ANTIGENS 35 



mum in size, and (b) they must be foreign to the circulation of the 

 animal in which they stimulate antibody production. 



It is not possible to give a definite figure for the minimal molecular 

 weight which a substance must possess to be an antigen, but good 

 antigens generally have a molecular weight of not less than 10,000. 

 It is also found that by adsorbing small molecules particulate material 

 may become antigenic, just as conjugated antigens may be produced 

 by coupling simpler compounds (haptens) to proteins. Some simple 

 substances not conjugated to a protein cause the production of anti- 

 bodies, but it is believed that they act by first combining with some 

 of the proteins of the body. 



Size alone does not seem to be enough. In addition to being 

 large, a molecule, to be antigenic, must possess other characteristics. 

 It has been suggested that a certain degree of internal complexity 

 may be required, and it has been found that sulfonated polystyrene 

 (Fig. 3-1), which is a large molecule polymer made up of a single 



CH2-CH2-(-CH-CH2).-C = CH., 



SO3H SO3H SO3H 



Fig. 3-1. Sulfonated polystyrene. 



repeated unit, is not antigenic (Boyd, 1952). Haurowitz (1952) sug- 

 gested that the necessary feature is a rigid structure of the determi- 

 nant groups of the antigen. In support of this idea, it has found 

 that gelatin, a non-rigid molecule, which is ordinarily a very poor 

 antigen, can be made into a relatively good antigen by being coupled 

 with chemical groupings which would be expected to increase the 

 rigidity of the molecule (Hooker and Boyd, 1932; Glutton, Haring- 

 ton, and Yuill, 1938; Sela and Arnon, 1960). Contrary to earlier 

 opinion, the introduced groups do not have to be aromatic (Sela 

 and Arnon, 1960). Haurowitz's suggestion is also supported by 

 studies on the antigenicity of synthetic polypeptides. Polyglutamic 

 acid was found to be non-antigenic (Maurer 1957), and most of the 

 polymers studied by Stahmann and his colleagues (references in 

 Sela, 1962) were either non-antigenic or poor antigens. On the 



