There are at least four well-confirmed genes responsible for Alzheimer's disease (AD), the most common form of dementia. In addition, many reports indicate an association between the disease and genetic variations in different gene candidates. The complexity and interpretation of these studies are discussed using, as an example, the recent discovery of the association between AD and the SORL1 gene. The knowledge obtained from AD genetics is applicable to many other forms of dementia, which are also genetically complex disorders and are almost all associated with the deposition of different aberrant proteins in the brain.

Introduction

The brain accumulation of a neurotoxic proteolytic derivative of the amyloid precursor protein (APP) (A-beta40/42 peptides) is the essential event in the pathogenesis of Alzheimer's disease (AD) leading to neuronal loss.^[1] Alzheimer's disease brain pathology is also characterized by the formation of intraneuronal tau-associated neurofibrillary tangles, which cause additional neurotoxic insult since mutations in the MAPT gene encoding the tau protein are responsible for frontotemporal dementia.^[2] The accumulation of neurofibrillary tangles is a feature of several neurodegenerative diseases. In some cases this accumulation is a primary disease-causing event (e.g., frontotemporal dementia). In other diseases (e.g., AD) tau accumulation is secondary to the accumulation of A-beta peptides.

The neurofibrillary tangles together with the longer and more toxic A-beta42 peptide appears to be elevated in the brains of individuals affected with either sporadic or familial AD, indicating a common pathogenetic mechanism for different forms of AD.

Approximately 5-10% of patients develop an early age-at-onset AD (before 65 years). The disease in up to 50% of such cases is explained by mutations in one of three genes: APP,^[3] presenilin 1 (PS1),^[4] and presenilin 2 (PS2)^[5] (Figure 1). Pathological mutations in these genes are responsible for an autosomal dominant trait and cause A-beta accumulation in the brain.^[1] However, the pathological consequence of some mutations detected in small AD families is uncertain and needs further investigation. Recently, we proposed to use a systematic algorithm to classify mutations in known AD genes as possibly, probably, or definitely pathogenic.^[6]

The genetics underlying the common late-onset form of AD is complex. The apolipoprotein E gene (APOE) is a major and well-replicated risk factor for both sporadic and familial late-onset AD.^[7] However, as discussed below, inherited variations in other genes could also be responsible for AD.