Immunodiffusion

http://210.36.18.48/gxujingpin/dwwswx/im/7.htm CHAPTER 7 ANTIGEN-ANTIBODY INTERACTIONS INTRODUCTION Antibodies constitute the humoral arm of acquired immunity that provides protection against infectious organisms and their toxic products. Therefore, the interaction between antigen and antibody is of paramount importance. In addition, because of the exquisite specificity of the immune response, the interaction between antigen and antibody in vitro is widely used for diagnostic purposes, for the detection and identification of either antigen or antibody. The utilization of the in vitro reaction between antigen and serum antibodies is termed serology. An example of the use of serology for the identification and classification of antigens is the serotyping of various microorganisms by the use of specific antisera. The interaction of antigen with antibodies may result in a variety of consequences, including precipitation (if the antigen is soluble), agglutination (if the antigen is particulate), and activation of complement. All of these outcomes are caused by the interactions between multivalent antigens and antibodies that have at least two combining sites per molecule. The consequences of antigen-antibody interaction listed above do not represent the primary interaction between antibodies and a given epitope but, rather, depend on secondary phenomena, which result from the interactions between multivalent antigens and antibodies. Such phenomena as the formation of precipitate, agglutination, and complement activation would not occur if the antibody with two or more combining sites reacted with a hapten (i.e., a unideterminant, univalent antigen), nor would they occur as a result of the interaction between a univalent fragment of antibody, such as Fab, and an antigen, even if the antigen is multivalent. The reasons for these differences are depicted in Figure 7.1A-E. Cross-linking of various antigen molecules by antibody is required for precipitation, agglutination, or complement activation, and it is possible only if the antigen is multivalent and the antibody is divalent [either intact, or F(ab')2] (see Fig. 7.lB, D, E). In contrast, no cross-linking is possible ifthe antigen or the antibody is univalent (Fig. 7. lA, C). There are many serologic reactions that demonstrate the binding between antigen and antibodies. This chapter describes selected reactions that are used in diagnosis; many others, not included here, are mostly variations of the reactions described here. PRIMARY INTERACTIONS BETWEEN ANTIBODY AND ANTIGEN No covalent bonds are involved in the interaction between antibody and an epitope. Consequently, the binding forces are relatively weak. They consist mainly of van der Waals forces, electrostatic forces, and hydrophobic forces, all of which require a very close proximity between the interacting moieties. Thus the interaction requires a very close fit between an epitope and the antibody, a fit that is often compared to that between a lock and a key. Because of the low levels of energy involved in the interaction between antigen and antibody, antigen-antibody complexes can be readily dissociated by low or high pH, by high salt concentrations, or by chaotropic ions, such as cyanates, which efficiently interfere with the hydrogen bonding of water molecules. Association Constant The reaction between an antibody and an epitope of an antigen is exemplified by the reaction between antibody and a univalent hapten. Because an antibody molecule is symmetric, with two identical Fab antigen combining sites, one antibody molecule binds with two identical hapten molecules, each Fab binding in an independent fashion with one hapten molecule. Affinity and Avidity The intrinsic association constant that characterizes the binding of an antibody with an epitope or a hapten is termed affinity. When the antigen consists of many repeating identical epitopes or when antigens are multivalent, the association between the entire antigen molecule and antibodies depends not only on the affinity between each epitope and its corresponding antibody but also on the sum of the affinities of all the epitopes involved. For example, the affinity of binding of antiA with multivalent A (shown in Fig. 7. lB) may be four or five orders of magnitude higher than between the same antibody (i.e., anti-A) and univalent A (Fig. 7. lA). This is because the pairing of anti-A with A (where A is multivalent) is influenced by the increased number of sites on A with which anti-A can react. While the term affinity denotes the intrinsic association constant between antibody and a univalent ligand such as a hapten, the term avidity is used to denote the overall binding energy between antibodies and a multivalent antigen. Thus, in general, IgM antibodies are of higher avidity than IgG antibodies, although the binding of each Fab in the IgM antibody with ligand may be of the same affinity as that of the Fab from IgG. SECONDARY INTERACTIONS BETWEEN ANTIBODY AND ANTIGEN Agglutination Reactions the reactions of antibody with a multivalent antigen that is particulate (i.e., an insoluble particle) results in the cross-linking of the various antigen particles by the antibodies. This cross-linking eventually results in the clumping or agglutination of the antigen particles by the antibodies. TITER. The agglutination of an antigen as a result of cross-linking by antibodies is dependent on the correct proportion of antigen to antibody. Figure 7.2 depicts an example of an agglutination test for antibodies to the bacterium Brucella abortus present in the serum of an infected individual. The figure shows 10 tubes containing twofold serial dilution of the serum, ranging from 1:4 to 1:2048, to which equal amounts of a suspension of B. abortus (a particulate antigen) are added. The plus and minus signs denote the presence or absence of agglutination. The results of the test (shown below each tube) indicate that agglutination occurs at dilutions of serum of 1:16 to 1:1024. There is no agglutination at higher dilutions because at such dilutions there are not enough antibodies to cause appreciable, visible agglutination. The highest dilution of serum that still causes agglutination, but beyond which no agglutination occurs, is termed the titer. PROZONE. shows tubes with no agglutination, although they contain a suspension of antigen and concentrated serum (diluted only 1:4 or 1:8). It is a common observation that agglutination may not occur at high concentrations of antibody, even though it does take place at higher dilutions of serum. The tubes with high concentrations of serum, where agglutination does not occur, represent a prozone. In the prozone, antibodies are present in excess. Agglutination may not occur at high ratio of antibody to antigen because every epitope on one particle may bind only to a single antibody molecule, preventing cross-linking between different particles (see Fig. 7.3). Because of the prozone phenomenon, in testing for the presence of agglutinating antibodies to a certain antigen, it is imperative that the antiserum be tested at several dilutions. Testing serum at only one concentration may give misleading conclusions if no agglutination occurs, because the absence of agglutination might reflect either a prozone or a lack of antibody. The agglutinating titer of a certain serum is only a semiquantitative expression of the antibodies present in the serum; it is not a quantitative measure of the concentration of antibody (weight/volume). Rather, the titer represents the ability of a certain dilution (i.e., volume) of the serum containing the antibodies to cause agglutination. As such, the titer of a given antisermn may be used for comparison with the agglutinating titer to another antiserum to the same antigen. For example, a change in titer of anti-B, abortus antibodies, in an individual, from 1:4 to 1:1024 would indicate an acute infection, while on the other hand a drop in titer might suggest that antimicrobial therapy was working. Thus, agglutination titers are useful for comparisons of the relative concentrations of agglutinating antibodies in various sera specific for the same antigen. Since the titer in any agglutination assay depends on a variety of factors, such as size, charge, and density of epitopes on an antigen, it is of little use to compare titers of antisera to different antigens. ZETA POTENTIAL. The surfaces of certain particulate antigens may possess an electrical charge, as, for example, the net negative charge on the surface of red blood cells caused by the presence of sialic acid. When such charged particles are suspended in saline solution, an electrical potential termed the zeta potential is created between particles, preventing them from getting very close to each other. This introduces a difficulty in agglutinating charged particles by antibodies, in particular red blood cells by IgG antibodies. The distance between the Fab arms of the IgG molecule, even in its most extended form, is too short to allow effective bridging between two red blood cells across the zeta potential. Thus, although IgG antibodies may be directed against antigens on the charged erythrocyte, agglutination may not occur because of the repulsion by the zeta potential. On the other hand, some of the Fab areas oflgMpentamers are far enough apart and can bridge red blood cells separated by the zeta potential. This property of IgM antibodies, together with their pentavalence, is a major reason for their effectiveness as agglutinating antibodies. Through the years attempts were made to improve agglutination reactions by decreasing the zeta potential in various ways, none of which was universally applicable or effective. However, an ingenious method was devised in the 1950s by Coombs to overcome this problem. This method, described below, facilitates the agglutination of erythrocytes by IgG antibodies specific for erythrocyte antigens. It is also useful for the detection of antibodies that are present on the surface of erythrocytes but that are unable to agglutinate them. PASSIVE AGGLUTINATION. The agglutination reaction can be used with particulate antigens (e.g., erythrocytes or bacteria) and also with soluble antigens, provided that the soluble antigen can be firmly attached to insoluble particles. For example, the soluble antigen thyroglobulin can be attached to latex particles, so that the addition of antibodies to the thyroglobulin antigen will cause agglutination of the latex particles coated with thyroglobulin. Of course, the addition of soluble antigen to the antibodies prior to the introduction of the thyroglobulin cortex latex particles will inhibit the agglutination because the antibodies will first combine with the soluble antigen, and if the soluble antigen is present in excess, the antibodies will not be able to bind with the particulate antigen. This latter example is referred to as agglutination inhibition. It should be distinguished from agglutination inhibition in which antibodies to certain viruses inhibit the aggluti nation of red blood cells by the virus. In these cases, the antibodies are directed to the area or areas on the virus that bind with the appropriate virus receptors on the red blood cells. When the antigen is a natural constituent of a particle, the agglutination reac tion is referred to as direct agglutination. When the agglutination reaction takes place between antibodies and soluble antigen that had been attached to an insolu ble particle, the reaction is referred to as passive agglutination. The agglutination reaction (direct or passive, either employing or not employ ing the Coombs test) is widely used clinically. In addition to the examples already given, major applications include erythrocyte typing in blood banks, diagnosis of various immunologically mediated hemolytic diseases, such as drug-induced auto hemolytic anemia, tests for rheumatoid factor (human IgM anti-human IgG), con firmatory test for syphilis, and the latex test for pregnancy, which involves the de tection of human chorionic gonadotropin (HCG) in the urine of pregnant women. Precipitation Reaction REACTION IN SOLUTIONS. In contrast to the agglutination reaction, which takes place between antibodies and particulate antigen, the precipitation reaction takes place when antibodies and soluble antigen are mixed. As in the case of agglutination, precipitation of antigen-antibody complexes occurs because the divalent antibody molecules cross-link multivalent antigen molecules to form a lattice. When it reaches a certain size, this antigen-antibody complex loses its solubility and precipitates out of solution. The phenomenon of precipitation is termed the precipitin reaction. When increasing concentrations of antigen are added to a series of tubes that contain a constant concentration of antibodies, variable amounts of precipitate form. The weight of the precipitate in each tube may be determined by a variety of methods. If the amount of the precipitate is plotted against the amount of antigen added, a precipitin curve like the one shown in Figure 7.6 is obtained. There are three important areas under the curve shown in Figure 7.6: (1) the zone of antibody excess, (2) the equivalence zone, and (3) the zone of antigen excess. In the equivalence zone, the proportion of antigen to antibody is optimal for maximal precipitation; in the zones of antibody excess or antigen excess, the proportions of the reactants do not lead to efficient cross-linking and formation of precipitate. It should be emphasized that the zones of the precipitin curve are based on the amount of antigen-antibody complexes precipitated. However, the zones of antigen or antibody excess may contain soluble antigen-antibody complexes, particularly the zone of antigen excess where a minimal amount of precipitate is formed, but large amounts of antigen--antibody complexes are present in the supematant. Thus, the amount of precipitate formed is dependent on the proportions of the reactant antigens and antibodies: the correct proportion of the reactions result in maximal formation of precipitate; excess of antigen (or antibody) results in soluble complexes. PRECIPITATION REACTIONS IN GELS. Precipitation reactions between soluble antigens and antibodies can take place not only in solution but also in semisolid media such as agar gels. When soluble antigen and antibodies are placed in wells cut in the gel (Fig. 7.7A), the reactants diffuse in the gel and form gradients of concentration, with the highest concentrations closest to the wells. Somewhere between the two wells, the reacting antigen and antibodies will be present at proportions that are optimal for formation of a precipitate. If the antibody well contains antibodies 1, 2, and 3 specific for antigens 1, 2, and 3, respectively, and if antigens 1, 2, and 3, placed in the antigen well, diffuse at different rates (with diffusion rates of 1 >> 2 >> 3), then three distinct precipitin lines will form. These three precipitin lines form because anti-1, anti-2, and anti3, which diffuse at the same rate, react independently with antigens 1, 2, and 3, respectively, to form three equivalence zones and thus three separate lines of precipitate (Fig. 7.7B). Different rates of diffusion of both antibody and antibody and antigen result from differences in concentration, molecular size, or shape. This double-diffusion method, developed by Ouchterlony, where antigen and antibody diffuse toward each other, is very useful for establishing the antigenic relationship between various substances, as shown in Figure 7.8. Three reaction patterns are seen in gel diffusion, each of which is illustrated in Figure 7.8: patterns of identity, patterns of nonidentity, and patterns of partial identity. Patterns of Identity. In the example given on the left in Figure 7.8, the central well contains antibodies and the peripheral wells contain identical antigens. The antibodies diffuse from the central well toward the antigens that, since they are identical, form one continuous, coalescing precipitin line. This pattern, formed when the two antigens are identical, is termed a pattern of identity. Patterns of Nonidentity. In the example in the center of Figure 7.8, the central well contains antibodies to antigen 1 and antibodies to antigen 2, two nonrelated antigens, and the peripheral wells contain the two nonrelated antigens, antigen 1 and antigen 2. The two antibody (immunoglobulin)populations diffuse toward the peripheral wells. Antigen 1 and antigen 2 diffuse from the two peripheral wells toward the antibodies. Each antigen forms an independent precipitin line with its corresponding antibody at an equivalent point. The precipitin lines cross each other since each antigen diffuses across the band formed by the other antigen until it meets its specific antibody diffusing toward it. A pattern where the precipitin lines cross each other denotes nonidentity of the two antigens. Patterns of Partial Identity. The pattern of partial identity is shown in the fight-hand portion of Figure 7.8, where the center well contains antibodies to various epitopes of antigen 1. The reaction of these antibodies with antigen 1 results in a precipitin line. Antigen 2, however, contains some (but not all) of the epitopes present on antigen 1. Thus, some of the antibodies to antigen 1 will also combine with antigen 2. This partial identity between the two antigens is responsible for the coalescence of the two lines to give a line of identity. However, antibodies that do not bind with antigen 2 will pass through this line of precipitate, combine with antigen 1 on the other side, and form a spur. This pattern, with the formation of a spur, denotes partial identity, signifying that antigen 1 and antigen 2 share epitopes, with antigen 1 having more epitopes (and being able to react with more antibody populations) than antigen 2. RADIAL IMMUNODIFFUSION. The radial immunodiffusion test, represents a variation of the diffusion test. The wells contain antigen at different concentrations, while the antibodies are distributed uniformly in the agar gel. Thus, the precipitin line is replaced by a precipitin ring around the well. The distance the precipitin ring migrates from the center of the antigen well is directly proportional to the concentration of antigen in the well. The relationship between concentration of antigen in a well and the diameter of the precipitin ring can be plotted .If wells, such as F and G, contain unknown amounts of the same antigen, the concentration of that antigen in these wells can be determined by comparing the diameter of the precipitin ring with the diameter of the ring formed by a known concentration of the antigen. An important application of radial immunodiffusion is its use clinically to measure concentrations of serum proteins. To do so, antiserum to various serum proteins is incorporated in the gel; concentration of a particular protein in a serum sample is determined by comparing the diameter of the resulting precipitin ring with the diameter obtained by known concentrations of the protein in question. IMMUNOELECTROPHORESIS.Immunoelectrophoresis involves separating a mixture of proteins in an electrical field (electrophoresis) followed by their detection with antibodies diffusing into the gel. It is very useful for the analysis of a mixture of antigens by antiserum that contains antibodies to the antigens in the mixture. For example, in the clinical characterization of human serum proteins, a small drop of human serum is placed in a well cut in the center of a slide that is coated with agar gel. The serum is then subjected to electrophoresis, which separates the various components according to their mobilities in the electrical field. After electrophoresis, a trough is cut along the side of the slides, and antibodies to human serum proteins are placed in the trough. The antibodies diffuse in the agar, as do the separated serum proteins. At an optimal antigen:antibody ratio for each antigen and its corresponding antibodies, they form precipitin lines. The result is a pattern similar to that depicted in Figure 7.10. Comparison of the pattern and intensity of lines of normal human serum with the patterns and intensity of lines obtained with sera of patients may reveal an absence, overabundance, or other abnormality of one or more serum proteins. WESTERN BLOTS (IMMUNOBLOTS). In the Western blot (immunoblot) technique, antigen (or a mixture of antigens) is first separated in gel. The separated material is then "blotted" omo nitrocellulose sheets to which the antigen binds strongly. Amibody, which is then applied to the nitrocellulose sheet, binds with its specific antigen. The antibody may be labeled (e.g., with radioactivity), or a labeled anti-immunoglobulin may be used to localize the antibody and the antigen to which the first antibody is bound. These so-called "Western blots" are gaining wide application in research and clinical laboratories for the characterization of antigen. IMMUNOASSAYS Direct Binding Immunoassays Radioimmunoassay (RIA) employs isotopically labeled molecules and permits measurements of extremely small amounts of antigen, antibody, or antigen-antibody complexes. The concentration of such labeled molecules is determined by measuring their radioactivity, rather than by chemical analysis. The sensitivity of detection is thus increased by several orders of magnitude. For the development of this highly sensitive analytical method that has tremendous applications in hormone assays as well as assays of other substances of biological importance, RosalynYalow received the Nobel Prize. The principle of radioimmunoassay is illustrated in Figure 7.11A,B. A known amount of radioactively labeled antigen is reacted with a limited amount of antibody. The solution now contains antibody-bound labeled antigen, as well as some unbound labeled antigen. After separating the antigen bound to antibody from free antigen, the amount of radioactivity bound to antibody is determined (Fig. 7.1 lA). The test continues with performance of a similar procedure in which the same amount of labeled antigen is premixed with unlabeled antigen (Fig. 7.1 lB). The mixture is reacted with the same amount of antibody as before, and the antibody-bound antigen is separated from the unbound antigen. The unlabeled antigen competes with the labeled antigen for the antibody and, as a result, less label is bound to antibody than in the absence of unlabeled antigen. The more unlabeled antigen present in the reaction mixture, the smaller the ratio of antibodybound, radiolabeled antigen to free, radiolabeled antigen. This ratio can be plotted as a function of the concentration of the unlabeled antigen used for competition. To determine an unknown concentration of antigen in a solution, a sample of the solution is mixed with predetermined amounts of labeled antigen and antibody. The ratio of bound/free radioactivity is compared with that obtained in the absence of unlabeled antigen (the latter value is set at 100%). An important step in performing a radioimmunoassay, as described above, is the separation of free antigen from that bound to antibody. Depending upon the antigen, this separation can be achieved in a variety of ways, principal among which is the anti-immunoglobulin procedure. The anti-immunoglobulin procedure is based on the fact that antigen (labeled or unlabeled) bound to immunoglobulin will also be precipitated, following the addition of anti-immunoglobul!n, antibodies, so that only unbound antigen remains in the supematant. Radlolmmunoassays commonly employ rabbit antibodies to the desired antigens. These rabbit antibody-antigen complexes may be precipitated by the addition of goat antibodies raised against rabbit immtmoglobulins. Since the amounts of antigen and antibody required for radioimmunoassay are extremely small, the antigen-antibody complexes reacted with anti-immunoglobulin would form only tiny precipitates. It is difficult, if not impossible, to recover these precipitates quantitatively by conventional means, in order to determine their radioactivity. To overcome this problem, it is customary to add nonspecific immunoglobulins to the reaction mixture, thereby increasing the amount of total immunoglobulins to an amount that can easily be precipitated by anti-immunoglobulins and recovered quantitatively. Such precipitates consist mainly of nonspecific immunoglobulins to which radioactive antigen does not bind. However, they also contain the extremely small amount of antigen-specific immunoglobulin and any radioactive antigen bound to it. An alternative method of separating complexes of antigen bound to antibody from free antigen is based on the fact that immunoglobulins become insoluble and precipitate in a solution containing 33% saturated ammonium sulfate. If the antigen does not precipitate in 33% ammonium sulfate, the addition of ammonium sulfate to 33% will cause the antibody complexed to antigen to precipitate, leaving the free antigen in solution. Here again, the amounts of antibodies reacting with antigen (or free antibodies) is so small, unable to form precipitates. As described for the radioimmunoassay where anti-immunoglobulins are used for the separation of antibody-bound antigen from free antigen, a sufficient amount of nonspecific immunoglobulins is added to the mixture so that an appreciable precipitate will form at 33% saturation ammonium sulfate to enable the separation of free antigen from antigen bound to antibody. Solid-Phase Immunoassays Solid-phase immunoassay is one of the most widely used immunologic techniques. It is now automated and is widely used in clinical medicine for the detection of antigen or antibody. Solid-phase immunoassays employ the property of various plastics (e.g., polyvinyl or polystyrene) to adsorb monomolecular layers of proteins onto their surface. Although the adsorbed molecules may lose some of their antigenic determinants, enough remain unaltered and can still react with their corresponding antibodies. The presence of these antibodies, bound to antigen adsorbed onto the plastic, may be detected by the use of anti-immunoglobulins (Fig. 7.13) labeled with a radioactive tracer or with an enzyme. A solid-phase assay that uses radioactive anti-immunoglobulins is termed a solid-phase radioimmunoassay (SPRIA). If the test uses anti-immunoglobulins that are labeled with an enzyme that can be detected by the appearance of a color on addition of proper substrate, the test is called an enzyme-linked immunosorbent assay (ELISA). Because of problems associated with disposal or radioactive waste and the cost of radiation measuring instruments, the ELISA is rapidly replacing SPRIA. It should be emphasized that after coating the plastic surface with antigen, it is imperative to "block" any uncoated plastic surface to prevent it from absorbing the other reagents, most importantly the labeled reagent. Such "blocking" is achieved by coating the plastic surface with a high concentration of an unrelated protein, such as gelatin, after the application of the antigen. Solid-phase immunoassay may be used to detect the presence of antibodies to the antigen that coats the plastic. Since the plastic wells are usually coated with relatively large amounts of antigen, the higher the concentration of antibodies bound with the antigen, the higher the amount of labeled anti-immunoglobulin that can bind to the antibodies. Thus, it is ~mportant always to use an excess of labeled anti-immunoglobulin to assure saturation. Solid-phase immunoassay may be used for the qualitative or quantitative determinations of antigen. Such determinations are performed by mixing the antiserum with varying known amounts of antigen before adding the antiserum to the antigen-coated plastic wells. This preliminary procedure results in the binding of the antibodies with the soluble antigen, decreasing the availability of free antibodies for binding with the antigen that is coating the plastic. The higher the concentration of the soluble antigen that reacts with antibodies prior to addition of the antibody to the wells, the lower the number of antibodies that can bind with the antigen on the plate, and the lower the number of labeled anti-immunoglobulin that can bind to these antibodies. The decrease in the amount of bound label as a function of the concentration of antigen used to cause this decrease can be plotted, and the amount of antigen in an unknown solution can then be determined from the graph by a comparison of the decrease in bound label caused by the unknown solution to the decrease caused by known concentrations of pure antigen. IMMUNOFLUORESCENCE A fluorescent compound has the property of emitting light of a certain wave length when it is excited by exposure to light of a shorter wavelength. Immunoflu orescence 'is a method for localizing an antigen by the use of fluorescence-labeled antibodies. The procedure, originally described by Coons, employs antibodies to which fluorescent groups have been covalently linked without any appreciable change in antibody activity. One fluorescent compound that is widely used in immunology isfiuorescein isothiocyanate (FITC), which fluoresces with a visible greenish color when ex cited by ultraviolet light. FITC is easily coupled to free amino groups. Another widely used fluorescent compound is tetramethyl rhodamine isothiocyanate (TRITe), which fluoresces red-orange and is also easily coupled to free amino groups. Specially constructed microscopes permit visualization of fluorescent antibody on a microscopic specimen, and fluorescent antibodies are widely used to localize antigens on various tissues and microorganisms. There are two important and related procedures that employ fluorescent antibodies: direct immunofluorescence and indirect immunofluorescence. Direct Immunofluorescence Direct immunofluorescence is primarily for detection of antigen and involves reacting the target tissue (or microorganism) with fluorescently labeled specific antibodies. It is widely used clinically for identifying lymphocytic subsets and for demonstrating the presence of specific protein deposition in certain tissues such as kidney and skin in cases of systemic lupus erythematosus (SLE). Indirect Immunofluorescence Indirect immunofluorescence involves first reacting the target with specific antibodies. This reaction is followed by subsequent reaction with fluorescently labeled anti-immunoglobulin. The indirect immunofluorescence method is more widely used than the direct method, because a single fluorescent anti-immunoglobulin antibody can be used to localize antibody of many different specificities. Moreover, since the antiimmunoglobulins contain antibodies to many epitopes on the specific immunoglobulin, the use of fluorescent anti-immunoglobulins amplifies manyfold the fluorescent signal. FLUORESCENCE-ACTIVATED CELL-SORTING ANALYSIS A very powerful tool has been developed around the use of fluorescent antibody specific for cell-surface antigens. This is the technique of fluorescence-activated cell sorting (FACS). A cell suspension labeled with specific fluorescent antibody is passed through an apparatus that forms a stream of small droplets each containing one cell. These droplets are passed between a laser beam of ultraviolet light and a detector for picking up emitted fluorescence when a labeled cell is present in the droplet. This emitted signal is passed to an electrode that charges the droplet, leading to its deflection (Fig. 7.14) in an electromagnetic field. Thus as all droplets fall through they are collected and counted depending on whether they emit a signal. By sophisticated electronics the fluorescein-stained and unstained cells may be counted, as well as the intensity of fluorescence on each. With this type of apparatus it is now possible to rapidly develop a profile of a pool of lymphocytes based on their differential expression of cell-surface molecules, the relative amount of cell-surface molecule expressed on each cell, and the size distribution and numbers of each cell type. It is also possible to use the apparatus to sort a collection of cells stained with five or more different fluorescent labels and obtain a very homogeneous sample of a particular cell type. A variation of this technique uses fluorescent antibodies coupled to magnetic beads to separate cell populations. Cells which bind to the fluorescent antibody can be separated from unstained cells by a magnet. Both FACS and magnetic bead separation methods have resulted in the isolation of very rare cells such as stem cells (see Chapter 8). IMMUNOABSORPTION AND IMMUNOADSORPTION Because of the specific binding between antigen and antibody, it is possible to "trap," or selectively remove, an antigen against which an antibody is directed from a mixture of antigens in solution. By the same token, it is possible to trap or remove selectively the antigen-specific antibodies from a mixture of antibodies, using the specific antigen. There are two general methods by which this removal can be achieved. The methods are related, but, in one method, the absorption is done with both reagents in solution (immunoabsorption); in the second method, it is performed with one reagent attached to an insoluble support (immunoadsorption). Immunoadsorption is of particular value because the adsorbed material can be recovered from the complex by careful treatments that dissociate antigen-antibody complexes, such as lowering the pH (HCl-glycine or acetic acid, pH 2-3) or adding chaotropic ions. This enables one effectively to purify antigens or antibodies of specific interest. MONOCLONAL AND GENETICALLY ENGINEERED ANTIBODIES Monoclonal Antibodies The specificity of the immune response has served as the basis for serologic reactions in which antibody specificity is used for the qualitative and quantitative determination of antigen. The discriminating power of serum antibody is not without limitations, however, because the immunizing antigen, which usually has many epitopes, leads to production of antisera that contain a mixture of antibodies with varying specificity for all the epitopes. Indeed, even antibodies to a single epitope are usually mixtures of immunoglobulins with different fine specificities, and therefore different affinities for the determinant. Furthermore, immunization with an antigen expands various populations of antibody-forming lymphocytes (see Chapter 11). These cells can be maintained in culture for only a short time (on the order of days), so it is impractical, if not impossible, to grow normal cells and obtain clones that produce antibodies of a single specificity A quantum leap in the resolution and discriminating power of antibodies was provided in the 1970s with the developmem of methods for the generation of monoclonal antibodies by Milstein and K6hler, who shared the Nobel Prize for this development. Monoclonal antibodies are homogeneous populations of antibody molecules, derived from a single antibody-producing cell, in which all antibodies are identical and of the same precise specificity for a given epitope. Milstein and Kohler took advantage of the properties of malignant plasma cells, which are "immortal" and can be maintained in culture for years. They selected a population of malignant plasma cells unable to secrete immunoglobulin and also deficient in the enzyme hypoxanthine phosphoribosyl transferase (HPRT) that would die unless HPRT were introduced. The malignant cells were then fused with antibody-producing cells, which have HPRT. The biotechnology for the production of hybridomas, developed by K6hler and Milstein, is illustrated in Figure 7.15. The fusion is generally accomplished by thc use of polyethylene glycol (PEG) or by inactivated Sendai virus. The nuclei of the hybrids also fuse, and the hybridoma cells then possess both the capacity to manufacture immunoglobulins and the ability to survive in culture in a select medium such as that containing hypoxanthine, aminopterin, and thymidine (HAT). The enzyme deficiency of the malignant cell results in its death in this medium unless that deficiency is corrected by the acquisition of the enzyme-producing genes derived from the antibody-producing cell. Thus, the hybrids can be separated from the contaminating malignant cells, which do not survive, because they have not acquired the enzyme. Those hybrid cells synthesizing specific antibody are selected by some test for antigen reactivity and then cloned from single cells and propagated in tissue culture, each clone synthesizing antibodies of a single specificity. These highly specific, monoclonal antibodies are used as reagents for numerous procedures, ranging from specific diagnostic tests to "magic bullets" in immunotherapy of cancer (see Chapter 21). In immunotherapy, various drugs or toxins are conjugated to monoclonal antibodies, which, in mm, "deliver" these substances to the tumor cells against which the antibodies are specifically directed. Genetically Engineered Antibodies To date most of the monoclonal antibodies are made in mouse cells. These are suitable for diagnostic and many other purposes. However, their administration into humans carries the complication that the patient will form antibodies to the mouse immunoglobulins. Attempts to develop in vitro human monoclonal antibodies are by and large quite difficult. Human monoclonal antibodies are being currently produced by genetic engineering utilizing several approaches. One method utilizes the technology of recombinant DNA to produce a chimeric monoclonal antibody. This molecule consists of the constant region of human immunoglobulin and a variable region of a mouse immunoglobulin. A more recent technology utilizes the polymerase chain reaction (PCR) to generate gene libraries of heavy and light chains from DNA obtained from hybridoma cells or plasma cells, joining at random numerous heavy and light chains and screening the resulting Fab clones for antibody activity against a desired antigen. With this technology it is now possible to produce millions of clones of different specificities, to rapidly screen them for the desired specificity and generate the desired monoclonal Fab constructs without immunization and without the difficulties encountered with the production of monoclonal antibodies, especially human monoclonal antibodies. Although this chapter deals with antibodies, it seems appropriate to mention at this point that the hybridoma technology is not limited to the production of monoclonal immunoglobulins. In the late 1970s, methods for producing hybridomas were also developed for T cells, fusing lines of malignant T cells with nonmalignant, antigen-specific T lymphocytes whose populations have been expanded by immunization with antigen. T-cell hybridomas have been very useful for studying the relationship between T cells of a single specificity with their corresponding epitope. SUMMARY 1. The reaction between an antibody and an epitope does not involve covalent forces; it involves weak forces of interaction such as electrostatic, hydrophobic, and van der Waals forces. Consequently, for a significant interaction, the antibody combining site and the epitope require a close steric fit like a lock and key. 2. Only the reaction between a multivalent antigen and at least a divalent antibody can bring about secondary antigen-antibody reactions that depend on cross-linking of antigen molecules by antibodies. These secondary reactions do not take place with haptens or monovalent Fab. 3. The interaction between a soluble antibody and an insoluble particulate antigen results in agglutination. The extent of agglutination depends on the proportions o.f the interacting antibody and antigen. At high antibody levels, excess agglutination may not occur. Th~s is referred to as a prozone. The term titer refers to the highest serum dilution at which agglutination still takes place and beyond which, at higher dilution, no agglutination occurs. 4. Because of the zeta potential, which precludes some antigenic particles from approaching each other closer than a certain distance, IgG antibodies may be incapable of causing agglutination. IgM antibodies, however, can bind such distant particles and agglutinate them. 5. The use of heterologous anti-immunoglobulin antibodies may bridge between antigenic particles that are bound to nonagglutinating antibodies, leading to agglutination. This sequence of events is the basis for the Coombs test. 6. Precipitation reactions occur on mixing, at the right proportions, of soluble multivalent antigen and (at least) divalent antibodies. The precipitation reaction may take place in aqueous media or in gels. 7. The reaction in gels, between soluble antigen and antibodies, may be used for the qualitative and quantitative analysis of antigen or antibody. Ex amples are gel diffusion tests, radial diffusion tests, and immunoelectrophoresis. 8. Radioimmunoassay is a test that is based on competitive inhibition of nonlabeled and labeled antigen for antibody (in antibody deficiency). Antibody-bound antigen must be separated from nonbound labeled antigen. Separation is usually achieved by precipitation with anti-immunoglobulins. 9. Solid-phase immunoassay is a test that employs the property of many proteins to adhere to plastic and form a monomolecular layer. Antigen is applied to plastic wells, antibodies are added, the well is washed, and any antibodies bound to antigen are measured by the use of radiolabeled or enzyme-linked anti-immunoglobulins. 10. ELISA. This enzyme-linked immunosorbent assay is essentially a solid-phase immunoassay in which an enzyme is linked to the anti-immunoglobulin. Quantitation of enzyme-linked anti-immunoglobulins is achieved by colorimetric evaluation, after the addition of a substrate, which changes color on the action of the enzyme. 11. Immunofiuorescence. In immunofiuorescence, the antigen is detected by the use of fluorescence-labeled immunoglobulins. In direct immunofiuorescence, the antibody to the antigen in question carries a fluorescent label. In indirect immunofiuorescence, the antigen-specific antibody is not labeled; it is detected by the addition of fiuorescently labeled anti-immunoglobulin. 12. Monoclonal antibodies are highly specific reagents consisting of homogeneous populations of antibodies, all of precisely the same specificity toward an epitope.