VARIATION AND ITS CHEMICAL CORRELATES 151 



A. Chemical Derivatives of TMV 



The two major components of plant viruses, protein and ribonucleic acid, 

 both contain reactive groups which can be substituted, altered, or otherwise 

 modified by treatment with specific chemical reagents. It was of consider- 

 able interest, first, whether or not significant changes could be made without 

 completely inactivating the virus, and second, whether or not the modifica- 

 tions would be inheritable. 



TMV has been treated with such reagents as formaldehyde (Ross and 

 Stanley, 1938; Fraenkel-Conrat, 1954; Cartwright et al., 1956), iodine 

 (Anson and Stanley, 1941; Fraenkel-Conrat, 1955), ketene, phenyl isocyanate, 

 carbobenzoxy chloride, jo-chlorobenzoyl chloride, benzenesulfonyl chloride 

 (Miller and Stanley, 1941, 1942), fluorodinitrobenzene (Knight, 1951), 

 iV-carboxyleucine anhydride (Fraenkel-Conrat, 1953), and benzyl and w-butyl 

 mustards (Wood et al., 1948). 



It was possible to get modified but infective virus in all of these cases, but 

 the disease caused by the chemical derivatives was indistinguishable from 

 that of untreated virus and the virus produced was also the same as the un- 

 treated; that is, none of the changes was inheritable. 



Another type of chemical modification is that caused by the enzymatic 

 removal of the C-terminal threonine residues from the peptide chains of 

 TMV and of 12 of its strains (Harris and Knight, 1955; Knight, 1955a). In 

 the case of TMV, which was most extensively studied, this kind of reaction 

 caused demonstrable changes in electrophoretic mobility and serological 

 reactivity, but resulted neither in measurable change in infectivity nor in 

 symptoms caused in several hosts. Moreover, the progeny from "dethreonin- 

 ated" virus were found to have the C-terminal characteristics of imtreated 

 virus. In the 12 strains of TMV tested, no alteration of the symptoms in 

 Turkish tobacco was caused by removal of the C-terminal amino acid 

 residues. 



With increasing e\adence for association of genetic function ^vith nucleic 

 acid (see Chapter VI, Vol. I) it can be reasoned that the above chemical and 

 enzymatic reactions failed to produce mutants because they involved pri- 

 marily the protein components. There is, of course, evidence that the for- 

 maldehyde and mustards reacted with the nucleic acid of TMV as well as 

 with the protein, but they may not have caused the type of change necessary 

 for a mutation. Hence, considerable interest is focused on attempts to cause 

 more deep-seated changes in the nucleic acid. In this connection, it has been 

 possible to introduce substituted purines and pyrimidines iti vivo into TMV 

 and TYMV nucleic acids (Matthews and Smith, 1955; Gordon and Staehelin, 

 1958), but there is no indication that mutants were obtained thereby, unless 

 the lesser infectivities of the modified viruses are interpreted as evidence for 



