The initial event in the replication cycle of parvovirus H-1 is conversion of the single-stranded linear viral DNA to the double-stranded linear replicative form. We describe here detection of an activity in uninfected cell extracts that carries out this reaction. The activity was purified and identified as DNA polymerase y.H-1 is an autonomous (helper-independent) member ofthe parvovirus group (1, 2). Its genome consists of a linear singlestranded (ss) DNA (==5 kilobases) with stable hairpin structures at both ends. After attachment to a susceptible cell and uptake of viral DNA, infecting viral ss DNA must first be converted to the double-stranded (ds) replicative form (RF) by synthesis of the complementary strand. The RF undergoes replication, probably by a continuous (as opposed to semidiscontinuous) 5' --3' displacement mechanism on both strands (1, 3-5), followed by synthesis and packaging of viral ss DNA. Given the small size of the genome and the known coding requirements for viral capsid polypeptides, it is assumed that replication must depend almost entirely on host functions, as is the case for several other prokaryotic and eukaryotic small DNA viruses.We are studying parvoviral DNA replication as a model for eukaryotic DNA replication and describe here studies on the enzymatic requirements for synthesis of parental RF from infecting viral ss DNA. Relatively little is known from in vivo experiments about this initial step in parvoviral DNA replication. The 3'-terminal foldback could provide a base-paired priming point for synthesis of the complementary strand (1, 2, 6-10), and it is assumed that this functions as the initiation site for synthesis of parental RF in vivo. Although evidence for the viral DNA to RF reaction in vivo has been reported (11,12), this step has been generally difficult to study, in part due to high particle/ infectivity ratios.In (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) mCi/Amol). In an alternative procedure, assay 4b, ATP, and phosphoenolpyruvate were replaced by 40 mM NaOAc and MgCl2 was reduced from 10 mM to 7 mM. All assay mixtures also contained 1 mM dithiothreitol and bovine plasma albumin (0.2 mg/ml).Results for response to inhibitors represent % activity remaining in the presence ofinhibitor compared with 100% in the absence of inhibitor. The concentration of 2',3'-dideoxythymidine-5'-triphosphate (ddTTP) was the same as that of dTJP; aphidicolin (gift of A. H. Todd, Imperial Chemical Industries, London) was at 3 Ag/ml; N-ethylmaleimide (MalNEt) was at 6 mM. All assays were carried out in a total volume of 30 ILI, at 370C, for 30 min (assays 1, 2, and 3) or 60 min (assay 4) (unless stated otherwise). DNA synthesis was determined by incorporation of radioactivity from dNTP into acid-insoluble material. One unit of polymerase activity = 1 nmol of total dNMP incorporated in 60 min at 37C.Fractionation of NB Cell Extracts. Cytosol was prepared from uninfected NB cells by disruption with a Dounce homogenizer followed by centrifugation to remove nuclei.Whole cell...