Common genetic variation of the APOE gene and personality

Alzheimer’s disease (AD) represents one of the most challenging disorders in modern society with high costs for society and tremendous pain for both patients afflicted and their relatives [1]. Patients suffering from AD often show a dramatic decline in cognitive functioning [2] and changes in personality [3]. In a small percentage of AD patients, the disease shows a particularly early age of onset and is monogenetically inherited. In the vast majority of patients, however, AD is a multifactorial disorder being influenced by both nature and nurture [4]. In the context of molecular genetics, the most prominent genetic risk factor for this “sporadic” type of AD is a polymorphism on the apolipoprotein E (APOE) gene located on chromosome 19 with its ϵ4 allele being associated with higher risk for AD [5]. In addition to this, especially carriers of the rare homozygous ϵ4/ϵ4 variant show an earlier onset of AD [6]. A study by Meyer et al. [7] even reported data showing that the here discussed APOE polymorphism is more linked to the onset of AD compared to “whether subjects will develop AD if they survive to late old age.” (p. 322). The investigation of the human brain in the context of the APOE gene and AD revealed an association between the APOE ϵ4 allele and a higher rate of hippocampus volume loss when suffering from AD [8]. Given the prominent role of the hippocampus for memory processes (e. g. [9]) and the reported negative relation between hippocampus volume and memory performance in patients with neurological disorders (e. g. [10]), the ϵ4 allele – hippocampus volume (loss) association constitutes a biological explanation for the decline of memory functions in AD patients. Interestingly, this same study by Schuff et al. [8] observed no influence of the APOE polymorphism on hippocampus volume loss in persons not suffering from dementia or patients with mild cognitive impairment. Noteworthy, the relationship between hippocampus volume and memory ability is still a matter of debate. A meta-analysis of healthy subjects across the life-span supports the “smaller is better” hypothesis at least in children, adolescents and young adults whereas in older adults a great variability was the most striking finding [10]. Apolipoprotein E is a major target in Alzheimer disease, because it is known to modulate the amyloid-β aggregation in the brain being involved in the neurodegeneration [11]. The mentioned APOE ϵ4 allele is associated with a much stronger deposition of amyloid-β into senile plaques, which are often observed in Alzheimer pathology [12].

As mentioned above personality changes can be observed in patients suffering from AD [3]. AD patients can become less extraverted (socially outgoing), more neurotic (anxious, moody), less open (curious) and less conscientious. In this context a longitudinal study of interest investigated potential associations between common genetic variants of the APOE gene and temperament traits such as hyperactivity, motor activity, sociability and mental vitality [13]. Of note, the investigated subjects were very young and the significant associations between the common APOE genetic variants and the investigated traits changed across the follow-up measures. Among others the link between ϵ4 and higher motor activity was only observable in childhood, but not in young adolescence. Interestingly, the ϵ4 variant was only associated with higher sociability when investigated in adolescence (but not in childhood). A newer study [14] investigated the role of the above discussed APOE genetic variation and individual differences in personality measured with the Temperament and Character Inventory in a small sample of young females (N = 135). Here, no significant association turned up. This null association could be explained by the different methods in assessing individual differences in temperament traits^a and differences in the investigated ethnic group. Moreover, only females were included in the Tsai et al. study [14]. Noteworthy is also the small number of participants in this study [14]. It is well known that personality is influenced by a large number of genetic variants with small impact. As a consequence larger sample sizes are needed to carve out potential main effects of genetic variants [15].

Dar Nimrod et al. [16] extended the findings from these first pioneer studies by reporting that elderly humans carrying at least one APOE ϵ4 allele suffered more often from cognitive decline and AD, when being high neurotic or high extraverted. Dar-Nimrod et al. [16] explained the results as follows: Neurotic or high anxious persons tend to have lower hippocampus volumes (see also a review [17]), which could be a result of an elevated activity of the hypothalamic-pituitary-adrenal-axis resulting in glucocorticoid induced brain atrophy. Therefore, lower hippocampus volumes in healthy participants together with a genetic risk for AD could lead to an earlier decline of cognitive functions and higher AD incidence. The Extraversion result was not hypothesized by the authors, but could be explained by the idea that extraverted humans exercise their cognitive functions via social interactions. The older a person gets, the smaller the chances for social interactions are, because large parts of one’s own social network are naturally getting smaller.

When summarizing the findings of the literature, it becomes clear that the APOE gene might play a role in personality, although this influence might be age dependent. The effects of common genetic variation of the APOE gene might change strongly across childhood to adulthood, because personality does not stabilize until the age of 30 [18]. It is also conceivable that stronger effects of APOE would be observed in adulthood, as heritability estimates for a range of traits increase from childhood to adulthood [19].

Given the large interest in finding early markers for AD, we investigated in the present study if the common genotypes of ϵ3/ϵ3, ϵ2/ϵ3 and ϵ3/ϵ4 of the APOE gene explain individual differences in personality. Given the small literature in the field, we investigated a much larger and mixed sex sample of young adults compared to the study by Tsai et al. [14] utilizing the same personality questionnaire as in the study by Dar-Nimrod et al. [16].

The present study investigated a potential influence of common genetic variation of the APOE gene on human personality. This is of importance, because the ϵ4 allele of this gene is associated with increased risk for developing AD, and personality has been suggested to moderate the association between the ϵ4 allele and AD [16]. Adding to this, two further studies searched for main effects of common genetic variation of the APOE gene on personality with mixed results [13, 14].

However, no significant effect of the investigated APOE polymorphism on personality could be detected in the present research, therefore making it unlikely that this polymorphism plays an important role for human personality in young adulthood. If detected, a positive association between APOE variants and personality would have been highly interesting from a diagnostic point of view to predict AD. Of interest, a personal communication via email between Ilan Dar-Nimrod and the principal investigator of this study yielded the additional information that no main effect of the APOE polymorphism could be observed on the personality dimensions in their sample consisting of N = 602 elderly humans [16]. This underlines the null findings of the present research and the findings by Tsai et al. [14].

Despite this we cannot rule out that polymorphisms on the APOE gene, together with other genetic variations or certain environmental influences, may shape human personality. Studies investigating such gene by environment effects or epigenetics in the context of personality will be of highest interest. Of note, the APOE ϵ4 allele is also associated with cardiovascular disease, whereas “…higher intake of fat in the Western diet may be partly responsible for the increased risk of AD associated with APOE ϵ4…” [22]; p. 729. As individual differences in both exercise behavior [23] and dietary behavior [24] have been also associated with personality, these variables might constitute interesting further factors interacting with APOE ϵ4 in the context of the present research topic.

Some more limitations of the present study need to be mentioned. First, we applied just one (although very prominent) measure of human personality stemming from a lexical approach [20]. More biologically oriented personality inventories such as the Affective Personality Neurosciences Scales (ANPS) could have yielded other results [25]. Second, our results are preliminary, because the highest risk constellation for AD – namely the homozygous ϵ4/ϵ4 genotype - occurs very rarely and could not be properly investigated in the current sample. Clearly huge sample sizes are needed to obtain a sufficient number of ϵ4/ϵ4 carriers to reveal an influence of this special genotype on human personality. Future studies should investigate the interplay of APOE polymorphisms, personality, hippocampus volume and memory functions in a longitudinal design using multiple age cohorts.

In sum, the present study provides no evidence for a role of the prominent APOE polymorphism on human personality in young adulthood.

^aKeltikangas-Järvinen et al. included mothers’ assessments of their children and in parts self-report data

^bThe values of the personality scales deviated from a normal distribution (tested with a Kolmogorow Smirnow test, Shapiro Wilk tests revealed the same results with the exception for Openness). Spearman’s Rho would have provided a non-significant result, here (Spearman’s rho = -.04, p = .38). Of importance for the following analyses in the result section: results did not differ when non-parametric tests were conducted. Therefore, and after visual inspection of the data (resembling normal distributed data), we decided to provide only data with parametric testing.