Abstract: "The High Fidelity of DNA Duplication".

Radman, Miroslav and Robert Wagner. "The High Fidelity of DNA Duplication". Scientific American. pp40-46.

Why are so few mistakes made when DNA is duplicated? The set of genetic instructions in humans is approximately 3 billion characters long. If mistakes are as rare as one in a million, 3,000 mistakes would be made during each duplication of the human genome. Since the genome replicates about a million billion times in the course of building a human being from a single fertilized egg, it is unlikely that the human organism could tolerate such a high rate of error. In fact, the actual rate of mistakes is more like one in 10 billion. How do cells achieve such high fidelity?

DNA is a double-stranded macro-molecule. The two strands are complementary. Thus, if one strand of the DNA becomes damaged, the other strand can be used as a model to replace the damaged strand. DNA is also replicated using a similar process: First, the strands split into two different strands, and each strand acts as a template for building another strand. The DNA is replicated by bonding of four complementary nucleotide bases. Without enzymes, mismatch errors would occur about once in every 100 bases. Three enzymatic systems improve the accuracy of DNA replication to about one error in every 10 billion bases.

The first enzyme system which improves the accuracy of DNA replication is the one which controls nucleotide selection. Called DNA polymerase, it moves along the DNA strand and synthesizes the complementary base from the cellular nucleotide. It is possible to insert any base opposite any other base, but the insertion of the complementary base is always the most energy effective. Nevertheless, in approximately one in every 100,000 cases, a non-complementary base is incorporated into the DNA strand.

The second systems acts as a "proofreader". If a non-complementary base has been bound into the DNA chain, this enzyme will immediately expel it, before more continuing down the strand. However, the accuracy of these first two enzyme system will be decreased if the four nucleotide bases (guanine, adenine, thymine, and cytosine) are located in the cell in disproportionate numbers.

The third system corrects errors in the DNA strand after the it has been synthesized. To correct the error, the enzyme must be able to discriminate between the parent and child strands. In some bacteria (such as Escherichia coli), a methyl group is added after the a certain base pattern is encountered. There is a small time lag between the synthesis of the strand, and the time which the methylization occurs. Thus, during this time lag, the enzyme can distinguish between the parent and child, and ensure that the base patterns of the child match those of the parent, and correct any errors which have occurred.

Many organisms, however, do not have a methylization process such as that found in the bacteria. Some alternate theories have been posed for DNA replication in the cells of other organisms. One theory is based upon the fact that newly replicated DNA is initially discontinuous, thus the organisms contain nicks. The enzyme can use these nicks to differentiate between the original, and its replicate.

One illustration included with this article depicted the "proofreading process". It consisted of three sub-drawings, showing in sequence the binding of non-complementary bases, the enzyme's removal of the 'bad' base, and the eventual binding of the proper base to the DNA strand. The drawing portrayed the different by using different colors and endings to clearly show which bases were paired together. This drawing enhances the article by providing a visualization of the process described in the article.

This article answers a small number of current questions, but raises many more. It answers questions concerning the DNA replication-error correction facilities in E. coli, but raises further questions concerning the mechanisms in other organisms. The article also opens up an entirely new field for debate: Why are there not more error-avoiding and error-correcting mechanisms located in the cells?