THE BIOCHEMISTRY OF PLANT VIRUSES 61 



on a weight basis, the difference being as much as 30 % for an ordinary 

 polymerized ribonucleic acid. The precise physical meaning of this difference 

 is uncertain, but it is evidently due to intermolecular structure of some kind, 

 such as hydrogen bonding between the absorbing rings. This attribute of 

 nucleic acids makes it very difficult to interpret absorption spectra of 

 substances such as viruses, because the contribution of the nucleic acid to 

 the spectrum is an unknown factor. Based on the moles of nucleotide residues 

 present the molar extuiction coefficients of the free nucleic acids in the region 

 of 260 nifjL are of the order of 7,000 to 11,000 and vary with temperature, 

 but on hydrolysis this rises to some 12,000, the value depending upon 

 the nucleotide composition of the nucleic acid and the pH of the 

 solution. 



Being polymers of phosphoric acid esters, which have one primary phos- 

 phate OH- per nucleotide residue, and having numbers of NH2+ groups 

 and enolic OH^ groups, the nucleic acids are in effect rather like linear ion 

 exchange resins, and have a great affinity for proteins. In fact, the com- 

 bination of the nucleic acid and the protein of viruses can be explained 

 simply on the assumption of electrovalent linkages existmg between the 

 two, although there is evidence that the phosphate groupings are not all 

 involved in such linkages and are to some extent neutralized by basic 

 compomids. The strong affinity of the nucleic acids for proteins is a cause of 

 considerable trouble in their study, because the nucleases, especially the 

 ribonucleases, are extremely stable proteins and are highly basic as well. 

 Consequently when nucleic acids are prepared, they adsorb any traces of 

 nuclease in the solution, and then are slowly digested. It is because of this that 

 while it is relatively easy to make highly polymerized virus ribonucleic acid 

 (Cohen and Stanley, 1942; Northrop and Sinsheimer, 1954), it is extremely 

 difficult to keep it from depolymerizing. The actual estimates for the molecular 

 weights are somewhat dubious. Hopkins and Sinsheimer (1955), using light 

 scattering, estimated the molecular weight of the tobacco mosiac virus to 

 be about 1.7 million, which is rather smaller than the anticipated value of 

 about 2.5 million, if the virus nucleic acid formed one single polynucleotide 

 chain. 



Because of their large size and their charge, which tends to make the 

 molecides elongate in water or dilute salt solutions, the nucleic acids form 

 highly viscous solutions or jellies, and show birefringence of flow. Fibers may 

 be dra\vn both from deoxyribonucleic acids and from ribonucleic acids. 

 The latter are, however, not suitable for detailed X-ray examination, so that 

 little is known about their finer structural details. The synthetic polymers of 

 ribonucleotides have, of course, been examined in this way (Rich, 1957), 

 but it is not altogether certain whether this information is relevant to virus 

 structure. 



