Introduction

It's well known that Alzheimer disease runs in families, with estimates of the prevalence of the inherited forms ranging from as little as 5-10% to as high as 50%. It's accepted that the presence of polymorphic alleles of apolipoprotein E predisposes to Alzheimer disease in the 60s and 70s, although geographic differences in prevalence indicate that genetic influences are readily affected, even over-ridden, by environmental factors. What's interesting is that the phenotypes of familial and apparently 'sporadic' cases of Alzheimer disease cannot usually be distinguished - apart from the early-onset autosomal dominant forms.

Early-onset Alzheimer disease

While the first specific genetic cause of Alzheimer disease to be recognized was mutation in the amyloid ß-protein precursor (APP) gene on chromosome 21, this is confined to about 25 families worldwide. APP mutations are generally associated with early-onset Alzheimer disease i.e. before the age of 65. They cause increased production of all Aß peptides or Aß42 peptides.

Down's syndrome, which affects chromosome 21, is associated with over-expression of structurally normal APP, which presumably accounts for the early appearance of many diffuse plaques, microgliosis, and neurofibrillary tangles in these individuals; the plaques consist of Aß42 peptides initially, with accrual of Aß40 in the subjects' late 20s and 30s.

The presenilin gene PS1 located on chromosome 14 has shown mutations in early-onset Alzheimer disease patients from some families. Another presenilin gene, PS2, is found on chromosome 1, and mutations have been associated with symptomatic Alzheimer disease in subjects in their 50s. Mutations in both cause increased production of Aß42 peptides.

Early-onset disease represents only a small fraction of all Alzheimer disease. It is therefore not practical to offer genetic testing using the APP or presenilin genes, except for members of suspected early-onset familial disease.

Late-onset Alzheimer disease

The apolipoprotein E gene on chromosome 19 has three alleles - E2, E3, and E4. The E4 allele is over-represented in Alzheimer disease patients. While the allele is present in 20-30% of the general population, it is found in 45-60% of Alzheimer patients. Homozygotes for this allele occur in 2-3% of the general population, but in 12-15% of Alzheimer patients.

APOE4 has been shown to bind immobilized Aß peptides in vitro, and may therefore cause increased density of Aß plaques.

APOE4 acts primarily as a modifier of age at onset in otherwise susceptible individuals. Its peak effect occurs in patients in their 60s, rather than in their 70s and 80s, when the disease is more common. Its presence, therefore, is a genetic risk factor for late-onset Alzheimer disease.

Other genetic alterations

In recent years, efforts have intensified to find other genes in which mutations might prove causative for Alzheimer disease. Over 40 candidate genes have been tested, as of April 2000¹. So far, none has been established as a risk factor. Linkage studies indicate that chromosome 12 might carry possible sites, with both alpha2-macroglobulin (A2M) and low-density lipoprotein receptor-related gene (LRP) as possible candidate genes. For neither, however, have polymorphisms been clearly associated with Alzheimer disease.

Future directions

The APP and presenilin mutations all result in increased Aß 42 production, albeit by different mechanisms. APOE4, on the other hand, appears to enhance the steady-state level of Aß40 among the Aß peptides; moreover, in mice, APOE4 is less supportive of normal neuronal form and function than APOE3 or E2.

Knowledge of the close relationship between presenilins and y-secretase, using Aß-secreting cells, has led to the screening of compounds for y-secretase-inhibitor activity, and at least one of these has been selected for clinical trials in man. It seems possible that there may be other substrates (e.g. Notch²) that require presenilins for proteolysis, some of which may also be involved in diseases like Alzheimer's.

While genetic aberrations can act as risk factors for the development of Alzheimer disease, additional steps may be required to permit actual progress of the condition to manifest disease. One of these under consideration today is the inflammatory component.

The author of the review summarized here postulates the following 'cascade' of events for familial forms of Alzheimer disease:

1. Mutations in APP, PS1 and PS2 genes
2. Altered proteolysis of APP
3. Increased production of Aß42
4. Accumulation of Aß42 in brain interstitial fluid
5. Deposition of Aß42 as diffuse plaques
6. Aggregation of Aß40 onto Aß42
7. Inflammatory response - microgliosis, astrocytosis, mediator-release
8. Progressive neuritic injury
9. Oxidative injury - free radical accumulation
10. Widespread neuronal dysfunction and cell death, transmitter deficiencies
11. Dementia

Most of these steps provide ample opportunity for research directed at trying to halt the cascade, and provide a suitable therapy or preventive measure for those at risk of this distressing disease.

Implications

Dr Selkoe, the author of this review, has speculated that, in the future, people entering their 50s may have an Alzheimer disease risk-assessment, including a family history, structural and functional brain imaging to detect presymptomatic lesions, and measurement of markers in cerebrospinal fluid and blood. Those at high risk would then be offered preventative therapy. The recent rate of scientific progress in this field entitles us to think this approach will be realized sooner rather than later.