11 DNA Replication: Topological Aspects and the Roles of DNA Topoisomerases

Abstract

The advent of the double-helix structure of DNA raised immediately the question of how the intertwined strands could come apart during replication (Delbruck 1954; Delbruck and Stent 1957). The problem was initially viewed as a kinetic one; as the strands separate, the unreplicated portion ahead of the replication fork would have to rotate rapidly around its helical axis. A rate of about 10,000 revolutions per minute was estimated based on the rate of replication in bacteria.

The problem of unraveling the parental strands during semiconservative replication became a topological one when the entire genome of the bacterium Escherichia coli was found to be in the form of a double-stranded ring. A duplex DNA ring can be viewed as two multiply-linked single-stranded rings of complementary nucleotide sequences, and the unlinking of such a pair is impossible without disrupting, at least transiently, the continuity of the strands (Cairns 1963Cairns 1964; Cairns and Davern 1967). Therefore, an enzyme or a group of enzymes must serve the role of a “swivel” to allow the unraveling of the intertwined single-stranded DNA (ssDNA) rings. The situation is not very different for long linear chromosomes organized into multiple loops, each of which appears to define a topological domain (for review, see Gross and Garrard 1988).

The topological problem described above for strand separation is basically one for the elongation step of replication. Because the linking number between the parental strands of a DNA ring or loop decreases continuously during replication, a swivel is required at most if...