242 THE ANTIOEN.ANTIBODY REACTIONS 



THE QUALITATIVE ASPECTS OF THE ANTIGEN-ANTIBODY 

 REACTIONS^IMMUNOLOGICAL SPECIFICITY 



This problem has been attacked mainly from the side of the antigens, but it 

 will facilitate our discussion of the recorded data to deal first with the probable 

 structure of the serum antibodies with which they react. 



The Nature and Properties of the Serum Antibodies. 



We have already made frequent references to the view .that the serum anti- 

 bodies are specialized globulins. The properties of normal serum globuHns have 

 been established by a variety of methods. The fractionation of serum by pre- 

 cipitants like ammonium sulphate yields a mixture of proteins differing from one 

 another physically, and chemically in lipin and carbohydrate content (see Tiselius 

 1937a, Stenhagen 1938, Hewitt 1938). It is difl&cult to decide how far these fractions 

 are the results of the chemical handling the native protein receives in its preparation. 

 The measurement of the rate of migration of serum proteins in an electrical field 

 (Tiselius 1937a) provides convincing evidence that native serum contains several 

 well-defined components. After a given period of exposure in the Tiselius electro- 

 phoretic apparatus, the proteins of native serum separate into a number of zones, 

 each containing particles having approximately the same electrophoretic mobility. 

 In normal horse and rabbit sera, the albumin forms a single well-defined zone, 

 and there are three components of globulin, a, /? and y, the a being the most, and 

 the y the least mobile. 



The current hypotheses of protein structure are none of them sufficiently 

 developed to warrant description in a book of this kind. The student may consult 

 the reviews of Pauling and Niemann (1939), Svedberg (1939) and Synge (1943) 

 for details of the controversies regarding the precise architecture of the proteins. 

 It is sufficient for our purpose to note those features of protein chemistry that are 

 relevant to the understanding of serology in its present state. In the first place, 

 it is generally accepted that the proteins are built up of polypeptide chains of 

 amino-acids, and that the pattern of recurrences, whether regular or irregular, 

 of the various amino-acids in the chains will vary from protein to protein according 

 to the relative proportion of constituent amino-acids. A secondary structure is 

 imposed on the protein molecule by the folding of the chains so that several of 

 them are held together by hydrogen-bonding and other interatomic electrical 

 forces. In the so-called fibrous proteins, the arrangement of chains, as revealed 

 by X-ray analysis, is markedly regular (see Astbury 1943) ; keratin and silk fibroin 

 are examples of this type of protein. The globulins belong to the class of non- 

 fibrous proteins, and here it is supposed that polypeptide chains are primarily 

 folded into a lamina that has approximately the thickness of the polypeptide 

 chain, and that these laminae are built up into a more or less globular protein 

 molecule, being held together by interatomic forces similar to those that hold the 

 chains in laminae. Whether the laminar structure exists or not, it is a fact that 

 certain proteins can be spread on an air-water interface into films whose thickness 

 is much less than the known diameter of the protein molecule ; some proteins 

 are denatured by this process, but others, including serum proteins, will re-form 

 into " globular " molecules after spreading. 



That is to say, although protein molecules compared with smaller molecules 



