The group examined mice with mutations in a gene that encodes a DNA repair enzyme called ERCC1 (short for excision repair cross-complementing group 1) and found that the animals developed age-dependent deficits in synaptic plasticity and cognition. “This paper is a nice addition to the body of evidence suggesting a link between DNA damage proteins and brain function,” said Karl Herrup at Rutgers University, who was not involved in the study. Several DNA repair proteins have been previously linked to neuronal function (McKinnon, 2009) and to AD (see ARF related news story on Stante et al., 2009). “What sets this study apart is that they have done electrophysiology and behavioral experiments,” said Peter McKinnon of St. Jude Children’s Research Hospital, Memphis, Tennessee.

A cell’s DNA is constantly under attack from a barrage of agents. Some are external, such as UV light and certain drugs, while others are the products of normal cellular functions, such as free radicals produced during metabolism. There is indirect evidence connecting such damage to the neuronal problems associated with AD. Feeding mice a high-cholesterol diet, for example, increases amyloid-β (Aβ) deposition in a transgenic mouse model of AD (see ARF related news story on Refolo et al., 2000 and Wolozin et al., 2000). “If you feed mice cholesterol and get more free radicals, you will have more DNA damage and accelerate AD in these mice,” said Elgersma. Subsequent studies have also linked cholesterol and oxidative stress to AD (see ARF related news story on Nicholson et al., 2009 and ARF related news story, as well as ARF live discussion from 2002).

To probe the connection between DNA damage and neuronal function, Elgersma and colleagues focused on ERCC1. This enzyme is involved primarily in a type of repair called nucleotide excision, which eliminates helix-distorting DNA lesions such as those caused by UV-induced photoproducts. Although ERCC1 does not play a role in AD, defects in nucleotide-excision repair cause xeroderma pigmentosum, a disorder that increases a person’s likelihood of developing cancer. A subset of patients with this disease also develop neurodegeneration.

Elgersma, Jaarsma, and colleagues generated mice with one allele of ERCC1 knocked out and the other mutated in such a way that the resulting protein contains a seven-amino-acid carboxy-terminal truncation that reduces its function. These mutant mice live until about six months of age with their brain architecture remaining close to intact, although their brains are small. The scientists examined these mice at one and four months. Their basal synaptic transmission was normal at both time points, but long-term potentiation (LTP)—a measure of synaptic plasticity and the main cellular mechanism involved in memory and learning—was significantly reduced in the four-month-old mutant mice compared to control littermates. The brains of the four-month-old mutants, but not the one-month-old ones, showed signs of genotoxic stress and neuronal degeneration, the authors write, such as increased staining for glial fibrillary acidic protein (GFAP) and upregulation of p53, as well as astrocytosis. But since baseline synaptic transmission was normal, the observed LTP deficit was not predicted to be due to neuronal cell death.

Elgersma and colleagues then confirmed these results in mice with neuronal-specific ERCC1 mutations. “With a global mutation maybe the mouse is not healthy and that is what is causing cognitive problems, so we wanted to restrict the mutation to the brain and see what happens,” said Elgersma. Using the Cre-loxP system, a genetic tool for knocking out genes in specific tissues, they generated mutant mice lacking ERCC1 in neurons of the hippocampus and cortex. These mice also exhibited signs of neurodegeneration and a reduction in LTP, but these changes occurred later than in the mice with the global mutation; for example, LTP was reduced in six-month-old mice, but not three-month-old ones. In addition, these mice performed less well than controls in the Morris water maze, which tests hippocampal function. Although the published paper only reports results in mice up to six months of age, the mice with the neuronal-specific mutation have now lived past a year. “The older they get, the more reduced plasticity we find,” said Elgersma.

The authors concluded that faulty DNA repair, which in people can occur as a result of aging or AD, directly affects neuronal function and cognition in the ERCC1-mutant mice. The model they propose is that “if a neuron has a considerable amount of DNA damage, it won’t be able to transcribe all its proteins. We know that LTP and memory formation require new protein synthesis, so LTP and learning will change as a result of DNA damage,” said Elgersma. “At some point, too much damage occurs, essential proteins are no longer made, and eventually the neurons die. The plasticity deficits are preceding massive cell death.”

Much of this model remains to be tested. Elgersma and colleagues will soon be publishing protein expression profiles of neurons from their transgenic mice. Their current study, however, did not formally establish that the effects on cognition are due to increased DNA damage. “They have shown that deficiency of the protein correlates with the neuronal deficits, but they have not shown that deficiency of the protein leads to increased DNA damage in the neurons under study,” said Herrup. Although those experiments are possible to do, measuring DNA damage is difficult, especially in the brain. “Most damage probably occurs by harvesting the tissues,” wrote Elgersma in an e-mail to ARF.

Despite its limitations, the overall study makes a valuable contribution to the field, according to Mark Mattson at the National Institute of Aging in Baltimore, Maryland. “The idea that in aging and in AD you have less effective DNA repair is viable,” he said. “These results are important because they show a direct link between a specific repair enzyme’s activity and cognitive function.”

To what extent this work relates to AD is subject to debate. “One major limitation of the study is that they have not looked at the hallmarks of AD pathology,” said Peter Davies at the Albert Einstein College of Medicine of Yeshiva University in New York City. Elgersma and colleagues did not measure amyloid deposition, one reason being that “mice do not develop amyloid plaques unless they express the human [amyloid precursor protein] APP gene,” explained Elgersma. However, Davies pointed out that wild-type mice do show changes in tau phosphorylation or in mouse APP processing; measuring that would have increased the relevance of the study to human disease. “The top experiment on my list of things to do next would be to cross these mice with APP transgenic mice and see if it more rapidly induces deposits,” said Elgersma.—Laura Bonetta.

Reference:
Borgesius NZ, de Waard MC, van der Pluijm I, Omrani A, Zondag GC, van der Horst GT, Melton DW, Hoeijmakers JH, Jaarsma D, Elgersma Y. Accelerated age-related cognitive decline and neurodegeneration, caused by deficient DNA repair. J Neurosci. 2011 Aug 31;31(35):12543-53. Abstract