Studies of DNA replication in vivo, II. Evidence for the second intermediate.

In the preceding paper,1 it was demonstrated that the initial product of bacterial DNA synthesis is isolated as single-stranded DNA. The characteristics of this first intermediate of DNA synthesis suggested possible models for the templating process of DNA replication. The isolated single-stranded DNA has a much smaller molecular weight than the bulk DNA, confirming the recent findings of Okazaki et al.2 3 However, the secondary structure of the replicating point still remains largely unknown despite preliminary indications of some unusual properties of replicating DNA.2-6 Kidson separated the newly synthesized DNA by countercurrent distribution and suggested that this DNA is partially denatured.6 It became important to try to trace the fate in the cells of the newly synthesized DNA, which is isolated in a single-stranded form. This report concerns the isolation of a second intermediate, from Bacillus subtilis, which may represent a second stage of DNA replication. This intermediate exhibits many of the characteristics of double-stranded DNA. However, there is evidence that part of it has a single-stranded structure. Materials and Methods.-Bacterial strains and media: A thymine-requiring B. subtilis strain (168-thy)7 was used throughout these studies. Medium Y' and medium Y -T were used. Labeling and isolation ofDNA: The same procedures were used as reported previously, except for the following changes. (1) Cells were resuspended in half a volume of Y T (25 ml) after being washed on Millipore filters. (2) Buffer C (0.01 M Tris, pH 7.8, 0.01 M EDTA, 0.015 M NaCl, 0.02 Al NaN3, and 0.02 M NaF) was used instead of buffer B. (3) Lysozyme treatment was carried out at 00C for 40 min and then at 370C for 10 min. Hydroxyapatite chromatography of DNA: Chromatography procedures8 and a batch procedure to fractionate DNA by hydroxyapatitel were reported previously. Nitrocellulose treatment of DNA: Nitrocellulose powder (Hercules Powder Co., kindly provided by Drs. T. Kasai and E. K. F. Bautz) was homogenized in 2 X SSC by a Waring Blendor for 3 min before use. Approximately 100 mg of nitrocellulose was mixed with DNA samples that had been dissolved in 2 X SSC, and the mixture was shaken for 5 min. After centrifugation, the nitrocellulose was shaken with 2 X SSC once more, and both supernatants were combined. To calculate the percentage of DNA bound to nitrocellulose, the supernatant radioactivity was subtracted from the total input-sample radioactivity before treatment. Approximately 95% of heat-denatured DNA and 20-40% of native DNA bind with nitrocellulose under these conditions. In a limited number of experiments, essentially the same results were obtained with commercially available nitrocellulose filters (S & S B6 filters). Cs2SO4-Hg(II) centrifugation: Cs2SO4-Hg(II) centrifugation was carried out according to Nandi et al.9 All the DNA samples were dialyzed against 0.02 Al Na2SO4, 0.002 M borate buffer at pH 8.7 before being mixed with HgC12. Results.-As reported in the preceding paper,1 under the conditions employed, most of the tritium-labeled DNA from cultures receiving the shortest pulse (5 sec) was isolated as single-stranded DNA for both B. subtilis and Escherichia coli.