causing erythroderma, alopecia, lymphadenopathy, eosinophilia, and elevated IgE — an autoimmune-like phenotype caused by dysregulated, oligoclonal lymphocyte expansion. The contrast between null RAG mutations (SCID with absent lymphocytes) and hypomorphic mutations (Omenn syndrome with activated oligoclonal lymphocytes) illustrates how the degree of V(D)J recombination impairment determines the clinical phenotype.

X-linked agammaglobulinaemia (XLA, Bruton's agammaglobulinaemia). Mutations in BTK (Bruton's tyrosine kinase) arrest B cell development at the pre-B cell stage. B cells successfully rearrange the heavy chain gene and express the pre-B cell receptor, but BTK-dependent signalling downstream of the pre-BCR fails, preventing proliferation and light chain rearrangement. The result is near-absent B cells and all immunoglobulin isotypes. BTK is also the target of the B cell malignancy drug ibrutinib.

Hyper-IgM syndromes. A group of disorders characterised by normal or elevated IgM with absent or low IgG, IgA, and IgE:

HIGM1 (X-linked): Mutations in CD40L (CD154) on T cells. CSR and SHM are impaired because T cells cannot deliver the CD40L signal to B cells. Patients also have defective T cell-mediated immunity (opportunistic infections with Pneumocystis, Cryptosporidium) because CD40L is also required for macrophage activation and dendritic cell licensing.
HIGM2 (autosomal recessive): Mutations in AICDA (AID), as discussed in Exercise 6. CSR and SHM are impaired. T cell immunity is preserved (unlike HIGM1).
HIGM3: Mutations in CD40 on B cells. Phenotypically identical to HIGM1 but autosomal recessive.
HIGM5: Mutations in UNG. CSR is impaired (UNG is required for processing AID-induced lesions into double-strand breaks in switch regions), but SHM is partially preserved with a characteristic mutation spectrum (increased C-to-T transitions at SHM hotspots, reflecting the failure to remove the AID-generated uracil).

Monoclonal antibody technology

Hybridoma technology (Kohler and Milstein, 1975). Spleen cells from a mouse immunised with a target antigen are fused with immortal myeloma cells (lacking hypoxanthine-guanine phosphoribosyltransferase, HGPRT) using polyethylene glycol or Sendai virus. The resulting hybridomas combine the antibody-producing capacity of the B cell with the immortality of the myeloma. Selection in HAT medium (hypoxanthine, aminopterin, thymidine) eliminates unfused myeloma cells (which lack HGPRT and cannot use the salvage pathway for nucleotide synthesis in the presence of aminopterin) and unfused B cells (which have a limited lifespan). Surviving hybridomas are screened for the desired antibody specificity and cloned by limiting dilution to obtain monoclonal populations. Each hybridoma produces a single, homogeneous antibody — hence monoclonal antibody (mAb).

Humanisation. Murine monoclonal antibodies elicit a human anti-mouse antibody (HAMA) response when administered to patients, limiting their therapeutic use. Humanisation strategies progressively replace murine sequences with human sequences:

Chimeric antibodies: Murine variable regions grafted onto human constant regions (e.g., rituximab, anti-CD20). Approximately 66% human. Still immunogenic in some patients.
Humanised antibodies: Only the murine CDRs (6 loops total) are grafted onto a human antibody framework (e.g., trastuzumab, anti-HER2; bevacizumab, anti-VEGF). Approximately 90-95% human. Framework residues that support CDR conformation may also be retained from the murine sequence.
Fully human antibodies: Generated by phage display of human antibody libraries or by immunising transgenic mice carrying human immunoglobulin loci (e.g., adalimumab, anti-TNF-alpha, the first fully human mAb approved by the FDA, generated by phage display).

Antibody nomenclature. Therapeutic monoclonal antibodies use the suffix -mab. The preceding letter indicates the source: -o- for murine (e.g., muromomab), -xi- for chimeric (e.g., rituximab), -zu- for humanised (e.g., trastuzumab), and -u- for fully human (e.g., adalimumab). The infix indicates the disease target: -ci- for cardiovascular, -li- for immune/inflammatory (adalimumab targets immune modulation), -tu- for tumour (trastuzumab targets HER2 in breast cancer), etc.

Major therapeutic antibodies

Adalimumab (Humira). Fully human anti-TNF-alpha IgG1. Blocks TNF-alpha signalling in rheumatoid arthritis, Crohn's disease, psoriasis, and other inflammatory conditions. Was the world's best-selling pharmaceutical by revenue for multiple consecutive years. Demonstrates that cytokine neutralisation by monoclonal antibodies is an effective therapeutic strategy.

Trastuzumab (Herceptin). Humanised anti-HER2 (ERBB2 receptor) IgG1. HER2 is overexpressed in approximately 20% of breast cancers (HER2-positive). Trastuzumab blocks HER2 signalling and recruits immune effector cells via its Fc region (antibody-dependent cellular cytotoxicity, ADCC). The combination of trastuzumab with chemotherapy dramatically improved survival in HER2-positive breast cancer. Resistance mechanisms include HER2 truncation (p95HER2, lacking the trastuzumab-binding extracellular domain) and activation of downstream signalling pathways (PI3K mutations).

Rituximab (Rituxan). Chimeric anti-CD20 IgG1. CD20 is expressed on the surface of mature B cells (but not stem cells or plasma cells). Rituximab depletes CD20+ B cells via complement-dependent cytotoxicity (CDC), ADCC, and direct apoptosis induction. Used for B cell lymphomas, rheumatoid arthritis, and autoimmune diseases. The B cell depletion is long-lasting (6-12 months) because the stem cell compartment is preserved and B cell reconstitution from stem cells is slow.

Pembrolizumab (Keytruda). Humanised anti-PD-1 IgG4. Blocks the PD-1 immune checkpoint, reactivating exhausted T cells in the tumour microenvironment. Approved for melanoma, lung cancer, and many other tumour types. Demonstrates that checkpoint inhibition (removing a brake on the immune response) can be more effective than directly targeting tumour antigens.

Bispecific antibodies

Blinatumomab (Blincyto). A bispecific T cell engager (BiTE) antibody that simultaneously binds CD3 on T cells and CD19 on B-lineage cells. This creates an artificial immunological synapse between a T cell and a target B cell, redirecting the T cell's cytotoxic machinery against the B cell regardless of TCR specificity. Approved for Philadelphia chromosome-negative relapsed/refractory B cell acute lymphoblastic leukaemia. The small molecular weight (approximately 55 kDa, lacking an Fc region) results in rapid renal clearance, requiring continuous intravenous infusion.

Antibody-drug conjugates (ADCs)

ADCs combine the targeting specificity of a monoclonal antibody with the cytotoxic potency of a chemotherapy drug. The antibody delivers the drug specifically to cells expressing the target antigen, minimising off-target toxicity.

Structure. Three components: (1) a monoclonal antibody targeting a tumour-associated antigen, (2) a cytotoxic payload (e.g., auristatin, maytansinoid, calicheamicin, or PBD dimer), and (3) a linker connecting the two. Linkers are either cleavable (released by tumour-specific conditions: low pH in endosomes/lysosomes, protease cleavage by cathepsins, or disulfide reduction by glutathione) or non-cleavable (the payload is released only after complete antibody degradation in the lysosome, resulting in a payload with an attached amino acid from the linker).

Tisagenlecleucel (Kymriah) and axicabtagene ciloleucel (Yescarta). These are not ADCs but CAR-T cell therapies — autologous T cells genetically engineered to express a chimeric antigen receptor (CAR) targeting CD19. The CAR consists of an anti-CD19 single-chain variable fragment (scFv, derived from a monoclonal antibody) fused to intracellular signalling domains (CD3-zeta plus one or two co-stimulatory domains such as CD28 or 4-1BB). CAR-T cells are living drugs that proliferate in vivo, traffic to sites of disease, and kill CD19+ B cells. Approved for relapsed/refractory B cell ALL and diffuse large B cell lymphoma. Toxicities include cytokine release syndrome (CRS, driven by massive T cell activation and IL-6 release) and immune effector cell-associated neurotoxicity syndrome (ICANS).

Camelid nanobodies

Nanobodies (single-domain antibodies, VHH) are derived from camelids (camels, llamas, alpacas), which produce a class of antibodies consisting only of heavy chains (no light chains). The single antigen-binding domain (VHH, approximately 15 kDa) is the smallest naturally occurring antibody fragment. Advantages over conventional antibodies include: small size (enabling tissue penetration and access to cryptic epitopes such as enzyme active sites), high stability (resistant to heat and denaturants, functional after oral administration for gastrointestinal targets), and ease of production in microbial systems (E. coli, yeast). Caplacizumab (anti-von Willebrand factor nanobody) was the first nanobody approved for human use (acquired thrombotic thrombocytopenic purpura).

CRISPR-based antibody engineering

CRISPR-Cas9 technology enables precise editing of antibody genes in B cells. Applications include: (1) knocking in broadly neutralising antibody sequences (e.g., anti-HIV antibodies VRC01 or bNAbs targeting the CD4 binding site) into the endogenous immunoglobulin locus, allowing physiological regulation of antibody expression; (2) disrupting immunosuppressive checkpoints in B cells to enhance antibody production; and (3) correcting loss-of-function mutations in antibody genes (e.g., RAG mutations in SCID). In vivo CRISPR-based antibody discovery uses library-based approaches where guide RNA libraries targeting the immunoglobulin loci are delivered to B cells, generating diverse antibody variants that can be screened for improved binding.

Connections Master

Adaptive immunity 17.10.02 pending. Antibody diversity is the molecular basis of the adaptive immune response. V(D)J recombination generates the diverse B cell repertoire during lymphocyte development, and clonal selection (activated by antigen binding to the B cell receptor) expands the specific B cell clone. The affinity-matured antibody response represents the primary effector mechanism of humoral adaptive immunity.
MHC and antigen presentation 17.10.03 pending. B cell activation requires T cell help: the B cell internalises antigen via its surface antibody, processes it, and presents peptides on MHC class II to helper T cells. This MHC-restricted antigen presentation by B cells is the bridge between the humoral (antibody-mediated) and cellular (T cell-mediated) arms of adaptive immunity. Without MHC class II presentation, B cells cannot receive T cell help for germinal center formation, somatic hypermutation, or class switch recombination.
Protein structure 17.01.02 pending. The immunoglobulin fold (beta-sandwich with 7-9 antiparallel beta strands stabilised by a disulfide bond) is one of the most common protein domains in the human genome. Antibodies are the paradigm for understanding how protein sequence determines three-dimensional structure, and how small sequence changes (somatic hypermutation) can fine-tune binding affinity. The CDR loops are a textbook example of how loop conformation determines molecular recognition.
DNA repair pathways 17.06.02 pending. V(D)J recombination and class switch recombination are specialised DNA repair reactions. Both generate intentional double-strand breaks and resolve them by non-homologous end joining (NHEJ). Deficiencies in NHEJ components (Ku70, Ku80, XRCC4, DNA ligase IV, Artemis) cause immunodeficiency due to failed V(D)J recombination, and also cause radiosensitivity and genomic instability due to defective repair of other double-strand breaks.
Transposable elements 17.06.03 pending. The RAG1/RAG2 system shares evolutionary ancestry with transposons. The RAG genes likely originated as a transposable element that inserted into a primordial immunoglobulin gene approximately 500 million years ago, establishing the V(D)J recombination system in jawed vertebrates. The signal joint formed during V(D)J recombination resembles a transposon excision product, and RAG1/RAG2 can catalyse transposition in vitro — a molecular fossil of their transposon origins.
Gene expression: transcription 17.05.02. Somatic hypermutation requires transcription through the antibody variable region, which generates single-stranded DNA for AID targeting. Class switch recombination requires germline transcription through the switch regions (sterile transcripts) to open chromatin and make the S regions accessible to AID. The cytokine-regulated germline transcription of specific S regions determines which isotype is selected during CSR.
Cell signaling 17.07.01. B cell receptor signalling involves Src-family kinases (Lyn, Blk, Fyn), Syk, and downstream adaptors (BLNK) and effectors (PLC-gamma2, PI3K). BTK (Bruton's tyrosine kinase) is a key signalling intermediate; its loss causes XLA. The pre-BCR checkpoint (testing successful heavy chain rearrangement) and the BCR checkpoint (testing functional light chain pairing) both involve receptor-mediated signalling analogous to RTK signalling cascades.

Historical notes Master

The discovery that antibody genes are assembled from separate gene segments by somatic recombination was made by Susumu Tonegawa in 1976. Working at the Basel Institute for Immunology, Tonegawa used Southern blot hybridisation to compare immunoglobulin gene organisation in embryonic cells (where the V and C regions are far apart on the chromosome) versus mature B cells (where a specific V segment has been joined to a J segment adjacent to the C region). This result, published in PNAS 73 (1976) 3628-3632, demonstrated that the genome is dynamically rearranged during lymphocyte development — a radical idea at the time. Tonegawa received the 1987 Nobel Prize in Physiology or Medicine for this discovery.

The two-gene model (separate V and C genes) had been proposed earlier by William Dreyer and Claude Bennett in 1965 (PNAS 54, 864-869), based on the observation that the variable and constant regions of antibody chains have very different sequence variability. They hypothesised that a single constant-region gene somehow associates with different variable-region genes during B cell development, but the mechanism (somatic DNA recombination) was not known.

The RAG genes were identified in 1989 by David Schatz, Marjorie Oettinger, and David Baltimore at the Whitehead Institute (Cell 59, 1035-1048). They transfected fibroblast cells with genomic DNA from B cells and screened for the ability to activate a V(D)J recombination reporter construct. This approach identified two closely linked genes, RAG1 and RAG2, that together are sufficient to confer V(D)J recombination activity on non-lymphoid cells. The tight linkage of RAG1 and RAG2 (they are arranged in opposite transcriptional orientation on chromosome 11 in humans, separated by only a few kilobases) and the similarity of their RSS recognition mechanism to transposase enzymes supported the transposon origin hypothesis.

Terminal deoxynucleotidyl transferase (TdT) was discovered by Bollum in 1960 as a template-independent DNA polymerase in calf thymus. Its role in adding N-nucleotides during V(D)J recombination was established in the 1980s-1990s: TdT knockout mice show a dramatic reduction in CDR3 diversity, with antibody junctions lacking N-nucleotides and showing limited length heterogeneity. TdT is not expressed in the fetus or neonate (it is upregulated after birth), which explains why the fetal antibody repertoire has shorter CDR3 loops and less diversity — a feature that may favour recognition of common neonatal pathogens at the cost of overall repertoire breadth.

AID was discovered by Tasuku Honjo's laboratory at Kyoto University in 1999 (Science 285, 1999, 240-243), in a screen for genes upregulated in B cells undergoing class switch recombination. AID-deficient mice lack both somatic hypermutation and class switch recombination, establishing AID as the single enzyme initiating both processes. The mechanism — cytidine deamination in DNA — was unexpected, as AID was initially thought to act on RNA (analogous to APOBEC enzymes that edit mRNA). The demonstration that AID acts on DNA (by Petersen-Mahrt and Neuberger in 2002) resolved this question and established the field of antibody diversification by targeted DNA damage and error-prone repair.

Hybridoma technology was developed by Georges Kohler and Cesar Milstein at the Laboratory of Molecular Biology in Cambridge, UK, published in Nature 256 (1975) 495-497. Their method for producing monoclonal antibodies of predefined specificity transformed immunology from a descriptive to an engineering discipline and earned them the 1984 Nobel Prize (shared with Niels Jerne). The first therapeutic monoclonal antibody, muromonab-CD3 (OKT3, anti-CD3 for transplant rejection), was approved by the FDA in 1986. It was a fully murine antibody and caused significant HAMA responses, motivating the development of chimeric, humanised, and fully human antibody technologies.

Phage display of antibody fragments was developed by Gregory Winter and colleagues at the Laboratory of Molecular Biology in Cambridge in the 1990s. Single-chain variable fragments (scFv) or Fab fragments are displayed on the surface of filamentous bacteriophage (M13), linking antibody phenotype (binding) to genotype (the encoding DNA packaged inside the phage). Panning against immobilised antigen enriches phage displaying high-affinity binders. Iterative rounds of selection, ideally from large combinatorial libraries ( $1 0^{9}$ to $1 0^{11}$ clones), can yield fully human antibodies without immunising animals. Adalimumab was discovered by this approach. Winter received the 2018 Nobel Prize in Chemistry (shared with Frances Arnold and George Smith) for phage display technology.

Bibliography Master

Tonegawa, S. et al. — Sequence of a mouse germline gene for a variable region of an immunoglobulin light chain, Proc. Natl. Acad. Sci. USA 73 (1976) 3628-3632.
Schatz, D. G., Oettinger, M. A. & Baltimore, D. — The V(D)J recombination activating gene, RAG-1, Cell 59 (1989) 1035-1048.
Muramatsu, M. et al. — Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme, Cell 102 (2000) 553-563.
Kohler, G. & Milstein, C. — Continuous cultures of fused cells secreting antibody of predefined specificity, Nature 256 (1975) 495-497.
Winter, G. et al. — Synthetic antibodies and the first fully human monoclonal antibody (adalimumab) from phage display, Annu. Rev. Immunol. 32 (2014) 1-27.
Dreyer, W. J. & Bennett, J. C. — The molecular basis of antibody formation: a paradox, Proc. Natl. Acad. Sci. USA 54 (1965) 864-869.
Petersen-Mahrt, S. K., Harris, R. S. & Neuberger, M. S. — AID mutates E. coli suggesting a DNA deamination mechanism for antibody diversification, Nature 418 (2002) 99-103.
Bollum, F. J. — Calf thymus polymerase, J. Biol. Chem. 235 (1960) 2399-2403.
Alberts, B. et al. — Molecular Biology of the Cell, 7th ed. (Garland Science, 2022), Ch. 24 The Adaptive Immune System.
Murphy, K. & Weaver, C. — Janeway's Immunobiology, 10th ed. (Garland Science, 2022), Ch. 5.
Lodish, H. et al. — Molecular Cell Biology, 9th ed. (W. H. Freeman, 2021), Ch. 24.
Tonegawa, S. — The generation of antibody diversity, Nature 302 (1983) 575-581.
Jung, D. & Alt, F. W. — Unraveling V(D)J recombination: insights into gene regulation, Cell 116 (2004) 299-311.
Di Noia, J. M. & Neuberger, M. S. — Molecular mechanisms of antibody somatic hypermutation, Annu. Rev. Biochem. 76 (2007) 1-22.
Hamers-Casterman, C. et al. — Naturally occurring antibodies devoid of light chains, Nature 363 (1993) 446-448.