THE BASIS OF SPECIFICITY 253 



1922, Landsteiner and Heidelberger 1923, Hektoen and Schulhof 1923), and 

 between different vegetable proteins (Wells and Osborne 1911, Wells 1915, Jones 

 and Gersdorff 1923, Lewis and Wells 1925, Wells et al. 1927). 



As the classical researches of Nuttall (1904) indicated in the first place, this 

 degree of specificity may be closely correlated with biological classifications, so 

 that we might expect cross-reactions between the serum proteins of man and 

 monkey, but none between those of man and ox. But, though this holds to a 

 large extent, later studies have revealed relationships between the antigens of 

 organisms as widely separated as man and Shiga's dysentery bacillus (see Chapter 8). 

 A serological relationship, therefore, between two organisms can strengthen a 

 relationship already established on other biological grounds, but it cannot by 

 itself be considered as good evidence of anything more than a similarity of perhaps 

 a small part of the metabolism of the two organisms. 



Analysis in many cases reveals chemical differences between native antigens, 

 but the complexity of the antigenic particles concerned is such that analyses are 

 necessarily crude. Far more significant information has been gained by altering 

 the chemical of protein antigens along certain limited and well-defined lines, and 

 noting the resulting changes in their immunological reactions. 



Obermayer and Pick (1906) (see also Pick 1912) showed that the nitration or halogena- 

 tion of proteins — that is, the introduction of the nitro-group or of a halogen element such as 

 iodine — profoundly altered the antigenic reactions of the treated j^rotein. Serum proteins 

 so treated lost their species specificity, but they gained a new specificity, shared by 

 normally unrelated serum proteins that had been chemically altered by the same procedure. 

 Thus an antiserum prepared against the nitrated serum of a particular animal species 

 failed to react with the unaltered serum of that species, but reacted with a wide range 

 of nitrated sera from other animals. Since it was known that the nitro-group and the 

 halogen elements entered into the benzene ring of certain of the amino-acids which build 

 up the complex protein molecule, Obermayer and Pick were led to attach particular import- 

 ance to these chemical groupings as factors determining immunological specificity (see also 

 Wormall 1930, Johnson and Wormall 1932, Snapper and Grunbaum 193(3, Shahrokh 1943). 



Landsteiner and his colleagues, however, have shown that the salt-forming groups of 

 the amino-acids (the carboxyl-, hydroxyl- and ammo-groupings) play an equaUy important 

 part. By esterification, methylation and acetylation, they have succeeded in altering the 

 immunological specificity of proteins (see Landsteiner and Prasek 1914, Landsteiner and 

 Lampl 1917a, Landsteiner 1917). 



This relatively simple modification of active groups on. the surface of the antigen 

 was largely extended by Landsteiner and his colleagues (see Landsteiner 1930, 

 1933, 1936), who utilized the diazo reaction to introduce larger modifying groups 

 into the protein molecule. The substances for introduction contain an — NHg 

 group, usually attached to a ring structure in the molecule. The — NHj group is 

 converted to — N : N-Cl in presence of NaNOg and HCl, and the resulting diazo 

 compound mixed with the protein. The diazo compound reacts with the phenolic 

 groups of tyrosine (see Fig. 41), with the imidazole ring of histidine, and probably 

 with the indole group of tryptophan, the — NH group of proline and hydroxy- 

 proline and all the aliphatic — NHj groups on the protein surface (Eagle and Vickers 

 1936). By varying the conditions of the reaction it is possible to control the 

 amount of modifying substance introduced into the protein. Fig. 41 illustrates 

 the coupling of atoxyl (j9-amino benzene arsenic acid) to a protein, on the assump- 

 tion that it unites with the phenolic group of the tyrosine residues in the protein 

 molecule. 



