Cerebral Cortex April 2011;21:877--883
Advance Access publication August 20, 2010
Effects of Risk Genes on BOLD Activation in Presymptomatic Carriers of Familial
Alzheimer’s Disease Mutations during a Novelty Encoding Task
John M. Ringman1, Luis D. Medina1, Meredith Braskie1,4, Yaneth Rodriguez-Agudelo2, Daniel H. Geschwind1, Miguel A. Macias-Islas3,
Jeffrey L. Cummings1and Susan Bookheimer4
1Mary S. Easton Center for Alzheimer’s Disease Research, Department of Neurology, David Geffen School of Medicine at University
of California at Los Angeles, Los Angeles, CA 90095, USA,2Laboratory of Experimental Psychology, National Institute of Neurology
and Neurosurgery, Mexico City 14269, Mexico,3Department of Neuroscience, University of Guadalajara, Guadalajara 44130, Mexico
and4Semel Institute for Psychiatry and Human Behavior, University of California at Los Angeles, Los Angeles, CA 90095, USA
Address correspondence to John M. Ringman, Mary S. Easton Center for Alzheimer’s Disease Research, Department of Neurology, David Geffen
School of Medicine at UCLA, 10911 Weyburn Avenue, #200, Los Angeles, CA 90095-7226, USA. Email: email@example.com.
Prior functional magnetic resonance imaging (fMRI) studies have
found increased activity-related blood oxygen level--dependent
(BOLD) signal in cognitively normal persons at genetic risk for
Alzheimer’s disease (AD). This has been interpreted as a compen-
satory response to incipient AD pathology. We studied the effects
of fully penetrant familial Alzheimer’s disease (FAD) mutations and
apolipoprotein E (APOE) genotype on BOLD fMRI during a novelty
encoding task in presymptomatic subjects. Twenty-three Mexican
or Mexican-American persons at-risk for inheriting FAD mutations
performed a block design novelty encoding task, and activation
exhibited by FAD mutation carriers (MCs) was contrasted with that
of noncarriers (NCs) and among APOE genotype groups. FAD MCs
(n 5 14) showed decreased BOLD activation in the anterior
cingulate gyrus relative to 9 NCs. No increased activation was seen
in MCs relative to NCs. Four APOE e3/4 carriers demonstrated
increased BOLD signal compared with 14 e3/3 carriers in the
occipital and perisylvian cortices bilaterally. There were no areas
where e3/3 carriers activated more than e3/4 carriers. Our findings
of increased fMRI activation associated with APOE genotype but
not with FAD mutations suggest that APOE exerts an effect on the
BOLD signal that is not readily explained as a compensatory
Keywords: Alzheimer’s disease, apolipoprotein E, familial, functional
magnetic resonance imaging, presymptomatic
As Alzheimer’s disease (AD) pathology begins decades before
the manifestation of symptoms (Troncoso et al. 1998), it should
be possible to diagnose the illness during its presymptomatic
stage. Cerebrospinal fluid indices (Mattsson et al. 2009) and
radioactively labeled ligands used in positron emission tomog-
raphy (PET) that bind to amyloid pathology (Okello et al. 2009)
show particular promise as preclinical markers of AD. The
relationship between such measurements and a given person’s
cognitive status is indirect as there is variability in individuals’
intellectual capacity and ability to adapt to brain damage. It has
been proposed that alternative neural networks are recruited
to maintain normal cognitive function in the face of neuro-
One correlate of cognitive function is focal change in
cerebral blood flow measured as the blood oxygenation level--
dependent (BOLD) signal by functional magnetic resonance
imaging (fMRI). Numerous fMRI studies have compared the
BOLD signal in persons at different levels of risk of having
presymptomatic AD neuropathology. A commonly studied
indicator for AD risk is apolipoprotein E (APOE) genotype,
with the e4 allele conferring a higher risk (Corder et al. 1993).
An early study, using fMRI during encoding and retrieval of
a series of unrelated word pairs, found increased magnitude
and extent of BOLD signal in the left hippocampus and parietal
and prefrontal cortex in APOE e4 carriers (Bookheimer et al.
2000). Similar results were obtained in subsequent studies
during encoding of novel versus familiar words (Fleisher et al.
2005) and novel versus familiar visual scenes (Bondi et al.
2005). Dickerson et al. (2004) found that APOE e4 carrier status
was associated with increased activation in entorhinal cortex
across controls, subjects with mild cognitive impairment and
AD. This effect has not been observed in all studies (Lind et al.
2006). At least one study suggested that a family history of AD
was associated with ‘‘increased’’ BOLD signal (Bassett et al.
2006) and another with ‘‘decreased’’ BOLD signal indepen-
dently of APOE genotype (Johnson et al. 2006). The underlying
basis for increased BOLD signal associated with the APOE e4
genotype is uncertain though a compensatory response
(Bookheimer et al. 2000; Bondi et al. 2005).
Unlike APOE genotype, which merely confers an increased
risk for AD, in persons who have inherited autosomal dominant
familial Alzheimer’s disease (FAD) due to mutations in the APP,
PSEN1, or PSEN2 genes, the future development of AD can be
predicted with essentially 100% certainty. In addition, the age
of onset may be predicted with some accuracy (Murrell et al.
2006). This population therefore provides the opportunity to
study task-related changes in fMRI response sensitively in
presymptomatic disease. A prior study found that a 20-year-old
presymptomatic PSEN1 mutation carrier (MC) had increased
BOLD signal during associative learning (left hemisphere) and
retrieval (bilaterally) in many cerebral areas (Mondadori et al.
2006). This was interpreted as representing a compensatory
effect. The goal of the current study is to look at the effects of
FAD mutation status and APOE genotype on fMRI activation
during a novelty encoding task in a larger number of
presymptomatic subjects at-risk for FAD mutations to differ-
entiate the effects of these genes.
Materials and Methods
Forty-three persons at-risk for inheriting pathogenic PSEN1 or APP
mutations underwent clinical, cognitive, and imaging evaluations.
? The Author 2010. Published by Oxford University Press. All rights reserved.
For permissions, please e-mail: firstname.lastname@example.org
Twenty-three asymptomatic subjects were included in the current
study. The remainder were excluded for the following reasons: no fMRI
data obtained (n = 5), presence of dementia (n = 5), presence of mild
cognitive impairment or Clinical Dementia Rating Scale (Morris 1997)
score > 0 (n = 8), genetic data unavailable (n = 1), and excessive
movement artifact (n = 1). All included subjects were of Mexican origin
and all spoke Spanish adequately to perform Spanish language neuro-
psychological testing. All study procedures were approved by the UCLA
Institutional Review Board, and all subjects provided written informed
Though the age of onset of dementia varies somewhat between
families with FAD, there is some consistency within families (Murrell
et al. 2006). It is therefore possible to estimate the length of time from
symptom onset and dementia diagnosis a given FAD MC is. In obtaining
the ages of symptom onset and dementia diagnosis within a family, we
have found substantial variability between family members’ reports of
age of symptom onset, even when describing the same patient.
Therefore, in order to compare subjects with regard to the interval
over which they would be expected to develop dementia, subjects’
ages relative to the median age of dementia diagnosis in their families
(adjusted age) was calculated and used as a covariate in fMRI analyses.
We also calculated and report subjects’ ages relative to the mean age of
reported symptom onset in their family (relative age).
Subjects underwent neuropsychological assessment by a fluently
bilingual psychometrician (LDM). Testing included measures that were
either nonlinguistic or available in Spanish. Composite z scores for
cognitive domains were calculated as previously reported (Ringman
et al. 2009). The domains consisted of Language (Category Fluency for
animals, Object Naming from the Spanish English Neuropsychological
Assessment Scale [Mungas et al. 2004]), Visuospatial (Rey--Osterrieth
Figure Copy [Wechsler 1987], Block Design from the Wechsler Memory
Scale-Revised [Loewenstein et al. 1995]), Verbal Memory (Word-
List Learning Delayed Recall, Memory Verbal Prose Delayed Recall
[Artiola-i-Fortuny et al. 1999]), and Frontal/Executive Function (Stroop
Interference Score [Stroop 1935], Color Trails Interference Score).
Mutation carriers who scored 1.5 standard deviations below non-
carriers (NCs) on these composite scores were defined as being
impaired in that domain and therefore having mild cognitive
impairment. Such subjects were excluded. As a global measure of
cognitive function, the Cognitive Abilities Screening Instrument,
a measure available in both English and Spanish on which scores range
from 0 (worst) to 100 (best) (Teng et al. 1994), was administered by
a clinician (J.M.R.). In all but one subject who had undergone clinical
presymptomatic genetic testing, ratings were performed blind to
subjects’ genetic status.
Blood samples were coded according to a unique identifier and
forwarded to the laboratory of D.H.G. DNA was extracted and APOE
genotyping performed using standard techniques.
Presenilin-1: The presence of A431E and L235V substitutions in PSEN1
were assessed using restriction fragment length polymorphism
Amyloid precursor protein: The presence of the V717I substitution in
APP was assessed with direct sequencing.
The incidental novelty encoding fMRI activation paradigm was the first
of 3 tasks performed in a single session. Subjects viewed 10 blocks of 16
scenes alternating between 5 blocks of repeated scenes and 5 blocks of
novel scenes. Each scene was presented for 2.55 s with each block
lasting 40.8 s. Subjects pressed 1 of 2 buttons depending on whether
the depicted scene was indoors or outdoors. Rest blocks lasting 30.6 s
were presented at the beginning and end of the session. Subjects
were instructed on how to perform the task the day before scanning.
While in the scanner, subjects were again given verbal and written
instructions on how to perform the task.
One-hundred thirty-eight, 30 slice whole-brain images were obtained
using echo planar imaging (EPI) on a 3 T Siemens Allegra scanner.
Thirty axial slices were acquired in the plane of the AC--PC line (time
repetition [TR] = 3400 ms, time echo [TE] = 35 ms, and flip angle of 90
degrees). Resolution was 3.1 3 3.1 3 3.0 mm. Subjects also underwent
T2-weighted imaging (TR = 5000 ms, TE = 30 ms) with bandwidth
matched to that of the EPI sequences to assist registration of functional
images to a standard template (see below).
Subjects also underwent structural MRI on a 1.5 T Siemens Sonata
scanner. Whole-brain T1-weighted images were obtained in the sagittal
plane using a magnetization prepared rapid gradient echo sequence
(TR = 1900 ms, TE = 4.38 ms, TI = 1100 ms, flip angle 15 degrees). Voxel
size was 1 3 1 3 1 mm3. Brain volumes were extracted from the
cranium and extracranial tissues using FMRIB’s Brain Extraction Tool
(Smith 2002). The percent of intracranial volume occupied by brain for
each subject was calculated using FMRIB’s SIENAX (Structural Image
Evaluation, using Normalization, of Atrophy) (Smith et al. 2002).
Demographic and Cognitive Variables
Mean age, adjusted age, relative age, and cognitive test scores were
compared between FAD MCs and NCs using 2-tailed t-tests. Gender,
proportion of persons of different APOE genotypes (e3/2, 3/3, or 3/4),
and the mutated gene for which they were at-risk of inheriting (PSEN1
vs. APP) were compared between FAD MCs and NCs using chi-square
tests and, where appropriate, Fisher’s exact test. Mean age, adjusted
age, and relative age were compared among APOE genotype groups
using analyses of variance. Gender, proportion of persons who were FAD
MCs or NCs, and the mutated gene they were at-risk of inheriting were
compared among APOE genotype groups using chi-square tests and,
where appropriate, Fisher’s exact test. A 95% confidence interval was
used for determining significance. These statistical analyses were
performed using Statistical Package for the Social Sciences version 11.0.2.
Areas of differential BOLD signal between novel and repeated blocks for
individual subjects were compared by comparing repeated blocks to
novel blocks using FEAT (FMRI Expert Analysis Tool) version 5.4, part of
FMRIB’s Software Library (http://www.fmrib.ox.ac.uk/fsl). Motion cor-
rection was performed using MCFLIRT (Smith 2002); non-brain structures
were removed using Brain Extraction Tool (Smith et al. 2002); spatial
smoothing was performed using a Gaussian kernel of full-width at half-
maximum 5 mm; mean-based intensity normalization of all volumes was
performed using the same factor; and high-pass temporal filtering
was applied (Gaussian-weighted least squares function straight-line fitting,
with sigma = 30.0 s). Time-series statistical analysis was carried out using
FILM with local autocorrelation correction (Woolrich et al. 2001).
Z (Gaussianized T/F) statistic images were thresholded using clusters
determined by Z >2.3 and a (corrected) cluster significance threshold of
P = 0.05 (Evans et al. 1992). Functional images were registered to
matched bandwidth images and then to the Montreal Neurological
Institute standard template using FLIRT (Jenkinson and Smith 2001).
Group comparisons were carried out with the output files of
individuals’ results using FLAME (FMRIB’s Local Analysis of Mixed
Effects) (Woolrich et al. 2004). Activation patterns for FAD MCs were
compared with those of NCs covarying for APOE genotype, adjusted
age, and the mutated gene they were at-risk for inheriting. The latter
variable also controlled, in part, for family of membership. Comparisons
were also made among APOE genotype groups covarying for adjusted
age, mutated gene they were at-risk for inheriting, and whether or not
they were an FAD MC. Z (Gaussianized T/F) statistic images were
thresholded using clusters determined by Z > 2.3 and a (corrected)
cluster significance threshold of P = 0.05 (Worsley et al. 1992). Percent
of intracranial volume occupied by brain calculated from 1.5 T
structural images as above were compared between FAD MCs and
NCs using a 2-tailed t-test.
In order to quantify signal changes in areas showing differential
responses between genetic groups, regions of interest masks were
FMRI in FAD
Ringman et al.
made from the activation maps from group comparisons. From the
APOE analyses, a mask derived from the common area differentially
activated between e4 and e3 carriers and between e3 carriers and e2
carriers was employed. Percent signal change between conditions in
these areas was then calculated for each subject.
Subjects came from 8 families: 2 families had the V717I
in PSEN1, and 5 had the A431E substitution in PSEN1,
representing a founder effect (Murrell et al. 2006; Yescas et al.
2006). Mean age of symptom onset among families ranged from
36 to 49 years.
Of the 23 subjects, 14 were MCs and 9 were NCs. There were
no differences in age adjusted for family-specific age of dementia
diagnosis, age relative to family-specific mean age of symptom
onset, gender, mutated gene they were at-risk for inheriting,
APOE genotype distribution, or mean percentage of intracranial
volume occupied by brain between FAD MCs and NCs (Table 1).
Mean absolute age of MCs was lower than that of NCs (29.9 vs.
37.3, P = 0.042). There were no differences in cognitive test
scores other than MCs having lower scores on the Memory
Verbal Prose Delayed Recall test (14.4 vs. 16.3, P = 0.05).
Of the 23 subjects, 5 had the e2/3, 14 had the e3/3, and 4 had
the e3/4 APOE genotype. There were no differences between
these groups in age adjusted for family-specific age of dementia
diagnosis, age relative to family-specific mean age of symptom
onset, gender, mutated gene they were at-risk for inheriting, or
whether or not they were FAD MCs (Table 2); 4 persons with
the e3/2 (80%), 8 with the e3/3 (57%), and 2 with the e3/4
(50%) APOE genotypes were FAD MCs.
During blocks of novel stimuli relative to blocks of repeated
stimuli, substantial increased BOLD signal was typically noted
in the inferior and medial temporal lobe, including the
hippocampus, and extending posteriorly into the inferior
occipital lobes and into primary visual cortex and widespread
areas of visual association cortex in the occipital lobes (Fig. 1).
Relative to repeated blocks, clusters of voxels in the anterior
cingulate gyrus bilaterally and in the left frontopolar region
were more activated in NCs than carriers of FAD mutations
during the novelty encoding trials (Fig. 2). The Z statistic had
a maximal value of 3.61 at Talairach coordinates –8, 34, 8,
corresponding to the anterior cingulate gyrus rostral to the
genu of the corpus callosum (Brodmann’s area 24) on the left
side. There were no areas where FAD MCs had greater
activation than NCs during the novelty blocks than the
Controlling for FAD MC status, carriers of the APOE e3/4
genotype had greater BOLD signal increase during novelty
blocks than carriers of the APOE e3/3 genotype in widespread
areas including primary visual cortex and other areas of
occipital lobe and perisylvian cortex bilaterally (Fig. 3). The Z
Demographic and neuropsychological variables in 14 FAD MCs and 9 nonmutation carrying family members
FAD mutation carriers (n 5 14) Noncarriers (n 5 9)P
Mean age in years (range)
Mean age in years, adjusted for median age of dementia diagnosis in the
Mean age in years, adjusted for mean age of symptom onset in the family
Gender, no. of female (%)
No. of subjects at-risk for PSEN1 mutations (vs. APP, %)
No. of APOE e3/2 (%)
No. of APOE e3/3 (%)
No. of APOE e3/4 (%)
Mean CASI score (SD)
Mean % ICV occupied by brain (SD)
Mean Category Fluency score (SD)
Mean Object Naming score (SD)
Mean Rey--O Copy score (SD)
Mean Block Design score (SD)
Mean Word List Learning Delayed Recall score (SD)
Mean Verbal Prose Delayed Recall (SD)
Mean Stroop Interference Score (SD)
Mean Color Trails Interference Score (SD)
29.9 (23, 43)
?15.6 (?31, ?5)
37.3 (19, 55)
?8.2 (?35, þ18)
?13.3 (?25, ?4)
?5.7 (?29, 19) 0.08*
Note: CASI, Cognitive Abilities Screening Instrument; ICV, intracranial volume; SD, standard deviation.
*P # 0.05.
Demographic data according to APOE gene status
APOE e3/2 (n 5 5)APOE e3/3 (n 5 14) APOE e3/4 (n 5 4)P
Mean age in years (range)
Mean age in years, adjusted for median age of dementia diagnosis in the
Mean age in years, adjusted for mean age of symptom onset in the family
Gender, no. of female (%)
No. of subjects at-risk for PSEN1 mutations (vs. APP, %)
No. of FAD MCs (%)
33.8 (26, 43)
?8.8 (?19, ?2)
30.4 (19, 46)
?16.0 (?35, ?2)
40.0 (29, 55)
?6.3 (?18, 18)
?7.0 (?16, 3)
?13.8 (?29, ?1)
?2.3 (?12, 19)0.09
Cerebral Cortex April 2011, V 21 N 4 879
statistic had a maximal value of 6.04 at Talairach coordinates
6, –86, 0, corresponding to Brodmann’s area 17 on the right.
APOE e3/3 carriers also activated more than e3/2 carriers in
primary visual cortex (Fig. 4, maximal Z statistic of 3.8 at
Talairach coordinates 8, –94, 8). Box plots of percent signal
change in area 17 on the right in APOE genotype groups are
shown in Figure 5. There were no areas where e3/2 carriers
activated more than e3/3 carriers, nor where e3/3 carriers
activated more than e3/4 carriers during the novelty blocks
relative to the repeated blocks.
We found decreased BOLD signal in the anterior cingulate
gyrus bilaterally and the left frontal pole in presymptomatic
FAD MCs compared with matched NC family members during
blocks of novel stimuli. No areas were more activated in FAD
MCs compared with NCs. We were therefore unable to
demonstrate increased brain activity in presymptomatic
persons destined to develop AD due to FAD mutations.
Interestingly, we found areas of increased BOLD signal in
carriers of the e3/4 APOE genotype compared with carriers of
the e3/3 genotype and in carriers of the e3/3 compared with
the e2/3 genotype. Our findings do not support the
hypothesis that increased activation seen in persons at
genetic risk for AD represents a compensatory phenomenon
but rather suggest that APOE exerts an effect on activation-
related BOLD signal that is at least partly independent of AD
risk per se.
Figure 1. Representative horizontal, right parasagittal, and coronal views of pattern of increased BOLD signal during blocks of novel stimuli relative to repeated stimuli averaged
across all 23 subjects. Widespread bilateral areas of inferior temporal and occipital lobes, lateral occipital lobes, and the hippocampi are activated during the novelty encoding
blocks. Areas shown are those with Z-scores greater of 4 or greater.
Figure 2. Areas where BOLD signal was greater during the novelty encoding task in 9 nonmutation carrying FAD family members than in their 14 presymptomatic mutation
carrying kin. No areas were more active in FAD MCs than their nonmutation carrying family members during this task.
Figure 3. Areas where BOLD signal was greater during the novelty encoding task in 4 persons with e3/4 alleles of the APOE gene than in 14 persons with carrying e3/3 alleles.
There were no areas where e3/3 carriers had greater BOLD signal increase than e3/4 carriers during this task.
FMRI in FAD
Ringman et al.
The novel versus repeated stimuli task used in this study is
a well-studied paradigm that induces increased cerebral blood
flow to the posterior hippocampus, parahippocampal gyrus,
and the fusiform and lingual gyri (Worsley et al. 1992;
Bondi et al. 2005). It is thought that the activation in the
hippocampus is related to the novelty of the stimuli, whereas
the activation seen in the fusiform and lingual gyri are related
to recognition of complex visual stimuli (Stern et al. 1996). We
found grossly similar anatomical patterns of activation across
subjects regardless of FAD mutation status and APOE genotype,
suggesting that there were no fundamental differences in the
way stimuli are processed in the 2 populations.
The decreased task-related fMRI signal seen in presymptom-
atic FAD MCs is consistent with what is observed in persons
with established AD (Dickerson et al. 2004). Using the novelty
encoding task in our population, the decreased BOLD signal
was seen in the anterior cingulate gyrus, whereas prior fMRI
studies have found decreased fMRI signal in the medial
temporal lobe (Dickerson et al. 2004) and frontal lobe (Li
et al. 2009), depending on what activation tasks are employed.
The anterior cingulate plays a role in many cognitive tasks and
is part of the salience network (Seeley et al. 2007) that might
be expected to be selectively activated during the processing
of novel stimuli. Decreases in baseline glucose metabolism and
cerebral blood flow occur in the anterior cingulate in
established AD and decreased resting cerebral blood flow to
the anterior cingulate has been described in presymptomatic
PSEN1 MCs (Johnson et al. 2001). The relatively increased
signal seen in NCs during blocks of novel stimuli in the anterior
cingulate was more related to decreased activation during
novel blocks in MCs than to decreased deactivation during
repeated trials (data not shown). As there were minimal true
rest periods during this task, it is not possible to address the
integrity of the intrinsic connectivity networks in this study.
A PET study using the amyloid-binding ligand Pittsburgh
Compound B (PIB) in persons with established variant FAD due
to a deletion in exon 9 of PSEN1 that features spastic
paraparesis found increased signal in the striatum and anterior
cingulate (Koivunen et al. 2008). Some persons carrying the
A431E substitution in PSEN1 also develop spastic paraparesis
(Murrell et al. 2006), and carriers of this mutation made up 43%
of our study population. A PET investigation of presymptomatic
PSEN1 MCs using PIB demonstrated high signal in the striatum
early in the presymptomatic period, presumably reflecting
amyloid pathology there (Klunk et al. 2007). In this study, some
asymptomatic subjects showing this pattern were more than 10
years younger than the age at which they would be expected to
begin to show cognitive decline, an age comparable to that of
the subjects in our study. We did not observe any differences in
BOLD signal in the striatum in FAD MCs though this area did
not tend to be activated during the task we employed. The
nucleus accumbens of the striatum has indirect connections
with the anterior cingulate via frontal-subcortical circuits
(Cummings 1995). Therefore, possible explanations for our
finding of decreased anterior cingulate activation include
involvement of the cingulate gyrus with amyloid or diaschisis
secondary to striatal pathology. Obtaining both fMRI and
amyloid imaging in the same subjects would be required to
directly address this question.
Multiple prior fMRI studies have found increased focal fMRI
response during various activation tasks in carriers of the APOE
e4 allele (Bookheimer et al. 2000; Dickerson et al. 2004; Bondi
et al. 2005; Fleisher et al. 2005; Johnson et al. 2006; Yassa et al.
2008). As this allele is the principal genetic susceptibility locus
for late-onset AD, such findings have been interpreted as
representing an increase in focal brain activity to compensate
for or differentiation of the BOLD signal due to subclinical AD
pathology. Familial AD, in which alterations in PSEN1 and APP
(and PSEN2), are fully penetrant and cause the onset of AD at
a young and somewhat predictable age (Fox et al. 1997; Murrell
et al. 2006) provide an alternative genetic model in which to
test this hypothesis. A prior fMRI study of 2 persons inheriting
the C410Y PSEN1 mutation found increased activation in
multiple brain areas during learning and retrieval of faces in the
young (20 years of age) but not the older (45 years of age)
preclinical carrier (Mondadori et al. 2006). As the age of onset
of symptoms in this family tended to be around age 48, the
authors interpreted this as possibly representing increased
focal cerebral blood flow 30 years prior to the development of
Figure 5. Box plots of percent BOLD signal change in right Brodmann’s area 17
during blocks of novel stimuli relative to repeated stimuli in various APOE genotypes.
Figure 4. Areas where BOLD signal was greater during the novelty encoding task in 14 persons with e3/3 alleles of the APOE gene than in 5 persons with carrying e2/3 alleles.
There were no areas where e2/3 carriers had greater BOLD signal increase than e3/3 carriers during this task.
Cerebral Cortex April 2011, V 21 N 4 881
symptoms. In our larger study, we were unable to demonstrate
a similar effect in similarly -aged presymptomatic FAD MCs.
Though at least one fMRI study suggested that having
a family history of AD was associated with increased fMRI
activity independently of APOE genotype (Bassett et al. 2006),
most fMRI findings in persons susceptible to AD due to APOE
genotype are equally compatible with an effect of APOE on
activation-related blood flow that is at least somewhat in-
dependent of AD pathology. Johnson et al. (2006), using an
activation task similar to ours, found increased activation in the
medial temporal lobes in response to novel items in APOE e4
carriers but decreased activation in nondemented persons with
a family history who did not carry the e4 genotype. This study
and ours are consistent with APOE genotype exerting a direct
effect on activation-related BOLD signal. This latter interpre-
tation is further supported by a recent study, also employing
a novelty encoding paradigm, that found increased activation in
the hippocampi and cerebellum in 18 APOE e4 carriers relative
to 18 NCs who were between 20 and 35 (mean 28.5) years of
age (Filippini et al. 2009). As this age is before the age at which
AD pathology would be expected to appear, even in APOE e4
carriers, it is consistent with our findings in which APOE had an
effect on focal fMRI activity in relatively young persons that
was greater that that of fully penetrant autosomal dominant
A limitation of our study is the small number of subjects in
the APOE e3/4 (n = 4) genotype subgroup. It is possible that
our findings are random effects of interindividual variability
in fMRI BOLD signal. Unfortunately, the absolute number of
subjects eligible to participate in studies such as this is
low due to the rarity of FAD mutations, and therefore,
the number of subjects at-risk for FAD of various APOE
genotypes is serendipitous. Because of the small numbers, we
included one APOE e3/4 carrier that was of an outlying age
(55 years) compared with the remainder of the cohort.
Exclusion of this subject did not change the fMRI results
Our study population was comprised of persons of varying
FAD and APOE genotypes from a limited number of Mexican or
Mexican American families. This raises issues regarding the
validity of cognitive testing and generalizability of our findings
to other ethnic groups. As we employed neuropsychological
measures that have demonstrated utility in Latinos living in the
United States, the norms for comparison were derived from the
nonmutation carrying family members, and we also indepen-
dently administered a widely used global clinical measure (the
Clinical Dementia Rating scale), the determination of clinical
status of our subjects should be valid. The use of fMRI as an
outcome measure has advantages with regard to the applica-
bility of our findings to other populations. Prior studies of
neuropsychological scores, clinical diagnosis, and structural
MRI measures among African-Americans, Caucasians, and
Hispanics (DeCarli et al. 2008; Mungas et al. 2009) showed
only subtle interethnic differences with the main relationships
among these variables being consistent between groups. We
are unaware, however, of any studies of interethnic differences
in BOLD activation or its relationship to cognition. Nonethe-
less, the novelty encoding activation task employed in our
study, which was essentially nonlinguistic, is unlikely to be
influenced by cultural or educational factors and therefore has
promise as an endophenotype of utility across diverse ethnic
In conclusion, we found that the AD risk--conferring APOE
e4 genotype was associated with increased BOLD signal during
a novelty encoding task in multiple brain areas of asymptomatic
persons, whereas no such increase was observable in pre-
symptomatic persons carrying FAD mutations. A parsimonious
explanation is that APOE e4 exerts an effect on activation-
related focal cerebral blood flow that is at least partly
independent of AD risk and parenchymal AD pathology.
Increases in BOLD signal seen in APOE e4 carriers may not
then be related to cognitive compensation or reserve but
instead to an unidentified effect of the allelic variant on
cerebral vascular reactivity.
Public Health Service (K08 AG-22228); California Department
of Health Services (#04--35522); University of California
Institute for Mexico and the United States; the National
Institute on Aging (Alzheimer’s Disease Research Center grant
P50 AG-16570); the Easton Consortium for Alzheimer’s Disease
Drug Discovery and Biomarkers; the General Clinical Research
Centers Program (M01-RR00865); an Alzheimer’s Disease
Research Center of California grant; the Sidell Kagan Founda-
tion; and the Shirley and Jack Goldberg Trust.
Conflict of Interest: None declared.
Artiola-i-Fortuny L, Hermosillo D, Heaton RK, Pardee RE. 1999. Manual
de Normal y Procedimientos para la Bateria Neuropsiclogia en
Espanol. Tucson (AZ): Neuropsychology Press.
Bassett SS, Yousem DM, Cristinzio C, Kusevic I, Yassa MA, Caffo BS,
Zeger SL. 2006. Familial risk for Alzheimer’s disease alters fMRI
activation patterns. Brain. 129:1229--1239.
Bondi MW, Houston WS, Eyler LT, Brown GG. 2005. fMRI evidence of
compensatory mechanisms in older adults at genetic risk for
Alzheimer disease. Neurology. 64:501--508.
Bookheimer SY, Magdalena H, Strojwas BS, Cohen MS, Saunders AM,
Pericak-Vance MA, Mazziotta JC, Small GW. 2000. Patterns of brain
activation in people at risk for Alzheimer’s disease. N Engl J Med.
Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC,
Small GW, Roses AD, Haines JL, Pericak-Vance MA. 1993. Gene dose
of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease
in late onset families. Science. 261:921--923.
Cummings JL. 1995. Anatomic and behavioral aspects of frontal-
subcortical circuits. Ann N Y Acad Sci. 769:1--13.
DeCarli C, Reed BR, Jagust W, Martinez O, Ortega M, Mungas D. 2008.
Brain behavior relationships among African Americans, Whites, and
Hispanics. Alzheimer Dis Assoc Disord. 22:382--391.
Dickerson BC, Salat DH, Bates JF, Atiya M, Killiany RJ, Greve DN,
Dale AM, Stern CE, Blacker D, Albert MS, et al. 2004. Medial temporal
lobe function and structure in mild cognitive impairment. Ann
Evans AC, Marrett S, Neelin P, Collins L, Worsley K, Dai W, Milot S,
Meyer E, Bub D. 1992. Anatomical mapping of functional activation
in stereotactic coordinate space. Neuroimage. 1:43--53.
Filippini N, MacIntosh BJ, Hough MG, Goodwin GM, Frisoni GB,
Smith SM, Matthews PM, Beckmann CF, Mackay CE. 2009. Distinct
patterns of brain activity in young carriers of the APOE-epsilon4
allele. Proc Natl Acad Sci U S A. 106:7209--7214.
Fleisher AS, Houston WS, Eyler LT, Frye S, Jenkins C, Thal LJ, Bondi MW.
2005. Identification of Alzheimer disease risk by functional
magnetic resonance imaging. Arch Neurol. 62:1881--1888.
FMRI in FAD
Ringman et al.
Fox NC, Kennedy AM, Harvey RJ, Lantos PL, Roques PK, Collinge J,
Hardy J, Hutton M, Stevens JM, Warrington EK, et al. 1997. Clinico-
pathological features of familial Alzheimer’s disease associated with
the M139V mutation in the presenilin 1 gene. Pedigree but not
factor. Brain. 120:491--501.
Jenkinson M, Smith S. 2001. A global optimisation method for robust
affine registration of brain images. Med Image Anal. 5:143--156.
Johnson KA, Lopera F, Jones K, Becker A, Sperling R, Hilson J,
Londono J, Siegert I, Arcos M, Moreno S, et al. 2001. Presenilin-1-
associated abnormalities in regional cerebral perfusion. Neurology.
Johnson SC, Schmitz TW, Trivedi MA, Ries ML, Torgerson BM,
Carlsson CM, Asthana S, Hermann BP, Sager MA. 2006. The influence
of Alzheimer disease family history and apolipoprotein E epsilon4
on mesial temporal lobe activation. J Neurosci. 26:6069--6076.
Klunk WE, Price JC, Mathis CA, Tsopelas ND, Lopresti BJ, Ziolko SK,
Bi W, Hoge JA, Cohen AD, Ikonomovic MD, et al. 2007. Amyloid
deposition begins in the striatum of presenilin-1 mutation carriers
from two unrelated pedigrees. J Neurosci. 27:6174--6184.
Koivunen J, Verkkoniemi A, Aalto S, Paetau A, Ahonen JP, Viitanen M,
Nagren K, Rokka J, Haaparanta M, Kalimo H, et al. 2008. PET amyloid
ligand [11C]PIB uptake shows predominantly striatal increase in
variant Alzheimer’s disease. Brain. 131:1845--1853.
Li C, Zheng J, Wang J, Gui L, Li C. 2009. An FMRI stroop task study
of prefrontal cortical function in normal aging, mild cognitive
Lind J, Persson J, Ingvar M, Larsson A, Cruts M, Van Broeckhoven C,
Adolfsson R, Backman L, Nilsson LG, Petersson KM, et al. 2006.
Reduced functional brain activity response in cognitively intact
apolipoprotein E epsilon4 carriers. Brain. 129:1240--1248.
Loewenstein DA, Rubert MP, Arguelles T, Duara R. 1995. Neuro-
psychological test performance and prediction of functional
capacities among Spanish-speaking and English-speaking patients
with dementia. Arch Clin Neuropsychol. 10:75--88.
Mattsson N, Zetterberg H, Hansson O, Andreasen N, Parnetti L,
Jonsson M, Herukka SK, van der Flier WM, Blankenstein MA,
Ewers M, et al. 2009. CSF biomarkers and incipient Alzheimer
disease in patients with mild cognitive impairment. JAMA.
Mondadori CR, Buchmann A, Mustovic H, Schmidt CF, Boesiger P,
Nitsch RM, Hock C, Streffer J, Henke K. 2006. Enhanced brain
activity may precede the diagnosis of Alzheimer’s disease by
30 years. Brain. 129:2908--2922.
Morris JC. 1997. Clinical dementia rating: a reliable and valid diagnostic
and staging measure for dementia of the Alzheimer type. Int
Psychogeriatr. 9(Suppl 1):173--176; discussion 177--178.
Mullan M, Tsuji S, Miki T, Katsuya T, Naruse S, Kaneko K, Shimizu T,
Kojima T, Nakano I, Ogihara T, et al. 1993. Clinical comparison of
Alzheimer’s disease in pedigrees with the codon 717 Val/Ile
mutation in the amyloid precursor protein gene. Neurobiol Aging.
Mungas D, Reed BR, Crane PK, Haan MN, Gonzalez H. 2004. Spanish and
English Neuropsychological Assessment Scales (SENAS): further
development and psychometric characteristics. Psychol Assess.
Mungas D, Reed BR, Farias ST, Decarli C. 2009. Age and education
effects on relationships of cognitive test scores with brain structure
disease.Curr Alzheimer Res.
Murrell J, Ghetti B, Cochran E, Macias-Islas MA, Medina L, Varpetian A,
Cummings JL, Mendez MF, Kawas C, Chui H, et al. 2006. The A431E
mutation in PSEN1 causing familial Alzheimer’s disease originating
in Jalisco State, Mexico: an additional fifteen families. Neurogenetics.
Okello A, Koivunen J, Edison P, Archer HA, Turkheimer FE, Nagren K,
Bullock R, Walker Z, Kennedy A, Fox NC, et al. 2009. Conversion of
amyloid positive and negative MCI to AD over 3 years: an 11C-PIB
PET study. Neurology. 73:754--760.
Ringman JM, Medina LD, Rodriguez-Agudelo Y, Chavez M, Lu P,
Cummings JL. 2009. Current concepts of mild cognitive impairment
and their applicability to persons at-risk for familial Alzheimer’s
disease. Curr Alzheimer Res. 6:341--346.
Seeley WW, Menon V, Schatzberg AF, Keller J, Glover GH, Kenna H,
Reiss AL, Greicius MD. 2007. Dissociable intrinsic connectivity
networks for salience processing and executive control. J Neurosci.
Smith SM. 2002. Fast robust automated brain extraction. Hum Brain
Smith SM, Zhang Y, Jenkinson M, Chen J, Matthews PM, Federico A,
De Stefano N. 2002. Accurate, robust, and automated longitudinal
and cross-sectional brain change analysis. Neuroimage. 17:479--489.
Stern CE, Corkin S, Gonzalez RG, Guimaraes AR, Baker JR, Jennings PJ,
Carr CA, Sugiura RM, Vedantham V, Rosen BR. 1996. The
hippocampal formation participates in novel picture encoding:
evidence from functional magnetic resonance imaging. Proc Natl
Acad Sci U S A. 93:8660--8665.
Stroop JR. 1935. Studies of interference in serial verbal reactions. J Exp
Teng EL, Hasegawa K, Homma A, Imai Y, Larson E, Graves A,
Sugimoto K, Yamaguchi T, Sasaki H, Chiu D, et al. 1994. The
Cognitive Abilities Screening Instrument (CASI): a practical test for
cross-cultural epidemiological studies of dementia. Int Psychoger-
iatr. 6:45--58; discussion 62.
Troncoso JC, Cataldo AM, Nixon RA, Barnett JL, Lee MK, Checler F,
Fowler DR, Smialek JE, Crain B, Martin LJ, et al. 1998. Neuropathol-
ogy of preclinical and clinical late-onset Alzheimer’s disease. Ann
Wechsler D. 1987. Wechsler Memory Scale-Revised manual. San
Antonio (CA): Psychological Corporation.
Woolrich MW, Behrens TE, Beckmann CF, Jenkinson M, Smith SM. 2004.
Multilevel linear modelling for FMRI group analysis using Bayesian
inference. Neuroimage. 21:1732--1747.
Woolrich MW, Ripley BD, Brady M, Smith SM. 2001. Temporal
autocorrelation in univariate linear modeling of FMRI data. Neuro-
Worsley KJ, Evans AC, Marrett S, Neelin P. 1992. A three-dimensional
statistical analysis for CBF activation studies in human brain. J Cereb
Blood Flow Metab. 12:900--918.
Yassa MA, Verduzco G, Cristinzio C, Bassett SS. 2008. Altered fMRI
activation during mental rotation in those at genetic risk for
Alzheimer disease. Neurology. 70:1898--1904.
Yescas P, Huertas-Vazquez A, Villarreal-Molina MT, Rasmussen A, Tusie-
Luna MT, Lopez M, Canizales-Quinteros S, Alonso ME. 2006. Founder
effect for the Ala431Glu mutation of the presenilin 1 gene causing
early-onset Alzheimer’s disease in Mexican families. Neurogenetics.
demographicallydiverseolder persons. Psychol Aging.
Cerebral Cortex April 2011, V 21 N 4 883