INACTIVATION OF VIRUSES 413 



(Henle and Zellat, 1941). Liquid 90 % propylene glycol applied for 24 

 hours at 4°C. caused a considerable reduction in infectivity of EMC virus 

 (Klarenbeek, 1954). 



Glycerol not only has no inactivating effect but actually seems to preserve 

 virus activity. This may possibly be explained by a replacement of hydration 

 water by glycerol, leading to a reduction in the rates of the various "spon- 

 taneous" inactivation processes, which all depend upon the presence of water. 



Ethyl ether is of particular interest on account of the relative specificity 

 of its effects (Andrewes and Horstmann, 1949). It seems to be completely 

 innocuous to lipid-free viruses. Of the lipid-containing viruses, some seem 

 to be ether-resistant, e.g., vaccinia (4 % Upids) and, surprisingly enough. 

 Western equine encephahtis (WEE) virus, reported to contain more than 

 50 % lipids (Beard, 1945). The group B arthropod-borne viruses are reported 

 to be rapidly and completely inactivated by 20 % ether. As shown by Hoyle 

 (1952), influenza A virus and, by Schafer and Zillig (1954; Schafer, 1957), 

 fowl plague virus are disrupted by ether treatment, with release of two com- 

 ponents with distinct biological and chemical properties: (a) nucleic acid- 

 free hemagglutinin (type specific antigen) and (b) nucleoprotein group 

 antigen. The latter is to all appearances identical with the "soluble antigen" 

 that can be extracted from infected cells. There remains little doubt that the 

 nucleoprotein represents the reproductive unit of the virus particle (Ada, 

 1957). So far, however, all attempts to demonstrate any infectivity of soluble 

 antigen, the nucleoprotein hberated from virus particles, or a mixture of the 

 disintegration products after ether treatment of the virus have failed. The 

 reasons for this failure are at present unknown. A priori, it does not seem 

 Hkely that exposure to ether would precipitate any significant structural 

 changes within the nucleoprotein unit resulting in inactivation. As a more 

 probable explanation it might be assumed that the nucleoprotein by itself 

 is completely nonavid and that its introduction into the ceU is entirely de- 

 pendent upon the receptor mechanism. The lipid would then mainly serv^e 

 the purpose of tying the two functionally different and equally essential 

 moieties together. In the case of WEE virus one would have to assume 

 either that the lipid is ether-insoluble or else that the virus hemagglutinin 

 plays a minor role in the mechanism of infection. 



Several nonpolar lipid solvents, in addition to the direct effect upon lipid- 

 containing viruses, have the capacity of denaturing proteins at the solvent- 

 water interface. As the virus protein seems to be very resistant to this type 

 of denaturation such solvents can be used for deproteinization and purifica- 

 tion of crude virus suspensions, with little or no damage to hpid-free viruses. 

 Methods analogous to Bachrach and Schwerdt's butanol technique were 

 appHed by Poison and Selzer (1954), using chloroform-amyl alcohol (Sevag- 

 treatment), and by Gessler et al. (1956), using fluorocarbons. 



