Ouchterlony's experiment is of particular interest as a project for amateurs because it enables one to observe the reactions of various antigens and antibodies without either the risk of accidental infection or the direct use of experimental animals. To do the experiment Ouchterlony cut a pair of shallow depressions or wells in a block of moist agar. He put a few drops of solution containing antigen in one well and a like amount of antibody solution in the other. The fluids diffused toward each other through the gelatinous agar and reacted at an intermediate zone. There a thin white crescent was precipitated in the otherwise clear gel. The crescent indicated the chemical neutralization of the antigen by the antibody. The technique is known both as immunodiffusion and as Ouchterlony analysis.

Subsequently Curtis A. Williams, Jr., of the U.S. and Pierre Grabar of France introduced a variation in the procedure for investigating the several species of proteins that make up each kind of antigen. After the antigen has been placed in its agar well an electric potential is applied across a plate of the moist gel. The potential exerts a force on the charged molecules of protein; the force varies with the amount of the characteristic charge associated with each species of protein in the antigen. Accordingly the fractions migrate through the agar at differing but characteristic rates and separate into well-defined zones. When antibody is applied to the edges of the agar, it diffuses across the plate and into contact with the zones. White crescents of precipitation form at the sites of reaction. Williams and Grabar called the technique immunoelectrophoresis [see "Immunoelectrophoresis," by Curtis A. Williams, Jr.; SCIENTIFIC AMERICAN, March, 1960].

Richard La Fond, a graduate student of biochemistry at Harvard University, has made several experiments with both techniques to illustrate the details of the procedures. He writes:

"The proteins used in my experiments were obtained from a supplier of biochemicals. The materials can of course be prepared in small quantities by means of experimental animals, but the cost would be excessive and the techniques required to ensure purity involve the use of apparatus that is not generally available. I bought my biochemicals from Pentex, Inc., 195 West Birch Street, P.O. Box 272, Kankakee, Ill. 60901. Antibody is available from this firm in the form of antiserum from a variety of animals, including human beings.

"I used rabbit antibovine serum, which is supplied in a minimum quantity of two milliliters. Antigen comes in the form of animal blood serum. I obtained minimum quantities from the following animals: cat, cattle, chicken, dog, duck, goat, goose, guinea pig and pig.

"All serums come in rubber-capped bottles and will remain usable for up to a year if stored in a refrigerator. The fluid is removed from the bottles with a sterilized hypodermic syringe. The needle of the syringe is pushed through the cap of the bottle. Sterilized air, trapped inside the syringe, is injected into the bottle. Serum is then drawn into the syringe for transfer to the agar well. The volume of injected air should equal the volume of serum that is removed, a procedure that maintains atmospheric pressure inside the bottle and prevents room air from leaking into the bottle and contaminating the serum.

"I usually prepare at one time enough agar for several experiments. I use Agar Noble from Difco Laboratories, Inc., 920 Henry Street, Detroit, Mich. 48201. This material comes in the form of dry granules. I put .95 gram of Agar Noble in a flask and dilute it to 100 milliliters with a solution compounded from 6.7 grams of boric acid and 13.4 grams of sodium tetraborate decahydrate diluted to 1,000 milliliters with distilled water. The mixture of agar and buffer is heated until it just starts to boil; by then most of the agar will have dissolved and the solution should be clear. Small particles of undissolved agar are removed by pouring the hot agar solution through two thicknesses of cheesecloth into a receptacle. The agar can be stored in test tubes in a refrigerator and remelted for use by putting the tubes in hot water.

"To prepare an agar plate for an Ouchterlony analysis I pour melted agar into a Petri dish to a depth of four millimeters. For a Petri dish nine centimeters in diameter the amount of agar is about 15 milliliters. I use disposable plastic dishes, which are available from dealers in laboratory supplies, along with disposable Pasteur pipettes, which are convenient for handling small volumes of serum.

"The gel should solidify within 15 minutes at ordinary room temperature. I cut wells in the agar block with a No. 1 cork borer, which is about four millimeters in diameter. The cylindrical plugs thus cut are removed with a needle. A well made in the center of the block for antiserum (antibody) can be surrounded by as many as six wells for holding , selected blood serums (antigens).

"Ordinarily I first make a full-scale drawing of the desired well layout. The Petri dish containing the transparent agar is

placed on top of the drawing, which serves as a guide for cutting the wells. The wells are spaced about seven millimeters apart. The central well is filled with antiserum and the neighboring wells are filled with blood serum from selected animals. A diagram of the Twenty years ago the Swedish biologist Orjan Ouchterlony devised a simple experiment that amateurs can perform to investigate the chemistry of biological immunity, which includes such phenomena as the resistance an animal acquires against infectious microorganisms and the animal's rejection of tissue grafted from another animal. The experiment involves the use of the two kinds of complex protein known as antigen and antibody. Included among the many antigens are the toxins liberated by microorganisms that cause diseases. Antigen that enters the body of an animal stimulates the production of antibody. The antibody may then unite chemically with the antigen in a neutralizing reaction. Although almost all antigens consist of mixtures of protein molecules, each antigen is chemically unique. Collectively antigen mixtures are as varied as the organisms that make them. Each kind of antibody, also a mixture, is unique, but antibody that forms in response to a specific antigen may react more or less vigorously with other similar antigens and so confer on the animal a measure of immunity against several diseases.

Ouchterlony's experiment is of particular interest as a project for amateurs because it enables one to observe the reactions of various antigens and antibodies without either the risk of accidental infection or the direct use of experimental animals. To do the experiment Ouchterlony cut a pair of shallow depressions or wells in a block of moist agar. He put a few drops of solution containing antigen in one well and a like amount of antibody solution in the other. The fluids diffused toward each other through the gelatinous agar and reacted at an intermediate zone. There a thin white crescent was precipitated in the blood serum with saline solution prepared by dissolving .9 gram of table salt in distilled water to make 100 milliliters. In some cases it has been necessary for good results to add as many as 30 parts of saline solution to one part of blood serum. As the solution is made more dilute the pattern moves from the antibody well toward the antigen well and the secondary lines tend to disappear. I dilute by trial and error until the pattern appears about midway between the wells. This position indicates that the antigen and the antibody are reacting in equivalent amounts.

"Antibody that appears in an animal in response to the invasion of an antigen reacts strongly with the antigen. The pair of substances can be thought of as being complementary, analogous to a lock and its key. Chemically active sites on the antibody molecule fit complementary sites on the antigen molecule. Such antigen is known as homologous antigen. The same antibody may also react chemically with other antigens having combining sites that are chemically similar to those on the molecules of homologous antigen. Indeed, the antibody may react with only a portion of a single chemically similar site. Antigens of this kind are known as heterologous antigens. Such heterogeneity is found among animals of the same species, although it may be absent in individuals of the species. Foreign antigens, containing no complementary combining sites, do not react with the antibody; the agar plate remains clear even though the antibody and the foreign antigen diffuse into contact.

"In general the white precipitin forms patterns of three major types when antigen in more than one well reacts with antibody in a neighboring well. For ex ample, if two of three wells in a triangular array contain homologous antigen and the third well contains antibody, the pattern takes the form of a small bow that bends around the antibody well [see Figure 1 ]. In fact, two distinct bands of precipitin form, but the inner ends fuse, creating the bow, because proteins in the antigen wells have identical complementary binding sites for antibody. This pattern indicates that the antigens are identical; it is called the identity reaction. Had the antigen wells contained an antigen B, unrelated to another antigen C, and had the antiserum contained antibodies that were specific to both antigens, the pattern would have been a pair of crossed lines [Figure 6]. In this case each type of antibody diffuses through areas of high concentration of noncomplementary antigen without being precipitated until it reaches an area into which complementary antigen has diffused in optimum concentration.

"Patterns of the third type consist of an asymmetrical T: a spur extends from the point where the lines of precipitin intersect. This pattern indicates partial identity. The spur forms when a highly specific antibody reacts with its homologous antigen and with another antigen that has an additional combining site. In the accompanying drawing [right in illustration in Figure 6] antigens that have combining sites in common are designated A. Note that one of the two has an additional combining site a. Some constituents of the antibody react with both A and Aa and form two lines of precipitin. The spur is formed by the combination of antibody constituents that react only with Aa but are not precipitated by the heterologous A antigens. The length of the spur is proportional to the dissimilarity between the antigens. Conversely, the spur becomes less pronounced as antigens of increasing similarity react; absence of the spur indicates the identity reaction.

"The several proteins in an antigen are almost identical in form and structure. The difference may be extremely subtle, as twin structures the size of the Empire State Building would differ if a single stone in either building were turned through a right angle. Yet the difference can be detected by Ouchterlony analysis. For example, antibovine serum (antibody) reacts with bovine blood serum (antigen) to form a nested family of precipitin lines or bands, each marking the site where a fraction of the material reacted. A similar pattern forms when this antiserum reacts with the heterologous antigen in the blood serum of a sheep, illustrated by the accompanying photograph [Figure 2], in which the central well, A, contains antibovine serum, well 2 contains sheep blood serum and well 6 contains bovine blood serum. No reaction appears adjacent to wells 1, 3, 4 and 5, which contain the immunologically unrelated blood serum of pig, dog, cat and horse respectively. Hence cattle may be more closely related to sheep, at least immunologically, than they are to the other animals.

"It is not always easy to make an accurate count of the number of fractions in an antigen that react to form patterns, because the lines of precipitin tend to crowd together and merge. Separations become increasingly distinct, however, as the material is made more dilute. When the most distinct separations are desired, the experimenter switches to immunoelectrophoresis.

"In this technique agar is poured on the level surface of a clean glass plate so that a pronounced meniscus forms $2 around the four edges of the glass. The fluid gels in 15 minutes at room temperature. A single well is cut in the middle of the agar for blood serum (antigen). My plates are 2 1/2 inches wide and 5 1/4 inches long. These dimensions are not critical.

"The plate is supported at the ends by the edges of a pair of rectangular trays of clear plastic that are 11 1/4 inches long, 3 1/2 inches wide and 2 1/2 inches high. A baffle is put in each tray about 3/4 inch from the edge that supports the glass plate [see Figure 7]. The trays are filled with buffer solution of the kind used for preparing the agar, and the solution in each tray is connected electrically to the neighboring end of the agar plate by a wick of moist filter paper. One end of the wick drips into the buffer and the other end rests on the agar [see Figure 8]. The buffer in each tray is connected to the power supply through electrodes that I made by winding a few turns of platinum wire around rods of clear plastic five millimeters in diameter. Plastic handles were attached to the ends of the rods to facilitate handling. "The apparatus is enclosed in a box of clear plastic 15 inches long, 10 inches wide and 5h inches high. A plastic cover prevents the evaporation of water from the gel. Parts for the box and the trays were sawed from 1/4-inch sheet plastic and assembled with special cement that was bought with the plastic. Details of the power supply have already been described in this department [see "The Amateur Scientist, SCIENTIFIC AMERICAN; June, 1962].

"After the apparatus is assembled and the trays are filled with buffer solution I put a few drops of blood serum (antigen) in the agar well, cover the enclosure, plug in the power supply and apply approximately 50 volts across the agar plate for 90 minutes. The electric potential across the plate is measured with a portable voltmeter. If all goes well, the antigen separates into fractions that come to rest at several spots along the center line of the plate. Initially these areas are about the size of the well, but they gradually expand as the fractions diffuse through the agar.

"After electrophoresis is complete I lift the plate from the apparatus and cut in the agar, along the sides of the plate, a pair of straight troughs, each about one millimeter wide and 15 millimeters from the center line of the plate. The cuts are made with a pair of razor blades embedded one millimeter apart in a block of polystyrene. The strips of agar thus cut are lifted from the trough with a needle.

"Each trough is filled with a few drops of antibovine serum (antibody). Antibody and antigen fractions diffuse into contact. Depending on the potency of the antiserum, patterns of precipitin form within 24 hours. A typical set of immunoelectrophoretic patterns that indicate the reaction of various fractions contained in bovine blood serum (antigen) with antibovine serum (antibody) appear in the accompanying photograph [Figure 3].

"If a preparation of pure antigen is available, it is possible to identify it in the immunoelectrophoretic pattern of a complex mixture of antigens. For example, to identify the prominent arcs closest to the origin, as depicted in the photograph, I made a second well in the agar adjacent to the arcs and filled it with a solution of alpha globulins. The resulting identity reaction between the unknown antigen and the alpha globulins is shown by a second photograph [left].

"My experiments had as their principal objective the comparison of various animal blood serums. As proved by the experiment I have described, antibovine serum reacts most strongly with its homologous antigen, bovine blood serum, and with the serum of a closely related animal, the sheep. When bovine and sheep serums occupy adjacent wells and react, the resulting precipitin bands suggest that sheep and cattle may have many immunologically identical serum antigens, as illustrated by the accompanying photograph [Figure 4], in which wells 1 and 2 respectively contain bovine and sheep serums and well C contains antibovine serum. The plate is so crowded with lines of precipitin, however, that the recognition of individual identity patterns is all but impossible.

"To ascertain if any antigens exist that are exclusively specific to bovine serum it is necessary to remove from the antiserum all antibodies that react with sheep-serum antigens. The separation can be made by a technique that immunologists call the absorption of antiserum. To demonstrate the technique I mixed approximately .7 milliliter of sheep serum with .5 milliliter of antibovine serum and put the mixture in the antiserum well. Only the antibodies that diffuse from the antiserum well can be bovine-specific, because antibodies that react with sheep antigen are precipitated in the antiserum well. I surrounded this well, D, with three others, 1, 2 and 3, respectively containing goat, sheep and bovine serums. A line of precipitin formed between wells D and 3, indicating that at least one of the antigens is associated exclusively with bovine serum [see Figure 5].

"The absorption technique greatly extends the analytical power of the immunodiffusion procedure. It can also be employed with immunoelectrophoresis. For example, I made an electrophoretic separation of bovine serum (antigen). Then I placed in the troughs antibovine serum (antibody) that had been absorbed with sheep serum. As in the preceding experiment, the resulting electrophoretic pattern indicated that the bovine-specific antigen is a single component. The fraction did not migrate far from the well, which suggests that it is a protein of high molecular weight. It would be interesting to test bovine serums of individuals I that differ in breed and age to learn if I this antigen is universally present.

"Antigen and antibody preparations from many species, as well as from various individuals of a given species, have become available commercially in recent years. Hence numerous experiments involving Ouchterlony analysis and immunoelectrophoresis are within the reach of the enterprising amateur."

Bibliography

EXPERIMENTAL IMMUNOCHEMISTRY: REVISED AND ENLARGED. Elvin A. Kabat and Manfred M. Mayer. Charles C Thomas, Publisher, 1961.

IMMUNODIFFUSION. Alfred J. Crowle. Academic Press, 1961.

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.