Replication

The two molecules of DNA polymerase used for the synthesis of both leading and lagging strands in bacteria are both DNA polymerase III. They are actually tethered together at the fork by one of the subunits of the protein, keeping their progress tightly coordinated. Many of the other players involved are also linked, so that the entire complex functions as a large molecular replicating machine.

DNA polymerase III has several special properties that make it suitable for its job. Replication of the leading strand of a bacterial chromosome requires the synthesis of a DNA strand several million bases in length. To prevent the DNA polymerase from "falling off" the template strand during this process, the polymerase has a ring-shaped clamp that encircles and slides along the DNA strand that is being replicated, holding the polymerase in place. This sliding clamp has to be opened like a bracelet in order to be loaded onto the DNA, and the polymerase also contains a special clamp loader that does this job.

A second important property of DNA polymerase III is that it is highly accurate. Any mistakes made in incorporating individual nucleotides cause mutations, which are changes in the DNA sequence. These mutations can be harmful to the organism. The accuracy of the DNA polymerase results both from its ability to select the correct nucleotide to incorporate, and from its ability to "proofread" its work.

Appropriate nucleotide selection depends on base-pairing of the incoming nucleotide with the template strand. At this step, the polymerase makes about one mistake per 1,000 to 10,000 incorporations. Following incorporation, the DNA polymerase has a way of checking to see that the nucleotide pairs with the template strand appropriately (that is, A only pairs with T, C only pairs with G). In the event that it does not, the DNA polymerase has a second enzymatic activity, called a proofreading exonuclease, or a 3′-to-5′ exonuclease, that allows it to back up and remove the incorrectly incorporated nucleotide. This ability to proofread reduces the overall error rate to about one error in a million nucleotides incorporated. Other mechanisms detect and remove mismatched base pairs that remain after proofreading and reduce the overall error rate to about one error in a billion.