Human DNA Polymerase [Iota]: Wrong for the Right Reasons.

In a recent collaborative study, Thomas A. Kunkel, director of the NIEHS Environmental Biology Program, NIEHS deputy director Samuel H. Wilson, and Roger Woodgate, an investigator at the National Institute of Child Health and Human Development, report novel findings about DNA repair and genome stability in humans. Woodgate's laboratory recently discovered and purified polymerase iota (pol [Iota], one of a superfamily of DNA polymerases that synthesize short stretches of DNA and are relatively inaccurate--that is, they may allow the A, C, G, or T base moieties to be matched to an incorrect mate (for instance, matching G with T or A, rather than with its proper mate, C). The scientists speculate that these enzymes' importance therefore lies in DNA repair and damage-induced mutagenesis, not in DNA replication.

The highly accurate DNA polymerases that carry out DNA replication are blocked by DNA damage. In contrast, some of the less accurate polymerases, also called translesion synthesis (TLS) polymerases, bypass damage in the DNA, fall off after inserting a few nucleotides, and then allow an accurate replicative polymerase to continue. Thus, TLS polymerases may allow damaged cells to survive at the cost of a slightly higher mutation rate in regions of DNA damage.

The results of the collaboration between the Kunkel, Wilson, and Woodgate laboratories, published in the 16 March 2001 issue of Science, indicate that pol [Iota] has a novel enzymatic activity called 5'-dRPase, which removes damage from the 5' end of a broken DNA strand. This activity was previously demonstrated in polymerase beta, which is active in a repair pathway called base excision repair (BER). dRPase activity is required during many BER reactions, suggesting that pol [Iota] could participate in BER.

In addition, Kunkel and colleagues show that the sequence of DNA in question greatly influences whether pol [Iota] forms a correct base pair (namely AT, TA, GC, or CG) or an incorrect base pair (for example, GT). Pol [Iota] forms an AT base pair inserting T opposite A 40- to 100-fold more efficiently than it forms the other correct base pairs, a property that had not been reported previously. However, says Kunkel, "The most unusual thing about pol [Iota] is its preference to insert G opposite T rather than A opposite T." The rate at which pol [Iota] makes this error during DNA synthesis is higher than observed for any base pair. Based on their results, Kunkel and colleagues suggest two novel roles for pol [Iota] in BER.

In the first possible role, pol [Iota] may play a role repairing uracil (U) in DNA. U is a normal, correct base in RNA that is not found in normal undamaged DNA. It forms in DNA spontaneously and enzymatically when a C base undergoes a normal damage process known as deamination. It can also be misinserted during normal DNA replication. In this role, U is removed enzymatically from an incorrect AU base pair, and then pol [Iota] replaces it with T, a reaction that calls on the dRPase activity of pol [Iota] and that is highly accurate.

In the second role, pol [Iota] could also prevent misrepair of GT mismatches. In this scenario, a GT base pair forms spontaneously when C undergoes deamination. When this occurs, G is the correct base and T is the incorrect base in the mispair. The incorrect T base should be removed and replaced with a C base, but if the correct G is removed instead, a mutation could result. In this case, Kunkel and colleagues suggest that pol [Iota] might reinsert G opposite T as a "correct" insertion event that stabilizes the genome, allowing the normal sequence to be restored.

This is a surprising suggestion, because it is thought that TLS polymerases are inaccurate and that they form mispairs that increase the mutation rate. Kunkel's idea is that pol [Iota] forms a mispair in a specific context where it actually decreases the mutation rate. Errol Friedberg, a professor of pathology at the University of Texas Southwestern Medical Center at Dallas, finds this "a very interesting idea" that suggests that these polymerases "may have very specialized functions that allow them to be error-free."

Friedberg points out that "the biochemical evidence needs to be backed up by genetic evidence" in order to establish the full extent of the role or roles of pol [Iota] in vivo. It is clear that pol [Iota] creates errors with high frequency in certain sequence contexts; thus, this enzyme, like other TLS polymerases, must be tightly regulated in vivo. More research is needed to clarify the biologic roles and the regulation of pol [Iota] and other TLS polymerases.