Tuesday, August 23, 2005

The vascular hypothesis of Alzheimer's disease

Alzheimer's disease affects about 4 million Americans, and the numbers of cases are expected to rise as the population ages. The exact causes of Alzheimer's disease are unknown, and it is likely that non-heritable cases are triggered by more than one kind of environmental insult. The most prevalent theory has been that accumulation of a protein fragment called "A-beta," which is toxic to neurons, starts a vicious cycle of neuronal death and inflammation.

Although blood circulation problems can also cause dementia, the difference in the sequence of cognitive losses have led neurologists to consider vascular dementia to be distinct from Alzheimer's. However, even from the earliest reports of the disease, it has been known that people who die from Alzheimer's have not only neuronal damage but also pathologies in the blood supply to the brain. A subset of researchers in this very big field are suggesting that perhaps problems with the circulation are the first problem, and that neuronal toxicity comes later. (There is an intermediate position, which could be called the neurovascular hypothesis, which would hold that clearance of toxic A-beta by the vasculature is faulty.) To test whether changes in the blood supply or accumulation and toxicity by A-beta(or both!) are the beginnings of Alzheimer's, it is critical to identify people with the very earliest stages of the disease.

The notion that circulatory disorders can contribute to Alzheimer's-like pathologies got a big boost in a paper to appear in next month's Nature Medicine. Wu et al. report that brain endothelial (blood vessel) cells taken at autopsy from people with Alzheimer's disease are deficient in the vascular transcription factor GAX/MEOX2 and are less able to form tubules in vitro. When expression of this transcription factor was reduced in normal endothelial cells, those cells developed some of the same difficulties as the cells from Alzheimer's victims. Moreover, mice lacking one copy of this gene show reduced ability to grow blood vessels in the brain, and impaired clearance of A-beta.

So the molecule GAX/MEOX2 is reduced in humans with the disease, and reduction in human cells in vitro and in mice in vivo can cause blood vessel abnormalities like those seen in the progression of Alzheimer's disease. There are three limitations to the data at present. The first is that the authors have not yet ruled out that GAX/MEOX2 is also expressed by neurons, and could also function on the neuronal side of the neurovascular unit in mice (analysis of this possibility appears to be in progress by that lab). The second is that there is no mention if the mice show any kind of dementia, or amyloid plaques (accumulations of A-beta which are seen in Alzheimer's disease). This may be because the mice do not live long enough. Lastly, although heriditary Alzheimer's disease is very frequently associated with abnormal generation of A-beta, there has not been, to my knowledge, any human genetic abnormality which only affects the vasculature. (Several proteins related to the generation of A-beta are expressed in neurons and blood vessels.)

My own reaction is that the importance of neuro-vascular interplay cannot be overstated, and the sort of mild global cognitive impairment which seems to represent the earliest stages of Alzheimer's is suggestive of circulatory problems.

More comments on this work are at What's Next and at Science Daily .

No comments: