Mechanisms of Disease: DNA Repair Defects and Neurological Disease

Kalluri Subba Rao


Nat Clin Pract Neurol. 2007;3(3):162-172. 

In This Article

Neurological Diseases Linked to DNA Repair Defects

Any recent text book of neurology will list at least 200 neurological disorders, with diverse etiologies and genetic characteristics. A number of these disorders have a definitive link to an inherited or acquired defect in one of the DNA repair pathways. Table 1 lists these disorders along with the known genetic DNA repair defect and the nature of the inheritance. In addition, there are a handful of syndromes with an etiological link to DNA damage and repair that exhibit symptoms of accelerated aging but have no striking neurological symptoms. Such syndromes include Werner's syndrome, Bloom syndrome and Rothmund–Thomson syndrome (none of which is included in Table 1 ).

The syndromes and diseases listed in Table 1 show some variation in the precise pathway or gene involved in the associated defective DNA repair mechanism. The first four disorders seem to result from defects in the excision repair pathway. In view of their overlapping symptoms, they are discussed together here.

Xeroderma Pigmentosum, Cockayne's Syndrome, Trichothiodystrophy and Down Syndrome

In the case of xeroderma pigmentosum, seven complementation groups (XPA, ERCC3 [XPB], XPC, ERCC2 [XPD], DDB1, ERCC4 and ERCC5) correspond to mutations in seven genes that have a role in NER (Figure 1). An additional variant, xeroderma pigmentosum V, is caused by a mutation in a novel DNA polymerase η belonging to the Y family of DNA polymerases, which supports translesion synthesis.[24,25] An interesting disease feature seen in some, but not all, patients with xeroderma pigmentosum is progressive neurodegeneration. Patients who experience severe neurodegeneration are often found to have mutations in components of the transcription-coupled repair system, such as the Cockayne's syndrome proteins ERCC8 (CSA) and ERCC6 (CSB; see below), and common genes such as XPA, ERCC2 (previously XPD), ERCC3 (XPB), ERCC4 (previously XPF) and ERCC5 (previously XPGC).

Cockayne's syndrome is an autosomal recessive disease, symptoms of which include growth retardation, deafness, dysmyelination in white matter, and retinal and Purkinje's cell degeneration. This syndrome is not associated with cancer or loss of personality. The two proteins found to be mutated in this syndrome, ERCC8 and ERCC6, have been shown to be required for transcription-coupled repair.

Some information regarding the precise role of the ERCC6 and ERCC8 protein factors in transcription-coupled repair is now known. The most recent studies, from two independent groups,[26,27] can be summarized as follows. ERCC8 is a subunit of E3 ubiquitin ligase complex, which is part of the ubiquitination machinery, and ERCC6 is the substrate for this ligase. When exposure to ultraviolet light stalls the RNA-polymerase-II-mediated transcription process, ERCC6 fulfills the role of attracting all the factors needed for transcription-coupled repair, which include the ERCC8–DDB1–ubiquitin ligase complex. Once the repair process is completed, ERCC6 is degraded in a proteasome in an ERCC8-dependent manner through ubiquitination. This degradation seems to be the tail-end process of DNA repair that allows transcription to proceed normally. There is also a syndrome closely related to Cockayne's syndrome known as cerebro-oculo-facio-skeletal syndrome, which involves mutations in ERCC2, ERCC5 and ERCC6. Details of this disorder are provided by Graham and co-workers.[28] Most of the symptoms of this autosomal recessive disease are similar to those of Cockayne's syndrome, except that cerebro-oculo-facio-skeletal syndrome is also associated with more-severe eye defects such as microcornea.

The main symptoms of cancer disposition in xeroderma pigmentosum are seen when mutations are found in genes that are unique to global genome repair and have no role in transcription-coupled repair; for example, XPC and DDB1 (XPE), and replication polymerase η. Considerable evidence is accumulating to indicate that lack of efficient NER leads to neurodegeneration. Mutations in three other genes, ERCC2, ERCC3 and ERCC5, can result in combined symptoms of either xeroderma pigmentosum and trichothiodystrophy or xeroderma pigmentosum and Cockayne's syndrome, depending on the type of mutation carried.[8,25] The symptoms of trichothiodystrophy are developmental abnormalities, with sulfur-deficient brittle hair and skin photosensitivity, growth retardation and neurological abnormalities. Recently, Andressoo et al.[29] reported a knock-in Ercc2 mouse model for combined xeroderma pigmentosum and Cockayne's syndrome that exhibited both cancer and segmental progeria, and the fibroblasts from these animals showed defective repair of oxidative DNA lesions.

In addition to the classical trisomy of chromosome 21, evidence is emerging that people with Down syndrome have defective DNA repair mechanisms, particularly with respect to NER. In an age-matched and sex-matched controlled study, Raji and Rao[30] showed that the activities of enzymes such as polymerase β, polymerase ε and two endo-DNases were adversely affected in the peripheral lymphocytes of patients with Down syndrome compared with normal controls. The capacity for lymphocyte repair also rapidly declined with age in patients with Down syndrome compared with normal subjects. Further evidence has emerged to indicate that there is an inherited defect in DNA repair in Down syndrome, particularly regarding the repair of oxidative damage. For example, unusually high levels of several gene products involved in the repair of oxidative damage, such as Cu–Zn superoxide dismutase (SODC, also known as SOD1),[31] XRCC1, ERCC2, ERCC3, transcription factor TAF–DBP and elongation factor EF1A,[32] have been found in patients with Down syndrome, indicating cellular oxidative stress. Recently, Zana et al.[33] reported finding elevated numbers of DNA single-strand breaks and oxidized bases in the lymphocytes of patients with Down syndrome. Unrepaired oxidative DNA damage in the brain might, therefore, be a causative factor in the mental retardation seen in Down syndrome.

Ataxia-telangiectasia, Nijmegen Breakage Syndrome and Ataxia-like Disorder

Ataxia-telangiectasia, Nijmegen breakage syndrome and ataxia-telangiectasia-like disorder are generally considered to be chromosome instability disorders, and the associated defective genes in these diseases are ATM,[34]NBN—also known as nibrin[35]—and MRE11,[36] respectively. All of these genes are involved in eliciting a response to DNA damage, in particular double-strand breaks that result from ionizing radiation. ATM belongs to the phosphatidylinositol kinase family, is activated by ionizing radiation, and phosphorylates nibrin in the MRN (MRE11–RAD50–nibrin) complex.[37] This complex is linked to DNA damage processing, in particular double-strand-break repair through recombination,[38] but also telomere-length maintenance.[39] It is not clear, however, whether ATM or the MRN complex have any role other than to activate the response to the DNA damage that might be connected to the cerebellar neurodegeneration seen in the above disorders. Lee et al.[40] have proposed that ATM might activate the apoptosis signaling pathway to eliminate neurons with a heavy load of DNA damage. Recently, Biton et al.,[41] working with human neuron-like cells in culture, observed that the ATM-mediated response to double-strand breaks is similar to that in proliferating cells. Knocking out ATM did not interfere with neuronal differentiation, but it abolished the ATM-mediated response to DNA damage. The damage response cascade, which is perhaps not yet completely understood, therefore seems to be important to effect repair, although the 'apoptosis' hypothesis also looks attractive.

Alzheimer's Disease and Parkinson's Disease

Of all the diseases listed in Table 1 , Alzheimer's disease and Parkinson's disease are probably the two that are gaining the most attention at present. Both of these disorders are devastating in nature, have familial and sporadic forms, and are attributable to a vulnerable genetic lineage combined with aging and possibly as-yet-unknown lifestyle factors. A voluminous literature is available on both of these disorders, particularly in relation to their association with aging. Only those observations that indicate a link with DNA repair defects will be mentioned here.

The hallmark symptom of Alzheimer's disease is cognitive decline; therefore, it was only natural that people would investigate, among other things, genomic damage and its repair in the CNS—whenever possible in affected patients, but also in other tissues or experimental models. In 1987, Robinson and co-workers[42] reported that alkylation damage is inefficiently repaired in cells from patients with Alzheimer's disease. They postulated that this could be the cause of late-onset familial Alzheimer's disease and the associated damage to the CNS. Since then, a large body of information has established that there is accumulated oxidative DNA damage in the cells of patients with Alzheimer's disease.[43] Many other studies have shown that there is both increased DNA damage and decreased DNA repair in patients with Alzheimer's disease (for a recent review see Fishel et al.[44]). It is noteworthy that the two DNA repair pathways that are most likely to be adversely affected in Alzheimer's disease are BER[44] and nonhomologous end joining.[45] So far, however, there has not been any unequivocal identification of a precise locus or gene (or genes) that is affected in these pathways.

Oxidative stress and DNA damage are also implicated in Parkinson's disease. Increased levels of oxidative stress as well as expression of the mitochondrial BER enzyme 8-oxoguanosine DNA glycosylase (OGG1) have been reported to occur in the substantia nigra region of the brain in patients with Parkinson's disease.[46] The cell death mechanism in this disorder is, however, proving elusive.

Huntington's Disease, Spinocerebellar Ataxias, Friedreich's Ataxia, and Myotonic Dystrophy Types 1 and 2

Huntington's disease, the spinocerebellar ataxias, Friedreich's ataxia and myotonic dystrophy types 1 and 2 are all hereditary disorders in which there is an expansion of repeat sequences in DNA. Several such neurological disorders are known, but only a few representative syndromes are mentioned here. In Huntington's disease and various spinocerebellar ataxias, the defect takes the form of an expanded CAG repeat sequence. In Huntington's disease, the unstable CAG expansion (more than 40 repeats) occurs within a gene located on chromosome 4 that was originally called IT15, but has since been renamed HD or huntingtin. The CAG expansion results in a polyglutamine (poly-Q) track in the N-terminal region of the huntingtin protein, which causes the abnormal degradation of this protein into small fragments that undergo ubiquitination; these fragments move from the cytosol to the nucleus and aggregate to cause damage and induce apoptosis.[47,48] It is not clear, however, why only some neurons in the corpus striatum and cerebral cortex are susceptible. There are two chromosomal loci—one at 6q23–24 and the other at 18q22—that are capable of modifying the age of onset of Huntington's disease.[49] Furthermore, it is becoming clear that in this disease transcriptional dysregulation via histone acetylation might be one of the factors causing neuronal dysfunction.[50] The same events are likely to take place in all of the other poly-Q disorders, including spinocerebellar ataxias and spinal and bulbar muscular atrophy, although the identity of the gene showing the CAG expansions varies.[51] In Friedreich's ataxia, expansion of the trinucleotide GAA occurs in the first intron of the gene on chromosome 9 that codes for frataxin (FRDA) protein, whereas in myotonic dystrophy either the CAG (type 1) or the CCTG (type 2) expansions are seen in a zinc finger protein 9 (CNBP) gene.[52]

Spinocerebellar Ataxia With Axonal Neuropathy-1 and Triple-A Syndrome

In spinocerebellar ataxia with axonal neuropathy-1 (SCAN1), there is a progressive degeneration of postmitotic neurons. El-Khamisy and co-workers[53] have recently demonstrated that this neurodegenerative disease results from a mutation in the gene encoding tyrosyl DNA phosphodiesterase 1 (TDP1). In lower eukaryotes, TDP1 is known to facilitate double-strand-break repair by removing the topoisomerase I peptide from DNA termini. A mutation in this enzyme produces no distinct phenotype, however, and is therefore unlikely to account for the progressive neuronal degeneration noticed in SCAN1. El-Khamisy and his group[53] have, therefore, looked for a different role for TDP1 in human cells. They found that TDP1 is required for the repair of chromosomal single-strand breaks arising from abortive topoisomerase I activity or oxidative stress. This group has also shown that TDP1 is part of the multiprotein single-strand-break repair complex and directly interacts with DNA ligase IIIα, and that this complex is inactive in SCAN1 cells. These findings are of considerable importance, as they indicate the existence of a TDP1-dependent single-strand-break repair pathway in differentiated neurons that is deficient in SCAN1 patients. Normally, single-strand breaks or gaps are repaired in neurons through BER, in which both DNA ligase IIIα and XRCC1 participate, along with polynucleotide kinase.[1] It therefore seems that the TDP1-dependent single-strand-break repair is a slightly different mode of single-strand-break repair, and could be of considerable importance in brain cells where it deals with single-strand breaks resulting from a variety of causes.

Yet another hereditary disease with neurological symptoms and a defect in the repair of DNA single-strand breaks has been recognized.[54] This disease is called triple-A (achalasia–addisonian–alacrima) syndrome, and it was found to be caused by a mutation in a gene called AAAS (located on chromosome 12q13), which codes for a protein named ALADIN. Triple-A syndrome shows considerable genetic heterogeneity. ALADIN is a component of the nuclear pore complex, and mutant ALADIN fails to target this complex. The consequences of this failure were recently investigated using fibroblasts from patients with triple-A syndrome.[55] Mutant ALADIN was found to decrease the nuclear accumulation of both apratoxin, a repair protein for DNA single-strand breaks, and DNA ligase I; this decrease was reversed by wild-type ALADIN.

Amyotrophic Lateral Sclerosis

The precise mechanism of motor neuron death in amyotrophic lateral sclerosis (ALS) remains elusive. A mutation in the SOD1 gene is known to be present in the familial form,[56] however, which is a definite indicator that oxidative stress could have an important role in the disease. Indeed, cell-permeable antioxidant peptides have been suggested as a potential therapeutic approach.[57] The SOD1 mutation might not be the sole cause of ALS, however, as Sod1 knockout mice have been found to survive for long periods without any motor neuron degeneration. The SOD1 mutation might, therefore, mediate other changes that lead to the death of motor neurons.[47] Further supporting oxidative stress as a major factor in ALS, Kikuchi et al.[58] have shown impaired mitochondrial repair of 8-oxo-guanine in the spinal motor neurons of patients with ALS. Similarly, Nagano et al.[59] observed decreased levels of phosphatidylinositol 3 kinase, its downstream effector Akt/protein kinase B, and AP endonuclease in transgenic mice expressing mutant SOD1. It should be noted that these factors are all involved in the normal DNA damage response and in BER. Furthermore, Oh et al.[60] reported increased expression of apoptosis-inducing factor (AIF) in the spinal cords of SOD1 Gly93Ala transgenic mice as their ALS disease progressed.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: