Saving brain cells from decline

BRAIN scans and a cancer drug point to new ways of managing Alzheimer's disease.

111029 Ellis

KATHRYN Ellis gets better at predicting Alzheimer's disease every day, but there's one problem: it costs a lot of money.

The Melbourne-based neuroscientist is co-leader of the Australian Imaging Biomarkers & Lifestyle study. The team's latest findings show that scanning the brain for specific biomarkers is a powerful early diagnostic tool for Alzheimer's. But the non-invasive positron emission tomography scanning technology needed to measure the brain in action and identify biomarkers is expensive and not widely available.

"It's probably one of the study's biggest challenges," she says." It's definitely an expensive technique to be working with."

Fortunately, Ellis and her colleagues' research should be greatly assisted by the opening of the $225 million Melbourne Brain Centre earlier this month. The centre is the fifth largest brain research facility in the world and promises to make Australia a key global player in investigating Alzheimer's, Ellis says.

"It's meant that Melbourne has become the world hub of this neuroimaging research and is probably one of the top three facilities in the world for these particular scans." Among the AIBL study's aims is finding a reliable diagnostic framework for Alzheimer's before symptoms appear, because by the time the disease is usually diagnosed there's already been extensive and irreversible neurodegeneration.

The AIBL team works with a group of 1112 volunteers, ranging from people in perfect health to those with severe cognitive impairment. They are examined at 18-month intervals to detect risk factors for developing Alzheimer's and possible biomarkers for it.

"It was started back in 2006 and we've now completed the third assessment," says Ellis, adding that in addition to finding presymptomatic biomarkers, the team is also looking for preventive diet and exercise targets.

So far, Ellis says the most reliable diagnostic marker is the level of a molecule called beta amyloid peptide in the brain. Beta amyloid is known for its destructive effects on neural tissue, and thus its role in neurodegenerative symptoms associated with Alzheimer's and dementia, but the stage at which it appears during the development of Alzheimer's has not been clear.

According to Ellis, the group's findings suggest that nearly three-quarters of people who have a significant positive brain scan for beta amyloid will develop Alzheimer's within three years, while only 10 per cent of those whose scans reveal no amyloid have Alzheimer's after three years.

"So there's quite a difference there," she says of the scans of healthy people as well as those with mild cognitive impairment who don't yet have dementia. "That's really the time to be giving drugs and, hopefully, delaying or even reversing the damage."

Effective treatments or cures for Alzheimer's remain elusive. But a team led by physician and cancer geneticist Peter Sicinski, with the Dana-Farber Cancer Institute and the Harvard Medical School in Boston, has discovered a possible means of halting neurodegeneration using, of all things, a protein associated with cancers.

Too much of a form of the so-called cyclin E protein in the body has been linked with poor prognoses in breast cancer, as well as a reduction in the effectiveness of letrozole, a drug used to treat patients with oestrogen-receptor-positive breast cancers. It has also been implicated in the growth of blood cancers and solid tumours.

Sicinski and his colleagues have found cyclin E seems to operate entirely differently in the brain compared with elsewhere in the body. There it plays a pivotal role in the development of neural connections and memories. In fact, when Sicinski and his fellow researchers reduced levels of cyclin E in mice brains, they observed impaired memory function and a reduction in the creation of new neural connections.

Tellingly, Alzheimer's disease has been strongly associated in the past with the overactivation in the brain of what's called the Cdk5 enzyme, a biochemical critical to the development of the nervous system. Animal experiments show that inhibiting the activation of Cdk5 reduces Alzheimer's symptoms.

Sicinski's team has pulled together both areas of research. Specifically, it has shown that cyclin E binds to Cdk5. "It inactivates Cdk5," he explains. "It takes it away from its activator. So, because Cdk5 overactivation has been postulated to play a role in the development of Alzheimer's disease, we're showing that cyclin E is an inhibitor of this molecule."

This means that by modulating cyclin E levels in the brain, it might be possible to ameliorate the symptoms of Alzheimer's.

Before they can do this, though, Sicinski needs to tease out exactly how cyclin E is activated and regulated in the neurons.

"What are that pathways that maintain such high cyclin E levels?" he asks. "We need to know exactly how those levels are controlled, so that we can then interfere with that process.

"I'm speculating on the clinical applications for this but, for example, once we've discovered the mechanism that operates the production of cyclin E in the neurons, we could then theoretically chemically interfere with that protein machinery."

That's just what Ellis is waiting for: drugs able to directly intervene in the process of memory loss and neurodegeneration. And it's what spurs on her work to develop effective, affordable techniques able to detect people likely to develop Alzheimer's before their brains are irreparably damaged. "If we can get to these people before the real damage occurs," she says, "there are so many more things we can do for them."