Cell number changes in Alzheimer's disease relate to dementia and not to plaques and tagles

Full text

BRAIN
A JOURNAL OF NEUROLOGY
Cell number changes in Alzheimer’s disease relate
to dementia, not to plaques and tangles
Carlos Humberto Andrade-Moraes,
1
Ana V. Oliveira-Pinto,
1
Emily Castro-Fonseca,
1
Camila G. da Silva,
1
Daniel M. Guimara
˜es,
1
Diego Szczupak,
1
Danielle R. Parente-Bruno,
1
Ludmila R. B. Carvalho,
1
Lı
´via Polichiso,
2,3
Bruna V. Gomes,
1
Lays M. Oliveira,
1
Roberta D. Rodriguez,
2
Renata E. P. Leite,
2
Renata E. L. Ferretti-Rebustini,
2,4
Wilson Jacob-Filho,
2,5
Carlos A. Pasqualucci,
2
Lea T. Grinberg
2,3
and Roberto Lent
1,6
1 Institute of Biomedical Sciences, Federal University of Rio de Janeiro, Brazil
2 Ageing Brain Study Group, Department of Pathology, LIM 22, University of Sa
˜o Paulo Medical School, Sa
˜o Paulo, Brazil
3 Department of Neurology, University of California, San Francisco, USA
4 University of Sa
˜o Paulo Nursing School, Sa
˜o Paulo, Brazil
5 Division of Geriatrics, University of Sa
˜o Paulo, Brazil
6 National Institute of Translational Neuroscience, Ministry of Science and Technology, Brazil
Correspondence to: Prof. Roberto Lent,
Instituto de Cie
ˆncias Biome
´dicas,
Universidade Federal do Rio de Janeiro,
Av. Carlos Chagas 373, Sl. F1-31,
Ilha do Funda
˜o, Rio de Janeiro,
CEP 21941-902, Brazil
E-mail: rlent@icb.ufrj.br
Alzheimer’s disease is the commonest cause of dementia in the elderly, but its pathological determinants are still debated.
Amyloid-bplaques and neurofibrillary tangles have been implicated either directly as disruptors of neural function, or indirectly
by precipitating neuronal death and thus causing a reduction in neuronal number. Alternatively, the initial cognitive decline has
been attributed to subtle intracellular events caused by amyloid-boligomers, resulting in dementia after massive synaptic
dysfunction followed by neuronal degeneration and death. To investigate whether Alzheimer’s disease is associated with
changes in the absolute cell numbers of ageing brains, we used the isotropic fractionator, a novel technique designed to
determine the absolute cellular composition of brain regions. We investigated whether plaques and tangles are associated
with neuronal loss, or whether it is dementia that relates to changes of absolute cell composition, by comparing cell numbers
in brains of patients severely demented with those of asymptomatic individuals—both groups histopathologically diagnosed as
Alzheimer’s—and normal subjects with no pathological signs of the disease. We found a great reduction of neuronal numbers in
the hippocampus and cerebral cortex of demented patients with Alzheimer’s disease, but not in asymptomatic subjects with
Alzheimer’s disease. We concluded that neuronal loss is associated with dementia and not the presence of plaques and tangles,
which may explain why subjects with histopathological features of Alzheimer’s disease can be asymptomatic; and exclude
amyloid-bdeposits as causes for the reduction of neuronal numbers in the brain. We found an increase of non-neuronal cell
numbers in the cerebral cortex and subcortical white matter of demented patients with Alzheimer’s disease when compared with
asymptomatic subjects with Alzheimer’s disease and control subjects, suggesting a reactive glial cell response in the former that
may be related to the symptoms they present.
Keywords: ageing; amyloid-b; dementia; neuronal loss; isotropic fractionator
doi:10.1093/brain/awt273 Brain 2013: 136; 3738–3752 |3738
Received November 11, 2012. Revised August 4, 2013. Accepted August 4, 2013. Advance Access publication October 17, 2013
ßThe Author (2013). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved.
For Permissions, please email: journals.permissions@oup.com
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

Introduction
Alzheimer’s disease is currently the commonest cause of dementia
in the elderly. Since its first description (Alzheimer, 1907; Fischer,
1907; Goedert, 2009), amyloid-bneuritic plaques and neurofibril-
lary tangles have often been considered as the main pathological
hallmarks of Alzheimer’s disease (Mattson, 2004; Nestor et al.,
2004; Palop and Mucke, 2010), despite growing post-mortem
evidence for amyloid-bdeposition in the cognitively normal elderly
(Price and Morris, 1999; Bennett et al., 2006; Negash et al.,
2011). Indeed, abundant amyloid-bneuritic plaques can be
found in the brains of ageing subjects with documented normal
cognition, a condition defined as asymptomatic Alzheimer’s dis-
ease. It is conceivable, therefore, that amyloid-bdeposition per
se is not the cause of cognitive impairment in Alzheimer’s disease.
In fact, intracellular transcriptional and post-transcriptional events
have been recently suggested as determinants of the first signs of
cognitive dysfunction (e.g. abnormal production of amyloid-b
oligomers: Gong et al., 2003; Lacor et al., 2004; Ferreira and
Klein, 2011; and tau hyperphosphorylation: De Felice et al.,
2008; Jin et al., 2011; Flunkert et al., 2012). Plaques and tangles,
then, would be late events of the pathological cascade.
Considering their late appearance in the brain, amyloid-bplaques
were even proposed to act as protective buffers against toxic
oligomers, being a possible antioxidant defence mechanism (Lee
et al., 2004; Lesne
´et al., 2008).
Alternatively, Alzheimer’s disease aetiopathogeny has been
related to anomalous cell cycle re-entrance before neuronal
death (Herrup, 2010, 2012), presumably causing a relevant reduc-
tion in the number of neurons, and the subsequent progressive
cognitive impairment that ends in dementia (Whitehouse et al.,
1982; Neniskyte et al., 2011). According to this view, the initial
cognitive decline would be caused by subtle intracellular events,
and dementia would follow due to massive cell death. Indeed, a
significant neuronal loss was reported in the hippocampus of
patients with Alzheimer’s disease (Simic et al., 1997, Korbo
et al., 2004; Giannakopoulos et al., 2009). For the neocortex as
a whole (Bundgaard et al., 2001) and cerebellum (Andersen et al.,
2003, 2012), despite hypometabolism and atrophy detected by
neuroimaging in demented patients (Thompson et al., 2003;
Che
´telat et al., 2008; Thomann et al., 2008), no evidence of sig-
nificant neuronal loss has been found.
To clarify these inconsistencies, we sought to use the isotropic
fractionator (Herculano-Houzel and Lent, 2005) to investigate
whether dementia is associated with changes in the quantitative
cellular composition of ageing brains. Using this novel technique, it
is possible to determine the absolute cell composition of the brain
and of brain regions with considerable precision (Azevedo et al.,
2009, 2013; Lent et al., 2012). Therefore, by comparing the ab-
solute neuronal and non-neuronal cell composition of the brains of
cognitively-normal elderly with those of severely demented and
asymptomatic patients with Alzheimer’s disease, we were able to
investigate whether neuronal loss severity is associated with
plaque and tangle burden, or whether it is dementia that could
be related to changes of absolute cell composition.
We found a great reduction of neurons in the hippocampus and
cerebral cortex of demented patients with Alzheimer’s disease
when compared with asymptomatic subjects with Alzheimer’s dis-
ease and the cognitively-normal elderly, but no significant differ-
ence between the two latter groups. In contrast, we found an
increase in non-neuronal cell numbers in the cortex and subcortical
white matter of the demented Alzheimer’s group when compared
to asymptomatic Alzheimer’s subjects and controls. The cerebel-
lum, however, showed no changes in its absolute cell composition
in all groups.
Materials and methods
Subjects
Fourteen female brains (Table 1) were obtained from the Brazilian
Brain Bank of the Aging Brain Study Group (Grinberg et al., 2007)
of the University of Sa
˜o Paulo Medical School. Brains were removed
and fixed within 24 h of death. All procedures were approved by the
local and national Ethics Committees. Informed consent for removal of
the brains was provided by a knowledgeable informant, who also
responded to semi-structured questionnaires designed to evaluate sev-
eral functional domains (Ferretti et al., 2010). Cognition was assessed
using the Clinical Dementia Rating Scale (Morris, 1993) and by the
Informant Questionnaire on Cognitive Decline in the Elderly–
Retrospective Version (IQCODE; Jorm and Jacomb, 1989).
The control group comprised five elderly females with no cognitive
impairment, who died of non-neurological causes. Five demented
elderly female brains with histopathological diagnosis of Alzheimer’s
disease were included in the demented-Alzheimer’s disease group.
Four non-demented female brains, but with histopathological features
of Alzheimer’s disease were included in the asymptomatic Alzheimer’s
disease group (Table 1).
Brain handling, fixation and dissection
Because no significant differences were found in mass or in neuronal,
non-neuronal and total cell numbers between right and left hemi-
spheres by Azevedo et al. (2009), each brain was split on the mid-
sagittal plane, and the left hemisphere immersed in 2%
paraformaldehyde at 4C, whereas samples of right hippocampus,
amygdala, cortical areas, brainstem and cerebellum were collected
for histopathological analysis, according to the Brazilian Brain Bank
protocol (Grinberg et al., 2007). After removing the meninges and
blood vessels (Fig. 1A), the left hemisphere was segmented into
seven regions of interest: cerebellum, hippocampal formation with
amygdala (hereafter termed ‘hippocampal formation’), frontal lobe
grey and white matters, grey matter and white matter from other
lobes, and remaining regions. The left cerebellar hemisphere was cut
through the cerebellar peduncles, allowing its separation from the
brainstem (Fig. 1B and C). The hippocampal formation was separated
from the hemisphere by sectioning along the collateral sulcus up to the
caudal-most coronal level of the callosal splenium (Fig. 1D). The fron-
tal lobe was separated from the rest of the brain by a section made
along the central sulcus (Fig. 1E), and the whole hemisphere was cut
coronally into slices 1-cm thick (Fig. 1E and F). Next, the cerebral
cortex and the subcortical white matter were separated from the
remaining regions (basal ganglia, diencephalon, mesencephalon, pons
and medulla oblongata) by cutting along the dorsolateral surface of
Cell numbers in Alzheimer’s brains Brain 2013: 136; 3738–3752 |3739
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

the striatum, diencephalon and mesencephalon in each section
(Fig. 1G). Finally, the coronal slices had the cortical grey matter dis-
sected away from the underlying white matter by careful shaving
around the gyri with a scalpel until grey and white matters were com-
pletely separated (not shown in Fig. 1).
The seven regions of interest were frozen in phosphate-buffered
saline (pH 7.4) at 20C and subjected, separately, to the isotropic
fractionator (Herculano-Houzel and Lent, 2005). Each region was
weighed and then cut into smaller pieces to be homogenized by an
automatic machine composed of tissue grinders operated by servo-
motors under control of the experimenters (Azevedo et al., 2013).
Histopathological analysis and
diagnostic criteria
Tissue blocks from the right hemisphere were embedded in paraffin,
cut and submitted to routine staining and immunohistochemistry
(Fig. 2) for diagnostic purposes, according to a previously published
protocol (Grinberg et al., 2007). In brief, immunohistochemistry was
performed using antibodies against amyloid-b(4G8, 1/5000, Signet
Laboratories), phospho-tau (PHF-1, 1/1000, provided by Peter
Davies, New York), and -synuclein (EQV-1, 1/10 000, provided by
Kenji Ueda, Tokyo, Japan). If frontotemporal lobar degeneration with
TDP-43 inclusions (FTLD-TDP) was suspected, immunostaining for
TDP-43 (1/500, ProteinTech) was performed, according to the brain
bank protocol (Grinberg et al., 2007). Staining was performed on
8-mm thick sections cut from paraffin blocks and mounted on glass
slides. For all antibodies, immunoperoxidase was employed using an
avidin-biotin complex detection system (VectastainÕABC kit; Vector
Laboratories) with 3,3’diaminobenzidine as the chromogen. Slides
were pretreated for antigen retrieval by immersion in citrate pH 6.0
in a steamer at 95C for 45 min. The sections were incubated over-
night at 4C with the primary antibodies and for 1 h at room tempera-
ture with species-specific biotinylated secondary antibodies.
For Alzheimer’s disease, amyloid-bneuritic plaques (Fig. 2) were as-
sessed following the neuropathological guidelines of the Consortium to
Establish a Registry for Alzheimer’s disease (CERAD: Mirra et al., 1991),
receiving semi-quantitative descriptors: none (0), scarce, moderate or
frequent (Table 1). Neurofibrillary tangles (Fig. 2) were assigned a
score (0–VI) according to the Braak and Braak (1991) staging system.
The control subjects had no history of cognitive and/or behavioural
deficits, cerebrovascular disease, or alcohol/drug abuse. On neuro-
pathological evaluation, the brains showed no neuritic plaques
(Fig. 2). Accordingly, the CERAD score was 0 for these subjects.
Neurofibrillary changes were confined to transentorhinal and entorh-
inal cortex, as well as hippocampus; thus, their Braak neurofibrillary
tangles scores ranged between 0 and II (Fig. 2 and Table 1). One
control case had mild argyrophilic grain disease and therefore the
Braak staging could not be reliably assessed in the hippocampus.
However, in this case there were no tangles outside the hippocampus,
assuring that the Braak stage was 5III.
The asymptomatic Alzheimer’s subjects had no history of behav-
ioural deficits, cerebrovascular disease, or alcohol/drug abuse,
and were cognitively intact. On neuropathological evaluation, how-
ever, their brains showed moderate or great quantities of neuritic
amyloid-bplaques, and the neurofibrillary tangles Braak scores were
at least IV (Fig. 2 and Table 1), indicating that the changes extended
beyond the limbic areas.
The demented patients with Alzheimer’s disease were individuals
who had received a clinical diagnosis of dementia (Clinical Dementia
Rating Scale = 3), and whose neuropathological brain examination
showed at least a moderate number of neuritic amyloid-bplaques,
and neurofibrillary tangles Braak scores 5III, indicating that changes
extended outside the hippocampal formation. The cases did not show
any other potential causes of cognitive decline (Fig. 2 and Table 1).
Chemomechanical dissociation, nuclei
staining, immunocytochemistry and
counting procedures
The isotropic fractionator was described previously by Herculano-
Houzel and Lent (2005), applied to the human brain by Azevedo
et al. (2009), and automated by Azevedo et al. (2013). Briefly, after
the neural tissue is properly fixated, small fragments are collected and
placed into the mortars of glass tissue grinders. A saline-detergent
solution (40 mM sodium citrate; 1% Triton
TM
X-100) was added,
Table 1 Subjects demographic data
Age
(years)
Group Mass (g) CDR IQCODE Braak CERAD Histopathology Cause of death
71 Control 1403.2 0 3.11 5III* 0 Normal Pulmonary oedema
73 Control 1363.9 0 3.30 I 0 Normal Myocardiosclerosis
74 Control 1289.32 0 3.00 II 0 Normal Cardiac tamponade
82 Control 1385.9 0 3.00 II 0 Normal Pulmonary oedema
84 Control 1013.7 0 3.00 5III* 0 Normal Hemoperitoneum
82 Asymptomatic Alzheimer’s disease 1338.2 0 3.00 V Frequent Alzheimer’s disease Myocardiosclerosis
82 Asymptomatic Alzheimer’s disease 1332.0 0 3.00 IV Moderate Alzheimer’s disease Pulmonary oedema
80 Asymptomatic Alzheimer’s disease 1209.86 0 3.00 IV Moderate Alzheimer’s disease Pulmonary oedema
80 Asymptomatic Alzheimer’s disease 1190.32 0 3.00 VI Frequent Alzheimer’s disease Pulmonary thromboembolism
74 Demented Alzheimer’s disease 1402.2 3 5.00 IV Frequent Alzheimer’s disease Bronchopneumonia
83 Demented Alzheimer’s disease 1255.7 3 5.00 III Moderate Alzheimer’s disease Pulmonary oedema
86 Demented Alzheimer’s disease 1136.84 3 5.00 VI Frequent Alzheimer’s disease Bronchopneumonia
88 Demented Alzheimer’s disease 1280.16 3 5.00 IV Moderate Alzheimer’s disease Serofibrinous pericarditis
88 Demented Alzheimer’s disease 1204.2 3 5.00 VI Moderate Alzheimer’s disease Pulmonary oedema
*Argyrophilic grain disease.
CDR = Clinical Dementia Rating scale; CERAD = Consortium to Establish a Registry for Alzheimer’s Disease.
3740 |Brain 2013: 136; 3738–3752 C. H. Andrade-Moraes et al.
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

and through careful and constant translation and rotation movements
of a tightly coupled pestle, the tissue was disrupted chemomechani-
cally by the combination of the dissociative effects of the detergent
solution with the turbulent flow produced inside the grinder. The cell
membranes are broken, releasing the intact nuclei into the fluid. The
result was an isotropic suspension containing all cell nuclei from the
fractioned region.
For the purpose of staining the nuclei of all cells, the fluorescent DNA
marker 4’-6-diamino-2-phenylindole dihydrochloride (DAPI) was added
to the suspension. Intense agitation was then performed to disperse the
nuclei and achieve isotropy. Aliquots from the isotropic suspension were
collected, deposited into a haemocytometer (Neubauer chamber) and
imaged using a fluorescence microscope (Zeiss Axioplan). The average
nuclei density was determined by counting the number of nuclei within
Figure 1 After removing the meninges and blood vessels (A), the left brainstem and cerebellum were separated from the brain (B), and
the cerebellar peduncles were cut to isolate the cerebellum (C). The hippocampal formation was removed out from the hemisphere by
sectioning along the collateral sulcus until the caudalmost coronal level of the callosal splenium (D). The frontal lobe was then separated
from the rest of the brain by a section made along the central sulcus (E), and cut coronally into slices of 1 cm thickness (Eand F). Finally
(F), the cerebral cortex was separated from the remaining regions (basal ganglia, diencephalon, mesencephalon, pons and medulla
oblongata).
Cell numbers in Alzheimer’s brains Brain 2013: 136; 3738–3752 |3741
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

sectors of the coverslipped haemocytometer (1 mm
2
area; 0.1 mm
depth) for four aliquots. The total number of cells originally present in
the analysed region is then obtained by multiplying density by the total
volume of the suspension. To identify the fraction of neuronal nuclei
among the total number of DAPI-stained nuclei, another aliquot of the
isotropic suspension was collected and selectively immunolabelled with
mouse primary antibody against neuronal nuclear protein (NeuN,
Chemicon; MAB 377B clone A60 against murine cells; 1:200 in PBS,
overnight incubation at room temperature). Anti-NeuN antibody recog-
nizes the majority of neuronal subtypes in a variety of vertebrate species,
with the exception of cerebellar Purkinje cells, mitral cells of the olfac-
tory bulb, inferior olivary nuclei of brainstem, dentate neurons of cere-
bellar deep nuclei, retinal photoreceptors, and nigral neurons in some
rodents (Mullen et al., 1992; Wolf et al., 1996; Sarnat et al., 1998;
Kumar and Buckmaster, 2007). The quantitative contribution of these
cells as compared with the total brain numbers, however, is negligible
(Herculano-Houzel and Lent, 2005; Herculano-Houzel et al., 2006).
Then, the nuclei were washed in saline and incubated at room tempera-
ture for at least 2 h with the secondary antibody (Alexa FluorÕ555 anti-
mouse goat IgG, Molecular Probes; 1:200 in PBS) and normal goat
serum (1:10). By counting the number of NeuN-labelled nuclei among
at least 500 DAPI-stained nuclei, the percentage of neurons present in
the sample was determined and thus the total number of neurons in the
analysed region was estimated. The non-neuronal cells were quantified
as the difference between the total number of cells and the total
number of neurons. Photomicrographs were taken for routine documen-
tation using a Zeiss Axioplan fluorescence microscope. For all illustra-
tions, contrast and brightness of the pictures were adjusted using Corel
Draw X3.
Statistical analyses
The statistical analyses were performed by use of the Graph Prism 5.0
software. Analysis of variance (one-way ANOVA) and Tukey’s multiple
comparison post hoc tests were used to compare results from these
groups. The means are shown in Supplementary Tables 1–10, and
means with standard deviations (SD) are represented in Figs 3-6.
Results
We compared the absolute cell number and density between con-
trol subjects, asymptomatic individuals with Alzheimer’s disease,
and severely demented patients with Alzheimer’s disease.
Subjects’ demographic data are shown in Table 1. Numbers
shown below refer to both hemispheres together, after multiplying
the counts obtained for the left hemispheres by 2, as explained in
the ‘Materials and methods’ section.
Cerebellum
Although representing only 8.1–9.1% of total brain mass, the
cerebellum contains 77.5% to 79.8% of all brain neurons in
elderly females. ANOVA and multiple comparison post hoc test,
however, showed no significant differences in cerebellar mass,
absolute cell number or density among groups (Fig. 3A and B;
Figure 2 Histopathological features of some cases, taken from sections through the inferior temporal cortex. Upper row (A,Cand E),
immunostaining for amyloid-b, and lower row (B,Dand F), immunostaining for phospho-Tau (see ‘Materials and methods’ section for
details). (Aand B) Control (CTRL; 82 years, Clinical Dementia Rating Scale = 0, Braak II; see Table 1), showing no signs of pathology.
(Cand D) Demented patient with Alzheimer’s disease (AD, 74 years, Clinical Dementia Rating Scale = 3, Braak IV), with an abundant
number of neuritic plaques (in brown). (Eand F) Asymptomatic subject with Alzheimer’s disease (ASYMAD, 82 years, Clinical Dementia
Rating Scale = 0, Braak V), showing an advanced pattern of Alzheimer’s pathology. Neurofibrillary tangles are indicated with arrows, and
neuritic plaques indicated with circles in Dand F. Scale bars: A,Cand E= 1 mm; B,Dand F= 100 mm.
3742 |Brain 2013: 136; 3738–3752 C. H. Andrade-Moraes et al.
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

Supplementary Table 1). The non-neuronal/neuronal ratio and the
fractional distribution of neurons were similar among groups.
These results suggest that this brain region seems not to be
affected by Alzheimer’s disease, at least regarding absolute cellular
composition and cell density.
Hippocampal formation and amygdala
Using one-way ANOVA to compare absolute neuronal number in
the hippocampal formation (Fig. 3C), significant differences be-
tween groups were found. Tukey’s test indicated that only the de-
mented patients with Alzheimer’s disease group (110 million)
differed statistically from control subjects (270 million), which was
not the case for the asymptomatic-Alzheimer’s disease group (220
million). Neuronal loss in demented patients with Alzheimer’s dis-
ease results in more than half the number neurons than in control
subjects. For the absolute number of non-neuronal cells, ANOVA
revealed no significant difference among groups (Fig. 3C and
Supplementary Table 2). To verify whether the difference in neur-
onal number was due to variability of brain mass or by uncertainties
in dissection, we focused on cell density. ANOVA and multiple
comparison tests showed a significant reduction in neuronal density
in demented patients with Alzheimer’s disease when compared with
the control and asymptomatic Alzheimer’s disease groups, but not
when asymptomatic-subjects with Alzheimer’s disease and control
subjects were compared (Fig. 3D).
On the other hand, ANOVA did not reveal any significant dif-
ference in non-neuronal cell density among groups (Fig. 3D). The
non-neuronal/neuronal ratio was higher in the demented patients
with Alzheimer’s disease group when compared with controls and
asymptomatic patients with Alzheimer’s disease. Inversely, the
fractional distribution of neurons was lower in the demented
Alzheimer’s disease group (Supplementary Table 2). These differ-
ences represent a great reduction of 450% in absolute neuronal
number and in neuronal density of demented patients with
Alzheimer’s disease compared with control subjects and asymp-
tomatic patients with Alzheimer’s disease.
Frontal lobe
Grey matter
ANOVA comparing the three groups showed significant differ-
ences in absolute neuronal composition of the frontal grey
Figure 3 Absolute bilateral cell number (Aand C) and density (Band D) of cerebellar (Aand B) and hippocampal formation (Cand D)
neuronal, non-neuronal, and total cells in control, asymptomatic Alzheimer’s disease (ASYMAD) and demented Alzheimer’s disease (AD)
groups. Each bar represents mean and standard deviation. Significant differences are indicated by *P50.05, **P50.01.
Cell numbers in Alzheimer’s brains Brain 2013: 136; 3738–3752 |3743
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

matter. However, using a multiple comparison test, the only
significant difference was found between controls (4.17 billion)
and demented patients with Alzheimer’s disease (2.46 billion)
(Fig. 4A and Supplementary Table 3). We analysed the neuronal
density in addition to absolute numbers. ANOVA showed signifi-
cant differences in neuronal density of frontal grey matter
among all groups. However, when Tukey’s test was used, no sig-
nificant difference in neuronal density was found between control
and asymptomatic Alzheimer’s disease groups (Fig. 4B and
Supplementary Table 3).
Thus, a significant reduction of 41% in the absolute number
of neurons was found in the demented Alzheimer’s disease group
when compared with control subjects. Similarly, a significant re-
duction of 34% neuronal density in frontal grey matter was found
in the demented Alzheimer’s disease group when compared with
control subjects, and 26% when compared to the asymptomatic
patients with Alzheimer’s disease group. The fractional distribution
of neurons in the demented patients with Alzheimer’s disease was
lower than in the other groups, such as the proportion of all
neurons found in the whole brain.
ANOVA and multiple comparison test showed a significant in-
crease of 47% in absolute non-neuronal cell number in demented
patients with Alzheimer’s disease (10.97 billion) when compared
with control subjects (7.48 billion), and 49% when compared with
asymptomatic patients with Alzheimer’s disease (7.34 billion), but
no significant difference was found when control and asymptom-
atic Alzheimer’s disease groups were compared. Correcting for
mass effects, ANOVA and post hoc test showed a significant
increase of 64% in non-neuronal cell density in demented
Alzheimer’s disease group (47.26 million/g) when compared
with controls (28.79 million/g), and 52% when compared with
asymptomatic subjects with Alzheimer’s disease (30.98 million/
g). However, no significant difference was found when asymp-
tomatic patients with Alzheimer’s disease and the control group
were compared (Fig. 4A).
The non-neuronal/neuronal ratio was higher in demented-
patients with Alzheimer’s disease when compared with the other
two groups. However, ANOVA showed no significant differences
of frontal lobe grey matter mass among groups (Supplementary
Table 3).
White matter
ANOVA showed no significant statistical difference of absolute cell
composition of the frontal lobe white matter between controls,
asymptomatic patients with Alzheimer’s disease and the demented
Alzheimer’s disease group (Fig. 4C). However, post hoc tests
showed a significant increase of 32% in non-neuronal cell density
in demented patients with Alzheimer’s disease (86.68 million/g)
when compared with controls (65.66 million/g), and 19% when
compared with asymptomatic subjects with Alzheimer’s disease
(72.61 million/g). However, no significant difference was found
when control subjects and asymptomatic patients with Alzheimer’s
disease were compared (Fig. 4D and Supplementary Table 4). The
non-neuronal/neuronal ratio was higher in demented-patients
with Alzheimer’s disease.
Other lobes
Grey matter
ANOVA and multiple comparison test found significant differences
in the absolute neuronal composition between control subjects
(7.62 billion) and demented patients with Alzheimer’s disease
(5.91 billion), and between asymptomatic subjects with
Alzheimer’s disease (7.69 billion) and the demented Alzheimer’s
disease group, but not between control subjects and asymptom-
atic patients with Alzheimer’s disease (Fig. 4E). Thus, a significant
reduction of 22% in the absolute number of neurons of the grey
matter was found in the demented patients with Alzheimer’s dis-
ease when compared with control subjects, and about the same
proportion when compared with the asymptomatic subjects with
Alzheimer’s disease. Using ANOVA and Tukey’s test, a significant
reduction of 24% in neuronal density was found in demented
patients with Alzheimer’s disease when compared with controls,
and almost 36% when compared with asymptomatic subjects with
Alzheimer’s disease (Fig. 4F). Thus, the proportion of neurons was
lower in Alzheimer’s disease when compared with the other two
groups (Supplementary Table 5).
When ANOVA and multiple comparison test of absolute non-
neuronal cell number was performed among all groups, a signifi-
cant increase of 45% in the demented patients with Alzheimer’s
disease (17.35 billion) was found when compared with the control
group (11.93 billion), and the same proportion when compared
with the asymptomatic subjects with Alzheimer’s disease (11.85
billion), but was not significant when control and asymptomatic
Alzheimer’s disease groups were compared. ANOVA and multiple
comparison test found a significant increase of 42% in non-
neuronal cell density in demented patients with Alzheimer’s dis-
ease (54.59 million/g) when compared with controls (38.47 mil-
lion/g), but not when compared to asymptomatic subjects with
Alzheimer’s disease (44.58 million/g), and no significant difference
was found when asymptomatic subjects with Alzheimer’s subjects
and controls were compared (Fig. 4B and Supplementary Table 5).
The non-neuronal/neuronal ratio was higher in demented-
Alzheimer’s disease when compared to the other groups.
White matter
ANOVA found significant differences in the absolute non-neuronal
cell number among all groups. Multiple comparison test analysis
showed a significant increase of 26% in absolute non-neuronal
cells in demented patients with Alzheimer’s disease (26.4 billion)
when compared with control subjects (20.93 billion), and the same
proportion when compared with asymptomatic patients with
Alzheimer’s disease (20.92 billion). However, no significant differ-
ence was found when control and asymptomatic Alzheimer’s
disease groups were compared (Fig. 4G). ANOVA showed a sig-
nificant difference in non-neuronal cell density of the white matter
among control, asymptomatic subjects with Alzheimer’s disease
and demented patients with Alzheimer’s disease (Fig. 4H).
Tukey’s post hoc multiple comparison test showed a significant
increase of 30% and 28% in the demented patients with
Alzheimer’s disease (99.92 million/g) when compared with control
subjects (76.71 million/g) and to asymptomatic subjects with
3744 |Brain 2013: 136; 3738–3752 C. H. Andrade-Moraes et al.
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

Figure 4 Absolute bilateral cell number (A,C,E, and G) and density (B,D,F, and H) for neuronal, non-neuronal, and total cells in the grey
(Aand B) and white (Cand D) matter of the frontal lobe (A–D) and of the other lobes as a whole (E–H) from control, asymptomatic
Alzheimer’s disease (ASYMAD) and demented Alzheimer’s disease (AD) groups. Each bar represents the mean and standard deviation.
Significant differences are indicated by *P50.05, **P50.01, and ***P50.001.
Cell numbers in Alzheimer’s brains Brain 2013: 136; 3738–3752 |3745
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

Alzheimer’s disease (78.17 million/g), respectively. However, no
significant difference was found between control and asymptom-
atic Alzheimer’s disease groups.
The non-neuronal/neuronal ratio was higher in demented
patients with Alzheimer’s disease when compared with the other
groups (Supplementary Table 6).
Cerebral cortex
The cerebral cortex as a whole (here comprising cortical grey and
subcortical white matter) in cognitively normal elderly females,
was found to represent 82% of brain mass. Neurons therein
amount to 12.7 billion (19%), and non-neuronal cells reach
54.9 billion (81%).
Grey matter
ANOVA of the absolute number of neurons showed significant
differences among groups (Fig. 5A and Supplementary Table 7).
Using Tukey’s multiple test, a significant reduction of 29%
was found in the demented-patients with Alzheimer’s disease
(8.38 billion) when compared with controls (11.8 billion), and
25% when compared with the asymptomatic subjects with
Alzheimer’s disease (11.12 billion), but no difference when asymp-
tomatic Alzheimer’s disease and control groups were compared.
ANOVA also showed significant differences of absolute non-
neuronal cell number in the cerebral cortex, among all groups.
Using the multiple comparison post hoc test, a significant increase
of 46% was revealed in demented patients with Alzheimer’s dis-
ease (28.32 billion) when compared with control subjects (19.4 bil-
lion), and a similar increase when compared with asymptomatic
patients with Alzheimer’s disease (19.2 billion). However, no signifi-
cant difference was found when control and asymptomatic subjects
with Alzheimer’s disease were compared. For non-neuronal cell dens-
ity, one-way ANOVA showed a significant difference among groups
(Fig. 5B). When compared with control subjects (34.04 million/g), a
significant increase of 51% was found in the demented patients with
Alzheimer’s disease (51.49 million/g) and of 35% when compared
with the asymptomatic subjects with Alzheimer’s subjects (38.19 mil-
lion/g). However, no significant difference was found between
asymptomatic Alzheimer’s disease and control groups.
Figure 5 Absolute bilateral cell number (Aand C) and density (Band D) for neuronal, non-neuronal, and total cells in the grey (Aand B)
and white (Cand D) matter of the entire cerebral cortex from control, asymptomatic Alzheimer’s disease (ASYMAD) and demented
Alzheimer’s disease (AD) groups. Each bar represents the mean and standard deviation. Significant differences are indicated by *P50.05,
**P50.01, and ***P50.001.
3746 |Brain 2013: 136; 3738–3752 C. H. Andrade-Moraes et al.
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

The non-neuronal/neuronal ratio was higher in demented
patients with Alzheimer’s disease when compared with the other
groups (Supplementary Table 7). Inversely, the whole cerebral
cortex of demented patients with Alzheimer’s disease contains a
lower proportion of all neurons in the brain (Supplementary
Tables 7 and 10). ANOVA showed no significant difference of
cortical mass among groups.
Correcting for mass, ANOVA revealed a significant difference of
neuronal density in the cerebral cortex among groups (Fig. 5B).
Post hoc analysis showed a significant reduction of 26% in
demented patients with Alzheimer’s disease (15.24 million/g)
when compared with control subjects (20.7 million/g) and
31% when compared with the asymptomatic subjects with
Alzheimer’s disease (22.12 million/g), but no difference when
asymptomatic Alzheimer’s disease and control groups were com-
pared (Supplementary Table 7).
White matter
ANOVA showed no significant difference in the absolute neuronal
composition and neuronal density of total subcortical white matter
between control, asymptomatic subjects with Alzheimer’s disease,
and demented patients with Alzheimer’s disease. The proportion
of neurons was low and similar in all groups (Fig. 5C and D).
On the other hand, although one-way ANOVA showed a sig-
nificant difference of absolute non-neuronal cell numbers among
groups, multiple comparison post hoc analysis revealed a signifi-
cant increase of almost 27% in the demented Alzheimer’s disease
group (44.97 billion) when compared with control subjects (35.51
billion), but neither when compared with asymptomatic patients
with Alzheimer’s disease (40.88 billion), nor when control and
asymptomatic Alzheimer’s disease groups were compared.
Furthermore, we found a significant increase of non-neuronal
cell density of 31% in the demented Alzheimer’s disease group
(94 million/g) when compared with control subjects (71.75 mil-
lion/g), and 25% when compared with asymptomatic subjects
with Alzheimer’s disease (75.36 million/g), but no difference was
found when control and asymptomatic Alzheimer’s disease groups
were compared.
The non-neuronal/neuronal ratio was higher in demented
patients with Alzheimer’s disease when compared with the other
groups (Supplementary Tables 8 and 10).
Remaining regions
The remaining regions comprise the basal nuclei, diencephalon and
brainstem considered together for counting purposes.
ANOVA and Tukey’s post hoc test indicated a significant
reduction of 45% in the absolute neuronal number of the remain-
ing regions in demented patients with Alzheimer’s disease (330
million) when compared with controls (600 million), and 47%
when compared with asymptomatic subjects with Alzheimer’s dis-
ease (620 million). Further, a significant reduction of 45% of neur-
onal density was found in the demented Alzheimer’s disease group
(3.3 million/g) when compared with control subjects (6 million/g),
and 49% when compared with asymptomatic subjects with
Alzheimer’s disease (6.5 million/g). However, there was no signifi-
cant difference between control and asymptomatic Alzheimer’s
disease groups (Supplementary Table 9). Remaining regions from
demented patients with Alzheimer’s disease contain a lower pro-
portion of all neurons in the brain when compared to the other
groups.
On the other hand, neither ANOVA nor Tukey’s test showed
any significant difference of absolute non-neuronal cell compos-
ition and non-neuronal cell density between these groups.
The non-neuronal/neuronal ratio was higher in demented
patients with Alzheimer’s disease when compared with the other
groups (Supplementary Table 9).
Whole brain
One-way ANOVA showed no significant difference of absolute neur-
onal composition and neuronal density in the whole brain between
controls (67.31 billion/52.13 million/g), asymptomatic patients with
Alzheimer’s disease (63.16 billion/49.83 million/g) and demented
patients with Alzheimer’s disease (67.82 billion/53.66 million/g),
and the same was found when using the multiple comparison test
(Fig. 6A and B; Supplementary Table 10). In contrast, ANOVA and
the post hoc test indicated a significant increase of 23% in absolute
non-neuronal cells in demented patients with Alzheimer’s disease
(96.8 billion) when compared with control subjects (78.6 billion),
but not when compared with asymptomatic Alzheimer’s disease
subjects (81.9 billion), or when asymptomatic Alzheimer’s disease
subjects and control groups were compared (Fig. 6A).
Further, ANOVA and post hoc analysis indicated a significant
increase of 26% in non-neuronal cell density in the demented
patients with Alzheimer’s disease (76.6 million/g) when compared
with the controls (60.86 million/g), and 18% when compared
with asymptomatic Alzheimer’s disease subjects (64.6 million/g)
(Fig. 6B and Supplementary Table 10).
Discussion
The classical hypothesis that Alzheimer’s dementia would be
caused by neuronal loss because of amyloid-bplaques and neuro-
fibrillary tangles has weakened, as these pathological deposits
were found in autopsied brains of cognitively-normal elderly.
Alternative explanations have been proposed, such as a synaptic
attack by amyloid-boligomers (Gong et al., 2003; Lacor et al.,
2004; Vieira et al., 2007; Ferreira and Klein, 2011; Sebollela et al.,
2012), and tau hyperphosphorylation (De Felice et al., 2008; Jin
et al., 2011; Flunkert et al., 2012), but the causal relation of
amyloid-bwith cell cycle re-entrance (Herrup, 2010, 2012) and
neuronal death (Neniskyte et al., 2011) has heated the debate
again. We here report a quantitative investigation of the relation
between neuronal loss and Alzheimer’s disease in different brain
regions of severely demented and asymptomatic patients with
Alzheimer’s disease.
We found a robust, significant reduction of neurons in the
hippocampus, cerebral cortex and subcortical regions (except the
cerebellum), and a concomitant increase of non-neuronal cell
number in cortical grey and subcortical white matter of demen-
ted-patients with Alzheimer’s disease as compared to asymptom-
atic-Alzheimer’s subjects. We conclude that it is dementia that
Cell numbers in Alzheimer’s brains Brain 2013: 136; 3738–3752 |3747
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

correlates with neuronal loss and glial increase, not Alzheimer’s
disease itself when defined by histopathological features. It will
be important for future work to confirm our data with larger
samples, and to relate the quantitative data and ascertain the
histopathological diagnosis with determined biochemical levels of
amyloid-b.
Neuronal loss in Alzheimer’s disease
brains correlates with dementia, not
with plaques and tangles
Because neurons in the human cortex seem not to be significantly
replaced after birth (Rakic, 1985; Bhardwaj et al., 2006), it is
conceivable that neuronal loss can lead to irreversible cognitive
changes. Nevertheless, a large wealth of data has associated cog-
nitive decline with progressive cortical mass atrophy, not neces-
sarily linked to cell death. Mouton et al. (1998) implicate synapse
degeneration, whereas Thompson et al. (2003) report neurofibril-
lary tangle accumulation. However, some studies have suggested
that distribution and severity of tangle accumulation closely fol-
lows the progress of neuronal loss (Gomez-Isla et al., 1996, 1997;
Morrison and Hof, 1997, 2007). Furthermore, others have attrib-
uted atrophy to a set of events besides neuronal loss, such as cell
shrinkage, reduced dendritic extent, and synaptic loss (DeKosky
and Scheff, 1990; Uylings and de Brabander, 2002).
We here report a significant reduction of neurons in the
demented Alzheimer’s disease group in both frontal and posterior
cortical regions when compared with asymptomatic subjects
with Alzheimer’s disease. These results indicate that cognitive
impairment correlates with lower neuronal numbers in the cerebral
cortex, but not to amyloid-band neurofibrillary tangles, as brains
from asymptomatic subjects with Alzheimer’s disease did not show
any difference compared with control subjects. It has been pro-
posed that the Alzheimer’s disease syndrome includes cognitive
impairment of higher-order functions, and that a pronounced,
widespread loss of cortical neurons takes place in these cases
(Gomez-Isla et al., 1997; Morrison and Hof, 1997, 2007;
Neniskyte et al., 2011); our quantitative data support this idea.
Similar results were obtained in the hippocampus. Classical stu-
dies have proposed that neuronal loss therein is a morphological
correlate of memory impairment in Alzheimer’s disease (Ball, 1977;
Hyman et al., 1984; Mani et al., 1986). Indeed, in demented
patients with Alzheimer’s disease, the hippocampus is severely
affected (Braak and Braak, 1991), and the consistency of hippo-
campal histopathology has led to a description of Alzheimer’s
disease as a ‘hippocampal dementia’ (Ball et al., 1985). Our results
are in agreement with these histopathological findings, and sup-
port previous stereological (Kril et al., 2004; West et al., 2004)
and neuroimaging studies (Apostolova et al., 2010; Stoub et al.,
2010).
Besides cerebral cortex and hippocampus involvement in
Alzheimer’s disease, some pathology studies have suggested that
Alzheimer’s syndrome actually begins in the brainstem and dis-
seminates throughout the brain (Yamamoto and Hirano, 1985;
Hardy et al., 1986; German et al., 1987; Grinberg et al., 2009;
Simic et al., 2009; Braak et al., 2011; Grinberg et al., 2011). These
authors argue that the current clinical criteria for Alzheimer’s
disease diagnosis are focused on cognitive deficits produced by
dysfunction of hippocampal and high order neocortical areas,
whereas other behavioural and psychological symptoms of demen-
tia such as disturbances in mood, emotion, wake-sleep cycle,
confusion and depression, may reflect brainstem involvement,
more specifically of serotonergic nuclei, in the pathogenesis of
Alzheimer’s disease (Mann and Yates, 1983; Aletrino et al.,
1992; Michelsen et al., 2008).
In addition to brainstem involvement, pathological (Whitehouse
et al., 1981; Braak and Braak, 1990) and neuroimaging studies
(Barber et al., 2002; Bruen et al., 2008; De Jong et al., 2008) have
found a correlation in the degree of cognitive decline with volume
reduction of basal ganglia and thalami in demented patients.
These changes were correlated linearly with impaired cognitive
performance, and strongly suggest that, besides cortical and hip-
pocampal atrophy, deep grey matter structures in Alzheimer’s
Figure 6 Absolute bilateral cell number (A) and density (B) for neuronal, non-neuronal, and total cells in the whole brain from control,
asymptomatic Alzheimer’s disease (ASYMAD) and demented Alzheimer’s disease (AD) groups. Each bar represents the mean and standard
deviation. Significant differences are indicated by *P50.05, **P50.01, and ***P50.001.
3748 |Brain 2013: 136; 3738–3752 C. H. Andrade-Moraes et al.
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

disease suffer atrophy as well, and that degenerative processes in
the putamen and thalamus may contribute to the cognitive decline
seen in Alzheimer’s disease.
Although we investigated the cellular composition of the brain-
stem together with basal ganglia (remaining regions), our results
may give support to the evidence described above. We found a
significant reduction of neurons in the remaining regions of the
demented patients with Alzheimer’s disease when compared with
asymptomatic patients with Alzheimer’s disease. These findings
strengthen the idea that plaques and tangles are not the main
cause for the clinical presentation of these patients, and that the
memory deficit is not the only symptom seen in Alzheimer’s dis-
ease, but one among other dysfunctions.
Although the cerebellum is traditionally considered a neural
region in charge of posture, balance, motor coordination and
learning (Marr, 1969; Thach, 1998; Wegiel et al., 1999; Boyden
et al., 2004; Morton and Bastian, 2004), recent studies implicate
it in non-motor functions, such as cognitive, behavioural and
affective processing (Ito, 1989; Leiner et al., 1991; Bower, 1997;
Timmann and Daum, 2007; Schmahmann, 2010). Nevertheless,
we failed to find any significant difference in the absolute cellular
composition and density among asymptomatic Alzheimer’s disease
and demented Alzheimer’s disease subjects compared with con-
trols, suggesting that the cerebellum is not quantitatively vulner-
able to Alzheimer’s disease.
Widespread increase of non-neuronal
cell number also correlates with
dementia, not with plaques and tangles
It has been proposed that reduction of white matter volume may
be associated with weakening of cognitive functions because of a
fall in propagation speed of electrical impulses (Marner et al.,
2003; Burns et al., 2005; Jørgensen et al., 2008). Indeed, reduc-
tion of white matter volume has been observed in demented pa-
tients with Alzheimer’s disease by use of both in vivo MRI and
post-mortem stereological techniques. Atrophy of the corpus cal-
losum has been recognized as a well-known feature of Alzheimer’s
disease (Capizzano et al., 2003; Chaim et al., 2007), and the same
for the fornix, cingulum bundle, perforant path and temporal
white matter (Hyman et al., 1986; Villain et al., 2008).
Despite these changes, few studies have elucidated the impact
of Alzheimer’s disease on glial cell numbers. We approached this
issue by quantifying the total number of non-neuronal cells in
Alzheimer’s disease. Although the isotropic fractionator does not
discern glial cells from other non-neuronal cells in the brain for
lack of specific nuclear markers, it is possible to infer the number
of glial cells using morphological criteria to count their nuclei.
We showed an increase of glial cell number in demented
patients with Alzheimer’s disease as compared with cognitively
normal and asymptomatic subjects with Alzheimer’s disease in
both white and grey matters of the cerebral cortex, more pro-
nounced in the frontal lobe than elsewhere, in contrast to another
report (Pelvig et al., 2003). These results can be correlated with
the frontal impairment seen in demented patients, and therefore
this would be because of glial cell increase, together with neuronal
reduction.
There is cellular and histological evidence in support of this
interpretation. Using post-mortem MRI and neuropathological
approaches, Polvikoski et al. (2010) found a correlation between
frontal white matter signal increase and tangles in demented patients
with Alzheimer’s disease. It could be, thus, that axonal degeneration
would follow neuronal atrophy in these cases. Other authors
(Sjo
¨beck and Englund, 2003; Sjo
¨beck et al., 2005) have reported
that astrocyte number and reactivity, and astrocyte/oligodendrocyte
ratio, are significantly greater in demented patients with Alzheimer’s
disease, whereas oligodendrocyte counts are significantly lower.
According to Sjo
¨beck and Englund (2003), astrocyte/oligodendro-
cyte ratio is positively correlated with severity of white matter dis-
ease. As these represent the majority of glial cell types in the brain
(Pelvig et al., 2008), functional, morphological (Rodrı
´guez et al.,
2009; Zhao et al., 2011), and cell number changes in Alzheimer’s
disease (Sjo
¨beck and Englund, 2003; Sjo
¨beck et al., 2005) could give
support to our interpretation.
On the other hand, although reactive astrocytes and activated
microglial cells are commonly associated with dense-core amyloid
plaques, thus suggesting that amyloid-btriggers gliosis (Itagaki
et al., 1989; Pike et al., 1995; Vehmas et al., 2003), we failed
to find any significant increase of glial cell numbers anywhere in
the cortex of asymptomatic subjects with Alzheimer’s disease.
Conclusion
Despite the relatively small sample, our findings show a robust
correlation between dementia and neuronal loss in hippocampus,
cortex and subcortical regions, and reinstate that the presence of
amyloid-bplaques and neurofibrillary tangles per se is not the
main feature responsible for the change in neuronal composition,
or for dementia of patients with Alzheimer’s disease. These results
emphasize the need to orient future research to look for a direct
link between the subtle initial disruption of cell function, resulting
in cell cycle re-entrance and apoptosis, as suggested by Herrup
(2010, 2012). In summary, our quantitative results may mean, at
least partially, that the asymptomatic subjects with histopatholo-
gical features of Alzheimer’s disease do not present with dementia
because they lack both neuronal loss and glial increase in their
brains.
Funding
This work was supported by grants provided to R.L. by the
Brazilian Council for Science and Technology Development
(CNPq), the Rio de Janeiro Foundation for the Support of
Science (FAPERJ), and the Brazilian Ministry of Science,
Technology and Innovation (Program of National Institutes of
Science and Technology, MCTI-INCTs). C.H.A.M. and A.V.O.P.
received CAPES PhD fellowships during the course of this work.
L.T.G. was funded by NIH grant R01AG040311-01 and the Sa
˜o
Paulo Foundation for the Support of Science (FAPESP). R.E.P.L.
and R.D.R. were funded by FAPESP. The Brain Bank of the
Cell numbers in Alzheimer’s brains Brain 2013: 136; 3738–3752 |3749
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

Brazilian Aging Brain Study Group was funded by LIM-22 FMUSP,
Hospital Israelita Albert Einstein, FAPESP and CAPES.
Supplementary material
Supplementary material is available at Brain online.
References
Aletrino MA, Vogels OJ, Van Domburg PH, Ten Donkelaar HJ. Cell loss
in the nucleus raphes dorsalis in Alzheimer’s disease. Neurobiol Aging
1992; 13: 461–8.
Alzheimer A. Uber einen eigenartige Erkranung der Himrinde. Allgem
Zeit Psychiatr Psychisch-Gerichtisch Med 1907; 64: 146–8.
Andersen K, Andersen BB, Pakkenberg B. Stereological quantification of
the cerebellum in patients with Alzheimer’s disease. Neurobiol Aging
2012; 33: 11–20.
Apostolova LG, Mosconi L, Thompson PM, Green AE, Hwang KS,
Ramirez A, et al. Subregional hippocampal atrophy predicts
Alzheimer’s dementia in the cognitively normal. Neurobiol Aging
2010; 31: 1077–88.
Azevedo FAC, Andrade-Moraes CH, Curado MR, Oliveira-Pinto AV,
Guimara
˜es DM, Szczupak D, et al. Automatic isotropic fractionation
for quantitative cell analysis of large brains and brain regions.
J Neurosci Meth 2013; 212: 72–8.
Azevedo FAC, Carvalho LR, Grinberg LT, Farfel JM, Ferretti R, Leite R,
et al. Equal numbers of neuronal and nonneuronal cells make the
human brain an isometrically scaled-up primate brain. J Comp
Neurol 2009; 513: 532–41.
Ball MJ. Neuronal loss, neurofibrillary tangles and granulovacuolar
degeneration in the hippocampus with ageing and dementia: a quan-
titative study. Acta Neuropathol 1977; 37: 111–8.
Ball MJ, Fisman M, Hachinski V, Blume W, Fox A, Kral VA, et al. A new
definition of Alzheimer’s disease: a hippocampal dementia. Lancet
1985; 325: 14–16.
Barber R, McKeith I, Ballard C, O’Brien J. Volumetric MRI study of the
caudate nucleus in patients with dementia with Lewy bodies,
Alzheimer’s disease, and vascular dementia. J Neurol Neurosurg
Psychiatry 2002; 72: 406–7.
Bhardwaj RD, Curtis MA, Spalding KL, Buchholz BA, Fink D, Bjo
¨rk-
Eriksson T, et al. Neocortical neurogenesis in humans is restricted to
development. Proc Natl Acad Sci USA 2006; 103: 12546–68.
Bennett DA, Schneider JA, Arvanitakis Z, Kelly JF, Aggarwal NT,
Shah RC, et al. Neuropathology of older persons without cognitive
impairment from two community-based studies. Neurology 2006;
66: 1837–44.
Bower JM. Control of sensory data acquisition. Int Rev Neurobiol 1997;
41: 489–513.
Boyden ES, Katoh A, Raymond JL. Cerebellum-dependent learning: the
role of multiple plasticity mechanisms. Annu Rev Neurosci 2004; 27:
581–609.
Braak H, Braak E. Alzheimer’s disease: striatal amyloid deposits and
neurofibrillary changes. J Neuropathol Exp Neurol 1990; 49: 215–24.
Braak H, Braak E. Neuropathological stageing of Alzheimer-related
changes. Acta Neuropathol (Berl) 1991; 82: 239–59.
Braak H, Thal DR, Ghebremedhin E, Del Tredici K. Stages of the patho-
logic process in Alzheimer disease: age categories from 1 to 100 years.
J Neuropathol Exp Neurol 2011; 70: 960–9.
Bruen PD, McGeown WJ, Shanks MF, Venneri A. Neuroanatomical
correlates of neuropsychiatric symptoms in Alzheimer’s disease. Brain
2008; 131: 2455–63.
Bundgaard MJ, Regeur L, Gundersen HJG, Pakkenberg B. Size of neo-
cortical neurons in control subjects and Alzheimer’s disease. J Anat
2001; 198: 481–9.
Burns JM, Church JA, Johnson DK, Xiong C, Marcus D, Fotenos AF, et al.
White matter lesions are prevalent but differentially related with
cognition in aging and early Alzheimer disease. Arch Neurol 2005;
62: 1870–6.
Capizzano AA, Acio
´n L, Bekinschtein T, Furman M, Gomila H,
Martı
´nez A, et al. White matter hyperintensities are significantly asso-
ciated with cortical atrophy in Alzheimer’s disease. Neurol Neurosurg
Psychiatry 2003; 75: 822–7.
Chaim TM, Duran FL, Uchida RR, Pe
´rico CA, De Castro CC, Busatto GF.
Volumetric reduction of the corpus callosum in Alzheimer’s disease
in vivo as assessed with voxel-based morphometry. Psychiatry Res
2007; 154: 59–68.
Che
´telat G, Desgranges B, Landeau B, Me
´zenge F, Poline JB, de la
Sayette V, et al. Direct voxel-based comparison between grey
matter hypometabolism and atrophy in Alzheimer’s disease. Brain
2008; 131: 60–71.
De Felice FG, Wu D, Lambert MP, Fernandez SJ, Velasco PT, Lacor PN,
et al. Alzheimer’s disease-type neuronal tau hyperphosphorylation
induced by A beta oligomers. Neurobiol Aging 2008; 9: 1334–47.
De Jong LW, Van der Hiele K, Veer IM, Houwing JJ, Westendorp RGJ,
Bollen ELEM, et al. Strongly reduced volumes of putamen and thal-
amus in Alzheimer’s disease: an MRI study. Brain 2008; 131: 3277–85.
DeKosky ST, Scheff SW. Synapse loss in frontal cortex biopsies in
Alzheimer’s disease: correlation with cognitive severity. Ann Neurol
1990; 27: 457–64.
Ferreira ST, Klein WL. The Aboligomer hypothesis for synaptic failure
and memory loss in Alzheimer’s disease. Neurobiol Learn Mem 2011;
96: 529–43.
Ferretti REL, Damin A, Brucki SMD, Morillo LS, Balbino ES, Lima MCA,
et al. Postmortem diagnosis of dementia by informant interview.
Dement Neuropsychol 2010; 4: 138–44.
Fischer O. Miliare Nekrosen mit drusigen Wucherungen der
Neurofibrillen, eine regelmassige Veranderung der Hirnrinde bei seniler
Demenz. Monatsschr Psychiatr Neurol 1907; 22: 361–72.
Flunkert S, Hierzer M, Lo
¨ffler T, Rabl R, Neddens J, Duller S, et al.
Elevated levels of soluble total and hyperphosphorylated Tau result
in early behavioral deficits and distinct changes in brain pathology in
a new Tau transgenic mouse model. Neurodegener Dis 2012[Epub
ahead of print].
German DC, White CL III, Sparkman DR. Alzheimer’s disease: neurofib-
rillary tangles in nuclei that project to the cerebral cortex. Neuroscience
1987; 21: 305–12.
Giannakopoulos P, Kovari E, Gold G, von Gunten A, Hof PR, Bouras C.
Pathological substrates of cognitive decline in Alzheimer’s disease.
Front Neurol Neurosci 2009; 24: 20–9.
Goedert M. Oskar Fischer and the study of dementia. Brain 2009; 132:
1102–1111.
Gomez-Isla T, Hollister R, West H, Mui S, Growdon JH, Petersen RC,
et al. Neuronal loss correlates with but exceeds neurofibrillary tangles
in Alzheimer’s disease. Ann Neurol 1997; 41: 17–24.
Gomez-Isla T, Price JL Jr, McKeel DW, Morris JC, Growdon JH,
Hyman BT. Profound loss of layer II entorhinal cortex neurons
occurs in very mild Alzheimer’s disease. J Neurosci 1996; 16:
4491–500.
Gong Y, Chang L, Viola KL, Lacor PN, Lambert MP, Finch CE, et al.
Alzheimer’s disease-affected brain: presence of oligomeric Abligands
(ADDLs) suggests a molecular basis for reversible memory loss. Proc
Natl Acad Sci 2003; 18: 10417–22.
Grinberg L, Ferreti RE, Farfel JM, Leite R, Pasqualucci CA, Rosemberg S,
et al. Brain bank of Brazilian aging brain study group: a milestone
reached and more than 1600 collected brains. Cell Tissue Bank
2007; 8: 151–62.
Grinberg LT, Ru¨ b U, Ferretti RE, Nitrini R, Farfel JM, Polichiso L, et al.
The dorsal raphe nucleus shows phospho-tau neurofibrillary changes
before the transentorhinal region in Alzheimer’s disease. A precocious
onset? Neuropathol Appl Neurobiol 2009; 4: 406–16.
Grinberg LT, Rueb U, Heinsen H. Brainstem: neglected locus in neuro-
degenerative diseases. Front Neurol 2011; 2: 42.
3750 |Brain 2013: 136; 3738–3752 C. H. Andrade-Moraes et al.
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

Hardy JA, Mann DM, Wester P. WinblAlzheimer’s disease B. An integra-
tive hypothesis concerning the pathogenesis and progression of
Alzheimer’s disease. Neurobiol Aging 1986; 7: 489–502.
Herculano-Houzel S, Lent R. Isotropic fractionator: a simple, rapid
method for the quantification of total cell and neuron numbers in
the brain. J Neurosci 2005; 25: 2518–21.
Herculano-Houzel S, Mota B, Lent R. Cellular scaling rules for rodent
brains. Proc Natl Acad Sci USA 2006; 103: 12138–43.
Herrup K. The involvement of cell cycle events in the pathogenesis of
Alzheimer’s disease. Alzheimers Res Ther 2010; 2: 13.
Herrup K. The contributions of unscheduled neuronal cell cycle events to
the death of neurons in Alzheimer’s disease. Front Biosci 2012; 4:
2101–9.
Hyman BT, Van Hoesen GW, Damasio AR, Barnes CL. Alzheimer’s
disease: cell-specific pathology isolates the hippocampal formation.
Science 1984; 225: 1168–70.
Hyman BT, Van Hoesen GW, Kromer LJ, Damasio AR. Perforant path-
way changes and the memory impairment of Alzheimer’s disease. Ann
Neurol 1986; 20: 472–81.
Itagaki S, McGeer PL, Akiyama H, Zhu S, Selkoe D. Relationship of
microglia and astrocytes to amyloid deposits of Alzheimer disease.
J Neuroimmunol 1989; 24: 173–82.
Ito M. Long-term depression. Annu Rev Neurosci 1989; 12: 85–102.
Jin M, Shepardson N, Yang T, Chen G, Walsh D, Selkoe DJ. Soluble
amyloid b-protein dimers isolated from Alzheimer cortex directly
induce Tau hyperphosphorylation and neuritic degeneration. Proc
Natl Acad Sci USA 2011; 108: 5819–24.
Jørgensen AMB, Marner L, Pakkenberg B. No change in total length of
white matter fibers in Alzheimer’s disease. J Neurosci 2008; 157:
878–83.
Jorm AF, Jacomb PA. The informant questionnaire on cognitive decline in
the elderly (IQCODE): socio-demographic correlates, reliability, validity
and some norms. Psychol Med 1989; 19: 1015–22.
Korbo L, Amrein I, Lipp H-P, Wolfer D, Regeur L, Oster S, et al. No
evidence for loss of hippocampal neurons in non-Alzheimer dementia
patients. Acta Neurol Scand 2004; 109: 132–39.
Kril JJ, Hodges J, Halliday G. Relationship between hippocampal volume
and CA1 neuron loss in brains of humans with and without
Alzheimer’s disease. Neurosci Lett 2004; 361: 9–12.
Kumar SS, Buckmaster PS. Neuron-specific nuclear antigen NeuN is not
detectable in gerbil substantia nigra pars reticulata. Brain Res 2007;
1142: 54–60.
Kusbeci OY, Bas O, Gocmen-Mas N, Karabekir HS, Yucel A, Ertekin T,
et al. Evaluation of cerebellar asymmetry in Alzheimer’s disease: a
stereological study. Dement Geriatr Cogn Disord 2009; 28: 1–5.
Lacor PN, Buniel MC, Chang L, Fernandez SJ, Gong Y, Viola KL, et al.
Synaptic targeting by Alzheimer’s-related amyloid b-oligomers.
J Neurosci 2004; 45: 10191–200.
Lee HG, Casadesus G, Zhu X, Takeda A, Perry G, Smith MA. Senile
plaques and amyloid-bas protective Alzheimer’s diseaseaptions to
Alzheimer’s disease. Ann NY Acad Sci 2004; 1019: 1–4.
Leiner HC, Leiner AL, Dow RS. The human cerebro-cerebellar system: its
computing, cognitive, and language skills. Behav Brain Res 1991; 44:
113–28.
Lent R, Azevedo FAC, Andrade-Moraes CH, Oliveira-Pinto AV. How
many neurons do you have? Some dogmas of quantitative neurosci-
ence under revision. Eur J Neurosci 2012; 35: 1–9.
Lesne
´S, Kotilinek L, Ashe KH. Plaque-bearing mice with reduced levels
of oligomeric amyloid-bassemblies have intact memory function.
Neurosci 2008; 151: 745–9.
Mani RB, Lohr JB, Jeste DV. Hippocampal pyramidal cells and aging in
the human: a quantitative study of neuronal loss in sectors CA1 to
CA4. Exp Neurol 1986; 94: 29–40.
Mann DM, Yates PO. Serotonin nerve cells in Alzheimer’s disease.
J Neurol Neurosurg Psychiatry 1983; 46: 96.
Marner L, Nyengaard JR, Tang Y, Pakkenberg B. Marked loss of mye-
linated nerve fibers in the human brain with age. J Comp Neurol 2003;
462: 144–52.
Marr D. A theory of cerebellar cortex. J Physiol 1969; 202: 437–70.
Mattson MP. Pathways towards and away from Alzheimer’s disease.
Nature 2004; 430: 631–9.
Michelsen KA, Prickaerts J, Steinbusch HW. The dorsal raphe nucleus and
serotonin: implications for neuroplasticity linked to major depression
and Alzheimer’s disease. Prog Brain Res 2008; 172: 233–64.
Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM, et al.
The Consortium to Establish a Registry for Alzheimer’s disease
(CERAD). Part II. Standardization of the neuropathologic assessment
of Alzheimer’s disease. Neurology 1991; 41: 479–86.
Morris JC. The CDR: current version and scoring rules. Neurology 1993;
43: 2412–3.
Morrison JH, Hof PR. Life and death of neurons in the aging brain.
Science 1997; 278: 412–19.
Morrison JH, Hof PR. Life and death of neurons in the aging cerebral
cortex. Int Rev Neurobiol 2007; 81: 41–57.
Morton SM, Bastian AJ. Prism adaptation during walking generalizes to
reaching and requires the cerebellum. J Neurophysiol 2004; 92:
2497–509.
Mouton PR, Martin LJ, Calhoun ME, Dal Forno G, Price DL. Cognitive
decline strongly correlates with cortical atrophy in Alzheimer’s demen-
tia. Neurobiol Aging 1998; 5: 371–7.
Mullen RJ, Buck CR, Smith AM. NeuN, a neuronal specific nuclear
protein in vertebrates. Development 1992; 116: 201–11.
Negash S, Bennett DA, Wilson RS, Schneider JA, Arnold SE. Cognition
and neuropathology in aging: multidimensional perspectives from the
Rush Religious Orders Study and Rush Memory and Aging Project.
Curr Alzheimer Res 2011; 4: 336–40.
Neniskyte U, Neher JJ, Brown GC. Neuronal death induced by nanomolar
amyloid-bis mediated by primary phagocytosis of neurons by micro-
glia. J Biol Chem 2011; 286: 39904–13.
Nestor PJ, Scheltens P, Hodges JR. Advances in the early detection of
Alzheimer’s disease. Nat Med 2004; 10 (Suppl): S34–41.
Palop JJ, Mucke L. Amyloid-b–induced neuronal dysfunction in
Alzheimer’s disease: from synapses toward neural networks. Nat Rev
Neurosci 2010; 7: 812–18.
Pelvig DP, Pakkenberg H, Regeurb L, Osterc S, Pakkenberg B.
Neocortical glial cell numbers in Alzheimer’s disease: a stereological
study. Dement Geriatr Cogn Disord 2003; 16: 212–19.
Pelvig DP, Pakkenberg H, Stark AK, Pakkenberg B. Neocortical glial cell
numbers in human brains. Neurobiol Aging 2008; 29: 1754–62.
Pike CJ, Cummings BJ, Cotman CW. Early association of reactive astro-
cytes with senile plaques in Alzheimer’s disease. Exp Neurol 1995; 132:
172–9.
Polvikoski TM, van Straaten ECW, Barkhof F, Sulkava R, Aronen HJ,
Niinisto L, et al. Frontal lobe white matter hyperintensities and
neurofibrillary pathology in the oldest old. Neurology 2010; 75:
2071–8.
Price JL, Morris JC. Tangles and plaques in nondemented aging and
“preclinical” Alzheimer’s disease. Ann Neurol 1999; 45: 358–68.
Rakic P. Limits of neurogenesis in primates. Science 1985; 227: 1054–6.
Rodrı
´guez JJ, Olabarria M, Chvatal A, Verkhratsky A. Astroglia in de-
mentia and Alzheimer’s disease. Cell Death Differ 2009; 16: 378–85.
Sarnat HB, Nochlin D, Born DE. Neuronal nuclear antigen (NeuN): a
marker of neuronal maturation in the early human fetal nervous
system. Brain Dev 1998; 20: 88–94.
Schmahmann JD. The role of the cerebellum in corgnition and emotion:
personal reflections since 1982 on the dysmetria of thought hypoth-
esis, and its historical evolution from theory to therapy. Neuropsychol
Rev 2010; 20: 236–60.
Sebollela A, Freitas-Correa L, Oliveira FF, Paula-Lima AC, Saraiva LM,
Martins SM, et al. Amyloid-boligomers induce differential gene
expression in Alzheimer’s diseaseult human brain slices. J Biol Chem
2012; 287: 5021–32.
Simic G, Kostovic I, Winblad B, Bogdanovic N. Volume and number of
neurons of the human hippocampal formation in normal aging and
Alzheimer’s disease. J Comp Neurol 1997; 379: 482–94.
Cell numbers in Alzheimer’s brains Brain 2013: 136; 3738–3752 |3751
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from

Simic G, Stanic
´G, Mladinov M, Jovanov-Milosevic N, Kostovic I, Hof PR.
Annotation - Does Alzheimer’s disease begin in the brainstem?
Neuropathol Appl Neurobiol 2009; 35: 532–54.
Sjo
¨beck M, Englund E. Glial levels determine severity of white matter
disease in Alzheimer’s disease: a neuropathological study of glial
changes. Neuropathol Appl Neurobiol 2003; 29: 159–69.
Sjo
¨beck M, Haglund M, Englund E. Decreasing myelin density reflected
increasing white matter pathology in Alzheimer’s disease: a neuro-
pathological study. Int J Geriatr Psychiatry 2005; 20: 919–26.
Stoub TR, Rogalski EJ, Leurgans S, Bennett DA, de Toledo-Morrell L.
Rate of entorhinal and hippocampal atrophy in incipient and
mild AD: relation to memory function. Neurobiol Aging 2010; 31:
1089–98.
Thach WT. A role for the cerebellum in learning movement coordination.
Neurobiol Learn Mem 1998; 70: 177–88.
Thomann PA, Schla
¨fera C, Seidl U, Santos VD, Essig M, Schro
¨der J. The
cerebellum in mild cognitive impairment and Alzheimer’s disease—a
structural MRI study. J Psychiatr Res 2008; 14: 1198–202.
Thompson PM, Hayashi KM, de Zubicaray G, Janke AL, Rose SE,
Semple J, et al. Dynamics of gray matter loss in Alzheimer’s disease.
J Neurosci 2003; 23: 994–1005.
Timmann D, Daum I. Cerebellar contributions to cognitive functions: a
progress report after two decades of research. Cerebellum 2007; 3:
159–62.
Uylings HB, de Brabander JM. Neuronal changes in normal human aging
and Alzheimer’s disease. Brain Cogn 2002; 49: 268–76.
Vehmas AK, Kawas CH, Stewart WF, Troncoso JC. Immune reactive cells
in senile plaques and cognitive decline in Alzheimer’s disease.
Neurobiol Aging 2003; 2: 321–31.
Vieira MNN, Forny-Germano L, Saraiva LM, Sebollela A, Martinez AMB,
Houzel JC, et al. Soluble oligomers from a non-disease related protein
mimic Ab-induced tau hyperphosphorylation and neurodegeneration.
J Neurochem 2007; 103: 736–48.
Villain N, Desgranges B, ViAlzheimer’s diseaseer F, de la Sayette V,
Me
´zenge F, Landeau B, et al. Relationships between hippocampal
atrophy, white matter disruption, and gray matter hypometabolism
in Alzheimer’s disease. J Neurosci 2008; 28: 6174–81.
Wegiel J, Wisniewski HM, Dziewiatkowski J, Badmajew E, Tarnawski M,
Reisberg B, et al. Cerebellar atrophy in Alzheimer’s disease—clinico-
pathological correlations. Brain Res 1999; 818: 41–50.
West MJ, Kawas CH, Stewart WF, Rudow GL, Troncoso JC.
Hippocampal neurons in pre-clinical Alzheimer’s disease. Neurobiol
Aging 2004; 25: 1205–12.
Whitehouse PJ, Price DL, Clark AW, Coyle JT, DeLong MR. Alzheimer’s
disease: evidence for selective loss of cholinergic neurons in the
nucleus basalis. Ann Neurol 1981; 10: 122–6.
Whitehouse PJ, Price DL, Struble RG, Clark AW, Coyle JT, Delon MR.
Alzheimer’s disease and senile dementia: loss of neurons in the basal
forebrain. Science 1982; 215: 1237–9.
Wolf HK, Buslei R, Schmidt-Kastner R, Shcmidt-Kastner PK, Pietsch T,
Wiestler OD, et al. NeuN: a useful neuronal marker for diagnostic
histopathology. J Histochem Cytochem 1996; 44: 1167–71.
Yamamoto T, Hirano A. Nucleus raphe dorsalis in Alzheimer’s disease:
neurofibrillary tangles and loss of large neurons. Ann Neurol 1985; 17:
3–7.
Zhao J, O’Connor T, Vassar R. The contribution of activated astrocytes
to Abproduction: implications for Alzheimer’s disease pathogenesis.
J Neuroinflammation 2011; 8: 150.
3752 |Brain 2013: 136; 3738–3752 C. H. Andrade-Moraes et al.
at Universidade Federal do Rio de Janeiro on December 17, 2013http://brain.oxfordjournals.org/Downloaded from