'Lots of things that once looked separate can now be brought into the same idea,' says Gareth Roberts, head of molecular neuropathology at the pharmaceutical company SmithKline Beecham.
More than half a million people in the UK suffer from Alzheimer's disease, a degenerative disorder in which brain cells slowly disintegrate, leaving the shrunken organ littered with plaques (large sticky clumps of aggregated protein) and tangles (scrambled filaments that once formed the framework of cells).
Since the turn of the century, pathologists have been approaching Alzheimer's by investigating changes in the brains of victims. They have catalogued the consequences but have been unable to find the cause. Clues have come instead from modern genetic studies.
A few years ago, John Hardy, then at St Mary's Hospital, London, studied a family in which all members suffered from the disease at an early age. He found its members shared a mutant gene. This gene normally controls the synthesis of amyloid precursor protein (APP), which is broken down inside brain cells to form a smaller unit, beta-amyloid, the major component of the sticky plaques that clog the brain. Those with a defective gene metabolise APP abnormally.
For the dozen or so families with a mutant gene, the link between APP and Alzheimer's is beyond dispute. For the rest, the APP gene is normal, and the cause of the cerebral deterioration remains unknown. None the less, some scientists believe that the genetic studies point to amyloid as the root of the problem for most, if not all, victims.
'Alzheimer's could be regarded as a disease of amyloid metabolism in the same way that heart disease is a disease of cholesterol metabolism. It is probably one disease with multiple causes,' Dr Hardy says.
As neuroscientists try to understand Alzheimer's by studying APP and how it changes in the disease, Allen Roses, professor of neurology at Duke University, North Carolina, has approached the problem from a different direction.
Last year he identified a second gene that influences the age at which a person develops Alzheimer's. The gene has no mutations and appears to have little to do with amyloid plaques or tangles. It controls the synthesis of a molecule (apolipoprotein E, ApoE) that transports fats in the body and comes in three forms, ApoE2, ApoE3 and ApoE4. Humans have two copies of the gene, one from each parent, which need not be the same.
Professor Roses and his colleagues found a strong link between the type of ApoE and the age at which Alzheimer's is likely to develop. People with two copies of ApoE4 will develop the disease, on average, at age 68; those with one copy each of ApoE3 and ApoE2 will be free of the illness, on average, until they are more than 90.
This link has been confirmed in other laboratories, and the abundance of ApoE4 in a population is related to the incidence of Alzheimer's. ApoE4 is less common in Japanese people, for example, and as a nation they develop Alzheimer's later than Europeans. The protective effect of ApoE2 is highlighted in a paper by Professor Roses and his colleagues in this month's Nature Genetics.
The difference between the three forms of ApoE is minuscule but can greatly affect the age at which mental faculties decline. 'A single amino acid change in a protein can determine two decades of difference in onset of the disease,' Professor Roses says. ApoE seems to be important in the progression and development of Alzheimer's but may not start it. Dr Roberts believes that Alzheimer's develops as a consequence of programmed changes in the brain.
'After the age of 20 we start losing about 1,000 nerve cells a day,' he says. 'By the age of 50, nerve cells automatically start resprouting.' This is the equivalent of a forest losing trees but the remaining ones gaining extra branches to compensate. For brain cells to repair and extend they need supplements of many construction materials. Fats, one of the essential building blocks of cell membranes, are ferried to the needy nerves by ApoE. Unlike other cells in the body, brain cells are not renewed or replaced. A 60-year-old cell suddenly having to increase its metabolic effort is like a senior citizen taking up jogging; the strain can prove too much.
The details of which parts of the machinery give out first have yet to be decided. Dr Hardy's corner believes that molecules such as ApoE and amyloid, which would normally be broken down and cleared away, begin to accumulate and stick to each other or to other structural cell components. Over a long time the slow accumulation of biochemical debris clogs up cell terminals, prevents them functioning and ultimately kills them, leaving only plaques or tangles behind. Professor Roses regards the deposition of amyloid as a consequence rather than a cause. More information will be needed before the picture is complete.
Athena Neurosciences, a diagnostics company in San Francisco, is developing a kit to test which forms of ApoE an individual carries. While two copies of ApoE4 would not be comforting news, the results do not predict who will get the disease but, rather, what a person's susceptibility might be. Scientists anticipate that the information could show how an individual's genetic profile controls the progress of the disease. Since Alzheimer's is regulated by many factors, this kind of test could help to predict potential treatments.
It is still ear1y days in the development of treatments. Current drugs aim only to compensate for missing chemicals rather than alter the course of the disease. But experts believe that now more is known about the biochemistry involved, more effective drugs could be developed - and they are not short of ideas.
'We could try to block production of APP, or prevent it being broken down to beta-amyloid,' Dr Roberts says. 'If we could prevent the interaction between APP and ApoE, that might slow the disease, too. Technically, some of these things might be a challenge, but for something as important as Alzheimer's disease, we have to try.'
Professor Roses is collaborating with Glaxo to set up new test-tube screening procedures, searching for drugs that interfere with the reactions of ApoE2, ApoE3 and APoE4. He believes an effective drug will be available before the end of the century. With most drugs taking at least 10 years from lab to prescription pad, this could be optimistic. 'Some researchers get very upset about giving people false hope, but when a drug company invests the money to start screening we believe there is some hope,' he says.