Figure 1. B cell populations. The same B cell precursor can probably give rise to different B cell populations. All subsets initially express IgM and most express IgD. Activated follicular B cells can develop into plasma cells that secrete IgM and some IgD (not shown). These B cells can also switch the class of immunoglobulin they express to IgG, IgA or IgE and can develop into plasma cells or memory cells. Activated marginal zone B cells can become plasma cells that make IgM antibodies, typically to carbohydrate antigens, but they do not switch to most other immunoglobulin classes or become memory cells. Some B‐1 B cells make natural IgM antibodies in the apparent absence of any antigenic stimulation and perhaps some IgA.
|
Figure 2. Antigen‐binding sites. When an immunoglobulin molecule folds into its tertiary structure, the six hypervariable regions (CDRs) – three each from the H and L chains – are brought together, they form the three‐dimensional antigen‐binding site, which can vary in structure from a flat surface as shown or a deep cleft (not shown).
|
Figure 3. Studies of immunoglobulin structure. When antibodies were digested with the protease papain, fragments of two sizes were obtained. One fragment could bind antigen and was called Fab; this fragment could not be crystallized. The other fragment did not bind antigen but could be crystallized and was called Fc. If however, immunoglobulin was digested with pepsin, a single large fragment was obtained that could bind two antigen molecules and was called F(ab′) 2. When disulphide bonds were reduced, two complete H and L chains were released from each antibody monomer. The V and C regions of each chain are indicated.
|
Figure 4. Generation of immunoglobulin genes. Complete immunoglobulin genes are found only in B cells. In germline DNA, three separate loci on different chromosomes code for the H chain and the κ and λ L chains. At each locus there are multiple gene segments that code for parts of the V region (and C regions). The H chain region contains numerous V H segments, a number of D H segments and a small number of J H segments. In B cells, one segment from each group is selected (largely randomly) and the selected segments are brought together to form the complete V region. The V region is paired initially with the µ (and δ; not shown) C region segments. For L chains the principle is similar, but there are no D region segments and the V region pairs with either κ or λ C region segments. During assembly of the complete H and L chain genes, additional bases can be inserted at the junctions between V, (D) and J segments, increasing junctional diversity (see Chapter , Figure 28 for details). The order of the C region genes is shown in a simplified form for humans.
|
Figure 5. Structures of different classes of antibodies. The five main classes of antibodies all have a similar monomeric structure. Some classes can also form multimers, such as IgM pentamers or IgA dimers in which the monomers are linked by another component called the J chain. However, the detailed structure of the monomers differs for each class (and subclass) of antibody. These differences lead to different properties for each antibody. For example, the arms of the monomer may be relatively rigid or flexible. The stem may be short or extended. The biological properties of antibodies are determined by the structure of the stem, enabling it to be recognized by different types of FcR, for example, or to activate complement. (Immunoglobulin subclasses for IgG and IgA are not shown.)
|
Figure 7. Tissue distribution of antibodies. IgM is confined to the blood plasma unless there is acute inflammation, when it can pass between endothelial cells into extravascular spaces. IgG can cross endothelia into many extravascular spaces and, in addition, can cross the human placenta to give protection to the neonate. Monomeric IgA tends to be present in the blood, but dimeric IgA is transported onto mucosal surfaces such as the gastro‐intestinal, respiratory and uro genital tracts, and the lactating breast. IgE is normally present at very low concentrations in blood but binds to mast cells, particularly those underlying the epithelia. IgD is usually only present in significant amounts on the surface of naïve B cells and is normally only present in the blood at exceedingly low levels; not shown.
|
Figure 8. Transport of IgA onto mucosal surfaces. Mucosal plasma cells secrete IgA as dimers, joined by the J chain. Dimeric IgA binds to poly‐immunoglobulin receptors on the baso‐lateral surfaces of mucosal epithelial cells and is endocytosed. These vacuoles are transported to the luminal surface of the epithelial cell where the poly‐immunoglobulin receptor is cleaved. This releases the dimeric IgA, with a portion of the receptor known as the secretory piece, which renders the IgA less susceptible to proteolytic degradation.
|
Figure 9. The common mucosal immune system. Antigens from the gut lumen, for example, cross M cells into Peyer's patches (or similar secondary lymphoid organ in other mucosal sites) and, with T cell help, develop into B cells that express IgA. These cells then leave these secondary lymphoid tissues and enter the blood. They express homing receptors for adhesion molecules present on all mucosal venule endothelial cells and thus can migrate, relatively randomly, to any mucosal tissue. Within these tissues they mature into plasma cells secreting dimeric IgA which is transported onto the luminal surface of any mucosal tissue.
|
Figure 10. Monoclonal antibody production. Mice (or another species such as rats) are immunized repeatedly with an antigen. Activated B cells (lymphoblasts) from the spleen are chemically fused (e.g. using polyethylene glycol) with non‐antibody‐secreting B tumour cells (myelomas), to form hybridomas. Each hybridoma secretes antibodies of a single specificity against the antigen used for immunization (i.e. they are monoclonal). The hybridomas are effectively immortal and will continue to proliferate indefinitely. They can be screened to select those of desired specificity (Figure 11).
|
Figure 11. Monoclonal antibody screening. The products of the cell fusion (See Figure 10) are divided into small wells of a culture vessel at limiting dilution, so most wells only contain one hybridoma cell. Non‐fused B cells die rapidly and the culture medium is designed so that non‐fused myeloma cells cannot survive. The fluid from individual cultures can then be tested (screened) for the presence of antibodies specific for a single epitope of the antigen (here, anti‐X). Cells from selected cultures are grown as clones and expanded. These hybridomas can be grown on industrial scales and generate sufficient monoclonal antibodies to be used clinically.
|
Figure 6. Antibody affinity and avidity. The affinity of an antibody refers to the strength of binding of a single antigen‐binding site to a single antigenic determinant (epitope). Antibody avidity refers to the overall strength of binding between all the antigen‐binding sites and an antigen that expresses more than one identical epitope. The avidity of a divalent F(ab′) 2 fragment is vastly greater than twice the affinity of a monomeric Fab fragment of the same antibody. Likewise, the avidity for a divalent IgG antibody may be 100 times its affinity, and for pentameric IgM 1000 times or more.
|
Figure 12. Toxin and pathogen neutralization by antibodies. Bacterial exotoxins such as cholera toxin, and pathogens such as influenza virus, need first to attach to receptors on cells to damage or modulate cells, or to replicate. Antibodies that bind to the toxin receptor or the pathogen‐binding site can block attachment and thus prevent toxin action or viral infection. This is known as neutralization.
|
Figure 13. Pathogen opsonization. Phagocytes (neutrophils and macrophages) can kill many phagocytosed pathogens intracellularly. Some bacteria avoid destruction by phagocytes because they do not express surface molecules that can be recognized by the phagocyte. If, however, the pathogen is coated with antibodies, it becomes opsonized and can be recognized either directly (through FcRs) or indirectly (through complement receptors) because the antibodies activate and bind complement. In this figure, Streptococcus pneumoniae is opsonized by anti‐capsule IgM which binds and activates complement, permitting phagocytosis and killing of the bacterium. A virus is shown being opsonized by IgG.
|
Figure 14. Antibodies and the induction of acute inflammation. Antibodies IgM or IgG (not shown) antibodies bound to pathogens (immune complexes) may activate complement. Complement fragments C3a and C5a (which are called anaphylatoxins) can bind to receptors on mast cells and activate the cell. The mast cells release multiple pro‐inflammatory mediators that act on local venules to increase permeability and recruit neutrophils and monocytes. C5a is also a powerful chemoattractant for neutrophils. (In some cases, uptake of immune complexes by phagocytes can also trigger the cells to secrete pro‐inflammatory molecules; not shown.)
|
Figure 15. Antibody‐dependent cell‐mediated cytotoxicity. Some cells with cytotoxic functions, such as NK cells, express FcRs for IgG. If a target cell is experimentally coated with IgG antibodies (e.g. that have been produced against a cell surface molecule of the target cell) it can be recognized through the FcRs of the cytotoxic cells. This stimulates the activation of cytotoxic mechanisms that include perforin and granzyme secretion, (the role of Fas‐Fas‐ligand interactions in ADCC is uncertain) that can induce apoptosis in the target cell. (Eosinophils may also use ADCC to kill some immature parasites, such as schistosomula, that are coated with IgE; not shown.)
|
Figure 16. Cellular interactions in T‐dependent B cell activation. (1) Initially a CD4 T cell is activated by an activated DC that expresses antigen‐derived peptides on MHC class II in the T cell area of a secondary lymphoid tissue. (2) An antigen‐specific B cell endocytoses the antigen and presents peptides on its own MHC class II. The activated T cell recognizes the peptide–MHC complex on the B cell and delivers activation signals to the B cell. This happens on the margins of the T cell area, close to the B cell follicle. (3) Within the follicle, further interactions between B cells and activated T cells can lead to isotype switching and somatic hypermutation.
|
Figure 17. Germinal centre reaction. Germinal centres form in B cell follicles during TD antibody responses. A very few activated B cells (possibly only one) enter each germinal centre. In the dark zone, the B cells (centroblasts) proliferate very rapidly to form clones that undergo somatic hypermutation. Many of these B cells apoptose. Surviving B cells enter the light zone as centrocytes where they interact with Follicular T helper cells (T fh) and follicular dendritic cells (FDCs), undergoing further differentiation and possibly isotype switching. Again, many B cells apoptose here. Some B cells may then re‐enter the dark zone, expand and undergo further mutation in an iterative process. Other B cells leave the germinal centre and may become plasma cells or memory cells.
|
Figure 18. Fates of activated B cells. B cells activated early in the response, outside follicles, migrate to areas such as the lymph node medulla or splenic red pulp and become short‐lived (a few days) plasma cells that secrete antibodies of different isotypes perhaps especially IgM and IgG. Later in the response, activated B cells originating from the germinal centre can migrate to areas such as the bone marrow and may become long‐lived plasma cells, secreting antibodies of different isotypes (IgG, IgA or IgE) for months or years. Other B cells activated late in the response become memory cells. Some of these are retained in tissues such as the spleen for long periods while others re‐enter the recirculating pool of B cells.
|
Figure 19. T‐independent antibody synthesis. TI antigens are of two types. TI‐1 antigens are exemplified by LPS. LPS is mitogenic for B cells at high concentrations and stimulates non‐specific immunoglobulin synthesis. This involves signalling via TLR4. At low concentrations, however, LPS induces specific anti‐LPS antibody synthesis perhaps by signalling through both antigen‐specific BCRs and TLRs. TI‐2 antigens are usually polymeric carbohydrates that can cross‐link surface immunoglobulin. Typically, the antibodies produced in TI responses are IgM, although some degree of switching to IgG3 and IgA can occur. TI responses show little if any memory or affinity maturation.
|
Figure 20. Intracellular signalling pathways in B cell activation. A highly over‐simplified scheme of some key intracellular signalling pathways in B cells. When the BCR is ligated or cross‐linked by antigen, signals are delivered via CD79 that activate several distinct intracellular signalling pathways. These induce cytoskeletal changes, changes in cell metabolism and the activation and nuclear translocation of several transcription factors, including AP‐1, NF‐κB and NF‐AT. Additional costimulatory signals may derive from the CD19–CD21–CD81 complex (Figure 21). In TD activation the activated T cell also delivers signals via CD40 and the secretion of cytokines (Figure 22).
|
Figure 21. Positive and negative regulation of B cell activation. If an antigen that binds a BCR is itself bound to complement C3d, the complement fragment can bind to CD21 (complement receptor type 2) complexed with CD19 and CD81. Signalling through CD19 then generates positive signals which can increase the sensitivity of the B cell to activation by 100 times or more. In contrast, at sufficiently high levels, IgG can bind to an inhibitory FcR on the B cell, FcγRIIB, which generates negative signals to inhibit further B cell activation and hence to prevent further IgG being produced.
|
Figure 22. T cell–B cell collaboration. Activated B cells (e.g. in germinal centres) can express co stimulatory molecules such as CD80 and CD86. These may help activate CD4 T cells (e.g. follicular T cells) that recognize peptide–MHC class II on the B cells. Other signals may then be generated (in either or both directions) through CD40 – CD40 ligand and other interactions (not shown) and through cytokines.
|
Figure 23. Isotype (class) switching. Isotype switching mainly occurs in TD responses. Naïve B cells express IgM and usually IgD. During activation a B cell may switch to producing a different immunoglobulin class. To bring this about a loop of DNA is formed that contains the intervening C region genes. The loop is excised and the free ends of DNA join so that the particular C region gene is apposed to the pre‐assembled V region gene. The cell can now produce a different antibody isotype (here IgE), but of the same antigen specificity. The C region gene segments are shown in a simplified manner.
|
Figure 24. Regulation of isotype switching. Which isotype a B cell expresses after activation is largely determined by cytokines secreted by its activated T helper cell, although cytokines secreted by other cells may also be important. Thus, in mice, IFN‐γ induces switching to IgG2a, IL‐4 to IgE, and IL‐5 plus TGF‐β to IgA.
|
Figure 25. Somatic hypermutation. During TD responses, random mutations are introduced in V region genes of B cells, particularly in the CDR regions (compare top and bottom showing Ig V regions of a naive B cell and one of its germinal centre progeny respectively. This is somatic hypermutation. These mutations may alter the structure of the antigen‐binding site. Most mutations will be neutral or decrease binding affinity, and the B cell will apoptose, but occasionally will result in higher affinity. In conditions of low antigen concentration, B cells expressing a higher affinity BCR will accumulate more antigen, and will have a selective advantage and survive. Thus, the average affinity of the secreted antibodies will increase. This is affinity maturation.
|
Figure 26. Antibody‐mediated resistance to re‐infection. At least three different mechanisms can be involved in the long‐term maintenance of antibody‐mediated resistance to re‐infection. In any immunocompetent individual, memory B cells specific for the antigen can be re‐activated rapidly following infection. This, however, takes time and resistance to some infections requires pre‐formed antibodies. Long‐lived plasma cells, particularly in the bone‐marrow, can also secrete antibodies for long periods (years), giving immediate protection. At the population level, repeated subclinical infection of immune individuals by an infection endemic in the community can serve to top up immunity without the individual being aware of the infection.
|
Figure 27. Follicular dendritic cells. FDCs are mesenchymal cells (not bone marrow‐derived). They are long‐lived resident (non‐migratory) cells found in B cell follicles in secondary lymphoid tissues. FDCs express complement and Fc receptors that do not mediate endocytosis. This enables them to retain native (i.e. non‐degraded) antigens that are complexed to antibodies on their surfaces for long periods. These complexes may be sampled by memory B cells and used to maintain low‐level stimulation of these cells. The B cells may be able to process and present the antigen they have acquired to memory T cells, which in turn activate the B cells. This serves to provide a continual supply of antibody and to maintain the memory B cell pool.
|
Figure 28. B cell differentiation. In adult mice and humans B cells are formed in the bone marrow from the common lymphoid precursor. B‐1 B cells leave the marrow and enter sites such as the peritoneal cavity, where they form a self‐renewing population. B‐2 B cells develop through pro‐B and pre‐B stages to become immature B cells. These leave the marrow as transitional B cells that migrate to the spleen. Here, a B‐2 B cell may become either a resident marginal zone B cell or leaves to become a recirculating follicular B cell.
|
Figure 29. B cell tolerance in the bone marrow. Immature B cells with BCRs that do not recognize self antigens enter the periphery and develop into mature B cells. Immature B cells that recognize multi‐valent self antigen in the marrow can, for a short time, attempt to form new BCRs that are non‐reactive through receptor editing. If the new receptor does not recognize self antigen, the B cell exits the marrow. If the BCR continues to recognize self antigen, the B cell apoptoses. If an immature B cell recognizes soluble self antigen in the marrow it may become an anergic B cell that is unresponsive to stimulation.
|
Figure 30. Antibody engineering. Non‐human monoclonal antibodies induce antibody responses in humans that render them ineffective. Several approaches are being used to overcome this problem. Humanized monoclonal antibodies. Genetic engineering can be used to replace all of a mouse immunoglobulin gene, except the CDRs, with human sequences. These antibodies retain their antigen specificity, but are much less immunogenic Human immunoglobulin transgenic mice. A mouse's immunoglobulin genes can be replaced with human genes and on immunization such mice make fully human antibodies. Phage display. Libraries of human V H and V L gene segments can be produced and the V H and V L segments joined to make genes that code for single‐chain variable fragment (ScFV) antibody fragments. These ScFV fragments can be expressed on filamentous bacteriophages, and phages expressing antibodies of interest can be selected by binding to antigen‐coated plates. The selected ScFV can then be joined to desired C region gene segments.
|