144 C. A. KNIGHT 



suggested that the individual particles of TMV resemble crystals and that they 

 are composed of subunits. Subsequent studies by the X-ray technique have 

 supported this idea (Frankhn, 1957). Further suggestion that TMV particles 

 might possess a substructure was provided by the observations of Schramm 

 and colleagues (Schramm, 1954) that TMV seemed to degrade in a regular, 

 rather than a random manner, and that certain degradation products could 

 be repolymerized to give back virus-like particles. Also, Takahashi and Ishii 

 (1952, 1953) discovered a low molecular weight protein in mosaic infected 

 tobacco (so-called X-protein) which could be polymerized to give virus-like 

 rods. Compelling chemical evidence for the presence of subunits in TMV 

 structure was provided when it was found (Harris and Knight, 1952, 1955) 

 that carboxypeptidase caused the release of about 2900 threonine residues 

 from each mole of TMV (assuming a molecular weight of 50 X 10^ for TMV). 

 These carboxyterminal (C-terminal) threonine groups are considered to 

 represent an equal number of peptide chains in the virus structure. Hence, 

 the virus is composed of subunits, each with a molecular weight of about 

 17,000, or, subtracting the nucleic acid, a protein subunit of about 16,000. 



The thirteen different strains of TMV illustrated in Fig. 2 were also treated 

 with carboxypeptidase, and it was found that they aU had C-terminal 

 threonine, and in approximately the same amount (Knight, 1955b). Hence, 

 it can be concluded that these strains of TMV probably contain the same 

 numbers of subunits in their structures and that the peptide chains terminate 

 in the same way. Since unrelated viruses gave results which were qualita- 

 tively and quantitatively different, the identity of C-terminal groups seems 

 to be a chemical criterion of strain relationship. 



Further information concerning the C-terminal end of the peptide chains 

 of the strains was obtained by a study of the peptides released from strain 

 proteins by chymotrypsin (Niu and Fraenkel-Conrat, 1955). C-terminal hexa- 

 peptides were obtained from the strains TMV, M, YA, and HR, and these 

 proved to have identical amino acid sequences in the cases of TMV, M, 

 and YA, while the hexapeptide from HR had the same sequence for the 

 last three residues, but differed in the other three. 



There appears to be no free amino terminal (iV-terminal) end of the peptide 

 chains of TMV (Fraenkel-Conrat and Singer, 1954). Recently, Narita (1958) 

 isolated an iV-acetylated peptide containing serine and tyrosine from enzy- 

 matic digests of TMV protein and this is thought to represent the iV-terminal 

 structure of the virus protein. This same peptide has also been obtained from 

 the strains M, YA, HR, and J14D1 and hence is probably a characteristic 

 feature of strains of TMV (Narita, 1958b). 



The following sketch appears now to represent the terminal features of 

 TMV strains: 



iV-acetylseryltyrosyl . . . prolylalanylthreonine 



