Antigens and Antibodies: Heidelberger and The Rise of Quantitative Immunochemistry, 1928-1954
Michael Heidelberger's studies of the immunochemistry of antibodies, antigen, and complement redefined the field of immunology and founded it on the precise, quantitative basis of analytical chemistry. In the 19th century, Robert Koch and Louis Pasteur had established that each infectious disease was caused by a particular microorganism, and Pasteur had found that inoculations of attenuated (weakened but living) forms of several of these microorganisms or of small, repeated doses of their toxic products induced an immune reaction that provided long-term protection against the diseases they cause. Aided by the discovery by Theobald Smith and Daniel Salmon that even bacteria killed by heat could impart protective immunity and that, contrary to longstanding belief, no interaction between the body and living microorganisms was required for protection, scientists in short order developed vaccinations and antitoxins against cholera, anthrax, rabies, diphtheria, and tetanus.
Emil von Behring suggested in the early 1890s that immunity stemmed from protective substances in blood serum (the clear liquid portion of the blood that separates out upon clotting), which he called antibodies. Moreover, he observed that antibodies were highly specific, meaning that they protected against one disease but not others. Further work during the 1890s revealed that in the presence of immune or antisera (sera that contained antibodies to specific diseases, which were obtained by inoculating animals such as rabbits with small doses of the disease-causing microorganism or toxin), bacteria clumped together or dissolved, and were thus rendered harmless. During the same decade, Jules Bordet hypothesized the existence of an important accessory factor in the blood serum of non-immunized individuals, which later came to be called complement and which, while not itself an antibody, combined with antibodies and activated them to destroy antigens, foreign substances that trigger immune reactions. According to Bordet's hypothesis, antibodies provided immunological specificity--the ability to bind to particular antigens--while complement acted as the destructive agent. Finally, scientists in the early part of the 20th century noted a second mechanism of immunity provided by specialized cells in the body that engulf and digest bacteria and other foreign matter. Such phagocytic cells, it was soon suggested, required that bacteria be coated by antibody before they could effectively kill them.
Despite these practical advances and conceptual breakthroughs, however, the processes and physiological changes by which immunity was acquired were not understood until well into the 20th century. Measurements of the amount of antibodies and antigen in solution were relative and comparative, not in absolute units of weight, which meant that scientists could not begin to infer their chemical constitution, or how they bound to one another. The chemical mechanisms which led to agglutination (clumping together) and cytolysis (breaking open) of bacteria by antibodies were unclear. Whether complement was indeed a separate substance, or rather an innate property of antibodies, remained controversial. Similarly, the precise role of antibodies in phagocytosis was a mystery. The term immunochemistry was introduced by the Swedish physical chemist Svante Arrhenius in 1907 when he chose it as the title for a book of lectures, but he did so with apparent irony because so little was then known about the chemical reactions and structures of the immune system.
Between the 1920s and the 1940s, Michael Heidelberger began to resolve several of these mysteries by placing immunology on the precise footing of modern biochemistry, and by introducing mathematical and statistical methods to the field. He was "convinced that it was necessary for progress in immunology to have some better methods than just titrations and dilution end points . . . , and that the main problems of immunology would never be solved unless you could apply analytical microchemistry and get some real data in absolute units." He developed a new experimental approach to immunology and analytical chemistry by using type-specific polysaccharides of pneumococcus as effective reagents to obtain immune precipitates--combinations of antibodies and polysaccharide antigens--from which quantitative methods for the estimation of the weight of antibodies could be derived. Heidelberger and his collaborators later expanded these methods to protein antigens. His methods led to a deeper understanding of the basic chemistry and molecular mechanisms of immune reactions, as well as to the production of much more effective antisera for the treatment and prevention of bacterial infections like pneumonia and meningitis caused by Haemophilus influenzae bacteria.
After fifteen years at the Rockefeller Institute for Medical Research and one year at Mount Sinai Hospital, Heidelberger was offered a position--the first of its kind in the United States--as consulting research chemist to the Department of Medicine of the College of Physicians and Surgeons at Columbia University. The offer came from the new department chairman, Walter W. Palmer, whom Heidelberger had first met in the early 1920s in Donald D. Van Slyke's laboratory at the Rockefeller Institute. Palmer not only created a congenial academic atmosphere in which Heidelberger thrived, he approached the financier Edward Harkness for money to set up the $500,000 Harkness Research Fund, which paid Heidelberger's salary during his 27 years at Columbia and freed him from the need to apply for research grants.
Heidelberger's research at Columbia University proceeded along a clearly defined line from his and Oswald Avery's finding in 1923 that type-specific antigens of pneumococcus bacteria were polysaccharides. That line moved towards unraveling the chemistry of pneumococcal and other bacterial and plant polysaccharides, and towards using them as factors in reactions that revealed the nature of other components of the immune system, in particular antibodies and complement.
Based on Heidelberger and Avery's discovery that type II and type III pneumococcal polysaccharides contained no nitrogen, it became possible for Heidelberger and his first collaborator at Columbia, Forrest E. Kendall, to measure the amount of antibody in washed antigen-antibody precipitates. In experiments conducted during the early 1930s, Heidelberger and Kendall added polysaccharide antigens of pneumococcal types II and III to an acid solution containing antiserum, obtained by inoculating horses with pneumococcal polysaccharides, in order to detect antibodies. The amount of antibodies, which many immunologists (but not all) had long assumed to be proteins, was indicated by the quantity of nitrogen found in the resulting antigen-antibody precipitates, because antibody proteins, like almost all other proteins, contained nitrogen, while polysaccharides did not. Conversely, Heidelberger and Kendall added antibody-containing antiserum to a solution of pneumococcal antigens in order to detect polysaccharide.
Their experiments revealed three distinct zones within the solution in which the precipitin reaction between antigen and antibody unfolded: a zone of antibody excess, in which added polysaccharide antigen precipitated more antibody; an equivalent zone, in which neither antigen nor antibody could be found as all of them formed antigen-antibody complexes and precipitated out; if more antigen was added, a third zone could be detected in which precipitation of antibody nitrogen declined and antigens were in excess. Heidelberger and Kendall found that the highest rate of precipitation of antibody was found at the antigen excess end of the equivalent zone. From these findings they developed a quantitative theory of immune precipitation, which expressed the course of the precipitin reaction curve as a linear equation derived from the Law of Mass Action. (The prominent physical chemist Linus Pauling arrived at a similar equation shortly afterwards.)
Heidelberger and his first graduate student and long-time Columbia University colleague, Elvin A. Kabat (1914-2000), subsequently extended this quantitative theory to agglutination, the clumping together of antigen-bearing cells, microorganisms, or particles in suspension by the introduction of specific antibodies (agglutinins). They showed that agglutination and precipitation were two functions of the same antibodies--that agglutination was a precipitin reaction on the surface of antigen-bearing cells--and not an indication of the existence of two distinct types of antibodies, as other immunologists had argued.
The principle that all of a given antigen is precipitated throughout the antibody excess and the equivalence zones was subsequently confirmed for other polysaccharide as well as for protein antigens by washing the precipitates and analyzing them for distinctive chemical constituents of the antigen. Similarly, by applying the precipitin curve and analyzing washed precipitates, Heidelberger was able to measure the amount of antigen in a biologic fluid, such as cerebrospinal fluid, by comparing it with the amount of antibody nitrogen precipitated, whose weight was known.
From their theory Heidelberger and Kendall predicted that analytically pure antibody could be dissociated from polysaccharide-antibody precipitates through the use of strong salt. After having obtained purified antibodies in this way for the first time, they were able to show conclusively that they were protein, until then a matter of scientific dispute. Specifically, they showed them to be modified serum globulins, a class of proteins that are insoluble in water but are soluble in saline solutions.
Their experiments further revealed the physiochemical heterogeneity of antibodies, which often combine in complex patterns in precipitating antisera (and may not precipitate with antigen when separated from other precipitating antibodies with which they are normally associated). Moreover, Heidelberger and Kendall's findings established that antigens and antibodies were multivalent, meaning that they could form two (in the case of antibodies) or more (in the case of antigens) chemical bonds with one another, bonds through which they combined into lattice-like structures. Taken together, Heidelberger, Kendall, and Kabat's theories and experimental results allowed much more accurate inferences regarding how and where antibodies are formed in animals, and what their physical and chemical properties are.
Of immediate clinical significance was the fact that the quantitative precipitin method made it possible to measure the total antibody nitrogen content of antisera, and thus to compare various antisera to a given antigen with regard to their combining proportions. Since antibodies contained in antisera can impart a level of immunity when injected into the bloodstream, Heidelberger was able to provide much more effective antiserum for use in immunizations by measuring the quantity of antibody in antiserum and by increasing it accordingly. For several years, he personally prepared all of the antibody solutions for type I, II, and III pneumococcus administered at Presbyterian Hospital, the teaching hospital of Columbia University. During treatment he measured the antibody content in patients' blood to test whether the solutions were effective, and adjusted the dosage accordingly.
In the early 1940s, just before the advent of penicillin, Heidelberger developed an antiserum for meningitis in infants caused by Haemophilus influenzae, a bacillus of the pharynx long thought to be the cause of epidemic influenza in humans. By varying the dosage and routes of injection of the bacilli into rabbits, and by measuring the agglutinin antibody count of the sera produced by the rabbits in response to the inoculations, he was able to provide an antiserum with ten times more antibodies than existing ones. The improved antiserum reduced the mortality of influenzal meningitis from ninety percent to five percent. In 1946, he and his collaborator on the project, Hattie E. Alexander, along with Catherine F. C. MacPherson and G. Leidy, isolated five groups of specific polysaccharide antigens of Haemophilus influenzae that made the bacillus pathogenic.
During the 1930s and 1940s, investigators in a number of countries--almost all of them trained by Heidelberger himself or by one of his collaborators--began using his quantitative methods to study other classes of polysaccharides from pathogenic bacteria of various species for type specificity, thereby launching a period of great advances in immunochemistry. Heidelberger remained at the forefront of these efforts. During the next several decades, in excess of one hundred type-specific capsular polysaccharides of pneumococcus alone were identified, several by Heidelberger and his collaborators.
Many newly-discovered polysaccharide antigens were found to cross-react with anti-pneumococcal sera and polysaccharides prepared from diverse species of microorganisms of clinical significance, as well as from plant resins and gums. For example, Heidelberger found that oxidized cotton contained multiple units of cellobiuronic acid, the same disaccharide of glucose and glucuronic acid contained in the capsular polysaccharides of type III and type VIII pneumococcus. He showed that this substance cross-reacted with type III and type VIII antipneumococcus horse and rabbit sera in the same way as the capsular polysaccharide, and that it could thus be used to confirm the correlation between chemical constitution and bacterial specificity. Heidelberger focused on such immunological cross reactions for the next six decades because they allowed inferences about the chemical structure of polysaccharide antigens, and about how their structure determined their reactions with other antigens and with antibodies.
In the late 1930s and early 1940s, Heidelberger extended his quantitative analytical method to the study of complement, a complex, essential component of host defense mechanisms against invading organisms. Since Bordet, the term had referred to the heat-sensitive factors in serum that trigger cytolysis, the dissolution of antibody-coated cells. Using his trademark microanalytical techniques, Heidelberger demonstrated that complement added weight to immune precipitates, thereby deciding the long-standing dispute as to whether complement had actual substance, or was merely an unstable, colloidal, finely divided state of serum proteins.
By measuring the weight of complement and evaluating the role of calcium and magnesium ions in complement fixation reactions (the consumption of complement upon reaction with immune complexes containing complement-fixing antibodies, and the basis of complement fixation tests widely used to detect antigens or antibodies), Heidelberger and his students Manfred Mayer, Abraham George Osler, and Otto Bier gave the first indications that complement was a functionally related system of several different serum proteins. Over twenty are known today. Mayer went on to show that complement plays a key enzymatic role not just in cytolysis, but in other immune responses, including phagocytosis, the engulfing of foreign matter by immune cells, and anaphylaxis, a form of hypersensitivity reaction to a specific antigen with often life-threatening consequences.
With the entry of the United States into World War II, Heidelberger became involved in military research, particularly in efforts by the U.S. Army to immunize soldiers against pneumonia with pneumococcal polysaccharides, which Heidelberger predicted would trigger the production of antipneumococcal antibodies in humans (as it did in horses and rabbits). In a preliminary trial Heidelberger and his assistants immunized several dozen student volunteers from Columbia University's medical school with single and mixed-type capsular pneumococcal polysaccharides, then applied the quantitative precipitin reaction to determine the antibody response. None of the immunized students contracted pneumonia, but since none of the students in a control group did, either, the effectiveness of the vaccine could not be proven until it was administered in 1944 to a much larger group of 17,000 airmen at a training camp in Sioux Falls, South Dakota, which had been struck by the disease. Using pneumococcal polysaccharides, by then produced in commercial quantities by Squibb, Heidelberger and his collaborators Colin MacLeod (who, with Avery and Maclyn McCarty, had discovered that DNA was the chemical substance of heredity) and Richard G. Hodges were able to report that recipients received protection against four types of pneumonia, whereas soldiers in a control group continued to show incidences of these types of pneumonia. During the 1950s, immunizations under the auspices of the World Health Organization with a mixture of pneumococcal and, later, of meningococcal polysaccharides devised by Emil Gotschlich helped greatly to reduce pneumonia and meningitis, especially in developing countries.
During the war, Heidelberger and Mayer also conducted the first controlled active immunization study for vivax malaria, to which many American soldiers engaged in the campaign on the Solomon Islands were falling victim. Heidelberger administered formalin-killed sporozoites (the malarial parasites plasmodium vivax in their infective stage) between relapses of the disease, but the results were discouraging and the relapse rate was not reduced.
Moreover, during the war Heidelberger and Kabat launched classified research on a vaccine against ricin, the powerful toxin of the castor bean and a potential biological weapons agent, after the Japanese government had asked its citizens to grow castor beans (to obtain their oil for use as lubricants in airplane engines, as it turned out). In order to study its antigenic effects, Heidelberger and Kabat first had to purify ricin, the results of which showed that it was a mixture of a highly toxic and a much less toxic form. By purifying ricin they made it more lethal, a blurring of offensive and defensive biological weapons research that troubled Heidelberger and spurred his fervent peace activism in the years after World War II.