Cortisol, a glucocorticoid, is released following stressful events, and is essential for human survival and for the ability of individuals to cope with stress. Stress response and the cortisol release are regulated via the hypothalamic-pituitary-adrenal (HPA) axis, which involves the hypothalamus, the anterior pituitary gland and the adrenal cortex (see Pomara et al., 2003
; Roozendaal, 2002
; Wingenfield & Wolf, 2010
, for recent reviews).
Stress and cortisol release have also been found to influence memory and they have been associated in particular with poorer memory performance (Wolf, 2009
). The neuronal substrate that connects HPA axis activity and memory performance is the hippocampus, which is considered critical for both episodic and spatial memory. The hippocampus has a high density of glucocorticoid receptors and has been implicated in the regulation of the HPA axis through negative feedback (de Kloet, 2003
). Importantly, the hippocampus is also susceptible to structural damage if abnormally elevated release of glucocorticoids extends over time due to chronic stress or other factors (Lupien et al. 1998
; Sapolsky, 2000
). In turn, damage may provoke a reduction in the density of glucocorticoid receptors, or impaired function, which could result in less feedback inhibition and increased glucocorticoids release, thus propagating a vicious cycle that may lead to additional insult to the hippocampus.
Early decline of hippocampal function and a distinct loss of episodic memory are also among warning signs of incipient Alzheimer’s disease (AD) (Convit et al. 1993
; Jack et al. 1997
; Killianny et al. 2000
). Some evidence also suggests an association between higher brain concentrations of glucocorticoids, HPA axis dysfunction and AD hippocampal pathology (see also Pomara et al. 2003
, for a review). For instance, much higher levels of cortisol have been found in ventricular cerebrospinal fluid (CSF), post mortem, in pre-senile (under 65 years of age) AD subjects compared to age-matched controls (Swaab et al. 1994
). Further evidence of an association between HPA axis abnormalities and AD comes from taking into consideration the APOE genotype. The apoliprotein E (apoE) has been found to regulate the synthesis of glucocorticoids (Poirier et al. 1995
). Critically, the ε4 allele of the APOE gene, i.e., the gene which encodes the apoliprotein E, is considered the strongest risk factor for the development of late-onset AD (Blennow et al. 1996
). Importantly, healthy individuals carrying the ε4 allele have been found to show structural and functional abnormalities of the hippocampus, consistent with preclinical brain changes related to AD (e.g., Crivello et al. 2010
; Donix et al. 2010
; Lu et al. 2011
; Nierenberg et al. 2005
A few recent studies have examined the relationship between APOE genotype, salivary cortisol concentrations and cognitive performance in elderly individuals, with differing outcomes (Beluche et al. 2010
; Gerritsen et al. in press
; Lee et al. 2008
), but, to the best of our knowledge, only two studies so far have examined whether different forms of APOE are associated with varying degrees of cortisol concentrations (Fiocco et al. 2008
; Peskind et al. 2001
). Peskind et al. (2001)
investigated CSF cortisol concentrations, which are thought to reflect brain levels more closely than plasma or salivary concentrations, in 64 AD sufferers and 34 healthy, nondemented, controls. They observed that CSF cortisol levels were higher in AD subjects than in controls (cf., Swaab et al. 1994
) and, critically, when combining the two groups, they found that the APOE ε4/ε4 genotype was associated with higher CSF cortisol levels than the APOE ε3/ε4 genotype, which, in turn, was associated with higher CSF cortisol than the APOE ε3/ε3 genotype. The lowest cortisol level was observed in the APOE ε2/ε3 group, although this group did not significantly differ from the APOE ε3/ε3 group. This general pattern was stronger in AD sufferers than in controls. Fiocco et al. (2008)
, however, failed to detect an APOE effect on peripheral cortisol in a longitudinal study, although they observed that ε4 carriers (n=13) presented higher cortisol levels than non-ε4 carriers (n=50) at the initial visit, in one of two cohorts. These results suggest that brain cortisol levels, as reflected by CSF concentrations, may be affected by APOE genotype, whereas peripheral cortisol levels may not be.
Recently, a variable length deoxythymidine homopolymer (poly-T), rs10542523, in the TOMM40 gene, which is in linkage disequilibrium with APOE, has also been reported to modulate risk and onset age of late-onset AD (Caselli et al. 2010
; Roses et al. 2009
). Roses and colleagues (see also Lutz et al. 2010
) have shown that APOE ε4 alleles are nearly exclusively linked to TOMM40 poly-T variants between 21 and 30 T residues in length (long
variants; i.e., L), whereas APOE ε3 alleles may be linked to either short
variants (20 or lower T residues in length; i.e., S) or very long
variants (31 or over T residues in length; i.e., VL). In individuals carrying the ε3 allele, VL poly-T variants were found to associate with earlier late-onset AD onset age, including in APOE ε3/ε3 carriers, a genotype previously considered to confer neutral risk for late-onset AD. In contrast, S poly-T variants are found to be associated with later late-onset AD onset age in ε3/ε4 and ε3/ε3 individuals and are considered to result in a protective effect.
Given the recently reported association between APOE and TOMM40 in influencing risk and age of onset of late-onset AD, we raised the question as to whether TOMM40 poly-T variants might mediate the effects of APOE alleles on CSF cortisol concentrations, as initially reported by Peskind et al. (2001)
. We hypothesized that shorter TOMM40 poly-T variants, as for onset age of AD, should be protective against the deleterious effects of APOE ε4 and, thus, might moderate the increase in CSF cortisol associated with this allele. Roses and colleagues (2009)
have suggested a number of mechanisms by which different TOMM40 poly-T variants may affect risk for AD, including interference with APOE transcription, but none of these mechanisms have yet been established. If our hypothesis is correct, it would be possible to advance an HPA axis-related mechanism to explain, at least in part, how TOMM40 variants influence AD risk.