X. TMV STUDIES IN GENETIC CODING 497 



the common TMV (Sugiyama and P'raenkcl-Conrat, 1962), even though 

 the proteins of two of these show differences in the N-terminal region 

 of the peptide chain. Further information has been obtained by Rushizky 

 and Knight (1960) through comparative studies of enzymatic digests of 

 various strain nucleic acids. Pancreatic ribonuclease, which splits only 

 next to pyrimidines, has been found useful for this purpose. By the 

 2-dimensional mapping procedure (electrophoresis and chromatography) 

 many of the smaller digestion products can be separated and quantitated 

 (cytidylic and uridylic acid, as well as di-, tri-, and tetranucleotides 

 composed of purines and terminal 5'-linked pyrimidines). These studies 

 have revealed definite differences in the frequency of recurrence of 

 some oligonucleotide sequences in natural strains not closely related to 

 TMV. 



VI. Chemically Produced Strains 



A. MODIFICATION OF THE RNA 



Ever since the early days of their isolation the viruses have been 

 subjected to various reagents or conditions which modified their struc- 

 ture (Stanley, 1936). Only more recently has it become evident that most 

 of those agents which inactivated vimses did so by force of their affinity 

 for, and reaction with, the viral RNA. The three main classes of modify- 

 ing agents are (1) physical agents, such as ultraviolet light. X-rays, heat, 

 and sonic oscillations; (2) the biological approach of allowing the virus 

 to replicate in the presence of base analogs which then tend to become 

 incorporated into the virus structure; and (3) a host of chemical reagents 

 which affect the amino groups or substitute the rings of the purines or 

 pyrimidines. 



This subject has been critically reviewed by Schuster (1960), and 

 will not be systematically dealt with here. The modification reactions 

 usually lead to inactivation of the viiTises and thus are of some interest 

 as antiviral agents, and for the purposes of vaccine preparation. In 

 connection with the present chapter, the main interest in these agents 

 resides in their ability to produce mutations and they will be discussed 

 primarily from this viewpoint. 



The effects of the physical agents are varied, ranging from chain 

 breakage to modification of the pyrimidine rings, but none of these reac- 

 tions appears to lead to the frequent production of mutants. The same 

 is true for the incorporation of base analogs. Azaguanine appears to 

 replace some guanine residues (Matthews and Smith, 1955) and 2- 

 thiouracil (.Teener and Rosseels, 1953; ]\Iandel et al., 1957) as well as 

 5-fluorouraeil (Gordon and Staehelin, 1958) replace uracil, the latter uji 



