54 AN INTRODUCTION TO THE STUDY OF VIRUSES 
Turnip Yellow Mosaic Virus 
The methods for isolating and purifying this virus have been 
described in the preceding chapter and the purified virus crystallizes 
in the form of small octahedra (Plate VIII, lower photograph). This 
virus is an unusual one in some respects and chemical studies upon it 
are therefore of great interest. Virus preparations made by the alcohol 
and ammonium sulphate precipitation methods described in Chapter V 
are highly infectious and are quite uniform in composition. They 
contain about 15 per cent nitrogen, 2°15 per cent phosphorus, and 
5 per cent of purine-bound pentrose, the latter two being entirely in 
the form of nucleic acid. The isoelectric point of the virus in solution 
is pH 3-75. The virus is unstable to weak alcohol and the solution is 
denatured by 334 per cent alcohol. The yield of virus from infected 
plants varies but may be as much as 1 g per litre. The infectivity is 
10o~® g¢ per millilitre. 
The most interesting point about this virus is the existence in the 
infected plant of two components which differ in their sedimentation 
constant. This means, therefore, that when a solution containing 
more than 10 mg per millilitre of virus protein is spun on a Sorvall 
S.S.1 centrifuge it separates off into two layers, a “top” and a “bottom” 
component. The boundary between these two layers is sufficiently 
stable to allow the top component to be removed by a pipette. The 
top and bottom components are similar, if not identical, in the following 
respects. Both crystallize from salt solutions in octahedra (Plate IX) 
and under the electron microscope are spheres of similar size. In 
solution, too, they seem to be the same size, and both are resistant to . 
digestion by pancreatic enzymes. 
The bottom component, however, is the only one containing 
nucleic acid, and the absorption spectra of the two are consequently 
very different. Although complete amino acid analyses are not yet 
available, preliminary paper chromatography using several solvents 
indicates that both proteins contain the same amino acids in similar 
proportions. All the evidence suggests that the protein parts of 
both components are identical and that the bottom component 
differs only in having some 28 per cent of pentose nucleic acid by 
weight embedded in it in such a way that it takes no part in 
forming the electrostatic surface of the molecules. The comparatively 
acid isoelectric point of both proteins seems to be due to the 
relative preponderance of the dicarboxyl’c amino acids in the 
protein. 
