Antibody

• 1. Introduction

  ›

  Antibodies are proteins shaped like the letter Y, and they are the reason most vaccines work. Inside every healthy human body, an estimated 10 billion distinct antibodies circulate, each one engineered to recognize a single target with a precision the rest of biology struggles to match. When a pathogen enters the body, these molecules do not wait. They tag, they bind, they signal for reinforcement. They cross from mother to fetus through the placenta. They persist in breast milk. They linger in the bone marrow for potentially an entire lifetime. How does a protein barely 10 nanometers in size manage all of that? And how does the body produce 10 billion different versions of the same basic molecule from a genome of finite size? The answers reach from the architecture of a Y-shaped protein all the way to chromosome-level genetic rearrangements that happen in real time inside individual cells.

• 2. The Y-Shaped Machine

  ›

  Antibodies weigh roughly 150 kilodaltons and measure about 10 nanometers in size, arranged across three globular regions that form their characteristic Y shape. Four polypeptide chains make up the full unit in humans and most other mammals: two identical heavy chains and two identical light chains, all joined by disulfide bonds. Each chain is itself a series of domains, with sequences of about 110 amino acids each, typically depicted in diagrams as simple rectangles. At the tips of the two upper arms of the Y sit the antigen-binding fragments, called Fab regions. The trunk of the Y is called the Fc region, and it serves a different purpose entirely.

  The Fab region contains variable domains whose shape determines what the antibody will recognize. Within each variable domain are three hypervariable regions called complementarity-determining regions, or CDRs. When the protein folds, these CDR loops cluster together on the antibody's surface and form a binding site whose shape can range from a small pocket to a flat surface to a protrusion. The fit between a CDR and its target antigen is described as a lock-and-key interaction: the antibody's paratope matches the antigen's epitope through spatial complementarity. The forces holding the two together are weak individually, including electrostatic forces, hydrogen bonds, hydrophobic interactions, and van der Waals forces. Because those forces are non-specific and reversible, an antibody can sometimes cross-react with structurally similar antigens.

  The Fc region in the trunk handles an entirely different set of tasks. Effector cells such as macrophages and natural killer cells carry Fc receptors on their surfaces, and these receptors bind the Fc region of an antibody that has already latched onto a target. This triggers the effector cell to act. The Fc region also binds a protein complex called C1q, which activates the classical complement pathway and can lead to the direct destruction of a bacterial cell. IgG antibodies benefit from one more Fc-region function: the neonatal Fc receptor, FcRn, binds to IgG's Fc region to transport it across the placenta from mother to fetus. The same FcRn binding endows IgG with a half-life of 3-4 weeks, which is exceptionally long relative to most plasma proteins.

  Antibodies do not only exist as single Y-shaped units. IgA can form dimers, IgM typically exists as a pentamer in its secreted form, and IgG has been observed forming hexamers as well. Larger antigen-antibody complexes also form when multivalent antigens carry multiple epitopes. An extreme example is the agglutination of red blood cells by antibodies during blood typing, where the clumps grow large enough to become insoluble and visually apparent.

• 3. Five Classes, Five Jobs

  ›

  Humans produce antibodies in five distinct classes: IgA, IgD, IgE, IgG, and IgM. Each class is defined by the type of heavy chain it carries and the effector functions that chain enables. IgG comes in four subclasses and provides the majority of antibody-based defense against invading pathogens. It is the only antibody class capable of crossing the placenta to give a fetus passive immunity, and it is the most commonly used molecular format in current antibody drugs.

  IgA stands guard at mucosal surfaces, including the gut, the respiratory tract, the urogenital tract, saliva, tears, and breast milk. Early clinical studies suggest IgA isotype antibodies may have potential as anti-cancer therapeutics, with demonstrated ability to reduce tumor growth. IgM is the first antibody produced in any immune response, expressed on the surface of B cells as a monomer but secreted as a pentamer with very high avidity. Its job is to eliminate pathogens in the early stages of an immune response before there is sufficient IgG.

  IgE is the least abundant class of immunoglobulin. Its variable region binds allergens such as house dust mite particles, while its Fc region engages a specific Fc receptor on mast cells, triggering degranulation and the release of histamine. IgE is considered a sole contributor to asthma in many cases. Humans and other animals are thought to have evolved IgE to protect against parasitic worms, though in the present environment, IgE is primarily associated with allergies. IgD functions mainly as an antigen receptor on B cells that have never encountered antigen, and has also been reported to activate basophils and mast cells to produce antimicrobial factors.

  The class an antibody belongs to is not fixed at birth. Through a process called isotype switching, or class switch recombination, an activated B cell can shift production from IgM or IgD to IgA, IgE, or IgG. Only the constant region of the heavy chain changes; the variable region and its antigen specificity remain exactly the same. The switch is triggered by cytokines present in the B cell's environment, meaning the type of immune threat shapes which class of antibody the cell produces.

• 4. B Cells and Their Offspring

  ›

  B cells themselves do not secrete antibody. They can only display immunoglobulin on their surface because they lack a critical transcription factor called BLIMP-1, which establishes the antibody-secreting program inside a cell. To become antibody factories, B cells must first differentiate into plasmablasts or plasma cells. Plasmablasts are short-lived, rapidly proliferating cells that emerge early in an immune response. Plasma cells are terminally differentiated, meaning they do not divide, and they secrete large quantities of antibody continuously, regardless of whether or not the original antigen is still present.

  Long-lived plasma cells can persist in survival niches, classically in the bone marrow, for potentially the entire lifetime of the organism. Other survival niches can form in mucosal tissues, with a different hierarchy of antibody classes from those in bone marrow. A typical human B cell before activation carries between 50,000 and 100,000 antibody molecules bound to its surface. When an antigen binds to those surface antibodies, they cluster in large patches that can exceed 1 micrometer in diameter on lipid rafts that isolate the B cell receptors from most other cell signaling receptors.

• 5. Building Ten Billion Variants

  ›

  Humans generate an estimated 10 billion different antibodies, each recognizing a distinct epitope, despite the human genome containing a limited number of antibody genes. The solution is a combinatorial gene-assembly process called V(D)J recombination. The variable region of each antibody heavy chain is encoded in separate gene segments called variable (V), diversity (D), and joining (J) segments. The heavy chain locus, found on chromosome 14 in humans, contains about 65 different variable domain genes. Lambda light chain genes are on chromosome 22 and kappa light chain genes are on chromosome 2.

  In the bone marrow, each developing B cell randomly selects and combines one V, one D, and one J segment to assemble its heavy chain variable region, and one V and one J segment for its light chain. The number of possible combinations across all gene segments is enormous. RAG proteins cut the DNA at specific sites to initiate this recombination; without RAG proteins, V(D)J recombination does not occur. Once a B cell has assembled a functional immunoglobulin gene, a process called allelic exclusion prevents it from rearranging any other variable region. Each B cell therefore produces antibodies with only one kind of variable chain.

  After a B cell is activated by antigen, a second diversification process begins. The genes encoding the variable domains undergo somatic hypermutation, accumulating approximately one nucleotide change per variable gene per cell division. Daughter B cells acquire slight amino acid differences in their antibody variable domains. Those that happen to carry mutations giving higher antigen-binding affinity receive stronger survival signals from interactions with other cells. Those with weaker affinity die by apoptosis. This selective pressure drives the average affinity of the antibody population upward over time, a process called affinity maturation. The result is that antibodies improve their fit for their target throughout the course of an immune response.

• 6. A History Written in Experiments

  ›

  The word antibody entered the scientific record in a text by Paul Ehrlich. The German term Antikörper appears in the conclusion of his article published in October 1891, where he wrote that if two substances give rise to two different Antikörper, then the substances themselves must be different. The word was not adopted immediately; competing terms including Immunkörper, Amboceptor, Zwischenkörper, and Immunisin were all proposed at various points.

  The scientific study of antibodies traces to 1890, when Emil von Behring and Kitasato Shibasaburō described antibody activity against diphtheria and tetanus toxins and proposed humoral immunity: the idea that a mediator in serum could react with a foreign antigen. Ehrlich built on this in 1897 with his side-chain theory, hypothesizing that receptors on cell surfaces could bind toxins in a lock-and-key interaction and that this binding triggers antibody production. In 1904, Almroth Wright proposed that soluble antibodies coat bacteria to label them for ingestion by phagocytes, a process he named opsoninization.

  In the 1920s, Michael Heidelberger and Oswald Avery showed that antibodies are made of protein. The lock-and-key theory received firm confirmation in the 1940s when Linus Pauling showed that antibody-antigen interactions depend more on shape than on chemical composition. In 1948, Astrid Fagraeus identified B cells in the form of plasma cells as the source of antibody production.

  Structural understanding advanced significantly in the early 1960s. Gerald Edelman and Joseph Gally discovered the antibody light chain and recognized it as the same protein as the Bence-Jones protein described by Henry Bence Jones in 1845. Edelman went on to show that antibodies are composed of disulfide bond-linked heavy and light chains. At the same time, Rodney Porter characterized the Fab and Fc regions of IgG. Together, these scientists determined the complete amino acid sequence of IgG, work for which they were jointly awarded the 1972 Nobel Prize in Physiology or Medicine. The other immunoglobulin classes were identified in the 1960s as well: Thomas Tomasi discovered secretory IgA, David S. Rowe and John L. Fahey discovered IgD, and Kimishige Ishizaka and Teruko Ishizaka discovered IgE and showed it was involved in allergic reactions. Then, in a landmark series of experiments beginning in 1976, Susumu Tonegawa demonstrated that genetic material can rearrange itself to generate the vast array of available antibodies.

• 7. Medicine Built on Antibodies

  ›

  Targeted monoclonal antibody therapy is now used to treat rheumatoid arthritis, multiple sclerosis, psoriasis, and many forms of cancer including non-Hodgkin's lymphoma, colorectal cancer, head and neck cancer, and breast cancer. Monoclonal antibodies are produced by hybridoma cell lines: antibody-secreting lymphocytes are isolated from an immunized animal and fused with a cancer cell line to create immortalized cells that continuously secrete a single, defined antibody. Single hybridoma cells are then isolated by dilution cloning, so all daughter cells produce the same antibody. The activation of natural killer cells through antibody-dependent cell-mediated cytotoxicity, triggered when antibodies coat a cancer cell and natural killer cells bind to the Fc region, may partly explain why monoclonal antibody therapies work against cancer.

  Diagnostic medicine relies on antibodies extensively. Blood tests for Epstein-Barr virus and Lyme disease work by measuring titers of antibodies directed against those pathogens. In clinical immunology, levels of individual immunoglobulin classes are measured by nephelometry or turbidimetry to characterize a patient's antibody profile; elevated IgM is often associated with primary biliary cirrhosis, while IgA deposition along hepatic sinusoids can suggest alcoholic liver disease. The Coombs test detects antibodies against red blood cell surface antigens in immune-mediated hemolytic anemia and is also used in blood transfusion preparation and antenatal antibody screening. Home pregnancy tests use antibodies directed at human chorionic gonadotropin.

  One striking prenatal application involves Rh factor, an antigen on red blood cells. When a mother is Rh-negative and carries an Rh-positive fetus, blood mixing can sensitize the mother's immune system to the Rh antigen, putting future pregnancies at risk for hemolytic disease of the newborn. A treatment called Rho(D) Immune Globulin, administered before and immediately after delivery, introduces anti-RhD antibodies that destroy fetal Rh antigen in the mother's system before her B cells can memorize the antigen and mount an enduring response. Newer radioactive antibody labeling using dioxaborolane chemistry enables fluoride-18 labeling, which allows positron emission tomography imaging of cancer.