54 MAURICE J. BESSMAN 



tides as primers for (looxyiihoimclcotidc incorporation (Bolluin, 1960b, c). 

 PolyriboiUK'k'otidcs cannot serve as i)riniers, nor ean sonicated DNA 

 unless it is heated after sonication to melt out interchain hydrogen 

 bonds. 



Preliminaiy studies with a series of deoxythymidylate homopolymers 

 of general structure (pT)„ wliere n^S showed that single deoxyribo- 

 nucleotides (incubated as triphosphates) added to the ends of these 

 synthetic oligonucleotides. All of llic four triphosphates were effective 

 but dGMP and dTAIP tended to form multiple products. Since the 

 incubation mixtures contained relatively large excesses of primer to 

 deoxyribonucleoside triphosphate, the most likely event was the addition 

 of single deoxynucleotides to the ends of the chains. The added deoxy- 

 ribonucleotides were removed from the polymer by snake venom 

 phosphodiesterase suggesting that authentic phosphodiester linkages were 

 established between the oligonucleotide and the incorporated deoxyribo- 

 nucleotide. Further investigations using chemically synthesized poly- 

 thymidylates as primer (Bollum, 1962) have extended this earlier work 

 and have shown that oligonucleotides with greater than two nucleotide 

 residues are required. The 3'-hydroxyl group of the terminal dcoxynu- 

 cleoside residue must be free for the condensation since polymers con- 

 taining an acetylated 3'-hydroxyl group are inactive as primers. Since 

 single deoxyribonucleotides add to the ends of chains in this reaction, 

 the mechanism seems similar to the "limited" reaction catalyzed by 

 E. coli polymerases (Section IV,C). The difference is that the latter 

 enzyme requires polymerized DNA and the deoxynucleotides added seem 

 to conform to the base-pairing relationships of the primer. 



These observations are part of the general investigation undertaken 

 by Bollum into what is the minimum length of jjrimer required for 

 replicative or hydrogen bond directed synthesis where base-pairing 

 relationships are obligatory. Preliminary observations by Furlong and 

 Bollum (cited in Bollum, 1962) suggest that replicative synthesis begins 

 when chain length exceeds 20 monomer units. 



Bollum has suggested (personal communication) that the addition of 

 specific deoxyribonucleotides to the ends of chemically prepared oligo- 

 nucleotides might provifle a useful means of synthesizing specific 

 sequences for testing in other biological systems. 



VI. Synthesis of DNA in Bacteriophage-Infected Escherichia coli 



A. BACTERIOPH.\r,E-SPECIFTC POLYMERASE 



The dramatic appearance of several new enzymatic activities soon 

 after infection of E. coli with the T-even bacteriophages has recently 



