Department of Internal Medicine and Medical Therapy, Chair of Geriatrics, University of Pavia, Piazza Borromeo 2, 271000 Pavia, Italy
Abstract: Dementia is a relatively well-defined condition characterized by a progressive decline of cognitive and performances, as a consequence of degenerative and/or vascular brain changes.
Although the definition of stress remains still problematic, it is now well known that a chronic exposure to stressors is usually able to disrupt the physiological balance both at the cellular or the organism level, and to play a role in the onset and progression of some pathological conditions.
Within this context, at systemic level stress includes all the neurohormonal and metabolic responses of the organism to external stressors; at cellular level, stress, mostly oxidative stress, may instead be a correlate of the aging process itself. The link between stress and cognitive impairment is probably to be found in the hippocampal changes, a crucial as well as vulnerable brain area involved in mood, cognitive and behavioural control, and in the mean time, a site with a very high density of glucocorticoid (GR) and mineralocorticoid (MR) receptors. Therefore, the hippocampal neuronal impairment is responsible for a continuous stress-induced activation of the hypothalamic-pituitary-adrenal (HPA) axis and an increased hypothalamic expression of corticotropin releasing factor (CRF) and vasopressin. Furthermore, the age-related changes of the adrenocortical secretory pattern could play a role in the pathophysiology of brain aging, fostering the brain exposure to a neurotoxic hormonal pattern.
In this chapter, we examine particularly the evidence for a link between dementia and HPA activity on the basis of the data in the literature as well as of our personal findings.
Likewise to other age-related diseases, dementia is a clinical condition characterized by increasing prevalence and incidence, due to both the improved diagnostic tools and the significant increase of life expectancy.
Indeed, due to its increased prevalence with age, especially till 80-85 years, senile dementia may be defined as an age-related disease, but it seems not to be caused by the aging process itself, being not an inevitable feature of aging (Ritchie and Kildea, 1995).
Among the different types of dementia, the most common is certainly Alzheimer's disease (AD),
»Corresponding author. Tel.: +39 382 27769; Fax: +39 38228827; E-mail: ferrari(a,unipv.it amounting for 50-80% of all causes (Lobo et al, 1995), followed by vascular dementia (10-24%, Ott et al, 1995), and by other degenerative dementias, such as the Lewy bodies disease and the fronto-temporal dementia, whose real prevalence is not easy to establish, due to the lack of sufficiently strict diagnostic criteria and the small number of specific studies.
In the early stage of senile dementia, the clinical picture may be often confused with the physiological age-related mnesic changes, but with the progression of the disease, the cognitive impairment becomes more and more evident and severe, and indeed patients show a decline of both recent and remote memory, together with behavioural changes and functional disabilities.
The clinical progressiveness of senile dementia makes it sometimes difficult to distinguish between physiological and pathological brain aging; therefore for some authors physiological brain aging and senile dementia might be considered as a continuum (Drachman, 1994), while others underline the differences between the two conditions, considering the aging only as a risk factor for the occurrence of cognitive impairments (Khachaturian, 2000). Anyway, between the clinical features of physiological and pathological brain aging there is often a grey area, within which it is difficult to set out the boundary markers (Mecocci et al, 2002).
Both for physiological and pathological aging, the cognitive impairment keeps up with significant morphological and metabolic changes of the brain and particularly of some regions such as the hippocampus, a crucial brain area involved in learning and memory control, particularly exposed to the effects of stress hormones, due to its high concentration of corticosteroid receptors.
The hippocampal formation seems to be always involved in Alzheimer's disease, which therefore has been defined as "hippocampal dementia" (Ball et al, 1985). However, recent data suggest the presence of alterations of the hippocampus also in vascular dementia, together with those of the white matter and the basal ganglia, in the widespread small ischemic or vascular lesions (Jellinger, 2002).
An age-related neuronal loss of the hippocampal formation may be observed in physiological aging and it is related to both mnesic deficits and impaired performances in verbal test (Golomb et al, 1993). However the neuronal impairment of the hippocampus becomes particularly evident in AD, especially affecting the CA1 pyramidal region. Indeed, according to a recent study (Gosche et al, 2002) the hippocampal volume measured by post-mortem magnetic resonance imaging (MRI) scans predicted the neuropathological criteria for AD according to the Braak's stage (Braak and Braak, 1991), which ia a histopathological index of severity based upon density and distribution of senile plaques and neurofibrillary tangles. Furthermore, the MRI hippocampal measures, particularly if pertinent to the left hippocampus, seem very sensitive in discriminating subjects with normal cognitive performances from those with mild cognitive impairment or with questionable dementia (Wolf et al, 2001). These findings may be clinically relevant, since the volumetric measures of the hippocampus could detect the earliest pathologic features of AD (Braak and Braak, 1997), a pathological condition well known for the long preclinical stage (Berg et al, 1992; Morris et al, 1993)
Similar degrees of CA1 pyramidal neuron loss and more generally of hippocampal atrophy have been described both in AD and in vascular dementia (Kril et al, 2002), even though the same changes probably result from different pathogenetic pathways, namely the abnormal amyloid protein deposition in AD and the microvascular damages in vascular dementia.
It seems important to remember that the adult brain retains a significant ability to remodel synaptic terminal region in terms of synaptic surface and numerical density, and of average area of synaptic contact zones (Bertoni-Freddari et al, 1996). Likewise to experimental data (DeToledo-Morrell et al, 1988; Fatioretti et al, 1992; Bertoni-Freddari et al, 1996), both the synaptic surface and the synaptic numerical density are reduced in human physiological aging and even more in senile dementia (Bertoni-Freddari et al, 1986; Bertoni-Freddari et al, 1990), with particular evidence for the hippocampal region (Scheff et al, 1991). Indeed a significant relationship between low synaptic numerical density and cognitive impairment has been found in AD (Dekosky and Scheff, 1990).
Together with the synaptic density, the synaptic size also may be relevant, since it has been suggested that a larger contact zone may release a greater amount of neurotransmitters, activate more postsynaptic receptors, finally improving neurotransmission. Indeed, due to an increase of a sub-population of larger synaptic contacts in hippocampal and cerebellum areas in aging (Bertoni-Freddari et al, 1996), the average synaptic area of contact may be higher both in aged animals and humans (Bertoni-Freddari et al, 1990; Dekosky and Scheff, 1990). This evidence could represent a physiological compensatory reaction (Bertoni-Freddari et al, 1990, Hillman and Chen, 1981) or, on the contrary, could be a marker of synaptic degeneration (Fattoretti et al, 1992).
In spite of their remarkable plasticity, the hippocampal neurons are particularly vulnerable to stress and to stress-related hormones, such as glucocorticoids.
Although in some rat strains basal corticosterone and ACTH concentrations did not change during life (Workel et al, 2001), in many other experimental animals the basal levels of glucocorticoids tend to increase with age, as a consequence of either an increased central activity and/or of a reduced sensitivity of the HPA-axis towards the steroid feedback (Sapolsky, 1992). In these animal models the extent of glucocorticoid hypersecretion is correlated with the severity of the degenerative changes of the hippo-campal neurons, which become also more vulnerable to metabolic and vascular challenges (Landfield et al, 1978; Sapolsky et al, 1985).
Further evidences about the detrimental effects of high Cortisol levels on hippocampal area derive from the observation that an intravenous bolus of 35 mg of hydrocortisone in elderly subjects results in a 12-16% reduction of hippocampal glucose metabolism, as measured by positron emission tomographic scans (McEwen et al, 1998).
A central action of hormones was suggested in the early 1950s, with the observation of cognitive impairment in subjects treated for a long time with corticosteroids and ACTH (Clark et al, 1952). About 10 years after this clinical observation, McEwen et al, (1968) showed a selective retention of corticosterone in the rodent brain and, especially at the level of limbic area. Furthermore, a high density of corticosteroid receptors was found in the hippocampus (Scoville and Milner, 1957).
Among hippocampal cells, the CA3 pyramidal ones seem to be the most vulnerable to glucocorticoid exposure, as well as to chronic stress and to aging processes (Uno et al, 1990).
Since the hippocampus plays an inhibitory role on the HPA activity and especially in its resiliency after stress activation, further experimental and clinical studies led to the "glucocorticoid cascade" hypothesis (Sapolsky et al, 1986). According to this hypothesis a chronic exposure to glucocorticoids throughout life, secondary to repeated stress, could downregulate the central glucocorticoid receptors especially at hippocampal level, with the consequent impairment of HPA sensitivity to the negative steroid feedback, fostering a neurotoxic steroidal milieu responsible for degenerative changes and neuronal loss.
Many although not all, reports deal with the increase of HPA activity after the removal of the hippocampal formation (Herman et al, 1992).
When considering these original findings, one could think of a functional link between hippocampus and adrenal steroids as a one-way relation, in which glucocorticoids negatively act at hippocampal level. Further evidence suggested that high glucocorticoid levels may not only be the cause, but also the consequence of hippocampal damage (Sapolsky et al, 1990, Lupien and Lepage, 2001). Indeed, Starkman et al, (1999) recently reported that the therapeutic reduction of Cortisol levels in Cushing's patients is associated with an increase of the hippocampal volume, so demonstrating the possible reversibility of the hippocampal atrophy occurring in Cushing syndrome as revealed by neuroimaging techniques such as MRI.
The adrenal steroids play a crucial role in modulating hippocampal plasticity, since they could biphasically modulate hippocampal excitability, long-term potentiation and depression (Kerr et al, 1994; McEwen, 2001), and consequently memory and learning; furthermore high levels of adrenal steroids, along with excitatory amino acid neurotransmission, play an inhibitory role on the neurogenesis of adult dentate gyrus (McEwen, 1999), possibly involved in fear-related learning and memory (McEwen, 2001). Finally, in experimental animals, the remodelling of dendrites in the hippocampal CA3 region is modulated by glucocorticoids and by excitatory amino acids, with shortening and debranching of the apical dendrites of CA3 pyramidal neurons and consequent memory impairment, particularly evident for spatial and short-term memory tasks (McEwen, 1999, 2001).
The CA3 region is the hippocampal sub-area at higher risk for the consequences of glucocorticoid overexposure: indeed, in experimental conditions, long-term glucocorticoid exposure is associated to different patterns of changes, ranging from less complex branching and reduction in length of dendritic trees (Magarinos et al, 1996), to enhanced vulnerability towards vascular and metabolic injuries (Lawrence and Sapolsky, 1994) and finally to cell death, probably throughout the apoptotic mechanism (McEwen, 1999).
Not all experimental work confirmed the gluco-corticoid-induced morphological and metabolic changes in CA3 pyramidal cells, for example, high doses of exogenous glucocorticoids for 1 year or a chronic psychosocial stress did not affect pyramidal neurons in monkey and tree shrew, respectively (Leverenz et al, 1999; Vollmann-Honsdorf et al, 1997). In rats less extreme stressors are not necessarily related to neuronal loss, but only to dendritic changes, often reversible (Magarinos et al, 1999).
At this point some considerations must be made, when extrapolating experimental data to human studies, one should take into account the different dosage of glucocorticoid, the duration of glucocorticoid treatment or exposure and finally the species differences, since differences in glucocorticoid receptor mRNA were found in the hippocampus of the different investigated species (Sanchez et al, 2000).
Also in human studies, the relationship between hypercortisolemia and hippocampal morphology is still debated. Besides studies dealing with the hypothesis that the exposure to high glucocorticoid levels induce hippocampal atrophy (Lupien et al, 1998), other evidence suggests that in some pathological conditions, such as major depression or steroid-treated patients, both characterized by increased glucocorticoid concentrations, the hippocampal atrophy or the pyramidal cell loss, even in areas at risk for glucocorticoid, is a minor event (Lucassen et al, 2001). Indeed in this study also the indirect post-mortem evaluation of synaptic density by the synaptophysin-like immunoreactivity, failed to demonstrate significant differences both in depressed and in steroid-treated patients. In the parallel paper, the same group of authors (Lucassen et al, 2001) suggested that corticosteroid over-exposure in humans is not associated with permanent damage of the hippocampal region and particularly of the CA3 region.
However, still in recent years the majority of psychoneuroendocrine studies associated chronically high glucocorticoid levels to memory impairment, with concomitant changes of hippocampal formation as well as of other cortical and subcortical regions (Lupien and Lepage, 2001).
The brain glucocorticoid binding sites include two different type of receptors (Veldhuis, et al, 1982; Reul and de Kloet, 1985). The first ones, named mineralocorticoid receptors (MRs), selectively expressed in the limbic system, show a high affinity for glucocorticoids and are already activated at trough steroidal level (for example, the Cortisol concentration at evening and night-time). The second ones called glucocorticoid receptors (GRs) are present in the pituitary, in the hypothalamic area and in prefrontal cortex, and are characterized by a low glucocorticoid affinity; therefore they are activated only in stress conditions or in coincidence with the circadian crest-time of Cortisol rhythm, but always after the complete saturation of MRs (Meaney and Aitken, 1985; Diorio et al, 1993).
The different affinity of glucocorticoid receptors and the ratio between their degree of occupancy may explain the relationship between stress hormones and cognitive performance, typically represented by an inverted-U shape function (Lupien and McEwen, 1997). Indeed, when the ratio of MRs/GRs occupation is low (such as during severe hypocortisolism, but also after exogenous supraphysiological doses of glucocorticoids), LTP significantly decreases. LTP represents a physiological correlate of long-lasting enhancement in synaptic efficacy in response to high-frequency electrical stimulation and, like memory, it is rapidly inducible and of long duration. On the contrary, the partial occupation of GRs with the MRs totally saturated (i.e. Cortisol levels mildly elevated) correspond to an optimal LTP (for review, see De Kloet, 1999) (Fig. 1).
Both in animals and in humans aging per se is able to affect hippocampal functions, making neurons more vulnerable to a great variety of injuries and neuropathological conditions. These effects may become more evident and severe when the aging organism is also exposed to cumulative stress and thus to a chronic activation of the HPA-axis, potentially involved in the pathogenesis of several neurodegenerative disorders, including AD (Viau, 2002).
Cortisol levels mildly elevated
MRs 100% saturated GRs 50 % saturated
= LTP optimal levels
MRs and GRs occupancy very low
Was this article helpful?
Discover The Secrets To Staying Young Forever. Discover How To Hinder The Aging Process On Your Body And In Your Life. Do you feel left out when it comes to trying to look young and keeping up with other people your age? Do you feel as though your body has been run down like an old vehicle on its last legs? Those feelings that you have not only affect you physically, but they can also affect you mentally. Thats not good.