Vol. 45, 1959 BIOCHEMISTRY: KORNBERG ET AL. 777 



each of the four deoxyriucleotides incorporated into phage T2 DNA reach values of 

 about the same magnitude. By contrast, only traces of dC-5-P kinase were 

 detected (Fig. 2 E). Furthermore, extracts of T2-infected cells Avere found to 

 inhibit the dC-5-P kinase activity of normal cell extracts when equal amounts of 

 infected and non-infected extracts were mixed. As will be described below, this 

 inhibitory effect is due to an enzyme (dCTPase) which splits dCTP. By use of 

 fluoride (8 X 10 ~^ M), which inhibits dCTPase more than 98 per cent, but the dC- 

 5-P kinase by 15 per cent or less, it was possible to show that there was actually 

 little or no change in the dC-5-P kinase levels upon infection. Extracts prepared by 

 Method II showed the same kinase patterns after T2 infection. 



Infection with phage T5, which contains cytosine rather than HMC, showed in- 

 creased kinase activities for the four deoxynucleotides which are present in its 

 DNA (Fig. 2 B, C, D, E). The 10-fold increase in dC-5-P kinase activity in the 

 extracts of T5-infected cells may be contrasted with the absence of any increase in 

 this activity in the extracts of T2-infected cells. 



Hydroxymcthylating activity was first detected at 4 min after T2 infection, and, 

 as predicted from the results of Flaks and Cohen,'' was absent from T5-infected cell 

 extracts (Fig. 2/^). 



An Enzyme which Destroys Deoxycytidine Triphosphate (dCTPase). — The in- 

 hibitor of dC-5-P kinase that develops upon T2 infection has been identified as 

 an enzyme which splits dCTP by removal of the terminal pyrophosphate group 

 (Fig. 3). Extracts of uninfected cells have 1 per cent or less of the dCTPase 

 activity observed in extracts of T2-infected cells; the level of dCTPase activity 

 of To-infected cells was the same as in normal cells. The relative insensitivity to 

 fluoride of the dCTPase activity in normal cells makes it doubtful that it represents 

 the same enzyme found in the T2-infected cells. 



After partial purification of the dCTPase, ^^ which reduced the level of an inor- 

 ganic pyrophosphatase to 2 per cent of the dCTPase, it was demonstrated that 

 the complete splitting of 0.92 ^tmole of dCTP was accompanied by the appearance 

 of 0.74 iumole of inorganic pyrophosphate identified as Norit-nonadsorbable, acid- 

 labile P and 0.15 Mmole of orthophosphate. With the purified enzyme preparation, 

 the rates of cleavage of cytidine triphosphate and ATP were less than 1 per cent of 

 that of dCTP ; the splitting of dHMC-TP was 1-2 per cent of that of dCTP. The 

 Km of dCTP for the enzyme is in the region of 10 ~^ M. In view of this high afl^nity 

 for dCTP and of the high level of enzyme activity compared to the dC-5-P kinase 

 in the extract (cf. Fig. 2 E and Fig. 3), it is reasonable to suppose that the dCTP 

 concentrations in the infected cell are reduced to an extremely low level. 



An Enzyme which Glucosylates the HMC of DNA. — An enzyme which transfers 

 glucose from UDPG to DNA containing HMC was observed in extracts of T2- 

 infected but not T5-infected or normal cells (Fig. 4). With the partially purified 

 enzyme'^ UDPG could not be replaced by glucose, glucose 1-phosphate or glucose 

 6-phosphate (Table I) ; similar results were also obtained with the crude extracts. 

 The reaction requires HMC-containing DNA, AAhich for these experiments was 

 enzymatically synthesized from dHMC-TP, dATP, dGTP, dTTP, purified poly- 

 merase, and primer DNA derived from any one of several sources (calf thymus, 

 E. coli, phage T2, phage <t)X174^^). When DNA enzymatically synthesized with 

 dCTP in place of dHMC-TP, or when the glucosylated DNA from phage T2 itself 



130 



