Identification of a previously uncharacterized genetic disease highlights DNA repair as a shared mechanism in neurodegenerative disorders, and suggests potential therapeutic approaches to tackling them. See Letter p.87
Neurodegenerative disorders are becoming increasingly prevalent worldwide, and are a growing burden on the ageing population. Research has highlighted some common themes, including roles for abnormal and misfolded proteins1. But the cellular mechanisms underlying these commonalities have remained stubbornly complex. On page 87, Hoch et al.2 report the identification of a neurodegenerative disease that involves DNA repair; their result highlights alteration of this process as a potential contributor to many neurodegenerative diseases.
The authors began with a 47-year-old woman, who for the past 19 years had had a progressive neurological syndrome involving an unstable gait and difficulty in coordinating eye and limb movements. These traits are consistent with neurodegeneration in the brain's cerebellum and basal ganglia. Hoch et al. found that the woman had mutations in the gene XRCC1 and decreased levels of the XRCC1 protein that it encodes. They also found that a second, unrelated person with similar symptoms had mutations in XRCC1. They termed this syndrome ataxia oculomotor apraxia XRCC1 (AOA-XRCC1).
XRCC1 is a crucial part of the molecular machinery that repairs single-strand breaks in DNA and DNA bases that have been oxidized by harmful reactive oxygen species (ROS). One of the first steps in the DNA-repair process is the synthesis of chains of poly-ADP ribose (PAR) molecules by the enzyme PAR polymerase (PARP). This initiates the assembly of several scaffolding factors on the damaged DNA, including XRCC1 and the protein ATM, which is mutated in another neurodegenerative disease, ataxia telangiectasia3. These factors, in turn, recruit the enzymes needed to repair the damage (Fig. 1).
In line with XRCC1's role in DNA repair, the authors demonstrated that levels of single-strand-break repair were much lower in the patients' cells than in normal cells. Next, they examined the mechanism by which XRCC1 mutations might cause neurodegeneration, turning to mice in which Xrcc1 in the brain had been genetically deleted. The mice had pronounced cerebellar degeneration and associated changes in movement and coordination; PARP activity was also increased.
What might be the role of PARP in AOA-XRCC1? When Hoch et al. generated mice that lacked both Xrcc1 and Parp, they observed a striking amelioration of the behavioural changes and cerebellar degeneration seen in animals lacking only Xrcc1, suggesting that PARP activity or hyperactivity is a mediator of the disease. The underlying mechanism remains uncertain, but perhaps enzymatic hyperactivity is driven by poor DNA repair due to mutated XRCC1 or low XRCC1 levels. This hyperactivity could exacerbate repair defects by trapping scaffolding complexes on the DNA, further increasing PARP hyperactivation in a feedback loop. The investigators suggest that the development of drugs to inhibit PARP activity might be a way of treating AOA-XRCC1, and possibly other neurodegenerative diseases.
DNA repair has been implicated in several other rare genetic diseases3. So far, all seem to share features of progressive movement disorders, which are typically seen when there is neurodegeneration in the cerebellum or the basal ganglia. Some of these diseases are caused by mutations in enzymes that process the ends of DNA strands to repair single-strand DNA breaks.
Intriguingly, DNA-repair mechanisms have also been implicated in the extensively studied neurodegenerative disorder Huntington's disease, which again affects the basal ganglia, and which is caused by an expanding stretch of glutamine amino-acid residues in the protein huntingtin4. This expansion causes the protein to adopt an abnormal conformation and to aggregate in the brain in a manner similar to that of proteins involved in some of the more prevalent neurodegenerative disorders, including Alzheimer's disease and Parkinson's disease. The age at which Huntington's disease arises is highly variable. Most of this variability is determined by the polyglutamine expansion, with longer repeats causing earlier onset, but a genome-wide analysis5 found several genomic regions in which variation can modulate the disease's age of onset — and these regions seem mostly to contain genes related to DNA repair, among other factors.
Another genetic study6 found evidence that DNA-repair enzymes also modify the age of onset of several other neurodegenerative diseases caused by polyglutamine expansion. In addition, DNA-repair mechanisms have been implicated in the pathogenesis of such polyglutamine diseases7. The detailed mechanisms by which DNA-repair defects might modulate these diseases require further study. Perhaps clustering of mutant polyglutamine-expanded proteins near DNA inhibits the recruitment and release of repair enzymes, interfering with cellular DNA repair.
Relationships between the pathways involved in DNA repair and in polyglutamine-expansion diseases are becoming apparent. Crossing mice that harbour mutated huntingtin with mice lacking one copy of ATM reduced the severity of traits associated with Huntington's disease in offspring8. Moreover, huntingtin forms a complex with ATM during DNA repair9, and the assembling complex recruits XRCC1 and other factors to sites of oxidative DNA damage.
How do DNA-repair defects bring about neurodegeneration? One possibility is that, over time, aberrant DNA repair results in a progressive accumulation of oxidative damage to DNA. This damage causes broad changes, both in DNA sequence and in epigenetic modifications, which alter gene expression without changing the underlying DNA. Over time, these changes could cause a loss of normal neuronal function — and, eventually, DNA-damage levels would cross a threshold, leading factors involved in maintaining genomic integrity, such as the protein p53, to trigger programmed cell death and neuronal loss.
These ideas, along with Hoch and colleagues' results, raise the question of whether DNA-repair mechanisms are relevant to Alzheimer's and Parkinson's diseases. Defective repair of oxidized DNA has been noted in the brains of people with Alzheimer's10. Perhaps here, too, the age of onset of disease is defined by DNA-repair defects combined with an increase in ROS levels that occurs during normal ageing. One caveat is that dementia affects the brain's cortex and hippocampus, rather than the basal ganglia and cerebellum. Nevertheless, there is at least indirect evidence for aberrant DNA-repair mechanisms in Alzheimer's — for instance, in the observation of elevated levels of oxidized DNA bases in patients' cerebrospinal fluid11.
In the future, PARP, ATM, p53 and other repair-related proteins could prove to be therapeutic targets for neurodegenerative disease. Hopes for treating these diseases 25 years ago focused on DNA and genetic studies — we may now have come full circle, back to DNA. Time, and further studies, will be needed to elucidate these mechanisms in more detail, and to determine whether such treatments could be useful in neurodegenerative diseases.
About this article
Nature Reviews Neuroscience (2017)