Estrogen Therapy and brain muscarinic receptor density in
healthy females: A SPET study
Ray Norburya,b,⁎, Michael J. Travisb,e, Kjell Erlandssonc, Wendy Waddingtonc,d,
Peter J. Ellc,d, Declan G.M. Murphyb
aPsychopharmacology and Emotion Research Laboratory (PERL), University of Oxford, UK
bInstitute of Psychiatry, King’s College, London, UK
cInstitute of Nuclear Medicine, Middlesex Hospital, London, UK
dInstitute of Nuclear Medicine, University College Hospital, London, UK
eWestern Psychiatric Institute and Clinic, UPMC, Pittsburgh, USA
Received 29 August 2006; revised 23 October 2006; accepted 23 October 2006
Available online 14 December 2006
Estrogen Therapy (ET) may protect against age-related cognitive decline and neuropsychiatric disorders (e.g. Alzheimer's disease). The
biological basis for this putative neuroprotective effect is not fully understood, but may include modulation of cholinergic systems. Cholinergic
dysfunction has been implicated in age-related memory impairment and Alzheimer's disease. However, to date no one has investigated the effect
of long-term ET on brain cholinergic muscarinic receptor aging, and related this to cognitive function. We used Single Photon Emission
Tomography (SPET) and (R,R)[123I]-I-QNB, a novel ligand with high affinity for m1/m4muscarinic receptors, to examine the effect of long-term
ET and age on brain m1/m4receptors in healthy females. We included 10 younger premenopausal subjects and 22 postmenopausal women; 11
long-term ET users (all treated following surgical menopause) and 11 ET never-users (surgical menopause, n=2). Also, verbal memory and
executive function was assessed in all postmenopausal subjects. Compared to young women, postmenopausal women (ET users and never-users
combined) had significantly lower muscarinic receptor density in all brain regions examined. ET users also had higher muscarinic receptor density
than ET never-users in all the brain regions, and this reached statistical significance in left striatum and hippocampus, lateral frontal cortex and
thalamus. Moreover, in ET users, (R,R)[123I]-I-QNB binding in left hippocampus and temporal cortex was significantly positively correlated with
plasma estradiol levels. We also found evidence for improved executive function in ET users as compared to ET never-users. However, there was
no significant relationship between receptor binding and cognitive function within any of the groups. In healthy postmenopausal women use of
long-term ET is associated with reduced age-related differences in muscarinic receptor binding, and this may be related to serum estradiol levels.
© 2006 Elsevier Inc. All rights reserved.
Keywords: (R,R)[123I]-I-QNB; SPET; Muscarinic; Estrogen; Aging
Increasing age is associated with a greater risk of cognitive
impairment and neuropsychiatric disorders such as Alzheimer's
disease and Parkinson's disease [AD and PD]. Hence, with the
growing size of the elderly population the prevalence of age-
related cognitive decline, and AD and PD, are likely to increase.
However, the few treatments available are of limited efficacy.
Some, but not all, studies suggest that in healthy post-
menopausal women the use of Estrogen Therapy (ET) may
protect against age-related differences in verbal memory and
executive function; and also decrease risk for AD (for a review,
Norbury et al., 2003). The variability in these findings may arise
from the heterogeneity of the populations studied. For example,
a number of earlier studies have included women taking both
unopposed estrogen and combined preparations—but animal
studies suggest that progesterones may counteract the effects of
Hormones and Behavior 51 (2007) 249–257
⁎Corresponding author. University of Oxford, Centre for Clinical Magnetic
Resonance Research,Level 0, MRS Unit, John Radcliffe Hospital, Oxford, OX3
9DU, UK. Fax: +44 1865 221111.
E-mail address: email@example.com (R. Norbury).
0018-506X/$ - see front matter © 2006 Elsevier Inc. All rights reserved.
estrogens (Maki and Hogervorst, 2003). Further, a number of
earlier studies differed in time of initiation of ET. Notably,
Verghese et al. (2000) reported improved verbal memory in
women who started hormone therapy (HT) at or near the time of
menopause. It has also been demonstrated that HT following
surgical menopause and “add-back estrogen” following treat-
(GnRH) agonists is associated with improved verbal memory
(Sherwin, 1988; Sherwin and Tulandi, 1996). By contrast,
suggest that HT initiated at age 65 or above does not protect
against cognitive decline or AD (Rapp et al., 2003; Schumaker
et al., 2004; Schumaker et al., 2003) and increases the risk for
all-cause dementia. Therefore, it is possible that the potential
beneficial effects of estrogen are realized only if treatment is
initiated at or near the menopause. Beyond this theoretical
‘window of opportunity’ neurons and/or chemical systems may
remain refractory to the putative positive effects of estrogen
that use of estrogen is associated with improved fine motor
control (Sherwin, 2002) and may be beneficial in controlling the
motor symptoms of PD (Craig et al., 2004).
There is mounting evidence to suggest that estrogen has posi-
human basal forebrain, the major cholinergic innervation to the
cerebral cortex, hippocampus and hypothalamus (Toran-Allerand
et al., 1992). Estrogen has been shown to provide trophic support
factors (Toran-Allerand et al., 1992).
In animals, estrogen modulates a number of cholinergic
parameters. For example, estrogen replacement has been shown
to: 1) enhance choline acetyltransferase activity and high affinity
choline uptake (Gibbs, 2000a); 2) modulate acetylcholine release
(Gibbs et al., 1997); and 3) ameliorate a scopolamine-induced
memory deficit (Gibbs et al., 1998). Moreover, a number of
studies have demonstrated that the effects of estradiol on
cholinergic and cognitive function diminish with age and time
following loss of ovarian function (Gibbs, 2000b; Gibbs and
these findings suggest that estrogen treatment in animals has
within a set period of time (referred to above as the ‘window of
opportunity’). Estrogen may also modulate cholinergic muscari-
nic receptors. In animals, muscarinic receptor density varies as a
function of estrous cycle, being highest during proestrous
(characterized by high estrogen levels) and lowest during
diestrous [low estrogen] (van Huizen et al., 1994). In addition,
ovariectomy up-regulates, and estrogen normalizes, rat hippo-
campal muscarinic m4receptors (El-Bakri et al., 2002).
In humans, data on estrogen and central cholinergic function
is limited. We recently reported that women taking long-term
ET, and who initiated ET at or near the menopause, had signi-
ficantly higher cholinergic responsivity (as measured by growth
hormone [GH] response to oral pyridostigmine) than women
who were ET naïve. Moreover, within long-term ET users there
was a significant positive correlation between GH response and
duration of estrogen exposure (van Amelsvoort et al., 2003).
Also, a recent SPET study (Smith et al., 2001) reported an
increased index of cortical cholinergic nerve terminal concen-
trations with increasing years of ET use in healthy postmeno-
pausal women, although no overall difference was observed
between users and non-users of ET. These studies (Smith et al.,
2001; van Amelsvoort et al., 2003) were important first steps.
However, the neuroendocrine challenge provides only an
indirect measure of central neurotransmitter responsivity and,
to our knowledge, no one has yet investigated the effect of
estrogen on cholinergic aging or muscarinic receptors. We
therefore used the novel m1/m4muscarinic receptor ligand (R,R)
[123I]-I-QNB and SPET toinvestigate the effectoflong-term ET
on brain m1 and m4 muscarinic receptor aging in healthy
females. Based on the extant literature we wished to test the
following specific predictions:
1. In healthy postmenopausal women use of long-term ET is
associated with reduced age-related differences in brain m1
and m4muscarinic receptor binding; and
2. ET use in healthy postmenopausal women is associated with
improved verbal memory and executive function.
Methods and materials
College London. Permission to administer radiotracer was obtained from the
United Kingdom Administration of Radioactive Substances Advisory Commit-
tee (ARSAC). Written informed consent was obtained from all subjects after the
experimental procedures had been fully explained. All subjects were in good
physical health and recruited from the same geographical area by local
advertisement and personal contact. No subject was taking any medication
or physical illness affecting brain function; 2) use of medications acting on the
central nervous system; 3) alcohol or drug dependence or abuse; 4) clinically
abnormal Magnetic Resonance scan [specifically, subjects were excluded if they
had evidence of intracerebral organic pathology, and all scans were defined by a
clinical radiologist (blind to group) as being within normal clinical limits. In
addition we excluded subjects with clinically identifiable cortical atrophy,
infarcts of any size, ventriculomegaly, or extensive white matter abnormalities
minimal white matter abnormalities (grade 1 or less as defined by Kozachuk
et al., 1990)]; 5) current use of any form of hormonal contraception; and 6)
current pregnancy. The study population consisted of three groups of women
(see Table 1): 10 healthy, premenopausal subjects (mean age 29, S.D. 4), 11
healthy surgically postmenopausal long-term ET users on standard doses of
estrogen-only preparations [mean age 65 (S.D. 8), age at menopause 47 (S.D. 8),
mean duration of ET 18 years (S.D. 9), 7 using slow-release estradiol implants, 2
on continuous oral conjugated equine estrogens, and 2 using transdermal
patches] and 11 healthy, postmenopausal ET never-users [mean age 65 (S.D. 6)
age at menopause 52 (S.D. 4)]. Menopause was defined as either: 1) absence of
menstrual periods for 12 months as reported by the subject; or 2) hysterectomy
with bilateral salpingo-oophorectomy. To ensure their status as ET users and
never-users, estradiol levels were measured using standard clinical assays in
the Department of Clinical Biochemistry, King's College Hospital, London.
The biochemist was blinded to subject status. The sensitivity of this assay is
10 pmol L−1and cross reactivity is <1%. In addition, standard methods were
used to assess apolipoprotein E (ApoE) polymorphisms in postmenopausal
subjects to establish if potential between-group differences were related to
ApoE allele frequencies. Premenopausal subjects had regular cycles, were not
250 R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257
on hormonal contraceptives and were scanned mid-cycle, when estradiol levels
are expected to be high and progesterone levels low.
Each postmenopausal subject received Magnetic Resonance Imaging (MRI)
scans to exclude clinical abnormalities and to ensure that differences in grey
matter volume did not account for any observed between-group differences in
muscarinic receptor density. All subjects underwent routine blood tests
(including full blood count, thyroid, liver and renal function, and serum cortisol
levels) to exclude any covert medical illness which may affect brain function.
Cognitive testing included the Wechsler Abbreviated Scale of Intelligence
[WASI] (Wechsler, 1997), after screening with the Beck Depression Inventory
[BDI] (Beck, 1978) and Annett Hand Preference Questionnaire [AHQ] (Annett,
1970). In addition, verbal memory was assessed in all postmenopausal subject
using the California Verbal Learning Test [CVLT] (Delis et al., 1987). The
California Verbal Learning Test was selected based on findings from previous
studiesthathavereportedaneffect ofestrogenon performanceof thistask (Maki
et al., 2001).
Radiosynthesis of (R,R)[123I]-I-QNB (IQNB) was performed by West of
Scotland Radionuclide Dispensary as previously described (Lee et al., 1996;
Norbury et al., 2004) All subjects were provided with a standard regime of
180 mg potassium iodate per day, beginning 2 days prior to scanning and
continuing for a total of 5 days to minimize radioactive iodine uptake by the
SPET data acquisition
Prior to SPET measurement a venous cannular was inserted into an ante-
cubital vein for radiotracer administration. Subjects were positioned supine on
the scanner bed and allowed to relax for 20 min. (R,R)[123I]-I-IQNB was admi-
nistered intravenously as a slow bolus over ∼20 s (Mean activity=177 MBq,
S.D. 8.45). The cannular was flushed with 20 mL normal saline immediately
after injection of radiotracer.
Dynamic SPET data (see also ROI analysis) was acquired using a Prism
3000XP triple-headed scanner equipped with a
(Philips Medical Systems, Cleveland, OH, USA). The three detectors were fitted
153Gd transmission source
Emission, transmission and scatter data were acquired as described in (Norbury
et al., 2004).
All image reconstruction, processing and analysis was performed using in-
house software implemented in IDL 5.2 (Interactive Data Language, Research
Systems Inc., Boulder, CO, USA). Transmission images were reconstructed
using a ordered subsets implementation of the convex ML algorithm (Lange and
with ramp-filter, after scatter correction with the triple energy window method
(Ogawa et al., 1991). A 3D Butterworth filter was applied to both the trans-
mission and emission data. Non-uniform attenuation correction, based on the
transmission images, was performed with two iterations of the method described
by Chang (Chang, 1978).
IQNB template and regions of interest
A study-specific IQNB template was generated from 10 young and 10 older
subjects (10 older subjects were included to ensure that the template was not
biased with respect to age). Transmission data for these subjects was scaled to
remove soft tissue and co-registered using a nine-parameter (three translation,
three rotation and three scaling) minimization of the sum of least-squares
difference in voxel intensity to create a mean transmission template. These
registration parameters were then applied to the corresponding emission data
from each subject and the realigned images averaged to produce a study specific
(R,R)[123I]-I-QNB template. Regions of interest (ROIs) demarcating hippo-
campus, thalamus, and medial frontal, lateral frontal, parietal and temporal
cortex taken from a standardized MRI template (Takeuchi et al., 2002) were then
co-registered to the IQNB template. ROIs were then co-registered to each
subject's emission data using an inverse of the registration parameters derived
from the realignment to the IQNB template. In addition ROIs corresponding to
striatum, anterior and posterior cingulate and cerebellum were generated
manually on each individual's data. These ROIs were created by an experienced
operatorbased onthe 50%isocontour line andwiththe aidof an anatomicalatlas
(Talairach and Tournoux, 1998).
Data acquired ∼405–450 min p.i. was averaged and analyzed using the ratio
method (ROI/REF)-1, where ROI is the region of interest and REF the reference
region (cerebellum). This approach has previouslybeen shownto provide robust
estimates of binding potential (BP) for this radiotracer (Norbury et al., 2004).
Statistical analyses were implemented using SPSS v.11.0 for windows
(SPSS Inc, Chicago, Illinois). ROI data (BP) were first assessed for normality of
distribution using the Kolmogorov–Smirnov statistic. These data violated the
assumption of normality (Kolmogorov–Smirnov statistic p<0.05) and were
transformed using a square root function to conform to a normal distribution.
Subsequent analysis was performed on the transformed data only. Between-
group differences in regional BP were analyzed using one-way analysis of
variance (ANOVA). Where the overall effect of group was significant, planned
comparisons were performed to examine the effect of age (young vs. post-
menopausal groups combined) and ET (ETusers vs. ET never-users). California
Verbal Learning Test scores were analyzed using Student's independent samples
t-tests. Correlations between BP and demographic and neuropsychological
variables were assessed using Pearson's coefficient. ApoE allele frequencies
were assessed using two-tailed Fisher's exact tests. No attempt was made to
correct for multiple comparisons.
MRI acquisition and analysis
Magnetic resonance imaging (MRI) scans were acquired (postmenopausal
subjects only) at the Maudsley Hospital using a GE Signal operating at 1.5
Tesla (Milwaukee, WI) to exclude any clinical abnormalities. High resolution
Data expressed as mean (S.D.)‡p<0.05, young vs. old (ERT users and ERT
never-users),#p<0.001, ERT never-users vs. young and ERT users
Age at menopause
Plasma oestradiol (pmol L−1)
Progesterone (nmol L−1)
Cortisol (nmol L−1)
ApoE allele frequency
N/A=not applicable; WASI=Wechsler Abbreviated Scale of Intelligence,
BDI=Beck Depression Inventory, AHQ=Annett Handedness Questionnaire.
ApoE allele frequency: values show % of subjects within each group expressing
either heterozygous or homozygous combinations.
251R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257
structural Magnetic Resonance Imaging (MRI) scans of the whole head were
also acquired to ensure that differences in grey matter volume did not account
for any observed between-group differences in muscarinic receptor density.
The MRI sequence employed was a 3D-Spoiled Grass (SPGR) volume scan.
There were 124 contiguous slices (256×192), a 1.5 mm slice thickness, field of
view of 220 mm, TR=13.8 ms, TE=2.8 ms.
MRI pre-processing and analysis
Structural MRI data were analyzed on a Sun Ultra 60 workstation (Sun
Microsystems, Mountain View, CA) using SPM2 (Wellcome Department of
Cognitive Neurology, London) implemented in MATLAB 6 (Mathworks,
Natick, MA). Image pre-processing was performed using the fully automated
optimized Voxel Based Morphometry Protocol (Good et al., 2001). Images were
subsequently smoothed with a Gaussian kernel (12 mm FWHM). The optimally
pre-processed images were then analyzed using SPM2. Regionally specific
differences in total grey matter volume were assessed using one-tailed t-tests
(namelytesting for increased or decreasedgrey matter in ETusers and ET never-
users) with α=0.05 (corrected) considered significant.
(WASI), mood (BDI), number of years spent in education, years
postmenopause or handedness (Table 1). Young subjects, how-
ever, had spent significantly longer in education than both
groups of postmenopausal subjects [χ2
larger proportion of subjects in the ET users group had received
hysterectomies with bilateral salpingo-oophorectomies (n=11
slightly lower (albeit non-significant [p=0.07]) age at meno-
pause in this group.
Mean plasma estradiol levels (Table 1) in premenopausal
women (414 pmol L−1, S.D. 87) and ET users (498 pmol L−1,
S.D. 196) were similar (t19=−0.89, p=0.384), indicating that
estradiol levels in the women on long-term ET can reach
physiological levels. In contrast, mean estradiol levels in
postmenopausal ET never-users confirmed postmenopausal
status (79 pmol L−1, S.D. 48) and were significantly lower
than those of young premenopausal women and ET users (both
ps<0.001). Progesterone levels were below the detectable thre-
shold of 5 nmol L−1in all subjects. No significant between-
group differences were observed in cortisol levels (F2,28=1.16,
p=0.328). Also, there was no significant between-group
difference in ApoE allele frequency (ET users and ET never-
users) (Table 1).
2=10.08, p=0.006). A
Effect of age ET on regional (R,R)[123I]-I-QNB binding
ANOVA revealed a significant main effect of group for BP in
all brain regions (F2,29>5.76, all ps<0.008). The planned
contrasts revealed significant age-related differences (young vs.
both postmenopausal groups combined) in all brain regions
examined (Table 2). Furthermore, ET users had numerically
higher muscarinic receptor density than never-users in all the
brain regions examined, and this reached statistical significance
and right lateral frontal cortex) and thalamus (mean and left
thalamus) (Table 3).
Relationship between BP and demographic variables in
Within ET users, there was no significant relationship be-
tween duration of ET and BP. There were, however, significant
positive correlations between plasma estradiol levels and BP in
hippocampus [mean and left hippocampus,] (r(11)=0.637,
p=0.035, and r(11)=0.627, p=0.039, respectively [nota bene;
this relationship did not remain significant if the data point
y=1.95, x=782 was removed) and temporal cortex [mean and
left temporal cortex] (r(11)=0.607, p=0.048, and r(11)=0.612,
p=0.045, respectively) (see Fig. 1). For ET never-users no
significant correlations were obtained between age at meno-
pause or years postmenopause and BP (all ps>0.05).
Estradiol and cognitive function
There were no significant between group differences (ET
users vs. ET never-users) in correct responses to measures of
learning or memory as assessed by the CVLT. However, ET
users made significantly fewer perseverative errors when
compared with ET never-users. Correlational analyses were
performed to investigate the relationship between BP and
California Verbal Learning Test scores. No significant correla-
tions were observed between BP in any brain region and correct
responses to measures of learning or memory scores in the
postmenopausal groups combined, or in a separate analysis of
ET users and ET never-users.
There were no significant between-group differences (ET
users vs. ET never users) in grey matter volume at p=0.05
corrected, or at the more relaxed threshold of p=0.001,
uncorrected. This finding suggests that the observed differences
due to differences in underlying grey matter volume (Table 4).
To our knowledge this is the first study to investigate the
effect of age and long-term ET on brain muscarinic receptor
aging. We have previously shown that long-term ET in healthy
postmenopausal women modulates central cholinergic respon-
study extend this work and provide preliminary evidence to
suggest that long-term ET initiated at or near the menopause
modulates brain muscarinic receptor aging. Consistent with a
number of previous findings (Zubieta et al., 2001), we found a
significant age-related difference (young vs. postmenopausal
women) in BP in all the brain regions we examined [the effectof
age on muscarinic receptor density in healthy females has been
detailed elsewhere (Norbury et al., 2005)]. Moreover, consistent
252 R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257
with our original hypothesis, long-term ET users had signifi-
cantly higher muscarinic receptor density in left striatum and
hippocampus, lateral frontal cortex (right and mean) and
thalamus (mean and left) as compared to ET never-users.
These brain regions are implicated in motor control, higher
cognitive function, Alzheimer's disease and Parkinson's
disease. It is possible therefore that modulation of brain m1/m4
muscarinic receptor aging may underlie, in part, the putative
Alzheimer's and Parkinson's disease. Contrary to our original
hypothesis, we found no significant effect of ET on verbal
memory. Consistent with earlier studies (Keenan et al., 2001),
however, we did find preliminary evidence to suggest that
estrogen modulates executive function.
Our study has a number of limitations and these should be
taken into consideration when interpreting the results. For
example, we used a cross-sectional design, and so we can only
examine age-related differences as opposed to aging per se.
Also, confounding variables may have contributed to our
findings. The most notable of these is the “healthy-user bias”,
i.e. the tendency for women who elect to take ET to be generally
healthier and better educated than non-recipients (Matthews
et al., 1996). Thisissuecanonlyproperlybeaddressedbylarge
randomized controlled trials. Also, the postmenopausal women
who took part in our study were in good health, recruited from
the same geographical area, similar in terms of years in educa-
that healthy-user bias can explain all our findings. Nevertheless,
there was a significant difference in the type of menopause
between the ET users and ET never-users. Specifically, more of
the women in the ET group had undergone surgical menopause
(n=11) as compared to the ET never-users group (n=2). During
natural menopause there is a gradual decline in estradiol and
androgen levels until circulating levels are maintained solely by
peripheral conversion of adrenal steroids. In contrast, surgical
menopause causes a complete cessation of both estradiol and
Effect of age on regional muscarinic receptor density (young vs. postmenopausal groups combined)
Brain regionYoung (n=10)Postmenopausal (n=22)t(29)p
Anterior cingulate1.8850.182 1.5480.232 4.04 <0.001
Posterior cingulate2.0060.1941.728 0.2293.39 0.001
Lateral frontal cortex
Medial frontal cortex
Values show t statistic and corresponding significance.
253R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257
testosterone levels although supplementation is usually imple-
mented almost immediately post-surgery. In addition, exogen-
ous estrogen supplementation in postmenopausal womencauses
an increase in sex hormone binding globulin, and a concomitant
reduction in the bioavailability of the remaining androgens. It is
unclear what effect this potential bias between the two groups
may have had on the current results as free testosterone was not
measured. Future studies are required to assess the gener-
lizeability of our findings to other groups of postmenopausal
women. In addition, although we found no significant diffe-
rences in total gray matter volume between ET users and ET
never users, we cannot exclude the possibility that some of the
observed differences in m1/m4muscarinic receptor density may
be due to regional differences in grey matter volume. However,
we observed highly significant between group differences in
m1/m4muscarinic receptor density in thalamus, a brain region
grey matter volume (Good et al., 2001). In addition, for volume
loss to account completely for the receptor loss reported here it
would have to be relatively specific to cholinergic neurons.
Thus, partial volume effects are unlikely to wholly explain
differences in m1/m4muscarinic receptor density reported here.
We also found significantly greater m1/m4binding in left hippo-
campus of ET users as compared to ET never-users. As the
hippocampus isa relativelysmall structure,itispossible that our
hippocampal measuremay include signalfrom adjacentregions.
However, a region anatomically corresponding to the hippo-
campus was clearly delineated on our IQNB template due to
the high uptake of the tracer in the study population as a
whole. In addition, the high uptake of IQNB in hippocampus is
consistent with the known distribution of m1/m4muscarinic
receptors in human brain. We also carried out multiple statis-
tical tests (and so may be open to Type 1 error). Nevertheless,
of the statistical comparisons between young and postmeno-
pausal women all were significant, and between ET users and
never-users most strongly trended towards significance, and 7
California Verbal Learning Test scores
Learning Test score
ERT users (n=11) ERT never-users
Mean S.D. Mean S.D.
Immediate free recall54.45
Short delay free recall 11.36 3.07
Short delay cued recall
Long delay free recall
Long delay cued recall13.18 2.27
Long delay recognition 15.180.75
Values show mean and S.D.
aERT never-users greater than ERT users (t(20)−4.033, p=0.001).
Effect of ERT on regional muscarinic receptor density (ERT users vs. ERT
Brain regionERT users
Anterior cingulate1.590 0.298 1.5070.141 0.90.19
Posterior cingulate1.7950.279 1.661 0.149 1.460.08
Lateral frontal cortex
Medial frontal cortex
Values show t statistic and corresponding significance (*p<0.05).
Fig. 1. Relationship between BP in hippocampus (averaged across hemispheres)
and plasma estradiol concentration (ERT users only). Pearson's r=0.637,
254R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257
reached significance. This is unlikely to be fully explained by
We observed significantly greater muscarinic m1/m4mus-
carinic receptor density in left hippocampus of ET users as
compared to ET never-users, in the absence of between group
differences in grey matter volume. Both m1and m4receptors
1997; Marino et al., 1998; Russo et al., 1993) and a loss of m1
people with AD (Mulugeta et al., 2003; Shiozaki et al., 2001).
There is also accumulating evidence to suggest that estrogen has
beneficial effects on hippocampal cholinergic function. In
ovariectomised rats, estrogen facilitates cholinergic neurotrans-
et al., 1997). In addition, estrogen has been shown to ameliorate
the cognitive impairment produced by intrahippocampal injec-
tions of the muscarinic antagonist scopolamine (Gibbs, 1999)
and modulate m4receptor density in rat hippocampus (El-Bakri
et al., 2002). Evidence from earlier neuroimaging studies also
point to the importance of the hippocampus in estrogen modula-
tion of cognitive function. Resnick et al. (1998) reported diffe-
rential activation of right hippocampal gyrus in HRT users
compared to non-users. In addition, our group has shown
hippocampal spectra of ET never-users compared to long-term
ET users and young premenopausal women (Robertson et al.,
2001). We observed significantly greater muscarinic m1/m4
receptor density in left hippocampus in ETusers as compared to
never-users. It would follow therefore, that modulation of
hippocampal m1and m4muscarinic receptor aging, may be one
mechanism whereby estrogen can protect against age-related
cognitive decline and perhaps AD. Unfortunately, we did not
observe any significant between group differences in verbal
memory nor any relationship between cognitive function and
muscarinic receptor density. Nevertheless, it remains possible
that estrogen-related preservation of hippocampal m1and m4
muscarinic receptors may protect against later cognitive decline
and risk for AD.
We also found significantly greater muscarinic m1/m4recep-
tor density in ET users as compared to ET never-users in
muscarinic antagonists impair (Diaz del Guante et al., 1991) and
cholinomimetics improve (Lazaris et al., 2003) performance on
memory tasks. In addition, age-related reductions in striatal
muscarinic receptors may contribute to motor deficits that
accompany old age (Strong et al., 1980). In humans, estrogen
use in healthy postmenopausal women is associated with
improved fine motor skills and memory (Sherwin, 2002). It is
possible therefore that by modulation of striatal muscarinic
receptors estrogen may have beneficial effects on both memory
and motor skills. Data from animal studies also suggest that
estrogen may play a role in modulating thalamic function. For
example, ovariectomy reduces acetylcholinesterase activity in
hypothalamus, thalamus and medulla—a deficit that can be
reversed by estrogen treatment (Das et al., 2002). Notably,
cholinergic basal forebrain neurons can regulate activity in
neocortex either through direct cortical projections or via their
projections to the thalamus (Cornwall and Phillipson, 1988) and
may play a role in gating the flow of information through the
thalamus (Guillery et al., 1998). Higher m1/m4 muscarinic
receptor density in thalamus may therefore positively affect
tasks that require selective attention—such as the perseverative
errors component of the CVLT. Our finding of significantly
greater m1/m4receptors in thalamus and fewer perseverative
errors in the ET users as compared to never-users may be
consistent with this hypothesis. We also observed significantly
greater muscarinic m1/m4receptor density in mean and right
lateral frontal cortex, brain regions strongly implicated in tasks
suggested that estrogen may play a role in tasks mediated by
frontal regions. For example, consistent with our findings
Keenan et al. (2001) reported improved executive function (as
measured by perseverative errors) in HT treated women as
compared women who were HT naïve. Neuroimaging studies
also point to the frontal cortex as a site of estrogen action.
Eberling et al. (2004) reported reduced glucose metabolism in
the frontal cortex of ET non-users as compared to ET users and
Kugaya etal. (2003) reportedincreased5-HT2Areceptor density
in right, medial and inferior frontal gyrus in healthy post-
menopausal women following treatment with estrogen. The
finding of increased m1/m4receptor density in mean and right
lateral frontal cortex reported here suggest that estrogen may
modulate frontal function through actions on the muscarinic
cholinergic system. However,we found no relationship between
cognitive function and muscarinic m1/m4receptor density in
this population of healthy women. Well designed longitudinal
studies, that include a wider range of memory and executive
tasks, are required to examine if the positive effects of estrogen
on muscarinic receptors reported here predict future cognitive
function and risk for AD and PD.
Significant relationships were observed between muscarinic
receptor density in hippocampus and temporal cortex and
circulating estradiol levels in the ETusers group. This finding is
consistent with an earlier report that muscarinic receptor density
in young rats was positively correlated with estrogen levels
across the estrus cycle (van Huizen et al., 1994). Thus, while
there is tentative evidence to suggest that, in this relatively small
group of ET-treated women, higher plasma estradiol levels may
be associated with increased muscarinic receptor density, future
studies that directly investigate the effects of differing doses and
regimes of ET are required.
In summary, we found a significant age-related difference
(young vs. postmenopausal women) in BP in all the brain
regions we examined. Moreover, consistent with our original
hypothesis, long-term ET users had significantly higher musca-
rinic receptor density in left striatum and hippocampus, lateral
frontal cortex (mean and right) and thalamus (mean and left
thalamus) as compared to ET never-users. These brain regions
are implicated in motor control, higher cognitive function,
Alzheimer's disease and Parkinson's disease. We also found
limited evidence to suggest that estrogen modulates executive
255 R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257
function. Thus, the findings from our study may explain in part,
the putative neuroprotective effects of estrogen on brain ageing
and risk for age-related cognitive decline and neuropsychiatric
The authors thank the staff at the Institute of Nuclear
Medicine for their technical assistance and all the individuals
who took part in this study.
This study was funded by grants from the Medical Research
holds a Junior Research Fellowship from Wolfson College,
University of Oxford. The findings of this paper were presented
in part at the 59th annual meeting of the Society of Biological
Psychiatry, New York, April 29–May 1, 2004 and at
Neuroreceptor Mapping, Vancouver, July 15th–18th, 2004.
Annett, M., 1970. A classification of hand preference by association analysis.
Br. J. Psychol. 61 (3), 303–321.
Beck, A.T., 1978. Beck Depression Inventory. Harcourt Brace and Company,
Chang, L.T., 1978. A method for attenuation correction in radionuclide
computed tomography. IEEE Trans. Nucl. Sci. 25, 638–643.
Cornwall, J., Phillipson, O.T., 1988. Afferent projections to the dorsal thalamus
of the rat as shown by retrograde lectic transport. 2. The midline nuclei.
Brain Res. Bull. 21, 147–161.
Craig, M.C., Cutter, W.j., Wickham, H., van Amelsvoort, T., Rymer, J.,
Whitehead, M., Murphy, D.G.M., 2004. Effect of long-term oestrogen
therapy on dopaminergic responsivity in post-menopausal women—A
preliminary study. Psychoneuroendocrinology 29, 1309–1316.
Das, A., Dikshit, M., Srivastava, S.R., Srivastava, U.K., Nath, C., 2002. Effect
of ovariectomy and oestrogen supplementation on brain acetylcholinesterase
and passive avoidance learning in rats. Can. J. Pharm. 80 (9), 907–914.
Delis, D.C., Kramer, J.H., Kaplan, E., Ober, B.A., 1987. California Verbal
Learning Test, Research ed. Psychological Corporation, New York.
Diaz del Guante, M.A., Cruz-Morales, S.E., Prado-Alcala, R.A., 1991. Time-
dependent effects of cholinergic blockade of the striatum on memory.
Neurosci. Lett. 122, 79–82.
Eberling, J.L., Wu, C., Tong-Turnbeaugh, R., Jagurst, J., 2004. Estrogen-and-
tamoxifen-associated effects on brain structure and function. NeuroImage
El-Bakri, N.K., Adem, A., Suliman, I.A., Mulugeta, E., Karlsson, E., Lindgren,
J.U., Winblad, B., Islam, A., 2002. Estrogen and progesterone treatment:
effects on muscarinic M4 receptor subtype in rat brain. Brain Res. Bull. 948,
Gibbs, R.B., 1999. Oestrogen replacement enhances acquisition of a spatial
memory task and reduces deficits associated with hippocampal muscarinic
receptor inhibition. Horm. Behav. 36, 222–233.
Gibbs, R.B., 2000a. Effects of oestrogen on basal forebrain cholinergic
neurones and cognition: implications for brain ageing and dementia in
women. In: Morrision, M. (Ed.), Hormones, Aging and Mental Disorders.
Cambridge Univ. Press, Cambridge.
Gibbs, R.B., 2000b. Long-term treatment with oestrogen and progesterone
enhances acquisition of a spatial memory task by ovariectomised aged rats.
Neurobiol. Aging 21, 107–116.
Gibbs, R.B., Gabor, R., 2003. Estrogen and cognition: applying preclinical
findings to clinical perspectives. J. Neurosci. Res. 74, 637–643.
Gibbs, R.B., Hashash, A., Johnstone, D.A., 1997. Effects of oestrogen on
potassium-stimulated acetylcholine release in the hippocampus and over-
lying cortex of adult rats. Brain Res. 749, 143–146.
Gibbs, R.B., Burke, A.M., Johnson, D.A., 1998. Estrogen attenuates effects of
scopolamine and lorazepam on memory acquisition and retention. Horm.
Behav. 34, 112–125.
Good, C., Johnsrude, I.S., Ashburner, J., Henosn, R.N.A., Friston, K.J.,
Frackowiak, S.J., 2001. Avoxel-based morphometric study of ageing in 465
normal adult human brains. NeuroImage 14, 21–36.
Guillery, R.W., Feig, S.L., Lozsadi, D.A., 1998. Paying attention to the thalamic
reticular nucleus. Trends Neurosci. 37, 31–44.
Jerulanski, D., Kornisiuk, E., Izquierdo, I., 1997. Cholinergic neurotransmission
and synaptic plasticity concerningmemory processing. Neurochem. Res. 22,
Keenan,P.A., Ezzat, W.H., Ginsberg, K., Moore, G.J., 2001. Prefrontal cortex as
the site of oestrogen's effect on cognition. Psychoneuroendocrinology 26,
Kozachuk, W.E., DeCarli, C., Schapiro, M.B., Wagner, E.E., Rapoport, S.I.,
Horwitz, B., 1990. White matter hyperintensities in dementia of Alzheimer's
typeand inhealthysubjects withoutcerebrovascular risk factors.Amagnetic
resonance imaging study. Arch. Neurol. 47 (12), 1306–1310.
Kugaya, A., Epperson, C.N., Zoghbi, S., van Dyck, C., Hou, Y., Fujita, M.,
Staley, J.K., Garg, P.K., Seibyl, J.P., Innis, R.B., 2003. Increase in prefrontal
cortex 5-HT2A receptors following estrogen treatment in postmenopausal
women. Am. J. Psych. 160, 1522–1524.
Lange, K., Fessler, J.A., 1995. Globally convergent algorithms for maximum
a posteri transmission tomography. IEEE Trans. Image Process. 4,
Lazaris, A., Cassel, S., Stemmlin, J., Cassel, C., Kelche, E., 2003. Intrastriatal
infusions of methoctramine improve memory in cognitively impaired aged
rats. Neurobiol. Aging 24, 379–383.
Lee, K.S., He, X.S., Jones, D.W., Coppola, R., Gorey, J.G., Knable, M.B.,
rapid and efficient radioiodination of iodine-123-IQNB. J. Nucl. Med. 37
Maki, P.M., Hogervorst, E., 2003. HRT and cognitive decline. Best Pract. Res.,
Clin. Enodocrinol. Metab. 17 (1), 105–122.
Maki, P.M., Zonderman, A.B., Resnick, S.M., 2001. Enhanced verbal memory
in nondemented elderly women receiving hormone-replacement therapy.
Am. J. Psych. 158, 227–233.
Marino, M.J., Rouse, S.T., Levey, A.I., Potter, L.T., Conn, P.J., 1998. Activation
of the genetically defined m1 muscarinic receptor potentiates N-methyl-
Natl. Acad. Sci. U. S. A. 95, 11465–11470.
Markowska, A.L., Savonenko, A.V., 2002. Effectiveness of estrogen replace-
mentin restorationof cognitivefunctionafter long-termestrogenwithdrawal
in aging rats. J. Neurosci. 22, 10985–10995.
Matthews, K.A., Kuller, L.H., Wing, R.R., Meilahn, E.N., Plantinga, P., 1996.
Prior to use of estrogen replacement therapy, are users healthier than
nonusers? Am. J. Epidemiol. 143 (10), 971–978.
Mulugeta, E., Karlsson, E., Islam, A., Kalaria, R., Mangat, H., Winbald, B.,
Abdu, A., 2003. Loss of muscarinic M4 receptors in hippocampus of
Alzheimer patients. Brain Res. 960, 259–262.
Norbury, R., Cutter, W.J., Compton, J., Robertson, M., Whitehead, M., Murphy,
D.G.M., 2003. The neuroprotective effects of oestrogen on the ageing brain.
Exp. Gerontol. 38, 109–117.
Norbury, R., Travis, M.J., Erlandsson, K., Waddington, W., Owens, J., Ell, P.J.,
Murphy, D., 2004. SPETimaging of central muscarinic receptors with (R,R)
[123I]-I-QNB: Methodological considerations. Nucl. Med. Biol. 31 (5),
Norbury, R., Travis, M.J., Erlandsson, K., Waddington, W., Owens, J., Pimlott,
S., Ell, P.J., Murphy, D.G., 2005. In vivo imaging of muscarinic receptors in
the aging female brain with (R,R)[123I]-I-QNB and single photon emission
tomography. Exp. Gerontol. 40 (3), 137–145.
Ogawa, K., Harat, Y., Ichihara, T., Kubo, A., Hashimoto, S., 1991. A practical
method for position-dependent Compton-scatter correction in single
emission CT. IEEE Trans. Med. Imag. 10, 408–412.
Rapp, S.R., Espeland, M.A., Shumaker, S.A., Henderson, V.W., Brunner, R.L.,
Manson, J.E., Gass, M.L., Stefanick, M.L., Lane, D.S., Hays, J., Johnson,
K.C., Coker, L.H., Dailey, M., Bowen, D., 2003. Effect of estrogen plus
progestin on global cognitive function in postmenopausal women: the
256 R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257
Women's Health Initiative Memory Study: a randomized controlled trial.
JAMA 289 (20), 2663–2672.
Resnick, S.M., Maki, P.M., Golski, S., Kraut, M.A., Zonderman, A.B., 1998.
Effects of oestrogen replacement therapy on PET cerebral blood flow and
neuropscychological performance. Horm. Behav. 34, 171–182.
Robertson, D.M.W.,vanAmelsvoort,T., Daly, E.,Simmons,A.,Whitehead,M.,
Morris, R.G., Murphy, K.C., Murphy, D.G.M., 2001. Effects of estrogen
replacement therapy on human brain aging: an in vivo 1H-MRS study.
Neurology 57, 2114–2117.
Russo, C., Marchi, M., Andrioli, G.C., Cavazzani, P., Raiteri, M., 1993.
Enhancement of glycine release form human brain synaptosomes by
acetylcholine action at m4 muscarinic receptors. J. Exp. Ther. 266, 142–146.
Schumaker, S.A., Legault, C., Rapp, S.R., Thal, L., Wallace, R.B., Ockene,J.K.,
Hendrix, S.L., Jones III, B.N., Assaf, A.R., Jackson, P.D., Kotchen, J.M.,
Wasswertheil-Smoller, S., Wactowski-Wende, J., 2003. Oestrogen and
progesterone and the incidence of dementia and mild cognitive impairment
in postmenopausal women the Womens Health Initiative Memory Study:a
randomised controlled trial. JAMA 289, 2651–2662.
Schumaker, S.A., Legault, C., Kuller, L., Rapp, S.R., Thal, L., Lane, D.S., Fillit,
H., Stefanick, M.L., Hendrix, S.L., Lewis, C.E., Masaki, K., Coker, L.H.,
2004. Conjugated equine oestrogens and incidence of probable dementia
and mild cognitive impairment in postmenopausal women. JAMA 291 (4),
Sherwin, B.B., 1988. Oestrogen and or androgen replacement therapy an
cognitive function in surgically menopausal women. Psychoneuroendocri-
Sherwin, B.B., 2002. Estrogen and cognitive aging in women. Trends Pharm.
Sci. 23 (11), 527–534.
Sherwin, B.B, Tulandi, T., 1996. “Add-back” estrogen reverses cognitive
deficits induced by a gonadotropin-releasing hormone agonist in women
with leiomyomata uteri. J. Clin. Endocrinol. Metab. 81, 2545–2549.
Shiozaki, K., Iseki, E., Hino, H., Kosaka, K., 2001. Distribution of m1
muscarinic acetylcholine receptors in the hippocampus of patients with
Alzheimer's disease and dementia with Lewy bodies—An immunohisto-
chemical study. J. Neurol. Sci. 193, 23–28.
Smith, Y.R., Satoshi, M., Kuhl, D.E., Zubieta, J., 2001. Effects of long-term
hormone replacement therapy on cholinergic synaptic concentrations in
healthy postmenopausal women. J. Clin. Endocrinol. Metab. 86, 679–684.
Strong, R., Hicks, P., Hsu, L., Bartus, R.T., Enna, S.J., 1980. Age-related
changes in the rodent brain cholinergic system and behavior. Neurobiol.
Aging 1, 59–63.
Takeuchi, R., Yonekura, Y., Matusda, H., Konishi, J., 2002. Usefulness of a
three-dimensional stereotaxic ROI template on anatomically standardised
99 mTc-ECD SPET. Eur. J. Nucl. Med. Mol. Imaging 29 (3), 331–341.
Talairach, J., Tournoux, P., 1998. Co-Planar Sterotaxic Atlas of the Human
Brain. Thieme Medical Publishers, New York.
Toran-Allerand, D., Miranda, R.C., Bentam, W.D.L., Sohrabji, F., Brown, T.J.,
Hochberg, R.B., McLusky, N.J., 1992. Oestrogen receptors co-localise with
low-affinity nerve growth factor receptors in cholinergic neurones of the
basal forebrain. Proc. Natl. Acad. Sci. U. S. A. 89, 4668–4672.
van Amelsvoort, T., Murphy, D.G.M., Robertson, D., Daly, E., Whitehead, M.,
Abel, K., 2003. Effects of long-term oestrogen replacement therapy on
growth hormone response to pyridostigmine challenge in healthy post-
menopausal women. Psychoneuroendocrinology 28, 101–112.
van Huizen, F., March, D., Cynader, M.S., Shaw, C., 1994. Muscarinic receptor
characteristics and regulation in rat cerebral cortex: changes during
development, aging and the oestrous cycle. Eur. J. Neurosci. 6 (2), 237–243.
Verghese, J., Kuslansky, G., Katz, M.J., Sliwinski, M., Crystal, H.A., Busche,
H., Lipton, R.B., 2000. Cognitive performance in surgically postmenopausal
women on estrogen. Neurology 55, 872–874.
Wechsler, D., 1997. Wechsler Abbreviated Scale of Intelligence (WASI).
Harcourt Brace and Company, San Antonio.
Zubieta, J.K., Koeppe, R.A., Frey, K.A., Kilbourn, M.R., Mangner, T.J., Foster,
N.L.,Kuhl, D.E.,2001.Assessment of muscarinicreceptorconcentrations in
aging and Alzheimer disease with [11C]NMPB and PET. Synapse 39 (4),
R. Norbury et al. / Hormones and Behavior 51 (2007) 249–257