The Chemistry of Antibody Globulins 193060

The history of protein chemistry is the history of a sequence of innovations in instrumentation and technique, each of them pointing to a new vein of knowledge to be worked. Most of these were fractionation methods; proteins could be separated in terms of charge or particle size, and biological activity correlated with the resulting serum fractions or protein fragments. For most of the period, efforts in this area were more or less independent of the work on immune specificity, but they too had their roots in colloid chemistry. Not until after World War II did protein chemists begin to feel that the high aspirations of colloid chemistry were an absurdity better forgotten.

The earliest and simplest preparative separation method for serum proteins was "salting-out," achieved by adding neutral salts to the protein solution, a technique introduced in the mid-nineteenth century. By 1930 there were still some chemists who were not convinced that the serum antibodies were actually globulins, and not some other material (Spiegel-Adolf 1930).

With their sophisticated techniques and mathematical approach, the physical chemists working at Uppsala in Sweden transformed colloid chemistry. The vitalistic enthusiasm of Pauli faded into a hardline physical science. However, both The Svedberg and Arne Tiselius thought of themselves as colloid chemists.

Svedberg's work on protein separation began to appear in 1925. Using a high-speed centrifuge, he found that proteins were forced to the bottom of a tube in order of their size. He assigned to each a sedimentation coefficient S that indicated molecular weight. The serum globulin mainly had a sedimen tation coefficient of 7 S, corresponding to a molecular weight of about 15,000, but there was a small amount of heavier 18 S globulin. As in the salting-out technique, antibody activity went along with the globulin fraction. Svedberg saw ultracentrifugal analysis as a contribution to classical colloid chemistry: Particle size and dispersal of proteins in solution were central to the colloid tradition.

In 1930 Svedberg's younger colleague Tiselius developed an apparatus for the separation of protein molecules in terms of their charge, based on the work of Landsteiner and Pauli. Tiselius, too, felt that the study of electrokinetic phenomena was among the most important of the tasks of colloid chemistry (Tiselius 1938).

Svedberg's and Tiselius's earlier methods had been useful for the analysis of protein mixtures, whereas the new techniques of the 1940s made it possible to prepare the products on a large scale. The name chromatography was coined by the Russian botanist M. Tswett in 1910 to describe his method of separating colored plant materials. The first good chromatographic method for proteins was introduced in 1944 by A. J. P. Martin and his colleagues at St. George's Hospital in London. Martin's method was to use a filter-paper sheet as the adsorbent, with the solvent containing the test material flowing slowly up a vertical sheet by capillary action. It also incorporated the novel idea of two-dimensional separation. When separation with one solvent was complete, the paper was turned 90 degrees and another separation with another solvent, or with an electric current, was performed. This method, called "fingerprinting," was good for identifying differently charged amino acids in a mixture of peptide fragments of proteins, and it became a favorite with protein geneticists.

After the war ended, a new group of methods for the separation of charged molecules appeared. Ion exchangers are essentially insoluble acids or bases in the form of a highly cross-linked matrix, usually a resin, through which a solvent carrying the material to be analyzed can percolate. Molecules with the same charge as the solid are repelled by it to various degrees and pass more or less quickly through, whereas those with the opposite charge are entangled and delayed. In 1951 an ion-exchange resin was brilliantly applied to the separation of amino acids by Stanford Moore and William H. Stein at the Rockefeller Institute in New York (Moore and Stein 1951). They not only obtained a quantitative separation of mixtures containing an astonishing number of different substances, up to 50 in some cases, but also automated the procedure so that it could be run unattended overnight. In the morning, the researchers would find a series of tubes filled by an automatic fraction collector, ready for them to analyze. This type of method with its greatly increased productivity made possible the rapid accumulation of information on amino acids and their sequence in many different proteins.

Protein molecules, however, easily lost their biological activity with the rough handling that resins gave them. Investigators might succeed in isolating a protein peak by a preparative method, only to find that all the antibody activity had disappeared, "lost on the column." It was not long before a better way was found. Cellulose-based ion exchangers were introduced by Herbert A. Sober and Elbert A. Peterson and by Jerker Porath at Uppsala (Sober et al. 1956). The anion exchanger was diethylaminoethyl cellulose, or DEAE, and the cation, carboxymethyl (or CM) cellulose. These materials turned out to be far less destructive to the protein molecule. A second type of material developed by Porath was a dextran gel called Sephadex by its manufacturer. It was not an ion exchanger; it acted, to use Porath's (1960) expression, as a kind of molecular sieve. Large molecules, such as the 18 S globulin, passed almost unhindered through it, excluded from the finer pores of the gel; smaller ones, such as albumin and 7 S globulin, wandered more slowly, exploring the labyrinth within. All these materials were soon available commercially, many from firms centered in Uppsala. As the techniques and the rather expensive equipment for column chromatography and automatic fraction collecting spread through laboratories, protein chemistry, particularly the chemistry of antibody globulins, moved very rapidly (Morris and Morris 1964).

The range of separation methods now available made possible an attack on the structure of the immunoglobulin molecule. Rodney Porter, working at St. Mary's Hospital Medical School in London, separated out a rabbit immunoglobulin by chromatography on DEAE, and then digested it with the proteolytic enzyme papain to break it into sections small enough to be understood. The digest was first put into the ultracentrifuge. There was only one peak, at 3.5 S, showing that the 7 S globulin had broken into smaller fragments of roughly equal size. On CM cellulose, there were three peaks. Amino acid analysis by the Moore and Stein method on an ion-exchange resin showed that peaks I and II were almost identical, and III, very different from them; antibody activity was present in peaks I and II. Porter's first interpretation of his results was that the molecule was a long single chain with three sections, supporting Linus Pauling's suggestion in 1940 of a long antibody molecule with a rigid centerpiece and two flexible ends, which folded, or molded, themselves on the antigen as a template to effect a specific match (Porter 1959). For Porter in 1959, a template theory of the generation of antibody diversity was still the most likely possibility.

In 1962 Porter tried a different dissection of the antibody globulin by opening its disulfide bonds. Separation of the result on a Sephadex column produced two peaks, one heavy and one light. He thought that these must be two chains of the globulin molecule, normally joined together by disulfide bonds. Amino acid analysis showed that there must be two of each chain. The relation of the chains to the papain pieces was determined immunologically, using goat antisera to the rabbit fragments. Porter then proposed a second model of the 7 S globulin molecule, with a pair of heavy chains joined by disulfide bonds and a light chain attached to each one, with the antibody site in fraction I, probably on the heavy chain. This model stood up satisfactorily to additional evidence, to the discovery of different types of heavy chain in different classes of immunoglobulin, and to the discovery of individual genetically determined polymorphisms of both chains (Porter 1973).

Using the fingerprinting technique, several laboratories in Britain and the United States, and Michael Sela and his group in Israel, found that the peptide maps of I and II from different antisera were very slightly different from each other. The chemists were getting nearer to finding out the amino acid sequence of an individual antibody; there was, however, still too much heterogeneity in normal antibody to make such detail possible.

The problem was solved by an accident of nature. To everyone's surprise, the fingerprints of normal light chains and the so-called Bence-Jones protein turned out to be almost identical. This material, named for its nineteenth-century discoverer, is found in the urine of patients with myeloma, a lymphoid tumor of the bone marrow. The only difference seemed to be that normal light chains were a mixture of two types, whereas the Bence-Jones chains were homogeneous, and therefore an ideal material available in large quantities for sequence studies on the amino acid analyzer. It appeared that each patient produced a different chain; each chain consisted of two sections, a variable region and a constant region. Of the 105 amino acids of the variable region, about 30 varied from case to case. In this variation lay the essence of antibody specificity.

Parallel with the detailed information that built up throughout the 1950s on amino acid sequences, evidence began to accumulate that these sequences were under genetic control. Information that specified each amino acid was found to be encoded in deoxynucleic acid, or DNA, transferred to a messenger ribonucleic acid, or RNA, and transcribed as an amino acid to be added to the growing peptide chain. The system appeared to be strictly directional. Francis Crick of Cambridge University stated what he called the "central dogma" of protein synthesis: Once genetic information has passed into protein, it cannot get out again. No protein could be formed by copying another protein (Crick 1957).

If this were so, it would make the template theory of antibody formation impossible. But the template theory had deep roots, and distinguished supporters, such as the U.S. immunochemists Pauling and Michael Heidelberger. Haurowitz, now at Indiana University, continued to hold that antigen could interfere, if not with the primary amino acid sequence of the protein, at least with its folding. In 1963 he was arguing that antibodies were normal globulins and that their amino acid sequences were shown by fingerprinting to be almost identical with one another, leaving only the folding to account for antibody specificity. By then, he knew that his theory was becoming less and less likely to be validated, and he often appeared to accept that its time had passed. But Haurowitz never laid down his arms. As he had said in 1960, he could not imagine the formation of a mold without a cast, or antibody without antigen.

Your Heart and Nutrition

Your Heart and Nutrition

Prevention is better than a cure. Learn how to cherish your heart by taking the necessary means to keep it pumping healthily and steadily through your life.

Get My Free Ebook

Post a comment