78 R. MARKHAM 



the C-terminal end was isolated from enzymatic (cliymotrypsin) digests of 

 several strains of tobacco mosaic virus, and, in addition, a similar penta- 

 peptide containing prolyl-alanine as its C-terminal end was isolated from 

 virus which had been treated previously with carboxypeptidase. Thus there 

 is every reason to suppose that this hexapeptide does in fact represent the 

 C-terminal sequence of the virus. The sequence of this peptide has been 

 determined by finding the N-terminal amino acid, which is threonine, and 

 then finding the relative positions of two of the remaining amino acids by 

 partial acid hydrolysis. The whole sequence is threonyl, seryl, glycyl, prolyl, 

 alanyl, threonine (Niu and Fraenkel-Conrat, 1955a,b). The Holmes ribgrass 

 virus gives a similar peptide which has not yet been fully analyzed. Its 

 sequence is threonyl (threonyl, alanyl), prolyl, alanyl, threonine, the sequence 

 in parentheses being uncertain. 



The determination of the nature of the other end of the chain gave much 

 more trouble. As Miller and Stanley (1941) found, the only groups reacting 

 with ketene and similar reagents were the basic groupings on the surface of 

 the virus, one of which is certainly an e-amino group of lysine. Treatment 

 of the virus with reagents such as fluorodinitrobenzene had no success in 

 revealing any a-amino groups in the virus, and there was a considerable 

 tendency to regard the a-amino group as blocked by being linked in some 

 way to the rest of the chain so that it was not available. A possible way in 

 which this could occur is by the formation of an amide link to a carboxyl 

 group, of which there are plenty. 



For some time it was thought, because of experiments reported by the 

 Tubingen group, that this end was occupied by proline. In order to reveal 

 it the latter had to subject the protein to rather harsh treatment. Using 

 hot trichloroacetic acid for preHminary hydrolysis, they detected some 2300 

 proline residues as the dinitrophenyl derivative, and also as the thiohydantoin, 

 in which case some 2400 residues were estimated, taking into account the 

 considerable losses in this estimation. As it is now fairly certain that the 

 N-terminal end is acetylated and consists of iV-acetyl serine, it might appear 

 that the German workers stumbled across an unusually specific chemical 

 hydrolytic method in this work (Schramm et ah, 1955c). 



The problem of the missing N-terminal end of tobacco mosaic virus now 

 seems to be fairly well resolved. The Berkeley workers originally thought 

 that this end was probably linked back on to the main chain, forming a 

 ring at one end, and attempts were made to isolate the piece of chain con- 

 taining this cyclic structure. Enzymatic digestion was used to break down 

 the chain, and as it was considered useful to try to isolate the fragment by 

 virtue of its having one carboxyl group at one end, and no terminal amino 

 group, chjonotrypsin and pepsin were used for the digestion, in the hope 

 of producing a neutral peptide, rather than trypsin, which would 



