The Oncologist 2005;10:361–362 www.TheOncologist.com Correspondence: David S. Goodsell, Ph.D., Associate Professor, The Scripps Research Institute, Department of Molecular Biology, 10550 North Torrey Pines Road, La Jolla, California 92037, USA. Telephone: 858-784-2839; Fax: 858-784-2860; e-mail: goodsell@scripps.edu Website: http://www.scripps.edu/pub/goodsell Received February 25, 2005; accepted for publication February 25, 2005. ©AlphaMed Press 1083-7159/2005/$12.00/0 Our long, delicate DNA strands are easily broken. Ionizing radiation, such as x-rays and gamma rays, as well as drugs like bleomycin (Blenoxane®; Bristol-Myers Squibb, Princeton, NJ, http://www.bms.com) create reactive forms of oxygen, which in turn attack DNA and cause breakage. Double-stranded DNA breaks can occur when DNA polymerase runs into an unrepaired nick in the DNA. Topoisomerase inhibitors can also cause breaks: topoisomerase breaks and rejoins DNA in the course of its function, and inhibitors can block the rejoining step. Cells also break their DNA on purpose for special functions, most notably during the gene shuffling that occurs as lymphocytes mature, which generates diversity in antibodies, T-cell receptors, and other highly variable immune system proteins. These breaks can cause serious problems. A single break in a key gene can kill a cell, or cause it to kill itself by apoptosis. So cells have powerful methods to repair this damage as soon as it happens. In your lifetime, each of your cells will have repaired, more or less successfully, several thousand double-stranded DNA breaks. Radiation therapy overwhelms this natural repair system, using high doses of radiation to fragment the DNA in cancer cells. Cells use two major methods to repair double-stranded DNA breaks. The first method—homologous recombination—uses the fact that we carry a duplicate set of DNA in our cells. The break is repaired using the duplicate set as a template. As you might imagine, this can be very precise, since the cell can use the undamaged DNA strand to ensure that the repair is correct. The second method—nonhomologous end joining—repairs the break directly, without any outside information. It is less accurate, and may result in the addition or removal of a few nucleotides at the repair site. Nonhomologous end joining requires the concerted action of a series of proteins. The process is thought to start with the Ku protein (Fig. 1), a dimer composed of two similar proteins. It is prevalent in the cell nucleus and binds readily to DNA ends. Ku then binds to DNA-dependent protein kinase and begins the process of synapsis that holds the two broken ends in close proximity. Other proteins, such as Artemis, and perhaps polymerases, then bind to the break, trimming the two ends and filling in gaps, making them ready for rejoining. Finally, the two ends are rejoined by DNA ligase IV with the help of XRCC4 (Fig. 2). Because of the trimming that occurs at each end, and because synapsis may occur between any two broken DNA ends, this process is imprecise. In the case of the antibody genes, this is a good thing, since it is the way that our immune system builds a large repertoire of slightly different antibodies. But for repair of accidental damage, these small (and large) errors can be dangerous, in some cases Fundamentals of Cancer Medicine The Oncologist