The amyloidogenic potential and behavioral correlates
C Catania1,5, I Sotiropoulos1,5, R Silva2, C Onofri1, KC Breen3,4, N Sousa2and OFX Almeida1
1Max Planck Institute of Psychiatry, Munich, Germany;2Life and Health Science Research Institute, University of Minho, Braga,
Portugal;3Alzheimer’s Disease Research Centre, University of Dundee, Dundee, UK and4Parkinson’s Disease Society,
Observations of elevated basal cortisol levels in Alzheimer’s disease (AD) patients prompted
the hypothesis that stress and glucocorticoids (GC) may contribute to the development and/or
maintenance of AD. Consistent with that hypothesis, we show that stress and GC provoke
misprocessing of amyloid precursor peptide in the rat hippocampus and prefrontal cortex,
resulting in increased levels of the peptide C-terminal fragment 99 (C99), whose further
proteolytic cleavage results in the generation of amyloid-b (Ab). We also show that exogenous
Ab can reproduce the effects of stress and GC on C99 production and that a history of stress
strikingly potentiates the C99-inducing effects of Ab and GC. Previous work has indicated a
role for Ab in disruption of synaptic function and cognitive behaviors, and AD patients
reportedly show signs of heightened anxiety. Here, behavioral analysis revealed that like
stress and GC, Ab administration causes spatial memory deficits that are exacerbated by
stress and GC; additionally, Ab, stress and GC induced a state of hyperanxiety. Given that the
intrinsic properties of C99 and Ab include neuroendangerment and behavioral impairment, our
findings suggest a causal role for stress and GC in the etiopathogenesis of AD, and
demonstrate that stressful life events and GC therapy can have a cumulative impact on the
course of AD development and progression.
Molecular Psychiatry (2009) 14, 95–105; doi:10.1038/sj.mp.4002101; published online 2 October 2007
Keywords: Alzheimer’s disease; amyloid precursor protein; amyloid-b; glucocorticoids; memory;
Amyloid-b (Ab) peptide is generated under normal
physiological conditions, but its overproduction and
accumulation give rise to senile plaques that, together
with aggregations of abnormally hyperphosphory-
lated t protein, constitute the main neuropathological
hallmarks of Alzheimer’s disease (AD). However, the
amount of deposited Ab correlates poorly with the
degree of cognitive impairment in AD patients,1and
central administration of Ab in nongenetically modi-
fied rodents results in cognitive impairments within a
relatively short period and before amyloid plaques
become detectable.2These observations suggest a
pivotal role for soluble nonaggregated Ab in the early
stages of the disease; they are supported by data
showing that soluble Ab can acutely disrupt synaptic
function and cognitive and emotional behavior.3–6
Ab peptide is produced through the sequential
proteolytic cleavage of amyloid precursor protein
(APP) by a-, b- and g-secretases; the last secretase is
a complex of at least four proteins, with nicastrin
making up the largest portion. Nicastrin may be
viewed as the gatekeeper of the g-secretase complex
because of its essential role in the recognition of
g-secretase substrates.7Cleavage of APP by a-secretase
precludes the generation of Ab, whereas b-secretase
(b-site APP-cleaving enzyme; BACE)-mediated clea-
vage generates the so-called C-terminal fragment 99
(C99); subsequent proteolysis of C99 by g-secretase
results in the generation of amyloid peptides (for
review, see Bayer et al.8). While the neurotoxic
potency of Ab is well known,9–11other studies have
demonstrated the neurotoxic and cognition-impairing
potential of C99.12–15Furthermore, in a vicious cascade
that characterizes AD, Ab interacts with its protein
precursor (APP), an interaction that is suggested to
contribute to the mechanism of Ab neurotoxicity.16,17
Observations that a large percentage of AD patients
display hypercortisolemia18,19suggest that glucocorti-
coids (GC) and stress may contribute to the develop-
ment or maintenance of AD. This view recently
gained support from studies in transgenic mouse
models of AD in which stress or GC exacerbated AD-
like neuropathology.20,21Indications of a link between
stress/GC and AD may also be inferred from previous
Received 29 March 2007; revised 19 July 2007; accepted 31 July
2007; published online 2 October 2007
Correspondence: Dr OFX Almeida, Max Planck Institute of
Psychiatry, Kraepelinstrasse 2-10, Munich D-80804, Germany.
5These authors contributed equally to this study
Molecular Psychiatry (2009) 14, 95–105
& 2009 Nature Publishing Group All rights reserved 1359-4184/09 $32.00
work in which a causal role for elevated GC levels and
stressors in cognitive impairment in humans and
animals was demonstrated,22,23GC-induced dendritic
atrophy and synaptic loss are thought to underlie the
GC-associated behavioral disruption.24,25
The present study focused on examining whether
stress or elevated GC levels can drive APP metabolism
middle-aged rats, and whether stressful life events
can influence the initiation/maintenance of AD-like
pathology. Our studies, which focused on the hippo-
campus and prefrontal cortex (PFC) because they rank
among the first brain areas to show AD-like neuro-
pathology,26disclosed that chronic stress and GC
similarly promote APP processing along the amylo-
idogenic pathway. These results provide a tentative
mechanism through which stress and/or GC exert
their effects on cognitive and emotional behavior. Of
note is the growing recognition of the importance
of emotional state in the regulation of cognition.27–29
In addition, it is pertinent to mention that stress-
related psychiatric disorders (for example, anxiety
and major depression) have been identified as a risk
for developing AD.30,31
Materials and methods
Male Wistar rats (Charles River, Barcelona, Spain),
aged 14 months, were used, in compliance with the
European Union Council’s Directive (86/609/EEC)
and local regulations on animal welfare. Animals
were housed 4–5 per cage under standard environ-
mental conditions (temperature 221C relative humi-
dity 70%; 12h light:dark cycle; ad libitum access to
food and drinking solution).
Treatments and their biological efficacy
Rats were allocated to one of two main treatment
groups: stressed and unstressed. Stressed animals
were exposed to a chronic, unpredictable stress
paradigm;32for 4 weeks, while their unstressed
counterparts were held undisturbed, under standard
laboratory conditions. The stress paradigm involved
random application of one of the following stressors,
daily: (1) hypertonic saline (9% NaCl; 1ml per 100g)
i.p. injection, (2) overcrowding for 1h, (3) placement
in a confined environment (30min) and (4) placement
on a vibrating/rocking platform (1h). After 5 days,
subgroups of stressed or unstressed animals were
given constant-rate i.c.v. infusions of soluble Ab1–40
(American Peptide Co., Inc., Sunnyvale, CA; 4.2nmol
Ab1–40 per 200ml; 0.5mlh?1) or vehicle (distilled
water) over a period of 14 days; half of each of these
subgroups additionally received daily s.c. injections
of a synthetic GC (dexamethasone (DEX; Fortecortin
Merck, Darmstadt, Germany); 300mgkg?1body weight
(BW) as an oily suspension in sesame oil (Sigma,
St Louis, MI, USA)). For the i.c.v. infusions, rats
were equipped with s.c. osmotic minipumps (Alzet
2002 minipumps; Alza Corp., Mountain View, CA,
USA) and cannulae (Alzet Brain Infusion Kit) that
were implanted in the left lateral ventricle using the
following stereotaxic coordinates: AP, ?1.0mm; DV,
?2.5mm and ML, þ1.5mm (right) with bregma as
a reference (Paxinos and Watson33); pump and
cannulae implantation was done under pentobarbital
(50mgkg?1) anesthesia. At the time of killing, mini-
pumps were removed and checked for patency and
functionality. Each treatment subgroup comprised
Efficacy of the stress paradigm was gauged by
measuring daytime serum corticosterone (CORT)
levels (Corticosterone RIA kit, ICN, Costa Mesa, CA,
USA), BWs and relative thymus weights. CORT levels
were significantly elevated in stressed vs nonstressed
animals (384.6765.9 and 54.774.8ngml?1, respec-
tively), and thymus weights at autopsy were consistent
with hypercorticalism in the stressed group (control,
CON: 7.670.9 vs stressed: 4.370.4 vs mgkg?1BW).
Whereas nonstressed animals showed a net gain in
BW (6.171.3g), stressed animals showed a significant
loss of body mass (?12.172.1g). GC-exposed rats
showed biometric changes that were similar to those
observed in the stressed group, whereas none of the
parameters monitored differed significantly between
CON and stressed animals after the superimposed
treatment with Ab.
Animals were rapidly killed and their brains were
immediately excised and divided along the midline.
One half of each brain was immersed in 4%
p-formaldehyde (2 days), embedded in paraffin and
saved for histochemical analysis. The PFC and
hippocampus were dissected out of the other half of
the brain, snap-frozen in liquid nitrogen and stored at
?801C until subsequent biochemical analyses.
Western blot analysis
Frozen hippocampal and PFC tissues were homo-
genized in lysis buffer (100mM Tris-HCl, 250mM
NaCl, 1mM EDTA, 5mM MgCl2, 1% NP-40, Complete
Protease Inhibitor (Roche, Mannheim, Germany)
and Phosphatase Inhibitor Cocktails I and II (Sigma,
St Louis, MO)) using a Dounce glass homogenizer;
extracts were cleared by centrifugation (14000g) and
their protein contents were estimated by the Lowry
assay. Lysates, in Laemmli buffer (250mM Tris-HCl,
pH 6.8, containing 4% sodium dodecyl sulfate, 10%
glycerol, 2% b-mercaptoethanol and 0.002% bromo-
phenol blue), were thereafter electrophoresed on 8
or 10% acrylamide gels, and transferred onto nitro-
cellulose membranes (Protran BA 85, Schleicher &
Schuell, Dassel, Germany). Membranes were blocked
in Tris-buffered saline containing 5% nonfat milk
powder and 0.2% Tween-20 before incubation with
the following antibodies: anti-APP369 (1:5000; kindly
provided by Dr Sam Gandy), anti-BACE-1 (1:500;
ProSci Inc., Poway, CA, USA), anti-nicastrin (1:5000;
Sigma) and anti-actin (1:2000; Chemicon, Temecula,
CA, USA) or anti-a-tubulin (1:2000; Calbiochem,
C Catania et al
San Diego, CA, USA). Antigens were revealed
by enhanced chemiluminescence (Amersham Bio-
sciences, Freiburg, Germany) after incubation with
appropriate horseradish peroxidase–immunoglobulin
G conjugates (Amersham Biosciences); blots were
scanned and quantified using TINA 3.0 bioimaging
software (Raytest, Straubenhardt, Germany) after
ascertaining linearity. All values were normalized
and expressed as percentages of control. To distin-
guish between mature and immature isoforms of APP,
tissue lysates were digested with endoglycosidase F
(Roche), according to the manufacturer’s instructions.
APP369 antibody using a mixture of Protein A and
G beads (Roche) before analysis by electrophoresis on
Tris-Tricine gradient gels (KMF Laborchemie, Loh-
mar, Germany) and immunoblotting with an APP
antibody (OPA1-01132; Affinity Bioreagents, Golden,
In situ hybridization
Paraffin sections (8mm) were deparaffinized (see
above), and fixed in 4% p-formaldehyde in phos-
phate-buffered saline (PBS) containing 0.1% diethyl-
pyrocarbonate (DEPC; Sigma). After washing (PBS–
DPEC), sections were acetylated (0.1 M triethanol-
amine, 0.25% acetic anhydride in DPEC), delipidated
in chloroform (5min), dehydrated through a graded
series of ethanols and air dried. A 40-mer oligo-
nucleotide probe (GCTGGCTGCCGTCGTGGGAACTC
GGACTACCTCCTCCACA) was used to detect APP695-
KPI.34,35Slides were hybridized with an antisense or
sense (control)35S-dATP-labeled oligoprobe (106c.p.m.
per slide; overnight at 421C), after which they were
stringently washed (1? saline sodium citrate (SSC)
for 15min at 551C). After sequential immersion 1?
SSC, H2O, 65% ethanol and 95% ethanol, slides were
air dried and dipped in photographic emulsion (1:1
Kodak NTB2 in distilled water) and exposed for 1
month. Sections were then developed, counterstained
(toluidine blue) and viewed under bright field or
polarized light to view cells and silver grains,
respectively. Hybridization signal on captured images
(all slides) was scored by two independent observers
(unaware of the treatment details) on a scale of 0–5; the
raters followed common scoring criteria and the results
shown in Table 1 are the means of the scores obtained
by the individual raters. Representative in situ hybridi-
zation (ISH) images are shown in Figure 1.
Assays of cognitive performance and emotionality
Hippocampus-dependent spatial reference memory
was assessed using the Morris water maze over 4 days
(four trials per day), as previously described.25Data
were recorded using a video-tracking system (View-
point, Champagne au Mont d’Or, France).
Emotional state was evaluated by monitoring
locomotion, exploratory behavior and anxiety. Loco-
motor and exploratory behavior were assessed (total
distances traveled and number and duration of
rearings) online in an open-field arena (43.2?
43.2cm; transparent acrylic walls and white floor;
protein levels of APP, C99, BACE and nicastrin in the hippocampus and PFC
Overall statistical analysis of treatment (stress history, the corticosteroid milieu, Ab and stressþAbþGC) effects on
Hippocampus (d.f.6,36)APPC99BACE Nicastrin
(F=132.2) (F=130.3) (F=25.07)
Stress history in (AbþGC)-treated
q=0.0336 (NS) q=1.608 (NS) q=30.03
q=2.903 (NS) q=1.293 (NS)
q=3.424 (NS) q=15.791 q=4.472
q=0.998 (NS) q=0.754 (NS)
Stress history in (AbþGC)-treated
Abbreviations: Ab, b-amyloid; APP, amyloid precursor peptide; BACE, b-site APP cleaving enzyme; C99, C-terminal fragment
99; GC, glucocorticoids; NS, not significant; PFC, prefrontal cortex.
Analyses were based on multiple one-way ANOVAs on ranks and Tukey’s all pairwise comparison tests.
C Catania et al
MedAssociates Inc., St Albans, VT, USA) over a
period of 5min. Levels of anxiety behavior were
evaluated in an elevated plus maze (EPM; a black
polypropylene plus-shaped platform elevated 72.4cm
above the floor with two open arms (50.8?10.2cm)
and two closed arms (50.8?10.2?40.6cm); Med-
Associates Inc.). The EPM test was carried out under
white light and animals were placed in the maze for a
total of 5min; times spent in the open vs closed arms
were recorded online, and data were computed to
yield the relative amount of time spent in the open
arms of the maze; the number of open and closed arms
entries were also recorded.
Results are expressed as group means7s.e.m. Data
were evaluated for their statistical significance using
one-way analysis of variance (ANOVA), followed
by Tukey’s all-pairwise multiple comparison test.
To avoid introducing confounds from the use of
multiple one-way ANOVAs on ranks, we applied the
Bonferroni correction (adjusting the P-values to
0.0125). For the proteins of interest in the hippocam-
pus and PFC, multiple-way ANOVAs were performed
to examine the effects of stress history, corticosteroid
milieu (GC), Ab and stressþAbþGC. Statistical ana-
lyses were conducted using SPSS (Chicago, IL, USA)
and SigmaStat (Systat, San Jose, CA, USA) software
packages; differences were considered to be signi-
ficant if P<0.05.
Overall statistical analysis of treatment effects
The results of an overall analysis of the effects of
stress history, the corticosteroid milieu (GC), Ab and
stressþAbþGC treatment on protein levels of APP,
C99, BACE and nicastrin in the hippocampus and
PFC are shown in Table 1.
Chronic stress drives APP processing along the
In light of clinical reports of GC hypersecretion in AD
patients,18,19and that stress exacerbates amyloid
production in transgenic mouse models of AD,20,21it
was of interest to examine whether stress stimulates
the amyloidogenic pathway in normal, middle-aged
rats. To this end, rats were exposed to a chronic
unpredictable stress protocol32before analysis of APP,
C99, BACE-1 and nicastrin mRNA and protein levels
in the hippocampus and PFC.
Exposure to stress resulted in a slight increase in
the expression levels of APP mRNA in the CA1 and
CA3 subfields of the hippocampus (see Table 2 for
levels of immature and mature APP protein in either
the hippocampus (Figure 2a) or PFC (Figure 2b).
However, exposure to stress resulted in a significant
increase in levels of the C99 fragment of APP in both
brain regions, with the magnitude of changes being
greater in the hippocampus than the PFC. Consistent
with the alterations in the levels of C99, stress was
associated with increased hippocampal and PFC
levels of BACE-1 and nicastrin. Thus, stress was
shown to drive APP processing along the amyloido-
butdid not alter
Proamyloidogenic actions of GC
Glucocorticoid secretion represents the major physio-
logical response to stress. Therefore, we next asked
whether GC mediate the above-reported effects of
stress on APP metabolism. This question was addres-
sed by administering rats with daily injections of the
synthetic glucocorticoid receptor (GR) agonist, DEX,
for a period of 14 days.
As shown in Table 2, GC treatment resulted in
increased APP mRNA levels (see semiquantitative
results in Table 2) and hippocampal and PFC levels of
APP and C99 (Figures 3a and b). BACE-1 expression
was, however, only significantly elevated in the
hippocampus (Figure 3a); nicastrin levels were not
altered by GC treatment (Figures 3a and b), suggesting
that signals, other than GC, are responsible for medi-
ating the above-reported ability of stress to increase
nicastrin levels (cf. Ni et al.36).
Exogenous Ab triggers APP misprocessing
Ab infusion is widely used to experimentally model
aspects of AD pathology in animals.37Since Ab is
images from hippocampal CA3 cells are shown. After ISHH,
silver grains were revealed by photo-emulsion dipping;
sections were counterstained with toluidine blue.
APP mRNA detected by ISHH. Representative
senting APP mRNA in the CA1 and CA3 subfields of the
hippocampus of rats exposed to various treatments involving
exposure to stress, Ab, GC or AbþGC, or a combination
Semiquantitative evaluation of silver grains repre-
Amyloid beta (Ab)
þ þ þ þ
þ þ þ þ þ
þ þ þ
þ þ þ þ
þ þ þ þ
þ þ þ
In situ hybridization histochemistry was performed as
described in ‘Materials and methods’, using a 40-mer oligo
probe designed to recognize APP695-KPI.
C Catania et al
known to induce AD pathology and to stimulate its
own production in vitro,38we were interested to
examine how Ab influences APP metabolism in vivo.
When middle-aged rats were given constant infu-
sions of Ab into one lateral ventricle (Frautschy
et al.39), there was a marked increase in APP mRNA
signal in the hippocampi of Ab-infused rats (semi-
quantitative data in Table 2). At the same time, Ab
treatment produced significant reductions in the
levels of immature and mature APP in the hippo-
campus (Figure 4a) and PFC (Figure 4b), changes that
were paralleled by increases in C99, BACE-1 and the
mature form of nicastrin; the Ab-induced increases in
C99 expression were twice greater in the hippocam-
pus than in the PFC. Together, this set of findings
metabolism toward amyloidogenesis. Hippocampal (a) and
prefrontal cortex (b) lysates from chronically stressed
middle-aged rats were analyzed for their contents of APP,
C99 (C-terminal fragment of APP), b-site APP-cleaving
enzyme (BACE)-1 and nicastrin, by immunoprecipitation
and western blotting. Representative blots from each brain
region are shown alongside the respective numerical data;
the latter are based on optical density evaluations, normali-
zed against actin or tubulin and are depicted relative
to control (CON) values as means7s.e.m. (n=6–8 rats).
Asterisks indicate Pp0.05 vs CON values. Stressors were
applied over a period of 4 weeks. In both brain areas,
nicastrin was upregulated by stress, indicating the potential
for C99 to be further processed to amyloid peptide. In
western blots, ‘m’ and ‘i’ refer to mature and immature
forms of the respective proteins.
Stress drives amyloid precursor protein (APP)
ment 99 (C99) production without upregulating nicastrin
levels. Shown are the changes in amyloid precursor protein
(APP), C99, b-site APP-cleaving enzyme (BACE)-1 and
nicastrin expression in the hippocampus (a) and PFC
(b) of middle-aged rats that had been treated with DEX, a
synthetic GC, for 14 days. Note that nicastrin levels were not
increased in either brain area. Means7s.e.m. are depicted
(n=6–8 rats). The data emerged from semiquantitative
assessment (optical densities) of immunoreactive bands,
normalized against actin or tubulin. Asterisks indicate
Pp0.05 vs control values. In western blots, ‘m’ and ‘i’ refer
to mature and immature forms of the respective proteins.
Glucocorticoids (GC) stimulate C-terminal frag-
C Catania et al
demonstrates that Ab treatment triggers BACE-1-
mediated APP misprocessing into C99 which can
be potentially cleaved by g-secretase into amyloid
peptides. These results show that the actions of
Ab closely resemble those of stress and GC.
Stress history exacerbates APP misprocessing
It was recently shown in transgenic mouse models
of AD that chronic stress potentiates Ab deposition
and induces cognitive deficits,20and that elevated
GC exacerbate amyloidogenesis.21Given that the
organism experiences stressful events throughout its
lifetime, and aged individuals tend to show exagge-
rated GC secretory responses to stress,22we here
examined the sequential effects of stress and GC
on AD-like pathology by simultaneously treating
previously stressed and nonstressed rats with Ab
(i.c.v. infusions) and/or GC (s.c. injections). Multiple-
way ANOVA analysis revealed that stress history
plays a significant role in determining subsequent
treatment effects on responses of the APP processing
Pp0.003; mature APP: F=11.1, Pp0.0001; C99:
F=224.2, Pp0.0001; immature nicastrin: F=3.5,
Pp0.043; mature nicastrin: F=11.9, Pp0.001) and
PFC (immature APP: F=8.9, Pp0.001; mature APP:
F=10.7, Pp0.0001; C99: F=11.5, Pp0.0001; BACE:
6.6, Pp0.004; immature nicastrin: F=8.9, Pp0.001;
mature nicastrin: F=10.7, Pp0.0001). Briefly, while
stress did not influence the effects of subsequently
applied Ab on APP metabolism (data not shown), the
amyloidogenic effects of AbþGC were further accen-
tuated in rats that had been previously exposed to
stress. Specifically, as shown in Figure 5, concomitant
treatment with Ab and GC triggered APP misproces-
sing (into C99) in both the hippocampus and PFC
of nonstressed rats, an effect that was significantly
exacerbated in previously stressed animals. BACE-1
levels in AbþGC-treated nonstressed and stressed
rats revealed increases that were only significant in
the PFC (Figure 5b). Mature nicastrin levels were
significant in both hippocampus and PFC, indicating
the greater likelihood of C99 processing along the
amyloidogenic pathway. In summary, the results
reported in this section demonstrate that a history of
stress increases vulnerability to subsequent exposures
to GC and Ab.
Impairments in learning and memory are the primary
behavioral manifestations of AD. Accordingly, it was
considered important to evaluate the extent to which
the various treatments used in the present study
impacted on cognitive performance. Examination
of escape latencies in the Morris water maze revealed
that stress and GC, as well as Ab, impair spatial
reference memory to similar extents (Table 3). A ten-
dency for further impairment of spatial reference
memory was observed when Ab and GC were applied
concomitantly, and even more pronounced deficits
were seen in animals that had been exposed to stress
before receiving AbþGC (Table 3).
Finally, in light of the evidence for interactions
between emotion and cognition27–29and evidence that
AD patients show hyperanxiety,30,31we monitored
the emotional state of animals subjected to the various
treatment regimens described above. Locomotor and
administered by chronic, constant-rate infusion into one
C-terminal fragment 99 (C99) production and, concomitantly,
in nicastrin levels in the hippocampus (a) and PFC (b),
suggesting the potential for further processing of C99 to
amyloid peptides(s). The treatment resulted in significant
reductions in amyloid precursor protein (APP) levels, most
likely due to its rapid processing into C99. Means7s.e.m.
(n=6–8 rats) are shown. Data represent optical densities of
immunoreactive bands, normalized against actin or tubulin.
Asterisks indicate Pp0.05 vs control values. In western
blots, ‘m’ and ‘i’ refer to mature and immature forms of the
Amyloid-b (Ab) stimulates amyloidogenesis. Ab
C Catania et al
exploratory behavior were assessed in an open-field
setup, and anxiety-like behavior in an EPM; the
results are summarized in Table 3. Stressed and
GC-treated rats were found to be more anxious than
controls; animals given Ab also displayed increased
signs of anxiety when compared to controls. Un-
expectedly, animals that were exposed to stress before
being treated with either GC or the combination of
Ab and GC appeared to be resistant to the anxiogenic
effects of GC and Ab (cf. data for Ab and GC in
nonstressed animals). Importantly,
phenotype cannot be attributed to locomotor deficits
as none of the experimental procedures produced
differences in total distances traveled in the open
field Table 3 or in the number of closed arm entries in
the EPM (data not shown). Interestingly, there were
impairments in exploratory behavior as assessed by
rearing activity. Briefly, stressed and GC-treated rats
displayed fewer rearings than controls. Ab-treated
rats also displayed decreased exploratory behavior
when compared to controls.
Stress and GC secretion may be regarded as insepar-
able phenomena since GC secretion represents a
had been preexposed (4 weeks) to a chronic unpredictable stress paradigm responded to the coadministration of AbþGC
(14 days) with markedly increased levels of C-terminal fragment 99 (C99), b-site APP-cleaving enzyme (BACE)-1 and
nicastrin in the hippocampus (a) and PFC (b). Means7s.e.m. are depicted (n=6–8 rats). Numerical data refer to optical
density readings from immunoreactive bands, normalized against actin. Asterisks indicate Pp0.05. In western blots, ‘m’ and
‘i’ refer to mature and immature forms of the respective proteins.
Stress history exacerbates the amyloidogenic effects of amyloid-b (Ab)þ glucocorticoids (GC) treatment. Rats that
C Catania et al
primary physiological response to stress. Both stress
and GC can induce neurodegenerative changes in the
hippocampus and PFC.23,40–45These brain regions
express GRs;46they are important for cognition and
are affected early in the development of AD.47Clinical
reports of hypercortisolism in a significant number of
AD patients18,19suggest a causal role for GC in AD, a
view supported by studies in laboratory rodents that
have shown that GC participate in the regulation of
APP levels.48,49Using middle-aged rats, we now show
that (1) stress and GC can drive APP metabolism
toward amyloidogenesis, (2) a history of stress, invol-
ving hypersecretion, biases APP processing toward
the amyloidogenic pathway and (3) that the proamy-
loidogenic effects of stress/GC are more pronounced
in the hippocampus than in the PFC, a finding that is
consistent with the temporal pattern of neuropatho-
logical events in the AD brain.47In addition, we
show that Ab induces APP misprocessing in a manner
similar to that observed after exposure to stress or
elevated GC levels. Lastly, we show that stress,
GC and Ab produce similar, as well as additive,
impairments in cognition and emotional behavior.
A considerable body of evidence indicates that the
amyloidogenic (mis)processing of APP plays a central
role in the neurodegenerative changes and behavioral
deficits that characterize AD.26Misprocessing of APP
commences with its endocytic cleavage by b-secretase
(BACE) to yield the bC-terminal fragment (bCTF),
C99. Subsequent cleavage of C99 by g-secretase
results in the generation of Ab peptides. While much
emphasis has been placed on examining the role and
mechanisms of action of Ab in AD, there is growing
evidence that C99 also makes a substantial contribu-
tion to AD pathology. In this context, it is worth
noting that C99 has intrinsic neurotoxic properties12
and causes synaptic degeneration;13furthermore, C99
can impair long-term potentiation14and cognition.15
The results from this study show that stress
upregulates steady-state APP mRNA levels without
inducing any change in APP protein levels. These
findings, together with our observation that stress
causes a significant increase in nicastrin levels,
suggest that stress promotes APP misprocessing. On
the other hand, our GC treatment pardigm seemed to
be a less potent inducer of APP misprocessing insofar
that while it led to increased expression of both APP
mRNA and protein, it did not elevate nicastrin levels.
Consistent with this interpretation are our findings
that both, stress and GC, upregulate BACE-1 levels as
well as those of its cleavage product, C99. A gluco-
corticoid response element in the promoter region of
the APP and BACE genes has been described,50(DK
Lahiri, personal communication), making it likely
that GC and GR mediate the regulatory actions of
stress on APP and BACE-1 expression. BACE-1 is
essential for C99 production, and previous studies
have shown that even slight increases in this enzyme
result in the generation of high levels of Ab.51–53
Although both GC and stress stimulate APP proces-
sing to C99, we found that whereas stress significantly
Tests of spatial memory and emotionality (exploratory and anxiety-like behavior)
Spatial memory (Morris water maze)
Mean escape latency (s; trials 2–4)
Locomotor and exploratory behavior (open field)
Total distance (cm)
No. of rearings/extensions
Anxiety level (elevated plus maze)
Time in open arms (%)
Abbreviations: Ab, b-amyloid; CON, control; DEX, dexamethasone.
Data shown are means7s.e.m. (n=6 rats per treatment group).
*Pp0.05 vs CON;zPp0.05 vs stressþAb.
C Catania et al
increases nicastrin levels, GC actions are limited to
the point of C99 production. Given the adverse
actions of C99 on neuronal function and survival, as
well as behavior,12–15the potential importance of
GC-induced increases in C99 production for AD-like
pathology is nevertheless significant.
Nicastrin, and in particular its glycosylated mature
form, is an essential component of the g-secretase
complex;54,55because of its role in the recognition and
presentation of substrate (C99) to presenilin-1 and -2,
nicastrin is crucial for the generation of Ab.7,56Our
observations that stress (but not GC) increases
nicastrin levels, especially those of the mature form,
suggest that stress can recruit additional mechanisms
that allow it to carry C99 metabolism through to
Ab production (cf. Ni et al.36) On the other hand,
the possibility that GC can eventually generate Ab
(for example, by acting on other members of the
g-secretase complex) cannot be excluded; technical
limitations of existing assays unfortunately precluded
direct measurements of rat Ab in the present study.
Since increased Ab production is causally related
to AD pathology, exogenous administration of Ab
peptides into the brain has been extensively used in
normal rodents and primates to reproduce the neuro-
patholgical and behavioral features of early stage
AD.37,39,57Using the Ab i.c.v. infusion paradigm in
middle-aged rats, we here observed that Ab upregu-
lates APP mRNA levels, but reduces APP protein
levels; at the same time, Ab treatment results in
increased tissue levels of C99 and nicastrin. Thus,
Ab appears to accelerate APP metabolism, increasing
the potential for amyloid peptide generation. The
resemblance of the Ab-induced changes to those
found after exposure to stress and GC supports the
view that stress and GC may contribute to the
etiopathogenesis of AD. Moreover, since Ab, stress
and GC all have negative effects on cognition,21,22,58it
would appear that Ab contributes to the molecular
signaling machinery that mediates the behavioral
actions of stress and GC. Lastly, it should be noted
that the Ab infusion protocol used here appropriately
models the preclinical stages of AD insofar that the
pathobiochemical changes observed here occurred in
a gradiential manner, being stronger in the hippo-
campus than in the PFC (cf. Braak and Braak47). While
our soluble Ab infusion paradigm did not result in
detectable amyloid depositions, it should be noted
that the role of Ab plaques in the pathogenesis of
AD is contentious.59,60Recent work indicates a strong
correlation between soluble Ab levels and the severity
of memory deficits in AD patients.61Importantly,
soluble Ab is known to trigger synapse loss,4an early
event in the pathogenesis of AD.
Interactions between endogenous and exogenous
factors (for example, age, mutations and Ab produc-
tion on the one hand, and stress on the other) are
important determinants of the onset and progress of
AD. The organism experiences intermittent stressors
throughout life, and its endogenous production of
Ab increases with age.62The results of recent studies
in transgenic animal models showed that chronic
stress potentiates Ab deposition and induces cogni-
tive deficits21and that amyloid production is exacer-
bated when GC levels are elevated.20In the present
work, we attempted to mimic typical lifetime events
by exposing rats to a chronic unpredictable stress
paradigm before subsequent treatment with GC
and/or central infusions of Ab. The results clearly
demonstrate that stress history exacerbates the APP
misprocessing effects of Ab and GC, and, strikingly,
also activated the amyloidogenic pathway in the PFC,
an area that had otherwise proven to be relatively
resilient to the individual stimuli. Importantly, the
combinatorial treatment resulted in a deterioration
of spatial reference memory that was greater than
that induced by the individual treatments. Thus,
stress history/previous exposure to high GC levels can
markedly worsen the effects of subsequent exposures
to stress/GC and Ab on AD-like biochemical and
behavioral pathology; accordingly, stress may be an
important precipitating and exacerbating factor in
AD. The present findings provide experimental
support for the implication of stress as a contributory
factor in early AD disease.63Based on this and
previous studies,64,65we suggest that stress (and GC)
make neurons more vulnerable to the pathological
actions of Ab, including Ab-induced stimulation of
To sum up, we have demonstrated that (1) stress/GC
can contribute to AD pathology by driving APP meta-
bolism toward the production of C99 and, eventually,
of Ab, and (2) that stress history ‘primes’ the brain to
generate greater amounts of amyloidogenic peptide(s)
upon subsequent exposures to GC and Ab. These
findings prompt the hypothesis that the effects of
stress/GC on neuronal atrophy may be mediated by
C99 and/or Ab, and that they may ultimately be
causally related to disruptions of behavior that are
typical of AD. The observation that Ab, like stress and
GC, can increase emotionality is interesting in view of
the increasing recognition of reciprocal regulatory
relationships between cognition and emotion,27–29
reports that many AD patients are hyperanxious,66
and the implication of stress and GC in disorders of
anxiety67and cognition.22,59Importantly, together
with previously established links between depression
and the increased risk for developing AD,31the
present findings suggest that Ab may be a common
denominator underlying these various stress-related
disorders, and indicate the potential importance
of including histories of stress and GC therapy in
anamnesic data. Lastly, our results reiterate the
need for judicial use of GC therapy in older subjects,
especially in light of the poor efficacy of GC to slow
the progression of AD.68,69
We thank Dieter Fischer and Rainer Stoffel for
excellent technical assistance, Carola Hetzel for
administrative help, and Drs Keiro Shirotani and
C Catania et al
Ayako Yamamoto for helpful advice. Isabel Matos and
Lucilia Pinto are thanked for help with histological
preparation and scoring, Dr Sam Gandy for providing
the 369 antibody and Dr Alexandre Patchev for help
with image preparation. CC and IS were supported by
stipends from the Max Planck Society and EU Marie
Curie Training Fellowships (at University College
London, UK). The collaboration between the German
and Portuguese laboratories was supported through
(DAAD/GRICES). This study was conducted within
the framework of the EU-supported integrated project
‘CRESCENDO’ (Contract FP6-018652).
1 Guillozet AL, Weintraub S, Mash DC, Mesulam MM. Neurofibril-
lary tangles, amyloid, and memory in aging and mild cognitive
impairment. Arch Neurol 2003; 60: 729–736.
2 Cleary JP, Walsh DM, Hofmeister JJ, Shankar GM, Kuskowski MA,
Selkoe DJ et al. Natural oligomers of the amyloid-beta protein
specifically disrupt cognitive function. Nat Neurosc 2005; 8:
3 Almeida CG, Tampellini D, Takahashi RH, Greengard P, Lin MT,
Snyder EM et al. Beta-amyloid accumulation in APP mutant
neurons reduces PSD-95 and GluR1 in synapses. Neurobiol Dis
2005; 20: 187–198.
4 Roselli F, Tirard M, Lu J, Hutzler P, Lamberti P, Livrea P et al.
Soluble beta-amyloid1-40 induces NMDA-dependent degradation
of postsynaptic density-95 at glutamatergic synapses. J Neurosci
2005; 25: 11061–11070.
5 Terry RD, Masliah E, Salmon DP, Butters N, DeTeresa R, Hill R
et al. Physical basis of cognitive alterations in Alzheimer’s disease:
synapse loss is the major correlate of cognitive impairment. Ann
Neurol 1991; 30: 572–580.
6 Olariu A, Tran MH, Yamada K, Mizuno M, Hefco V, Nabeshima T.
Memory deficits and increased emotionality induced by b-amyloid
(25–35) are correlated with the reduced acetylcholine release and
altered phorbol dibutyrate binding in the hippocampus. J Neural
Transm 2001; 108: 1065–1079.
7 De Strooper B. Nicastrin: gatekeeper of the gamma-secretase
complex. Cell 2005; 122: 318–320.
8 Bayer TA, Wirths O, Majtenyi K, Hartmann T, Multhaup G,
Beyreuther K et al. Key factors in Alzheimer’s disease: beta-
amyloid precursor protein processing, metabolism and intra-
neuronal transport. Brain Pathol 2001; 11: 1–11.
9 Lorenzo A, Yankner BA. Beta-amyloid neurotoxicity requires fibril
formation and is inhibited by congo red. Proc Natl Acad Sci USA
1994; 91: 12243–12247.
10 Weldon DT, Rogers SD, Ghilardi JR, Finke MP, Cleary JP, O’Hare E
et al. Fibrillar beta-amyloid induces microglial phagocytosis,
expression of inducible nitric oxide synthase, and loss of a select
population of neurons in the rat CNS in vivo. J Neurosci 1998; 18:
11 Mattson MP. Pathways towards and away from Alzheimer’s
disease. Nature 2004; 430: 631–639.
12 Yankner BA, Dawes LR, Fisher S, Villa-Komaroff L, Oster-Granite
ML, Neve RL. Neurotoxicity of a fragment of the amyloid precursor
associated with Alzheimer’s disease. Science 1989; 245: 417–420.
13 Oster-Granite ML, McPhie DL, Greenan J, Neve RL. Age-dependent
neuronal and synaptic degeneration in mice transgenic for the C
terminus of the amyloid precursor protein. J Neurosci 1996; 16:
14 Nalbantoglu J, Tirado-Santiago G, Lahsaini A, Poirier J, Goncalves
O, Verge G et al. Impaired learning and LTP in mice expressing the
carboxy terminus of the Alzheimer amyloid precursor protein.
Nature 1997; 387: 500–505.
15 Berger-Sweeney J, McPhie DL, Arters JA, Greenan J, Oster-Granite
ML, Neve RL. Impairments in learning and memory accompanied
by neurodegeneration in mice transgenic for the carboxyl-terminus
of the amyloid precursor protein. Mol Brain Res 1999; 66: 150–162.
16 Lorenzo A, Yuan M, Zhang Z, Paganetti PA, Sturchler-Pierrat C,
Staufenbiel M et al. Amyloid beta interacts with the amyloid
precursor protein: a potential toxic mechanism in Alzheimer’s
disease. Nat Neurosci 2000; 3: 460–464.
17 Heredia L, Lin R, Vigo FS, Kedikian G, Busciglio J, Lorenzo A.
Deposition of amyloid fibrils promotes cell-surface accumulation
of amyloid beta precursor protein. Neurobiol Dis 2004; 16:
18 Hartmann A, Veldhuis JD, Deuschle M, Standhardt H, Heuser I.
Twenty-four hour cortisol release profiles in patients with
Alzheimer’s and Parkinson’s disease compared to normal controls:
ultradian secretory pulsatility and diurnal variation. Neurobiol
Aging 1997; 18: 285–289.
19 Elgh E, Lindqvist Astot A, Fagerlund M, Eriksson S, Olsson T,
Nasman B. Cognitive dysfunction, hippocampal atrophy and
glucocorticoid feedback in Alzheimer’s disease. Biol Psychiatry
2006; 59: 155–161.
20 Green KN, Billings LM, Roozendaal B, McGaugh JL, LaFerla FM.
Glucocorticoids increase amyloid-beta and tau pathology in
a mouse model of Alzheimer’s disease. J Neurosci 2006; 26:
21 Jeong YH, Park CH, Yoo J, Shin KY, Ahn SM, Kim HS et al. Chronic
stress accelerates learning and memory impairments and increases
amyloid deposition in APPV717I-CT100 transgenic mice, an
Alzheimer’s disease model. FASEB J 2006; 20: 729–731.
22 Lupien SJ, Fiocco A, Wan N, Maheu F, Lord C, Schramek T et al.
Stress hormones and human memory function across the lifespan.
Psychoneuroendocrinology 2005; 30: 225–242.
23 Cerqueira JJ, Pego JM, Taipa R, Bessa JM, Almeida OF, Sousa N.
in prefrontal cortex-dependent behaviors. J Neurosci 2005; 25:
24 Sousa N, Almeida OFX. Corticosteroids: sculptors of the hippo-
campal formation. Rev Neurosci 2002; 13: 59–84.
25 Cerqueira JJ, Taipa R, Almeida OFX, Sousa N. Specific configura-
tion of dendritic degeneration in pyramidal neurons of the medial
prefrontal cortex induced by differing corticosteroid regimen.
Cereb Cortex 2006; 17: 1998–2006.
26 Selkoe DJ. Alzheimer’s disease: genotypes, phenotypes, and
treatments. Science 1997; 275: 630–631.
27 Dolan RJ. Emotion, cognition, and behavior. Science 2002; 298:
28 Ochsner KN, Gross JJ. The cognitive control of emotion. Trends
Cogn Sci 2005; 9: 242–249.
29 Phelps EA. Emotion and cognition: insights from studies of the
human amygdala. Annu Rev Psychol 2006; 57: 27–53.
30 Ownby RL, Harwood DG, Barker WW, Duara R. Predictors of
anxiety in patients with Alzheimer’s disease. Depress Anxiety
2000; 11: 38–42.
31 Ownby RL, Crocco E, Acevedo A, John V, Loewenstein D.
Depression and risk for Alzheimer disease: systematic review,
meta-analysis, and metaregression analysis. Arch Gen Psychiatry
2006; 63: 530–538.
32 Sousa N, Almeida OFX, Holsboer F, Paula-Barbosa MM, Madeira
MD. Maintenance of hippocampal cell numbers in young and aged
rats submitted to chronic unpredictable stress. Comparison with
the effects of corticosterone treatment. Stress 1998; 2: 237–249.
33 Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates,4th
edn. Academic Press: San Diego, 1998.
34 Sola ` C, Mengod G, Probst A, Palacios KM. Differential regional and
cellular distribution of a ˆ-amyloid precursor RNAs containing
and lacking the Kunitz protease inhibitor domain in the brain of
human, rat and mouse. Neuroscience 1993; 53: 267–295.
35 Panegyres PK. The effects of excitotoxicity on the expression of the
amyloid precursor protein gene in the brain and its modulation by
neuroprotective agents. J Neural Transm 1998; 105: 463–478.
36 Ni Y, Zhao X, Bao G, Zou L, Teng L, Wang Z et al. Activation of
beta(2)-adrenergic receptor stimulates gamma-secretase activity
and accelerates amyloid plaque formation. Nat Med 2007; 12:
37 Stephan A, Phillips AG. A case for a non-transgenic animal model
of Alzheimer’s disease. Genes Brain Behav 2005; 4: 157–172.
C Catania et al
38 Davis-Salinas J, Saporito-Irwin SM, Cotman CW, Van Nostrand Download full-text
WE. Amyloid beta-protein induces its own production in cultured
degenerating cerebrovascular smooth muscle cells. J Neurochem
1995; 65: 931–934.
39 Frautschy SA, Yang F, Calderon L, Cole GM. Rodent models of
Alzheimer’s disease: rat A beta infusion approaches to amyloid
deposits. Neurobiol Aging 1996; 17: 311–321.
40 Magarinos AM, Verdugo JM, McEwen BS. Chronic stress alters
synaptic terminal structure in hippocampus. Proc Natl Acad Sci
USA 1997; 94: 14002–14008.
41 Lupien SJ, de Leon M, de Santi S, Convit A, Tarshish C, Nair N et
al. Cortisol levels during human aging predict hippocampal
atrophy and memory deficits. Nat Neurosci 1998; 1: 69–73.
42 Sousa N, Paula-Barbosa MM, Almeida OFX. Ligand and subfield
specificity of corticoid-induced neuronal loss in the rat hippo-
campal formation. Neuroscience 1999; 89: 1079–1087.
43 Wellman CL. Dendritic reorganization in pyramidal neurons in
medial prefrontal cortex after chronic corticosterone administra-
tion. J Neurobiol 2001; 49: 245–253.
44 Cerqueira JJ, Catania C, Sotiropoulos I, Schubert M, Kalisch R,
Almeida OFX et al. Corticosteroid status influences the volume of
the rat cingulate cortex—a magnetic resonance imaging study.
J Psychiatr Res 2005; 39: 451–460.
45 Radley JJ, Morrison JH. Repeated stress and structural plasticity in
the brain. Ageing Res Rev 2005; 4: 271–287.
46 McEwen BS, De Kloet ER, Rostene W. Adrenal steroid receptors
and actions in the nervous system. Physiol Rev 1986; 66:
47 Braak H, Braak E. Neuropathological stageing of Alzheimer-related
changes. Acta Neuropathol (Berl) 1991; 82: 239–259.
48 Islam A, Kalaria RN, Winblad B, Adem A. Enhanced localization
of amyloid beta precursor protein in the rat hippocampus
following long-term adrenalectomy. Brain Res 1998; 806: 108–112.
49 Budas G, Coughlan CM, Seckl JR, Breen KC. The effect of
corticosteroids on amyloid beta precursor protein/amyloid pre-
cursor-like protein expression and processing in vivo. Neurosci
Lett 1999; 276: 61–64.
50 Sambamurti K, Kinsey R, Maloney B, Ge YW, Lahiri DK. Gene
structure and organization of the human beta-secretase (BACE)
promoter. FASEB J 2004; 18: 1034–1036.
51 Haass C. Take five—BACE and the gamma-secretase quartet
conduct Alzheimer’s amyloid beta-peptide generation. EMBO J
2004; 23: 483–488.
52 Johnston JA, Liu WW, Todd SA, Coulson DT, Murphy S, Irvine GB
et al. Expression and activity of beta-site amyloid precursor
protein cleaving enzyme in Alzheimer’s disease. Biochem Soc
Trans 2005; 33: 1096–1100.
53 Li Y, Zhou W, Tong Y, He G, Song W. Control of APP processing
and Abeta generation level by BACE1 enzymatic activity and
transcription. FASEB J 2006; 20: 285–292.
54 Yang DS, Tandon A, Chen F, Yu G, Yu H, Arawaka S et al. Mature
glycosylation and trafficking of nicastrin modulate its binding to
presenilins. J Biol Chem 2002; 277: 28135–28142.
55 Herreman A, Van Gassen G, Bentahir M, Nyabi O, Craessaerts K,
Mueller U et al. g-secretase activity requires the presenilin-
dependent trafficking of nicastrin through the Golgi apparatus
but not its complex glycosylation. J Cell Sci 2003; 116: 1127–1136.
56 Shah S, Lee SF, Tabuchi K, Hao YH, Yu C, LaPlant Q et al.
Nicastrin functions as a gamma-secretase-substrate receptor. Cell
2005; 122: 435–447.
57 Geula C, Wu CK, Saroff D, Lorenzo A, Yuan M, Yankner BA. Aging
renders the brain vulnerable to amyloid beta-protein neurotoxi-
city. Nat Med 1998; 4: 827–831.
58 Starkman MN, Giordani B, Berent S, Schork MA, Schteingart DE.
Elevated cortisol levels in Cushing’s disease are associated with
cognitive decrements. Psychosom Med 2001; 63: 985–993.
59 Lue LF, Kuo YM, Roher AE, Brachova L, Shen Y, Sue L et al.
Soluble amyloid beta peptide concentration as a predictor of
synaptic change in Alzheimer’s disease. Am J Pathol 1999; 155:
60 Edison P, Archer HA, Hinz R, Hammers A, Pavese N, Tai YF
et al. Amyloid, hypometabolism, and cognition in Alzheimer
disease: an [11C]PIB and [18F]FDG PETstudy. Neurology 2007; 68:
61 McLean CA, Cherny RA, Fraser FW, Fuller SJ, Smith MJ,
Beyreuther K et al. Soluble pool of Abeta amyloid as a determinant
of severity of neurodegeneration in Alzheimer’s disease. Ann
Neurol 1999; 46: 860–866.
62 Braak H, Braak E. Diagnostic criteria for neuropathologic assess-
ment of Alzheimer’s disease. Neurobiol Aging 1997; 18(Suppl 4):
63 Swanwick GR, Kirby M, Bruce I, Buggy F, Coen RF, Coakley D
Alzheimer’s disease: lack of association between longitudinal
64 Behl C, Lezoualch F, Trapp T, Widmann M, Skutella T, Holsboer F.
Glucocorticoids enhance oxidative stress-induced cell death in
hippocampal neurons in vitro. Endocrinology 1997; 138: 101–106.
65 Sapolsky RM. The possibility of neurotoxicity in the hippocampus
in major depression: a primer on neuron death. Biol Psychiatry
2000; 48: 755–765.
66 Grossberg GT. Diagnosis and treatment of Alzheimer’s disease.
J Clin Psychiatry 2003; 64(Suppl 9): S3–S6.
67 Tatsch MF, Bottino CM, Azevedo D, Hototian SR, Moscoso MA,
Folquitto JC et al. Neuropsychiatric symptoms in Alzheimer
disease and cognitively impaired, nondemented elderly from a
community-based sample in Brazil: prevalence and relation-
ship with dementia severity. Am J Geriatr Psychiatry 2006; 14:
68 in ‘t Veld BA, Ruitenberg A, Hofman A, Launer LJ, van Duijn CM,
Stijnen T et al. Nonsteroidal antiinflammatory drugs and the risk
of Alzheimer’s disease. N Engl J Med 2001; 345: 1515–1521.
69 Aisen PS. The potential of anti-inflammatory drugs for the
treatment of Alzheimer’s disease. Lancet Neurol 2000; 1: 279–284.
C Catania et al