Combined effects of hyperglycemic conditions and HIV-1 Nef: a potential model for induced HIV neuropathogenesis.

Abstract

Hyperglycemic conditions associated with diabetes mellitus (DM) or with the use of antiretroviral therapy may increase the risk of central nervous system (CNS) disorders in HIV-1 infected patients. In support of this hypothesis, we investigated the combined effects of hyperglycemic conditions and HIV-1 accessory protein Nef on the CNS using both in vitro and in vivo models. Astrocytes, the most abundant glial cell type required for normal synaptic transmission and other functions were selected for our in vitro study. The results show that in vitro hyperglycemic conditions enhance the expression of proinflammatory cytokines including caspase-3, complement factor 3 (C3), and the production of total nitrate and 8-iso-PGF2 alpha as reactive oxygen species (ROS) in human astrocytes leading to cell death in a dose-dependent manner. Delivery of purified recombinant HIV-1 Nef protein, or Nef expressed via HIV-1-based vectors in astrocytes showed similar results. The expression of Nef protein delivered via HIV-1 vectors in combination with hyperglycemia further augmented the production of ROS, C3, activation of caspase-3, modulation of filamentous protein (F-protein), depolarization of the mitochondria, and loss of astrocytes. To further verify the effects of hyperglycemia and HIV-1 Nef protein on CNS individually or in combination, in vivo studies were performed in streptozotocin (STZ) induced diabetic mice, by injecting HIV-1 Nef expressing viral particles into the sub-cortical region of the brain. Our in vivo results were similar to in vitro findings indicating an enhanced production of caspases-3, ROS (lipid oxidation and total nitrate), and C3 in the brain tissues of these animals. Interestingly, the delivery of HIV-1 Nef protein alone caused similar damage to CNS as augmented by hyperglycemia conditions. Taken together, the data suggests that HIV-1 infected individuals with hyperglycemia could potentially be at a higher risk of developing CNS related complications.

Full text

Page 1

BioMed Central
Page 1 of 14
(page number not for citation purposes)
Virology Journal
Open Access
Research
Combined effects of hyperglycemic conditions and HIV-1 Nef: a
potential model for induced HIV neuropathogenesis
Edward A Acheampong1, Cassandra Roschel2, Muhammad Mukhtar3,
Alagarsamy Srinivasan4, Mohammad Rafi5, Roger J Pomerantz6 and
Zahida Parveen*1
Address: 1The Dorrance H. Hamilton Laboratories, Division of Infectious Diseases and Environmental Medicine, PA 19107, USA, 2Bioscience
Technologies - Biotechnology, Thomas Jefferson University, Philadelphia, PA 19107, USA, 3Department of Biochemistry, Pir Mehr Ali Shah Arid
Agriculture University Rawalpindi, 46300 Pakistan, 4NanoBio Diagnostics, West Chester, PA 19382, USA, 5Department of Neurology, Jefferson
Medical College, Thomas Jefferson University, Philadelphia, PA 19107, USA and 6Tibotec Inc. 1020 Stony Hill Road, Suite 300, Yardley, PA 19067,
USA
Email: Edward A Acheampong - eacheamp@yahoo.com; Cassandra Roschel - Cassandra.Roschel@gmail.com;
Muhammad Mukhtar - mukhtar.muhammad@gmail.com; Alagarsamy Srinivasan - alagarsamy.srinivasan@gmail.com;
Mohammad Rafi - Mohammad.Rafi@jefferson.edu; Roger J Pomerantz - RPomeran@its.jnj.com;
Zahida Parveen* - zahida.parveen@jefferson.edu
* Corresponding author
Abstract
Hyperglycemic conditions associated with diabetes mellitus (DM) or with the use of antiretroviral
therapy may increase the risk of central nervous system (CNS) disorders in HIV-1 infected patients.
In support of this hypothesis, we investigated the combined effects of hyperglycemic conditions and
HIV-1 accessory protein Nef on the CNS using both in vitro and in vivo models. Astrocytes, the most
abundant glial cell type required for normal synaptic transmission and other functions were
selected for our in vitro study. The results show that in vitro hyperglycemic conditions enhance the
expression of proinflammatory cytokines including caspase-3, complement factor 3 (C3), and the
production of total nitrate and 8-iso-PGF2 α as reactive oxygen species (ROS) in human astrocytes
leading to cell death in a dose-dependent manner. Delivery of purified recombinant HIV-1 Nef
protein, or Nef expressed via HIV-1-based vectors in astrocytes showed similar results. The
expression of Nef protein delivered via HIV-1 vectors in combination with hyperglycemia further
augmented the production of ROS, C3, activation of caspase-3, modulation of filamentous protein
(F-protein), depolarization of the mitochondria, and loss of astrocytes. To further verify the effects
of hyperglycemia and HIV-1 Nef protein on CNS individually or in combination, in vivo studies were
performed in streptozotocin (STZ) induced diabetic mice, by injecting HIV-1 Nef expressing viral
particles into the sub-cortical region of the brain. Our in vivo results were similar to in vitro findings
indicating an enhanced production of caspases-3, ROS (lipid oxidation and total nitrate), and C3 in
the brain tissues of these animals. Interestingly, the delivery of HIV-1 Nef protein alone caused
similar damage to CNS as augmented by hyperglycemia conditions. Taken together, the data
suggests that HIV-1 infected individuals with hyperglycemia could potentially be at a higher risk of
developing CNS related complications.
Published: 30 October 2009
Virology Journal 2009, 6:183doi:10.1186/1743-422X-6-183
Received: 4 May 2009
Accepted: 30 October 2009
This article is available from: http://www.virologyj.com/content/6/1/183
© 2009 Acheampong et al; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Page 2

Virology Journal 2009, 6:183http://www.virologyj.com/content/6/1/183
Page 2 of 14
(page number not for citation purposes)
Introduction
Antiretroviral therapy has been linked to insulin resist-
ance and dyslipidemia in HIV infected individuals under
treatment [1-4]. Since glucose is a major nutrient utilized
by the brain[5], diabetes or HAART-associated hyperglyc-
emic conditions may become a potential risk factor in the
brain [6-8], and could lead to a series of devastating clini-
cal conditions in the CNS of HIV-1 infected individu-
als[9]. Several studies have described hyperglycemia-
induced neuronal and astrocytic glial cell death leading to
various neurological
patients[7,10,11]. However, limited information is avail-
able regarding the combined effects of hyperglycemia and
HIV-1 infection on the CNS. Astrocytes play a critical role
in the provision of nutrients and strength to the CNS via
the foot processes protecting the blood brain barrier [12].
In this study, we selected astrocytes as target cells to eval-
uate the cumulative toxic effects of hyperglycemia and
HIV-1 Nef protein. Previous studies have shown that
hyperglycemia increases the production of proinflamma-
tory cytokines, oxidative reactive species and activation of
CD4+ and CD8 T lymphocytes in the peripheral blood
system [13]. Of the proteins encoded by HIV-1, Env, Vpr,
Vif, Tat, and Nef are known to exhibit cytopathic effects
[14-16]. Specifically, the data from previous studies sug-
gest a potentially important role of Nef in cellular dys-
functions and its contribution to the development of the
neuropathology associated with AIDS. HIV-1 Nef expres-
sion has been shown to be essential in maintaining high
replication level of the virus and promoting the develop-
ment of AIDS in SIV-infected monkeys[17]. Skowronski
and others have shown that the expression of Nef in trans-
genic mice is associated with the development of a severe
AIDS like disease [18,19]. Nef and gp120 have been
detected in the CSF of HIV-1 infected individuals and are
known to be involved in the induction of complement
factor C3 [9,20,21]. HIV-1 infection, thus affects the cellu-
lar processes in the brain by activating signaling pathways
and the production of cytokines [22,23]. It has been
reported that extracellular release of Nef protein could
exert its effects on non-infected bystander cells in brain
tissues of HIV-1 infected individuals and could be
detected in distant brain regions [14,17]. HIV-1 proteins
also cause an increase in systemic oxidative/nitrosative
stress, by enhancing the deleterious effects of secondary
infections [9]. The molecular mechanism involved in
HIV-1 associated neuropathogenesis is not completely
understood due to the inaccessibility of the brain paren-
chyma during the course of AIDS. Hence, limited infor-
mation is available regarding the contributions of Nef
alone and or in combination with hyperglycemic condi-
tions to the pathogenesis of the CNS in the context of
HIV-1 infection. The focus of this study was to evaluate
the cytopathic effects of hyperglycemic conditions in the
presence of HIV-1 Nef delivered either through HIV-1-
disorders in diabetic
based vector systems (intracellular) or in the form of
recombinant protein (extracellular) in human astrocytes
(in vitro) and STZ induced diabetic mice used as an in vivo
model[24]. The delivery of Nef protein via viral injection
into the STZ induced diabetic mice brain increased oxida-
tive reactions as well as the production of inflammatory
cytokines, complement factor C3, and depolarization of
mitochondria. Induction of in vitro and in vivo hyperglyc-
emia alone induced similar cytopathic effects in astrocytes
and in diabetes induced mice. Further, the data involving
astrocytes suggests that the presence of extracellular Nef
protein further increased the risk of toxicity and cell death
in a dose-dependent manner under hyperglycemic condi-
tions.
Materials and methods
Cell Culture
Primary cultures of human fetal brain astrocytes and
astrocytes medium were purchased from Cambrex, Inc
(Walkersville, MD) and Sciencell (San Diego, CA). The
cells were maintained in astrocyte media (AM) in a water-
jacketed incubator at 37°C, with 5% CO2 in a humid envi-
ronment. The cells were passaged at a confluence of 80-
85%. The human glioblastoma/astrocytoma cell line
U87-MG, and human kidney cell line 293-T were
obtained from American Type Culture Collection (ATCC)
and cultured in Dubelcco's Modified Eagle's Medium
(DMEM) supplemented with 10% fetal bovine serum
(Sigma Aldrich, St. Louis, MO), penicillin-streptomycin
(100 U/ml and 100 μg/ml, respectively), and 2 mM L-
glutamate (Mediatech Corp, MD).
Generation of Nef expressing viral particles
andtransduction
HIV-1 Nef expressing recombinant viral particles were
generated by triple transfection of plasmids using Cal-
cium phosphate transfection kit (Promega Corp, Madi-
son, WI) following the manufacturer's protocol. Briefly,
293 T cells were seeded in 100 mm culture plates over-
night. The cells were transfected with reagents of Mamma-
lian Calcium Phosphate transfection kit (from Promega)
in the presence of HIV-1 based vectors DNA; pHR'CMV
Nef, pHR CMV delta 8.2, and pMD.G encoding VSV.G as
an envelope protein. In addition, viral particles expressing
HIV-1 Nef generated from spleen necrosis virus (SNV)
packaging vector pZP32, transfer vector expressing HIV-1
Nef pZP35, and envelope vector VSV.G [14,25] were used
as control. The supernatants from HIV-1 and SNV based
viral vectors were harvested 3 days post transfection and
frozen at -80°C. For some experiments both viruses were
concentrated by ultracentrifugation at 25,000 rpm for one
hour. The pellets were resuspended in 1% phosphate-
buffered saline (PBS) containing 5% sucrose and stored at
-80°C. The viral yield for HIV-1 was determined by p24
antigen enzyme-linked immunosorbent assay (ELISA) kit

Page 3

Virology Journal 2009, 6:183http://www.virologyj.com/content/6/1/183
Page 3 of 14
(page number not for citation purposes)
(Perkin Elmer, Boston, MA). Based on the quantification,
equal amount of viral particles were used for the experi-
ments. Culture supernatants without hyperglycemia and
Nef were collected from the astrocytes and used as a
source for mock treatment. Astrocytes were plated at 60%
confluency over night before tranduction. Primary human
astrocytes or U87-MG cells were distributed into 4-well
chamber slides or plates at a cell density of 1.0 × 105 cells
per well and allowed to stabilize in AM media for 24
hours prior to the addition of glucose media. After the sta-
bilization period, the cells in each well were washed with
pre-warmed 1× PBS. To mimic the in vivo hyperglycemia,
glucose stock solutions were added to glucose-free media
(contained 1.0 mM sodium pyruvate, 1% strep/pen, and
5% FBS) to achieve 10 mM, 15 mM, and 20 mM glucose
concentrations. Of note, 10, 15, and 20 mM glucose rep-
resents the 180, 200, 350 mg glucose/dl blood in diabetic
patients. The medium with 5.0 mM glucose was used as a
control. The astrocytes were exposed to in vitro hyperglyc-
emic conditions for 12 hours and washed with 1× PBS.
The astrocytes were then transduced with viral superna-
tant mixed with 8 ug/ml polybrene for astrocytic cell
line(U87-MG) and 4 ug/ml for primary astrocytes for 3
hours followed by washing to remove the virus and incu-
bated with complete medium. Astrocytes were harvested
and supernatants were collected after 48 hours for various
analyses. Non- transduced astrocytes were used as a con-
trol.
In vitro effects of hyperglycemia and recombinant Nef
protein on human astrocytes
Individual and cummulative effects of hyperglycemia and
recombinant Nef protein on primary human fetal astro-
cytes were evaluated by observing changes in the F-actin,
a protein involved in mitochondrial and cellular integrity
[26]. Astrocytes were seeded into 4-well chamber slides at
a cell density of 1.0 × 105 cells per well and exposed to var-
ious hyperglycemic conditions for 12 hours, followed by
extensive washing with 1× PBS. The cells were fixed with
4% paraformaldehyde for 10 minutes and washed several
times with 1× PBS to remove the fixative. The astrocytes
were then stained with BODIPY phallacidin (Invitrogen
Corporation, Carsbad, CA) cytoskeleton staining dye fol-
lowing protocol suggested by the manufacturer, and
observed under fluorescence-microscope.
The effect of recombinant Nef protein on mitochondria
was studied by using Mitotracker dye (Invitrogen Corpo-
ration, Carsbad, CA). The dye stains mitochondria only
upon depolarization. For this, astrocytes were seeded in
chamber slides in AM medium and allowed to attach for
24 hours prior to Nef protein treatments. Recombinant
Nef protein generated in our laboratory [14] was added at
concentrations of 1 nM, 3 nM, and 25 nM in 500 μl of glu-
cose free DMEM containing 1.0 mM sodium pyruvate, 1%
strep/pen, and 5% FBS and incubated with astrocytes for
24 hours. The cells were washed with 1XPBS and stained
live with 200 nM Mitotracker Red fluorochrome (Invitro-
gen Corp., Carlsbald, CA), to detect the depolarization of
mitochondria.
Induction of Diabetes
Mice with C57/BL6 genetic background were purchased
from Jackson Laboratories to study the effects of hypergly-
cemic variations either alone or in combination with
intracellularly expressed HIV-1 Nef protein in vivo[27]. To
rule out the effect of other accessory proteins associated
with HIV-1 virus, SNV vector virus encoding Nef was also
included in the study. The exact physiological concentra-
tion of Nef is not clear in HIV-1 infected individuals.
However extracellular recombinant Nef protein generated
in our laboratory was added to astrocytes at concentra-
tions mentioned in previous studies [28]. Diabetes was
induced in 12 mice by a single subcutaneous injection of
40 mg/kg body weight streptozotocin (Sigma Aldrich
Corp., St.Louis, MO) dissolved in freshly prepared 0.1 M
citrate buffer pH 4.5. The blood glucose level was assessed
by a glucometer using a drop of blood drawn at 1, 2, 4, 6,
8, 10, and 12 hours post injection. The mean elevated glu-
cose level was 325 mg/dl after injection of STZ. Upon con-
firmation of induction of hyperglycemia in mice, 2-ul of
concentrated viral particles (1 × 107) generated through
HIV-1 and SNV-based vectors systems were injected into
the brain of mice via the cortex as described previously
[29]. Age-matched non-diabetic and STZ treated (diabe-
tes) mice injected with an equal volume of citrate buffer,
served as controls. The animals were housed under path-
ogenic free conditions in Thomas Jefferson Animal Facil-
ity. Mice from the ages of 1-2 weeks of the same sex were
used in the experiments. All procedures were conducted in
accordance with federal guidelines using animal protocols
approved by the Thomas Jefferson University Institutional
Animal Care and Use Committee (IACUC). The mice were
sacrificed eight weeks post-injection and their brains were
detached, washed in cold 1× PBS and used for various
analyses (25).
Western Blot Analyses
Astrocytes exposed to various glucose solutions to induce
hyperglycemia, or in combination with HIV-1 Nef, or
transduced with Nef alone, as well as non-treated control
astrocytes were washed and lysed in radio Immunoprecip-
itation assay (RIPA) buffer containing protease inhibitors.
The protein concentrations were determined with the
bicinchoninic acid protein assay kit (Pierce Biotechnolo-
gies, Rockford, IL). Approximately 25 μg of each protein
preparation was resolved on 10% sodium dodecyl sulfate
polyacrylamide gels (Bio-Rad) and transferred to polyvi-
nylidene difluoride (PVDF) membranes (Amersham Bio-
sciences, Piscataway, NJ) using electroblotting method.

Page 4

Virology Journal 2009, 6:183http://www.virologyj.com/content/6/1/183
Page 4 of 14
(page number not for citation purposes)
The membranes were washed in PBS containing 0.01%
Tween 20 (Sigma-Aldrich, St. Louis, MO.). Non-specific
proteins were blocked with PBS-based blocking buffer
(Pierce Biotechnologies, Rockford, IL) and the mem-
branes were probed with specific monoclonal antibodies
against GFAP at a concentration of 1:1000, mouse anti-
Caspase 3 antibody at a concentration of 1:1000 as pri-
mary antibodies and horseradish peroxidase labeled anti-
mouse immunoglobulin G (heavy plus light chains) as
secondary antibodies. The protein-antibody complexes
were visualized by autoradiography of the membranes
after incubating with the ECL blotting detection system
(Pierce Biotechnologies, Rockford, IL) and subsequently
exposing them to BioMax MS (Kodak, Rochester, N.Y.)
film. HIV-1 Nef protein was detected in the transduced
astrocytes by treating the blots with anti-HIV-1 Nef anti-
body (NIH AIDS Repository). For in vivo analysis of the
expression of Nef, viral vector expressing Nef was identi-
fied from mice brain tissues using immunoprecipitation
method. The Seize X Immunoprecipitation Kit (Pierce
Biotechnologies, Rockford, IL) was used following the
manufacturer's protocol. The purified Nef protein was
then subjected to Western Blot analysis using the method
described earlier.
Enzyme Linked Immunosorbent Assays (ELISAs)
The production of nitric oxide and lipid oxidation reac-
tion in the form of total nitrate and 8-iso- PGF-2α, respec-
tively, were measured to determine the level of reactive
oxidative species induced as a result of exposure to either
hyperglycemic conditions alone or in combination with
HIV-1 Nef protein or Nef alone. The U87-MG cells were
exposed to various concentrations of glucose (10, 15, 20,
25 mM glucose solutions) followed by transduction with
HIV-1 Nef expressing viral particles. Control astrocytes
were cultured in normal medium. For in vivo studies, 10-
day-old mice were injected with a single dose of STZ for
induction of hyperglycemia followed by delivery of HIV-1
based Nef expressing virus via injection in the brain tis-
sues. Non-diabetic mice, hyperglycemic mice, or HIV-1
Nef injected mice were used as controls. To rule out the
influence of other accessory proteins of HIV-1 and for
exclusive effect of Nef protein, mice were also injected
with virus generated by SNV vector systems [27]. The cell/
tissue lysates and supernatants from the treated cells as
well as from brain tissues from the hyperglycemic and Nef
treated mice were collected. Samples were analyzed for
the presence of nitric oxide (NO), 8-isoprostaglandin-F2-
α, or complement factor C3 using respective ELISA kits
(Stressgen Biotechnologies, Victoria, BC, Canada) as well
as the manufacturer's suggested protocol [30].
Results
In this study, we utilized in vitro and in vivo models to eval-
uate the combined cytopathic effects of hyperglycemia
and HIV-1 proteins on the CNS to mimic the conditions
in individuals with diabetes or hyperglycemia associated
with the use of highly active antiretroviral therapy
(HAART). For in vitro studies, U87-MG/primary astrocytes
were exposed to various hyperglycemic conditions by
adding the appropriate amount of glucose in medium
[31] and tranduced with HIV-1 Nef expressing virus. For
the in vivo studies, diabetes was induced in mice with STZ
and recombinant viral particles expressing HIV-1 Nef were
injected in both STZ induced diabetic and normal mice
brains. The combined cytopathic effects of hyperglycemia
and HIV-1 Nef protein on the CNS were determined by
evaluating the expression of complement factor 3, pro-
duction of oxidative species (ROS), caspase activity,
changes in F-actin protein, and the depolarization of
mitochondria.
Effect of Hyperglycemia and HIV-1 Nef on Complement
Factor 3(C3)
The cerebral complement system has been known as a
contributor to AIDS-associated neurological disorders. To
evaluate the inflammatory response during hyperglyc-
emia and/or HIV- infection in the CNS, complement fac-
tor 3 was used as an indirect measure of immune response
[20,21]. Our results indicate that the exposure of astro-
cytes to 10, 15, 20 and 25 mM glucose increased the
expression of C3 (2.0, 4.0, 10 and 10.7 fold) in a dose-
dependent fashion respectively. Expression of Nef via
HIV-1 vectors in astrocytes exposed to 10, 15, 20 and 25
mM glucose resulted in an increase of more than 4.0, 6.0,
16 and 12 fold respectively in the production of C3 (Fig-
ure 1A). Exposure of astrocytes to HIV-1 Nef alone also
enhanced the production of C3 to more than 4 fold, sug-
gesting that HIV-1 Nef itself is capable of inducing
immune response[20]. The effect of hyperglycemia on C3
production was also studied in vivo using STZ-induced
diabetic mouse model. Our results from diabetic mice
were similar to the results obtained from the in vitro study
in astrocytes. We observed more than 6-fold increase in
the production of C3 in diabetic mice brain as compared
to the normal mice. The expression of Nef particles deliv-
ered via injection into normal mice brain resulted in 8-
fold increase in C3, while the expression of Nef in diabetic
mice resulted in more than 10-fold increase in C3 produc-
tion (Figure 1B) as compared to normal mice used as con-
trol. Delivery of HIV-1 Nef via SNV vectors into mice brain
also showed similar increase in C3, suggesting the exclu-
sive effect of Nef in enhancing the C3 production. These
results indicate that in vitro hyperglycemia or in vivo dia-
betic conditions increase the immune response in the
form of complement factor 3 production in CNS, whereas
the expression of Nef under normal glycemia or in combi-
nation with hyperglycemia further enhanced the produc-
tion of C3 as a consequence of severe immune reaction
[12].

Page 5

Virology Journal 2009, 6:183http://www.virologyj.com/content/6/1/183
Page 5 of 14
(page number not for citation purposes)
Detection of Reactive Oxygen Species (ROS)
Effect of Hyperglycemia and HIV-1 Nef on Nitric Oxide Production
The ability of hyperglycemia to induce reactive oxygen
species (ROS) thereby enhancing the production of nitric
oxide (total nitrate) and lipid peroxidation in the form of
8-iso-prostaglandin F2 alpha (8-iso-PGF2 alpha) are well
documented, and have been previously used as biological
markers to detect the oxidative stress levels [14,32,33]. In
this study, we investigated the production of reactive oxy-
gen species by determining the level of total nitrates pro-
duced due to in vitro hyperglycemic conditions or due to
the expression of HIV-1 Nef protein in astrocytes or com-
Figure 1
Hyperglycemic conditions and HIV-1 Nef significantly enhance the production of complement factor C3 in
vitro and in vivo. (A) To mimic hyperglycemic conditions close to the blood glucose levels of 180, 270 and 360 mg/dl, U87-
MG human astroglioma cells were cultured with 10,15, 20 and 25 mM glucose containing medium for 12 hours. Astrocytes
with 5 mM glucose treatment were used as control. The cells were washed and transduced with HIV-1 Nef expressing virus.
48 hours later, the astrocytes and cellular supernatants were collected and subjected to ELISA using manufacturer's protocol
to quantify the complement factor 3 (Assay Designs, Ann Arbor, MI) (B). Hyperglycemic conditions and expression of HIV-1
Nef significantly enhanced the production of complement factor 3 in mice brain. Diabetes was induced in C57/BL6 mice by a
subcutaneous injection of a single dose of 40 mg/kg body weight streptozotocin (Sigma Chemicals, St.Louis, MO), which has
been freshly dissolved in 0.1 mol/L citrate buffer at pH of 4.5. Upon confirmation of diabetes induction (325-425 mg/dl glucose)
by a glucometer in these mice, 1 × 107 viral particles generated through HIV-1 based vectors or SNV-based vectors were
injected into the mice brain via the mid ventricle, cortex, or the cerebellum as described previously. Age-matched non-diabetic
mice injected with an equal volume of citrate buffer were served as control. After eight weeks the mice were sacrificed and the
brain and other organs were harvested. ELISA was performed on brain tissue extracts to determine the release of C3 into the
brain. The results are the mean values for triplicate samples ± standard errors of the means. The data presented are averages
of three independent experiments.