THE BIOCHEMISTRY OF PLANT VIRUSES 77 



proline, and alanine, of which he finds relatively less, and of isoleucine, 

 phenylalanine, leucine, serine, arginine, threonine, and glutamic acid, of 

 which he finds somewhat more. He has expressed his results in the form of 

 integral numbers of residues per subunit, and deduces the existence of a 

 subunit having a molecular weight of 18,420 having 164 residues, these 

 being: 19 (16) aspartic acids; 17 (15) threonines; 18 (16) serines; 8 (7) prolmes; 

 17 (14) glutamic acids; 6 (5) glycines; 15 (13) alanines; 14(12) valines; 9 (8) 

 isoleucines; 13 (11) leucines; 4 (3) tyrosines; 8 (7) phenylalanines; 2 (2) 

 lysines; 11 (9) argmines; 2 (2) tryptophans; and 1 (1) cysteine; plus 20 amide 

 groups. The figures in parentheses are the result of altering the data of 

 Ramachandran in such a way as to make arginine have the more probable 

 value of 9 residues, and then fitting the rest of the data to the nearest whole 

 numbers of residues. It will be noted that the size of the subunit is then 

 about the same as that deduced from the data of Newmark and Fraser (141 

 amino acid residues compared with 145), but that the composition is slightly 

 different. 



It should be remarked here that if expressed on a weight basis, the quantity 

 of arginine found by Ramachandran (1958), Newmark and Fraser (1956), 

 and also by Black and Knight (1953) is essentially the same, so that if the 

 data obtained by tryptic digestion (in the same laboratory, and confirmed 

 elsewhere) are to be believed, there must be a systematic deviation in the 

 results, which makes the subunit appear to be slightly too large. Alterna- 

 tively, of course, there may be some structure of an unknown type which 

 may prevent 2 of the arginine peptides from being released by trypsin 

 digestion. This could be confirmed by makmg semiquantitative analyses 

 on the 9 arginine-containing peptides. The virtue of counting the actual 

 numbers of residues in each subunit, rather than attempting to deduce it 

 from quantitative estimates, should be evident, and any estimate of the 

 size of the protein subunit based upon such evidence should be very reliable. 



With regard to the possible snags in such a procedure, it has been noted 

 above that one of the two lysine groups seems to be blocked, both with 

 regard to trypsin action, and also to chemical substitution of the e-amino 

 group, and a reason for this blocking has been suggested. 



A number of peptides have also been obtained by the action of pepsin 

 and of chymotrypsin, and they are being investigated. One such peptide has 

 been obtained by the action of both of these enzymes separately, and it is 

 thought to represent the blocked N-terminal sequence of the protein, which 

 has been referred to earlier (Narita, 1958b). 



E. The Chain Ends 



The carboxyhc end of the polypeptide chain has already been mentioned, 

 and is -j)rolyl-alanyl-tlireonine. A hexapeptide containing this structure as 



