I. EEPLICATION OF DNA IN CELL-FREE SYSTEMS 25 



(see Section IV,r) favor the latter interpretation since it can be shown 

 that each of the four deoxyribonucleotides attaches to chains ending in 

 each of the bases. This analysis also rules out the possibility that the 

 incorporation of single deoxyribonucleotides is due to the foraiation of 

 homopolymers of the individual deoxyribonucleotides (poly-dCMP, for 

 example), and demonstrates that the deoxyribonucleotide is attached to 

 the DNA in covalent linkage through its i)hosphorus atom and not by 

 some nonspecific adsorption or inclusion, etc. 



What is the significance of the limited reaction? One possibility is 

 that this is the prototype of the reaction that occurs when all four 

 deoxyribonucleotides are present. If DNA replicates by adding deoxy- 

 ribonucleotides to its end, these new units must loop back in some 

 manner to allow the directing influence of the DNA chain to be exerted. 

 But this imposes an additional requirement, that is some means of break- 

 ing this loop when the DNA has been duplicated. We know that in reac- 

 tions in which there has been a 20-fold synthesis over the DNA originally 

 present as primer, the molecular weight of the product is not increased 

 20-fold. 



Another possibility is that the chains of the double helix of primer 

 DNA as isolated from natural sources are of unequal length (possibly 

 through damage during preparation) and the limited reaction is the 

 repair of these broken ends. This could mean that the sequence of deoxy- 

 ribonucleotides added to the shorter ends w^ould be dictated by the longer 

 end. In keeping with this notion is the recent observation (A. Kornberg, 

 personal communication) that in the limited reaction using the d-AT 

 polymer as primer (see Section IV,G) dATP or dTTP is incorporated, 

 whereas dGTP and dCTP are not. This would indicate that there is 

 some specificity imposed by the primer even for the limited additions to 

 the ends of the chains. This "repair" mechanism would not be pertinent 

 to a scheme for the replication of DNA which does not involve a covalent 

 linkage at any stage between primer and product. 



A third possibility is that the enzyme which carries out the limited 

 reaction is different from E. coli polymerase. This possibility seems 

 remote. 



D. NATURE OF THE PRIMER 



So far, using highly purified fractions of E. coli polymerase, it has 

 not been possible to synthesize a product containing all four deoxyribo- 

 nucleotides in the absence of a primer. The priming DNA can be ob- 

 tained from many sources, e.g., E. coli, calf thymus, salmon spenii, 

 bacteriophage, crab testes, wheat germ, etc., but at present no chemically 

 synthesized product has been able to substitute for the natural polymer. 



