Group Title: Journal of Neuroinflammation 2007, 4:20
Title: Microglial activation in the hippocampus of hypercholesterolemic rabbits occurs independent of increased amyloid production
Full Citation
Permanent Link:
 Material Information
Title: Microglial activation in the hippocampus of hypercholesterolemic rabbits occurs independent of increased amyloid production
Series Title: Journal of Neuroinflammation 2007, 4:20
Physical Description: Archival
Creator: Xue QS
Sparks DL
Streit WJ
Publication Date: 39318
 Record Information
Bibliographic ID: UF00100241
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: Open Access:


This item has the following downloads:

microglial ( PDF )

Full Text

Journal of Neuroinflammation Bioed Ce


Microglial activation in the hippocampus of hypercholesterolemic
rabbits occurs independent of increased amyloid production
Qing-Shan Xue', D Larry Sparks2 and Wolfgang J Streit* I

Address: 'Department of Neuroscience, University of Florida College of Medicine and McKnight Brain Institute, 100 Newell Drive, Gainesville FL
32611, USA and 2Roberts Laboratory for Neurodegenerative Disease Research, Sun Health Research Institute, Sun City, AZ, USA
Email: Qing-Shan Xue; D Larry Sparks; Wolfgang J Streit*
* Corresponding author

Published: 24 August 2007 Received: 5 June 2007
journal of Neuroinflammation 2007, 4:20 doi:10.1 186/1742-2094-4-20 Accepted: 24 August 2007
This article is available from:
2007 Xue et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Rabbits maintained on high-cholesterol diets are known to show increased
immunoreactivity for amyloid beta protein in cortex and hippocampus, an effect that is amplified
by presence of copper in the drinking water. Hypercholesterolemic rabbits also develop sporadic
neuroinflammatory changes. The purpose of this study was to survey microglial activation in rabbits
fed cholesterol in the presence or absence of copper or other metal ions, such as zinc and
Methods: Vibratome sections of the rabbit hippocampus and overlying cerebral cortex were
examined for microglial activation using histochemistry with isolectin B4 from Griffonia simplicifolia.
Animals were scored as showing either focal or diffuse microglial activation with or without
presence of rod cells.
Results: Approximately one quarter of all rabbits fed high-cholesterol diets showed evidence of
microglial activation, which was always present in the hippocampus and not in the cortex. Microglial
activation was not correlated spatially with increased amyloid immunoreactivity or with
neurodegenerative changes and was most pronounced in hypercholesterolemic animals whose
drinking water had been supplemented with either copper or zinc. Controls maintained on normal
chow were largely devoid of neuroinflammatory changes, but revealed minimal microglial activation
in one case.
Conclusion: Because the increase in intraneuronal amyloid immunoreactivity that results from
administration of cholesterol occurs in both cerebral cortex and hippocampus, we deduce that the
microglial activation reported here, which is limited to the hippocampus, occurs independent of
amyloid accumulation. Furthermore, since neuroinflammation occurred in the absence of
detectable neurodegenerative changes, and was also not accompanied by increased astrogliosis, we
conclude that microglial activation occurs because of metabolic or biochemical derangements that
are influenced by dietary factors.

Page 1 of 10
(page number not for citation purposes)

Open Access

Journal of Neuroinflammation 2007, 4:20

A number of neuropathological changes similar to those
characteristically associated with Alzheimer's disease have
been reported in hypercholesterolemic rabbits, and thus
the cholesterol-fed rabbit offers a pertinent animal model
for investigating some of the mechanisms that underlie
disease pathogenesis [1,2]. Perhaps most relevant is the
fact that addition of cholesterol to the diet consistently
results in increased immunoreactivity for amyloid beta
protein within neurons of the cerebral and hippocampal
cortices of these animals [3,4]. Inflammatory changes,
such as microglial activation and leukocyte extravasation,
have also been reported in cholesterol-fed rabbits, but
unlike the enhanced accumulation of amyloid neuroin-
flammatory changes are not found uniformly in all hyper-
cholesterolemic animals [5]. When neuroinflammation
does occur it tends to be limited affecting relatively small
areas rather than an entire region. In the past, we have
assumed that the inciting stimulus for neuroinflamma-
tion is provided by the increase in amyloid beta protein
that results from high serum cholesterol levels. This
assumption seemed reasonable in light of large numbers
of studies reporting proinflammatory effects of amyloid
beta peptides over many years [6-14].

The finding that addition of small amounts of copper to
the drinking water of cholesterol-fed rabbits amplifies the
accumulation ofintraneuronal amyloid in cortex and hip-
pocampus and leads to cognitive dysfunction [15] has
prompted us to reexamine brains from animals treated in
this fashion for neuroinflammatory changes. Our expecta-
tion was that concomitant with the enhanced accumula-
tion of amyloid there would be increased
neuroinflammation. At the same time, since zinc-supple-
mented drinking water does not have a significant effect
on amyloid accumulation [16], we expected to see no
change in neuroinflammation in rabbits receiving zinc.
However, contrary to this hypothesis our current findings
now show that animals from both copper and zinc-sup-
plemented groups show similar levels ofmicroglial activa-
tion. In addition, microglial activation in all animals
maintained on cholesterol diets, regardless of metals
added, was confined to the hippocampal region. This
leads us to think that microglial activation in the choles-
terol-fed rabbit is unrelated to intraneuronal amyloid
accumulation, but is triggered instead by metabolic or
biochemical abnormalities in the hippocampus caused by
elevated serum cholesterol levels.

New Zealand white rabbits
Adolescent male New Zealand white rabbits (3000-4000
g) were housed in the rabbit facility at SHRI operating
under the guidelines of the USDA with a 12:12 light cycle,
at 67 7'F, and 45-50% humidity. Animals were ran-

domly assigned to one of seven groups as a subset of a
larger IACUC approved experimental protocol. Some ani-
mals received normal chow and allowed either distilled
water or distilled water with 0.12 PPM copper added (n =
8) ad libitum. Other animals were administered 2% cho-
lesterol diet and allowed tap water (n = 4) or distilled
water (n = 4), or distilled water with 0.12 PPM copper ion
(as sulfate, n = 4), 0.36 PPM zinc (as sulfate, n = 5) or 0.36
PPM aluminum (as sulfate, n = 5) ad libitum. Control and
cholesterol diets were commercially obtained from Purina
Mills, Inc. (Laboratory Rabbit Diet with and without 2%
cholesterol) and were administered for 10 weeks. Dietary
food intake was limited to one cup per day (8 oz) and ad
libitum water consumption varied between 32 and 40 oz/
day. The animal protocol (# 0403) was approved by the
Sun Health Research Institute Institutional Animal Care
and Use Committee.

Water analysis
Water was analyzed by US Filters (Vivendi Environment),
an EPA Certified Water Quality Testing Laboratory for lev-
els of Arsenic (EPA 200.9), Mercury (EPA 245.1), and
organic (total organic carbo-TOC; SM5310C) as special
studies, and for a 'Standard A' assessment (EPA 200.7,
EPA 300.0) to include levels of aluminum, calcium, mag-
nesium, sodium, potassium, barium, strontium, iron,
copper, manganese, zinc, chloride, sulfate, nitrate, fluo-
ride, and silica.

Tissue processing
Animals in each group were sacrificed ten weeks after ini-
tiating the experimental dietary (food and water) proto-
col. On the day of sacrifice, animals were administered a
cocktail of Ketamine and Xylazine (IM; 45-75 mg/kg and
5-10 mg/kg respectively). Anesthetized animals were
secured to a stainless steel surgical apparatus, the heart
was exposed and a butterfly needle was inserted in the left
apex, and blood was collected in purple top (EDTA) vacu-
tainer tubes for chemical analysis. Thereafter, a needle
attached to the perfusion apparatus was inserted and
secured in the left apex of the heart, the vena cava was
incised and perfusion was initiated. Animals were per-
fused under pressure with 120 ml of 4% paraformalde-
hyde at a constant rate of 5 ml/min using a constant
pressure pump. A full necropsy was performed on each
animal. Fifty-micron vibratome sections of hippocampus
and hippocampal cortex of the brain were prepared for
subsequent staining.

Lectin histochemistry
Microglial cells were visualized in brain sections using lec-
tin binding, as described [17,18]. Following a rinse in
PBS, sections were incubated in lectin GSA I-B4-HRP
(Sigma Chemical Co., L5391), diluted to 5 lg/ml in 0.1%
Triton/PBS overnight at 4 C. After washing with PBS, lec-

Page 2 of 10
(page number not for citation purposes)

Journal of Neuroinflammation 2007, 4:20

tin binding sites were visualized with 3,3'-diabimobenzi-
dine (DAB)-H202 substrate. All sections were dehydrated
through ascending alcohols, cleared in xylenes and cover-
slipped with Permount. Selected sections were counter-
stained with 0.5% cresyl violet.

Double fluorescent labeling of microglia and astrocytes
In order to determine if microglial activation was accom-
panied by astrogliosis, double-labeling for both glial cell
types was performed. Sections were rinsed in PBS, fol-
lowed by blockage of non-specific binding of antibodies
in 10% normal goat serum in PBS for 1 hr at 370C. Sec-
tions were then incubated in a mixed solution of rabbit
polyclonal anti-glial fibrillary acidic protein (GFAP,
DakoCytomation, Denmark A/S, diluted at 1:200) and
biotinylated isolectin B4 (5 pg/ml, Sigma, L2140) in 5%
goat serum with 0.1% Triton X-100 in PBS at 4 C for over-
night. After three washes with PBS, sections were incu-
bated in a mixed solution of highly cross-adsorbed goat
anti-rabbit IgG conjugated with Alexa fluor 488 (Molecu-
lar Probes, A11034, diluted at 1:300) and avidin conju-
gated with Alexa fluor 594 (Invitrogen, S32356, diluted at
1:500) in 5% goat serum with 0.1% Triton X-100 in PBS
for 1 hr at room temperature. Following three washes, sec-
tions were mounted onto glass slides and coverslipped
with GEL/MOUNT (Biomeda corp., Foster City, CA).

Immunolabeling for ubiquitin
In order to detect ubiquitinated neurons or neurites indic-
ative ofneurodegeneration, single staining was performed
using a monoclonal antibody against ubiquitin (hybrid-
oma supernatant provided by Dr. Gerry Shaw [191). Bind-
ing sites were visualized using biotinylated goat anti-
mouse IgG antibodies (Vector Laboratories, Cat. No. BA-
9200), amplified by avidin conjugated with HRP, and
DAB-H202 substratum. Ubiquitin immunolabeling was
also performed with goat anti-mouse IgG conjugated with
Alexa fluor 488 (Molecular Probes, A11029, diluted at

Observation and imaging
Slides were examined with a Zeiss Axioskop 2 microscope.
Digital images were captured with a Spot RT3 digital cam-
era (Diagnostic Instruments Inc.; Sterling Heights, MI).
For double fluorescence labeling, images were originally
captured in black and white. Images were pseudo-colored
and/or digitally merged from images captured at single
fluorochrome using Adobe Photoshop software (Adobe
Systems Inc.; San Jose, CA).

Similar to our prior observations regarding microgliosis in
hypercholesterolemic rabbits [5], the current results
revealed considerable variability in neuroinflammation
among animals in any given group of animals fed a cho-
lesterol-containing diet. Most animals on cholesterol diets
did not show any signs of microglial activation while
some showed focal and/or diffuse patterns of activation,
as detailed below. Controls maintained on regular chow
and dHO2 with or without copper ion added showed no
evidence of microglial activation in 7 out of 8 animals
(Table 1, Figs. 1A,C,D). However, in one case we were able
to observe small foci of enhanced staining intensity in the
dentate gyrus indicative of low level activation (Fig. 1B).
For purposes of scoring and comparing the intensity of
microglial activation in individual animals, we designated
the pattern observed in this control animal as "minimal"

More pronounced microglial activation was observed in 5
out of the 22 animals that had been fed a cholesterol diet
(Table 1). In these animals, microglial activation was evi-
dent by the presence of multiple foci of intensified lectin
staining (e.g. Figs. 2A,B) and/or by a more diffuse pres-
ence of activated microglial cells throughout the dentate
gyrus and the stratum lacunosum molecular (Fig. 2C).
Round spots of microglial activation measuring about
300-400 pm in diameter were most often seen in the
hilus of the dentate gyrus (Figs. 2A,B, 3C,D). Their

Table I: Qualitative assessment of microglial activation in hippocampi of rabbits fed different diets.

Animal\Group Reg Chow/

Reg Chow/Cu Chol/dH20



* Sections showing patches unstained with lectin.

Page 3 of 10
(page number not for citation purposes)












9 2s
A e, j.... 4 -


I V. .' I

J 1c' ': V.


Reg Chow/dHO..

A" A 'St


.. Reg Chow/dH2O

.4 *t' q

k. 4 .
U' .i
I) ..'

Figure I
Lectin staining for microglia in the hippocampus (A-C) and cerebral cortex (D) of control rabbits receiving regular chow. A,
C, D, microglia are distributed evenly throughout the parenchyma as resting cells showing slightly greater density in the sub-
granular zone of the dentate gyrus (triangle in C) and the stratum lacunosum molecular (star in C). One of the control ani-
mals shows small foci of minimal microglial activation (arrows in B). Scale bar: 1,000 gtm

increased staining intensity was due to the accumulation
of activated microglia displaying cell hypertrophy (Figs
3B,C,G). In one instance, very small foci of microglial acti-
vation could be observed in the stratum pyramidale (Fig.
2B). Animals that showed microglial activation also dis-
played conspicuous microglial rod cells, which were
prominent in the stratum radiatum (Fig. 3F).

It is interesting to note that all animals, regardless of their
diets, showed enhanced microglial staining in the sub-
granular zone clearly delineating the fascia dentata (Fig.
1A-C). This enhanced staining appeared to be due to a
greater density of microglial cells in the subgranular layer,
rather than to microglial activation as there was no evi-
dence of cell hypertrophy. The stratum lacunosum molec-

ulare was also well delineated by microglial staining in
that cellular density there appeared to be slightly greater
than elsewhere in the hippocampus (e.g. Figs. 1A-C).
None of the animals demonstrated any evidence ofmicro-
glial activation in the cerebral cortex overlying the hippoc-
ampal formation (Figs. 1D, 2F). Sections of the cerebral
cortex revealed an even distribution of ramified microglia

When foci of microglial activation were examined at
higher power, it was evident that the cells present in these
areas were hypertrophied and their ramified processes
retracted (Figs. 3B-D,G). Typically, these activated micro-
glia were seen as round, focal formations within the hilus
(Fig. 3C), but could also be seen to be distributed in a

Page 4 of 10
(page number not for citation purposes)

Journal of Neuroinfiammation 2007, 4:20

lilt' '

I:-' -. .

*, C. *

s '''4- '

4 " rl~
Choit: '
* . q
* J*.-- .

-*.r ..o- ACholi/ Al. ."r' Chol/Zn

..:. 9 /1' -.
L. . o : :',-v ..
,,, 1. .. .. ..
,,,-;..I,. :- **', . .., ; ... .*.
,-. ., ., ." ....

-., -""" J""S..... '' '" "-

r *A I*

Lectin staining for microglia in the hippocampus (A-E) and cerebral cortex (F) of rabbits receiving cholesterol diets. The figure
shows focal microglial activation evident as hyperintense spots (arrows in A, B, D), and a more diffuse pattern covering most
the dentate hilus (asterisks in C). The arrowhead in panel B points to small spot of activated microglia in the CAI pyramidal
layer. Animals receiving drinking water supplemented with aluminum did not show significant microglial activation (E). None of

the cholesterol-fed animals showed microglial activation in the cerebral cortex (F). Scale bar: 1,000 gtm

Page 5 of 10
(page number not for citation purposes)

Journal of Neuroinflammation 2007, 4:20

Figure 3
Comparison of resting and activated microglia in control and cholesterol-fed rabbits. A-D, dentate gyrus showing granule cell
layer (near top) and hilus. Normal distribution of resting microglia in control rabbit (A) stands in contrast to activated micro-
glia in hypercholesterolemic rabbits (B-D). Panel B shows diffuse distribution of activated microglia; panel C shows focal accu-
mulations of activated microglia; panel D shows a combination of diffuse and focal patterns. E-G, high power views of various
microglial morphologies, including resting cells (E), rod cells in stratum radiatum (F), and activated cells (G). Cresyl violet
counterstain. Scale bars: 200 im (A-D); 50 im (E-G)

Page 6 of 10
(page number not for citation purposes)

Journal of Neuroinflammation 2007, 4:20

cc~i~ ~,~dLE

Journal of Neuroinflammation 2007, 4:20

more widespread and diffuse fashion throughout the hilar
gray and white matter strata (Figs. 3B,D). Microglial rod
cells were frequently encountered in the stratum radiatum
of those animals showing focal and diffuse activation pat-
terns (Fig. 3F). Counterstaining with cresyl violet clearly
revealed the neuronal layers of the hippocampal forma-
tion, but failed to show any evidence of neuronal damage
or loss. In order to detect neurodegenerative changes, we
also stained sections from cholesterol-fed animals that
showed microglial activation for ubiquitin, but these
studies failed to reveal any specific staining of neurons or
their processes. To further analyze areas showing micro-
glial activation we performed double fluorescent staining
for both microglia and astrocytes, using a combination of
lectin staining and GFAP immunostaining (Fig. 4).
Included in these experiments were all five animals that
had shown microglial activation in the hippocampus.
Examination of double-stained preparations revealed that
foci of microglial activation did not show concomitant
increases in GFAP immunoreactivity (Figs. 4D-F). The
intensity and distribution of GFAP immunoreactivity in
these foci was no different from that observed elsewhere
in the section or as seen in control animals (Figs. 4A-C),
leading us to conclude that neuroinflammatory foci
revealed by microglial activation were not subject to reac-
tive astrogliosis.

A final observation that was evident in some of the choles-
terol-fed animals, but not in controls, concerns the pres-
ence of microglia-free patches in lectin-stained sections
(Table 1; Fig. 5). These patches are areas in any given sec-
tion that are devoid of lectin staining, suggesting a local-
ized loss of microglial cells. Shown in Fig. 5A is the most
dramatic example of patchiness we were able to observe,
and clearly the density of microglia in this particular sec-
tion is much lower than what was normally seen in hip-
pocampal sections (compare to Fig. 1). Examination of
cell-free patches at high power did not reveal any signs of
microglial cell death, and microglia in the vicinity of
unstained patches were perfectly ramified and appeared
normal and non-activated (Fig. 5B). No abnormalities
could be detected in patchy areas using either cresyl violet
staining or ubiquitin immunohistochemistry, and thus
we attribute the spotty lectin staining to a tissue process-
ing artifact, possibly related to fixation, rather than to a
loss of microglial cells.

The current study serves to extend prior work from our
and other laboratories regarding the sporadic neuroin-
flammation that occurs in hypercholesterolemic rabbits
[5,20]. While on one hand confirming the intermittent
nature of microglial activation and showing that it

Figure 4
Double fluoresecent labeling for microglia with isolectin B4 and for astrocytes with anti-GFAP in rabbit hippocampus. A-C,
uniform distributions of both glial cell types are evident in a control animal. D-F, focus of microglial activation in a cholesterol-
fed rabbit shows normal staining pattern for astrocytes. Scale bar: 100 gIm

Page 7 of 10
(page number not for citation purposes)

Reg Chow/dH 2 0 1B 4


Journal of Neuroinflammation 2007, 4:20

1'a 'Y t:
A~~;. - _

Figure 5
Patchy lectin staining of microglia throughout the hippocam-
pus is shown at low power (A), and at high power with cre-
syl violet counterstaining (B). No evidence for microglial cell
loss or other degenerative changes was detectable. Patchy
staining is most likely an artifact of fixation. Scale bars: 1,000
gtm (A); 100 gtm (B).

affected only 23% of all cholesterol-fed rabbits, compared
to 30% in the study by Zatta et al. [20], the current find-
ings also call attention to a previously unsuspected dis-
connection between increased amyloid production in
hypercholesterolemic rabbits and microglial activation.
There are at least four observations derived from the cur-
rent study supporting the notion that increased amyloid
production in hypercholesterolemic is not a direct stimu-
lus for microglial activation. First is the mismatch
between regions affected by increased amyloid immuno-
reactivity and neuroinflammation. Prior work has shown
that the increase in amyloid induced by cholesterol occurs
prominently in neuronal layers II, IV-V of the cerebral cor-
tex, as well as in those of the hippocampus, including
pyramidal and granule cell layers [3,4,15]. In contrast, the

~I "

Page 8 of 10
(page number not for citation purposes)




neuroinflammatory changes reported here are limited to
the hippocampus with the dentate gyrus being affected to
the greatest extent. Second, cholesterol-induced increases
in intraneuronal amyloid immunoreactivity occur con-
sistently in all hypercholesterolemic rabbits, while micro-
glial activation occurs only in a relatively small fraction of
these animals. Third, prior work has shown that supple-
mentation of the drinking water with copper amplifies
intraneuronal amyloid immunoreactivity, while addition
of zinc does not [16]. This influence of metal ions over
Alzheimer-like pathology is not mirrored by concomitant
changes in neuroinflammation, as shown by our current
results. In fact, it appears that neuroinflammation is most
pronounced in animals that consumed zinc-supple-
mented drinking water. Fourth, we observed some, albeit
minimal microglial activation in a control animal con-
suming standard rabbit chow and dHO2. Since control
animals do not show intraneuronal amyloid immunore-
activity, the observed glial activation could not possibly
represent a direct microglial response to amyloid.
Although unexpected, this latter observation in a control
animal serves to make an important point, namely, that
studying microglial activation and distribution patterns is
a very sensitive method for detecting subtle and localized
changes in brain homeostasis that may not be detectable
by other assays. Thus, microglia are indeed keen sensors of
brain pathology [21]. This point is underscored further by
our inability to uncover any evidence for neurodegenera-
tive changes or neuronal loss in the five hypercholestero-
lemic rabbits that showed pronounced microglial
activation. Neither counterstaining with cresyl violet nor
ubiquitin immunolabeling revealed any neuronal abnor-
malities. In addition, there was a striking absence of reac-
tive astrocytes in those focal areas demonstrating
microglial activation, which leads us to think that the dis-
turbance that triggered microglial activation was not suffi-
ciently severe to cause serious neuronal damage and
subsequent astroglial scarring.

There are numerous reports, most of them in vitro, describ-
ing how amyloid peptides stimulate detrimental micro-
glial activation (e.g. [12,14,22,23]). These in vitro studies
have been critical for supporting the notion that presence
of amyloid plaques in AD brain leads to a chronic neu-
roinflammatory response, which many believe plays a
central role in the development of Alzheimer's disease
[6,8,10,11,24-27]. Our current findings in cholesterol-fed
rabbits do not offer additional support for the idea that
amyloid directly triggers neuroinflammation, as already
explained. However, it is important to point out that most
of the amyloid accumulation in hypercholesterolemic
rabbits is intraneuronal and that deposition of amyloid in
the extracellular space occurs only rarely [5,15,20]. This,
of course, could mean that most of the intracellular amy-
loid never reaches microglial cells surveying the extracel-

Journal of Neuroinflammation 2007, 4:20

lular milieu. As far as the ability of copper in the drinking
water (but not zinc or aluminum) to amplify cholesterol-
induced amyloid accumulation, we hypothesize that this
is due to copper's unique ability to inhibit amyloid clear-
ance from brain [28].

So what might be the nature of an underlying perturba-
tion that triggers the sporadic and localized neuroinflam-
matory reactions observed? One possibility that comes to
mind is vascular inflammation and an associated breach
in the blood brain barrier (BBB). Previous studies in the
hypercholesterolemic rabbits have shown leakage of
Evans Blue dye into the brain parenchyma, as well as
increased vascular immunoreactivity with MECA-32 [1],
an antibody which recognizes an endothelial cell epitope
that is downregulated as the BBB matures during develop-
ment [29]. Reexpression of the MECA-32 antigen has been
found to occur during experimentally induced neuroin-
flammation [30]. Thus, high levels of serum cholesterol in
rabbits may induce vascular changes similar to early
inflammatory lesions of atherosclerosis, and this vascular
inflammation may trigger microglial activation. However,
in other animal models, an induction of peripheral
inflammation and increased BBB permeability associated
with extravasation of serum proteins has been shown to
occur without reactive microgliosis or astrogliosis [31].
Thus, further studies focused specifically on examining
the relationship of vascular inflammation and microglial
activation in hypercholesterolemic rabbits seem to be

A final consideration pertains to the basic understanding
of the functional significance of microglial activation and
neuroinflammation, i.e. whether it is beneficial or harm-
ful. Given the great abundance ofmicroglial cells through-
out the CNS, as shown in the micrographs presented here,
it is difficult to see an evolutionary advantage in having
this many potentially dangerous immune effector cells
populate an organ that is relatively incapable of regenera-
tion. In our view, the only way to reconcile microglial
abundance with an evolutionary advantage is to accept
that these cells are constitutively neuroprotective, and that
the spatially restricted microglial activation observed here
is a reflection of an ongoing rescue effort [32,33]. In other
words, microglia get activated when neurons get dam-
aged, rather than the other way around. Thus, we believe
that the current findings demonstrating focal microglial
activation in the hippocampus are a reflection of focal
neuronal damage, which is likely to be minor since it is
not demonstrable with routine histological stains, with
specific markers of neurodegeneration, or with markers of
astrogliosis. The increased and sporadic occurrence of
microglial activation in rabbits on cholesterol diets sug-
gests that dietary factors can directly affect the hippocam-

The current histopathological analysis underscores the
extreme sensitivity of microglial reactions they are truly
biological sensors of neuropathology. The sporadic and
focal nature of the microglial activation observed in
hypercholesterolemic rabbits suggests that any damage
inflicted on hippocampal neurons is very slight, and
potentially reversible. We suspect that high-cholesterol
diets, which are very atypical for rabbits and rodents in
general, are sufficiently adverse to upset the metabolism
of some neurons in some animals to trigger a microglial
response. By analogy, it now seems reasonable to think
that dietary factors in humans may subtly influence brain
homeostasis, and that diet-induced disturbances are
demonstrable through analysis of microglia during post-
mortem examination.

Competing interests
The authors) declare that they have no competing inter-

Authors' contributions
QX carried out the histopathological studies and drafted
the manuscript. DS initiated this collaborative study. WS
participated in analysis of histopathological findings and
design of figures. All authors completed the final version
of the manuscript.

Supported by NIH grant AG023665 (WJS), and by The Arizona Biomedical
Research Commission (DLS).

I. Sparks DL, Kuo YM, Roher A, Martin T, Lukas RJ: Alterations of
Alzheimer's disease in the cholesterol-fed rabbit, including
vascular inflammation. Preliminary observations. Ann N Y
Acad Sci 2000, 903:335-344.
2. Sparks DL, Liu H, Gross DR, Scheff SW: Increased density of cor-
tical apolipoprotein E immunoreactive neurons in rabbit
brain after dietary administration of cholesterol. Neurosci Lett
1995, 187(2): 142-144.
3. Sparks DL, Scheff SW, HunsakerJC 3rd, Liu H, Landers T, Gross DR:
Induction of Alzheimer-like beta-amyloid immunoreactivity
in the brains of rabbits with dietary cholesterol. Exp Neurol
1994, 126(1):88-94.
4. Sparks DL: Intraneuronal beta-amyloid immunoreactivity in
the CNS. NeurobiolAging 1996, 17(2):291-299.
5. Streit WJ, Sparks DL: Activation of microglia in the brains of
humans with heart disease and hypercholesterolemic rab-
bits. j Mol Med 1997, 75(2): 30-138.
6. Rogers J, Strohmeyer R, Kovelowski CJ, Li R: Microglia and inflam-
matory mechanisms in the clearance of amyloid beta pep-
tide. Glia 2002, 40(2):260-269.
7. Meda L, Bernasconi S, Bonaiuto C, Sozzani S, Zhou D, Otvos L Jr.,
Mantovani A, Rossi F, Cassatella MA: Beta-amyloid (25-35) pep-
tide and IFN-gamma synergistically induce the production of
the chemotactic cytokine MCP-I/JE in monocytes and
microglial cells. j Immunol 1996, 157(3): 1213-1218.
8. McGeer PL, McGeer EG: Inflammation, autotoxicity and Alzhe-
imer disease. Neurobiol Aging 2001, 22(6):799-809.
9. Eikelenboom P, Bate C, Van Gool WA, HoozemansJJ, RozemullerJM,
Veerhuis R, Williams A: Neuroinflammation in Alzheimer's dis-
ease and prion disease. Glia 2002, 40(2):232-239.

Page 9 of 10
(page number not for citation purposes)

Journal of Neuroinflammation 2007, 4:20

10. Akiyama H, Barger S, Barnum S, Bradt B, Bauer j, Cole GM, Cooper
NR, Eikelenboom P, Emmerling M, Fiebich BL, Finch CE, Frautschy S,
Griffin WS, Hampel H, Hull M, Landreth G, Lue L, Mrak R, Mackenzie
IR, McGeer PL, O'Banion MK, Pachter Pasinetti G, Plata-Salaman C,
Rogers J, Rydel R, Shen Y, Streit W, Strohmeyer R, Tooyoma I, Van
Muiswinkel FL, Veerhuis R, Walker D, Webster S, Wegrzyniak B,
Wenk G, Wyss-Coray T: Inflammation and Alzheimer's dis-
ease. Neurobiol Aging 2000, 21(3):383-421.
I I. Streit WJ: Microglia and Alzheimer's disease pathogenesis. J
Neurosci Res 2004, 77(1):1-8.
12. McDonald DR, Brunden KR, Landreth GE: Amyloid fibrils activate
tyrosine kinase-dependent signaling and superoxide produc-
tion in microglia. J Neurosci 1997, 17(7):2284-2294.
13. Lue LF, Brachova L, Civin WH, Rogers J: Inflammation, A beta
deposition, and neurofibrillary tangle formation as corre-
lates of Alzheimer's disease neurodegeneration. J Neuropathol
Exp Neurol 1996, 55(10):1083-1088.
14. Meda L, Cassatella MA, Szendrei GI, Otvos LJr., Baron P, Villalba M,
Ferrari D, Rossi F: Activation of microglial cells by beta-amy-
loid protein and interferon-gamma. Nature 1995,
15. Sparks DL, Schreurs BG: Trace amounts of copper in water
induce beta-amyloid plaques and learning deficits in a rabbit
model of Alzheimer's disease. Proc Notl Acad Sci U S A 2003,
16. Sparks DL, Friedland R, Petanceska S, Schreurs BG, Shi J, Perry G,
Smith MA, Sharma A, Derosa S, Ziolkowski C, Stankovic G: Trace
copper levels in the drinking water, but not zinc or alumi-
num influence CNS Alzheimer-like pathology. J Nutr Health
Aging 2006, 10(4):247-254.
17. Streit WJ, Kreutzberg GW: Lectin binding by resting and reac-
tive microglia. J Neurocytol 1987, 16(2):249-260.
18. Streit WJ: An improved staining method for rat microglial
cells using the lectin from Griffonia simplicifolia (GSA 1-B4).
J Histochem Cytochem 1990, 38(1 I):1683-1686.
19. Perry G, Friedman R, Shaw G, Chau V: Ubiquitin is detected in
neurofibrillary tangles and senile plaque neurites of Alzhe-
imer disease brains. Proc Notl Acad Sci U S A 1987,
20. Zatta P, Zambenedetti P, Stella MP, Licastro F: Astrocytosis,
microgliosis, metallothionein-1-ll and amyloid expression in
high cholesterol-fed rabbits. J Alzheimers Dis 2002, 4(1):1-9.
21. Kreutzberg GW: Microglia: a sensor for pathological events in
the CNS. Trends Neurosci 1996, 19(8):312-318.
22. Ito S, Sawada M, Haneda M, Fujii S, Oh-Hashi K, Kiuchi K, Takahashi
M, Isobe K: Amyloid-beta peptides induce cell proliferation
and macrophage colony-stimulating factor expression via
the P13-kinase/Akt pathway in cultured Ra2 microglial cells.
FEBS Lett 2005, 579(9):1995-2000.
23. Floden AM, Li S, Combs CK: Beta-amyloid-stimulated microglia
induce neuron death via synergistic stimulation of tumor
necrosis factor alpha and NMDA receptors. J Neurosci 2005,
24. Streit WJ, Mrak RE, Griffin WS: Microglia and neuroinflamma-
tion: a pathological perspective. J Neuroinflammation 2004,
25. Aguirre N, Beal MF, Matson WR, Bogdanov MB: Increased oxida-
tive damage to DNA in an animal model of amyotrophic lat-
eral sclerosis. Free Radic Res 2005, 39(4):383-388.
26. Eikelenboom P, Veerhuis R, Scheper W, Rozemuller AJ, van Gool
WA, Hoozemans JJ: The significance of neuroinflammation in
understanding Alzheimer's disease. J Neural Transm 2006,
113(1 1):1685-1695.
27. Town T, Nikolic V, Tan J: The microglial "activation" contin-
uum: from innate to adaptive responses. J Neuroinflammation
2005, 2:24.
28. Sparks DL: Cholesterol metabolism and brain amyloidosis:
evidence for a role of copper in the clearance of Abeta
through the liver. Curr Alzheimer Res 2007, 4(2): 165-169.
29. Hallmann R, Mayer DN, Berg EL, Broermann R, Butcher EC: Novel
mouse endothelial cell surface marker is suppressed during
differentiation of the blood brain barrier. Dev Dyn 1995,
30. Engelhardt B, Conley FK, Butcher EC: Cell adhesion molecules on
vessels during inflammation in the mouse central nervous
system. J Neuroimmunol 1994, 51(2): 199-208.

31. Rabchevsky AG, Degos JD, Dreyfus PA: Peripheral injections of
Freund's adjuvant in mice provoke leakage of serum pro-
teins through the blood-brain barrier without inducing reac-
tive gliosis. Brain Res 1999, 832(1-2):84-96.
32. StreitWJ: Microglial senescence: does the brain's immune sys-
tem have an expiration date? Trends Neurosci 2006,
33. Streit WJ: Microglia as neuroprotective, immunocompetent
cells of the CNS. Glia 2002, 40(2):133-139.

Page 10 of 10
(page number not for citation purposes)

Publish with BioMed Central and every
scientist can read your work free of charge
"BioMed Central will be the most significant development for
disseminating the results of biomedical research in our lifetime."
Sir Paul Nurse, Cancer Research UK
Your research papers will be:
available free of charge to the entire biomedical community
peer reviewed and published immediately upon acceptance
cited in PubMed and archived on PubMed Central
yours you keep the copyright

Submit your manuscript here: BioMedcentral

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs