INFECTIVITY OF TOBACCO MOSAIC VIRUS 447 



such imdegraded nucleic acid particle-complements, possibly in the form of 

 a single polynucleotide chain composed of about 7000 mononucleotides. 



In contrast to this concept, some sedimentation analyses performed at 

 Berkeley with the SDS-type of infectious nucleic acid preparation had given 

 much lower S values, comparable to those obtained by Cohen and Stanley in 

 1942, and interpreted m terms of a molecular weight of 0.2-0.3 x 10^. 

 Further unpublished studies performed by Sue Hanlon, Pearl Appel, and 

 H. K. Schachman showed that the sedimentation of TMV-KNA was to a 

 remarkable extent dependent upon the salt concentration of the medium. 

 Thus, low sedimentation constants (S20 = 5-10) were obtained only at very 

 low ionic strengths and were then very concentration-dependent, extrapolat- 

 ing at zero concentration to values similar to those obtained in the more 

 concentrated ionic media (S20 = 20-30). There appeared to be no basis for 

 the earlier behef that there was a difference in the two types of nucleic acid 

 preparations, since the sedimentability of both was found to depend in a 

 similarly crucial manner upon the salt concentration. 



One mterprctation of the observed effects was that it represented an 

 indication of the formation of higher molecular aggregates under the influence 

 of the salt. It thus appeared possible that the various tecliniques listed above 

 had measured the aggregate weight of a number of subunit pol}Tiucleotide 

 chains, possibly of about 0.25 X 10^ molecular weight, and existing in free 

 state only at low concentrations m very low ionic strength media. However, 

 an alternate interpretation of the observed sedimentation behavior is that 

 sedimentation at very low ionic strength is abnormal, owing to the primary 

 charge effect (Svedberg and Pedersen, 1940), and that it therefore cannot be 

 regarded as evidence for the formation of aggregates at the higher ionic 

 strength. A similar though somewhat less marked effect of ionic strength 

 was noted also for the sedimentation behaviour of DNA. 



Thus, while the weight of evidence may favor the concept that the infec- 

 tious component of nucleic acid preparations consists of intact nucleic acid 

 units of 2-3 X 10^ molecular weight, several independent experimental 

 observations are definitely not in accord with this concept, but seem to 

 point to the existence of aggregating subunits. One of these is the hypo- 

 cliromic effect observed with TMV-RNA in the presence of salts. When salt 

 (0. 1 M) is added to a mixture of the synthetic polynucleotides, polyadenylic 

 and polyuridylic acid, the absorbance (A) decreases by as much as 30 %, an 

 effect which is interpreted as an expression of the base-pairmg aggregation 

 of two or more polymer strands (Warner, 1957; Felsenfeld and Rich, 1957). 

 Quite a similar effect is noted with infectious TMV-RNA (17-20 % depression 

 at 0.01 M or higher salt concentrations, be it NaCl, KCl, phosphate, etc., or 

 at 10-" M MgCla), but to a smaller extent with degraded TMV-RNA, and 

 yet less with yeast RNA. When it is taken into consideration that these same 

 VOL. I — 30 



