A segment of DNA whose replication starts from a replication origin and proceeds unidirectionally or bidirectionally to one or two sites of termination of DNA replication is called a replicon, a unit of DNA replication. In each replicon, replication is continuous from the origin to the terminus and is accompanied by the movement of the replicating point, called the replication fork. Both parental DNA strands are replicated concurrently at the fork. However, replication at a fork is semidiscontinuous: DNA synthesis is continuous on one strand, the leading strand, and discontinuous on the other, the lagging strand (see Discontinuous DNA Replication). This occurs because the two chains of double helical DNA are antiparallel, and DNA polymerase can extend a DNA chain only in the 5′ ^ 3′ direction.
The parent strand that runs 5′ 3′ in the reverse direction of fork movement is termed the leading strand, and it serves as a template for the continuous DNA synthesis, in which the DNA polymerase carries out chain elongation in a highly processive manner. The other parent strand runs 5′ ^ 3′ in the direction of fork movement and is termed the lagging strand; it serves as a template for the discontinuous DNA synthesis. Short pieces of DNA, called Okazaki Fragments, are repeatedly synthesized on the lagging strand; these Okazaki fragments are a few thousand nucleotides in bacterial cells and a few hundred in eukaryotic cells.
In Escherichia coli, Pol III holoenzyme is the major replicative DNA polymerase for both leading-and lagging-strand synthesis. The Pol III holoenzyme is a huge multiprotein complex that consists of 10 distinct polypeptide chains (1, 2). This enzyme extends the DNA chain with a high processivity (>500 kb of DNA can be synthesized continuously without the dissociation of polymerase from the template) and high catalytic efficiency (the velocity of chain elongation is 1000 nucleotides per second at 37°C). The catalytic core, composed of three subunits, contains the polymerase activity and a 3′ ^ 5′ exonuclease for proofreading (3). The remaining seven auxiliary subunits enhance the processivity of the core by clamping it onto the template (4). They also promote the repeated association of the polymerase necessary for discontinuous synthesis of the lagging strand. Structural analysis of the Pol III holoenzyme and studies on a reconstituted replication fork suggest that the holoenzyme is an asymmetric dimer with twin polymerase active sites: One half of the dimer has high processivity and might be the polymerase for continuous synthesis of the leading strand, whereas the other half has the recycling capacity needed for lagging-strand synthesis (5). Thus, it seems likely that a single molecule of Pol III holoenzyme acts at the replication fork catalyzing concurrently both leading- and lagging-strand synthesis (6, 7).
In eukaryotic DNA replication, the division of labor among the polymerases remains ambiguous. Pol a (Pol I in yeast) is apparently involved in DNA replication, since mutant cells defective in this polymerase activity are inviable. However, Pol a lacks the 3′ ^ 5′ exonuclease activity, so its DNA synthesis is inaccurate and shows a low processivity. These enzymatic characteristics make Pol a a poor candidate for the major replicative polymerase. In addition, Pol a is unique in possessing a primase activity, the only such activity thus far identified in eukaryotic cells, suggesting that Pol a may play a role in the priming of DNA synthesis (8). On the other hand, Pol d and Pol e each possess a 3′ ^ 5′ exonuclease activity and are highly processive polymerases in the presence of proliferating cell nuclear antigen (PCNA) and replication factor C (9, 10). Their yeast counterparts, Pol III and Pol II, respectively, are essential for cell growth and DNA replication. Therefore, these polymerases are better suited for chromosome replication. However, the specific roles of Pol d and Pol e at the replication fork remain controversial.
There is another difference in the enzymatic processes of synthesizing the leading and lagging strands. Leading-strand DNA synthesis requires RNA primer only once in the replication of each replicon, but a frequent priming process is associated with lagging-strand DNA synthesis. RNA primer must be laid down as the initiation step of each cycle of synthesis of Okazaki fragments. Therefore, the priming proteins, including the primase, are required for lagging-strand DNA synthesis, along with the replication fork.
