DNA and Dementia

Jay H. Robbins Defects in the body's DNA repair system long have been linked to skin cancer, but evidence is now emerging that similar problems play a role in Alzheimer's disease and other neurodegenerative disorders. New findings by senior investigator and medical officer Jay H. Robbins and colleagues in the dermatology branch at the National Cancer Institute (NCI) go far toward validating a hypothesis broached by Robbins a quarter-century ago. That's how long his team has been following clinical cases of xeroderma pigmentosum (XP), a rare, autosomal recessive disorder that causes at least 1,000 times the normal risk of skin cancer1 and also sometimes involves premature neuronal death. Certain DNA lesions found in Alzheimer's exist in some XP patients. In both diseases, free radicals that arise as by-products of oxygen metabolism in the brain appear to cause these lesions. Mutations in nucleotide excision repair (NER), one of several endogenous methods of rehabilitating damaged DNA, prevent mending of the lesions. The new findings indicate that Robbins' theory, which links defective DNA repair to neurodegeneration in XP,2 now may also apply to Alzheimer's and other dementias. "I knew that the neuronal degeneration had to be explained on the basis of mutation," Robbins says of his early thinking about XP. Yet the source of such mutation had to be different from ultraviolet (UV) rays of sunlight known to cause DNA repair defects that elevate the skin cancer risk among XP patients. Harmful UV light can't penetrate the skull to reach neurons, but several by-products of endogenous processes could cause central nervous system damage if DNA repair were defective. Over the years, evidence mounted that free radicals, especially hydroxyl radicals, were the culprits. One such free radical-induced lesion is 8,5-(S)-cyclo-2-deoxyadenosine (cyclo-dA). Obtaining tissue with such lesions has long been a stumbling block, but finally, Robbins and coworkers reported in a recent abstract that they had developed a synthesis for cyclo-dA and incorporated it into DNA to test its repair in mammalian cell extracts and living cells. They found that the lesion is repaired by the same NER process that is deficient in XP, and not by the base excision repair pathway that usually responds to free radical-induced lesions. Moreover, cells from XP patients with neurodegeneration fared extremely poorly at repairing cyclo-dA, implicating the lesion as an endogenous cause of XP neurodegeneration. The full paper has just been published,3 corroborating similar findings by a European group that was completing its paper when the NCI team's abstract appeared.4 One difference between the two projects is that Robbins' group developed an assay that allowed them to insert the lesion into living cells via the luciferase reporter gene rather than work with DNA plasmid. About 30 DNA repair proteins have been identified from in vitro models, and many have been cloned, but Robbins thinks the actual number in a cell will be between 100 and 150 proteins. In the repair process, chromatin is somehow "finessed," perhaps to provide clearance for a preassembled protein machine or repairosome, although another hypothesis holds that the proteins required for repair are individually recruited to the damaged site. "Probably the least understood area of repair is how the chromatin is modified and what has to be modified," comments Randy J. Legerski, an associate professor of molecular genetics at M.D. Anderson Cancer Center in Houston. Transcription-Coupled Repair Incisions are made on the transcribed strand for removal of an approximately 28-nucleotide sequence that includes a "bulky" lesion. The gap is then repaired by DNA polymerase, and a ligase seals the newly synthesized patch to the parental DNA. This process, called transcription-coupled repair, is supplemented by the random, slower function of global genome repair that occurs on the untranscribed strand. Errol C. Friedberg XP and either of two other rare diseases, trichothiodystrophy and Cockayne syndrome, share various defective genes. Although the latter two are associated with DNA repair deficiency, neither is characterized by a predisposition to cancer. All the genes common to these disorders belong to a transcription complex subunit, notes Errol C. Friedberg, professor and chairman of pathology at the University of Texas Southwestern Medical Center in Dallas. Transcription is plainly at least part of the disease phenotype, but the fact that patients with these genetic defects proceed through embryogenesis and are born indicates that the deficient transcription factors are not essential to development. The use of mouse models has been "incredibly informative" in recent years, Friedberg says. In collaboration with a Dutch team from the University of Leiden, his lab recently found that mice with the C form of XP disease develop an age-dependent decrease in DNA repair capability, resulting in a 30 percent increase in spontaneous mutations compared to normal mice. Publication of this research is pending. "Clearly, XPC mice are accumulating mutations spontaneously as they get older," he comments, adding that this implies a potential relationship between XPC and neurodegenerative disease. The importance of maintaining a stable genome through successful repair pathways is underscored by the estimated 20,000 times daily that each human cell incurs endogenous DNA damage.5 "NER repairs some types of oxidative damage that occur routinely in cells and may occur even more often in some brain cells," Legerski points out. "In the body, we know there is specificity of repair of one type of bulky lesion over another," Robbins notes. Using reporter genes to carry lesions into a host cell will help to answer whether one lesion is repaired by NER while others are not. "In Alzheimer's disease, we know there's no problem with UV damage, but we're looking for different lesions that no one yet has looked at. That's the basis of my theory." Neurodegeneration in XP There are seven forms, or complementation groups, of XP, ranging from XPA to XPG, plus a variant that is not NER related. For years, researchers thought neuronal damage was confined to XPA, but Robbins stresses that "complementation" only means that the same gene is affected in a particular group, not that the same mutation occurs. "It's a question of how many mutations there are, and of what clinical signs each one gives." His team is now preparing a review paper that describes the occurrence of XP neurological disease in three groups, as well as Cockayne syndrome-related neurodegeneration in two other groups containing patients with an XP-Cockayne complex. XP symptoms often start at a very early age. Aside from skin lesions and mental deterioration, they can include ataxia, spasticity, and deafness. Yet, Robbins says, "We've had patients who went to school, drove a car, and others who, until they were adults, had no evidence of the disease." In the United States, XP occurs in one of about every 250,000 people. "It takes 20 to 30 years before the establishment in a field like neurology really appreciates a very rare disease," Robbins believes. The range and difficulty of neurological studies preclude much attention given to rare and complex disorders such as XP, he explains. "This hurts research." Nowadays, federal government data show that more than 65 research projects are under way concerning XP. Anne Pilaro of the division of clinical design and trial analysis in the Food and Drug Administration Center for Biologics Evaluation and Research identifies DNA excision repair as one of the fastest- growing areas of preclinical gene therapy research. Robbins expresses satisfaction that his long-held views concerning DNA repair and neurodegeneration are finally being put to the test. Steve Bunk (sbunk@uswest.net) is a contributing editor for The Scientist. References 1. P. Smaglik, "Delving into the choreography of DNA repair," The Scientist, 13[7]:14-5, March 29, 1999. 2. J.H. Robbins et al., "Xeroderma pigmentosum: an inherited disease with sun sensitivity, multiple cutaneous neoplasms and abnormal DNA repair," Annals of Internal Medicine, 80:221-48, 1974. 3. P.J. Brooks et al., "The oxidative DNA lesion 8,5'-cyclo-2'-deoxyadenosine is repaired by the nucleotide excision repair pathway and blocks gene expression in mammalian cells," Journal of Biological Chemistry, 275:22355-62, July 21, 2000. 4. I. Kuraoka et al., "Removal of oxygen free-radical-induced 5',8-purine cyclodeoxynucleosides from DNA by the nucleotide excision-repair pathway in human cells," Proceedings of the National Academy of Sciences, 97:3832-7, April 11, 2000. 5. T. Lindahl, "Instability and decay of the primary structure of DNA," Nature, 362:709-15, 1993.

DNA and Dementia

Interested in reading more?

Become a Member of