European Journal of Endocrinology (2001) 144 319±329 ISSN 0804-4643

INVITED REVIEW

Age-related changes of the hypothalamic±pituitary±adrenal

axis: pathophysiological correlates
E Ferrari, L Cravello, B Muzzoni, D Casarotti, M Paltro, S B Solerte, M Fioravanti, G Cuzzoni, B Pontiggia
and F Magri
Department of Internal Medicine and Medical Therapeutics, University of Pavia, Pavia, Italy
(Correspondence should be addressed to E Ferrari, Department of Internal Medicine and Medical Therapeutics, University of Pavia, Piazza Borromeo 2,
27100 Pavia, Italy; Email: ferrari@unipv.it)

Abstract
The aim of this review was to examine the evidence for age-related changes of the hypothalamic±
pituitary±adrenal (HPA) axis in both physiological and pathological aging, on the basis of the many
data in the literature, as well as of our personal findings.
A statistically significant circadian rhythmicity of serum cortisol was maintained in elderly
subjects, even if with a reduced amplitude of the 24 h fluctuations and a trend to an increase of the
serum levels in the evening and at night-time, in comparison with young controls. Furthermore, an
age-related impairment of HPA sensitivity to steroid feedback was present in elderly people.
The occurrence of senile dementia amplified the changes already present in physiological aging.
While the cortisol secretion was generally well maintained in aging, the adrenal production of
dehydroepiandrosterone and of its sulfate (DHEAS) exhibited an age-related decline. Therefore, the
cortisol/DHEAS molar ratio was significantly higher in elderly subjects and even more in demented
ones, than in young controls.
Due to the opposite effects of cortisol and DHEAS on the brain and particularly on the hippocampal
region, the imbalance between glucocorticoids and androgens occurring in physiological and even
more in pathological aging, may have adverse effects on the function of this region, whose key role in
learning and memory is well known.

European Journal of Endocrinology 144 319±329

neural cell impairment and the compensatory gliosis

Introduction
The hypothalamic±pituitary±adrenal (HPA) axis is an
auto-regulating system with many modulatory
mechanisms; due to such regulation, the circulating
levels of glucocorticoids are highly variable, according
both to the spontaneous rhythmic fluctuations and to
the responses towards stressful conditions. Further-
more, the HPA axis realizes a tight integration among
the endocrine, nervous and immune systems, being its
key point.
In this way, the HPA axis is one of the most
important allostatic (or adaptive) systems; when active,
it allows the organism to respond to a challenge, or in
general to a stressful condition, but its inactivation is
equally important, since the allostatic load can over-
expose the organism to the mediators of neural,
endocrine and immune stress, with different patho-
physiological consequences (1, 2).
Significant morphological and functional changes
affect with age the HPA axis at different levels both
in experimental animals and in human beings. The
are particularly evident at the level of the hypothala-
mus and of the hippocampal and limbic system. In
particular the neuronal impairment of the suprachias-
matic nucleus (3) may be responsible for alterations of
the circadian temporal structure of the aged organism,
due to the role played by the same nucleus as central
pacemaker of several circadian rhythms. Furthermore,
the loss of neurons, especially affecting both the
hypothalamus as well as the hippocampus and the
limbic region, implies a reduction of the glucocorticoid
receptors and hence an impaired regulation of the
adrenocortical secretion (4, 5). Indeed, the role played
by the hippocampus, both in animals and in humans,
in maintaining the adrenocortical circadian rhythmi-
city (6), as well as in modulating the glucocorticoid
feedback control of adrenocorticotropin (ACTH) secre-
tion (7±9) and the adrenocortical responses to stressful
conditions (10) is well known.
Consequently, the hippocampal degenerative changes
occurring in physiological and even more in pathological
aging may be responsible for the impaired sensitivity to

q 2001 Society of the European Journal of Endocrinology Online version via http://www.eje.org
320 E Ferrari and others
the steroid feedback and for a certain degree of
hyperactivity of the HPA axis in elderly subjects (11).
However, the heterogeneity of the age-related HPA
patterns must be underlined; indeed, some individuals
exhibit substantial changes while others maintain an
HPA function similar to younger individuals.
Many factors might be potentially involved in the
aging process and, in particular, in brain aging, with
several metabolic, immune and neuroendocrine
changes playing a pathogenetic role. Due to the
primary role of the adrenal steroids in maintaining
the homeostasis and the adaptational organism
responses to stress, it seems very interesting to study
the link between the corticosteroid secretion and the
limbic±hippocampal age-related changes.
In this review we particularly focus on the age-
related changes of the glucocorticoid and androgen
secretion, by considering separately the two hormonal
patterns, due to the peculiar modifications affecting the
different adrenocortical zones in aging. For such
studies we favored the evaluation of the circadian
fluctuations of blood corticosteroids, which may
provide better information about the pathophysiology
of the endocrine secretions in comparison with the
pharmacological tests, which sometimes do not reflect
the spontaneous hormonal release. Furthermore, the
chronobiological evaluation of neuroendocrine func-
tions in elderly people may represent a clinically
reliable tool to investigate the biology of the CNS and
particularly of the limbic±hypothalamic system, and
hence reveal the pathophysiology of the cerebral aging.
Together with physiological aging, senile dementia
represents a clinical model very interesting for chron-
obiological studies, either for the frequency of changes
of the central neurotransmitter pathways, or because of
the frequent occurrence of degenerative alterations of
the hypothalamic suprachiasmatic nucleus.
Cortisol secretion and aging
Much evidence suggests that the primary defect in
allostatic systems in aging might be a prolonged
response to stressful conditions, due to the inability to
shut off the allostatic response after the end of the
stress (12±16). Consequently the HPA axis would
become less resilient with age in responding to
stimulations.
However, in spite of the subtle age-related changes of
HPA function, the circulating levels of glucocorticoid
show a relative constancy or even a trend toward an
increase with aging. Indeed, the blood levels of both
cortisol and ACTH usually fall within the normal range
in physiological aging, but a trend towards higher
nocturnal levels is often reported (17±19).
The possible role of changes of the glucocorticoid
signal in the age-related modifications of the CNS has
been often suggested. Indeed, since 1968, a link
between high glucocorticoid levels and neuronal loss
www.eje.org
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
had been hypothesized in Alzheimer's disease (AD)
(20); in the following years much experimental work
confirmed this suggestion (21, 22). Similar findings
were found both in primates (23) and in humans; in
particular, a significant relationship between the
increasing cortisol levels over a period of 4 years and
the impairment of explicit memory and selective
attention performances has been described in a long-
itudinal study (24).
Furthermore, high cortisol levels with an impaired
HPA sensitivity to steroid feedback are often described
in senile dementia, both of degenerative and vascular
types (25±28).
Besides, a positive relationship of the cortisol
response to an i.v. glucose load with the degree of
hippocampal atrophy and the severity of dementia has
been described in several demented patients (29);
moreover the cortisol increase seems to be a very
sensitive marker of cognitive decline in AD patients
(30).
The conventional measures of total cortisol levels
might underestimate the real degree of hypercortiso-
lism in AD patients, since a reduction of corticosteroid-
binding globulin (CBG) levels has been found in about
30% of the patients with this kind of dementia (31). In
any case, the CBG plasma levels were similar in old
subjects with high, low or normal cortisol values, and a
strong correlation between total and unbound cortisol
levels across subjects has been described (32).
A cross-sectional approach in studying the age-
related changes of the HPA axis may offer only tentative
suggestions. However, information arising from serial
blood samples collected every 4 h during the day and
every 2 h during the night, may give a true picture of
the hormonal plasma changes in groups of different age
which may underline the interindividual variations.
Furthermore, the inclusion in the study of subjects
drug-free and without overt acute or chronic disease
allowed us to investigate the role of age and dementia in
HPA changes, independently of individual health status.
On these basis, we studied the circadian rhythm of
serum cortisol in a wide number of healthy old subjects
…n ˆ 52; age 69±90 years), of old demented patients,
including both AD and multi-infarct dementia (vascu-
lar dementia (VD)) …n ˆ 35; age 69±90 years) and of
young controls …n ˆ 22; age 19±43 years).The physio-
logical cortisol circadian rhythmicity was maintained,
with a significant age-related increase of the nocturnal
cortisol levels in physiological aging and even more in
senile dementia, in comparison with young subjects
(Fig. 1).
Our data agree with other studies, showing the age-
related increase of the cortisol nadir values, together
with the phase-advance of the onset of the circadian
cortisol rise (33). Due to these changes, a reduction of
the quiescent period of the cortisol secretion and hence
an increase of the cortisol signal throughout the 24 h
occur in elderly people in comparison with young one.
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
The relative increase of cortisol serum levels in
the evening and at night-time is responsible for the
clear flattening of the cortisol circadian profile.
Indeed, the nocturnal increase of serum cortisol
(calculated as difference between the values recorded
at 0800 and 0000 h) was significantly reduced in
both physiological and pathological aging; moreover
this reduction was negatively correlated to both the
age (r ˆ 20:3718, P , 0:01) and the cognitive
Aging and adrenal steroids 321
Figure 1 Serum DHEAS and cortisol circadian rhythm
(before and after DXM administration, 1 mg orally at
2300 h) in physiological and pathological brain aging and
in young controls (means^S.E.M.). Old subjects vs
young controls: a ˆ P , 0:05; b ˆ P , 0:01;
c ˆ P , 0:001: AD patients vs young controls:
d ˆ P , 0:05; e ˆ P , 0:01; f ˆ P , 0:001: AD patients
vs old subjects g ˆ P , 0:05; h ˆ P , 0:01;
i ˆ P , 0:001: L/D, light dark.
performance of subjects (r ˆ 0:3580, P , 0:01),
the latter being evaluated by the Mini Mental
State Examination (Table 1).
Since the crucial role played by the central cholin-
ergic and serotoninergic pathways in the circadian
activation of the ACTH-secreting system at night-time
is well known (33, 34), the impairment of the cortisol
nocturnal increase in elderly subjects may be a marker
of a reduced activity of these pathways.
www.eje.org
322 E Ferrari and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144

Table 1 Cortisol nocturnal increase in Interesting results have been also reported (44) by
physiological and pathological brain aging and in the CRH stimulation under DXM suppression. Indeed,
young controls (means ^ S.E.M.).
in spite of the higher cortisol levels recorded in old
Young controls 16.10 ^ 1.39a,b subjects under basal conditions, the levels of both
Old subjects 9.26 ^ 0.88 ACTH and cortisol after DXM were not different in old
Demented patients 8.02 ^ 1.83 and in young subjects, whereas a significantly higher
a cortisol response to CRH occurred in old subjects.
P , 0:001 (Student's t-test) vs demented patients.
b
P , 0:001 vs old subjects. Some data in the literature (45, 46) suggest that
aging does not modify the pituitary and adrenal
responsiveness to exogenous CRH. However, other
A reduction of the melatonin signal, widely described studies concerning cortisol and ACTH response to
in aging, could also contribute to the changes of both CRH and vasopressin, suggest that the pituitary
cortisol and ACTH nocturnal increase in elderly gland might become more sensitive with aging to CRH
subjects. Indeed the pineal gland seems to play an alone or in association with vasopressin (12, 16, 47) in
inhibitory role on the HPA axis (35, 36), by modulating the presence of stressful conditions and of age-related
the adrenal sensitivity to the corticotropic stimulation diseases (48). In our experience (45), the response of
(37) and even by limiting the corticotropin response to cortisol and ACTH to a CRH stimulation test (ovine
stressful conditions (38). Our data concerning the CRH, 1 mg/kg i.v.) was not significantly different in 14
simultaneous evaluation of cortisol and melatonin healthy old women (69±83 years) and in 11 young
secretions in elderly subjects confirm the opposite women (20±40 years) (Table 3).
behavior of the two secretions in both physiological Age-related changes of the endogenous CRH or
and pathological aging (39, 40) (Table 2). vasopressin secretion may also occur (49). Indeed in
Age-related changes of the dynamic secretory AD patients the CRH immunoreactivity has been
pattern of HPA axis have also been described. In fact, described as increased in the hypothalamus and
the cortisol peak values recorded during insulin- decreased in other cerebral areas, suggesting the
induced hypoglycemia were generally higher in old existence of alterations of the glucocorticoid regulation
than in young subjects, in spite of the smaller glycemic of this peptide (49). Similar neuroendocrine changes
fall occurring in the elderly (41). Thus, the HPA axis in may also be suggested in physiological aging (50).
aged people seems to recover slowly from certain types The adrenal sensitivity to exogenous ACTH seems
of stress. not to be significantly modified during physiological
Higher cortisol levels in old than in young subjects aging and in the early stage of AD (51). In our
have been described during some pharmacological experience (52) the cortisol response to pulse injection
challenges, such as the dexamethasone (DXM) suppres- of a small dose of synthetic ACTH (Synacthen,
sion test, the stimulation by human or ovine cortico- 2500 ng, i.v., at 2030 h) was quite similar in young
tropin-releasing hormone (CRH) or by physostigmine and old healthy women; however, patients with overt
(12, 42), probably due to the age-related changes of the senile dementia exhibited a significant increase of both
HPA sensitivity to both exogenous and endogenous the duration and the amplitude of cortisol response, as
stimuli. However, it seems noteworthy that the basal well as a significant delay of the time to peak …73 ^
cortisol levels exhibit a high intraperson stability and 4:7 min†; when compared with old healthy subjects
that they are closely related to the feedback sensitivity …46 ^ 4:9 min† and with young controls …45 ^ 5 min†:
of the HPA axis towards a low dose (0.25 mg) of DXM Since the cortisol response to a pulse injection of a
(43), suggesting a genetic influence on the set point of small dose of Synacthen is similar to that induced by a
the HPA axis. stressor, it is possible to argue that also in humans the
Table 2 Relationship between melatonin and cortisol secretion in
adrenocortical responsiveness to stressful conditions is
physiological and pathological brain aging and in young controls increased and prolonged only when `senescence is
(means ^ S.E.M.). coupled with pathological conditions', in agreement
with previous studies on old animals (14).
Ratio between Ratio between An age-related decrease of the HPA sensitivity to the
cortisol and cortisol nadir
melatonin and melatonin
steroid feedback signal has been reported. In particular
circadian mesors peak both the rate of non-suppression and the residual
cortisol levels after DXM were higher in demented
Young controls 0.58 ^ 0.17a 0.05 ^ 0.01b patients than in age-matched controls, possibly in
Old subjects 0.93 ^ 0.14 0.34 ^ 0.09c relation to anxiety, depression, panic or restlessness
AD patients 1.33 ^ 0.20 0.58 ^ 0.11
(53). Since these clinical symptoms and the high
AD patients and young controls: age vs cortisol nadir/melatonin peak ratio cortisol levels may be modified by the selective
r ˆ 0:64; P , 0:001:
a
serotonin re-uptake blockers, it is possible to suggest
P , 0:001 vs AD patients.
b
P , 0:01 vs AD patients. that the enhanced activity of the HPA axis in senile
c
P , 0:001 vs AD patients. dementia may be related to a central neurotransmitter

www.eje.org
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
Table 3 Plasma ACTH and cortisol response to CRH stimulation
test (ovine-CRH, 1 mg/kg i.v.) in young and old healthy women
(means ^ S.E.M.).
Old healthy Young healthy
women women
…n ˆ 14; …n ˆ 11;
age 69±83 years) age 20±40 years)
ACTH (pg/ml)
Baseline 27.1 ^ 3.7 19.4 ^ 1.5
Delta (max.) 29.2 ^ 8.4 28.1 ^ 10
Delta (%) 117.9 ^ 22.3 144.8 ^ 33.5
Cortisol (mg/dl)
Baseline 15.7 ^ 1.3 9.0 ^ 1.1
Delta (max.) 12.6 ^ 2.4 8.4 ^ 1.5
Delta (%) 90.4 ^ 18.5 93.3 ^ 21.2
imbalance, probably responsible also for the impaired
sensitivity to the steroid feedback (53, 54).
It is now well established that the glucocorticoid
feedback on the HPA axis includes at least three time
domains: fast (seconds or minutes), intermediate
(usually less than 10 h) and slow (within hours or
days) (55), and that for all of them the sites of action
are the hypothalamus and the hippocampus (56), two
brain areas in which the corticosteroid receptors (both
type I and type II) are widely represented (57).
In order to study the first phase of the negative
feedback, Boscaro et al. (9) recently evaluated in aged
people and in young controls the ACTH and cortisol
suppressibility by the infusion of hydrocortisone, which
differs from DXM in the affinity for corticosteroid
receptors subtypes (16, 58±61). In spite of a similar
cortisol increase in both groups of subjects, the old ones
exhibited a slight and not significant decrease of ACTH
levels during the first phase of the response to
hydrocortisone infusion, in comparison with young
controls; on the contrary, a pronounced and significant
decline of ACTH concentrations occurred thereafter in
old subjects.
These findings suggest that the age-related changes
of the HPA activity in aging may be related to several
factors, namely a decrease with aging of the brain
corticosteroid receptor concentration (62), but also the
age-related differences in the cerebrospinal fluid (CSF)
steroid concentrations (63), probably secondary to
changes of the cortisol clearance rate in CSF and/or
of the blood±brain barrier (64, 65). Thus, also vascular
factors may play a pathogenetic role in the decline of
the HPA sensitivity to the steroid feedback.
In physiological and pathological brain aging and in
young controls, the study of the circadian rhythm of
serum cortisol, both before and after DXM administra-
tion (1 mg at 2300 h, orally) (Fig. 1), allowed us to
detect a high percentage of `non-responders' (serum
cortisol levels above 5 mg/dl the morning following
DXM (66), in healthy old subjects (about 30%) and in
old demented patients (50%). Indeed, the mean cortisol
values recorded after DXM were significantly higher in
Aging and adrenal steroids 323
elderly subjects and especially in demented patients
than in young controls; moreover, patients with senile
dementia also exhibited a delay of the cortisol response
to DXM (67).
These findings underline the relevance of the `age'
factor in the pathogenesis of the impaired HPA
sensitivity towards the negative steroid feedback, and
the additional effect of the occurrence of senile
dementia.
In conclusion, the changes of the cortisol secretory
pattern observed in elderly subjects and especially in
patients with senile dementia (namely the flattening of
the circadian profile and the increase of the nadir
levels, together with the impaired sensitivity to the
steroid feedback) are reminiscent of the data described
in senescent animals or in animals with experimental
hippocampal lesions (68).
Dehydroepiandrosterone sulfate
(DHEAS) and aging
Only in humans and a few primates do the adrenals
secrete dehydroepiandrosterone (DHEA) and its sulfate
(DHEAS) (69, 70); in particular, the zona reticularis of
the adrenal cortex is the exclusive source for DHEA and
DHEAS, and the level of expression of 3b-hydroxyster-
oid dehydrogenase is the most important determinant
for their biosynthesis (71).
In contrast with the minimal changes of the cortisol
secretion occurring with age, the last decades of life are
characterized by a progressive reduction of the secre-
tion rate of DHEA and DHEAS, the main adrenal
androgens.
In fact, the blood levels of these steroids, very low in
the first years of life, begin to rise at the adrenarche and
progressively increase till the third decade; after the age
of 30 the steroid levels declines at a rate of 1±2% per
year, even if with marked interindividual differences,
and by the age of 70±80 they correspond only to 20±
30% of the peak concentration occurring in young
people (72±74). The DHEAS/DHEA ratio does not
change with aging.
In spite of the great interindividual variability, the
DHEAS levels show a low within-subjects changes;
therefore they may be considered as highly specific
individual markers (75).
The morphological correlates of the physiological
age-related changes of DHEA and DHEAS secretion are
the modifications of the zona reticularis. Indeed, the
development of this zone of the adrenals begins at the
adrenarche and shows a clear width reduction during
aging (76). The mechanisms responsible for this
reduction are still unknown; furthermore, a significant
correlation between the DHEAS decline in aging and
the morphological changes of the zona reticularis was
never found (77).
However, the zona reticularis is particularly suscep-
tible to the intra-adrenal gradient of autocrine and
www.eje.org
324 E Ferrari and others
paracrine factors created by the blood supply, and to
vascular damages, such as multiple microhemorragic
events, with subsequent necrotic damage. Taken
together, all these factors might explain the age-related
decline and the interindividual changes in DHEAS
secretion.
In addition to the reduced secretion in basal
conditions, the DHEAS response to exogenous ACTH
is significantly impaired in old subjects, whereas the
cortisol response is generally maintained (78). Thus, a
specific, but unexplained, defect of the desmolase
activity of P450c17 cytochrome in the zona reticularis
seems to occur with aging (79).
Extra-adrenal factors may also affect the adrenal
androgen biosynthesis. For example, some proopio-
melanocortin-related peptides, and in particular the
joint peptide and beta-endorphin, could stimulate the
androgen secretion (80, 81), although some authors
did not confirm such action (82, 83). The possible role
of a specific pituitary adrenal androgen-stimulating
factor, identified by Parker et al. (81), is still debated.
Furthermore, a possible influence of insulin in the
age-related decline of DHEAS has been suggested, due
to the inhibitory activity of insulin on DHEAS secretion,
and the trend to increase of insulin during aging, but
this effect seems well-evident only in men (84).
In addition to the doubts concerning the patho-
physiology of the age-related decline of DHEAS secre-
tion, many questions still remain about the central and
peripheral effects of adrenal androgens, since no
specific cellular or molecular targets for DHEA and
DHEAS have until now been identified (79). However,
the results of DHEA administration in experimental
animals or during clinical trials, suggest a pleiotropic
effect of this hormone.
Indeed, DHEAS, but not DHEA, is able to activate the
hepatic peroxisome proliferator-activated receptor
alpha (85), an intracellular receptor of the steroid
receptors family; this activation could modulate the
fatty acid metabolism and the expression of peroxi-
somal enzymes, and thus could explain the DHEAS
anticarcinogenetic and chemoprotective effects (85).
Antidiabetic and antiobesity activities of DHEAS have
been also described in experimental animals, but they
are not always confirmed in humans. Furthermore,
high DHEAS levels have been related to a significant
decrease of mortality for cardiovascular disease in men
but not in women (86).
The most important findings, however, concern the
effects of DHEAS on the CNS, where these steroids
may be synthesized de novo, namely independently
from the gonads and the adrenal glands, and thus
they are called `neurosteroids' (87). At the present it
is unknown if changes in plasma DHEAS levels can
directly affect CNS functions. However, at the CNS
level, DHEAS and pregnenolone sulfate on the one
hand and tetrahydroprogesterone, tetrahydrodeoxy-
corticosterone and androsterone on the other, effect
www.eje.org
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
an important homeostatic control mechanism for the
neuronal activity, by binding to the g-aminobutyric
acid A (GABAA) receptor (88). In particular, DHEA
and DHEAS, acting as allosteric antagonists of
GABAA, receptor, may enhance the neuronal and
glial survival and may improve learning and memory
capacities (89, 90). Indeed in experimental animals,
DHEA administration has a memory-enhancing effect
and in vitro studies suggest a neurotropic effect on
neurons and glial cells (91).
Furthermore, the noradrenergic release of N-methyl-
d-aspartate in the rat hippocampus may be enhanced
by DHEA (92). This action, together with the proser-
otoninergic activity of DHEAS, could be responsible for
the antidepressive effects, and for the improvement of
memory reported after DHEA administration in
humans, due to the key role played by noradrenaline
on this function (93±95).
In rat primary hippocampal neurons, the addition of
DHEAS, but not of DHEA, was able to enhance the
neuronal resistance to stress relevant for both acute
and chronic neurodegenerative conditions, by the
activation of a kB-binding transcription factor (96)
which is involved in the neuroprotection against
oxidative stress, calcium and b-amyloid (97, 98).
Thus, the age-related decline of DHEAS might play a
role in the occurrence of AD and other neurodegen-
erative diseases.
Beside the possible role of DHEA on cognitive
functions, many studies suggest an involvement of
the adrenal androgens also in the maintenance of
functional abilities (99, 100) and in the pathogenesis of
several age-related diseases (101). Furthermore, the
reported lack of relationship between the age and the
DHEAS levels in subjects over 90 years old (102),
suggests that high DHEAS levels might be associated
with a longer survival. Indeed, when age is factored out
by appropriate polynomial regression, a significant
component of DHEAS variations seems to exist (103).
The data in the literature concerning DHEAS
secretion in senile dementia are still conflicting (104±
110), although at the moment strong evidence points
to the possible role of DHEAS in limiting the cortisol
neurotoxic effect on hippocampal cells, in this way
preventing the brain degeneration in AD (108).
In a community-based study, no significant relation-
ships between DHEAS levels and cognitive function
have been found (111). In another longitudinal study,
lower DHEAS values were recorded in women with
functional limitations, depressive symptomatology and
poor well-being, but not in men (112).
Our results, obtained by the simultaneous evaluation
of the circadian rhythm of blood cortisol and DHEAS in
physiological and pathological brain aging, including
both AD and VD, indicated a clear decline in DHEAS
secretion throughout the 24 h cycle in elderly subjects
…n ˆ 23; age 76±96 years), and especially in old
demented patients …n ˆ 23; age 68±91 years), without
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144 Aging and adrenal steroids 325

sex-related differences, in comparison with young

controls …n ˆ 10; age 25±32 years). Concerning the
type of dementia, patients with VD exhibited the lowest
DHEAS levels (Fig. 1).
The statistical analysis of our data by the population
mean cosinor method demonstrated the statistical
significance of the serum DHEAS circadian rhythm in
young and elderly subjects, even if with a marked
reduction of both mesor and amplitude in old subjects,
particularly if demented. However, a good maintenance
of the temporal relationship between the DHEAS and
cortisol crest-times persisted in both physiological and
pathological brain aging, as already described in
children and in young adults (113).
The molar ratio between cortisol and DHEAS levels
throughout the 24 h cycle was significantly higher in
elderly subjects, particularly if demented, in compar-
ison with young controls. Furthermore, old demented
patients exhibited a further increase of this ratio at
night-time, when compared with old controls, suggest-
ing the existence of a deeper neurotoxic effect of the
adrenocortical secretory imbalance (Fig. 1). Figure 2 Mean hippocampal volumes in physiological and
pathological brain aging and in young controls. Right: volume vs
This finding seems particularly interesting if we age, r ˆ 20:88; P , 0:001; vs cortisol nocturnal increase r ˆ 0:40;
consider that, in spite of the decline of DHEAS P , 0:05; vs DHEAS mesor, r ˆ 0:63; P , 0:01: Left: volume vs
secretion, well evident also in extreme senility, we age, r ˆ 20:82; P , 0:01; vs cortisol nocturnal increase, r ˆ 0:45;
have recently found similar cortisol levels in healthy P , 0:05; vs DHEAS mesor, r ˆ 0:64; P , 0:01: ***P , 0:001:
centenarians …n ˆ 20; age 100±106 years) and in
younger controls. Thus the cortisol/DHEAS molar ratio
was similar in centenarians and in healthy old controls hippocampal volumes and a progressive enlargement of
(data not shown). the ventricular size, which represents an index of
Due to the opposite changes of cortisol and androgen cerebral atrophy and volume loss (120). An important
secretions occurring in physiological and even more in finding to be emphasized is that the reduction of the
pathological brain aging, it seemed interesting to study, hippocampal volumes seems to be significantly related
by morphometric measures, the cerebral changes to the impairment of the cortisol nocturnal increase
occurring in these conditions, in order to try to and to the mean circadian value of serum DHEAS
correlate the neuroendocrinological features with the (Fig. 2). Even if in humans it is hard to demonstrate a
cerebral volumetric modifications. relationship between the hippocampal modifications
In particular we have considered the hippocampal and the adrenal secretion, our data suggest the
changes, since it is well known that both the age- existence of a link between the reduction of hippo-
related neuronal impairment and the degenerative campal volumes occurring with age and the trend to
changes of AD are especially evident in this area. increase of cortisol in the evening and at night-time
Indeed, the hippocampus is a particularly vulnerable and the decline of DHEAS secretion.
brain area, with a high expression of glucocorticoid Among the other cerebral volumes evaluated, only
receptors, and it is deeply involved in the regulation of the temporal lobe was significantly lower in AD
the HPA axis (114). Thus, it represents a vulnerable patients than in age-matched controls, suggesting the
link in the regulation of both HPA function and specificity for AD of these changes (121).
cognition (115).
Both in experimental animals and in humans, the
hippocampal volume decreases with aging, due not Conclusions
only to neuronal loss, but also to the effects of According to the results of our research and of many
neurotransmitters imbalance, excitatory amino acids studies in the literature, a clear dissociation of the
(116) and adrenal steroids (117, 118). To assess in vivo adrenocortical secretory pattern occurs with aging, due
the main cerebral volumes in healthy young and old to good maintenance of cortisol secretion and clear
subjects and in AD patients, we have recently impairment of that of androgens in elderly subjects,
performed a cerebral morphometric analysis by the with particular evidence in the demented ones.
magnetic resonance imaging method (119). Consequently, a significant increase of the cortisol/
In spite of the great interindividual variability, DHEAS molar ratio occurs in both physiological and
our study demonstrated an age-related decline in pathological aging.

www.eje.org
326 E Ferrari and others
It is now well known that both glucocorticoids and
androgens have relevant effects on the brain, most
notably on the hippocampus, a critical area for learning
and memory, as well as a crucial center for the
glucocorticoid inhibition of their own release (115,
122, 123). The effects of the two kinds of steroids on the
hippocampus are opposite, since cortisol promotes the
neuronal degeneration and death, whereas DHEAS
plays a protective role towards the structural damages
and the functional impairment due to aging itself or to
pathological situations, and stimulates the neuronal
long-term potentiation (124). Thus, the age-related
decrease of DHEAS secretion necessarily results in an
impairment of the DHEAS-mediated antiglucocorticoid
activity, and consequently in an enhanced neurotoxicity
of glucocorticoids and of excitatory amino acids (123).
As a whole our data suggest that, as in experimental
animals (125, 126), the brain aging may depend upon
the cumulative exposure to increasing cortisol levels
throughout life (32, 127), particularly if parallel to a
reduced androgen secretion.
The age-related hormonal changes, such as the
opposite modifications of cortisol and melatonin secre-
tion, may play a role in the pathophysiology of the CNS
degenerative changes also through the enhancement of
5-lipoxygenase (5-LOX) expression in the CNS (128);
this is the key enzyme responsible for the synthesis of
inflammatory leukotrienes, and probably involved in
neurodegeneration. Indeed, it is well known that 5-LOX
expression, prominent in hippocampal and cerebellar
neurons, is stimulated by glucocorticoids (129) and
depressed by melatonin (130). Therefore, the subtle but
significant changes of the hormonal balance occurring
in aging may contribute to the onset and progression of
the aging-associated neurodegenerative diseases.
References
1 McEwen BS. Protective and damaging effects of stress mediators.
New England Journal of Medicine 1998 338 171±179.
2 McEwen BS & Stellar E. Stress and the individual: mechanisms
leading to disease. Archives of Internal Medicine 1993 153
2093±2101.
3 Swaab DF, Fliers E & Partiman TS. The suprachiasmatic nucleus
of the human brain in relation to sex, age and senile dementia.
Brain Research 1985 342 37±44.
4 Ball MJ. Neuronal loss, neurofibrillary tangles and granulova-
cuolar degeneration in the hippocampus with ageing and
dementia. Acta Neuropathologica 1977 37 111±118.
5 Mani R, Lohr J & Jeste D. Hippocampal pyramidal cells and
aging in the human: a quantitative study of neuronal loss in
sectors CA1 to CA4. Experimental Neurology 1986 94 29±40.
6 Moberg GP, Scapagnini U, de Groot J & Ganong WF. Effect of
sectioning the fornix on diurnal fluctuation in plasma
corticosterone levels in the rat. Neuroendocrinology 1971 7
11±15.
7 Feldman S & Conforti N. Participation of the dorsal hippocam-
pus in the glucocorticoid feed-back effect on adrenocortical
activity. Neuroendocrinology 1980 30 42±55.
8 Vermeulen A, Deslypere JP, Schelfhout W, Verdonck L &
Rubens R. Adrenocortical function in old age: response to
www.eje.org
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
acute adrenocorticotropin stimulation. Journal of Clinical Endo-
crinology and Metabolism 1982 54 187±191.
9 Boscaro M, Paoletta A, Scarpa E, Barzon L, Fusaro P, Fallo F et al.
Age-related changes in glucocorticoid fast feedback inhibition of
adrenocorticotropin in man. Journal of Clinical Endocrinology and
Metabolism 1998 83 1380±1383.
10 Conforti N & Feldman S. Effects of dorsal fornix section and
hippocampectomy on adrenocortical responses to sensory
stimulation in the rat. Neuroendocrinology 1976 22 1±7.
11 Sapolsky RM, Krey LC & McEwen BS. The neuroendocrinology
of stress and ageing. The glucocorticoid cascade hypothesis.
Endocrine Reviews 1986 7 284±301.
12 Seeman TE & Robbins RJ. Aging and hypothalamic±pituitary±
adrenal response to challenge in humans. Endocrine Reviews
1994 15 233±260.
13 McCarty R. Sympathetic-adrenal medullary and cardiovascular
responses to acute cold stress in adult and aged rats. Journal of
the Autonomic Nervous System 1985 12 15±22.
14 Sapolsky RM. Stress, the Aging Brain and the Mechanisms of
Neuron Death. Cambridge, MA: MIT Press, 1992.
15 McEwen BS. Re-examination of the glucocorticoid hypothesis of
stress and aging. In Progress in Brain Research, Vol. 93: The
Human Hypothalamus in Health and Disease, pp 365±383. Eds DF
Swaab, MA Hofman, M Mirmiran, R Ravid & FW van Leeuwen.
Amsterdam: Elsevier Science, 1992.
16 Wilkinson CW, Peskind ER & Raskind MA. Decreased hypo-
thalamic±pituitary±adrenal axis sensitivity to cortisol feed-
back inhibition in human aging. Neuroendocrinology 1997 65
79±90.
17 Van Cauter E, Leproult R & Kupfer DJ. Effects of gender and age
on the levels and circadian rhythmicity of plasma cortisol.
Journal of Clinical Endocrinology and Metabolism 1996 81 2468±
2473.
18 Kern W, Dodt C, Born J & Fehm HL. Changes in cortisol and
growth hormone secretion during nocturnal sleep in the course
of aging. Journal of Gerontology Series A. Biological Sciences and
Medical Sciences 1996 51 M3±M9.
19 Hatzinger M, Z'Brunn A, Hemmeter U, Seifritz E, Baumann F,
Holsboer-Trachsler E et al. Hypothalamic±pituitary±adrenal
system function in patients with Alzheimer's disease. Neurobiol-
ogy of Aging 1995 16 205±209.
20 McEwen BE, Weiss JM & Schwartz LS. Selective retention of
corticosterone by limbic structures in rat brain. Nature 1968
220 911±912.
21 Sapolsky RM & McEwen BS. Stress, glucocorticoids and their
role in degenerative changes in the aging hippocampus. In
Treatment Development Strategies for Alzheimer's Disease, pp
151±172. Eds T Crook, R Bartus, S Ferris & C Flickers. New
Cannan, CT: Mark Powley Associates, 1986.
22 Meaney M, Aitken D, Bhatnagar S, Van Berkel C & Sapolsky R.
Postnatal handling attenuates neuroendocrine, anatomical and
cognitive impairments related to the aged hippocampus. Science
1988 238 766±768.
23 Sapolsky R, Uno H, Rebert CS & Finch CE. Hippocampal damage
associated with prolonged glucocorticoid exposure in primates.
Journal of Neuroscience 1990 10 2897±2902.
24 Lupien S, Lecours AR, Lussoer I, Schwartz G, Nair NP &
Meaney M. Basal cortisol levels and cognitive deficits in human
aging. Journal of Neuroscience 1994 14 2893±2903.
25 Balladin J, Gottfries CG, Karlsson I, Lindstet G, Langstrom G &
Walinder J. Dexamethasone suppression test and serum
prolactin in dementia disorders. British Journal of Psychiatry
1983 143 277±281.
26 Lawlor BA, Tsuboyama G, Ryan TM, Mohs RC, Davis BM,
Davidson M et al. Agitation and postdexamethasone cortisol
levels in Alzheimer's disease. American Journal of Psychiatry
1992 149 546±548.
27 Davis KL, Davis BM, Greenwald BS, Mohs RC, MatheÁ AA,
Johns CA et al. Cortisol and Alzheimer's disease: I. Basal studies.
American Journal of Psychiatry 1986 143 300±305.
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
28 Maeda K, Tanimoto K, Terada T, Shintani T & Kakigi T. Elevated
urinary free cortisol in patients with dementia. Neurobiology of
Aging 1991 12 161±163.
29 De Leon MJ, McRae T, Tsai JR, George AJ, Marcus DL,
Freedman M et al. Abnormal cortisol response in Alzheimer's
disease linked to hippocampal atrophy. Lancet 1988 ii 391±392.
30 Weiner MF, Vobach S, Olsson K, Svetlik D & Risser RC. Cortisol
secretion and Alzheimer's disease progression. Biological
Psychiatry 1997 42 1030±1038.
31 Meaney MJ, O'Donnel D, Rowe W, Tannenbaum B, Steverman A,
Walker M et al. Individual differences in hypothalamic±
pituitary±adrenal activity in later life and hippocampal aging.
Experimental Gerontology 1995 30 229±251.
32 Lupien S, Lecours AR, Schwartz G, Sharma S, Hauger RL,
Meaney MJ et al. Longitudinal study of basal cortisol levels in
healthy elderly subjects: evidence for subgroups. Neurobiology of
Aging 1996 17 95±105.
33 Krieger DT, Silverberg AI, Rizzo F & Krieger HP. Abolition of
circadian periodicity of plasma 17-OH-cs levels in the cat.
American Journal of Physiology 1968 215 959±967.
34 Ferrari E, Bossolo PA, Vailati A, Martinelli I, Rea A & Nosari I.
Variations circadiennes des effects d'une substance vagolytique
sur le systeÁme ACTH-secreÂtant chez l'homme. Annales d'Endo-
crinologie 1977 38 203±213.
35 Dickson KL & Hasty DL. Effects of the pineal gland in unilaterally
adrenalectomized rats. Acta Endocrinologica 1972 70 438±442.
36 Wetteberg L. The relationship between the pineal gland and
pituitary±adrenal axis in health, endocrine and psychiatric
conditions. Psychoneuroendocrinology 1983 8 75±80.
37 Sanchez De La Pena S, Halberg F & Hungar F. Pineal
chronomodulation. The feed-sideward. Clinical Chemistry News-
letter 1982 2 129±135.
38 Milin J, Djakovic-Svaicer K & Demajo M. Rat pineal gland
suppresses the injection stress-reactive ACTH outflow. Hormone
and Metabolic Research 1993 25 149±151.
39 Magri F, Locatelli M, Balza G, Molla G, Cuzzoni G, Fioravanti M
et al. Changes in endocrine circadian rhythms as markers of
physiological and pathological brain aging. Chronobiology
International 1997 14 385±396.
40 Ferrari E, Magri F, Locatelli M, Balza G, Nescis T, Battegazzore C
et al. Chrononeuroendocrine markers of the aging brain. Aging
1996 8 320±327.
41 Meneilly GS, Minaker KL, Young JB, Landesberg L & Rowe JW.
Counterregulatory responses to insulin-induced glucose reduc-
tion in the elderly. Journal of Clinical Endocrinology and
Metabolism 1985 61 178±182.
42 Raskind MA, Peskind ER & Wilkinson CW. Hypothalamic±
pituitary±adrenal axis regulation and human aging. Annals of
the New York Academy of Sciences 1994 746 327±335.
43 Huizenga NA, Koper JW, De Lange P, Pols HA, Stolk RP,
Grobbee DE et al. Interperson variability but intraperson stability
of baseline plasma cortisol concentrations, and its relation to
feedback sensitivity of the hypothalamo±pituitary±adrenal axis
to a low dose of dexamethasone in elderly individuals. Journal of
Clinical Endocrinology and Metabolism 1998 83 47±54.
44 Heuser IJ, Gotthardt U, Schweiger U, Schmider J, Lammers C-H,
Dettling M et al. Age-associated changes of pituitary-adrenocor-
tical hormone regulation in men: importance of gender.
Neurobiology of Aging 1994 15 227±231.
45 Ferrari E, Solerte SB, Magri F, Fioravanti M, Magni P, Dori D et al.
Age-related changes of pituitary±adrenal axis in humans. In
Stress and the Aging Brain, pp 39±51. Ed. G Nappi. New York:
Raven Press, 1990.
46 Pavlov EP, Harman SM, Chrousos GP, Loriaux DL &
Blackman MR. Responses of plasma adrenocorticotropin,
cortisol, dehydroepiandrosterone to ovine corticotropin releas-
ing hormone in healthy aging men. Journal of Clinical
Endocrinology and Metabolism 1986 62 767±772.
47 Holboer F, Grasser A, Friess E & Wiedemann K. Steroid effects on
central neurons and implications for psychiatric and neurological
Aging and adrenal steroids 327
disorders. Annals of the New York Academy of Sciences 1994 746
354±359.
48 Nichols NR. Glucocorticoids and aging. In Functional Endocri-
nology of Aging, vol 29, pp 1±26. Eds CV Mobbs & PR Hof. Basel:
Karger, 1998.
49 Sadow TF & Rubin RT. Effects of hypothalamic peptides on the
aging brain. Psychoneuroendocrinology 1992 17 293±314.
50 Holboer F, Spengler D & Heuser I. The role of corticotropin-
releasing hormone in the pathogenesis of Cushing's disease,
anorexia nervosa, alcoholism, affective disorders and dementia.
Progress in Brain Research 1992 93 385±417.
51 Rasmuson S, Nasman B, Eriksson S, Carlstrom K & Olsson T.
Adrenal responsivity in normal aging and mild to moderate
Alzheimer's disease. Biological Psychiatry 1998 43 401±407.
52 Ferrari E, Borri R, Pontiggia B, Camozzi D, Spina ML, Stabile M
et al. Changes of the adrenocortical secretory pattern in
physiological and pathological brain aging. Gerontologia 1998
36 Suppl 1.
53 Gottfries CG, Balldin J, Blennow K, Brane G, Karsson I, Regland B
& Wallin A. Regulation of the hypothalamic±pituitary±adrenal
axis in dementia disorders. Annals of the New York Academy of
Sciences 1994 746 336±344.
54 Gottfries CG, Balldin J, Blennow K, Brane G, Karlsson I,
Regland B et al. Hypothalamic dysfunction in dementia. Journal
of Neural Transmitters 43 (Suppl) 1994 203±209.
55 Keller-Wood ME & Dallman MF. Corticosteroid inhibition of
ACTH secretion. Endocrine Review 1984 5 1±24.
56 Aneko M & Hiroshige T. Fast, rate sensitive corticosteroid
negative feed-back during stress. American Journal of Physiology
1978 234 39±46.
57 Armanini D. Corticosteroid receptors in lymphocytes: a possible
marker of brain involution? Journal of Steroid Biochemistry and
Molecular Biology 1994 49 429±434.
58 De Kloet ER. Brain corticosteroid receptor balance and
homeostatic control. Frontiers in Neuroendocrinology 1991 12
95±164.
59 Miller AH, Spencer RL, Pulera M, Kang S, McEwen BS & Stein M.
Adrenal steroid receptor activation in rat brain and pituitary
following dexamethasone: implications for the dexamethasone
suppression test. Biological Psychiatry 1992 32 850±869.
60 De Kloet ER, Wallach G & McEwen BS. Differences in
corticosterone and dexamethasone binding to rat brain and
pituitary. Endocrinology 1975 96 598±609.
61 Kovacs KJ & Makara GB. Corticosterone and dexamethasone act
at different brain sites to inhibit adrenalectomy-induced
adrenocorticotropin hypersecretion. Brain Research 1988 474
205±210.
62 Morano MI, Vazquez DM & Akil H. The role of hippocampal
mineralocorticoid and glucocorticoid receptors in the hypotha-
lamo±pituitary±adrenal axis of the aged Fisher rat. Molecular
and Cellular Neuroscience 1994 5 400±412.
63 Guazzo EP, Kirkpatrick PJ, Goodyer IM, Shiers HM & Herbert J.
Cortisol, dehydroepiandrosterone (DHEA), and DHEA sulfate in
the cerebrospinal fluid of man: relation to blood levels and the
effects of age. Journal of Clinical Endocrinology and Metabolism
1996 81 3951±3960.
64 Walker MD, Henkin RI, Harlan AB & Caspar ATG. Distribution
of tritiated cortisol in blood brain CSF and other tissues of the
cat. Endocrinology 1971 88 224±232.
65 Belanger A, Candas B, Dupont A, Cusan L, Diamond P &
Gomez JL. Changes in serum concentrations of conjugated
and unconjugated steroids in 40 to 80-year old men.
Journal of Clinical Endocrinology and Metabolism 1994 79
1086±1090.
66 Carrol BJ, Feinberg M & Greden JF. A specific laboratory test for
diagnosis of melancholia: standardization, validation and
clinical utility. Archives of General Psychiatry 1981 38 15±22.
67 Ferrari E, Magri F, Dori D, Migliorati G, Nescis T, Molla G et al.
Neuroendocrine correlates of the aging brain in humans.
Neuroendocrinology 1995 61 464±470.
www.eje.org
328 E Ferrari and others
68 Sapolsky RM, Krey LC & McEwen BS. Prolonged glucocorticoid
exposure reduces hippocampal neuron number: implications for
aging. Journal of Neuroscience 1985 5 1221±1226.
69 Hornsby PJ. DHEA: a biologist's perspective. Journal of the
American Geriatrics Society 1997 45 1395±1401.
70 Guillemette C, Hum DW & Belanger A. Levels of plasma C19
steroids and 5-alpha-reduced C19 steroid glucuridones in
primates, rodents, and domestic animals. American Journal of
Physiology 1996 271 E348±-E353.
71 Endoh A, Kristiansen SB, Casson PR, Busters JE & Hornsby PJ.
The zona reticularis is the site of biosynthesis of dehydroepian-
drosterone and dehydroepiandrosterone sulfate in the adult
human adrenal cortex, resulting from its low expression of
3beta-hydroxysteroid dehydrogenase. Journal of Clinical Endocri-
nology and Metabolism 1996 81 3558±3565.
72 Labrie F, Belanger A, Cusan L, Gomez JL & Candas B. Major decline
in serum concentrations of adrenal C19 sex steroid precursors
and conjugated androgen metabolites during aging. Journal of
Clinical Endocrinology and Metabolism 1997 82 2396±2402.
73 Ohashi M, Kato K, Nawata H & Ibayashi H. Adrenocortical
responsiveness to graded ACTH infusion in normal young and
elderly human subjects. Gerontology 1986 32 43±51.
74 Kroboth PD, Salek FS, Pittenger AL, Fabian TJ & Frye RF. DHEA
and DHEAS: a review. Journal of Clinical Pharmacology 1999 39
327±348.
75 Thomas G, Frenoy N, Legrain S, Sebag-Lanoe R, Baulieu EE &
Debuire B. Serum dehydroepiandrosterone sulfate levels as an
individual marker. Journal of Clinical Endocrinology and Metabo-
lism 1994 79 1273±1276.
76 Parker CR, Mixon RL, Brissie RM & Grizzle WE. Aging alters
zonation in the adrenal cortex of men. Journal of Clinical
Endocrinology and Metabolism 1997 82 3898±3901.
77 Hornsby PJ. Biosynthesis of DHEAS by the human adrenal
cortex and its age-related decline. Annals of the New York
Academy of Sciences 1995 774 29±46.
78 Baulieu EE. Dehydroepiandrosteronea (DHEA): fountain of
youth? Journal of Clinical Endocrinology and Metabolism 1996
81 3147±3151.
79 Robel P & Baulieu EE. Dehydroepiandrosterone (DHEA) is a
neuractive neurosteroid. Annals of the New York Academy of
Sciences 1995 774 82±110.
80 Clarke D, Fearon U, Cunnigham SK & McKenna TJ. The
steroidogenic effects of beta-endorphin and joining peptide: a
potential role in the modulation of adrenal androgen produc-
tion. Journal of Endocrinology 1996 151 301±307.
81 Parker LN, Lifrak E, Gelfand R, Lee J, Shively T & Kaplan V.
Isolation, purification synthesis and binding of human adrenal
gland cortical androgen stimulating hormone. Endocrine Journal
1993 1 441±445.
82 Penhoat A, Sanchez P, Jaillard C, Langlois D, Begeot M & Saez JM.
Human pro-opiomelanocortin (79±96) a proposed cortical
androgen-stimulating hormone, does not affect steroidogenesis
in cultured human adult adrenal cells. Journal of Clinical
Endocrinology and Metabolism 1991 72 23±26.
83 Mellon SH, Shively JE & Miller WL. Human pro-opiomelano-
cortin (79±96) a proposed cortical androgen-stimulating
hormone, does not affect steroidogenesis in cultured human
foetal adrenal cells. Journal of Clinical Endocrinology and
Metabolism 1991 72 19±21.
84 Beer NA, Jakubowicz DJ, Beer RM & Nestler JE. Disparate effects
of insulin reduction with diltiazem on serum dehydroepian-
drosterone sulfate levels in obese, hypertensive men and women.
Journal of Clinical Endocrinology and Metabolism 1994 79 1079±
1081.
85 Peters JM, Zhou YC, Ram PA, Lee SS, Gonzales FJ & Waxman DJ.
Peroxisome proliferator-activated receptor alpha required for
gene induction by dehydroepiandrosterone-3beta-sulfate.
Molecular Pharmacology 1996 50 67±74.
86 Barrett-Connor E & Khawt KT. Absence of an inverse relation of
dehydroepiandrosterone sulfate with cardiovascular mortality in
www.eje.org
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
postmenopausal women [letter]. New England Journal of Medicine
1987 317±711.
87 Baulieu EE, Robel P, Vatier O, Haug A, Le Goascogne C &
Bourreau E. Neurosteroids: pregnenolone and dehydroepian-
drosterone in the rat brain. In Receptor±Receptor Interaction. A
New Intramembrane Integrative Mechanism, pp 89±104. Eds K
Fuxe & LF Agnati. Basingstoke: MacMillan, 1987.
88 Majewska MD. Neuronal actions of dehydroepiandrosterone.
Possible roles in the brain development, aging, memory, and
affect. Annals of the New York Academy of Sciences 1995 774
111±120.
89 Abadie J, Wright B, Correa G, Browne ES, Porter JR & Swec F.
Effects of dehydroepiandrosterone on neurotransmitter levels
and appetite regulation of the obese Zucker rat. Diabetes 1993
42 662±669.
90 Roberts E, Bologa L, Flood JF & Smith GE. Effects of
dehydroepiandrosterone and its sulfate on brain tissues in
culture and on memory in mice. Brain Research 1987 406 357±
362.
91 Thijnssen JHH & Nieuwenhuyse H. DHEA: a Comprehensive
Review. New York: The Parthenon Publishing Group, 1999.
92 Monnet FP, Mahe V, Robel P & Baulieu EE. Neurosteroids, via
sigma receptors, modulate the [3H]norepinephrine release
evoked by N-methyl-d-aspartate in the rat hippocampus. PNAS
1995 92 3774±3778.
93 Wolkowitz OM, Reus VI, Roberts E, Manfredi F, Chan T,
Ormitson S et al. Antidepressant and cognition-enhancing
effects of DHEA in major depression. Annals of the New York
Academy of Sciences 1995 774 337±339.
94 Stanton PK & Survey JM. Depletion of norepinephrine, but not
serotonin, reduces long-term potentiation in dentae gyrus of rat
hippocampal slices. Journal of Neuroscience 1986 5 2169±2176.
95 Sara SJ, Vankov A & Herve A. Locus coeruleus-evoked responses
in behaving rats. A clue to the role of noradrenaline in memory.
Brain Research Bulletin 1994 35 457±465.
96 Mao X & Barger SW. Neuroprotection by dehydroepiandroster-
one-sulfate: role of an NFkB-like factor. NeuroReport 1998 9
759±763.
97 Barger SW & Mattson MP. Induction of neuroprotective kappa B-
dependent transcription by secreted forms of the Alzheimer's beta
amyloid precursor. Molecular Brain Research 1996 40 116±126.
98 Barger SW. Tumor necrosis factor: the good, the bad and the
umbra. In Neuroprotective Signal Transduction, pp 163±183. Ed.
MP Mattson. Totawa, NJ, USA: Humana Press, 1997.
99 Rudman D, Shetty KR & Mattson DE. Plasma dehydroepian-
drosterone sulfate in nursing home man. Journal of the American
Geriatric Society 1990 38 421±427.
100 Morales AJ, Nolan JJ, Nelson JC & Yen SSC. Effects of
replacement dose of dehydroepiandrosterone in men and
women of advancing age. Journal of Clinical Endocrinology and
Metabolism 1994 78 1360±1367.
101 Schneider EC & Reed JD. Life extension. New England Journal of
Medicine 1985 312 1159±1168.
102 Ravaglia G, Morini P, Forti P, Maioli F, Boschi F, Mernardi M et al.
Anthropometric characteristics of healthy Italian nonagenarians
and centenarians. British Journal of Nutrition 1997 77 9±17.
103 Rotter JI, Wong FL, Lifrak ET & Parker LN. A genetic component
to the variation of dehydroepiandrosterone sulfate. Metabolism
1985 34 731±736.
104 Legrain S, Berr C, Frenoy N, Gourlet V, Bebuire B & Baulieu EE.
Dehydroepiandrosterone sulfate in a long-term care aged
population. Gerontology 1995 41 343±351.
105 Nasman B, Olsson T, Viitanen M & Carlstrom K. A subtle
disturbance in the feedback regulation of the hypothalamic±
pituitary±adrenal axis in the early phase of Alzheimer's disease.
Psychoneuroendocrinology 1995 20 211±220.
106 Sunderland T, Merril CR, Harrington MG, Lawlor BA,
Molchan SE, Martinez R et al. Reduced plasma dehydroepian-
drosterone concentrations in Alzheimer's disease. Lancet 1989 ii
570.
EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 144
107 Leblhuber F, Windhanger E, Reisecker F, Steinparz FX & Dienstl V.
Dehydroepiandrosterone sulfate in Alzheimer's disease. Lancet
1990 i 449.
108 Leblhuber F, Windhager F, Steinparz FX, Dientsl E, Reisecker F &
Rittmannsberger H. Dehydroepiandrosterone-sulfate (DHEA-S)
serum levels in normal aging and Alzheimer's disease (AD). Age
1991 14 116±118.
109 Schneider LS, Hinsey M & Lyness S. Plasma dehydroepiandros-
terone in Alzheimer's disease. Biological Psychology 1992 31
205±208.
110 Regelson W & Kalimi M. Dehydorepiandrosterone (DHEA) ± the
multifunctional steroid. II Effects on CNS, cell proliferation,
metabolic and vascular, clinical and other effects. Mechanism of
action. Annals of the New York Academy of Sciences 1994 719
564±575.
111 Dobbie JW. Adrenocortical nodular hyperplasias in ageing
adrenal. Journal of Pathology 1969 99 1±18.
112 Parker CR. Adrenal function in aging. Current Opinion in
Endocrinology and Diabetes 1999 3 210±215.
113 Haus H, Nicolau G, Lakatua DJ, Sackett-Lundeen L & Petrescu E.
Circadian rhythm parameters of endocrine functions in elderly
subjects during the seventh to the ninth decade of life.
Chronobiology 1989 16 331±352.
114 McEwen BS, de Leon MJ & Meaney MJ. Corticosteroids, the aging
brain and cognition. Trends in Endocrinology and Metabolism
1999 10 92±96.
115 Miller AH, Sastry G, Speranza AJ, Lawlor BA, Mohs RC, Ryan TM
et al. Lack of association between cortisol hypersecretion and
nonsuppression on the DST in patients with Alzheimer's disease.
American Journal of Psychiatry 1994 151 267±270.
116 McEwen BS, Albeck D, Cameron H, Chao HM, Gould E &
Hastings N. Stress and the brain: a paradoxical role for adrenal
steroids. In Vitamins and Hormones, pp 371±402. Ed. GD
Litwack. New York: Academic Press, 1995.
117 Galea LA & McEwen BS. Sex and seasonal differences in the rate
of cell proliferation in the dentate gyrus of adult meadow voles.
Neuroscience 1999 83 955±964.
118 Gould E, McEwen BS, Tanapat P, Galea LAM & Fuchs E.
Neurogenesis in the dentate gyrus of the adult tree shrew is
regulated by psychosocial stress and NMDA. Journal of
Neuroscience 1997 17 2492±2498.
119 Magri F, Terenzi F, Ricciardi T, Fioravanti M, Solerte SB, Stabile M
et al. Association between changes in adrenal secretion and
Aging and adrenal steroids 329
cerebral morphometric correlates in normal aging and senile
dementia. Dementia and Geriatric Cognitive Disorders 2000 11
90±99.
120 De Leon MJ, Golomb J, George AE, Convit A, Tarshish CM,
McRae T et al. The radiologic prediction of Alzheimer's disease:
the atrophic hippocampal formation. American Journal of
Neuroradiology 1993 14 897±906.
121 Double KL, Halliday GM, Kril JJ, Harasty JA, Cullen K, Beooks WS
et al. Topography of brain atrophy during normal aging and
Alzheimer's disease. Neurobiology of Aging 1996 17 513±521.
122 Scheff SW, Scott SA & DeKoski ST. Quantification of synaptic
density in the septal nuclei of young and aged Fischer 344 rats.
Neurobiology of Aging 1991 12 3±12.
123 McEwen BS. Stress and the aging hippocampus. Frontiers in
Neuroendocrinology 1999 20 49±70.
124 Yoo A, Harris J & Dubrovsky B. Dose±response study of
dehydroepiandrosterone sulfate on dentate gyrus long-term
potentiation. Experimental Neurology 1996 137 151±156.
125 Landfield PW & Elridge JC. The glucocorticoid hypothesis of
brain aging and neurodegeneration: recent modifications. Acta
Endocrinologica 1991 125 54±64.
126 Landfield PW & Eldrige JC. The glucocorticoid hypothesis of age-
related hippocampal neurodegeneration: role of dysregulated
intraneuronal calcium. Annals of the New York Academy of
Sciences 1994 746 308±321.
127 Lupien SJ & McEwen BS. The acute effects of corticosteroids on
cognition: integration of animal and human model studies.
Brain Research Review 1997 24 1±27.
128 Qu T, Uz T & Manev H. Inflammatory 5-LOX mRNA and protein
are increased in brain of aging rats. Neurobiology of Aging 2000
21 647±652.
129 Zhang W, Lei ZM & Rao ChV. Immortalized hippocampal cells
contain functional luteinizing hormone/human chorionic
gonadotropin receptors. Life Science 1999 65 2083±2098.
130 Steinhilberg D, Brungs M, Wertz O, Wiesenberg I, Danielsson C,
Kahlen JP et al. The nuclear receptors for melatonin repress 5-
lipoxygenase gene expression in human B lymphocytes. Journal
of Biological Chemistry 1995 270 7037±7040.
Received 15 May 2000
Accepted 18 December 2000
www.eje.org