Alzheimer Genes: One Step Forward for the Nucleus, One Step Back for Mitochondria


Sources: De Strooper B, et al. Nature 1998;391:339-340; Holcomb L, et al. Nature Med 1998;4:97-100; Hirano M, et al. PNAS 1997;94:14894-14899; Wallace DC, et al. PNAS 1997;94:14900-14905.

New studies suggest that the presenilin 1 (PS1) gene may provide a potential target for Alzheimer's disease (AD) therapy, while a reported link between AD and mitochondrial DNA mutations has now been cast into doubt.

Mutations of the PS1 gene on chromosome 14 are carried by a significant portion of those afflicted with early onset familial AD (EOFAD). Transgenic mice expressing the human PS1 gene do not develop Alzheimer pathology, although they do express increased levels of the 42-43 amino acid form of beta amyloid. In contrast, transgenic mice expressing the human amyloid precursor protein (APP) gene develop amyloid deposits in multiple brain regions. Such deposits usually take 6-12 months to develop. Holcomb and colleagues crossbred PS1 and APP transgenics and found that the doubly transgenic animals had significantly more fibrillar amyloid deposition and developed this pathology significantly earlier than mice with only one of the two mutations. They suggest that their results indicate that a modest increase in the amyloidogenic 42-43 form of beta amyloid by PS1 mutations can profoundly enhance pathogenicity when combined with another factor that influences amyloid production, such as the APP mutation. Both singly and doubly transgenic animals showed behavioral alterations even before beta amyloid deposits became evident. These findings raise the possibility that the PS-1 and APP mutations may have other effects on cerebral function in addition to promotion of amyloid deposition.

The function of the PS1 gene has been difficult to study because animals bred without the PS-1 locus die in utero. De Strooper and coworkers recently studied neuronal cells taken from PS1 knockout embryos before they died and found that they produced approximately five times less amyloid beta protein than wild type cells in response to human APP introduced via a viral vector. The authors suggest that the production of beta amyloid in the brain might therefore be reduced by agents that inhibit PS-1 function.

Less than a year ago, Davis and associates (PNAS 1997;94:4526-4531) made the surprising report that mutated forms of mitochondrial cytochrome C oxidase genes could be found in normal adults but were present at an unprecedentedly high level in patients with AD. This finding generated considerable interest, since the documentation of increased mitochondrial DNA mutations in AD would significantly alter current understanding of the pathogenesis and transmission of this disease. Independent studies by two groups, including authorities in mitochondrial DNA research, now suggest that the finding was an artifact resulting from the use of an idiosyncratic DNA isolation technique. Hirano et al and Wallace et al demonstrate that the novel boiling technique used by Davis et al for extraction of genetic material produces authentic mitochondrial DNA mixed with similar sequences of pseudogenes derived from nuclear DNA that were mistakenly identified as mutated mitochondrial DNA. While these studies did not convincingly explain why the proportion of these nuclear pseudogenes recovered from AD patients exceeded that of normals, it was suggested that AD patients might have nuclei that are more easily disrupted by the boiling technique and, therefore, released more of the pseudogenes than was the case in normals. Whatever the explanation, these pseudogenes are not transcribed into mRNA, and, as such, they are unlikely to be of functional significance in AD pathogenesis.


A number of potential targets for AD pharmacotherapy have emerged from the discovery of the three known disease-causing genes for EOFAD, namely APP, PS1, and PS2. Increasing evidence suggests that these autosomal dominant mutations promote amyloidogenesis. Several pharmaceutical companies are developing agents intended to prevent or retard senile plaque formation by reducing beta amyloid production or amyloid fibrillogenesis. Inhibition of PS1 may be a logical strategy to pursue in treating and/or preventing AD, particularly in the tens of thousands of individuals worldwide carrying PS1 mutations. -nrr