Replication of DNA in Vitro 



285 



poxanthine has the same H-bonding groups 

 as thymine, it does not replace thymine, 

 probably because the A-hypoxanthine pair, 

 being composed of two purines, takes up too 

 much space to fit the regular double-helix 

 configuration. These results support the 

 hypothesis that normally the /'// vitro syn- 

 thesis of DNA is dependent upon the forma- 

 tion of complementary purine-pyrimidine 

 pairs — A with T and C with G — just as is 

 the case in vivo. 



The DNA synthesized in vitro from the 

 usual four deoxyriboside 5'-triphosphates 

 can be analyzed chemically. The analysis 

 shows that in the in v/fro-synthesized DNA, 

 A equals T and C equals G, just as they do 

 in natural DNA, even though the relative 

 concentration of the four triphosphates in 

 the substrate is widely distorted. Not only 

 do total pyrimidines equal total purines in 

 "synthetic'" DNA, as described — whether a 

 moderate or a large amount is synthesized — 

 A + T 



but the particular 



ratio of the primer- 



G + C 



template is reproduced faithfully in the syn- 

 thesized product (see Figure 8 in Supple- 

 ment V). In this respect, the product is a 

 replica of the primer-template, as is strik- 

 ingly illustrated by the following experiment. 

 After a long period of incubation of all the 

 usual components except pre-existing DNA, 

 a linear deoxyribotide polymer composed 

 only of A and T is formed spontaneously. 

 When this polymer is added as the pre-exist- 

 ing DNA, even though all four usual triphos- 

 phates are present in the substrate, the result 

 is an extensive synthesis of material which 

 contains A and T only, with no trace of C 

 and G (again see Figure 8 in Supplement V). 

 It has been mentioned that for each nu- 

 cleotide added to the end of the DNA strand, 

 an inorganic pyrophosphate, PP. is liberated. 

 When PP is added to the usual synthesizing 

 complex in great excess (about one hundred 

 times the concentration of the triphos- 

 phates), the synthetic reaction is inhibited 



by about 50%. This observation implies 

 the reversibility of DNA synthesis in vitro. 

 It was also mentioned earlier that in a 

 limited reaction, </CP*(PP) can add onto 

 a strand terminating in all four types of 

 nucleotides (dCP, TP, d AP, and </GP). 

 However, this statement does not suggest 

 that dCP* joins linearly to each nucleotide 

 with equal frequency, or that any nucleotide 

 joins to all others with equal frequency. 

 Other results indicate that the limited re- 

 action does not add nucleotides to the end 

 at random; this reaction probably involves 

 the repair of the shorter strand of a double 

 helix, the particular nucleotide added being 

 specified in the usual way by the bases pres- 

 ent in the longer strand. In other words, 

 the shorter strand acts as a primer and the 

 longer strand as a template. 



Dinucleotide Sequences 



What is the linear arrangement of nucleo- 

 tides in the DNA synthesized in an extended 

 reaction? If it is genetic material, different 

 linear segments of DNA can represent dif- 

 ferent genes, and the differences among 

 genes would lie in the sequence of their 

 organic bases. Considering only the four 

 usual deoxyribotides, how many different 

 sequences of two nucleotides are possible? 

 The first nucleotide can be one of four, and 

 so can the second, making a possible 4 times 

 4. or 16 different linear arrangements in di- 

 nucleotides. The orders in dinucleotides 

 can be determined experimentally as follows: 

 one of the four triphosphates added as sub- 

 strate is labeled with P 32 in the innermost 

 phosphate, the other three are not. Ex- 

 tended synthesis is permitted during which 

 the P* attaches to the 3' of the sugar of the 

 nucleotide which is its linear neighbor (refer 

 to the left part of Figure 21-2 and to Figure 

 9 in Supplement V). This linear neighbor 

 can be identified by digesting the synthesized 

 product with micrococcal DNase and splenic 

 phosphodiesterase. Recall that the latter 



