ABSTRACr Because of their association with protein, short nascent DNA chains in Escherichia coli can be separated from other cellular DNA by chromatography on hydroxylapatite. Protein-free DNA chains of less than 500 nucleotides in length are resistant to degradation from the 5'-end by alkaline phosphatase [orthophosphoric-monoester phosphohydrolase (alkaline optimum); EC 3.1.3.11 and spleen phosphodiesterase (oligonucleate 3'-nucleotidohydrolase; EC 3.1.4.18). In contrast, DNA chains containing more than 500 nucleotides are degradable. From these results we conclude that short nascent DNA chains are structurally modified at their 5'-ends. The nature of this structure and its possible functions are discussed.Okazaki pieces, short DNA chains of about 1000 nucleotides in length, have long been considered the first intermediates in DNA replication (1). Recent evidence, however, shows that Okazaki pieces arise through the joining of many much shorter DNA chains (2, 3). Such short DNA chains accumulate in certain dna mutants (4, 5), suggesting that more than one reaction is involved in the synthesis of Okazaki pieces. The formation of a single Okazaki piece seems to involve many independent initiation events. Since all known DNA polymerases require a primer for the synthesis of a new DNA strand, we decided to study the 5'-ends of the short DNA chains, hoping that their structure might provide some clue to the mechanism of DNA chain initiation. In the course of these studies we found that the shortest replication intermediates, containing fewer than 500 nucleotides, carry an unusual group at-their 5'-ends, which may function in the initiation or elongation of nascent DNA chains.
MATERIALS AND METHODSBacterial Strains and Culture Conditions. Escherichla colh strain 15 TAMT (thy-, arg-, met-, trp-) was grown at 140 in LSTL medium supplemented with 5 ,tg/ml of thymine and with 40,g/ml each of arginine, methionine, and tryptophan, as described (2).Labeling and Cell Lysis. In experiments where intracellular thymidine nucleotide pools were depleted, bacteria were incubated for 30 min in the absence of thymine and then pulselabeled with 1 ,gg/ml of [3H]thymine (50 Ci/mmol) for 10 sec as described (2). For steady-state labeling, an equal volume of LSTL medium, containing 5 ,Ag/ml of [3H]thymine, was added to the culture, and incubation was continued for 0.3-10 min (2). Incorporation of label was terminated with phenol-ethanol, and the cells were lysed as described (2)