Replication of DNA in Vitro 



289 



For example, trypsin breaks peptide bonds 

 only at places in a polypeptide chain where 

 lysine or arginine arc present. DNA poly- 

 merases are unique in that they take direc- 

 tions from a long template; the strand act- 

 ing as template dictates which particular 

 monomers can be added by DNA polymerase 

 to the strand acting as primer. It should be 

 noted that, in vitro, ribonucleotides can be 

 incorporated into terminal positions in 

 DNA ' and that, using DNA as a primer- 

 template, E. coli DNA polymerase can use 

 a mixture of riboside and deoxyriboside 

 triphosphates to synthesize complementary 

 strands that contain both ribo- and deoxy- 

 ribotides.' ; No evidence exists, however, 

 that such mixed RNA-DNA strands have 

 biological or genetic significance. 



DNA Synthesis from RNA 



Using poly {A + U), the double helix com- 

 posed of the homopolymer of adenylic acid 

 base-paired with the homopolymer of uri- 

 dylic acid, E. coli DNA polymerase activated 

 by MgCL can be tested ' for ability to utilize 

 various substrates in an in vitro synthesis. 

 Using dAPPP and TPPP in the substrate, 

 base-paired homopolymers of deoxyadenylic 

 acid and thymidylic acid (designated as poly 



u See J. S. Krakow, H. O. Kammen, and E. S. 



Canellakis (1961). 



,; See P. Berg, H. Fancher, and M. Chamberlin 



(1963). 



7 See S. Lee-Huang and L. F. Cavalieri (1963), 



and L. F. Cavalieri (1963). 



[dA -(- T]) are synthesized without an ap- 

 preciable lag period. Since each DNA 

 strand of the product is homopolymeric 

 (proved by nearest-neighbor analysis), it is 

 clear that poly (A -f- U) is the template- 

 primer. No evidence is found for the for- 

 mation of a complex of poly A with poly T 

 or of poly U with poly dA. Moreover, 

 neither single-stranded poly A, nor single- 

 stranded poly U, nor triple-stranded poly- 

 ribotides can act as primer-template for 

 DNA synthesis; using poly (A -f- U) as 

 primer-template, no synthesis occurs with 

 either JAPPP or TPPP alone. Finally, even 

 after a one-fold synthesis, most — if not all — 

 of the original poly (A -f- U) is present in 

 original form. Consequently, no DNA-RNA 

 double-helix hybrid molecules seem to be 

 formed, no complete strand separation of 

 poly (A + U) appears to be involved in 

 synthesizing poly (dA + T), and both DNA 

 strands need to be synthesized simultane- 

 ously. Other helical, double-stranded, 

 homopolymeric polyribotides can also be 

 used with DNA polymerase to produce 

 other double-stranded. homopolymeric 

 DNAs (one is probably dGdC). 



Such in vitro studies may have important 

 bearings upon the fate of RNA in vivo. Is 

 RNA (genetic or nongenetic) ever used in 

 vivo to make DNA (genetic or nongenetic)? 

 To what extent does nucleic acid replication 

 occur in vivo without complete strand sep- 

 aration and the formation of template-prod- 

 uct hybrid molecules? 



SUMMARY AND CONCLUSIONS 



DNA can be synthesized in vitro. Extended synthesis can be obtained with pre-existing 

 single-stranded DNA; the 5'-triphosphates of deoxyadenosine, deoxycytidine. deoxy- 

 guanosine, and thymidine; Mg++ ions; and a DNA polymerase. In making the prod- 

 uct, the polymerase takes directions from pre-existing single-stranded DNA acting as a 

 template. //; vitro, strands lengthen at their nucleoside end, the strand lengthened serv- 

 ing as a primer. No lag period occurs in syntheses requiring base-pairing; the amounts 

 of new DNA produced by these syntheses are either limited (when a pre-existing strand 



