The immediate effect of a temperature shift on a mutant with a mutation in a DNA replication gene depends on whether the product of the gene is continuously required for replication at the replication forks or is involved only in the initiation of new rounds of replication. For example, if the mutation is in a gene for DNA polymerase III or in the gene for the DnaG primase, replication ceases immediately. However, if the temperature-sensitive mutation is in a gene whose product is required only for initiation of DNA replication, for example, the gene for DnaA or DnaC (see “Initiation of Chromosome Replication” below), the replication rate for the population will slowly decline. Unless the cells have been somehow synchronized in their cell cycles, each cell is at a different stage of replication, with some cells having just finished a round of replication and other cells having just begun a new round. Cells in which rounds of chromosome replication were under way at the time of the temperature shift will complete their replication cycle but will not start a new round. Therefore, the rate of replication decreases until the rounds of replication in all the cells are completed.
Replication Errors
To maintain the stability of a species, replication of the DNA must be almost free of error. Changes in the DNA sequence that are passed on to subsequent generations are called mutations (see chapter 3). Depending on where these changes occur, they can severely alter the protein products of genes or other cellular functions. To avoid such instability, the cell has mechanisms that reduce the error rate.
As DNA replicates, the wrong base is sometimes inserted into the growing DNA chain. For example, Figure 1.11 shows the incorrect incorporation of a T opposite a G. Such a base pair in which the bases are paired wrongly is called a mismatch. Mismatches can occur when the bases take on forms called tautomers, which pair differently from the normal form of the base (see chapter 3). After the first replication shown in Figure 1.11, the mispaired T pairs with an A, causing a GC-to-AT change in the sequence of one of the two progeny DNAs and thus changing the base pair at that position on all subsequent copies of the mutated DNA molecule.
Figure 1.11 Mistakes in base pairing can lead to changes in the DNA sequence called mutations. If a T is mistakenly placed opposite a G during replication (A), it can lead to an AT base pair replacing a GC base pair in the progeny DNA (B to D).
Editing
One way the cell reduces mistakes during replication is through editing functions. With some DNA polymerases the editing function resides in the same protein, while in other cases a separate protein performs the editing function. Editing proteins are aptly named because they go back over the newly replicated DNA looking for mistakes and then recognize and remove incorrectly inserted bases (Figure 1.12). If the last nucleotide inserted in the growing DNA chain creates a mismatch, the editing function stops replication until the offending nucleotide is removed. DNA replication then continues, inserting the correct nucleotide. Because the DNA chain grows in the 5′-to-3′ direction, the last nucleotide added is at the 3′ end. The enzyme activity found in a DNA polymerase or one of its accessory proteins that removes this nucleotide is therefore called a 3′ exonuclease. The editing proteins probably recognize a mismatch because the mispairing (between T and G in the example in Figure 1.11) causes a minor distortion in the structure of the double-stranded helix of the DNA.
DNA polymerase I is an example of a DNA polymerase in which the 3′ exonuclease editing activity is part of the DNA polymerase itself. However, in DNA polymerase III, which replicates the bacterial chromosome, the editing function resides in an accessory protein encoded by a separate gene whose product travels along the DNA with the DNA polymerase during replication. In E. coli, the 3' exonuclease editing function is encoded by the dnaQ gene (Table 1.1), and dnaQ mutants, also called mutD mutants (i.e., cells with a mutation in this gene that inactivates the 3' exonuclease function), show much higher rates of spontaneous mutagenesis than do cells containing the wild-type, or normally functioning, dnaQ gene product. Because of their high spontaneous mutation rates, mutD mutants of E. coli can be used as a tool for mutagenesis, often combined with mutations in other genes whose products normally contribute to the correction of mismatches (see chapter 10).
Figure 1.12 Editing function of DNA polymerase. (A) A G is mistakenly placed opposite an A while the DNA is replicating. (B and C) The DNA polymerase stops while the G is removed and replaced by a T before replication continues.
RNA Primers and Editing
The importance of the editing functions in lowering the number of mistakes during replication may explain why DNA replication is primed by RNA rather than by DNA. When the replication of a DNA chain has just initiated, the helix may be too short for distortions in its structure to be easily recognized by the editing proteins. The mistakes may then go uncorrected. However, if the first nucleotides inserted in a growing chain are ribonucleotides rather than deoxynucleotides, an RNA primer is synthesized rather than a DNA primer. The RNA primer can be removed and resynthesized as DNA by using preexisting upstream DNA as a primer. Under these conditions, a distortion in the helix can be detected by the editing functions, and mistakes are avoided.
Another important system that safeguards the fidelity of the replication process is responsible for fixing mismatches after the growing DNA strand leaves the polymerase. In E. coli and its closest relatives, this process is guided by methylation and is termed methyl-directed mismatch repair. Related mismatch repair systems are used across all three domains of life, but the use of methylation signals is not widespread. The methyl-directed mismatch repair system is discussed in chapter 10.
Impediments to DNA Replication
While the process described above and diagrammed