The human APOE gene is located on chromosome 19 (19q13.2) and encodes the protein apolipoprotein E (ApoE).(1) There are 3 known APOE variants, epsilon2, epsilon3 & epsilon4 with single amino acid substitutions occurring at amino acid residue 112 (arginine for cysteine in epsilon4) or 158 (cysteine for arginine in epsilon3).(2) APOE is mainly expressed in the liver, with the brain having the second highest concentration.(2) ApoE is involved in lipid transport around the body, is a constituent of VLDL & HDL particles and of intestinally formed chylomicrons.(2) It also modifies binding of lipoprotein particles containing ApoE to LDL receptors(2) and in the brain is thought to be involved in lipid recycling, neuronal repair and maintenance, with E4 being the least efficient isoform.(3, 4)

Possession of an e4 allele is a strong genetic risk factor for the development of Alzheimer's disease (AD), although the mechanisms remain unclear.(5, 6) Cumulative inefficiences of repair and maintenance may contribute to the increased risk in e4 carriers. Early neuropathological changes of AD including deposition of amyloid-beta (Abeta), FDG PET abnormalities and cognitive changes have been found even in early middle age in e4 carriers, decades before AD is clinically evident.(7-9) Possession of e4 (without diagnosed dementia) is also linked to higher levels of Abeta and neurofibrillary tangles at autopsy,(10, 11) accelerated volume loss,(12) reduced synaptic protein concentration(13) and decreased PET glucose metabolism in areas affected in AD.

A meta-analysis identified that e4 carriage (in the absence of dementia) was associated with significantly poorer global cognitive functioning, episodic memory and executive functioning;(14) however, most of the studies were small, some used insensitive tests, several cognitive domains were less studied, information on confounding variables was frequently lacking and most examined the over-60s only. From the small number of studies of younger age groups there is evidence that e4 may be beneficial for early cognitive function although one study reported that young adult e4 carriers had a thinner entorrhinal cortex.(15-17)

In addition to the link with AD, APOE genotype has been shown to influence LDL-C levels and cardiovascular disease risk with e4 homozygotes having the highest risk and e2 homozygotes the lowest.(18, 19) A non-linear relationship between APOE genotype and stroke was recently reported in a meta-analysis to which members of our group contributed.(20)

Recently the hypothesis that mitochondrial dysfunction precedes the development of AD neuropathology has again gained support.(24) Roses et al recently demonstrated that a variable length polymorphism in TOMM40, a mitochondrial protein essential for protein import, "predicts the age of onset of late-onset Alzheimer's disease".(23, 25) TOMM40 is very close to the APOE gene on Chr19 and they are in linkage disequilibrium. Roses et al proposed that the APOE/AD link may actually be due to a haplotype including a long poly-T repeat in TOMM40.(23) Improving our understanding of the early neurobiological effects of APOE and TOMM40 may help in predicting AD and to enhance the effectiveness of lifestyle-based and targeted pharmacotherapeutic interventions.

Most of our current knowledge regarding APOE and cognition relates to older adults and the end-stage pathology seen in AD. It is currently unclear when the e4-related changes in brain structure and function begin and which changes occur first. There is a dearth of information on e4 effects in earlier adult life, although the existing evidence suggests that there are differences between e4 and non e4 carriers. Even less is known about TOMM40 and pre-dementia changes.

Many other studies in this area have been criticised for using insensitive neuropsychological tests or not examining specific cognitive areas. I intend to carry out a battery of validated, sensitive neuropsychological tests to avoid this problem. Neuropsychological testing using genotype based recall from ALSPAC gives the potential to have much greater numbers ofe4 homozygotes, thus improving statistical power. Few other studies have included such young participants, had access to such a high number of individuals with rare genotypes or sufficient information on potential confounders, e.g. cardiovascular variables, whereas the use of ALSPAC data would permit the inclusion a wealth of information on confounders in addition to greater statistical power.