THE BIOCHEMISTRY OF PLANT VIRUSES 75 



reactive, and gives rise to — S — S — linkages, but in the tobacco mosaic 

 virus is prevented from doing this until denaturation or disaggregation has 

 taken place. 



The type virus contains 2 tryptophan residues per subunit, although, as 

 mentioned above, the related cucumber viruses 3 and 4 have only half a 

 residue per 17,000 molecular weight. There are also 2 lysine residues. One of 

 these is blocked so as to resist the action of fluorodinitrobenzene, even after 

 the virus has been disaggregated (Fraenkel-Conrat and Singer, 1954). 

 Presumably this may mean that the e-amino group is already substituted 

 by one of the acetyl groups detected by Miller and Stanley (1941). This 

 observation would also explain the fact that only one lysine peptide is 

 liberated by trypsin instead of the anticipated two (Sclu-amm and Anderer, 

 1955). The other lysine e-amino group is presumably available and is one 

 of the two groups substituted by carbobenzoxy, phenylureido, ;p-chloro- 

 benzoyl, and benzylsuKonyl groups, a substitution which does not involve 

 appreciable loss of virus activity (Miller and Stanley, 1942). This sub- 

 stitution does block the two free amino groups which react with nitrous 

 acid, and it increases the electrophoretic mobihty by about 10 % at neutral 

 pH values. It is surprising that these basic groups do not react readily with 

 acidic dyes, but this may be due to their steric arrangement. It is also a 

 matter of some interest to know what the second amino group is attached to, 

 because only one would be anticipated, the terminal amino group of the 

 polypeptide chain being blocked by an acetyl group. It is, of course, always 

 possible that the lack of reactivity of the second lysine e-amino group is 

 due to steric effects, but this would hardly explain the results of the tryptic 

 digestion. 



There are 4 tyrosine residues in the subunit. Proline accoimts for 6 residues, 

 thus allowing some 6 major bends in the poljrpeptide chains. One of these 

 residues is near the C-terminal end of the chains, while another is exposed 

 very readily by the action of 5 % trichloroacetic acid at 85°C. (Schramm and 

 Braunitzer, 1953). 



There are some 6 glycine residues, and 7 each of valine, phenylalanine, 

 and isoleucine. Next come 9 arginine residues, which are probably the most 

 important factors in estimating the actual chain size, because each gives 

 rise on tryptic digestion to a break in the chain, which can be detected by the 

 isolation of a peptide containing 1 arginine residue. Nine peptides of this 

 type are in fact found (Fraenkel-Conrat and Narita, 1957). (This is, of course, 

 easily demonstrated by a colorimetric spray for arginine). The proportion of 

 arginine is rather low for a nucleoproteia, but is more than adequate to neut- 

 rahze the 3 phosphate groups corresponding to each subunit. The remaining 

 6 guanidino groups of the arginines must be well buried in the protein, 

 because the virus as a whole has a predominantly acidic character. This also 



