SEROLOGICAL VARIATION 253 



tissue ill which virus is propagated undoubtedly affects the fullness of anti- 

 genic expression and contributes tissue components which may be antigenic 

 in another species. If virus is inactivated before use in serological tests, the 

 antigenic character may be altered. 



Antibody response in the animal undergoing active infection can be 

 expected to reflect the full array of viral components, especially if extensive 

 tissue involvement occurs. Activities related to adaptation and virulence are 

 not well recognized serologically as yet. Parenteral injection of nonmultiply- 

 ing virus in animals may presumably anticipate a narrower serum response, 

 but hyperimmunization consistently increases the breadth of reaction with 

 secondarily related strains. Susceptible and nonsusceptible species may differ 

 in this respect. The time of collection of serum from the test animal has been 

 consistently shown to affect its reactive character; antibodies measured by 

 complement fixation (CF) or hemagglutination inhibition (HI) may appear in 

 workable amounts at different times from neutralizing, agglutinating, or 

 precipating antibody. This is particularly true after infection. The antibodies 

 may not have the same immunological significance if they are reacting with 

 constituents of different functional and biochemical character. Thus, the 

 technical procedures must be conducted with awareness of the nonspecific as 

 well as specific reactants of serum, the character of the virus preparation, 

 and the quantitative considerations which permit valid interpretations. 



Since influenza viruses have been most extensively studied and illustrate 

 the numerous features involved in the interpretation of serological variation 

 and its significance, the ensuing discussion will be largely limited to that 

 group of viral agents. 



III. Serological Variation Among Influenza Viruses 

 A. Viral Structure 



Serological behavior of influenza viruses has led from the early periods of 

 antigenic study to the suggestion that characteristic antigenic components of 

 a strain were located at or near the surface of the virus particle. Examination 

 of the chemical structure of influenza A virus has been particularly pursued 

 by Hoyle and his co-workers (Hoyle, 1952; Hoyle and Frisch-Niggemeyer, 

 1955; Frisch-Niggemeyer and Hoyle, 1956). The virus particle is considered 

 to consist of a central core of type-specific nucleoprotein (RNA), or "soluble" 

 antigen, surrounded by strain-specific mucoprotein (hemagglutinating, 

 immunizing antigen) bound together by lipo- and mucoprotein, presumably 

 derived from the cell wall. The serological behavior of this matrix is essentially 

 unstudied, but one may speculate that it plays a role in certain of the phase 

 variations by covering active surface antigens, and may also be involved in 

 nonspecific reactions. 



